import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout
from tensorflow.keras.preprocessing.image import ImageDataGenerator
import matplotlib.pyplot as plt

# Paths
train_dir = 'Tomato_Plant_Disease'

# Image Data Generator with Augmentation for training
datagen = ImageDataGenerator(
    rescale=1./255,               # Normalize pixel values
    rotation_range=20,
    width_shift_range=0.2,
    height_shift_range=0.2,
    shear_range=0.2,
    zoom_range=0.2,
    horizontal_flip=True,
    validation_split=0.2          # Use 20% for validation
)

# Training and Validation Data
train_generator = datagen.flow_from_directory(
    train_dir,
    target_size=(128, 128),
    batch_size=32,
    class_mode='binary',  # 'binary' since it's a binary classification
    subset='training'
)

validation_generator = datagen.flow_from_directory(
    train_dir,
    target_size=(128, 128),
    batch_size=32,
    class_mode='binary',
    subset='validation'
)

# Define the CNN model
model = Sequential()

# 1st Convolution Layer
model.add(Conv2D(32, (3, 3), activation='relu', input_shape=(128, 128, 3)))
model.add(MaxPooling2D(pool_size=(2, 2)))

# 2nd Convolution Layer
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))

# 3rd Convolution Layer
model.add(Conv2D(128, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))

# Flatten the output
model.add(Flatten())

# Fully connected layer
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))  # Dropout to prevent overfitting

# Output layer
model.add(Dense(1, activation='sigmoid'))  # Binary classification, hence 'sigmoid'

# Compile the model
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

# Summary
model.summary()

history = model.fit(
    train_generator,
    steps_per_epoch=train_generator.samples // 32,
    validation_data=validation_generator,
    validation_steps=validation_generator.samples // 32,
    epochs=20  # You can adjust based on performance
)

# Plot training & validation accuracy values
plt.plot(history.history['accuracy'])
plt.plot(history.history['val_accuracy'])
plt.title('Model accuracy')
plt.ylabel('Accuracy')
plt.xlabel('Epoch')
plt.legend(['Train', 'Validation'], loc='upper left')
plt.show()

# Plot training & validation loss values
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('Model loss')
plt.ylabel('Loss')
plt.xlabel('Epoch')
plt.legend(['Train', 'Validation'], loc='upper left')
plt.show()

# Assuming you have a test directory
test_dir = 'Tomato_Plant_Disease'  # Path to your test dataset

# Create a test data generator (without augmentation)
test_datagen = ImageDataGenerator(rescale=1./255)

# Load test data
test_generator = test_datagen.flow_from_directory(
    test_dir,
    target_size=(128, 128),
    batch_size=32,
    class_mode='binary'
)

test_loss, test_acc = model.evaluate(test_generator, steps=test_generator.samples // 32)
print(f"Test Accuracy: {test_acc}")

model.save('tomato_disease_detection_model.h5')

from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing import image
import numpy as np
from PIL import Image

# Load the model
model = load_model('tomato_disease_detection_model.h5')

def predict_disease(img_path):
    # Load and preprocess the image
    img = Image.open(img_path)
    img = img.resize((128, 128))  # Resize image to match model input size
    img = np.array(img)  # Convert to array
    img = img / 255.0    # Normalize pixel values
    img = np.expand_dims(img, axis=0)  # Add batch dimension

    # Make prediction
    prediction = model.predict(img)

    # Determine if infected or healthy
    if prediction[0][0] > 0.5:
        print("The plant is healthy.")
    else:
        print("The plant is infected.")

# Example usage:
img_path = 'Tomato_Plant_Disease/0/0045ba29-ed1b-43b4-afde-719cc7adefdb___GCREC_Bact.Sp 6254.JPG'
predict_disease(img_path)

import gradio as gr
from tensorflow.keras.models import load_model
from PIL import Image
import numpy as np

# Load the pre-trained model
model = load_model('tomato_disease_detection_model.h5')

# Prediction function to be used in Gradio
def predict_disease(img):
    # Preprocess the image
    img = img.resize((128, 128))  # Resize to match model input size
    img = np.array(img)  # Convert to array
    img = img / 255.0  # Normalize pixel values
    img = np.expand_dims(img, axis=0)  # Add batch dimension

    # Make prediction
    prediction = model.predict(img)

    # Determine if infected or healthy
    if prediction[0][0] > 0.5:
        return "The plant is healthy."
    else:
        return "The plant is infected."

# Gradio interface
interface = gr.Interface(
    fn=predict_disease,
    inputs=gr.Image(type="pil"),  # Corrected input without 'shape' argument
    outputs=gr.Textbox(),  # Output component
    title="Tomato Plant Disease Detection",
    description="Upload an image of a tomato leaf to determine if it's infected or healthy."
)

# Launch the Gradio interface
interface.launch()