A first look at the new timeseries forecasting module of pyCaret

pycaret_timeseries.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6c1cdf13-d268-43fd-85a8-8e7a8f0ba86c",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    },
    "tags": []
   },
   "source": [
    "# Pycaret time_series module\n",
    "\n",
    "- available since pycaret 3.0.0 released March 18th 2023\n",
    "- based on sktime\n",
    "\n",
    "**Warning 1**\n",
    "\n",
    "Mac is offically not supported. Before installing pycaret do\n",
    "\n",
    "    brew install cmake\n",
    "    \n",
    "**Warning 2**  \n",
    "\n",
    "pycaret 3.0.0 forgot to pin the version of sktime.\n",
    "To make it work one has to install `sktime<0.18`\n",
    "    \n",
    "    pip install pycaret \"sktime<0.18\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e3e1cae8-d4ab-4445-9053-3b3bb2c41bc0",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import pycaret.time_series as pyts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "11a7cbc0-5bfd-4cbd-9370-419b161cda3a",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"20230503_090829.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d27ff8f8-e16a-4d8f-98ca-749e954b61da",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 394 entries, 0 to 393\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype  \n",
      "---  ------      --------------  -----  \n",
      " 0   Time        394 non-null    object \n",
      " 1   MAX(count)  394 non-null    float64\n",
      "dtypes: float64(1), object(1)\n",
      "memory usage: 6.3+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "beeb01c1-ede7-4700-a73c-a97684fe661b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df.index = pd.to_datetime(df.Time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "3668f9fd-0d04-48e2-b9b8-12e3fe58f80d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df2 = df.drop(\"Time\", axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "68c880fc-7820-498e-abb9-cbf72fbd97a2",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "df3 = df2.rename(columns={\"MAX(count)\": \"timeoff\"})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "42e0d597-9ba9-4eed-8202-e80fde1f0c8d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='Time'>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x550 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df3.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e57a0a69-7273-4d40-8de8-649556b22a17",
   "metadata": {},
   "source": [
    "## Object oriented interface\n",
    "\n",
    "- new in pycaret 3.0  \n",
    "- recommended by me\n",
    "\n",
    "Instead of global imports\n",
    "\n",
    "    from pycaret.time_series import *\n",
    "    s = setup(df3, fh=30, session_id=42)\n",
    "    \n",
    "Better use\n",
    "\n",
    "    import pycaret.timeseries as pyts\n",
    "    exp = pyts.TSForecastingExperiment()\n",
    "    exp.setup(df3, fh=30, session_id=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1f7100cb-ac04-4f98-a65d-5036d01b352b",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "exp = pyts.TSForecastingExperiment()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d8b90f30-2e00-4cb8-80d5-0c50cfe2a9c4",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_36f20_row25_col1 {\n",
       "  background-color: lightgreen;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_36f20\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_36f20_level0_col0\" class=\"col_heading level0 col0\" >Description</th>\n",
       "      <th id=\"T_36f20_level0_col1\" class=\"col_heading level0 col1\" >Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_36f20_row0_col0\" class=\"data row0 col0\" >session_id</td>\n",
       "      <td id=\"T_36f20_row0_col1\" class=\"data row0 col1\" >42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_36f20_row1_col0\" class=\"data row1 col0\" >Target</td>\n",
       "      <td id=\"T_36f20_row1_col1\" class=\"data row1 col1\" >timeoff</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_36f20_row2_col0\" class=\"data row2 col0\" >Approach</td>\n",
       "      <td id=\"T_36f20_row2_col1\" class=\"data row2 col1\" >Univariate</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_36f20_row3_col0\" class=\"data row3 col0\" >Exogenous Variables</td>\n",
       "      <td id=\"T_36f20_row3_col1\" class=\"data row3 col1\" >Not Present</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_36f20_row4_col0\" class=\"data row4 col0\" >Original data shape</td>\n",
       "      <td id=\"T_36f20_row4_col1\" class=\"data row4 col1\" >(394, 1)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_36f20_row5_col0\" class=\"data row5 col0\" >Transformed data shape</td>\n",
       "      <td id=\"T_36f20_row5_col1\" class=\"data row5 col1\" >(394, 1)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_36f20_row6_col0\" class=\"data row6 col0\" >Transformed train set shape</td>\n",
       "      <td id=\"T_36f20_row6_col1\" class=\"data row6 col1\" >(364, 1)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_36f20_row7_col0\" class=\"data row7 col0\" >Transformed test set shape</td>\n",
       "      <td id=\"T_36f20_row7_col1\" class=\"data row7 col1\" >(30, 1)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_36f20_row8_col0\" class=\"data row8 col0\" >Rows with missing values</td>\n",
       "      <td id=\"T_36f20_row8_col1\" class=\"data row8 col1\" >0.0%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_36f20_row9_col0\" class=\"data row9 col0\" >Fold Generator</td>\n",
       "      <td id=\"T_36f20_row9_col1\" class=\"data row9 col1\" >ExpandingWindowSplitter</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "      <td id=\"T_36f20_row10_col0\" class=\"data row10 col0\" >Fold Number</td>\n",
       "      <td id=\"T_36f20_row10_col1\" class=\"data row10 col1\" >3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "      <td id=\"T_36f20_row11_col0\" class=\"data row11 col0\" >Enforce Prediction Interval</td>\n",
       "      <td id=\"T_36f20_row11_col1\" class=\"data row11 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "      <td id=\"T_36f20_row12_col0\" class=\"data row12 col0\" >Splits used for hyperparameters</td>\n",
       "      <td id=\"T_36f20_row12_col1\" class=\"data row12 col1\" >all</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "      <td id=\"T_36f20_row13_col0\" class=\"data row13 col0\" >User Defined Seasonal Period(s)</td>\n",
       "      <td id=\"T_36f20_row13_col1\" class=\"data row13 col1\" >[7, 30, 31, 365, 366]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "      <td id=\"T_36f20_row14_col0\" class=\"data row14 col0\" >Ignore Seasonality Test</td>\n",
       "      <td id=\"T_36f20_row14_col1\" class=\"data row14 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "      <td id=\"T_36f20_row15_col0\" class=\"data row15 col0\" >Seasonality Detection Algo</td>\n",
       "      <td id=\"T_36f20_row15_col1\" class=\"data row15 col1\" >user_defined</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "      <td id=\"T_36f20_row16_col0\" class=\"data row16 col0\" >Max Period to Consider</td>\n",
       "      <td id=\"T_36f20_row16_col1\" class=\"data row16 col1\" >400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "      <td id=\"T_36f20_row17_col0\" class=\"data row17 col0\" >Seasonal Period(s) Tested</td>\n",
       "      <td id=\"T_36f20_row17_col1\" class=\"data row17 col1\" >[7, 30, 31, 365, 366]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row18\" class=\"row_heading level0 row18\" >18</th>\n",
       "      <td id=\"T_36f20_row18_col0\" class=\"data row18 col0\" >Significant Seasonal Period(s)</td>\n",
       "      <td id=\"T_36f20_row18_col1\" class=\"data row18 col1\" >[7]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row19\" class=\"row_heading level0 row19\" >19</th>\n",
       "      <td id=\"T_36f20_row19_col0\" class=\"data row19 col0\" >Significant Seasonal Period(s) without Harmonics</td>\n",
       "      <td id=\"T_36f20_row19_col1\" class=\"data row19 col1\" >[7]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row20\" class=\"row_heading level0 row20\" >20</th>\n",
       "      <td id=\"T_36f20_row20_col0\" class=\"data row20 col0\" >Remove Harmonics</td>\n",
       "      <td id=\"T_36f20_row20_col1\" class=\"data row20 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row21\" class=\"row_heading level0 row21\" >21</th>\n",
       "      <td id=\"T_36f20_row21_col0\" class=\"data row21 col0\" >Harmonics Order Method</td>\n",
       "      <td id=\"T_36f20_row21_col1\" class=\"data row21 col1\" >harmonic_max</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row22\" class=\"row_heading level0 row22\" >22</th>\n",
       "      <td id=\"T_36f20_row22_col0\" class=\"data row22 col0\" >Num Seasonalities to Use</td>\n",
       "      <td id=\"T_36f20_row22_col1\" class=\"data row22 col1\" >1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row23\" class=\"row_heading level0 row23\" >23</th>\n",
       "      <td id=\"T_36f20_row23_col0\" class=\"data row23 col0\" >All Seasonalities to Use</td>\n",
       "      <td id=\"T_36f20_row23_col1\" class=\"data row23 col1\" >[7]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row24\" class=\"row_heading level0 row24\" >24</th>\n",
       "      <td id=\"T_36f20_row24_col0\" class=\"data row24 col0\" >Primary Seasonality</td>\n",
       "      <td id=\"T_36f20_row24_col1\" class=\"data row24 col1\" >7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row25\" class=\"row_heading level0 row25\" >25</th>\n",
       "      <td id=\"T_36f20_row25_col0\" class=\"data row25 col0\" >Seasonality Present</td>\n",
       "      <td id=\"T_36f20_row25_col1\" class=\"data row25 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row26\" class=\"row_heading level0 row26\" >26</th>\n",
       "      <td id=\"T_36f20_row26_col0\" class=\"data row26 col0\" >Target Strictly Positive</td>\n",
       "      <td id=\"T_36f20_row26_col1\" class=\"data row26 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row27\" class=\"row_heading level0 row27\" >27</th>\n",
       "      <td id=\"T_36f20_row27_col0\" class=\"data row27 col0\" >Target White Noise</td>\n",
       "      <td id=\"T_36f20_row27_col1\" class=\"data row27 col1\" >No</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row28\" class=\"row_heading level0 row28\" >28</th>\n",
       "      <td id=\"T_36f20_row28_col0\" class=\"data row28 col0\" >Recommended d</td>\n",
       "      <td id=\"T_36f20_row28_col1\" class=\"data row28 col1\" >0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row29\" class=\"row_heading level0 row29\" >29</th>\n",
       "      <td id=\"T_36f20_row29_col0\" class=\"data row29 col0\" >Recommended Seasonal D</td>\n",
       "      <td id=\"T_36f20_row29_col1\" class=\"data row29 col1\" >0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row30\" class=\"row_heading level0 row30\" >30</th>\n",
       "      <td id=\"T_36f20_row30_col0\" class=\"data row30 col0\" >Preprocess</td>\n",
       "      <td id=\"T_36f20_row30_col1\" class=\"data row30 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row31\" class=\"row_heading level0 row31\" >31</th>\n",
       "      <td id=\"T_36f20_row31_col0\" class=\"data row31 col0\" >CPU Jobs</td>\n",
       "      <td id=\"T_36f20_row31_col1\" class=\"data row31 col1\" >-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row32\" class=\"row_heading level0 row32\" >32</th>\n",
       "      <td id=\"T_36f20_row32_col0\" class=\"data row32 col0\" >Use GPU</td>\n",
       "      <td id=\"T_36f20_row32_col1\" class=\"data row32 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row33\" class=\"row_heading level0 row33\" >33</th>\n",
       "      <td id=\"T_36f20_row33_col0\" class=\"data row33 col0\" >Log Experiment</td>\n",
       "      <td id=\"T_36f20_row33_col1\" class=\"data row33 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row34\" class=\"row_heading level0 row34\" >34</th>\n",
       "      <td id=\"T_36f20_row34_col0\" class=\"data row34 col0\" >Experiment Name</td>\n",
       "      <td id=\"T_36f20_row34_col1\" class=\"data row34 col1\" >ts-default-name</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_36f20_level0_row35\" class=\"row_heading level0 row35\" >35</th>\n",
       "      <td id=\"T_36f20_row35_col0\" class=\"data row35 col0\" >USI</td>\n",
       "      <td id=\"T_36f20_row35_col1\" class=\"data row35 col1\" >8e74</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x284b6a580>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<pycaret.time_series.forecasting.oop.TSForecastingExperiment at 0x284b2d8e0>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exp.setup(df3, fh=30, session_id=42, seasonal_period=[7,30,31,365,366], max_sp_to_consider=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "535e5086-e10e-46be-9f32-d5f8e98ee131",
   "metadata": {
    "tags": []
   },
   "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>Test</th>\n",
       "      <th>Test Name</th>\n",
       "      <th>Data</th>\n",
       "      <th>Property</th>\n",
       "      <th>Setting</th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Summary</td>\n",
       "      <td>Statistics</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Length</td>\n",
       "      <td></td>\n",
       "      <td>394.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Summary</td>\n",
       "      <td>Statistics</td>\n",
       "      <td>Transformed</td>\n",
       "      <td># Missing Values</td>\n",
       "      <td></td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Summary</td>\n",
       "      <td>Statistics</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Mean</td>\n",
       "      <td></td>\n",
       "      <td>15.598985</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Summary</td>\n",
       "      <td>Statistics</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Median</td>\n",
       "      <td></td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Summary</td>\n",
       "      <td>Statistics</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Standard Deviation</td>\n",
       "      <td></td>\n",
       "      <td>9.059408</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Summary</td>\n",
       "      <td>Statistics</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Variance</td>\n",
       "      <td></td>\n",
       "      <td>82.072874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Summary</td>\n",
       "      <td>Statistics</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Kurtosis</td>\n",
       "      <td></td>\n",
       "      <td>12.447098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Summary</td>\n",
       "      <td>Statistics</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Skewness</td>\n",
       "      <td></td>\n",
       "      <td>2.690653</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Summary</td>\n",
       "      <td>Statistics</td>\n",
       "      <td>Transformed</td>\n",
       "      <td># Distinct Values</td>\n",
       "      <td></td>\n",
       "      <td>39.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>White Noise</td>\n",
       "      <td>Ljung-Box</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Test Statictic</td>\n",
       "      <td>{'alpha': 0.05, 'K': 24}</td>\n",
       "      <td>898.583563</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>White Noise</td>\n",
       "      <td>Ljung-Box</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Test Statictic</td>\n",
       "      <td>{'alpha': 0.05, 'K': 48}</td>\n",
       "      <td>1006.762155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>White Noise</td>\n",
       "      <td>Ljung-Box</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>p-value</td>\n",
       "      <td>{'alpha': 0.05, 'K': 24}</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>White Noise</td>\n",
       "      <td>Ljung-Box</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>p-value</td>\n",
       "      <td>{'alpha': 0.05, 'K': 48}</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>White Noise</td>\n",
       "      <td>Ljung-Box</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>White Noise</td>\n",
       "      <td>{'alpha': 0.05, 'K': 24}</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>White Noise</td>\n",
       "      <td>Ljung-Box</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>White Noise</td>\n",
       "      <td>{'alpha': 0.05, 'K': 48}</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>ADF</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Stationarity</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>ADF</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>p-value</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>0.000565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>ADF</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Test Statistic</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>-4.239263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>ADF</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Critical Value 1%</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>-3.447815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>ADF</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Critical Value 5%</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>-2.869237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>ADF</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Critical Value 10%</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>-2.57087</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>KPSS</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Trend Stationarity</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>KPSS</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>p-value</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>0.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>KPSS</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Test Statistic</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>0.083097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>KPSS</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Critical Value 10%</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>0.119</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>KPSS</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Critical Value 5%</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>0.146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>KPSS</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Critical Value 2.5%</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>0.176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Stationarity</td>\n",
       "      <td>KPSS</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Critical Value 1%</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>0.216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>Normality</td>\n",
       "      <td>Shapiro</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>Normality</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Normality</td>\n",
       "      <td>Shapiro</td>\n",
       "      <td>Transformed</td>\n",
       "      <td>p-value</td>\n",
       "      <td>{'alpha': 0.05}</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Test   Test Name         Data             Property  \\\n",
       "0        Summary  Statistics  Transformed               Length   \n",
       "1        Summary  Statistics  Transformed     # Missing Values   \n",
       "2        Summary  Statistics  Transformed                 Mean   \n",
       "3        Summary  Statistics  Transformed               Median   \n",
       "4        Summary  Statistics  Transformed   Standard Deviation   \n",
       "5        Summary  Statistics  Transformed             Variance   \n",
       "6        Summary  Statistics  Transformed             Kurtosis   \n",
       "7        Summary  Statistics  Transformed             Skewness   \n",
       "8        Summary  Statistics  Transformed    # Distinct Values   \n",
       "9    White Noise   Ljung-Box  Transformed       Test Statictic   \n",
       "10   White Noise   Ljung-Box  Transformed       Test Statictic   \n",
       "11   White Noise   Ljung-Box  Transformed              p-value   \n",
       "12   White Noise   Ljung-Box  Transformed              p-value   \n",
       "13   White Noise   Ljung-Box  Transformed          White Noise   \n",
       "14   White Noise   Ljung-Box  Transformed          White Noise   \n",
       "15  Stationarity         ADF  Transformed         Stationarity   \n",
       "16  Stationarity         ADF  Transformed              p-value   \n",
       "17  Stationarity         ADF  Transformed       Test Statistic   \n",
       "18  Stationarity         ADF  Transformed    Critical Value 1%   \n",
       "19  Stationarity         ADF  Transformed    Critical Value 5%   \n",
       "20  Stationarity         ADF  Transformed   Critical Value 10%   \n",
       "21  Stationarity        KPSS  Transformed   Trend Stationarity   \n",
       "22  Stationarity        KPSS  Transformed              p-value   \n",
       "23  Stationarity        KPSS  Transformed       Test Statistic   \n",
       "24  Stationarity        KPSS  Transformed   Critical Value 10%   \n",
       "25  Stationarity        KPSS  Transformed    Critical Value 5%   \n",
       "26  Stationarity        KPSS  Transformed  Critical Value 2.5%   \n",
       "27  Stationarity        KPSS  Transformed    Critical Value 1%   \n",
       "28     Normality     Shapiro  Transformed            Normality   \n",
       "29     Normality     Shapiro  Transformed              p-value   \n",
       "\n",
       "                     Setting        Value  \n",
       "0                                   394.0  \n",
       "1                                     0.0  \n",
       "2                               15.598985  \n",
       "3                                    14.0  \n",
       "4                                9.059408  \n",
       "5                               82.072874  \n",
       "6                               12.447098  \n",
       "7                                2.690653  \n",
       "8                                    39.0  \n",
       "9   {'alpha': 0.05, 'K': 24}   898.583563  \n",
       "10  {'alpha': 0.05, 'K': 48}  1006.762155  \n",
       "11  {'alpha': 0.05, 'K': 24}          0.0  \n",
       "12  {'alpha': 0.05, 'K': 48}          0.0  \n",
       "13  {'alpha': 0.05, 'K': 24}        False  \n",
       "14  {'alpha': 0.05, 'K': 48}        False  \n",
       "15           {'alpha': 0.05}         True  \n",
       "16           {'alpha': 0.05}     0.000565  \n",
       "17           {'alpha': 0.05}    -4.239263  \n",
       "18           {'alpha': 0.05}    -3.447815  \n",
       "19           {'alpha': 0.05}    -2.869237  \n",
       "20           {'alpha': 0.05}     -2.57087  \n",
       "21           {'alpha': 0.05}         True  \n",
       "22           {'alpha': 0.05}          0.1  \n",
       "23           {'alpha': 0.05}     0.083097  \n",
       "24           {'alpha': 0.05}        0.119  \n",
       "25           {'alpha': 0.05}        0.146  \n",
       "26           {'alpha': 0.05}        0.176  \n",
       "27           {'alpha': 0.05}        0.216  \n",
       "28           {'alpha': 0.05}        False  \n",
       "29           {'alpha': 0.05}          0.0  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exp.check_stats()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "301bc2fc-d849-43a4-b969-dea523bf323c",
   "metadata": {},
   "source": [
    "## What does all this mean\n",
    "\n",
    "- See documentation sktime <https://www.sktime.net/en/latest/api_reference/param_est.html>\n",
    "- Sktime refers to statsmodel <https://www.statsmodels.org/stable/tsa.html#descriptive-statistics-and-tests>\n",
    "- statsmodel results have been verified with at least one other statistical package: R, Stata or SAS."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a9790aef-1289-4b19-bd42-13ca42cd3503",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_ae44d th {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_ae44d_row0_col0, #T_ae44d_row0_col2, #T_ae44d_row0_col4, #T_ae44d_row0_col6, #T_ae44d_row1_col0, #T_ae44d_row1_col1, #T_ae44d_row1_col3, #T_ae44d_row1_col5, #T_ae44d_row1_col7, #T_ae44d_row2_col0, #T_ae44d_row2_col1, #T_ae44d_row2_col2, #T_ae44d_row2_col3, #T_ae44d_row2_col4, #T_ae44d_row2_col5, #T_ae44d_row2_col6, #T_ae44d_row2_col7, #T_ae44d_row3_col0, #T_ae44d_row3_col1, #T_ae44d_row3_col2, #T_ae44d_row3_col3, #T_ae44d_row3_col4, #T_ae44d_row3_col5, #T_ae44d_row3_col6, #T_ae44d_row3_col7, #T_ae44d_row4_col0, #T_ae44d_row4_col1, #T_ae44d_row4_col2, #T_ae44d_row4_col3, #T_ae44d_row4_col4, #T_ae44d_row4_col5, #T_ae44d_row4_col6, #T_ae44d_row4_col7, #T_ae44d_row5_col0, #T_ae44d_row5_col1, #T_ae44d_row5_col2, #T_ae44d_row5_col3, #T_ae44d_row5_col4, #T_ae44d_row5_col5, #T_ae44d_row5_col6, #T_ae44d_row5_col7, #T_ae44d_row6_col0, #T_ae44d_row6_col1, #T_ae44d_row6_col2, #T_ae44d_row6_col3, #T_ae44d_row6_col4, #T_ae44d_row6_col5, #T_ae44d_row6_col6, #T_ae44d_row6_col7, #T_ae44d_row7_col0, #T_ae44d_row7_col1, #T_ae44d_row7_col2, #T_ae44d_row7_col3, #T_ae44d_row7_col4, #T_ae44d_row7_col5, #T_ae44d_row7_col6, #T_ae44d_row7_col7, #T_ae44d_row8_col0, #T_ae44d_row8_col1, #T_ae44d_row8_col2, #T_ae44d_row8_col3, #T_ae44d_row8_col4, #T_ae44d_row8_col5, #T_ae44d_row8_col6, #T_ae44d_row8_col7, #T_ae44d_row9_col0, #T_ae44d_row9_col1, #T_ae44d_row9_col2, #T_ae44d_row9_col3, #T_ae44d_row9_col4, #T_ae44d_row9_col5, #T_ae44d_row9_col6, #T_ae44d_row9_col7, #T_ae44d_row10_col0, #T_ae44d_row10_col1, #T_ae44d_row10_col2, #T_ae44d_row10_col3, #T_ae44d_row10_col4, #T_ae44d_row10_col5, #T_ae44d_row10_col6, #T_ae44d_row10_col7, #T_ae44d_row11_col0, #T_ae44d_row11_col1, #T_ae44d_row11_col2, #T_ae44d_row11_col3, #T_ae44d_row11_col4, #T_ae44d_row11_col5, #T_ae44d_row11_col6, #T_ae44d_row11_col7, #T_ae44d_row12_col0, #T_ae44d_row12_col1, #T_ae44d_row12_col2, #T_ae44d_row12_col3, #T_ae44d_row12_col4, #T_ae44d_row12_col5, #T_ae44d_row12_col6, #T_ae44d_row12_col7, #T_ae44d_row13_col0, #T_ae44d_row13_col1, #T_ae44d_row13_col2, #T_ae44d_row13_col3, #T_ae44d_row13_col4, #T_ae44d_row13_col5, #T_ae44d_row13_col6, #T_ae44d_row13_col7, #T_ae44d_row14_col0, #T_ae44d_row14_col1, #T_ae44d_row14_col2, #T_ae44d_row14_col3, #T_ae44d_row14_col4, #T_ae44d_row14_col5, #T_ae44d_row14_col6, #T_ae44d_row14_col7, #T_ae44d_row15_col0, #T_ae44d_row15_col1, #T_ae44d_row15_col2, #T_ae44d_row15_col3, #T_ae44d_row15_col4, #T_ae44d_row15_col5, #T_ae44d_row15_col6, #T_ae44d_row15_col7, #T_ae44d_row16_col0, #T_ae44d_row16_col1, #T_ae44d_row16_col2, #T_ae44d_row16_col3, #T_ae44d_row16_col4, #T_ae44d_row16_col5, #T_ae44d_row16_col6, #T_ae44d_row16_col7, #T_ae44d_row17_col0, #T_ae44d_row17_col1, #T_ae44d_row17_col2, #T_ae44d_row17_col3, #T_ae44d_row17_col4, #T_ae44d_row17_col5, #T_ae44d_row17_col6, #T_ae44d_row17_col7, #T_ae44d_row18_col0, #T_ae44d_row18_col1, #T_ae44d_row18_col2, #T_ae44d_row18_col3, #T_ae44d_row18_col4, #T_ae44d_row18_col5, #T_ae44d_row18_col6, #T_ae44d_row18_col7, #T_ae44d_row19_col0, #T_ae44d_row19_col1, #T_ae44d_row19_col2, #T_ae44d_row19_col3, #T_ae44d_row19_col4, #T_ae44d_row19_col5, #T_ae44d_row19_col6, #T_ae44d_row19_col7, #T_ae44d_row20_col0, #T_ae44d_row20_col1, #T_ae44d_row20_col2, #T_ae44d_row20_col3, #T_ae44d_row20_col4, #T_ae44d_row20_col5, #T_ae44d_row20_col6, #T_ae44d_row20_col7, #T_ae44d_row21_col0, #T_ae44d_row21_col1, #T_ae44d_row21_col2, #T_ae44d_row21_col3, #T_ae44d_row21_col4, #T_ae44d_row21_col5, #T_ae44d_row21_col6, #T_ae44d_row21_col7, #T_ae44d_row22_col0, #T_ae44d_row22_col1, #T_ae44d_row22_col2, #T_ae44d_row22_col3, #T_ae44d_row22_col4, #T_ae44d_row22_col5, #T_ae44d_row22_col6, #T_ae44d_row22_col7, #T_ae44d_row23_col0, #T_ae44d_row23_col1, #T_ae44d_row23_col2, #T_ae44d_row23_col3, #T_ae44d_row23_col4, #T_ae44d_row23_col5, #T_ae44d_row23_col6, #T_ae44d_row23_col7, #T_ae44d_row24_col0, #T_ae44d_row24_col1, #T_ae44d_row24_col2, #T_ae44d_row24_col3, #T_ae44d_row24_col4, #T_ae44d_row24_col5, #T_ae44d_row24_col6, #T_ae44d_row24_col7, #T_ae44d_row25_col0, #T_ae44d_row25_col1, #T_ae44d_row25_col2, #T_ae44d_row25_col3, #T_ae44d_row25_col4, #T_ae44d_row25_col5, #T_ae44d_row25_col6, #T_ae44d_row25_col7, #T_ae44d_row26_col0, #T_ae44d_row26_col1, #T_ae44d_row26_col2, #T_ae44d_row26_col3, #T_ae44d_row26_col4, #T_ae44d_row26_col5, #T_ae44d_row26_col6, #T_ae44d_row26_col7, #T_ae44d_row27_col0, #T_ae44d_row27_col1, #T_ae44d_row27_col2, #T_ae44d_row27_col3, #T_ae44d_row27_col4, #T_ae44d_row27_col5, #T_ae44d_row27_col6, #T_ae44d_row27_col7 {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_ae44d_row0_col1, #T_ae44d_row0_col3, #T_ae44d_row0_col5, #T_ae44d_row0_col7, #T_ae44d_row1_col2, #T_ae44d_row1_col4, #T_ae44d_row1_col6 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "}\n",
       "#T_ae44d_row0_col8, #T_ae44d_row1_col8, #T_ae44d_row2_col8, #T_ae44d_row3_col8, #T_ae44d_row4_col8, #T_ae44d_row5_col8, #T_ae44d_row6_col8, #T_ae44d_row7_col8, #T_ae44d_row8_col8, #T_ae44d_row9_col8, #T_ae44d_row10_col8, #T_ae44d_row11_col8, #T_ae44d_row12_col8, #T_ae44d_row13_col8, #T_ae44d_row14_col8, #T_ae44d_row15_col8, #T_ae44d_row16_col8, #T_ae44d_row17_col8, #T_ae44d_row18_col8, #T_ae44d_row19_col8, #T_ae44d_row20_col8, #T_ae44d_row21_col8, #T_ae44d_row22_col8, #T_ae44d_row24_col8, #T_ae44d_row25_col8, #T_ae44d_row26_col8, #T_ae44d_row27_col8 {\n",
       "  text-align: left;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "#T_ae44d_row23_col8 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_ae44d\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_ae44d_level0_col0\" class=\"col_heading level0 col0\" >Model</th>\n",
       "      <th id=\"T_ae44d_level0_col1\" class=\"col_heading level0 col1\" >MASE</th>\n",
       "      <th id=\"T_ae44d_level0_col2\" class=\"col_heading level0 col2\" >RMSSE</th>\n",
       "      <th id=\"T_ae44d_level0_col3\" class=\"col_heading level0 col3\" >MAE</th>\n",
       "      <th id=\"T_ae44d_level0_col4\" class=\"col_heading level0 col4\" >RMSE</th>\n",
       "      <th id=\"T_ae44d_level0_col5\" class=\"col_heading level0 col5\" >MAPE</th>\n",
       "      <th id=\"T_ae44d_level0_col6\" class=\"col_heading level0 col6\" >SMAPE</th>\n",
       "      <th id=\"T_ae44d_level0_col7\" class=\"col_heading level0 col7\" >R2</th>\n",
       "      <th id=\"T_ae44d_level0_col8\" class=\"col_heading level0 col8\" >TT (Sec)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row0\" class=\"row_heading level0 row0\" >dt_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row0_col0\" class=\"data row0 col0\" >Decision Tree w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row0_col1\" class=\"data row0 col1\" >0.7840</td>\n",
       "      <td id=\"T_ae44d_row0_col2\" class=\"data row0 col2\" >0.5818</td>\n",
       "      <td id=\"T_ae44d_row0_col3\" class=\"data row0 col3\" >4.6563</td>\n",
       "      <td id=\"T_ae44d_row0_col4\" class=\"data row0 col4\" >5.6513</td>\n",
       "      <td id=\"T_ae44d_row0_col5\" class=\"data row0 col5\" >0.4271</td>\n",
       "      <td id=\"T_ae44d_row0_col6\" class=\"data row0 col6\" >0.3493</td>\n",
       "      <td id=\"T_ae44d_row0_col7\" class=\"data row0 col7\" >-0.5516</td>\n",
       "      <td id=\"T_ae44d_row0_col8\" class=\"data row0 col8\" >0.0933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row1\" class=\"row_heading level0 row1\" >gbr_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row1_col0\" class=\"data row1 col0\" >Gradient Boosting w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row1_col1\" class=\"data row1 col1\" >0.7851</td>\n",
       "      <td id=\"T_ae44d_row1_col2\" class=\"data row1 col2\" >0.5651</td>\n",
       "      <td id=\"T_ae44d_row1_col3\" class=\"data row1 col3\" >4.6759</td>\n",
       "      <td id=\"T_ae44d_row1_col4\" class=\"data row1 col4\" >5.5123</td>\n",
       "      <td id=\"T_ae44d_row1_col5\" class=\"data row1 col5\" >0.4511</td>\n",
       "      <td id=\"T_ae44d_row1_col6\" class=\"data row1 col6\" >0.3406</td>\n",
       "      <td id=\"T_ae44d_row1_col7\" class=\"data row1 col7\" >-0.5596</td>\n",
       "      <td id=\"T_ae44d_row1_col8\" class=\"data row1 col8\" >0.1100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row2\" class=\"row_heading level0 row2\" >xgboost_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row2_col0\" class=\"data row2 col0\" >Extreme Gradient Boosting w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row2_col1\" class=\"data row2 col1\" >0.8564</td>\n",
       "      <td id=\"T_ae44d_row2_col2\" class=\"data row2 col2\" >0.6094</td>\n",
       "      <td id=\"T_ae44d_row2_col3\" class=\"data row2 col3\" >5.1303</td>\n",
       "      <td id=\"T_ae44d_row2_col4\" class=\"data row2 col4\" >5.9715</td>\n",
       "      <td id=\"T_ae44d_row2_col5\" class=\"data row2 col5\" >0.4938</td>\n",
       "      <td id=\"T_ae44d_row2_col6\" class=\"data row2 col6\" >0.3701</td>\n",
       "      <td id=\"T_ae44d_row2_col7\" class=\"data row2 col7\" >-0.9327</td>\n",
       "      <td id=\"T_ae44d_row2_col8\" class=\"data row2 col8\" >0.1133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row3\" class=\"row_heading level0 row3\" >huber_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row3_col0\" class=\"data row3 col0\" >Huber w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row3_col1\" class=\"data row3 col1\" >0.8584</td>\n",
       "      <td id=\"T_ae44d_row3_col2\" class=\"data row3 col2\" >0.6283</td>\n",
       "      <td id=\"T_ae44d_row3_col3\" class=\"data row3 col3\" >5.1065</td>\n",
       "      <td id=\"T_ae44d_row3_col4\" class=\"data row3 col4\" >6.1222</td>\n",
       "      <td id=\"T_ae44d_row3_col5\" class=\"data row3 col5\" >0.4944</td>\n",
       "      <td id=\"T_ae44d_row3_col6\" class=\"data row3 col6\" >0.3682</td>\n",
       "      <td id=\"T_ae44d_row3_col7\" class=\"data row3 col7\" >-0.8949</td>\n",
       "      <td id=\"T_ae44d_row3_col8\" class=\"data row3 col8\" >0.1400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row4\" class=\"row_heading level0 row4\" >grand_means</th>\n",
       "      <td id=\"T_ae44d_row4_col0\" class=\"data row4 col0\" >Grand Means Forecaster</td>\n",
       "      <td id=\"T_ae44d_row4_col1\" class=\"data row4 col1\" >0.9570</td>\n",
       "      <td id=\"T_ae44d_row4_col2\" class=\"data row4 col2\" >0.6527</td>\n",
       "      <td id=\"T_ae44d_row4_col3\" class=\"data row4 col3\" >5.6907</td>\n",
       "      <td id=\"T_ae44d_row4_col4\" class=\"data row4 col4\" >6.3576</td>\n",
       "      <td id=\"T_ae44d_row4_col5\" class=\"data row4 col5\" >0.5903</td>\n",
       "      <td id=\"T_ae44d_row4_col6\" class=\"data row4 col6\" >0.4209</td>\n",
       "      <td id=\"T_ae44d_row4_col7\" class=\"data row4 col7\" >-1.0972</td>\n",
       "      <td id=\"T_ae44d_row4_col8\" class=\"data row4 col8\" >0.4233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row5\" class=\"row_heading level0 row5\" >knn_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row5_col0\" class=\"data row5 col0\" >K Neighbors w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row5_col1\" class=\"data row5 col1\" >0.9905</td>\n",
       "      <td id=\"T_ae44d_row5_col2\" class=\"data row5 col2\" >0.7135</td>\n",
       "      <td id=\"T_ae44d_row5_col3\" class=\"data row5 col3\" >5.8843</td>\n",
       "      <td id=\"T_ae44d_row5_col4\" class=\"data row5 col4\" >6.9445</td>\n",
       "      <td id=\"T_ae44d_row5_col5\" class=\"data row5 col5\" >0.5715</td>\n",
       "      <td id=\"T_ae44d_row5_col6\" class=\"data row5 col6\" >0.4053</td>\n",
       "      <td id=\"T_ae44d_row5_col7\" class=\"data row5 col7\" >-1.3989</td>\n",
       "      <td id=\"T_ae44d_row5_col8\" class=\"data row5 col8\" >0.1167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row6\" class=\"row_heading level0 row6\" >et_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row6_col0\" class=\"data row6 col0\" >Extra Trees w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row6_col1\" class=\"data row6 col1\" >0.9937</td>\n",
       "      <td id=\"T_ae44d_row6_col2\" class=\"data row6 col2\" >0.7140</td>\n",
       "      <td id=\"T_ae44d_row6_col3\" class=\"data row6 col3\" >5.8995</td>\n",
       "      <td id=\"T_ae44d_row6_col4\" class=\"data row6 col4\" >6.9412</td>\n",
       "      <td id=\"T_ae44d_row6_col5\" class=\"data row6 col5\" >0.5782</td>\n",
       "      <td id=\"T_ae44d_row6_col6\" class=\"data row6 col6\" >0.4078</td>\n",
       "      <td id=\"T_ae44d_row6_col7\" class=\"data row6 col7\" >-1.3476</td>\n",
       "      <td id=\"T_ae44d_row6_col8\" class=\"data row6 col8\" >0.1767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row7\" class=\"row_heading level0 row7\" >catboost_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row7_col0\" class=\"data row7 col0\" >CatBoost Regressor w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row7_col1\" class=\"data row7 col1\" >1.0142</td>\n",
       "      <td id=\"T_ae44d_row7_col2\" class=\"data row7 col2\" >0.7218</td>\n",
       "      <td id=\"T_ae44d_row7_col3\" class=\"data row7 col3\" >6.0176</td>\n",
       "      <td id=\"T_ae44d_row7_col4\" class=\"data row7 col4\" >7.0189</td>\n",
       "      <td id=\"T_ae44d_row7_col5\" class=\"data row7 col5\" >0.5917</td>\n",
       "      <td id=\"T_ae44d_row7_col6\" class=\"data row7 col6\" >0.4149</td>\n",
       "      <td id=\"T_ae44d_row7_col7\" class=\"data row7 col7\" >-1.4032</td>\n",
       "      <td id=\"T_ae44d_row7_col8\" class=\"data row7 col8\" >0.5333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row8\" class=\"row_heading level0 row8\" >llar_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row8_col0\" class=\"data row8 col0\" >Lasso Least Angular Regressor w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row8_col1\" class=\"data row8 col1\" >1.0349</td>\n",
       "      <td id=\"T_ae44d_row8_col2\" class=\"data row8 col2\" >0.7368</td>\n",
       "      <td id=\"T_ae44d_row8_col3\" class=\"data row8 col3\" >6.1565</td>\n",
       "      <td id=\"T_ae44d_row8_col4\" class=\"data row8 col4\" >7.1753</td>\n",
       "      <td id=\"T_ae44d_row8_col5\" class=\"data row8 col5\" >0.5896</td>\n",
       "      <td id=\"T_ae44d_row8_col6\" class=\"data row8 col6\" >0.4324</td>\n",
       "      <td id=\"T_ae44d_row8_col7\" class=\"data row8 col7\" >-1.5957</td>\n",
       "      <td id=\"T_ae44d_row8_col8\" class=\"data row8 col8\" >0.0933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row9\" class=\"row_heading level0 row9\" >lasso_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row9_col0\" class=\"data row9 col0\" >Lasso w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row9_col1\" class=\"data row9 col1\" >1.0350</td>\n",
       "      <td id=\"T_ae44d_row9_col2\" class=\"data row9 col2\" >0.7368</td>\n",
       "      <td id=\"T_ae44d_row9_col3\" class=\"data row9 col3\" >6.1570</td>\n",
       "      <td id=\"T_ae44d_row9_col4\" class=\"data row9 col4\" >7.1758</td>\n",
       "      <td id=\"T_ae44d_row9_col5\" class=\"data row9 col5\" >0.5896</td>\n",
       "      <td id=\"T_ae44d_row9_col6\" class=\"data row9 col6\" >0.4325</td>\n",
       "      <td id=\"T_ae44d_row9_col7\" class=\"data row9 col7\" >-1.5959</td>\n",
       "      <td id=\"T_ae44d_row9_col8\" class=\"data row9 col8\" >0.0933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row10\" class=\"row_heading level0 row10\" >en_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row10_col0\" class=\"data row10 col0\" >Elastic Net w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row10_col1\" class=\"data row10 col1\" >1.0786</td>\n",
       "      <td id=\"T_ae44d_row10_col2\" class=\"data row10 col2\" >0.7631</td>\n",
       "      <td id=\"T_ae44d_row10_col3\" class=\"data row10 col3\" >6.4032</td>\n",
       "      <td id=\"T_ae44d_row10_col4\" class=\"data row10 col4\" >7.4182</td>\n",
       "      <td id=\"T_ae44d_row10_col5\" class=\"data row10 col5\" >0.6066</td>\n",
       "      <td id=\"T_ae44d_row10_col6\" class=\"data row10 col6\" >0.4582</td>\n",
       "      <td id=\"T_ae44d_row10_col7\" class=\"data row10 col7\" >-1.7171</td>\n",
       "      <td id=\"T_ae44d_row10_col8\" class=\"data row10 col8\" >0.5400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row11\" class=\"row_heading level0 row11\" >auto_arima</th>\n",
       "      <td id=\"T_ae44d_row11_col0\" class=\"data row11 col0\" >Auto ARIMA</td>\n",
       "      <td id=\"T_ae44d_row11_col1\" class=\"data row11 col1\" >1.1070</td>\n",
       "      <td id=\"T_ae44d_row11_col2\" class=\"data row11 col2\" >0.7763</td>\n",
       "      <td id=\"T_ae44d_row11_col3\" class=\"data row11 col3\" >6.5406</td>\n",
       "      <td id=\"T_ae44d_row11_col4\" class=\"data row11 col4\" >7.5086</td>\n",
       "      <td id=\"T_ae44d_row11_col5\" class=\"data row11 col5\" >0.6193</td>\n",
       "      <td id=\"T_ae44d_row11_col6\" class=\"data row11 col6\" >0.4955</td>\n",
       "      <td id=\"T_ae44d_row11_col7\" class=\"data row11 col7\" >-1.6275</td>\n",
       "      <td id=\"T_ae44d_row11_col8\" class=\"data row11 col8\" >2.6533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row12\" class=\"row_heading level0 row12\" >omp_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row12_col0\" class=\"data row12 col0\" >Orthogonal Matching Pursuit w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row12_col1\" class=\"data row12 col1\" >1.1098</td>\n",
       "      <td id=\"T_ae44d_row12_col2\" class=\"data row12 col2\" >0.7852</td>\n",
       "      <td id=\"T_ae44d_row12_col3\" class=\"data row12 col3\" >6.5709</td>\n",
       "      <td id=\"T_ae44d_row12_col4\" class=\"data row12 col4\" >7.6138</td>\n",
       "      <td id=\"T_ae44d_row12_col5\" class=\"data row12 col5\" >0.6459</td>\n",
       "      <td id=\"T_ae44d_row12_col6\" class=\"data row12 col6\" >0.4434</td>\n",
       "      <td id=\"T_ae44d_row12_col7\" class=\"data row12 col7\" >-1.7825</td>\n",
       "      <td id=\"T_ae44d_row12_col8\" class=\"data row12 col8\" >0.5300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row13\" class=\"row_heading level0 row13\" >polytrend</th>\n",
       "      <td id=\"T_ae44d_row13_col0\" class=\"data row13 col0\" >Polynomial Trend Forecaster</td>\n",
       "      <td id=\"T_ae44d_row13_col1\" class=\"data row13 col1\" >1.1587</td>\n",
       "      <td id=\"T_ae44d_row13_col2\" class=\"data row13 col2\" >0.7919</td>\n",
       "      <td id=\"T_ae44d_row13_col3\" class=\"data row13 col3\" >6.8715</td>\n",
       "      <td id=\"T_ae44d_row13_col4\" class=\"data row13 col4\" >7.6948</td>\n",
       "      <td id=\"T_ae44d_row13_col5\" class=\"data row13 col5\" >0.7231</td>\n",
       "      <td id=\"T_ae44d_row13_col6\" class=\"data row13 col6\" >0.4794</td>\n",
       "      <td id=\"T_ae44d_row13_col7\" class=\"data row13 col7\" >-1.9075</td>\n",
       "      <td id=\"T_ae44d_row13_col8\" class=\"data row13 col8\" >0.4100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row14\" class=\"row_heading level0 row14\" >br_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row14_col0\" class=\"data row14 col0\" >Bayesian Ridge w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row14_col1\" class=\"data row14 col1\" >1.1640</td>\n",
       "      <td id=\"T_ae44d_row14_col2\" class=\"data row14 col2\" >0.8108</td>\n",
       "      <td id=\"T_ae44d_row14_col3\" class=\"data row14 col3\" >6.8848</td>\n",
       "      <td id=\"T_ae44d_row14_col4\" class=\"data row14 col4\" >7.8572</td>\n",
       "      <td id=\"T_ae44d_row14_col5\" class=\"data row14 col5\" >0.6455</td>\n",
       "      <td id=\"T_ae44d_row14_col6\" class=\"data row14 col6\" >0.5092</td>\n",
       "      <td id=\"T_ae44d_row14_col7\" class=\"data row14 col7\" >-1.9573</td>\n",
       "      <td id=\"T_ae44d_row14_col8\" class=\"data row14 col8\" >0.5233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row15\" class=\"row_heading level0 row15\" >ada_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row15_col0\" class=\"data row15 col0\" >AdaBoost w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row15_col1\" class=\"data row15 col1\" >1.1762</td>\n",
       "      <td id=\"T_ae44d_row15_col2\" class=\"data row15 col2\" >0.8178</td>\n",
       "      <td id=\"T_ae44d_row15_col3\" class=\"data row15 col3\" >6.8570</td>\n",
       "      <td id=\"T_ae44d_row15_col4\" class=\"data row15 col4\" >7.8220</td>\n",
       "      <td id=\"T_ae44d_row15_col5\" class=\"data row15 col5\" >0.6588</td>\n",
       "      <td id=\"T_ae44d_row15_col6\" class=\"data row15 col6\" >0.4395</td>\n",
       "      <td id=\"T_ae44d_row15_col7\" class=\"data row15 col7\" >-1.8145</td>\n",
       "      <td id=\"T_ae44d_row15_col8\" class=\"data row15 col8\" >0.1133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row16\" class=\"row_heading level0 row16\" >ridge_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row16_col0\" class=\"data row16 col0\" >Ridge w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row16_col1\" class=\"data row16 col1\" >1.2041</td>\n",
       "      <td id=\"T_ae44d_row16_col2\" class=\"data row16 col2\" >0.8346</td>\n",
       "      <td id=\"T_ae44d_row16_col3\" class=\"data row16 col3\" >7.1102</td>\n",
       "      <td id=\"T_ae44d_row16_col4\" class=\"data row16 col4\" >8.0768</td>\n",
       "      <td id=\"T_ae44d_row16_col5\" class=\"data row16 col5\" >0.6639</td>\n",
       "      <td id=\"T_ae44d_row16_col6\" class=\"data row16 col6\" >0.5326</td>\n",
       "      <td id=\"T_ae44d_row16_col7\" class=\"data row16 col7\" >-2.0881</td>\n",
       "      <td id=\"T_ae44d_row16_col8\" class=\"data row16 col8\" >0.5200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row17\" class=\"row_heading level0 row17\" >lar_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row17_col0\" class=\"data row17 col0\" >Least Angular Regressor w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row17_col1\" class=\"data row17 col1\" >1.2043</td>\n",
       "      <td id=\"T_ae44d_row17_col2\" class=\"data row17 col2\" >0.8347</td>\n",
       "      <td id=\"T_ae44d_row17_col3\" class=\"data row17 col3\" >7.1113</td>\n",
       "      <td id=\"T_ae44d_row17_col4\" class=\"data row17 col4\" >8.0780</td>\n",
       "      <td id=\"T_ae44d_row17_col5\" class=\"data row17 col5\" >0.6640</td>\n",
       "      <td id=\"T_ae44d_row17_col6\" class=\"data row17 col6\" >0.5327</td>\n",
       "      <td id=\"T_ae44d_row17_col7\" class=\"data row17 col7\" >-2.0888</td>\n",
       "      <td id=\"T_ae44d_row17_col8\" class=\"data row17 col8\" >0.1200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row18\" class=\"row_heading level0 row18\" >lr_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row18_col0\" class=\"data row18 col0\" >Linear w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row18_col1\" class=\"data row18 col1\" >1.2043</td>\n",
       "      <td id=\"T_ae44d_row18_col2\" class=\"data row18 col2\" >0.8347</td>\n",
       "      <td id=\"T_ae44d_row18_col3\" class=\"data row18 col3\" >7.1113</td>\n",
       "      <td id=\"T_ae44d_row18_col4\" class=\"data row18 col4\" >8.0780</td>\n",
       "      <td id=\"T_ae44d_row18_col5\" class=\"data row18 col5\" >0.6640</td>\n",
       "      <td id=\"T_ae44d_row18_col6\" class=\"data row18 col6\" >0.5327</td>\n",
       "      <td id=\"T_ae44d_row18_col7\" class=\"data row18 col7\" >-2.0888</td>\n",
       "      <td id=\"T_ae44d_row18_col8\" class=\"data row18 col8\" >0.5367</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row19\" class=\"row_heading level0 row19\" >theta</th>\n",
       "      <td id=\"T_ae44d_row19_col0\" class=\"data row19 col0\" >Theta Forecaster</td>\n",
       "      <td id=\"T_ae44d_row19_col1\" class=\"data row19 col1\" >1.3256</td>\n",
       "      <td id=\"T_ae44d_row19_col2\" class=\"data row19 col2\" >0.9353</td>\n",
       "      <td id=\"T_ae44d_row19_col3\" class=\"data row19 col3\" >7.7370</td>\n",
       "      <td id=\"T_ae44d_row19_col4\" class=\"data row19 col4\" >8.9518</td>\n",
       "      <td id=\"T_ae44d_row19_col5\" class=\"data row19 col5\" >0.7157</td>\n",
       "      <td id=\"T_ae44d_row19_col6\" class=\"data row19 col6\" >0.4629</td>\n",
       "      <td id=\"T_ae44d_row19_col7\" class=\"data row19 col7\" >-2.6523</td>\n",
       "      <td id=\"T_ae44d_row19_col8\" class=\"data row19 col8\" >0.5067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row20\" class=\"row_heading level0 row20\" >naive</th>\n",
       "      <td id=\"T_ae44d_row20_col0\" class=\"data row20 col0\" >Naive Forecaster</td>\n",
       "      <td id=\"T_ae44d_row20_col1\" class=\"data row20 col1\" >1.3461</td>\n",
       "      <td id=\"T_ae44d_row20_col2\" class=\"data row20 col2\" >0.9211</td>\n",
       "      <td id=\"T_ae44d_row20_col3\" class=\"data row20 col3\" >7.9000</td>\n",
       "      <td id=\"T_ae44d_row20_col4\" class=\"data row20 col4\" >8.8715</td>\n",
       "      <td id=\"T_ae44d_row20_col5\" class=\"data row20 col5\" >0.7694</td>\n",
       "      <td id=\"T_ae44d_row20_col6\" class=\"data row20 col6\" >0.4899</td>\n",
       "      <td id=\"T_ae44d_row20_col7\" class=\"data row20 col7\" >-2.7436</td>\n",
       "      <td id=\"T_ae44d_row20_col8\" class=\"data row20 col8\" >0.5633</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row21\" class=\"row_heading level0 row21\" >rf_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row21_col0\" class=\"data row21 col0\" >Random Forest w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row21_col1\" class=\"data row21 col1\" >1.3849</td>\n",
       "      <td id=\"T_ae44d_row21_col2\" class=\"data row21 col2\" >0.9491</td>\n",
       "      <td id=\"T_ae44d_row21_col3\" class=\"data row21 col3\" >8.0766</td>\n",
       "      <td id=\"T_ae44d_row21_col4\" class=\"data row21 col4\" >9.0822</td>\n",
       "      <td id=\"T_ae44d_row21_col5\" class=\"data row21 col5\" >0.7651</td>\n",
       "      <td id=\"T_ae44d_row21_col6\" class=\"data row21 col6\" >0.4861</td>\n",
       "      <td id=\"T_ae44d_row21_col7\" class=\"data row21 col7\" >-2.7598</td>\n",
       "      <td id=\"T_ae44d_row21_col8\" class=\"data row21 col8\" >0.1833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row22\" class=\"row_heading level0 row22\" >exp_smooth</th>\n",
       "      <td id=\"T_ae44d_row22_col0\" class=\"data row22 col0\" >Exponential Smoothing</td>\n",
       "      <td id=\"T_ae44d_row22_col1\" class=\"data row22 col1\" >1.8392</td>\n",
       "      <td id=\"T_ae44d_row22_col2\" class=\"data row22 col2\" >1.2175</td>\n",
       "      <td id=\"T_ae44d_row22_col3\" class=\"data row22 col3\" >10.7538</td>\n",
       "      <td id=\"T_ae44d_row22_col4\" class=\"data row22 col4\" >11.6661</td>\n",
       "      <td id=\"T_ae44d_row22_col5\" class=\"data row22 col5\" >0.9858</td>\n",
       "      <td id=\"T_ae44d_row22_col6\" class=\"data row22 col6\" >0.5957</td>\n",
       "      <td id=\"T_ae44d_row22_col7\" class=\"data row22 col7\" >-5.2042</td>\n",
       "      <td id=\"T_ae44d_row22_col8\" class=\"data row22 col8\" >0.4333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row23\" class=\"row_heading level0 row23\" >arima</th>\n",
       "      <td id=\"T_ae44d_row23_col0\" class=\"data row23 col0\" >ARIMA</td>\n",
       "      <td id=\"T_ae44d_row23_col1\" class=\"data row23 col1\" >1.9264</td>\n",
       "      <td id=\"T_ae44d_row23_col2\" class=\"data row23 col2\" >1.2888</td>\n",
       "      <td id=\"T_ae44d_row23_col3\" class=\"data row23 col3\" >11.2028</td>\n",
       "      <td id=\"T_ae44d_row23_col4\" class=\"data row23 col4\" >12.2847</td>\n",
       "      <td id=\"T_ae44d_row23_col5\" class=\"data row23 col5\" >0.9996</td>\n",
       "      <td id=\"T_ae44d_row23_col6\" class=\"data row23 col6\" >0.5835</td>\n",
       "      <td id=\"T_ae44d_row23_col7\" class=\"data row23 col7\" >-6.1658</td>\n",
       "      <td id=\"T_ae44d_row23_col8\" class=\"data row23 col8\" >0.0233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row24\" class=\"row_heading level0 row24\" >croston</th>\n",
       "      <td id=\"T_ae44d_row24_col0\" class=\"data row24 col0\" >Croston</td>\n",
       "      <td id=\"T_ae44d_row24_col1\" class=\"data row24 col1\" >1.9302</td>\n",
       "      <td id=\"T_ae44d_row24_col2\" class=\"data row24 col2\" >1.2406</td>\n",
       "      <td id=\"T_ae44d_row24_col3\" class=\"data row24 col3\" >11.0885</td>\n",
       "      <td id=\"T_ae44d_row24_col4\" class=\"data row24 col4\" >11.6822</td>\n",
       "      <td id=\"T_ae44d_row24_col5\" class=\"data row24 col5\" >1.0485</td>\n",
       "      <td id=\"T_ae44d_row24_col6\" class=\"data row24 col6\" >0.5636</td>\n",
       "      <td id=\"T_ae44d_row24_col7\" class=\"data row24 col7\" >-6.1902</td>\n",
       "      <td id=\"T_ae44d_row24_col8\" class=\"data row24 col8\" >0.4133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row25\" class=\"row_heading level0 row25\" >lightgbm_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row25_col0\" class=\"data row25 col0\" >Light Gradient Boosting w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row25_col1\" class=\"data row25 col1\" >2.4842</td>\n",
       "      <td id=\"T_ae44d_row25_col2\" class=\"data row25 col2\" >1.6088</td>\n",
       "      <td id=\"T_ae44d_row25_col3\" class=\"data row25 col3\" >14.3037</td>\n",
       "      <td id=\"T_ae44d_row25_col4\" class=\"data row25 col4\" >15.1940</td>\n",
       "      <td id=\"T_ae44d_row25_col5\" class=\"data row25 col5\" >1.3002</td>\n",
       "      <td id=\"T_ae44d_row25_col6\" class=\"data row25 col6\" >0.6469</td>\n",
       "      <td id=\"T_ae44d_row25_col7\" class=\"data row25 col7\" >-10.7331</td>\n",
       "      <td id=\"T_ae44d_row25_col8\" class=\"data row25 col8\" >0.1033</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row26\" class=\"row_heading level0 row26\" >snaive</th>\n",
       "      <td id=\"T_ae44d_row26_col0\" class=\"data row26 col0\" >Seasonal Naive Forecaster</td>\n",
       "      <td id=\"T_ae44d_row26_col1\" class=\"data row26 col1\" >2.7172</td>\n",
       "      <td id=\"T_ae44d_row26_col2\" class=\"data row26 col2\" >1.8540</td>\n",
       "      <td id=\"T_ae44d_row26_col3\" class=\"data row26 col3\" >15.5222</td>\n",
       "      <td id=\"T_ae44d_row26_col4\" class=\"data row26 col4\" >17.3566</td>\n",
       "      <td id=\"T_ae44d_row26_col5\" class=\"data row26 col5\" >1.3198</td>\n",
       "      <td id=\"T_ae44d_row26_col6\" class=\"data row26 col6\" >0.6134</td>\n",
       "      <td id=\"T_ae44d_row26_col7\" class=\"data row26 col7\" >-16.4375</td>\n",
       "      <td id=\"T_ae44d_row26_col8\" class=\"data row26 col8\" >0.4167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_ae44d_level0_row27\" class=\"row_heading level0 row27\" >par_cds_dt</th>\n",
       "      <td id=\"T_ae44d_row27_col0\" class=\"data row27 col0\" >Passive Aggressive w/ Cond. Deseasonalize & Detrending</td>\n",
       "      <td id=\"T_ae44d_row27_col1\" class=\"data row27 col1\" >946.6825</td>\n",
       "      <td id=\"T_ae44d_row27_col2\" class=\"data row27 col2\" >1200.3324</td>\n",
       "      <td id=\"T_ae44d_row27_col3\" class=\"data row27 col3\" >5308.9380</td>\n",
       "      <td id=\"T_ae44d_row27_col4\" class=\"data row27 col4\" >11006.0074</td>\n",
       "      <td id=\"T_ae44d_row27_col5\" class=\"data row27 col5\" >317.4566</td>\n",
       "      <td id=\"T_ae44d_row27_col6\" class=\"data row27 col6\" >1.2519</td>\n",
       "      <td id=\"T_ae44d_row27_col7\" class=\"data row27 col7\" >-10726566.4294</td>\n",
       "      <td id=\"T_ae44d_row27_col8\" class=\"data row27 col8\" >0.0967</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x284b0f130>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "best = exp.compare_models()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "041ebfbf-9cd9-4d83-afe7-c755870591c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "        <script type=\"text/javascript\">\n",
       "        window.PlotlyConfig = {MathJaxConfig: 'local'};\n",
       "        if (window.MathJax && window.MathJax.Hub && window.MathJax.Hub.Config) {window.MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}\n",
       "        if (typeof require !== 'undefined') {\n",
       "        require.undef(\"plotly\");\n",
       "        define('plotly', function(require, exports, module) {\n",
       "            /**\n",
       "* plotly.js v2.20.0\n",
       "* Copyright 2012-2023, Plotly, Inc.\n",
       "* All rights reserved.\n",
       "* Licensed under the MIT license\n",
       "*/\n",
       "/*! For license information please see plotly.min.js.LICENSE.txt */\n",