notebooks/t2_with_q_SKAB.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pipeline of the anomaly detection on the SKAB using combined Hotelling's ($T^2$) and Q (SPE) statistics based on the PCA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Details regarding the SKAB one can find in the [SKAB repository](https://github.com/waico/SKAB)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Libraries importing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# libraries importing\n",
    "import sys\n",
    "import warnings\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "sys.path.append(\"..\")\n",
    "from core.t2 import T2\n",
    "from core.metrics import chp_score\n",
    "from core.utils import load_preprocess_skab, plot_results\n",
    "\n",
    "warnings.filterwarnings(\"ignore\", category=UserWarning)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Xy_traintest_list = load_preprocess_skab()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Method"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Method initialization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# classifier initializing\n",
    "model = T2(scaling=True, using_pca=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# inference\n",
    "predicted_outlier, predicted_cp = [], []\n",
    "true_outlier, true_cp = [], []\n",
    "for X_train, X_test, y_train, y_test in Xy_traintest_list:\n",
    "    # classifier fitting\n",
    "    model.fit(X_train)\n",
    "\n",
    "    # results predicting\n",
    "    model.predict(\n",
    "        X_test,\n",
    "        window_size=5,\n",
    "        plot_fig=False,\n",
    "    )\n",
    "    prediction = pd.Series(\n",
    "        (\n",
    "            (model.t2[\"T2\"].values > model.t2_ucl)\n",
    "            | (model.q[\"Q\"] > model.q_ucl)\n",
    "        ).astype(int),\n",
    "        index=X_test.index,\n",
    "    ).fillna(0)\n",
    "\n",
    "    # predicted outliers saving\n",
    "    predicted_outlier.append(prediction)\n",
    "\n",
    "    # predicted CPs saving\n",
    "    prediction_cp = abs(prediction.diff())\n",
    "    prediction_cp[0] = prediction[0]\n",
    "    predicted_cp.append(prediction_cp)\n",
    "\n",
    "    true_outlier.append(y_test[\"anomaly\"])\n",
    "    true_cp.append(y_test[\"changepoint\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Results visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# [donotremove]\n",
    "plot_results(\n",
    "    (true_outlier[1], predicted_outlier[1]),\n",
    "    (true_cp[1], predicted_cp[1]),\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation (metrics calculation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "\n",
    "pickle.dump(\n",
    "    predicted_outlier,\n",
    "    open(f\"../results/results-{model.__class__.__name__}-q.pkl\", \"wb\"),\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Binary classification (outlier detection) metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "False Alarm Rate 26.62 %\n",
      "Missing Alarm Rate 24.92 %\n",
      "F1 metric 0.76\n"
     ]
    }
   ],
   "source": [
    "# [donotremove]\n",
    "# binary classification metrics calculation\n",
    "binary = chp_score(true_outlier, predicted_outlier, metric=\"binary\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Changepoint detection metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Amount of true anomalies 127\n",
      "A number of missed CPs = 72\n",
      "A number of FPs = 232\n",
      "Average time 0 days 00:00:20.927272727\n"
     ]
    }
   ],
   "source": [
    "# [donotremove]\n",
    "# average detection delay metric calculation\n",
    "add = chp_score(\n",
    "    true_cp,\n",
    "    predicted_cp,\n",
    "    metric=\"average_time\",\n",
    "    window_width=\"60s\",\n",
    "    anomaly_window_destination=\"righter\",\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Standard  -  25.35\n",
      "LowFP  -  14.51\n",
      "LowFN  -  31.33\n"
     ]
    }
   ],
   "source": [
    "# [donotremove]\n",
    "# nab metric calculation\n",
    "nab = chp_score(\n",
    "    true_cp,\n",
    "    predicted_cp,\n",
    "    metric=\"nab\",\n",
    "    window_width=\"60s\",\n",
    "    anomaly_window_destination=\"righter\",\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "",
   "language": "python",
   "name": ""
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.14"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}