Alzheimer's Disease Prediction (VGG16).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "dd27db85",
   "metadata": {},
   "source": [
    "# <center> Alzheimer's Disease Prediction (VGG16) </center>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "024681cf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.8.0\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "print(tf.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ea9f75af",
   "metadata": {},
   "outputs": [],
   "source": [
    "# import the libraries\n",
    "\n",
    "from tensorflow.keras.layers import Input, Lambda, Dense, Flatten\n",
    "from tensorflow.keras.models import Model\n",
    "from tensorflow.keras.applications.vgg16 import VGG16\n",
    "from tensorflow.keras.preprocessing import image\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator,load_img\n",
    "from tensorflow.keras.models import Sequential\n",
    "import numpy as np\n",
    "from glob import glob\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "7696a143",
   "metadata": {},
   "outputs": [],
   "source": [
    "# re-size all the images to this\n",
    "IMAGE_SIZE = [224, 224]\n",
    "\n",
    "train_path = 'Alzheimer_s Dataset/train'\n",
    "valid_path = 'Alzheimer_s Dataset/test'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1473edaf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "A local file was found, but it seems to be incomplete or outdated because the auto file hash does not match the original value of 6d6bbae143d832006294945121d1f1fc so we will re-download the data.\n",
      "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg16/vgg16_weights_tf_dim_ordering_tf_kernels_notop.h5\n",
      "58892288/58889256 [==============================] - 902s 15us/step\n",
      "58900480/58889256 [==============================] - 902s 15us/step\n"
     ]
    }
   ],
   "source": [
    "# Import the VGG16 library as shown below and add preprocessing layer to the front of VGG\n",
    "# Here we will be using imagenet weights\n",
    "\n",
    "vgg16 = VGG16(input_shape=IMAGE_SIZE + [3], weights='imagenet', include_top=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0c2eb838",
   "metadata": {},
   "outputs": [],
   "source": [
    "# don't train existing weights\n",
    "for layer in vgg16.layers:\n",
    "    layer.trainable = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "cf91a401",
   "metadata": {},
   "outputs": [],
   "source": [
    "  # useful for getting number of output classes\n",
    "folders = glob('Alzheimer_s Dataset/train/*')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c294a87a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Alzheimer_s Dataset/train\\\\MildDemented',\n",
       " 'Alzheimer_s Dataset/train\\\\ModerateDemented',\n",
       " 'Alzheimer_s Dataset/train\\\\NonDemented',\n",
       " 'Alzheimer_s Dataset/train\\\\VeryMildDemented']"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "folders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "0c00ba58",
   "metadata": {},
   "outputs": [],
   "source": [
    "# our layers - you can add more if you want\n",
    "x = Flatten()(vgg16.output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8174ef3b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(folders)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "bc3a701a",
   "metadata": {},
   "outputs": [],
   "source": [
    "prediction = Dense(len(folders), activation='softmax')(x)\n",
    "\n",
    "# create a model object\n",
    "model = Model(inputs=vgg16.input, outputs=prediction)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e2dbe879",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_1 (InputLayer)        [(None, 224, 224, 3)]     0         \n",
      "                                                                 \n",
      " block1_conv1 (Conv2D)       (None, 224, 224, 64)      1792      \n",
      "                                                                 \n",
      " block1_conv2 (Conv2D)       (None, 224, 224, 64)      36928     \n",
      "                                                                 \n",
      " block1_pool (MaxPooling2D)  (None, 112, 112, 64)      0         \n",
      "                                                                 \n",
      " block2_conv1 (Conv2D)       (None, 112, 112, 128)     73856     \n",
      "                                                                 \n",
      " block2_conv2 (Conv2D)       (None, 112, 112, 128)     147584    \n",
      "                                                                 \n",
      " block2_pool (MaxPooling2D)  (None, 56, 56, 128)       0         \n",
      "                                                                 \n",
      " block3_conv1 (Conv2D)       (None, 56, 56, 256)       295168    \n",
      "                                                                 \n",
      " block3_conv2 (Conv2D)       (None, 56, 56, 256)       590080    \n",
      "                                                                 \n",
      " block3_conv3 (Conv2D)       (None, 56, 56, 256)       590080    \n",
      "                                                                 \n",
      " block3_pool (MaxPooling2D)  (None, 28, 28, 256)       0         \n",
      "                                                                 \n",
      " block4_conv1 (Conv2D)       (None, 28, 28, 512)       1180160   \n",
      "                                                                 \n",
      " block4_conv2 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
      "                                                                 \n",
      " block4_conv3 (Conv2D)       (None, 28, 28, 512)       2359808   \n",
      "                                                                 \n",
      " block4_pool (MaxPooling2D)  (None, 14, 14, 512)       0         \n",
      "                                                                 \n",
      " block5_conv1 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
      "                                                                 \n",
      " block5_conv2 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
      "                                                                 \n",
      " block5_conv3 (Conv2D)       (None, 14, 14, 512)       2359808   \n",
      "                                                                 \n",
      " block5_pool (MaxPooling2D)  (None, 7, 7, 512)         0         \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 25088)             0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 4)                 100356    \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 14,815,044\n",
      "Trainable params: 100,356\n",
      "Non-trainable params: 14,714,688\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# view the structure of the model\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ddaa4714",
   "metadata": {},
   "outputs": [],
   "source": [
    "# tell the model what cost and optimization method to use\n",
    "model.compile(\n",
    "  loss='categorical_crossentropy',\n",
    "  optimizer='adam',\n",
    "  metrics=['accuracy']\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "78c9f9be",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use the Image Data Generator to import the images from the dataset\n",
    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
    "\n",
    "train_datagen = ImageDataGenerator(rescale = 1./255,\n",
    "                                   shear_range = 0.2,\n",
    "                                   zoom_range = 0.2,\n",
    "                                   horizontal_flip = True)\n",
    "\n",
    "test_datagen = ImageDataGenerator(rescale = 1./255)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "6645278a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 5121 images belonging to 4 classes.\n"
     ]
    }
   ],
   "source": [
    "# Make sure you provide the same target size as initialied for the image size\n",
    "training_set = train_datagen.flow_from_directory('Alzheimer_s Dataset/train',\n",
    "                                                 target_size = (224, 224),\n",
    "                                                 batch_size = 32,\n",
    "                                                 class_mode = 'categorical')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "81c2f1c1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 1279 images belonging to 4 classes.\n"
     ]
    }
   ],
   "source": [
    "test_set = test_datagen.flow_from_directory('Alzheimer_s Dataset/test',\n",
    "                                            target_size = (224, 224),\n",
    "                                            batch_size = 32,\n",
    "                                            class_mode = 'categorical')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "59cf0e89",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\sanja\\AppData\\Local\\Temp/ipykernel_5640/2724883676.py:3: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
      "  r = model.fit_generator(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "161/161 [==============================] - 1044s 6s/step - loss: 1.0667 - accuracy: 0.5556 - val_loss: 0.9595 - val_accuracy: 0.6044\n",
      "Epoch 2/10\n",
      "161/161 [==============================] - 1014s 6s/step - loss: 0.8417 - accuracy: 0.6401 - val_loss: 0.8355 - val_accuracy: 0.6192\n",
      "Epoch 3/10\n",
      "161/161 [==============================] - 1016s 6s/step - loss: 0.7507 - accuracy: 0.6778 - val_loss: 0.8582 - val_accuracy: 0.6286\n",
      "Epoch 4/10\n",
      "161/161 [==============================] - 1011s 6s/step - loss: 0.6230 - accuracy: 0.7295 - val_loss: 0.8144 - val_accuracy: 0.6458\n",
      "Epoch 5/10\n",
      "161/161 [==============================] - 968s 6s/step - loss: 0.6311 - accuracy: 0.7290 - val_loss: 0.9031 - val_accuracy: 0.6357\n",
      "Epoch 6/10\n",
      "161/161 [==============================] - 966s 6s/step - loss: 0.5846 - accuracy: 0.7467 - val_loss: 0.9719 - val_accuracy: 0.6302\n",
      "Epoch 7/10\n",
      "161/161 [==============================] - 969s 6s/step - loss: 0.5760 - accuracy: 0.7538 - val_loss: 0.9247 - val_accuracy: 0.6278\n",
      "Epoch 8/10\n",
      "161/161 [==============================] - 997s 6s/step - loss: 0.5099 - accuracy: 0.7840 - val_loss: 0.8322 - val_accuracy: 0.6497\n",
      "Epoch 9/10\n",
      "161/161 [==============================] - 999s 6s/step - loss: 0.5178 - accuracy: 0.7852 - val_loss: 0.9416 - val_accuracy: 0.6372\n",
      "Epoch 10/10\n",
      "161/161 [==============================] - 1005s 6s/step - loss: 0.4700 - accuracy: 0.7995 - val_loss: 1.0831 - val_accuracy: 0.6122\n"
     ]
    }
   ],
   "source": [
    "# fit the model\n",
    "# Run the cell. It will take some time to execute\n",
    "r = model.fit_generator(\n",
    "  training_set,\n",
    "  validation_data=test_set,\n",
    "  epochs=10,\n",
    "  steps_per_epoch=len(training_set),\n",
    "  validation_steps=len(test_set)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c1301a9c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the loss\n",
    "plt.plot(r.history['loss'], label='train loss')\n",
    "plt.plot(r.history['val_loss'], label='val loss')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "plt.savefig('LossVal_loss')\n",
    "\n",
    "# plot the accuracy\n",
    "plt.plot(r.history['accuracy'], label='train acc')\n",
    "plt.plot(r.history['val_accuracy'], label='val acc')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "plt.savefig('AccVal_acc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "117e418e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# save it as a h5 file\n",
    "from tensorflow.keras.models import load_model\n",
    "model.save('model_vgg16.h5')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}