소스 검색

Default FP16 TensorRT export (#6798)

* Assert engine precision #6777

* Default to FP32 inputs for TensorRT engines

* Default to FP16 TensorRT exports #6777

* Remove wrong line #6777

* Automatically adjust detect.py input precision #6777

* Automatically adjust val.py input precision #6777

* Add missing colon

* Cleanup

* Cleanup

* Remove default trt_fp16_input definition

* Experiment

* Reorder detect.py if statement to after half checks

* Update common.py

* Update export.py

* Cleanup

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
modifyDataloader
DavidB GitHub 2 년 전
부모
커밋
596de6d5a0
No known key found for this signature in database GPG Key ID: 4AEE18F83AFDEB23
4개의 변경된 파일13개의 추가작업 그리고 3개의 파일을 삭제
  1. +4
    -0
      detect.py
  2. +2
    -3
      export.py
  3. +3
    -0
      models/common.py
  4. +4
    -0
      val.py

+ 4
- 0
detect.py 파일 보기

@@ -97,6 +97,10 @@ def run(weights=ROOT / 'yolov5s.pt', # model.pt path(s)
half &= (pt or jit or onnx or engine) and device.type != 'cpu' # FP16 supported on limited backends with CUDA
if pt or jit:
model.model.half() if half else model.model.float()
elif engine and model.trt_fp16_input != half:
LOGGER.info('model ' + (
'requires' if model.trt_fp16_input else 'incompatible with') + ' --half. Adjusting automatically.')
half = model.trt_fp16_input

# Dataloader
if webcam:

+ 2
- 3
export.py 파일 보기

@@ -233,9 +233,8 @@ def export_engine(model, im, file, train, half, simplify, workspace=4, verbose=F
for out in outputs:
LOGGER.info(f'{prefix}\toutput "{out.name}" with shape {out.shape} and dtype {out.dtype}')

half &= builder.platform_has_fast_fp16
LOGGER.info(f'{prefix} building FP{16 if half else 32} engine in {f}')
if half:
LOGGER.info(f'{prefix} building FP{16 if builder.platform_has_fast_fp16 else 32} engine in {f}')
if builder.platform_has_fast_fp16:
config.set_flag(trt.BuilderFlag.FP16)
with builder.build_engine(network, config) as engine, open(f, 'wb') as t:
t.write(engine.serialize())

+ 3
- 0
models/common.py 파일 보기

@@ -338,6 +338,7 @@ class DetectMultiBackend(nn.Module):
import tensorrt as trt # https://developer.nvidia.com/nvidia-tensorrt-download
check_version(trt.__version__, '7.0.0', hard=True) # require tensorrt>=7.0.0
Binding = namedtuple('Binding', ('name', 'dtype', 'shape', 'data', 'ptr'))
trt_fp16_input = False
logger = trt.Logger(trt.Logger.INFO)
with open(w, 'rb') as f, trt.Runtime(logger) as runtime:
model = runtime.deserialize_cuda_engine(f.read())
@@ -348,6 +349,8 @@ class DetectMultiBackend(nn.Module):
shape = tuple(model.get_binding_shape(index))
data = torch.from_numpy(np.empty(shape, dtype=np.dtype(dtype))).to(device)
bindings[name] = Binding(name, dtype, shape, data, int(data.data_ptr()))
if model.binding_is_input(index) and dtype == np.float16:
trt_fp16_input = True
binding_addrs = OrderedDict((n, d.ptr) for n, d in bindings.items())
context = model.create_execution_context()
batch_size = bindings['images'].shape[0]

+ 4
- 0
val.py 파일 보기

@@ -144,6 +144,10 @@ def run(data,
model.model.half() if half else model.model.float()
elif engine:
batch_size = model.batch_size
if model.trt_fp16_input != half:
LOGGER.info('model ' + (
'requires' if model.trt_fp16_input else 'incompatible with') + ' --half. Adjusting automatically.')
half = model.trt_fp16_input
else:
half = False
batch_size = 1 # export.py models default to batch-size 1

Loading…
취소
저장