app.py

import gradio as gr
from tensorflow.keras.models import load_model
import numpy as np
import os

# Load the model (ensure it's in the same directory as app.py)
model_path = os.path.join(os.path.dirname(__file__), 'waste_classifier_mobilenetv2.h5')
model = load_model(model_path)

# Prediction function
def classify_image(image):
    if image is None:
        return "No image provided."
    
    image = np.array(image)

    if image.shape != (224, 224, 3):
        image = np.resize(image, (224, 224, 3))

    image = image / 255.0
    image = np.expand_dims(image, axis=0)

    prediction = model.predict(image)
    class_label = 'Reyclabble' if prediction[0][0] > 0.5 else 'Organic'
    return class_label

# Create Gradio interface
interface = gr.Interface(fn=classify_image, inputs=gr.Image(type="numpy"), outputs="text")
interface.launch(share=True)