error in run.py

[SanSLee]

Traceback (most recent call last):
File "run.py", line 172, in
moduleNetwork = Network().cuda().eval()
File "run.py", line 124, in init
self.load_state_dict(torch.load('network-' + arguments_strModel + '.pytorch'))
File "/home/ubuntu/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/serialization.py", line 358, in load
return _load(f, map_location, pickle_module)
File "/home/ubuntu/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/serialization.py", line 532, in _load
magic_number = pickle_module.load(f)
_pickle.UnpicklingError: invalid load key, '<'.

When I run the file run.py, the error above occurred. Hope someone can help me.

[sniklaus]

It could be that the state dict got corrupted. The md5 hashes of them should be as follows.

ca2cea416fb99108bb08a8bfd31993cb  network-l1.pytorch
9121087a0550b77ebd6ae4100e4398a3  network-lf.pytorch

Can you check and see whether the md5 hashes of yours matches these?

[SanSLee]

Thanks! These two files have the same md5 value. That may be the problem and how can I solve it?

[sniklaus]

I am afraid that I am confused. You said that your files have the same md5 hash and note that this could be the underlying problem. Did you mean to say that your files have different md5 hashes?

Should they differ then I would recommend downloading the models again. Should they match then you might want to consider the following discussion: CSAILVision/places365#25 (comment)

[sniklaus]

Are you behind a firewall that alters your downloads by chance?

[SanSLee]

The problem is that the model files were downloaded incompletely before. I download them again in another way. Thanks for your help and your wonderful work!