"""
src/model.py — GPT-style transformer. You write this.

Target architecture (nanoGPT-style):
  - CausalSelfAttention: multi-head attention with causal mask
  - MLP: two linear layers with GELU activation
  - TransformerBlock: attention + MLP with residual connections and LayerNorm
  - GPT: embedding + N blocks + final projection to vocab

Reference: https://github.com/karpathy/nanoGPT/blob/master/model.py
Paper: "Attention Is All You Need" (Vaswani et al., 2017) — Sections 3.1–3.4

Start here:
  1. Define a GPTConfig dataclass (n_layer, n_head, n_embd, vocab_size, block_size)
  2. Write CausalSelfAttention.__init__ and .forward
  3. Write MLP
  4. Write TransformerBlock
  5. Write GPT with token + position embeddings, stack of blocks, LM head

Sanity check (paste into a scratch script when done):
  config = GPTConfig(n_layer=6, n_head=6, n_embd=384, vocab_size=50257, block_size=1024)
  model = GPT(config)
  x = torch.randint(0, 50257, (4, 1024))
  logits, loss = model(x, x)
  print(logits.shape)  # expect: torch.Size([4, 1024, 50257])
  print(sum(p.numel() for p in model.parameters()) / 1e6, "M params")  # expect ~85M
"""

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


@dataclass
class GPTConfig:
    block_size: int = 1024
    vocab_size: int = 50257
    n_layer: int = 12
    n_head: int = 12
    n_embd: int = 768
    dropout: float = 0.0
    bias: bool = True


# --- Write your code below ---
class CausalSelfAttention(nn.Module):
    def __init__(self, config):
        super().__init__()
        assert config.n_embd % config.n_head == 0
        self.c_attn = nn.Linear(config.n_embd, 3 * config.n_embd, bias=config.bias)
        self.c_proj = nn.Linear(config.n_embd, config.n_embd, bias=config.bias)
        self.attn_dropout = nn.Dropout(config.dropout)
        self.resid_dropout = nn.Dropout(config.dropout)
        self.n_head = config.n_head
        self.n_embd = config.n_embd
        self.dropout = config.dropout
        self.flash = hasattr(torch.nn.functional, 'scaled_dot_product_attention')
        if not self.flash:
            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()
        q, k, v = self.c_attn(x).split(self.n_embd, dim=2)
        k = k.view(B, T, self.n_head, C // self.n_head).transpose(1, 2)
        q = q.view(B, T, self.n_head, C // self.n_head).transpose(1, 2)
        v = v.view(B, T, self.n_head, C // self.n_head).transpose(1, 2)

        if self.flash:
            y = torch.nn.functional.scaled_dot_product_attention(
                    q, k, v, attn_mask=None,
                    dropout_p=self.dropout if self.training else 0,
                    is_causal=True)
        else:
            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)
            att = self.attn_dropout(att)
            y = att @ v
        y = y.transpose(1, 2).contiguous().view(B, T, C)
        y = self.resid_dropout(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, bias=config.bias)
        self.gelu    = nn.GELU()
        self.c_proj  = nn.Linear(4 * config.n_embd, config.n_embd, bias=config.bias)
        self.dropout = nn.Dropout(config.dropout)
    def forward(self, x):
        x = self.c_fc(x)
        x = self.gelu(x)
        x = self.c_proj(x)
        x = self.dropout(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
    

class GPT(nn.Module):

    def __init__(self, config):
        super().__init__()
        self.transformer = nn.ModuleDict(dict(
            wte  = nn.Embedding(config.vocab_size, config.n_embd),
            wpe  = nn.Embedding(config.block_size, config.n_embd),
            drop = nn.Dropout(config.dropout),
            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)
    def forward(self, idx, targets=None):
        b, t = idx.size()
        pos = torch.arange(0, t, dtype=torch.long, device=idx.device)

        tok_emb = self.transformer.wte(idx)
        pos_emb = self.transformer.wpe(pos)
        x = self.transformer.drop(tok_emb + pos_emb)
        for block in self.transformer.h:
            x = block(x)
        x = self.transformer.ln_f(x)

        if targets is not None:
            logits = self.lm_head(x)
            loss = F.cross_entropy(logits.view(-1, logits.size(-1)), targets.view(-1))
        else:
            logits = self.lm_head(x[:, [-1], :])
            loss = None

        return logits, loss

    def configure_optimizers(self, weight_decay, learning_rate, betas, device_type):
        # weight decay only on 2D params (weights), not biases or layernorm
        decay = {p for n, p in self.named_parameters() if p.requires_grad and p.dim() >= 2}
        no_decay = {p for n, p in self.named_parameters() if p.requires_grad and p.dim() < 2}
        groups = [{"params": list(decay), "weight_decay": weight_decay},
                  {"params": list(no_decay), "weight_decay": 0.0}]
        return torch.optim.AdamW(groups, lr=learning_rate, betas=betas)