architecture.py

import torch
import torch.nn as nn
from torch.nn import functional as F
from transformers import GPT2LMHeadModel, GPT2Config as HF_GPT2Config

def get_device():
    if torch.cuda.is_available():
        return torch.device("cuda")
    try:
        import intel_extension_for_pytorch as ipex
        if torch.xpu.is_available():
            return torch.device("xpu")
    except:
        try:
            if torch.xpu.is_available():
                return torch.device("xpu")
        except:
            pass
    try:
        import torch_directml
        return torch_directml.device()
    except:
        pass
    return torch.device("cpu")

DEVICE = get_device()

class GPTConfig:
    def __init__(self, block_size=256, n_embd=384, n_head=6, n_layer=6, dropout=0.0, vocab_size=100277, model_mode='scratch'):
        self.block_size = block_size
        self.n_embd = n_embd
        self.n_head = n_head
        self.n_layer = n_layer
        self.dropout = dropout
        self.vocab_size = vocab_size
        self.model_mode = model_mode

class Head(nn.Module):
    def __init__(self, config, head_size):
        super().__init__()
        self.key = nn.Linear(config.n_embd, head_size, bias=False)
        self.query = nn.Linear(config.n_embd, head_size, bias=False)
        self.value = nn.Linear(config.n_embd, head_size, bias=False)
        self.register_buffer('tril', torch.tril(torch.ones(config.block_size, config.block_size)))
        self.dropout = nn.Dropout(config.dropout)
    def forward(self, x):
        B, T, C = x.shape
        k = self.key(x)
        q = self.query(x)
        wei = q @ k.transpose(-2, -1) * C**-0.5
        wei = wei.masked_fill(self.tril[:T, :T] == 0, float('-inf'))
        wei = F.softmax(wei, dim=-1)
        wei = self.dropout(wei)
        v = self.value(x)
        out = wei @ v
        return out

class MultiHeadAttention(nn.Module):
    def __init__(self, config, head_size):
        super().__init__()
        self.heads = nn.ModuleList([Head(config, head_size) for _ in range(config.n_head)])
        self.proj = nn.Linear(config.n_head * head_size, config.n_embd)
        self.dropout = nn.Dropout(config.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

class FeedForward(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.net = nn.Sequential(
            nn.Linear(config.n_embd, 4 * config.n_embd),
            nn.GELU(),
            nn.Linear(4 * config.n_embd, config.n_embd),
            nn.Dropout(config.dropout),
        )
    def forward(self, x):
        return self.net(x)

class Block(nn.Module):
    def __init__(self, config):
        super().__init__()
        head_size = config.n_embd // config.n_head
        self.sa = MultiHeadAttention(config, head_size)
        self.ffwd = FeedForward(config)
        self.ln1 = nn.LayerNorm(config.n_embd)
        self.ln2 = nn.LayerNorm(config.n_embd)
    def forward(self, x):
        x = x + self.sa(self.ln1(x))
        x = x + self.ffwd(self.ln2(x))
        return x

class ScratchGPT(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.config = config
        self.token_embedding_table = nn.Embedding(config.vocab_size, config.n_embd)
        self.position_embedding_table = nn.Embedding(config.block_size, config.n_embd)
        self.blocks = nn.Sequential(*[Block(config) for _ in range(config.n_layer)])
        self.ln_f = nn.LayerNorm(config.n_embd)
        self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)
        self.token_embedding_table.weight = self.lm_head.weight
    def forward(self, idx, targets=None):
        B, T = idx.shape
        tok_emb = self.token_embedding_table(idx)
        pos_emb = self.position_embedding_table(torch.arange(T, device=DEVICE))
        x = tok_emb + pos_emb
        x = self.blocks(x)
        x = self.ln_f(x)
        logits = self.lm_head(x)
        if targets is not None:
            loss = F.cross_entropy(logits.view(-1, logits.size(-1)), targets.view(-1))
        else:
            logits = logits[:, [-1], :]
            loss = None
        return logits, loss
    @torch.no_grad()
    def generate(self, idx, max_new_tokens, temperature=1.0):
        for _ in range(max_new_tokens):
            idx_cond = idx if idx.size(1) <= self.config.block_size else idx[:, -self.config.block_size:]
            logits, _ = self(idx_cond)
            logits = logits[:, -1, :] / (temperature + 1e-9)
            probs = F.softmax(logits, dim=-1)
            idx_next = torch.multinomial(probs, num_samples=1)
            idx = torch.cat((idx, idx_next), dim=1)
        return idx

class FineTuneGPT(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.config = config
        if config.model_mode == 'gpt2_pretrained':
            self.model = GPT2LMHeadModel.from_pretrained('gpt2')
        else:
            hf_config = HF_GPT2Config(
                n_embd=config.n_embd, n_layer=config.n_layer, n_head=config.n_head,
                vocab_size=config.vocab_size, n_positions=config.block_size
            )
            self.model = GPT2LMHeadModel(hf_config)
    def forward(self, idx, targets=None):
        outputs = self.model(input_ids=idx, labels=targets)
        return outputs.logits, outputs.loss
    @torch.no_grad()
    def generate(self, idx, max_new_tokens, temperature=1.0):
        return self.model.generate(
            input_ids=idx,
            max_new_tokens=max_new_tokens,
            temperature=temperature,
            do_sample=True,
            pad_token_id=50256
        )