Windows下pytorch读取数据出现Broken pipe

原创已于 2024-03-29 14:45:04 修改 · 置顶 · 1.1k 阅读

3 ·

CC 4.0 BY-SA版权

文章标签：

#pytorch #深度学习 #windows #python

于 2021-02-26 22:58:44 首次发布

人工智能专栏收录该内容

1 篇文章

订阅专栏

在Windows上使用PyTorch进行深度学习训练时遇到RuntimeError，错误源于尝试在当前进程初始化阶段启动新进程。解决方案是将数据加载器的num_workers参数设置为0，避免在Windows系统中使用多线程读取数据。

部署运行你感兴趣的模型镜像

完整错误：

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 “H:\Desktop\PythonCode\ModelDeploy\pytorch-slimming-master\main.py”, line 160, in
train(epoch)
File “H:\Desktop\PythonCode\ModelDeploy\pytorch-slimming-master\main.py”, line 114, in train
for batch_idx, (data, target) in enumerate(train_loader):
File “F:\Anaconda3\lib\site-packages\torch\utils\data\dataloader.py”, line 291, in iter
return _MultiProcessingDataLoaderIter(self)
File “F:\Anaconda3\lib\site-packages\torch\utils\data\dataloader.py”, line 737, in init
w.start()
File “F:\Anaconda3\lib\multiprocessing\process.py”, line 112, in start
self._popen = self._Popen(self)
File “F:\Anaconda3\lib\multiprocessing\context.py”, line 223, in _Popen
return _default_context.get_context().Process._Popen(process_obj)
File “F:\Anaconda3\lib\multiprocessing\context.py”, line 322, in _Popen
return Popen(process_obj)
File “F:\Anaconda3\lib\multiprocessing\popen_spawn_win32.py”, line 89, in init
reduction.dump(process_obj, to_child)
File “F:\Anaconda3\lib\multiprocessing\reduction.py”, line 60, in dump
ForkingPickler(file, protocol).dump(obj)
BrokenPipeError: [Errno 32] Broken pipe