app.py

import gradio as gr
from transformers import BlipProcessor, BlipForConditionalGeneration
from PIL import Image
import torch

# Example with BLIP (replace with your fine-tuned model)
processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")

def caption_image(image):
    if image is None:
        return "No image provided"
    inputs = processor(images=image, return_tensors="pt")
    with torch.no_grad():
        out = model.generate(**inputs)
    caption = processor.decode(out[0], skip_special_tokens=True)
    return caption

demo = gr.Interface(
    fn=caption_image,
    inputs=gr.Image(type="pil"),
    outputs="text",
    title="Custom UI Action Description"
)

if __name__ == "__main__":
    demo.launch()