app.py

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

model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32")
processor = CLIPProcessor.from_pretrained("tokenizer")

def generate_answer(image):

    clas = ['high_natural_light','neutral_natural_light','No_Light', 'artificial_lights']
    classes = ['A picture of a room filled with abundant natural light with a lot windows, without objects that prevent the light from passing through regardless of whether it is night','A picture of a room with neutral natural light with few windows, without objects that prevent the light from passing through regardless of whether it is night','Nolights','A picture of a room with Artificial lights like lamps or headlamps']
    inputs = processor(text=classes, images=image, return_tensors="pt", padding=True)
    outputs = model(**inputs)
    logits_per_image = outputs.logits_per_image
    probs = logits_per_image.softmax(dim=1)

    probabilities_list = probs.squeeze().tolist()

    result_dict = {class_name: probability for class_name, probability in zip(clas, probabilities_list)}
    
    return result_dict

image_input = gr.Image(type="pil", label="Upload Image")

iface = gr.Interface(
    fn=generate_answer, 
    inputs=[image_input], 
    outputs="text",
    title="Room Lightning Score",
    description="Upload an room image"
)

iface.launch()