app.py

import gradio as gr
import tensorflow as tf
import numpy as np
from PIL import Image
 
 
model_path = "pokemon_classifier_xception.keras"
model = tf.keras.models.load_model(model_path)
 
# Define the core prediction function
def predict_pokemon(image):
    # Preprocess image
    print(type(image))
    image = Image.fromarray(image.astype('uint8'))  # Convert numpy array to PIL image
    image = image.resize((150, 150)) #resize the image to 150x150
    image = np.array(image)
    image = np.expand_dims(image, axis=0) # same as image[None, ...]
    # Predict
    prediction = model.predict(image)
    # Apply softmax to get probabilities for each class
    prediction = tf.nn.softmax(prediction)
    # Create a dictionary with the probabilities for each Pokemon
    abra = np.round(float(prediction[0][0]), 2)
    aerodactyl = np.round(float(prediction[0][1]), 2)
    arcanine = np.round(float(prediction[0][2]), 2)
    return {'abra': abra, 'aerodactyl': aerodactyl, 'arcanine': arcanine}
 
 
input_image = gr.Image()
iface = gr.Interface(
    fn=predict_pokemon,
    inputs=input_image, 
    outputs=gr.Label(),
    examples=[], 
    description="A simple mlp classification model for image classification using the mnist dataset.")
iface.launch(share=True)