3_Feature_Engineering.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "103376a5-74f2-441b-a544-f59dd215a51d",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "# 1 Housekeeping"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "bb099034-7712-4889-9b82-5596b6d996a1",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "### 1.1 Load libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "9b54d8dc-12aa-443a-b449-d2730ee0cfdf",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "54f78081-0d46-4472-a4b2-dbb435089856",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "### 1.2 Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "c3bd799d-5ebf-45a9-b8d2-49b27a440654",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#set directory\n",
    "train_directory = \"C:/Users/Meyke/WiDS/data/train.csv\"\n",
    "test_directory = \"C:/Users/Meyke/WiDS/data/test.csv\"\n",
    "imputed_data_directory = \"C:/Users/Meyke/WiDS/data/clean_data_2801_imputed_star_rating_and_generous_building_id.csv\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "30a9d46f-e071-4497-b4c0-96c778a180ed",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#import data\n",
    "raw_train = pd.read_csv(train_directory)\n",
    "raw_test = pd.read_csv (test_directory)\n",
    "\n",
    "#version 14/1/22 - imputed dataset\n",
    "df = pd.read_csv(imputed_data_directory)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Unnamed: 0', 'year_factor', 'state_factor', 'building_class',\n",
       "       'facility_type', 'floor_area', 'elevation', 'january_min_temp',\n",
       "       'january_avg_temp', 'january_max_temp', 'february_min_temp',\n",
       "       'february_avg_temp', 'february_max_temp', 'march_min_temp',\n",
       "       'march_avg_temp', 'march_max_temp', 'april_min_temp', 'april_avg_temp',\n",
       "       'april_max_temp', 'may_min_temp', 'may_avg_temp', 'may_max_temp',\n",
       "       'june_min_temp', 'june_avg_temp', 'june_max_temp', 'july_min_temp',\n",
       "       'july_avg_temp', 'july_max_temp', 'august_min_temp', 'august_avg_temp',\n",
       "       'august_max_temp', 'september_min_temp', 'september_avg_temp',\n",
       "       'september_max_temp', 'october_min_temp', 'october_avg_temp',\n",
       "       'october_max_temp', 'november_min_temp', 'november_avg_temp',\n",
       "       'november_max_temp', 'december_min_temp', 'december_avg_temp',\n",
       "       'december_max_temp', 'cooling_degree_days', 'heating_degree_days',\n",
       "       'precipitation_inches', 'snowfall_inches', 'snowdepth_inches',\n",
       "       'avg_temp', 'days_below_30F', 'days_below_20F', 'days_below_10F',\n",
       "       'days_below_0F', 'days_above_80F', 'days_above_90F', 'days_above_100F',\n",
       "       'days_above_110F', 'max_wind_speed', 'days_with_fog', 'site_eui', 'id',\n",
       "       'building_id', 'energy_star_rating', 'peak_NE', 'peak_SE', 'peak_SW',\n",
       "       'peak_NW', 'max_NE', 'max_SE', 'max_SW', 'max_NW', 'facility_group',\n",
       "       'year_built'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "d58112eb-1de0-4145-b296-bf3c0c0e7905",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "**Drop, rename and set data types**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "0cfad86a-c1e6-4823-b2cb-4ca8620606af",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>peak_NE</th>\n",
       "      <th>peak_SE</th>\n",
       "      <th>peak_SW</th>\n",
       "      <th>peak_NW</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1955.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Retail_Enclosed_mall</td>\n",
       "      <td>280025.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>retail</td>\n",
       "      <td>1951.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Education_Other_classroom</td>\n",
       "      <td>55325.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>eduation</td>\n",
       "      <td>1980.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Nonrefrigerated</td>\n",
       "      <td>66000.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1985.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 72 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "2            1      State_1     Commercial   \n",
       "3            1      State_1     Commercial   \n",
       "4            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "2                       Retail_Enclosed_mall    280025.0        1.8   \n",
       "3                  Education_Other_classroom     55325.0        1.8   \n",
       "4                  Warehouse_Nonrefrigerated     66000.0        2.4   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "2                36              50.5                68                 35   \n",
       "3                36              50.5                68                 35   \n",
       "4                36              50.5                68                 35   \n",
       "\n",
       "   ...  peak_NE  peak_SE  peak_SW  peak_NW  max_NE  max_SE  max_SW  max_NW  \\\n",
       "0  ...        1        0        0        0       1       0       0       0   \n",
       "1  ...        0        0        0        0       1       0       0       0   \n",
       "2  ...        0        0        0        0       1       0       0       0   \n",
       "3  ...        0        0        0        0       1       0       0       0   \n",
       "4  ...        1        0        0        0       1       0       0       0   \n",
       "\n",
       "          facility_group  year_built  \n",
       "0     food_sales_service      1942.0  \n",
       "1  distribution_shipping      1955.0  \n",
       "2                 retail      1951.0  \n",
       "3               eduation      1980.0  \n",
       "4  distribution_shipping      1985.0  \n",
       "\n",
       "[5 rows x 72 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#drop redundant columns\n",
    "df = df.drop(['Unnamed: 0'], axis=1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>peak_NE</th>\n",
       "      <th>peak_SE</th>\n",
       "      <th>peak_SW</th>\n",
       "      <th>peak_NW</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Retail_Enclosed_mall</td>\n",
       "      <td>280025.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>retail</td>\n",
       "      <td>1951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Education_Other_classroom</td>\n",
       "      <td>55325.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>eduation</td>\n",
       "      <td>1980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Nonrefrigerated</td>\n",
       "      <td>66000.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1985</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 72 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "2            1      State_1     Commercial   \n",
       "3            1      State_1     Commercial   \n",
       "4            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "2                       Retail_Enclosed_mall    280025.0        1.8   \n",
       "3                  Education_Other_classroom     55325.0        1.8   \n",
       "4                  Warehouse_Nonrefrigerated     66000.0        2.4   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "2                36              50.5                68                 35   \n",
       "3                36              50.5                68                 35   \n",
       "4                36              50.5                68                 35   \n",
       "\n",
       "   ...  peak_NE  peak_SE  peak_SW  peak_NW  max_NE  max_SE  max_SW  max_NW  \\\n",
       "0  ...        1        0        0        0       1       0       0       0   \n",
       "1  ...        0        0        0        0       1       0       0       0   \n",
       "2  ...        0        0        0        0       1       0       0       0   \n",
       "3  ...        0        0        0        0       1       0       0       0   \n",
       "4  ...        1        0        0        0       1       0       0       0   \n",
       "\n",
       "          facility_group  year_built  \n",
       "0     food_sales_service        1942  \n",
       "1  distribution_shipping        1955  \n",
       "2                 retail        1951  \n",
       "3               eduation        1980  \n",
       "4  distribution_shipping        1985  \n",
       "\n",
       "[5 rows x 72 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['year_built'] = df['year_built'].astype(int)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "a6807151-134a-43d3-b7ae-4e23f8459693",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "# 2 Feature engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "4f37163b-4498-4ebf-be00-4471870900af",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "### 2.1 Floor area binned"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "48d85b32-c5d6-4370-91b5-a207f359c319",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "20012022_Submission_8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "b0d57aa0-29a2-4e5a-b49d-a0443f47cdd2",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>peak_SE</th>\n",
       "      <th>peak_SW</th>\n",
       "      <th>peak_NW</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1955</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Retail_Enclosed_mall</td>\n",
       "      <td>280025.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>retail</td>\n",
       "      <td>1951</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Education_Other_classroom</td>\n",
       "      <td>55325.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>eduation</td>\n",
       "      <td>1980</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Nonrefrigerated</td>\n",
       "      <td>66000.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1985</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 73 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "2            1      State_1     Commercial   \n",
       "3            1      State_1     Commercial   \n",
       "4            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "2                       Retail_Enclosed_mall    280025.0        1.8   \n",
       "3                  Education_Other_classroom     55325.0        1.8   \n",
       "4                  Warehouse_Nonrefrigerated     66000.0        2.4   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "2                36              50.5                68                 35   \n",
       "3                36              50.5                68                 35   \n",
       "4                36              50.5                68                 35   \n",
       "\n",
       "   ...  peak_SE  peak_SW  peak_NW  max_NE  max_SE  max_SW  max_NW  \\\n",
       "0  ...        0        0        0       1       0       0       0   \n",
       "1  ...        0        0        0       1       0       0       0   \n",
       "2  ...        0        0        0       1       0       0       0   \n",
       "3  ...        0        0        0       1       0       0       0   \n",
       "4  ...        0        0        0       1       0       0       0   \n",
       "\n",
       "          facility_group  year_built  floor_area_binned  \n",
       "0     food_sales_service        1942                  1  \n",
       "1  distribution_shipping        1955                  4  \n",
       "2                 retail        1951                  4  \n",
       "3               eduation        1980                  1  \n",
       "4  distribution_shipping        1985                  1  \n",
       "\n",
       "[5 rows x 73 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#create bins for floor area\n",
    "cut_labels = [1, 2, 3, 4, 5]\n",
    "cut_bins = [0,  75000, 100000, 200000, 500000, 1000000000]\n",
    "df['floor_area_binned'] = pd.cut(df['floor_area'], bins=cut_bins, labels=cut_labels)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>peak_SE</th>\n",
       "      <th>peak_SW</th>\n",
       "      <th>peak_NW</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>0 rows × 73 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [year_factor, state_factor, building_class, facility_type, floor_area, elevation, january_min_temp, january_avg_temp, january_max_temp, february_min_temp, february_avg_temp, february_max_temp, march_min_temp, march_avg_temp, march_max_temp, april_min_temp, april_avg_temp, april_max_temp, may_min_temp, may_avg_temp, may_max_temp, june_min_temp, june_avg_temp, june_max_temp, july_min_temp, july_avg_temp, july_max_temp, august_min_temp, august_avg_temp, august_max_temp, september_min_temp, september_avg_temp, september_max_temp, october_min_temp, october_avg_temp, october_max_temp, november_min_temp, november_avg_temp, november_max_temp, december_min_temp, december_avg_temp, december_max_temp, cooling_degree_days, heating_degree_days, precipitation_inches, snowfall_inches, snowdepth_inches, avg_temp, days_below_30F, days_below_20F, days_below_10F, days_below_0F, days_above_80F, days_above_90F, days_above_100F, days_above_110F, max_wind_speed, days_with_fog, site_eui, id, building_id, energy_star_rating, peak_NE, peak_SE, peak_SW, peak_NW, max_NE, max_SE, max_SW, max_NW, facility_group, year_built, floor_area_binned]\n",
       "Index: []\n",
       "\n",
       "[0 rows x 73 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['floor_area_binned'].isna()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['floor_area_binned'] = df['floor_area_binned'].fillna(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "98805785-4388-4a6d-b15d-6735ca5f4c34",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "### 2.2 Energy intensity floor area"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "61638294-c1b9-4671-830f-1c09c6e418aa",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "20012022_Submission_4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "97135b0d-1363-4db2-b31f-1373fd31fd4a",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#split into (new) train and test set based on id\n",
    "df_train = df[df[\"id\"].isin(raw_train[\"id\"])]\n",
    "df_test = df[df[\"id\"].isin(raw_test[\"id\"])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "ee1f675c-2625-47fc-937b-0e4620c39225",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#calculate q1\n",
    "q1 = df_train[['floor_area_binned', 'site_eui']]\n",
    "q1= q1.groupby('floor_area_binned')['site_eui'].quantile(0.25).reset_index()\n",
    "q1  = q1.rename(columns ={'site_eui': 'q1'})\n",
    "\n",
    "#calculate q2\n",
    "q3 = df_train[['floor_area_binned', 'site_eui']]\n",
    "q3= q3.groupby('floor_area_binned')['site_eui'].quantile(0.75).reset_index()\n",
    "q3  = q3.rename(columns ={'site_eui': 'q3'})\n",
    "\n",
    "#calcualte 'normal' range\n",
    "ranges = q1.merge(q3, how = 'left', on = 'floor_area_binned')\n",
    "ranges['iqr'] = ranges['q3'] - ranges ['q1']\n",
    "ranges['left_border'] = ranges['q1'] - 0.1 * ranges['iqr']\n",
    "ranges ['right_border'] = ranges['q3'] + 1.1* ranges['iqr']\n",
    "\n",
    "#merge\n",
    "df_train = df_train.merge(ranges[['floor_area_binned', 'left_border', 'right_border']], how = 'left', on = 'floor_area_binned')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "b22179a5-b0e4-4110-b1ff-d7598635ee24",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#calculate energy intensity in train set\n",
    "df_train['energy_intensity_floor_area'] = np.where( df_train['site_eui'] < df_train['left_border'], 1, 2)\n",
    "df_train['energy_intensity_floor_area'] = np.where (df_train['site_eui']> df_train['right_border'], 3, df_train['energy_intensity_floor_area'])\n",
    "\n",
    "#drop redundant\n",
    "df_train = df_train.drop(['left_border', 'right_border'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "8dd01004-fce5-480b-a031-429f0a6d5f52",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#groupby building id and keep for each building the last record (based on year_factor)\n",
    "building_id = df_train [['building_id', 'year_factor', 'energy_intensity_floor_area']]\n",
    "building_id = building_id.sort_values(['building_id', 'year_factor'])\n",
    "building_id = building_id.groupby('building_id')['energy_intensity_floor_area'].last().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "ac28b57e-de06-4e89-a228-a19442ad7936",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>peak_SW</th>\n",
       "      <th>peak_NW</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "      <th>energy_intensity_floor_area</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>28484.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>21906.0</td>\n",
       "      <td>45.7</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1961</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>16138.0</td>\n",
       "      <td>59.1</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1950</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>97422.0</td>\n",
       "      <td>35.4</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1971</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 73 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class                 facility_type  \\\n",
       "0            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "1            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "2            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "3            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "4            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "\n",
       "   floor_area  elevation  january_min_temp  january_avg_temp  \\\n",
       "0     28484.0        2.4                38         50.596774   \n",
       "1     21906.0       45.7                38         50.596774   \n",
       "2     16138.0       59.1                38         50.596774   \n",
       "3     97422.0       35.4                38         50.596774   \n",
       "4     61242.0        1.8                38         50.596774   \n",
       "\n",
       "   january_max_temp  february_min_temp  ...  peak_SW  peak_NW  max_NE  max_SE  \\\n",
       "0                64                 40  ...        0        0       0       0   \n",
       "1                64                 40  ...        0        0       0       0   \n",
       "2                64                 40  ...        0        0       0       0   \n",
       "3                64                 40  ...        0        0       0       0   \n",
       "4                64                 40  ...        0        1       0       0   \n",
       "\n",
       "   max_SW  max_NW      facility_group  year_built  floor_area_binned  \\\n",
       "0       0       0  food_sales_service        1994                  1   \n",
       "1       0       0  food_sales_service        1961                  1   \n",
       "2       0       0  food_sales_service        1950                  1   \n",
       "3       0       0  food_sales_service        1971                  2   \n",
       "4       0       1  food_sales_service        1942                  1   \n",
       "\n",
       "   energy_intensity_floor_area  \n",
       "0                            3  \n",
       "1                            3  \n",
       "2                            2  \n",
       "3                            3  \n",
       "4                            3  \n",
       "\n",
       "[5 rows x 73 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#merge based on building id, and fill missing recors (around 700) with 2\n",
    "df_test  = df_test.drop(['site_eui'],axis=1)\n",
    "df_test = df_test.merge(building_id, how = 'left', on = 'building_id').fillna(2)\n",
    "df_test['energy_intensity_floor_area'] = df_test['energy_intensity_floor_area'].astype('int')\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "3d33421f-582b-49dc-a2b3-8dec428a5ffc",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>peak_SW</th>\n",
       "      <th>peak_NW</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "      <th>energy_intensity_floor_area</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1955</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 74 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "\n",
       "   ...  peak_SW  peak_NW  max_NE  max_SE  max_SW  max_NW  \\\n",
       "0  ...        0        0       1       0       0       0   \n",
       "1  ...        0        0       1       0       0       0   \n",
       "\n",
       "          facility_group  year_built  floor_area_binned  \\\n",
       "0     food_sales_service        1942                  1   \n",
       "1  distribution_shipping        1955                  4   \n",
       "\n",
       "   energy_intensity_floor_area  \n",
       "0                            3  \n",
       "1                            1  \n",
       "\n",
       "[2 rows x 74 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#append train and test\n",
    "df = df_train.append(df_test)\n",
    "df.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "d9b843d0-b275-4852-a239-71e29221587f",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "### 2.3 Energy intenstiy built year"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "20a7df78-e512-4ffd-8c64-63773a526e4b",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "20012022_Submission_11"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "68c6ea98-8d21-4d3a-aaf1-69651efd39f2",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#create built year bins\n",
    "cut_labels = [1, 2, 3, 4, 5]\n",
    "cut_bins = [0,  1930, 1950, 1970, 1990, 2022]\n",
    "df['year_built_binned'] = pd.cut(df['year_built'], bins=cut_bins, labels=cut_labels)\n",
    "df['year_built_binned'] = df['year_built_binned'].astype('int')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "db5a948e-a78c-4b45-8656-2fe2aa252166",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#split into (new) train and test set based on id\n",
    "df_train = df[df[\"id\"].isin(raw_train[\"id\"])]\n",
    "df_test = df[df[\"id\"].isin(raw_test[\"id\"])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "e623dbe6-2ee1-4c04-adde-37683e37b206",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#set iqr to 1.2\n",
    "q1 = df_train[['year_built_binned', 'site_eui']]\n",
    "q1= q1.groupby('year_built_binned')['site_eui'].quantile(0.25).reset_index()\n",
    "q1  = q1.rename(columns ={'site_eui': 'q1'})\n",
    "\n",
    "q3 = df_train[['year_built_binned', 'site_eui']]\n",
    "q3= q3.groupby('year_built_binned')['site_eui'].quantile(0.75).reset_index()\n",
    "q3  = q3.rename(columns ={'site_eui': 'q3'})\n",
    "\n",
    "ranges = q1.merge(q3, how = 'left', on = 'year_built_binned')\n",
    "ranges['iqr'] = ranges['q3'] - ranges ['q1']\n",
    "\n",
    "#first borders:  1qr =1.2\n",
    "ranges['left_border'] = ranges['q1'] - 0.1 * ranges['iqr']\n",
    "ranges ['right_border'] = ranges['q3'] + 1.5 * ranges['iqr']\n",
    "\n",
    "#merge\n",
    "df_train = df_train.merge(ranges[['year_built_binned', 'left_border', 'right_border']], how = 'left', on = 'year_built_binned')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "b99ed483-2ab1-42b9-a671-47638c8281a1",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>peak_NW</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "      <th>energy_intensity_floor_area</th>\n",
       "      <th>energy_intensity_built_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1955</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Retail_Enclosed_mall</td>\n",
       "      <td>280025.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>retail</td>\n",
       "      <td>1951</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Education_Other_classroom</td>\n",
       "      <td>55325.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>eduation</td>\n",
       "      <td>1980</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Nonrefrigerated</td>\n",
       "      <td>66000.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1985</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 75 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "2            1      State_1     Commercial   \n",
       "3            1      State_1     Commercial   \n",
       "4            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "2                       Retail_Enclosed_mall    280025.0        1.8   \n",
       "3                  Education_Other_classroom     55325.0        1.8   \n",
       "4                  Warehouse_Nonrefrigerated     66000.0        2.4   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "2                36              50.5                68                 35   \n",
       "3                36              50.5                68                 35   \n",
       "4                36              50.5                68                 35   \n",
       "\n",
       "   ...  peak_NW  max_NE  max_SE  max_SW  max_NW         facility_group  \\\n",
       "0  ...        0       1       0       0       0     food_sales_service   \n",
       "1  ...        0       1       0       0       0  distribution_shipping   \n",
       "2  ...        0       1       0       0       0                 retail   \n",
       "3  ...        0       1       0       0       0               eduation   \n",
       "4  ...        0       1       0       0       0  distribution_shipping   \n",
       "\n",
       "   year_built  floor_area_binned  energy_intensity_floor_area  \\\n",
       "0        1942                  1                            3   \n",
       "1        1955                  4                            1   \n",
       "2        1951                  4                            1   \n",
       "3        1980                  1                            2   \n",
       "4        1985                  1                            1   \n",
       "\n",
       "   energy_intensity_built_year  \n",
       "0                            3  \n",
       "1                            1  \n",
       "2                            1  \n",
       "3                            2  \n",
       "4                            1  \n",
       "\n",
       "[5 rows x 75 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#calculate energy intenstiy built year in train set\n",
    "df_train['energy_intensity_built_year'] = np.where( df_train['site_eui'] < df_train['left_border'], 1, 2)\n",
    "df_train['energy_intensity_built_year'] = np.where (df_train['site_eui']> df_train['right_border'], 3, df_train['energy_intensity_built_year'])\n",
    "\n",
    "#drop redundant\n",
    "df_train = df_train.drop(['year_built_binned', 'left_border', 'right_border'], axis=1)\n",
    "\n",
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "8063c704-75b3-4e70-9395-53c34f3dd54b",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#groupby building id and keep for each building the last record (based on year_factor)\n",
    "building_id = df_train [['building_id', 'year_factor', 'energy_intensity_built_year']]\n",
    "building_id = building_id.sort_values(['building_id', 'year_factor'])\n",
    "building_id = building_id.groupby('building_id')['energy_intensity_built_year'].last().reset_index()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "348102a9-50b0-440f-b440-32b1e40e3c52",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>peak_NW</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "      <th>energy_intensity_floor_area</th>\n",
       "      <th>energy_intensity_built_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>28484.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>21906.0</td>\n",
       "      <td>45.7</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1961</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>16138.0</td>\n",
       "      <td>59.1</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1950</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>97422.0</td>\n",
       "      <td>35.4</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1971</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 74 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class                 facility_type  \\\n",
       "0            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "1            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "2            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "3            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "4            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "\n",
       "   floor_area  elevation  january_min_temp  january_avg_temp  \\\n",
       "0     28484.0        2.4                38         50.596774   \n",
       "1     21906.0       45.7                38         50.596774   \n",
       "2     16138.0       59.1                38         50.596774   \n",
       "3     97422.0       35.4                38         50.596774   \n",
       "4     61242.0        1.8                38         50.596774   \n",
       "\n",
       "   january_max_temp  february_min_temp  ...  peak_NW  max_NE  max_SE  max_SW  \\\n",
       "0                64                 40  ...        0       0       0       0   \n",
       "1                64                 40  ...        0       0       0       0   \n",
       "2                64                 40  ...        0       0       0       0   \n",
       "3                64                 40  ...        0       0       0       0   \n",
       "4                64                 40  ...        1       0       0       0   \n",
       "\n",
       "   max_NW      facility_group  year_built  floor_area_binned  \\\n",
       "0       0  food_sales_service        1994                  1   \n",
       "1       0  food_sales_service        1961                  1   \n",
       "2       0  food_sales_service        1950                  1   \n",
       "3       0  food_sales_service        1971                  2   \n",
       "4       1  food_sales_service        1942                  1   \n",
       "\n",
       "   energy_intensity_floor_area  energy_intensity_built_year  \n",
       "0                            3                            2  \n",
       "1                            3                            3  \n",
       "2                            2                            2  \n",
       "3                            3                            3  \n",
       "4                            3                            3  \n",
       "\n",
       "[5 rows x 74 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#merge based on building id, and fill missing recors (around 700) with 2\n",
    "df_test = df_test.drop('site_eui', axis=1)\n",
    "df_test = df_test.merge(building_id, how = 'left', on = 'building_id').fillna(2)\n",
    "df_test['energy_intensity_built_year'] = df_test['energy_intensity_built_year'].astype('int')\n",
    "\n",
    "#drop redundant\n",
    "df_test = df_test.drop(['year_built_binned'],axis=1)\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "334425ea-daa1-44c4-ae04-acd0a7146b0e",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>peak_NW</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "      <th>energy_intensity_floor_area</th>\n",
       "      <th>energy_intensity_built_year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1955</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 75 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "\n",
       "   ...  peak_NW  max_NE  max_SE  max_SW  max_NW         facility_group  \\\n",
       "0  ...        0       1       0       0       0     food_sales_service   \n",
       "1  ...        0       1       0       0       0  distribution_shipping   \n",
       "\n",
       "   year_built  floor_area_binned  energy_intensity_floor_area  \\\n",
       "0        1942                  1                            3   \n",
       "1        1955                  4                            1   \n",
       "\n",
       "   energy_intensity_built_year  \n",
       "0                            3  \n",
       "1                            1  \n",
       "\n",
       "[2 rows x 75 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#append train and test\n",
    "df = df_train.append(df_test)\n",
    "df.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "d090751c-328f-467b-98fe-70a369eba806",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "### 2.4 Energy intenstiy facility type"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "0a57356a-6240-4113-8c50-3a848e8736c8",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "22012022-Submission_3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "e28477cc-9570-4b5e-9dd2-95641caa3372",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#split into (new) train and test set based on id\n",
    "df_train = df[df[\"id\"].isin(raw_train[\"id\"])]\n",
    "df_test = df[df[\"id\"].isin(raw_test[\"id\"])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "b0833c50-1ac2-41f0-ae3c-3e3d3d075403",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#calculate q1\n",
    "q1 = df_train[['facility_type', 'site_eui']]\n",
    "q1= q1.groupby('facility_type')['site_eui'].quantile(0.25).reset_index()\n",
    "q1  = q1.rename(columns ={'site_eui': 'q1'})\n",
    "\n",
    "#calculate q2\n",
    "q3 = df_train[['facility_type', 'site_eui']]\n",
    "q3= q3.groupby('facility_type')['site_eui'].quantile(0.75).reset_index()\n",
    "q3  = q3.rename(columns ={'site_eui': 'q3'})\n",
    "\n",
    "#calcualte 'normal' range\n",
    "ranges = q1.merge(q3, how = 'left', on = 'facility_type')\n",
    "ranges['iqr'] = ranges['q3'] - ranges ['q1']\n",
    "ranges['left_border'] = ranges['q1'] - 0.1 * ranges['iqr']\n",
    "ranges ['right_border'] = ranges['q3'] + 1.1* ranges['iqr']\n",
    "\n",
    "#merge\n",
    "df_train = df_train.merge(ranges[['facility_type', 'left_border', 'right_border']], how = 'left', on = 'facility_type')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "d982062c-ad80-4094-8ab5-aea31b75cf3c",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "      <th>energy_intensity_floor_area</th>\n",
       "      <th>energy_intensity_built_year</th>\n",
       "      <th>energy_intensity_facility_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1955</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Retail_Enclosed_mall</td>\n",
       "      <td>280025.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>retail</td>\n",
       "      <td>1951</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Education_Other_classroom</td>\n",
       "      <td>55325.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>eduation</td>\n",
       "      <td>1980</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Nonrefrigerated</td>\n",
       "      <td>66000.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1985</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 76 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "2            1      State_1     Commercial   \n",
       "3            1      State_1     Commercial   \n",
       "4            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "2                       Retail_Enclosed_mall    280025.0        1.8   \n",
       "3                  Education_Other_classroom     55325.0        1.8   \n",
       "4                  Warehouse_Nonrefrigerated     66000.0        2.4   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "2                36              50.5                68                 35   \n",
       "3                36              50.5                68                 35   \n",
       "4                36              50.5                68                 35   \n",
       "\n",
       "   ...  max_NE  max_SE  max_SW  max_NW         facility_group  year_built  \\\n",
       "0  ...       1       0       0       0     food_sales_service        1942   \n",
       "1  ...       1       0       0       0  distribution_shipping        1955   \n",
       "2  ...       1       0       0       0                 retail        1951   \n",
       "3  ...       1       0       0       0               eduation        1980   \n",
       "4  ...       1       0       0       0  distribution_shipping        1985   \n",
       "\n",
       "   floor_area_binned  energy_intensity_floor_area  \\\n",
       "0                  1                            3   \n",
       "1                  4                            1   \n",
       "2                  4                            1   \n",
       "3                  1                            2   \n",
       "4                  1                            1   \n",
       "\n",
       "   energy_intensity_built_year  energy_intensity_facility_type  \n",
       "0                            3                               2  \n",
       "1                            1                               2  \n",
       "2                            1                               2  \n",
       "3                            2                               1  \n",
       "4                            1                               1  \n",
       "\n",
       "[5 rows x 76 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#calculate energy intensity in train set\n",
    "df_train['energy_intensity_facility_type'] = np.where( df_train['site_eui'] < df_train['left_border'], 1, 2)\n",
    "df_train['energy_intensity_facility_type'] = np.where (df_train['site_eui']> df_train['right_border'], 3, df_train['energy_intensity_facility_type'])\n",
    "\n",
    "#drop redundant\n",
    "df_train = df_train.drop(['left_border', 'right_border'],  axis=1)\n",
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "ede6bbe8-ebd3-4e21-9c65-584a9a310adf",
     "showTitle": false,
     "title": ""
    }
   },
   "outputs": [],
   "source": [
    "#groupby building id and keep for each building the last record (based on year_factor)\n",
    "building_id = df_train [['building_id', 'year_factor', 'energy_intensity_facility_type']]\n",
    "building_id = building_id.sort_values(['building_id', 'year_factor'])\n",
    "building_id = building_id.groupby('building_id')['energy_intensity_facility_type'].last().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "30ba53ea-a0ea-4e0b-be95-6651da652400",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "      <th>energy_intensity_floor_area</th>\n",
       "      <th>energy_intensity_built_year</th>\n",
       "      <th>energy_intensity_facility_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>28484.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>21906.0</td>\n",
       "      <td>45.7</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1961</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>16138.0</td>\n",
       "      <td>59.1</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1950</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>97422.0</td>\n",
       "      <td>35.4</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1971</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>38</td>\n",
       "      <td>50.596774</td>\n",
       "      <td>64</td>\n",
       "      <td>40</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 75 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class                 facility_type  \\\n",
       "0            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "1            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "2            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "3            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "4            7      State_1     Commercial  Grocery_store_or_food_market   \n",
       "\n",
       "   floor_area  elevation  january_min_temp  january_avg_temp  \\\n",
       "0     28484.0        2.4                38         50.596774   \n",
       "1     21906.0       45.7                38         50.596774   \n",
       "2     16138.0       59.1                38         50.596774   \n",
       "3     97422.0       35.4                38         50.596774   \n",
       "4     61242.0        1.8                38         50.596774   \n",
       "\n",
       "   january_max_temp  february_min_temp  ...  max_NE  max_SE  max_SW  max_NW  \\\n",
       "0                64                 40  ...       0       0       0       0   \n",
       "1                64                 40  ...       0       0       0       0   \n",
       "2                64                 40  ...       0       0       0       0   \n",
       "3                64                 40  ...       0       0       0       0   \n",
       "4                64                 40  ...       0       0       0       1   \n",
       "\n",
       "       facility_group  year_built  floor_area_binned  \\\n",
       "0  food_sales_service        1994                  1   \n",
       "1  food_sales_service        1961                  1   \n",
       "2  food_sales_service        1950                  1   \n",
       "3  food_sales_service        1971                  2   \n",
       "4  food_sales_service        1942                  1   \n",
       "\n",
       "   energy_intensity_floor_area  energy_intensity_built_year  \\\n",
       "0                            3                            2   \n",
       "1                            3                            3   \n",
       "2                            2                            2   \n",
       "3                            3                            3   \n",
       "4                            3                            3   \n",
       "\n",
       "   energy_intensity_facility_type  \n",
       "0                               1  \n",
       "1                               2  \n",
       "2                               1  \n",
       "3                               2  \n",
       "4                               2  \n",
       "\n",
       "[5 rows x 75 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#merge based on building id, and fill missing recors (around 700) with 2\n",
    "df_test = df_test.drop('site_eui', axis=1)\n",
    "df_test = df_test.merge(building_id, how = 'left', on = 'building_id').fillna(2)\n",
    "df_test['energy_intensity_facility_type'] = df_test['energy_intensity_facility_type'].astype ('int')\n",
    "\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "d15afbfd-bf44-4700-b293-0da00d2d6536",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>max_NE</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "      <th>energy_intensity_floor_area</th>\n",
       "      <th>energy_intensity_built_year</th>\n",
       "      <th>energy_intensity_facility_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1955</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 76 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "\n",
       "   ...  max_NE  max_SE  max_SW  max_NW         facility_group  year_built  \\\n",
       "0  ...       1       0       0       0     food_sales_service        1942   \n",
       "1  ...       1       0       0       0  distribution_shipping        1955   \n",
       "\n",
       "   floor_area_binned  energy_intensity_floor_area  \\\n",
       "0                  1                            3   \n",
       "1                  4                            1   \n",
       "\n",
       "   energy_intensity_built_year  energy_intensity_facility_type  \n",
       "0                            3                               2  \n",
       "1                            1                               2  \n",
       "\n",
       "[2 rows x 76 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#append train and test\n",
    "df = df_train.append(df_test)\n",
    "df.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "10c37899-72f0-4f8c-8fcd-7475c89ed7bf",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "### 2.5 Mean energy rating"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "5d019305-6f18-4711-b6d7-696c9dc4d1c9",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "22012022-Submission_8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "604cd767-ab47-49fe-8a4b-5221635a8760",
     "showTitle": false,
     "title": ""
    }
   },
   "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>mean_energy_rating_building_id</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>building_id</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>56.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>56.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>22.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>32.3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             mean_energy_rating_building_id\n",
       "building_id                                \n",
       "0                                      56.4\n",
       "1                                      57.0\n",
       "2                                      56.4\n",
       "3                                      22.8\n",
       "4                                      32.3"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "agg = df.groupby('building_id')['energy_star_rating'].agg({'mean'}).round(1)\n",
    "agg = agg.rename(columns={'mean': 'mean_energy_rating_building_id'})\n",
    "agg.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "72919b82-26ee-4ac5-a43d-6f499156098e",
     "showTitle": false,
     "title": ""
    }
   },
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>max_SE</th>\n",
       "      <th>max_SW</th>\n",
       "      <th>max_NW</th>\n",
       "      <th>facility_group</th>\n",
       "      <th>year_built</th>\n",
       "      <th>floor_area_binned</th>\n",
       "      <th>energy_intensity_floor_area</th>\n",
       "      <th>energy_intensity_built_year</th>\n",
       "      <th>energy_intensity_facility_type</th>\n",
       "      <th>mean_energy_rating_building_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>food_sales_service</td>\n",
       "      <td>1942</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>29.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1955</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>40.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Retail_Enclosed_mall</td>\n",
       "      <td>280025.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>retail</td>\n",
       "      <td>1951</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>97.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Education_Other_classroom</td>\n",
       "      <td>55325.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>eduation</td>\n",
       "      <td>1980</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>49.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Nonrefrigerated</td>\n",
       "      <td>66000.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>distribution_shipping</td>\n",
       "      <td>1985</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 77 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "2            1      State_1     Commercial   \n",
       "3            1      State_1     Commercial   \n",
       "4            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "2                       Retail_Enclosed_mall    280025.0        1.8   \n",
       "3                  Education_Other_classroom     55325.0        1.8   \n",
       "4                  Warehouse_Nonrefrigerated     66000.0        2.4   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "2                36              50.5                68                 35   \n",
       "3                36              50.5                68                 35   \n",
       "4                36              50.5                68                 35   \n",
       "\n",
       "   ...  max_SE  max_SW  max_NW         facility_group  year_built  \\\n",
       "0  ...       0       0       0     food_sales_service        1942   \n",
       "1  ...       0       0       0  distribution_shipping        1955   \n",
       "2  ...       0       0       0                 retail        1951   \n",
       "3  ...       0       0       0               eduation        1980   \n",
       "4  ...       0       0       0  distribution_shipping        1985   \n",
       "\n",
       "   floor_area_binned  energy_intensity_floor_area  \\\n",
       "0                  1                            3   \n",
       "1                  4                            1   \n",
       "2                  4                            1   \n",
       "3                  1                            2   \n",
       "4                  1                            1   \n",
       "\n",
       "   energy_intensity_built_year  energy_intensity_facility_type  \\\n",
       "0                            3                               2   \n",
       "1                            1                               2   \n",
       "2                            1                               2   \n",
       "3                            2                               1   \n",
       "4                            1                               1   \n",
       "\n",
       "   mean_energy_rating_building_id  \n",
       "0                            29.0  \n",
       "1                            40.6  \n",
       "2                            97.8  \n",
       "3                            49.1  \n",
       "4                           100.0  \n",
       "\n",
       "[5 rows x 77 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#merge to df\n",
    "df = df.merge(agg, how = 'left', on = 'building_id')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "application/vnd.databricks.v1+cell": {
     "inputWidgets": {},
     "nuid": "5cddc46d-30cf-4b3b-8849-acb976ed0177",
     "showTitle": false,
     "title": ""
    }
   },
   "source": [
    "### 2.6 month_number (12x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_january(row):\n",
    "    max_min = row.january_max_temp - row.january_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.january_avg_temp\n",
    "    min_3 = row.january_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.january_max_temp\n",
    "    \n",
    "    smallest = 0.0227*31\n",
    "    middle = 0.136*31\n",
    "    biggest = 0.3413*31\n",
    "    \n",
    "    january_total =0\n",
    "    january_total += ((min_3+min_2)/2)*smallest\n",
    "    january_total += ((plus_3+plus_2)/2)*smallest\n",
    "    january_total += ((min_2+min_1)/2)*middle\n",
    "    january_total += ((plus_2+plus_1)/2)*middle\n",
    "    january_total += ((min_1+mean)/2)*biggest\n",
    "    january_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return january_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_january'] = df.apply(get_number_january, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_february(row):\n",
    "    max_min = row.february_max_temp - row.february_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.february_avg_temp\n",
    "    min_3 = row.february_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.february_max_temp\n",
    "    \n",
    "    smallest = 0.0227*28\n",
    "    middle = 0.136*28\n",
    "    biggest = 0.3413*28\n",
    "    \n",
    "    february_total =0\n",
    "    february_total += ((min_3+min_2)/2)*smallest\n",
    "    february_total += ((plus_3+plus_2)/2)*smallest\n",
    "    february_total += ((min_2+min_1)/2)*middle\n",
    "    february_total += ((plus_2+plus_1)/2)*middle\n",
    "    february_total += ((min_1+mean)/2)*biggest\n",
    "    february_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return february_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_february'] = df.apply(get_number_february, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_march(row):\n",
    "    max_min = row.march_max_temp - row.march_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.march_avg_temp\n",
    "    min_3 = row.march_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.march_max_temp\n",
    "    \n",
    "    smallest = 0.0227*31\n",
    "    middle = 0.136*31\n",
    "    biggest = 0.3413*31\n",
    "    \n",
    "    march_total =0\n",
    "    march_total += ((min_3+min_2)/2)*smallest\n",
    "    march_total += ((plus_3+plus_2)/2)*smallest\n",
    "    march_total += ((min_2+min_1)/2)*middle\n",
    "    march_total += ((plus_2+plus_1)/2)*middle\n",
    "    march_total += ((min_1+mean)/2)*biggest\n",
    "    march_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return march_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_march'] = df.apply(get_number_march, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_april(row):\n",
    "    max_min = row.april_max_temp - row.april_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.april_avg_temp\n",
    "    min_3 = row.april_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.april_max_temp\n",
    "    \n",
    "    smallest = 0.0227*30\n",
    "    middle = 0.136*30\n",
    "    biggest = 0.3413*30\n",
    "    \n",
    "    april_total =0\n",
    "    april_total += ((min_3+min_2)/2)*smallest\n",
    "    april_total += ((plus_3+plus_2)/2)*smallest\n",
    "    april_total += ((min_2+min_1)/2)*middle\n",
    "    april_total += ((plus_2+plus_1)/2)*middle\n",
    "    april_total += ((min_1+mean)/2)*biggest\n",
    "    april_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return april_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_april'] = df.apply(get_number_april, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_may(row):\n",
    "    max_min = row.may_max_temp - row.may_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.may_avg_temp\n",
    "    min_3 = row.may_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.may_max_temp\n",
    "    \n",
    "    smallest = 0.0227*31\n",
    "    middle = 0.136*31\n",
    "    biggest = 0.3413*31\n",
    "    \n",
    "    may_total =0\n",
    "    may_total += ((min_3+min_2)/2)*smallest\n",
    "    may_total += ((plus_3+plus_2)/2)*smallest\n",
    "    may_total += ((min_2+min_1)/2)*middle\n",
    "    may_total += ((plus_2+plus_1)/2)*middle\n",
    "    may_total += ((min_1+mean)/2)*biggest\n",
    "    may_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return may_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_may'] = df.apply(get_number_may, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_june(row):\n",
    "    max_min = row.june_max_temp - row.june_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.june_avg_temp\n",
    "    min_3 = row.june_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.june_max_temp\n",
    "    \n",
    "    smallest = 0.0227*30\n",
    "    middle = 0.136*30\n",
    "    biggest = 0.3413*30\n",
    "    \n",
    "    june_total =0\n",
    "    june_total += ((min_3+min_2)/2)*smallest\n",
    "    june_total += ((plus_3+plus_2)/2)*smallest\n",
    "    june_total += ((min_2+min_1)/2)*middle\n",
    "    june_total += ((plus_2+plus_1)/2)*middle\n",
    "    june_total += ((min_1+mean)/2)*biggest\n",
    "    june_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return june_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_june'] = df.apply(get_number_june, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_july(row):\n",
    "    max_min = row.july_max_temp - row.july_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.july_avg_temp\n",
    "    min_3 = row.july_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.july_max_temp\n",
    "    \n",
    "    smallest = 0.0227*31\n",
    "    middle = 0.136*31\n",
    "    biggest = 0.3413*31\n",
    "    \n",
    "    july_total =0\n",
    "    july_total += ((min_3+min_2)/2)*smallest\n",
    "    july_total += ((plus_3+plus_2)/2)*smallest\n",
    "    july_total += ((min_2+min_1)/2)*middle\n",
    "    july_total += ((plus_2+plus_1)/2)*middle\n",
    "    july_total += ((min_1+mean)/2)*biggest\n",
    "    july_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return july_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_july'] = df.apply(get_number_july, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_august(row):\n",
    "    max_min = row.august_max_temp - row.august_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.august_avg_temp\n",
    "    min_3 = row.august_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.august_max_temp\n",
    "    \n",
    "    smallest = 0.0227*31\n",
    "    middle = 0.136*31\n",
    "    biggest = 0.3413*31\n",
    "    \n",
    "    august_total =0\n",
    "    august_total += ((min_3+min_2)/2)*smallest\n",
    "    august_total += ((plus_3+plus_2)/2)*smallest\n",
    "    august_total += ((min_2+min_1)/2)*middle\n",
    "    august_total += ((plus_2+plus_1)/2)*middle\n",
    "    august_total += ((min_1+mean)/2)*biggest\n",
    "    august_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return august_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_august'] = df.apply(get_number_august, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_september(row):\n",
    "    max_min = row.september_max_temp - row.september_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.september_avg_temp\n",
    "    min_3 = row.september_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.september_max_temp\n",
    "    \n",
    "    smallest = 0.0227*30\n",
    "    middle = 0.136*30\n",
    "    biggest = 0.3413*30\n",
    "    \n",
    "    september_total =0\n",
    "    september_total += ((min_3+min_2)/2)*smallest\n",
    "    september_total += ((plus_3+plus_2)/2)*smallest\n",
    "    september_total += ((min_2+min_1)/2)*middle\n",
    "    september_total += ((plus_2+plus_1)/2)*middle\n",
    "    september_total += ((min_1+mean)/2)*biggest\n",
    "    september_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return september_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_september'] = df.apply(get_number_september, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_october(row):\n",
    "    max_min = row.october_max_temp - row.october_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.october_avg_temp\n",
    "    min_3 = row.october_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.october_max_temp\n",
    "    \n",
    "    smallest = 0.0227*31\n",
    "    middle = 0.136*31\n",
    "    biggest = 0.3413*31\n",
    "    \n",
    "    october_total =0\n",
    "    october_total += ((min_3+min_2)/2)*smallest\n",
    "    october_total += ((plus_3+plus_2)/2)*smallest\n",
    "    october_total += ((min_2+min_1)/2)*middle\n",
    "    october_total += ((plus_2+plus_1)/2)*middle\n",
    "    october_total += ((min_1+mean)/2)*biggest\n",
    "    october_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return october_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_october'] = df.apply(get_number_october, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_november(row):\n",
    "    max_min = row.november_max_temp - row.november_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.november_avg_temp\n",
    "    min_3 = row.november_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.november_max_temp\n",
    "    \n",
    "    smallest = 0.0227*30\n",
    "    middle = 0.136*30\n",
    "    biggest = 0.3413*30\n",
    "    \n",
    "    november_total =0\n",
    "    november_total += ((min_3+min_2)/2)*smallest\n",
    "    november_total += ((plus_3+plus_2)/2)*smallest\n",
    "    november_total += ((min_2+min_1)/2)*middle\n",
    "    november_total += ((plus_2+plus_1)/2)*middle\n",
    "    november_total += ((min_1+mean)/2)*biggest\n",
    "    november_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return november_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_november'] = df.apply(get_number_november, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_number_december(row):\n",
    "    max_min = row.december_max_temp - row.december_min_temp\n",
    "    st_dev = max_min/6\n",
    "    mean = row.december_avg_temp\n",
    "    min_3 = row.december_min_temp\n",
    "    min_1 = mean - st_dev\n",
    "    min_2 = min_1 - st_dev\n",
    "    plus_1 = mean + st_dev\n",
    "    plus_2 = plus_1 + st_dev\n",
    "    plus_3 = row.december_max_temp\n",
    "    \n",
    "    smallest = 0.0227*31\n",
    "    middle = 0.136*31\n",
    "    biggest = 0.3413*31\n",
    "    \n",
    "    december_total =0\n",
    "    december_total += ((min_3+min_2)/2)*smallest\n",
    "    december_total += ((plus_3+plus_2)/2)*smallest\n",
    "    december_total += ((min_2+min_1)/2)*middle\n",
    "    december_total += ((plus_2+plus_1)/2)*middle\n",
    "    december_total += ((min_1+mean)/2)*biggest\n",
    "    december_total += ((plus_1+mean)/2)*biggest\n",
    "    \n",
    "    return december_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_december'] = df.apply(get_number_december, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "correlation = df[['number_january','number_february','number_march','number_april','number_may','number_june',\n",
    "                  'number_july','number_august','number_september','number_october', 'number_november',\n",
    "                  'number_december','site_eui']].corr(method=\"pearson\")\n",
    "mask = np.triu(np.ones_like(correlation, dtype=bool))\n",
    "fig = plt.figure(figsize=(9,6), dpi=100, facecolor = 'w', edgecolor=\"k\")\n",
    "figure = sns.heatmap(correlation, mask=mask, cmap= \"Spectral\", center=0, linewidths=1, fmt=\".2f\",annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.7 total_number"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['total_number'] = df.apply(lambda row:\n",
    "                             row.number_december + row.number_november + row.number_october + row.number_september + row.number_august + row.number_july\n",
    "                             + row.number_june + row.number_may + row.number_april + row.number_march + row.number_february + row.number_january, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.8 number_season (4x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['number_winter'] = df.apply(lambda row:\n",
    "                             row.number_march + row.number_february + row.number_january, axis=1)\n",
    "df['number_spring'] = df.apply(lambda row:\n",
    "                             + row.number_june + row.number_may + row.number_april, axis=1)\n",
    "df['number_summer'] = df.apply(lambda row:\n",
    "                             row.number_september + row.number_august + row.number_july, axis=1)\n",
    "df['number_autumn'] = df.apply(lambda row:\n",
    "                             row.number_december + row.number_november + row.number_october, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "correlation = df[['number_winter','number_spring','number_summer','number_autumn','total_number','heating_degree_days','cooling_degree_days','site_eui']].corr(method=\"pearson\")\n",
    "mask = np.triu(np.ones_like(correlation, dtype=bool))\n",
    "fig = plt.figure(figsize=(5,3), dpi=100, facecolor = 'w', edgecolor=\"k\")\n",
    "figure = sns.heatmap(correlation, mask=mask, cmap= \"Spectral\", center=0, linewidths=1, fmt=\".2f\",annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.9 eui_improved"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "def bool_eui_BUILD(row):\n",
    "    temp_df = df[df['building_id'] ==  row.building_id]\n",
    "    e_star = temp_df['site_eui'].to_list()\n",
    "    bool_star = e_star[-1] - e_star[0]\n",
    "    if bool_star > 0:\n",
    "        prev_e_star = 1\n",
    "    elif bool_star == 0:\n",
    "        prev_e_star = 0\n",
    "    else:\n",
    "        prev_e_star = -1\n",
    "    \n",
    "    return prev_e_star"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['eui_improved'] = df.apply(bool_eui_BUILD, axis =1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.10 eui_diff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "def diff_eui(row):\n",
    "    temp_df = df[df['building_id'] ==  row.building_id]\n",
    "    years = temp_df['year_factor'].to_list()\n",
    "    e_star = temp_df['site_eui'].to_list()\n",
    "    if len(years) > 1:\n",
    "        years_2 = [x for x in years if x < row.year_factor]\n",
    "        if len(years_2) > 0:\n",
    "            prev_e_star = row.site_eui - e_star[years.index(max(years_2))]\n",
    "        else:\n",
    "            prev_e_star = 0\n",
    "    else:\n",
    "        prev_e_star = 0\n",
    "    \n",
    "    return prev_e_star"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['eui_diff'] = df.apply(diff_eui, axis =1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['eui_diff'] = df.eui_diff.fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 300x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "correlation = df[['eui_diff','eui_improved','site_eui']].corr(method=\"pearson\")\n",
    "mask = np.triu(np.ones_like(correlation, dtype=bool))\n",
    "fig = plt.figure(figsize=(3,2), dpi=100, facecolor = 'w', edgecolor=\"k\")\n",
    "figure = sns.heatmap(correlation, mask=mask, cmap= \"Spectral\", center=0, linewidths=1, fmt=\".2f\",annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.11 hdd_extreme\n",
    "#### 2.11 A hdd_extreme"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "def hdd(row):\n",
    "    hdd = 0 \n",
    "    if row.january_avg_temp < 65:\n",
    "        hdd += 65 - row.january_avg_temp\n",
    "    if row.january_min_temp < 65:\n",
    "        hdd += 65 - row.january_min_temp    \n",
    "    \n",
    "    if row.february_avg_temp < 65:\n",
    "        hdd += 65 - row.february_avg_temp\n",
    "    if row.february_min_temp < 65:\n",
    "        hdd += 65 - row.february_min_temp\n",
    "    \n",
    "    if row.march_avg_temp <65:\n",
    "        hdd += 65 - row.march_avg_temp\n",
    "    if row.march_min_temp < 65:\n",
    "        hdd += 65 - row.march_min_temp\n",
    "        \n",
    "    if row.april_avg_temp < 65:\n",
    "        hdd += 65 - row.april_avg_temp\n",
    "    if row.april_min_temp < 65:\n",
    "        hdd += 65 - row.april_min_temp\n",
    "    \n",
    "    if row.may_avg_temp < 65:\n",
    "        hdd += 65 - row.may_avg_temp\n",
    "    if row.may_min_temp < 65:\n",
    "        hdd += 65 - row.may_min_temp\n",
    "        \n",
    "    if row.june_avg_temp < 65:\n",
    "        hdd += 65 - row.june_avg_temp\n",
    "    if row.june_min_temp < 65:\n",
    "        hdd += 65 - row.june_min_temp\n",
    "        \n",
    "    if row.july_avg_temp < 65:\n",
    "        hdd += 65 - row.july_avg_temp\n",
    "    if row.july_min_temp < 65:\n",
    "        hdd += 65 - row.july_min_temp\n",
    "    \n",
    "    if row.august_avg_temp < 65:\n",
    "        hdd += 65 - row.august_avg_temp\n",
    "    if row.august_min_temp < 65:\n",
    "        hdd += 65 - row.august_min_temp\n",
    "        \n",
    "    if row.september_avg_temp < 65:\n",
    "        hdd += 65 - row.september_avg_temp\n",
    "    if row.september_min_temp < 65:\n",
    "        hdd += 65 - row.september_min_temp\n",
    "    \n",
    "    if row.october_avg_temp < 65:\n",
    "        hdd += 65 - row.october_avg_temp\n",
    "    if row.october_min_temp < 65:\n",
    "        hdd += 65 - row.october_min_temp\n",
    "        \n",
    "    if row.november_avg_temp < 65:\n",
    "        hdd += 65 - row.november_avg_temp\n",
    "    if row.november_min_temp < 65:\n",
    "        hdd += 65 - row.november_min_temp\n",
    "    \n",
    "    if row.december_avg_temp < 65:\n",
    "        hdd += 65 - row.december_avg_temp\n",
    "    if row.december_min_temp < 65:\n",
    "        hdd += 65 - row.december_min_temp\n",
    "        \n",
    "    return hdd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['hdd_extreme'] = df.apply(hdd, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.11 B hdd_extreme_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "def hdd_2ext(row):\n",
    "    hdd = 0 \n",
    "    if row.january_avg_temp < 65:\n",
    "        hdd += 65 - row.january_avg_temp\n",
    "    if row.january_min_temp < 65:\n",
    "        hdd += (65 - row.january_min_temp)*2\n",
    "    \n",
    "    if row.february_avg_temp < 65:\n",
    "        hdd += 65 - row.february_avg_temp\n",
    "    if row.february_min_temp < 65:\n",
    "        hdd += (65 - row.february_min_temp)*2\n",
    "    \n",
    "    if row.march_avg_temp <65:\n",
    "        hdd += 65 - row.march_avg_temp\n",
    "    if row.march_min_temp < 65:\n",
    "        hdd += (65 - row.march_min_temp)*2\n",
    "        \n",
    "    if row.april_avg_temp < 65:\n",
    "        hdd += 65 - row.april_avg_temp\n",
    "    if row.april_min_temp < 65:\n",
    "        hdd += (65 - row.april_min_temp)*2\n",
    "    \n",
    "    if row.may_avg_temp < 65:\n",
    "        hdd += 65 - row.may_avg_temp\n",
    "    if row.may_min_temp < 65:\n",
    "        hdd += (65 - row.may_min_temp)*2\n",
    "        \n",
    "    if row.june_avg_temp < 65:\n",
    "        hdd += 65 - row.june_avg_temp\n",
    "    if row.june_min_temp < 65:\n",
    "        hdd += (65 - row.june_min_temp)*2\n",
    "        \n",
    "    if row.july_avg_temp < 65:\n",
    "        hdd += 65 - row.july_avg_temp\n",
    "    if row.july_min_temp < 65:\n",
    "        hdd += (65 - row.july_min_temp)*2\n",
    "    \n",
    "    if row.august_avg_temp < 65:\n",
    "        hdd += 65 - row.august_avg_temp\n",
    "    if row.august_min_temp < 65:\n",
    "        hdd += (65 - row.august_min_temp)*2\n",
    "        \n",
    "    if row.september_avg_temp < 65:\n",
    "        hdd += 65 - row.september_avg_temp\n",
    "    if row.september_min_temp < 65:\n",
    "        hdd += (65 - row.september_min_temp)*2\n",
    "    \n",
    "    if row.october_avg_temp < 65:\n",
    "        hdd += 65 - row.october_avg_temp\n",
    "    if row.october_min_temp < 65:\n",
    "        hdd += (65 - row.october_min_temp)*2\n",
    "        \n",
    "    if row.november_avg_temp < 65:\n",
    "        hdd += 65 - row.november_avg_temp\n",
    "    if row.november_min_temp < 65:\n",
    "        hdd += (65 - row.november_min_temp)*2\n",
    "    \n",
    "    if row.december_avg_temp < 65:\n",
    "        hdd += 65 - row.december_avg_temp\n",
    "    if row.december_min_temp < 65:\n",
    "        hdd += (65 - row.december_min_temp)*2\n",
    "        \n",
    "    return hdd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['hdd_extreme_2'] = df.apply(hdd_2ext, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 300x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "correlation = df[['hdd_extreme','hdd_extreme_2','site_eui']].corr(method=\"pearson\")\n",
    "mask = np.triu(np.ones_like(correlation, dtype=bool))\n",
    "fig = plt.figure(figsize=(3,2), dpi=100, facecolor = 'w', edgecolor=\"k\")\n",
    "figure = sns.heatmap(correlation, mask=mask, cmap= \"Spectral\", center=0, linewidths=1, fmt=\".2f\",annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.12 calculated_eui\n",
    "#### 2.12 avg_total_number_building"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "building_id_list = df['building_id'].unique()\n",
    "\n",
    "def temp_number_building(row):\n",
    "    temp_df = df[df['building_id'] ==  row.building_id]\n",
    "    total_numbers = temp_df['total_number'].to_list()\n",
    "    total_number_sum =0\n",
    "    for x in total_numbers:\n",
    "        total_number_sum += x\n",
    "    \n",
    "    return total_number_sum/len(total_numbers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['avg_total_number_building'] = df.apply(temp_number_building, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.12 avg_eui_building\n",
    "Heeft nog 719 np.nan values maar kan gebruikt worden voor het eventueel mean encoden van building_id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "def average_eui_building(row):\n",
    "    temp_df = df[df['building_id'] ==  row.building_id]\n",
    "    total_euis = temp_df['site_eui'].to_list()\n",
    "    total_euis = [x for x in total_euis if x==x]\n",
    "    total_eui = 0\n",
    "    for x in total_euis:\n",
    "        total_eui += x\n",
    "    if len(total_euis) > 0:\n",
    "        return total_eui/len(total_euis)\n",
    "    else:\n",
    "        return np.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['avg_eui_building'] = df.apply(average_eui_building, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['avg_total_number_building_DIV_avg_eui_building'] = df.apply(lambda row: row.avg_total_number_building/row.avg_eui_building, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.12 year_built_bin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "cut_labels = [1, 2, 3, 4, 5]\n",
    "cut_bins = [0,  1930, 1950, 1970, 1990, 2022]\n",
    "df['year_built_bin'] = pd.cut(df['year_built'], bins=cut_bins, labels=cut_labels)\n",
    "df['year_built_bin'] = df['year_built_bin'].astype('int')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_feature2(row):\n",
    "    temp_df = df[df['year_built_bin'] == row.year_built_bin]\n",
    "    temp_df_eui_list = temp_df['site_eui'].to_list()\n",
    "    temp_df_eui_list.sort()\n",
    "    p1 = np.nanpercentile(temp_df_eui_list, 25, axis=0)\n",
    "    p3 = np.nanpercentile(temp_df_eui_list, 75, axis=0)\n",
    "    InterQ = abs(p3-p1)\n",
    "    if row.site_eui < (p1 - (1.5*InterQ)):\n",
    "        return 1\n",
    "    elif row.site_eui > (p3 + (1.5*InterQ)):\n",
    "        return 3\n",
    "    else:\n",
    "        return 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['yearB_E_int']= df.apply(create_feature2, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.12 calculated_eui"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "site_eui                                          9705\n",
       "avg_total_number_building_DIV_avg_eui_building     716\n",
       "avg_eui_building                                   716\n",
       "year_factor                                          0\n",
       "peak_NW                                              0\n",
       "                                                  ... \n",
       "september_avg_temp                                   0\n",
       "september_min_temp                                   0\n",
       "august_max_temp                                      0\n",
       "august_avg_temp                                      0\n",
       "yearB_E_int                                          0\n",
       "Length: 103, dtype: int64"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['calculated_eui'] = df.apply(lambda row: row.total_number / row.avg_total_number_building_DIV_avg_eui_building, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "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>facility_type</th>\n",
       "      <th>year_built_bin</th>\n",
       "      <th>site_eui</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2to4_Unit_Building</td>\n",
       "      <td>1</td>\n",
       "      <td>44.290212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2to4_Unit_Building</td>\n",
       "      <td>2</td>\n",
       "      <td>36.697574</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2to4_Unit_Building</td>\n",
       "      <td>3</td>\n",
       "      <td>28.579536</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2to4_Unit_Building</td>\n",
       "      <td>4</td>\n",
       "      <td>26.166490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2to4_Unit_Building</td>\n",
       "      <td>5</td>\n",
       "      <td>31.317190</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        facility_type  year_built_bin   site_eui\n",
       "0  2to4_Unit_Building               1  44.290212\n",
       "1  2to4_Unit_Building               2  36.697574\n",
       "2  2to4_Unit_Building               3  28.579536\n",
       "3  2to4_Unit_Building               4  26.166490\n",
       "4  2to4_Unit_Building               5  31.317190"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_df = df.groupby(['facility_type','year_built_bin'])['site_eui'].mean()\n",
    "temp_df = temp_df.reset_index()\n",
    "temp_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "def calculated_eui(row):\n",
    "    if row.calculated_eui != row.calculated_eui:\n",
    "        return temp_df[(temp_df['facility_type'] == row.facility_type) & (temp_df['year_built_bin'] == row.year_built_bin)]['site_eui'].values[0]\n",
    "    else:\n",
    "        return row.calculated_eui"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['calculated_eui'] = df.apply(calculated_eui, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "def calculated_eui_2(row):\n",
    "    if row.calculated_eui != row.calculated_eui:\n",
    "        return temp_df[(temp_df['facility_type'] == row.facility_type)]['site_eui'].values[0]\n",
    "    else:\n",
    "        return row.calculated_eui"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['calculated_eui'] = df.apply(calculated_eui_2, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['calculated_eui'] = df['calculated_eui'].fillna(df['calculated_eui'].median())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "del df['avg_total_number_building_DIV_avg_eui_building']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.13 calc_eui_perc_off"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "def percentage_off(row):\n",
    "    diff_eui = row.site_eui - row.calculated_eui\n",
    "    perc = diff_eui/row.site_eui\n",
    "    return perc*-1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['calc_eui_perc_off'] = df.apply(percentage_off, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fill_calc_eui_perc_off(row):\n",
    "    if row.calc_eui_perc_off != row.calc_eui_perc_off:\n",
    "        df_temp = df[df['building_id'] == row.building_id]\n",
    "        list_calc_eui_perc_off = df_temp['calc_eui_perc_off'].to_list()\n",
    "        list_calc_eui_perc_off = list_calc_eui_perc_off[:-1]\n",
    "        if len(list_calc_eui_perc_off) >1:\n",
    "            #print(list_calc_eui_perc_off)\n",
    "            return sum(list_calc_eui_perc_off) / len(list_calc_eui_perc_off)\n",
    "        else:\n",
    "            return np.nan\n",
    "    else:\n",
    "        return row.calc_eui_perc_off"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['calc_eui_perc_off'] = df.apply(fill_calc_eui_perc_off, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['calc_eui_perc_off'] = df['calc_eui_perc_off'].fillna(df['calc_eui_perc_off'].mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.14 binned_calc_eui\n",
    "#### 2.14 binned_calc_eui"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "list_eui = df['calculated_eui'].to_list()\n",
    "p0 = np.percentile(list_eui, 0)\n",
    "p1 = np.percentile(list_eui, 10)\n",
    "p2 = np.percentile(list_eui, 20)\n",
    "p3 = np.percentile(list_eui, 30)\n",
    "p4 = np.percentile(list_eui, 40)\n",
    "p5 = np.percentile(list_eui, 50)\n",
    "p6 = np.percentile(list_eui, 60)\n",
    "p7 = np.percentile(list_eui, 70)\n",
    "p8 = np.percentile(list_eui, 80)\n",
    "p9 = np.percentile(list_eui, 90)\n",
    "p10 = np.percentile(list_eui, 100)\n",
    " \n",
    "bins = [p0,p1,p2,p3,p4,p5,p6,p7,p8,p9,p10]\n",
    "labels = [1,2,3,4,5,6,7,8,9,10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['binned_calc_eui'] = pd.cut(df['calculated_eui'], bins=bins, labels=labels)\n",
    "df['binned_calc_eui'] = df['binned_calc_eui'].fillna(1)\n",
    "df['binned_calc_eui'] = df.binned_calc_eui.astype('int64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (10,7)\n",
    "ax = sns.barplot(x=\"binned_calc_eui\", y=\"site_eui\", data=df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2.14 binned_calc_eui_20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "list_eui = df['calculated_eui'].to_list()\n",
    "p0 = np.percentile(list_eui, 0)\n",
    "p1 = np.percentile(list_eui, 5)\n",
    "p2 = np.percentile(list_eui, 10)\n",
    "p3 = np.percentile(list_eui, 15)\n",
    "p4 = np.percentile(list_eui, 20)\n",
    "p5 = np.percentile(list_eui, 25)\n",
    "p6 = np.percentile(list_eui, 30)\n",
    "p7 = np.percentile(list_eui, 35)\n",
    "p8 = np.percentile(list_eui, 40)\n",
    "p9 = np.percentile(list_eui, 45)\n",
    "p10 = np.percentile(list_eui, 50)\n",
    "p11 = np.percentile(list_eui, 55)\n",
    "p12 = np.percentile(list_eui, 60)\n",
    "p13 = np.percentile(list_eui, 65)\n",
    "p14 = np.percentile(list_eui, 70)\n",
    "p15 = np.percentile(list_eui, 75)\n",
    "p16 = np.percentile(list_eui, 80)\n",
    "p17 = np.percentile(list_eui, 85)\n",
    "p18 = np.percentile(list_eui, 90)\n",
    "p19 = np.percentile(list_eui, 95)\n",
    "p20 = np.percentile(list_eui, 100)\n",
    " \n",
    "bins = [p0,p1,p2,p3,p4,p5,p6,p7,p8,p9,p10,p11,p12,p13,p14,p15,p16,p17,p18,p19,p20]\n",
    "labels = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['binned_calc_eui_20'] = pd.cut(df['calculated_eui'], bins=bins, labels=labels)\n",
    "df['binned_calc_eui_20'] = df['binned_calc_eui_20'].fillna(1)\n",
    "df['binned_calc_eui_20'] = df.binned_calc_eui.astype('int64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "correlation = df[['binned_calc_eui','binned_calc_eui_20','calc_eui_perc_off','calculated_eui','site_eui']].corr(method=\"pearson\")\n",
    "mask = np.triu(np.ones_like(correlation, dtype=bool))\n",
    "fig = plt.figure(figsize=(5,2), dpi=100, facecolor = 'w', edgecolor=\"k\")\n",
    "figure = sns.heatmap(correlation, mask=mask, cmap= \"Spectral\", center=0, linewidths=1, fmt=\".2f\",annot=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.15 verschil_uitstoot_per_temp_tov_jaar_state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "table = pd.pivot_table(df, values=['january_min_temp',\n",
    " 'january_avg_temp',\n",
    " 'january_max_temp',\n",
    " 'february_min_temp',\n",
    " 'february_avg_temp',\n",
    " 'february_max_temp',\n",
    " 'march_min_temp',\n",
    " 'march_avg_temp',\n",
    " 'march_max_temp',\n",
    " 'april_min_temp',\n",
    " 'april_avg_temp',\n",
    " 'april_max_temp',\n",
    " 'may_min_temp',\n",
    " 'may_avg_temp',\n",
    " 'may_max_temp',\n",
    " 'june_min_temp',\n",
    " 'june_avg_temp',\n",
    " 'june_max_temp',\n",
    " 'july_min_temp',\n",
    " 'july_avg_temp',\n",
    " 'july_max_temp',\n",
    " 'august_min_temp',\n",
    " 'august_avg_temp',\n",
    " 'august_max_temp',\n",
    " 'september_min_temp',\n",
    " 'september_avg_temp',\n",
    " 'september_max_temp',\n",
    " 'october_min_temp',\n",
    " 'october_avg_temp',\n",
    " 'october_max_temp',\n",
    " 'november_min_temp',\n",
    " 'november_avg_temp',\n",
    " 'november_max_temp',\n",
    " 'december_min_temp',\n",
    " 'december_avg_temp',\n",
    " 'december_max_temp','heating_degree_days','cooling_degree_days','total_number',\n",
    " 'number_winter',\n",
    " 'number_spring',\n",
    " 'number_summer',\n",
    " 'number_autumn','site_eui'], index=['state_factor', 'year_factor'],\n",
    "                    aggfunc={'january_min_temp': np.mean ,\n",
    " 'january_avg_temp': np.mean,\n",
    " 'january_max_temp': np.mean,\n",
    " 'february_min_temp': np.mean,\n",
    " 'february_avg_temp': np.mean,\n",
    " 'february_max_temp': np.mean,\n",
    " 'march_min_temp': np.mean,\n",
    " 'march_avg_temp': np.mean,\n",
    " 'march_max_temp': np.mean,\n",
    " 'april_min_temp': np.mean,\n",
    " 'april_avg_temp': np.mean,\n",
    " 'april_max_temp': np.mean,\n",
    " 'may_min_temp': np.mean,\n",
    " 'may_avg_temp': np.mean,\n",
    " 'may_max_temp': np.mean,\n",
    " 'june_min_temp': np.mean,\n",
    " 'june_avg_temp': np.mean,\n",
    " 'june_max_temp': np.mean,\n",
    " 'july_min_temp': np.mean,\n",
    " 'july_avg_temp': np.mean,\n",
    " 'july_max_temp': np.mean,\n",
    " 'august_min_temp': np.mean,\n",
    " 'august_avg_temp': np.mean,\n",
    " 'august_max_temp': np.mean,\n",
    " 'september_min_temp': np.mean,\n",
    " 'september_avg_temp': np.mean,\n",
    " 'september_max_temp': np.mean,\n",
    " 'october_min_temp': np.mean,\n",
    " 'october_avg_temp': np.mean,\n",
    " 'october_max_temp': np.mean,\n",
    " 'november_min_temp': np.mean,\n",
    " 'november_avg_temp': np.mean,\n",
    " 'november_max_temp': np.mean,\n",
    " 'december_min_temp': np.mean,\n",
    " 'december_avg_temp': np.mean,\n",
    " 'december_max_temp': np.mean,\n",
    " 'heating_degree_days': np.mean,\n",
    " 'cooling_degree_days': np.mean,'total_number': np.mean,\n",
    " 'number_winter': np.mean,\n",
    " 'number_spring': np.mean,\n",
    " 'number_summer': np.mean,\n",
    " 'number_autumn': np.mean, 'site_eui': np.mean})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "table = table.reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>state_factor</th>\n",
       "      <th>year_factor</th>\n",
       "      <th>april_avg_temp</th>\n",
       "      <th>april_max_temp</th>\n",
       "      <th>april_min_temp</th>\n",
       "      <th>august_avg_temp</th>\n",
       "      <th>august_max_temp</th>\n",
       "      <th>august_min_temp</th>\n",
       "      <th>cooling_degree_days</th>\n",
       "      <th>december_avg_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>number_summer</th>\n",
       "      <th>number_winter</th>\n",
       "      <th>october_avg_temp</th>\n",
       "      <th>october_max_temp</th>\n",
       "      <th>october_min_temp</th>\n",
       "      <th>september_avg_temp</th>\n",
       "      <th>september_max_temp</th>\n",
       "      <th>september_min_temp</th>\n",
       "      <th>site_eui</th>\n",
       "      <th>total_number</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>State_1</td>\n",
       "      <td>1</td>\n",
       "      <td>55.500000</td>\n",
       "      <td>78.000000</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>62.161290</td>\n",
       "      <td>85.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>115.000000</td>\n",
       "      <td>49.274194</td>\n",
       "      <td>...</td>\n",
       "      <td>5823.496350</td>\n",
       "      <td>4654.161250</td>\n",
       "      <td>63.016129</td>\n",
       "      <td>83.000000</td>\n",
       "      <td>47.000000</td>\n",
       "      <td>64.650000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>70.148330</td>\n",
       "      <td>20836.972300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>State_1</td>\n",
       "      <td>2</td>\n",
       "      <td>58.966416</td>\n",
       "      <td>89.877941</td>\n",
       "      <td>36.927941</td>\n",
       "      <td>72.491224</td>\n",
       "      <td>100.123529</td>\n",
       "      <td>52.408824</td>\n",
       "      <td>872.120588</td>\n",
       "      <td>48.139354</td>\n",
       "      <td>...</td>\n",
       "      <td>6529.767616</td>\n",
       "      <td>4621.379011</td>\n",
       "      <td>64.253925</td>\n",
       "      <td>99.764706</td>\n",
       "      <td>43.182353</td>\n",
       "      <td>69.176656</td>\n",
       "      <td>97.291176</td>\n",
       "      <td>49.491176</td>\n",
       "      <td>66.543553</td>\n",
       "      <td>22140.545271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>State_1</td>\n",
       "      <td>3</td>\n",
       "      <td>60.217771</td>\n",
       "      <td>87.844605</td>\n",
       "      <td>36.847021</td>\n",
       "      <td>69.487742</td>\n",
       "      <td>94.402576</td>\n",
       "      <td>49.250403</td>\n",
       "      <td>640.805958</td>\n",
       "      <td>45.205850</td>\n",
       "      <td>...</td>\n",
       "      <td>6336.734571</td>\n",
       "      <td>4476.664246</td>\n",
       "      <td>59.099922</td>\n",
       "      <td>85.400966</td>\n",
       "      <td>38.419485</td>\n",
       "      <td>67.863899</td>\n",
       "      <td>96.642512</td>\n",
       "      <td>44.437198</td>\n",
       "      <td>60.261259</td>\n",
       "      <td>21513.691395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>State_1</td>\n",
       "      <td>4</td>\n",
       "      <td>60.524645</td>\n",
       "      <td>90.264076</td>\n",
       "      <td>45.635210</td>\n",
       "      <td>68.575825</td>\n",
       "      <td>83.245837</td>\n",
       "      <td>57.236320</td>\n",
       "      <td>707.289453</td>\n",
       "      <td>56.148588</td>\n",
       "      <td>...</td>\n",
       "      <td>6346.894880</td>\n",
       "      <td>5085.608160</td>\n",
       "      <td>68.316349</td>\n",
       "      <td>94.915940</td>\n",
       "      <td>53.067407</td>\n",
       "      <td>69.414209</td>\n",
       "      <td>86.038858</td>\n",
       "      <td>55.398097</td>\n",
       "      <td>58.362694</td>\n",
       "      <td>22877.366604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>State_1</td>\n",
       "      <td>5</td>\n",
       "      <td>60.966667</td>\n",
       "      <td>91.000000</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>74.774194</td>\n",
       "      <td>104.000000</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>1319.000000</td>\n",
       "      <td>48.580645</td>\n",
       "      <td>...</td>\n",
       "      <td>6848.484400</td>\n",
       "      <td>5089.055550</td>\n",
       "      <td>69.580645</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>73.583333</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>63.259263</td>\n",
       "      <td>23167.388150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>State_1</td>\n",
       "      <td>6</td>\n",
       "      <td>60.283333</td>\n",
       "      <td>86.000000</td>\n",
       "      <td>47.000000</td>\n",
       "      <td>63.790323</td>\n",
       "      <td>76.000000</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>142.000000</td>\n",
       "      <td>50.661290</td>\n",
       "      <td>...</td>\n",
       "      <td>5880.955150</td>\n",
       "      <td>5018.946600</td>\n",
       "      <td>63.354839</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>64.800000</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>62.512324</td>\n",
       "      <td>21801.495450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>State_1</td>\n",
       "      <td>7</td>\n",
       "      <td>58.450000</td>\n",
       "      <td>78.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>65.854839</td>\n",
       "      <td>88.000000</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>342.000000</td>\n",
       "      <td>51.887097</td>\n",
       "      <td>...</td>\n",
       "      <td>6139.393600</td>\n",
       "      <td>4860.146350</td>\n",
       "      <td>64.193548</td>\n",
       "      <td>91.000000</td>\n",
       "      <td>48.000000</td>\n",
       "      <td>69.766667</td>\n",
       "      <td>104.000000</td>\n",
       "      <td>54.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21892.417400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>State_10</td>\n",
       "      <td>3</td>\n",
       "      <td>58.883333</td>\n",
       "      <td>91.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>77.080645</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>1719.000000</td>\n",
       "      <td>42.290323</td>\n",
       "      <td>...</td>\n",
       "      <td>7048.078850</td>\n",
       "      <td>3683.881700</td>\n",
       "      <td>62.403226</td>\n",
       "      <td>91.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>71.316667</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>93.232511</td>\n",
       "      <td>21509.816550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>State_10</td>\n",
       "      <td>4</td>\n",
       "      <td>57.333333</td>\n",
       "      <td>85.000000</td>\n",
       "      <td>34.000000</td>\n",
       "      <td>77.661290</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>62.000000</td>\n",
       "      <td>1712.000000</td>\n",
       "      <td>43.725806</td>\n",
       "      <td>...</td>\n",
       "      <td>7085.464150</td>\n",
       "      <td>3390.677400</td>\n",
       "      <td>62.887097</td>\n",
       "      <td>82.000000</td>\n",
       "      <td>44.000000</td>\n",
       "      <td>73.833333</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>54.000000</td>\n",
       "      <td>68.255412</td>\n",
       "      <td>21385.916899</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>State_10</td>\n",
       "      <td>5</td>\n",
       "      <td>59.383333</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>39.000000</td>\n",
       "      <td>79.274194</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>63.000000</td>\n",
       "      <td>1999.000000</td>\n",
       "      <td>51.177419</td>\n",
       "      <td>...</td>\n",
       "      <td>7229.964750</td>\n",
       "      <td>3362.207250</td>\n",
       "      <td>58.854839</td>\n",
       "      <td>79.000000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>74.866667</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>55.000000</td>\n",
       "      <td>61.224291</td>\n",
       "      <td>22008.001150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>State_10</td>\n",
       "      <td>6</td>\n",
       "      <td>56.866667</td>\n",
       "      <td>86.000000</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>82.741935</td>\n",
       "      <td>101.000000</td>\n",
       "      <td>67.000000</td>\n",
       "      <td>1994.000000</td>\n",
       "      <td>41.790323</td>\n",
       "      <td>...</td>\n",
       "      <td>7411.088400</td>\n",
       "      <td>3861.399300</td>\n",
       "      <td>63.080645</td>\n",
       "      <td>87.000000</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>75.983333</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>59.000000</td>\n",
       "      <td>57.834971</td>\n",
       "      <td>22078.946449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>State_10</td>\n",
       "      <td>7</td>\n",
       "      <td>58.050000</td>\n",
       "      <td>87.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>71.290323</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>55.000000</td>\n",
       "      <td>988.000000</td>\n",
       "      <td>34.467742</td>\n",
       "      <td>...</td>\n",
       "      <td>6543.943850</td>\n",
       "      <td>3624.864950</td>\n",
       "      <td>59.266667</td>\n",
       "      <td>83.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>66.450000</td>\n",
       "      <td>88.000000</td>\n",
       "      <td>47.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20154.800564</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>State_11</td>\n",
       "      <td>5</td>\n",
       "      <td>51.822433</td>\n",
       "      <td>77.975556</td>\n",
       "      <td>36.975556</td>\n",
       "      <td>68.391990</td>\n",
       "      <td>89.020370</td>\n",
       "      <td>50.991852</td>\n",
       "      <td>429.458164</td>\n",
       "      <td>43.402432</td>\n",
       "      <td>...</td>\n",
       "      <td>6135.914802</td>\n",
       "      <td>4330.436909</td>\n",
       "      <td>57.484953</td>\n",
       "      <td>74.004074</td>\n",
       "      <td>45.983704</td>\n",
       "      <td>60.410519</td>\n",
       "      <td>80.004074</td>\n",
       "      <td>43.971482</td>\n",
       "      <td>53.484372</td>\n",
       "      <td>20332.190632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>State_11</td>\n",
       "      <td>6</td>\n",
       "      <td>56.222177</td>\n",
       "      <td>88.929215</td>\n",
       "      <td>39.956535</td>\n",
       "      <td>68.562438</td>\n",
       "      <td>90.972058</td>\n",
       "      <td>51.970817</td>\n",
       "      <td>258.530270</td>\n",
       "      <td>38.471568</td>\n",
       "      <td>...</td>\n",
       "      <td>6043.568649</td>\n",
       "      <td>4209.560561</td>\n",
       "      <td>55.346081</td>\n",
       "      <td>67.933561</td>\n",
       "      <td>38.986961</td>\n",
       "      <td>61.328039</td>\n",
       "      <td>78.956535</td>\n",
       "      <td>43.978889</td>\n",
       "      <td>54.489394</td>\n",
       "      <td>20139.244590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>State_11</td>\n",
       "      <td>7</td>\n",
       "      <td>50.754345</td>\n",
       "      <td>66.941248</td>\n",
       "      <td>36.941248</td>\n",
       "      <td>69.685186</td>\n",
       "      <td>91.970624</td>\n",
       "      <td>52.931457</td>\n",
       "      <td>342.698898</td>\n",
       "      <td>40.472243</td>\n",
       "      <td>...</td>\n",
       "      <td>6185.074928</td>\n",
       "      <td>3798.125195</td>\n",
       "      <td>53.108086</td>\n",
       "      <td>71.931457</td>\n",
       "      <td>35.960832</td>\n",
       "      <td>64.806283</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>47.941248</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19487.121402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>State_2</td>\n",
       "      <td>1</td>\n",
       "      <td>59.400000</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>33.000000</td>\n",
       "      <td>76.951613</td>\n",
       "      <td>96.000000</td>\n",
       "      <td>57.000000</td>\n",
       "      <td>1631.000000</td>\n",
       "      <td>41.816667</td>\n",
       "      <td>...</td>\n",
       "      <td>7008.240150</td>\n",
       "      <td>3421.884050</td>\n",
       "      <td>55.306452</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>69.833333</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>81.831620</td>\n",
       "      <td>21104.728178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>State_2</td>\n",
       "      <td>2</td>\n",
       "      <td>55.650000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>34.000000</td>\n",
       "      <td>77.435484</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>63.000000</td>\n",
       "      <td>1509.000000</td>\n",
       "      <td>42.887097</td>\n",
       "      <td>...</td>\n",
       "      <td>6987.869150</td>\n",
       "      <td>4028.213450</td>\n",
       "      <td>56.903226</td>\n",
       "      <td>82.000000</td>\n",
       "      <td>34.000000</td>\n",
       "      <td>69.066667</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>94.532905</td>\n",
       "      <td>21399.019800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>State_2</td>\n",
       "      <td>3</td>\n",
       "      <td>56.583333</td>\n",
       "      <td>91.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>73.783333</td>\n",
       "      <td>93.000000</td>\n",
       "      <td>55.000000</td>\n",
       "      <td>1306.000000</td>\n",
       "      <td>39.500000</td>\n",
       "      <td>...</td>\n",
       "      <td>6762.261842</td>\n",
       "      <td>3444.799250</td>\n",
       "      <td>58.816667</td>\n",
       "      <td>88.000000</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>67.160000</td>\n",
       "      <td>93.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>80.064677</td>\n",
       "      <td>20457.926670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>State_2</td>\n",
       "      <td>4</td>\n",
       "      <td>54.450000</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>73.550000</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>55.000000</td>\n",
       "      <td>1170.000000</td>\n",
       "      <td>40.350000</td>\n",
       "      <td>...</td>\n",
       "      <td>6711.364641</td>\n",
       "      <td>3014.856000</td>\n",
       "      <td>59.183333</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>69.172414</td>\n",
       "      <td>93.000000</td>\n",
       "      <td>46.000000</td>\n",
       "      <td>78.439699</td>\n",
       "      <td>20006.722339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>State_2</td>\n",
       "      <td>5</td>\n",
       "      <td>55.379310</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>75.816667</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>59.000000</td>\n",
       "      <td>1451.000000</td>\n",
       "      <td>48.850000</td>\n",
       "      <td>...</td>\n",
       "      <td>6927.132878</td>\n",
       "      <td>2827.570172</td>\n",
       "      <td>55.767857</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>72.448276</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>77.675325</td>\n",
       "      <td>20614.355553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>State_2</td>\n",
       "      <td>6</td>\n",
       "      <td>54.870370</td>\n",
       "      <td>85.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>79.550000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>61.000000</td>\n",
       "      <td>1570.000000</td>\n",
       "      <td>39.033333</td>\n",
       "      <td>...</td>\n",
       "      <td>7143.270665</td>\n",
       "      <td>3545.752359</td>\n",
       "      <td>60.053571</td>\n",
       "      <td>86.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>73.275862</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>76.533208</td>\n",
       "      <td>20994.655781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>State_2</td>\n",
       "      <td>7</td>\n",
       "      <td>61.983333</td>\n",
       "      <td>93.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>76.838710</td>\n",
       "      <td>93.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>1643.000000</td>\n",
       "      <td>38.048387</td>\n",
       "      <td>...</td>\n",
       "      <td>7024.015742</td>\n",
       "      <td>3964.509180</td>\n",
       "      <td>63.160714</td>\n",
       "      <td>85.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>71.016667</td>\n",
       "      <td>91.000000</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>21706.274470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>State_4</td>\n",
       "      <td>4</td>\n",
       "      <td>48.322409</td>\n",
       "      <td>79.560345</td>\n",
       "      <td>25.525862</td>\n",
       "      <td>73.318479</td>\n",
       "      <td>87.741379</td>\n",
       "      <td>51.599138</td>\n",
       "      <td>773.650862</td>\n",
       "      <td>31.904616</td>\n",
       "      <td>...</td>\n",
       "      <td>6352.453704</td>\n",
       "      <td>1903.521793</td>\n",
       "      <td>51.850042</td>\n",
       "      <td>76.969828</td>\n",
       "      <td>31.814655</td>\n",
       "      <td>63.705086</td>\n",
       "      <td>87.693966</td>\n",
       "      <td>40.663793</td>\n",
       "      <td>109.719905</td>\n",
       "      <td>17294.285563</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>State_4</td>\n",
       "      <td>5</td>\n",
       "      <td>48.730918</td>\n",
       "      <td>77.920330</td>\n",
       "      <td>27.189560</td>\n",
       "      <td>71.254342</td>\n",
       "      <td>92.853022</td>\n",
       "      <td>49.945742</td>\n",
       "      <td>771.043956</td>\n",
       "      <td>39.479418</td>\n",
       "      <td>...</td>\n",
       "      <td>6475.203019</td>\n",
       "      <td>2151.381838</td>\n",
       "      <td>54.010369</td>\n",
       "      <td>78.130495</td>\n",
       "      <td>29.073489</td>\n",
       "      <td>68.861699</td>\n",
       "      <td>92.466346</td>\n",
       "      <td>45.587912</td>\n",
       "      <td>100.612697</td>\n",
       "      <td>18172.453644</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>State_4</td>\n",
       "      <td>6</td>\n",
       "      <td>48.876348</td>\n",
       "      <td>86.949464</td>\n",
       "      <td>22.856049</td>\n",
       "      <td>77.282356</td>\n",
       "      <td>96.496172</td>\n",
       "      <td>53.933384</td>\n",
       "      <td>1319.392802</td>\n",
       "      <td>24.588623</td>\n",
       "      <td>...</td>\n",
       "      <td>6891.746472</td>\n",
       "      <td>2965.281432</td>\n",
       "      <td>57.517117</td>\n",
       "      <td>82.889357</td>\n",
       "      <td>33.966692</td>\n",
       "      <td>70.545300</td>\n",
       "      <td>95.170750</td>\n",
       "      <td>48.342649</td>\n",
       "      <td>90.891978</td>\n",
       "      <td>19368.469985</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>State_4</td>\n",
       "      <td>7</td>\n",
       "      <td>51.857143</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>69.274194</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>46.000000</td>\n",
       "      <td>761.000000</td>\n",
       "      <td>25.919355</td>\n",
       "      <td>...</td>\n",
       "      <td>6383.705500</td>\n",
       "      <td>3030.477850</td>\n",
       "      <td>56.112903</td>\n",
       "      <td>83.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>66.333333</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18481.270272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>State_6</td>\n",
       "      <td>1</td>\n",
       "      <td>54.300000</td>\n",
       "      <td>83.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>75.274194</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>59.000000</td>\n",
       "      <td>1331.000000</td>\n",
       "      <td>43.306452</td>\n",
       "      <td>...</td>\n",
       "      <td>6924.111100</td>\n",
       "      <td>3248.377500</td>\n",
       "      <td>57.096774</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>33.000000</td>\n",
       "      <td>70.033333</td>\n",
       "      <td>86.000000</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>98.764624</td>\n",
       "      <td>20647.308150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>State_6</td>\n",
       "      <td>2</td>\n",
       "      <td>54.766509</td>\n",
       "      <td>87.999163</td>\n",
       "      <td>37.999641</td>\n",
       "      <td>76.693384</td>\n",
       "      <td>90.999761</td>\n",
       "      <td>60.999641</td>\n",
       "      <td>1276.977403</td>\n",
       "      <td>41.467858</td>\n",
       "      <td>...</td>\n",
       "      <td>6885.901123</td>\n",
       "      <td>3881.200100</td>\n",
       "      <td>58.048632</td>\n",
       "      <td>77.999761</td>\n",
       "      <td>37.999880</td>\n",
       "      <td>68.799922</td>\n",
       "      <td>90.998804</td>\n",
       "      <td>52.999044</td>\n",
       "      <td>84.003879</td>\n",
       "      <td>20973.599715</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>State_6</td>\n",
       "      <td>3</td>\n",
       "      <td>52.982166</td>\n",
       "      <td>81.999036</td>\n",
       "      <td>32.998313</td>\n",
       "      <td>74.595458</td>\n",
       "      <td>89.998554</td>\n",
       "      <td>58.998433</td>\n",
       "      <td>1271.868884</td>\n",
       "      <td>38.547141</td>\n",
       "      <td>...</td>\n",
       "      <td>6826.606464</td>\n",
       "      <td>3285.388700</td>\n",
       "      <td>60.192248</td>\n",
       "      <td>85.998554</td>\n",
       "      <td>39.997710</td>\n",
       "      <td>67.865202</td>\n",
       "      <td>95.997831</td>\n",
       "      <td>46.998554</td>\n",
       "      <td>89.019186</td>\n",
       "      <td>20258.377302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>State_6</td>\n",
       "      <td>4</td>\n",
       "      <td>52.331945</td>\n",
       "      <td>76.998921</td>\n",
       "      <td>30.998813</td>\n",
       "      <td>74.514460</td>\n",
       "      <td>89.998597</td>\n",
       "      <td>60.997519</td>\n",
       "      <td>1127.845183</td>\n",
       "      <td>40.515033</td>\n",
       "      <td>...</td>\n",
       "      <td>6765.200797</td>\n",
       "      <td>2942.912872</td>\n",
       "      <td>59.627680</td>\n",
       "      <td>76.999676</td>\n",
       "      <td>41.997519</td>\n",
       "      <td>69.731870</td>\n",
       "      <td>91.998813</td>\n",
       "      <td>51.997519</td>\n",
       "      <td>91.177892</td>\n",
       "      <td>19905.006669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>State_6</td>\n",
       "      <td>5</td>\n",
       "      <td>54.248747</td>\n",
       "      <td>79.999113</td>\n",
       "      <td>31.998620</td>\n",
       "      <td>78.966389</td>\n",
       "      <td>94.998423</td>\n",
       "      <td>62.998423</td>\n",
       "      <td>1580.840824</td>\n",
       "      <td>50.773097</td>\n",
       "      <td>...</td>\n",
       "      <td>7126.149969</td>\n",
       "      <td>2776.723434</td>\n",
       "      <td>58.047333</td>\n",
       "      <td>77.999212</td>\n",
       "      <td>34.998620</td>\n",
       "      <td>74.448789</td>\n",
       "      <td>96.998817</td>\n",
       "      <td>55.997930</td>\n",
       "      <td>87.275406</td>\n",
       "      <td>20747.320307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>State_6</td>\n",
       "      <td>6</td>\n",
       "      <td>53.332196</td>\n",
       "      <td>81.999232</td>\n",
       "      <td>25.998464</td>\n",
       "      <td>79.175997</td>\n",
       "      <td>95.998617</td>\n",
       "      <td>60.998156</td>\n",
       "      <td>1488.848199</td>\n",
       "      <td>38.337505</td>\n",
       "      <td>...</td>\n",
       "      <td>7047.954140</td>\n",
       "      <td>3636.733764</td>\n",
       "      <td>58.756840</td>\n",
       "      <td>84.998924</td>\n",
       "      <td>37.998464</td>\n",
       "      <td>71.798625</td>\n",
       "      <td>90.998464</td>\n",
       "      <td>53.997388</td>\n",
       "      <td>84.749005</td>\n",
       "      <td>20907.957574</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>State_8</td>\n",
       "      <td>3</td>\n",
       "      <td>51.600000</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>71.741935</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>1081.000000</td>\n",
       "      <td>34.177419</td>\n",
       "      <td>...</td>\n",
       "      <td>6587.808250</td>\n",
       "      <td>3117.641550</td>\n",
       "      <td>56.806452</td>\n",
       "      <td>86.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>64.466667</td>\n",
       "      <td>93.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>92.077496</td>\n",
       "      <td>19387.723350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>State_8</td>\n",
       "      <td>4</td>\n",
       "      <td>54.629218</td>\n",
       "      <td>83.713580</td>\n",
       "      <td>31.997531</td>\n",
       "      <td>74.751892</td>\n",
       "      <td>91.713580</td>\n",
       "      <td>57.854321</td>\n",
       "      <td>1357.804938</td>\n",
       "      <td>40.800677</td>\n",
       "      <td>...</td>\n",
       "      <td>6855.124261</td>\n",
       "      <td>2978.433508</td>\n",
       "      <td>60.341458</td>\n",
       "      <td>79.856790</td>\n",
       "      <td>38.711111</td>\n",
       "      <td>70.629547</td>\n",
       "      <td>93.856790</td>\n",
       "      <td>49.567901</td>\n",
       "      <td>91.700065</td>\n",
       "      <td>20196.178929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>State_8</td>\n",
       "      <td>5</td>\n",
       "      <td>49.626389</td>\n",
       "      <td>79.052632</td>\n",
       "      <td>25.057018</td>\n",
       "      <td>71.275962</td>\n",
       "      <td>91.127193</td>\n",
       "      <td>41.780702</td>\n",
       "      <td>825.982456</td>\n",
       "      <td>43.155065</td>\n",
       "      <td>...</td>\n",
       "      <td>6526.890937</td>\n",
       "      <td>2299.216645</td>\n",
       "      <td>51.209111</td>\n",
       "      <td>78.000000</td>\n",
       "      <td>23.184211</td>\n",
       "      <td>68.752836</td>\n",
       "      <td>93.030702</td>\n",
       "      <td>44.236842</td>\n",
       "      <td>89.280948</td>\n",
       "      <td>18626.967816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>State_8</td>\n",
       "      <td>6</td>\n",
       "      <td>49.993096</td>\n",
       "      <td>81.263385</td>\n",
       "      <td>19.580311</td>\n",
       "      <td>75.819266</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>54.000000</td>\n",
       "      <td>1128.222798</td>\n",
       "      <td>35.226740</td>\n",
       "      <td>...</td>\n",
       "      <td>6787.545557</td>\n",
       "      <td>3309.888969</td>\n",
       "      <td>55.521450</td>\n",
       "      <td>84.473230</td>\n",
       "      <td>25.580311</td>\n",
       "      <td>68.450216</td>\n",
       "      <td>95.473230</td>\n",
       "      <td>41.580311</td>\n",
       "      <td>79.645753</td>\n",
       "      <td>19707.254042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>State_8</td>\n",
       "      <td>7</td>\n",
       "      <td>55.098753</td>\n",
       "      <td>85.192744</td>\n",
       "      <td>28.578231</td>\n",
       "      <td>70.867103</td>\n",
       "      <td>89.192744</td>\n",
       "      <td>52.513983</td>\n",
       "      <td>906.090703</td>\n",
       "      <td>30.327249</td>\n",
       "      <td>...</td>\n",
       "      <td>6516.309696</td>\n",
       "      <td>3243.817648</td>\n",
       "      <td>58.511411</td>\n",
       "      <td>83.064248</td>\n",
       "      <td>33.578231</td>\n",
       "      <td>66.583018</td>\n",
       "      <td>91.064248</td>\n",
       "      <td>40.835223</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19343.829364</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>37 rows × 46 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   state_factor  year_factor  april_avg_temp  april_max_temp  april_min_temp  \\\n",
       "0       State_1            1       55.500000       78.000000       41.000000   \n",
       "1       State_1            2       58.966416       89.877941       36.927941   \n",
       "2       State_1            3       60.217771       87.844605       36.847021   \n",
       "3       State_1            4       60.524645       90.264076       45.635210   \n",
       "4       State_1            5       60.966667       91.000000       41.000000   \n",
       "5       State_1            6       60.283333       86.000000       47.000000   \n",
       "6       State_1            7       58.450000       78.000000       45.000000   \n",
       "7      State_10            3       58.883333       91.000000       31.000000   \n",
       "8      State_10            4       57.333333       85.000000       34.000000   \n",
       "9      State_10            5       59.383333       84.000000       39.000000   \n",
       "10     State_10            6       56.866667       86.000000       30.000000   \n",
       "11     State_10            7       58.050000       87.000000       32.000000   \n",
       "12     State_11            5       51.822433       77.975556       36.975556   \n",
       "13     State_11            6       56.222177       88.929215       39.956535   \n",
       "14     State_11            7       50.754345       66.941248       36.941248   \n",
       "15      State_2            1       59.400000       89.000000       33.000000   \n",
       "16      State_2            2       55.650000       90.000000       34.000000   \n",
       "17      State_2            3       56.583333       91.000000       26.000000   \n",
       "18      State_2            4       54.450000       84.000000       31.000000   \n",
       "19      State_2            5       55.379310       81.000000       32.000000   \n",
       "20      State_2            6       54.870370       85.000000       28.000000   \n",
       "21      State_2            7       61.983333       93.000000       38.000000   \n",
       "22      State_4            4       48.322409       79.560345       25.525862   \n",
       "23      State_4            5       48.730918       77.920330       27.189560   \n",
       "24      State_4            6       48.876348       86.949464       22.856049   \n",
       "25      State_4            7       51.857143       81.000000       29.000000   \n",
       "26      State_6            1       54.300000       83.000000       35.000000   \n",
       "27      State_6            2       54.766509       87.999163       37.999641   \n",
       "28      State_6            3       52.982166       81.999036       32.998313   \n",
       "29      State_6            4       52.331945       76.998921       30.998813   \n",
       "30      State_6            5       54.248747       79.999113       31.998620   \n",
       "31      State_6            6       53.332196       81.999232       25.998464   \n",
       "32      State_8            3       51.600000       89.000000       24.000000   \n",
       "33      State_8            4       54.629218       83.713580       31.997531   \n",
       "34      State_8            5       49.626389       79.052632       25.057018   \n",
       "35      State_8            6       49.993096       81.263385       19.580311   \n",
       "36      State_8            7       55.098753       85.192744       28.578231   \n",
       "\n",
       "    august_avg_temp  august_max_temp  august_min_temp  cooling_degree_days  \\\n",
       "0         62.161290        85.000000        52.000000           115.000000   \n",
       "1         72.491224       100.123529        52.408824           872.120588   \n",
       "2         69.487742        94.402576        49.250403           640.805958   \n",
       "3         68.575825        83.245837        57.236320           707.289453   \n",
       "4         74.774194       104.000000        58.000000          1319.000000   \n",
       "5         63.790323        76.000000        53.000000           142.000000   \n",
       "6         65.854839        88.000000        53.000000           342.000000   \n",
       "7         77.080645        94.000000        60.000000          1719.000000   \n",
       "8         77.661290        96.000000        62.000000          1712.000000   \n",
       "9         79.274194        96.000000        63.000000          1999.000000   \n",
       "10        82.741935       101.000000        67.000000          1994.000000   \n",
       "11        71.290323        89.000000        55.000000           988.000000   \n",
       "12        68.391990        89.020370        50.991852           429.458164   \n",
       "13        68.562438        90.972058        51.970817           258.530270   \n",
       "14        69.685186        91.970624        52.931457           342.698898   \n",
       "15        76.951613        96.000000        57.000000          1631.000000   \n",
       "16        77.435484        95.000000        63.000000          1509.000000   \n",
       "17        73.783333        93.000000        55.000000          1306.000000   \n",
       "18        73.550000        89.000000        55.000000          1170.000000   \n",
       "19        75.816667        94.000000        59.000000          1451.000000   \n",
       "20        79.550000       100.000000        61.000000          1570.000000   \n",
       "21        76.838710        93.000000        60.000000          1643.000000   \n",
       "22        73.318479        87.741379        51.599138           773.650862   \n",
       "23        71.254342        92.853022        49.945742           771.043956   \n",
       "24        77.282356        96.496172        53.933384          1319.392802   \n",
       "25        69.274194        89.000000        46.000000           761.000000   \n",
       "26        75.274194        94.000000        59.000000          1331.000000   \n",
       "27        76.693384        90.999761        60.999641          1276.977403   \n",
       "28        74.595458        89.998554        58.998433          1271.868884   \n",
       "29        74.514460        89.998597        60.997519          1127.845183   \n",
       "30        78.966389        94.998423        62.998423          1580.840824   \n",
       "31        79.175997        95.998617        60.998156          1488.848199   \n",
       "32        71.741935        89.000000        51.000000          1081.000000   \n",
       "33        74.751892        91.713580        57.854321          1357.804938   \n",
       "34        71.275962        91.127193        41.780702           825.982456   \n",
       "35        75.819266        95.000000        54.000000          1128.222798   \n",
       "36        70.867103        89.192744        52.513983           906.090703   \n",
       "\n",
       "    december_avg_temp  ...  number_summer  number_winter  october_avg_temp  \\\n",
       "0           49.274194  ...    5823.496350    4654.161250         63.016129   \n",
       "1           48.139354  ...    6529.767616    4621.379011         64.253925   \n",
       "2           45.205850  ...    6336.734571    4476.664246         59.099922   \n",
       "3           56.148588  ...    6346.894880    5085.608160         68.316349   \n",
       "4           48.580645  ...    6848.484400    5089.055550         69.580645   \n",
       "5           50.661290  ...    5880.955150    5018.946600         63.354839   \n",
       "6           51.887097  ...    6139.393600    4860.146350         64.193548   \n",
       "7           42.290323  ...    7048.078850    3683.881700         62.403226   \n",
       "8           43.725806  ...    7085.464150    3390.677400         62.887097   \n",
       "9           51.177419  ...    7229.964750    3362.207250         58.854839   \n",
       "10          41.790323  ...    7411.088400    3861.399300         63.080645   \n",
       "11          34.467742  ...    6543.943850    3624.864950         59.266667   \n",
       "12          43.402432  ...    6135.914802    4330.436909         57.484953   \n",
       "13          38.471568  ...    6043.568649    4209.560561         55.346081   \n",
       "14          40.472243  ...    6185.074928    3798.125195         53.108086   \n",
       "15          41.816667  ...    7008.240150    3421.884050         55.306452   \n",
       "16          42.887097  ...    6987.869150    4028.213450         56.903226   \n",
       "17          39.500000  ...    6762.261842    3444.799250         58.816667   \n",
       "18          40.350000  ...    6711.364641    3014.856000         59.183333   \n",
       "19          48.850000  ...    6927.132878    2827.570172         55.767857   \n",
       "20          39.033333  ...    7143.270665    3545.752359         60.053571   \n",
       "21          38.048387  ...    7024.015742    3964.509180         63.160714   \n",
       "22          31.904616  ...    6352.453704    1903.521793         51.850042   \n",
       "23          39.479418  ...    6475.203019    2151.381838         54.010369   \n",
       "24          24.588623  ...    6891.746472    2965.281432         57.517117   \n",
       "25          25.919355  ...    6383.705500    3030.477850         56.112903   \n",
       "26          43.306452  ...    6924.111100    3248.377500         57.096774   \n",
       "27          41.467858  ...    6885.901123    3881.200100         58.048632   \n",
       "28          38.547141  ...    6826.606464    3285.388700         60.192248   \n",
       "29          40.515033  ...    6765.200797    2942.912872         59.627680   \n",
       "30          50.773097  ...    7126.149969    2776.723434         58.047333   \n",
       "31          38.337505  ...    7047.954140    3636.733764         58.756840   \n",
       "32          34.177419  ...    6587.808250    3117.641550         56.806452   \n",
       "33          40.800677  ...    6855.124261    2978.433508         60.341458   \n",
       "34          43.155065  ...    6526.890937    2299.216645         51.209111   \n",
       "35          35.226740  ...    6787.545557    3309.888969         55.521450   \n",
       "36          30.327249  ...    6516.309696    3243.817648         58.511411   \n",
       "\n",
       "    october_max_temp  october_min_temp  september_avg_temp  \\\n",
       "0          83.000000         47.000000           64.650000   \n",
       "1          99.764706         43.182353           69.176656   \n",
       "2          85.400966         38.419485           67.863899   \n",
       "3          94.915940         53.067407           69.414209   \n",
       "4          96.000000         49.000000           73.583333   \n",
       "5          84.000000         51.000000           64.800000   \n",
       "6          91.000000         48.000000           69.766667   \n",
       "7          91.000000         35.000000           71.316667   \n",
       "8          82.000000         44.000000           73.833333   \n",
       "9          79.000000         37.000000           74.866667   \n",
       "10         87.000000         41.000000           75.983333   \n",
       "11         83.000000         35.000000           66.450000   \n",
       "12         74.004074         45.983704           60.410519   \n",
       "13         67.933561         38.986961           61.328039   \n",
       "14         71.931457         35.960832           64.806283   \n",
       "15         75.000000         31.000000           69.833333   \n",
       "16         82.000000         34.000000           69.066667   \n",
       "17         88.000000         30.000000           67.160000   \n",
       "18         81.000000         38.000000           69.172414   \n",
       "19         80.000000         32.000000           72.448276   \n",
       "20         86.000000         35.000000           73.275862   \n",
       "21         85.000000         38.000000           71.016667   \n",
       "22         76.969828         31.814655           63.705086   \n",
       "23         78.130495         29.073489           68.861699   \n",
       "24         82.889357         33.966692           70.545300   \n",
       "25         83.000000         32.000000           66.333333   \n",
       "26         84.000000         33.000000           70.033333   \n",
       "27         77.999761         37.999880           68.799922   \n",
       "28         85.998554         39.997710           67.865202   \n",
       "29         76.999676         41.997519           69.731870   \n",
       "30         77.999212         34.998620           74.448789   \n",
       "31         84.998924         37.998464           71.798625   \n",
       "32         86.000000         28.000000           64.466667   \n",
       "33         79.856790         38.711111           70.629547   \n",
       "34         78.000000         23.184211           68.752836   \n",
       "35         84.473230         25.580311           68.450216   \n",
       "36         83.064248         33.578231           66.583018   \n",
       "\n",
       "    september_max_temp  september_min_temp    site_eui  total_number  \n",
       "0            90.000000           52.000000   70.148330  20836.972300  \n",
       "1            97.291176           49.491176   66.543553  22140.545271  \n",
       "2            96.642512           44.437198   60.261259  21513.691395  \n",
       "3            86.038858           55.398097   58.362694  22877.366604  \n",
       "4           105.000000           53.000000   63.259263  23167.388150  \n",
       "5            94.000000           52.000000   62.512324  21801.495450  \n",
       "6           104.000000           54.000000         NaN  21892.417400  \n",
       "7            94.000000           51.000000   93.232511  21509.816550  \n",
       "8            95.000000           54.000000   68.255412  21385.916899  \n",
       "9            96.000000           55.000000   61.224291  22008.001150  \n",
       "10           98.000000           59.000000   57.834971  22078.946449  \n",
       "11           88.000000           47.000000         NaN  20154.800564  \n",
       "12           80.004074           43.971482   53.484372  20332.190632  \n",
       "13           78.956535           43.978889   54.489394  20139.244590  \n",
       "14           90.000000           47.941248         NaN  19487.121402  \n",
       "15           90.000000           45.000000   81.831620  21104.728178  \n",
       "16           95.000000           49.000000   94.532905  21399.019800  \n",
       "17           93.000000           45.000000   80.064677  20457.926670  \n",
       "18           93.000000           46.000000   78.439699  20006.722339  \n",
       "19           94.000000           51.000000   77.675325  20614.355553  \n",
       "20           95.000000           52.000000   76.533208  20994.655781  \n",
       "21           91.000000           51.000000         NaN  21706.274470  \n",
       "22           87.693966           40.663793  109.719905  17294.285563  \n",
       "23           92.466346           45.587912  100.612697  18172.453644  \n",
       "24           95.170750           48.342649   90.891978  19368.469985  \n",
       "25           95.000000           40.000000         NaN  18481.270272  \n",
       "26           86.000000           51.000000   98.764624  20647.308150  \n",
       "27           90.998804           52.999044   84.003879  20973.599715  \n",
       "28           95.997831           46.998554   89.019186  20258.377302  \n",
       "29           91.998813           51.997519   91.177892  19905.006669  \n",
       "30           96.998817           55.997930   87.275406  20747.320307  \n",
       "31           90.998464           53.997388   84.749005  20907.957574  \n",
       "32           93.000000           40.000000   92.077496  19387.723350  \n",
       "33           93.856790           49.567901   91.700065  20196.178929  \n",
       "34           93.030702           44.236842   89.280948  18626.967816  \n",
       "35           95.473230           41.580311   79.645753  19707.254042  \n",
       "36           91.064248           40.835223         NaN  19343.829364  \n",
       "\n",
       "[37 rows x 46 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='total_number', ylabel='site_eui'>"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.scatterplot(data=table, x=\"total_number\", y=\"site_eui\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "# uitstoot per temperatuur\n",
    "table['avg_eui_totalNO'] = table.apply(lambda row: row.site_eui/row.total_number *100, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "# uitstoot per temperatuur\n",
    "df['avg_eui_totalNO'] = df.apply(lambda row: row.site_eui/row.total_number *100, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "def calc_difference_uitstoot_per_temp_tov_jaar_state(row):\n",
    "    return row.avg_eui_totalNO - table[(table['state_factor'] == row.state_factor) & (table['year_factor'] == row.year_factor)]['avg_eui_totalNO'].values[0]\n",
    "\n",
    "df['verschil_uitstoot_per_temp_tov_jaar_state']= df.apply(calc_difference_uitstoot_per_temp_tov_jaar_state, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='verschil_uitstoot_per_temp_tov_jaar_state', ylabel='Count'>"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.histplot(data=df, x=\"verschil_uitstoot_per_temp_tov_jaar_state\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_verschil_uitstoot_per_temp_tov_jaar_state_from_building(row):\n",
    "    if row.verschil_uitstoot_per_temp_tov_jaar_state == row.verschil_uitstoot_per_temp_tov_jaar_state:\n",
    "        return row.verschil_uitstoot_per_temp_tov_jaar_state\n",
    "    else:\n",
    "        temp_df = df[df['building_id'] == row.building_id]\n",
    "        temp_df_list = temp_df['verschil_uitstoot_per_temp_tov_jaar_state'].to_list()\n",
    "        if len(temp_df_list) > 1:\n",
    "            return temp_df_list[-2]\n",
    "        else:\n",
    "            return 0\n",
    "\n",
    "df['verschil_uitstoot_per_temp_tov_jaar_state'] = df.apply(get_verschil_uitstoot_per_temp_tov_jaar_state_from_building, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['verschil_uitstoot_per_temp_tov_jaar_state'] = df['verschil_uitstoot_per_temp_tov_jaar_state'].fillna(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "del df['avg_eui_totalNO']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.16 StateYear_E_Int"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'State_1': [-0.14397914926889885, 0.044751209585454865, 0.1887303588543537],\n",
       " 'State_2': [-0.11878063777956323, 0.04097145627061849, 0.1597520940501817],\n",
       " 'State_4': [-0.16076011020040304, 0.0422102516561606, 0.20297036185656364],\n",
       " 'State_6': [-0.11509766097313048, 0.06447219601445692, 0.1795698569875874],\n",
       " 'State_8': [-0.20803747732214825, 0.0653155080397419, 0.2733529853618901],\n",
       " 'State_10': [-0.04954064622170248, 0.023568890988139468, 0.07310953720984195],\n",
       " 'State_11': [-0.13934786247281813, 0.018683777384450595, 0.15803163985726873]}"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "state_dict ={}\n",
    "def create_feature():\n",
    "    for x in df['state_factor'].unique():\n",
    "        temp_df = df[df['state_factor'] == x]\n",
    "        temp_df_list = temp_df['verschil_uitstoot_per_temp_tov_jaar_state'].to_list()\n",
    "        temp_df_list.sort()\n",
    "        p1 = np.nanpercentile(temp_df_list, 25, axis=0)\n",
    "        p3 = np.nanpercentile(temp_df_list, 75, axis=0)\n",
    "        InterQ = abs(p3-p1)\n",
    "        state_dict[x] = [p1,p3,InterQ]\n",
    "\n",
    "create_feature()\n",
    "state_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['StateYear_E_Int'] = df.apply(lambda row: 1 if row.verschil_uitstoot_per_temp_tov_jaar_state < (state_dict[row.state_factor][0]-state_dict[row.state_factor][2]) else\n",
    "                                3 if row.verschil_uitstoot_per_temp_tov_jaar_state > (1.5* state_dict[row.state_factor][2] + state_dict[row.state_factor][1]) else \n",
    "                                 2 if (row.verschil_uitstoot_per_temp_tov_jaar_state > (state_dict[row.state_factor][0]-state_dict[row.state_factor][2])) & (row.verschil_uitstoot_per_temp_tov_jaar_state < (1.5* state_dict[row.state_factor][2] + state_dict[row.state_factor][1])) else np.nan \n",
    "                                 , axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_StateYear_E_Int_from_building(row):\n",
    "    if row.StateYear_E_Int == row.StateYear_E_Int:\n",
    "        return row.StateYear_E_Int\n",
    "    else:\n",
    "        temp_df = df[df['building_id'] == row.building_id]\n",
    "        temp_df_list = temp_df['StateYear_E_Int'].to_list()\n",
    "        if len(temp_df_list) > 1:\n",
    "            return temp_df_list[-2]\n",
    "        else:\n",
    "            return 2\n",
    "\n",
    "df['StateYear_E_Int'] = df.apply(get_StateYear_E_Int_from_building, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['StateYear_E_Int'] = df['StateYear_E_Int'].fillna(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 500x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "correlation = df[['StateYear_E_Int','verschil_uitstoot_per_temp_tov_jaar_state','site_eui']].corr(method=\"pearson\")\n",
    "mask = np.triu(np.ones_like(correlation, dtype=bool))\n",
    "fig = plt.figure(figsize=(5,2), dpi=100, facecolor = 'w', edgecolor=\"k\")\n",
    "figure = sns.heatmap(correlation, mask=mask, cmap= \"Spectral\", center=0, linewidths=1, fmt=\".2f\",annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "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>year_factor</th>\n",
       "      <th>state_factor</th>\n",
       "      <th>building_class</th>\n",
       "      <th>facility_type</th>\n",
       "      <th>floor_area</th>\n",
       "      <th>elevation</th>\n",
       "      <th>january_min_temp</th>\n",
       "      <th>january_avg_temp</th>\n",
       "      <th>january_max_temp</th>\n",
       "      <th>february_min_temp</th>\n",
       "      <th>...</th>\n",
       "      <th>avg_total_number_building</th>\n",
       "      <th>avg_eui_building</th>\n",
       "      <th>year_built_bin</th>\n",
       "      <th>yearB_E_int</th>\n",
       "      <th>calculated_eui</th>\n",
       "      <th>calc_eui_perc_off</th>\n",
       "      <th>binned_calc_eui</th>\n",
       "      <th>binned_calc_eui_20</th>\n",
       "      <th>verschil_uitstoot_per_temp_tov_jaar_state</th>\n",
       "      <th>StateYear_E_Int</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Grocery_store_or_food_market</td>\n",
       "      <td>61242.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>22078.118536</td>\n",
       "      <td>245.522877</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>231.720532</td>\n",
       "      <td>-0.068208</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>0.856815</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Distribution_or_Shipping_center</td>\n",
       "      <td>274000.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>22078.118536</td>\n",
       "      <td>27.609132</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>26.057054</td>\n",
       "      <td>-0.016720</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.209475</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Retail_Enclosed_mall</td>\n",
       "      <td>280025.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>22068.147067</td>\n",
       "      <td>23.481589</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>22.171559</td>\n",
       "      <td>-0.102134</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.218145</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Education_Other_classroom</td>\n",
       "      <td>55325.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>22050.354236</td>\n",
       "      <td>43.272466</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>40.891278</td>\n",
       "      <td>-0.155260</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>-0.104341</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>State_1</td>\n",
       "      <td>Commercial</td>\n",
       "      <td>Warehouse_Nonrefrigerated</td>\n",
       "      <td>66000.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>36</td>\n",
       "      <td>50.5</td>\n",
       "      <td>68</td>\n",
       "      <td>35</td>\n",
       "      <td>...</td>\n",
       "      <td>22109.068725</td>\n",
       "      <td>5.196438</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>4.897449</td>\n",
       "      <td>0.255951</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.317939</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 108 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_factor state_factor building_class  \\\n",
       "0            1      State_1     Commercial   \n",
       "1            1      State_1     Commercial   \n",
       "2            1      State_1     Commercial   \n",
       "3            1      State_1     Commercial   \n",
       "4            1      State_1     Commercial   \n",
       "\n",
       "                               facility_type  floor_area  elevation  \\\n",
       "0               Grocery_store_or_food_market     61242.0        2.4   \n",
       "1  Warehouse_Distribution_or_Shipping_center    274000.0        1.8   \n",
       "2                       Retail_Enclosed_mall    280025.0        1.8   \n",
       "3                  Education_Other_classroom     55325.0        1.8   \n",
       "4                  Warehouse_Nonrefrigerated     66000.0        2.4   \n",
       "\n",
       "   january_min_temp  january_avg_temp  january_max_temp  february_min_temp  \\\n",
       "0                36              50.5                68                 35   \n",
       "1                36              50.5                68                 35   \n",
       "2                36              50.5                68                 35   \n",
       "3                36              50.5                68                 35   \n",
       "4                36              50.5                68                 35   \n",
       "\n",
       "   ...  avg_total_number_building  avg_eui_building  year_built_bin  \\\n",
       "0  ...               22078.118536        245.522877               2   \n",
       "1  ...               22078.118536         27.609132               3   \n",
       "2  ...               22068.147067         23.481589               3   \n",
       "3  ...               22050.354236         43.272466               4   \n",
       "4  ...               22109.068725          5.196438               4   \n",
       "\n",
       "   yearB_E_int  calculated_eui  calc_eui_perc_off  binned_calc_eui  \\\n",
       "0            3      231.720532          -0.068208               10   \n",
       "1            2       26.057054          -0.016720                1   \n",
       "2            2       22.171559          -0.102134                1   \n",
       "3            2       40.891278          -0.155260                2   \n",
       "4            2        4.897449           0.255951                1   \n",
       "\n",
       "   binned_calc_eui_20  verschil_uitstoot_per_temp_tov_jaar_state  \\\n",
       "0                  10                                   0.856815   \n",
       "1                   1                                  -0.209475   \n",
       "2                   1                                  -0.218145   \n",
       "3                   2                                  -0.104341   \n",
       "4                   1                                  -0.317939   \n",
       "\n",
       "   StateYear_E_Int  \n",
       "0                3  \n",
       "1                2  \n",
       "2                2  \n",
       "3                2  \n",
       "4                2  \n",
       "\n",
       "[5 rows x 108 columns]"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "site_eui              9705\n",
       "avg_eui_building       716\n",
       "year_factor              0\n",
       "max_SW                   0\n",
       "number_february          0\n",
       "                      ... \n",
       "september_max_temp       0\n",
       "september_avg_temp       0\n",
       "september_min_temp       0\n",
       "august_max_temp          0\n",
       "StateYear_E_Int          0\n",
       "Length: 108, dtype: int64"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv(\"C:/Users/Meyke/WiDS/data/features_data_0402_imputed_star_rating_and_generous_building_id.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "application/vnd.databricks.v1+notebook": {
   "dashboards": [],
   "language": "python",
   "notebookMetadata": {
    "pythonIndentUnit": 2
   },
   "notebookName": "feature_engineering",
   "notebookOrigID": 4075831666413441,
   "widgets": {}
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}