RuntimeError

[sisyphus229]

Initializing BaseSegmenter to cuda:0
Traceback (most recent call last):
File "D:\pythonProject\Track-Anything-master\app.py", line 383, in
model = TrackingAnything(SAM_checkpoint, xmem_checkpoint, e2fgvi_checkpoint,args)
File "D:\pythonProject\Track-Anything-master\track_anything.py", line 18, in init
self.samcontroler = SamControler(self.sam_checkpoint, args.sam_model_type, args.device)
File "D:\pythonProject\Track-Anything-master\tools\interact_tools.py", line 37, in init
self.sam_controler = BaseSegmenter(SAM_checkpoint, model_type, device)
File "D:\pythonProject\Track-Anything-master\tools\base_segmenter.py", line 25, in init
self.model = sam_model_registrymodel_type
File "D:\anaconda\envs\my_pytorch\lib\site-packages\segment_anything\build_sam.py", line 15, in build_sam_vit_h
return _build_sam(
File "D:\anaconda\envs\my_pytorch\lib\site-packages\segment_anything\build_sam.py", line 105, in _build_sam
state_dict = torch.load(f)
File "D:\anaconda\envs\my_pytorch\lib\site-packages\torch\serialization.py", line 777, in load
with _open_zipfile_reader(opened_file) as opened_zipfile:
File "D:\anaconda\envs\my_pytorch\lib\site-packages\torch\serialization.py", line 282, in init
super(_open_zipfile_reader, self).init(torch._C.PyTorchFileReader(name_or_buffer))
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

进程已结束,退出代码1

What is this problem?

[gaomingqi]

Hello, it seems pre-trained weights for SAM are not downloaded completely. Please clear the cached files in checkpoints folder and try again. Thanks.

[sisyphus229]

Hello, it seems pre-trained weights for SAM are not downloaded completely. Please clear the cached files in checkpoints folder and try again. Thanks.

yes, I just solved that problem, too. Thank you very much. And then I want to ask you how much memory i need to run the TAM, which always say is out of memory.

[gaomingqi]

Hello, please check our previous discussions for more details about memory: #4 (comment) (run TAM for tracking), #4 (comment) (run TAM for inpainting). Thanks.