深度学习算法中的长短期记忆网络（Long Short-Term Memory）

pythonCopy codeimport torch
import torch.nn as nn
# 定义LSTM模型
class LSTMModel(nn.Module):
    def __init__(self, input_size, hidden_size, num_layers, output_size):
        super(LSTMModel, 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, output_size)
    def forward(self, x):
        h0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(x.device)
        c0 = torch.zeros(self.num_layers, x.size(0), self.hidden_size).to(x.device)
        out, _ = self.lstm(x, (h0, c0))
        out = self.fc(out[:, -1, :])
        return out
# 定义输入数据
input_size = 10
hidden_size = 32
num_layers = 2
output_size = 1
batch_size = 16
sequence_length = 20
x = torch.randn(batch_size, sequence_length, input_size)
# 创建LSTM模型
model = LSTMModel(input_size, hidden_size, num_layers, output_size)
# 前向传播
output = model(x)
print(output.shape)

pythonCopy codeimport torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt
# 生成sin函数的训练数据
def generate_data(seq_length):
    time = np.linspace(0, 10, seq_length+1)
    data = np.sin(time)
    x = data[:-1]
    y = data[1:]
    return x, y
# 数据预处理
def preprocess_data(x, y):
    x = torch.from_numpy(x).unsqueeze(1).float()
    y = torch.from_numpy(y).unsqueeze(1).float()
    return x, y
# 定义LSTM模型
class LSTMModel(nn.Module):
    def __init__(self, input_size, hidden_size, output_size):
        super(LSTMModel, self).__init__()
        self.hidden_size = hidden_size
        self.lstm = nn.LSTM(input_size, hidden_size, batch_first=True)
        self.fc = nn.Linear(hidden_size, output_size)
    def forward(self, x):
        out, _ = self.lstm(x)
        out = self.fc(out[:, -1, :])
        return out
# 设置训练参数
input_size = 1
hidden_size = 32
output_size = 1
num_epochs = 100
learning_rate = 0.01
# 生成训练数据
seq_length = 100
x_train, y_train = generate_data(seq_length)
x_train, y_train = preprocess_data(x_train, y_train)
# 创建LSTM模型
model = LSTMModel(input_size, hidden_size, output_size)
# 定义损失函数和优化器
criterion = nn.MSELoss()
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)
# 训练模型
for epoch in range(num_epochs):
    model.train()
    outputs = model(x_train)
    loss = criterion(outputs, y_train)
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()
    if (epoch+1) % 10 == 0:
        print(f'Epoch [{epoch+1}/{num_epochs}], Loss: {loss.item():.4f}')
# 测试模型
model.eval()
x_test, y_test = generate_data(seq_length)
x_test, y_test = preprocess_data(x_test, y_test)
with torch.no_grad():
    predictions = model(x_test)
# 可视化结果
plt.plot(y_test.numpy(), label='Ground Truth')
plt.plot(predictions.numpy(), label='Predictions')
plt.xlabel('Time')
plt.ylabel('Value')
plt.legend()
plt.show()