Hi,
I selected CPU option with this command:
pip3 install torch torchvision torchaudio
and opted for track w/o identities. still, it generates this error:
ERROR An error occurred, saving data before printing traceback and exiting the run_idtrackerai.py:139
program
INFO Saving video object in video.py:504
C:\Users\Crop2x\Desktop\IDTRacker.ai\session_no_name\video_object.json
11:49:26 INFO Saving ListOfBlobs at list_of_blobs.py:133
C:\Users\Crop2x\Desktop\IDTRacker.ai\session_no_name\preprocessing\list_of_blobs
.pickle
CRITICAL module 'torch._C' has no attribute '_cuda_setDevice' init_logger.py:73
Traceback (most recent call last):
File
"C:\Users\Crop2x\Desktop\IDTRacker.ai\myenv\Lib\site-packages\idtrackerai\utils\ini
t_logger.py", line 68, in applicator
return main_function(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File
"C:\Users\Crop2x\Desktop\IDTRacker.ai\myenv\Lib\site-packages\idtrackerai_start_app
\__main__.py", line 112, in main
return RunIdTrackerAi(parameters).track_video()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File
"C:\Users\Crop2x\Desktop\IDTRacker.ai\myenv\Lib\site-packages\idtrackerai_start_app
\run_idtrackerai.py", line 144, in track_video
raise error
File
"C:\Users\Crop2x\Desktop\IDTRacker.ai\myenv\Lib\site-packages\idtrackerai_start_app
\run_idtrackerai.py", line 90, in track_video
crossings_detection_API(self.video, self.list_of_blobs)
File
"C:\Users\Crop2x\Desktop\IDTRacker.ai\myenv\Lib\site-packages\idtrackerai\crossings
_detection\crossings_detection.py", line 74, in crossings_detection_API
detect_crossings(list_of_blobs, video)
File
"C:\Users\Crop2x\Desktop\IDTRacker.ai\myenv\Lib\site-packages\idtrackerai\crossings
_detection\crossing_detector.py", line 141, in detect_crossings
torch.cuda.set_device(0)
File
"C:\Users\Crop2x\Desktop\IDTRacker.ai\myenv\Lib\site-packages\torch\cuda\__init__.p
y", line 350, in set_device
torch._C._cuda_setDevice(device)
^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: module 'torch._C' has no attribute '_cuda_setDevice'