app.py

import gradio as gr
from transformers import CLIPModel, CLIPProcessor
from PIL import Image
import torch

# Load model and processor
model_name = "jinaai/jina-clip-v1"
model = CLIPModel.from_pretrained(model_name)
processor = CLIPProcessor.from_pretrained(model_name)

def compute_similarity(image, text):
    image = Image.fromarray(image)  # Convert NumPy array to PIL image

    # Process inputs
    inputs = processor(text=[text], images=image, return_tensors="pt", padding=True, truncation=True)

    with torch.no_grad():
        outputs = model(**inputs)
        logits_per_image = outputs.logits_per_image  # Image-to-text similarity
        similarity_score = logits_per_image.item()
    
    return similarity_score

# Gradio UI
demo = gr.Interface(
    fn=compute_similarity,
    inputs=[gr.Image(type="numpy"), gr.Textbox(label="Enter text")],
    outputs=gr.Number(label="Similarity Score"),
    title="CLIP Image-Text Similarity",
    description="Upload an image and enter a text prompt to get the similarity score."
)

demo.launch()