bark/model_fine.py

"""
Much of this code is adapted from Andrej Karpathy's NanoGPT
(https://github.com/karpathy/nanoGPT)
"""
from dataclasses import dataclass
import math

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

from .model import GPT, GPTConfig, MLP


class NonCausalSelfAttention(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, bias=config.bias)
        # output projection
        self.c_proj = nn.Linear(config.n_embd, config.n_embd, bias=config.bias)
        # regularization
        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
        # flash attention make GPU go brrrrr but support is only in PyTorch >= 2.0
        self.flash = (
            hasattr(torch.nn.functional, "scaled_dot_product_attention")
        )

    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
        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)  # (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)

        # causal self-attention; Self-attend: (B, nh, T, hs) x (B, nh, hs, T) -> (B, nh, T, T)
        if self.flash:
            # efficient attention using Flash Attention CUDA kernels
            y = torch.nn.functional.scaled_dot_product_attention(
                q, k, v, attn_mask=None, dropout_p=self.dropout, is_causal=False
            )
        else:
            # manual implementation of attention
            att = (q @ k.transpose(-2, -1)) * (1.0 / math.sqrt(k.size(-1)))
            att = F.softmax(att, dim=-1)
            att = self.attn_dropout(att)
            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.resid_dropout(self.c_proj(y))
        return y


class FineBlock(nn.Module):
    def __init__(self, config):
        super().__init__()
        self.ln_1 = nn.LayerNorm(config.n_embd)
        self.attn = NonCausalSelfAttention(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 FineGPT(GPT):
    def __init__(self, config):
        super().__init__(config)
        del self.lm_head
        self.config = config
        self.n_codes_total = config.n_codes_total
        self.transformer = nn.ModuleDict(
            dict(
                wtes=nn.ModuleList(
                    [
                        nn.Embedding(config.input_vocab_size, config.n_embd)
                        for _ in range(config.n_codes_total)
                    ]
                ),
                wpe=nn.Embedding(config.block_size, config.n_embd),
                drop=nn.Dropout(config.dropout),
                h=nn.ModuleList([FineBlock(config) for _ in range(config.n_layer)]),
                ln_f=nn.LayerNorm(config.n_embd),
            )
        )
        self.lm_heads = nn.ModuleList(
            [
                nn.Linear(config.n_embd, config.output_vocab_size, bias=False)
                for _ in range(config.n_codes_given, self.n_codes_total)
            ]
        )
        for i in range(self.n_codes_total - config.n_codes_given):
            self.transformer.wtes[i + 1].weight = self.lm_heads[i].weight

    def forward(self, pred_idx, idx):
        device = idx.device
        b, t, codes = idx.size()
        assert (
            t <= self.config.block_size
        ), f"Cannot forward sequence of length {t}, block size is only {self.config.block_size}"
        assert pred_idx > 0, "cannot predict 0th codebook"
        assert codes == self.n_codes_total, (b, t, codes)
        pos = torch.arange(0, t, dtype=torch.long, device=device).unsqueeze(0)  # shape (1, t)

        # forward the GPT model itself
        tok_embs = [
            wte(idx[:, :, i]).unsqueeze(-1) for i, wte in enumerate(self.transformer.wtes)
        ]  # token embeddings of shape (b, t, n_embd)
        tok_emb = torch.cat(tok_embs, dim=-1)
        pos_emb = self.transformer.wpe(pos)  # position embeddings of shape (1, t, n_embd)
        x = tok_emb[:, :, :, : pred_idx + 1].sum(dim=-1)
        x = self.transformer.drop(x + pos_emb)
        for block in self.transformer.h:
            x = block(x)
        x = self.transformer.ln_f(x)
        logits = self.lm_heads[pred_idx - self.config.n_codes_given](x)
        return logits

    def get_num_params(self, non_embedding=True):
        """
        Return the number of parameters in the model.
        For non-embedding count (default), the position embeddings get subtracted.
        The token embeddings would too, except due to the parameter sharing these
        params are actually used as weights in the final layer, so we include them.
        """
        n_params = sum(p.numel() for p in self.parameters())
        if non_embedding:
            for wte in self.transformer.wtes:
                n_params -= wte.weight.numel()
            n_params -= self.transformer.wpe.weight.numel()
        return n_params


@dataclass
class FineGPTConfig(GPTConfig):
    n_codes_total: int = 8
    n_codes_given: int = 1