在本教程中,我们将使用CIFAR10数据集。它有类别:“飞机”、“汽车”、“鸟”、“猫”、“鹿”、“狗”、“青蛙”、“马”、“船”、“卡车”。CIFAR-10中的图像大小为3x32x32,即3通道彩色图像大小为32x32像素。
我们将按以下顺序进行:
1.使用torchvision加载和规范CIFAR10培训和测试数据集
2.定义一个卷积神经网络
3.定义损失函数
4.根据培训数据对网络进行培训
5.在测试数据上测试网络
1.
import torch
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')
Out:
Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz
Extracting ./data/cifar-10-python.tar.gz to ./data
Files already downloaded and verified
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)))
plt.show()
# 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)))
Out:
car deer bird car
2.
import torch.nn as nn
import torch.nn.functional as F
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()
3.
import torch.optim as optim
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)
4.
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()
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')
Out:
[1, 2000] loss: 2.247
[1, 4000] loss: 1.899
[1, 6000] loss: 1.702
[1, 8000] loss: 1.574
[1, 10000] loss: 1.504
[1, 12000] loss: 1.489
[2, 2000] loss: 1.401
[2, 4000] loss: 1.391
[2, 6000] loss: 1.353
[2, 8000] loss: 1.332
[2, 10000] loss: 1.309
[2, 12000] loss: 1.291
Finished Training
保存训练的模型
PATH = './cifar_net.pth'
torch.save(net.state_dict(), PATH)
5.
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]))
Out:
Accuracy of plane : 62 %
Accuracy of car : 66 %
Accuracy of bird : 40 %
Accuracy of cat : 42 %
Accuracy of deer : 59 %
Accuracy of dog : 32 %
Accuracy of frog : 46 %
Accuracy of horse : 64 %
Accuracy of ship : 75 %
Accuracy of truck : 63 %
就像你把一个张量转移到GPU上一样,你把神经网络转移到GPU上。
让我们首先定义我们的设备为第一个可见的cuda设备,如果我们有cuda可用:
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# Assuming that we are on a CUDA machine, this should print a CUDA device:
print(device)
Out:
cuda:0
然后这些方法将递归遍历所有模块,并将其参数和缓冲区转换为CUDA张量,也要记得把输入转移到GPU
net.to(device)
inputs, labels = data[0].to(device), data[1].to(device)
会发现GPU提速不大,这是因为网络规模小