defaulter.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "defaulter.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "metadata": {
        "id": "DBK6Z7jFxZCI",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Loading the Dataset"
      ]
    },
    {
      "metadata": {
        "id": "IxOtj-1saaxN",
        "colab_type": "code",
        "outputId": "2f4bd4db-c497-46a9-d137-dd3f3b74398d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "df = pd.read_csv(\"Avail_hmeq.csv\")\n",
        "df.head()"
      ],
      "execution_count": 0,
      "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>BAD</th>\n",
              "      <th>LOAN</th>\n",
              "      <th>MORTDUE</th>\n",
              "      <th>VALUE</th>\n",
              "      <th>REASON</th>\n",
              "      <th>JOB</th>\n",
              "      <th>YOJ</th>\n",
              "      <th>DEROG</th>\n",
              "      <th>DELINQ</th>\n",
              "      <th>CLAGE</th>\n",
              "      <th>NINQ</th>\n",
              "      <th>CLNO</th>\n",
              "      <th>DEBTINC</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1100</td>\n",
              "      <td>25860.0</td>\n",
              "      <td>39025.0</td>\n",
              "      <td>HomeImp</td>\n",
              "      <td>Other</td>\n",
              "      <td>10.5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>94.366667</td>\n",
              "      <td>1.0</td>\n",
              "      <td>9.0</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>1300</td>\n",
              "      <td>70053.0</td>\n",
              "      <td>68400.0</td>\n",
              "      <td>HomeImp</td>\n",
              "      <td>Other</td>\n",
              "      <td>7.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>121.833333</td>\n",
              "      <td>0.0</td>\n",
              "      <td>14.0</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>1500</td>\n",
              "      <td>13500.0</td>\n",
              "      <td>16700.0</td>\n",
              "      <td>HomeImp</td>\n",
              "      <td>Other</td>\n",
              "      <td>4.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>149.466667</td>\n",
              "      <td>1.0</td>\n",
              "      <td>10.0</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>1500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>1700</td>\n",
              "      <td>97800.0</td>\n",
              "      <td>112000.0</td>\n",
              "      <td>HomeImp</td>\n",
              "      <td>Office</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>93.333333</td>\n",
              "      <td>0.0</td>\n",
              "      <td>14.0</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   BAD  LOAN  MORTDUE     VALUE   REASON     JOB   YOJ  DEROG  DELINQ  \\\n",
              "0    1  1100  25860.0   39025.0  HomeImp   Other  10.5    0.0     0.0   \n",
              "1    1  1300  70053.0   68400.0  HomeImp   Other   7.0    0.0     2.0   \n",
              "2    1  1500  13500.0   16700.0  HomeImp   Other   4.0    0.0     0.0   \n",
              "3    1  1500      NaN       NaN      NaN     NaN   NaN    NaN     NaN   \n",
              "4    0  1700  97800.0  112000.0  HomeImp  Office   3.0    0.0     0.0   \n",
              "\n",
              "        CLAGE  NINQ  CLNO  DEBTINC  \n",
              "0   94.366667   1.0   9.0      NaN  \n",
              "1  121.833333   0.0  14.0      NaN  \n",
              "2  149.466667   1.0  10.0      NaN  \n",
              "3         NaN   NaN   NaN      NaN  \n",
              "4   93.333333   0.0  14.0      NaN  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "metadata": {
        "id": "rpwgoO5Txj1a",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### First understand the Dataset"
      ]
    },
    {
      "metadata": {
        "id": "iZUIx-ywntpw",
        "colab_type": "code",
        "outputId": "b90003e5-7d17-4366-d716-31135054467e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "df.shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(5960, 13)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "metadata": {
        "id": "I_sNmzqpeRea",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Data Type"
      ]
    },
    {
      "metadata": {
        "id": "65_FbXKXn0wF",
        "colab_type": "code",
        "outputId": "dd7e3556-aa92-4b54-ffcc-bff5a06612e0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        }
      },
      "cell_type": "code",
      "source": [
        "df.dtypes.value_counts()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "float64    9\n",
              "object     2\n",
              "int64      2\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "metadata": {
        "id": "9fJEi_dMpauJ",
        "colab_type": "code",
        "outputId": "ddb5f4ab-ca05-4365-d233-0e5d47cd339e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "df.columns"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['BAD', 'LOAN', 'MORTDUE', 'VALUE', 'REASON', 'JOB', 'YOJ', 'DEROG',\n",
              "       'DELINQ', 'CLAGE', 'NINQ', 'CLNO', 'DEBTINC'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "metadata": {
        "id": "4njgb08xx0yu",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Statistical values for dataset"
      ]
    },
    {
      "metadata": {
        "id": "h6x6JozIoCSd",
        "colab_type": "code",
        "outputId": "082987b8-8451-4f8a-c851-c9d2d0d054f0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        }
      },
      "cell_type": "code",
      "source": [
        "df.describe()"
      ],
      "execution_count": 0,
      "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>BAD</th>\n",
              "      <th>LOAN</th>\n",
              "      <th>MORTDUE</th>\n",
              "      <th>VALUE</th>\n",
              "      <th>YOJ</th>\n",
              "      <th>DEROG</th>\n",
              "      <th>DELINQ</th>\n",
              "      <th>CLAGE</th>\n",
              "      <th>NINQ</th>\n",
              "      <th>CLNO</th>\n",
              "      <th>DEBTINC</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>5960.000000</td>\n",
              "      <td>5960.000000</td>\n",
              "      <td>5442.000000</td>\n",
              "      <td>5848.000000</td>\n",
              "      <td>5445.000000</td>\n",
              "      <td>5252.000000</td>\n",
              "      <td>5380.000000</td>\n",
              "      <td>5652.000000</td>\n",
              "      <td>5450.000000</td>\n",
              "      <td>5738.000000</td>\n",
              "      <td>4693.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.199497</td>\n",
              "      <td>18607.969799</td>\n",
              "      <td>73760.817200</td>\n",
              "      <td>101776.048741</td>\n",
              "      <td>8.922268</td>\n",
              "      <td>0.254570</td>\n",
              "      <td>0.449442</td>\n",
              "      <td>179.766275</td>\n",
              "      <td>1.186055</td>\n",
              "      <td>21.296096</td>\n",
              "      <td>33.779915</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.399656</td>\n",
              "      <td>11207.480417</td>\n",
              "      <td>44457.609458</td>\n",
              "      <td>57385.775334</td>\n",
              "      <td>7.573982</td>\n",
              "      <td>0.846047</td>\n",
              "      <td>1.127266</td>\n",
              "      <td>85.810092</td>\n",
              "      <td>1.728675</td>\n",
              "      <td>10.138933</td>\n",
              "      <td>8.601746</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>1100.000000</td>\n",
              "      <td>2063.000000</td>\n",
              "      <td>8000.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.524499</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>11100.000000</td>\n",
              "      <td>46276.000000</td>\n",
              "      <td>66075.500000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>115.116702</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>15.000000</td>\n",
              "      <td>29.140031</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>16300.000000</td>\n",
              "      <td>65019.000000</td>\n",
              "      <td>89235.500000</td>\n",
              "      <td>7.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>173.466667</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>20.000000</td>\n",
              "      <td>34.818262</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>23300.000000</td>\n",
              "      <td>91488.000000</td>\n",
              "      <td>119824.250000</td>\n",
              "      <td>13.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>231.562278</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>26.000000</td>\n",
              "      <td>39.003141</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>89900.000000</td>\n",
              "      <td>399550.000000</td>\n",
              "      <td>855909.000000</td>\n",
              "      <td>41.000000</td>\n",
              "      <td>10.000000</td>\n",
              "      <td>15.000000</td>\n",
              "      <td>1168.233561</td>\n",
              "      <td>17.000000</td>\n",
              "      <td>71.000000</td>\n",
              "      <td>203.312149</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "               BAD          LOAN        MORTDUE          VALUE          YOJ  \\\n",
              "count  5960.000000   5960.000000    5442.000000    5848.000000  5445.000000   \n",
              "mean      0.199497  18607.969799   73760.817200  101776.048741     8.922268   \n",
              "std       0.399656  11207.480417   44457.609458   57385.775334     7.573982   \n",
              "min       0.000000   1100.000000    2063.000000    8000.000000     0.000000   \n",
              "25%       0.000000  11100.000000   46276.000000   66075.500000     3.000000   \n",
              "50%       0.000000  16300.000000   65019.000000   89235.500000     7.000000   \n",
              "75%       0.000000  23300.000000   91488.000000  119824.250000    13.000000   \n",
              "max       1.000000  89900.000000  399550.000000  855909.000000    41.000000   \n",
              "\n",
              "             DEROG       DELINQ        CLAGE         NINQ         CLNO  \\\n",
              "count  5252.000000  5380.000000  5652.000000  5450.000000  5738.000000   \n",
              "mean      0.254570     0.449442   179.766275     1.186055    21.296096   \n",
              "std       0.846047     1.127266    85.810092     1.728675    10.138933   \n",
              "min       0.000000     0.000000     0.000000     0.000000     0.000000   \n",
              "25%       0.000000     0.000000   115.116702     0.000000    15.000000   \n",
              "50%       0.000000     0.000000   173.466667     1.000000    20.000000   \n",
              "75%       0.000000     0.000000   231.562278     2.000000    26.000000   \n",
              "max      10.000000    15.000000  1168.233561    17.000000    71.000000   \n",
              "\n",
              "           DEBTINC  \n",
              "count  4693.000000  \n",
              "mean     33.779915  \n",
              "std       8.601746  \n",
              "min       0.524499  \n",
              "25%      29.140031  \n",
              "50%      34.818262  \n",
              "75%      39.003141  \n",
              "max     203.312149  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "metadata": {
        "id": "O0KV5-Y8x82v",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### **Target value**"
      ]
    },
    {
      "metadata": {
        "id": "ClMsXLKvyEN-",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "As part of data, Target data set in 80/20 ratio. the dataset is **Imbalance.**"
      ]
    },
    {
      "metadata": {
        "id": "VfhvM7Ezpps2",
        "colab_type": "code",
        "outputId": "bee53f85-dd91-48a5-9f07-d1dc35e3995f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "df[\"BAD\"].value_counts(normalize=True)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    0.800503\n",
              "1    0.199497\n",
              "Name: BAD, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "metadata": {
        "id": "kj8VjzckyY-e",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Later on in sheet we will use SMOTE oversampling method to balance out the data to get a better results from model."
      ]
    },
    {
      "metadata": {
        "id": "p0EpkPd1ypwD",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Pre porcessing and Understanding the data Features"
      ]
    },
    {
      "metadata": {
        "id": "C4OvPsHiy6hg",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Missing value in the dataset **"
      ]
    },
    {
      "metadata": {
        "id": "qDiZQ7zXoU7i",
        "colab_type": "code",
        "outputId": "5f99189b-a18c-447f-f21a-15648b13023d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 255
        }
      },
      "cell_type": "code",
      "source": [
        "df.isna().sum()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "BAD           0\n",
              "LOAN          0\n",
              "MORTDUE     518\n",
              "VALUE       112\n",
              "REASON      252\n",
              "JOB         279\n",
              "YOJ         515\n",
              "DEROG       708\n",
              "DELINQ      580\n",
              "CLAGE       308\n",
              "NINQ        510\n",
              "CLNO        222\n",
              "DEBTINC    1267\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "metadata": {
        "id": "DcSPLeYuqwFK",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Let's check categorial feature count wise**"
      ]
    },
    {
      "metadata": {
        "id": "yKhw7Bfop6Yi",
        "colab_type": "code",
        "outputId": "7de11183-b680-4c6b-bf15-cec2432f92c5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "df[\"REASON\"].value_counts()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "DebtCon    3928\n",
              "HomeImp    1780\n",
              "Name: REASON, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "metadata": {
        "id": "s2FFJCpDLaTk",
        "colab_type": "code",
        "outputId": "777cf9b6-8048-49b6-f559-46eac6eaddee",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 145
        }
      },
      "cell_type": "code",
      "source": [
        "reason = df.groupby('REASON')\n",
        "reason['BAD'].value_counts(normalize=True).to_frame().style.format(\"{:.1%}\")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<style  type=\"text/css\" >\n",
              "</style>  \n",
              "<table id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002\" > \n",
              "<thead>    <tr> \n",
              "        <th class=\"blank\" ></th> \n",
              "        <th class=\"blank level0\" ></th> \n",
              "        <th class=\"col_heading level0 col0\" >BAD</th> \n",
              "    </tr>    <tr> \n",
              "        <th class=\"index_name level0\" >REASON</th> \n",
              "        <th class=\"index_name level1\" >BAD</th> \n",
              "        <th class=\"blank\" ></th> \n",
              "    </tr></thead> \n",
              "<tbody>    <tr> \n",
              "        <th id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002level0_row0\" class=\"row_heading level0 row0\" rowspan=2>DebtCon</th> \n",
              "        <th id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002level1_row0\" class=\"row_heading level1 row0\" >0</th> \n",
              "        <td id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002row0_col0\" class=\"data row0 col0\" >81.0%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002level1_row1\" class=\"row_heading level1 row1\" >1</th> \n",
              "        <td id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002row1_col0\" class=\"data row1 col0\" >19.0%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002level0_row2\" class=\"row_heading level0 row2\" rowspan=2>HomeImp</th> \n",
              "        <th id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002level1_row2\" class=\"row_heading level1 row2\" >0</th> \n",
              "        <td id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002row2_col0\" class=\"data row2 col0\" >77.8%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002level1_row3\" class=\"row_heading level1 row3\" >1</th> \n",
              "        <td id=\"T_47674be0_364f_11e9_a9de_0242ac1c0002row3_col0\" class=\"data row3 col0\" >22.2%</td> \n",
              "    </tr></tbody> \n",
              "</table> "
            ],
            "text/plain": [
              "<pandas.io.formats.style.Styler at 0x7f65b1165320>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "metadata": {
        "id": "Ms1zgCqxLqkk",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Comment**\n",
        "\n",
        "loan do not seem to depend strongly on the REASON."
      ]
    },
    {
      "metadata": {
        "id": "wnKr8yhxqimp",
        "colab_type": "code",
        "outputId": "717a79da-e151-48b9-f303-1db716050b1f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        }
      },
      "cell_type": "code",
      "source": [
        "df[\"JOB\"].value_counts()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Other      2388\n",
              "ProfExe    1276\n",
              "Office      948\n",
              "Mgr         767\n",
              "Self        193\n",
              "Sales       109\n",
              "Name: JOB, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "metadata": {
        "id": "TYiSFOyEDcPE",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Comment**\n",
        "\n",
        "Office worker and professional executives have the highest probability to pay their loans while sales and self employed have the highest probability to default. The occupation shows a good discriminating power and it will most likely be an important feature of our classification model\n"
      ]
    },
    {
      "metadata": {
        "id": "ISUfENS1Jp3y",
        "colab_type": "code",
        "outputId": "453dc446-8bd0-4149-8259-c427adbcb1ce",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        }
      },
      "cell_type": "code",
      "source": [
        "job = df.groupby('JOB')\n",
        "job['BAD'].value_counts(normalize=True).to_frame().style.format(\"{:.1%}\")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<style  type=\"text/css\" >\n",
              "</style>  \n",
              "<table id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002\" > \n",
              "<thead>    <tr> \n",
              "        <th class=\"blank\" ></th> \n",
              "        <th class=\"blank level0\" ></th> \n",
              "        <th class=\"col_heading level0 col0\" >BAD</th> \n",
              "    </tr>    <tr> \n",
              "        <th class=\"index_name level0\" >JOB</th> \n",
              "        <th class=\"index_name level1\" >BAD</th> \n",
              "        <th class=\"blank\" ></th> \n",
              "    </tr></thead> \n",
              "<tbody>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level0_row0\" class=\"row_heading level0 row0\" rowspan=2>Mgr</th> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row0\" class=\"row_heading level1 row0\" >0</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row0_col0\" class=\"data row0 col0\" >76.7%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row1\" class=\"row_heading level1 row1\" >1</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row1_col0\" class=\"data row1 col0\" >23.3%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level0_row2\" class=\"row_heading level0 row2\" rowspan=2>Office</th> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row2\" class=\"row_heading level1 row2\" >0</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row2_col0\" class=\"data row2 col0\" >86.8%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row3\" class=\"row_heading level1 row3\" >1</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row3_col0\" class=\"data row3 col0\" >13.2%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level0_row4\" class=\"row_heading level0 row4\" rowspan=2>Other</th> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row4\" class=\"row_heading level1 row4\" >0</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row4_col0\" class=\"data row4 col0\" >76.8%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row5\" class=\"row_heading level1 row5\" >1</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row5_col0\" class=\"data row5 col0\" >23.2%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level0_row6\" class=\"row_heading level0 row6\" rowspan=2>ProfExe</th> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row6\" class=\"row_heading level1 row6\" >0</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row6_col0\" class=\"data row6 col0\" >83.4%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row7\" class=\"row_heading level1 row7\" >1</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row7_col0\" class=\"data row7 col0\" >16.6%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level0_row8\" class=\"row_heading level0 row8\" rowspan=2>Sales</th> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row8\" class=\"row_heading level1 row8\" >0</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row8_col0\" class=\"data row8 col0\" >65.1%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row9\" class=\"row_heading level1 row9\" >1</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row9_col0\" class=\"data row9 col0\" >34.9%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level0_row10\" class=\"row_heading level0 row10\" rowspan=2>Self</th> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row10\" class=\"row_heading level1 row10\" >0</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row10_col0\" class=\"data row10 col0\" >69.9%</td> \n",
              "    </tr>    <tr> \n",
              "        <th id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002level1_row11\" class=\"row_heading level1 row11\" >1</th> \n",
              "        <td id=\"T_b35cddc0_364e_11e9_a9de_0242ac1c0002row11_col0\" class=\"data row11 col0\" >30.1%</td> \n",
              "    </tr></tbody> \n",
              "</table> "
            ],
            "text/plain": [
              "<pandas.io.formats.style.Styler at 0x7f65b171ccf8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "metadata": {
        "id": "BJkxM-7fzVTB",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "later on for imputing missing value for above  **categorial features**  we will use most frequest ( Mode ) technique."
      ]
    },
    {
      "metadata": {
        "id": "APtahiryz4T8",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Let's check the distribution of the features"
      ]
    },
    {
      "metadata": {
        "id": "1K1dKbOIqt9n",
        "colab_type": "code",
        "outputId": "cb42e860-61af-4d87-a1ab-8137b8945705",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 688
        }
      },
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "plt.figure(figsize=(10,10))\n",
        "plt.subplot(321)\n",
        "ax1=sns.distplot(df['LOAN'])\n",
        "ax1.set_title('Loan')\n",
        "plt.subplot(322)\n",
        "ax2=df[\"DEBTINC\"].plot.hist(bins = 20)\n",
        "ax2.set_title('DEBTINC')\n",
        "plt.subplot(323)\n",
        "ax3=df[\"CLAGE\"].plot.hist(bins = 20)\n",
        "ax3.set_title('CLAGE')\n",
        "plt.subplot(324)\n",
        "ax4=df[\"CLNO\"].plot.hist(bins = 20)\n",
        "ax4.set_title('CLNO')\n",
        "plt.subplot(325)\n",
        "ax5=df[\"VALUE\"].plot.hist(bins = 80)\n",
        "ax5.set_xlabel('VALUE')\n",
        "plt.subplot(326)\n",
        "ax6=df[\"YOJ\"].plot.hist(bins = 40)\n",
        "ax6.set_xlabel('YOJ')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/matplotlib/axes/_axes.py:6521: MatplotlibDeprecationWarning: \n",
            "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
            "  alternative=\"'density'\", removal=\"3.1\")\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'YOJ')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmcAAAJbCAYAAAChCjPHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 720x720 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "Yf2ULbTA0IG6",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Comment**\n",
        "\n",
        "We can use cappping technique to remobe extreme value and reshaping it to normal distribution.\n",
        "\n",
        "'YOG' feature which is hight skewed we will transform it using log transformation.\n",
        "\n",
        "'VALUE' and 'CLAG' feature value more than 400k and 500 respectively can be capped.\n",
        "\n",
        "'CLNO' feature is centered around good distribution\n"
      ]
    },
    {
      "metadata": {
        "id": "cuW44Wzq0QaG",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Imputation "
      ]
    },
    {
      "metadata": {
        "id": "ZlWSnlV10hOw",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "By understanding the distribution and box plot below we can take the decision for imputing missing value."
      ]
    },
    {
      "metadata": {
        "id": "F3gCA1bPNFC6",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "For Categorical feature we are using Mode(most frequent)  imputation menthod.\n",
        "\n",
        "For Numerical feature we are going to use Mean( rest) and Median('for YOG') method"
      ]
    },
    {
      "metadata": {
        "id": "wt1mIc_51g4X",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "df[\"REASON\"].fillna(value = \"DebtCon\",inplace = True)\n",
        "df[\"JOB\"].fillna(value = \"Other\",inplace = True)\n",
        "df[\"DEROG\"].fillna(value=0,inplace=True)\n",
        "df[\"DELINQ\"].fillna(value=0,inplace=True)\n",
        "df['YOJ'].fillna(value=df.median(),inplace=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Qmz7jBUNs7Sg",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "df.fillna(value=df.mean(),inplace=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "gW_v8ih502U5",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "###Box plot visulization for Outlier and feature transformation"
      ]
    },
    {
      "metadata": {
        "id": "UOyGlqSjNxNE",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "for capping feature as well as for outlier let's plot "
      ]
    },
    {
      "metadata": {
        "id": "lpmL7bBqytQ6",
        "colab_type": "code",
        "outputId": "bf8ef4dc-6eda-40f8-ae8d-97c402b7dca5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 609
        }
      },
      "cell_type": "code",
      "source": [
        "plt.figure(figsize=(10,10))\n",
        "plt.subplot(321)\n",
        "ax1=sns.boxplot(y=df['LOAN'])\n",
        "#title('Loan')\n",
        "plt.subplot(322)\n",
        "\n",
        "ax2=sns.boxplot(y=df['DEBTINC'])\n",
        "plt.subplot(323)\n",
        "\n",
        "ax3=sns.boxplot(y=df['CLAGE'])\n",
        "plt.subplot(324)\n",
        "\n",
        "ax4=sns.boxplot(y=df['CLNO'])\n",
        "plt.subplot(325)\n",
        "\n",
        "ax5=sns.boxplot(y=df['VALUE'])\n",
        "plt.subplot(326)\n",
        "\n",
        "ax1=sns.boxplot(y=df['YOJ'])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/seaborn/categorical.py:454: FutureWarning: remove_na is deprecated and is a private function. Do not use.\n",
            "  box_data = remove_na(group_data)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnMAAAIuCAYAAADUoI4nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nREREIvb/DDQ1mNL9uBQAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 720x720 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "znqIFoTdOHOo",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Form above we can see quite a few feature have a little amount of outlier.\n",
        "\n",
        "To improve model from outlier we are capping feature and further remove some as part of feature selection."
      ]
    },
    {
      "metadata": {
        "id": "4mvpsbR61Gp9",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### **Feature transform**"
      ]
    },
    {
      "metadata": {
        "id": "nCQetfcE2s7j",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "df.loc[df[\"CLAGE\"]>=500,\"CLAGE\"] = 500\n",
        "df.loc[df[\"VALUE\"]>=400000,\"VALUE\"] = 400000\n",
        "df.loc[df[\"MORTDUE\"]>=300000,\"MORTDUE\"] = 300000\n",
        "df.loc[df[\"DEBTINC\"]>=100,\"DEBTINC\"] = 100"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "AtB1mcqD1O-L",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Feature \"YOG\" have highly skewed as we have seen in distribution plot as we are using **box-cox transformation.**"
      ]
    },
    {
      "metadata": {
        "id": "Jgn1Wx9a5bJf",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "df[\"YOJ\"] = df[\"YOJ\"].apply(lambda x: np.log(x+1))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "8fazFaFax44N",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "#df['LTV']=df['LOAN']/df['VALUE']"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "llOeKC5U0T-m",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "#df.drop(['VALUE'], axis=1, inplace=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "18EL8mQQ6Lsk",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**One hot encoding for categorial feature**"
      ]
    },
    {
      "metadata": {
        "id": "vSTgybl45gpQ",
        "colab_type": "code",
        "outputId": "6e72ae9c-4e15-435c-d075-6c0dcb03ec7f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        }
      },
      "cell_type": "code",
      "source": [
        "df=pd.get_dummies(data=df, columns=['REASON', 'JOB'])\n",
        "df.columns"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['BAD', 'LOAN', 'MORTDUE', 'VALUE', 'YOJ', 'DEROG', 'DELINQ', 'CLAGE',\n",
              "       'NINQ', 'CLNO', 'DEBTINC', 'REASON_DebtCon', 'REASON_HomeImp',\n",
              "       'JOB_Mgr', 'JOB_Office', 'JOB_Other', 'JOB_ProfExe', 'JOB_Sales',\n",
              "       'JOB_Self'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "metadata": {
        "id": "YBghDNHa74jg",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### **Feature selection**\n",
        "\n",
        "We will use the following for this purpose :\n",
        "\n",
        "**Pearson correlation**\n",
        "\n",
        "**chi square test **\n",
        "\n",
        "### ** First plot Correlation matrix for numerical features**"
      ]
    },
    {
      "metadata": {
        "id": "_5_6NZ0J7ts8",
        "colab_type": "code",
        "outputId": "b243e9b8-462e-4d3e-a22f-0671e457e2bc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 388
        }
      },
      "cell_type": "code",
      "source": [
        "df1=df.loc[:, :'DEBTINC']\n",
        "ax=sns.heatmap(df1.corr(),annot=True,fmt='0.2f',cmap='coolwarm')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe0AAAFzCAYAAAAT2FHIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n8I4gjLZAIBAIBO8I/wcZ0EVVQD589gAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 576x396 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "niIsQOaSO8ms",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "As we can see 'VALUE',  'MORTDUE' are **highly correlated. we are removing** 'MORTDUE' from dataset"
      ]
    },
    {
      "metadata": {
        "id": "PzFTvxXB5SRV",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "df.drop(['MORTDUE'], axis=1, inplace=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "Ykz5uc8AJcWG",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "y = df['BAD']\n",
        "df.drop(['BAD'], axis=1, inplace=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "FbDoiExd2IfH",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Chi-Square based feature selection**"
      ]
    },
    {
      "metadata": {
        "id": "FS2_DkRgPaEP",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "we will use Chi square base feature selection method for most important feature."
      ]
    },
    {
      "metadata": {
        "id": "JkKZ-7Yx-bXK",
        "colab_type": "code",
        "outputId": "f1bbb0a7-422d-4e0e-be4c-7e039444c28c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "cell_type": "code",
      "source": [
        "from sklearn.feature_selection import SelectKBest\n",
        "from sklearn.feature_selection import chi2\n",
        "\n",
        "\n",
        "chi2_selector = SelectKBest(chi2, k=12)\n",
        "X_kbest = chi2_selector.fit_transform(df,y)\n",
        "df_chi = pd.DataFrame(X_kbest)\n",
        "df_chi.head()"
      ],
      "execution_count": 0,
      "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>0</th>\n",
              "      <td>1100.0</td>\n",
              "      <td>39025.000000</td>\n",
              "      <td>2.442347</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>94.366667</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>33.779915</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1300.0</td>\n",
              "      <td>68400.000000</td>\n",
              "      <td>2.079442</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>121.833333</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>33.779915</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1500.0</td>\n",
              "      <td>16700.000000</td>\n",
              "      <td>1.609438</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>149.466667</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>33.779915</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1500.0</td>\n",
              "      <td>101776.048741</td>\n",
              "      <td>2.294782</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>179.766275</td>\n",
              "      <td>1.186055</td>\n",
              "      <td>33.779915</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1700.0</td>\n",
              "      <td>112000.000000</td>\n",
              "      <td>1.386294</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>93.333333</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>33.779915</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       0              1         2    3    4           5         6          7   \\\n",
              "0  1100.0   39025.000000  2.442347  0.0  0.0   94.366667  1.000000  33.779915   \n",
              "1  1300.0   68400.000000  2.079442  0.0  2.0  121.833333  0.000000  33.779915   \n",
              "2  1500.0   16700.000000  1.609438  0.0  0.0  149.466667  1.000000  33.779915   \n",
              "3  1500.0  101776.048741  2.294782  0.0  0.0  179.766275  1.186055  33.779915   \n",
              "4  1700.0  112000.000000  1.386294  0.0  0.0   93.333333  0.000000  33.779915   \n",
              "\n",
              "    8    9    10   11  \n",
              "0  0.0  0.0  0.0  0.0  \n",
              "1  0.0  0.0  0.0  0.0  \n",
              "2  0.0  0.0  0.0  0.0  \n",
              "3  0.0  0.0  0.0  0.0  \n",
              "4  1.0  0.0  0.0  0.0  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        }
      ]
    },
    {
      "metadata": {
        "id": "kCDIud1cd2CP",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "We are choosing best 12 feature form chi square test."
      ]
    },
    {
      "metadata": {
        "id": "QijMft5QPobV",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Later on we will use Model based most feature_importance  those will be part of feature selection.**"
      ]
    },
    {
      "metadata": {
        "id": "cE1eEcq3EOS3",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Data Partition "
      ]
    },
    {
      "metadata": {
        "id": "UmH2AuoTcVQk",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "x_tr,x_te,y_tr,y_te = train_test_split(df_chi,y,test_size = .33,random_state=1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "8AGptKiIEedu",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "for the training part of data set we can use oversampling method."
      ]
    },
    {
      "metadata": {
        "id": "TFFch4672ba0",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**SMOTE based oversampling technique**"
      ]
    },
    {
      "metadata": {
        "id": "YcX-bzuCQJgl",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "As we have earlier seen data is **imbalance** so for some model pipe. it tries to give better results."
      ]
    },
    {
      "metadata": {
        "id": "O2IRpQ5Qbtm7",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from imblearn.over_sampling import SMOTE\n",
        "sm = SMOTE(random_state=2)\n",
        "x_tr_sm, y_tr_sm = sm.fit_sample(x_tr, y_tr.ravel())"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "BBoJIWKFcv5U",
        "colab_type": "code",
        "outputId": "cc0b61e3-5edb-415d-c6c7-b4a88c1e7cd6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "cell_type": "code",
      "source": [
        "print(\"After OverSampling, counts of label '1': {}\".format(sum(y_tr_sm==1)))\n",
        "print(\"After OverSampling, counts of label '0': {}\".format(sum(y_tr_sm==0)))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "After OverSampling, counts of label '1': 3210\n",
            "After OverSampling, counts of label '0': 3210\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "5qjj8AMJQmEX",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Now we have two kind of dataset-**\n",
        "\n",
        "**1.Origional Imbalance dataset**\n",
        "\n",
        "**2.SMOTE base oversampled dataset**\n",
        "\n",
        "We will fit both kind of dataset over a model. which ever gives good result we will proceed for that part."
      ]
    },
    {
      "metadata": {
        "id": "aOo8V14A2pJ4",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## **Model building and selection**"
      ]
    },
    {
      "metadata": {
        "id": "fCvi6LB-NOgG",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Comment for model selection**\n",
        "\n",
        "As we can see the dataset above we are able to convert feature in normal way and reduce the features. Earlier dataset consist a amont  of outliers, so not moving with SVM. \n",
        "\n",
        "We can go with logistic regression ,Tree based model and boosting algorithms"
      ]
    },
    {
      "metadata": {
        "id": "ag6IhwM42uWG",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**1.Logistic Regression**"
      ]
    },
    {
      "metadata": {
        "id": "yz7J7q--Rk2_",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Here we are assuming a cutoff of 0.5 as threshold, once we able to select better model then we will play with probablity threshold."
      ]
    },
    {
      "metadata": {
        "id": "aCRe4_5lHXZ7",
        "colab_type": "code",
        "outputId": "e849211a-044f-44b8-d8b3-191ac5e05987",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 275
        }
      },
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "\n",
        "logreg = LogisticRegression()\n",
        "logreg.fit(x_tr_sm,y_tr_sm)\n",
        "\n",
        "y_pred_proba = logreg.predict_proba(x_te)\n",
        "dum=pd.DataFrame(y_pred_proba[:,1])\n",
        "print(dum.head(5))\n",
        "\n",
        "y_pre = logreg.predict(x_te)\n",
        "a5 = accuracy_score(y_te,y_pre)\n",
        "f5 = f1_score(y_te, y_pre, average=\"macro\")\n",
        "p5 = precision_score(y_te, y_pre, average=\"macro\")\n",
        "r5 = recall_score(y_te, y_pre, average=\"macro\")\n",
        "print(\"accuracy score : \",a5)\n",
        "print(\"f1 score : \",f5)\n",
        "print(\"precision score : \",p5)\n",
        "print(\"recall score : \",r5)\n",
        "\n",
        "# Computing Confusion matrix for the above algorithm\n",
        "\n",
        "cnf_matrix = confusion_matrix(y_te, y_pre)\n",
        "np.set_printoptions(precision=2)\n",
        "print(cnf_matrix)\n",
        "# Plot non-normalized confusion matrix\n",
        "#plt.figure()\n",
        "#plot_confusion_matrix(cnf_matrix, classes=[\"BAD\"],\n",
        "# title='Confusion matrix - Logistic Regression Algorithm with chi2 test')\n",
        "\n",
        "#plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "          0\n",
            "0  0.988859\n",
            "1  0.146216\n",
            "2  0.475800\n",
            "3  0.083643\n",
            "4  0.331319\n",
            "accuracy score :  0.7707168276563294\n",
            "f1 score :  0.6912318988126005\n",
            "precision score :  0.6776673346389155\n",
            "recall score :  0.7215920387019815\n",
            "[[1257  304]\n",
            " [ 147  259]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
            "  FutureWarning)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "2qS5o3igRMB7",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Logistic Regression is giving better results **in case over-sampled data**. but **not as good as **upcomming model.\n",
        "\n",
        "Logistic Regression got above results only after feature selection process."
      ]
    },
    {
      "metadata": {
        "id": "hqRY79odTRS9",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Comment on Accuracy Score selection\n",
        "\n",
        "Generally Accuracy test can be used to evaluate algorithms. But in this case, using the MAJORITY CLASS(0) to predict the output will give high(79.2%) accuracy. but we would end up **PREDICTING ALL 1's AS O's INSTEAD**.\n",
        "\n",
        "we want to decrease the number of **FN or FP** error as We predict that credit will be repaid but it actually being a fraudant one. as per busniess understanding we can tackle up with Recall or Precision optimization.\n",
        "\n",
        "hence we will performance metrics must be used to evaluate the model. those would be\n",
        "**F1 score , \n",
        "Recall ,\n",
        "Precision **Confusion matrix parameter."
      ]
    },
    {
      "metadata": {
        "id": "dY7rlGOYxKw3",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "y_pred_proba = logreg.predict_proba(x_te)\n",
        "from sklearn import metrics\n",
        "from sklearn.metrics import roc_auc_score\n",
        "\n",
        "preds = y_pred_proba[:,1]\n",
        "fpr, tpr, threshold = metrics.roc_curve(y_te, preds)\n",
        "roc_auc = metrics.auc(fpr, tpr)\n",
        "\n",
        "print(\"ROC AUC score for model': {}\".format(roc_auc))\n",
        "import matplotlib.pyplot as plt\n",
        "plt.title('Receiver Operating Characteristic')\n",
        "plt.plot(fpr, tpr, 'b', label = 'AUC = %0.2f' % roc_auc)\n",
        "plt.legend(loc = 'lower right')\n",
        "plt.plot([0, 1], [0, 1],'r--')\n",
        "plt.xlim([0, 1])\n",
        "plt.ylim([0, 1])\n",
        "plt.ylabel('True Positive Rate')\n",
        "plt.xlabel('False Positive Rate')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "b4MZ5rqW24E-",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### **Decision Tree Model**"
      ]
    },
    {
      "metadata": {
        "id": "6XDHG013ICpT",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Default assumption:**\n",
        "\n",
        "**The threshold is defaultly set to .5 in these Models**\n",
        "\n",
        "**Obviously,changing the threshold affects the performance of the model and this can be observed in the next section.** "
      ]
    },
    {
      "metadata": {
        "id": "0ydSvct3KEt4",
        "colab_type": "code",
        "outputId": "b865d506-8f2d-454d-bcce-b2efd4eb3cab",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        }
      },
      "cell_type": "code",
      "source": [
        "clf_tree=DecisionTreeClassifier()\n",
        "clf_tree.max_depth = 10\n",
        "clf_tree.fit(x_tr,y_tr)\n",
        "y_pre = clf_tree.predict(x_te)\n",
        "a8 = accuracy_score(y_te,y_pre)\n",
        "f8 = f1_score(y_te, y_pre, average=\"macro\")\n",
        "p8 = precision_score(y_te, y_pre, average=\"macro\")\n",
        "r8 = recall_score(y_te, y_pre, average=\"macro\")\n",
        "print(\"accuracy score : \",a8)\n",
        "print(\"f1 score : \",f8)\n",
        "print(\"precision score : \",p8)\n",
        "print(\"recall score : \",r8)\n",
        "# Computing Confusion matrix for the above algorithm\n",
        "\n",
        "cnf_matrix = confusion_matrix(y_te, y_pre)\n",
        "print(cnf_matrix)\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "accuracy score :  0.8820538891713269\n",
            "f1 score :  0.8040607846774248\n",
            "precision score :  0.8376679189435337\n",
            "recall score :  0.7808047449689632\n",
            "[[1488   73]\n",
            " [ 159  247]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "wwd4yhenHSzB",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Decision Tree dominated over Logistic Regression**"
      ]
    },
    {
      "metadata": {
        "id": "1U84A_A22_Di",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Random Forest**"
      ]
    },
    {
      "metadata": {
        "id": "e3o1iM9IMqQi",
        "colab_type": "code",
        "outputId": "73159547-27e0-4565-cc0e-5563b55a914c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 190
        }
      },
      "cell_type": "code",
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "random_forest = RandomForestClassifier(n_estimators=200, min_samples_leaf=10, min_samples_split=2, criterion='entropy', \n",
        "                               max_depth=None, bootstrap=True, max_features=9)\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "scaler = StandardScaler()\n",
        "scaled_data = scaler.fit_transform(df)\n",
        "\n",
        "x_tr,x_te,y_tr,y_te = train_test_split(scaled_data,y,test_size = .33,random_state=1)\n",
        "random_forest.fit(x_tr, y_tr)\n",
        "\n",
        "y_pred = random_forest.predict(x_te)\n",
        "a = accuracy_score(y_te,y_pre)\n",
        "f = f1_score(y_te, y_pre, average=\"macro\")\n",
        "p = precision_score(y_te, y_pre, average=\"macro\")\n",
        "r = recall_score(y_te, y_pre, average=\"macro\")\n",
        "#print(\"accuracy score : \",a)\n",
        "print(\"f1 score : \",f)\n",
        "print(\"precision score : \",p)\n",
        "print(\"recall score : \",r)\n",
        "\n",
        "\n",
        "cnf_matrix = confusion_matrix(y_te, y_pre)\n",
        "print(cnf_matrix)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/preprocessing/data.py:625: DataConversionWarning: Data with input dtype uint8, int64, float64 were all converted to float64 by StandardScaler.\n",
            "  return self.partial_fit(X, y)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/base.py:462: DataConversionWarning: Data with input dtype uint8, int64, float64 were all converted to float64 by StandardScaler.\n",
            "  return self.fit(X, **fit_params).transform(X)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "f1 score :  0.8040607846774248\n",
            "precision score :  0.8376679189435337\n",
            "recall score :  0.7808047449689632\n",
            "[[1488   73]\n",
            " [ 159  247]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "PeGEvds3FpIJ",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Here we are getting better results."
      ]
    },
    {
      "metadata": {
        "id": "HkpB_HaKFw4Y",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Let's plot ROC For the same."
      ]
    },
    {
      "metadata": {
        "id": "YUg6tovsKivw",
        "colab_type": "code",
        "outputId": "4cb68acd-6683-4051-b767-fd05d41ef62c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 311
        }
      },
      "cell_type": "code",
      "source": [
        "from sklearn import metrics\n",
        "from sklearn.metrics import roc_auc_score\n",
        "probs = random_forest.predict_proba(x_te)\n",
        "preds = probs[:,1]\n",
        "fpr, tpr, threshold = metrics.roc_curve(y_te, preds)\n",
        "roc_auc = metrics.auc(fpr, tpr)\n",
        "\n",
        "print(\"ROC AUC score for model': {}\".format(roc_auc))\n",
        "import matplotlib.pyplot as plt\n",
        "plt.figure(figsize=(5,4))\n",
        "plt.title('Receiver Operating Characteristic')\n",
        "plt.plot(fpr, tpr, 'b', label = 'AUC = %0.2f' % roc_auc)\n",
        "plt.legend(loc = 'lower right')\n",
        "plt.plot([0, 1], [0, 1],'r--')\n",
        "plt.xlim([0, 1])\n",
        "plt.ylim([0, 1])\n",
        "plt.ylabel('True Positive Rate')\n",
        "plt.xlabel('False Positive Rate')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "ROC AUC score for model': 0.9230788650700733\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAU8AAAEVCAYAAABg2GdjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkOIphBCl8P/APuGI8OE3fwAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 360x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "ukXb-oiqWNYi",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### K-fold cross validation to improve performance of model."
      ]
    },
    {
      "metadata": {
        "id": "RLfrTuyFOowp",
        "colab_type": "code",
        "outputId": "21f6e7bf-b470-455c-8800-0204dcd1795f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import KFold,cross_val_score\n",
        "#k_fold = KFold(len(y), n_splits=5, shuffle=True, random_state=0)\n",
        "K=3\n",
        "k_fold = KFold(n_splits = K, random_state = 0, shuffle = True)\n",
        "#clf = RandomForestClassifier(n_estimators=est, min_samples_split=min_samples)\n",
        "forest = RandomForestClassifier(n_estimators=200, min_samples_leaf=10, min_samples_split=2, criterion='entropy', \n",
        "                               max_depth=None, bootstrap=True, max_features=9)\n",
        "scoring = 'precision_macro'\n",
        "results = cross_val_score(forest, df_chi, y, cv=k_fold, n_jobs=1, scoring=scoring)\n",
        "print(results)\n",
        "print(\"Precision Score in 3 fold CV': {}\".format(results.mean()))\n",
        "results_1 = cross_val_score(forest, df_chi, y, cv=k_fold, n_jobs=1, scoring='recall_macro')\n",
        "print(\"Recall Score in 3 fold CV': {}\".format(results_1.mean()))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[0.84 0.84 0.85]\n",
            "Precision Score in 3 fold CV': 0.8419494732464274\n",
            "Recall Score in 3 fold CV': 0.7950440316391854\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "ELzqlDVvSryf",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Slightly impove in score using k fold."
      ]
    },
    {
      "metadata": {
        "id": "8Km41rUVWbuV",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Model over various threshold "
      ]
    },
    {
      "metadata": {
        "id": "KzoW7ikqF90O",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "We can change the threshold of the model as per our busniess understanding and by looking precision recall score below."
      ]
    },
    {
      "metadata": {
        "id": "LNHBREN4k1FY",
        "colab_type": "code",
        "outputId": "f852081b-3304-4973-f577-748938271663",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 187
        }
      },
      "cell_type": "code",
      "source": [
        "forest.fit(x_tr,y_tr)\n",
        "y_pred_proba = forest.predict_proba(x_te)\n",
        "\n",
        "thresholds = [0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9]\n",
        "recall,prec,acc=[],[],[]\n",
        "for i in thresholds:\n",
        "  y_test_predictions_high_recall = y_pred_proba[:,1] >= i\n",
        "  cnf_matrix = confusion_matrix(y_te,y_test_predictions_high_recall)\n",
        "  rec1 = recall_score(y_te, y_test_predictions_high_recall)\n",
        "  recall.append(rec1)\n",
        "  pres1=precision_score(y_te, y_test_predictions_high_recall)\n",
        "  prec.append(pres1)\n",
        "  acc.append(accuracy_score(y_te,y_test_predictions_high_recall))\n",
        "  #print(\"Recall metric in the testing dataset: \",rec1)\n",
        "\n",
        "dash=pd.DataFrame({'Threshold':thresholds,'Recall':recall,'Precision':prec, 'Accuracy':acc})\n",
        "print(dash)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "   Accuracy  Precision    Recall  Threshold\n",
            "0  0.762583   0.462668  0.931034        0.1\n",
            "1  0.860702   0.620879  0.834975        0.2\n",
            "2  0.884596   0.704805  0.758621        0.3\n",
            "3  0.891713   0.762943  0.689655        0.4\n",
            "4  0.890696   0.819398  0.603448        0.5\n",
            "5  0.882562   0.857143  0.517241        0.6\n",
            "6  0.863244   0.874317  0.394089        0.7\n",
            "7  0.845450   0.955357  0.263547        0.8\n",
            "8  0.808338   1.000000  0.071429        0.9\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "HQcrjvAQJIRl",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "The values of recall and accuracy vary with the threshold selected.\n",
        "\n",
        "Based on the required accuracy and recall values , one has to decide and select a threshold."
      ]
    },
    {
      "metadata": {
        "id": "QO3Da672GUNJ",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Plot the above collected data to visulize**"
      ]
    },
    {
      "metadata": {
        "id": "msdujiu96rC_",
        "colab_type": "code",
        "outputId": "2732fb98-b5aa-4f62-c369-5c30fe3a292a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 364
        }
      },
      "cell_type": "code",
      "source": [
        "plt.plot( 'Threshold', 'Precision', data=dash)\n",
        "plt.plot( 'Threshold', 'Recall', data=dash, color='olive')\n",
        "plt.plot( 'Threshold', 'Accuracy', data=dash, )\n",
        "plt.legend()\n",
        "\n",
        "#ax = sns.lineplot(x='Threshold', y='Precision', data=dash)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f65aa9c6b00>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 41
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeEAAAFKCAYAAAAqkecjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "lgARCMHNGnlk",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "from graph we can set optimum value of threshold as per requirement."
      ]
    },
    {
      "metadata": {
        "id": "O3_Gch9n3IWd",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Let's check with Light Boosting  LGB Model"
      ]
    },
    {
      "metadata": {
        "id": "eMtl2mRnY3yl",
        "colab_type": "code",
        "outputId": "a5a730dd-58bc-4e7b-d27b-21e714434748",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        }
      },
      "cell_type": "code",
      "source": [
        "import lightgbm as lgb\n",
        "model = lgb.LGBMClassifier(objective='binary', boosting_type = 'goss', n_estimators = 10000, class_weight = 'balanced')\n",
        "\n",
        "for i in range(2):\n",
        "    \n",
        "    # Split into training and validation set\n",
        "    #train_features, valid_features, train_y, valid_y = train_test_split(train, train_labels, test_size = 0.25, random_state = i)\n",
        "    \n",
        "    # Train using early stopping\n",
        "    model.fit(x_tr, y_tr, early_stopping_rounds=100, eval_set = [(x_te, y_te)], \n",
        "              eval_metric = 'auc', verbose = 200)\n",
        "    "
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training until validation scores don't improve for 100 rounds.\n",
            "[200]\tvalid_0's auc: 0.950434\tvalid_0's binary_logloss: 0.217307\n",
            "Early stopping, best iteration is:\n",
            "[195]\tvalid_0's auc: 0.950613\tvalid_0's binary_logloss: 0.216592\n",
            "Training until validation scores don't improve for 100 rounds.\n",
            "[200]\tvalid_0's auc: 0.950434\tvalid_0's binary_logloss: 0.217307\n",
            "Early stopping, best iteration is:\n",
            "[195]\tvalid_0's auc: 0.950613\tvalid_0's binary_logloss: 0.216592\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "JJzSbClhPE4r",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "Whooaa!!"
      ]
    },
    {
      "metadata": {
        "id": "0tpqSMCgcyQQ",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "We can think for the LGB model also we good set of parameter and hyper parameter tuning."
      ]
    },
    {
      "metadata": {
        "id": "aNqs-oHkblwa",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "**Comment**\n",
        "\n",
        "Performance of the modeld mentioned can be **further improved**  as part of **Tuning Hyer-parameter** and a good Feature engineering as per understanding.\n",
        "\n",
        "I am leaving that part as i am occupied with other activities and cetain time constraints.\n",
        "\n",
        "Hopefully you like the rest part."
      ]
    },
    {
      "metadata": {
        "id": "gBlRAq3wcVCG",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Thanks"
      ]
    }
  ]
}