app.py

import gradio as gr
import numpy as np
from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing import image
import tensorflow as tf

# Load the saved model
model = load_model('acres-ppdc-01.keras')

# Define the classes the model was trained on
class_labels = ['Potato___Early_blight', 'Potato___Late_blight', 'Potato___healthy']

def classify_potato_plant(img):
    # Preprocess the image for the model
    img = img.resize((128, 128))  # Resize to the same size the model was trained on
    img = image.img_to_array(img)
    img = np.expand_dims(img, axis=0)
    img = img / 255.0  # Normalize the image

    # Make the prediction
    predictions = model.predict(img)
    predicted_class = np.argmax(predictions[0])
    confidence = predictions[0][predicted_class]
    model_output = "None"

    if class_labels[predicted_class] == "Potato__Early_blight":
        model_output = "Early blight"
    elif class_labels[predicted_class] == "Potato__Late_blight":
        model_output = "Late blight"
    elif class_labels[predicted_class] == "Potate__healthy":
        model_output = "Healthy"

    return model_output, confidence

# Create the Gradio interface
interface = gr.Interface(
    fn=classify_potato_plant,
    inputs=gr.Image(type="pil"),
    outputs=[gr.Label(label="Predicted output", num_top_classes=1), gr.Textbox(label="Confidence Score")],
    title="Acres - PPDC",
    description="Acres PPDC, is our Potato Plant Disease Classification vision model, capable of accurately classifying potato plant disease, based on a single image."
)

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