Student_Performance_Analysis.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "TARP_project",
      "provenance": [],
      "collapsed_sections": [
        "_197dAffXJn9",
        "XCjUkDZfXNoZ",
        "L1wuPUOQXSy_",
        "5crx74xDZXOW",
        "_BwlA13Rg0gn",
        "cGdWuDD7oJcP",
        "LOGK8LFUg652",
        "4XglJ_W5ZdzN"
      ],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/sumitajmera/Student-Performance-analysis/blob/main/Student_Performance_Analysis.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QP6I2gTTHY62"
      },
      "source": [
        "# Student Performance Mining"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oWJZxlWma3Ez",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "48240875-1acb-4a07-d438-480afbfda01d"
      },
      "source": [
        "import numpy as np  \n",
        "import pandas as pd\n",
        "import random\n",
        "\n",
        "# Import libraries for plotting data\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Import the classifiers used for creating the models\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.naive_bayes import BernoulliNB\n",
        "from sklearn.tree import ExtraTreeClassifier\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.naive_bayes import GaussianNB\n",
        "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
        "from sklearn.semi_supervised import LabelSpreading\n",
        "from sklearn.svm import LinearSVC\n",
        "from sklearn.neighbors.nearest_centroid import NearestCentroid\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "#from sklearn.ensemble import Classifier\n",
        "from lightgbm import LGBMClassifier\n",
        "\n",
        "# Import the metrics libraries for comparinf the results from these algorithms\n",
        "from sklearn.metrics import f1_score\n",
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.model_selection import KFold"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:144: FutureWarning: The sklearn.neighbors.nearest_centroid module is  deprecated in version 0.22 and will be removed in version 0.24. The corresponding classes / functions should instead be imported from sklearn.neighbors. Anything that cannot be imported from sklearn.neighbors is now part of the private API.\n",
            "  warnings.warn(message, FutureWarning)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TXfY6erOHTFD"
      },
      "source": [
        "## 1. Data Import and Pre-processing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2V9eI1TDbtjw"
      },
      "source": [
        "df_por = pd.read_csv('student-por.csv', sep=\";\")\n",
        "df_mat  = pd.read_csv('student-mat.csv', sep = \";\")\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PCdfeyfKb1jR",
        "outputId": "e24852e8-befe-4c25-b240-0ac69a50c193"
      },
      "source": [
        "df_por.head()\n",
        "df_por[\"address\"].value_counts()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "U    452\n",
              "R    197\n",
              "Name: address, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "id": "aOnAvQvHb5f6",
        "outputId": "07705a20-5688-4652-a628-ae6c6e2db68f"
      },
      "source": [
        "df_mat.head()"
      ],
      "execution_count": null,
      "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>school</th>\n",
              "      <th>sex</th>\n",
              "      <th>age</th>\n",
              "      <th>address</th>\n",
              "      <th>famsize</th>\n",
              "      <th>Pstatus</th>\n",
              "      <th>Medu</th>\n",
              "      <th>Fedu</th>\n",
              "      <th>Mjob</th>\n",
              "      <th>Fjob</th>\n",
              "      <th>reason</th>\n",
              "      <th>guardian</th>\n",
              "      <th>traveltime</th>\n",
              "      <th>studytime</th>\n",
              "      <th>failures</th>\n",
              "      <th>schoolsup</th>\n",
              "      <th>famsup</th>\n",
              "      <th>paid</th>\n",
              "      <th>activities</th>\n",
              "      <th>nursery</th>\n",
              "      <th>higher</th>\n",
              "      <th>internet</th>\n",
              "      <th>romantic</th>\n",
              "      <th>famrel</th>\n",
              "      <th>freetime</th>\n",
              "      <th>goout</th>\n",
              "      <th>Dalc</th>\n",
              "      <th>Walc</th>\n",
              "      <th>health</th>\n",
              "      <th>absences</th>\n",
              "      <th>G1</th>\n",
              "      <th>G2</th>\n",
              "      <th>G3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>18</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>A</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>at_home</td>\n",
              "      <td>teacher</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>6</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>17</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>at_home</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>father</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>5</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>15</td>\n",
              "      <td>U</td>\n",
              "      <td>LE3</td>\n",
              "      <td>T</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>at_home</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>10</td>\n",
              "      <td>7</td>\n",
              "      <td>8</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>15</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>health</td>\n",
              "      <td>services</td>\n",
              "      <td>home</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>2</td>\n",
              "      <td>15</td>\n",
              "      <td>14</td>\n",
              "      <td>15</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>16</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>home</td>\n",
              "      <td>father</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>6</td>\n",
              "      <td>10</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  school sex  age address famsize Pstatus  ...  Walc  health absences  G1  G2  G3\n",
              "0     GP   F   18       U     GT3       A  ...     1       3        6   5   6   6\n",
              "1     GP   F   17       U     GT3       T  ...     1       3        4   5   5   6\n",
              "2     GP   F   15       U     LE3       T  ...     3       3       10   7   8  10\n",
              "3     GP   F   15       U     GT3       T  ...     1       5        2  15  14  15\n",
              "4     GP   F   16       U     GT3       T  ...     2       5        4   6  10  10\n",
              "\n",
              "[5 rows x 33 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EdMq6oFjcCv5",
        "outputId": "44dd3043-13bd-4562-b9e1-f38478bcd2e3"
      },
      "source": [
        "df_por.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 649 entries, 0 to 648\n",
            "Data columns (total 33 columns):\n",
            " #   Column      Non-Null Count  Dtype \n",
            "---  ------      --------------  ----- \n",
            " 0   school      649 non-null    object\n",
            " 1   sex         649 non-null    object\n",
            " 2   age         649 non-null    int64 \n",
            " 3   address     649 non-null    object\n",
            " 4   famsize     649 non-null    object\n",
            " 5   Pstatus     649 non-null    object\n",
            " 6   Medu        649 non-null    int64 \n",
            " 7   Fedu        649 non-null    int64 \n",
            " 8   Mjob        649 non-null    object\n",
            " 9   Fjob        649 non-null    object\n",
            " 10  reason      649 non-null    object\n",
            " 11  guardian    649 non-null    object\n",
            " 12  traveltime  649 non-null    int64 \n",
            " 13  studytime   649 non-null    int64 \n",
            " 14  failures    649 non-null    int64 \n",
            " 15  schoolsup   649 non-null    object\n",
            " 16  famsup      649 non-null    object\n",
            " 17  paid        649 non-null    object\n",
            " 18  activities  649 non-null    object\n",
            " 19  nursery     649 non-null    object\n",
            " 20  higher      649 non-null    object\n",
            " 21  internet    649 non-null    object\n",
            " 22  romantic    649 non-null    object\n",
            " 23  famrel      649 non-null    int64 \n",
            " 24  freetime    649 non-null    int64 \n",
            " 25  goout       649 non-null    int64 \n",
            " 26  Dalc        649 non-null    int64 \n",
            " 27  Walc        649 non-null    int64 \n",
            " 28  health      649 non-null    int64 \n",
            " 29  absences    649 non-null    int64 \n",
            " 30  G1          649 non-null    int64 \n",
            " 31  G2          649 non-null    int64 \n",
            " 32  G3          649 non-null    int64 \n",
            "dtypes: int64(16), object(17)\n",
            "memory usage: 167.4+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XvDjdwaycJF9",
        "outputId": "3e9ca0a9-f8af-498c-e5f1-00fa887603c8"
      },
      "source": [
        "df_por.columns"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['school', 'sex', 'age', 'address', 'famsize', 'Pstatus', 'Medu', 'Fedu',\n",
              "       'Mjob', 'Fjob', 'reason', 'guardian', 'traveltime', 'studytime',\n",
              "       'failures', 'schoolsup', 'famsup', 'paid', 'activities', 'nursery',\n",
              "       'higher', 'internet', 'romantic', 'famrel', 'freetime', 'goout', 'Dalc',\n",
              "       'Walc', 'health', 'absences', 'G1', 'G2', 'G3'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NMncPcm3cRuV",
        "outputId": "4e9160a7-5855-4586-ba19-0521f2714ea0"
      },
      "source": [
        "len(df_por.columns)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "33"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NN3TdBKlcWJu",
        "outputId": "ac6d2a15-8fb2-441d-eac9-21191ad0e2fe"
      },
      "source": [
        "df_por.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(649, 33)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VcX0eIUhciXC",
        "outputId": "aee1f910-1fa4-4e3c-f06f-ea910b7c786c"
      },
      "source": [
        "df_mat.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(395, 33)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m0-JS_jCk5DX"
      },
      "source": [
        "cleanup_nums = {\"yes\": 1, \"no\": 0, \"F\": 1, \"M\": 0, \"GP\": 1, \"MS\": 0, \"R\": 1, \"U\" : 0, \"GT3\": 1, \"LE3\" :0, \"A\":1, \"T\":0}\n",
        "\n",
        "df_por.replace(cleanup_nums, inplace=True)\n",
        "df_mat.replace(cleanup_nums, inplace=True)\n",
        "\n",
        "\n",
        "#dataMatExt= pd.get_dummies(dataMat, columns=[\"school\", \"sex\", \"address\", \"famsize\", \"Pstatus\", \"Mjob\", \"Fjob\", \"reason\", \"guardian\"])\n",
        "#dataPorExt= pd.get_dummies(dataPor, columns=[\"school\", \"sex\", \"address\", \"famsize\", \"Pstatus\", \"Mjob\", \"Fjob\", \"reason\", \"guardian\"])\n",
        "def encode(df):\n",
        "  if \"Mjob\" in df.columns:\n",
        "    df = pd.get_dummies(df, columns=[ \"Mjob\" ])\n",
        "  if \"Fjob\" in df.columns:\n",
        "    df = pd.get_dummies(df, columns=[ \"Fjob\" ])\n",
        "  if \"reason\" in df.columns:\n",
        "    df = pd.get_dummies(df, columns=[ \"reason\" ])\n",
        "  if \"guardian\" in df.columns:\n",
        "    df = pd.get_dummies(df, columns=[ \"guardian\" ])\n",
        "  return df\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "05xkgL2g8hls"
      },
      "source": [
        "### 5 Level of grade label encoding"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 439
        },
        "id": "0agiZ-9UGw3R",
        "outputId": "9bb150cb-b1e0-4b30-da0e-14910ef15097"
      },
      "source": [
        "def gradesTo5L(x):\n",
        "    if(x<10):\n",
        "        return 0\n",
        "    elif 10 <= x and x < 13:\n",
        "      return 1\n",
        "    elif 13 <= x and x < 16:\n",
        "      return 2\n",
        "    elif 16 <= x and x < 18:\n",
        "      return 3\n",
        "    else:\n",
        "        return 4\n",
        "\n",
        "dataMat5L = df_mat.copy()\n",
        "dataPor5L = df_por.copy()\n",
        "\n",
        "dataMat5L['G3'] = dataMat5L['G3'].map(gradesTo5L)\n",
        "dataMat5L['G2'] = dataMat5L['G2'].map(gradesTo5L)\n",
        "dataMat5L['G1'] = dataMat5L['G1'].map(gradesTo5L)\n",
        "\n",
        "dataPor5L['G3'] = dataPor5L['G3'].map(gradesTo5L)\n",
        "dataPor5L['G2'] = dataPor5L['G2'].map(gradesTo5L)\n",
        "dataPor5L['G1'] = dataPor5L['G1'].map(gradesTo5L)\n",
        "\n",
        "dataPor5L"
      ],
      "execution_count": null,
      "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>school</th>\n",
              "      <th>sex</th>\n",
              "      <th>age</th>\n",
              "      <th>address</th>\n",
              "      <th>famsize</th>\n",
              "      <th>Pstatus</th>\n",
              "      <th>Medu</th>\n",
              "      <th>Fedu</th>\n",
              "      <th>Mjob</th>\n",
              "      <th>Fjob</th>\n",
              "      <th>reason</th>\n",
              "      <th>guardian</th>\n",
              "      <th>traveltime</th>\n",
              "      <th>studytime</th>\n",
              "      <th>failures</th>\n",
              "      <th>schoolsup</th>\n",
              "      <th>famsup</th>\n",
              "      <th>paid</th>\n",
              "      <th>activities</th>\n",
              "      <th>nursery</th>\n",
              "      <th>higher</th>\n",
              "      <th>internet</th>\n",
              "      <th>romantic</th>\n",
              "      <th>famrel</th>\n",
              "      <th>freetime</th>\n",
              "      <th>goout</th>\n",
              "      <th>Dalc</th>\n",
              "      <th>Walc</th>\n",
              "      <th>health</th>\n",
              "      <th>absences</th>\n",
              "      <th>G1</th>\n",
              "      <th>G2</th>\n",
              "      <th>G3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>at_home</td>\n",
              "      <td>teacher</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>17</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>at_home</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>father</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>15</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>at_home</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>6</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>15</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>health</td>\n",
              "      <td>services</td>\n",
              "      <td>home</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>16</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>home</td>\n",
              "      <td>father</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>644</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>19</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>services</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>645</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>teacher</td>\n",
              "      <td>services</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>646</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>647</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>17</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>services</td>\n",
              "      <td>services</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>648</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>services</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>649 rows × 33 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     school  sex  age  address  famsize  ...  health  absences  G1 G2 G3\n",
              "0         1    1   18        0        1  ...       3         4   0  1  1\n",
              "1         1    1   17        0        1  ...       3         2   0  1  1\n",
              "2         1    1   15        0        0  ...       3         6   1  2  1\n",
              "3         1    1   15        0        1  ...       5         0   2  2  2\n",
              "4         1    1   16        0        1  ...       5         0   1  2  2\n",
              "..      ...  ...  ...      ...      ...  ...     ...       ...  .. .. ..\n",
              "644       0    1   19        1        1  ...       5         4   1  1  1\n",
              "645       0    1   18        0        0  ...       1         4   2  2  3\n",
              "646       0    1   18        0        1  ...       5         6   1  1  0\n",
              "647       0    0   17        0        0  ...       2         6   1  1  1\n",
              "648       0    0   18        1        0  ...       5         4   1  1  1\n",
              "\n",
              "[649 rows x 33 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nXLhnW2z8tGp"
      },
      "source": [
        "### 2 Level of grade encoding"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ntVsX14F8rV_",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 439
        },
        "outputId": "c720f07e-1f32-460c-91c0-1e39621de8df"
      },
      "source": [
        "def gradesTo2L(x):\n",
        "    if(x<10):\n",
        "        return 0\n",
        "    else:\n",
        "        return 1\n",
        "\n",
        "dataMat5L = df_mat.copy()\n",
        "dataPor5L = df_por.copy()\n",
        "\n",
        "dataMat5L['G3'] = dataMat5L['G3'].map(gradesTo2L)\n",
        "dataMat5L['G2'] = dataMat5L['G2'].map(gradesTo2L)\n",
        "dataMat5L['G1'] = dataMat5L['G1'].map(gradesTo2L)\n",
        "\n",
        "dataPor5L['G3'] = dataPor5L['G3'].map(gradesTo2L)\n",
        "dataPor5L['G2'] = dataPor5L['G2'].map(gradesTo2L)\n",
        "dataPor5L['G1'] = dataPor5L['G1'].map(gradesTo2L)\n",
        "\n",
        "dataPor5L"
      ],
      "execution_count": null,
      "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>school</th>\n",
              "      <th>sex</th>\n",
              "      <th>age</th>\n",
              "      <th>address</th>\n",
              "      <th>famsize</th>\n",
              "      <th>Pstatus</th>\n",
              "      <th>Medu</th>\n",
              "      <th>Fedu</th>\n",
              "      <th>Mjob</th>\n",
              "      <th>Fjob</th>\n",
              "      <th>reason</th>\n",
              "      <th>guardian</th>\n",
              "      <th>traveltime</th>\n",
              "      <th>studytime</th>\n",
              "      <th>failures</th>\n",
              "      <th>schoolsup</th>\n",
              "      <th>famsup</th>\n",
              "      <th>paid</th>\n",
              "      <th>activities</th>\n",
              "      <th>nursery</th>\n",
              "      <th>higher</th>\n",
              "      <th>internet</th>\n",
              "      <th>romantic</th>\n",
              "      <th>famrel</th>\n",
              "      <th>freetime</th>\n",
              "      <th>goout</th>\n",
              "      <th>Dalc</th>\n",
              "      <th>Walc</th>\n",
              "      <th>health</th>\n",
              "      <th>absences</th>\n",
              "      <th>G1</th>\n",
              "      <th>G2</th>\n",
              "      <th>G3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>at_home</td>\n",
              "      <td>teacher</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>17</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>at_home</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>father</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>15</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>at_home</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>6</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>15</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>health</td>\n",
              "      <td>services</td>\n",
              "      <td>home</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>16</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>home</td>\n",
              "      <td>father</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>644</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>19</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>services</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>645</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>teacher</td>\n",
              "      <td>services</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>646</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>647</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>17</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>services</td>\n",
              "      <td>services</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>6</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>648</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>18</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>services</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>649 rows × 33 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     school  sex  age  address  famsize  ...  health  absences  G1 G2 G3\n",
              "0         1    1   18        0        1  ...       3         4   0  1  1\n",
              "1         1    1   17        0        1  ...       3         2   0  1  1\n",
              "2         1    1   15        0        0  ...       3         6   1  1  1\n",
              "3         1    1   15        0        1  ...       5         0   1  1  1\n",
              "4         1    1   16        0        1  ...       5         0   1  1  1\n",
              "..      ...  ...  ...      ...      ...  ...     ...       ...  .. .. ..\n",
              "644       0    1   19        1        1  ...       5         4   1  1  1\n",
              "645       0    1   18        0        0  ...       1         4   1  1  1\n",
              "646       0    1   18        0        1  ...       5         6   1  1  0\n",
              "647       0    0   17        0        0  ...       2         6   1  1  1\n",
              "648       0    0   18        1        0  ...       5         4   1  1  1\n",
              "\n",
              "[649 rows x 33 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ouHbO7Mo-IQg",
        "outputId": "1d17a16e-2a0e-4606-f032-70e7be398692"
      },
      "source": [
        "copy_data = dataPor5L.copy()\n",
        "Y_label = dataPor5L[\"G3\"]\n",
        "Y_label"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0      1\n",
              "1      1\n",
              "2      1\n",
              "3      1\n",
              "4      1\n",
              "      ..\n",
              "644    1\n",
              "645    1\n",
              "646    0\n",
              "647    1\n",
              "648    1\n",
              "Name: G3, Length: 649, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hTD39skaHLJV"
      },
      "source": [
        "copy_data.drop(\"G3\",axis='columns', inplace=True)\n",
        "X_label = copy_data"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7_CWHsbNHAL9"
      },
      "source": [
        "## 2. Algorithms"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_197dAffXJn9"
      },
      "source": [
        "### 2.1 Decision Tree Classifier"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "au6aFds5_8A9"
      },
      "source": [
        "def DecisionTreeClassifier_Model(X_train, Y_train, X_test, Y_test):\n",
        "\n",
        "    # Create a Decision Tree Classification Model and fit it on the training data\n",
        "    dec_tree_clf = DecisionTreeClassifier().fit(X_train, Y_train)\n",
        "    Y_pred = dec_tree_clf.predict(X_test)\n",
        "    acc_score = accuracy_score(Y_test, Y_pred)\n",
        "    return acc_score\n",
        "\n",
        "#DecisionTreeClassifier_Model(encode(X_train_1),Y_train_1,encode(X_test_1), Y_test_1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XCjUkDZfXNoZ"
      },
      "source": [
        "### 2.2 Bernoulli Naive Bayes Classifier"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QDOKuU4bClLH"
      },
      "source": [
        "def BernoulliNB_Model(X_train, Y_train, X_test, Y_test):\n",
        "\n",
        "    # Create a Bernoulli Naive Bayes Classification Model and fit it on the training data\n",
        "    bernoulli_clf = BernoulliNB().fit(X_train, Y_train)\n",
        "    Y_pred = bernoulli_clf.predict(X_test)\n",
        "    acc_score = accuracy_score(Y_test, Y_pred)\n",
        "    return acc_score\n",
        "   \n",
        "# BernoulliNB_Model(encode(X_train_1),Y_train_1,encode(X_test_1), Y_test_1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "L1wuPUOQXSy_"
      },
      "source": [
        "### 2.3 K Nearest Neighbor Classifier"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZtsITJqMR-1K"
      },
      "source": [
        "def KNearestNeighborsClassifier_Model(X_train, Y_train, X_test, Y_test):\n",
        "\n",
        "    # Create a K - Nearest Neighbor Classification Model and fit it on the training data\n",
        "    k_neighbor_clf = KNeighborsClassifier().fit(X_train, Y_train) \n",
        "    Y_pred = k_neighbor_clf.predict(X_test)\n",
        "    #print(\"F1_score: \",f1_score(Y_test, Y_pred, average=None))\n",
        "    acc_score = accuracy_score(Y_test, Y_pred)\n",
        "    return acc_score\n",
        "\n",
        "# KNearestNeighborsClassifier_Model(encode(X_train_1),Y_train_1, encode(X_test_1), Y_test_1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5crx74xDZXOW"
      },
      "source": [
        "### 2.4 Light GBM Classifier"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0V07cJ75YwwQ"
      },
      "source": [
        "def LightGBM_Model(X_train, Y_train, X_test, Y_test):\n",
        "\n",
        "    # Create a Decision Tree Classification Model and fit it on the training data\n",
        "    lgbm_clf = LGBMClassifier().fit(X_train, Y_train)\n",
        "    Y_pred = lgbm_clf.predict(X_test)\n",
        "    acc_score = accuracy_score(Y_test, Y_pred)\n",
        "    return acc_score\n",
        "\n",
        "# LightGBM_Model(encode(X_train_1),Y_train_1, encode(X_test_1), Y_test_1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_BwlA13Rg0gn"
      },
      "source": [
        "### 2.5 Random Forest Classifier"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g3rUGEo5Ywtl"
      },
      "source": [
        "def RandomForestClassifier_Model(X_train, Y_train, X_test, Y_test):\n",
        "\n",
        "    RandomForest_clf = RandomForestClassifier().fit(X_train, Y_train)\n",
        "    Y_pred = RandomForest_clf.predict(X_test)\n",
        "    acc_score = accuracy_score(Y_test, Y_pred)\n",
        "    return acc_score\n",
        "   \n",
        "# GaussianNB_Model(encode(X_train_1),Y_train_1,encode(X_test_1), Y_test_1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cGdWuDD7oJcP"
      },
      "source": [
        "### 2.6 Gaussian Naive Bayes Classifier"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WW4e6tk6YwqU"
      },
      "source": [
        "def GaussianNB_Model(X_train, Y_train, X_test, Y_test):\n",
        "\n",
        "    # Create a Gaussian Naive Bayes Classification Model and fit it on the training data\n",
        "    Gaussian_clf = GaussianNB().fit(X_train, Y_train)\n",
        "    Y_pred = Gaussian_clf.predict(X_test)\n",
        "    acc_score = accuracy_score(Y_test, Y_pred)\n",
        "    return acc_score"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LOGK8LFUg652"
      },
      "source": [
        "### 2.7 Linear Support Vector Classifier"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wrUVIvXzhCXC"
      },
      "source": [
        "def LinearSVC_Model(X_train, Y_train, X_test, Y_test):\n",
        "\n",
        "    LinearSVC_clf = LinearSVC(max_iter=100000).fit(X_train, Y_train)\n",
        "    Y_pred = LinearSVC_clf.predict(X_test)\n",
        "    acc_score = accuracy_score(Y_test, Y_pred)\n",
        "    return acc_score\n",
        "\n",
        "# LinearSVC_Model(encode(X_train_1),Y_train_1, encode(X_test_1), Y_test_1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ium5ntUPXYiN"
      },
      "source": [
        "### 2.11 Genetic Algorithm"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XiAZUfWuHNTw"
      },
      "source": [
        "#defining various steps required for the genetic algorithm\n",
        "def initilization_of_population(size,n_feat):\n",
        "    population = []\n",
        "    for i in range(size):\n",
        "        chromosome = np.ones(n_feat,dtype=np.bool)\n",
        "        chromosome[:int(0.3*n_feat)]=False\n",
        "        np.random.shuffle(chromosome)\n",
        "        population.append(chromosome)\n",
        "    return population\n",
        "\n",
        "def fitness_score(population, X_train, Y_train, X_test, Y_test, ML_Model):\n",
        "    global parameters_list\n",
        "    scores = []\n",
        "    for chromosome in population:\n",
        "        X_train_2 = encode(X_train.iloc[:,chromosome])\n",
        "        X_test_2 = encode(X_test.iloc[:,chromosome])\n",
        "        parameters_list = list(X_train_2.columns)\n",
        "        scores.append(ML_Model(X_train_2, Y_train, X_test_2, Y_test))\n",
        "    scores, population = np.array(scores), np.array(population) \n",
        "    inds = np.argsort(scores)\n",
        "    return list(scores[inds][::-1]), list(population[inds,:][::-1])\n",
        "\n",
        "def selection(pop_after_fit,n_parents):\n",
        "    population_nextgen = [] \n",
        "    for i in range(n_parents):\n",
        "        population_nextgen.append(pop_after_fit[i])\n",
        "    return population_nextgen\n",
        "\n",
        "def crossover(pop_after_sel):\n",
        "    population_nextgen=pop_after_sel\n",
        "    for i in range(len(pop_after_sel)):\n",
        "        child=pop_after_sel[i]\n",
        "        child[3:7]=pop_after_sel[(i+1)%len(pop_after_sel)][3:7]\n",
        "        population_nextgen.append(child)\n",
        "    return population_nextgen\n",
        "\n",
        "def mutation(pop_after_cross,mutation_rate):\n",
        "    population_nextgen = []\n",
        "    for i in range(0,len(pop_after_cross)):\n",
        "        chromosome = pop_after_cross[i]\n",
        "        for j in range(len(chromosome)):\n",
        "            if random.random() < mutation_rate:\n",
        "                chromosome[j]= not chromosome[j]\n",
        "        population_nextgen.append(chromosome)\n",
        "    #print(population_nextgen)\n",
        "    return population_nextgen\n",
        "\n",
        "def generations(size,n_feat,n_parents,mutation_rate,n_gen,X_train,\n",
        "                                   X_test, y_train, y_test, ML_Model):\n",
        "    best_chromo= []\n",
        "    best_score= []\n",
        "    population_nextgen=initilization_of_population(size,n_feat)\n",
        "    #print(population_nextgen)\n",
        "    for i in range(n_gen):\n",
        "        scores, pop_after_fit = fitness_score(population_nextgen, X_train, y_train, X_test, y_test, ML_Model)\n",
        "        #print(scores[0])\n",
        "        pop_after_sel = selection(pop_after_fit,n_parents)\n",
        "        pop_after_cross = crossover(pop_after_sel)\n",
        "        population_nextgen = mutation(pop_after_cross,mutation_rate)\n",
        "        best_chromo.append(pop_after_fit[0])\n",
        "        best_score.append(scores[0])\n",
        "    return best_chromo,best_score\n",
        "\n",
        "def GeneticAlgorithm(size,n_feat,n_parents,mutation_rate,n_gen,X_train,\n",
        "                                   X_test, y_train, y_test, ML_Model=BernoulliNB_Model):\n",
        "    best_parameters, best_scores = generations(size,n_feat,n_parents,mutation_rate,n_gen,X_train, X_test, y_train, y_test, ML_Model)\n",
        "    # Print the parameters\n",
        "    global parameters_list\n",
        "    top_chromosome = best_parameters[0]\n",
        "    print(\"The choosen chromosome in raw format :\")\n",
        "    print(top_chromosome)\n",
        "    choosen_parameters = [parameters_list[i] for i in range(len(parameters_list)) if top_chromosome[i]]\n",
        "    print(\"The choosen set of parameters after feature selection :\")\n",
        "    for i in range(len(choosen_parameters)) :\n",
        "        print(choosen_parameters[i])\n",
        "    \n",
        "    # Returns the best accuracy score\n",
        "    return max(best_scores)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JdewUmCBbEnU",
        "outputId": "ee8271e3-983e-45e9-9589-7641ca2d81b9"
      },
      "source": [
        "parameters_list = []\n",
        "\n",
        "GeneticAlgorithm(\n",
        "    size=20, n_feat=len(X_train_1.columns), n_parents=10, mutation_rate=0.10, n_gen=38, \n",
        "    X_train=X_train_1, X_test=X_test_1, y_train=Y_train_1, y_test=Y_test_1, ML_Model=BernoulliNB_Model\n",
        ")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The choosen chromosome in raw format :\n",
            "[ True False  True  True  True  True False  True  True  True  True  True\n",
            " False  True  True False  True  True  True  True False  True False  True\n",
            " False  True False False  True False  True  True]\n",
            "The choosen set of parameters after feature selection :\n",
            "school\n",
            "age\n",
            "address\n",
            "famsize\n",
            "Pstatus\n",
            "schoolsup\n",
            "activities\n",
            "nursery\n",
            "internet\n",
            "freetime\n",
            "Dalc\n",
            "Walc\n",
            "G2\n",
            "Mjob_at_home\n",
            "Mjob_health\n",
            "Mjob_other\n",
            "Mjob_teacher\n",
            "Fjob_health\n",
            "Fjob_services\n",
            "guardian_mother\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4XglJ_W5ZdzN"
      },
      "source": [
        "### 2.12 Results Analysis"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P2LWykFTZdTC"
      },
      "source": [
        "class MLResultsAnalysis :\n",
        "\n",
        "    def __init__(self) :\n",
        "        \"\"\"\n",
        "        Initializes a dictoonary to store the results of all algorithms for a particular scenario\n",
        "        \"\"\"\n",
        "\n",
        "        # Initialize a variable to store the overall results\n",
        "        self.results = {}\n",
        "\n",
        "\n",
        "    def addResult(self, algo_name, accuracy_score_value) :\n",
        "        \"\"\"\n",
        "        Given the prediction values of an algorithm, calculates and stores the metric scores of the prediction made.\n",
        "\n",
        "        Parameters\n",
        "        ----------\n",
        "        algo_name - Name of the ML Algorithm which has invoked this function\n",
        "        accuracy_score_value - The value of accuracy obtained by the algorithm\n",
        "        \"\"\"\n",
        "\n",
        "        # Store the accuracy score of this algorithm into the class variable\n",
        "        self.results[algo_name] = accuracy_score_value\n",
        "    \n",
        "\n",
        "    def plotScores(self) :\n",
        "        \"\"\"\n",
        "        After calculating the scores of all the algorithms, plots the scores in a graph.\n",
        "        \"\"\"\n",
        "\n",
        "        # Initialize an empty array to stores the accuracy scores of all algorithms\n",
        "        scores = []\n",
        "\n",
        "        # Initialize an array to store the names of all the algorithms used\n",
        "        algorithms_available = []\n",
        "\n",
        "        # Fill the algorithms and scores data structure\n",
        "        for key, value in self.results.items() :\n",
        "            algorithms_available.append(key)\n",
        "            scores.append(value)\n",
        "\n",
        "        # Plot the accuracy scores\n",
        "        plt.plot(algorithms_available, scores)\n",
        "\n",
        "        # Set the labels and titles\n",
        "        plt.xlabel(\"ML Algorithms\")\n",
        "        plt.ylabel(\"accuracy score\")\n",
        "        plt.title(\"Results\")\n",
        "    \n",
        "\n",
        "    def displayResults(self) :\n",
        "        \"\"\"\n",
        "        After the calculations, this function prints the results of all the algorithms and their metrics.\n",
        "        \"\"\"\n",
        "\n",
        "        # Print the results for each algorithm\n",
        "        for key, value in self.results.items() :\n",
        "            #name = \" \".join([word.capitalize() for word in key.split(\"_\")]) + \" Classifier\"\n",
        "            name = key + \" Classifier\"\n",
        "            print(name)\n",
        "            print(\"--------------------------------\")\n",
        "            print(f\"Accuracy : {value}\")\n",
        "            print(\"---------------------------------------------------------------------------------------------\\n\\n\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SqAlJSyIZdP7"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EWUIC3YCXr8M"
      },
      "source": [
        "## 3. Training and Validation"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bG3qVN1-o-J2"
      },
      "source": [
        "### 3.1 Training Utility Function"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6WWM03-jpEIH"
      },
      "source": [
        "# The various options available to us\n",
        "# Each model will be trained and validated on its own and as well as with genetic algorithm\n",
        "\n",
        "options = {\n",
        "    \"Decision Tree\" : DecisionTreeClassifier_Model,\n",
        "    \"Bernoulli NB\" : BernoulliNB_Model,\n",
        "    \"Gaussian NB\" : GaussianNB_Model,\n",
        "    \"K Nearest Neighbor\" : KNearestNeighborsClassifier_Model,\n",
        "    \"Light GBM\" : LightGBM_Model,\n",
        "    \"Random Forest\" : RandomForestClassifier_Model,\n",
        "    \"Linear SVC\" : LinearSVC_Model\n",
        "}"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "paF1QZulp02T"
      },
      "source": [
        "def trainAndValidateModels(results_obj, options, dataset) :\n",
        "\n",
        "    (X_train, X_test, Y_train, Y_test) = dataset\n",
        "\n",
        "    for name, model in options.items() :\n",
        "        results_obj.addResult(name, model(encode(X_train),Y_train,encode(X_test), Y_test))\n",
        "        results_obj.addResult(\n",
        "            (name + \" with Genetics\"),\n",
        "            GeneticAlgorithm(\n",
        "                size=20, n_feat=len(X_train.columns), n_parents=10, mutation_rate=0.10, n_gen=38, \n",
        "                X_train=X_train, X_test=X_test, y_train=Y_train, y_test=Y_test, ML_Model=model\n",
        "            )\n",
        "        )"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zDes79-1o0JN"
      },
      "source": [
        "### 3.2 Using General Train-Test Split"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3EUn8WoFr4eI"
      },
      "source": [
        "# Perform a test train split with 10% as test set\n",
        "\n",
        "X_train_1, X_test_1, Y_train_1, Y_test_1 = train_test_split(\n",
        "    X_label,\n",
        "    Y_label, \n",
        "    test_size = 0.1,\n",
        "    random_state=42\n",
        ")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TYqmxnNLiXvC"
      },
      "source": [
        "### Classification for 5-Level without K-Fold verification"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LNXkmzufiDes",
        "outputId": "817677e2-a8a3-4a5d-a630-0fe699f9bd63"
      },
      "source": [
        "# Perform the analysis\n",
        "\n",
        "dataset = (X_train_1, X_test_1, Y_train_1, Y_test_1)\n",
        "test_train_split_results = MLResultsAnalysis()\n",
        "trainAndValidateModels(test_train_split_results, options, dataset)\n",
        "test_train_split_results.displayResults()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Decision Tree Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.6153846153846154\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Decision Tree with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.7846153846153846\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.46153846153846156\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.6307692307692307\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.3230769230769231\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.7538461538461538\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.4461538461538462\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.6923076923076923\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.676923076923077\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.6923076923076923\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.7846153846153846\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.5384615384615384\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.7538461538461538\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 476
        },
        "id": "ARy_6wPVwBzZ",
        "outputId": "1df91ab7-0805-4de3-88b6-95ec855b8884"
      },
      "source": [
        "# Set the size of plot\n",
        "plt.rcParams[\"figure.figsize\"] = (30,10)\n",
        "\n",
        "# Plot the scores obtained by these algorithms\n",
        "test_train_split_results.plotScores()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 2160x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GVojlFnruq2w"
      },
      "source": [
        "### 3.3 Using K- Fold Cross Validation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EOrNhk5lGTtX"
      },
      "source": [
        "# Initialize the cross validator\n",
        "\n",
        "kf = KFold(n_splits = 10, shuffle = True, random_state = 2)\n",
        "split_indices = next(kf.split(X_label))\n",
        "\n",
        "X_train_2 = X_label.iloc[split_indices[0]]\n",
        "Y_train_2 = Y_label.iloc[split_indices[0]]\n",
        "X_test_2 = X_label.iloc[split_indices[1]]\n",
        "Y_test_2 = Y_label.iloc[split_indices[1]]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "saStRtRFiBTN"
      },
      "source": [
        "### Classification for 2-Level with using K-fold Validation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "background_save": true
        },
        "id": "1jeYRN1-wzla",
        "outputId": "fe641e64-4e50-42d4-98b2-68cf648ed2e2"
      },
      "source": [
        "# Perform the analysis\n",
        "\n",
        "dataset = (X_train_2, X_test_2, Y_train_2, Y_test_2)\n",
        "k_fold_split_results = MLResultsAnalysis()\n",
        "trainAndValidateModels(k_fold_split_results, options, dataset)\n",
        "k_fold_split_results.displayResults()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Decision Tree Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9076923076923077\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Decision Tree with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9846153846153847\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8923076923076924\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9538461538461539\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8923076923076924\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9538461538461539\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8153846153846154\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9384615384615385\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9384615384615385\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9692307692307692\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9384615384615385\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9846153846153847\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8923076923076924\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9384615384615385\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FeCGATYExO1X"
      },
      "source": [
        "# Set the size of plot\n",
        "plt.rcParams[\"figure.figsize\"] = (30,10)\n",
        "\n",
        "# Plot the scores obtained by these algorithms\n",
        "k_fold_split_results.plotScores()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Eo4fsy8_HEUO"
      },
      "source": [
        "## 8. Other Ideas"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IP-3WxDu_Gt3"
      },
      "source": [
        "For PCA \n",
        "* https://github.com/sharmaroshan/Students-Performance-Analytics/blob/master/Student_Performance.ipynb\n",
        "\n",
        "For SVR and NN(sklearn based)\n",
        "* https://github.com/divyansh21april/Student-Performance-Analysis-/blob/master/Student%20Performance%20Analysis.ipynb\n",
        "* https://github.com/Tsmith5151/udacity-student-intervention/blob/master/student_intervention.ipynb\n",
        "\n",
        "For Feature Selection\n",
        "* https://github.com/sachanganesh/student-performance-prediction/blob/master/model.py\n",
        "\n",
        "Neural Network\n",
        "* https://github.com/sean-huni/deep-learning/blob/master/part-1/deepl-stud.py"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fPeaIGiNHI8x"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IYtWMzOvLQ3s"
      },
      "source": [
        "## Prevent random disconnects\n",
        "\n",
        "This cell runs JS code to automatic reconnect to runtime."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "id": "jtClEMAMLVHw",
        "outputId": "379a3e29-5936-447c-f64b-7b1f022cb5e0"
      },
      "source": [
        "import IPython\n",
        "from google.colab import output\n",
        "\n",
        "display(IPython.display.Javascript('''\n",
        " function ClickConnect(){\n",
        "   btn = document.querySelector(\"colab-connect-button\")\n",
        "   if (btn != null){\n",
        "     console.log(\"Click colab-connect-button\"); \n",
        "     btn.click() \n",
        "     }\n",
        "   \n",
        "   btn = document.getElementById('ok')\n",
        "   if (btn != null){\n",
        "     console.log(\"Click reconnect\"); \n",
        "     btn.click() \n",
        "     }\n",
        "  }\n",
        "  \n",
        "setInterval(ClickConnect,60000)\n",
        "'''))\n",
        "\n",
        "print(\"Done.\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/javascript": [
              "\n",
              " function ClickConnect(){\n",
              "   btn = document.querySelector(\"colab-connect-button\")\n",
              "   if (btn != null){\n",
              "     console.log(\"Click colab-connect-button\"); \n",
              "     btn.click() \n",
              "     }\n",
              "   \n",
              "   btn = document.getElementById('ok')\n",
              "   if (btn != null){\n",
              "     console.log(\"Click reconnect\"); \n",
              "     btn.click() \n",
              "     }\n",
              "  }\n",
              "  \n",
              "setInterval(ClickConnect,60000)\n"
            ],
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Done.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hBEugqdJgZAe"
      },
      "source": [
        "### Classification with 5-Level with k-fold validation`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hAD0v6WYEXbU",
        "outputId": "72bff987-2780-49b6-ea01-1743493d73e0"
      },
      "source": [
        "dataset = (X_train_2, X_test_2, Y_train_2, Y_test_2)\n",
        "k_fold_split_results = MLResultsAnalysis()\n",
        "trainAndValidateModels(k_fold_split_results, options, dataset)\n",
        "k_fold_split_results.displayResults()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Decision Tree Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.6461538461538462\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Decision Tree with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.6615384615384615\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.6923076923076923\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.27692307692307694\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.7538461538461538\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.5692307692307692\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.7846153846153846\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.7692307692307693\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8307692307692308\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8307692307692308\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.6923076923076923\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8153846153846154\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pT4qgo4MeA5-"
      },
      "source": [
        "### CLASSIFICATION ALGORITHM FOR 2 LEVEL WITHOUT K-FOLD VALIDATION"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XqoLCgTuEdRo",
        "outputId": "3d1fd255-8c99-49b4-db0d-e7d499d1daf4"
      },
      "source": [
        "dataset = (X_train_1, X_test_1, Y_train_1, Y_test_1)\n",
        "test_train_split_results = MLResultsAnalysis()\n",
        "trainAndValidateModels(test_train_split_results, options, dataset)\n",
        "test_train_split_results.displayResults()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Decision Tree Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9076923076923077\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Decision Tree with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 1.0\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9230769230769231\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9692307692307692\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9076923076923077\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9692307692307692\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8923076923076924\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9692307692307692\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9230769230769231\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9846153846153847\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8923076923076924\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9846153846153847\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9384615384615385\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9846153846153847\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DQ2ZWX7xgFQu"
      },
      "source": [
        "### Classification 2-Level without G2 and G1, without k-fold valiation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_gSAniuvd6xD",
        "outputId": "e6d5d7f6-b10a-4b4a-b665-f362a7f8ccb1"
      },
      "source": [
        "X_train_1.drop(\"G2\",axis='columns', inplace=True)\n",
        "X_train_1.drop(\"G1\",axis='columns', inplace=True)\n",
        "X_test_1.drop(\"G2\",axis='columns', inplace=True)\n",
        "X_test_1.drop(\"G1\",axis='columns', inplace=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/pandas/core/frame.py:4174: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  errors=errors,\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iEPZwWqWfsPN",
        "outputId": "65076000-d22e-441c-c0b5-632591cbc5f2"
      },
      "source": [
        "dataset = (X_train_1, X_test_1, Y_train_1, Y_test_1)\n",
        "test_train_split_results = MLResultsAnalysis()\n",
        "trainAndValidateModels(test_train_split_results, options, dataset)\n",
        "test_train_split_results.displayResults()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Decision Tree Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8461538461538461\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Decision Tree with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9230769230769231\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9230769230769231\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Bernoulli NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9384615384615385\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8615384615384616\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Gaussian NB with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9384615384615385\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8769230769230769\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "K Nearest Neighbor with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9384615384615385\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8923076923076924\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Light GBM with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9538461538461539\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.8923076923076924\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Random Forest with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9538461538461539\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9230769230769231\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n",
            "Linear SVC with Genetics Classifier\n",
            "--------------------------------\n",
            "Accuracy : 0.9384615384615385\n",
            "---------------------------------------------------------------------------------------------\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vOnrm4C7f6jp",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "1423f40d-c0e8-432d-e921-ce13b3ce8268"
      },
      "source": [
        "import random\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from matplotlib.font_manager import FontProperties\n",
        "\n",
        "fontP = FontProperties()\n",
        "fontP.set_size('small')\n",
        "\n",
        "index = np.arange(7)\n",
        "bar_width = 0.35\n",
        "\n",
        "fig, ax = plt.subplots()\n",
        "summer = ax.bar(index, [0.815, 0.907, 0.892, 0.892, 0.938, 0.892, 0.938  ], bar_width,\n",
        "                label=\"Without Genetic Algorithm\")\n",
        "\n",
        "winter = ax.bar(index+bar_width, [0.938, 0.984, 0.953, 0.953, 0.969, 0.938, 0.9846],\n",
        "                 bar_width, label=\"With Genetic Algorithm\")\n",
        "\n",
        "ax.set_xlabel('Classification Algorithm')\n",
        "ax.set_ylabel('Accuracy')\n",
        "ax.set_title('Accuracy score for 2 Level Classification of Grades')\n",
        "ax.set_xticks(index + bar_width / 2)\n",
        "ax.set_xticklabels([\"KNN\", \"J48\", \"BNB\", \"GNB\", \"LGBM\",\"LSVC\",\"RF\"])\n",
        "ax.legend(prop=fontP)\n",
        "\n",
        "plt.ylim(0.7, 1.05)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "2Ts7hmch4dFx",
        "outputId": "d8774984-e058-4a5d-a91a-06891167a164"
      },
      "source": [
        "import random\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "\n",
        "index = np.arange(7)\n",
        "bar_width = 0.35\n",
        "\n",
        "fig, ax = plt.subplots()\n",
        "summer = ax.bar(index, [0.569, 0.646, 0.661, 0.276,0.769, 0.692, 0.800  ], bar_width,\n",
        "                label=\"Without Genetic Algorithm\")\n",
        "\n",
        "winter = ax.bar(index+bar_width, [0.784, 0.800, 0.692, 0.753, 0.830, 0.815, 0.831],\n",
        "                 bar_width, label=\"With Genetic Algorithm\")\n",
        "\n",
        "ax.set_xlabel('Classification Algorithm')\n",
        "ax.set_ylabel('Accuracy')\n",
        "ax.set_title('Accuracy score for 5 Level Classification of Grades')\n",
        "ax.set_xticks(index + bar_width / 2)\n",
        "ax.set_xticklabels([\"KNN\", \"J48\", \"BNB\", \"GNB\", \"LGBM\",\"LSVC\",\"RF\"])\n",
        "ax.legend()\n",
        "\n",
        "plt.ylim(0, 1.2)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1Dfiiw075nRh"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}