transformer.py

import math
import torch
import torch.nn as nn
import torch.nn.functional as F
from dataclasses import dataclass


class CausalSelfAttention(nn.Module):

    def __init__(self, config):
        super().__init__()
        assert config.n_embd % config.n_head == 0
        # key, query, value projections for all heads, but in a batch
        self.c_attn = nn.Linear(config.n_embd, 3 * config.n_embd)
        # output projection
        self.c_proj = nn.Linear(config.n_embd, config.n_embd)
        self.c_proj.NANGPT_SCALE_INIT = 1
        # regularization
        self.n_head = config.n_head
        self.n_embd = config.n_embd
        self.register_buffer("bias", torch.tril(torch.ones(config.block_size, config.block_size)).view(1, 1, config.block_size, config.block_size))

    def forward(self, x):
        B, T, C = x.size() # batch size, sequence length, embedding dimensionality (n_embd)
        # calculate query, key, values for all heads in batch and move head forward to be the batch dim
        # nh is "number of heads", hs is "head size", and C (number of channels) = nh * hs
        # e.g. in GPT-2 (124M), n_head=12, hs=64, so nh*hs=C=768 channels in the Transformer
        qkv = self.c_attn(x)
        q, k, v = qkv.split(self.n_embd, dim=2)
        k = k.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # (B, nh, T, hs)
        q = q.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # (B, nh, T, hs)
        v = v.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # (B, nh, T, hs)

        att = (q @ k.transpose(-2, -1)) * (1.0 / math.sqrt(k.size(-1)))
        att = att.masked_fill(self.bias[:, :, :T, :T] == 0, float('-inf'))
        att = F.softmax(att, dim=-1)
        y = att @ v # (B, nh, T, T) x (B, nh, T, hs) -> (B, nh, T, hs)

        y = y.transpose(1, 2).contiguous().view(B, T, C) # re-assemble all head outputs side by side
        # output projection
        y = self.c_proj(y)
        return y


class MLP(nn.Module):

    def __init__(self, config):
        super().__init__()
        self.c_fc    = nn.Linear(config.n_embd, 4 * config.n_embd)
        self.gelu    = nn.GELU(approximate='tanh')
        self.c_proj  = nn.Linear(4 * config.n_embd, config.n_embd)
        self.c_proj.NANOGPT_SCALE_INIT = 1

    def forward(self, x):
        x = self.c_fc(x)
        x = self.gelu(x)
        x = self.c_proj(x)
        return x

class Block(nn.Module):

    def __init__(self, config):
        super().__init__()
        self.ln_1 = nn.LayerNorm(config.n_embd)
        self.attn = CausalSelfAttention(config)
        self.ln_2 = nn.LayerNorm(config.n_embd)
        self.mlp = MLP(config)

    def forward(self, x):
        x = x + self.attn(self.ln_1(x))
        x = x + self.mlp(self.ln_2(x))
        return x


@dataclass
class GPTConfig:
    block_size: int = 1024 # max sequence length
    vocab_size: int = 50257 # number of tokens: 50,000 BPE merges + 256 bytes tokens + 1 <|endoftext|> token
    n_layer: int = 12 # number of layers
    n_head: int = 12 # number of heads
    n_embd: int = 768 # embedding dimension


class GPT(nn.Module):

    def __init__(self, config):
        super().__init__()
        self.config = config

        self.transformer = nn.ModuleDict(dict(
            wte = nn.Embedding(config.vocab_size, config.n_embd),
            wpe = nn.Embedding(config.block_size, config.n_embd),
            h = nn.ModuleList([Block(config) for _ in range(config.n_layer)]),
            ln_f = nn.LayerNorm(config.n_embd),
        ))
        self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)

        # weight sharing
        self.transformer.wte.weight = self.lm_head.weight

        # weight initialization
        self.apply(self._init_weights)

    def _init_weights(self, module):
        if isinstance(module, nn.Linear):
            std = 0.02
            if hasattr(module, 'NANGPT_SCALE_INIT'):
                std *= (2 * self.config.n_layer) ** -0.5
            torch.nn.init.normal_(module.weight, mean = 0.0, std = std)
            if module.bias is not None:
                torch.nn.init.zeros_(module.bias)
        elif isinstance(module, nn.Embedding):
            torch.nn.init.normal_(module.weight, mean=0.0, std = 0.02)



    def forward(self, idx, targets=None):
        # idx is of shape (B, T)
        B, T = idx.size()
        assert T <= self.config.block_size, f"Cannot forward sequence of length {T}, block size is only {self.config.block_size}"
        # forward the token and posisition embeddings
        pos = torch.arange(0, T, dtype=torch.long, device=idx.device) # shape (T)
        pos_emb = self.transformer.wpe(pos) # position embeddings of shape (T, n_embd)
        tok_emb = self.transformer.wte(idx) # token embeddings of shape (B, T, n_embd)
        x = tok_emb + pos_emb
        # forward the blocks of the transformer
        for block in self.transformer.h:
            x = block(x)
        # forward the final layernorm and the classifier
        x = self.transformer.ln_f(x)
        logits = self.lm_head(x) # (B, T, vocab_size)
        loss = None
        if targets is not None:
            loss = F.cross_entropy(logits.view(-1, logits.size(-1)), targets.view(-1))
        return logits, loss

    @classmethod
    def from_pretrained(cls, model_type):
        """Loads pretrained GPT-2 model weights from huggingface"""
        assert model_type in {'gpt2', 'gpt2-medium', 'gpt2-large', 'gpt2-xl'}
        from transformers import GPT2LMHeadModel
        print("loading weights from pretrained gpt: %s" % model_type)

        # n_layer, n_head and n_embd are determined from model_type
        config_args = {
            'gpt2':         dict(n_layer=12, n_head=12, n_embd=768),  # 124M params
            'gpt2-medium':  dict(n_layer=24, n_head=16, n_embd=1024), # 350M params
            'gpt2-large':   dict(n_layer=36, n_head=20, n_embd=1280), # 774M params
            'gpt2-xl':      dict(n_layer=48, n_head=25, n_embd=1600), # 1558M params
        }[model_type]
        config_args['vocab_size'] = 50257 # always 50257 for GPT model checkpoints
        config_args['block_size'] = 1024 # always 1024 for GPT model checkpoints
        # create a from-scratch initialized minGPT model
        config = GPTConfig(**config_args)
        model = GPT(config)
        sd = model.state_dict()
        sd_keys = sd.keys()
        sd_keys = [k for k in sd_keys if not k.endswith('.attn.bias')] # discard this mask / buffer, not a param

        # init a huggingface/transformers model
        model_hf = GPT2LMHeadModel.from_pretrained(model_type)
        sd_hf = model_hf.state_dict()

        # copy while ensuring all of the parameters are aligned and match in names and shapes
        sd_keys_hf = sd_hf.keys()
        sd_keys_hf = [k for k in sd_keys_hf if not k.endswith('.attn.masked_bias')] # ignore these, just a buffer
        sd_keys_hf = [k for k in sd_keys_hf if not k.endswith('.attn.bias')] # same, just the mask (buffer)
        transposed = ['attn.c_attn.weight', 'attn.c_proj.weight', 'mlp.c_fc.weight', 'mlp.c_proj.weight']
        # basically the openai checkpoints use a "Conv1D" module, but we only want to use a vanilla Linear
        # this means that we have to transpose these weights when we import them
        assert len(sd_keys_hf) == len(sd_keys), f"mismatched keys: {len(sd_keys_hf)} != {len(sd_keys)}"
        for k in sd_keys_hf:
            if any(k.endswith(w) for w in transposed):
                # special treatment for the Conv1D weights we need to transpose
                assert sd_hf[k].shape[::-1] == sd[k].shape
                with torch.no_grad():
                    sd[k].copy_(sd_hf[k].t())
            else:
                # vanilla copy over the other parameters
                assert sd_hf[k].shape == sd[k].shape
                with torch.no_grad():
                    sd[k].copy_(sd_hf[k])

        return model


@dataclass
class Config:
    vocab_size: int = 50257
    max_seq_len: int = 2048
    dim: int = 768
    num_layers: int = 12
    num_heads: int = 12
    dropout: float = 0.1

class MultiHeadAttention(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.config = config
        self.n_head = config.num_heads
        self.n_embd = config.dim
        
        # Linear projections for Q, K, V
        self.c_attn = nn.Linear(config.dim, 3 * config.dim) # [n_embd, 3 * n_embd]
        self.c_proj = nn.Linear(config.dim, config.dim) # [n_embd, n_embd]
        
        self.attn_dropout = nn.Dropout(config.dropout)
        self.resid_dropout = nn.Dropout(config.dropout)

    def forward(self, x):
        B, T, C = x.size() # [B, T, n_embd]
        
        # Linear projection and split into Q, K, V
        q, k, v = self.c_attn(x).split(self.n_embd, dim=2) # [B, T, n_embd] each
        
        # Reshape for multi-head attention
        k = k.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # [B, n_head, T, n_embd/n_head]
        q = q.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # [B, n_head, T, n_embd/n_head]
        v = v.view(B, T, self.n_head, C // self.n_head).transpose(1, 2) # [B, n_head, T, n_embd/n_head]
        
        # Attention scores
        att = (q @ k.transpose(-2, -1)) * (1.0 / (k.size(-1) ** 0.5)) # [B, n_head, T, T]
        att = F.softmax(att, dim=-1) # [B, n_head, T, T]
        att = self.attn_dropout(att) # [B, n_head, T, T]
        
        # Weighted sum of values
        y = att @ v # [B, n_head, T, n_embd/n_head]
        
        # Reshape and project
        y = y.transpose(1, 2).contiguous().view(B, T, C) # [B, T, n_embd]
        y = self.c_proj(y) # [B, T, n_embd]
        y = self.resid_dropout(y) # [B, T, n_embd]
        
        return y

class FeedForward(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.c_fc = nn.Linear(config.dim, 4 * config.dim) # [n_embd, 4 * n_embd]
        self.c_proj = nn.Linear(4 * config.dim, config.dim) # [4 * n_embd, n_embd]
        self.dropout = nn.Dropout(config.dropout)

    def forward(self, x):
        x = self.c_fc(x) # [B, T, 4 * n_embd]
        x = F.gelu(x) # [B, T, 4 * n_embd]
        x = self.c_proj(x) # [B, T, n_embd]
        x = self.dropout(x) # [B, T, n_embd]
        return x

class TransformerBlock(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.ln_1 = nn.LayerNorm(config.dim) # [n_embd]
        self.attn = MultiHeadAttention(config)
        self.ln_2 = nn.LayerNorm(config.dim) # [n_embd]
        self.mlp = FeedForward(config)

    def forward(self, x):
        x = x + self.attn(self.ln_1(x)) # [B, T, n_embd]
        x = x + self.mlp(self.ln_2(x)) # [B, T, n_embd]
        return x

class DecoderOnlyTransformer(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.config = config
        self.wte = nn.Embedding(config.vocab_size, config.dim) # [vocab_size, n_embd]
        self.wpe = nn.Embedding(config.max_seq_len, config.dim) # [max_seq_len, n_embd]
        self.drop = nn.Dropout(config.dropout)
        self.blocks = nn.ModuleList([TransformerBlock(config) for _ in range(config.num_layers)])
        self.ln_f = nn.LayerNorm(config.dim) # [n_embd]
        self.lm_head = nn.Linear(config.dim, config.vocab_size, bias=False) # [n_embd, vocab_size]
        
        self.apply(self._init_weights)

    def _init_weights(self, module):
        if isinstance(module, (nn.Linear, nn.Embedding)):
            module.weight.data.normal_(mean=0.0, std=0.02)
            if isinstance(module, nn.Linear) and module.bias is not None:
                module.bias.data.zero_()
        elif isinstance(module, nn.LayerNorm):
            module.bias.data.zero_()
            module.weight.data.fill_(1.0)

    def forward(self, idx):
        B, T = idx.size() # [B, T]
        
        # Positional embeddings
        pos = torch.arange(0, T, dtype=torch.long, device=idx.device).unsqueeze(0) # [1, T]
        
        # Token and position embeddings
        tok_emb = self.wte(idx) # [B, T, n_embd]
        pos_emb = self.wpe(pos) # [1, T, n_embd]
        
        # Combine embeddings and apply dropout
        x = self.drop(tok_emb + pos_emb) # [B, T, n_embd]
        
        # Transformer blocks
        for block in self.blocks:
            x = block(x) # [B, T, n_embd]
        
        # Final layer norm and linear projection
        x = self.ln_f(x) # [B, T, n_embd]
        logits = self.lm_head(x) # [B, T, vocab_size]
        
        return logits

# if __name__ == '__main__':
#     config = Config()
#     model = DecoderOnlyTransformer(config)
    
#     # Example usage
#     batch_size = 4
#     seq_len = 128
    
#     # Generate random input
#     input_ids = torch.randint(0, config.vocab_size, (batch_size, seq_len))
    
#     # Forward pass
#     logits = model(input_ids)
    
#     print("Input shape:", input_ids.shape)
#     print("Output shape:", logits.shape)