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')
数据处理使用torch.ultis.data
类与torchvision.transforms
类。
其中,transforms.Compose()
是将各种数据变换组合起来使用,由各种变换构成的列表。
torchvision.datasets.CIFAR10()
可以使用自己下载好的,解压后放在./data
内,并设置download=False
。
class torch.utils.data.DataLoader(dataset, batch_size=1, shuffle=False, sampler=None, num_workers=0, collate_fn=<function default_collate>, pin_memory=False, drop_last=False)
torch.utils.data.DataLoader
类是数据加载器,它组合数据集和采样器,并在数据集上提供单进程或多进程迭代器。
下面的代码用来显示部分图像:
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)))
# 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)))
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()
super()
构造函数待写
使用交叉熵criterion和SGD优化。
import torch.optim as optim
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)
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
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')
enumerate()
函数enumerate()
函数用于将一个可遍历的数据对象(如列表、元组或字符串)组合为一个索引序列,同时列出数据和数据下标,一般用在 for
循环当中。如下是enumerate的用法:
enumerate(sequence, [start=0])
>>>seq = ['one', 'two', 'three']
>>> for i, element in enumerate(seq):
... print i, element
...
0 one
1 two
2 three
iter()
与enumerate()
lst = [3,60,9]
strlst = ['a','box','c']
lst_enum = enumerate(lst)
lst_iter = iter(lst)
strlst_enum = enumerate(strlst)
strlst_iter = iter(strlst)
for a in lst_enum:
print(a)
>>> (0, 3)
(1, 60)
(2, 9)
for b in lst_iter:
print(b)
>>> 3
60
9
for c in strlst_enum:
print(c)
>>> (0, 'a')
(1, 'box')
(2, 'c')
for d in strlst_iter:
print(d)
>>> a
box
c
for e,element in lst_enum:
print(e,element)
>>> 0 a
1 box
2 c
dataiter = iter(testloader)
images, labels = dataiter.next()
# print images
# 可视化显示4张图像的预测结果
imshow(torchvision.utils.make_grid(images))
print('GroundTruth: ', ' '.join('%5s' % classes[labels[j]] for j in range(4)))
outputs = net(images)
_, predicted = torch.max(outputs, 1)
print('Predicted: ', ' '.join('%5s' % classes[predicted[j]]
for j in range(4)))
# 在整个测试集上测试准确率
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))
iter()
函数iter()
函数用来生成迭代器。
list
、tuple
等都是可迭代对象,我们可以通过iter()
函数获取这些可迭代对象的迭代器。然后我们可以对获取到的迭代器不断使⽤next()
函数来获取下⼀条数据。iter()
函数实际上就是调⽤了可迭代对象的__iter__
⽅法。
iter()
用法如下:
iter(object[, sentinel])
>>>lst = [1, 2, 3]
>>> for i in iter(lst):
... print(i)
...
1
2
3
next()
方法next()
方法在文件使用迭代器时会使用到,在循环中,next()
方法会在每次循环中调用,该方法返回文件的下一行,如果到达结尾(EOF),则触发 StopIteration
。
使用方法:
fileObject.next();
next()
函数next() 返回迭代器的下一个项目。用法如下:
next(iterator[, default])
iterator
– 可迭代对象default
– 可选,用于设置在没有下一个元素时返回该默认值,如果不设置,又没有下一个元素则会触发 StopIteration
异常。# 首先获得Iterator对象:
it = iter([1, 2, 3, 4, 5])
# 循环:
while True:
try:
# 获得下一个值:
x = next(it)
print(x)
except StopIteration:
# 遇到StopIteration就退出循环
break
输出:
1
2
3
4
5
iter(trainloader)
与enumerate(trainloader)
在可视化图像的时候用的是:
dataiter = iter(trainloader)
images, labels = dataiter.next()
这样一来images, labels的size都是4,是trainloader的1个batch_size的大小。
但是在训练网络的时候用的是
for i, data in enumerate(trainloader, 0):
这样遍历的i
是0-12499
[1] https://pytorch.org/tutorials/beginner/blitz/cifar10_tutorial.html#sphx-glr-beginner-blitz-cifar10-tutorial-py
[2] PyTorch文档中文版:https://pytorch-cn.readthedocs.io/zh/latest/
[3] http://www.runoob.com/python/