app.py

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

# Load the dog breed classifier model
model_path = "best_model.keras"
model = tf.keras.models.load_model(model_path)
labels = ['Beagle', 'French Bulldog', 'German Shepherd', 'Golden Retriever', 'Labrador Retriever']

# Define function for dog breed classification with data augmentation
def preprocess_image(image):
    image = Image.fromarray(image.astype('uint8'), 'RGB')
    image = image.resize((150, 150))
    image = np.array(image)
    image = image / 255.0  # Normalization of pixel values
    return image

# Prediction function
def predict_dog_breed(image):
    image = preprocess_image(image)
    prediction = model.predict(np.expand_dims(image, axis=0))
    predicted_class = labels[np.argmax(prediction)]
    confidence = np.round(np.max(prediction) * 100, 2)
    result = f"{predicted_class}, Confidence: {confidence}%"
    return result

# Create Gradio interface
input_image = gr.Image()
output_text = gr.Textbox(label="Dog Breed")
interface = gr.Interface(fn=predict_dog_breed, inputs=input_image, outputs=output_text, 
                         description="A dog breed classifier using a custom CNN model with data augmentation.", 
                         theme="default")

if __name__ == "__main__":
    interface.launch(share=True)