PlantDiseasesVGG16.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "import keras\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Conv2D, MaxPooling2D\n",
    "from keras.layers import Activation, Dropout, Flatten, Dense\n",
    "from keras.applications import VGG16\n",
    "\n",
    "import tensorflow as tf\n",
    "\n",
    "import numpy as np \n",
    "import pandas as pd \n",
    "\n",
    "import os\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "val_path = \"data/extracted/New Plant Diseases Dataset(Augmented)/New Plant Diseases Dataset(Augmented)/valid\"\n",
    "train_path = \"data/extracted/New Plant Diseases Dataset(Augmented)/New Plant Diseases Dataset(Augmented)/train\"\n",
    "dirs = os.listdir(train_path)\n",
    "\n",
    "labels = [folder for folder in dirs]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 70295 images belonging to 38 classes.\n",
      "Found 17572 images belonging to 38 classes.\n"
     ]
    }
   ],
   "source": [
    "train_dir = 'data/extracted/New Plant Diseases Dataset(Augmented)/New Plant Diseases Dataset(Augmented)/train/'\n",
    "val_dir = 'data/extracted/New Plant Diseases Dataset(Augmented)/New Plant Diseases Dataset(Augmented)/valid/'\n",
    "\n",
    "train_dirs = [train_dir + lb for lb in labels]\n",
    "val_dirs = [train_dir + lb for lb in labels]\n",
    "batch_size = 32\n",
    "image_width,image_height = 128,128\n",
    "\n",
    "train_datagen = ImageDataGenerator(rescale=1./255)\n",
    "\n",
    "train_generator = train_datagen.flow_from_directory(\n",
    "    train_dir,\n",
    "    target_size=(image_width,image_height),\n",
    "    batch_size=batch_size,\n",
    "    class_mode='categorical')\n",
    "\n",
    "validation_generator = train_datagen.flow_from_directory(\n",
    "    val_dir,\n",
    "    target_size=(image_width,image_height),\n",
    "    batch_size=batch_size,\n",
    "    class_mode='categorical')\n",
    "\n",
    "\n",
    "nb_train_samples = 70295\n",
    "nb_validation_samples = 17572"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"vgg16\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_2 (InputLayer)         [(None, 128, 128, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 128, 128, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 128, 128, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 64, 64, 64)        0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 64, 64, 128)       73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 64, 64, 128)       147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 32, 32, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 32, 32, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 32, 32, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 32, 32, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 16, 16, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 16, 16, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 16, 16, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 16, 16, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 8, 8, 512)         0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 8, 8, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 8, 8, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 8, 8, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 4, 4, 512)         0         \n",
      "=================================================================\n",
      "Total params: 14,714,688\n",
      "Trainable params: 14,714,688\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "vgg16 = VGG16(weights='imagenet',\n",
    "             include_top=False,\n",
    "             input_shape=(128,128,3))\n",
    "\n",
    "vgg16.summary()\n",
    "vgg16.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(vgg16)\n",
    "model.add(Flatten())\n",
    "model.add(Dense(64))\n",
    "model.add(Activation('relu'))\n",
    "model.add(Dense(38))\n",
    "model.add(Activation('softmax'))\n",
    "\n",
    "model.compile(loss='categorical_crossentropy',\n",
    "              optimizer='rmsprop',\n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "vgg16 (Functional)           (None, 4, 4, 512)         14714688  \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 8192)              0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 64)                524352    \n",
      "_________________________________________________________________\n",
      "activation_2 (Activation)    (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 38)                2470      \n",
      "_________________________________________________________________\n",
      "activation_3 (Activation)    (None, 38)                0         \n",
      "=================================================================\n",
      "Total params: 15,241,510\n",
      "Trainable params: 526,822\n",
      "Non-trainable params: 14,714,688\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "2196/2196 [==============================] - 311s 142ms/step - loss: 0.7966 - accuracy: 0.7636 - val_loss: 0.4046 - val_accuracy: 0.8678\n",
      "Epoch 2/20\n",
      "2196/2196 [==============================] - 311s 142ms/step - loss: 0.3319 - accuracy: 0.8887 - val_loss: 0.3451 - val_accuracy: 0.8863\n",
      "Epoch 3/20\n",
      "2196/2196 [==============================] - 311s 142ms/step - loss: 0.2521 - accuracy: 0.9150 - val_loss: 0.3314 - val_accuracy: 0.8932\n",
      "Epoch 4/20\n",
      "2196/2196 [==============================] - 311s 142ms/step - loss: 0.2046 - accuracy: 0.9307 - val_loss: 0.2911 - val_accuracy: 0.9077\n",
      "Epoch 5/20\n",
      "2196/2196 [==============================] - 309s 141ms/step - loss: 0.1754 - accuracy: 0.9405 - val_loss: 0.4334 - val_accuracy: 0.8765\n",
      "Epoch 6/20\n",
      "2196/2196 [==============================] - 310s 141ms/step - loss: 0.1540 - accuracy: 0.9478 - val_loss: 0.3578 - val_accuracy: 0.8987\n",
      "Epoch 7/20\n",
      "2196/2196 [==============================] - 310s 141ms/step - loss: 0.1349 - accuracy: 0.9535 - val_loss: 0.4236 - val_accuracy: 0.8874\n",
      "Epoch 8/20\n",
      "2196/2196 [==============================] - 313s 143ms/step - loss: 0.1219 - accuracy: 0.9579 - val_loss: 0.3290 - val_accuracy: 0.9106\n",
      "Epoch 9/20\n",
      "2196/2196 [==============================] - 314s 143ms/step - loss: 0.1103 - accuracy: 0.9621 - val_loss: 0.3436 - val_accuracy: 0.9117\n",
      "Epoch 10/20\n",
      "2196/2196 [==============================] - 304s 138ms/step - loss: 0.0981 - accuracy: 0.9662 - val_loss: 0.4054 - val_accuracy: 0.9022\n",
      "Epoch 11/20\n",
      "2196/2196 [==============================] - 305s 139ms/step - loss: 0.0914 - accuracy: 0.9695 - val_loss: 0.3936 - val_accuracy: 0.9099\n",
      "Epoch 12/20\n",
      "2196/2196 [==============================] - 312s 142ms/step - loss: 0.0837 - accuracy: 0.9713 - val_loss: 0.3500 - val_accuracy: 0.9191\n",
      "Epoch 13/20\n",
      "2196/2196 [==============================] - 313s 143ms/step - loss: 0.0761 - accuracy: 0.9743 - val_loss: 0.3862 - val_accuracy: 0.9162\n",
      "Epoch 14/20\n",
      "2196/2196 [==============================] - 312s 142ms/step - loss: 0.0710 - accuracy: 0.9759 - val_loss: 0.4893 - val_accuracy: 0.9016\n",
      "Epoch 15/20\n",
      "2196/2196 [==============================] - 313s 142ms/step - loss: 0.0646 - accuracy: 0.9779 - val_loss: 0.4946 - val_accuracy: 0.9028\n",
      "Epoch 16/20\n",
      "2196/2196 [==============================] - 311s 142ms/step - loss: 0.0602 - accuracy: 0.9796 - val_loss: 0.4464 - val_accuracy: 0.9117\n",
      "Epoch 17/20\n",
      "2196/2196 [==============================] - 312s 142ms/step - loss: 0.0564 - accuracy: 0.9806 - val_loss: 0.3890 - val_accuracy: 0.9235\n",
      "Epoch 18/20\n",
      "2196/2196 [==============================] - 313s 142ms/step - loss: 0.0525 - accuracy: 0.9819 - val_loss: 0.5163 - val_accuracy: 0.9035\n",
      "Epoch 19/20\n",
      "2196/2196 [==============================] - 310s 141ms/step - loss: 0.0506 - accuracy: 0.9834 - val_loss: 0.4099 - val_accuracy: 0.9266\n",
      "Epoch 20/20\n",
      "2196/2196 [==============================] - 311s 142ms/step - loss: 0.0481 - accuracy: 0.9842 - val_loss: 0.4605 - val_accuracy: 0.9198\n"
     ]
    }
   ],
   "source": [
    "history = model.fit(\n",
    "        train_generator,\n",
    "        steps_per_epoch=nb_train_samples // batch_size,\n",
    "        epochs=20,\n",
    "        validation_data=validation_generator,\n",
    "        validation_steps=nb_validation_samples // batch_size,\n",
    "        verbose=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Accuracy per epoch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f70f46ef750>]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(np.arange(0,20),history.history['accuracy'])\n",
    "plt.plot(np.arange(0,20),history.history['val_accuracy'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loss per epoch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f70f46c5190>]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(np.arange(0,20),history.history['loss'])\n",
    "plt.plot(np.arange(0,20),history.history['val_loss'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Citation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ferentinos, Konstantinos. (2018). Deep learning models for plant disease detection and diagnosis. Computers and Electronics in Agriculture. 145. 311-318. 10.1016/j.compag.2018.01.009. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[ResearchGate Link](https://www.researchgate.net/publication/322941653_Deep_learning_models_for_plant_disease_detection_and_diagnosis)"
   ]
  }
 ],
 "metadata": {
  "instance_type": "ml.g4dn.xlarge",
  "kernelspec": {
   "display_name": "Python 3 (TensorFlow 2.3 Python 3.7 GPU Optimized)",
   "language": "python",
   "name": "python3__SAGEMAKER_INTERNAL__arn:aws:sagemaker:eu-central-1:936697816551:image/tensorflow-2.3-gpu-py37-cu110-ubuntu18.04-v3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}