# -*- coding: utf-8 -*-
"""gradio.ipynb

Automatically generated by Colab.

Original file is located at
    https://colab.research.google.com/drive/1B520JUHmyofueyUqotN2yj6Gad69uavo
"""

import gradio as gr
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import models, layers


input_shape = (256, 256, 3)

def create_model():
    model = models.Sequential([
        
    layers.Input(shape=(256, 256, 3)),  # Define input shape here
    layers.Resizing(256, 256),       # Resize images 
    layers.Rescaling(1.0/255),       # Rescale pixel values
    layers.RandomFlip('horizontal_and_vertical'),
    layers.RandomRotation(0.2),
    layers.Conv2D(32, (3, 3), activation='relu'),  # Remove input_shape here
    layers.MaxPooling2D((2, 2)),
    layers.Conv2D(64, (3, 3), activation='relu'),
    layers.MaxPooling2D((2, 2)),
    layers.Conv2D(64, (3, 3), activation='relu'),
    layers.MaxPooling2D((2, 2)),
    layers.Conv2D(64, (3, 3), activation='relu'),
    layers.MaxPooling2D((2, 2)),
    layers.Conv2D(64, (3, 3), activation='relu'),
    layers.MaxPooling2D((2, 2)),
    layers.Flatten(),
    layers.Dense(64, activation='relu'),
    layers.Dense(3, activation='softmax')
        
    ])
    
    return model

model = create_model()
model.load_weights('my_model_weights1.weights.h5')

def predict_disease(image):
    """
    Preprocesses the image, performs prediction, and returns the predicted class with confidence score.
    """
    img = tf.image.resize(image, [256, 256])
    img = tf.expand_dims(img, axis=0)  # Add batch dimension

    # Predict probabilities
    prediction = model.predict(img)[0]  # Extract first element (batch of size 1)
    labels = ['Early Blight', 'Late Blight', 'Healthy']

    # Create a dictionary of class probabilities
    class_probs = {label: float(prob) for label, prob in zip(labels, prediction)}

    return class_probs
# ... (Rest of your code) ...

iface = gr.Interface(
    fn=predict_disease,
    inputs=gr.Image(),  # Allow image editing
    outputs=gr.Label(num_top_classes=3),          # Display all three probabilities
    title="Potato Disease Classifier 🍟🥔",
    description=(
        "Upload an image of a potato leaf, and I'll classify it as one of the following:\n"
        "- **Early Blight** 🌱\n"
        "- **Late Blight** 🌧️\n"
        "- **Healthy** 💚\n\n"
        "Check out the probability bars for more details!"
    ),
)
iface.launch(share=True)