エラー 「DataLoader worker (pid 80830) is killed by signal: Unknown signal: 0. 」と表示される

Question

Pytorchのチュートリアルのコードを実行したところエラーが出てきたのですが, 解決方法がわからず困っています。

ソースコード

import torch
import torch.nn as nn
import torch.nn.functional as F
import torchvision
import torchvision.transforms as transforms

transform = transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])

trainset = torchvision.datasets.CIFAR10(root='./data', train=True,download=True, transform=transform)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=4,shuffle=True, num_workers=2)

testset = torchvision.datasets.CIFAR10(root='./data', train=False, download=True, transform=transform)
testloader = torch.utils.data.DataLoader(testset, batch_size=4,shuffle=False, num_workers=2)


classes = ('plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck')

import matplotlib.pyplot as plt
import numpy as np

def imshow(img):
    img = img / 2 + 0.5   # 標準化を戻す
    npimg = img.numpy()   # NumPy配列に変換
    plt.imshow(np.transpose(npimg, (1, 2, 0)))   # (高さ, 横幅, チャネル数)となるよう整形
    plt.show()   #画像の表示
    
# get some random training images
dataiter = iter(trainloader)
images, labels = dataiter.next()

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

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(3, 6, 5)
        self.pool = nn.MaxPool2d(2, 2)
        self.conv2 = nn.Conv2d(6, 16, 5)
        self.fc1 = nn.Linear(16 * 5 * 5, 120)
        self.fc2 = nn.Linear(120, 84)
        self.fc3 = nn.Linear(84, 10)

    def forward(self, x):
        x = self.pool(F.relu(self.conv1(x)))
        x = self.pool(F.relu(self.conv2(x)))
        x = x.view(-1, 16 * 5 * 5)
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x


net = Net()

import torch.optim as optim

criterion = nn.CrossEntropyLoss()   # 損失関数を交差エントロピーに設定
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)   # 最適化関数をSGDにしてmomentumを指定


for epoch in range(2):  # loop over the dataset multiple times

    running_loss = 0.0
    for i, data in enumerate(trainloader, 0):
        # get the inputs; data is a list of [inputs, labels]
        inputs, labels = data

        # zero the parameter gradients
        optimizer.zero_grad()

        # forward + backward + optimize
        outputs = net(inputs)
        loss = criterion(outputs, labels)
        loss.backward()
        optimizer.step()

        # print statistics
        running_loss += loss.item()
        
        #plot_history(running_loss,i,epoch)
        
        if i % 2000 == 1999:    # print every 2000 mini-batches
            print('[%d, %5d] loss: %.3f' % (epoch + 1, i + 1, running_loss / 2000))
            running_loss = 0.0

print('Finished Training')

PATH = './cifar_net.pth'
torch.save(net.state_dict(), PATH)

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

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

net = Net()
net.load_state_dict(torch.load(PATH))

outputs = net(images)

print("outputs:\n{}".format(outputs))

_, predicted = torch.max(outputs, 1)
print("max value:{}".format(_))
print("Predicted:{}".format(predicted))

print('Predicted: ', ' '.join('%5s' % classes[predicted[j]] for j in range(4)))

correct = 0
total = 0
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(images)
        _, predicted = torch.max(outputs.data, 1)
        total += labels.size(0)
        correct += (predicted == labels).sum().item()

print('Accuracy of the network on the 10000 test images: %d %%' % (100 * correct / total))

class_correct = list(0. for i in range(10))
class_total = list(0. for i in range(10))
with torch.no_grad():
    for data in testloader:
        images, labels = data
        outputs = net(images)
        _, predicted = torch.max(outputs, 1)
        c = (predicted == labels).squeeze()
        for i in range(4):
            label = labels[i]
            class_correct[label] += c[i].item()
            class_total[label] += 1


for i in range(10):
    print('Accuracy of %5s : %2d %%' % (
        classes[i], 100 * class_correct[i] / class_total[i]))

エラー内容

---

RuntimeError Traceback (most recent call last)
<ipython-input-9-99ffcbefe663> in <module>
26
27 # get some random training images
---> 28 dataiter = iter(trainloader)
29 images, labels = dataiter.next()
30

~/opt/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py in iter(self)
277 return _SingleProcessDataLoaderIter(self)
278 else:
--> 279 return _MultiProcessingDataLoaderIter(self)
280
281 @property

~/opt/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py in init(self, loader)
744 # prime the prefetch loop
745 for _ in range(2 * self._num_workers):
--> 746 self._try_put_index()
747
748 def _try_get_data(self, timeout=_utils.MP_STATUS_CHECK_INTERVAL):

~/opt/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py in _try_put_index(self)
859 assert self._tasks_outstanding < 2 * self._num_workers
860 try:
--> 861 index = self._next_index()
862 except StopIteration:
863 return

~/opt/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py in _next_index(self)
337
338 def _next_index(self):
--> 339 return next(self._sampler_iter) # may raise StopIteration
340
341 def _next_data(self):

~/opt/anaconda3/lib/python3.7/site-packages/torch/utils/data/sampler.py in iter(self)
198 def iter(self):
199 batch = []
--> 200 for idx in self.sampler:
201 batch.append(idx)
202 if len(batch) == self.batch_size:

~/opt/anaconda3/lib/python3.7/site-packages/torch/utils/data/sampler.py in iter(self)
105 if self.replacement:
106 return iter(torch.randint(high=n, size=(self.num_samples,), dtype=torch.int64).tolist())
--> 107 return iter(torch.randperm(n).tolist())
108
109 def len(self):

~/opt/anaconda3/lib/python3.7/site-packages/torch/utils/data/_utils/signal_handling.py in handler(signum, frame)
64 # This following call uses waitid with WNOHANG from C side. Therefore,
65 # Python can still get and update the process status successfully.
---> 66 _error_if_any_worker_fails()
67 if previous_handler is not None:
68 previous_handler(signum, frame)

**RuntimeError: DataLoader worker (pid 80830) is killed by signal: Unknown signal: 0. **

---

同じコードで昨日動かしていたときには, 問題なく動作していたので, コードの中身の問題ではないと思うのですが, 何かわかる方いましたらお願いします。

Answer 1

DataLoader設定のnum_workersの設定を２→0に変更してみてはいかがでしょうか。