app.py

import tensorflow as tf
import gradio as gr
import numpy as np
from PIL import Image

# Load the model
model = tf.saved_model.load('./saved_model')

# Define the prediction function
def predict(image):
    # Preprocess the image to the required input format
    img = np.array(image).astype(np.float32)
    img = np.expand_dims(img, axis=0)  # Add batch dimension
    img = tf.image.resize(img, (640, 640))  # Resize if needed

    # Perform inference
    predictions = model(img)
    return predictions.numpy()  # Adjust output processing as needed

# Set up the Gradio interface
interface = gr.Interface(fn=predict, inputs=gr.inputs.Image(type="pil"), outputs="label")
interface.launch()