注意力机制加gru代码

以下是注意力机制加GRU的代码示例：

import torch
import torch.nn as nn

class AttentionGRU(nn.Module):
    def __init__(self, input_size, hidden_size, num_layers=1, dropout=0.0):
        super(AttentionGRU, self).__init__()
        self.hidden_size = hidden_size
        self.num_layers = num_layers
        self.dropout = dropout

        self.gru = nn.GRU(input_size, hidden_size, num_layers, batch_first=True, dropout=dropout, bidirectional=True)
        self.attention = nn.Linear(hidden_size * 2, 1)

    def forward(self, input):
        # input shape: (batch_size, seq_len, input_size)
        batch_size, seq_len, input_size = input.size()

        # output shape: (batch_size, seq_len, hidden_size * 2)
        output, hidden = self.gru(input)

        # attention_weights shape: (batch_size, seq_len, 1)
        attention_weights = self.attention(output).squeeze(-1)
        attention_weights = torch.softmax(attention_weights, dim=-1)

        # context shape: (batch_size, hidden_size * 2)
        context = torch.sum(output * attention_weights.unsqueeze(-1), dim=1)

        return context, hidden

希望这个代码示例能够帮助你！