运行时错误：DataLoader 工作线程（pid(s) 15876、2756）意外退出

Question

我正在编译 PyTorch 教程网站上的一些现有示例。我主要在没有 GPU 的 CPU 设备上工作。

运行程序时，会显示以下错误类型。我正在处理 CPU 设备或设置问题吗？raise RuntimeError('DataLoader worker (pid(s) {}) exited unexpectedly'.format(pids_str)) from e 运行时错误：DataLoader 工作线程（pid(s) 15876, 2756）意外退出`。我该如何解决？

import torch
import torch.functional as F
import torch.nn as nn
import torch.optim as optim

import torchvision
import torchvision.transforms as transforms

import matplotlib.pyplot as plt
import numpy as np

from torch.utils.tensorboard import SummaryWriter
from torch.utils.data import DataLoader
from torchvision import datasets

device = 'cpu' if torch.cuda.is_available() else 'cuda'
print(device)

transform = transforms.Compose(
[transforms.ToTensor(),
 transforms.Normalize((0.5,), (0.5,))]
)
#Store separate training and validations splits in data
training_set = datasets.FashionMNIST(
 root='data',
 train=True,
 download=True,
 transform=transform
)
validation_set = datasets.FashionMNIST(
root='data',
train=False,
download=True,
transform=transform
)
training_loader = DataLoader(training_set, batch_size=4, shuffle=True, num_workers=2)
validation_loader = DataLoader(validation_set, batch_size=4, shuffle=False, num_workers=2)
classes = ('T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',
    'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle Boot')


def matplotlib_imshow(img, one_channel=False):
  if one_channel:
     img = img.mean(dim=0)
img = img/2+0.5 #unnormalize
npimg = img.numpy()
if one_channel:
    plt.imshow(npimg, cmap="Greys")
else:
    plt.imshow(np.transpose(npimg, (1, 2, 0)))


dataiter = iter(training_loader)
images, labels = dataiter.next()

img_grid = torchvision.utils.make_grid(images)
matplotlib_imshow(img_grid, one_channel=True)

Answer 1

set num_workers=0 在 Windows 上，由于多处理限制，将 num_workers 设置为 > 0 会导致错误。这是预料之中的。

Github上也有一个问题：