app.py

import gradio as gr
import tensorflow as tf
from tensorflow.keras.preprocessing import image
import numpy as np
from PIL import Image
from keras import layers


# Load your trained Xception model
model = tf.keras.models.load_model("inception_acc_0.989001-_val_acc_0.98252.h5")

# Define the labels for your classification
class_labels = ['arm', 'hand', 'foot',  'legs','fullbody','head','backside', 'torso', 'stake', 'plastic'] # Replace with your actual class names

def classify_image(img):
    # Preprocess the image to fit the model input shape
    img = img.resize((299, 299))  # Xception takes 299x299 input size
    img = np.array(img) / 255.0   # Normalize the image
    img = np.expand_dims(img, axis=0)

    # Make prediction
    predictions = model.predict(img)
    predicted_class = np.argmax(predictions, axis=1)[0]
    confidence = np.max(predictions)
    return {class_labels[i]: float(predictions[0][i]) for i in range(len(class_labels))}, confidence

# Example images (local paths or URLs)
example_images = [
    'head.jpg',  # Replace with actual local file paths or URLs
    'torso.jpg' 
]

# Gradio interface
demo = gr.Interface(
    fn=classify_image,
    title="Human Bodypart Image Classification",
    description = "Predict the bodypart of human bodypart images. This is a demo of our human bodypart image <a href=\"https://huggingface.co/icputrd/Inception-V3-Human-Bodypart-Classifier\">classifier</a>.",
    inputs=gr.Image(type="pil"),
    outputs=[gr.Label(num_top_classes=len(class_labels)), gr.Number()],
    examples=example_images,
    cache_examples=False,
    live=True,
)

if __name__ == "__main__":
    demo.launch()