handler.py

from typing import Dict, List, Any
from transformers import pipeline

class EndpointHandler():
    def __init__(self, path=""):
        self.pipeline = pipeline("visual-question-answering", model=path)

    def __call__(self, data: Dict[str, Any]) -> List[Dict[str, Any]]:
        """
       data args:
            inputs (:obj: `str`)
            date (:obj: `str`)
      Return:
            A :obj:`list` | `dict`: will be serialized and returned
        """
        # get inputs
        inputs = data.pop("inputs", data)
        top_k = inputs.get("top_k", 10)
        question = inputs.get("question", 'Whats the negative of "You provided a question"?')
        image = inputs.get("image", "https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg")

        # run normal prediction
        prediction = self.pipeline(image, question, top_k=10)
        return prediction