notebooks/workshop/1_An_Experimental_EDA_for_a_classification_task.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "1_An_Experimental_EDA_for_a_classification_task.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/claudio1975/SDS2020/blob/master/notebooks/workshop/1_An_Experimental_EDA_for_a_classification_task.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HlcrvwrX70Uv",
        "colab_type": "text"
      },
      "source": [
        "# **An Experimental Exploratory Data Analysis for a Classification Task step 1**\n",
        "\n",
        "### ***From Visualization to Statistical Analysis***\n",
        "\n",
        "### ***From Feature Engineering to Feature Selection***\n",
        "\n",
        "### ***From the Best Model Selection to Interpretability***\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vngHrhnzA7qr",
        "colab_type": "text"
      },
      "source": [
        "To start the exploration set up the environment with libraries, upload the data set (it's stored in a github repository) and split it into target variable and features variables. No more set up is required using Google Colab. Look at the guidelines: https://colab.research.google.com/notebooks/welcome.ipynb"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "E2o9mujoz5by",
        "colab_type": "text"
      },
      "source": [
        "#### **Contents**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LRwDUZ019MxF",
        "colab_type": "text"
      },
      "source": [
        "The goal of this challenge, launched by CrowdAnalytix, is to develop a model to predict whether a mortgage will be funded or not based on certain factors in a customer’s application data. \n",
        "The evaluation metric used is the F1 score.\n",
        "The data set is made up by 45.642 observations with predictor variables (21 features) and the target variable. It's a classification task with the goal to predict the 'Result' target variable for every row (Funded, Not Funded). Look at the competition: https://www.crowdanalytix.com/contests/propensity-to-fund-mortgages\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z9DkyQZ2Ls-4",
        "colab_type": "text"
      },
      "source": [
        "### **Exploratory Data Analysis (EDA) Pipeline**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZVQahUb-8995",
        "colab_type": "text"
      },
      "source": [
        "![](http://www.theleader.info/wp-content/uploads/2017/07/Mortgage-rates.jpg)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hs9lEhRM0Mne",
        "colab_type": "text"
      },
      "source": [
        "# Prepare Workspace"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zJlI9RQG-A8w",
        "colab_type": "text"
      },
      "source": [
        "#####- Upload libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XQysnLU9RUKI",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "86161e01-6fd1-4486-f75a-398bccfa7fe9"
      },
      "source": [
        "# Upload libraries\n",
        "\n",
        "# to handle data set\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "\n",
        "# to plot\n",
        "%matplotlib inline\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from sklearn.metrics import plot_confusion_matrix\n",
        "\n",
        "# statistics\n",
        "import statistics\n",
        "import scipy.stats as stats\n",
        "import statsmodels.api as sm\n",
        "from statsmodels.formula.api import logit\n",
        "from scipy.stats import chi2_contingency\n",
        "from scipy.stats import kurtosis \n",
        "from scipy.stats import skew\n",
        "from statistics import stdev \n",
        "\n",
        "# to split data set \n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# standardization\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "# to build models\n",
        "from sklearn.model_selection import StratifiedKFold\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.ensemble import BaggingClassifier\n",
        "from sklearn.ensemble import AdaBoostClassifier\n",
        "from sklearn.ensemble import GradientBoostingClassifier\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.neural_network import MLPClassifier\n",
        "\n",
        "# to evaluate models\n",
        "from sklearn.metrics import f1_score\n",
        "\n",
        "# to handle imbalanced data set\n",
        "from imblearn.over_sampling import RandomOverSampler\n",
        "from collections import Counter\n",
        "\n",
        "# feature engineering\n",
        "!pip install feature-engine\n",
        "import feature_engine\n",
        "from sklearn.preprocessing import KBinsDiscretizer\n",
        "\n",
        "# feature importance \n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "!pip install eli5 \n",
        "import eli5\n",
        "from eli5.sklearn import PermutationImportance\n",
        "!pip install shap\n",
        "import shap\n",
        "import eli5\n",
        "from eli5.sklearn import PermutationImportance\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/externals/six.py:31: FutureWarning: The module is deprecated in version 0.21 and will be removed in version 0.23 since we've dropped support for Python 2.7. Please rely on the official version of six (https://pypi.org/project/six/).\n",
            "  \"(https://pypi.org/project/six/).\", FutureWarning)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:144: FutureWarning: The sklearn.neighbors.base module is  deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.neighbors. Anything that cannot be imported from sklearn.neighbors is now part of the private API.\n",
            "  warnings.warn(message, FutureWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Collecting feature-engine\n",
            "  Downloading https://files.pythonhosted.org/packages/57/cb/afbd5caa894ced21b8990b2b4bf7f6aea6ca392e787d8b37a542cb13a4e0/feature_engine-0.4.31-py2.py3-none-any.whl\n",
            "Requirement already satisfied: pandas>=1.0.3 in /usr/local/lib/python3.6/dist-packages (from feature-engine) (1.0.4)\n",
            "Requirement already satisfied: numpy>=1.18.2 in /usr/local/lib/python3.6/dist-packages (from feature-engine) (1.18.5)\n",
            "Collecting statsmodels>=0.11.1\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/cb/83/540fd83238a18abe6c2d280fa8e489ac5fcefa1f370f0ca1acd16ae1b860/statsmodels-0.11.1-cp36-cp36m-manylinux1_x86_64.whl (8.7MB)\n",
            "\u001b[K     |████████████████████████████████| 8.7MB 5.4MB/s \n",
            "\u001b[?25hRequirement already satisfied: scikit-learn>=0.22.2 in /usr/local/lib/python3.6/dist-packages (from feature-engine) (0.22.2.post1)\n",
            "Requirement already satisfied: scipy>=1.4.1 in /usr/local/lib/python3.6/dist-packages (from feature-engine) (1.4.1)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas>=1.0.3->feature-engine) (2018.9)\n",
            "Requirement already satisfied: python-dateutil>=2.6.1 in /usr/local/lib/python3.6/dist-packages (from pandas>=1.0.3->feature-engine) (2.8.1)\n",
            "Requirement already satisfied: patsy>=0.5 in /usr/local/lib/python3.6/dist-packages (from statsmodels>=0.11.1->feature-engine) (0.5.1)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.6/dist-packages (from scikit-learn>=0.22.2->feature-engine) (0.15.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.6/dist-packages (from python-dateutil>=2.6.1->pandas>=1.0.3->feature-engine) (1.12.0)\n",
            "Installing collected packages: statsmodels, feature-engine\n",
            "  Found existing installation: statsmodels 0.10.2\n",
            "    Uninstalling statsmodels-0.10.2:\n",
            "      Successfully uninstalled statsmodels-0.10.2\n",
            "Successfully installed feature-engine-0.4.31 statsmodels-0.11.1\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.colab-display-data+json": {
              "pip_warning": {
                "packages": [
                  "statsmodels"
                ]
              }
            }
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Collecting eli5\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/97/2f/c85c7d8f8548e460829971785347e14e45fa5c6617da374711dec8cb38cc/eli5-0.10.1-py2.py3-none-any.whl (105kB)\n",
            "\r\u001b[K     |███                             | 10kB 19.6MB/s eta 0:00:01\r\u001b[K     |██████▏                         | 20kB 1.8MB/s eta 0:00:01\r\u001b[K     |█████████▎                      | 30kB 2.3MB/s eta 0:00:01\r\u001b[K     |████████████▍                   | 40kB 2.6MB/s eta 0:00:01\r\u001b[K     |███████████████▌                | 51kB 2.0MB/s eta 0:00:01\r\u001b[K     |██████████████████▋             | 61kB 2.3MB/s eta 0:00:01\r\u001b[K     |█████████████████████▊          | 71kB 2.6MB/s eta 0:00:01\r\u001b[K     |████████████████████████▊       | 81kB 2.8MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▉    | 92kB 3.0MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████ | 102kB 2.9MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 112kB 2.9MB/s \n",
            "\u001b[?25hRequirement already satisfied: attrs>16.0.0 in /usr/local/lib/python3.6/dist-packages (from eli5) (19.3.0)\n",
            "Requirement already satisfied: tabulate>=0.7.7 in /usr/local/lib/python3.6/dist-packages (from eli5) (0.8.7)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.6/dist-packages (from eli5) (1.4.1)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.6/dist-packages (from eli5) (2.11.2)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from eli5) (1.12.0)\n",
            "Requirement already satisfied: scikit-learn>=0.18 in /usr/local/lib/python3.6/dist-packages (from eli5) (0.22.2.post1)\n",
            "Requirement already satisfied: graphviz in /usr/local/lib/python3.6/dist-packages (from eli5) (0.10.1)\n",
            "Requirement already satisfied: numpy>=1.9.0 in /usr/local/lib/python3.6/dist-packages (from eli5) (1.18.5)\n",
            "Requirement already satisfied: MarkupSafe>=0.23 in /usr/local/lib/python3.6/dist-packages (from jinja2->eli5) (1.1.1)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.6/dist-packages (from scikit-learn>=0.18->eli5) (0.15.1)\n",
            "Installing collected packages: eli5\n",
            "Successfully installed eli5-0.10.1\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:144: FutureWarning: The sklearn.metrics.scorer module is  deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.metrics. Anything that cannot be imported from sklearn.metrics is now part of the private API.\n",
            "  warnings.warn(message, FutureWarning)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/deprecation.py:144: FutureWarning: The sklearn.feature_selection.base module is  deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.feature_selection. Anything that cannot be imported from sklearn.feature_selection is now part of the private API.\n",
            "  warnings.warn(message, FutureWarning)\n",
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Collecting shap\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/a8/77/b504e43e21a2ba543a1ac4696718beb500cfa708af2fb57cb54ce299045c/shap-0.35.0.tar.gz (273kB)\n",
            "\r\u001b[K     |█▏                              | 10kB 16.0MB/s eta 0:00:01\r\u001b[K     |██▍                             | 20kB 1.8MB/s eta 0:00:01\r\u001b[K     |███▋                            | 30kB 2.3MB/s eta 0:00:01\r\u001b[K     |████▉                           | 40kB 2.6MB/s eta 0:00:01\r\u001b[K     |██████                          | 51kB 2.0MB/s eta 0:00:01\r\u001b[K     |███████▏                        | 61kB 2.3MB/s eta 0:00:01\r\u001b[K     |████████▍                       | 71kB 2.6MB/s eta 0:00:01\r\u001b[K     |█████████▋                      | 81kB 2.8MB/s eta 0:00:01\r\u001b[K     |██████████▉                     | 92kB 3.0MB/s eta 0:00:01\r\u001b[K     |████████████                    | 102kB 2.9MB/s eta 0:00:01\r\u001b[K     |█████████████▏                  | 112kB 2.9MB/s eta 0:00:01\r\u001b[K     |██████████████▍                 | 122kB 2.9MB/s eta 0:00:01\r\u001b[K     |███████████████▋                | 133kB 2.9MB/s eta 0:00:01\r\u001b[K     |████████████████▉               | 143kB 2.9MB/s eta 0:00:01\r\u001b[K     |██████████████████              | 153kB 2.9MB/s eta 0:00:01\r\u001b[K     |███████████████████▏            | 163kB 2.9MB/s eta 0:00:01\r\u001b[K     |████████████████████▍           | 174kB 2.9MB/s eta 0:00:01\r\u001b[K     |█████████████████████▋          | 184kB 2.9MB/s eta 0:00:01\r\u001b[K     |██████████████████████▉         | 194kB 2.9MB/s eta 0:00:01\r\u001b[K     |████████████████████████        | 204kB 2.9MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▏      | 215kB 2.9MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▍     | 225kB 2.9MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▋    | 235kB 2.9MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▉   | 245kB 2.9MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████  | 256kB 2.9MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▏| 266kB 2.9MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 276kB 2.9MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from shap) (1.18.5)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.6/dist-packages (from shap) (1.4.1)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.6/dist-packages (from shap) (0.22.2.post1)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.6/dist-packages (from shap) (1.0.4)\n",
            "Requirement already satisfied: tqdm>4.25.0 in /usr/local/lib/python3.6/dist-packages (from shap) (4.41.1)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.6/dist-packages (from scikit-learn->shap) (0.15.1)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas->shap) (2018.9)\n",
            "Requirement already satisfied: python-dateutil>=2.6.1 in /usr/local/lib/python3.6/dist-packages (from pandas->shap) (2.8.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.6/dist-packages (from python-dateutil>=2.6.1->pandas->shap) (1.12.0)\n",
            "Building wheels for collected packages: shap\n",
            "  Building wheel for shap (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for shap: filename=shap-0.35.0-cp36-cp36m-linux_x86_64.whl size=394115 sha256=4644a230afea06d6c0699cc822a1e4b2232b972da2ef3b5374a7f6c549fd16e9\n",
            "  Stored in directory: /root/.cache/pip/wheels/e7/f7/0f/b57055080cf8894906b3bd3616d2fc2bfd0b12d5161bcb24ac\n",
            "Successfully built shap\n",
            "Installing collected packages: shap\n",
            "Successfully installed shap-0.35.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZLFLLtY2-I06",
        "colab_type": "text"
      },
      "source": [
        "#####- Upload data set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pgQMY8HqDrGQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Upload dataset\n",
        "url ='https://raw.githubusercontent.com/claudio1975/SDS2020/master/data/CAX_MortgageModeling_Train.csv'\n",
        "df = pd.read_csv(url)"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Q2_fumrJ-NCl",
        "colab_type": "text"
      },
      "source": [
        "#####- Split data set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wwPBNz5pDvDZ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Split data set between target and features\n",
        "X_full = df\n",
        "y = X_full.RESULT\n",
        "X_full = X_full.drop(['RESULT'], axis=1)\n"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cIEjdLM1BGti",
        "colab_type": "text"
      },
      "source": [
        "# Summarize Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hvXmGWII7I_8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 502
        },
        "outputId": "5c5edde9-a5ec-43f8-b50c-f5ece8f892d2"
      },
      "source": [
        "# Look at dimension of data set and types of each attribute\n",
        "df.info()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 45642 entries, 0 to 45641\n",
            "Data columns (total 22 columns):\n",
            " #   Column             Non-Null Count  Dtype  \n",
            "---  ------             --------------  -----  \n",
            " 0   Unique_ID          45642 non-null  object \n",
            " 1   MORTGAGE NUMBER    45642 non-null  int64  \n",
            " 2   PROPERTY VALUE     45642 non-null  int64  \n",
            " 3   MORTGAGE PAYMENT   45642 non-null  int64  \n",
            " 4   GDS                45642 non-null  float64\n",
            " 5   LTV                45642 non-null  float64\n",
            " 6   TDS                45642 non-null  float64\n",
            " 7   AMORTIZATION       45642 non-null  int64  \n",
            " 8   MORTGAGE AMOUNT    45642 non-null  float64\n",
            " 9   RATE               45642 non-null  float64\n",
            " 10  MORTGAGE PURPOSE   45642 non-null  object \n",
            " 11  PAYMENT FREQUENCY  45642 non-null  object \n",
            " 12  PROPERTY TYPE      45642 non-null  object \n",
            " 13  TERM               45642 non-null  int64  \n",
            " 14  FSA                45642 non-null  object \n",
            " 15  AGE RANGE          45642 non-null  object \n",
            " 16  GENDER             45642 non-null  object \n",
            " 17  INCOME             45642 non-null  int64  \n",
            " 18  INCOME TYPE        45642 non-null  int64  \n",
            " 19  NAICS CODE         45642 non-null  object \n",
            " 20  CREDIT SCORE       45642 non-null  int64  \n",
            " 21  RESULT             45642 non-null  object \n",
            "dtypes: float64(5), int64(8), object(9)\n",
            "memory usage: 7.7+ MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uDB8l79ZLldm",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 429
        },
        "outputId": "b57cc833-1315-4536-c941-59e349087408"
      },
      "source": [
        "# Summarize attribute distributions of the data frame\n",
        "df.describe(include='all')"
      ],
      "execution_count": 5,
      "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>Unique_ID</th>\n",
              "      <th>MORTGAGE NUMBER</th>\n",
              "      <th>PROPERTY VALUE</th>\n",
              "      <th>MORTGAGE PAYMENT</th>\n",
              "      <th>GDS</th>\n",
              "      <th>LTV</th>\n",
              "      <th>TDS</th>\n",
              "      <th>AMORTIZATION</th>\n",
              "      <th>MORTGAGE AMOUNT</th>\n",
              "      <th>RATE</th>\n",
              "      <th>MORTGAGE PURPOSE</th>\n",
              "      <th>PAYMENT FREQUENCY</th>\n",
              "      <th>PROPERTY TYPE</th>\n",
              "      <th>TERM</th>\n",
              "      <th>FSA</th>\n",
              "      <th>AGE RANGE</th>\n",
              "      <th>GENDER</th>\n",
              "      <th>INCOME</th>\n",
              "      <th>INCOME TYPE</th>\n",
              "      <th>NAICS CODE</th>\n",
              "      <th>CREDIT SCORE</th>\n",
              "      <th>RESULT</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>45642</td>\n",
              "      <td>45642.00000</td>\n",
              "      <td>4.564200e+04</td>\n",
              "      <td>45642.000000</td>\n",
              "      <td>4.564200e+04</td>\n",
              "      <td>45642.000000</td>\n",
              "      <td>4.564200e+04</td>\n",
              "      <td>45642.000000</td>\n",
              "      <td>4.564200e+04</td>\n",
              "      <td>45642.000000</td>\n",
              "      <td>45642</td>\n",
              "      <td>45642</td>\n",
              "      <td>45642</td>\n",
              "      <td>45642.000000</td>\n",
              "      <td>45642</td>\n",
              "      <td>45642</td>\n",
              "      <td>45642</td>\n",
              "      <td>4.564200e+04</td>\n",
              "      <td>45642.000000</td>\n",
              "      <td>45642</td>\n",
              "      <td>45642.000000</td>\n",
              "      <td>45642</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>unique</th>\n",
              "      <td>45642</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>8</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1141</td>\n",
              "      <td>11</td>\n",
              "      <td>3</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>19</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>top</th>\n",
              "      <td>CAX_Train_33578</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Purchase</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>NaN</td>\n",
              "      <td>L7A</td>\n",
              "      <td>45-49</td>\n",
              "      <td>Male</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>56</td>\n",
              "      <td>NaN</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>freq</th>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>24419</td>\n",
              "      <td>42785</td>\n",
              "      <td>29377</td>\n",
              "      <td>NaN</td>\n",
              "      <td>898</td>\n",
              "      <td>7645</td>\n",
              "      <td>24849</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>11245</td>\n",
              "      <td>NaN</td>\n",
              "      <td>36023</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>NaN</td>\n",
              "      <td>22821.50000</td>\n",
              "      <td>5.826110e+05</td>\n",
              "      <td>2392.148043</td>\n",
              "      <td>2.138033e+02</td>\n",
              "      <td>70.632932</td>\n",
              "      <td>4.125670e+02</td>\n",
              "      <td>355.309583</td>\n",
              "      <td>4.255704e+05</td>\n",
              "      <td>4.651149</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15.286578</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.251093e+05</td>\n",
              "      <td>5.315017</td>\n",
              "      <td>NaN</td>\n",
              "      <td>662.662811</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>NaN</td>\n",
              "      <td>13175.85483</td>\n",
              "      <td>4.382082e+05</td>\n",
              "      <td>1688.208721</td>\n",
              "      <td>3.263766e+04</td>\n",
              "      <td>11.566444</td>\n",
              "      <td>6.425859e+04</td>\n",
              "      <td>18.011798</td>\n",
              "      <td>3.020469e+05</td>\n",
              "      <td>0.642371</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>7.443857</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.149974e+05</td>\n",
              "      <td>3.267761</td>\n",
              "      <td>NaN</td>\n",
              "      <td>132.818234</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>NaN</td>\n",
              "      <td>1.00000</td>\n",
              "      <td>1.400000e+04</td>\n",
              "      <td>141.000000</td>\n",
              "      <td>-3.704000e+01</td>\n",
              "      <td>6.667000</td>\n",
              "      <td>-1.070000e+01</td>\n",
              "      <td>60.000000</td>\n",
              "      <td>1.500000e+04</td>\n",
              "      <td>2.500000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>6.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.000000e+03</td>\n",
              "      <td>-1.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>NaN</td>\n",
              "      <td>11411.25000</td>\n",
              "      <td>3.400000e+05</td>\n",
              "      <td>1406.000000</td>\n",
              "      <td>1.984000e+01</td>\n",
              "      <td>65.000000</td>\n",
              "      <td>3.284000e+01</td>\n",
              "      <td>360.000000</td>\n",
              "      <td>2.459250e+05</td>\n",
              "      <td>4.250000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>12.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>6.800000e+04</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>623.000000</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>NaN</td>\n",
              "      <td>22821.50000</td>\n",
              "      <td>4.770000e+05</td>\n",
              "      <td>2033.000000</td>\n",
              "      <td>2.921000e+01</td>\n",
              "      <td>75.000000</td>\n",
              "      <td>3.830000e+01</td>\n",
              "      <td>360.000000</td>\n",
              "      <td>3.575000e+05</td>\n",
              "      <td>4.500000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>12.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>9.700000e+04</td>\n",
              "      <td>6.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>677.000000</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>NaN</td>\n",
              "      <td>34231.75000</td>\n",
              "      <td>6.850000e+05</td>\n",
              "      <td>2866.000000</td>\n",
              "      <td>3.587000e+01</td>\n",
              "      <td>80.000000</td>\n",
              "      <td>4.217000e+01</td>\n",
              "      <td>360.000000</td>\n",
              "      <td>5.120000e+05</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>12.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.450000e+05</td>\n",
              "      <td>8.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>736.000000</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>NaN</td>\n",
              "      <td>45642.00000</td>\n",
              "      <td>1.800000e+07</td>\n",
              "      <td>46479.000000</td>\n",
              "      <td>6.872300e+06</td>\n",
              "      <td>95.000000</td>\n",
              "      <td>1.346256e+07</td>\n",
              "      <td>480.000000</td>\n",
              "      <td>7.500000e+06</td>\n",
              "      <td>10.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>60.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3.860000e+06</td>\n",
              "      <td>20.000000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>899.000000</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "              Unique_ID  MORTGAGE NUMBER  ...  CREDIT SCORE  RESULT\n",
              "count             45642      45642.00000  ...  45642.000000   45642\n",
              "unique            45642              NaN  ...           NaN       2\n",
              "top     CAX_Train_33578              NaN  ...           NaN  FUNDED\n",
              "freq                  1              NaN  ...           NaN   36023\n",
              "mean                NaN      22821.50000  ...    662.662811     NaN\n",
              "std                 NaN      13175.85483  ...    132.818234     NaN\n",
              "min                 NaN          1.00000  ...      0.000000     NaN\n",
              "25%                 NaN      11411.25000  ...    623.000000     NaN\n",
              "50%                 NaN      22821.50000  ...    677.000000     NaN\n",
              "75%                 NaN      34231.75000  ...    736.000000     NaN\n",
              "max                 NaN      45642.00000  ...    899.000000     NaN\n",
              "\n",
              "[11 rows x 22 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vl_waIVssaPb",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 566
        },
        "outputId": "6a5fdd6d-6011-47d3-f874-49f25cd8a973"
      },
      "source": [
        "# Take a peek at the first rows of the data\n",
        "df.head(10)"
      ],
      "execution_count": 6,
      "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>Unique_ID</th>\n",
              "      <th>MORTGAGE NUMBER</th>\n",
              "      <th>PROPERTY VALUE</th>\n",
              "      <th>MORTGAGE PAYMENT</th>\n",
              "      <th>GDS</th>\n",
              "      <th>LTV</th>\n",
              "      <th>TDS</th>\n",
              "      <th>AMORTIZATION</th>\n",
              "      <th>MORTGAGE AMOUNT</th>\n",
              "      <th>RATE</th>\n",
              "      <th>MORTGAGE PURPOSE</th>\n",
              "      <th>PAYMENT FREQUENCY</th>\n",
              "      <th>PROPERTY TYPE</th>\n",
              "      <th>TERM</th>\n",
              "      <th>FSA</th>\n",
              "      <th>AGE RANGE</th>\n",
              "      <th>GENDER</th>\n",
              "      <th>INCOME</th>\n",
              "      <th>INCOME TYPE</th>\n",
              "      <th>NAICS CODE</th>\n",
              "      <th>CREDIT SCORE</th>\n",
              "      <th>RESULT</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>CAX_Train_1</td>\n",
              "      <td>1</td>\n",
              "      <td>900000</td>\n",
              "      <td>5429</td>\n",
              "      <td>61.98</td>\n",
              "      <td>65.000</td>\n",
              "      <td>71.63</td>\n",
              "      <td>360</td>\n",
              "      <td>1040000.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Refinance</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>12</td>\n",
              "      <td>L4C</td>\n",
              "      <td>Under 25</td>\n",
              "      <td>Male</td>\n",
              "      <td>108000</td>\n",
              "      <td>8</td>\n",
              "      <td>44-45</td>\n",
              "      <td>681</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>CAX_Train_2</td>\n",
              "      <td>2</td>\n",
              "      <td>386000</td>\n",
              "      <td>2179</td>\n",
              "      <td>35.22</td>\n",
              "      <td>74.290</td>\n",
              "      <td>40.65</td>\n",
              "      <td>360</td>\n",
              "      <td>390000.0</td>\n",
              "      <td>4.5</td>\n",
              "      <td>Refinance</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>12</td>\n",
              "      <td>L9T</td>\n",
              "      <td>70 and over</td>\n",
              "      <td>Male</td>\n",
              "      <td>78000</td>\n",
              "      <td>2</td>\n",
              "      <td>56</td>\n",
              "      <td>710</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>CAX_Train_3</td>\n",
              "      <td>3</td>\n",
              "      <td>531000</td>\n",
              "      <td>2152</td>\n",
              "      <td>30.97</td>\n",
              "      <td>80.000</td>\n",
              "      <td>35.41</td>\n",
              "      <td>360</td>\n",
              "      <td>424800.0</td>\n",
              "      <td>3.5</td>\n",
              "      <td>Purchase</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>6</td>\n",
              "      <td>M1N</td>\n",
              "      <td>35-39</td>\n",
              "      <td>Female</td>\n",
              "      <td>87000</td>\n",
              "      <td>2</td>\n",
              "      <td>72</td>\n",
              "      <td>709</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>CAX_Train_4</td>\n",
              "      <td>4</td>\n",
              "      <td>1200000</td>\n",
              "      <td>5410</td>\n",
              "      <td>19.04</td>\n",
              "      <td>75.000</td>\n",
              "      <td>34.14</td>\n",
              "      <td>360</td>\n",
              "      <td>960000.0</td>\n",
              "      <td>5.5</td>\n",
              "      <td>Purchase</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>12</td>\n",
              "      <td>M2M</td>\n",
              "      <td>45-49</td>\n",
              "      <td>Male</td>\n",
              "      <td>300000</td>\n",
              "      <td>8</td>\n",
              "      <td>54</td>\n",
              "      <td>761</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>CAX_Train_5</td>\n",
              "      <td>5</td>\n",
              "      <td>350000</td>\n",
              "      <td>3342</td>\n",
              "      <td>29.59</td>\n",
              "      <td>80.000</td>\n",
              "      <td>34.85</td>\n",
              "      <td>360</td>\n",
              "      <td>592000.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>Refinance</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>12</td>\n",
              "      <td>L7G</td>\n",
              "      <td>50-54</td>\n",
              "      <td>Male</td>\n",
              "      <td>147000</td>\n",
              "      <td>8</td>\n",
              "      <td>62</td>\n",
              "      <td>762</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>CAX_Train_6</td>\n",
              "      <td>6</td>\n",
              "      <td>420000</td>\n",
              "      <td>2179</td>\n",
              "      <td>26.16</td>\n",
              "      <td>69.050</td>\n",
              "      <td>26.16</td>\n",
              "      <td>240</td>\n",
              "      <td>290000.0</td>\n",
              "      <td>5.5</td>\n",
              "      <td>Refinance</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>24</td>\n",
              "      <td>M1M</td>\n",
              "      <td>40-44</td>\n",
              "      <td>Female</td>\n",
              "      <td>106000</td>\n",
              "      <td>2</td>\n",
              "      <td>53</td>\n",
              "      <td>570</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>CAX_Train_7</td>\n",
              "      <td>7</td>\n",
              "      <td>1600000</td>\n",
              "      <td>5275</td>\n",
              "      <td>32.12</td>\n",
              "      <td>69.230</td>\n",
              "      <td>33.90</td>\n",
              "      <td>360</td>\n",
              "      <td>900000.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Refinance</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>12</td>\n",
              "      <td>L7B</td>\n",
              "      <td>55-59</td>\n",
              "      <td>Female</td>\n",
              "      <td>185000</td>\n",
              "      <td>8</td>\n",
              "      <td>56</td>\n",
              "      <td>812</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>CAX_Train_8</td>\n",
              "      <td>8</td>\n",
              "      <td>1500000</td>\n",
              "      <td>4383</td>\n",
              "      <td>35.68</td>\n",
              "      <td>55.170</td>\n",
              "      <td>37.81</td>\n",
              "      <td>360</td>\n",
              "      <td>644000.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Refinance</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>12</td>\n",
              "      <td>L3P</td>\n",
              "      <td>45-49</td>\n",
              "      <td>Male</td>\n",
              "      <td>160000</td>\n",
              "      <td>8</td>\n",
              "      <td>48-49</td>\n",
              "      <td>683</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>CAX_Train_9</td>\n",
              "      <td>9</td>\n",
              "      <td>790000</td>\n",
              "      <td>3092</td>\n",
              "      <td>35.29</td>\n",
              "      <td>78.481</td>\n",
              "      <td>35.40</td>\n",
              "      <td>360</td>\n",
              "      <td>620000.0</td>\n",
              "      <td>3.5</td>\n",
              "      <td>Refinance</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>60</td>\n",
              "      <td>M4C</td>\n",
              "      <td>45-49</td>\n",
              "      <td>Female</td>\n",
              "      <td>89000</td>\n",
              "      <td>2</td>\n",
              "      <td>61</td>\n",
              "      <td>684</td>\n",
              "      <td>NOT FUNDED</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>CAX_Train_10</td>\n",
              "      <td>10</td>\n",
              "      <td>650000</td>\n",
              "      <td>1839</td>\n",
              "      <td>38.78</td>\n",
              "      <td>41.850</td>\n",
              "      <td>38.98</td>\n",
              "      <td>360</td>\n",
              "      <td>272000.0</td>\n",
              "      <td>5.5</td>\n",
              "      <td>Purchase</td>\n",
              "      <td>Monthly</td>\n",
              "      <td>Single Detached</td>\n",
              "      <td>12</td>\n",
              "      <td>L4W</td>\n",
              "      <td>40-44</td>\n",
              "      <td>Male</td>\n",
              "      <td>60000</td>\n",
              "      <td>2</td>\n",
              "      <td>56</td>\n",
              "      <td>716</td>\n",
              "      <td>FUNDED</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "      Unique_ID  MORTGAGE NUMBER  ...  CREDIT SCORE      RESULT\n",
              "0   CAX_Train_1                1  ...           681      FUNDED\n",
              "1   CAX_Train_2                2  ...           710      FUNDED\n",
              "2   CAX_Train_3                3  ...           709      FUNDED\n",
              "3   CAX_Train_4                4  ...           761      FUNDED\n",
              "4   CAX_Train_5                5  ...           762      FUNDED\n",
              "5   CAX_Train_6                6  ...           570      FUNDED\n",
              "6   CAX_Train_7                7  ...           812      FUNDED\n",
              "7   CAX_Train_8                8  ...           683      FUNDED\n",
              "8   CAX_Train_9                9  ...           684  NOT FUNDED\n",
              "9  CAX_Train_10               10  ...           716      FUNDED\n",
              "\n",
              "[10 rows x 22 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "H6N4eR9vEXQi",
        "colab_type": "text"
      },
      "source": [
        "Explanatory variables are grouped into categorical variables and numerical variables and for each one let's do a graphical and non-graphical analysis, but before this split let's run some some data preparation activities."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UqhoHvdpPKse",
        "colab_type": "text"
      },
      "source": [
        "# Formatting Features"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bbHzfBaj2J5t",
        "colab_type": "text"
      },
      "source": [
        "If necessary, it's a good practice to format data, after have taken a peek of it. Missing values on numeric features are marked by \"-1\", meanwhile for categorical features they are marked with \"Unknown\"; let's imput these values with \"NA\".  "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BoC1p8FWLFbF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Replaced both '-1' and 'Unknown' values with NA's\n",
        "X_full[X_full== -1] = np.nan\n",
        "X_full[X_full==\"Unknown\"] = np.nan"
      ],
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Cfr3gw7vQeWB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Format data into float and object types and split mixed variables\n",
        "X_full['PROPERTY VALUE'] = X_full['PROPERTY VALUE'].astype(float)\n",
        "X_full['MORTGAGE PAYMENT'] = X_full['MORTGAGE PAYMENT'].astype(float)\n",
        "X_full['AMORTIZATION'] = X_full['AMORTIZATION'].astype(float)\n",
        "X_full['TERM'] = X_full['TERM'].astype(float)\n",
        "X_full['INCOME'] = X_full['INCOME'].astype(float)\n",
        "X_full['INCOME TYPE'] = X_full['INCOME TYPE'].astype(object)\n",
        "X_full['CREDIT SCORE'] = X_full['CREDIT SCORE'].astype(float)\n",
        "X_full['FSA_num'] = X_full['FSA'].str.extract('(\\d+)') # extract numerical part\n",
        "X_full['FSA_let'] = X_full['FSA'].str[0] # extract the first letter"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Sa67IEEHGdPg",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Rename some features for a practical use\n",
        "X_full = X_full.rename(columns={\"MORTGAGE PURPOSE\":\"MORTGAGE_PURPOSE\",\"PAYMENT FREQUENCY\":\"PAYMENT_FREQUENCY\",\"PROPERTY TYPE\":\"PROPERTY_TYPE\",\"AGE RANGE\":\"AGE_RANGE\",\"PROPERTY VALUE\": \"PROPERTY_VALUE\",\n",
        "                                \"MORTGAGE PAYMENT\": \"MORTGAGE_PAYMENT\", \"MORTGAGE AMOUNT\":\"MORTGAGE_AMOUNT\",\"INCOME TYPE\":\"INCOME_TYPE\",\"CREDIT SCORE\":\"CREDIT_SCORE\"})"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uqcTsDB4OkMK",
        "colab_type": "text"
      },
      "source": [
        "# Handling Categorical Features\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BMN3HpKV3S10",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 201
        },
        "outputId": "1c452230-97de-424a-b348-962c8015ddf4"
      },
      "source": [
        "# let's have a look at how many labels for categorical features\n",
        "for col in X_full.columns:\n",
        "  if X_full[col].dtype ==\"object\":\n",
        "    print(col, ': ', len(X_full[col].unique()), ' labels')"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Unique_ID :  45642  labels\n",
            "MORTGAGE_PURPOSE :  2  labels\n",
            "PAYMENT_FREQUENCY :  6  labels\n",
            "PROPERTY_TYPE :  8  labels\n",
            "FSA :  1141  labels\n",
            "AGE_RANGE :  11  labels\n",
            "GENDER :  3  labels\n",
            "INCOME_TYPE :  21  labels\n",
            "NAICS CODE :  19  labels\n",
            "FSA_num :  10  labels\n",
            "FSA_let :  17  labels\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NREoC4kEBP17",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Select categorical columns with relatively low cardinality (convenient but arbitrary)\n",
        "categorical_cols = [cname for cname in X_full.columns if\n",
        "                    X_full[cname].nunique() <= 15 and \n",
        "                    X_full[cname].dtype == \"object\"]"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IfZUySHMrLiS",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 67
        },
        "outputId": "9e80cbff-4891-488a-a84c-0ae6eb0aa67d"
      },
      "source": [
        "# Subset with categorical features\n",
        "cat = X_full[categorical_cols]\n",
        "cat.columns\n"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['MORTGAGE_PURPOSE', 'PAYMENT_FREQUENCY', 'PROPERTY_TYPE', 'AGE_RANGE',\n",
              "       'GENDER', 'FSA_num'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y9notscfuyPD",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 234
        },
        "outputId": "67e8f8aa-b45f-44c3-bb0a-5af1a606e044"
      },
      "source": [
        "cat.info()"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 45642 entries, 0 to 45641\n",
            "Data columns (total 6 columns):\n",
            " #   Column             Non-Null Count  Dtype \n",
            "---  ------             --------------  ----- \n",
            " 0   MORTGAGE_PURPOSE   45642 non-null  object\n",
            " 1   PAYMENT_FREQUENCY  45642 non-null  object\n",
            " 2   PROPERTY_TYPE      45642 non-null  object\n",
            " 3   AGE_RANGE          45642 non-null  object\n",
            " 4   GENDER             39224 non-null  object\n",
            " 5   FSA_num            45642 non-null  object\n",
            "dtypes: object(6)\n",
            "memory usage: 2.1+ MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aHJVRnRmT7-j",
        "colab_type": "text"
      },
      "source": [
        "#####- Feature Engineering on categorical features: one-hot encoding into k-1 dummy variables"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CSvpWEpNMQuJ",
        "colab_type": "text"
      },
      "source": [
        "Let's transform categorical features into numerical variables with one-hot encoding methodology to afford a better understanding of variables by machine learning models."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i7UsSPk3EdMd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# One-hot encode the data (to shorten the code, I use pandas)\n",
        "HOcat = pd.concat([pd.get_dummies(cat.MORTGAGE_PURPOSE, drop_first=True), pd.get_dummies(cat.PAYMENT_FREQUENCY, drop_first=True), pd.get_dummies(cat.PROPERTY_TYPE, drop_first=True),\n",
        "                  pd.get_dummies(cat.AGE_RANGE, drop_first=True), pd.get_dummies(cat.GENDER, drop_first=True), pd.get_dummies(cat.FSA_num,prefix='FSA_num',drop_first=True)], axis=1)\n",
        "\n",
        "HOcat = HOcat.astype(int)\n"
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Nsr5D7xTHNdc",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 33
        },
        "outputId": "4f1aabda-cfc8-4838-bd67-5a2d70e203a6"
      },
      "source": [
        "# Look at the new subset\n",
        "HOcat.shape"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(45642, 33)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7pUe0kjijI5i",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 250
        },
        "outputId": "1325a45b-0df5-4daa-b782-73743db587e4"
      },
      "source": [
        "HOcat.head()"
      ],
      "execution_count": 16,
      "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>Refinance</th>\n",
              "      <th>Bi-Weekly Accelerated</th>\n",
              "      <th>Monthly</th>\n",
              "      <th>Semi-Monthly</th>\n",
              "      <th>Weekly</th>\n",
              "      <th>Weekly Accelerated</th>\n",
              "      <th>Fourplex</th>\n",
              "      <th>High Rise</th>\n",
              "      <th>Semi-Detached</th>\n",
              "      <th>Single Detached</th>\n",
              "      <th>Townhouse Condominium</th>\n",
              "      <th>Townhouse Freehold</th>\n",
              "      <th>Triplex</th>\n",
              "      <th>30-34</th>\n",
              "      <th>35-39</th>\n",
              "      <th>40-44</th>\n",
              "      <th>45-49</th>\n",
              "      <th>50-54</th>\n",
              "      <th>55-59</th>\n",
              "      <th>60-64</th>\n",
              "      <th>65-69</th>\n",
              "      <th>70 and over</th>\n",
              "      <th>Under 25</th>\n",
              "      <th>Male</th>\n",
              "      <th>FSA_num_1</th>\n",
              "      <th>FSA_num_2</th>\n",
              "      <th>FSA_num_3</th>\n",
              "      <th>FSA_num_4</th>\n",
              "      <th>FSA_num_5</th>\n",
              "      <th>FSA_num_6</th>\n",
              "      <th>FSA_num_7</th>\n",
              "      <th>FSA_num_8</th>\n",
              "      <th>FSA_num_9</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Refinance  Bi-Weekly Accelerated  Monthly  ...  FSA_num_7  FSA_num_8  FSA_num_9\n",
              "0          1                      0        1  ...          0          0          0\n",
              "1          1                      0        1  ...          0          0          1\n",
              "2          0                      0        1  ...          0          0          0\n",
              "3          0                      0        1  ...          0          0          0\n",
              "4          1                      0        1  ...          1          0          0\n",
              "\n",
              "[5 rows x 33 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kiqq6UMKYy_T",
        "colab_type": "text"
      },
      "source": [
        "# Numerical Features"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c_0FZBcsU8eg",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Select numerical columns\n",
        "numerical_cols = [cname for cname in X_full.columns if \n",
        "                X_full[cname].dtype in ['float64']]"
      ],
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fg-ru4uxU88C",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 84
        },
        "outputId": "505bb0c4-86e4-45e9-a2e1-1fc4d33407c1"
      },
      "source": [
        "# Subset with numerical features\n",
        "num = X_full[numerical_cols]\n",
        "num.columns"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['PROPERTY_VALUE', 'MORTGAGE_PAYMENT', 'GDS', 'LTV', 'TDS',\n",
              "       'AMORTIZATION', 'MORTGAGE_AMOUNT', 'RATE', 'TERM', 'INCOME',\n",
              "       'CREDIT_SCORE'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HPeULBW3U9Fo",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 318
        },
        "outputId": "4b745279-de64-45a0-fa00-33881f7a0ecd"
      },
      "source": [
        "num.info()"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 45642 entries, 0 to 45641\n",
            "Data columns (total 11 columns):\n",
            " #   Column            Non-Null Count  Dtype  \n",
            "---  ------            --------------  -----  \n",
            " 0   PROPERTY_VALUE    45642 non-null  float64\n",
            " 1   MORTGAGE_PAYMENT  45642 non-null  float64\n",
            " 2   GDS               45642 non-null  float64\n",
            " 3   LTV               45642 non-null  float64\n",
            " 4   TDS               45642 non-null  float64\n",
            " 5   AMORTIZATION      45642 non-null  float64\n",
            " 6   MORTGAGE_AMOUNT   45642 non-null  float64\n",
            " 7   RATE              45642 non-null  float64\n",
            " 8   TERM              45642 non-null  float64\n",
            " 9   INCOME            45642 non-null  float64\n",
            " 10  CREDIT_SCORE      45642 non-null  float64\n",
            "dtypes: float64(11)\n",
            "memory usage: 3.8 MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eSRSSCLGVEh5",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Grasp all\n",
        "X_all = pd.concat([HOcat, num], axis=1, join='inner')"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e3g7VOS0w2lD",
        "colab_type": "text"
      },
      "source": [
        "# Split data set"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "n_Hw64zwxu_6",
        "colab_type": "text"
      },
      "source": [
        "To analyze the performance of a model is a good manner to split the data set into the training set and the test set. It's been decided to split it into three parts: training set, validation set and test set for a better understanding of models. The training set is a sample of data used to fit the model, meanwhile the validation set is a sample of data used to provide an unbiased evaluation of the model that fit on the training set and to tune the model hyperparameters (not in this explorative phase). The test set is a sample of data used to provide an unbiased evaluation of the model applied on data never seen before."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "csVOGV7Qw5VG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Break off validation and test set from training data\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_all, y, train_size=0.8, test_size=0.2,\n",
        "                                                                random_state=0)\n",
        "X_train, X_valid, y_train, y_valid = train_test_split(X_train, y_train, train_size=0.8, test_size=0.2,\n",
        "                                                                random_state=0)"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Tfvk77arw9dH",
        "colab_type": "text"
      },
      "source": [
        "# Standardization"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "geBXV2Ah45YO",
        "colab_type": "text"
      },
      "source": [
        "Since values of the features are not uniform and may be neagatively impact the skill of some models, the same models are evaluated with a standardized copy of the data set. It means, data are transformed such that each feature has a mean value of 0 and a standard deviation of 1. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rLARBi9ow-OD",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Standardization of data\n",
        "sc = StandardScaler()\n",
        "X_train_sc = sc.fit_transform(X_train)\n",
        "X_valid_sc = sc.fit_transform(X_valid)\n",
        "X_test_sc = sc.transform(X_test)"
      ],
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qe16GLXqopQl",
        "colab_type": "text"
      },
      "source": [
        "# Modeling Part"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "z5VfzCYGcWgq",
        "colab_type": "text"
      },
      "source": [
        "The traditional data exploration is extended looking at the behaviour of several baseline models and which features can be relevant for the prediction. This exploration is splitted in two parts: without handling the imbalanced target variable (scaled baseline models) and handling it (scaled baseline models).  "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6S3Jt8sxLF2b",
        "colab_type": "text"
      },
      "source": [
        "- Evaluation Metric and Confusion Matrix"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9zebre5UP0u7",
        "colab_type": "text"
      },
      "source": [
        "The confusion matrix is a summary table representation of prediction results for a classification problem. The number of correct and incorrect predictions are summarized with count values and broken down by each class. The diagonal elements represent the number of points for which the predicted label is equal to the true label, while off-diagonal elements are those that are mislabeled by the classifier. Good predictions coming from the higher diagonal values of the confusion matrix. For this imbalanced classification task is not used Accuracy metric but more appropriately the F1 score metric that combines both precision and recall, it's an harmonic mean between them, it's indicates how precise is the classifier (precision) and how robust it is (recall). F1 score equal to 0.00 indicates a poor model, instead F1 score equal 1.00 indicates a perfect model.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EQWGVT8QwwI8",
        "colab_type": "text"
      },
      "source": [
        "#  Modeling Part I: without handling imbalanced data set"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9fHQdwCHxrIL",
        "colab_type": "text"
      },
      "source": [
        "The analysis is based on six baseline models: Logistic Regression as the easiest model and as well as benchmark, then other five models: Bagging, Random Forest, AdaBoost, Gradient Boosting Machine and Neural Networks (MLP)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mwCD7aCLxEc6",
        "colab_type": "text"
      },
      "source": [
        "#####- Baseline Models"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6d8vIEE4xFMF",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "b3c0a2b6-f387-422d-fe8b-6e45f3a066bd"
      },
      "source": [
        "# Spot Check Algorithms\n",
        "models = []\n",
        "models.append(('LogisticRegression', LogisticRegression(random_state=0)))\n",
        "models.append(('Bagging', BaggingClassifier(random_state=0)))\n",
        "models.append(('RandomForest', RandomForestClassifier(random_state=0)))\n",
        "models.append(('AdaBoost', AdaBoostClassifier(random_state=0)))\n",
        "models.append(('GBM', GradientBoostingClassifier(random_state=0)))\n",
        "models.append(('NN', MLPClassifier(random_state=0)))\n",
        "results_tr = []\n",
        "results_v = []\n",
        "results_t = []\n",
        "names = []\n",
        "score = []\n",
        "skf = StratifiedKFold(n_splits=5)\n",
        "for (name, model) in models:\n",
        "    param_grid = {}\n",
        "    my_model = GridSearchCV(model,param_grid,cv=skf)\n",
        "    my_model.fit(X_train_sc, y_train)\n",
        "    predictions_tr = my_model.predict(X_train_sc) \n",
        "    predictions_v = my_model.predict(X_valid_sc)\n",
        "    predictions_t = my_model.predict(X_test_sc)\n",
        "    f1_train = f1_score(y_train, predictions_tr, average='macro') \n",
        "    f1_valid = f1_score(y_valid, predictions_v,average='macro') \n",
        "    f1_test = f1_score(y_test, predictions_t,average='macro') \n",
        "    results_tr.append(f1_train)\n",
        "    results_v.append(f1_valid)\n",
        "    results_t.append(f1_test)\n",
        "    \n",
        "    names.append(name)\n",
        "    f_dict = {\n",
        "        'model': name,\n",
        "        'f1_train': f1_train,\n",
        "        'f1_valid': f1_valid,\n",
        "        'f1_test': f1_test\n",
        "    }\n",
        "    score.append(f_dict)\n",
        "    # Computing Confusion matrix for the above algorithms\n",
        "    sns.set( rc = {'figure.figsize': (5, 5)})\n",
        "    plt.figure()\n",
        "    plot_confusion_matrix(my_model,X_test_sc, y_test,values_format= '.2f', cmap='Blues')\n",
        "    plt.title(name)\n",
        "    plt.show()   \n",
        "score = pd.DataFrame(score, columns = ['model','f1_train', 'f1_valid', 'f1_test'])"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x360 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x360 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x360 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x360 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x360 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x360 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oyNrSed6x_68",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 134
        },
        "outputId": "eb3b55fc-c7d7-41c8-b2b9-6105f5677cd9"
      },
      "source": [
        "# Look at the F1 score for each model and for each data set\n",
        "print(score)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "                model  f1_train  f1_valid   f1_test\n",
            "0  LogisticRegression  0.457653  0.460203  0.457607\n",
            "1             Bagging  0.962447  0.420451  0.421311\n",
            "2        RandomForest  0.999949  0.416236  0.413635\n",
            "3            AdaBoost  0.493106  0.431048  0.416958\n",
            "4                 GBM  0.497147  0.354919  0.346063\n",
            "5                  NN  0.658520  0.499372  0.489637\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BBOAoq8kyDJe",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 465
        },
        "outputId": "da46879f-fd2c-4806-ebf5-dc41900cc3be"
      },
      "source": [
        "# Plot results for a graphical comparison\n",
        "print(\"Spot Check Algorithms\")\n",
        "sns.set( rc = {'figure.figsize': (15, 5)})\n",
        "plt.figure()\n",
        "plt.subplot(1,3,1)  \n",
        "sns.stripplot(x=\"model\", y=\"f1_train\",data=score,size=15)\n",
        "plt.xticks(rotation=90)\n",
        "plt.title('Train results')\n",
        "axes = plt.gca()\n",
        "axes.set_ylim([0,1.1])\n",
        "plt.subplot(1,3,2)\n",
        "sns.stripplot(x=\"model\", y=\"f1_valid\",data=score,size=15)\n",
        "plt.xticks(rotation=90)\n",
        "plt.title('Validation results')\n",
        "axes = plt.gca()\n",
        "axes.set_ylim([0,1.1])\n",
        "plt.subplot(1,3,3)\n",
        "sns.stripplot(x=\"model\", y=\"f1_test\",data=score,size=15)\n",
        "plt.xticks(rotation=90)\n",
        "plt.title('Test results')\n",
        "axes = plt.gca()\n",
        "axes.set_ylim([0,1.1])\n",
        "plt.show()"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Spot Check Algorithms\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}