TEP_experiment_with_NAB_metric.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from glob import glob\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "from tqdm import tnrange, tqdm_notebook\n",
    "from IPython.display import clear_output\n",
    "from datetime import datetime, timedelta\n",
    "\n",
    "import ruptures as rpt\n",
    "from utils.custom_cost import CostNew\n",
    "from SKAB_data.evaluating import evaluating_change_point"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1 Data importing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "TEP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "files = sorted(glob('TEP_data/*_te.dat'))\n",
    "\n",
    "columns=[]\n",
    "for i in range(1,42):\n",
    "    columns.append(\"XMEAS({})\".format(i))\n",
    "for i in range(1,12):\n",
    "    columns.append(\"XMV({})\".format(i))\n",
    "    \n",
    "# We do not use the first file because it does not contain a defect\n",
    "test = []\n",
    "for i, j in enumerate(files[1:], start=1):\n",
    "    test.append(pd.read_table(j, sep=\"\\s+\", names=columns))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>XMEAS(1)</th>\n",
       "      <th>XMEAS(2)</th>\n",
       "      <th>XMEAS(3)</th>\n",
       "      <th>XMEAS(4)</th>\n",
       "      <th>XMEAS(5)</th>\n",
       "      <th>XMEAS(6)</th>\n",
       "      <th>XMEAS(7)</th>\n",
       "      <th>XMEAS(8)</th>\n",
       "      <th>XMEAS(9)</th>\n",
       "      <th>XMEAS(10)</th>\n",
       "      <th>...</th>\n",
       "      <th>XMV(2)</th>\n",
       "      <th>XMV(3)</th>\n",
       "      <th>XMV(4)</th>\n",
       "      <th>XMV(5)</th>\n",
       "      <th>XMV(6)</th>\n",
       "      <th>XMV(7)</th>\n",
       "      <th>XMV(8)</th>\n",
       "      <th>XMV(9)</th>\n",
       "      <th>XMV(10)</th>\n",
       "      <th>XMV(11)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.25025</td>\n",
       "      <td>3657.2</td>\n",
       "      <td>4520.1</td>\n",
       "      <td>9.3965</td>\n",
       "      <td>26.715</td>\n",
       "      <td>42.191</td>\n",
       "      <td>2704.5</td>\n",
       "      <td>74.593</td>\n",
       "      <td>120.42</td>\n",
       "      <td>0.33701</td>\n",
       "      <td>...</td>\n",
       "      <td>53.850</td>\n",
       "      <td>24.670</td>\n",
       "      <td>61.839</td>\n",
       "      <td>22.101</td>\n",
       "      <td>40.078</td>\n",
       "      <td>33.041</td>\n",
       "      <td>48.969</td>\n",
       "      <td>47.459</td>\n",
       "      <td>41.841</td>\n",
       "      <td>18.049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.25135</td>\n",
       "      <td>3662.1</td>\n",
       "      <td>4532.3</td>\n",
       "      <td>9.4020</td>\n",
       "      <td>26.644</td>\n",
       "      <td>42.812</td>\n",
       "      <td>2704.9</td>\n",
       "      <td>75.044</td>\n",
       "      <td>120.39</td>\n",
       "      <td>0.33723</td>\n",
       "      <td>...</td>\n",
       "      <td>53.705</td>\n",
       "      <td>24.562</td>\n",
       "      <td>61.348</td>\n",
       "      <td>22.264</td>\n",
       "      <td>40.050</td>\n",
       "      <td>39.154</td>\n",
       "      <td>49.870</td>\n",
       "      <td>47.403</td>\n",
       "      <td>41.188</td>\n",
       "      <td>18.008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.25045</td>\n",
       "      <td>3681.9</td>\n",
       "      <td>4493.6</td>\n",
       "      <td>9.2395</td>\n",
       "      <td>26.971</td>\n",
       "      <td>42.282</td>\n",
       "      <td>2705.0</td>\n",
       "      <td>75.405</td>\n",
       "      <td>120.42</td>\n",
       "      <td>0.33643</td>\n",
       "      <td>...</td>\n",
       "      <td>54.167</td>\n",
       "      <td>24.659</td>\n",
       "      <td>61.826</td>\n",
       "      <td>22.312</td>\n",
       "      <td>40.147</td>\n",
       "      <td>39.294</td>\n",
       "      <td>44.511</td>\n",
       "      <td>47.265</td>\n",
       "      <td>41.290</td>\n",
       "      <td>18.009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.25056</td>\n",
       "      <td>3667.9</td>\n",
       "      <td>4466.0</td>\n",
       "      <td>9.3602</td>\n",
       "      <td>26.937</td>\n",
       "      <td>42.194</td>\n",
       "      <td>2705.9</td>\n",
       "      <td>75.397</td>\n",
       "      <td>120.41</td>\n",
       "      <td>0.33891</td>\n",
       "      <td>...</td>\n",
       "      <td>54.498</td>\n",
       "      <td>24.648</td>\n",
       "      <td>63.168</td>\n",
       "      <td>22.288</td>\n",
       "      <td>39.845</td>\n",
       "      <td>36.334</td>\n",
       "      <td>46.168</td>\n",
       "      <td>47.410</td>\n",
       "      <td>40.997</td>\n",
       "      <td>17.771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.28659</td>\n",
       "      <td>3693.0</td>\n",
       "      <td>4502.5</td>\n",
       "      <td>9.5652</td>\n",
       "      <td>26.999</td>\n",
       "      <td>42.540</td>\n",
       "      <td>2704.3</td>\n",
       "      <td>74.749</td>\n",
       "      <td>120.39</td>\n",
       "      <td>0.34173</td>\n",
       "      <td>...</td>\n",
       "      <td>54.142</td>\n",
       "      <td>28.430</td>\n",
       "      <td>57.566</td>\n",
       "      <td>22.142</td>\n",
       "      <td>40.806</td>\n",
       "      <td>34.882</td>\n",
       "      <td>45.576</td>\n",
       "      <td>47.424</td>\n",
       "      <td>41.290</td>\n",
       "      <td>19.371</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 52 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   XMEAS(1)  XMEAS(2)  XMEAS(3)  XMEAS(4)  XMEAS(5)  XMEAS(6)  XMEAS(7)  \\\n",
       "0   0.25025    3657.2    4520.1    9.3965    26.715    42.191    2704.5   \n",
       "1   0.25135    3662.1    4532.3    9.4020    26.644    42.812    2704.9   \n",
       "2   0.25045    3681.9    4493.6    9.2395    26.971    42.282    2705.0   \n",
       "3   0.25056    3667.9    4466.0    9.3602    26.937    42.194    2705.9   \n",
       "4   0.28659    3693.0    4502.5    9.5652    26.999    42.540    2704.3   \n",
       "\n",
       "   XMEAS(8)  XMEAS(9)  XMEAS(10)  ...  XMV(2)  XMV(3)  XMV(4)  XMV(5)  XMV(6)  \\\n",
       "0    74.593    120.42    0.33701  ...  53.850  24.670  61.839  22.101  40.078   \n",
       "1    75.044    120.39    0.33723  ...  53.705  24.562  61.348  22.264  40.050   \n",
       "2    75.405    120.42    0.33643  ...  54.167  24.659  61.826  22.312  40.147   \n",
       "3    75.397    120.41    0.33891  ...  54.498  24.648  63.168  22.288  39.845   \n",
       "4    74.749    120.39    0.34173  ...  54.142  28.430  57.566  22.142  40.806   \n",
       "\n",
       "   XMV(7)  XMV(8)  XMV(9)  XMV(10)  XMV(11)  \n",
       "0  33.041  48.969  47.459   41.841   18.049  \n",
       "1  39.154  49.870  47.403   41.188   18.008  \n",
       "2  39.294  44.511  47.265   41.290   18.009  \n",
       "3  36.334  46.168  47.410   40.997   17.771  \n",
       "4  34.882  45.576  47.424   41.290   19.371  \n",
       "\n",
       "[5 rows x 52 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test[0].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2 Testing ensembling functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "test[19].plot(legend=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "signal = StandardScaler().fit_transform(test[19].values)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.1 Testing Dynp and DynpEnsembling algorithms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "INDEX = pd.to_datetime(\n",
    "        [datetime(2020,1,1)+timedelta(minutes=m) for m in test[1].index*3]\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dynp algorithm:\n",
      "Average delay nan\n",
      "A number of missed CPs = 1\n",
      "Standart  -  0.0\n",
      "LowFP  -  0.0\n",
      "LowFN  -  0.0\n"
     ]
    }
   ],
   "source": [
    "print('Dynp algorithm:')\n",
    "algo = rpt.Dynp(custom_cost='l1', jump=2).fit(signal)\n",
    "\n",
    "my_bkps = algo.predict(n_bkps=1)\n",
    "\n",
    "true_cp = pd.Series(data=0, index=INDEX)\n",
    "true_cp.iloc[160] = 1\n",
    "\n",
    "predicted_cp = pd.Series(data=0, index=INDEX)\n",
    "predicted_cp[predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "\n",
    "add = evaluating_change_point(true_cp, predicted_cp, metric='average_delay', numenta_time='288 min')\n",
    "nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DynpEnsembling algorithm:\n",
      "Average delay 0 days 03:42:00\n",
      "A number of missed CPs = 0\n",
      "Standart  -  99.54\n",
      "LowFP  -  99.5\n",
      "LowFN  -  99.7\n"
     ]
    }
   ],
   "source": [
    "print('DynpEnsembling algorithm:')\n",
    "c = CostNew()\n",
    "algo = rpt.DynpEnsembling(custom_cost=c, jump=2, ensembling=5).fit(signal)\n",
    "\n",
    "my_bkps = algo.predict(n_bkps=1)\n",
    "\n",
    "true_cp = pd.Series(data=0, index=INDEX)\n",
    "true_cp.iloc[160] = 1\n",
    "\n",
    "predicted_cp = pd.Series(data=0, index=INDEX)\n",
    "predicted_cp[predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "\n",
    "add = evaluating_change_point(true_cp, predicted_cp, metric='average_delay', numenta_time='288 min')\n",
    "nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.2 Testing Window and WindowEnsembling algorithms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Window algorithm:\n",
      "Average delay nan\n",
      "A number of missed CPs = 1\n",
      "Standart  -  0.0\n",
      "LowFP  -  0.0\n",
      "LowFN  -  0.0\n"
     ]
    }
   ],
   "source": [
    "print('Window algorithm:')\n",
    "algo = rpt.Window(custom_cost='ar', jump=1, width=20).fit(signal)\n",
    "\n",
    "my_bkps = algo.predict(n_bkps=1)\n",
    "\n",
    "true_cp = pd.Series(data=0, index=INDEX)\n",
    "true_cp.iloc[160] = 1\n",
    "\n",
    "predicted_cp = pd.Series(data=0, index=INDEX)\n",
    "predicted_cp[predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "\n",
    "add = evaluating_change_point(true_cp, predicted_cp, metric='average_delay', numenta_time='288 min')\n",
    "nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WindowEnsembling algorithm:\n",
      "Average delay nan\n",
      "A number of missed CPs = 1\n",
      "Standart  -  0.0\n",
      "LowFP  -  0.0\n",
      "LowFN  -  0.0\n"
     ]
    }
   ],
   "source": [
    "print('WindowEnsembling algorithm:')\n",
    "c = CostNew()\n",
    "algo = rpt.WindowEnsembling(custom_cost=c, jump=1, width=20, ensembling=5).fit(signal)\n",
    "\n",
    "my_bkps = algo.predict(n_bkps=1)\n",
    "\n",
    "true_cp = pd.Series(data=0, index=INDEX)\n",
    "true_cp.iloc[160] = 1\n",
    "\n",
    "predicted_cp = pd.Series(data=0, index=INDEX)\n",
    "predicted_cp[predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "\n",
    "add = evaluating_change_point(true_cp, predicted_cp, metric='average_delay', numenta_time='288 min')\n",
    "nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2.3 Testing BinSeg and BinSegEnsembling algorithms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BinSeg algorithm:\n",
      "Average delay nan\n",
      "A number of missed CPs = 1\n",
      "Standart  -  0.0\n",
      "LowFP  -  0.0\n",
      "LowFN  -  0.0\n"
     ]
    }
   ],
   "source": [
    "print('BinSeg algorithm:')\n",
    "algo = rpt.Binseg(custom_cost='l2', jump=1).fit(signal)\n",
    "\n",
    "my_bkps = algo.predict(n_bkps=1)\n",
    "\n",
    "true_cp = pd.Series(data=0, index=INDEX)\n",
    "true_cp.iloc[160] = 1\n",
    "\n",
    "predicted_cp = pd.Series(data=0, index=INDEX)\n",
    "predicted_cp[predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "\n",
    "add = evaluating_change_point(true_cp, predicted_cp, metric='average_delay', numenta_time='288 min')\n",
    "nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BinSegEnsembling algorithm:\n",
      "Average delay 0 days 03:42:00\n",
      "A number of missed CPs = 0\n",
      "Standart  -  99.54\n",
      "LowFP  -  99.5\n",
      "LowFN  -  99.7\n"
     ]
    }
   ],
   "source": [
    "print('BinSegEnsembling algorithm:')\n",
    "c = CostNew()\n",
    "algo = rpt.BinsegEnsembling(custom_cost=c, jump=1, ensembling=5).fit(signal)\n",
    "\n",
    "my_bkps = algo.predict(n_bkps=1)\n",
    "\n",
    "true_cp = pd.Series(data=0, index=INDEX)\n",
    "true_cp.iloc[160] = 1\n",
    "\n",
    "predicted_cp = pd.Series(data=0, index=INDEX)\n",
    "predicted_cp[predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "\n",
    "add = evaluating_change_point(true_cp, predicted_cp, metric='average_delay', numenta_time='288 min')\n",
    "nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "NUM_CPDE = 17"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "true_cp = pd.Series(data=0, index=INDEX)\n",
    "true_cp.iloc[160] = 1\n",
    "true_cp = [true_cp]*len(test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 Dynp experiment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.1 Individual algorithms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "def dynp(cost, data, params, true_cp=true_cp):\n",
    "    predicted_cp = []\n",
    "    for dataset in data:\n",
    "        stsc = StandardScaler()\n",
    "        signal = stsc.fit_transform(dataset)\n",
    "        algo = rpt.Dynp(model=cost,\n",
    "                        params=params,\n",
    "                        jump=1)\n",
    "        algo.fit(signal)\n",
    "        my_bkps = algo.predict(n_bkps=1)\n",
    "        \n",
    "        single_predicted_cp = pd.Series(data=0, index=true_cp[0].index)\n",
    "        single_predicted_cp[single_predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "        predicted_cp.append(single_predicted_cp)\n",
    "\n",
    "    nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')\n",
    "    return nab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "models = (\n",
    "    {'cost':'ar', 'params':{'order':1}},\n",
    "    {'cost':'mahalanobis', 'params':{}},\n",
    "    {'cost':'l1', 'params':{}},\n",
    "    {'cost':'l2', 'params':{}},\n",
    "    {'cost':'linear', 'params':{}},\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 4min 21s, sys: 37.3 s, total: 4min 58s\n",
      "Wall time: 1min 8s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "table = []\n",
    "for model in tqdm_notebook(models):\n",
    "    results = dynp(data=test, **model)\n",
    "    clear_output()\n",
    "    table.append(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow0_col0{\n",
       "            background-color:  #1966ad;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow0_col1{\n",
       "            background-color:  #1b69af;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow0_col2,#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow2_col2{\n",
       "            background-color:  #0a549e;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow1_col0,#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow1_col1,#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow1_col2{\n",
       "            background-color:  #08306b;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow2_col0{\n",
       "            background-color:  #0a539e;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow2_col1{\n",
       "            background-color:  #08509b;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow3_col0{\n",
       "            background-color:  #1865ac;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow3_col1{\n",
       "            background-color:  #1764ab;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow3_col2{\n",
       "            background-color:  #135fa7;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow4_col0,#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow4_col1,#T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow4_col2{\n",
       "            background-color:  #f7fbff;\n",
       "            color:  #000000;\n",
       "        }</style><table id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9d\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Standart</th>        <th class=\"col_heading level0 col1\" >LowFP</th>        <th class=\"col_heading level0 col2\" >LowFN</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9dlevel0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow0_col0\" class=\"data row0 col0\" >30.150000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow0_col1\" class=\"data row0 col1\" >28.890000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow0_col2\" class=\"data row0 col2\" >32.800000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9dlevel0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow1_col0\" class=\"data row1 col0\" >36.880000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow1_col1\" class=\"data row1 col1\" >35.820000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow1_col2\" class=\"data row1 col2\" >37.290000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9dlevel0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow2_col0\" class=\"data row2 col0\" >32.530000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow2_col1\" class=\"data row2 col1\" >31.980000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow2_col2\" class=\"data row2 col2\" >32.800000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9dlevel0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow3_col0\" class=\"data row3 col0\" >30.300000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow3_col1\" class=\"data row3 col1\" >29.520000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow3_col2\" class=\"data row3 col2\" >31.310000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9dlevel0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow4_col0\" class=\"data row4 col0\" >4.500000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow4_col1\" class=\"data row4 col1\" >4.240000</td>\n",
       "                        <td id=\"T_34e3aa70_9081_11eb_9aeb_f218985f9c9drow4_col2\" class=\"data row4 col2\" >4.590000</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x1395c4040>"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(table).style.background_gradient(cmap='Blues')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.2 Ensemble"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "def dynpEnsemble(cost, data, num_agg_func, true_cp=true_cp):\n",
    "    predicted_cp = []\n",
    "    for dataset in data:\n",
    "        stsc = StandardScaler()\n",
    "        signal = stsc.fit_transform(dataset)\n",
    "        algo = rpt.DynpEnsembling(custom_cost=cost, \n",
    "                                  jump=1, \n",
    "                                  ensembling=num_agg_func)\n",
    "        single_predicted_cp = pd.Series(data=0, index=true_cp[0].index)\n",
    "        try:\n",
    "            algo.fit(signal)\n",
    "            my_bkps = algo.predict(n_bkps=1)\n",
    "\n",
    "            single_predicted_cp[single_predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "            predicted_cp.append(single_predicted_cp)\n",
    "        except:\n",
    "            predicted_cp.append(single_predicted_cp)\n",
    "\n",
    "    nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')\n",
    "    return nab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1h 58min 49s, sys: 23min 13s, total: 2h 22min 3s\n",
      "Wall time: 24min 47s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "cost = CostNew()\n",
    "table1 = []\n",
    "\n",
    "for n in tnrange(1, NUM_CPDE, desc='agg functions loop'):\n",
    "    table1.append(dynpEnsemble(cost=cost, data=test, num_agg_func=n))\n",
    "    clear_output()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow0_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow0_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow0_col2,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow3_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow3_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow3_col2{\n",
       "            background-color:  #08306b;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow1_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow11_col0{\n",
       "            background-color:  #3989c1;\n",
       "            color:  #000000;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow1_col1{\n",
       "            background-color:  #3181bd;\n",
       "            color:  #000000;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow1_col2,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow11_col2{\n",
       "            background-color:  #3888c1;\n",
       "            color:  #000000;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow2_col0{\n",
       "            background-color:  #4a98c9;\n",
       "            color:  #000000;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow2_col1{\n",
       "            background-color:  #4090c5;\n",
       "            color:  #000000;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow2_col2{\n",
       "            background-color:  #4493c7;\n",
       "            color:  #000000;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow4_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow5_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow8_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow9_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow13_col0{\n",
       "            background-color:  #0c56a0;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow4_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow4_col2,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow5_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow5_col2,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow8_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow8_col2,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow9_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow9_col2,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow13_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow13_col2{\n",
       "            background-color:  #0a539e;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow6_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow10_col0{\n",
       "            background-color:  #0d57a1;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow6_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow10_col1{\n",
       "            background-color:  #0b559f;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow6_col2,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow10_col2{\n",
       "            background-color:  #0a549e;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow7_col0{\n",
       "            background-color:  #1f6eb3;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow7_col1{\n",
       "            background-color:  #1967ad;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow7_col2{\n",
       "            background-color:  #1d6cb1;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow11_col1{\n",
       "            background-color:  #3080bd;\n",
       "            color:  #000000;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow12_col0{\n",
       "            background-color:  #125da6;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow12_col1{\n",
       "            background-color:  #0f5aa3;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow12_col2{\n",
       "            background-color:  #0e58a2;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow14_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow14_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow14_col2,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow15_col0,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow15_col1,#T_c01b057c_9084_11eb_9aeb_f218985f9c9drow15_col2{\n",
       "            background-color:  #f7fbff;\n",
       "            color:  #000000;\n",
       "        }</style><table id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9d\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Standart</th>        <th class=\"col_heading level0 col1\" >LowFP</th>        <th class=\"col_heading level0 col2\" >LowFN</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow0_col0\" class=\"data row0 col0\" >41.810000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow0_col1\" class=\"data row0 col1\" >41.000000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow0_col2\" class=\"data row0 col2\" >42.160000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow1_col0\" class=\"data row1 col0\" >25.660000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow1_col1\" class=\"data row1 col1\" >24.900000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow1_col2\" class=\"data row1 col2\" >26.630000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow2_col0\" class=\"data row2 col0\" >22.850000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow2_col1\" class=\"data row2 col1\" >21.820000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow2_col2\" class=\"data row2 col2\" >24.760000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow3_col0\" class=\"data row3 col0\" >41.810000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow3_col1\" class=\"data row3 col1\" >41.000000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow3_col2\" class=\"data row3 col2\" >42.160000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow4_col0\" class=\"data row4 col0\" >34.800000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow4_col1\" class=\"data row4 col1\" >34.000000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow4_col2\" class=\"data row4 col2\" >35.900000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow5_col0\" class=\"data row5 col0\" >34.830000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow5_col1\" class=\"data row5 col1\" >34.030000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow5_col2\" class=\"data row5 col2\" >35.920000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow6_col0\" class=\"data row6 col0\" >34.730000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow6_col1\" class=\"data row6 col1\" >33.680000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow6_col2\" class=\"data row6 col2\" >35.850000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow7_col0\" class=\"data row7 col0\" >30.470000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow7_col1\" class=\"data row7 col1\" >29.950000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow7_col2\" class=\"data row7 col2\" >31.420000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow8_col0\" class=\"data row8 col0\" >34.800000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow8_col1\" class=\"data row8 col1\" >34.000000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow8_col2\" class=\"data row8 col2\" >35.900000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow9_col0\" class=\"data row9 col0\" >34.830000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow9_col1\" class=\"data row9 col1\" >34.030000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow9_col2\" class=\"data row9 col2\" >35.920000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow10_col0\" class=\"data row10 col0\" >34.730000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow10_col1\" class=\"data row10 col1\" >33.680000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow10_col2\" class=\"data row10 col2\" >35.850000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow11_col0\" class=\"data row11 col0\" >25.590000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow11_col1\" class=\"data row11 col1\" >25.060000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow11_col2\" class=\"data row11 col2\" >26.590000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow12_col0\" class=\"data row12 col0\" >33.510000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow12_col1\" class=\"data row12 col1\" >32.580000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow12_col2\" class=\"data row12 col2\" >35.040000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow13_col0\" class=\"data row13 col0\" >34.830000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow13_col1\" class=\"data row13 col1\" >34.030000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow13_col2\" class=\"data row13 col2\" >35.920000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow14_col0\" class=\"data row14 col0\" >-5.500000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow14_col1\" class=\"data row14 col1\" >-11.000000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow14_col2\" class=\"data row14 col2\" >-3.670000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9dlevel0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow15_col0\" class=\"data row15 col0\" >-5.500000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow15_col1\" class=\"data row15 col1\" >-11.000000</td>\n",
       "                        <td id=\"T_c01b057c_9084_11eb_9aeb_f218985f9c9drow15_col2\" class=\"data row15 col2\" >-3.670000</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x13adc4430>"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(table1).style.background_gradient(cmap='Blues')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4 Win experiment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.1 Individual algorithms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "models = (\n",
    "    {'cost':'ar', 'params':{'order':1}, 'width':10},\n",
    "    {'cost':'ar', 'params':{'order':1}, 'width':15},\n",
    "    {'cost':'ar', 'params':{'order':5}, 'width':20},\n",
    "    {'cost':'mahalanobis', 'params':{}, 'width':10},\n",
    "    {'cost':'mahalanobis', 'params':{}, 'width':15},\n",
    "    {'cost':'mahalanobis', 'params':{}, 'width':20},\n",
    "    {'cost':'l1', 'params':{}, 'width':10},\n",
    "    {'cost':'l1', 'params':{}, 'width':15},\n",
    "    {'cost':'l1', 'params':{}, 'width':20},\n",
    "    {'cost':'l2', 'params':{}, 'width':10},\n",
    "    {'cost':'l2', 'params':{}, 'width':15},\n",
    "    {'cost':'l2', 'params':{}, 'width':20},\n",
    "#     {'cost':'linear', 'params':{}, 'width':10},\n",
    "#     {'cost':'linear', 'params':{}, 'width':40},\n",
    "#     {'cost':'linear', 'params':{}, 'width':100}\n",
    "#     {'model':'rbf', 'params':{}, 'width':40},\n",
    "#     {'model':'rbf', 'params':{}, 'width':100},\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "def wind(cost, data, params, width, true_cp=true_cp):\n",
    "    predicted_cp = []\n",
    "    for dataset in data:\n",
    "        stsc = StandardScaler()\n",
    "        signal = stsc.fit_transform(dataset)\n",
    "        algo = rpt.Window(model=cost, \n",
    "                          params=params, \n",
    "                          width=width,\n",
    "                          jump=1)\n",
    "        algo.fit(signal)\n",
    "        my_bkps = algo.predict(n_bkps=1)\n",
    "        \n",
    "        single_predicted_cp = pd.Series(data=0, index=true_cp[0].index)\n",
    "        single_predicted_cp[single_predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "        predicted_cp.append(single_predicted_cp)\n",
    "\n",
    "    nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')\n",
    "    return nab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1min 39s, sys: 10 s, total: 1min 49s\n",
      "Wall time: 49.3 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "table = []\n",
    "for model in tqdm_notebook(models, desc='agg functions loop'):\n",
    "    table.append(wind(**model, data=test))\n",
    "    clear_output()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow0_col0,#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow0_col1,#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow0_col2{\n",
       "            background-color:  #f7fbff;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow1_col0{\n",
       "            background-color:  #539ecd;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow1_col1{\n",
       "            background-color:  #4997c9;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow1_col2{\n",
       "            background-color:  #66abd4;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow2_col0{\n",
       "            background-color:  #aed1e7;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow2_col1{\n",
       "            background-color:  #afd1e7;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow2_col2{\n",
       "            background-color:  #bcd7eb;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow3_col0{\n",
       "            background-color:  #0b559f;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow3_col1{\n",
       "            background-color:  #125ea6;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow3_col2{\n",
       "            background-color:  #0e59a2;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow4_col0{\n",
       "            background-color:  #084082;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow4_col1{\n",
       "            background-color:  #084285;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow4_col2{\n",
       "            background-color:  #083877;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow5_col0,#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow5_col1,#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow7_col2{\n",
       "            background-color:  #08306b;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow5_col2{\n",
       "            background-color:  #084184;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow6_col0{\n",
       "            background-color:  #2575b7;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow6_col1{\n",
       "            background-color:  #2272b6;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow6_col2{\n",
       "            background-color:  #3181bd;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow7_col0,#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow7_col1{\n",
       "            background-color:  #08326e;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow8_col0{\n",
       "            background-color:  #3686c0;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow8_col1{\n",
       "            background-color:  #3585bf;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow8_col2{\n",
       "            background-color:  #3b8bc2;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow9_col0{\n",
       "            background-color:  #57a0ce;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow9_col1{\n",
       "            background-color:  #56a0ce;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow9_col2{\n",
       "            background-color:  #69add5;\n",
       "            color:  #000000;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow10_col0{\n",
       "            background-color:  #084b93;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow10_col1{\n",
       "            background-color:  #08478d;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow10_col2{\n",
       "            background-color:  #09529d;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow11_col0{\n",
       "            background-color:  #2474b7;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow11_col1{\n",
       "            background-color:  #206fb4;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow11_col2{\n",
       "            background-color:  #3080bd;\n",
       "            color:  #000000;\n",
       "        }</style><table id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9d\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Standart</th>        <th class=\"col_heading level0 col1\" >LowFP</th>        <th class=\"col_heading level0 col2\" >LowFN</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow0_col0\" class=\"data row0 col0\" >6.160000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow0_col1\" class=\"data row0 col1\" >5.110000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow0_col2\" class=\"data row0 col2\" >7.280000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow1_col0\" class=\"data row1 col0\" >18.580000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow1_col1\" class=\"data row1 col1\" >18.290000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow1_col2\" class=\"data row1 col2\" >18.730000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow2_col0\" class=\"data row2 col0\" >13.230000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow2_col1\" class=\"data row2 col1\" >12.190000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow2_col2\" class=\"data row2 col2\" >13.590000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow3_col0\" class=\"data row3 col0\" >24.680000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow3_col1\" class=\"data row3 col1\" >23.110000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow3_col2\" class=\"data row3 col2\" >25.980000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow4_col0\" class=\"data row4 col0\" >26.460000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow4_col1\" class=\"data row4 col1\" >25.540000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow4_col2\" class=\"data row4 col2\" >28.750000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow5_col0\" class=\"data row5 col0\" >27.790000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow5_col1\" class=\"data row5 col1\" >27.000000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow5_col2\" class=\"data row5 col2\" >28.050000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow6_col0\" class=\"data row6 col0\" >22.040000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow6_col1\" class=\"data row6 col1\" >21.390000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow6_col2\" class=\"data row6 col2\" >22.630000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow7_col0\" class=\"data row7 col0\" >27.580000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow7_col1\" class=\"data row7 col1\" >26.770000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow7_col2\" class=\"data row7 col2\" >29.500000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow8_col0\" class=\"data row8 col0\" >20.630000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow8_col1\" class=\"data row8 col1\" >19.850000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow8_col2\" class=\"data row8 col2\" >21.690000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow9_col0\" class=\"data row9 col0\" >18.260000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow9_col1\" class=\"data row9 col1\" >17.470000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow9_col2\" class=\"data row9 col2\" >18.520000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow10_col0\" class=\"data row10 col0\" >25.590000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow10_col1\" class=\"data row10 col1\" >25.050000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow10_col2\" class=\"data row10 col2\" >26.580000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9dlevel0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow11_col0\" class=\"data row11 col0\" >22.090000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow11_col1\" class=\"data row11 col1\" >21.680000</td>\n",
       "                        <td id=\"T_dd8453a2_9084_11eb_9aeb_f218985f9c9drow11_col2\" class=\"data row11 col2\" >22.660000</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x13adc4970>"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(table).style.background_gradient(cmap='Blues')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.2 Ensemble"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "def windowEnsemble(cost, data, num_agg_func, width, true_cp=true_cp):\n",
    "    predicted_cp = []\n",
    "    for dataset in data:\n",
    "        stsc = StandardScaler()\n",
    "        signal = stsc.fit_transform(dataset)\n",
    "        algo = rpt.WindowEnsembling(custom_cost=cost, \n",
    "                                    jump=1, \n",
    "                                    ensembling=num_agg_func, \n",
    "                                    width=width)\n",
    "        single_predicted_cp = pd.Series(data=0, index=true_cp[0].index)\n",
    "        try:\n",
    "            algo.fit(signal)\n",
    "            my_bkps = algo.predict(n_bkps=1)\n",
    "\n",
    "            single_predicted_cp[single_predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "            predicted_cp.append(single_predicted_cp)\n",
    "        except:\n",
    "            predicted_cp.append(single_predicted_cp)\n",
    "\n",
    "    nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')\n",
    "    return nab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 24min 3s, sys: 2min 6s, total: 26min 10s\n",
      "Wall time: 16min 13s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "cost = CostNew()\n",
    "table1 = []\n",
    "\n",
    "for n in tnrange(1, NUM_CPDE, desc='agg functions loop'):\n",
    "    for w in tqdm_notebook([10, 20, 30], desc='width loop', leave=False):\n",
    "        table1.append(windowEnsemble(cost=cost, data=test, num_agg_func=n, width=w))\n",
    "        clear_output()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "#T_21821524_9087_11eb_9aeb_f218985f9c9drow0_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow21_col2{\n",
       "            background-color:  #2e7ebc;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow0_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow8_col2{\n",
       "            background-color:  #3080bd;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow0_col2{\n",
       "            background-color:  #3181bd;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow1_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow2_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow5_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow13_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow16_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow25_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow28_col0{\n",
       "            background-color:  #084c95;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow1_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow7_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow13_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow16_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow25_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow28_col1{\n",
       "            background-color:  #084a91;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow1_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow13_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow16_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow25_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow28_col2{\n",
       "            background-color:  #08519c;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow2_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow5_col0{\n",
       "            background-color:  #084e98;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow2_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow3_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow4_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow5_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow9_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow22_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow34_col2{\n",
       "            background-color:  #0a539e;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow3_col1{\n",
       "            background-color:  #0a549e;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow3_col2{\n",
       "            background-color:  #0c56a0;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow4_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow4_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow9_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow9_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow34_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow34_col1{\n",
       "            background-color:  #084f99;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow6_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow6_col2{\n",
       "            background-color:  #3888c1;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow6_col1{\n",
       "            background-color:  #3a8ac2;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow7_col1{\n",
       "            background-color:  #084990;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow7_col2{\n",
       "            background-color:  #084184;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow8_col0{\n",
       "            background-color:  #2d7dbb;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow8_col1{\n",
       "            background-color:  #2c7cba;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow10_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow10_col1{\n",
       "            background-color:  #09529d;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow10_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow22_col0{\n",
       "            background-color:  #0b559f;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow11_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow11_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow11_col2{\n",
       "            background-color:  #08306b;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow12_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow33_col0{\n",
       "            background-color:  #65aad4;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow12_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow18_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow33_col1{\n",
       "            background-color:  #69add5;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow12_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow33_col2{\n",
       "            background-color:  #66abd4;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow14_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow23_col2{\n",
       "            background-color:  #1764ab;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow14_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow35_col2{\n",
       "            background-color:  #1663aa;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow14_col2{\n",
       "            background-color:  #1561a9;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow15_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow15_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow24_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow24_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow27_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow27_col2{\n",
       "            background-color:  #3686c0;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow15_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow24_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow27_col1{\n",
       "            background-color:  #3e8ec4;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow17_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow26_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow35_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow35_col1{\n",
       "            background-color:  #1966ad;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow17_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow26_col1{\n",
       "            background-color:  #1967ad;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow17_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow26_col2{\n",
       "            background-color:  #1562a9;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow18_col0{\n",
       "            background-color:  #68acd5;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow18_col1{\n",
       "            background-color:  #6fb0d7;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow19_col0{\n",
       "            background-color:  #084b93;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow19_col1{\n",
       "            background-color:  #084d96;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow19_col2{\n",
       "            background-color:  #084285;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow20_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow32_col0{\n",
       "            background-color:  #125ea6;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow20_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow20_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow32_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow32_col2{\n",
       "            background-color:  #125da6;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow21_col0{\n",
       "            background-color:  #2979b9;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow21_col1{\n",
       "            background-color:  #2575b7;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow22_col2{\n",
       "            background-color:  #0d57a1;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow23_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow23_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow30_col2{\n",
       "            background-color:  #1a68ae;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow29_col0{\n",
       "            background-color:  #2b7bba;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow29_col1{\n",
       "            background-color:  #2f7fbc;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow29_col2{\n",
       "            background-color:  #2070b4;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow30_col0{\n",
       "            background-color:  #1f6eb3;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow30_col1{\n",
       "            background-color:  #2777b8;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow31_col0{\n",
       "            background-color:  #083c7d;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow31_col1{\n",
       "            background-color:  #084082;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow31_col2{\n",
       "            background-color:  #083877;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow36_col0{\n",
       "            background-color:  #d0e1f2;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow36_col1{\n",
       "            background-color:  #cbdef1;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow36_col2{\n",
       "            background-color:  #d2e3f3;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow37_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow38_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow40_col0{\n",
       "            background-color:  #7db8da;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow37_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow38_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow40_col1{\n",
       "            background-color:  #7ab6d9;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow37_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow38_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow40_col2{\n",
       "            background-color:  #77b5d9;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow39_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow39_col2{\n",
       "            background-color:  #abd0e6;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow39_col1{\n",
       "            background-color:  #a5cde3;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow41_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow46_col1{\n",
       "            background-color:  #bfd8ed;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow41_col1{\n",
       "            background-color:  #bed8ec;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow41_col2,#T_21821524_9087_11eb_9aeb_f218985f9c9drow46_col2{\n",
       "            background-color:  #b9d6ea;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow42_col0{\n",
       "            background-color:  #a4cce3;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow42_col1{\n",
       "            background-color:  #a1cbe2;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow42_col2{\n",
       "            background-color:  #a6cee4;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow43_col0{\n",
       "            background-color:  #5fa6d1;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow43_col1{\n",
       "            background-color:  #5ca4d0;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow43_col2{\n",
       "            background-color:  #63a8d3;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow44_col0,#T_21821524_9087_11eb_9aeb_f218985f9c9drow44_col1,#T_21821524_9087_11eb_9aeb_f218985f9c9drow44_col2{\n",
       "            background-color:  #f7fbff;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow45_col0{\n",
       "            background-color:  #9dcae1;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow45_col1{\n",
       "            background-color:  #91c3de;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow45_col2{\n",
       "            background-color:  #a3cce3;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow46_col0{\n",
       "            background-color:  #c1d9ed;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow47_col0{\n",
       "            background-color:  #d0e2f2;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow47_col1{\n",
       "            background-color:  #ccdff1;\n",
       "            color:  #000000;\n",
       "        }#T_21821524_9087_11eb_9aeb_f218985f9c9drow47_col2{\n",
       "            background-color:  #d3e3f3;\n",
       "            color:  #000000;\n",
       "        }</style><table id=\"T_21821524_9087_11eb_9aeb_f218985f9c9d\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Standart</th>        <th class=\"col_heading level0 col1\" >LowFP</th>        <th class=\"col_heading level0 col2\" >LowFN</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow0_col0\" class=\"data row0 col0\" >18.260000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow0_col1\" class=\"data row0 col1\" >17.480000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow0_col2\" class=\"data row0 col2\" >18.520000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow1_col0\" class=\"data row1 col0\" >23.550000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow1_col1\" class=\"data row1 col1\" >23.280000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow1_col2\" class=\"data row1 col2\" >23.630000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow2_col0\" class=\"data row2 col0\" >23.300000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow2_col1\" class=\"data row2 col1\" >23.020000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow2_col2\" class=\"data row2 col2\" >23.470000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow3_col0\" class=\"data row3 col0\" >22.780000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow3_col1\" class=\"data row3 col1\" >22.200000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow3_col2\" class=\"data row3 col2\" >23.120000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow4_col0\" class=\"data row4 col0\" >23.280000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow4_col1\" class=\"data row4 col1\" >22.760000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow4_col2\" class=\"data row4 col2\" >23.460000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow5_col0\" class=\"data row5 col0\" >23.300000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow5_col1\" class=\"data row5 col1\" >23.020000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow5_col2\" class=\"data row5 col2\" >23.470000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow6_col0\" class=\"data row6 col0\" >17.200000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow6_col1\" class=\"data row6 col1\" >16.310000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow6_col2\" class=\"data row6 col2\" >17.820000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow7_col0\" class=\"data row7 col0\" >23.820000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow7_col1\" class=\"data row7 col1\" >23.350000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow7_col2\" class=\"data row7 col2\" >25.400000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow8_col0\" class=\"data row8 col0\" >18.440000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow8_col1\" class=\"data row8 col1\" >17.910000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow8_col2\" class=\"data row8 col2\" >18.640000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow9_col0\" class=\"data row9 col0\" >23.280000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow9_col1\" class=\"data row9 col1\" >22.750000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow9_col2\" class=\"data row9 col2\" >23.450000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow10_col0\" class=\"data row10 col0\" >22.930000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow10_col1\" class=\"data row10 col1\" >22.370000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow10_col2\" class=\"data row10 col2\" >23.230000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow11_col0\" class=\"data row11 col0\" >26.530000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow11_col1\" class=\"data row11 col1\" >26.090000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow11_col2\" class=\"data row11 col2\" >27.210000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow12_col0\" class=\"data row12 col0\" >13.240000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow12_col1\" class=\"data row12 col1\" >12.190000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow12_col2\" class=\"data row12 col2\" >13.590000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow13_col0\" class=\"data row13 col0\" >23.540000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow13_col1\" class=\"data row13 col1\" >23.280000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow13_col2\" class=\"data row13 col2\" >23.630000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow14_col0\" class=\"data row14 col0\" >20.990000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow14_col1\" class=\"data row14 col1\" >20.480000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow14_col2\" class=\"data row14 col2\" >21.930000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow15_col0\" class=\"data row15 col0\" >17.480000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow15_col1\" class=\"data row15 col1\" >15.900000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow15_col2\" class=\"data row15 col2\" >18.000000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow16_col0\" class=\"data row16 col0\" >23.550000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow16_col1\" class=\"data row16 col1\" >23.280000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow16_col2\" class=\"data row16 col2\" >23.630000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow17_col0\" class=\"data row17 col0\" >20.810000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow17_col1\" class=\"data row17 col1\" >20.040000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow17_col2\" class=\"data row17 col2\" >21.810000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row18\" class=\"row_heading level0 row18\" >18</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow18_col0\" class=\"data row18 col0\" >12.980000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow18_col1\" class=\"data row18 col1\" >11.670000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow18_col2\" class=\"data row18 col2\" >13.410000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row19\" class=\"row_heading level0 row19\" >19</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow19_col0\" class=\"data row19 col0\" >23.680000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow19_col1\" class=\"data row19 col1\" >22.960000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow19_col2\" class=\"data row19 col2\" >25.310000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row20\" class=\"row_heading level0 row20\" >20</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow20_col0\" class=\"data row20 col0\" >21.630000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow20_col1\" class=\"data row20 col1\" >21.180000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow20_col2\" class=\"data row20 col2\" >22.360000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row21\" class=\"row_heading level0 row21\" >21</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow21_col0\" class=\"data row21 col0\" >18.780000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow21_col1\" class=\"data row21 col1\" >18.520000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow21_col2\" class=\"data row21 col2\" >18.870000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row22\" class=\"row_heading level0 row22\" >22</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow22_col0\" class=\"data row22 col0\" >22.620000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow22_col1\" class=\"data row22 col1\" >22.270000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow22_col2\" class=\"data row22 col2\" >23.020000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row23\" class=\"row_heading level0 row23\" >23</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow23_col0\" class=\"data row23 col0\" >20.520000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow23_col1\" class=\"data row23 col1\" >19.950000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow23_col2\" class=\"data row23 col2\" >21.610000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row24\" class=\"row_heading level0 row24\" >24</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow24_col0\" class=\"data row24 col0\" >17.480000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow24_col1\" class=\"data row24 col1\" >15.900000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow24_col2\" class=\"data row24 col2\" >18.000000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row25\" class=\"row_heading level0 row25\" >25</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow25_col0\" class=\"data row25 col0\" >23.550000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow25_col1\" class=\"data row25 col1\" >23.280000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow25_col2\" class=\"data row25 col2\" >23.630000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row26\" class=\"row_heading level0 row26\" >26</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow26_col0\" class=\"data row26 col0\" >20.810000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow26_col1\" class=\"data row26 col1\" >20.040000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow26_col2\" class=\"data row26 col2\" >21.810000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row27\" class=\"row_heading level0 row27\" >27</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow27_col0\" class=\"data row27 col0\" >17.480000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow27_col1\" class=\"data row27 col1\" >15.900000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow27_col2\" class=\"data row27 col2\" >18.000000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row28\" class=\"row_heading level0 row28\" >28</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow28_col0\" class=\"data row28 col0\" >23.550000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow28_col1\" class=\"data row28 col1\" >23.280000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow28_col2\" class=\"data row28 col2\" >23.630000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row29\" class=\"row_heading level0 row29\" >29</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow29_col0\" class=\"data row29 col0\" >18.580000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow29_col1\" class=\"data row29 col1\" >17.590000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow29_col2\" class=\"data row29 col2\" >20.320000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row30\" class=\"row_heading level0 row30\" >30</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow30_col0\" class=\"data row30 col0\" >19.900000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow30_col1\" class=\"data row30 col1\" >18.330000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow30_col2\" class=\"data row30 col2\" >21.200000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row31\" class=\"row_heading level0 row31\" >31</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow31_col0\" class=\"data row31 col0\" >25.140000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow31_col1\" class=\"data row31 col1\" >24.330000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow31_col2\" class=\"data row31 col2\" >26.290000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row32\" class=\"row_heading level0 row32\" >32</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow32_col0\" class=\"data row32 col0\" >21.630000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow32_col1\" class=\"data row32 col1\" >21.180000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow32_col2\" class=\"data row32 col2\" >22.360000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row33\" class=\"row_heading level0 row33\" >33</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow33_col0\" class=\"data row33 col0\" >13.240000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow33_col1\" class=\"data row33 col1\" >12.190000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow33_col2\" class=\"data row33 col2\" >13.590000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row34\" class=\"row_heading level0 row34\" >34</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow34_col0\" class=\"data row34 col0\" >23.280000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow34_col1\" class=\"data row34 col1\" >22.760000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow34_col2\" class=\"data row34 col2\" >23.460000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row35\" class=\"row_heading level0 row35\" >35</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow35_col0\" class=\"data row35 col0\" >20.720000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow35_col1\" class=\"data row35 col1\" >20.170000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow35_col2\" class=\"data row35 col2\" >21.750000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row36\" class=\"row_heading level0 row36\" >36</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow36_col0\" class=\"data row36 col0\" >4.500000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow36_col1\" class=\"data row36 col1\" >4.240000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow36_col2\" class=\"data row36 col2\" >4.590000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row37\" class=\"row_heading level0 row37\" >37</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow37_col0\" class=\"data row37 col0\" >11.460000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow37_col1\" class=\"data row37 col1\" >10.950000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow37_col2\" class=\"data row37 col2\" >12.400000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row38\" class=\"row_heading level0 row38\" >38</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow38_col0\" class=\"data row38 col0\" >11.450000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow38_col1\" class=\"data row38 col1\" >10.940000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow38_col2\" class=\"data row38 col2\" >12.400000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row39\" class=\"row_heading level0 row39\" >39</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow39_col0\" class=\"data row39 col0\" >8.210000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow39_col1\" class=\"data row39 col1\" >7.840000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow39_col2\" class=\"data row39 col2\" >8.650000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row40\" class=\"row_heading level0 row40\" >40</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow40_col0\" class=\"data row40 col0\" >11.460000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow40_col1\" class=\"data row40 col1\" >10.950000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow40_col2\" class=\"data row40 col2\" >12.400000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row41\" class=\"row_heading level0 row41\" >41</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow41_col0\" class=\"data row41 col0\" >6.430000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow41_col1\" class=\"data row41 col1\" >5.650000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow41_col2\" class=\"data row41 col2\" >7.460000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row42\" class=\"row_heading level0 row42\" >42</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow42_col0\" class=\"data row42 col0\" >8.790000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow42_col1\" class=\"data row42 col1\" >8.250000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow42_col2\" class=\"data row42 col2\" >9.030000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row43\" class=\"row_heading level0 row43\" >43</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow43_col0\" class=\"data row43 col0\" >13.750000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow43_col1\" class=\"data row43 col1\" >13.220000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow43_col2\" class=\"data row43 col2\" >13.930000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row44\" class=\"row_heading level0 row44\" >44</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow44_col0\" class=\"data row44 col0\" >-1.050000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow44_col1\" class=\"data row44 col1\" >-2.100000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow44_col2\" class=\"data row44 col2\" >-0.700000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row45\" class=\"row_heading level0 row45\" >45</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow45_col0\" class=\"data row45 col0\" >9.390000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow45_col1\" class=\"data row45 col1\" >9.370000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow45_col2\" class=\"data row45 col2\" >9.430000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row46\" class=\"row_heading level0 row46\" >46</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow46_col0\" class=\"data row46 col0\" >6.380000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow46_col1\" class=\"data row46 col1\" >5.590000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow46_col2\" class=\"data row46 col2\" >7.430000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_21821524_9087_11eb_9aeb_f218985f9c9dlevel0_row47\" class=\"row_heading level0 row47\" >47</th>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow47_col0\" class=\"data row47 col0\" >4.410000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow47_col1\" class=\"data row47 col1\" >4.140000</td>\n",
       "                        <td id=\"T_21821524_9087_11eb_9aeb_f218985f9c9drow47_col2\" class=\"data row47 col2\" >4.530000</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x13a9a9130>"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(table1).style.background_gradient(cmap='Blues')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5 BinSeg experiment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.1 Individual algorithms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "models = (\n",
    "    {'cost':'ar', 'params':{'order':1}},\n",
    "    {'cost':'mahalanobis', 'params':{}},\n",
    "    {'cost':'l1', 'params':{}},\n",
    "    {'cost':'l2', 'params':{}},\n",
    "    {'cost':'linear', 'params':{}},\n",
    "#     {'model':'rbf', 'params':{}},\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "def binseg(cost, data, params, true_cp=true_cp):\n",
    "    predicted_cp = []\n",
    "    for dataset in data:\n",
    "        stsc = StandardScaler()\n",
    "        signal = stsc.fit_transform(dataset)\n",
    "        algo = rpt.Binseg(model=cost, \n",
    "                          params=params,\n",
    "                          jump=1)\n",
    "        algo.fit(signal)\n",
    "        my_bkps = algo.predict(n_bkps=1)\n",
    "        \n",
    "        single_predicted_cp = pd.Series(data=0, index=true_cp[0].index)\n",
    "        single_predicted_cp[single_predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "        predicted_cp.append(single_predicted_cp)\n",
    "\n",
    "    nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')\n",
    "    return nab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "table = []\n",
    "for model in tqdm_notebook(models, desc='agg functions loop'):\n",
    "    table.append(binseg(**model, data=test))\n",
    "    clear_output()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow0_col0{\n",
       "            background-color:  #1966ad;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow0_col1{\n",
       "            background-color:  #1b69af;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow0_col2,#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow2_col2{\n",
       "            background-color:  #0a549e;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow1_col0,#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow1_col1,#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow1_col2{\n",
       "            background-color:  #08306b;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow2_col0{\n",
       "            background-color:  #0a539e;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow2_col1{\n",
       "            background-color:  #08509b;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow3_col0{\n",
       "            background-color:  #1865ac;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow3_col1{\n",
       "            background-color:  #1764ab;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow3_col2{\n",
       "            background-color:  #135fa7;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow4_col0,#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow4_col1,#T_c6a38588_9087_11eb_9aeb_f218985f9c9drow4_col2{\n",
       "            background-color:  #f7fbff;\n",
       "            color:  #000000;\n",
       "        }</style><table id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9d\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Standart</th>        <th class=\"col_heading level0 col1\" >LowFP</th>        <th class=\"col_heading level0 col2\" >LowFN</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9dlevel0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow0_col0\" class=\"data row0 col0\" >30.150000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow0_col1\" class=\"data row0 col1\" >28.890000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow0_col2\" class=\"data row0 col2\" >32.800000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9dlevel0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow1_col0\" class=\"data row1 col0\" >36.880000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow1_col1\" class=\"data row1 col1\" >35.820000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow1_col2\" class=\"data row1 col2\" >37.290000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9dlevel0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow2_col0\" class=\"data row2 col0\" >32.530000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow2_col1\" class=\"data row2 col1\" >31.980000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow2_col2\" class=\"data row2 col2\" >32.800000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9dlevel0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow3_col0\" class=\"data row3 col0\" >30.300000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow3_col1\" class=\"data row3 col1\" >29.520000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow3_col2\" class=\"data row3 col2\" >31.310000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9dlevel0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow4_col0\" class=\"data row4 col0\" >4.500000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow4_col1\" class=\"data row4 col1\" >4.240000</td>\n",
       "                        <td id=\"T_c6a38588_9087_11eb_9aeb_f218985f9c9drow4_col2\" class=\"data row4 col2\" >4.590000</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x139af7100>"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(table).style.background_gradient(cmap='Blues')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.2 Ensemble"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "def binsegEnsemble(cost, data, num_agg_func, true_cp=true_cp):\n",
    "    predicted_cp = []\n",
    "    for dataset in data:\n",
    "        stsc = StandardScaler()\n",
    "        signal = stsc.fit_transform(dataset)\n",
    "        algo = rpt.BinsegEnsembling(custom_cost=cost, \n",
    "                                    jump=1, \n",
    "                                    ensembling=num_agg_func)\n",
    "        single_predicted_cp = pd.Series(data=0, index=true_cp[0].index)\n",
    "        try:\n",
    "            algo.fit(signal)\n",
    "            my_bkps = algo.predict(n_bkps=1)\n",
    "\n",
    "            single_predicted_cp[single_predicted_cp.index[my_bkps[:-1]]] = 1\n",
    "            predicted_cp.append(single_predicted_cp)\n",
    "        except:\n",
    "            predicted_cp.append(single_predicted_cp)\n",
    "\n",
    "    nab = evaluating_change_point(true_cp, predicted_cp, metric='nab', numenta_time='288 min')\n",
    "    return nab"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 3h 49min 44s, sys: 1h 16min 26s, total: 5h 6min 10s\n",
      "Wall time: 55min 27s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "cost = CostNew()\n",
    "table1 = []\n",
    "\n",
    "for n in tnrange(1, NUM_CPDE, desc='agg functions loop'):\n",
    "    table1.append(binsegEnsemble(cost, data=test, num_agg_func=n))\n",
    "    clear_output()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "#T_861601d2_908f_11eb_9aeb_f218985f9c9drow0_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow0_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow0_col2,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow3_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow3_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow3_col2{\n",
       "            background-color:  #08306b;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow1_col0{\n",
       "            background-color:  #4594c7;\n",
       "            color:  #000000;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow1_col1{\n",
       "            background-color:  #4896c8;\n",
       "            color:  #000000;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow1_col2,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow11_col2{\n",
       "            background-color:  #4090c5;\n",
       "            color:  #000000;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow2_col0{\n",
       "            background-color:  #5ca4d0;\n",
       "            color:  #000000;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow2_col1{\n",
       "            background-color:  #60a7d2;\n",
       "            color:  #000000;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow2_col2{\n",
       "            background-color:  #4e9acb;\n",
       "            color:  #000000;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow4_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow5_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow6_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow8_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow9_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow10_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow12_col2,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow13_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow14_col2{\n",
       "            background-color:  #105ba4;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow4_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow5_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow6_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow8_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow9_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow10_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow13_col1{\n",
       "            background-color:  #115ca5;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow4_col2,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow6_col2,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow8_col2,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow10_col2{\n",
       "            background-color:  #0d57a1;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow5_col2,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow9_col2,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow13_col2{\n",
       "            background-color:  #0c56a0;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow7_col0{\n",
       "            background-color:  #2676b8;\n",
       "            color:  #000000;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow7_col1{\n",
       "            background-color:  #2575b7;\n",
       "            color:  #000000;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow7_col2{\n",
       "            background-color:  #2272b6;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow11_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow11_col1{\n",
       "            background-color:  #4695c8;\n",
       "            color:  #000000;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow12_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow14_col0{\n",
       "            background-color:  #1663aa;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow12_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow14_col1{\n",
       "            background-color:  #1764ab;\n",
       "            color:  #f1f1f1;\n",
       "        }#T_861601d2_908f_11eb_9aeb_f218985f9c9drow15_col0,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow15_col1,#T_861601d2_908f_11eb_9aeb_f218985f9c9drow15_col2{\n",
       "            background-color:  #f7fbff;\n",
       "            color:  #000000;\n",
       "        }</style><table id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9d\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >Standart</th>        <th class=\"col_heading level0 col1\" >LowFP</th>        <th class=\"col_heading level0 col2\" >LowFN</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow0_col0\" class=\"data row0 col0\" >41.810000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow0_col1\" class=\"data row0 col1\" >41.000000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow0_col2\" class=\"data row0 col2\" >42.160000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow1_col0\" class=\"data row1 col0\" >25.660000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow1_col1\" class=\"data row1 col1\" >24.900000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow1_col2\" class=\"data row1 col2\" >26.630000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow2_col0\" class=\"data row2 col0\" >22.850000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow2_col1\" class=\"data row2 col1\" >21.820000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow2_col2\" class=\"data row2 col2\" >24.760000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow3_col0\" class=\"data row3 col0\" >41.810000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow3_col1\" class=\"data row3 col1\" >41.000000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow3_col2\" class=\"data row3 col2\" >42.160000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow4_col0\" class=\"data row4 col0\" >34.800000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow4_col1\" class=\"data row4 col1\" >34.000000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow4_col2\" class=\"data row4 col2\" >35.900000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow5_col0\" class=\"data row5 col0\" >34.830000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow5_col1\" class=\"data row5 col1\" >34.030000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow5_col2\" class=\"data row5 col2\" >35.920000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow6_col0\" class=\"data row6 col0\" >34.800000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow6_col1\" class=\"data row6 col1\" >34.000000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow6_col2\" class=\"data row6 col2\" >35.900000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow7_col0\" class=\"data row7 col0\" >30.470000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow7_col1\" class=\"data row7 col1\" >29.950000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow7_col2\" class=\"data row7 col2\" >31.420000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow8_col0\" class=\"data row8 col0\" >34.800000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow8_col1\" class=\"data row8 col1\" >34.000000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow8_col2\" class=\"data row8 col2\" >35.900000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow9_col0\" class=\"data row9 col0\" >34.830000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow9_col1\" class=\"data row9 col1\" >34.030000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow9_col2\" class=\"data row9 col2\" >35.920000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow10_col0\" class=\"data row10 col0\" >34.800000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow10_col1\" class=\"data row10 col1\" >34.000000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow10_col2\" class=\"data row10 col2\" >35.900000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow11_col0\" class=\"data row11 col0\" >25.590000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow11_col1\" class=\"data row11 col1\" >25.060000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow11_col2\" class=\"data row11 col2\" >26.590000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow12_col0\" class=\"data row12 col0\" >33.640000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow12_col1\" class=\"data row12 col1\" >32.730000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow12_col2\" class=\"data row12 col2\" >35.130000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow13_col0\" class=\"data row13 col0\" >34.830000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow13_col1\" class=\"data row13 col1\" >34.030000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow13_col2\" class=\"data row13 col2\" >35.920000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow14_col0\" class=\"data row14 col0\" >33.640000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow14_col1\" class=\"data row14 col1\" >32.730000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow14_col2\" class=\"data row14 col2\" >35.130000</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9dlevel0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow15_col0\" class=\"data row15 col0\" >0.000000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow15_col1\" class=\"data row15 col1\" >0.000000</td>\n",
       "                        <td id=\"T_861601d2_908f_11eb_9aeb_f218985f9c9drow15_col2\" class=\"data row15 col2\" >0.000000</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x13afda1c0>"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(table1).style.background_gradient(cmap='Blues')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}