app.py

import gradio as gr
from transformers.utils import logging
logging.set_verbosity_error()

import warnings
warnings.filterwarnings("ignore", message="Using the model-agnostic default `max_length`")

from transformers import BlipForQuestionAnswering
from transformers import AutoProcessor

def qa(image, question):
    model = BlipForQuestionAnswering.from_pretrained(
    "./models/Salesforce/blip-vqa-base")
    processor = AutoProcessor.from_pretrained(
    "./models/Salesforce/blip-vqa-base")
    
    inputs = processor(image, question, return_tensors="pt")

    out = model.generate(**inputs)

    result  = processor.decode(out[0], skip_special_tokens=True)
    return result

# def greet(name):
#     return "Hello " + name + "!!"

iface = gr.Interface(fn=qa, inputs=["image","text"], outputs="textbox")
iface.launch()