Pytorch BrokenPipeError: [Errno 32] Broken pipe 报错解决

一、报错原因

Windows下多线程的问题，和torch.utils.data.DataLoader类有关。num_workers参数设置不当

from torch.utils.data import DataLoader
...
dataset_train = DataLoader(train_data, batch_size=batch_size, shuffle=True, num_workers=16)
dataset_test = DataLoader(test_data, batch_size=batch_size, shuffle=False, num_workers=16)

num_workers参数官方API解释：num_workers (int, optional) – how many subprocesses to use for data loading. 0 means that the data will be loaded in the main process. (default: 0)

该参数是指在进行数据集加载时，启用的线程数目。num_workers参数必须大于等于0，0的话表示数据集加载在主进程中进行，大于0表示通过多个进程来提升数据集加载速度。默认值为0。

二、解决方法

1. 将num_workers值设为0

from torch.utils.data import DataLoader
...
dataset_train = DataLoader(train_data, batch_size=batch_size, shuffle=True, num_workers=0)
dataset_test = DataLoader(test_data, batch_size=batch_size, shuffle=False, num_workers=0)

2. 如果num_workers的值大于0，要将运行的部分放进if __name__ == '__main__':才不会报错：

from torch.utils.data import DataLoader
...
if __name__ == '__main__':
	dataset_train = DataLoader(train_data, batch_size=batch_size, shuffle=True, num_workers=16)
	dataset_test = DataLoader(test_data, batch_size=batch_size, shuffle=False, num_workers=16)

3. 如果将运行部分放进main方法里面还报错，一般是num_workers设置太大了。可以调小一点

   OSError: [WinError 1455] 页面文件太小，无法完成操作。 Error loading “F:\anaconda3\envs\xxx\lib\site-packages\torch\lib\caffe2_detectron_ops_gpu.dll” or one of its dependencies.
   train(model, device, dataset_train, optimizer, epoch + 1, FocalLoss, batch_size)

---

num_workers参数设置技巧：

数据集较小时（小于2W）建议num_works不用管默认就行，因为用了反而比没用慢。
当数据集较大时建议采用，num_works一般设置为（CPU线程数±1）为最佳，可以用以下代码找出最佳num_works：

import time
import torch.utils.data as d
import torchvision
import torchvision.transforms as transforms

if __name__ == '__main__':
    BATCH_SIZE = 100
    transform = transforms.Compose([transforms.ToTensor(),
                                    transforms.Normalize((0.5,), (0.5,))])
    train_set = torchvision.datasets.MNIST('\mnist', download=True, train=True, transform=transform)

    # data loaders
    train_loader = d.DataLoader(train_set, batch_size=BATCH_SIZE, shuffle=True)

    for num_workers in range(20):
        train_loader = d.DataLoader(train_set, batch_size=BATCH_SIZE, shuffle=True, num_workers=num_workers)
        # training ...
        start = time.time()
        for epoch in range(1):
            for step, (batch_x, batch_y) in enumerate(train_loader):
                pass
        end = time.time()
        print('num_workers is {} and it took {} seconds'.format(num_workers, end - start))

参考文章：

https://blog.csdn.net/Ginomica_xyx/article/details/113745596
https://blog.csdn.net/qq_41196472/article/details/106393994