import tensorflow as tf
from tensorflow import keras
from tensorflow.keras.preprocessing import image_dataset_from_directory
from tensorflow.keras import models, layers
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import gradio as gr
from transformers import AutoTokenizer, AutoModelForImageClassification
from PIL import Image
import requests
import torch

# Load model from Hugging Face model hub
model_name = "NKASG/GNN"  # Replace with your model's name on Hugging Face
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForImageClassification.from_pretrained(model_name)

# Define function for image preprocessing and prediction
def process_image(image):
    
    img_4d=img.reshape(-1,256,256,3)
    prediction=model.predict(img_4d)[0]
  # return {class_names[i]: float(prediction[i]) for i in range(3)}

    image = gr.inputs.Image(shape=(256,256))
    label = gr.outputs.Label(num_top_classes=1)

    gr.Interface(fn=predict_image, inputs=image, outputs=label,interpretation='default').launch()