src/lmms-eval/lmms_eval/models/chat/huggingface.py

import base64
import re
import time
from io import BytesIO
from typing import List, Optional, Tuple, Union

import decord
import numpy as np
import torch
from accelerate import Accelerator, DistributedType
from loguru import logger as eval_logger
from PIL import Image
from tqdm import tqdm
from transformers import (
    AutoConfig,
    AutoModel,
    AutoModelForCausalLM,
    AutoModelForImageTextToText,
    AutoProcessor,
    AutoTokenizer,
)

from lmms_eval import utils
from lmms_eval.api.instance import Instance
from lmms_eval.api.model import lmms
from lmms_eval.api.registry import register_model
from lmms_eval.models.model_utils.gen_metrics import log_metrics
from lmms_eval.models.model_utils.reasoning_model_utils import (
    parse_reasoning_model_answer,
)
from lmms_eval.protocol import ChatMessages

try:
    from qwen_vl_utils import process_vision_info
except ImportError:
    eval_logger.warning("Failed to import qwen_vl_utils; Please install it via `pip install qwen-vl-utils`")


@register_model("huggingface")
class Huggingface(lmms):
    is_simple = False
    """
    Qwen2.5_VL Model
    "https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct"
    """

    def __init__(
        self,
        pretrained: str = "Qwen/Qwen2.5-VL-3B-Instruct",
        device: Optional[str] = "cuda",
        device_map: Optional[str] = "auto",
        batch_size: Optional[Union[int, str]] = 1,
        use_cache=True,
        attn_implementation: Optional[str] = None,
        max_num_frames: int = 32,
        use_custom_video_loader: Optional[bool] = False,
        fps: Optional[float] = None,  # Only applicable if use_custom_video_loader is True
        max_image_size: Optional[int] = None,  # Only applicable if use_custom_video_loader is True
        system_prompt: Optional[str] = "You are a helpful assistant.",
        interleave_visuals: Optional[bool] = False,
        reasoning_prompt: Optional[str] = None,
        **kwargs,
    ) -> None:
        super().__init__()
        # Do not use kwargs for now
        assert kwargs == {}, f"Unexpected kwargs: {kwargs}"

        # Validate attention implementation
        valid_attn_implementations = [None, "flash_attention_2", "sdpa", "eager"]
        if attn_implementation not in valid_attn_implementations:
            raise ValueError(f"attn_implementation must be one of {valid_attn_implementations}, got {attn_implementation}")

        self.use_custom_video_loader = use_custom_video_loader
        self.fps = fps
        # if self.fps and not self.use_custom_video_loader:
        #     raise ValueError("FPS is only applicable if use_custom_video_loader is True")
        self.max_image_size = max_image_size
        if self.max_image_size and not self.use_custom_video_loader:
            raise ValueError("max_image_size is only applicable if use_custom_video_loader is True")

        accelerator = Accelerator()
        self.accelerator = accelerator
        if accelerator.num_processes > 1:
            self._device = torch.device(f"cuda:{accelerator.local_process_index}")
            self.device_map = f"cuda:{accelerator.local_process_index}"
        else:
            self._device = torch.device(device)
            self.device_map = device_map if device_map else device

        # Prepare model loading arguments
        model_kwargs = {
            "torch_dtype": "bfloat16",
            "device_map": self.device_map,
        }

        # Add attention implementation if specified
        if attn_implementation is not None:
            model_kwargs["attn_implementation"] = attn_implementation

        config = AutoConfig.from_pretrained(pretrained)
        if config.model_type in AutoModelForCausalLM._model_mapping.keys():
            model_cls = AutoModelForCausalLM
        elif config.model_type in AutoModelForImageTextToText._model_mapping.keys():
            model_cls = AutoModelForImageTextToText
        else:
            model_cls = AutoModel

        self._model = model_cls.from_pretrained(pretrained, **model_kwargs).eval()
        self.max_num_frames = max_num_frames

        if reasoning_prompt:
            self.reasoning_prompt = reasoning_prompt.replace("\\n", "\n")
        else:
            self.reasoning_prompt = None
        self.processor = AutoProcessor.from_pretrained(pretrained)
        self._tokenizer = AutoTokenizer.from_pretrained(pretrained)
        self.system_prompt = system_prompt
        self.interleave_visuals = interleave_visuals

        self._config = self.model.config
        self._max_length = kwargs.get("max_length", 2048)
        self.batch_size_per_gpu = int(batch_size)
        self.use_cache = use_cache

        if accelerator.num_processes > 1:
            assert accelerator.distributed_type in [
                DistributedType.FSDP,
                DistributedType.MULTI_GPU,
            ], "Unsupported distributed type provided. Only DDP and FSDP are supported."
            if accelerator.distributed_type == DistributedType.FSDP:
                self._model = accelerator.prepare(self.model)
            else:
                self._model = accelerator.prepare_model(self.model, evaluation_mode=True)
            self.accelerator = accelerator
            if self.accelerator.is_local_main_process:
                eval_logger.info(f"Using {accelerator.num_processes} devices with data parallelism")
            self._rank = self.accelerator.local_process_index
            self._world_size = self.accelerator.num_processes
        else:
            self._rank = 0
            self._world_size = 1

    @property
    def config(self):
        # return the associated transformers.AutoConfig for the given pretrained model.
        return self._config

    @property
    def tokenizer(self):
        return self._tokenizer

    @property
    def model(self):
        # returns the model, unwrapping it if using Accelerate
        if hasattr(self, "accelerator"):
            return self.accelerator.unwrap_model(self._model)
        else:
            return self._model

    @property
    def eot_token_id(self):
        return self.tokenizer.eos_token_id

    @property
    def max_length(self):
        return self._max_length

    @property
    def batch_size(self):
        return self.batch_size_per_gpu

    @property
    def device(self):
        return self._device

    @property
    def rank(self):
        return self._rank

    @property
    def world_size(self):
        return self._world_size

    def loglikelihood(self, requests: List[Instance]) -> List[Tuple[float, bool]]:
        raise NotImplementedError("Loglikelihood is not implemented for Qwen2.5_VL")

    def flatten(self, input):
        new_list = []
        for i in input:
            for j in i:
                new_list.append(j)
        return new_list

    def generate_until(self, requests: List[Instance]) -> List[str]:
        res = []

        # A dummy collate here to sort by doc id
        def _collate(x):
            return x[2], x[2]

        # we group requests by their generation_kwargs,
        # so that we don't try to execute e.g. greedy sampling and temp=0.8 sampling
        # in the same batch.
        re_ords = utils.Collator([reg.args for reg in requests], _collate, group_fn=lambda x: x[2], grouping=True)
        chunks = re_ords.get_batched(n=self.batch_size, batch_fn=None)
        num_iters = len(requests) // self.batch_size if len(requests) % self.batch_size == 0 else len(requests) // self.batch_size + 1
        pbar = tqdm(total=num_iters, disable=(self.rank != 0), desc="Model Responding")
        e2e_latency = 0
        total_tokens = 0
        for chunk in chunks:
            ctx, doc_to_messages, all_gen_kwargs, doc_id, task, split = zip(*chunk)
            chat_messages = [doc_to_messages[0](self.task_dict[task][split][ids]) for ids, task, split in zip(doc_id, task, split)]
            chat_messages: List[ChatMessages] = [ChatMessages(**{"messages": message}) for message in chat_messages]
            visuals = []
            videos = []
            for messages in chat_messages:
                visual, video, _ = messages.extract_media()
                visuals.append(visual)
                videos.append(video)
            visuals = self.flatten(visuals)
            videos = self.flatten(videos)
            gen_kwargs = all_gen_kwargs[0]

            # Apply chat template
            batched_messages = [chat_message.to_hf_messages() for chat_message in chat_messages]
            texts = [self.processor.apply_chat_template(msg, tokenize=False, add_generation_prompt=True) for msg in batched_messages]
            images = []
            videos = []
            audios = []
            for messages in chat_messages:
                image, video, audio = messages.extract_media()
                images.append(image)
                videos.append(video)
                audios.append(audio)
            images = self.flatten(images)
            videos = self.flatten(videos)
            audios = self.flatten(audios)
            kwargs = {"images": images, "videos": videos, "audios": audios}
            inputs = self.processor(text=texts, padding=True, return_tensors="pt", **kwargs)

            if self.device_map == "auto":
                inputs = inputs.to("cuda")
            else:
                inputs = inputs.to(self.device)

            # Set default generation kwargs
            default_gen_kwargs = {
                "max_new_tokens": 4096,
                "temperature": 0.0,  # Set to 0 for greedy default
                "top_p": None,
                "num_beams": 1,
            }
            # Update with provided kwargs
            current_gen_kwargs = {**default_gen_kwargs, **gen_kwargs}
            pad_token_id = self.tokenizer.pad_token_id

            if current_gen_kwargs["temperature"] > 0:
                current_gen_kwargs["do_sample"] = True
            else:
                current_gen_kwargs["do_sample"] = False
                current_gen_kwargs["temperature"] = None
                current_gen_kwargs["top_p"] = None

            start_time = time.time()
            cont = self.model.generate(
                **inputs,
                eos_token_id=self.tokenizer.eos_token_id,
                pad_token_id=pad_token_id,
                do_sample=current_gen_kwargs["do_sample"],
                temperature=current_gen_kwargs["temperature"],
                top_p=current_gen_kwargs["top_p"],
                num_beams=current_gen_kwargs["num_beams"],
                max_new_tokens=current_gen_kwargs["max_new_tokens"],
                use_cache=self.use_cache,
            )
            end_time = time.time()

            generated_ids_trimmed = [out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, cont)]
            answers = self.processor.batch_decode(generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False)

            # Calculate timing metrics for batch
            e2e_latency += end_time - start_time
            total_tokens += sum(len(ids) for ids in generated_ids_trimmed)

            for ans, context in zip(answers, texts):
                clean_ans = parse_reasoning_model_answer(ans)
                res.append(clean_ans)
                self.cache_hook.add_partial("generate_until", (context, gen_kwargs), clean_ans)
                pbar.update(1)

                eval_logger.debug(f"Question: {context}")
                eval_logger.debug(f"Model Raw Response: {ans}")
                eval_logger.debug(f"Model Clean Response: {clean_ans}")
            # reorder this group of results back to original unsorted form
        res = re_ords.get_original(res)
        # Calculate average speed
        avg_speed = total_tokens / e2e_latency if e2e_latency > 0 else 0
        # Log metrics
        metric_dict = {
            "total_tokens": total_tokens,
            "e2e_latency": e2e_latency,
            "avg_speed": avg_speed,
        }
        log_metrics(**metric_dict)

        pbar.close()
        return res

    def generate_until_multi_round(self, requests) -> List[str]:
        raise NotImplementedError("TODO: Implement multi-round generation")