First Test Model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "02f48d44",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pycaret\n",
    "import pandas as pd\n",
    "from pycaret.classification import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "357e71f5",
   "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>time</th>\n",
       "      <th>location</th>\n",
       "      <th>consumption</th>\n",
       "      <th>temperature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>bergen</td>\n",
       "      <td>1.113325</td>\n",
       "      <td>-0.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>oslo</td>\n",
       "      <td>4.092830</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>stavanger</td>\n",
       "      <td>2.057858</td>\n",
       "      <td>1.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>tromsø</td>\n",
       "      <td>1.246582</td>\n",
       "      <td>-3.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>trondheim</td>\n",
       "      <td>1.970098</td>\n",
       "      <td>-2.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49489</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>helsingfors</td>\n",
       "      <td>6.333000</td>\n",
       "      <td>-1.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49490</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>oslo</td>\n",
       "      <td>12.134655</td>\n",
       "      <td>-1.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49491</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>stavanger</td>\n",
       "      <td>5.622820</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49492</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>tromsø</td>\n",
       "      <td>2.018333</td>\n",
       "      <td>-2.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49493</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>trondheim</td>\n",
       "      <td>3.749047</td>\n",
       "      <td>-3.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>49494 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      time     location  consumption  temperature\n",
       "0      2022-04-07 21:00:00       bergen     1.113325         -0.3\n",
       "1      2022-04-07 21:00:00         oslo     4.092830          1.0\n",
       "2      2022-04-07 21:00:00    stavanger     2.057858          1.3\n",
       "3      2022-04-07 21:00:00       tromsø     1.246582         -3.9\n",
       "4      2022-04-07 21:00:00    trondheim     1.970098         -2.8\n",
       "...                    ...          ...          ...          ...\n",
       "49489  2023-04-02 21:00:00  helsingfors     6.333000         -1.1\n",
       "49490  2023-04-02 21:00:00         oslo    12.134655         -1.1\n",
       "49491  2023-04-02 21:00:00    stavanger     5.622820          0.5\n",
       "49492  2023-04-02 21:00:00       tromsø     2.018333         -2.5\n",
       "49493  2023-04-02 21:00:00    trondheim     3.749047         -3.8\n",
       "\n",
       "[49494 rows x 4 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load Dataset\n",
    "df = pd.read_csv('consumption_temp.csv')  \n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b84b7d0d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['bergen' 'oslo' 'stavanger' 'tromsø' 'trondheim' 'helsingfors']\n",
      "--------------\n",
      "Index([], dtype='object')\n",
      "--------------\n",
      "Highest value: 30.5, Lowest value: -17.299999237060547\n",
      "--------------\n",
      "time           2022-07-14 21:00:00\n",
      "location               helsingfors\n",
      "consumption                    0.0\n",
      "temperature                   15.8\n",
      "Name: 11761, dtype: object\n",
      "--------------\n"
     ]
    }
   ],
   "source": [
    "#Just some stuff to look into the Dataset\n",
    "\n",
    "#Distinct Values \n",
    "print(df['location'].unique()) \n",
    "print(\"--------------\")\n",
    "\n",
    "#Are there empty data values?\n",
    "print(df.columns[df.isnull().any()]) # Answer is no!\n",
    "print(\"--------------\")\n",
    "\n",
    "#Highest lowest\n",
    "print(f\"Highest value: {df['temperature'].max()}, Lowest value: {df['temperature'].min()}\")\n",
    "print(\"--------------\")\n",
    "\n",
    "#Find first entry of helsingfors\n",
    "value_to_find = 'helsingfors'\n",
    "print(df[df['location'] == value_to_find].iloc[0])\n",
    "print(\"--------------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "49e24618",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "#T_a3e7b_row10_col1 {\n",
       "  background-color: lightgreen;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_a3e7b\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_a3e7b_level0_col0\" class=\"col_heading level0 col0\" >Description</th>\n",
       "      <th id=\"T_a3e7b_level0_col1\" class=\"col_heading level0 col1\" >Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_a3e7b_row0_col0\" class=\"data row0 col0\" >Session id</td>\n",
       "      <td id=\"T_a3e7b_row0_col1\" class=\"data row0 col1\" >123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_a3e7b_row1_col0\" class=\"data row1 col0\" >Target</td>\n",
       "      <td id=\"T_a3e7b_row1_col1\" class=\"data row1 col1\" >location</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_a3e7b_row2_col0\" class=\"data row2 col0\" >Target type</td>\n",
       "      <td id=\"T_a3e7b_row2_col1\" class=\"data row2 col1\" >Multiclass</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_a3e7b_row3_col0\" class=\"data row3 col0\" >Target mapping</td>\n",
       "      <td id=\"T_a3e7b_row3_col1\" class=\"data row3 col1\" >bergen: 0, helsingfors: 1, oslo: 2, stavanger: 3, tromsø: 4, trondheim: 5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_a3e7b_row4_col0\" class=\"data row4 col0\" >Original data shape</td>\n",
       "      <td id=\"T_a3e7b_row4_col1\" class=\"data row4 col1\" >(49494, 4)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_a3e7b_row5_col0\" class=\"data row5 col0\" >Transformed data shape</td>\n",
       "      <td id=\"T_a3e7b_row5_col1\" class=\"data row5 col1\" >(49494, 4)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_a3e7b_row6_col0\" class=\"data row6 col0\" >Transformed train set shape</td>\n",
       "      <td id=\"T_a3e7b_row6_col1\" class=\"data row6 col1\" >(34645, 4)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_a3e7b_row7_col0\" class=\"data row7 col0\" >Transformed test set shape</td>\n",
       "      <td id=\"T_a3e7b_row7_col1\" class=\"data row7 col1\" >(14849, 4)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_a3e7b_row8_col0\" class=\"data row8 col0\" >Numeric features</td>\n",
       "      <td id=\"T_a3e7b_row8_col1\" class=\"data row8 col1\" >2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_a3e7b_row9_col0\" class=\"data row9 col0\" >Categorical features</td>\n",
       "      <td id=\"T_a3e7b_row9_col1\" class=\"data row9 col1\" >1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row10\" class=\"row_heading level0 row10\" >10</th>\n",
       "      <td id=\"T_a3e7b_row10_col0\" class=\"data row10 col0\" >Preprocess</td>\n",
       "      <td id=\"T_a3e7b_row10_col1\" class=\"data row10 col1\" >True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row11\" class=\"row_heading level0 row11\" >11</th>\n",
       "      <td id=\"T_a3e7b_row11_col0\" class=\"data row11 col0\" >Imputation type</td>\n",
       "      <td id=\"T_a3e7b_row11_col1\" class=\"data row11 col1\" >simple</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row12\" class=\"row_heading level0 row12\" >12</th>\n",
       "      <td id=\"T_a3e7b_row12_col0\" class=\"data row12 col0\" >Numeric imputation</td>\n",
       "      <td id=\"T_a3e7b_row12_col1\" class=\"data row12 col1\" >mean</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row13\" class=\"row_heading level0 row13\" >13</th>\n",
       "      <td id=\"T_a3e7b_row13_col0\" class=\"data row13 col0\" >Categorical imputation</td>\n",
       "      <td id=\"T_a3e7b_row13_col1\" class=\"data row13 col1\" >mode</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row14\" class=\"row_heading level0 row14\" >14</th>\n",
       "      <td id=\"T_a3e7b_row14_col0\" class=\"data row14 col0\" >Maximum one-hot encoding</td>\n",
       "      <td id=\"T_a3e7b_row14_col1\" class=\"data row14 col1\" >25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row15\" class=\"row_heading level0 row15\" >15</th>\n",
       "      <td id=\"T_a3e7b_row15_col0\" class=\"data row15 col0\" >Encoding method</td>\n",
       "      <td id=\"T_a3e7b_row15_col1\" class=\"data row15 col1\" >None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row16\" class=\"row_heading level0 row16\" >16</th>\n",
       "      <td id=\"T_a3e7b_row16_col0\" class=\"data row16 col0\" >Fold Generator</td>\n",
       "      <td id=\"T_a3e7b_row16_col1\" class=\"data row16 col1\" >StratifiedKFold</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row17\" class=\"row_heading level0 row17\" >17</th>\n",
       "      <td id=\"T_a3e7b_row17_col0\" class=\"data row17 col0\" >Fold Number</td>\n",
       "      <td id=\"T_a3e7b_row17_col1\" class=\"data row17 col1\" >10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row18\" class=\"row_heading level0 row18\" >18</th>\n",
       "      <td id=\"T_a3e7b_row18_col0\" class=\"data row18 col0\" >CPU Jobs</td>\n",
       "      <td id=\"T_a3e7b_row18_col1\" class=\"data row18 col1\" >-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row19\" class=\"row_heading level0 row19\" >19</th>\n",
       "      <td id=\"T_a3e7b_row19_col0\" class=\"data row19 col0\" >Use GPU</td>\n",
       "      <td id=\"T_a3e7b_row19_col1\" class=\"data row19 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row20\" class=\"row_heading level0 row20\" >20</th>\n",
       "      <td id=\"T_a3e7b_row20_col0\" class=\"data row20 col0\" >Log Experiment</td>\n",
       "      <td id=\"T_a3e7b_row20_col1\" class=\"data row20 col1\" >False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row21\" class=\"row_heading level0 row21\" >21</th>\n",
       "      <td id=\"T_a3e7b_row21_col0\" class=\"data row21 col0\" >Experiment Name</td>\n",
       "      <td id=\"T_a3e7b_row21_col1\" class=\"data row21 col1\" >clf-default-name</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a3e7b_level0_row22\" class=\"row_heading level0 row22\" >22</th>\n",
       "      <td id=\"T_a3e7b_row22_col0\" class=\"data row22 col0\" >USI</td>\n",
       "      <td id=\"T_a3e7b_row22_col1\" class=\"data row22 col1\" >aba1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x2c562625160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#supervised machine learning Classification for Location\n",
    "\n",
    "s = setup(df, target = 'location', session_id = 123)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "8c653a26",
   "metadata": {},
   "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_1e62b th {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_1e62b_row0_col0, #T_1e62b_row1_col0, #T_1e62b_row1_col1, #T_1e62b_row1_col2, #T_1e62b_row1_col3, #T_1e62b_row1_col4, #T_1e62b_row1_col5, #T_1e62b_row1_col6, #T_1e62b_row1_col7, #T_1e62b_row2_col0, #T_1e62b_row2_col1, #T_1e62b_row2_col2, #T_1e62b_row2_col3, #T_1e62b_row2_col4, #T_1e62b_row2_col5, #T_1e62b_row2_col6, #T_1e62b_row2_col7, #T_1e62b_row3_col0, #T_1e62b_row3_col1, #T_1e62b_row3_col2, #T_1e62b_row3_col3, #T_1e62b_row3_col4, #T_1e62b_row3_col5, #T_1e62b_row3_col6, #T_1e62b_row3_col7, #T_1e62b_row4_col0, #T_1e62b_row4_col1, #T_1e62b_row4_col2, #T_1e62b_row4_col3, #T_1e62b_row4_col4, #T_1e62b_row4_col5, #T_1e62b_row4_col6, #T_1e62b_row4_col7, #T_1e62b_row5_col0, #T_1e62b_row5_col1, #T_1e62b_row5_col2, #T_1e62b_row5_col3, #T_1e62b_row5_col4, #T_1e62b_row5_col5, #T_1e62b_row5_col6, #T_1e62b_row5_col7, #T_1e62b_row6_col0, #T_1e62b_row6_col1, #T_1e62b_row6_col2, #T_1e62b_row6_col3, #T_1e62b_row6_col4, #T_1e62b_row6_col5, #T_1e62b_row6_col6, #T_1e62b_row6_col7, #T_1e62b_row7_col0, #T_1e62b_row7_col1, #T_1e62b_row7_col2, #T_1e62b_row7_col3, #T_1e62b_row7_col4, #T_1e62b_row7_col5, #T_1e62b_row7_col6, #T_1e62b_row7_col7, #T_1e62b_row8_col0, #T_1e62b_row8_col1, #T_1e62b_row8_col2, #T_1e62b_row8_col3, #T_1e62b_row8_col4, #T_1e62b_row8_col5, #T_1e62b_row8_col6, #T_1e62b_row8_col7, #T_1e62b_row9_col0, #T_1e62b_row9_col1, #T_1e62b_row9_col2, #T_1e62b_row9_col3, #T_1e62b_row9_col4, #T_1e62b_row9_col5, #T_1e62b_row9_col6, #T_1e62b_row9_col7, #T_1e62b_row10_col0, #T_1e62b_row10_col1, #T_1e62b_row10_col2, #T_1e62b_row10_col3, #T_1e62b_row10_col4, #T_1e62b_row10_col5, #T_1e62b_row10_col6, #T_1e62b_row10_col7, #T_1e62b_row11_col0, #T_1e62b_row11_col1, #T_1e62b_row11_col2, #T_1e62b_row11_col3, #T_1e62b_row11_col4, #T_1e62b_row11_col5, #T_1e62b_row11_col6, #T_1e62b_row11_col7, #T_1e62b_row12_col0, #T_1e62b_row12_col1, #T_1e62b_row12_col2, #T_1e62b_row12_col3, #T_1e62b_row12_col4, #T_1e62b_row12_col5, #T_1e62b_row12_col6, #T_1e62b_row12_col7, #T_1e62b_row13_col0, #T_1e62b_row13_col1, #T_1e62b_row13_col2, #T_1e62b_row13_col3, #T_1e62b_row13_col4, #T_1e62b_row13_col5, #T_1e62b_row13_col6, #T_1e62b_row13_col7 {\n",
       "  text-align: left;\n",
       "}\n",
       "#T_1e62b_row0_col1, #T_1e62b_row0_col2, #T_1e62b_row0_col3, #T_1e62b_row0_col4, #T_1e62b_row0_col5, #T_1e62b_row0_col6, #T_1e62b_row0_col7 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "}\n",
       "#T_1e62b_row0_col8, #T_1e62b_row1_col8, #T_1e62b_row2_col8, #T_1e62b_row3_col8, #T_1e62b_row4_col8, #T_1e62b_row5_col8, #T_1e62b_row7_col8, #T_1e62b_row8_col8, #T_1e62b_row9_col8, #T_1e62b_row11_col8, #T_1e62b_row12_col8 {\n",
       "  text-align: left;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "#T_1e62b_row6_col8, #T_1e62b_row10_col8, #T_1e62b_row13_col8 {\n",
       "  text-align: left;\n",
       "  background-color: yellow;\n",
       "  background-color: lightgrey;\n",
       "}\n",
       "</style>\n",
       "<table id=\"T_1e62b\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_1e62b_level0_col0\" class=\"col_heading level0 col0\" >Model</th>\n",
       "      <th id=\"T_1e62b_level0_col1\" class=\"col_heading level0 col1\" >Accuracy</th>\n",
       "      <th id=\"T_1e62b_level0_col2\" class=\"col_heading level0 col2\" >AUC</th>\n",
       "      <th id=\"T_1e62b_level0_col3\" class=\"col_heading level0 col3\" >Recall</th>\n",
       "      <th id=\"T_1e62b_level0_col4\" class=\"col_heading level0 col4\" >Prec.</th>\n",
       "      <th id=\"T_1e62b_level0_col5\" class=\"col_heading level0 col5\" >F1</th>\n",
       "      <th id=\"T_1e62b_level0_col6\" class=\"col_heading level0 col6\" >Kappa</th>\n",
       "      <th id=\"T_1e62b_level0_col7\" class=\"col_heading level0 col7\" >MCC</th>\n",
       "      <th id=\"T_1e62b_level0_col8\" class=\"col_heading level0 col8\" >TT (Sec)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row0\" class=\"row_heading level0 row0\" >knn</th>\n",
       "      <td id=\"T_1e62b_row0_col0\" class=\"data row0 col0\" >K Neighbors Classifier</td>\n",
       "      <td id=\"T_1e62b_row0_col1\" class=\"data row0 col1\" >0.6015</td>\n",
       "      <td id=\"T_1e62b_row0_col2\" class=\"data row0 col2\" >0.8588</td>\n",
       "      <td id=\"T_1e62b_row0_col3\" class=\"data row0 col3\" >0.6015</td>\n",
       "      <td id=\"T_1e62b_row0_col4\" class=\"data row0 col4\" >0.5970</td>\n",
       "      <td id=\"T_1e62b_row0_col5\" class=\"data row0 col5\" >0.5980</td>\n",
       "      <td id=\"T_1e62b_row0_col6\" class=\"data row0 col6\" >0.5211</td>\n",
       "      <td id=\"T_1e62b_row0_col7\" class=\"data row0 col7\" >0.5216</td>\n",
       "      <td id=\"T_1e62b_row0_col8\" class=\"data row0 col8\" >0.4220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row1\" class=\"row_heading level0 row1\" >gbc</th>\n",
       "      <td id=\"T_1e62b_row1_col0\" class=\"data row1 col0\" >Gradient Boosting Classifier</td>\n",
       "      <td id=\"T_1e62b_row1_col1\" class=\"data row1 col1\" >0.5076</td>\n",
       "      <td id=\"T_1e62b_row1_col2\" class=\"data row1 col2\" >0.8250</td>\n",
       "      <td id=\"T_1e62b_row1_col3\" class=\"data row1 col3\" >0.5076</td>\n",
       "      <td id=\"T_1e62b_row1_col4\" class=\"data row1 col4\" >0.5137</td>\n",
       "      <td id=\"T_1e62b_row1_col5\" class=\"data row1 col5\" >0.5102</td>\n",
       "      <td id=\"T_1e62b_row1_col6\" class=\"data row1 col6\" >0.4074</td>\n",
       "      <td id=\"T_1e62b_row1_col7\" class=\"data row1 col7\" >0.4076</td>\n",
       "      <td id=\"T_1e62b_row1_col8\" class=\"data row1 col8\" >1.5470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row2\" class=\"row_heading level0 row2\" >dt</th>\n",
       "      <td id=\"T_1e62b_row2_col0\" class=\"data row2 col0\" >Decision Tree Classifier</td>\n",
       "      <td id=\"T_1e62b_row2_col1\" class=\"data row2 col1\" >0.4888</td>\n",
       "      <td id=\"T_1e62b_row2_col2\" class=\"data row2 col2\" >0.6909</td>\n",
       "      <td id=\"T_1e62b_row2_col3\" class=\"data row2 col3\" >0.4888</td>\n",
       "      <td id=\"T_1e62b_row2_col4\" class=\"data row2 col4\" >0.4972</td>\n",
       "      <td id=\"T_1e62b_row2_col5\" class=\"data row2 col5\" >0.4913</td>\n",
       "      <td id=\"T_1e62b_row2_col6\" class=\"data row2 col6\" >0.3843</td>\n",
       "      <td id=\"T_1e62b_row2_col7\" class=\"data row2 col7\" >0.3846</td>\n",
       "      <td id=\"T_1e62b_row2_col8\" class=\"data row2 col8\" >0.0280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row3\" class=\"row_heading level0 row3\" >lightgbm</th>\n",
       "      <td id=\"T_1e62b_row3_col0\" class=\"data row3 col0\" >Light Gradient Boosting Machine</td>\n",
       "      <td id=\"T_1e62b_row3_col1\" class=\"data row3 col1\" >0.4654</td>\n",
       "      <td id=\"T_1e62b_row3_col2\" class=\"data row3 col2\" >0.7935</td>\n",
       "      <td id=\"T_1e62b_row3_col3\" class=\"data row3 col3\" >0.4654</td>\n",
       "      <td id=\"T_1e62b_row3_col4\" class=\"data row3 col4\" >0.4820</td>\n",
       "      <td id=\"T_1e62b_row3_col5\" class=\"data row3 col5\" >0.4726</td>\n",
       "      <td id=\"T_1e62b_row3_col6\" class=\"data row3 col6\" >0.3563</td>\n",
       "      <td id=\"T_1e62b_row3_col7\" class=\"data row3 col7\" >0.3566</td>\n",
       "      <td id=\"T_1e62b_row3_col8\" class=\"data row3 col8\" >0.6040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row4\" class=\"row_heading level0 row4\" >rf</th>\n",
       "      <td id=\"T_1e62b_row4_col0\" class=\"data row4 col0\" >Random Forest Classifier</td>\n",
       "      <td id=\"T_1e62b_row4_col1\" class=\"data row4 col1\" >0.4548</td>\n",
       "      <td id=\"T_1e62b_row4_col2\" class=\"data row4 col2\" >0.7970</td>\n",
       "      <td id=\"T_1e62b_row4_col3\" class=\"data row4 col3\" >0.4548</td>\n",
       "      <td id=\"T_1e62b_row4_col4\" class=\"data row4 col4\" >0.4761</td>\n",
       "      <td id=\"T_1e62b_row4_col5\" class=\"data row4 col5\" >0.4637</td>\n",
       "      <td id=\"T_1e62b_row4_col6\" class=\"data row4 col6\" >0.3432</td>\n",
       "      <td id=\"T_1e62b_row4_col7\" class=\"data row4 col7\" >0.3437</td>\n",
       "      <td id=\"T_1e62b_row4_col8\" class=\"data row4 col8\" >0.1920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row5\" class=\"row_heading level0 row5\" >ada</th>\n",
       "      <td id=\"T_1e62b_row5_col0\" class=\"data row5 col0\" >Ada Boost Classifier</td>\n",
       "      <td id=\"T_1e62b_row5_col1\" class=\"data row5 col1\" >0.4521</td>\n",
       "      <td id=\"T_1e62b_row5_col2\" class=\"data row5 col2\" >0.8108</td>\n",
       "      <td id=\"T_1e62b_row5_col3\" class=\"data row5 col3\" >0.4521</td>\n",
       "      <td id=\"T_1e62b_row5_col4\" class=\"data row5 col4\" >0.4880</td>\n",
       "      <td id=\"T_1e62b_row5_col5\" class=\"data row5 col5\" >0.4386</td>\n",
       "      <td id=\"T_1e62b_row5_col6\" class=\"data row5 col6\" >0.3386</td>\n",
       "      <td id=\"T_1e62b_row5_col7\" class=\"data row5 col7\" >0.3465</td>\n",
       "      <td id=\"T_1e62b_row5_col8\" class=\"data row5 col8\" >0.1220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row6\" class=\"row_heading level0 row6\" >qda</th>\n",
       "      <td id=\"T_1e62b_row6_col0\" class=\"data row6 col0\" >Quadratic Discriminant Analysis</td>\n",
       "      <td id=\"T_1e62b_row6_col1\" class=\"data row6 col1\" >0.4487</td>\n",
       "      <td id=\"T_1e62b_row6_col2\" class=\"data row6 col2\" >0.8131</td>\n",
       "      <td id=\"T_1e62b_row6_col3\" class=\"data row6 col3\" >0.4487</td>\n",
       "      <td id=\"T_1e62b_row6_col4\" class=\"data row6 col4\" >0.4604</td>\n",
       "      <td id=\"T_1e62b_row6_col5\" class=\"data row6 col5\" >0.4440</td>\n",
       "      <td id=\"T_1e62b_row6_col6\" class=\"data row6 col6\" >0.3381</td>\n",
       "      <td id=\"T_1e62b_row6_col7\" class=\"data row6 col7\" >0.3407</td>\n",
       "      <td id=\"T_1e62b_row6_col8\" class=\"data row6 col8\" >0.0180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row7\" class=\"row_heading level0 row7\" >et</th>\n",
       "      <td id=\"T_1e62b_row7_col0\" class=\"data row7 col0\" >Extra Trees Classifier</td>\n",
       "      <td id=\"T_1e62b_row7_col1\" class=\"data row7 col1\" >0.4378</td>\n",
       "      <td id=\"T_1e62b_row7_col2\" class=\"data row7 col2\" >0.7907</td>\n",
       "      <td id=\"T_1e62b_row7_col3\" class=\"data row7 col3\" >0.4378</td>\n",
       "      <td id=\"T_1e62b_row7_col4\" class=\"data row7 col4\" >0.4609</td>\n",
       "      <td id=\"T_1e62b_row7_col5\" class=\"data row7 col5\" >0.4477</td>\n",
       "      <td id=\"T_1e62b_row7_col6\" class=\"data row7 col6\" >0.3229</td>\n",
       "      <td id=\"T_1e62b_row7_col7\" class=\"data row7 col7\" >0.3234</td>\n",
       "      <td id=\"T_1e62b_row7_col8\" class=\"data row7 col8\" >0.2300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row8\" class=\"row_heading level0 row8\" >lr</th>\n",
       "      <td id=\"T_1e62b_row8_col0\" class=\"data row8 col0\" >Logistic Regression</td>\n",
       "      <td id=\"T_1e62b_row8_col1\" class=\"data row8 col1\" >0.3641</td>\n",
       "      <td id=\"T_1e62b_row8_col2\" class=\"data row8 col2\" >0.7677</td>\n",
       "      <td id=\"T_1e62b_row8_col3\" class=\"data row8 col3\" >0.3641</td>\n",
       "      <td id=\"T_1e62b_row8_col4\" class=\"data row8 col4\" >0.3766</td>\n",
       "      <td id=\"T_1e62b_row8_col5\" class=\"data row8 col5\" >0.3653</td>\n",
       "      <td id=\"T_1e62b_row8_col6\" class=\"data row8 col6\" >0.2325</td>\n",
       "      <td id=\"T_1e62b_row8_col7\" class=\"data row8 col7\" >0.2337</td>\n",
       "      <td id=\"T_1e62b_row8_col8\" class=\"data row8 col8\" >1.7260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row9\" class=\"row_heading level0 row9\" >svm</th>\n",
       "      <td id=\"T_1e62b_row9_col0\" class=\"data row9 col0\" >SVM - Linear Kernel</td>\n",
       "      <td id=\"T_1e62b_row9_col1\" class=\"data row9 col1\" >0.3368</td>\n",
       "      <td id=\"T_1e62b_row9_col2\" class=\"data row9 col2\" >0.0000</td>\n",
       "      <td id=\"T_1e62b_row9_col3\" class=\"data row9 col3\" >0.3368</td>\n",
       "      <td id=\"T_1e62b_row9_col4\" class=\"data row9 col4\" >0.3200</td>\n",
       "      <td id=\"T_1e62b_row9_col5\" class=\"data row9 col5\" >0.2985</td>\n",
       "      <td id=\"T_1e62b_row9_col6\" class=\"data row9 col6\" >0.1982</td>\n",
       "      <td id=\"T_1e62b_row9_col7\" class=\"data row9 col7\" >0.2068</td>\n",
       "      <td id=\"T_1e62b_row9_col8\" class=\"data row9 col8\" >0.1150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row10\" class=\"row_heading level0 row10\" >nb</th>\n",
       "      <td id=\"T_1e62b_row10_col0\" class=\"data row10 col0\" >Naive Bayes</td>\n",
       "      <td id=\"T_1e62b_row10_col1\" class=\"data row10 col1\" >0.3340</td>\n",
       "      <td id=\"T_1e62b_row10_col2\" class=\"data row10 col2\" >0.6787</td>\n",
       "      <td id=\"T_1e62b_row10_col3\" class=\"data row10 col3\" >0.3340</td>\n",
       "      <td id=\"T_1e62b_row10_col4\" class=\"data row10 col4\" >0.3796</td>\n",
       "      <td id=\"T_1e62b_row10_col5\" class=\"data row10 col5\" >0.3282</td>\n",
       "      <td id=\"T_1e62b_row10_col6\" class=\"data row10 col6\" >0.2011</td>\n",
       "      <td id=\"T_1e62b_row10_col7\" class=\"data row10 col7\" >0.2064</td>\n",
       "      <td id=\"T_1e62b_row10_col8\" class=\"data row10 col8\" >0.0180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row11\" class=\"row_heading level0 row11\" >lda</th>\n",
       "      <td id=\"T_1e62b_row11_col0\" class=\"data row11 col0\" >Linear Discriminant Analysis</td>\n",
       "      <td id=\"T_1e62b_row11_col1\" class=\"data row11 col1\" >0.3071</td>\n",
       "      <td id=\"T_1e62b_row11_col2\" class=\"data row11 col2\" >0.7118</td>\n",
       "      <td id=\"T_1e62b_row11_col3\" class=\"data row11 col3\" >0.3071</td>\n",
       "      <td id=\"T_1e62b_row11_col4\" class=\"data row11 col4\" >0.3542</td>\n",
       "      <td id=\"T_1e62b_row11_col5\" class=\"data row11 col5\" >0.3232</td>\n",
       "      <td id=\"T_1e62b_row11_col6\" class=\"data row11 col6\" >0.1635</td>\n",
       "      <td id=\"T_1e62b_row11_col7\" class=\"data row11 col7\" >0.1650</td>\n",
       "      <td id=\"T_1e62b_row11_col8\" class=\"data row11 col8\" >0.0190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row12\" class=\"row_heading level0 row12\" >ridge</th>\n",
       "      <td id=\"T_1e62b_row12_col0\" class=\"data row12 col0\" >Ridge Classifier</td>\n",
       "      <td id=\"T_1e62b_row12_col1\" class=\"data row12 col1\" >0.2618</td>\n",
       "      <td id=\"T_1e62b_row12_col2\" class=\"data row12 col2\" >0.0000</td>\n",
       "      <td id=\"T_1e62b_row12_col3\" class=\"data row12 col3\" >0.2618</td>\n",
       "      <td id=\"T_1e62b_row12_col4\" class=\"data row12 col4\" >0.2900</td>\n",
       "      <td id=\"T_1e62b_row12_col5\" class=\"data row12 col5\" >0.2209</td>\n",
       "      <td id=\"T_1e62b_row12_col6\" class=\"data row12 col6\" >0.1057</td>\n",
       "      <td id=\"T_1e62b_row12_col7\" class=\"data row12 col7\" >0.1104</td>\n",
       "      <td id=\"T_1e62b_row12_col8\" class=\"data row12 col8\" >0.0190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_1e62b_level0_row13\" class=\"row_heading level0 row13\" >dummy</th>\n",
       "      <td id=\"T_1e62b_row13_col0\" class=\"data row13 col0\" >Dummy Classifier</td>\n",
       "      <td id=\"T_1e62b_row13_col1\" class=\"data row13 col1\" >0.1745</td>\n",
       "      <td id=\"T_1e62b_row13_col2\" class=\"data row13 col2\" >0.5000</td>\n",
       "      <td id=\"T_1e62b_row13_col3\" class=\"data row13 col3\" >0.1745</td>\n",
       "      <td id=\"T_1e62b_row13_col4\" class=\"data row13 col4\" >0.0304</td>\n",
       "      <td id=\"T_1e62b_row13_col5\" class=\"data row13 col5\" >0.0518</td>\n",
       "      <td id=\"T_1e62b_row13_col6\" class=\"data row13 col6\" >0.0000</td>\n",
       "      <td id=\"T_1e62b_row13_col7\" class=\"data row13 col7\" >0.0000</td>\n",
       "      <td id=\"T_1e62b_row13_col8\" class=\"data row13 col8\" >0.0180</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x2c55fe156a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Processing:   0%|          | 0/61 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "best = compare_models()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "90817fdc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x550 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot confusion matrix\n",
    "plot_model(best, plot = 'confusion_matrix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "e7c51c30",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x550 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot AUC\n",
    "plot_model(best, plot = 'auc')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "5bc2305d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_30ff7\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_30ff7_level0_col0\" class=\"col_heading level0 col0\" >Model</th>\n",
       "      <th id=\"T_30ff7_level0_col1\" class=\"col_heading level0 col1\" >Accuracy</th>\n",
       "      <th id=\"T_30ff7_level0_col2\" class=\"col_heading level0 col2\" >AUC</th>\n",
       "      <th id=\"T_30ff7_level0_col3\" class=\"col_heading level0 col3\" >Recall</th>\n",
       "      <th id=\"T_30ff7_level0_col4\" class=\"col_heading level0 col4\" >Prec.</th>\n",
       "      <th id=\"T_30ff7_level0_col5\" class=\"col_heading level0 col5\" >F1</th>\n",
       "      <th id=\"T_30ff7_level0_col6\" class=\"col_heading level0 col6\" >Kappa</th>\n",
       "      <th id=\"T_30ff7_level0_col7\" class=\"col_heading level0 col7\" >MCC</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_30ff7_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_30ff7_row0_col0\" class=\"data row0 col0\" >K Neighbors Classifier</td>\n",
       "      <td id=\"T_30ff7_row0_col1\" class=\"data row0 col1\" >0.5960</td>\n",
       "      <td id=\"T_30ff7_row0_col2\" class=\"data row0 col2\" >0.8587</td>\n",
       "      <td id=\"T_30ff7_row0_col3\" class=\"data row0 col3\" >0.5960</td>\n",
       "      <td id=\"T_30ff7_row0_col4\" class=\"data row0 col4\" >0.5910</td>\n",
       "      <td id=\"T_30ff7_row0_col5\" class=\"data row0 col5\" >0.5923</td>\n",
       "      <td id=\"T_30ff7_row0_col6\" class=\"data row0 col6\" >0.5144</td>\n",
       "      <td id=\"T_30ff7_row0_col7\" class=\"data row0 col7\" >0.5149</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x2c56274eb80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# predict on test set\n",
    "holdout_pred = predict_model(best)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "4a3a83bb",
   "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>time</th>\n",
       "      <th>consumption</th>\n",
       "      <th>temperature</th>\n",
       "      <th>location</th>\n",
       "      <th>prediction_label</th>\n",
       "      <th>prediction_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>31813</th>\n",
       "      <td>2022-12-01 03:00:00</td>\n",
       "      <td>6.473000</td>\n",
       "      <td>1.100000</td>\n",
       "      <td>1</td>\n",
       "      <td>helsingfors</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9556</th>\n",
       "      <td>2022-06-26 12:00:00</td>\n",
       "      <td>2.717178</td>\n",
       "      <td>23.200001</td>\n",
       "      <td>2</td>\n",
       "      <td>oslo</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6316</th>\n",
       "      <td>2022-05-30 12:00:00</td>\n",
       "      <td>2.235322</td>\n",
       "      <td>14.400000</td>\n",
       "      <td>2</td>\n",
       "      <td>oslo</td>\n",
       "      <td>0.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37854</th>\n",
       "      <td>2023-01-12 02:00:00</td>\n",
       "      <td>2.655275</td>\n",
       "      <td>3.800000</td>\n",
       "      <td>0</td>\n",
       "      <td>bergen</td>\n",
       "      <td>0.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40793</th>\n",
       "      <td>2023-02-01 11:00:00</td>\n",
       "      <td>3.741430</td>\n",
       "      <td>-7.500000</td>\n",
       "      <td>5</td>\n",
       "      <td>bergen</td>\n",
       "      <td>0.6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      time  consumption  temperature  location  \\\n",
       "31813  2022-12-01 03:00:00     6.473000     1.100000         1   \n",
       "9556   2022-06-26 12:00:00     2.717178    23.200001         2   \n",
       "6316   2022-05-30 12:00:00     2.235322    14.400000         2   \n",
       "37854  2023-01-12 02:00:00     2.655275     3.800000         0   \n",
       "40793  2023-02-01 11:00:00     3.741430    -7.500000         5   \n",
       "\n",
       "      prediction_label  prediction_score  \n",
       "31813      helsingfors               1.0  \n",
       "9556              oslo               1.0  \n",
       "6316              oslo               0.8  \n",
       "37854           bergen               0.8  \n",
       "40793           bergen               0.6  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# show predictions df\n",
    "holdout_pred.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "191ecc66",
   "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>time</th>\n",
       "      <th>consumption</th>\n",
       "      <th>temperature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>1.113325</td>\n",
       "      <td>-0.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>4.092830</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>2.057858</td>\n",
       "      <td>1.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>1.246582</td>\n",
       "      <td>-3.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>1.970098</td>\n",
       "      <td>-2.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49489</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>6.333000</td>\n",
       "      <td>-1.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49490</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>12.134655</td>\n",
       "      <td>-1.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49491</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>5.622820</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49492</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>2.018333</td>\n",
       "      <td>-2.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49493</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>3.749047</td>\n",
       "      <td>-3.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>49494 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      time  consumption  temperature\n",
       "0      2022-04-07 21:00:00     1.113325         -0.3\n",
       "1      2022-04-07 21:00:00     4.092830          1.0\n",
       "2      2022-04-07 21:00:00     2.057858          1.3\n",
       "3      2022-04-07 21:00:00     1.246582         -3.9\n",
       "4      2022-04-07 21:00:00     1.970098         -2.8\n",
       "...                    ...          ...          ...\n",
       "49489  2023-04-02 21:00:00     6.333000         -1.1\n",
       "49490  2023-04-02 21:00:00    12.134655         -1.1\n",
       "49491  2023-04-02 21:00:00     5.622820          0.5\n",
       "49492  2023-04-02 21:00:00     2.018333         -2.5\n",
       "49493  2023-04-02 21:00:00     3.749047         -3.8\n",
       "\n",
       "[49494 rows x 3 columns]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# copy data and drop Class variable\n",
    "\n",
    "new_data = df.copy()\n",
    "new_data.drop('location', axis=1, inplace=True)\n",
    "new_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "16a447f5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>time</th>\n",
       "      <th>consumption</th>\n",
       "      <th>temperature</th>\n",
       "      <th>prediction_label</th>\n",
       "      <th>prediction_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>1.113325</td>\n",
       "      <td>-0.3</td>\n",
       "      <td>tromsø</td>\n",
       "      <td>0.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>4.092830</td>\n",
       "      <td>1.0</td>\n",
       "      <td>trondheim</td>\n",
       "      <td>0.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>2.057858</td>\n",
       "      <td>1.3</td>\n",
       "      <td>tromsø</td>\n",
       "      <td>0.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>1.246582</td>\n",
       "      <td>-3.9</td>\n",
       "      <td>tromsø</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>1.970098</td>\n",
       "      <td>-2.8</td>\n",
       "      <td>tromsø</td>\n",
       "      <td>0.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49489</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>6.333000</td>\n",
       "      <td>-1.1</td>\n",
       "      <td>helsingfors</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49490</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>12.134655</td>\n",
       "      <td>-1.1</td>\n",
       "      <td>oslo</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49491</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>5.622820</td>\n",
       "      <td>0.5</td>\n",
       "      <td>stavanger</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49492</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>2.018333</td>\n",
       "      <td>-2.5</td>\n",
       "      <td>tromsø</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49493</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>3.749047</td>\n",
       "      <td>-3.8</td>\n",
       "      <td>trondheim</td>\n",
       "      <td>0.6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>49494 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      time  consumption  temperature prediction_label  \\\n",
       "0      2022-04-07 21:00:00     1.113325         -0.3           tromsø   \n",
       "1      2022-04-07 21:00:00     4.092830          1.0        trondheim   \n",
       "2      2022-04-07 21:00:00     2.057858          1.3           tromsø   \n",
       "3      2022-04-07 21:00:00     1.246582         -3.9           tromsø   \n",
       "4      2022-04-07 21:00:00     1.970098         -2.8           tromsø   \n",
       "...                    ...          ...          ...              ...   \n",
       "49489  2023-04-02 21:00:00     6.333000         -1.1      helsingfors   \n",
       "49490  2023-04-02 21:00:00    12.134655         -1.1             oslo   \n",
       "49491  2023-04-02 21:00:00     5.622820          0.5        stavanger   \n",
       "49492  2023-04-02 21:00:00     2.018333         -2.5           tromsø   \n",
       "49493  2023-04-02 21:00:00     3.749047         -3.8        trondheim   \n",
       "\n",
       "       prediction_score  \n",
       "0                   0.8  \n",
       "1                   0.8  \n",
       "2                   0.6  \n",
       "3                   1.0  \n",
       "4                   0.8  \n",
       "...                 ...  \n",
       "49489               1.0  \n",
       "49490               1.0  \n",
       "49491               1.0  \n",
       "49492               1.0  \n",
       "49493               0.6  \n",
       "\n",
       "[49494 rows x 5 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# predict model on new_data\n",
    "predictions = predict_model(best, data = new_data)\n",
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "052a8994",
   "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>time</th>\n",
       "      <th>location</th>\n",
       "      <th>consumption</th>\n",
       "      <th>temperature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>bergen</td>\n",
       "      <td>1.113325</td>\n",
       "      <td>-0.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>oslo</td>\n",
       "      <td>4.092830</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>stavanger</td>\n",
       "      <td>2.057858</td>\n",
       "      <td>1.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>tromsø</td>\n",
       "      <td>1.246582</td>\n",
       "      <td>-3.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-04-07 21:00:00</td>\n",
       "      <td>trondheim</td>\n",
       "      <td>1.970098</td>\n",
       "      <td>-2.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49489</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>helsingfors</td>\n",
       "      <td>6.333000</td>\n",
       "      <td>-1.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49490</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>oslo</td>\n",
       "      <td>12.134655</td>\n",
       "      <td>-1.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49491</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>stavanger</td>\n",
       "      <td>5.622820</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49492</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>tromsø</td>\n",
       "      <td>2.018333</td>\n",
       "      <td>-2.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49493</th>\n",
       "      <td>2023-04-02 21:00:00</td>\n",
       "      <td>trondheim</td>\n",
       "      <td>3.749047</td>\n",
       "      <td>-3.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>49494 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      time     location  consumption  temperature\n",
       "0      2022-04-07 21:00:00       bergen     1.113325         -0.3\n",
       "1      2022-04-07 21:00:00         oslo     4.092830          1.0\n",
       "2      2022-04-07 21:00:00    stavanger     2.057858          1.3\n",
       "3      2022-04-07 21:00:00       tromsø     1.246582         -3.9\n",
       "4      2022-04-07 21:00:00    trondheim     1.970098         -2.8\n",
       "...                    ...          ...          ...          ...\n",
       "49489  2023-04-02 21:00:00  helsingfors     6.333000         -1.1\n",
       "49490  2023-04-02 21:00:00         oslo    12.134655         -1.1\n",
       "49491  2023-04-02 21:00:00    stavanger     5.622820          0.5\n",
       "49492  2023-04-02 21:00:00       tromsø     2.018333         -2.5\n",
       "49493  2023-04-02 21:00:00    trondheim     3.749047         -3.8\n",
       "\n",
       "[49494 rows x 4 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "da8fa6fc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}