"""A deliberately tiny GPT-style language model for CPU experiments."""
from __future__ import annotations

import torch
import torch.nn as nn
from torch.nn import functional as F


class TinyGPTConfig:
    def __init__(
        self,
        vocab_size: int,
        block_size: int = 64,
        n_layer: int = 2,
        n_head: int = 2,
        n_embd: int = 64,
        dropout: float = 0.1,
    ):
        self.vocab_size = vocab_size
        self.block_size = block_size
        self.n_layer = n_layer
        self.n_head = n_head
        self.n_embd = n_embd
        self.dropout = dropout


class CausalSelfAttention(nn.Module):
    def __init__(self, cfg: TinyGPTConfig):
        super().__init__()
        assert cfg.n_embd % cfg.n_head == 0
        self.n_head = cfg.n_head
        self.head_dim = cfg.n_embd // cfg.n_head
        self.qkv = nn.Linear(cfg.n_embd, 3 * cfg.n_embd)
        self.proj = nn.Linear(cfg.n_embd, cfg.n_embd)
        self.dropout = nn.Dropout(cfg.dropout)
        self.register_buffer(
            "mask",
            torch.tril(torch.ones(cfg.block_size, cfg.block_size)).view(1, 1, cfg.block_size, cfg.block_size),
            persistent=False,
        )

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        b, t, c = x.shape
        q, k, v = self.qkv(x).split(c, dim=2)
        q = q.view(b, t, self.n_head, self.head_dim).transpose(1, 2)
        k = k.view(b, t, self.n_head, self.head_dim).transpose(1, 2)
        v = v.view(b, t, self.n_head, self.head_dim).transpose(1, 2)

        att = (q @ k.transpose(-2, -1)) * (self.head_dim ** -0.5)
        att = att.masked_fill(self.mask[:, :, :t, :t] == 0, float("-inf"))
        att = F.softmax(att, dim=-1)
        att = self.dropout(att)
        y = att @ v
        y = y.transpose(1, 2).contiguous().view(b, t, c)
        return self.dropout(self.proj(y))


class Block(nn.Module):
    def __init__(self, cfg: TinyGPTConfig):
        super().__init__()
        self.ln1 = nn.LayerNorm(cfg.n_embd)
        self.attn = CausalSelfAttention(cfg)
        self.ln2 = nn.LayerNorm(cfg.n_embd)
        self.mlp = nn.Sequential(
            nn.Linear(cfg.n_embd, 4 * cfg.n_embd),
            nn.GELU(),
            nn.Linear(4 * cfg.n_embd, cfg.n_embd),
            nn.Dropout(cfg.dropout),
        )

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        x = x + self.attn(self.ln1(x))
        x = x + self.mlp(self.ln2(x))
        return x


class TinyGPT(nn.Module):
    def __init__(self, cfg: TinyGPTConfig):
        super().__init__()
        self.cfg = cfg
        self.token_embedding = nn.Embedding(cfg.vocab_size, cfg.n_embd)
        self.position_embedding = nn.Embedding(cfg.block_size, cfg.n_embd)
        self.drop = nn.Dropout(cfg.dropout)
        self.blocks = nn.Sequential(*[Block(cfg) for _ in range(cfg.n_layer)])
        self.ln_f = nn.LayerNorm(cfg.n_embd)
        self.head = nn.Linear(cfg.n_embd, cfg.vocab_size, bias=False)

        # Weight tying: common in GPT-style LMs.
        self.head.weight = self.token_embedding.weight
        self.apply(self._init_weights)

    def _init_weights(self, module: nn.Module) -> None:
        if isinstance(module, nn.Linear):
            nn.init.normal_(module.weight, mean=0.0, std=0.02)
            if module.bias is not None:
                nn.init.zeros_(module.bias)
        elif isinstance(module, nn.Embedding):
            nn.init.normal_(module.weight, mean=0.0, std=0.02)

    def forward(self, idx: torch.Tensor, targets: torch.Tensor | None = None):
        b, t = idx.shape
        if t > self.cfg.block_size:
            raise ValueError(f"sequence length {t} > block_size {self.cfg.block_size}")
        pos = torch.arange(0, t, device=idx.device)
        x = self.token_embedding(idx) + self.position_embedding(pos)[None, :, :]
        x = self.drop(x)
        x = self.blocks(x)
        x = self.ln_f(x)
        logits = self.head(x)
        loss = None
        if targets is not None:
            loss = F.cross_entropy(logits.view(-1, logits.size(-1)), targets.view(-1))
        return logits, loss

    @torch.no_grad()
    def generate(self, idx: torch.Tensor, max_new_tokens: int, temperature: float = 0.8, top_k: int | None = None):
        for _ in range(max_new_tokens):
            idx_cond = idx[:, -self.cfg.block_size :]
            logits, _ = self(idx_cond)
            logits = logits[:, -1, :] / max(temperature, 1e-6)
            if top_k is not None:
                v, _ = torch.topk(logits, min(top_k, logits.size(-1)))
                logits[logits < v[:, [-1]]] = -float("inf")
            probs = F.softmax(logits, dim=-1)
            next_idx = torch.multinomial(probs, num_samples=1)
            idx = torch.cat((idx, next_idx), dim=1)
        return idx