2_Predictive_Analytics_Regression_Boston_House_Price_Prediction.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "2_Predictive Analytics (Regression).ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "UAcxFlqAzByp"
      },
      "source": [
        "# Predictive Modeling - Regression"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wc3aoVBl-Wo0",
        "colab_type": "text"
      },
      "source": [
        "This workshop contains predictive analytics for house price prediction using Boston dataset.  \n",
        "\n",
        "Step 1: Predict house prices directly from the dataset using linear regression   \n",
        "1.  Load the Boston dataset.  \n",
        "2.  Check for missing values.  \n",
        "3.  Run Linear Regression Model without any data preprocessing.  \n",
        "  \n",
        "Step 2:  Pre-process data for improving accuracy of the predictive model    \n",
        "1. Explore the dataset using boxplots.   \n",
        "2. Standardize features due to findings from boxplots.  \n",
        "3. Run Linear regression model  \n",
        "4. Explore the dataset using correlation plots.  \n",
        "\n",
        "Step 3:   Neural Network Regression using pre-processed data    \n",
        "1. Run Neural Network to predict house price.  "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cBWer2oZz6N1",
        "colab_type": "text"
      },
      "source": [
        "## Step 1: Predict house prices using linear regression"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AW7QMtZb-eaq",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Data Manipulation libraries\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "\n",
        "# Visualization libraries\n",
        "import matplotlib as mpl\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "%matplotlib inline\n",
        "\n",
        "# Predictive Analytics models\n",
        "import sklearn\n",
        "from sklearn import linear_model\n",
        "from sklearn.metrics import mean_squared_error\n",
        "from sklearn.model_selection  import train_test_split"
      ],
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kCNaGig_0now",
        "colab_type": "text"
      },
      "source": [
        "### Load the dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dod-h3sF-ihE",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df = pd.read_csv('tp3_boston_data.csv')  # Load the dataset"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_XgmlTQg-z9s",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "e7fa1489-c4c3-4ea1-c703-36f9c65c637b"
      },
      "source": [
        "print(df.keys())  # We print the value of the boston_dataset to understand what it contains. print(boston_dataset.keys()) gives"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Index(['crim', 'zn', 'indus', 'chas', 'nox', 'rm', 'age', 'dis', 'rad', 'tax',\n",
            "       'ptratio', 'lstat', 'medv'],\n",
            "      dtype='object')\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6RiFFtg8rLGk",
        "colab_type": "text"
      },
      "source": [
        "**Description of the dataset:**  \n",
        "CRIM - per capita crime rate by town  \n",
        "ZN - proportion of residential land zoned for lots over 25,000 sq.ft.  \n",
        "INDUS - proportion of non-retail business acres per town.  \n",
        "CHAS - Charles River dummy variable (1 if tract bounds river; 0 otherwise)  \n",
        "NOX - nitric oxides concentration (parts per 10 million)  \n",
        "RM - average number of rooms per dwelling  \n",
        "AGE - proportion of owner-occupied units built prior to 1940  \n",
        "DIS - weighted distances to five Boston employment centres  \n",
        "RAD - index of accessibility to radial highways  \n",
        "TAX - full-value property-tax rate per 10,000 dollars  \n",
        "PTRATIO - pupil-teacher ratio by town  \n",
        "LSTAT - % lower status of the population  \n",
        "MEDV - Median value of owner-occupied homes in $1000's (target)   "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O63NMQVw-3hV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# split the dataset into predictors (df_x) and target variable (df_y)\n",
        "df_x = df[['crim', 'zn', 'indus', 'chas', 'nox', 'rm', 'age', 'dis', 'rad', 'tax', 'ptratio', 'lstat']]\n",
        "df_y = df[['medv']]"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sHr3EF2i06MM",
        "colab_type": "text"
      },
      "source": [
        "### Check for missing values"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5aqxT0vb08uE",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "outputId": "d279511c-9b5c-4ac9-fd3b-0c588e8ad17c"
      },
      "source": [
        "# Check for missing values\n",
        "df_x.isnull().sum()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "crim       0\n",
              "zn         0\n",
              "indus      0\n",
              "chas       0\n",
              "nox        0\n",
              "rm         0\n",
              "age        0\n",
              "dis        0\n",
              "rad        0\n",
              "tax        0\n",
              "ptratio    0\n",
              "lstat      0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dPYnTwQ5SUsb",
        "colab_type": "text"
      },
      "source": [
        "No missing values for predictors."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ekb_n3hp1GmF",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "c663329d-c9f8-48c4-af38-7a8ee57c47d5"
      },
      "source": [
        "df_y.isnull().sum()"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "medv    0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-GHQFOVX1KEC",
        "colab_type": "text"
      },
      "source": [
        "No missing values.  "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rfnfHqZv_PJs",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "ddb568c9-ddc9-4b97-98d6-3af1f2371611"
      },
      "source": [
        "# Have a look at the predictors\n",
        "df_x.head()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>crim</th>\n",
              "      <th>zn</th>\n",
              "      <th>indus</th>\n",
              "      <th>chas</th>\n",
              "      <th>nox</th>\n",
              "      <th>rm</th>\n",
              "      <th>age</th>\n",
              "      <th>dis</th>\n",
              "      <th>rad</th>\n",
              "      <th>tax</th>\n",
              "      <th>ptratio</th>\n",
              "      <th>lstat</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.00632</td>\n",
              "      <td>18.0</td>\n",
              "      <td>2.31</td>\n",
              "      <td>0</td>\n",
              "      <td>0.538</td>\n",
              "      <td>6.575</td>\n",
              "      <td>65.2</td>\n",
              "      <td>4.0900</td>\n",
              "      <td>1</td>\n",
              "      <td>296</td>\n",
              "      <td>15.3</td>\n",
              "      <td>4.98</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.02731</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.07</td>\n",
              "      <td>0</td>\n",
              "      <td>0.469</td>\n",
              "      <td>6.421</td>\n",
              "      <td>78.9</td>\n",
              "      <td>4.9671</td>\n",
              "      <td>2</td>\n",
              "      <td>242</td>\n",
              "      <td>17.8</td>\n",
              "      <td>9.14</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.02729</td>\n",
              "      <td>0.0</td>\n",
              "      <td>7.07</td>\n",
              "      <td>0</td>\n",
              "      <td>0.469</td>\n",
              "      <td>7.185</td>\n",
              "      <td>61.1</td>\n",
              "      <td>4.9671</td>\n",
              "      <td>2</td>\n",
              "      <td>242</td>\n",
              "      <td>17.8</td>\n",
              "      <td>4.03</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.03237</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.18</td>\n",
              "      <td>0</td>\n",
              "      <td>0.458</td>\n",
              "      <td>6.998</td>\n",
              "      <td>45.8</td>\n",
              "      <td>6.0622</td>\n",
              "      <td>3</td>\n",
              "      <td>222</td>\n",
              "      <td>18.7</td>\n",
              "      <td>2.94</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.06905</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.18</td>\n",
              "      <td>0</td>\n",
              "      <td>0.458</td>\n",
              "      <td>7.147</td>\n",
              "      <td>54.2</td>\n",
              "      <td>6.0622</td>\n",
              "      <td>3</td>\n",
              "      <td>222</td>\n",
              "      <td>18.7</td>\n",
              "      <td>5.33</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "      crim    zn  indus  chas    nox  ...     dis  rad  tax  ptratio  lstat\n",
              "0  0.00632  18.0   2.31     0  0.538  ...  4.0900    1  296     15.3   4.98\n",
              "1  0.02731   0.0   7.07     0  0.469  ...  4.9671    2  242     17.8   9.14\n",
              "2  0.02729   0.0   7.07     0  0.469  ...  4.9671    2  242     17.8   4.03\n",
              "3  0.03237   0.0   2.18     0  0.458  ...  6.0622    3  222     18.7   2.94\n",
              "4  0.06905   0.0   2.18     0  0.458  ...  6.0622    3  222     18.7   5.33\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lt-lfnTEb6-E",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 340
        },
        "outputId": "8b568300-a643-4e57-b441-29702ee08cf0"
      },
      "source": [
        "df_x.info()"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 506 entries, 0 to 505\n",
            "Data columns (total 12 columns):\n",
            " #   Column   Non-Null Count  Dtype  \n",
            "---  ------   --------------  -----  \n",
            " 0   crim     506 non-null    float64\n",
            " 1   zn       506 non-null    float64\n",
            " 2   indus    506 non-null    float64\n",
            " 3   chas     506 non-null    int64  \n",
            " 4   nox      506 non-null    float64\n",
            " 5   rm       506 non-null    float64\n",
            " 6   age      506 non-null    float64\n",
            " 7   dis      506 non-null    float64\n",
            " 8   rad      506 non-null    int64  \n",
            " 9   tax      506 non-null    int64  \n",
            " 10  ptratio  506 non-null    float64\n",
            " 11  lstat    506 non-null    float64\n",
            "dtypes: float64(9), int64(3)\n",
            "memory usage: 47.6 KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pXs6RfGe1Uyo",
        "colab_type": "text"
      },
      "source": [
        "### Split the data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "x_WYOR8h_QWd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "X_train, X_test, Y_train, Y_test = train_test_split(df_x, df_y, test_size = 0.33, random_state = 5)"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V_hr87CD1Zlh",
        "colab_type": "text"
      },
      "source": [
        "### Linear regression modeling"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KyXFJXYwo4DD",
        "colab_type": "text"
      },
      "source": [
        "Method API: https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yThjltzP_Wo9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Setup the machine learning model\n",
        "reg = linear_model.LinearRegression()\n",
        "\n",
        "# Fit the machine learning model\n",
        "reg.fit(X_train,Y_train)\n",
        "\n",
        "# Predict from the machine leanring model\n",
        "Y_predict = reg.predict(X_test)"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hgmjZF0W_bME",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "c155db05-df80-497b-a577-8b8c6fdf7b0e"
      },
      "source": [
        "plt.scatter(Y_test, Y_predict, color='blue')\n",
        "plt.xlabel(\"Prices\")\n",
        "plt.ylabel(\"Predicted prices\")\n",
        "plt.title(\"Prices vs Predicted prices\")\n",
        "plt.show()"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "adPMWCTY_gP0",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "3825f181-38af-49dc-e080-909d7d713b40"
      },
      "source": [
        "# Calculate \n",
        "error = np.sqrt(np.mean((Y_predict - Y_test)**2))\n",
        "print('Root Mean Squared Error: ' + str(error))"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Root Mean Squared Error: medv    5.309472\n",
            "dtype: float64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "B8Mg50bJ_kH1",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "229176de-7eef-4462-b480-75ccadc73e86"
      },
      "source": [
        "error_1 = np.sqrt(mean_squared_error(Y_test, Y_predict))\n",
        "print('Root Mean Squared Error: ' + str(error_1))"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Root Mean Squared Error: 5.309471828719486\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7hRykIz71gAf",
        "colab_type": "text"
      },
      "source": [
        "## Step 2: Pre-process data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vTpX1vNW10Do",
        "colab_type": "text"
      },
      "source": [
        "### Box plot analysis"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8jQJeGIReqO_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "67a30db5-7448-4077-ddc0-a0dc424f142e"
      },
      "source": [
        "# Box plot analysis\n",
        "df_x.boxplot()"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fc80a7ed550>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD4CAYAAAAXUaZHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAf6UlEQVR4nO3df5xcdX3v8debzS/IIpEfdx8IIcu1PDQ0yI9NFSRqNikSLG3yaCm3qEnA+MjFi6m1iolyq0JFk9JeLhZLi4abBCEr4uWHC0ZosqsFLmqiEELij4gEEikUTVI3kJAfn/vH+e4y2czuzs7O7O6cvJ+PxzzmnO/5nvP5zpkzn/nO98zMUURgZmb5csRQN8DMzCrPyd3MLIec3M3McsjJ3cwsh5zczcxyaMRQNwDg+OOPj8bGxn6vt2vXLsaOHVv5Bjme49VQLMc7fOOtW7fu5Yg4oejCiBjyW1NTU5Sjra2trPXK5XiONxxjOd7hGw9YGz3kVQ/LmJnlkJO7mVkOObmbmeWQk7uZWQ45uZuZ5ZCTu5lZDjm5m5nlkJO7mVkO9fkLVUlvAb5RUPRfgc8CK1J5I/AscGlEbJck4CbgfcArwOUR8ePKNtvM8i5LJT0LX4uiV3323CPiZxFxVkScBTSRJex7gEXA6og4DVid5gEuAk5Lt/nALdVouJnlW+GvLScsbD3kF5jWu/4Oy0wHfhkRW4CZwPJUvhyYlaZnAivSr2MfB8ZJOrEirTUzs5KoP++Akm4DfhwRN0vaERHjUrmA7RExTlIrsDgiHknLVgMLI2Jtt23NJ+vZ09DQ0NTS0tLvxnd0dFBfX9/v9crleI43HGMdDvEuX7WLZTMG74+8amV/Njc3r4uIyUUX9vSnM0U+Ao0CXgYa0vyObsu3p/tWYEpB+Wpgcm/b9h+HOV7e4uX5sQ1FvAkLWwc1Xq3sTyr0x2EXkfXaX0zzL3YOt6T7l1L5NmB8wXonpzIzMxsk/UnulwErC+bvB+am6bnAfQXlc5Q5F9gZES8MuKVmZlayki7WIWkscAHw3wuKFwN3SZoHbAEuTeUPkn0NcjPZN2uuqFhrzcysJCUl94jYBRzXrew3ZN+e6V43gKsq0jozMyuLf6FqZpZDTu5mZjnk5G5mlkNO7mZmOVTSCVUzs2o789qH2Pnq3h6XNy56oGj5MUeO5MnPvbdazapZTu5mNizsfHUvzy7+o6LL2tvbmTp1atFlPSX9w52HZczMcsjJ3cwsh5zczcxyyMndzCyHnNzNzHLIyd3MLIec3M3McsjJ3cwsh5zczcxyyMndzCyHnNzNzHLIyd3MLIec3M3Mcqik5C5pnKS7Jf1U0iZJ50k6VtLDkn6R7t+Y6krSlyVtlrRe0jnVfQhmZtZdqT33m4BVEfFW4ExgE7AIWB0RpwGr0zzARcBp6TYfuKWiLTYzsz71mdwlHQO8G1gKEBGvRcQOYCawPFVbDsxK0zOBFZF5HBgn6cSKt9zMzHqkiOi9gnQWcCuwkazXvg74GLAtIsalOgK2R8Q4Sa3A4oh4JC1bDSyMiLXdtjufrGdPQ0NDU0tLS78b39HRQX19fb/XK5fjOd5wjJWXeJev2sWyGWP7Ha+39cpVK/uzubl5XURMLrowInq9AZOBfcA70vxNwN8CO7rV257uW4EpBeWrgcm9xWhqaopytLW1lbVeuRzP8YZjrLzEm7Cwtax4va1XrlrZn8Da6CGvljLmvhXYGhE/SPN3A+cAL3YOt6T7l9LybcD4gvVPTmVmZjZI+kzuEfHvwPOS3pKKppMN0dwPzE1lc4H70vT9wJz0rZlzgZ0R8UJlm21mZr0p9QLZC4A7JI0CngGuIHtjuEvSPGALcGmq+yDwPmAz8Eqqa2Zmg6ik5B4RT5CNvXc3vUjdAK4aYLvMzGwA/AtVM7MccnI3M8shJ3czsxxycjczyyEndzOzHHJyNzPLISd3M7MccnI3M8shJ3czsxxycjczyyEndzOzHHJyNzPLISd3M7MccnI3M8shJ3czsxxycjczyyEndzOzHHJyNzPLISd3M7McKim5S3pW0lOSnpC0NpUdK+lhSb9I929M5ZL0ZUmbJa2XdE41H4CZmR2qPz335og4KyI6L5S9CFgdEacBq9M8wEXAaek2H7ilUo01M7PSDGRYZiawPE0vB2YVlK+IzOPAOEknDiCOmZn1kyKi70rSr4DtQAD/EhG3StoREePScgHbI2KcpFZgcUQ8kpatBhZGxNpu25xP1rOnoaGhqaWlpd+N7+jooL6+vt/rlcvxHG84xspLvMtX7WLZjLH9jtfbeuWqlf3Z3Ny8rmA05WAR0ecNOCnd/xfgSeDdwI5udban+1ZgSkH5amByb9tvamqKcrS1tZW1Xrkcz/GGY6y8xJuwsLWseL2tV65a2Z/A2ughr5Y0LBMR29L9S8A9wNuBFzuHW9L9S6n6NmB8weonpzIzMxskfSZ3SWMlHd05DbwX2ADcD8xN1eYC96Xp+4E56Vsz5wI7I+KFirfczMx6NKKEOg3APdmwOiOAOyNilaQfAXdJmgdsAS5N9R8E3gdsBl4Brqh4q83MrFd9JveIeAY4s0j5b4DpRcoDuKoirTMzs7L4F6pmZjnk5G5mlkNO7mZmOeTkbmaWQ07uZmY55ORuZpZDTu5mZjnk5G5mlkNO7mZmOeTkbmaWQ07uZmY55ORuZpZDTu5mZjnk5G5mlkNO7mZmOeTkbmaWQ07uZmY55ORuZpZDTu5mZjlUcnKXVCfpJ5Ja0/ypkn4gabOkb0galcpHp/nNaXljdZpuZmY96U/P/WPApoL5JcCNEfF7wHZgXiqfB2xP5TememZmNohKSu6STgb+CPhamhcwDbg7VVkOzErTM9M8afn0VN/MzAaJIqLvStLdwJeAo4FPApcDj6feOZLGA9+JiEmSNgAzImJrWvZL4B0R8XK3bc4H5gM0NDQ0tbS09LvxHR0d1NfX93u9cjme4w3HWHmJd/mqXSybMbbf8Xpbr1y1sj+bm5vXRcTkogsjotcbcDHwT2l6KtAKHA9sLqgzHtiQpjcAJxcs+yVwfG8xmpqaohxtbW1lrVcux3O84RgrL/EmLGwtK15v65WrVvYnsDZ6yKsjSnhzOB/4E0nvA8YAbwBuAsZJGhER+4CTgW2p/raU7LdKGgEcA/ymH29GZmY2QH2OuUfEpyPi5IhoBP4CWBMRHwDagEtStbnAfWn6/jRPWr4mvcOYmdkgGcj33BcCfy1pM3AcsDSVLwWOS+V/DSwaWBPNzKy/ShmW6RIR7UB7mn4GeHuROruBP69A28zMrEz+haqZWQ45uZuZ5ZCTu5lZDjm5m5nlkJO7mVkOObmbmeWQk7uZWQ45uZuZ5ZCTu5lZDjm5m5nlkJO7mVkOObmbmeWQk7uZWQ71618hzcyq5eiJizhjeS//EL68ePHREyG7xLMVcnI3s2Hhd5sW8+zi4km6vb2dqVOnFl3WuOiBKraqdnlYxswsh5zczcxyyMndzCyH+kzuksZI+qGkJyU9LenaVH6qpB9I2izpG5JGpfLRaX5zWt5Y3YdgZmbdldJz3wNMi4gzgbOAGZLOBZYAN0bE7wHbgXmp/jxgeyq/MdUzM7NB1Gdyj0xHmh2ZbgFMA+5O5cuBWWl6Jq9/aeluYLokVazFZmbWp5LG3CXVSXoCeAl4GPglsCMi9qUqW4GT0vRJwPMAaflO4LhKNtrMzHqniCi9sjQOuAf4G2BZGnpB0njgOxExSdIGYEZEbE3Lfgm8IyJe7rat+cB8gIaGhqaWlpZ+N76jo4P6+vp+r1cux3O84RgrL/EuX7WLZTPG9jteb+uVq1b2Z3Nz87qImFx0YUT06wZ8FrgaeBkYkcrOA76bpr8LnJemR6R66m2bTU1NUY62tray1iuX4znecIyVl3gTFraWFa+39cpVK/sTWBs95NVSvi1zQuqxI+lI4AJgE9AGXJKqzQXuS9P3p3nS8jWpEWZmNkhK+fuBE4HlkurIxujviohWSRuBFklfAH4CLE31lwK3S9oM/Bb4iyq028zMetFnco+I9cDZRcqfAd5epHw38OcVaZ2ZmZXFfxxmZsNGr38Ctqr4smOOHFml1tQ2J3czGxZ6+kdIyJJ+b8vtUP5vGTOzHHJyNzPLISd3M7MccnI3M8shJ3czsxxycjczyyEndzOzHHJyNzPLISd3M7MccnI3M8shJ3czsxxycjczyyEndzOzHHJyNzPLISd3M7MccnI3M8shJ3czsxzqM7lLGi+pTdJGSU9L+lgqP1bSw5J+ke7fmMol6cuSNktaL+mcaj8IMzM7WCk9933AJyLidOBc4CpJpwOLgNURcRqwOs0DXASclm7zgVsq3mozM+tVn8k9Il6IiB+n6d8Bm4CTgJnA8lRtOTArTc8EVkTmcWCcpBMr3nIzM+uRIqL0ylIj8H1gEvBcRIxL5QK2R8Q4Sa3A4oh4JC1bDSyMiLXdtjWfrGdPQ0NDU0tLS78b39HRQX19fb/XK5fjOd5wjHU4xLt81S6WzRg7aPFqZX82Nzevi4jJRRdGREk3oB5YB/xpmt/Rbfn2dN8KTCkoXw1M7m3bTU1NUY62tray1iuX4znecIx1OMSbsLB1UOPVyv4E1kYPebWkb8tIGgl8C7gjIv5vKn6xc7gl3b+UyrcB4wtWPzmVmZnZICnl2zIClgKbIuJ/FSy6H5ibpucC9xWUz0nfmjkX2BkRL1SwzWZm1ocRJdQ5H5gNPCXpiVT2GWAxcJekecAW4NK07EHgfcBm4BXgioq22MwOC1m/smB+ycHLox/nCw9HfSb3yE6MqofF04vUD+CqAbbLzA5zhcm7vb2dqVOnDl1japB/oWpmlkNO7mZmOeTkbmaWQ07uZmY55ORuZpZDTu5mZjnk5G5mlkNO7mZmOeTkbmaWQ07uZmY55ORuZpZDTu5mZjnk5G5mlkNO7mZmOeTkbmaWQ07uZmY55ORuZpZDTu5mZjnk5G5mlkN9JndJt0l6SdKGgrJjJT0s6Rfp/o2pXJK+LGmzpPWSzqlm4wfLypUrmTRpEtOnT2fSpEmsXLlyqJtkZtarPi+QDSwDbgZWFJQtAlZHxGJJi9L8QuAi4LR0ewdwS7qvWStXruSaa65h6dKl7N+/n7q6OubNmwfAZZddNsStMzMrrs+ee0R8H/htt+KZwPI0vRyYVVC+IjKPA+MknVipxg6F66+/HoBp06ZxwQUXMG3atIPKzcyGI0VE35WkRqA1Iial+R0RMS5NC9geEeMktQKLI+KRtGw1sDAi1hbZ5nxgPkBDQ0NTS0tLyY1evXo1X//613nuuec45ZRT+OAHP8j06dNLXr8/mpube1zW1tZWlZidOjo6qK+vr2oMx6uc3o4VqO7xkrd96XilaW5uXhcRk4sujIg+b0AjsKFgfke35dvTfSswpaB8NTC5r+03NTVFqe6888449dRTY82aNfHwww/HmjVr4tRTT40777yz5G30BxBASDroPtt11dXW1lb1GI5XHRMWtg5arIh870vH6xmwNnrIq6WMuRfzoqQTI+KFNOzyUirfBowvqHdyKquY66+/nqVLl9Lc3Ex7eztTp05l6dKlLFiwoKpj4JKIiK57szOvfYidr+7tcXnjogeKlh9z5Eie/Nx7q9UsM6C0E6rF3A/MBRan+/sKyj8qqYXsROrOiHhhwK0ssGnTJqZMmXJQ2ZQpU9i0aVMlwxziwIEDB92b7Xx1L88u/qOiyzo7HsX0lPTNKqnP5C5pJTAVOF7SVuBzZEn9LknzgC3Apan6g8D7gM3AK8AVlW7wxIkTufbaa7n33nvZtGkTEydOZNasWUycOLHSoczMalafyT0iehrrOOQMZhoDumqgjepNc3MzS5YsYcmSJZx++uls3LiRhQsXcuWVV1YzrJlZTSl3WGbItLW1cfHFF/OZz3yGPXv2MHr0aC6++OKqf3PFrLujJy7ijOWLeq6wvHjx0RMBig/nmFVKzSX3jRs3smvXLr7zne90/ajoQx/6EFu2bKlq3Lq6uq54+/fvr2osqw2/27TYY+42bNXcf8uMGjWKBQsW0NzczIgRI2hubmbBggWMGjWqqnE7E7oTu5nVgprrub/22mvcfPPNnH322ezfv5+2tjZuvvlmXnvttaFumpnZsFFzyf30009n1qxZLFiwoOvbMu9///u59957h7ppZmbDRs0l92uuuYaPfexjjB07FoBdu3Zx6623ctNNNw1xy8zMho+aS+4Au3fvZseOHRw4cIBt27YxZsyYoW6SmdmwUnMnVD/1qU9RV1fHSSedhCROOukk6urq+NSnPlXVuPX19dxyyy2D+mdCZmblqrnkvnXrVvbs2cO2bduICLZt28aePXvYunVrVeN2dHTwkY98hI6OjqrGMbPDRzUvBFSTwzJ79uzhhhtu6PqF6tVXXz3UTbLDVK/fWV/V8x+Hma1cuZLZs2d3fb366aefZvbs2UBlLgRUk8n9qKOO6voq5Nlnn81RRx3lHrUNup5+wARZ0u9tudmcOXPYv38/Y8aMYffu3V33c+bMOXyT+/79+7nwwgvZu3cvI0eOZMSI6j+M+vp6brjhBq6++mq/kZjZgO3bt49Ro0bx4IMPdv36fcaMGRX7zU7NjbnX1dXx6quvsndv9j/ae/fu5dVXX6Wurq6qcT3mbmaVtmjRooN+bb9oUS//VdRPNZfce/o/9Wr/z3pjYyO33347jY2NVY1jZvmXXZ0UrrvuOiTR3NyMJK677rqKxai5YZmeroJUjasjdT4BAM8++2zXyY5qKYxXjK8AZZYPEcHIkSPZt2/fIWPulRpmrrnkPpgigre97W089dRThyw744wzqhKvkE/K2VDqfhnBLUsu7rX+hIWtQG1fRrBYB6tanaoVK1Ywe/Zsdu/eDWQ/zqyrq2PFihUV2b6Tex/Wr19/SII/44wzWL9+/RC2yvrjwgsv5OGHH+66Bu4FF1zAd7/73aFuVkWccsopPP/8813z48eP57nnnqvItg+5jODi15NcHv7SuNQ3r86EX+k3r85vxFx//fU8vXETv3/6RK655pqKXQvayb0EnYm8Gj1pX2S5ui688EIeeuihrvmI4KGHHuLCCy+s+QTfmdjf+c538vGPf5wbb7yRxx57jFNOOaViCT7PDjR+gqML5ictm9THGtnJzuzs3qGf5vtS/LX+Brh4CRMuhg7g00/Cp588+DVf7mvdyb2bwU62vshydRUm9lLKa0lnYn/00Udpb2/n0Ucf5fzzz+exxx4b6qbVhKfmHpygq33Oa7Bf61VJ7pJmADcBdcDXImJxNeIUiVtzT8BwuFTbYI4zDpWGhgYWL17MokWLePHFF4e6ORXz2GOPFX3+KvFaOBxJ4sCBA12v9SOOOIKIqMi+HOzXesWTu6Q64CvABcBW4EeS7o+IjZWO1V0tHsy/21Te+16lfsJemBguueQS7r777q7yWtyfZyw/+ER34Uftf+AfOGHJCZzACUXrdu/JDTfFPlWOfNNbedPsv++a//Xtn2Tvr3/KhIWtXR0OD+GVLiI44ogj+NKXvsS0adMq+hoY7Nd6NXrubwc2R8QzAJJagJlAWcm9r2GSQoW951o5oIfLT9gjgvb2dr75zW/2+fF0OCtM0IP91dLu8bSksvF6HiN+vTf4li+OACYdVFbuGPFw+FQ5mDpPuEfEQT8mqtRxUvhaHoxjU1U4wC8BZkTEh9P8bOAdEfHRbvXmA/MBGhoamlpaWopub8GWBWW35R8n/GO/1xnseIWam5t7Xd7W1jag7cPhtT+h931aif3Zk46Ojqr/PfRgHC9DGa/QYOzPWozX3Ny8LiImF13YOZ5UqRtwCdk4e+f8bODm3tZpamqKcrS1tZW1XrnyGA+I7DB4PV5hWTUNxuMbPXp01+MpvI0ePbqqcfN4rDje8IsHrI0e8mo1/n5gGzC+YP7kVGbDmCS+8pWv1PSQTDG7d+9m9OjRB5WNHj2664cjZnlVjeT+I+A0SadKGgX8BXB/FeJYBUTBsFznydTu5bVu9+7dRARtbW1EhBO7HRYqntwjYh/wUeC7wCbgroh4utJxrHI6P8Z1Jr88JXazw1VVvuceEQ8CD1Zj22Zm1rea+8tfMzPrm5O7mVkOObmbmeWQk7uZWQ5V/BeqZTVC+g9gSxmrHg+8XOHmOJ7j1Vosxzt8402IiBOKLRgWyb1cktZGTz+9dTzHG8J4eX5sjlcb8TwsY2aWQ07uZmY5VOvJ/VbHc7xhGi/Pj83xaiBeTY+5m5lZcbXeczczsyKc3M3Mcqgmk7ukP5HUy/W/hh9J/bokvaSpklqr1JZl6YpZVuMkfV7SJyVdJ+kPh0F7yjpuJY2T9D+q0aZucf5K0lFlrHe5pDcVzH9N0uklrNfRx/LPlBi/pHqFai65SxoREfdHRHlXmx0iEfHOoW7D4UiZmjvO+ysiPhsR/1qt7Q/CfhwHVD25A38FFE3ukup6We9yoCu5R8SHI6Ks60J3U2rSzkdylzRH0npJT0q6PfU0/1nSD4C/S++iN6e6yyTdIulxSc+knsNtkjZJWjaANlwp6Yl0+5WkNkkdkq5P7XpcUkM/tteR7qdKapd0t6SfSrpD6fJHkmaksh8Df1qw7uclfbJgfoOkRkljJT2Q2rNB0n/rIfZB+zMVv1vSY2mfXZLq1UtaLenHkp6SNDOVlxSnW8zG9Bx8VdLTkh6SdKSks9K+Wy/pHklvlDRB0i8kHS/pCEn/Jqnsq5un2D+TtALoAH6ZjpOfp/39h5IeTTHfPoA490palx7f/FQ2L8X5YXrsncfpCZK+JelH6Xb+AOJek2I8ArwllXV9GpO0WNLGtI//fgBxCvfjBmCppLXp8V5bUK/ocdtPi4E3p9fbjT0ch3+QHtOYdEw+LWlSsY2ltne+vjal19tfkiXoNkltqV6HpH+Q9CRwnqTPpudng6RblbkEmAzckdp3ZHoNT07buCy1c4PU/bLoXe05UdL30/obJL1L0mLgyFR2R6pX7Jg6pF5Jerr+3lDdgN8Hfg4cn+aPBZYBrUBdKrucdF3WtKwFEDAT+E/gDLI3rnXAWQNsz0jg34A/Jrv+5h+n8r8D/mc/ttOR7qcCO8kuP3gE8P+AKcAY4HngtPRY7gJa0zqfBz5ZsK0NQCPwZ8BXC8qP6cf+/GaKfzqwOS0bAbwhTR8PbE5t6TNOkbiNwL7O/Z8ezweB9cB7Utl1wP9O0x9Obboa+JcBPmeNwAHg3IJ2FB4TtxUcL/cOIM6x6f7I9JycBDyb9nHncdN5nN4JTEnTpwCbyozZBDxF1vt8Q3qOPpme00uA44Cf8fo34cZVYj92e7x1QDvwtt6O2zJibejtOEzzXwD+HvgK8Ok+thfA+Wn+trSfniW9FlJ5AJd2f07T9O28/npvByYXLGsnS/hvAp4DTkjtXgPMKvK6/wRwTcH+O7pweS/H1HHF6pVyG44992nANyPiZYCI+G0q/2ZE7O9hnW9HtgeeAl6MiKci4gDwNNmTPBA3AWsi4tvAa2RvMpAliXK3/cOI2Jra+ETazluBX0XEL9Jj+XoJ23kKuEDSEknvioidRer0tD/vjYgDkX207PwEIuCLktYD/0qWrBpKjFPMryLiiTS9DngzWbL5XipbDrw7tetrZMnqSrIX4UBtiYjHC9pReEysLjheGgcQ4y9Tj+9xsusGzwa+FxG/jYi9ZG9Wnf4QuFnSE2SXnXyDpP5f7h7eBdwTEa9ExH9y6CUsdwK7yXrZfwq8UkaMQoX78dLUO/8JWafhdMo7bvvS03EIWYfgArLE+nd9bOf5iHg0TX+drBPV3X7gWwXzzZJ+IOkpstfO7/cR4w+A9oj4j8iuQncH6Zju5kfAFZI+D5wREb/rYXvdj6nT+ojfo+GY3Huyq5dle9L9gYLpzvmyrzYl6XJgAtD5EXRvOoAhOyjK3XZhG0vZzj4Ofq7GAETEz4FzyJLUFyR9tsw2dF4V+wNkPZCmiDgLeBEYM4A43R/nuJ4qKjvJdXKaLSfpdVd4vHQ/JgqPl7KeQ0lTyRL2eRFxJlnC+2kvqxxB1gM+K91OioheT7aVIyWYtwN3AxcDqwa4yV0Akk4le9OdHhFvAx4gHYdVUPQ4TMuOIzs+ji4hfvcf8RT7Uc/uzk6jpDHAPwGXRMQZwFdLiFGSiPg+WdLfBiyTNKd7nR6OqbLjD8fkvgb4c0nHAUg6digaIamJ7GD+YOrxVdtPgUZJb07zlxUse5YsuSLpHODUNP0m4JWI+DpwQ2edbvqzP48BXoqIvZKayd7YSo1Tip3AdknvSvOzgc5e/BKyXs9nyV5Uw90xwPaIeEXSW8mGgMYC71F2HmEE2XBWp4eABZ0zks4qM+73gVlp3PdosuHCLunTwDGRXery48CZZcbp7g1kiX6nsnNNF6Xy3o7b/vgdWcKGHo7D5F+AvyE7VoqObxc4RdJ5afr9wCPd4nTXmUhfTvux8BtlPa33Q7Ln/HhlJ2Qv4/VjuoukCWSjCl8Fvsbrr6G9kkam6WLHFEXqlaQq11AdiIh4WtL1wPck7Sd79xoKHyUbO21Tdr5zbTWDRcTudALlAUmvkI3Xdh5M3wLmSHoa+AHZGDpk48g3SDoA7AU+UmS7/dmfdwDfTh9J1/J6T7TPOP0wF/jn1FN/huyj6nvIPt6eHxH7Jf2ZpCsi4v8MIE61rQKulLSJbIz7cbJe2RfJXvC/Jdt/nUNYfwl8JQ01jCBL0lf2N2hE/FjSN4AngZfIPu4XOhq4L/VCBfx1f2P0EPdJSZ2fTp4HHk3lvR23/dn+b5Sd5N5A9pje2v04TL3dvRFxZ0qkj0maFhFretjsz4CrJN0GbARuIRtaXSXp1xHR3K0NOyR9lWys+985eN8uIztuXwXOK1jnBWVfy24j298PRMR9RdoyFbha0l6yk/ydPfdbgfVpuOtDHHpM0b1eRHygh8d7EP/9gFkFSaqPiI7Uc78HuC0i7hnqdh1uJDWSndgt+m2aw8FwHJYxq2WfTydNNwC/Au4d4vbYYco9dzOzHHLP3cwsh5zczcxyyMndzCyHnNzNzHLIyd3MLIf+P7gzH1knYaC+AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bgWI1CFYgmdS",
        "colab_type": "text"
      },
      "source": [
        "Data dispersion is extremely different among different features. Thus, scaling/standardization is required."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Yezu8Dng13Dp",
        "colab_type": "text"
      },
      "source": [
        "### Data Standardization"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0hEglTmGflZG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "fa69aaac-0f8f-4ca1-8f61-6752e26cab50"
      },
      "source": [
        "# Use standard scaling from sklearn library\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "scaler = StandardScaler()\n",
        "scaler.fit(df_x)"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "StandardScaler(copy=True, with_mean=True, with_std=True)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4AP375gGg_44",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "df_x_scaled = scaler.transform(df_x)\n",
        "df_x_scaled = pd.DataFrame(df_x_scaled, columns=df_x.columns)"
      ],
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-d0KaBibhKpM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "3861b30a-bc19-453f-e3e6-65af54c05f0f"
      },
      "source": [
        "# Show boxplot for scaled predictors\n",
        "df_x_scaled.boxplot()"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fc80a4e8518>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l9Lsh0Wj186C",
        "colab_type": "text"
      },
      "source": [
        "### Regression Modeling"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eH9X1xlSihV9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Run the predictive model to see whether accuracy has improved.\n",
        "X_train, X_test, Y_train, Y_test = train_test_split(df_x_scaled, df_y, test_size = 0.33, random_state = 5)"
      ],
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "90GHM-cCisEO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "reg = linear_model.LinearRegression()\n",
        "reg.fit(X_train,Y_train)\n",
        "Y_predict = reg.predict(X_test)"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7sfbrg3DivVm",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "a0eedd08-a196-440e-eba8-b13a8ff78d52"
      },
      "source": [
        "plt.scatter(Y_test, Y_predict, color='blue')\n",
        "plt.xlabel(\"Prices\")\n",
        "plt.ylabel(\"Predicted prices\")\n",
        "plt.title(\"Prices vs Predicted prices\")\n",
        "plt.show()"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "23VX1P2xivnj",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "5f431f87-4f78-4d23-f49d-d6df16c9724d"
      },
      "source": [
        "error_2 = np.sqrt(mean_squared_error(Y_test, Y_predict))\n",
        "print('Root Mean Squared Error: ' + str(error_2))"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Root Mean Squared Error: 5.309471828719491\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "F8Iqoe9CjR3_",
        "colab_type": "text"
      },
      "source": [
        "Attempt 1 (original data) RMSE: 5.309471828719486    \n",
        "Attempt 2 (with scaling) RMSE: 5.309471828719491  \n",
        "We do not see any difference in scaling in this case."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IM0MLFfK2AVA",
        "colab_type": "text"
      },
      "source": [
        "### Correlation Analaysis"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "GUMNFzWeX5Om",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 601
        },
        "outputId": "6e37df71-0f76-47dc-cb23-aac028897241"
      },
      "source": [
        "# Correlation plot\n",
        "plt.figure(figsize=(10, 10))\n",
        "corr = df_x_scaled.corr()\n",
        "sns.heatmap(corr, annot = True,\n",
        "    vmin=-1, vmax=1, center=0,\n",
        "    cmap=sns.diverging_palette(20, 220, n=200),\n",
        "    square=True)"
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fc80a4dc8d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cuEC-n542GTK",
        "colab_type": "text"
      },
      "source": [
        "## Step 3: Neural Networks"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ulJPc0xaTcU4",
        "colab_type": "text"
      },
      "source": [
        "Model an advanced predictive model using standard neural network."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tUykC1ynjrm4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.neural_network import MLPRegressor"
      ],
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FETgZhpPkDjS",
        "colab_type": "text"
      },
      "source": [
        "Method API: https://scikit-learn.org/stable/modules/generated/sklearn.neural_network.MLPRegressor.html"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vQAgbKxUkFui",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "mlp = MLPRegressor(hidden_layer_sizes=(60), max_iter=1000)"
      ],
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KCYJ-EzQT-vO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 241
        },
        "outputId": "0512cb5f-46ea-4798-e654-939658d5da66"
      },
      "source": [
        "mlp.fit(X_train, Y_train)"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:1342: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/neural_network/_multilayer_perceptron.py:571: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (1000) reached and the optimization hasn't converged yet.\n",
            "  % self.max_iter, ConvergenceWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "MLPRegressor(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,\n",
              "             beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
              "             hidden_layer_sizes=60, learning_rate='constant',\n",
              "             learning_rate_init=0.001, max_fun=15000, max_iter=1000,\n",
              "             momentum=0.9, n_iter_no_change=10, nesterovs_momentum=True,\n",
              "             power_t=0.5, random_state=None, shuffle=True, solver='adam',\n",
              "             tol=0.0001, validation_fraction=0.1, verbose=False,\n",
              "             warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kiDm74PIklhz",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "Y_predict = mlp.predict(X_test)"
      ],
      "execution_count": 47,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NCapva4mlKdt",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "d76adf57-ad95-4b51-aa11-cafc73505329"
      },
      "source": [
        "# Plot the predicted values\n",
        "plt.scatter(Y_test, Y_predict, color='blue')\n",
        "plt.xlabel(\"Prices\")\n",
        "plt.ylabel(\"Predicted prices\")\n",
        "plt.title(\"Prices vs Predicted prices\")\n",
        "plt.show()"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEWCAYAAABhffzLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de7hddX3n8fcnhxMhIZ2Qk8igkBMtTH3QEYRUFGwHAasiih2V6gQaFUsRn5qOWm+xaluZp7ZeSp16iaJmzGmV4rVKBURar5UmChWl80AFgsg1kJFw0ZB854+1Nmdnn7XWXvuy9m19Xs+znuy99mX99oLz3Wt/f7/f96eIwMzM6mPRsBtgZmaD5cBvZlYzDvxmZjXjwG9mVjMO/GZmNePAb2ZWMw781jNJuyQ9ftjtGBWSbpJ0Snr7rZI+NoBjnijpp318v3WSLuvX+9loceC3BdLA9WAa0O+Q9ElJB+Y9PyIOjIifDLKNvZC0RlKkn29X+nnfXMWxIuJ/RcSrSrTpk5LeVUUbuhERcxHxW8Nuh1XDgd/yPD8iDgSOAdYCb2t9gqT9Bt6q/lqefsaXAW+X9JzWJ0zAZ+xYHT9z3TjwW6GIuBX4R+BJAOmV8mskXQ9c37Tv8PT2AZLeK+lmSf9P0rckHZA+9jRJ35G0U9I1kk5sHEfSyyX9RNJ9km6UtK61LZIek/4SWdG07ymS7pY0LelwSf+cHvduSZ8p+Rm/C/wIeFIjZSLpTZJuBz4haZGkN0v6D0k7JF3U0oaz0s+7Q9LGlja/U9KWpvvPaDoHt6Sf+xxgHfDG9BfIPzR93s9Kuis9J69tep8D0l8J90r6MfDrRZ8x/W/02vQc3y3pLyUtajr335b0fkk7gHem+77V9PonSrpc0j3pr8C3pvtzz42k/SVtSffvlPSvkg4u89/EquXAb4UkHQacCvygafcLgeOAIzNe8h7gWOB4YAXwRmCvpMcCXwHele5/A/BZSaskLQX+GnhuRCxLX3t16xtHxM+A7wIvatr9P4CLI2I38GfAZcBBwKHAB0p8Pkk6AXhi02f8z2kbZ4FzgD9IP/N/Ax4D3Av8Tfr6I4EPAWelj82kx8461izJl+gHgFXA0cDVEbEJmAP+Ik2bPT8Nyv8AXAM8FjgZ+ENJz07f7h3Ar6bbs4H17T4r8Nskv96OAU4HXtn02HHAT4CDgfNb2r0M+Brw1fQzHg5ckT6ce27SNv0n4LD0vJwLPFiinVa1iPDmbZ8NuAnYBewEbgY+CByQPhbASS3PD5JgsIjkD/uojPd8E/Cpln2XkgSHpemxXtQ4TkHbXgV8Pb0t4BbgN9P7/wfYBBza5j3WpG3eSRKorgNemz52IvBLYP+m518HnNx0/xBgN7Af8Hbg002PLU1ff0p6/53AlvT2W4DP57Tpk8C7mu4fB2xvec5bgE+kt38CPKfpsXOAnxZ85mh5/nnAFentl2cc6+XAt9LbLwN+kPO+RefmlcB3gCcP+/9pb/tuzuVZnhdGxNdyHrslZ/9KYH/gPzIemwVeIun5TfumgSsj4n5Jv0PyK+BCSd8GXh8R/57xPp8FPiDpEOC/AHuBb6aPvZHkqv8qSfcC742Ij+d/RFZGxMMZ+++KiIda2v55SXub9u0huTp+DE3nI/0sO3KOdxjZ5ybLLPAYSTub9k0x/1n3OS7JF3Q7rc9/TM5jrYraXXRuPpW+9tOSlgNbgI2R/DqzIXKqx7qRV9L1buAhkvRDq1tIrviXN21LI+LPASLi0oh4FskV478DH808cMS9JOmc3yFJ83w60kvNiLg9In4vIh4D/D7wwUbfQ4+f7xaSNFRz2/ePpP/jNpLgBoCkJSRpjSy3kH1u8o55Y8sxl0XEqenj+xwXWF3ic7U+/2cFx29tS95w3dxzExG7I+JPIuJIkvTdacDvlminVcyB3/omIvYCHwfel3ZMTkl6uqRHkVztPV/Ss9P9+6cdqYdKOljS6Wmu/xckaaa9BYf6W5IA8uL0NgCSXiKpkV+/lySYFb1PWR8Gzk9z9KT9Eqenj10MnJZ22i4G/pT8v6s54BRJZ0jaT9KMpKPTx+5g3+B6FXBf2sl8QHrOniSp0Yl7EfAWSQeln/kPSnyOP0qffxiwASjV+Q18GThE0h9KepSkZZKOSx/LPTeSninpv0qaAn5OkgLqx38P65EDv/XbG4AfAv8K3AO8G1gUEbeQdCi+FbiL5Erxj0j+H1wEvI7kCvQeko7CVxcc40vAEcDtEXFN0/5fB74naVf6nA3Rn/kFF6Tvd5mk+4B/IcnBExE/Al5D8gV0G8kXTuZEqojYTtJR/nqSz3k1cFT68IXAkenoly9ExB6SK+SjgRtJfk19jKSzFOBPSNI1N5L8AvpUic/xRWBbetyvpMdsKyLuA54FPB+4nWQ01zPTh3PPDUkn+cUkQf864J9LttMqpvRXsplNMEkBHBERNwy7LTZ8vuI3M6sZB34zs5pxqsfMrGZ8xW9mVjNjMYFr5cqVsWbNmmE3w8xsrGzbtu3uiFjVur/SwC/pJuA+kpl8D0fE2rSA02dIps3fBJyRTsrJtWbNGrZu3VplU83MJo6kzBndg0j1PDMijo6Iten9N5PUCDmCpNBTJXXQzcws2zBy/KcDm9Pbm0kq+5mZ2YBUHfiDZEbftrTmOMDBEXFbevt2kmJOZmY2IFV37j4jIm6V9Gjgckn7VFuMiEhnFC6QflGcA7B6dZn6U2ZmVkalV/xp9UIi4k7g88BTgTvSkrqk/96Z89pNEbE2ItauWrWgU9rMzLpUWeCXtDRduYe06uJvAdeSFHRqrBa0nqRwlJlZbczNwZo1sGhR8u/c3GCPX2Wq52CSBRoax/nbiPiqpH8FLpJ0Nkl1wTMqbIOZ2UiZm4NzzoEHHkju33xzch9g3YKVpqsxFiUb1q5dGx7Hb2aTYM2aJNi3mp2Fm27q77EkbWsaSv8Il2wwMxug7dvb7686FeTAb2Y2QHmDFBv7G6mgm2+GiPlUUD+DvwO/mdkAnX8+LFmy774lS5L9ABs3zuf/Gx54INnfLw78ZmYDtG4dbNqU5PSl5N9Nm+Y7dsukgno1FtU5zcwmybp1+SN4Vq/O7vzt5zxWX/GbmY2QdqmgfnDgNzMbIevWwfr1MDWV3J+aSu73c4y/A7+Z2QiZm4PNm2HPnuT+nj3JfY/qMTObUB7VY2ZWM4MY1ePAb2a1N+yiac3aTfDqBwd+M6u1QcyU7YRH9ZiZVWwQOfVOtJvg1Q8O/GZWa4PIqY8aB34zq7VB5NQ74SJtZmYVG0ROvRMezmlmVrFB5NQ74SJtZmYDUFQ0bdBcpM3MrE9Gaax+EQ/nNDPrg1Ebq19kEKknL7ZuZhNvkAucjxIvtm5mfTUuqROo51j9Ig78ZtaxcUqdwOiN1W+n6i9VB34z69iolTloZ9TG6hfxBC4zG0njljoZtbH6RQbxpepx/GbWsUGMNe+3URqrX8T1+M1sJI1T6mTcuB6/mY2kcUqdjJtBfKk61WNmXRmX1Mm4aZzTjRuT9M7q1UnQdz1+MzPrmq/4zcxGSGM4Z2NkT2M4J/Tvqt9X/GZmI8T1+M3MasbDOc3MamYihnNKmpL0A0lfTu8/TtL3JN0g6TOSFlfdBjOzcTEp9fg3ANc13X838P6IOBy4Fzh7AG0wMxsLg5gjUWngl3Qo8DzgY+l9AScBF6dP2Qy8sMo2mJmNm3XrknUC9u5N/u33fImqr/j/CngjsDe9PwPsjIiH0/s/BR5bcRvMzKxJZYFf0mnAnRGxrcvXnyNpq6Std911V59bZ2ZWX1Ve8Z8AvEDSTcCnSVI8FwDLJTUmjh0K3Jr14ojYFBFrI2LtqlWrKmymmVm9VBb4I+ItEXFoRKwBXgp8PSLWAVcCL06fth74YlVtMDOzhYYxjv9NwOsk3UCS879wCG0wMxtZVS+9OJBaPRHxT8A/pbd/Ajx1EMc1Mxs3rtVjZlYzrtVjZlYzrtVjZlYzE1Grx8zMypuUWj1mZl2reoTLqBn7Wj1mNnzjHDgbI1xuvhki5ke4jNNn6Ma41+oxsyHKC5znnTceXwaDGOFSR4qIYbehrbVr18bWrVuH3QyzsbNmTRLsW0nJF0HDkiX9Tyf0w6JF+7azQUquhq2YpG0RsbZ1v6/4zcZUmRRO3hDA1mA6qlfRgxjhUkcO/GZjqGzuu5MA2c9x4v0yiBEudeTAbzaGyua+swKnlP2eo3gVPYgRLnXkwG82RN2OuCk7u7M1cM7MLPwigOSxU0/tpOWDU/UIlzpy4Dcbkl6GKnaS+24Ezk99Ch58EO6/f+FzImDz5tEd3WP95cBvNiS9DFXsJveddbxujm3jz4HfbEh6KcbVnMIBmJqaD9x5V+1l3ncUO3jrqOpJdw78ZkPS61DFdevmr/z37En2FaWLyrzvKHbw1s0gZis78JsNST+GKnaSLso6Xi/Htmq4Hr/ZBOvHUMVO0kWtx1u6NEklQJIqWr/eI2ZGwUjU45f0EknL0ttvk/Q5Scf0rwlm9ZCVt+11qGKn6aLmET4R82UP9uzxqJ5RMSr1+P84Iu6T9AzgFJLF0T/UvyaYTb6q8rbdpotc/Gx0jUo9/rTbiOcBmyLiK8Di/jXBbPJVFWi7TRcNIp1g3RnEbOW21TklfRm4FXgWcAzwIHBVRBzVv2YUc3VOG3ejVmUyr2rn7GySCrLJ0Et1zjOAS4FnR8ROYAXwR31un9lE6zZvW9V47rx0wqmnjkedfutN28AfEQ8AdwLPSHc9DFxfZaPMJk03edsqx3NnpRPWr086eOu22lUdlRnV8w7gTcBb0l3TwJYqG2U2abrJ25btF+j2V0HriKJLLnGHb12UyfFfDTwF+H5EPCXd928R8eQBtA9wjt/qqUy/QONXQXPA7nY1rVHrh7De9ZLj/2Uk3w6RvtHSfjfOzBYq0y/Qz9FCXu2qPsoE/oskfQRYLun3gK8BH622WWb11Jy22bULpqf3fby1X6CfwzK92lV9lOncfQ9wMfBZ4NeAt0fEB6pumFndtHbm7tgxv3hKXr9AP6/SvdpVfZTJ8T8OuC0iHkrvHwAcHBE3Vd+8hHP8VgfdjK3vZ47fJk8vOf6/B5q7dvak+8ysj7pJ2/gq3bpRJvDvFxG/bNxJb7tkg1mfdZu26aTQW9ULfNh4KBP475L0gsYdSacDd1fXJLN6qrpzdRALfNh4KBP4zwXeKmm7pFtIJnP9frXNMqufqtM2rshpDW07dx95onQgQETsqrRFGdy5a9Y7T9Cqn7zO3f0KXnBmRGyR9LqW/QBExPvaHHB/4BvAo9LjXBwR70hHCX0amAG2AWc19yGYWTVWr84eNeQJWvVTlOppzNBdlrO18wvgpLR889HAcyQ9DXg38P6IOBy4Fzi7y7abWQc8QWt8VN4JHxG5GzAF/M+i55TZgCXA94HjSDqG90v3Px24tN3rjz322DAbpi1bImZnI6Tk3y1bht2i7kzK55hkW7ZELFkSkSTmkm3Jku7+WwFbIyOmlpnAdVVEPLWbLxVJUyTpnMOBvwH+EviXSK72kXQY8I8R8aSM154DnAOwevXqY2/O+o1qNgCeJGWD1M9FcnqZwPVtSf9b0m9IOqaxlTloROyJiKOBQ4GnAk8o2+CI2BQRayNi7apVq8q+zKzv8kbDbNjgMfHWf4NYFrNM4D8aeCLwp8B70+09nRwkkpW7riRJ7SyX1OhUPpRkWUezvqgiN5r3B7djR/kx8Z44ZWUNpEpqVv6nHxuwClie3j4A+CZwGkm5h5em+z8MnNfuvZzjt1ZZuep+5kabzc7u+55F2+xsdluraJdNpkHk+MsE8Bngr0k6Z7cBFwAzJV73ZOAHwL8B15JU9QR4PHAVcEP6JfCodu/lwG/N8v4wZmbKB+Nej5e3SQtfn/fF0Wu7bHL1qxM+L/CX6dy9nGQ8fmO5xXXAiRFxSm+/NcrzBC5rltf5lafdBKW5uSSPv3178nP6/PMXdtq2PmfXriTV0yqrAy5v4hTAli3uILbq9NK5e0hE/FlE3Jhu7wIO7n8TzcrptJOrKDdatn5NayG0Cy4oPya+6PiulWPDUCbwXybppZIWpdsZwKVVN8wsy9xccgWdZWam8wlK3davaa2rMzMDBxwAZ521sPM2a+JUJ8cy67cyqZ77SGbx7kl3TQH3p7cjIn6luuYlnOoxyB5P3zA9DZ/4RHK7XdqmWT/q15QZ5z83B2eemf1618qxqnSd6omIZRGxKCKm021Rum/ZIIK+WUPW1XlDWkKKdeuSYL96dRL8N24sTqX0Y+hcmV8N69Ylvw56PZZZP5RJ9ZiNhKLc/i9/OR/kO6k534/6NWUn3LhWjo0KB34bG+2ujBtX+FlX32eemT1xqh818Mv+avAyiTYqHPhtpBTNcC3qJAVYsaJ4mGeZETvnn598eXQyw7aTK/lOlkk0q0zW4P60w3dF0Zb3uio2T+CqhzIzFrdsyZ6otXhxxPR097Nryx6/qO2uemmjhk4ncEm6EQhAwGqS2vkClgPbI+Jx1X4lzfOonnropCph2QlVWfJG0fSzKqLZKOh4VE9EPC4iHg98DXh+RKyMiBmSejuXVddUq6u8NE3W/taUyT33lD9OXk5+EFURzUZBmRz/0yLiksadiPhH4PjqmmR1NTXV2f5mZYdEFo2iqbIqoqtz2igpE/h/Jultktak20bgZ1U3zOpnz57O9jdr1/ELyRdI0SiaqoZbdjrE1KxqZQL/y0hKLH8e+Fx6+2VVNsrqKW+CU97+hka+/4EHin8d7N2bXXytcSW+cSOsX9//4ZbdloUwq0rbkg2PPFFaGhH3t39m/7lztx6ySh9IyVXy7Gx+1cy817Rq7aQ97zz48If3fW4VSyr2oyyEWTe6Ltkg6XhJPwauS+8fJemDFbTRaq55ghPsG8Dz0iNZV9MR8yUcGlpTNnNzC4M+VHMlPpAVlcw6UCbV837g2cAOgIi4BvjNKhtl9dUYrTM7Wy4o540EavxKyEvZbNyYXyO/36N4XKrBRk2pmbsRcUvLrhLdbTbpqhypUmZoZ7vjnX9+/gzZouDe7ytxl2qwUVMm8N8i6XggJE1LegNp2sfqq58jVbK+QPJq7jc6bxvHL1LUnrzgLlVzJe5SDTZKytTjX0myzu4pJDN3LwNeGxEdTJnpjTt3R0+/Zrlmdc4uXpxU28wTUX75xbz2nHcefOhDC/effDJ87Wvt39dsHOR17u5X4rW/FhH7XJ9IOgH4dr8aZ+OnX7Ncszpni4L+1FTyZVH2OHnPu+SS7P033FDufc3GWZlUzwdK7rMa6ddIlU6/KPbsSX4htJus1a49Ls9gdZYb+CU9XdLrgVWSXte0vZNk+UWrsXYjVdp1/DYeLzmNZB8PPAAPPtj+ef0sz+CSCzZJiq74FwMHkqSDljVtPwdeXH3TbJQVjVRp1/Hb/Hi3iiY+lRk508kQS5dcsElTpnN3NiJ6+BPtnTt3x0tex+vMDNx9d/mO2W500rncWto5b2F2l2u2cdX1zF3gY5KWN73RQZIu7WvrbKLk5cl37OisY7addrNz2yk7xNL9ATZpygT+lRGxs3EnIu4FHl1dk2zcFXXwbtiQP0a/U+1m5/aLSy7YpCnzJ7hX0iP/i0uaJVmZyyZAFZ2Wp56a/9iOHeXKLJcxNdU+TdMPLrlgk6bMOP6NwLck/TPJBK7fANrMmbRx0Dp5qtFpCb0F0bwx8kX23x8eeqiz1zS+QPrV7jyN9yzTH2A2DkqVZU5n7z4tvfsvEXF3pa1q4c7danTaaVm2MzSvDHGR6ekkZVM0eav5/bNG9biz1WxfHXfuSnpC+u8xJIut/yzdVqf7bMx10mnZyZDGojo4eXbvhmXLyi2zmDeU052tZuUU5fhfn/773oztPRW3y/osK5ffSadl3ipSGzYsfN+8nPi55yZX9nka+f+iL4giixZ5gpVZKREx8tuxxx4b1r0tWyKWLIlIrtWTbcmSiFe/Onv/li0LX9/8nKJNSt53y5aI2dnk/uzs/HvOzJR/n7LHzNqyPodZ3QBbIyOm5gZb4L8XbXmvq2Jz4M8PpGXMzmYHx8b7FL1v1pdGJ9vMzL7v2clrp6aK37fR7rznzc72ds7Nxl1e4M/t3JX0ifTmo4Hjga+n958JfCciTqvoR8gCde/czSpd3MnasL2s+bpyZZKC6cXixfDxjye3zzqru/o8rbZsmf/sXtPWLFvHnbsR8YqIeAUwDRwZES+KiBcBT0z3tTvgYZKulPRjST+StCHdv0LS5ZKuT/89qPuPVQ95+fWya8N2OwFpbq73oA/JSJ0NG5Ktk6Cf19E7M7PvF54nWJl1pswErsMi4ram+3eQjPJp52Hg9RFxJMlQ0NdIOhJ4M3BFRBwBXJHetwK9lgzodgJSPxcd37Gj+Eskq31Z5ZeXLIELLpi/PzcHu3Zlv58nWJllKxP4r5B0qaSXS3o58BWg7RpFEXFbRHw/vX0fyXKNjwVOBzanT9sMvLCbhtdJr1e0WZU0169PAnvRKJhBDY9slFtoLb/wwQ8Wr1XbSIG1fqHMzHhNW7NCWYn/1g34beD96fbbZV7T8vo1wHbgV4CdTfvVfL/lNecAW4Gtq1evrqLfY2zkjcrpdtRK2ffL6xTu91Y0CqhIUae1meV37pYN3LPAKentJcCyMq9Ln38gsI10JFBroAfubfceHtXT26ieVmUDZq8jespu09MRixd3/sWWN+RT6v7cmE2SvMDfNtUj6feAi4GPpLseC3yhzK8JSdPAZ4G5iPhcuvsOSYekjx8C3FnmveqmdcIVlCshnPf65lRO2T6D1hRRVXbvXliqoUzntTt1zbpTJsf/GuAEkpW3iIjrKVGWWZKAC4HrIuJ9TQ99CVif3l4PfLGTBtdBrys+tXt9JwGzuWb97GxXH6dr7foYXDXTrDtlAv8vIuKR6zFJ+0GpsswnAGcBJ0m6Ot1OBf4ceJak64FT0vvWJG/45plnlitH0G74Z17Z5Ob9Wb8YsgJtO2Vq7+RZsaK4ZHTR8o9mlq/M0ot/AewEfhf4A+A84McR0cfBfsXqNoGrXVql3eStdhOa2i2NWDRhDJIRQWVq6ktw0knw3e8u/CJqtXjxvumexYuTz7B798I2OLCbldPL0otvAu4Cfgj8PnAJ8Lb+Ns+atbtKbpf/zkvlNK6g89a73bEDzjuv+BfDunWweXMSmFsdeeS+X1oRSdBfv744TTQzk8zsbb5yX7Zs36Df3AYz601h4Jc0RZKj/2hEvCQiXpze7sOke8tT5mp6+/bkynzlyiRYSsntvJTM4sVw773tFzn/8Ifzn9Occ2/9P2B6Gu64Y+H+Bx5IFma56aakzELehKzW9W/vuad9G8ysO4WBPyL2AP+3eelFq16ZTtQVK+AVr9h38tKOHfDKVya3W3Pf09Pl6tZE5P/iaPyS2Lhx4dX47t35M3MbwbqTnLxH7JhVp0yq5yDgR5KukPSlxlZ1w+qktSP11FOLO1Ebj7UGX0jy5I2UTPMV9P33l2/Pnj0Ljy/Nd/52etXdHKxb25WXr/eIHbPqlFlz948rb0WNzc0lV+mNjs2bb4YLL4Szz05SJNu3J1f3kKQ/GksennVW/nv2mg6ZnU2C/Ic+NL8vImnXCSckbchKBy1dmjyvtVO4m2DtdW7NqlNUlnl/4FzgcJKO3Qsj4uEBtu0RkzyqJ6/scWOETZ6iTtqstWfLlldujJzZsCH/+TMzsHPnwr6I6Wl41avmv7AcrM2Gq5tRPZuBtSRB/7kkSy5anzTSO3nBtV2QPv/87GUMFy+ev8JuTiEB7Lffwue++tXzfQpTU/MjZ4qO31gisdXu3fMduWVnGJvZEGTVcUh/Bfyw6fZ+wPfznlv1Nmm1esrWwCnzPs1LGTavdpV1jOnp5Dmt9X76XZPHzEYDXdTqeaTrMIaU4ulFUa2aYcsaJ9+qTG2cdeuSoZCNUTIHHlh8jN27k+e0Xo2XaU9Zi8oMFzCzoSr6Mz1K0s/T7T7gyY3bkn4+qAZ2o9daN/1sR9aXT5nO14j27S36nJ0s3tLPsfF7947Wl6yZLdS2ZMMo6LRzN6/jM6vTsypFZQ82bmw/kQqSq/hzz00WJMmS9zmnpmD58uw8fdY5KCrhcOCB2Y9J+csoDvI8m1m+Xko2jJ0yV7tVp4KKyh6ULXYWkQypXLmys/LKe/bAffct7PyVkiDe+j5ZHcXT00kaKautjRo8eSZldu0opwvNepKV+B+1rdPO3XYLjfR7Rass7RYJaV5YpbmDtszW6MRtt0LWzMz8c1rb0/x5t2xZuBDK4sXFx5idzW/3JKyANYj/R8yqRi8rcA176zTwt/ujHcSSfZ0eo9Pgv2RJsmRhu9E4ZdpS9HjRF9gkB0cv62iTIC/wT2Sqp11NmE46PrvVruRAaxrhoYc6e/9G8bNNm/Jr60jlOnqLirIV1cyZ5Hr4g/h/xGxosr4NRm3r9zj+QV3N5a2T269x881po7wr89nZ4s/b7rWTfFVfxFf8NgmoU6qnnWEEs+Yvgamp3oN+axAqel7R580LcI1UTmvbe13ofVzU9QvPJosDf4tBBrNur/CXLIk4+eSIRYuyH2tuc96XydRU8efNu9qH6s7HuKjjF55NlrzAP5Hj+EdNUUG1PIsWwQEHZJdTnpmZX7ykoWimb9F/4lGY82Bm1ajVOP5R002H4N69+TX0DzxwYQdq3uIt7RZ1cd17s/px4B+AvJEx7dbWzZP1RdJtAJ/kkTlmls2Bv0+KZnnmBeXNm7PXoW0n64uklwBedlUsM5sMZVbgsjZa6/I0iqVBEkTLrCbV/NiuXfn18Iuu4puPZWaWx527fdDvDtLzztt32cOGpUvhIx9xcDezcty524N2xbr6Pcvzkkuy969c6aBvZr1zqqeNdmkcSBZDz0rNNBZJ75TLBZhZlXzF30ZReeWqFNXHMTPrlQN/G2Wuvu+5J/s5efvbGdbYetefN6uHWgf+MoGuzNV3v6/QhzG2flSWqzSzAciq4zBqW1W1esoU4SrzvEko6OVqlGaThzrV4y+jbO6+zNX3JMx+dYeyWX3Udhz/okXZxcukZAZr3bhYm9nk8Tj+Fh45sy8XazOrj9oGfg0rXoUAAAdlSURBVAe6fU1CusrMyqks8Ev6uKQ7JV3btG+FpMslXZ/+e1BVx2/HgW4hF2szq4cqr/g/CTynZd+bgSsi4gjgivT+0DjQmVkdVRb4I+IbQOsUptOBzentzcALqzq+mZllG3SO/+CIuC29fTtwcN4TJZ0jaaukrXfddddgWmdmVgND69xNJxfkjiWNiE0RsTYi1q5ataqSNvRaoqDM610GwcxGzaAD/x2SDgFI/71zwMd/RK8lCrJef+aZSenkxnu4DIKZjaJBB/4vAevT2+uBLw74+I/otepm1ushKc/cCO7DqOxpZtZOZTN3Jf0dcCKwErgDeAfwBeAiYDVwM3BGRLStYTmKM3fzXt8wO5uUO/DsYDMbloHP3I2Il0XEIRExHRGHRsSFEbEjIk6OiCMi4pQyQb8qeTN0Fy0ql49vt8hKY/3cTo5tZjYInrnbYs+e/uTjGwuqe3awmY2a2gb+1pm7U1MLn1OUjy9aZKUR3D072MxGUW2rc7bqNOefV81yago2b3ZwN7Phc3XONjrNx+elcRz0zWzUOfCnOs3HO41jZuNqv2E3YFQ0AvbGjfMjchp5+qLXONCb2bjxFX+TTqp1uhSDmY0rX/F3oVGKoTErtzH0E/wLwMxGn6/4u+BSDGY2zhz4u7B9e2f7zcxGiQN/F1yKwczGmQN/F1yKwczGmQN/FzyG38zGmUf1dMlj+M1sXPmK38ysZhz4zcxqxoHfzKxmHPjNzGrGgd/MrGYc+M3MasaB38ysZhz4zcxqZmIDv+vlm5llm8iZu66Xb2aWbyKv+F0v38ws30QGftfLNzPLN5GB3/XyzczyTWTgd718M7N8Exn4XS/fzCzfRI7qAdfLNzPLM5FX/GZmls+B38ysZhz4zcxqxoHfzKxmHPjNzGpGETHsNrQl6S7g5mG3o8lK4O5hN2JE+dzk87nJ53OTrdfzMhsRq1p3jkXgHzWStkbE2mG3YxT53OTzucnnc5OtqvPiVI+ZWc048JuZ1YwDf3c2DbsBI8znJp/PTT6fm2yVnBfn+M3MasZX/GZmNePAb2ZWMw78bUj6uKQ7JV3btG+FpMslXZ/+e9Aw2zgMkg6TdKWkH0v6kaQN6X6fG2l/SVdJuiY9N3+S7n+cpO9JukHSZyQtHnZbh0XSlKQfSPpyet/nBpB0k6QfSrpa0tZ0X9//phz42/sk8JyWfW8GroiII4Ar0vt18zDw+og4Enga8BpJR+JzA/AL4KSIOAo4GniOpKcB7wbeHxGHA/cCZw+xjcO2Abiu6b7PzbxnRsTRTeP3+/435cDfRkR8A7inZffpwOb09mbghQNt1AiIiNsi4vvp7ftI/ogfi88NkdiV3p1OtwBOAi5O99fy3ABIOhR4HvCx9L7wuSnS978pB/7uHBwRt6W3bwcOHmZjhk3SGuApwPfwuQEeSWVcDdwJXA78B7AzIh5On/JTki/KOvor4I3A3vT+DD43DQFcJmmbpHPSfX3/m5rYFbgGJSJCUm3HxEo6EPgs8IcR8fPk4i1R53MTEXuAoyUtBz4PPGHITRoJkk4D7oyIbZJOHHZ7RtAzIuJWSY8GLpf0780P9utvylf83blD0iEA6b93Drk9QyFpmiToz0XE59LdPjdNImIncCXwdGC5pMbF1qHArUNr2PCcALxA0k3Ap0lSPBfgcwNARNya/nsnyQXDU6ngb8qBvztfAtant9cDXxxiW4YizcteCFwXEe9resjnRlqVXukj6QDgWSR9IFcCL06fVstzExFviYhDI2IN8FLg6xGxDp8bJC2VtKxxG/gt4Foq+JvyzN02JP0dcCJJedQ7gHcAXwAuAlaTlIs+IyJaO4AnmqRnAN8Efsh8rvatJHn+up+bJ5N0wk2RXFxdFBF/KunxJFe5K4AfAGdGxC+G19LhSlM9b4iI03xuID0Hn0/v7gf8bUScL2mGPv9NOfCbmdWMUz1mZjXjwG9mVjMO/GZmNePAb2ZWMw78ZmY148BvBkjak1ZEvFbS30takvO87wy6bWb95sBvlngwrYj4JOCXwLnNDzZmlUbE8cNonFk/OfCbLfRN4HBJJ0r6pqQvAT8GkNSouomkN6W106+R9Ofpvl+V9NW0yNY3JT0h3f+S9NfENZK+MYwPZdbgIm1mTdIr++cCX013HQM8KSJubHnec0nK5R4XEQ9IWpE+tAk4NyKul3Qc8EGSejRvB56dFuBaPojPYpbHgd8scUBaRhmSK/4LgeOBq1qDfuoU4BMR8QBARNyTVio9Hvj7piqlj0r//TbwSUkXAZ9rfTOzQXLgN0s8GBFHN+9Ig/f9HbzHIpK68ke3PhAR56a/AJ4HbJN0bETs6KXBZt1yjt+sO5cDr2iM/pG0IiJ+Dtwo6SXpPkk6Kr39qxHxvYh4O3AXcNiwGm7mwG/WhYj4Kkm53K1piugN6UPrgLMlXQP8iKQfAOAv047ga4HvANcMus1mDa7OaWZWM77iNzOrGQd+M7OaceA3M6sZB34zs5px4DczqxkHfjOzmnHgNzOrmf8Px0In3xm3OhAAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JGCh5QDqmfXl",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "3dd95ba5-404a-487b-dbe3-84a9c5a248bb"
      },
      "source": [
        "error_3 = np.sqrt(mean_squared_error(Y_test, Y_predict))\n",
        "print('Root Mean Squared Error: ' + str(error_3))"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Root Mean Squared Error: 3.5675317610112574\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hsAVZVmSnmNe",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "outputId": "9c07389d-774c-4cf5-bc7a-3588a842011d"
      },
      "source": [
        "models = ['LR1', 'LR2', 'ANN']\n",
        "y_pos = np.arange(len(models))\n",
        "performance = [error_1, error_2, error_3]\n",
        " \n",
        "plt.bar(y_pos, performance, align='center', alpha=0.5)\n",
        "plt.xticks(y_pos, models)\n",
        "plt.ylabel('RMSE (Error)')\n",
        "plt.title('Evaluation for house price prediction')\n",
        " \n",
        "plt.show()"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QAYeMKDZWOlm",
        "colab_type": "text"
      },
      "source": [
        "## Exercise  \n",
        "We can improve the accuracy by focusing on the data pre-processing and feature engineering. On approach is to transform skewed data (as you already exposed in Predictive Analytics).  \n",
        "Therefore, as an expercise you can attempt to visualize the data skewness of the dataset, and transform the data accordingly (e.g., log or square transformation.)  \n",
        "An example can be found from this medium article on [Transforming Skewed Data for Machine Learning](https://medium.com/@ODSC/transforming-skewed-data-for-machine-learning-90e6cc364b0).  \n",
        "Inition few steps are as follows:\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KUfyAK4EWSaG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 828
        },
        "outputId": "5a76aaa4-9c16-452d-b6ed-7a5bd10d9f53"
      },
      "source": [
        "# Visualize the data discribtion using pandas dataframe:\n",
        "df_x.hist(figsize=(10, 10))"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x7fc80a090ef0>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7fc80a048208>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7fc809ff9588>],\n",
              "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7fc809faa908>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7fc809f5cc88>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7fc809f8ffd0>],\n",
              "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7fc809f4b3c8>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7fc809eff710>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7fc809eff780>],\n",
              "       [<matplotlib.axes._subplots.AxesSubplot object at 0x7fc809e64e10>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7fc809e221d0>,\n",
              "        <matplotlib.axes._subplots.AxesSubplot object at 0x7fc809e52550>]],\n",
              "      dtype=object)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 51
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 12 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I9XGweG1Ydjh",
        "colab_type": "text"
      },
      "source": [
        "Calculate skewness for all the continous variables."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3n9Hlg24XS6P",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "outputId": "bd839942-cf61-4bad-f70b-f72e1cf73ac3"
      },
      "source": [
        "# determine index for continouse variables\n",
        "cts_vars = df_x.dtypes[df_x.dtypes != 'object'].index\n",
        "\n",
        "# Calculate the skewness and then sort\n",
        "skew_vars = df_x[cts_vars].skew().sort_values(ascending=False)\n",
        "print(skew_vars)"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "crim       5.223149\n",
            "chas       3.405904\n",
            "zn         2.225666\n",
            "dis        1.011781\n",
            "rad        1.004815\n",
            "lstat      0.906460\n",
            "nox        0.729308\n",
            "tax        0.669956\n",
            "rm         0.403612\n",
            "indus      0.295022\n",
            "age       -0.598963\n",
            "ptratio   -0.802325\n",
            "dtype: float64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "z9AthMjWZTI-",
        "colab_type": "text"
      },
      "source": [
        "We can address skewed variables by transforming them (i.e. applying the same function to each value). Common transformations include square root (sqrt(x)), logarithmic (log(x)), and reciprocal (1/x)."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UqWmNnY-ZDfl",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        },
        "outputId": "92c4fb0b-2c7b-40b2-bc9c-f1feb5a2524f"
      },
      "source": [
        "sns.distplot(df_x['crim'])"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fc80a390780>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z_aTvQpCZmOV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "crim_var = df_x['crim']"
      ],
      "execution_count": 54,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "u5QTgkImaKeH",
        "colab_type": "text"
      },
      "source": [
        "**Square root transformation.**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gMcz4PJnZyC6",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "sqrt_crim = crim_var ** (1/2)"
      ],
      "execution_count": 55,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qoYzHxedZ5n_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        },
        "outputId": "cef8811c-3a43-40f0-914d-95b65155548b"
      },
      "source": [
        "sns.distplot(sqrt_crim)"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fc80a383048>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 56
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7GkRhcD5aNvm",
        "colab_type": "text"
      },
      "source": [
        "**Log Transformation.**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fMWd7LbZZ9mV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "log_crim = np.log(crim_var)"
      ],
      "execution_count": 57,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xOocagOyaWsw",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 296
        },
        "outputId": "56792db0-ca02-427f-e9ef-017d088af33a"
      },
      "source": [
        "sns.distplot(log_crim)"
      ],
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fc80aede898>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 58
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "U-U74Sp1adTT",
        "colab_type": "text"
      },
      "source": [
        "The log transformation seems to have de-skewed, as the distribution of transformed crime is the most symmetrical. \n",
        "\n",
        "Therefore, we can construct the new variable (i.e., log_crim) instead of crim.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3ef0c7SoaZPN",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# create a new varaiable for log_crim\n",
        "df_x['log_crim'] = log_crim"
      ],
      "execution_count": 59,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ACWUDRMlbADT",
        "colab_type": "text"
      },
      "source": [
        "Important: Once you create the new variable, make sure delete the original variable when using the dataframe as a predictor. (Discuss why?)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vtiMXkSGa_y4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "del df_x['crim']"
      ],
      "execution_count": 60,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "B3CYc2ZQa55S",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "b0b6c10a-1c4a-4e55-a3a3-be8a29c482d4"
      },
      "source": [
        "df_x.keys()"
      ],
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['zn', 'indus', 'chas', 'nox', 'rm', 'age', 'dis', 'rad', 'tax',\n",
              "       'ptratio', 'lstat', 'log_crim'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 61
        }
      ]
    }
  ]
}