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

# Load the trained model
load_model = tf.keras.models.load_model('checkpoint_model.h5')  # replace with actual model path

# Assuming 'label_names' is a list of class names
label_names = ['asoka', 'kecubung', 'krokot', 'periwinkle', 'telang', 'zinnia']  # replace with actual class names

# Image size to match the input size of your model
IMG_SIZE = (224, 224)  # replace with your actual image size

def predict_image(image):
    img = Image.fromarray(image)
    img = img.resize(IMG_SIZE)
    x = tf.keras.preprocessing.image.img_to_array(img)
    x = np.expand_dims(x, axis=0)

    classes = load_model.predict(x, batch_size=10)
    outclass = np.argmax(classes)

    return label_names[outclass], {label: float(classes[0][i]) for i, label in enumerate(label_names)}

def classify_image(image):
    image = np.array(image)
    outclass, classes = predict_image(image)
    return outclass, classes

css = """
body {
    background-color: #D9FO8absF;
}
"""

with gr.Blocks(css=css) as demo:
    gr.Markdown('<div style="display: flex; justify-content: center; margin-top:-50px"><img src="https://media.giphy.com/media/X5PsaxTP6U3h9dUSxd/giphy.gif" alt="GIF" width="100%" style="max-width: 200px;"></div>')
    gr.Markdown('<p style="background-color: #658626; color: white; font-size: 30px; padding: 20px; border-radius: 10px; text-align: center; box-shadow: 0px 6px 8px rgba(0, 0, 0, 0.1);">Bunga Herbal App</p>')
    gr.Markdown('<p style="background-color: #8FB447; color: white; font-size: 20px; padding: 10px; border-radius: 5px; text-align: center; box-shadow: 0px 2px 3px rgba(0, 0, 0, 0.1);"> Tipe Bunga </p>')
    # Optional: Display a placeholder image
    gr.Image("/content/drive/MyDrive/bismillah.png")

    with gr.Row():
        with gr.Column():
            image_input = gr.Image(type="pil", label="Upload an image...")

        with gr.Column():
            label_output = gr.Label(label="Predicted Label")
            confidence_output = gr.JSON(label="Prediction Confidence")

    # Inform the user about capturing an image
    gr.Markdown("**To capture an image using your phone's camera:**", label="**OR**")
    gr.Markdown(
        "1. Open your device's camera app."
        "2. Take a picture of the flower you want to identify."
        "3. Save the image."
        "4. Upload the saved image here using the upload button above."
    )

    image_input.change(fn=classify_image, inputs=image_input, outputs=[label_output, confidence_output])

demo.launch()