notebooks/LSTM_AE.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pipeline for the anomaly detection on the SKAB using LSTM-Autoencoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Details regarding the SKAB one can find in the [SKAB repository](https://github.com/waico/SKAB)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# libraries importing\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "# additional modules\n",
    "import sys\n",
    "sys.path.append('../algorithms')\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\", category=UserWarning)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Put the `data` folder from the SKAB repository in the root of this repository. So the path will be following: `../data/`. Or select your own path."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "path_to_data = '../data/'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# benchmark files checking\n",
    "all_files=[]\n",
    "import os\n",
    "for root, dirs, files in os.walk(path_to_data):\n",
    "    for file in files:\n",
    "        if file.endswith(\".csv\"):\n",
    "             all_files.append(os.path.join(root, file))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# datasets with anomalies loading\n",
    "list_of_df = [pd.read_csv(file, \n",
    "                          sep=';', \n",
    "                          index_col='datetime', \n",
    "                          parse_dates=True) for file in all_files if 'anomaly-free' not in file]\n",
    "# anomaly-free df loading\n",
    "anomaly_free_df = pd.read_csv([file for file in all_files if 'anomaly-free' in file][0], \n",
    "                            sep=';', \n",
    "                            index_col='datetime', \n",
    "                            parse_dates=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data description"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SKAB v0.9\n",
      "- A number of datasets: 34\n",
      "- Shape of the random dataset: (1147, 10)\n",
      "- A number of changepoints: 129\n",
      "- A number of outliers (point anomalies): 13067\n",
      "\n",
      "Head of the random dataset:\n"
     ]
    },
    {
     "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>Accelerometer1RMS</th>\n",
       "      <th>Accelerometer2RMS</th>\n",
       "      <th>Current</th>\n",
       "      <th>Pressure</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Thermocouple</th>\n",
       "      <th>Voltage</th>\n",
       "      <th>Volume Flow RateRMS</th>\n",
       "      <th>anomaly</th>\n",
       "      <th>changepoint</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>datetime</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2020-03-01 15:44:06</th>\n",
       "      <td>0.082065</td>\n",
       "      <td>0.133521</td>\n",
       "      <td>1.27794</td>\n",
       "      <td>0.054711</td>\n",
       "      <td>92.2562</td>\n",
       "      <td>22.2577</td>\n",
       "      <td>209.639</td>\n",
       "      <td>76.0197</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Accelerometer1RMS  Accelerometer2RMS  Current  Pressure  \\\n",
       "datetime                                                                       \n",
       "2020-03-01 15:44:06           0.082065           0.133521  1.27794  0.054711   \n",
       "\n",
       "                     Temperature  Thermocouple  Voltage  Volume Flow RateRMS  \\\n",
       "datetime                                                                       \n",
       "2020-03-01 15:44:06      92.2562       22.2577  209.639              76.0197   \n",
       "\n",
       "                     anomaly  changepoint  \n",
       "datetime                                   \n",
       "2020-03-01 15:44:06      0.0          0.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# dataset characteristics printing\n",
    "print('SKAB v0.9')\n",
    "print(f'- A number of datasets: {len(list_of_df)}')\n",
    "print(f'- Shape of the random dataset: {list_of_df[10].shape}')\n",
    "n_cp = sum([len(df[df.changepoint==1.]) for df in list_of_df])\n",
    "n_outlier = sum([len(df[df.anomaly==1.]) for df in list_of_df])\n",
    "print(f'- A number of changepoints: {n_cp}')\n",
    "print(f'- A number of outliers (point anomalies): {n_outlier}\\n')\n",
    "print(f'Head of the random dataset:')\n",
    "display(list_of_df[0].head(1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# random dataset visualizing\n",
    "list_of_df[1].plot(figsize=(12, 3))\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('Value')\n",
    "plt.title('Signals')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Labels visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting the labels both for outlier and changepoint detection problems\n",
    "list_of_df[1].anomaly.plot(figsize=(12, 3))\n",
    "list_of_df[1].changepoint.plot()\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Method"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Method initialization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from LSTM_AE import LSTM_AE \n",
    "from sklearn.preprocessing import StandardScaler\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# hyperparameters selection\n",
    "EPOCHS = 100\n",
    "BATCH_SIZE = 32\n",
    "VAL_SPLIT = 0.1\n",
    "N_STEPS = 10\n",
    "Q = 0.99 # quantile for upper control limit (UCL) selection\n",
    "PARAM = [EPOCHS, BATCH_SIZE, VAL_SPLIT]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = LSTM_AE(PARAM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generated training sequences for use in the model.\n",
    "def create_sequences(values, time_steps=N_STEPS):\n",
    "    output = []\n",
    "    for i in range(len(values) - time_steps + 1):\n",
    "        output.append(values[i : (i + time_steps)])\n",
    "    return np.stack(output)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Method fitting and applying"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13/13 [==============================] - 0s 5ms/step\n",
      "23/23 [==============================] - 0s 6ms/step\n",
      "13/13 [==============================] - 0s 6ms/step\n",
      "42/42 [==============================] - 0s 6ms/step\n",
      "13/13 [==============================] - 1s 6ms/step\n",
      "37/37 [==============================] - 0s 6ms/step\n",
      "13/13 [==============================] - 0s 6ms/step\n",
      "33/33 [==============================] - 0s 6ms/step\n",
      "13/13 [==============================] - 1s 8ms/step\n",
      "29/29 [==============================] - 0s 8ms/step\n",
      "13/13 [==============================] - 0s 7ms/step\n",
      "28/28 [==============================] - 0s 6ms/step\n",
      "13/13 [==============================] - 0s 6ms/step\n",
      "25/25 [==============================] - 0s 6ms/step\n",
      "13/13 [==============================] - 1s 10ms/step\n",
      "36/36 [==============================] - 0s 10ms/step\n",
      "13/13 [==============================] - 1s 11ms/step\n",
      "37/37 [==============================] - 0s 10ms/step\n",
      "13/13 [==============================] - 0s 8ms/step\n",
      "36/36 [==============================] - 0s 8ms/step\n",
      "13/13 [==============================] - 0s 7ms/step\n",
      "36/36 [==============================] - 0s 7ms/step\n",
      "13/13 [==============================] - 0s 11ms/step\n",
      "34/34 [==============================] - 0s 10ms/step\n",
      "13/13 [==============================] - 0s 7ms/step\n",
      "36/36 [==============================] - 0s 8ms/step\n",
      "13/13 [==============================] - 0s 7ms/step\n",
      "36/36 [==============================] - 0s 8ms/step\n",
      "13/13 [==============================] - 0s 7ms/step\n",
      "36/36 [==============================] - 0s 7ms/step\n",
      "13/13 [==============================] - 0s 10ms/step\n",
      "36/36 [==============================] - 0s 10ms/step\n",
      "13/13 [==============================] - 0s 8ms/step\n",
      "36/36 [==============================] - 0s 9ms/step\n",
      "13/13 [==============================] - 0s 7ms/step\n",
      "36/36 [==============================] - 0s 9ms/step\n",
      "13/13 [==============================] - 0s 10ms/step\n",
      "36/36 [==============================] - 0s 10ms/step\n",
      "13/13 [==============================] - 0s 8ms/step\n",
      "36/36 [==============================] - 0s 9ms/step\n",
      "13/13 [==============================] - 0s 8ms/step\n",
      "36/36 [==============================] - 0s 8ms/step\n",
      "13/13 [==============================] - 0s 8ms/step\n",
      "36/36 [==============================] - 0s 8ms/step\n",
      "13/13 [==============================] - 0s 9ms/step\n",
      "34/34 [==============================] - 0s 10ms/step\n",
      "13/13 [==============================] - 0s 8ms/step\n",
      "36/36 [==============================] - 0s 8ms/step\n",
      "13/13 [==============================] - 0s 9ms/step\n",
      "34/34 [==============================] - 0s 8ms/step\n",
      "13/13 [==============================] - 0s 8ms/step\n",
      "36/36 [==============================] - 0s 8ms/step\n",
      "13/13 [==============================] - 0s 10ms/step\n",
      "36/36 [==============================] - 0s 11ms/step\n",
      "13/13 [==============================] - 1s 13ms/step\n",
      "34/34 [==============================] - 0s 10ms/step\n",
      "13/13 [==============================] - 1s 10ms/step\n",
      "36/36 [==============================] - 0s 12ms/step\n",
      "13/13 [==============================] - 1s 15ms/step\n",
      "36/36 [==============================] - 1s 15ms/step\n",
      "13/13 [==============================] - 1s 9ms/step\n",
      "35/35 [==============================] - 0s 10ms/step\n",
      "13/13 [==============================] - 1s 9ms/step\n",
      "33/33 [==============================] - 0s 9ms/step\n",
      "13/13 [==============================] - 1s 9ms/step\n",
      "35/35 [==============================] - 0s 10ms/step\n",
      "13/13 [==============================] - 1s 12ms/step\n",
      "31/31 [==============================] - 0s 12ms/step\n",
      "CPU times: total: 17min 38s\n",
      "Wall time: 9min 18s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# inference\n",
    "predicted_outlier, predicted_cp = [], []\n",
    "for df in list_of_df:\n",
    "    X_train = df[:400].drop(['anomaly','changepoint'], axis=1)\n",
    "    \n",
    "    # scaler init and fitting\n",
    "    StSc = StandardScaler()\n",
    "    StSc.fit(X_train)\n",
    "    \n",
    "    # convert into input/output\n",
    "    X = create_sequences(StSc.transform(X_train), N_STEPS)\n",
    "    \n",
    "    # model fitting\n",
    "    # history, model = arch(X)\n",
    "    model.fit(X)\n",
    "    \n",
    "    # results predicting\n",
    "    residuals = pd.Series(np.sum(np.mean(np.abs(X - model.predict(X)), axis=1), axis=1))\n",
    "    UCL = residuals.quantile(Q) * 3/2\n",
    "    \n",
    "    # results predicting\n",
    "    X = create_sequences(StSc.transform(df.drop(['anomaly','changepoint'], axis=1)), N_STEPS)\n",
    "    cnn_residuals = pd.Series(np.sum(np.mean(np.abs(X - model.predict(X)), axis=1), axis=1))\n",
    "    \n",
    "    # data i is an anomaly if samples [(i - timesteps + 1) to (i)] are anomalies\n",
    "    anomalous_data = cnn_residuals > UCL\n",
    "    anomalous_data_indices = []\n",
    "    for data_idx in range(N_STEPS - 1, len(X) - N_STEPS + 1):\n",
    "        if np.all(anomalous_data[data_idx - N_STEPS + 1 : data_idx]):\n",
    "            anomalous_data_indices.append(data_idx)\n",
    "    \n",
    "    prediction = pd.Series(data=0, index=df.index)\n",
    "    prediction.iloc[anomalous_data_indices] = 1\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)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Results visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# true outlier indices selection\n",
    "true_outlier = [df.anomaly for df in list_of_df]\n",
    "\n",
    "predicted_outlier[0].plot(figsize=(12,3), label='predictions', marker='o', markersize=5)\n",
    "true_outlier[0].plot(marker='o', markersize=2)\n",
    "plt.legend();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# true changepoint indices selection\n",
    "true_cp = [df.changepoint for df in list_of_df]\n",
    "\n",
    "predicted_cp[0].plot(figsize=(12,3), label='predictions', marker='o', markersize=5)\n",
    "true_cp[0].plot(marker='o', markersize=2)\n",
    "plt.legend();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation (metrics calculation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# 'pip install tsad' if needed\n",
    "from tsad.evaluating.evaluating import evaluating"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Binary classification (outlier detection) metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Since you not choose numenta_time and portion, then portion will be 0.1\n",
      "False Alarm Rate 13.97 %\n",
      "Missing Alarm Rate 26.72 %\n",
      "F1 metric 0.74\n"
     ]
    }
   ],
   "source": [
    "# binary classification metrics calculation\n",
    "binary = evaluating(\n",
    "    true_outlier, \n",
    "    predicted_outlier, \n",
    "    metric='binary'\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Changepoint detection metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Intersection of scoring windows(Timestamp('2020-03-01 15:58:10'), Timestamp('2020-03-01 15:54:50'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 18:02:23.600000'), Timestamp('2020-02-08 17:58:38'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 18:09:01.600000'), Timestamp('2020-02-08 18:08:03'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 18:24:52.800000'), Timestamp('2020-02-08 18:20:46'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 18:31:06.800000'), Timestamp('2020-02-08 18:28:37'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 18:48:51.600000'), Timestamp('2020-02-08 18:46:33'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 18:55:37.600000'), Timestamp('2020-02-08 18:51:55'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 19:01:15.800000'), Timestamp('2020-02-08 18:58:30'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 19:06:15.800000'), Timestamp('2020-02-08 19:03:36'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 19:30:25.750000'), Timestamp('2020-02-08 19:26:37'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-01 16:37:42.600000'), Timestamp('2020-03-01 16:34:52'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-01 16:43:24.600000'), Timestamp('2020-03-01 16:40:53'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-01 16:57:52.600000'), Timestamp('2020-03-01 16:54:53'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-01 17:03:52.600000'), Timestamp('2020-03-01 17:00:53'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-01 17:19:07'), Timestamp('2020-03-01 17:14:53'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-01 17:20:07'), Timestamp('2020-03-01 17:19:53'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 16:41:09'), Timestamp('2020-02-08 16:38:12'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 16:47:08'), Timestamp('2020-02-08 16:44:10'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 17:00:59.200000'), Timestamp('2020-02-08 16:59:15'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 17:03:03.200000'), Timestamp('2020-02-08 17:02:12'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 17:06:00.200000'), Timestamp('2020-02-08 17:03:14'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 17:21:12.400000'), Timestamp('2020-02-08 17:17:13'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 17:28:11.400000'), Timestamp('2020-02-08 17:24:14'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 17:41:17.800000'), Timestamp('2020-02-08 17:38:17'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 17:47:18.800000'), Timestamp('2020-02-08 17:44:19'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 10:28:32.800000'), Timestamp('2020-03-09 10:25:33'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 10:34:32.800000'), Timestamp('2020-03-09 10:31:33'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 10:48:33'), Timestamp('2020-03-09 10:45:33'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 10:54:33'), Timestamp('2020-03-09 10:51:34'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 13:48:38.800000'), Timestamp('2020-03-09 13:45:39'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 13:54:38.800000'), Timestamp('2020-03-09 13:51:39'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 14:08:39'), Timestamp('2020-03-09 14:05:40'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 14:14:40'), Timestamp('2020-03-09 14:11:41'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 14:28:40'), Timestamp('2020-03-09 14:25:40'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 14:34:40'), Timestamp('2020-03-09 14:31:41'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 14:48:40.800000'), Timestamp('2020-03-09 14:45:41'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 14:54:40.800000'), Timestamp('2020-03-09 14:51:41'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 15:08:40.800000'), Timestamp('2020-03-09 15:05:41'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 15:14:40.800000'), Timestamp('2020-03-09 15:11:41'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 15:28:41'), Timestamp('2020-03-09 15:25:42'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 15:34:42'), Timestamp('2020-03-09 15:31:42'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 11:10:39.750000'), Timestamp('2020-03-09 11:10:34'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 11:15:33.750000'), Timestamp('2020-03-09 11:11:34'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 11:28:34.200000'), Timestamp('2020-03-09 11:25:35'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 11:34:35.200000'), Timestamp('2020-03-09 11:31:36'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 11:48:35.800000'), Timestamp('2020-03-09 11:45:36'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 11:54:35.800000'), Timestamp('2020-03-09 11:51:36'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 12:08:35.800000'), Timestamp('2020-03-09 12:05:36'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 12:14:36.800000'), Timestamp('2020-03-09 12:11:36'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 12:28:35.800000'), Timestamp('2020-03-09 12:25:36'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 12:34:35.800000'), Timestamp('2020-03-09 12:31:37'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 12:48:37'), Timestamp('2020-03-09 12:45:37'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 12:54:37'), Timestamp('2020-03-09 12:51:38'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 13:08:37.600000'), Timestamp('2020-03-09 13:05:38'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 13:14:37.600000'), Timestamp('2020-03-09 13:11:38'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 13:28:38'), Timestamp('2020-03-09 13:25:38'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 13:34:38'), Timestamp('2020-03-09 13:31:39'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 16:10:29.800000'), Timestamp('2020-03-09 16:07:30'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 16:16:29.800000'), Timestamp('2020-03-09 16:13:30'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 16:30:30'), Timestamp('2020-03-09 16:27:30'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 16:36:32'), Timestamp('2020-03-09 16:33:31'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 16:50:30.800000'), Timestamp('2020-03-09 16:47:31'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 16:56:30.800000'), Timestamp('2020-03-09 16:53:31'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 17:10:02.600000'), Timestamp('2020-03-09 17:07:32'))\n",
      "Intersection of scoring windows(Timestamp('2020-03-09 17:16:03.600000'), Timestamp('2020-03-09 17:13:32'))\n",
      "Amount of true anomalies 129\n",
      "A number of missed CPs = 73\n",
      "A number of FPs = 33\n",
      "Average time 0 days 00:01:14.790178571\n"
     ]
    }
   ],
   "source": [
    "# average detection delay metric calculation\n",
    "add = evaluating(\n",
    "    true_cp, \n",
    "    predicted_cp, \n",
    "    metric='average_time',\n",
    "    anomaly_window_destenation='righter', \n",
    "    portion=1\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Intersection of scoring windows(Timestamp('2020-02-08 18:20:57'), Timestamp('2020-02-08 18:20:31'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 18:51:52'), Timestamp('2020-02-08 18:51:40'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 19:26:43'), Timestamp('2020-02-08 19:26:22'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 17:17:27'), Timestamp('2020-02-08 17:16:58'))\n",
      "Intersection of scoring windows(Timestamp('2020-02-08 17:24:26'), Timestamp('2020-02-08 17:23:59'))\n",
      "Standart  -  20.2\n",
      "LowFP  -  15.33\n",
      "LowFN  -  24.32\n"
     ]
    }
   ],
   "source": [
    "# nab metric calculation\n",
    "nab = evaluating(\n",
    "    true_cp, \n",
    "    predicted_cp, \n",
    "    metric='nab', \n",
    "    numenta_time='30S',\n",
    "    anomaly_window_destenation='center', \n",
    ")"
   ]
  }
 ],
 "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": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}