app.py

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

# Load the BLIP model and processor
try:
    processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
    model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
    model = model.to(device='cuda' if torch.cuda.is_available() else 'cpu')
    model.eval()
except Exception as e:
    print(f"Error loading model or processor: {e}")
    exit()

def process_image(image, question):
    device = 'cuda' if torch.cuda.is_available() else 'cpu'

    # Convert Gradio image to PIL Image
    image = Image.fromarray(image).convert('RGB')
    
    # Preprocess the image and question
    inputs = processor(image, question, return_tensors="pt").to(device)
    
    # Perform inference
    try:
        with torch.no_grad():
            outputs = model.generate(**inputs)
        answer = processor.decode(outputs[0], skip_special_tokens=True)
        return answer
    except Exception as e:
        return f"Error during model inference: {e}"

# Define the Gradio interface
interface = gr.Interface(
    fn=process_image,
    inputs=[gr.Image(type='numpy'), gr.Textbox(label="Question")],
    outputs=gr.Textbox(),
    title="Image Question Answering",
    description="Upload an image and ask a question about it. The model will provide an answer."
)

# Launch the Gradio app
interface.launch()