self-attention代码

### Self-Attention Mechanism Code Implementation

In deep learning, self-attention mechanisms play a crucial role in various models, especially those involving sequence data processing. Below is an example of how one can implement a simple version of the self-attention mechanism using Python and PyTorch[^1].

The following code snippet demonstrates creating a basic self-attention layer:

import torch
import torch.nn as nn
import math

class SelfAttention(nn.Module):
    def __init__(self, embed_size, heads):
        super(SelfAttention, self).__init__()
        self.embed_size = embed_size
        self.heads = heads
        self.head_dim = embed_size // heads
        
        assert (
            self.head_dim * heads == embed_size
        ), "Embedding size needs to be divisible by heads"
        
        self.values = nn.Linear(self.head_dim, embed_size, bias=False)
        self.keys = nn.Linear(self.head_dim, embed_size, bias=False)
        self.queries = nn.Linear(self.head_dim, embed_size, bias=False)
        self.fc_out = nn.Linear(embed_size, embed_size)

    def forward(self, values, keys, query, mask):
        N = query.shape[0]
        value_len, key_len, query_len = values.shape[1], keys.shape[1], query.shape[1]

        # Split embedding into self.heads pieces
        values = values.reshape(N, value_len, self.heads, self.head_dim)
        keys = keys.reshape(N, key_len, self.heads, self.head_dim)
        queries = query.reshape(N, query_len, self.heads, self.head_dim)

        energy = torch.einsum("nqhd,nkhd->nhqk", [queries, keys])

        if mask is not None:
            energy = energy.masked_fill(mask == 0, float("-1e20"))

        attention = torch.softmax(energy / (self.embed_size ** (1 / 2)), dim=3)

        out = torch.einsum("nhql,nlhd->nqhd", [attention, values]).reshape(
            N, query_len, self.embed_size
        )

        out = self.fc_out(out)
        return out

This implementation includes several components necessary for constructing a functional self-attention module within neural networks designed for natural language processing or other sequential data applications.