train.py 中TypeError: can't pickle Environment objects错误？

[joeqincai]

Traceback (most recent call last): File "train.py", line 322, in train_iter = iter(train_loader) File "D:\Anaconda\AnacondaRunning\lib\site-packages\torch\utils\data\dataloader.py", line 193, in iter return _DataLoaderIter(self) File "D:\Anaconda\AnacondaRunning\lib\site-packages\torch\utils\data\dataloader.py", line 469, in init w.start() File "D:\Anaconda\AnacondaRunning\lib\multiprocessing\process.py", line 105, in start self._popen = self._Popen(self) File "D:\Anaconda\AnacondaRunning\lib\multiprocessing\context.py", line 223, in _Popen return _default_context.get_context().Process._Popen(process_obj) File "D:\Anaconda\AnacondaRunning\lib\multiprocessing\context.py", line 322, in _Popen return Popen(process_obj) File "D:\Anaconda\AnacondaRunning\lib\multiprocessing\popen_spawn_win32.py", line 65, in init reduction.dump(process_obj, to_child) File "D:\Anaconda\AnacondaRunning\lib\multiprocessing\reduction.py", line 60, in dump ForkingPickler(file, protocol).dump(obj) TypeError: can't pickle Environment objects Traceback (most recent call last): File "", line 1, in File "D:\Anaconda\AnacondaRunning\lib\multiprocessing\spawn.py", line 99, in spawn_main new_handle = reduction.steal_handle(parent_pid, pipe_handle) File "D:\Anaconda\AnacondaRunning\lib\multiprocessing\reduction.py", line 87, in steal_handle _winapi.DUPLICATE_SAME_ACCESS | _winapi.DUPLICATE_CLOSE_SOURCE) PermissionError: [WinError 5] ▒ܾ▒▒▒▒ʡ▒

TypeError: can't pickle Environment objects 不知道是什么原因导致的？求讨论

[joeqincai]

for epoch in range(Config.epoch):

    train_iter = iter(train_loader)

    i = 0

    while i < len(train_loader):

        for p in net.parameters():

错误出在执行 iter(train_loader)以后；

[qlw-05]

解决了吗？我也遇到了这个问题，求教，求讨论。

[qlw-05]

查了资料应该是操作系统的问题，在Unix/Linux下，multiprocessing模块封装了fork（）调用，是我们不需要关注fork（）的细节。由于windows没有fork调用，详情参考https://blog.csdn.net/tpc4289/article/details/79280659。我现在先把Config.py里的data_worker设为0，编程但进程的。有进展在交流。