Default DataLoader `shuffle=True` for training (#5623)

* Fix shuffle DataLoader argument

* Add shuffle argument

* Disable shuffle when rect

* Cleanup, add rect warning

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Cleanup2

* Cleanup3

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

train.py

@@ -212,7 +212,7 @@ def train(hyp,  # path/to/hyp.yaml or hyp dictionary
     train_loader, dataset = create_dataloader(train_path, imgsz, batch_size // WORLD_SIZE, gs, single_cls,
                                               hyp=hyp, augment=True, cache=opt.cache, rect=opt.rect, rank=LOCAL_RANK,
                                               workers=workers, image_weights=opt.image_weights, quad=opt.quad,
-                                              prefix=colorstr('train: '))
+                                              prefix=colorstr('train: '), shuffle=True)
     mlc = int(np.concatenate(dataset.labels, 0)[:, 0].max())  # max label class
     nb = len(train_loader)  # number of batches
     assert mlc < nc, f'Label class {mlc} exceeds nc={nc} in {data}. Possible class labels are 0-{nc - 1}'

utils/datasets.py

@@ -22,7 +22,7 @@ import torch
 import torch.nn.functional as F
 import yaml
 from PIL import ExifTags, Image, ImageOps
-from torch.utils.data import Dataset
+from torch.utils.data import DataLoader, Dataset, dataloader, distributed
 from tqdm import tqdm
 
 from utils.augmentations import Albumentations, augment_hsv, copy_paste, letterbox, mixup, random_perspective
@@ -93,13 +93,15 @@ def exif_transpose(image):
 
 
 def create_dataloader(path, imgsz, batch_size, stride, single_cls=False, hyp=None, augment=False, cache=False, pad=0.0,
-                      rect=False, rank=-1, workers=8, image_weights=False, quad=False, prefix=''):
+                      rect=False, rank=-1, workers=8, image_weights=False, quad=False, prefix='', shuffle=False):
-    # Make sure only the first process in DDP process the dataset first, and the following others can use the cache
+    if rect and shuffle:
-    with torch_distributed_zero_first(rank):
+        LOGGER.warning('WARNING: --rect is incompatible with DataLoader shuffle, setting shuffle=False')
+        shuffle = False
+    with torch_distributed_zero_first(rank):  # init dataset *.cache only once if DDP
         dataset = LoadImagesAndLabels(path, imgsz, batch_size,
-                                      augment=augment,  # augment images
+                                      augment=augment,  # augmentation
-                                      hyp=hyp,  # augmentation hyperparameters
+                                      hyp=hyp,  # hyperparameters
-                                      rect=rect,  # rectangular training
+                                      rect=rect,  # rectangular batches
                                       cache_images=cache,
                                       single_cls=single_cls,
                                       stride=int(stride),
@@ -109,19 +111,18 @@ def create_dataloader(path, imgsz, batch_size, stride, single_cls=False, hyp=Non
 
     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
+    sampler = None if rank == -1 else distributed.DistributedSampler(dataset, shuffle=shuffle)
-    loader = torch.utils.data.DataLoader if image_weights else InfiniteDataLoader
+    loader = DataLoader if image_weights else InfiniteDataLoader  # only DataLoader allows for attribute updates
-    # Use torch.utils.data.DataLoader() if dataset.properties will update during training else InfiniteDataLoader()
+    return loader(dataset,
-    dataloader = loader(dataset,
+                  batch_size=batch_size,
-                        batch_size=batch_size,
+                  shuffle=shuffle and sampler is None,
-                        num_workers=nw,
+                  num_workers=nw,
-                        sampler=sampler,
+                  sampler=sampler,
-                        pin_memory=True,
+                  pin_memory=True,
-                        collate_fn=LoadImagesAndLabels.collate_fn4 if quad else LoadImagesAndLabels.collate_fn)
+                  collate_fn=LoadImagesAndLabels.collate_fn4 if quad else LoadImagesAndLabels.collate_fn), dataset
-    return dataloader, dataset
 
 
-class InfiniteDataLoader(torch.utils.data.dataloader.DataLoader):
+class InfiniteDataLoader(dataloader.DataLoader):
     """ Dataloader that reuses workers
 
     Uses same syntax as vanilla DataLoader