veremi-fed-learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "# Misbehavior Detection in Vehicular Networks with Federated Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "pycharm": {
     "name": "#%% md\n"
    }
   },
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "pycharm": {
     "is_executing": true,
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>.container { width:65% !important; }</style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import os\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"0\"\n",
    "os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'\n",
    "\n",
    "import locale\n",
    "import flwr as fl\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from abc import ABC\n",
    "from metrics import *\n",
    "import seaborn as sns\n",
    "import tensorflow as tf\n",
    "from itertools import cycle\n",
    "from tensorflow import keras\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn import preprocessing\n",
    "from sklearn.utils import shuffle\n",
    "from typing import Optional, Tuple, Dict, Any\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import MultiLabelBinarizer\n",
    "from sklearn.metrics import precision_recall_curve, classification_report\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, auc\n",
    "from sklearn.metrics import PrecisionRecallDisplay, RocCurveDisplay, roc_curve\n",
    "from sklearn.metrics import precision_score, recall_score, accuracy_score, f1_score\n",
    "\n",
    "physical_devices = tf.config.list_physical_devices('GPU')\n",
    "try:\n",
    "  tf.config.set_visible_devices(physical_devices[1:], 'GPU')\n",
    "  logical_devices = tf.config.list_logical_devices('GPU')\n",
    "  assert len(logical_devices) == len(physical_devices) - 1\n",
    "except:\n",
    "  pass\n",
    "\n",
    "tf.get_logger().setLevel('ERROR')\n",
    "locale.setlocale(locale.LC_ALL, 'pt_BR.UTF-8')\n",
    "locale._override_localeconv = {'thousands_sep': '.'}\n",
    "\n",
    "from IPython.display import display, HTML\n",
    "display(HTML(\"<style>.container { width:65% !important; }</style>\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Support Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_veremi(csv_file: str, feature: str, label: str, delimiter=','):\n",
    "    # Import VeReMi Dataset\n",
    "    data = pd.read_csv(csv_file, delimiter=delimiter)\n",
    "\n",
    "    # select columns\n",
    "    columns = []\n",
    "    for column in data.columns.values:\n",
    "        if feature == 'feat1':\n",
    "            if 'RSSI' in column:\n",
    "                columns.append(column)\n",
    "            elif 'distance' in column:\n",
    "                columns.append(column)\n",
    "        elif feature == 'feat2':\n",
    "            if 'conformity' in column and '0' not in column:\n",
    "                columns.append(column)\n",
    "        elif feature == 'feat3':\n",
    "            if 'RSSI' in column and '0' not in column:\n",
    "                columns.append(column)\n",
    "            elif 'distance' in column and '0' not in column:\n",
    "                columns.append(column)\n",
    "            elif 'conformity' in column and '0' not in column:\n",
    "                columns.append(column)\n",
    "        elif feature == 'feat4':\n",
    "            if 'RSSI' in column:\n",
    "                columns.append(column)\n",
    "            elif 'aoa' in column:\n",
    "                columns.append(column)\n",
    "            elif 'distance' in column:\n",
    "                columns.append(column)\n",
    "            elif 'conformity' in column and '0' not in column:\n",
    "                columns.append(column)\n",
    "    columns.append('attackerType')\n",
    "\n",
    "    # process target values\n",
    "    if label == 'multiclass':\n",
    "        data = data[columns]\n",
    "    elif label == 'binary':\n",
    "        pos_label = 1\n",
    "        data = data[columns]\n",
    "        data['attackerType'].loc[data['attackerType'] != 0] = pos_label\n",
    "    else:\n",
    "        pos_label = int(label.split(\"_\")[1])\n",
    "        data = data[columns]\n",
    "        data = data.loc[(data['attackerType'] == 0) | (data['attackerType'] == pos_label)]\n",
    "\n",
    "    data_normal = data.loc[data['attackerType'] == 0]\n",
    "    data_atk = data.loc[data['attackerType'] != 0]\n",
    "    # atk_size = int(data_atk.shape[0] * 1.5)\n",
    "    atk_size = int(data_atk.shape[0])\n",
    "    data = pd.concat([data_normal.sample(atk_size), data_atk])\n",
    "    data = shuffle(data)\n",
    "\n",
    "    dataset = data\n",
    "    target = data[data.columns[-1:]]\n",
    "    data = data[data.columns[0:-1]]\n",
    "\n",
    "    # normalize data\n",
    "    data = (data - data.mean()) / data.std()\n",
    "\n",
    "    # label binarize one-hot style\n",
    "    lb = preprocessing.LabelBinarizer()\n",
    "    lb.fit(target)\n",
    "    if label == 'multiclass':\n",
    "        target = lb.transform(target)\n",
    "    else:\n",
    "        target = lb.transform(target)\n",
    "        target = MultiLabelBinarizer().fit_transform(target)\n",
    "\n",
    "    # Create training and test data\n",
    "    train_data, test_data, train_labels, test_labels = train_test_split(\n",
    "        data,\n",
    "        target,\n",
    "        train_size=Config.data_train_size,\n",
    "        test_size=Config.data_test_size,\n",
    "        # random_state=42\n",
    "    )\n",
    "\n",
    "    return train_data, test_data, train_labels, test_labels, lb, dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Config Class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Config:\n",
    "    csv = \"../dismiss-bsm-globecom2022-main/models/globecom/dismiss/02bsm/preprocessing/allmsg/allMsg-new-preds.csv\"\n",
    "    model_type = \"mlp\"\n",
    "    label = \"multiclass\"\n",
    "    feature = \"feat4\"\n",
    "    batch_size = 128\n",
    "    epochs = 50\n",
    "    rounds = 80\n",
    "    learning_rate = 3e-4 #1e-3\n",
    "    min_available_clients = 2\n",
    "    fraction_fit = 0.1\n",
    "    early_stop_patience = 3\n",
    "    early_stop_monitor = \"loss\"\n",
    "    early_stop_min_delta = 1e-4\n",
    "    early_stop_restore_best_weights = True\n",
    "    data_train_size = 0.8\n",
    "    data_test_size = 0.2\n",
    "    output_path = f\"results/{feature}/{label}/\"\n",
    "    performance_file = \"performance.csv\"\n",
    "    weights_file = \"model_weights.npz\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def recall(y_true, y_pred):\n",
    "    true_positives = K.sum(K.round(K.clip(y_true * y_pred, 0, 1)))\n",
    "    possible_positives = K.sum(K.round(K.clip(y_true, 0, 1)))\n",
    "    recall_keras = true_positives / (possible_positives + K.epsilon())\n",
    "    return recall_keras\n",
    "\n",
    "\n",
    "def precision(y_true, y_pred):\n",
    "    true_positives = K.sum(K.round(K.clip(y_true * y_pred, 0, 1)))\n",
    "    predicted_positives = K.sum(K.round(K.clip(y_pred, 0, 1)))\n",
    "    precision_keras = true_positives / (predicted_positives + K.epsilon())\n",
    "    return precision_keras\n",
    "\n",
    "\n",
    "def f1(y_true, y_pred):\n",
    "    p = precision(y_true, y_pred)\n",
    "    r = recall(y_true, y_pred)\n",
    "    return 2 * ((p * r) / (p + r + K.epsilon()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## VeReMi Base Class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "class VeremiBase(ABC):\n",
    "    def __init__(self, data_file: str, model_type: str, label: str, feature: str, activation: str = \"softmax\"):\n",
    "        \"\"\" The Veremi Client Constructor\n",
    "            :param model_type: Keras Model Type ('mlp' or 'lstm'\n",
    "            :param label: Model label type ('binary', 'multiclass', 'atk_1', 'atk_2', 'atk_4', 'atk_8', 'atk_16')\n",
    "            :param feature: Feature to evaluate ('feat1', 'feat2', 'feat3')\n",
    "        \"\"\"\n",
    "        self.lb = None\n",
    "        self.dataset = None\n",
    "        self.train_data = None\n",
    "        self.test_data = None\n",
    "        self.train_labels = None\n",
    "        self.test_labels = None\n",
    "        self.model = None\n",
    "        self.data_file = data_file\n",
    "        self.label = label\n",
    "        self.feature = feature\n",
    "        self.model_type = model_type\n",
    "        self.activation = activation\n",
    "\n",
    "        self.load_veremi()\n",
    "        self.create_model()\n",
    "\n",
    "    def create_model(self):\n",
    "        layer1, layer2, layer3, layer4, layer5, layer6, layer7, layer8, layer9, output = \\\n",
    "            None, None, None, None, None, None, None, None, None, None\n",
    "        if self.model_type == 'mlp':\n",
    "            layer1 = keras.layers.Input(shape=(self.train_data.shape[1],))\n",
    "            layer2 = keras.layers.Dense(256, activation=\"relu\")(layer1)\n",
    "            layer3 = keras.layers.Dense(256, activation=\"relu\")(layer2)\n",
    "            layer4 = keras.layers.Dropout(0.5)(layer3)\n",
    "            output = keras.layers.Dense(self.train_labels.shape[1], activation=self.activation)(layer4)\n",
    "        else:\n",
    "            pass\n",
    "\n",
    "        # ML Model\n",
    "        name = self.label + \"-\" + self.model_type + \"-\" + self.feature\n",
    "        self.model = keras.Model(inputs=layer1, outputs=output, name=name)\n",
    "        self.model.compile(\n",
    "            loss=keras.losses.CategoricalCrossentropy(),\n",
    "            optimizer=keras.optimizers.Adam(learning_rate=Config.learning_rate),\n",
    "            metrics=[f1]\n",
    "        )\n",
    "        self.model.summary()\n",
    "\n",
    "    def load_veremi(self):\n",
    "        print(\"Loading dataset in \" + self.__class__.__name__ + \"...\")\n",
    "        self.train_data, self.test_data, self.train_labels, self.test_labels, self.lb, self.dataset = load_veremi(\n",
    "            self.data_file,\n",
    "            feature=self.feature,\n",
    "            label=self.label\n",
    "        )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## VeReMi Client Class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "class VeremiClient(VeremiBase, fl.client.NumPyClient):\n",
    "    def __init__(self, data_file: str, model_type: str, label: str, feature: str):\n",
    "        VeremiBase.__init__(self, data_file, model_type, label, feature)\n",
    "        self.history = None\n",
    "\n",
    "    def get_parameters(self, config):\n",
    "        return self.model.get_weights()\n",
    "    \n",
    "    def fit(self, parameters, config, verbose):\n",
    "        print(\"Training...\")\n",
    "        early_stopping = keras.callbacks.EarlyStopping(\n",
    "            monitor=Config.early_stop_monitor,\n",
    "            patience=Config.early_stop_patience,\n",
    "            min_delta=Config.early_stop_min_delta,\n",
    "            restore_best_weights=Config.early_stop_restore_best_weights\n",
    "        )\n",
    "        if parameters is not None:\n",
    "            self.model.set_weights(parameters)\n",
    "        self.history = self.model.fit(\n",
    "            self.train_data,\n",
    "            self.train_labels,\n",
    "            batch_size=config[\"batch_size\"],\n",
    "            epochs=config[\"epochs\"],\n",
    "            # callbacks=[early_stopping],\n",
    "            validation_data=(self.test_data, self.test_labels),\n",
    "            verbose=verbose,\n",
    "        )\n",
    "        result = {\n",
    "            \"f1_score:\": float(self.history.history['f1'][-1]),\n",
    "            \"f1_val\": float(self.history.history['val_f1'][-1]),\n",
    "        }\n",
    "        print(\"Finished!\")\n",
    "        return self.model.get_weights(), len(self.train_data), result"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create Client"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading dataset in VeremiClient...\n",
      "Model: \"multiclass-mlp-feat4\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_1 (InputLayer)        [(None, 7)]               0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 256)               2048      \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 256)               65792     \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 256)               0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 6)                 1542      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 69,382\n",
      "Trainable params: 69,382\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "client = VeremiClient(Config.csv, Config.model_type, Config.label, Config.feature)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## VeReMi DataSet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": false
   },
   "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>aoa0</th>\n",
       "      <th>aoa1</th>\n",
       "      <th>RSSI0</th>\n",
       "      <th>distance0</th>\n",
       "      <th>RSSI1</th>\n",
       "      <th>distance1</th>\n",
       "      <th>conformity1</th>\n",
       "      <th>attackerType</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1002433</th>\n",
       "      <td>0.138260</td>\n",
       "      <td>0.152957</td>\n",
       "      <td>5.394733e-09</td>\n",
       "      <td>53.162378</td>\n",
       "      <td>4.120589e-08</td>\n",
       "      <td>54.912082</td>\n",
       "      <td>4.276343</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1085519</th>\n",
       "      <td>0.063239</td>\n",
       "      <td>0.028495</td>\n",
       "      <td>8.316230e-09</td>\n",
       "      <td>244.250429</td>\n",
       "      <td>8.497413e-09</td>\n",
       "      <td>244.441396</td>\n",
       "      <td>1.629658</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2250629</th>\n",
       "      <td>0.740756</td>\n",
       "      <td>0.625101</td>\n",
       "      <td>1.406403e-09</td>\n",
       "      <td>148.340674</td>\n",
       "      <td>3.034993e-09</td>\n",
       "      <td>162.209074</td>\n",
       "      <td>0.393618</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>433381</th>\n",
       "      <td>1.159103</td>\n",
       "      <td>1.066895</td>\n",
       "      <td>3.043372e-09</td>\n",
       "      <td>507.826810</td>\n",
       "      <td>4.145469e-09</td>\n",
       "      <td>442.220065</td>\n",
       "      <td>1.082238</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1357341</th>\n",
       "      <td>0.727151</td>\n",
       "      <td>1.547215</td>\n",
       "      <td>1.954551e-09</td>\n",
       "      <td>433.274199</td>\n",
       "      <td>3.432482e-09</td>\n",
       "      <td>299.933505</td>\n",
       "      <td>193.769991</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>495080</th>\n",
       "      <td>0.147886</td>\n",
       "      <td>0.305241</td>\n",
       "      <td>5.322183e-09</td>\n",
       "      <td>194.581156</td>\n",
       "      <td>1.089669e-08</td>\n",
       "      <td>196.801445</td>\n",
       "      <td>0.369332</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2431721</th>\n",
       "      <td>0.077762</td>\n",
       "      <td>1.112885</td>\n",
       "      <td>2.371038e-09</td>\n",
       "      <td>376.907880</td>\n",
       "      <td>3.573118e-09</td>\n",
       "      <td>420.532068</td>\n",
       "      <td>30.158390</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>710112</th>\n",
       "      <td>0.426267</td>\n",
       "      <td>0.447813</td>\n",
       "      <td>2.860802e-09</td>\n",
       "      <td>558.542740</td>\n",
       "      <td>3.496405e-09</td>\n",
       "      <td>492.008956</td>\n",
       "      <td>14.213322</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>991808</th>\n",
       "      <td>1.206235</td>\n",
       "      <td>1.239401</td>\n",
       "      <td>1.440438e-09</td>\n",
       "      <td>401.754144</td>\n",
       "      <td>1.340471e-08</td>\n",
       "      <td>254.636633</td>\n",
       "      <td>0.196537</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1487338</th>\n",
       "      <td>0.390249</td>\n",
       "      <td>0.405875</td>\n",
       "      <td>1.134450e-08</td>\n",
       "      <td>322.145467</td>\n",
       "      <td>3.046832e-09</td>\n",
       "      <td>307.282886</td>\n",
       "      <td>0.930178</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>992438 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             aoa0      aoa1         RSSI0   distance0         RSSI1  \\\n",
       "1002433  0.138260  0.152957  5.394733e-09   53.162378  4.120589e-08   \n",
       "1085519  0.063239  0.028495  8.316230e-09  244.250429  8.497413e-09   \n",
       "2250629  0.740756  0.625101  1.406403e-09  148.340674  3.034993e-09   \n",
       "433381   1.159103  1.066895  3.043372e-09  507.826810  4.145469e-09   \n",
       "1357341  0.727151  1.547215  1.954551e-09  433.274199  3.432482e-09   \n",
       "...           ...       ...           ...         ...           ...   \n",
       "495080   0.147886  0.305241  5.322183e-09  194.581156  1.089669e-08   \n",
       "2431721  0.077762  1.112885  2.371038e-09  376.907880  3.573118e-09   \n",
       "710112   0.426267  0.447813  2.860802e-09  558.542740  3.496405e-09   \n",
       "991808   1.206235  1.239401  1.440438e-09  401.754144  1.340471e-08   \n",
       "1487338  0.390249  0.405875  1.134450e-08  322.145467  3.046832e-09   \n",
       "\n",
       "          distance1  conformity1  attackerType  \n",
       "1002433   54.912082     4.276343           0.0  \n",
       "1085519  244.441396     1.629658           0.0  \n",
       "2250629  162.209074     0.393618           0.0  \n",
       "433381   442.220065     1.082238           0.0  \n",
       "1357341  299.933505   193.769991           0.0  \n",
       "...             ...          ...           ...  \n",
       "495080   196.801445     0.369332           0.0  \n",
       "2431721  420.532068    30.158390           0.0  \n",
       "710112   492.008956    14.213322           0.0  \n",
       "991808   254.636633     0.196537           0.0  \n",
       "1487338  307.282886     0.930178           0.0  \n",
       "\n",
       "[992438 rows x 8 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "client.dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>aoa0</th>\n",
       "      <th>aoa1</th>\n",
       "      <th>RSSI0</th>\n",
       "      <th>distance0</th>\n",
       "      <th>RSSI1</th>\n",
       "      <th>distance1</th>\n",
       "      <th>conformity1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>793950.000000</td>\n",
       "      <td>793950.000000</td>\n",
       "      <td>793950.000000</td>\n",
       "      <td>793950.000000</td>\n",
       "      <td>793950.000000</td>\n",
       "      <td>793950.000000</td>\n",
       "      <td>793950.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.000217</td>\n",
       "      <td>0.000754</td>\n",
       "      <td>0.000859</td>\n",
       "      <td>-0.000906</td>\n",
       "      <td>0.000498</td>\n",
       "      <td>-0.000778</td>\n",
       "      <td>-0.000313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.000023</td>\n",
       "      <td>1.000272</td>\n",
       "      <td>1.012868</td>\n",
       "      <td>0.998497</td>\n",
       "      <td>1.071826</td>\n",
       "      <td>0.998686</td>\n",
       "      <td>0.999941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-1.352462</td>\n",
       "      <td>-1.355454</td>\n",
       "      <td>-0.094098</td>\n",
       "      <td>-0.547932</td>\n",
       "      <td>-0.058071</td>\n",
       "      <td>-0.540818</td>\n",
       "      <td>-0.324825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.946884</td>\n",
       "      <td>-0.954011</td>\n",
       "      <td>-0.092890</td>\n",
       "      <td>-0.428503</td>\n",
       "      <td>-0.057287</td>\n",
       "      <td>-0.424873</td>\n",
       "      <td>-0.324745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-0.200205</td>\n",
       "      <td>-0.181305</td>\n",
       "      <td>-0.090278</td>\n",
       "      <td>-0.344394</td>\n",
       "      <td>-0.055654</td>\n",
       "      <td>-0.344246</td>\n",
       "      <td>-0.323539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.002203</td>\n",
       "      <td>0.992884</td>\n",
       "      <td>-0.082442</td>\n",
       "      <td>-0.197108</td>\n",
       "      <td>-0.050918</td>\n",
       "      <td>-0.204802</td>\n",
       "      <td>-0.296791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.613994</td>\n",
       "      <td>1.614379</td>\n",
       "      <td>396.918037</td>\n",
       "      <td>6.754025</td>\n",
       "      <td>779.546138</td>\n",
       "      <td>6.724057</td>\n",
       "      <td>7.301259</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                aoa0           aoa1          RSSI0      distance0  \\\n",
       "count  793950.000000  793950.000000  793950.000000  793950.000000   \n",
       "mean        0.000217       0.000754       0.000859      -0.000906   \n",
       "std         1.000023       1.000272       1.012868       0.998497   \n",
       "min        -1.352462      -1.355454      -0.094098      -0.547932   \n",
       "25%        -0.946884      -0.954011      -0.092890      -0.428503   \n",
       "50%        -0.200205      -0.181305      -0.090278      -0.344394   \n",
       "75%         1.002203       0.992884      -0.082442      -0.197108   \n",
       "max         1.613994       1.614379     396.918037       6.754025   \n",
       "\n",
       "               RSSI1      distance1    conformity1  \n",
       "count  793950.000000  793950.000000  793950.000000  \n",
       "mean        0.000498      -0.000778      -0.000313  \n",
       "std         1.071826       0.998686       0.999941  \n",
       "min        -0.058071      -0.540818      -0.324825  \n",
       "25%        -0.057287      -0.424873      -0.324745  \n",
       "50%        -0.055654      -0.344246      -0.323539  \n",
       "75%        -0.050918      -0.204802      -0.296791  \n",
       "max       779.546138       6.724057       7.301259  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "client.train_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 454,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot: xlabel='aoa0', ylabel='aoa1'>"
      ]
     },
     "execution_count": 454,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "client.dataset.plot.scatter(x='aoa0', y='aoa1', c='attackerType', colormap='viridis')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Attackers:\n",
      "    Total: 992,438\n",
      "    Attack: 496,219 (50.00% of total)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "attack = len(client.dataset[client.dataset.attackerType != 0])\n",
    "normal = len(client.dataset[client.dataset.attackerType == 0])\n",
    "total = attack + normal\n",
    "print('Attackers:\\n    Total: {:,d}\\n    Attack: {:,d} ({:.2f}% of total)\\n'.format(\n",
    "    total, attack, 100 * attack / total))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>aoa0</th>\n",
       "      <th>aoa1</th>\n",
       "      <th>RSSI0</th>\n",
       "      <th>distance0</th>\n",
       "      <th>RSSI1</th>\n",
       "      <th>distance1</th>\n",
       "      <th>conformity1</th>\n",
       "      <th>attackerType</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>971305</th>\n",
       "      <td>-0.595873</td>\n",
       "      <td>-0.604092</td>\n",
       "      <td>-0.139288</td>\n",
       "      <td>-0.472145</td>\n",
       "      <td>-0.227778</td>\n",
       "      <td>-0.475812</td>\n",
       "      <td>-0.453426</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1312667</th>\n",
       "      <td>-0.815763</td>\n",
       "      <td>-0.644929</td>\n",
       "      <td>-0.141341</td>\n",
       "      <td>0.184232</td>\n",
       "      <td>-0.237442</td>\n",
       "      <td>1.737417</td>\n",
       "      <td>2.096229</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>901043</th>\n",
       "      <td>-0.743733</td>\n",
       "      <td>1.013438</td>\n",
       "      <td>-0.109856</td>\n",
       "      <td>-0.545921</td>\n",
       "      <td>-0.220559</td>\n",
       "      <td>-0.417239</td>\n",
       "      <td>-0.451197</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>309776</th>\n",
       "      <td>-0.832594</td>\n",
       "      <td>-0.753864</td>\n",
       "      <td>1.177272</td>\n",
       "      <td>-0.307935</td>\n",
       "      <td>0.330270</td>\n",
       "      <td>-0.306931</td>\n",
       "      <td>-0.452903</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>457617</th>\n",
       "      <td>-0.778091</td>\n",
       "      <td>-0.668504</td>\n",
       "      <td>-0.129908</td>\n",
       "      <td>0.251826</td>\n",
       "      <td>-0.204924</td>\n",
       "      <td>0.311376</td>\n",
       "      <td>-0.448471</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1554467</th>\n",
       "      <td>1.514189</td>\n",
       "      <td>0.274243</td>\n",
       "      <td>-0.137460</td>\n",
       "      <td>-0.570573</td>\n",
       "      <td>-0.222941</td>\n",
       "      <td>-0.615905</td>\n",
       "      <td>-0.395816</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1444396</th>\n",
       "      <td>1.748665</td>\n",
       "      <td>0.066308</td>\n",
       "      <td>-0.139043</td>\n",
       "      <td>0.841497</td>\n",
       "      <td>-0.211853</td>\n",
       "      <td>1.934713</td>\n",
       "      <td>1.391938</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>713431</th>\n",
       "      <td>-0.300315</td>\n",
       "      <td>-0.242710</td>\n",
       "      <td>11.062685</td>\n",
       "      <td>-0.524590</td>\n",
       "      <td>0.325845</td>\n",
       "      <td>-0.545510</td>\n",
       "      <td>-0.453056</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1356880</th>\n",
       "      <td>0.074765</td>\n",
       "      <td>0.293578</td>\n",
       "      <td>-0.123397</td>\n",
       "      <td>2.632511</td>\n",
       "      <td>-0.192787</td>\n",
       "      <td>3.042495</td>\n",
       "      <td>3.408485</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1403136</th>\n",
       "      <td>-0.263606</td>\n",
       "      <td>-0.052494</td>\n",
       "      <td>-0.132743</td>\n",
       "      <td>2.503883</td>\n",
       "      <td>-0.234450</td>\n",
       "      <td>1.690049</td>\n",
       "      <td>0.418905</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>128 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             aoa0      aoa1      RSSI0  distance0     RSSI1  distance1  \\\n",
       "971305  -0.595873 -0.604092  -0.139288  -0.472145 -0.227778  -0.475812   \n",
       "1312667 -0.815763 -0.644929  -0.141341   0.184232 -0.237442   1.737417   \n",
       "901043  -0.743733  1.013438  -0.109856  -0.545921 -0.220559  -0.417239   \n",
       "309776  -0.832594 -0.753864   1.177272  -0.307935  0.330270  -0.306931   \n",
       "457617  -0.778091 -0.668504  -0.129908   0.251826 -0.204924   0.311376   \n",
       "...           ...       ...        ...        ...       ...        ...   \n",
       "1554467  1.514189  0.274243  -0.137460  -0.570573 -0.222941  -0.615905   \n",
       "1444396  1.748665  0.066308  -0.139043   0.841497 -0.211853   1.934713   \n",
       "713431  -0.300315 -0.242710  11.062685  -0.524590  0.325845  -0.545510   \n",
       "1356880  0.074765  0.293578  -0.123397   2.632511 -0.192787   3.042495   \n",
       "1403136 -0.263606 -0.052494  -0.132743   2.503883 -0.234450   1.690049   \n",
       "\n",
       "         conformity1  attackerType  \n",
       "971305     -0.453426           2.0  \n",
       "1312667     2.096229           4.0  \n",
       "901043     -0.451197           2.0  \n",
       "309776     -0.452903           1.0  \n",
       "457617     -0.448471           1.0  \n",
       "...              ...           ...  \n",
       "1554467    -0.395816           8.0  \n",
       "1444396     1.391938           4.0  \n",
       "713431     -0.453056           2.0  \n",
       "1356880     3.408485           4.0  \n",
       "1403136     0.418905           4.0  \n",
       "\n",
       "[128 rows x 8 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "atk_df = client.dataset.loc[client.dataset['attackerType'] != 0].sample(128)\n",
    "normal_df = client.dataset.loc[client.dataset['attackerType'] == 0].sample(128)\n",
    "\n",
    "# normalize data\n",
    "columns = atk_df.columns[0:-1]\n",
    "atk_df[columns] = (atk_df[columns] - atk_df[columns].mean()) / atk_df[columns].std()\n",
    "\n",
    "columns = normal_df.columns[0:-1]\n",
    "normal_df[columns] = (normal_df[columns] - normal_df[columns].mean()) / normal_df[columns].std()\n",
    "\n",
    "atk_df"
   ]
  },
  {
   "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>aoa0</th>\n",
       "      <th>aoa1</th>\n",
       "      <th>RSSI0</th>\n",
       "      <th>distance0</th>\n",
       "      <th>RSSI1</th>\n",
       "      <th>distance1</th>\n",
       "      <th>conformity1</th>\n",
       "      <th>attackerType</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.280000e+02</td>\n",
       "      <td>1.280000e+02</td>\n",
       "      <td>1.280000e+02</td>\n",
       "      <td>1.280000e+02</td>\n",
       "      <td>128.000000</td>\n",
       "      <td>128.000000</td>\n",
       "      <td>1.280000e+02</td>\n",
       "      <td>128.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.857226e-17</td>\n",
       "      <td>-6.938894e-17</td>\n",
       "      <td>-6.938894e-18</td>\n",
       "      <td>5.204170e-17</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.163336e-17</td>\n",
       "      <td>5.929688</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>5.432744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-1.378752e+00</td>\n",
       "      <td>-1.356575e+00</td>\n",
       "      <td>-1.435830e-01</td>\n",
       "      <td>-6.581674e-01</td>\n",
       "      <td>-0.239196</td>\n",
       "      <td>-0.684918</td>\n",
       "      <td>-4.534258e-01</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-8.135285e-01</td>\n",
       "      <td>-7.592344e-01</td>\n",
       "      <td>-1.394745e-01</td>\n",
       "      <td>-5.253949e-01</td>\n",
       "      <td>-0.235329</td>\n",
       "      <td>-0.566232</td>\n",
       "      <td>-4.512734e-01</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-1.730955e-01</td>\n",
       "      <td>-2.399409e-01</td>\n",
       "      <td>-1.327856e-01</td>\n",
       "      <td>-4.132956e-01</td>\n",
       "      <td>-0.225323</td>\n",
       "      <td>-0.413719</td>\n",
       "      <td>-4.402093e-01</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>7.226010e-01</td>\n",
       "      <td>7.773163e-01</td>\n",
       "      <td>-1.148317e-01</td>\n",
       "      <td>-1.200326e-02</td>\n",
       "      <td>-0.204624</td>\n",
       "      <td>0.166655</td>\n",
       "      <td>-3.051479e-01</td>\n",
       "      <td>8.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.922284e+00</td>\n",
       "      <td>2.019748e+00</td>\n",
       "      <td>1.106269e+01</td>\n",
       "      <td>3.703821e+00</td>\n",
       "      <td>7.323738</td>\n",
       "      <td>4.442400</td>\n",
       "      <td>4.281688e+00</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               aoa0          aoa1         RSSI0     distance0       RSSI1  \\\n",
       "count  1.280000e+02  1.280000e+02  1.280000e+02  1.280000e+02  128.000000   \n",
       "mean   4.857226e-17 -6.938894e-17 -6.938894e-18  5.204170e-17    0.000000   \n",
       "std    1.000000e+00  1.000000e+00  1.000000e+00  1.000000e+00    1.000000   \n",
       "min   -1.378752e+00 -1.356575e+00 -1.435830e-01 -6.581674e-01   -0.239196   \n",
       "25%   -8.135285e-01 -7.592344e-01 -1.394745e-01 -5.253949e-01   -0.235329   \n",
       "50%   -1.730955e-01 -2.399409e-01 -1.327856e-01 -4.132956e-01   -0.225323   \n",
       "75%    7.226010e-01  7.773163e-01 -1.148317e-01 -1.200326e-02   -0.204624   \n",
       "max    1.922284e+00  2.019748e+00  1.106269e+01  3.703821e+00    7.323738   \n",
       "\n",
       "        distance1   conformity1  attackerType  \n",
       "count  128.000000  1.280000e+02    128.000000  \n",
       "mean     0.000000  4.163336e-17      5.929688  \n",
       "std      1.000000  1.000000e+00      5.432744  \n",
       "min     -0.684918 -4.534258e-01      1.000000  \n",
       "25%     -0.566232 -4.512734e-01      2.000000  \n",
       "50%     -0.413719 -4.402093e-01      4.000000  \n",
       "75%      0.166655 -3.051479e-01      8.000000  \n",
       "max      4.442400  4.281688e+00     16.000000  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "atk_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>aoa0</th>\n",
       "      <th>aoa1</th>\n",
       "      <th>RSSI0</th>\n",
       "      <th>distance0</th>\n",
       "      <th>RSSI1</th>\n",
       "      <th>distance1</th>\n",
       "      <th>conformity1</th>\n",
       "      <th>attackerType</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2074774</th>\n",
       "      <td>-1.483545</td>\n",
       "      <td>-1.526624</td>\n",
       "      <td>-0.276904</td>\n",
       "      <td>1.727279</td>\n",
       "      <td>-0.244672</td>\n",
       "      <td>1.728277</td>\n",
       "      <td>-0.202042</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2313286</th>\n",
       "      <td>1.213163</td>\n",
       "      <td>1.237717</td>\n",
       "      <td>-0.134657</td>\n",
       "      <td>-1.011312</td>\n",
       "      <td>-0.239282</td>\n",
       "      <td>-0.230963</td>\n",
       "      <td>-0.221379</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>563557</th>\n",
       "      <td>-0.613088</td>\n",
       "      <td>-0.507789</td>\n",
       "      <td>-0.288871</td>\n",
       "      <td>-0.112390</td>\n",
       "      <td>-0.216478</td>\n",
       "      <td>-0.025133</td>\n",
       "      <td>-0.200195</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1948676</th>\n",
       "      <td>-0.640919</td>\n",
       "      <td>-0.721677</td>\n",
       "      <td>-0.269738</td>\n",
       "      <td>-0.864862</td>\n",
       "      <td>-0.247231</td>\n",
       "      <td>-0.729553</td>\n",
       "      <td>-0.202100</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1933294</th>\n",
       "      <td>0.861014</td>\n",
       "      <td>0.849481</td>\n",
       "      <td>0.141877</td>\n",
       "      <td>-1.161676</td>\n",
       "      <td>-0.166583</td>\n",
       "      <td>-1.014234</td>\n",
       "      <td>-0.165892</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>592623</th>\n",
       "      <td>1.206566</td>\n",
       "      <td>1.081213</td>\n",
       "      <td>-0.192866</td>\n",
       "      <td>-0.562665</td>\n",
       "      <td>-0.204714</td>\n",
       "      <td>-1.244263</td>\n",
       "      <td>-0.217526</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>615776</th>\n",
       "      <td>1.006325</td>\n",
       "      <td>0.881764</td>\n",
       "      <td>-0.240655</td>\n",
       "      <td>0.396043</td>\n",
       "      <td>-0.244815</td>\n",
       "      <td>0.404734</td>\n",
       "      <td>-0.221045</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1696611</th>\n",
       "      <td>1.039258</td>\n",
       "      <td>1.067194</td>\n",
       "      <td>-0.231427</td>\n",
       "      <td>0.248086</td>\n",
       "      <td>-0.232031</td>\n",
       "      <td>0.255555</td>\n",
       "      <td>-0.213967</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1199115</th>\n",
       "      <td>0.526742</td>\n",
       "      <td>1.147191</td>\n",
       "      <td>1.807280</td>\n",
       "      <td>-1.404036</td>\n",
       "      <td>-0.006745</td>\n",
       "      <td>-1.360844</td>\n",
       "      <td>-0.096420</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2074086</th>\n",
       "      <td>0.594301</td>\n",
       "      <td>0.563051</td>\n",
       "      <td>-0.277878</td>\n",
       "      <td>1.075587</td>\n",
       "      <td>-0.249542</td>\n",
       "      <td>1.004752</td>\n",
       "      <td>-0.194008</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>128 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             aoa0      aoa1     RSSI0  distance0     RSSI1  distance1  \\\n",
       "2074774 -1.483545 -1.526624 -0.276904   1.727279 -0.244672   1.728277   \n",
       "2313286  1.213163  1.237717 -0.134657  -1.011312 -0.239282  -0.230963   \n",
       "563557  -0.613088 -0.507789 -0.288871  -0.112390 -0.216478  -0.025133   \n",
       "1948676 -0.640919 -0.721677 -0.269738  -0.864862 -0.247231  -0.729553   \n",
       "1933294  0.861014  0.849481  0.141877  -1.161676 -0.166583  -1.014234   \n",
       "...           ...       ...       ...        ...       ...        ...   \n",
       "592623   1.206566  1.081213 -0.192866  -0.562665 -0.204714  -1.244263   \n",
       "615776   1.006325  0.881764 -0.240655   0.396043 -0.244815   0.404734   \n",
       "1696611  1.039258  1.067194 -0.231427   0.248086 -0.232031   0.255555   \n",
       "1199115  0.526742  1.147191  1.807280  -1.404036 -0.006745  -1.360844   \n",
       "2074086  0.594301  0.563051 -0.277878   1.075587 -0.249542   1.004752   \n",
       "\n",
       "         conformity1  attackerType  \n",
       "2074774    -0.202042           0.0  \n",
       "2313286    -0.221379           0.0  \n",
       "563557     -0.200195           0.0  \n",
       "1948676    -0.202100           0.0  \n",
       "1933294    -0.165892           0.0  \n",
       "...              ...           ...  \n",
       "592623     -0.217526           0.0  \n",
       "615776     -0.221045           0.0  \n",
       "1696611    -0.213967           0.0  \n",
       "1199115    -0.096420           0.0  \n",
       "2074086    -0.194008           0.0  \n",
       "\n",
       "[128 rows x 8 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "normal_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.JointGrid at 0x7f98144fd960>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot\n",
    "sns.jointplot(x=atk_df['RSSI0'], y=atk_df['distance0'], kind='hex')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.JointGrid at 0x7f9d26c0aad0>"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot\n",
    "sns.jointplot(x=atk_df['RSSI1'], y=atk_df['distance1'], kind='hex')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fit the MLP Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load Model Params"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 450,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Setting model params...\n"
     ]
    }
   ],
   "source": [
    "initial_parameters = None\n",
    "file = Config.output_path + Config.weights_file\n",
    "if os.path.exists(file):\n",
    "    npzfile = np.load(file)\n",
    "    params = [npzfile[x] for x in npzfile]\n",
    "    params = fl.common.ndarrays_to_parameters(params)\n",
    "    initial_parameters = fl.common.parameters_to_ndarrays(params)\n",
    "    print(\"Setting model params...\")\n",
    "    client.model.set_weights(initial_parameters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training...\n",
      "Epoch 1/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2926 - f1: 0.8995 - val_loss: 0.2710 - val_f1: 0.9055\n",
      "Epoch 2/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2931 - f1: 0.8991 - val_loss: 0.2739 - val_f1: 0.9041\n",
      "Epoch 3/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2922 - f1: 0.8992 - val_loss: 0.2725 - val_f1: 0.9052\n",
      "Epoch 4/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2920 - f1: 0.8993 - val_loss: 0.2731 - val_f1: 0.9044\n",
      "Epoch 5/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2918 - f1: 0.8992 - val_loss: 0.2748 - val_f1: 0.9055\n",
      "Epoch 6/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2911 - f1: 0.8999 - val_loss: 0.2733 - val_f1: 0.9041\n",
      "Epoch 7/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2910 - f1: 0.8996 - val_loss: 0.2791 - val_f1: 0.9027\n",
      "Epoch 8/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2907 - f1: 0.8997 - val_loss: 0.2722 - val_f1: 0.9064\n",
      "Epoch 9/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2903 - f1: 0.9000 - val_loss: 0.2786 - val_f1: 0.9022\n",
      "Epoch 10/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2916 - f1: 0.8999 - val_loss: 0.2695 - val_f1: 0.9063\n",
      "Epoch 11/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2897 - f1: 0.8999 - val_loss: 0.2717 - val_f1: 0.9055\n",
      "Epoch 12/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2894 - f1: 0.9000 - val_loss: 0.2744 - val_f1: 0.9048\n",
      "Epoch 13/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2893 - f1: 0.9001 - val_loss: 0.2677 - val_f1: 0.9071\n",
      "Epoch 14/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2896 - f1: 0.9002 - val_loss: 0.2668 - val_f1: 0.9074\n",
      "Epoch 15/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2893 - f1: 0.9002 - val_loss: 0.2671 - val_f1: 0.9069\n",
      "Epoch 16/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2883 - f1: 0.9005 - val_loss: 0.2720 - val_f1: 0.9059\n",
      "Epoch 17/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2887 - f1: 0.9003 - val_loss: 0.2647 - val_f1: 0.9076\n",
      "Epoch 18/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2875 - f1: 0.9006 - val_loss: 0.2698 - val_f1: 0.9061\n",
      "Epoch 19/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2878 - f1: 0.9005 - val_loss: 0.2682 - val_f1: 0.9077\n",
      "Epoch 20/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2873 - f1: 0.9009 - val_loss: 0.2714 - val_f1: 0.9072\n",
      "Epoch 21/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2869 - f1: 0.9011 - val_loss: 0.2666 - val_f1: 0.9070\n",
      "Epoch 22/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2866 - f1: 0.9010 - val_loss: 0.2692 - val_f1: 0.9061\n",
      "Epoch 23/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2864 - f1: 0.9012 - val_loss: 0.2712 - val_f1: 0.9057\n",
      "Epoch 24/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2871 - f1: 0.9011 - val_loss: 0.2712 - val_f1: 0.9051\n",
      "Epoch 25/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2861 - f1: 0.9013 - val_loss: 0.2669 - val_f1: 0.9073\n",
      "Epoch 26/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2854 - f1: 0.9012 - val_loss: 0.2669 - val_f1: 0.9060\n",
      "Epoch 27/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2862 - f1: 0.9011 - val_loss: 0.2740 - val_f1: 0.9056\n",
      "Epoch 28/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2861 - f1: 0.9016 - val_loss: 0.2709 - val_f1: 0.9069\n",
      "Epoch 29/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2859 - f1: 0.9012 - val_loss: 0.2701 - val_f1: 0.9066\n",
      "Epoch 30/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2851 - f1: 0.9014 - val_loss: 0.2609 - val_f1: 0.9099\n",
      "Epoch 31/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2853 - f1: 0.9016 - val_loss: 0.2698 - val_f1: 0.9059\n",
      "Epoch 32/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2852 - f1: 0.9016 - val_loss: 0.2679 - val_f1: 0.9075\n",
      "Epoch 33/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2847 - f1: 0.9016 - val_loss: 0.2630 - val_f1: 0.9082\n",
      "Epoch 34/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2843 - f1: 0.9016 - val_loss: 0.2631 - val_f1: 0.9088\n",
      "Epoch 35/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2851 - f1: 0.9019 - val_loss: 0.2659 - val_f1: 0.9064\n",
      "Epoch 36/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2836 - f1: 0.9021 - val_loss: 0.2721 - val_f1: 0.9058\n",
      "Epoch 37/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2831 - f1: 0.9023 - val_loss: 0.2638 - val_f1: 0.9079\n",
      "Epoch 38/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2831 - f1: 0.9023 - val_loss: 0.2634 - val_f1: 0.9083\n",
      "Epoch 39/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2848 - f1: 0.9024 - val_loss: 0.2652 - val_f1: 0.9085\n",
      "Epoch 40/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2835 - f1: 0.9021 - val_loss: 0.2600 - val_f1: 0.9099\n",
      "Epoch 41/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2827 - f1: 0.9023 - val_loss: 0.2595 - val_f1: 0.9110\n",
      "Epoch 42/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2825 - f1: 0.9021 - val_loss: 0.2775 - val_f1: 0.9024\n",
      "Epoch 43/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2839 - f1: 0.9024 - val_loss: 0.2629 - val_f1: 0.9080\n",
      "Epoch 44/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2823 - f1: 0.9027 - val_loss: 0.2618 - val_f1: 0.9090\n",
      "Epoch 45/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2817 - f1: 0.9026 - val_loss: 0.2623 - val_f1: 0.9088\n",
      "Epoch 46/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2827 - f1: 0.9027 - val_loss: 0.2601 - val_f1: 0.9085\n",
      "Epoch 47/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2815 - f1: 0.9027 - val_loss: 0.2605 - val_f1: 0.9087\n",
      "Epoch 48/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2812 - f1: 0.9028 - val_loss: 0.2633 - val_f1: 0.9079\n",
      "Epoch 49/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2814 - f1: 0.9029 - val_loss: 0.2656 - val_f1: 0.9077\n",
      "Epoch 50/50\n",
      "6203/6203 [==============================] - 9s 1ms/step - loss: 0.2813 - f1: 0.9028 - val_loss: 0.2631 - val_f1: 0.9093\n",
      "Finished!\n"
     ]
    }
   ],
   "source": [
    "results = client.fit(\n",
    "    parameters=None, # initial_parameters,\n",
    "    config={\n",
    "        \"batch_size\": Config.batch_size,\n",
    "        \"epochs\": Config.epochs\n",
    "    },\n",
    "    verbose=1\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_f1_score = client.history.history['f1']\n",
    "test_f1_score = client.history.history['val_f1']\n",
    "train_loss = client.history.history['loss']\n",
    "test_loss = client.history.history['val_loss']\n",
    "epochs = range(1, Config.epochs + 1)\n",
    "loss, num_examples, metrics = results\n",
    "\n",
    "fig, (ax1, ax2) = plt.subplots(1,2)\n",
    "fig.suptitle('Model Performance')\n",
    "fig.set_figwidth(15)\n",
    "fig.set_figheight(5)\n",
    "\n",
    "ax1.plot(epochs, train_loss, '-g', label=\"Train Loss\")\n",
    "ax1.plot(epochs, test_loss, '-b', label=\"Test Loss\")\n",
    "ax1.legend()\n",
    "ax1.set(xlabel='Epochs', ylabel='Loss')\n",
    "ax1.set_title('Loss')\n",
    "\n",
    "ax2.plot(epochs, train_f1_score, '-g', label=\"Train F1 Score\")\n",
    "ax2.plot(epochs, test_f1_score, '-b', label=\"Test F1 Score\")\n",
    "ax2.legend()\n",
    "ax2.set(xlabel='Epochs', ylabel='F1 Score')\n",
    "ax2.set_title('F1-Score')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_pr_roc_curves(probabilities: Any):\n",
    "    n_classes = client.test_labels.shape[1]\n",
    "    # Compute ROC curve and ROC area for each class\n",
    "    fpr = dict()\n",
    "    tpr = dict()\n",
    "    roc_auc = dict()\n",
    "    precision = dict()\n",
    "    recall = dict()\n",
    "    pr_auc = dict()\n",
    "        \n",
    "    for i in range(n_classes):\n",
    "        fpr[i], tpr[i], _ = roc_curve(client.test_labels[:, i], probabilities[:, i])\n",
    "        roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "\n",
    "        precision[i], recall[i], _ = precision_recall_curve(client.test_labels[:, i], probabilities[:, i])\n",
    "        pr_auc[i] = auc(recall[i], precision[i])\n",
    "\n",
    "    # First aggregate all false positive rates\n",
    "    all_precision = np.unique(np.concatenate([precision[i] for i in range(n_classes)]))\n",
    "    all_fpr = np.unique(np.concatenate([fpr[i] for i in range(n_classes)]))\n",
    "\n",
    "    # Then interpolate all ROC curves at this points\n",
    "    mean_tpr = np.zeros_like(all_fpr)\n",
    "    mean_recall = np.zeros_like(all_precision)\n",
    "    for i in range(n_classes):\n",
    "        mean_tpr += np.interp(all_fpr, fpr[i], tpr[i])\n",
    "        mean_recall += np.interp(all_precision, precision[i], recall[i])\n",
    "\n",
    "    # Finally average it and compute AUC\n",
    "    mean_tpr /= n_classes\n",
    "    mean_recall /= n_classes\n",
    "\n",
    "    fpr[\"macro\"] = all_fpr\n",
    "    tpr[\"macro\"] = mean_tpr\n",
    "    roc_auc[\"macro\"] = auc(fpr[\"macro\"], tpr[\"macro\"])\n",
    "    \n",
    "    precision[\"macro\"] = all_precision\n",
    "    recall[\"macro\"] = mean_recall\n",
    "    pr_auc[\"macro\"] = auc(recall[\"macro\"], precision[\"macro\"])\n",
    "\n",
    "    lw = 2\n",
    "    \n",
    "    fig, (ax1, ax2) = plt.subplots(1,2)\n",
    "    fig.suptitle('PR and ROC Curves - Multiclass')\n",
    "    fig.set_figwidth(15)\n",
    "    fig.set_figheight(5)\n",
    "\n",
    "    colors = cycle([\"b\", \"g\", \"r\", \"c\", \"m\", \"y\"])\n",
    "    \n",
    "    # ROC Curve\n",
    "    ax1.plot(\n",
    "        fpr[\"macro\"],\n",
    "        tpr[\"macro\"],\n",
    "        label=\"Macro Avg (area = {0:0.2f})\".format(roc_auc[\"macro\"]),\n",
    "        color=\"navy\",\n",
    "        linestyle=\":\",\n",
    "        linewidth=4,\n",
    "        alpha=0.5,        \n",
    "    )\n",
    "    for i, color in zip(range(n_classes), colors):\n",
    "        label_classes = int(client.lb.classes_[i])\n",
    "        ax1.plot(\n",
    "            fpr[i],\n",
    "            tpr[i],\n",
    "            color=color,\n",
    "            lw=lw,\n",
    "            label=\"Class {0} (area = {1:0.2f})\".format(label_classes, roc_auc[i]),\n",
    "            alpha=0.5\n",
    "        )\n",
    "    ax1.set_title(f\"ROC Curve - {Config.feature} - {Config.label}\")\n",
    "    ax1.set_xlabel(\"False Positive Rate\")\n",
    "    ax1.set_ylabel(\"True Positive Rate\")\n",
    "    ax1.legend()\n",
    "\n",
    "    # PR curve\n",
    "    ax2.plot(\n",
    "        precision[\"macro\"],\n",
    "        recall[\"macro\"],\n",
    "        label=\"Macro Avg (area = {0:0.2f})\".format(pr_auc[\"macro\"]),\n",
    "        color=\"navy\",\n",
    "        linestyle=\":\",\n",
    "        linewidth=4,\n",
    "        alpha=0.5,\n",
    "    )\n",
    "    for i, color in zip(range(n_classes), colors):\n",
    "        label_classes = int(client.lb.classes_[i])\n",
    "        ax2.plot(\n",
    "            precision[i],\n",
    "            recall[i],\n",
    "            color=color,\n",
    "            lw=lw,\n",
    "            label=\"Class {0} (area = {1:0.2f})\".format(label_classes, pr_auc[i]),\n",
    "            alpha=0.5\n",
    "        )\n",
    "    ax2.set_title(f\"PR Curve - {Config.feature} - {Config.label}\")    \n",
    "    ax2.set_xlabel(\"False Positive Rate\")\n",
    "    ax2.set_ylabel(\"True Positive Rate\")\n",
    "    ax2.legend()\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6203/6203 [==============================] - 4s 568us/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "probabilities = client.model.predict(client.test_data)\n",
    "inverse_target = client.lb.inverse_transform(client.test_labels)\n",
    "prediction = None\n",
    "\n",
    "if Config.label == 'multiclass':\n",
    "    prediction = client.lb.inverse_transform(probabilities)\n",
    "    # TODO: PLOT MULTICLASS \n",
    "    plot_pr_roc_curves(probabilities)\n",
    "else:\n",
    "    pos_label = 1 if Config.label == 'binary' else int(Config.label.split(\"_\")[1])\n",
    "    # Best threshold\n",
    "    precision, recall, thresholds = precision_recall_curve(\n",
    "        inverse_target,\n",
    "        probabilities[:, 1],\n",
    "        pos_label=pos_label\n",
    "    )\n",
    "    # convert to f score\n",
    "    np.seterr(divide='ignore', invalid='ignore')\n",
    "    fscore = (2 * precision * recall) / (precision + recall)\n",
    "    np.nan_to_num(fscore, copy=False)\n",
    "    # locate the index of the largest f score\n",
    "    ix = np.argmax(fscore)\n",
    "    print('Best Threshold=%f, F-Score=%.3f' % (thresholds[ix], fscore[ix]))\n",
    "    print(\"-\" * 70)\n",
    "    \n",
    "    prediction = np.where(np.array(probabilities[:, 1]) >= thresholds[ix], pos_label, 0)\n",
    "    \n",
    "    fig, (ax1, ax2) = plt.subplots(1,2)\n",
    "    fig.suptitle('PR and ROC Curves')\n",
    "    fig.set_figwidth(15)\n",
    "    fig.set_figheight(5)\n",
    "\n",
    "    # PR Curve\n",
    "    PrecisionRecallDisplay.from_predictions(inverse_target, probabilities[:, 1], pos_label=pos_label, ax=ax1)\n",
    "    no_skill = len(inverse_target[inverse_target == 1]) / len(inverse_target)\n",
    "    ax1.plot([0, 1], [no_skill, no_skill], linestyle='--', color=\"grey\", label='No Skill')\n",
    "    ax1.scatter(recall[ix], precision[ix], marker='o', color='black', label='Best threshold')\n",
    "    ax1.set_title(f\"PR Curve - {Config.feature} - {Config.label}\")\n",
    "    ax1.legend()\n",
    "\n",
    "    # ROC curve\n",
    "    RocCurveDisplay.from_predictions(inverse_target, probabilities[:, 1], pos_label=pos_label, ax=ax2)\n",
    "    ax2.plot([0, 1], [0, 1], color=\"grey\", lw=1, linestyle=\"--\")\n",
    "    ax2.set_title(f\"ROC Curve - {Config.feature} - {Config.label}\")\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "     class 0      0.889     0.974     0.929     99043\n",
      "     class 1      0.928     0.894     0.911     20652\n",
      "     class 2      0.875     0.641     0.740     20391\n",
      "     class 4      1.000     0.999     1.000     20698\n",
      "     class 8      0.898     0.950     0.923     20551\n",
      "    class 16      0.866     0.638     0.735     17153\n",
      "\n",
      "    accuracy                          0.903    198488\n",
      "   macro avg      0.909     0.849     0.873    198488\n",
      "weighted avg      0.902     0.903     0.898    198488\n",
      "\n",
      "----------------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "classlist = []\n",
    "for cl in client.lb.classes_:\n",
    "    classlist.append('class ' + str(int(cl)))\n",
    "\n",
    "print(classification_report(inverse_target, prediction, target_names=classlist, digits=3, zero_division=0))\n",
    "print(\"-\" * 70)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(inverse_target, prediction, labels=client.lb.classes_)\n",
    "disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=client.lb.classes_)\n",
    "disp.plot()\n",
    "plt.title(f\"Confusion Matrix - {Config.feature} - {Config.label}\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>precision</th>\n",
       "      <th>recall</th>\n",
       "      <th>f1score</th>\n",
       "      <th>accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>multiclass-mlp-feat4</th>\n",
       "      <td>0.909226</td>\n",
       "      <td>0.849492</td>\n",
       "      <td>0.872977</td>\n",
       "      <td>0.90272</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      precision    recall   f1score  accuracy\n",
       "multiclass-mlp-feat4   0.909226  0.849492  0.872977   0.90272"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name = Config.label + \"-\" + Config.model_type + \"-\" + Config.feature\n",
    "if Config.label == 'multiclass':\n",
    "    prscore = precision_score(inverse_target, prediction, average='macro', zero_division=0)\n",
    "    rcscore = recall_score(inverse_target, prediction, average='macro', zero_division=0)\n",
    "    f1score = f1_score(inverse_target, prediction, average='macro', zero_division=0)\n",
    "    accscore = accuracy_score(inverse_target, prediction)\n",
    "else:\n",
    "    prscore = precision_score(inverse_target, prediction, pos_label=pos_label, zero_division=0)\n",
    "    rcscore = recall_score(inverse_target, prediction, pos_label=pos_label, zero_division=0)\n",
    "    f1score = f1_score(inverse_target, prediction, pos_label=pos_label, zero_division=0)\n",
    "    accscore = accuracy_score(inverse_target, prediction)\n",
    "data_performance = {name: [prscore, rcscore, f1score, accscore]}\n",
    "df_performance = pd.DataFrame.from_dict(data_performance, orient='index', columns=[\"precision\", \"recall\", \"f1score\", \"accuracy\"])\n",
    "df_performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}