循环神经网络 (RNN) 代码示例：入门指南

当然可以！以下是一个使用循环神经网络 (RNN) 的简单代码示例，使用 PyTorch 框架实现：

import torch
import torch.nn as nn

# 定义循环神经网络模型
class RNN(nn.Module):
    def __init__(self, input_size, hidden_size, output_size):
        super(RNN, self).__init__()

        self.hidden_size = hidden_size

        self.i2h = nn.Linear(input_size + hidden_size, hidden_size)
        self.i2o = nn.Linear(input_size + hidden_size, output_size)
        self.softmax = nn.LogSoftmax(dim=1)

    def forward(self, input, hidden):
        combined = torch.cat((input, hidden), 1)
        hidden = self.i2h(combined)
        output = self.i2o(combined)
        output = self.softmax(output)
        return output, hidden

    def init_hidden(self):
        return torch.zeros(1, self.hidden_size)

# 定义参数
input_size = 10
hidden_size = 20
output_size = 5

# 创建模型实例
rnn = RNN(input_size, hidden_size, output_size)

# 定义损失函数
criterion = nn.NLLLoss()

# 定义优化器
learning_rate = 0.01
optimizer = torch.optim.SGD(rnn.parameters(), lr=learning_rate)

# 定义训练数据
input_sequence = [torch.randn(1, input_size) for _ in range(100)]
target_sequence = [torch.LongTensor([torch.randint(output_size, (1,)).item()]) for _ in range(100)]

# 训练模型
for input, target in zip(input_sequence, target_sequence):
    hidden = rnn.init_hidden()

    rnn.zero_grad()

    output = None
    for i in range(input.shape[0]):
        output, hidden = rnn(input[i], hidden)

    loss = criterion(output, target)
    loss.backward()

    optimizer.step()

# 使用模型进行预测
input = torch.randn(1, input_size)
hidden = rnn.init_hidden()
output = None
for i in range(input.shape[0]):
    output, hidden = rnn(input[i], hidden)

predicted_class = torch.argmax(output).item()
print('预测的类别为:', predicted_class)

这是一个简单的循环神经网络模型，你可以根据你的具体需求进行修改和扩展。请注意，这只是一个基本示例，并未完全展示循环神经网络的全部功能。