train tacotron error

[cycycy-g]

Got everything setup as told in the readme guide. Got this error when running python train_tacotron.py. Has anybody encountered with this error?

Traceback (most recent call last):
  File "train_tacotron.py", line 202, in <module>
    main()
  File "train_tacotron.py", line 98, in main
    tts_train_loop(paths, model, optimizer, train_set, lr, training_steps, attn_example)
  File "train_tacotron.py", line 126, in tts_train_loop
    for i, (x, m, ids, _) in enumerate(train_set, 1):
  File "F:\conda\envs\env_pytorch\lib\site-packages\torch\utils\data\dataloader.py", line 279, in __iter__
    return _MultiProcessingDataLoaderIter(self)
  File "F:\conda\envs\env_pytorch\lib\site-packages\torch\utils\data\dataloader.py", line 719, in __init__
    w.start()
  File "F:\conda\envs\env_pytorch\lib\multiprocessing\process.py", line 121, in start
    self._popen = self._Popen(self)
  File "F:\conda\envs\env_pytorch\lib\multiprocessing\context.py", line 224, in _Popen
    return _default_context.get_context().Process._Popen(process_obj)
  File "F:\conda\envs\env_pytorch\lib\multiprocessing\context.py", line 327, in _Popen
    return Popen(process_obj)
  File "F:\conda\envs\env_pytorch\lib\multiprocessing\popen_spawn_win32.py", line 93, in __init__
    reduction.dump(process_obj, to_child)
  File "F:\conda\envs\env_pytorch\lib\multiprocessing\reduction.py", line 60, in dump
    ForkingPickler(file, protocol).dump(obj)
AttributeError: Can't pickle local object 'get_tts_datasets.<locals>.<lambda>'

C:\Users\user\Desktop\WaveRNN-master>Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "F:\conda\envs\env_pytorch\lib\multiprocessing\spawn.py", line 116, in spawn_main
    exitcode = _main(fd, parent_sentinel)
  File "F:\conda\envs\env_pytorch\lib\multiprocessing\spawn.py", line 126, in _main
    self = reduction.pickle.load(from_parent)
EOFError: Ran out of input

[FaWeeeZ]

Same issue for me

[serg06]

Same issue for me

[serg06]

It looks like this is due to the fact that Python lambdas can't be pickled on Windows. I am currently trying to find a workaround, but if none exists, then I guess we have to do it on linux.

[serg06]

I was able to fix this by changing this:

In dataset.py change DataLoader( ... num_workers=1 ) to DataLoader( ... num_workers=0 )

[FaWeeeZ]

@serg06 thank you so much, fixed it for me

[fatchord]

thanks @serg06 for helping out here :slightly_smiling_face:

[serg06]

Glad to help 😄