layers.py

# -*- coding: utf-8 -*-
"""
Created on Tue Mar 12 13:08:32 2024

@author: Mels
"""
import torch
import torch.nn as nn
from torch.nn import functional as F

#%% Head
class Head(nn.Module):
    """
    Simple verion of the Attention mechanism as described in the Attention is all you need paper"
    Can be added together for the Multi-Headed Attention mechanism. 
    """
    def __init__(self, n_embd, head_size, block_size, dropout):
        super().__init__()
        self.key = nn.Linear(n_embd, head_size, bias=False)
        self.query = nn.Linear(n_embd, head_size, bias=False)
        self.value = nn.Linear(n_embd, head_size, bias=False)
        self.register_buffer('tril', torch.tril(torch.ones(block_size, block_size)))

        self.dropout = nn.Dropout(dropout)

    def forward(self, x):
        B,T,C = x.shape
        k = self.key(x)
        q = self.query(x)
        
        # compute attention scores ("affinities")
        wei = q @ k.transpose(-2,-1) * C**-0.5 # (B, T, C) @ (B, C, T) -> (B, T, T)
        wei = wei.masked_fill(self.tril[:T, :T] == 0, float('-inf')) # (B, T, T)
        wei = F.softmax(wei, dim=-1) # (B, T, T)
        wei = self.dropout(wei)
        
        # perform the weighted aggregation of the values
        v = self.value(x) # (B,T,C)
        out = wei @ v # (B, T, T) @ (B, T, C) -> (B, T, C)
        return out


#%% MultiHeadAttention
class MultiHeadAttention(nn.Module):
    """
    Adds together multiple Heads to form the multi head attention 
    """

    def __init__(self, n_heads, n_embd, head_size, block_size, dropout):
        super().__init__()
        self.heads = nn.ModuleList([Head(n_embd=n_embd, head_size=head_size, 
                                         block_size=block_size, dropout=dropout) for _ in range(n_heads)])
        #self.proj = nn.Linear(n_embd, n_embd)
        self.proj = nn.Linear(n_heads * head_size, n_embd)
        self.dropout = nn.Dropout(dropout)

    def forward(self, x):
        out = torch.cat([h(x) for h in self.heads], dim=-1)
        out = self.dropout(self.proj(out))
        return out


#%% FeedFoward
class FeedFoward(nn.Module):
    """ a simple linear layer followed by a non-linearity """

    def __init__(self, n_embd, dropout):
        super().__init__()
        self.net = nn.Sequential(
            nn.Linear(n_embd, 4 * n_embd),
            nn.ReLU(),
            nn.Linear(4 * n_embd, n_embd),
            nn.Dropout(dropout),
        )

    def forward(self, x):
        return self.net(x)