base.py

import sys
import time
from pathlib import Path
from typing import Any, Literal, Optional

import lightning as L
import torch
import torch._dynamo.config
import torch._inductor.config
from lightning.fabric.plugins import BitsandbytesPrecision
from lightning.fabric.strategies import FSDPStrategy

# support running without installing as a package
wd = Path(__file__).parent.parent.resolve()
sys.path.append(str(wd))

from model import *
from utils import *
from tokenizer import *

def multinomial_num_samples_1(probs: torch.Tensor) -> torch.Tensor:
    if torch._dynamo.is_compiling():
        # Faster alternative to `torch.multinomial(probs, num_samples=1)` that is also CUDAGraph friendly
        distribution = torch.empty_like(probs).exponential_(1)
        return torch.argmax(probs / distribution, dim=-1, keepdim=True)
    return torch.multinomial(probs, num_samples=1)


def sample(logits: torch.Tensor, temperature: float = 1.0, top_k: Optional[int] = None) -> torch.Tensor:
    logits = logits[0, -1]
    # optionally crop the logits to only the top k options
    if top_k is not None:
        v, i = torch.topk(logits, min(top_k, logits.size(-1)))
        # do not use `torch.where` as in nanogpt because it will repeat top-k collisions
        logits = torch.full_like(logits, float("-inf")).scatter_(-1, i, v)
    # optionally scale the logits and sample from a probability distribution
    if temperature > 0.0:
        probs = torch.nn.functional.softmax(logits / temperature, dim=-1)
        return multinomial_num_samples_1(probs)
    return torch.argmax(logits, dim=-1, keepdim=True)


def next_token(model: GPT, input_pos: torch.Tensor, x: torch.Tensor, **kwargs: Any) -> torch.Tensor:
    logits = model(x, input_pos)
    next = sample(logits, **kwargs)
    return next.type_as(x)


@torch.inference_mode()
def generate(
    model: GPT,
    prompt: torch.Tensor,
    max_returned_tokens: int,
    *,
    temperature: float = 1.0,
    top_k: Optional[int] = None,
    eos_id: Optional[int] = None,
) -> torch.Tensor:
    """Takes a conditioning sequence (prompt) as input and continues to generate as many tokens as requested.

    The implementation of this function is modified from A. Karpathy's nanoGPT.

    Args:
        model: The model to use.
        prompt: Tensor of shape (T) with indices of the prompt sequence.
        max_returned_tokens: The maximum number of tokens to return (given plus generated).
        temperature: Scales the predicted logits by 1 / temperature.
        top_k: If specified, only sample among the tokens with the k highest probabilities.
        eos_id: If specified, stop generating any more token once the <eos> token is triggered.
    """
    T = prompt.size(0)
    assert max_returned_tokens > T
    if model.max_seq_length < max_returned_tokens - 1:
        # rolling the kv cache based on the `input_pos` value would be necessary. However, doing so would introduce a
        # data dependency on the `input_pos` tensor and impact model compilation. Since this setting is uncommon, we do
        # not support it to avoid negatively impacting the overall speed
        raise NotImplementedError(f"max_seq_length {model.max_seq_length} needs to be >= {max_returned_tokens - 1}")

    device = prompt.device
    tokens = [prompt]
    input_pos = torch.tensor([T], device=device)
    token = next_token(
        model, torch.arange(0, T, device=device), prompt.view(1, -1), temperature=temperature, top_k=top_k
    ).clone()
    tokens.append(token)
    for _ in range(2, max_returned_tokens - T + 1):
        token = next_token(model, input_pos, token.view(1, -1), temperature=temperature, top_k=top_k).clone()
        tokens.append(token)
        if token == eos_id:
            break
        input_pos = input_pos.add_(1)
    return torch.cat(tokens)


def main(
    prompt: str = "What food do llamas eat?",
    *,
    num_samples: int = 1,
    max_new_tokens: int = 50,
    top_k: Optional[int] = 200,
    temperature: float = 0.8,
    checkpoint_dir: Path = Path("checkpoints/stabilityai/stablelm-base-alpha-3b"),
    quantize: Optional[Literal["bnb.nf4", "bnb.nf4-dq", "bnb.fp4", "bnb.fp4-dq", "bnb.int8", "gptq.int4"]] = None,
    strategy: str = "auto",
    devices: int = 1,
    precision: Optional[str] = None,
    compile: bool = False,
    model_name: str = "pythia_160m_hf"
) -> None:
    """Generates text samples based on a pre-trained model and tokenizer.

    Args:
        prompt: The prompt string to use for generating the samples.
        num_samples: The number of text samples to generate.
        max_new_tokens: The number of generation steps to take.
        top_k: The number of top most probable tokens to consider in the sampling process.
        temperature: A value controlling the randomness of the sampling process. Higher values result in more random
            samples.
        checkpoint_dir: The checkpoint directory to load.
        quantize: Whether to quantize the model and using which method:
            - bnb.nf4, bnb.nf4-dq, bnb.fp4, bnb.fp4-dq: 4-bit quantization from bitsandbytes
            - bnb.int8: 8-bit quantization from bitsandbytes
            - gptq.int4: 4-bit quantization from GPTQ
            for more details, see https://github.com/Lightning-AI/lit-gpt/blob/main/tutorials/quantize.md
        strategy: Indicates the Fabric strategy setting to use.
        devices: How many devices to use.
        precision: Indicates the Fabric precision setting to use.
        compile: Whether to compile the model.
    """
    precision = precision or get_default_supported_precision(training=False)

    plugins = None
    if quantize is not None:
        if devices > 1:
            raise NotImplementedError(
                "Quantization is currently not supported for multi-GPU training. Please set devices=1 when using the"
                " --quantize flag."
            )
        if quantize.startswith("bnb."):
            if "mixed" in precision:
                raise ValueError("Quantization and mixed precision is not supported.")
            dtype = {"16-true": torch.float16, "bf16-true": torch.bfloat16, "32-true": torch.float32}[precision]
            plugins = BitsandbytesPrecision(quantize[4:], dtype)
            precision = None

    if strategy == "fsdp":
        strategy = FSDPStrategy(auto_wrap_policy={Block}, cpu_offload=False)

    fabric = L.Fabric(devices=devices, precision=precision, strategy=strategy, plugins=plugins)
    fabric.launch()

    check_valid_checkpoint_dir(checkpoint_dir, model_name)

    config = Config.from_json(checkpoint_dir / "lit_config.json")

    if quantize == "gptq.int4":
        model_file = "lit_model_gptq.4bit.pth"
        if not (checkpoint_dir / model_file).is_file():
            raise ValueError("Please run `python quantize/gptq.py` first")
    else:
        if model_name == "pythia_160m_deduped_huggingface":
            model_file = "pythia_160m_deduped_hf.pth"
        elif model_name == "pythia_160m_deduped_custom":
            model_file = "pythia_160m_deduped_custom.pth"
        else:
            model_file = "lit_model.pth"
    checkpoint_path = checkpoint_dir / model_file

    tokenizer = Tokenizer(checkpoint_dir)
    encoded = tokenizer.encode(prompt, device=fabric.device)
    prompt_length = encoded.size(0)
    max_returned_tokens = prompt_length + max_new_tokens

    fabric.print(f"Loading model {str(checkpoint_path)!r} with {config.__dict__}", file=sys.stderr)
    t0 = time.perf_counter()
    with fabric.init_module(empty_init=True), gptq_quantization(quantize == "gptq.int4"):
        model = GPT(config)
    fabric.print(f"Time to instantiate model: {time.perf_counter() - t0:.02f} seconds.", file=sys.stderr)
    with fabric.init_tensor():
        # set the max_seq_length to limit the memory usage to what we need
        model.max_seq_length = max_returned_tokens
        # enable the kv cache
        model.set_kv_cache(batch_size=1)
    model.eval()

    if compile:
        torch._dynamo.config.automatic_dynamic_shapes = True
        torch._inductor.config.triton.unique_kernel_names = True
        torch._inductor.config.coordinate_descent_tuning = True
        global next_token
        next_token = torch.compile(next_token, mode="reduce-overhead")

    model = fabric.setup_module(model)

    t0 = time.perf_counter()
    load_checkpoint(fabric, model, checkpoint_path)
    fabric.print(f"Time to load the model weights: {time.perf_counter() - t0:.02f} seconds.", file=sys.stderr)

    L.seed_everything(1234)
    print(f'num_samples is {num_samples}')
    output_msg_list = []
    for i in range(num_samples):
        t0 = time.perf_counter()
        y = generate(model, encoded, max_returned_tokens, temperature=temperature, top_k=top_k)
        t = time.perf_counter() - t0
        for block in model.transformer.h:
            block.attn.kv_cache.reset_parameters()
        output_msg = tokenizer.decode(y)
        tokens_generated = y.size(0) - prompt_length
        output_msg_list.append(output_msg)
        fabric.print(
            f"Time for inference {i + 1}: {t:.02f} sec total, {tokens_generated / t:.02f} tokens/sec", file=sys.stderr
        )
    if fabric.device.type == "cuda":
        fabric.print(f"Memory used: {torch.cuda.max_memory_allocated() / 1e9:.02f} GB", file=sys.stderr)
    return output_msg_list

if __name__ == "__main__":
    from jsonargparse import CLI

    torch.set_float32_matmul_precision("high")
    output_msg_list = CLI(main)