AE SWAT.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "# import libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import pickle\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy import stats\n",
    "import tensorflow as tf\n",
    "import seaborn as sns\n",
    "from pylab import rcParams\n",
    "from sklearn.model_selection import train_test_split\n",
    "from keras.models import Model, load_model, Sequential\n",
    "from keras.layers import Input, Dense\n",
    "from keras.layers import Dropout\n",
    "from keras.callbacks import ModelCheckpoint, TensorBoard\n",
    "from keras import regularizers\n",
    "\n",
    "%matplotlib inline\n",
    "sns.set(style='whitegrid', palette='muted', font_scale=1.5)\n",
    "rcParams['figure.figsize'] = 14, 8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_excel(\"C:/Users/Ruben/Datasets/SWAT/SWaT_Dataset_Attack_v0.xlsx\")\n",
    "\n",
    "df.columns = [\n",
    "\t'TIMESTAMP','FIT101','LIT101','MV101','P101','P102','AIT201','AIT202','AIT203','FIT201','MV201','P201','P202','P203',\n",
    "    'P204','P205','P206','DPIT301','FIT301','LIT301','MV301','MV302','MV303','MV304','P301','P302','AIT401','AIT402','FIT401',\n",
    "    'LIT401','P401','P402','P403','P404','UV401','AIT501','AIT502','AIT503','AIT504','FIT501','FIT502','FIT503','FIT504',\n",
    "    'P501','P502','PIT501','PIT502','PIT503','FIT601','P601','P602','P603','OUTCOME'\n",
    "]\n",
    "\n",
    "# The first row only contains labels\n",
    "# 'TIMESTAMP' is irrelevant for the thesis\n",
    "# The other dropped columns contain either only 0s, only 1s or only 2s and are therefore irrelevant\n",
    "\n",
    "df = df.iloc[1:]\n",
    "df = df.drop(['TIMESTAMP', 'P202', 'P301', 'P401', 'P404', 'P502', 'P601', 'P603'], axis = 1)\n",
    "\n",
    "# The dataset labels attacks by 'A ttack' and 'Attack', and labels normal data as 'Normal'\n",
    "# To keep the same structure in all datasets, the 'A ttack' and 'Attack' values are changed to '-1' and normal values to '1'\n",
    "\n",
    "df['OUTCOME'].replace(to_replace = ['A ttack', 'Attack'], value = 1, inplace = True)\n",
    "df['OUTCOME'].replace(to_replace = ['Normal'], value = 0, inplace = True)\n",
    "\n",
    "# data types need to be numeric to be encoded to z-scores --> convert column object data types to numerics\n",
    "\n",
    "cols = df.columns[df.columns != 'OUTCOME']\n",
    "df[cols] = df[cols].apply(pd.to_numeric, errors='coerce')\n",
    "\n",
    "# Encoding the feature vectors to z-scores\n",
    "\n",
    "cols = list(df.columns[df.columns != 'OUTCOME'])\n",
    "for col in cols:\n",
    "    df[col] = ((df[col] - df[col].mean())/df[col].std(ddof=0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating normal and attack masks\n",
    "normal_mask = df[df.OUTCOME == 0]\n",
    "attack_mask = df[df.OUTCOME == 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(316353, 44)\n",
      "(89984, 44)\n",
      "(89984,)\n"
     ]
    }
   ],
   "source": [
    "# Creating X_training and X_testing datasets\n",
    "X_train, X_test = train_test_split(df, test_size = 0.2, random_state = 42)\n",
    "\n",
    "X_train = X_train[X_train.OUTCOME == 0]\n",
    "X_train = X_train.drop(['OUTCOME'], axis=1)\n",
    "\n",
    "y_test = X_test['OUTCOME']\n",
    "X_test = X_test.drop(['OUTCOME'], axis=1)\n",
    "\n",
    "X_train = X_train.values\n",
    "X_test = X_test.values\n",
    "\n",
    "print(X_train.shape)\n",
    "print(X_test.shape)\n",
    "print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense_1 (Dense)              (None, 35)                1575      \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 30)                1080      \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 25)                775       \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 30)                780       \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 35)                1085      \n",
      "_________________________________________________________________\n",
      "dense_6 (Dense)              (None, 44)                1584      \n",
      "=================================================================\n",
      "Total params: 6,879\n",
      "Trainable params: 6,879\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Building the autoencoder model and showing its summary\n",
    "ae = model = Sequential()\n",
    "model.add(Dense(35, input_dim=X_train.shape[1], activation='relu'))\n",
    "model.add(Dense(30, activation='relu'))\n",
    "model.add(Dense(25, activation='relu'))\n",
    "model.add(Dense(30, activation='relu'))\n",
    "model.add(Dense(35, activation='relu'))\n",
    "model.add(Dense(X_train.shape[1]))\n",
    "\n",
    "ae.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 316353 samples, validate on 89984 samples\n",
      "Epoch 1/50\n",
      "316353/316353 [==============================] - 21s 65us/step - loss: 0.0477 - accuracy: 0.7711 - val_loss: 0.3813 - val_accuracy: 0.5939\n",
      "Epoch 2/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0145 - accuracy: 0.8687 - val_loss: 0.1943 - val_accuracy: 0.6975\n",
      "Epoch 3/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0101 - accuracy: 0.8782 - val_loss: 0.1206 - val_accuracy: 0.8315\n",
      "Epoch 4/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0081 - accuracy: 0.8843 - val_loss: 0.1225 - val_accuracy: 0.8384\n",
      "Epoch 5/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0069 - accuracy: 0.8884 - val_loss: 0.0931 - val_accuracy: 0.8631\n",
      "Epoch 6/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0063 - accuracy: 0.8951 - val_loss: 0.1134 - val_accuracy: 0.8778\n",
      "Epoch 7/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0061 - accuracy: 0.9061 - val_loss: 0.0977 - val_accuracy: 0.8763\n",
      "Epoch 8/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0047 - accuracy: 0.9138 - val_loss: 0.1173 - val_accuracy: 0.8469\n",
      "Epoch 9/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0044 - accuracy: 0.9250 - val_loss: 0.1067 - val_accuracy: 0.8553\n",
      "Epoch 10/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0038 - accuracy: 0.9325 - val_loss: 0.0975 - val_accuracy: 0.9033\n",
      "Epoch 11/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0037 - accuracy: 0.9367 - val_loss: 0.0999 - val_accuracy: 0.8858\n",
      "Epoch 12/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0035 - accuracy: 0.9401 - val_loss: 0.1147 - val_accuracy: 0.8587\n",
      "Epoch 13/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0031 - accuracy: 0.9443 - val_loss: 0.1133 - val_accuracy: 0.9214\n",
      "Epoch 14/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0029 - accuracy: 0.9454 - val_loss: 0.1045 - val_accuracy: 0.8992\n",
      "Epoch 15/50\n",
      "316353/316353 [==============================] - 21s 67us/step - loss: 0.0028 - accuracy: 0.9458 - val_loss: 0.1407 - val_accuracy: 0.9098\n",
      "Epoch 16/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0027 - accuracy: 0.9467 - val_loss: 0.1115 - val_accuracy: 0.9096\n",
      "Epoch 17/50\n",
      "316353/316353 [==============================] - 21s 65us/step - loss: 0.0025 - accuracy: 0.9494 - val_loss: 0.1140 - val_accuracy: 0.9062\n",
      "Epoch 18/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0026 - accuracy: 0.9488 - val_loss: 0.1207 - val_accuracy: 0.8545\n",
      "Epoch 19/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0024 - accuracy: 0.9508 - val_loss: 0.0987 - val_accuracy: 0.9175\n",
      "Epoch 20/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0022 - accuracy: 0.9525 - val_loss: 0.1076 - val_accuracy: 0.9203\n",
      "Epoch 21/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0023 - accuracy: 0.9540 - val_loss: 0.1158 - val_accuracy: 0.8791\n",
      "Epoch 22/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0021 - accuracy: 0.9559 - val_loss: 0.1024 - val_accuracy: 0.9240\n",
      "Epoch 23/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0020 - accuracy: 0.9559 - val_loss: 0.1119 - val_accuracy: 0.9148\n",
      "Epoch 24/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0020 - accuracy: 0.9568 - val_loss: 0.1043 - val_accuracy: 0.9263\n",
      "Epoch 25/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0023 - accuracy: 0.9579 - val_loss: 0.1074 - val_accuracy: 0.9242\n",
      "Epoch 26/50\n",
      "316353/316353 [==============================] - 21s 65us/step - loss: 0.0017 - accuracy: 0.9600 - val_loss: 0.1194 - val_accuracy: 0.9258\n",
      "Epoch 27/50\n",
      "316353/316353 [==============================] - 21s 65us/step - loss: 0.0017 - accuracy: 0.9588 - val_loss: 0.1153 - val_accuracy: 0.8478\n",
      "Epoch 28/50\n",
      "316353/316353 [==============================] - 20s 65us/step - loss: 0.0018 - accuracy: 0.9585 - val_loss: 0.1105 - val_accuracy: 0.9202\n",
      "Epoch 29/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0018 - accuracy: 0.9594 - val_loss: 0.1105 - val_accuracy: 0.9287\n",
      "Epoch 30/50\n",
      "316353/316353 [==============================] - 20s 65us/step - loss: 0.0017 - accuracy: 0.9598 - val_loss: 0.1062 - val_accuracy: 0.9044\n",
      "Epoch 31/50\n",
      "316353/316353 [==============================] - 20s 65us/step - loss: 0.0019 - accuracy: 0.9589 - val_loss: 0.1172 - val_accuracy: 0.9264\n",
      "Epoch 32/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0017 - accuracy: 0.9607 - val_loss: 0.1083 - val_accuracy: 0.9183\n",
      "Epoch 33/50\n",
      "316353/316353 [==============================] - 21s 66us/step - loss: 0.0017 - accuracy: 0.9605 - val_loss: 0.1239 - val_accuracy: 0.9053\n",
      "Epoch 34/50\n",
      "316353/316353 [==============================] - 21s 65us/step - loss: 0.0016 - accuracy: 0.9615 - val_loss: 0.1207 - val_accuracy: 0.9331\n",
      "Epoch 35/50\n",
      "316353/316353 [==============================] - 21s 66us/step - loss: 0.0017 - accuracy: 0.9608 - val_loss: 0.1190 - val_accuracy: 0.8777\n",
      "Epoch 36/50\n",
      "316353/316353 [==============================] - 19s 61us/step - loss: 0.0016 - accuracy: 0.9612 - val_loss: 0.1187 - val_accuracy: 0.8920\n",
      "Epoch 37/50\n",
      "316353/316353 [==============================] - 19s 61us/step - loss: 0.0016 - accuracy: 0.9618 - val_loss: 0.1168 - val_accuracy: 0.9110\n",
      "Epoch 38/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0015 - accuracy: 0.9610 - val_loss: 0.1171 - val_accuracy: 0.9349\n",
      "Epoch 39/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0015 - accuracy: 0.9613 - val_loss: 0.1161 - val_accuracy: 0.9156\n",
      "Epoch 40/50\n",
      "316353/316353 [==============================] - 20s 63us/step - loss: 0.0016 - accuracy: 0.9613 - val_loss: 0.1200 - val_accuracy: 0.9210\n",
      "Epoch 41/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0016 - accuracy: 0.9616 - val_loss: 0.1201 - val_accuracy: 0.9215\n",
      "Epoch 42/50\n",
      "316353/316353 [==============================] - 20s 62us/step - loss: 0.0015 - accuracy: 0.9620 - val_loss: 0.1266 - val_accuracy: 0.9267\n",
      "Epoch 43/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0014 - accuracy: 0.9623 - val_loss: 0.1302 - val_accuracy: 0.9259\n",
      "Epoch 44/50\n",
      "316353/316353 [==============================] - 21s 66us/step - loss: 0.0015 - accuracy: 0.9619 - val_loss: 0.1354 - val_accuracy: 0.9320\n",
      "Epoch 45/50\n",
      "316353/316353 [==============================] - 21s 65us/step - loss: 0.0014 - accuracy: 0.9630 - val_loss: 0.1217 - val_accuracy: 0.9231\n",
      "Epoch 46/50\n",
      "316353/316353 [==============================] - 19s 61us/step - loss: 0.0014 - accuracy: 0.9631 - val_loss: 0.1282 - val_accuracy: 0.8995\n",
      "Epoch 47/50\n",
      "316353/316353 [==============================] - 21s 65us/step - loss: 0.0015 - accuracy: 0.9625 - val_loss: 0.1364 - val_accuracy: 0.9054\n",
      "Epoch 48/50\n",
      "316353/316353 [==============================] - 21s 66us/step - loss: 0.0014 - accuracy: 0.9633 - val_loss: 0.1351 - val_accuracy: 0.9046\n",
      "Epoch 49/50\n",
      "316353/316353 [==============================] - 20s 65us/step - loss: 0.0014 - accuracy: 0.9629 - val_loss: 0.1334 - val_accuracy: 0.9249\n",
      "Epoch 50/50\n",
      "316353/316353 [==============================] - 20s 64us/step - loss: 0.0013 - accuracy: 0.9632 - val_loss: 0.1548 - val_accuracy: 0.8465\n"
     ]
    }
   ],
   "source": [
    "# Train the model\n",
    "nb_epoch = 50\n",
    "batch_size = 64\n",
    "\n",
    "ae.compile(optimizer='adam', loss='mean_squared_error', metrics=['accuracy'])\n",
    "\n",
    "checkpointer = ModelCheckpoint(filepath=\"model.SWAT\", verbose=0, save_best_only=True)\n",
    "\n",
    "tensorboard = TensorBoard(log_dir='./logs', histogram_freq=0, write_graph=True, write_images=True)\n",
    "\n",
    "history = ae.fit(X_train, X_train, epochs=nb_epoch, batch_size=batch_size, shuffle=True,\n",
    "                 validation_data=(X_test, X_test), verbose=1, callbacks=[checkpointer, tensorboard]).history"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save the model \n",
    "ae = load_model('model.SWAT')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot model loss over the amount of epochs\n",
    "plt.plot(history['loss'])\n",
    "plt.plot(history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper right');\n",
    "# plt.savefig('loss_SWAT_GoodModel.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting model accuracy over the amount of epochs\n",
    "plt.plot(history['accuracy'])\n",
    "plt.plot(history['val_accuracy'])\n",
    "plt.title('model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'test'], loc='upper right');\n",
    "# plt.savefig('accuracy_SWaT_GoodModel.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculating the predictions of the autoencoder model on X_test testing sample\n",
    "predictions = ae.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculating MSE and reconstruction error\n",
    "mse = np.mean(np.power(X_test - predictions, 2), axis=1)\n",
    "error_df = pd.DataFrame({'reconstruction_error': mse,\n",
    "                        'true_class': y_test})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>reconstruction_error</th>\n",
       "      <th>true_class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>89984.000000</td>\n",
       "      <td>89984.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.093134</td>\n",
       "      <td>0.122677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>3.922728</td>\n",
       "      <td>0.328068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000175</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000827</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.001232</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.002406</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>409.113012</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       reconstruction_error    true_class\n",
       "count          89984.000000  89984.000000\n",
       "mean               0.093134      0.122677\n",
       "std                3.922728      0.328068\n",
       "min                0.000175      0.000000\n",
       "25%                0.000827      0.000000\n",
       "50%                0.001232      0.000000\n",
       "75%                0.002406      0.000000\n",
       "max              409.113012      1.000000"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "error_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "normal_error_df = error_df[(error_df['true_class'] == 0) & (error_df['reconstruction_error'] < 0.1)]\n",
    "_ = ax.hist(normal_error_df.reconstruction_error.values, bins=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "fraud_error_df = error_df[error_df['true_class'] == 1]\n",
    "_ = ax.hist(fraud_error_df.reconstruction_error.values, bins=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import (confusion_matrix, precision_recall_curve, auc,\n",
    "                             roc_curve, recall_score, classification_report, f1_score,\n",
    "                             precision_recall_fscore_support)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Calculating and plotting the ROC curve\n",
    "fpr, tpr, thresholds = roc_curve(error_df.true_class, error_df.reconstruction_error)\n",
    "roc_auc = auc(fpr, tpr)\n",
    "\n",
    "plt.title('Receiver Operating Characteristic')\n",
    "plt.plot(fpr, tpr, label='AUC = %0.4f'% roc_auc)\n",
    "plt.legend(loc='lower right')\n",
    "plt.plot([-1,1],[-1,1],'r--')\n",
    "plt.xlim([-0.001, 1])\n",
    "plt.ylim([0, 1.001])\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.xlabel('False Positive Rate')\n",
    "# plt.savefig('ROC_SWAT2.png')\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting Recall vs Precision plot\n",
    "precision, recall, th = precision_recall_curve(error_df.true_class, error_df.reconstruction_error)\n",
    "plt.plot(recall, precision, 'b', label='Precision-Recall curve')\n",
    "plt.title('Recall vs Precision')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.show()\n",
    "# high precision = low false positive\n",
    "# high recall = low false negative "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting precision metrics for different threshold values to find optimal threshold\n",
    "plt.plot(th, precision[1:], 'b', label='Threshold-Precision curve')\n",
    "plt.title('Precision for different threshold values')\n",
    "plt.xlabel('Threshold')\n",
    "plt.ylabel('Precision')\n",
    "# plt.savefig('precision_threshold_SWAT.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting recall metrics for different threshold values to find optimal threshold\n",
    "plt.plot(th, recall[1:], 'b', label='Threshold-Recall curve')\n",
    "plt.title('Recall for different threshold values')\n",
    "plt.xlabel('Reconstruction error')\n",
    "plt.ylabel('Recall')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setting the manual threshold\n",
    "threshold = 0.03"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot reconstruction error per datapoint index for the manually defined threshold\n",
    "groups = error_df.groupby('true_class')\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "for name, group in groups:\n",
    "    ax.plot(group.index, group.reconstruction_error, marker='o', ms=3.5, linestyle='',\n",
    "            label= \"Attack\" if name == 1 else \"Normal\")\n",
    "ax.hlines(threshold, ax.get_xlim()[0], ax.get_xlim()[1], colors=\"r\", zorder=100, label='Threshold')\n",
    "ax.legend()\n",
    "plt.title(\"Reconstruction error for different classes\")\n",
    "plt.ylabel(\"Reconstruction error\")\n",
    "plt.xlabel(\"Data point index\")\n",
    "plt.show();\n",
    "# plt.savefig('Threshold SWaT.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting the confusion matrix\n",
    "threshold = 0.01\n",
    "LABELS = [\"Normal\", \"Attack\"]\n",
    "y_pred = [1 if e > threshold else 0 for e in error_df.reconstruction_error.values]\n",
    "conf_matrix = confusion_matrix(error_df.true_class, y_pred)\n",
    "\n",
    "plt.figure(figsize=(12, 12))\n",
    "sns.heatmap(conf_matrix, xticklabels=LABELS, yticklabels=LABELS, annot=True, fmt=\"d\");\n",
    "plt.title(\"Confusion matrix\")\n",
    "plt.ylabel('True class')\n",
    "plt.xlabel('Predicted class')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7 (tensorflow)",
   "language": "python",
   "name": "tensorflow"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}