# model.py
import torch
import torch.nn as nn
import math

class SelfAttention(nn.Module):
    def __init__(self, embed_dim, num_heads):
        super().__init__()
        assert embed_dim % num_heads == 0
        self.head_dim = embed_dim // num_heads
        self.num_heads = num_heads

        self.query = nn.Linear(embed_dim, embed_dim)
        self.key = nn.Linear(embed_dim, embed_dim)
        self.value = nn.Linear(embed_dim, embed_dim)
        self.out_proj = nn.Linear(embed_dim, embed_dim)

    def forward(self, x):
        B, T, C = x.size()
        q = self.query(x).view(B, T, self.num_heads, self.head_dim).transpose(1, 2)  # (B, heads, T, head_dim)
        k = self.key(x).view(B, T, self.num_heads, self.head_dim).transpose(1, 2)
        v = self.value(x).view(B, T, self.num_heads, self.head_dim).transpose(1, 2)

        scores = (q @ k.transpose(-2, -1)) / math.sqrt(self.head_dim)  # (B, heads, T, T)
        mask = torch.tril(torch.ones(T, T)).to(x.device)
        scores = scores.masked_fill(mask == 0, float('-inf'))
        attn = torch.softmax(scores, dim=-1)

        out = attn @ v  # (B, heads, T, head_dim)
        out = out.transpose(1, 2).contiguous().view(B, T, C)
        return self.out_proj(out)

class TransformerBlock(nn.Module):
    def __init__(self, embed_dim, num_heads):
        super().__init__()
        self.attn = SelfAttention(embed_dim, num_heads)
        self.ln1 = nn.LayerNorm(embed_dim)
        self.ff = nn.Sequential(
            nn.Linear(embed_dim, embed_dim * 4),
            nn.GELU(),
            nn.Linear(embed_dim * 4, embed_dim)
        )
        self.ln2 = nn.LayerNorm(embed_dim)

    def forward(self, x):
        x = x + self.attn(self.ln1(x))
        x = x + self.ff(self.ln2(x))
        return x

class TinyTransformer(nn.Module):
    def __init__(self, vocab_size, max_len, embed_dim=128, num_heads=2, num_layers=1):
        super().__init__()
        self.token_embed = nn.Embedding(vocab_size, embed_dim)
        self.pos_embed = nn.Parameter(torch.zeros(1, max_len, embed_dim))
        self.blocks = nn.ModuleList([
            TransformerBlock(embed_dim, num_heads) for _ in range(num_layers)
        ])
        self.ln_final = nn.LayerNorm(embed_dim)
        self.head = nn.Linear(embed_dim, vocab_size)

    def forward(self, x):
        B, T = x.size()
        tok_emb = self.token_embed(x)           # (B, T, C)
        pos_emb = self.pos_embed[:, :T, :]      # (1, T, C)
        x = tok_emb + pos_emb                   # (B, T, C)

        for block in self.blocks:
            x = block(x)

        x = self.ln_final(x)
        logits = self.head(x)                   # (B, T, vocab_size)
        return logits