UnicodeDecodeError on Windows10

[TimeHunter618]

Thank you sincerely for your work. I got a stuck in running the code, which is as follow :

My env is: win10, anaconda, pytorch=1.1.0 with cuda auto installed by conda

An UnicodeDecodeError happened when I running: (ad) λ p python 1_train_predictor.py --data ecg --filename chfdb_chf01_275.pkl

The error is: UnicodeDecodeError: 'utf-8' codec can't decode byte 0xc8 in position 93: invalid continuation byte

The detail is: `> building the model The using device is : cuda Traceback (most recent call last):

File "1_train_predictor.py", line 115, in model = model.to(device(args.device))

File "E:\ProgramSoftW\Anaconda\Installed\envs\ad\lib\site-packages\torch\nn\modules\module.py", line 386, in to return self._apply(convert)

File "E:\ProgramSoftW\Anaconda\Installed\envs\ad\lib\site-packages\torch\nn\modules\module.py", line 193, in _apply module._apply(fn)

File "E:\ProgramSoftW\Anaconda\Installed\envs\ad\lib\site-packages\torch\nn\modules\rnn.py", line 127, in _apply self.flatten_parameters()

File "E:\ProgramSoftW\Anaconda\Installed\envs\ad\lib\site-packages\torch\nn\modules\rnn.py", line 100, in flatten_parameters if not any_param.is_cuda or not torch.backends.cudnn.is_acceptable(any_param):

File "E:\ProgramSoftW\Anaconda\Installed\envs\ad\lib\site-packages\torch\backends\cudnn__init__.py", line 105, in is_acceptable if _libcudnn() is None:

File "E:\ProgramSoftW\Anaconda\Installed\envs\ad\lib\site-packages\torch\backends\cudnn__init__.py", line 48, in _libcudnn lib = find_cudnn_windows_lib()

File "E:\ProgramSoftW\Anaconda\Installed\envs\ad\lib\site-packages\torch\backends\cudnn__init__.py", line 32, in find_cudnn_windows_lib out = out.decode().strip()

UnicodeDecodeError: 'utf-8' codec can't decode byte 0xc8 in position 93: invalid continuation byte`

The solutions I had try are:

• I tried to use pytorch=0.4.0 and pytorch=1.1.0 seperately in python 3.5 and 3.6 3.7,but it doesn't work

• I tried to run it use cpu. It works. But It takes too long.

• I tried to change the encoding way, encoding='utf-8'. It worked partly

I cann't figure out what's happened in cuda or pytorch with the error. Looking for Ur reply~

[ivanokeeffe]

Hi did you solve this issue yet?

[TimeHunter618]

Hi did you solve this issue yet?

I haven't solve it yet.

[HYixuann]

Try to download the dataset on the official website.