import gradio as gr
import tensorflow as tf
import numpy as np
from tensorflow.keras.preprocessing.image import load_img, img_to_array

# Load the trained model
model = tf.keras.models.load_model("my_model.h5")

# Class labels (modify according to your dataset)
class_labels = ["Angry", "Disgust", "Fear", "Happy", "Neutral", "Sad", "Surprise"]

# Function to make predictions
def predict_emotion(image):
    # Resize image to match model's expected input
    image = image.resize((224, 224))  
    img_array = img_to_array(image) / 255.0  # Normalize
    img_array = np.expand_dims(img_array, axis=0)  # Add batch dimension

    # Make prediction
    prediction = model.predict(img_array)
    predicted_class = class_labels[np.argmax(prediction)]
    confidence = np.max(prediction)

    return f"Emotion: {predicted_class} (Confidence: {confidence:.2f})"

# Create Gradio interface
interface = gr.Interface(
    fn=predict_emotion,
    inputs=gr.Image(type="pil"),
    outputs=gr.Textbox(label="Prediction"),
    title="Facial Emotion Detection",
    description="Upload an image, and the model will predict the emotion.",
)

# Run the app
if __name__ == "__main__":
    interface.launch()