app.py

import gradio as gr
from transformers import pipeline

generator = pipeline("visual-question-answering", model="jihadzakki/blip1-medvqa")

def format_answer(image, question):
    result = generator(image, question)
    print(result)  # Print the result to see its structure
    predicted_answer = result[0].get('answer', 'No answer found')  # Adjust this key if necessary

    return f"Predicted Answer: {predicted_answer}"

VisualQAApp = gr.Interface(
    fn=format_answer,
    inputs=[
        gr.Image(label="Upload image", type="pil"),
        gr.Textbox(label="Question"),
    ],
    outputs=[gr.Textbox(label="Answer")],
    title="Visual Question Answering using BLIP Model",
    description="VQA",
    allow_flagging="never"
)

VisualQAApp.launch(share=True)