diff --git a/AIlib/AI.py b/AIlib/AI.py deleted file mode 100644 index fc65de0..0000000 --- a/AIlib/AI.py +++ /dev/null @@ -1,90 +0,0 @@ - -import cv2,os,time -from models.experimental import attempt_load -from segutils.segmodel import SegModel,get_largest_contours -from utils.torch_utils import select_device -from utilsK.queRiver import get_labelnames,get_label_arrays,post_process_ -from utils.datasets import letterbox -import numpy as np -import torch - -def AI_process(im0s,model,segmodel,names,label_arraylist,rainbows,half=True,device=' cuda:0',conf_thres=0.25, iou_thres=0.45,allowedList=[0,1,2,3]): - #输入参数 - # im0s---原始图像列表 - # model---检测模型，segmodel---分割模型 - #输出：两个元素（列表，字符）构成的元组，[im0s[0],im0,det_xywh,iframe],strout - # [im0s[0],im0,det_xywh,iframe]中， - # im0s[0]--原始图像，im0--AI处理后的图像,iframe--帧号/暂时不需用到。 - # det_xywh--检测结果，是一个列表。 - # 其中每一个元素表示一个目标构成如：[float(cls_c), xc,yc,w,h, float(conf_c)] - # #cls_c--类别，如0,1,2,3; xc,yc,w,h--中心点坐标及宽；conf_c--得分, 取值范围在0-1之间 - # #strout---统计AI处理个环节的时间 - - # Letterbox - img = [letterbox(x, 640, auto=True, stride=32)[0] for x in im0s] - # Stack - img = np.stack(img, 0) - # Convert - img = img[:, :, :, ::-1].transpose(0, 3, 1, 2) # BGR to RGB, to bsx3x416x416 - img = np.ascontiguousarray(img) - - - img = torch.from_numpy(img).to(device) - img = img.half() if half else img.float() # uint8 to fp16/32 - - img /= 255.0 # 0 - 255 to 0.0 - 1.0 - - seg_pred,segstr = segmodel.eval(im0s[0] ) - pred = model(img,augment=False)[0] - datas = [[''], img, im0s, None,pred,seg_pred,10] - - p_result,timeOut = post_process_(datas,conf_thres, iou_thres,names,label_arraylist,rainbows,10,object_config=allowedList) - - return p_result,timeOut - -def main(): - ##预先设置的参数 - device_='1' ##选定模型，可选 cpu,'0','1' - - ##以下参数目前不可改 - Detweights = "weights/yolov5/class5/best_5classes.pt" - seg_nclass = 2 - Segweights = "weights/BiSeNet/checkpoint.pth" - conf_thres,iou_thres,classes= 0.25,0.45,5 - labelnames = "weights/yolov5/class5/labelnames.json" - rainbows = [ [0,0,255],[0,255,0],[255,0,0],[255,0,255],[255,255,0],[255,129,0],[255,0,127],[127,255,0],[0,255,127],[0,127,255],[127,0,255],[255,127,255],[255,255,127],[127,255,255],[0,255,255],[255,127,255],[127,255,255], [0,127,0],[0,0,127],[0,255,255]] - allowedList=[0,1,2,3] - - - ##加载模型，准备好显示字符 - device = select_device(device_) - names=get_labelnames(labelnames) - label_arraylist = get_label_arrays(names,rainbows,outfontsize=40,fontpath="conf/platech.ttf") - half = device.type != 'cpu' # half precision only supported on CUDA - model = attempt_load(Detweights, map_location=device) # load FP32 model - if half: model.half() - segmodel = SegModel(nclass=seg_nclass,weights=Segweights,device=device) - - - - ##图像测试 - #url='images/examples/20220624_响水河_12300_1621.jpg' - impth = 'images/examples/' - outpth = 'images/results/' - folders = os.listdir(impth) - for i in range(len(folders)): - imgpath = os.path.join(impth, folders[i]) - im0s=[cv2.imread(imgpath)] - time00 = time.time() - p_result,timeOut = AI_process(im0s,model,segmodel,names,label_arraylist,rainbows,half,device,conf_thres, iou_thres,allowedList) - time11 = time.time() - image_array = p_result[1] - cv2.imwrite( os.path.join( outpth,folders[i] ) ,image_array ) - print('----process:%s'%(folders[i]), (time.time() - time11) * 1000) - - - - - -if __name__=="__main__": - main() diff --git a/AIlib/__pycache__/AI.cpython-38.pyc b/AIlib/__pycache__/AI.cpython-38.pyc deleted file mode 100644 index b805d49..0000000 Binary files a/AIlib/__pycache__/AI.cpython-38.pyc and /dev/null differ diff --git a/AIlib/conf/bak/model_5class.json b/AIlib/conf/bak/model_5class.json deleted file mode 100644 index e1c111d..0000000 --- a/AIlib/conf/bak/model_5class.json +++ /dev/null @@ -1,16 +0,0 @@ -{ - - "gpu_process":{"det_weights":"../yolov5/weights/best_5classes.pt","seg_nclass":2,"seg_weights": "../yolov5/weights/segmentation/BiSeNet/checkpoint.pth" }, - - "post_process":{ "name":"post_process","conf_thres":0.25,"iou_thres":0.45,"classes":5,"labelnames":"../yolov5/config/labelnames.json","fpsample":240,"debug":false , "rainbows":[ [0,0,255],[0,255,0],[255,0,0],[255,0,255],[255,255,0],[255,129,0],[255,0,127],[127,255,0],[0,255,127],[0,127,255],[127,0,255],[255,127,255],[255,255,127],[127,255,255],[0,255,255],[255,127,255],[127,255,255], [0,127,0],[0,0,127],[0,255,255]],"outImaDir":"problems/images_tmp","outVideoDir":"problems/videos_save" }, - - "push_process":{ "OutVideoW":1920, "OutVideoH":1080 }, - "AI_video_save": {"onLine":false,"offLine":true }, - "imageTxtFile":true, - "logChildProcessOffline":"logs/logChildProcess/offline", - "logChildProcessOnline":"logs/logChildProcess/online", - "StreamWaitingTime":240, - "StreamRecoveringTime":180 - - -} diff --git a/AIlib/conf/bak/model_9class.json b/AIlib/conf/bak/model_9class.json deleted file mode 100644 index 2b60325..0000000 --- a/AIlib/conf/bak/model_9class.json +++ /dev/null @@ -1,16 +0,0 @@ -{ - - "gpu_process":{"det_weights":"../weights/yolov5/class9/weights/best.pt","seg_nclass":2,"seg_weights": "../yolov5/weights/segmentation/BiSeNet/checkpoint.pth" }, - - "post_process":{ "name":"post_process","conf_thres":0.25,"iou_thres":0.45,"classes":5,"labelnames":"../weights/yolov5/class9/labelnames.json","fpsample":240,"debug":false , "rainbows":[ [0,0,255],[0,255,0],[255,0,0],[255,0,255],[255,255,0],[255,129,0],[255,0,127],[127,255,0],[0,255,127],[0,127,255],[127,0,255],[255,127,255],[255,255,127],[127,255,255],[0,255,255],[255,127,255],[127,255,255], [0,127,0],[0,0,127],[0,255,255]],"outImaDir":"problems/images_tmp","outVideoDir":"problems/videos_save" }, - - "push_process":{ "OutVideoW":1920, "OutVideoH":1080 }, - "AI_video_save": {"onLine":false,"offLine":true }, - "imageTxtFile":true, - "logChildProcessOffline":"logs/logChildProcess/offline", - "logChildProcessOnline":"logs/logChildProcess/online", - "StreamWaitingTime":240, - "StreamRecoveringTime":180 - - -} diff --git a/AIlib/conf/errorDic.json b/AIlib/conf/errorDic.json deleted file mode 100644 index 5809a4f..0000000 --- a/AIlib/conf/errorDic.json +++ /dev/null @@ -1,6 +0,0 @@ -{ - "101":"video uploading failure", - "102":"Stream or video ERROR", - "": - -} diff --git a/AIlib/conf/master.json b/AIlib/conf/master.json deleted file mode 100644 index 0cf452d..0000000 --- a/AIlib/conf/master.json +++ /dev/null @@ -1,14 +0,0 @@ -{ -"par":{ - "server":"212.129.223.66:19092", - "server2":"101.132.127.1:19092", - "server3":"192.168.11.242:9092", - "topic": ["dsp-alg-online-tasks","dsp-alg-offline-tasks","dsp-alg-task-results"], - "group_id":"testWw", - "kafka":"mintors/kafka", - "modelJson":"conf/model.json", - "logDir":"logs/master", - "StreamWaitingTime":240, - "logPrintInterval":60 - } -} diff --git a/AIlib/conf/model.json b/AIlib/conf/model.json deleted file mode 100644 index b35e451..0000000 --- a/AIlib/conf/model.json +++ /dev/null @@ -1,17 +0,0 @@ -{ - - "gpu_process":{"det_weights":"weights/yolov5/class5/best_5classes.pt","seg_nclass":2,"seg_weights": "weights/BiSeNet/checkpoint.pth" }, - - "post_process":{ "name":"post_process","conf_thres":0.25,"iou_thres":0.45,"classes":5,"labelnames":"weights/yolov5/class5/labelnames.json","fpsample":240,"debug":false , "rainbows":[ [0,0,255],[0,255,0],[255,0,0],[255,0,255],[255,255,0],[255,129,0],[255,0,127],[127,255,0],[0,255,127],[0,127,255],[127,0,255],[255,127,255],[255,255,127],[127,255,255],[0,255,255],[255,127,255],[127,255,255], [0,127,0],[0,0,127],[0,255,255]],"outImaDir":"problems/images_tmp","outVideoDir":"problems/videos_save" }, - - "push_process":{ "OutVideoW":1920, "OutVideoH":1080 }, - "AI_video_save": {"onLine":false,"offLine":true }, - "imageTxtFile":true, - "logChildProcessOffline":"logs/logChildProcess/offline", - "logChildProcessOnline":"logs/logChildProcess/online", - "TaskStatusQueryUrl":"http://192.168.11.241:1011/api/web/serviceInst", - "StreamWaitingTime":240, - "StreamRecoveringTime":600 - - -} diff --git a/AIlib/conf/platech.ttf b/AIlib/conf/platech.ttf deleted file mode 100644 index d66a970..0000000 Binary files a/AIlib/conf/platech.ttf and /dev/null differ diff --git a/AIlib/conf/send_oss.json b/AIlib/conf/send_oss.json deleted file mode 100644 index 2600528..0000000 --- a/AIlib/conf/send_oss.json +++ /dev/null @@ -1,20 +0,0 @@ -{ - "indir":"problems/images_tmp", - "outdir":"problems/images_save", - "jsonDir" : "mintors/kafka/", - "hearBeatTimeMs":30, - "logdir":"logs/send", - "videoBakDir":"problems/videos_save", - "ossPar":{"Epoint":"http://oss-cn-shanghai.aliyuncs.com", - "AId":"LTAI5tSJ62TLMUb4SZuf285A", - "ASt":"MWYynm30filZ7x0HqSHlU3pdLVNeI7", - "bucketName":"ta-tech-image" - }, - "vodPar":{ - "AId":"LTAI5tE7KWN9fsuGU7DyfYF4", - "ASt":"yPPCyfsqWgrTuoz5H4sisY0COclx8E" - }, - "kafkaPar":{"boostServer1":["192.168.11.242:9092"] ,"boostServer2":["101.132.127.1:19092"], "boostServer":["212.129.223.66:19092"] ,"topic":"dsp-alg-task-results"}, - "labelnamesFile":"weights/yolov5/class5/labelnames.json" - -} diff --git a/AIlib/models/__init__.py b/AIlib/models/__init__.py deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/models/__pycache__/__init__.cpython-37.pyc b/AIlib/models/__pycache__/__init__.cpython-37.pyc deleted file mode 100644 index 2d10de3..0000000 Binary files a/AIlib/models/__pycache__/__init__.cpython-37.pyc and /dev/null differ diff --git a/AIlib/models/__pycache__/__init__.cpython-38.pyc b/AIlib/models/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index d34bc01..0000000 Binary files a/AIlib/models/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/models/__pycache__/common.cpython-37.pyc b/AIlib/models/__pycache__/common.cpython-37.pyc deleted file mode 100644 index 8a36cab..0000000 Binary files a/AIlib/models/__pycache__/common.cpython-37.pyc and /dev/null differ diff --git a/AIlib/models/__pycache__/common.cpython-38.pyc b/AIlib/models/__pycache__/common.cpython-38.pyc deleted file mode 100644 index a166853..0000000 Binary files a/AIlib/models/__pycache__/common.cpython-38.pyc and /dev/null differ diff --git a/AIlib/models/__pycache__/experimental.cpython-37.pyc b/AIlib/models/__pycache__/experimental.cpython-37.pyc deleted file mode 100644 index 75ca459..0000000 Binary files a/AIlib/models/__pycache__/experimental.cpython-37.pyc and /dev/null differ diff --git a/AIlib/models/__pycache__/experimental.cpython-38.pyc b/AIlib/models/__pycache__/experimental.cpython-38.pyc deleted file mode 100644 index 5b62e41..0000000 Binary files a/AIlib/models/__pycache__/experimental.cpython-38.pyc and /dev/null differ diff --git a/AIlib/models/__pycache__/yolo.cpython-38.pyc b/AIlib/models/__pycache__/yolo.cpython-38.pyc deleted file mode 100644 index ddd93b4..0000000 Binary files a/AIlib/models/__pycache__/yolo.cpython-38.pyc and /dev/null differ diff --git a/AIlib/models/common.py b/AIlib/models/common.py deleted file mode 100644 index 028dedd..0000000 --- a/AIlib/models/common.py +++ /dev/null @@ -1,405 +0,0 @@ -# YOLOv5 common modules - -import math -from copy import copy -from pathlib import Path - -import numpy as np -import pandas as pd -import requests -import torch -import torch.nn as nn -from PIL import Image -from torch.cuda import amp - -from utils.datasets import letterbox -from utils.general import non_max_suppression, make_divisible, scale_coords, increment_path, xyxy2xywh -from utils.plots import color_list, plot_one_box -from utils.torch_utils import time_synchronized - -import warnings - -class SPPF(nn.Module): - # Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv5 by Glenn Jocher - def __init__(self, c1, c2, k=5): # equivalent to SPP(k=(5, 9, 13)) - super().__init__() - c_ = c1 // 2 # hidden channels - self.cv1 = Conv(c1, c_, 1, 1) - self.cv2 = Conv(c_ * 4, c2, 1, 1) - self.m = nn.MaxPool2d(kernel_size=k, stride=1, padding=k // 2) - - def forward(self, x): - x = self.cv1(x) - with warnings.catch_warnings(): - warnings.simplefilter('ignore') # suppress torch 1.9.0 max_pool2d() warning - y1 = self.m(x) - y2 = self.m(y1) - return self.cv2(torch.cat([x, y1, y2, self.m(y2)], 1)) - - -def autopad(k, p=None): # kernel, padding - # Pad to 'same' - if p is None: - p = k // 2 if isinstance(k, int) else [x // 2 for x in k] # auto-pad - return p - - -def DWConv(c1, c2, k=1, s=1, act=True): - # Depthwise convolution - return Conv(c1, c2, k, s, g=math.gcd(c1, c2), act=act) - - -class Conv(nn.Module): - # Standard convolution - def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups - super(Conv, self).__init__() - self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False) - self.bn = nn.BatchNorm2d(c2) - self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity()) - - def forward(self, x): - return self.act(self.bn(self.conv(x))) - - def fuseforward(self, x): - return self.act(self.conv(x)) - - -class TransformerLayer(nn.Module): - # Transformer layer https://arxiv.org/abs/2010.11929 (LayerNorm layers removed for better performance) - def __init__(self, c, num_heads): - super().__init__() - self.q = nn.Linear(c, c, bias=False) - self.k = nn.Linear(c, c, bias=False) - self.v = nn.Linear(c, c, bias=False) - self.ma = nn.MultiheadAttention(embed_dim=c, num_heads=num_heads) - self.fc1 = nn.Linear(c, c, bias=False) - self.fc2 = nn.Linear(c, c, bias=False) - - def forward(self, x): - x = self.ma(self.q(x), self.k(x), self.v(x))[0] + x - x = self.fc2(self.fc1(x)) + x - return x - - -class TransformerBlock(nn.Module): - # Vision Transformer https://arxiv.org/abs/2010.11929 - def __init__(self, c1, c2, num_heads, num_layers): - super().__init__() - self.conv = None - if c1 != c2: - self.conv = Conv(c1, c2) - self.linear = nn.Linear(c2, c2) # learnable position embedding - self.tr = nn.Sequential(*[TransformerLayer(c2, num_heads) for _ in range(num_layers)]) - self.c2 = c2 - - def forward(self, x): - if self.conv is not None: - x = self.conv(x) - b, _, w, h = x.shape - p = x.flatten(2) - p = p.unsqueeze(0) - p = p.transpose(0, 3) - p = p.squeeze(3) - e = self.linear(p) - x = p + e - - x = self.tr(x) - x = x.unsqueeze(3) - x = x.transpose(0, 3) - x = x.reshape(b, self.c2, w, h) - return x - - -class Bottleneck(nn.Module): - # Standard bottleneck - def __init__(self, c1, c2, shortcut=True, g=1, e=0.5): # ch_in, ch_out, shortcut, groups, expansion - super(Bottleneck, self).__init__() - c_ = int(c2 * e) # hidden channels - self.cv1 = Conv(c1, c_, 1, 1) - self.cv2 = Conv(c_, c2, 3, 1, g=g) - self.add = shortcut and c1 == c2 - - def forward(self, x): - return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x)) - - -class BottleneckCSP(nn.Module): - # CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks - def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion - super(BottleneckCSP, self).__init__() - c_ = int(c2 * e) # hidden channels - self.cv1 = Conv(c1, c_, 1, 1) - self.cv2 = nn.Conv2d(c1, c_, 1, 1, bias=False) - self.cv3 = nn.Conv2d(c_, c_, 1, 1, bias=False) - self.cv4 = Conv(2 * c_, c2, 1, 1) - self.bn = nn.BatchNorm2d(2 * c_) # applied to cat(cv2, cv3) - self.act = nn.LeakyReLU(0.1, inplace=True) - self.m = nn.Sequential(*[Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)]) - - def forward(self, x): - y1 = self.cv3(self.m(self.cv1(x))) - y2 = self.cv2(x) - return self.cv4(self.act(self.bn(torch.cat((y1, y2), dim=1)))) - - -class C3(nn.Module): - # CSP Bottleneck with 3 convolutions - def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion - super(C3, self).__init__() - c_ = int(c2 * e) # hidden channels - self.cv1 = Conv(c1, c_, 1, 1) - self.cv2 = Conv(c1, c_, 1, 1) - self.cv3 = Conv(2 * c_, c2, 1) # act=FReLU(c2) - self.m = nn.Sequential(*[Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)]) - # self.m = nn.Sequential(*[CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n)]) - - def forward(self, x): - return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), dim=1)) - - -class C3TR(C3): - # C3 module with TransformerBlock() - def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): - super().__init__(c1, c2, n, shortcut, g, e) - c_ = int(c2 * e) - self.m = TransformerBlock(c_, c_, 4, n) - - -class SPP(nn.Module): - # Spatial pyramid pooling layer used in YOLOv3-SPP - def __init__(self, c1, c2, k=(5, 9, 13)): - super(SPP, self).__init__() - c_ = c1 // 2 # hidden channels - self.cv1 = Conv(c1, c_, 1, 1) - self.cv2 = Conv(c_ * (len(k) + 1), c2, 1, 1) - self.m = nn.ModuleList([nn.MaxPool2d(kernel_size=x, stride=1, padding=x // 2) for x in k]) - - def forward(self, x): - x = self.cv1(x) - return self.cv2(torch.cat([x] + [m(x) for m in self.m], 1)) - - -class Focus(nn.Module): - # Focus wh information into c-space - def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups - super(Focus, self).__init__() - self.conv = Conv(c1 * 4, c2, k, s, p, g, act) - # self.contract = Contract(gain=2) - - def forward(self, x): # x(b,c,w,h) -> y(b,4c,w/2,h/2) - return self.conv(torch.cat([x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]], 1)) - # return self.conv(self.contract(x)) - - -class Contract(nn.Module): - # Contract width-height into channels, i.e. x(1,64,80,80) to x(1,256,40,40) - def __init__(self, gain=2): - super().__init__() - self.gain = gain - - def forward(self, x): - N, C, H, W = x.size() # assert (H / s == 0) and (W / s == 0), 'Indivisible gain' - s = self.gain - x = x.view(N, C, H // s, s, W // s, s) # x(1,64,40,2,40,2) - x = x.permute(0, 3, 5, 1, 2, 4).contiguous() # x(1,2,2,64,40,40) - return x.view(N, C * s * s, H // s, W // s) # x(1,256,40,40) - - -class Expand(nn.Module): - # Expand channels into width-height, i.e. x(1,64,80,80) to x(1,16,160,160) - def __init__(self, gain=2): - super().__init__() - self.gain = gain - - def forward(self, x): - N, C, H, W = x.size() # assert C / s ** 2 == 0, 'Indivisible gain' - s = self.gain - x = x.view(N, s, s, C // s ** 2, H, W) # x(1,2,2,16,80,80) - x = x.permute(0, 3, 4, 1, 5, 2).contiguous() # x(1,16,80,2,80,2) - return x.view(N, C // s ** 2, H * s, W * s) # x(1,16,160,160) - - -class Concat(nn.Module): - # Concatenate a list of tensors along dimension - def __init__(self, dimension=1): - super(Concat, self).__init__() - self.d = dimension - - def forward(self, x): - return torch.cat(x, self.d) - - -class NMS(nn.Module): - # Non-Maximum Suppression (NMS) module - conf = 0.25 # confidence threshold - iou = 0.45 # IoU threshold - classes = None # (optional list) filter by class - - def __init__(self): - super(NMS, self).__init__() - - def forward(self, x): - return non_max_suppression(x[0], conf_thres=self.conf, iou_thres=self.iou, classes=self.classes) - - -class autoShape(nn.Module): - # input-robust model wrapper for passing cv2/np/PIL/torch inputs. Includes preprocessing, inference and NMS - conf = 0.25 # NMS confidence threshold - iou = 0.45 # NMS IoU threshold - classes = None # (optional list) filter by class - - def __init__(self, model): - super(autoShape, self).__init__() - self.model = model.eval() - - def autoshape(self): - print('autoShape already enabled, skipping... ') # model already converted to model.autoshape() - return self - - @torch.no_grad() - def forward(self, imgs, size=640, augment=False, profile=False): - # Inference from various sources. For height=640, width=1280, RGB images example inputs are: - # filename: imgs = 'data/images/zidane.jpg' - # URI: = 'https://github.com/ultralytics/yolov5/releases/download/v1.0/zidane.jpg' - # OpenCV: = cv2.imread('image.jpg')[:,:,::-1] # HWC BGR to RGB x(640,1280,3) - # PIL: = Image.open('image.jpg') # HWC x(640,1280,3) - # numpy: = np.zeros((640,1280,3)) # HWC - # torch: = torch.zeros(16,3,320,640) # BCHW (scaled to size=640, 0-1 values) - # multiple: = [Image.open('image1.jpg'), Image.open('image2.jpg'), ...] # list of images - - t = [time_synchronized()] - p = next(self.model.parameters()) # for device and type - if isinstance(imgs, torch.Tensor): # torch - with amp.autocast(enabled=p.device.type != 'cpu'): - return self.model(imgs.to(p.device).type_as(p), augment, profile) # inference - - # Pre-process - n, imgs = (len(imgs), imgs) if isinstance(imgs, list) else (1, [imgs]) # number of images, list of images - shape0, shape1, files = [], [], [] # image and inference shapes, filenames - for i, im in enumerate(imgs): - f = f'image{i}' # filename - if isinstance(im, str): # filename or uri - im, f = np.asarray(Image.open(requests.get(im, stream=True).raw if im.startswith('http') else im)), im - elif isinstance(im, Image.Image): # PIL Image - im, f = np.asarray(im), getattr(im, 'filename', f) or f - files.append(Path(f).with_suffix('.jpg').name) - if im.shape[0] < 5: # image in CHW - im = im.transpose((1, 2, 0)) # reverse dataloader .transpose(2, 0, 1) - im = im[:, :, :3] if im.ndim == 3 else np.tile(im[:, :, None], 3) # enforce 3ch input - s = im.shape[:2] # HWC - shape0.append(s) # image shape - g = (size / max(s)) # gain - shape1.append([y * g for y in s]) - imgs[i] = im if im.data.contiguous else np.ascontiguousarray(im) # update - shape1 = [make_divisible(x, int(self.stride.max())) for x in np.stack(shape1, 0).max(0)] # inference shape - x = [letterbox(im, new_shape=shape1, auto=False)[0] for im in imgs] # pad - x = np.stack(x, 0) if n > 1 else x[0][None] # stack - x = np.ascontiguousarray(x.transpose((0, 3, 1, 2))) # BHWC to BCHW - x = torch.from_numpy(x).to(p.device).type_as(p) / 255. # uint8 to fp16/32 - t.append(time_synchronized()) - - with amp.autocast(enabled=p.device.type != 'cpu'): - # Inference - y = self.model(x, augment, profile)[0] # forward - t.append(time_synchronized()) - - # Post-process - y = non_max_suppression(y, conf_thres=self.conf, iou_thres=self.iou, classes=self.classes) # NMS - for i in range(n): - scale_coords(shape1, y[i][:, :4], shape0[i]) - - t.append(time_synchronized()) - return Detections(imgs, y, files, t, self.names, x.shape) - - -class Detections: - # detections class for YOLOv5 inference results - def __init__(self, imgs, pred, files, times=None, names=None, shape=None): - super(Detections, self).__init__() - d = pred[0].device # device - gn = [torch.tensor([*[im.shape[i] for i in [1, 0, 1, 0]], 1., 1.], device=d) for im in imgs] # normalizations - self.imgs = imgs # list of images as numpy arrays - self.pred = pred # list of tensors pred[0] = (xyxy, conf, cls) - self.names = names # class names - self.files = files # image filenames - self.xyxy = pred # xyxy pixels - self.xywh = [xyxy2xywh(x) for x in pred] # xywh pixels - self.xyxyn = [x / g for x, g in zip(self.xyxy, gn)] # xyxy normalized - self.xywhn = [x / g for x, g in zip(self.xywh, gn)] # xywh normalized - self.n = len(self.pred) # number of images (batch size) - self.t = tuple((times[i + 1] - times[i]) * 1000 / self.n for i in range(3)) # timestamps (ms) - self.s = shape # inference BCHW shape - - def display(self, pprint=False, show=False, save=False, render=False, save_dir=''): - colors = color_list() - for i, (img, pred) in enumerate(zip(self.imgs, self.pred)): - str = f'image {i + 1}/{len(self.pred)}: {img.shape[0]}x{img.shape[1]} ' - if pred is not None: - for c in pred[:, -1].unique(): - n = (pred[:, -1] == c).sum() # detections per class - str += f"{n} {self.names[int(c)]}{'s' * (n > 1)}, " # add to string - if show or save or render: - for *box, conf, cls in pred: # xyxy, confidence, class - label = f'{self.names[int(cls)]} {conf:.2f}' - plot_one_box(box, img, label=label, color=colors[int(cls) % 10]) - img = Image.fromarray(img.astype(np.uint8)) if isinstance(img, np.ndarray) else img # from np - if pprint: - print(str.rstrip(', ')) - if show: - img.show(self.files[i]) # show - if save: - f = self.files[i] - img.save(Path(save_dir) / f) # save - print(f"{'Saved' * (i == 0)} {f}", end=',' if i < self.n - 1 else f' to {save_dir}\n') - if render: - self.imgs[i] = np.asarray(img) - - def print(self): - self.display(pprint=True) # print results - print(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {tuple(self.s)}' % self.t) - - def show(self): - self.display(show=True) # show results - - def save(self, save_dir='runs/hub/exp'): - save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/hub/exp') # increment save_dir - Path(save_dir).mkdir(parents=True, exist_ok=True) - self.display(save=True, save_dir=save_dir) # save results - - def render(self): - self.display(render=True) # render results - return self.imgs - - def pandas(self): - # return detections as pandas DataFrames, i.e. print(results.pandas().xyxy[0]) - new = copy(self) # return copy - ca = 'xmin', 'ymin', 'xmax', 'ymax', 'confidence', 'class', 'name' # xyxy columns - cb = 'xcenter', 'ycenter', 'width', 'height', 'confidence', 'class', 'name' # xywh columns - for k, c in zip(['xyxy', 'xyxyn', 'xywh', 'xywhn'], [ca, ca, cb, cb]): - a = [[x[:5] + [int(x[5]), self.names[int(x[5])]] for x in x.tolist()] for x in getattr(self, k)] # update - setattr(new, k, [pd.DataFrame(x, columns=c) for x in a]) - return new - - def tolist(self): - # return a list of Detections objects, i.e. 'for result in results.tolist():' - x = [Detections([self.imgs[i]], [self.pred[i]], self.names, self.s) for i in range(self.n)] - for d in x: - for k in ['imgs', 'pred', 'xyxy', 'xyxyn', 'xywh', 'xywhn']: - setattr(d, k, getattr(d, k)[0]) # pop out of list - return x - - def __len__(self): - return self.n - - -class Classify(nn.Module): - # Classification head, i.e. x(b,c1,20,20) to x(b,c2) - def __init__(self, c1, c2, k=1, s=1, p=None, g=1): # ch_in, ch_out, kernel, stride, padding, groups - super(Classify, self).__init__() - self.aap = nn.AdaptiveAvgPool2d(1) # to x(b,c1,1,1) - self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g) # to x(b,c2,1,1) - self.flat = nn.Flatten() - - def forward(self, x): - z = torch.cat([self.aap(y) for y in (x if isinstance(x, list) else [x])], 1) # cat if list - return self.flat(self.conv(z)) # flatten to x(b,c2) diff --git a/AIlib/models/experimental.py b/AIlib/models/experimental.py deleted file mode 100644 index 548353c..0000000 --- a/AIlib/models/experimental.py +++ /dev/null @@ -1,134 +0,0 @@ -# YOLOv5 experimental modules - -import numpy as np -import torch -import torch.nn as nn - -from models.common import Conv, DWConv -from utils.google_utils import attempt_download - - -class CrossConv(nn.Module): - # Cross Convolution Downsample - def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False): - # ch_in, ch_out, kernel, stride, groups, expansion, shortcut - super(CrossConv, self).__init__() - c_ = int(c2 * e) # hidden channels - self.cv1 = Conv(c1, c_, (1, k), (1, s)) - self.cv2 = Conv(c_, c2, (k, 1), (s, 1), g=g) - self.add = shortcut and c1 == c2 - - def forward(self, x): - return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x)) - - -class Sum(nn.Module): - # Weighted sum of 2 or more layers https://arxiv.org/abs/1911.09070 - def __init__(self, n, weight=False): # n: number of inputs - super(Sum, self).__init__() - self.weight = weight # apply weights boolean - self.iter = range(n - 1) # iter object - if weight: - self.w = nn.Parameter(-torch.arange(1., n) / 2, requires_grad=True) # layer weights - - def forward(self, x): - y = x[0] # no weight - if self.weight: - w = torch.sigmoid(self.w) * 2 - for i in self.iter: - y = y + x[i + 1] * w[i] - else: - for i in self.iter: - y = y + x[i + 1] - return y - - -class GhostConv(nn.Module): - # Ghost Convolution https://github.com/huawei-noah/ghostnet - def __init__(self, c1, c2, k=1, s=1, g=1, act=True): # ch_in, ch_out, kernel, stride, groups - super(GhostConv, self).__init__() - c_ = c2 // 2 # hidden channels - self.cv1 = Conv(c1, c_, k, s, None, g, act) - self.cv2 = Conv(c_, c_, 5, 1, None, c_, act) - - def forward(self, x): - y = self.cv1(x) - return torch.cat([y, self.cv2(y)], 1) - - -class GhostBottleneck(nn.Module): - # Ghost Bottleneck https://github.com/huawei-noah/ghostnet - def __init__(self, c1, c2, k=3, s=1): # ch_in, ch_out, kernel, stride - super(GhostBottleneck, self).__init__() - c_ = c2 // 2 - self.conv = nn.Sequential(GhostConv(c1, c_, 1, 1), # pw - DWConv(c_, c_, k, s, act=False) if s == 2 else nn.Identity(), # dw - GhostConv(c_, c2, 1, 1, act=False)) # pw-linear - self.shortcut = nn.Sequential(DWConv(c1, c1, k, s, act=False), - Conv(c1, c2, 1, 1, act=False)) if s == 2 else nn.Identity() - - def forward(self, x): - return self.conv(x) + self.shortcut(x) - - -class MixConv2d(nn.Module): - # Mixed Depthwise Conv https://arxiv.org/abs/1907.09595 - def __init__(self, c1, c2, k=(1, 3), s=1, equal_ch=True): - super(MixConv2d, self).__init__() - groups = len(k) - if equal_ch: # equal c_ per group - i = torch.linspace(0, groups - 1E-6, c2).floor() # c2 indices - c_ = [(i == g).sum() for g in range(groups)] # intermediate channels - else: # equal weight.numel() per group - b = [c2] + [0] * groups - a = np.eye(groups + 1, groups, k=-1) - a -= np.roll(a, 1, axis=1) - a *= np.array(k) ** 2 - a[0] = 1 - c_ = np.linalg.lstsq(a, b, rcond=None)[0].round() # solve for equal weight indices, ax = b - - self.m = nn.ModuleList([nn.Conv2d(c1, int(c_[g]), k[g], s, k[g] // 2, bias=False) for g in range(groups)]) - self.bn = nn.BatchNorm2d(c2) - self.act = nn.LeakyReLU(0.1, inplace=True) - - def forward(self, x): - return x + self.act(self.bn(torch.cat([m(x) for m in self.m], 1))) - - -class Ensemble(nn.ModuleList): - # Ensemble of models - def __init__(self): - super(Ensemble, self).__init__() - - def forward(self, x, augment=False): - y = [] - for module in self: - y.append(module(x, augment)[0]) - # y = torch.stack(y).max(0)[0] # max ensemble - # y = torch.stack(y).mean(0) # mean ensemble - y = torch.cat(y, 1) # nms ensemble - return y, None # inference, train output - - -def attempt_load(weights, map_location=None): - # Loads an ensemble of models weights=[a,b,c] or a single model weights=[a] or weights=a - model = Ensemble() - for w in weights if isinstance(weights, list) else [weights]: - attempt_download(w) - ckpt = torch.load(w, map_location=map_location) # load - model.append(ckpt['ema' if ckpt.get('ema') else 'model'].float().fuse().eval()) # FP32 model - - # Compatibility updates - for m in model.modules(): - if type(m) in [nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU]: - m.inplace = True # pytorch 1.7.0 compatibility - elif type(m) is Conv: - m._non_persistent_buffers_set = set() # pytorch 1.6.0 compatibility - - if len(model) == 1: - return model[-1] # return model - else: - print('Ensemble created with %s\n' % weights) - for k in ['names', 'stride']: - setattr(model, k, getattr(model[-1], k)) - return model # return ensemble diff --git a/AIlib/models/export.py b/AIlib/models/export.py deleted file mode 100644 index c527a47..0000000 --- a/AIlib/models/export.py +++ /dev/null @@ -1,123 +0,0 @@ -"""Exports a YOLOv5 *.pt model to ONNX and TorchScript formats - -Usage: - $ export PYTHONPATH="$PWD" && python models/export.py --weights yolov5s.pt --img 640 --batch 1 -""" - -import argparse -import sys -import time - -sys.path.append('./') # to run '$ python *.py' files in subdirectories - -import torch -import torch.nn as nn - -import models -from models.experimental import attempt_load -from utils.activations import Hardswish, SiLU -from utils.general import colorstr, check_img_size, check_requirements, set_logging -from utils.torch_utils import select_device - -if __name__ == '__main__': - parser = argparse.ArgumentParser() - parser.add_argument('--weights', type=str, default='./yolov5s.pt', help='weights path') - parser.add_argument('--img-size', nargs='+', type=int, default=[640, 640], help='image size') # height, width - parser.add_argument('--batch-size', type=int, default=1, help='batch size') - parser.add_argument('--grid', action='store_true', help='export Detect() layer grid') - parser.add_argument('--device', default='cpu', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') - parser.add_argument('--dynamic', action='store_true', help='dynamic ONNX axes') # ONNX-only - parser.add_argument('--simplify', action='store_true', help='simplify ONNX model') # ONNX-only - opt = parser.parse_args() - opt.img_size *= 2 if len(opt.img_size) == 1 else 1 # expand - print(opt) - set_logging() - t = time.time() - - # Load PyTorch model - device = select_device(opt.device) - model = attempt_load(opt.weights, map_location=device) # load FP32 model - labels = model.names - - # Checks - gs = int(max(model.stride)) # grid size (max stride) - opt.img_size = [check_img_size(x, gs) for x in opt.img_size] # verify img_size are gs-multiples - - # Input - img = torch.zeros(opt.batch_size, 3, *opt.img_size).to(device) # image size(1,3,320,192) iDetection - - # Update model - for k, m in model.named_modules(): - m._non_persistent_buffers_set = set() # pytorch 1.6.0 compatibility - if isinstance(m, models.common.Conv): # assign export-friendly activations - if isinstance(m.act, nn.Hardswish): - m.act = Hardswish() - elif isinstance(m.act, nn.SiLU): - m.act = SiLU() - # elif isinstance(m, models.yolo.Detect): - # m.forward = m.forward_export # assign forward (optional) - model.model[-1].export = not opt.grid # set Detect() layer grid export - y = model(img) # dry run - - # TorchScript export ----------------------------------------------------------------------------------------------- - prefix = colorstr('TorchScript:') - try: - print(f'\n{prefix} starting export with torch {torch.__version__}...') - f = opt.weights.replace('.pt', '.torchscript.pt') # filename - ts = torch.jit.trace(model, img, strict=False) - ts.save(f) - print(f'{prefix} export success, saved as {f}') - except Exception as e: - print(f'{prefix} export failure: {e}') - - # ONNX export ------------------------------------------------------------------------------------------------------ - prefix = colorstr('ONNX:') - try: - import onnx - - print(f'{prefix} starting export with onnx {onnx.__version__}...') - f = opt.weights.replace('.pt', '.onnx') # filename - torch.onnx.export(model, img, f, verbose=False, opset_version=12, input_names=['images'], - output_names=['classes', 'boxes'] if y is None else ['output'], - dynamic_axes={'images': {0: 'batch', 2: 'height', 3: 'width'}, # size(1,3,640,640) - 'output': {0: 'batch', 2: 'y', 3: 'x'}} if opt.dynamic else None) - - # Checks - model_onnx = onnx.load(f) # load onnx model - onnx.checker.check_model(model_onnx) # check onnx model - # print(onnx.helper.printable_graph(model_onnx.graph)) # print - - # Simplify - if opt.simplify: - try: - check_requirements(['onnx-simplifier']) - import onnxsim - - print(f'{prefix} simplifying with onnx-simplifier {onnxsim.__version__}...') - model_onnx, check = onnxsim.simplify(model_onnx, - dynamic_input_shape=opt.dynamic, - input_shapes={'images': list(img.shape)} if opt.dynamic else None) - assert check, 'assert check failed' - onnx.save(model_onnx, f) - except Exception as e: - print(f'{prefix} simplifier failure: {e}') - print(f'{prefix} export success, saved as {f}') - except Exception as e: - print(f'{prefix} export failure: {e}') - - # CoreML export ---------------------------------------------------------------------------------------------------- - prefix = colorstr('CoreML:') - try: - import coremltools as ct - - print(f'{prefix} starting export with coremltools {onnx.__version__}...') - # convert model from torchscript and apply pixel scaling as per detect.py - model = ct.convert(ts, inputs=[ct.ImageType(name='image', shape=img.shape, scale=1 / 255.0, bias=[0, 0, 0])]) - f = opt.weights.replace('.pt', '.mlmodel') # filename - model.save(f) - print(f'{prefix} export success, saved as {f}') - except Exception as e: - print(f'{prefix} export failure: {e}') - - # Finish - print(f'\nExport complete ({time.time() - t:.2f}s). Visualize with https://github.com/lutzroeder/netron.') diff --git a/AIlib/models/hub/anchors.yaml b/AIlib/models/hub/anchors.yaml deleted file mode 100644 index a07a4dc..0000000 --- a/AIlib/models/hub/anchors.yaml +++ /dev/null @@ -1,58 +0,0 @@ -# Default YOLOv5 anchors for COCO data - - -# P5 ------------------------------------------------------------------------------------------------------------------- -# P5-640: -anchors_p5_640: - - [ 10,13, 16,30, 33,23 ] # P3/8 - - [ 30,61, 62,45, 59,119 ] # P4/16 - - [ 116,90, 156,198, 373,326 ] # P5/32 - - -# P6 ------------------------------------------------------------------------------------------------------------------- -# P6-640: thr=0.25: 0.9964 BPR, 5.54 anchors past thr, n=12, img_size=640, metric_all=0.281/0.716-mean/best, past_thr=0.469-mean: 9,11, 21,19, 17,41, 43,32, 39,70, 86,64, 65,131, 134,130, 120,265, 282,180, 247,354, 512,387 -anchors_p6_640: - - [ 9,11, 21,19, 17,41 ] # P3/8 - - [ 43,32, 39,70, 86,64 ] # P4/16 - - [ 65,131, 134,130, 120,265 ] # P5/32 - - [ 282,180, 247,354, 512,387 ] # P6/64 - -# P6-1280: thr=0.25: 0.9950 BPR, 5.55 anchors past thr, n=12, img_size=1280, metric_all=0.281/0.714-mean/best, past_thr=0.468-mean: 19,27, 44,40, 38,94, 96,68, 86,152, 180,137, 140,301, 303,264, 238,542, 436,615, 739,380, 925,792 -anchors_p6_1280: - - [ 19,27, 44,40, 38,94 ] # P3/8 - - [ 96,68, 86,152, 180,137 ] # P4/16 - - [ 140,301, 303,264, 238,542 ] # P5/32 - - [ 436,615, 739,380, 925,792 ] # P6/64 - -# P6-1920: thr=0.25: 0.9950 BPR, 5.55 anchors past thr, n=12, img_size=1920, metric_all=0.281/0.714-mean/best, past_thr=0.468-mean: 28,41, 67,59, 57,141, 144,103, 129,227, 270,205, 209,452, 455,396, 358,812, 653,922, 1109,570, 1387,1187 -anchors_p6_1920: - - [ 28,41, 67,59, 57,141 ] # P3/8 - - [ 144,103, 129,227, 270,205 ] # P4/16 - - [ 209,452, 455,396, 358,812 ] # P5/32 - - [ 653,922, 1109,570, 1387,1187 ] # P6/64 - - -# P7 ------------------------------------------------------------------------------------------------------------------- -# P7-640: thr=0.25: 0.9962 BPR, 6.76 anchors past thr, n=15, img_size=640, metric_all=0.275/0.733-mean/best, past_thr=0.466-mean: 11,11, 13,30, 29,20, 30,46, 61,38, 39,92, 78,80, 146,66, 79,163, 149,150, 321,143, 157,303, 257,402, 359,290, 524,372 -anchors_p7_640: - - [ 11,11, 13,30, 29,20 ] # P3/8 - - [ 30,46, 61,38, 39,92 ] # P4/16 - - [ 78,80, 146,66, 79,163 ] # P5/32 - - [ 149,150, 321,143, 157,303 ] # P6/64 - - [ 257,402, 359,290, 524,372 ] # P7/128 - -# P7-1280: thr=0.25: 0.9968 BPR, 6.71 anchors past thr, n=15, img_size=1280, metric_all=0.273/0.732-mean/best, past_thr=0.463-mean: 19,22, 54,36, 32,77, 70,83, 138,71, 75,173, 165,159, 148,334, 375,151, 334,317, 251,626, 499,474, 750,326, 534,814, 1079,818 -anchors_p7_1280: - - [ 19,22, 54,36, 32,77 ] # P3/8 - - [ 70,83, 138,71, 75,173 ] # P4/16 - - [ 165,159, 148,334, 375,151 ] # P5/32 - - [ 334,317, 251,626, 499,474 ] # P6/64 - - [ 750,326, 534,814, 1079,818 ] # P7/128 - -# P7-1920: thr=0.25: 0.9968 BPR, 6.71 anchors past thr, n=15, img_size=1920, metric_all=0.273/0.732-mean/best, past_thr=0.463-mean: 29,34, 81,55, 47,115, 105,124, 207,107, 113,259, 247,238, 222,500, 563,227, 501,476, 376,939, 749,711, 1126,489, 801,1222, 1618,1227 -anchors_p7_1920: - - [ 29,34, 81,55, 47,115 ] # P3/8 - - [ 105,124, 207,107, 113,259 ] # P4/16 - - [ 247,238, 222,500, 563,227 ] # P5/32 - - [ 501,476, 376,939, 749,711 ] # P6/64 - - [ 1126,489, 801,1222, 1618,1227 ] # P7/128 diff --git a/AIlib/models/hub/yolov3-spp.yaml b/AIlib/models/hub/yolov3-spp.yaml deleted file mode 100644 index 38dcc44..0000000 --- a/AIlib/models/hub/yolov3-spp.yaml +++ /dev/null @@ -1,51 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: - - [10,13, 16,30, 33,23] # P3/8 - - [30,61, 62,45, 59,119] # P4/16 - - [116,90, 156,198, 373,326] # P5/32 - -# darknet53 backbone -backbone: - # [from, number, module, args] - [[-1, 1, Conv, [32, 3, 1]], # 0 - [-1, 1, Conv, [64, 3, 2]], # 1-P1/2 - [-1, 1, Bottleneck, [64]], - [-1, 1, Conv, [128, 3, 2]], # 3-P2/4 - [-1, 2, Bottleneck, [128]], - [-1, 1, Conv, [256, 3, 2]], # 5-P3/8 - [-1, 8, Bottleneck, [256]], - [-1, 1, Conv, [512, 3, 2]], # 7-P4/16 - [-1, 8, Bottleneck, [512]], - [-1, 1, Conv, [1024, 3, 2]], # 9-P5/32 - [-1, 4, Bottleneck, [1024]], # 10 - ] - -# YOLOv3-SPP head -head: - [[-1, 1, Bottleneck, [1024, False]], - [-1, 1, SPP, [512, [5, 9, 13]]], - [-1, 1, Conv, [1024, 3, 1]], - [-1, 1, Conv, [512, 1, 1]], - [-1, 1, Conv, [1024, 3, 1]], # 15 (P5/32-large) - - [-2, 1, Conv, [256, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 8], 1, Concat, [1]], # cat backbone P4 - [-1, 1, Bottleneck, [512, False]], - [-1, 1, Bottleneck, [512, False]], - [-1, 1, Conv, [256, 1, 1]], - [-1, 1, Conv, [512, 3, 1]], # 22 (P4/16-medium) - - [-2, 1, Conv, [128, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 6], 1, Concat, [1]], # cat backbone P3 - [-1, 1, Bottleneck, [256, False]], - [-1, 2, Bottleneck, [256, False]], # 27 (P3/8-small) - - [[27, 22, 15], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) - ] diff --git a/AIlib/models/hub/yolov3-tiny.yaml b/AIlib/models/hub/yolov3-tiny.yaml deleted file mode 100644 index ff7638c..0000000 --- a/AIlib/models/hub/yolov3-tiny.yaml +++ /dev/null @@ -1,41 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: - - [10,14, 23,27, 37,58] # P4/16 - - [81,82, 135,169, 344,319] # P5/32 - -# YOLOv3-tiny backbone -backbone: - # [from, number, module, args] - [[-1, 1, Conv, [16, 3, 1]], # 0 - [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 1-P1/2 - [-1, 1, Conv, [32, 3, 1]], - [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 3-P2/4 - [-1, 1, Conv, [64, 3, 1]], - [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 5-P3/8 - [-1, 1, Conv, [128, 3, 1]], - [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 7-P4/16 - [-1, 1, Conv, [256, 3, 1]], - [-1, 1, nn.MaxPool2d, [2, 2, 0]], # 9-P5/32 - [-1, 1, Conv, [512, 3, 1]], - [-1, 1, nn.ZeroPad2d, [[0, 1, 0, 1]]], # 11 - [-1, 1, nn.MaxPool2d, [2, 1, 0]], # 12 - ] - -# YOLOv3-tiny head -head: - [[-1, 1, Conv, [1024, 3, 1]], - [-1, 1, Conv, [256, 1, 1]], - [-1, 1, Conv, [512, 3, 1]], # 15 (P5/32-large) - - [-2, 1, Conv, [128, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 8], 1, Concat, [1]], # cat backbone P4 - [-1, 1, Conv, [256, 3, 1]], # 19 (P4/16-medium) - - [[19, 15], 1, Detect, [nc, anchors]], # Detect(P4, P5) - ] diff --git a/AIlib/models/hub/yolov3.yaml b/AIlib/models/hub/yolov3.yaml deleted file mode 100644 index f2e7613..0000000 --- a/AIlib/models/hub/yolov3.yaml +++ /dev/null @@ -1,51 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: - - [10,13, 16,30, 33,23] # P3/8 - - [30,61, 62,45, 59,119] # P4/16 - - [116,90, 156,198, 373,326] # P5/32 - -# darknet53 backbone -backbone: - # [from, number, module, args] - [[-1, 1, Conv, [32, 3, 1]], # 0 - [-1, 1, Conv, [64, 3, 2]], # 1-P1/2 - [-1, 1, Bottleneck, [64]], - [-1, 1, Conv, [128, 3, 2]], # 3-P2/4 - [-1, 2, Bottleneck, [128]], - [-1, 1, Conv, [256, 3, 2]], # 5-P3/8 - [-1, 8, Bottleneck, [256]], - [-1, 1, Conv, [512, 3, 2]], # 7-P4/16 - [-1, 8, Bottleneck, [512]], - [-1, 1, Conv, [1024, 3, 2]], # 9-P5/32 - [-1, 4, Bottleneck, [1024]], # 10 - ] - -# YOLOv3 head -head: - [[-1, 1, Bottleneck, [1024, False]], - [-1, 1, Conv, [512, [1, 1]]], - [-1, 1, Conv, [1024, 3, 1]], - [-1, 1, Conv, [512, 1, 1]], - [-1, 1, Conv, [1024, 3, 1]], # 15 (P5/32-large) - - [-2, 1, Conv, [256, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 8], 1, Concat, [1]], # cat backbone P4 - [-1, 1, Bottleneck, [512, False]], - [-1, 1, Bottleneck, [512, False]], - [-1, 1, Conv, [256, 1, 1]], - [-1, 1, Conv, [512, 3, 1]], # 22 (P4/16-medium) - - [-2, 1, Conv, [128, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 6], 1, Concat, [1]], # cat backbone P3 - [-1, 1, Bottleneck, [256, False]], - [-1, 2, Bottleneck, [256, False]], # 27 (P3/8-small) - - [[27, 22, 15], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) - ] diff --git a/AIlib/models/hub/yolov5-fpn.yaml b/AIlib/models/hub/yolov5-fpn.yaml deleted file mode 100644 index e772bff..0000000 --- a/AIlib/models/hub/yolov5-fpn.yaml +++ /dev/null @@ -1,42 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: - - [10,13, 16,30, 33,23] # P3/8 - - [30,61, 62,45, 59,119] # P4/16 - - [116,90, 156,198, 373,326] # P5/32 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [[-1, 1, Focus, [64, 3]], # 0-P1/2 - [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 - [-1, 3, Bottleneck, [128]], - [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 - [-1, 9, BottleneckCSP, [256]], - [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 - [-1, 9, BottleneckCSP, [512]], - [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 - [-1, 1, SPP, [1024, [5, 9, 13]]], - [-1, 6, BottleneckCSP, [1024]], # 9 - ] - -# YOLOv5 FPN head -head: - [[-1, 3, BottleneckCSP, [1024, False]], # 10 (P5/32-large) - - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 6], 1, Concat, [1]], # cat backbone P4 - [-1, 1, Conv, [512, 1, 1]], - [-1, 3, BottleneckCSP, [512, False]], # 14 (P4/16-medium) - - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 4], 1, Concat, [1]], # cat backbone P3 - [-1, 1, Conv, [256, 1, 1]], - [-1, 3, BottleneckCSP, [256, False]], # 18 (P3/8-small) - - [[18, 14, 10], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) - ] diff --git a/AIlib/models/hub/yolov5-p2.yaml b/AIlib/models/hub/yolov5-p2.yaml deleted file mode 100644 index 0633a90..0000000 --- a/AIlib/models/hub/yolov5-p2.yaml +++ /dev/null @@ -1,54 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: 3 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [ [ -1, 1, Focus, [ 64, 3 ] ], # 0-P1/2 - [ -1, 1, Conv, [ 128, 3, 2 ] ], # 1-P2/4 - [ -1, 3, C3, [ 128 ] ], - [ -1, 1, Conv, [ 256, 3, 2 ] ], # 3-P3/8 - [ -1, 9, C3, [ 256 ] ], - [ -1, 1, Conv, [ 512, 3, 2 ] ], # 5-P4/16 - [ -1, 9, C3, [ 512 ] ], - [ -1, 1, Conv, [ 1024, 3, 2 ] ], # 7-P5/32 - [ -1, 1, SPP, [ 1024, [ 5, 9, 13 ] ] ], - [ -1, 3, C3, [ 1024, False ] ], # 9 - ] - -# YOLOv5 head -head: - [ [ -1, 1, Conv, [ 512, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 6 ], 1, Concat, [ 1 ] ], # cat backbone P4 - [ -1, 3, C3, [ 512, False ] ], # 13 - - [ -1, 1, Conv, [ 256, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 4 ], 1, Concat, [ 1 ] ], # cat backbone P3 - [ -1, 3, C3, [ 256, False ] ], # 17 (P3/8-small) - - [ -1, 1, Conv, [ 128, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 2 ], 1, Concat, [ 1 ] ], # cat backbone P2 - [ -1, 1, C3, [ 128, False ] ], # 21 (P2/4-xsmall) - - [ -1, 1, Conv, [ 128, 3, 2 ] ], - [ [ -1, 18 ], 1, Concat, [ 1 ] ], # cat head P3 - [ -1, 3, C3, [ 256, False ] ], # 24 (P3/8-small) - - [ -1, 1, Conv, [ 256, 3, 2 ] ], - [ [ -1, 14 ], 1, Concat, [ 1 ] ], # cat head P4 - [ -1, 3, C3, [ 512, False ] ], # 27 (P4/16-medium) - - [ -1, 1, Conv, [ 512, 3, 2 ] ], - [ [ -1, 10 ], 1, Concat, [ 1 ] ], # cat head P5 - [ -1, 3, C3, [ 1024, False ] ], # 30 (P5/32-large) - - [ [ 24, 27, 30 ], 1, Detect, [ nc, anchors ] ], # Detect(P3, P4, P5) - ] diff --git a/AIlib/models/hub/yolov5-p6.yaml b/AIlib/models/hub/yolov5-p6.yaml deleted file mode 100644 index 3728a11..0000000 --- a/AIlib/models/hub/yolov5-p6.yaml +++ /dev/null @@ -1,56 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: 3 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [ [ -1, 1, Focus, [ 64, 3 ] ], # 0-P1/2 - [ -1, 1, Conv, [ 128, 3, 2 ] ], # 1-P2/4 - [ -1, 3, C3, [ 128 ] ], - [ -1, 1, Conv, [ 256, 3, 2 ] ], # 3-P3/8 - [ -1, 9, C3, [ 256 ] ], - [ -1, 1, Conv, [ 512, 3, 2 ] ], # 5-P4/16 - [ -1, 9, C3, [ 512 ] ], - [ -1, 1, Conv, [ 768, 3, 2 ] ], # 7-P5/32 - [ -1, 3, C3, [ 768 ] ], - [ -1, 1, Conv, [ 1024, 3, 2 ] ], # 9-P6/64 - [ -1, 1, SPP, [ 1024, [ 3, 5, 7 ] ] ], - [ -1, 3, C3, [ 1024, False ] ], # 11 - ] - -# YOLOv5 head -head: - [ [ -1, 1, Conv, [ 768, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 8 ], 1, Concat, [ 1 ] ], # cat backbone P5 - [ -1, 3, C3, [ 768, False ] ], # 15 - - [ -1, 1, Conv, [ 512, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 6 ], 1, Concat, [ 1 ] ], # cat backbone P4 - [ -1, 3, C3, [ 512, False ] ], # 19 - - [ -1, 1, Conv, [ 256, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 4 ], 1, Concat, [ 1 ] ], # cat backbone P3 - [ -1, 3, C3, [ 256, False ] ], # 23 (P3/8-small) - - [ -1, 1, Conv, [ 256, 3, 2 ] ], - [ [ -1, 20 ], 1, Concat, [ 1 ] ], # cat head P4 - [ -1, 3, C3, [ 512, False ] ], # 26 (P4/16-medium) - - [ -1, 1, Conv, [ 512, 3, 2 ] ], - [ [ -1, 16 ], 1, Concat, [ 1 ] ], # cat head P5 - [ -1, 3, C3, [ 768, False ] ], # 29 (P5/32-large) - - [ -1, 1, Conv, [ 768, 3, 2 ] ], - [ [ -1, 12 ], 1, Concat, [ 1 ] ], # cat head P6 - [ -1, 3, C3, [ 1024, False ] ], # 32 (P5/64-xlarge) - - [ [ 23, 26, 29, 32 ], 1, Detect, [ nc, anchors ] ], # Detect(P3, P4, P5, P6) - ] diff --git a/AIlib/models/hub/yolov5-p7.yaml b/AIlib/models/hub/yolov5-p7.yaml deleted file mode 100644 index ca8f849..0000000 --- a/AIlib/models/hub/yolov5-p7.yaml +++ /dev/null @@ -1,67 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: 3 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [ [ -1, 1, Focus, [ 64, 3 ] ], # 0-P1/2 - [ -1, 1, Conv, [ 128, 3, 2 ] ], # 1-P2/4 - [ -1, 3, C3, [ 128 ] ], - [ -1, 1, Conv, [ 256, 3, 2 ] ], # 3-P3/8 - [ -1, 9, C3, [ 256 ] ], - [ -1, 1, Conv, [ 512, 3, 2 ] ], # 5-P4/16 - [ -1, 9, C3, [ 512 ] ], - [ -1, 1, Conv, [ 768, 3, 2 ] ], # 7-P5/32 - [ -1, 3, C3, [ 768 ] ], - [ -1, 1, Conv, [ 1024, 3, 2 ] ], # 9-P6/64 - [ -1, 3, C3, [ 1024 ] ], - [ -1, 1, Conv, [ 1280, 3, 2 ] ], # 11-P7/128 - [ -1, 1, SPP, [ 1280, [ 3, 5 ] ] ], - [ -1, 3, C3, [ 1280, False ] ], # 13 - ] - -# YOLOv5 head -head: - [ [ -1, 1, Conv, [ 1024, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 10 ], 1, Concat, [ 1 ] ], # cat backbone P6 - [ -1, 3, C3, [ 1024, False ] ], # 17 - - [ -1, 1, Conv, [ 768, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 8 ], 1, Concat, [ 1 ] ], # cat backbone P5 - [ -1, 3, C3, [ 768, False ] ], # 21 - - [ -1, 1, Conv, [ 512, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 6 ], 1, Concat, [ 1 ] ], # cat backbone P4 - [ -1, 3, C3, [ 512, False ] ], # 25 - - [ -1, 1, Conv, [ 256, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 4 ], 1, Concat, [ 1 ] ], # cat backbone P3 - [ -1, 3, C3, [ 256, False ] ], # 29 (P3/8-small) - - [ -1, 1, Conv, [ 256, 3, 2 ] ], - [ [ -1, 26 ], 1, Concat, [ 1 ] ], # cat head P4 - [ -1, 3, C3, [ 512, False ] ], # 32 (P4/16-medium) - - [ -1, 1, Conv, [ 512, 3, 2 ] ], - [ [ -1, 22 ], 1, Concat, [ 1 ] ], # cat head P5 - [ -1, 3, C3, [ 768, False ] ], # 35 (P5/32-large) - - [ -1, 1, Conv, [ 768, 3, 2 ] ], - [ [ -1, 18 ], 1, Concat, [ 1 ] ], # cat head P6 - [ -1, 3, C3, [ 1024, False ] ], # 38 (P6/64-xlarge) - - [ -1, 1, Conv, [ 1024, 3, 2 ] ], - [ [ -1, 14 ], 1, Concat, [ 1 ] ], # cat head P7 - [ -1, 3, C3, [ 1280, False ] ], # 41 (P7/128-xxlarge) - - [ [ 29, 32, 35, 38, 41 ], 1, Detect, [ nc, anchors ] ], # Detect(P3, P4, P5, P6, P7) - ] diff --git a/AIlib/models/hub/yolov5-panet.yaml b/AIlib/models/hub/yolov5-panet.yaml deleted file mode 100644 index 340f95a..0000000 --- a/AIlib/models/hub/yolov5-panet.yaml +++ /dev/null @@ -1,48 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: - - [10,13, 16,30, 33,23] # P3/8 - - [30,61, 62,45, 59,119] # P4/16 - - [116,90, 156,198, 373,326] # P5/32 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [[-1, 1, Focus, [64, 3]], # 0-P1/2 - [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 - [-1, 3, BottleneckCSP, [128]], - [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 - [-1, 9, BottleneckCSP, [256]], - [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 - [-1, 9, BottleneckCSP, [512]], - [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 - [-1, 1, SPP, [1024, [5, 9, 13]]], - [-1, 3, BottleneckCSP, [1024, False]], # 9 - ] - -# YOLOv5 PANet head -head: - [[-1, 1, Conv, [512, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 6], 1, Concat, [1]], # cat backbone P4 - [-1, 3, BottleneckCSP, [512, False]], # 13 - - [-1, 1, Conv, [256, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 4], 1, Concat, [1]], # cat backbone P3 - [-1, 3, BottleneckCSP, [256, False]], # 17 (P3/8-small) - - [-1, 1, Conv, [256, 3, 2]], - [[-1, 14], 1, Concat, [1]], # cat head P4 - [-1, 3, BottleneckCSP, [512, False]], # 20 (P4/16-medium) - - [-1, 1, Conv, [512, 3, 2]], - [[-1, 10], 1, Concat, [1]], # cat head P5 - [-1, 3, BottleneckCSP, [1024, False]], # 23 (P5/32-large) - - [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) - ] diff --git a/AIlib/models/hub/yolov5l6.yaml b/AIlib/models/hub/yolov5l6.yaml deleted file mode 100644 index 11298b0..0000000 --- a/AIlib/models/hub/yolov5l6.yaml +++ /dev/null @@ -1,60 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: - - [ 19,27, 44,40, 38,94 ] # P3/8 - - [ 96,68, 86,152, 180,137 ] # P4/16 - - [ 140,301, 303,264, 238,542 ] # P5/32 - - [ 436,615, 739,380, 925,792 ] # P6/64 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [ [ -1, 1, Focus, [ 64, 3 ] ], # 0-P1/2 - [ -1, 1, Conv, [ 128, 3, 2 ] ], # 1-P2/4 - [ -1, 3, C3, [ 128 ] ], - [ -1, 1, Conv, [ 256, 3, 2 ] ], # 3-P3/8 - [ -1, 9, C3, [ 256 ] ], - [ -1, 1, Conv, [ 512, 3, 2 ] ], # 5-P4/16 - [ -1, 9, C3, [ 512 ] ], - [ -1, 1, Conv, [ 768, 3, 2 ] ], # 7-P5/32 - [ -1, 3, C3, [ 768 ] ], - [ -1, 1, Conv, [ 1024, 3, 2 ] ], # 9-P6/64 - [ -1, 1, SPP, [ 1024, [ 3, 5, 7 ] ] ], - [ -1, 3, C3, [ 1024, False ] ], # 11 - ] - -# YOLOv5 head -head: - [ [ -1, 1, Conv, [ 768, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 8 ], 1, Concat, [ 1 ] ], # cat backbone P5 - [ -1, 3, C3, [ 768, False ] ], # 15 - - [ -1, 1, Conv, [ 512, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 6 ], 1, Concat, [ 1 ] ], # cat backbone P4 - [ -1, 3, C3, [ 512, False ] ], # 19 - - [ -1, 1, Conv, [ 256, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 4 ], 1, Concat, [ 1 ] ], # cat backbone P3 - [ -1, 3, C3, [ 256, False ] ], # 23 (P3/8-small) - - [ -1, 1, Conv, [ 256, 3, 2 ] ], - [ [ -1, 20 ], 1, Concat, [ 1 ] ], # cat head P4 - [ -1, 3, C3, [ 512, False ] ], # 26 (P4/16-medium) - - [ -1, 1, Conv, [ 512, 3, 2 ] ], - [ [ -1, 16 ], 1, Concat, [ 1 ] ], # cat head P5 - [ -1, 3, C3, [ 768, False ] ], # 29 (P5/32-large) - - [ -1, 1, Conv, [ 768, 3, 2 ] ], - [ [ -1, 12 ], 1, Concat, [ 1 ] ], # cat head P6 - [ -1, 3, C3, [ 1024, False ] ], # 32 (P6/64-xlarge) - - [ [ 23, 26, 29, 32 ], 1, Detect, [ nc, anchors ] ], # Detect(P3, P4, P5, P6) - ] diff --git a/AIlib/models/hub/yolov5m6.yaml b/AIlib/models/hub/yolov5m6.yaml deleted file mode 100644 index 48afc86..0000000 --- a/AIlib/models/hub/yolov5m6.yaml +++ /dev/null @@ -1,60 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 0.67 # model depth multiple -width_multiple: 0.75 # layer channel multiple - -# anchors -anchors: - - [ 19,27, 44,40, 38,94 ] # P3/8 - - [ 96,68, 86,152, 180,137 ] # P4/16 - - [ 140,301, 303,264, 238,542 ] # P5/32 - - [ 436,615, 739,380, 925,792 ] # P6/64 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [ [ -1, 1, Focus, [ 64, 3 ] ], # 0-P1/2 - [ -1, 1, Conv, [ 128, 3, 2 ] ], # 1-P2/4 - [ -1, 3, C3, [ 128 ] ], - [ -1, 1, Conv, [ 256, 3, 2 ] ], # 3-P3/8 - [ -1, 9, C3, [ 256 ] ], - [ -1, 1, Conv, [ 512, 3, 2 ] ], # 5-P4/16 - [ -1, 9, C3, [ 512 ] ], - [ -1, 1, Conv, [ 768, 3, 2 ] ], # 7-P5/32 - [ -1, 3, C3, [ 768 ] ], - [ -1, 1, Conv, [ 1024, 3, 2 ] ], # 9-P6/64 - [ -1, 1, SPP, [ 1024, [ 3, 5, 7 ] ] ], - [ -1, 3, C3, [ 1024, False ] ], # 11 - ] - -# YOLOv5 head -head: - [ [ -1, 1, Conv, [ 768, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 8 ], 1, Concat, [ 1 ] ], # cat backbone P5 - [ -1, 3, C3, [ 768, False ] ], # 15 - - [ -1, 1, Conv, [ 512, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 6 ], 1, Concat, [ 1 ] ], # cat backbone P4 - [ -1, 3, C3, [ 512, False ] ], # 19 - - [ -1, 1, Conv, [ 256, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 4 ], 1, Concat, [ 1 ] ], # cat backbone P3 - [ -1, 3, C3, [ 256, False ] ], # 23 (P3/8-small) - - [ -1, 1, Conv, [ 256, 3, 2 ] ], - [ [ -1, 20 ], 1, Concat, [ 1 ] ], # cat head P4 - [ -1, 3, C3, [ 512, False ] ], # 26 (P4/16-medium) - - [ -1, 1, Conv, [ 512, 3, 2 ] ], - [ [ -1, 16 ], 1, Concat, [ 1 ] ], # cat head P5 - [ -1, 3, C3, [ 768, False ] ], # 29 (P5/32-large) - - [ -1, 1, Conv, [ 768, 3, 2 ] ], - [ [ -1, 12 ], 1, Concat, [ 1 ] ], # cat head P6 - [ -1, 3, C3, [ 1024, False ] ], # 32 (P6/64-xlarge) - - [ [ 23, 26, 29, 32 ], 1, Detect, [ nc, anchors ] ], # Detect(P3, P4, P5, P6) - ] diff --git a/AIlib/models/hub/yolov5s-transformer.yaml b/AIlib/models/hub/yolov5s-transformer.yaml deleted file mode 100644 index f2d6667..0000000 --- a/AIlib/models/hub/yolov5s-transformer.yaml +++ /dev/null @@ -1,48 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 0.33 # model depth multiple -width_multiple: 0.50 # layer channel multiple - -# anchors -anchors: - - [10,13, 16,30, 33,23] # P3/8 - - [30,61, 62,45, 59,119] # P4/16 - - [116,90, 156,198, 373,326] # P5/32 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [[-1, 1, Focus, [64, 3]], # 0-P1/2 - [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 - [-1, 3, C3, [128]], - [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 - [-1, 9, C3, [256]], - [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 - [-1, 9, C3, [512]], - [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 - [-1, 1, SPP, [1024, [5, 9, 13]]], - [-1, 3, C3TR, [1024, False]], # 9 <-------- C3TR() Transformer module - ] - -# YOLOv5 head -head: - [[-1, 1, Conv, [512, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 6], 1, Concat, [1]], # cat backbone P4 - [-1, 3, C3, [512, False]], # 13 - - [-1, 1, Conv, [256, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 4], 1, Concat, [1]], # cat backbone P3 - [-1, 3, C3, [256, False]], # 17 (P3/8-small) - - [-1, 1, Conv, [256, 3, 2]], - [[-1, 14], 1, Concat, [1]], # cat head P4 - [-1, 3, C3, [512, False]], # 20 (P4/16-medium) - - [-1, 1, Conv, [512, 3, 2]], - [[-1, 10], 1, Concat, [1]], # cat head P5 - [-1, 3, C3, [1024, False]], # 23 (P5/32-large) - - [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) - ] diff --git a/AIlib/models/hub/yolov5s6.yaml b/AIlib/models/hub/yolov5s6.yaml deleted file mode 100644 index 1df577a..0000000 --- a/AIlib/models/hub/yolov5s6.yaml +++ /dev/null @@ -1,60 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 0.33 # model depth multiple -width_multiple: 0.50 # layer channel multiple - -# anchors -anchors: - - [ 19,27, 44,40, 38,94 ] # P3/8 - - [ 96,68, 86,152, 180,137 ] # P4/16 - - [ 140,301, 303,264, 238,542 ] # P5/32 - - [ 436,615, 739,380, 925,792 ] # P6/64 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [ [ -1, 1, Focus, [ 64, 3 ] ], # 0-P1/2 - [ -1, 1, Conv, [ 128, 3, 2 ] ], # 1-P2/4 - [ -1, 3, C3, [ 128 ] ], - [ -1, 1, Conv, [ 256, 3, 2 ] ], # 3-P3/8 - [ -1, 9, C3, [ 256 ] ], - [ -1, 1, Conv, [ 512, 3, 2 ] ], # 5-P4/16 - [ -1, 9, C3, [ 512 ] ], - [ -1, 1, Conv, [ 768, 3, 2 ] ], # 7-P5/32 - [ -1, 3, C3, [ 768 ] ], - [ -1, 1, Conv, [ 1024, 3, 2 ] ], # 9-P6/64 - [ -1, 1, SPP, [ 1024, [ 3, 5, 7 ] ] ], - [ -1, 3, C3, [ 1024, False ] ], # 11 - ] - -# YOLOv5 head -head: - [ [ -1, 1, Conv, [ 768, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 8 ], 1, Concat, [ 1 ] ], # cat backbone P5 - [ -1, 3, C3, [ 768, False ] ], # 15 - - [ -1, 1, Conv, [ 512, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 6 ], 1, Concat, [ 1 ] ], # cat backbone P4 - [ -1, 3, C3, [ 512, False ] ], # 19 - - [ -1, 1, Conv, [ 256, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 4 ], 1, Concat, [ 1 ] ], # cat backbone P3 - [ -1, 3, C3, [ 256, False ] ], # 23 (P3/8-small) - - [ -1, 1, Conv, [ 256, 3, 2 ] ], - [ [ -1, 20 ], 1, Concat, [ 1 ] ], # cat head P4 - [ -1, 3, C3, [ 512, False ] ], # 26 (P4/16-medium) - - [ -1, 1, Conv, [ 512, 3, 2 ] ], - [ [ -1, 16 ], 1, Concat, [ 1 ] ], # cat head P5 - [ -1, 3, C3, [ 768, False ] ], # 29 (P5/32-large) - - [ -1, 1, Conv, [ 768, 3, 2 ] ], - [ [ -1, 12 ], 1, Concat, [ 1 ] ], # cat head P6 - [ -1, 3, C3, [ 1024, False ] ], # 32 (P6/64-xlarge) - - [ [ 23, 26, 29, 32 ], 1, Detect, [ nc, anchors ] ], # Detect(P3, P4, P5, P6) - ] diff --git a/AIlib/models/hub/yolov5x6.yaml b/AIlib/models/hub/yolov5x6.yaml deleted file mode 100644 index 5ebc021..0000000 --- a/AIlib/models/hub/yolov5x6.yaml +++ /dev/null @@ -1,60 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.33 # model depth multiple -width_multiple: 1.25 # layer channel multiple - -# anchors -anchors: - - [ 19,27, 44,40, 38,94 ] # P3/8 - - [ 96,68, 86,152, 180,137 ] # P4/16 - - [ 140,301, 303,264, 238,542 ] # P5/32 - - [ 436,615, 739,380, 925,792 ] # P6/64 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [ [ -1, 1, Focus, [ 64, 3 ] ], # 0-P1/2 - [ -1, 1, Conv, [ 128, 3, 2 ] ], # 1-P2/4 - [ -1, 3, C3, [ 128 ] ], - [ -1, 1, Conv, [ 256, 3, 2 ] ], # 3-P3/8 - [ -1, 9, C3, [ 256 ] ], - [ -1, 1, Conv, [ 512, 3, 2 ] ], # 5-P4/16 - [ -1, 9, C3, [ 512 ] ], - [ -1, 1, Conv, [ 768, 3, 2 ] ], # 7-P5/32 - [ -1, 3, C3, [ 768 ] ], - [ -1, 1, Conv, [ 1024, 3, 2 ] ], # 9-P6/64 - [ -1, 1, SPP, [ 1024, [ 3, 5, 7 ] ] ], - [ -1, 3, C3, [ 1024, False ] ], # 11 - ] - -# YOLOv5 head -head: - [ [ -1, 1, Conv, [ 768, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 8 ], 1, Concat, [ 1 ] ], # cat backbone P5 - [ -1, 3, C3, [ 768, False ] ], # 15 - - [ -1, 1, Conv, [ 512, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 6 ], 1, Concat, [ 1 ] ], # cat backbone P4 - [ -1, 3, C3, [ 512, False ] ], # 19 - - [ -1, 1, Conv, [ 256, 1, 1 ] ], - [ -1, 1, nn.Upsample, [ None, 2, 'nearest' ] ], - [ [ -1, 4 ], 1, Concat, [ 1 ] ], # cat backbone P3 - [ -1, 3, C3, [ 256, False ] ], # 23 (P3/8-small) - - [ -1, 1, Conv, [ 256, 3, 2 ] ], - [ [ -1, 20 ], 1, Concat, [ 1 ] ], # cat head P4 - [ -1, 3, C3, [ 512, False ] ], # 26 (P4/16-medium) - - [ -1, 1, Conv, [ 512, 3, 2 ] ], - [ [ -1, 16 ], 1, Concat, [ 1 ] ], # cat head P5 - [ -1, 3, C3, [ 768, False ] ], # 29 (P5/32-large) - - [ -1, 1, Conv, [ 768, 3, 2 ] ], - [ [ -1, 12 ], 1, Concat, [ 1 ] ], # cat head P6 - [ -1, 3, C3, [ 1024, False ] ], # 32 (P6/64-xlarge) - - [ [ 23, 26, 29, 32 ], 1, Detect, [ nc, anchors ] ], # Detect(P3, P4, P5, P6) - ] diff --git a/AIlib/models/yolo.py b/AIlib/models/yolo.py deleted file mode 100644 index f730a1e..0000000 --- a/AIlib/models/yolo.py +++ /dev/null @@ -1,277 +0,0 @@ -# YOLOv5 YOLO-specific modules - -import argparse -import logging -import sys -from copy import deepcopy - -sys.path.append('./') # to run '$ python *.py' files in subdirectories -logger = logging.getLogger(__name__) - -from models.common import * -from models.experimental import * -from utils.autoanchor import check_anchor_order -from utils.general import make_divisible, check_file, set_logging -from utils.torch_utils import time_synchronized, fuse_conv_and_bn, model_info, scale_img, initialize_weights, \ - select_device, copy_attr - -try: - import thop # for FLOPS computation -except ImportError: - thop = None - - -class Detect(nn.Module): - stride = None # strides computed during build - export = False # onnx export - - def __init__(self, nc=80, anchors=(), ch=()): # detection layer - super(Detect, self).__init__() - self.nc = nc # number of classes - self.no = nc + 5 # number of outputs per anchor - self.nl = len(anchors) # number of detection layers - self.na = len(anchors[0]) // 2 # number of anchors - self.grid = [torch.zeros(1)] * self.nl # init grid - a = torch.tensor(anchors).float().view(self.nl, -1, 2) - self.register_buffer('anchors', a) # shape(nl,na,2) - self.register_buffer('anchor_grid', a.clone().view(self.nl, 1, -1, 1, 1, 2)) # shape(nl,1,na,1,1,2) - self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv - - def forward(self, x): - # x = x.copy() # for profiling - z = [] # inference output - self.training |= self.export - for i in range(self.nl): - x[i] = self.m[i](x[i]) # conv - bs, _, ny, nx = x[i].shape # x(bs,255,20,20) to x(bs,3,20,20,85) - x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous() - - if not self.training: # inference - if self.grid[i].shape[2:4] != x[i].shape[2:4]: - self.grid[i] = self._make_grid(nx, ny).to(x[i].device) - - y = x[i].sigmoid() - y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i] # xy - y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh - z.append(y.view(bs, -1, self.no)) - - return x if self.training else (torch.cat(z, 1), x) - - @staticmethod - def _make_grid(nx=20, ny=20): - yv, xv = torch.meshgrid([torch.arange(ny), torch.arange(nx)]) - return torch.stack((xv, yv), 2).view((1, 1, ny, nx, 2)).float() - - -class Model(nn.Module): - def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, anchors=None): # model, input channels, number of classes - super(Model, self).__init__() - if isinstance(cfg, dict): - self.yaml = cfg # model dict - else: # is *.yaml - import yaml # for torch hub - self.yaml_file = Path(cfg).name - with open(cfg) as f: - self.yaml = yaml.load(f, Loader=yaml.SafeLoader) # model dict - - # Define model - ch = self.yaml['ch'] = self.yaml.get('ch', ch) # input channels - if nc and nc != self.yaml['nc']: - logger.info(f"Overriding model.yaml nc={self.yaml['nc']} with nc={nc}") - self.yaml['nc'] = nc # override yaml value - if anchors: - logger.info(f'Overriding model.yaml anchors with anchors={anchors}') - self.yaml['anchors'] = round(anchors) # override yaml value - self.model, self.save = parse_model(deepcopy(self.yaml), ch=[ch]) # model, savelist - self.names = [str(i) for i in range(self.yaml['nc'])] # default names - # print([x.shape for x in self.forward(torch.zeros(1, ch, 64, 64))]) - - # Build strides, anchors - m = self.model[-1] # Detect() - if isinstance(m, Detect): - s = 256 # 2x min stride - m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))]) # forward - m.anchors /= m.stride.view(-1, 1, 1) - check_anchor_order(m) - self.stride = m.stride - self._initialize_biases() # only run once - # print('Strides: %s' % m.stride.tolist()) - - # Init weights, biases - initialize_weights(self) - self.info() - logger.info('') - - def forward(self, x, augment=False, profile=False): - if augment: - img_size = x.shape[-2:] # height, width - s = [1, 0.83, 0.67] # scales - f = [None, 3, None] # flips (2-ud, 3-lr) - y = [] # outputs - for si, fi in zip(s, f): - xi = scale_img(x.flip(fi) if fi else x, si, gs=int(self.stride.max())) - yi = self.forward_once(xi)[0] # forward - # cv2.imwrite(f'img_{si}.jpg', 255 * xi[0].cpu().numpy().transpose((1, 2, 0))[:, :, ::-1]) # save - yi[..., :4] /= si # de-scale - if fi == 2: - yi[..., 1] = img_size[0] - yi[..., 1] # de-flip ud - elif fi == 3: - yi[..., 0] = img_size[1] - yi[..., 0] # de-flip lr - y.append(yi) - return torch.cat(y, 1), None # augmented inference, train - else: - return self.forward_once(x, profile) # single-scale inference, train - - def forward_once(self, x, profile=False): - y, dt = [], [] # outputs - for m in self.model: - if m.f != -1: # if not from previous layer - x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers - - if profile: - o = thop.profile(m, inputs=(x,), verbose=False)[0] / 1E9 * 2 if thop else 0 # FLOPS - t = time_synchronized() - for _ in range(10): - _ = m(x) - dt.append((time_synchronized() - t) * 100) - print('%10.1f%10.0f%10.1fms %-40s' % (o, m.np, dt[-1], m.type)) - - x = m(x) # run - y.append(x if m.i in self.save else None) # save output - - if profile: - print('%.1fms total' % sum(dt)) - return x - - def _initialize_biases(self, cf=None): # initialize biases into Detect(), cf is class frequency - # https://arxiv.org/abs/1708.02002 section 3.3 - # cf = torch.bincount(torch.tensor(np.concatenate(dataset.labels, 0)[:, 0]).long(), minlength=nc) + 1. - m = self.model[-1] # Detect() module - for mi, s in zip(m.m, m.stride): # from - b = mi.bias.view(m.na, -1) # conv.bias(255) to (3,85) - b.data[:, 4] += math.log(8 / (640 / s) ** 2) # obj (8 objects per 640 image) - b.data[:, 5:] += math.log(0.6 / (m.nc - 0.99)) if cf is None else torch.log(cf / cf.sum()) # cls - mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True) - - def _print_biases(self): - m = self.model[-1] # Detect() module - for mi in m.m: # from - b = mi.bias.detach().view(m.na, -1).T # conv.bias(255) to (3,85) - print(('%6g Conv2d.bias:' + '%10.3g' * 6) % (mi.weight.shape[1], *b[:5].mean(1).tolist(), b[5:].mean())) - - # def _print_weights(self): - # for m in self.model.modules(): - # if type(m) is Bottleneck: - # print('%10.3g' % (m.w.detach().sigmoid() * 2)) # shortcut weights - - def fuse(self): # fuse model Conv2d() + BatchNorm2d() layers - print('Fusing layers... ') - for m in self.model.modules(): - if type(m) is Conv and hasattr(m, 'bn'): - m.conv = fuse_conv_and_bn(m.conv, m.bn) # update conv - delattr(m, 'bn') # remove batchnorm - m.forward = m.fuseforward # update forward - self.info() - return self - - def nms(self, mode=True): # add or remove NMS module - present = type(self.model[-1]) is NMS # last layer is NMS - if mode and not present: - print('Adding NMS... ') - m = NMS() # module - m.f = -1 # from - m.i = self.model[-1].i + 1 # index - self.model.add_module(name='%s' % m.i, module=m) # add - self.eval() - elif not mode and present: - print('Removing NMS... ') - self.model = self.model[:-1] # remove - return self - - def autoshape(self): # add autoShape module - print('Adding autoShape... ') - m = autoShape(self) # wrap model - copy_attr(m, self, include=('yaml', 'nc', 'hyp', 'names', 'stride'), exclude=()) # copy attributes - return m - - def info(self, verbose=False, img_size=640): # print model information - model_info(self, verbose, img_size) - - -def parse_model(d, ch): # model_dict, input_channels(3) - logger.info('\n%3s%18s%3s%10s %-40s%-30s' % ('', 'from', 'n', 'params', 'module', 'arguments')) - anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'] - na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors - no = na * (nc + 5) # number of outputs = anchors * (classes + 5) - - layers, save, c2 = [], [], ch[-1] # layers, savelist, ch out - for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): # from, number, module, args - m = eval(m) if isinstance(m, str) else m # eval strings - for j, a in enumerate(args): - try: - args[j] = eval(a) if isinstance(a, str) else a # eval strings - except: - pass - - n = max(round(n * gd), 1) if n > 1 else n # depth gain - if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, - C3, C3TR]: - c1, c2 = ch[f], args[0] - if c2 != no: # if not output - c2 = make_divisible(c2 * gw, 8) - - args = [c1, c2, *args[1:]] - if m in [BottleneckCSP, C3, C3TR]: - args.insert(2, n) # number of repeats - n = 1 - elif m is nn.BatchNorm2d: - args = [ch[f]] - elif m is Concat: - c2 = sum([ch[x] for x in f]) - elif m is Detect: - args.append([ch[x] for x in f]) - if isinstance(args[1], int): # number of anchors - args[1] = [list(range(args[1] * 2))] * len(f) - elif m is Contract: - c2 = ch[f] * args[0] ** 2 - elif m is Expand: - c2 = ch[f] // args[0] ** 2 - else: - c2 = ch[f] - - m_ = nn.Sequential(*[m(*args) for _ in range(n)]) if n > 1 else m(*args) # module - t = str(m)[8:-2].replace('__main__.', '') # module type - np = sum([x.numel() for x in m_.parameters()]) # number params - m_.i, m_.f, m_.type, m_.np = i, f, t, np # attach index, 'from' index, type, number params - logger.info('%3s%18s%3s%10.0f %-40s%-30s' % (i, f, n, np, t, args)) # print - save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1) # append to savelist - layers.append(m_) - if i == 0: - ch = [] - ch.append(c2) - return nn.Sequential(*layers), sorted(save) - - -if __name__ == '__main__': - parser = argparse.ArgumentParser() - parser.add_argument('--cfg', type=str, default='yolov5s.yaml', help='model.yaml') - parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu') - opt = parser.parse_args() - opt.cfg = check_file(opt.cfg) # check file - set_logging() - device = select_device(opt.device) - - # Create model - model = Model(opt.cfg).to(device) - model.train() - - # Profile - # img = torch.rand(8 if torch.cuda.is_available() else 1, 3, 640, 640).to(device) - # y = model(img, profile=True) - - # Tensorboard - # from torch.utils.tensorboard import SummaryWriter - # tb_writer = SummaryWriter() - # print("Run 'tensorboard --logdir=models/runs' to view tensorboard at http://localhost:6006/") - # tb_writer.add_graph(model.model, img) # add model to tensorboard - # tb_writer.add_image('test', img[0], dataformats='CWH') # add model to tensorboard diff --git a/AIlib/models/yolov5l.yaml b/AIlib/models/yolov5l.yaml deleted file mode 100644 index 71ebf86..0000000 --- a/AIlib/models/yolov5l.yaml +++ /dev/null @@ -1,48 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.0 # model depth multiple -width_multiple: 1.0 # layer channel multiple - -# anchors -anchors: - - [10,13, 16,30, 33,23] # P3/8 - - [30,61, 62,45, 59,119] # P4/16 - - [116,90, 156,198, 373,326] # P5/32 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [[-1, 1, Focus, [64, 3]], # 0-P1/2 - [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 - [-1, 3, C3, [128]], - [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 - [-1, 9, C3, [256]], - [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 - [-1, 9, C3, [512]], - [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 - [-1, 1, SPP, [1024, [5, 9, 13]]], - [-1, 3, C3, [1024, False]], # 9 - ] - -# YOLOv5 head -head: - [[-1, 1, Conv, [512, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 6], 1, Concat, [1]], # cat backbone P4 - [-1, 3, C3, [512, False]], # 13 - - [-1, 1, Conv, [256, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 4], 1, Concat, [1]], # cat backbone P3 - [-1, 3, C3, [256, False]], # 17 (P3/8-small) - - [-1, 1, Conv, [256, 3, 2]], - [[-1, 14], 1, Concat, [1]], # cat head P4 - [-1, 3, C3, [512, False]], # 20 (P4/16-medium) - - [-1, 1, Conv, [512, 3, 2]], - [[-1, 10], 1, Concat, [1]], # cat head P5 - [-1, 3, C3, [1024, False]], # 23 (P5/32-large) - - [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) - ] diff --git a/AIlib/models/yolov5m.yaml b/AIlib/models/yolov5m.yaml deleted file mode 100644 index 3c749c9..0000000 --- a/AIlib/models/yolov5m.yaml +++ /dev/null @@ -1,48 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 0.67 # model depth multiple -width_multiple: 0.75 # layer channel multiple - -# anchors -anchors: - - [10,13, 16,30, 33,23] # P3/8 - - [30,61, 62,45, 59,119] # P4/16 - - [116,90, 156,198, 373,326] # P5/32 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [[-1, 1, Focus, [64, 3]], # 0-P1/2 - [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 - [-1, 3, C3, [128]], - [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 - [-1, 9, C3, [256]], - [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 - [-1, 9, C3, [512]], - [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 - [-1, 1, SPP, [1024, [5, 9, 13]]], - [-1, 3, C3, [1024, False]], # 9 - ] - -# YOLOv5 head -head: - [[-1, 1, Conv, [512, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 6], 1, Concat, [1]], # cat backbone P4 - [-1, 3, C3, [512, False]], # 13 - - [-1, 1, Conv, [256, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 4], 1, Concat, [1]], # cat backbone P3 - [-1, 3, C3, [256, False]], # 17 (P3/8-small) - - [-1, 1, Conv, [256, 3, 2]], - [[-1, 14], 1, Concat, [1]], # cat head P4 - [-1, 3, C3, [512, False]], # 20 (P4/16-medium) - - [-1, 1, Conv, [512, 3, 2]], - [[-1, 10], 1, Concat, [1]], # cat head P5 - [-1, 3, C3, [1024, False]], # 23 (P5/32-large) - - [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) - ] diff --git a/AIlib/models/yolov5s.yaml b/AIlib/models/yolov5s.yaml deleted file mode 100644 index aca669d..0000000 --- a/AIlib/models/yolov5s.yaml +++ /dev/null @@ -1,48 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 0.33 # model depth multiple -width_multiple: 0.50 # layer channel multiple - -# anchors -anchors: - - [10,13, 16,30, 33,23] # P3/8 - - [30,61, 62,45, 59,119] # P4/16 - - [116,90, 156,198, 373,326] # P5/32 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [[-1, 1, Focus, [64, 3]], # 0-P1/2 - [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 - [-1, 3, C3, [128]], - [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 - [-1, 9, C3, [256]], - [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 - [-1, 9, C3, [512]], - [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 - [-1, 1, SPP, [1024, [5, 9, 13]]], - [-1, 3, C3, [1024, False]], # 9 - ] - -# YOLOv5 head -head: - [[-1, 1, Conv, [512, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 6], 1, Concat, [1]], # cat backbone P4 - [-1, 3, C3, [512, False]], # 13 - - [-1, 1, Conv, [256, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 4], 1, Concat, [1]], # cat backbone P3 - [-1, 3, C3, [256, False]], # 17 (P3/8-small) - - [-1, 1, Conv, [256, 3, 2]], - [[-1, 14], 1, Concat, [1]], # cat head P4 - [-1, 3, C3, [512, False]], # 20 (P4/16-medium) - - [-1, 1, Conv, [512, 3, 2]], - [[-1, 10], 1, Concat, [1]], # cat head P5 - [-1, 3, C3, [1024, False]], # 23 (P5/32-large) - - [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) - ] diff --git a/AIlib/models/yolov5x.yaml b/AIlib/models/yolov5x.yaml deleted file mode 100644 index d3babdf..0000000 --- a/AIlib/models/yolov5x.yaml +++ /dev/null @@ -1,48 +0,0 @@ -# parameters -nc: 80 # number of classes -depth_multiple: 1.33 # model depth multiple -width_multiple: 1.25 # layer channel multiple - -# anchors -anchors: - - [10,13, 16,30, 33,23] # P3/8 - - [30,61, 62,45, 59,119] # P4/16 - - [116,90, 156,198, 373,326] # P5/32 - -# YOLOv5 backbone -backbone: - # [from, number, module, args] - [[-1, 1, Focus, [64, 3]], # 0-P1/2 - [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 - [-1, 3, C3, [128]], - [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 - [-1, 9, C3, [256]], - [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 - [-1, 9, C3, [512]], - [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 - [-1, 1, SPP, [1024, [5, 9, 13]]], - [-1, 3, C3, [1024, False]], # 9 - ] - -# YOLOv5 head -head: - [[-1, 1, Conv, [512, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 6], 1, Concat, [1]], # cat backbone P4 - [-1, 3, C3, [512, False]], # 13 - - [-1, 1, Conv, [256, 1, 1]], - [-1, 1, nn.Upsample, [None, 2, 'nearest']], - [[-1, 4], 1, Concat, [1]], # cat backbone P3 - [-1, 3, C3, [256, False]], # 17 (P3/8-small) - - [-1, 1, Conv, [256, 3, 2]], - [[-1, 14], 1, Concat, [1]], # cat head P4 - [-1, 3, C3, [512, False]], # 20 (P4/16-medium) - - [-1, 1, Conv, [512, 3, 2]], - [[-1, 10], 1, Concat, [1]], # cat head P5 - [-1, 3, C3, [1024, False]], # 23 (P5/32-large) - - [[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5) - ] diff --git a/AIlib/segutils/GPUtils.py b/AIlib/segutils/GPUtils.py deleted file mode 100644 index 72d8088..0000000 --- a/AIlib/segutils/GPUtils.py +++ /dev/null @@ -1,501 +0,0 @@ -#@@ -1,43 +1,43 @@ -# GPUtil - GPU utilization -# -# A Python module for programmically getting the GPU utilization from NVIDA GPUs using nvidia-smi -# -# Author: Anders Krogh Mortensen (anderskm) -# Date: 16 January 2017 -# Web: https://github.com/anderskm/gputil -# -# LICENSE -# -# MIT License -# -# Copyright (c) 2017 anderskm -# -# Permission is hereby granted, free of charge, to any person obtaining a copy -# of this software and associated documentation files (the "Software"), to deal -# in the Software without restriction, including without limitation the rights -# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell -# copies of the Software, and to permit persons to whom the Software is -# furnished to do so, subject to the following conditions: -# -# The above copyright notice and this permission notice shall be included in all -# copies or substantial portions of the Software. -# -# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR -# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, -# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE -# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER -# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, -# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE -# SOFTWARE. - -from subprocess import Popen, PIPE -from distutils import spawn -import os -import math -import random -import time -import sys -import platform -import subprocess -import numpy as np - - -__version__ = '1.4.0' -class GPU: - def __init__(self, ID, uuid, load, memoryTotal, memoryUsed, memoryFree, driver, gpu_name, serial, display_mode, display_active, temp_gpu): - self.id = ID - self.uuid = uuid - self.load = load - self.memoryUtil = float(memoryUsed)/float(memoryTotal) - self.memoryTotal = memoryTotal - self.memoryUsed = memoryUsed - self.memoryFree = memoryFree - self.driver = driver - self.name = gpu_name - self.serial = serial - self.display_mode = display_mode - self.display_active = display_active - self.temperature = temp_gpu - - def __str__(self): - return str(self.__dict__) - - -class GPUProcess: - def __init__(self, pid, processName, gpuId, gpuUuid, gpuName, usedMemory, - uid, uname): - self.pid = pid - self.processName = processName - self.gpuId = gpuId - self.gpuUuid = gpuUuid - self.gpuName = gpuName - self.usedMemory = usedMemory - self.uid = uid - self.uname = uname - - def __str__(self): - return str(self.__dict__) - -def safeFloatCast(strNumber): - try: - number = float(strNumber) - except ValueError: - number = float('nan') - return number - -#def getGPUs(): -def getNvidiaSmiCmd(): - if platform.system() == "Windows": - # If the platform is Windows and nvidia-smi - # could not be found from the environment path, - #@@ -75,57 +94,97 @@ def getGPUs(): - nvidia_smi = "%s\\Program Files\\NVIDIA Corporation\\NVSMI\\nvidia-smi.exe" % os.environ['systemdrive'] - else: - nvidia_smi = "nvidia-smi" - return nvidia_smi - - -def getGPUs(): - # Get ID, processing and memory utilization for all GPUs - nvidia_smi = getNvidiaSmiCmd() - try: - p = Popen([nvidia_smi,"--query-gpu=index,uuid,utilization.gpu,memory.total,memory.used,memory.free,driver_version,name,gpu_serial,display_active,display_mode,temperature.gpu", "--format=csv,noheader,nounits"], stdout=PIPE) - stdout, stderror = p.communicate() - p = subprocess.run([ - nvidia_smi, - "--query-gpu=index,uuid,utilization.gpu,memory.total,memory.used,memory.free,driver_version,name,gpu_serial,display_active,display_mode,temperature.gpu", - "--format=csv,noheader,nounits" - ], stdout=subprocess.PIPE, encoding='utf8') - stdout, stderror = p.stdout, p.stderr - except: - return [] - output = stdout;#output = stdout.decode('UTF-8') - # output = output[2:-1] # Remove b' and ' from string added by python - #print(output) - output = stdout - ## Parse output - # Split on line break - lines = output.split(os.linesep) - #print(lines) - numDevices = len(lines)-1 - GPUs = [] - for g in range(numDevices): - line = lines[g] - #print(line) - vals = line.split(', ') - #print(vals) - for i in range(12): - # print(vals[i]) - if (i == 0): - deviceIds = int(vals[i]) - elif (i == 1): - uuid = vals[i] - elif (i == 2): - gpuUtil = safeFloatCast(vals[i])/100 - elif (i == 3): - memTotal = safeFloatCast(vals[i]) - elif (i == 4): - memUsed = safeFloatCast(vals[i]) - elif (i == 5): - memFree = safeFloatCast(vals[i]) - elif (i == 6): - driver = vals[i] - elif (i == 7): - gpu_name = vals[i] - elif (i == 8): - serial = vals[i] - elif (i == 9): - display_active = vals[i] - elif (i == 10): - display_mode = vals[i] - elif (i == 11): - temp_gpu = safeFloatCast(vals[i]); - deviceIds = int(vals[0]) - uuid = vals[1] - gpuUtil = safeFloatCast(vals[2]) / 100 - memTotal = safeFloatCast(vals[3]) - memUsed = safeFloatCast(vals[4]) - memFree = safeFloatCast(vals[5]) - driver = vals[6] - gpu_name = vals[7] - serial = vals[8] - display_active = vals[9] - display_mode = vals[10] - temp_gpu = safeFloatCast(vals[11]); - GPUs.append(GPU(deviceIds, uuid, gpuUtil, memTotal, memUsed, memFree, driver, gpu_name, serial, display_mode, display_active, temp_gpu)) - return GPUs # (deviceIds, gpuUtil, memUtil) - - - -def getGPUProcesses(): - """Get all gpu compute processes.""" - - global gpuUuidToIdMap - gpuUuidToIdMap = {} - try: - gpus = getGPUs() - for gpu in gpus: - gpuUuidToIdMap[gpu.uuid] = gpu.id - del gpus - except: - pass - - - nvidia_smi = getNvidiaSmiCmd() - try: - p = subprocess.run([ - nvidia_smi, - "--query-compute-apps=pid,process_name,gpu_uuid,gpu_name,used_memory", - "--format=csv,noheader,nounits" - ], stdout=subprocess.PIPE, encoding='utf8') - stdout, stderror = p.stdout, p.stderr - except: - return [] - output = stdout - ## Parse output - # Split on line break - lines = output.split(os.linesep) - numProcesses = len(lines) - 1 - processes = [] - for g in range(numProcesses): - line = lines[g] - #print(line) - vals = line.split(', ') - #print(vals) - pid = int(vals[0]) - processName = vals[1] - gpuUuid = vals[2] - gpuName = vals[3] - usedMemory = safeFloatCast(vals[4]) - gpuId = gpuUuidToIdMap[gpuUuid] - if gpuId is None: - gpuId = -1 - - # get uid and uname owner of the pid - try: - p = subprocess.run(['ps', f'-p{pid}', '-oruid=,ruser='], - stdout=subprocess.PIPE, encoding='utf8') - uid, uname = p.stdout.split() - uid = int(uid) - except: - uid, uname = -1, '' - - processes.append(GPUProcess(pid, processName, gpuId, gpuUuid, - gpuName, usedMemory, uid, uname)) - return processes - - -def getAvailable(order = 'first', limit=1, maxLoad=0.5, maxMemory=0.5, memoryFree=0, includeNan=False, excludeID=[], excludeUUID=[]): - # order = first | last | random | load | memory - # first --> select the GPU with the lowest ID (DEFAULT) - # last --> select the GPU with the highest ID - # random --> select a random available GPU - # load --> select the GPU with the lowest load - # memory --> select the GPU with the most memory available - # limit = 1 (DEFAULT), 2, ..., Inf - # Limit sets the upper limit for the number of GPUs to return. E.g. if limit = 2, but only one is available, only one is returned. - # Get device IDs, load and memory usage - GPUs = getGPUs() - # Determine, which GPUs are available - GPUavailability = getAvailability(GPUs, maxLoad=maxLoad, maxMemory=maxMemory, memoryFree=memoryFree, includeNan=includeNan, excludeID=excludeID, excludeUUID=excludeUUID) - availAbleGPUindex = [idx for idx in range(0,len(GPUavailability)) if (GPUavailability[idx] == 1)] - # Discard unavailable GPUs - GPUs = [GPUs[g] for g in availAbleGPUindex] - # Sort available GPUs according to the order argument - if (order == 'first'): - GPUs.sort(key=lambda x: float('inf') if math.isnan(x.id) else x.id, reverse=False) - elif (order == 'last'): - GPUs.sort(key=lambda x: float('-inf') if math.isnan(x.id) else x.id, reverse=True) - elif (order == 'random'): - GPUs = [GPUs[g] for g in random.sample(range(0,len(GPUs)),len(GPUs))] - elif (order == 'load'): - GPUs.sort(key=lambda x: float('inf') if math.isnan(x.load) else x.load, reverse=False) - elif (order == 'memory'): - GPUs.sort(key=lambda x: float('inf') if math.isnan(x.memoryUtil) else x.memoryUtil, reverse=False) - # Extract the number of desired GPUs, but limited to the total number of available GPUs - GPUs = GPUs[0:min(limit, len(GPUs))] - # Extract the device IDs from the GPUs and return them - deviceIds = [gpu.id for gpu in GPUs] - return deviceIds -#def getAvailability(GPUs, maxLoad = 0.5, maxMemory = 0.5, includeNan = False): -# # Determine, which GPUs are available -# GPUavailability = np.zeros(len(GPUs)) -# for i in range(len(GPUs)): -# if (GPUs[i].load < maxLoad or (includeNan and np.isnan(GPUs[i].load))) and (GPUs[i].memoryUtil < maxMemory or (includeNan and np.isnan(GPUs[i].memoryUtil))): -# GPUavailability[i] = 1 -def getAvailability(GPUs, maxLoad=0.5, maxMemory=0.5, memoryFree=0, includeNan=False, excludeID=[], excludeUUID=[]): - # Determine, which GPUs are available - GPUavailability = [1 if (gpu.memoryFree>=memoryFree) and (gpu.load < maxLoad or (includeNan and math.isnan(gpu.load))) and (gpu.memoryUtil < maxMemory or (includeNan and math.isnan(gpu.memoryUtil))) and ((gpu.id not in excludeID) and (gpu.uuid not in excludeUUID)) else 0 for gpu in GPUs] - return GPUavailability -def getFirstAvailable(order = 'first', maxLoad=0.5, maxMemory=0.5, attempts=1, interval=900, verbose=False, includeNan=False, excludeID=[], excludeUUID=[]): - #GPUs = getGPUs() - #firstAvailableGPU = np.NaN - #for i in range(len(GPUs)): - # if (GPUs[i].load < maxLoad) & (GPUs[i].memory < maxMemory): - # firstAvailableGPU = GPUs[i].id - # break - #return firstAvailableGPU - for i in range(attempts): - if (verbose): - print('Attempting (' + str(i+1) + '/' + str(attempts) + ') to locate available GPU.') - # Get first available GPU - available = getAvailable(order=order, limit=1, maxLoad=maxLoad, maxMemory=maxMemory, includeNan=includeNan, excludeID=excludeID, excludeUUID=excludeUUID) - # If an available GPU was found, break for loop. - if (available): - if (verbose): - print('GPU ' + str(available) + ' located!') - break - # If this is not the last attempt, sleep for 'interval' seconds - if (i != attempts-1): - time.sleep(interval) - # Check if an GPU was found, or if the attempts simply ran out. Throw error, if no GPU was found - if (not(available)): - raise RuntimeError('Could not find an available GPU after ' + str(attempts) + ' attempts with ' + str(interval) + ' seconds interval.') - # Return found GPU - return available -def showUtilization(all=False, attrList=None, useOldCode=False): - GPUs = getGPUs() - if (all): - if (useOldCode): - print(' ID | Name | Serial | UUID || GPU util. | Memory util. || Memory total | Memory used | Memory free || Display mode | Display active |') - print('------------------------------------------------------------------------------------------------------------------------------') - for gpu in GPUs: - print(' {0:2d} | {1:s} | {2:s} | {3:s} || {4:3.0f}% | {5:3.0f}% || {6:.0f}MB | {7:.0f}MB | {8:.0f}MB || {9:s} | {10:s}'.format(gpu.id,gpu.name,gpu.serial,gpu.uuid,gpu.load*100,gpu.memoryUtil*100,gpu.memoryTotal,gpu.memoryUsed,gpu.memoryFree,gpu.display_mode,gpu.display_active)) - else: - attrList = [[{'attr':'id','name':'ID'}, - {'attr':'name','name':'Name'}, - {'attr':'serial','name':'Serial'}, - {'attr':'uuid','name':'UUID'}], - [{'attr':'temperature','name':'GPU temp.','suffix':'C','transform': lambda x: x,'precision':0}, - {'attr':'load','name':'GPU util.','suffix':'%','transform': lambda x: x*100,'precision':0}, - {'attr':'memoryUtil','name':'Memory util.','suffix':'%','transform': lambda x: x*100,'precision':0}], - [{'attr':'memoryTotal','name':'Memory total','suffix':'MB','precision':0}, - {'attr':'memoryUsed','name':'Memory used','suffix':'MB','precision':0}, - {'attr':'memoryFree','name':'Memory free','suffix':'MB','precision':0}], - [{'attr':'display_mode','name':'Display mode'}, - {'attr':'display_active','name':'Display active'}]] - - else: - if (useOldCode): - print(' ID GPU MEM') - print('--------------') - for gpu in GPUs: - print(' {0:2d} {1:3.0f}% {2:3.0f}%'.format(gpu.id, gpu.load*100, gpu.memoryUtil*100)) - else: - attrList = [[{'attr':'id','name':'ID'}, - {'attr':'load','name':'GPU','suffix':'%','transform': lambda x: x*100,'precision':0}, - {'attr':'memoryUtil','name':'MEM','suffix':'%','transform': lambda x: x*100,'precision':0}], - ] - - if (not useOldCode): - if (attrList is not None): - headerString = '' - GPUstrings = ['']*len(GPUs) - for attrGroup in attrList: - #print(attrGroup) - for attrDict in attrGroup: - headerString = headerString + '| ' + attrDict['name'] + ' ' - headerWidth = len(attrDict['name']) - minWidth = len(attrDict['name']) - - attrPrecision = '.' + str(attrDict['precision']) if ('precision' in attrDict.keys()) else '' - attrSuffix = str(attrDict['suffix']) if ('suffix' in attrDict.keys()) else '' - attrTransform = attrDict['transform'] if ('transform' in attrDict.keys()) else lambda x : x - for gpu in GPUs: - attr = getattr(gpu,attrDict['attr']) - - attr = attrTransform(attr) - - if (isinstance(attr,float)): - attrStr = ('{0:' + attrPrecision + 'f}').format(attr) - elif (isinstance(attr,int)): - attrStr = ('{0:d}').format(attr) - elif (isinstance(attr,str)): - attrStr = attr; - elif (sys.version_info[0] == 2): - if (isinstance(attr,unicode)): - attrStr = attr.encode('ascii','ignore') - else: - raise TypeError('Unhandled object type (' + str(type(attr)) + ') for attribute \'' + attrDict['name'] + '\'') - - attrStr += attrSuffix - - minWidth = max(minWidth,len(attrStr)) - - headerString += ' '*max(0,minWidth-headerWidth) - - minWidthStr = str(minWidth - len(attrSuffix)) - - for gpuIdx,gpu in enumerate(GPUs): - attr = getattr(gpu,attrDict['attr']) - - attr = attrTransform(attr) - - if (isinstance(attr,float)): - attrStr = ('{0:'+ minWidthStr + attrPrecision + 'f}').format(attr) - elif (isinstance(attr,int)): - attrStr = ('{0:' + minWidthStr + 'd}').format(attr) - elif (isinstance(attr,str)): - attrStr = ('{0:' + minWidthStr + 's}').format(attr); - elif (sys.version_info[0] == 2): - if (isinstance(attr,unicode)): - attrStr = ('{0:' + minWidthStr + 's}').format(attr.encode('ascii','ignore')) - else: - raise TypeError('Unhandled object type (' + str(type(attr)) + ') for attribute \'' + attrDict['name'] + '\'') - - attrStr += attrSuffix - - GPUstrings[gpuIdx] += '| ' + attrStr + ' ' - - headerString = headerString + '|' - for gpuIdx,gpu in enumerate(GPUs): - GPUstrings[gpuIdx] += '|' - - headerSpacingString = '-' * len(headerString) - print(headerString) - print(headerSpacingString) - for GPUstring in GPUstrings: - print(GPUstring) - - -# Generate gpu uuid to id map -gpuUuidToIdMap = {} -try: - gpus = getGPUs() - for gpu in gpus: - gpuUuidToIdMap[gpu.uuid] = gpu.id - del gpus -except: - pass -def getGPUInfos(): - ###返回gpus：list,一个GPU为一个元素-对象 - ###########：有属性，'id','load','memoryFree', - ###########：'memoryTotal','memoryUsed','memoryUtil','name','serial''temperature','uuid',process - ###其中process：每一个计算进程是一个元素--对象 - ############：有属性，'gpuId','gpuName','gpuUuid', - ############：'gpuid','pid','processName','uid', 'uname','usedMemory' - gpus = getGPUs() - gpuUuidToIdMap={} - for gpu in gpus: - gpuUuidToIdMap[gpu.uuid] = gpu.id - gpu.process=[] - indexx = [x.id for x in gpus ] - - process = getGPUProcesses() - for pre in process: - pre.gpuid = gpuUuidToIdMap[pre.gpuUuid] - gpuId = indexx.index(pre.gpuid ) - gpus[gpuId].process.append(pre ) - return gpus - -def get_available_gpu(gpuStatus): - ##判断是否有空闲的显卡，如果有返回id，没有返回None - cuda=None - for gpus in gpuStatus: - if len(gpus.process) == 0: - cuda = gpus.id - return cuda - return cuda -def get_whether_gpuProcess(): - ##判断是否有空闲的显卡，如果有返回id，没有返回None - gpuStatus=getGPUInfos() - gpuProcess=True - for gpus in gpuStatus: - if len(gpus.process) != 0: - gpuProcess = False - return gpuProcess - -def get_offlineProcess_gpu(gpuStatus,pidInfos): - gpu_onLine = [] - for gpu in gpuStatus: - for gpuProcess in gpu.process: - pid = gpuProcess.pid - if pid in pidInfos.keys(): - pidType = pidInfos[pid]['type'] - if pidType == 'onLine': - gpu_onLine.append(gpu) - gpu_offLine = set(gpuStatus) - set(gpu_onLine) - return list(gpu_offLine) -def arrange_offlineProcess(gpuStatus,pidInfos,modelMemory=1500): - cudaArrange=[] - gpu_offLine = get_offlineProcess_gpu(gpuStatus,pidInfos) - for gpu in gpu_offLine: - leftMemory = gpu.memoryTotal*0.9 - gpu.memoryUsed - modelCnt = int(leftMemory// modelMemory) - - cudaArrange.extend( [gpu.id] * modelCnt ) - return cudaArrange -def get_potential_gpu(gpuStatus,pidInfos): - ###所有GPU上都有计算。需要为“在线任务”空出一块显卡。 - ###step1：查看所有显卡上是否有“在线任务” - - gpu_offLine = get_offlineProcess_gpu(gpuStatus,pidInfos) - if len(gpu_offLine) == 0 : - return False - - ###step2,找出每张显卡上离线进程的数目 - offLineCnt = [ len(gpu.process) for gpu in gpu_offLine ] - minCntIndex =offLineCnt.index( min(offLineCnt)) - - pids = [x.pid for x in gpu_offLine[minCntIndex].process] - return {'cuda':gpu_offLine[minCntIndex].id,'pids':pids } -if __name__=='__main__': - #pres = getGPUProcesses() - #print('###line404:',pres) - gpus = getGPUs() - for gpu in gpus: - gpuUuidToIdMap[gpu.uuid] = gpu.id - print(gpu) - print(gpuUuidToIdMap) - pres = getGPUProcesses() - print('###line404:',pres) - for pre in pres: - print('#'*20) - for ken in ['gpuName','gpuUuid','pid','processName','uid','uname','usedMemory' ]: - print(ken,' ',pre.__getattribute__(ken )) - print(' ') - - diff --git a/AIlib/segutils/__pycache__/GPUtils.cpython-38.pyc b/AIlib/segutils/__pycache__/GPUtils.cpython-38.pyc deleted file mode 100644 index 0873cfb..0000000 Binary files a/AIlib/segutils/__pycache__/GPUtils.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/__pycache__/segWaterBuilding.cpython-38.pyc b/AIlib/segutils/__pycache__/segWaterBuilding.cpython-38.pyc deleted file mode 100644 index 537afb9..0000000 Binary files a/AIlib/segutils/__pycache__/segWaterBuilding.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/__pycache__/segmodel.cpython-38.pyc b/AIlib/segutils/__pycache__/segmodel.cpython-38.pyc deleted file mode 100644 index 5cbd6cd..0000000 Binary files a/AIlib/segutils/__pycache__/segmodel.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/__init__.py b/AIlib/segutils/core/__init__.py deleted file mode 100644 index 453f410..0000000 --- a/AIlib/segutils/core/__init__.py +++ /dev/null @@ -1 +0,0 @@ -from . import nn, models, utils, data \ No newline at end of file diff --git a/AIlib/segutils/core/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index 6ab6e7d..0000000 Binary files a/AIlib/segutils/core/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/__pycache__/__init__.cpython-38.pyc b/AIlib/segutils/core/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index 07f6414..0000000 Binary files a/AIlib/segutils/core/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/__init__.py b/AIlib/segutils/core/data/__init__.py deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/segutils/core/data/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/data/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index c18d558..0000000 Binary files a/AIlib/segutils/core/data/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/__pycache__/__init__.cpython-38.pyc b/AIlib/segutils/core/data/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index 263da84..0000000 Binary files a/AIlib/segutils/core/data/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/dataloader/__init__.py b/AIlib/segutils/core/data/dataloader/__init__.py deleted file mode 100644 index b22f962..0000000 --- a/AIlib/segutils/core/data/dataloader/__init__.py +++ /dev/null @@ -1,23 +0,0 @@ -""" -This module provides data loaders and transformers for popular vision datasets. -""" -from .mscoco import COCOSegmentation -from .cityscapes import CitySegmentation -from .ade import ADE20KSegmentation -from .pascal_voc import VOCSegmentation -from .pascal_aug import VOCAugSegmentation -from .sbu_shadow import SBUSegmentation - -datasets = { - 'ade20k': ADE20KSegmentation, - 'pascal_voc': VOCSegmentation, - 'pascal_aug': VOCAugSegmentation, - 'coco': COCOSegmentation, - 'citys': CitySegmentation, - 'sbu': SBUSegmentation, -} - - -def get_segmentation_dataset(name, **kwargs): - """Segmentation Datasets""" - return datasets[name.lower()](**kwargs) diff --git a/AIlib/segutils/core/data/dataloader/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/data/dataloader/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index 5029d98..0000000 Binary files a/AIlib/segutils/core/data/dataloader/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/dataloader/__pycache__/ade.cpython-36.pyc b/AIlib/segutils/core/data/dataloader/__pycache__/ade.cpython-36.pyc deleted file mode 100644 index 34b59dd..0000000 Binary files a/AIlib/segutils/core/data/dataloader/__pycache__/ade.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/dataloader/__pycache__/cityscapes.cpython-36.pyc b/AIlib/segutils/core/data/dataloader/__pycache__/cityscapes.cpython-36.pyc deleted file mode 100644 index bdff2fc..0000000 Binary files a/AIlib/segutils/core/data/dataloader/__pycache__/cityscapes.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/dataloader/__pycache__/mscoco.cpython-36.pyc b/AIlib/segutils/core/data/dataloader/__pycache__/mscoco.cpython-36.pyc deleted file mode 100644 index 2c0d2e3..0000000 Binary files a/AIlib/segutils/core/data/dataloader/__pycache__/mscoco.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/dataloader/__pycache__/pascal_aug.cpython-36.pyc b/AIlib/segutils/core/data/dataloader/__pycache__/pascal_aug.cpython-36.pyc deleted file mode 100644 index fa3e95f..0000000 Binary files a/AIlib/segutils/core/data/dataloader/__pycache__/pascal_aug.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/dataloader/__pycache__/pascal_voc.cpython-36.pyc b/AIlib/segutils/core/data/dataloader/__pycache__/pascal_voc.cpython-36.pyc deleted file mode 100644 index 1743082..0000000 Binary files a/AIlib/segutils/core/data/dataloader/__pycache__/pascal_voc.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/dataloader/__pycache__/sbu_shadow.cpython-36.pyc b/AIlib/segutils/core/data/dataloader/__pycache__/sbu_shadow.cpython-36.pyc deleted file mode 100644 index 1c3b63f..0000000 Binary files a/AIlib/segutils/core/data/dataloader/__pycache__/sbu_shadow.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/dataloader/__pycache__/segbase.cpython-36.pyc b/AIlib/segutils/core/data/dataloader/__pycache__/segbase.cpython-36.pyc deleted file mode 100644 index e5e253b..0000000 Binary files a/AIlib/segutils/core/data/dataloader/__pycache__/segbase.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/data/dataloader/ade.py b/AIlib/segutils/core/data/dataloader/ade.py deleted file mode 100644 index 522ecbd..0000000 --- a/AIlib/segutils/core/data/dataloader/ade.py +++ /dev/null @@ -1,172 +0,0 @@ -"""Pascal ADE20K Semantic Segmentation Dataset.""" -import os -import torch -import numpy as np - -from PIL import Image -from .segbase import SegmentationDataset - - -class ADE20KSegmentation(SegmentationDataset): - """ADE20K Semantic Segmentation Dataset. - - Parameters - ---------- - root : string - Path to ADE20K folder. Default is './datasets/ade' - split: string - 'train', 'val' or 'test' - transform : callable, optional - A function that transforms the image - Examples - -------- - >>> from torchvision import transforms - >>> import torch.utils.data as data - >>> # Transforms for Normalization - >>> input_transform = transforms.Compose([ - >>> transforms.ToTensor(), - >>> transforms.Normalize((.485, .456, .406), (.229, .224, .225)), - >>> ]) - >>> # Create Dataset - >>> trainset = ADE20KSegmentation(split='train', transform=input_transform) - >>> # Create Training Loader - >>> train_data = data.DataLoader( - >>> trainset, 4, shuffle=True, - >>> num_workers=4) - """ - BASE_DIR = 'ADEChallengeData2016' - NUM_CLASS = 150 - - def __init__(self, root='../datasets/ade', split='test', mode=None, transform=None, **kwargs): - super(ADE20KSegmentation, self).__init__(root, split, mode, transform, **kwargs) - root = os.path.join(root, self.BASE_DIR) - assert os.path.exists(root), "Please setup the dataset using ../datasets/ade20k.py" - self.images, self.masks = _get_ade20k_pairs(root, split) - assert (len(self.images) == len(self.masks)) - if len(self.images) == 0: - raise RuntimeError("Found 0 images in subfolders of:" + root + "\n") - print('Found {} images in the folder {}'.format(len(self.images), root)) - - def __getitem__(self, index): - img = Image.open(self.images[index]).convert('RGB') - if self.mode == 'test': - img = self._img_transform(img) - if self.transform is not None: - img = self.transform(img) - return img, os.path.basename(self.images[index]) - mask = Image.open(self.masks[index]) - # synchrosized transform - if self.mode == 'train': - img, mask = self._sync_transform(img, mask) - elif self.mode == 'val': - img, mask = self._val_sync_transform(img, mask) - else: - assert self.mode == 'testval' - img, mask = self._img_transform(img), self._mask_transform(mask) - # general resize, normalize and to Tensor - if self.transform is not None: - img = self.transform(img) - return img, mask, os.path.basename(self.images[index]) - - def _mask_transform(self, mask): - return torch.LongTensor(np.array(mask).astype('int32') - 1) - - def __len__(self): - return len(self.images) - - @property - def pred_offset(self): - return 1 - - @property - def classes(self): - """Category names.""" - return ("wall", "building, edifice", "sky", "floor, flooring", "tree", - "ceiling", "road, route", "bed", "windowpane, window", "grass", - "cabinet", "sidewalk, pavement", - "person, individual, someone, somebody, mortal, soul", - "earth, ground", "door, double door", "table", "mountain, mount", - "plant, flora, plant life", "curtain, drape, drapery, mantle, pall", - "chair", "car, auto, automobile, machine, motorcar", - "water", "painting, picture", "sofa, couch, lounge", "shelf", - "house", "sea", "mirror", "rug, carpet, carpeting", "field", "armchair", - "seat", "fence, fencing", "desk", "rock, stone", "wardrobe, closet, press", - "lamp", "bathtub, bathing tub, bath, tub", "railing, rail", "cushion", - "base, pedestal, stand", "box", "column, pillar", "signboard, sign", - "chest of drawers, chest, bureau, dresser", "counter", "sand", "sink", - "skyscraper", "fireplace, hearth, open fireplace", "refrigerator, icebox", - "grandstand, covered stand", "path", "stairs, steps", "runway", - "case, display case, showcase, vitrine", - "pool table, billiard table, snooker table", "pillow", - "screen door, screen", "stairway, staircase", "river", "bridge, span", - "bookcase", "blind, screen", "coffee table, cocktail table", - "toilet, can, commode, crapper, pot, potty, stool, throne", - "flower", "book", "hill", "bench", "countertop", - "stove, kitchen stove, range, kitchen range, cooking stove", - "palm, palm tree", "kitchen island", - "computer, computing machine, computing device, data processor, " - "electronic computer, information processing system", - "swivel chair", "boat", "bar", "arcade machine", - "hovel, hut, hutch, shack, shanty", - "bus, autobus, coach, charabanc, double-decker, jitney, motorbus, " - "motorcoach, omnibus, passenger vehicle", - "towel", "light, light source", "truck, motortruck", "tower", - "chandelier, pendant, pendent", "awning, sunshade, sunblind", - "streetlight, street lamp", "booth, cubicle, stall, kiosk", - "television receiver, television, television set, tv, tv set, idiot " - "box, boob tube, telly, goggle box", - "airplane, aeroplane, plane", "dirt track", - "apparel, wearing apparel, dress, clothes", - "pole", "land, ground, soil", - "bannister, banister, balustrade, balusters, handrail", - "escalator, moving staircase, moving stairway", - "ottoman, pouf, pouffe, puff, hassock", - "bottle", "buffet, counter, sideboard", - "poster, posting, placard, notice, bill, card", - "stage", "van", "ship", "fountain", - "conveyer belt, conveyor belt, conveyer, conveyor, transporter", - "canopy", "washer, automatic washer, washing machine", - "plaything, toy", "swimming pool, swimming bath, natatorium", - "stool", "barrel, cask", "basket, handbasket", "waterfall, falls", - "tent, collapsible shelter", "bag", "minibike, motorbike", "cradle", - "oven", "ball", "food, solid food", "step, stair", "tank, storage tank", - "trade name, brand name, brand, marque", "microwave, microwave oven", - "pot, flowerpot", "animal, animate being, beast, brute, creature, fauna", - "bicycle, bike, wheel, cycle", "lake", - "dishwasher, dish washer, dishwashing machine", - "screen, silver screen, projection screen", - "blanket, cover", "sculpture", "hood, exhaust hood", "sconce", "vase", - "traffic light, traffic signal, stoplight", "tray", - "ashcan, trash can, garbage can, wastebin, ash bin, ash-bin, ashbin, " - "dustbin, trash barrel, trash bin", - "fan", "pier, wharf, wharfage, dock", "crt screen", - "plate", "monitor, monitoring device", "bulletin board, notice board", - "shower", "radiator", "glass, drinking glass", "clock", "flag") - - -def _get_ade20k_pairs(folder, mode='train'): - img_paths = [] - mask_paths = [] - if mode == 'train': - img_folder = os.path.join(folder, 'images/training') - mask_folder = os.path.join(folder, 'annotations/training') - else: - img_folder = os.path.join(folder, 'images/validation') - mask_folder = os.path.join(folder, 'annotations/validation') - for filename in os.listdir(img_folder): - basename, _ = os.path.splitext(filename) - if filename.endswith(".jpg"): - imgpath = os.path.join(img_folder, filename) - maskname = basename + '.png' - maskpath = os.path.join(mask_folder, maskname) - if os.path.isfile(maskpath): - img_paths.append(imgpath) - mask_paths.append(maskpath) - else: - print('cannot find the mask:', maskpath) - - return img_paths, mask_paths - - -if __name__ == '__main__': - train_dataset = ADE20KSegmentation() diff --git a/AIlib/segutils/core/data/dataloader/cityscapes.py b/AIlib/segutils/core/data/dataloader/cityscapes.py deleted file mode 100644 index 7d5de71..0000000 --- a/AIlib/segutils/core/data/dataloader/cityscapes.py +++ /dev/null @@ -1,137 +0,0 @@ -"""Prepare Cityscapes dataset""" -import os -import torch -import numpy as np - -from PIL import Image -from .segbase import SegmentationDataset - - -class CitySegmentation(SegmentationDataset): - """Cityscapes Semantic Segmentation Dataset. - - Parameters - ---------- - root : string - Path to Cityscapes folder. Default is './datasets/citys' - split: string - 'train', 'val' or 'test' - transform : callable, optional - A function that transforms the image - Examples - -------- - >>> from torchvision import transforms - >>> import torch.utils.data as data - >>> # Transforms for Normalization - >>> input_transform = transforms.Compose([ - >>> transforms.ToTensor(), - >>> transforms.Normalize((.485, .456, .406), (.229, .224, .225)), - >>> ]) - >>> # Create Dataset - >>> trainset = CitySegmentation(split='train', transform=input_transform) - >>> # Create Training Loader - >>> train_data = data.DataLoader( - >>> trainset, 4, shuffle=True, - >>> num_workers=4) - """ - BASE_DIR = 'cityscapes' - NUM_CLASS = 19 - - def __init__(self, root='../datasets/citys', split='train', mode=None, transform=None, **kwargs): - super(CitySegmentation, self).__init__(root, split, mode, transform, **kwargs) - # self.root = os.path.join(root, self.BASE_DIR) - assert os.path.exists(self.root), "Please setup the dataset using ../datasets/cityscapes.py" - self.images, self.mask_paths = _get_city_pairs(self.root, self.split) - assert (len(self.images) == len(self.mask_paths)) - if len(self.images) == 0: - raise RuntimeError("Found 0 images in subfolders of:" + root + "\n") - self.valid_classes = [7, 8, 11, 12, 13, 17, 19, 20, 21, 22, - 23, 24, 25, 26, 27, 28, 31, 32, 33] - self._key = np.array([-1, -1, -1, -1, -1, -1, - -1, -1, 0, 1, -1, -1, - 2, 3, 4, -1, -1, -1, - 5, -1, 6, 7, 8, 9, - 10, 11, 12, 13, 14, 15, - -1, -1, 16, 17, 18]) - self._mapping = np.array(range(-1, len(self._key) - 1)).astype('int32') - - def _class_to_index(self, mask): - # assert the value - values = np.unique(mask) - for value in values: - assert (value in self._mapping) - index = np.digitize(mask.ravel(), self._mapping, right=True) - return self._key[index].reshape(mask.shape) - - def __getitem__(self, index): - img = Image.open(self.images[index]).convert('RGB') - if self.mode == 'test': - if self.transform is not None: - img = self.transform(img) - return img, os.path.basename(self.images[index]) - mask = Image.open(self.mask_paths[index]) - # synchrosized transform - if self.mode == 'train': - img, mask = self._sync_transform(img, mask) - elif self.mode == 'val': - img, mask = self._val_sync_transform(img, mask) - else: - assert self.mode == 'testval' - img, mask = self._img_transform(img), self._mask_transform(mask) - # general resize, normalize and toTensor - if self.transform is not None: - img = self.transform(img) - return img, mask, os.path.basename(self.images[index]) - - def _mask_transform(self, mask): - target = self._class_to_index(np.array(mask).astype('int32')) - return torch.LongTensor(np.array(target).astype('int32')) - - def __len__(self): - return len(self.images) - - @property - def pred_offset(self): - return 0 - - -def _get_city_pairs(folder, split='train'): - def get_path_pairs(img_folder, mask_folder): - img_paths = [] - mask_paths = [] - for root, _, files in os.walk(img_folder): - for filename in files: - if filename.endswith('.png'): - imgpath = os.path.join(root, filename) - foldername = os.path.basename(os.path.dirname(imgpath)) - maskname = filename.replace('leftImg8bit', 'gtFine_labelIds') - maskpath = os.path.join(mask_folder, foldername, maskname) - if os.path.isfile(imgpath) and os.path.isfile(maskpath): - img_paths.append(imgpath) - mask_paths.append(maskpath) - else: - print('cannot find the mask or image:', imgpath, maskpath) - print('Found {} images in the folder {}'.format(len(img_paths), img_folder)) - return img_paths, mask_paths - - if split in ('train', 'val'): - img_folder = os.path.join(folder, 'leftImg8bit/' + split) - mask_folder = os.path.join(folder, 'gtFine/' + split) - img_paths, mask_paths = get_path_pairs(img_folder, mask_folder) - return img_paths, mask_paths - else: - assert split == 'trainval' - print('trainval set') - train_img_folder = os.path.join(folder, 'leftImg8bit/train') - train_mask_folder = os.path.join(folder, 'gtFine/train') - val_img_folder = os.path.join(folder, 'leftImg8bit/val') - val_mask_folder = os.path.join(folder, 'gtFine/val') - train_img_paths, train_mask_paths = get_path_pairs(train_img_folder, train_mask_folder) - val_img_paths, val_mask_paths = get_path_pairs(val_img_folder, val_mask_folder) - img_paths = train_img_paths + val_img_paths - mask_paths = train_mask_paths + val_mask_paths - return img_paths, mask_paths - - -if __name__ == '__main__': - dataset = CitySegmentation() diff --git a/AIlib/segutils/core/data/dataloader/lip_parsing.py b/AIlib/segutils/core/data/dataloader/lip_parsing.py deleted file mode 100644 index 245beda..0000000 --- a/AIlib/segutils/core/data/dataloader/lip_parsing.py +++ /dev/null @@ -1,90 +0,0 @@ -"""Look into Person Dataset""" -import os -import torch -import numpy as np - -from PIL import Image -from core.data.dataloader.segbase import SegmentationDataset - - -class LIPSegmentation(SegmentationDataset): - """Look into person parsing dataset """ - - BASE_DIR = 'LIP' - NUM_CLASS = 20 - - def __init__(self, root='../datasets/LIP', split='train', mode=None, transform=None, **kwargs): - super(LIPSegmentation, self).__init__(root, split, mode, transform, **kwargs) - _trainval_image_dir = os.path.join(root, 'TrainVal_images') - _testing_image_dir = os.path.join(root, 'Testing_images') - _trainval_mask_dir = os.path.join(root, 'TrainVal_parsing_annotations') - if split == 'train': - _image_dir = os.path.join(_trainval_image_dir, 'train_images') - _mask_dir = os.path.join(_trainval_mask_dir, 'train_segmentations') - _split_f = os.path.join(_trainval_image_dir, 'train_id.txt') - elif split == 'val': - _image_dir = os.path.join(_trainval_image_dir, 'val_images') - _mask_dir = os.path.join(_trainval_mask_dir, 'val_segmentations') - _split_f = os.path.join(_trainval_image_dir, 'val_id.txt') - elif split == 'test': - _image_dir = os.path.join(_testing_image_dir, 'testing_images') - _split_f = os.path.join(_testing_image_dir, 'test_id.txt') - else: - raise RuntimeError('Unknown dataset split.') - - self.images = [] - self.masks = [] - with open(os.path.join(_split_f), 'r') as lines: - for line in lines: - _image = os.path.join(_image_dir, line.rstrip('\n') + '.jpg') - assert os.path.isfile(_image) - self.images.append(_image) - if split != 'test': - _mask = os.path.join(_mask_dir, line.rstrip('\n') + '.png') - assert os.path.isfile(_mask) - self.masks.append(_mask) - - if split != 'test': - assert (len(self.images) == len(self.masks)) - print('Found {} {} images in the folder {}'.format(len(self.images), split, root)) - - def __getitem__(self, index): - img = Image.open(self.images[index]).convert('RGB') - if self.mode == 'test': - img = self._img_transform(img) - if self.transform is not None: - img = self.transform(img) - return img, os.path.basename(self.images[index]) - mask = Image.open(self.masks[index]) - # synchronized transform - if self.mode == 'train': - img, mask = self._sync_transform(img, mask) - elif self.mode == 'val': - img, mask = self._val_sync_transform(img, mask) - else: - assert self.mode == 'testval' - img, mask = self._img_transform(img), self._mask_transform(mask) - # general resize, normalize and toTensor - if self.transform is not None: - img = self.transform(img) - - return img, mask, os.path.basename(self.images[index]) - - def __len__(self): - return len(self.images) - - def _mask_transform(self, mask): - target = np.array(mask).astype('int32') - return torch.from_numpy(target).long() - - @property - def classes(self): - """Category name.""" - return ('background', 'hat', 'hair', 'glove', 'sunglasses', 'upperclothes', - 'dress', 'coat', 'socks', 'pants', 'jumpsuits', 'scarf', 'skirt', - 'face', 'leftArm', 'rightArm', 'leftLeg', 'rightLeg', 'leftShoe', - 'rightShoe') - - -if __name__ == '__main__': - dataset = LIPSegmentation(base_size=280, crop_size=256) \ No newline at end of file diff --git a/AIlib/segutils/core/data/dataloader/mscoco.py b/AIlib/segutils/core/data/dataloader/mscoco.py deleted file mode 100644 index 6e280c8..0000000 --- a/AIlib/segutils/core/data/dataloader/mscoco.py +++ /dev/null @@ -1,136 +0,0 @@ -"""MSCOCO Semantic Segmentation pretraining for VOC.""" -import os -import pickle -import torch -import numpy as np - -from tqdm import trange -from PIL import Image -from .segbase import SegmentationDataset - - -class COCOSegmentation(SegmentationDataset): - """COCO Semantic Segmentation Dataset for VOC Pre-training. - - Parameters - ---------- - root : string - Path to ADE20K folder. Default is './datasets/coco' - split: string - 'train', 'val' or 'test' - transform : callable, optional - A function that transforms the image - Examples - -------- - >>> from torchvision import transforms - >>> import torch.utils.data as data - >>> # Transforms for Normalization - >>> input_transform = transforms.Compose([ - >>> transforms.ToTensor(), - >>> transforms.Normalize((.485, .456, .406), (.229, .224, .225)), - >>> ]) - >>> # Create Dataset - >>> trainset = COCOSegmentation(split='train', transform=input_transform) - >>> # Create Training Loader - >>> train_data = data.DataLoader( - >>> trainset, 4, shuffle=True, - >>> num_workers=4) - """ - CAT_LIST = [0, 5, 2, 16, 9, 44, 6, 3, 17, 62, 21, 67, 18, 19, 4, - 1, 64, 20, 63, 7, 72] - NUM_CLASS = 21 - - def __init__(self, root='../datasets/coco', split='train', mode=None, transform=None, **kwargs): - super(COCOSegmentation, self).__init__(root, split, mode, transform, **kwargs) - # lazy import pycocotools - from pycocotools.coco import COCO - from pycocotools import mask - if split == 'train': - print('train set') - ann_file = os.path.join(root, 'annotations/instances_train2017.json') - ids_file = os.path.join(root, 'annotations/train_ids.mx') - self.root = os.path.join(root, 'train2017') - else: - print('val set') - ann_file = os.path.join(root, 'annotations/instances_val2017.json') - ids_file = os.path.join(root, 'annotations/val_ids.mx') - self.root = os.path.join(root, 'val2017') - self.coco = COCO(ann_file) - self.coco_mask = mask - if os.path.exists(ids_file): - with open(ids_file, 'rb') as f: - self.ids = pickle.load(f) - else: - ids = list(self.coco.imgs.keys()) - self.ids = self._preprocess(ids, ids_file) - self.transform = transform - - def __getitem__(self, index): - coco = self.coco - img_id = self.ids[index] - img_metadata = coco.loadImgs(img_id)[0] - path = img_metadata['file_name'] - img = Image.open(os.path.join(self.root, path)).convert('RGB') - cocotarget = coco.loadAnns(coco.getAnnIds(imgIds=img_id)) - mask = Image.fromarray(self._gen_seg_mask( - cocotarget, img_metadata['height'], img_metadata['width'])) - # synchrosized transform - if self.mode == 'train': - img, mask = self._sync_transform(img, mask) - elif self.mode == 'val': - img, mask = self._val_sync_transform(img, mask) - else: - assert self.mode == 'testval' - img, mask = self._img_transform(img), self._mask_transform(mask) - # general resize, normalize and toTensor - if self.transform is not None: - img = self.transform(img) - return img, mask, os.path.basename(self.ids[index]) - - def _mask_transform(self, mask): - return torch.LongTensor(np.array(mask).astype('int32')) - - def _gen_seg_mask(self, target, h, w): - mask = np.zeros((h, w), dtype=np.uint8) - coco_mask = self.coco_mask - for instance in target: - rle = coco_mask.frPyObjects(instance['Segmentation'], h, w) - m = coco_mask.decode(rle) - cat = instance['category_id'] - if cat in self.CAT_LIST: - c = self.CAT_LIST.index(cat) - else: - continue - if len(m.shape) < 3: - mask[:, :] += (mask == 0) * (m * c) - else: - mask[:, :] += (mask == 0) * (((np.sum(m, axis=2)) > 0) * c).astype(np.uint8) - return mask - - def _preprocess(self, ids, ids_file): - print("Preprocessing mask, this will take a while." + \ - "But don't worry, it only run once for each split.") - tbar = trange(len(ids)) - new_ids = [] - for i in tbar: - img_id = ids[i] - cocotarget = self.coco.loadAnns(self.coco.getAnnIds(imgIds=img_id)) - img_metadata = self.coco.loadImgs(img_id)[0] - mask = self._gen_seg_mask(cocotarget, img_metadata['height'], img_metadata['width']) - # more than 1k pixels - if (mask > 0).sum() > 1000: - new_ids.append(img_id) - tbar.set_description('Doing: {}/{}, got {} qualified images'. \ - format(i, len(ids), len(new_ids))) - print('Found number of qualified images: ', len(new_ids)) - with open(ids_file, 'wb') as f: - pickle.dump(new_ids, f) - return new_ids - - @property - def classes(self): - """Category names.""" - return ('background', 'airplane', 'bicycle', 'bird', 'boat', 'bottle', - 'bus', 'car', 'cat', 'chair', 'cow', 'diningtable', 'dog', 'horse', - 'motorcycle', 'person', 'potted-plant', 'sheep', 'sofa', 'train', - 'tv') diff --git a/AIlib/segutils/core/data/dataloader/pascal_aug.py b/AIlib/segutils/core/data/dataloader/pascal_aug.py deleted file mode 100644 index 1cbe238..0000000 --- a/AIlib/segutils/core/data/dataloader/pascal_aug.py +++ /dev/null @@ -1,104 +0,0 @@ -"""Pascal Augmented VOC Semantic Segmentation Dataset.""" -import os -import torch -import scipy.io as sio -import numpy as np - -from PIL import Image -from .segbase import SegmentationDataset - - -class VOCAugSegmentation(SegmentationDataset): - """Pascal VOC Augmented Semantic Segmentation Dataset. - - Parameters - ---------- - root : string - Path to VOCdevkit folder. Default is './datasets/voc' - split: string - 'train', 'val' or 'test' - transform : callable, optional - A function that transforms the image - Examples - -------- - >>> from torchvision import transforms - >>> import torch.utils.data as data - >>> # Transforms for Normalization - >>> input_transform = transforms.Compose([ - >>> transforms.ToTensor(), - >>> transforms.Normalize([.485, .456, .406], [.229, .224, .225]), - >>> ]) - >>> # Create Dataset - >>> trainset = VOCAugSegmentation(split='train', transform=input_transform) - >>> # Create Training Loader - >>> train_data = data.DataLoader( - >>> trainset, 4, shuffle=True, - >>> num_workers=4) - """ - BASE_DIR = 'VOCaug/dataset/' - NUM_CLASS = 21 - - def __init__(self, root='../datasets/voc', split='train', mode=None, transform=None, **kwargs): - super(VOCAugSegmentation, self).__init__(root, split, mode, transform, **kwargs) - # train/val/test splits are pre-cut - _voc_root = os.path.join(root, self.BASE_DIR) - _mask_dir = os.path.join(_voc_root, 'cls') - _image_dir = os.path.join(_voc_root, 'img') - if split == 'train': - _split_f = os.path.join(_voc_root, 'trainval.txt') - elif split == 'val': - _split_f = os.path.join(_voc_root, 'val.txt') - else: - raise RuntimeError('Unknown dataset split: {}'.format(split)) - - self.images = [] - self.masks = [] - with open(os.path.join(_split_f), "r") as lines: - for line in lines: - _image = os.path.join(_image_dir, line.rstrip('\n') + ".jpg") - assert os.path.isfile(_image) - self.images.append(_image) - _mask = os.path.join(_mask_dir, line.rstrip('\n') + ".mat") - assert os.path.isfile(_mask) - self.masks.append(_mask) - - assert (len(self.images) == len(self.masks)) - print('Found {} images in the folder {}'.format(len(self.images), _voc_root)) - - def __getitem__(self, index): - img = Image.open(self.images[index]).convert('RGB') - target = self._load_mat(self.masks[index]) - # synchrosized transform - if self.mode == 'train': - img, target = self._sync_transform(img, target) - elif self.mode == 'val': - img, target = self._val_sync_transform(img, target) - else: - raise RuntimeError('unknown mode for dataloader: {}'.format(self.mode)) - # general resize, normalize and toTensor - if self.transform is not None: - img = self.transform(img) - return img, target, os.path.basename(self.images[index]) - - def _mask_transform(self, mask): - return torch.LongTensor(np.array(mask).astype('int32')) - - def _load_mat(self, filename): - mat = sio.loadmat(filename, mat_dtype=True, squeeze_me=True, struct_as_record=False) - mask = mat['GTcls'].Segmentation - return Image.fromarray(mask) - - def __len__(self): - return len(self.images) - - @property - def classes(self): - """Category names.""" - return ('background', 'airplane', 'bicycle', 'bird', 'boat', 'bottle', - 'bus', 'car', 'cat', 'chair', 'cow', 'diningtable', 'dog', 'horse', - 'motorcycle', 'person', 'potted-plant', 'sheep', 'sofa', 'train', - 'tv') - - -if __name__ == '__main__': - dataset = VOCAugSegmentation() \ No newline at end of file diff --git a/AIlib/segutils/core/data/dataloader/pascal_voc.py b/AIlib/segutils/core/data/dataloader/pascal_voc.py deleted file mode 100644 index 94db82c..0000000 --- a/AIlib/segutils/core/data/dataloader/pascal_voc.py +++ /dev/null @@ -1,112 +0,0 @@ -"""Pascal VOC Semantic Segmentation Dataset.""" -import os -import torch -import numpy as np - -from PIL import Image -from .segbase import SegmentationDataset - - -class VOCSegmentation(SegmentationDataset): - """Pascal VOC Semantic Segmentation Dataset. - - Parameters - ---------- - root : string - Path to VOCdevkit folder. Default is './datasets/VOCdevkit' - split: string - 'train', 'val' or 'test' - transform : callable, optional - A function that transforms the image - Examples - -------- - >>> from torchvision import transforms - >>> import torch.utils.data as data - >>> # Transforms for Normalization - >>> input_transform = transforms.Compose([ - >>> transforms.ToTensor(), - >>> transforms.Normalize([.485, .456, .406], [.229, .224, .225]), - >>> ]) - >>> # Create Dataset - >>> trainset = VOCSegmentation(split='train', transform=input_transform) - >>> # Create Training Loader - >>> train_data = data.DataLoader( - >>> trainset, 4, shuffle=True, - >>> num_workers=4) - """ - BASE_DIR = 'VOC2012' - NUM_CLASS = 21 - - def __init__(self, root='../datasets/voc', split='train', mode=None, transform=None, **kwargs): - super(VOCSegmentation, self).__init__(root, split, mode, transform, **kwargs) - _voc_root = os.path.join(root, self.BASE_DIR) - _mask_dir = os.path.join(_voc_root, 'SegmentationClass') - _image_dir = os.path.join(_voc_root, 'JPEGImages') - # train/val/test splits are pre-cut - _splits_dir = os.path.join(_voc_root, 'ImageSets/Segmentation') - if split == 'train': - _split_f = os.path.join(_splits_dir, 'train.txt') - elif split == 'val': - _split_f = os.path.join(_splits_dir, 'val.txt') - elif split == 'test': - _split_f = os.path.join(_splits_dir, 'test.txt') - else: - raise RuntimeError('Unknown dataset split.') - - self.images = [] - self.masks = [] - with open(os.path.join(_split_f), "r") as lines: - for line in lines: - _image = os.path.join(_image_dir, line.rstrip('\n') + ".jpg") - assert os.path.isfile(_image) - self.images.append(_image) - if split != 'test': - _mask = os.path.join(_mask_dir, line.rstrip('\n') + ".png") - assert os.path.isfile(_mask) - self.masks.append(_mask) - - if split != 'test': - assert (len(self.images) == len(self.masks)) - print('Found {} images in the folder {}'.format(len(self.images), _voc_root)) - - def __getitem__(self, index): - img = Image.open(self.images[index]).convert('RGB') - if self.mode == 'test': - img = self._img_transform(img) - if self.transform is not None: - img = self.transform(img) - return img, os.path.basename(self.images[index]) - mask = Image.open(self.masks[index]) - # synchronized transform - if self.mode == 'train': - img, mask = self._sync_transform(img, mask) - elif self.mode == 'val': - img, mask = self._val_sync_transform(img, mask) - else: - assert self.mode == 'testval' - img, mask = self._img_transform(img), self._mask_transform(mask) - # general resize, normalize and toTensor - if self.transform is not None: - img = self.transform(img) - - return img, mask, os.path.basename(self.images[index]) - - def __len__(self): - return len(self.images) - - def _mask_transform(self, mask): - target = np.array(mask).astype('int32') - target[target == 255] = -1 - return torch.from_numpy(target).long() - - @property - def classes(self): - """Category names.""" - return ('background', 'airplane', 'bicycle', 'bird', 'boat', 'bottle', - 'bus', 'car', 'cat', 'chair', 'cow', 'diningtable', 'dog', 'horse', - 'motorcycle', 'person', 'potted-plant', 'sheep', 'sofa', 'train', - 'tv') - - -if __name__ == '__main__': - dataset = VOCSegmentation() \ No newline at end of file diff --git a/AIlib/segutils/core/data/dataloader/sbu_shadow.py b/AIlib/segutils/core/data/dataloader/sbu_shadow.py deleted file mode 100644 index 0cf4ca9..0000000 --- a/AIlib/segutils/core/data/dataloader/sbu_shadow.py +++ /dev/null @@ -1,88 +0,0 @@ -"""SBU Shadow Segmentation Dataset.""" -import os -import torch -import numpy as np - -from PIL import Image -from .segbase import SegmentationDataset - - -class SBUSegmentation(SegmentationDataset): - """SBU Shadow Segmentation Dataset - """ - NUM_CLASS = 2 - - def __init__(self, root='../datasets/sbu', split='train', mode=None, transform=None, **kwargs): - super(SBUSegmentation, self).__init__(root, split, mode, transform, **kwargs) - assert os.path.exists(self.root) - self.images, self.masks = _get_sbu_pairs(self.root, self.split) - assert (len(self.images) == len(self.masks)) - if len(self.images) == 0: - raise RuntimeError("Found 0 images in subfolders of:" + root + "\n") - - def __getitem__(self, index): - img = Image.open(self.images[index]).convert('RGB') - if self.mode == 'test': - if self.transform is not None: - img = self.transform(img) - return img, os.path.basename(self.images[index]) - mask = Image.open(self.masks[index]) - # synchrosized transform - if self.mode == 'train': - img, mask = self._sync_transform(img, mask) - elif self.mode == 'val': - img, mask = self._val_sync_transform(img, mask) - else: - assert self.mode == 'testval' - img, mask = self._img_transform(img), self._mask_transform(mask) - # general resize, normalize and toTensor - if self.transform is not None: - img = self.transform(img) - return img, mask, os.path.basename(self.images[index]) - - def _mask_transform(self, mask): - target = np.array(mask).astype('int32') - target[target > 0] = 1 - return torch.from_numpy(target).long() - - def __len__(self): - return len(self.images) - - @property - def pred_offset(self): - return 0 - - -def _get_sbu_pairs(folder, split='train'): - def get_path_pairs(img_folder, mask_folder): - img_paths = [] - mask_paths = [] - for root, _, files in os.walk(img_folder): - print(root) - for filename in files: - if filename.endswith('.jpg'): - imgpath = os.path.join(root, filename) - maskname = filename.replace('.jpg', '.png') - maskpath = os.path.join(mask_folder, maskname) - if os.path.isfile(imgpath) and os.path.isfile(maskpath): - img_paths.append(imgpath) - mask_paths.append(maskpath) - else: - print('cannot find the mask or image:', imgpath, maskpath) - print('Found {} images in the folder {}'.format(len(img_paths), img_folder)) - return img_paths, mask_paths - - if split == 'train': - img_folder = os.path.join(folder, 'SBUTrain4KRecoveredSmall/ShadowImages') - mask_folder = os.path.join(folder, 'SBUTrain4KRecoveredSmall/ShadowMasks') - img_paths, mask_paths = get_path_pairs(img_folder, mask_folder) - else: - assert split in ('val', 'test') - img_folder = os.path.join(folder, 'SBU-Test/ShadowImages') - mask_folder = os.path.join(folder, 'SBU-Test/ShadowMasks') - img_paths, mask_paths = get_path_pairs(img_folder, mask_folder) - return img_paths, mask_paths - - -if __name__ == '__main__': - dataset = SBUSegmentation(base_size=280, crop_size=256) \ No newline at end of file diff --git a/AIlib/segutils/core/data/dataloader/segbase.py b/AIlib/segutils/core/data/dataloader/segbase.py deleted file mode 100644 index 823436d..0000000 --- a/AIlib/segutils/core/data/dataloader/segbase.py +++ /dev/null @@ -1,93 +0,0 @@ -"""Base segmentation dataset""" -import random -import numpy as np - -from PIL import Image, ImageOps, ImageFilter - -__all__ = ['SegmentationDataset'] - - -class SegmentationDataset(object): - """Segmentation Base Dataset""" - - def __init__(self, root, split, mode, transform, base_size=520, crop_size=480): - super(SegmentationDataset, self).__init__() - self.root = root - self.transform = transform - self.split = split - self.mode = mode if mode is not None else split - self.base_size = base_size - self.crop_size = crop_size - - def _val_sync_transform(self, img, mask): - outsize = self.crop_size - short_size = outsize - w, h = img.size - if w > h: - oh = short_size - ow = int(1.0 * w * oh / h) - else: - ow = short_size - oh = int(1.0 * h * ow / w) - img = img.resize((ow, oh), Image.BILINEAR) - mask = mask.resize((ow, oh), Image.NEAREST) - # center crop - w, h = img.size - x1 = int(round((w - outsize) / 2.)) - y1 = int(round((h - outsize) / 2.)) - img = img.crop((x1, y1, x1 + outsize, y1 + outsize)) - mask = mask.crop((x1, y1, x1 + outsize, y1 + outsize)) - # final transform - img, mask = self._img_transform(img), self._mask_transform(mask) - return img, mask - - def _sync_transform(self, img, mask): - # random mirror - if random.random() < 0.5: - img = img.transpose(Image.FLIP_LEFT_RIGHT) - mask = mask.transpose(Image.FLIP_LEFT_RIGHT) - crop_size = self.crop_size - # random scale (short edge) - short_size = random.randint(int(self.base_size * 0.5), int(self.base_size * 2.0)) - w, h = img.size - if h > w: - ow = short_size - oh = int(1.0 * h * ow / w) - else: - oh = short_size - ow = int(1.0 * w * oh / h) - img = img.resize((ow, oh), Image.BILINEAR) - mask = mask.resize((ow, oh), Image.NEAREST) - # pad crop - if short_size < crop_size: - padh = crop_size - oh if oh < crop_size else 0 - padw = crop_size - ow if ow < crop_size else 0 - img = ImageOps.expand(img, border=(0, 0, padw, padh), fill=0) - mask = ImageOps.expand(mask, border=(0, 0, padw, padh), fill=0) - # random crop crop_size - w, h = img.size - x1 = random.randint(0, w - crop_size) - y1 = random.randint(0, h - crop_size) - img = img.crop((x1, y1, x1 + crop_size, y1 + crop_size)) - mask = mask.crop((x1, y1, x1 + crop_size, y1 + crop_size)) - # gaussian blur as in PSP - if random.random() < 0.5: - img = img.filter(ImageFilter.GaussianBlur(radius=random.random())) - # final transform - img, mask = self._img_transform(img), self._mask_transform(mask) - return img, mask - - def _img_transform(self, img): - return np.array(img) - - def _mask_transform(self, mask): - return np.array(mask).astype('int32') - - @property - def num_class(self): - """Number of categories.""" - return self.NUM_CLASS - - @property - def pred_offset(self): - return 0 diff --git a/AIlib/segutils/core/data/dataloader/utils.py b/AIlib/segutils/core/data/dataloader/utils.py deleted file mode 100644 index c0bd1ad..0000000 --- a/AIlib/segutils/core/data/dataloader/utils.py +++ /dev/null @@ -1,69 +0,0 @@ -import os -import hashlib -import errno -import tarfile -from six.moves import urllib -from torch.utils.model_zoo import tqdm - -def gen_bar_updater(): - pbar = tqdm(total=None) - - def bar_update(count, block_size, total_size): - if pbar.total is None and total_size: - pbar.total = total_size - progress_bytes = count * block_size - pbar.update(progress_bytes - pbar.n) - - return bar_update - -def check_integrity(fpath, md5=None): - if md5 is None: - return True - if not os.path.isfile(fpath): - return False - md5o = hashlib.md5() - with open(fpath, 'rb') as f: - # read in 1MB chunks - for chunk in iter(lambda: f.read(1024 * 1024), b''): - md5o.update(chunk) - md5c = md5o.hexdigest() - if md5c != md5: - return False - return True - -def makedir_exist_ok(dirpath): - try: - os.makedirs(dirpath) - except OSError as e: - if e.errno == errno.EEXIST: - pass - else: - pass - -def download_url(url, root, filename=None, md5=None): - """Download a file from a url and place it in root.""" - root = os.path.expanduser(root) - if not filename: - filename = os.path.basename(url) - fpath = os.path.join(root, filename) - - makedir_exist_ok(root) - - # downloads file - if os.path.isfile(fpath) and check_integrity(fpath, md5): - print('Using downloaded and verified file: ' + fpath) - else: - try: - print('Downloading ' + url + ' to ' + fpath) - urllib.request.urlretrieve(url, fpath, reporthook=gen_bar_updater()) - except OSError: - if url[:5] == 'https': - url = url.replace('https:', 'http:') - print('Failed download. Trying https -> http instead.' - ' Downloading ' + url + ' to ' + fpath) - urllib.request.urlretrieve(url, fpath, reporthook=gen_bar_updater()) - -def download_extract(url, root, filename, md5): - download_url(url, root, filename, md5) - with tarfile.open(os.path.join(root, filename), "r") as tar: - tar.extractall(path=root) \ No newline at end of file diff --git a/AIlib/segutils/core/data/downloader/__init__.py b/AIlib/segutils/core/data/downloader/__init__.py deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/segutils/core/data/downloader/ade20k.py b/AIlib/segutils/core/data/downloader/ade20k.py deleted file mode 100644 index 8187c48..0000000 --- a/AIlib/segutils/core/data/downloader/ade20k.py +++ /dev/null @@ -1,51 +0,0 @@ -"""Prepare ADE20K dataset""" -import os -import sys -import argparse -import zipfile - -# TODO: optim code -cur_path = os.path.abspath(os.path.dirname(__file__)) -root_path = os.path.split(os.path.split(os.path.split(cur_path)[0])[0])[0] -sys.path.append(root_path) - -from core.utils import download, makedirs - -_TARGET_DIR = os.path.expanduser('~/.torch/datasets/ade') - - -def parse_args(): - parser = argparse.ArgumentParser( - description='Initialize ADE20K dataset.', - epilog='Example: python setup_ade20k.py', - formatter_class=argparse.ArgumentDefaultsHelpFormatter) - parser.add_argument('--download-dir', default=None, help='dataset directory on disk') - args = parser.parse_args() - return args - - -def download_ade(path, overwrite=False): - _AUG_DOWNLOAD_URLS = [ - ('http://data.csail.mit.edu/places/ADEchallenge/ADEChallengeData2016.zip', - '219e1696abb36c8ba3a3afe7fb2f4b4606a897c7'), - ( - 'http://data.csail.mit.edu/places/ADEchallenge/release_test.zip', - 'e05747892219d10e9243933371a497e905a4860c'), ] - download_dir = os.path.join(path, 'downloads') - makedirs(download_dir) - for url, checksum in _AUG_DOWNLOAD_URLS: - filename = download(url, path=download_dir, overwrite=overwrite, sha1_hash=checksum) - # extract - with zipfile.ZipFile(filename, "r") as zip_ref: - zip_ref.extractall(path=path) - - -if __name__ == '__main__': - args = parse_args() - makedirs(os.path.expanduser('~/.torch/datasets')) - if args.download_dir is not None: - if os.path.isdir(_TARGET_DIR): - os.remove(_TARGET_DIR) - # make symlink - os.symlink(args.download_dir, _TARGET_DIR) - download_ade(_TARGET_DIR, overwrite=False) diff --git a/AIlib/segutils/core/data/downloader/cityscapes.py b/AIlib/segutils/core/data/downloader/cityscapes.py deleted file mode 100644 index 3b65b88..0000000 --- a/AIlib/segutils/core/data/downloader/cityscapes.py +++ /dev/null @@ -1,54 +0,0 @@ -"""Prepare Cityscapes dataset""" -import os -import sys -import argparse -import zipfile - -# TODO: optim code -cur_path = os.path.abspath(os.path.dirname(__file__)) -root_path = os.path.split(os.path.split(os.path.split(cur_path)[0])[0])[0] -sys.path.append(root_path) - -from core.utils import download, makedirs, check_sha1 - -_TARGET_DIR = os.path.expanduser('~/.torch/datasets/citys') - - -def parse_args(): - parser = argparse.ArgumentParser( - description='Initialize ADE20K dataset.', - epilog='Example: python prepare_cityscapes.py', - formatter_class=argparse.ArgumentDefaultsHelpFormatter) - parser.add_argument('--download-dir', default=None, help='dataset directory on disk') - args = parser.parse_args() - return args - - -def download_city(path, overwrite=False): - _CITY_DOWNLOAD_URLS = [ - ('gtFine_trainvaltest.zip', '99f532cb1af174f5fcc4c5bc8feea8c66246ddbc'), - ('leftImg8bit_trainvaltest.zip', '2c0b77ce9933cc635adda307fbba5566f5d9d404')] - download_dir = os.path.join(path, 'downloads') - makedirs(download_dir) - for filename, checksum in _CITY_DOWNLOAD_URLS: - if not check_sha1(filename, checksum): - raise UserWarning('File {} is downloaded but the content hash does not match. ' \ - 'The repo may be outdated or download may be incomplete. ' \ - 'If the "repo_url" is overridden, consider switching to ' \ - 'the default repo.'.format(filename)) - # extract - with zipfile.ZipFile(filename, "r") as zip_ref: - zip_ref.extractall(path=path) - print("Extracted", filename) - - -if __name__ == '__main__': - args = parse_args() - makedirs(os.path.expanduser('~/.torch/datasets')) - if args.download_dir is not None: - if os.path.isdir(_TARGET_DIR): - os.remove(_TARGET_DIR) - # make symlink - os.symlink(args.download_dir, _TARGET_DIR) - else: - download_city(_TARGET_DIR, overwrite=False) diff --git a/AIlib/segutils/core/data/downloader/mscoco.py b/AIlib/segutils/core/data/downloader/mscoco.py deleted file mode 100644 index 6d509b6..0000000 --- a/AIlib/segutils/core/data/downloader/mscoco.py +++ /dev/null @@ -1,69 +0,0 @@ -"""Prepare MS COCO datasets""" -import os -import sys -import argparse -import zipfile - -# TODO: optim code -cur_path = os.path.abspath(os.path.dirname(__file__)) -root_path = os.path.split(os.path.split(os.path.split(cur_path)[0])[0])[0] -sys.path.append(root_path) - -from core.utils import download, makedirs, try_import_pycocotools - -_TARGET_DIR = os.path.expanduser('~/.torch/datasets/coco') - - -def parse_args(): - parser = argparse.ArgumentParser( - description='Initialize MS COCO dataset.', - epilog='Example: python mscoco.py --download-dir ~/mscoco', - formatter_class=argparse.ArgumentDefaultsHelpFormatter) - parser.add_argument('--download-dir', type=str, default='~/mscoco/', help='dataset directory on disk') - parser.add_argument('--no-download', action='store_true', help='disable automatic download if set') - parser.add_argument('--overwrite', action='store_true', - help='overwrite downloaded files if set, in case they are corrupted') - args = parser.parse_args() - return args - - -def download_coco(path, overwrite=False): - _DOWNLOAD_URLS = [ - ('http://images.cocodataset.org/zips/train2017.zip', - '10ad623668ab00c62c096f0ed636d6aff41faca5'), - ('http://images.cocodataset.org/annotations/annotations_trainval2017.zip', - '8551ee4bb5860311e79dace7e79cb91e432e78b3'), - ('http://images.cocodataset.org/zips/val2017.zip', - '4950dc9d00dbe1c933ee0170f5797584351d2a41'), - # ('http://images.cocodataset.org/annotations/stuff_annotations_trainval2017.zip', - # '46cdcf715b6b4f67e980b529534e79c2edffe084'), - # test2017.zip, for those who want to attend the competition. - # ('http://images.cocodataset.org/zips/test2017.zip', - # '4e443f8a2eca6b1dac8a6c57641b67dd40621a49'), - ] - makedirs(path) - for url, checksum in _DOWNLOAD_URLS: - filename = download(url, path=path, overwrite=overwrite, sha1_hash=checksum) - # extract - with zipfile.ZipFile(filename) as zf: - zf.extractall(path=path) - - -if __name__ == '__main__': - args = parse_args() - path = os.path.expanduser(args.download_dir) - if not os.path.isdir(path) or not os.path.isdir(os.path.join(path, 'train2017')) \ - or not os.path.isdir(os.path.join(path, 'val2017')) \ - or not os.path.isdir(os.path.join(path, 'annotations')): - if args.no_download: - raise ValueError(('{} is not a valid directory, make sure it is present.' - ' Or you should not disable "--no-download" to grab it'.format(path))) - else: - download_coco(path, overwrite=args.overwrite) - - # make symlink - makedirs(os.path.expanduser('~/.torch/datasets')) - if os.path.isdir(_TARGET_DIR): - os.remove(_TARGET_DIR) - os.symlink(path, _TARGET_DIR) - try_import_pycocotools() diff --git a/AIlib/segutils/core/data/downloader/pascal_voc.py b/AIlib/segutils/core/data/downloader/pascal_voc.py deleted file mode 100644 index 849c95b..0000000 --- a/AIlib/segutils/core/data/downloader/pascal_voc.py +++ /dev/null @@ -1,100 +0,0 @@ -"""Prepare PASCAL VOC datasets""" -import os -import sys -import shutil -import argparse -import tarfile - -# TODO: optim code -cur_path = os.path.abspath(os.path.dirname(__file__)) -root_path = os.path.split(os.path.split(os.path.split(cur_path)[0])[0])[0] -sys.path.append(root_path) - -from core.utils import download, makedirs - -_TARGET_DIR = os.path.expanduser('~/.torch/datasets/voc') - - -def parse_args(): - parser = argparse.ArgumentParser( - description='Initialize PASCAL VOC dataset.', - epilog='Example: python pascal_voc.py --download-dir ~/VOCdevkit', - formatter_class=argparse.ArgumentDefaultsHelpFormatter) - parser.add_argument('--download-dir', type=str, default='~/VOCdevkit/', help='dataset directory on disk') - parser.add_argument('--no-download', action='store_true', help='disable automatic download if set') - parser.add_argument('--overwrite', action='store_true', - help='overwrite downloaded files if set, in case they are corrupted') - args = parser.parse_args() - return args - - -##################################################################################### -# Download and extract VOC datasets into ``path`` - -def download_voc(path, overwrite=False): - _DOWNLOAD_URLS = [ - ('http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtrainval_06-Nov-2007.tar', - '34ed68851bce2a36e2a223fa52c661d592c66b3c'), - ('http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtest_06-Nov-2007.tar', - '41a8d6e12baa5ab18ee7f8f8029b9e11805b4ef1'), - ('http://host.robots.ox.ac.uk/pascal/VOC/voc2012/VOCtrainval_11-May-2012.tar', - '4e443f8a2eca6b1dac8a6c57641b67dd40621a49')] - makedirs(path) - for url, checksum in _DOWNLOAD_URLS: - filename = download(url, path=path, overwrite=overwrite, sha1_hash=checksum) - # extract - with tarfile.open(filename) as tar: - tar.extractall(path=path) - - -##################################################################################### -# Download and extract the VOC augmented segmentation dataset into ``path`` - -def download_aug(path, overwrite=False): - _AUG_DOWNLOAD_URLS = [ - ('http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/semantic_contours/benchmark.tgz', - '7129e0a480c2d6afb02b517bb18ac54283bfaa35')] - makedirs(path) - for url, checksum in _AUG_DOWNLOAD_URLS: - filename = download(url, path=path, overwrite=overwrite, sha1_hash=checksum) - # extract - with tarfile.open(filename) as tar: - tar.extractall(path=path) - shutil.move(os.path.join(path, 'benchmark_RELEASE'), - os.path.join(path, 'VOCaug')) - filenames = ['VOCaug/dataset/train.txt', 'VOCaug/dataset/val.txt'] - # generate trainval.txt - with open(os.path.join(path, 'VOCaug/dataset/trainval.txt'), 'w') as outfile: - for fname in filenames: - fname = os.path.join(path, fname) - with open(fname) as infile: - for line in infile: - outfile.write(line) - - -if __name__ == '__main__': - args = parse_args() - path = os.path.expanduser(args.download_dir) - if not os.path.isfile(path) or not os.path.isdir(os.path.join(path, 'VOC2007')) \ - or not os.path.isdir(os.path.join(path, 'VOC2012')): - if args.no_download: - raise ValueError(('{} is not a valid directory, make sure it is present.' - ' Or you should not disable "--no-download" to grab it'.format(path))) - else: - download_voc(path, overwrite=args.overwrite) - shutil.move(os.path.join(path, 'VOCdevkit', 'VOC2007'), os.path.join(path, 'VOC2007')) - shutil.move(os.path.join(path, 'VOCdevkit', 'VOC2012'), os.path.join(path, 'VOC2012')) - shutil.rmtree(os.path.join(path, 'VOCdevkit')) - - if not os.path.isdir(os.path.join(path, 'VOCaug')): - if args.no_download: - raise ValueError(('{} is not a valid directory, make sure it is present.' - ' Or you should not disable "--no-download" to grab it'.format(path))) - else: - download_aug(path, overwrite=args.overwrite) - - # make symlink - makedirs(os.path.expanduser('~/.torch/datasets')) - if os.path.isdir(_TARGET_DIR): - os.remove(_TARGET_DIR) - os.symlink(path, _TARGET_DIR) diff --git a/AIlib/segutils/core/data/downloader/sbu_shadow.py b/AIlib/segutils/core/data/downloader/sbu_shadow.py deleted file mode 100644 index cdcbdde..0000000 --- a/AIlib/segutils/core/data/downloader/sbu_shadow.py +++ /dev/null @@ -1,56 +0,0 @@ -"""Prepare SBU Shadow datasets""" -import os -import sys -import argparse -import zipfile - -# TODO: optim code -cur_path = os.path.abspath(os.path.dirname(__file__)) -root_path = os.path.split(os.path.split(os.path.split(cur_path)[0])[0])[0] -sys.path.append(root_path) - -from core.utils import download, makedirs - -_TARGET_DIR = os.path.expanduser('~/.torch/datasets/sbu') - - -def parse_args(): - parser = argparse.ArgumentParser( - description='Initialize SBU Shadow dataset.', - epilog='Example: python sbu_shadow.py --download-dir ~/SBU-shadow', - formatter_class=argparse.ArgumentDefaultsHelpFormatter) - parser.add_argument('--download-dir', type=str, default=None, help='dataset directory on disk') - parser.add_argument('--no-download', action='store_true', help='disable automatic download if set') - parser.add_argument('--overwrite', action='store_true', - help='overwrite downloaded files if set, in case they are corrupted') - args = parser.parse_args() - return args - - -##################################################################################### -# Download and extract SBU shadow datasets into ``path`` - -def download_sbu(path, overwrite=False): - _DOWNLOAD_URLS = [ - ('http://www3.cs.stonybrook.edu/~cvl/content/datasets/shadow_db/SBU-shadow.zip'), - ] - download_dir = os.path.join(path, 'downloads') - makedirs(download_dir) - for url in _DOWNLOAD_URLS: - filename = download(url, path=path, overwrite=overwrite) - # extract - with zipfile.ZipFile(filename, "r") as zf: - zf.extractall(path=path) - print("Extracted", filename) - - -if __name__ == '__main__': - args = parse_args() - makedirs(os.path.expanduser('~/.torch/datasets')) - if args.download_dir is not None: - if os.path.isdir(_TARGET_DIR): - os.remove(_TARGET_DIR) - # make symlink - os.symlink(args.download_dir, _TARGET_DIR) - else: - download_sbu(_TARGET_DIR, overwrite=False) diff --git a/AIlib/segutils/core/lib/psa/__pycache__/functional.cpython-36.pyc b/AIlib/segutils/core/lib/psa/__pycache__/functional.cpython-36.pyc deleted file mode 100644 index bf41206..0000000 Binary files a/AIlib/segutils/core/lib/psa/__pycache__/functional.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/lib/psa/functional.py b/AIlib/segutils/core/lib/psa/functional.py deleted file mode 100644 index 8e66088..0000000 --- a/AIlib/segutils/core/lib/psa/functional.py +++ /dev/null @@ -1,5 +0,0 @@ -from . import functions - - -def psa_mask(input, psa_type=0, mask_H_=None, mask_W_=None): - return functions.psa_mask(input, psa_type, mask_H_, mask_W_) diff --git a/AIlib/segutils/core/lib/psa/functions/__init__.py b/AIlib/segutils/core/lib/psa/functions/__init__.py deleted file mode 100644 index 1b4726b..0000000 --- a/AIlib/segutils/core/lib/psa/functions/__init__.py +++ /dev/null @@ -1 +0,0 @@ -from .psamask import * diff --git a/AIlib/segutils/core/lib/psa/functions/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/lib/psa/functions/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index f6e5711..0000000 Binary files a/AIlib/segutils/core/lib/psa/functions/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/lib/psa/functions/__pycache__/psamask.cpython-36.pyc b/AIlib/segutils/core/lib/psa/functions/__pycache__/psamask.cpython-36.pyc deleted file mode 100644 index ba6b6f1..0000000 Binary files a/AIlib/segutils/core/lib/psa/functions/__pycache__/psamask.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/lib/psa/functions/psamask.py b/AIlib/segutils/core/lib/psa/functions/psamask.py deleted file mode 100644 index 26f34a2..0000000 --- a/AIlib/segutils/core/lib/psa/functions/psamask.py +++ /dev/null @@ -1,39 +0,0 @@ -import torch -from torch.autograd import Function -from .. import src - - -class PSAMask(Function): - @staticmethod - def forward(ctx, input, psa_type=0, mask_H_=None, mask_W_=None): - assert psa_type in [0, 1] # 0-col, 1-dis - assert (mask_H_ is None and mask_W_ is None) or (mask_H_ is not None and mask_W_ is not None) - num_, channels_, feature_H_, feature_W_ = input.size() - if mask_H_ is None and mask_W_ is None: - mask_H_, mask_W_ = 2 * feature_H_ - 1, 2 * feature_W_ - 1 - assert (mask_H_ % 2 == 1) and (mask_W_ % 2 == 1) - assert channels_ == mask_H_ * mask_W_ - half_mask_H_, half_mask_W_ = (mask_H_ - 1) // 2, (mask_W_ - 1) // 2 - output = torch.zeros([num_, feature_H_ * feature_W_, feature_H_, feature_W_], dtype=input.dtype, device=input.device) - if not input.is_cuda: - src.cpu.psamask_forward(psa_type, input, output, num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_) - else: - output = output.cuda() - src.gpu.psamask_forward(psa_type, input, output, num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_) - ctx.psa_type, ctx.num_, ctx.channels_, ctx.feature_H_, ctx.feature_W_ = psa_type, num_, channels_, feature_H_, feature_W_ - ctx.mask_H_, ctx.mask_W_, ctx.half_mask_H_, ctx.half_mask_W_ = mask_H_, mask_W_, half_mask_H_, half_mask_W_ - return output - - @staticmethod - def backward(ctx, grad_output): - psa_type, num_, channels_, feature_H_, feature_W_ = ctx.psa_type, ctx.num_, ctx.channels_, ctx.feature_H_, ctx.feature_W_ - mask_H_, mask_W_, half_mask_H_, half_mask_W_ = ctx.mask_H_, ctx.mask_W_, ctx.half_mask_H_, ctx.half_mask_W_ - grad_input = torch.zeros([num_, channels_, feature_H_, feature_W_], dtype=grad_output.dtype, device=grad_output.device) - if not grad_output.is_cuda: - src.cpu.psamask_backward(psa_type, grad_output, grad_input, num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_) - else: - src.gpu.psamask_backward(psa_type, grad_output, grad_input, num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_) - return grad_input, None, None, None - - -psa_mask = PSAMask.apply diff --git a/AIlib/segutils/core/lib/psa/modules/__init__.py b/AIlib/segutils/core/lib/psa/modules/__init__.py deleted file mode 100644 index 1b4726b..0000000 --- a/AIlib/segutils/core/lib/psa/modules/__init__.py +++ /dev/null @@ -1 +0,0 @@ -from .psamask import * diff --git a/AIlib/segutils/core/lib/psa/modules/psamask.py b/AIlib/segutils/core/lib/psa/modules/psamask.py deleted file mode 100644 index 58ea4d9..0000000 --- a/AIlib/segutils/core/lib/psa/modules/psamask.py +++ /dev/null @@ -1,15 +0,0 @@ -from torch import nn -from .. import functional as F - - -class PSAMask(nn.Module): - def __init__(self, psa_type=0, mask_H_=None, mask_W_=None): - super(PSAMask, self).__init__() - assert psa_type in [0, 1] # 0-col, 1-dis - assert (mask_H_ in None and mask_W_ is None) or (mask_H_ is not None and mask_W_ is not None) - self.psa_type = psa_type - self.mask_H_ = mask_H_ - self.mask_W_ = mask_W_ - - def forward(self, input): - return F.psa_mask(input, self.psa_type, self.mask_H_, self.mask_W_) diff --git a/AIlib/segutils/core/lib/psa/src/__init__.py b/AIlib/segutils/core/lib/psa/src/__init__.py deleted file mode 100644 index ead1cfe..0000000 --- a/AIlib/segutils/core/lib/psa/src/__init__.py +++ /dev/null @@ -1,18 +0,0 @@ -import os -import torch -from torch.utils.cpp_extension import load - -cwd = os.path.dirname(os.path.realpath(__file__)) -cpu_path = os.path.join(cwd, 'cpu') -gpu_path = os.path.join(cwd, 'gpu') -print(cpu_path,gpu_path) -cpu = load('psamask_cpu', [ - os.path.join(cpu_path, 'operator.cpp'), - os.path.join(cpu_path, 'psamask.cpp'), -], build_directory=cpu_path, verbose=False) - -if torch.cuda.is_available(): - gpu = load('psamask_gpu', [ - os.path.join(gpu_path, 'operator.cpp'), - os.path.join(gpu_path, 'psamask_cuda.cu'), - ], build_directory=gpu_path, verbose=False) \ No newline at end of file diff --git a/AIlib/segutils/core/lib/psa/src/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/lib/psa/src/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index 2532ecb..0000000 Binary files a/AIlib/segutils/core/lib/psa/src/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/lib/psa/src/cpu/operator.cpp b/AIlib/segutils/core/lib/psa/src/cpu/operator.cpp deleted file mode 100644 index e7b9f6c..0000000 --- a/AIlib/segutils/core/lib/psa/src/cpu/operator.cpp +++ /dev/null @@ -1,6 +0,0 @@ -#include "operator.h" - -PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { - m.def("psamask_forward", &psamask_forward_cpu, "PSAMASK forward (CPU)"); - m.def("psamask_backward", &psamask_backward_cpu, "PSAMASK backward (CPU)"); -} diff --git a/AIlib/segutils/core/lib/psa/src/cpu/operator.h b/AIlib/segutils/core/lib/psa/src/cpu/operator.h deleted file mode 100644 index abc43cb..0000000 --- a/AIlib/segutils/core/lib/psa/src/cpu/operator.h +++ /dev/null @@ -1,4 +0,0 @@ -#include - -void psamask_forward_cpu(const int psa_type, const at::Tensor& input, at::Tensor& output, const int num_, const int feature_H_, const int feature_W_, const int mask_H_, const int mask_W_, const int half_mask_H_, const int half_mask_W_); -void psamask_backward_cpu(const int psa_type, const at::Tensor& grad_output, at::Tensor& grad_input, const int num_, const int feature_H_, const int feature_W_, const int mask_H_, const int mask_W_, const int half_mask_H_, const int half_mask_W_); \ No newline at end of file diff --git a/AIlib/segutils/core/lib/psa/src/cpu/psamask.cpp b/AIlib/segutils/core/lib/psa/src/cpu/psamask.cpp deleted file mode 100644 index eb33694..0000000 --- a/AIlib/segutils/core/lib/psa/src/cpu/psamask.cpp +++ /dev/null @@ -1,133 +0,0 @@ -#include - -#ifndef min -#define min(a,b) (((a) < (b)) ? (a) : (b)) -#endif - -#ifndef max -#define max(a,b) (((a) > (b)) ? (a) : (b)) -#endif - -void psamask_collect_forward(const int num_, - const int feature_H_, const int feature_W_, - const int mask_H_, const int mask_W_, - const int half_mask_H_, const int half_mask_W_, - const float* mask_data, float* buffer_data) { - for(int n = 0; n < num_; n++) { - for(int h = 0; h < feature_H_; h++) { - for(int w = 0; w < feature_W_; w++) { - // effective mask region : [hstart, hend) x [wstart, wend) with mask-indexed - const int hstart = max(0, half_mask_H_ - h); - const int hend = min(mask_H_, feature_H_ + half_mask_H_ - h); - const int wstart = max(0, half_mask_W_ - w); - const int wend = min(mask_W_, feature_W_ + half_mask_W_ - w); - // (hidx, widx ) with mask-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - buffer_data[(n * feature_H_ * feature_W_ + (hidx + h - half_mask_H_) * feature_W_ + (widx + w - half_mask_W_)) * feature_H_ * feature_W_ + h * feature_W_ + w] = - mask_data[((n * mask_H_ * mask_W_ + hidx * mask_W_ + widx) * feature_H_ + h) * feature_W_ + w]; - } - } - } - } - } -} - -void psamask_distribute_forward(const int num_, - const int feature_H_, const int feature_W_, - const int mask_H_, const int mask_W_, - const int half_mask_H_, const int half_mask_W_, - const float* mask_data, float* buffer_data) { - for(int n = 0; n < num_; n++) { - for(int h = 0; h < feature_H_; h++) { - for(int w = 0; w < feature_W_; w++) { - // effective mask region : [hstart, hend) x [wstart, wend) with mask-indexed - const int hstart = max(0, half_mask_H_ - h); - const int hend = min(mask_H_, feature_H_ + half_mask_H_ - h); - const int wstart = max(0, half_mask_W_ - w); - const int wend = min(mask_W_, feature_W_ + half_mask_W_ - w); - // (hidx, widx ) with mask-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - buffer_data[(n * feature_H_ * feature_W_ + h * feature_W_ + w) * feature_H_ * feature_W_ + (hidx + h - half_mask_H_) * feature_W_ + (widx + w - half_mask_W_)] = - mask_data[((n * mask_H_ * mask_W_ + hidx * mask_W_ + widx) * feature_H_ + h) * feature_W_ + w]; - } - } - } - } - } -} - -void psamask_collect_backward(const int num_, - const int feature_H_, const int feature_W_, - const int mask_H_, const int mask_W_, - const int half_mask_H_, const int half_mask_W_, - const float* buffer_diff, float* mask_diff) { - for(int n = 0; n < num_; n++) { - for(int h = 0; h < feature_H_; h++) { - for(int w = 0; w < feature_W_; w++) { - // effective mask region : [hstart, hend) x [wstart, wend) with mask-indexed - const int hstart = max(0, half_mask_H_ - h); - const int hend = min(mask_H_, feature_H_ + half_mask_H_ - h); - const int wstart = max(0, half_mask_W_ - w); - const int wend = min(mask_W_, feature_W_ + half_mask_W_ - w); - // (hidx, widx ) with mask-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - mask_diff[((n * mask_H_ * mask_W_ + hidx * mask_W_ + widx) * feature_H_ + h) * feature_W_ + w] = - buffer_diff[(n * feature_H_ * feature_W_ + (hidx + h - half_mask_H_) * feature_W_ + (widx + w - half_mask_W_)) * feature_H_ * feature_W_ + h * feature_W_ + w]; - } - } - } - } - } -} - -void psamask_distribute_backward(const int num_, - const int feature_H_, const int feature_W_, - const int mask_H_, const int mask_W_, - const int half_mask_H_, const int half_mask_W_, - const float* buffer_diff, float* mask_diff) { - for(int n = 0; n < num_; n++) { - for(int h = 0; h < feature_H_; h++) { - for(int w = 0; w < feature_W_; w++) { - // effective mask region : [hstart, hend) x [wstart, wend) with mask-indexed - const int hstart = max(0, half_mask_H_ - h); - const int hend = min(mask_H_, feature_H_ + half_mask_H_ - h); - const int wstart = max(0, half_mask_W_ - w); - const int wend = min(mask_W_, feature_W_ + half_mask_W_ - w); - // (hidx, widx ) with mask-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - mask_diff[((n * mask_H_ * mask_W_ + hidx * mask_W_ + widx) * feature_H_ + h) * feature_W_ + w] = - buffer_diff[(n * feature_H_ * feature_W_ + h * feature_W_ + w) * feature_H_ * feature_W_ + (hidx + h - half_mask_H_) * feature_W_ + (widx + w - half_mask_W_)]; - } - } - } - } - } -} - -void psamask_forward_cpu(const int psa_type, const at::Tensor& input, at::Tensor& output, const int num_, const int feature_H_, const int feature_W_, const int mask_H_, const int mask_W_, const int half_mask_H_, const int half_mask_W_) -{ - const float* input_data = input.data(); - float* output_data = output.data(); - if(psa_type == 0) - psamask_collect_forward(num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_, input_data, output_data); - else - psamask_distribute_forward(num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_, input_data, output_data); -} - -void psamask_backward_cpu(const int psa_type, const at::Tensor& grad_output, at::Tensor& grad_input, const int num_, const int feature_H_, const int feature_W_, const int mask_H_, const int mask_W_, const int half_mask_H_, const int half_mask_W_) -{ - const float* grad_output_data = grad_output.data(); - float* grad_input_data = grad_input.data(); - if(psa_type == 0) - psamask_collect_backward(num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_, grad_output_data, grad_input_data); - else - psamask_distribute_backward(num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_, grad_output_data, grad_input_data); -} diff --git a/AIlib/segutils/core/lib/psa/src/gpu/operator.cpp b/AIlib/segutils/core/lib/psa/src/gpu/operator.cpp deleted file mode 100644 index 5a52f4a..0000000 --- a/AIlib/segutils/core/lib/psa/src/gpu/operator.cpp +++ /dev/null @@ -1,6 +0,0 @@ -#include "operator.h" - -PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { - m.def("psamask_forward", &psamask_forward_cuda, "PSAMASK forward (GPU)"); - m.def("psamask_backward", &psamask_backward_cuda, "PSAMASK backward (GPU)"); -} diff --git a/AIlib/segutils/core/lib/psa/src/gpu/operator.h b/AIlib/segutils/core/lib/psa/src/gpu/operator.h deleted file mode 100644 index 235a9e1..0000000 --- a/AIlib/segutils/core/lib/psa/src/gpu/operator.h +++ /dev/null @@ -1,4 +0,0 @@ -#include - -void psamask_forward_cuda(const int psa_type, const at::Tensor& input, at::Tensor& output, const int num_, const int feature_H_, const int feature_W_, const int mask_H_, const int mask_W_, const int half_mask_H_, const int half_mask_W_); -void psamask_backward_cuda(const int psa_type, const at::Tensor& grad_output, at::Tensor& grad_input, const int num_, const int feature_H_, const int feature_W_, const int mask_H_, const int mask_W_, const int half_mask_H_, const int half_mask_W_); diff --git a/AIlib/segutils/core/lib/psa/src/gpu/psamask_cuda.cu b/AIlib/segutils/core/lib/psa/src/gpu/psamask_cuda.cu deleted file mode 100644 index f3fcb93..0000000 --- a/AIlib/segutils/core/lib/psa/src/gpu/psamask_cuda.cu +++ /dev/null @@ -1,128 +0,0 @@ -#include - -// CUDA: grid stride looping -#ifndef CUDA_KERNEL_LOOP -#define CUDA_KERNEL_LOOP(i, n) for (int i = blockIdx.x * blockDim.x + threadIdx.x; i < (n); i += blockDim.x * gridDim.x) -#endif - -__global__ void psamask_collect_forward_cuda(const int nthreads, - const int feature_H_, const int feature_W_, - const int mask_H_, const int mask_W_, - const int half_mask_H_, const int half_mask_W_, - const float* mask_data, float* buffer_data) { - CUDA_KERNEL_LOOP(index, nthreads) { - const int w = index % feature_W_; - const int h = (index / feature_W_) % feature_H_; - const int n = index / feature_W_ / feature_H_; - // effective mask region : [hstart, hend) x [wstart, wend) with mask-indexed - const int hstart = max(0, half_mask_H_ - h); - const int hend = min(mask_H_, feature_H_ + half_mask_H_ - h); - const int wstart = max(0, half_mask_W_ - w); - const int wend = min(mask_W_, feature_W_ + half_mask_W_ - w); - // (hidx, widx ) with mask-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - buffer_data[(n * feature_H_ * feature_W_ + (hidx + h - half_mask_H_) * feature_W_ + (widx + w - half_mask_W_)) * feature_H_ * feature_W_ + h * feature_W_ + w] = - mask_data[((n * mask_H_ * mask_W_ + hidx * mask_W_ + widx) * feature_H_ + h) * feature_W_ + w]; - } - } - } -} - -__global__ void psamask_distribute_forward_cuda(const int nthreads, - const int feature_H_, const int feature_W_, - const int mask_H_, const int mask_W_, - const int half_mask_H_, const int half_mask_W_, - const float* mask_data, float* buffer_data) { - CUDA_KERNEL_LOOP(index, nthreads) { - const int w = index % feature_W_; - const int h = (index / feature_W_) % feature_H_; - const int n = index / feature_W_ / feature_H_; - // effective mask region : [hstart, hend) x [wstart, wend) with mask-indexed - const int hstart = max(0, half_mask_H_ - h); - const int hend = min(mask_H_, feature_H_ + half_mask_H_ - h); - const int wstart = max(0, half_mask_W_ - w); - const int wend = min(mask_W_, feature_W_ + half_mask_W_ - w); - // (hidx, widx ) with mask-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - buffer_data[(n * feature_H_ * feature_W_ + h * feature_W_ + w) * feature_H_ * feature_W_ + (hidx + h - half_mask_H_) * feature_W_ + (widx + w - half_mask_W_)] = - mask_data[((n * mask_H_ * mask_W_ + hidx * mask_W_ + widx) * feature_H_ + h) * feature_W_ + w]; - } - } - } -} - -__global__ void psamask_collect_backward_cuda(const int nthreads, - const int feature_H_, const int feature_W_, - const int mask_H_, const int mask_W_, - const int half_mask_H_, const int half_mask_W_, - const float* buffer_diff, float* mask_diff) { - CUDA_KERNEL_LOOP(index, nthreads) { - const int w = index % feature_W_; - const int h = (index / feature_W_) % feature_H_; - const int n = index / feature_W_ / feature_H_; - // effective mask region : [hstart, hend) x [wstart, wend) with mask-indexed - const int hstart = max(0, half_mask_H_ - h); - const int hend = min(mask_H_, feature_H_ + half_mask_H_ - h); - const int wstart = max(0, half_mask_W_ - w); - const int wend = min(mask_W_, feature_W_ + half_mask_W_ - w); - // (hidx, widx ) with mask-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - mask_diff[((n * mask_H_ * mask_W_ + hidx * mask_W_ + widx) * feature_H_ + h) * feature_W_ + w] = - buffer_diff[(n * feature_H_ * feature_W_ + (hidx + h - half_mask_H_) * feature_W_ + (widx + w - half_mask_W_)) * feature_H_ * feature_W_ + h * feature_W_ + w]; - } - } - } -} - -__global__ void psamask_distribute_backward_cuda(const int nthreads, - const int feature_H_, const int feature_W_, - const int mask_H_, const int mask_W_, - const int half_mask_H_, const int half_mask_W_, - const float* buffer_diff, float* mask_diff) { - CUDA_KERNEL_LOOP(index, nthreads) { - const int w = index % feature_W_; - const int h = (index / feature_W_) % feature_H_; - const int n = index / feature_W_ / feature_H_; - // effective mask region : [hstart, hend) x [wstart, wend) with mask-indexed - const int hstart = max(0, half_mask_H_ - h); - const int hend = min(mask_H_, feature_H_ + half_mask_H_ - h); - const int wstart = max(0, half_mask_W_ - w); - const int wend = min(mask_W_, feature_W_ + half_mask_W_ - w); - // (hidx, widx ) with mask-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - mask_diff[((n * mask_H_ * mask_W_ + hidx * mask_W_ + widx) * feature_H_ + h) * feature_W_ + w] = - buffer_diff[(n * feature_H_ * feature_W_ + h * feature_W_ + w) * feature_H_ * feature_W_ + (hidx + h - half_mask_H_) * feature_W_ + (widx + w - half_mask_W_)]; - } - } - } -} - -void psamask_forward_cuda(const int psa_type, const at::Tensor& input, at::Tensor& output, const int num_, const int feature_H_, const int feature_W_, const int mask_H_, const int mask_W_, const int half_mask_H_, const int half_mask_W_) -{ - int nthreads = num_ * feature_H_ * feature_W_; - const float* input_data = input.data(); - float* output_data = output.data(); - if(psa_type == 0) - psamask_collect_forward_cuda<<>>(nthreads, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_, input_data, output_data); - else - psamask_distribute_forward_cuda<<>>(nthreads, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_, input_data, output_data); -} - -void psamask_backward_cuda(const int psa_type, const at::Tensor& grad_output, at::Tensor& grad_input, const int num_, const int feature_H_, const int feature_W_, const int mask_H_, const int mask_W_, const int half_mask_H_, const int half_mask_W_) -{ - int nthreads = num_ * feature_H_ * feature_W_; - const float* grad_output_data = grad_output.data(); - float* grad_input_data = grad_input.data(); - if(psa_type == 0) - psamask_collect_backward_cuda<<>>(nthreads, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_, grad_output_data, grad_input_data); - else - psamask_distribute_backward_cuda<<>>(nthreads, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_, grad_output_data, grad_input_data); -} diff --git a/AIlib/segutils/core/models/__init__.py b/AIlib/segutils/core/models/__init__.py deleted file mode 100644 index 2a8b222..0000000 --- a/AIlib/segutils/core/models/__init__.py +++ /dev/null @@ -1,2 +0,0 @@ -"""Model Zoo""" -from .model_zoo import get_model, get_model_list \ No newline at end of file diff --git a/AIlib/segutils/core/models/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index a439fcc..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/__init__.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index 1391e84..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/bisenet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/bisenet.cpython-36.pyc deleted file mode 100644 index 6b3558c..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/bisenet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/bisenet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/bisenet.cpython-38.pyc deleted file mode 100644 index 3f394a1..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/bisenet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/ccnet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/ccnet.cpython-36.pyc deleted file mode 100644 index 32d987b..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/ccnet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/ccnet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/ccnet.cpython-38.pyc deleted file mode 100644 index bd31a78..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/ccnet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/cgnet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/cgnet.cpython-36.pyc deleted file mode 100644 index 524084d..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/cgnet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/cgnet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/cgnet.cpython-38.pyc deleted file mode 100644 index efeac5c..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/cgnet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/danet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/danet.cpython-36.pyc deleted file mode 100644 index 2878430..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/danet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/danet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/danet.cpython-38.pyc deleted file mode 100644 index 6083319..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/danet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/deeplabv3.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/deeplabv3.cpython-36.pyc deleted file mode 100644 index 25da2f6..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/deeplabv3.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/deeplabv3.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/deeplabv3.cpython-38.pyc deleted file mode 100644 index 8bed563..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/deeplabv3.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/deeplabv3_plus.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/deeplabv3_plus.cpython-36.pyc deleted file mode 100644 index 0793537..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/deeplabv3_plus.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/deeplabv3_plus.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/deeplabv3_plus.cpython-38.pyc deleted file mode 100644 index 57da98f..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/deeplabv3_plus.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/denseaspp.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/denseaspp.cpython-36.pyc deleted file mode 100644 index 795d512..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/denseaspp.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/denseaspp.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/denseaspp.cpython-38.pyc deleted file mode 100644 index 4f461a9..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/denseaspp.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/dfanet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/dfanet.cpython-36.pyc deleted file mode 100644 index b81fd89..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/dfanet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/dfanet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/dfanet.cpython-38.pyc deleted file mode 100644 index 62b873e..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/dfanet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/dinknet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/dinknet.cpython-38.pyc deleted file mode 100644 index e8a8ba8..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/dinknet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/dunet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/dunet.cpython-36.pyc deleted file mode 100644 index f5a46bd..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/dunet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/dunet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/dunet.cpython-38.pyc deleted file mode 100644 index 3c984c3..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/dunet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/encnet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/encnet.cpython-36.pyc deleted file mode 100644 index 91b5bcd..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/encnet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/encnet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/encnet.cpython-38.pyc deleted file mode 100644 index 1727e10..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/encnet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/enet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/enet.cpython-36.pyc deleted file mode 100644 index dcbff56..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/enet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/enet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/enet.cpython-38.pyc deleted file mode 100644 index 89764f5..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/enet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/espnet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/espnet.cpython-36.pyc deleted file mode 100644 index 74c47b4..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/espnet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/espnet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/espnet.cpython-38.pyc deleted file mode 100644 index ceac53b..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/espnet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/fcn.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/fcn.cpython-36.pyc deleted file mode 100644 index 1631b50..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/fcn.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/fcn.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/fcn.cpython-38.pyc deleted file mode 100644 index 40c6d21..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/fcn.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/fcnv2.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/fcnv2.cpython-36.pyc deleted file mode 100644 index 246374e..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/fcnv2.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/fcnv2.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/fcnv2.cpython-38.pyc deleted file mode 100644 index be39402..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/fcnv2.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/icnet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/icnet.cpython-36.pyc deleted file mode 100644 index 345b469..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/icnet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/icnet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/icnet.cpython-38.pyc deleted file mode 100644 index a7c7e71..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/icnet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/lednet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/lednet.cpython-36.pyc deleted file mode 100644 index 186d60f..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/lednet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/lednet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/lednet.cpython-38.pyc deleted file mode 100644 index b2e205d..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/lednet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/model_store.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/model_store.cpython-36.pyc deleted file mode 100644 index b337749..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/model_store.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/model_store.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/model_store.cpython-38.pyc deleted file mode 100644 index 873ad70..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/model_store.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/model_zoo.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/model_zoo.cpython-36.pyc deleted file mode 100644 index 199642f..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/model_zoo.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/model_zoo.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/model_zoo.cpython-38.pyc deleted file mode 100644 index 6f67f13..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/model_zoo.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/ocnet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/ocnet.cpython-36.pyc deleted file mode 100644 index 78b3cea..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/ocnet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/ocnet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/ocnet.cpython-38.pyc deleted file mode 100644 index 4c4812a..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/ocnet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/psanet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/psanet.cpython-36.pyc deleted file mode 100644 index 9fd6713..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/psanet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/psanet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/psanet.cpython-38.pyc deleted file mode 100644 index d9fb676..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/psanet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/pspnet.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/pspnet.cpython-36.pyc deleted file mode 100644 index 4580707..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/pspnet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/pspnet.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/pspnet.cpython-38.pyc deleted file mode 100644 index 2e04bb4..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/pspnet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/segbase.cpython-36.pyc b/AIlib/segutils/core/models/__pycache__/segbase.cpython-36.pyc deleted file mode 100644 index 3b23196..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/segbase.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/__pycache__/segbase.cpython-38.pyc b/AIlib/segutils/core/models/__pycache__/segbase.cpython-38.pyc deleted file mode 100644 index 62bd258..0000000 Binary files a/AIlib/segutils/core/models/__pycache__/segbase.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models.zip b/AIlib/segutils/core/models/base_models.zip deleted file mode 100644 index b7fb6b1..0000000 Binary files a/AIlib/segutils/core/models/base_models.zip and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__init__.py b/AIlib/segutils/core/models/base_models/__init__.py deleted file mode 100644 index 562aa28..0000000 --- a/AIlib/segutils/core/models/base_models/__init__.py +++ /dev/null @@ -1,6 +0,0 @@ -from .densenet import * -from .resnet import * -from .resnetv1b import * -from .vgg import * -from .eespnet import * -from .xception import * diff --git a/AIlib/segutils/core/models/base_models/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/models/base_models/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index f7e8ec8..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/__init__.cpython-38.pyc b/AIlib/segutils/core/models/base_models/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index fbca07b..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/densenet.cpython-36.pyc b/AIlib/segutils/core/models/base_models/__pycache__/densenet.cpython-36.pyc deleted file mode 100644 index c291f3e..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/densenet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/densenet.cpython-38.pyc b/AIlib/segutils/core/models/base_models/__pycache__/densenet.cpython-38.pyc deleted file mode 100644 index 3db8c4d..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/densenet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/eespnet.cpython-36.pyc b/AIlib/segutils/core/models/base_models/__pycache__/eespnet.cpython-36.pyc deleted file mode 100644 index bc6bad7..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/eespnet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/eespnet.cpython-38.pyc b/AIlib/segutils/core/models/base_models/__pycache__/eespnet.cpython-38.pyc deleted file mode 100644 index e9b146c..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/eespnet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/resnet.cpython-36.pyc b/AIlib/segutils/core/models/base_models/__pycache__/resnet.cpython-36.pyc deleted file mode 100644 index 95158df..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/resnet.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/resnet.cpython-38.pyc b/AIlib/segutils/core/models/base_models/__pycache__/resnet.cpython-38.pyc deleted file mode 100644 index f19f885..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/resnet.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/resnetv1b.cpython-36.pyc b/AIlib/segutils/core/models/base_models/__pycache__/resnetv1b.cpython-36.pyc deleted file mode 100644 index 7c92fe0..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/resnetv1b.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/resnetv1b.cpython-38.pyc b/AIlib/segutils/core/models/base_models/__pycache__/resnetv1b.cpython-38.pyc deleted file mode 100644 index 7859838..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/resnetv1b.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/vgg.cpython-36.pyc b/AIlib/segutils/core/models/base_models/__pycache__/vgg.cpython-36.pyc deleted file mode 100644 index c1d9f59..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/vgg.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/vgg.cpython-38.pyc b/AIlib/segutils/core/models/base_models/__pycache__/vgg.cpython-38.pyc deleted file mode 100644 index 41d67fb..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/vgg.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/xception.cpython-36.pyc b/AIlib/segutils/core/models/base_models/__pycache__/xception.cpython-36.pyc deleted file mode 100644 index 1c1dfae..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/xception.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/__pycache__/xception.cpython-38.pyc b/AIlib/segutils/core/models/base_models/__pycache__/xception.cpython-38.pyc deleted file mode 100644 index 1d2349a..0000000 Binary files a/AIlib/segutils/core/models/base_models/__pycache__/xception.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/models/base_models/densenet.py b/AIlib/segutils/core/models/base_models/densenet.py deleted file mode 100644 index 733f21d..0000000 --- a/AIlib/segutils/core/models/base_models/densenet.py +++ /dev/null @@ -1,237 +0,0 @@ -import re -import torch -import torch.nn as nn -import torch.nn.functional as F -import torch.utils.model_zoo as model_zoo - -from collections import OrderedDict - -__all__ = ['DenseNet', 'densenet121', 'densenet161', 'densenet169', 'densenet201', - 'dilated_densenet121', 'dilated_densenet161', 'dilated_densenet169', 'dilated_densenet201'] - -model_urls = { - 'densenet121': 'https://download.pytorch.org/models/densenet121-a639ec97.pth', - 'densenet169': 'https://download.pytorch.org/models/densenet169-b2777c0a.pth', - 'densenet201': 'https://download.pytorch.org/models/densenet201-c1103571.pth', - 'densenet161': 'https://download.pytorch.org/models/densenet161-8d451a50.pth', -} - - -class _DenseLayer(nn.Sequential): - def __init__(self, num_input_features, growth_rate, bn_size, drop_rate, dilation=1, norm_layer=nn.BatchNorm2d): - super(_DenseLayer, self).__init__() - self.add_module('norm1', norm_layer(num_input_features)), - self.add_module('relu1', nn.ReLU(True)), - self.add_module('conv1', nn.Conv2d(num_input_features, bn_size * growth_rate, 1, 1, bias=False)), - self.add_module('norm2', norm_layer(bn_size * growth_rate)), - self.add_module('relu2', nn.ReLU(True)), - self.add_module('conv2', nn.Conv2d(bn_size * growth_rate, growth_rate, 3, 1, dilation, dilation, bias=False)), - self.drop_rate = drop_rate - - def forward(self, x): - new_features = super(_DenseLayer, self).forward(x) - if self.drop_rate > 0: - new_features = F.dropout(new_features, p=self.drop_rate, training=self.training) - return torch.cat([x, new_features], 1) - - -class _DenseBlock(nn.Sequential): - def __init__(self, num_layers, num_input_features, bn_size, - growth_rate, drop_rate, dilation=1, norm_layer=nn.BatchNorm2d): - super(_DenseBlock, self).__init__() - for i in range(num_layers): - layer = _DenseLayer(num_input_features + i * growth_rate, - growth_rate, bn_size, drop_rate, dilation, norm_layer) - self.add_module('denselayer%d' % (i + 1), layer) - - -class _Transition(nn.Sequential): - def __init__(self, num_input_features, num_output_features, norm_layer=nn.BatchNorm2d): - super(_Transition, self).__init__() - self.add_module('norm', norm_layer(num_input_features)) - self.add_module('relu', nn.ReLU(True)) - self.add_module('conv', nn.Conv2d(num_input_features, num_output_features, 1, 1, bias=False)) - self.add_module('pool', nn.AvgPool2d(2, 2)) - - -# Net -class DenseNet(nn.Module): #这是一个全新的构建模型的方法，<先构造模块后两步传递数据：features和classifier>；另一种常见的是，<边构造边传递数据> - - def __init__(self, growth_rate=12, block_config=(6, 12, 24, 16), num_init_features=64, - bn_size=4, drop_rate=0, num_classes=1000, norm_layer=nn.BatchNorm2d, **kwargs): - super(DenseNet, self).__init__() - - # First convolution - self.features = nn.Sequential(OrderedDict([ - ('conv0', nn.Conv2d(3, num_init_features, 7, 2, 3, bias=False)), - ('norm0', norm_layer(num_init_features)), - ('relu0', nn.ReLU(True)), - ('pool0', nn.MaxPool2d(3, 2, 1)), - ])) - - # Each denseblock - num_features = num_init_features - for i, num_layers in enumerate(block_config): - block = _DenseBlock(num_layers, num_features, bn_size, growth_rate, drop_rate, norm_layer=norm_layer) - self.features.add_module('denseblock%d' % (i + 1), block) - num_features = num_features + num_layers * growth_rate - if i != len(block_config) - 1: - trans = _Transition(num_features, num_features // 2, norm_layer=norm_layer) - self.features.add_module('transition%d' % (i + 1), trans) - num_features = num_features // 2 - self.num_features = num_features - - # Final batch norm - self.features.add_module('norm5', norm_layer(num_features)) - - # Linear layer - self.classifier = nn.Linear(num_features, num_classes) - - # Official init from torch repo. - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_(m.weight) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.constant_(m.bias, 0) - - def forward(self, x): - features = self.features(x) - print('11',features.shape) - out = F.relu(features, True) - out = F.adaptive_avg_pool2d(out, (1, 1)).view(features.size(0), -1) - out = self.classifier(out) - return out - - -class DilatedDenseNet(DenseNet): - def __init__(self, growth_rate=12, block_config=(6, 12, 24, 16), num_init_features=64, - bn_size=4, drop_rate=0, num_classes=1000, dilate_scale=8, norm_layer=nn.BatchNorm2d, **kwargs): - super(DilatedDenseNet, self).__init__(growth_rate, block_config, num_init_features, - bn_size, drop_rate, num_classes, norm_layer) - assert (dilate_scale == 8 or dilate_scale == 16), "dilate_scale can only set as 8 or 16" - from functools import partial - if dilate_scale == 8: # output_stride - self.features.denseblock3.apply(partial(self._conv_dilate, dilate=2))#利用partial函数给 - self.features.denseblock4.apply(partial(self._conv_dilate, dilate=4)) - del self.features.transition2.pool - del self.features.transition3.pool - elif dilate_scale == 16: - self.features.denseblock4.apply(partial(self._conv_dilate, dilate=2)) - del self.features.transition3.pool - - def _conv_dilate(self, m, dilate): - classname = m.__class__.__name__ - if classname.find('Conv') != -1: - if m.kernel_size == (3, 3): - m.padding = (dilate, dilate) - m.dilation = (dilate, dilate) - - -# Specification -densenet_spec = {121: (64, 32, [6, 12, 24, 16]), - 161: (96, 48, [6, 12, 36, 24]), - 169: (64, 32, [6, 12, 32, 32]), - 201: (64, 32, [6, 12, 48, 32])} - - -# Constructor -def get_densenet(num_layers, pretrained=False, **kwargs): - r"""Densenet-BC model from the - `"Densely Connected Convolutional Networks" `_ paper. - - Parameters - ---------- - num_layers : int - Number of layers for the variant of densenet. Options are 121, 161, 169, 201. - pretrained : bool or str - Boolean value controls whether to load the default pretrained weights for model. - String value represents the hashtag for a certain version of pretrained weights. - root : str, default $TORCH_HOME/models - Location for keeping the model parameters. - """ - num_init_features, growth_rate, block_config = densenet_spec[num_layers] - model = DenseNet(growth_rate, block_config, num_init_features, **kwargs) - if pretrained: - # '.'s are no longer allowed in module names, but pervious _DenseLayer - # has keys 'norm.1', 'relu.1', 'conv.1', 'norm.2', 'relu.2', 'conv.2'. - # They are also in the checkpoints in model_urls. This pattern is used - # to find such keys. - pattern = re.compile( - r'^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$') - state_dict = model_zoo.load_url(model_urls['densenet%d' % num_layers]) - for key in list(state_dict.keys()): - res = pattern.match(key) - if res: - new_key = res.group(1) + res.group(2) - state_dict[new_key] = state_dict[key] - del state_dict[key] - model.load_state_dict(state_dict) #初始化（加载权重） - return model - - -def get_dilated_densenet(num_layers, dilate_scale, pretrained=False, **kwargs): - num_init_features, growth_rate, block_config = densenet_spec[num_layers] - model = DilatedDenseNet(growth_rate, block_config, num_init_features, dilate_scale=dilate_scale) - if pretrained: - # '.'s are no longer allowed in module names, but pervious _DenseLayer - # has keys 'norm.1', 'relu.1', 'conv.1', 'norm.2', 'relu.2', 'conv.2'. - # They are also in the checkpoints in model_urls. This pattern is used - # to find such keys. - pattern = re.compile( - r'^(.*denselayer\d+\.(?:norm|relu|conv))\.((?:[12])\.(?:weight|bias|running_mean|running_var))$') - state_dict = model_zoo.load_url(model_urls['densenet%d' % num_layers]) - for key in list(state_dict.keys()): - res = pattern.match(key) - if res: - new_key = res.group(1) + res.group(2) - state_dict[new_key] = state_dict[key] - del state_dict[key] - model.load_state_dict(state_dict) - return model - - -def densenet121(**kwargs): - return get_densenet(121, **kwargs) - - -def densenet161(**kwargs): - return get_densenet(161, **kwargs) - - -def densenet169(**kwargs): - return get_densenet(169, **kwargs) - - -def densenet201(**kwargs): - return get_densenet(201, **kwargs) - - -def dilated_densenet121(dilate_scale, **kwargs): - return get_dilated_densenet(121, dilate_scale, **kwargs) - - -def dilated_densenet161(dilate_scale, **kwargs): - return get_dilated_densenet(161, dilate_scale, **kwargs) - - -def dilated_densenet169(dilate_scale, **kwargs): - return get_dilated_densenet(169, dilate_scale, **kwargs) - - -def dilated_densenet201(dilate_scale, **kwargs): - return get_dilated_densenet(201, dilate_scale, **kwargs) - - -if __name__ == '__main__': - img = torch.randn(2, 3, 512, 512).cuda() - model = dilated_densenet121(8).cuda() - outputs = model(img) - print(outputs.shape) - from torchsummary import summary - - summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - for name, parameters in model.named_parameters(): - print(name, ':', parameters.size()) diff --git a/AIlib/segutils/core/models/base_models/eespnet.py b/AIlib/segutils/core/models/base_models/eespnet.py deleted file mode 100644 index 7d087fd..0000000 --- a/AIlib/segutils/core/models/base_models/eespnet.py +++ /dev/null @@ -1,202 +0,0 @@ -import math -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.nn import _ConvBNPReLU, _ConvBN, _BNPReLU - -__all__ = ['EESP', 'EESPNet', 'eespnet'] - - -class EESP(nn.Module): - - def __init__(self, in_channels, out_channels, stride=1, k=4, r_lim=7, down_method='esp', norm_layer=nn.BatchNorm2d): - super(EESP, self).__init__() - self.stride = stride - n = int(out_channels / k) - n1 = out_channels - (k - 1) * n - assert down_method in ['avg', 'esp'], 'One of these is suppported (avg or esp)' - assert n == n1, "n(={}) and n1(={}) should be equal for Depth-wise Convolution ".format(n, n1) - self.proj_1x1 = _ConvBNPReLU(in_channels, n, 1, stride=1, groups=k, norm_layer=norm_layer) - - map_receptive_ksize = {3: 1, 5: 2, 7: 3, 9: 4, 11: 5, 13: 6, 15: 7, 17: 8} - self.k_sizes = list() - for i in range(k): - ksize = int(3 + 2 * i) - ksize = ksize if ksize <= r_lim else 3 - self.k_sizes.append(ksize) - self.k_sizes.sort() - self.spp_dw = nn.ModuleList() - for i in range(k): - dilation = map_receptive_ksize[self.k_sizes[i]] - self.spp_dw.append(nn.Conv2d(n, n, 3, stride, dilation, dilation=dilation, groups=n, bias=False)) - self.conv_1x1_exp = _ConvBN(out_channels, out_channels, 1, 1, groups=k, norm_layer=norm_layer) - self.br_after_cat = _BNPReLU(out_channels, norm_layer) - self.module_act = nn.PReLU(out_channels) - self.downAvg = True if down_method == 'avg' else False - - def forward(self, x): - output1 = self.proj_1x1(x) - output = [self.spp_dw[0](output1)] - for k in range(1, len(self.spp_dw)): - out_k = self.spp_dw[k](output1) - out_k = out_k + output[k - 1] - output.append(out_k) - expanded = self.conv_1x1_exp(self.br_after_cat(torch.cat(output, 1))) - del output - if self.stride == 2 and self.downAvg: - return expanded - - if expanded.size() == x.size(): - expanded = expanded + x - - return self.module_act(expanded) - - -class DownSampler(nn.Module): - - def __init__(self, in_channels, out_channels, k=4, r_lim=9, reinf=True, inp_reinf=3, norm_layer=None): - super(DownSampler, self).__init__() - channels_diff = out_channels - in_channels - self.eesp = EESP(in_channels, channels_diff, stride=2, k=k, - r_lim=r_lim, down_method='avg', norm_layer=norm_layer) - self.avg = nn.AvgPool2d(kernel_size=3, padding=1, stride=2) - if reinf: - self.inp_reinf = nn.Sequential( - _ConvBNPReLU(inp_reinf, inp_reinf, 3, 1, 1), - _ConvBN(inp_reinf, out_channels, 1, 1)) - self.act = nn.PReLU(out_channels) - - def forward(self, x, x2=None): - avg_out = self.avg(x) - eesp_out = self.eesp(x) - output = torch.cat([avg_out, eesp_out], 1) - if x2 is not None: - w1 = avg_out.size(2) - while True: - x2 = F.avg_pool2d(x2, kernel_size=3, padding=1, stride=2) - w2 = x2.size(2) - if w2 == w1: - break - output = output + self.inp_reinf(x2) - - return self.act(output) - - -class EESPNet(nn.Module): - def __init__(self, num_classes=1000, scale=1, reinf=True, norm_layer=nn.BatchNorm2d): - super(EESPNet, self).__init__() - inp_reinf = 3 if reinf else None - reps = [0, 3, 7, 3] - r_lim = [13, 11, 9, 7, 5] - K = [4] * len(r_lim) - - # set out_channels - base, levels, base_s = 32, 5, 0 - out_channels = [base] * levels - for i in range(levels): - if i == 0: - base_s = int(base * scale) - base_s = math.ceil(base_s / K[0]) * K[0] - out_channels[i] = base if base_s > base else base_s - else: - out_channels[i] = base_s * pow(2, i) - if scale <= 1.5: - out_channels.append(1024) - elif scale in [1.5, 2]: - out_channels.append(1280) - else: - raise ValueError("Unknown scale value.") - - self.level1 = _ConvBNPReLU(3, out_channels[0], 3, 2, 1, norm_layer=norm_layer) - - self.level2_0 = DownSampler(out_channels[0], out_channels[1], k=K[0], r_lim=r_lim[0], - reinf=reinf, inp_reinf=inp_reinf, norm_layer=norm_layer) - - self.level3_0 = DownSampler(out_channels[1], out_channels[2], k=K[1], r_lim=r_lim[1], - reinf=reinf, inp_reinf=inp_reinf, norm_layer=norm_layer) - self.level3 = nn.ModuleList() - for i in range(reps[1]): - self.level3.append(EESP(out_channels[2], out_channels[2], k=K[2], r_lim=r_lim[2], - norm_layer=norm_layer)) - - self.level4_0 = DownSampler(out_channels[2], out_channels[3], k=K[2], r_lim=r_lim[2], - reinf=reinf, inp_reinf=inp_reinf, norm_layer=norm_layer) - self.level4 = nn.ModuleList() - for i in range(reps[2]): - self.level4.append(EESP(out_channels[3], out_channels[3], k=K[3], r_lim=r_lim[3], - norm_layer=norm_layer)) - - self.level5_0 = DownSampler(out_channels[3], out_channels[4], k=K[3], r_lim=r_lim[3], - reinf=reinf, inp_reinf=inp_reinf, norm_layer=norm_layer) - self.level5 = nn.ModuleList() - for i in range(reps[2]): - self.level5.append(EESP(out_channels[4], out_channels[4], k=K[4], r_lim=r_lim[4], - norm_layer=norm_layer)) - - self.level5.append(_ConvBNPReLU(out_channels[4], out_channels[4], 3, 1, 1, - groups=out_channels[4], norm_layer=norm_layer)) - self.level5.append(_ConvBNPReLU(out_channels[4], out_channels[5], 1, 1, 0, - groups=K[4], norm_layer=norm_layer)) - - self.fc = nn.Linear(out_channels[5], num_classes) - - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, std=0.001) - if m.bias is not None: - nn.init.constant_(m.bias, 0) - - def forward(self, x, seg=True): - out_l1 = self.level1(x) - - out_l2 = self.level2_0(out_l1, x) - - out_l3_0 = self.level3_0(out_l2, x) - for i, layer in enumerate(self.level3): - if i == 0: - out_l3 = layer(out_l3_0) - else: - out_l3 = layer(out_l3) - - out_l4_0 = self.level4_0(out_l3, x) - for i, layer in enumerate(self.level4): - if i == 0: - out_l4 = layer(out_l4_0) - else: - out_l4 = layer(out_l4) - - if not seg: - out_l5_0 = self.level5_0(out_l4) # down-sampled - for i, layer in enumerate(self.level5): - if i == 0: - out_l5 = layer(out_l5_0) - else: - out_l5 = layer(out_l5) - - output_g = F.adaptive_avg_pool2d(out_l5, output_size=1) - output_g = F.dropout(output_g, p=0.2, training=self.training) - output_1x1 = output_g.view(output_g.size(0), -1) - - return self.fc(output_1x1) - return out_l1, out_l2, out_l3, out_l4 - - -def eespnet(pretrained=False, **kwargs): - model = EESPNet(**kwargs) - if pretrained: - raise ValueError("Don't support pretrained") - return model - - -if __name__ == '__main__': - img = torch.randn(1, 3, 224, 224) - model = eespnet() - out = model(img) diff --git a/AIlib/segutils/core/models/base_models/hrnet.py b/AIlib/segutils/core/models/base_models/hrnet.py deleted file mode 100644 index 775b809..0000000 --- a/AIlib/segutils/core/models/base_models/hrnet.py +++ /dev/null @@ -1,371 +0,0 @@ -import torch -import torch.nn as nn - - -class BasicBlock(nn.Module): - expansion = 1 - - def __init__(self, inplanes, planes, stride=1, downsample=None, norm_layer=nn.BatchNorm2d): - super(BasicBlock, self).__init__() - self.conv1 = nn.Conv2d(inplanes, planes, 3, stride, padding=1, bias=False) - self.bn1 = norm_layer(planes) - self.relu = nn.ReLU(True) - self.conv2 = nn.Conv2d(planes, planes, 3, padding=1, bias=False) - self.bn2 = norm_layer(planes) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - identity = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - - if self.downsample is not None: - identity = self.downsample(x) - - out += identity - out = self.relu(out) - - return out - - -class Bottleneck(nn.Module): - expansion = 4 - - def __init__(self, inplanes, planes, stride=1, downsample=None, norm_layer=nn.BatchNorm2d): - super(Bottleneck, self).__init__() - self.conv1 = nn.Conv2d(inplanes, planes, 1, bias=False) - self.bn1 = norm_layer(planes) - self.conv2 = nn.Conv2d(planes, planes, 3, stride, 1, bias=False) - self.bn2 = norm_layer(planes) - self.conv3 = nn.Conv2d(planes, planes * self.expansion, 1, bias=False) - self.bn3 = norm_layer(planes * self.expansion) - self.relu = nn.ReLU(True) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - identity = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - identity = self.downsample(x) - - out += identity - out = self.relu(out) - - return out - - -class HighResolutionModule(nn.Module): - def __init__(self, num_branches, blocks, num_blocks, num_inchannels, num_channels, - fuse_method, multi_scale_output=True, norm_layer=nn.BatchNorm2d): - super(HighResolutionModule, self).__init__() - assert num_branches == len(num_blocks) - assert num_branches == len(num_channels) - assert num_branches == len(num_inchannels) - - self.num_inchannels = num_inchannels - self.fuse_method = fuse_method - self.num_branches = num_branches - self.multi_scale_output = multi_scale_output - - self.branches = self._make_branches(num_branches, blocks, num_blocks, num_channels, norm_layer=norm_layer) - self.fuse_layers = self._make_fuse_layers(norm_layer) - self.relu = nn.ReLU(True) - - def _make_one_branch(self, branch_index, block, num_blocks, num_channels, - stride=1, norm_layer=nn.BatchNorm2d): - downsample = None - if stride != 1 or self.num_inchannels[branch_index] != num_channels[branch_index] * block.expansion: - downsample = nn.Sequential( - nn.Conv2d(self.num_inchannels[branch_index], num_channels[branch_index] * block.expansion, - 1, stride, bias=False), - norm_layer(num_channels[branch_index] * block.expansion)) - - layers = list() - layers.append(block(self.num_inchannels[branch_index], num_channels[branch_index], - stride, downsample, norm_layer=norm_layer)) - self.num_inchannels[branch_index] = num_channels[branch_index] * block.expansion - for i in range(1, num_blocks[branch_index]): - layers.append(block(self.num_inchannels[branch_index], num_channels[branch_index], norm_layer=norm_layer)) - - return nn.Sequential(*layers) - - def _make_branches(self, num_branches, block, num_blocks, num_channels, norm_layer=nn.BatchNorm2d): - branches = list() - for i in range(num_branches): - branches.append( - self._make_one_branch(i, block, num_blocks, num_channels, norm_layer=norm_layer)) - - return nn.ModuleList(branches) - - def _make_fuse_layers(self, norm_layer=nn.BatchNorm2d): - if self.num_branches == 1: - return None - - num_branches = self.num_branches - num_inchannels = self.num_inchannels - fuse_layers = [] - for i in range(num_branches if self.multi_scale_output else 1): - fuse_layer = list() - for j in range(num_branches): - if j > i: - fuse_layer.append(nn.Sequential( - nn.Conv2d(num_inchannels[j], num_inchannels[i], 1, bias=False), - norm_layer(num_inchannels[i]), - nn.Upsample(scale_factor=2 ** (j - i), mode='nearest'))) - elif j == i: - fuse_layer.append(None) - else: - conv3x3s = list() - for k in range(i - j): - if k == i - j - 1: - num_outchannels_conv3x3 = num_inchannels[i] - conv3x3s.append(nn.Sequential( - nn.Conv2d(num_inchannels[j], num_outchannels_conv3x3, 3, 2, 1, bias=False), - norm_layer(num_outchannels_conv3x3))) - else: - num_outchannels_conv3x3 = num_inchannels[j] - conv3x3s.append(nn.Sequential( - nn.Conv2d(num_inchannels[j], num_outchannels_conv3x3, 3, 2, 1, bias=False), - norm_layer(num_outchannels_conv3x3), - nn.ReLU(False))) - fuse_layer.append(nn.Sequential(*conv3x3s)) - fuse_layers.append(nn.ModuleList(fuse_layer)) - - return nn.ModuleList(fuse_layers) - - def get_num_inchannels(self): - return self.num_inchannels - - def forward(self, x): - if self.num_branches == 1: - return [self.branches[0](x[0])] - - for i in range(self.num_branches): - x[i] = self.branches[i](x[i]) - - x_fuse = list() - for i in range(len(self.fuse_layers)): - y = x[0] if i == 0 else self.fuse_layers[i][0](x[0]) - for j in range(1, self.num_branches): - if i == j: - y = y + x[j] - else: - y = y + self.fuse_layers[i][j](x[j]) - x_fuse.append(self.relu(y)) - - return x_fuse - - -class HighResolutionNet(nn.Module): - def __init__(self, blocks, num_channels, num_modules, num_branches, num_blocks, - fuse_method, norm_layer=nn.BatchNorm2d, **kwargs): - super(HighResolutionNet, self).__init__() - self.num_branches = num_branches - - # deep stem - self.conv1 = nn.Sequential( - nn.Conv2d(3, 64, 3, 2, 1, bias=False), - norm_layer(64), - nn.ReLU(True), - nn.Conv2d(64, 64, 3, 2, 1, bias=False), - norm_layer(64), - nn.ReLU(True)) - - self.layer1 = self._make_layer(Bottleneck, 64, 64, 4, norm_layer=norm_layer) - - # stage 2 - num_channel, block = num_channels[0], blocks[0] - channels = [channel * block.expansion for channel in num_channel] - self.transition1 = self._make_transition_layer([256], channels, norm_layer) - self.stage2, pre_stage_channels = self._make_stage(num_modules[0], num_branches[0], - num_blocks[0], channels, block, - fuse_method[0], channels, - norm_layer=norm_layer) - - # stage 3 - num_channel, block = num_channels[1], blocks[1] - channels = [channel * block.expansion for channel in num_channel] - self.transition1 = self._make_transition_layer(pre_stage_channels, channels, norm_layer) - self.stage3, pre_stage_channels = self._make_stage(num_modules[1], num_branches[1], - num_blocks[1], channels, block, - fuse_method[1], channels, - norm_layer=norm_layer) - - # stage 4 - num_channel, block = num_channels[2], blocks[2] - channels = [channel * block.expansion for channel in num_channel] - self.transition1 = self._make_transition_layer(pre_stage_channels, channels, norm_layer) - self.stage4, pre_stage_channels = self._make_stage(num_modules[2], num_branches[2], - num_blocks[2], channels, block, - fuse_method[2], channels, - norm_layer=norm_layer) - - self.incre_modules, self.downsamp_modules, self.final_layer = self._make_head(pre_stage_channels, norm_layer) - - self.classifier = nn.Linear(2048, 1000) - - def _make_layer(self, block, inplanes, planes, blocks, stride=1, norm_layer=nn.BatchNorm2d): - downsample = None - if stride != 1 or inplanes != planes * block.expansion: - downsample = nn.Sequential( - nn.Conv2d(inplanes, planes * block.expansion, 1, stride, bias=False), - norm_layer(planes * block.expansion)) - - layers = list() - layers.append(block(inplanes, planes, stride, downsample=downsample, norm_layer=norm_layer)) - inplanes = planes * block.expansion - for _ in range(1, blocks): - layers.append(block(inplanes, planes, norm_layer=norm_layer)) - - return nn.Sequential(*layers) - - def _make_transition_layer(self, num_channels_pre_layer, num_channels_cur_layer, norm_layer=nn.BatchNorm2d): - num_branches_cur = len(num_channels_cur_layer) - num_branches_pre = len(num_channels_pre_layer) - - transition_layers = list() - for i in range(num_branches_cur): - if i < num_branches_pre: - if num_channels_cur_layer[i] != num_channels_pre_layer[i]: - transition_layers.append(nn.Sequential( - nn.Conv2d(num_channels_pre_layer[i], num_channels_cur_layer[i], 3, padding=1, bias=False), - norm_layer(num_channels_cur_layer[i]), - nn.ReLU(True))) - else: - transition_layers.append(None) - else: - conv3x3s = list() - for j in range(i + 1 - num_branches_pre): - in_channels = num_channels_pre_layer[-1] - out_channels = num_channels_cur_layer[i] if j == i - num_branches_pre else in_channels - conv3x3s.append(nn.Sequential( - nn.Conv2d(in_channels, out_channels, 3, 2, 1, bias=False), - norm_layer(out_channels), - nn.ReLU(True))) - transition_layers.append(nn.Sequential(*conv3x3s)) - - return nn.ModuleList(transition_layers) - - def _make_stage(self, num_modules, num_branches, num_blocks, num_channels, block, - fuse_method, num_inchannels, multi_scale_output=True, norm_layer=nn.BatchNorm2d): - modules = list() - for i in range(num_modules): - # multi_scale_output is only used last module - if not multi_scale_output and i == num_modules - 1: - reset_multi_scale_output = False - else: - reset_multi_scale_output = True - - modules.append(HighResolutionModule(num_branches, block, num_blocks, num_inchannels, num_channels, - fuse_method, reset_multi_scale_output, norm_layer=norm_layer)) - num_inchannels = modules[-1].get_num_inchannels() - - return nn.Sequential(*modules), num_inchannels - - def _make_head(self, pre_stage_channels, norm_layer=nn.BatchNorm2d): - head_block = Bottleneck - head_channels = [32, 64, 128, 256] - - # Increasing the #channels on each resolution - # from C, 2C, 4C, 8C to 128, 256, 512, 1024 - incre_modules = list() - for i, channels in enumerate(pre_stage_channels): - incre_module = self._make_layer(head_block, channels, head_channels[i], 1) - incre_modules.append(incre_module) - incre_modules = nn.ModuleList(incre_modules) - - # downsampling modules - downsamp_modules = [] - for i in range(len(pre_stage_channels) - 1): - in_channels = head_channels[i] * head_block.expansion - out_channels = head_channels[i + 1] * head_block.expansion - - downsamp_module = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 3, 2, 1), - norm_layer(out_channels), - nn.ReLU(True)) - - downsamp_modules.append(downsamp_module) - downsamp_modules = nn.ModuleList(downsamp_modules) - - final_layer = nn.Sequential( - nn.Conv2d(head_channels[3] * head_block.expansion, 2048, 1), - norm_layer(2048), - nn.ReLU(True)) - - return incre_modules, downsamp_modules, final_layer - - def forward(self, x): - x = self.conv1(x) - x = self.layer1(x) - - x_list = list() - for i in range(self.num_branches[0]): - if self.transition1[i] is not None: - tmp = self.transition1[i](x) - print(tmp.size()) - x_list.append(self.transition1[i](x)) - else: - x_list.append(x) - y_list = self.stage2(x_list) - - x_list = [] - for i in range(self.num_branches[1]): - if self.transition2[i] is not None: - x_list.append(self.transition2[i](y_list[-1])) - else: - x_list.append(y_list[i]) - y_list = self.stage3(x_list) - - x_list = [] - for i in range(self.num_branches[2]): - if self.transition3[i] is not None: - x_list.append(self.transition3[i](y_list[-1])) - else: - x_list.append(y_list[i]) - y_list = self.stage4(x_list) - - # Classification Head - y = self.incre_modules[0](y_list[0]) - for i in range(len(self.downsamp_modules)): - y = self.incre_modules[i + 1](y_list[i + 1]) + self.downsamp_modules[i](y) - - y = self.final_layer(y) - - y = F.avg_pool2d(y, kernel_size=y.size() - [2:]).view(y.size(0), -1) - - y = self.classifier(y) - - return y - - -blocks = [BasicBlock, BasicBlock, BasicBlock] -num_modules = [1, 1, 1] -num_branches = [2, 3, 4] -num_blocks = [[4, 4], [4, 4, 4], [4, 4, 4, 4]] -num_channels = [[256, 256], [32, 64, 128], [32, 64, 128, 256]] -fuse_method = ['sum', 'sum', 'sum'] - -if __name__ == '__main__': - img = torch.randn(1, 3, 256, 256) - model = HighResolutionNet(blocks, num_channels, num_modules, num_branches, num_blocks, fuse_method) - output = model(img) diff --git a/AIlib/segutils/core/models/base_models/mobilenetv2.py b/AIlib/segutils/core/models/base_models/mobilenetv2.py deleted file mode 100644 index 4e4c093..0000000 --- a/AIlib/segutils/core/models/base_models/mobilenetv2.py +++ /dev/null @@ -1,158 +0,0 @@ -"""MobileNet and MobileNetV2.""" -import torch -import torch.nn as nn - -from core.nn import _ConvBNReLU, _DepthwiseConv, InvertedResidual - -__all__ = ['MobileNet', 'MobileNetV2', 'get_mobilenet', 'get_mobilenet_v2', - 'mobilenet1_0', 'mobilenet_v2_1_0', 'mobilenet0_75', 'mobilenet_v2_0_75', - 'mobilenet0_5', 'mobilenet_v2_0_5', 'mobilenet0_25', 'mobilenet_v2_0_25'] - - -class MobileNet(nn.Module): - def __init__(self, num_classes=1000, multiplier=1.0, norm_layer=nn.BatchNorm2d, **kwargs): - super(MobileNet, self).__init__() - conv_dw_setting = [ - [64, 1, 1], - [128, 2, 2], - [256, 2, 2], - [512, 6, 2], - [1024, 2, 2]] - input_channels = int(32 * multiplier) if multiplier > 1.0 else 32 - features = [_ConvBNReLU(3, input_channels, 3, 2, 1, norm_layer=norm_layer)] - - for c, n, s in conv_dw_setting: - out_channels = int(c * multiplier) - for i in range(n): - stride = s if i == 0 else 1 - features.append(_DepthwiseConv(input_channels, out_channels, stride, norm_layer)) - input_channels = out_channels - features.append(nn.AdaptiveAvgPool2d(1)) - self.features = nn.Sequential(*features) - - self.classifier = nn.Linear(int(1024 * multiplier), num_classes) - - # weight initialization - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_(m.weight, mode='fan_out') - if m.bias is not None: - nn.init.zeros_(m.bias) - elif isinstance(m, nn.BatchNorm2d): - nn.init.ones_(m.weight) - nn.init.zeros_(m.bias) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - nn.init.zeros_(m.bias) - - def forward(self, x): - x = self.features(x) - x = self.classifier(x.view(x.size(0), x.size(1))) - return x - - -class MobileNetV2(nn.Module): - def __init__(self, num_classes=1000, multiplier=1.0, norm_layer=nn.BatchNorm2d, **kwargs): - super(MobileNetV2, self).__init__() - inverted_residual_setting = [ - # t, c, n, s - [1, 16, 1, 1], - [6, 24, 2, 2], - [6, 32, 3, 2], - [6, 64, 4, 2], - [6, 96, 3, 1], - [6, 160, 3, 2], - [6, 320, 1, 1]] - # building first layer - input_channels = int(32 * multiplier) if multiplier > 1.0 else 32 - last_channels = int(1280 * multiplier) if multiplier > 1.0 else 1280 - features = [_ConvBNReLU(3, input_channels, 3, 2, 1, relu6=True, norm_layer=norm_layer)] - - # building inverted residual blocks - for t, c, n, s in inverted_residual_setting: - out_channels = int(c * multiplier) - for i in range(n): - stride = s if i == 0 else 1 - features.append(InvertedResidual(input_channels, out_channels, stride, t, norm_layer)) - input_channels = out_channels - - # building last several layers - features.append(_ConvBNReLU(input_channels, last_channels, 1, relu6=True, norm_layer=norm_layer)) - features.append(nn.AdaptiveAvgPool2d(1)) - self.features = nn.Sequential(*features) - - self.classifier = nn.Sequential( - nn.Dropout2d(0.2), - nn.Linear(last_channels, num_classes)) - - # weight initialization - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_(m.weight, mode='fan_out') - if m.bias is not None: - nn.init.zeros_(m.bias) - elif isinstance(m, nn.BatchNorm2d): - nn.init.ones_(m.weight) - nn.init.zeros_(m.bias) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - if m.bias is not None: - nn.init.zeros_(m.bias) - - def forward(self, x): - x = self.features(x) - x = self.classifier(x.view(x.size(0), x.size(1))) - return x - - -# Constructor -def get_mobilenet(multiplier=1.0, pretrained=False, root='~/.torch/models', **kwargs): - model = MobileNet(multiplier=multiplier, **kwargs) - - if pretrained: - raise ValueError("Not support pretrained") - return model - - -def get_mobilenet_v2(multiplier=1.0, pretrained=False, root='~/.torch/models', **kwargs): - model = MobileNetV2(multiplier=multiplier, **kwargs) - - if pretrained: - raise ValueError("Not support pretrained") - return model - - -def mobilenet1_0(**kwargs): - return get_mobilenet(1.0, **kwargs) - - -def mobilenet_v2_1_0(**kwargs): - return get_mobilenet_v2(1.0, **kwargs) - - -def mobilenet0_75(**kwargs): - return get_mobilenet(0.75, **kwargs) - - -def mobilenet_v2_0_75(**kwargs): - return get_mobilenet_v2(0.75, **kwargs) - - -def mobilenet0_5(**kwargs): - return get_mobilenet(0.5, **kwargs) - - -def mobilenet_v2_0_5(**kwargs): - return get_mobilenet_v2(0.5, **kwargs) - - -def mobilenet0_25(**kwargs): - return get_mobilenet(0.25, **kwargs) - - -def mobilenet_v2_0_25(**kwargs): - return get_mobilenet_v2(0.25, **kwargs) - - -if __name__ == '__main__': - model = mobilenet0_5() diff --git a/AIlib/segutils/core/models/base_models/resnet.py b/AIlib/segutils/core/models/base_models/resnet.py deleted file mode 100644 index b95b8c6..0000000 --- a/AIlib/segutils/core/models/base_models/resnet.py +++ /dev/null @@ -1,226 +0,0 @@ -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['ResNet', 'resnet18', 'resnet34', 'resnet50', 'resnet101', - 'resnet152'] - -model_urls = { - 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth', - 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', - 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', - 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', - 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', -} - - -def conv3x3(in_planes, out_planes, stride=1): - """3x3 convolution with padding""" - return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride, - padding=1, bias=False) - - -def conv1x1(in_planes, out_planes, stride=1): - """1x1 convolution""" - return nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False) - - -class BasicBlock(nn.Module): - expansion = 1 - - def __init__(self, inplanes, planes, stride=1, downsample=None, norm_layer=nn.BatchNorm2d): - super(BasicBlock, self).__init__() - # Both self.conv1 and self.downsample layers downsample the input when stride != 1 - self.conv1 = conv3x3(inplanes, planes, stride) - self.bn1 = norm_layer(planes) - self.relu = nn.ReLU(inplace=True) - self.conv2 = conv3x3(planes, planes) - self.bn2 = norm_layer(planes) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - identity = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - - if self.downsample is not None: - identity = self.downsample(x) - - out += identity - out = self.relu(out) - - return out - - -class Bottleneck(nn.Module): - expansion = 4 - - def __init__(self, inplanes, planes, stride=1, downsample=None, norm_layer=nn.BatchNorm2d): - super(Bottleneck, self).__init__() - # Both self.conv2 and self.downsample layers downsample the input when stride != 1 - self.conv1 = conv1x1(inplanes, planes) - self.bn1 = norm_layer(planes) - self.conv2 = conv3x3(planes, planes, stride) - self.bn2 = norm_layer(planes) - self.conv3 = conv1x1(planes, planes * self.expansion) - self.bn3 = norm_layer(planes * self.expansion) - self.relu = nn.ReLU(inplace=True) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - identity = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - identity = self.downsample(x) - - out += identity - out = self.relu(out) - - return out - - -class ResNet(nn.Module): - - def __init__(self, block, layers, num_classes=1000, zero_init_residual=False, norm_layer=nn.BatchNorm2d): - super(ResNet, self).__init__() - self.inplanes = 64 - self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, - bias=False) - self.bn1 = norm_layer(64) - self.relu = nn.ReLU(inplace=True) - self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1) - self.layer1 = self._make_layer(block, 64, layers[0], norm_layer=norm_layer) - self.layer2 = self._make_layer(block, 128, layers[1], stride=2, norm_layer=norm_layer) - self.layer3 = self._make_layer(block, 256, layers[2], stride=2, norm_layer=norm_layer) - self.layer4 = self._make_layer(block, 512, layers[3], stride=2, norm_layer=norm_layer) - self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) - self.fc = nn.Linear(512 * block.expansion, num_classes) - - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - - # Zero-initialize the last BN in each residual branch, - # so that the residual branch starts with zeros, and each residual block behaves like an identity. - # This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677 - if zero_init_residual: - for m in self.modules(): - if isinstance(m, Bottleneck): - nn.init.constant_(m.bn3.weight, 0) - elif isinstance(m, BasicBlock): - nn.init.constant_(m.bn2.weight, 0) - - def _make_layer(self, block, planes, blocks, stride=1, norm_layer=nn.BatchNorm2d): - downsample = None - if stride != 1 or self.inplanes != planes * block.expansion: - downsample = nn.Sequential( - conv1x1(self.inplanes, planes * block.expansion, stride), - norm_layer(planes * block.expansion), - ) - - layers = [] - layers.append(block(self.inplanes, planes, stride, downsample)) - self.inplanes = planes * block.expansion - for _ in range(1, blocks): - layers.append(block(self.inplanes, planes)) - - return nn.Sequential(*layers) - - def forward(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - x = self.maxpool(x) - - x = self.layer1(x) - x = self.layer2(x) - x = self.layer3(x) - x = self.layer4(x) - - x = self.avgpool(x) - x = x.view(x.size(0), -1) - x = self.fc(x) - - return x - - -def resnet18(pretrained=False, **kwargs): - """Constructs a ResNet-18 model. - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - model = ResNet(BasicBlock, [2, 2, 2, 2], **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['resnet18'])) - return model - - -def resnet34(pretrained=False, **kwargs): - """Constructs a ResNet-34 model. - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - model = ResNet(BasicBlock, [3, 4, 6, 3], **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['resnet34'])) - return model - - -def resnet50(pretrained=False, **kwargs): - """Constructs a ResNet-50 model. - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - model = ResNet(Bottleneck, [3, 4, 6, 3], **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['resnet50'])) - return model - - -def resnet101(pretrained=False, **kwargs): - """Constructs a ResNet-101 model. - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - model = ResNet(Bottleneck, [3, 4, 23, 3], **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['resnet101'])) - return model - - -def resnet152(pretrained=False, **kwargs): - """Constructs a ResNet-152 model. - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - model = ResNet(Bottleneck, [3, 8, 36, 3], **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['resnet152'])) - return model - - -if __name__ == '__main__': - import torch - img = torch.randn(4, 3, 224, 224) - model = resnet50(True) - output = model(img) \ No newline at end of file diff --git a/AIlib/segutils/core/models/base_models/resnetv1b.py b/AIlib/segutils/core/models/base_models/resnetv1b.py deleted file mode 100644 index 21d67b7..0000000 --- a/AIlib/segutils/core/models/base_models/resnetv1b.py +++ /dev/null @@ -1,264 +0,0 @@ -import torch -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['ResNetV1b', 'resnet18_v1b', 'resnet34_v1b', 'resnet50_v1b', - 'resnet101_v1b', 'resnet152_v1b', 'resnet152_v1s', 'resnet101_v1s', 'resnet50_v1s'] - -model_urls = { - 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth', - 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth', - 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth', - 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth', - 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth', -} - - -class BasicBlockV1b(nn.Module): - expansion = 1 - - def __init__(self, inplanes, planes, stride=1, dilation=1, downsample=None, - previous_dilation=1, norm_layer=nn.BatchNorm2d): - super(BasicBlockV1b, self).__init__() - self.conv1 = nn.Conv2d(inplanes, planes, 3, stride, - dilation, dilation, bias=False) - self.bn1 = norm_layer(planes) - self.relu = nn.ReLU(True) - self.conv2 = nn.Conv2d(planes, planes, 3, 1, previous_dilation, - dilation=previous_dilation, bias=False) - self.bn2 = norm_layer(planes) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - identity = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - - if self.downsample is not None: - identity = self.downsample(x) - - out += identity - out = self.relu(out) - - return out - - -class BottleneckV1b(nn.Module): - expansion = 4 - - def __init__(self, inplanes, planes, stride=1, dilation=1, downsample=None, - previous_dilation=1, norm_layer=nn.BatchNorm2d): - super(BottleneckV1b, self).__init__() - self.conv1 = nn.Conv2d(inplanes, planes, 1, bias=False) - self.bn1 = norm_layer(planes) - self.conv2 = nn.Conv2d(planes, planes, 3, stride, - dilation, dilation, bias=False) - self.bn2 = norm_layer(planes) - self.conv3 = nn.Conv2d(planes, planes * self.expansion, 1, bias=False) - self.bn3 = norm_layer(planes * self.expansion) - self.relu = nn.ReLU(True) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - identity = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - identity = self.downsample(x) - - out += identity - out = self.relu(out) - - return out - - -class ResNetV1b(nn.Module): - - def __init__(self, block, layers, num_classes=1000, dilated=True, deep_stem=False, - zero_init_residual=False, norm_layer=nn.BatchNorm2d): - self.inplanes = 128 if deep_stem else 64 - super(ResNetV1b, self).__init__() - if deep_stem: - self.conv1 = nn.Sequential( - nn.Conv2d(3, 64, 3, 2, 1, bias=False), - norm_layer(64), - nn.ReLU(True), - nn.Conv2d(64, 64, 3, 1, 1, bias=False), - norm_layer(64), - nn.ReLU(True), - nn.Conv2d(64, 128, 3, 1, 1, bias=False) - ) - else: - self.conv1 = nn.Conv2d(3, 64, 7, 2, 3, bias=False) - self.bn1 = norm_layer(self.inplanes) - self.relu = nn.ReLU(True) - self.maxpool = nn.MaxPool2d(3, 2, 1) - self.layer1 = self._make_layer(block, 64, layers[0], norm_layer=norm_layer) - self.layer2 = self._make_layer(block, 128, layers[1], stride=2, norm_layer=norm_layer) - if dilated: - self.layer3 = self._make_layer(block, 256, layers[2], stride=1, dilation=2, norm_layer=norm_layer) - self.layer4 = self._make_layer(block, 512, layers[3], stride=1, dilation=4, norm_layer=norm_layer) - else: - self.layer3 = self._make_layer(block, 256, layers[2], stride=2, norm_layer=norm_layer) - self.layer4 = self._make_layer(block, 512, layers[3], stride=2, norm_layer=norm_layer) - self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) - self.fc = nn.Linear(512 * block.expansion, num_classes) - - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - - if zero_init_residual: - for m in self.modules(): - if isinstance(m, BottleneckV1b): - nn.init.constant_(m.bn3.weight, 0) - elif isinstance(m, BasicBlockV1b): - nn.init.constant_(m.bn2.weight, 0) - - def _make_layer(self, block, planes, blocks, stride=1, dilation=1, norm_layer=nn.BatchNorm2d): - downsample = None - if stride != 1 or self.inplanes != planes * block.expansion: - downsample = nn.Sequential( - nn.Conv2d(self.inplanes, planes * block.expansion, 1, stride, bias=False), - norm_layer(planes * block.expansion), - ) - - layers = [] - if dilation in (1, 2): - layers.append(block(self.inplanes, planes, stride, dilation=1, downsample=downsample, - previous_dilation=dilation, norm_layer=norm_layer)) - elif dilation == 4: - layers.append(block(self.inplanes, planes, stride, dilation=2, downsample=downsample, - previous_dilation=dilation, norm_layer=norm_layer)) - else: - raise RuntimeError("=> unknown dilation size: {}".format(dilation)) - self.inplanes = planes * block.expansion - for _ in range(1, blocks): - layers.append(block(self.inplanes, planes, dilation=dilation, - previous_dilation=dilation, norm_layer=norm_layer)) - - return nn.Sequential(*layers) - - def forward(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - x = self.maxpool(x) - - x = self.layer1(x) - x = self.layer2(x) - x = self.layer3(x) - x = self.layer4(x) - - x = self.avgpool(x) - x = x.view(x.size(0), -1) - x = self.fc(x) - - return x - - -def resnet18_v1b(pretrained=False, **kwargs): - model = ResNetV1b(BasicBlockV1b, [2, 2, 2, 2], **kwargs) - if pretrained: - old_dict = model_zoo.load_url(model_urls['resnet18']) - model_dict = model.state_dict() - old_dict = {k: v for k, v in old_dict.items() if (k in model_dict)} - model_dict.update(old_dict) - model.load_state_dict(model_dict) - return model - - -def resnet34_v1b(pretrained=False, **kwargs): - model = ResNetV1b(BasicBlockV1b, [3, 4, 6, 3], **kwargs) - if pretrained: - old_dict = model_zoo.load_url(model_urls['resnet34']) - model_dict = model.state_dict() - old_dict = {k: v for k, v in old_dict.items() if (k in model_dict)} - model_dict.update(old_dict) - model.load_state_dict(model_dict) - return model - - -def resnet50_v1b(pretrained=False, **kwargs): - model = ResNetV1b(BottleneckV1b, [3, 4, 6, 3], **kwargs) - if pretrained: - old_dict = model_zoo.load_url(model_urls['resnet50']) - model_dict = model.state_dict() - old_dict = {k: v for k, v in old_dict.items() if (k in model_dict)} - model_dict.update(old_dict) - model.load_state_dict(model_dict) - return model - - -def resnet101_v1b(pretrained=False, **kwargs): - model = ResNetV1b(BottleneckV1b, [3, 4, 23, 3], **kwargs) - if pretrained: - old_dict = model_zoo.load_url(model_urls['resnet101']) - model_dict = model.state_dict() - old_dict = {k: v for k, v in old_dict.items() if (k in model_dict)} - model_dict.update(old_dict) - model.load_state_dict(model_dict) - return model - - -def resnet152_v1b(pretrained=False, **kwargs): - model = ResNetV1b(BottleneckV1b, [3, 8, 36, 3], **kwargs) - if pretrained: - old_dict = model_zoo.load_url(model_urls['resnet152']) - model_dict = model.state_dict() - old_dict = {k: v for k, v in old_dict.items() if (k in model_dict)} - model_dict.update(old_dict) - model.load_state_dict(model_dict) - return model - - -def resnet50_v1s(pretrained=False, root='~/.torch/models', **kwargs): - model = ResNetV1b(BottleneckV1b, [3, 4, 6, 3], deep_stem=True, **kwargs) - if pretrained: - from ..model_store import get_resnet_file - model.load_state_dict(torch.load(get_resnet_file('resnet50', root=root)), strict=False) - return model - - -def resnet101_v1s(pretrained=False, root='~/.torch/models', **kwargs): - model = ResNetV1b(BottleneckV1b, [3, 4, 23, 3], deep_stem=True, **kwargs) - if pretrained: - from ..model_store import get_resnet_file - model.load_state_dict(torch.load(get_resnet_file('resnet101', root=root)), strict=False) - return model - - -def resnet152_v1s(pretrained=False, root='~/.torch/models', **kwargs): - model = ResNetV1b(BottleneckV1b, [3, 8, 36, 3], deep_stem=True, **kwargs) - if pretrained: - from ..model_store import get_resnet_file - model.load_state_dict(torch.load(get_resnet_file('resnet152', root=root)), strict=False) - return model - - -if __name__ == '__main__': - import torch - - img = torch.randn(4, 3, 224, 224) - model = resnet50_v1b(True) - output = model(img) diff --git a/AIlib/segutils/core/models/base_models/resnext.py b/AIlib/segutils/core/models/base_models/resnext.py deleted file mode 100644 index 8daf287..0000000 --- a/AIlib/segutils/core/models/base_models/resnext.py +++ /dev/null @@ -1,154 +0,0 @@ -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = ['ResNext', 'resnext50_32x4d', 'resnext101_32x8d'] - -model_urls = { - 'resnext50_32x4d': 'https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth', - 'resnext101_32x8d': 'https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth', -} - - -class Bottleneck(nn.Module): - expansion = 4 - - def __init__(self, inplanes, planes, stride=1, downsample=None, groups=1, - base_width=64, dilation=1, norm_layer=None, **kwargs): - super(Bottleneck, self).__init__() - width = int(planes * (base_width / 64.)) * groups - - self.conv1 = nn.Conv2d(inplanes, width, 1, bias=False) - self.bn1 = norm_layer(width) - self.conv2 = nn.Conv2d(width, width, 3, stride, dilation, dilation, groups, bias=False) - self.bn2 = norm_layer(width) - self.conv3 = nn.Conv2d(width, planes * self.expansion, 1, bias=False) - self.bn3 = norm_layer(planes * self.expansion) - self.relu = nn.ReLU(True) - self.downsample = downsample - self.stride = stride - - def forward(self, x): - identity = x - - out = self.conv1(x) - out = self.bn1(out) - out = self.relu(out) - - out = self.conv2(out) - out = self.bn2(out) - out = self.relu(out) - - out = self.conv3(out) - out = self.bn3(out) - - if self.downsample is not None: - identity = self.downsample(x) - - out += identity - out = self.relu(out) - - return out - - -class ResNext(nn.Module): - - def __init__(self, block, layers, num_classes=1000, zero_init_residual=False, groups=1, - width_per_group=64, dilated=False, norm_layer=nn.BatchNorm2d, **kwargs): - super(ResNext, self).__init__() - self.inplanes = 64 - self.groups = groups - self.base_width = width_per_group - - self.conv1 = nn.Conv2d(3, self.inplanes, 7, 2, 3, bias=False) - self.bn1 = norm_layer(self.inplanes) - self.relu = nn.ReLU(True) - self.maxpool = nn.MaxPool2d(3, 2, 1) - - self.layer1 = self._make_layer(block, 64, layers[0], norm_layer=norm_layer) - self.layer2 = self._make_layer(block, 128, layers[1], stride=2, norm_layer=norm_layer) - if dilated: - self.layer3 = self._make_layer(block, 256, layers[2], stride=1, dilation=2, norm_layer=norm_layer) - self.layer4 = self._make_layer(block, 512, layers[3], stride=1, dilation=4, norm_layer=norm_layer) - else: - self.layer3 = self._make_layer(block, 256, layers[2], stride=2, norm_layer=norm_layer) - self.layer4 = self._make_layer(block, 512, layers[3], stride=2, norm_layer=norm_layer) - - self.avgpool = nn.AdaptiveAvgPool2d((1, 1)) - self.fc = nn.Linear(512 * block.expansion, num_classes) - - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - - if zero_init_residual: - for m in self.modules(): - if isinstance(m, Bottleneck): - nn.init.constant_(m.bn3.weight, 0) - - def _make_layer(self, block, planes, blocks, stride=1, dilation=1, norm_layer=nn.BatchNorm2d): - downsample = None - if stride != 1 or self.inplanes != planes * block.expansion: - downsample = nn.Sequential( - nn.Conv2d(self.inplanes, planes * block.expansion, 1, stride, bias=False), - norm_layer(planes * block.expansion) - ) - - layers = list() - if dilation in (1, 2): - layers.append(block(self.inplanes, planes, stride, downsample, self.groups, - self.base_width, norm_layer=norm_layer)) - elif dilation == 4: - layers.append(block(self.inplanes, planes, stride, downsample, self.groups, - self.base_width, dilation=2, norm_layer=norm_layer)) - else: - raise RuntimeError("=> unknown dilation size: {}".format(dilation)) - self.inplanes = planes * block.expansion - for _ in range(1, blocks): - layers.append(block(self.inplanes, planes, groups=self.groups, base_width=self.base_width, - dilation=dilation, norm_layer=norm_layer)) - - return nn.Sequential(*layers) - - def forward(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - x = self.maxpool(x) - - x = self.layer1(x) - x = self.layer2(x) - x = self.layer3(x) - x = self.layer4(x) - - x = self.avgpool(x) - x = x.view(x.size(0), -1) - x = self.fc(x) - - return x - - -def resnext50_32x4d(pretrained=False, **kwargs): - kwargs['groups'] = 32 - kwargs['width_per_group'] = 4 - model = ResNext(Bottleneck, [3, 4, 6, 3], **kwargs) - if pretrained: - state_dict = model_zoo.load_url(model_urls['resnext50_32x4d']) - model.load_state_dict(state_dict) - return model - - -def resnext101_32x8d(pretrained=False, **kwargs): - kwargs['groups'] = 32 - kwargs['width_per_group'] = 8 - model = ResNext(Bottleneck, [3, 4, 23, 3], **kwargs) - if pretrained: - state_dict = model_zoo.load_url(model_urls['resnext101_32x8d']) - model.load_state_dict(state_dict) - return model - - -if __name__ == '__main__': - model = resnext101_32x8d() diff --git a/AIlib/segutils/core/models/base_models/vgg.py b/AIlib/segutils/core/models/base_models/vgg.py deleted file mode 100644 index fe5c163..0000000 --- a/AIlib/segutils/core/models/base_models/vgg.py +++ /dev/null @@ -1,191 +0,0 @@ -import torch -import torch.nn as nn -import torch.utils.model_zoo as model_zoo - -__all__ = [ - 'VGG', 'vgg11', 'vgg11_bn', 'vgg13', 'vgg13_bn', 'vgg16', 'vgg16_bn', - 'vgg19_bn', 'vgg19', -] - -model_urls = { - 'vgg11': 'https://download.pytorch.org/models/vgg11-bbd30ac9.pth', - 'vgg13': 'https://download.pytorch.org/models/vgg13-c768596a.pth', - 'vgg16': 'https://download.pytorch.org/models/vgg16-397923af.pth', - 'vgg19': 'https://download.pytorch.org/models/vgg19-dcbb9e9d.pth', - 'vgg11_bn': 'https://download.pytorch.org/models/vgg11_bn-6002323d.pth', - 'vgg13_bn': 'https://download.pytorch.org/models/vgg13_bn-abd245e5.pth', - 'vgg16_bn': 'https://download.pytorch.org/models/vgg16_bn-6c64b313.pth', - 'vgg19_bn': 'https://download.pytorch.org/models/vgg19_bn-c79401a0.pth', -} - - -class VGG(nn.Module): - def __init__(self, features, num_classes=1000, init_weights=True): - super(VGG, self).__init__() - self.features = features - self.avgpool = nn.AdaptiveAvgPool2d((7, 7)) - self.classifier = nn.Sequential( - nn.Linear(512 * 7 * 7, 4096), - nn.ReLU(True), - nn.Dropout(), - nn.Linear(4096, 4096), - nn.ReLU(True), - nn.Dropout(), - nn.Linear(4096, num_classes) - ) - if init_weights: - self._initialize_weights() - - def forward(self, x): - x = self.features(x) - x = self.avgpool(x) - x = x.view(x.size(0), -1) - x = self.classifier(x) - return x - - def _initialize_weights(self): - for m in self.modules(): - if isinstance(m, nn.Conv2d): - nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') - if m.bias is not None: - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.BatchNorm2d): - nn.init.constant_(m.weight, 1) - nn.init.constant_(m.bias, 0) - elif isinstance(m, nn.Linear): - nn.init.normal_(m.weight, 0, 0.01) - nn.init.constant_(m.bias, 0) - - -def make_layers(cfg, batch_norm=False): - layers = [] - in_channels = 3 - for v in cfg: - if v == 'M': - layers += [nn.MaxPool2d(kernel_size=2, stride=2)] - else: - conv2d = nn.Conv2d(in_channels, v, kernel_size=3, padding=1) - if batch_norm: - layers += (conv2d, nn.BatchNorm2d(v), nn.ReLU(inplace=True)) - else: - layers += [conv2d, nn.ReLU(inplace=True)] - in_channels = v - return nn.Sequential(*layers) - - -cfg = { - 'A': [64, 'M', 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'], - 'B': [64, 64, 'M', 128, 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M'], - 'D': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 'M', 512, 512, 512, 'M', 512, 512, 512, 'M'], - 'E': [64, 64, 'M', 128, 128, 'M', 256, 256, 256, 256, 'M', 512, 512, 512, 512, 'M', 512, 512, 512, 512, 'M'], -} - - -def vgg11(pretrained=False, **kwargs): - """VGG 11-layer model (configuration "A") - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - if pretrained: - kwargs['init_weights'] = False - model = VGG(make_layers(cfg['A']), **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['vgg11'])) - return model - - -def vgg11_bn(pretrained=False, **kwargs): - """VGG 11-layer model (configuration "A") with batch normalization - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - if pretrained: - kwargs['init_weights'] = False - model = VGG(make_layers(cfg['A'], batch_norm=True), **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['vgg11_bn'])) - return model - - -def vgg13(pretrained=False, **kwargs): - """VGG 13-layer model (configuration "B") - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - if pretrained: - kwargs['init_weights'] = False - model = VGG(make_layers(cfg['B']), **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['vgg13'])) - return model - - -def vgg13_bn(pretrained=False, **kwargs): - """VGG 13-layer model (configuration "B") with batch normalization - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - if pretrained: - kwargs['init_weights'] = False - model = VGG(make_layers(cfg['B'], batch_norm=True), **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['vgg13_bn'])) - return model - - -def vgg16(pretrained=False, **kwargs): - """VGG 16-layer model (configuration "D") - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - if pretrained: - kwargs['init_weights'] = False - model = VGG(make_layers(cfg['D']), **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['vgg16'])) - return model - - -def vgg16_bn(pretrained=False, **kwargs): - """VGG 16-layer model (configuration "D") with batch normalization - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - if pretrained: - kwargs['init_weights'] = False - model = VGG(make_layers(cfg['D'], batch_norm=True), **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['vgg16_bn'])) - return model - - -def vgg19(pretrained=False, **kwargs): - """VGG 19-layer model (configuration "E") - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - if pretrained: - kwargs['init_weights'] = False - model = VGG(make_layers(cfg['E']), **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['vgg19'])) - return model - - -def vgg19_bn(pretrained=False, **kwargs): - """VGG 19-layer model (configuration 'E') with batch normalization - Args: - pretrained (bool): If True, returns a model pre-trained on ImageNet - """ - if pretrained: - kwargs['init_weights'] = False - model = VGG(make_layers(cfg['E'], batch_norm=True), **kwargs) - if pretrained: - model.load_state_dict(model_zoo.load_url(model_urls['vgg19_bn'])) - return model - - -if __name__ == '__main__': - img = torch.randn((4, 3, 480, 480)) - model = vgg16(pretrained=False) - out = model(img) diff --git a/AIlib/segutils/core/models/base_models/xception.py b/AIlib/segutils/core/models/base_models/xception.py deleted file mode 100644 index 51832f1..0000000 --- a/AIlib/segutils/core/models/base_models/xception.py +++ /dev/null @@ -1,411 +0,0 @@ -import torch -import torch.nn as nn -import torch.nn.functional as F - -__all__ = ['Enc', 'FCAttention', 'Xception65', 'Xception71', 'get_xception', 'get_xception_71', 'get_xception_a'] - - -class SeparableConv2d(nn.Module): - def __init__(self, in_channels, out_channels, kernel_size=3, stride=1, dilation=1, bias=False, norm_layer=None): - super(SeparableConv2d, self).__init__() - self.kernel_size = kernel_size - self.dilation = dilation - - self.conv1 = nn.Conv2d(in_channels, in_channels, kernel_size, stride, 0, dilation, groups=in_channels, - bias=bias) - self.bn = norm_layer(in_channels) - self.pointwise = nn.Conv2d(in_channels, out_channels, 1, bias=bias) - - def forward(self, x): - x = self.fix_padding(x, self.kernel_size, self.dilation) - x = self.conv1(x) - x = self.bn(x) - x = self.pointwise(x) - - return x - - def fix_padding(self, x, kernel_size, dilation): - kernel_size_effective = kernel_size + (kernel_size - 1) * (dilation - 1) - pad_total = kernel_size_effective - 1 - pad_beg = pad_total // 2 - pad_end = pad_total - pad_beg - padded_inputs = F.pad(x, (pad_beg, pad_end, pad_beg, pad_end)) - return padded_inputs - - -class Block(nn.Module): - def __init__(self, in_channels, out_channels, reps, stride=1, dilation=1, norm_layer=None, - start_with_relu=True, grow_first=True, is_last=False): - super(Block, self).__init__() - if out_channels != in_channels or stride != 1: - self.skip = nn.Conv2d(in_channels, out_channels, 1, stride, bias=False) - self.skipbn = norm_layer(out_channels) - else: - self.skip = None - self.relu = nn.ReLU(True) - rep = list() - filters = in_channels - if grow_first: - if start_with_relu: - rep.append(self.relu) - rep.append(SeparableConv2d(in_channels, out_channels, 3, 1, dilation, norm_layer=norm_layer)) - rep.append(norm_layer(out_channels)) - filters = out_channels - for i in range(reps - 1): - if grow_first or start_with_relu: - rep.append(self.relu) - rep.append(SeparableConv2d(filters, filters, 3, 1, dilation, norm_layer=norm_layer)) - rep.append(norm_layer(filters)) - if not grow_first: - rep.append(self.relu) - rep.append(SeparableConv2d(in_channels, out_channels, 3, 1, dilation, norm_layer=norm_layer)) - if stride != 1: - rep.append(self.relu) - rep.append(SeparableConv2d(out_channels, out_channels, 3, stride, norm_layer=norm_layer)) - rep.append(norm_layer(out_channels)) - elif is_last: - rep.append(self.relu) - rep.append(SeparableConv2d(out_channels, out_channels, 3, 1, dilation, norm_layer=norm_layer)) - rep.append(norm_layer(out_channels)) - self.rep = nn.Sequential(*rep) - - def forward(self, x): - out = self.rep(x) - if self.skip is not None: - skip = self.skipbn(self.skip(x)) - else: - skip = x - out = out + skip - return out - - -class Xception65(nn.Module): - """Modified Aligned Xception - """ - - def __init__(self, num_classes=1000, output_stride=32, norm_layer=nn.BatchNorm2d): - super(Xception65, self).__init__() - if output_stride == 32: - entry_block3_stride = 2 - exit_block20_stride = 2 - middle_block_dilation = 1 - exit_block_dilations = (1, 1) - elif output_stride == 16: - entry_block3_stride = 2 - exit_block20_stride = 1 - middle_block_dilation = 1 - exit_block_dilations = (1, 2) - elif output_stride == 8: - entry_block3_stride = 1 - exit_block20_stride = 1 - middle_block_dilation = 2 - exit_block_dilations = (2, 4) - else: - raise NotImplementedError - # Entry flow - self.conv1 = nn.Conv2d(3, 32, 3, 2, 1, bias=False) - self.bn1 = norm_layer(32) - self.relu = nn.ReLU(True) - - self.conv2 = nn.Conv2d(32, 64, 3, 1, 1, bias=False) - self.bn2 = norm_layer(64) - - self.block1 = Block(64, 128, reps=2, stride=2, norm_layer=norm_layer, start_with_relu=False) - self.block2 = Block(128, 256, reps=2, stride=2, norm_layer=norm_layer, start_with_relu=False, grow_first=True) - self.block3 = Block(256, 728, reps=2, stride=entry_block3_stride, norm_layer=norm_layer, - start_with_relu=True, grow_first=True, is_last=True) - - # Middle flow - midflow = list() - for i in range(4, 20): - midflow.append(Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation, norm_layer=norm_layer, - start_with_relu=True, grow_first=True)) - self.midflow = nn.Sequential(*midflow) - - # Exit flow - self.block20 = Block(728, 1024, reps=2, stride=exit_block20_stride, dilation=exit_block_dilations[0], - norm_layer=norm_layer, start_with_relu=True, grow_first=False, is_last=True) - self.conv3 = SeparableConv2d(1024, 1536, 3, 1, dilation=exit_block_dilations[1], norm_layer=norm_layer) - self.bn3 = norm_layer(1536) - self.conv4 = SeparableConv2d(1536, 1536, 3, stride=1, dilation=exit_block_dilations[1], norm_layer=norm_layer) - self.bn4 = norm_layer(1536) - self.conv5 = SeparableConv2d(1536, 2048, 3, 1, dilation=exit_block_dilations[1], norm_layer=norm_layer) - self.bn5 = norm_layer(2048) - self.avgpool = nn.AdaptiveAvgPool2d(1) - self.fc = nn.Linear(2048, num_classes) - - def forward(self, x): - # Entry flow - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - - x = self.conv2(x) - x = self.bn2(x) - x = self.relu(x) - - x = self.block1(x) - x = self.relu(x) - # c1 = x - x = self.block2(x) - # c2 = x - x = self.block3(x) - - # Middle flow - x = self.midflow(x) - # c3 = x - - # Exit flow - x = self.block20(x) - x = self.relu(x) - x = self.conv3(x) - x = self.bn3(x) - x = self.relu(x) - - x = self.conv4(x) - x = self.bn4(x) - x = self.relu(x) - - x = self.conv5(x) - x = self.bn5(x) - x = self.relu(x) - - x = self.avgpool(x) - x = x.view(x.size(0), -1) - x = self.fc(x) - - return x - - -class Xception71(nn.Module): - """Modified Aligned Xception - """ - - def __init__(self, num_classes=1000, output_stride=32, norm_layer=nn.BatchNorm2d): - super(Xception71, self).__init__() - if output_stride == 32: - entry_block3_stride = 2 - exit_block20_stride = 2 - middle_block_dilation = 1 - exit_block_dilations = (1, 1) - elif output_stride == 16: - entry_block3_stride = 2 - exit_block20_stride = 1 - middle_block_dilation = 1 - exit_block_dilations = (1, 2) - elif output_stride == 8: - entry_block3_stride = 1 - exit_block20_stride = 1 - middle_block_dilation = 2 - exit_block_dilations = (2, 4) - else: - raise NotImplementedError - # Entry flow - self.conv1 = nn.Conv2d(3, 32, 3, 2, 1, bias=False) - self.bn1 = norm_layer(32) - self.relu = nn.ReLU(True) - - self.conv2 = nn.Conv2d(32, 64, 3, 1, 1, bias=False) - self.bn2 = norm_layer(64) - - self.block1 = Block(64, 128, reps=2, stride=2, norm_layer=norm_layer, start_with_relu=False) - self.block2 = nn.Sequential( - Block(128, 256, reps=2, stride=2, norm_layer=norm_layer, start_with_relu=False, grow_first=True), - Block(256, 728, reps=2, stride=2, norm_layer=norm_layer, start_with_relu=False, grow_first=True)) - self.block3 = Block(728, 728, reps=2, stride=entry_block3_stride, norm_layer=norm_layer, - start_with_relu=True, grow_first=True, is_last=True) - - # Middle flow - midflow = list() - for i in range(4, 20): - midflow.append(Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation, norm_layer=norm_layer, - start_with_relu=True, grow_first=True)) - self.midflow = nn.Sequential(*midflow) - - # Exit flow - self.block20 = Block(728, 1024, reps=2, stride=exit_block20_stride, dilation=exit_block_dilations[0], - norm_layer=norm_layer, start_with_relu=True, grow_first=False, is_last=True) - self.conv3 = SeparableConv2d(1024, 1536, 3, 1, dilation=exit_block_dilations[1], norm_layer=norm_layer) - self.bn3 = norm_layer(1536) - self.conv4 = SeparableConv2d(1536, 1536, 3, stride=1, dilation=exit_block_dilations[1], norm_layer=norm_layer) - self.bn4 = norm_layer(1536) - self.conv5 = SeparableConv2d(1536, 2048, 3, 1, dilation=exit_block_dilations[1], norm_layer=norm_layer) - self.bn5 = norm_layer(2048) - self.avgpool = nn.AdaptiveAvgPool2d(1) - self.fc = nn.Linear(2048, num_classes) - - def forward(self, x): - # Entry flow - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - - x = self.conv2(x) - x = self.bn2(x) - x = self.relu(x) - - x = self.block1(x) - x = self.relu(x) - # c1 = x - x = self.block2(x) - # c2 = x - x = self.block3(x) - - # Middle flow - x = self.midflow(x) - # c3 = x - - # Exit flow - x = self.block20(x) - x = self.relu(x) - x = self.conv3(x) - x = self.bn3(x) - x = self.relu(x) - - x = self.conv4(x) - x = self.bn4(x) - x = self.relu(x) - - x = self.conv5(x) - x = self.bn5(x) - x = self.relu(x) - - x = self.avgpool(x) - x = x.view(x.size(0), -1) - x = self.fc(x) - - return x - - -# ------------------------------------------------- -# For DFANet -# ------------------------------------------------- -class BlockA(nn.Module): - def __init__(self, in_channels, out_channels, stride=1, dilation=1, norm_layer=None, start_with_relu=True): - super(BlockA, self).__init__() - if out_channels != in_channels or stride != 1: - self.skip = nn.Conv2d(in_channels, out_channels, 1, stride, bias=False) - self.skipbn = norm_layer(out_channels) - else: - self.skip = None - self.relu = nn.ReLU(False) - rep = list() - inter_channels = out_channels // 4 - - if start_with_relu: - rep.append(self.relu) - rep.append(SeparableConv2d(in_channels, inter_channels, 3, 1, dilation, norm_layer=norm_layer)) - rep.append(norm_layer(inter_channels)) - - rep.append(self.relu) - rep.append(SeparableConv2d(inter_channels, inter_channels, 3, 1, dilation, norm_layer=norm_layer)) - rep.append(norm_layer(inter_channels)) - - if stride != 1: - rep.append(self.relu) - rep.append(SeparableConv2d(inter_channels, out_channels, 3, stride, norm_layer=norm_layer)) - rep.append(norm_layer(out_channels)) - else: - rep.append(self.relu) - rep.append(SeparableConv2d(inter_channels, out_channels, 3, 1, norm_layer=norm_layer)) - rep.append(norm_layer(out_channels)) - self.rep = nn.Sequential(*rep) - - def forward(self, x): - out = self.rep(x) - if self.skip is not None: - skip = self.skipbn(self.skip(x)) - else: - skip = x - out = out + skip - return out - - -class Enc(nn.Module): - def __init__(self, in_channels, out_channels, blocks, norm_layer=nn.BatchNorm2d): - super(Enc, self).__init__() - block = list() - block.append(BlockA(in_channels, out_channels, 2, norm_layer=norm_layer)) - for i in range(blocks - 1): - block.append(BlockA(out_channels, out_channels, 1, norm_layer=norm_layer)) - self.block = nn.Sequential(*block) - - def forward(self, x): - return self.block(x) - - -class FCAttention(nn.Module): - def __init__(self, in_channels, norm_layer=nn.BatchNorm2d): - super(FCAttention, self).__init__() - self.avgpool = nn.AdaptiveAvgPool2d(1) - self.fc = nn.Linear(in_channels, 1000) - self.conv = nn.Sequential( - nn.Conv2d(1000, in_channels, 1, bias=False), - norm_layer(in_channels), - nn.ReLU(False)) - - def forward(self, x): - n, c, _, _ = x.size() - att = self.avgpool(x).view(n, c) - att = self.fc(att).view(n, 1000, 1, 1) - att = self.conv(att) - return x * att.expand_as(x) - - -class XceptionA(nn.Module): - def __init__(self, num_classes=1000, norm_layer=nn.BatchNorm2d): - super(XceptionA, self).__init__() - self.conv1 = nn.Sequential(nn.Conv2d(3, 8, 3, 2, 1, bias=False), - norm_layer(8), - nn.ReLU(True)) - - self.enc2 = Enc(8, 48, 4, norm_layer=norm_layer) - self.enc3 = Enc(48, 96, 6, norm_layer=norm_layer) - self.enc4 = Enc(96, 192, 4, norm_layer=norm_layer) - - self.fca = FCAttention(192, norm_layer=norm_layer) - self.avgpool = nn.AdaptiveAvgPool2d(1) - self.fc = nn.Linear(192, num_classes) - - def forward(self, x): - x = self.conv1(x) - - x = self.enc2(x) - x = self.enc3(x) - x = self.enc4(x) - x = self.fca(x) - - x = self.avgpool(x) - x = x.view(x.size(0), -1) - x = self.fc(x) - - return x - - -# Constructor -def get_xception(pretrained=False, root='~/.torch/models', **kwargs): - model = Xception65(**kwargs) - if pretrained: - from ..model_store import get_model_file - model.load_state_dict(torch.load(get_model_file('xception', root=root))) - return model - - -def get_xception_71(pretrained=False, root='~/.torch/models', **kwargs): - model = Xception71(**kwargs) - if pretrained: - from ..model_store import get_model_file - model.load_state_dict(torch.load(get_model_file('xception71', root=root))) - return model - - -def get_xception_a(pretrained=False, root='~/.torch/models', **kwargs): - model = XceptionA(**kwargs) - if pretrained: - from ..model_store import get_model_file - model.load_state_dict(torch.load(get_model_file('xception_a', root=root))) - return model - - -if __name__ == '__main__': - model = get_xception_a() diff --git a/AIlib/segutils/core/models/bisenet.py b/AIlib/segutils/core/models/bisenet.py deleted file mode 100644 index 09d335d..0000000 --- a/AIlib/segutils/core/models/bisenet.py +++ /dev/null @@ -1,298 +0,0 @@ -"""Bilateral Segmentation Network""" -import torch -import torch.nn as nn -import torch.nn.functional as F -import numpy as np -from core.models.base_models.resnet import resnet18,resnet50 -from core.nn import _ConvBNReLU - -__all__ = ['BiSeNet', 'get_bisenet', 'get_bisenet_resnet18_citys'] - - -class BiSeNet(nn.Module): - def __init__(self, nclass, backbone='resnet18', aux=False, jpu=False, pretrained_base=True, **kwargs): - super(BiSeNet, self).__init__() - self.aux = aux - self.spatial_path = SpatialPath(3, 128, **kwargs) - self.context_path = ContextPath(backbone, pretrained_base, **kwargs) - self.ffm = FeatureFusion(256, 256, 4, **kwargs) - self.head = _BiSeHead(256, 64, nclass, **kwargs) - if aux: - self.auxlayer1 = _BiSeHead(128, 256, nclass, **kwargs) - self.auxlayer2 = _BiSeHead(128, 256, nclass, **kwargs) - - self.__setattr__('exclusive', - ['spatial_path', 'context_path', 'ffm', 'head', 'auxlayer1', 'auxlayer2'] if aux else [ - 'spatial_path', 'context_path', 'ffm', 'head']) - - def forward(self, x,outsize=None,test_flag=False): - size = x.size()[2:] - spatial_out = self.spatial_path(x) - context_out = self.context_path(x) - fusion_out = self.ffm(spatial_out, context_out[-1]) - outputs = [] - x = self.head(fusion_out) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - - - if outsize: - print('######using torch resize#######',outsize) - x = F.interpolate(x, outsize, mode='bilinear', align_corners=True) - outputs.append(x) - - if self.aux: - auxout1 = self.auxlayer1(context_out[0]) - auxout1 = F.interpolate(auxout1, size, mode='bilinear', align_corners=True) - outputs.append(auxout1) - auxout2 = self.auxlayer2(context_out[1]) - auxout2 = F.interpolate(auxout2, size, mode='bilinear', align_corners=True) - outputs.append(auxout2) - if test_flag: - outputs = [torch.argmax(outputx ,axis=1) for outputx in outputs] - #return tuple(outputs) - return outputs[0] -class BiSeNet_MultiOutput(nn.Module): - def __init__(self, nclass, backbone='resnet18', aux=False, jpu=False, pretrained_base=True, **kwargs): - super(BiSeNet_MultiOutput, self).__init__() - self.aux = aux - self.spatial_path = SpatialPath(3, 128, **kwargs) - self.context_path = ContextPath(backbone, pretrained_base, **kwargs) - self.ffm = FeatureFusion(256, 256, 4, **kwargs) - assert isinstance(nclass,list) - self.outCnt = len(nclass) - for ii,nclassii in enumerate(nclass): - setattr(self,'head%d'%(ii) , _BiSeHead(256, 64, nclassii, **kwargs)) - - if aux: - self.auxlayer1 = _BiSeHead(128, 256, nclass, **kwargs) - self.auxlayer2 = _BiSeHead(128, 256, nclass, **kwargs) - - self.__setattr__('exclusive', - ['spatial_path', 'context_path', 'ffm', 'head', 'auxlayer1', 'auxlayer2'] if aux else [ - 'spatial_path', 'context_path', 'ffm', 'head']) - - def forward(self, x,outsize=None,test_flag=False,smooth_kernel=0): - size = x.size()[2:] - spatial_out = self.spatial_path(x) - context_out = self.context_path(x) - fusion_out = self.ffm(spatial_out, context_out[-1]) - outputs = [] - for ii in range(self.outCnt): - x = getattr(self,'head%d'%(ii))(fusion_out) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - - if self.aux: - auxout1 = self.auxlayer1(context_out[0]) - auxout1 = F.interpolate(auxout1, size, mode='bilinear', align_corners=True) - outputs.append(auxout1) - auxout2 = self.auxlayer2(context_out[1]) - auxout2 = F.interpolate(auxout2, size, mode='bilinear', align_corners=True) - outputs.append(auxout2) - if test_flag: - outputs = [torch.argmax(outputx ,axis=1) for outputx in outputs] - if smooth_kernel>0: - gaussian_kernel = torch.from_numpy(np.ones((1,1,smooth_kernel,smooth_kernel)) ) - - pad = int((smooth_kernel - 1)/2) - if not gaussian_kernel.is_cuda: - gaussian_kernel = gaussian_kernel.to(x.device) - #print(gaussian_kernel.dtype,gaussian_kernel,outputs[0].dtype) - outputs = [ x.unsqueeze(1).double() for x in outputs] - outputs = [torch.conv2d(x, gaussian_kernel, padding=pad) for x in outputs ] - outputs = [ x.squeeze(1).long() for x in outputs] - #return tuple(outputs) - return outputs - -class _BiSeHead(nn.Module): - def __init__(self, in_channels, inter_channels, nclass, norm_layer=nn.BatchNorm2d, **kwargs): - super(_BiSeHead, self).__init__() - self.block = nn.Sequential( - _ConvBNReLU(in_channels, inter_channels, 3, 1, 1, norm_layer=norm_layer), - nn.Dropout(0.1), - nn.Conv2d(inter_channels, nclass, 1) - ) - - def forward(self, x): - x = self.block(x) - return x - - -class SpatialPath(nn.Module): - """Spatial path""" - - def __init__(self, in_channels, out_channels, norm_layer=nn.BatchNorm2d, **kwargs): - super(SpatialPath, self).__init__() - inter_channels = 64 - self.conv7x7 = _ConvBNReLU(in_channels, inter_channels, 7, 2, 3, norm_layer=norm_layer) - self.conv3x3_1 = _ConvBNReLU(inter_channels, inter_channels, 3, 2, 1, norm_layer=norm_layer) - self.conv3x3_2 = _ConvBNReLU(inter_channels, inter_channels, 3, 2, 1, norm_layer=norm_layer) - self.conv1x1 = _ConvBNReLU(inter_channels, out_channels, 1, 1, 0, norm_layer=norm_layer) - - def forward(self, x): - x = self.conv7x7(x) - x = self.conv3x3_1(x) - x = self.conv3x3_2(x) - x = self.conv1x1(x) - - return x - - -class _GlobalAvgPooling(nn.Module): - def __init__(self, in_channels, out_channels, norm_layer, **kwargs): - super(_GlobalAvgPooling, self).__init__() - self.gap = nn.Sequential( - nn.AdaptiveAvgPool2d(1), - nn.Conv2d(in_channels, out_channels, 1, bias=False), - norm_layer(out_channels), - nn.ReLU(True) - ) - - def forward(self, x): - size = x.size()[2:] - pool = self.gap(x) - out = F.interpolate(pool, size, mode='bilinear', align_corners=True) - return out - - -class AttentionRefinmentModule(nn.Module): - def __init__(self, in_channels, out_channels, norm_layer=nn.BatchNorm2d, **kwargs): - super(AttentionRefinmentModule, self).__init__() - self.conv3x3 = _ConvBNReLU(in_channels, out_channels, 3, 1, 1, norm_layer=norm_layer) - self.channel_attention = nn.Sequential( - nn.AdaptiveAvgPool2d(1), - _ConvBNReLU(out_channels, out_channels, 1, 1, 0, norm_layer=norm_layer), - nn.Sigmoid() - ) - - def forward(self, x): - x = self.conv3x3(x) - attention = self.channel_attention(x) - x = x * attention - return x - - -class ContextPath(nn.Module): - def __init__(self, backbone='resnet18', pretrained_base=True, norm_layer=nn.BatchNorm2d, **kwargs): - super(ContextPath, self).__init__() - if backbone == 'resnet18': - pretrained = resnet18(pretrained=pretrained_base, **kwargs) - elif backbone=='resnet50': - pretrained = resnet50(pretrained=pretrained_base, **kwargs) - else: - raise RuntimeError('unknown backbone: {}'.format(backbone)) - self.conv1 = pretrained.conv1 - self.bn1 = pretrained.bn1 - self.relu = pretrained.relu - self.maxpool = pretrained.maxpool - self.layer1 = pretrained.layer1 - self.layer2 = pretrained.layer2 - self.layer3 = pretrained.layer3 - self.layer4 = pretrained.layer4 - - inter_channels = 128 - self.global_context = _GlobalAvgPooling(512, inter_channels, norm_layer) - - self.arms = nn.ModuleList( - [AttentionRefinmentModule(512, inter_channels, norm_layer, **kwargs), - AttentionRefinmentModule(256, inter_channels, norm_layer, **kwargs)] - ) - self.refines = nn.ModuleList( - [_ConvBNReLU(inter_channels, inter_channels, 3, 1, 1, norm_layer=norm_layer), - _ConvBNReLU(inter_channels, inter_channels, 3, 1, 1, norm_layer=norm_layer)] - ) - - def forward(self, x): - x = self.conv1(x) - x = self.bn1(x) - x = self.relu(x) - x = self.maxpool(x) - x = self.layer1(x) - - context_blocks = [] - context_blocks.append(x) - x = self.layer2(x) - context_blocks.append(x) - c3 = self.layer3(x) - context_blocks.append(c3) - c4 = self.layer4(c3) - context_blocks.append(c4) - context_blocks.reverse() - - global_context = self.global_context(c4) - last_feature = global_context - context_outputs = [] - for i, (feature, arm, refine) in enumerate(zip(context_blocks[:2], self.arms, self.refines)): - feature = arm(feature) - feature += last_feature - last_feature = F.interpolate(feature, size=context_blocks[i + 1].size()[2:], - mode='bilinear', align_corners=True) - last_feature = refine(last_feature) - context_outputs.append(last_feature) - - return context_outputs - - -class FeatureFusion(nn.Module): - def __init__(self, in_channels, out_channels, reduction=1, norm_layer=nn.BatchNorm2d, **kwargs): - super(FeatureFusion, self).__init__() - self.conv1x1 = _ConvBNReLU(in_channels, out_channels, 1, 1, 0, norm_layer=norm_layer, **kwargs) - self.channel_attention = nn.Sequential( - nn.AdaptiveAvgPool2d(1), - _ConvBNReLU(out_channels, out_channels // reduction, 1, 1, 0, norm_layer=norm_layer), - _ConvBNReLU(out_channels // reduction, out_channels, 1, 1, 0, norm_layer=norm_layer), - nn.Sigmoid() - ) - - def forward(self, x1, x2): - fusion = torch.cat([x1, x2], dim=1) - out = self.conv1x1(fusion) - attention = self.channel_attention(out) - out = out + out * attention - return out - - -def get_bisenet(dataset='citys', backbone='resnet18', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = BiSeNet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('bisenet_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_bisenet_resnet18_citys(**kwargs): - return get_bisenet('citys', 'resnet18', **kwargs) - - -if __name__ == '__main__': - # img = torch.randn(2, 3, 224, 224) - # model = BiSeNet(19, backbone='resnet18') - # print(model.exclusive) - input = torch.rand(2, 3, 224, 224) - model = BiSeNet(4, pretrained_base=True) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) diff --git a/AIlib/segutils/core/models/ccnet.py b/AIlib/segutils/core/models/ccnet.py deleted file mode 100644 index b06ca03..0000000 --- a/AIlib/segutils/core/models/ccnet.py +++ /dev/null @@ -1,166 +0,0 @@ -"""Criss-Cross Network""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.nn import CrissCrossAttention -from core.models.segbase import SegBaseModel -from core.models.fcn import _FCNHead - -#失败：NameError: name '_C' is not defined - -__all__ = ['CCNet', 'get_ccnet', 'get_ccnet_resnet50_citys', 'get_ccnet_resnet101_citys', - 'get_ccnet_resnet152_citys', 'get_ccnet_resnet50_ade', 'get_ccnet_resnet101_ade', - 'get_ccnet_resnet152_ade'] - - -class CCNet(SegBaseModel): - r"""CCNet - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - - Reference: - Zilong Huang, et al. "CCNet: Criss-Cross Attention for Semantic Segmentation." - arXiv preprint arXiv:1811.11721 (2018). - """ - - def __init__(self, nclass, backbone='resnet50', aux=False, pretrained_base=True, **kwargs): - super(CCNet, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.head = _CCHead(nclass, **kwargs) - if aux: - self.auxlayer = _FCNHead(1024, nclass, **kwargs) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - _, _, c3, c4 = self.base_forward(x) - outputs = list() - x = self.head(c4) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - - if self.aux: - auxout = self.auxlayer(c3) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - return tuple(outputs) - - -class _CCHead(nn.Module): - def __init__(self, nclass, norm_layer=nn.BatchNorm2d, **kwargs): - super(_CCHead, self).__init__() - self.rcca = _RCCAModule(2048, 512, norm_layer, **kwargs) - self.out = nn.Conv2d(512, nclass, 1) - - def forward(self, x): - x = self.rcca(x) - x = self.out(x) - return x - - -class _RCCAModule(nn.Module): - def __init__(self, in_channels, out_channels, norm_layer, **kwargs): - super(_RCCAModule, self).__init__() - inter_channels = in_channels // 4 - self.conva = nn.Sequential( - nn.Conv2d(in_channels, inter_channels, 3, padding=1, bias=False), - norm_layer(inter_channels), - nn.ReLU(True)) - self.cca = CrissCrossAttention(inter_channels) - self.convb = nn.Sequential( - nn.Conv2d(inter_channels, inter_channels, 3, padding=1, bias=False), - norm_layer(inter_channels), - nn.ReLU(True)) - - self.bottleneck = nn.Sequential( - nn.Conv2d(in_channels + inter_channels, out_channels, 3, padding=1, bias=False), - norm_layer(out_channels), - nn.Dropout2d(0.1)) - - def forward(self, x, recurrence=1): - out = self.conva(x) - for i in range(recurrence): - out = self.cca(out) - out = self.convb(out) - out = torch.cat([x, out], dim=1) - out = self.bottleneck(out) - - return out - - -def get_ccnet(dataset='pascal_voc', backbone='resnet50', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = CCNet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('ccnet_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_ccnet_resnet50_citys(**kwargs): - return get_ccnet('citys', 'resnet50', **kwargs) - - -def get_ccnet_resnet101_citys(**kwargs): - return get_ccnet('citys', 'resnet101', **kwargs) - - -def get_ccnet_resnet152_citys(**kwargs): - return get_ccnet('citys', 'resnet152', **kwargs) - - -def get_ccnet_resnet50_ade(**kwargs): - return get_ccnet('ade20k', 'resnet50', **kwargs) - - -def get_ccnet_resnet101_ade(**kwargs): - return get_ccnet('ade20k', 'resnet101', **kwargs) - - -def get_ccnet_resnet152_ade(**kwargs): - return get_ccnet('ade20k', 'resnet152', **kwargs) - - -if __name__ == '__main__': - # model = get_ccnet_resnet50_citys() - # img = torch.randn(1, 3, 480, 480) - # outputs = model(img) - input = torch.rand(2, 3, 224, 224) - model = CCNet(4, pretrained_base=False) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) \ No newline at end of file diff --git a/AIlib/segutils/core/models/cgnet.py b/AIlib/segutils/core/models/cgnet.py deleted file mode 100644 index 85cb4e6..0000000 --- a/AIlib/segutils/core/models/cgnet.py +++ /dev/null @@ -1,228 +0,0 @@ -"""Context Guided Network for Semantic Segmentation""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.nn import _ConvBNPReLU, _BNPReLU - -__all__ = ['CGNet', 'get_cgnet', 'get_cgnet_citys'] - - -class CGNet(nn.Module): - r"""CGNet - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - - Reference: - Tianyi Wu, et al. "CGNet: A Light-weight Context Guided Network for Semantic Segmentation." - arXiv preprint arXiv:1811.08201 (2018). - """ - - def __init__(self, nclass, backbone='', aux=False, jpu=False, pretrained_base=True, M=3, N=21, **kwargs): - super(CGNet, self).__init__() - # stage 1 - self.stage1_0 = _ConvBNPReLU(3, 32, 3, 2, 1, **kwargs) - self.stage1_1 = _ConvBNPReLU(32, 32, 3, 1, 1, **kwargs) - self.stage1_2 = _ConvBNPReLU(32, 32, 3, 1, 1, **kwargs) - - self.sample1 = _InputInjection(1) - self.sample2 = _InputInjection(2) - self.bn_prelu1 = _BNPReLU(32 + 3, **kwargs) - - # stage 2 - self.stage2_0 = ContextGuidedBlock(32 + 3, 64, dilation=2, reduction=8, down=True, residual=False, **kwargs) - self.stage2 = nn.ModuleList() - for i in range(0, M - 1): - self.stage2.append(ContextGuidedBlock(64, 64, dilation=2, reduction=8, **kwargs)) - self.bn_prelu2 = _BNPReLU(128 + 3, **kwargs) - - # stage 3 - self.stage3_0 = ContextGuidedBlock(128 + 3, 128, dilation=4, reduction=16, down=True, residual=False, **kwargs) - self.stage3 = nn.ModuleList() - for i in range(0, N - 1): - self.stage3.append(ContextGuidedBlock(128, 128, dilation=4, reduction=16, **kwargs)) - self.bn_prelu3 = _BNPReLU(256, **kwargs) - - self.head = nn.Sequential( - nn.Dropout2d(0.1, False), - nn.Conv2d(256, nclass, 1)) - - self.__setattr__('exclusive', ['stage1_0', 'stage1_1', 'stage1_2', 'sample1', 'sample2', - 'bn_prelu1', 'stage2_0', 'stage2', 'bn_prelu2', 'stage3_0', - 'stage3', 'bn_prelu3', 'head']) - - def forward(self, x): - size = x.size()[2:] - # stage1 - out0 = self.stage1_0(x) - out0 = self.stage1_1(out0) - out0 = self.stage1_2(out0) - - inp1 = self.sample1(x) - inp2 = self.sample2(x) - - # stage 2 - out0_cat = self.bn_prelu1(torch.cat([out0, inp1], dim=1)) - out1_0 = self.stage2_0(out0_cat) - for i, layer in enumerate(self.stage2): - if i == 0: - out1 = layer(out1_0) - else: - out1 = layer(out1) - out1_cat = self.bn_prelu2(torch.cat([out1, out1_0, inp2], dim=1)) - - # stage 3 - out2_0 = self.stage3_0(out1_cat) - for i, layer in enumerate(self.stage3): - if i == 0: - out2 = layer(out2_0) - else: - out2 = layer(out2) - out2_cat = self.bn_prelu3(torch.cat([out2_0, out2], dim=1)) - - outputs = [] - out = self.head(out2_cat) - out = F.interpolate(out, size, mode='bilinear', align_corners=True) - outputs.append(out) - #return tuple(outputs) - return outputs[0] - - -class _ChannelWiseConv(nn.Module): - def __init__(self, in_channels, out_channels, dilation=1, **kwargs): - super(_ChannelWiseConv, self).__init__() - self.conv = nn.Conv2d(in_channels, out_channels, 3, 1, dilation, dilation, groups=in_channels, bias=False) - - def forward(self, x): - x = self.conv(x) - return x - - -class _FGlo(nn.Module): - def __init__(self, in_channels, reduction=16, **kwargs): - super(_FGlo, self).__init__() - self.gap = nn.AdaptiveAvgPool2d(1) - self.fc = nn.Sequential( - nn.Linear(in_channels, in_channels // reduction), - nn.ReLU(True), - nn.Linear(in_channels // reduction, in_channels), - nn.Sigmoid()) - - def forward(self, x): - n, c, _, _ = x.size() - out = self.gap(x).view(n, c) - out = self.fc(out).view(n, c, 1, 1) - return x * out - - -class _InputInjection(nn.Module): - def __init__(self, ratio): - super(_InputInjection, self).__init__() - self.pool = nn.ModuleList() - for i in range(0, ratio): - self.pool.append(nn.AvgPool2d(3, 2, 1)) - - def forward(self, x): - for pool in self.pool: - x = pool(x) - return x - - -class _ConcatInjection(nn.Module): - def __init__(self, in_channels, norm_layer=nn.BatchNorm2d, **kwargs): - super(_ConcatInjection, self).__init__() - self.bn = norm_layer(in_channels) - self.prelu = nn.PReLU(in_channels) - - def forward(self, x1, x2): - out = torch.cat([x1, x2], dim=1) - out = self.bn(out) - out = self.prelu(out) - return out - - -class ContextGuidedBlock(nn.Module): - def __init__(self, in_channels, out_channels, dilation=2, reduction=16, down=False, - residual=True, norm_layer=nn.BatchNorm2d, **kwargs): - super(ContextGuidedBlock, self).__init__() - inter_channels = out_channels // 2 if not down else out_channels - if down: - self.conv = _ConvBNPReLU(in_channels, inter_channels, 3, 2, 1, norm_layer=norm_layer, **kwargs) - self.reduce = nn.Conv2d(inter_channels * 2, out_channels, 1, bias=False) - else: - self.conv = _ConvBNPReLU(in_channels, inter_channels, 1, 1, 0, norm_layer=norm_layer, **kwargs) - self.f_loc = _ChannelWiseConv(inter_channels, inter_channels, **kwargs) - self.f_sur = _ChannelWiseConv(inter_channels, inter_channels, dilation, **kwargs) - self.bn = norm_layer(inter_channels * 2) - self.prelu = nn.PReLU(inter_channels * 2) - self.f_glo = _FGlo(out_channels, reduction, **kwargs) - self.down = down - self.residual = residual - - def forward(self, x): - out = self.conv(x) - loc = self.f_loc(out) - sur = self.f_sur(out) - - joi_feat = torch.cat([loc, sur], dim=1) - joi_feat = self.prelu(self.bn(joi_feat)) - if self.down: - joi_feat = self.reduce(joi_feat) - - out = self.f_glo(joi_feat) - if self.residual: - out = out + x - - return out - - -def get_cgnet(dataset='citys', backbone='', pretrained=False, root='~/.torch/models', pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from core.data.dataloader import datasets - model = CGNet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('cgnet_%s' % (acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_cgnet_citys(**kwargs): - return get_cgnet('citys', '', **kwargs) - - -if __name__ == '__main__': - # model = get_cgnet_citys() - # print(model) - input = torch.rand(2, 3, 224, 224) - model = CGNet(4, pretrained_base=True) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) diff --git a/AIlib/segutils/core/models/danet.py b/AIlib/segutils/core/models/danet.py deleted file mode 100644 index 7dae5d3..0000000 --- a/AIlib/segutils/core/models/danet.py +++ /dev/null @@ -1,232 +0,0 @@ -"""Dual Attention Network""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.models.segbase import SegBaseModel - -__all__ = ['DANet', 'get_danet', 'get_danet_resnet50_citys', - 'get_danet_resnet101_citys', 'get_danet_resnet152_citys'] - - -class DANet(SegBaseModel): - r"""Pyramid Scene Parsing Network - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`mxnet.gluon.nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - Reference: - Jun Fu, Jing Liu, Haijie Tian, Yong Li, Yongjun Bao, Zhiwei Fang,and Hanqing Lu. - "Dual Attention Network for Scene Segmentation." *CVPR*, 2019 - """ - - def __init__(self, nclass, backbone='resnet50', aux=True, pretrained_base=True, **kwargs): - super(DANet, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.head = _DAHead(2048, nclass, aux, **kwargs) - - self.__setattr__('exclusive', ['head']) - - def forward(self, x): - size = x.size()[2:] - _, _, c3, c4 = self.base_forward(x) - outputs = [] - x = self.head(c4) - x0 = F.interpolate(x[0], size, mode='bilinear', align_corners=True) - outputs.append(x0) - - if self.aux: - x1 = F.interpolate(x[1], size, mode='bilinear', align_corners=True) - x2 = F.interpolate(x[2], size, mode='bilinear', align_corners=True) - outputs.append(x1) - outputs.append(x2) - #return outputs - return outputs[0] - -class _PositionAttentionModule(nn.Module): - """ Position attention module""" - - def __init__(self, in_channels, **kwargs): - super(_PositionAttentionModule, self).__init__() - self.conv_b = nn.Conv2d(in_channels, in_channels // 8, 1) - self.conv_c = nn.Conv2d(in_channels, in_channels // 8, 1) - self.conv_d = nn.Conv2d(in_channels, in_channels, 1) - self.alpha = nn.Parameter(torch.zeros(1)) - self.softmax = nn.Softmax(dim=-1) - - def forward(self, x): - batch_size, _, height, width = x.size() - feat_b = self.conv_b(x).view(batch_size, -1, height * width).permute(0, 2, 1) - feat_c = self.conv_c(x).view(batch_size, -1, height * width) - attention_s = self.softmax(torch.bmm(feat_b, feat_c)) - feat_d = self.conv_d(x).view(batch_size, -1, height * width) - feat_e = torch.bmm(feat_d, attention_s.permute(0, 2, 1)).view(batch_size, -1, height, width) - out = self.alpha * feat_e + x - - return out - - -class _ChannelAttentionModule(nn.Module): - """Channel attention module""" - - def __init__(self, **kwargs): - super(_ChannelAttentionModule, self).__init__() - self.beta = nn.Parameter(torch.zeros(1)) - self.softmax = nn.Softmax(dim=-1) - - def forward(self, x): - batch_size, _, height, width = x.size() - feat_a = x.view(batch_size, -1, height * width) - feat_a_transpose = x.view(batch_size, -1, height * width).permute(0, 2, 1) - attention = torch.bmm(feat_a, feat_a_transpose) - attention_new = torch.max(attention, dim=-1, keepdim=True)[0].expand_as(attention) - attention - attention = self.softmax(attention_new) - - feat_e = torch.bmm(attention, feat_a).view(batch_size, -1, height, width) - out = self.beta * feat_e + x - - return out - - -class _DAHead(nn.Module): - def __init__(self, in_channels, nclass, aux=True, norm_layer=nn.BatchNorm2d, norm_kwargs=None, **kwargs): - super(_DAHead, self).__init__() - self.aux = aux - inter_channels = in_channels // 4 - self.conv_p1 = nn.Sequential( - nn.Conv2d(in_channels, inter_channels, 3, padding=1, bias=False), - norm_layer(inter_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - self.conv_c1 = nn.Sequential( - nn.Conv2d(in_channels, inter_channels, 3, padding=1, bias=False), - norm_layer(inter_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - self.pam = _PositionAttentionModule(inter_channels, **kwargs) - self.cam = _ChannelAttentionModule(**kwargs) - self.conv_p2 = nn.Sequential( - nn.Conv2d(inter_channels, inter_channels, 3, padding=1, bias=False), - norm_layer(inter_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - self.conv_c2 = nn.Sequential( - nn.Conv2d(inter_channels, inter_channels, 3, padding=1, bias=False), - norm_layer(inter_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - self.out = nn.Sequential( - nn.Dropout(0.1), - nn.Conv2d(inter_channels, nclass, 1) - ) - if aux: - self.conv_p3 = nn.Sequential( - nn.Dropout(0.1), - nn.Conv2d(inter_channels, nclass, 1) - ) - self.conv_c3 = nn.Sequential( - nn.Dropout(0.1), - nn.Conv2d(inter_channels, nclass, 1) - ) - - def forward(self, x): - feat_p = self.conv_p1(x) - feat_p = self.pam(feat_p) - feat_p = self.conv_p2(feat_p) - - feat_c = self.conv_c1(x) - feat_c = self.cam(feat_c) - feat_c = self.conv_c2(feat_c) - - feat_fusion = feat_p + feat_c - - outputs = [] - fusion_out = self.out(feat_fusion) - outputs.append(fusion_out) - if self.aux: - p_out = self.conv_p3(feat_p) - c_out = self.conv_c3(feat_c) - outputs.append(p_out) - outputs.append(c_out) - - return tuple(outputs) - - -def get_danet(dataset='citys', backbone='resnet50', pretrained=False, - root='~/.torch/models', pretrained_base=True, **kwargs): - r"""Dual Attention Network - - Parameters - ---------- - dataset : str, default pascal_voc - The dataset that model pretrained on. (pascal_voc, ade20k) - pretrained : bool or str - Boolean value controls whether to load the default pretrained weights for model. - String value represents the hashtag for a certain version of pretrained weights. - root : str, default '~/.torch/models' - Location for keeping the model parameters. - pretrained_base : bool or str, default True - This will load pretrained backbone network, that was trained on ImageNet. - Examples - -------- - >>> model = get_danet(dataset='pascal_voc', backbone='resnet50', pretrained=False) - >>> print(model) - """ - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = DANet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('danet_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_danet_resnet50_citys(**kwargs): - return get_danet('citys', 'resnet50', **kwargs) - - -def get_danet_resnet101_citys(**kwargs): - return get_danet('citys', 'resnet101', **kwargs) - - -def get_danet_resnet152_citys(**kwargs): - return get_danet('citys', 'resnet152', **kwargs) - - -if __name__ == '__main__': - # img = torch.randn(2, 3, 480, 480) - # model = get_danet_resnet50_citys() - # outputs = model(img) - input = torch.rand(2, 3,512,512) - model = DANet(4, pretrained_base=False) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) \ No newline at end of file diff --git a/AIlib/segutils/core/models/deeplabv3.py b/AIlib/segutils/core/models/deeplabv3.py deleted file mode 100644 index 98d0c02..0000000 --- a/AIlib/segutils/core/models/deeplabv3.py +++ /dev/null @@ -1,185 +0,0 @@ -"""Pyramid Scene Parsing Network""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from .segbase import SegBaseModel -from .fcn import _FCNHead - -__all__ = ['DeepLabV3', 'get_deeplabv3', 'get_deeplabv3_resnet50_voc', 'get_deeplabv3_resnet101_voc', - 'get_deeplabv3_resnet152_voc', 'get_deeplabv3_resnet50_ade', 'get_deeplabv3_resnet101_ade', - 'get_deeplabv3_resnet152_ade'] - - -class DeepLabV3(SegBaseModel): - r"""DeepLabV3 - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - - Reference: - Chen, Liang-Chieh, et al. "Rethinking atrous convolution for semantic image segmentation." - arXiv preprint arXiv:1706.05587 (2017). - """ - - def __init__(self, nclass, backbone='resnet50', aux=False, pretrained_base=True, **kwargs): - super(DeepLabV3, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.head = _DeepLabHead(nclass, **kwargs) - if self.aux: - self.auxlayer = _FCNHead(1024, nclass, **kwargs) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - _, _, c3, c4 = self.base_forward(x) - outputs = [] - x = self.head(c4) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - - if self.aux: - auxout = self.auxlayer(c3) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - return tuple(outputs) - - -class _DeepLabHead(nn.Module): - def __init__(self, nclass, norm_layer=nn.BatchNorm2d, norm_kwargs=None, **kwargs): - super(_DeepLabHead, self).__init__() - self.aspp = _ASPP(2048, [12, 24, 36], norm_layer=norm_layer, norm_kwargs=norm_kwargs, **kwargs) - self.block = nn.Sequential( - nn.Conv2d(256, 256, 3, padding=1, bias=False), - norm_layer(256, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True), - nn.Dropout(0.1), - nn.Conv2d(256, nclass, 1) - ) - - def forward(self, x): - x = self.aspp(x) - return self.block(x) - - -class _ASPPConv(nn.Module): - def __init__(self, in_channels, out_channels, atrous_rate, norm_layer, norm_kwargs): - super(_ASPPConv, self).__init__() - self.block = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 3, padding=atrous_rate, dilation=atrous_rate, bias=False), - norm_layer(out_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - - def forward(self, x): - return self.block(x) - - -class _AsppPooling(nn.Module): - def __init__(self, in_channels, out_channels, norm_layer, norm_kwargs, **kwargs): - super(_AsppPooling, self).__init__() - self.gap = nn.Sequential( - nn.AdaptiveAvgPool2d(1), - nn.Conv2d(in_channels, out_channels, 1, bias=False), - norm_layer(out_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - - def forward(self, x): - size = x.size()[2:] - pool = self.gap(x) - out = F.interpolate(pool, size, mode='bilinear', align_corners=True) - return out - - -class _ASPP(nn.Module): - def __init__(self, in_channels, atrous_rates, norm_layer, norm_kwargs, **kwargs): - super(_ASPP, self).__init__() - out_channels = 256 - self.b0 = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 1, bias=False), - norm_layer(out_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - - rate1, rate2, rate3 = tuple(atrous_rates) - self.b1 = _ASPPConv(in_channels, out_channels, rate1, norm_layer, norm_kwargs) - self.b2 = _ASPPConv(in_channels, out_channels, rate2, norm_layer, norm_kwargs) - self.b3 = _ASPPConv(in_channels, out_channels, rate3, norm_layer, norm_kwargs) - self.b4 = _AsppPooling(in_channels, out_channels, norm_layer=norm_layer, norm_kwargs=norm_kwargs) - - self.project = nn.Sequential( - nn.Conv2d(5 * out_channels, out_channels, 1, bias=False), - norm_layer(out_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True), - nn.Dropout(0.5) - ) - - def forward(self, x): - feat1 = self.b0(x) - feat2 = self.b1(x) - feat3 = self.b2(x) - feat4 = self.b3(x) - feat5 = self.b4(x) - x = torch.cat((feat1, feat2, feat3, feat4, feat5), dim=1) - x = self.project(x) - return x - - -def get_deeplabv3(dataset='pascal_voc', backbone='resnet50', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = DeepLabV3(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('deeplabv3_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_deeplabv3_resnet50_voc(**kwargs): - return get_deeplabv3('pascal_voc', 'resnet50', **kwargs) - - -def get_deeplabv3_resnet101_voc(**kwargs): - return get_deeplabv3('pascal_voc', 'resnet101', **kwargs) - - -def get_deeplabv3_resnet152_voc(**kwargs): - return get_deeplabv3('pascal_voc', 'resnet152', **kwargs) - - -def get_deeplabv3_resnet50_ade(**kwargs): - return get_deeplabv3('ade20k', 'resnet50', **kwargs) - - -def get_deeplabv3_resnet101_ade(**kwargs): - return get_deeplabv3('ade20k', 'resnet101', **kwargs) - - -def get_deeplabv3_resnet152_ade(**kwargs): - return get_deeplabv3('ade20k', 'resnet152', **kwargs) - - -if __name__ == '__main__': - model = get_deeplabv3_resnet50_voc() - img = torch.randn(2, 3, 480, 480) - output = model(img) diff --git a/AIlib/segutils/core/models/deeplabv3_plus.py b/AIlib/segutils/core/models/deeplabv3_plus.py deleted file mode 100644 index 9b5a703..0000000 --- a/AIlib/segutils/core/models/deeplabv3_plus.py +++ /dev/null @@ -1,142 +0,0 @@ -import torch -import torch.nn as nn -import torch.nn.functional as F - -from .base_models.xception import get_xception -from .deeplabv3 import _ASPP -from .fcn import _FCNHead -from ..nn import _ConvBNReLU - -__all__ = ['DeepLabV3Plus', 'get_deeplabv3_plus', 'get_deeplabv3_plus_xception_voc'] - - -class DeepLabV3Plus(nn.Module): - r"""DeepLabV3Plus - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'xception'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - - Reference: - Chen, Liang-Chieh, et al. "Encoder-Decoder with Atrous Separable Convolution for Semantic - Image Segmentation." - """ - - def __init__(self, nclass, backbone='xception', aux=True, pretrained_base=True, dilated=True, **kwargs): - super(DeepLabV3Plus, self).__init__() - self.aux = aux - self.nclass = nclass - output_stride = 8 if dilated else 32 - - self.pretrained = get_xception(pretrained=pretrained_base, output_stride=output_stride, **kwargs) - - # deeplabv3 plus - self.head = _DeepLabHead(nclass, **kwargs) - if aux: - self.auxlayer = _FCNHead(728, nclass, **kwargs) - - def base_forward(self, x): - # Entry flow - x = self.pretrained.conv1(x) - x = self.pretrained.bn1(x) - x = self.pretrained.relu(x) - - x = self.pretrained.conv2(x) - x = self.pretrained.bn2(x) - x = self.pretrained.relu(x) - - x = self.pretrained.block1(x) - # add relu here - x = self.pretrained.relu(x) - low_level_feat = x - - x = self.pretrained.block2(x) - x = self.pretrained.block3(x) - - # Middle flow - x = self.pretrained.midflow(x) - mid_level_feat = x - - # Exit flow - x = self.pretrained.block20(x) - x = self.pretrained.relu(x) - x = self.pretrained.conv3(x) - x = self.pretrained.bn3(x) - x = self.pretrained.relu(x) - - x = self.pretrained.conv4(x) - x = self.pretrained.bn4(x) - x = self.pretrained.relu(x) - - x = self.pretrained.conv5(x) - x = self.pretrained.bn5(x) - x = self.pretrained.relu(x) - return low_level_feat, mid_level_feat, x - - def forward(self, x): - size = x.size()[2:] - c1, c3, c4 = self.base_forward(x) - outputs = list() - x = self.head(c4, c1) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - if self.aux: - auxout = self.auxlayer(c3) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - return tuple(outputs) - - -class _DeepLabHead(nn.Module): - def __init__(self, nclass, c1_channels=128, norm_layer=nn.BatchNorm2d, **kwargs): - super(_DeepLabHead, self).__init__() - self.aspp = _ASPP(2048, [12, 24, 36], norm_layer=norm_layer, **kwargs) - self.c1_block = _ConvBNReLU(c1_channels, 48, 3, padding=1, norm_layer=norm_layer) - self.block = nn.Sequential( - _ConvBNReLU(304, 256, 3, padding=1, norm_layer=norm_layer), - nn.Dropout(0.5), - _ConvBNReLU(256, 256, 3, padding=1, norm_layer=norm_layer), - nn.Dropout(0.1), - nn.Conv2d(256, nclass, 1)) - - def forward(self, x, c1): - size = c1.size()[2:] - c1 = self.c1_block(c1) - x = self.aspp(x) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - return self.block(torch.cat([x, c1], dim=1)) - - -def get_deeplabv3_plus(dataset='pascal_voc', backbone='xception', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = DeepLabV3Plus(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict( - torch.load(get_model_file('deeplabv3_plus_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_deeplabv3_plus_xception_voc(**kwargs): - return get_deeplabv3_plus('pascal_voc', 'xception', **kwargs) - - -if __name__ == '__main__': - model = get_deeplabv3_plus_xception_voc() diff --git a/AIlib/segutils/core/models/denseaspp.py b/AIlib/segutils/core/models/denseaspp.py deleted file mode 100644 index 1582375..0000000 --- a/AIlib/segutils/core/models/denseaspp.py +++ /dev/null @@ -1,198 +0,0 @@ -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.models.base_models.densenet import * -from core.models.fcn import _FCNHead - -__all__ = ['DenseASPP', 'get_denseaspp', 'get_denseaspp_densenet121_citys', - 'get_denseaspp_densenet161_citys', 'get_denseaspp_densenet169_citys', 'get_denseaspp_densenet201_citys'] - - -class DenseASPP(nn.Module): - def __init__(self, nclass, backbone='densenet121', aux=False, jpu=False, - pretrained_base=True, dilate_scale=8, **kwargs): - super(DenseASPP, self).__init__() - self.nclass = nclass - self.aux = aux - self.dilate_scale = dilate_scale - if backbone == 'densenet121': - self.pretrained = dilated_densenet121(dilate_scale, pretrained=pretrained_base, **kwargs) - elif backbone == 'densenet161': - self.pretrained = dilated_densenet161(dilate_scale, pretrained=pretrained_base, **kwargs) - elif backbone == 'densenet169': - self.pretrained = dilated_densenet169(dilate_scale, pretrained=pretrained_base, **kwargs) - elif backbone == 'densenet201': - self.pretrained = dilated_densenet201(dilate_scale, pretrained=pretrained_base, **kwargs) - else: - raise RuntimeError('unknown backbone: {}'.format(backbone)) - in_channels = self.pretrained.num_features - - self.head = _DenseASPPHead(in_channels, nclass) - - if aux: - self.auxlayer = _FCNHead(in_channels, nclass, **kwargs) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - #print('size', size) #torch.Size([512, 512]) - features = self.pretrained.features(x) - #print('22',features.shape) #torch.Size([2, 1024, 64, 64]) - if self.dilate_scale > 8: - features = F.interpolate(features, scale_factor=2, mode='bilinear', align_corners=True) - outputs = [] - x = self.head(features) #torch.Size([2, 4, 64, 64]) - #print('x.shape',x.shape) - x = F.interpolate(x, size, mode='bilinear', align_corners=True)#直接64到512。。。。效果还这么好！ - outputs.append(x) - - if self.aux: - auxout = self.auxlayer(features) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - #return tuple(outputs) - return outputs[0] - -class _DenseASPPHead(nn.Module): - def __init__(self, in_channels, nclass, norm_layer=nn.BatchNorm2d, norm_kwargs=None, **kwargs): - super(_DenseASPPHead, self).__init__() - self.dense_aspp_block = _DenseASPPBlock(in_channels, 256, 64, norm_layer, norm_kwargs) - self.block = nn.Sequential( - nn.Dropout(0.1), - nn.Conv2d(in_channels + 5 * 64, nclass, 1) - ) - - def forward(self, x): - x = self.dense_aspp_block(x) - return self.block(x) - - -class _DenseASPPConv(nn.Sequential): - def __init__(self, in_channels, inter_channels, out_channels, atrous_rate, - drop_rate=0.1, norm_layer=nn.BatchNorm2d, norm_kwargs=None): - super(_DenseASPPConv, self).__init__() - self.add_module('conv1', nn.Conv2d(in_channels, inter_channels, 1)), - self.add_module('bn1', norm_layer(inter_channels, **({} if norm_kwargs is None else norm_kwargs))), - self.add_module('relu1', nn.ReLU(True)), - self.add_module('conv2', nn.Conv2d(inter_channels, out_channels, 3, dilation=atrous_rate, padding=atrous_rate)), - self.add_module('bn2', norm_layer(out_channels, **({} if norm_kwargs is None else norm_kwargs))), - self.add_module('relu2', nn.ReLU(True)), - self.drop_rate = drop_rate - - def forward(self, x): - features = super(_DenseASPPConv, self).forward(x) - if self.drop_rate > 0: - features = F.dropout(features, p=self.drop_rate, training=self.training) - return features - - -class _DenseASPPBlock(nn.Module): - def __init__(self, in_channels, inter_channels1, inter_channels2, - norm_layer=nn.BatchNorm2d, norm_kwargs=None): - super(_DenseASPPBlock, self).__init__() - self.aspp_3 = _DenseASPPConv(in_channels, inter_channels1, inter_channels2, 3, 0.1, - norm_layer, norm_kwargs) - self.aspp_6 = _DenseASPPConv(in_channels + inter_channels2 * 1, inter_channels1, inter_channels2, 6, 0.1, - norm_layer, norm_kwargs) - self.aspp_12 = _DenseASPPConv(in_channels + inter_channels2 * 2, inter_channels1, inter_channels2, 12, 0.1, - norm_layer, norm_kwargs) - self.aspp_18 = _DenseASPPConv(in_channels + inter_channels2 * 3, inter_channels1, inter_channels2, 18, 0.1, - norm_layer, norm_kwargs) - self.aspp_24 = _DenseASPPConv(in_channels + inter_channels2 * 4, inter_channels1, inter_channels2, 24, 0.1, - norm_layer, norm_kwargs) - - def forward(self, x): - aspp3 = self.aspp_3(x) - x = torch.cat([aspp3, x], dim=1) - - aspp6 = self.aspp_6(x) - x = torch.cat([aspp6, x], dim=1) - - aspp12 = self.aspp_12(x) - x = torch.cat([aspp12, x], dim=1) - - aspp18 = self.aspp_18(x) - x = torch.cat([aspp18, x], dim=1) - - aspp24 = self.aspp_24(x) - x = torch.cat([aspp24, x], dim=1) - - return x - - -def get_denseaspp(dataset='citys', backbone='densenet121', pretrained=False, - root='~/.torch/models', pretrained_base=True, **kwargs): - r"""DenseASPP - - Parameters - ---------- - dataset : str, default citys - The dataset that model pretrained on. (pascal_voc, ade20k) - pretrained : bool or str - Boolean value controls whether to load the default pretrained weights for model. - String value represents the hashtag for a certain version of pretrained weights. - root : str, default '~/.torch/models' - Location for keeping the model parameters. - pretrained_base : bool or str, default True - This will load pretrained backbone network, that was trained on ImageNet. - Examples - -------- - # >>> model = get_denseaspp(dataset='citys', backbone='densenet121', pretrained=False) - # >>> print(model) - """ - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = DenseASPP(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('denseaspp_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_denseaspp_densenet121_citys(**kwargs): - return get_denseaspp('citys', 'densenet121', **kwargs) - - -def get_denseaspp_densenet161_citys(**kwargs): - return get_denseaspp('citys', 'densenet161', **kwargs) - - -def get_denseaspp_densenet169_citys(**kwargs): - return get_denseaspp('citys', 'densenet169', **kwargs) - - -def get_denseaspp_densenet201_citys(**kwargs): - return get_denseaspp('citys', 'densenet201', **kwargs) - - -if __name__ == '__main__': - # img = torch.randn(2, 3, 480, 480) - # model = get_denseaspp_densenet121_citys() - # outputs = model(img) - input = torch.rand(2, 3, 512, 512) - model = DenseASPP(4, pretrained_base=True) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) \ No newline at end of file diff --git a/AIlib/segutils/core/models/dfanet.py b/AIlib/segutils/core/models/dfanet.py deleted file mode 100644 index 15e3be0..0000000 --- a/AIlib/segutils/core/models/dfanet.py +++ /dev/null @@ -1,129 +0,0 @@ -""" Deep Feature Aggregation""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.models.base_models import Enc, FCAttention, get_xception_a -from core.nn import _ConvBNReLU - -__all__ = ['DFANet', 'get_dfanet', 'get_dfanet_citys'] - - -class DFANet(nn.Module): - def __init__(self, nclass, backbone='', aux=False, jpu=False, pretrained_base=False, **kwargs): - super(DFANet, self).__init__() - self.pretrained = get_xception_a(pretrained_base, **kwargs) - - self.enc2_2 = Enc(240, 48, 4, **kwargs) - self.enc3_2 = Enc(144, 96, 6, **kwargs) - self.enc4_2 = Enc(288, 192, 4, **kwargs) - self.fca_2 = FCAttention(192, **kwargs) - - self.enc2_3 = Enc(240, 48, 4, **kwargs) - self.enc3_3 = Enc(144, 96, 6, **kwargs) - self.enc3_4 = Enc(288, 192, 4, **kwargs) - self.fca_3 = FCAttention(192, **kwargs) - - self.enc2_1_reduce = _ConvBNReLU(48, 32, 1, **kwargs) - self.enc2_2_reduce = _ConvBNReLU(48, 32, 1, **kwargs) - self.enc2_3_reduce = _ConvBNReLU(48, 32, 1, **kwargs) - self.conv_fusion = _ConvBNReLU(32, 32, 1, **kwargs) - - self.fca_1_reduce = _ConvBNReLU(192, 32, 1, **kwargs) - self.fca_2_reduce = _ConvBNReLU(192, 32, 1, **kwargs) - self.fca_3_reduce = _ConvBNReLU(192, 32, 1, **kwargs) - self.conv_out = nn.Conv2d(32, nclass, 1) - - self.__setattr__('exclusive', ['enc2_2', 'enc3_2', 'enc4_2', 'fca_2', 'enc2_3', 'enc3_3', 'enc3_4', 'fca_3', - 'enc2_1_reduce', 'enc2_2_reduce', 'enc2_3_reduce', 'conv_fusion', 'fca_1_reduce', - 'fca_2_reduce', 'fca_3_reduce', 'conv_out']) - - def forward(self, x): - # backbone - stage1_conv1 = self.pretrained.conv1(x) - stage1_enc2 = self.pretrained.enc2(stage1_conv1) - stage1_enc3 = self.pretrained.enc3(stage1_enc2) - stage1_enc4 = self.pretrained.enc4(stage1_enc3) - stage1_fca = self.pretrained.fca(stage1_enc4) - stage1_out = F.interpolate(stage1_fca, scale_factor=4, mode='bilinear', align_corners=True) - - # stage2 - stage2_enc2 = self.enc2_2(torch.cat([stage1_enc2, stage1_out], dim=1)) - stage2_enc3 = self.enc3_2(torch.cat([stage1_enc3, stage2_enc2], dim=1)) - stage2_enc4 = self.enc4_2(torch.cat([stage1_enc4, stage2_enc3], dim=1)) - stage2_fca = self.fca_2(stage2_enc4) - stage2_out = F.interpolate(stage2_fca, scale_factor=4, mode='bilinear', align_corners=True) - - # stage3 - stage3_enc2 = self.enc2_3(torch.cat([stage2_enc2, stage2_out], dim=1)) - stage3_enc3 = self.enc3_3(torch.cat([stage2_enc3, stage3_enc2], dim=1)) - stage3_enc4 = self.enc3_4(torch.cat([stage2_enc4, stage3_enc3], dim=1)) - stage3_fca = self.fca_3(stage3_enc4) - - stage1_enc2_decoder = self.enc2_1_reduce(stage1_enc2) - stage2_enc2_docoder = F.interpolate(self.enc2_2_reduce(stage2_enc2), scale_factor=2, - mode='bilinear', align_corners=True) - stage3_enc2_decoder = F.interpolate(self.enc2_3_reduce(stage3_enc2), scale_factor=4, - mode='bilinear', align_corners=True) - fusion = stage1_enc2_decoder + stage2_enc2_docoder + stage3_enc2_decoder - fusion1 = self.conv_fusion(fusion) - - stage1_fca_decoder = F.interpolate(self.fca_1_reduce(stage1_fca), scale_factor=4, - mode='bilinear', align_corners=True) - stage2_fca_decoder = F.interpolate(self.fca_2_reduce(stage2_fca), scale_factor=8, - mode='bilinear', align_corners=True) - stage3_fca_decoder = F.interpolate(self.fca_3_reduce(stage3_fca), scale_factor=16, - mode='bilinear', align_corners=True) - #print(fusion.shape,stage1_fca_decoder.shape,stage2_fca_decoder.shape,stage3_fca_decoder.shape) - fusion2 = fusion1 + stage1_fca_decoder + stage2_fca_decoder + stage3_fca_decoder - - outputs = list() - out = self.conv_out(fusion2) - out1 = F.interpolate(out, scale_factor=4, mode='bilinear', align_corners=True) - outputs.append(out1) - - #return tuple(outputs) - return outputs[0] - -def get_dfanet(dataset='citys', backbone='', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = DFANet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('dfanet_%s' % (acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_dfanet_citys(**kwargs): - return get_dfanet('citys', **kwargs) - - -if __name__ == '__main__': - #model = get_dfanet_citys() - input = torch.rand(2, 3, 512, 512) - model = DFANet(4, pretrained_base=False) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) diff --git a/AIlib/segutils/core/models/dinknet.py b/AIlib/segutils/core/models/dinknet.py deleted file mode 100644 index a36b90c..0000000 --- a/AIlib/segutils/core/models/dinknet.py +++ /dev/null @@ -1,359 +0,0 @@ -""" -Codes of LinkNet based on https://github.com/snakers4/spacenet-three -""" -import torch -import torch.nn as nn -from torch.autograd import Variable -from torchvision import models -import torch.nn.functional as F - -from functools import partial - -nonlinearity = partial(F.relu,inplace=True) - -class Dblock_more_dilate(nn.Module): - def __init__(self,channel): - super(Dblock_more_dilate, self).__init__() - self.dilate1 = nn.Conv2d(channel, channel, kernel_size=3, dilation=1, padding=1) - self.dilate2 = nn.Conv2d(channel, channel, kernel_size=3, dilation=2, padding=2) - self.dilate3 = nn.Conv2d(channel, channel, kernel_size=3, dilation=4, padding=4) - self.dilate4 = nn.Conv2d(channel, channel, kernel_size=3, dilation=8, padding=8) - self.dilate5 = nn.Conv2d(channel, channel, kernel_size=3, dilation=16, padding=16) - for m in self.modules(): - if isinstance(m, nn.Conv2d) or isinstance(m, nn.ConvTranspose2d): - if m.bias is not None: - m.bias.data.zero_() - - def forward(self, x): - dilate1_out = nonlinearity(self.dilate1(x)) - dilate2_out = nonlinearity(self.dilate2(dilate1_out)) - dilate3_out = nonlinearity(self.dilate3(dilate2_out)) - dilate4_out = nonlinearity(self.dilate4(dilate3_out)) - dilate5_out = nonlinearity(self.dilate5(dilate4_out)) - out = x + dilate1_out + dilate2_out + dilate3_out + dilate4_out + dilate5_out - return out - -class Dblock(nn.Module): - def __init__(self,channel): - super(Dblock, self).__init__() - self.dilate1 = nn.Conv2d(channel, channel, kernel_size=3, dilation=1, padding=1) - self.dilate2 = nn.Conv2d(channel, channel, kernel_size=3, dilation=2, padding=2) - self.dilate3 = nn.Conv2d(channel, channel, kernel_size=3, dilation=4, padding=4) - self.dilate4 = nn.Conv2d(channel, channel, kernel_size=3, dilation=8, padding=8) - #self.dilate5 = nn.Conv2d(channel, channel, kernel_size=3, dilation=16, padding=16) - for m in self.modules(): - if isinstance(m, nn.Conv2d) or isinstance(m, nn.ConvTranspose2d): - if m.bias is not None: - m.bias.data.zero_() - - def forward(self, x): - dilate1_out = nonlinearity(self.dilate1(x)) - dilate2_out = nonlinearity(self.dilate2(dilate1_out)) - dilate3_out = nonlinearity(self.dilate3(dilate2_out)) - dilate4_out = nonlinearity(self.dilate4(dilate3_out)) - #dilate5_out = nonlinearity(self.dilate5(dilate4_out)) - out = x + dilate1_out + dilate2_out + dilate3_out + dilate4_out# + dilate5_out - return out - -class DecoderBlock(nn.Module): - def __init__(self, in_channels, n_filters): - super(DecoderBlock,self).__init__() - - self.conv1 = nn.Conv2d(in_channels, in_channels // 4, 1) - self.norm1 = nn.BatchNorm2d(in_channels // 4) - self.relu1 = nonlinearity - - self.deconv2 = nn.ConvTranspose2d(in_channels // 4, in_channels // 4, 3, stride=2, padding=1, output_padding=1) - self.norm2 = nn.BatchNorm2d(in_channels // 4) - self.relu2 = nonlinearity - - self.conv3 = nn.Conv2d(in_channels // 4, n_filters, 1) - self.norm3 = nn.BatchNorm2d(n_filters) - self.relu3 = nonlinearity - - def forward(self, x): - x = self.conv1(x) - x = self.norm1(x) - x = self.relu1(x) - x = self.deconv2(x) - x = self.norm2(x) - x = self.relu2(x) - x = self.conv3(x) - x = self.norm3(x) - x = self.relu3(x) - return x - -class DinkNet34_less_pool(nn.Module): - def __init__(self, num_classes=1): - super(DinkNet34_more_dilate, self).__init__() - - filters = [64, 128, 256, 512] - resnet = models.resnet34(pretrained=True) - - self.firstconv = resnet.conv1 - self.firstbn = resnet.bn1 - self.firstrelu = resnet.relu - self.firstmaxpool = resnet.maxpool - self.encoder1 = resnet.layer1 - self.encoder2 = resnet.layer2 - self.encoder3 = resnet.layer3 - - self.dblock = Dblock_more_dilate(256) - - self.decoder3 = DecoderBlock(filters[2], filters[1]) - self.decoder2 = DecoderBlock(filters[1], filters[0]) - self.decoder1 = DecoderBlock(filters[0], filters[0]) - - self.finaldeconv1 = nn.ConvTranspose2d(filters[0], 32, 4, 2, 1) - self.finalrelu1 = nonlinearity - self.finalconv2 = nn.Conv2d(32, 32, 3, padding=1) - self.finalrelu2 = nonlinearity - self.finalconv3 = nn.Conv2d(32, num_classes, 3, padding=1) - - def forward(self, x): - # Encoder - x = self.firstconv(x) - x = self.firstbn(x) - x = self.firstrelu(x) - x = self.firstmaxpool(x) - e1 = self.encoder1(x) - e2 = self.encoder2(e1) - e3 = self.encoder3(e2) - - #Center - e3 = self.dblock(e3) - - # Decoder - d3 = self.decoder3(e3) + e2 - d2 = self.decoder2(d3) + e1 - d1 = self.decoder1(d2) - - # Final Classification - out = self.finaldeconv1(d1) - out = self.finalrelu1(out) - out = self.finalconv2(out) - out = self.finalrelu2(out) - out = self.finalconv3(out) - - #return F.sigmoid(out) - return out - -class DinkNet34(nn.Module): - def __init__(self, num_classes=1, num_channels=3): - super(DinkNet34, self).__init__() - - filters = [64, 128, 256, 512] - resnet = models.resnet34(pretrained=True) - self.firstconv = resnet.conv1 - self.firstbn = resnet.bn1 - self.firstrelu = resnet.relu - self.firstmaxpool = resnet.maxpool - self.encoder1 = resnet.layer1 - self.encoder2 = resnet.layer2 - self.encoder3 = resnet.layer3 - self.encoder4 = resnet.layer4 - - self.dblock = Dblock(512) - - self.decoder4 = DecoderBlock(filters[3], filters[2]) - self.decoder3 = DecoderBlock(filters[2], filters[1]) - self.decoder2 = DecoderBlock(filters[1], filters[0]) - self.decoder1 = DecoderBlock(filters[0], filters[0]) - - self.finaldeconv1 = nn.ConvTranspose2d(filters[0], 32, 4, 2, 1) - self.finalrelu1 = nonlinearity - self.finalconv2 = nn.Conv2d(32, 32, 3, padding=1) - self.finalrelu2 = nonlinearity - self.finalconv3 = nn.Conv2d(32, num_classes, 3, padding=1) - - def forward(self, x): - # Encoder - x = self.firstconv(x) - x = self.firstbn(x) - x = self.firstrelu(x) - x = self.firstmaxpool(x) - e1 = self.encoder1(x) - e2 = self.encoder2(e1) - e3 = self.encoder3(e2) - e4 = self.encoder4(e3) - - # Center - e4 = self.dblock(e4) - - # Decoder - d4 = self.decoder4(e4) + e3 - d3 = self.decoder3(d4) + e2 - d2 = self.decoder2(d3) + e1 - d1 = self.decoder1(d2) - - out = self.finaldeconv1(d1) - out = self.finalrelu1(out) - out = self.finalconv2(out) - out = self.finalrelu2(out) - out = self.finalconv3(out) - - #return F.sigmoid(out) - return out - -class DinkNet50(nn.Module): - def __init__(self, num_classes=1): - super(DinkNet50, self).__init__() - - filters = [256, 512, 1024, 2048] - resnet = models.resnet50(pretrained=True) - self.firstconv = resnet.conv1 - self.firstbn = resnet.bn1 - self.firstrelu = resnet.relu - self.firstmaxpool = resnet.maxpool - self.encoder1 = resnet.layer1 - self.encoder2 = resnet.layer2 - self.encoder3 = resnet.layer3 - self.encoder4 = resnet.layer4 - - self.dblock = Dblock_more_dilate(2048) - - self.decoder4 = DecoderBlock(filters[3], filters[2]) - self.decoder3 = DecoderBlock(filters[2], filters[1]) - self.decoder2 = DecoderBlock(filters[1], filters[0]) - self.decoder1 = DecoderBlock(filters[0], filters[0]) - - self.finaldeconv1 = nn.ConvTranspose2d(filters[0], 32, 4, 2, 1) - self.finalrelu1 = nonlinearity - self.finalconv2 = nn.Conv2d(32, 32, 3, padding=1) - self.finalrelu2 = nonlinearity - self.finalconv3 = nn.Conv2d(32, num_classes, 3, padding=1) - - def forward(self, x): - # Encoder - x = self.firstconv(x) - x = self.firstbn(x) - x = self.firstrelu(x) - x = self.firstmaxpool(x) - e1 = self.encoder1(x) - e2 = self.encoder2(e1) - e3 = self.encoder3(e2) - e4 = self.encoder4(e3) - - # Center - e4 = self.dblock(e4) - - # Decoder - d4 = self.decoder4(e4) + e3 - d3 = self.decoder3(d4) + e2 - d2 = self.decoder2(d3) + e1 - d1 = self.decoder1(d2) - out = self.finaldeconv1(d1) - out = self.finalrelu1(out) - out = self.finalconv2(out) - out = self.finalrelu2(out) - out = self.finalconv3(out) - - #return F.sigmoid(out) - return out - -class DinkNet101(nn.Module): - def __init__(self, num_classes=1): - super(DinkNet101, self).__init__() - - filters = [256, 512, 1024, 2048] - resnet = models.resnet101(pretrained=True) - self.firstconv = resnet.conv1 - self.firstbn = resnet.bn1 - self.firstrelu = resnet.relu - self.firstmaxpool = resnet.maxpool - self.encoder1 = resnet.layer1 - self.encoder2 = resnet.layer2 - self.encoder3 = resnet.layer3 - self.encoder4 = resnet.layer4 - - self.dblock = Dblock_more_dilate(2048) - - self.decoder4 = DecoderBlock(filters[3], filters[2]) - self.decoder3 = DecoderBlock(filters[2], filters[1]) - self.decoder2 = DecoderBlock(filters[1], filters[0]) - self.decoder1 = DecoderBlock(filters[0], filters[0]) - - self.finaldeconv1 = nn.ConvTranspose2d(filters[0], 32, 4, 2, 1) - self.finalrelu1 = nonlinearity - self.finalconv2 = nn.Conv2d(32, 32, 3, padding=1) - self.finalrelu2 = nonlinearity - self.finalconv3 = nn.Conv2d(32, num_classes, 3, padding=1) - - def forward(self, x): - # Encoder - x = self.firstconv(x) - x = self.firstbn(x) - x = self.firstrelu(x) - x = self.firstmaxpool(x) - e1 = self.encoder1(x) - e2 = self.encoder2(e1) - e3 = self.encoder3(e2) - e4 = self.encoder4(e3) - - # Center - e4 = self.dblock(e4) - - # Decoder - d4 = self.decoder4(e4) + e3 - d3 = self.decoder3(d4) + e2 - d2 = self.decoder2(d3) + e1 - d1 = self.decoder1(d2) - out = self.finaldeconv1(d1) - out = self.finalrelu1(out) - out = self.finalconv2(out) - out = self.finalrelu2(out) - out = self.finalconv3(out) - - #return F.sigmoid(out) - return out - -class LinkNet34(nn.Module): - def __init__(self, num_classes=1): - super(LinkNet34, self).__init__() - - filters = [64, 128, 256, 512] - resnet = models.resnet34(pretrained=True) - self.firstconv = resnet.conv1 - self.firstbn = resnet.bn1 - self.firstrelu = resnet.relu - self.firstmaxpool = resnet.maxpool - self.encoder1 = resnet.layer1 - self.encoder2 = resnet.layer2 - self.encoder3 = resnet.layer3 - self.encoder4 = resnet.layer4 - - self.decoder4 = DecoderBlock(filters[3], filters[2]) - self.decoder3 = DecoderBlock(filters[2], filters[1]) - self.decoder2 = DecoderBlock(filters[1], filters[0]) - self.decoder1 = DecoderBlock(filters[0], filters[0]) - - self.finaldeconv1 = nn.ConvTranspose2d(filters[0], 32, 3, stride=2) - self.finalrelu1 = nonlinearity - self.finalconv2 = nn.Conv2d(32, 32, 3) - self.finalrelu2 = nonlinearity - self.finalconv3 = nn.Conv2d(32, num_classes, 2, padding=1) - - def forward(self, x): - # Encoder - x = self.firstconv(x) - x = self.firstbn(x) - x = self.firstrelu(x) - x = self.firstmaxpool(x) - e1 = self.encoder1(x) - e2 = self.encoder2(e1) - e3 = self.encoder3(e2) - e4 = self.encoder4(e3) - - # Decoder - d4 = self.decoder4(e4) + e3 - d3 = self.decoder3(d4) + e2 - d2 = self.decoder2(d3) + e1 - d1 = self.decoder1(d2) - out = self.finaldeconv1(d1) - out = self.finalrelu1(out) - out = self.finalconv2(out) - out = self.finalrelu2(out) - out = self.finalconv3(out) - - #return F.sigmoid(out) - return out \ No newline at end of file diff --git a/AIlib/segutils/core/models/dunet.py b/AIlib/segutils/core/models/dunet.py deleted file mode 100644 index affc476..0000000 --- a/AIlib/segutils/core/models/dunet.py +++ /dev/null @@ -1,172 +0,0 @@ -"""Decoders Matter for Semantic Segmentation""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.models.segbase import SegBaseModel -from core.models.fcn import _FCNHead - -__all__ = ['DUNet', 'get_dunet', 'get_dunet_resnet50_pascal_voc', - 'get_dunet_resnet101_pascal_voc', 'get_dunet_resnet152_pascal_voc'] - - -# The model may be wrong because lots of details missing in paper. -class DUNet(SegBaseModel): - """Decoders Matter for Semantic Segmentation - - Reference: - Zhi Tian, Tong He, Chunhua Shen, and Youliang Yan. - "Decoders Matter for Semantic Segmentation: - Data-Dependent Decoding Enables Flexible Feature Aggregation." CVPR, 2019 - """ - - def __init__(self, nclass, backbone='resnet50', aux=True, pretrained_base=True, **kwargs): - super(DUNet, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.head = _DUHead(2144, **kwargs) - self.dupsample = DUpsampling(256, nclass, scale_factor=8, **kwargs) - if aux: - self.auxlayer = _FCNHead(1024, 256, **kwargs) - self.aux_dupsample = DUpsampling(256, nclass, scale_factor=8, **kwargs) - - self.__setattr__('exclusive', - ['dupsample', 'head', 'auxlayer', 'aux_dupsample'] if aux else ['dupsample', 'head']) - - def forward(self, x): - c1, c2, c3, c4 = self.base_forward(x)#继承自SegBaseModel；返回的是resnet的layer1,2,3,4的输出 - outputs = [] - x = self.head(c2, c3, c4) - x = self.dupsample(x) - outputs.append(x) - - if self.aux: - auxout = self.auxlayer(c3) - auxout = self.aux_dupsample(auxout) - outputs.append(auxout) - #return tuple(outputs) - return outputs[0] - -class FeatureFused(nn.Module): - """Module for fused features""" - - def __init__(self, inter_channels=48, norm_layer=nn.BatchNorm2d, **kwargs): - super(FeatureFused, self).__init__() - self.conv2 = nn.Sequential( - nn.Conv2d(512, inter_channels, 1, bias=False), - norm_layer(inter_channels), - nn.ReLU(True) - ) - self.conv3 = nn.Sequential( - nn.Conv2d(1024, inter_channels, 1, bias=False), - norm_layer(inter_channels), - nn.ReLU(True) - ) - - def forward(self, c2, c3, c4): - size = c4.size()[2:] - c2 = self.conv2(F.interpolate(c2, size, mode='bilinear', align_corners=True)) - c3 = self.conv3(F.interpolate(c3, size, mode='bilinear', align_corners=True)) - fused_feature = torch.cat([c4, c3, c2], dim=1) - return fused_feature - - -class _DUHead(nn.Module): - def __init__(self, in_channels, norm_layer=nn.BatchNorm2d, **kwargs): - super(_DUHead, self).__init__() - self.fuse = FeatureFused(norm_layer=norm_layer, **kwargs) - self.block = nn.Sequential( - nn.Conv2d(in_channels, 256, 3, padding=1, bias=False), - norm_layer(256), - nn.ReLU(True), - nn.Conv2d(256, 256, 3, padding=1, bias=False), - norm_layer(256), - nn.ReLU(True) - ) - - def forward(self, c2, c3, c4): - fused_feature = self.fuse(c2, c3, c4) - out = self.block(fused_feature) - return out - - -class DUpsampling(nn.Module): - """DUsampling module""" - - def __init__(self, in_channels, out_channels, scale_factor=2, **kwargs): - super(DUpsampling, self).__init__() - self.scale_factor = scale_factor - self.conv_w = nn.Conv2d(in_channels, out_channels * scale_factor * scale_factor, 1, bias=False) - - def forward(self, x): - x = self.conv_w(x) - n, c, h, w = x.size() - - # N, C, H, W --> N, W, H, C - x = x.permute(0, 3, 2, 1).contiguous() - - # N, W, H, C --> N, W, H * scale, C // scale - x = x.view(n, w, h * self.scale_factor, c // self.scale_factor) - - # N, W, H * scale, C // scale --> N, H * scale, W, C // scale - x = x.permute(0, 2, 1, 3).contiguous() - - # N, H * scale, W, C // scale --> N, H * scale, W * scale, C // (scale ** 2) - x = x.view(n, h * self.scale_factor, w * self.scale_factor, c // (self.scale_factor * self.scale_factor)) - - # N, H * scale, W * scale, C // (scale ** 2) -- > N, C // (scale ** 2), H * scale, W * scale - x = x.permute(0, 3, 1, 2) - - return x - -def get_dunet(dataset='pascal_voc', backbone='resnet50', pretrained=False, - root='~/.torch/models', pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = DUNet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('dunet_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_dunet_resnet50_pascal_voc(**kwargs): - return get_dunet('pascal_voc', 'resnet50', **kwargs) - - -def get_dunet_resnet101_pascal_voc(**kwargs): - return get_dunet('pascal_voc', 'resnet101', **kwargs) - - -def get_dunet_resnet152_pascal_voc(**kwargs): - return get_dunet('pascal_voc', 'resnet152', **kwargs) - - -if __name__ == '__main__': - # img = torch.randn(2, 3, 256, 256) - # model = get_dunet_resnet50_pascal_voc() - # outputs = model(img) - input = torch.rand(2, 3, 224, 224) - model = DUNet(4, pretrained_base=False) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - input = torch.randn(1, 3, 512, 512) - flop, params = profile(model, inputs=(input, )) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) \ No newline at end of file diff --git a/AIlib/segutils/core/models/encnet.py b/AIlib/segutils/core/models/encnet.py deleted file mode 100644 index 585557b..0000000 --- a/AIlib/segutils/core/models/encnet.py +++ /dev/null @@ -1,212 +0,0 @@ -"""Context Encoding for Semantic Segmentation""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from .segbase import SegBaseModel -from .fcn import _FCNHead - -__all__ = ['EncNet', 'EncModule', 'get_encnet', 'get_encnet_resnet50_ade', - 'get_encnet_resnet101_ade', 'get_encnet_resnet152_ade'] - - -class EncNet(SegBaseModel): - def __init__(self, nclass, backbone='resnet50', aux=True, se_loss=True, lateral=False, - pretrained_base=True, **kwargs): - super(EncNet, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.head = _EncHead(2048, nclass, se_loss=se_loss, lateral=lateral, **kwargs) - if aux: - self.auxlayer = _FCNHead(1024, nclass, **kwargs) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - features = self.base_forward(x) - - x = list(self.head(*features)) - x[0] = F.interpolate(x[0], size, mode='bilinear', align_corners=True) - if self.aux: - auxout = self.auxlayer(features[2]) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - x.append(auxout) - return tuple(x) - - -class _EncHead(nn.Module): - def __init__(self, in_channels, nclass, se_loss=True, lateral=True, - norm_layer=nn.BatchNorm2d, norm_kwargs=None, **kwargs): - super(_EncHead, self).__init__() - self.lateral = lateral - self.conv5 = nn.Sequential( - nn.Conv2d(in_channels, 512, 3, padding=1, bias=False), - norm_layer(512, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - if lateral: - self.connect = nn.ModuleList([ - nn.Sequential( - nn.Conv2d(512, 512, 1, bias=False), - norm_layer(512, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True)), - nn.Sequential( - nn.Conv2d(1024, 512, 1, bias=False), - norm_layer(512, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True)), - ]) - self.fusion = nn.Sequential( - nn.Conv2d(3 * 512, 512, 3, padding=1, bias=False), - norm_layer(512, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - self.encmodule = EncModule(512, nclass, ncodes=32, se_loss=se_loss, - norm_layer=norm_layer, norm_kwargs=norm_kwargs, **kwargs) - self.conv6 = nn.Sequential( - nn.Dropout(0.1, False), - nn.Conv2d(512, nclass, 1) - ) - - def forward(self, *inputs): - feat = self.conv5(inputs[-1]) - if self.lateral: - c2 = self.connect[0](inputs[1]) - c3 = self.connect[1](inputs[2]) - feat = self.fusion(torch.cat([feat, c2, c3], 1)) - outs = list(self.encmodule(feat)) - outs[0] = self.conv6(outs[0]) - return tuple(outs) - - -class EncModule(nn.Module): - def __init__(self, in_channels, nclass, ncodes=32, se_loss=True, - norm_layer=nn.BatchNorm2d, norm_kwargs=None, **kwargs): - super(EncModule, self).__init__() - self.se_loss = se_loss - self.encoding = nn.Sequential( - nn.Conv2d(in_channels, in_channels, 1, bias=False), - norm_layer(in_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True), - Encoding(D=in_channels, K=ncodes), - nn.BatchNorm1d(ncodes), - nn.ReLU(True), - Mean(dim=1) - ) - self.fc = nn.Sequential( - nn.Linear(in_channels, in_channels), - nn.Sigmoid() - ) - if self.se_loss: - self.selayer = nn.Linear(in_channels, nclass) - - def forward(self, x): - en = self.encoding(x) - b, c, _, _ = x.size() - gamma = self.fc(en) - y = gamma.view(b, c, 1, 1) - outputs = [F.relu_(x + x * y)] - if self.se_loss: - outputs.append(self.selayer(en)) - return tuple(outputs) - - -class Encoding(nn.Module): - def __init__(self, D, K): - super(Encoding, self).__init__() - # init codewords and smoothing factor - self.D, self.K = D, K - self.codewords = nn.Parameter(torch.Tensor(K, D), requires_grad=True) - self.scale = nn.Parameter(torch.Tensor(K), requires_grad=True) - self.reset_params() - - def reset_params(self): - std1 = 1. / ((self.K * self.D) ** (1 / 2)) - self.codewords.data.uniform_(-std1, std1) - self.scale.data.uniform_(-1, 0) - - def forward(self, X): - # input X is a 4D tensor - assert (X.size(1) == self.D) - B, D = X.size(0), self.D - if X.dim() == 3: - # BxDxN -> BxNxD - X = X.transpose(1, 2).contiguous() - elif X.dim() == 4: - # BxDxHxW -> Bx(HW)xD - X = X.view(B, D, -1).transpose(1, 2).contiguous() - else: - raise RuntimeError('Encoding Layer unknown input dims!') - # assignment weights BxNxK - A = F.softmax(self.scale_l2(X, self.codewords, self.scale), dim=2) - # aggregate - E = self.aggregate(A, X, self.codewords) - return E - - def __repr__(self): - return self.__class__.__name__ + '(' \ - + 'N x' + str(self.D) + '=>' + str(self.K) + 'x' \ - + str(self.D) + ')' - - @staticmethod - def scale_l2(X, C, S): - S = S.view(1, 1, C.size(0), 1) - X = X.unsqueeze(2).expand(X.size(0), X.size(1), C.size(0), C.size(1)) - C = C.unsqueeze(0).unsqueeze(0) - SL = S * (X - C) - SL = SL.pow(2).sum(3) - return SL - - @staticmethod - def aggregate(A, X, C): - A = A.unsqueeze(3) - X = X.unsqueeze(2).expand(X.size(0), X.size(1), C.size(0), C.size(1)) - C = C.unsqueeze(0).unsqueeze(0) - E = A * (X - C) - E = E.sum(1) - return E - - -class Mean(nn.Module): - def __init__(self, dim, keep_dim=False): - super(Mean, self).__init__() - self.dim = dim - self.keep_dim = keep_dim - - def forward(self, input): - return input.mean(self.dim, self.keep_dim) - - -def get_encnet(dataset='pascal_voc', backbone='resnet50', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = EncNet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('encnet_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_encnet_resnet50_ade(**kwargs): - return get_encnet('ade20k', 'resnet50', **kwargs) - - -def get_encnet_resnet101_ade(**kwargs): - return get_encnet('ade20k', 'resnet101', **kwargs) - - -def get_encnet_resnet152_ade(**kwargs): - return get_encnet('ade20k', 'resnet152', **kwargs) - - -if __name__ == '__main__': - img = torch.randn(2, 3, 224, 224) - model = get_encnet_resnet50_ade() - outputs = model(img) diff --git a/AIlib/segutils/core/models/enet.py b/AIlib/segutils/core/models/enet.py deleted file mode 100644 index 853fc65..0000000 --- a/AIlib/segutils/core/models/enet.py +++ /dev/null @@ -1,243 +0,0 @@ -"""Efficient Neural Network""" -import torch -import torch.nn as nn - -__all__ = ['ENet', 'get_enet', 'get_enet_citys'] - - -class ENet(nn.Module): - """Efficient Neural Network""" - - def __init__(self, nclass, backbone='', aux=False, jpu=False, pretrained_base=None, **kwargs): - super(ENet, self).__init__() - self.initial = InitialBlock(13, **kwargs) - - self.bottleneck1_0 = Bottleneck(16, 16, 64, downsampling=True, **kwargs) - self.bottleneck1_1 = Bottleneck(64, 16, 64, **kwargs) - self.bottleneck1_2 = Bottleneck(64, 16, 64, **kwargs) - self.bottleneck1_3 = Bottleneck(64, 16, 64, **kwargs) - self.bottleneck1_4 = Bottleneck(64, 16, 64, **kwargs) - - self.bottleneck2_0 = Bottleneck(64, 32, 128, downsampling=True, **kwargs) - self.bottleneck2_1 = Bottleneck(128, 32, 128, **kwargs) - self.bottleneck2_2 = Bottleneck(128, 32, 128, dilation=2, **kwargs) - self.bottleneck2_3 = Bottleneck(128, 32, 128, asymmetric=True, **kwargs) - self.bottleneck2_4 = Bottleneck(128, 32, 128, dilation=4, **kwargs) - self.bottleneck2_5 = Bottleneck(128, 32, 128, **kwargs) - self.bottleneck2_6 = Bottleneck(128, 32, 128, dilation=8, **kwargs) - self.bottleneck2_7 = Bottleneck(128, 32, 128, asymmetric=True, **kwargs) - self.bottleneck2_8 = Bottleneck(128, 32, 128, dilation=16, **kwargs) - - self.bottleneck3_1 = Bottleneck(128, 32, 128, **kwargs) - self.bottleneck3_2 = Bottleneck(128, 32, 128, dilation=2, **kwargs) - self.bottleneck3_3 = Bottleneck(128, 32, 128, asymmetric=True, **kwargs) - self.bottleneck3_4 = Bottleneck(128, 32, 128, dilation=4, **kwargs) - self.bottleneck3_5 = Bottleneck(128, 32, 128, **kwargs) - self.bottleneck3_6 = Bottleneck(128, 32, 128, dilation=8, **kwargs) - self.bottleneck3_7 = Bottleneck(128, 32, 128, asymmetric=True, **kwargs) - self.bottleneck3_8 = Bottleneck(128, 32, 128, dilation=16, **kwargs) - - self.bottleneck4_0 = UpsamplingBottleneck(128, 16, 64, **kwargs) - self.bottleneck4_1 = Bottleneck(64, 16, 64, **kwargs) - self.bottleneck4_2 = Bottleneck(64, 16, 64, **kwargs) - - self.bottleneck5_0 = UpsamplingBottleneck(64, 4, 16, **kwargs) - self.bottleneck5_1 = Bottleneck(16, 4, 16, **kwargs) - - self.fullconv = nn.ConvTranspose2d(16, nclass, 2, 2, bias=False) - - self.__setattr__('exclusive', ['bottleneck1_0', 'bottleneck1_1', 'bottleneck1_2', 'bottleneck1_3', - 'bottleneck1_4', 'bottleneck2_0', 'bottleneck2_1', 'bottleneck2_2', - 'bottleneck2_3', 'bottleneck2_4', 'bottleneck2_5', 'bottleneck2_6', - 'bottleneck2_7', 'bottleneck2_8', 'bottleneck3_1', 'bottleneck3_2', - 'bottleneck3_3', 'bottleneck3_4', 'bottleneck3_5', 'bottleneck3_6', - 'bottleneck3_7', 'bottleneck3_8', 'bottleneck4_0', 'bottleneck4_1', - 'bottleneck4_2', 'bottleneck5_0', 'bottleneck5_1', 'fullconv']) - - def forward(self, x): - # init - x = self.initial(x) - - # stage 1 - x, max_indices1 = self.bottleneck1_0(x) - x = self.bottleneck1_1(x) - x = self.bottleneck1_2(x) - x = self.bottleneck1_3(x) - x = self.bottleneck1_4(x) - - # stage 2 - x, max_indices2 = self.bottleneck2_0(x) - x = self.bottleneck2_1(x) - x = self.bottleneck2_2(x) - x = self.bottleneck2_3(x) - x = self.bottleneck2_4(x) - x = self.bottleneck2_5(x) - x = self.bottleneck2_6(x) - x = self.bottleneck2_7(x) - x = self.bottleneck2_8(x) - - # stage 3 - x = self.bottleneck3_1(x) - x = self.bottleneck3_2(x) - x = self.bottleneck3_3(x) - x = self.bottleneck3_4(x) - x = self.bottleneck3_6(x) - x = self.bottleneck3_7(x) - x = self.bottleneck3_8(x) - - # stage 4 - x = self.bottleneck4_0(x, max_indices2) - x = self.bottleneck4_1(x) - x = self.bottleneck4_2(x) - - # stage 5 - x = self.bottleneck5_0(x, max_indices1) - x = self.bottleneck5_1(x) - - # out - x = self.fullconv(x) - return tuple([x]) - - -class InitialBlock(nn.Module): - """ENet initial block""" - - def __init__(self, out_channels, norm_layer=nn.BatchNorm2d, **kwargs): - super(InitialBlock, self).__init__() - self.conv = nn.Conv2d(3, out_channels, 3, 2, 1, bias=False) - self.maxpool = nn.MaxPool2d(2, 2) - self.bn = norm_layer(out_channels + 3) - self.act = nn.PReLU() - - def forward(self, x): - x_conv = self.conv(x) - x_pool = self.maxpool(x) - x = torch.cat([x_conv, x_pool], dim=1) - x = self.bn(x) - x = self.act(x) - return x - - -class Bottleneck(nn.Module): - """Bottlenecks include regular, asymmetric, downsampling, dilated""" - - def __init__(self, in_channels, inter_channels, out_channels, dilation=1, asymmetric=False, - downsampling=False, norm_layer=nn.BatchNorm2d, **kwargs): - super(Bottleneck, self).__init__() - self.downsamping = downsampling - if downsampling: - self.maxpool = nn.MaxPool2d(2, 2, return_indices=True) - self.conv_down = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 1, bias=False), - norm_layer(out_channels) - ) - - self.conv1 = nn.Sequential( - nn.Conv2d(in_channels, inter_channels, 1, bias=False), - norm_layer(inter_channels), - nn.PReLU() - ) - - if downsampling: - self.conv2 = nn.Sequential( - nn.Conv2d(inter_channels, inter_channels, 2, stride=2, bias=False), - norm_layer(inter_channels), - nn.PReLU() - ) - else: - if asymmetric: - self.conv2 = nn.Sequential( - nn.Conv2d(inter_channels, inter_channels, (5, 1), padding=(2, 0), bias=False), - nn.Conv2d(inter_channels, inter_channels, (1, 5), padding=(0, 2), bias=False), - norm_layer(inter_channels), - nn.PReLU() - ) - else: - self.conv2 = nn.Sequential( - nn.Conv2d(inter_channels, inter_channels, 3, dilation=dilation, padding=dilation, bias=False), - norm_layer(inter_channels), - nn.PReLU() - ) - self.conv3 = nn.Sequential( - nn.Conv2d(inter_channels, out_channels, 1, bias=False), - norm_layer(out_channels), - nn.Dropout2d(0.1) - ) - self.act = nn.PReLU() - - def forward(self, x): - identity = x - if self.downsamping: - identity, max_indices = self.maxpool(identity) - identity = self.conv_down(identity) - - out = self.conv1(x) - out = self.conv2(out) - out = self.conv3(out) - out = self.act(out + identity) - - if self.downsamping: - return out, max_indices - else: - return out - - -class UpsamplingBottleneck(nn.Module): - """upsampling Block""" - - def __init__(self, in_channels, inter_channels, out_channels, norm_layer=nn.BatchNorm2d, **kwargs): - super(UpsamplingBottleneck, self).__init__() - self.conv = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 1, bias=False), - norm_layer(out_channels) - ) - self.upsampling = nn.MaxUnpool2d(2) - - self.block = nn.Sequential( - nn.Conv2d(in_channels, inter_channels, 1, bias=False), - norm_layer(inter_channels), - nn.PReLU(), - nn.ConvTranspose2d(inter_channels, inter_channels, 2, 2, bias=False), - norm_layer(inter_channels), - nn.PReLU(), - nn.Conv2d(inter_channels, out_channels, 1, bias=False), - norm_layer(out_channels), - nn.Dropout2d(0.1) - ) - self.act = nn.PReLU() - - def forward(self, x, max_indices): - out_up = self.conv(x) - out_up = self.upsampling(out_up, max_indices) - - out_ext = self.block(x) - out = self.act(out_up + out_ext) - return out - - -def get_enet(dataset='citys', backbone='', pretrained=False, root='~/.torch/models', pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from core.data.dataloader import datasets - model = ENet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('enet_%s' % (acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_enet_citys(**kwargs): - return get_enet('citys', '', **kwargs) - - -if __name__ == '__main__': - img = torch.randn(1, 3, 512, 512) - model = get_enet_citys() - output = model(img) diff --git a/AIlib/segutils/core/models/espnet.py b/AIlib/segutils/core/models/espnet.py deleted file mode 100644 index 82651f4..0000000 --- a/AIlib/segutils/core/models/espnet.py +++ /dev/null @@ -1,134 +0,0 @@ -"ESPNetv2: A Light-weight, Power Efficient, and General Purpose for Semantic Segmentation" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.models.base_models import eespnet, EESP -from core.nn import _ConvBNPReLU, _BNPReLU - - -class ESPNetV2(nn.Module): - r"""ESPNetV2 - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - - Reference: - Sachin Mehta, et al. "ESPNetv2: A Light-weight, Power Efficient, and General Purpose Convolutional Neural Network." - arXiv preprint arXiv:1811.11431 (2018). - """ - - def __init__(self, nclass, backbone='', aux=False, jpu=False, pretrained_base=False, **kwargs): - super(ESPNetV2, self).__init__() - self.pretrained = eespnet(pretrained=pretrained_base, **kwargs) - self.proj_L4_C = _ConvBNPReLU(256, 128, 1, **kwargs) - self.pspMod = nn.Sequential( - EESP(256, 128, stride=1, k=4, r_lim=7, **kwargs), - _PSPModule(128, 128, **kwargs)) - self.project_l3 = nn.Sequential( - nn.Dropout2d(0.1), - nn.Conv2d(128, nclass, 1, bias=False)) - self.act_l3 = _BNPReLU(nclass, **kwargs) - self.project_l2 = _ConvBNPReLU(64 + nclass, nclass, 1, **kwargs) - self.project_l1 = nn.Sequential( - nn.Dropout2d(0.1), - nn.Conv2d(32 + nclass, nclass, 1, bias=False)) - - self.aux = aux - - self.__setattr__('exclusive', ['proj_L4_C', 'pspMod', 'project_l3', 'act_l3', 'project_l2', 'project_l1']) - - def forward(self, x): - size = x.size()[2:] - out_l1, out_l2, out_l3, out_l4 = self.pretrained(x, seg=True) - out_l4_proj = self.proj_L4_C(out_l4) - up_l4_to_l3 = F.interpolate(out_l4_proj, scale_factor=2, mode='bilinear', align_corners=True) - merged_l3_upl4 = self.pspMod(torch.cat([out_l3, up_l4_to_l3], 1)) - proj_merge_l3_bef_act = self.project_l3(merged_l3_upl4) - proj_merge_l3 = self.act_l3(proj_merge_l3_bef_act) - out_up_l3 = F.interpolate(proj_merge_l3, scale_factor=2, mode='bilinear', align_corners=True) - merge_l2 = self.project_l2(torch.cat([out_l2, out_up_l3], 1)) - out_up_l2 = F.interpolate(merge_l2, scale_factor=2, mode='bilinear', align_corners=True) - merge_l1 = self.project_l1(torch.cat([out_l1, out_up_l2], 1)) - - outputs = list() - merge1_l1 = F.interpolate(merge_l1, scale_factor=2, mode='bilinear', align_corners=True) - outputs.append(merge1_l1) - if self.aux: - # different from paper - auxout = F.interpolate(proj_merge_l3_bef_act, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - - #return tuple(outputs) - return outputs[0] - -# different from PSPNet -class _PSPModule(nn.Module): - def __init__(self, in_channels, out_channels=1024, sizes=(1, 2, 4, 8), **kwargs): - super(_PSPModule, self).__init__() - self.stages = nn.ModuleList( - [nn.Conv2d(in_channels, in_channels, 3, 1, 1, groups=in_channels, bias=False) for _ in sizes]) - self.project = _ConvBNPReLU(in_channels * (len(sizes) + 1), out_channels, 1, 1, **kwargs) - - def forward(self, x): - size = x.size()[2:] - feats = [x] - for stage in self.stages: - x = F.avg_pool2d(x, kernel_size=3, stride=2, padding=1) - upsampled = F.interpolate(stage(x), size, mode='bilinear', align_corners=True) - feats.append(upsampled) - return self.project(torch.cat(feats, dim=1)) - - -def get_espnet(dataset='pascal_voc', backbone='', pretrained=False, root='~/.torch/models', - pretrained_base=False, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from core.data.dataloader import datasets - model = ESPNetV2(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('espnet_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_espnet_citys(**kwargs): - return get_espnet('citys', **kwargs) - - -if __name__ == '__main__': - #model = get_espnet_citys() - input = torch.rand(2, 3, 224, 224) - model =ESPNetV2(4, pretrained_base=False) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) diff --git a/AIlib/segutils/core/models/fcn.py b/AIlib/segutils/core/models/fcn.py deleted file mode 100644 index bc54fb4..0000000 --- a/AIlib/segutils/core/models/fcn.py +++ /dev/null @@ -1,235 +0,0 @@ -import os -import torch -import torch.nn as nn -import torch.nn.functional as F -import sys -sys.path.extend(['/home/thsw2/WJ/src/yolov5/segutils/','../..','..' ]) -from core.models.base_models.vgg import vgg16 - -__all__ = ['get_fcn32s', 'get_fcn16s', 'get_fcn8s', - 'get_fcn32s_vgg16_voc', 'get_fcn16s_vgg16_voc', 'get_fcn8s_vgg16_voc'] - - -class FCN32s(nn.Module): - """There are some difference from original fcn""" - - def __init__(self, nclass, backbone='vgg16', aux=False, pretrained_base=True, - norm_layer=nn.BatchNorm2d, **kwargs): - super(FCN32s, self).__init__() - self.aux = aux - if backbone == 'vgg16': - self.pretrained = vgg16(pretrained=pretrained_base).features - else: - raise RuntimeError('unknown backbone: {}'.format(backbone)) - self.head = _FCNHead(512, nclass, norm_layer) - if aux: - self.auxlayer = _FCNHead(512, nclass, norm_layer) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - pool5 = self.pretrained(x) - - outputs = [] - out = self.head(pool5) - out = F.interpolate(out, size, mode='bilinear', align_corners=True) - outputs.append(out) - - if self.aux: - auxout = self.auxlayer(pool5) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - - return tuple(outputs) - - -class FCN16s(nn.Module): - def __init__(self, nclass, backbone='vgg16', aux=False, pretrained_base=True, norm_layer=nn.BatchNorm2d, **kwargs): - super(FCN16s, self).__init__() - self.aux = aux - if backbone == 'vgg16': - self.pretrained = vgg16(pretrained=pretrained_base).features - else: - raise RuntimeError('unknown backbone: {}'.format(backbone)) - self.pool4 = nn.Sequential(*self.pretrained[:24]) - self.pool5 = nn.Sequential(*self.pretrained[24:]) - self.head = _FCNHead(512, nclass, norm_layer) - self.score_pool4 = nn.Conv2d(512, nclass, 1) - if aux: - self.auxlayer = _FCNHead(512, nclass, norm_layer) - - self.__setattr__('exclusive', ['head', 'score_pool4', 'auxlayer'] if aux else ['head', 'score_pool4']) - - def forward(self, x): - pool4 = self.pool4(x) - pool5 = self.pool5(pool4) - - outputs = [] - score_fr = self.head(pool5) - - score_pool4 = self.score_pool4(pool4) - - upscore2 = F.interpolate(score_fr, score_pool4.size()[2:], mode='bilinear', align_corners=True) - fuse_pool4 = upscore2 + score_pool4 - - out = F.interpolate(fuse_pool4, x.size()[2:], mode='bilinear', align_corners=True) - outputs.append(out) - - if self.aux: - auxout = self.auxlayer(pool5) - auxout = F.interpolate(auxout, x.size()[2:], mode='bilinear', align_corners=True) - outputs.append(auxout) - - #return tuple(outputs) - return outputs[0] - -class FCN8s(nn.Module): - def __init__(self, nclass, backbone='vgg16', aux=False, pretrained_base=True, norm_layer=nn.BatchNorm2d, **kwargs): - super(FCN8s, self).__init__() - self.aux = aux - if backbone == 'vgg16': - self.pretrained = vgg16(pretrained=pretrained_base).features - else: - raise RuntimeError('unknown backbone: {}'.format(backbone)) - self.pool3 = nn.Sequential(*self.pretrained[:17]) - self.pool4 = nn.Sequential(*self.pretrained[17:24]) - self.pool5 = nn.Sequential(*self.pretrained[24:]) - self.head = _FCNHead(512, nclass, norm_layer) - self.score_pool3 = nn.Conv2d(256, nclass, 1) - self.score_pool4 = nn.Conv2d(512, nclass, 1) - if aux: - self.auxlayer = _FCNHead(512, nclass, norm_layer) - - self.__setattr__('exclusive', - ['head', 'score_pool3', 'score_pool4', 'auxlayer'] if aux else ['head', 'score_pool3', - 'score_pool4']) - - def forward(self, x): - pool3 = self.pool3(x) - pool4 = self.pool4(pool3) - pool5 = self.pool5(pool4) - - outputs = [] - score_fr = self.head(pool5) - - score_pool4 = self.score_pool4(pool4) - score_pool3 = self.score_pool3(pool3) - - upscore2 = F.interpolate(score_fr, score_pool4.size()[2:], mode='bilinear', align_corners=True) - fuse_pool4 = upscore2 + score_pool4 - - upscore_pool4 = F.interpolate(fuse_pool4, score_pool3.size()[2:], mode='bilinear', align_corners=True) - fuse_pool3 = upscore_pool4 + score_pool3 - - out = F.interpolate(fuse_pool3, x.size()[2:], mode='bilinear', align_corners=True) - outputs.append(out) - - if self.aux: - auxout = self.auxlayer(pool5) - auxout = F.interpolate(auxout, x.size()[2:], mode='bilinear', align_corners=True) - outputs.append(auxout) - - return tuple(outputs) - - -class _FCNHead(nn.Module): - def __init__(self, in_channels, channels, norm_layer=nn.BatchNorm2d, **kwargs): - super(_FCNHead, self).__init__() - inter_channels = in_channels // 4 - self.block = nn.Sequential( - nn.Conv2d(in_channels, inter_channels, 3, padding=1, bias=False), - norm_layer(inter_channels), - nn.ReLU(inplace=True), - nn.Dropout(0.1), - nn.Conv2d(inter_channels, channels, 1) - ) - - def forward(self, x): - return self.block(x) - - -def get_fcn32s(dataset='pascal_voc', backbone='vgg16', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = FCN32s(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('fcn32s_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_fcn16s(dataset='pascal_voc', backbone='vgg16', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = FCN16s(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('fcn16s_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_fcn8s(dataset='pascal_voc', backbone='vgg16', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = FCN8s(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('fcn8s_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_fcn32s_vgg16_voc(**kwargs): - return get_fcn32s('pascal_voc', 'vgg16', **kwargs) - - -def get_fcn16s_vgg16_voc(**kwargs): - return get_fcn16s('pascal_voc', 'vgg16', **kwargs) - - -def get_fcn8s_vgg16_voc(**kwargs): - return get_fcn8s('pascal_voc', 'vgg16', **kwargs) - - -if __name__ == "__main__": - model = FCN16s(21) - print(model) - input = torch.rand(2, 3, 224,224) - #target = torch.zeros(4, 512, 512).cuda() - #model.eval() - #print(model) - loss = model(input) - print(loss) - print(loss.shape) - import torch - from thop import profile - from torchsummary import summary - flop,params=profile(model,input_size=(1,3,512,512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop/1e9, params/1e6)) diff --git a/AIlib/segutils/core/models/fcnv2.py b/AIlib/segutils/core/models/fcnv2.py deleted file mode 100644 index 6bc4954..0000000 --- a/AIlib/segutils/core/models/fcnv2.py +++ /dev/null @@ -1,82 +0,0 @@ -"""Fully Convolutional Network with Stride of 8""" -from __future__ import division - -import torch -import torch.nn as nn -import torch.nn.functional as F - -from .segbase import SegBaseModel - -__all__ = ['FCN', 'get_fcn', 'get_fcn_resnet50_voc', - 'get_fcn_resnet101_voc', 'get_fcn_resnet152_voc'] - - -class FCN(SegBaseModel): - def __init__(self, nclass, backbone='resnet50', aux=True, pretrained_base=True, **kwargs): - super(FCN, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.head = _FCNHead(2048, nclass, **kwargs) - if aux: - self.auxlayer = _FCNHead(1024, nclass, **kwargs) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - _, _, c3, c4 = self.base_forward(x) - - outputs = [] - x = self.head(c4) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - if self.aux: - auxout = self.auxlayer(c3) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - return tuple(outputs) - - -class _FCNHead(nn.Module): - def __init__(self, in_channels, channels, norm_layer=nn.BatchNorm2d, norm_kwargs=None, **kwargs): - super(_FCNHead, self).__init__() - inter_channels = in_channels // 4 - self.block = nn.Sequential( - nn.Conv2d(in_channels, inter_channels, 3, padding=1, bias=False), - norm_layer(inter_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True), - nn.Dropout(0.1), - nn.Conv2d(inter_channels, channels, 1) - ) - - def forward(self, x): - return self.block(x) - - -def get_fcn(dataset='pascal_voc', backbone='resnet50', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = FCN(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('fcn_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_fcn_resnet50_voc(**kwargs): - return get_fcn('pascal_voc', 'resnet50', **kwargs) - - -def get_fcn_resnet101_voc(**kwargs): - return get_fcn('pascal_voc', 'resnet101', **kwargs) - - -def get_fcn_resnet152_voc(**kwargs): - return get_fcn('pascal_voc', 'resnet152', **kwargs) diff --git a/AIlib/segutils/core/models/hrnet.py b/AIlib/segutils/core/models/hrnet.py deleted file mode 100644 index 8ad08e3..0000000 --- a/AIlib/segutils/core/models/hrnet.py +++ /dev/null @@ -1,29 +0,0 @@ -"""High-Resolution Representations for Semantic Segmentation""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -class HRNet(nn.Module): - """HRNet - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - Reference: - Ke Sun. "High-Resolution Representations for Labeling Pixels and Regions." - arXiv preprint arXiv:1904.04514 (2019). - """ - def __init__(self, nclass, backbone='', aux=False, pretrained_base=False, **kwargs): - super(HRNet, self).__init__() - - def forward(self, x): - pass \ No newline at end of file diff --git a/AIlib/segutils/core/models/icnet.py b/AIlib/segutils/core/models/icnet.py deleted file mode 100644 index fed14a4..0000000 --- a/AIlib/segutils/core/models/icnet.py +++ /dev/null @@ -1,180 +0,0 @@ -"""Image Cascade Network""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.models.segbase import SegBaseModel - -__all__ = ['ICNet', 'get_icnet', 'get_icnet_resnet50_citys', - 'get_icnet_resnet101_citys', 'get_icnet_resnet152_citys'] - - -class ICNet(SegBaseModel): - """Image Cascade Network""" - - def __init__(self, nclass, backbone='resnet50', aux=False, jpu=False, pretrained_base=True, **kwargs): - super(ICNet, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.conv_sub1 = nn.Sequential( - _ConvBNReLU(3, 32, 3, 2, **kwargs), - _ConvBNReLU(32, 32, 3, 2, **kwargs), - _ConvBNReLU(32, 64, 3, 2, **kwargs) - ) - - self.ppm = PyramidPoolingModule() - - self.head = _ICHead(nclass, **kwargs) - - self.__setattr__('exclusive', ['conv_sub1', 'head']) - - def forward(self, x): - # sub 1 - x_sub1 = self.conv_sub1(x) - - # sub 2 - x_sub2 = F.interpolate(x, scale_factor=0.5, mode='bilinear', align_corners=True) - _, x_sub2, _, _ = self.base_forward(x_sub2) - - # sub 4 - x_sub4 = F.interpolate(x, scale_factor=0.25, mode='bilinear', align_corners=True) - _, _, _, x_sub4 = self.base_forward(x_sub4) - # add PyramidPoolingModule - x_sub4 = self.ppm(x_sub4) - outputs = self.head(x_sub1, x_sub2, x_sub4) - - return tuple(outputs) - -class PyramidPoolingModule(nn.Module): - def __init__(self, pyramids=[1,2,3,6]): - super(PyramidPoolingModule, self).__init__() - self.pyramids = pyramids - - def forward(self, input): - feat = input - height, width = input.shape[2:] - for bin_size in self.pyramids: - x = F.adaptive_avg_pool2d(input, output_size=bin_size) - x = F.interpolate(x, size=(height, width), mode='bilinear', align_corners=True) - feat = feat + x - return feat - -class _ICHead(nn.Module): - def __init__(self, nclass, norm_layer=nn.BatchNorm2d, **kwargs): - super(_ICHead, self).__init__() - #self.cff_12 = CascadeFeatureFusion(512, 64, 128, nclass, norm_layer, **kwargs) - self.cff_12 = CascadeFeatureFusion(128, 64, 128, nclass, norm_layer, **kwargs) - self.cff_24 = CascadeFeatureFusion(2048, 512, 128, nclass, norm_layer, **kwargs) - - self.conv_cls = nn.Conv2d(128, nclass, 1, bias=False) - - def forward(self, x_sub1, x_sub2, x_sub4): - outputs = list() - x_cff_24, x_24_cls = self.cff_24(x_sub4, x_sub2) - outputs.append(x_24_cls) - #x_cff_12, x_12_cls = self.cff_12(x_sub2, x_sub1) - x_cff_12, x_12_cls = self.cff_12(x_cff_24, x_sub1) - outputs.append(x_12_cls) - - up_x2 = F.interpolate(x_cff_12, scale_factor=2, mode='bilinear', align_corners=True) - up_x2 = self.conv_cls(up_x2) - outputs.append(up_x2) - up_x8 = F.interpolate(up_x2, scale_factor=4, mode='bilinear', align_corners=True) - outputs.append(up_x8) - # 1 -> 1/4 -> 1/8 -> 1/16 - outputs.reverse() - - return outputs - - -class _ConvBNReLU(nn.Module): - def __init__(self, in_channels, out_channels, kernel_size=3, stride=1, padding=1, dilation=1, - groups=1, norm_layer=nn.BatchNorm2d, bias=False, **kwargs): - super(_ConvBNReLU, self).__init__() - self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding, dilation, groups, bias) - self.bn = norm_layer(out_channels) - self.relu = nn.ReLU(True) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.relu(x) - return x - - -class CascadeFeatureFusion(nn.Module): - """CFF Unit""" - - def __init__(self, low_channels, high_channels, out_channels, nclass, norm_layer=nn.BatchNorm2d, **kwargs): - super(CascadeFeatureFusion, self).__init__() - self.conv_low = nn.Sequential( - nn.Conv2d(low_channels, out_channels, 3, padding=2, dilation=2, bias=False), - norm_layer(out_channels) - ) - self.conv_high = nn.Sequential( - nn.Conv2d(high_channels, out_channels, 1, bias=False), - norm_layer(out_channels) - ) - self.conv_low_cls = nn.Conv2d(out_channels, nclass, 1, bias=False) - - def forward(self, x_low, x_high): - x_low = F.interpolate(x_low, size=x_high.size()[2:], mode='bilinear', align_corners=True) - x_low = self.conv_low(x_low) - x_high = self.conv_high(x_high) - x = x_low + x_high - x = F.relu(x, inplace=True) - x_low_cls = self.conv_low_cls(x_low) - - return x, x_low_cls - - -def get_icnet(dataset='citys', backbone='resnet50', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = ICNet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('icnet_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_icnet_resnet50_citys(**kwargs): - return get_icnet('citys', 'resnet50', **kwargs) - - -def get_icnet_resnet101_citys(**kwargs): - return get_icnet('citys', 'resnet101', **kwargs) - - -def get_icnet_resnet152_citys(**kwargs): - return get_icnet('citys', 'resnet152', **kwargs) - - -if __name__ == '__main__': - # img = torch.randn(1, 3, 256, 256) - # model = get_icnet_resnet50_citys() - # outputs = model(img) - input = torch.rand(2, 3, 224, 224) - model = ICNet(4, pretrained_base=False) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - #print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) \ No newline at end of file diff --git a/AIlib/segutils/core/models/lednet.py b/AIlib/segutils/core/models/lednet.py deleted file mode 100644 index 03c05bd..0000000 --- a/AIlib/segutils/core/models/lednet.py +++ /dev/null @@ -1,211 +0,0 @@ -"""LEDNet: A Lightweight Encoder-Decoder Network for Real-time Semantic Segmentation""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.nn import _ConvBNReLU - -__all__ = ['LEDNet', 'get_lednet', 'get_lednet_citys'] - -class LEDNet(nn.Module): - r"""LEDNet - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - - Reference: - Yu Wang, et al. "LEDNet: A Lightweight Encoder-Decoder Network for Real-Time Semantic Segmentation." - arXiv preprint arXiv:1905.02423 (2019). - """ - - def __init__(self, nclass, backbone='', aux=False, jpu=False, pretrained_base=True, **kwargs): - super(LEDNet, self).__init__() - self.encoder = nn.Sequential( - Downsampling(3, 32), - SSnbt(32, **kwargs), SSnbt(32, **kwargs), SSnbt(32, **kwargs), - Downsampling(32, 64), - SSnbt(64, **kwargs), SSnbt(64, **kwargs), - Downsampling(64, 128), - SSnbt(128, **kwargs), - SSnbt(128, 2, **kwargs), - SSnbt(128, 5, **kwargs), - SSnbt(128, 9, **kwargs), - SSnbt(128, 2, **kwargs), - SSnbt(128, 5, **kwargs), - SSnbt(128, 9, **kwargs), - SSnbt(128, 17, **kwargs), - ) - self.decoder = APNModule(128, nclass) - - self.__setattr__('exclusive', ['encoder', 'decoder']) - - def forward(self, x): - size = x.size()[2:] - x = self.encoder(x) - x = self.decoder(x) - outputs = list() - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - - #return tuple(outputs) - return outputs[0] - -class Downsampling(nn.Module): - def __init__(self, in_channels, out_channels, **kwargs): - super(Downsampling, self).__init__() - self.conv1 = nn.Conv2d(in_channels, out_channels // 2, 3, 2, 2, bias=False) - self.conv2 = nn.Conv2d(in_channels, out_channels // 2, 3, 2, 2, bias=False) - self.pool = nn.MaxPool2d(kernel_size=2, stride=1) - - def forward(self, x): - x1 = self.conv1(x) - x1 = self.pool(x1) - - x2 = self.conv2(x) - x2 = self.pool(x2) - - return torch.cat([x1, x2], dim=1) - - -class SSnbt(nn.Module): - def __init__(self, in_channels, dilation=1, norm_layer=nn.BatchNorm2d, **kwargs): - super(SSnbt, self).__init__() - inter_channels = in_channels // 2 - self.branch1 = nn.Sequential( - nn.Conv2d(inter_channels, inter_channels, (3, 1), padding=(1, 0), bias=False), - nn.ReLU(True), - nn.Conv2d(inter_channels, inter_channels, (1, 3), padding=(0, 1), bias=False), - norm_layer(inter_channels), - nn.ReLU(True), - nn.Conv2d(inter_channels, inter_channels, (3, 1), padding=(dilation, 0), dilation=(dilation, 1), - bias=False), - nn.ReLU(True), - nn.Conv2d(inter_channels, inter_channels, (1, 3), padding=(0, dilation), dilation=(1, dilation), - bias=False), - norm_layer(inter_channels), - nn.ReLU(True)) - - self.branch2 = nn.Sequential( - nn.Conv2d(inter_channels, inter_channels, (1, 3), padding=(0, 1), bias=False), - nn.ReLU(True), - nn.Conv2d(inter_channels, inter_channels, (3, 1), padding=(1, 0), bias=False), - norm_layer(inter_channels), - nn.ReLU(True), - nn.Conv2d(inter_channels, inter_channels, (1, 3), padding=(0, dilation), dilation=(1, dilation), - bias=False), - nn.ReLU(True), - nn.Conv2d(inter_channels, inter_channels, (3, 1), padding=(dilation, 0), dilation=(dilation, 1), - bias=False), - norm_layer(inter_channels), - nn.ReLU(True)) - - self.relu = nn.ReLU(True) - - @staticmethod - def channel_shuffle(x, groups): - n, c, h, w = x.size() - - channels_per_group = c // groups - x = x.view(n, groups, channels_per_group, h, w) - x = torch.transpose(x, 1, 2).contiguous() - x = x.view(n, -1, h, w) - - return x - - def forward(self, x): - # channels split - x1, x2 = x.split(x.size(1) // 2, 1) - - x1 = self.branch1(x1) - x2 = self.branch2(x2) - - out = torch.cat([x1, x2], dim=1) - out = self.relu(out + x) - out = self.channel_shuffle(out, groups=2) - - return out - - -class APNModule(nn.Module): - def __init__(self, in_channels, nclass, norm_layer=nn.BatchNorm2d, **kwargs): - super(APNModule, self).__init__() - self.conv1 = _ConvBNReLU(in_channels, in_channels, 3, 2, 1, norm_layer=norm_layer) - self.conv2 = _ConvBNReLU(in_channels, in_channels, 5, 2, 2, norm_layer=norm_layer) - self.conv3 = _ConvBNReLU(in_channels, in_channels, 7, 2, 3, norm_layer=norm_layer) - self.level1 = _ConvBNReLU(in_channels, nclass, 1, norm_layer=norm_layer) - self.level2 = _ConvBNReLU(in_channels, nclass, 1, norm_layer=norm_layer) - self.level3 = _ConvBNReLU(in_channels, nclass, 1, norm_layer=norm_layer) - self.level4 = _ConvBNReLU(in_channels, nclass, 1, norm_layer=norm_layer) - self.level5 = nn.Sequential( - nn.AdaptiveAvgPool2d(1), - _ConvBNReLU(in_channels, nclass, 1)) - - def forward(self, x): - w, h = x.size()[2:] - branch3 = self.conv1(x) - branch2 = self.conv2(branch3) - branch1 = self.conv3(branch2) - - out = self.level1(branch1) - out = F.interpolate(out, ((w + 3) // 4, (h + 3) // 4), mode='bilinear', align_corners=True) - out = self.level2(branch2) + out - out = F.interpolate(out, ((w + 1) // 2, (h + 1) // 2), mode='bilinear', align_corners=True) - out = self.level3(branch3) + out - out = F.interpolate(out, (w, h), mode='bilinear', align_corners=True) - out = self.level4(x) * out - out = self.level5(x) + out - return out - - -def get_lednet(dataset='citys', backbone='', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = LEDNet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('lednet_%s' % (acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_lednet_citys(**kwargs): - return get_lednet('citys', **kwargs) - - -if __name__ == '__main__': - #model = get_lednet_citys() - input = torch.rand(2, 3, 224, 224) - model =LEDNet(4, pretrained_base=True) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) \ No newline at end of file diff --git a/AIlib/segutils/core/models/model_store.py b/AIlib/segutils/core/models/model_store.py deleted file mode 100644 index 9e64675..0000000 --- a/AIlib/segutils/core/models/model_store.py +++ /dev/null @@ -1,68 +0,0 @@ -"""Model store which provides pretrained models.""" -from __future__ import print_function - -import os -import zipfile - -from ..utils.download import download, check_sha1 - -__all__ = ['get_model_file', 'get_resnet_file'] - -_model_sha1 = {name: checksum for checksum, name in [ - ('25c4b50959ef024fcc050213a06b614899f94b3d', 'resnet50'), - ('2a57e44de9c853fa015b172309a1ee7e2d0e4e2a', 'resnet101'), - ('0d43d698c66aceaa2bc0309f55efdd7ff4b143af', 'resnet152'), -]} - -encoding_repo_url = 'https://hangzh.s3.amazonaws.com/' -_url_format = '{repo_url}encoding/models/{file_name}.zip' - - -def short_hash(name): - if name not in _model_sha1: - raise ValueError('Pretrained model for {name} is not available.'.format(name=name)) - return _model_sha1[name][:8] - - -def get_resnet_file(name, root='~/.torch/models'): - file_name = '{name}-{short_hash}'.format(name=name, short_hash=short_hash(name)) - root = os.path.expanduser(root) - - file_path = os.path.join(root, file_name + '.pth') - sha1_hash = _model_sha1[name] - if os.path.exists(file_path): - if check_sha1(file_path, sha1_hash): - return file_path - else: - print('Mismatch in the content of model file {} detected.' + - ' Downloading again.'.format(file_path)) - else: - print('Model file {} is not found. Downloading.'.format(file_path)) - - if not os.path.exists(root): - os.makedirs(root) - - zip_file_path = os.path.join(root, file_name + '.zip') - repo_url = os.environ.get('ENCODING_REPO', encoding_repo_url) - if repo_url[-1] != '/': - repo_url = repo_url + '/' - download(_url_format.format(repo_url=repo_url, file_name=file_name), - path=zip_file_path, - overwrite=True) - with zipfile.ZipFile(zip_file_path) as zf: - zf.extractall(root) - os.remove(zip_file_path) - - if check_sha1(file_path, sha1_hash): - return file_path - else: - raise ValueError('Downloaded file has different hash. Please try again.') - - -def get_model_file(name, root='~/.torch/models'): - root = os.path.expanduser(root) - file_path = os.path.join(root, name + '.pth') - if os.path.exists(file_path): - return file_path - else: - raise ValueError('Model file is not found. Downloading or trainning.') diff --git a/AIlib/segutils/core/models/model_zoo.py b/AIlib/segutils/core/models/model_zoo.py deleted file mode 100644 index 7f8cd11..0000000 --- a/AIlib/segutils/core/models/model_zoo.py +++ /dev/null @@ -1,122 +0,0 @@ -"""Model store which handles pretrained models """ -from .fcn import * -from .fcnv2 import * -from .pspnet import * -from .deeplabv3 import * -from .deeplabv3_plus import * -from .danet import * -from .denseaspp import * -from .bisenet import * -from .encnet import * -from .dunet import * -from .icnet import * -from .enet import * -from .ocnet import * -from .ccnet import * -from .psanet import * -from .cgnet import * -from .espnet import * -from .lednet import * -from .dfanet import * - -__all__ = ['get_model', 'get_model_list', 'get_segmentation_model'] - -_models = { - 'fcn32s_vgg16_voc': get_fcn32s_vgg16_voc, - 'fcn16s_vgg16_voc': get_fcn16s_vgg16_voc, - 'fcn8s_vgg16_voc': get_fcn8s_vgg16_voc, - 'fcn_resnet50_voc': get_fcn_resnet50_voc, - 'fcn_resnet101_voc': get_fcn_resnet101_voc, - 'fcn_resnet152_voc': get_fcn_resnet152_voc, - 'psp_resnet50_voc': get_psp_resnet50_voc, - 'psp_resnet50_ade': get_psp_resnet50_ade, - 'psp_resnet101_voc': get_psp_resnet101_voc, - 'psp_resnet101_ade': get_psp_resnet101_ade, - 'psp_resnet101_citys': get_psp_resnet101_citys, - 'psp_resnet101_coco': get_psp_resnet101_coco, - 'deeplabv3_resnet50_voc': get_deeplabv3_resnet50_voc, - 'deeplabv3_resnet101_voc': get_deeplabv3_resnet101_voc, - 'deeplabv3_resnet152_voc': get_deeplabv3_resnet152_voc, - 'deeplabv3_resnet50_ade': get_deeplabv3_resnet50_ade, - 'deeplabv3_resnet101_ade': get_deeplabv3_resnet101_ade, - 'deeplabv3_resnet152_ade': get_deeplabv3_resnet152_ade, - 'deeplabv3_plus_xception_voc': get_deeplabv3_plus_xception_voc, - 'danet_resnet50_ciyts': get_danet_resnet50_citys, - 'danet_resnet101_citys': get_danet_resnet101_citys, - 'danet_resnet152_citys': get_danet_resnet152_citys, - 'denseaspp_densenet121_citys': get_denseaspp_densenet121_citys, - 'denseaspp_densenet161_citys': get_denseaspp_densenet161_citys, - 'denseaspp_densenet169_citys': get_denseaspp_densenet169_citys, - 'denseaspp_densenet201_citys': get_denseaspp_densenet201_citys, - 'bisenet_resnet18_citys': get_bisenet_resnet18_citys, - 'encnet_resnet50_ade': get_encnet_resnet50_ade, - 'encnet_resnet101_ade': get_encnet_resnet101_ade, - 'encnet_resnet152_ade': get_encnet_resnet152_ade, - 'dunet_resnet50_pascal_voc': get_dunet_resnet50_pascal_voc, - 'dunet_resnet101_pascal_voc': get_dunet_resnet101_pascal_voc, - 'dunet_resnet152_pascal_voc': get_dunet_resnet152_pascal_voc, - 'icnet_resnet50_citys': get_icnet_resnet50_citys, - 'icnet_resnet101_citys': get_icnet_resnet101_citys, - 'icnet_resnet152_citys': get_icnet_resnet152_citys, - 'enet_citys': get_enet_citys, - 'base_ocnet_resnet101_citys': get_base_ocnet_resnet101_citys, - 'pyramid_ocnet_resnet101_citys': get_pyramid_ocnet_resnet101_citys, - 'asp_ocnet_resnet101_citys': get_asp_ocnet_resnet101_citys, - 'ccnet_resnet50_citys': get_ccnet_resnet50_citys, - 'ccnet_resnet101_citys': get_ccnet_resnet101_citys, - 'ccnet_resnet152_citys': get_ccnet_resnet152_citys, - 'ccnet_resnet50_ade': get_ccnet_resnet50_ade, - 'ccnet_resnet101_ade': get_ccnet_resnet101_ade, - 'ccnet_resnet152_ade': get_ccnet_resnet152_ade, - 'psanet_resnet50_voc': get_psanet_resnet50_voc, - 'psanet_resnet101_voc': get_psanet_resnet101_voc, - 'psanet_resnet152_voc': get_psanet_resnet152_voc, - 'psanet_resnet50_citys': get_psanet_resnet50_citys, - 'psanet_resnet101_citys': get_psanet_resnet101_citys, - 'psanet_resnet152_citys': get_psanet_resnet152_citys, - 'cgnet_citys': get_cgnet_citys, - 'espnet_citys': get_espnet_citys, - 'lednet_citys': get_lednet_citys, - 'dfanet_citys': get_dfanet_citys, -} - - -def get_model(name, **kwargs): - name = name.lower() - if name not in _models: - err_str = '"%s" is not among the following model list:\n\t' % (name) - err_str += '%s' % ('\n\t'.join(sorted(_models.keys()))) - raise ValueError(err_str) - net = _models[name](**kwargs) - return net - - -def get_model_list(): - return _models.keys() - - -def get_segmentation_model(model, **kwargs): - models = { - 'fcn32s': get_fcn32s, - 'fcn16s': get_fcn16s, - 'fcn8s': get_fcn8s, - 'fcn': get_fcn, - 'psp': get_psp, - 'deeplabv3': get_deeplabv3, - 'deeplabv3_plus': get_deeplabv3_plus, - 'danet': get_danet, - 'denseaspp': get_denseaspp, - 'bisenet': get_bisenet, - 'encnet': get_encnet, - 'dunet': get_dunet, - 'icnet': get_icnet, - 'enet': get_enet, - 'ocnet': get_ocnet, - 'ccnet': get_ccnet, - 'psanet': get_psanet, - 'cgnet': get_cgnet, - 'espnet': get_espnet, - 'lednet': get_lednet, - 'dfanet': get_dfanet, - } - return models[model](**kwargs) diff --git a/AIlib/segutils/core/models/ocnet.py b/AIlib/segutils/core/models/ocnet.py deleted file mode 100644 index 1e1e85c..0000000 --- a/AIlib/segutils/core/models/ocnet.py +++ /dev/null @@ -1,361 +0,0 @@ -""" Object Context Network for Scene Parsing""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.models.segbase import SegBaseModel -from core.models.fcn import _FCNHead - -__all__ = ['OCNet', 'get_ocnet', 'get_base_ocnet_resnet101_citys', - 'get_pyramid_ocnet_resnet101_citys', 'get_asp_ocnet_resnet101_citys'] - - -class OCNet(SegBaseModel): - r"""OCNet - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - Reference: - Yuhui Yuan, Jingdong Wang. "OCNet: Object Context Network for Scene Parsing." - arXiv preprint arXiv:1809.00916 (2018). - """ - - def __init__(self, nclass, backbone='resnet101', oc_arch='base', aux=False, pretrained_base=True, **kwargs): - super(OCNet, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.head = _OCHead(nclass, oc_arch, **kwargs) - if self.aux: - self.auxlayer = _FCNHead(1024, nclass, **kwargs) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - _, _, c3, c4 = self.base_forward(x) - outputs = [] - x = self.head(c4) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - - if self.aux: - auxout = self.auxlayer(c3) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - #return tuple(outputs) - return outputs[0] - -class _OCHead(nn.Module): - def __init__(self, nclass, oc_arch, norm_layer=nn.BatchNorm2d, **kwargs): - super(_OCHead, self).__init__() - if oc_arch == 'base': - self.context = nn.Sequential( - nn.Conv2d(2048, 512, 3, 1, padding=1, bias=False), - norm_layer(512), - nn.ReLU(True), - BaseOCModule(512, 512, 256, 256, scales=([1]), norm_layer=norm_layer, **kwargs)) - elif oc_arch == 'pyramid': - self.context = nn.Sequential( - nn.Conv2d(2048, 512, 3, 1, padding=1, bias=False), - norm_layer(512), - nn.ReLU(True), - PyramidOCModule(512, 512, 256, 512, scales=([1, 2, 3, 6]), norm_layer=norm_layer, **kwargs)) - elif oc_arch == 'asp': - self.context = ASPOCModule(2048, 512, 256, 512, norm_layer=norm_layer, **kwargs) - else: - raise ValueError("Unknown OC architecture!") - - self.out = nn.Conv2d(512, nclass, 1) - - def forward(self, x): - x = self.context(x) - return self.out(x) - - -class BaseAttentionBlock(nn.Module): - """The basic implementation for self-attention block/non-local block.""" - - def __init__(self, in_channels, out_channels, key_channels, value_channels, - scale=1, norm_layer=nn.BatchNorm2d, **kwargs): - super(BaseAttentionBlock, self).__init__() - self.scale = scale - self.key_channels = key_channels - self.value_channels = value_channels - if scale > 1: - self.pool = nn.MaxPool2d(scale) - - self.f_value = nn.Conv2d(in_channels, value_channels, 1) - self.f_key = nn.Sequential( - nn.Conv2d(in_channels, key_channels, 1), - norm_layer(key_channels), - nn.ReLU(True) - ) - self.f_query = self.f_key - self.W = nn.Conv2d(value_channels, out_channels, 1) - nn.init.constant_(self.W.weight, 0) - nn.init.constant_(self.W.bias, 0) - - def forward(self, x): - batch_size, c, w, h = x.size() - if self.scale > 1: - x = self.pool(x) - - value = self.f_value(x).view(batch_size, self.value_channels, -1).permute(0, 2, 1) - query = self.f_query(x).view(batch_size, self.key_channels, -1).permute(0, 2, 1) - key = self.f_key(x).view(batch_size, self.key_channels, -1) - - sim_map = torch.bmm(query, key) * (self.key_channels ** -.5) - sim_map = F.softmax(sim_map, dim=-1) - - context = torch.bmm(sim_map, value).permute(0, 2, 1).contiguous() - context = context.view(batch_size, self.value_channels, *x.size()[2:]) - context = self.W(context) - if self.scale > 1: - context = F.interpolate(context, size=(w, h), mode='bilinear', align_corners=True) - - return context - - -class BaseOCModule(nn.Module): - """Base-OC""" - - def __init__(self, in_channels, out_channels, key_channels, value_channels, - scales=([1]), norm_layer=nn.BatchNorm2d, concat=True, **kwargs): - super(BaseOCModule, self).__init__() - self.stages = nn.ModuleList([ - BaseAttentionBlock(in_channels, out_channels, key_channels, value_channels, scale, norm_layer, **kwargs) - for scale in scales]) - in_channels = in_channels * 2 if concat else in_channels - self.project = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 1), - norm_layer(out_channels), - nn.ReLU(True), - nn.Dropout2d(0.05) - ) - self.concat = concat - - def forward(self, x): - priors = [stage(x) for stage in self.stages] - context = priors[0] - for i in range(1, len(priors)): - context += priors[i] - if self.concat: - context = torch.cat([context, x], 1) - out = self.project(context) - return out - - -class PyramidAttentionBlock(nn.Module): - """The basic implementation for pyramid self-attention block/non-local block""" - - def __init__(self, in_channels, out_channels, key_channels, value_channels, - scale=1, norm_layer=nn.BatchNorm2d, **kwargs): - super(PyramidAttentionBlock, self).__init__() - self.scale = scale - self.value_channels = value_channels - self.key_channels = key_channels - - self.f_value = nn.Conv2d(in_channels, value_channels, 1) - self.f_key = nn.Sequential( - nn.Conv2d(in_channels, key_channels, 1), - norm_layer(key_channels), - nn.ReLU(True) - ) - self.f_query = self.f_key - self.W = nn.Conv2d(value_channels, out_channels, 1) - nn.init.constant_(self.W.weight, 0) - nn.init.constant_(self.W.bias, 0) - - def forward(self, x): - batch_size, c, w, h = x.size() - - local_x = list() - local_y = list() - step_w, step_h = w // self.scale, h // self.scale - for i in range(self.scale): - for j in range(self.scale): - start_x, start_y = step_w * i, step_h * j - end_x, end_y = min(start_x + step_w, w), min(start_y + step_h, h) - if i == (self.scale - 1): - end_x = w - if j == (self.scale - 1): - end_y = h - local_x += [start_x, end_x] - local_y += [start_y, end_y] - - value = self.f_value(x) - query = self.f_query(x) - key = self.f_key(x) - - local_list = list() - local_block_cnt = (self.scale ** 2) * 2 - for i in range(0, local_block_cnt, 2): - value_local = value[:, :, local_x[i]:local_x[i + 1], local_y[i]:local_y[i + 1]] - query_local = query[:, :, local_x[i]:local_x[i + 1], local_y[i]:local_y[i + 1]] - key_local = key[:, :, local_x[i]:local_x[i + 1], local_y[i]:local_y[i + 1]] - - w_local, h_local = value_local.size(2), value_local.size(3) - value_local = value_local.contiguous().view(batch_size, self.value_channels, -1).permute(0, 2, 1) - query_local = query_local.contiguous().view(batch_size, self.key_channels, -1).permute(0, 2, 1) - key_local = key_local.contiguous().view(batch_size, self.key_channels, -1) - - sim_map = torch.bmm(query_local, key_local) * (self.key_channels ** -.5) - sim_map = F.softmax(sim_map, dim=-1) - - context_local = torch.bmm(sim_map, value_local).permute(0, 2, 1).contiguous() - context_local = context_local.view(batch_size, self.value_channels, w_local, h_local) - local_list.append(context_local) - - context_list = list() - for i in range(0, self.scale): - row_tmp = list() - for j in range(self.scale): - row_tmp.append(local_list[j + i * self.scale]) - context_list.append(torch.cat(row_tmp, 3)) - - context = torch.cat(context_list, 2) - context = self.W(context) - - return context - - -class PyramidOCModule(nn.Module): - """Pyramid-OC""" - - def __init__(self, in_channels, out_channels, key_channels, value_channels, - scales=([1]), norm_layer=nn.BatchNorm2d, **kwargs): - super(PyramidOCModule, self).__init__() - self.stages = nn.ModuleList([ - PyramidAttentionBlock(in_channels, out_channels, key_channels, value_channels, scale, norm_layer, **kwargs) - for scale in scales]) - self.up_dr = nn.Sequential( - nn.Conv2d(in_channels, in_channels * len(scales), 1), - norm_layer(in_channels * len(scales)), - nn.ReLU(True) - ) - self.project = nn.Sequential( - nn.Conv2d(in_channels * len(scales) * 2, out_channels, 1), - norm_layer(out_channels), - nn.ReLU(True), - nn.Dropout2d(0.05) - ) - - def forward(self, x): - priors = [stage(x) for stage in self.stages] - context = [self.up_dr(x)] - for i in range(len(priors)): - context += [priors[i]] - context = torch.cat(context, 1) - out = self.project(context) - return out - - -class ASPOCModule(nn.Module): - """ASP-OC""" - - def __init__(self, in_channels, out_channels, key_channels, value_channels, - atrous_rates=(12, 24, 36), norm_layer=nn.BatchNorm2d, **kwargs): - super(ASPOCModule, self).__init__() - self.context = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 3, padding=1), - norm_layer(out_channels), - nn.ReLU(True), - BaseOCModule(out_channels, out_channels, key_channels, value_channels, ([2]), norm_layer, False, **kwargs)) - - rate1, rate2, rate3 = tuple(atrous_rates) - self.b1 = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 3, padding=rate1, dilation=rate1, bias=False), - norm_layer(out_channels), - nn.ReLU(True)) - self.b2 = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 3, padding=rate2, dilation=rate2, bias=False), - norm_layer(out_channels), - nn.ReLU(True)) - self.b3 = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 3, padding=rate3, dilation=rate3, bias=False), - norm_layer(out_channels), - nn.ReLU(True)) - self.b4 = nn.Sequential( - nn.Conv2d(in_channels, out_channels, 1, bias=False), - norm_layer(out_channels), - nn.ReLU(True)) - - self.project = nn.Sequential( - nn.Conv2d(out_channels * 5, out_channels, 1, bias=False), - norm_layer(out_channels), - nn.ReLU(True), - nn.Dropout2d(0.1) - ) - - def forward(self, x): - feat1 = self.context(x) - feat2 = self.b1(x) - feat3 = self.b2(x) - feat4 = self.b3(x) - feat5 = self.b4(x) - out = torch.cat((feat1, feat2, feat3, feat4, feat5), dim=1) - out = self.project(out) - return out - - -def get_ocnet(dataset='citys', backbone='resnet50', oc_arch='base', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = OCNet(datasets[dataset].NUM_CLASS, backbone=backbone, oc_arch=oc_arch, - pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('%s_ocnet_%s_%s' % ( - oc_arch, backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_base_ocnet_resnet101_citys(**kwargs): - return get_ocnet('citys', 'resnet101', 'base', **kwargs) - - -def get_pyramid_ocnet_resnet101_citys(**kwargs): - return get_ocnet('citys', 'resnet101', 'pyramid', **kwargs) - - -def get_asp_ocnet_resnet101_citys(**kwargs): - return get_ocnet('citys', 'resnet101', 'asp', **kwargs) - - -if __name__ == '__main__': - #img = torch.randn(1, 3, 256, 256) - #model = get_asp_ocnet_resnet101_citys() - # outputs = model(img) - input = torch.rand(1, 3, 224,224) - model=OCNet(4,pretrained_base=False) - #target = torch.zeros(4, 512, 512).cuda() - #model.eval() - #print(model) - loss = model(input) - print(loss,loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - flop,params=profile(model,input_size=(1,3,512,512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop/1e9, params/1e6)) \ No newline at end of file diff --git a/AIlib/segutils/core/models/psanet.py b/AIlib/segutils/core/models/psanet.py deleted file mode 100644 index 82361f3..0000000 --- a/AIlib/segutils/core/models/psanet.py +++ /dev/null @@ -1,163 +0,0 @@ -"""Point-wise Spatial Attention Network""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.nn import _ConvBNReLU -from core.models.segbase import SegBaseModel -from core.models.fcn import _FCNHead - -__all__ = ['PSANet', 'get_psanet', 'get_psanet_resnet50_voc', 'get_psanet_resnet101_voc', - 'get_psanet_resnet152_voc', 'get_psanet_resnet50_citys', 'get_psanet_resnet101_citys', - 'get_psanet_resnet152_citys'] - - -class PSANet(SegBaseModel): - r"""PSANet - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - - Reference: - Hengshuang Zhao, et al. "PSANet: Point-wise Spatial Attention Network for Scene Parsing." - ECCV-2018. - """ - - def __init__(self, nclass, backbone='resnet50', aux=False, pretrained_base=True, **kwargs): - super(PSANet, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.head = _PSAHead(nclass, **kwargs) - if aux: - self.auxlayer = _FCNHead(1024, nclass, **kwargs) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - _, _, c3, c4 = self.base_forward(x) - outputs = list() - x = self.head(c4) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - - if self.aux: - auxout = self.auxlayer(c3) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - #return tuple(outputs) - return outputs[0] - -class _PSAHead(nn.Module): - def __init__(self, nclass, norm_layer=nn.BatchNorm2d, **kwargs): - super(_PSAHead, self).__init__() - # psa_out_channels = crop_size // 8 ** 2 - self.psa = _PointwiseSpatialAttention(2048, 3600, norm_layer) - - self.conv_post = _ConvBNReLU(1024, 2048, 1, norm_layer=norm_layer) - self.project = nn.Sequential( - _ConvBNReLU(4096, 512, 3, padding=1, norm_layer=norm_layer), - nn.Dropout2d(0.1, False), - nn.Conv2d(512, nclass, 1)) - - def forward(self, x): - global_feature = self.psa(x) - out = self.conv_post(global_feature) - out = torch.cat([x, out], dim=1) - out = self.project(out) - - return out - - -class _PointwiseSpatialAttention(nn.Module):# - def __init__(self, in_channels, out_channels, norm_layer=nn.BatchNorm2d, **kwargs): - super(_PointwiseSpatialAttention, self).__init__() - reduced_channels = 512 - self.collect_attention = _AttentionGeneration(in_channels, reduced_channels, out_channels, norm_layer) - self.distribute_attention = _AttentionGeneration(in_channels, reduced_channels, out_channels, norm_layer) - - def forward(self, x): - collect_fm = self.collect_attention(x) - distribute_fm = self.distribute_attention(x) - psa_fm = torch.cat([collect_fm, distribute_fm], dim=1) - return psa_fm - - -class _AttentionGeneration(nn.Module):#-->Z:(n,C2,H,W),不是原文over-completed的做法。 - def __init__(self, in_channels, reduced_channels, out_channels, norm_layer, **kwargs): - super(_AttentionGeneration, self).__init__() - self.conv_reduce = _ConvBNReLU(in_channels, reduced_channels, 1, norm_layer=norm_layer) - self.attention = nn.Sequential( - _ConvBNReLU(reduced_channels, reduced_channels, 1, norm_layer=norm_layer), - nn.Conv2d(reduced_channels, out_channels, 1, bias=False)) - - self.reduced_channels = reduced_channels - - def forward(self, x): - reduce_x = self.conv_reduce(x) - attention = self.attention(reduce_x) - n, c, h, w = attention.size()#c=out_channels=3600, - attention = attention.view(n, c, -1)#(n,3600,H*W) - reduce_x = reduce_x.view(n, self.reduced_channels, -1)#(n,512,H*W) - print(reduce_x.shape,attention.shape) - fm = torch.bmm(reduce_x, torch.softmax(attention, dim=1)) - fm = fm.view(n, self.reduced_channels, h, w)#(n,512,60,60) - - return fm - - -def get_psanet(dataset='pascal_voc', backbone='resnet50', pretrained=False, root='~/.torch/models', - pretrained_base=False, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from core.data.dataloader import datasets - model = PSANet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('deeplabv3_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_psanet_resnet50_voc(**kwargs): - return get_psanet('pascal_voc', 'resnet50', **kwargs) - - -def get_psanet_resnet101_voc(**kwargs): - return get_psanet('pascal_voc', 'resnet101', **kwargs) - - -def get_psanet_resnet152_voc(**kwargs): - return get_psanet('pascal_voc', 'resnet152', **kwargs) - - -def get_psanet_resnet50_citys(**kwargs): - return get_psanet('citys', 'resnet50', **kwargs) - - -def get_psanet_resnet101_citys(**kwargs): - return get_psanet('citys', 'resnet101', **kwargs) - - -def get_psanet_resnet152_citys(**kwargs): - return get_psanet('citys', 'resnet152', **kwargs) - - -if __name__ == '__main__': - model = get_psanet_resnet50_voc() - img = torch.randn(1, 3, 480, 480) - output = model(img) diff --git a/AIlib/segutils/core/models/psanet_offical.py b/AIlib/segutils/core/models/psanet_offical.py deleted file mode 100644 index 54531a3..0000000 --- a/AIlib/segutils/core/models/psanet_offical.py +++ /dev/null @@ -1,255 +0,0 @@ -import torch -from torch import nn -import torch.nn.functional as F -import core.lib.psa.functional as PF -import modeling.backbone.resnet_real as models - -#运行失败，compact可以运行，但over-completed运行不了。也是跟psamask的实现有关：用到了自定义的torch.autograd.Function（里面用到了cpp文件，导入不了_C模块出错） -# -# from . import functions -# -# -# def psa_mask(input, psa_type=0, mask_H_=None, mask_W_=None): -# return functions.psa_mask(input, psa_type, mask_H_, mask_W_) -# -# -# import torch -# from torch.autograd import Function -# from .. import src - - -# class PSAMask(Function): -# @staticmethod -# def forward(ctx, input, psa_type=0, mask_H_=None, mask_W_=None): -# assert psa_type in [0, 1] # 0-col, 1-dis -# assert (mask_H_ is None and mask_W_ is None) or (mask_H_ is not None and mask_W_ is not None) -# num_, channels_, feature_H_, feature_W_ = input.size() -# if mask_H_ is None and mask_W_ is None: -# mask_H_, mask_W_ = 2 * feature_H_ - 1, 2 * feature_W_ - 1 -# assert (mask_H_ % 2 == 1) and (mask_W_ % 2 == 1) -# assert channels_ == mask_H_ * mask_W_ -# half_mask_H_, half_mask_W_ = (mask_H_ - 1) // 2, (mask_W_ - 1) // 2 -# output = torch.zeros([num_, feature_H_ * feature_W_, feature_H_, feature_W_], dtype=input.dtype, device=input.device) -# if not input.is_cuda: -# src.cpu.psamask_forward(psa_type, input, output, num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_) -# else: -# output = output.cuda() -# src.gpu.psamask_forward(psa_type, input, output, num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_) -# ctx.psa_type, ctx.num_, ctx.channels_, ctx.feature_H_, ctx.feature_W_ = psa_type, num_, channels_, feature_H_, feature_W_ -# ctx.mask_H_, ctx.mask_W_, ctx.half_mask_H_, ctx.half_mask_W_ = mask_H_, mask_W_, half_mask_H_, half_mask_W_ -# return output -# -# @staticmethod -# def backward(ctx, grad_output): -# psa_type, num_, channels_, feature_H_, feature_W_ = ctx.psa_type, ctx.num_, ctx.channels_, ctx.feature_H_, ctx.feature_W_ -# mask_H_, mask_W_, half_mask_H_, half_mask_W_ = ctx.mask_H_, ctx.mask_W_, ctx.half_mask_H_, ctx.half_mask_W_ -# grad_input = torch.zeros([num_, channels_, feature_H_, feature_W_], dtype=grad_output.dtype, device=grad_output.device) -# if not grad_output.is_cuda: -# src.cpu.psamask_backward(psa_type, grad_output, grad_input, num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_) -# else: -# src.gpu.psamask_backward(psa_type, grad_output, grad_input, num_, feature_H_, feature_W_, mask_H_, mask_W_, half_mask_H_, half_mask_W_) -# return grad_input, None, None, None - - -# psa_mask = PSAMask.apply - - -class PSA(nn.Module): - def __init__(self, in_channels=2048, mid_channels=512, psa_type=2, compact=False, shrink_factor=2, mask_h=59, - mask_w=59, normalization_factor=1.0, psa_softmax=True): - super(PSA, self).__init__() - assert psa_type in [0, 1, 2] - self.psa_type = psa_type - self.compact = compact - self.shrink_factor = shrink_factor - self.mask_h = mask_h - self.mask_w = mask_w - self.psa_softmax = psa_softmax - if normalization_factor is None: - normalization_factor = mask_h * mask_w - self.normalization_factor = normalization_factor - - self.reduce = nn.Sequential( - nn.Conv2d(in_channels, mid_channels, kernel_size=1, bias=False), - nn.BatchNorm2d(mid_channels), - nn.ReLU(inplace=True) - ) - self.attention = nn.Sequential( - nn.Conv2d(mid_channels, mid_channels, kernel_size=1, bias=False), - nn.BatchNorm2d(mid_channels), - nn.ReLU(inplace=True), - nn.Conv2d(mid_channels, mask_h*mask_w, kernel_size=1, bias=False), - ) - if psa_type == 2: - self.reduce_p = nn.Sequential( - nn.Conv2d(in_channels, mid_channels, kernel_size=1, bias=False), - nn.BatchNorm2d(mid_channels), - nn.ReLU(inplace=True) - ) - self.attention_p = nn.Sequential( - nn.Conv2d(mid_channels, mid_channels, kernel_size=1, bias=False), - nn.BatchNorm2d(mid_channels), - nn.ReLU(inplace=True), - nn.Conv2d(mid_channels, mask_h*mask_w, kernel_size=1, bias=False), - ) - self.proj = nn.Sequential( - nn.Conv2d(mid_channels * (2 if psa_type == 2 else 1), in_channels, kernel_size=1, bias=False), - nn.BatchNorm2d(in_channels), - nn.ReLU(inplace=True) - ) - - def forward(self, x): - out = x - if self.psa_type in [0, 1]: - x = self.reduce(x) - n, c, h, w = x.size() - if self.shrink_factor != 1: - h = (h - 1) // self.shrink_factor + 1#可以理解为这样做的目的是向上取整。 - w = (w - 1) // self.shrink_factor + 1 - x = F.interpolate(x, size=(h, w), mode='bilinear', align_corners=True) - y = self.attention(x) - if self.compact: - if self.psa_type == 1: - y = y.view(n, h * w, h * w).transpose(1, 2).view(n, h * w, h, w) - else: - y = PF.psa_mask(y, self.psa_type, self.mask_h, self.mask_w) - if self.psa_softmax: - y = F.softmax(y, dim=1) - x = torch.bmm(x.view(n, c, h * w), y.view(n, h * w, h * w)).view(n, c, h, w) * (1.0 / self.normalization_factor) - elif self.psa_type == 2: - x_col = self.reduce(x) - x_dis = self.reduce_p(x) - n, c, h, w = x_col.size() - if self.shrink_factor != 1: - h = (h - 1) // self.shrink_factor + 1 - w = (w - 1) // self.shrink_factor + 1 - x_col = F.interpolate(x_col, size=(h, w), mode='bilinear', align_corners=True) - x_dis = F.interpolate(x_dis, size=(h, w), mode='bilinear', align_corners=True) - y_col = self.attention(x_col) - y_dis = self.attention_p(x_dis) - if self.compact: - y_dis = y_dis.view(n, h * w, h * w).transpose(1, 2).view(n, h * w, h, w) - else: - y_col = PF.psa_mask(y_col, 0, self.mask_h, self.mask_w) - y_dis = PF.psa_mask(y_dis, 1, self.mask_h, self.mask_w) - if self.psa_softmax: - y_col = F.softmax(y_col, dim=1) - y_dis = F.softmax(y_dis, dim=1) - x_col = torch.bmm(x_col.view(n, c, h * w), y_col.view(n, h * w, h * w)).view(n, c, h, w) * (1.0 / self.normalization_factor) - x_dis = torch.bmm(x_dis.view(n, c, h * w), y_dis.view(n, h * w, h * w)).view(n, c, h, w) * (1.0 / self.normalization_factor) - x = torch.cat([x_col, x_dis], 1) - x = self.proj(x) - if self.shrink_factor != 1: - h = (h - 1) * self.shrink_factor + 1 - w = (w - 1) * self.shrink_factor + 1 - x = F.interpolate(x, size=(h, w), mode='bilinear', align_corners=True) - return torch.cat((out, x), 1) - - -class PSANet(nn.Module): - def __init__(self, layers=50, dropout=0.1, classes=2, zoom_factor=8, use_psa=True, psa_type=2, compact=False, - shrink_factor=2, mask_h=59, mask_w=59, normalization_factor=1.0, psa_softmax=True, - criterion=nn.CrossEntropyLoss(ignore_index=255), pretrained=True): - super(PSANet, self).__init__() - assert layers in [50, 101, 152] - assert classes > 1 - assert zoom_factor in [1, 2, 4, 8] - assert psa_type in [0, 1, 2] - self.zoom_factor = zoom_factor - self.use_psa = use_psa - self.criterion = criterion - - if layers == 50: - resnet = models.resnet50(pretrained=pretrained,deep_base=True) - elif layers == 101: - resnet = models.resnet101(pretrained=pretrained) - else: - resnet = models.resnet152(pretrained=pretrained) - self.layer0 = nn.Sequential(resnet.conv1, resnet.bn1, resnet.relu, resnet.conv2, resnet.bn2, resnet.relu, resnet.conv3, resnet.bn3, resnet.relu, resnet.maxpool) - self.layer1, self.layer2, self.layer3, self.layer4 = resnet.layer1, resnet.layer2, resnet.layer3, resnet.layer4 - - for n, m in self.layer3.named_modules(): - if 'conv2' in n: - m.dilation, m.padding, m.stride = (2, 2), (2, 2), (1, 1) - elif 'downsample.0' in n: - m.stride = (1, 1) - for n, m in self.layer4.named_modules(): - if 'conv2' in n: - m.dilation, m.padding, m.stride = (4, 4), (4, 4), (1, 1) - elif 'downsample.0' in n: - m.stride = (1, 1) - - fea_dim = 2048 - if use_psa: - self.psa = PSA(fea_dim, 512, psa_type, compact, shrink_factor, mask_h, mask_w, normalization_factor, psa_softmax) - fea_dim *= 2 - self.cls = nn.Sequential( - nn.Conv2d(fea_dim, 512, kernel_size=3, padding=1, bias=False), - nn.BatchNorm2d(512), - nn.ReLU(inplace=True), - nn.Dropout2d(p=dropout), - nn.Conv2d(512, classes, kernel_size=1) - ) - if self.training: - self.aux = nn.Sequential( - nn.Conv2d(1024, 256, kernel_size=3, padding=1, bias=False), - nn.BatchNorm2d(256), - nn.ReLU(inplace=True), - nn.Dropout2d(p=dropout), - nn.Conv2d(256, classes, kernel_size=1) - ) - - def forward(self, x, y=None): - x_size = x.size() - assert (x_size[2] - 1) % 8 == 0 and (x_size[3] - 1) % 8 == 0 - h = int((x_size[2] - 1) / 8 * self.zoom_factor + 1) - w = int((x_size[3] - 1) / 8 * self.zoom_factor + 1) - - x = self.layer0(x) - x = self.layer1(x) - x = self.layer2(x) - x_tmp = self.layer3(x) - x = self.layer4(x_tmp) - if self.use_psa: - x = self.psa(x) - x = self.cls(x) - if self.zoom_factor != 1: - x = F.interpolate(x, size=(h, w), mode='bilinear', align_corners=True) - - if self.training: - aux = self.aux(x_tmp) - if self.zoom_factor != 1: - aux = F.interpolate(aux, size=(h, w), mode='bilinear', align_corners=True) - main_loss = self.criterion(x, y) - aux_loss = self.criterion(aux, y) - return x.max(1)[1], main_loss, aux_loss - else: - return x - - -if __name__ == '__main__': - import os - os.environ["CUDA_VISIBLE_DEVICES"] = '0' - crop_h = crop_w = 465 - input = torch.rand(4, 3, crop_h, crop_w).cuda() - compact = False - mask_h, mask_w = None, None - shrink_factor = 2 - if compact: - mask_h = (crop_h - 1) // (8 * shrink_factor) + 1 - mask_w = (crop_w - 1) // (8 * shrink_factor) + 1 - else: - assert (mask_h is None and mask_w is None) or (mask_h is not None and mask_w is not None) - if mask_h is None and mask_w is None: - mask_h = 2 * ((crop_h - 1) // (8 * shrink_factor) + 1) - 1 - mask_w = 2 * ((crop_w - 1) // (8 * shrink_factor) + 1) - 1 - else: - assert (mask_h % 2 == 1) and (mask_h >= 3) and (mask_h <= 2 * ((crop_h - 1) // (8 * shrink_factor) + 1) - 1) - assert (mask_w % 2 == 1) and (mask_w >= 3) and (mask_w <= 2 * ((crop_h - 1) // (8 * shrink_factor) + 1) - 1) - - model = PSANet(layers=50, dropout=0.1, classes=21, zoom_factor=8, use_psa=True, psa_type=2, compact=compact, - shrink_factor=shrink_factor, mask_h=mask_h, mask_w=mask_w, psa_softmax=True, pretrained=False).cuda() - print(model) - model.eval() - output = model(input) - print('PSANet', output.size()) diff --git a/AIlib/segutils/core/models/psanet_old.py b/AIlib/segutils/core/models/psanet_old.py deleted file mode 100644 index 71a6db7..0000000 --- a/AIlib/segutils/core/models/psanet_old.py +++ /dev/null @@ -1,208 +0,0 @@ -"""Point-wise Spatial Attention Network""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.nn import CollectAttention, DistributeAttention -from core.models.segbase import SegBaseModel -from core.models.fcn import _FCNHead - - -#运行失败，name '_C' is not defined。也是跟psa_block模块的实现有关：用到了自定义的torch.autograd.Function（里面用到了cpp文件，找不到文件出错） - - -__all__ = ['PSANet', 'get_psanet', 'get_psanet_resnet50_voc', 'get_psanet_resnet101_voc', - 'get_psanet_resnet152_voc', 'get_psanet_resnet50_citys', 'get_psanet_resnet101_citys', - 'get_psanet_resnet152_citys'] - - -class PSANet(SegBaseModel): - r"""PSANet - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - - Reference: - Hengshuang Zhao, et al. "PSANet: Point-wise Spatial Attention Network for Scene Parsing." - ECCV-2018. - """ - - def __init__(self, nclass, backbone='resnet', aux=False, pretrained_base=False, **kwargs): - super(PSANet, self).__init__(nclass, aux, backbone, pretrained_base, **kwargs) - self.head = _PSAHead(nclass, **kwargs) - if aux: - self.auxlayer = _FCNHead(1024, nclass, **kwargs) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - _, _, c3, c4 = self.base_forward(x) - outputs = list() - x = self.head(c4) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - - if self.aux: - auxout = self.auxlayer(c3) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - return tuple(outputs) - - -class _PSAHead(nn.Module): - def __init__(self, nclass, norm_layer=nn.BatchNorm2d, **kwargs): - super(_PSAHead, self).__init__() - self.collect = _CollectModule(2048, 512, 60, 60, norm_layer, **kwargs) - self.distribute = _DistributeModule(2048, 512, 60, 60, norm_layer, **kwargs) - - self.conv_post = nn.Sequential( - nn.Conv2d(1024, 2048, 1, bias=False), - norm_layer(2048), - nn.ReLU(True)) - self.project = nn.Sequential( - nn.Conv2d(4096, 512, 3, padding=1, bias=False), - norm_layer(512), - nn.ReLU(True), - nn.Conv2d(512, nclass, 1) - ) - - def forward(self, x): - global_feature_collect = self.collect(x) - global_feature_distribute = self.distribute(x) - - global_feature = torch.cat([global_feature_collect, global_feature_distribute], dim=1) - out = self.conv_post(global_feature) - out = F.interpolate(out, scale_factor=2, mode='bilinear', align_corners=True) - out = torch.cat([x, out], dim=1) - out = self.project(out) - - return out - - -class _CollectModule(nn.Module): - def __init__(self, in_channels, reduced_channels, feat_w, feat_h, norm_layer, **kwargs): - super(_CollectModule, self).__init__() - self.conv_reduce = nn.Sequential( - nn.Conv2d(in_channels, reduced_channels, 1, bias=False), - norm_layer(reduced_channels), - nn.ReLU(True)) - self.conv_adaption = nn.Sequential( - nn.Conv2d(reduced_channels, reduced_channels, 1, bias=False), - norm_layer(reduced_channels), - nn.ReLU(True), - nn.Conv2d(reduced_channels, (feat_w - 1) * (feat_h), 1, bias=False)) - self.collect_attention = CollectAttention() - - self.reduced_channels = reduced_channels - self.feat_w = feat_w - self.feat_h = feat_h - - def forward(self, x): - x = self.conv_reduce(x) - # shrink - x_shrink = F.interpolate(x, scale_factor=1 / 2, mode='bilinear', align_corners=True) - x_adaption = self.conv_adaption(x_shrink) - ca = self.collect_attention(x_adaption) - global_feature_collect_list = list() - for i in range(x_shrink.shape[0]): - x_shrink_i = x_shrink[i].view(self.reduced_channels, -1) - ca_i = ca[i].view(ca.shape[1], -1) - global_feature_collect_list.append( - torch.mm(x_shrink_i, ca_i).view(1, self.reduced_channels, self.feat_h // 2, self.feat_w // 2)) - global_feature_collect = torch.cat(global_feature_collect_list) - - return global_feature_collect - - -class _DistributeModule(nn.Module): - def __init__(self, in_channels, reduced_channels, feat_w, feat_h, norm_layer, **kwargs): - super(_DistributeModule, self).__init__() - self.conv_reduce = nn.Sequential( - nn.Conv2d(in_channels, reduced_channels, 1, bias=False), - norm_layer(reduced_channels), - nn.ReLU(True)) - self.conv_adaption = nn.Sequential( - nn.Conv2d(reduced_channels, reduced_channels, 1, bias=False), - norm_layer(reduced_channels), - nn.ReLU(True), - nn.Conv2d(reduced_channels, (feat_w - 1) * (feat_h), 1, bias=False)) - self.distribute_attention = DistributeAttention() - - self.reduced_channels = reduced_channels - self.feat_w = feat_w - self.feat_h = feat_h - - def forward(self, x): - x = self.conv_reduce(x) - x_shrink = F.interpolate(x, scale_factor=1 / 2, mode='bilinear', align_corners=True) - x_adaption = self.conv_adaption(x_shrink) - da = self.distribute_attention(x_adaption) - global_feature_distribute_list = list() - for i in range(x_shrink.shape[0]): - x_shrink_i = x_shrink[i].view(self.reduced_channels, -1) - da_i = da[i].view(da.shape[1], -1) - global_feature_distribute_list.append( - torch.mm(x_shrink_i, da_i).view(1, self.reduced_channels, self.feat_h // 2, self.feat_w // 2)) - global_feature_distribute = torch.cat(global_feature_distribute_list) - - return global_feature_distribute - - -def get_psanet(dataset='pascal_voc', backbone='resnet50', pretrained=False, root='~/.torch/models', - pretrained_base=False, **kwargs): - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - # from ..data.dataloader import datasets - model = PSANet(4, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - # if pretrained: - # from .model_store import get_model_file - # device = torch.device(kwargs['local_rank']) - # model.load_state_dict(torch.load(get_model_file('deeplabv3_%s_%s' % (backbone, acronyms[dataset]), root=root), - # map_location=device)) - return model - - -def get_psanet_resnet50_voc(**kwargs): - return get_psanet('pascal_voc', 'resnet50', **kwargs) - - -def get_psanet_resnet101_voc(**kwargs): - return get_psanet('pascal_voc', 'resnet101', **kwargs) - - -def get_psanet_resnet152_voc(**kwargs): - return get_psanet('pascal_voc', 'resnet152', **kwargs) - - -def get_psanet_resnet50_citys(**kwargs): - return get_psanet('citys', 'resnet50', **kwargs) - - -def get_psanet_resnet101_citys(**kwargs): - return get_psanet('citys', 'resnet101', **kwargs) - - -def get_psanet_resnet152_citys(**kwargs): - return get_psanet('citys', 'resnet152', **kwargs) - - -if __name__ == '__main__': - model = get_psanet_resnet50_voc() - img = torch.randn(1, 3, 480, 480) - output = model(img) diff --git a/AIlib/segutils/core/models/pspnet.py b/AIlib/segutils/core/models/pspnet.py deleted file mode 100644 index 6960e57..0000000 --- a/AIlib/segutils/core/models/pspnet.py +++ /dev/null @@ -1,185 +0,0 @@ -"""Pyramid Scene Parsing Network""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from core.models.segbase import SegBaseModel -from core.models.fcn import _FCNHead - -__all__ = ['PSPNet', 'get_psp', 'get_psp_resnet50_voc', 'get_psp_resnet50_ade', 'get_psp_resnet101_voc', - 'get_psp_resnet101_ade', 'get_psp_resnet101_citys', 'get_psp_resnet101_coco'] - - -class PSPNet(SegBaseModel): - r"""Pyramid Scene Parsing Network - - Parameters - ---------- - nclass : int - Number of categories for the training dataset. - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - norm_layer : object - Normalization layer used in backbone network (default: :class:`nn.BatchNorm`; - for Synchronized Cross-GPU BachNormalization). - aux : bool - Auxiliary loss. - - Reference: - Zhao, Hengshuang, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, and Jiaya Jia. - "Pyramid scene parsing network." *CVPR*, 2017 - """ - - def __init__(self, nclass, backbone='resnet50', aux=False, pretrained_base=True, **kwargs): - super(PSPNet, self).__init__(nclass, aux, backbone, pretrained_base=pretrained_base, **kwargs) - self.head = _PSPHead(nclass, **kwargs) - if self.aux: - self.auxlayer = _FCNHead(1024, nclass, **kwargs) - - self.__setattr__('exclusive', ['head', 'auxlayer'] if aux else ['head']) - - def forward(self, x): - size = x.size()[2:] - _, _, c3, c4 = self.base_forward(x) - outputs = [] - x = self.head(c4) - x = F.interpolate(x, size, mode='bilinear', align_corners=True) - outputs.append(x) - - if self.aux: - auxout = self.auxlayer(c3) - auxout = F.interpolate(auxout, size, mode='bilinear', align_corners=True) - outputs.append(auxout) - #return tuple(outputs) - return outputs[0] - -def _PSP1x1Conv(in_channels, out_channels, norm_layer, norm_kwargs): - return nn.Sequential( - nn.Conv2d(in_channels, out_channels, 1, bias=False), - norm_layer(out_channels, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True) - ) - - -class _PyramidPooling(nn.Module): - def __init__(self, in_channels, **kwargs): - super(_PyramidPooling, self).__init__() - out_channels = int(in_channels / 4) - self.avgpool1 = nn.AdaptiveAvgPool2d(1) - self.avgpool2 = nn.AdaptiveAvgPool2d(2) - self.avgpool3 = nn.AdaptiveAvgPool2d(3) - self.avgpool4 = nn.AdaptiveAvgPool2d(6) - self.conv1 = _PSP1x1Conv(in_channels, out_channels, **kwargs) - self.conv2 = _PSP1x1Conv(in_channels, out_channels, **kwargs) - self.conv3 = _PSP1x1Conv(in_channels, out_channels, **kwargs) - self.conv4 = _PSP1x1Conv(in_channels, out_channels, **kwargs) - - def forward(self, x): - size = x.size()[2:] - feat1 = F.interpolate(self.conv1(self.avgpool1(x)), size, mode='bilinear', align_corners=True) - feat2 = F.interpolate(self.conv2(self.avgpool2(x)), size, mode='bilinear', align_corners=True) - feat3 = F.interpolate(self.conv3(self.avgpool3(x)), size, mode='bilinear', align_corners=True) - feat4 = F.interpolate(self.conv4(self.avgpool4(x)), size, mode='bilinear', align_corners=True) - return torch.cat([x, feat1, feat2, feat3, feat4], dim=1) - - -class _PSPHead(nn.Module): - def __init__(self, nclass, norm_layer=nn.BatchNorm2d, norm_kwargs=None, **kwargs): - super(_PSPHead, self).__init__() - self.psp = _PyramidPooling(2048, norm_layer=norm_layer, norm_kwargs=norm_kwargs) - self.block = nn.Sequential( - nn.Conv2d(4096, 512, 3, padding=1, bias=False), - norm_layer(512, **({} if norm_kwargs is None else norm_kwargs)), - nn.ReLU(True), - nn.Dropout(0.1), - nn.Conv2d(512, nclass, 1) - ) - - def forward(self, x): - x = self.psp(x) - return self.block(x) - - -def get_psp(dataset='pascal_voc', backbone='resnet50', pretrained=False, root='~/.torch/models', - pretrained_base=True, **kwargs): - r"""Pyramid Scene Parsing Network - - Parameters - ---------- - dataset : str, default pascal_voc - The dataset that model pretrained on. (pascal_voc, ade20k) - pretrained : bool or str - Boolean value controls whether to load the default pretrained weights for model. - String value represents the hashtag for a certain version of pretrained weights. - root : str, default '~/.torch/models' - Location for keeping the model parameters. - pretrained_base : bool or str, default True - This will load pretrained backbone network, that was trained on ImageNet. - Examples - -------- - >>> model = get_psp(dataset='pascal_voc', backbone='resnet50', pretrained=False) - >>> print(model) - """ - acronyms = { - 'pascal_voc': 'pascal_voc', - 'pascal_aug': 'pascal_aug', - 'ade20k': 'ade', - 'coco': 'coco', - 'citys': 'citys', - } - from ..data.dataloader import datasets - model = PSPNet(datasets[dataset].NUM_CLASS, backbone=backbone, pretrained_base=pretrained_base, **kwargs) - if pretrained: - from .model_store import get_model_file - device = torch.device(kwargs['local_rank']) - model.load_state_dict(torch.load(get_model_file('psp_%s_%s' % (backbone, acronyms[dataset]), root=root), - map_location=device)) - return model - - -def get_psp_resnet50_voc(**kwargs): - return get_psp('pascal_voc', 'resnet50', **kwargs) - - -def get_psp_resnet50_ade(**kwargs): - return get_psp('ade20k', 'resnet50', **kwargs) - - -def get_psp_resnet101_voc(**kwargs): - return get_psp('pascal_voc', 'resnet101', **kwargs) - - -def get_psp_resnet101_ade(**kwargs): - return get_psp('ade20k', 'resnet101', **kwargs) - - -def get_psp_resnet101_citys(**kwargs): - return get_psp('citys', 'resnet101', **kwargs) - - -def get_psp_resnet101_coco(**kwargs): - return get_psp('coco', 'resnet101', **kwargs) - - -if __name__ == '__main__': - # model = get_psp_resnet50_voc() - # img = torch.randn(4, 3, 480, 480) - # output = model(img) - input = torch.rand(2, 3, 512, 512) - model = PSPNet(4, pretrained_base=False) - # target = torch.zeros(4, 512, 512).cuda() - # model.eval() - # print(model) - loss = model(input) - print(loss, loss.shape) - - # from torchsummary import summary - # - # summary(model, (3, 224, 224)) # 打印表格，按顺序输出每层的输出形状和参数 - import torch - from thop import profile - from torchsummary import summary - - flop, params = profile(model, input_size=(1, 3, 512, 512)) - print('flops:{:.3f}G\nparams:{:.3f}M'.format(flop / 1e9, params / 1e6)) \ No newline at end of file diff --git a/AIlib/segutils/core/models/segbase.py b/AIlib/segutils/core/models/segbase.py deleted file mode 100644 index dd06266..0000000 --- a/AIlib/segutils/core/models/segbase.py +++ /dev/null @@ -1,60 +0,0 @@ -"""Base Model for Semantic Segmentation""" -import torch.nn as nn - -from ..nn import JPU -from .base_models.resnetv1b import resnet50_v1s, resnet101_v1s, resnet152_v1s - -__all__ = ['SegBaseModel'] - - -class SegBaseModel(nn.Module): - r"""Base Model for Semantic Segmentation - - Parameters - ---------- - backbone : string - Pre-trained dilated backbone network type (default:'resnet50'; 'resnet50', - 'resnet101' or 'resnet152'). - """ - - def __init__(self, nclass, aux, backbone='resnet50', jpu=False, pretrained_base=True, **kwargs): - super(SegBaseModel, self).__init__() - dilated = False if jpu else True - self.aux = aux - self.nclass = nclass - if backbone == 'resnet50': - self.pretrained = resnet50_v1s(pretrained=pretrained_base, dilated=dilated, **kwargs) - elif backbone == 'resnet101': - self.pretrained = resnet101_v1s(pretrained=pretrained_base, dilated=dilated, **kwargs) - elif backbone == 'resnet152': - self.pretrained = resnet152_v1s(pretrained=pretrained_base, dilated=dilated, **kwargs) - else: - raise RuntimeError('unknown backbone: {}'.format(backbone)) - - self.jpu = JPU([512, 1024, 2048], width=512, **kwargs) if jpu else None - - def base_forward(self, x): - """forwarding pre-trained network""" - x = self.pretrained.conv1(x) - x = self.pretrained.bn1(x) - x = self.pretrained.relu(x) - x = self.pretrained.maxpool(x) - c1 = self.pretrained.layer1(x) - c2 = self.pretrained.layer2(c1) - c3 = self.pretrained.layer3(c2) - c4 = self.pretrained.layer4(c3) - - if self.jpu: - return self.jpu(c1, c2, c3, c4) - else: - return c1, c2, c3, c4 #返回的是layer1,2,3,4的输出 - - def evaluate(self, x): - """evaluating network with inputs and targets""" - return self.forward(x)[0] - - def demo(self, x): - pred = self.forward(x) - if self.aux: - pred = pred[0] - return pred diff --git a/AIlib/segutils/core/nn.zip b/AIlib/segutils/core/nn.zip deleted file mode 100644 index eea3167..0000000 Binary files a/AIlib/segutils/core/nn.zip and /dev/null differ diff --git a/AIlib/segutils/core/nn/__init__.py b/AIlib/segutils/core/nn/__init__.py deleted file mode 100644 index 218bee9..0000000 --- a/AIlib/segutils/core/nn/__init__.py +++ /dev/null @@ -1,7 +0,0 @@ -"""Seg NN Modules""" -#from .sync_bn.syncbn import * -#from .syncbn import * -from .ca_block import * -from .psa_block import * -from .jpu import * -from .basic import * \ No newline at end of file diff --git a/AIlib/segutils/core/nn/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/nn/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index 1a7ec4d..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/__pycache__/__init__.cpython-38.pyc b/AIlib/segutils/core/nn/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index 3b1e898..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/__pycache__/basic.cpython-36.pyc b/AIlib/segutils/core/nn/__pycache__/basic.cpython-36.pyc deleted file mode 100644 index 16fd916..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/basic.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/__pycache__/basic.cpython-38.pyc b/AIlib/segutils/core/nn/__pycache__/basic.cpython-38.pyc deleted file mode 100644 index 218c506..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/basic.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/__pycache__/ca_block.cpython-36.pyc b/AIlib/segutils/core/nn/__pycache__/ca_block.cpython-36.pyc deleted file mode 100644 index 73d0d24..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/ca_block.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/__pycache__/ca_block.cpython-38.pyc b/AIlib/segutils/core/nn/__pycache__/ca_block.cpython-38.pyc deleted file mode 100644 index f3fcc8e..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/ca_block.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/__pycache__/jpu.cpython-36.pyc b/AIlib/segutils/core/nn/__pycache__/jpu.cpython-36.pyc deleted file mode 100644 index c2795d4..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/jpu.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/__pycache__/jpu.cpython-38.pyc b/AIlib/segutils/core/nn/__pycache__/jpu.cpython-38.pyc deleted file mode 100644 index 3110eb3..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/jpu.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/__pycache__/psa_block.cpython-36.pyc b/AIlib/segutils/core/nn/__pycache__/psa_block.cpython-36.pyc deleted file mode 100644 index 27c7aa8..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/psa_block.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/__pycache__/psa_block.cpython-38.pyc b/AIlib/segutils/core/nn/__pycache__/psa_block.cpython-38.pyc deleted file mode 100644 index db8b954..0000000 Binary files a/AIlib/segutils/core/nn/__pycache__/psa_block.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/basic.py b/AIlib/segutils/core/nn/basic.py deleted file mode 100644 index 3b5a186..0000000 --- a/AIlib/segutils/core/nn/basic.py +++ /dev/null @@ -1,134 +0,0 @@ -"""Basic Module for Semantic Segmentation""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -__all__ = ['_ConvBNPReLU', '_ConvBN', '_BNPReLU', '_ConvBNReLU', '_DepthwiseConv', 'InvertedResidual'] - - -class _ConvBNReLU(nn.Module): - def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding=0, - dilation=1, groups=1, relu6=False, norm_layer=nn.BatchNorm2d, **kwargs): - super(_ConvBNReLU, self).__init__() - self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding, dilation, groups, bias=False) - self.bn = norm_layer(out_channels) - self.relu = nn.ReLU6(False) if relu6 else nn.ReLU(False) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.relu(x) - return x - - -class _ConvBNPReLU(nn.Module): - def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding=0, - dilation=1, groups=1, norm_layer=nn.BatchNorm2d, **kwargs): - super(_ConvBNPReLU, self).__init__() - self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding, dilation, groups, bias=False) - self.bn = norm_layer(out_channels) - self.prelu = nn.PReLU(out_channels) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.prelu(x) - return x - - -class _ConvBN(nn.Module): - def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding=0, - dilation=1, groups=1, norm_layer=nn.BatchNorm2d, **kwargs): - super(_ConvBN, self).__init__() - self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding, dilation, groups, bias=False) - self.bn = norm_layer(out_channels) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - return x - - -class _BNPReLU(nn.Module): - def __init__(self, out_channels, norm_layer=nn.BatchNorm2d, **kwargs): - super(_BNPReLU, self).__init__() - self.bn = norm_layer(out_channels) - self.prelu = nn.PReLU(out_channels) - - def forward(self, x): - x = self.bn(x) - x = self.prelu(x) - return x - - -# ----------------------------------------------------------------- -# For PSPNet -# ----------------------------------------------------------------- -class _PSPModule(nn.Module): - def __init__(self, in_channels, sizes=(1, 2, 3, 6), **kwargs): - super(_PSPModule, self).__init__() - out_channels = int(in_channels / 4) - self.avgpools = nn.ModuleList() - self.convs = nn.ModuleList() - for size in sizes: - self.avgpool.append(nn.AdaptiveAvgPool2d(size)) - self.convs.append(_ConvBNReLU(in_channels, out_channels, 1, **kwargs)) - - def forward(self, x): - size = x.size()[2:] - feats = [x] - for (avgpool, conv) in enumerate(zip(self.avgpools, self.convs)): - feats.append(F.interpolate(conv(avgpool(x)), size, mode='bilinear', align_corners=True)) - return torch.cat(feats, dim=1) - - -# ----------------------------------------------------------------- -# For MobileNet -# ----------------------------------------------------------------- -class _DepthwiseConv(nn.Module): - """conv_dw in MobileNet""" - - def __init__(self, in_channels, out_channels, stride, norm_layer=nn.BatchNorm2d, **kwargs): - super(_DepthwiseConv, self).__init__() - self.conv = nn.Sequential( - _ConvBNReLU(in_channels, in_channels, 3, stride, 1, groups=in_channels, norm_layer=norm_layer), - _ConvBNReLU(in_channels, out_channels, 1, norm_layer=norm_layer)) - - def forward(self, x): - return self.conv(x) - - -# ----------------------------------------------------------------- -# For MobileNetV2 -# ----------------------------------------------------------------- -class InvertedResidual(nn.Module): - def __init__(self, in_channels, out_channels, stride, expand_ratio, norm_layer=nn.BatchNorm2d, **kwargs): - super(InvertedResidual, self).__init__() - assert stride in [1, 2] - self.use_res_connect = stride == 1 and in_channels == out_channels - - layers = list() - inter_channels = int(round(in_channels * expand_ratio)) - if expand_ratio != 1: - # pw - layers.append(_ConvBNReLU(in_channels, inter_channels, 1, relu6=True, norm_layer=norm_layer)) - layers.extend([ - # dw - _ConvBNReLU(inter_channels, inter_channels, 3, stride, 1, - groups=inter_channels, relu6=True, norm_layer=norm_layer), - # pw-linear - nn.Conv2d(inter_channels, out_channels, 1, bias=False), - norm_layer(out_channels)]) - self.conv = nn.Sequential(*layers) - - def forward(self, x): - if self.use_res_connect: - return x + self.conv(x) - else: - return self.conv(x) - - -if __name__ == '__main__': - x = torch.randn(1, 32, 64, 64) - model = InvertedResidual(32, 64, 2, 1) - out = model(x) diff --git a/AIlib/segutils/core/nn/ca_block.py b/AIlib/segutils/core/nn/ca_block.py deleted file mode 100644 index 954c293..0000000 --- a/AIlib/segutils/core/nn/ca_block.py +++ /dev/null @@ -1,72 +0,0 @@ -import torch -import torch.nn as nn -import torch.nn.functional as F - -from torch.autograd.function import once_differentiable -#from core.nn import _C - -__all__ = ['CrissCrossAttention', 'ca_weight', 'ca_map'] - - -class _CAWeight(torch.autograd.Function): - @staticmethod - def forward(ctx, t, f): - weight = _C.ca_forward(t, f) - - ctx.save_for_backward(t, f) - - return weight - - @staticmethod - @once_differentiable - def backward(ctx, dw): - t, f = ctx.saved_tensors - - dt, df = _C.ca_backward(dw, t, f) - return dt, df - - -class _CAMap(torch.autograd.Function): - @staticmethod - def forward(ctx, weight, g): - out = _C.ca_map_forward(weight, g) - - ctx.save_for_backward(weight, g) - - return out - - @staticmethod - @once_differentiable - def backward(ctx, dout): - weight, g = ctx.saved_tensors - - dw, dg = _C.ca_map_backward(dout, weight, g) - - return dw, dg - - -ca_weight = _CAWeight.apply -ca_map = _CAMap.apply - - -class CrissCrossAttention(nn.Module): - """Criss-Cross Attention Module""" - - def __init__(self, in_channels): - super(CrissCrossAttention, self).__init__() - self.query_conv = nn.Conv2d(in_channels, in_channels // 8, 1) - self.key_conv = nn.Conv2d(in_channels, in_channels // 8, 1) - self.value_conv = nn.Conv2d(in_channels, in_channels, 1) - self.gamma = nn.Parameter(torch.zeros(1)) - - def forward(self, x): - proj_query = self.query_conv(x) - proj_key = self.key_conv(x) - proj_value = self.value_conv(x) - - energy = ca_weight(proj_query, proj_key) - attention = F.softmax(energy, 1) - out = ca_map(attention, proj_value) - out = self.gamma * out + x - - return out diff --git a/AIlib/segutils/core/nn/csrc/ca.h b/AIlib/segutils/core/nn/csrc/ca.h deleted file mode 100644 index 1a93b36..0000000 --- a/AIlib/segutils/core/nn/csrc/ca.h +++ /dev/null @@ -1,58 +0,0 @@ -#pragma once - -#include "cpu/vision.h" - -#ifdef WITH_CUDA -#include "cuda/vision.h" -#endif - -// Interface for Python -at::Tensor ca_forward(const at::Tensor& t, - const at::Tensor& f) { - if (t.type().is_cuda()) { - #ifdef WITH_CUDA - return ca_forward_cuda(t, f); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - return ca_forward_cpu(t, f); -} - -std::tuple ca_backward(const at::Tensor& dw, - const at::Tensor& t, - const at::Tensor& f) { - if (dw.type().is_cuda()) { - #ifdef WITH_CUDA - return ca_backward_cuda(dw, t, f); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - return ca_backward_cpu(dw, t, f); -} - -at::Tensor ca_map_forward(const at::Tensor& weight, - const at::Tensor& g) { - if (weight.type().is_cuda()) { - #ifdef WITH_CUDA - return ca_map_forward_cuda(weight, g); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - return ca_map_forward_cpu(weight, g); -} - -std::tuple ca_map_backward(const at::Tensor& dout, - const at::Tensor& weight, - const at::Tensor& g) { - if (dout.type().is_cuda()) { - #ifdef WITH_CUDA - return ca_map_backward_cuda(dout, weight, g); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - return ca_map_backward_cpu(dout, weight, g); -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/cpu/ca_cpu.cpp b/AIlib/segutils/core/nn/csrc/cpu/ca_cpu.cpp deleted file mode 100644 index 6029c51..0000000 --- a/AIlib/segutils/core/nn/csrc/cpu/ca_cpu.cpp +++ /dev/null @@ -1,24 +0,0 @@ -#include "cpu/vision.h" - - -at::Tensor ca_forward_cpu( - const torch::Tensor& t, - const torch::Tensor& f) { - AT_ERROR("Not implemented on the CPU");} - -std::tuple ca_backward_cpu( - const at::Tensor& dw, - const at::Tensor& t, - const at::Tensor& f) { - AT_ERROR("Not implemented on the CPU");} - -at::Tensor ca_map_forward_cpu( - const at::Tensor& weight, - const at::Tensor& g) { - AT_ERROR("Not implemented on the CPU");} - -std::tuple ca_map_backward_cpu( - const at::Tensor& dout, - const at::Tensor& weight, - const at::Tensor& g) { - AT_ERROR("Not implemented on the CPU");} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/cpu/psa_cpu.cpp b/AIlib/segutils/core/nn/csrc/cpu/psa_cpu.cpp deleted file mode 100644 index 9e0e765..0000000 --- a/AIlib/segutils/core/nn/csrc/cpu/psa_cpu.cpp +++ /dev/null @@ -1,13 +0,0 @@ -#include "cpu/vision.h" - - -at::Tensor psa_forward_cpu( - const torch::Tensor& hc, - const int forward_type) { - AT_ERROR("Not implemented on the CPU");} - -at::Tensor psa_backward_cpu( - const at::Tensor& dout, - const at::Tensor& hc, - const int forward_type) { - AT_ERROR("Not implemented on the CPU");} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/cpu/syncbn_cpu.cpp b/AIlib/segutils/core/nn/csrc/cpu/syncbn_cpu.cpp deleted file mode 100644 index 70b5db4..0000000 --- a/AIlib/segutils/core/nn/csrc/cpu/syncbn_cpu.cpp +++ /dev/null @@ -1,45 +0,0 @@ -#include -#include -#include - -at::Tensor broadcast_to(at::Tensor v, at::Tensor x) { - if (x.ndimension() == 2) { - return v; - } else { - std::vector broadcast_size = {1, -1}; - for (int64_t i = 2; i < x.ndimension(); ++i) - broadcast_size.push_back(1); - - return v.view(broadcast_size); - } -} - -at::Tensor batchnorm_forward_cpu( - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - auto output = (input_ - broadcast_to(ex_, input_)) / broadcast_to(exs_, input_); - output = output * broadcast_to(gamma_, input_) + broadcast_to(beta_, input_); - return output; -} - -// Not implementing CPU backward for now -std::vector batchnorm_backward_cpu( - const at::Tensor gradoutput_, - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - /* outputs*/ - at::Tensor gradinput = at::zeros_like(input_); - at::Tensor gradgamma = at::zeros_like(gamma_); - at::Tensor gradbeta = at::zeros_like(beta_); - at::Tensor gradMean = at::zeros_like(ex_); - at::Tensor gradStd = at::zeros_like(exs_); - return {gradinput, gradMean, gradStd, gradgamma, gradbeta}; -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/cpu/vision.h b/AIlib/segutils/core/nn/csrc/cpu/vision.h deleted file mode 100644 index 8a824fe..0000000 --- a/AIlib/segutils/core/nn/csrc/cpu/vision.h +++ /dev/null @@ -1,47 +0,0 @@ -#pragma once -#include - - -at::Tensor ca_forward_cpu( - const at::Tensor& t, - const at::Tensor& f); - -std::tuple ca_backward_cpu( - const at::Tensor& dw, - const at::Tensor& t, - const at::Tensor& f); - -at::Tensor ca_map_forward_cpu( - const at::Tensor& weight, - const at::Tensor& g); - -std::tuple ca_map_backward_cpu( - const at::Tensor& dout, - const at::Tensor& weight, - const at::Tensor& g); - -at::Tensor psa_forward_cpu( - const at::Tensor& hc, - const int forward_type); - -at::Tensor psa_backward_cpu( - const at::Tensor& dout, - const at::Tensor& hc, - const int forward_type); - -at::Tensor batchnorm_forward_cpu( - const at::Tensor input_, - const at::Tensor mean_, - const at::Tensor std_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -std::vector batchnorm_backward_cpu( - const at::Tensor gradoutput_, - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/cuda/ca_cuda.cu b/AIlib/segutils/core/nn/csrc/cuda/ca_cuda.cu deleted file mode 100644 index ba459fa..0000000 --- a/AIlib/segutils/core/nn/csrc/cuda/ca_cuda.cu +++ /dev/null @@ -1,324 +0,0 @@ -#include -#include - -#include -#include -#include - -template -__global__ void ca_forward_kernel(const T *t, const T *f, T *weight, int num, int chn, int height, int width) { - int x = blockIdx.x * blockDim.x + threadIdx.x; - int y = blockIdx.y * blockDim.y + threadIdx.y; - int sp = height * width; - int len = height + width - 1; - int z = blockIdx.z; - - if (x < width && y < height && z < height+width-1) { - for (int batch = 0; batch < num; ++batch) { - for (int plane = 0; plane < chn; ++plane) { - T _t = t[(batch * chn + plane) * sp + y * width + x]; - - if (z < width) { - int i = z; - T _f = f[(batch * chn + plane) * sp + y * width + i]; - weight[(batch * len + i) * sp + y*width + x] += _t*_f; - } - else { - int i = z - width; - int j = i -__global__ void ca_backward_kernel_t(const T *dw, const T *t, const T *f, T *dt, - int num, int chn, int height, int width) { - int x = blockIdx.x * blockDim.x + threadIdx.x; - int y = blockIdx.y * blockDim.y + threadIdx.y; - int sp = height * width; - int len = height + width - 1; - int plane = blockIdx.z; - - if (x < width && y < height && plane < chn) { - for (int batch = 0; batch < num; ++batch) { - for (int i = 0; i < width; ++i) { - T _dw = dw[(batch * len + i) * sp + y*width + x]; - T _f = f[(batch * chn + plane) * sp + y*width + i]; - dt[(batch * chn + plane) * sp + y*width + x] += _dw * _f; - } - for (int i = 0; i < height; ++i) { - if (i == y) continue; - int j = i -__global__ void ca_backward_kernel_f(const T *dw, const T *t, const T *f, T *df, - int num, int chn, int height, int width) { - int x = blockIdx.x * blockDim.x + threadIdx.x; - int y = blockIdx.y * blockDim.y + threadIdx.y; - int sp = height * width; - int len = height + width - 1; - int plane = blockIdx.z; - - if (x < width && y < height && plane < chn) { - for (int batch = 0; batch < num; ++batch) { - for (int i = 0; i < width; ++i) { - T _dw = dw[(batch * len + x) * sp + y*width + i]; - T _t = t[(batch * chn + plane) * sp + y*width + i]; - df[(batch * chn + plane) * sp + y*width + x] += _dw * _t; - } - for (int i = 0; i < height; ++i) { - if (i == y) continue; - int j = i>y ? y : y-1; - - T _dw = dw[(batch * len + width + j) * sp + i*width + x]; - T _t = t[(batch * chn + plane) * sp + i*width + x]; - df[(batch * chn + plane) * sp + y*width + x] += _dw * _t; - } - } - } -} - -template -__global__ void ca_map_forward_kernel(const T *weight, const T *g, T *out, int num, int chn, int height, int width) { - int x = blockIdx.x * blockDim.x + threadIdx.x; - int y = blockIdx.y * blockDim.y + threadIdx.y; - int sp = height * width; - int len = height + width - 1; - int plane = blockIdx.z; - - if (x < width && y < height && plane < chn) { - for (int batch = 0; batch < num; ++batch) { - for (int i = 0; i < width; ++i) { - T _g = g[(batch * chn + plane) * sp + y*width + i]; - T _w = weight[(batch * len + i) * sp + y*width + x]; - out[(batch * chn + plane) * sp + y*width + x] += _g * _w; - } - for (int i = 0; i < height; ++i) { - if (i == y) continue; - - int j = i -__global__ void ca_map_backward_kernel_w(const T *dout, const T *weight, const T *g, T *dw, int num, int chn, int height, int width) { - int x = blockIdx.x * blockDim.x + threadIdx.x; - int y = blockIdx.y * blockDim.y + threadIdx.y; - int sp = height * width; - int len = height + width - 1; - int z = blockIdx.z; - - if (x < width && y < height && z < height+width-1) { - for (int batch = 0; batch < num; ++batch) { - for (int plane = 0; plane < chn; ++plane) { - T _dout = dout[(batch * chn + plane) * sp + y*width + x]; - - if (z < width) { - int i = z; - T _g = g[(batch * chn + plane) * sp + y*width + i]; - dw[(batch * len + i) * sp + y*width + x] += _dout * _g; - } - else { - int i = z - width; - int j = i -__global__ void ca_map_backward_kernel_g(const T *dout, const T *weight, const T *g, T *dg, int num, int chn, int height, int width) { - int x = blockIdx.x * blockDim.x + threadIdx.x; - int y = blockIdx.y * blockDim.y + threadIdx.y; - int sp = height * width; - int len = height + width - 1; - int plane = blockIdx.z; - - if (x < width && y < height && plane < chn) { - for (int batch = 0; batch < num; ++batch) { - for (int i = 0; i < width; ++i) { - T _dout = dout[(batch * chn + plane) * sp + y*width + i]; - T _w = weight[(batch * len + x) * sp + y*width + i]; - dg[(batch * chn + plane) * sp + y*width + x] += _dout * _w; - } - for (int i = 0; i < height; ++i) { - if (i == y) continue; - int j = i>y ? y : y-1; - - T _dout = dout[(batch * chn + plane) * sp + i*width + x]; - T _w = weight[(batch * len + width + j) * sp + i*width + x]; - dg[(batch * chn + plane) * sp + y*width + x] += _dout * _w; - } - } - } -} - -/* - * Implementations - */ -at::Tensor ca_forward_cuda(const at::Tensor& t, const at::Tensor& f) { - AT_ASSERTM(t.type().is_cuda(), "input must be a CUDA tensor"); - AT_ASSERTM(f.type().is_cuda(), "input must be a CUDA tensor"); - - auto n = t.size(0); - auto c = t.size(1); - auto h = t.size(2); - auto w = t.size(3); - - at::Tensor weight = at::zeros({n, h + w - 1, h, w}, t.options()); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - - // Run kernel - dim3 threads(32, 32); - int d1 = (w + threads.x - 1) / threads.x; - int d2 = (h + threads.y - 1) / threads.y; - int d3 = h + w; - dim3 blocks(d1, d2, d3); - - AT_DISPATCH_FLOATING_TYPES(t.type(), "ca_forward", [&] { - ca_forward_kernel<<>>( - t.contiguous().data(), - f.contiguous().data(), - weight.contiguous().data(), - n, c, h, w); - }); - THCudaCheck(cudaGetLastError()); - return weight; -} - -std::tuple ca_backward_cuda(const at::Tensor& dw, const at::Tensor& t, const at::Tensor& f) { - AT_ASSERTM(dw.type().is_cuda(), "input must be a CUDA tensor"); - AT_ASSERTM(t.type().is_cuda(), "input must be a CUDA tensor"); - AT_ASSERTM(f.type().is_cuda(), "input must be a CUDA tensor"); - - auto n = t.size(0); - auto c = t.size(1); - auto h = t.size(2); - auto w = t.size(3); - - at::Tensor dt = at::zeros_like(t); - at::Tensor df = at::zeros_like(f); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - - // Run kernel - dim3 threads(32, 32); - int d1 = (w + threads.x - 1) / threads.x; - int d2 = (h + threads.y - 1) / threads.y; - int d3 = c; - dim3 blocks(d1, d2, d3); - - AT_DISPATCH_FLOATING_TYPES(t.type(), "ca_backward_kernel_t", [&] { - ca_backward_kernel_t<<>> ( - dw.contiguous().data(), - t.contiguous().data(), - f.contiguous().data(), - dt.contiguous().data(), - n, c, h, w); - }); - - AT_DISPATCH_FLOATING_TYPES(f.type(), "ca_backward_kernel_f", [&] { - ca_backward_kernel_f<<>> ( - dw.contiguous().data(), - t.contiguous().data(), - f.contiguous().data(), - df.contiguous().data(), - n, c, h, w); - }); - THCudaCheck(cudaGetLastError()); - return std::make_tuple(dt, df); -} - -at::Tensor ca_map_forward_cuda(const at::Tensor& weight, const at::Tensor& g) { - AT_ASSERTM(weight.type().is_cuda(), "input must be a CUDA tensor"); - AT_ASSERTM(g.type().is_cuda(), "input must be a CUDA tensor"); - - auto n = g.size(0); - auto c = g.size(1); - auto h = g.size(2); - auto w = g.size(3); - - at::Tensor out = at::zeros_like(g); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - - // Run kernel - dim3 threads(32, 32); - int d1 = (w + threads.x - 1) / threads.x; - int d2 = (h + threads.y - 1) / threads.y; - int d3 = c; - dim3 blocks(d1, d2, d3); - - AT_DISPATCH_FLOATING_TYPES(g.type(), "ca_map_forward", [&] { - ca_map_forward_kernel<<>>( - weight.contiguous().data(), - g.contiguous().data(), - out.contiguous().data(), - n, c, h, w); - }); - THCudaCheck(cudaGetLastError()); - return out; -} - -std::tuple ca_map_backward_cuda(const at::Tensor& dout, const at::Tensor& weight, const at::Tensor& g) { - AT_ASSERTM(dout.type().is_cuda(), "input must be a CUDA tensor"); - AT_ASSERTM(weight.type().is_cuda(), "input must be a CUDA tensor"); - AT_ASSERTM(g.type().is_cuda(), "input must be a CUDA tensor"); - - auto n = dout.size(0); - auto c = dout.size(1); - auto h = dout.size(2); - auto w = dout.size(3); - - at::Tensor dw = at::zeros_like(weight); - at::Tensor dg = at::zeros_like(g); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - - // Run kernel - dim3 threads(32, 32); - int d1 = (w + threads.x - 1) / threads.x; - int d2 = (h + threads.y - 1) / threads.y; - int d3 = h + w; - dim3 blocks(d1, d2, d3); - - AT_DISPATCH_FLOATING_TYPES(weight.type(), "ca_map_backward_kernel_w", [&] { - ca_map_backward_kernel_w<<>> ( - dout.contiguous().data(), - weight.contiguous().data(), - g.contiguous().data(), - dw.contiguous().data(), - n, c, h, w); - }); - - AT_DISPATCH_FLOATING_TYPES(g.type(), "ca_map_backward_kernel_g", [&] { - ca_map_backward_kernel_g<<>> ( - dout.contiguous().data(), - weight.contiguous().data(), - g.contiguous().data(), - dg.contiguous().data(), - n, c, h, w); - }); - THCudaCheck(cudaGetLastError()); - return std::make_tuple(dw, dg); -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/cuda/helper.h b/AIlib/segutils/core/nn/csrc/cuda/helper.h deleted file mode 100644 index cc5ea88..0000000 --- a/AIlib/segutils/core/nn/csrc/cuda/helper.h +++ /dev/null @@ -1,334 +0,0 @@ -#include -#include -#include - -static const unsigned WARP_SIZE = 32; - -// The maximum number of threads in a block -static const unsigned MAX_BLOCK_SIZE = 512U; - -template -struct ScalarConvert { - static __host__ __device__ __forceinline__ Out to(const In v) { return (Out) v; } -}; - -// Number of threads in a block given an input size up to MAX_BLOCK_SIZE -static int getNumThreads(int nElem) { - int threadSizes[5] = { 32, 64, 128, 256, MAX_BLOCK_SIZE }; - for (int i = 0; i != 5; ++i) { - if (nElem <= threadSizes[i]) { - return threadSizes[i]; - } - } - return MAX_BLOCK_SIZE; -} - -// Returns the index of the most significant 1 bit in `val`. -__device__ __forceinline__ int getMSB(int val) { - return 31 - __clz(val); -} - -template -__device__ __forceinline__ T WARP_SHFL_XOR(T value, int laneMask, int width = warpSize, unsigned int mask = 0xffffffff) -{ -#if CUDA_VERSION >= 9000 - return __shfl_xor_sync(mask, value, laneMask, width); -#else - return __shfl_xor(value, laneMask, width); -#endif -} - -// Sum across all threads within a warp -template -static __device__ __forceinline__ T warpSum(T val) { -#if __CUDA_ARCH__ >= 300 - for (int i = 0; i < getMSB(WARP_SIZE); ++i) { - val += WARP_SHFL_XOR(val, 1 << i, WARP_SIZE); - } -#else - __shared__ T values[MAX_BLOCK_SIZE]; - values[threadIdx.x] = val; - __threadfence_block(); - const int base = (threadIdx.x / WARP_SIZE) * WARP_SIZE; - for (int i = 1; i < WARP_SIZE; i++) { - val += values[base + ((i + threadIdx.x) % WARP_SIZE)]; - } -#endif - return val; -} - -template -struct Float2 { - Acctype v1, v2; - __device__ Float2() {} - __device__ Float2(DType v1, DType v2) : v1(ScalarConvert::to(v1)), v2(ScalarConvert::to(v2)) {} - __device__ Float2(DType v) : v1(ScalarConvert::to(v)), v2(ScalarConvert::to(v)) {} - __device__ Float2(int v) : v1(ScalarConvert::to(v)), v2(ScalarConvert::to(v)) {} - __device__ Float2& operator+=(const Float2& a) { - v1 += a.v1; - v2 += a.v2; - return *this; - } -}; - -template -static __device__ __forceinline__ Float2 warpSum(Float2 value) { - value.v1 = warpSum(value.v1); - value.v2 = warpSum(value.v2); - return value; -} - -template -__device__ T reduceD( - Op op, int b, int i, int k, int D) { - T sum = 0; - for (int x = threadIdx.x; x < D; x += blockDim.x) { - sum += op(b,i,k,x); - } - // sum over NumThreads within a warp - sum = warpSum(sum); - - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - if (threadIdx.x / WARP_SIZE < 32) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T) 0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} - -template -__device__ T reduceN( - Op op, int b, int k, int d, int N) { - T sum = 0; - for (int x = threadIdx.x; x < N; x += blockDim.x) { - sum += op(b,x,k,d); - } - // sum over NumThreads within a warp - sum = warpSum(sum); - - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - if (threadIdx.x / WARP_SIZE < 32) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T) 0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} - -template -__device__ T reduceK( - Op op, int b, int i, int d, int K) { - T sum = 0; - for (int x = threadIdx.x; x < K; x += blockDim.x) { - sum += op(b,i,x,d); - } - // sum over NumThreads within a warp - sum = warpSum(sum); - - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - if (threadIdx.x / WARP_SIZE < 32) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T) 0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} - -template -__device__ T reduceBN( - Op op, - int k, int d, int B, int N) { - T sum = 0; - for (int batch = 0; batch < B; ++batch) { - for (int x = threadIdx.x; x < N; x += blockDim.x) { - sum += op(batch,x,k,d); - } - } - // sum over NumThreads within a warp - sum = warpSum(sum); - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - if (threadIdx.x / WARP_SIZE < 32) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T) 0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} - -template -struct DeviceTensor { - public: - inline __device__ __host__ DeviceTensor(DType *p, const int *size) - : dptr_(p) { - for (int i = 0; i < Dim; ++i) { - size_[i] = size ? size[i] : 0; - } - } - - inline __device__ __host__ unsigned getSize(const int i) const { - assert(i < Dim); - return size_[i]; - } - - inline __device__ __host__ int numElements() const { - int n = 1; - for (int i = 0; i < Dim; ++i) { - n *= size_[i]; - } - return n; - } - - inline __device__ __host__ DeviceTensor select(const size_t x) const { - assert(Dim > 1); - int offset = x; - for (int i = 1; i < Dim; ++i) { - offset *= size_[i]; - } - DeviceTensor tensor(dptr_ + offset, nullptr); - for (int i = 0; i < Dim - 1; ++i) { - tensor.size_[i] = this->size_[i+1]; - } - return tensor; - } - - inline __device__ __host__ DeviceTensor operator[](const size_t x) const { - assert(Dim > 1); - int offset = x; - for (int i = 1; i < Dim; ++i) { - offset *= size_[i]; - } - DeviceTensor tensor(dptr_ + offset, nullptr); - for (int i = 0; i < Dim - 1; ++i) { - tensor.size_[i] = this->size_[i+1]; - } - return tensor; - } - - inline __device__ __host__ size_t InnerSize() const { - assert(Dim >= 3); - size_t sz = 1; - for (size_t i = 2; i < Dim; ++i) { - sz *= size_[i]; - } - return sz; - } - - inline __device__ __host__ size_t ChannelCount() const { - assert(Dim >= 3); - return size_[1]; - } - - inline __device__ __host__ DType* data_ptr() const { - return dptr_; - } - - DType *dptr_; - int size_[Dim]; -}; - -template -struct DeviceTensor { - inline __device__ __host__ DeviceTensor(DType *p, const int *size) - : dptr_(p) { - size_[0] = size ? size[0] : 0; - } - - inline __device__ __host__ unsigned getSize(const int i) const { - assert(i == 0); - return size_[0]; - } - - inline __device__ __host__ int numElements() const { - return size_[0]; - } - - inline __device__ __host__ DType &operator[](const size_t x) const { - return *(dptr_ + x); - } - - inline __device__ __host__ DType* data_ptr() const { - return dptr_; - } - - DType *dptr_; - int size_[1]; -}; - -template -static DeviceTensor devicetensor(const at::Tensor &blob) { - DType *data = blob.data(); - DeviceTensor tensor(data, nullptr); - for (int i = 0; i < Dim; ++i) { - tensor.size_[i] = blob.size(i); - } - return tensor; -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/cuda/psa_cuda.cu b/AIlib/segutils/core/nn/csrc/cuda/psa_cuda.cu deleted file mode 100644 index c47c98a..0000000 --- a/AIlib/segutils/core/nn/csrc/cuda/psa_cuda.cu +++ /dev/null @@ -1,214 +0,0 @@ -#include -#include - -#include -#include -#include - -#define PSA_TYPE_COLLECT 1 -#define PSA_TYPE_DISTRIBUTE 2 - -const int CUDA_NUM_THREADS = 512; - -inline int GET_BLOCKS(const int N) { - return (N + CUDA_NUM_THREADS - 1) / CUDA_NUM_THREADS; -} - -template -__global__ void psa_collect_forward_kernel(const T *hc, T *out, int num, int height, int width) { - const int out_h = 2 * height - 1; - const int out_w = 2 * width - 1; - const int half_out_h = (out_h - 1) / 2; - const int half_out_w = (out_w - 1) / 2; - - int x = blockIdx.x * blockDim.x + threadIdx.x; - int nthreads = num * height * width; - - for (int i = x; i < nthreads; i += blockDim.x * gridDim.x) { - const int w = i % width; - const int h = (i / width) % height; - const int n = i / width / height; - - // effective mask region : [hstart, hend) x [wstart, wend) with out-indexed - const int hstart = max(0, half_out_h - h); - const int hend = min(out_h, height + half_out_h - h); - const int wstart = max(0, half_out_w - w); - const int wend = min(out_w, width + half_out_w - w); - - // (hidx, widx) with out-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - out[(n * height * width + (hidx + h - half_out_h) * width + (widx + w - half_out_w)) * height * width + h * width + w] = - hc[((n * out_h * out_w + hidx * out_w + widx) * height + h) * width + w]; - } - } - } -} - -template -__global__ void psa_distribute_forward_kernel(const T *hc, T *out, int num, int height, int width) { - const int out_h = 2 * height - 1; - const int out_w = 2 * width - 1; - const int half_out_h = (out_h - 1) / 2; - const int half_out_w = (out_w - 1) / 2; - - int x = blockIdx.x * blockDim.x + threadIdx.x; - int nthreads = num * height * width; - - for (int i = x; i < nthreads; i += blockDim.x * gridDim.x) { - const int w = i % width; - const int h = (i / width) % height; - const int n = i / width / height; - - // effective mask region : [hstart, hend) x [wstart, wend) with out-indexed - const int hstart = max(0, half_out_h - h); - const int hend = min(out_h, height + half_out_h - h); - const int wstart = max(0, half_out_w - w); - const int wend = min(out_w, width + half_out_w - w); - - // (hidx, widx) with out-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - out[(n * height * width + h * width + w) * height * width + (hidx + h - half_out_h) * width + (widx + w - half_out_w)] = - hc[((n * out_h * out_w + hidx * out_w + widx) * height + h) * width + w]; - } - } - } -} - -template -__global__ void psa_collect_backward_kernel(const T *dout, T *dhc, int num, int height, int width) { - const int out_h = 2 * height - 1; - const int out_w = 2 * width - 1; - const int half_out_h = (out_h - 1) / 2; - const int half_out_w = (out_w - 1) / 2; - - int x = blockIdx.x * blockDim.x + threadIdx.x; - int nthreads = num * height * width; - - for (int i = x; i < nthreads; i += blockDim.x * gridDim.x) { - const int w = i % width; - const int h = (i / width) % height; - const int n = i / width / height; - - // effective mask region : [hstart, hend) x [wstart, wend) with out-indexed - const int hstart = max(0, half_out_h - h); - const int hend = min(out_h, height + half_out_h - h); - const int wstart = max(0, half_out_w - w); - const int wend = min(out_w, width + half_out_w - w); - - // (hidx, widx) with out-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - dhc[((h * out_h * out_w + hidx * out_w + widx) * height + h) * width + w] = - dout[(n * height * width + (hidx + h - half_out_h) * width + (widx + w - half_out_w)) * height * width + h * width + w]; - } - } - } -} - -template -__global__ void psa_distribute_backward_kernel(const T *dout, T *dhc, int num, int height, int width) { - const int out_h = 2 * height - 1; - const int out_w = 2 * width - 1; - const int half_out_h = (out_h - 1) / 2; - const int half_out_w = (out_w - 1) / 2; - - int x = blockIdx.x * blockDim.x + threadIdx.x; - int nthreads = num * height * width; - - for (int i = x; i < nthreads; i += blockDim.x * gridDim.x) { - const int w = i % width; - const int h = (i / width) % height; - const int n = i / width / height; - - // effective mask region : [hstart, hend) x [wstart, wend) with out-indexed - const int hstart = max(0, half_out_h - h); - const int hend = min(out_h, height + half_out_h - h); - const int wstart = max(0, half_out_w - w); - const int wend = min(out_w, width + half_out_w - w); - - // (hidx, widx) with out-indexed - // (hidx + h - half_mask_H_, widx + w - half_mask_W_) with feature-indexed - for (int hidx = hstart; hidx < hend; hidx++) { - for (int widx = wstart; widx < wend; widx++) { - dhc[((n * out_h * out_w + hidx * out_w + widx) * height + h) * width + w] = - dout[(n * height * width + h * width + w) * height * width + (hidx + h - half_out_h) * width + (widx + w - half_out_w)]; - } - } - } -} - -at::Tensor psa_forward_cuda(const at::Tensor& hc, const int forward_type) { - AT_ASSERTM(hc.type().is_cuda(), "input must be a CUDA tensor"); - - auto n = hc.size(0); - auto c = hc.size(1); - auto h = hc.size(2); - auto w = hc.size(3); - - at::Tensor out = at::zeros({n, h * w, h * w}, hc.options()); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - - int nthreads = n * h * w; - - switch (forward_type) { - case PSA_TYPE_COLLECT: - AT_DISPATCH_FLOATING_TYPES(hc.type(), "psa_forward", [&] { - psa_collect_forward_kernel<<>>( - hc.contiguous().data(), - out.contiguous().data(), - n, h, w); - }); - break; - case PSA_TYPE_DISTRIBUTE: - AT_DISPATCH_FLOATING_TYPES(hc.type(), "psa_forward", [&] { - psa_distribute_forward_kernel<<>>( - hc.contiguous().data(), - out.contiguous().data(), - n, h, w); - }); - break; - } - THCudaCheck(cudaGetLastError()); - return out; -} - -at::Tensor psa_backward_cuda(const at::Tensor& dout, const at::Tensor& hc, const int forward_type) { - AT_ASSERTM(dout.type().is_cuda(), "input must be a CUDA tensor"); - AT_ASSERTM(hc.type().is_cuda(), "input must be a CUDA tensor"); - - auto n = hc.size(0); - auto c = hc.size(1); - auto h = hc.size(2); - auto w = hc.size(3); - - at::Tensor dhc = at::zeros_like(hc); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - - int nthreads = n * h * w; - - switch (forward_type) { - case PSA_TYPE_COLLECT: - AT_DISPATCH_FLOATING_TYPES(hc.type(), "psa_backward", [&] { - psa_collect_backward_kernel<<>>( - dout.contiguous().data(), - dhc.contiguous().data(), - n, h, w); - }); - break; - case PSA_TYPE_DISTRIBUTE: - AT_DISPATCH_FLOATING_TYPES(hc.type(), "psa_backward", [&] { - psa_distribute_backward_kernel<<>>( - dout.contiguous().data(), - dhc.contiguous().data(), - n, h, w); - }); - break; - } - THCudaCheck(cudaGetLastError()); - return dhc; -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/cuda/syncbn_cuda.cu b/AIlib/segutils/core/nn/csrc/cuda/syncbn_cuda.cu deleted file mode 100644 index dcaed67..0000000 --- a/AIlib/segutils/core/nn/csrc/cuda/syncbn_cuda.cu +++ /dev/null @@ -1,488 +0,0 @@ -#include -// #include -#include -#include -#include - -#include -#include - -#include "helper.h" - -namespace { - -template -struct GradOp { - __device__ GradOp(Acctype m, const DeviceTensor3 i, const DeviceTensor3 g) - : beta(m), output(i), gradOutput(g) {} - __device__ __forceinline__ Float2 operator()(int batch, int plane, int n) { - DType g = gradOutput[batch][plane][n]; - DType c = ScalarConvert::to(output[batch][plane][n] - beta); - return Float2(g, g * c); - } - const Acctype beta; - const DeviceTensor3 output; - const DeviceTensor3 gradOutput; -}; - -template -struct SumOp { - __device__ SumOp(DeviceTensor i) : input(i){} - __device__ __forceinline__ Float2 operator()(int batch, int plane, int n) { - DType g = input[batch][plane][n]; - return Float2(g, g * g); - } - DType mean; - DeviceTensor input; -}; - -// Sum across (batch, x/y/z) applying Op() pointwise -template -__device__ T reduce(Op op, DeviceTensor3 tensor, int plane) { - T sum = (T)0; - for (int batch = 0; batch < tensor.getSize(0); ++batch) { - for (int x = threadIdx.x; x < tensor.getSize(2); x += blockDim.x) { - sum += op(batch, plane, x); - } - } - - // sum over NumThreads within a warp - sum = warpSum(sum); - - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T)0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} - -template -__global__ void batchnorm_forward_kernel ( - DeviceTensor output, - DeviceTensor input, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta) { - int c = blockIdx.x; - /* main operation */ - for (int b = 0; b < input.getSize(0); ++b) { - for (int x = threadIdx.x; x < input.getSize(2); x += blockDim.x) { - DType inp = input[b][c][x]; - output[b][c][x] = gamma[c] * (inp - mean[c]) / - std[c] + beta[c]; - } - } -} - -template -__global__ void inp_batchnorm_forward_kernel ( - DeviceTensor input, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta) { - int c = blockIdx.x; - /* main operation */ - for (int b = 0; b < input.getSize(0); ++b) { - for (int x = threadIdx.x; x < input.getSize(2); x += blockDim.x) { - DType inp = input[b][c][x]; - input[b][c][x] = gamma[c] * (inp - mean[c]) / - std[c] + beta[c]; - } - } -} - -template -__global__ void expectation_forward_kernel ( - DeviceTensor input, - DeviceTensor ex, - DeviceTensor exs, - DType norm) { - int c = blockIdx.x; - /* main operation */ - SumOp g(input); - Float2 res = reduce, - SumOp, DeviceTensor>(g, input, c); - DType xsum = res.v1; - DType xsquare = res.v2; - if (threadIdx.x == 0) { - ex[c] = xsum * norm; - exs[c] = xsquare * norm; - } -} - -template -__global__ void batchnorm_backward_kernel ( - DeviceTensor gradoutput, - DeviceTensor input, - DeviceTensor gradinput, - DeviceTensor gradgamma, - DeviceTensor gradbeta, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta, - DeviceTensor gradEx, - DeviceTensor gradExs) { - /* declarations of the variables */ - /* Get the index and channels */ - int c = blockIdx.x; - /* main operation */ - GradOp> g(mean[c], input, gradoutput); - Float2 res = reduce, - GradOp>, - DeviceTensor>(g, gradoutput, c); - DType gradOutputSum = res.v1; - DType dotP = res.v2; - DType invstd = DType(1.0) / std[c]; - DType gradScale = invstd * gamma[c]; - if (threadIdx.x == 0) { - gradEx[c] = - gradOutputSum * gradScale + mean[c] * invstd * invstd * dotP * gradScale; - gradExs[c] = - 0.5 * invstd * invstd * dotP * gradScale; - } - if (gradinput.numElements() > 0) { - for (int batch = 0; batch < gradoutput.getSize(0); ++batch) { - for (int x = threadIdx.x; x < gradoutput.getSize(2); x += blockDim.x) { - gradinput[batch][c][x] = gradoutput[batch][c][x] * gradScale; - } - } - } - if (gradgamma.numElements() > 0) { - if (threadIdx.x == 0) { - gradgamma[c] += dotP * invstd; - } - } - if (gradbeta.numElements() > 0) { - if (threadIdx.x == 0) { - gradbeta[c] += gradOutputSum; - } - } -} - -template -__global__ void inp_batchnorm_backward_kernel ( - DeviceTensor gradoutput, - DeviceTensor output, - DeviceTensor gradinput, - DeviceTensor gradgamma, - DeviceTensor gradbeta, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta, - DeviceTensor gradEx, - DeviceTensor gradExs) { - /* declarations of the variables */ - /* Get the index and channels */ - int c = blockIdx.x; - /* main operation */ - GradOp> g(beta[c], output, gradoutput); - Float2 res = reduce, - GradOp>, - DeviceTensor>(g, gradoutput, c); - DType gradOutputSum = res.v1; - DType dotP = res.v2; - DType invstd = DType(1.0) / std[c]; - DType gradScale = invstd * gamma[c]; - if (threadIdx.x == 0) { - gradEx[c] = - gradOutputSum * gradScale + mean[c] * invstd * invstd * dotP; - gradExs[c] = - 0.5 * invstd * invstd * dotP; - } - if (gradinput.numElements() > 0) { - for (int batch = 0; batch < gradoutput.getSize(0); ++batch) { - for (int x = threadIdx.x; x < gradoutput.getSize(2); x += blockDim.x) { - gradinput[batch][c][x] = gradoutput[batch][c][x] * gradScale; - } - } - } - if (gradgamma.numElements() > 0) { - if (threadIdx.x == 0) { - gradgamma[c] += dotP / gamma[c]; - } - } - if (gradbeta.numElements() > 0) { - if (threadIdx.x == 0) { - gradbeta[c] += gradOutputSum; - } - } -} - -template -__global__ void expectation_backward_kernel ( - DeviceTensor gradInput, - DeviceTensor input, - DeviceTensor gradEx, - DeviceTensor gradExs, - DType norm) { - int c = blockIdx.x; - /* main operation */ - for (int batch = 0; batch < gradInput.getSize(0); ++batch) { - for (int x = threadIdx.x; x < gradInput.getSize(2); x += blockDim.x) { - gradInput[batch][c][x] = gradEx[c] * norm + 2 * gradExs[c] * - input[batch][c][x] * norm; - } - } -} - -template -__global__ void inp_expectation_backward_kernel ( - DeviceTensor gradInput, - DeviceTensor output, - DeviceTensor gradEx, - DeviceTensor gradExs, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta, - DType norm) { - int c = blockIdx.x; - /* main operation */ - for (int batch = 0; batch < gradInput.getSize(0); ++batch) { - for (int x = threadIdx.x; x < gradInput.getSize(2); x += blockDim.x) { - gradInput[batch][c][x] += gradEx[c] * norm + 2 * gradExs[c] * - ((output[batch][c][x] - beta[c]) / gamma[c] * std[c] + mean[c]) * norm; - } - } -} - -} // namespace - -at::Tensor batchnorm_forward_cuda( - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - auto output_ = at::zeros_like(input_); - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "batchnorm_forward_cuda", ([&] { - /* Device tensors */ - DeviceTensor output = devicetensor(output_); - DeviceTensor input = devicetensor(input_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - /* kernel function */ - batchnorm_forward_kernel<<>>( - output, input, ex, std, gamma, beta); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return output_; -} - -at::Tensor inp_batchnorm_forward_cuda( - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "inp_batchnorm_forward_cuda", ([&] { - /* Device tensors */ - DeviceTensor input = devicetensor(input_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - /* kernel function */ - inp_batchnorm_forward_kernel<<>>( - input, ex, std, gamma, beta); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return input_; -} - -std::vector batchnorm_backward_cuda( - const at::Tensor gradoutput_, - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - /* outputs*/ - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - auto gradinput_ = at::zeros_like(input_); - auto gradgamma_ = at::zeros_like(gamma_); - auto gradbeta_ = at::zeros_like(beta_); - auto gradEx_ = at::zeros_like(ex_); - auto gradExs_ = at::zeros_like(std_); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "batchnorm_backward_cuda", ([&] { - /* Device tensors */ - DeviceTensor gradoutput = devicetensor(gradoutput_); - DeviceTensor input = devicetensor(input_); - DeviceTensor gradinput = devicetensor(gradinput_); - DeviceTensor gradgamma = devicetensor(gradgamma_); - DeviceTensor gradbeta = devicetensor(gradbeta_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - DeviceTensor gradEx = devicetensor(gradEx_); - DeviceTensor gradExs = devicetensor(gradExs_); - /* kernel function */ - batchnorm_backward_kernel - <<>>( - gradoutput, input, gradinput, gradgamma, gradbeta, ex, std, - gamma, beta, gradEx, gradExs); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return {gradinput_, gradEx_, gradExs_, gradgamma_, gradbeta_}; -} - -std::vector inp_batchnorm_backward_cuda( - const at::Tensor gradoutput_, - const at::Tensor output_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - /* outputs*/ - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - auto gradinput_ = at::zeros_like(output_); - auto gradgamma_ = at::zeros_like(gamma_); - auto gradbeta_ = at::zeros_like(beta_); - auto gradEx_ = at::zeros_like(ex_); - auto gradExs_ = at::zeros_like(std_); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(output_.size(1)); - dim3 threads(getNumThreads(output_.size(2))); - AT_DISPATCH_FLOATING_TYPES(output_.type(), "inp_batchnorm_backward_cuda", ([&] { - /* Device tensors */ - DeviceTensor gradoutput = devicetensor(gradoutput_); - DeviceTensor output = devicetensor(output_); - DeviceTensor gradinput = devicetensor(gradinput_); - DeviceTensor gradgamma = devicetensor(gradgamma_); - DeviceTensor gradbeta = devicetensor(gradbeta_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - DeviceTensor gradEx = devicetensor(gradEx_); - DeviceTensor gradExs = devicetensor(gradExs_); - /* kernel function */ - inp_batchnorm_backward_kernel - <<>>( - gradoutput, output, gradinput, gradgamma, gradbeta, ex, std, - gamma, beta, gradEx, gradExs); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return {gradinput_, gradEx_, gradExs_, gradgamma_, gradbeta_}; -} - -std::vector expectation_forward_cuda( - const at::Tensor input_) { - /* outputs */ - auto ex_ = torch::zeros({input_.size(1)}, input_.options()); - auto exs_ = torch::zeros({input_.size(1)}, input_.options()); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "expectation_forward_cuda", ([&] { - scalar_t norm = scalar_t(1) / (input_.size(0) * input_.size(2)); - /* Device tensors */ - DeviceTensor input = devicetensor(input_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor exs = devicetensor(exs_); - /* kernel function */ - expectation_forward_kernel - <<>>(input, ex, exs, norm); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return {ex_, exs_}; -} - -at::Tensor expectation_backward_cuda( - const at::Tensor input_, - const at::Tensor gradEx_, - const at::Tensor gradExs_) { - /* outputs */ - at::Tensor gradInput_ = at::zeros_like(input_); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "expectation_backward_cuda", ([&] { - scalar_t norm = scalar_t(1) / (input_.size(0) * input_.size(2)); - /* Device tensors */ - DeviceTensor gradInput = devicetensor(gradInput_); - DeviceTensor input = devicetensor(input_); - DeviceTensor gradEx = devicetensor(gradEx_); - DeviceTensor gradExs =devicetensor(gradExs_); - /* kernel function */ - expectation_backward_kernel - <<>>(gradInput, input, gradEx, gradExs, norm); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return gradInput_; -} - -at::Tensor inp_expectation_backward_cuda( - const at::Tensor gradInput_, - const at::Tensor output_, - const at::Tensor gradEx_, - const at::Tensor gradExs_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - /* outputs */ - //auto gradInput_ = at::zeros_like(output_); - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(output_.size(1)); - dim3 threads(getNumThreads(output_.size(2))); - AT_DISPATCH_FLOATING_TYPES(output_.type(), "inp_expectation_backward_cuda", ([&] { - scalar_t norm = scalar_t(1) / (output_.size(0) * output_.size(2)); - /* Device tensors */ - DeviceTensor gradInput = devicetensor(gradInput_); - DeviceTensor input = devicetensor(output_); - DeviceTensor gradEx = devicetensor(gradEx_); - DeviceTensor gradExs =devicetensor(gradExs_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - /* kernel function */ - inp_expectation_backward_kernel - <<>>(gradInput, input, gradEx, gradExs, - ex, std, gamma, beta, norm); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return gradInput_; -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/cuda/vision.h b/AIlib/segutils/core/nn/csrc/cuda/vision.h deleted file mode 100644 index 6696840..0000000 --- a/AIlib/segutils/core/nn/csrc/cuda/vision.h +++ /dev/null @@ -1,84 +0,0 @@ -#pragma once -#include -#include - - -at::Tensor ca_forward_cuda( - const at::Tensor& t, - const at::Tensor& f); - -std::tuple ca_backward_cuda( - const at::Tensor& dw, - const at::Tensor& t, - const at::Tensor& f); - -at::Tensor ca_map_forward_cuda( - const at::Tensor& weight, - const at::Tensor& g); - -std::tuple ca_map_backward_cuda( - const at::Tensor& dout, - const at::Tensor& weight, - const at::Tensor& g); - -at::Tensor psa_forward_cuda( - const at::Tensor& hc, - const int forward_type); - -at::Tensor psa_backward_cuda( - const at::Tensor& dout, - const at::Tensor& hc, - const int forward_type); - -at::Tensor batchnorm_forward_cuda( - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -at::Tensor inp_batchnorm_forward_cuda( - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -std::vector batchnorm_backward_cuda( - const at::Tensor gradoutput_, - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -std::vector inp_batchnorm_backward_cuda( - const at::Tensor gradoutput_, - const at::Tensor output_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -std::vector expectation_forward_cuda( - const at::Tensor input_); - -at::Tensor expectation_backward_cuda( - const at::Tensor input_, - const at::Tensor gradEx_, - const at::Tensor gradExs_); - -at::Tensor inp_expectation_backward_cuda( - const at::Tensor gradInput_, - const at::Tensor output_, - const at::Tensor gradEx_, - const at::Tensor gradExs_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/psa.h b/AIlib/segutils/core/nn/csrc/psa.h deleted file mode 100644 index 1702581..0000000 --- a/AIlib/segutils/core/nn/csrc/psa.h +++ /dev/null @@ -1,33 +0,0 @@ -#pragma once - -#include "cpu/vision.h" - -#ifdef WITH_CUDA -#include "cuda/vision.h" -#endif - -// Interface for Python -at::Tensor psa_forward(const at::Tensor& hc, - const int forward_type) { - if (hc.type().is_cuda()) { - #ifdef WITH_CUDA - return psa_forward_cuda(hc, forward_type); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - return psa_forward_cpu(hc, forward_type); -} - -at::Tensor psa_backward(const at::Tensor& dout, - const at::Tensor& hc, - const int forward_type) { - if (hc.type().is_cuda()) { - #ifdef WITH_CUDA - return psa_backward_cuda(dout, hc, forward_type); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - return psa_backward_cpu(dout, hc, forward_type); -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/syncbn.h b/AIlib/segutils/core/nn/csrc/syncbn.h deleted file mode 100644 index fbcf695..0000000 --- a/AIlib/segutils/core/nn/csrc/syncbn.h +++ /dev/null @@ -1,118 +0,0 @@ -#pragma once - -#include -#include "cpu/vision.h" - -#ifdef WITH_CUDA -#include "cuda/vision.h" -#endif - -// Interface for Python -at::Tensor batchnorm_forward(const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - if (input_.type().is_cuda()) { - #ifdef WITH_CUDA - return batchnorm_forward_cuda(input_, ex_, exs_, gamma_, beta_, eps); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - return batchnorm_forward_cpu(input_, ex_, exs_, gamma_, beta_, eps); -} - -at::Tensor inp_batchnorm_forward(const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - if (input_.type().is_cuda()) { - #ifdef WITH_CUDA - return inp_batchnorm_forward_cuda(input_, ex_, exs_, gamma_, beta_, eps); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - AT_ERROR("Not implemented on the CPU"); -} - -std::vector batchnorm_backward(const at::Tensor gradoutput_, - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - if (gradoutput_.type().is_cuda()) { - #ifdef WITH_CUDA - return batchnorm_backward_cuda(gradoutput_, input_, ex_, exs_, gamma_, beta_, eps); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - return batchnorm_backward_cpu(gradoutput_, input_, ex_, exs_, gamma_, beta_, eps); -} - -std::vector inp_batchnorm_backward(const at::Tensor gradoutput_, - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - if (gradoutput_.type().is_cuda()) { - #ifdef WITH_CUDA - return inp_batchnorm_backward_cuda(gradoutput_, input_, ex_, exs_, gamma_, beta_, eps); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - AT_ERROR("Not implemented on the CPU"); -} - -std::vector expectation_forward(const at::Tensor input_) { - if (input_.type().is_cuda()) { - #ifdef WITH_CUDA - return expectation_forward_cuda(input_); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - AT_ERROR("Not implemented on the CPU"); -} - -at::Tensor expectation_backward(const at::Tensor input_, - const at::Tensor gradEx_, - const at::Tensor gradExs_) { - if (input_.type().is_cuda()) { - #ifdef WITH_CUDA - return expectation_backward_cuda(input_, gradEx_, gradExs_); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - AT_ERROR("Not implemented on the CPU"); -} - -at::Tensor inp_expectation_backward(const at::Tensor gradInput_, - const at::Tensor output_, - const at::Tensor gradEx_, - const at::Tensor gradExs_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - if (output_.type().is_cuda()) { - #ifdef WITH_CUDA - return inp_expectation_backward_cuda(gradInput_, output_, gradEx_, gradExs_, ex_, exs_, gamma_, beta_, eps); - #else - AT_ERROR("Not compiled with GPU support"); - #endif - } - AT_ERROR("Not implemented on the CPU"); -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/csrc/vision.cpp b/AIlib/segutils/core/nn/csrc/vision.cpp deleted file mode 100644 index c369176..0000000 --- a/AIlib/segutils/core/nn/csrc/vision.cpp +++ /dev/null @@ -1,19 +0,0 @@ -#include "ca.h" -#include "psa.h" -#include "syncbn.h" - -PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { - m.def("ca_forward", &ca_forward, "ca_forward"); - m.def("ca_backward", &ca_backward, "ca_backward"); - m.def("ca_map_forward", &ca_map_forward, "ca_map_forward"); - m.def("ca_map_backward", &ca_map_backward, "ca_map_backward"); - m.def("psa_forward", &psa_forward, "psa_forward"); - m.def("psa_backward", &psa_backward, "psa_backward"); - m.def("batchnorm_forward", &batchnorm_forward, "batchnorm_forward"); - m.def("inp_batchnorm_forward", &inp_batchnorm_forward, "inp_batchnorm_forward"); - m.def("batchnorm_backward", &batchnorm_backward, "batchnorm_backward"); - m.def("inp_batchnorm_backward", &inp_batchnorm_backward, "inp_batchnorm_backward"); - m.def("expectation_forward", &expectation_forward, "expectation_forward"); - m.def("expectation_backward", &expectation_backward, "expectation_backward"); - m.def("inp_expectation_backward", &inp_expectation_backward, "inp_expectation_backward"); -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/jpu.py b/AIlib/segutils/core/nn/jpu.py deleted file mode 100644 index db23bab..0000000 --- a/AIlib/segutils/core/nn/jpu.py +++ /dev/null @@ -1,68 +0,0 @@ -"""Joint Pyramid Upsampling""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -__all__ = ['JPU'] - - -class SeparableConv2d(nn.Module): - def __init__(self, inplanes, planes, kernel_size=3, stride=1, padding=1, - dilation=1, bias=False, norm_layer=nn.BatchNorm2d): - super(SeparableConv2d, self).__init__() - self.conv = nn.Conv2d(inplanes, inplanes, kernel_size, stride, padding, dilation, groups=inplanes, bias=bias) - self.bn = norm_layer(inplanes) - self.pointwise = nn.Conv2d(inplanes, planes, 1, bias=bias) - - def forward(self, x): - x = self.conv(x) - x = self.bn(x) - x = self.pointwise(x) - return x - - -# copy from: https://github.com/wuhuikai/FastFCN/blob/master/encoding/nn/customize.py -class JPU(nn.Module): - def __init__(self, in_channels, width=512, norm_layer=nn.BatchNorm2d, **kwargs): - super(JPU, self).__init__() - - self.conv5 = nn.Sequential( - nn.Conv2d(in_channels[-1], width, 3, padding=1, bias=False), - norm_layer(width), - nn.ReLU(True)) - self.conv4 = nn.Sequential( - nn.Conv2d(in_channels[-2], width, 3, padding=1, bias=False), - norm_layer(width), - nn.ReLU(True)) - self.conv3 = nn.Sequential( - nn.Conv2d(in_channels[-3], width, 3, padding=1, bias=False), - norm_layer(width), - nn.ReLU(True)) - - self.dilation1 = nn.Sequential( - SeparableConv2d(3 * width, width, 3, padding=1, dilation=1, bias=False), - norm_layer(width), - nn.ReLU(True)) - self.dilation2 = nn.Sequential( - SeparableConv2d(3 * width, width, 3, padding=2, dilation=2, bias=False), - norm_layer(width), - nn.ReLU(True)) - self.dilation3 = nn.Sequential( - SeparableConv2d(3 * width, width, 3, padding=4, dilation=4, bias=False), - norm_layer(width), - nn.ReLU(True)) - self.dilation4 = nn.Sequential( - SeparableConv2d(3 * width, width, 3, padding=8, dilation=8, bias=False), - norm_layer(width), - nn.ReLU(True)) - - def forward(self, *inputs): - feats = [self.conv5(inputs[-1]), self.conv4(inputs[-2]), self.conv3(inputs[-3])] - size = feats[-1].size()[2:] - feats[-2] = F.interpolate(feats[-2], size, mode='bilinear', align_corners=True) - feats[-3] = F.interpolate(feats[-3], size, mode='bilinear', align_corners=True) - feat = torch.cat(feats, dim=1) - feat = torch.cat([self.dilation1(feat), self.dilation2(feat), self.dilation3(feat), self.dilation4(feat)], - dim=1) - - return inputs[0], inputs[1], inputs[2], feat diff --git a/AIlib/segutils/core/nn/psa_block.py b/AIlib/segutils/core/nn/psa_block.py deleted file mode 100644 index c8ff11b..0000000 --- a/AIlib/segutils/core/nn/psa_block.py +++ /dev/null @@ -1,71 +0,0 @@ -import torch -import torch.nn as nn - -from torch.autograd.function import once_differentiable -#from core.nn import _C - -__all__ = ['CollectAttention', 'DistributeAttention', 'psa_collect', 'psa_distribute'] - - -class _PSACollect(torch.autograd.Function): - @staticmethod - def forward(ctx, hc): - out = _C.psa_forward(hc, 1) - - ctx.save_for_backward(hc) - - return out - - @staticmethod - @once_differentiable - def backward(ctx, dout): - hc = ctx.saved_tensors - - dhc = _C.psa_backward(dout, hc[0], 1) - - return dhc - - -class _PSADistribute(torch.autograd.Function): - @staticmethod - def forward(ctx, hc): - out = _C.psa_forward(hc, 2) - - ctx.save_for_backward(hc) - - return out - - @staticmethod - @once_differentiable - def backward(ctx, dout): - hc = ctx.saved_tensors - - dhc = _C.psa_backward(dout, hc[0], 2) - - return dhc - - -psa_collect = _PSACollect.apply -psa_distribute = _PSADistribute.apply - - -class CollectAttention(nn.Module): - """Collect Attention Generation Module""" - - def __init__(self): - super(CollectAttention, self).__init__() - - def forward(self, x): - out = psa_collect(x) - return out - - -class DistributeAttention(nn.Module): - """Distribute Attention Generation Module""" - - def __init__(self): - super(DistributeAttention, self).__init__() - - def forward(self, x): - out = psa_distribute(x) - return out diff --git a/AIlib/segutils/core/nn/setup.py b/AIlib/segutils/core/nn/setup.py deleted file mode 100644 index ec800c4..0000000 --- a/AIlib/segutils/core/nn/setup.py +++ /dev/null @@ -1,56 +0,0 @@ -# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. -# !/usr/bin/env python -# reference: https://github.com/facebookresearch/maskrcnn-benchmark/blob/90c226cf10e098263d1df28bda054a5f22513b4f/setup.py - -import os -import glob -import torch - -from setuptools import setup -from torch.utils.cpp_extension import BuildExtension, CppExtension, CUDAExtension, CUDA_HOME - -requirements = ["torch"] - - -def get_extension(): - this_dir = os.path.dirname(os.path.abspath(__file__)) - extensions_dir = os.path.join(this_dir, "csrc") - - main_file = glob.glob(os.path.join(extensions_dir, "*.cpp")) - source_cpu = glob.glob(os.path.join(extensions_dir, "cpu", "*.cpp")) - source_cuda = glob.glob(os.path.join(extensions_dir, "cuda", "*.cu")) - - sources = main_file + source_cpu - extension = CppExtension - - define_macros = [] - - if (torch.cuda.is_available() and CUDA_HOME is not None) or os.getenv("FORCE_CUDA", "0") == "1": - extension = CUDAExtension - sources += source_cuda - define_macros += [("WITH_CUDA", None)] - - sources = [os.path.join(extensions_dir, s) for s in sources] - - include_dirs = [extensions_dir] - - ext_modules = [ - extension( - "._C", - sources, - include_dirs=include_dirs, - define_macros=define_macros, - ) - ] - - return ext_modules - - -setup( - name="semantic_segmentation", - version="0.1", - author="tramac", - description="semantic segmentation in pytorch", - ext_modules=get_extension(), - cmdclass={"build_ext": BuildExtension} -) \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/__init__.py b/AIlib/segutils/core/nn/sync_bn/__init__.py deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/segutils/core/nn/sync_bn/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/nn/sync_bn/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index fd85555..0000000 Binary files a/AIlib/segutils/core/nn/sync_bn/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/sync_bn/__pycache__/functions.cpython-36.pyc b/AIlib/segutils/core/nn/sync_bn/__pycache__/functions.cpython-36.pyc deleted file mode 100644 index 063b818..0000000 Binary files a/AIlib/segutils/core/nn/sync_bn/__pycache__/functions.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/sync_bn/__pycache__/syncbn.cpython-36.pyc b/AIlib/segutils/core/nn/sync_bn/__pycache__/syncbn.cpython-36.pyc deleted file mode 100644 index f75150f..0000000 Binary files a/AIlib/segutils/core/nn/sync_bn/__pycache__/syncbn.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/sync_bn/functions.py b/AIlib/segutils/core/nn/sync_bn/functions.py deleted file mode 100644 index b0102e6..0000000 --- a/AIlib/segutils/core/nn/sync_bn/functions.py +++ /dev/null @@ -1,285 +0,0 @@ -##+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ -## Created by: Hang Zhang -## Email: zhanghang0704@gmail.com -## Copyright (c) 2018 -## -## This source code is licensed under the MIT-style license found in the -## LICENSE file in the root directory of this source tree -##+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - -"""Synchronized Cross-GPU Batch Normalization functions""" -import torch.cuda.comm as comm - -from torch.autograd import Function -from torch.autograd.function import once_differentiable -from core.nn.sync_bn import lib - -__all__ = ['syncbatchnorm', 'inp_syncbatchnorm'] - - -class syncbatchnorm_(Function): - @classmethod - def forward(cls, ctx, x, gamma, beta, running_mean, running_var, - extra, sync=True, training=True, momentum=0.1, eps=1e-05, - activation="none", slope=0.01): - # save context - cls._parse_extra(ctx, extra) - ctx.sync = sync - ctx.training = training - ctx.momentum = momentum - ctx.eps = eps - ctx.activation = activation - ctx.slope = slope - assert activation == 'none' - - # continous inputs - x = x.contiguous() - gamma = gamma.contiguous() - beta = beta.contiguous() - - if ctx.training: - if x.is_cuda: - _ex, _exs = lib.gpu.expectation_forward(x) - else: - raise NotImplemented - - if ctx.sync: - if ctx.is_master: - _ex, _exs = [_ex.unsqueeze(0)], [_exs.unsqueeze(0)] - for _ in range(ctx.master_queue.maxsize): - _ex_w, _exs_w = ctx.master_queue.get() - ctx.master_queue.task_done() - _ex.append(_ex_w.unsqueeze(0)) - _exs.append(_exs_w.unsqueeze(0)) - - _ex = comm.gather(_ex).mean(0) - _exs = comm.gather(_exs).mean(0) - - tensors = comm.broadcast_coalesced((_ex, _exs), [_ex.get_device()] + ctx.worker_ids) - for ts, queue in zip(tensors[1:], ctx.worker_queues): - queue.put(ts) - else: - ctx.master_queue.put((_ex, _exs)) - _ex, _exs = ctx.worker_queue.get() - ctx.worker_queue.task_done() - - # Update running stats - _var = _exs - _ex ** 2 - running_mean.mul_((1 - ctx.momentum)).add_(ctx.momentum * _ex) - running_var.mul_((1 - ctx.momentum)).add_(ctx.momentum * _var) - - # Mark in-place modified tensors - ctx.mark_dirty(running_mean, running_var) - else: - _ex, _var = running_mean.contiguous(), running_var.contiguous() - _exs = _var + _ex ** 2 - - # BN forward - if x.is_cuda: - y = lib.gpu.batchnorm_forward(x, _ex, _exs, gamma, beta, ctx.eps) - else: - y = lib.cpu.batchnorm_forward(x, _ex, _exs, gamma, beta, ctx.eps) - - # Output - ctx.save_for_backward(x, _ex, _exs, gamma, beta) - return y - - @staticmethod - @once_differentiable - def backward(ctx, dz): - x, _ex, _exs, gamma, beta = ctx.saved_tensors - dz = dz.contiguous() - - # BN backward - if dz.is_cuda: - dx, _dex, _dexs, dgamma, dbeta = lib.gpu.batchnorm_backward(dz, x, _ex, _exs, gamma, beta, ctx.eps) - else: - raise NotImplemented - - if ctx.training: - if ctx.sync: - if ctx.is_master: - _dex, _dexs = [_dex.unsqueeze(0)], [_dexs.unsqueeze(0)] - for _ in range(ctx.master_queue.maxsize): - _dex_w, _dexs_w = ctx.master_queue.get() - ctx.master_queue.task_done() - _dex.append(_dex_w.unsqueeze(0)) - _dexs.append(_dexs_w.unsqueeze(0)) - - _dex = comm.gather(_dex).mean(0) - _dexs = comm.gather(_dexs).mean(0) - - tensors = comm.broadcast_coalesced((_dex, _dexs), [_dex.get_device()] + ctx.worker_ids) - for ts, queue in zip(tensors[1:], ctx.worker_queues): - queue.put(ts) - else: - ctx.master_queue.put((_dex, _dexs)) - _dex, _dexs = ctx.worker_queue.get() - ctx.worker_queue.task_done() - - if x.is_cuda: - dx_ = lib.gpu.expectation_backward(x, _dex, _dexs) - else: - raise NotImplemented - dx = dx + dx_ - - return dx, dgamma, dbeta, None, None, None, None, None, None, None, None, None - - @staticmethod - def _parse_extra(ctx, extra): - ctx.is_master = extra["is_master"] - if ctx.is_master: - ctx.master_queue = extra["master_queue"] - ctx.worker_queues = extra["worker_queues"] - ctx.worker_ids = extra["worker_ids"] - else: - ctx.master_queue = extra["master_queue"] - ctx.worker_queue = extra["worker_queue"] - - -def _act_forward(ctx, x): - if ctx.activation.lower() == "leaky_relu": - if x.is_cuda: - lib.gpu.leaky_relu_forward(x, ctx.slope) - else: - raise NotImplemented - else: - assert ctx.activation == 'none' - - -def _act_backward(ctx, x, dx): - if ctx.activation.lower() == "leaky_relu": - if x.is_cuda: - lib.gpu.leaky_relu_backward(x, dx, ctx.slope) - else: - raise NotImplemented - else: - assert ctx.activation == 'none' - - -class inp_syncbatchnorm_(Function): - @classmethod - def forward(cls, ctx, x, gamma, beta, running_mean, running_var, - extra, sync=True, training=True, momentum=0.1, eps=1e-5, - activation='none', slope=0.01): - # save context - cls._parse_extra(ctx, extra) - ctx.sync = sync - ctx.training = training - ctx.momentum = momentum - ctx.eps = eps - ctx.activation = activation - ctx.slope = slope - - # continous inputs - x = x.contiguous() - gamma = gamma.contiguous() - beta = beta.contiguous() - - if ctx.training: - if x.is_cuda: - _ex, _exs = lib.gpu.expectation_forward(x) - else: - raise NotImplemented - - if ctx.sync: - if ctx.is_master: - _ex, _exs = [_ex.unsqueeze(0)], [_exs.unsqueeze(0)] - for _ in range(ctx.master_queue.maxsize): - _ex_w, _exs_w = ctx.master_queue.get() - ctx.master_queue.task_done() - _ex.append(_ex_w.unsqueeze(0)) - _exs.append(_exs_w.unsuqeeze(0)) - - _ex = comm.gather(_ex).mean(0) - _exs = comm.gather(_exs).mean(0) - - tensors = comm.broadcast_coalesced((_ex, _exs), [_ex.get_device()] + ctx.worker_ids) - for ts, queue in zip(tensors[1:], ctx.worker_queues): - queue.put(ts) - else: - ctx.master_queue.put((_ex, _exs)) - _ex, _exs = ctx.worker_queue.get() - ctx.worker_queue.task_done() - - # Update running stats - _var = _exs - _ex ** 2 - running_mean.mul_((1 - ctx.momentum)).add_(ctx.momentum * _ex) - running_var.mul_((1 - ctx.momentum)).add_(ctx.momentum * _var) - - # Mark in-place modified tensors - ctx.mark_dirty(x, running_mean, running_var) - else: - _ex, _var = running_mean.contiguous(), running_var.contiguous() - _exs = _var + _ex ** 2 - ctx.mark_dirty(x) - - # BN forward + activation - if x.is_cuda: - lib.gpu.batchnorm_inp_forward(x, _ex, _exs, gamma, beta, ctx.eps) - else: - raise NotImplemented - - _act_forward(ctx, x) - - # Output - ctx.save_for_backward(x, _ex, _exs, gamma, beta) - return x - - @staticmethod - @once_differentiable - def backward(ctx, dz): - z, _ex, _exs, gamma, beta = ctx.saved_tensors - dz = dz.contiguous() - - # Undo activation - _act_backward(ctx, z, dz) - - # BN backward - if dz.is_cuda: - dx, _dex, _dexs, dgamma, dbeta = lib.gpu.batchnorm_inp_backward(dz, z, _ex, _exs, gamma, beta, ctx.eps) - else: - raise NotImplemented - - if ctx.training: - if ctx.sync: - if ctx.is_master: - _dex, _dexs = [_dex.unsqueeze(0)], [_dexs.unsqueeze(0)] - for _ in range(ctx.master_queue.maxsize): - _dex_w, _dexs_w = ctx.master_queue.get() - ctx.master_queue.task_done() - _dex.append(_dex_w.unsqueeze(0)) - _dexs.append(_dexs_w.unsqueeze(0)) - - _dex = comm.gather(_dex).mean(0) - _dexs = comm.gather(_dexs).mean(0) - - tensors = comm.broadcast_coalesced((_dex, _dexs), [_dex.get_device()] + ctx.worker_ids) - for ts, queue in zip(tensors[1:], ctx.worker_queues): - queue.put(ts) - else: - ctx.master_queue.put((_dex, _dexs)) - _dex, _dexs = ctx.worker_queue.get() - ctx.worker_queue.task_done() - - if z.is_cuda: - lib.gpu.expectation_inp_backward(dx, z, _dex, _dexs, _ex, _exs, gamma, beta, ctx.eps) - else: - raise NotImplemented - - return dx, dgamma, dbeta, None, None, None, None, None, None, None, None, None - - @staticmethod - def _parse_extra(ctx, extra): - ctx.is_master = extra["is_master"] - if ctx.is_master: - ctx.master_queue = extra["master_queue"] - ctx.worker_queues = extra["worker_queues"] - ctx.worker_ids = extra["worker_ids"] - else: - ctx.master_queue = extra["master_queue"] - ctx.worker_queue = extra["worker_queue"] - - -syncbatchnorm = syncbatchnorm_.apply -inp_syncbatchnorm = inp_syncbatchnorm_.apply diff --git a/AIlib/segutils/core/nn/sync_bn/lib/__init__.py b/AIlib/segutils/core/nn/sync_bn/lib/__init__.py deleted file mode 100644 index 98c3374..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/__init__.py +++ /dev/null @@ -1,20 +0,0 @@ -import os -import torch -from torch.utils.cpp_extension import load - -cwd = os.path.dirname(os.path.realpath(__file__)) -cpu_path = os.path.join(cwd, 'cpu') -gpu_path = os.path.join(cwd, 'gpu') - -cpu = load('sync_cpu', [ - os.path.join(cpu_path, 'operator.cpp'), - os.path.join(cpu_path, 'syncbn_cpu.cpp'), -], build_directory=cpu_path, verbose=False) - -if torch.cuda.is_available(): - gpu = load('sync_gpu', [ - os.path.join(gpu_path, 'operator.cpp'), - os.path.join(gpu_path, 'activation_kernel.cu'), - os.path.join(gpu_path, 'syncbn_kernel.cu'), - ], extra_cuda_cflags=["--expt-extended-lambda"], - build_directory=gpu_path, verbose=False) diff --git a/AIlib/segutils/core/nn/sync_bn/lib/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/nn/sync_bn/lib/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index 0110085..0000000 Binary files a/AIlib/segutils/core/nn/sync_bn/lib/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/nn/sync_bn/lib/cpu/.ninja_deps b/AIlib/segutils/core/nn/sync_bn/lib/cpu/.ninja_deps deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/segutils/core/nn/sync_bn/lib/cpu/.ninja_log b/AIlib/segutils/core/nn/sync_bn/lib/cpu/.ninja_log deleted file mode 100644 index d4c4d9d..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/cpu/.ninja_log +++ /dev/null @@ -1,7 +0,0 @@ -# ninja log v5 -0 6679 1555417150 syncbn_cpu.o b884354b4810778d -0 7702 1555417151 operator.o df6e270344a1d164 -7703 8115 1555417151 sync_cpu.so d148b4e40b0af67e -0 5172 1557113015 syncbn_cpu.o 9052547bb175072 -0 6447 1557113016 operator.o 209836e0b0c1e97e -6447 6613 1557113016 sync_cpu.so d148b4e40b0af67e diff --git a/AIlib/segutils/core/nn/sync_bn/lib/cpu/build.ninja b/AIlib/segutils/core/nn/sync_bn/lib/cpu/build.ninja deleted file mode 100644 index e432f66..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/cpu/build.ninja +++ /dev/null @@ -1,21 +0,0 @@ -ninja_required_version = 1.3 -cxx = c++ - -cflags = -DTORCH_EXTENSION_NAME=sync_cpu -DTORCH_API_INCLUDE_EXTENSION_H -isystem /home/tramac/.pyenv/versions/anaconda3-4.4.0/lib/python3.6/site-packages/torch/include -isystem /home/tramac/.pyenv/versions/anaconda3-4.4.0/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -isystem /home/tramac/.pyenv/versions/anaconda3-4.4.0/lib/python3.6/site-packages/torch/include/TH -isystem /home/tramac/.pyenv/versions/anaconda3-4.4.0/lib/python3.6/site-packages/torch/include/THC -isystem /home/tramac/.pyenv/versions/anaconda3-4.4.0/include/python3.6m -D_GLIBCXX_USE_CXX11_ABI=0 -fPIC -std=c++11 -ldflags = -shared - -rule compile - command = $cxx -MMD -MF $out.d $cflags -c $in -o $out - depfile = $out.d - deps = gcc - -rule link - command = $cxx $in $ldflags -o $out - -build operator.o: compile /home/tramac/PycharmProjects/awesome-semantic-segmentation-pytorch/core/nn/sync_bn/lib/cpu/operator.cpp -build syncbn_cpu.o: compile /home/tramac/PycharmProjects/awesome-semantic-segmentation-pytorch/core/nn/sync_bn/lib/cpu/syncbn_cpu.cpp - -build sync_cpu.so: link operator.o syncbn_cpu.o - -default sync_cpu.so - diff --git a/AIlib/segutils/core/nn/sync_bn/lib/cpu/operator.cpp b/AIlib/segutils/core/nn/sync_bn/lib/cpu/operator.cpp deleted file mode 100644 index 5981ffc..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/cpu/operator.cpp +++ /dev/null @@ -1,8 +0,0 @@ -#include "operator.h" - -PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { - m.def("batchnorm_forward", &BatchNorm_Forward_CPU, "BatchNorm forward (CPU)"); - m.def("batchnorm_backward", &BatchNorm_Backward_CPU, "BatchNorm backward (CPU)"); - m.def("sumsquare_forward", &Sum_Square_Forward_CPU, "SumSqu forward (CPU)"); - m.def("sumsquare_backward", &Sum_Square_Backward_CPU, "SumSqu backward (CPU)"); -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/lib/cpu/operator.h b/AIlib/segutils/core/nn/sync_bn/lib/cpu/operator.h deleted file mode 100644 index 215fd53..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/cpu/operator.h +++ /dev/null @@ -1,26 +0,0 @@ -#include -#include - -at::Tensor BatchNorm_Forward_CPU( - const at::Tensor input_, - const at::Tensor mean_, - const at::Tensor std_, - const at::Tensor gamma_, - const at::Tensor beta_); - -std::vector BatchNorm_Backward_CPU( - const at::Tensor gradoutput_, - const at::Tensor input_, - const at::Tensor mean_, - const at::Tensor std_, - const at::Tensor gamma_, - const at::Tensor beta_, - bool train); - -std::vector Sum_Square_Forward_CPU( - const at::Tensor input_); - -at::Tensor Sum_Square_Backward_CPU( - const at::Tensor input_, - const at::Tensor gradSum_, - const at::Tensor gradSquare_); \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/lib/cpu/operator.o b/AIlib/segutils/core/nn/sync_bn/lib/cpu/operator.o deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/segutils/core/nn/sync_bn/lib/cpu/setup.py b/AIlib/segutils/core/nn/sync_bn/lib/cpu/setup.py deleted file mode 100644 index b0ecd6c..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/cpu/setup.py +++ /dev/null @@ -1,14 +0,0 @@ -from setuptools import setup -from torch.utils.cpp_extension import BuildExtension, CppExtension - -setup( - name='syncbn_cpu', - ext_modules=[ - CppExtension('syncbn_cpu', [ - 'operator.cpp', - 'syncbn_cpu.cpp', - ]), - ], - cmdclass={ - 'build_ext': BuildExtension - }) diff --git a/AIlib/segutils/core/nn/sync_bn/lib/cpu/syncbn_cpu.cpp b/AIlib/segutils/core/nn/sync_bn/lib/cpu/syncbn_cpu.cpp deleted file mode 100644 index 6b6bb73..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/cpu/syncbn_cpu.cpp +++ /dev/null @@ -1,61 +0,0 @@ -#include -#include -#include - -at::Tensor broadcast_to(at::Tensor v, at::Tensor x) { - if (x.ndimension() == 2) { - return v; - } else { - std::vector broadcast_size = {1, -1}; - for (int64_t i = 2; i < x.ndimension(); ++i) - broadcast_size.push_back(1); - - return v.view(broadcast_size); - } -} - -at::Tensor BatchNorm_Forward_CPU( - const at::Tensor input, - const at::Tensor mean, - const at::Tensor std, - const at::Tensor gamma, - const at::Tensor beta) { - auto output = (input - broadcast_to(mean, input)) / broadcast_to(std, input); - output = output * broadcast_to(gamma, input) + broadcast_to(beta, input); - return output; -} - -// Not implementing CPU backward for now -std::vector BatchNorm_Backward_CPU( - const at::Tensor gradoutput, - const at::Tensor input, - const at::Tensor mean, - const at::Tensor std, - const at::Tensor gamma, - const at::Tensor beta, - bool train) { - /* outputs*/ - at::Tensor gradinput = at::zeros_like(input); - at::Tensor gradgamma = at::zeros_like(gamma); - at::Tensor gradbeta = at::zeros_like(beta); - at::Tensor gradMean = at::zeros_like(mean); - at::Tensor gradStd = at::zeros_like(std); - return {gradinput, gradMean, gradStd, gradgamma, gradbeta}; -} - -std::vector Sum_Square_Forward_CPU( - const at::Tensor input) { - /* outputs */ - at::Tensor sum = torch::zeros({input.size(1)}, input.options()); - at::Tensor square = torch::zeros({input.size(1)}, input.options()); - return {sum, square}; -} - -at::Tensor Sum_Square_Backward_CPU( - const at::Tensor input, - const at::Tensor gradSum, - const at::Tensor gradSquare) { - /* outputs */ - at::Tensor gradInput = at::zeros_like(input); - return gradInput; -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/lib/cpu/syncbn_cpu.o b/AIlib/segutils/core/nn/sync_bn/lib/cpu/syncbn_cpu.o deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/segutils/core/nn/sync_bn/lib/gpu/__init__.py b/AIlib/segutils/core/nn/sync_bn/lib/gpu/__init__.py deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/segutils/core/nn/sync_bn/lib/gpu/activation_kernel.cu b/AIlib/segutils/core/nn/sync_bn/lib/gpu/activation_kernel.cu deleted file mode 100644 index e696667..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/gpu/activation_kernel.cu +++ /dev/null @@ -1,46 +0,0 @@ -#include -// #include -#include -#include -#include - -#include - -#include -#include - - -namespace { - -template -inline void leaky_relu_backward_impl(T *z, T *dz, float slope, int64_t count) { - // Create thrust pointers - thrust::device_ptr th_z = thrust::device_pointer_cast(z); - thrust::device_ptr th_dz = thrust::device_pointer_cast(dz); - - thrust::transform_if(th_dz, th_dz + count, th_z, th_dz, - [slope] __device__ (const T& dz) { return dz * slope; }, - [] __device__ (const T& z) { return z < 0; }); - thrust::transform_if(th_z, th_z + count, th_z, - [slope] __device__ (const T& z) { return z / slope; }, - [] __device__ (const T& z) { return z < 0; }); -} - -} - -void LeakyRelu_Forward_CUDA(at::Tensor z, float slope) { - at::leaky_relu_(z, slope); -} - -void LeakyRelu_Backward_CUDA(at::Tensor z, at::Tensor dz, float slope) { - int64_t count = z.numel(); - - AT_DISPATCH_FLOATING_TYPES(z.type(), "LeakyRelu_Backward_CUDA", ([&] { - leaky_relu_backward_impl(z.data(), dz.data(), slope, count); - })); - /* - // unstable after scaling - at::leaky_relu_(z, 1.0 / slope); - at::leaky_relu_backward(dz, z, slope); - */ -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/lib/gpu/common.h b/AIlib/segutils/core/nn/sync_bn/lib/gpu/common.h deleted file mode 100644 index aa38296..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/gpu/common.h +++ /dev/null @@ -1,224 +0,0 @@ -#include -#include - -static const unsigned WARP_SIZE = 32; - -// The maximum number of threads in a block -static const unsigned MAX_BLOCK_SIZE = 512U; - -template -struct ScalarConvert { - static __host__ __device__ __forceinline__ Out to(const In v) { return (Out) v; } -}; - -// Number of threads in a block given an input size up to MAX_BLOCK_SIZE -static int getNumThreads(int nElem) { - int threadSizes[5] = { 32, 64, 128, 256, MAX_BLOCK_SIZE }; - for (int i = 0; i != 5; ++i) { - if (nElem <= threadSizes[i]) { - return threadSizes[i]; - } - } - return MAX_BLOCK_SIZE; -} - -// Returns the index of the most significant 1 bit in `val`. -__device__ __forceinline__ int getMSB(int val) { - return 31 - __clz(val); -} - -template -__device__ __forceinline__ T WARP_SHFL_XOR(T value, int laneMask, int width = warpSize, unsigned int mask = 0xffffffff) -{ -#if CUDA_VERSION >= 9000 - return __shfl_xor_sync(mask, value, laneMask, width); -#else - return __shfl_xor(value, laneMask, width); -#endif -} - -// Sum across all threads within a warp -template -static __device__ __forceinline__ T warpSum(T val) { -#if __CUDA_ARCH__ >= 300 - for (int i = 0; i < getMSB(WARP_SIZE); ++i) { - val += WARP_SHFL_XOR(val, 1 << i, WARP_SIZE); - } -#else - __shared__ T values[MAX_BLOCK_SIZE]; - values[threadIdx.x] = val; - __threadfence_block(); - const int base = (threadIdx.x / WARP_SIZE) * WARP_SIZE; - for (int i = 1; i < WARP_SIZE; i++) { - val += values[base + ((i + threadIdx.x) % WARP_SIZE)]; - } -#endif - return val; -} - -template -struct Float2 { - Acctype v1, v2; - __device__ Float2() {} - __device__ Float2(DType v1, DType v2) : v1(ScalarConvert::to(v1)), v2(ScalarConvert::to(v2)) {} - __device__ Float2(DType v) : v1(ScalarConvert::to(v)), v2(ScalarConvert::to(v)) {} - __device__ Float2(int v) : v1(ScalarConvert::to(v)), v2(ScalarConvert::to(v)) {} - __device__ Float2& operator+=(const Float2& a) { - v1 += a.v1; - v2 += a.v2; - return *this; - } -}; - -template -static __device__ __forceinline__ Float2 warpSum(Float2 value) { - value.v1 = warpSum(value.v1); - value.v2 = warpSum(value.v2); - return value; -} - -template -__device__ T reduceD( - Op op, int b, int i, int k, int D) { - T sum = 0; - for (int x = threadIdx.x; x < D; x += blockDim.x) { - sum += op(b,i,k,x); - } - // sum over NumThreads within a warp - sum = warpSum(sum); - - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - if (threadIdx.x / WARP_SIZE < 32) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T) 0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} - -template -__device__ T reduceN( - Op op, int b, int k, int d, int N) { - T sum = 0; - for (int x = threadIdx.x; x < N; x += blockDim.x) { - sum += op(b,x,k,d); - } - // sum over NumThreads within a warp - sum = warpSum(sum); - - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - if (threadIdx.x / WARP_SIZE < 32) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T) 0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} - -template -__device__ T reduceK( - Op op, int b, int i, int d, int K) { - T sum = 0; - for (int x = threadIdx.x; x < K; x += blockDim.x) { - sum += op(b,i,x,d); - } - // sum over NumThreads within a warp - sum = warpSum(sum); - - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - if (threadIdx.x / WARP_SIZE < 32) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T) 0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} - -template -__device__ T reduceBN( - Op op, - int k, int d, int B, int N) { - T sum = 0; - for (int batch = 0; batch < B; ++batch) { - for (int x = threadIdx.x; x < N; x += blockDim.x) { - sum += op(batch,x,k,d); - } - } - // sum over NumThreads within a warp - sum = warpSum(sum); - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - if (threadIdx.x / WARP_SIZE < 32) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T) 0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/lib/gpu/device_tensor.h b/AIlib/segutils/core/nn/sync_bn/lib/gpu/device_tensor.h deleted file mode 100644 index c67dfae..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/gpu/device_tensor.h +++ /dev/null @@ -1,110 +0,0 @@ -#include - -template -struct DeviceTensor { - public: - inline __device__ __host__ DeviceTensor(DType *p, const int *size) - : dptr_(p) { - for (int i = 0; i < Dim; ++i) { - size_[i] = size ? size[i] : 0; - } - } - - inline __device__ __host__ unsigned getSize(const int i) const { - assert(i < Dim); - return size_[i]; - } - - inline __device__ __host__ int numElements() const { - int n = 1; - for (int i = 0; i < Dim; ++i) { - n *= size_[i]; - } - return n; - } - - inline __device__ __host__ DeviceTensor select(const size_t x) const { - assert(Dim > 1); - int offset = x; - for (int i = 1; i < Dim; ++i) { - offset *= size_[i]; - } - DeviceTensor tensor(dptr_ + offset, nullptr); - for (int i = 0; i < Dim - 1; ++i) { - tensor.size_[i] = this->size_[i+1]; - } - return tensor; - } - - inline __device__ __host__ DeviceTensor operator[](const size_t x) const { - assert(Dim > 1); - int offset = x; - for (int i = 1; i < Dim; ++i) { - offset *= size_[i]; - } - DeviceTensor tensor(dptr_ + offset, nullptr); - for (int i = 0; i < Dim - 1; ++i) { - tensor.size_[i] = this->size_[i+1]; - } - return tensor; - } - - inline __device__ __host__ size_t InnerSize() const { - assert(Dim >= 3); - size_t sz = 1; - for (size_t i = 2; i < Dim; ++i) { - sz *= size_[i]; - } - return sz; - } - - inline __device__ __host__ size_t ChannelCount() const { - assert(Dim >= 3); - return size_[1]; - } - - inline __device__ __host__ DType* data_ptr() const { - return dptr_; - } - - DType *dptr_; - int size_[Dim]; -}; - -template -struct DeviceTensor { - inline __device__ __host__ DeviceTensor(DType *p, const int *size) - : dptr_(p) { - size_[0] = size ? size[0] : 0; - } - - inline __device__ __host__ unsigned getSize(const int i) const { - assert(i == 0); - return size_[0]; - } - - inline __device__ __host__ int numElements() const { - return size_[0]; - } - - inline __device__ __host__ DType &operator[](const size_t x) const { - return *(dptr_ + x); - } - - inline __device__ __host__ DType* data_ptr() const { - return dptr_; - } - - DType *dptr_; - int size_[1]; -}; - -template -static DeviceTensor devicetensor(const at::Tensor &blob) { - DType *data = blob.data(); - DeviceTensor tensor(data, nullptr); - for (int i = 0; i < Dim; ++i) { - tensor.size_[i] = blob.size(i); - } - return tensor; -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/lib/gpu/operator.cpp b/AIlib/segutils/core/nn/sync_bn/lib/gpu/operator.cpp deleted file mode 100644 index 48e28fe..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/gpu/operator.cpp +++ /dev/null @@ -1,13 +0,0 @@ -#include "operator.h" - -PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) { - m.def("batchnorm_forward", &BatchNorm_Forward_CUDA, "BatchNorm forward (CUDA)"); - m.def("batchnorm_inp_forward", &BatchNorm_Forward_Inp_CUDA, "BatchNorm forward (CUDA)"); - m.def("batchnorm_backward", &BatchNorm_Backward_CUDA, "BatchNorm backward (CUDA)"); - m.def("batchnorm_inp_backward", &BatchNorm_Inp_Backward_CUDA, "BatchNorm backward (CUDA)"); - m.def("expectation_forward", &Expectation_Forward_CUDA, "Expectation forward (CUDA)"); - m.def("expectation_backward", &Expectation_Backward_CUDA, "Expectation backward (CUDA)"); - m.def("expectation_inp_backward", &Expectation_Inp_Backward_CUDA, "Inplace Expectation backward (CUDA)"); - m.def("leaky_relu_forward", &LeakyRelu_Forward_CUDA, "Learky ReLU forward (CUDA)"); - m.def("leaky_relu_backward", &LeakyRelu_Backward_CUDA, "Learky ReLU backward (CUDA)"); -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/lib/gpu/operator.h b/AIlib/segutils/core/nn/sync_bn/lib/gpu/operator.h deleted file mode 100644 index 246570d..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/gpu/operator.h +++ /dev/null @@ -1,59 +0,0 @@ -#include -#include - -at::Tensor BatchNorm_Forward_CUDA( - const at::Tensor input_, - const at::Tensor mean_, - const at::Tensor std_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -at::Tensor BatchNorm_Forward_Inp_CUDA( - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -std::vector BatchNorm_Backward_CUDA( - const at::Tensor gradoutput_, - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -std::vector BatchNorm_Inp_Backward_CUDA( - const at::Tensor gradoutput_, - const at::Tensor output_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -std::vector Expectation_Forward_CUDA( - const at::Tensor input_); - -at::Tensor Expectation_Backward_CUDA( - const at::Tensor input_, - const at::Tensor gradEx_, - const at::Tensor gradExs_); - -at::Tensor Expectation_Inp_Backward_CUDA( - const at::Tensor gradInput_, - const at::Tensor output_, - const at::Tensor gradEx_, - const at::Tensor gradExs_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps); - -void LeakyRelu_Forward_CUDA(at::Tensor z, float slope); - -void LeakyRelu_Backward_CUDA(at::Tensor z, at::Tensor dz, float slope); \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/lib/gpu/setup.py b/AIlib/segutils/core/nn/sync_bn/lib/gpu/setup.py deleted file mode 100644 index 14c01f6..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/gpu/setup.py +++ /dev/null @@ -1,15 +0,0 @@ -from setuptools import setup -from torch.utils.cpp_extension import BuildExtension, CUDAExtension - -setup( - name='syncbn_gpu', - ext_modules=[ - CUDAExtension('sync_gpu', [ - 'operator.cpp', - 'activation_kernel.cu', - 'syncbn_kernel.cu', - ]), - ], - cmdclass={ - 'build_ext': BuildExtension - }) \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/lib/gpu/syncbn_kernel.cu b/AIlib/segutils/core/nn/sync_bn/lib/gpu/syncbn_kernel.cu deleted file mode 100644 index 2a7e840..0000000 --- a/AIlib/segutils/core/nn/sync_bn/lib/gpu/syncbn_kernel.cu +++ /dev/null @@ -1,489 +0,0 @@ -#include -// #include -#include -#include -#include - -#include "common.h" -#include "device_tensor.h" - -namespace { - -template -struct GradOp { - __device__ GradOp(Acctype m, const DeviceTensor3 i, const DeviceTensor3 g) - : beta(m), output(i), gradOutput(g) {} - __device__ __forceinline__ Float2 operator()(int batch, int plane, int n) { - DType g = gradOutput[batch][plane][n]; - DType c = ScalarConvert::to(output[batch][plane][n] - beta); - return Float2(g, g * c); - } - const Acctype beta; - const DeviceTensor3 output; - const DeviceTensor3 gradOutput; -}; - -template -struct SumOp { - __device__ SumOp(DeviceTensor i) : input(i){} - __device__ __forceinline__ Float2 operator()(int batch, int plane, int n) { - DType g = input[batch][plane][n]; - return Float2(g, g * g); - } - DType mean; - DeviceTensor input; -}; - -// Sum across (batch, x/y/z) applying Op() pointwise -template -__device__ T reduce(Op op, DeviceTensor3 tensor, int plane) { - T sum = (T)0; - for (int batch = 0; batch < tensor.getSize(0); ++batch) { - for (int x = threadIdx.x; x < tensor.getSize(2); x += blockDim.x) { - sum += op(batch, plane, x); - } - } - - // sum over NumThreads within a warp - sum = warpSum(sum); - - // 'transpose', and reduce within warp again - __shared__ T shared[32]; - __syncthreads(); - if (threadIdx.x % WARP_SIZE == 0) { - shared[threadIdx.x / WARP_SIZE] = sum; - } - if (threadIdx.x >= blockDim.x / WARP_SIZE && threadIdx.x < WARP_SIZE) { - // zero out the other entries in shared - shared[threadIdx.x] = (T)0; - } - __syncthreads(); - if (threadIdx.x / WARP_SIZE == 0) { - sum = warpSum(shared[threadIdx.x]); - if (threadIdx.x == 0) { - shared[0] = sum; - } - } - __syncthreads(); - - // Everyone picks it up, should be broadcast into the whole gradInput - return shared[0]; -} - -template -__global__ void BatchNorm_Forward_kernel ( - DeviceTensor output, - DeviceTensor input, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta) { - int c = blockIdx.x; - /* main operation */ - for (int b = 0; b < input.getSize(0); ++b) { - for (int x = threadIdx.x; x < input.getSize(2); x += blockDim.x) { - DType inp = input[b][c][x]; - output[b][c][x] = gamma[c] * (inp - mean[c]) / - std[c] + beta[c]; - } - } -} - -template -__global__ void BatchNorm_Forward_Inp_kernel ( - DeviceTensor input, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta) { - int c = blockIdx.x; - /* main operation */ - for (int b = 0; b < input.getSize(0); ++b) { - for (int x = threadIdx.x; x < input.getSize(2); x += blockDim.x) { - DType inp = input[b][c][x]; - input[b][c][x] = gamma[c] * (inp - mean[c]) / - std[c] + beta[c]; - } - } -} - -template -__global__ void BatchNorm_Backward_Inp_kernel ( - DeviceTensor gradoutput, - DeviceTensor output, - DeviceTensor gradinput, - DeviceTensor gradgamma, - DeviceTensor gradbeta, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta, - DeviceTensor gradEx, - DeviceTensor gradExs) { - /* declarations of the variables */ - /* Get the index and channels */ - int c = blockIdx.x; - /* main operation */ - GradOp> g(beta[c], output, gradoutput); - Float2 res = reduce, - GradOp>, - DeviceTensor>(g, gradoutput, c); - DType gradOutputSum = res.v1; - DType dotP = res.v2; - DType invstd = DType(1.0) / std[c]; - DType gradScale = invstd * gamma[c]; - if (threadIdx.x == 0) { - gradEx[c] = - gradOutputSum * gradScale + mean[c] * invstd * invstd * dotP; - gradExs[c] = - 0.5 * invstd * invstd * dotP; - } - if (gradinput.numElements() > 0) { - for (int batch = 0; batch < gradoutput.getSize(0); ++batch) { - for (int x = threadIdx.x; x < gradoutput.getSize(2); x += blockDim.x) { - gradinput[batch][c][x] = gradoutput[batch][c][x] * gradScale; - } - } - } - if (gradgamma.numElements() > 0) { - if (threadIdx.x == 0) { - gradgamma[c] += dotP / gamma[c]; - } - } - if (gradbeta.numElements() > 0) { - if (threadIdx.x == 0) { - gradbeta[c] += gradOutputSum; - } - } -} - -template -__global__ void BatchNorm_Backward_kernel ( - DeviceTensor gradoutput, - DeviceTensor input, - DeviceTensor gradinput, - DeviceTensor gradgamma, - DeviceTensor gradbeta, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta, - DeviceTensor gradEx, - DeviceTensor gradExs) { - /* declarations of the variables */ - /* Get the index and channels */ - int c = blockIdx.x; - /* main operation */ - GradOp> g(mean[c], input, gradoutput); - Float2 res = reduce, - GradOp>, - DeviceTensor>(g, gradoutput, c); - DType gradOutputSum = res.v1; - DType dotP = res.v2; - DType invstd = DType(1.0) / std[c]; - DType gradScale = invstd * gamma[c]; - if (threadIdx.x == 0) { - gradEx[c] = - gradOutputSum * gradScale + mean[c] * invstd * invstd * dotP * gradScale; - gradExs[c] = - 0.5 * invstd * invstd * dotP * gradScale; - } - if (gradinput.numElements() > 0) { - for (int batch = 0; batch < gradoutput.getSize(0); ++batch) { - for (int x = threadIdx.x; x < gradoutput.getSize(2); x += blockDim.x) { - gradinput[batch][c][x] = gradoutput[batch][c][x] * gradScale; - } - } - } - if (gradgamma.numElements() > 0) { - if (threadIdx.x == 0) { - gradgamma[c] += dotP * invstd; - } - } - if (gradbeta.numElements() > 0) { - if (threadIdx.x == 0) { - gradbeta[c] += gradOutputSum; - } - } -} - - -template -__global__ void Expectation_Forward_kernel ( - DeviceTensor input, - DeviceTensor ex, - DeviceTensor exs, - DType norm) { - int c = blockIdx.x; - /* main operation */ - SumOp g(input); - Float2 res = reduce, - SumOp, DeviceTensor>(g, input, c); - DType xsum = res.v1; - DType xsquare = res.v2; - if (threadIdx.x == 0) { - ex[c] = xsum * norm; - exs[c] = xsquare * norm; - } -} - -template -__global__ void Expectation_Backward_kernel ( - DeviceTensor gradInput, - DeviceTensor input, - DeviceTensor gradEx, - DeviceTensor gradExs, - DType norm) { - int c = blockIdx.x; - /* main operation */ - for (int batch = 0; batch < gradInput.getSize(0); ++batch) { - for (int x = threadIdx.x; x < gradInput.getSize(2); x += blockDim.x) { - gradInput[batch][c][x] = gradEx[c] * norm + 2 * gradExs[c] * - input[batch][c][x] * norm; - } - } -} - -template -__global__ void Expectation_Backward_Inp_kernel ( - DeviceTensor gradInput, - DeviceTensor output, - DeviceTensor gradEx, - DeviceTensor gradExs, - DeviceTensor mean, - DeviceTensor std, - DeviceTensor gamma, - DeviceTensor beta, - DType norm) { - int c = blockIdx.x; - /* main operation */ - for (int batch = 0; batch < gradInput.getSize(0); ++batch) { - for (int x = threadIdx.x; x < gradInput.getSize(2); x += blockDim.x) { - gradInput[batch][c][x] += gradEx[c] * norm + 2 * gradExs[c] * - ((output[batch][c][x] - beta[c]) / gamma[c] * std[c] + mean[c]) * norm; - } - } -} - -} // namespace - -at::Tensor BatchNorm_Forward_CUDA( - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - auto output_ = at::zeros_like(input_); - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "BatchNorm_Forward_CUDA", ([&] { - /* Device tensors */ - DeviceTensor output = devicetensor(output_); - DeviceTensor input = devicetensor(input_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - /* kernel function */ - BatchNorm_Forward_kernel<<>>( - output, input, ex, std, gamma, beta); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return output_; -} - -at::Tensor BatchNorm_Forward_Inp_CUDA( - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "BatchNorm_Forward_CUDA", ([&] { - /* Device tensors */ - DeviceTensor input = devicetensor(input_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - /* kernel function */ - BatchNorm_Forward_Inp_kernel<<>>( - input, ex, std, gamma, beta); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return input_; -} - - -std::vector BatchNorm_Inp_Backward_CUDA( - const at::Tensor gradoutput_, - const at::Tensor output_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - /* outputs*/ - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - auto gradinput_ = at::zeros_like(output_); - auto gradgamma_ = at::zeros_like(gamma_); - auto gradbeta_ = at::zeros_like(beta_); - auto gradEx_ = at::zeros_like(ex_); - auto gradExs_ = at::zeros_like(std_); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(output_.size(1)); - dim3 threads(getNumThreads(output_.size(2))); - AT_DISPATCH_FLOATING_TYPES(output_.type(), "BatchNorm_Inp_Backward_CUDA", ([&] { - /* Device tensors */ - DeviceTensor gradoutput = devicetensor(gradoutput_); - DeviceTensor output = devicetensor(output_); - DeviceTensor gradinput = devicetensor(gradinput_); - DeviceTensor gradgamma = devicetensor(gradgamma_); - DeviceTensor gradbeta = devicetensor(gradbeta_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - DeviceTensor gradEx = devicetensor(gradEx_); - DeviceTensor gradExs = devicetensor(gradExs_); - /* kernel function */ - BatchNorm_Backward_Inp_kernel - <<>>( - gradoutput, output, gradinput, gradgamma, gradbeta, ex, std, - gamma, beta, gradEx, gradExs); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return {gradinput_, gradEx_, gradExs_, gradgamma_, gradbeta_}; -} - - -std::vector BatchNorm_Backward_CUDA( - const at::Tensor gradoutput_, - const at::Tensor input_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - /* outputs*/ - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - auto gradinput_ = at::zeros_like(input_); - auto gradgamma_ = at::zeros_like(gamma_); - auto gradbeta_ = at::zeros_like(beta_); - auto gradEx_ = at::zeros_like(ex_); - auto gradExs_ = at::zeros_like(std_); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "BatchNorm_Inp_Backward_CUDA", ([&] { - /* Device tensors */ - DeviceTensor gradoutput = devicetensor(gradoutput_); - DeviceTensor input = devicetensor(input_); - DeviceTensor gradinput = devicetensor(gradinput_); - DeviceTensor gradgamma = devicetensor(gradgamma_); - DeviceTensor gradbeta = devicetensor(gradbeta_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - DeviceTensor gradEx = devicetensor(gradEx_); - DeviceTensor gradExs = devicetensor(gradExs_); - /* kernel function */ - BatchNorm_Backward_kernel - <<>>( - gradoutput, input, gradinput, gradgamma, gradbeta, ex, std, - gamma, beta, gradEx, gradExs); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return {gradinput_, gradEx_, gradExs_, gradgamma_, gradbeta_}; -} - -std::vector Expectation_Forward_CUDA( - const at::Tensor input_) { - /* outputs */ - auto ex_ = torch::zeros({input_.size(1)}, input_.options()); - auto exs_ = torch::zeros({input_.size(1)}, input_.options()); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "SumSquare_forward_CUDA", ([&] { - scalar_t norm = scalar_t(1) / (input_.size(0) * input_.size(2)); - /* Device tensors */ - DeviceTensor input = devicetensor(input_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor exs = devicetensor(exs_); - /* kernel function */ - Expectation_Forward_kernel - <<>>(input, ex, exs, norm); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return {ex_, exs_}; -} - -at::Tensor Expectation_Backward_CUDA( - const at::Tensor input_, - const at::Tensor gradEx_, - const at::Tensor gradExs_) { - /* outputs */ - at::Tensor gradInput_ = at::zeros_like(input_); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(input_.size(1)); - dim3 threads(getNumThreads(input_.size(2))); - AT_DISPATCH_FLOATING_TYPES(input_.type(), "SumSquare_Backward_CUDA", ([&] { - scalar_t norm = scalar_t(1) / (input_.size(0) * input_.size(2)); - /* Device tensors */ - DeviceTensor gradInput = devicetensor(gradInput_); - DeviceTensor input = devicetensor(input_); - DeviceTensor gradEx = devicetensor(gradEx_); - DeviceTensor gradExs =devicetensor(gradExs_); - /* kernel function */ - Expectation_Backward_kernel - <<>>(gradInput, input, gradEx, gradExs, norm); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return gradInput_; -} - -at::Tensor Expectation_Inp_Backward_CUDA( - const at::Tensor gradInput_, - const at::Tensor output_, - const at::Tensor gradEx_, - const at::Tensor gradExs_, - const at::Tensor ex_, - const at::Tensor exs_, - const at::Tensor gamma_, - const at::Tensor beta_, - float eps) { - /* outputs */ - //auto gradInput_ = at::zeros_like(output_); - auto std_ = (exs_ - ex_ * ex_ + eps).sqrt(); - /* cuda utils*/ - cudaStream_t stream = at::cuda::getCurrentCUDAStream(); - dim3 blocks(output_.size(1)); - dim3 threads(getNumThreads(output_.size(2))); - AT_DISPATCH_FLOATING_TYPES(output_.type(), "SumSquare_Backward_CUDA", ([&] { - scalar_t norm = scalar_t(1) / (output_.size(0) * output_.size(2)); - /* Device tensors */ - DeviceTensor gradInput = devicetensor(gradInput_); - DeviceTensor input = devicetensor(output_); - DeviceTensor gradEx = devicetensor(gradEx_); - DeviceTensor gradExs =devicetensor(gradExs_); - DeviceTensor ex = devicetensor(ex_); - DeviceTensor std = devicetensor(std_); - DeviceTensor gamma = devicetensor(gamma_); - DeviceTensor beta = devicetensor(beta_); - /* kernel function */ - Expectation_Backward_Inp_kernel - <<>>(gradInput, input, gradEx, gradExs, - ex, std, gamma, beta, norm); - })); - AT_ASSERT(cudaGetLastError() == cudaSuccess); - return gradInput_; -} \ No newline at end of file diff --git a/AIlib/segutils/core/nn/sync_bn/syncbn.py b/AIlib/segutils/core/nn/sync_bn/syncbn.py deleted file mode 100644 index f1247af..0000000 --- a/AIlib/segutils/core/nn/sync_bn/syncbn.py +++ /dev/null @@ -1,124 +0,0 @@ -##+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ -## Created by: Hang Zhang -## ECE Department, Rutgers University -## Email: zhang.hang@rutgers.edu -## Copyright (c) 2017 -## -## This source code is licensed under the MIT-style license found in the -## LICENSE file in the root directory of this source tree -##+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ - -"""Synchronized Cross-GPU Batch Normalization Module""" -import warnings -import torch - -from torch.nn.modules.batchnorm import _BatchNorm -from queue import Queue -from .functions import * - -__all__ = ['SyncBatchNorm', 'BatchNorm1d', 'BatchNorm2d', 'BatchNorm3d'] - - -# Adopt from https://github.com/zhanghang1989/PyTorch-Encoding/blob/master/encoding/nn/syncbn.py -class SyncBatchNorm(_BatchNorm): - """Cross-GPU Synchronized Batch normalization (SyncBN) - - Parameters: - num_features: num_features from an expected input of - size batch_size x num_features x height x width - eps: a value added to the denominator for numerical stability. - Default: 1e-5 - momentum: the value used for the running_mean and running_var - computation. Default: 0.1 - sync: a boolean value that when set to ``True``, synchronize across - different gpus. Default: ``True`` - activation : str - Name of the activation functions, one of: `leaky_relu` or `none`. - slope : float - Negative slope for the `leaky_relu` activation. - - Shape: - - Input: :math:`(N, C, H, W)` - - Output: :math:`(N, C, H, W)` (same shape as input) - Reference: - .. [1] Ioffe, Sergey, and Christian Szegedy. "Batch normalization: Accelerating deep network training by reducing internal covariate shift." *ICML 2015* - .. [2] Hang Zhang, Kristin Dana, Jianping Shi, Zhongyue Zhang, Xiaogang Wang, Ambrish Tyagi, and Amit Agrawal. "Context Encoding for Semantic Segmentation." *CVPR 2018* - Examples: - >>> m = SyncBatchNorm(100) - >>> net = torch.nn.DataParallel(m) - >>> output = net(input) - """ - - def __init__(self, num_features, eps=1e-5, momentum=0.1, sync=True, activation='none', slope=0.01, inplace=True): - super(SyncBatchNorm, self).__init__(num_features, eps=eps, momentum=momentum, affine=True) - self.activation = activation - self.inplace = False if activation == 'none' else inplace - self.slope = slope - self.devices = list(range(torch.cuda.device_count())) - self.sync = sync if len(self.devices) > 1 else False - # Initialize queues - self.worker_ids = self.devices[1:] - self.master_queue = Queue(len(self.worker_ids)) - self.worker_queues = [Queue(1) for _ in self.worker_ids] - - def forward(self, x): - # resize the input to (B, C, -1) - input_shape = x.size() - x = x.view(input_shape[0], self.num_features, -1) - if x.get_device() == self.devices[0]: - # Master mode - extra = { - "is_master": True, - "master_queue": self.master_queue, - "worker_queues": self.worker_queues, - "worker_ids": self.worker_ids - } - else: - # Worker mode - extra = { - "is_master": False, - "master_queue": self.master_queue, - "worker_queue": self.worker_queues[self.worker_ids.index(x.get_device())] - } - if self.inplace: - return inp_syncbatchnorm(x, self.weight, self.bias, self.running_mean, self.running_var, - extra, self.sync, self.training, self.momentum, self.eps, - self.activation, self.slope).view(input_shape) - else: - return syncbatchnorm(x, self.weight, self.bias, self.running_mean, self.running_var, - extra, self.sync, self.training, self.momentum, self.eps, - self.activation, self.slope).view(input_shape) - - def extra_repr(self): - if self.activation == 'none': - return 'sync={}'.format(self.sync) - else: - return 'sync={}, act={}, slope={}, inplace={}'.format( - self.sync, self.activation, self.slope, self.inplace) - - -class BatchNorm1d(SyncBatchNorm): - """BatchNorm1d is deprecated in favor of :class:`core.nn.sync_bn.SyncBatchNorm`.""" - - def __init__(self, *args, **kwargs): - warnings.warn("core.nn.sync_bn.{} is now deprecated in favor of core.nn.sync_bn.{}." - .format('BatchNorm1d', SyncBatchNorm.__name__), DeprecationWarning) - super(BatchNorm1d, self).__init__(*args, **kwargs) - - -class BatchNorm2d(SyncBatchNorm): - """BatchNorm1d is deprecated in favor of :class:`core.nn.sync_bn.SyncBatchNorm`.""" - - def __init__(self, *args, **kwargs): - warnings.warn("core.nn.sync_bn.{} is now deprecated in favor of core.nn.sync_bn.{}." - .format('BatchNorm2d', SyncBatchNorm.__name__), DeprecationWarning) - super(BatchNorm2d, self).__init__(*args, **kwargs) - - -class BatchNorm3d(SyncBatchNorm): - """BatchNorm1d is deprecated in favor of :class:`core.nn.sync_bn.SyncBatchNorm`.""" - - def __init__(self, *args, **kwargs): - warnings.warn("core.nn.sync_bn.{} is now deprecated in favor of core.nn.sync_bn.{}." - .format('BatchNorm3d', SyncBatchNorm.__name__), DeprecationWarning) - super(BatchNorm3d, self).__init__(*args, **kwargs) diff --git a/AIlib/segutils/core/nn/syncbn.py b/AIlib/segutils/core/nn/syncbn.py deleted file mode 100644 index c52ec1a..0000000 --- a/AIlib/segutils/core/nn/syncbn.py +++ /dev/null @@ -1,223 +0,0 @@ -# Adopt from https://github.com/zhanghang1989/PyTorch-Encoding/blob/master/encoding/nn/syncbn.py -"""Synchronized Cross-GPU Batch Normalization Module""" -import warnings -import torch -import torch.cuda.comm as comm - -from queue import Queue -from torch.autograd import Function -from torch.nn.modules.batchnorm import _BatchNorm -from torch.autograd.function import once_differentiable -from core.nn import _C - -__all__ = ['SyncBatchNorm', 'BatchNorm1d', 'BatchNorm2d', 'BatchNorm3d'] - - -class _SyncBatchNorm(Function): - @classmethod - def forward(cls, ctx, x, gamma, beta, running_mean, running_var, - extra, sync=True, training=True, momentum=0.1, eps=1e-05, - activation="none", slope=0.01): - # save context - cls._parse_extra(ctx, extra) - ctx.sync = sync - ctx.training = training - ctx.momentum = momentum - ctx.eps = eps - ctx.activation = activation - ctx.slope = slope - assert activation == 'none' - - # continous inputs - x = x.contiguous() - gamma = gamma.contiguous() - beta = beta.contiguous() - - if ctx.training: - _ex, _exs = _C.expectation_forward(x) - - if ctx.sync: - if ctx.is_master: - _ex, _exs = [_ex.unsqueeze(0)], [_exs.unsqueeze(0)] - for _ in range(ctx.master_queue.maxsize): - _ex_w, _exs_w = ctx.master_queue.get() - ctx.master_queue.task_done() - _ex.append(_ex_w.unsqueeze(0)) - _exs.append(_exs_w.unsqueeze(0)) - - _ex = comm.gather(_ex).mean(0) - _exs = comm.gather(_exs).mean(0) - - tensors = comm.broadcast_coalesced((_ex, _exs), [_ex.get_device()] + ctx.worker_ids) - for ts, queue in zip(tensors[1:], ctx.worker_queues): - queue.put(ts) - else: - ctx.master_queue.put((_ex, _exs)) - _ex, _exs = ctx.worker_queue.get() - ctx.worker_queue.task_done() - - # Update running stats - _var = _exs - _ex ** 2 - running_mean.mul_((1 - ctx.momentum)).add_(ctx.momentum * _ex) - running_var.mul_((1 - ctx.momentum)).add_(ctx.momentum * _var) - - # Mark in-place modified tensors - ctx.mark_dirty(running_mean, running_var) - else: - _ex, _var = running_mean.contiguous(), running_var.contiguous() - _exs = _var + _ex ** 2 - - # BN forward - y = _C.batchnorm_forward(x, _ex, _exs, gamma, beta, ctx.eps) - - # Output - ctx.save_for_backward(x, _ex, _exs, gamma, beta) - return y - - @staticmethod - @once_differentiable - def backward(ctx, dz): - x, _ex, _exs, gamma, beta = ctx.saved_tensors - dz = dz.contiguous() - - # BN backward - dx, _dex, _dexs, dgamma, dbeta = _C.batchnorm_backward(dz, x, _ex, _exs, gamma, beta, ctx.eps) - - if ctx.training: - if ctx.sync: - if ctx.is_master: - _dex, _dexs = [_dex.unsqueeze(0)], [_dexs.unsqueeze(0)] - for _ in range(ctx.master_queue.maxsize): - _dex_w, _dexs_w = ctx.master_queue.get() - ctx.master_queue.task_done() - _dex.append(_dex_w.unsqueeze(0)) - _dexs.append(_dexs_w.unsqueeze(0)) - - _dex = comm.gather(_dex).mean(0) - _dexs = comm.gather(_dexs).mean(0) - - tensors = comm.broadcast_coalesced((_dex, _dexs), [_dex.get_device()] + ctx.worker_ids) - for ts, queue in zip(tensors[1:], ctx.worker_queues): - queue.put(ts) - else: - ctx.master_queue.put((_dex, _dexs)) - _dex, _dexs = ctx.worker_queue.get() - ctx.worker_queue.task_done() - - dx_ = _C.expectation_backward(x, _dex, _dexs) - dx = dx + dx_ - - return dx, dgamma, dbeta, None, None, None, None, None, None, None, None, None - - @staticmethod - def _parse_extra(ctx, extra): - ctx.is_master = extra["is_master"] - if ctx.is_master: - ctx.master_queue = extra["master_queue"] - ctx.worker_queues = extra["worker_queues"] - ctx.worker_ids = extra["worker_ids"] - else: - ctx.master_queue = extra["master_queue"] - ctx.worker_queue = extra["worker_queue"] - - -syncbatchnorm = _SyncBatchNorm.apply - - -class SyncBatchNorm(_BatchNorm): - """Cross-GPU Synchronized Batch normalization (SyncBN) - - Parameters: - num_features: num_features from an expected input of - size batch_size x num_features x height x width - eps: a value added to the denominator for numerical stability. - Default: 1e-5 - momentum: the value used for the running_mean and running_var - computation. Default: 0.1 - sync: a boolean value that when set to ``True``, synchronize across - different gpus. Default: ``True`` - activation : str - Name of the activation functions, one of: `leaky_relu` or `none`. - slope : float - Negative slope for the `leaky_relu` activation. - - Shape: - - Input: :math:`(N, C, H, W)` - - Output: :math:`(N, C, H, W)` (same shape as input) - Reference: - .. [1] Ioffe, Sergey, and Christian Szegedy. "Batch normalization: Accelerating deep network training by reducing internal covariate shift." *ICML 2015* - .. [2] Hang Zhang, Kristin Dana, Jianping Shi, Zhongyue Zhang, Xiaogang Wang, Ambrish Tyagi, and Amit Agrawal. "Context Encoding for Semantic Segmentation." *CVPR 2018* - Examples: - >>> m = SyncBatchNorm(100) - >>> net = torch.nn.DataParallel(m) - >>> output = net(input) - """ - - def __init__(self, num_features, eps=1e-5, momentum=0.1, sync=True, activation='none', slope=0.01): - super(SyncBatchNorm, self).__init__(num_features, eps=eps, momentum=momentum, affine=True) - self.activation = activation - self.slope = slope - self.devices = list(range(torch.cuda.device_count())) - self.sync = sync if len(self.devices) > 1 else False - # Initialize queues - self.worker_ids = self.devices[1:] - self.master_queue = Queue(len(self.worker_ids)) - self.worker_queues = [Queue(1) for _ in self.worker_ids] - - def forward(self, x): - # resize the input to (B, C, -1) - input_shape = x.size() - x = x.view(input_shape[0], self.num_features, -1) - if x.get_device() == self.devices[0]: - # Master mode - extra = { - "is_master": True, - "master_queue": self.master_queue, - "worker_queues": self.worker_queues, - "worker_ids": self.worker_ids - } - else: - # Worker mode - extra = { - "is_master": False, - "master_queue": self.master_queue, - "worker_queue": self.worker_queues[self.worker_ids.index(x.get_device())] - } - - return syncbatchnorm(x, self.weight, self.bias, self.running_mean, self.running_var, - extra, self.sync, self.training, self.momentum, self.eps, - self.activation, self.slope).view(input_shape) - - def extra_repr(self): - if self.activation == 'none': - return 'sync={}'.format(self.sync) - else: - return 'sync={}, act={}, slope={}'.format( - self.sync, self.activation, self.slope) - - -class BatchNorm1d(SyncBatchNorm): - """BatchNorm1d is deprecated in favor of :class:`core.nn.sync_bn.SyncBatchNorm`.""" - - def __init__(self, *args, **kwargs): - warnings.warn("core.nn.sync_bn.{} is now deprecated in favor of core.nn.sync_bn.{}." - .format('BatchNorm1d', SyncBatchNorm.__name__), DeprecationWarning) - super(BatchNorm1d, self).__init__(*args, **kwargs) - - -class BatchNorm2d(SyncBatchNorm): - """BatchNorm1d is deprecated in favor of :class:`core.nn.sync_bn.SyncBatchNorm`.""" - - def __init__(self, *args, **kwargs): - warnings.warn("core.nn.sync_bn.{} is now deprecated in favor of core.nn.sync_bn.{}." - .format('BatchNorm2d', SyncBatchNorm.__name__), DeprecationWarning) - super(BatchNorm2d, self).__init__(*args, **kwargs) - - -class BatchNorm3d(SyncBatchNorm): - """BatchNorm1d is deprecated in favor of :class:`core.nn.sync_bn.SyncBatchNorm`.""" - - def __init__(self, *args, **kwargs): - warnings.warn("core.nn.sync_bn.{} is now deprecated in favor of core.nn.sync_bn.{}." - .format('BatchNorm3d', SyncBatchNorm.__name__), DeprecationWarning) - super(BatchNorm3d, self).__init__(*args, **kwargs) diff --git a/AIlib/segutils/core/utils/__init__.py b/AIlib/segutils/core/utils/__init__.py deleted file mode 100644 index 067a8d0..0000000 --- a/AIlib/segutils/core/utils/__init__.py +++ /dev/null @@ -1,5 +0,0 @@ -"""Utility functions.""" -from __future__ import absolute_import - -from .download import download, check_sha1 -from .filesystem import makedirs, try_import_pycocotools diff --git a/AIlib/segutils/core/utils/__pycache__/__init__.cpython-36.pyc b/AIlib/segutils/core/utils/__pycache__/__init__.cpython-36.pyc deleted file mode 100644 index 94e930b..0000000 Binary files a/AIlib/segutils/core/utils/__pycache__/__init__.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/utils/__pycache__/__init__.cpython-38.pyc b/AIlib/segutils/core/utils/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index daec83b..0000000 Binary files a/AIlib/segutils/core/utils/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/utils/__pycache__/distributed.cpython-36.pyc b/AIlib/segutils/core/utils/__pycache__/distributed.cpython-36.pyc deleted file mode 100644 index d8b7f1d..0000000 Binary files a/AIlib/segutils/core/utils/__pycache__/distributed.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/utils/__pycache__/download.cpython-36.pyc b/AIlib/segutils/core/utils/__pycache__/download.cpython-36.pyc deleted file mode 100644 index 26e9a4a..0000000 Binary files a/AIlib/segutils/core/utils/__pycache__/download.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/utils/__pycache__/download.cpython-38.pyc b/AIlib/segutils/core/utils/__pycache__/download.cpython-38.pyc deleted file mode 100644 index d26b696..0000000 Binary files a/AIlib/segutils/core/utils/__pycache__/download.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/utils/__pycache__/filesystem.cpython-36.pyc b/AIlib/segutils/core/utils/__pycache__/filesystem.cpython-36.pyc deleted file mode 100644 index 72efcec..0000000 Binary files a/AIlib/segutils/core/utils/__pycache__/filesystem.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/utils/__pycache__/filesystem.cpython-38.pyc b/AIlib/segutils/core/utils/__pycache__/filesystem.cpython-38.pyc deleted file mode 100644 index 68e18d4..0000000 Binary files a/AIlib/segutils/core/utils/__pycache__/filesystem.cpython-38.pyc and /dev/null differ diff --git a/AIlib/segutils/core/utils/__pycache__/loss.cpython-36.pyc b/AIlib/segutils/core/utils/__pycache__/loss.cpython-36.pyc deleted file mode 100644 index 75fab07..0000000 Binary files a/AIlib/segutils/core/utils/__pycache__/loss.cpython-36.pyc and /dev/null differ diff --git a/AIlib/segutils/core/utils/distributed.py b/AIlib/segutils/core/utils/distributed.py deleted file mode 100644 index 257cdf9..0000000 --- a/AIlib/segutils/core/utils/distributed.py +++ /dev/null @@ -1,258 +0,0 @@ -""" -This file contains primitives for multi-gpu communication. -This is useful when doing distributed training. -""" -import math -import pickle -import torch -import torch.utils.data as data -import torch.distributed as dist - -from torch.utils.data.sampler import Sampler, BatchSampler - -__all__ = ['get_world_size', 'get_rank', 'synchronize', 'is_main_process', - 'all_gather', 'make_data_sampler', 'make_batch_data_sampler', - 'reduce_dict', 'reduce_loss_dict'] - - -# reference: https://github.com/facebookresearch/maskrcnn-benchmark/blob/master/maskrcnn_benchmark/utils/comm.py -def get_world_size(): - if not dist.is_available(): - return 1 - if not dist.is_initialized(): - return 1 - return dist.get_world_size() - - -def get_rank(): - if not dist.is_available(): - return 0 - if not dist.is_initialized(): - return 0 - return dist.get_rank() - - -def is_main_process(): - return get_rank() == 0 - - -def synchronize(): - """ - Helper function to synchronize (barrier) among all processes when - using distributed training - """ - if not dist.is_available(): - return - if not dist.is_initialized(): - return - world_size = dist.get_world_size() - if world_size == 1: - return - dist.barrier() - - -def all_gather(data): - """ - Run all_gather on arbitrary picklable data (not necessarily tensors) - Args: - data: any picklable object - Returns: - list[data]: list of data gathered from each rank - """ - world_size = get_world_size() - if world_size == 1: - return [data] - - # serialized to a Tensor - buffer = pickle.dumps(data) - storage = torch.ByteStorage.from_buffer(buffer) - tensor = torch.ByteTensor(storage).to("cuda") - - # obtain Tensor size of each rank - local_size = torch.IntTensor([tensor.numel()]).to("cuda") - size_list = [torch.IntTensor([0]).to("cuda") for _ in range(world_size)] - dist.all_gather(size_list, local_size) - size_list = [int(size.item()) for size in size_list] - max_size = max(size_list) - - # receiving Tensor from all ranks - # we pad the tensor because torch all_gather does not support - # gathering tensors of different shapes - tensor_list = [] - for _ in size_list: - tensor_list.append(torch.ByteTensor(size=(max_size,)).to("cuda")) - if local_size != max_size: - padding = torch.ByteTensor(size=(max_size - local_size,)).to("cuda") - tensor = torch.cat((tensor, padding), dim=0) - dist.all_gather(tensor_list, tensor) - - data_list = [] - for size, tensor in zip(size_list, tensor_list): - buffer = tensor.cpu().numpy().tobytes()[:size] - data_list.append(pickle.loads(buffer)) - - return data_list - - -def reduce_dict(input_dict, average=True): - """ - Args: - input_dict (dict): all the values will be reduced - average (bool): whether to do average or sum - Reduce the values in the dictionary from all processes so that process with rank - 0 has the averaged results. Returns a dict with the same fields as - input_dict, after reduction. - """ - world_size = get_world_size() - if world_size < 2: - return input_dict - with torch.no_grad(): - names = [] - values = [] - # sort the keys so that they are consistent across processes - for k in sorted(input_dict.keys()): - names.append(k) - values.append(input_dict[k]) - values = torch.stack(values, dim=0) - dist.reduce(values, dst=0) - if dist.get_rank() == 0 and average: - # only main process gets accumulated, so only divide by - # world_size in this case - values /= world_size - reduced_dict = {k: v for k, v in zip(names, values)} - return reduced_dict - - -def reduce_loss_dict(loss_dict): - """ - Reduce the loss dictionary from all processes so that process with rank - 0 has the averaged results. Returns a dict with the same fields as - loss_dict, after reduction. - """ - world_size = get_world_size() - if world_size < 2: - return loss_dict - with torch.no_grad(): - loss_names = [] - all_losses = [] - for k in sorted(loss_dict.keys()): - loss_names.append(k) - all_losses.append(loss_dict[k]) - all_losses = torch.stack(all_losses, dim=0) - dist.reduce(all_losses, dst=0) - if dist.get_rank() == 0: - # only main process gets accumulated, so only divide by - # world_size in this case - all_losses /= world_size - reduced_losses = {k: v for k, v in zip(loss_names, all_losses)} - return reduced_losses - - -def make_data_sampler(dataset, shuffle, distributed): - if distributed: - return DistributedSampler(dataset, shuffle=shuffle) - if shuffle: - sampler = data.sampler.RandomSampler(dataset) - else: - sampler = data.sampler.SequentialSampler(dataset) - return sampler - - -def make_batch_data_sampler(sampler, images_per_batch, num_iters=None, start_iter=0): - batch_sampler = data.sampler.BatchSampler(sampler, images_per_batch, drop_last=True) - if num_iters is not None: - batch_sampler = IterationBasedBatchSampler(batch_sampler, num_iters, start_iter) - return batch_sampler - - -# Code is copy-pasted from https://github.com/facebookresearch/maskrcnn-benchmark/blob/master/maskrcnn_benchmark/data/samplers/distributed.py -class DistributedSampler(Sampler): - """Sampler that restricts data loading to a subset of the dataset. - It is especially useful in conjunction with - :class:`torch.nn.parallel.DistributedDataParallel`. In such case, each - process can pass a DistributedSampler instance as a DataLoader sampler, - and load a subset of the original dataset that is exclusive to it. - .. note:: - Dataset is assumed to be of constant size. - Arguments: - dataset: Dataset used for sampling. - num_replicas (optional): Number of processes participating in - distributed training. - rank (optional): Rank of the current process within num_replicas. - """ - - def __init__(self, dataset, num_replicas=None, rank=None, shuffle=True): - if num_replicas is None: - if not dist.is_available(): - raise RuntimeError("Requires distributed package to be available") - num_replicas = dist.get_world_size() - if rank is None: - if not dist.is_available(): - raise RuntimeError("Requires distributed package to be available") - rank = dist.get_rank() - self.dataset = dataset - self.num_replicas = num_replicas - self.rank = rank - self.epoch = 0 - self.num_samples = int(math.ceil(len(self.dataset) * 1.0 / self.num_replicas)) - self.total_size = self.num_samples * self.num_replicas - self.shuffle = shuffle - - def __iter__(self): - if self.shuffle: - # deterministically shuffle based on epoch - g = torch.Generator() - g.manual_seed(self.epoch) - indices = torch.randperm(len(self.dataset), generator=g).tolist() - else: - indices = torch.arange(len(self.dataset)).tolist() - - # add extra samples to make it evenly divisible - indices += indices[: (self.total_size - len(indices))] - assert len(indices) == self.total_size - - # subsample - offset = self.num_samples * self.rank - indices = indices[offset: offset + self.num_samples] - assert len(indices) == self.num_samples - - return iter(indices) - - def __len__(self): - return self.num_samples - - def set_epoch(self, epoch): - self.epoch = epoch - - -class IterationBasedBatchSampler(BatchSampler): - """ - Wraps a BatchSampler, resampling from it until - a specified number of iterations have been sampled - """ - - def __init__(self, batch_sampler, num_iterations, start_iter=0): - self.batch_sampler = batch_sampler - self.num_iterations = num_iterations - self.start_iter = start_iter - - def __iter__(self): - iteration = self.start_iter - while iteration <= self.num_iterations: - # if the underlying sampler has a set_epoch method, like - # DistributedSampler, used for making each process see - # a different split of the dataset, then set it - if hasattr(self.batch_sampler.sampler, "set_epoch"): - self.batch_sampler.sampler.set_epoch(iteration) - for batch in self.batch_sampler: - iteration += 1 - if iteration > self.num_iterations: - break - yield batch - - def __len__(self): - return self.num_iterations - - -if __name__ == '__main__': - pass diff --git a/AIlib/segutils/core/utils/download.py b/AIlib/segutils/core/utils/download.py deleted file mode 100644 index fec8bb4..0000000 --- a/AIlib/segutils/core/utils/download.py +++ /dev/null @@ -1,88 +0,0 @@ -"""Download files with progress bar.""" -import os -import hashlib -import requests -from tqdm import tqdm - -def check_sha1(filename, sha1_hash): - """Check whether the sha1 hash of the file content matches the expected hash. - Parameters - ---------- - filename : str - Path to the file. - sha1_hash : str - Expected sha1 hash in hexadecimal digits. - Returns - ------- - bool - Whether the file content matches the expected hash. - """ - sha1 = hashlib.sha1() - with open(filename, 'rb') as f: - while True: - data = f.read(1048576) - if not data: - break - sha1.update(data) - - sha1_file = sha1.hexdigest() - l = min(len(sha1_file), len(sha1_hash)) - return sha1.hexdigest()[0:l] == sha1_hash[0:l] - -def download(url, path=None, overwrite=False, sha1_hash=None): - """Download an given URL - Parameters - ---------- - url : str - URL to download - path : str, optional - Destination path to store downloaded file. By default stores to the - current directory with same name as in url. - overwrite : bool, optional - Whether to overwrite destination file if already exists. - sha1_hash : str, optional - Expected sha1 hash in hexadecimal digits. Will ignore existing file when hash is specified - but doesn't match. - Returns - ------- - str - The file path of the downloaded file. - """ - if path is None: - fname = url.split('/')[-1] - else: - path = os.path.expanduser(path) - if os.path.isdir(path): - fname = os.path.join(path, url.split('/')[-1]) - else: - fname = path - - if overwrite or not os.path.exists(fname) or (sha1_hash and not check_sha1(fname, sha1_hash)): - dirname = os.path.dirname(os.path.abspath(os.path.expanduser(fname))) - if not os.path.exists(dirname): - os.makedirs(dirname) - - print('Downloading %s from %s...'%(fname, url)) - r = requests.get(url, stream=True) - if r.status_code != 200: - raise RuntimeError("Failed downloading url %s"%url) - total_length = r.headers.get('content-length') - with open(fname, 'wb') as f: - if total_length is None: # no content length header - for chunk in r.iter_content(chunk_size=1024): - if chunk: # filter out keep-alive new chunks - f.write(chunk) - else: - total_length = int(total_length) - for chunk in tqdm(r.iter_content(chunk_size=1024), - total=int(total_length / 1024. + 0.5), - unit='KB', unit_scale=False, dynamic_ncols=True): - f.write(chunk) - - if sha1_hash and not check_sha1(fname, sha1_hash): - raise UserWarning('File {} is downloaded but the content hash does not match. ' \ - 'The repo may be outdated or download may be incomplete. ' \ - 'If the "repo_url" is overridden, consider switching to ' \ - 'the default repo.'.format(fname)) - - return fname \ No newline at end of file diff --git a/AIlib/segutils/core/utils/filesystem.py b/AIlib/segutils/core/utils/filesystem.py deleted file mode 100644 index ab2510d..0000000 --- a/AIlib/segutils/core/utils/filesystem.py +++ /dev/null @@ -1,123 +0,0 @@ -"""Filesystem utility functions.""" -from __future__ import absolute_import -import os -import errno - - -def makedirs(path): - """Create directory recursively if not exists. - Similar to `makedir -p`, you can skip checking existence before this function. - Parameters - ---------- - path : str - Path of the desired dir - """ - try: - os.makedirs(path) - except OSError as exc: - if exc.errno != errno.EEXIST: - raise - - -def try_import(package, message=None): - """Try import specified package, with custom message support. - Parameters - ---------- - package : str - The name of the targeting package. - message : str, default is None - If not None, this function will raise customized error message when import error is found. - Returns - ------- - module if found, raise ImportError otherwise - """ - try: - return __import__(package) - except ImportError as e: - if not message: - raise e - raise ImportError(message) - - -def try_import_cv2(): - """Try import cv2 at runtime. - Returns - ------- - cv2 module if found. Raise ImportError otherwise - """ - msg = "cv2 is required, you can install by package manager, e.g. 'apt-get', \ - or `pip install opencv-python --user` (note that this is unofficial PYPI package)." - return try_import('cv2', msg) - - -def import_try_install(package, extern_url=None): - """Try import the specified package. - If the package not installed, try use pip to install and import if success. - Parameters - ---------- - package : str - The name of the package trying to import. - extern_url : str or None, optional - The external url if package is not hosted on PyPI. - For example, you can install a package using: - "pip install git+http://github.com/user/repo/tarball/master/egginfo=xxx". - In this case, you can pass the url to the extern_url. - Returns - ------- - - The imported python module. - """ - try: - return __import__(package) - except ImportError: - try: - from pip import main as pipmain - except ImportError: - from pip._internal import main as pipmain - - # trying to install package - url = package if extern_url is None else extern_url - pipmain(['install', '--user', url]) # will raise SystemExit Error if fails - - # trying to load again - try: - return __import__(package) - except ImportError: - import sys - import site - user_site = site.getusersitepackages() - if user_site not in sys.path: - sys.path.append(user_site) - return __import__(package) - return __import__(package) - - -"""Import helper for pycocotools""" - - -# NOTE: for developers -# please do not import any pycocotools in __init__ because we are trying to lazy -# import pycocotools to avoid install it for other users who may not use it. -# only import when you actually use it - - -def try_import_pycocotools(): - """Tricks to optionally install and import pycocotools""" - # first we can try import pycocotools - try: - import pycocotools as _ - except ImportError: - import os - # we need to install pycootools, which is a bit tricky - # pycocotools sdist requires Cython, numpy(already met) - import_try_install('cython') - # pypi pycocotools is not compatible with windows - win_url = 'git+https://github.com/zhreshold/cocoapi.git#subdirectory=PythonAPI' - try: - if os.name == 'nt': - import_try_install('pycocotools', win_url) - else: - import_try_install('pycocotools') - except ImportError: - faq = 'cocoapi FAQ' - raise ImportError('Cannot import or install pycocotools, please refer to %s.' % faq) diff --git a/AIlib/segutils/core/utils/logger.py b/AIlib/segutils/core/utils/logger.py deleted file mode 100644 index a2de227..0000000 --- a/AIlib/segutils/core/utils/logger.py +++ /dev/null @@ -1,30 +0,0 @@ -# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved. -import logging -import os -import sys - -__all__ = ['setup_logger'] - - -# reference from: https://github.com/facebookresearch/maskrcnn-benchmark/blob/master/maskrcnn_benchmark/utils/logger.py -def setup_logger(name, save_dir, distributed_rank, filename="log.txt", mode='w'): - logger = logging.getLogger(name) - logger.setLevel(logging.DEBUG) - # don't log results for the non-master process - if distributed_rank > 0: - return logger - ch = logging.StreamHandler(stream=sys.stdout) - ch.setLevel(logging.DEBUG) - formatter = logging.Formatter("%(asctime)s %(name)s %(levelname)s: %(message)s") - ch.setFormatter(formatter) - logger.addHandler(ch) - - if save_dir: - if not os.path.exists(save_dir): - os.makedirs(save_dir) - fh = logging.FileHandler(os.path.join(save_dir, filename), mode=mode) # 'a+' for add, 'w' for overwrite - fh.setLevel(logging.DEBUG) - fh.setFormatter(formatter) - logger.addHandler(fh) - - return logger diff --git a/AIlib/segutils/core/utils/loss.py b/AIlib/segutils/core/utils/loss.py deleted file mode 100644 index aab5314..0000000 --- a/AIlib/segutils/core/utils/loss.py +++ /dev/null @@ -1,196 +0,0 @@ -"""Custom losses.""" -import torch -import torch.nn as nn -import torch.nn.functional as F - -from torch.autograd import Variable - -__all__ = ['MixSoftmaxCrossEntropyLoss', 'MixSoftmaxCrossEntropyOHEMLoss', - 'EncNetLoss', 'ICNetLoss', 'get_segmentation_loss'] - - -# TODO: optim function -class MixSoftmaxCrossEntropyLoss(nn.CrossEntropyLoss): - def __init__(self, aux=True, aux_weight=0.2, ignore_index=-1, **kwargs): - super(MixSoftmaxCrossEntropyLoss, self).__init__(ignore_index=ignore_index) - self.aux = aux - self.aux_weight = aux_weight - - def _aux_forward(self, *inputs, **kwargs): - *preds, target = tuple(inputs) - - loss = super(MixSoftmaxCrossEntropyLoss, self).forward(preds[0], target) - for i in range(1, len(preds)): - aux_loss = super(MixSoftmaxCrossEntropyLoss, self).forward(preds[i], target) - loss += self.aux_weight * aux_loss - return loss - - def forward(self, *inputs, **kwargs): - preds, target = tuple(inputs) - inputs = tuple(list(preds) + [target]) - if self.aux: - return dict(loss=self._aux_forward(*inputs)) - else: - return dict(loss=super(MixSoftmaxCrossEntropyLoss, self).forward(*inputs)) - - -# reference: https://github.com/zhanghang1989/PyTorch-Encoding/blob/master/encoding/nn/loss.py -class EncNetLoss(nn.CrossEntropyLoss): - """2D Cross Entropy Loss with SE Loss""" - - def __init__(self, se_loss=True, se_weight=0.2, nclass=19, aux=False, - aux_weight=0.4, weight=None, ignore_index=-1, **kwargs): - super(EncNetLoss, self).__init__(weight, None, ignore_index) - self.se_loss = se_loss - self.aux = aux - self.nclass = nclass - self.se_weight = se_weight - self.aux_weight = aux_weight - self.bceloss = nn.BCELoss(weight) - - def forward(self, *inputs): - preds, target = tuple(inputs) - inputs = tuple(list(preds) + [target]) - if not self.se_loss and not self.aux: - return super(EncNetLoss, self).forward(*inputs) - elif not self.se_loss: - pred1, pred2, target = tuple(inputs) - loss1 = super(EncNetLoss, self).forward(pred1, target) - loss2 = super(EncNetLoss, self).forward(pred2, target) - return dict(loss=loss1 + self.aux_weight * loss2) - elif not self.aux: - print (inputs) - pred, se_pred, target = tuple(inputs) - se_target = self._get_batch_label_vector(target, nclass=self.nclass).type_as(pred) - loss1 = super(EncNetLoss, self).forward(pred, target) - loss2 = self.bceloss(torch.sigmoid(se_pred), se_target) - return dict(loss=loss1 + self.se_weight * loss2) - else: - pred1, se_pred, pred2, target = tuple(inputs) - se_target = self._get_batch_label_vector(target, nclass=self.nclass).type_as(pred1) - loss1 = super(EncNetLoss, self).forward(pred1, target) - loss2 = super(EncNetLoss, self).forward(pred2, target) - loss3 = self.bceloss(torch.sigmoid(se_pred), se_target) - return dict(loss=loss1 + self.aux_weight * loss2 + self.se_weight * loss3) - - @staticmethod - def _get_batch_label_vector(target, nclass): - # target is a 3D Variable BxHxW, output is 2D BxnClass - batch = target.size(0) - tvect = Variable(torch.zeros(batch, nclass)) - for i in range(batch): - hist = torch.histc(target[i].cpu().data.float(), - bins=nclass, min=0, - max=nclass - 1) - vect = hist > 0 - tvect[i] = vect - return tvect - - -# TODO: optim function -class ICNetLoss(nn.CrossEntropyLoss): - """Cross Entropy Loss for ICNet""" - - def __init__(self, nclass, aux_weight=0.4, ignore_index=-1, **kwargs): - super(ICNetLoss, self).__init__(ignore_index=ignore_index) - self.nclass = nclass - self.aux_weight = aux_weight - - def forward(self, *inputs): - preds, target = tuple(inputs) - inputs = tuple(list(preds) + [target]) - - pred, pred_sub4, pred_sub8, pred_sub16, target = tuple(inputs) - # [batch, W, H] -> [batch, 1, W, H] - target = target.unsqueeze(1).float() - target_sub4 = F.interpolate(target, pred_sub4.size()[2:], mode='bilinear', align_corners=True).squeeze(1).long() - target_sub8 = F.interpolate(target, pred_sub8.size()[2:], mode='bilinear', align_corners=True).squeeze(1).long() - target_sub16 = F.interpolate(target, pred_sub16.size()[2:], mode='bilinear', align_corners=True).squeeze( - 1).long() - loss1 = super(ICNetLoss, self).forward(pred_sub4, target_sub4) - loss2 = super(ICNetLoss, self).forward(pred_sub8, target_sub8) - loss3 = super(ICNetLoss, self).forward(pred_sub16, target_sub16) - return dict(loss=loss1 + loss2 * self.aux_weight + loss3 * self.aux_weight) - - -class OhemCrossEntropy2d(nn.Module): - def __init__(self, ignore_index=-1, thresh=0.7, min_kept=100000, use_weight=True, **kwargs): - super(OhemCrossEntropy2d, self).__init__() - self.ignore_index = ignore_index - self.thresh = float(thresh) - self.min_kept = int(min_kept) - if use_weight: - weight = torch.FloatTensor([0.8373, 0.918, 0.866, 1.0345, 1.0166, 0.9969, 0.9754, - 1.0489, 0.8786, 1.0023, 0.9539, 0.9843, 1.1116, 0.9037, 1.0865, 1.0955, - 1.0865, 1.1529, 1.0507]) - self.criterion = torch.nn.CrossEntropyLoss(weight=weight, ignore_index=ignore_index) - else: - self.criterion = torch.nn.CrossEntropyLoss(ignore_index=ignore_index) - - def forward(self, pred, target): - n, c, h, w = pred.size() - target = target.view(-1) - valid_mask = target.ne(self.ignore_index) - target = target * valid_mask.long() - num_valid = valid_mask.sum() - - prob = F.softmax(pred, dim=1) - prob = prob.transpose(0, 1).reshape(c, -1) - - if self.min_kept > num_valid: - print("Lables: {}".format(num_valid)) - elif num_valid > 0: - prob = prob.masked_fill_(1 - valid_mask, 1) - mask_prob = prob[target, torch.arange(len(target), dtype=torch.long)] - threshold = self.thresh - if self.min_kept > 0: - index = mask_prob.argsort() - threshold_index = index[min(len(index), self.min_kept) - 1] - if mask_prob[threshold_index] > self.thresh: - threshold = mask_prob[threshold_index] - kept_mask = mask_prob.le(threshold) - valid_mask = valid_mask * kept_mask - target = target * kept_mask.long() - - target = target.masked_fill_(1 - valid_mask, self.ignore_index) - target = target.view(n, h, w) - - return self.criterion(pred, target) - - -class MixSoftmaxCrossEntropyOHEMLoss(OhemCrossEntropy2d): - def __init__(self, aux=False, aux_weight=0.4, weight=None, ignore_index=-1, **kwargs): - super(MixSoftmaxCrossEntropyOHEMLoss, self).__init__(ignore_index=ignore_index) - self.aux = aux - self.aux_weight = aux_weight - self.bceloss = nn.BCELoss(weight) - - def _aux_forward(self, *inputs, **kwargs): - *preds, target = tuple(inputs) - - loss = super(MixSoftmaxCrossEntropyOHEMLoss, self).forward(preds[0], target) - for i in range(1, len(preds)): - aux_loss = super(MixSoftmaxCrossEntropyOHEMLoss, self).forward(preds[i], target) - loss += self.aux_weight * aux_loss - return loss - - def forward(self, *inputs): - preds, target = tuple(inputs) - inputs = tuple(list(preds) + [target]) - if self.aux: - return dict(loss=self._aux_forward(*inputs)) - else: - return dict(loss=super(MixSoftmaxCrossEntropyOHEMLoss, self).forward(*inputs)) - - -def get_segmentation_loss(model, use_ohem=False, **kwargs): - if use_ohem: - return MixSoftmaxCrossEntropyOHEMLoss(**kwargs) - - model = model.lower() - if model == 'encnet': - return EncNetLoss(**kwargs) - elif model == 'icnet': - return ICNetLoss(nclass=4, **kwargs) - else: - return MixSoftmaxCrossEntropyLoss(**kwargs) diff --git a/AIlib/segutils/core/utils/lr_scheduler.py b/AIlib/segutils/core/utils/lr_scheduler.py deleted file mode 100644 index 32b3795..0000000 --- a/AIlib/segutils/core/utils/lr_scheduler.py +++ /dev/null @@ -1,179 +0,0 @@ -"""Popular Learning Rate Schedulers""" -from __future__ import division -import math -import torch - -from bisect import bisect_right - -__all__ = ['LRScheduler', 'WarmupMultiStepLR', 'WarmupPolyLR'] - - -class LRScheduler(object): - r"""Learning Rate Scheduler - - Parameters - ---------- - mode : str - Modes for learning rate scheduler. - Currently it supports 'constant', 'step', 'linear', 'poly' and 'cosine'. - base_lr : float - Base learning rate, i.e. the starting learning rate. - target_lr : float - Target learning rate, i.e. the ending learning rate. - With constant mode target_lr is ignored. - niters : int - Number of iterations to be scheduled. - nepochs : int - Number of epochs to be scheduled. - iters_per_epoch : int - Number of iterations in each epoch. - offset : int - Number of iterations before this scheduler. - power : float - Power parameter of poly scheduler. - step_iter : list - A list of iterations to decay the learning rate. - step_epoch : list - A list of epochs to decay the learning rate. - step_factor : float - Learning rate decay factor. - """ - - def __init__(self, mode, base_lr=0.01, target_lr=0, niters=0, nepochs=0, iters_per_epoch=0, - offset=0, power=0.9, step_iter=None, step_epoch=None, step_factor=0.1, warmup_epochs=0): - super(LRScheduler, self).__init__() - assert (mode in ['constant', 'step', 'linear', 'poly', 'cosine']) - - if mode == 'step': - assert (step_iter is not None or step_epoch is not None) - self.niters = niters - self.step = step_iter - epoch_iters = nepochs * iters_per_epoch - if epoch_iters > 0: - self.niters = epoch_iters - if step_epoch is not None: - self.step = [s * iters_per_epoch for s in step_epoch] - - self.step_factor = step_factor - self.base_lr = base_lr - self.target_lr = base_lr if mode == 'constant' else target_lr - self.offset = offset - self.power = power - self.warmup_iters = warmup_epochs * iters_per_epoch - self.mode = mode - - def __call__(self, optimizer, num_update): - self.update(num_update) - assert self.learning_rate >= 0 - self._adjust_learning_rate(optimizer, self.learning_rate) - - def update(self, num_update): - N = self.niters - 1 - T = num_update - self.offset - T = min(max(0, T), N) - - if self.mode == 'constant': - factor = 0 - elif self.mode == 'linear': - factor = 1 - T / N - elif self.mode == 'poly': - factor = pow(1 - T / N, self.power) - elif self.mode == 'cosine': - factor = (1 + math.cos(math.pi * T / N)) / 2 - elif self.mode == 'step': - if self.step is not None: - count = sum([1 for s in self.step if s <= T]) - factor = pow(self.step_factor, count) - else: - factor = 1 - else: - raise NotImplementedError - - # warm up lr schedule - if self.warmup_iters > 0 and T < self.warmup_iters: - factor = factor * 1.0 * T / self.warmup_iters - - if self.mode == 'step': - self.learning_rate = self.base_lr * factor - else: - self.learning_rate = self.target_lr + (self.base_lr - self.target_lr) * factor - - def _adjust_learning_rate(self, optimizer, lr): - optimizer.param_groups[0]['lr'] = lr - # enlarge the lr at the head - for i in range(1, len(optimizer.param_groups)): - optimizer.param_groups[i]['lr'] = lr * 10 - - -# separating MultiStepLR with WarmupLR -# but the current LRScheduler design doesn't allow it -# reference: https://github.com/facebookresearch/maskrcnn-benchmark/blob/master/maskrcnn_benchmark/solver/lr_scheduler.py -class WarmupMultiStepLR(torch.optim.lr_scheduler._LRScheduler): - def __init__(self, optimizer, milestones, gamma=0.1, warmup_factor=1.0 / 3, - warmup_iters=500, warmup_method="linear", last_epoch=-1): - super(WarmupMultiStepLR, self).__init__(optimizer, last_epoch) - if not list(milestones) == sorted(milestones): - raise ValueError( - "Milestones should be a list of" " increasing integers. Got {}", milestones) - if warmup_method not in ("constant", "linear"): - raise ValueError( - "Only 'constant' or 'linear' warmup_method accepted got {}".format(warmup_method)) - - self.milestones = milestones - self.gamma = gamma - self.warmup_factor = warmup_factor - self.warmup_iters = warmup_iters - self.warmup_method = warmup_method - - def get_lr(self): - warmup_factor = 1 - if self.last_epoch < self.warmup_iters: - if self.warmup_method == 'constant': - warmup_factor = self.warmup_factor - elif self.warmup_factor == 'linear': - alpha = float(self.last_epoch) / self.warmup_iters - warmup_factor = self.warmup_factor * (1 - alpha) + alpha - return [base_lr * warmup_factor * self.gamma ** bisect_right(self.milestones, self.last_epoch) - for base_lr in self.base_lrs] - - -class WarmupPolyLR(torch.optim.lr_scheduler._LRScheduler): - def __init__(self, optimizer, target_lr=0, max_iters=0, power=0.9, warmup_factor=1.0 / 3, - warmup_iters=500, warmup_method='linear', last_epoch=-1): - if warmup_method not in ("constant", "linear"): - raise ValueError( - "Only 'constant' or 'linear' warmup_method accepted " - "got {}".format(warmup_method)) - - self.target_lr = target_lr - self.max_iters = max_iters - self.power = power - self.warmup_factor = warmup_factor - self.warmup_iters = warmup_iters - self.warmup_method = warmup_method - - super(WarmupPolyLR, self).__init__(optimizer, last_epoch) - - def get_lr(self): - N = self.max_iters - self.warmup_iters - T = self.last_epoch - self.warmup_iters - if self.last_epoch < self.warmup_iters: - if self.warmup_method == 'constant': - warmup_factor = self.warmup_factor - elif self.warmup_method == 'linear': - alpha = float(self.last_epoch) / self.warmup_iters - warmup_factor = self.warmup_factor * (1 - alpha) + alpha - else: - raise ValueError("Unknown warmup type.") - return [self.target_lr + (base_lr - self.target_lr) * warmup_factor for base_lr in self.base_lrs] - factor = pow(1 - T / N, self.power) - return [self.target_lr + (base_lr - self.target_lr) * factor for base_lr in self.base_lrs] - - -if __name__ == '__main__': - import torch - import torch.nn as nn - - model = nn.Conv2d(16, 16, 3, 1, 1) - optimizer = torch.optim.Adam(model.parameters(), lr=0.01) - lr_scheduler = WarmupPolyLR(optimizer, niters=1000) diff --git a/AIlib/segutils/core/utils/parallel.py b/AIlib/segutils/core/utils/parallel.py deleted file mode 100644 index cb9e896..0000000 --- a/AIlib/segutils/core/utils/parallel.py +++ /dev/null @@ -1,162 +0,0 @@ -"""Utils for Semantic Segmentation""" -import threading -import torch -import torch.cuda.comm as comm -from torch.nn.parallel.data_parallel import DataParallel -from torch.nn.parallel._functions import Broadcast -from torch.autograd import Function - -__all__ = ['DataParallelModel', 'DataParallelCriterion'] - - -class Reduce(Function): - @staticmethod - def forward(ctx, *inputs): - ctx.target_gpus = [inputs[i].get_device() for i in range(len(inputs))] - inputs = sorted(inputs, key=lambda i: i.get_device()) - return comm.reduce_add(inputs) - - @staticmethod - def backward(ctx, gradOutputs): - return Broadcast.apply(ctx.target_gpus, gradOutputs) - - -class DataParallelModel(DataParallel): - """Data parallelism - - Hide the difference of single/multiple GPUs to the user. - In the forward pass, the module is replicated on each device, - and each replica handles a portion of the input. During the backwards - pass, gradients from each replica are summed into the original module. - - The batch size should be larger than the number of GPUs used. - - Parameters - ---------- - module : object - Network to be parallelized. - sync : bool - enable synchronization (default: False). - Inputs: - - **inputs**: list of input - Outputs: - - **outputs**: list of output - Example:: - >>> net = DataParallelModel(model, device_ids=[0, 1, 2]) - >>> output = net(input_var) # input_var can be on any device, including CPU - """ - - def gather(self, outputs, output_device): - return outputs - - def replicate(self, module, device_ids): - modules = super(DataParallelModel, self).replicate(module, device_ids) - return modules - - -# Reference: https://github.com/zhanghang1989/PyTorch-Encoding/blob/master/encoding/parallel.py -class DataParallelCriterion(DataParallel): - """ - Calculate loss in multiple-GPUs, which balance the memory usage for - Semantic Segmentation. - - The targets are splitted across the specified devices by chunking in - the batch dimension. Please use together with :class:`encoding.parallel.DataParallelModel`. - - Example:: - >>> net = DataParallelModel(model, device_ids=[0, 1, 2]) - >>> criterion = DataParallelCriterion(criterion, device_ids=[0, 1, 2]) - >>> y = net(x) - >>> loss = criterion(y, target) - """ - - def forward(self, inputs, *targets, **kwargs): - # the inputs should be the outputs of DataParallelModel - if not self.device_ids: - return self.module(inputs, *targets, **kwargs) - targets, kwargs = self.scatter(targets, kwargs, self.device_ids) - if len(self.device_ids) == 1: - return self.module(inputs, *targets[0], **kwargs[0]) - replicas = self.replicate(self.module, self.device_ids[:len(inputs)]) - outputs = criterion_parallel_apply(replicas, inputs, targets, kwargs) - return Reduce.apply(*outputs) / len(outputs) - - -def get_a_var(obj): - if isinstance(obj, torch.Tensor): - return obj - - if isinstance(obj, list) or isinstance(obj, tuple): - for result in map(get_a_var, obj): - if isinstance(result, torch.Tensor): - return result - - if isinstance(obj, dict): - for result in map(get_a_var, obj.items()): - if isinstance(result, torch.Tensor): - return result - return None - - -def criterion_parallel_apply(modules, inputs, targets, kwargs_tup=None, devices=None): - r"""Applies each `module` in :attr:`modules` in parallel on arguments - contained in :attr:`inputs` (positional), attr:'targets' (positional) and :attr:`kwargs_tup` (keyword) - on each of :attr:`devices`. - - Args: - modules (Module): modules to be parallelized - inputs (tensor): inputs to the modules - targets (tensor): targets to the modules - devices (list of int or torch.device): CUDA devices - :attr:`modules`, :attr:`inputs`, :attr:'targets' :attr:`kwargs_tup` (if given), and - :attr:`devices` (if given) should all have same length. Moreover, each - element of :attr:`inputs` can either be a single object as the only argument - to a module, or a collection of positional arguments. - """ - assert len(modules) == len(inputs) - assert len(targets) == len(inputs) - if kwargs_tup is not None: - assert len(modules) == len(kwargs_tup) - else: - kwargs_tup = ({},) * len(modules) - if devices is not None: - assert len(modules) == len(devices) - else: - devices = [None] * len(modules) - lock = threading.Lock() - results = {} - grad_enabled = torch.is_grad_enabled() - - def _worker(i, module, input, target, kwargs, device=None): - torch.set_grad_enabled(grad_enabled) - if device is None: - device = get_a_var(input).get_device() - try: - with torch.cuda.device(device): - output = module(*(list(input) + target), **kwargs) - with lock: - results[i] = output - except Exception as e: - with lock: - results[i] = e - - if len(modules) > 1: - threads = [threading.Thread(target=_worker, - args=(i, module, input, target, kwargs, device)) - for i, (module, input, target, kwargs, device) in - enumerate(zip(modules, inputs, targets, kwargs_tup, devices))] - - for thread in threads: - thread.start() - for thread in threads: - thread.join() - else: - _worker(0, modules[0], inputs[0], targets[0], kwargs_tup[0], devices[0]) - - outputs = [] - for i in range(len(inputs)): - output = results[i] - if isinstance(output, Exception): - raise output - outputs.append(output) - return outputs diff --git a/AIlib/segutils/core/utils/score.py b/AIlib/segutils/core/utils/score.py deleted file mode 100644 index a037e65..0000000 --- a/AIlib/segutils/core/utils/score.py +++ /dev/null @@ -1,161 +0,0 @@ -"""Evaluation Metrics for Semantic Segmentation""" -import torch -import numpy as np - -__all__ = ['SegmentationMetric', 'batch_pix_accuracy', 'batch_intersection_union', - 'pixelAccuracy', 'intersectionAndUnion', 'hist_info', 'compute_score'] - - -class SegmentationMetric(object): - """Computes pixAcc and mIoU metric scores - """ - - def __init__(self, nclass): - super(SegmentationMetric, self).__init__() - self.nclass = nclass - self.reset() - - def update(self, preds, labels): - """Updates the internal evaluation result. - - Parameters - ---------- - labels : 'NumpyArray' or list of `NumpyArray` - The labels of the data. - preds : 'NumpyArray' or list of `NumpyArray` - Predicted values. - """ - - def evaluate_worker(self, pred, label): - correct, labeled = batch_pix_accuracy(pred, label) - inter, union = batch_intersection_union(pred, label, self.nclass) - - self.total_correct += correct - self.total_label += labeled - if self.total_inter.device != inter.device: - self.total_inter = self.total_inter.to(inter.device) - self.total_union = self.total_union.to(union.device) - self.total_inter += inter - self.total_union += union - - if isinstance(preds, torch.Tensor): - evaluate_worker(self, preds, labels) - elif isinstance(preds, (list, tuple)): - for (pred, label) in zip(preds, labels): - evaluate_worker(self, pred, label) - - def get(self): - """Gets the current evaluation result. - - Returns - ------- - metrics : tuple of float - pixAcc and mIoU - """ - pixAcc = 1.0 * self.total_correct / (2.220446049250313e-16 + self.total_label) # remove np.spacing(1) - IoU = 1.0 * self.total_inter / (2.220446049250313e-16 + self.total_union) - mIoU = IoU.mean().item() - return pixAcc, mIoU - - def reset(self): - """Resets the internal evaluation result to initial state.""" - self.total_inter = torch.zeros(self.nclass) - self.total_union = torch.zeros(self.nclass) - self.total_correct = 0 - self.total_label = 0 - - -# pytorch version -def batch_pix_accuracy(output, target): - """PixAcc""" - # inputs are numpy array, output 4D, target 3D - predict = torch.argmax(output.long(), 1) + 1 - target = target.long() + 1 - - pixel_labeled = torch.sum(target > 0).item() - pixel_correct = torch.sum((predict == target) * (target > 0)).item() - assert pixel_correct <= pixel_labeled, "Correct area should be smaller than Labeled" - return pixel_correct, pixel_labeled - - -def batch_intersection_union(output, target, nclass): - """mIoU""" - # inputs are numpy array, output 4D, target 3D - mini = 1 - maxi = nclass - nbins = nclass - predict = torch.argmax(output, 1) + 1 - target = target.float() + 1 - - predict = predict.float() * (target > 0).float() - intersection = predict * (predict == target).float() - # areas of intersection and union - # element 0 in intersection occur the main difference from np.bincount. set boundary to -1 is necessary. - area_inter = torch.histc(intersection.cpu(), bins=nbins, min=mini, max=maxi) - area_pred = torch.histc(predict.cpu(), bins=nbins, min=mini, max=maxi) - area_lab = torch.histc(target.cpu(), bins=nbins, min=mini, max=maxi) - area_union = area_pred + area_lab - area_inter - assert torch.sum(area_inter > area_union).item() == 0, "Intersection area should be smaller than Union area" - return area_inter.float(), area_union.float() - - -def pixelAccuracy(imPred, imLab): - """ - This function takes the prediction and label of a single image, returns pixel-wise accuracy - To compute over many images do: - for i = range(Nimages): - (pixel_accuracy[i], pixel_correct[i], pixel_labeled[i]) = \ - pixelAccuracy(imPred[i], imLab[i]) - mean_pixel_accuracy = 1.0 * np.sum(pixel_correct) / (np.spacing(1) + np.sum(pixel_labeled)) - """ - # Remove classes from unlabeled pixels in gt image. - # We should not penalize detections in unlabeled portions of the image. - pixel_labeled = np.sum(imLab >= 0) - pixel_correct = np.sum((imPred == imLab) * (imLab >= 0)) - pixel_accuracy = 1.0 * pixel_correct / pixel_labeled - return (pixel_accuracy, pixel_correct, pixel_labeled) - - -def intersectionAndUnion(imPred, imLab, numClass): - """ - This function takes the prediction and label of a single image, - returns intersection and union areas for each class - To compute over many images do: - for i in range(Nimages): - (area_intersection[:,i], area_union[:,i]) = intersectionAndUnion(imPred[i], imLab[i]) - IoU = 1.0 * np.sum(area_intersection, axis=1) / np.sum(np.spacing(1)+area_union, axis=1) - """ - # Remove classes from unlabeled pixels in gt image. - # We should not penalize detections in unlabeled portions of the image. - imPred = imPred * (imLab >= 0) - - # Compute area intersection: - intersection = imPred * (imPred == imLab) - (area_intersection, _) = np.histogram(intersection, bins=numClass, range=(1, numClass)) - - # Compute area union: - (area_pred, _) = np.histogram(imPred, bins=numClass, range=(1, numClass)) - (area_lab, _) = np.histogram(imLab, bins=numClass, range=(1, numClass)) - area_union = area_pred + area_lab - area_intersection - return (area_intersection, area_union) - - -def hist_info(pred, label, num_cls): - assert pred.shape == label.shape - k = (label >= 0) & (label < num_cls) - labeled = np.sum(k) - correct = np.sum((pred[k] == label[k])) - - return np.bincount(num_cls * label[k].astype(int) + pred[k], minlength=num_cls ** 2).reshape(num_cls, - num_cls), labeled, correct - - -def compute_score(hist, correct, labeled): - iu = np.diag(hist) / (hist.sum(1) + hist.sum(0) - np.diag(hist)) - mean_IU = np.nanmean(iu) - mean_IU_no_back = np.nanmean(iu[1:]) - freq = hist.sum(1) / hist.sum() - freq_IU = (iu[freq > 0] * freq[freq > 0]).sum() - mean_pixel_acc = correct / labeled - - return iu, mean_IU, mean_IU_no_back, mean_pixel_acc diff --git a/AIlib/segutils/core/utils/visualize.py b/AIlib/segutils/core/utils/visualize.py deleted file mode 100644 index c63d6c9..0000000 --- a/AIlib/segutils/core/utils/visualize.py +++ /dev/null @@ -1,158 +0,0 @@ -import os -import numpy as np -from PIL import Image - -__all__ = ['get_color_pallete', 'print_iou', 'set_img_color', - 'show_prediction', 'show_colorful_images', 'save_colorful_images'] - - -def print_iou(iu, mean_pixel_acc, class_names=None, show_no_back=False): - n = iu.size - lines = [] - for i in range(n): - if class_names is None: - cls = 'Class %d:' % (i + 1) - else: - cls = '%d %s' % (i + 1, class_names[i]) - # lines.append('%-8s: %.3f%%' % (cls, iu[i] * 100)) - mean_IU = np.nanmean(iu) - mean_IU_no_back = np.nanmean(iu[1:]) - if show_no_back: - lines.append('mean_IU: %.3f%% || mean_IU_no_back: %.3f%% || mean_pixel_acc: %.3f%%' % ( - mean_IU * 100, mean_IU_no_back * 100, mean_pixel_acc * 100)) - else: - lines.append('mean_IU: %.3f%% || mean_pixel_acc: %.3f%%' % (mean_IU * 100, mean_pixel_acc * 100)) - lines.append('=================================================') - line = "\n".join(lines) - - print(line) - - -def set_img_color(img, label, colors, background=0, show255=False): - for i in range(len(colors)): - if i != background: - img[np.where(label == i)] = colors[i] - if show255: - img[np.where(label == 255)] = 255 - - return img - - -def show_prediction(img, pred, colors, background=0): - im = np.array(img, np.uint8) - set_img_color(im, pred, colors, background) - out = np.array(im) - - return out - - -def show_colorful_images(prediction, palettes): - im = Image.fromarray(palettes[prediction.astype('uint8').squeeze()]) - im.show() - - -def save_colorful_images(prediction, filename, output_dir, palettes): - ''' - :param prediction: [B, H, W, C] - ''' - im = Image.fromarray(palettes[prediction.astype('uint8').squeeze()]) - fn = os.path.join(output_dir, filename) - out_dir = os.path.split(fn)[0] - if not os.path.exists(out_dir): - os.mkdir(out_dir) - im.save(fn) - - -def get_color_pallete(npimg, dataset='pascal_voc'): - """Visualize image. - - Parameters - ---------- - npimg : numpy.ndarray - Single channel image with shape `H, W, 1`. - dataset : str, default: 'pascal_voc' - The dataset that model pretrained on. ('pascal_voc', 'ade20k') - Returns - ------- - out_img : PIL.Image - Image with color pallete - """ - # recovery boundary - if dataset in ('pascal_voc', 'pascal_aug'): - npimg[npimg == -1] = 255 - # put colormap - if dataset == 'ade20k': - npimg = npimg + 1 - out_img = Image.fromarray(npimg.astype('uint8')) - out_img.putpalette(adepallete) - return out_img - elif dataset == 'citys': - out_img = Image.fromarray(npimg.astype('uint8')) - out_img.putpalette(cityspallete) - return out_img - out_img = Image.fromarray(npimg.astype('uint8')) - out_img.putpalette(vocpallete) - return out_img - - -def _getvocpallete(num_cls): - n = num_cls - pallete = [0] * (n * 3) - for j in range(0, n): - lab = j - pallete[j * 3 + 0] = 0 - pallete[j * 3 + 1] = 0 - pallete[j * 3 + 2] = 0 - i = 0 - while (lab > 0): - pallete[j * 3 + 0] |= (((lab >> 0) & 1) << (7 - i)) - pallete[j * 3 + 1] |= (((lab >> 1) & 1) << (7 - i)) - pallete[j * 3 + 2] |= (((lab >> 2) & 1) << (7 - i)) - i = i + 1 - lab >>= 3 - return pallete - - -vocpallete = _getvocpallete(256) - -adepallete = [ - 0, 0, 0, 120, 120, 120, 180, 120, 120, 6, 230, 230, 80, 50, 50, 4, 200, 3, 120, 120, 80, 140, 140, 140, 204, - 5, 255, 230, 230, 230, 4, 250, 7, 224, 5, 255, 235, 255, 7, 150, 5, 61, 120, 120, 70, 8, 255, 51, 255, 6, 82, - 143, 255, 140, 204, 255, 4, 255, 51, 7, 204, 70, 3, 0, 102, 200, 61, 230, 250, 255, 6, 51, 11, 102, 255, 255, - 7, 71, 255, 9, 224, 9, 7, 230, 220, 220, 220, 255, 9, 92, 112, 9, 255, 8, 255, 214, 7, 255, 224, 255, 184, 6, - 10, 255, 71, 255, 41, 10, 7, 255, 255, 224, 255, 8, 102, 8, 255, 255, 61, 6, 255, 194, 7, 255, 122, 8, 0, 255, - 20, 255, 8, 41, 255, 5, 153, 6, 51, 255, 235, 12, 255, 160, 150, 20, 0, 163, 255, 140, 140, 140, 250, 10, 15, - 20, 255, 0, 31, 255, 0, 255, 31, 0, 255, 224, 0, 153, 255, 0, 0, 0, 255, 255, 71, 0, 0, 235, 255, 0, 173, 255, - 31, 0, 255, 11, 200, 200, 255, 82, 0, 0, 255, 245, 0, 61, 255, 0, 255, 112, 0, 255, 133, 255, 0, 0, 255, 163, - 0, 255, 102, 0, 194, 255, 0, 0, 143, 255, 51, 255, 0, 0, 82, 255, 0, 255, 41, 0, 255, 173, 10, 0, 255, 173, 255, - 0, 0, 255, 153, 255, 92, 0, 255, 0, 255, 255, 0, 245, 255, 0, 102, 255, 173, 0, 255, 0, 20, 255, 184, 184, 0, - 31, 255, 0, 255, 61, 0, 71, 255, 255, 0, 204, 0, 255, 194, 0, 255, 82, 0, 10, 255, 0, 112, 255, 51, 0, 255, 0, - 194, 255, 0, 122, 255, 0, 255, 163, 255, 153, 0, 0, 255, 10, 255, 112, 0, 143, 255, 0, 82, 0, 255, 163, 255, - 0, 255, 235, 0, 8, 184, 170, 133, 0, 255, 0, 255, 92, 184, 0, 255, 255, 0, 31, 0, 184, 255, 0, 214, 255, 255, - 0, 112, 92, 255, 0, 0, 224, 255, 112, 224, 255, 70, 184, 160, 163, 0, 255, 153, 0, 255, 71, 255, 0, 255, 0, - 163, 255, 204, 0, 255, 0, 143, 0, 255, 235, 133, 255, 0, 255, 0, 235, 245, 0, 255, 255, 0, 122, 255, 245, 0, - 10, 190, 212, 214, 255, 0, 0, 204, 255, 20, 0, 255, 255, 255, 0, 0, 153, 255, 0, 41, 255, 0, 255, 204, 41, 0, - 255, 41, 255, 0, 173, 0, 255, 0, 245, 255, 71, 0, 255, 122, 0, 255, 0, 255, 184, 0, 92, 255, 184, 255, 0, 0, - 133, 255, 255, 214, 0, 25, 194, 194, 102, 255, 0, 92, 0, 255] - -cityspallete = [ - 128, 64, 128, - 244, 35, 232, - 70, 70, 70, - 102, 102, 156, - 190, 153, 153, - 153, 153, 153, - 250, 170, 30, - 220, 220, 0, - 107, 142, 35, - 152, 251, 152, - 0, 130, 180, - 220, 20, 60, - 255, 0, 0, - 0, 0, 142, - 0, 0, 70, - 0, 60, 100, - 0, 80, 100, - 0, 0, 230, - 119, 11, 32, -] diff --git a/AIlib/segutils/segMultiOutModel.py b/AIlib/segutils/segMultiOutModel.py deleted file mode 100644 index e38838d..0000000 --- a/AIlib/segutils/segMultiOutModel.py +++ /dev/null @@ -1,377 +0,0 @@ -import torch -from core.models.bisenet import BiSeNet,BiSeNet_MultiOutput -from torchvision import transforms -import cv2,os,glob -import numpy as np -from core.models.dinknet import DinkNet34 -import matplotlib.pyplot as plt - -import matplotlib.pyplot as plt -import time -class SegModel(object): - def __init__(self, nclass=2,model = None,weights=None,modelsize=512,device='cuda:3',multiOutput=False): - #self.args = args - self.model = model - #self.model = DinkNet34(nclass) - checkpoint = torch.load(weights) - self.modelsize = modelsize - self.model.load_state_dict(checkpoint['model']) - self.device = device - self.multiOutput = multiOutput - self.model= self.model.to(self.device) - '''self.composed_transforms = transforms.Compose([ - - transforms.Normalize(mean=(0.335, 0.358, 0.332), std=(0.141, 0.138, 0.143)), - transforms.ToTensor()]) ''' - self.mean = (0.335, 0.358, 0.332) - self.std = (0.141, 0.138, 0.143) - #mean=(0.335, 0.358, 0.332), std=(0.141, 0.138, 0.143) - def eval(self,image,outsize=None,smooth_kernel=0): - imageH,imageW,imageC = image.shape - time0 = time.time() - image = self.preprocess_image(image) - time1 = time.time() - self.model.eval() - image = image.to(self.device) - with torch.no_grad(): - output = self.model(image,test_flag=True,smooth_kernel = 0) - - time2 = time.time() - - if self.multiOutput: - pred = [outputx.data.cpu().numpy()[0] for outputx in output] - else: - pred = output.data.cpu().numpy() - pred = pred[0] - - time3 = time.time() - - if self.multiOutput: - pred = [ cv2.blur(predx,(smooth_kernel,smooth_kernel) ) for predx in pred] - pred = [cv2.resize(predx.astype(np.uint8),(imageW,imageH)) for predx in pred[0:2]] - else: - pred = cv2.blur(pred,(smooth_kernel,smooth_kernel) ) - pred = cv2.resize(pred.astype(np.uint8),(imageW,imageH),interpolation = cv2.INTER_NEAREST) - time4 = time.time() - print('##line52:pre-precess:%.1f ,infer:%.1f ,post-precess:%.1f ,post-resize:%.1f '%( self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3) )) - - return pred - def get_ms(self,t1,t0): - return (t1-t0)*1000.0 - def preprocess_image(self,image): - - time0 = time.time() - image = cv2.resize(image,(self.modelsize,self.modelsize)) - - time1 = time.time() - image = image.astype(np.float32) - image /= 255.0 - - time2 = time.time() - #image = image * 3.2 - 1.6 - image[:,:,0] -=self.mean[0] - image[:,:,1] -=self.mean[1] - image[:,:,2] -=self.mean[2] - - time3 = time.time() - image[:,:,0] /= self.std[0] - image[:,:,1] /= self.std[1] - image[:,:,2] /= self.std[2] - - - time4 = time.time() - image = np.transpose(image, ( 2, 0, 1)) - time5 = time.time() - image = torch.from_numpy(image).float() - image = image.unsqueeze(0) - print('###line84: in preprocess: resize:%.1f norm:%.1f mean:%.1f std:%.1f trans:%.f '%(self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3) ,self.get_ms(time5,time4) ) ) - - return image - - - -def get_ms(t1,t0): - return (t1-t0)*1000.0 - -def test(): - #os.environ["CUDA_VISIBLE_DEVICES"] = str('4') - ''' - image_url = '../../data/landcover/corp512/test/images/N-33-139-C-d-2-4_169.jpg' - nclass = 5 - weights = 'runs/landcover/DinkNet34_save/experiment_wj_loss-10-10-1/checkpoint.pth' - ''' - - - image_url = 'temp_pics/DJI_0645.JPG' - nclass = 2 - #weights = '../weights/segmentation/BiSeNet/checkpoint.pth' - weights = 'runs/THriver/BiSeNet/train/experiment_0/checkpoint.pth' - #weights = 'runs/segmentation/BiSeNet_test/experiment_10/checkpoint.pth' - - model = BiSeNet(nclass) - segmodel = SegModel(model=model,nclass=nclass,weights=weights,device='cuda:4') - for i in range(10): - image_array0 = cv2.imread(image_url) - imageH,imageW,_ = image_array0.shape - #print('###line84:',image_array0.shape) - image_array = cv2.cvtColor( image_array0,cv2.COLOR_RGB2BGR) - #image_in = segmodel.preprocess_image(image_array) - pred = segmodel.eval(image_array,outsize=None) - time0=time.time() - binary = pred.copy() - time1=time.time() - contours, hierarchy = cv2.findContours(binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - time2=time.time() - print(pred.shape,' time copy:%.1f finccontour:%.1f '%(get_ms(time1,time0),get_ms(time2,time1) )) - -label_dic={'landcover':[[0, 0, 0], [255, 0, 0], [0,255,0], [0,0,255], [255,255,0]], - 'deepRoad':[[0,0,0],[255,0,0]], - 'water':[[0,0,0],[255,255,255]], - 'water_building':[[0,0,0],[0,0,255],[255,0,0]], - 'floater':[[0,0,0], [0,255,0],[255,255,0],[255,0,255],[0,128, 255], [255,0,0], [0,255,255] ] - - - - - } - -def index2color(label_mask,label_colours): - r = label_mask.copy() - g = label_mask.copy() - b = label_mask.copy() - label_cnt = len(label_colours) - for ll in range(0, label_cnt): - r[label_mask == ll] = label_colours[ll][0] - g[label_mask == ll] = label_colours[ll][1] - b[label_mask == ll] = label_colours[ll][2] - rgb = np.stack((b, g,r), axis=-1) - return rgb.astype(np.uint8) -def get_largest_contours(contours): - areas = [cv2.contourArea(x) for x in contours] - max_area = max(areas) - max_id = areas.index(max_area) - - return max_id -def result_merge_sep(image,mask_colors): - #mask_colors=[{ 'mask':mask_map,'index':[1],'color':[255,255,255] }] - for mask_color in mask_colors: - mask_map,indexes,colors = mask_color['mask'], mask_color['index'], mask_color['color'] - ishow = 2 - #plt.figure(1);plt.imshow(mask_map); - for index,color in zip(indexes,colors): - mask_binaray = (mask_map == index).astype(np.uint8) - contours, hierarchy = cv2.findContours(mask_binaray,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - if len(contours)>0: - d=hierarchy[0,:,3]<0 ; - contours = np.array(contours,dtype=object)[d] - cv2.drawContours(image,contours,-1,color[::-1],3) - #plt.figure(ishow);plt.imshow(mask_binaray);ishow+=1 - #plt.show() - return image -def result_merge(image,mask_colors): - #mask_colors=[{ 'mask':mask_map,'index':[1],'color':[255,255,255] }] - for mask_color in mask_colors: - mask_map,indexes,colors = mask_color['mask'], mask_color['index'], mask_color['color'] - mask_binary = (mask_map>0).astype(np.uint8) - contours, hierarchy = cv2.findContours(mask_binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - - if len(contours)>0: - d=hierarchy[0,:,3]<0 ; contours = np.array(contours)[d] - cv2.drawContours(image,contours,-1,colors[0][::-1],3) - - coors = np.array([(np.mean(contours_x ,axis=0)+0.5).astype(np.int32)[0] for contours_x in contours]) - #print(mask_map.shape,coors.shape) - typess = mask_map[ coors[:,1],coors[:,0]] - #for jj,iclass in enumerate(typess): - #print(iclass,colors) - # cv2.drawContours(image,contours,-1, colors[iclass][::-1],3) - - - - return image - -def test_floater(): - from core.models.dinknet import DinkNet34_MultiOutput - #create_model('DinkNet34_MultiOutput',[2,5]) - - image_url = 'temp_pics/DJI_0645.JPG' - nclass = [2,7] - outresult=True - weights = 'runs/thFloater/BiSeNet_MultiOutput/train/experiment_4/checkpoint.pth' - model = BiSeNet_MultiOutput(nclass) - outdir='temp' - image_dir = '/host/workspace/WJ/data/thFloater/val/images/' - image_url_list=glob.glob('%s/*'%(image_dir)) - segmodel = SegModel(model=model,nclass=nclass,weights=weights,device='cuda:9',multiOutput=True) - - - for i,image_url in enumerate(image_url_list[0:10]) : - image_array0 = cv2.imread(image_url) - image_array0 = cv2.cvtColor(image_array0, cv2.COLOR_BGR2RGB) # cv2默认为bgr顺序 - imageH,imageW,_ = image_array0.shape - #image_array = cv2.cvtColor( image_array0,cv2.COLOR_RGB2BGR) - pred = segmodel.eval(image_array,outsize=None) - - - time0=time.time() - if isinstance(pred,list): - binary = [predx.copy() for predx in pred] - time1=time.time() - - mask_colors=[ { 'mask':pred[0] ,'index':range(1,2),'color':label_dic['water'][0:] }, - { 'mask':pred[1] ,'index':[1,2,3,4,5,6],'color':label_dic['floater'][0:] } ] - result_draw = result_merge(image_array0,mask_colors) - - - time2=time.time() - - - if outresult: - basename=os.path.splitext( os.path.basename(image_url))[0] - outname=os.path.join(outdir,basename+'_draw.png') - cv2.imwrite(outname,result_draw[:,:,:]) - - - - - print('##line151: time copy:%.1f finccontour:%.1f '%(get_ms(time1,time0),get_ms(time2,time1) )) -def test_water_buildings(): - from core.models.bisenet import BiSeNet - #image_url = 'temp_pics/DJI_0645.JPG' - nclass = 3 - outresult=True - weights = 'runs/thWaterBuilding/BiSeNet/train/experiment_2/checkpoint.pth' - model = BiSeNet(nclass) - outdir='temp' - image_dir = '/home/thsw/WJ/data/river_buildings/' - #image_dir = '/home/thsw/WJ/data/THWaterBuilding/val/images' - image_url_list=glob.glob('%s/*'%(image_dir)) - segmodel = SegModel(model=model,nclass=nclass,weights=weights,device='cuda:0',multiOutput=False) - - - for i,image_url in enumerate(image_url_list[0:]) : - #image_url = '/home/thsw/WJ/data/THWaterBuilding/val/images/0anWqgmO9rGe1n8P.png' - image_array0 = cv2.imread(image_url) - imageH,imageW,_ = image_array0.shape - image_array = cv2.cvtColor( image_array0,cv2.COLOR_RGB2BGR) - pred = segmodel.eval(image_array,outsize=None) - - time0=time.time() - if isinstance(pred,list): - binary = [predx.copy() for predx in pred] - #print(binary[0].shape) - time1=time.time() - - - mask_colors=[ { 'mask':pred ,'index':range(1,3),'color':label_dic['water_building'][1:] }, - #{ 'mask':pred[1] ,'index':[1,2,3,4,5,6],'color':label_dic['floater'][0:] } - ] - result_draw = result_merge_sep(image_array0,mask_colors) - - - time2=time.time() - if outresult: - basename=os.path.splitext( os.path.basename(image_url))[0] - outname=os.path.join(outdir,basename+'_draw.png') - cv2.imwrite(outname,result_draw[:,:,:]) - - print('##line294: time copy:%.1f finccontour:%.1f '%(get_ms(time1,time0),get_ms(time2,time1) )) - -def get_illegal_index(contours,hierarchy,water_dilate,overlap_threshold): - out_index=[] - if len(contours)>0: - d=hierarchy[0,:,3]<0 ; - contours = np.array(contours,dtype=object)[d] - imageH,imageW = water_dilate.shape - for ii,cont in enumerate(contours): - build_area=np.zeros((imageH,imageW )) - cv2.fillPoly(build_area,[cont[:,0,:]],1) - area1=np.sum(build_area);area2=np.sum(build_area*water_dilate) - if (area2/area1) >overlap_threshold: - out_index.append(ii) - - - return out_index - - -def test_water_building_seperately(): - from core.models.dinknet import DinkNet34_MultiOutput - #create_model('DinkNet34_MultiOutput',[2,5]) - - image_url = 'temp_pics/DJI_0645.JPG' - nclass = [2,2] - outresult=True - weights = 'runs/thWaterBuilding_seperate/BiSeNet_MultiOutput/train/experiment_0/checkpoint.pth' - model = BiSeNet_MultiOutput(nclass) - outdir='temp' - image_dir = '/home/thsw/WJ/data/river_buildings/' - #image_dir = '/home/thsw/WJ/data/THWaterBuilding/val/images' - image_url_list=glob.glob('%s/*'%(image_dir)) - segmodel = SegModel(model=model,nclass=nclass,weights=weights,device='cuda:1',multiOutput=True) - - print('###line307 image cnt:',len(image_url_list)) - for i,image_url in enumerate(image_url_list[0:1]) : - image_url = '/home/thsw/WJ/data/river_buildings/DJI_20210904092044_0001_S_output896.jpg' - image_array0 = cv2.imread(image_url) - imageH,imageW,_ = image_array0.shape - image_array = cv2.cvtColor( image_array0,cv2.COLOR_RGB2BGR) - pred = segmodel.eval(image_array,outsize=None,smooth_kernel=20) - - ##画出水体区域 - contours, hierarchy = cv2.findContours(pred[0],cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - max_id = get_largest_contours(contours); - water = pred[0].copy(); water[:,:] = 0 - cv2.fillPoly(water, [contours[max_id][:,0,:]], 1) - cv2.drawContours(image_array0,contours,max_id,(0,255,255),3) - - - - - ##画出水体膨胀后的蓝线区域。 - kernel = np.ones((100,100),np.uint8) - water_dilate = cv2.dilate(water,kernel,iterations = 1) - contours, hierarchy = cv2.findContours(water_dilate,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - #print('####line310:',contours) - cv2.drawContours(image_array0,contours,-1,(255,0,0),3) - - - ###逐个建筑判断是否与蓝线内区域有交叉。如果交叉面积占本身面积超过0.1,则认为是违法建筑。 - contours, hierarchy = cv2.findContours(pred[1],cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - outIndex=get_illegal_index(contours,hierarchy,water_dilate,0.1) - - for ii in outIndex: - cv2.drawContours(image_array0,contours,ii,(0,0,255),3) - - - plt.imshow(image_array0);plt.show() - ## - - time0=time.time() - - time1=time.time() - - - mask_colors=[ { 'mask':pred[0],'index':[1],'color':label_dic['water_building'][1:2]}, - { 'mask':pred[1],'index':[1],'color':label_dic['water_building'][2:3]} - ] - result_draw = result_merge_sep(image_array0,mask_colors) - time2=time.time() - - if outresult: - basename=os.path.splitext( os.path.basename(image_url))[0] - outname=os.path.join(outdir,basename+'_draw.png') - cv2.imwrite(outname,result_draw[:,:,:]) - - print('##line151: time copy:%.1f finccontour:%.1f '%(get_ms(time1,time0),get_ms(time2,time1) )) - -if __name__=='__main__': - #test() - #test_floater() - #test_water_buildings() - test_water_building_seperately() - - - - - - diff --git a/AIlib/segutils/segWaterBuilding.py b/AIlib/segutils/segWaterBuilding.py deleted file mode 100644 index f087227..0000000 --- a/AIlib/segutils/segWaterBuilding.py +++ /dev/null @@ -1,387 +0,0 @@ -import torch -from core.models.bisenet import BiSeNet,BiSeNet_MultiOutput -from torchvision import transforms -import cv2,os,glob -import numpy as np -from core.models.dinknet import DinkNet34 -import matplotlib.pyplot as plt - -import matplotlib.pyplot as plt -import time -class SegModel(object): - def __init__(self, nclass=2,model = None,weights=None,modelsize=512,device='cuda:3',multiOutput=False): - #self.args = args - self.model = model - #self.model = DinkNet34(nclass) - checkpoint = torch.load(weights) - self.modelsize = modelsize - self.model.load_state_dict(checkpoint['model']) - self.device = device - self.multiOutput = multiOutput - self.model= self.model.to(self.device) - '''self.composed_transforms = transforms.Compose([ - - transforms.Normalize(mean=(0.335, 0.358, 0.332), std=(0.141, 0.138, 0.143)), - transforms.ToTensor()]) ''' - self.mean = (0.335, 0.358, 0.332) - self.std = (0.141, 0.138, 0.143) - #mean=(0.335, 0.358, 0.332), std=(0.141, 0.138, 0.143) - def eval(self,image,outsize=None,smooth_kernel=0): - imageH,imageW,imageC = image.shape - time0 = time.time() - image = self.preprocess_image(image) - time1 = time.time() - self.model.eval() - image = image.to(self.device) - with torch.no_grad(): - output = self.model(image,test_flag=True,smooth_kernel = 0) - - time2 = time.time() - - if self.multiOutput: - pred = [outputx.data.cpu().numpy()[0] for outputx in output] - else: - pred = output.data.cpu().numpy() - pred = pred[0] - - time3 = time.time() - - if self.multiOutput: - pred = [ cv2.blur(predx,(smooth_kernel,smooth_kernel) ) for predx in pred] - pred = [cv2.resize(predx.astype(np.uint8),(imageW,imageH)) for predx in pred[0:2]] - else: - pred = cv2.blur(pred,(smooth_kernel,smooth_kernel) ) - pred = cv2.resize(pred.astype(np.uint8),(imageW,imageH),interpolation = cv2.INTER_NEAREST) - time4 = time.time() - outStr= '##line52:pre-precess:%.1f ,infer:%.1f ,post-precess:%.1f ,post-resize:%.1f '%( self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3) ) - #print('##line52:pre-precess:%.1f ,infer:%.1f ,post-precess:%.1f ,post-resize:%.1f '%( self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3) )) - - return pred - def get_ms(self,t1,t0): - return (t1-t0)*1000.0 - def preprocess_image(self,image): - - time0 = time.time() - image = cv2.resize(image,(self.modelsize,self.modelsize)) - - time1 = time.time() - image = image.astype(np.float32) - image /= 255.0 - - time2 = time.time() - #image = image * 3.2 - 1.6 - image[:,:,0] -=self.mean[0] - image[:,:,1] -=self.mean[1] - image[:,:,2] -=self.mean[2] - - time3 = time.time() - image[:,:,0] /= self.std[0] - image[:,:,1] /= self.std[1] - image[:,:,2] /= self.std[2] - - - time4 = time.time() - image = np.transpose(image, ( 2, 0, 1)) - time5 = time.time() - image = torch.from_numpy(image).float() - image = image.unsqueeze(0) - outStr='###line84: in preprocess: resize:%.1f norm:%.1f mean:%.1f std:%.1f trans:%.f '%(self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3) ,self.get_ms(time5,time4) ) - #print('###line84: in preprocess: resize:%.1f norm:%.1f mean:%.1f std:%.1f trans:%.f '%(self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3) ,self.get_ms(time5,time4) ) ) - - return image - - - -def get_ms(t1,t0): - return (t1-t0)*1000.0 - -def test(): - #os.environ["CUDA_VISIBLE_DEVICES"] = str('4') - ''' - image_url = '../../data/landcover/corp512/test/images/N-33-139-C-d-2-4_169.jpg' - nclass = 5 - weights = 'runs/landcover/DinkNet34_save/experiment_wj_loss-10-10-1/checkpoint.pth' - ''' - - - image_url = 'temp_pics/DJI_0645.JPG' - nclass = 2 - #weights = '../weights/segmentation/BiSeNet/checkpoint.pth' - weights = 'runs/THriver/BiSeNet/train/experiment_0/checkpoint.pth' - #weights = 'runs/segmentation/BiSeNet_test/experiment_10/checkpoint.pth' - - model = BiSeNet(nclass) - segmodel = SegModel(model=model,nclass=nclass,weights=weights,device='cuda:4') - for i in range(10): - image_array0 = cv2.imread(image_url) - imageH,imageW,_ = image_array0.shape - #print('###line84:',image_array0.shape) - image_array = cv2.cvtColor( image_array0,cv2.COLOR_RGB2BGR) - #image_in = segmodel.preprocess_image(image_array) - pred = segmodel.eval(image_array,outsize=None) - time0=time.time() - binary = pred.copy() - time1=time.time() - contours, hierarchy = cv2.findContours(binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - time2=time.time() - print(pred.shape,' time copy:%.1f finccontour:%.1f '%(get_ms(time1,time0),get_ms(time2,time1) )) - -label_dic={'landcover':[[0, 0, 0], [255, 0, 0], [0,255,0], [0,0,255], [255,255,0]], - 'deepRoad':[[0,0,0],[255,0,0]], - 'water':[[0,0,0],[255,255,255]], - 'water_building':[[0,0,0],[0,0,255],[255,0,0]], - 'floater':[[0,0,0], [0,255,0],[255,255,0],[255,0,255],[0,128, 255], [255,0,0], [0,255,255] ] - - - - - } - -def index2color(label_mask,label_colours): - r = label_mask.copy() - g = label_mask.copy() - b = label_mask.copy() - label_cnt = len(label_colours) - for ll in range(0, label_cnt): - r[label_mask == ll] = label_colours[ll][0] - g[label_mask == ll] = label_colours[ll][1] - b[label_mask == ll] = label_colours[ll][2] - rgb = np.stack((b, g,r), axis=-1) - return rgb.astype(np.uint8) -def get_largest_contours(contours): - areas = [cv2.contourArea(x) for x in contours] - max_area = max(areas) - max_id = areas.index(max_area) - - return max_id -def result_merge_sep(image,mask_colors): - #mask_colors=[{ 'mask':mask_map,'index':[1],'color':[255,255,255] }] - for mask_color in mask_colors: - mask_map,indexes,colors = mask_color['mask'], mask_color['index'], mask_color['color'] - ishow = 2 - #plt.figure(1);plt.imshow(mask_map); - for index,color in zip(indexes,colors): - mask_binaray = (mask_map == index).astype(np.uint8) - contours, hierarchy = cv2.findContours(mask_binaray,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - if len(contours)>0: - d=hierarchy[0,:,3]<0 ; - contours = np.array(contours,dtype=object)[d] - cv2.drawContours(image,contours,-1,color[::-1],3) - #plt.figure(ishow);plt.imshow(mask_binaray);ishow+=1 - #plt.show() - return image -def result_merge(image,mask_colors): - #mask_colors=[{ 'mask':mask_map,'index':[1],'color':[255,255,255] }] - for mask_color in mask_colors: - mask_map,indexes,colors = mask_color['mask'], mask_color['index'], mask_color['color'] - mask_binary = (mask_map>0).astype(np.uint8) - contours, hierarchy = cv2.findContours(mask_binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - - if len(contours)>0: - d=hierarchy[0,:,3]<0 ; contours = np.array(contours)[d] - cv2.drawContours(image,contours,-1,colors[0][::-1],3) - - coors = np.array([(np.mean(contours_x ,axis=0)+0.5).astype(np.int32)[0] for contours_x in contours]) - #print(mask_map.shape,coors.shape) - typess = mask_map[ coors[:,1],coors[:,0]] - #for jj,iclass in enumerate(typess): - #print(iclass,colors) - # cv2.drawContours(image,contours,-1, colors[iclass][::-1],3) - - - - return image - -def test_floater(): - from core.models.dinknet import DinkNet34_MultiOutput - #create_model('DinkNet34_MultiOutput',[2,5]) - - image_url = 'temp_pics/DJI_0645.JPG' - nclass = [2,7] - outresult=True - weights = 'runs/thFloater/BiSeNet_MultiOutput/train/experiment_4/checkpoint.pth' - model = BiSeNet_MultiOutput(nclass) - outdir='temp' - image_dir = '/host/workspace/WJ/data/thFloater/val/images/' - image_url_list=glob.glob('%s/*'%(image_dir)) - segmodel = SegModel(model=model,nclass=nclass,weights=weights,device='cuda:9',multiOutput=True) - - - for i,image_url in enumerate(image_url_list[0:10]) : - image_array0 = cv2.imread(image_url) - image_array0 = cv2.cvtColor(image_array0, cv2.COLOR_BGR2RGB) # cv2默认为bgr顺序 - imageH,imageW,_ = image_array0.shape - #image_array = cv2.cvtColor( image_array0,cv2.COLOR_RGB2BGR) - pred = segmodel.eval(image_array,outsize=None) - - - time0=time.time() - if isinstance(pred,list): - binary = [predx.copy() for predx in pred] - time1=time.time() - - mask_colors=[ { 'mask':pred[0] ,'index':range(1,2),'color':label_dic['water'][0:] }, - { 'mask':pred[1] ,'index':[1,2,3,4,5,6],'color':label_dic['floater'][0:] } ] - result_draw = result_merge(image_array0,mask_colors) - - - time2=time.time() - - - if outresult: - basename=os.path.splitext( os.path.basename(image_url))[0] - outname=os.path.join(outdir,basename+'_draw.png') - cv2.imwrite(outname,result_draw[:,:,:]) - - - - - print('##line151: time copy:%.1f finccontour:%.1f '%(get_ms(time1,time0),get_ms(time2,time1) )) -def test_water_buildings(): - from core.models.bisenet import BiSeNet - #image_url = 'temp_pics/DJI_0645.JPG' - nclass = 3 - outresult=True - weights = 'runs/thWaterBuilding/BiSeNet/train/experiment_2/checkpoint.pth' - model = BiSeNet(nclass) - outdir='temp' - image_dir = '/home/thsw/WJ/data/river_buildings/' - #image_dir = '/home/thsw/WJ/data/THWaterBuilding/val/images' - image_url_list=glob.glob('%s/*'%(image_dir)) - segmodel = SegModel(model=model,nclass=nclass,weights=weights,device='cuda:0',multiOutput=False) - - - for i,image_url in enumerate(image_url_list[0:]) : - #image_url = '/home/thsw/WJ/data/THWaterBuilding/val/images/0anWqgmO9rGe1n8P.png' - image_array0 = cv2.imread(image_url) - imageH,imageW,_ = image_array0.shape - image_array = cv2.cvtColor( image_array0,cv2.COLOR_RGB2BGR) - pred = segmodel.eval(image_array,outsize=None) - - time0=time.time() - if isinstance(pred,list): - binary = [predx.copy() for predx in pred] - #print(binary[0].shape) - time1=time.time() - - - mask_colors=[ { 'mask':pred ,'index':range(1,3),'color':label_dic['water_building'][1:] }, - #{ 'mask':pred[1] ,'index':[1,2,3,4,5,6],'color':label_dic['floater'][0:] } - ] - result_draw = result_merge_sep(image_array0,mask_colors) - - - time2=time.time() - if outresult: - basename=os.path.splitext( os.path.basename(image_url))[0] - outname=os.path.join(outdir,basename+'_draw.png') - cv2.imwrite(outname,result_draw[:,:,:]) - - print('##line294: time copy:%.1f finccontour:%.1f '%(get_ms(time1,time0),get_ms(time2,time1) )) - -def get_illegal_index(contours,hierarchy,water_dilate,overlap_threshold): - out_index=[] - if len(contours)>0: - d=hierarchy[0,:,3]<0 ; - contours = np.array(contours,dtype=object)[d] - imageH,imageW = water_dilate.shape - for ii,cont in enumerate(contours): - cont = cont.astype(np.int32) - build_area=np.zeros((imageH,imageW )) - try: - cv2.fillPoly(build_area,[cont[:,0,:]],1) - area1=np.sum(build_area);area2=np.sum(build_area*water_dilate) - if (area2/area1) >overlap_threshold: - out_index.append(ii) - except Exception as e: - print('###read error:%s '%(e)) - print(cont.shape,type(cont),cont.dtype) - - - return out_index - -def illBuildings(pred,image_array0): - ##画出水体区域 - contours, hierarchy = cv2.findContours(pred[0],cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - water = pred[0].copy(); water[:,:] = 0 - if len(contours)==0: - return image_array0,water - max_id = get_largest_contours(contours); - - cv2.fillPoly(water, [contours[max_id][:,0,:]], 1) - cv2.drawContours(image_array0,contours,max_id,(0,255,255),3) - - ##画出水体膨胀后的蓝线区域。 - kernel = np.ones((100,100),np.uint8) - water_dilate = cv2.dilate(water,kernel,iterations = 1) - contours, hierarchy = cv2.findContours(water_dilate,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - #print('####line310:',contours) - cv2.drawContours(image_array0,contours,-1,(255,0,0),3) - - ##确定违法建筑并绘图 - ###逐个建筑判断是否与蓝线内区域有交叉。如果交叉面积占本身面积超过0.1,则认为是违法建筑。 - contours, hierarchy = cv2.findContours(pred[1],cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - outIndex=get_illegal_index(contours,hierarchy,water_dilate,0.1) - - for ii in outIndex: - cv2.drawContours(image_array0,contours,ii,(0,0,255),3) - return image_array0,water - -def test_water_building_seperately(): - from core.models.dinknet import DinkNet34_MultiOutput - #create_model('DinkNet34_MultiOutput',[2,5]) - - image_url = 'temp_pics/DJI_0645.JPG' - nclass = [2,2] - outresult=True - weights = 'runs/thWaterBuilding_seperate/BiSeNet_MultiOutput/train/experiment_0/checkpoint.pth' - model = BiSeNet_MultiOutput(nclass) - outdir='temp' - image_dir = '/home/thsw/WJ/data/river_buildings/' - #image_dir = '/home/thsw/WJ/data/THWaterBuilding/val/images' - image_url_list=glob.glob('%s/*'%(image_dir)) - segmodel = SegModel(model=model,nclass=nclass,weights=weights,device='cuda:1',multiOutput=True) - - print('###line307 image cnt:',len(image_url_list)) - for i,image_url in enumerate(image_url_list[0:1]) : - image_url = '/home/thsw/WJ/data/river_buildings/DJI_20210904092044_0001_S_output896.jpg' - image_array0 = cv2.imread(image_url) - imageH,imageW,_ = image_array0.shape - image_array = cv2.cvtColor( image_array0,cv2.COLOR_RGB2BGR) - pred = segmodel.eval(image_array,outsize=None,smooth_kernel=20) - - image_array0,water = illBuildings(pred,image_array0) - - - plt.imshow(image_array0);plt.show() - ## - - time0=time.time() - - time1=time.time() - - - mask_colors=[ { 'mask':pred[0],'index':[1],'color':label_dic['water_building'][1:2]}, - { 'mask':pred[1],'index':[1],'color':label_dic['water_building'][2:3]} - ] - result_draw = result_merge_sep(image_array0,mask_colors) - time2=time.time() - - if outresult: - basename=os.path.splitext( os.path.basename(image_url))[0] - outname=os.path.join(outdir,basename+'_draw.png') - cv2.imwrite(outname,result_draw[:,:,:]) - - print('##line151: time copy:%.1f finccontour:%.1f '%(get_ms(time1,time0),get_ms(time2,time1) )) - -if __name__=='__main__': - #test() - #test_floater() - #test_water_buildings() - test_water_building_seperately() - - - - - - diff --git a/AIlib/segutils/seg_detect.py b/AIlib/segutils/seg_detect.py deleted file mode 100644 index 84611b7..0000000 --- a/AIlib/segutils/seg_detect.py +++ /dev/null @@ -1,132 +0,0 @@ -import torch -from core.models.bisenet import BiSeNet -from torchvision import transforms -import cv2,os -import numpy as np -from core.models.dinknet import DinkNet34 -import matplotlib.pyplot as plt - -import matplotlib.pyplot as plt -import time -class SegModel(object): - def __init__(self, nclass=2,weights=None,modelsize=512,device='cuda:3'): - #self.args = args - self.model = BiSeNet(nclass) - #self.model = DinkNet34(nclass) - checkpoint = torch.load(weights) - self.modelsize = modelsize - self.model.load_state_dict(checkpoint['model']) - self.device = device - self.model= self.model.to(self.device) - '''self.composed_transforms = transforms.Compose([ - - transforms.Normalize(mean=(0.335, 0.358, 0.332), std=(0.141, 0.138, 0.143)), - transforms.ToTensor()]) ''' - self.mean = (0.335, 0.358, 0.332) - self.std = (0.141, 0.138, 0.143) - def eval(self,image,outsize=None): - imageW,imageH,imageC = image.shape - time0 = time.time() - image = self.preprocess_image(image) - time1 = time.time() - self.model.eval() - image = image.to(self.device) - with torch.no_grad(): - output = self.model(image,outsize=outsize) - - time2 = time.time() - pred = output.data.cpu().numpy() - pred = np.argmax(pred, axis=1)[0]#得到每行 - time3 = time.time() - pred = cv2.resize(pred.astype(np.uint8),(imageW,imageH)) - time4 = time.time() - print('pre-precess:%.1f ,infer:%.1f ,post-precess:%.1f ,post-resize:%.1f '%( self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3) )) - return pred - def get_ms(self,t1,t0): - return (t1-t0)*1000.0 - def preprocess_image(self,image): - - time0 = time.time() - image = cv2.resize(image,(self.modelsize,self.modelsize)) - - time1 = time.time() - image = image.astype(np.float32) - image /= 255.0 - - time2 = time.time() - #image -= self.mean - image[:,:,0] -=self.mean[0] - image[:,:,1] -=self.mean[1] - image[:,:,2] -=self.mean[2] - - time3 = time.time() - #image /= self.std - - image[:,:,0] /= self.std[0] - image[:,:,1] /= self.std[1] - image[:,:,2] /= self.std[2] - - - time4 = time.time() - image = np.transpose(image, ( 2, 0, 1)) - time5 = time.time() - image = torch.from_numpy(image).float() - image = image.unsqueeze(0) - print('resize:%.1f norm:%.1f mean:%.1f std:%.1f trans:%.f '%(self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3) ,self.get_ms(time5,time4) ) ) - - return image - - - -def get_ms(t1,t0): - return (t1-t0)*1000.0 - -if __name__=='__main__': - - - - #os.environ["CUDA_VISIBLE_DEVICES"] = str('4') - ''' - image_url = '../../data/landcover/corp512/test/images/N-33-139-C-d-2-4_169.jpg' - nclass = 5 - weights = 'runs/landcover/DinkNet34_save/experiment_wj_loss-10-10-1/checkpoint.pth' - ''' - - - image_url = 'temp_pics/DJI_0645.JPG' - nclass = 2 - #weights = '../weights/segmentation/BiSeNet/checkpoint.pth' - weights = 'runs/THriver/BiSeNet/train/experiment_0/checkpoint.pth' - #weights = 'runs/segmentation/BiSeNet_test/experiment_10/checkpoint.pth' - - - - segmodel = SegModel(nclass=nclass,weights=weights,device='cuda:4') - for i in range(10): - image_array0 = cv2.imread(image_url) - imageH,imageW,_ = image_array0.shape - #print('###line84:',image_array0.shape) - image_array = cv2.cvtColor( image_array0,cv2.COLOR_RGB2BGR) - #image_in = segmodel.preprocess_image(image_array) - pred = segmodel.eval(image_array,outsize=None) - time0=time.time() - binary = pred.copy() - time1=time.time() - contours, hierarchy = cv2.findContours(binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - time2=time.time() - print(pred.shape,' time copy:%.1f finccontour:%.1f '%(get_ms(time1,time0),get_ms(time2,time1) )) - - - ##计算findconturs时间与大小的关系 - binary0 = binary.copy() - for ii,ss in enumerate([22,256,512,1024,2048]): - time0=time.time() - image = cv2.resize(binary0,(ss,ss)) - time1=time.time() - if ii ==0: - contours, hierarchy = cv2.findContours(binary0,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - else: - contours, hierarchy = cv2.findContours(binary,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - time2=time.time() - print('size:%d resize:%.1f ,findtime:%.1f '%(ss, get_ms(time1,time0),get_ms(time2,time1))) - \ No newline at end of file diff --git a/AIlib/segutils/segmodel.py b/AIlib/segutils/segmodel.py deleted file mode 100644 index 5393290..0000000 --- a/AIlib/segutils/segmodel.py +++ /dev/null @@ -1,140 +0,0 @@ -import torch -import sys,os -sys.path.extend(['../AIlib/segutils']) -from core.models.bisenet import BiSeNet -from torchvision import transforms -import cv2,glob -import numpy as np -from core.models.dinknet import DinkNet34 -import matplotlib.pyplot as plt -import time -class SegModel(object): - def __init__(self, nclass=2,weights=None,modelsize=512,device='cuda:0'): - #self.args = args - self.model = BiSeNet(nclass) - #self.model = DinkNet34(nclass) - checkpoint = torch.load(weights) - self.modelsize = modelsize - self.model.load_state_dict(checkpoint['model']) - self.device = device - self.model= self.model.to(self.device) - '''self.composed_transforms = transforms.Compose([ - - transforms.Normalize(mean=(0.335, 0.358, 0.332), std=(0.141, 0.138, 0.143)), - transforms.ToTensor()]) ''' - self.mean = (0.335, 0.358, 0.332) - self.std = (0.141, 0.138, 0.143) - def eval(self,image): - time0 = time.time() - imageH,imageW,imageC = image.shape - image = self.preprocess_image(image) - time1 = time.time() - self.model.eval() - image = image.to(self.device) - with torch.no_grad(): - output = self.model(image) - - time2 = time.time() - pred = output.data.cpu().numpy() - pred = np.argmax(pred, axis=1)[0]#得到每行 - time3 = time.time() - pred = cv2.resize(pred.astype(np.uint8),(imageW,imageH)) - time4 = time.time() - outstr= 'pre-precess:%.1f ,infer:%.1f ,post-precess:%.1f ,post-resize:%.1f, total:%.1f \n '%( self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3),self.get_ms(time4,time0) ) - - #print('pre-precess:%.1f ,infer:%.1f ,post-precess:%.1f ,post-resize:%.1f, total:%.1f '%( self.get_ms(time1,time0),self.get_ms(time2,time1),self.get_ms(time3,time2),self.get_ms(time4,time3),self.get_ms(time4,time0) )) - return pred,outstr - def get_ms(self,t1,t0): - return (t1-t0)*1000.0 - def preprocess_image(self,image): - - time0 = time.time() - image = cv2.resize(image,(self.modelsize,self.modelsize)) - time0 = time.time() - image = image.astype(np.float32) - image /= 255.0 - - image[:,:,0] -=self.mean[0] - image[:,:,1] -=self.mean[1] - image[:,:,2] -=self.mean[2] - - image[:,:,0] /= self.std[0] - image[:,:,1] /= self.std[1] - image[:,:,2] /= self.std[2] - image = cv2.cvtColor( image,cv2.COLOR_RGB2BGR) - #image -= self.mean - #image /= self.std - image = np.transpose(image, ( 2, 0, 1)) - - image = torch.from_numpy(image).float() - image = image.unsqueeze(0) - - - return image - -def get_ms(t1,t0): - return (t1-t0)*1000.0 - - -def get_largest_contours(contours): - areas = [cv2.contourArea(x) for x in contours] - max_area = max(areas) - max_id = areas.index(max_area) - - return max_id - -if __name__=='__main__': - image_url = '/home/thsw2/WJ/data/THexit/val/images/DJI_0645.JPG' - nclass = 2 - weights = '../weights/segmentation/BiSeNet/checkpoint.pth' - - segmodel = SegModel(nclass=nclass,weights=weights) - - image_urls=glob.glob('/home/thsw2/WJ/data/THexit/val/images/*') - out_dir ='../runs/detect/exp2-seg';os.makedirs(out_dir,exist_ok=True) - for image_url in image_urls[0:1]: - image_url = '/home/thsw2/WJ/data/THexit/val/images/54(199).JPG' - image_array0 = cv2.imread(image_url) - pred = segmodel.eval(image_array0 ) - - #plt.figure(1);plt.imshow(pred); - #plt.show() - binary0 = pred.copy() - - - time0 = time.time() - contours, hierarchy = cv2.findContours(binary0,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - max_id = -1 - if len(contours)>0: - max_id = get_largest_contours(contours) - binary0[:,:] = 0 - print(contours[0].shape,contours[1].shape,contours[0]) - cv2.fillPoly(binary0, [contours[max_id][:,0,:]], 1) - - time1 = time.time() - - #num_labels,_,Areastats,centroids = cv2.connectedComponentsWithStats(binary0,connectivity=4) - time2 = time.time() - - cv2.drawContours(image_array0,contours,max_id,(0,255,255),3) - time3 = time.time() - out_url='%s/%s'%(out_dir,os.path.basename(image_url)) - ret = cv2.imwrite(out_url,image_array0) - time4 = time.time() - - print('image:%s findcontours:%.1f ms , connect:%.1f ms ,draw:%.1f save:%.1f'%(os.path.basename(image_url),get_ms(time1,time0),get_ms(time2,time1), get_ms(time3,time2),get_ms(time4,time3), ) ) - plt.figure(0);plt.imshow(pred) - plt.figure(1);plt.imshow(image_array0) - plt.figure(2);plt.imshow(binary0) - plt.show() - - #print(out_url,ret) - - - - - - - - - diff --git a/AIlib/utils/SendLog/platformQueryOfftask.json b/AIlib/utils/SendLog/platformQueryOfftask.json deleted file mode 100644 index 70aadd1..0000000 --- a/AIlib/utils/SendLog/platformQueryOfftask.json +++ /dev/null @@ -1 +0,0 @@ -{"code": 0, "msg": "操作成功", "data": []} \ No newline at end of file diff --git a/AIlib/utils/__init__.py b/AIlib/utils/__init__.py deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/utils/__pycache__/__init__.cpython-38.pyc b/AIlib/utils/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index fa14c4b..0000000 Binary files a/AIlib/utils/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/__pycache__/autoanchor.cpython-38.pyc b/AIlib/utils/__pycache__/autoanchor.cpython-38.pyc deleted file mode 100644 index 90e65d5..0000000 Binary files a/AIlib/utils/__pycache__/autoanchor.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/__pycache__/datasets.cpython-38.pyc b/AIlib/utils/__pycache__/datasets.cpython-38.pyc deleted file mode 100644 index 767adf9..0000000 Binary files a/AIlib/utils/__pycache__/datasets.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/__pycache__/general.cpython-38.pyc b/AIlib/utils/__pycache__/general.cpython-38.pyc deleted file mode 100644 index dfc8a4a..0000000 Binary files a/AIlib/utils/__pycache__/general.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/__pycache__/get_offline_url.cpython-38.pyc b/AIlib/utils/__pycache__/get_offline_url.cpython-38.pyc deleted file mode 100644 index 844c185..0000000 Binary files a/AIlib/utils/__pycache__/get_offline_url.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/__pycache__/google_utils.cpython-38.pyc b/AIlib/utils/__pycache__/google_utils.cpython-38.pyc deleted file mode 100644 index 9418f69..0000000 Binary files a/AIlib/utils/__pycache__/google_utils.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/__pycache__/loss.cpython-38.pyc b/AIlib/utils/__pycache__/loss.cpython-38.pyc deleted file mode 100644 index ac552e4..0000000 Binary files a/AIlib/utils/__pycache__/loss.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/__pycache__/metrics.cpython-38.pyc b/AIlib/utils/__pycache__/metrics.cpython-38.pyc deleted file mode 100644 index 00772c3..0000000 Binary files a/AIlib/utils/__pycache__/metrics.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/__pycache__/plots.cpython-38.pyc b/AIlib/utils/__pycache__/plots.cpython-38.pyc deleted file mode 100644 index bbef36a..0000000 Binary files a/AIlib/utils/__pycache__/plots.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/__pycache__/torch_utils.cpython-38.pyc b/AIlib/utils/__pycache__/torch_utils.cpython-38.pyc deleted file mode 100644 index 2a7d638..0000000 Binary files a/AIlib/utils/__pycache__/torch_utils.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/activations.py b/AIlib/utils/activations.py deleted file mode 100644 index aa3ddf0..0000000 --- a/AIlib/utils/activations.py +++ /dev/null @@ -1,72 +0,0 @@ -# Activation functions - -import torch -import torch.nn as nn -import torch.nn.functional as F - - -# SiLU https://arxiv.org/pdf/1606.08415.pdf ---------------------------------------------------------------------------- -class SiLU(nn.Module): # export-friendly version of nn.SiLU() - @staticmethod - def forward(x): - return x * torch.sigmoid(x) - - -class Hardswish(nn.Module): # export-friendly version of nn.Hardswish() - @staticmethod - def forward(x): - # return x * F.hardsigmoid(x) # for torchscript and CoreML - return x * F.hardtanh(x + 3, 0., 6.) / 6. # for torchscript, CoreML and ONNX - - -class MemoryEfficientSwish(nn.Module): - class F(torch.autograd.Function): - @staticmethod - def forward(ctx, x): - ctx.save_for_backward(x) - return x * torch.sigmoid(x) - - @staticmethod - def backward(ctx, grad_output): - x = ctx.saved_tensors[0] - sx = torch.sigmoid(x) - return grad_output * (sx * (1 + x * (1 - sx))) - - def forward(self, x): - return self.F.apply(x) - - -# Mish https://github.com/digantamisra98/Mish -------------------------------------------------------------------------- -class Mish(nn.Module): - @staticmethod - def forward(x): - return x * F.softplus(x).tanh() - - -class MemoryEfficientMish(nn.Module): - class F(torch.autograd.Function): - @staticmethod - def forward(ctx, x): - ctx.save_for_backward(x) - return x.mul(torch.tanh(F.softplus(x))) # x * tanh(ln(1 + exp(x))) - - @staticmethod - def backward(ctx, grad_output): - x = ctx.saved_tensors[0] - sx = torch.sigmoid(x) - fx = F.softplus(x).tanh() - return grad_output * (fx + x * sx * (1 - fx * fx)) - - def forward(self, x): - return self.F.apply(x) - - -# FReLU https://arxiv.org/abs/2007.11824 ------------------------------------------------------------------------------- -class FReLU(nn.Module): - def __init__(self, c1, k=3): # ch_in, kernel - super().__init__() - self.conv = nn.Conv2d(c1, c1, k, 1, 1, groups=c1, bias=False) - self.bn = nn.BatchNorm2d(c1) - - def forward(self, x): - return torch.max(x, self.bn(self.conv(x))) diff --git a/AIlib/utils/autoanchor.py b/AIlib/utils/autoanchor.py deleted file mode 100644 index 5777746..0000000 --- a/AIlib/utils/autoanchor.py +++ /dev/null @@ -1,160 +0,0 @@ -# Auto-anchor utils - -import numpy as np -import torch -import yaml -from scipy.cluster.vq import kmeans -from tqdm import tqdm - -from utils.general import colorstr - - -def check_anchor_order(m): - # Check anchor order against stride order for YOLOv5 Detect() module m, and correct if necessary - a = m.anchor_grid.prod(-1).view(-1) # anchor area - da = a[-1] - a[0] # delta a - ds = m.stride[-1] - m.stride[0] # delta s - if da.sign() != ds.sign(): # same order - print('Reversing anchor order') - m.anchors[:] = m.anchors.flip(0) - m.anchor_grid[:] = m.anchor_grid.flip(0) - - -def check_anchors(dataset, model, thr=4.0, imgsz=640): - # Check anchor fit to data, recompute if necessary - prefix = colorstr('autoanchor: ') - print(f'\n{prefix}Analyzing anchors... ', end='') - m = model.module.model[-1] if hasattr(model, 'module') else model.model[-1] # Detect() - shapes = imgsz * dataset.shapes / dataset.shapes.max(1, keepdims=True) - scale = np.random.uniform(0.9, 1.1, size=(shapes.shape[0], 1)) # augment scale - wh = torch.tensor(np.concatenate([l[:, 3:5] * s for s, l in zip(shapes * scale, dataset.labels)])).float() # wh - - def metric(k): # compute metric - r = wh[:, None] / k[None] - x = torch.min(r, 1. / r).min(2)[0] # ratio metric - best = x.max(1)[0] # best_x - aat = (x > 1. / thr).float().sum(1).mean() # anchors above threshold - bpr = (best > 1. / thr).float().mean() # best possible recall - return bpr, aat - - anchors = m.anchor_grid.clone().cpu().view(-1, 2) # current anchors - bpr, aat = metric(anchors) - print(f'anchors/target = {aat:.2f}, Best Possible Recall (BPR) = {bpr:.4f}', end='') - if bpr < 0.98: # threshold to recompute - print('. Attempting to improve anchors, please wait...') - na = m.anchor_grid.numel() // 2 # number of anchors - try: - anchors = kmean_anchors(dataset, n=na, img_size=imgsz, thr=thr, gen=1000, verbose=False) - except Exception as e: - print(f'{prefix}ERROR: {e}') - new_bpr = metric(anchors)[0] - if new_bpr > bpr: # replace anchors - anchors = torch.tensor(anchors, device=m.anchors.device).type_as(m.anchors) - m.anchor_grid[:] = anchors.clone().view_as(m.anchor_grid) # for inference - m.anchors[:] = anchors.clone().view_as(m.anchors) / m.stride.to(m.anchors.device).view(-1, 1, 1) # loss - check_anchor_order(m) - print(f'{prefix}New anchors saved to model. Update model *.yaml to use these anchors in the future.') - else: - print(f'{prefix}Original anchors better than new anchors. Proceeding with original anchors.') - print('') # newline - - -def kmean_anchors(path='./data/coco128.yaml', n=9, img_size=640, thr=4.0, gen=1000, verbose=True): - """ Creates kmeans-evolved anchors from training dataset - - Arguments: - path: path to dataset *.yaml, or a loaded dataset - n: number of anchors - img_size: image size used for training - thr: anchor-label wh ratio threshold hyperparameter hyp['anchor_t'] used for training, default=4.0 - gen: generations to evolve anchors using genetic algorithm - verbose: print all results - - Return: - k: kmeans evolved anchors - - Usage: - from utils.autoanchor import *; _ = kmean_anchors() - """ - thr = 1. / thr - prefix = colorstr('autoanchor: ') - - def metric(k, wh): # compute metrics - r = wh[:, None] / k[None] - x = torch.min(r, 1. / r).min(2)[0] # ratio metric - # x = wh_iou(wh, torch.tensor(k)) # iou metric - return x, x.max(1)[0] # x, best_x - - def anchor_fitness(k): # mutation fitness - _, best = metric(torch.tensor(k, dtype=torch.float32), wh) - return (best * (best > thr).float()).mean() # fitness - - def print_results(k): - k = k[np.argsort(k.prod(1))] # sort small to large - x, best = metric(k, wh0) - bpr, aat = (best > thr).float().mean(), (x > thr).float().mean() * n # best possible recall, anch > thr - print(f'{prefix}thr={thr:.2f}: {bpr:.4f} best possible recall, {aat:.2f} anchors past thr') - print(f'{prefix}n={n}, img_size={img_size}, metric_all={x.mean():.3f}/{best.mean():.3f}-mean/best, ' - f'past_thr={x[x > thr].mean():.3f}-mean: ', end='') - for i, x in enumerate(k): - print('%i,%i' % (round(x[0]), round(x[1])), end=', ' if i < len(k) - 1 else '\n') # use in *.cfg - return k - - if isinstance(path, str): # *.yaml file - with open(path) as f: - data_dict = yaml.load(f, Loader=yaml.SafeLoader) # model dict - from utils.datasets import LoadImagesAndLabels - dataset = LoadImagesAndLabels(data_dict['train'], augment=True, rect=True) - else: - dataset = path # dataset - - # Get label wh - shapes = img_size * dataset.shapes / dataset.shapes.max(1, keepdims=True) - wh0 = np.concatenate([l[:, 3:5] * s for s, l in zip(shapes, dataset.labels)]) # wh - - # Filter - i = (wh0 < 3.0).any(1).sum() - if i: - print(f'{prefix}WARNING: Extremely small objects found. {i} of {len(wh0)} labels are < 3 pixels in size.') - wh = wh0[(wh0 >= 2.0).any(1)] # filter > 2 pixels - # wh = wh * (np.random.rand(wh.shape[0], 1) * 0.9 + 0.1) # multiply by random scale 0-1 - - # Kmeans calculation - print(f'{prefix}Running kmeans for {n} anchors on {len(wh)} points...') - s = wh.std(0) # sigmas for whitening - k, dist = kmeans(wh / s, n, iter=30) # points, mean distance - assert len(k) == n, print(f'{prefix}ERROR: scipy.cluster.vq.kmeans requested {n} points but returned only {len(k)}') - k *= s - wh = torch.tensor(wh, dtype=torch.float32) # filtered - wh0 = torch.tensor(wh0, dtype=torch.float32) # unfiltered - k = print_results(k) - - # Plot - # k, d = [None] * 20, [None] * 20 - # for i in tqdm(range(1, 21)): - # k[i-1], d[i-1] = kmeans(wh / s, i) # points, mean distance - # fig, ax = plt.subplots(1, 2, figsize=(14, 7), tight_layout=True) - # ax = ax.ravel() - # ax[0].plot(np.arange(1, 21), np.array(d) ** 2, marker='.') - # fig, ax = plt.subplots(1, 2, figsize=(14, 7)) # plot wh - # ax[0].hist(wh[wh[:, 0]<100, 0],400) - # ax[1].hist(wh[wh[:, 1]<100, 1],400) - # fig.savefig('wh.png', dpi=200) - - # Evolve - npr = np.random - f, sh, mp, s = anchor_fitness(k), k.shape, 0.9, 0.1 # fitness, generations, mutation prob, sigma - pbar = tqdm(range(gen), desc=f'{prefix}Evolving anchors with Genetic Algorithm:') # progress bar - for _ in pbar: - v = np.ones(sh) - while (v == 1).all(): # mutate until a change occurs (prevent duplicates) - v = ((npr.random(sh) < mp) * npr.random() * npr.randn(*sh) * s + 1).clip(0.3, 3.0) - kg = (k.copy() * v).clip(min=2.0) - fg = anchor_fitness(kg) - if fg > f: - f, k = fg, kg.copy() - pbar.desc = f'{prefix}Evolving anchors with Genetic Algorithm: fitness = {f:.4f}' - if verbose: - print_results(k) - - return print_results(k) diff --git a/AIlib/utils/aws/__init__.py b/AIlib/utils/aws/__init__.py deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/utils/aws/mime.sh b/AIlib/utils/aws/mime.sh deleted file mode 100644 index c319a83..0000000 --- a/AIlib/utils/aws/mime.sh +++ /dev/null @@ -1,26 +0,0 @@ -# AWS EC2 instance startup 'MIME' script https://aws.amazon.com/premiumsupport/knowledge-center/execute-user-data-ec2/ -# This script will run on every instance restart, not only on first start -# --- DO NOT COPY ABOVE COMMENTS WHEN PASTING INTO USERDATA --- - -Content-Type: multipart/mixed; boundary="//" -MIME-Version: 1.0 - ---// -Content-Type: text/cloud-config; charset="us-ascii" -MIME-Version: 1.0 -Content-Transfer-Encoding: 7bit -Content-Disposition: attachment; filename="cloud-config.txt" - -#cloud-config -cloud_final_modules: -- [scripts-user, always] - ---// -Content-Type: text/x-shellscript; charset="us-ascii" -MIME-Version: 1.0 -Content-Transfer-Encoding: 7bit -Content-Disposition: attachment; filename="userdata.txt" - -#!/bin/bash -# --- paste contents of userdata.sh here --- ---// diff --git a/AIlib/utils/aws/resume.py b/AIlib/utils/aws/resume.py deleted file mode 100644 index faad8d2..0000000 --- a/AIlib/utils/aws/resume.py +++ /dev/null @@ -1,37 +0,0 @@ -# Resume all interrupted trainings in yolov5/ dir including DDP trainings -# Usage: $ python utils/aws/resume.py - -import os -import sys -from pathlib import Path - -import torch -import yaml - -sys.path.append('./') # to run '$ python *.py' files in subdirectories - -port = 0 # --master_port -path = Path('').resolve() -for last in path.rglob('*/**/last.pt'): - ckpt = torch.load(last) - if ckpt['optimizer'] is None: - continue - - # Load opt.yaml - with open(last.parent.parent / 'opt.yaml') as f: - opt = yaml.load(f, Loader=yaml.SafeLoader) - - # Get device count - d = opt['device'].split(',') # devices - nd = len(d) # number of devices - ddp = nd > 1 or (nd == 0 and torch.cuda.device_count() > 1) # distributed data parallel - - if ddp: # multi-GPU - port += 1 - cmd = f'python -m torch.distributed.launch --nproc_per_node {nd} --master_port {port} train.py --resume {last}' - else: # single-GPU - cmd = f'python train.py --resume {last}' - - cmd += ' > /dev/null 2>&1 &' # redirect output to dev/null and run in daemon thread - print(cmd) - os.system(cmd) diff --git a/AIlib/utils/aws/userdata.sh b/AIlib/utils/aws/userdata.sh deleted file mode 100644 index 890606b..0000000 --- a/AIlib/utils/aws/userdata.sh +++ /dev/null @@ -1,27 +0,0 @@ -#!/bin/bash -# AWS EC2 instance startup script https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/user-data.html -# This script will run only once on first instance start (for a re-start script see mime.sh) -# /home/ubuntu (ubuntu) or /home/ec2-user (amazon-linux) is working dir -# Use >300 GB SSD - -cd home/ubuntu -if [ ! -d yolov5 ]; then - echo "Running first-time script." # install dependencies, download COCO, pull Docker - git clone https://github.com/ultralytics/yolov5 && sudo chmod -R 777 yolov5 - cd yolov5 - bash data/scripts/get_coco.sh && echo "Data done." & - sudo docker pull ultralytics/yolov5:latest && echo "Docker done." & - python -m pip install --upgrade pip && pip install -r requirements.txt && python detect.py && echo "Requirements done." & - wait && echo "All tasks done." # finish background tasks -else - echo "Running re-start script." # resume interrupted runs - i=0 - list=$(sudo docker ps -qa) # container list i.e. $'one\ntwo\nthree\nfour' - while IFS= read -r id; do - ((i++)) - echo "restarting container $i: $id" - sudo docker start $id - # sudo docker exec -it $id python train.py --resume # single-GPU - sudo docker exec -d $id python utils/aws/resume.py # multi-scenario - done <<<"$list" -fi diff --git a/AIlib/utils/datasets.py b/AIlib/utils/datasets.py deleted file mode 100644 index ed44569..0000000 --- a/AIlib/utils/datasets.py +++ /dev/null @@ -1,1074 +0,0 @@ -# Dataset utils and dataloaders - -import glob -import logging -import math -import os -import random -import shutil -import time -from itertools import repeat -from multiprocessing.pool import ThreadPool -from pathlib import Path -from threading import Thread - -import cv2 -import numpy as np -import torch -import torch.nn.functional as F -from PIL import Image, ExifTags -from torch.utils.data import Dataset -from tqdm import tqdm - -from utils.general import check_requirements, xyxy2xywh, xywh2xyxy, xywhn2xyxy, xyn2xy, segment2box, segments2boxes, \ - resample_segments, clean_str -from utils.torch_utils import torch_distributed_zero_first - -# Parameters -help_url = 'https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data' -img_formats = ['bmp', 'jpg', 'jpeg', 'png', 'tif', 'tiff', 'dng', 'webp', 'mpo'] # acceptable image suffixes -vid_formats = ['mov', 'avi', 'mp4', 'mpg', 'mpeg', 'm4v', 'wmv', 'mkv'] # acceptable video suffixes -logger = logging.getLogger(__name__) - -# Get orientation exif tag -for orientation in ExifTags.TAGS.keys(): - if ExifTags.TAGS[orientation] == 'Orientation': - break - - -def get_hash(files): - # Returns a single hash value of a list of files - return sum(os.path.getsize(f) for f in files if os.path.isfile(f)) - - -def exif_size(img): - # Returns exif-corrected PIL size - s = img.size # (width, height) - try: - rotation = dict(img._getexif().items())[orientation] - if rotation == 6: # rotation 270 - s = (s[1], s[0]) - elif rotation == 8: # rotation 90 - s = (s[1], s[0]) - except: - pass - - return s - - -def create_dataloader(path, imgsz, batch_size, stride, opt, hyp=None, augment=False, cache=False, pad=0.0, rect=False, - rank=-1, world_size=1, workers=8, image_weights=False, quad=False, prefix=''): - # Make sure only the first process in DDP process the dataset first, and the following others can use the cache - with torch_distributed_zero_first(rank): - dataset = LoadImagesAndLabels(path, imgsz, batch_size, - augment=augment, # augment images - hyp=hyp, # augmentation hyperparameters - rect=rect, # rectangular training - cache_images=cache, - single_cls=opt.single_cls, - stride=int(stride), - pad=pad, - image_weights=image_weights, - prefix=prefix) - - batch_size = min(batch_size, len(dataset)) - nw = min([os.cpu_count() // world_size, batch_size if batch_size > 1 else 0, workers]) # number of workers - sampler = torch.utils.data.distributed.DistributedSampler(dataset) if rank != -1 else None - loader = torch.utils.data.DataLoader if image_weights else InfiniteDataLoader - # Use torch.utils.data.DataLoader() if dataset.properties will update during training else InfiniteDataLoader() - dataloader = loader(dataset, - batch_size=batch_size, - num_workers=nw, - sampler=sampler, - pin_memory=True, - collate_fn=LoadImagesAndLabels.collate_fn4 if quad else LoadImagesAndLabels.collate_fn) - return dataloader, dataset - - -class InfiniteDataLoader(torch.utils.data.dataloader.DataLoader): - """ Dataloader that reuses workers - - Uses same syntax as vanilla DataLoader - """ - - def __init__(self, *args, **kwargs): - super().__init__(*args, **kwargs) - object.__setattr__(self, 'batch_sampler', _RepeatSampler(self.batch_sampler)) - self.iterator = super().__iter__() - - def __len__(self): - return len(self.batch_sampler.sampler) - - def __iter__(self): - for i in range(len(self)): - yield next(self.iterator) - - -class _RepeatSampler(object): - """ Sampler that repeats forever - - Args: - sampler (Sampler) - """ - - def __init__(self, sampler): - self.sampler = sampler - - def __iter__(self): - while True: - yield from iter(self.sampler) - - -class LoadImages: # for inference - def __init__(self, path, img_size=640, stride=32): - p = str(Path(path).absolute()) # os-agnostic absolute path - if '*' in p: - files = sorted(glob.glob(p, recursive=True)) # glob - elif os.path.isdir(p): - files = sorted(glob.glob(os.path.join(p, '*.*'))) # dir - elif os.path.isfile(p): - files = [p] # files - else: - raise Exception(f'ERROR: {p} does not exist') - - images = [x for x in files if x.split('.')[-1].lower() in img_formats] - videos = [x for x in files if x.split('.')[-1].lower() in vid_formats] - ni, nv = len(images), len(videos) - - self.img_size = img_size - self.stride = stride - self.files = images + videos - self.nf = ni + nv # number of files - self.video_flag = [False] * ni + [True] * nv - self.mode = 'image' - if any(videos): - self.new_video(videos[0]) # new video - else: - self.cap = None - assert self.nf > 0, f'No images or videos found in {p}. ' \ - f'Supported formats are:\nimages: {img_formats}\nvideos: {vid_formats}' - - def __iter__(self): - self.count = 0 - return self - - def __next__(self): - if self.count == self.nf: - raise StopIteration - path = self.files[self.count] - - if self.video_flag[self.count]: - # Read video - self.mode = 'video' - ret_val, img0 = self.cap.read() - if not ret_val: - self.count += 1 - self.cap.release() - if self.count == self.nf: # last video - raise StopIteration - else: - path = self.files[self.count] - self.new_video(path) - ret_val, img0 = self.cap.read() - - self.frame += 1 - print(f'video {self.count + 1}/{self.nf} ({self.frame}/{self.nframes}) {path}: ', end='') - - else: - # Read image - self.count += 1 - img0 = cv2.imread(path) # BGR - assert img0 is not None, 'Image Not Found ' + path - print(f'image {self.count}/{self.nf} {path}: ', end='') - - # Padded resize - img = letterbox(img0, self.img_size, stride=self.stride)[0] - - # Convert - img = img[:, :, ::-1].transpose(2, 0, 1) # BGR to RGB, to 3x416x416 - img = np.ascontiguousarray(img) - - return path, img, img0, self.cap - - def new_video(self, path): - self.frame = 0 - self.cap = cv2.VideoCapture(path) - self.nframes = int(self.cap.get(cv2.CAP_PROP_FRAME_COUNT)) - - def __len__(self): - return self.nf # number of files - - -class LoadWebcam: # for inference - def __init__(self, pipe='0', img_size=640, stride=32): - self.img_size = img_size - self.stride = stride - - if pipe.isnumeric(): - pipe = eval(pipe) # local camera - # pipe = 'rtsp://192.168.1.64/1' # IP camera - # pipe = 'rtsp://username:password@192.168.1.64/1' # IP camera with login - # pipe = 'http://wmccpinetop.axiscam.net/mjpg/video.mjpg' # IP golf camera - - self.pipe = pipe - self.cap = cv2.VideoCapture(pipe) # video capture object - self.cap.set(cv2.CAP_PROP_BUFFERSIZE, 3) # set buffer size - - def __iter__(self): - self.count = -1 - return self - - def __next__(self): - self.count += 1 - if cv2.waitKey(1) == ord('q'): # q to quit - self.cap.release() - cv2.destroyAllWindows() - raise StopIteration - - # Read frame - if self.pipe == 0: # local camera - ret_val, img0 = self.cap.read() - img0 = cv2.flip(img0, 1) # flip left-right - else: # IP camera - n = 0 - while True: - n += 1 - self.cap.grab() - if n % 30 == 0: # skip frames - ret_val, img0 = self.cap.retrieve() - if ret_val: - break - - # Print - assert ret_val, f'Camera Error {self.pipe}' - img_path = 'webcam.jpg' - print(f'webcam {self.count}: ', end='') - - # Padded resize - img = letterbox(img0, self.img_size, stride=self.stride)[0] - - # Convert - img = img[:, :, ::-1].transpose(2, 0, 1) # BGR to RGB, to 3x416x416 - img = np.ascontiguousarray(img) - - return img_path, img, img0, None - - def __len__(self): - return 0 - - -class LoadStreams: # multiple IP or RTSP cameras - def __init__(self, sources='streams.txt', img_size=640, stride=32): - self.mode = 'stream' - self.img_size = img_size - self.stride = stride - - if os.path.isfile(sources): - with open(sources, 'r') as f: - sources = [x.strip() for x in f.read().strip().splitlines() if len(x.strip())] - else: - sources = [sources] - - n = len(sources) - self.imgs = [None] * n - self.sources = [clean_str(x) for x in sources] # clean source names for later - for i, s in enumerate(sources): # index, source - assert i==0 - # Start thread to read frames from video stream - print(f'{i + 1}/{n}: {s}... ', end='') - if 'youtube.com/' in s or 'youtu.be/' in s: # if source is YouTube video - check_requirements(('pafy', 'youtube_dl')) - import pafy - s = pafy.new(s).getbest(preftype="mp4").url # YouTube URL - s = eval(s) if s.isnumeric() else s # i.e. s = '0' local webcam - cap = cv2.VideoCapture(s) - assert cap.isOpened(), f'Failed to open {s}' - w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH)) - h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)) - self.fps = cap.get(cv2.CAP_PROP_FPS) % 100 - self.cap = cap - _, self.imgs[i] = cap.read() # guarantee first frame - - print(f' success ({w}x{h} at {self.fps:.2f} FPS).') - - print('') # newline - - # check for common shapes - s = np.stack([letterbox(x, self.img_size, stride=self.stride)[0].shape for x in self.imgs], 0) # shapes - self.rect = np.unique(s, axis=0).shape[0] == 1 # rect inference if all shapes equal - if not self.rect: - print('WARNING: Different stream shapes detected. For optimal performance supply similarly-shaped streams.') - - def update(self, index, cap): - frames=2 - # Read next stream frame in a daemon thread - n = 0 - iframe=0 - while cap.isOpened(): - n += 1 - _, self.imgs[index] = cap.read() - iframe +=1 - '''cap.grab() - if n == frames: # read every 4th frame - success, im = cap.retrieve() - self.imgs[index] = im if success else self.imgs[index] * 0 - n = 0''' - #print('###sleep:%.1f ms ,index:%d ,n:%d, iframe:%d'%(1/self.fps*1000,index,n,iframe) ) - #time.sleep(1 / self.fps) # wait time - return self.imgs - def __iter__(self): - self.count = -1 - return self - - def __next__(self): - self.count += 1 - #img0 = self.imgs.copy() - - img0 = self.update(0,self.cap).copy() - if not isinstance(img0[0],np.ndarray): - #print('####video stream :%s error or video ends#####',self.sources) - return False, None, None, None - #if cv2.waitKey(1) == ord('q'): # q to quit - # cv2.destroyAllWindows() - # raise StopIteration - - # Letterbox - img = [letterbox(x, self.img_size, auto=self.rect, stride=self.stride)[0] for x in img0] - - # Stack - img = np.stack(img, 0) - - # Convert - img = img[:, :, :, ::-1].transpose(0, 3, 1, 2) # BGR to RGB, to bsx3x416x416 - img = np.ascontiguousarray(img) - - return self.sources, img, img0, None - - def __len__(self): - return 0 # 1E12 frames = 32 streams at 30 FPS for 30 years - - -def img2label_paths(img_paths): - # Define label paths as a function of image paths - sa, sb = os.sep + 'images' + os.sep, os.sep + 'labels' + os.sep # /images/, /labels/ substrings - return ['txt'.join(x.replace(sa, sb, 1).rsplit(x.split('.')[-1], 1)) for x in img_paths] - - -class LoadImagesAndLabels(Dataset): # for training/testing - def __init__(self, path, img_size=640, batch_size=16, augment=False, hyp=None, rect=False, image_weights=False, - cache_images=False, single_cls=False, stride=32, pad=0.0, prefix=''): - self.img_size = img_size - self.augment = augment - self.hyp = hyp - self.image_weights = image_weights - self.rect = False if image_weights else rect - self.mosaic = self.augment and not self.rect # load 4 images at a time into a mosaic (only during training) - self.mosaic_border = [-img_size // 2, -img_size // 2] - self.stride = stride - self.path = path - - try: - f = [] # image files - for p in path if isinstance(path, list) else [path]: - p = Path(p) # os-agnostic - if p.is_dir(): # dir - f += glob.glob(str(p / '**' / '*.*'), recursive=True) - # f = list(p.rglob('**/*.*')) # pathlib - elif p.is_file(): # file - with open(p, 'r') as t: - t = t.read().strip().splitlines() - parent = str(p.parent) + os.sep - f += [x.replace('./', parent) if x.startswith('./') else x for x in t] # local to global path - # f += [p.parent / x.lstrip(os.sep) for x in t] # local to global path (pathlib) - else: - raise Exception(f'{prefix}{p} does not exist') - self.img_files = sorted([x.replace('/', os.sep) for x in f if x.split('.')[-1].lower() in img_formats]) - # self.img_files = sorted([x for x in f if x.suffix[1:].lower() in img_formats]) # pathlib - assert self.img_files, f'{prefix}No images found' - except Exception as e: - raise Exception(f'{prefix}Error loading data from {path}: {e}\nSee {help_url}') - - # Check cache - self.label_files = img2label_paths(self.img_files) # labels - cache_path = (p if p.is_file() else Path(self.label_files[0]).parent).with_suffix('.cache') # cached labels - if cache_path.is_file(): - cache, exists = torch.load(cache_path), True # load - if cache['hash'] != get_hash(self.label_files + self.img_files) or 'version' not in cache: # changed - cache, exists = self.cache_labels(cache_path, prefix), False # re-cache - else: - cache, exists = self.cache_labels(cache_path, prefix), False # cache - - # Display cache - nf, nm, ne, nc, n = cache.pop('results') # found, missing, empty, corrupted, total - if exists: - d = f"Scanning '{cache_path}' images and labels... {nf} found, {nm} missing, {ne} empty, {nc} corrupted" - tqdm(None, desc=prefix + d, total=n, initial=n) # display cache results - assert nf > 0 or not augment, f'{prefix}No labels in {cache_path}. Can not train without labels. See {help_url}' - - # Read cache - cache.pop('hash') # remove hash - cache.pop('version') # remove version - labels, shapes, self.segments = zip(*cache.values()) - self.labels = list(labels) - self.shapes = np.array(shapes, dtype=np.float64) - self.img_files = list(cache.keys()) # update - self.label_files = img2label_paths(cache.keys()) # update - if single_cls: - for x in self.labels: - x[:, 0] = 0 - - n = len(shapes) # number of images - bi = np.floor(np.arange(n) / batch_size).astype(np.int) # batch index - nb = bi[-1] + 1 # number of batches - self.batch = bi # batch index of image - self.n = n - self.indices = range(n) - - # Rectangular Training - if self.rect: - # Sort by aspect ratio - s = self.shapes # wh - ar = s[:, 1] / s[:, 0] # aspect ratio - irect = ar.argsort() - self.img_files = [self.img_files[i] for i in irect] - self.label_files = [self.label_files[i] for i in irect] - self.labels = [self.labels[i] for i in irect] - self.shapes = s[irect] # wh - ar = ar[irect] - - # Set training image shapes - shapes = [[1, 1]] * nb - for i in range(nb): - ari = ar[bi == i] - mini, maxi = ari.min(), ari.max() - if maxi < 1: - shapes[i] = [maxi, 1] - elif mini > 1: - shapes[i] = [1, 1 / mini] - - self.batch_shapes = np.ceil(np.array(shapes) * img_size / stride + pad).astype(np.int) * stride - - # Cache images into memory for faster training (WARNING: large datasets may exceed system RAM) - self.imgs = [None] * n - if cache_images: - gb = 0 # Gigabytes of cached images - self.img_hw0, self.img_hw = [None] * n, [None] * n - results = ThreadPool(8).imap(lambda x: load_image(*x), zip(repeat(self), range(n))) # 8 threads - pbar = tqdm(enumerate(results), total=n) - for i, x in pbar: - self.imgs[i], self.img_hw0[i], self.img_hw[i] = x # img, hw_original, hw_resized = load_image(self, i) - gb += self.imgs[i].nbytes - pbar.desc = f'{prefix}Caching images ({gb / 1E9:.1f}GB)' - pbar.close() - - def cache_labels(self, path=Path('./labels.cache'), prefix=''): - # Cache dataset labels, check images and read shapes - x = {} # dict - nm, nf, ne, nc = 0, 0, 0, 0 # number missing, found, empty, duplicate - pbar = tqdm(zip(self.img_files, self.label_files), desc='Scanning images', total=len(self.img_files)) - for i, (im_file, lb_file) in enumerate(pbar): - try: - # verify images - im = Image.open(im_file) - im.verify() # PIL verify - shape = exif_size(im) # image size - segments = [] # instance segments - assert (shape[0] > 9) & (shape[1] > 9), f'image size {shape} <10 pixels' - assert im.format.lower() in img_formats, f'invalid image format {im.format}' - - # verify labels - if os.path.isfile(lb_file): - nf += 1 # label found - with open(lb_file, 'r') as f: - l = [x.split() for x in f.read().strip().splitlines()] - if any([len(x) > 8 for x in l]): # is segment - classes = np.array([x[0] for x in l], dtype=np.float32) - segments = [np.array(x[1:], dtype=np.float32).reshape(-1, 2) for x in l] # (cls, xy1...) - l = np.concatenate((classes.reshape(-1, 1), segments2boxes(segments)), 1) # (cls, xywh) - l = np.array(l, dtype=np.float32) - if len(l): - assert l.shape[1] == 5, 'labels require 5 columns each' - assert (l >= 0).all(), 'negative labels' - assert (l[:, 1:] <= 1).all(), 'non-normalized or out of bounds coordinate labels' - assert np.unique(l, axis=0).shape[0] == l.shape[0], 'duplicate labels' - else: - ne += 1 # label empty - l = np.zeros((0, 5), dtype=np.float32) - else: - nm += 1 # label missing - l = np.zeros((0, 5), dtype=np.float32) - x[im_file] = [l, shape, segments] - except Exception as e: - nc += 1 - print(f'{prefix}WARNING: Ignoring corrupted image and/or label {im_file}: {e}') - - pbar.desc = f"{prefix}Scanning '{path.parent / path.stem}' images and labels... " \ - f"{nf} found, {nm} missing, {ne} empty, {nc} corrupted" - pbar.close() - - if nf == 0: - print(f'{prefix}WARNING: No labels found in {path}. See {help_url}') - - x['hash'] = get_hash(self.label_files + self.img_files) - x['results'] = nf, nm, ne, nc, i + 1 - x['version'] = 0.1 # cache version - torch.save(x, path) # save for next time - logging.info(f'{prefix}New cache created: {path}') - return x - - def __len__(self): - return len(self.img_files) - - # def __iter__(self): - # self.count = -1 - # print('ran dataset iter') - # #self.shuffled_vector = np.random.permutation(self.nF) if self.augment else np.arange(self.nF) - # return self - - def __getitem__(self, index): - index = self.indices[index] # linear, shuffled, or image_weights - - hyp = self.hyp - mosaic = self.mosaic and random.random() < hyp['mosaic'] - if mosaic: - # Load mosaic - img, labels = load_mosaic(self, index) - shapes = None - - # MixUp https://arxiv.org/pdf/1710.09412.pdf - if random.random() < hyp['mixup']: - img2, labels2 = load_mosaic(self, random.randint(0, self.n - 1)) - r = np.random.beta(8.0, 8.0) # mixup ratio, alpha=beta=8.0 - img = (img * r + img2 * (1 - r)).astype(np.uint8) - labels = np.concatenate((labels, labels2), 0) - - else: - # Load image - img, (h0, w0), (h, w) = load_image(self, index) - - # Letterbox - shape = self.batch_shapes[self.batch[index]] if self.rect else self.img_size # final letterboxed shape - img, ratio, pad = letterbox(img, shape, auto=False, scaleup=self.augment) - shapes = (h0, w0), ((h / h0, w / w0), pad) # for COCO mAP rescaling - - labels = self.labels[index].copy() - if labels.size: # normalized xywh to pixel xyxy format - labels[:, 1:] = xywhn2xyxy(labels[:, 1:], ratio[0] * w, ratio[1] * h, padw=pad[0], padh=pad[1]) - - if self.augment: - # Augment imagespace - if not mosaic: - img, labels = random_perspective(img, labels, - degrees=hyp['degrees'], - translate=hyp['translate'], - scale=hyp['scale'], - shear=hyp['shear'], - perspective=hyp['perspective']) - - # Augment colorspace - augment_hsv(img, hgain=hyp['hsv_h'], sgain=hyp['hsv_s'], vgain=hyp['hsv_v']) - - # Apply cutouts - # if random.random() < 0.9: - # labels = cutout(img, labels) - - nL = len(labels) # number of labels - if nL: - labels[:, 1:5] = xyxy2xywh(labels[:, 1:5]) # convert xyxy to xywh - labels[:, [2, 4]] /= img.shape[0] # normalized height 0-1 - labels[:, [1, 3]] /= img.shape[1] # normalized width 0-1 - - if self.augment: - # flip up-down - if random.random() < hyp['flipud']: - img = np.flipud(img) - if nL: - labels[:, 2] = 1 - labels[:, 2] - - # flip left-right - if random.random() < hyp['fliplr']: - img = np.fliplr(img) - if nL: - labels[:, 1] = 1 - labels[:, 1] - - labels_out = torch.zeros((nL, 6)) - if nL: - labels_out[:, 1:] = torch.from_numpy(labels) - - # Convert - img = img[:, :, ::-1].transpose(2, 0, 1) # BGR to RGB, to 3x416x416 - img = np.ascontiguousarray(img) - - return torch.from_numpy(img), labels_out, self.img_files[index], shapes - - @staticmethod - def collate_fn(batch): - img, label, path, shapes = zip(*batch) # transposed - for i, l in enumerate(label): - l[:, 0] = i # add target image index for build_targets() - return torch.stack(img, 0), torch.cat(label, 0), path, shapes - - @staticmethod - def collate_fn4(batch): - img, label, path, shapes = zip(*batch) # transposed - n = len(shapes) // 4 - img4, label4, path4, shapes4 = [], [], path[:n], shapes[:n] - - ho = torch.tensor([[0., 0, 0, 1, 0, 0]]) - wo = torch.tensor([[0., 0, 1, 0, 0, 0]]) - s = torch.tensor([[1, 1, .5, .5, .5, .5]]) # scale - for i in range(n): # zidane torch.zeros(16,3,720,1280) # BCHW - i *= 4 - if random.random() < 0.5: - im = F.interpolate(img[i].unsqueeze(0).float(), scale_factor=2., mode='bilinear', align_corners=False)[ - 0].type(img[i].type()) - l = label[i] - else: - im = torch.cat((torch.cat((img[i], img[i + 1]), 1), torch.cat((img[i + 2], img[i + 3]), 1)), 2) - l = torch.cat((label[i], label[i + 1] + ho, label[i + 2] + wo, label[i + 3] + ho + wo), 0) * s - img4.append(im) - label4.append(l) - - for i, l in enumerate(label4): - l[:, 0] = i # add target image index for build_targets() - - return torch.stack(img4, 0), torch.cat(label4, 0), path4, shapes4 - - -# Ancillary functions -------------------------------------------------------------------------------------------------- -def load_image(self, index): - # loads 1 image from dataset, returns img, original hw, resized hw - img = self.imgs[index] - if img is None: # not cached - path = self.img_files[index] - img = cv2.imread(path) # BGR - assert img is not None, 'Image Not Found ' + path - h0, w0 = img.shape[:2] # orig hw - r = self.img_size / max(h0, w0) # resize image to img_size - if r != 1: # always resize down, only resize up if training with augmentation - interp = cv2.INTER_AREA if r < 1 and not self.augment else cv2.INTER_LINEAR - img = cv2.resize(img, (int(w0 * r), int(h0 * r)), interpolation=interp) - return img, (h0, w0), img.shape[:2] # img, hw_original, hw_resized - else: - return self.imgs[index], self.img_hw0[index], self.img_hw[index] # img, hw_original, hw_resized - - -def augment_hsv(img, hgain=0.5, sgain=0.5, vgain=0.5): - r = np.random.uniform(-1, 1, 3) * [hgain, sgain, vgain] + 1 # random gains - hue, sat, val = cv2.split(cv2.cvtColor(img, cv2.COLOR_BGR2HSV)) - dtype = img.dtype # uint8 - - x = np.arange(0, 256, dtype=np.int16) - lut_hue = ((x * r[0]) % 180).astype(dtype) - lut_sat = np.clip(x * r[1], 0, 255).astype(dtype) - lut_val = np.clip(x * r[2], 0, 255).astype(dtype) - - img_hsv = cv2.merge((cv2.LUT(hue, lut_hue), cv2.LUT(sat, lut_sat), cv2.LUT(val, lut_val))).astype(dtype) - cv2.cvtColor(img_hsv, cv2.COLOR_HSV2BGR, dst=img) # no return needed - - -def hist_equalize(img, clahe=True, bgr=False): - # Equalize histogram on BGR image 'img' with img.shape(n,m,3) and range 0-255 - yuv = cv2.cvtColor(img, cv2.COLOR_BGR2YUV if bgr else cv2.COLOR_RGB2YUV) - if clahe: - c = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8)) - yuv[:, :, 0] = c.apply(yuv[:, :, 0]) - else: - yuv[:, :, 0] = cv2.equalizeHist(yuv[:, :, 0]) # equalize Y channel histogram - return cv2.cvtColor(yuv, cv2.COLOR_YUV2BGR if bgr else cv2.COLOR_YUV2RGB) # convert YUV image to RGB - - -def load_mosaic(self, index): - # loads images in a 4-mosaic - - labels4, segments4 = [], [] - s = self.img_size - yc, xc = [int(random.uniform(-x, 2 * s + x)) for x in self.mosaic_border] # mosaic center x, y - indices = [index] + random.choices(self.indices, k=3) # 3 additional image indices - for i, index in enumerate(indices): - # Load image - img, _, (h, w) = load_image(self, index) - - # place img in img4 - if i == 0: # top left - img4 = np.full((s * 2, s * 2, img.shape[2]), 114, dtype=np.uint8) # base image with 4 tiles - x1a, y1a, x2a, y2a = max(xc - w, 0), max(yc - h, 0), xc, yc # xmin, ymin, xmax, ymax (large image) - x1b, y1b, x2b, y2b = w - (x2a - x1a), h - (y2a - y1a), w, h # xmin, ymin, xmax, ymax (small image) - elif i == 1: # top right - x1a, y1a, x2a, y2a = xc, max(yc - h, 0), min(xc + w, s * 2), yc - x1b, y1b, x2b, y2b = 0, h - (y2a - y1a), min(w, x2a - x1a), h - elif i == 2: # bottom left - x1a, y1a, x2a, y2a = max(xc - w, 0), yc, xc, min(s * 2, yc + h) - x1b, y1b, x2b, y2b = w - (x2a - x1a), 0, w, min(y2a - y1a, h) - elif i == 3: # bottom right - x1a, y1a, x2a, y2a = xc, yc, min(xc + w, s * 2), min(s * 2, yc + h) - x1b, y1b, x2b, y2b = 0, 0, min(w, x2a - x1a), min(y2a - y1a, h) - - img4[y1a:y2a, x1a:x2a] = img[y1b:y2b, x1b:x2b] # img4[ymin:ymax, xmin:xmax] - padw = x1a - x1b - padh = y1a - y1b - - # Labels - labels, segments = self.labels[index].copy(), self.segments[index].copy() - if labels.size: - labels[:, 1:] = xywhn2xyxy(labels[:, 1:], w, h, padw, padh) # normalized xywh to pixel xyxy format - segments = [xyn2xy(x, w, h, padw, padh) for x in segments] - labels4.append(labels) - segments4.extend(segments) - - # Concat/clip labels - labels4 = np.concatenate(labels4, 0) - for x in (labels4[:, 1:], *segments4): - np.clip(x, 0, 2 * s, out=x) # clip when using random_perspective() - # img4, labels4 = replicate(img4, labels4) # replicate - - # Augment - img4, labels4 = random_perspective(img4, labels4, segments4, - degrees=self.hyp['degrees'], - translate=self.hyp['translate'], - scale=self.hyp['scale'], - shear=self.hyp['shear'], - perspective=self.hyp['perspective'], - border=self.mosaic_border) # border to remove - - return img4, labels4 - - -def load_mosaic9(self, index): - # loads images in a 9-mosaic - - labels9, segments9 = [], [] - s = self.img_size - indices = [index] + random.choices(self.indices, k=8) # 8 additional image indices - for i, index in enumerate(indices): - # Load image - img, _, (h, w) = load_image(self, index) - - # place img in img9 - if i == 0: # center - img9 = np.full((s * 3, s * 3, img.shape[2]), 114, dtype=np.uint8) # base image with 4 tiles - h0, w0 = h, w - c = s, s, s + w, s + h # xmin, ymin, xmax, ymax (base) coordinates - elif i == 1: # top - c = s, s - h, s + w, s - elif i == 2: # top right - c = s + wp, s - h, s + wp + w, s - elif i == 3: # right - c = s + w0, s, s + w0 + w, s + h - elif i == 4: # bottom right - c = s + w0, s + hp, s + w0 + w, s + hp + h - elif i == 5: # bottom - c = s + w0 - w, s + h0, s + w0, s + h0 + h - elif i == 6: # bottom left - c = s + w0 - wp - w, s + h0, s + w0 - wp, s + h0 + h - elif i == 7: # left - c = s - w, s + h0 - h, s, s + h0 - elif i == 8: # top left - c = s - w, s + h0 - hp - h, s, s + h0 - hp - - padx, pady = c[:2] - x1, y1, x2, y2 = [max(x, 0) for x in c] # allocate coords - - # Labels - labels, segments = self.labels[index].copy(), self.segments[index].copy() - if labels.size: - labels[:, 1:] = xywhn2xyxy(labels[:, 1:], w, h, padx, pady) # normalized xywh to pixel xyxy format - segments = [xyn2xy(x, w, h, padx, pady) for x in segments] - labels9.append(labels) - segments9.extend(segments) - - # Image - img9[y1:y2, x1:x2] = img[y1 - pady:, x1 - padx:] # img9[ymin:ymax, xmin:xmax] - hp, wp = h, w # height, width previous - - # Offset - yc, xc = [int(random.uniform(0, s)) for _ in self.mosaic_border] # mosaic center x, y - img9 = img9[yc:yc + 2 * s, xc:xc + 2 * s] - - # Concat/clip labels - labels9 = np.concatenate(labels9, 0) - labels9[:, [1, 3]] -= xc - labels9[:, [2, 4]] -= yc - c = np.array([xc, yc]) # centers - segments9 = [x - c for x in segments9] - - for x in (labels9[:, 1:], *segments9): - np.clip(x, 0, 2 * s, out=x) # clip when using random_perspective() - # img9, labels9 = replicate(img9, labels9) # replicate - - # Augment - img9, labels9 = random_perspective(img9, labels9, segments9, - degrees=self.hyp['degrees'], - translate=self.hyp['translate'], - scale=self.hyp['scale'], - shear=self.hyp['shear'], - perspective=self.hyp['perspective'], - border=self.mosaic_border) # border to remove - - return img9, labels9 - - -def replicate(img, labels): - # Replicate labels - h, w = img.shape[:2] - boxes = labels[:, 1:].astype(int) - x1, y1, x2, y2 = boxes.T - s = ((x2 - x1) + (y2 - y1)) / 2 # side length (pixels) - for i in s.argsort()[:round(s.size * 0.5)]: # smallest indices - x1b, y1b, x2b, y2b = boxes[i] - bh, bw = y2b - y1b, x2b - x1b - yc, xc = int(random.uniform(0, h - bh)), int(random.uniform(0, w - bw)) # offset x, y - x1a, y1a, x2a, y2a = [xc, yc, xc + bw, yc + bh] - img[y1a:y2a, x1a:x2a] = img[y1b:y2b, x1b:x2b] # img4[ymin:ymax, xmin:xmax] - labels = np.append(labels, [[labels[i, 0], x1a, y1a, x2a, y2a]], axis=0) - - return img, labels - - -def letterbox(img, new_shape=(640, 640), color=(114, 114, 114), auto=True, scaleFill=False, scaleup=True, stride=32): - # Resize and pad image while meeting stride-multiple constraints - shape = img.shape[:2] # current shape [height, width] - if isinstance(new_shape, int): - new_shape = (new_shape, new_shape) - - # Scale ratio (new / old) - r = min(new_shape[0] / shape[0], new_shape[1] / shape[1]) - if not scaleup: # only scale down, do not scale up (for better test mAP) - r = min(r, 1.0) - - # Compute padding - ratio = r, r # width, height ratios - new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r)) - dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1] # wh padding - if auto: # minimum rectangle - dw, dh = np.mod(dw, stride), np.mod(dh, stride) # wh padding - elif scaleFill: # stretch - dw, dh = 0.0, 0.0 - new_unpad = (new_shape[1], new_shape[0]) - ratio = new_shape[1] / shape[1], new_shape[0] / shape[0] # width, height ratios - - dw /= 2 # divide padding into 2 sides - dh /= 2 - - if shape[::-1] != new_unpad: # resize - img = cv2.resize(img, new_unpad, interpolation=cv2.INTER_LINEAR) - top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1)) - left, right = int(round(dw - 0.1)), int(round(dw + 0.1)) - img = cv2.copyMakeBorder(img, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color) # add border - return img, ratio, (dw, dh) - - -def random_perspective(img, targets=(), segments=(), degrees=10, translate=.1, scale=.1, shear=10, perspective=0.0, - border=(0, 0)): - # torchvision.transforms.RandomAffine(degrees=(-10, 10), translate=(.1, .1), scale=(.9, 1.1), shear=(-10, 10)) - # targets = [cls, xyxy] - - height = img.shape[0] + border[0] * 2 # shape(h,w,c) - width = img.shape[1] + border[1] * 2 - - # Center - C = np.eye(3) - C[0, 2] = -img.shape[1] / 2 # x translation (pixels) - C[1, 2] = -img.shape[0] / 2 # y translation (pixels) - - # Perspective - P = np.eye(3) - P[2, 0] = random.uniform(-perspective, perspective) # x perspective (about y) - P[2, 1] = random.uniform(-perspective, perspective) # y perspective (about x) - - # Rotation and Scale - R = np.eye(3) - a = random.uniform(-degrees, degrees) - # a += random.choice([-180, -90, 0, 90]) # add 90deg rotations to small rotations - s = random.uniform(1 - scale, 1 + scale) - # s = 2 ** random.uniform(-scale, scale) - R[:2] = cv2.getRotationMatrix2D(angle=a, center=(0, 0), scale=s) - - # Shear - S = np.eye(3) - S[0, 1] = math.tan(random.uniform(-shear, shear) * math.pi / 180) # x shear (deg) - S[1, 0] = math.tan(random.uniform(-shear, shear) * math.pi / 180) # y shear (deg) - - # Translation - T = np.eye(3) - T[0, 2] = random.uniform(0.5 - translate, 0.5 + translate) * width # x translation (pixels) - T[1, 2] = random.uniform(0.5 - translate, 0.5 + translate) * height # y translation (pixels) - - # Combined rotation matrix - M = T @ S @ R @ P @ C # order of operations (right to left) is IMPORTANT - if (border[0] != 0) or (border[1] != 0) or (M != np.eye(3)).any(): # image changed - if perspective: - img = cv2.warpPerspective(img, M, dsize=(width, height), borderValue=(114, 114, 114)) - else: # affine - img = cv2.warpAffine(img, M[:2], dsize=(width, height), borderValue=(114, 114, 114)) - - # Visualize - # import matplotlib.pyplot as plt - # ax = plt.subplots(1, 2, figsize=(12, 6))[1].ravel() - # ax[0].imshow(img[:, :, ::-1]) # base - # ax[1].imshow(img2[:, :, ::-1]) # warped - - # Transform label coordinates - n = len(targets) - if n: - use_segments = any(x.any() for x in segments) - new = np.zeros((n, 4)) - if use_segments: # warp segments - segments = resample_segments(segments) # upsample - for i, segment in enumerate(segments): - xy = np.ones((len(segment), 3)) - xy[:, :2] = segment - xy = xy @ M.T # transform - xy = xy[:, :2] / xy[:, 2:3] if perspective else xy[:, :2] # perspective rescale or affine - - # clip - new[i] = segment2box(xy, width, height) - - else: # warp boxes - xy = np.ones((n * 4, 3)) - xy[:, :2] = targets[:, [1, 2, 3, 4, 1, 4, 3, 2]].reshape(n * 4, 2) # x1y1, x2y2, x1y2, x2y1 - xy = xy @ M.T # transform - xy = (xy[:, :2] / xy[:, 2:3] if perspective else xy[:, :2]).reshape(n, 8) # perspective rescale or affine - - # create new boxes - x = xy[:, [0, 2, 4, 6]] - y = xy[:, [1, 3, 5, 7]] - new = np.concatenate((x.min(1), y.min(1), x.max(1), y.max(1))).reshape(4, n).T - - # clip - new[:, [0, 2]] = new[:, [0, 2]].clip(0, width) - new[:, [1, 3]] = new[:, [1, 3]].clip(0, height) - - # filter candidates - i = box_candidates(box1=targets[:, 1:5].T * s, box2=new.T, area_thr=0.01 if use_segments else 0.10) - targets = targets[i] - targets[:, 1:5] = new[i] - - return img, targets - - -def box_candidates(box1, box2, wh_thr=2, ar_thr=20, area_thr=0.1, eps=1e-16): # box1(4,n), box2(4,n) - # Compute candidate boxes: box1 before augment, box2 after augment, wh_thr (pixels), aspect_ratio_thr, area_ratio - w1, h1 = box1[2] - box1[0], box1[3] - box1[1] - w2, h2 = box2[2] - box2[0], box2[3] - box2[1] - ar = np.maximum(w2 / (h2 + eps), h2 / (w2 + eps)) # aspect ratio - return (w2 > wh_thr) & (h2 > wh_thr) & (w2 * h2 / (w1 * h1 + eps) > area_thr) & (ar < ar_thr) # candidates - - -def cutout(image, labels): - # Applies image cutout augmentation https://arxiv.org/abs/1708.04552 - h, w = image.shape[:2] - - def bbox_ioa(box1, box2): - # Returns the intersection over box2 area given box1, box2. box1 is 4, box2 is nx4. boxes are x1y1x2y2 - box2 = box2.transpose() - - # Get the coordinates of bounding boxes - b1_x1, b1_y1, b1_x2, b1_y2 = box1[0], box1[1], box1[2], box1[3] - b2_x1, b2_y1, b2_x2, b2_y2 = box2[0], box2[1], box2[2], box2[3] - - # Intersection area - inter_area = (np.minimum(b1_x2, b2_x2) - np.maximum(b1_x1, b2_x1)).clip(0) * \ - (np.minimum(b1_y2, b2_y2) - np.maximum(b1_y1, b2_y1)).clip(0) - - # box2 area - box2_area = (b2_x2 - b2_x1) * (b2_y2 - b2_y1) + 1e-16 - - # Intersection over box2 area - return inter_area / box2_area - - # create random masks - scales = [0.5] * 1 + [0.25] * 2 + [0.125] * 4 + [0.0625] * 8 + [0.03125] * 16 # image size fraction - for s in scales: - mask_h = random.randint(1, int(h * s)) - mask_w = random.randint(1, int(w * s)) - - # box - xmin = max(0, random.randint(0, w) - mask_w // 2) - ymin = max(0, random.randint(0, h) - mask_h // 2) - xmax = min(w, xmin + mask_w) - ymax = min(h, ymin + mask_h) - - # apply random color mask - image[ymin:ymax, xmin:xmax] = [random.randint(64, 191) for _ in range(3)] - - # return unobscured labels - if len(labels) and s > 0.03: - box = np.array([xmin, ymin, xmax, ymax], dtype=np.float32) - ioa = bbox_ioa(box, labels[:, 1:5]) # intersection over area - labels = labels[ioa < 0.60] # remove >60% obscured labels - - return labels - - -def create_folder(path='./new'): - # Create folder - if os.path.exists(path): - shutil.rmtree(path) # delete output folder - os.makedirs(path) # make new output folder - - -def flatten_recursive(path='../coco128'): - # Flatten a recursive directory by bringing all files to top level - new_path = Path(path + '_flat') - create_folder(new_path) - for file in tqdm(glob.glob(str(Path(path)) + '/**/*.*', recursive=True)): - shutil.copyfile(file, new_path / Path(file).name) - - -def extract_boxes(path='../coco128/'): # from utils.datasets import *; extract_boxes('../coco128') - # Convert detection dataset into classification dataset, with one directory per class - - path = Path(path) # images dir - shutil.rmtree(path / 'classifier') if (path / 'classifier').is_dir() else None # remove existing - files = list(path.rglob('*.*')) - n = len(files) # number of files - for im_file in tqdm(files, total=n): - if im_file.suffix[1:] in img_formats: - # image - im = cv2.imread(str(im_file))[..., ::-1] # BGR to RGB - h, w = im.shape[:2] - - # labels - lb_file = Path(img2label_paths([str(im_file)])[0]) - if Path(lb_file).exists(): - with open(lb_file, 'r') as f: - lb = np.array([x.split() for x in f.read().strip().splitlines()], dtype=np.float32) # labels - - for j, x in enumerate(lb): - c = int(x[0]) # class - f = (path / 'classifier') / f'{c}' / f'{path.stem}_{im_file.stem}_{j}.jpg' # new filename - if not f.parent.is_dir(): - f.parent.mkdir(parents=True) - - b = x[1:] * [w, h, w, h] # box - # b[2:] = b[2:].max() # rectangle to square - b[2:] = b[2:] * 1.2 + 3 # pad - b = xywh2xyxy(b.reshape(-1, 4)).ravel().astype(np.int) - - b[[0, 2]] = np.clip(b[[0, 2]], 0, w) # clip boxes outside of image - b[[1, 3]] = np.clip(b[[1, 3]], 0, h) - assert cv2.imwrite(str(f), im[b[1]:b[3], b[0]:b[2]]), f'box failure in {f}' - - -def autosplit(path='../coco128', weights=(0.9, 0.1, 0.0), annotated_only=False): - """ Autosplit a dataset into train/val/test splits and save path/autosplit_*.txt files - Usage: from utils.datasets import *; autosplit('../coco128') - Arguments - path: Path to images directory - weights: Train, val, test weights (list) - annotated_only: Only use images with an annotated txt file - """ - path = Path(path) # images dir - files = sum([list(path.rglob(f"*.{img_ext}")) for img_ext in img_formats], []) # image files only - n = len(files) # number of files - indices = random.choices([0, 1, 2], weights=weights, k=n) # assign each image to a split - - txt = ['autosplit_train.txt', 'autosplit_val.txt', 'autosplit_test.txt'] # 3 txt files - [(path / x).unlink() for x in txt if (path / x).exists()] # remove existing - - print(f'Autosplitting images from {path}' + ', using *.txt labeled images only' * annotated_only) - for i, img in tqdm(zip(indices, files), total=n): - if not annotated_only or Path(img2label_paths([str(img)])[0]).exists(): # check label - with open(path / txt[i], 'a') as f: - f.write(str(img) + '\n') # add image to txt file diff --git a/AIlib/utils/flask_rest_api/README.md b/AIlib/utils/flask_rest_api/README.md deleted file mode 100644 index 0cdc51b..0000000 --- a/AIlib/utils/flask_rest_api/README.md +++ /dev/null @@ -1,51 +0,0 @@ -# Flask REST API -[REST](https://en.wikipedia.org/wiki/Representational_state_transfer) [API](https://en.wikipedia.org/wiki/API)s are commonly used to expose Machine Learning (ML) models to other services. This folder contains an example REST API created using Flask to expose the `yolov5s` model from [PyTorch Hub](https://pytorch.org/hub/ultralytics_yolov5/). - -## Requirements - -[Flask](https://palletsprojects.com/p/flask/) is required. Install with: -```shell -$ pip install Flask -``` - -## Run - -After Flask installation run: - -```shell -$ python3 restapi.py --port 5000 -``` - -Then use [curl](https://curl.se/) to perform a request: - -```shell -$ curl -X POST -F image=@zidane.jpg 'http://localhost:5000/v1/object-detection/yolov5s'` -``` - -The model inference results are returned: - -```shell -[{'class': 0, - 'confidence': 0.8197850585, - 'name': 'person', - 'xmax': 1159.1403808594, - 'xmin': 750.912902832, - 'ymax': 711.2583007812, - 'ymin': 44.0350036621}, - {'class': 0, - 'confidence': 0.5667674541, - 'name': 'person', - 'xmax': 1065.5523681641, - 'xmin': 116.0448303223, - 'ymax': 713.8904418945, - 'ymin': 198.4603881836}, - {'class': 27, - 'confidence': 0.5661227107, - 'name': 'tie', - 'xmax': 516.7975463867, - 'xmin': 416.6880187988, - 'ymax': 717.0524902344, - 'ymin': 429.2020568848}] -``` - -An example python script to perform inference using [requests](https://docs.python-requests.org/en/master/) is given in `example_request.py` diff --git a/AIlib/utils/flask_rest_api/example_request.py b/AIlib/utils/flask_rest_api/example_request.py deleted file mode 100644 index ff21f30..0000000 --- a/AIlib/utils/flask_rest_api/example_request.py +++ /dev/null @@ -1,13 +0,0 @@ -"""Perform test request""" -import pprint - -import requests - -DETECTION_URL = "http://localhost:5000/v1/object-detection/yolov5s" -TEST_IMAGE = "zidane.jpg" - -image_data = open(TEST_IMAGE, "rb").read() - -response = requests.post(DETECTION_URL, files={"image": image_data}).json() - -pprint.pprint(response) diff --git a/AIlib/utils/flask_rest_api/restapi.py b/AIlib/utils/flask_rest_api/restapi.py deleted file mode 100644 index 9d88f61..0000000 --- a/AIlib/utils/flask_rest_api/restapi.py +++ /dev/null @@ -1,38 +0,0 @@ -""" -Run a rest API exposing the yolov5s object detection model -""" -import argparse -import io - -import torch -from PIL import Image -from flask import Flask, request - -app = Flask(__name__) - -DETECTION_URL = "/v1/object-detection/yolov5s" - - -@app.route(DETECTION_URL, methods=["POST"]) -def predict(): - if not request.method == "POST": - return - - if request.files.get("image"): - image_file = request.files["image"] - image_bytes = image_file.read() - - img = Image.open(io.BytesIO(image_bytes)) - - results = model(img, size=640) - data = results.pandas().xyxy[0].to_json(orient="records") - return data - - -if __name__ == "__main__": - parser = argparse.ArgumentParser(description="Flask api exposing yolov5 model") - parser.add_argument("--port", default=5000, type=int, help="port number") - args = parser.parse_args() - - model = torch.hub.load("ultralytics/yolov5", "yolov5s", force_reload=True).autoshape() # force_reload to recache - app.run(host="0.0.0.0", port=args.port) # debug=True causes Restarting with stat diff --git a/AIlib/utils/general.py b/AIlib/utils/general.py deleted file mode 100644 index ac3a698..0000000 --- a/AIlib/utils/general.py +++ /dev/null @@ -1,604 +0,0 @@ -# YOLOv5 general utils - -import glob -import logging -import math -import os -import platform -import random -import re -import subprocess -import time -from pathlib import Path - -import cv2 -import numpy as np -import pandas as pd -import torch -import torchvision -import yaml - -from utils.google_utils import gsutil_getsize -from utils.metrics import fitness -from utils.torch_utils import init_torch_seeds - -# Settings -torch.set_printoptions(linewidth=320, precision=5, profile='long') -np.set_printoptions(linewidth=320, formatter={'float_kind': '{:11.5g}'.format}) # format short g, %precision=5 -pd.options.display.max_columns = 10 -cv2.setNumThreads(0) # prevent OpenCV from multithreading (incompatible with PyTorch DataLoader) -os.environ['NUMEXPR_MAX_THREADS'] = str(min(os.cpu_count(), 8)) # NumExpr max threads - - -def set_logging(rank=-1): - logging.basicConfig( - format="%(message)s", - level=logging.INFO if rank in [-1, 0] else logging.WARN) - - -def init_seeds(seed=0): - # Initialize random number generator (RNG) seeds - random.seed(seed) - np.random.seed(seed) - init_torch_seeds(seed) - - -def get_latest_run(search_dir='.'): - # Return path to most recent 'last.pt' in /runs (i.e. to --resume from) - last_list = glob.glob(f'{search_dir}/**/last*.pt', recursive=True) - return max(last_list, key=os.path.getctime) if last_list else '' - - -def isdocker(): - # Is environment a Docker container - return Path('/workspace').exists() # or Path('/.dockerenv').exists() - - -def emojis(str=''): - # Return platform-dependent emoji-safe version of string - return str.encode().decode('ascii', 'ignore') if platform.system() == 'Windows' else str - - -def check_online(): - # Check internet connectivity - import socket - try: - socket.create_connection(("1.1.1.1", 443), 5) # check host accesability - return True - except OSError: - return False - - -def check_git_status(): - # Recommend 'git pull' if code is out of date - print(colorstr('github: '), end='') - try: - assert Path('.git').exists(), 'skipping check (not a git repository)' - assert not isdocker(), 'skipping check (Docker image)' - assert check_online(), 'skipping check (offline)' - - cmd = 'git fetch && git config --get remote.origin.url' - url = subprocess.check_output(cmd, shell=True).decode().strip().rstrip('.git') # github repo url - branch = subprocess.check_output('git rev-parse --abbrev-ref HEAD', shell=True).decode().strip() # checked out - n = int(subprocess.check_output(f'git rev-list {branch}..origin/master --count', shell=True)) # commits behind - if n > 0: - s = f"⚠️ WARNING: code is out of date by {n} commit{'s' * (n > 1)}. " \ - f"Use 'git pull' to update or 'git clone {url}' to download latest." - else: - s = f'up to date with {url} ✅' - print(emojis(s)) # emoji-safe - except Exception as e: - print(e) - - -def check_requirements(requirements='requirements.txt', exclude=()): - # Check installed dependencies meet requirements (pass *.txt file or list of packages) - import pkg_resources as pkg - prefix = colorstr('red', 'bold', 'requirements:') - if isinstance(requirements, (str, Path)): # requirements.txt file - file = Path(requirements) - if not file.exists(): - print(f"{prefix} {file.resolve()} not found, check failed.") - return - requirements = [f'{x.name}{x.specifier}' for x in pkg.parse_requirements(file.open()) if x.name not in exclude] - else: # list or tuple of packages - requirements = [x for x in requirements if x not in exclude] - - n = 0 # number of packages updates - for r in requirements: - try: - pkg.require(r) - except Exception as e: # DistributionNotFound or VersionConflict if requirements not met - n += 1 - print(f"{prefix} {e.req} not found and is required by YOLOv5, attempting auto-update...") - print(subprocess.check_output(f"pip install {e.req}", shell=True).decode()) - - if n: # if packages updated - source = file.resolve() if 'file' in locals() else requirements - s = f"{prefix} {n} package{'s' * (n > 1)} updated per {source}\n" \ - f"{prefix} ⚠️ {colorstr('bold', 'Restart runtime or rerun command for updates to take effect')}\n" - print(emojis(s)) # emoji-safe - - -def check_img_size(img_size, s=32): - # Verify img_size is a multiple of stride s - new_size = make_divisible(img_size, int(s)) # ceil gs-multiple - if new_size != img_size: - print('WARNING: --img-size %g must be multiple of max stride %g, updating to %g' % (img_size, s, new_size)) - return new_size - - -def check_imshow(): - # Check if environment supports image displays - try: - assert not isdocker(), 'cv2.imshow() is disabled in Docker environments' - cv2.imshow('test', np.zeros((1, 1, 3))) - cv2.waitKey(1) - cv2.destroyAllWindows() - cv2.waitKey(1) - return True - except Exception as e: - print(f'WARNING: Environment does not support cv2.imshow() or PIL Image.show() image displays\n{e}') - return False - - -def check_file(file): - # Search for file if not found - if Path(file).is_file() or file == '': - return file - else: - files = glob.glob('./**/' + file, recursive=True) # find file - assert len(files), f'File Not Found: {file}' # assert file was found - assert len(files) == 1, f"Multiple files match '{file}', specify exact path: {files}" # assert unique - return files[0] # return file - - -def check_dataset(dict): - # Download dataset if not found locally - val, s = dict.get('val'), dict.get('download') - if val and len(val): - val = [Path(x).resolve() for x in (val if isinstance(val, list) else [val])] # val path - if not all(x.exists() for x in val): - print('\nWARNING: Dataset not found, nonexistent paths: %s' % [str(x) for x in val if not x.exists()]) - if s and len(s): # download script - print('Downloading %s ...' % s) - if s.startswith('http') and s.endswith('.zip'): # URL - f = Path(s).name # filename - torch.hub.download_url_to_file(s, f) - r = os.system('unzip -q %s -d ../ && rm %s' % (f, f)) # unzip - else: # bash script - r = os.system(s) - print('Dataset autodownload %s\n' % ('success' if r == 0 else 'failure')) # analyze return value - else: - raise Exception('Dataset not found.') - - -def make_divisible(x, divisor): - # Returns x evenly divisible by divisor - return math.ceil(x / divisor) * divisor - - -def clean_str(s): - # Cleans a string by replacing special characters with underscore _ - return re.sub(pattern="[|@#!¡·$€%&()=?¿^*;:,¨´><+]", repl="_", string=s) - - -def one_cycle(y1=0.0, y2=1.0, steps=100): - # lambda function for sinusoidal ramp from y1 to y2 - return lambda x: ((1 - math.cos(x * math.pi / steps)) / 2) * (y2 - y1) + y1 - - -def colorstr(*input): - # Colors a string https://en.wikipedia.org/wiki/ANSI_escape_code, i.e. colorstr('blue', 'hello world') - *args, string = input if len(input) > 1 else ('blue', 'bold', input[0]) # color arguments, string - colors = {'black': '\033[30m', # basic colors - 'red': '\033[31m', - 'green': '\033[32m', - 'yellow': '\033[33m', - 'blue': '\033[34m', - 'magenta': '\033[35m', - 'cyan': '\033[36m', - 'white': '\033[37m', - 'bright_black': '\033[90m', # bright colors - 'bright_red': '\033[91m', - 'bright_green': '\033[92m', - 'bright_yellow': '\033[93m', - 'bright_blue': '\033[94m', - 'bright_magenta': '\033[95m', - 'bright_cyan': '\033[96m', - 'bright_white': '\033[97m', - 'end': '\033[0m', # misc - 'bold': '\033[1m', - 'underline': '\033[4m'} - return ''.join(colors[x] for x in args) + f'{string}' + colors['end'] - - -def labels_to_class_weights(labels, nc=80): - # Get class weights (inverse frequency) from training labels - if labels[0] is None: # no labels loaded - return torch.Tensor() - - labels = np.concatenate(labels, 0) # labels.shape = (866643, 5) for COCO - classes = labels[:, 0].astype(np.int) # labels = [class xywh] - weights = np.bincount(classes, minlength=nc) # occurrences per class - - # Prepend gridpoint count (for uCE training) - # gpi = ((320 / 32 * np.array([1, 2, 4])) ** 2 * 3).sum() # gridpoints per image - # weights = np.hstack([gpi * len(labels) - weights.sum() * 9, weights * 9]) ** 0.5 # prepend gridpoints to start - - weights[weights == 0] = 1 # replace empty bins with 1 - weights = 1 / weights # number of targets per class - weights /= weights.sum() # normalize - return torch.from_numpy(weights) - - -def labels_to_image_weights(labels, nc=80, class_weights=np.ones(80)): - # Produces image weights based on class_weights and image contents - class_counts = np.array([np.bincount(x[:, 0].astype(np.int), minlength=nc) for x in labels]) - image_weights = (class_weights.reshape(1, nc) * class_counts).sum(1) - # index = random.choices(range(n), weights=image_weights, k=1) # weight image sample - return image_weights - - -def coco80_to_coco91_class(): # converts 80-index (val2014) to 91-index (paper) - # https://tech.amikelive.com/node-718/what-object-categories-labels-are-in-coco-dataset/ - # a = np.loadtxt('data/coco.names', dtype='str', delimiter='\n') - # b = np.loadtxt('data/coco_paper.names', dtype='str', delimiter='\n') - # x1 = [list(a[i] == b).index(True) + 1 for i in range(80)] # darknet to coco - # x2 = [list(b[i] == a).index(True) if any(b[i] == a) else None for i in range(91)] # coco to darknet - x = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 27, 28, 31, 32, 33, 34, - 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, - 64, 65, 67, 70, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 84, 85, 86, 87, 88, 89, 90] - return x - - -def xyxy2xywh(x): - # Convert nx4 boxes from [x1, y1, x2, y2] to [x, y, w, h] where xy1=top-left, xy2=bottom-right - y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) - y[:, 0] = (x[:, 0] + x[:, 2]) / 2 # x center - y[:, 1] = (x[:, 1] + x[:, 3]) / 2 # y center - y[:, 2] = x[:, 2] - x[:, 0] # width - y[:, 3] = x[:, 3] - x[:, 1] # height - return y - - -def xywh2xyxy(x): - # Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right - y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) - y[:, 0] = x[:, 0] - x[:, 2] / 2 # top left x - y[:, 1] = x[:, 1] - x[:, 3] / 2 # top left y - y[:, 2] = x[:, 0] + x[:, 2] / 2 # bottom right x - y[:, 3] = x[:, 1] + x[:, 3] / 2 # bottom right y - return y - - -def xywhn2xyxy(x, w=640, h=640, padw=0, padh=0): - # Convert nx4 boxes from [x, y, w, h] normalized to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right - y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) - y[:, 0] = w * (x[:, 0] - x[:, 2] / 2) + padw # top left x - y[:, 1] = h * (x[:, 1] - x[:, 3] / 2) + padh # top left y - y[:, 2] = w * (x[:, 0] + x[:, 2] / 2) + padw # bottom right x - y[:, 3] = h * (x[:, 1] + x[:, 3] / 2) + padh # bottom right y - return y - - -def xyn2xy(x, w=640, h=640, padw=0, padh=0): - # Convert normalized segments into pixel segments, shape (n,2) - y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x) - y[:, 0] = w * x[:, 0] + padw # top left x - y[:, 1] = h * x[:, 1] + padh # top left y - return y - - -def segment2box(segment, width=640, height=640): - # Convert 1 segment label to 1 box label, applying inside-image constraint, i.e. (xy1, xy2, ...) to (xyxy) - x, y = segment.T # segment xy - inside = (x >= 0) & (y >= 0) & (x <= width) & (y <= height) - x, y, = x[inside], y[inside] - return np.array([x.min(), y.min(), x.max(), y.max()]) if any(x) else np.zeros((1, 4)) # xyxy - - -def segments2boxes(segments): - # Convert segment labels to box labels, i.e. (cls, xy1, xy2, ...) to (cls, xywh) - boxes = [] - for s in segments: - x, y = s.T # segment xy - boxes.append([x.min(), y.min(), x.max(), y.max()]) # cls, xyxy - return xyxy2xywh(np.array(boxes)) # cls, xywh - - -def resample_segments(segments, n=1000): - # Up-sample an (n,2) segment - for i, s in enumerate(segments): - x = np.linspace(0, len(s) - 1, n) - xp = np.arange(len(s)) - segments[i] = np.concatenate([np.interp(x, xp, s[:, i]) for i in range(2)]).reshape(2, -1).T # segment xy - return segments - - -def scale_coords(img1_shape, coords, img0_shape, ratio_pad=None): - # Rescale coords (xyxy) from img1_shape to img0_shape - if ratio_pad is None: # calculate from img0_shape - gain = min(img1_shape[0] / img0_shape[0], img1_shape[1] / img0_shape[1]) # gain = old / new - pad = (img1_shape[1] - img0_shape[1] * gain) / 2, (img1_shape[0] - img0_shape[0] * gain) / 2 # wh padding - else: - gain = ratio_pad[0][0] - pad = ratio_pad[1] - - coords[:, [0, 2]] -= pad[0] # x padding - coords[:, [1, 3]] -= pad[1] # y padding - coords[:, :4] /= gain - clip_coords(coords, img0_shape) - return coords - - -def clip_coords(boxes, img_shape): - # Clip bounding xyxy bounding boxes to image shape (height, width) - boxes[:, 0].clamp_(0, img_shape[1]) # x1 - boxes[:, 1].clamp_(0, img_shape[0]) # y1 - boxes[:, 2].clamp_(0, img_shape[1]) # x2 - boxes[:, 3].clamp_(0, img_shape[0]) # y2 - - -def bbox_iou(box1, box2, x1y1x2y2=True, GIoU=False, DIoU=False, CIoU=False, eps=1e-7): - # Returns the IoU of box1 to box2. box1 is 4, box2 is nx4 - box2 = box2.T - - # Get the coordinates of bounding boxes - if x1y1x2y2: # x1, y1, x2, y2 = box1 - b1_x1, b1_y1, b1_x2, b1_y2 = box1[0], box1[1], box1[2], box1[3] - b2_x1, b2_y1, b2_x2, b2_y2 = box2[0], box2[1], box2[2], box2[3] - else: # transform from xywh to xyxy - b1_x1, b1_x2 = box1[0] - box1[2] / 2, box1[0] + box1[2] / 2 - b1_y1, b1_y2 = box1[1] - box1[3] / 2, box1[1] + box1[3] / 2 - b2_x1, b2_x2 = box2[0] - box2[2] / 2, box2[0] + box2[2] / 2 - b2_y1, b2_y2 = box2[1] - box2[3] / 2, box2[1] + box2[3] / 2 - - # Intersection area - inter = (torch.min(b1_x2, b2_x2) - torch.max(b1_x1, b2_x1)).clamp(0) * \ - (torch.min(b1_y2, b2_y2) - torch.max(b1_y1, b2_y1)).clamp(0) - - # Union Area - w1, h1 = b1_x2 - b1_x1, b1_y2 - b1_y1 + eps - w2, h2 = b2_x2 - b2_x1, b2_y2 - b2_y1 + eps - union = w1 * h1 + w2 * h2 - inter + eps - - iou = inter / union - if GIoU or DIoU or CIoU: - cw = torch.max(b1_x2, b2_x2) - torch.min(b1_x1, b2_x1) # convex (smallest enclosing box) width - ch = torch.max(b1_y2, b2_y2) - torch.min(b1_y1, b2_y1) # convex height - if CIoU or DIoU: # Distance or Complete IoU https://arxiv.org/abs/1911.08287v1 - c2 = cw ** 2 + ch ** 2 + eps # convex diagonal squared - rho2 = ((b2_x1 + b2_x2 - b1_x1 - b1_x2) ** 2 + - (b2_y1 + b2_y2 - b1_y1 - b1_y2) ** 2) / 4 # center distance squared - if DIoU: - return iou - rho2 / c2 # DIoU - elif CIoU: # https://github.com/Zzh-tju/DIoU-SSD-pytorch/blob/master/utils/box/box_utils.py#L47 - v = (4 / math.pi ** 2) * torch.pow(torch.atan(w2 / h2) - torch.atan(w1 / h1), 2) - with torch.no_grad(): - alpha = v / (v - iou + (1 + eps)) - return iou - (rho2 / c2 + v * alpha) # CIoU - else: # GIoU https://arxiv.org/pdf/1902.09630.pdf - c_area = cw * ch + eps # convex area - return iou - (c_area - union) / c_area # GIoU - else: - return iou # IoU - - -def box_iou(box1, box2): - # https://github.com/pytorch/vision/blob/master/torchvision/ops/boxes.py - """ - Return intersection-over-union (Jaccard index) of boxes. - Both sets of boxes are expected to be in (x1, y1, x2, y2) format. - Arguments: - box1 (Tensor[N, 4]) - box2 (Tensor[M, 4]) - Returns: - iou (Tensor[N, M]): the NxM matrix containing the pairwise - IoU values for every element in boxes1 and boxes2 - """ - - def box_area(box): - # box = 4xn - return (box[2] - box[0]) * (box[3] - box[1]) - - area1 = box_area(box1.T) - area2 = box_area(box2.T) - - # inter(N,M) = (rb(N,M,2) - lt(N,M,2)).clamp(0).prod(2) - inter = (torch.min(box1[:, None, 2:], box2[:, 2:]) - torch.max(box1[:, None, :2], box2[:, :2])).clamp(0).prod(2) - return inter / (area1[:, None] + area2 - inter) # iou = inter / (area1 + area2 - inter) - - -def wh_iou(wh1, wh2): - # Returns the nxm IoU matrix. wh1 is nx2, wh2 is mx2 - wh1 = wh1[:, None] # [N,1,2] - wh2 = wh2[None] # [1,M,2] - inter = torch.min(wh1, wh2).prod(2) # [N,M] - return inter / (wh1.prod(2) + wh2.prod(2) - inter) # iou = inter / (area1 + area2 - inter) - - -def non_max_suppression(prediction, conf_thres=0.25, iou_thres=0.45, classes=None, agnostic=False, multi_label=False, - labels=()): - """Runs Non-Maximum Suppression (NMS) on inference results - - Returns: - list of detections, on (n,6) tensor per image [xyxy, conf, cls] - """ - - nc = prediction.shape[2] - 5 # number of classes - xc = prediction[..., 4] > conf_thres # candidates - - # Settings - min_wh, max_wh = 2, 4096 # (pixels) minimum and maximum box width and height - max_det = 300 # maximum number of detections per image - max_nms = 30000 # maximum number of boxes into torchvision.ops.nms() - time_limit = 10.0 # seconds to quit after - redundant = True # require redundant detections - multi_label &= nc > 1 # multiple labels per box (adds 0.5ms/img) - merge = False # use merge-NMS - - t = time.time() - output = [torch.zeros((0, 6), device=prediction.device)] * prediction.shape[0] - for xi, x in enumerate(prediction): # image index, image inference - # Apply constraints - # x[((x[..., 2:4] < min_wh) | (x[..., 2:4] > max_wh)).any(1), 4] = 0 # width-height - x = x[xc[xi]] # confidence - - # Cat apriori labels if autolabelling - if labels and len(labels[xi]): - l = labels[xi] - v = torch.zeros((len(l), nc + 5), device=x.device) - v[:, :4] = l[:, 1:5] # box - v[:, 4] = 1.0 # conf - v[range(len(l)), l[:, 0].long() + 5] = 1.0 # cls - x = torch.cat((x, v), 0) - - # If none remain process next image - if not x.shape[0]: - continue - - # Compute conf - x[:, 5:] *= x[:, 4:5] # conf = obj_conf * cls_conf - - # Box (center x, center y, width, height) to (x1, y1, x2, y2) - box = xywh2xyxy(x[:, :4]) - - # Detections matrix nx6 (xyxy, conf, cls) - if multi_label: - i, j = (x[:, 5:] > conf_thres).nonzero(as_tuple=False).T - x = torch.cat((box[i], x[i, j + 5, None], j[:, None].float()), 1) - else: # best class only - conf, j = x[:, 5:].max(1, keepdim=True) - x = torch.cat((box, conf, j.float()), 1)[conf.view(-1) > conf_thres] - - # Filter by class - if classes is not None: - x = x[(x[:, 5:6] == torch.tensor(classes, device=x.device)).any(1)] - - # Apply finite constraint - # if not torch.isfinite(x).all(): - # x = x[torch.isfinite(x).all(1)] - - # Check shape - n = x.shape[0] # number of boxes - if not n: # no boxes - continue - elif n > max_nms: # excess boxes - x = x[x[:, 4].argsort(descending=True)[:max_nms]] # sort by confidence - - # Batched NMS - c = x[:, 5:6] * (0 if agnostic else max_wh) # classes - boxes, scores = x[:, :4] + c, x[:, 4] # boxes (offset by class), scores - i = torchvision.ops.nms(boxes, scores, iou_thres) # NMS - if i.shape[0] > max_det: # limit detections - i = i[:max_det] - if merge and (1 < n < 3E3): # Merge NMS (boxes merged using weighted mean) - # update boxes as boxes(i,4) = weights(i,n) * boxes(n,4) - iou = box_iou(boxes[i], boxes) > iou_thres # iou matrix - weights = iou * scores[None] # box weights - x[i, :4] = torch.mm(weights, x[:, :4]).float() / weights.sum(1, keepdim=True) # merged boxes - if redundant: - i = i[iou.sum(1) > 1] # require redundancy - - output[xi] = x[i] - if (time.time() - t) > time_limit: - print(f'WARNING: NMS time limit {time_limit}s exceeded') - break # time limit exceeded - - return output - - -def strip_optimizer(f='best.pt', s=''): # from utils.general import *; strip_optimizer() - # Strip optimizer from 'f' to finalize training, optionally save as 's' - x = torch.load(f, map_location=torch.device('cpu')) - if x.get('ema'): - x['model'] = x['ema'] # replace model with ema - for k in 'optimizer', 'training_results', 'wandb_id', 'ema', 'updates': # keys - x[k] = None - x['epoch'] = -1 - x['model'].half() # to FP16 - for p in x['model'].parameters(): - p.requires_grad = False - torch.save(x, s or f) - mb = os.path.getsize(s or f) / 1E6 # filesize - print(f"Optimizer stripped from {f},{(' saved as %s,' % s) if s else ''} {mb:.1f}MB") - - -def print_mutation(hyp, results, yaml_file='hyp_evolved.yaml', bucket=''): - # Print mutation results to evolve.txt (for use with train.py --evolve) - a = '%10s' * len(hyp) % tuple(hyp.keys()) # hyperparam keys - b = '%10.3g' * len(hyp) % tuple(hyp.values()) # hyperparam values - c = '%10.4g' * len(results) % results # results (P, R, mAP@0.5, mAP@0.5:0.95, val_losses x 3) - print('\n%s\n%s\nEvolved fitness: %s\n' % (a, b, c)) - - if bucket: - url = 'gs://%s/evolve.txt' % bucket - if gsutil_getsize(url) > (os.path.getsize('evolve.txt') if os.path.exists('evolve.txt') else 0): - os.system('gsutil cp %s .' % url) # download evolve.txt if larger than local - - with open('evolve.txt', 'a') as f: # append result - f.write(c + b + '\n') - x = np.unique(np.loadtxt('evolve.txt', ndmin=2), axis=0) # load unique rows - x = x[np.argsort(-fitness(x))] # sort - np.savetxt('evolve.txt', x, '%10.3g') # save sort by fitness - - # Save yaml - for i, k in enumerate(hyp.keys()): - hyp[k] = float(x[0, i + 7]) - with open(yaml_file, 'w') as f: - results = tuple(x[0, :7]) - c = '%10.4g' * len(results) % results # results (P, R, mAP@0.5, mAP@0.5:0.95, val_losses x 3) - f.write('# Hyperparameter Evolution Results\n# Generations: %g\n# Metrics: ' % len(x) + c + '\n\n') - yaml.dump(hyp, f, sort_keys=False) - - if bucket: - os.system('gsutil cp evolve.txt %s gs://%s' % (yaml_file, bucket)) # upload - - -def apply_classifier(x, model, img, im0): - # applies a second stage classifier to yolo outputs - im0 = [im0] if isinstance(im0, np.ndarray) else im0 - for i, d in enumerate(x): # per image - if d is not None and len(d): - d = d.clone() - - # Reshape and pad cutouts - b = xyxy2xywh(d[:, :4]) # boxes - b[:, 2:] = b[:, 2:].max(1)[0].unsqueeze(1) # rectangle to square - b[:, 2:] = b[:, 2:] * 1.3 + 30 # pad - d[:, :4] = xywh2xyxy(b).long() - - # Rescale boxes from img_size to im0 size - scale_coords(img.shape[2:], d[:, :4], im0[i].shape) - - # Classes - pred_cls1 = d[:, 5].long() - ims = [] - for j, a in enumerate(d): # per item - cutout = im0[i][int(a[1]):int(a[3]), int(a[0]):int(a[2])] - im = cv2.resize(cutout, (224, 224)) # BGR - # cv2.imwrite('test%i.jpg' % j, cutout) - - im = im[:, :, ::-1].transpose(2, 0, 1) # BGR to RGB, to 3x416x416 - im = np.ascontiguousarray(im, dtype=np.float32) # uint8 to float32 - im /= 255.0 # 0 - 255 to 0.0 - 1.0 - ims.append(im) - - pred_cls2 = model(torch.Tensor(ims).to(d.device)).argmax(1) # classifier prediction - x[i] = x[i][pred_cls1 == pred_cls2] # retain matching class detections - - return x - - -def increment_path(path, exist_ok=True, sep=''): - # Increment path, i.e. runs/exp --> runs/exp{sep}0, runs/exp{sep}1 etc. - path = Path(path) # os-agnostic - if (path.exists() and exist_ok) or (not path.exists()): - return str(path) - else: - dirs = glob.glob(f"{path}{sep}*") # similar paths - matches = [re.search(rf"%s{sep}(\d+)" % path.stem, d) for d in dirs] - i = [int(m.groups()[0]) for m in matches if m] # indices - n = max(i) + 1 if i else 2 # increment number - return f"{path}{sep}{n}" # update path diff --git a/AIlib/utils/get_offline_url.py b/AIlib/utils/get_offline_url.py deleted file mode 100644 index 934e367..0000000 --- a/AIlib/utils/get_offline_url.py +++ /dev/null @@ -1,106 +0,0 @@ -from PIL import Image -import numpy as np -import cv2 -import base64 -import io,os -import requests -import time,json -import string,random -import glob -##for CeKanYuan -def get_offlineUrls(taskUrl,offlineFile,jsonfile='SendLog/platformQueryOfftask.json'): - with open(offlineFile,'r') as fp: - lines=fp.readlines() - doneCodes=[line.strip().split(' ')[2] for line in lines] - try: - res = requests.get(taskUrl,timeout=10).json() - offlines=res['data'] ##offlines[0]['code'],offlines[0]['videoUrl'] - with open(jsonfile,'w') as fp: - json.dump(res,fp, ensure_ascii=False) - except Exception as ee: - timestr=time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) - print('###line43 %s read taskUrl:%s error:%s '%(timestr,taskUrl,ee)) - offlines=[] - outOfflines=[] - for off in offlines: - off['port']=1935; - off.update({'name':'off-' +off.pop("code")}) - if off['name'] in doneCodes: - continue - off.update({'url': off.pop("videoUrl")}) - outOfflines.append(off) - #off['url']=off['videoUrl'] - - return outOfflines -def platurlToJsonfile(taskUrl,jsonfile='SendLog/platformQuery.json'): - try: - res = requests.get(taskUrl,timeout=10).json() - offlines=res['data'] ##offlines[0]['code'],offlines[0]['videoUrl'] - with open(jsonfile,'w') as fp: - json.dump(res,fp, ensure_ascii=False) - except Exception as ee: - timestr=time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) - print('###line43 %s read taskUrl:%s error:%s '%(timestr,taskUrl,ee)) - -def get_websource_fromTxt(txtfile): - with open(txtfile,'r') as fp: - lines = fp.readlines() - sources=[ ] - for line in lines: - sous={} - try: - sps = line.strip().split(' ') - sous['url']=sps[0];sous['port']=sps[1] - - #webs.append(sps[0]) - if 'rtmp' in sps[0]: - name = sps[0].split('/')[4] - - else: - name = sps[0][-3:] - sous['name']='live-'+name.replace('_','') - sous['port']=sps[1] - sources.append(sous) - except Exception as ee: - - print('####format error : %s, line:%s , in file:%s#####'%(ee,line,txtfile)) - assert len(sources)>0 - return sources - -def update_websource_offAndLive(platform_query_url,sourceFile,offlineFile,jsonfile='SendLog/platformQuery.json'): - - #platform_query_url='http://47.96.182.154:9051/api/suanfa/getPlatformInfo' - txtSource=get_websource_fromTxt(sourceFile) - try: - res = requests.get(platform_query_url,timeout=10).json() - questionUrl = res['data']['questionUrl'] ###直播流时，问题图片的推送地址 - offlineUrl = res['data']['offlineUrl'] ###http离线视频时，问题图片的推送地址 - taskUrl = res['data']['taskUrl'] ###http离线视频时，离线视频存放的地址 - with open(jsonfile,'w') as fp: - json.dump(res,fp, ensure_ascii=False) - except Exception as ee: - timestr=time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) - print('######line83: %s: file:geturlPlatform: error %s ,url:%s #####'%(timestr,ee,platform_query_url)) - taskUrl='http://121.40.249.52:9050/api/analyse/getAiInspectionList' - if taskUrl: - offlines=get_offlineUrls(taskUrl,offlineFile) - txtSource.extend(offlines) - #[{'url': 'rtmp://demoplay.yunhengzhizao.cn/live/THSA_HD5M', 'port': '1935', 'name': 'live-THSAHD5M'}] - outlist=[] - for sourss in txtSource : - source_url = sourss['url'] - vid = cv2.VideoCapture(source_url) - if vid.isOpened(): - outlist.append( sourss ) - return outlist - #print('##line65:',txtSource) - - - -if __name__=='__main__': - platform_query_url='http://47.96.182.154:9051/api/suanfa/getPlatformInfo' - sourceFile='../config/source.txt' - offlineFile='../mintors/offlines/doneCodes.txt' - jsonfile='../SendLog/platformQuery.json' - txtSource=update_websource_offAndLive(platform_query_url,sourceFile,offlineFile,jsonfile=jsonfile) - print(txtSource) diff --git a/AIlib/utils/google_app_engine/Dockerfile b/AIlib/utils/google_app_engine/Dockerfile deleted file mode 100644 index 0155618..0000000 --- a/AIlib/utils/google_app_engine/Dockerfile +++ /dev/null @@ -1,25 +0,0 @@ -FROM gcr.io/google-appengine/python - -# Create a virtualenv for dependencies. This isolates these packages from -# system-level packages. -# Use -p python3 or -p python3.7 to select python version. Default is version 2. -RUN virtualenv /env -p python3 - -# Setting these environment variables are the same as running -# source /env/bin/activate. -ENV VIRTUAL_ENV /env -ENV PATH /env/bin:$PATH - -RUN apt-get update && apt-get install -y python-opencv - -# Copy the application's requirements.txt and run pip to install all -# dependencies into the virtualenv. -ADD requirements.txt /app/requirements.txt -RUN pip install -r /app/requirements.txt - -# Add the application source code. -ADD . /app - -# Run a WSGI server to serve the application. gunicorn must be declared as -# a dependency in requirements.txt. -CMD gunicorn -b :$PORT main:app diff --git a/AIlib/utils/google_app_engine/additional_requirements.txt b/AIlib/utils/google_app_engine/additional_requirements.txt deleted file mode 100644 index 5fcc305..0000000 --- a/AIlib/utils/google_app_engine/additional_requirements.txt +++ /dev/null @@ -1,4 +0,0 @@ -# add these requirements in your app on top of the existing ones -pip==18.1 -Flask==1.0.2 -gunicorn==19.9.0 diff --git a/AIlib/utils/google_app_engine/app.yaml b/AIlib/utils/google_app_engine/app.yaml deleted file mode 100644 index ac29d10..0000000 --- a/AIlib/utils/google_app_engine/app.yaml +++ /dev/null @@ -1,14 +0,0 @@ -runtime: custom -env: flex - -service: yolov5app - -liveness_check: - initial_delay_sec: 600 - -manual_scaling: - instances: 1 -resources: - cpu: 1 - memory_gb: 4 - disk_size_gb: 20 \ No newline at end of file diff --git a/AIlib/utils/google_utils.py b/AIlib/utils/google_utils.py deleted file mode 100644 index 0a7ca3b..0000000 --- a/AIlib/utils/google_utils.py +++ /dev/null @@ -1,122 +0,0 @@ -# Google utils: https://cloud.google.com/storage/docs/reference/libraries - -import os -import platform -import subprocess -import time -from pathlib import Path - -import requests -import torch - - -def gsutil_getsize(url=''): - # gs://bucket/file size https://cloud.google.com/storage/docs/gsutil/commands/du - s = subprocess.check_output(f'gsutil du {url}', shell=True).decode('utf-8') - return eval(s.split(' ')[0]) if len(s) else 0 # bytes - - -def attempt_download(file, repo='ultralytics/yolov5'): - # Attempt file download if does not exist - file = Path(str(file).strip().replace("'", '').lower()) - - if not file.exists(): - try: - response = requests.get(f'https://api.github.com/repos/{repo}/releases/latest').json() # github api - assets = [x['name'] for x in response['assets']] # release assets, i.e. ['yolov5s.pt', 'yolov5m.pt', ...] - tag = response['tag_name'] # i.e. 'v1.0' - except: # fallback plan - assets = ['yolov5s.pt', 'yolov5m.pt', 'yolov5l.pt', 'yolov5x.pt'] - tag = subprocess.check_output('git tag', shell=True).decode().split()[-1] - - name = file.name - if name in assets: - msg = f'{file} missing, try downloading from https://github.com/{repo}/releases/' - redundant = False # second download option - try: # GitHub - url = f'https://github.com/{repo}/releases/download/{tag}/{name}' - print(f'Downloading {url} to {file}...') - torch.hub.download_url_to_file(url, file) - assert file.exists() and file.stat().st_size > 1E6 # check - except Exception as e: # GCP - print(f'Download error: {e}') - assert redundant, 'No secondary mirror' - url = f'https://storage.googleapis.com/{repo}/ckpt/{name}' - print(f'Downloading {url} to {file}...') - os.system(f'curl -L {url} -o {file}') # torch.hub.download_url_to_file(url, weights) - finally: - if not file.exists() or file.stat().st_size < 1E6: # check - file.unlink(missing_ok=True) # remove partial downloads - print(f'ERROR: Download failure: {msg}') - print('') - return - - -def gdrive_download(id='16TiPfZj7htmTyhntwcZyEEAejOUxuT6m', file='tmp.zip'): - # Downloads a file from Google Drive. from yolov5.utils.google_utils import *; gdrive_download() - t = time.time() - file = Path(file) - cookie = Path('cookie') # gdrive cookie - print(f'Downloading https://drive.google.com/uc?export=download&id={id} as {file}... ', end='') - file.unlink(missing_ok=True) # remove existing file - cookie.unlink(missing_ok=True) # remove existing cookie - - # Attempt file download - out = "NUL" if platform.system() == "Windows" else "/dev/null" - os.system(f'curl -c ./cookie -s -L "drive.google.com/uc?export=download&id={id}" > {out}') - if os.path.exists('cookie'): # large file - s = f'curl -Lb ./cookie "drive.google.com/uc?export=download&confirm={get_token()}&id={id}" -o {file}' - else: # small file - s = f'curl -s -L -o {file} "drive.google.com/uc?export=download&id={id}"' - r = os.system(s) # execute, capture return - cookie.unlink(missing_ok=True) # remove existing cookie - - # Error check - if r != 0: - file.unlink(missing_ok=True) # remove partial - print('Download error ') # raise Exception('Download error') - return r - - # Unzip if archive - if file.suffix == '.zip': - print('unzipping... ', end='') - os.system(f'unzip -q {file}') # unzip - file.unlink() # remove zip to free space - - print(f'Done ({time.time() - t:.1f}s)') - return r - - -def get_token(cookie="./cookie"): - with open(cookie) as f: - for line in f: - if "download" in line: - return line.split()[-1] - return "" - -# def upload_blob(bucket_name, source_file_name, destination_blob_name): -# # Uploads a file to a bucket -# # https://cloud.google.com/storage/docs/uploading-objects#storage-upload-object-python -# -# storage_client = storage.Client() -# bucket = storage_client.get_bucket(bucket_name) -# blob = bucket.blob(destination_blob_name) -# -# blob.upload_from_filename(source_file_name) -# -# print('File {} uploaded to {}.'.format( -# source_file_name, -# destination_blob_name)) -# -# -# def download_blob(bucket_name, source_blob_name, destination_file_name): -# # Uploads a blob from a bucket -# storage_client = storage.Client() -# bucket = storage_client.get_bucket(bucket_name) -# blob = bucket.blob(source_blob_name) -# -# blob.download_to_filename(destination_file_name) -# -# print('Blob {} downloaded to {}.'.format( -# source_blob_name, -# destination_file_name)) diff --git a/AIlib/utils/loss.py b/AIlib/utils/loss.py deleted file mode 100644 index 9e78df1..0000000 --- a/AIlib/utils/loss.py +++ /dev/null @@ -1,216 +0,0 @@ -# Loss functions - -import torch -import torch.nn as nn - -from utils.general import bbox_iou -from utils.torch_utils import is_parallel - - -def smooth_BCE(eps=0.1): # https://github.com/ultralytics/yolov3/issues/238#issuecomment-598028441 - # return positive, negative label smoothing BCE targets - return 1.0 - 0.5 * eps, 0.5 * eps - - -class BCEBlurWithLogitsLoss(nn.Module): - # BCEwithLogitLoss() with reduced missing label effects. - def __init__(self, alpha=0.05): - super(BCEBlurWithLogitsLoss, self).__init__() - self.loss_fcn = nn.BCEWithLogitsLoss(reduction='none') # must be nn.BCEWithLogitsLoss() - self.alpha = alpha - - def forward(self, pred, true): - loss = self.loss_fcn(pred, true) - pred = torch.sigmoid(pred) # prob from logits - dx = pred - true # reduce only missing label effects - # dx = (pred - true).abs() # reduce missing label and false label effects - alpha_factor = 1 - torch.exp((dx - 1) / (self.alpha + 1e-4)) - loss *= alpha_factor - return loss.mean() - - -class FocalLoss(nn.Module): - # Wraps focal loss around existing loss_fcn(), i.e. criteria = FocalLoss(nn.BCEWithLogitsLoss(), gamma=1.5) - def __init__(self, loss_fcn, gamma=1.5, alpha=0.25): - super(FocalLoss, self).__init__() - self.loss_fcn = loss_fcn # must be nn.BCEWithLogitsLoss() - self.gamma = gamma - self.alpha = alpha - self.reduction = loss_fcn.reduction - self.loss_fcn.reduction = 'none' # required to apply FL to each element - - def forward(self, pred, true): - loss = self.loss_fcn(pred, true) - # p_t = torch.exp(-loss) - # loss *= self.alpha * (1.000001 - p_t) ** self.gamma # non-zero power for gradient stability - - # TF implementation https://github.com/tensorflow/addons/blob/v0.7.1/tensorflow_addons/losses/focal_loss.py - pred_prob = torch.sigmoid(pred) # prob from logits - p_t = true * pred_prob + (1 - true) * (1 - pred_prob) - alpha_factor = true * self.alpha + (1 - true) * (1 - self.alpha) - modulating_factor = (1.0 - p_t) ** self.gamma - loss *= alpha_factor * modulating_factor - - if self.reduction == 'mean': - return loss.mean() - elif self.reduction == 'sum': - return loss.sum() - else: # 'none' - return loss - - -class QFocalLoss(nn.Module): - # Wraps Quality focal loss around existing loss_fcn(), i.e. criteria = FocalLoss(nn.BCEWithLogitsLoss(), gamma=1.5) - def __init__(self, loss_fcn, gamma=1.5, alpha=0.25): - super(QFocalLoss, self).__init__() - self.loss_fcn = loss_fcn # must be nn.BCEWithLogitsLoss() - self.gamma = gamma - self.alpha = alpha - self.reduction = loss_fcn.reduction - self.loss_fcn.reduction = 'none' # required to apply FL to each element - - def forward(self, pred, true): - loss = self.loss_fcn(pred, true) - - pred_prob = torch.sigmoid(pred) # prob from logits - alpha_factor = true * self.alpha + (1 - true) * (1 - self.alpha) - modulating_factor = torch.abs(true - pred_prob) ** self.gamma - loss *= alpha_factor * modulating_factor - - if self.reduction == 'mean': - return loss.mean() - elif self.reduction == 'sum': - return loss.sum() - else: # 'none' - return loss - - -class ComputeLoss: - # Compute losses - def __init__(self, model, autobalance=False): - super(ComputeLoss, self).__init__() - device = next(model.parameters()).device # get model device - h = model.hyp # hyperparameters - - # Define criteria - BCEcls = nn.BCEWithLogitsLoss(pos_weight=torch.tensor([h['cls_pw']], device=device)) - BCEobj = nn.BCEWithLogitsLoss(pos_weight=torch.tensor([h['obj_pw']], device=device)) - - # Class label smoothing https://arxiv.org/pdf/1902.04103.pdf eqn 3 - self.cp, self.cn = smooth_BCE(eps=h.get('label_smoothing', 0.0)) # positive, negative BCE targets - - # Focal loss - g = h['fl_gamma'] # focal loss gamma - if g > 0: - BCEcls, BCEobj = FocalLoss(BCEcls, g), FocalLoss(BCEobj, g) - - det = model.module.model[-1] if is_parallel(model) else model.model[-1] # Detect() module - self.balance = {3: [4.0, 1.0, 0.4]}.get(det.nl, [4.0, 1.0, 0.25, 0.06, .02]) # P3-P7 - self.ssi = list(det.stride).index(16) if autobalance else 0 # stride 16 index - self.BCEcls, self.BCEobj, self.gr, self.hyp, self.autobalance = BCEcls, BCEobj, model.gr, h, autobalance - for k in 'na', 'nc', 'nl', 'anchors': - setattr(self, k, getattr(det, k)) - - def __call__(self, p, targets): # predictions, targets, model - device = targets.device - lcls, lbox, lobj = torch.zeros(1, device=device), torch.zeros(1, device=device), torch.zeros(1, device=device) - tcls, tbox, indices, anchors = self.build_targets(p, targets) # targets - - # Losses - for i, pi in enumerate(p): # layer index, layer predictions - b, a, gj, gi = indices[i] # image, anchor, gridy, gridx - tobj = torch.zeros_like(pi[..., 0], device=device) # target obj - - n = b.shape[0] # number of targets - if n: - ps = pi[b, a, gj, gi] # prediction subset corresponding to targets - - # Regression - pxy = ps[:, :2].sigmoid() * 2. - 0.5 - pwh = (ps[:, 2:4].sigmoid() * 2) ** 2 * anchors[i] - pbox = torch.cat((pxy, pwh), 1) # predicted box - iou = bbox_iou(pbox.T, tbox[i], x1y1x2y2=False, CIoU=True) # iou(prediction, target) - lbox += (1.0 - iou).mean() # iou loss - - # Objectness - tobj[b, a, gj, gi] = (1.0 - self.gr) + self.gr * iou.detach().clamp(0).type(tobj.dtype) # iou ratio - - # Classification - if self.nc > 1: # cls loss (only if multiple classes) - t = torch.full_like(ps[:, 5:], self.cn, device=device) # targets - t[range(n), tcls[i]] = self.cp - lcls += self.BCEcls(ps[:, 5:], t) # BCE - - # Append targets to text file - # with open('targets.txt', 'a') as file: - # [file.write('%11.5g ' * 4 % tuple(x) + '\n') for x in torch.cat((txy[i], twh[i]), 1)] - - obji = self.BCEobj(pi[..., 4], tobj) - lobj += obji * self.balance[i] # obj loss - if self.autobalance: - self.balance[i] = self.balance[i] * 0.9999 + 0.0001 / obji.detach().item() - - if self.autobalance: - self.balance = [x / self.balance[self.ssi] for x in self.balance] - lbox *= self.hyp['box'] - lobj *= self.hyp['obj'] - lcls *= self.hyp['cls'] - bs = tobj.shape[0] # batch size - - loss = lbox + lobj + lcls - return loss * bs, torch.cat((lbox, lobj, lcls, loss)).detach() - - def build_targets(self, p, targets): - # Build targets for compute_loss(), input targets(image,class,x,y,w,h) - na, nt = self.na, targets.shape[0] # number of anchors, targets - tcls, tbox, indices, anch = [], [], [], [] - gain = torch.ones(7, device=targets.device) # normalized to gridspace gain - ai = torch.arange(na, device=targets.device).float().view(na, 1).repeat(1, nt) # same as .repeat_interleave(nt) - targets = torch.cat((targets.repeat(na, 1, 1), ai[:, :, None]), 2) # append anchor indices - - g = 0.5 # bias - off = torch.tensor([[0, 0], - [1, 0], [0, 1], [-1, 0], [0, -1], # j,k,l,m - # [1, 1], [1, -1], [-1, 1], [-1, -1], # jk,jm,lk,lm - ], device=targets.device).float() * g # offsets - - for i in range(self.nl): - anchors = self.anchors[i] - gain[2:6] = torch.tensor(p[i].shape)[[3, 2, 3, 2]] # xyxy gain - - # Match targets to anchors - t = targets * gain - if nt: - # Matches - r = t[:, :, 4:6] / anchors[:, None] # wh ratio - j = torch.max(r, 1. / r).max(2)[0] < self.hyp['anchor_t'] # compare - # j = wh_iou(anchors, t[:, 4:6]) > model.hyp['iou_t'] # iou(3,n)=wh_iou(anchors(3,2), gwh(n,2)) - t = t[j] # filter - - # Offsets - gxy = t[:, 2:4] # grid xy - gxi = gain[[2, 3]] - gxy # inverse - j, k = ((gxy % 1. < g) & (gxy > 1.)).T - l, m = ((gxi % 1. < g) & (gxi > 1.)).T - j = torch.stack((torch.ones_like(j), j, k, l, m)) - t = t.repeat((5, 1, 1))[j] - offsets = (torch.zeros_like(gxy)[None] + off[:, None])[j] - else: - t = targets[0] - offsets = 0 - - # Define - b, c = t[:, :2].long().T # image, class - gxy = t[:, 2:4] # grid xy - gwh = t[:, 4:6] # grid wh - gij = (gxy - offsets).long() - gi, gj = gij.T # grid xy indices - - # Append - a = t[:, 6].long() # anchor indices - indices.append((b, a, gj.clamp_(0, gain[3] - 1), gi.clamp_(0, gain[2] - 1))) # image, anchor, grid indices - tbox.append(torch.cat((gxy - gij, gwh), 1)) # box - anch.append(anchors[a]) # anchors - tcls.append(c) # class - - return tcls, tbox, indices, anch diff --git a/AIlib/utils/metrics.py b/AIlib/utils/metrics.py deleted file mode 100644 index 666b8c7..0000000 --- a/AIlib/utils/metrics.py +++ /dev/null @@ -1,223 +0,0 @@ -# Model validation metrics - -from pathlib import Path - -import matplotlib.pyplot as plt -import numpy as np -import torch - -from . import general - - -def fitness(x): - # Model fitness as a weighted combination of metrics - w = [0.0, 0.0, 0.1, 0.9] # weights for [P, R, mAP@0.5, mAP@0.5:0.95] - return (x[:, :4] * w).sum(1) - - -def ap_per_class(tp, conf, pred_cls, target_cls, plot=False, save_dir='.', names=()): - """ Compute the average precision, given the recall and precision curves. - Source: https://github.com/rafaelpadilla/Object-Detection-Metrics. - # Arguments - tp: True positives (nparray, nx1 or nx10). - conf: Objectness value from 0-1 (nparray). - pred_cls: Predicted object classes (nparray). - target_cls: True object classes (nparray). - plot: Plot precision-recall curve at mAP@0.5 - save_dir: Plot save directory - # Returns - The average precision as computed in py-faster-rcnn. - """ - - # Sort by objectness - i = np.argsort(-conf) - tp, conf, pred_cls = tp[i], conf[i], pred_cls[i] - - # Find unique classes - unique_classes = np.unique(target_cls) - nc = unique_classes.shape[0] # number of classes, number of detections - - # Create Precision-Recall curve and compute AP for each class - px, py = np.linspace(0, 1, 1000), [] # for plotting - ap, p, r = np.zeros((nc, tp.shape[1])), np.zeros((nc, 1000)), np.zeros((nc, 1000)) - for ci, c in enumerate(unique_classes): - i = pred_cls == c - n_l = (target_cls == c).sum() # number of labels - n_p = i.sum() # number of predictions - - if n_p == 0 or n_l == 0: - continue - else: - # Accumulate FPs and TPs - fpc = (1 - tp[i]).cumsum(0) - tpc = tp[i].cumsum(0) - - # Recall - recall = tpc / (n_l + 1e-16) # recall curve - r[ci] = np.interp(-px, -conf[i], recall[:, 0], left=0) # negative x, xp because xp decreases - - # Precision - precision = tpc / (tpc + fpc) # precision curve - p[ci] = np.interp(-px, -conf[i], precision[:, 0], left=1) # p at pr_score - - # AP from recall-precision curve - for j in range(tp.shape[1]): - ap[ci, j], mpre, mrec = compute_ap(recall[:, j], precision[:, j]) - if plot and j == 0: - py.append(np.interp(px, mrec, mpre)) # precision at mAP@0.5 - - # Compute F1 (harmonic mean of precision and recall) - f1 = 2 * p * r / (p + r + 1e-16) - if plot: - plot_pr_curve(px, py, ap, Path(save_dir) / 'PR_curve.png', names) - plot_mc_curve(px, f1, Path(save_dir) / 'F1_curve.png', names, ylabel='F1') - plot_mc_curve(px, p, Path(save_dir) / 'P_curve.png', names, ylabel='Precision') - plot_mc_curve(px, r, Path(save_dir) / 'R_curve.png', names, ylabel='Recall') - - i = f1.mean(0).argmax() # max F1 index - return p[:, i], r[:, i], ap, f1[:, i], unique_classes.astype('int32') - - -def compute_ap(recall, precision): - """ Compute the average precision, given the recall and precision curves - # Arguments - recall: The recall curve (list) - precision: The precision curve (list) - # Returns - Average precision, precision curve, recall curve - """ - - # Append sentinel values to beginning and end - mrec = np.concatenate(([0.], recall, [recall[-1] + 0.01])) - mpre = np.concatenate(([1.], precision, [0.])) - - # Compute the precision envelope - mpre = np.flip(np.maximum.accumulate(np.flip(mpre))) - - # Integrate area under curve - method = 'interp' # methods: 'continuous', 'interp' - if method == 'interp': - x = np.linspace(0, 1, 101) # 101-point interp (COCO) - ap = np.trapz(np.interp(x, mrec, mpre), x) # integrate - else: # 'continuous' - i = np.where(mrec[1:] != mrec[:-1])[0] # points where x axis (recall) changes - ap = np.sum((mrec[i + 1] - mrec[i]) * mpre[i + 1]) # area under curve - - return ap, mpre, mrec - - -class ConfusionMatrix: - # Updated version of https://github.com/kaanakan/object_detection_confusion_matrix - def __init__(self, nc, conf=0.25, iou_thres=0.45): - self.matrix = np.zeros((nc + 1, nc + 1)) - self.nc = nc # number of classes - self.conf = conf - self.iou_thres = iou_thres - - def process_batch(self, detections, labels): - """ - Return intersection-over-union (Jaccard index) of boxes. - Both sets of boxes are expected to be in (x1, y1, x2, y2) format. - Arguments: - detections (Array[N, 6]), x1, y1, x2, y2, conf, class - labels (Array[M, 5]), class, x1, y1, x2, y2 - Returns: - None, updates confusion matrix accordingly - """ - detections = detections[detections[:, 4] > self.conf] - gt_classes = labels[:, 0].int() - detection_classes = detections[:, 5].int() - iou = general.box_iou(labels[:, 1:], detections[:, :4]) - - x = torch.where(iou > self.iou_thres) - if x[0].shape[0]: - matches = torch.cat((torch.stack(x, 1), iou[x[0], x[1]][:, None]), 1).cpu().numpy() - if x[0].shape[0] > 1: - matches = matches[matches[:, 2].argsort()[::-1]] - matches = matches[np.unique(matches[:, 1], return_index=True)[1]] - matches = matches[matches[:, 2].argsort()[::-1]] - matches = matches[np.unique(matches[:, 0], return_index=True)[1]] - else: - matches = np.zeros((0, 3)) - - n = matches.shape[0] > 0 - m0, m1, _ = matches.transpose().astype(np.int16) - for i, gc in enumerate(gt_classes): - j = m0 == i - if n and sum(j) == 1: - self.matrix[gc, detection_classes[m1[j]]] += 1 # correct - else: - self.matrix[self.nc, gc] += 1 # background FP - - if n: - for i, dc in enumerate(detection_classes): - if not any(m1 == i): - self.matrix[dc, self.nc] += 1 # background FN - - def matrix(self): - return self.matrix - - def plot(self, save_dir='', names=()): - try: - import seaborn as sn - - array = self.matrix / (self.matrix.sum(0).reshape(1, self.nc + 1) + 1E-6) # normalize - array[array < 0.005] = np.nan # don't annotate (would appear as 0.00) - - fig = plt.figure(figsize=(12, 9), tight_layout=True) - sn.set(font_scale=1.0 if self.nc < 50 else 0.8) # for label size - labels = (0 < len(names) < 99) and len(names) == self.nc # apply names to ticklabels - sn.heatmap(array, annot=self.nc < 30, annot_kws={"size": 8}, cmap='Blues', fmt='.2f', square=True, - xticklabels=names + ['background FP'] if labels else "auto", - yticklabels=names + ['background FN'] if labels else "auto").set_facecolor((1, 1, 1)) - fig.axes[0].set_xlabel('True') - fig.axes[0].set_ylabel('Predicted') - fig.savefig(Path(save_dir) / 'confusion_matrix.png', dpi=250) - except Exception as e: - pass - - def print(self): - for i in range(self.nc + 1): - print(' '.join(map(str, self.matrix[i]))) - - -# Plots ---------------------------------------------------------------------------------------------------------------- - -def plot_pr_curve(px, py, ap, save_dir='pr_curve.png', names=()): - # Precision-recall curve - fig, ax = plt.subplots(1, 1, figsize=(9, 6), tight_layout=True) - py = np.stack(py, axis=1) - - if 0 < len(names) < 21: # display per-class legend if < 21 classes - for i, y in enumerate(py.T): - ax.plot(px, y, linewidth=1, label=f'{names[i]} {ap[i, 0]:.3f}') # plot(recall, precision) - else: - ax.plot(px, py, linewidth=1, color='grey') # plot(recall, precision) - - ax.plot(px, py.mean(1), linewidth=3, color='blue', label='all classes %.3f mAP@0.5' % ap[:, 0].mean()) - ax.set_xlabel('Recall') - ax.set_ylabel('Precision') - ax.set_xlim(0, 1) - ax.set_ylim(0, 1) - plt.legend(bbox_to_anchor=(1.04, 1), loc="upper left") - fig.savefig(Path(save_dir), dpi=250) - - -def plot_mc_curve(px, py, save_dir='mc_curve.png', names=(), xlabel='Confidence', ylabel='Metric'): - # Metric-confidence curve - fig, ax = plt.subplots(1, 1, figsize=(9, 6), tight_layout=True) - - if 0 < len(names) < 21: # display per-class legend if < 21 classes - for i, y in enumerate(py): - ax.plot(px, y, linewidth=1, label=f'{names[i]}') # plot(confidence, metric) - else: - ax.plot(px, py.T, linewidth=1, color='grey') # plot(confidence, metric) - - y = py.mean(0) - ax.plot(px, y, linewidth=3, color='blue', label=f'all classes {y.max():.2f} at {px[y.argmax()]:.3f}') - ax.set_xlabel(xlabel) - ax.set_ylabel(ylabel) - ax.set_xlim(0, 1) - ax.set_ylim(0, 1) - plt.legend(bbox_to_anchor=(1.04, 1), loc="upper left") - fig.savefig(Path(save_dir), dpi=250) diff --git a/AIlib/utils/platech.ttf b/AIlib/utils/platech.ttf deleted file mode 100644 index d66a970..0000000 Binary files a/AIlib/utils/platech.ttf and /dev/null differ diff --git a/AIlib/utils/plots.py b/AIlib/utils/plots.py deleted file mode 100644 index f16dac2..0000000 --- a/AIlib/utils/plots.py +++ /dev/null @@ -1,575 +0,0 @@ -# Plotting utils - -import glob -import math -import os -import random -from copy import copy -from pathlib import Path - -import cv2 -import matplotlib -import matplotlib.pyplot as plt -import numpy as np -import pandas as pd -import seaborn as sns -import torch -import yaml -from PIL import Image, ImageDraw, ImageFont -from scipy.signal import butter, filtfilt,savgol_filter - -from utils.general import xywh2xyxy, xyxy2xywh -from utils.metrics import fitness - -# Settings -matplotlib.rc('font', **{'size': 11}) -#matplotlib.use('Agg') # for writing to files only - -def smooth_outline(contours,p1,p2): - arcontours=np.array(contours) - coors_x=arcontours[0,:,0,0] - coors_y=arcontours[0,:,0,1] - coors_x_smooth= savgol_filter(coors_x,p1,p2) - coors_y_smooth= savgol_filter(coors_y,p1,p2) - arcontours[0,:,0,0] = coors_x_smooth - arcontours[0,:,0,1] = coors_y_smooth - return arcontours -def smooth_outline_auto(contours): - cnt = len(contours[0]) - p1 = int(cnt/12)*2+1 - p2 =3 - if p10 - return webs,ports,streamNames - - -def get_label_array( color=None, label=None,outfontsize=None,fontpath="conf/platech.ttf"): - - # Plots one bounding box on image 'im' using PIL - fontsize = 48 - font = ImageFont.truetype(fontpath, fontsize,encoding='utf-8') - - txt_width, txt_height = font.getsize(label) - im = np.zeros((txt_height,txt_width,3),dtype=np.uint8) - im = Image.fromarray(im) - draw = ImageDraw.Draw(im) - draw.rectangle([0, 0 , txt_width, txt_height ], fill=tuple(color)) - draw.text(( 0 , -3 ), label, fill=(255, 255, 255), font=font) - im_array = np.asarray(im) - - if outfontsize: - scaley = outfontsize / txt_height - im_array= cv2.resize(im_array,(0,0),fx = scaley ,fy =scaley) - return im_array -def get_label_arrays(labelnames,colors,outfontsize=40,fontpath="conf/platech.ttf"): - label_arraylist = [] - if len(labelnames) > len(colors): - print('#####labelnames cnt > colors cnt#####') - for ii,labelname in enumerate(labelnames): - - color = colors[ii%20] - label_arraylist.append(get_label_array(color=color,label=labelname,outfontsize=40,fontpath=fontpath)) - - return label_arraylist -def color_list(): - # Return first 10 plt colors as (r,g,b) https://stackoverflow.com/questions/51350872/python-from-color-name-to-rgb - def hex2rgb(h): - return tuple(int(h[1 + i:1 + i + 2], 16) for i in (0, 2, 4)) - - return [hex2rgb(h) for h in matplotlib.colors.TABLEAU_COLORS.values()] # or BASE_ (8), CSS4_ (148), XKCD_ (949) - - -def hist2d(x, y, n=100): - # 2d histogram used in labels.png and evolve.png - xedges, yedges = np.linspace(x.min(), x.max(), n), np.linspace(y.min(), y.max(), n) - hist, xedges, yedges = np.histogram2d(x, y, (xedges, yedges)) - xidx = np.clip(np.digitize(x, xedges) - 1, 0, hist.shape[0] - 1) - yidx = np.clip(np.digitize(y, yedges) - 1, 0, hist.shape[1] - 1) - return np.log(hist[xidx, yidx]) - - -def butter_lowpass_filtfilt(data, cutoff=1500, fs=50000, order=5): - # https://stackoverflow.com/questions/28536191/how-to-filter-smooth-with-scipy-numpy - def butter_lowpass(cutoff, fs, order): - nyq = 0.5 * fs - normal_cutoff = cutoff / nyq - return butter(order, normal_cutoff, btype='low', analog=False) - - b, a = butter_lowpass(cutoff, fs, order=order) - return filtfilt(b, a, data) # forward-backward filter - - - '''image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB) - pil_image = Image.fromarray(image) - draw = ImageDraw.Draw(pil_image) - font = ImageFont.truetype('./font/platech.ttf', 40, encoding='utf-8') - for info in infos: - detect = info['bndbox'] - text = ','.join(list(info['attributes'].values())) - temp = -50 - if info['name'] == 'vehicle': - temp = 20 - draw.text((detect[0], detect[1] + temp), text, (0, 255, 255), font=font) - if 'scores' in info: - draw.text((detect[0], detect[3]), info['scores'], (0, 255, 0), font=font) - if 'pscore' in info: - draw.text((detect[2], detect[3]), str(round(info['pscore'],3)), (0, 255, 0), font=font) - image = cv2.cvtColor(np.array(pil_image), cv2.COLOR_RGB2BGR) - for info in infos: - detect = info['bndbox'] - cv2.rectangle(image, (detect[0], detect[1]), (detect[2], detect[3]), (0, 255, 0), 1, cv2.LINE_AA) - return image''' - -'''def plot_one_box_PIL(x, im, color=None, label=None, line_thickness=3): - # Plots one bounding box on image 'im' using OpenCV - assert im.data.contiguous, 'Image not contiguous. Apply np.ascontiguousarray(im) to plot_on_box() input image.' - tl = line_thickness or round(0.002 * (im.shape[0] + im.shape[1]) / 2) + 1 # line/font thickness - color = color or [random.randint(0, 255) for _ in range(3)] - c1, c2 = (int(x[0]), int(x[1])), (int(x[2]), int(x[3])) - - - - cv2.rectangle(im, c1, c2, color, thickness=tl, lineType=cv2.LINE_AA) - - - if label: - tf = max(tl - 1, 1) # font thickness - t_size = cv2.getTextSize(label, 0, fontScale=tl / 3, thickness=tf)[0] - c2 = c1[0] + t_size[0], c1[1] - t_size[1] - 3 - cv2.rectangle(im, c1, c2, color, -1, cv2.LINE_AA) # filled - - im = Image.fromarray(im) - draw = ImageDraw.Draw(im) - font = ImageFont.truetype('./font/platech.ttf', t_size, encoding='utf-8') - draw.text((c1[0], c1[1] - 2), label, (0, 255, 0), font=font) - - #cv2.putText(im, label, (c1[0], c1[1] - 2), 0, tl / 3, [225, 255, 255], thickness=tf, lineType=cv2.LINE_AA) - return np.array(im) ''' - -def plot_one_box(x, im, color=None, label=None, line_thickness=3): - # Plots one bounding box on image 'im' using OpenCV - assert im.data.contiguous, 'Image not contiguous. Apply np.ascontiguousarray(im) to plot_on_box() input image.' - tl = line_thickness or round(0.002 * (im.shape[0] + im.shape[1]) / 2) + 1 # line/font thickness - color = color or [random.randint(0, 255) for _ in range(3)] - c1, c2 = (int(x[0]), int(x[1])), (int(x[2]), int(x[3])) - cv2.rectangle(im, c1, c2, color, thickness=tl, lineType=cv2.LINE_AA) - - if label: - tf = max(tl - 1, 1) # font thickness - t_size = cv2.getTextSize(label, 0, fontScale=tl / 3, thickness=tf)[0] - c2 = c1[0] + t_size[0], c1[1] - t_size[1] - 3 - cv2.rectangle(im, c1, c2, color, -1, cv2.LINE_AA) # filled - cv2.putText(im, label, (c1[0], c1[1] - 2), 0, tl / 3, [225, 255, 255], thickness=tf, lineType=cv2.LINE_AA) - - -def plot_one_box_PIL(box, im, color=None, label=None, line_thickness=None): - # Plots one bounding box on image 'im' using PIL - print('##line149:',box) - im = Image.fromarray(im) - draw = ImageDraw.Draw(im) - line_thickness = line_thickness or max(int(min(im.size) / 200), 2) - draw.rectangle(box, width=line_thickness, outline=tuple(color)) # plot - - if label: - fontsize = max(round(max(im.size) / 40), 12) - font = ImageFont.truetype("platech.ttf", fontsize,encoding='utf-8') - txt_width, txt_height = font.getsize(label) - draw.rectangle([box[0], box[1] - txt_height + 4, box[0] + txt_width, box[1]], fill=tuple(color)) - draw.text((box[0], box[1] - txt_height + 1), label, fill=(255, 255, 255), font=font) - im_array = np.asarray(im) - - return np.asarray(im) - -def draw_painting_joint(box,img,label_array,score=0.5,color=None,line_thickness=None): - box = [int(xx.cpu()) for xx in box] - lh, lw, lc = label_array.shape - imh, imw, imc = img.shape - x0 , y1 = box[0:2] - x1 , y0 = x0 + lw , y1 - lh - y0 = max(0,y0) ; y1 = y0 + lh - x1 = min(imw, x1); x0 = x1 - lw - img[y0:y1,x0:x1,:] = label_array - tl = line_thickness or round(0.002 * (img.shape[0] + img.shape[1]) / 2) + 1 # line/font thickness - - c1, c2 = (int(box[0]), int(box[1])), (int(box[2]), int(box[3])) - cv2.rectangle(img, c1, c2, color, thickness=tl, lineType=cv2.LINE_AA) - label = ' %.2f'%(score) - tf = max(tl - 1, 1) # font thickness - fontScale = tl * 0.33 - t_size = cv2.getTextSize(label, 0, fontScale=fontScale , thickness=tf)[0] - c2 = c1[0]+ lw + t_size[0], c1[1] - lh - cv2.rectangle(img, (int(box[0])+lw,int(box[1])) , c2, color, -1, cv2.LINE_AA) # filled - cv2.putText(img, label, (c1[0]+lw, c1[1] - (lh-t_size[1])//2 ), 0, fontScale, [225, 255, 255], thickness=tf, lineType=cv2.LINE_AA) - - return img - -def plot_wh_methods(): # from utils.plots import *; plot_wh_methods() - # Compares the two methods for width-height anchor multiplication - # https://github.com/ultralytics/yolov3/issues/168 - x = np.arange(-4.0, 4.0, .1) - ya = np.exp(x) - yb = torch.sigmoid(torch.from_numpy(x)).numpy() * 2 - - fig = plt.figure(figsize=(6, 3), tight_layout=True) - plt.plot(x, ya, '.-', label='YOLOv3') - plt.plot(x, yb ** 2, '.-', label='YOLOv5 ^2') - plt.plot(x, yb ** 1.6, '.-', label='YOLOv5 ^1.6') - plt.xlim(left=-4, right=4) - plt.ylim(bottom=0, top=6) - plt.xlabel('input') - plt.ylabel('output') - plt.grid() - plt.legend() - fig.savefig('comparison.png', dpi=200) - - -def output_to_target(output): - # Convert model output to target format [batch_id, class_id, x, y, w, h, conf] - targets = [] - for i, o in enumerate(output): - for *box, conf, cls in o.cpu().numpy(): - targets.append([i, cls, *list(*xyxy2xywh(np.array(box)[None])), conf]) - return np.array(targets) - - -def plot_images(images, targets, paths=None, fname='images.jpg', names=None, max_size=640, max_subplots=16): - # Plot image grid with labels - - if isinstance(images, torch.Tensor): - images = images.cpu().float().numpy() - if isinstance(targets, torch.Tensor): - targets = targets.cpu().numpy() - - # un-normalise - if np.max(images[0]) <= 1: - images *= 255 - - tl = 3 # line thickness - tf = max(tl - 1, 1) # font thickness - bs, _, h, w = images.shape # batch size, _, height, width - bs = min(bs, max_subplots) # limit plot images - ns = np.ceil(bs ** 0.5) # number of subplots (square) - - # Check if we should resize - scale_factor = max_size / max(h, w) - if scale_factor < 1: - h = math.ceil(scale_factor * h) - w = math.ceil(scale_factor * w) - - colors = color_list() # list of colors - mosaic = np.full((int(ns * h), int(ns * w), 3), 255, dtype=np.uint8) # init - for i, img in enumerate(images): - if i == max_subplots: # if last batch has fewer images than we expect - break - - block_x = int(w * (i // ns)) - block_y = int(h * (i % ns)) - - img = img.transpose(1, 2, 0) - if scale_factor < 1: - img = cv2.resize(img, (w, h)) - - mosaic[block_y:block_y + h, block_x:block_x + w, :] = img - if len(targets) > 0: - image_targets = targets[targets[:, 0] == i] - boxes = xywh2xyxy(image_targets[:, 2:6]).T - classes = image_targets[:, 1].astype('int') - labels = image_targets.shape[1] == 6 # labels if no conf column - conf = None if labels else image_targets[:, 6] # check for confidence presence (label vs pred) - - if boxes.shape[1]: - if boxes.max() <= 1.01: # if normalized with tolerance 0.01 - boxes[[0, 2]] *= w # scale to pixels - boxes[[1, 3]] *= h - elif scale_factor < 1: # absolute coords need scale if image scales - boxes *= scale_factor - boxes[[0, 2]] += block_x - boxes[[1, 3]] += block_y - for j, box in enumerate(boxes.T): - cls = int(classes[j]) - color = colors[cls % len(colors)] - cls = names[cls] if names else cls - if labels or conf[j] > 0.25: # 0.25 conf thresh - label = '%s' % cls if labels else '%s %.1f' % (cls, conf[j]) - plot_one_box(box, mosaic, label=label, color=color, line_thickness=tl) - - # Draw image filename labels - if paths: - label = Path(paths[i]).name[:40] # trim to 40 char - t_size = cv2.getTextSize(label, 0, fontScale=tl / 3, thickness=tf)[0] - cv2.putText(mosaic, label, (block_x + 5, block_y + t_size[1] + 5), 0, tl / 3, [220, 220, 220], thickness=tf, - lineType=cv2.LINE_AA) - - # Image border - cv2.rectangle(mosaic, (block_x, block_y), (block_x + w, block_y + h), (255, 255, 255), thickness=3) - - if fname: - r = min(1280. / max(h, w) / ns, 1.0) # ratio to limit image size - mosaic = cv2.resize(mosaic, (int(ns * w * r), int(ns * h * r)), interpolation=cv2.INTER_AREA) - # cv2.imwrite(fname, cv2.cvtColor(mosaic, cv2.COLOR_BGR2RGB)) # cv2 save - Image.fromarray(mosaic).save(fname) # PIL save - return mosaic - - -def plot_lr_scheduler(optimizer, scheduler, epochs=300, save_dir=''): - # Plot LR simulating training for full epochs - optimizer, scheduler = copy(optimizer), copy(scheduler) # do not modify originals - y = [] - for _ in range(epochs): - scheduler.step() - y.append(optimizer.param_groups[0]['lr']) - plt.plot(y, '.-', label='LR') - plt.xlabel('epoch') - plt.ylabel('LR') - plt.grid() - plt.xlim(0, epochs) - plt.ylim(0) - plt.savefig(Path(save_dir) / 'LR.png', dpi=200) - plt.close() - - -def plot_test_txt(): # from utils.plots import *; plot_test() - # Plot test.txt histograms - x = np.loadtxt('test.txt', dtype=np.float32) - box = xyxy2xywh(x[:, :4]) - cx, cy = box[:, 0], box[:, 1] - - fig, ax = plt.subplots(1, 1, figsize=(6, 6), tight_layout=True) - ax.hist2d(cx, cy, bins=600, cmax=10, cmin=0) - ax.set_aspect('equal') - plt.savefig('hist2d.png', dpi=300) - - fig, ax = plt.subplots(1, 2, figsize=(12, 6), tight_layout=True) - ax[0].hist(cx, bins=600) - ax[1].hist(cy, bins=600) - plt.savefig('hist1d.png', dpi=200) - - -def plot_targets_txt(): # from utils.plots import *; plot_targets_txt() - # Plot targets.txt histograms - x = np.loadtxt('targets.txt', dtype=np.float32).T - s = ['x targets', 'y targets', 'width targets', 'height targets'] - fig, ax = plt.subplots(2, 2, figsize=(8, 8), tight_layout=True) - ax = ax.ravel() - for i in range(4): - ax[i].hist(x[i], bins=100, label='%.3g +/- %.3g' % (x[i].mean(), x[i].std())) - ax[i].legend() - ax[i].set_title(s[i]) - plt.savefig('targets.jpg', dpi=200) - - -def plot_study_txt(path='', x=None): # from utils.plots import *; plot_study_txt() - # Plot study.txt generated by test.py - fig, ax = plt.subplots(2, 4, figsize=(10, 6), tight_layout=True) - # ax = ax.ravel() - - fig2, ax2 = plt.subplots(1, 1, figsize=(8, 4), tight_layout=True) - # for f in [Path(path) / f'study_coco_{x}.txt' for x in ['yolov5s6', 'yolov5m6', 'yolov5l6', 'yolov5x6']]: - for f in sorted(Path(path).glob('study*.txt')): - y = np.loadtxt(f, dtype=np.float32, usecols=[0, 1, 2, 3, 7, 8, 9], ndmin=2).T - x = np.arange(y.shape[1]) if x is None else np.array(x) - s = ['P', 'R', 'mAP@.5', 'mAP@.5:.95', 't_inference (ms/img)', 't_NMS (ms/img)', 't_total (ms/img)'] - # for i in range(7): - # ax[i].plot(x, y[i], '.-', linewidth=2, markersize=8) - # ax[i].set_title(s[i]) - - j = y[3].argmax() + 1 - ax2.plot(y[6, 1:j], y[3, 1:j] * 1E2, '.-', linewidth=2, markersize=8, - label=f.stem.replace('study_coco_', '').replace('yolo', 'YOLO')) - - ax2.plot(1E3 / np.array([209, 140, 97, 58, 35, 18]), [34.6, 40.5, 43.0, 47.5, 49.7, 51.5], - 'k.-', linewidth=2, markersize=8, alpha=.25, label='EfficientDet') - - ax2.grid(alpha=0.2) - ax2.set_yticks(np.arange(20, 60, 5)) - ax2.set_xlim(0, 57) - ax2.set_ylim(30, 55) - ax2.set_xlabel('GPU Speed (ms/img)') - ax2.set_ylabel('COCO AP val') - ax2.legend(loc='lower right') - plt.savefig(str(Path(path).name) + '.png', dpi=300) - - -def plot_labels(labels, names=(), save_dir=Path(''), loggers=None): - # plot dataset labels - print('Plotting labels... ') - c, b = labels[:, 0], labels[:, 1:].transpose() # classes, boxes - nc = int(c.max() + 1) # number of classes - colors = color_list() - x = pd.DataFrame(b.transpose(), columns=['x', 'y', 'width', 'height']) - - # seaborn correlogram - sns.pairplot(x, corner=True, diag_kind='auto', kind='hist', diag_kws=dict(bins=50), plot_kws=dict(pmax=0.9)) - plt.savefig(save_dir / 'labels_correlogram.jpg', dpi=200) - plt.close() - - # matplotlib labels - matplotlib.use('svg') # faster - ax = plt.subplots(2, 2, figsize=(8, 8), tight_layout=True)[1].ravel() - ax[0].hist(c, bins=np.linspace(0, nc, nc + 1) - 0.5, rwidth=0.8) - ax[0].set_ylabel('instances') - if 0 < len(names) < 30: - ax[0].set_xticks(range(len(names))) - ax[0].set_xticklabels(names, rotation=90, fontsize=10) - else: - ax[0].set_xlabel('classes') - sns.histplot(x, x='x', y='y', ax=ax[2], bins=50, pmax=0.9) - sns.histplot(x, x='width', y='height', ax=ax[3], bins=50, pmax=0.9) - - # rectangles - labels[:, 1:3] = 0.5 # center - labels[:, 1:] = xywh2xyxy(labels[:, 1:]) * 2000 - img = Image.fromarray(np.ones((2000, 2000, 3), dtype=np.uint8) * 255) - for cls, *box in labels[:1000]: - ImageDraw.Draw(img).rectangle(box, width=1, outline=colors[int(cls) % 10]) # plot - ax[1].imshow(img) - ax[1].axis('off') - - for a in [0, 1, 2, 3]: - for s in ['top', 'right', 'left', 'bottom']: - ax[a].spines[s].set_visible(False) - - plt.savefig(save_dir / 'labels.jpg', dpi=200) - matplotlib.use('Agg') - plt.close() - - # loggers - for k, v in loggers.items() or {}: - if k == 'wandb' and v: - v.log({"Labels": [v.Image(str(x), caption=x.name) for x in save_dir.glob('*labels*.jpg')]}, commit=False) - - -def plot_evolution(yaml_file='data/hyp.finetune.yaml'): # from utils.plots import *; plot_evolution() - # Plot hyperparameter evolution results in evolve.txt - with open(yaml_file) as f: - hyp = yaml.load(f, Loader=yaml.SafeLoader) - x = np.loadtxt('evolve.txt', ndmin=2) - f = fitness(x) - # weights = (f - f.min()) ** 2 # for weighted results - plt.figure(figsize=(10, 12), tight_layout=True) - matplotlib.rc('font', **{'size': 8}) - for i, (k, v) in enumerate(hyp.items()): - y = x[:, i + 7] - # mu = (y * weights).sum() / weights.sum() # best weighted result - mu = y[f.argmax()] # best single result - plt.subplot(6, 5, i + 1) - plt.scatter(y, f, c=hist2d(y, f, 20), cmap='viridis', alpha=.8, edgecolors='none') - plt.plot(mu, f.max(), 'k+', markersize=15) - plt.title('%s = %.3g' % (k, mu), fontdict={'size': 9}) # limit to 40 characters - if i % 5 != 0: - plt.yticks([]) - print('%15s: %.3g' % (k, mu)) - plt.savefig('evolve.png', dpi=200) - print('\nPlot saved as evolve.png') - - -def profile_idetection(start=0, stop=0, labels=(), save_dir=''): - # Plot iDetection '*.txt' per-image logs. from utils.plots import *; profile_idetection() - ax = plt.subplots(2, 4, figsize=(12, 6), tight_layout=True)[1].ravel() - s = ['Images', 'Free Storage (GB)', 'RAM Usage (GB)', 'Battery', 'dt_raw (ms)', 'dt_smooth (ms)', 'real-world FPS'] - files = list(Path(save_dir).glob('frames*.txt')) - for fi, f in enumerate(files): - try: - results = np.loadtxt(f, ndmin=2).T[:, 90:-30] # clip first and last rows - n = results.shape[1] # number of rows - x = np.arange(start, min(stop, n) if stop else n) - results = results[:, x] - t = (results[0] - results[0].min()) # set t0=0s - results[0] = x - for i, a in enumerate(ax): - if i < len(results): - label = labels[fi] if len(labels) else f.stem.replace('frames_', '') - a.plot(t, results[i], marker='.', label=label, linewidth=1, markersize=5) - a.set_title(s[i]) - a.set_xlabel('time (s)') - # if fi == len(files) - 1: - # a.set_ylim(bottom=0) - for side in ['top', 'right']: - a.spines[side].set_visible(False) - else: - a.remove() - except Exception as e: - print('Warning: Plotting error for %s; %s' % (f, e)) - - ax[1].legend() - plt.savefig(Path(save_dir) / 'idetection_profile.png', dpi=200) - - -def plot_results_overlay(start=0, stop=0): # from utils.plots import *; plot_results_overlay() - # Plot training 'results*.txt', overlaying train and val losses - s = ['train', 'train', 'train', 'Precision', 'mAP@0.5', 'val', 'val', 'val', 'Recall', 'mAP@0.5:0.95'] # legends - t = ['Box', 'Objectness', 'Classification', 'P-R', 'mAP-F1'] # titles - for f in sorted(glob.glob('results*.txt') + glob.glob('../../Downloads/results*.txt')): - results = np.loadtxt(f, usecols=[2, 3, 4, 8, 9, 12, 13, 14, 10, 11], ndmin=2).T - n = results.shape[1] # number of rows - x = range(start, min(stop, n) if stop else n) - fig, ax = plt.subplots(1, 5, figsize=(14, 3.5), tight_layout=True) - ax = ax.ravel() - for i in range(5): - for j in [i, i + 5]: - y = results[j, x] - ax[i].plot(x, y, marker='.', label=s[j]) - # y_smooth = butter_lowpass_filtfilt(y) - # ax[i].plot(x, np.gradient(y_smooth), marker='.', label=s[j]) - - ax[i].set_title(t[i]) - ax[i].legend() - ax[i].set_ylabel(f) if i == 0 else None # add filename - fig.savefig(f.replace('.txt', '.png'), dpi=200) - - -def plot_results(start=0, stop=0, bucket='', id=(), labels=(), save_dir=''): - # Plot training 'results*.txt'. from utils.plots import *; plot_results(save_dir='runs/train/exp') - fig, ax = plt.subplots(2, 5, figsize=(12, 6), tight_layout=True) - ax = ax.ravel() - s = ['Box', 'Objectness', 'Classification', 'Precision', 'Recall', - 'val Box', 'val Objectness', 'val Classification', 'mAP@0.5', 'mAP@0.5:0.95'] - if bucket: - # files = ['https://storage.googleapis.com/%s/results%g.txt' % (bucket, x) for x in id] - files = ['results%g.txt' % x for x in id] - c = ('gsutil cp ' + '%s ' * len(files) + '.') % tuple('gs://%s/results%g.txt' % (bucket, x) for x in id) - os.system(c) - else: - files = list(Path(save_dir).glob('results*.txt')) - assert len(files), 'No results.txt files found in %s, nothing to plot.' % os.path.abspath(save_dir) - for fi, f in enumerate(files): - try: - results = np.loadtxt(f, usecols=[2, 3, 4, 8, 9, 12, 13, 14, 10, 11], ndmin=2).T - n = results.shape[1] # number of rows - x = range(start, min(stop, n) if stop else n) - for i in range(10): - y = results[i, x] - if i in [0, 1, 2, 5, 6, 7]: - y[y == 0] = np.nan # don't show zero loss values - # y /= y[0] # normalize - label = labels[fi] if len(labels) else f.stem - ax[i].plot(x, y, marker='.', label=label, linewidth=2, markersize=8) - ax[i].set_title(s[i]) - # if i in [5, 6, 7]: # share train and val loss y axes - # ax[i].get_shared_y_axes().join(ax[i], ax[i - 5]) - except Exception as e: - print('Warning: Plotting error for %s; %s' % (f, e)) - - ax[1].legend() - fig.savefig(Path(save_dir) / 'results.png', dpi=200) diff --git a/AIlib/utils/torch_utils.py b/AIlib/utils/torch_utils.py deleted file mode 100644 index 9991e5e..0000000 --- a/AIlib/utils/torch_utils.py +++ /dev/null @@ -1,303 +0,0 @@ -# YOLOv5 PyTorch utils - -import datetime -import logging -import math -import os -import platform -import subprocess -import time -from contextlib import contextmanager -from copy import deepcopy -from pathlib import Path - -import torch -import torch.backends.cudnn as cudnn -import torch.nn as nn -import torch.nn.functional as F -import torchvision - -try: - import thop # for FLOPS computation -except ImportError: - thop = None -logger = logging.getLogger(__name__) - - -@contextmanager -def torch_distributed_zero_first(local_rank: int): - """ - Decorator to make all processes in distributed training wait for each local_master to do something. - """ - if local_rank not in [-1, 0]: - torch.distributed.barrier() - yield - if local_rank == 0: - torch.distributed.barrier() - - -def init_torch_seeds(seed=0): - # Speed-reproducibility tradeoff https://pytorch.org/docs/stable/notes/randomness.html - torch.manual_seed(seed) - if seed == 0: # slower, more reproducible - cudnn.benchmark, cudnn.deterministic = False, True - else: # faster, less reproducible - cudnn.benchmark, cudnn.deterministic = True, False - - -def date_modified(path=__file__): - # return human-readable file modification date, i.e. '2021-3-26' - t = datetime.datetime.fromtimestamp(Path(path).stat().st_mtime) - return f'{t.year}-{t.month}-{t.day}' - - -def git_describe(path=Path(__file__).parent): # path must be a directory - # return human-readable git description, i.e. v5.0-5-g3e25f1e https://git-scm.com/docs/git-describe - s = f'git -C {path} describe --tags --long --always' - try: - return subprocess.check_output(s, shell=True, stderr=subprocess.STDOUT).decode()[:-1] - except subprocess.CalledProcessError as e: - return '' # not a git repository - - -def select_device(device='', batch_size=None): - # device = 'cpu' or '0' or '0,1,2,3' - s = f'YOLOv5 🚀 {git_describe() or date_modified()} torch {torch.__version__} ' # string - cpu = device.lower() == 'cpu' - if cpu: - os.environ['CUDA_VISIBLE_DEVICES'] = '-1' # force torch.cuda.is_available() = False - elif device: # non-cpu device requested - os.environ['CUDA_VISIBLE_DEVICES'] = device # set environment variable - assert torch.cuda.is_available(), f'CUDA unavailable, invalid device {device} requested' # check availability - - cuda = not cpu and torch.cuda.is_available() - if cuda: - n = torch.cuda.device_count() - if n > 1 and batch_size: # check that batch_size is compatible with device_count - assert batch_size % n == 0, f'batch-size {batch_size} not multiple of GPU count {n}' - space = ' ' * len(s) - for i, d in enumerate(device.split(',') if device else range(n)): - p = torch.cuda.get_device_properties(i) - s += f"{'' if i == 0 else space}CUDA:{d} ({p.name}, {p.total_memory / 1024 ** 2}MB)\n" # bytes to MB - else: - s += 'CPU\n' - - logger.info(s.encode().decode('ascii', 'ignore') if platform.system() == 'Windows' else s) # emoji-safe - return torch.device('cuda:0' if cuda else 'cpu') - - -def time_synchronized(): - # pytorch-accurate time - if torch.cuda.is_available(): - torch.cuda.synchronize() - return time.time() - - -def profile(x, ops, n=100, device=None): - # profile a pytorch module or list of modules. Example usage: - # x = torch.randn(16, 3, 640, 640) # input - # m1 = lambda x: x * torch.sigmoid(x) - # m2 = nn.SiLU() - # profile(x, [m1, m2], n=100) # profile speed over 100 iterations - - device = device or torch.device('cuda:0' if torch.cuda.is_available() else 'cpu') - x = x.to(device) - x.requires_grad = True - print(torch.__version__, device.type, torch.cuda.get_device_properties(0) if device.type == 'cuda' else '') - print(f"\n{'Params':>12s}{'GFLOPS':>12s}{'forward (ms)':>16s}{'backward (ms)':>16s}{'input':>24s}{'output':>24s}") - for m in ops if isinstance(ops, list) else [ops]: - m = m.to(device) if hasattr(m, 'to') else m # device - m = m.half() if hasattr(m, 'half') and isinstance(x, torch.Tensor) and x.dtype is torch.float16 else m # type - dtf, dtb, t = 0., 0., [0., 0., 0.] # dt forward, backward - try: - flops = thop.profile(m, inputs=(x,), verbose=False)[0] / 1E9 * 2 # GFLOPS - except: - flops = 0 - - for _ in range(n): - t[0] = time_synchronized() - y = m(x) - t[1] = time_synchronized() - try: - _ = y.sum().backward() - t[2] = time_synchronized() - except: # no backward method - t[2] = float('nan') - dtf += (t[1] - t[0]) * 1000 / n # ms per op forward - dtb += (t[2] - t[1]) * 1000 / n # ms per op backward - - s_in = tuple(x.shape) if isinstance(x, torch.Tensor) else 'list' - s_out = tuple(y.shape) if isinstance(y, torch.Tensor) else 'list' - p = sum(list(x.numel() for x in m.parameters())) if isinstance(m, nn.Module) else 0 # parameters - print(f'{p:12}{flops:12.4g}{dtf:16.4g}{dtb:16.4g}{str(s_in):>24s}{str(s_out):>24s}') - - -def is_parallel(model): - return type(model) in (nn.parallel.DataParallel, nn.parallel.DistributedDataParallel) - - -def intersect_dicts(da, db, exclude=()): - # Dictionary intersection of matching keys and shapes, omitting 'exclude' keys, using da values - return {k: v for k, v in da.items() if k in db and not any(x in k for x in exclude) and v.shape == db[k].shape} - - -def initialize_weights(model): - for m in model.modules(): - t = type(m) - if t is nn.Conv2d: - pass # nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu') - elif t is nn.BatchNorm2d: - m.eps = 1e-3 - m.momentum = 0.03 - elif t in [nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6]: - m.inplace = True - - -def find_modules(model, mclass=nn.Conv2d): - # Finds layer indices matching module class 'mclass' - return [i for i, m in enumerate(model.module_list) if isinstance(m, mclass)] - - -def sparsity(model): - # Return global model sparsity - a, b = 0., 0. - for p in model.parameters(): - a += p.numel() - b += (p == 0).sum() - return b / a - - -def prune(model, amount=0.3): - # Prune model to requested global sparsity - import torch.nn.utils.prune as prune - print('Pruning model... ', end='') - for name, m in model.named_modules(): - if isinstance(m, nn.Conv2d): - prune.l1_unstructured(m, name='weight', amount=amount) # prune - prune.remove(m, 'weight') # make permanent - print(' %.3g global sparsity' % sparsity(model)) - - -def fuse_conv_and_bn(conv, bn): - # Fuse convolution and batchnorm layers https://tehnokv.com/posts/fusing-batchnorm-and-conv/ - fusedconv = nn.Conv2d(conv.in_channels, - conv.out_channels, - kernel_size=conv.kernel_size, - stride=conv.stride, - padding=conv.padding, - groups=conv.groups, - bias=True).requires_grad_(False).to(conv.weight.device) - - # prepare filters - w_conv = conv.weight.clone().view(conv.out_channels, -1) - w_bn = torch.diag(bn.weight.div(torch.sqrt(bn.eps + bn.running_var))) - fusedconv.weight.copy_(torch.mm(w_bn, w_conv).view(fusedconv.weight.shape)) - - # prepare spatial bias - b_conv = torch.zeros(conv.weight.size(0), device=conv.weight.device) if conv.bias is None else conv.bias - b_bn = bn.bias - bn.weight.mul(bn.running_mean).div(torch.sqrt(bn.running_var + bn.eps)) - fusedconv.bias.copy_(torch.mm(w_bn, b_conv.reshape(-1, 1)).reshape(-1) + b_bn) - - return fusedconv - - -def model_info(model, verbose=False, img_size=640): - # Model information. img_size may be int or list, i.e. img_size=640 or img_size=[640, 320] - n_p = sum(x.numel() for x in model.parameters()) # number parameters - n_g = sum(x.numel() for x in model.parameters() if x.requires_grad) # number gradients - if verbose: - print('%5s %40s %9s %12s %20s %10s %10s' % ('layer', 'name', 'gradient', 'parameters', 'shape', 'mu', 'sigma')) - for i, (name, p) in enumerate(model.named_parameters()): - name = name.replace('module_list.', '') - print('%5g %40s %9s %12g %20s %10.3g %10.3g' % - (i, name, p.requires_grad, p.numel(), list(p.shape), p.mean(), p.std())) - - try: # FLOPS - from thop import profile - stride = max(int(model.stride.max()), 32) if hasattr(model, 'stride') else 32 - img = torch.zeros((1, model.yaml.get('ch', 3), stride, stride), device=next(model.parameters()).device) # input - flops = profile(deepcopy(model), inputs=(img,), verbose=False)[0] / 1E9 * 2 # stride GFLOPS - img_size = img_size if isinstance(img_size, list) else [img_size, img_size] # expand if int/float - fs = ', %.1f GFLOPS' % (flops * img_size[0] / stride * img_size[1] / stride) # 640x640 GFLOPS - except (ImportError, Exception): - fs = '' - - logger.info(f"Model Summary: {len(list(model.modules()))} layers, {n_p} parameters, {n_g} gradients{fs}") - - -def load_classifier(name='resnet101', n=2): - # Loads a pretrained model reshaped to n-class output - model = torchvision.models.__dict__[name](pretrained=True) - - # ResNet model properties - # input_size = [3, 224, 224] - # input_space = 'RGB' - # input_range = [0, 1] - # mean = [0.485, 0.456, 0.406] - # std = [0.229, 0.224, 0.225] - - # Reshape output to n classes - filters = model.fc.weight.shape[1] - model.fc.bias = nn.Parameter(torch.zeros(n), requires_grad=True) - model.fc.weight = nn.Parameter(torch.zeros(n, filters), requires_grad=True) - model.fc.out_features = n - return model - - -def scale_img(img, ratio=1.0, same_shape=False, gs=32): # img(16,3,256,416) - # scales img(bs,3,y,x) by ratio constrained to gs-multiple - if ratio == 1.0: - return img - else: - h, w = img.shape[2:] - s = (int(h * ratio), int(w * ratio)) # new size - img = F.interpolate(img, size=s, mode='bilinear', align_corners=False) # resize - if not same_shape: # pad/crop img - h, w = [math.ceil(x * ratio / gs) * gs for x in (h, w)] - return F.pad(img, [0, w - s[1], 0, h - s[0]], value=0.447) # value = imagenet mean - - -def copy_attr(a, b, include=(), exclude=()): - # Copy attributes from b to a, options to only include [...] and to exclude [...] - for k, v in b.__dict__.items(): - if (len(include) and k not in include) or k.startswith('_') or k in exclude: - continue - else: - setattr(a, k, v) - - -class ModelEMA: - """ Model Exponential Moving Average from https://github.com/rwightman/pytorch-image-models - Keep a moving average of everything in the model state_dict (parameters and buffers). - This is intended to allow functionality like - https://www.tensorflow.org/api_docs/python/tf/train/ExponentialMovingAverage - A smoothed version of the weights is necessary for some training schemes to perform well. - This class is sensitive where it is initialized in the sequence of model init, - GPU assignment and distributed training wrappers. - """ - - def __init__(self, model, decay=0.9999, updates=0): - # Create EMA - self.ema = deepcopy(model.module if is_parallel(model) else model).eval() # FP32 EMA - # if next(model.parameters()).device.type != 'cpu': - # self.ema.half() # FP16 EMA - self.updates = updates # number of EMA updates - self.decay = lambda x: decay * (1 - math.exp(-x / 2000)) # decay exponential ramp (to help early epochs) - for p in self.ema.parameters(): - p.requires_grad_(False) - - def update(self, model): - # Update EMA parameters - with torch.no_grad(): - self.updates += 1 - d = self.decay(self.updates) - - msd = model.module.state_dict() if is_parallel(model) else model.state_dict() # model state_dict - for k, v in self.ema.state_dict().items(): - if v.dtype.is_floating_point: - v *= d - v += (1. - d) * msd[k].detach() - - def update_attr(self, model, include=(), exclude=('process_group', 'reducer')): - # Update EMA attributes - copy_attr(self.ema, model, include, exclude) diff --git a/AIlib/utils/wandb_logging/__init__.py b/AIlib/utils/wandb_logging/__init__.py deleted file mode 100644 index e69de29..0000000 diff --git a/AIlib/utils/wandb_logging/__pycache__/__init__.cpython-38.pyc b/AIlib/utils/wandb_logging/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index 71491cf..0000000 Binary files a/AIlib/utils/wandb_logging/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/wandb_logging/__pycache__/wandb_utils.cpython-38.pyc b/AIlib/utils/wandb_logging/__pycache__/wandb_utils.cpython-38.pyc deleted file mode 100644 index 9d3e648..0000000 Binary files a/AIlib/utils/wandb_logging/__pycache__/wandb_utils.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utils/wandb_logging/log_dataset.py b/AIlib/utils/wandb_logging/log_dataset.py deleted file mode 100644 index d7a521f..0000000 --- a/AIlib/utils/wandb_logging/log_dataset.py +++ /dev/null @@ -1,24 +0,0 @@ -import argparse - -import yaml - -from wandb_utils import WandbLogger - -WANDB_ARTIFACT_PREFIX = 'wandb-artifact://' - - -def create_dataset_artifact(opt): - with open(opt.data) as f: - data = yaml.load(f, Loader=yaml.SafeLoader) # data dict - logger = WandbLogger(opt, '', None, data, job_type='Dataset Creation') - - -if __name__ == '__main__': - parser = argparse.ArgumentParser() - parser.add_argument('--data', type=str, default='data/coco128.yaml', help='data.yaml path') - parser.add_argument('--single-cls', action='store_true', help='train as single-class dataset') - parser.add_argument('--project', type=str, default='YOLOv5', help='name of W&B Project') - opt = parser.parse_args() - opt.resume = False # Explicitly disallow resume check for dataset upload job - - create_dataset_artifact(opt) diff --git a/AIlib/utils/wandb_logging/wandb_utils.py b/AIlib/utils/wandb_logging/wandb_utils.py deleted file mode 100644 index d8f50ae..0000000 --- a/AIlib/utils/wandb_logging/wandb_utils.py +++ /dev/null @@ -1,306 +0,0 @@ -import json -import sys -from pathlib import Path - -import torch -import yaml -from tqdm import tqdm - -sys.path.append(str(Path(__file__).parent.parent.parent)) # add utils/ to path -from utils.datasets import LoadImagesAndLabels -from utils.datasets import img2label_paths -from utils.general import colorstr, xywh2xyxy, check_dataset - -try: - import wandb - from wandb import init, finish -except ImportError: - wandb = None - -WANDB_ARTIFACT_PREFIX = 'wandb-artifact://' - - -def remove_prefix(from_string, prefix=WANDB_ARTIFACT_PREFIX): - return from_string[len(prefix):] - - -def check_wandb_config_file(data_config_file): - wandb_config = '_wandb.'.join(data_config_file.rsplit('.', 1)) # updated data.yaml path - if Path(wandb_config).is_file(): - return wandb_config - return data_config_file - - -def get_run_info(run_path): - run_path = Path(remove_prefix(run_path, WANDB_ARTIFACT_PREFIX)) - run_id = run_path.stem - project = run_path.parent.stem - model_artifact_name = 'run_' + run_id + '_model' - return run_id, project, model_artifact_name - - -def check_wandb_resume(opt): - process_wandb_config_ddp_mode(opt) if opt.global_rank not in [-1, 0] else None - if isinstance(opt.resume, str): - if opt.resume.startswith(WANDB_ARTIFACT_PREFIX): - if opt.global_rank not in [-1, 0]: # For resuming DDP runs - run_id, project, model_artifact_name = get_run_info(opt.resume) - api = wandb.Api() - artifact = api.artifact(project + '/' + model_artifact_name + ':latest') - modeldir = artifact.download() - opt.weights = str(Path(modeldir) / "last.pt") - return True - return None - - -def process_wandb_config_ddp_mode(opt): - with open(opt.data) as f: - data_dict = yaml.load(f, Loader=yaml.SafeLoader) # data dict - train_dir, val_dir = None, None - if isinstance(data_dict['train'], str) and data_dict['train'].startswith(WANDB_ARTIFACT_PREFIX): - api = wandb.Api() - train_artifact = api.artifact(remove_prefix(data_dict['train']) + ':' + opt.artifact_alias) - train_dir = train_artifact.download() - train_path = Path(train_dir) / 'data/images/' - data_dict['train'] = str(train_path) - - if isinstance(data_dict['val'], str) and data_dict['val'].startswith(WANDB_ARTIFACT_PREFIX): - api = wandb.Api() - val_artifact = api.artifact(remove_prefix(data_dict['val']) + ':' + opt.artifact_alias) - val_dir = val_artifact.download() - val_path = Path(val_dir) / 'data/images/' - data_dict['val'] = str(val_path) - if train_dir or val_dir: - ddp_data_path = str(Path(val_dir) / 'wandb_local_data.yaml') - with open(ddp_data_path, 'w') as f: - yaml.dump(data_dict, f) - opt.data = ddp_data_path - - -class WandbLogger(): - def __init__(self, opt, name, run_id, data_dict, job_type='Training'): - # Pre-training routine -- - self.job_type = job_type - self.wandb, self.wandb_run, self.data_dict = wandb, None if not wandb else wandb.run, data_dict - # It's more elegant to stick to 1 wandb.init call, but useful config data is overwritten in the WandbLogger's wandb.init call - if isinstance(opt.resume, str): # checks resume from artifact - if opt.resume.startswith(WANDB_ARTIFACT_PREFIX): - run_id, project, model_artifact_name = get_run_info(opt.resume) - model_artifact_name = WANDB_ARTIFACT_PREFIX + model_artifact_name - assert wandb, 'install wandb to resume wandb runs' - # Resume wandb-artifact:// runs here| workaround for not overwriting wandb.config - self.wandb_run = wandb.init(id=run_id, project=project, resume='allow') - opt.resume = model_artifact_name - elif self.wandb: - self.wandb_run = wandb.init(config=opt, - resume="allow", - project='YOLOv5' if opt.project == 'runs/train' else Path(opt.project).stem, - name=name, - job_type=job_type, - id=run_id) if not wandb.run else wandb.run - if self.wandb_run: - if self.job_type == 'Training': - if not opt.resume: - wandb_data_dict = self.check_and_upload_dataset(opt) if opt.upload_dataset else data_dict - # Info useful for resuming from artifacts - self.wandb_run.config.opt = vars(opt) - self.wandb_run.config.data_dict = wandb_data_dict - self.data_dict = self.setup_training(opt, data_dict) - if self.job_type == 'Dataset Creation': - self.data_dict = self.check_and_upload_dataset(opt) - else: - prefix = colorstr('wandb: ') - print(f"{prefix}Install Weights & Biases for YOLOv5 logging with 'pip install wandb' (recommended)") - - def check_and_upload_dataset(self, opt): - assert wandb, 'Install wandb to upload dataset' - check_dataset(self.data_dict) - config_path = self.log_dataset_artifact(opt.data, - opt.single_cls, - 'YOLOv5' if opt.project == 'runs/train' else Path(opt.project).stem) - print("Created dataset config file ", config_path) - with open(config_path) as f: - wandb_data_dict = yaml.load(f, Loader=yaml.SafeLoader) - return wandb_data_dict - - def setup_training(self, opt, data_dict): - self.log_dict, self.current_epoch, self.log_imgs = {}, 0, 16 # Logging Constants - self.bbox_interval = opt.bbox_interval - if isinstance(opt.resume, str): - modeldir, _ = self.download_model_artifact(opt) - if modeldir: - self.weights = Path(modeldir) / "last.pt" - config = self.wandb_run.config - opt.weights, opt.save_period, opt.batch_size, opt.bbox_interval, opt.epochs, opt.hyp = str( - self.weights), config.save_period, config.total_batch_size, config.bbox_interval, config.epochs, \ - config.opt['hyp'] - data_dict = dict(self.wandb_run.config.data_dict) # eliminates the need for config file to resume - if 'val_artifact' not in self.__dict__: # If --upload_dataset is set, use the existing artifact, don't download - self.train_artifact_path, self.train_artifact = self.download_dataset_artifact(data_dict.get('train'), - opt.artifact_alias) - self.val_artifact_path, self.val_artifact = self.download_dataset_artifact(data_dict.get('val'), - opt.artifact_alias) - self.result_artifact, self.result_table, self.val_table, self.weights = None, None, None, None - if self.train_artifact_path is not None: - train_path = Path(self.train_artifact_path) / 'data/images/' - data_dict['train'] = str(train_path) - if self.val_artifact_path is not None: - val_path = Path(self.val_artifact_path) / 'data/images/' - data_dict['val'] = str(val_path) - self.val_table = self.val_artifact.get("val") - self.map_val_table_path() - if self.val_artifact is not None: - self.result_artifact = wandb.Artifact("run_" + wandb.run.id + "_progress", "evaluation") - self.result_table = wandb.Table(["epoch", "id", "prediction", "avg_confidence"]) - if opt.bbox_interval == -1: - self.bbox_interval = opt.bbox_interval = (opt.epochs // 10) if opt.epochs > 10 else 1 - return data_dict - - def download_dataset_artifact(self, path, alias): - if isinstance(path, str) and path.startswith(WANDB_ARTIFACT_PREFIX): - dataset_artifact = wandb.use_artifact(remove_prefix(path, WANDB_ARTIFACT_PREFIX) + ":" + alias) - assert dataset_artifact is not None, "'Error: W&B dataset artifact doesn\'t exist'" - datadir = dataset_artifact.download() - return datadir, dataset_artifact - return None, None - - def download_model_artifact(self, opt): - if opt.resume.startswith(WANDB_ARTIFACT_PREFIX): - model_artifact = wandb.use_artifact(remove_prefix(opt.resume, WANDB_ARTIFACT_PREFIX) + ":latest") - assert model_artifact is not None, 'Error: W&B model artifact doesn\'t exist' - modeldir = model_artifact.download() - epochs_trained = model_artifact.metadata.get('epochs_trained') - total_epochs = model_artifact.metadata.get('total_epochs') - assert epochs_trained < total_epochs, 'training to %g epochs is finished, nothing to resume.' % ( - total_epochs) - return modeldir, model_artifact - return None, None - - def log_model(self, path, opt, epoch, fitness_score, best_model=False): - model_artifact = wandb.Artifact('run_' + wandb.run.id + '_model', type='model', metadata={ - 'original_url': str(path), - 'epochs_trained': epoch + 1, - 'save period': opt.save_period, - 'project': opt.project, - 'total_epochs': opt.epochs, - 'fitness_score': fitness_score - }) - model_artifact.add_file(str(path / 'last.pt'), name='last.pt') - wandb.log_artifact(model_artifact, - aliases=['latest', 'epoch ' + str(self.current_epoch), 'best' if best_model else '']) - print("Saving model artifact on epoch ", epoch + 1) - - def log_dataset_artifact(self, data_file, single_cls, project, overwrite_config=False): - with open(data_file) as f: - data = yaml.load(f, Loader=yaml.SafeLoader) # data dict - nc, names = (1, ['item']) if single_cls else (int(data['nc']), data['names']) - names = {k: v for k, v in enumerate(names)} # to index dictionary - self.train_artifact = self.create_dataset_table(LoadImagesAndLabels( - data['train']), names, name='train') if data.get('train') else None - self.val_artifact = self.create_dataset_table(LoadImagesAndLabels( - data['val']), names, name='val') if data.get('val') else None - if data.get('train'): - data['train'] = WANDB_ARTIFACT_PREFIX + str(Path(project) / 'train') - if data.get('val'): - data['val'] = WANDB_ARTIFACT_PREFIX + str(Path(project) / 'val') - path = data_file if overwrite_config else '_wandb.'.join(data_file.rsplit('.', 1)) # updated data.yaml path - data.pop('download', None) - with open(path, 'w') as f: - yaml.dump(data, f) - - if self.job_type == 'Training': # builds correct artifact pipeline graph - self.wandb_run.use_artifact(self.val_artifact) - self.wandb_run.use_artifact(self.train_artifact) - self.val_artifact.wait() - self.val_table = self.val_artifact.get('val') - self.map_val_table_path() - else: - self.wandb_run.log_artifact(self.train_artifact) - self.wandb_run.log_artifact(self.val_artifact) - return path - - def map_val_table_path(self): - self.val_table_map = {} - print("Mapping dataset") - for i, data in enumerate(tqdm(self.val_table.data)): - self.val_table_map[data[3]] = data[0] - - def create_dataset_table(self, dataset, class_to_id, name='dataset'): - # TODO: Explore multiprocessing to slpit this loop parallely| This is essential for speeding up the the logging - artifact = wandb.Artifact(name=name, type="dataset") - img_files = tqdm([dataset.path]) if isinstance(dataset.path, str) and Path(dataset.path).is_dir() else None - img_files = tqdm(dataset.img_files) if not img_files else img_files - for img_file in img_files: - if Path(img_file).is_dir(): - artifact.add_dir(img_file, name='data/images') - labels_path = 'labels'.join(dataset.path.rsplit('images', 1)) - artifact.add_dir(labels_path, name='data/labels') - else: - artifact.add_file(img_file, name='data/images/' + Path(img_file).name) - label_file = Path(img2label_paths([img_file])[0]) - artifact.add_file(str(label_file), - name='data/labels/' + label_file.name) if label_file.exists() else None - table = wandb.Table(columns=["id", "train_image", "Classes", "name"]) - class_set = wandb.Classes([{'id': id, 'name': name} for id, name in class_to_id.items()]) - for si, (img, labels, paths, shapes) in enumerate(tqdm(dataset)): - height, width = shapes[0] - labels[:, 2:] = (xywh2xyxy(labels[:, 2:].view(-1, 4))) * torch.Tensor([width, height, width, height]) - box_data, img_classes = [], {} - for cls, *xyxy in labels[:, 1:].tolist(): - cls = int(cls) - box_data.append({"position": {"minX": xyxy[0], "minY": xyxy[1], "maxX": xyxy[2], "maxY": xyxy[3]}, - "class_id": cls, - "box_caption": "%s" % (class_to_id[cls]), - "scores": {"acc": 1}, - "domain": "pixel"}) - img_classes[cls] = class_to_id[cls] - boxes = {"ground_truth": {"box_data": box_data, "class_labels": class_to_id}} # inference-space - table.add_data(si, wandb.Image(paths, classes=class_set, boxes=boxes), json.dumps(img_classes), - Path(paths).name) - artifact.add(table, name) - return artifact - - def log_training_progress(self, predn, path, names): - if self.val_table and self.result_table: - class_set = wandb.Classes([{'id': id, 'name': name} for id, name in names.items()]) - box_data = [] - total_conf = 0 - for *xyxy, conf, cls in predn.tolist(): - if conf >= 0.25: - box_data.append( - {"position": {"minX": xyxy[0], "minY": xyxy[1], "maxX": xyxy[2], "maxY": xyxy[3]}, - "class_id": int(cls), - "box_caption": "%s %.3f" % (names[cls], conf), - "scores": {"class_score": conf}, - "domain": "pixel"}) - total_conf = total_conf + conf - boxes = {"predictions": {"box_data": box_data, "class_labels": names}} # inference-space - id = self.val_table_map[Path(path).name] - self.result_table.add_data(self.current_epoch, - id, - wandb.Image(self.val_table.data[id][1], boxes=boxes, classes=class_set), - total_conf / max(1, len(box_data)) - ) - - def log(self, log_dict): - if self.wandb_run: - for key, value in log_dict.items(): - self.log_dict[key] = value - - def end_epoch(self, best_result=False): - if self.wandb_run: - wandb.log(self.log_dict) - self.log_dict = {} - if self.result_artifact: - train_results = wandb.JoinedTable(self.val_table, self.result_table, "id") - self.result_artifact.add(train_results, 'result') - wandb.log_artifact(self.result_artifact, aliases=['latest', 'epoch ' + str(self.current_epoch), - ('best' if best_result else '')]) - self.result_table = wandb.Table(["epoch", "id", "prediction", "avg_confidence"]) - self.result_artifact = wandb.Artifact("run_" + wandb.run.id + "_progress", "evaluation") - - def finish_run(self): - if self.wandb_run: - if self.log_dict: - wandb.log(self.log_dict) - wandb.run.finish() diff --git a/AIlib/utilsK/GPUtils.py b/AIlib/utilsK/GPUtils.py deleted file mode 100644 index bcbafff..0000000 --- a/AIlib/utilsK/GPUtils.py +++ /dev/null @@ -1,501 +0,0 @@ -#@@ -1,43 +1,43 @@ -# GPUtil - GPU utilization -# -# A Python module for programmically getting the GPU utilization from NVIDA GPUs using nvidia-smi -# -# Author: Anders Krogh Mortensen (anderskm) -# Date: 16 January 2017 -# Web: https://github.com/anderskm/gputil -# -# LICENSE -# -# MIT License -# -# Copyright (c) 2017 anderskm -# -# Permission is hereby granted, free of charge, to any person obtaining a copy -# of this software and associated documentation files (the "Software"), to deal -# in the Software without restriction, including without limitation the rights -# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell -# copies of the Software, and to permit persons to whom the Software is -# furnished to do so, subject to the following conditions: -# -# The above copyright notice and this permission notice shall be included in all -# copies or substantial portions of the Software. -# -# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR -# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, -# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE -# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER -# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, -# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE -# SOFTWARE. - -from subprocess import Popen, PIPE -from distutils import spawn -import os -import math -import random -import time -import sys -import platform -import subprocess -import numpy as np - - -__version__ = '1.4.0' -class GPU: - def __init__(self, ID, uuid, load, memoryTotal, memoryUsed, memoryFree, driver, gpu_name, serial, display_mode, display_active, temp_gpu): - self.id = ID - self.uuid = uuid - self.load = load - self.memoryUtil = float(memoryUsed)/float(memoryTotal) - self.memoryTotal = memoryTotal - self.memoryUsed = memoryUsed - self.memoryFree = memoryFree - self.driver = driver - self.name = gpu_name - self.serial = serial - self.display_mode = display_mode - self.display_active = display_active - self.temperature = temp_gpu - - def __str__(self): - return str(self.__dict__) - - -class GPUProcess: - def __init__(self, pid, processName, gpuId, gpuUuid, gpuName, usedMemory, - uid, uname): - self.pid = pid - self.processName = processName - self.gpuId = gpuId - self.gpuUuid = gpuUuid - self.gpuName = gpuName - self.usedMemory = usedMemory - self.uid = uid - self.uname = uname - - def __str__(self): - return str(self.__dict__) - -def safeFloatCast(strNumber): - try: - number = float(strNumber) - except ValueError: - number = float('nan') - return number - -#def getGPUs(): -def getNvidiaSmiCmd(): - if platform.system() == "Windows": - # If the platform is Windows and nvidia-smi - # could not be found from the environment path, - #@@ -75,57 +94,97 @@ def getGPUs(): - nvidia_smi = "%s\\Program Files\\NVIDIA Corporation\\NVSMI\\nvidia-smi.exe" % os.environ['systemdrive'] - else: - nvidia_smi = "nvidia-smi" - return nvidia_smi - - -def getGPUs(): - # Get ID, processing and memory utilization for all GPUs - nvidia_smi = getNvidiaSmiCmd() - try: - p = Popen([nvidia_smi,"--query-gpu=index,uuid,utilization.gpu,memory.total,memory.used,memory.free,driver_version,name,gpu_serial,display_active,display_mode,temperature.gpu", "--format=csv,noheader,nounits"], stdout=PIPE) - stdout, stderror = p.communicate() - p = subprocess.run([ - nvidia_smi, - "--query-gpu=index,uuid,utilization.gpu,memory.total,memory.used,memory.free,driver_version,name,gpu_serial,display_active,display_mode,temperature.gpu", - "--format=csv,noheader,nounits" - ], stdout=subprocess.PIPE, encoding='utf8') - stdout, stderror = p.stdout, p.stderr - except: - return [] - output = stdout;#output = stdout.decode('UTF-8') - # output = output[2:-1] # Remove b' and ' from string added by python - #print(output) - output = stdout - ## Parse output - # Split on line break - lines = output.split(os.linesep) - #print(lines) - numDevices = len(lines)-1 - GPUs = [] - for g in range(numDevices): - line = lines[g] - #print(line) - vals = line.split(', ') - #print(vals) - for i in range(12): - # print(vals[i]) - if (i == 0): - deviceIds = int(vals[i]) - elif (i == 1): - uuid = vals[i] - elif (i == 2): - gpuUtil = safeFloatCast(vals[i])/100 - elif (i == 3): - memTotal = safeFloatCast(vals[i]) - elif (i == 4): - memUsed = safeFloatCast(vals[i]) - elif (i == 5): - memFree = safeFloatCast(vals[i]) - elif (i == 6): - driver = vals[i] - elif (i == 7): - gpu_name = vals[i] - elif (i == 8): - serial = vals[i] - elif (i == 9): - display_active = vals[i] - elif (i == 10): - display_mode = vals[i] - elif (i == 11): - temp_gpu = safeFloatCast(vals[i]); - deviceIds = int(vals[0]) - uuid = vals[1] - gpuUtil = safeFloatCast(vals[2]) / 100 - memTotal = safeFloatCast(vals[3]) - memUsed = safeFloatCast(vals[4]) - memFree = safeFloatCast(vals[5]) - driver = vals[6] - gpu_name = vals[7] - serial = vals[8] - display_active = vals[9] - display_mode = vals[10] - temp_gpu = safeFloatCast(vals[11]); - GPUs.append(GPU(deviceIds, uuid, gpuUtil, memTotal, memUsed, memFree, driver, gpu_name, serial, display_mode, display_active, temp_gpu)) - return GPUs # (deviceIds, gpuUtil, memUtil) - - - -def getGPUProcesses(): - """Get all gpu compute processes.""" - - global gpuUuidToIdMap - gpuUuidToIdMap = {} - try: - gpus = getGPUs() - for gpu in gpus: - gpuUuidToIdMap[gpu.uuid] = gpu.id - del gpus - except: - pass - - - nvidia_smi = getNvidiaSmiCmd() - try: - p = subprocess.run([ - nvidia_smi, - "--query-compute-apps=pid,process_name,gpu_uuid,gpu_name,used_memory", - "--format=csv,noheader,nounits" - ], stdout=subprocess.PIPE, encoding='utf8') - stdout, stderror = p.stdout, p.stderr - except: - return [] - output = stdout - ## Parse output - # Split on line break - lines = output.split(os.linesep) - numProcesses = len(lines) - 1 - processes = [] - for g in range(numProcesses): - line = lines[g] - #print(line) - vals = line.split(', ') - #print(vals) - pid = int(vals[0]) - processName = vals[1] - gpuUuid = vals[2] - gpuName = vals[3] - usedMemory = safeFloatCast(vals[4]) - gpuId = gpuUuidToIdMap[gpuUuid] - if gpuId is None: - gpuId = -1 - - # get uid and uname owner of the pid - try: - p = subprocess.run(['ps', f'-p{pid}', '-oruid=,ruser='], - stdout=subprocess.PIPE, encoding='utf8') - uid, uname = p.stdout.split() - uid = int(uid) - except: - uid, uname = -1, '' - - processes.append(GPUProcess(pid, processName, gpuId, gpuUuid, - gpuName, usedMemory, uid, uname)) - return processes - - -def getAvailable(order = 'first', limit=1, maxLoad=0.5, maxMemory=0.5, memoryFree=0, includeNan=False, excludeID=[], excludeUUID=[]): - # order = first | last | random | load | memory - # first --> select the GPU with the lowest ID (DEFAULT) - # last --> select the GPU with the highest ID - # random --> select a random available GPU - # load --> select the GPU with the lowest load - # memory --> select the GPU with the most memory available - # limit = 1 (DEFAULT), 2, ..., Inf - # Limit sets the upper limit for the number of GPUs to return. E.g. if limit = 2, but only one is available, only one is returned. - # Get device IDs, load and memory usage - GPUs = getGPUs() - # Determine, which GPUs are available - GPUavailability = getAvailability(GPUs, maxLoad=maxLoad, maxMemory=maxMemory, memoryFree=memoryFree, includeNan=includeNan, excludeID=excludeID, excludeUUID=excludeUUID) - availAbleGPUindex = [idx for idx in range(0,len(GPUavailability)) if (GPUavailability[idx] == 1)] - # Discard unavailable GPUs - GPUs = [GPUs[g] for g in availAbleGPUindex] - # Sort available GPUs according to the order argument - if (order == 'first'): - GPUs.sort(key=lambda x: float('inf') if math.isnan(x.id) else x.id, reverse=False) - elif (order == 'last'): - GPUs.sort(key=lambda x: float('-inf') if math.isnan(x.id) else x.id, reverse=True) - elif (order == 'random'): - GPUs = [GPUs[g] for g in random.sample(range(0,len(GPUs)),len(GPUs))] - elif (order == 'load'): - GPUs.sort(key=lambda x: float('inf') if math.isnan(x.load) else x.load, reverse=False) - elif (order == 'memory'): - GPUs.sort(key=lambda x: float('inf') if math.isnan(x.memoryUtil) else x.memoryUtil, reverse=False) - # Extract the number of desired GPUs, but limited to the total number of available GPUs - GPUs = GPUs[0:min(limit, len(GPUs))] - # Extract the device IDs from the GPUs and return them - deviceIds = [gpu.id for gpu in GPUs] - return deviceIds -#def getAvailability(GPUs, maxLoad = 0.5, maxMemory = 0.5, includeNan = False): -# # Determine, which GPUs are available -# GPUavailability = np.zeros(len(GPUs)) -# for i in range(len(GPUs)): -# if (GPUs[i].load < maxLoad or (includeNan and np.isnan(GPUs[i].load))) and (GPUs[i].memoryUtil < maxMemory or (includeNan and np.isnan(GPUs[i].memoryUtil))): -# GPUavailability[i] = 1 -def getAvailability(GPUs, maxLoad=0.5, maxMemory=0.5, memoryFree=0, includeNan=False, excludeID=[], excludeUUID=[]): - # Determine, which GPUs are available - GPUavailability = [1 if (gpu.memoryFree>=memoryFree) and (gpu.load < maxLoad or (includeNan and math.isnan(gpu.load))) and (gpu.memoryUtil < maxMemory or (includeNan and math.isnan(gpu.memoryUtil))) and ((gpu.id not in excludeID) and (gpu.uuid not in excludeUUID)) else 0 for gpu in GPUs] - return GPUavailability -def getFirstAvailable(order = 'first', maxLoad=0.5, maxMemory=0.5, attempts=1, interval=900, verbose=False, includeNan=False, excludeID=[], excludeUUID=[]): - #GPUs = getGPUs() - #firstAvailableGPU = np.NaN - #for i in range(len(GPUs)): - # if (GPUs[i].load < maxLoad) & (GPUs[i].memory < maxMemory): - # firstAvailableGPU = GPUs[i].id - # break - #return firstAvailableGPU - for i in range(attempts): - if (verbose): - print('Attempting (' + str(i+1) + '/' + str(attempts) + ') to locate available GPU.') - # Get first available GPU - available = getAvailable(order=order, limit=1, maxLoad=maxLoad, maxMemory=maxMemory, includeNan=includeNan, excludeID=excludeID, excludeUUID=excludeUUID) - # If an available GPU was found, break for loop. - if (available): - if (verbose): - print('GPU ' + str(available) + ' located!') - break - # If this is not the last attempt, sleep for 'interval' seconds - if (i != attempts-1): - time.sleep(interval) - # Check if an GPU was found, or if the attempts simply ran out. Throw error, if no GPU was found - if (not(available)): - raise RuntimeError('Could not find an available GPU after ' + str(attempts) + ' attempts with ' + str(interval) + ' seconds interval.') - # Return found GPU - return available -def showUtilization(all=False, attrList=None, useOldCode=False): - GPUs = getGPUs() - if (all): - if (useOldCode): - print(' ID | Name | Serial | UUID || GPU util. | Memory util. || Memory total | Memory used | Memory free || Display mode | Display active |') - print('------------------------------------------------------------------------------------------------------------------------------') - for gpu in GPUs: - print(' {0:2d} | {1:s} | {2:s} | {3:s} || {4:3.0f}% | {5:3.0f}% || {6:.0f}MB | {7:.0f}MB | {8:.0f}MB || {9:s} | {10:s}'.format(gpu.id,gpu.name,gpu.serial,gpu.uuid,gpu.load*100,gpu.memoryUtil*100,gpu.memoryTotal,gpu.memoryUsed,gpu.memoryFree,gpu.display_mode,gpu.display_active)) - else: - attrList = [[{'attr':'id','name':'ID'}, - {'attr':'name','name':'Name'}, - {'attr':'serial','name':'Serial'}, - {'attr':'uuid','name':'UUID'}], - [{'attr':'temperature','name':'GPU temp.','suffix':'C','transform': lambda x: x,'precision':0}, - {'attr':'load','name':'GPU util.','suffix':'%','transform': lambda x: x*100,'precision':0}, - {'attr':'memoryUtil','name':'Memory util.','suffix':'%','transform': lambda x: x*100,'precision':0}], - [{'attr':'memoryTotal','name':'Memory total','suffix':'MB','precision':0}, - {'attr':'memoryUsed','name':'Memory used','suffix':'MB','precision':0}, - {'attr':'memoryFree','name':'Memory free','suffix':'MB','precision':0}], - [{'attr':'display_mode','name':'Display mode'}, - {'attr':'display_active','name':'Display active'}]] - - else: - if (useOldCode): - print(' ID GPU MEM') - print('--------------') - for gpu in GPUs: - print(' {0:2d} {1:3.0f}% {2:3.0f}%'.format(gpu.id, gpu.load*100, gpu.memoryUtil*100)) - else: - attrList = [[{'attr':'id','name':'ID'}, - {'attr':'load','name':'GPU','suffix':'%','transform': lambda x: x*100,'precision':0}, - {'attr':'memoryUtil','name':'MEM','suffix':'%','transform': lambda x: x*100,'precision':0}], - ] - - if (not useOldCode): - if (attrList is not None): - headerString = '' - GPUstrings = ['']*len(GPUs) - for attrGroup in attrList: - #print(attrGroup) - for attrDict in attrGroup: - headerString = headerString + '| ' + attrDict['name'] + ' ' - headerWidth = len(attrDict['name']) - minWidth = len(attrDict['name']) - - attrPrecision = '.' + str(attrDict['precision']) if ('precision' in attrDict.keys()) else '' - attrSuffix = str(attrDict['suffix']) if ('suffix' in attrDict.keys()) else '' - attrTransform = attrDict['transform'] if ('transform' in attrDict.keys()) else lambda x : x - for gpu in GPUs: - attr = getattr(gpu,attrDict['attr']) - - attr = attrTransform(attr) - - if (isinstance(attr,float)): - attrStr = ('{0:' + attrPrecision + 'f}').format(attr) - elif (isinstance(attr,int)): - attrStr = ('{0:d}').format(attr) - elif (isinstance(attr,str)): - attrStr = attr; - elif (sys.version_info[0] == 2): - if (isinstance(attr,unicode)): - attrStr = attr.encode('ascii','ignore') - else: - raise TypeError('Unhandled object type (' + str(type(attr)) + ') for attribute \'' + attrDict['name'] + '\'') - - attrStr += attrSuffix - - minWidth = max(minWidth,len(attrStr)) - - headerString += ' '*max(0,minWidth-headerWidth) - - minWidthStr = str(minWidth - len(attrSuffix)) - - for gpuIdx,gpu in enumerate(GPUs): - attr = getattr(gpu,attrDict['attr']) - - attr = attrTransform(attr) - - if (isinstance(attr,float)): - attrStr = ('{0:'+ minWidthStr + attrPrecision + 'f}').format(attr) - elif (isinstance(attr,int)): - attrStr = ('{0:' + minWidthStr + 'd}').format(attr) - elif (isinstance(attr,str)): - attrStr = ('{0:' + minWidthStr + 's}').format(attr); - elif (sys.version_info[0] == 2): - if (isinstance(attr,unicode)): - attrStr = ('{0:' + minWidthStr + 's}').format(attr.encode('ascii','ignore')) - else: - raise TypeError('Unhandled object type (' + str(type(attr)) + ') for attribute \'' + attrDict['name'] + '\'') - - attrStr += attrSuffix - - GPUstrings[gpuIdx] += '| ' + attrStr + ' ' - - headerString = headerString + '|' - for gpuIdx,gpu in enumerate(GPUs): - GPUstrings[gpuIdx] += '|' - - headerSpacingString = '-' * len(headerString) - print(headerString) - print(headerSpacingString) - for GPUstring in GPUstrings: - print(GPUstring) - - -# Generate gpu uuid to id map -gpuUuidToIdMap = {} -try: - gpus = getGPUs() - for gpu in gpus: - gpuUuidToIdMap[gpu.uuid] = gpu.id - del gpus -except: - pass -def getGPUInfos(): - ###返回gpus：list,一个GPU为一个元素-对象 - ###########：有属性，'id','load','memoryFree', - ###########：'memoryTotal','memoryUsed','memoryUtil','name','serial''temperature','uuid',process - ###其中process：每一个计算进程是一个元素--对象 - ############：有属性，'gpuId','gpuName','gpuUuid', - ############：'gpuid','pid','processName','uid', 'uname','usedMemory' - gpus = getGPUs() - gpuUuidToIdMap={} - for gpu in gpus: - gpuUuidToIdMap[gpu.uuid] = gpu.id - gpu.process=[] - indexx = [x.id for x in gpus ] - - process = getGPUProcesses() - for pre in process: - pre.gpuid = gpuUuidToIdMap[pre.gpuUuid] - gpuId = indexx.index(pre.gpuid ) - gpus[gpuId].process.append(pre ) - return gpus - -def get_available_gpu(gpuStatus): - ##判断是否有空闲的显卡，如果有返回id，没有返回None - cuda=None - for gpus in gpuStatus: - if len(gpus.process) == 0: - cuda = gpus.id - return str(cuda) - return cuda -def get_whether_gpuProcess(): - ##判断是否有空闲的显卡，如果有返回id，没有返回None - gpuStatus=getGPUInfos() - gpuProcess=True - for gpus in gpuStatus: - if len(gpus.process) != 0: - gpuProcess = False - return gpuProcess - -def get_offlineProcess_gpu(gpuStatus,pidInfos): - gpu_onLine = [] - for gpu in gpuStatus: - for gpuProcess in gpu.process: - pid = gpuProcess.pid - if pid in pidInfos.keys(): - pidType = pidInfos[pid]['type'] - if pidType == 'onLine': - gpu_onLine.append(gpu) - gpu_offLine = set(gpuStatus) - set(gpu_onLine) - return list(gpu_offLine) -def arrange_offlineProcess(gpuStatus,pidInfos,modelMemory=1500): - cudaArrange=[] - gpu_offLine = get_offlineProcess_gpu(gpuStatus,pidInfos) - for gpu in gpu_offLine: - leftMemory = gpu.memoryTotal*0.9 - gpu.memoryUsed - modelCnt = int(leftMemory// modelMemory) - - cudaArrange.extend( [gpu.id] * modelCnt ) - return cudaArrange -def get_potential_gpu(gpuStatus,pidInfos): - ###所有GPU上都有计算。需要为“在线任务”空出一块显卡。 - ###step1：查看所有显卡上是否有“在线任务” - - gpu_offLine = get_offlineProcess_gpu(gpuStatus,pidInfos) - if len(gpu_offLine) == 0 : - return False - - ###step2,找出每张显卡上离线进程的数目 - offLineCnt = [ len(gpu.process) for gpu in gpu_offLine ] - minCntIndex =offLineCnt.index( min(offLineCnt)) - - pids = [x.pid for x in gpu_offLine[minCntIndex].process] - return {'cuda':gpu_offLine[minCntIndex].id,'pids':pids } -if __name__=='__main__': - #pres = getGPUProcesses() - #print('###line404:',pres) - gpus = getGPUs() - for gpu in gpus: - gpuUuidToIdMap[gpu.uuid] = gpu.id - print(gpu) - print(gpuUuidToIdMap) - pres = getGPUProcesses() - print('###line404:',pres) - for pre in pres: - print('#'*20) - for ken in ['gpuName','gpuUuid','pid','processName','uid','uname','usedMemory' ]: - print(ken,' ',pre.__getattribute__(ken )) - print(' ') - - diff --git a/AIlib/utilsK/__pycache__/GPUtils.cpython-38.pyc b/AIlib/utilsK/__pycache__/GPUtils.cpython-38.pyc deleted file mode 100644 index 2c63946..0000000 Binary files a/AIlib/utilsK/__pycache__/GPUtils.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utilsK/__pycache__/maskUtils.cpython-38.pyc b/AIlib/utilsK/__pycache__/maskUtils.cpython-38.pyc deleted file mode 100644 index f88b60a..0000000 Binary files a/AIlib/utilsK/__pycache__/maskUtils.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utilsK/__pycache__/masterUtils.cpython-38.pyc b/AIlib/utilsK/__pycache__/masterUtils.cpython-38.pyc deleted file mode 100644 index 59098a6..0000000 Binary files a/AIlib/utilsK/__pycache__/masterUtils.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utilsK/__pycache__/modelEval.cpython-38.pyc b/AIlib/utilsK/__pycache__/modelEval.cpython-38.pyc deleted file mode 100644 index 720bfcb..0000000 Binary files a/AIlib/utilsK/__pycache__/modelEval.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utilsK/__pycache__/queRiver.cpython-38.pyc b/AIlib/utilsK/__pycache__/queRiver.cpython-38.pyc deleted file mode 100644 index e871c8b..0000000 Binary files a/AIlib/utilsK/__pycache__/queRiver.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utilsK/__pycache__/sendUtils.cpython-38.pyc b/AIlib/utilsK/__pycache__/sendUtils.cpython-38.pyc deleted file mode 100644 index 1724baa..0000000 Binary files a/AIlib/utilsK/__pycache__/sendUtils.cpython-38.pyc and /dev/null differ diff --git a/AIlib/utilsK/masterUtils.py b/AIlib/utilsK/masterUtils.py deleted file mode 100644 index 236c64d..0000000 --- a/AIlib/utilsK/masterUtils.py +++ /dev/null @@ -1,303 +0,0 @@ -from kafka import KafkaProducer, KafkaConsumer,TopicPartition -from kafka.errors import kafka_errors -import os,cv2,sys,json,time -import numpy as np -import requests -def query_channel_status(channelIndex): - channel_query_api='https://streaming.t-aaron.com/livechannel/getLiveStatus/%s'%(channelIndex) - #https://streaming.t-aaron.com/livechannel/getLiveStatus/LC001 - try: - res = requests.get(channel_query_api,timeout=10).json() - if res['data']['status']==2:#1空闲中 2使用中 3停用 4待关闭 - taskEnd=False - else: - taskEnd=True - infos='channel_query_api connected' - except Exception as e: - taskEnd=True - infos='channel_query_api not connected:%s'%(e) - return infos, taskEnd - -def query_request_status(request_url): - #channel_query_api='https://streaming.t-aaron.com/livechannel/getLiveStatus/%s'%(channelIndex) - channel_request_api=request_url - - try: - res = requests.get(channel_request_api,timeout=10).json() - if res['data']['status']==5:#5：执行中 10：待停止分析 15：执行结束 - taskEnd=False - else: - taskEnd=True - infos='channel_request_api connected' - except Exception as e: - taskEnd=True - infos='channel_request_api not connected:%s'%(e) - return infos, taskEnd - -def get_needed_objectsIndex(object_config): - needed_objectsIndex=[] - - for model in object_config: - try: - needed_objectsIndex.append(int(model['id'])) - except Exception as e: - a=1 - allowedList_str=[str(x) for x in needed_objectsIndex] - allowedList_string=','.join(allowedList_str) - - return needed_objectsIndex , allowedList_string - - -def get_infos(taskId, msgId,msg_h,key_str='waiting stream or video, send heartbeat'): - outStrList={} - outStrList['success']= '%s, taskId:%s msgId:%s send:%s'%(key_str,taskId, msgId,msg_h); - outStrList['failure']='kafka ERROR, %s'%(key_str) - outStrList['Refailure']='kafka Re-send ERROR ,%s'%(key_str) - return outStrList -def writeTxtEndFlag(outImaDir,streamName,imageTxtFile,endFlag='结束'): -#time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) - EndUrl='%s/%s_frame-9999-9999_type-%s_9999999999999999_s-%s_AI.jpg'%(outImaDir,time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()),endFlag,streamName) - EndUrl = EndUrl.replace(' ','-').replace(':','-') - img_end=np.zeros((100,100),dtype=np.uint8);cv2.imwrite(EndUrl,img_end) - if imageTxtFile: - EndUrl_txt = EndUrl.replace('.jpg','.txt') - fp_t=open(EndUrl_txt,'w');fp_t.write(EndUrl+'\n');fp_t.close() - - EndUrl='%s/%s_frame-9999-9999_type-%s_9999999999999999_s-%s_OR.jpg'%(outImaDir,time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()),endFlag,streamName) - EndUrl = EndUrl.replace(' ','-').replace(':','-') - ret = cv2.imwrite(EndUrl,img_end) - if imageTxtFile: - EndUrl_txt = EndUrl.replace('.jpg','.txt') - fp_t=open(EndUrl_txt,'w');fp_t.write(EndUrl+'\n');fp_t.close() -def get_current_time(): - """[summary] 获取当前时间 - - [description] 用time.localtime()+time.strftime()实现 - :returns: [description] 返回str类型 - """ - ct = time.time() - local_time = time.localtime(ct) - data_head = time.strftime("%Y-%m-%d %H:%M:%S", local_time) - data_secs = (ct - int(ct)) * 1000 - time_stamp = "%s.%03d" % (data_head, data_secs) - return time_stamp - - - -def send_kafka(producer,par,msg,outStrList,fp_log,logger,line='000',thread='detector',printFlag=False ): - future = producer.send(par['topic'], msg) - try: - record_metadata = future.get() - outstr=outStrList['success'] - - #outstr=wrtiteLog(fp_log,outstr);print( outstr); - writeELK_log(outstr,fp_log,level='INFO',thread=thread,line=line,logger=logger,printFlag=printFlag) - - except Exception as e: - outstr='%s , warning: %s'%( outStrList['failure'],str(e)) - writeELK_log(outstr,fp_log,level='WARNING',thread=thread,line=line,logger=logger,printFlag=printFlag) - try: - producer.close() - producer = KafkaProducer(bootstrap_servers=par['server'], value_serializer=lambda v: v.encode('utf-8')).get() - future = producer.send(par['topic'], msg).get() - except Exception as e: - outstr='%s, error: %s'%( outStrList['Refailure'],str(e)) - #outstr=wrtiteLog(fp_log,outstr);print( outstr); - writeELK_log(outstr,fp_log,level='ERROR',thread=thread,line=line,logger=logger,printFlag=printFlag) - -def check_time_interval(time0_beg,time_interval): - time_2 = time.time() - if time_2 - time0_beg>time_interval: - return time_2,True - else: - return time0_beg,False -def addTime(strs): - timestr=time.strftime("%Y-%m-%d %H:%M:%S ", time.localtime()) - - outstr='\n %s %s '%(timestr,strs) - return - - -def get_file(): - print("文件名 :",__file__,sys._getframe().f_lineno) - print("函数名: ", sys._getframe().f_code.co_name) - print("模块名: ", sys._getframe().f_back.f_code.co_name) - -def writeELK_log(msg,fp,level='INFO',thread='detector',logger='kafka_yolov5',line=9999,newLine=False,printFlag=True): - #timestr=time.strftime("%Y-%m-%d %H:%M:%S ", time.localtime()) - timestr=get_current_time() - outstr='%s [%s][%s][%d][%s]- %s'%(timestr,level,thread,line,logger,msg) - - if newLine: - outstr = '\n'+outstr - - fp.write(outstr+'\n') - fp.flush() - if printFlag: - print(outstr) - return outstr - - -def wrtiteLog(fp,strs,newLine=False): - timestr=time.strftime("%Y-%m-%d %H:%M:%S ", time.localtime()) - if newLine: - outstr='\n %s %s '%(timestr,strs) - else: - outstr='%s %s '%(timestr,strs) - fp.write(outstr+'\n') - fp.flush() - return outstr - -def create_logFile(logdir='logdir',name=None): - if name: - logname =logdir+'/'+ name - else: - logname =logdir+'/'+ time.strftime("%Y-%m-%d.txt", time.localtime()) - if os.path.exists(logname): - fp_log = open(logname,'a+') - else: - fp_log = open(logname,'w') - return fp_log -def get_boradcast_address(outResource): - #rtmp://live.push.t-aaron.com/live/THSB，阿里云，1945 - #rtmp://demopush.yunhengzhizao.cn/live/THSB，腾讯云，1935 - if '1945' in outResource: - return 'rtmp://live.play.t-aaron.com/live/THSB' - else: - return 'rtmp://demoplay.yunhengzhizao.cn/live/THSB_HD5M' -def save_message(kafka_dir,msg): - outtxt=os.path.join(kafka_dir,msg['request_id']+'.json') - assert os.path.exists(kafka_dir) - with open(outtxt,'w') as fp: - json.dump(msg,fp,ensure_ascii=False) - - - -def get_push_address(outResource): - #rtmp://live.push.t-aaron.com/live/THSB，阿里云，1945 - #rtmp://demopush.yunhengzhizao.cn/live/THSB，腾讯云，1935 - #终端推流地址：rtmp://live.push.t-aaron.com/live/THSAa - #终端拉流地址：rtmp://live.play.t-aaron.com/live/THSAa_hd - #AI推流地址：rtmp://live.push.t-aaron.com/live/THSBa - #AI拉流地址：rtmp://live.play.t-aaron.com/live/THSBa_hd - - if 't-aaron' in outResource: - if 'THSBa' in outResource: port=1975 - elif 'THSBb' in outResource: port=1991 - elif 'THSBc' in outResource: port=1992 - elif 'THSBd' in outResource: port=1993 - elif 'THSBe' in outResource: port=1994 - elif 'THSBf' in outResource: port=1995 - elif 'THSBg' in outResource: port=1996 - elif 'THSBh' in outResource: port=1997 - else: port=1945 - else: port=1935 - return 'rtmp://127.0.0.1:%d/live/test'%(port) - return outResource -def getAllRecord_poll(consumer): - msgs = consumer.poll(5000) - keys=msgs.keys() - out = [ msgs[x] for x in keys] - out = [y for x in out for y in x] - - - for key in keys: - out.extend(msgs[key]) - return out -def getAllRecords(consumer,topics): - leftCnt = 0 - for topic in topics[0:2]: - leftCnt+=get_left_cnt(consumer,topic) - out = [] - if leftCnt == 0: - return [] - for ii,msg in enumerate(consumer): - consumer.commit() - out.append(msg) - if ii== (leftCnt-1): - break###断流或者到终点 - return out - -def get_left_cnt(consumer,topic): - partitions = [TopicPartition(topic, p) for p in consumer.partitions_for_topic(topic)] - - # total - toff = consumer.end_offsets(partitions) - toff = [(key.partition, toff[key]) for key in toff.keys()] - toff.sort() - - # current - coff = [(x.partition, consumer.committed(x)) for x in partitions] - coff.sort() - - # cal sum and left - toff_sum = sum([x[1] for x in toff]) - cur_sum = sum([x[1] for x in coff if x[1] is not None]) - left_sum = toff_sum - cur_sum - - return left_sum -def view_bar(num, total,time1,prefix='prefix'): - rate = num / total - time_n=time.time() - rate_num = int(rate * 30) - rate_nums = np.round(rate * 100) - r = '\r %s %d / %d [%s%s] %.2f s'%(prefix,num,total, ">" * rate_num, " " * (30 - rate_num), time_n-time1 ) - sys.stdout.write(r) - sys.stdout.flush() -def get_total_cnt(inSource): - cap=cv2.VideoCapture(inSource) - assert cap.isOpened() - cnt=cap.get(7) - fps = cap.get(cv2.CAP_PROP_FPS) - cap.release() - return cnt,fps -def check_stream(inSource,producer,par,msg,outStrList ,fp_log,logger,line='000',thread='detector',timeMs=120,): - cnt =(timeMs-1)//10 + 1 - Stream_ok=False - - for icap in range(cnt): - cap=cv2.VideoCapture(inSource) - - if cap.isOpened() and get_fps_rtmp(inSource,video=False)[0] : - Stream_ok=True ;cap.release();break; - #Stream_ok,_= get_fps_rtmp(inSource,video=False) - #if Stream_ok:cap.release();break; - else: - Stream_ok=False - timestr=time.strftime("%Y-%m-%d %H:%M:%S ", time.localtime()) - outstr='Waiting stream %d s'%(10*icap) - writeELK_log(msg=outstr,fp=fp_log,thread=thread,line=line,logger=logger) - time.sleep(10) - if icap%3==0: - send_kafka(producer,par,msg,outStrList,fp_log,logger=logger,line=line,thread=thread ) - - - return Stream_ok - - - - -def get_fps_rtmp(inSource,video=False): - cap=cv2.VideoCapture(inSource) - if not cap.isOpened(): - print('#####error url:',inSource) - return False,[0,0,0,0] - - fps = cap.get(cv2.CAP_PROP_FPS) - width = cap.get(cv2.CAP_PROP_FRAME_WIDTH ) - height = cap.get(cv2.CAP_PROP_FRAME_HEIGHT) - cnt = 0 - if video: cnt=cap.get(7) - - if width*height==0 or fps>30: - return False,[0,0,0,0] - cap.release() - try: - outx = [fps,width,height,cnt] - outx = [int(x+0.5) for x in outx] - - return True,outx - except: - return False, [0,0,0,0] - - diff --git a/AIlib/utilsK/modelEval.py b/AIlib/utilsK/modelEval.py deleted file mode 100644 index 7fb7c1f..0000000 --- a/AIlib/utilsK/modelEval.py +++ /dev/null @@ -1,110 +0,0 @@ -import sys -sys.path.extend(['/home/thsw2/WJ/src/yolov5']) -import utils,json,time,torch -import numpy as np -from segutils.segmodel import SegModel,get_largest_contours -from models.experimental import attempt_load -from utils.torch_utils import select_device, load_classifier, time_synchronized -import subprocess as sp -import cv2 -from utils.datasets import LoadStreams, LoadImages -from queRiver import get_labelnames,get_label_arrays,post_process_,save_problem_images,time_str -def get_total_cnt(inSource): - cap=cv2.VideoCapture(inSource) - cnt=cap.get(7) - cap.release() - return cnt -def onlineModelProcess(parIn ): - streamName = parIn['streamName'] - childCallback=parIn['callback'] - try: - - inSource,outSource=parIn['inSource'],parIn['outSource'] - weights='../yolov5/weights/1230_last.pt' - device = select_device('0') - half = device.type != 'cpu' # half precision only supported on CUDA - model = attempt_load(weights, map_location=device) # load FP32 model - if half: model.half() - seg_nclass = 2 - weights = '../yolov5/weights/segmentation/BiSeNet/checkpoint.pth' - segmodel = SegModel(nclass=seg_nclass,weights=weights,device=device) - jsonfile='../yolov5/config/queRiver.json' - with open(jsonfile,'r') as fp: - parAll = json.load(fp) - - - resource=parAll['prep_process']['source'] - if outSource: - command=['ffmpeg','-y','-f', 'rawvideo','-vcodec','rawvideo','-pix_fmt', 'bgr24', - '-s', "{}x{}".format(parAll["push_process"]['OutVideoW'],parAll["push_process"]['OutVideoH']),# 图片分辨率 - '-r', str(30),# 视频帧率 - '-i', '-','-c:v', 'libx264','-pix_fmt', 'yuv420p', - '-f', 'flv',outSource - ] - txtname='mintors/%s.txt'%( time.strftime("%Y-%m-%d", time.localtime()) ) - fp_out = open( txtname,'a+' ) - outstr='%s stream:%s starts \n'%( time_str(),parAll['push_process']['rtmpUrl']) - fp_out.write(outstr);fp_out.flush() - - - - - # 管道配置，其中用到管道 - if outSource: - ppipe = sp.Popen(command, stdin=sp.PIPE) - - ##后处理参数 - par=parAll['post_process'] - conf_thres,iou_thres,classes=par['conf_thres'],par['iou_thres'],par['classes'] - labelnames=par['labelnames'] - rainbows=par['rainbows'] - fpsample = par['fpsample'] - names=get_labelnames(labelnames) - label_arraylist = get_label_arrays(names,rainbows,outfontsize=40) - - dataset = LoadStreams(inSource, img_size=640, stride=32) - if (inSource.endswith('.MP4')) or (inSource.endswith('.mp4')): - totalcnt=get_total_cnt(inSource) - childCallback.send('####model load success####') - iframe = 0;post_results=[];time_beg=time.time() - print('###line71 modelEval.py',totalcnt,len(dataset), inSource) - for path, img, im0s, vid_cap in dataset: - print(path) - if not path:childCallback.send('####strem ends####'); break###断流或者到终点 - if not outSource:###如果不推流，则显示进度条 - view_bar(iframe,totalcnt,time_beg ) - time0=time.time() - iframe +=1 - time1=time.time() - img = torch.from_numpy(img).to(device) - img = img.half() if half else img.float() # uint8 to fp16/32 - img /= 255.0 # 0 - 255 to 0.0 - 1.0 - time2 = time.time() - pred = model(img,augment=False)[0] - time3 = time.time() - seg_pred,segstr = segmodel.eval(im0s[0] ) - - time4 = time.time() - datas = [path, img, im0s, vid_cap,pred,seg_pred,iframe] - - p_result,timeOut = post_process_(datas,conf_thres, iou_thres,names,label_arraylist,rainbows,iframe) - ##每隔 fpsample帧处理一次，如果有问题就保存图片 - if (iframe % fpsample == 0) and (len(post_results)>0) : - parImage=save_problem_images(post_results,iframe,names,streamName=streamName) - - #parOut = {}; parOut['imgOR'] = img_send; parOut['imgAR'] = img_send; parOut['uid']=uid - #parOut['imgORname']=os.path.basename(outnameOR);parOut['imgARname']=os.path.basename(outnameAR); - #parOut['time_str'] = time_str;parOut['type'] = names[cls_max] - - - - post_results=[] - - if len(p_result[2] )>0: ## - post_results.append(p_result) - - image_array = p_result[1] - if outSource: - ppipe.stdin.write(image_array.tostring()) - except Exception as e: - childCallback.send(e) #将异常通过管道送出 diff --git a/AIlib/utilsK/queRiver.py b/AIlib/utilsK/queRiver.py deleted file mode 100644 index 3167a89..0000000 --- a/AIlib/utilsK/queRiver.py +++ /dev/null @@ -1,307 +0,0 @@ -from kafka import KafkaProducer, KafkaConsumer -from kafka.errors import kafka_errors -import traceback -import json, base64,os -import numpy as np -from multiprocessing import Process,Queue -import time,cv2,string,random -import subprocess as sp - -import matplotlib.pyplot as plt -from utils.datasets import LoadStreams, LoadImages -from models.experimental import attempt_load -from utils.general import check_img_size, check_requirements, check_imshow, non_max_suppression, apply_classifier, scale_coords, xyxy2xywh, strip_optimizer, set_logging, increment_path -import torch,sys -#from segutils.segmodel import SegModel,get_largest_contours -#sys.path.extend(['../yolov5/segutils']) - -from segutils.segWaterBuilding import SegModel,get_largest_contours,illBuildings - -#from segutils.core.models.bisenet import BiSeNet -from segutils.core.models.bisenet import BiSeNet_MultiOutput - -from utils.plots import plot_one_box,plot_one_box_PIL,draw_painting_joint,get_label_arrays,get_websource -from collections import Counter -#import matplotlib -import matplotlib.pyplot as plt -# get_labelnames,get_label_arrays,post_process_,save_problem_images,time_str -#FP_DEBUG=open('debut.txt','w') -def bsJpgCode(image_ori): - jpgCode = cv2.imencode('.jpg',image_ori)[-1]###np.array,(4502009,1) - bsCode = str(base64.b64encode(jpgCode))[2:-1] ###str,长6002680 - return bsCode -def bsJpgDecode(bsCode): - bsDecode = base64.b64decode(bsCode)###types,长4502009 - npString = np.frombuffer(bsDecode,np.uint8)###np.array,(长4502009,) - jpgDecode = cv2.imdecode(npString,cv2.IMREAD_COLOR)###np.array,(3000,4000,3) - return jpgDecode -def get_ms(time0,time1): - str_time ='%.2f ms'%((time1-time0)*1000) - return str_time -rainbows=[ - (0,0,255),(0,255,0),(255,0,0),(255,0,255),(255,255,0),(255,127,0),(255,0,127), - (127,255,0),(0,255,127),(0,127,255),(127,0,255),(255,127,255),(255,255,127), - (127,255,255),(0,255,255),(255,127,255),(127,255,255), - (0,127,0),(0,0,127),(0,255,255) - ] - - -def get_labelnames(labelnames): - with open(labelnames,'r') as fp: - namesjson=json.load(fp) - names_fromfile=namesjson['labelnames'] - names = names_fromfile - return names - -def check_stream(stream): - cap = cv2.VideoCapture(stream) - if cap.isOpened(): - return True - else: - return False -##### -def drawWater(pred,image_array0):####pred是模型的输出，只有水分割的任务 - ##画出水体区域 - contours, hierarchy = cv2.findContours(pred,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - water = pred.copy(); water[:,:] = 0 - - if len(contours)==0: - return image_array0,water - max_id = get_largest_contours(contours); - cv2.fillPoly(water, [contours[max_id][:,0,:]], 1) - cv2.drawContours(image_array0,contours,max_id,(0,255,255),3) - return image_array0,water - - - -def post_process_(datas,conf_thres, iou_thres,names,label_arraylist,rainbows,iframe,object_config=[0,1,2,3,4]): - ##输入dataset genereate 生成的数据，model预测的结果pred，nms参数 - ##主要操作NMS ---> 坐标转换 ---> 画图 - ##输出原图、AI处理后的图、检测结果 - time0=time.time() - path, img, im0s, vid_cap ,pred,seg_pred= datas[0:6]; - segmodel=True - pred = non_max_suppression(pred, conf_thres, iou_thres, classes=None, agnostic=False) - time1=time.time() - i=0;det=pred[0]###一次检测一张图片 - p, s, im0 = path[i], '%g: ' % i, im0s[i].copy() - gn = torch.tensor(im0.shape)[[1, 0, 1, 0]] # normalization gain whwh - det_xywh=[]; - #im0_brg=cv2.cvtColor(im0,cv2.COLOR_RGB2BGR); - if len(seg_pred)==2: - im0,water = illBuildings(seg_pred,im0) - else: - im0,water = drawWater(seg_pred,im0) - time2=time.time() - #plt.imshow(im0);plt.show() - if len(det)>0: - # Rescale boxes from img_size to im0 size - det[:, :4] = scale_coords(img.shape[2:], det[:, :4],im0.shape).round() - #用seg模型，确定有效检测匡及河道轮廓线 - if segmodel: - '''contours, hierarchy = cv2.findContours(seg_pred,cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE) - if len(contours)>0: - max_id = get_largest_contours(contours) - seg_pred[:,:] = 0 - cv2.fillPoly(seg_pred, [contours[max_id][:,0,:]], 1) - cv2.drawContours(im0,contours,max_id,(0,255,255),3)''' - det_c = det.clone(); det_c=det_c.cpu().numpy() - area_factors = np.array([np.sum(water[int(x[1]):int(x[3]), int(x[0]):int(x[2])] )/((x[2]-x[0])*(x[3]-x[1])) for x in det_c] ) - det = det[area_factors>0.1] - #对检测匡绘图 - for *xyxy, conf, cls in reversed(det): - xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist() # normalized xywh - cls_c = cls.cpu().numpy() - if int(cls_c) not in object_config: ###如果不是所需要的目标，则不显示 - continue - conf_c = conf.cpu().numpy() - line = [float(cls_c), *xywh, float(conf_c)] # label format - det_xywh.append(line) - label = f'{names[int(cls)]} {conf:.2f}' - - im0 = draw_painting_joint(xyxy,im0,label_arraylist[int(cls)],score=conf,color=rainbows[int(cls)%20],line_thickness=None) - time3=time.time() - strout='nms:%s illBuilding:%s detDraw:%s '%(get_ms(time0,time1),get_ms(time1,time2), get_ms(time2,time3) ) - return [im0s[0],im0,det_xywh,iframe],strout - - - -def preprocess(par): - print('#####process:',par['name']) - ##负责读取视频，生成原图及供检测的使用图，numpy格式 - #source='rtmp://liveplay.yunhengzhizao.cn/live/demo_HD5M' - #img_size=640; stride=32 - while True: - cap = cv2.VideoCapture(par['source']) - iframe = 0 - if cap.isOpened(): - print( '#### read %s success!'%(par['source'])) - try: - dataset = LoadStreams(par['source'], img_size=640, stride=32) - for path, img, im0s, vid_cap in dataset: - datas=[path, img, im0s, vid_cap,iframe] - par['queOut'].put(datas) - iframe +=1 - except Exception as e: - print('###read error:%s '%(par['source'])) - time.sleep(10) - iframe = 0 - - else: - print('###read error:%s '%(par['source'] )) - time.sleep(10) - iframe = 0 - -def gpu_process(par): - print('#####process:',par['name']) - half=True - ##gpu运算,检测模型 - weights = par['weights'] - device = par['device'] - print('###line127:',par['device']) - model = attempt_load(par['weights'], map_location=par['device']) # load FP32 model - if half: - model.half() - - ##gpu运算，分割模型 - seg_nclass = par['seg_nclass'] - seg_weights = par['seg_weights'] - - #segmodel = SegModel(nclass=seg_nclass,weights=seg_weights,device=device) - - - nclass = [2,2] - Segmodel = BiSeNet_MultiOutput(nclass) - weights='weights/segmentation/WaterBuilding.pth' - segmodel = SegModel(model=Segmodel,nclass=nclass,weights=weights,device='cuda:0',multiOutput=True) - while True: - if not par['queIn'].empty(): - time0=time.time() - datas = par['queIn'].get() - path, img, im0s, vid_cap,iframe = datas[0:5] - time1=time.time() - img = torch.from_numpy(img).to(device) - img = img.half() if half else img.float() # uint8 to fp16/32 - img /= 255.0 # 0 - 255 to 0.0 - 1.0 - time2 = time.time() - pred = model(img,augment=False)[0] - time3 = time.time() - seg_pred = segmodel.eval(im0s[0],outsize=None,smooth_kernel=20) - time4 = time.time() - fpStr= 'process:%s ,iframe:%d,getdata:%s,copygpu:%s,dettime:%s,segtime:%s , time:%s, queLen:%d '%( par['name'],iframe,get_ms(time0,time1) ,get_ms(time1,time2) ,get_ms(time2,time3) ,get_ms(time3,time4),get_ms(time0,time4) ,par['queIn'].qsize() ) - #FP_DEBUG.write( fpStr+'\n' ) - datasOut = [path, img, im0s, vid_cap,pred,seg_pred,iframe] - par['queOut'].put(datasOut) - if par['debug']: - print('#####process:',par['name'],' line107') - else: - time.sleep(1/300) -def get_cls(array): - dcs = Counter(array) - keys = list(dcs.keys()) - values = list(dcs.values()) - max_index = values.index(max(values)) - cls = int(keys[max_index]) - return cls -def save_problem_images(post_results,iimage_cnt,names,streamName='live-THSAHD5M',outImaDir='problems/images_tmp',imageTxtFile=False): - ## [cls, x,y,w,h, conf] - problem_image=[[] for i in range(6)] - - - dets_list = [x[2] for x in post_results] - - mean_scores=[ np.array(x)[:,5].mean() for x in dets_list ] ###mean conf - - best_index = mean_scores.index(max(mean_scores)) ##获取该批图片里，问题图片的index - best_frame = post_results[ best_index][3] ##获取绝对帧号 - img_send = post_results[best_index][1]##AI处理后的图 - img_bak = post_results[best_index][0]##原图 - cls_max = get_cls( x[0] for x in dets_list[best_index] ) - - - time_str = time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime()) - uid=''.join(random.sample(string.ascii_letters + string.digits, 16)) - #ori_name = '2022-01-20-15-57-36_frame-368-720_type-漂浮物_qVh4zI08ZlwJN9on_s-live-THSAHD5M_OR.jpg' - #2022-01-13-15-07-57_frame-9999-9999_type-结束_9999999999999999_s-off-XJRW20220110115904_AI.jpg - outnameOR= '%s/%s_frame-%d-%d_type-%s_%s_s-%s_AI.jpg'%(outImaDir,time_str,best_frame,iimage_cnt,names[cls_max],uid,streamName) - outnameAR= '%s/%s_frame-%d-%d_type-%s_%s_s-%s_OR.jpg'%(outImaDir,time_str,best_frame,iimage_cnt,names[cls_max],uid,streamName) - - cv2.imwrite(outnameOR,img_send) - cv2.imwrite(outnameAR,img_bak) - if imageTxtFile: - outnameOR_txt = outnameOR.replace('.jpg','.txt') - fp=open(outnameOR_txt,'w');fp.write(outnameOR+'\n');fp.close() - outnameAI_txt = outnameAR.replace('.jpg','.txt') - fp=open(outnameAI_txt,'w');fp.write(outnameAR+'\n');fp.close() - - parOut = {}; parOut['imgOR'] = img_send; parOut['imgAR'] = img_send; parOut['uid']=uid - parOut['imgORname']=os.path.basename(outnameOR);parOut['imgARname']=os.path.basename(outnameAR); - parOut['time_str'] = time_str;parOut['type'] = names[cls_max] - return parOut - - - - -def post_process(par): - - print('#####process:',par['name']) - ###post-process参数 - conf_thres,iou_thres,classes=par['conf_thres'],par['iou_thres'],par['classes'] - labelnames=par['labelnames'] - rainbows=par['rainbows'] - fpsample = par['fpsample'] - names=get_labelnames(labelnames) - label_arraylist = get_label_arrays(names,rainbows,outfontsize=40) - iimage_cnt = 0 - post_results=[] - while True: - if not par['queIn'].empty(): - time0=time.time() - datas = par['queIn'].get() - iframe = datas[6] - if par['debug']: - print('#####process:',par['name'],' line129') - p_result,timeOut = post_process_(datas,conf_thres, iou_thres,names,label_arraylist,rainbows,iframe) - par['queOut'].put(p_result) - ##输出结果 - - - - ##每隔 fpsample帧处理一次，如果有问题就保存图片 - if (iframe % fpsample == 0) and (len(post_results)>0) : - #print('####line204:',iframe,post_results) - save_problem_images(post_results,iframe,names) - post_results=[] - - if len(p_result[2] )>0: ## - #post_list = p_result.append(iframe) - post_results.append(p_result) - #print('####line201:',type(p_result)) - - time1=time.time() - outstr='process:%s ,iframe:%d,%s , time:%s, queLen:%d '%( par['name'],iframe,timeOut,get_ms(time0,time1) ,par['queIn'].qsize() ) - #FP_DEBUG.write(outstr +'\n') - #print( 'process:%s ,iframe:%d,%s , time:%s, queLen:%d '%( par['name'],iframe,timeOut,get_ms(time0,time1) ,par['queIn'].qsize() ) ) - else: - time.sleep(1/300) - - -def save_logfile(name,txt): - if os.path.exists(name): - fp=open(name,'r+') - else: - fp=open(name,'w') - - fp.write('%s %s \n'%(time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()),txt)) - fp.close() -def time_str(): - return time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) - - - -if __name__=='__main__': - jsonfile='config/queRiver.json' - #image_encode_decode() - work_stream(jsonfile) - #par={'name':'preprocess'} - #preprocess(par) diff --git a/AIlib/utilsK/sendUtils.py b/AIlib/utilsK/sendUtils.py deleted file mode 100644 index d42a84e..0000000 --- a/AIlib/utilsK/sendUtils.py +++ /dev/null @@ -1,207 +0,0 @@ -from aliyunsdkvod.request.v20170321 import GetPlayInfoRequest -import json -import traceback -from aliyunsdkcore.client import AcsClient - -from PIL import Image -import numpy as np -import cv2 -import base64 -import io,os,copy -import requests -import time,json -import string,random -import glob,string,sys -from multiprocessing import Process,Queue -import oss2 -from kafka import KafkaProducer, KafkaConsumer -from voduploadsdk.UploadVideoRequest import UploadVideoRequest -from voduploadsdk.AliyunVodUtils import * -from voduploadsdk.AliyunVodUploader import AliyunVodUploader -from datetime import datetime, date, timedelta -def get_today(): - return date.today().strftime("%Y-%m-%d") -def get_yesterday(beforeday=-1): - return (date.today() + timedelta(days =beforeday)).strftime("%Y-%m-%d") - -def get_videoUurl(videoBakDir,filename): - ###七天时间内 - potentialUrls=[ os.path.join( videoBakDir,get_yesterday(beforeday=-x),filename) for x in range(7) ] - existsList=[os.path.exists(x ) for x in potentialUrls] - for i,flag in enumerate(existsList): - if flag: return potentialUrls[i] - return potentialUrls[0] - -def getNamedic(jsonfile): - with open(jsonfile) as fp: - dataDic=json.load(fp) - #"labelnames":["排口","排污口","水生植被","漂浮物","其它"], - #"labelIndexs":["SL014","SL011","SL013","SL001","SL001" ] - - assert 'labelnames' in dataDic.keys() , 'labelnames is not the key in %s'%(jsonfile) - assert 'labelIndexs' in dataDic.keys() , 'labelIndexs is not the key in %s'%(jsonfile) - assert len(dataDic['labelnames'])==len(dataDic['labelIndexs']) - nameDic={} - for key,value in zip(dataDic['labelnames'],dataDic['labelIndexs']): - nameDic[key]=value - return nameDic - - -def get_play_info(clt, videoId): - request = GetPlayInfoRequest.GetPlayInfoRequest() - request.set_accept_format('JSON') - request.set_VideoId(videoId) - request.set_AuthTimeout(3600*5) - response = json.loads(clt.do_action_with_exception(request)) - return response - -def create_status_msg(msg_dict_off,taskInfos,sts='waiting'): - msg= copy.deepcopy(msg_dict_off) - msg=update_json(taskInfos,msg,offkeys=["request_id"] ) - msg['status']=sts - msg = json.dumps(msg, ensure_ascii=False) - return msg -# 填入AccessKey信息 -def init_vod_client(accessKeyId, accessKeySecret): - regionId = 'cn-shanghai' # 点播服务接入地域 - connectTimeout = 3 # 连接超时，单位为秒 - return AcsClient(accessKeyId, accessKeySecret, regionId, auto_retry=True, max_retry_time=3, timeout=connectTimeout) -def update_json(jsonOri,jsonNew,offkeys=["request_id" ]): - #{'biz_id': 'hehuzhang', 'mod_id': 'ai', 'request_id': 'bblvgyntTsZCamqjuLArkiSYIbKXEeWx', 'offering_id': 'http://vod.play.t-aaron.com/customerTrans/c49a2c620795d124f2ae4b10197b8d0e/303b7a58-17f3ef4494e-0004-f90c-f2c-7ec68.mp4', 'offering_type': 'mp4', 'results_base_dir': 'XJRW20220317153547', 'inSource': 'http://vod.play.t-aaron.com/customerTrans/c49a2c620795d124f2ae4b10197b8d0e/303b7a58-17f3ef4494e-0004-f90c-f2c-7ec68.mp4', 'outSource': 'NO'} - for key in offkeys: - jsonNew[key] = jsonOri[key] - return jsonNew -def get_time(filename): - #2021-10-09-11-44-51_frame-598-720_type-水生植被.jpg - sps=filename.strip().split('_')[0] - tsps=sps.split('-') - return '%s-%s-%s %s:%s:%s'%(tsps[0],tsps[1],tsps[2],tsps[3],tsps[4],tsps[5]) -def get_ms(time0,time1): - str_time ='%.2f ms'%((time1-time0)*1000) - return str_time - -def get_urls( platform_query_url,fp_log ): - try: - if os.path.exists(platform_query_url): - #print('###line49') - with open('SendLog/platformQuery.json','r') as fp: - res = json.load(fp) - else: - res = requests.get(platform_query_url,timeout=10).json() - #print('###line54') - questionUrl = res['data']['questionUrl'] ###直播流时，问题图片的推送地址 - offlineUrl = res['data']['offlineUrl'] ###http离线视频时，问题图片的推送地址 - except Exception as ee: - timestr=time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) - print('###### %s: file:send_transfer: error %s ,url:%s #####'%(timestr,ee,platform_query_url)) - outstr = '\n %s ###### get url platform error : update error:%s , url:%s'%( time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) ,ee,platform_query_url) - fp_log.write(outstr);fp_log.flush() - questionUrl="http://47.96.182.154:9040/api/taskFile/submitUAVKHQuestion" - offlineUrl ="http://47.96.182.154:9040/api/taskFile/submitUAVKHQuestion" - return questionUrl,offlineUrl -def parse_filename(filename_base): - #etc:2022-01-13-16-04-17_frame-823-1440_type-水生植被_hgYFEulc0dPIrG1S_s-off-XJRW20220113154959_AI.jpg - uid =filename_base.split('.')[0].split('_')[3].strip() - sourceType=filename_base.split('_')[4].split('-')[1] - sourceId=filename_base.split('_')[4].split('-')[2] - typename=filename_base.split('.')[0].split('_')[2].split('-')[1].strip() - return uid,sourceType,sourceId,typename -def b64encode_function(filename, filename_OR): - if os.path.exists(filename): - image_ori=cv2.imread(filename) - image_ori_OR=cv2.imread(filename_OR) - else: - image_ori = filename.copy() - image_ori_OR = image_ori_OR.copy() - image_pngcode = cv2.imencode('.jpg',image_ori)[-1] - image_pngcode_OR = cv2.imencode('.jpg',image_ori_OR)[-1] - image_code = str(base64.b64encode(image_pngcode))[2:-1] - image_code_OR = str(base64.b64encode(image_pngcode_OR))[2:-1] - return image_code, image_code_OR -def JsonSend(parIn): - - fp_log = parIn['fp_log'] - try: - response=requests.post(parIn['api'],json=parIn['input_'],timeout=10).json() - t3 = time.time() - print('\n file:%s encodetime:%.5f request time:%.5f,send to %s ,return code:%s, size:%.2f M \n'%(parIn['filename_base'],parIn['t2']-parIn['t1'],t3-parIn['t2'],api,response['code'],parIn['sizeImage'])) - outstr = '%s file:%s encodetime:%.5f request time:%.5f,send to %s ,return code:%s,size:%.2f M ,%s\n'%( time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()),parIn['filename_base'],parIn['t2']-parIn['t1'],t3-parIn['t2'],parIn['api'],response['code'],parIn['sizeImage'],parIn['dic_str']) - fp_log.write(outstr);fp_log.flush() - - except Exception as ee: - print('\n ######file:%s: upload error:%s,size:%.2f M'%(parIn['filename_base'],ee, parIn['sizeImage'])) - outstr = '\n%s ###### file:%s: upload error:%s , size:%.2f M'%( time.strftime("%Y-%m-%d %H:%M:%S", time.localtime()) ,parIn['filename_base'],ee,parIn['sizeImage']) - fp_log.write(outstr);fp_log.flush() - - -def dic2str(dic): - st='' - for key in dic.keys(): - st='%s %s:%s,'%(st,key,dic[key]) - return st -def createJsonInput(filename,offlineUrl,questionUrl): - flag = True - filename_base = os.path.basename(filename) - filename_OR=filename.replace('_AI.','_OR.') - if not os.path.exists(filename_OR ): - return False - - uid,sourceType, sourceId,typename = parse_filename(filename_base) - if (typename not in name_dic.keys()) or (typename == '排口'): - return False - api = questionUrl if sourceType=='live' else offlineUrl - - time_str = get_time(filename_base) - input_ ={ - 'imgName':os.path.basename(filename), - 'imgNameOriginal':os.path.basename(filename_OR), - 'time':time_str, - 'fid':uid, ###随机16位字符 - 'type':name_dic[typename],###这次先采用 ["排口","污口","水生植被","漂浮物","其它"] - 'typeId':nameID_dic[typename] - - } - if sourceType!='live': - input_['code']=sourceId;###只有离线视频才需要code， - - dic_str = dic2str(input_) - t1 = time.time() - - image_code, image_code_OR = b64encode_function(filename, filename_OR) - input_['imgData']=image_code - input_['imgDataOriginal']=image_code_OR - - sizeImage = (len(image_code) + len(image_code_OR) )/1000000.0 - - parOut={};parOut['flag']=True;parOut['input_']=input_; - parOut['sizeImage']=sizeImage;parOut['dic_str']=dic_str; - parOut['filename']=filename;parOut['filename_OR']=filename_OR; - parOut['api']=api ; parOut['t1']=t1 ; parOut['filename_base']= filename_base - return parOut - -def getLogFileFp(streamName): - logname ='SendLog/'+ time.strftime("%Y-%m-%d", time.localtime())+'_%s.txt'%(streamName) - if os.path.exists(logname): - fp_log = open(logname,'a+') - else: - fp_log = open(logname,'w') - return - -def lodaMsgInfos(jsonDir,msgId): - jsonUrl = os.path.join(jsonDir,msgId+'.json') - with open(jsonUrl,'r') as fp: - data=json.load(fp) - return data - -def parse_filename_for_oss(name): - splts=name.split('_') - typename=splts[2].split('-')[1].strip() - msgId=splts[4].split('-')[3] - onLineType=splts[4].split('-')[1] - return typename,msgId,onLineType -def percentage(consumed_bytes, total_bytes): - if total_bytes: - rate = int(100 * (float(consumed_bytes) / float(total_bytes))) - print('\r{0}% '.format(rate), end='') - sys.stdout.flush() - \ No newline at end of file diff --git a/AIlib/voduploadsdk/AliyunVodUploader.py b/AIlib/voduploadsdk/AliyunVodUploader.py deleted file mode 100644 index c0356c9..0000000 --- a/AIlib/voduploadsdk/AliyunVodUploader.py +++ /dev/null @@ -1,670 +0,0 @@ -# -*- coding: UTF-8 -*- -import json -import oss2 -import base64 -import requests -from oss2 import compat -import time - -from aliyunsdkcore import client -from aliyunsdkvod.request.v20170321 import CreateUploadVideoRequest -from aliyunsdkvod.request.v20170321 import RefreshUploadVideoRequest -from aliyunsdkvod.request.v20170321 import CreateUploadImageRequest -from aliyunsdkvod.request.v20170321 import CreateUploadAttachedMediaRequest -from voduploadsdk.AliyunVodUtils import * -from voduploadsdk.UploadVideoRequest import UploadVideoRequest - -VOD_MAX_TITLE_LENGTH = 128 -VOD_MAX_DESCRIPTION_LENGTH = 1024 - -class AliyunVodUploader: - - def __init__(self, accessKeyId, accessKeySecret, ecsRegionId=None): - """ - constructor for VodUpload - :param accessKeyId: string, access key id - :param accessKeySecret: string, access key secret - :param ecsRegion: string, 部署迁移脚本的ECS所在的Region，详细参考：https://help.aliyun.com/document_detail/40654.html，如：cn-beijing - :return - """ - self.__accessKeyId = accessKeyId - self.__accessKeySecret = accessKeySecret - self.__ecsRegion = ecsRegionId - self.__vodApiRegion = None - self.__connTimeout = 3 - self.__bucketClient = None - self.__maxRetryTimes = 3 - self.__vodClient = None - self.__EnableCrc = True - - # 分片上传参数 - self.__multipartThreshold = 10 * 1024 * 1024 # 分片上传的阈值，超过此值开启分片上传 - self.__multipartPartSize = 10 * 1024 * 1024 # 分片大小，单位byte - self.__multipartThreadsNum = 3 # 分片上传时并行上传的线程数，暂时为串行上传，不支持并行，后续会支持。 - - self.setApiRegion('cn-shanghai') - - - def setApiRegion(self, apiRegion): - """ - 设置VoD的接入地址，中国大陆为cn-shanghai，海外支持ap-southeast-1(新加坡)等区域，详情参考：https://help.aliyun.com/document_detail/98194.html - :param apiRegion: 接入地址的Region英文表示 - :return: - """ - self.__vodApiRegion = apiRegion - self.__vodClient = self.__initVodClient() - - - def setMultipartUpload(self, multipartThreshold=10*1024*1024, multipartPartSize=10*1024*1024, multipartThreadsNum=1): - if multipartThreshold > 0: - self.__multipartThreshold = multipartThreshold - if multipartPartSize > 0: - self.__multipartPartSize = multipartPartSize - if multipartThreadsNum > 0: - self.__multipartThreadsNum = multipartThreadsNum - - def setEnableCrc(self, isEnable=False): - self.__EnableCrc = True if isEnable else False - - @catch_error - def uploadLocalVideo(self, uploadVideoRequest, startUploadCallback=None): - """ - 上传本地视频或音频文件到点播，最大支持48.8TB的单个文件，暂不支持断点续传 - :param uploadVideoRequest: UploadVideoRequest类的实例，注意filePath为本地文件的绝对路径 - :param startUploadCallback为获取到上传地址和凭证(uploadInfo)后开始进行文件上传时的回调，可用于记录上传日志等；uploadId为设置的上传ID，可用于关联导入视频。 - :return - """ - uploadInfo = self.__createUploadVideo(uploadVideoRequest) - if startUploadCallback: - startUploadCallback(uploadVideoRequest.uploadId, uploadInfo) - headers = self.__getUploadHeaders(uploadVideoRequest) - self.__uploadOssObjectWithRetry(uploadVideoRequest.filePath, uploadInfo['UploadAddress']['FileName'], uploadInfo, headers) - return uploadInfo - - @catch_error - def uploadWebVideo(self, uploadVideoRequest, startUploadCallback=None): - """ - 上传网络视频或音频文件到点播，最大支持48.8TB的单个文件(需本地磁盘空间足够)；会先下载到本地临时目录，再上传到点播存储 - :param uploadVideoRequest: UploadVideoRequest类的实例，注意filePath为网络文件的URL地址 - :return - """ - # 下载文件 - uploadVideoRequest = self.__downloadWebMedia(uploadVideoRequest) - - # 上传到点播 - uploadInfo = self.__createUploadVideo(uploadVideoRequest) - if startUploadCallback: - startUploadCallback(uploadVideoRequest.uploadId, uploadInfo) - headers = self.__getUploadHeaders(uploadVideoRequest) - self.__uploadOssObjectWithRetry(uploadVideoRequest.filePath, uploadInfo['UploadAddress']['FileName'], uploadInfo, headers) - - # 删除本地临时文件 - os.remove(uploadVideoRequest.filePath) - - return uploadInfo['VideoId'] - - @catch_error - def uploadLocalM3u8(self, uploadVideoRequest, sliceFilePaths=None): - """ - 上传本地m3u8视频或音频文件到点播，m3u8文件和分片文件默认在同一目录 - :param uploadVideoRequest: UploadVideoRequest类的实例，注意filePath为本地m3u8索引文件的绝对路径， - 且m3u8文件的分片信息必须是相对地址，不能含有URL或本地绝对路径 - :param sliceFilePaths: list, 分片文件的本地路径列表，例如：['/opt/m3u8_video/sample_001.ts', '/opt/m3u8_video/sample_002.ts'] - sliceFilePaths为None时，会按照同一目录去解析分片地址；如不在同一目录等原因导致解析有误，可自行组装分片地址 - :return - """ - - if sliceFilePaths is None: - sliceFilePaths = self.parseLocalM3u8(uploadVideoRequest.filePath) - - if (not isinstance(sliceFilePaths, list)) or len(sliceFilePaths) <= 0: - raise AliyunVodException('InvalidM3u8SliceFile', 'M3u8 slice files invalid', 'sliceFilePaths invalid or m3u8 index file error') - - # 上传到点播的m3u8索引文件会重写，以此确保分片地址都为相对地址 - downloader = AliyunVodDownloader() - m3u8LocalDir = downloader.getSaveLocalDir() + '/' + AliyunVodUtils.getStringMd5(uploadVideoRequest.fileName) - downloader.setSaveLocalDir(m3u8LocalDir) - m3u8LocalPath = m3u8LocalDir + '/' + os.path.basename(uploadVideoRequest.fileName) - self.__rewriteM3u8File(uploadVideoRequest.filePath, m3u8LocalPath, True) - - # 获取上传凭证 - uploadVideoRequest.setFilePath(m3u8LocalPath) - uploadInfo = self.__createUploadVideo(uploadVideoRequest) - uploadAddress = uploadInfo['UploadAddress'] - headers = self.__getUploadHeaders(uploadVideoRequest) - - # 依次上传分片文件 - for sliceFilePath in sliceFilePaths: - tempFilePath, sliceFileName = AliyunVodUtils.getFileBriefPath(sliceFilePath) - self.__uploadOssObjectWithRetry(sliceFilePath, uploadAddress['ObjectPrefix'] + sliceFileName, uploadInfo, headers) - - # 上传m3u8文件 - self.__uploadOssObjectWithRetry(m3u8LocalPath, uploadAddress['FileName'], uploadInfo, headers) - - # 删除重写到本地的m3u8文件 - if os.path.exists(m3u8LocalPath): - os.remove(m3u8LocalPath) - if not os.listdir(m3u8LocalDir): - os.rmdir(m3u8LocalDir) - - return uploadInfo['VideoId'] - - @catch_error - def uploadWebM3u8(self, uploadVideoRequest, sliceFileUrls=None): - """ - 上传网络m3u8视频或音频文件到点播，需本地磁盘空间足够，会先下载到本地临时目录，再上传到点播存储 - :param uploadVideoRequest: UploadVideoRequest类的实例，注意filePath为m3u8网络文件的URL地址 - :param sliceFileUrls: list, 分片文件的url，例如：['http://host/sample_001.ts', 'http://host/sample_002.ts'] - sliceFileUrls为None时，会按照同一前缀解析分片地址；如分片路径和m3u8索引文件前缀不同等原因导致解析有误，可自行组装分片地址 - :return - """ - if sliceFileUrls is None: - sliceFileUrls = self.parseWebM3u8(uploadVideoRequest.filePath) - - if (not isinstance(sliceFileUrls, list)) or len(sliceFileUrls) <= 0: - raise AliyunVodException('InvalidM3u8SliceFile', 'M3u8 slice urls invalid', - 'sliceFileUrls invalid or m3u8 index file error') - - # 下载m3u8文件和所有ts分片文件到本地；上传到点播的m3u8索引文件会重写，以此确保分片地址都为相对地址 - downloader = AliyunVodDownloader() - m3u8LocalDir = downloader.getSaveLocalDir() + '/' + AliyunVodUtils.getStringMd5(uploadVideoRequest.fileName) - downloader.setSaveLocalDir(m3u8LocalDir) - m3u8LocalPath = m3u8LocalDir + '/' + os.path.basename(uploadVideoRequest.fileName) - self.__rewriteM3u8File(uploadVideoRequest.filePath, m3u8LocalPath, False) - - sliceList = [] - for sliceFileUrl in sliceFileUrls: - tempFilePath, sliceFileName = AliyunVodUtils.getFileBriefPath(sliceFileUrl) - err, sliceLocalPath = downloader.downloadFile(sliceFileUrl, sliceFileName) - if sliceLocalPath is None: - raise AliyunVodException('FileDownloadError', 'Download M3u8 File Error', '') - sliceList.append((sliceLocalPath, sliceFileName)) - - # 获取上传凭证 - uploadVideoRequest.setFilePath(m3u8LocalPath) - uploadInfo = self.__createUploadVideo(uploadVideoRequest) - uploadAddress = uploadInfo['UploadAddress'] - headers = self.__getUploadHeaders(uploadVideoRequest) - - # 依次上传分片文件 - for sliceFile in sliceList: - self.__uploadOssObjectWithRetry(sliceFile[0], uploadAddress['ObjectPrefix'] + sliceFile[1], uploadInfo, headers) - - # 上传m3u8文件 - self.__uploadOssObjectWithRetry(m3u8LocalPath, uploadAddress['FileName'], uploadInfo, headers) - - # 删除下载到本地的m3u8文件和分片文件 - if os.path.exists(m3u8LocalPath): - os.remove(m3u8LocalPath) - for sliceFile in sliceList: - if os.path.exists(sliceFile[0]): - os.remove(sliceFile[0]) - if not os.listdir(m3u8LocalDir): - os.rmdir(m3u8LocalDir) - - return uploadInfo['VideoId'] - - - @catch_error - def uploadImage(self, uploadImageRequest, isLocalFile=True): - """ - 上传图片文件到点播，不支持断点续传；该接口可支持上传本地图片或网络图片 - :param uploadImageRequest: UploadImageRequest，注意filePath为本地文件的绝对路径或网络文件的URL地址 - :param isLocalFile: bool, 是否为本地文件。True：本地文件，False：网络文件 - :return - """ - # 网络图片需要先下载到本地 - if not isLocalFile: - uploadImageRequest = self.__downloadWebMedia(uploadImageRequest) - - # 上传到点播 - uploadInfo = self.__createUploadImage(uploadImageRequest) - self.__uploadOssObject(uploadImageRequest.filePath, uploadInfo['UploadAddress']['FileName'], uploadInfo, None) - - # 删除本地临时文件 - if not isLocalFile: - os.remove(uploadImageRequest.filePath) - - return uploadInfo['ImageId'], uploadInfo['ImageURL'] - - @catch_error - def uploadAttachedMedia(self, uploadAttachedRequest, isLocalFile=True): - """ - 上传辅助媒资文件(如水印、字幕文件)到点播，不支持断点续传；该接口可支持上传本地或网络文件 - :param uploadAttachedRequest: UploadAttachedMediaRequest，注意filePath为本地文件的绝对路径或网络文件的URL地址 - :param isLocalFile: bool, 是否为本地文件。True：本地文件，False：网络文件 - :return - """ - # 网络文件需要先下载到本地 - if not isLocalFile: - uploadAttachedRequest = self.__downloadWebMedia(uploadAttachedRequest) - - # 上传到点播 - uploadInfo = self.__createUploadAttachedMedia(uploadAttachedRequest) - self.__uploadOssObject(uploadAttachedRequest.filePath, uploadInfo['UploadAddress']['FileName'], uploadInfo, None) - - # 删除本地临时文件 - if not isLocalFile: - os.remove(uploadAttachedRequest.filePath) - - result = {'MediaId': uploadInfo['MediaId'], 'MediaURL': uploadInfo['MediaURL'], 'FileURL': uploadInfo['FileURL']} - return result - - @catch_error - def parseWebM3u8(self, m3u8FileUrl): - """ - 解析网络m3u8文件得到所有分片文件地址，原理是将m3u8地址前缀拼接ts分片名称作为后者的下载url，适用于url不带签名或分片与m3u8文件签名相同的情况 - 本函数解析时会默认分片文件和m3u8文件位于同一目录，如不是则请自行拼接分片文件的地址列表 - :param m3u8FileUrl: string, m3u8网络文件url，例如：http://host/sample.m3u8 - :return sliceFileUrls - """ - sliceFileUrls = [] - res = requests.get(m3u8FileUrl) - res.raise_for_status() - for line in res.iter_lines(): - if line.startswith('#'): - continue - sliceFileUrl = AliyunVodUtils.replaceFileName(m3u8FileUrl, line.strip()) - sliceFileUrls.append(sliceFileUrl) - - return sliceFileUrls - - @catch_error - def parseLocalM3u8(self, m3u8FilePath): - """ - 解析本地m3u8文件得到所有分片文件地址，原理是将m3u8地址前缀拼接ts分片名称作为后者的本地路径 - 本函数解析时会默认分片文件和m3u8文件位于同一目录，如不是则请自行拼接分片文件的地址列表 - :param m3u8FilePath: string, m3u8本地文件路径，例如：/opt/videos/sample.m3u8 - :return sliceFilePaths - """ - sliceFilePaths = [] - m3u8FilePath = AliyunVodUtils.toUnicode(m3u8FilePath) - for line in open(m3u8FilePath): - if line.startswith('#'): - continue - sliceFileName = line.strip() - sliceFilePath = AliyunVodUtils.replaceFileName(m3u8FilePath, sliceFileName) - sliceFilePaths.append(sliceFilePath) - - return sliceFilePaths - - - # 定义进度条回调函数；consumedBytes: 已经上传的数据量，totalBytes：总数据量 - def uploadProgressCallback(self, consumedBytes, totalBytes): - - if totalBytes: - rate = int(100 * (float(consumedBytes) / float(totalBytes))) - else: - rate = 0 - - print ("[%s]uploaded %s bytes, percent %s%s" % (AliyunVodUtils.getCurrentTimeStr(), consumedBytes, format(rate), '%')) - sys.stdout.flush() - - - def __initVodClient(self): - return client.AcsClient(self.__accessKeyId, self.__accessKeySecret, self.__vodApiRegion, - auto_retry=True, max_retry_time=self.__maxRetryTimes, timeout=self.__connTimeout) - - def __downloadWebMedia(self, request): - - # 下载媒体文件到本地临时目录 - downloader = AliyunVodDownloader() - localFileName = "%s.%s" % (AliyunVodUtils.getStringMd5(request.fileName), request.mediaExt) - fileUrl = request.filePath - err, localFilePath = downloader.downloadFile(fileUrl, localFileName) - if err < 0: - raise AliyunVodException('FileDownloadError', 'Download File Error', '') - - # 重新设置上传请求对象 - request.setFilePath(localFilePath) - return request - - def __rewriteM3u8File(self, srcM3u8File, dstM3u8File, isSrcLocal=True): - newM3u8Text = '' - if isSrcLocal: - for line in open(AliyunVodUtils.toUnicode(srcM3u8File)): - item = self.__processM3u8Line(line) - if item is not None: - newM3u8Text += item + "\n" - else: - res = requests.get(srcM3u8File) - res.raise_for_status() - for line in res.iter_lines(): - item = self.__processM3u8Line(line) - if item is not None: - newM3u8Text += item + "\n" - - AliyunVodUtils.mkDir(dstM3u8File) - with open(dstM3u8File, 'w') as f: - f.write(newM3u8Text) - - - def __processM3u8Line(self, line): - item = line.strip() - if len(item) <= 0: - return None - - if item.startswith('#'): - return item - - tempFilePath, fileName = AliyunVodUtils.getFileBriefPath(item) - return fileName - - - def __requestUploadInfo(self, request, mediaType): - request.set_accept_format('JSON') - result = json.loads(self.__vodClient.do_action_with_exception(request).decode('utf-8')) - result['OriUploadAddress'] = result['UploadAddress'] - result['OriUploadAuth'] = result['UploadAuth'] - - result['UploadAddress'] = json.loads(base64.b64decode(result['OriUploadAddress']).decode('utf-8')) - result['UploadAuth'] = json.loads(base64.b64decode(result['OriUploadAuth']).decode('utf-8')) - - result['MediaType'] = mediaType - if mediaType == 'video': - result['MediaId'] = result['VideoId'] - elif mediaType == 'image': - result['MediaId'] = result['ImageId'] - result['MediaURL'] = result['ImageURL'] - - return result - - - # 获取视频上传地址和凭证 - def __createUploadVideo(self, uploadVideoRequest): - request = CreateUploadVideoRequest.CreateUploadVideoRequest() - - title = AliyunVodUtils.subString(uploadVideoRequest.title, VOD_MAX_TITLE_LENGTH) - request.set_Title(title) - request.set_FileName(uploadVideoRequest.fileName) - - if uploadVideoRequest.description: - description = AliyunVodUtils.subString(uploadVideoRequest.description, VOD_MAX_DESCRIPTION_LENGTH) - request.set_Description(description) - if uploadVideoRequest.coverURL: - request.set_CoverURL(uploadVideoRequest.coverURL) - if uploadVideoRequest.tags: - request.set_Tags(uploadVideoRequest.tags) - if uploadVideoRequest.cateId: - request.set_CateId(uploadVideoRequest.cateId) - if uploadVideoRequest.templateGroupId: - request.set_TemplateGroupId(uploadVideoRequest.templateGroupId) - if uploadVideoRequest.storageLocation: - request.set_StorageLocation(uploadVideoRequest.storageLocation) - if uploadVideoRequest.userData: - request.set_UserData(uploadVideoRequest.userData) - if uploadVideoRequest.appId: - request.set_AppId(uploadVideoRequest.appId) - if uploadVideoRequest.workflowId: - request.set_WorkflowId(uploadVideoRequest.workflowId) - - result = self.__requestUploadInfo(request, 'video') - logger.info("CreateUploadVideo, FilePath: %s, VideoId: %s" % (uploadVideoRequest.filePath, result['VideoId'])) - return result - - # 刷新上传凭证 - def __refresh_upload_video(self, videoId): - request = RefreshUploadVideoRequest.RefreshUploadVideoRequest(); - request.set_VideoId(videoId) - - result = self.__requestUploadInfo(request, 'video') - logger.info("RefreshUploadVideo, VideoId %s" % (result['VideoId'])) - return result - - # 获取图片上传地址和凭证 - def __createUploadImage(self, uploadImageRequest): - request = CreateUploadImageRequest.CreateUploadImageRequest() - - request.set_ImageType(uploadImageRequest.imageType) - request.set_ImageExt(uploadImageRequest.imageExt) - if uploadImageRequest.title: - title = AliyunVodUtils.subString(uploadImageRequest.title, VOD_MAX_TITLE_LENGTH) - request.set_Title(title) - if uploadImageRequest.description: - description = AliyunVodUtils.subString(uploadImageRequest.description, VOD_MAX_DESCRIPTION_LENGTH) - request.set_Description(description) - if uploadImageRequest.tags: - request.set_Tags(uploadImageRequest.tags) - if uploadImageRequest.cateId: - request.set_CateId(uploadImageRequest.cateId) - if uploadImageRequest.storageLocation: - request.set_StorageLocation(uploadImageRequest.storageLocation) - if uploadImageRequest.userData: - request.set_UserData(uploadImageRequest.userData) - if uploadImageRequest.appId: - request.set_AppId(uploadImageRequest.appId) - if uploadImageRequest.workflowId: - request.set_WorkflowId(uploadImageRequest.workflowId) - - result = self.__requestUploadInfo(request, 'image') - logger.info("CreateUploadImage, FilePath: %s, ImageId: %s, ImageUrl: %s" % ( - uploadImageRequest.filePath, result['ImageId'], result['ImageURL'])) - return result - - def __createUploadAttachedMedia(self, uploadAttachedRequest): - request = CreateUploadAttachedMediaRequest.CreateUploadAttachedMediaRequest() - request.set_BusinessType(uploadAttachedRequest.businessType) - request.set_MediaExt(uploadAttachedRequest.mediaExt) - - if uploadAttachedRequest.title: - title = AliyunVodUtils.subString(uploadAttachedRequest.title, VOD_MAX_TITLE_LENGTH) - request.set_Title(title) - if uploadAttachedRequest.description: - description = AliyunVodUtils.subString(uploadAttachedRequest.description, VOD_MAX_DESCRIPTION_LENGTH) - request.set_Description(description) - if uploadAttachedRequest.tags: - request.set_Tags(uploadAttachedRequest.tags) - if uploadAttachedRequest.cateId: - request.set_CateId(uploadAttachedRequest.cateId) - if uploadAttachedRequest.storageLocation: - request.set_StorageLocation(uploadAttachedRequest.storageLocation) - if uploadAttachedRequest.userData: - request.set_UserData(uploadAttachedRequest.userData) - if uploadAttachedRequest.appId: - request.set_AppId(uploadAttachedRequest.appId) - if uploadAttachedRequest.workflowId: - request.set_WorkflowId(uploadAttachedRequest.workflowId) - - result = self.__requestUploadInfo(request, 'attached') - logger.info("CreateUploadImage, FilePath: %s, MediaId: %s, MediaURL: %s" % ( - uploadAttachedRequest.filePath, result['MediaId'], result['MediaURL'])) - return result - - - def __getUploadHeaders(self, uploadVideoRequest): - if uploadVideoRequest.isShowWatermark is None: - return None - else: - userData = "{\"Vod\":{\"UserData\":{\"IsShowWaterMark\": \"%s\"}}}" % (uploadVideoRequest.isShowWatermark) - return {'x-oss-notification': base64.b64encode(userData, 'utf-8')} - - # uploadType，可选：multipart, put, web - def __uploadOssObjectWithRetry(self, filePath, object, uploadInfo, headers=None): - retryTimes = 0 - while retryTimes < self.__maxRetryTimes: - try: - return self.__uploadOssObject(filePath, object, uploadInfo, headers) - except OssError as e: - # 上传凭证过期需要重新获取凭证 - if e.code == 'SecurityTokenExpired' or e.code == 'InvalidAccessKeyId': - uploadInfo = self.__refresh_upload_video(uploadInfo['MediaId']) - except Exception as e: - raise e - except: - raise AliyunVodException('UnkownError', repr(e), traceback.format_exc()) - finally: - retryTimes += 1 - - - def __uploadOssObject(self, filePath, object, uploadInfo, headers=None): - self.__createOssClient(uploadInfo['UploadAuth'], uploadInfo['UploadAddress']) - """ - p = os.path.dirname(os.path.realpath(__file__)) - store = os.path.dirname(p) + '/osstmp' - return oss2.resumable_upload(self.__bucketClient, object, filePath, - store=oss2.ResumableStore(root=store), headers=headers, - multipart_threshold=self.__multipartThreshold, part_size=self.__multipartPartSize, - num_threads=self.__multipartThreadsNum, progress_callback=self.uploadProgressCallback) - """ - uploader = _VodResumableUploader(self.__bucketClient, filePath, object, uploadInfo, headers, - self.uploadProgressCallback, self.__refreshUploadAuth) - uploader.setMultipartInfo(self.__multipartThreshold, self.__multipartPartSize, self.__multipartThreadsNum) - uploader.setClientId(self.__accessKeyId) - res = uploader.upload() - - uploadAddress = uploadInfo['UploadAddress'] - bucketHost = uploadAddress['Endpoint'].replace('://', '://' + uploadAddress['Bucket'] + ".") - logger.info("UploadFile %s Finish, MediaId: %s, FilePath: %s, Destination: %s/%s" % ( - uploadInfo['MediaType'], uploadInfo['MediaId'], filePath, bucketHost, object)) - return res - - # 使用上传凭证和地址信息初始化OSS客户端（注意需要先Base64解码并Json Decode再传入） - # 如果上传的ECS位于点播相同的存储区域（如上海），则可以指定internal为True，通过内网上传更快且免费 - def __createOssClient(self, uploadAuth, uploadAddress): - auth = oss2.StsAuth(uploadAuth['AccessKeyId'], uploadAuth['AccessKeySecret'], uploadAuth['SecurityToken']) - endpoint = AliyunVodUtils.convertOssInternal(uploadAddress['Endpoint'], self.__ecsRegion) - self.__bucketClient = oss2.Bucket(auth, endpoint, uploadAddress['Bucket'], - connect_timeout=self.__connTimeout, enable_crc=self.__EnableCrc) - return self.__bucketClient - - def __refreshUploadAuth(self, videoId): - uploadInfo = self.__refresh_upload_video(videoId) - uploadAuth = uploadInfo['UploadAuth'] - uploadAddress = uploadInfo['UploadAddress'] - return self.__createOssClient(uploadAuth, uploadAddress) - - -from oss2 import SizedFileAdapter, determine_part_size -from oss2.models import PartInfo -from aliyunsdkcore.utils import parameter_helper as helper -class _VodResumableUploader: - def __init__(self, bucket, filePath, object, uploadInfo, headers, progressCallback, refreshAuthCallback): - self.__bucket = bucket - self.__filePath = filePath - self.__object = object - self.__uploadInfo = uploadInfo - self.__totalSize = None - self.__headers = headers - self.__mtime = os.path.getmtime(filePath) - self.__progressCallback = progressCallback - self.__refreshAuthCallback = refreshAuthCallback - - self.__threshold = None - self.__partSize = None - self.__threadsNum = None - self.__uploadId = 0 - - self.__record = {} - self.__finishedSize = 0 - self.__finishedParts = [] - self.__filePartHash = None - self.__clientId = None - - def setMultipartInfo(self, threshold, partSize, threadsNum): - self.__threshold = threshold - self.__partSize = partSize - self.__threadsNum = threadsNum - - - def setClientId(self, clientId): - self.__clientId = clientId - - - def upload(self): - self.__totalSize = os.path.getsize(self.__filePath) - if self.__threshold and self.__totalSize <= self.__threshold: - return self.simpleUpload() - else: - return self.multipartUpload() - - - def simpleUpload(self): - with open(AliyunVodUtils.toUnicode(self.__filePath), 'rb') as f: - result = self.__bucket.put_object(self.__object, f, headers=self.__headers, progress_callback=None) - if self.__uploadInfo['MediaType'] == 'video': - self.__reportUploadProgress('put', 1, self.__totalSize) - - return result - - def multipartUpload(self): - psize = oss2.determine_part_size(self.__totalSize, preferred_size=self.__partSize) - - # 初始化分片 - self.__uploadId = self.__bucket.init_multipart_upload(self.__object).upload_id - - startTime = time.time() - expireSeconds = 2500 # 上传凭证有效期3000秒，提前刷新 - # 逐个上传分片 - with open(AliyunVodUtils.toUnicode(self.__filePath), 'rb') as fileObj: - partNumber = 1 - offset = 0 - - while offset < self.__totalSize: - uploadSize = min(psize, self.__totalSize - offset) - #logger.info("UploadPart, FilePath: %s, VideoId: %s, UploadId: %s, PartNumber: %s, PartSize: %s" % (self.__fileName, self.__videoId, self.__uploadId, partNumber, uploadSize)) - result = self.__bucket.upload_part(self.__object, self.__uploadId, partNumber, SizedFileAdapter(fileObj,uploadSize)) - #print(result.request_id) - self.__finishedParts.append(PartInfo(partNumber, result.etag)) - offset += uploadSize - partNumber += 1 - - # 上传进度回调 - self.__progressCallback(offset, self.__totalSize) - - if self.__uploadInfo['MediaType'] == 'video': - # 上报上传进度 - self.__reportUploadProgress('multipart', partNumber - 1, offset) - - # 检测上传凭证是否过期 - nowTime = time.time() - if nowTime - startTime >= expireSeconds: - self.__bucket = self.__refreshAuthCallback(self.__uploadInfo['MediaId']) - startTime = nowTime - - - # 完成分片上传 - self.__bucket.complete_multipart_upload(self.__object, self.__uploadId, self.__finishedParts, headers=self.__headers) - - return result - - - def __reportUploadProgress(self, uploadMethod, donePartsCount, doneBytes): - reportHost = 'vod.cn-shanghai.aliyuncs.com' - sdkVersion = '1.3.1' - reportKey = 'HBL9nnSwhtU2$STX' - - uploadPoint = {'upMethod': uploadMethod, 'partSize': self.__partSize, 'doneBytes': doneBytes} - timestamp = int(time.time()) - authInfo = AliyunVodUtils.getStringMd5("%s|%s|%s" % (self.__clientId, reportKey, timestamp)) - - fields = {'Action': 'ReportUploadProgress', 'Format': 'JSON', 'Version': '2017-03-21', - 'Timestamp': helper.get_iso_8061_date(), 'SignatureNonce': helper.get_uuid(), - 'VideoId': self.__uploadInfo['MediaId'], 'Source': 'PythonSDK', 'ClientId': self.__clientId, - 'BusinessType': 'UploadVideo', 'TerminalType': 'PC', 'DeviceModel': 'Server', - 'AppVersion': sdkVersion, 'AuthTimestamp': timestamp, 'AuthInfo': authInfo, - - 'FileName': self.__filePath, 'FileHash': self.__getFilePartHash(self.__clientId, self.__filePath, self.__totalSize), - 'FileSize': self.__totalSize, 'FileCreateTime': timestamp, 'UploadRatio': 0, 'UploadId': self.__uploadId, - 'DonePartsCount': donePartsCount, 'PartSize': self.__partSize, 'UploadPoint': json.dumps(uploadPoint), - 'UploadAddress': self.__uploadInfo['OriUploadAddress'] - } - requests.post('http://' + reportHost, fields, timeout=1) - - - def __getFilePartHash(self, clientId, filePath, fileSize): - if self.__filePartHash: - return self.__filePartHash - - length = 1 * 1024 * 1024 - if fileSize < length: - length = fileSize - - try: - fp = open(AliyunVodUtils.toUnicode(filePath), 'rb') - strVal = fp.read(length) - self.__filePartHash = AliyunVodUtils.getStringMd5(strVal, False) - fp.close() - except: - self.__filePartHash = "%s|%s|%s" % (clientId, filePath, self.__mtime) - - return self.__filePartHash diff --git a/AIlib/voduploadsdk/AliyunVodUtils.py b/AIlib/voduploadsdk/AliyunVodUtils.py deleted file mode 100644 index 5a477b9..0000000 --- a/AIlib/voduploadsdk/AliyunVodUtils.py +++ /dev/null @@ -1,325 +0,0 @@ -# -*- coding: UTF-8 -*- -import os,sys -import hashlib -import datetime -import functools -import logging -from oss2.exceptions import OssError -from aliyunsdkcore.acs_exception.exceptions import ServerException -from aliyunsdkcore.acs_exception.exceptions import ClientException -import traceback -import requests - -if sys.version_info[0] == 3: - import urllib.parse -else: - from urllib import unquote - - -VOD_PRINT_INFO_LOG_SWITCH = 1 - -class AliyunVodLog: - """ - VOD日志类，基于logging实现 - """ - @staticmethod - def printLogStr(msg, *args, **kwargs): - if VOD_PRINT_INFO_LOG_SWITCH: - print("[%s]%s" % (AliyunVodUtils.getCurrentTimeStr(), msg)) - - @staticmethod - def info(msg, *args, **kwargs): - logging.info(msg, *args, **kwargs) - AliyunVodLog.printLogStr(msg, *args, **kwargs) - - @staticmethod - def error(msg, *args, **kwargs): - logging.error(msg, *args, **kwargs) - AliyunVodLog.printLogStr(msg, *args, **kwargs) - - @staticmethod - def warning(msg, *args, **kwargs): - logging.warning(msg, *args, **kwargs) - AliyunVodLog.printLogStr(msg, *args, **kwargs) - -logger = AliyunVodLog - - -class AliyunVodUtils: - """ - VOD上传SDK的工具类，提供截取字符串、获取扩展名、获取文件名等静态函数 - """ - - # 截取字符串，在不超过最大字节数前提下确保中文字符不被截断出现乱码（先转换成unicode，再取子串，然后转换成utf-8） - @staticmethod - def subString(strVal, maxBytes, charSet='utf-8'): - i = maxBytes - if sys.version_info[0] == 3: - while len(strVal.encode(charSet)) > maxBytes: - if i < 0: - return '' - strVal = strVal[:i] - i -= 1 - else: - while len(strVal) > maxBytes: - if i < 0: - return '' - strVal = strVal.decode(charSet)[:i].encode(charSet) - i -= 1 - return strVal - - @staticmethod - def getFileExtension(fileName): - end = fileName.rfind('?') - if end <= 0: - end = len(fileName) - - i = fileName.rfind('.') - if i >= 0: - return fileName[i+1:end].lower() - else: - return None - - # urldecode - @staticmethod - def urlDecode(fileUrl): - if sys.version_info[0] == 3: - return urllib.parse.unquote(fileUrl) - else: - return unquote(fileUrl) - - # urlencode - @staticmethod - def urlEncode(fileUrl): - if sys.version_info[0] == 3: - return urllib.parse.urlencode(fileUrl) - else: - return urllib.urlencode(fileUrl) - - # 获取Url的摘要地址（去除?后的参数，如果有）以及文件名 - @staticmethod - def getFileBriefPath(fileUrl): - #fileUrl = AliyunVodUtils.urlDecode(fileUrl) - i = fileUrl.rfind('?') - if i > 0: - briefPath = fileUrl[:i] - else: - briefPath = fileUrl - - briefName = os.path.basename(briefPath) - return briefPath, AliyunVodUtils.urlDecode(briefName) - - @staticmethod - def getStringMd5(strVal, isEncode=True): - m = hashlib.md5() - m.update(strVal.encode('utf-8') if isEncode else strVal) - return m.hexdigest() - - @staticmethod - def getCurrentTimeStr(): - now = datetime.datetime.now() - return now.strftime("%Y-%m-%d %H:%M:%S") - - # 将oss地址转换为内网地址（如果脚本部署的ecs与oss bucket在同一区域） - @staticmethod - def convertOssInternal(ossUrl, ecsRegion=None, isVpc=False): - if (not ossUrl) or (not ecsRegion): - return ossUrl - - availableRegions = ['cn-qingdao', 'cn-beijing', 'cn-zhangjiakou', 'cn-huhehaote', 'cn-hangzhou', 'cn-shanghai', 'cn-shenzhen', - 'cn-hongkong', 'ap-southeast-1', 'ap-southeast-2', 'ap-southeast-3', - 'ap-northeast-1', 'us-west-1', 'us-east-1', 'eu-central-1', 'me-east-1'] - if ecsRegion not in availableRegions: - return ossUrl - - ossUrl = ossUrl.replace("https:", "http:") - if isVpc: - return ossUrl.replace("oss-%s.aliyuncs.com" % (ecsRegion), "vpc100-oss-%s.aliyuncs.com" % (ecsRegion)) - else: - return ossUrl.replace("oss-%s.aliyuncs.com" % (ecsRegion), "oss-%s-internal.aliyuncs.com" % (ecsRegion)) - - # 把输入转换为unicode - @staticmethod - def toUnicode(data): - if isinstance(data, bytes): - return data.decode('utf-8') - else: - return data - - # 替换路径中的文件名；考虑分隔符为"/" 或 "\"(windows) - @staticmethod - def replaceFileName(filePath, replace): - if len(filePath) <= 0 or len(replace) <= 0: - return filePath - - filePath = AliyunVodUtils.urlDecode(filePath) - separator = '/' - start = filePath.rfind(separator) - if start < 0: - separator = '\\' - start = filePath.rfind(separator) - if start < 0: - return None - - result = "%s%s%s" % (filePath[0:start], separator, replace) - return result - - # 创建文件中的目录 - @staticmethod - def mkDir(filePath): - if len(filePath) <= 0: - return -1 - - separator = '/' - i = filePath.rfind(separator) - if i < 0: - separator = '\\' - i = filePath.rfind(separator) - if i < 0: - return -2 - - dirs = filePath[:i] - if os.path.exists(dirs) and os.path.isdir(dirs): - return 0 - - os.makedirs(dirs) - return 1 - - - -class AliyunVodException(Exception): - """ - VOD上传SDK的异常类，做统一的异常处理，外部捕获此异常即可 - """ - - def __init__(self, type, code, msg, http_status=None, request_id=None): - Exception.__init__(self) - self.type = type or 'UnkownError' - self.code = code - self.message = msg - self.http_status = http_status or 'NULL' - self.request_id = request_id or 'NULL' - - def __str__(self): - return "Type: %s, Code: %s, Message: %s, HTTPStatus: %s, RequestId: %s" % ( - self.type, self.code, self.message, str(self.http_status), self.request_id) - -def catch_error(method): - """ - 装饰器，将内部异常转换成统一的异常类AliyunVodException - """ - - @functools.wraps(method) - def wrapper(self, *args, **kwargs): - try: - return method(self, *args, **kwargs) - except ServerException as e: - # 可能原因：AK错误、账号无权限、参数错误等 - raise AliyunVodException('ServerException', e.get_error_code(), e.get_error_msg(), e.get_http_status(), e.get_request_id()) - logger.error("ServerException: %s", e) - except ClientException as e: - # 可能原因：本地网络故障（如不能连接外网）等 - raise AliyunVodException('ClientException', e.get_error_code(), e.get_error_msg()) - logger.error("ClientException: %s", e) - except OssError as e: - # 可能原因：上传凭证过期等 - raise AliyunVodException('OssError', e.code, e.message, e.status, e.request_id) - logger.error("OssError: %s", e) - except IOError as e: - # 可能原因：文件URL不能访问、本地文件无法读取等 - raise AliyunVodException('IOError', repr(e), traceback.format_exc()) - logger.error("IOError: %s", traceback.format_exc()) - except OSError as e: - # 可能原因：本地文件不存在等 - raise AliyunVodException('OSError', repr(e), traceback.format_exc()) - logger.error("OSError: %s", traceback.format_exc()) - except AliyunVodException as e: - # 可能原因：参数错误 - raise e - logger.error("VodException: %s", e) - except Exception as e: - raise AliyunVodException('UnkownException', repr(e), traceback.format_exc()) - logger.error("UnkownException: %s", traceback.format_exc()) - except: - raise AliyunVodException('UnkownError', 'UnkownError', traceback.format_exc()) - logger.error("UnkownError: %s", traceback.format_exc()) - - return wrapper - - -class AliyunVodDownloader: - """ - VOD网络文件的下载类，上传网络文件时会先下载到本地临时目录，再上传到点播 - """ - - def __init__(self, localDir=None): - if localDir: - self.__localDir = localDir - else: - p = os.path.dirname(os.path.realpath(__file__)) - self.__localDir = os.path.dirname(p) + '/dlfiles' - - def setSaveLocalDir(self, localDir): - self.__localDir = localDir - - def getSaveLocalDir(self): - return self.__localDir - - def downloadFile(self, fileUrl, localFileName, fileSize=None): - localPath = self.__localDir + '/' + localFileName - logger.info("Download %s To %s" % (fileUrl, localPath)) - try: - lsize = self.getFileSize(localPath) - if fileSize and lsize == fileSize: - logger.info('Download OK, File Exists') - return 0, localPath - - AliyunVodUtils.mkDir(self.__localDir) - - err, webPage = self.__openWebFile(fileUrl, lsize) - if err == 0: - logger.info('Download OK, File Exists') - webPage.close() - return 0, localPath - - fileObj = open(localPath, 'ab+') - for chunk in webPage.iter_content(chunk_size=8 * 1024): - if chunk: - fileObj.write(chunk) - except Exception as e: - logger.error("Download fail: %s" % (e)) - return -1, None - - fileObj.close() - webPage.close() - logger.info('Download OK') - return 1, localPath - - def getFileSize(self, filePath): - try: - lsize = os.stat(filePath).st_size - except: - lsize = 0 - - return lsize - - - def __openWebFile(self, fileUrl, offset): - webPage = None - try: - headers = {'Range': 'bytes=%d-' % offset} - webPage = requests.get(fileUrl, stream=True, headers=headers, timeout=120, verify=False) - status_code = webPage.status_code - err = -1 - if status_code in [200, 206]: - err = 1 - elif status_code == 416: - err = 0 - else: - logger.error("Download offset %s fail, invalid url, status: %s" % (offset, status_code)) - except Exception as e: - logger.error("Download offset %s fail: %s" % (offset, e)) - err = -2 - finally: - return err, webPage - diff --git a/AIlib/voduploadsdk/ChangeLog.txt b/AIlib/voduploadsdk/ChangeLog.txt deleted file mode 100755 index 084c65b..0000000 --- a/AIlib/voduploadsdk/ChangeLog.txt +++ /dev/null @@ -1,20 +0,0 @@ -2019-04-12 Version: 1.3.1 -1. 上传时可指定应用ID，以实现多应用体系的资源隔离 -2. 支持上传时指定工作流ID，可自动化媒体处理 - -2019-02-12 Version: 1.3.0 -1. 可指定点播中心(默认为上海)和存储区域，便于海外上传 -2. 支持辅助媒资(水印、字幕文件等)的上传 -3. 支持上传时设置UserData等个性化配置和更多元数据 -4. 上传网络文件调整为先下载到本地，再上传到点播，以支持大文件上传(最大48.8TB) -5. 改进m3u8视频的上传，提供默认解析m3u8分片信息的接口，也可自定义分片列表 - -2018-07-05 Version: 1.2.1 -1. 支持设置视频存储区域UploadVideoRequest.setStorageLocation -2. 修复上传大文件时上传凭证过期未刷新的问题 - -2017-12-21 Version: 1.1.1 -1. 支持上传本地单个视频、m3u8视频(含ts分片)、图片文件到点播 -2. 支持上传网络上的(HTTP/HTTPS链接，含OSS链接)的单个视频、m3u8视频(含ts分片)、图片文件到点播. -3. 支持Python2、Python3版本 - diff --git a/AIlib/voduploadsdk/UploadAttachedMediaRequest.py b/AIlib/voduploadsdk/UploadAttachedMediaRequest.py deleted file mode 100644 index a09cfe9..0000000 --- a/AIlib/voduploadsdk/UploadAttachedMediaRequest.py +++ /dev/null @@ -1,87 +0,0 @@ -# -*- coding: UTF-8 -*- -""" - # Class UploadAttachedMediaRequest - # - # Aliyun VoD's Upload Attached Media(such as watermark,subtitle files) Request class, which wraps parameters to upload an media file into VoD. - # Users could pass parameters to AliyunVodUploader, including File Path,Title,etc. via an UploadAttachedMediaRequest instance. - # For more details, please check out the VoD API document: https://help.aliyun.com/document_detail/98467.html -""" - -from voduploadsdk.AliyunVodUtils import * -class UploadAttachedMediaRequest: - def __init__(self, filePath, businessType, title=None, fileExt=None): - """ - constructor for UploadAttachedMediaRequest - :param filePath: string, 文件的绝对路径，或者网络文件的URL，必须含有扩展名 - :return - """ - self.businessType = businessType - self.filePath = None - self.fileName = None - self.mediaExt = None - self.title = None - self.setFilePath(filePath, title, fileExt) - - self.fileSize = None - self.cateId = None - self.tags = None - self.description = None - self.userData = None - self.storageLocation = None - self.appId = None - self.workflowId = None - - - def setFilePath(self, filePath, title=None, fileExt=None): - if fileExt is None: - fileExt = AliyunVodUtils.getFileExtension(filePath) - if not fileExt: - raise AliyunVodException('ParameterError', 'InvalidParameter', 'filePath has no Extension') - - fileExt = fileExt.lstrip('.') - self.mediaExt = fileExt - self.filePath = AliyunVodUtils.toUnicode(filePath) - - briefPath, briefName = AliyunVodUtils.getFileBriefPath(self.filePath) - self.fileName = briefPath - if fileExt and (not self.fileName.endswith('.' + fileExt)): - self.fileName = self.fileName + '.' + fileExt - - if title: - self.title = title - else: - if self.title is None: - self.title = briefName - - - def setBusinessType(self, businessType): - self.businessType = businessType - - def setTitle(self, title): - self.title = title - - def setFileSize(self, fileSize): - self.fileSize = fileSize - - def setCateId(self, cateId): - self.cateId = cateId - - def setTags(self, tags): - self.tags = tags - - def setDescription(self, description): - self.description = description - - def setStorageLocation(self, storageLocation): - self.storageLocation = storageLocation - - def setUserData(self, userData): - self.userData = userData - - def setAppId(self, appId): - self.appId = appId - - def setWorkflowId(self, workflowId): - self.workflowId = workflowId - - diff --git a/AIlib/voduploadsdk/UploadImageRequest.py b/AIlib/voduploadsdk/UploadImageRequest.py deleted file mode 100644 index 73f6baa..0000000 --- a/AIlib/voduploadsdk/UploadImageRequest.py +++ /dev/null @@ -1,84 +0,0 @@ -# -*- coding: UTF-8 -*- -""" - # Class UploadImageRequest - # - # Aliyun VoD's Upload Image Request class, which wraps parameters to upload an image into VoD. - # Users could pass parameters to AliyunVodUploader, including File Path,Title,etc. via an UploadImageRequest instance. - # For more details, please check out the VoD API document: https://help.aliyun.com/document_detail/55619.html -""" - -from voduploadsdk.AliyunVodUtils import * -class UploadImageRequest: - def __init__(self, filePath, title=None, fileExt=None): - """ - constructor for UploadVideoRequest - :param filePath: string, 文件的绝对路径，或者网络文件的URL，必须含有扩展名 - :param title: string, 图片标题 - :return - """ - self.filePath = None - self.fileName = None - self.imageExt = None - self.mediaExt = None - self.title = None - self.setFilePath(filePath, title, fileExt) - - self.imageType = 'default' - self.cateId = None - self.tags = None - self.description = None - self.userData = None - self.storageLocation = None - self.appId = None - self.workflowId = None - - def setFilePath(self, filePath, title=None, fileExt=None): - if fileExt is None: - fileExt = AliyunVodUtils.getFileExtension(filePath) - if not fileExt: - raise AliyunVodException('ParameterError', 'InvalidParameter', 'filePath has no Extension') - - fileExt = fileExt.lstrip('.') - self.imageExt = fileExt - self.mediaExt = fileExt - self.filePath = AliyunVodUtils.toUnicode(filePath) - - briefPath, briefName = AliyunVodUtils.getFileBriefPath(self.filePath) - self.fileName = briefPath - - if fileExt and (not self.fileName.endswith('.' + fileExt)): - self.fileName = self.fileName + '.' + fileExt - - if title: - self.title = title - else: - if self.title is None: - self.title = briefName - - - def setImageType(self, imageType): - self.imageType = imageType - - def setTitle(self, title): - self.title = title - - def setCateId(self, cateId): - self.cateId = cateId - - def setTags(self, tags): - self.tags = tags - - def setDescription(self, description): - self.description = description - - def setStorageLocation(self, storageLocation): - self.storageLocation = storageLocation - - def setUserData(self, userData): - self.userData = userData - - def setAppId(self, appId): - self.appId = appId - - def setWorkflowId(self, workflowId): - self.workflowId = workflowId diff --git a/AIlib/voduploadsdk/UploadVideoRequest.py b/AIlib/voduploadsdk/UploadVideoRequest.py deleted file mode 100644 index afe3b35..0000000 --- a/AIlib/voduploadsdk/UploadVideoRequest.py +++ /dev/null @@ -1,86 +0,0 @@ -# -*- coding: UTF-8 -*- -""" - # Class UploadVideoRequest - # - # Aliyun VoD's Upload Video Request class, which wraps parameters to upload a video into VoD. - # Users could pass parameters to AliyunVodUploader, including File Path,Title,etc. via an UploadVideoRequest instance. - # For more details, please check out the VoD API document: https://help.aliyun.com/document_detail/55407.html -""" - -from voduploadsdk.AliyunVodUtils import * -class UploadVideoRequest: - def __init__(self, filePath, title=None, fileExt=None): - """ - constructor for UploadVideoRequest - :param filePath: string, 文件的绝对路径，或者网络文件的URL，必须含有扩展名 - :param title: string, 视频标题，最长128字节，不传则使用文件名为标题 - :return - """ - self.filePath = None - self.fileName = None - self.mediaExt = None - self.title = None - self.setFilePath(filePath, title, fileExt) - - self.cateId = None - self.tags = None - self.description = None - self.coverURL = None - self.templateGroupId = None - self.isShowWatermark = None - self.userData = None - self.storageLocation = None - self.uploadId = None - self.appId = None - self.workflowId = None - - def setFilePath(self, filePath, title=None, fileExt=None): - if fileExt is None: - fileExt = AliyunVodUtils.getFileExtension(filePath) - if not fileExt: - raise AliyunVodException('ParameterError', 'InvalidParameter', 'filePath has no Extension') - - fileExt = fileExt.lstrip('.') - self.mediaExt = fileExt - self.filePath = AliyunVodUtils.toUnicode(filePath) - - briefPath, briefName = AliyunVodUtils.getFileBriefPath(self.filePath) - self.fileName = briefPath - if fileExt and (not self.fileName.endswith('.' + fileExt)): - self.fileName = self.fileName + '.' + fileExt - - if title: - self.title = title - else: - if self.title is None: - self.title = briefName - - - def setCateId(self, cateId): - self.cateId = cateId - - def setTags(self, tags): - self.tags = tags - - def setDescription(self, description): - self.description = description - - def setCoverURL(self, coverURL): - self.coverURL = coverURL - - def setTemplateGroupId(self, templateGroupId): - self.templateGroupId = templateGroupId - - # 关闭水印，仅用于配置全局水印且转码模板开启水印后，单次上传时关闭水印 - def shutdownWatermark(self): - self.isShowWatermark = False - - # 设置上传ID，可用于关联导入视频 - def setUploadId(self, uploadId): - self.uploadId = uploadId - - def setAppId(self, appId): - self.appId = appId - - def setWorkflowId(self, workflowId): - self.workflowId = workflowId diff --git a/AIlib/voduploadsdk/__init__.py b/AIlib/voduploadsdk/__init__.py deleted file mode 100644 index 7bd766b..0000000 --- a/AIlib/voduploadsdk/__init__.py +++ /dev/null @@ -1,3 +0,0 @@ -__version__ = '1.3.1' - - diff --git a/AIlib/voduploadsdk/__pycache__/AliyunVodUploader.cpython-38.pyc b/AIlib/voduploadsdk/__pycache__/AliyunVodUploader.cpython-38.pyc deleted file mode 100644 index 0022e16..0000000 Binary files a/AIlib/voduploadsdk/__pycache__/AliyunVodUploader.cpython-38.pyc and /dev/null differ diff --git a/AIlib/voduploadsdk/__pycache__/AliyunVodUtils.cpython-38.pyc b/AIlib/voduploadsdk/__pycache__/AliyunVodUtils.cpython-38.pyc deleted file mode 100644 index 21a4b0a..0000000 Binary files a/AIlib/voduploadsdk/__pycache__/AliyunVodUtils.cpython-38.pyc and /dev/null differ diff --git a/AIlib/voduploadsdk/__pycache__/UploadAttachedMediaRequest.cpython-38.pyc b/AIlib/voduploadsdk/__pycache__/UploadAttachedMediaRequest.cpython-38.pyc deleted file mode 100644 index 1ff2ce9..0000000 Binary files a/AIlib/voduploadsdk/__pycache__/UploadAttachedMediaRequest.cpython-38.pyc and /dev/null differ diff --git a/AIlib/voduploadsdk/__pycache__/UploadImageRequest.cpython-38.pyc b/AIlib/voduploadsdk/__pycache__/UploadImageRequest.cpython-38.pyc deleted file mode 100644 index 47b84f9..0000000 Binary files a/AIlib/voduploadsdk/__pycache__/UploadImageRequest.cpython-38.pyc and /dev/null differ diff --git a/AIlib/voduploadsdk/__pycache__/UploadVideoRequest.cpython-38.pyc b/AIlib/voduploadsdk/__pycache__/UploadVideoRequest.cpython-38.pyc deleted file mode 100644 index 5c7233a..0000000 Binary files a/AIlib/voduploadsdk/__pycache__/UploadVideoRequest.cpython-38.pyc and /dev/null differ diff --git a/AIlib/voduploadsdk/__pycache__/__init__.cpython-38.pyc b/AIlib/voduploadsdk/__pycache__/__init__.cpython-38.pyc deleted file mode 100644 index 5de1595..0000000 Binary files a/AIlib/voduploadsdk/__pycache__/__init__.cpython-38.pyc and /dev/null differ diff --git a/AIlib/weights/BiSeNet/checkpoint.pth b/AIlib/weights/BiSeNet/checkpoint.pth deleted file mode 100644 index e7b49ec..0000000 Binary files a/AIlib/weights/BiSeNet/checkpoint.pth and /dev/null differ diff --git a/AIlib/weights/yolov5/class5/best_5classes.pt b/AIlib/weights/yolov5/class5/best_5classes.pt deleted file mode 100644 index f4db424..0000000 Binary files a/AIlib/weights/yolov5/class5/best_5classes.pt and /dev/null differ diff --git a/AIlib/weights/yolov5/class5/labelnames.json b/AIlib/weights/yolov5/class5/labelnames.json deleted file mode 100644 index 99d6440..0000000 --- a/AIlib/weights/yolov5/class5/labelnames.json +++ /dev/null @@ -1,4 +0,0 @@ -{ - "labelnames":["排口","排污口","水生植被","漂浮物","其它"], - "labelIndexs":["SL014","SL011","SL013","SL001","SL001" ] -} diff --git a/AIlib/weights/yolov5/class9/F1_curve.png b/AIlib/weights/yolov5/class9/F1_curve.png deleted file mode 100644 index cad3d9e..0000000 Binary files a/AIlib/weights/yolov5/class9/F1_curve.png and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/PR_curve.png b/AIlib/weights/yolov5/class9/PR_curve.png deleted file mode 100644 index 4bd833c..0000000 Binary files a/AIlib/weights/yolov5/class9/PR_curve.png and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/P_curve.png b/AIlib/weights/yolov5/class9/P_curve.png deleted file mode 100644 index 882c62a..0000000 Binary files a/AIlib/weights/yolov5/class9/P_curve.png and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/R_curve.png b/AIlib/weights/yolov5/class9/R_curve.png deleted file mode 100644 index 5721b37..0000000 Binary files a/AIlib/weights/yolov5/class9/R_curve.png and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/best.pt b/AIlib/weights/yolov5/class9/best.pt deleted file mode 100644 index 9b24f6e..0000000 Binary files a/AIlib/weights/yolov5/class9/best.pt and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/confusion_matrix.png b/AIlib/weights/yolov5/class9/confusion_matrix.png deleted file mode 100644 index 24393f7..0000000 Binary files a/AIlib/weights/yolov5/class9/confusion_matrix.png and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/events.out.tfevents.1655953657.f899b1b7e8da.17376.0 b/AIlib/weights/yolov5/class9/events.out.tfevents.1655953657.f899b1b7e8da.17376.0 deleted file mode 100644 index f5ff0f0..0000000 Binary files a/AIlib/weights/yolov5/class9/events.out.tfevents.1655953657.f899b1b7e8da.17376.0 and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/hyp.yaml b/AIlib/weights/yolov5/class9/hyp.yaml deleted file mode 100644 index bd4aa8c..0000000 --- a/AIlib/weights/yolov5/class9/hyp.yaml +++ /dev/null @@ -1,28 +0,0 @@ -lr0: 0.01 -lrf: 0.01 -momentum: 0.937 -weight_decay: 0.0005 -warmup_epochs: 3.0 -warmup_momentum: 0.8 -warmup_bias_lr: 0.1 -box: 0.05 -cls: 0.5 -cls_pw: 1.0 -obj: 1.0 -obj_pw: 1.0 -iou_t: 0.2 -anchor_t: 4.0 -fl_gamma: 0.0 -hsv_h: 0.015 -hsv_s: 0.7 -hsv_v: 0.4 -degrees: 0.35 -translate: 0.2 -scale: 0.6 -shear: 0.5 -perspective: 0.001 -flipud: 0.5 -fliplr: 0.5 -mosaic: 1.0 -mixup: 0.243 -copy_paste: 0.0 diff --git a/AIlib/weights/yolov5/class9/labelnames.json b/AIlib/weights/yolov5/class9/labelnames.json deleted file mode 100644 index 26b634b..0000000 --- a/AIlib/weights/yolov5/class9/labelnames.json +++ /dev/null @@ -1,3 +0,0 @@ -{ - "labelnames":["排口","水生植被", "其它" ,"漂浮物","污口","菜地","违建","垃圾","河床"] -} diff --git a/AIlib/weights/yolov5/class9/labels.jpg b/AIlib/weights/yolov5/class9/labels.jpg deleted file mode 100644 index c0e4fce..0000000 Binary files a/AIlib/weights/yolov5/class9/labels.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/labels_correlogram.jpg b/AIlib/weights/yolov5/class9/labels_correlogram.jpg deleted file mode 100644 index 9dace54..0000000 Binary files a/AIlib/weights/yolov5/class9/labels_correlogram.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/opt.yaml b/AIlib/weights/yolov5/class9/opt.yaml deleted file mode 100644 index 1c4ef4d..0000000 --- a/AIlib/weights/yolov5/class9/opt.yaml +++ /dev/null @@ -1,39 +0,0 @@ -weights: yolov5s.pt -cfg: '' -data: data/River_detection_data.yaml -hyp: data/hyps/hyp.scratch-low.yaml -epochs: 200 -batch_size: 8 -imgsz: 640 -rect: false -resume: false -nosave: false -noval: false -noautoanchor: false -noplots: false -evolve: null -bucket: '' -cache: null -image_weights: false -device: '' -multi_scale: false -single_cls: false -optimizer: SGD -sync_bn: false -workers: 16 -project: runs/train -name: exp -exist_ok: false -quad: false -cos_lr: false -label_smoothing: 0.0 -patience: 100 -freeze: -- 0 -save_period: -1 -local_rank: -1 -entity: null -upload_dataset: false -bbox_interval: -1 -artifact_alias: latest -save_dir: runs/train/exp7 diff --git a/AIlib/weights/yolov5/class9/results.csv b/AIlib/weights/yolov5/class9/results.csv deleted file mode 100644 index e48f369..0000000 --- a/AIlib/weights/yolov5/class9/results.csv +++ /dev/null @@ -1,201 +0,0 @@ - epoch, train/box_loss, train/obj_loss, train/cls_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss, x/lr0, x/lr1, x/lr2 - 0, 0.1163, 0.078471, 0.060603, 0.0091424, 0.053632, 0.0032591, 0.00068739, 0.10289, 0.073506, 0.052194, 0.0033077, 0.0033077, 0.070231 - 1, 0.096567, 0.095328, 0.046852, 0.48139, 0.071425, 0.019772, 0.0038298, 0.089826, 0.087395, 0.038753, 0.0066082, 0.0066082, 0.040198 - 2, 0.092639, 0.096597, 0.03695, 0.60906, 0.11302, 0.044254, 0.0098758, 0.088638, 0.079321, 0.028477, 0.0098756, 0.0098756, 0.010132 - 3, 0.088354, 0.093358, 0.031132, 0.53835, 0.13594, 0.075434, 0.019504, 0.082118, 0.071749, 0.023135, 0.0098515, 0.0098515, 0.0098515 - 4, 0.084038, 0.095268, 0.026507, 0.59598, 0.16171, 0.12862, 0.037239, 0.075409, 0.070509, 0.017207, 0.0098515, 0.0098515, 0.0098515 - 5, 0.081941, 0.095587, 0.023229, 0.58377, 0.20923, 0.15806, 0.049477, 0.071079, 0.073671, 0.015088, 0.009802, 0.009802, 0.009802 - 6, 0.07997, 0.093637, 0.021312, 0.54366, 0.20309, 0.1499, 0.04285, 0.073025, 0.071019, 0.0136, 0.0097525, 0.0097525, 0.0097525 - 7, 0.078453, 0.097821, 0.02096, 0.4899, 0.20122, 0.13173, 0.033875, 0.077732, 0.068175, 0.012874, 0.009703, 0.009703, 0.009703 - 8, 0.077643, 0.094062, 0.020389, 0.61771, 0.183, 0.18765, 0.06058, 0.071304, 0.072565, 0.013283, 0.0096535, 0.0096535, 0.0096535 - 9, 0.077455, 0.092403, 0.018928, 0.68859, 0.17815, 0.19471, 0.057794, 0.070921, 0.075355, 0.012061, 0.009604, 0.009604, 0.009604 - 10, 0.076227, 0.095143, 0.018652, 0.54529, 0.22365, 0.2113, 0.07751, 0.068299, 0.070042, 0.011792, 0.0095545, 0.0095545, 0.0095545 - 11, 0.075621, 0.096919, 0.017805, 0.60475, 0.22383, 0.22727, 0.084465, 0.066867, 0.070019, 0.01066, 0.009505, 0.009505, 0.009505 - 12, 0.075278, 0.092505, 0.017624, 0.318, 0.38178, 0.2386, 0.088241, 0.06641, 0.069216, 0.010871, 0.0094555, 0.0094555, 0.0094555 - 13, 0.074596, 0.092825, 0.017083, 0.57635, 0.23396, 0.22908, 0.069304, 0.06832, 0.070965, 0.0096148, 0.009406, 0.009406, 0.009406 - 14, 0.074336, 0.090165, 0.01704, 0.34469, 0.35945, 0.24647, 0.093281, 0.067523, 0.067457, 0.010291, 0.0093565, 0.0093565, 0.0093565 - 15, 0.074199, 0.091868, 0.016145, 0.53659, 0.24416, 0.2452, 0.089367, 0.066737, 0.06926, 0.0096651, 0.009307, 0.009307, 0.009307 - 16, 0.073709, 0.091386, 0.016228, 0.58383, 0.27631, 0.26264, 0.095983, 0.065483, 0.06832, 0.0089162, 0.0092575, 0.0092575, 0.0092575 - 17, 0.073127, 0.092894, 0.015744, 0.37985, 0.32613, 0.24739, 0.087963, 0.068443, 0.066683, 0.0084403, 0.009208, 0.009208, 0.009208 - 18, 0.072616, 0.088464, 0.015487, 0.56692, 0.28385, 0.27279, 0.10299, 0.064793, 0.06803, 0.0089056, 0.0091585, 0.0091585, 0.0091585 - 19, 0.073291, 0.092542, 0.015206, 0.56882, 0.26733, 0.25371, 0.094288, 0.065888, 0.067756, 0.0085961, 0.009109, 0.009109, 0.009109 - 20, 0.072293, 0.090466, 0.015108, 0.57946, 0.30154, 0.27734, 0.10642, 0.0646, 0.067511, 0.0085873, 0.0090595, 0.0090595, 0.0090595 - 21, 0.072206, 0.092318, 0.015516, 0.6099, 0.27121, 0.26615, 0.097604, 0.066132, 0.067926, 0.0087777, 0.00901, 0.00901, 0.00901 - 22, 0.0725, 0.089683, 0.015057, 0.57866, 0.29392, 0.27893, 0.10599, 0.064014, 0.068478, 0.0088765, 0.0089605, 0.0089605, 0.0089605 - 23, 0.07183, 0.087295, 0.014614, 0.52155, 0.2966, 0.27902, 0.10725, 0.066293, 0.067119, 0.0092347, 0.008911, 0.008911, 0.008911 - 24, 0.072084, 0.09317, 0.014796, 0.49656, 0.32987, 0.28109, 0.11037, 0.064766, 0.067627, 0.0081019, 0.0088615, 0.0088615, 0.0088615 - 25, 0.07202, 0.09287, 0.014637, 0.52159, 0.3276, 0.29186, 0.11875, 0.063771, 0.067623, 0.0075126, 0.008812, 0.008812, 0.008812 - 26, 0.071227, 0.09077, 0.01366, 0.51896, 0.29976, 0.27309, 0.099052, 0.06738, 0.068287, 0.0089357, 0.0087625, 0.0087625, 0.0087625 - 27, 0.071496, 0.08987, 0.014003, 0.54018, 0.3222, 0.29007, 0.11414, 0.064039, 0.068832, 0.0080796, 0.008713, 0.008713, 0.008713 - 28, 0.070916, 0.093107, 0.014703, 0.47723, 0.34298, 0.29723, 0.11451, 0.064219, 0.066814, 0.0076718, 0.0086635, 0.0086635, 0.0086635 - 29, 0.070427, 0.088971, 0.013238, 0.54622, 0.31494, 0.30256, 0.11241, 0.06479, 0.067063, 0.0074613, 0.008614, 0.008614, 0.008614 - 30, 0.071102, 0.08735, 0.013443, 0.56721, 0.30333, 0.29104, 0.098722, 0.067164, 0.066289, 0.0075761, 0.0085645, 0.0085645, 0.0085645 - 31, 0.070841, 0.086304, 0.013342, 0.60408, 0.30416, 0.31094, 0.11943, 0.063399, 0.067872, 0.0075819, 0.008515, 0.008515, 0.008515 - 32, 0.071053, 0.090118, 0.014227, 0.53406, 0.32637, 0.31198, 0.12228, 0.064635, 0.066901, 0.0076553, 0.0084655, 0.0084655, 0.0084655 - 33, 0.069379, 0.090244, 0.013159, 0.55659, 0.31976, 0.31874, 0.12192, 0.064553, 0.06729, 0.0073647, 0.008416, 0.008416, 0.008416 - 34, 0.069977, 0.088556, 0.013386, 0.51855, 0.35026, 0.32884, 0.13784, 0.063374, 0.065856, 0.0072325, 0.0083665, 0.0083665, 0.0083665 - 35, 0.069105, 0.08728, 0.012953, 0.54258, 0.30226, 0.30634, 0.11756, 0.064441, 0.066554, 0.0070503, 0.008317, 0.008317, 0.008317 - 36, 0.069297, 0.087483, 0.013251, 0.46931, 0.36652, 0.32269, 0.13031, 0.063113, 0.067067, 0.0072046, 0.0082675, 0.0082675, 0.0082675 - 37, 0.069017, 0.089875, 0.012903, 0.53854, 0.31763, 0.32451, 0.12181, 0.064756, 0.068596, 0.0070815, 0.008218, 0.008218, 0.008218 - 38, 0.070204, 0.087944, 0.013116, 0.55487, 0.31301, 0.31483, 0.1313, 0.062245, 0.067089, 0.007585, 0.0081685, 0.0081685, 0.0081685 - 39, 0.068852, 0.088236, 0.012347, 0.45483, 0.37616, 0.32405, 0.13517, 0.062677, 0.067016, 0.0069761, 0.008119, 0.008119, 0.008119 - 40, 0.068883, 0.087472, 0.012411, 0.3262, 0.4185, 0.32651, 0.12907, 0.064086, 0.066548, 0.0071477, 0.0080695, 0.0080695, 0.0080695 - 41, 0.068694, 0.087602, 0.012271, 0.58739, 0.33346, 0.34273, 0.12701, 0.064972, 0.066784, 0.0071221, 0.00802, 0.00802, 0.00802 - 42, 0.067869, 0.08467, 0.012307, 0.50697, 0.35637, 0.33964, 0.13608, 0.06288, 0.067245, 0.0071069, 0.0079705, 0.0079705, 0.0079705 - 43, 0.068691, 0.085801, 0.012611, 0.53732, 0.3621, 0.34559, 0.1272, 0.064582, 0.067814, 0.0069667, 0.007921, 0.007921, 0.007921 - 44, 0.069254, 0.090675, 0.012805, 0.59737, 0.35245, 0.35435, 0.13788, 0.062864, 0.067236, 0.0066713, 0.0078715, 0.0078715, 0.0078715 - 45, 0.068376, 0.085859, 0.011942, 0.50395, 0.37409, 0.34755, 0.13974, 0.062393, 0.067137, 0.007546, 0.007822, 0.007822, 0.007822 - 46, 0.06861, 0.089459, 0.012456, 0.50414, 0.3914, 0.3425, 0.13007, 0.063929, 0.067608, 0.0067689, 0.0077725, 0.0077725, 0.0077725 - 47, 0.068593, 0.088558, 0.012249, 0.60456, 0.35743, 0.35891, 0.15363, 0.061952, 0.067407, 0.0071866, 0.007723, 0.007723, 0.007723 - 48, 0.06807, 0.08762, 0.011871, 0.51236, 0.37961, 0.36165, 0.15042, 0.061985, 0.066772, 0.0065586, 0.0076735, 0.0076735, 0.0076735 - 49, 0.069001, 0.089288, 0.012612, 0.50785, 0.41838, 0.37019, 0.15476, 0.061558, 0.066265, 0.006661, 0.007624, 0.007624, 0.007624 - 50, 0.067667, 0.086656, 0.011313, 0.46651, 0.46262, 0.37364, 0.1524, 0.061764, 0.067662, 0.0066976, 0.0075745, 0.0075745, 0.0075745 - 51, 0.067883, 0.087964, 0.011696, 0.55395, 0.3629, 0.36086, 0.13546, 0.063416, 0.066609, 0.0065205, 0.007525, 0.007525, 0.007525 - 52, 0.067801, 0.08903, 0.011977, 0.53392, 0.39244, 0.35415, 0.14072, 0.063178, 0.066399, 0.006225, 0.0074755, 0.0074755, 0.0074755 - 53, 0.067799, 0.0835, 0.011333, 0.5501, 0.38187, 0.35995, 0.13908, 0.063433, 0.066017, 0.006538, 0.007426, 0.007426, 0.007426 - 54, 0.067447, 0.088694, 0.011632, 0.54091, 0.38112, 0.36278, 0.14907, 0.061905, 0.066289, 0.0065171, 0.0073765, 0.0073765, 0.0073765 - 55, 0.067586, 0.085356, 0.011703, 0.52286, 0.40525, 0.36899, 0.14865, 0.06177, 0.066779, 0.0062129, 0.007327, 0.007327, 0.007327 - 56, 0.066878, 0.083479, 0.011399, 0.60013, 0.38319, 0.36966, 0.15262, 0.061542, 0.067507, 0.0064337, 0.0072775, 0.0072775, 0.0072775 - 57, 0.068031, 0.091226, 0.012066, 0.50613, 0.44063, 0.36909, 0.15244, 0.061868, 0.066166, 0.0065633, 0.007228, 0.007228, 0.007228 - 58, 0.067418, 0.085757, 0.011622, 0.56517, 0.40061, 0.36943, 0.14651, 0.062586, 0.066619, 0.0062601, 0.0071785, 0.0071785, 0.0071785 - 59, 0.066718, 0.085864, 0.011158, 0.55397, 0.38493, 0.36709, 0.14512, 0.063075, 0.066393, 0.0065429, 0.007129, 0.007129, 0.007129 - 60, 0.066731, 0.087387, 0.011068, 0.58739, 0.38874, 0.37592, 0.15718, 0.061648, 0.067504, 0.0069853, 0.0070795, 0.0070795, 0.0070795 - 61, 0.067061, 0.086715, 0.011811, 0.60299, 0.37194, 0.3742, 0.14995, 0.062568, 0.066529, 0.0066613, 0.00703, 0.00703, 0.00703 - 62, 0.066752, 0.084938, 0.010905, 0.61709, 0.35762, 0.36524, 0.14986, 0.062524, 0.067757, 0.0065433, 0.0069805, 0.0069805, 0.0069805 - 63, 0.066791, 0.084643, 0.010893, 0.62242, 0.41571, 0.39725, 0.16297, 0.061438, 0.066376, 0.0059869, 0.006931, 0.006931, 0.006931 - 64, 0.066845, 0.089024, 0.010864, 0.51756, 0.42181, 0.37814, 0.15135, 0.062043, 0.067269, 0.0060459, 0.0068815, 0.0068815, 0.0068815 - 65, 0.066513, 0.085811, 0.011389, 0.57159, 0.40589, 0.38348, 0.1593, 0.061298, 0.067347, 0.0068386, 0.006832, 0.006832, 0.006832 - 66, 0.066189, 0.084214, 0.010857, 0.52411, 0.43847, 0.37121, 0.15857, 0.061873, 0.066419, 0.00703, 0.0067825, 0.0067825, 0.0067825 - 67, 0.066244, 0.084572, 0.010595, 0.56425, 0.42456, 0.38402, 0.16437, 0.061371, 0.066397, 0.0064646, 0.006733, 0.006733, 0.006733 - 68, 0.066373, 0.086031, 0.010447, 0.55259, 0.39495, 0.36706, 0.13834, 0.063205, 0.065358, 0.0063739, 0.0066835, 0.0066835, 0.0066835 - 69, 0.066427, 0.086668, 0.010629, 0.61355, 0.38296, 0.36921, 0.15198, 0.062032, 0.066947, 0.0064047, 0.006634, 0.006634, 0.006634 - 70, 0.066244, 0.082839, 0.01088, 0.54489, 0.40873, 0.36629, 0.14945, 0.062507, 0.066586, 0.0066613, 0.0065845, 0.0065845, 0.0065845 - 71, 0.065914, 0.083775, 0.010386, 0.57188, 0.40664, 0.38227, 0.15763, 0.061905, 0.065879, 0.0061664, 0.006535, 0.006535, 0.006535 - 72, 0.065846, 0.083656, 0.010348, 0.61598, 0.39283, 0.37813, 0.15818, 0.061937, 0.067622, 0.0063666, 0.0064855, 0.0064855, 0.0064855 - 73, 0.065892, 0.085602, 0.010373, 0.62465, 0.367, 0.37421, 0.15764, 0.062358, 0.066741, 0.0063928, 0.006436, 0.006436, 0.006436 - 74, 0.065946, 0.086742, 0.010123, 0.61873, 0.38325, 0.37564, 0.16515, 0.062037, 0.067166, 0.0060427, 0.0063865, 0.0063865, 0.0063865 - 75, 0.065979, 0.084694, 0.010405, 0.59271, 0.40393, 0.37981, 0.15395, 0.06157, 0.067572, 0.0063874, 0.006337, 0.006337, 0.006337 - 76, 0.065851, 0.08633, 0.010715, 0.601, 0.38391, 0.37534, 0.15587, 0.061249, 0.067263, 0.0060625, 0.0062875, 0.0062875, 0.0062875 - 77, 0.065805, 0.085466, 0.01038, 0.61726, 0.36675, 0.37618, 0.15247, 0.061583, 0.066389, 0.0056583, 0.006238, 0.006238, 0.006238 - 78, 0.064958, 0.085547, 0.010087, 0.60557, 0.37817, 0.37016, 0.15905, 0.061912, 0.067135, 0.005876, 0.0061885, 0.0061885, 0.0061885 - 79, 0.065643, 0.08418, 0.010639, 0.63468, 0.38455, 0.36693, 0.15109, 0.061682, 0.068052, 0.0059359, 0.006139, 0.006139, 0.006139 - 80, 0.065556, 0.085941, 0.01048, 0.57815, 0.40519, 0.37027, 0.14387, 0.062216, 0.066558, 0.0059213, 0.0060895, 0.0060895, 0.0060895 - 81, 0.065182, 0.083699, 0.010384, 0.59147, 0.40593, 0.38949, 0.16124, 0.061847, 0.066603, 0.0060445, 0.00604, 0.00604, 0.00604 - 82, 0.065421, 0.085555, 0.010329, 0.61879, 0.37087, 0.3649, 0.1519, 0.06202, 0.066681, 0.0061421, 0.0059905, 0.0059905, 0.0059905 - 83, 0.065366, 0.086068, 0.010347, 0.55812, 0.39094, 0.36624, 0.15682, 0.062595, 0.067239, 0.0059068, 0.005941, 0.005941, 0.005941 - 84, 0.065097, 0.084055, 0.010372, 0.53991, 0.42806, 0.38254, 0.15846, 0.06136, 0.066975, 0.0057967, 0.0058915, 0.0058915, 0.0058915 - 85, 0.064374, 0.086181, 0.010123, 0.59861, 0.40198, 0.39171, 0.16501, 0.062239, 0.068405, 0.006259, 0.005842, 0.005842, 0.005842 - 86, 0.065197, 0.085523, 0.010152, 0.64409, 0.37603, 0.38836, 0.16013, 0.061907, 0.066445, 0.0059465, 0.0057925, 0.0057925, 0.0057925 - 87, 0.063779, 0.084176, 0.0089023, 0.63461, 0.37991, 0.38777, 0.15515, 0.0622, 0.06603, 0.0055444, 0.005743, 0.005743, 0.005743 - 88, 0.064477, 0.087534, 0.010278, 0.42722, 0.40099, 0.36072, 0.15154, 0.062104, 0.067563, 0.006415, 0.0056935, 0.0056935, 0.0056935 - 89, 0.064742, 0.084291, 0.0098472, 0.61324, 0.357, 0.36387, 0.13968, 0.063151, 0.067409, 0.0057912, 0.005644, 0.005644, 0.005644 - 90, 0.064703, 0.081543, 0.0099267, 0.6087, 0.38481, 0.37916, 0.16156, 0.062021, 0.067536, 0.0059791, 0.0055945, 0.0055945, 0.0055945 - 91, 0.064966, 0.086366, 0.010074, 0.61269, 0.38964, 0.39016, 0.15864, 0.061301, 0.067608, 0.0058913, 0.005545, 0.005545, 0.005545 - 92, 0.063874, 0.084096, 0.010313, 0.65604, 0.38164, 0.39434, 0.16384, 0.061271, 0.067982, 0.0055352, 0.0054955, 0.0054955, 0.0054955 - 93, 0.064593, 0.084133, 0.010451, 0.62026, 0.38759, 0.39417, 0.16559, 0.061492, 0.067733, 0.0062295, 0.005446, 0.005446, 0.005446 - 94, 0.064062, 0.081924, 0.0093699, 0.62588, 0.39245, 0.38855, 0.15933, 0.062091, 0.06688, 0.0056458, 0.0053965, 0.0053965, 0.0053965 - 95, 0.064347, 0.083627, 0.010303, 0.68128, 0.37355, 0.39119, 0.16429, 0.061663, 0.067386, 0.0060654, 0.005347, 0.005347, 0.005347 - 96, 0.064265, 0.083972, 0.0099317, 0.58813, 0.37959, 0.37562, 0.16018, 0.06199, 0.066333, 0.0057443, 0.0052975, 0.0052975, 0.0052975 - 97, 0.063302, 0.083542, 0.0094549, 0.59221, 0.40444, 0.38679, 0.16226, 0.061384, 0.066408, 0.0060679, 0.005248, 0.005248, 0.005248 - 98, 0.063991, 0.082154, 0.0095181, 0.56433, 0.4257, 0.38644, 0.16509, 0.061664, 0.067919, 0.0061783, 0.0051985, 0.0051985, 0.0051985 - 99, 0.064648, 0.086819, 0.010362, 0.62136, 0.38195, 0.39053, 0.15959, 0.061459, 0.066701, 0.006003, 0.005149, 0.005149, 0.005149 - 100, 0.063241, 0.081441, 0.0093159, 0.56769, 0.41884, 0.37789, 0.15233, 0.062306, 0.067186, 0.0058342, 0.0050995, 0.0050995, 0.0050995 - 101, 0.062968, 0.0836, 0.0089659, 0.64785, 0.37956, 0.38932, 0.16422, 0.061863, 0.067672, 0.0059326, 0.00505, 0.00505, 0.00505 - 102, 0.06284, 0.078934, 0.0090736, 0.57377, 0.40535, 0.38972, 0.16505, 0.061779, 0.067911, 0.0057767, 0.0050005, 0.0050005, 0.0050005 - 103, 0.064299, 0.082681, 0.010138, 0.61472, 0.40008, 0.38363, 0.1538, 0.061702, 0.066864, 0.0059251, 0.004951, 0.004951, 0.004951 - 104, 0.063765, 0.08581, 0.0096124, 0.62708, 0.38161, 0.38659, 0.16376, 0.061379, 0.067747, 0.0065927, 0.0049015, 0.0049015, 0.0049015 - 105, 0.064029, 0.086371, 0.0094885, 0.44298, 0.42325, 0.38235, 0.16022, 0.0611, 0.067532, 0.0059926, 0.004852, 0.004852, 0.004852 - 106, 0.063933, 0.083755, 0.0092936, 0.57128, 0.40904, 0.37533, 0.15955, 0.061493, 0.067812, 0.0060433, 0.0048025, 0.0048025, 0.0048025 - 107, 0.063925, 0.083809, 0.0096163, 0.55432, 0.41368, 0.37486, 0.16001, 0.061462, 0.068598, 0.005863, 0.004753, 0.004753, 0.004753 - 108, 0.062201, 0.083453, 0.0085994, 0.57548, 0.41776, 0.38787, 0.16373, 0.061403, 0.067542, 0.0060933, 0.0047035, 0.0047035, 0.0047035 - 109, 0.06275, 0.08118, 0.008878, 0.54169, 0.42289, 0.37186, 0.15456, 0.061683, 0.068245, 0.0057971, 0.004654, 0.004654, 0.004654 - 110, 0.063428, 0.082393, 0.0098573, 0.54333, 0.42902, 0.39734, 0.16654, 0.061067, 0.067816, 0.0057777, 0.0046045, 0.0046045, 0.0046045 - 111, 0.063, 0.080144, 0.0090875, 0.65566, 0.38288, 0.40283, 0.17192, 0.061325, 0.067714, 0.0060259, 0.004555, 0.004555, 0.004555 - 112, 0.06263, 0.083642, 0.0093641, 0.41437, 0.42205, 0.37776, 0.15887, 0.061718, 0.067645, 0.0063483, 0.0045055, 0.0045055, 0.0045055 - 113, 0.063118, 0.084509, 0.0096402, 0.62197, 0.38187, 0.38449, 0.16554, 0.060996, 0.067075, 0.0060038, 0.004456, 0.004456, 0.004456 - 114, 0.062873, 0.082505, 0.0088858, 0.60133, 0.3802, 0.38675, 0.1559, 0.060974, 0.068178, 0.0056237, 0.0044065, 0.0044065, 0.0044065 - 115, 0.062607, 0.082866, 0.008828, 0.57434, 0.3973, 0.37809, 0.17002, 0.061668, 0.068287, 0.0055489, 0.004357, 0.004357, 0.004357 - 116, 0.062916, 0.08195, 0.0095846, 0.59209, 0.39282, 0.37895, 0.15746, 0.061779, 0.067903, 0.0057769, 0.0043075, 0.0043075, 0.0043075 - 117, 0.062263, 0.080514, 0.0084992, 0.54274, 0.4159, 0.38896, 0.16493, 0.061529, 0.067438, 0.0059834, 0.004258, 0.004258, 0.004258 - 118, 0.063117, 0.080655, 0.0093576, 0.56906, 0.43282, 0.40171, 0.17407, 0.061155, 0.068624, 0.0057789, 0.0042085, 0.0042085, 0.0042085 - 119, 0.062702, 0.082241, 0.0088405, 0.52481, 0.42562, 0.38037, 0.15888, 0.061468, 0.068739, 0.0058799, 0.004159, 0.004159, 0.004159 - 120, 0.062472, 0.080464, 0.0090944, 0.57049, 0.39505, 0.37787, 0.16326, 0.061481, 0.068984, 0.0057145, 0.0041095, 0.0041095, 0.0041095 - 121, 0.062513, 0.082724, 0.0090526, 0.58679, 0.41222, 0.38838, 0.16373, 0.061139, 0.06887, 0.0058018, 0.00406, 0.00406, 0.00406 - 122, 0.06212, 0.080603, 0.0080341, 0.59962, 0.4259, 0.40007, 0.17061, 0.0617, 0.068887, 0.0057023, 0.0040105, 0.0040105, 0.0040105 - 123, 0.062704, 0.08208, 0.0084618, 0.58263, 0.42756, 0.39383, 0.15642, 0.061847, 0.068772, 0.0057635, 0.003961, 0.003961, 0.003961 - 124, 0.062541, 0.080097, 0.0091065, 0.60489, 0.4021, 0.38306, 0.15652, 0.06178, 0.06825, 0.0057683, 0.0039115, 0.0039115, 0.0039115 - 125, 0.06197, 0.081429, 0.0083321, 0.64762, 0.39049, 0.38741, 0.15672, 0.062131, 0.067943, 0.0056695, 0.003862, 0.003862, 0.003862 - 126, 0.062301, 0.081651, 0.0087167, 0.61908, 0.40452, 0.3906, 0.16294, 0.06121, 0.068418, 0.0056092, 0.0038125, 0.0038125, 0.0038125 - 127, 0.062221, 0.079613, 0.0084921, 0.61394, 0.39371, 0.38955, 0.16974, 0.061417, 0.06805, 0.0058537, 0.003763, 0.003763, 0.003763 - 128, 0.062123, 0.081567, 0.0090106, 0.59164, 0.39826, 0.38816, 0.16825, 0.061512, 0.06776, 0.0058033, 0.0037135, 0.0037135, 0.0037135 - 129, 0.062216, 0.080156, 0.008595, 0.66416, 0.37233, 0.39473, 0.16427, 0.061477, 0.068075, 0.0054373, 0.003664, 0.003664, 0.003664 - 130, 0.062499, 0.082136, 0.0089943, 0.5806, 0.40678, 0.38811, 0.16412, 0.061353, 0.068901, 0.0055116, 0.0036145, 0.0036145, 0.0036145 - 131, 0.06183, 0.079962, 0.0087681, 0.56545, 0.40643, 0.37984, 0.1655, 0.061294, 0.068665, 0.0058172, 0.003565, 0.003565, 0.003565 - 132, 0.061493, 0.079948, 0.0080891, 0.54577, 0.40385, 0.37962, 0.16849, 0.060879, 0.068566, 0.0059687, 0.0035155, 0.0035155, 0.0035155 - 133, 0.06131, 0.078567, 0.0086257, 0.60001, 0.38724, 0.3776, 0.16521, 0.061301, 0.06911, 0.0061088, 0.003466, 0.003466, 0.003466 - 134, 0.061697, 0.081441, 0.0085654, 0.6225, 0.39107, 0.3943, 0.16714, 0.061478, 0.06788, 0.0060577, 0.0034165, 0.0034165, 0.0034165 - 135, 0.061623, 0.079985, 0.0085603, 0.59139, 0.39628, 0.3785, 0.16194, 0.060983, 0.068897, 0.0057355, 0.003367, 0.003367, 0.003367 - 136, 0.061479, 0.079573, 0.0086732, 0.59881, 0.3988, 0.38341, 0.16106, 0.061175, 0.068809, 0.0055626, 0.0033175, 0.0033175, 0.0033175 - 137, 0.061413, 0.079408, 0.0084376, 0.61248, 0.39101, 0.3826, 0.15207, 0.061572, 0.068304, 0.0058183, 0.003268, 0.003268, 0.003268 - 138, 0.061046, 0.082144, 0.0089468, 0.55422, 0.42804, 0.38352, 0.1605, 0.06132, 0.069155, 0.0055395, 0.0032185, 0.0032185, 0.0032185 - 139, 0.061471, 0.07995, 0.0081675, 0.64321, 0.3717, 0.38275, 0.16285, 0.061362, 0.068849, 0.0055299, 0.003169, 0.003169, 0.003169 - 140, 0.060823, 0.080007, 0.0086115, 0.61921, 0.38559, 0.39302, 0.16613, 0.061234, 0.068911, 0.0052536, 0.0031195, 0.0031195, 0.0031195 - 141, 0.062534, 0.080749, 0.0094048, 0.58546, 0.40447, 0.38576, 0.16285, 0.061139, 0.069, 0.0056188, 0.00307, 0.00307, 0.00307 - 142, 0.060494, 0.080903, 0.0081691, 0.58388, 0.39437, 0.37923, 0.16453, 0.06183, 0.068096, 0.005568, 0.0030205, 0.0030205, 0.0030205 - 143, 0.061919, 0.083564, 0.0094533, 0.57831, 0.41394, 0.38564, 0.16276, 0.061135, 0.068703, 0.0057063, 0.002971, 0.002971, 0.002971 - 144, 0.060503, 0.077934, 0.0082693, 0.64683, 0.38783, 0.38866, 0.16462, 0.06107, 0.06893, 0.0057078, 0.0029215, 0.0029215, 0.0029215 - 145, 0.060227, 0.079083, 0.0080657, 0.58952, 0.41561, 0.39865, 0.16874, 0.061168, 0.068317, 0.0057771, 0.002872, 0.002872, 0.002872 - 146, 0.060703, 0.0805, 0.008231, 0.6058, 0.39931, 0.39018, 0.16269, 0.061424, 0.069072, 0.0056074, 0.0028225, 0.0028225, 0.0028225 - 147, 0.060809, 0.079416, 0.0084209, 0.58098, 0.42434, 0.39792, 0.17197, 0.060992, 0.069045, 0.0056534, 0.002773, 0.002773, 0.002773 - 148, 0.061181, 0.082232, 0.0086708, 0.6496, 0.37019, 0.38515, 0.16745, 0.061349, 0.070234, 0.0058444, 0.0027235, 0.0027235, 0.0027235 - 149, 0.060623, 0.083097, 0.0083962, 0.61744, 0.3864, 0.39006, 0.16122, 0.061336, 0.069071, 0.0055872, 0.002674, 0.002674, 0.002674 - 150, 0.061013, 0.078397, 0.0081109, 0.55319, 0.40884, 0.38888, 0.16894, 0.061426, 0.068424, 0.0055361, 0.0026245, 0.0026245, 0.0026245 - 151, 0.060906, 0.081067, 0.0087698, 0.58694, 0.40101, 0.38418, 0.16181, 0.061248, 0.069114, 0.0055448, 0.002575, 0.002575, 0.002575 - 152, 0.059971, 0.078016, 0.0082625, 0.54703, 0.42067, 0.38596, 0.15784, 0.060987, 0.068635, 0.0055849, 0.0025255, 0.0025255, 0.0025255 - 153, 0.060802, 0.080326, 0.0084251, 0.57626, 0.40981, 0.38869, 0.16595, 0.06094, 0.069052, 0.0055836, 0.002476, 0.002476, 0.002476 - 154, 0.060876, 0.080142, 0.0084623, 0.64034, 0.39032, 0.39224, 0.17052, 0.061258, 0.069313, 0.0057674, 0.0024265, 0.0024265, 0.0024265 - 155, 0.061197, 0.082908, 0.0082472, 0.62902, 0.40014, 0.39678, 0.17092, 0.060819, 0.069366, 0.0055072, 0.002377, 0.002377, 0.002377 - 156, 0.060326, 0.081083, 0.008418, 0.50159, 0.41416, 0.40481, 0.16998, 0.061234, 0.069954, 0.0053375, 0.0023275, 0.0023275, 0.0023275 - 157, 0.060145, 0.079134, 0.008171, 0.47626, 0.42082, 0.39914, 0.16988, 0.06098, 0.068871, 0.0056765, 0.002278, 0.002278, 0.002278 - 158, 0.059991, 0.078631, 0.0080613, 0.43714, 0.42794, 0.39301, 0.17251, 0.061006, 0.069104, 0.0055834, 0.0022285, 0.0022285, 0.0022285 - 159, 0.060168, 0.077415, 0.0081355, 0.4283, 0.43196, 0.39506, 0.1677, 0.06154, 0.069498, 0.0056595, 0.002179, 0.002179, 0.002179 - 160, 0.060013, 0.079332, 0.0078322, 0.48257, 0.40145, 0.39793, 0.16728, 0.06111, 0.069211, 0.0056615, 0.0021295, 0.0021295, 0.0021295 - 161, 0.059797, 0.076389, 0.0085948, 0.5621, 0.40947, 0.38788, 0.16676, 0.061514, 0.06914, 0.0058107, 0.00208, 0.00208, 0.00208 - 162, 0.05953, 0.079817, 0.0073943, 0.58375, 0.41093, 0.39391, 0.16412, 0.061514, 0.069279, 0.0055135, 0.0020305, 0.0020305, 0.0020305 - 163, 0.060089, 0.077, 0.0075225, 0.46733, 0.41864, 0.39178, 0.16438, 0.061474, 0.068694, 0.0057247, 0.001981, 0.001981, 0.001981 - 164, 0.060458, 0.082368, 0.0081539, 0.6308, 0.40624, 0.39519, 0.16866, 0.061197, 0.070284, 0.0057369, 0.0019315, 0.0019315, 0.0019315 - 165, 0.060204, 0.080686, 0.0078595, 0.58455, 0.40365, 0.39361, 0.16931, 0.061302, 0.070159, 0.0055198, 0.001882, 0.001882, 0.001882 - 166, 0.059445, 0.077755, 0.0076423, 0.57182, 0.44039, 0.40162, 0.16522, 0.061931, 0.069638, 0.005605, 0.0018325, 0.0018325, 0.0018325 - 167, 0.059385, 0.07858, 0.0077505, 0.5719, 0.4126, 0.38621, 0.16781, 0.061554, 0.069619, 0.0055225, 0.001783, 0.001783, 0.001783 - 168, 0.059453, 0.077021, 0.0081007, 0.6831, 0.38097, 0.39635, 0.16965, 0.061492, 0.069905, 0.0054181, 0.0017335, 0.0017335, 0.0017335 - 169, 0.059746, 0.07885, 0.0076966, 0.47484, 0.43266, 0.39458, 0.16799, 0.061473, 0.070302, 0.0055999, 0.001684, 0.001684, 0.001684 - 170, 0.060082, 0.080544, 0.0080393, 0.60734, 0.4093, 0.39553, 0.16867, 0.061064, 0.070145, 0.0056745, 0.0016345, 0.0016345, 0.0016345 - 171, 0.059243, 0.078759, 0.0079038, 0.59986, 0.41792, 0.3949, 0.17225, 0.061454, 0.070232, 0.0056883, 0.001585, 0.001585, 0.001585 - 172, 0.06003, 0.079498, 0.0080035, 0.58769, 0.41244, 0.40095, 0.17154, 0.061297, 0.069892, 0.0056825, 0.0015355, 0.0015355, 0.0015355 - 173, 0.05939, 0.077445, 0.0076199, 0.63194, 0.39694, 0.40471, 0.17308, 0.061558, 0.070094, 0.0056222, 0.001486, 0.001486, 0.001486 - 174, 0.058707, 0.078033, 0.0074211, 0.5845, 0.42189, 0.40162, 0.16921, 0.061337, 0.070011, 0.0058099, 0.0014365, 0.0014365, 0.0014365 - 175, 0.05945, 0.077614, 0.0084547, 0.5733, 0.42687, 0.40296, 0.17239, 0.061514, 0.06996, 0.0057241, 0.001387, 0.001387, 0.001387 - 176, 0.059536, 0.078818, 0.0078729, 0.64568, 0.39487, 0.40622, 0.1751, 0.061238, 0.06992, 0.0056799, 0.0013375, 0.0013375, 0.0013375 - 177, 0.059717, 0.07678, 0.0082633, 0.46492, 0.42903, 0.40088, 0.17398, 0.061486, 0.06953, 0.0054343, 0.001288, 0.001288, 0.001288 - 178, 0.058306, 0.075138, 0.0075247, 0.45756, 0.4475, 0.40696, 0.17211, 0.061272, 0.069687, 0.0055336, 0.0012385, 0.0012385, 0.0012385 - 179, 0.05981, 0.078592, 0.007826, 0.61119, 0.40411, 0.4047, 0.17375, 0.061233, 0.070374, 0.0056963, 0.001189, 0.001189, 0.001189 - 180, 0.059516, 0.07941, 0.0078784, 0.44491, 0.429, 0.39605, 0.16979, 0.061378, 0.070209, 0.0056918, 0.0011395, 0.0011395, 0.0011395 - 181, 0.059005, 0.080269, 0.0074365, 0.61792, 0.40438, 0.40462, 0.17199, 0.061282, 0.069842, 0.0056687, 0.00109, 0.00109, 0.00109 - 182, 0.059388, 0.0773, 0.0079592, 0.63607, 0.39592, 0.39934, 0.17166, 0.061353, 0.06989, 0.0055814, 0.0010405, 0.0010405, 0.0010405 - 183, 0.058627, 0.077392, 0.0073762, 0.65207, 0.39113, 0.39734, 0.16954, 0.061314, 0.070393, 0.0056366, 0.000991, 0.000991, 0.000991 - 184, 0.059958, 0.081703, 0.0084177, 0.60146, 0.41408, 0.40499, 0.17261, 0.061612, 0.070475, 0.0056101, 0.0009415, 0.0009415, 0.0009415 - 185, 0.059215, 0.079674, 0.0077917, 0.58245, 0.40966, 0.39842, 0.16618, 0.061862, 0.070253, 0.005696, 0.000892, 0.000892, 0.000892 - 186, 0.058119, 0.07413, 0.0074922, 0.58256, 0.41551, 0.39886, 0.16668, 0.061806, 0.070799, 0.0057426, 0.0008425, 0.0008425, 0.0008425 - 187, 0.059019, 0.08088, 0.0071937, 0.43465, 0.43227, 0.39255, 0.16627, 0.061765, 0.070654, 0.0057164, 0.000793, 0.000793, 0.000793 - 188, 0.05869, 0.0781, 0.0074471, 0.6328, 0.39031, 0.39202, 0.16877, 0.061584, 0.070445, 0.0056742, 0.0007435, 0.0007435, 0.0007435 - 189, 0.058051, 0.076634, 0.0070859, 0.62861, 0.39466, 0.39986, 0.17171, 0.061221, 0.070759, 0.005697, 0.000694, 0.000694, 0.000694 - 190, 0.059161, 0.078122, 0.007997, 0.6439, 0.38945, 0.39997, 0.16842, 0.061556, 0.070606, 0.0056897, 0.0006445, 0.0006445, 0.0006445 - 191, 0.059396, 0.079488, 0.0075062, 0.65862, 0.38499, 0.39836, 0.16873, 0.061452, 0.070499, 0.005647, 0.000595, 0.000595, 0.000595 - 192, 0.059279, 0.077722, 0.0074302, 0.62596, 0.40425, 0.40382, 0.17101, 0.061447, 0.070135, 0.0056625, 0.0005455, 0.0005455, 0.0005455 - 193, 0.058946, 0.077397, 0.0073196, 0.58967, 0.42201, 0.40048, 0.17041, 0.061256, 0.070125, 0.005794, 0.000496, 0.000496, 0.000496 - 194, 0.058338, 0.077359, 0.0075816, 0.60158, 0.40707, 0.3958, 0.17155, 0.06142, 0.070538, 0.0056139, 0.0004465, 0.0004465, 0.0004465 - 195, 0.058652, 0.079973, 0.0076763, 0.61455, 0.39871, 0.39545, 0.17112, 0.061313, 0.070764, 0.0056792, 0.000397, 0.000397, 0.000397 - 196, 0.058626, 0.076117, 0.0075257, 0.58013, 0.41444, 0.39569, 0.16962, 0.06143, 0.070803, 0.00566, 0.0003475, 0.0003475, 0.0003475 - 197, 0.058519, 0.07889, 0.0073568, 0.60364, 0.39333, 0.3906, 0.16771, 0.061568, 0.070652, 0.005606, 0.000298, 0.000298, 0.000298 - 198, 0.058598, 0.079198, 0.0076465, 0.61812, 0.39651, 0.3927, 0.17121, 0.061296, 0.070684, 0.0056408, 0.0002485, 0.0002485, 0.0002485 - 199, 0.058884, 0.077597, 0.0077865, 0.59517, 0.40793, 0.39424, 0.16871, 0.061455, 0.070537, 0.0056641, 0.000199, 0.000199, 0.000199 diff --git a/AIlib/weights/yolov5/class9/results.png b/AIlib/weights/yolov5/class9/results.png deleted file mode 100644 index 2ebc8ce..0000000 Binary files a/AIlib/weights/yolov5/class9/results.png and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/train_batch0.jpg b/AIlib/weights/yolov5/class9/train_batch0.jpg deleted file mode 100644 index 7c3d5e8..0000000 Binary files a/AIlib/weights/yolov5/class9/train_batch0.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/train_batch1.jpg b/AIlib/weights/yolov5/class9/train_batch1.jpg deleted file mode 100644 index 3c2a24e..0000000 Binary files a/AIlib/weights/yolov5/class9/train_batch1.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/train_batch2.jpg b/AIlib/weights/yolov5/class9/train_batch2.jpg deleted file mode 100644 index 84c4db6..0000000 Binary files a/AIlib/weights/yolov5/class9/train_batch2.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/val_batch0_labels.jpg b/AIlib/weights/yolov5/class9/val_batch0_labels.jpg deleted file mode 100644 index 95382b7..0000000 Binary files a/AIlib/weights/yolov5/class9/val_batch0_labels.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/val_batch0_pred.jpg b/AIlib/weights/yolov5/class9/val_batch0_pred.jpg deleted file mode 100644 index 49d0068..0000000 Binary files a/AIlib/weights/yolov5/class9/val_batch0_pred.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/val_batch1_labels.jpg b/AIlib/weights/yolov5/class9/val_batch1_labels.jpg deleted file mode 100644 index 42e060e..0000000 Binary files a/AIlib/weights/yolov5/class9/val_batch1_labels.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/val_batch1_pred.jpg b/AIlib/weights/yolov5/class9/val_batch1_pred.jpg deleted file mode 100644 index 06a2458..0000000 Binary files a/AIlib/weights/yolov5/class9/val_batch1_pred.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/val_batch2_labels.jpg b/AIlib/weights/yolov5/class9/val_batch2_labels.jpg deleted file mode 100644 index 94fb870..0000000 Binary files a/AIlib/weights/yolov5/class9/val_batch2_labels.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/val_batch2_pred.jpg b/AIlib/weights/yolov5/class9/val_batch2_pred.jpg deleted file mode 100644 index 574c332..0000000 Binary files a/AIlib/weights/yolov5/class9/val_batch2_pred.jpg and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/weights/best.pt b/AIlib/weights/yolov5/class9/weights/best.pt deleted file mode 100644 index 9b24f6e..0000000 Binary files a/AIlib/weights/yolov5/class9/weights/best.pt and /dev/null differ diff --git a/AIlib/weights/yolov5/class9/weights/last.pt b/AIlib/weights/yolov5/class9/weights/last.pt deleted file mode 100644 index baec14a..0000000 Binary files a/AIlib/weights/yolov5/class9/weights/last.pt and /dev/null differ diff --git a/test/demo.py b/demo.py similarity index 100% rename from test/demo.py rename to demo.py diff --git a/test/images/examples/2022-07-27-13-35-23_frame-160-240_type-排口_UfNOMh78smt0wezB_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-35-23_frame-160-240_type-排口_UfNOMh78smt0wezB_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-35-23_frame-160-240_type-排口_UfNOMh78smt0wezB_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-35-23_frame-160-240_type-排口_UfNOMh78smt0wezB_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-35-32_frame-428-480_type-排污口_jFuIUO357Z6kBxXg_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-35-32_frame-428-480_type-排污口_jFuIUO357Z6kBxXg_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-35-32_frame-428-480_type-排污口_jFuIUO357Z6kBxXg_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-35-32_frame-428-480_type-排污口_jFuIUO357Z6kBxXg_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-35-41_frame-516-720_type-排污口_o2QNRX4xrpcziV9w_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-35-41_frame-516-720_type-排污口_o2QNRX4xrpcziV9w_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-35-41_frame-516-720_type-排污口_o2QNRX4xrpcziV9w_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-35-41_frame-516-720_type-排污口_o2QNRX4xrpcziV9w_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-35-50_frame-931-960_type-排污口_h6Xr2u0T4Y5ZkFUA_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-35-50_frame-931-960_type-排污口_h6Xr2u0T4Y5ZkFUA_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-35-50_frame-931-960_type-排污口_h6Xr2u0T4Y5ZkFUA_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-35-50_frame-931-960_type-排污口_h6Xr2u0T4Y5ZkFUA_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-35-59_frame-1068-1200_type-排污口_epkNQdPqMcgsC8Du_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-35-59_frame-1068-1200_type-排污口_epkNQdPqMcgsC8Du_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-35-59_frame-1068-1200_type-排污口_epkNQdPqMcgsC8Du_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-35-59_frame-1068-1200_type-排污口_epkNQdPqMcgsC8Du_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-36-08_frame-1315-1440_type-排污口_28X6uDsAYzi0VmaG_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-36-08_frame-1315-1440_type-排污口_28X6uDsAYzi0VmaG_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-36-08_frame-1315-1440_type-排污口_28X6uDsAYzi0VmaG_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-36-08_frame-1315-1440_type-排污口_28X6uDsAYzi0VmaG_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-36-16_frame-1519-1680_type-排污口_LUiuGqmd10XMTtKY_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-36-16_frame-1519-1680_type-排污口_LUiuGqmd10XMTtKY_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-36-16_frame-1519-1680_type-排污口_LUiuGqmd10XMTtKY_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-36-16_frame-1519-1680_type-排污口_LUiuGqmd10XMTtKY_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-36-25_frame-1750-1920_type-排口_mVkyctd4HIuzNreF_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-36-25_frame-1750-1920_type-排口_mVkyctd4HIuzNreF_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-36-25_frame-1750-1920_type-排口_mVkyctd4HIuzNreF_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-36-25_frame-1750-1920_type-排口_mVkyctd4HIuzNreF_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-36-34_frame-2116-2160_type-水生植被_QuXwfae6vck3j4qJ_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-36-34_frame-2116-2160_type-水生植被_QuXwfae6vck3j4qJ_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-36-34_frame-2116-2160_type-水生植被_QuXwfae6vck3j4qJ_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-36-34_frame-2116-2160_type-水生植被_QuXwfae6vck3j4qJ_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-36-43_frame-2329-2400_type-水生植被_EqPuHXg2hsxUA5Y7_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-36-43_frame-2329-2400_type-水生植被_EqPuHXg2hsxUA5Y7_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-36-43_frame-2329-2400_type-水生植被_EqPuHXg2hsxUA5Y7_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-36-43_frame-2329-2400_type-水生植被_EqPuHXg2hsxUA5Y7_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-36-52_frame-2604-2640_type-水生植被_QjZ4FYy3twCdr5ga_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-36-52_frame-2604-2640_type-水生植被_QjZ4FYy3twCdr5ga_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-36-52_frame-2604-2640_type-水生植被_QjZ4FYy3twCdr5ga_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-36-52_frame-2604-2640_type-水生植被_QjZ4FYy3twCdr5ga_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-37-00_frame-2728-2880_type-水生植被_RC5sTu1qOdXH2WBh_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-37-00_frame-2728-2880_type-水生植被_RC5sTu1qOdXH2WBh_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-37-00_frame-2728-2880_type-水生植被_RC5sTu1qOdXH2WBh_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-37-00_frame-2728-2880_type-水生植被_RC5sTu1qOdXH2WBh_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-37-10_frame-3117-3120_type-水生植被_genKJUQ2XFxYGSA3_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-37-10_frame-3117-3120_type-水生植被_genKJUQ2XFxYGSA3_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-37-10_frame-3117-3120_type-水生植被_genKJUQ2XFxYGSA3_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-37-10_frame-3117-3120_type-水生植被_genKJUQ2XFxYGSA3_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-37-19_frame-3131-3360_type-排口_AC6d4PocQ7W2DzrS_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-37-19_frame-3131-3360_type-排口_AC6d4PocQ7W2DzrS_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-37-19_frame-3131-3360_type-排口_AC6d4PocQ7W2DzrS_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-37-19_frame-3131-3360_type-排口_AC6d4PocQ7W2DzrS_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-37-28_frame-3517-3600_type-排污口_BPJyiMcu5X1mZfGw_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-37-28_frame-3517-3600_type-排污口_BPJyiMcu5X1mZfGw_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-37-28_frame-3517-3600_type-排污口_BPJyiMcu5X1mZfGw_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-37-28_frame-3517-3600_type-排污口_BPJyiMcu5X1mZfGw_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-37-37_frame-3792-3840_type-水生植被_HvUOAF1CncGgRiqJ_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-37-37_frame-3792-3840_type-水生植被_HvUOAF1CncGgRiqJ_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-37-37_frame-3792-3840_type-水生植被_HvUOAF1CncGgRiqJ_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-37-37_frame-3792-3840_type-水生植被_HvUOAF1CncGgRiqJ_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-37-46_frame-3992-4080_type-排污口_bdoy79MHKY6SqJz4_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-37-46_frame-3992-4080_type-排污口_bdoy79MHKY6SqJz4_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-37-46_frame-3992-4080_type-排污口_bdoy79MHKY6SqJz4_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-37-46_frame-3992-4080_type-排污口_bdoy79MHKY6SqJz4_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-37-55_frame-4174-4320_type-排污口_fiKbtqDLXPZ5ANoe_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-37-55_frame-4174-4320_type-排污口_fiKbtqDLXPZ5ANoe_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-37-55_frame-4174-4320_type-排污口_fiKbtqDLXPZ5ANoe_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-37-55_frame-4174-4320_type-排污口_fiKbtqDLXPZ5ANoe_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-38-05_frame-4485-4560_type-排口_ZPxLFlRT9ds1crzh_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-38-05_frame-4485-4560_type-排口_ZPxLFlRT9ds1crzh_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-38-05_frame-4485-4560_type-排口_ZPxLFlRT9ds1crzh_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-38-05_frame-4485-4560_type-排口_ZPxLFlRT9ds1crzh_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-38-14_frame-4631-4800_type-排口_pyq8su0Y4xZvGNAc_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-38-14_frame-4631-4800_type-排口_pyq8su0Y4xZvGNAc_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-38-14_frame-4631-4800_type-排口_pyq8su0Y4xZvGNAc_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-38-14_frame-4631-4800_type-排口_pyq8su0Y4xZvGNAc_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-38-23_frame-4857-5040_type-水生植被_gB3oJ7ivRfOe0Lrj_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-38-23_frame-4857-5040_type-水生植被_gB3oJ7ivRfOe0Lrj_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-38-23_frame-4857-5040_type-水生植被_gB3oJ7ivRfOe0Lrj_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-38-23_frame-4857-5040_type-水生植被_gB3oJ7ivRfOe0Lrj_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg diff --git a/test/images/examples/2022-07-27-13-38-32_frame-5124-5280_type-水生植被_f836RQ9PDzvH5qIK_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg b/images/examples/2022-07-27-13-38-32_frame-5124-5280_type-水生植被_f836RQ9PDzvH5qIK_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg similarity index 100% rename from test/images/examples/2022-07-27-13-38-32_frame-5124-5280_type-水生植被_f836RQ9PDzvH5qIK_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg rename to images/examples/2022-07-27-13-38-32_frame-5124-5280_type-水生植被_f836RQ9PDzvH5qIK_s-off-P20220727133456697-30e25a4f15eb4756abd4571a2fcad2de_OR.jpg