PyTorch - Torchvision - BrokenPipeError：[Errno 32] 断管答案

【问题标题】：PyTorch - Torchvision - BrokenPipeError: [Errno 32] Broken pipePyTorch - Torchvision - BrokenPipeError：[Errno 32] 断管
【发布时间】：2018-11-01 23:45:22
【问题描述】：

我正在尝试使用 PyTorch 执行名为 "Training a classifier" 的教程。当尝试调试这部分代码时：

import matplotlib.pyplot as plt
import numpy as np

# functions to show an image


def imshow(img):
    img = img / 2 + 0.5     # unnormalize
    npimg = img.numpy()
    plt.imshow(np.transpose(npimg, (1, 2, 0)))


# get some random training images
dataiter = iter(trainloader)
images, labels = dataiter.next()

# show images
imshow(torchvision.utils.make_grid(images))
# print labels
print(' '.join('%5s' % classes[labels[j]] for j in range(4)))

我收到此错误消息：

Files already downloaded and verified Files already downloaded and verified 
Files already downloaded and verified Files already downloaded and verified Traceback (most recent call last):   
File "<string>", line 1, in <module>   
File "D:\Anaconda\lib\multiprocessing\spawn.py", line 105, in spawn_main
    exitcode = _main(fd)   
File "D:\Anaconda\lib\multiprocessing\spawn.py", line 114, in _main
    prepare(preparation_data)   
File "D:\Anaconda\lib\multiprocessing\spawn.py", line 225, in prepare
    _fixup_main_from_path(data['init_main_from_path'])   
File "D:\Anaconda\lib\multiprocessing\spawn.py", line 277, in
_fixup_main_from_path
    run_name="__mp_main__")   
File "D:\Anaconda\lib\runpy.py", line 263, in run_path
    pkg_name=pkg_name, script_name=fname)   
File "D:\Anaconda\lib\runpy.py", line 96, in _run_module_code
    mod_name, mod_spec, pkg_name, script_name)   
File "D:\Anaconda\lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)   
File "d:\Yggdrasil\Programmation\PyTorch\TutorialCIFAR10.py", line 36, in <module>
    dataiter = iter(trainloader)   
File "D:\Anaconda\lib\site-packages\torch\utils\data\dataloader.py", line 451, in __iter__
    return _DataLoaderIter(self)   
File "D:\Anaconda\lib\site-packages\torch\utils\data\dataloader.py", line 239, in __init__
    w.start()   
File "D:\Anaconda\lib\multiprocessing\process.py", line 105, in start
    self._popen = self._Popen(self)   
File "D:\Anaconda\lib\multiprocessing\context.py", line 223, in _Popen
    return _default_context.get_context().Process._Popen(process_obj)     
File "D:\Anaconda\lib\multiprocessing\context.py", line 322, in _Popen
    return Popen(process_obj)   
File "D:\Anaconda\lib\multiprocessing\popen_spawn_win32.py", line 33, in __init__
    prep_data = spawn.get_preparation_data(process_obj._name)   
File "D:\Anaconda\lib\multiprocessing\spawn.py", line 143, in get_preparation_data
    _check_not_importing_main()   
File "D:\Anaconda\lib\multiprocessing\spawn.py", line 136, in
_check_not_importing_main
    is not going to be frozen to produce an executable.) 
RuntimeError:
        An attempt has been made to start a new process before the
        current process has finished its bootstrapping phase.

        This probably means that you are not using fork to start your
        child processes and you have forgotten to use the proper idiom
        in the main module:

            if __name__ == '__main__':
                freeze_support()
                ...

        The "freeze_support()" line can be omitted if the program
        is not going to be frozen to produce an executable. 
 Traceback (most recent call last):   
 File "d:\Yggdrasil\Programmation\PyTorch\TutorialCIFAR10.py", line 36, in <module>
    dataiter = iter(trainloader)   
 File "D:\Anaconda\lib\site-packages\torch\utils\data\dataloader.py", line 451, in __iter__
    return _DataLoaderIter(self)   
 File "D:\Anaconda\lib\site-packages\torch\utils\data\dataloader.py", line 239, in __init__
    w.start()   
 File "D:\Anaconda\lib\multiprocessing\process.py", line 105, in start
    self._popen = self._Popen(self)   
 File "D:\Anaconda\lib\multiprocessing\context.py", line 223, in _Popen
    return _default_context.get_context().Process._Popen(process_obj)   File "D:\Anaconda\lib\multiprocessing\context.py", line 322, in _Popen
    return Popen(process_obj)   
 File "D:\Anaconda\lib\multiprocessing\popen_spawn_win32.py", line 65, in
__init__
    reduction.dump(process_obj, to_child)   
 File "D:\Anaconda\lib\multiprocessing\reduction.py", line 60, in dump
    ForkingPickler(file, protocol).dump(obj) 
 BrokenPipeError: [Errno 32] Broken pipe

本教程中的所有前几行都运行良好。

请问有人知道怎么解决吗？提前非常感谢

【问题讨论】：

标签： python-3.x pytorch

【解决方案1】：

出现问题是因为 Windows 无法在 'num_workers' 中运行此 DataLoader 超过 0。你可以看到trainloader是从哪里来的。我们可以看到

trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,
                                      shuffle=True, num_workers=2)

我们需要将 'num_workers' 更改为 0。像这样：

trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,
                                      shuffle=True, num_workers=0)

每个装车机都需要这样改变。

【讨论】：

不只是trainloader，所有DataLoader都需要这样改，
但是有没有办法使用多线程通过 pytorch 在 Windows 上加载数据？
我不能使用多线程在带有 pytorch 的 cpu 版本的 windows 上加载数据。也许你可以试试 GPU 版本。你也可以在 jupyter notebook 中运行它。
在 GPU 上运行也有同样的问题。我会尝试在 jupyter notebook 中运行它。
如果您只是在 Windows 上从命令行运行，您将能够使用多线程

【解决方案2】：

遇到同样的错误。以下解决方法对我有用：

def run():
   # code goes here

if __name__ == '__main__':
    run()

【讨论】：

【解决方案3】：

这看起来不是 PyTorch 问题。尝试在 Jupyter 笔记本和其他环境故障排除中执行代码。

【讨论】：

【解决方案4】：

您需要添加一个 if 子句保护，如 pytorch 文档中所述： https://pytorch.org/docs/stable/notes/windows.html#usage-multiprocessing

【讨论】：