《PyTorch深度学习实践》第十课（卷积神经网络CNN）

b站刘二视频，地址：

《PyTorch深度学习实践》完结合集_哔哩哔哩_bilibili

CNN模型

 卷积运算

 卷积核运算

黄色方块为fliter（卷积核n  * 3* 3）,要想输出通道数为m，需要m个卷积核

 

import torch
in_channel, out_channel = 5, 10 #输入通道数，输出通道数（图层数）
width, height = 100, 100 #输入一张图层的大小
kernel_size = 3 #卷积核的大小(3 * 3)
batch_size = 1

input = torch.randn(batch_size, in_channel, width, height)
                    # B           N          W       H
conv_layer = torch.nn.Conv2d(in_channel, out_channel, kernel_size=kernel_size)
                    #            N           M              (3 * 3)

output = conv_layer(input)

print(input.shape)
print(output.shape)
print(conv_layer.weight.shape)

#输出结果
# torch.Size([1, 5, 100, 100])
        # batch大小 通道数 一个图层的大小
# torch.Size([1, 10, 98, 98])
# torch.Size([10, 5, 3, 3])
        #10个卷积核  每个卷积核有5个通道  卷积核大小为3 * 3

padding

保持输出图像大小不变，进行零填充

 stride

跳一格扫描

 maxpooling最大池化层

 网络整体

 

作业，手写MNIST识别

import torch
from torchvision import datasets
from torchvision import transforms
from torch.utils.data import DataLoader
import torch.nn.functional as F
import matplotlib.pyplot as plt

BATCH_SIZE = 64
transforms = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))])
train_set = datasets.MNIST(download=False, root='mnist', train=True, transform=transforms)
train_loader = DataLoader(dataset=train_set, batch_size=BATCH_SIZE, shuffle=True)
test_set = datasets.MNIST(download=False, root='mnist', train=False, transform=transforms)
test_loader = DataLoader(dataset=test_set, batch_size=BATCH_SIZE, shuffle=False)

class Net(torch.nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = torch.nn.Conv2d(1, 10, kernel_size=5)
        self.conv2 = torch.nn.Conv2d(10, 20, kernel_size=5)
        self.pooling = torch.nn.MaxPool2d(2)
        self.fc = torch.nn.Linear(320, 10)

    def forward(self, x):
        batch_size = x.size(0)
        x = F.relu(self.pooling(self.conv1(x)))
        x = F.relu(self.pooling(self.conv2(x)))
        x = x.view(batch_size, -1)
        x = self.fc(x)
        return x

modle = Net()

criteration = torch.nn.CrossEntropyLoss()
optimizor = torch.optim.SGD(modle.parameters(), lr=0.01, momentum=0.5)

def train():
    sum = 0
    for i, data in enumerate(train_loader, 0):
        inputs, lables = data
        y_pred = modle(inputs)
        loss = criteration(y_pred, lables)
        sum += loss

        optimizor.zero_grad()
        loss.backward()
        optimizor.step()

        if(i % 300 == 299):
            sum /= 300
            loss_lst.append(sum)
            sum = 0

def test():
    correct = 0
    totall = 0
    with torch.no_grad():
        for i, data in enumerate(test_loader, 0):
            inputs, lables = data
            y_pred = modle(inputs)
            _, predicted = torch.max(y_pred, dim=-1)
            correct += (lables == predicted).sum().item()
            totall += lables.size(0)

        acc_lst.append(correct / totall * 100)

if __name__ == '__main__':
    loss_lst = []
    acc_lst = []

    for epoch in range(10):
        train()
        test()

    num_lst = [i for i in range(len(loss_lst))]
    plt.plot(num_lst, loss_lst)
    plt.xlabel("i")
    plt.ylabel("loss")
    plt.show()

    num_lst = [i for i in range(len(acc_lst))]
    plt.plot(num_lst, acc_lst)
    plt.xlabel("epoch")
    plt.ylabel("acc")
    plt.show()