【PyTorch】yunjey/pytorch-tutorial——2.5-循環神經網絡

專欄【PyTorch】

原文鏈接：https://github.com/yunjey/pytorch-tutorial

# 導入需要的庫
import torch
import torch.nn as nn
import torchvision
import torchvision.transforms as transforms

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

# Hyper-parameters
# 超參數
sequence_length = 28
input_size = 28
hidden_size = 128
num_layers = 2
num_classes = 10
batch_size = 100
num_epochs = 2
learning_rate = 0.01

# MNIST dataset
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
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)

# Recurrent neural network (many-to-one)
# 循環神經網絡（多對一）
class RNN(nn.Module):
    def __init__(self, input_size, hidden_size, num_layers, num_classes):
        super(RNN, self).__init__()
        self.hidden_size = hidden_size
        self.num_layers = num_layers
        self.lstm = nn.LSTM(
            input_size,
            hidden_size,
            num_layers,
            batch_first=True)
        self.fc = nn.Linear(hidden_size, num_classes)

    def forward(self, x):
        # Set initial hidden and cell states
        # 設置初始隱藏和單元格狀態
        h0 = torch.zeros(
            self.num_layers,
            x.size(0),
            self.hidden_size).to(device)
        c0 = torch.zeros(
            self.num_layers,
            x.size(0),
            self.hidden_size).to(device)

        # Forward propagate LSTM
        # 前向傳播LSTM
        # out: tensor of shape (batch_size, seq_length, hidden_size)
        out, _ = self.lstm(x, (h0, c0))

        # Decode the hidden state of the last time step
        # 解碼最後一個時間步驟的隱藏狀態
        out = self.fc(out[:, -1, :])
        return out


model = RNN(input_size, hidden_size, num_layers, num_classes).to(device)

# Loss and optimizer
# 損失和優化器
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

# Train the model
# 訓練模型
total_step = len(train_loader)
for epoch in range(num_epochs):
    for i, (images, labels) in enumerate(train_loader):
        images = images.reshape(-1, sequence_length, input_size).to(device)
        labels = labels.to(device)

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

        # Backward and optimize
        # 反向傳播和優化
        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
# 測試模型
with torch.no_grad():
    correct = 0
    total = 0
    for images, labels in test_loader:
        images = images.reshape(-1, sequence_length, input_size).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(
        'Test Accuracy of the model on the 10000 test images: {} %'.format(
            100 *
            correct /
            total))

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

代碼

Epoch [1/2], Step [100/600], Loss: 0.6361
Epoch [1/2], Step [200/600], Loss: 0.3237
Epoch [1/2], Step [300/600], Loss: 0.3490
Epoch [1/2], Step [400/600], Loss: 0.0732
Epoch [1/2], Step [500/600], Loss: 0.1803
Epoch [1/2], Step [600/600], Loss: 0.1077
Epoch [2/2], Step [100/600], Loss: 0.0433
Epoch [2/2], Step [200/600], Loss: 0.1484
Epoch [2/2], Step [300/600], Loss: 0.1480
Epoch [2/2], Step [400/600], Loss: 0.1173
Epoch [2/2], Step [500/600], Loss: 0.0247
Epoch [2/2], Step [600/600], Loss: 0.0960
Test Accuracy of the model on the 10000 test images: 98.07 %