layer.py

import torch
import torch.nn as nn

from .sublayer import FeedForward, MultiHeadAttention, Norm


class EncoderLayer(nn.Module):
    def __init__(self, d_model, heads, dropout=0.1):
        super().__init__()
        self.norm_1 = Norm(d_model)
        self.norm_2 = Norm(d_model)
        self.attn = MultiHeadAttention(heads, d_model, dropout=dropout)
        self.ff = FeedForward(d_model, dropout=dropout)
        self.dropout_1 = nn.Dropout(dropout)
        self.dropout_2 = nn.Dropout(dropout)
        
    def forward(self, x, mask):
        x2 = self.norm_1(x)
        # encoder self-attention
        x = x + self.dropout_1(self.attn(x2, x2, x2, mask))
        x2 = self.norm_2(x)
        x = x + self.dropout_2(self.ff(x2))
        return x


class DecoderLayer(nn.Module):
    '''build a decoder layer with 2 multi-head attention layers and 1 feed-forward layer'''
    def __init__(self, d_model, heads, dropout=0.1):
        super().__init__()
        self.norm_1 = Norm(d_model)
        self.norm_2 = Norm(d_model)
        self.norm_3 = Norm(d_model)
        
        self.dropout_1 = nn.Dropout(dropout)
        self.dropout_2 = nn.Dropout(dropout)
        self.dropout_3 = nn.Dropout(dropout)
        
        self.attn_1 = MultiHeadAttention(heads, d_model, dropout=dropout)
        self.attn_2 = MultiHeadAttention(heads, d_model, dropout=dropout)
        self.ff = FeedForward(d_model, dropout=dropout)

    def forward(self, x, e_outputs, src_mask, trg_mask):
        x2 = self.norm_1(x)
        # decoder self-attention with target masking
        x = x + self.dropout_1(self.attn_1(x2, x2, x2, trg_mask))
        x2 = self.norm_2(x)
        # encoder-decoder attention
        # q is previous position decoder output, k and v is from encoder output
        x = x + self.dropout_2(self.attn_2(x2, e_outputs, e_outputs, src_mask))
        x2 = self.norm_3(x)
        x = x + self.dropout_3(self.ff(x2))
        return x