pytorch用rnn模型拟合sin函数

可以使用PyTorch中的RNN模型来拟合sin函数。以下是一个简单的例子：

import torch
import torch.nn as nn
import matplotlib.pyplot as plt
import numpy as np

# 定义模型
class RNN(nn.Module):
    def __init__(self, input_size, hidden_size, output_size):
        super(RNN, self).__init__()
        self.hidden_size = hidden_size
        self.rnn = nn.RNN(input_size, hidden_size, batch_first=True)
        self.fc = nn.Linear(hidden_size, output_size)

    def forward(self, x, hidden):
        out, hidden = self.rnn(x, hidden)
        out = self.fc(out)
        return out, hidden

# 训练模型
input_size = 1
hidden_size = 32
output_size = 1
sequence_length = 20
num_epochs = 1000
learning_rate = 0.01

rnn = RNN(input_size, hidden_size, output_size)
criterion = nn.MSELoss()
optimizer = torch.optim.Adam(rnn.parameters(), lr=learning_rate)

hidden = None
for epoch in range(num_epochs):
    # 生成随机序列
    inputs = torch.from_numpy(np.random.uniform(-np.pi, np.pi, size=(1, sequence_length, input_size))).float()
    outputs = torch.sin(inputs)

    # 前向传播
    rnn.zero_grad()
    hidden = hidden.detach() if hidden is not None else None
    outputs, hidden = rnn(inputs, hidden)
    loss = criterion(outputs, outputs)

    # 反向传播
    optimizer.zero_grad()
    loss.backward()
    optimizer.step()

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

# 测试模型
with torch.no_grad():
    inputs = torch.from_numpy(np.linspace(-np.pi, np.pi, num=100).reshape(1, -1, 1)).float()
    outputs = torch.sin(inputs)
    hidden = None
    preds, hidden = rnn(inputs, hidden)
    plt.plot(inputs.squeeze().numpy(), outputs.squeeze().numpy(), label='target')
    plt.plot(inputs.squeeze().numpy(), preds.squeeze().numpy(), label='prediction')
    plt.legend()
    plt.show()

这个例子中，我们使用一个有一个隐藏层的RNN模型来拟合sin函数。我们生成随机序列作为输入，并将其作为sin函数的输出进行训练。在1000个epoch之后，我们使用训练好的模型来预测sin函数，并将其可视化。