pytorch 手写数字图片识别代码

1、训练文件

import torch
from torch import nn
from torch.nn import functional as F
from torch import optim
import torchvision

import matplotlib.pyplot as plt


# 辅助函数

def plot_curve(data):
    fig = plt.figure()
    plt.plot(range(len(data)), data, color='blue')
    plt.legend(['loss'], loc='upper right')
    plt.xlabel('step')
    plt.ylabel('loss')
    plt.show()


def plot_image(img, label, name):
    fig = plt.figure()
    for i in range(6):
        plt.subplot(2, 3, i + 1)
        plt.tight_layout()
        plt.imshow(img[i][0] * 0.3081 + 0.1307, cmap='gray', interpolation='none')
        plt.title('{}:{}'.format(name, label[i].item()))
        plt.xticks([])
        plt.yticks([])
    plt.show()

def one_hot(label, depth=10):
    out = torch.zeros(label.size(0), depth)
    idx = torch.LongTensor(label).view(-1, 1)
    out.scatter_(dim=1, index=idx, value=1)
    return out


# Step1.
# 加载数据

batch_size = 64
train_loader = torch.utils.data.DataLoader(torchvision.datasets.MNIST('/tmp/pycharm_project_225/datasets/mnist_data/',
                train=True,
                download=True,
                transform=torchvision.transforms.Compose([
                torchvision.transforms.ToTensor(),                       # 数据类型转化
                torchvision.transforms.Normalize((0.1307, ), (0.3081, )) # 数据归一化处理
    ])), batch_size=batch_size,shuffle=True)

test_loader = torch.utils.data.DataLoader(torchvision.datasets.MNIST('/tmp/pycharm_project_225/datasets/mnist_data/',
                train=False,
                download=True,
                transform=torchvision.transforms.Compose([
                torchvision.transforms.ToTensor(),
                torchvision.transforms.Normalize((0.1307, ), (0.3081, ))
    ])),batch_size=batch_size,shuffle=False)

x,y = next(iter(train_loader))
print(x.shape,y.shape)

plot_image(x,y,'image smaple')


# Step2.
# 构建网络

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.fc1 = nn.Linear(28 * 28, 256)
        self.fc2 = nn.Linear(256, 64)
        self.fc3 = nn.Linear(64, 10)

    def forward(self, x):
        # img : [batch,1,28,28]
        # x : [batch,784]
        # h1 = relu(xw1 + b1)
        x = F.relu(self.fc1(x))
        # h2 = relu(h1w2 + b2)
        x = F.relu(self.fc2(x))
        # h3 = h2w3 + b3
        x = self.fc3(x)

        return x


# step3.
# 训练训练样本

net = Net()

# [w1,b1,w2,b2,w3,b3]
optimizer = optim.SGD(net.parameters(), lr=0.01, momentum=0.9)

train_loss = []
for epoch in range(4):
    for batch_idx, (x, y) in enumerate(train_loader):
        # x : [batch,1,28,28] y:[batch]
        # flatten [batch,1,28,28] - > [batch,feature]
        # x : [batch,784]
        x = x.view(x.size(0), 28 * 28)
        out = net(x)

        # one-hot
        y_onehot = one_hot(y)

        # loss : mse(out,y_onehot)
        loss = F.mse_loss(out, y_onehot)

        # 清零梯度
        optimizer.zero_grad()

        # 计算梯度
        loss.backward()

        # 更新梯度
        # w' = w - lr * grad
        optimizer.step()

        train_loss.append(loss.item())

        if batch_idx % 10 == 0:
            print(f"epoch:{epoch} step:{batch_idx} loss:{loss.item()}")

# %%

plot_curve(train_loss)

# %%

# Step4.
# 测试集验证

total_correct = 0
for x, y in test_loader:
    x = x.view(x.size(0), 28 * 28)
    out = net(x)
    # out : [batch,10] - > pred : [batch]
    y_pred = out.argmax(dim=1)
    correct = y_pred.eq(y).sum().float().item()
    total_correct += correct
total_num = len(test_loader.dataset)
print(f"accuracy : {total_correct / total_num}")

# %%

x, y = next(iter(test_loader))
out = net(x.view(x.size(0), 28 * 28))
pred = out.argmax(dim=1)
plot_image(x, pred, 'test')

# %%

torch.save(net.state_dict(),"mnist.pth")

2、测试一张图片

import torch
from torch import nn
from torch.nn import functional as F
import numpy as np
import cv2

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.fc1 = nn.Linear(28 * 28, 256)
        self.fc2 = nn.Linear(256, 64)
        self.fc3 = nn.Linear(64, 10)

    def forward(self, x):
        # img : [batch,1,28,28]
        # x : [batch,784]
        # h1 = relu(xw1 + b1)
        x = F.relu(self.fc1(x))
        # h2 = relu(h1w2 + b2)
        x = F.relu(self.fc2(x))
        # h3 = h2w3 + b3
        x = self.fc3(x)
        return x

net = Net()
net.load_state_dict(torch.load("mnist.pth"))

img = cv2.imread("./3.png")  #读取要预测的图片  分辨率为28*28
img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)#图片转为灰度图，因为mnist数据集都是灰度图
img=np.array(img).astype(np.float32)
img=np.expand_dims(img,0)
img=np.expand_dims(img,0)#扩展后，为[1，1，28，28]
img=torch.from_numpy(img)
x = img.view(img.size(0), 28 * 28)
out = net(x)
# out : [batch,10] - > pred : [batch]
y_pred = out.argmax(dim=1)
print(y_pred)