pytorch教程实现mnist手写数字识别代码示例

时间:2021-07-30 09:01:10

1.构建网络

nn.Moudle是pytorch官方指定的编写Net模块,在init函数中添加需要使用的层,在foeword中定义网络流向。

下面详细解释各层:

conv1层:输入channel = 1 ,输出chanael = 10,滤波器5*5

maxpooling = 2*2

conv2层:输入channel = 10 ,输出chanael = 20,滤波器5*5,

dropout

maxpooling = 2*2

fc1层:输入320 个神经单元,输出50个神经单元
fc1层:输入50个神经单元 ,输出10个神经单元
?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 10, kernel_size=5)        
        self.conv2 = nn.Conv2d(10, 20, kernel_size=5)
        self.conv2_drop = nn.Dropout2d()
        self.fc1 = nn.Linear(320, 50)
        self.fc2 = nn.Linear(50, 10)
    def forward(self, x):                                 #x.size() = 28*28*1
        x = F.relu(F.max_pool2d(self.conv1(x), 2))
        x = F.relu(F.max_pool2d(self.conv2_drop(self.conv2(x)), 2))  #x.size() =12*12*10   
        x = x.view(-1, 320)                                          #x.size() =1*320
        x = F.relu(self.fc1(x))
        x = F.dropout(x, training=self.training)
        x = self.fc2(x)
        return F.log_softmax(x, dim=1)

2.编写训练代码

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
model = Net()                                 #调用写好的网络
if args.cuda:                                 #如果有GPU使用CPU
    model.cuda()
optimizer = optim.SGD(model.parameters(), lr=args.lr, momentum=args.momentum)   #设置SGD随机梯度下降算法
def train(epoch):                                                
    model.train()                            
    for batch_idx, (data, target) in enumerate(train_loader):
        if args.cuda:
            data, target = data.cuda(), target.cuda()
        data, target = Variable(data), Variable(target)
        optimizer.zero_grad()                                  #梯度初始化为O
        output = model(data)
        loss = F.nll_loss(output, target)                      #简历loss function
        loss.backward()                                        #反向传播,计算梯度
        optimizer.step()                                       #更新权重
        if batch_idx % args.log_interval == 0:                 #输出信息
            print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
                epoch, batch_idx * len(data), len(train_loader.dataset),
                100. * batch_idx / len(train_loader), loss.data[0]))

3.编写测试代码

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
def test():
    model.eval()
    test_loss = 0
    correct = 0
    for data, target in test_loader:
        if args.cuda:
            data, target = data.cuda(), target.cuda()
        data, target = Variable(data, volatile=True), Variable(target)
        output = model(data)
        test_loss += F.nll_loss(output, target, size_average=False).data[0] # sum up batch loss
        pred = output.data.max(1, keepdim=True)[1] # get the index of the max log-probability
        correct += pred.eq(target.data.view_as(pred)).long().cpu().sum()
     test_loss /= len(test_loader.dataset)
    print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
        test_loss, correct, len(test_loader.dataset),
        100. * correct / len(test_loader.dataset)))

4.指导程序train和test

?
1
2
3
for epoch in range(1, args.epochs + 1):
    train(epoch)                              #训练N个epoch
    test()                                    #检验在测试集上的表现

5.完整代码

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
# -*- coding: utf-8 -*-
from __future__ import print_function
import argparse
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torchvision import datasets, transforms
from torch.autograd import Variable
# Training settings
parser = argparse.ArgumentParser(description='PyTorch MNIST Example')
parser.add_argument('--batch-size', type=int, default=64, metavar='N',
                    help='input batch size for training (default: 64)')
parser.add_argument('--test-batch-size', type=int, default=1000, metavar='N',
                    help='input batch size for testing (default: 1000)')
parser.add_argument('--epochs', type=int, default=10, metavar='N',
                    help='number of epochs to train (default: 10)')
parser.add_argument('--lr', type=float, default=0.01, metavar='LR',
                    help='learning rate (default: 0.01)')
parser.add_argument('--momentum', type=float, default=0.5, metavar='M',
                    help='SGD momentum (default: 0.5)')
parser.add_argument('--no-cuda', action='store_true', default=False,
                    help='disables CUDA training')
parser.add_argument('--seed', type=int, default=1, metavar='S',
                    help='random seed (default: 1)')
parser.add_argument('--log-interval', type=int, default=10, metavar='N',
                    help='how many batches to wait before logging training status')
args = parser.parse_args()
args.cuda = not args.no_cuda and torch.cuda.is_available()
 
torch.manual_seed(args.seed)
if args.cuda:
    torch.cuda.manual_seed(args.seed)
kwargs = {'num_workers': 1, 'pin_memory': True} if args.cuda else {}
train_loader = torch.utils.data.DataLoader(
    datasets.MNIST('../data', train=True, download=True,
                   transform=transforms.Compose([
                       transforms.ToTensor(),
                       transforms.Normalize((0.1307,), (0.3081,))
                   ])),
    batch_size=args.batch_size, shuffle=True, **kwargs)
test_loader = torch.utils.data.DataLoader(
    datasets.MNIST('../data', train=False, transform=transforms.Compose([
                       transforms.ToTensor(),
                       transforms.Normalize((0.1307,), (0.3081,))
                   ])),
    batch_size=args.test_batch_size, shuffle=True, **kwargs)
class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 10, kernel_size=5)
        self.conv2 = nn.Conv2d(10, 20, kernel_size=5)
        self.conv2_drop = nn.Dropout2d()
        self.fc1 = nn.Linear(320, 50)
        self.fc2 = nn.Linear(50, 10)
    def forward(self, x):
        print (x.size())
        x = F.relu(F.max_pool2d(self.conv1(x), 2))
        print(x.size())
        x = F.relu(F.max_pool2d(self.conv2_drop(self.conv2(x)), 2))
        print(x.size())
        x = x.view(-1, 320)
        x = F.relu(self.fc1(x))
        x = F.dropout(x, training=self.training)
        x = self.fc2(x)
        return F.log_softmax(x, dim=1)
 model = Net()                                 #调用写好的网络
if args.cuda:                                 #如果有GPU使用CPU
    model.cuda()
optimizer = optim.SGD(model.parameters(), lr=args.lr, momentum=args.momentum)   #设置SGD随机梯度下降算法
def train(epoch):
    model.train()
    for batch_idx, (data, target) in enumerate(train_loader):
        if args.cuda:
            data, target = data.cuda(), target.cuda()
        data, target = Variable(data), Variable(target)
        optimizer.zero_grad()                                  #梯度初始化为O
        output = model(data)
        loss = F.nll_loss(output, target)                      #简历loss function
        loss.backward()                                        #反向传播,计算梯度
        optimizer.step()                                       #更新权重
        if batch_idx % args.log_interval == 0:                 #输出信息
            print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
                epoch, batch_idx * len(data), len(train_loader.dataset),
                100. * batch_idx / len(train_loader), loss.data[0]))
def test():
    model.eval()
    test_loss = 0
    correct = 0
    for data, target in test_loader:
        if args.cuda:
            data, target = data.cuda(), target.cuda()
        data, target = Variable(data, volatile=True), Variable(target)
        output = model(data)
        test_loss += F.nll_loss(output, target, size_average=False).data[0] # sum up batch loss
        pred = output.data.max(1, keepdim=True)[1] # get the index of the max log-probability
        correct += pred.eq(target.data.view_as(pred)).long().cpu().sum()
 
    test_loss /= len(test_loader.dataset)
    print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
        test_loss, correct, len(test_loader.dataset),
        100. * correct / len(test_loader.dataset)))
for epoch in range(1, args.epochs + 1):
    train(epoch)
    test()

以上就是pytorch教程实现mnist手写数字识别代码示例的详细内容,更多关于pytorch实现mnist手写数字识别的资料请关注服务器之家其它相关文章!

原文链接:https://blog.csdn.net/xz1308579340/article/details/79520969