handler.py

import torch
from typing import Dict, List, Any
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
import logging

# get dtype
dtype = torch.bfloat16 if torch.cuda.get_device_capability()[0] == 8 else torch.float16

class EndpointHandler:
    def __init__(self, path=""):
        # load the model
        self.tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code=True)
        self.model = AutoModelForCausalLM.from_pretrained(path, device_map="auto", torch_dtype=dtype, trust_remote_code=True)
        # create inference pipeline
        self.pipeline = pipeline("text-generation", model=self.model, tokenizer=self.tokenizer)

    def __call__(self, data: Any) -> List[List[Dict[str, float]]]:
        inputs = data.pop("inputs", data)
        parameters = data.pop("parameters", None)

        # pass inputs with all kwargs in data
        if parameters is not None:
            prediction = self.pipeline(inputs, **parameters)
        else:
            prediction = self.pipeline(inputs)
        logging.warn("---start---")
        logging.warn(prediction)
        logging.warn("---end---")

        # ignoring parameters! Default to configs in generation_config.json.
        messages = [{"role": "user", "content": data.pop("inputs", data)}]
        response = self.model.chat(self.tokenizer, messages)
        logging.warn("---start chat response---")
        logging.warn(response)
        logging.warn("---end chat response---")

        return [[{response: 1.0}]]