源碼：github code
pytorch簡(jiǎn)單神經(jīng)網(wǎng)絡(luò)手寫(xiě)數(shù)字識(shí)別

import torch
import torch.nn as nn
import torchvision
import torchvision.transforms as transforms

# Device configuration
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

# Hyper-parameters
# 超參設(shè)置
input_size = 784
hidden_size = 500 ##有一個(gè)500個(gè)神經(jīng)元的隱藏層神經(jīng)網(wǎng)絡(luò)
num_classes = 10
num_epochs = 5
batch_size = 100
learning_rate = 0.001

# MNIST dataset
# 手寫(xiě)數(shù)字識(shí)別數(shù)據(jù)下載
train_dataset = torchvision.datasets.MNIST(root='../../data',
                                           train=True,
                                           transform=transforms.ToTensor(),
                                           download=True)

test_dataset = torchvision.datasets.MNIST(root='../../data',
                                          train=False,
                                          transform=transforms.ToTensor())

# Data loader
# 數(shù)據(jù)加載
train_loader = torch.utils.data.DataLoader(dataset=train_dataset,
                                           batch_size=batch_size,
                                           shuffle=True)

test_loader = torch.utils.data.DataLoader(dataset=test_dataset,
                                          batch_size=batch_size,
                                          shuffle=False)


# Fully connected neural network with one hidden layer
# 全連接網(wǎng)絡(luò)模型自定義
class NeuralNet(nn.Module):
    def __init__(self, input_size, hidden_size, num_classes):
        super(NeuralNet, self).__init__()
        self.fc1 = nn.Linear(input_size, hidden_size)
        self.relu = nn.ReLU()
        self.fc2 = nn.Linear(hidden_size, num_classes)

    def forward(self, x):
        out = self.fc1(x)
        out = self.relu(out)
        out = self.fc2(out)
        return out

# 模型初始化
model = NeuralNet(input_size, hidden_size, num_classes).to(device)

# Loss and optimizer
# 交叉熵?fù)p失函數(shù)和優(yōu)化器
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

# Train the model
# 模型訓(xùn)練
total_step = len(train_loader)
for epoch in range(num_epochs):
    for i, (images, labels) in enumerate(train_loader):
        # Move tensors to the configured device
        images = images.reshape(-1, 28 * 28).to(device)
        labels = labels.to(device)

        # Forward pass
        outputs = model(images)
        loss = criterion(outputs, labels)

        # Backward and optimize
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        if (i + 1) % 100 == 0:
            print('Epoch [{}/{}], Step [{}/{}], Loss: {:.4f}'
                  .format(epoch + 1, num_epochs, i + 1, total_step, loss.item()))

# Test the model
# 模型測(cè)試
# In test phase, we don't need to compute gradients (for memory efficiency)
with torch.no_grad():
    correct = 0
    total = 0
    for images, labels in test_loader:
        images = images.reshape(-1, 28 * 28).to(device)
        labels = labels.to(device)
        outputs = model(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: {} %'.format(100 * correct / total))

# Save the model checkpoint
# 模型保存
torch.save(model.state_dict(), 'model.ckpt')