Pytorch框架学校记录11——搭建小实战完整细节

1. 搭建小实战和Sequential的使用

我们搭建了一个CIFAR10模型，下面的代码是未使用Sequential的情况。

import torch
from torch import nn


class Test(nn.Module):
    def __init__(self):
        super(Test, self).__init__()
        self.conv1 = nn.Conv2d(in_channels=3, out_channels=32, kernel_size=5, padding=2)
        self.maxpool1 = nn.MaxPool2d(kernel_size=2)
        self.conv2 = nn.Conv2d(in_channels=32, out_channels=32, kernel_size=5, padding=2)
        self.maxpool2 = nn.MaxPool2d(kernel_size=2)
        self.conv3 = nn.Conv2d(in_channels=32, out_channels=64, kernel_size=5, padding=2)
        self.maxpool3 = nn.MaxPool2d(kernel_size=2)
        self.flatten = nn.Flatten()
        self.hidden = nn.Linear(in_features=1024, out_features=64)
        self.fc = nn.Linear(in_features=64, out_features=10)

    def forward(self, input):
        input = self.conv1(input)
        input = self.maxpool1(input)
        input = self.conv2(input)
        input = self.maxpool2(input)
        input = self.conv3(input)
        input = self.maxpool3(input)

        input = self.flatten(input)
        input = self.hidden(input)
        output = self.fc(input)
        return output


x = torch.tensor([0.1, 0.2, 0.3])
print(x.shape)
y = torch.tensor([1])
x = torch.reshape(x, (1, 3))
print(x)
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

2. 损失函数和反向传播

torch.nn.CrossEntropyLoss(weight=None, size_average=None, ignore_index=- 100, reduce=None, reduction=‘mean’, label_smoothing=0.0)

参数：

• Input: Shape ©(C), (N, C)(N,C) or (N, C, d_1, d_2, …, d_K)(N,C,d1,d2,…,d**K) with K \geq 1K≥1 in the case of K-dimensional loss.
• Target: If containing class indices, shape ()(), (N)(N) or (N, d_1, d_2, …, d_K)(N,d1,d2,…,d**K) with K \geq 1K≥1 in the case of K-dimensional loss where each value should be between [0, C)[0,C). If containing class probabilities, same shape as the input and each value should be between [0, 1][0,1].

在这里我们以交叉熵损失函数为例，backward()方法为反向传播算法。

from torch import nn
import torchvision
from torch.utils.data import DataLoader

dataset = torchvision.datasets.CIFAR10(root='dataset', train=False, transform=torchvision.transforms.ToTensor(), download=True)

dataloader = DataLoader(dataset, 64)


class Test(nn.Module):
    def __init__(self):
        super(Test, self).__init__()
        self.conv1 = nn.Conv2d(in_channels=3, out_channels=32, kernel_size=5, padding=2)
        self.maxpool1 = nn.MaxPool2d(kernel_size=2)
        self.conv2 = nn.Conv2d(in_channels=32, out_channels=32, kernel_size=5, padding=2)
        self.maxpool2 = nn.MaxPool2d(kernel_size=2)
        self.conv3 = nn.Conv2d(in_channels=32, out_channels=64, kernel_size=5, padding=2)
        self.maxpool3 = nn.MaxPool2d(kernel_size=2)
        self.flatten = nn.Flatten()
        self.hidden = nn.Linear(in_features=1024, out_features=64)
        self.fc = nn.Linear(in_features=64, out_features=10)

    def forward(self, input):
        input = self.conv1(input)
        input = self.maxpool1(input)
        input = self.conv2(input)
        input = self.maxpool2(input)
        input = self.conv3(input)
        input = self.maxpool3(input)

        input = self.flatten(input)
        input = self.hidden(input)
        output = self.fc(input)
        return output


test = Test()
loss = nn.CrossEntropyLoss()
step = 0
for data in dataloader:
    imgs, target = data
    output = test(imgs)
    res = loss(output, target)
    res.backward()
    print(res)
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

3. 优化器

优化器的作用：将模型的中的参数根据要求进行实时调整更新，使得模型变得更加优良。

在这里我们使用的是随机梯度下降法（SGD）作为优化器的优化依据。

from torch import nn
import torchvision
import torch
from torch.utils.data import DataLoader

dataset = torchvision.datasets.CIFAR10(root='dataset', train=False, transform=torchvision.transforms.ToTensor(), download=True)

dataloader = DataLoader(dataset, 64)


class Test(nn.Module):
    def __init__(self):
        super(Test, self).__init__()
        self.conv1 = nn.Conv2d(in_channels=3, out_channels=32, kernel_size=5, padding=2)
        self.maxpool1 = nn.MaxPool2d(kernel_size=2)
        self.conv2 = nn.Conv2d(in_channels=32, out_channels=32, kernel_size=5, padding=2)
        self.maxpool2 = nn.MaxPool2d(kernel_size=2)
        self.conv3 = nn.Conv2d(in_channels=32, out_channels=64, kernel_size=5, padding=2)
        self.maxpool3 = nn.MaxPool2d(kernel_size=2)
        self.flatten = nn.Flatten()
        self.hidden = nn.Linear(in_features=1024, out_features=64)
        self.fc = nn.Linear(in_features=64, out_features=10)

    def forward(self, input):
        input = self.conv1(input)
        input = self.maxpool1(input)
        input = self.conv2(input)
        input = self.maxpool2(input)
        input = self.conv3(input)
        input = self.maxpool3(input)

        input = self.flatten(input)
        input = self.hidden(input)
        output = self.fc(input)
        return output


test = Test()
loss = nn.CrossEntropyLoss()
step = 0
optimer = torch.optim.SGD(params=test.parameters(), lr=0.01)
for epoch in range(20):
    loss_sum = 0.0
    for data in dataloader:
        imgs, target = data
        output = test(imgs)
        res = loss(output, target)
        optimer.zero_grad()
        res.backward()
        optimer.step()
        loss_sum += res
    print(loss_sum)
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

4. 现有网络模型的使用及修改

我们使用Pytorch框架中的VGG16模型，并将该模型的全连接层的输出特征的个数设置为10，

torchvision.models.vgg16(***, weights: Optional[torchvision.models.vgg.VGG16_Weights] = None, progress: bool = True, **kwargs: Any)

参数：

• pretrained：设置为True代表加载预训练模型

对现有网络模型的修改可分为两种方式，一种为添加，另一种为修改。

详细的操作方法如下：

import torch
from torch import nn
import torchvision

vgg_true = torchvision.models.vgg16(pretrained=True)
vgg_false = torchvision.models.vgg16(pretrained=False)


vgg_true.add_module("add_model", nn.Linear(in_features=1000, out_features=10))
print(vgg_true)
vgg_false.classifier[6] = nn.Linear(in_features=4096, out_features=10)
print(vgg_false)
1
2
3
4
5
6
7
8
9
10
11
12

5. 模型的保存与读取

模型的保存有两种方式：

import torchvision
import torch

vgg16 = torchvision.models.vgg16(pretrained=False)

# 保存方式1 模型结构+模型参数
torch.save(vgg16, "vgg16_methods1.pth")


# 保存方式2 模型参数（官方推荐）
torch.save(vgg16.state_dict(), "vgg16_methods2.pth")
1
2
3
4
5
6
7
8
9
10
11

模型的加载方式也有两种方式

import torch
import torchvision
# 方式1 加载模型
vgg16_model1 = torch.load("vgg16_methods1.pth")
print(vgg16_model1)

# 方式2 加载模型
vgg16 = torchvision.models.vgg16(pretrained=False)
vgg16.state_dict(torch.load("vgg16_methods2.pth"))
print(vgg16)
1
2
3
4
5
6
7
8
9
10