handler.py

import os
import torch
from PIL import Image
from io import BytesIO
from typing import Dict, Any
from transformers import LlamaTokenizer, GenerationConfig
from robohusky.model.modeling_husky_embody2 import HuskyForConditionalGeneration
from robohusky.video_transformers import (
    GroupNormalize, GroupScale, GroupCenterCrop,
    Stack, ToTorchFormatTensor, get_index
)
from decord import VideoReader, cpu
import torchvision.transforms as T
from torchvision.transforms.functional import InterpolationMode

DEFAULT_IMG_START_TOKEN = "<img>"
DEFAULT_IMG_END_TOKEN = "</img>"
DEFAULT_VIDEO_START_TOKEN = "<vid>"
DEFAULT_VIDEO_END_TOKEN = "</vid>"

class EndpointHandler:
    def __init__(self, model_path: str = "."):
        self.device = "cuda" if torch.cuda.is_available() else "cpu"
        self.tokenizer = LlamaTokenizer.from_pretrained(model_path, use_fast=False)
        self.model = HuskyForConditionalGeneration.from_pretrained(
            model_path, torch_dtype=torch.float16 if self.device == "cuda" else torch.float32
        ).to(self.device).eval()

        self.gen_config = GenerationConfig(
            bos_token_id=1,
            do_sample=True,
            temperature=0.7,
            max_new_tokens=1024
        )

    def __call__(self, data: Dict[str, Any]) -> Dict[str, str]:
        # Hugging Face 会调用这个函数，data 是原始输入
        inputs = self.preprocess(data)
        prediction = self.inference(inputs)
        return self.postprocess(prediction)

    def preprocess(self, request: Dict[str, Any]) -> Dict[str, Any]:
        prompt = request["inputs"]
        image = request.get("image", None)
        video = request.get("video", None)

        if image:
            pixel_values = self._load_image(image).unsqueeze(0).to(self.device)
            prompt = prompt.replace("<image>", DEFAULT_IMG_START_TOKEN + DEFAULT_IMG_END_TOKEN)
        elif video:
            pixel_values = self._load_video(video).unsqueeze(0).to(self.device)
            prompt = prompt.replace("<video>", DEFAULT_VIDEO_START_TOKEN + DEFAULT_VIDEO_END_TOKEN)
        else:
            pixel_values = None

        return {
            "prompt": prompt,
            "pixel_values": pixel_values
        }

    def inference(self, inputs: Dict[str, Any]) -> str:
        prompt = inputs["prompt"]
        pixel_values = inputs["pixel_values"]

        model_inputs = self.tokenizer([prompt], return_tensors="pt")
        model_inputs.pop("token_type_ids", None)
        model_inputs = {k: v.to(self.device) for k, v in model_inputs.items()}

        if pixel_values is not None:
            output = self.model.generate(
                **model_inputs,
                pixel_values=pixel_values,
                generation_config=self.gen_config,
                return_dict_in_generate=True,
                output_scores=True
            )
        else:
            output = self.model.language_model.generate(
                **model_inputs,
                generation_config=self.gen_config,
                return_dict_in_generate=True,
                output_scores=True
            )

        return self.tokenizer.decode(output.sequences[0], skip_special_tokens=True)

    def postprocess(self, output: str) -> Dict[str, str]:
        return {"output": output.strip()}

    def _load_image(self, image_bytes: bytes) -> torch.Tensor:
        image = Image.open(BytesIO(image_bytes)).convert('RGB')
        crop_pct = 224 / 256
        size = int(224 / crop_pct)
        transform = T.Compose([
            T.Resize(size, interpolation=InterpolationMode.BICUBIC),
            T.CenterCrop(224),
            T.ToTensor(),
            T.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225))
        ])
        return transform(image)

    def _load_video(self, video_bytes: bytes, num_segments=8) -> torch.Tensor:
        with open("/tmp/temp_video.mp4", "wb") as f:
            f.write(video_bytes)
        vr = VideoReader("/tmp/temp_video.mp4", ctx=cpu(0))
        frame_indices = get_index(len(vr), num_segments)
        frames = [Image.fromarray(vr[idx].asnumpy()) for idx in frame_indices]

        transform = T.Compose([
            GroupScale(224),
            GroupCenterCrop(224),
            Stack(),
            ToTorchFormatTensor(),
            GroupNormalize([0.48145466, 0.4578275, 0.40821073], [0.26862954, 0.26130258, 0.27577711])
        ])
        return transform(frames)