Question about training output

[byungjikim]

Hello, thanks for sharing your lane detection model.

I don't know about what is real output like culane_18.pth.
In my output folder, there are many files cfg.txt, ep000.pth ... and events.out.~ (binary file).
In my guess, 'events.out~~' is output like culane_18.pth but it does not work when trying to demo.py to my own dataset.

Error msg:
Traceback (most recent call last):
File "demo.py", line 33, in
state_dict = torch.load(cfg.test_model, map_location='cpu')['model']
File "/home/byungji/anaconda3/envs/lanedet/lib/python3.8/site-packages/torch/serialization.py", line 815, in load
return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
File "/home/byungji/anaconda3/envs/lanedet/lib/python3.8/site-packages/torch/serialization.py", line 1033, in _legacy_load
magic_number = pickle_module.load(f, **pickle_load_args)
_pickle.UnpicklingError: invalid load key, 'H'.

1. Is it right the output file like culane_18.pth is events.out binary file?
2. If it is right, can i get information about invalid load key?