-
Notifications
You must be signed in to change notification settings - Fork 386
Description
I use python ./tools/train.py configs/swin\cascade_mask_rcnn_swin_small_patch4_window7_mstrain_480-800_giou_4conv1f_adamw_3x_coco.py --work-dir=runs/1/ to train,and get an error:
File "C:\Users\anaconda3\envs\mmd\lib\site-packages\mmcv\parallel_functions.py", line 75, in forward
streams = [_get_stream(device) for device in target_gpus]
File "C:\Users\anaconda3\envs\mmd\lib\site-packages\mmcv\parallel_functions.py", line 75, in
streams = [_get_stream(device) for device in target_gpus]
File "C:\Users\anaconda3\envs\mmd\lib\site-packages\torch\nn\parallel_functions.py", line 118, in _get_stream
if device.type == "cpu":
AttributeError: 'int' object has no attribute 'type'
and I find the code is:
@staticmethod
def forward(target_gpus: List[int], input: Union[List, Tensor]) -> tuple:
input_device = get_input_device(input)
streams = None
if input_device == -1 and target_gpus != [-1]:
# Perform CPU to GPU copies in a background stream
streams = [_get_stream(device) for device in target_gpus]
and the device here is int 0
I believe it's the problem of gpu,I just run on 1 gpu but it make error in synchronize_stream,what should I do?