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

batch_size = 32
block_size = 128
max_iters = 1000
learning_rate = 3e-4
eval_steps = 200
n_embd = 384
n_head = 4
n_layer = 4
dropout = 0.2


class Block(nn.Module):
    """Transformer block: communication followed by computation"""

    def __init__(self, n_embd, n_head):
        # n_embd: embedding dimension, n_head: the number of heads we'd like
        super().__init__()
        head_size = n_embd // n_head
        self.sa = MultiHeadAttention(n_head, head_size)
        self.ffwd = FeedFoward(n_embd)
        self.ln1 = nn.LayerNorm(n_embd)
        self.ln2 = nn.LayerNorm(n_embd)

    def forward(self, x):
        y = self.sa(x)
        x = self.ln1(x + y)
        y = self.ffwd(x)
        x = self.ln2(x + y)
        return x


class MultiHeadAttention(nn.Module):
    """multiple heads of self-attention in parallel"""

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

    def forward(self, x):
        # (B, T, F) -> (B, T, [h1, h1, h1, h1, h2, h2, h2, h2, h3, h3, h3, h3])
        out = torch.cat([h(x) for h in self.heads], dim=-1)
        out = self.dropout(self.proj(out))
        return out


class Head(nn.Module):
    """one head of self-attention"""

    def __init__(self, head_size):
        super().__init__()
        self.key = nn.Linear(n_embd, head_size, bias=False)
        self.query = nn.Linear(n_embd, head_size, bias=False)
        self.value = nn.Linear(n_embd, head_size, bias=False)
        self.register_buffer(
            "tril", torch.tril(torch.ones(block_size, block_size))
        )  # noqa 5501

        self.dropout = nn.Dropout(dropout)

    def forward(self, x):
        # input of size (batch, time-step, channels)
        # output of size (batch, time-step, head size)
        B, T, C = x.shape
        k = self.key(x)  # (B,T,hs)
        q = self.query(x)  # (B,T,hs)
        # compute attention scores ("affinities")
        wei = (
            q @ k.transpose(-2, -1) * k.shape[-1] ** -0.5
        )  # (B, T, hs) @ (B, hs, T) -> (B, T, T) # noqa 5501
        wei = wei.masked_fill(
            self.tril[:T, :T] == 0, float("-inf")
        )  # (B, T, T) # noqa 5501
        wei = F.softmax(wei, dim=-1)  # (B, T, T)
        wei = self.dropout(wei)
        # perform the weighted aggregation of the values
        v = self.value(x)  # (B,T,hs)
        out = wei @ v  # (B, T, T) @ (B, T, hs) -> (B, T, hs)
        return out


class FeedFoward(nn.Module):
    """Simple linear layer followed by non_linear layer"""

    def __init__(self, n_embd):
        super().__init__()
        self.net = nn.Sequential(
            nn.Linear(n_embd, 4 * n_embd),
            nn.ReLU(),
            nn.Linear(4 * n_embd, n_embd),
            nn.Dropout(dropout),
        )

    def forward(self, x):
        return self.net(x)


class GPTLanguageModel(nn.Module):
    def __init__(self, vocab_size, device):
        super().__init__()
        self.device = device
        self.token_embedding_table = nn.Embedding(vocab_size, n_embd)
        self.position_embedding_table = nn.Embedding(block_size, n_embd)
        self.blocks = nn.Sequential(
            *[Block(n_embd, n_head=n_head) for _ in range(n_layer)]
        )  # noqa 5501
        self.ln_f = nn.LayerNorm(n_embd)
        self.lm_head = nn.Linear(n_embd, vocab_size)

        self.apply(self._init_weights)

    def _init_weights(self, module):
        if isinstance(module, nn.Linear):
            torch.nn.init.normal_(module.weight, mean=0.0, std=0.02)
            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, index, targets=None):
        B, T = index.shape

        # idx and targets are both (B,T) tensor of integers
        tok_emb = self.token_embedding_table(index)  # (B,T,C)
        pos_emb = self.position_embedding_table(
            torch.arange(T, device=self.device)
        )  # (T,C) # noqa 5501
        x = tok_emb + pos_emb  # (B,T,C)
        x = self.blocks(x)  # (B,T,C)
        x = self.ln_f(x)  # (B,T,C)
        logits = self.lm_head(x)  # (B,T,vocab_size)

        if targets is None:
            loss = None
        else:
            B, T, C = logits.shape
            logits = logits.view(B * T, C)
            targets = targets.view(B * T)
            loss = F.cross_entropy(logits, targets)

        return logits, loss

    def generate(self, index, max_new_tokens):
        # index is (B, T) array of indices in the current context
        for _ in range(max_new_tokens):
            # crop idx to the last block_size tokens
            index_cond = index[:, -block_size:]
            # get the predictions
            logits, loss = self.forward(index_cond)
            # focus only on the last time step
            logits = logits[:, -1, :]  # becomes (B, C)
            # apply softmax to get probabilities
            probs = F.softmax(logits, dim=-1)  # (B, C)
            # sample from the distribution
            index_next = torch.multinomial(probs, num_samples=1)  # (B, 1)
            # append sampled index to the running sequence
            index = torch.cat((index, index_next), dim=1)  # (B, T+1)
        return index


def create_GPT_model(vocab_size, device):
    model = GPTLanguageModel(vocab_size=vocab_size, device=device)
    model = model.to(device)
    return model