credit_risk_analysis.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "HACKANONS COLAB 25GB RAM.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "machine_shape": "hm"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WrYP16czjBz6"
      },
      "source": [
        "# **1. Overview**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nmPtZP7EjBjg"
      },
      "source": [
        "### **Context :** \r\n",
        "Banks need to protect their interest before it can take risk on you and issue credit card to you. Banks use their previous credit card holders records for understanding the patterns of the card holders. It is a lot more complex process to predict whether a person who they do not know at personal level, will be a defaulter or not. Banks, along with the data from their own records, also use CIBIL data.\r\n",
        "Based on all this data, banks want to develop a pattern that will tell them who are likely to be a defaulter and who are not.\r\n",
        "\r\n",
        "### **Attribute Information :** \r\n",
        "The dataset has 13 features with 50636 observations. The features are:\r\n",
        "\r\n",
        "Here <b>age, gender</b> are the age and gender of the card holder.<br>\r\n",
        "<b>education</b> is the last acquired educational qualification of the card holder.<br>\r\n",
        "<b>occupation</b> can be salaried, or self employed or business etc. <br>\r\n",
        "<b>organization_type</b> can be tire 1, 2, 3 etc.<br> \r\n",
        "<b>seniority</b> denotes at which career level the card holder is in.<br> \r\n",
        "<b>annual_income</b> is the gross annual income of the card holder.<br> \r\n",
        "<b>disposable_income</b> is annual income - recurring expenses.<br> \r\n",
        "<b>house_type</b> is owned or rented or company provided etc.<br> \r\n",
        "<b>vehicle_type</b> is 4-wheeler or two-wheeler or none.<br> \r\n",
        "<b>marital_status</b> is of the card holder.<br> \r\n",
        "<b>no_card</b> has the information of the number of other credit cards that the card holder already holds. <br>\r\n",
        "And at the end of each row, we have a <b>defaulter</b> indicator indicating whether the card holder was a defaulter or not. It is 1 if the card holder was a defaulter, 0 otherwise.\r\n",
        "### **Objective :**\r\n",
        "We have to use this dataset to generate a classification model that can successfully predict for a new applicant with recorded data for given parameters in the data set, if he is likely to be a defaulter."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kOL_PF6UjfhP"
      },
      "source": [
        "# **2. Import Libraries:**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TqSUzglmhLps"
      },
      "source": [
        "# import necessary libraries\r\n",
        "import pandas as pd\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "import numpy as np\r\n",
        "import seaborn as sns\r\n",
        "\r\n",
        "from sklearn.preprocessing import LabelEncoder\r\n",
        "from sklearn.preprocessing import StandardScaler\r\n",
        "from sklearn.model_selection import train_test_split\r\n",
        "\r\n",
        "import xgboost as xgb\r\n",
        "import lightgbm as lgb\r\n",
        "from sklearn.svm import SVC\r\n",
        "from sklearn.ensemble import BaggingClassifier\r\n",
        "from sklearn.tree import DecisionTreeClassifier\r\n",
        "from sklearn.ensemble import RandomForestClassifier, ExtraTreesClassifier\r\n",
        "from sklearn.model_selection import RandomizedSearchCV,GridSearchCV\r\n",
        "from sklearn.model_selection import KFold\r\n",
        "from datetime import datetime\r\n",
        "import warnings"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-qzi4pOewgEJ"
      },
      "source": [
        "# **3. Load Dataset:**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IZ35IRoNhLvr",
        "outputId": "8469b19e-3c89-4aa0-9a11-40b5d1bbea36",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        }
      },
      "source": [
        "# Load the dataset\r\n",
        "credit_df=pd.read_csv(\"credit_data.csv\")\r\n",
        "credit_df.head()"
      ],
      "execution_count": 2,
      "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>age</th>\n",
              "      <th>gender</th>\n",
              "      <th>education</th>\n",
              "      <th>occupation</th>\n",
              "      <th>organization_type</th>\n",
              "      <th>seniority</th>\n",
              "      <th>annual_income</th>\n",
              "      <th>disposable_income</th>\n",
              "      <th>house_type</th>\n",
              "      <th>vehicle_type</th>\n",
              "      <th>marital_status</th>\n",
              "      <th>no_card</th>\n",
              "      <th>default</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19</td>\n",
              "      <td>Male</td>\n",
              "      <td>Graduate</td>\n",
              "      <td>Professional</td>\n",
              "      <td>None</td>\n",
              "      <td>None</td>\n",
              "      <td>186319</td>\n",
              "      <td>21625</td>\n",
              "      <td>Family</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>18</td>\n",
              "      <td>Male</td>\n",
              "      <td>Under Graduate</td>\n",
              "      <td>Professional</td>\n",
              "      <td>None</td>\n",
              "      <td>None</td>\n",
              "      <td>277022</td>\n",
              "      <td>20442</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>29</td>\n",
              "      <td>Male</td>\n",
              "      <td>Under Graduate</td>\n",
              "      <td>Salaried</td>\n",
              "      <td>None</td>\n",
              "      <td>Entry</td>\n",
              "      <td>348676</td>\n",
              "      <td>24404</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>18</td>\n",
              "      <td>Male</td>\n",
              "      <td>Graduate</td>\n",
              "      <td>Student</td>\n",
              "      <td>None</td>\n",
              "      <td>None</td>\n",
              "      <td>165041</td>\n",
              "      <td>2533</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>26</td>\n",
              "      <td>Male</td>\n",
              "      <td>Post Graduate</td>\n",
              "      <td>Salaried</td>\n",
              "      <td>None</td>\n",
              "      <td>Mid-level 1</td>\n",
              "      <td>348745</td>\n",
              "      <td>19321</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   age gender       education  ... marital_status no_card default\n",
              "0   19   Male        Graduate  ...        Married       0       1\n",
              "1   18   Male  Under Graduate  ...        Married       0       1\n",
              "2   29   Male  Under Graduate  ...        Married       1       1\n",
              "3   18   Male        Graduate  ...        Married       0       1\n",
              "4   26   Male   Post Graduate  ...        Married       1       1\n",
              "\n",
              "[5 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RBCHC0wKjwnd"
      },
      "source": [
        "# **4. Exploratory Data Analysis**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WcyJJFR0hLxd",
        "outputId": "9172d2f5-4f0a-4ed2-f61d-f125d1232b32",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "credit_df.shape"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(50636, 13)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "za1uSEHahL1Y",
        "outputId": "d86106cf-005f-491a-df3e-cecfc3262a6d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "credit_df.info()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 50636 entries, 0 to 50635\n",
            "Data columns (total 13 columns):\n",
            " #   Column             Non-Null Count  Dtype \n",
            "---  ------             --------------  ----- \n",
            " 0   age                50636 non-null  int64 \n",
            " 1   gender             50636 non-null  object\n",
            " 2   education          50636 non-null  object\n",
            " 3   occupation         50636 non-null  object\n",
            " 4   organization_type  50636 non-null  object\n",
            " 5   seniority          50636 non-null  object\n",
            " 6   annual_income      50636 non-null  int64 \n",
            " 7   disposable_income  50636 non-null  int64 \n",
            " 8   house_type         50636 non-null  object\n",
            " 9   vehicle_type       50636 non-null  object\n",
            " 10  marital_status     50636 non-null  object\n",
            " 11  no_card            50636 non-null  int64 \n",
            " 12  default            50636 non-null  int64 \n",
            "dtypes: int64(5), object(8)\n",
            "memory usage: 5.0+ MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eZO2Sgv2hL3h",
        "outputId": "9cf452f9-2895-49c1-c592-2758112ae1be",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "# check for null values\r\n",
        "credit_df.isnull().sum()"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "age                  0\n",
              "gender               0\n",
              "education            0\n",
              "occupation           0\n",
              "organization_type    0\n",
              "seniority            0\n",
              "annual_income        0\n",
              "disposable_income    0\n",
              "house_type           0\n",
              "vehicle_type         0\n",
              "marital_status       0\n",
              "no_card              0\n",
              "default              0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AG-E83GZhL73",
        "outputId": "39cf3e0a-7fb6-4cff-a550-e21f1a2edc62",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        }
      },
      "source": [
        "# see the statistical analysis of the dataset\r\n",
        "credit_df.describe()"
      ],
      "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>age</th>\n",
              "      <th>annual_income</th>\n",
              "      <th>disposable_income</th>\n",
              "      <th>no_card</th>\n",
              "      <th>default</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>50636.000000</td>\n",
              "      <td>50636.000000</td>\n",
              "      <td>50636.000000</td>\n",
              "      <td>50636.000000</td>\n",
              "      <td>50636.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>29.527411</td>\n",
              "      <td>277243.989889</td>\n",
              "      <td>18325.788569</td>\n",
              "      <td>0.509815</td>\n",
              "      <td>0.158425</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>8.816532</td>\n",
              "      <td>153838.973755</td>\n",
              "      <td>12677.864844</td>\n",
              "      <td>0.669883</td>\n",
              "      <td>0.365142</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>18.000000</td>\n",
              "      <td>50000.000000</td>\n",
              "      <td>1000.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>25.000000</td>\n",
              "      <td>154052.250000</td>\n",
              "      <td>8317.750000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>27.000000</td>\n",
              "      <td>258860.500000</td>\n",
              "      <td>15770.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>30.000000</td>\n",
              "      <td>385071.500000</td>\n",
              "      <td>24135.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>64.000000</td>\n",
              "      <td>999844.000000</td>\n",
              "      <td>49999.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                age  annual_income  ...       no_card       default\n",
              "count  50636.000000   50636.000000  ...  50636.000000  50636.000000\n",
              "mean      29.527411  277243.989889  ...      0.509815      0.158425\n",
              "std        8.816532  153838.973755  ...      0.669883      0.365142\n",
              "min       18.000000   50000.000000  ...      0.000000      0.000000\n",
              "25%       25.000000  154052.250000  ...      0.000000      0.000000\n",
              "50%       27.000000  258860.500000  ...      0.000000      0.000000\n",
              "75%       30.000000  385071.500000  ...      1.000000      0.000000\n",
              "max       64.000000  999844.000000  ...      2.000000      1.000000\n",
              "\n",
              "[8 rows x 5 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vwNwZQHQhL9_",
        "outputId": "4c676267-cfcd-4ca3-c5c5-638487aa4a63",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        }
      },
      "source": [
        "# Separate columns having categorical value\r\n",
        "obj_df=credit_df.select_dtypes(include=['object']).copy()\r\n",
        "obj_df.head()"
      ],
      "execution_count": 7,
      "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>gender</th>\n",
              "      <th>education</th>\n",
              "      <th>occupation</th>\n",
              "      <th>organization_type</th>\n",
              "      <th>seniority</th>\n",
              "      <th>house_type</th>\n",
              "      <th>vehicle_type</th>\n",
              "      <th>marital_status</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Male</td>\n",
              "      <td>Graduate</td>\n",
              "      <td>Professional</td>\n",
              "      <td>None</td>\n",
              "      <td>None</td>\n",
              "      <td>Family</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Male</td>\n",
              "      <td>Under Graduate</td>\n",
              "      <td>Professional</td>\n",
              "      <td>None</td>\n",
              "      <td>None</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Male</td>\n",
              "      <td>Under Graduate</td>\n",
              "      <td>Salaried</td>\n",
              "      <td>None</td>\n",
              "      <td>Entry</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Male</td>\n",
              "      <td>Graduate</td>\n",
              "      <td>Student</td>\n",
              "      <td>None</td>\n",
              "      <td>None</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Male</td>\n",
              "      <td>Post Graduate</td>\n",
              "      <td>Salaried</td>\n",
              "      <td>None</td>\n",
              "      <td>Mid-level 1</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  gender       education    occupation  ... house_type vehicle_type marital_status\n",
              "0   Male        Graduate  Professional  ...     Family         None        Married\n",
              "1   Male  Under Graduate  Professional  ...     Rented         None        Married\n",
              "2   Male  Under Graduate      Salaried  ...     Rented         None        Married\n",
              "3   Male        Graduate       Student  ...     Rented         None        Married\n",
              "4   Male   Post Graduate      Salaried  ...     Rented         None        Married\n",
              "\n",
              "[5 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wL2BEPf5hMCh",
        "outputId": "166c4289-0d37-40f1-e693-9ae157df93c8",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "# find unique values present in \"no_card\" column\r\n",
        "credit_df['no_card'].value_counts()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    29855\n",
              "1    15747\n",
              "2     5034\n",
              "Name: no_card, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jxiwFCzokDqb"
      },
      "source": [
        "Here we saw that column **'no_card'** has categorical values but the values are integer. So we'd add that column to the **obj_df** dataframe."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D7wetJyphMFS"
      },
      "source": [
        "# add column \"no_card\" to the dataframe \"obj_df\"\r\n",
        "obj_df['no_card']=credit_df['no_card']"
      ],
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pLHUH_bIhMJn",
        "outputId": "00cdf25e-8877-4f6f-9b32-cad48ae4ad9d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        }
      },
      "source": [
        "obj_df.head()"
      ],
      "execution_count": 10,
      "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>gender</th>\n",
              "      <th>education</th>\n",
              "      <th>occupation</th>\n",
              "      <th>organization_type</th>\n",
              "      <th>seniority</th>\n",
              "      <th>house_type</th>\n",
              "      <th>vehicle_type</th>\n",
              "      <th>marital_status</th>\n",
              "      <th>no_card</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Male</td>\n",
              "      <td>Graduate</td>\n",
              "      <td>Professional</td>\n",
              "      <td>None</td>\n",
              "      <td>None</td>\n",
              "      <td>Family</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Male</td>\n",
              "      <td>Under Graduate</td>\n",
              "      <td>Professional</td>\n",
              "      <td>None</td>\n",
              "      <td>None</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Male</td>\n",
              "      <td>Under Graduate</td>\n",
              "      <td>Salaried</td>\n",
              "      <td>None</td>\n",
              "      <td>Entry</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Male</td>\n",
              "      <td>Graduate</td>\n",
              "      <td>Student</td>\n",
              "      <td>None</td>\n",
              "      <td>None</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Male</td>\n",
              "      <td>Post Graduate</td>\n",
              "      <td>Salaried</td>\n",
              "      <td>None</td>\n",
              "      <td>Mid-level 1</td>\n",
              "      <td>Rented</td>\n",
              "      <td>None</td>\n",
              "      <td>Married</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  gender       education    occupation  ... vehicle_type marital_status no_card\n",
              "0   Male        Graduate  Professional  ...         None        Married       0\n",
              "1   Male  Under Graduate  Professional  ...         None        Married       0\n",
              "2   Male  Under Graduate      Salaried  ...         None        Married       1\n",
              "3   Male        Graduate       Student  ...         None        Married       0\n",
              "4   Male   Post Graduate      Salaried  ...         None        Married       1\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rWp6VsexhML_",
        "outputId": "61be435f-5de9-49d9-c97d-e8b139b11233",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "obj_df.columns"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['gender', 'education', 'occupation', 'organization_type', 'seniority',\n",
              "       'house_type', 'vehicle_type', 'marital_status', 'no_card'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fZDjqr7KhMPc",
        "outputId": "46c4be86-ccc0-4986-b01c-45b79aff5e34",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "# Find number of unique value present in all columns\r\n",
        "obj_df.nunique()"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "gender               2\n",
              "education            4\n",
              "occupation           4\n",
              "organization_type    4\n",
              "seniority            6\n",
              "house_type           4\n",
              "vehicle_type         3\n",
              "marital_status       3\n",
              "no_card              3\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "43v99isRkSdx",
        "outputId": "f9d97511-c8f1-4b56-8774-e96a2114ca3b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "def find_unique(col):\r\n",
        "  print(col,\":\",obj_df[col].unique())\r\n",
        "for col in obj_df.columns:\r\n",
        "  find_unique(col)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "gender : ['Male' 'Female']\n",
            "education : ['Graduate' 'Under Graduate' 'Post Graduate' 'Other']\n",
            "occupation : ['Professional' 'Salaried' 'Student' 'Business']\n",
            "organization_type : ['None' 'Tier 3' 'Tier 2' 'Tier 1']\n",
            "seniority : ['None' 'Entry' 'Mid-level 1' 'Junior' 'Mid-level 2' 'Senior']\n",
            "house_type : ['Family' 'Rented' 'Company provided' 'Owned']\n",
            "vehicle_type : ['None' 'Two Wheeler' 'Four Wheeler']\n",
            "marital_status : ['Married' 'Single' 'Other']\n",
            "no_card : [0 1 2]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rFmGelDqkVua"
      },
      "source": [
        "# **5. Data Visualization**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ybYYrsZqkhPO"
      },
      "source": [
        "**Frequency distribution of columns having categorical values**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T8pmKxqGkSi4"
      },
      "source": [
        "def plot_bar_graph(column_name):\r\n",
        "    ed_count = column_name.value_counts()\r\n",
        "    sns.set(style=\"darkgrid\")\r\n",
        "    sns.barplot(ed_count.index, ed_count.values, alpha=0.9)\r\n",
        "    plt.title('Frequency Distribution of {} Levels using Bar Plot'.format(column_name.name))\r\n",
        "    plt.ylabel('Number of Occurrences', fontsize=12)\r\n",
        "    plt.xlabel('{}'.format(column_name.name), fontsize=12)\r\n",
        "    plt.show() \r\n",
        "    "
      ],
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dTSfmuYXkSlO",
        "outputId": "b76b9d77-ebba-49da-f0e8-846fb6ca475a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "for i in obj_df.columns:\r\n",
        "  plot_bar_graph(obj_df[i])"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/_decorators.py:43: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
            "  FutureWarning\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9RGu949EkSpY",
        "outputId": "ab4604f8-cc01-429a-da19-558e0d290fc0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        }
      },
      "source": [
        "numeric_df=credit_df.select_dtypes(include=['int64']).copy()\r\n",
        "numeric_df.drop(columns=['no_card','default'],axis=1,inplace=True)\r\n",
        "numeric_df.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>age</th>\n",
              "      <th>annual_income</th>\n",
              "      <th>disposable_income</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19</td>\n",
              "      <td>186319</td>\n",
              "      <td>21625</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>18</td>\n",
              "      <td>277022</td>\n",
              "      <td>20442</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>29</td>\n",
              "      <td>348676</td>\n",
              "      <td>24404</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>18</td>\n",
              "      <td>165041</td>\n",
              "      <td>2533</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>26</td>\n",
              "      <td>348745</td>\n",
              "      <td>19321</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   age  annual_income  disposable_income\n",
              "0   19         186319              21625\n",
              "1   18         277022              20442\n",
              "2   29         348676              24404\n",
              "3   18         165041               2533\n",
              "4   26         348745              19321"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UUg8X98LkSr-",
        "outputId": "0a392925-7358-447d-a5a5-af4d92777241",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 990
        }
      },
      "source": [
        "def distplot_graph(column_name):\r\n",
        "    sns.set(style=\"darkgrid\")\r\n",
        "    sns.distplot(numeric_df[column_name], color=\"g\")\r\n",
        "    plt.title('Normal Distribution of {} Levels using Bar Plot'.format(column_name))    \r\n",
        "    plt.xlabel('{}'.format(column_name), fontsize=12)\r\n",
        "    plt.show() \r\n",
        "    \r\n",
        "for col in numeric_df.columns:\r\n",
        "  distplot_graph(col)"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2557: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TvKu-QLBkuk5"
      },
      "source": [
        "# **6. Data Preprocessing**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JHbeHDF5kSvY",
        "outputId": "c6d1768f-9181-42a8-df6a-4ac2b7a2db34",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 215
        }
      },
      "source": [
        "# encode the categorical values into numeric values\r\n",
        "le=LabelEncoder()\r\n",
        "for col in obj_df.columns:\r\n",
        "    credit_df[col]=le.fit_transform(credit_df[col])\r\n",
        "credit_df.head()"
      ],
      "execution_count": 18,
      "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>age</th>\n",
              "      <th>gender</th>\n",
              "      <th>education</th>\n",
              "      <th>occupation</th>\n",
              "      <th>organization_type</th>\n",
              "      <th>seniority</th>\n",
              "      <th>annual_income</th>\n",
              "      <th>disposable_income</th>\n",
              "      <th>house_type</th>\n",
              "      <th>vehicle_type</th>\n",
              "      <th>marital_status</th>\n",
              "      <th>no_card</th>\n",
              "      <th>default</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>186319</td>\n",
              "      <td>21625</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>277022</td>\n",
              "      <td>20442</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>29</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>348676</td>\n",
              "      <td>24404</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>165041</td>\n",
              "      <td>2533</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>26</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>348745</td>\n",
              "      <td>19321</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   age  gender  education  ...  marital_status  no_card  default\n",
              "0   19       1          0  ...               0        0        1\n",
              "1   18       1          3  ...               0        0        1\n",
              "2   29       1          3  ...               0        1        1\n",
              "3   18       1          0  ...               0        0        1\n",
              "4   26       1          2  ...               0        1        1\n",
              "\n",
              "[5 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hldL1wd1kzoQ"
      },
      "source": [
        "### **6.1. Outlier Detection**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tAb77yZ9fzMG",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "78f1b9fd-79e3-448c-9ffa-4df430ce8953"
      },
      "source": [
        "# find outlier in all columns\r\n",
        "for i in credit_df.select_dtypes(include=['float64','int64']).columns:\r\n",
        "  max_thresold=credit_df[i].quantile(0.95)\r\n",
        "  min_thresold=credit_df[i].quantile(0.05)\r\n",
        "  credit_df_no_outlier=credit_df[(credit_df[i] < max_thresold) & (credit_df[i] > min_thresold)].shape\r\n",
        "  print(\" outlier in \",i,\"is\" ,int(((credit_df.shape[0]-credit_df_no_outlier[0])/credit_df.shape[0])*100),\"%\")"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            " outlier in  age is 10 %\n",
            " outlier in  gender is 100 %\n",
            " outlier in  education is 60 %\n",
            " outlier in  occupation is 25 %\n",
            " outlier in  organization_type is 88 %\n",
            " outlier in  seniority is 53 %\n",
            " outlier in  annual_income is 10 %\n",
            " outlier in  disposable_income is 10 %\n",
            " outlier in  house_type is 67 %\n",
            " outlier in  vehicle_type is 100 %\n",
            " outlier in  marital_status is 95 %\n",
            " outlier in  no_card is 68 %\n",
            " outlier in  default is 100 %\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pafL7Li0jyXW"
      },
      "source": [
        "# remove outliers from columns having nearly 10% outlier\r\n",
        "max_thresold_annual_income=credit_df[\"annual_income\"].quantile(0.95)\r\n",
        "min_thresold_annual_income=credit_df[\"annual_income\"].quantile(0.05)\r\n",
        "max_thresold_disposable_income=credit_df[\"disposable_income\"].quantile(0.95)\r\n",
        "min_thresold_disposable_income=credit_df[\"disposable_income\"].quantile(0.05)\r\n",
        "credit_df_no_outlier=credit_df[(credit_df[\"annual_income\"] < max_thresold_annual_income) & (credit_df[\"annual_income\"] > min_thresold_annual_income) &  (credit_df[\"disposable_income\"] < max_thresold_disposable_income) & (credit_df[\"disposable_income\"] > min_thresold_disposable_income)]\r\n"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LyTC0FVqk9MJ",
        "outputId": "f4a3220f-57c8-4bdf-ba02-72ae6b5319bc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 215
        }
      },
      "source": [
        "credit_df_no_outlier.head()"
      ],
      "execution_count": 21,
      "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>age</th>\n",
              "      <th>gender</th>\n",
              "      <th>education</th>\n",
              "      <th>occupation</th>\n",
              "      <th>organization_type</th>\n",
              "      <th>seniority</th>\n",
              "      <th>annual_income</th>\n",
              "      <th>disposable_income</th>\n",
              "      <th>house_type</th>\n",
              "      <th>vehicle_type</th>\n",
              "      <th>marital_status</th>\n",
              "      <th>no_card</th>\n",
              "      <th>default</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>186319</td>\n",
              "      <td>21625</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>277022</td>\n",
              "      <td>20442</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>29</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>348676</td>\n",
              "      <td>24404</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>165041</td>\n",
              "      <td>2533</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>26</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>348745</td>\n",
              "      <td>19321</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   age  gender  education  ...  marital_status  no_card  default\n",
              "0   19       1          0  ...               0        0        1\n",
              "1   18       1          3  ...               0        0        1\n",
              "2   29       1          3  ...               0        1        1\n",
              "3   18       1          0  ...               0        0        1\n",
              "4   26       1          2  ...               0        1        1\n",
              "\n",
              "[5 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Nlv4-F9Vk9OO",
        "outputId": "1a5db247-2657-4a65-a69a-3617990e65a1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 215
        }
      },
      "source": [
        "credit_df_no_outlier.reset_index(inplace=True)\r\n",
        "credit_df_no_outlier.head()"
      ],
      "execution_count": 22,
      "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>index</th>\n",
              "      <th>age</th>\n",
              "      <th>gender</th>\n",
              "      <th>education</th>\n",
              "      <th>occupation</th>\n",
              "      <th>organization_type</th>\n",
              "      <th>seniority</th>\n",
              "      <th>annual_income</th>\n",
              "      <th>disposable_income</th>\n",
              "      <th>house_type</th>\n",
              "      <th>vehicle_type</th>\n",
              "      <th>marital_status</th>\n",
              "      <th>no_card</th>\n",
              "      <th>default</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>19</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>186319</td>\n",
              "      <td>21625</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>277022</td>\n",
              "      <td>20442</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>29</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>348676</td>\n",
              "      <td>24404</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>165041</td>\n",
              "      <td>2533</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>26</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>348745</td>\n",
              "      <td>19321</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   index  age  gender  ...  marital_status  no_card  default\n",
              "0      0   19       1  ...               0        0        1\n",
              "1      1   18       1  ...               0        0        1\n",
              "2      2   29       1  ...               0        1        1\n",
              "3      3   18       1  ...               0        0        1\n",
              "4      4   26       1  ...               0        1        1\n",
              "\n",
              "[5 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rXZzS7C0k9Sb",
        "outputId": "ba9815aa-288a-40d3-8063-b1967d798244",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 215
        }
      },
      "source": [
        "credit_df_no_outlier=credit_df_no_outlier.drop(columns=[\"index\"],axis=1)\r\n",
        "credit_df_no_outlier.head()"
      ],
      "execution_count": 23,
      "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>age</th>\n",
              "      <th>gender</th>\n",
              "      <th>education</th>\n",
              "      <th>occupation</th>\n",
              "      <th>organization_type</th>\n",
              "      <th>seniority</th>\n",
              "      <th>annual_income</th>\n",
              "      <th>disposable_income</th>\n",
              "      <th>house_type</th>\n",
              "      <th>vehicle_type</th>\n",
              "      <th>marital_status</th>\n",
              "      <th>no_card</th>\n",
              "      <th>default</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>19</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>186319</td>\n",
              "      <td>21625</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>277022</td>\n",
              "      <td>20442</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>29</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>348676</td>\n",
              "      <td>24404</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>165041</td>\n",
              "      <td>2533</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>26</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>348745</td>\n",
              "      <td>19321</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   age  gender  education  ...  marital_status  no_card  default\n",
              "0   19       1          0  ...               0        0        1\n",
              "1   18       1          3  ...               0        0        1\n",
              "2   29       1          3  ...               0        1        1\n",
              "3   18       1          0  ...               0        0        1\n",
              "4   26       1          2  ...               0        1        1\n",
              "\n",
              "[5 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "62-SWXfblF01"
      },
      "source": [
        "### **6.2. Checking Data Correlation**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LAMnJUt0k9T5",
        "outputId": "20666f6f-e942-49b1-e014-bdc811fcf83a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 694
        }
      },
      "source": [
        "# correlation matrix of DataFrame\r\n",
        "plt.figure(figsize=(20,10))\r\n",
        "corn=credit_df_no_outlier.corr()\r\n",
        "sns.heatmap(corn,annot=True,cmap=\"BuPu\")"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f03aa73cf90>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-tctRXD7k9Xw"
      },
      "source": [
        "X=credit_df_no_outlier.drop(columns=[\"default\"])\r\n",
        "y=credit_df_no_outlier[\"default\"]"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2gF4zUpVlH83"
      },
      "source": [
        "# scale the DataFrame\r\n",
        "scalar=StandardScaler()\r\n",
        "X=scalar.fit_transform(X)\r\n",
        "X=pd.DataFrame(X)"
      ],
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GZTl-HLXlRXU"
      },
      "source": [
        "# **7. Split the dataset**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gBvkn95alUel"
      },
      "source": [
        "Here before going for <code>hyper-parameter</code> tuning we'll apply <code>LazyClassifier</code> algorithm to find which algorithm will give how much of accuracy."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "375EfjAJlH_W"
      },
      "source": [
        "x_train,x_test,y_train,y_test=train_test_split(X,y,test_size=0.25,random_state=21)"
      ],
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HaGuRDpjla2X",
        "outputId": "5b7bfce0-62bf-451c-9844-9858539a4dc3",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "print(x_train.shape, y_train.shape)\r\n",
        "print(x_test.shape, y_test.shape)"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(30985, 12) (30985,)\n",
            "(10329, 12) (10329,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NCXJOL7ElfSH"
      },
      "source": [
        "# **8. Apply Lazypredict**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GxhfumP-mJnb"
      },
      "source": [
        "As the dataset is too big when we apply <code>LazyClassifier</code> algorithm our execution may crash due to less RAM. Google colab provides 12gb RAM for free but to execute this algorithm with big dataset we need more RAM. So i'm using Google colab pro to execute this algorithm. Don't worry if you have not Colab pro. I'll provide the output in a csv file."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_bgfDKe9la4p",
        "outputId": "ac0cd5b2-7650-48be-acd6-53b8188afc1e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "!pip install lazypredict==0.2.7"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting lazypredict==0.2.7\n",
            "  Downloading https://files.pythonhosted.org/packages/5f/db/1566dca1050ea74e9474dca0f1e7bbffcb0c3e694cf92e7e6e7ef9fca3af/lazypredict-0.2.7-py2.py3-none-any.whl\n",
            "Requirement already satisfied: Click>=7.0 in /usr/local/lib/python3.7/dist-packages (from lazypredict==0.2.7) (7.1.2)\n",
            "Installing collected packages: lazypredict\n",
            "Successfully installed lazypredict-0.2.7\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FKQH5U0rla9U",
        "outputId": "a96f9a75-76a0-4d19-830e-e49cbd1df45f",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "!pip install lightgbm"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: lightgbm in /usr/local/lib/python3.7/dist-packages (2.2.3)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from lightgbm) (1.19.5)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.7/dist-packages (from lightgbm) (0.22.2.post1)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from lightgbm) (1.4.1)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->lightgbm) (1.0.1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U9TOIaQ6la_h",
        "outputId": "d768e50c-ea9c-4f0e-dc70-259fd8bf5a8d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "import lazypredict\r\n",
        "from lazypredict.Supervised import LazyClassifier"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:144: FutureWarning: The sklearn.utils.testing 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.utils. Anything that cannot be imported from sklearn.utils is now part of the private API.\n",
            "  warnings.warn(message, FutureWarning)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_cCQV0rrlbDO",
        "outputId": "a6824244-8925-45e2-f7a7-0684eefee88a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "clf= LazyClassifier(verbose=0,ignore_warnings=True,custom_metric=None)\r\n",
        "train,test=clf.fit(x_train,x_test,y_train,y_test)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "100%|██████████| 30/30 [02:34<00:00,  5.15s/it]\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PVHjunXglIDk",
        "outputId": "34ca2d5c-0107-4c01-9a89-1f8a182e4353",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 876
        }
      },
      "source": [
        "train"
      ],
      "execution_count": 33,
      "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>Accuracy</th>\n",
              "      <th>Balanced Accuracy</th>\n",
              "      <th>ROC AUC</th>\n",
              "      <th>F1 Score</th>\n",
              "      <th>Time Taken</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Model</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>NearestCentroid</th>\n",
              "      <td>0.62</td>\n",
              "      <td>0.64</td>\n",
              "      <td>0.64</td>\n",
              "      <td>0.67</td>\n",
              "      <td>0.06</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LGBMClassifier</th>\n",
              "      <td>0.86</td>\n",
              "      <td>0.63</td>\n",
              "      <td>0.63</td>\n",
              "      <td>0.84</td>\n",
              "      <td>0.49</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>RandomForestClassifier</th>\n",
              "      <td>0.86</td>\n",
              "      <td>0.62</td>\n",
              "      <td>0.62</td>\n",
              "      <td>0.83</td>\n",
              "      <td>3.53</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>XGBClassifier</th>\n",
              "      <td>0.86</td>\n",
              "      <td>0.61</td>\n",
              "      <td>0.61</td>\n",
              "      <td>0.83</td>\n",
              "      <td>2.18</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>DecisionTreeClassifier</th>\n",
              "      <td>0.78</td>\n",
              "      <td>0.61</td>\n",
              "      <td>0.61</td>\n",
              "      <td>0.79</td>\n",
              "      <td>0.23</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>BaggingClassifier</th>\n",
              "      <td>0.85</td>\n",
              "      <td>0.60</td>\n",
              "      <td>0.60</td>\n",
              "      <td>0.82</td>\n",
              "      <td>1.12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>AdaBoostClassifier</th>\n",
              "      <td>0.86</td>\n",
              "      <td>0.60</td>\n",
              "      <td>0.60</td>\n",
              "      <td>0.83</td>\n",
              "      <td>1.15</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>ExtraTreesClassifier</th>\n",
              "      <td>0.85</td>\n",
              "      <td>0.60</td>\n",
              "      <td>0.60</td>\n",
              "      <td>0.82</td>\n",
              "      <td>2.49</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>ExtraTreeClassifier</th>\n",
              "      <td>0.78</td>\n",
              "      <td>0.59</td>\n",
              "      <td>0.59</td>\n",
              "      <td>0.78</td>\n",
              "      <td>0.05</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LabelSpreading</th>\n",
              "      <td>0.79</td>\n",
              "      <td>0.58</td>\n",
              "      <td>0.58</td>\n",
              "      <td>0.78</td>\n",
              "      <td>54.39</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>KNeighborsClassifier</th>\n",
              "      <td>0.84</td>\n",
              "      <td>0.58</td>\n",
              "      <td>0.58</td>\n",
              "      <td>0.81</td>\n",
              "      <td>2.17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LabelPropagation</th>\n",
              "      <td>0.79</td>\n",
              "      <td>0.58</td>\n",
              "      <td>0.58</td>\n",
              "      <td>0.78</td>\n",
              "      <td>39.02</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>PassiveAggressiveClassifier</th>\n",
              "      <td>0.76</td>\n",
              "      <td>0.56</td>\n",
              "      <td>0.56</td>\n",
              "      <td>0.76</td>\n",
              "      <td>0.07</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>QuadraticDiscriminantAnalysis</th>\n",
              "      <td>0.83</td>\n",
              "      <td>0.55</td>\n",
              "      <td>0.55</td>\n",
              "      <td>0.79</td>\n",
              "      <td>0.09</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>GaussianNB</th>\n",
              "      <td>0.85</td>\n",
              "      <td>0.53</td>\n",
              "      <td>0.53</td>\n",
              "      <td>0.79</td>\n",
              "      <td>0.08</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LogisticRegression</th>\n",
              "      <td>0.85</td>\n",
              "      <td>0.52</td>\n",
              "      <td>0.52</td>\n",
              "      <td>0.78</td>\n",
              "      <td>0.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>CalibratedClassifierCV</th>\n",
              "      <td>0.85</td>\n",
              "      <td>0.52</td>\n",
              "      <td>0.52</td>\n",
              "      <td>0.78</td>\n",
              "      <td>7.69</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>SVC</th>\n",
              "      <td>0.85</td>\n",
              "      <td>0.52</td>\n",
              "      <td>0.52</td>\n",
              "      <td>0.78</td>\n",
              "      <td>37.02</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Perceptron</th>\n",
              "      <td>0.80</td>\n",
              "      <td>0.52</td>\n",
              "      <td>0.52</td>\n",
              "      <td>0.77</td>\n",
              "      <td>0.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>DummyClassifier</th>\n",
              "      <td>0.74</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.03</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LinearDiscriminantAnalysis</th>\n",
              "      <td>0.84</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.77</td>\n",
              "      <td>0.13</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LinearSVC</th>\n",
              "      <td>0.84</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.77</td>\n",
              "      <td>1.84</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>CheckingClassifier</th>\n",
              "      <td>0.84</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.77</td>\n",
              "      <td>0.04</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>RidgeClassifier</th>\n",
              "      <td>0.84</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.77</td>\n",
              "      <td>0.06</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>RidgeClassifierCV</th>\n",
              "      <td>0.84</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.77</td>\n",
              "      <td>0.08</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>SGDClassifier</th>\n",
              "      <td>0.84</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.77</td>\n",
              "      <td>0.15</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>BernoulliNB</th>\n",
              "      <td>0.84</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.50</td>\n",
              "      <td>0.77</td>\n",
              "      <td>0.06</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                               Accuracy  ...  Time Taken\n",
              "Model                                    ...            \n",
              "NearestCentroid                    0.62  ...        0.06\n",
              "LGBMClassifier                     0.86  ...        0.49\n",
              "RandomForestClassifier             0.86  ...        3.53\n",
              "XGBClassifier                      0.86  ...        2.18\n",
              "DecisionTreeClassifier             0.78  ...        0.23\n",
              "BaggingClassifier                  0.85  ...        1.12\n",
              "AdaBoostClassifier                 0.86  ...        1.15\n",
              "ExtraTreesClassifier               0.85  ...        2.49\n",
              "ExtraTreeClassifier                0.78  ...        0.05\n",
              "LabelSpreading                     0.79  ...       54.39\n",
              "KNeighborsClassifier               0.84  ...        2.17\n",
              "LabelPropagation                   0.79  ...       39.02\n",
              "PassiveAggressiveClassifier        0.76  ...        0.07\n",
              "QuadraticDiscriminantAnalysis      0.83  ...        0.09\n",
              "GaussianNB                         0.85  ...        0.08\n",
              "LogisticRegression                 0.85  ...        0.10\n",
              "CalibratedClassifierCV             0.85  ...        7.69\n",
              "SVC                                0.85  ...       37.02\n",
              "Perceptron                         0.80  ...        0.10\n",
              "DummyClassifier                    0.74  ...        0.03\n",
              "LinearDiscriminantAnalysis         0.84  ...        0.13\n",
              "LinearSVC                          0.84  ...        1.84\n",
              "CheckingClassifier                 0.84  ...        0.04\n",
              "RidgeClassifier                    0.84  ...        0.06\n",
              "RidgeClassifierCV                  0.84  ...        0.08\n",
              "SGDClassifier                      0.84  ...        0.15\n",
              "BernoulliNB                        0.84  ...        0.06\n",
              "\n",
              "[27 rows x 5 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4YeuULmtmVfn"
      },
      "source": [
        "NOTE: If the execution got failed or crashed then run the below cell to see the output of upper cells and make comment the above five cells"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p74Zh0NOlIF6"
      },
      "source": [
        "#  all_algorithm_df=pd.read_csv(\"lazypredict_algo.csv\")\r\n",
        "#  all_algorithm_df"
      ],
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HeNEq2tinKje"
      },
      "source": [
        "# **9. Hyperparameter Tuning**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u_E9Gh_KlIKV"
      },
      "source": [
        "# this code is to show how much time required to train the model using different algorithms\r\n",
        "def timer(start_time= None):\r\n",
        "  if not start_time:\r\n",
        "    start_time=datetime.now()\r\n",
        "    return start_time\r\n",
        "  elif start_time:\r\n",
        "    thour,temp_sec=divmod((datetime.now()-start_time).total_seconds(),3600)\r\n",
        "    tmin,tsec=divmod(temp_sec,60)\r\n",
        "    print('\\n Time taken: %i hours %i minutes and %s seconds. '% (thour,tmin,round(tsec,2)))"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5ZxHP8p3nNr1"
      },
      "source": [
        "# parameters of all classification algorithms respectively\r\n",
        "model_param={    \r\n",
        "    'random_forest':{\r\n",
        "        'model':RandomForestClassifier(),\r\n",
        "        'params':{\r\n",
        "            'n_estimators':[100,1000]\r\n",
        "        }\r\n",
        "    },\r\n",
        "    'LGBMClassifier':{\r\n",
        "        'model':lgb.LGBMClassifier(),\r\n",
        "        'params':{\r\n",
        "            'max_depth': [20, 30],\r\n",
        "            'max_bin':[255,300],                      \r\n",
        "            'n_estimators' : [10,100,1000]\r\n",
        "        }\r\n",
        "    },\r\n",
        "    'ExtraTreesClfr':{\r\n",
        "        'model':ExtraTreesClassifier(),\r\n",
        "        'params':{\r\n",
        "            \"max_depth\": [None],\r\n",
        "            \"max_features\": [8,12],\r\n",
        "            \"min_samples_split\": [10,20,30],\r\n",
        "            \"bootstrap\": [False, True],\r\n",
        "            \"n_estimators\" :[10,100,200],\r\n",
        "            \"criterion\": [\"gini\"]\r\n",
        "        }\r\n",
        "    }\r\n",
        "}     "
      ],
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ewpeGc-GqraM"
      },
      "source": [
        "This below cell may take a little more time to execute."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZtNpN54xnNug",
        "outputId": "7497e5aa-8c3f-4cee-fdbd-786296ae883a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "start_time=timer(None)\r\n",
        "scores=[]\r\n",
        "for model_name,mp in model_param.items():\r\n",
        "    # Apply GridSearchCV\r\n",
        "    rs=GridSearchCV(mp['model'],mp['params'],cv=5,return_train_score=False)\r\n",
        "    rs.fit(X,y)\r\n",
        "    print( model_name,\" : successfully execute\")\r\n",
        "    scores.append({\r\n",
        "        'model':model_name, # it'll retrive the best model name\r\n",
        "        \"best_score\":rs.best_score_, # it'll retrive the best accuracy score\r\n",
        "        'best_params':rs.best_params_ # it'll retrive the best parameter\r\n",
        "    })\r\n",
        "timer(start_time)"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "random_forest  : successfully execute\n",
            "LGBMClassifier  : successfully execute\n",
            "ExtraTreesClfr  : successfully execute\n",
            "\n",
            " Time taken: 0 hours 15 minutes and 44.74 seconds. \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n5t-oPqlnNyd",
        "outputId": "12a25993-3093-4310-edb4-b777358f094a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        }
      },
      "source": [
        "df=pd.DataFrame(scores,columns=['model','best_score','best_params'])\r\n",
        "df"
      ],
      "execution_count": 38,
      "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>model</th>\n",
              "      <th>best_score</th>\n",
              "      <th>best_params</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>random_forest</td>\n",
              "      <td>0.86</td>\n",
              "      <td>{'n_estimators': 1000}</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>LGBMClassifier</td>\n",
              "      <td>0.86</td>\n",
              "      <td>{'max_bin': 300, 'max_depth': 20, 'n_estimator...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ExtraTreesClfr</td>\n",
              "      <td>0.86</td>\n",
              "      <td>{'bootstrap': True, 'criterion': 'gini', 'max_...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            model  ...                                        best_params\n",
              "0   random_forest  ...                             {'n_estimators': 1000}\n",
              "1  LGBMClassifier  ...  {'max_bin': 300, 'max_depth': 20, 'n_estimator...\n",
              "2  ExtraTreesClfr  ...  {'bootstrap': True, 'criterion': 'gini', 'max_...\n",
              "\n",
              "[3 rows x 3 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QOCAopTDnN6M",
        "outputId": "0110aae7-4593-47e6-c5b5-4c049b010e1d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "# Find the parameter of LGBMClassifier\r\n",
        "# rs.best_params_\r\n",
        "# rs.best_estimator_\r\n",
        "scores[1]"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'best_params': {'max_bin': 300, 'max_depth': 20, 'n_estimators': 100},\n",
              " 'best_score': 0.86433061634238,\n",
              " 'model': 'LGBMClassifier'}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Yr801SNEu8Ld"
      },
      "source": [
        "Now we are going to split the dataset into two parts where 90% of data will be use in KFold Cross validation and rest 10% data will be use for sample prediction "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yh-tFqyolINM"
      },
      "source": [
        "# Split the dataset\r\n",
        "x_train,x_test,y_train,y_test=train_test_split(X,y,test_size=0.10,random_state=21)"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7KdOsogllIeD",
        "outputId": "403fdf08-6246-4304-e81a-b528631b8428",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        }
      },
      "source": [
        "# reset the index of x_train\r\n",
        "x_train.reset_index(inplace=True)\r\n",
        "# drop \"index\" column\r\n",
        "x_train.drop(columns=['index'],axis=1,inplace=True)\r\n",
        "x_train.tail()"
      ],
      "execution_count": 41,
      "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>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>37177</th>\n",
              "      <td>1.41</td>\n",
              "      <td>0.63</td>\n",
              "      <td>0.85</td>\n",
              "      <td>0.57</td>\n",
              "      <td>-0.61</td>\n",
              "      <td>-0.29</td>\n",
              "      <td>0.21</td>\n",
              "      <td>0.28</td>\n",
              "      <td>0.71</td>\n",
              "      <td>-0.48</td>\n",
              "      <td>-0.82</td>\n",
              "      <td>-0.76</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37178</th>\n",
              "      <td>0.28</td>\n",
              "      <td>-1.59</td>\n",
              "      <td>-0.04</td>\n",
              "      <td>0.57</td>\n",
              "      <td>-0.61</td>\n",
              "      <td>-0.29</td>\n",
              "      <td>1.11</td>\n",
              "      <td>0.63</td>\n",
              "      <td>-2.38</td>\n",
              "      <td>-0.48</td>\n",
              "      <td>1.28</td>\n",
              "      <td>0.73</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37179</th>\n",
              "      <td>1.75</td>\n",
              "      <td>0.63</td>\n",
              "      <td>0.85</td>\n",
              "      <td>-1.71</td>\n",
              "      <td>-0.61</td>\n",
              "      <td>1.08</td>\n",
              "      <td>0.08</td>\n",
              "      <td>0.58</td>\n",
              "      <td>0.71</td>\n",
              "      <td>-0.48</td>\n",
              "      <td>-0.82</td>\n",
              "      <td>-0.76</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37180</th>\n",
              "      <td>1.75</td>\n",
              "      <td>0.63</td>\n",
              "      <td>-0.94</td>\n",
              "      <td>0.57</td>\n",
              "      <td>0.22</td>\n",
              "      <td>-0.29</td>\n",
              "      <td>-0.82</td>\n",
              "      <td>0.35</td>\n",
              "      <td>-2.38</td>\n",
              "      <td>-2.38</td>\n",
              "      <td>-0.82</td>\n",
              "      <td>-0.76</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37181</th>\n",
              "      <td>-0.06</td>\n",
              "      <td>0.63</td>\n",
              "      <td>-0.94</td>\n",
              "      <td>-0.57</td>\n",
              "      <td>-0.61</td>\n",
              "      <td>1.08</td>\n",
              "      <td>-1.50</td>\n",
              "      <td>0.27</td>\n",
              "      <td>0.71</td>\n",
              "      <td>1.41</td>\n",
              "      <td>1.28</td>\n",
              "      <td>-0.76</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         0     1     2     3     4     5     6    7     8     9     10    11\n",
              "37177  1.41  0.63  0.85  0.57 -0.61 -0.29  0.21 0.28  0.71 -0.48 -0.82 -0.76\n",
              "37178  0.28 -1.59 -0.04  0.57 -0.61 -0.29  1.11 0.63 -2.38 -0.48  1.28  0.73\n",
              "37179  1.75  0.63  0.85 -1.71 -0.61  1.08  0.08 0.58  0.71 -0.48 -0.82 -0.76\n",
              "37180  1.75  0.63 -0.94  0.57  0.22 -0.29 -0.82 0.35 -2.38 -2.38 -0.82 -0.76\n",
              "37181 -0.06  0.63 -0.94 -0.57 -0.61  1.08 -1.50 0.27  0.71  1.41  1.28 -0.76"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RCBSYMvBu589",
        "outputId": "806de899-11bf-40f4-c40e-ce3519d9bea5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        }
      },
      "source": [
        "y_train=pd.DataFrame(y_train)\r\n",
        "y_train.reset_index(inplace=True)\r\n",
        "y_train.drop(columns=['index'],axis=1,inplace=True)\r\n",
        "y_train.tail()"
      ],
      "execution_count": 42,
      "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>default</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>37177</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37178</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37179</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37180</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37181</th>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       default\n",
              "37177        0\n",
              "37178        0\n",
              "37179        0\n",
              "37180        0\n",
              "37181        0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JmSOBxlwu5_p"
      },
      "source": [
        "# convert \"y_train\" dataframe into an array \r\n",
        "y_train=np.array(y_train)"
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "41rx4k39vQoM"
      },
      "source": [
        "# **10. Model Building**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "55KulOAYu6Dk",
        "outputId": "8e9ff17a-a972-41e3-fd9f-3f16ad9b1bfb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "# Hide warning\r\n",
        "warnings.filterwarnings('ignore')\r\n",
        "\r\n",
        "# Apply KFold\r\n",
        "skf=KFold(n_splits=5)\r\n",
        "LGBMClf= lgb.LGBMClassifier(max_bin= 300, max_depth = 20, n_estimators = 100)\r\n",
        "\r\n",
        "for train_index,test_index in skf.split(x_train,y_train):\r\n",
        "    x_train_kf,x_test_kf,y_train_kf,y_test_kf = x_train.loc[train_index],x_train.loc[test_index],y_train[train_index],y_train[test_index]\r\n",
        "    LGBMClf.fit(x_train_kf, y_train_kf)\r\n",
        "\r\n",
        "# Find Accuracy\r\n",
        "print(\"============================================================\")\r\n",
        "print(\"Training accuracy of data is :\", LGBMClf.score(x_train_kf, y_train_kf))\r\n",
        "print(\"============================================================\")\r\n",
        "print(\"Testng accuracy of data is :\", LGBMClf.score(x_test_kf, y_test_kf))\r\n",
        "print(\"============================================================\")"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "============================================================\n",
            "Training accuracy of data is : 0.8817319975795065\n",
            "============================================================\n",
            "Testng accuracy of data is : 0.865115653577192\n",
            "============================================================\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "h27oxmCnvWiQ"
      },
      "source": [
        "# **11. Save the Model**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7z9DzyoNu6Fg",
        "outputId": "d2fd3a37-22d1-46e1-b3c9-a93426892b66",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "#Saving Scikitlearn models\r\n",
        "import joblib\r\n",
        "joblib.dump(LGBMClf, \"credit_risk_analyzer.pkl\")"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['credit_risk_analyzer.pkl']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "89Qu115uu6J1",
        "outputId": "a45c8829-0e86-43b8-d9a5-4aeeeac56e30",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "# load the model\r\n",
        "model=joblib.load(\"credit_risk_analyzer.pkl\")\r\n",
        "model"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LGBMClassifier(boosting_type='gbdt', class_weight=None, colsample_bytree=1.0,\n",
              "               importance_type='split', learning_rate=0.1, max_bin=300,\n",
              "               max_depth=20, min_child_samples=20, min_child_weight=0.001,\n",
              "               min_split_gain=0.0, n_estimators=100, n_jobs=-1, num_leaves=31,\n",
              "               objective=None, random_state=None, reg_alpha=0.0, reg_lambda=0.0,\n",
              "               silent=True, subsample=1.0, subsample_for_bin=200000,\n",
              "               subsample_freq=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fc61p-eCvbdw"
      },
      "source": [
        "# **12. Find Prediction**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bt6nq3ayu6NK"
      },
      "source": [
        "y_pred=model.predict(x_test)"
      ],
      "execution_count": 47,
      "outputs": []
    }
  ]
}