File size: 1,605 Bytes
ccff75d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 |
# evo_model.py — Defines EvoDecoderModel used in inference and training
import torch
import torch.nn as nn
import math
class PositionalEncoding(nn.Module):
def __init__(self, d_model, max_len=128):
super().__init__()
pe = torch.zeros(max_len, d_model)
position = torch.arange(0, max_len).unsqueeze(1)
div_term = torch.exp(torch.arange(0, d_model, 2) * (-math.log(10000.0) / d_model))
pe[:, 0::2] = torch.sin(position * div_term)
pe[:, 1::2] = torch.cos(position * div_term)
pe = pe.unsqueeze(0) # [1, max_len, d_model]
self.register_buffer('pe', pe)
def forward(self, x):
return x + self.pe[:, :x.size(1)]
class EvoDecoderModel(nn.Module):
def __init__(self, vocab_size, d_model=512, nhead=8, num_layers=6, dim_ff=2048, max_len=128):
super().__init__()
self.token_embed = nn.Embedding(vocab_size, d_model)
self.pos_encoder = PositionalEncoding(d_model, max_len)
decoder_layer = nn.TransformerDecoderLayer(d_model, nhead, dim_ff, batch_first=True)
self.decoder = nn.TransformerDecoder(decoder_layer, num_layers=num_layers)
self.lm_head = nn.Linear(d_model, vocab_size)
def generate_square_subsequent_mask(self, sz):
return torch.triu(torch.full((sz, sz), float('-inf')), diagonal=1)
def forward(self, input_ids):
x = self.token_embed(input_ids)
x = self.pos_encoder(x)
tgt_mask = self.generate_square_subsequent_mask(x.size(1)).to(x.device)
x = self.decoder(x, x, tgt_mask=tgt_mask)
return self.lm_head(x)
|