184 lines
7.2 KiB
Python
184 lines
7.2 KiB
Python
import torch
|
|
import torch.nn as nn
|
|
import threading
|
|
import os
|
|
import json
|
|
|
|
|
|
# Simple Tokenizer
|
|
class SimpleTokenizer:
|
|
def __init__(self):
|
|
self.token2idx = {'<PAD>': 0, '<UNK>': 1, '<SOS>': 2, '<EOS>': 3}
|
|
self.idx2token = {idx: token for token, idx in self.token2idx.items()}
|
|
self.lock = threading.Lock()
|
|
|
|
def build_vocab(self, texts):
|
|
with self.lock:
|
|
for text in texts:
|
|
tokens = text.split()
|
|
for token in tokens:
|
|
if token not in self.token2idx:
|
|
idx = len(self.token2idx)
|
|
self.token2idx[token] = idx
|
|
self.idx2token[idx] = token
|
|
|
|
def encode(self, text):
|
|
with self.lock:
|
|
return [self.token2idx.get(token, self.token2idx['<UNK>']) for token in text.split()]
|
|
|
|
def decode(self, indices):
|
|
with self.lock:
|
|
return ' '.join([self.idx2token.get(idx, '<UNK>') for idx in indices])
|
|
|
|
def save_vocab(self, path):
|
|
with open(path, 'w') as f:
|
|
json.dump({'token2idx': self.token2idx, 'idx2token': self.idx2token}, f)
|
|
|
|
def load_vocab(self, path):
|
|
if os.path.exists(path):
|
|
with open(path, 'r') as f:
|
|
vocab = json.load(f)
|
|
self.token2idx = vocab['token2idx']
|
|
self.idx2token = {int(k): v for k, v in vocab['idx2token'].items()}
|
|
print('Tokenizer vocabulary loaded from', path)
|
|
# Ensure special tokens are present
|
|
special_tokens = {'<PAD>': 0, '<UNK>': 1, '<SOS>': 2, '<EOS>': 3}
|
|
for token, idx in special_tokens.items():
|
|
if token not in self.token2idx:
|
|
self.token2idx[token] = idx
|
|
self.idx2token[idx] = token
|
|
else:
|
|
print('No existing tokenizer vocabulary found. Starting fresh.')
|
|
self.token2idx = {'<PAD>': 0, '<UNK>': 1, '<SOS>': 2, '<EOS>': 3}
|
|
self.idx2token = {idx: token for token, idx in self.token2idx.items()}
|
|
|
|
|
|
# Positional Encoding
|
|
class PositionalEncoding(nn.Module):
|
|
def __init__(self, d_model, max_len=5000):
|
|
super(PositionalEncoding, self).__init__()
|
|
pe = torch.zeros(max_len, d_model)
|
|
position = torch.arange(0, max_len).unsqueeze(1).float()
|
|
div_term = torch.exp(torch.arange(0, d_model, 2).float() * (-torch.log(torch.tensor(10000.0)) / d_model))
|
|
pe[:, 0::2] = torch.sin(position * div_term)
|
|
if d_model % 2 == 1:
|
|
pe[:, -1] = torch.cos(position.squeeze() * div_term[-1])
|
|
else:
|
|
pe[:, 1::2] = torch.cos(position * div_term)
|
|
pe = pe.unsqueeze(1) # Shape: [max_len, 1, d_model]
|
|
self.register_buffer('pe', pe)
|
|
|
|
def forward(self, x):
|
|
# x: [seq_len, batch_size, d_model]
|
|
x = x + self.pe[:x.size(0)]
|
|
return x
|
|
|
|
|
|
# GPT Model
|
|
class GPTModel(nn.Module):
|
|
def __init__(self, vocab_size, d_model=128, nhead=8, num_layers=2):
|
|
super(GPTModel, self).__init__()
|
|
self.model_type = 'Transformer'
|
|
self.d_model = d_model
|
|
self.embedding = nn.Embedding(vocab_size, d_model)
|
|
self.pos_encoder = PositionalEncoding(d_model)
|
|
encoder_layers = nn.TransformerEncoderLayer(d_model, nhead)
|
|
self.transformer_encoder = nn.TransformerEncoder(encoder_layers, num_layers)
|
|
self.fc_out = nn.Linear(d_model, vocab_size)
|
|
self.src_mask = None
|
|
|
|
def _generate_square_subsequent_mask(self, sz):
|
|
mask = torch.triu(torch.ones(sz, sz) == 1).transpose(0, 1)
|
|
mask = mask.float().masked_fill(mask == 0, float('-inf')).masked_fill(mask == 1, float(0.0))
|
|
return mask
|
|
|
|
def forward(self, src):
|
|
# src: [seq_len, batch_size]
|
|
src = src.transpose(0, 1) # Shape: [seq_len, batch_size]
|
|
src = self.embedding(src) * torch.sqrt(torch.tensor(self.d_model, dtype=torch.float32))
|
|
src = self.pos_encoder(src)
|
|
if self.src_mask is None or self.src_mask.size(0) != src.size(0):
|
|
device = src.device
|
|
self.src_mask = self._generate_square_subsequent_mask(src.size(0)).to(device)
|
|
output = self.transformer_encoder(src, self.src_mask)
|
|
logits = self.fc_out(output)
|
|
return logits # Shape: [seq_len, batch_size, vocab_size]
|
|
|
|
|
|
# Training function
|
|
def train_step(model, optimizer, criterion, input_tensor, target_tensor):
|
|
model.train()
|
|
optimizer.zero_grad()
|
|
output = model(input_tensor) # [seq_len, batch_size, vocab_size]
|
|
output = output.view(-1, output.size(-1)) # [seq_len * batch_size, vocab_size]
|
|
target = target_tensor.transpose(0,1).contiguous().view(-1) # [seq_len * batch_size]
|
|
loss = criterion(output, target)
|
|
loss.backward()
|
|
optimizer.step()
|
|
print(f'Training loss: {loss.item():.4f}')
|
|
return loss.item()
|
|
|
|
|
|
def initialize_model(tokenizer, device):
|
|
vocab_size = len(tokenizer.token2idx)
|
|
model = GPTModel(vocab_size=vocab_size).to(device)
|
|
optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
|
|
criterion = nn.CrossEntropyLoss(ignore_index=0)
|
|
model_path = 'gpt_model.pth'
|
|
|
|
# Load existing model if available
|
|
if os.path.exists(model_path):
|
|
model.load_state_dict(torch.load(model_path, map_location=device))
|
|
print('Model loaded from', model_path)
|
|
else:
|
|
print('No existing model found. Starting fresh.')
|
|
return model, optimizer, criterion
|
|
|
|
|
|
def save_model(model):
|
|
model_path = 'gpt_model.pth'
|
|
torch.save(model.state_dict(), model_path)
|
|
|
|
|
|
def update_model_vocab(model, tokenizer, device):
|
|
vocab_size = len(tokenizer.token2idx)
|
|
|
|
old_embedding_weight = model.embedding.weight.data
|
|
old_vocab_size, embedding_dim = old_embedding_weight.shape
|
|
new_embedding = nn.Embedding(vocab_size, model.d_model).to(device)
|
|
new_embedding.weight.data[:old_vocab_size] = old_embedding_weight
|
|
model.embedding = new_embedding
|
|
|
|
old_fc_out_weight = model.fc_out.weight.data
|
|
old_fc_out_bias = model.fc_out.bias.data
|
|
new_fc_out = nn.Linear(model.d_model, vocab_size).to(device)
|
|
new_fc_out.weight.data[:old_vocab_size] = old_fc_out_weight
|
|
new_fc_out.bias.data[:old_vocab_size] = old_fc_out_bias
|
|
model.fc_out = new_fc_out
|
|
|
|
return model
|
|
|
|
|
|
def train_on_conversation(model, optimizer, criterion, tokenizer, input_text, target_text, device):
|
|
tokenizer.build_vocab([input_text, target_text])
|
|
input_indices = tokenizer.encode(input_text)
|
|
target_indices = tokenizer.encode(target_text)
|
|
|
|
# Concatenate input and target indices to create a single sequence
|
|
full_indices = input_indices + target_indices
|
|
|
|
# Create input and target sequences for training
|
|
input_sequence = full_indices[:-1] # All tokens except the last
|
|
target_sequence = full_indices[1:] # All tokens except the first
|
|
|
|
# Update model if vocabulary has changed
|
|
if len(tokenizer.token2idx) != model.embedding.num_embeddings:
|
|
model = update_model_vocab(model, tokenizer, device)
|
|
optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
|
|
|
|
input_tensor = torch.tensor([input_sequence], dtype=torch.long, device=device)
|
|
target_tensor = torch.tensor([target_sequence], dtype=torch.long, device=device)
|
|
|
|
loss = train_step(model, optimizer, criterion, input_tensor, target_tensor)
|
|
return loss
|