04-machine-learning-pipeline-model-training.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Machine Learning Pipeline - Model Training\n",
    "\n",
    "In this notebook, we pick up the transformed datasets and the selected variables that we saved in the previous notebooks."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Reproducibility: Setting the seed\n",
    "\n",
    "With the aim to ensure reproducibility between runs of the same notebook, but also between the research and production environment, for each step that includes some element of randomness, it is extremely important that we **set the seed**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# to handle datasets\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# for plotting\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# to save the model\n",
    "import joblib\n",
    "\n",
    "# to build the model\n",
    "from sklearn.linear_model import Lasso\n",
    "\n",
    "# to evaluate the model\n",
    "from sklearn.metrics import mean_squared_error, r2_score\n",
    "\n",
    "# to visualise al the columns in the dataframe\n",
    "pd.pandas.set_option('display.max_columns', None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>LotFrontage</th>\n",
       "      <th>LotArea</th>\n",
       "      <th>Street</th>\n",
       "      <th>Alley</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",
       "      <th>OverallQual</th>\n",
       "      <th>OverallCond</th>\n",
       "      <th>YearBuilt</th>\n",
       "      <th>YearRemodAdd</th>\n",
       "      <th>RoofStyle</th>\n",
       "      <th>RoofMatl</th>\n",
       "      <th>Exterior1st</th>\n",
       "      <th>Exterior2nd</th>\n",
       "      <th>MasVnrType</th>\n",
       "      <th>MasVnrArea</th>\n",
       "      <th>ExterQual</th>\n",
       "      <th>ExterCond</th>\n",
       "      <th>Foundation</th>\n",
       "      <th>BsmtQual</th>\n",
       "      <th>BsmtCond</th>\n",
       "      <th>BsmtExposure</th>\n",
       "      <th>BsmtFinType1</th>\n",
       "      <th>BsmtFinSF1</th>\n",
       "      <th>BsmtFinType2</th>\n",
       "      <th>BsmtFinSF2</th>\n",
       "      <th>BsmtUnfSF</th>\n",
       "      <th>TotalBsmtSF</th>\n",
       "      <th>Heating</th>\n",
       "      <th>HeatingQC</th>\n",
       "      <th>CentralAir</th>\n",
       "      <th>Electrical</th>\n",
       "      <th>1stFlrSF</th>\n",
       "      <th>2ndFlrSF</th>\n",
       "      <th>LowQualFinSF</th>\n",
       "      <th>GrLivArea</th>\n",
       "      <th>BsmtFullBath</th>\n",
       "      <th>BsmtHalfBath</th>\n",
       "      <th>FullBath</th>\n",
       "      <th>HalfBath</th>\n",
       "      <th>BedroomAbvGr</th>\n",
       "      <th>KitchenAbvGr</th>\n",
       "      <th>KitchenQual</th>\n",
       "      <th>TotRmsAbvGrd</th>\n",
       "      <th>Functional</th>\n",
       "      <th>Fireplaces</th>\n",
       "      <th>FireplaceQu</th>\n",
       "      <th>GarageType</th>\n",
       "      <th>GarageYrBlt</th>\n",
       "      <th>GarageFinish</th>\n",
       "      <th>GarageCars</th>\n",
       "      <th>GarageArea</th>\n",
       "      <th>GarageQual</th>\n",
       "      <th>GarageCond</th>\n",
       "      <th>PavedDrive</th>\n",
       "      <th>WoodDeckSF</th>\n",
       "      <th>OpenPorchSF</th>\n",
       "      <th>EnclosedPorch</th>\n",
       "      <th>3SsnPorch</th>\n",
       "      <th>ScreenPorch</th>\n",
       "      <th>PoolArea</th>\n",
       "      <th>PoolQC</th>\n",
       "      <th>Fence</th>\n",
       "      <th>MiscFeature</th>\n",
       "      <th>MiscVal</th>\n",
       "      <th>MoSold</th>\n",
       "      <th>SaleType</th>\n",
       "      <th>SaleCondition</th>\n",
       "      <th>LotFrontage_na</th>\n",
       "      <th>MasVnrArea_na</th>\n",
       "      <th>GarageYrBlt_na</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.461171</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.863636</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.777778</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.014706</td>\n",
       "      <td>0.049180</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.002835</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.673479</td>\n",
       "      <td>0.239935</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.559760</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.523250</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.375</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.416667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.018692</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.430183</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.116686</td>\n",
       "      <td>0.032907</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.545455</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.456066</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.363636</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.360294</td>\n",
       "      <td>0.049180</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.03375</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.8</td>\n",
       "      <td>0.142807</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.114724</td>\n",
       "      <td>0.172340</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.434539</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.406196</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.375</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.457944</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.25</td>\n",
       "      <td>0.220028</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.636364</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.916667</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.394699</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.954545</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.888889</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.036765</td>\n",
       "      <td>0.098361</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.25750</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.080794</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.601951</td>\n",
       "      <td>0.286743</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.627205</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.586296</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.8</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.046729</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.406206</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.228705</td>\n",
       "      <td>0.149909</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.090909</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.445002</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.454545</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.066176</td>\n",
       "      <td>0.163934</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.255670</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.018114</td>\n",
       "      <td>0.242553</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.566920</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.529943</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.375</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.084112</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.362482</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.469078</td>\n",
       "      <td>0.045704</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.636364</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.577658</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.363636</td>\n",
       "      <td>0.4</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.555556</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.323529</td>\n",
       "      <td>0.737705</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.17000</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.6</td>\n",
       "      <td>0.086818</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.434278</td>\n",
       "      <td>0.233224</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.75</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.549026</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.513216</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.375</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.416667</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.8</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.411215</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.406206</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.545455</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.75</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   MSSubClass  MSZoning  LotFrontage  LotArea  Street  Alley  LotShape  \\\n",
       "0    0.750000      0.75     0.461171      0.0     1.0    1.0  0.333333   \n",
       "1    0.750000      0.75     0.456066      0.0     1.0    1.0  0.333333   \n",
       "2    0.916667      0.75     0.394699      0.0     1.0    1.0  0.000000   \n",
       "3    0.750000      0.75     0.445002      0.0     1.0    1.0  0.666667   \n",
       "4    0.750000      0.75     0.577658      0.0     1.0    1.0  0.333333   \n",
       "\n",
       "   LandContour  Utilities  LotConfig  LandSlope  Neighborhood  Condition1  \\\n",
       "0     1.000000        1.0        0.0        0.0      0.863636         0.4   \n",
       "1     0.333333        1.0        0.0        0.0      0.363636         0.4   \n",
       "2     0.333333        1.0        0.0        0.0      0.954545         0.4   \n",
       "3     0.666667        1.0        0.0        0.0      0.454545         0.4   \n",
       "4     0.333333        1.0        0.0        0.0      0.363636         0.4   \n",
       "\n",
       "   Condition2  BldgType  HouseStyle  OverallQual  OverallCond  YearBuilt  \\\n",
       "0         1.0      0.75         0.6     0.777778         0.50   0.014706   \n",
       "1         1.0      0.75         0.6     0.444444         0.75   0.360294   \n",
       "2         1.0      1.00         0.6     0.888889         0.50   0.036765   \n",
       "3         1.0      0.75         0.6     0.666667         0.50   0.066176   \n",
       "4         1.0      0.75         0.6     0.555556         0.50   0.323529   \n",
       "\n",
       "   YearRemodAdd  RoofStyle  RoofMatl  Exterior1st  Exterior2nd  MasVnrType  \\\n",
       "0      0.049180        0.0       0.0          1.0          1.0    0.333333   \n",
       "1      0.049180        0.0       0.0          0.6          0.6    0.666667   \n",
       "2      0.098361        1.0       0.0          0.3          0.2    0.666667   \n",
       "3      0.163934        0.0       0.0          1.0          1.0    0.333333   \n",
       "4      0.737705        0.0       0.0          0.6          0.7    0.666667   \n",
       "\n",
       "   MasVnrArea  ExterQual  ExterCond  Foundation  BsmtQual  BsmtCond  \\\n",
       "0     0.00000   0.666667        0.5         1.0  0.666667  0.666667   \n",
       "1     0.03375   0.666667        0.5         0.5  0.333333  0.666667   \n",
       "2     0.25750   1.000000        0.5         1.0  1.000000  0.666667   \n",
       "3     0.00000   0.666667        0.5         1.0  0.666667  0.666667   \n",
       "4     0.17000   0.333333        0.5         0.5  0.333333  0.666667   \n",
       "\n",
       "   BsmtExposure  BsmtFinType1  BsmtFinSF1  BsmtFinType2  BsmtFinSF2  \\\n",
       "0      0.666667           1.0    0.002835           0.0         0.0   \n",
       "1      0.000000           0.8    0.142807           0.0         0.0   \n",
       "2      0.000000           1.0    0.080794           0.0         0.0   \n",
       "3      1.000000           1.0    0.255670           0.0         0.0   \n",
       "4      0.000000           0.6    0.086818           0.0         0.0   \n",
       "\n",
       "   BsmtUnfSF  TotalBsmtSF  Heating  HeatingQC  CentralAir  Electrical  \\\n",
       "0   0.673479     0.239935      1.0       1.00         1.0         1.0   \n",
       "1   0.114724     0.172340      1.0       1.00         1.0         1.0   \n",
       "2   0.601951     0.286743      1.0       1.00         1.0         1.0   \n",
       "3   0.018114     0.242553      1.0       1.00         1.0         1.0   \n",
       "4   0.434278     0.233224      1.0       0.75         1.0         1.0   \n",
       "\n",
       "   1stFlrSF  2ndFlrSF  LowQualFinSF  GrLivArea  BsmtFullBath  BsmtHalfBath  \\\n",
       "0  0.559760       0.0           0.0   0.523250      0.000000           0.0   \n",
       "1  0.434539       0.0           0.0   0.406196      0.333333           0.0   \n",
       "2  0.627205       0.0           0.0   0.586296      0.333333           0.0   \n",
       "3  0.566920       0.0           0.0   0.529943      0.333333           0.0   \n",
       "4  0.549026       0.0           0.0   0.513216      0.000000           0.0   \n",
       "\n",
       "   FullBath  HalfBath  BedroomAbvGr  KitchenAbvGr  KitchenQual  TotRmsAbvGrd  \\\n",
       "0  0.666667       0.0         0.375      0.333333     0.666667      0.416667   \n",
       "1  0.333333       0.5         0.375      0.333333     0.666667      0.250000   \n",
       "2  0.666667       0.0         0.250      0.333333     1.000000      0.333333   \n",
       "3  0.666667       0.0         0.375      0.333333     0.666667      0.250000   \n",
       "4  0.666667       0.0         0.375      0.333333     0.333333      0.416667   \n",
       "\n",
       "   Functional  Fireplaces  FireplaceQu  GarageType  GarageYrBlt  GarageFinish  \\\n",
       "0         1.0    0.000000          0.0        0.75     0.018692           1.0   \n",
       "1         1.0    0.000000          0.0        0.75     0.457944           0.5   \n",
       "2         1.0    0.333333          0.8        0.75     0.046729           0.5   \n",
       "3         1.0    0.333333          0.4        0.75     0.084112           0.5   \n",
       "4         1.0    0.333333          0.8        0.75     0.411215           0.5   \n",
       "\n",
       "   GarageCars  GarageArea  GarageQual  GarageCond  PavedDrive  WoodDeckSF  \\\n",
       "0        0.75    0.430183         0.5         0.5         1.0    0.116686   \n",
       "1        0.25    0.220028         0.5         0.5         1.0    0.000000   \n",
       "2        0.50    0.406206         0.5         0.5         1.0    0.228705   \n",
       "3        0.50    0.362482         0.5         0.5         1.0    0.469078   \n",
       "4        0.50    0.406206         0.5         0.5         1.0    0.000000   \n",
       "\n",
       "   OpenPorchSF  EnclosedPorch  3SsnPorch  ScreenPorch  PoolArea  PoolQC  \\\n",
       "0     0.032907            0.0        0.0          0.0       0.0     0.0   \n",
       "1     0.000000            0.0        0.0          0.0       0.0     0.0   \n",
       "2     0.149909            0.0        0.0          0.0       0.0     0.0   \n",
       "3     0.045704            0.0        0.0          0.0       0.0     0.0   \n",
       "4     0.000000            0.0        1.0          0.0       0.0     0.0   \n",
       "\n",
       "   Fence  MiscFeature  MiscVal    MoSold  SaleType  SaleCondition  \\\n",
       "0   0.00          1.0      0.0  0.545455  0.666667           0.75   \n",
       "1   0.75          1.0      0.0  0.636364  0.666667           0.75   \n",
       "2   0.00          1.0      0.0  0.090909  0.666667           0.75   \n",
       "3   0.00          1.0      0.0  0.636364  0.666667           0.75   \n",
       "4   0.00          1.0      0.0  0.545455  0.666667           0.75   \n",
       "\n",
       "   LotFrontage_na  MasVnrArea_na  GarageYrBlt_na  \n",
       "0             0.0            0.0             0.0  \n",
       "1             0.0            0.0             0.0  \n",
       "2             0.0            0.0             0.0  \n",
       "3             0.0            0.0             0.0  \n",
       "4             0.0            0.0             0.0  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load the train and test set with the engineered variables\n",
    "\n",
    "# we built and saved these datasets in a previous notebook.\n",
    "# If you haven't done so, go ahead and check the previous notebooks (step 2)\n",
    "# to find out how to create these datasets\n",
    "\n",
    "X_train = pd.read_csv('xtrain.csv')\n",
    "X_test = pd.read_csv('xtest.csv')\n",
    "\n",
    "X_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>12.211060</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11.887931</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>12.675764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12.278393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>12.103486</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   SalePrice\n",
       "0  12.211060\n",
       "1  11.887931\n",
       "2  12.675764\n",
       "3  12.278393\n",
       "4  12.103486"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load the target (remember that the target is log transformed)\n",
    "y_train = pd.read_csv('ytrain.csv')\n",
    "y_test = pd.read_csv('ytest.csv')\n",
    "\n",
    "y_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['MSSubClass',\n",
       " 'MSZoning',\n",
       " 'LotFrontage',\n",
       " 'LotShape',\n",
       " 'LandContour',\n",
       " 'LotConfig',\n",
       " 'Neighborhood',\n",
       " 'OverallQual',\n",
       " 'OverallCond',\n",
       " 'YearRemodAdd',\n",
       " 'RoofStyle',\n",
       " 'Exterior1st',\n",
       " 'ExterQual',\n",
       " 'Foundation',\n",
       " 'BsmtQual',\n",
       " 'BsmtExposure',\n",
       " 'BsmtFinType1',\n",
       " 'HeatingQC',\n",
       " 'CentralAir',\n",
       " '1stFlrSF',\n",
       " '2ndFlrSF',\n",
       " 'GrLivArea',\n",
       " 'BsmtFullBath',\n",
       " 'HalfBath',\n",
       " 'KitchenQual',\n",
       " 'TotRmsAbvGrd',\n",
       " 'Functional',\n",
       " 'Fireplaces',\n",
       " 'FireplaceQu',\n",
       " 'GarageFinish',\n",
       " 'GarageCars',\n",
       " 'GarageArea',\n",
       " 'PavedDrive',\n",
       " 'WoodDeckSF',\n",
       " 'ScreenPorch',\n",
       " 'SaleCondition']"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load the pre-selected features\n",
    "# ==============================\n",
    "\n",
    "# we selected the features in the previous notebook (step 3)\n",
    "\n",
    "# if you haven't done so, go ahead and visit the previous notebook\n",
    "# to find out how to select the features\n",
    "\n",
    "features = pd.read_csv('selected_features.csv')\n",
    "features = features['0'].to_list() \n",
    "\n",
    "# display final feature set\n",
    "features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# reduce the train and test set to the selected features\n",
    "\n",
    "X_train = X_train[features]\n",
    "X_test = X_test[features]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Regularised linear regression: Lasso\n",
    "\n",
    "Remember to set the seed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Lasso(alpha=0.001, random_state=0)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# set up the model\n",
    "# remember to set the random_state / seed\n",
    "\n",
    "lin_model = Lasso(alpha=0.001, random_state=0)\n",
    "\n",
    "# train the model\n",
    "\n",
    "lin_model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train mse: 781396538\n",
      "train rmse: 27953\n",
      "train r2: 0.8748530463468015\n",
      "\n",
      "test mse: 1060767982\n",
      "test rmse: 32569\n",
      "test r2: 0.8456417073258412\n",
      "\n",
      "Average house price:  163000\n"
     ]
    }
   ],
   "source": [
    "# evaluate the model:\n",
    "# ====================\n",
    "\n",
    "# remember that we log transformed the output (SalePrice)\n",
    "# in our feature engineering notebook (step 2).\n",
    "\n",
    "# In order to get the true performance of the Lasso\n",
    "# we need to transform both the target and the predictions\n",
    "# back to the original house prices values.\n",
    "\n",
    "# We will evaluate performance using the mean squared error and\n",
    "# the root of the mean squared error and r2\n",
    "\n",
    "# make predictions for train set\n",
    "pred = lin_model.predict(X_train)\n",
    "\n",
    "# determine mse, rmse and r2\n",
    "print('train mse: {}'.format(int(\n",
    "    mean_squared_error(np.exp(y_train), np.exp(pred)))))\n",
    "print('train rmse: {}'.format(int(\n",
    "    mean_squared_error(np.exp(y_train), np.exp(pred), squared=False))))\n",
    "print('train r2: {}'.format(\n",
    "    r2_score(np.exp(y_train), np.exp(pred))))\n",
    "print()\n",
    "\n",
    "# make predictions for test set\n",
    "pred = lin_model.predict(X_test)\n",
    "\n",
    "# determine mse, rmse and r2\n",
    "print('test mse: {}'.format(int(\n",
    "    mean_squared_error(np.exp(y_test), np.exp(pred)))))\n",
    "print('test rmse: {}'.format(int(\n",
    "    mean_squared_error(np.exp(y_test), np.exp(pred), squared=False))))\n",
    "print('test r2: {}'.format(\n",
    "    r2_score(np.exp(y_test), np.exp(pred))))\n",
    "print()\n",
    "\n",
    "print('Average house price: ', int(np.exp(y_train).median()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Evaluation of Lasso Predictions')"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY4AAAEWCAYAAABxMXBSAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAygUlEQVR4nO3deZhcdZ3v8fcnTQMdQBskMKQlBB2MskgCPcCIVwUusogYQFQGFZURvTPcGdDJI4ij4c5iNIp6ZWYUlQGviFGByKbAsIgyoiYkLJFEWYUOkiCJLGmlk3zvH+dUqK4+p+pUV1VXdffn9Tz1pOpXZ/nV6cr51m9XRGBmZlbUlHZnwMzMxhcHDjMzq4sDh5mZ1cWBw8zM6uLAYWZmdXHgMDOzujhw2JiQdJukv27RsT8u6eutOHaN854g6TFJz0maM9bn70Tlf2dJp0q6cZTH+aGk05qbO2sWBw4bRtIjkgbTm2HpcWG781Ui6U2SHi9Pi4h/jYiWBKUaPgecGRHbR8SyyjclhaQ/b0O+qkrz9Xz6tx2QdIGkrmafJyIui4g3F8jPfEnfqtj3mIi4tNl5subYqt0ZsI701oj4r3ZnYhzYA1jR7kyM0v4R8YCkVwO3Ab8GvlK+gaStImJjOzJnnc0lDitE0jaS1kvatyxtWlo62UXSjpKulbRW0rr0+ctzjjXsF6akmemv4K3S1++XdL+kZyU9JOlDafp2wA+B6WWloekZxzte0oo0v7dJek3Ze49I+gdJ90j6g6RFkrbNyecUSZ+Q9KikNZK+Keml6bV4DugC7pb0YJ3X8pWSbpH0e0lPSbpMUm/Z+x9LSwLPSlol6Yg0/SBJSyQ9I+lJSRcU+czVRMRK4CfAvmV/h9Ml/Ra4JT32B9K/xzpJN0jao+y8R0pamV7LCwGVvfc+ST8te72PpJskPZ3m/+OSjgY+Drwz/XvenW5bXuWV+XdI3yvl+TRJv02v53ll58y9ZjZ6DhxWSET8CbgSOKUs+R3AjyNiDcl36T9JfoXPAAaB0VZxrQGOA14CvB/4gqQDIuJ54BhgdVo9tH1ErC7fUdKrgMuBs4BpwPXANZK2rsj30cCewGuB9+Xk433p4zDgFcD2wIUR8aeI2D7dZv+IeGWdn0/Ap4HpwGuA3YH5af5nAWcCfxEROwBHAY+k+30J+FJEvAR4JfDdOj5zdkakvYH/AZRXtb0xzddRkuaS3NhPTI/9k/RcSNoZuAL4BLAz8CBwaM55dgD+C/hR+rn/HLg5In4E/CuwKP177p+x+/vI+DtUbPN6YBZwBPDJssCZec2sMQ4clmVx+su19Phgmv5thgeOv0rTiIjfR8QVEbEhIp4F/oXkBlS3iLguIh6MxI+BG0lubkW8E7guIm6KiCGSdoge4HVl2/zfiFgdEU8D1wCzc451KnBBRDwUEc8B5wLvKpWMRisiHkjz96eIWAtcwIvXahOwDbC3pO6IeCQiSiWaIeDPJe0cEc9FxJ11fOZKd0laR/L5v04S9EvmR8TzETEIfAj4dETcn1Zb/SswOy11HAv8KiK+n573i8Dvcs53HPC7iPh8RPwxIp6NiJ8Xu2KF/g7nR8RgRNwN3A2UAlDeNbMGOHBYlrkR0Vv2+FqafgvQI+ng9MYxG7gKQNJUSV9NqxOeAW4HejWKRldJx0i6M63SWE9yg9q54O7TgUdLLyJiM/AY0Fe2TfnNbQPJL9iax0qfbwXsWjAvmZRU7X0nrY56BvgW6eeLiAdISg7zgTXpdtPTXU8HXgWslPRLScdl5TPnM1c6ICJ2jIhXRsQn0n1KHit7vgfwpdKPCOBpkhJTX3reLdtGMmNq+b7ldicpkYxGkb9D3t8075pZAxw4rLD05vJdklLHXwHXpqULgI+SVBUcnFYLvCFN14gDwfPA1LLXf1Z6ImkbkuqPzwG7RkQvSdVL6Ti1pnNeTXKzKx1PJDetgRr71TwWSRXcRuDJURyr3KdJPsdr02v1bsquU0R8OyJen547gM+k6b+JiFOAXdK07ytp92nmZ4bh1/gx4EMVPyR6IuK/gSfS81SeN8tjJFVFtc6XZdR/hyrXzBrgwGH1+jZJ1cip6fOSHUjaNdZL2gn4VJVjLAfeIGlG2sh5btl7W5NU1awFNko6Bijv0vkk8LJS42iG7wJvkXSEpG6SgPYn4L8Lfr5ylwNnS9pT0va8WBdfT0+jrSVtW/boIrlWz5Fcqz5gXmljSbMkHZ4G0D+SXNNN6XvvljQtDeDr0102NfkzV/oKcK6kfdI8vFTSyel71wH7SDoxrTb6O8p+BFS4FvgzSWcp6Vywg6SD0/eeBGZKyrsfjfrvUOWaWQMcOCzLNRo+juOq0htpvfTzJNUHPyzb54sk9epPAXeSNIJmioibgEXAPcBSkptK6b1nSW5A3wXWkZRsri57fyXJjeShtPpketmhiYhVJL/gv5zm5a0k3YtfqPMaAFwM/D+SareHSW7k/7vOY6wgufmXHu8HzgcOAP5AcvO9smz7bYAFad5/R/JL+ePpe0cDK5T06PoS8K60vaCZn3mYiLiK5Jf6d9JqtftIOigQEU8BJ6f5/T2wF3BHznGeBY5M8/Y74Dckjd0A30v//b2kuzJ2b+TvkHnNCu5rOeSFnMzMrB4ucZiZWV0cOMzMrC4OHGZmVhcHDjMzq8ukmORw5513jpkzZ7Y7G2Zm48rSpUufiohplemTInDMnDmTJUuWtDsbZmbjiqRHs9JdVWVmZnVx4DAzs7o4cJiZWV0cOMzMrC4OHGZmVpdJ0avKzGyyWbxsgIU3rGL1+kGm9/Yw76hZzJ1TbYmW4hw4zMwmmMXLBjj3ynsZHEpmkB9YP8i5V94L0JTg4aoqM7MJZuENq7YEjZLBoU0svGFVU47vwGFmNsGsXj9YV3q9HDjMzCaY6b09daXXy4HDzGyCmXfULHq6u4al9XR3Me+oWU05vhvHzcwmmFIDuHtVmZlZYXPn9DUtUFRqWVWVpIslrZF0X1naP0m6R9JySTdKmp6z79GSVkl6QNI5Zek7SbpJ0m/Sf3dsVf7NzCxbK9s4LgGOrkhbGBGvjYjZwLXAJyt3ktQF/BtwDLA3cIqkvdO3zwFujoi9gJvT12ZmNoZaFjgi4nbg6Yq0Z8pebgdExq4HAQ9ExEMR8QLwHeBt6XtvAy5Nn18KzG1mns3MrLYxb+OQ9C/Ae4E/AIdlbNIHPFb2+nHg4PT5rhHxBEBEPCFplyrnOQM4A2DGjBlNyLmZmUEbuuNGxHkRsTtwGXBmxibK2m0U57koIvojon/atBErH5qZ2Si1cxzHt4GTMtIfB3Yve/1yYHX6/ElJuwGk/65paQ7NzGyEMQ0ckvYqe3k8sDJjs18Ce0naU9LWwLuAq9P3rgZOS5+fBvygVXk1M7NsLWvjkHQ58CZgZ0mPA58CjpU0C9gMPAp8ON12OvD1iDg2IjZKOhO4AegCLo6IFelhFwDflXQ68Fvg5Fbl38zMsimi7uaDcae/vz+WLFnS7myYmY0rkpZGRH9luueqMjOzujhwmJlZXRw4zMysLp7k0MzGVCvXwrax4cBhZmOm1Wth29hwVZWZjZlWr4VtY8MlDjMbM0XXwnZ1VmdzicPMxkyRtbBL1VkD6wcJXqzOWrxsYIxyabU4cJjZmCmyFrarszqfq6rMbMwUWQu7aHWWtY8Dh5mNqVprYU/v7WEgI0jkVXPZ2HNVlZl1lCLVWdZeLnGYWUcpUp1l7eXAYWYdp1Z1lrWXq6rMzKwuLnGYWUfyIMDO5cBhZh3Hc1p1NldVmVnH8SDAztaywCHpYklrJN1XlrZQ0kpJ90i6SlJvxn6zJC0vezwj6az0vfmSBsreO7ZV+Tez9vEgwM7WyhLHJcDRFWk3AftGxGuBXwPnVu4UEasiYnZEzAYOBDYAV5Vt8oXS+xFxfUtybmZtVWROK2uflgWOiLgdeLoi7caI2Ji+vBN4eY3DHAE8GBGPtiCLZtahPAiws7WzcfwDwKIa27wLuLwi7UxJ7wWWAB+NiHWtyJyZja3KXlQnHdjHrSvXuldVB1JEtO7g0kzg2ojYtyL9PKAfODFyMiBpa2A1sE9EPJmm7Qo8BQTwT8BuEfGBnP3PAM4AmDFjxoGPPupCi1mnquxFBUkJ49Mn7udg0UaSlkZEf2X6mPeqknQacBxwal7QSB0D3FUKGgAR8WREbIqIzcDXgIPydo6IiyKiPyL6p02b1qzsm1kLuBfV+DKmVVWSjgY+BrwxIjbU2PwUKqqpJO0WEU+kL08A7huxl5k1bKwH37kX1fjSyu64lwM/A2ZJelzS6cCFwA7ATWl32q+k206XdH3ZvlOBI4ErKw77WUn3SroHOAw4u1X5N5us2rECn3tRjS8tK3FExCkZyd/I2XY1cGzZ6w3AyzK2e0/TMmhmmapVG7Wq1DHvqFmZbRzuRdWZPOWImQ3TjmojT6U+vjhwmNkw7VqBz1Opjx+eq8rMhvHgO6vFJQ4zG8bVRlaLA4dZB+i0tSdcbWTVFKqqkvR6Se9Pn0+TtGdrs2U2ebSj+6tZI2oGDkmfIhm0V5rJthv4ViszZTaZTORR04uXDXDoglvY85zrOHTBLQ6GE0SRqqoTgDnAXZCMuZC0Q0tzZTaJTNRR017Fb+IqUlX1QjqnVABI2q61WTKbXCbqqOmJXJKa7IoEju9K+irQK+mDwH+RTDBoZk3QaPfXTq0OmqglKStQVRURn5N0JPAMMAv4ZETc1PKcmU0SjXR/7eTqoHYNJLTWqxk40h5UPykFC0k9kmZGxCOtzpzZZDHa7q/tmFeqKM8/NXEVqar6HrC57PWmNM3M2qyTq4Pmzunj0yfuR19vDwL6enu8MNMEUaRX1VYR8ULpRUS8kK7OZ2Zt1ju1m3Ubhkakd0p1kAcSTkxFShxrJR1feiHpbSTLt5pZGy1eNsBzf9w4Ir27S64OspYqUuL4MHCZpAsBAY8B721prsyspoU3rGJo88jVl7fbeqtR/8rvtKlPrDMV6VX1IHCIpO0BRcSzrc+WmdWS147xh8GRVVdFdHIPLessuYFD0rsj4luSPlKRDkBEXNDivJlZFc3u7trJPbSss1Rr4yiNEN8h52FmbdTsdTM6uYeWdZbcEkdEfFVSF/BMRHyh3gNLuhg4DlgTEfumaQuBtwIvAA8C74+I9Rn7PgI8S9L1d2NE9KfpOwGLgJnAI8A7ImJdvXkzmwiavW6GB+xZUUqmoaqygXRrRBxW94GlNwDPAd8sCxxvBm6JiI2SPgMQER/L2PcRoD8inqpI/yzwdEQskHQOsGPW/pX6+/tjyZIl9X4Es0mlso0DkhKMx15MXpKWln64lyvSq+q/0x5Vi4DnS4kRcVe1nSLidkkzK9JuLHt5J/D2Aucv9zbgTenzS4HbSKZ8N5twsno4QetW5vPKf1ZUoRJHRnJExOE1D54EjmtLJY6K964BFkXEiLU9JD0MrCOZkferEXFRmr4+InrLtlsXETvmnPsM4AyAGTNmHPjoo4/Wyq5Zx8j69d89RSAY2vTi/9nxXCJw19/O10iJ4+TKKqMmZOY8YCNwWc4mh6brfuwC3CRpZUTcXs850mBzESRVVQ1l2KyGRm6CWftm9XDKGrPR7F5PY3Uzd9ff8S23V5Wkt0paC9wj6XFJr2vGCSWdRtJofmrkFHciYnX67xrgKuCg9K0nJe2WHmc3YE0z8mTWiEaWfs3bN6uROk+zej2N5RK2XqtjfKvWHfdfgP8REdOBk4BPN3oySUeTtEkcHxEbcrbZrrTCYLpo1JuB+9K3rwZOS5+fBvyg0TyZNaqRm2Devl3peKkiSr2eGl2XYyxv5u76O75VCxwbI2IlQET8nDrHbki6HPgZMCstsZwOXJge5yZJyyV9Jd12uqTr0113BX4q6W7gF8B1EfGj9L0FwJGSfgMcmb42a6tGboJ522yKGDFGo3uK6O4aHlBK4zaaUVoYy5v5RF31cLKo1saxS8Wo8WGva40cj4hTMpK/kbPtauDY9PlDwP452/0eOKLaec1qaXY9fiPjH/L27Str6yjSq+rQBbc0POp7LMdxeK2O8a1a4Pgaw0sZla/Nxp1WNMo2chOstm/elORZac0oLYzlzdxdf8e3aiPHzx/LjJiNhVbMx9TITbBZN9BmlBbG+mbutTrGr5rjOCYCjxy3kj3PuY6sb7yAhxe8Zayz0zQe9W2tkDeOo8hCTmYTxkRtlPUyrTaWigwANJswJnKjrKt+bKzULHFI2lXSNyT9MH29d9q11mzc8S9zs8YVKXFcAvwncF76+tckEx5mdq0163T+ZW7WmCKBY+eI+K6kcwHSKdE31drJzBrjSQCtUxVpHH9e0stIZqpF0iHAH1qaK7NJLmsk+FmLljP7/BtbMneUWT2KlDg+QjJH1Csl3QFMo/51NMyM4qWIrPEmAOsHhzyLrLVdzcAREXdJeiMwi6S7+6qIGGp5zszGoWqBoZ5R69VGfDd7KnWzetUMHJJOBn4UESskfQI4QNI/11oB0GyiqwwSh716GlcsHcgNDPWMWs8bCV7iWWStnYq0cfxjRDwr6fXAUSRLtv5Ha7Nl1lqNTkGe1QZx2Z2/rToteT3zSc07ataI2XHLTe/tafgzmI1WkTaO0v+EtwD/ERE/kDS/dVkya61qVUZQbK6mrNJD3uQ9pcBQz3xSpXOef80K1m0YXjPc093FYa+e5hX0rG2KlDgGJH0VeAdwvaRtCu5n1pHyqozmX72i8JoW9VQVlQJDVimi2qj1uXP6WPbJN/PFd84eMWDx1pVrvYKetU2REsc7gKOBz0XE+nTJ1nmtzZZZ6+Td9NcPjuzzUX4zLi+J9E7tHlESyNI9RVsCw2hnn80asHj2ouWZ27rtw8ZCkZLDzsAS4E+SZgDdwMqW5sqsTLPr8uud0LB8HfBSSeS5P24csRpflu233WrYTX/unD7uOOdwvvDO2UASAEbzmSbqZI02PhQpcVxHUn0rYFtgT2AVsE8L82UGNHfhpVIvqIH1g4jhbRI93V2IYMPQ5sx9K6uFhjYne08RbK6yMsH6jFJJMz7TRJ6s0TpfzRJHROwXEa9N/90LOAj4aa39JF0saY2k+8rSFkpaKekeSVdJ6s3Yb3dJt0q6X9IKSX9f9t58SQPpeuXLJR1b+JPauFStC2s9yntBwYu/hODFdoNtqvRiyrM5oLtL9PZ0Z76fVQJoxmfyZI3WTnVPq54OCPyLApteAlwIfLMs7Sbg3HS+q88A5wIfq9hvI/DR9Dw7AEsl3RQRv0rf/0JEfK7efNv4VKsLayMjsYPkhnvHOYcD+e0GtQxtCqTkF3+REkAzlnkFT9Zo7VNkWvWPlD3+QdK3gbW19ouI24GnK9JujIiN6cs7gZdn7PdEaXBhRDwL3A/4f8ckVa0uP2ssRb29oMrTq7UP1GrNWL9hqHAJwO0TNt4VaRzfoeyxDUmbx9uacO4PAD+stoGkmcAc4OdlyWemVV0XS9qxyr5nSFoiacnatTXjnHWoal1Y66nyKXKznnfUrMwG7+4p4tRDZtCl/PAxvbeHuXP6mHfULKb39rB6/SALb1iVGcTq7ZZr1mmKtHGcHxHnAxcAX4qIyyLij42cVNJ5JFVSl1XZZnvgCuCsiHgmTf4P4JXAbOAJ4PNV8n1RRPRHRP+0adMaya61UbW6/EZHYlferOfO6WPh2/dnx6kvtlf09nSz8OT9+ee5+3HIK7J/pyg9ftESkNsnbLwrMlfVvsD/A3ZKXz8FnBYR91XdMf94pwHHAUdERGZ/FEndJEHjsoi4spQeEU+WbfM14NrR5MHGl7y6/NGMxK7VHlKt3eDOh9ZlZ1DJfocuuKXwXFRun7DxrEjj+EXARyLiVgBJb0rTXlfvySQdTdIY/saI2JCzjUhWF7w/Ii6oeG+3iHgifXkCMKrgZRNDkS6pi5cNDJu2o7enmy+8czZz5/RtGR9SHkggP7hsyv6dQym5WY3eZp2uSODYrhQ0ACLiNknb1dpJ0uXAm4CdJT0OfIqkF9U2wE1JfODOiPiwpOnA1yPiWOBQ4D3AvZKWp4f7eERcD3xW0mySDjGPAB8q8iFtYqpVili8bIB537+boU0v3vDXDw5x1qLlnFXRg2pg/SAfWbScri5t2b5yfEWXlBk8Sm0f9ZSAzMazIoHjIUn/SFJdBfBu4OFaO0XEKRnJmeuUR8Rq4Nj0+U/J6cQSEe8pkF+bRCqDR6lhvDSNeXnQqGUzsLli+/KqplMO3p1v3fnbEfudcvDugAfl2eRRJHB8ADgfuJLkhn478P5WZsosTz1rYFRbz6Iepaqmf567HwCX//wxNkXQJXHKwbtvSR/tXFRm441y2qcnlP7+/liyZEm7s2ENqpyqAxgxdUhJT/cU/ji0OXeq83qUDxI0m0wkLY2I/sr03BKHpGvIX2KAiDi+SXkzK6SeNTAGc+acqld3l1zVZFahWlVVaVoPAV8D/rr12THL15beSRO/QG5Wt9zAERE/Lj2X9Fz5a7N2qLUOdysMbY7McRhmk1nRlfz8u8tapuh6G3mjv7fbOntW29qrZRTjcRhmw1Vr49ip7GVXOi/Ulv+LEfH0yL3M6lPP2hR5vZaAzG6wJx3Yt6UHVCNemjNlutlkldurStLDDF+2oFxExCtambFmcq+qznXoglsyq5/q7cmUNUJ8/vHJWmOVQaVeU5SMDnf3Wpts8npV5VZVRcSeEfGK9N/Kx7gJGtbZmjlNxx/LelKtHxzi7EXLWfLo08MmFByNzUHNadvNJpOibRxmLdGstSnyuupelo70vuOcw3l4wVvoq3LcatOml4xm9UGzicaBw9pq3lGz6J4y/IbdPaX+sRN5JZSALTf6xcsG2PDCxhHb9HR38cV3zubz79h/RON7PecymyzqXjrWrOkqf+hn/PCvtURsta66q9cPZo46hxfbQsqPVTrPlJxJDT1poU12uSUOSTtVe4xlJm3iypqIcGhTDKsOKrJA0ryjZuW2YUzv7cmsygLYbputhgWNuXP6tlRrZZVAPGmhWfWqqqXAkvTftcCvgd+kz5e2Pms2GRRpHC+6ROzUjPEcpRv9aBrhvVKfWbZqI8f3BJD0FeDqdD0MJB0D/M+xyZ5NdEXWsKh108+rhgLYtntK4fNk8Up9ZiMVaRz/i1LQAIiIHwJvbF2WbKLKGiGeNRpcwGGvfnGd+Fo9r/KqoQDWbRji3Cvv5bBXT3O1k1mTFAkcT0n6hKSZkvaQdB7w+1ZnzCaWvHYKgJMO7BvWPhHAFUsHtrRh5E01Urrp1+rlNDi0ict//hiDQ5u2dLl1tZPZ6BXpVXUKybKvV5H8n749TTMrrFY7RWXfpcGhTXz0u3cDtRdIKjL5Yal31KaILUHHQcNsdAov5CRp+4h4rvCBpYuB44A1EbFvmrYQeCvwAvAg8P6IWJ+x79HAl4AukrXIF6TpOwGLgJkka46/IyLW1cqLpxxpvz3PuS5zpsxSSSPvW9jT3TWsZJDVLRfqn1bEizOZ1Vb3lCNlO75O0q+AX6Wv95f07wXOeQlwdEXaTcC+EfFakl5a52acrwv4N+AYYG/gFEl7p2+fA9wcEXsBN6evrcMtXjbAlJxR2dN7e6o2UJeXSqpVd5V6PxXlQXxmo1ekjeMLwFGk7RoRcTfwhlo7RcTtwNMVaTdGRGno7p3AyzN2PQh4ICIeiogXgO8Ab0vfextwafr8UmBugfxbmyxeNsDs82/krEXLMwfSlaqMyhvCs5SqoapVd5XGXxQNHh7EZzZ6hUaOR8RjGv6LcfRTjb7oAyTVTpX6gMfKXj8OHJw+3zUinkjz9ISkXZqQD2uBal1kIZkXqlQFVWvup1KDdl4pYWD9IIcuuIXVaUmkFvemMmtMkcDxmKTXASFpa+DvgPsbOWnaM2sjcFnW2xlpdS+oIOkM4AyAGTNm1Lu7FZQ3FUi1LrIAmyO2tFvUqjYqlVbyGsEFhVcG7PPU6GYNKxI4PkzSUN1H8uv/RuBvRntCSaeRNJofEdkt848Du5e9fjmwOn3+pKTd0tLGbsCavPNExEXARZA0jo82v5av2iJMtYLBFIk9z7mO6b09vLSnm/WDQ7nblqqf5h01K7MUU+uPK+DUQ2bwz3P3q7GlmRVRJHDMiohTyxMkHQrcUe/J0t5SHwPeGBEbcjb7JbCXpD2BAeBdwF+l710NnAYsSP/9Qb15sOap1uZQq4tsqRQxsH6Q7i7RPUUMbc5vB4GR3XJ7p3ZvWbgpi2BYz6tSdZYXZDJrTJHG8S8XTBtG0uXAz4BZkh6XdDpwIbADcJOk5el0JkiaLul6gLTx/EzgBpIqse9GxIr0sAuAIyX9BjgyfW1tUq3NIWvQHiSr6VUa2hRsv+1WW0oW1QbplU9COHXr/N89fb09PLzgLVu63NaaJBGKr31uNtlVW3P8L4HXAdMkfaTsrZeQjK+oKiKyBgl+I2fb1cCxZa+vB67P2O73wBG1zj0Z1Zp2vBWqtTlA0kW2Mk9nL1qeeaz1G4ZY9sk313X+atVh5Y3ftXpjQX1rn5tNdtWqqrYGtk+32aEs/Rng7a3MlNVnLG965QHqpT3dmdsEcNai5ZkN0QtvWDWqyQaz5AWuHad2DztnozPwOnCYDVdtzfEfR8T5wCERcX7Z44KI+M0Y5tFqKDrteKMqB+BVa9CG/HUzmjXZYN6xPvXWfYalFVmetplrn5tNdEXaOL4uqbf0QtKOkm5oXZasXmN106vVxTZLac6pUvBo5hoXRY9VJFg1a+1zs8mgSK+qncvnk4qIdR5411lGu9ZEvUYbiDZFDKs6y1vjYjTtNEXWy6g1SSJkd/X1QEGzbEUCx2ZJMyLitwCS9mAUA/KsdcbqpldkFto8WbPdlmt1O02tAFMkuJhZokjgOA/4qaQfp6/fQDoi2zpDvTe90fbAyhuAV1RlyaNcJzROe7U/s2JqBo6I+JGkA4BDSHpanh0RT7U8Z1aXoje9Rn7ZVwaoKVLm5IXV5AWDvJLMwPpBFi8b8A3drIPkNo5LenX67wHADJJpPwaAGWmajUON9sAqH4BXb9AoqWwrWbxsIHOCspKswXpm1j7VShwfBT4IfD7jvQC8Cs441KweWKWb/WhCR2Wj/cIbVlU9jsdTmHWW3MARER9M/z1s7LJjrTaaHlhZbSK1bvZ5shrtiwSt0tTpbrA2a79qU46cWG3HiLiy+dmxVstq4BbkLqaU1yYymgby8jU4yhXtreVpQMw6Q7UBgG9NH6eTzDF1avr4OvDu1mfNWmHunD5OOrBvWJtCAFcsHchsR8hrE8lTGoyXJa9NJG9CxCytGBFvZvWpVlX1fgBJ1wJ7l1beS9fB+LexyZ61wq0r146oZsprR6in7UOwpRorrwQx73t3c/41K1i/YWhEV+DyObAkcqdM9zQgZu1VZBzHzFLQSD0JvKpF+bExUE8DeT2D/gI4e9FyXtrTTXeXGNo0soQxtDm2BITKqqfKoHXoglvGZES8mdWnyFxVt0m6QdL70tX7rgNubXG+rIXqmZcpqxqpWtfZLZMfFmw5r1b11MwJEc2seWoGjog4E/gKsD8wG7goIv53i/NlLTTvqFl0V6yo1D1FI27Ipd5Ug0Obhi2udOohM2q2SQxtji371JJXAmrmhIhm1jxFqqoA7gKejYj/kjRV0g4R8WwrM2YtVnlPr3hd2ZtqU8SWX/tz5/TRv8dOW9ok8goXmyJyl4QtV63qydOAmHWemiUOSR8Evg98NU3qAxa3ME9W0GiXOl14w6oR7Q9Dm2JYlVGtEeblI8j7cm78vT3d1eu1cNWT2XhUpI3jb4FDSVb+I13EydOqt1nlokoD6wc5a9Fy5vyfG6sGkMXLBnIbu8urjOppQM9ri5DIbCDvklz1ZDaOFamq+lNEvKC0vlrSVhRo+pR0MXAcsCYi9k3TTgbmA68BDoqIJRn7zQIWlSW9AvhkRHxR0nySaVDWpu99PF2ffNLJW1Rp3YYh5n3/buZfvYI/DA7v8rp42QDzvn937jGnSOx5znVM7+2hd2p3ZnfYrGqlvNl589YX3xzBwwveUvCTmlmnKRI4fizp40CPpCOBvwGuKbDfJcCFwDfL0u4DTuTFaq8RImIVSSM8krpIJla8qmyTL0TE5wqcf0KrNpZhaFNsWda1vMtrVhVVudIAvYH1g3RP0YgutdWqlbLaIpq5vriZdY4iVVUfI/mFfy/wIeB64BO1doqI24GnK9LuTwNDUUcAD0bEo3XsMynUc/MttU3UM3BuaHOw3dZb5fZoKtK+4u60ZhNT1RKHpCnAPWlV09fGJkvDvAu4vCLtTEnvBZYAH42IdVk7SjqDdMGpGTNmtDST7VDvokqlKqR6VvD7w+AQyz/15hHp1db0gOFVVicd2MetK9c2vKreaBefMrPmU9RYU0HSZcC5paVj6zq4NBO4ttTGUZZ+G/APWW0cZdtsTbIGyD4R8WSativwFEkbyz8Bu0XEB2rlo7+/P5YsyT3VuLV42QDzr16xpVqqmr70Zjvv+3dXra6q3OeOc16cPb90884LPjtO7eaPQ5tHLGHbaAN4ZaBq1nHNrDpJSyOivzK9SFXVbsAKSTdLurr0aH4WRzgGuKsUNAAi4smI2BQRm0lKQAeNQT461tw5fSz/1Jv54jtnb6lS2nFq94jBfeXjLxa+fX92nNq95b3enm7efcgMursqBgR2DR8QWN6LK8+6DUMNLRKVp9HFp8ysuYo0jp/f8lxkO4WKaipJu5XNm3UCSWP7pFfZMF2tWierEXvxsgEW/eKx4QetKJTk9eIqotFJCZu1+JSZNUe19Ti2BT4M/DlJw/g3ImJj0QNLuhx4E7CzpMeBT5E0ln8ZmAZcJ2l5RBwlaTrw9Yg4Nt13KnAkSWN8uc9Kmk1yW3sk432j/tHWC29YNWJ099DmYP7VK2qODi+i0V5Uo1l8ysxap1qJ41JgCPgJSbXR3sDfFz1wRJyS89ZVlQkRsRo4tuz1BuBlGdu9p+j5J5tGGo/zfrmvHxwq1H5STdYcWPXK6gjg3llm7VMtcOwdEfsBSPoG8IuxyZLVq1ovpyLBo97eVvXYftutGm7Azhtg6IZxs/aoFji2/NSMiI0qONOpjb1qjcdFbq7zjprFvO/dXXMywtFYn7MYU7082aFZ56jWq2p/Sc+kj2eB15aeS3pmrDJotTXaeDx3Th/bb1tsouTy3w+lzlt9vT3DemqVczuE2cRTbenYYotAW9vV03ic1xZStGRQPuxnm61eHEuRN9bC7RBmE0+RcRzW4YpO7ZE1o+65V97L4mUD9OaUGKqpnGbdiy6ZTQ5FF3KyDla08TivLeT8a1bw3B8L97Qeprw6zO0QZpODA8cEUeSmndfmkTV9erkdp3YTQWbXXLdhmE0+rqqaREZ7k5+69VbMP34fz3RrZoADx6SS1xbS21O9fWP1+kG3YZjZFq6qmkTmzuljyaNPc/nPH2NTBF0SJx3YR/8eO1Wdor1UUnEbhpmBSxzjUpFFlPL2u2LpwJaV/jZFcMXSZN9Pn7hf5lgMV0eZWSWXOFqs2QsQNTK9SLUR5necc/iW8Rie2sPMqnHgaKFG55DK0sj0IkVGmLs6ysxqcVVVC7ViAaJGphfJ61XlLrVmVg8HjhZqxQJEjdz8i44wNzOrxoGjAbUaqVvxC7+Rm7+71JpZM7iNY5SKtF+0agGibbaasuWYU7unsE33FM5etJyFN6yq2ZjtNgwza5QDxygVaaRu9gJEWTPQbhjazIahzUB28HIvKTNrtpYFDkkXA8cBayJi3zTtZGA+8BrgoIhYkrPvI8CzwCZgY0T0p+k7AYuAmSRrjr8jIta16jNUU7T9opm/8LOCVaXy4NWKXl1mZq1s47gEOLoi7T7gROD2AvsfFhGzS0EjdQ5wc0TsBdycvm6LdvRQKtqoXtquFb26zMxaFjgi4nbg6Yq0+yOikbvW24BL0+eXAnMbOFZD2tFDqWhQKm3Xil5dZmad2qsqgBslLZV0Rln6rhHxBED67y55B5B0hqQlkpasXbu26RlsRw+lrGCV5fk/bWTxsgGP2zCzlujUxvFDI2K1pF2AmyStTEswhUXERcBFAP39/VFj81EZ6x5KWY3th716Gtfd88SwNTXWDw5x7pX3ctKBfVyxdMDLuZpZU3Vk4IiI1em/ayRdBRxE0i7ypKTdIuIJSbsBa9qZz3bICla3rlw7YjGmwaFN3LpyLZ8+cT/3qjKzpuq4wCFpO2BKRDybPn8z8H/St68GTgMWpP/+oFX5GG031nZ0f63WluFxG2bWbC1r45B0OfAzYJakxyWdLukESY8DfwlcJ+mGdNvpkq5Pd90V+Kmku4FfANdFxI/S9xYAR0r6DXBk+rrpSt1YB9YPErzYjbXW9OWj3a9Rbssws7GkiJZU/3eU/v7+WLIkc8hIpkMX3MJAxq/4vt4e7jjn8Kbv16isgYE93V3DGus9ENDM6iVpacWQCKADq6o6QdFurJU346ygUe14zVJrhLoHAppZMzlwZMgLAuVVP1k3Y5H0I662X6tUa8toZA0PM7NKDhwZikxOmHUzDhgRPNrR/bVTSkJmNjE5cGQoMjlh3k03SNo02tWW0GklITObeBw4ctTqxpr3S77VDeG1dHJJyMwmhk6dcqTjdepqerVKQl7Aycwa5RLHKDV7rY1m6dSSkJlNHA4cDejEUdmtWnXQzKzEgWOC6dSSkJlNHA4cE1AnloTMbOJw4OhAnh7EzDqZA8coterm7ulBzKzTOXCMQitv7kWnB3GpxMzaxeM4RqHazb1RRSZYbNf07WZm4MAxKkVnzx2NImtrtDJwmZnV4sAxCq1cOKnIiPRWBi4zs1ocOEahldONzJ3Tx6dP3K/q9CBe8c/M2smN4wVVNkafdGAft65c25LG6VrjMDw63MzaqWWBQ9LFwHHAmojYN007GZgPvAY4KCJGrOcqaXfgm8CfAZuBiyLiS+l784EPAmvTzT8eEddXHqPZsnpRXbF0oG0TBXp0uJm1UytLHJcAF5IEgZL7gBOBr1bZbyPw0Yi4S9IOwFJJN0XEr9L3vxARn2tFhvN04gp6Hh1uZu3SssAREbdLmlmRdj+ApGr7PQE8kT5/VtL9QB/wq9ydWsyN0WZmL+roxvE08MwBfl6WfKakeyRdLGnHsciHG6PNzF7UsYFD0vbAFcBZEfFMmvwfwCuB2SSlks9X2f8MSUskLVm7dm3eZoV06qJNZmbt0JGBQ1I3SdC4LCKuLKVHxJMRsSkiNgNfAw7KO0ZEXBQR/RHRP23atIbyU6SLrJnZZNFx3XGVNIB8A7g/Ii6oeG+3tA0E4ASSxvYx4cZoM7NEy0ocki4HfgbMkvS4pNMlnSDpceAvgesk3ZBuO11SqVvtocB7gMMlLU8fx6bvfVbSvZLuAQ4Dzm5V/s3MLJsiot15aLn+/v5YsmTEkBEzM6tC0tKI6K9M78g2DjMz61wOHGZmVhcHDjMzq8ukaOOQtBZ4tN35qNPOwFPtzkQH8fUYztdjOF+P4Zp1PfaIiBHjGSZF4BiPJC3JapSarHw9hvP1GM7XY7hWXw9XVZmZWV0cOMzMrC4OHJ3ronZnoMP4egzn6zGcr8dwLb0ebuMwM7O6uMRhZmZ1ceAwM7O6OHCMsXQBqjWS7itLO1nSCkmbJeV2oZP0SDrJ43JJE2LyrZzrsVDSynTBrqsk9ebse7SkVZIekHTOmGW6hRq8HpPl+/FP6bVYLulGSdNz9p0s34+i16N534+I8GMMH8AbgAOA+8rSXgPMAm4D+qvs+wiwc7s/wxhcjzcDW6XPPwN8JmO/LuBB4BXA1sDdwN7t/jztuh6T7PvxkrLnfwd8ZZJ/P2pej2Z/P1ziGGMRcTvwdEXa/RGxqk1Zaquc63FjRGxMX94JvDxj14OAByLioYh4AfgO8LaWZnYMNHA9JqSc6/FM2cvtgKwePpPp+1HkejSVA8f4EsCNkpZKOqPdmRkjHwB+mJHeBzxW9vrxNG2iy7seMIm+H5L+RdJjwKnAJzM2mVTfjwLXA5r4/XDgGF8OjYgDgGOAv5X0hnZnqJUknQdsBC7LejsjbUL3La9xPWASfT8i4ryI2J3kWpyZscmk+n4UuB7QxO+HA8c4EhGr03/XAFdRZc318U7SacBxwKmRVtBWeBzYvez1y4HVY5G3dihwPSbV96PMt4GTMtIn1fejTN71aOr3w4FjnJC0naQdSs9JGkzHbM31sSTpaOBjwPERsSFns18Ce0naU9LWwLuAq8cqj2OpyPWYZN+PvcpeHg+szNhsMn0/al6Ppn8/2t1LYLI9gMuBJ4Ahkl9FpwMnpM//BDwJ3JBuOx24Pn3+CpKeIXcDK4Dz2v1ZWng9HiCpn16ePr5SeT3S18cCvybpPTOpr8ck+35ckd707gGuAfom+fej5vVo9vfDU46YmVldXFVlZmZ1ceAwM7O6OHCYmVldHDjMzKwuDhxmZlYXBw6bsCS9LJ0JdLmk30kaKHu9dZPOcVv5jMaSZpbPXDpWJL1P0tr0s/1K0gdztjt+oswUa+2zVbszYNYqEfF7YDaApPnAcxHxudL7kraKFycPnAgWRcSZknYBVki6OiKeLL2Zft6rmaAD4WzsuMRhk4qkSyRdIOlW4DOS5kv6h7L375M0M33+bkm/SH/Ff1VSV53n2lbSf6ZrICyTdFia/j5JF5Ztd62kN0nqSvN3X7rP2en7r5T0o3Ryup9IenW180YypcSDwB4Zn3fLuSXtmq7vcXf6eF0zPrdNfC5x2GT0KuB/RsSmtCQygqTXAO8kmRhuSNK/k8w8+s2MzS+TNJg+3xrYnD7/W4CI2C+92d8o6VVV8jWbZNTvvmkeetP0i4APR8RvJB0M/DtweN5BJL2CZKTwAxmf931lm/5f4McRcUIaHLav83PbJOXAYZPR9yJiU41tjgAOBH4pCaAHWJOz7akRsQSSNg7g2jT99cCXASJipaRHSW7ieR4CXiHpy8B1JIFme+B1wPfSfABsk7P/OyW9nmTqmg9FxNPpPnmf93DgvWn+NgF/kPSeOj63TVIOHDYZPV/2fCPDq2y3Tf8VcGlEnNvAebKm9s49Z0Ssk7Q/cBRJaeUdwFnA+oiYXeB8iyIia0rt5zPS8jTjc9sE5zYOm+weIVmKE0kHAHum6TcDb08bmpG0k6Q96jz27STVPKRVVDOAVek5Z0uaIml30umtJe0MTImIK4B/BA6IZHW3hyWdnG6jNLg0w83A/0qP2yXpJTTnc9sE58Bhk90VwE6SlpPcRH8NEBG/Aj5BUl10D3ATsFudx/53oEvSvcAi4H0R8SfgDuBh4F7gc8Bd6fZ9wG1pXi4BSr/6TwVOl1Sa2bRZS6D+PXBYmr+lwD5N+tw2wXl2XDMzq4tLHGZmVhcHDjMzq4sDh5mZ1cWBw8zM6uLAYWZmdXHgMDOzujhwmJlZXf4/zkr7HVA/FH0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's evaluate our predictions respect to the real sale price\n",
    "plt.scatter(y_test, lin_model.predict(X_test))\n",
    "plt.xlabel('True House Price')\n",
    "plt.ylabel('Predicted House Price')\n",
    "plt.title('Evaluation of Lasso Predictions')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that our model is doing a pretty good job at estimating house prices."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>12.209188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11.798104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11.608236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12.165251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11.385092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141</th>\n",
       "      <td>11.884489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>142</th>\n",
       "      <td>12.287653</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>143</th>\n",
       "      <td>11.921718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>144</th>\n",
       "      <td>11.598727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>12.017331</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>146 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     SalePrice\n",
       "0    12.209188\n",
       "1    11.798104\n",
       "2    11.608236\n",
       "3    12.165251\n",
       "4    11.385092\n",
       "..         ...\n",
       "141  11.884489\n",
       "142  12.287653\n",
       "143  11.921718\n",
       "144  11.598727\n",
       "145  12.017331\n",
       "\n",
       "[146 rows x 1 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>12.209188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11.798104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11.608236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12.165251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11.385092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141</th>\n",
       "      <td>11.884489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>142</th>\n",
       "      <td>12.287653</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>143</th>\n",
       "      <td>11.921718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>144</th>\n",
       "      <td>11.598727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>12.017331</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>146 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     SalePrice\n",
       "0    12.209188\n",
       "1    11.798104\n",
       "2    11.608236\n",
       "3    12.165251\n",
       "4    11.385092\n",
       "..         ...\n",
       "141  11.884489\n",
       "142  12.287653\n",
       "143  11.921718\n",
       "144  11.598727\n",
       "145  12.017331\n",
       "\n",
       "[146 rows x 1 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      12.148226\n",
       "1      11.919326\n",
       "2      11.677107\n",
       "3      12.304289\n",
       "4      11.447473\n",
       "         ...    \n",
       "141    11.775100\n",
       "142    12.316546\n",
       "143    11.955957\n",
       "144    11.757571\n",
       "145    12.072691\n",
       "Length: 146, dtype: float64"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# let's evaluate the distribution of the errors: \n",
    "# they should be fairly normally distributed\n",
    "\n",
    "y_test.reset_index(drop=True, inplace=True)\n",
    "\n",
    "preds = pd.Series(lin_model.predict(X_test))\n",
    "\n",
    "preds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAD4CAYAAAD1jb0+AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAATC0lEQVR4nO3df4zkdX3H8ee7/IgHi4A9HelBu9paGnuLlhstVWN3RZsTjGhiUglaaGk2ttHS5kjFmNY0jSm1xf5Im5iLXqCRsG0RlYJtpciWNAK6h8iCh6L2onfiXSlyuHgVz777x87VY25n57sz35n9fo7nI9nczHe+8/285ns7r/3ud2c+E5mJJKk8P7beASRJg7HAJalQFrgkFcoCl6RCWeCSVKjjxznYxo0bc3JycqRjPPnkk5x88skjHWNQTc4Gzc7X5GzQ7HxNzgbNzteUbDt37nw0M5971A2ZObavLVu25KjdcccdIx9jUE3OltnsfE3OltnsfE3OltnsfE3JBizkCp3qKRRJKpQFLkmFssAlqVAWuCQVygKXpEJZ4JJUqL4FHhE7ImJ/RDzQtfxdEfHliHgwIj4wuoiSpJVUOQK/Fth65IKImAEuAs7JzJ8H/rz+aJKk1fQt8My8E3isa/FvAVdn5vc76+wfQTZJ0ioiK3ygQ0RMArdk5ubO9fuAT7J8ZP4/wJWZ+fke950FZgFardaWubm5WoL3srS0xMTExEjHGFSTs8Fo8y3uPVBpvalNp664/Jm874bV5GzQ7HxNyTYzM7MzM9vdywedC+V44HTgPOBlwD9ExAtzhZ8Gmbkd2A7Qbrdzenp6wCGrmZ+fZ9RjDKrJ2WC0+S676tZK6+2+ZOXxn8n7blhNzgbNztfkbDD4q1D2ADd13qb/OeB/gY31xZIk9TNogX8CeA1ARPwscCLwaE2ZJEkV9D2FEhE3ANPAxojYA7wP2AHs6Ly08Cng0pVOn0iSRqdvgWfmxT1uelvNWSRJa+A7MSWpUBa4JBXKApekQlngklQoC1ySCmWBS1KhLHBJKtSgc6FIz0iTPeZ02TZ16Kj5XnZffeE4IukZzCNwSSqUBS5JhbLAJalQFrgkFcoCl6RCWeCSVCgLXJIKZYFLUqH6FnhE7IiI/Z1P3+m+7cqIyIjw8zAlacyqHIFfC2ztXhgRZwGvA75RcyZJUgV9Czwz7wQeW+GmvwB+H/CzMCVpHQx0Djwi3gjszcwv1pxHklRRVPkw+YiYBG7JzM0RcRJwB/ArmXkgInYD7cx8tMd9Z4FZgFartWVubq6u7CtaWlpiYmJipGMMqsnZYLB8i3sP1JphatOpKy5vyr7r9XhbG2Dfwacv6/VYxq0p+66XJudrSraZmZmdmdnuXj5IgU8BtwPf69x8JvAt4OWZ+e3VttNut3NhYWGt2ddkfn6e6enpkY4xqCZng8Hy9Zqdb1C9ZvBryr5bbTbCaxafPrlnU2YjbMq+66XJ+ZqSLSJWLPA1TyebmYvA847Y8G5WOQKXJI1GlZcR3gDcBZwdEXsi4vLRx5Ik9dP3CDwzL+5z+2RtaSRJlflOTEkqlAUuSYWywCWpUBa4JBXKApekQlngklQoC1ySCmWBS1KhLHBJKpQFLkmFssAlqVAWuCQVygKXpEJZ4JJUKAtckgplgUtSoSxwSSpUlY9U2xER+yPigSOW/VlEPBQR90fExyPitJGmlCQdpcoR+LXA1q5ltwGbM/Mc4CvAe2rOJUnqo2+BZ+adwGNdyz6dmYc6V+8GzhxBNknSKiIz+68UMQnckpmbV7jtn4C/z8yP9rjvLDAL0Gq1tszNzQ0VuJ+lpSUmJiZGOsagmpwNBsu3uPdArRmmNp264vKm7Ltej7e1AfYdfPqyXo9l3Jqy73ppcr6mZJuZmdmZme3u5X0/lX41EfFe4BBwfa91MnM7sB2g3W7n9PT0MEP2NT8/z6jHGFSTs8Fg+S676tZaM+y+ZOXxm7Lvej3ebVOHuGbx6U+nXo9l3Jqy73ppcr4mZ4MhCjwiLgXeAJyfVQ7jJUm1GqjAI2Ir8G7glzPze/VGkiRVUeVlhDcAdwFnR8SeiLgc+BvgFOC2iLgvIj404pySpC59j8Az8+IVFn9kBFkkSWvgOzElqVAWuCQVygKXpEJZ4JJUKAtckgplgUtSoSxwSSrUUHOhSOtlsuIcLLuvvrDW7UlN4hG4JBXKApekQlngklQoC1ySCmWBS1KhLHBJKpQFLkmFssAlqVAWuCQVqspHqu2IiP0R8cARy54TEbdFxMOdf08fbUxJUrcqR+DXAlu7ll0F3J6ZLwJu71yXJI1R3wLPzDuBx7oWXwRc17l8HfCmemNJkvqJzOy/UsQkcEtmbu5cfzwzTzvi9u9k5oqnUSJiFpgFaLVaW+bm5mqI3dvS0hITExMjHWNQTc4Gg+Vb3Hug1gxTm05dcXl3tqrj9tpet2EfR2sD7Ds42Nijdix+341LU7LNzMzszMx29/KRz0aYmduB7QDtdjunp6dHOt78/DyjHmNQTc4Gg+W7rOZZ/HZfsvL43dmqjttre92GfRzbpg5xzeLTn05Vxx61Y/H7blyanA0GfxXKvog4A6Dz7/76IkmSqhi0wG8GLu1cvhT4ZD1xJElVVXkZ4Q3AXcDZEbEnIi4HrgZeFxEPA6/rXJckjVHfc+CZeXGPm86vOYskaQ18J6YkFcoCl6RCWeCSVCgLXJIKZYFLUqEscEkqlAUuSYUa+Vwo0lpM9piTZNvUoYHmK+m1vXGoOvbuqy8ccRIdqzwCl6RCWeCSVCgLXJIKZYFLUqEscEkqlAUuSYWywCWpUBa4JBXKApekQg1V4BHxexHxYEQ8EBE3RMSz6gomSVrdwAUeEZuA3wHambkZOA54a13BJEmrG/YUyvHAhog4HjgJ+NbwkSRJVURmDn7niCuA9wMHgU9n5iUrrDMLzAK0Wq0tc3NzA49XxdLSEhMTEyMdY1DrnW1x74FVb29tgH0HxxRmjZqcDYbLN7Xp1HrDdFnv77t+mpyvKdlmZmZ2Zma7e/nABR4RpwMfA34VeBz4R+DGzPxor/u02+1cWFgYaLyq5ufnmZ6eHukYg1rvbP1mx9s2dYhrFps5QWWTs8Fw+UY9G+F6f9/10+R8TckWESsW+DCnUF4L/Gdm/ldm/gC4CXjFENuTJK3BMAX+DeC8iDgpIgI4H9hVTyxJUj8DF3hm3gPcCNwLLHa2tb2mXJKkPoY6qZiZ7wPeV1MWSdIa+E5MSSqUBS5JhbLAJalQFrgkFcoCl6RCWeCSVCgLXJIKZYFLUqEscEkqlAUuSYWywCWpUBa4JBXKApekQlngklQoC1ySCmWBS1KhLHBJKtRQBR4Rp0XEjRHxUETsiohfqiuYJGl1Q32kGvBXwL9k5lsi4kTgpBoySZIqGLjAI+LZwKuBywAy8yngqXpiSZL6icwc7I4RL2X5U+i/BLwE2AlckZlPdq03C8wCtFqtLXNzc8Pk7WtpaYmJiYmRjjGoUWVb3Huglu20NsC+g7VsqnZNzgbD5ZvadGq9Ybo0+TkBzc7XlGwzMzM7M7PdvXyYAm8DdwOvzMx7IuKvgCcy8w963afdbufCwsJA41U1Pz/P9PT0SMcY1KiyTV51ay3b2TZ1iGsWhz2rNhpNzgbD5dt99YU1p3m6Jj8noNn5mpItIlYs8GH+iLkH2JOZ93Su3wicO8T2JElrMHCBZ+a3gW9GxNmdReezfDpFkjQGw/5O+i7g+s4rUL4O/PrwkSRJVQxV4Jl5H3DUeRlJ0uj5TkxJKpQFLkmFssAlqVAWuCQVygKXpEJZ4JJUKAtckgrV3MkljmFV5y4Z9RwZksrmEbgkFcoCl6RCWeCSVCgLXJIKZYFLUqEscEkqlAUuSYWywCWpUBa4JBVq6AKPiOMi4gsRcUsdgSRJ1dRxBH4FsKuG7UiS1mCoAo+IM4ELgQ/XE0eSVFVk5uB3jrgR+BPgFODKzHzDCuvMArMArVZry9zc3MDjVbG0tMTExMRIxxjU4WyLew+sd5QVtTbAvoPrnWJlTc4Gw+Wb2nRqpfWqft90b6/Jzwlodr6mZJuZmdmZmUd9gPzAsxFGxBuA/Zm5MyKme62XmduB7QDtdjunp3uuWov5+XlGPcagDme7rOJshOO2beoQ1yw2c4LKJmeD4fLtvmS60npVv2+6t9fk5wQ0O1+Ts8Fwp1BeCbwxInYDc8BrIuKjtaSSJPU1cIFn5nsy88zMnATeCnwmM99WWzJJ0qp8HbgkFaqWk4qZOQ/M17EtSVI1HoFLUqEscEkqlAUuSYWywCWpUBa4JBXKApekQlngklSo5k4uIT1DTK7T3DhVx9199YUjTqJBeQQuSYWywCWpUBa4JBXKApekQlngklQoC1ySCmWBS1KhLHBJKpQFLkmFGrjAI+KsiLgjInZFxIMRcUWdwSRJqxvmrfSHgG2ZeW9EnALsjIjbMvNLNWWTJK1imE+lfyQz7+1c/i6wC9hUVzBJ0uoiM4ffSMQkcCewOTOf6LptFpgFaLVaW+bm5oYebzVLS0tMTEz0XW9x74HK25zadOowkf7f4WxrGXucWhtg38H1TrGyJmeDZucbZ7ZBnitVn7ProSnZZmZmdmZmu3v50AUeERPAvwPvz8ybVlu33W7nwsLCUOP1Mz8/z/T0dN/11jIDXF2zsR3Otl6zz/WzbeoQ1yw2c4LKJmeDZucbZ7ZBnitVn7ProSnZImLFAh/qVSgRcQLwMeD6fuUtSarXMK9CCeAjwK7M/GB9kSRJVQxzBP5K4O3AayLivs7XBTXlkiT1MfCJscz8DyBqzCJJWgPfiSlJhbLAJalQFrgkFcoCl6RCWeCSVCgLXJIKZYFLUqGaOXnDCqrOH3Lt1pNHnKS3fhm3TR3isobOgyKN05HPldWeF3XNQ7RWh/PV+ZwdxWPxCFySCmWBS1KhLHBJKpQFLkmFssAlqVAWuCQVygKXpEJZ4JJUKAtckgo17Icab42IL0fEVyPiqrpCSZL6G+ZDjY8D/hZ4PfBi4OKIeHFdwSRJqxvmCPzlwFcz8+uZ+RQwB1xUTyxJUj+RmYPdMeItwNbM/M3O9bcDv5iZ7+xabxaY7Vw9G/jy4HEr2Qg8OuIxBtXkbNDsfE3OBs3O1+Rs0Ox8Tcn2U5n53O6Fw8xGuNIn0h/10yAztwPbhxhnTSJiITPb4xpvLZqcDZqdr8nZoNn5mpwNmp2vydlguFMoe4Czjrh+JvCt4eJIkqoapsA/D7woIl4QEScCbwVurieWJKmfgU+hZOahiHgn8K/AccCOzHywtmSDG9vpmgE0ORs0O1+Ts0Gz8zU5GzQ7X5OzDf5HTEnS+vKdmJJUKAtckgpVfIFHxHMi4raIeLjz7+krrPOsiPhcRHwxIh6MiD9qULazIuKOiNjVyXbFOLJVzddZb0dE7I+IB8aQadXpGWLZX3duvz8izh11pjVk+7mIuCsivh8RV44r1xryXdLZZ/dHxGcj4iUNynZRJ9d9EbEQEa8aV7Yq+Y5Y72UR8cPO+2DWX2YW/QV8ALiqc/kq4E9XWCeAic7lE4B7gPMaku0M4NzO5VOArwAvbsq+69z2auBc4IER5zkO+BrwQuBE4Ivd+wK4APjnzv/pecA9Y9pXVbI9D3gZ8H7gynHkWmO+VwCndy6/vmH7boIf/U3uHOChJu27I9b7DPAp4C3j/P/t9VX8ETjLb9+/rnP5OuBN3SvksqXO1RM6X+P4622VbI9k5r2dy98FdgGbxpCtUj6AzLwTeGwMeapMz3AR8Hed/9O7gdMi4owmZMvM/Zn5eeAHY8gzSL7PZuZ3OlfvZvm9G03JtpSdlgROZjzPz8r5Ot4FfAzYP8ZsqzoWCryVmY/AchmyfBR0lIg4LiLuY3nn35aZ9zQl22ERMQn8Asu/IYzDmvKNwSbgm0dc38PRP8yqrDMK6zVuVWvNdznLv8mMQ6VsEfHmiHgIuBX4jTFlgwr5ImIT8GbgQ2PM1dcwb6Ufm4j4N+D5K9z03qrbyMwfAi+NiNOAj0fE5swc+pxuHdk625lg+af772bmE8PmOmK7teQbkyrTM1SawmEE1mvcqirni4gZlgt8XOeZq0678XGWn5uvBv4YeO2og3VUyfeXwLsz84cRK62+Pooo8Mzs+R8ZEfsi4ozMfKTzq/Sqv95k5uMRMQ9sBYYu8DqyRcQJLJf39Zl507CZ6s43RlWmZ1ivKRyaPnVEpXwRcQ7wYeD1mfnfTcp2WGbeGRE/HREbM3McE0lVydcG5jrlvRG4ICIOZeYnxpCvp2PhFMrNwKWdy5cCn+xeISKe2znyJiI2sPyT/aGGZAvgI8CuzPzgGDIdqW++MasyPcPNwK91Xo1yHnDg8GmgBmRbT33zRcRPAjcBb8/MrzQs2890ngt0Xll0IjCuHzB982XmCzJzMjMngRuB317v8gaOiVeh/DhwO/Bw59/ndJb/BPCp/NFftb8A3M/yUfcfNijbq1j+de1+4L7O1wVNyde5fgPwCMt/nNsDXD7CTBew/EqcrwHv7Sx7B/COzuVg+YNEvgYsAu0xfq/1y/b8zv55Ani8c/nZDcr3YeA7R3yfLTQo27uBBzu57gJeNa5sVfJ1rXstDXkVim+ll6RCHQunUCTpGckCl6RCWeCSVCgLXJIKZYFLUqEscEkqlAUuSYX6P/KEn1VbO70FAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's evaluate the distribution of the errors: \n",
    "# they should be fairly normally distributed\n",
    "\n",
    "errors = y_test['SalePrice'] - preds\n",
    "errors.hist(bins=30)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The distribution of the errors follows quite closely a gaussian distribution. That suggests that our model is doing a good job as well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.03962646, 0.10524604, 0.0049094 , 0.02031924, 0.00886536,\n",
       "       0.0244373 , 0.24923157, 0.46945057, 0.28749129, 0.01991704,\n",
       "       0.01447198, 0.02216442, 0.01902546, 0.03124932, 0.07560124,\n",
       "       0.05338536, 0.05869799, 0.04266609, 0.0664621 , 0.38327925,\n",
       "       0.02840977, 0.82388482, 0.096785  , 0.03540985, 0.08241899,\n",
       "       0.0466837 , 0.06404299, 0.01865053, 0.05455405, 0.0128952 ,\n",
       "       0.18728765, 0.00628094, 0.01645246, 0.04327042, 0.03501549,\n",
       "       0.12598851])"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.abs(lin_model.coef_.ravel())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Feature Importance')"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Finally, just for fun, let's look at the feature importance\n",
    "\n",
    "importance = pd.Series(np.abs(lin_model.coef_.ravel()))\n",
    "importance.index = features\n",
    "importance.sort_values(inplace=True, ascending=False)\n",
    "importance.plot.bar(figsize=(18,6))\n",
    "plt.ylabel('Lasso Coefficients')\n",
    "plt.title('Feature Importance')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Save the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['linear_regression.joblib']"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# we are happy to our model, so we save it to be able\n",
    "# to score new data\n",
    "\n",
    "joblib.dump(lin_model, 'linear_regression.joblib') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "583px",
    "left": "0px",
    "right": "1324px",
    "top": "107px",
    "width": "212px"
   },
   "toc_section_display": "block",
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}