Udacity_capstone.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Udacity Data Scientist Capstone\n",
    "## Kaggle Competition: House Prices - Advanced Regression Techniques\n",
    "\n",
    "by Umrbek Allakulov"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Table of contents:\n",
    "* [Introduction](#first-bullet)\n",
    "* [Data preprocessing and descriptive statistics](#second-bullet)\n",
    "     - [Preprocessing and descriptive statistics of categorical attributes](#third-bullet)\n",
    "     - [Preprocessing and descriptive statistics of numerical attributes](#fourth-bullet)\n",
    "         - [Continuous attributes](#5-bullet)\n",
    "         - [Ordinal attributes](#6-bullet)\n",
    "         - [Descriptive statistics of numerical attributes](#7-bullet)\n",
    "* [Prepararation of datasets for modelling](#9-bullet)\n",
    "* [Unsupervised learning](#10-bullet)\n",
    "    - [Principal component analysis](#11-bullet)\n",
    "    - [KMeans clustering](#13-bullet)\n",
    "- [Supervised learning](#14-bullet)\n",
    "    - [Feature selection](#15-bullet)\n",
    "    - [Tune Hyperparameters with Bayes Search](#17-bullet)\n",
    "        - [Search strategy 1](#18-bullet)\n",
    "        - [Search strategy 2](#19-bullet)\n",
    "    - [Evaluation of feature importances](#20-bullet)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introduction <a class=\"anchor\" id=\"first-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This note book contains all the code and analysis, and is submitted as part of the Capstone Project for the Data Scientist Nanodegree. The project uses datasets from a Kaggle Competition titled \"House Prices - Advanced Regression Techniques\". <br>\n",
    "\n",
    "\n",
    "**Data descriptions as provided in the Kaggle competition webpage** <br>\n",
    "\n",
    "The Ames Housing dataset was compiled by Dean De Cock for use in data science education. It's an incredible alternative for data scientists looking for a modernized and expanded version of the often cited Boston Housing dataset. \n",
    "\n",
    "train.csv - the training set <br>\n",
    "test.csv - the test set <br>\n",
    "data_description.txt - full description of each column, originally prepared by Dean De Cock but lightly edited to match the column names used here <br>\n",
    "sample_submission.csv - a benchmark submission from a linear regression on year and month of sale, lot square footage, and number of bedrooms"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# install modules\n",
    "!pip install plotly\n",
    "!pip install --upgrade setuptools\n",
    "!pip install --upgrade pip\n",
    "!pip install xgboost\n",
    "!pip install scikit-optimize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1365,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import libraries\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# magic word for producing visualizations in notebook\n",
    "%matplotlib inline\n",
    "\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn import tree\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.impute import KNNImputer\n",
    "from numpy import mean\n",
    "from numpy import std\n",
    "from sklearn.datasets import make_classification\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import OrdinalEncoder\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "from matplotlib import pyplot\n",
    "from plotly.offline import init_notebook_mode, iplot\n",
    "from plotly.graph_objs import *\n",
    "import plotly.offline as py\n",
    "from scipy.stats import uniform, truncnorm, randint\n",
    "import warnings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1101,
   "metadata": {},
   "outputs": [],
   "source": [
    "#import datasets\n",
    "df_train = pd.read_csv(r\"train.csv\")\n",
    "df_test = pd.read_csv(r\"test.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1102,
   "metadata": {},
   "outputs": [],
   "source": [
    "# replace NAs with np.nan\n",
    "\n",
    "def fix_na(df):\n",
    "    '''\n",
    "    Replaces NA's with np.nan\n",
    "    \n",
    "    Args:\n",
    "    df - DataFrame \n",
    "    \n",
    "    Returns: \n",
    "    df - DataFrame where NA's are replace with np.nan\n",
    "    '''\n",
    "    for col in categorical_col:\n",
    "        df[col] = df[col].replace(\"NA\", np.nan)\n",
    "    return df\n",
    "\n",
    "df_train = fix_na(df_train)\n",
    "df_test = fix_na(df_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1793,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1460, 81), (1459, 80))"
      ]
     },
     "execution_count": 1793,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check the dimensions of the datasets\n",
    "df_train.shape, df_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1543,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Attribute</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Type</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>int64</th>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>float64</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>object</th>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Attribute\n",
       "Type              \n",
       "int64           35\n",
       "float64          3\n",
       "object          43"
      ]
     },
     "execution_count": 1543,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#count data types\n",
    "pd.options.display.max_rows = 2000\n",
    "types_df = pd.DataFrame(df_train.dtypes).reset_index()\n",
    "types_df.columns = ['Attribute', 'Type']\n",
    "types_df.groupby('Type').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1012,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSSubClass: Identifies the type of dwelling involved in the sale.\t\n",
      "\n",
      "        20\t1-STORY 1946 & NEWER ALL STYLES\n",
      "        30\t1-STORY 1945 & OLDER\n",
      "        40\t1-STORY W/FINISHED ATTIC ALL AGES\n",
      "        45\t1-1/2 STORY - UNFINISHED ALL AGES\n",
      "        50\t1-1/2 STORY FINISHED ALL AGES\n",
      "        60\t2-STORY 1946 & NEWER\n",
      "        70\t2-STORY 1945 & OLDER\n",
      "        75\t2-1/2 STORY ALL AGES\n",
      "        80\tSPLIT OR MULTI-LEVEL\n",
      "        85\tSPLIT FOYER\n",
      "        90\tDUPLEX - ALL STYLES AND AGES\n",
      "       120\t1-STORY PUD (Planned Unit Development) - 1946 & NEWER\n",
      "       150\t1-1/2 STORY PUD - ALL AGES\n",
      "       160\t2-STORY PUD - 1946 & NEWER\n",
      "       180\tPUD - MULTILEVEL - INCL SPLIT LEV/FOYER\n",
      "       190\t2 FAMILY CONVERSION - ALL STYLES AND AGES\n",
      "\n",
      "MSZoning: Identifies the general zoning classification of the sale.\n",
      "\t\t\n",
      "       A\tAgriculture\n",
      "       C\tCommercial\n",
      "       FV\tFloating Village Residential\n",
      "       I\tIndustrial\n",
      "       RH\tResidential High Density\n",
      "       RL\tResidential Low Density\n",
      "       RP\tResidential Low Density Park \n",
      "       RM\tResidential Medium Density\n",
      "\t\n",
      "LotFrontage: Linear feet of street connected to property\n",
      "\n",
      "LotArea: Lot size in square feet\n",
      "\n",
      "Street: Type of road access to property\n",
      "\n",
      "       Grvl\tGravel\t\n",
      "       Pave\tPaved\n",
      "       \t\n",
      "Alley: Type of alley access to property\n",
      "\n",
      "       Grvl\tGravel\n",
      "       Pave\tPaved\n",
      "       NA \tNo alley access\n",
      "\t\t\n",
      "LotShape: General shape of property\n",
      "\n",
      "       Reg\tRegular\t\n",
      "       IR1\tSlightly irregular\n",
      "       IR2\tModerately Irregular\n",
      "       IR3\tIrregular\n",
      "       \n",
      "LandContour: Flatness of the property\n",
      "\n",
      "       Lvl\tNear Flat/Level\t\n",
      "       Bnk\tBanked - Quick and significant rise from street grade to building\n",
      "       HLS\tHillside - Significant slope from side to side\n",
      "       Low\tDepression\n",
      "\t\t\n",
      "Utilities: Type of utilities available\n",
      "\t\t\n",
      "       AllPub\tAll public Utilities (E,G,W,& S)\t\n",
      "       NoSewr\tElectricity, Gas, and Water (Septic Tank)\n",
      "       NoSeWa\tElectricity and Gas Only\n",
      "       ELO\tElectricity only\t\n",
      "\t\n",
      "LotConfig: Lot configuration\n",
      "\n",
      "       Inside\tInside lot\n",
      "       Corner\tCorner lot\n",
      "       CulDSac\tCul-de-sac\n",
      "       FR2\tFrontage on 2 sides of property\n",
      "       FR3\tFrontage on 3 sides of property\n",
      "\t\n",
      "LandSlope: Slope of property\n",
      "\t\t\n",
      "       Gtl\tGentle slope\n",
      "       Mod\tModerate Slope\t\n",
      "       Sev\tSevere Slope\n",
      "\t\n",
      "Neighborhood: Physical locations within Ames city limits\n",
      "\n",
      "       Blmngtn\tBloomington Heights\n",
      "       Blueste\tBluestem\n",
      "       BrDale\tBriardale\n",
      "       BrkSide\tBrookside\n",
      "       ClearCr\tClear Creek\n",
      "       CollgCr\tCollege Creek\n",
      "       Crawfor\tCrawford\n",
      "       Edwards\tEdwards\n",
      "       Gilbert\tGilbert\n",
      "       IDOTRR\tIowa DOT and Rail Road\n",
      "       MeadowV\tMeadow Village\n",
      "       Mitchel\tMitchell\n",
      "       Names\tNorth Ames\n",
      "       NoRidge\tNorthridge\n",
      "       NPkVill\tNorthpark Villa\n",
      "       NridgHt\tNorthridge Heights\n",
      "       NWAmes\tNorthwest Ames\n",
      "       OldTown\tOld Town\n",
      "       SWISU\tSouth & West of Iowa State University\n",
      "       Sawyer\tSawyer\n",
      "       SawyerW\tSawyer West\n",
      "       Somerst\tSomerset\n",
      "       StoneBr\tStone Brook\n",
      "       Timber\tTimberland\n",
      "       Veenker\tVeenker\n",
      "\t\t\t\n",
      "Condition1: Proximity to various conditions\n",
      "\t\n",
      "       Artery\tAdjacent to arterial street\n",
      "       Feedr\tAdjacent to feeder street\t\n",
      "       Norm\tNormal\t\n",
      "       RRNn\tWithin 200' of North-South Railroad\n",
      "       RRAn\tAdjacent to North-South Railroad\n",
      "       PosN\tNear positive off-site feature--park, greenbelt, etc.\n",
      "       PosA\tAdjacent to postive off-site feature\n",
      "       RRNe\tWithin 200' of East-West Railroad\n",
      "       RRAe\tAdjacent to East-West Railroad\n",
      "\t\n",
      "Condition2: Proximity to various conditions (if more than one is present)\n",
      "\t\t\n",
      "       Artery\tAdjacent to arterial street\n",
      "       Feedr\tAdjacent to feeder street\t\n",
      "       Norm\tNormal\t\n",
      "       RRNn\tWithin 200' of North-South Railroad\n",
      "       RRAn\tAdjacent to North-South Railroad\n",
      "       PosN\tNear positive off-site feature--park, greenbelt, etc.\n",
      "       PosA\tAdjacent to postive off-site feature\n",
      "       RRNe\tWithin 200' of East-West Railroad\n",
      "       RRAe\tAdjacent to East-West Railroad\n",
      "\t\n",
      "BldgType: Type of dwelling\n",
      "\t\t\n",
      "       1Fam\tSingle-family Detached\t\n",
      "       2FmCon\tTwo-family Conversion; originally built as one-family dwelling\n",
      "       Duplx\tDuplex\n",
      "       TwnhsE\tTownhouse End Unit\n",
      "       TwnhsI\tTownhouse Inside Unit\n",
      "\t\n",
      "HouseStyle: Style of dwelling\n",
      "\t\n",
      "       1Story\tOne story\n",
      "       1.5Fin\tOne and one-half story: 2nd level finished\n",
      "       1.5Unf\tOne and one-half story: 2nd level unfinished\n",
      "       2Story\tTwo story\n",
      "       2.5Fin\tTwo and one-half story: 2nd level finished\n",
      "       2.5Unf\tTwo and one-half story: 2nd level unfinished\n",
      "       SFoyer\tSplit Foyer\n",
      "       SLvl\tSplit Level\n",
      "\t\n",
      "OverallQual: Rates the overall material and finish of the house\n",
      "\n",
      "       10\tVery Excellent\n",
      "       9\tExcellent\n",
      "       8\tVery Good\n",
      "       7\tGood\n",
      "       6\tAbove Average\n",
      "       5\tAverage\n",
      "       4\tBelow Average\n",
      "       3\tFair\n",
      "       2\tPoor\n",
      "       1\tVery Poor\n",
      "\t\n",
      "OverallCond: Rates the overall condition of the house\n",
      "\n",
      "       10\tVery Excellent\n",
      "       9\tExcellent\n",
      "       8\tVery Good\n",
      "       7\tGood\n",
      "       6\tAbove Average\t\n",
      "       5\tAverage\n",
      "       4\tBelow Average\t\n",
      "       3\tFair\n",
      "       2\tPoor\n",
      "       1\tVery Poor\n",
      "\t\t\n",
      "YearBuilt: Original construction date\n",
      "\n",
      "YearRemodAdd: Remodel date (same as construction date if no remodeling or additions)\n",
      "\n",
      "RoofStyle: Type of roof\n",
      "\n",
      "       Flat\tFlat\n",
      "       Gable\tGable\n",
      "       Gambrel\tGabrel (Barn)\n",
      "       Hip\tHip\n",
      "       Mansard\tMansard\n",
      "       Shed\tShed\n",
      "\t\t\n",
      "RoofMatl: Roof material\n",
      "\n",
      "       ClyTile\tClay or Tile\n",
      "       CompShg\tStandard (Composite) Shingle\n",
      "       Membran\tMembrane\n",
      "       Metal\tMetal\n",
      "       Roll\tRoll\n",
      "       Tar&Grv\tGravel & Tar\n",
      "       WdShake\tWood Shakes\n",
      "       WdShngl\tWood Shingles\n",
      "\t\t\n",
      "Exterior1st: Exterior covering on house\n",
      "\n",
      "       AsbShng\tAsbestos Shingles\n",
      "       AsphShn\tAsphalt Shingles\n",
      "       BrkComm\tBrick Common\n",
      "       BrkFace\tBrick Face\n",
      "       CBlock\tCinder Block\n",
      "       CemntBd\tCement Board\n",
      "       HdBoard\tHard Board\n",
      "       ImStucc\tImitation Stucco\n",
      "       MetalSd\tMetal Siding\n",
      "       Other\tOther\n",
      "       Plywood\tPlywood\n",
      "       PreCast\tPreCast\t\n",
      "       Stone\tStone\n",
      "       Stucco\tStucco\n",
      "       VinylSd\tVinyl Siding\n",
      "       Wd Sdng\tWood Siding\n",
      "       WdShing\tWood Shingles\n",
      "\t\n",
      "Exterior2nd: Exterior covering on house (if more than one material)\n",
      "\n",
      "       AsbShng\tAsbestos Shingles\n",
      "       AsphShn\tAsphalt Shingles\n",
      "       BrkComm\tBrick Common\n",
      "       BrkFace\tBrick Face\n",
      "       CBlock\tCinder Block\n",
      "       CemntBd\tCement Board\n",
      "       HdBoard\tHard Board\n",
      "       ImStucc\tImitation Stucco\n",
      "       MetalSd\tMetal Siding\n",
      "       Other\tOther\n",
      "       Plywood\tPlywood\n",
      "       PreCast\tPreCast\n",
      "       Stone\tStone\n",
      "       Stucco\tStucco\n",
      "       VinylSd\tVinyl Siding\n",
      "       Wd Sdng\tWood Siding\n",
      "       WdShing\tWood Shingles\n",
      "\t\n",
      "MasVnrType: Masonry veneer type\n",
      "\n",
      "       BrkCmn\tBrick Common\n",
      "       BrkFace\tBrick Face\n",
      "       CBlock\tCinder Block\n",
      "       None\tNone\n",
      "       Stone\tStone\n",
      "\t\n",
      "MasVnrArea: Masonry veneer area in square feet\n",
      "\n",
      "ExterQual: Evaluates the quality of the material on the exterior \n",
      "\t\t\n",
      "       Ex\tExcellent\n",
      "       Gd\tGood\n",
      "       TA\tAverage/Typical\n",
      "       Fa\tFair\n",
      "       Po\tPoor\n",
      "\t\t\n",
      "ExterCond: Evaluates the present condition of the material on the exterior\n",
      "\t\t\n",
      "       Ex\tExcellent\n",
      "       Gd\tGood\n",
      "       TA\tAverage/Typical\n",
      "       Fa\tFair\n",
      "       Po\tPoor\n",
      "\t\t\n",
      "Foundation: Type of foundation\n",
      "\t\t\n",
      "       BrkTil\tBrick & Tile\n",
      "       CBlock\tCinder Block\n",
      "       PConc\tPoured Contrete\t\n",
      "       Slab\tSlab\n",
      "       Stone\tStone\n",
      "       Wood\tWood\n",
      "\t\t\n",
      "BsmtQual: Evaluates the height of the basement\n",
      "\n",
      "       Ex\tExcellent (100+ inches)\t\n",
      "       Gd\tGood (90-99 inches)\n",
      "       TA\tTypical (80-89 inches)\n",
      "       Fa\tFair (70-79 inches)\n",
      "       Po\tPoor (<70 inches\n",
      "       NA\tNo Basement\n",
      "\t\t\n",
      "BsmtCond: Evaluates the general condition of the basement\n",
      "\n",
      "       Ex\tExcellent\n",
      "       Gd\tGood\n",
      "       TA\tTypical - slight dampness allowed\n",
      "       Fa\tFair - dampness or some cracking or settling\n",
      "       Po\tPoor - Severe cracking, settling, or wetness\n",
      "       NA\tNo Basement\n",
      "\t\n",
      "BsmtExposure: Refers to walkout or garden level walls\n",
      "\n",
      "       Gd\tGood Exposure\n",
      "       Av\tAverage Exposure (split levels or foyers typically score average or above)\t\n",
      "       Mn\tMimimum Exposure\n",
      "       No\tNo Exposure\n",
      "       NA\tNo Basement\n",
      "\t\n",
      "BsmtFinType1: Rating of basement finished area\n",
      "\n",
      "       GLQ\tGood Living Quarters\n",
      "       ALQ\tAverage Living Quarters\n",
      "       BLQ\tBelow Average Living Quarters\t\n",
      "       Rec\tAverage Rec Room\n",
      "       LwQ\tLow Quality\n",
      "       Unf\tUnfinshed\n",
      "       NA\tNo Basement\n",
      "\t\t\n",
      "BsmtFinSF1: Type 1 finished square feet\n",
      "\n",
      "BsmtFinType2: Rating of basement finished area (if multiple types)\n",
      "\n",
      "       GLQ\tGood Living Quarters\n",
      "       ALQ\tAverage Living Quarters\n",
      "       BLQ\tBelow Average Living Quarters\t\n",
      "       Rec\tAverage Rec Room\n",
      "       LwQ\tLow Quality\n",
      "       Unf\tUnfinshed\n",
      "       NA\tNo Basement\n",
      "\n",
      "BsmtFinSF2: Type 2 finished square feet\n",
      "\n",
      "BsmtUnfSF: Unfinished square feet of basement area\n",
      "\n",
      "TotalBsmtSF: Total square feet of basement area\n",
      "\n",
      "Heating: Type of heating\n",
      "\t\t\n",
      "       Floor\tFloor Furnace\n",
      "       GasA\tGas forced warm air furnace\n",
      "       GasW\tGas hot water or steam heat\n",
      "       Grav\tGravity furnace\t\n",
      "       OthW\tHot water or steam heat other than gas\n",
      "       Wall\tWall furnace\n",
      "\t\t\n",
      "HeatingQC: Heating quality and condition\n",
      "\n",
      "       Ex\tExcellent\n",
      "       Gd\tGood\n",
      "       TA\tAverage/Typical\n",
      "       Fa\tFair\n",
      "       Po\tPoor\n",
      "\t\t\n",
      "CentralAir: Central air conditioning\n",
      "\n",
      "       N\tNo\n",
      "       Y\tYes\n",
      "\t\t\n",
      "Electrical: Electrical system\n",
      "\n",
      "       SBrkr\tStandard Circuit Breakers & Romex\n",
      "       FuseA\tFuse Box over 60 AMP and all Romex wiring (Average)\t\n",
      "       FuseF\t60 AMP Fuse Box and mostly Romex wiring (Fair)\n",
      "       FuseP\t60 AMP Fuse Box and mostly knob & tube wiring (poor)\n",
      "       Mix\tMixed\n",
      "\t\t\n",
      "1stFlrSF: First Floor square feet\n",
      " \n",
      "2ndFlrSF: Second floor square feet\n",
      "\n",
      "LowQualFinSF: Low quality finished square feet (all floors)\n",
      "\n",
      "GrLivArea: Above grade (ground) living area square feet\n",
      "\n",
      "BsmtFullBath: Basement full bathrooms\n",
      "\n",
      "BsmtHalfBath: Basement half bathrooms\n",
      "\n",
      "FullBath: Full bathrooms above grade\n",
      "\n",
      "HalfBath: Half baths above grade\n",
      "\n",
      "Bedroom: Bedrooms above grade (does NOT include basement bedrooms)\n",
      "\n",
      "Kitchen: Kitchens above grade\n",
      "\n",
      "KitchenQual: Kitchen quality\n",
      "\n",
      "       Ex\tExcellent\n",
      "       Gd\tGood\n",
      "       TA\tTypical/Average\n",
      "       Fa\tFair\n",
      "       Po\tPoor\n",
      "       \t\n",
      "TotRmsAbvGrd: Total rooms above grade (does not include bathrooms)\n",
      "\n",
      "Functional: Home functionality (Assume typical unless deductions are warranted)\n",
      "\n",
      "       Typ\tTypical Functionality\n",
      "       Min1\tMinor Deductions 1\n",
      "       Min2\tMinor Deductions 2\n",
      "       Mod\tModerate Deductions\n",
      "       Maj1\tMajor Deductions 1\n",
      "       Maj2\tMajor Deductions 2\n",
      "       Sev\tSeverely Damaged\n",
      "       Sal\tSalvage only\n",
      "\t\t\n",
      "Fireplaces: Number of fireplaces\n",
      "\n",
      "FireplaceQu: Fireplace quality\n",
      "\n",
      "       Ex\tExcellent - Exceptional Masonry Fireplace\n",
      "       Gd\tGood - Masonry Fireplace in main level\n",
      "       TA\tAverage - Prefabricated Fireplace in main living area or Masonry Fireplace in basement\n",
      "       Fa\tFair - Prefabricated Fireplace in basement\n",
      "       Po\tPoor - Ben Franklin Stove\n",
      "       NA\tNo Fireplace\n",
      "\t\t\n",
      "GarageType: Garage location\n",
      "\t\t\n",
      "       2Types\tMore than one type of garage\n",
      "       Attchd\tAttached to home\n",
      "       Basment\tBasement Garage\n",
      "       BuiltIn\tBuilt-In (Garage part of house - typically has room above garage)\n",
      "       CarPort\tCar Port\n",
      "       Detchd\tDetached from home\n",
      "       NA\tNo Garage\n",
      "\t\t\n",
      "GarageYrBlt: Year garage was built\n",
      "\t\t\n",
      "GarageFinish: Interior finish of the garage\n",
      "\n",
      "       Fin\tFinished\n",
      "       RFn\tRough Finished\t\n",
      "       Unf\tUnfinished\n",
      "       NA\tNo Garage\n",
      "\t\t\n",
      "GarageCars: Size of garage in car capacity\n",
      "\n",
      "GarageArea: Size of garage in square feet\n",
      "\n",
      "GarageQual: Garage quality\n",
      "\n",
      "       Ex\tExcellent\n",
      "       Gd\tGood\n",
      "       TA\tTypical/Average\n",
      "       Fa\tFair\n",
      "       Po\tPoor\n",
      "       NA\tNo Garage\n",
      "\t\t\n",
      "GarageCond: Garage condition\n",
      "\n",
      "       Ex\tExcellent\n",
      "       Gd\tGood\n",
      "       TA\tTypical/Average\n",
      "       Fa\tFair\n",
      "       Po\tPoor\n",
      "       NA\tNo Garage\n",
      "\t\t\n",
      "PavedDrive: Paved driveway\n",
      "\n",
      "       Y\tPaved \n",
      "       P\tPartial Pavement\n",
      "       N\tDirt/Gravel\n",
      "\t\t\n",
      "WoodDeckSF: Wood deck area in square feet\n",
      "\n",
      "OpenPorchSF: Open porch area in square feet\n",
      "\n",
      "EnclosedPorch: Enclosed porch area in square feet\n",
      "\n",
      "3SsnPorch: Three season porch area in square feet\n",
      "\n",
      "ScreenPorch: Screen porch area in square feet\n",
      "\n",
      "PoolArea: Pool area in square feet\n",
      "\n",
      "PoolQC: Pool quality\n",
      "\t\t\n",
      "       Ex\tExcellent\n",
      "       Gd\tGood\n",
      "       TA\tAverage/Typical\n",
      "       Fa\tFair\n",
      "       NA\tNo Pool\n",
      "\t\t\n",
      "Fence: Fence quality\n",
      "\t\t\n",
      "       GdPrv\tGood Privacy\n",
      "       MnPrv\tMinimum Privacy\n",
      "       GdWo\tGood Wood\n",
      "       MnWw\tMinimum Wood/Wire\n",
      "       NA\tNo Fence\n",
      "\t\n",
      "MiscFeature: Miscellaneous feature not covered in other categories\n",
      "\t\t\n",
      "       Elev\tElevator\n",
      "       Gar2\t2nd Garage (if not described in garage section)\n",
      "       Othr\tOther\n",
      "       Shed\tShed (over 100 SF)\n",
      "       TenC\tTennis Court\n",
      "       NA\tNone\n",
      "\t\t\n",
      "MiscVal: $Value of miscellaneous feature\n",
      "\n",
      "MoSold: Month Sold (MM)\n",
      "\n",
      "YrSold: Year Sold (YYYY)\n",
      "\n",
      "SaleType: Type of sale\n",
      "\t\t\n",
      "       WD \tWarranty Deed - Conventional\n",
      "       CWD\tWarranty Deed - Cash\n",
      "       VWD\tWarranty Deed - VA Loan\n",
      "       New\tHome just constructed and sold\n",
      "       COD\tCourt Officer Deed/Estate\n",
      "       Con\tContract 15% Down payment regular terms\n",
      "       ConLw\tContract Low Down payment and low interest\n",
      "       ConLI\tContract Low Interest\n",
      "       ConLD\tContract Low Down\n",
      "       Oth\tOther\n",
      "\t\t\n",
      "SaleCondition: Condition of sale\n",
      "\n",
      "       Normal\tNormal Sale\n",
      "       Abnorml\tAbnormal Sale -  trade, foreclosure, short sale\n",
      "       AdjLand\tAdjoining Land Purchase\n",
      "       Alloca\tAllocation - two linked properties with separate deeds, typically condo with a garage unit\t\n",
      "       Family\tSale between family members\n",
      "       Partial\tHome was not completed when last assessed (associated with New Homes)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#data description\n",
    "desc = open(\"data_description.txt\", \"r\")\n",
    "print(desc.read())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data preprocessing and descriptive statistics <a class=\"anchor\" id=\"second-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I begin by checking the presence of missing values in the training dataset and identify columns to drop. I drop the selected columns in both the training and testing datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1105,
   "metadata": {},
   "outputs": [],
   "source": [
    "def percent_missing(df):\n",
    "    '''\n",
    "    Calculates the percentage of missing values in a dataset.\n",
    "    \n",
    "    Args:\n",
    "    inputs: Dataset\n",
    "    returns: Dataset showing the percentage of missing values per column.\n",
    "    '''\n",
    "    percent_missing = df.isnull().sum()* 100/len(df)\n",
    "    percent_missing_df = pd.DataFrame({'col': df.columns,\n",
    "                                 'percent_missing': percent_missing})\n",
    "    return percent_missing_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1106,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate the percentage of missing values\n",
    "missing_train = percent_missing(df_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1557,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<BarContainer object of 10 artists>"
      ]
     },
     "execution_count": 1557,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# top 15 attributes in terms of missing values\n",
    "missing_df = pd.DataFrame(missing_train[\"percent_missing\"].sort_values(ascending = False).head(10)).reset_index()\n",
    "missing_df.columns = ['Attribute', 'Missing']\n",
    "\n",
    "#vizualize missing data\n",
    "plt.figure(figsize=(14,6))\n",
    "plt.bar(missing_df.Attribute, missing_df.Missing)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most variables have less than 6% values missing. For this analysis, I will drop variables that have more than 18% values missing, resulting in the elimination of 5 attributes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Alley', 'FireplaceQu', 'PoolQC', 'Fence', 'MiscFeature']"
      ]
     },
     "execution_count": 1108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#select columns to remove\n",
    "missing_remove = list(missing_train[missing_train[\"percent_missing\"]>19][\"col\"])\n",
    "missing_remove"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1109,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = df_train.drop(missing_remove, inplace = False, axis = 1)\n",
    "test = df_test.drop(missing_remove, inplace = False, axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1460, 76), (1459, 75))"
      ]
     },
     "execution_count": 1110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.shape, test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing and descriptive statistics of categorical attributes <a class=\"anchor\" id=\"third-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1111,
   "metadata": {},
   "outputs": [],
   "source": [
    "# list of categorical variables\n",
    "categorical_col = [\"MSSubClass\",\"MSZoning\", \"Street\", \"LotShape\", \"LandContour\", \n",
    "                   \"Utilities\", \"LotConfig\", \"LandSlope\", \"Neighborhood\", \"Condition1\",\n",
    "                   \"Condition2\", \"BldgType\", \"HouseStyle\", \"RoofStyle\", \"Exterior1st\",\n",
    "                   \"Exterior2nd\", \"MasVnrType\", \"Foundation\", \"Heating\", \"Electrical\", \n",
    "                   \"GarageType\", \"SaleType\", \"SaleCondition\", \"CentralAir\", \"RoofMatl\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1112,
   "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>MSSubClass</th>\n",
       "      <th>MSZoning</th>\n",
       "      <th>Street</th>\n",
       "      <th>LotShape</th>\n",
       "      <th>LandContour</th>\n",
       "      <th>Utilities</th>\n",
       "      <th>LotConfig</th>\n",
       "      <th>LandSlope</th>\n",
       "      <th>Neighborhood</th>\n",
       "      <th>Condition1</th>\n",
       "      <th>Condition2</th>\n",
       "      <th>BldgType</th>\n",
       "      <th>HouseStyle</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>15</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>25</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>20</td>\n",
       "      <td>RL</td>\n",
       "      <td>Pave</td>\n",
       "      <td>Reg</td>\n",
       "      <td>Lvl</td>\n",
       "      <td>AllPub</td>\n",
       "      <td>Inside</td>\n",
       "      <td>Gtl</td>\n",
       "      <td>NAmes</td>\n",
       "      <td>Norm</td>\n",
       "      <td>Norm</td>\n",
       "      <td>1Fam</td>\n",
       "      <td>1Story</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>536</td>\n",
       "      <td>1151</td>\n",
       "      <td>1454</td>\n",
       "      <td>925</td>\n",
       "      <td>1311</td>\n",
       "      <td>1459</td>\n",
       "      <td>1052</td>\n",
       "      <td>1382</td>\n",
       "      <td>225</td>\n",
       "      <td>1260</td>\n",
       "      <td>1445</td>\n",
       "      <td>1220</td>\n",
       "      <td>726</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       MSSubClass MSZoning Street LotShape LandContour Utilities LotConfig  \\\n",
       "count        1460     1460   1460     1460        1460      1460      1460   \n",
       "unique         15        5      2        4           4         2         5   \n",
       "top            20       RL   Pave      Reg         Lvl    AllPub    Inside   \n",
       "freq          536     1151   1454      925        1311      1459      1052   \n",
       "\n",
       "       LandSlope Neighborhood Condition1 Condition2 BldgType HouseStyle  \n",
       "count       1460         1460       1460       1460     1460       1460  \n",
       "unique         3           25          9          8        5          8  \n",
       "top          Gtl        NAmes       Norm       Norm     1Fam     1Story  \n",
       "freq        1382          225       1260       1445     1220        726  "
      ]
     },
     "execution_count": 1112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# descriptive statistics of categorical variables - part1\n",
    "train[categorical_col].iloc[:, 0:13].astype(str).describe(include = 'all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1113,
   "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>RoofStyle</th>\n",
       "      <th>Exterior1st</th>\n",
       "      <th>Exterior2nd</th>\n",
       "      <th>MasVnrType</th>\n",
       "      <th>Foundation</th>\n",
       "      <th>Heating</th>\n",
       "      <th>Electrical</th>\n",
       "      <th>GarageType</th>\n",
       "      <th>SaleType</th>\n",
       "      <th>SaleCondition</th>\n",
       "      <th>CentralAir</th>\n",
       "      <th>RoofMatl</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1452</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1459</td>\n",
       "      <td>1379</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "      <td>1460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>16</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>Gable</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>VinylSd</td>\n",
       "      <td>None</td>\n",
       "      <td>PConc</td>\n",
       "      <td>GasA</td>\n",
       "      <td>SBrkr</td>\n",
       "      <td>Attchd</td>\n",
       "      <td>WD</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Y</td>\n",
       "      <td>CompShg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>1141</td>\n",
       "      <td>515</td>\n",
       "      <td>504</td>\n",
       "      <td>864</td>\n",
       "      <td>647</td>\n",
       "      <td>1428</td>\n",
       "      <td>1334</td>\n",
       "      <td>870</td>\n",
       "      <td>1267</td>\n",
       "      <td>1198</td>\n",
       "      <td>1365</td>\n",
       "      <td>1434</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       RoofStyle Exterior1st Exterior2nd MasVnrType Foundation Heating  \\\n",
       "count       1460        1460        1460       1452       1460    1460   \n",
       "unique         6          15          16          4          6       6   \n",
       "top        Gable     VinylSd     VinylSd       None      PConc    GasA   \n",
       "freq        1141         515         504        864        647    1428   \n",
       "\n",
       "       Electrical GarageType SaleType SaleCondition CentralAir RoofMatl  \n",
       "count        1459       1379     1460          1460       1460     1460  \n",
       "unique          5          6        9             6          2        8  \n",
       "top         SBrkr     Attchd       WD        Normal          Y  CompShg  \n",
       "freq         1334        870     1267          1198       1365     1434  "
      ]
     },
     "execution_count": 1113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# descriptive statistics of categorical variables - part2\n",
    "train[categorical_col].iloc[:, 13:].describe(include = 'all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1114,
   "metadata": {},
   "outputs": [],
   "source": [
    "# impute categorical attributes using the most frequent value\n",
    "simple_imputer = SimpleImputer(missing_values=np.nan, strategy='most_frequent')\n",
    "\n",
    "#fit imputer on the combined dataset\n",
    "for col in categorical_col:\n",
    "    simple_imputer.fit(pd.concat([train[col], test[col]], axis=0, sort=False).values.reshape(-1, 1))\n",
    "    \n",
    "    #apply the imputer separately to the train and test dataset\n",
    "    train[col] = simple_imputer.transform(train[col].values.reshape(-1, 1))\n",
    "    test[col] = simple_imputer.transform(test[col].values.reshape(-1, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1115,
   "metadata": {},
   "outputs": [],
   "source": [
    "#length of the train set - I will needed to separate the concatenated set later\n",
    "train_objs_num = len(train)\n",
    "\n",
    "#concatenate the test and train sets\n",
    "dataset = pd.concat(objs = [train, test], axis=0)\n",
    "\n",
    "# get dummies on the combined data first\n",
    "dataset_preprocessed  = pd.get_dummies(dataset, columns = categorical_col, prefix = categorical_col, dummy_na = False, drop_first = True)\n",
    "\n",
    "# separate the test and train sets from the combined dataset\n",
    "train = dataset_preprocessed[:train_objs_num]\n",
    "test = dataset_preprocessed[train_objs_num:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Preprocessing and descriptive statistics of numerical attributes <a class=\"anchor\" id=\"fourth-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Continuous attributes <a class=\"anchor\" id=\"5-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1116,
   "metadata": {},
   "outputs": [],
   "source": [
    "# engineer features using numeric attributes\n",
    "def numeric_features(df):\n",
    "    '''\n",
    "    Creates new features based on continuous numerical attributes \n",
    "    \n",
    "    Args:\n",
    "    df: DataFrame\n",
    "    \n",
    "    Returns:\n",
    "    df: DataFrame with new features\n",
    "    '''\n",
    "    \n",
    "    #create new features by aggregating several existing features\n",
    "    df['Total_indoor_area'] = df['TotalBsmtSF'] + df['1stFlrSF'] + df['2ndFlrSF']\n",
    "    df['Total_area'] = df['TotalBsmtSF'] + df['1stFlrSF'] + df['2ndFlrSF'] + df['GrLivArea'] + df['GarageArea']\n",
    "    df['Total_bathrooms'] = (df['FullBath'] + (0.5 * df['HalfBath']) + df['BsmtFullBath'] + (0.5 * df['BsmtHalfBath']))\n",
    "    df['Total_baths'] = df['FullBath'] + 0.5 * df['HalfBath']\n",
    "    df['Total_lot_area'] = df['LotFrontage'] + df['LotArea']\n",
    "    df['Total_porch_area'] = df['OpenPorchSF'] + df['EnclosedPorch'] + df['ScreenPorch']\n",
    "    df['Total_finished_basement'] = df['BsmtFinSF1'] + df['BsmtFinSF2']\n",
    "    df['Total_porch_area'] = (df['OpenPorchSF'] + df['3SsnPorch'] + df['EnclosedPorch'] + df['ScreenPorch'] + df['WoodDeckSF'])\n",
    "    \n",
    "    # create dummy variables that take 1 if the detail is present, 0 if absent\n",
    "    df['Pool_dummy'] = df['PoolArea'].apply(lambda x: 1 if x > 0 else 0)\n",
    "    df['2floor_dummy'] = df['2ndFlrSF'].apply(lambda x: 1 if x > 0 else 0)\n",
    "    df['Garage_dummy'] = df['GarageArea'].apply(lambda x: 1 if x > 0 else 0)\n",
    "    df['Basement_dummy'] = df['TotalBsmtSF'].apply(lambda x: 1 if x > 0 else 0)\n",
    "    df['Fireplace_dummy'] = df['Fireplaces'].apply(lambda x: 1 if x > 0 else 0)\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1117,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = numeric_features(train)\n",
    "test = numeric_features(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Total_lot_area     259\n",
       "LotFrontage        259\n",
       "GarageYrBlt         81\n",
       "GarageFinish        81\n",
       "GarageQual          81\n",
       "GarageCond          81\n",
       "BsmtExposure        38\n",
       "BsmtFinType2        38\n",
       "BsmtQual            37\n",
       "BsmtCond            37\n",
       "BsmtFinType1        37\n",
       "MasVnrArea           8\n",
       "MSZoning_FV          0\n",
       "LandContour_Lvl      0\n",
       "MSSubClass_150       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 1120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check for null values in attributes\n",
    "pd.isnull(train).sum().sort_values(ascending=False).head(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LotFrontage          70.049958\n",
       "Total_lot_area    10021.748543\n",
       "MasVnrArea          103.685262\n",
       "GarageYrBlt        1978.506164\n",
       "dtype: float64"
      ]
     },
     "execution_count": 1121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check the means of the numeric variables before imputing \n",
    "train[['LotFrontage','Total_lot_area','MasVnrArea', 'GarageYrBlt']].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1123,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Impute numerical continuous attributes using the closest neighbors\n",
    "# numeric attibutes with missing values\n",
    "num_col = ['LotFrontage','Total_lot_area','MasVnrArea', 'GarageYrBlt', 'Total_area', 'Total_bathrooms', 'BsmtFullBath', 'BsmtHalfBath',\n",
    "          'TotalBsmtSF', 'BsmtUnfSF', 'BsmtFinSF1', 'Total_indoor_area', 'BsmtFinSF2', 'Total_finished_basement', 'GarageArea', 'GarageCars']\n",
    "\n",
    "def knn_imputer(train, test, num_col):\n",
    "    '''\n",
    "    Imputes null values using KNN imputer for selected columns\n",
    "    \n",
    "    Args:\n",
    "    train: DataFrame containing training data\n",
    "    test: DataFrame containing test data\n",
    "    num_cols: continuous numerical columns with null values\n",
    "    \n",
    "    Returns:\n",
    "    train and test datasets where the columns in num_cols are imputed\n",
    "    '''\n",
    "    #initialize imputer\n",
    "    imputer = KNNImputer(n_neighbors=5, weights='uniform', metric='nan_euclidean')\n",
    "    # fit imputer on the combined dataset\n",
    "    for col in num_col:\n",
    "        imputer.fit(pd.concat([train[col], test[col]], axis=0, sort=False).values.reshape(-1, 1))\n",
    "    \n",
    "    #apply the imputer separately to the train and test dataset\n",
    "        train[col] = imputer.transform(train[col].values.reshape(-1, 1))\n",
    "        test[col] = imputer.transform(test[col].values.reshape(-1, 1))\n",
    "    \n",
    "    return train, test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1125,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(        Id  LotFrontage  LotArea  OverallQual  OverallCond  YearBuilt  \\\n",
       " 0        1         65.0     8450            7            5       2003   \n",
       " 1        2         80.0     9600            6            8       1976   \n",
       " 2        3         68.0    11250            7            5       2001   \n",
       " 3        4         60.0     9550            7            5       1915   \n",
       " 4        5         84.0    14260            8            5       2000   \n",
       " ...    ...          ...      ...          ...          ...        ...   \n",
       " 1455  1456         62.0     7917            6            5       1999   \n",
       " 1456  1457         85.0    13175            6            6       1978   \n",
       " 1457  1458         66.0     9042            7            9       1941   \n",
       " 1458  1459         68.0     9717            5            6       1950   \n",
       " 1459  1460         75.0     9937            5            6       1965   \n",
       " \n",
       "       YearRemodAdd  MasVnrArea ExterQual ExterCond  ... Total_bathrooms  \\\n",
       " 0             2003       196.0        Gd        TA  ...             3.5   \n",
       " 1             1976         0.0        TA        TA  ...             2.5   \n",
       " 2             2002       162.0        Gd        TA  ...             3.5   \n",
       " 3             1970         0.0        TA        TA  ...             2.0   \n",
       " 4             2000       350.0        Gd        TA  ...             3.5   \n",
       " ...            ...         ...       ...       ...  ...             ...   \n",
       " 1455          2000         0.0        TA        TA  ...             2.5   \n",
       " 1456          1988       119.0        TA        TA  ...             3.0   \n",
       " 1457          2006         0.0        Ex        Gd  ...             2.0   \n",
       " 1458          1996         0.0        TA        TA  ...             2.0   \n",
       " 1459          1965         0.0        Gd        TA  ...             2.5   \n",
       " \n",
       "      Total_baths Total_lot_area Total_porch_area  Total_finished_basement  \\\n",
       " 0            2.5         8515.0               61                    706.0   \n",
       " 1            2.0         9680.0              298                    978.0   \n",
       " 2            2.5        11318.0               42                    486.0   \n",
       " 3            1.0         9610.0              307                    216.0   \n",
       " 4            2.5        14344.0              276                    655.0   \n",
       " ...          ...            ...              ...                      ...   \n",
       " 1455         2.5         7979.0               40                      0.0   \n",
       " 1456         2.0        13260.0              349                    953.0   \n",
       " 1457         2.0         9108.0               60                    275.0   \n",
       " 1458         1.0         9785.0              478                   1078.0   \n",
       " 1459         1.5        10012.0              804                   1120.0   \n",
       " \n",
       "      Pool_dummy  2floor_dummy  Garage_dummy  Basement_dummy Fireplace_dummy  \n",
       " 0             0             1             1               1               0  \n",
       " 1             0             0             1               1               1  \n",
       " 2             0             1             1               1               1  \n",
       " 3             0             1             1               1               1  \n",
       " 4             0             1             1               1               1  \n",
       " ...         ...           ...           ...             ...             ...  \n",
       " 1455          0             1             1               1               1  \n",
       " 1456          0             0             1               1               1  \n",
       " 1457          0             1             1               1               1  \n",
       " 1458          0             0             1               1               0  \n",
       " 1459          0             0             1               1               0  \n",
       " \n",
       " [1460 rows x 223 columns],\n",
       "         Id  LotFrontage  LotArea  OverallQual  OverallCond  YearBuilt  \\\n",
       " 0     1461         80.0    11622            5            6       1961   \n",
       " 1     1462         81.0    14267            6            6       1958   \n",
       " 2     1463         74.0    13830            5            5       1997   \n",
       " 3     1464         78.0     9978            6            6       1998   \n",
       " 4     1465         43.0     5005            8            5       1992   \n",
       " ...    ...          ...      ...          ...          ...        ...   \n",
       " 1454  2915         21.0     1936            4            7       1970   \n",
       " 1455  2916         21.0     1894            4            5       1970   \n",
       " 1456  2917        160.0    20000            5            7       1960   \n",
       " 1457  2918         62.0    10441            5            5       1992   \n",
       " 1458  2919         74.0     9627            7            5       1993   \n",
       " \n",
       "       YearRemodAdd  MasVnrArea ExterQual ExterCond  ... Total_bathrooms  \\\n",
       " 0             1961         0.0        TA        TA  ...             1.0   \n",
       " 1             1958       108.0        TA        TA  ...             1.5   \n",
       " 2             1998         0.0        TA        TA  ...             2.5   \n",
       " 3             1998        20.0        TA        TA  ...             2.5   \n",
       " 4             1992         0.0        Gd        TA  ...             2.0   \n",
       " ...            ...         ...       ...       ...  ...             ...   \n",
       " 1454          1970         0.0        TA        TA  ...             1.5   \n",
       " 1455          1970         0.0        TA        TA  ...             1.5   \n",
       " 1456          1996         0.0        TA        TA  ...             2.0   \n",
       " 1457          1992         0.0        TA        TA  ...             1.5   \n",
       " 1458          1994        94.0        TA        TA  ...             2.5   \n",
       " \n",
       "      Total_baths Total_lot_area Total_porch_area  Total_finished_basement  \\\n",
       " 0            1.0        11702.0              260                    612.0   \n",
       " 1            1.5        14348.0              429                    923.0   \n",
       " 2            2.5        13904.0              246                    791.0   \n",
       " 3            2.5        10056.0              396                    602.0   \n",
       " 4            2.0         5048.0              226                    263.0   \n",
       " ...          ...            ...              ...                      ...   \n",
       " 1454         1.5         1957.0                0                      0.0   \n",
       " 1455         1.5         1915.0               24                    252.0   \n",
       " 1456         1.0        20160.0              474                   1224.0   \n",
       " 1457         1.0        10503.0              112                    337.0   \n",
       " 1458         2.5         9701.0              238                    758.0   \n",
       " \n",
       "      Pool_dummy  2floor_dummy  Garage_dummy  Basement_dummy Fireplace_dummy  \n",
       " 0             0             0             1               1               0  \n",
       " 1             0             0             1               1               0  \n",
       " 2             0             1             1               1               1  \n",
       " 3             0             1             1               1               1  \n",
       " 4             0             0             1               1               0  \n",
       " ...         ...           ...           ...             ...             ...  \n",
       " 1454          0             1             0               1               0  \n",
       " 1455          0             1             1               1               0  \n",
       " 1456          0             0             1               1               1  \n",
       " 1457          0             0             0               1               0  \n",
       " 1458          0             1             1               1               1  \n",
       " \n",
       " [1459 rows x 223 columns])"
      ]
     },
     "execution_count": 1125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "knn_imputer(train, test, num_col)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1126,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LotFrontage         69.917946\n",
       "Total_lot_area    9981.619489\n",
       "MasVnrArea         103.677130\n",
       "GarageYrBlt       1978.484374\n",
       "dtype: float64"
      ]
     },
     "execution_count": 1126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check that the means of the numerical attributes have decreased after imputation \n",
    "train[['LotFrontage','Total_lot_area','MasVnrArea', 'GarageYrBlt']].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1127,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GarageFinish        81\n",
       "GarageQual          81\n",
       "GarageCond          81\n",
       "BsmtFinType2        38\n",
       "BsmtExposure        38\n",
       "BsmtQual            37\n",
       "BsmtFinType1        37\n",
       "BsmtCond            37\n",
       "LandContour_HLS      0\n",
       "LandContour_Low      0\n",
       "Utilities_NoSeWa     0\n",
       "LandContour_Lvl      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 1127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# verify that only categorical attributes have missing values - to be imputed later\n",
    "pd.isnull(train).sum().sort_values(ascending=False).head(12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ordinal attributes <a class=\"anchor\" id=\"6-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1128,
   "metadata": {},
   "outputs": [],
   "source": [
    "# list of variables to be treated as ordinal\n",
    "ordinal_col = [\"OverallQual\", \"OverallCond\", \"ExterQual\", \"ExterCond\", \"BsmtQual\", \"BsmtCond\", \n",
    "               \"BsmtExposure\", \"BsmtFinType1\", \"BsmtFinType2\", \"HeatingQC\", \"KitchenQual\", \n",
    "               \"Functional\", \"GarageFinish\", \"GarageQual\", \"GarageCond\", \n",
    "               \"PavedDrive\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1129,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GarageCond      81\n",
       "GarageQual      81\n",
       "GarageFinish    81\n",
       "BsmtFinType2    38\n",
       "BsmtExposure    38\n",
       "BsmtFinType1    37\n",
       "BsmtCond        37\n",
       "BsmtQual        37\n",
       "PavedDrive       0\n",
       "Functional       0\n",
       "KitchenQual      0\n",
       "HeatingQC        0\n",
       "ExterCond        0\n",
       "ExterQual        0\n",
       "OverallCond      0\n",
       "OverallQual      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 1129,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check for missing values in the ordianl attributes\n",
    "pd.isnull(train[ordinal_col]).sum().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1131,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(        Id  LotFrontage  LotArea  OverallQual  OverallCond  YearBuilt  \\\n",
       " 0        1         65.0     8450            7            5       2003   \n",
       " 1        2         80.0     9600            6            8       1976   \n",
       " 2        3         68.0    11250            7            5       2001   \n",
       " 3        4         60.0     9550            7            5       1915   \n",
       " 4        5         84.0    14260            8            5       2000   \n",
       " ...    ...          ...      ...          ...          ...        ...   \n",
       " 1455  1456         62.0     7917            6            5       1999   \n",
       " 1456  1457         85.0    13175            6            6       1978   \n",
       " 1457  1458         66.0     9042            7            9       1941   \n",
       " 1458  1459         68.0     9717            5            6       1950   \n",
       " 1459  1460         75.0     9937            5            6       1965   \n",
       " \n",
       "       YearRemodAdd  MasVnrArea ExterQual ExterCond  ... Total_bathrooms  \\\n",
       " 0             2003       196.0        Gd        TA  ...             3.5   \n",
       " 1             1976         0.0        TA        TA  ...             2.5   \n",
       " 2             2002       162.0        Gd        TA  ...             3.5   \n",
       " 3             1970         0.0        TA        TA  ...             2.0   \n",
       " 4             2000       350.0        Gd        TA  ...             3.5   \n",
       " ...            ...         ...       ...       ...  ...             ...   \n",
       " 1455          2000         0.0        TA        TA  ...             2.5   \n",
       " 1456          1988       119.0        TA        TA  ...             3.0   \n",
       " 1457          2006         0.0        Ex        Gd  ...             2.0   \n",
       " 1458          1996         0.0        TA        TA  ...             2.0   \n",
       " 1459          1965         0.0        Gd        TA  ...             2.5   \n",
       " \n",
       "      Total_baths Total_lot_area Total_porch_area  Total_finished_basement  \\\n",
       " 0            2.5         8515.0               61                    706.0   \n",
       " 1            2.0         9680.0              298                    978.0   \n",
       " 2            2.5        11318.0               42                    486.0   \n",
       " 3            1.0         9610.0              307                    216.0   \n",
       " 4            2.5        14344.0              276                    655.0   \n",
       " ...          ...            ...              ...                      ...   \n",
       " 1455         2.5         7979.0               40                      0.0   \n",
       " 1456         2.0        13260.0              349                    953.0   \n",
       " 1457         2.0         9108.0               60                    275.0   \n",
       " 1458         1.0         9785.0              478                   1078.0   \n",
       " 1459         1.5        10012.0              804                   1120.0   \n",
       " \n",
       "      Pool_dummy  2floor_dummy  Garage_dummy  Basement_dummy Fireplace_dummy  \n",
       " 0             0             1             1               1               0  \n",
       " 1             0             0             1               1               1  \n",
       " 2             0             1             1               1               1  \n",
       " 3             0             1             1               1               1  \n",
       " 4             0             1             1               1               1  \n",
       " ...         ...           ...           ...             ...             ...  \n",
       " 1455          0             1             1               1               1  \n",
       " 1456          0             0             1               1               1  \n",
       " 1457          0             1             1               1               1  \n",
       " 1458          0             0             1               1               0  \n",
       " 1459          0             0             1               1               0  \n",
       " \n",
       " [1460 rows x 223 columns],\n",
       "         Id  LotFrontage  LotArea  OverallQual  OverallCond  YearBuilt  \\\n",
       " 0     1461         80.0    11622            5            6       1961   \n",
       " 1     1462         81.0    14267            6            6       1958   \n",
       " 2     1463         74.0    13830            5            5       1997   \n",
       " 3     1464         78.0     9978            6            6       1998   \n",
       " 4     1465         43.0     5005            8            5       1992   \n",
       " ...    ...          ...      ...          ...          ...        ...   \n",
       " 1454  2915         21.0     1936            4            7       1970   \n",
       " 1455  2916         21.0     1894            4            5       1970   \n",
       " 1456  2917        160.0    20000            5            7       1960   \n",
       " 1457  2918         62.0    10441            5            5       1992   \n",
       " 1458  2919         74.0     9627            7            5       1993   \n",
       " \n",
       "       YearRemodAdd  MasVnrArea ExterQual ExterCond  ... Total_bathrooms  \\\n",
       " 0             1961         0.0        TA        TA  ...             1.0   \n",
       " 1             1958       108.0        TA        TA  ...             1.5   \n",
       " 2             1998         0.0        TA        TA  ...             2.5   \n",
       " 3             1998        20.0        TA        TA  ...             2.5   \n",
       " 4             1992         0.0        Gd        TA  ...             2.0   \n",
       " ...            ...         ...       ...       ...  ...             ...   \n",
       " 1454          1970         0.0        TA        TA  ...             1.5   \n",
       " 1455          1970         0.0        TA        TA  ...             1.5   \n",
       " 1456          1996         0.0        TA        TA  ...             2.0   \n",
       " 1457          1992         0.0        TA        TA  ...             1.5   \n",
       " 1458          1994        94.0        TA        TA  ...             2.5   \n",
       " \n",
       "      Total_baths Total_lot_area Total_porch_area  Total_finished_basement  \\\n",
       " 0            1.0        11702.0              260                    612.0   \n",
       " 1            1.5        14348.0              429                    923.0   \n",
       " 2            2.5        13904.0              246                    791.0   \n",
       " 3            2.5        10056.0              396                    602.0   \n",
       " 4            2.0         5048.0              226                    263.0   \n",
       " ...          ...            ...              ...                      ...   \n",
       " 1454         1.5         1957.0                0                      0.0   \n",
       " 1455         1.5         1915.0               24                    252.0   \n",
       " 1456         1.0        20160.0              474                   1224.0   \n",
       " 1457         1.0        10503.0              112                    337.0   \n",
       " 1458         2.5         9701.0              238                    758.0   \n",
       " \n",
       "      Pool_dummy  2floor_dummy  Garage_dummy  Basement_dummy Fireplace_dummy  \n",
       " 0             0             0             1               1               0  \n",
       " 1             0             0             1               1               0  \n",
       " 2             0             1             1               1               1  \n",
       " 3             0             1             1               1               1  \n",
       " 4             0             0             1               1               0  \n",
       " ...         ...           ...           ...             ...             ...  \n",
       " 1454          0             1             0               1               0  \n",
       " 1455          0             1             1               1               0  \n",
       " 1456          0             0             1               1               1  \n",
       " 1457          0             0             0               1               0  \n",
       " 1458          0             1             1               1               1  \n",
       " \n",
       " [1459 rows x 223 columns])"
      ]
     },
     "execution_count": 1131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# impute ordinal attributes using the most frequent value\n",
    "def simple_imputer(train, test, ordinal_col):\n",
    "    '''\n",
    "    Imputes null values using simple imputer for selected columns\n",
    "    \n",
    "    Args:\n",
    "    train: DataFrame containing training data\n",
    "    test: DataFrame containing test data\n",
    "    ordinal_col: list of ordinal attributes with null values\n",
    "    \n",
    "    Returns:\n",
    "    train and test datasets where the columns in ordinal_col are imputed\n",
    "    '''\n",
    "    #initialize imputer\n",
    "    imputer = SimpleImputer(missing_values=np.nan, strategy='most_frequent')\n",
    "    # fit imputer on the combined dataset\n",
    "    for col in ordinal_col:\n",
    "        imputer.fit(pd.concat([train[col], test[col]], axis=0, sort=False).values.reshape(-1, 1))\n",
    "    \n",
    "    #apply the imputer separately to the train and test dataset\n",
    "        train[col] = imputer.transform(train[col].values.reshape(-1, 1))\n",
    "        test[col] = imputer.transform(test[col].values.reshape(-1, 1))\n",
    "    \n",
    "    return train, test\n",
    "\n",
    "simple_imputer(train, test, ordinal_col)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PavedDrive      0\n",
       "GarageCond      0\n",
       "GarageQual      0\n",
       "GarageFinish    0\n",
       "Functional      0\n",
       "KitchenQual     0\n",
       "HeatingQC       0\n",
       "BsmtFinType2    0\n",
       "BsmtFinType1    0\n",
       "BsmtExposure    0\n",
       "BsmtCond        0\n",
       "BsmtQual        0\n",
       "ExterCond       0\n",
       "ExterQual       0\n",
       "OverallCond     0\n",
       "OverallQual     0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 1133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check if any null values left in ordinal attributes\n",
    "pd.isnull(test[ordinal_col]).sum().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1134,
   "metadata": {},
   "outputs": [],
   "source": [
    "#create a dictionary of columns\n",
    "dict_cols = {0: [\"OverallQual\", \"OverallCond\"],\n",
    "1: [\"ExterQual\", \"ExterCond\", \"BsmtQual\", \"BsmtCond\", 'HeatingQC', 'KitchenQual', 'GarageQual', 'GarageCond'], \n",
    "2: ['BsmtExposure'], \n",
    "3: ['BsmtFinType1', 'BsmtFinType2'],\n",
    "4: ['Functional'],\n",
    "5: ['GarageFinish'], \n",
    "6: ['PavedDrive']}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1135,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create an ordered dictionary of categories\n",
    "dict_cats = {0: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10],\n",
    "1: ['Po', 'Fa', 'TA', 'Gd', 'Ex'],\n",
    "2: ['No', 'Mn', 'Av', 'Gd'],\n",
    "3: ['Unf', 'LwQ', 'Rec', 'BLQ', 'ALQ', 'GLQ'], \n",
    "4: ['Sal', 'Sev', 'Maj2', 'Maj1', 'Mod', 'Min2', 'Min1', 'Typ'], \n",
    "5: ['Unf', 'RFn', 'Fin'],\n",
    "6: ['N', 'P', 'Y']}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1136,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 2, 3, 4, 5, 6, 7, 8, 9, 10]\n",
      "OverallQual\n",
      "Finished\n",
      "OverallCond\n",
      "Finished\n",
      "['Po', 'Fa', 'TA', 'Gd', 'Ex']\n",
      "ExterQual\n",
      "Finished\n",
      "ExterCond\n",
      "Finished\n",
      "BsmtQual\n",
      "Finished\n",
      "BsmtCond\n",
      "Finished\n",
      "HeatingQC\n",
      "Finished\n",
      "KitchenQual\n",
      "Finished\n",
      "GarageQual\n",
      "Finished\n",
      "GarageCond\n",
      "Finished\n",
      "['No', 'Mn', 'Av', 'Gd']\n",
      "BsmtExposure\n",
      "Finished\n",
      "['Unf', 'LwQ', 'Rec', 'BLQ', 'ALQ', 'GLQ']\n",
      "BsmtFinType1\n",
      "Finished\n",
      "BsmtFinType2\n",
      "Finished\n",
      "['Sal', 'Sev', 'Maj2', 'Maj1', 'Mod', 'Min2', 'Min1', 'Typ']\n",
      "Functional\n",
      "Finished\n",
      "['Unf', 'RFn', 'Fin']\n",
      "GarageFinish\n",
      "Finished\n",
      "['N', 'P', 'Y']\n",
      "PavedDrive\n",
      "Finished\n"
     ]
    }
   ],
   "source": [
    "# iterate through columns in dict_cols and encode ordinal variables of each attribute using the specific order given in dict_cats\n",
    "for i in range(0, 7):\n",
    "    print(dict_cats[i])\n",
    "    encoder = OrdinalEncoder(categories = [dict_cats[i]])\n",
    "    for col in dict_cols[i]:\n",
    "        print(col)\n",
    "        encoder.fit(pd.concat([train[col], test[col]], axis=0, sort=False).values.reshape(-1, 1))\n",
    "        train[col] = encoder.transform(train[col].values.reshape(-1, 1))\n",
    "        test[col] = encoder.transform(test[col].values.reshape(-1, 1))\n",
    "        print(\"Finished\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1137,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(BsmtFinType2\n",
       " 0.0             1294\n",
       " 2.0               54\n",
       " 1.0               46\n",
       " 3.0               33\n",
       " 4.0               19\n",
       " 5.0               14\n",
       " dtype: int64,\n",
       " BsmtFinType2\n",
       " 0.0             1279\n",
       " 2.0               51\n",
       " 1.0               41\n",
       " 3.0               35\n",
       " 4.0               33\n",
       " 5.0               20\n",
       " dtype: int64)"
      ]
     },
     "execution_count": 1137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check if encoding was successful\n",
    "train[['BsmtFinType2']].value_counts(), test[['BsmtFinType2']].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1140,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(Fireplace_dummy        0\n",
       " MSZoning_RM            0\n",
       " LotConfig_FR3          0\n",
       " LotConfig_FR2          0\n",
       " LotConfig_CulDSac      0\n",
       "                       ..\n",
       " Exterior1st_HdBoard    0\n",
       " Exterior1st_CemntBd    0\n",
       " Exterior1st_CBlock     0\n",
       " Exterior1st_BrkFace    0\n",
       " Id                     0\n",
       " Length: 223, dtype: int64,\n",
       " SalePrice              1459\n",
       " Fireplace_dummy           0\n",
       " MSZoning_RM               0\n",
       " LotConfig_FR3             0\n",
       " LotConfig_FR2             0\n",
       "                        ... \n",
       " Exterior1st_HdBoard       0\n",
       " Exterior1st_CemntBd       0\n",
       " Exterior1st_CBlock        0\n",
       " Exterior1st_BrkFace       0\n",
       " Id                        0\n",
       " Length: 223, dtype: int64)"
      ]
     },
     "execution_count": 1140,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check for presence of any null values in the da\n",
    "pd.isnull(train).sum().sort_values(ascending=False), pd.isnull(test).sum().sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Descriptive statistics of numerical attributes <a class=\"anchor\" id=\"7-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1772,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['LotFrontage',\n",
       " 'Total_lot_area',\n",
       " 'MasVnrArea',\n",
       " 'GarageYrBlt',\n",
       " 'Total_area',\n",
       " 'Total_bathrooms',\n",
       " 'BsmtFullBath',\n",
       " 'BsmtHalfBath',\n",
       " 'TotalBsmtSF',\n",
       " 'BsmtUnfSF',\n",
       " 'BsmtFinSF1',\n",
       " 'Total_indoor_area',\n",
       " 'BsmtFinSF2',\n",
       " 'Total_finished_basement',\n",
       " 'GarageArea',\n",
       " 'GarageCars',\n",
       " 'OverallQual',\n",
       " 'OverallCond',\n",
       " 'ExterQual',\n",
       " 'ExterCond',\n",
       " 'BsmtQual',\n",
       " 'BsmtCond',\n",
       " 'BsmtExposure',\n",
       " 'BsmtFinType1',\n",
       " 'BsmtFinType2',\n",
       " 'HeatingQC',\n",
       " 'KitchenQual',\n",
       " 'Functional',\n",
       " 'GarageFinish',\n",
       " 'GarageQual',\n",
       " 'GarageCond',\n",
       " 'PavedDrive',\n",
       " 'SalePrice']"
      ]
     },
     "execution_count": 1772,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# gather all numerical attributes by combinining\n",
    "num_attributes = num_col + ordinal_col\n",
    "num_attributes.append('SalePrice')\n",
    "num_attributes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1773,
   "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>LotFrontage</th>\n",
       "      <th>Total_lot_area</th>\n",
       "      <th>MasVnrArea</th>\n",
       "      <th>GarageYrBlt</th>\n",
       "      <th>Total_area</th>\n",
       "      <th>Total_bathrooms</th>\n",
       "      <th>BsmtFullBath</th>\n",
       "      <th>BsmtHalfBath</th>\n",
       "      <th>TotalBsmtSF</th>\n",
       "      <th>BsmtUnfSF</th>\n",
       "      <th>BsmtFinSF1</th>\n",
       "      <th>Total_indoor_area</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>69.69</td>\n",
       "      <td>9922.43</td>\n",
       "      <td>102.25</td>\n",
       "      <td>1978.42</td>\n",
       "      <td>4537.36</td>\n",
       "      <td>2.21</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0.06</td>\n",
       "      <td>1052.36</td>\n",
       "      <td>567.23</td>\n",
       "      <td>438.46</td>\n",
       "      <td>2556.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>21.05</td>\n",
       "      <td>7019.85</td>\n",
       "      <td>178.50</td>\n",
       "      <td>23.99</td>\n",
       "      <td>1364.97</td>\n",
       "      <td>0.78</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.24</td>\n",
       "      <td>415.17</td>\n",
       "      <td>442.32</td>\n",
       "      <td>433.15</td>\n",
       "      <td>773.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>21.00</td>\n",
       "      <td>1324.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1900.00</td>\n",
       "      <td>668.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>334.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>60.00</td>\n",
       "      <td>8043.75</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1962.00</td>\n",
       "      <td>3552.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>795.00</td>\n",
       "      <td>222.50</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2008.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>69.31</td>\n",
       "      <td>9795.54</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1978.11</td>\n",
       "      <td>4379.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>991.00</td>\n",
       "      <td>477.50</td>\n",
       "      <td>381.00</td>\n",
       "      <td>2472.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>79.00</td>\n",
       "      <td>10752.50</td>\n",
       "      <td>162.25</td>\n",
       "      <td>2001.00</td>\n",
       "      <td>5312.25</td>\n",
       "      <td>2.50</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1296.25</td>\n",
       "      <td>808.00</td>\n",
       "      <td>709.00</td>\n",
       "      <td>3000.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>313.00</td>\n",
       "      <td>215395.00</td>\n",
       "      <td>1600.00</td>\n",
       "      <td>2010.00</td>\n",
       "      <td>12161.00</td>\n",
       "      <td>6.00</td>\n",
       "      <td>3.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>3206.00</td>\n",
       "      <td>2336.00</td>\n",
       "      <td>2188.00</td>\n",
       "      <td>6872.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       LotFrontage  Total_lot_area  MasVnrArea  GarageYrBlt  Total_area  \\\n",
       "count      1456.00         1456.00     1456.00      1456.00     1456.00   \n",
       "mean         69.69         9922.43      102.25      1978.42     4537.36   \n",
       "std          21.05         7019.85      178.50        23.99     1364.97   \n",
       "min          21.00         1324.00        0.00      1900.00      668.00   \n",
       "25%          60.00         8043.75        0.00      1962.00     3552.00   \n",
       "50%          69.31         9795.54        0.00      1978.11     4379.00   \n",
       "75%          79.00        10752.50      162.25      2001.00     5312.25   \n",
       "max         313.00       215395.00     1600.00      2010.00    12161.00   \n",
       "\n",
       "       Total_bathrooms  BsmtFullBath  BsmtHalfBath  TotalBsmtSF  BsmtUnfSF  \\\n",
       "count          1456.00       1456.00       1456.00      1456.00    1456.00   \n",
       "mean              2.21          0.42          0.06      1052.36     567.23   \n",
       "std               0.78          0.52          0.24       415.17     442.32   \n",
       "min               1.00          0.00          0.00         0.00       0.00   \n",
       "25%               2.00          0.00          0.00       795.00     222.50   \n",
       "50%               2.00          0.00          0.00       991.00     477.50   \n",
       "75%               2.50          1.00          0.00      1296.25     808.00   \n",
       "max               6.00          3.00          2.00      3206.00    2336.00   \n",
       "\n",
       "       BsmtFinSF1  Total_indoor_area  \n",
       "count     1456.00            1456.00  \n",
       "mean       438.46            2556.04  \n",
       "std        433.15             773.91  \n",
       "min          0.00             334.00  \n",
       "25%          0.00            2008.00  \n",
       "50%        381.00            2472.00  \n",
       "75%        709.00            3000.75  \n",
       "max       2188.00            6872.00  "
      ]
     },
     "execution_count": 1773,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# descriptive statistics of numerica values - part 1\n",
    "round(train[num_attributes].iloc[:, :12].describe(), 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1774,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>BsmtFinSF2</th>\n",
       "      <th>Total_finished_basement</th>\n",
       "      <th>GarageArea</th>\n",
       "      <th>GarageCars</th>\n",
       "      <th>OverallQual</th>\n",
       "      <th>OverallCond</th>\n",
       "      <th>ExterQual</th>\n",
       "      <th>ExterCond</th>\n",
       "      <th>BsmtQual</th>\n",
       "      <th>BsmtCond</th>\n",
       "      <th>BsmtExposure</th>\n",
       "      <th>BsmtFinType1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>46.68</td>\n",
       "      <td>485.14</td>\n",
       "      <td>471.79</td>\n",
       "      <td>1.77</td>\n",
       "      <td>5.09</td>\n",
       "      <td>4.58</td>\n",
       "      <td>2.39</td>\n",
       "      <td>2.08</td>\n",
       "      <td>2.56</td>\n",
       "      <td>2.01</td>\n",
       "      <td>0.65</td>\n",
       "      <td>2.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>161.52</td>\n",
       "      <td>454.75</td>\n",
       "      <td>212.17</td>\n",
       "      <td>0.75</td>\n",
       "      <td>1.38</td>\n",
       "      <td>1.11</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.68</td>\n",
       "      <td>0.28</td>\n",
       "      <td>1.04</td>\n",
       "      <td>2.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>329.50</td>\n",
       "      <td>1.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.00</td>\n",
       "      <td>464.00</td>\n",
       "      <td>478.50</td>\n",
       "      <td>2.00</td>\n",
       "      <td>5.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>3.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>3.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.00</td>\n",
       "      <td>789.00</td>\n",
       "      <td>576.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>6.00</td>\n",
       "      <td>5.00</td>\n",
       "      <td>3.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>3.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>5.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1474.00</td>\n",
       "      <td>2306.00</td>\n",
       "      <td>1390.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>9.00</td>\n",
       "      <td>8.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>3.00</td>\n",
       "      <td>3.00</td>\n",
       "      <td>5.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       BsmtFinSF2  Total_finished_basement  GarageArea  GarageCars  \\\n",
       "count     1456.00                  1456.00     1456.00     1456.00   \n",
       "mean        46.68                   485.14      471.79        1.77   \n",
       "std        161.52                   454.75      212.17        0.75   \n",
       "min          0.00                     0.00        0.00        0.00   \n",
       "25%          0.00                     0.00      329.50        1.00   \n",
       "50%          0.00                   464.00      478.50        2.00   \n",
       "75%          0.00                   789.00      576.00        2.00   \n",
       "max       1474.00                  2306.00     1390.00        4.00   \n",
       "\n",
       "       OverallQual  OverallCond  ExterQual  ExterCond  BsmtQual  BsmtCond  \\\n",
       "count      1456.00      1456.00    1456.00    1456.00   1456.00   1456.00   \n",
       "mean          5.09         4.58       2.39       2.08      2.56      2.01   \n",
       "std           1.38         1.11       0.57       0.35      0.68      0.28   \n",
       "min           0.00         0.00       1.00       0.00      1.00      0.00   \n",
       "25%           4.00         4.00       2.00       2.00      2.00      2.00   \n",
       "50%           5.00         4.00       2.00       2.00      3.00      2.00   \n",
       "75%           6.00         5.00       3.00       2.00      3.00      2.00   \n",
       "max           9.00         8.00       4.00       4.00      4.00      3.00   \n",
       "\n",
       "       BsmtExposure  BsmtFinType1  \n",
       "count       1456.00       1456.00  \n",
       "mean           0.65          2.56  \n",
       "std            1.04          2.07  \n",
       "min            0.00          0.00  \n",
       "25%            0.00          0.00  \n",
       "50%            0.00          3.00  \n",
       "75%            1.00          5.00  \n",
       "max            3.00          5.00  "
      ]
     },
     "execution_count": 1774,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# descriptive statistics of numerica values - part 2\n",
    "round(train[num_attributes].iloc[:, 12:24].describe(), 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1775,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>BsmtFinType2</th>\n",
       "      <th>HeatingQC</th>\n",
       "      <th>KitchenQual</th>\n",
       "      <th>Functional</th>\n",
       "      <th>GarageFinish</th>\n",
       "      <th>GarageQual</th>\n",
       "      <th>GarageCond</th>\n",
       "      <th>PavedDrive</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "      <td>1456.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.27</td>\n",
       "      <td>3.14</td>\n",
       "      <td>2.51</td>\n",
       "      <td>6.84</td>\n",
       "      <td>0.77</td>\n",
       "      <td>1.98</td>\n",
       "      <td>1.98</td>\n",
       "      <td>1.86</td>\n",
       "      <td>180810.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.87</td>\n",
       "      <td>0.96</td>\n",
       "      <td>0.66</td>\n",
       "      <td>0.67</td>\n",
       "      <td>0.81</td>\n",
       "      <td>0.24</td>\n",
       "      <td>0.23</td>\n",
       "      <td>0.50</td>\n",
       "      <td>79462.62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>34900.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>7.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>129900.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>7.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>163000.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>3.00</td>\n",
       "      <td>7.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>214000.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>7.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.00</td>\n",
       "      <td>2.00</td>\n",
       "      <td>755000.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       BsmtFinType2  HeatingQC  KitchenQual  Functional  GarageFinish  \\\n",
       "count       1456.00    1456.00      1456.00     1456.00       1456.00   \n",
       "mean           0.27       3.14         2.51        6.84          0.77   \n",
       "std            0.87       0.96         0.66        0.67          0.81   \n",
       "min            0.00       0.00         1.00        1.00          0.00   \n",
       "25%            0.00       2.00         2.00        7.00          0.00   \n",
       "50%            0.00       4.00         2.00        7.00          1.00   \n",
       "75%            0.00       4.00         3.00        7.00          1.00   \n",
       "max            5.00       4.00         4.00        7.00          2.00   \n",
       "\n",
       "       GarageQual  GarageCond  PavedDrive  SalePrice  \n",
       "count     1456.00     1456.00     1456.00    1456.00  \n",
       "mean         1.98        1.98        1.86  180810.06  \n",
       "std          0.24        0.23        0.50   79462.62  \n",
       "min          0.00        0.00        0.00   34900.00  \n",
       "25%          2.00        2.00        2.00  129900.00  \n",
       "50%          2.00        2.00        2.00  163000.00  \n",
       "75%          2.00        2.00        2.00  214000.00  \n",
       "max          4.00        4.00        2.00  755000.00  "
      ]
     },
     "execution_count": 1775,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# descriptive statistics of numerica values - part 3\n",
    "round(train[num_attributes].iloc[:, 24:].describe(), 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1354,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:4: SettingWithCopyWarning:\n",
      "\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(1456, 35)"
      ]
     },
     "execution_count": 1354,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dataset of numerical values for constructing a correllation matrix\n",
    "num_attributes.extend(('YearBuilt', 'YearRemodAdd'))\n",
    "df_corr = train[num_attributes]\n",
    "df_corr['SalePrice'] = train['SalePrice']\n",
    "df_corr.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1785,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 1785,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# compute correlation values\n",
    "corr = df_corr.corr()\n",
    "\n",
    "#graph\n",
    "mask = np.triu(corr)\n",
    "plt.figure(figsize=(15, 15))\n",
    "sns.heatmap(corr,\n",
    "            annot=True,\n",
    "            fmt='.1f',\n",
    "            cmap='coolwarm',\n",
    "            square=True,\n",
    "            mask=mask,\n",
    "            linewidths=1,\n",
    "            cbar=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The last row shows the correlation values of different attributes with the sale price. As can be seen, some attributes (dark red) are positively correlation with the sale prices, whereas others (dark blue) are negatively correlated. In the meantime, some attributes have zero correlation with the sale price, meaning they will bring almost no explanatory power in our predictive model. Also, there are a few strong correlations within the explanatory variables, which may cause multicollinearity in regression models. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Prepararation of datasets for modelling <a class=\"anchor\" id=\"9-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To begin with, let's check for outliers and remove them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1783,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Total_area', ylabel='SalePrice'>"
      ]
     },
     "execution_count": 1783,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# check for outliers\n",
    "plt.figure(figsize=(13,5))\n",
    "sns.regplot(x=train['Total_area'], y=train['SalePrice'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1149,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='SalePrice'>"
      ]
     },
     "execution_count": 1149,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# check for the 2 outlier sales price datapoints in a boxplot\n",
    "sns.boxplot(x=train['SalePrice'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1781,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([], dtype=int64)"
      ]
     },
     "execution_count": 1781,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# capture four outliers\n",
    "outliers = np.where(train['GrLivArea'] > 4500)[0]\n",
    "outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1153,
   "metadata": {},
   "outputs": [],
   "source": [
    "#drop the outliers from the train set\n",
    "train = train.drop(train.index[outliers])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, prepare the training and test datasets by removing the unused attributes and making sure the data types are consistent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1154,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert all columns to float\n",
    "SalePrice = train['SalePrice']\n",
    "X = train.drop([\"Id\", \"SalePrice\"], axis = 1, inplace = False).astype(float)\n",
    "\n",
    "# targets \n",
    "y = SalePrice.values\n",
    "\n",
    "# set for Kaggle prediction\n",
    "X_test = test.drop([\"Id\", \"SalePrice\"], axis = 1, inplace = False).astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1362,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1456, 221), (1459, 221))"
      ]
     },
     "execution_count": 1362,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Now the number of attributes in train and test sets should match\n",
    "X.shape, X_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Unsupervised learning <a class=\"anchor\" id=\"10-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Principal component analysis <a class=\"anchor\" id=\"11-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, I appply Principal component analysis (PCA) to the training dataset. PCA is an unsupervised technique helps reduce the dimensionality of high-dimensional datasets while preserving the original structure and relationships inherent to the original dataset. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1158,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">1 0.535 (0.105)\n",
      ">2 0.602 (0.087)\n",
      ">3 0.664 (0.098)\n",
      ">4 0.687 (0.086)\n",
      ">5 0.737 (0.066)\n",
      ">6 0.745 (0.063)\n",
      ">7 0.754 (0.060)\n",
      ">8 0.756 (0.055)\n",
      ">9 0.757 (0.061)\n",
      ">10 0.753 (0.063)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD+CAYAAAAqP/5ZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Il7ecAAAACXBIWXMAAAsTAAALEwEAmpwYAAAZHUlEQVR4nO3dfZAc9X3n8fdHix6IA1iy5NgGBFuJHOshdbHZA8dWnb1nY4PtCFw4iWT8oJNiQZWlXPx0glvOlnBUtuKUE98GR8as4iSOVkUpieDKBO7qdq8uSpxEC0YKQsFWyAOCO2tBijGH5V1pv/dH94rRMLs70sx09/R8XlVTbPc001/Nzn76N7/+df8UEZiZWfublXcBZmbWHA50M7OScKCbmZWEA93MrCQc6GZmJXFBXjteuHBhXHnllXnt3sysLT388MPPRsSiWs/lFuhXXnklIyMjee3ezKwtSfrnqZ5zl4uZWUk40M3MSsKBbmZWEg50M7OScKCbmZWEA93MrCQc6GZmJeFANzMridwuLDIzaxVJdW1XtvkgHOhmVjq1glpS5gFez4GlmTU50M3MWqQ6rFt9UHEfuplZSbiFblYiWX/Ft2JxoJuVSNZf8WvxQSU/DnRre0UIkCLUUBRFOKh0Kge6tb0iBEgRajDzSVEzs5JwoJuZlYQD3cysJBzoZmYl4ZOidt48ssOsWOpqoUu6TtITko5Iuq3G84slDUv6jqSDkt7T/FKtaCLirMdU68wsGzMGuqQu4C7gemAZsEbSsqrN7gDujYg3AquBrza7UDMzm149LfSrgSMR8WREjAG7gRuqtgng4vTnS4BnmleimZnVo55AvxR4qmL5aLqu0hbgQ5KOAg8Am2q9kKQNkkYkjYyOjp5HuWY2acGCBUia9gFM+/yCBQty/lc0h9+LRLNGuawBvhERlwHvAf5I0steOyLujoieiOhZtGhRk3bdmWb68NZ7g39rXydOnHjZOYtzfZw4caLhOmYKU5j589pomPq9SNQzyuVp4PKK5cvSdZXWA9cBRMS3Jc0DFgLHzrsym5YvNbeimAzTRpSlAZL3e1FPC30/sERSt6Q5JCc976/a5l+Ad6TFLAXmAe5TsdLKuyVmVsuMLfSIOCVpI/AQ0AXsjIhDku4ERiLifuBTwNclfYLkBOnacHPRWmDBggV1fTWerpUzf/58jh8/3lAdebfEzGqp68KiiHiA5GRn5brPVvz8OPDW5pZm9nIOUrOp+UpRM2tIfO5i2HJJ46/R5jUUgfLqGenp6YmRkZFc9l1GRTgpmkUNzdhHWV6jCDUU5TWKUAPQ8EHlpdf5wZRPSXo4InpqPecWuplZk2jr8805sGw5v//XgW7WptzNYNUc6GbnoQhhmndr0IrHgW52HhymVkQOdGsrRWgZW/sZfXGUz/zvz/Bbb/stFl64MO9yWsaBbm3FLeP6FSHEilADwI6DO3jk+4+w48AO7njzHbnV0Wqegs6spCpDrJNrGH1xlPuO3EcQ7D2yl2d/9Gxudax9cG1L9+9ANyuhIoRYEWqA5KAyERMATMREbgeXLA5uDnSzEipCiBWhhsmDyvjEOADjE+O5HFyyOrg50M2aLIuv1jPtP+8QK0INcPZBZVIeB5esDm4OdLMmy7vfuAghVoQaAA4cO3DmoDJpfGKcR489mlkNWR7cPMrFSiXvURXVX61v/Te3Zl5HEUKsCDUA7Fm1J9P91TLdwa3ZI24c6FYqeQ9Pq/XVOus6ihBiRaihKLI8uPluiyWRxZ0O651cYjqNTi4x3b9z9MVRrv/T6/nx6R8zt2suD970YM3WcavuzFe5/0mtrKModxgswmsUoYasXmO6uy26D93qVpSJeKeS96iKovQbW+dyoFspFGFURVH6ja1zuQ/dSiHLE09Tcb+x5c0tdCsFt47N3EK3knDr2MwtdDOz0nCgm5mVhLtczKwUJDX0/8+fP79JleTHgd4G6r2gZ7oPdKMX9IBnCyoih1iinot5srj4Lm8O9DYweUFPIxr9wwfPFlQ0DjGr5kA/R/UEo/+AzCwPdQW6pOuArwBdwD0R8cWq538b6E0XfwJ4dUS8sol1FkZ1WLsFZOaun0p5vhczBrqkLuAu4FrgKLBf0v0R8fjkNhHxiYrtNwFvPO+KzKytzNSg6aRGT97vRT3DFq8GjkTEkxExBuwGbphm+zXAYDOKMysySQ09ytQqtWKop8vlUuCpiuWjwDW1NpR0BdANDE3x/AZgA8DixYvPqVCzIsm7JWZWS7MvLFoN7ImI07WejIi7I6InInoWLVrU5F2bmXW2egL9aeDyiuXL0nW1rMbdLR0py4mR3dVhVls9XS77gSWSukmCfDXwweqNJL0BmA98u6kVWltc0JPV1G8ee202tRkDPSJOSdoIPEQybHFnRBySdCcwEhH3p5uuBnaH/5KarugX9BRhYmQzq3McekQ8ADxQte6zVctbmleWtZMiTIxsZm12t8V6+kctW0WY+s3MEm0V6NUTDk+1zrLjiZHNiqOtAt2Kx1O/mRWHb85lDfHUb2bF4RZ6CWQ5BtzMisuBXgKVY8DNrHM50GewYMGCGUfVzDTyZsGCBS2rr3oMuFvpZp3LgT6DydmCGnnUM33c+ao1BtzMOpMDvY15DLiZVXKgtzGPATezSh622MbyGAPuqcbMisuB3sayHgPuSR2sHrUO+tXr/DlpDQe6mTWVwzo/7kM3MysJt9DbhPuuzWwmDvQ24Fl6zKwehe1ymekKzXqu0mzlFZpmZkVT2Bb65BWajfCEF2bWSQrbQjczs3NT2BZ6UcTnLoYtlzT+GmaWmam+nZd9PLwDfQba+nxTun48hbZZdsoW1PVyl4uZWUk40M3MSsKBbmZWEg50M7OS8ElRM7MWyfrOk26hN2D0xVHWPrjWMwSZWU31TFHZTHUFuqTrJD0h6Yik26bY5pclPS7pkKRdTa2yoHYc3MEj33/EMwSZWSHMGOiSuoC7gOuBZcAaScuqtlkC3A68NSKWA7/e/FKLZXI+zyA8j6eZFUI9LfSrgSMR8WREjAG7gRuqtvkYcFdEnACIiGPNLbN4Kufz9DyeZlYE9QT6pcBTFctH03WVXg+8XtJfSvprSdfVeiFJGySNSBoZHR09v4oLYLJ1Pjmf5/jEuFvpZpa7Zp0UvQBYArwdWAN8XdIrqzeKiLsjoiciehYtWtSkXWevsnU+ya30zlbPrZ07qQ7LRz3DFp8GLq9YvixdV+ko8DcRMQ78o6TvkgT8/qZUWTAHjh040zqfND4xzqPHHs2noA5XhEmJi3LvkKLUYfmoJ9D3A0skdZME+Wrgg1Xb7CVpmf++pIUkXTBPNrHOQtmzak/eJVgFh5hZYsYul4g4BWwEHgIOA/dGxCFJd0palW72EPCcpMeBYeAzEfFcq4o2M7OXU16tm56enhgZGZny+WbMkVmm1yjCPtqhBrOyk/RwRPTUes5XipqZlYQD3cysJBzoZmYl4UA3MysJB7qZWUk40M3MSsKBbmZWEp6xqA6N3v9i/vz5TarEzGxqDvQZzHShjC+mMbOicJeLmVlJONDNzErCgW5mVhIOdDOzknCgm5mVRNsG+uiLo6x9cG3HzuPpqcbMrFrbBvqOgzt45PuPdOw8nhEx48PMOktbBvroi6Pcd+Q+gmDvkb0d20o3M6vUloG+4+AOJmICgImY6NhWuplZpcJOQceWS2quHu2axfWXvY4fz3rpWDR3YoIHjz7DwtMTNV7nB42WOi1fKfoSvxdmrTfdFHSFvfRfW5+vGQ47/vrzTHzvz2Bi/My6iQvmsuPaT3HHm+84+zUkYkurK7XBwUG2bdsGwIoVK+jr62PNmjU5V2XWedquy+XAsQOMV4Q5wPjEOI8eezSfgjrc4OAgfX199Pf3A9Df309fXx+Dg4M5V2bWeQrb5dKMr+9ZdAF0cjdDPUMjO/W9MWuV6bpc2q6FbsUREUiiu7uboaEhxsbGGBoaoru7u6MPdGZ5caBbQ+bMmcPGjRvp7e1l9uzZ9Pb2snHjRubMmZN3aWYdx4FuDRkbG6O/v5/h4WHGx8cZHh6mv7+fsbGxvEsz6ziFHeVi7WHZsmXceOONbNq0icOHD7N06VJuvvlm9u7dm3dpZh3HLXRrSF9fH7t27aK/v5+TJ0/S39/Prl276Ovry7s0s45TVwtd0nXAV4Au4J6I+GLV82uBLwFPp6t+NyLuaWKdVlCT480rW+jbtm3zOHSzHMw4bFFSF/Bd4FrgKLAfWBMRj1dssxboiYiN9e7YwxbNzM5do8MWrwaORMSTETEG7AZuaGaB1t4GBwdZsWIFXV1drFixwhcVmeWknkC/FHiqYvlouq7aTZIOStoj6fJaLyRpg6QRSSOjo6PnUa5VyztMK68UnexD95WiZjmp457aHyDpN59c/jBJH3nlNq8C5qY/3wIMzfS6V111VUwnKa0xzXiNIuxjKrt27Yru7u4YGhqKsbGxGBoaiu7u7ti1a1dmNSxfvjyGhobOWjc0NBTLly/PrAazTgKMxBS5Wk8f+i8AWyLi3eny7emB4AtTbN8FHI+I2rdLTLkPvXErVqygv7+f3t7eM+uGh4fZtGkTjz32WCY1dHV1cfLkSWbPnn1m3fj4OPPmzeP06dOZ1GDWSRrtQ98PLJHULWkOsBq4v2oHr61YXAUcPt9irX6HDx9m5cqVZ61buXIlhw9n9/YvXbqUffv2nbVu3759LF26NLMazCwxY6BHxClgI/AQSVDfGxGHJN0paVW62a9JOiTpAPBrwNpWFWwvKUKY9vX1sX79+rOuFF2/fr3HoZvlYaq+mFY/3IfeuCL0oU/WsXz58pg1a1YsX7488/2bdRIa6UNvFfehN8fk5BKTF/V4cgmzcvPtc83MOoBvztXGJseADwwMsHLlSvbt28f69esB3Eo360Buobexbdu2MTAwcNa9yAcGBs7M72lmncV96A3Ksw/dY8DNOo/70Etq6dKlbN269axL/7du3eox4GYdqtCBLqmhx/z58/P+J7RUb28v27dvZ926dfzwhz9k3bp1bN++/awrR82scxQ20KcaZ1n5mGm748eP5/yvaK3h4WE2b97Mzp07ueiii9i5cyebN29meHg479LMLAeF7UOvR95jwPOuwX3oZp3HfeglVYRL/82sOBzobcz3UTGzSr6wqI15Pk8zq+Q+9AYVoQYz6xzuQzcz6wAOdDOzknCgm5mVhAPdzKwkHOhmZiXhYYvnSNKM6zzqxczy4EA/Rw5rMysqd7mYmZWEA93MrCQc6GZmJeFANzMrCQe6mVlJONDb3ODg4Flzig4ODuZdkpnlxMMW29jg4CB9fX0MDAywcuVK9u3bx/r16wF8C12zDlRXC13SdZKekHRE0m3TbHeTpJBU89aO1lzbtm1jYGCA3t5eZs+eTW9vLwMDA2zbti3v0swsBzPeD11SF/Bd4FrgKLAfWBMRj1dtdxHwLWAOsDEipr3ZeVnuh54nzylq1nkavR/61cCRiHgyIsaA3cANNbb7PLAdOHneldo58ZyiZlapnkC/FHiqYvlouu4MSW8CLo+Ib033QpI2SBqRNDI6OnrOxRZJEU5Gek5RM6vU8ElRSbOALwNrZ9o2Iu4G7oaky6XRfeelKCcjPaeomVWqpw/9F4AtEfHudPl2gIj4Qrp8CfAPwAvp//Ia4Diwarp+9HbuQ1+xYgX9/f309vaeWTc8PMymTZt47LHHMq/HzDrHdH3o9QT6BSQnRd8BPE1yUvSDEXFoiu3/F/DpMp8U9clIM8tLQydFI+IUsBF4CDgM3BsRhyTdKWlVc0ttDz4ZaWZFVFcfekQ8ADxQte6zU2z79sbLKrbJk5HVfege/21mefKVoufBJyPNrIhm7ENvlXbuQzczy0ujFxaZmVkbcKCbmZWEA93MrCQc6GZmJeFANzMrCQe6mVlJONDNzErCgW5mVhIOdDOzknCgm5mVhAPdzKwkHOhmZiXhQDczKwkHuplZSTjQzcxKwoFuZlYSDnQzs5JwoJuZlYQD3cysJBzoZmYl4UA3MysJB7qZWUk40M3MSsKBbmZWEg50M7OSqCvQJV0n6QlJRyTdVuP5WyX9naRHJe2TtKz5pZqZ2XRmDHRJXcBdwPXAMmBNjcDeFRE/FxE/D/wm8OVmF2pmZtOrp4V+NXAkIp6MiDFgN3BD5QYR8XzF4iuAaF6JZmZWjwvq2OZS4KmK5aPANdUbSfo48ElgDvDva72QpA3ABoDFixefa61mZjaNpp0UjYi7IuKngc3AHVNsc3dE9EREz6JFi5q1azMzo75Afxq4vGL5snTdVHYDNzZQ05QknfWYap2ZWSeqJ9D3A0skdUuaA6wG7q/cQNKSisX3At9rXokviYgZH2ZmnWrGPvSIOCVpI/AQ0AXsjIhDku4ERiLifmCjpHcC48AJ4KOtLNrMzF6unpOiRMQDwANV6z5b8fN/bHJdZmZ2jnylqJlZSTjQzcxKwoFuZlYSDnQzs5JwoJuZlYQD3cysJJTXxTiSRoF/bvBlFgLPNqGcdq8BilFHEWqAYtRRhBqgGHUUoQYoRh3NqOGKiKh575TcAr0ZJI1ERE+n11CUOopQQ1HqKEINRamjCDUUpY5W1+AuFzOzknCgm5mVRLsH+t15F0AxaoBi1FGEGqAYdRShBihGHUWoAYpRR0traOs+dDMze0m7t9DNzCzlQDczKwkHujWNPGUUkl5RgBpe499FZ2rLQJfUlfP+f0ZSj6S5OdawXNLbJL0qrxrSOlZK+jBAREQeQSLpFyXlfk9+STcA2yW9Osca3g38GWdPG5lHHW+W9OH0v3NyqmFJ+nc6K+/MqNaqv5O2CnRJrweIiNN5/YIkvQ/4U+BLwDcma8q4huuBQeATwB9Kek0ONcyS9JPA14DbJd0KZ0I9s8+VpHcBnwcez2qfU9TxNmA7cF9EHMuphnelNbwW+FQeNaR1rCIZzfFO4NPAFTnUcCOwB7gd+DJwS57fniRdkzbA/i20rvHTNoGeBumjknZBPqEu6S0kQf7RiOglmW7vtoxreDvwFeBXI+JGYAxYkWUNABExEREvAH8ADABvkfSJyeeyqCH9ffwRsCEi/oekSyRdIeknsth/lauAe9I6Xifp2vSP+JIsdp5OAflV4GZgCbBU0r/LYt9VdbwK+DjwwYj4KPA88POSXi1pXoY13AKsiYibgIPAfwA+KemiLGqoqud64Jskv5v/LGkAWhPqbRHo6ZF1I/DrwJikb0JuLfXtEfGd9OfPAQsy7nr5PnBLRPxt2jK/hmRO169J+kAOXR6nSL7e/wFwtaQvS/qCEq3+fD1HMo/ta9M/4r3A75F8c8r6vThV8fMeYB3JZ/YuSfMz2H8X8JGIOAS8AngCWA6Zn9s4BVwIvEHSxcDbgY8AvwPckVEr+RTwk8BrACJiJ/BPJPdReV8G+z8jzaePAndGxIb05zdI2pPW1tRQb4tAj4j/R/IHsovkK9y8ylDPsJS/IelumfxFzSX5Onlxuq7l/dkRcTgihtPF9cBX05b6t4EPkHxos3Qf8H8j4n8CI8CtwMWRaGlLPSKeAN4L/DZwgOTz8T7gQeAmIIsgnTQMfEzSbuDrEbGG5ID/AnB1q3ceEQ9FxF9JmhUR/wp8C/icpJ+LDC82iYgfAP+VpKvjvwO/HxG/CNwDXAb8TEY1/DGwLu3H3wb8mKRb7p2t3n9VLaeB71QsPx8RbwV+StLX0nVN+/20RaADRMQzEfFCRDxL8nXqwslQl/QmSW/IoIbTEfF8uijgX4HjETEq6WbgNyRd2Oo6KurZFhG/kf78DZIDS9Ynw34E/Kykj5GE+ReBxZJuyWLnEXGAJMS/GBFfT7uCdpKE+eIsakjr+DuSxsY1QHe67kmSlnPNO+O1qI6J9L8PkvRjvy+jb0uVNewhCc6/IA2ziBgCLiK7/vRB4M+BXuDCiPhQRHyNJEgvbvXOq86tPQ1sllT5eXw/8CpJy5u53wua+WJZiYjn0sD4kqS/J/mj6c24hlPAC5KekvQF4F3A2oj4URb7l6TKI7ukm4CfAp7JYv+TIuIZSU8B/wX4eET8N0m9wJEMa3icipOi6XuxCPg/WdWQ+nOSVvkWSZO3hn4jyUEuDwdITpz/ZsbfZImIE5KGgF+WNAbMIznQHcxo/z8A/ljS4ORBTtJHgAVAS9+L9HzfvZLuj4jVEfFNST8L/KWkt0bEv0TEs5JOkXSPNW/f7Xzpf3oSbjNwbdpCynLfAmYDh9P/viMivpdlDWkdc4EPAZ8EfiUiHsuhhsuBV0fEw+nyrKxOjFbVIZKTX58GfintT86cpDeRdH/NBb6R9WezqpZ7gf8UEf+Uw75fSdJ/fhNwMq3jQNZ1pLWsI/lc/Eorfx/pOYI/IemafQswN+1+Q9LngVUkJ68XkpwkfW9E/GPT9t+ugZ6eaLoX+FREZHLUn6KOtcD+HMNjNnAt8A9pn3Juqr815LF/4G0kffp/n1cdRZD376JSOrJEFd2VedRwBTA7Ilr+zVHS60hG98wDdgDjFaH+fpKTtVcBv9PsBljbBjqApHkRcTLnGgrzh2NmxZIOlLgbGIuINWmf+QsR0ehsbTW1zUnRWvIO87QGh7mZ1RQRz5EM4jgp6QmSUWEt68Nv60A3Myu6dGTeQeAS4P0RcbRV+3Kgm5m1UHq+7z3Au1p9gryt+9DNzNpBVuf7HOhmZiXhLhczs5JwoJuZlYQD3cysJBzoZmYl4UA3MysJB7qZWUn8f8ZKBP01rllkAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#ignore warnings\n",
    "warnings.simplefilter('ignore')\n",
    "\n",
    "# get a list of models to evaluate\n",
    "def get_models():\n",
    "    models = dict()\n",
    "\n",
    "    for i in range(1, 11):\n",
    "        steps = [('scaler', MinMaxScaler()), ('pca', PCA(n_components=i)), ('m', GradientBoostingRegressor(random_state = 10))]\n",
    "        models[str(i)] = Pipeline(steps=steps)\n",
    "    return models\n",
    " \n",
    "# evaluate a given model using cross-validation\n",
    "def evaluate_model(model, X, y):\n",
    "    cv = RepeatedStratifiedKFold(n_splits=10, n_repeats=3, random_state=1)\n",
    "    scores = cross_val_score(model, X, y, scoring='r2', cv=cv, n_jobs=-1, error_score=np.nan)\n",
    "    return scores\n",
    " \n",
    "# get the models to evaluate\n",
    "models = get_models()\n",
    "\n",
    "# evaluate the models and store results\n",
    "results, names = list(), list()\n",
    "for name, model in models.items():\n",
    "    scores = evaluate_model(model, X, y)\n",
    "    results.append(scores)\n",
    "    names.append(name)\n",
    "    print('>%s %.3f (%.3f)' % (name, mean(scores), std(scores)))\n",
    "# plot model performance for comparison\n",
    "pyplot.boxplot(results, labels=names, showmeans=True)\n",
    "pyplot.xticks(rotation=45)\n",
    "pyplot.show()\n",
    "\n",
    "warnings.simplefilter('default')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As can be seen above, the explained variance ratio starts to level off after 5 components and reaches a peak at 8 components. Next, I'll use 8 components to do k-means clustering. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## KMeans clustering <a class=\"anchor\" id=\"13-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, I apply the k-means clustering method to the selected components from the PCA analysis. Using K-means clustering, my objective is to identify clusters of houses with similar charateristics. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1164,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# merge explanatory attributes and target values for cluster analysis\n",
    "df_kmeans = X.merge(pd.Series(y).rename('SalePrice'), left_index=True, right_index=True)\n",
    "\n",
    "# keep column names which will be necessary after scaling the dataset\n",
    "df_kmeans_columns = df_kmeans.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1165,
   "metadata": {},
   "outputs": [],
   "source": [
    "# scale the dataset \n",
    "df_kmeans = MinMaxScaler().fit_transform(df_kmeans)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1216,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# identify the opminal number of clusters using the elbow method\n",
    "inertia = []\n",
    "K = range(1,12)\n",
    "for k in K:\n",
    "    kmeanModel = KMeans(n_clusters=k, random_state = 42)\n",
    "    kmeanModel.fit(df_kmeans)\n",
    "    inertia.append(kmeanModel.inertia_)\n",
    "    \n",
    "plt.figure(figsize=(16,8))\n",
    "plt.plot(K, inertia, 'bx-')\n",
    "plt.xlabel('k')\n",
    "plt.ylabel('Inertia')\n",
    "plt.title('The Elbow Method showing the optimal k')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1217,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# run k-means clustering with 4 clusters\n",
    "kmeans = KMeans(n_clusters=4, random_state=0).fit(df_kmeans)\n",
    "\n",
    "#get labels\n",
    "labels = kmeans.labels_\n",
    "\n",
    "#merge labels \n",
    "df_kmeans = pd.DataFrame(df_kmeans, columns = df_kmeans_columns)\n",
    "df_kmeans = df_kmeans.merge(pd.Series(labels).rename('Labels'), left_index=True, right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1789,
   "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>Cluster</th>\n",
       "      <th>Number of observations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>286</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Cluster  Number of observations\n",
       "0        3                     458\n",
       "1        1                     393\n",
       "2        2                     315\n",
       "3        0                     286"
      ]
     },
     "execution_count": 1789,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check how the four clusters divide the dataset \n",
    "clusters = pd.DataFrame(df_kmeans['Labels'].value_counts()).reset_index()\n",
    "clusters.columns = [\"Cluster\", \"Number of observations\"]\n",
    "clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1219,
   "metadata": {},
   "outputs": [],
   "source": [
    "# the highest correlated attributes with price\n",
    "high_corr = corr['SalePrice'][abs(corr['SalePrice'])> 0.6].sort_values().index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1220,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>OverallQual</th>\n",
       "      <th>OverallCond</th>\n",
       "      <th>YearBuilt</th>\n",
       "      <th>YearRemodAdd</th>\n",
       "      <th>MasVnrArea</th>\n",
       "      <th>ExterQual</th>\n",
       "      <th>ExterCond</th>\n",
       "      <th>BsmtQual</th>\n",
       "      <th>...</th>\n",
       "      <th>Total_baths</th>\n",
       "      <th>Total_lot_area</th>\n",
       "      <th>Total_porch_area</th>\n",
       "      <th>Total_finished_basement</th>\n",
       "      <th>Pool_dummy</th>\n",
       "      <th>2floor_dummy</th>\n",
       "      <th>Garage_dummy</th>\n",
       "      <th>Basement_dummy</th>\n",
       "      <th>Fireplace_dummy</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Labels</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.131</td>\n",
       "      <td>0.031</td>\n",
       "      <td>0.486</td>\n",
       "      <td>0.616</td>\n",
       "      <td>0.479</td>\n",
       "      <td>0.384</td>\n",
       "      <td>0.023</td>\n",
       "      <td>0.360</td>\n",
       "      <td>0.527</td>\n",
       "      <td>0.371</td>\n",
       "      <td>...</td>\n",
       "      <td>0.435</td>\n",
       "      <td>0.032</td>\n",
       "      <td>0.139</td>\n",
       "      <td>0.097</td>\n",
       "      <td>0.003</td>\n",
       "      <td>0.735</td>\n",
       "      <td>0.878</td>\n",
       "      <td>0.959</td>\n",
       "      <td>0.328</td>\n",
       "      <td>0.173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.181</td>\n",
       "      <td>0.052</td>\n",
       "      <td>0.483</td>\n",
       "      <td>0.597</td>\n",
       "      <td>0.671</td>\n",
       "      <td>0.425</td>\n",
       "      <td>0.049</td>\n",
       "      <td>0.365</td>\n",
       "      <td>0.527</td>\n",
       "      <td>0.416</td>\n",
       "      <td>...</td>\n",
       "      <td>0.390</td>\n",
       "      <td>0.045</td>\n",
       "      <td>0.150</td>\n",
       "      <td>0.274</td>\n",
       "      <td>0.007</td>\n",
       "      <td>0.048</td>\n",
       "      <td>0.937</td>\n",
       "      <td>0.959</td>\n",
       "      <td>0.472</td>\n",
       "      <td>0.183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.169</td>\n",
       "      <td>0.038</td>\n",
       "      <td>0.693</td>\n",
       "      <td>0.507</td>\n",
       "      <td>0.939</td>\n",
       "      <td>0.879</td>\n",
       "      <td>0.095</td>\n",
       "      <td>0.643</td>\n",
       "      <td>0.510</td>\n",
       "      <td>0.742</td>\n",
       "      <td>...</td>\n",
       "      <td>0.539</td>\n",
       "      <td>0.039</td>\n",
       "      <td>0.206</td>\n",
       "      <td>0.283</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.986</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.612</td>\n",
       "      <td>0.235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.189</td>\n",
       "      <td>0.048</td>\n",
       "      <td>0.671</td>\n",
       "      <td>0.538</td>\n",
       "      <td>0.888</td>\n",
       "      <td>0.779</td>\n",
       "      <td>0.109</td>\n",
       "      <td>0.576</td>\n",
       "      <td>0.513</td>\n",
       "      <td>0.661</td>\n",
       "      <td>...</td>\n",
       "      <td>0.711</td>\n",
       "      <td>0.045</td>\n",
       "      <td>0.230</td>\n",
       "      <td>0.191</td>\n",
       "      <td>0.006</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.994</td>\n",
       "      <td>0.775</td>\n",
       "      <td>0.238</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4 rows × 222 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        LotFrontage  LotArea  OverallQual  OverallCond  YearBuilt  \\\n",
       "Labels                                                              \n",
       "1             0.131    0.031        0.486        0.616      0.479   \n",
       "3             0.181    0.052        0.483        0.597      0.671   \n",
       "0             0.169    0.038        0.693        0.507      0.939   \n",
       "2             0.189    0.048        0.671        0.538      0.888   \n",
       "\n",
       "        YearRemodAdd  MasVnrArea  ExterQual  ExterCond  BsmtQual  ...  \\\n",
       "Labels                                                            ...   \n",
       "1              0.384       0.023      0.360      0.527     0.371  ...   \n",
       "3              0.425       0.049      0.365      0.527     0.416  ...   \n",
       "0              0.879       0.095      0.643      0.510     0.742  ...   \n",
       "2              0.779       0.109      0.576      0.513     0.661  ...   \n",
       "\n",
       "        Total_baths  Total_lot_area  Total_porch_area  \\\n",
       "Labels                                                  \n",
       "1             0.435           0.032             0.139   \n",
       "3             0.390           0.045             0.150   \n",
       "0             0.539           0.039             0.206   \n",
       "2             0.711           0.045             0.230   \n",
       "\n",
       "        Total_finished_basement  Pool_dummy  2floor_dummy  Garage_dummy  \\\n",
       "Labels                                                                    \n",
       "1                         0.097       0.003         0.735         0.878   \n",
       "3                         0.274       0.007         0.048         0.937   \n",
       "0                         0.283       0.000         0.000         0.986   \n",
       "2                         0.191       0.006         1.000         1.000   \n",
       "\n",
       "        Basement_dummy  Fireplace_dummy  SalePrice  \n",
       "Labels                                              \n",
       "1                0.959            0.328      0.173  \n",
       "3                0.959            0.472      0.183  \n",
       "0                1.000            0.612      0.235  \n",
       "2                0.994            0.775      0.238  \n",
       "\n",
       "[4 rows x 222 columns]"
      ]
     },
     "execution_count": 1220,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# compare cluster differences across all features\n",
    "table_all = round(df_kmeans.groupby('Labels').mean().sort_values('SalePrice'),3)\n",
    "table_all"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In terms of prices, there are two main categories in the clusters. Clusters 0 and 1 are similarly low priced, whereas clusters 2 and 3 are similarly high priced"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1241,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>SaleType_Con</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Condition1_RRNe</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Neighborhood_CollgCr</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Neighborhood_NoRidge</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Neighborhood_NridgHt</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Exterior2nd_AsphShn</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Neighborhood_StoneBr</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Neighborhood_Timber</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Neighborhood_Veenker</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Condition1_PosA</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RoofMatl_WdShake</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MSSubClass_85</th>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MSSubClass_60</th>\n",
       "      <td>184.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Neighborhood_Gilbert</th>\n",
       "      <td>68.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LotConfig_CulDSac</th>\n",
       "      <td>36.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Neighborhood_NWAmes</th>\n",
       "      <td>31.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SaleType_New</th>\n",
       "      <td>26.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SaleCondition_Partial</th>\n",
       "      <td>16.625000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GarageType_BuiltIn</th>\n",
       "      <td>15.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GarageFinish</th>\n",
       "      <td>11.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            Value\n",
       "SaleType_Con                  inf\n",
       "Condition1_RRNe               inf\n",
       "Neighborhood_CollgCr          inf\n",
       "Neighborhood_NoRidge          inf\n",
       "Neighborhood_NridgHt          inf\n",
       "Exterior2nd_AsphShn           inf\n",
       "Neighborhood_StoneBr          inf\n",
       "Neighborhood_Timber           inf\n",
       "Neighborhood_Veenker          inf\n",
       "Condition1_PosA               inf\n",
       "RoofMatl_WdShake              inf\n",
       "MSSubClass_85                 inf\n",
       "MSSubClass_60          184.800000\n",
       "Neighborhood_Gilbert    68.666667\n",
       "LotConfig_CulDSac       36.000000\n",
       "Neighborhood_NWAmes     31.666667\n",
       "SaleType_New            26.600000\n",
       "SaleCondition_Partial   16.625000\n",
       "GarageType_BuiltIn      15.666667\n",
       "GarageFinish            11.500000"
      ]
     },
     "execution_count": 1241,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# the highest relative differences in attributes between cluster 2 and 0\n",
    "differ = pd.DataFrame(table_all.loc[2]/table_all.loc[1], columns = ['Value']).sort_values('Value', ascending = False).head(20)\n",
    "differ"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1356,
   "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>TotalBsmtSF</th>\n",
       "      <th>GarageArea</th>\n",
       "      <th>Total_bathrooms</th>\n",
       "      <th>GarageCars</th>\n",
       "      <th>BsmtQual</th>\n",
       "      <th>KitchenQual</th>\n",
       "      <th>ExterQual</th>\n",
       "      <th>Total_indoor_area</th>\n",
       "      <th>OverallQual</th>\n",
       "      <th>Total_area</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Labels</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.238</td>\n",
       "      <td>0.248</td>\n",
       "      <td>0.149</td>\n",
       "      <td>0.327</td>\n",
       "      <td>0.371</td>\n",
       "      <td>0.394</td>\n",
       "      <td>0.360</td>\n",
       "      <td>0.281</td>\n",
       "      <td>0.486</td>\n",
       "      <td>0.285</td>\n",
       "      <td>0.173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.333</td>\n",
       "      <td>0.306</td>\n",
       "      <td>0.190</td>\n",
       "      <td>0.383</td>\n",
       "      <td>0.416</td>\n",
       "      <td>0.400</td>\n",
       "      <td>0.365</td>\n",
       "      <td>0.304</td>\n",
       "      <td>0.483</td>\n",
       "      <td>0.290</td>\n",
       "      <td>0.183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.463</td>\n",
       "      <td>0.435</td>\n",
       "      <td>0.299</td>\n",
       "      <td>0.568</td>\n",
       "      <td>0.742</td>\n",
       "      <td>0.691</td>\n",
       "      <td>0.643</td>\n",
       "      <td>0.406</td>\n",
       "      <td>0.693</td>\n",
       "      <td>0.385</td>\n",
       "      <td>0.235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.310</td>\n",
       "      <td>0.416</td>\n",
       "      <td>0.378</td>\n",
       "      <td>0.556</td>\n",
       "      <td>0.661</td>\n",
       "      <td>0.618</td>\n",
       "      <td>0.576</td>\n",
       "      <td>0.405</td>\n",
       "      <td>0.671</td>\n",
       "      <td>0.425</td>\n",
       "      <td>0.238</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        TotalBsmtSF  GarageArea  Total_bathrooms  GarageCars  BsmtQual  \\\n",
       "Labels                                                                   \n",
       "1             0.238       0.248            0.149       0.327     0.371   \n",
       "3             0.333       0.306            0.190       0.383     0.416   \n",
       "0             0.463       0.435            0.299       0.568     0.742   \n",
       "2             0.310       0.416            0.378       0.556     0.661   \n",
       "\n",
       "        KitchenQual  ExterQual  Total_indoor_area  OverallQual  Total_area  \\\n",
       "Labels                                                                       \n",
       "1             0.394      0.360              0.281        0.486       0.285   \n",
       "3             0.400      0.365              0.304        0.483       0.290   \n",
       "0             0.691      0.643              0.406        0.693       0.385   \n",
       "2             0.618      0.576              0.405        0.671       0.425   \n",
       "\n",
       "        SalePrice  \n",
       "Labels             \n",
       "1           0.173  \n",
       "3           0.183  \n",
       "0           0.235  \n",
       "2           0.238  "
      ]
     },
     "execution_count": 1356,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check how the highest correlated numerical attributes are different across the four clusters\n",
    "table_highcorr = round(df_kmeans.groupby('Labels')[high_corr].mean().sort_values('SalePrice'),3)\n",
    "table_highcorr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1228,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Total_bathrooms</th>\n",
       "      <td>2.536913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtQual</th>\n",
       "      <td>1.781671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GarageCars</th>\n",
       "      <td>1.700306</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GarageArea</th>\n",
       "      <td>1.677419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExterQual</th>\n",
       "      <td>1.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KitchenQual</th>\n",
       "      <td>1.568528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_area</th>\n",
       "      <td>1.491228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_indoor_area</th>\n",
       "      <td>1.441281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OverallQual</th>\n",
       "      <td>1.380658</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SalePrice</th>\n",
       "      <td>1.375723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TotalBsmtSF</th>\n",
       "      <td>1.302521</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Value\n",
       "Total_bathrooms    2.536913\n",
       "BsmtQual           1.781671\n",
       "GarageCars         1.700306\n",
       "GarageArea         1.677419\n",
       "ExterQual          1.600000\n",
       "KitchenQual        1.568528\n",
       "Total_area         1.491228\n",
       "Total_indoor_area  1.441281\n",
       "OverallQual        1.380658\n",
       "SalePrice          1.375723\n",
       "TotalBsmtSF        1.302521"
      ]
     },
     "execution_count": 1228,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#difference between high and low price clusters\n",
    "pd.DataFrame(table_highcorr.loc[2]/table_highcorr.loc[1], columns = ['Value']).sort_values('Value', ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1229,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Total_bathrooms</th>\n",
       "      <td>1.264214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_area</th>\n",
       "      <td>1.103896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SalePrice</th>\n",
       "      <td>1.012766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_indoor_area</th>\n",
       "      <td>0.997537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GarageCars</th>\n",
       "      <td>0.978873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OverallQual</th>\n",
       "      <td>0.968254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GarageArea</th>\n",
       "      <td>0.956322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExterQual</th>\n",
       "      <td>0.895801</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KitchenQual</th>\n",
       "      <td>0.894356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtQual</th>\n",
       "      <td>0.890836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TotalBsmtSF</th>\n",
       "      <td>0.669546</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Value\n",
       "Total_bathrooms    1.264214\n",
       "Total_area         1.103896\n",
       "SalePrice          1.012766\n",
       "Total_indoor_area  0.997537\n",
       "GarageCars         0.978873\n",
       "OverallQual        0.968254\n",
       "GarageArea         0.956322\n",
       "ExterQual          0.895801\n",
       "KitchenQual        0.894356\n",
       "BsmtQual           0.890836\n",
       "TotalBsmtSF        0.669546"
      ]
     },
     "execution_count": 1229,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#difference between 2 high price clusters\n",
    "pd.DataFrame(table_highcorr.loc[2]/table_highcorr.loc[0], columns = ['Value']).sort_values('Value', ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1230,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>TotalBsmtSF</th>\n",
       "      <td>1.399160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_bathrooms</th>\n",
       "      <td>1.275168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GarageArea</th>\n",
       "      <td>1.233871</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>GarageCars</th>\n",
       "      <td>1.171254</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BsmtQual</th>\n",
       "      <td>1.121294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_indoor_area</th>\n",
       "      <td>1.081851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SalePrice</th>\n",
       "      <td>1.057803</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_area</th>\n",
       "      <td>1.017544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KitchenQual</th>\n",
       "      <td>1.015228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ExterQual</th>\n",
       "      <td>1.013889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OverallQual</th>\n",
       "      <td>0.993827</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Value\n",
       "TotalBsmtSF        1.399160\n",
       "Total_bathrooms    1.275168\n",
       "GarageArea         1.233871\n",
       "GarageCars         1.171254\n",
       "BsmtQual           1.121294\n",
       "Total_indoor_area  1.081851\n",
       "SalePrice          1.057803\n",
       "Total_area         1.017544\n",
       "KitchenQual        1.015228\n",
       "ExterQual          1.013889\n",
       "OverallQual        0.993827"
      ]
     },
     "execution_count": 1230,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#difference between 2 low price clusters\n",
    "pd.DataFrame(table_highcorr.loc[3]/table_highcorr.loc[1], columns = ['Value']).sort_values('Value', ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1226,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Labels'>"
      ]
     },
     "execution_count": 1226,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plot the highly correlated features across the four clusters\n",
    "table_highcorr.plot.bar(figsize=(15,7))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Clusters: <br><br>\n",
    "\n",
    "Clusters 0 and 2 are close in price and total area <br>\n",
    "Cluster 0: expensive segment, more bathrooms <br>\n",
    "Cluster 2: expensive segment, large basement <br>\n",
    "\n",
    "Clusters 1 and 3 are close in price and total area <br>\n",
    "Cluster 3: cheaper segment, more basement and bathroom <br>\n",
    "Cluster 1: cheaper segment, all attributes inferior to cluster 3 <br><br>\n",
    "\n",
    "Overall, clusters 0 and 2 houses are more expensive, bigger, and better quality than clusters 1 and 3."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Supervised learning <a class=\"anchor\" id=\"14-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature selection <a class=\"anchor\" id=\"15-bullet\"></a>\n",
    "\n",
    "First, I reduce the number of features using recursive feature elimination (RFE).To do this, I evaluate a Gradient Boosting Regressor model with different numbers of features and select the number of features that yield the smallest root mean squared error."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1249,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">100 -21908.521 (3268.933)\n",
      ">205 -21877.138 (3314.816)\n",
      ">208 -21800.031 (3229.133)\n",
      ">221 -21984.383 (3293.911)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#use recursive feature elimination (RFE) to select features\n",
    "\n",
    "warnings.simplefilter('ignore')\n",
    "\n",
    "# get a list of models to evaluate\n",
    "def get_models():\n",
    "    models = dict()\n",
    "\n",
    "    for i in [100, 205, 208, 221]:\n",
    "        steps = [('scaler', MinMaxScaler()), ('rfe', RFE(estimator = GradientBoostingRegressor(random_state = 10), n_features_to_select= i))]\n",
    "        models[str(i)] = Pipeline(steps=steps)\n",
    "    return models\n",
    "\n",
    "# evaluate a given model using cross-validation\n",
    "def evaluate_model(model, X, y):\n",
    "    cv = RepeatedStratifiedKFold(n_splits=10, n_repeats=3, random_state=1)\n",
    "    scores = cross_val_score(model, X, y, scoring= 'neg_root_mean_squared_error', cv=cv, n_jobs=-1, error_score=np.nan)\n",
    "    return scores\n",
    " \n",
    "# get the models to evaluate\n",
    "models = get_models()\n",
    "\n",
    "# evaluate the models and store results\n",
    "results, names = list(), list()\n",
    "for name, model in models.items():\n",
    "    scores = evaluate_model(model, X, y)\n",
    "    results.append(scores)\n",
    "    names.append(name)\n",
    "    print('>%s %.3f (%.3f)' % (name, mean(scores), std(scores)))\n",
    "# plot model performance for comparison\n",
    "pyplot.boxplot(results, labels=names, showmeans=True)\n",
    "pyplot.xticks(rotation=45)\n",
    "pyplot.show()\n",
    "\n",
    "warnings.simplefilter('default')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Reducing the attributes to 208 results in the lowest RMSE. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1250,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:\n",
      "\n",
      "`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([ True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True, False, False,\n",
       "        True, False, False,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True, False, False,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True, False, False, False,  True, False, False, False, False,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True])"
      ]
     },
     "execution_count": 1250,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# eliminate features \n",
    "estimator = GradientBoostingRegressor(random_state = 10)\n",
    "selector = RFE(estimator, n_features_to_select=208)\n",
    "selector = selector.fit(X, y)\n",
    "selector.support_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1255,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1456, 208), (1459, 208))"
      ]
     },
     "execution_count": 1255,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# reduce features based on the RFE selection\n",
    "X_reduced = X[X.columns[selector.support_]]\n",
    "X_test_reduced = X_test[X_test.columns[selector.support_]]\n",
    "X_reduced.shape, X_test_reduced.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tune Hyperparameters with Bayes Search <a class=\"anchor\" id=\"17-bullet\"></a>\n",
    "\n",
    "Next, I use two distinct strategies to tune hyperparameters of the model and evaluate different models using cross-validation. First, I perform BayesSearchCV GradientBoostingRegressor on the reduced number of features. Next, I perform BayesSearchCV using XGBRegressor on the full list of features. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Search strategy 1 <a class=\"anchor\" id=\"18-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1706,
   "metadata": {},
   "outputs": [],
   "source": [
    "# pipeline of the scaler and regression model\n",
    "pipeline_gbr = Pipeline([\n",
    "        ('scaler', MinMaxScaler()),\n",
    "        ('GBR', GradientBoostingRegressor())\n",
    "    ])\n",
    "\n",
    "# parameters to search on\n",
    "parameters_gbr = {\n",
    "        'GBR__random_state': [10],\n",
    "        'GBR__learning_rate': (0.001, 0.06),\n",
    "        'GBR__n_estimators': (1000, 5000),\n",
    "        'GBR__subsample': (0.01, 0.8),\n",
    "        'GBR__max_depth': (3, 8),\n",
    "        'GBR__criterion': (\"friedman_mse\", \"mse\", \"mae\"), \n",
    "        'GBR__tol': [0.001],\n",
    "        'GBR__n_iter_no_change': [10]\n",
    "    }\n",
    "\n",
    "# initiate instance\n",
    "gbr_bayes_search = BayesSearchCV(estimator=pipeline_gbr, search_spaces=parameters_gbr, n_iter=100, n_jobs=-1, scoring= 'neg_root_mean_squared_error', cv = 5, verbose=1, random_state =10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1707,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   11.1s remaining:   16.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   12.6s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   14.0s remaining:   21.0s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   18.1s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   28.8s remaining:   43.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   39.5s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.2s remaining:    3.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.4s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    7.1s remaining:   10.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    9.5s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    0.7s remaining:    1.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    0.9s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.0min remaining:  1.5min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.1min finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  5.1min remaining:  7.7min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  5.7min finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   17.2s remaining:   25.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   22.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.6s remaining:    2.5s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.4s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   53.6s remaining:  1.3min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.1min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    8.7s remaining:   13.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   10.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   10.2s remaining:   15.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   12.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.1s remaining:    1.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    1.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.3s remaining:    2.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    1.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   24.5s remaining:   36.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   31.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    7.3s remaining:   10.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    9.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    0.5s remaining:    0.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    0.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  3.9min remaining:  5.8min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  4.0min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    3.7s remaining:    5.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    5.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    0.2s remaining:    0.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    0.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   36.7s remaining:   55.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   38.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.5s remaining:    2.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.1min remaining:  3.1min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.1min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    5.4s remaining:    8.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    7.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   42.7s remaining:  1.1min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   43.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.9s remaining:    2.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    0.1s remaining:    0.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    0.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   37.2s remaining:   55.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   39.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed: 11.6min remaining: 17.5min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed: 11.9min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    6.6s remaining:   10.0s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    9.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   16.8s remaining:   25.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   17.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.5s remaining:    2.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   12.9s remaining:   19.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   15.4s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    8.4s remaining:   12.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    9.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    6.9s remaining:   10.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    8.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   59.9s remaining:  1.5min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.0min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   10.0s remaining:   15.0s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   11.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  3.8min remaining:  5.7min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  4.8min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    6.0s remaining:    9.0s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    6.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    4.3s remaining:    6.5s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    5.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   12.5s remaining:   18.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   13.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    9.2s remaining:   13.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   10.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    4.5s remaining:    6.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    5.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.5s remaining:    3.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    5.4s remaining:    8.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    7.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.6min remaining:  3.8min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  3.0min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    5.8s remaining:    8.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    7.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    4.3s remaining:    6.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    5.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    7.1s remaining:   10.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    8.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.3s remaining:    3.5s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   12.8s remaining:   19.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   14.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    4.3s remaining:    6.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    6.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   16.8s remaining:   25.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   20.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.5s remaining:    2.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.4s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   10.6s remaining:   15.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   11.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    5.6s remaining:    8.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    7.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    3.2s remaining:    4.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    4.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    9.3s remaining:   13.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   11.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    5.5s remaining:    8.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    7.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.1s remaining:    3.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    5.1s remaining:    7.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    5.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    4.9s remaining:    7.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    7.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.5s remaining:    3.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    0.6s remaining:    1.0s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    1.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.6s remaining:    3.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   51.3s remaining:  1.3min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   56.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    9.0s remaining:   13.5s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   12.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    8.4s remaining:   12.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   12.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    9.3s remaining:   14.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   10.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    0.5s remaining:    0.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    0.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    3.4s remaining:    5.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    6.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.1s remaining:    3.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.7s remaining:    2.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.7s remaining:    4.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    3.7s remaining:    5.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    4.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.1s remaining:    3.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.6s remaining:    4.0s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    9.5s remaining:   14.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   10.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  3.0min remaining:  4.5min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  3.2min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    4.8s remaining:    7.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    6.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    7.1s remaining:   10.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    8.4s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.0s remaining:    1.5s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    1.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.0s remaining:    3.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    9.7s remaining:   14.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   11.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    4.2s remaining:    6.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    5.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    8.8s remaining:   13.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   12.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    6.2s remaining:    9.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    8.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    3.3s remaining:    4.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    4.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    9.0s remaining:   13.5s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   10.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.9s remaining:    4.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    4.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    5.5s remaining:    8.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    6.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    7.7s remaining:   11.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   10.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    4.1s remaining:    6.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    6.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    7.9s remaining:   11.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   11.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.3s remaining:    3.5s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    3.7s remaining:    5.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    4.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    1.6s remaining:    2.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    2.1s remaining:    3.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    3.0s remaining:    4.5s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.7s finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "BayesSearchCV(cv=5,\n",
       "              estimator=Pipeline(steps=[('scaler', MinMaxScaler()),\n",
       "                                        ('GBR', GradientBoostingRegressor())]),\n",
       "              n_iter=100, n_jobs=-1, random_state=10,\n",
       "              scoring='neg_root_mean_squared_error',\n",
       "              search_spaces={'GBR__criterion': ('friedman_mse', 'mse', 'mae'),\n",
       "                             'GBR__learning_rate': (0.001, 0.06),\n",
       "                             'GBR__max_depth': (3, 8),\n",
       "                             'GBR__n_estimators': (1000, 5000),\n",
       "                             'GBR__n_iter_no_change': [10],\n",
       "                             'GBR__random_state': [10],\n",
       "                             'GBR__subsample': (0.01, 0.8),\n",
       "                             'GBR__tol': [0.001]},\n",
       "              verbose=1)"
      ]
     },
     "execution_count": 1707,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# random grid search\n",
    "gbr_bayes_search.fit(X_reduced, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1708,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OrderedDict([('GBR__criterion', 'friedman_mse'),\n",
       "             ('GBR__learning_rate', 0.008862930368187014),\n",
       "             ('GBR__max_depth', 8),\n",
       "             ('GBR__n_estimators', 3038),\n",
       "             ('GBR__n_iter_no_change', 10),\n",
       "             ('GBR__random_state', 10),\n",
       "             ('GBR__subsample', 0.32556277999210126),\n",
       "             ('GBR__tol', 0.001)])"
      ]
     },
     "execution_count": 1708,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#get best parameters\n",
    "gbr_bayes_search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1790,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('scaler', MinMaxScaler()),\n",
       "                ('GBR',\n",
       "                 GradientBoostingRegressor(learning_rate=0.008862930368187014,\n",
       "                                           max_depth=8, n_estimators=3038,\n",
       "                                           n_iter_no_change=10, random_state=10,\n",
       "                                           subsample=0.32556277999210126,\n",
       "                                           tol=0.001))])"
      ]
     },
     "execution_count": 1790,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gbr_bayes_search.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1709,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-21990.305691805817"
      ]
     },
     "execution_count": 1709,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# cross-validation rmse score of the best model\n",
    "gbr_bayes_search.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1711,
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict on the training set\n",
    "y_hat = gbr_bayes_search.predict(X_reduced)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1712,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9821011092219659"
      ]
     },
     "execution_count": 1712,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# R squared on the training set\n",
    "from sklearn.metrics import r2_score\n",
    "r2_score(y_hat, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1710,
   "metadata": {},
   "outputs": [],
   "source": [
    "# predict for Kaggle submission\n",
    "pred_gbr = gbr_bayes_search.predict(X_test_reduced)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1713,
   "metadata": {},
   "outputs": [],
   "source": [
    "# export prediction to a csv file for Kaggle submission\n",
    "submission_gbr = pd.DataFrame()\n",
    "submission_gbr['Id'] = df_test['Id']\n",
    "submission_gbr['SalePrice'] = pred_gbr\n",
    "submission_gbr.to_csv('submission_gbr.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Search strategy 2 <a class=\"anchor\" id=\"19-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1736,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n"
     ]
    }
   ],
   "source": [
    "# BayesSearchCV using XGBRegressor on the full list of features\n",
    "from skopt import BayesSearchCV\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "\n",
    "estimator = Pipeline([\n",
    "        ('scaler', MinMaxScaler()),\n",
    "        ('XGB', XGBRegressor())\n",
    "    ])\n",
    "\n",
    "parameters_opt = {'XGB__random_state': [10],\n",
    "        'XGB__learning_rate': (0.001, 0.06),\n",
    "        'XGB__n_estimators': (1000, 5000),\n",
    "        'XGB__subsample': (0.1, 0.8),\n",
    "        'XGB__max_depth': (3, 8),\n",
    "        'XGB__booster': ('gbtree', 'gblinear'), \n",
    "        'XGB__eval_metric': ('rmse', 'mae'), \n",
    "        'XGB__seed': [10]}\n",
    "\n",
    "bayes_search_full = BayesSearchCV(estimator=estimator, search_spaces=parameters_opt, n_iter=80, n_jobs=-1, scoring= 'neg_root_mean_squared_error', cv = 5, verbose=1, random_state =10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1737,
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   19.8s remaining:   29.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   20.1s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.1min remaining:  3.2min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.2min finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   34.5s remaining:   51.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   34.8s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.0min remaining:  3.1min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.1min finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   34.2s remaining:   51.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   35.4s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.9min remaining:  2.9min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.9min finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   12.4s remaining:   18.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   12.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   17.1s remaining:   25.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   17.4s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    8.6s remaining:   13.0s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    8.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.6min remaining:  2.5min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.7min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.4min remaining:  2.1min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.4min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.7min remaining:  2.6min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.7min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.7min remaining:  4.0min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.7min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.6min remaining:  4.0min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.7min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   17.5s remaining:   26.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   18.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.3min remaining:  2.0min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.4min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.6min remaining:  2.4min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.6min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   14.1s remaining:   21.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   14.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   46.9s remaining:  1.2min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   47.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   22.3s remaining:   33.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   22.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   22.0s remaining:   33.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   22.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.6min remaining:  3.9min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.6min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.0min remaining:  3.0min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.1min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.5min remaining:  2.2min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.6min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   38.0s remaining:   57.1s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   39.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   40.6s remaining:  1.0min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   41.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.5min remaining:  2.3min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.5min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   53.3s remaining:  1.3min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   54.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   40.8s remaining:  1.0min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   40.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   55.3s remaining:  1.4min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   56.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   44.1s remaining:  1.1min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   44.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.1min remaining:  3.2min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.2min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   50.3s remaining:  1.3min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   50.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   54.6s remaining:  1.4min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   54.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   33.2s remaining:   49.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   33.4s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   32.4s remaining:   48.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   32.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   48.3s remaining:  1.2min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   49.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   33.7s remaining:   50.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   34.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   46.4s remaining:  1.2min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   48.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   53.0s remaining:  1.3min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   53.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   55.2s remaining:  1.4min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   55.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.3min remaining:  2.0min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.3min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   35.1s remaining:   52.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   35.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   37.9s remaining:   56.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   39.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   37.2s remaining:   55.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   37.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.1min remaining:  1.6min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.1min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  3.1min remaining:  4.6min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  3.2min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   36.6s remaining:   55.0s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   36.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.4min remaining:  3.6min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.5min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   18.8s remaining:   28.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   19.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.1min remaining:  1.7min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.1min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   47.1s remaining:  1.2min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   47.4s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.0min remaining:  1.5min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.0min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   28.5s remaining:   42.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   28.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.2min remaining:  1.7min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.2min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.6min remaining:  2.4min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.6min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   15.8s remaining:   23.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   15.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   52.0s remaining:  1.3min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   52.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.9min remaining:  2.8min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.9min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   35.9s remaining:   53.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   36.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   11.0s remaining:   16.6s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   11.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   22.4s remaining:   33.7s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   23.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.2min remaining:  1.8min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.2min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  1.9min remaining:  2.9min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.0min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   16.9s remaining:   25.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   17.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   45.6s remaining:  1.1min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   45.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   17.2s remaining:   25.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   17.3s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  3.0min remaining:  4.4min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  3.0min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   37.2s remaining:   55.8s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   37.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:  2.0min remaining:  3.0min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.0min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   58.5s remaining:  1.5min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   59.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   23.9s remaining:   35.9s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   24.1s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   16.8s remaining:   25.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   16.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   39.6s remaining:   59.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   39.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   56.1s remaining:  1.4min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   56.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   16.7s remaining:   25.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   16.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   44.6s remaining:  1.1min\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   46.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:    5.6s remaining:    8.4s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    5.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   28.8s remaining:   43.2s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   28.9s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1 candidates, totalling 5 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done   2 out of   5 | elapsed:   36.8s remaining:   55.3s\n",
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   37.0s finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "BayesSearchCV(cv=5,\n",
       "              estimator=Pipeline(steps=[('scaler', MinMaxScaler()),\n",
       "                                        ('XGB',\n",
       "                                         XGBRegressor(base_score=None,\n",
       "                                                      booster=None,\n",
       "                                                      colsample_bylevel=None,\n",
       "                                                      colsample_bynode=None,\n",
       "                                                      colsample_bytree=None,\n",
       "                                                      gamma=None, gpu_id=None,\n",
       "                                                      importance_type='gain',\n",
       "                                                      interaction_constraints=None,\n",
       "                                                      learning_rate=None,\n",
       "                                                      max_delta_step=None,\n",
       "                                                      max_depth=None,\n",
       "                                                      min_child_weight=None,\n",
       "                                                      missing=nan,...\n",
       "                                                      validate_parameters=None,\n",
       "                                                      verbosity=None))]),\n",
       "              n_iter=80, n_jobs=-1, random_state=10,\n",
       "              scoring='neg_root_mean_squared_error',\n",
       "              search_spaces={'XGB__booster': ('gbtree', 'gblinear'),\n",
       "                             'XGB__eval_metric': ('rmse', 'mae'),\n",
       "                             'XGB__learning_rate': (0.001, 0.06),\n",
       "                             'XGB__max_depth': (3, 8),\n",
       "                             'XGB__n_estimators': (1000, 5000),\n",
       "                             'XGB__random_state': [10], 'XGB__seed': [10],\n",
       "                             'XGB__subsample': (0.1, 0.8)},\n",
       "              verbose=1)"
      ]
     },
     "execution_count": 1737,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# fit model\n",
    "bayes_search_full.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1742,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "-21068.05488106399"
      ]
     },
     "execution_count": 1742,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#obtain best cross-validation RMSE \n",
    "bayes_search_full.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1791,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('scaler', MinMaxScaler()),\n",
       "                ('XGB',\n",
       "                 XGBRegressor(base_score=0.5, booster='gbtree',\n",
       "                              colsample_bylevel=1, colsample_bynode=1,\n",
       "                              colsample_bytree=1, eval_metric='mae', gamma=0,\n",
       "                              gpu_id=-1, importance_type='gain',\n",
       "                              interaction_constraints='',\n",
       "                              learning_rate=0.009355177335003302,\n",
       "                              max_delta_step=0, max_depth=4, min_child_weight=1,\n",
       "                              missing=nan, monotone_constraints='()',\n",
       "                              n_estimators=4749, n_jobs=8, num_parallel_tree=1,\n",
       "                              random_state=10, reg_alpha=0, reg_lambda=1,\n",
       "                              scale_pos_weight=1, seed=10,\n",
       "                              subsample=0.27990084664978787,\n",
       "                              tree_method='exact', validate_parameters=1,\n",
       "                              verbosity=None))])"
      ]
     },
     "execution_count": 1791,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#obtain all parameters of the best estimator\n",
    "bayes_search_full.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1741,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9965137238172347"
      ]
     },
     "execution_count": 1741,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#calculate R squared on the training set\n",
    "y_hat_full = bayes_search_full.predict(X)\n",
    "r2_score(y_hat_full, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1738,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n"
     ]
    }
   ],
   "source": [
    "#predict\n",
    "pred_full = bayes_search_full.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1739,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " C:\\Users\\uallakulov\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel\\ipkernel.py:287: DeprecationWarning:`should_run_async` will not call `transform_cell` automatically in the future. Please pass the result to `transformed_cell` argument and any exception that happen during thetransform in `preprocessing_exc_tuple` in IPython 7.17 and above.\n"
     ]
    }
   ],
   "source": [
    "#submission for Kaggle\n",
    "submission_full = pd.DataFrame()\n",
    "submission_full['Id'] = df_test['Id']\n",
    "submission_full['SalePrice'] = pred_full\n",
    "submission_full.to_csv('submission_full.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluation of feature importances <a class=\"anchor\" id=\"20-bullet\"></a>\n",
    "\n",
    "Finally, I evaluate feature importances by assigning scores to input features in the above model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1763,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#get feature importances\n",
    "importances = gbr_bayes_search.best_estimator_._final_estimator.feature_importances_\n",
    "#reverse the list - greatest to least important\n",
    "sorted_index = np.argsort(importances)[::-1]\n",
    "x = range(0,20)\n",
    "\n",
    "#tick labels\n",
    "labels = np.array(X_reduced.columns)[sorted_index]\n",
    "\n",
    "#plot\n",
    "plt.figure(figsize=(8,4))\n",
    "plt.bar(x, importances[sorted_index][0:20], tick_label = labels[0:20])\n",
    "plt.xticks(rotation = 90)\n",
    "\n",
    "axes = plt.gca()\n",
    "axes.set_ylim([0,0.3])\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1764,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#get feature importances\n",
    "importances = bayes_search_full.best_estimator_._final_estimator.feature_importances_\n",
    "#reverse the list - greatest to least important\n",
    "sorted_index = np.argsort(importances)[::-1]\n",
    "x = range(0,20)\n",
    "\n",
    "#tick labels\n",
    "labels = np.array(X.columns)[sorted_index]\n",
    "\n",
    "#plot\n",
    "plt.figure(figsize=(8,4))\n",
    "plt.bar(x, importances[sorted_index][0:20], tick_label = labels[0:20])\n",
    "plt.xticks(rotation = 90)\n",
    "\n",
    "axes = plt.gca()\n",
    "axes.set_ylim([0,0.3])\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*End of notebook*"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}