FinalPRMLcode.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "FinalPRMLcode.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/vaishnavipatil29/Career-Guidance-ML-Project/blob/master/FinalPRMLcode.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5Ce14OgkfaAc"
      },
      "source": [
        "**Career Guidance System**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 316
        },
        "id": "VREYLjokfSaN",
        "outputId": "85f55979-5173-4f3f-abd6-978fb3c60674"
      },
      "source": [
        "#importing data\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "url='https://raw.githubusercontent.com/Nantha-1998/404-CODER/master/career_compute_dataset.csv'\n",
        "dataset = pd.read_csv(url)\n",
        "print(np.shape(dataset))\n",
        "dataset.head()"
      ],
      "execution_count": 141,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(7525, 26)\n"
          ],
          "name": "stdout"
        },
        {
          "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>sslc</th>\n",
              "      <th>hsc</th>\n",
              "      <th>cgpa</th>\n",
              "      <th>school_type</th>\n",
              "      <th>no_of_miniprojects</th>\n",
              "      <th>no_of_projects</th>\n",
              "      <th>coresub_skill</th>\n",
              "      <th>aptitude_skill</th>\n",
              "      <th>problemsolving_skill</th>\n",
              "      <th>programming_skill</th>\n",
              "      <th>abstractthink_skill</th>\n",
              "      <th>design_skill</th>\n",
              "      <th>first_computer</th>\n",
              "      <th>first_program</th>\n",
              "      <th>lab_programs</th>\n",
              "      <th>ds_coding</th>\n",
              "      <th>technology_used</th>\n",
              "      <th>sympos_attend</th>\n",
              "      <th>sympos_won</th>\n",
              "      <th>extracurricular</th>\n",
              "      <th>learning_style</th>\n",
              "      <th>college_bench</th>\n",
              "      <th>clg_teachers_know</th>\n",
              "      <th>college_performence</th>\n",
              "      <th>college_skills</th>\n",
              "      <th>ROLE</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>Technical Support</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>Software Developer</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>UI/UX Designer</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>2</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>2</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>Technical Support</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>Data Analyst</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   sslc  hsc  cgpa  ...  college_performence   college_skills                ROLE\n",
              "0     2    3     2  ...                     4               5   Technical Support\n",
              "1     3    1     2  ...                     3               4  Software Developer\n",
              "2     3    4     2  ...                     3               4      UI/UX Designer\n",
              "3     3    4     2  ...                     4               3   Technical Support\n",
              "4     2    3     2  ...                     3               1        Data Analyst\n",
              "\n",
              "[5 rows x 26 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 141
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7c1CCk-EhQic",
        "outputId": "8f7d0bb9-49bd-4825-c1f0-58b142a5d18f"
      },
      "source": [
        "data = dataset.iloc[:49,:-1].values\n",
        "label = dataset.iloc[:49,-1]\n",
        "#-------------------------------Label Encoding--------------------------#\n",
        "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
        "labelencoder = LabelEncoder()\n",
        "df = dataset\n",
        "label = df.iloc[:49,-1]\n",
        "original=label.unique() \n",
        "label=label.values\n",
        "label2 = labelencoder.fit_transform(label)\n",
        "y=pd.DataFrame(label2,columns=[\"ROLE\"])\n",
        "numeric=y[\"ROLE\"].unique() \n",
        "y1 = pd.DataFrame({'ROLE':original, 'Associated Number':numeric})\n",
        "print(y1)\n",
        "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
        "labelencoder = LabelEncoder()\n",
        "label = labelencoder.fit_transform(label)\n",
        "y=pd.DataFrame(label,columns=[\"role\"])\n",
        "X = pd.DataFrame(data,columns=['sslc','hsc','cgpa','school_type','no_of_miniprojects','no_of_projects',\n",
        "                                'coresub_skill','aptitude_skill','problemsolving_skill','programming_skill','abstractthink_skill',\n",
        "                                'design_skill','first_computer','first_program','lab_programs','ds_coding','technology_used',\n",
        "                                'sympos_attend','sympos_won','extracurricular','learning_style','college_bench','clg_teachers_know','college_performence','college_skills'])"
      ],
      "execution_count": 175,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "                 ROLE  Associated Number\n",
            "0   Technical Support                  4\n",
            "1  Software Developer                  2\n",
            "2      UI/UX Designer                  6\n",
            "3        Data Analyst                  1\n",
            "4    Technical Writer                  5\n",
            "5       Web Developer                  7\n",
            "6     Software Tester                  3\n",
            "7    Business Analyst                  0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YZzTh9Yrf_x6"
      },
      "source": [
        "Thus, we have categorical data and categorical target.\n",
        "\n",
        "For  Feature Selection : \n",
        "\n",
        "We propose to use: \n",
        "\n",
        "1. Chi-Squared Statistic.\n",
        "2. Mutual Information Statistic.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EkszPWfkf_dz"
      },
      "source": [
        "\n",
        "from pandas import read_csv\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.preprocessing import OrdinalEncoder\n",
        "from sklearn.feature_selection import SelectKBest\n",
        "from sklearn.feature_selection import chi2\n",
        "from matplotlib import pyplot"
      ],
      "execution_count": 143,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "An3KlNcZglav",
        "outputId": "be6e46b5-63a0-44d1-a8a3-e8be21b8a621"
      },
      "source": [
        "# feature selection\n",
        "# example of chi squared feature selection for categorical data\n",
        "def select_features(X_train, y_train, X_test):\n",
        "\tfs = SelectKBest(score_func=chi2, k='all')\n",
        "\tfs.fit(X_train, y_train)\n",
        "\tX_train_fs = fs.transform(X_train)\n",
        "\tX_test_fs = fs.transform(X_test)\n",
        "\treturn X_train_fs, X_test_fs, fs\n",
        "\n",
        "# split into train and test sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=20)#Decision tree\n",
        "X_train2,X_test2,y_train2,y_test2=train_test_split(X,y,test_size=0.3,random_state=10)#XGBoost\n",
        "X_train6,X_test6,y_train6,y_test6=train_test_split(X,y,test_size=0.2,random_state=15)#SVM\n",
        "\n",
        "X_train1, X_test1, fs1 = select_features(X_train, y_train, X_test)\n",
        "X_train3, X_test3, fs3 = select_features(X_train2, y_train2, X_test2)\n",
        "X_train5, X_test5, fs5 = select_features(X_train2, y_train2, X_test2)\n",
        "\n",
        "# what are scores for the features\n",
        "for i in range(len(fs1.scores_)):\n",
        "\tprint('Feature %d: %f' % (i, fs.scores_[i]))\n",
        "# plot the scores\n",
        "pyplot.bar([i for i in range(len(fs.scores_))], fs.scores_)\n",
        "pyplot.show()\n",
        "\n",
        "count = 0\n",
        "for i in fs.scores_ : \n",
        "    if i > 1 : \n",
        "        count = count + 1\n",
        "k1=count\n",
        "print(\" The number of important features with threshold as 1 :{} \".format(k1))\n",
        "\n",
        "\n",
        "for i in range(len(fs3.scores_)):\n",
        "\tprint('Feature %d: %f' % (i, fs3.scores_[i]))\n",
        "# plot the scores\n",
        "pyplot.bar([i for i in range(len(fs3.scores_))], fs3.scores_)\n",
        "pyplot.show()\n",
        "\n",
        "count = 0\n",
        "for i in fs3.scores_ : \n",
        "    if i > 1 : \n",
        "        count = count + 1\n",
        "k3=count\n",
        "\n",
        "print(\" The number of important features with threshold as 1 :{} \".format(k3))"
      ],
      "execution_count": 144,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Feature 0: 0.056432\n",
            "Feature 1: 0.026606\n",
            "Feature 2: 0.264380\n",
            "Feature 3: 0.381868\n",
            "Feature 4: 1.256479\n",
            "Feature 5: 1.807401\n",
            "Feature 6: 1.264217\n",
            "Feature 7: 0.566296\n",
            "Feature 8: 0.000000\n",
            "Feature 9: 0.000000\n",
            "Feature 10: 0.676191\n",
            "Feature 11: 0.898628\n",
            "Feature 12: 1.021494\n",
            "Feature 13: 0.000000\n",
            "Feature 14: 1.712591\n",
            "Feature 15: 0.273604\n",
            "Feature 16: 2.650362\n",
            "Feature 17: 0.150529\n",
            "Feature 18: 0.279142\n",
            "Feature 19: 0.146479\n",
            "Feature 20: 0.246907\n",
            "Feature 21: 0.181558\n",
            "Feature 22: 0.243139\n",
            "Feature 23: 0.814039\n",
            "Feature 24: 0.079684\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAANH0lEQVR4nO3db4hl9X3H8fenuumDKE3KDomsq5O20gcpbbWLJiQUoaT4J8SWSqoPTAwNW4JShTyo8YGGQMGWVkpqULZVosWalGjtBjekQgXNg4izy8Y/u9huwwZ32eqoxT8kNGzz7YM5ptPJ3L13Z+/M7Hzn/YJh7jnnt+d+f3vWj7/53fM7k6pCkrTx/dx6FyBJmg4DXZKaMNAlqQkDXZKaMNAlqYkz1+uNt27dWrOzs+v19pK0Ie3du/fVqppZ7ti6Bfrs7Cxzc3Pr9faStCEl+cGoY065SFITBrokNWGgS1ITBrokNWGgS1ITBrokNWGgS1ITBrokNWGgS1IT67ZSVNL0zN7y2ETtDt9x5SpXovXkCF2SmjDQJakJA12SmjDQJakJA12SmjDQJakJA12Smhgb6Em2J3kiyYEkLyS5aZk2lyZ5I8n+4eu21SlXkjTKJAuLjgOfr6p9Sc4G9iZ5vKoOLGn3VFV9fPolSpImMXaEXlXHqmrf8Pot4CCwbbULkySdnJOaQ08yC1wIPL3M4Q8n+V6SbyX54Ig/vzPJXJK5+fn5ky5WkjTaxIGe5CzgYeDmqnpzyeF9wPlV9RvA3wCPLneOqtpVVTuqasfMzMxKa5YkLWOiQE+yhYUwf7CqHll6vKrerKq3h9d7gC1Jtk61UknSCU1yl0uAe4GDVXXniDbvH9qR5OLhvK9Ns1BJ0olNcpfLR4DrgOeS7B/23QqcB1BV9wBXA59Lchz4EXBNVdUq1CtJGmFsoFfVd4CMaXMXcNe0ipIknTxXikpSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDVhoEtSEwa6JDUxNtCTbE/yRJIDSV5IctMybZLky0kOJXk2yUWrU64kaZQzJ2hzHPh8Ve1LcjawN8njVXVgUZvLgQuGr0uAu4fvkqQ1MnaEXlXHqmrf8Pot4CCwbUmzq4AHasF3gfckOWfq1UqSRjqpOfQks8CFwNNLDm0DXlq0fYSfDX1J0iqaONCTnAU8DNxcVW+u5M2S7Ewyl2Rufn5+JaeQJI0wUaAn2cJCmD9YVY8s0+QosH3R9rnDvv+nqnZV1Y6q2jEzM7OSeiVJI0xyl0uAe4GDVXXniGa7gU8Nd7t8CHijqo5NsU5J0hiT3OXyEeA64Lkk+4d9twLnAVTVPcAe4ArgEPBD4DPTL1WSdCJjA72qvgNkTJsCbphWUZKkk+dKUUlqwkCXpCYMdElqwkCXpCYMdElqwkCXpCYMdElqwkCXpCYMdElqwkCXpCYmeZaLThOztzw2UbvDd1y5ypVIOh05QpekJgx0SWrCQJekJgx0SWrCQJekJgx0SWrC2xalVebtplorjtAlqQkDXZKaMNAlqQkDXZKaMNAlqQkDXZKaMNAlqQkDXZKaMNAlqQkDXZKaMNAlqQkDXZKaGBvoSe5L8kqS50ccvzTJG0n2D1+3Tb9MSdI4kzxt8avAXcADJ2jzVFV9fCoVSZJWZOwIvaqeBF5fg1okSadgWnPoH07yvSTfSvLBKZ1TknQSpvELLvYB51fV20muAB4FLliuYZKdwE6A8847bwpvLUl6xymP0Kvqzap6e3i9B9iSZOuItruqakdV7ZiZmTnVt5YkLXLKgZ7k/UkyvL54OOdrp3peSdLJGTvlkuQh4FJga5IjwO3AFoCquge4GvhckuPAj4BrqqpWrWJJ0rLGBnpVXTvm+F0s3NYoSVpHrhSVpCYMdElqwkCXpCYMdElqwkCXpCYMdElqwkCXpCYMdElqwkCXpCYMdElqwkCXpCYMdElqwkCXpCam8RuLdJqaveWxidodvuPKVa5E0low0NfRJIFr2EqalFMuktSEgS5JTRjoktSEgS5JTRjoktSEgS5JTRjoktSEgS5JTRjoktSEgS5JTRjoktSEgS5JTRjoktSEgS5JTRjoktSEgS5JTYwN9CT3JXklyfMjjifJl5McSvJskoumX6YkaZxJfmPRV4G7gAdGHL8cuGD4ugS4e/gubQj+qj51MXaEXlVPAq+foMlVwAO14LvAe5KcM60CJUmTmcYc+jbgpUXbR4Z9PyPJziRzSebm5+en8NaSpHes6YeiVbWrqnZU1Y6ZmZm1fGtJam8agX4U2L5o+9xhnyRpDU0j0HcDnxrudvkQ8EZVHZvCeSVJJ2HsXS5JHgIuBbYmOQLcDmwBqKp7gD3AFcAh4IfAZ1arWEnSaGMDvaquHXO8gBumVpEkaUVcKSpJTRjoktTEJCtFpQ3DVZ/azByhS1ITBrokNWGgS1ITBrokNWGgS1IT3uUiSSNstLumHKFLUhMGuiQ1YaBLUhMGuiQ1YaBLUhMGuiQ1YaBLUhMGuiQ1YaBLUhMGuiQ14dJ/ndY22tJraT05QpekJgx0SWrCQJekJgx0SWrCQJekJgx0SWrCQJekJgx0SWrChUX6KRfxSBubI3RJamKiQE9yWZIXkxxKcssyx69PMp9k//D12emXKkk6kbFTLknOAL4CfAw4AjyTZHdVHVjS9OtVdeMq1ChJmsAkI/SLgUNV9f2q+jHwNeCq1S1LknSyJgn0bcBLi7aPDPuW+oMkzyb5RpLty50oyc4kc0nm5ufnV1CuJGmUaX0o+k1gtqp+HXgcuH+5RlW1q6p2VNWOmZmZKb21JAkmC/SjwOIR97nDvp+qqteq6r+Hzb8Dfms65UmSJjVJoD8DXJDkA0neBVwD7F7cIMk5izY/ARycXomSpEmMvculqo4nuRH4NnAGcF9VvZDkS8BcVe0G/iTJJ4DjwOvA9atYsyRpGROtFK2qPcCeJftuW/T6C8AXpluaJOlkuFJUkpow0CWpCQNdkprwaYtT4pMKJa03R+iS1ISBLklNGOiS1ISBLklNGOiS1ISBLklNGOiS1IT3oUunoUnWNazlmgbXWWwMBvoI/gOWtNE45SJJTThCl3RaWO2fijfDT92O0CWpCQNdkpow0CWpCQNdkprwQ1FJG9Jm+JDzZDlCl6QmHKFLm9BajG5Pt9Wum4EjdElqwkCXpCYMdElqwkCXpCYMdElqwrtcJGlK1vve+A0Z6Ov9lyZJpyOnXCSpiYkCPcllSV5McijJLcsc//kkXx+OP51kdtqFSpJObOyUS5IzgK8AHwOOAM8k2V1VBxY1+yPgv6rqV5JcA/w58IerUfBKOU0jqbtJRugXA4eq6vtV9WPga8BVS9pcBdw/vP4G8DtJMr0yJUnjpKpO3CC5Grisqj47bF8HXFJVNy5q8/zQ5siw/R9Dm1eXnGsnsHPY/FXgxWl1BNgKvDq2VT+bsd/2efPYjP0e1+fzq2pmuQNrepdLVe0Cdq3GuZPMVdWO1Tj36Wwz9ts+bx6bsd+n0udJplyOAtsXbZ877Fu2TZIzgV8AXltJQZKklZkk0J8BLkjygSTvAq4Bdi9psxv49PD6auBfa9xcjiRpqsZOuVTV8SQ3At8GzgDuq6oXknwJmKuq3cC9wN8nOQS8zkLor7VVmcrZADZjv+3z5rEZ+73iPo/9UFSStDG4UlSSmjDQJamJFoE+7tEEHSU5nOS5JPuTzK13PaslyX1JXhnWOryz7xeTPJ7k34fv713PGqdtRJ+/mOTocL33J7liPWuctiTbkzyR5ECSF5LcNOzvfq1H9XtF13vDz6EPjyb4NxY9mgC4dsmjCdpJchjYsXTxVjdJfht4G3igqn5t2PcXwOtVdcfwP/D3VtWfrmed0zSiz18E3q6qv1zP2lZLknOAc6pqX5Kzgb3A7wHX0/taj+r3J1nB9e4wQp/k0QTaoKrqSRbunFps8aMm7mfhP4A2RvS5tao6VlX7htdvAQeBbfS/1qP6vSIdAn0b8NKi7SOcwl/IBlLAvyTZOzxSYTN5X1UdG17/J/C+9SxmDd2Y5NlhSqbV1MNiw9NaLwSeZhNd6yX9hhVc7w6Bvll9tKouAi4Hbhh+TN90hgVsG3vecDJ3A78M/CZwDPir9S1ndSQ5C3gYuLmq3lx8rPO1XqbfK7reHQJ9kkcTtFNVR4fvrwD/xMLU02bx8jD3+M4c5CvrXM+qq6qXq+p/quonwN/S8Hon2cJCqD1YVY8Mu9tf6+X6vdLr3SHQJ3k0QStJ3j18gEKSdwO/Czx/4j/VyuJHTXwa+Od1rGVNvBNqg9+n2fUeHrd9L3Cwqu5cdKj1tR7V75Ve7w1/lwvAcEvPX/N/jyb4s3UuaVUl+SUWRuWw8PiGf+ja5yQPAZey8EjRl4HbgUeBfwTOA34AfLKq2nyIOKLPl7Lw43cBh4E/XjS3vOEl+SjwFPAc8JNh960szCd3vtaj+n0tK7jeLQJdktRjykWShIEuSW0Y6JLUhIEuSU0Y6JLUhIEuSU0Y6JLUxP8Cabq8hsNCgo8AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            " The number of important features with threshold as 1 :6 \n",
            "Feature 0: 2.647197\n",
            "Feature 1: 0.606646\n",
            "Feature 2: 0.987726\n",
            "Feature 3: 3.816667\n",
            "Feature 4: 2.366213\n",
            "Feature 5: 1.146154\n",
            "Feature 6: 1.424897\n",
            "Feature 7: 4.506504\n",
            "Feature 8: 1.653086\n",
            "Feature 9: 1.386111\n",
            "Feature 10: 3.688753\n",
            "Feature 11: 2.377637\n",
            "Feature 12: 1.653040\n",
            "Feature 13: 1.869637\n",
            "Feature 14: 0.573077\n",
            "Feature 15: 2.759781\n",
            "Feature 16: 1.664387\n",
            "Feature 17: 2.188087\n",
            "Feature 18: 3.249751\n",
            "Feature 19: 4.095988\n",
            "Feature 20: 1.730137\n",
            "Feature 21: 0.844985\n",
            "Feature 22: 0.326944\n",
            "Feature 23: 0.498964\n",
            "Feature 24: 7.232533\n"
          ],
          "name": "stdout"
        },
        {
          "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"
          }
        },
        {
          "output_type": "stream",
          "text": [
            " The number of important features with threshold as 1 :19 \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 702
        },
        "id": "wqvizUr3R_-d",
        "outputId": "8bc5bb23-d53c-46ad-aa27-1acee0c2e2cb"
      },
      "source": [
        "#X_train4, X_test4, fs = select_features(X_train2, y_train2, X_test2)\n",
        "for i in range(len(fs5.scores_)):\n",
        "\tprint('Feature %d: %f' % (i, fs5.scores_[i]))\n",
        "# plot the scores\n",
        "pyplot.bar([i for i in range(len(fs5.scores_))], fs5.scores_)\n",
        "pyplot.show()\n",
        "\n",
        "count = 0\n",
        "for i in fs5.scores_ : \n",
        "    if i > 1 : \n",
        "        count = count + 1\n",
        "k5=count\n",
        "\n",
        "print(\" The number of important features with threshold as 1 : {}\" .format(k5))"
      ],
      "execution_count": 145,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Feature 0: 2.647197\n",
            "Feature 1: 0.606646\n",
            "Feature 2: 0.987726\n",
            "Feature 3: 3.816667\n",
            "Feature 4: 2.366213\n",
            "Feature 5: 1.146154\n",
            "Feature 6: 1.424897\n",
            "Feature 7: 4.506504\n",
            "Feature 8: 1.653086\n",
            "Feature 9: 1.386111\n",
            "Feature 10: 3.688753\n",
            "Feature 11: 2.377637\n",
            "Feature 12: 1.653040\n",
            "Feature 13: 1.869637\n",
            "Feature 14: 0.573077\n",
            "Feature 15: 2.759781\n",
            "Feature 16: 1.664387\n",
            "Feature 17: 2.188087\n",
            "Feature 18: 3.249751\n",
            "Feature 19: 4.095988\n",
            "Feature 20: 1.730137\n",
            "Feature 21: 0.844985\n",
            "Feature 22: 0.326944\n",
            "Feature 23: 0.498964\n",
            "Feature 24: 7.232533\n"
          ],
          "name": "stdout"
        },
        {
          "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"
          }
        },
        {
          "output_type": "stream",
          "text": [
            " The number of important features with threshold as 1 : 19\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yQK5DajNzrYD"
      },
      "source": [
        "A bar chart of the feature importance scores for each input feature is created.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "p-Pll3-10ASq"
      },
      "source": [
        "So, from above plot we remove the features with lower importance scores (<1.5).\n",
        "\n",
        "Hence features 2,5,8,9,14,16,21,22,23 are not considered,\n",
        "\n",
        "We could set k=16 i.e (25-9)  When configuring the SelectKBest to select these top 16 features"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IcSMQuHA3rzW",
        "outputId": "738d5820-976e-43cb-ed68-288e32affd8b"
      },
      "source": [
        "def select_features(X_train, y_train, X_test):\n",
        "\tfs = SelectKBest(score_func=chi2,k=k1)\n",
        "\tfs.fit(X_train, y_train)\n",
        "\tX_train_fs = fs.transform(X_train)\n",
        "\tX_test_fs = fs.transform(X_test)\n",
        "\treturn X_train_fs, X_test_fs, fs\n",
        "\n",
        "def select_features2(X_train, y_train, X_test):\n",
        "\tfs = SelectKBest(score_func=chi2,k=k3)\n",
        "\tfs.fit(X_train, y_train)\n",
        "\tX_train_fs = fs.transform(X_train)\n",
        "\tX_test_fs = fs.transform(X_test)\n",
        "\treturn X_train_fs, X_test_fs, fs\n",
        "\n",
        "def select_features3(X_train, y_train, X_test):\n",
        "\tfs = SelectKBest(score_func=chi2,k=k5)\n",
        "\tfs.fit(X_train, y_train)\n",
        "\tX_train_fs = fs.transform(X_train)\n",
        "\tX_test_fs = fs.transform(X_test)\n",
        "\treturn X_train_fs, X_test_fs, fs\n",
        "\n",
        "\n",
        "X_train1, X_test1, fs = select_features(X_train, y_train, X_test)\n",
        "X_train3, X_test3, fs = select_features2(X_train2, y_train2, X_test2)\n",
        "X_train5, X_test5, fs = select_features3(X_train6, y_train6, X_test6)\n",
        "\n",
        " \n",
        "#---------------SVM------------------#\n",
        "#\n",
        "print(\"Without feature Selection : \")\n",
        "acc = svm(X_train6,y_train6,X_test6,y_test6)\n",
        "\n",
        "print(\"With feature Selection : \")\n",
        "acc1 = svm(X_train5,y_train6,X_test5,y_test6)\n",
        "\n",
        "#-------------Decision Tree-----------------#\n",
        "print(\"Without feature Selection : \")\n",
        "acc1,clf1 = Dec_tree(X_train1,y_train,X_test1,y_test)\n",
        "\n",
        "\n",
        "print(\"With feature Selection : \")\n",
        "acc,clf = Dec_tree(X_train,y_train,X_test,y_test)\n",
        "\n",
        "print(' ')\n",
        "\n",
        "#-------------XGBoost-----------------#\n",
        "print(\"Boosting the Decision Tree  \")\n",
        "acc = xgboost(X_train2,y_train2,X_test2,y_test2,clf)\n"
      ],
      "execution_count": 149,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Without feature Selection : \n",
            "Model accuracy score with default hyperparameters: 60.0000\n",
            "With feature Selection : \n",
            "Model accuracy score with default hyperparameters: 70.0000\n",
            "Without feature Selection : \n",
            "Model accuracy score with Decision Tree 40.0\n",
            "With feature Selection : \n",
            "Model accuracy score with Decision Tree 40.0\n",
            " \n",
            "Boosting the Decision Tree  \n",
            "accuracy= 93.33333333333333\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/preprocessing/_label.py:235: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/preprocessing/_label.py:268: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w_GJF49U0ABF"
      },
      "source": [
        "Mutual Info Classification"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "xtsFDLnYiWNP",
        "outputId": "a58dfc84-955e-4e5f-fcf9-b25a35747fe1"
      },
      "source": [
        "# example of mutual information feature selection for categorical data\n",
        "from sklearn.feature_selection import mutual_info_classif\n",
        "\n",
        "def select_features(X_train, y_train, X_test):\n",
        "\tfs = SelectKBest(score_func=mutual_info_classif, k='all')\n",
        "\tfs.fit(X_train, y_train)\n",
        "\tX_train_fs = fs.transform(X_train)\n",
        "\tX_test_fs = fs.transform(X_test)\n",
        "\treturn X_train_fs, X_test_fs, fs\n",
        "\n",
        "# split into train and test sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=20)#Decision tree\n",
        "X_train2,X_test2,y_train2,y_test2=train_test_split(X,y,test_size=0.3,random_state=10)#XGBoost\n",
        "X_train6,X_test6,y_train6,y_test6=train_test_split(X,y,test_size=0.2,random_state=15)#SVM\n",
        "\n",
        "X_train1, X_test1, fs1 = select_features(X_train, y_train, X_test)\n",
        "X_train3, X_test3, fs3 = select_features(X_train2, y_train2, X_test2)\n",
        "X_train5, X_test5, fs5 = select_features(X_train2, y_train2, X_test2)\n",
        "\n",
        "# what are scores for the features\n",
        "for i in range(len(fs1.scores_)):\n",
        "\tprint('Feature %d: %f' % (i, fs.scores_[i]))\n",
        "# plot the scores\n",
        "pyplot.bar([i for i in range(len(fs.scores_))], fs.scores_)\n",
        "pyplot.show()\n",
        "\n",
        "count = 0\n",
        "for i in fs.scores_ : \n",
        "    if i > 0.2 : \n",
        "        count = count + 1\n",
        "k1=count\n",
        "print(\" The number of important features with threshold as 0.2 :{} \".format(k1))\n",
        "\n",
        "\n",
        "for i in range(len(fs3.scores_)):\n",
        "\tprint('Feature %d: %f' % (i, fs3.scores_[i]))\n",
        "# plot the scores\n",
        "pyplot.bar([i for i in range(len(fs3.scores_))], fs3.scores_)\n",
        "pyplot.show()\n",
        "\n",
        "count = 0\n",
        "for i in fs3.scores_ : \n",
        "    if i > 0.2 : \n",
        "        count = count + 1\n",
        "k3=count\n",
        "\n",
        "print(\" The number of important features with threshold as 0.2 :{} \".format(k3))\n",
        "\n",
        "#X_train4, X_test4, fs = select_features(X_train2, y_train2, X_test2)\n",
        "for i in range(len(fs5.scores_)):\n",
        "\tprint('Feature %d: %f' % (i, fs5.scores_[i]))\n",
        "# plot the scores\n",
        "pyplot.bar([i for i in range(len(fs5.scores_))], fs5.scores_)\n",
        "pyplot.show()\n",
        "\n",
        "count = 0\n",
        "for i in fs5.scores_ : \n",
        "    if i > 0.2 : \n",
        "        count = count + 1\n",
        "k5=count\n",
        "\n",
        "print(\" The number of important features with threshold as 0.2 : {}\" .format(k5))"
      ],
      "execution_count": 147,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Feature 0: 1.939425\n",
            "Feature 1: 0.621032\n",
            "Feature 2: 1.190205\n",
            "Feature 3: 2.361067\n",
            "Feature 4: 2.723434\n",
            "Feature 5: 1.019886\n",
            "Feature 6: 1.630621\n",
            "Feature 7: 5.962755\n",
            "Feature 8: 0.948673\n",
            "Feature 9: 1.391667\n",
            "Feature 10: 3.995109\n",
            "Feature 11: 3.001412\n",
            "Feature 12: 1.301361\n",
            "Feature 13: 1.730517\n",
            "Feature 14: 0.916034\n",
            "Feature 15: 2.627847\n",
            "Feature 16: 0.703328\n",
            "Feature 17: 2.405423\n",
            "Feature 18: 3.379566\n",
            "Feature 19: 3.567291\n",
            "Feature 20: 2.512279\n",
            "Feature 21: 1.074299\n",
            "Feature 22: 0.271954\n",
            "Feature 23: 0.452889\n",
            "Feature 24: 8.890845\n"
          ],
          "name": "stdout"
        },
        {
          "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"
          }
        },
        {
          "output_type": "stream",
          "text": [
            " The number of important features with threshold as 0.2 :25 \n",
            "Feature 0: 0.206512\n",
            "Feature 1: 0.000000\n",
            "Feature 2: 0.030565\n",
            "Feature 3: 0.000000\n",
            "Feature 4: 1.123318\n",
            "Feature 5: 1.615520\n",
            "Feature 6: 0.763666\n",
            "Feature 7: 0.226821\n",
            "Feature 8: 0.021102\n",
            "Feature 9: 0.000000\n",
            "Feature 10: 0.605446\n",
            "Feature 11: 0.231260\n",
            "Feature 12: 1.023677\n",
            "Feature 13: 0.055574\n",
            "Feature 14: 2.088958\n",
            "Feature 15: 0.304069\n",
            "Feature 16: 1.360764\n",
            "Feature 17: 0.000000\n",
            "Feature 18: 0.297629\n",
            "Feature 19: 0.143156\n",
            "Feature 20: 0.305350\n",
            "Feature 21: 0.058968\n",
            "Feature 22: 0.230731\n",
            "Feature 23: 0.455775\n",
            "Feature 24: 0.175940\n"
          ],
          "name": "stdout"
        },
        {
          "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"
          }
        },
        {
          "output_type": "stream",
          "text": [
            " The number of important features with threshold as 0.2 :15 \n",
            "Feature 0: 0.143944\n",
            "Feature 1: 0.000000\n",
            "Feature 2: 0.102765\n",
            "Feature 3: 0.000000\n",
            "Feature 4: 0.600839\n",
            "Feature 5: 0.759564\n",
            "Feature 6: 0.593741\n",
            "Feature 7: 0.182509\n",
            "Feature 8: 0.276503\n",
            "Feature 9: 0.000000\n",
            "Feature 10: 0.173822\n",
            "Feature 11: 0.328662\n",
            "Feature 12: 0.483945\n",
            "Feature 13: 0.066697\n",
            "Feature 14: 1.777437\n",
            "Feature 15: 0.209989\n",
            "Feature 16: 1.585256\n",
            "Feature 17: 0.333728\n",
            "Feature 18: 0.291603\n",
            "Feature 19: 0.127384\n",
            "Feature 20: 0.202855\n",
            "Feature 21: 0.144670\n",
            "Feature 22: 0.202034\n",
            "Feature 23: 0.443774\n",
            "Feature 24: 0.208607\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAQ40lEQVR4nO3dfYwdV33G8e9TJwE1IGpqN41iExtqtYSXJGhlWhGBUYlxoI1BRdQupaElcoXivkpVk1aKo6BKKbSFtgSCC1agLQk0EHCFQxIJaFDTUK9pmjcaMMY0a6V4iSkvBRE5+fWPHcPNZnfv9e7dXfvs9yOt9s45Z+b+JqM8d/bM3HGqCklSu35ssQuQJM0vg16SGmfQS1LjDHpJapxBL0mNO2WxC5jKihUras2aNYtdhiSdNPbt2/eNqlo5Vd8JGfRr1qxhdHR0scuQpJNGkq9N1+fUjSQ1zqCXpMYZ9JLUOINekhpn0EtS4wx6SWqcQS9JjTPoJalxBr0kNa7vN2OT7AJ+CThcVc+fov+PgDf0bO+5wMqqOpLkIPAd4DHgaFWNDKtwqQVrLv9k3zEHr3n1AlSilg1yRn89sGm6zqp6e1WdV1XnAVcA/1JVR3qGvLzrN+QlaRH0DfqqugM40m9cZytww5wqkiQN1dDm6JP8OBNn/h/taS7gtiT7kmzrs/62JKNJRsfHx4dVliQtecO8GPvLwL9Omra5oKpeBFwEXJbkpdOtXFU7q2qkqkZWrpzySZuSpFkYZtBvYdK0TVUd6n4fBm4G1g/x/SRJAxhK0Cd5BvAy4BM9bacnefqx18BG4L5hvJ8kaXCD3F55A7ABWJFkDNgBnApQVdd1w14L3FZV/9ez6hnAzUmOvc+HqupTwytdkjSIvkFfVVsHGHM9E7dh9rYdAM6dbWGSpOHwm7GS1DiDXpIaZ9BLUuMMeklqnEEvSY0z6CWpcQa9JDXOoJekxhn0ktQ4g16SGmfQS1LjDHpJapxBL0mN6/v0SkknrzWXf3KgcQevefU8V6LF5Bm9JDXOoJekxhn0ktQ4g16SGmfQS1LjDHpJalzfoE+yK8nhJPdN078hybeS3N39XNnTtynJg0n2J7l8mIVLkgYzyBn99cCmPmM+V1XndT9XAyRZBlwLXAScA2xNcs5cipUkHb++QV9VdwBHZrHt9cD+qjpQVY8CNwKbZ7EdSdIcDGuO/heS/GeSW5I8r2s7C3ioZ8xY1zalJNuSjCYZHR8fH1JZkqRhBP0XgLOr6lzgb4GPz2YjVbWzqkaqamTlypVDKEuSBEMI+qr6dlV9t3u9Bzg1yQrgELC6Z+iqrk2StIDmHPRJfjpJutfru20+AuwF1iVZm+Q0YAuwe67vJ0k6Pn2fXpnkBmADsCLJGLADOBWgqq4DXge8JclR4PvAlqoq4GiS7cCtwDJgV1XdPy97IUmaVt+gr6qtffrfBbxrmr49wJ7ZlSZJGga/GStJjTPoJalxBr0kNc6gl6TGGfSS1DiDXpIaZ9BLUuMMeklqnEEvSY0z6CWpcQa9JDXOoJekxhn0ktQ4g16SGmfQS1LjDHpJapxBL0mNM+glqXEGvSQ1zqCXpMb1Dfoku5IcTnLfNP1vSHJPknuT3Jnk3J6+g1373UlGh1m4JGkwg5zRXw9smqH/q8DLquoFwFuBnZP6X15V51XVyOxKlCTNxSn9BlTVHUnWzNB/Z8/iXcCquZclSRqWYc/Rvxm4pWe5gNuS7EuybaYVk2xLMppkdHx8fMhlSdLS1feMflBJXs5E0F/Q03xBVR1K8lPA7Un+q6rumGr9qtpJN+0zMjJSw6pLkpa6oZzRJ3kh8D5gc1U9cqy9qg51vw8DNwPrh/F+kqTBzTnokzwL+Bjwxqr6Uk/76Umefuw1sBGY8s4dSdL86Tt1k+QGYAOwIskYsAM4FaCqrgOuBH4SeHcSgKPdHTZnADd3bacAH6qqT83DPkiSZjDIXTdb+/RfClw6RfsB4NwnryFJWkh+M1aSGmfQS1LjDHpJapxBL0mNM+glqXEGvSQ1zqCXpMYZ9JLUOINekhpn0EtS4wx6SWqcQS9JjTPoJalxBr0kNc6gl6TGGfSS1DiDXpIaZ9BLUuMMeklqnEEvSY0bKOiT7EpyOMl90/Qnyd8k2Z/kniQv6um7JMmXu59LhlW4JGkwg57RXw9smqH/ImBd97MNeA9AkmcCO4AXA+uBHUmWz7ZYSdLxGyjoq+oO4MgMQzYDH6wJdwE/keRM4JXA7VV1pKq+CdzOzB8YkqQhG9Yc/VnAQz3LY13bdO1PkmRbktEko+Pj40MqS5J0wlyMraqdVTVSVSMrV65c7HIkqRnDCvpDwOqe5VVd23TtkqQFMqyg3w38Rnf3zc8D36qqh4FbgY1JlncXYTd2bZKkBXLKIIOS3ABsAFYkGWPiTppTAarqOmAP8CpgP/A94De7viNJ3grs7TZ1dVXNdFFXkjRkAwV9VW3t01/AZdP07QJ2HX9pkqRhOGEuxkqS5odBL0mNM+glqXEGvSQ1zqCXpMYZ9JLUOINekhpn0EtS4wx6SWqcQS9JjTPoJalxBr0kNc6gl6TGGfSS1DiDXpIaZ9BLUuMMeklqnEEvSY0z6CWpcQa9JDVuoKBPsinJg0n2J7l8iv53JLm7+/lSkv/t6Xusp2/3MIuXJPV3Sr8BSZYB1wIXAmPA3iS7q+qBY2Oq6g96xv8OcH7PJr5fVecNr2RJ0vHoG/TAemB/VR0ASHIjsBl4YJrxW4EdwylPg1hz+ScHGnfwmlfPcyWSTkSDTN2cBTzUszzWtT1JkrOBtcCne5qfmmQ0yV1JXjPdmyTZ1o0bHR8fH6AsSdIghn0xdgtwU1U91tN2dlWNAL8GvDPJc6Zasap2VtVIVY2sXLlyyGVJ0tI1SNAfAlb3LK/q2qayBbiht6GqDnW/DwCf5Ynz95KkeTZI0O8F1iVZm+Q0JsL8SXfPJPk5YDnwbz1ty5M8pXu9AngJ08/tS5LmQd+LsVV1NMl24FZgGbCrqu5PcjUwWlXHQn8LcGNVVc/qzwXem+RxJj5Urum9W0eSNP8GueuGqtoD7JnUduWk5aumWO9O4AVzqE+SNEd+M1aSGmfQS1LjDHpJapxBL0mNM+glqXEGvSQ1zqCXpMYZ9JLUOINekhpn0EtS4wx6SWqcQS9JjRvooWZaWP7TgJKGyaBfgvwgkZYWp24kqXEGvSQ1zqCXpMYZ9JLUOINekhpn0EtS4wYK+iSbkjyYZH+Sy6fof1OS8SR3dz+X9vRdkuTL3c8lwyxektRf3/vokywDrgUuBMaAvUl2V9UDk4Z+uKq2T1r3mcAOYAQoYF+37jeHUr0kqa9BzujXA/ur6kBVPQrcCGwecPuvBG6vqiNduN8ObJpdqZKk2Rgk6M8CHupZHuvaJvuVJPckuSnJ6uNcV5I0T4Z1MfafgTVV9UImzto/cLwbSLItyWiS0fHx8SGVJUka5Fk3h4DVPcururYfqqpHehbfB7ytZ90Nk9b97FRvUlU7gZ0AIyMjNUBd0rzymUBqxSBn9HuBdUnWJjkN2ALs7h2Q5MyexYuBL3avbwU2JlmeZDmwsWuTJC2Qvmf0VXU0yXYmAnoZsKuq7k9yNTBaVbuB301yMXAUOAK8qVv3SJK3MvFhAXB1VR2Zh/2QJE1joMcUV9UeYM+ktit7Xl8BXDHNuruAXXOoUZI0B34zVpIa5z88IknH6WS7UO8ZvSQ1zqCXpMYZ9JLUOINekhpn0EtS4wx6SWqcQS9JjTPoJalxBr0kNc6gl6TGGfSS1DiDXpIa50PNJP3QyfawLg3GoNeSYYhpqXLqRpIaZ9BLUuMMeklqnHP0kubEax8nPs/oJalxA53RJ9kE/DWwDHhfVV0zqf8PgUuBo8A48FtV9bWu7zHg3m7of1fVxUOqXScwz/KkE0ffoE+yDLgWuBAYA/Ym2V1VD/QM+w9gpKq+l+QtwNuAX+36vl9V5w25bknSgAaZulkP7K+qA1X1KHAjsLl3QFV9pqq+1y3eBawabpmSpNkaJOjPAh7qWR7r2qbzZuCWnuWnJhlNcleS10y3UpJt3bjR8fHxAcqSJA1iqHfdJPl1YAR4WU/z2VV1KMmzgU8nubeqvjJ53araCewEGBkZqWHWJUlL2SBn9IeA1T3Lq7q2J0jyCuBPgYur6gfH2qvqUPf7APBZ4Pw51CtJOk6DBP1eYF2StUlOA7YAu3sHJDkfeC8TIX+4p315kqd0r1cALwF6L+JKkuZZ36mbqjqaZDtwKxO3V+6qqvuTXA2MVtVu4O3A04B/SgI/uo3yucB7kzzOxIfKNZPu1pF0HLxt9eS02MdtoDn6qtoD7JnUdmXP61dMs96dwAvmUqCkpW0hQnKxg3i++c1YSWqcQS9JjfOhZhpI63/aSi0z6HXSGuTDxw8eyakbSWqeQS9JjWtu6sa5ZElO6z2RZ/SS1Ljmzuglndj8q3vheUYvSY0z6CWpcU7dHCf/7JR0svGMXpIaZ9BLUuMMeklqnHP00pB4/UYnKs/oJalxBr0kNc6gl6TGGfSS1LiBgj7JpiQPJtmf5PIp+p+S5MNd/+eTrOnpu6JrfzDJK4dXuiRpEH2DPsky4FrgIuAcYGuScyYNezPwzar6GeAdwJ93654DbAGeB2wC3t1tT5K0QAY5o18P7K+qA1X1KHAjsHnSmM3AB7rXNwG/mCRd+41V9YOq+iqwv9ueJGmBpKpmHpC8DthUVZd2y28EXlxV23vG3NeNGeuWvwK8GLgKuKuq/qFrfz9wS1XdNMX7bAO2dYs/Czw4t117ghXAN4a4vZOB+7x0LMX9Xor7DDPv99lVtXKqjhPmC1NVtRPYOR/bTjJaVSPzse0Tlfu8dCzF/V6K+wyz3+9Bpm4OAat7lld1bVOOSXIK8AzgkQHXlSTNo0GCfi+wLsnaJKcxcXF196Qxu4FLutevAz5dE3NCu4Et3V05a4F1wL8Pp3RJ0iD6Tt1U1dEk24FbgWXArqq6P8nVwGhV7QbeD/x9kv3AESY+DOjGfQR4ADgKXFZVj83TvsxkXqaETnDu89KxFPd7Ke4zzHK/+16MlSSd3PxmrCQ1zqCXpMY1HfT9Ht3QqiQHk9yb5O4ko4tdz3xIsivJ4e47HMfanpnk9iRf7n4vX8wa58M0+31VkkPd8b47yasWs8ZhS7I6yWeSPJDk/iS/17U3e7xn2OdZHetm5+i7Ry18CbgQGGPi7qGtVfXAoha2AJIcBEaqqtkvlCR5KfBd4INV9fyu7W3Akaq6pvtgX15Vf7yYdQ7bNPt9FfDdqvqLxaxtviQ5Ezizqr6Q5OnAPuA1wJto9HjPsM+vZxbHuuUz+kEe3aCTVFXdwcQdXr16H8XxASb+x2jKNPvdtKp6uKq+0L3+DvBF4CwaPt4z7POstBz0ZwEP9SyPMYf/UCeZAm5Lsq97tMRScUZVPdy9/h/gjMUsZoFtT3JPN7XTzBTGZN2Tcc8HPs8SOd6T9hlmcaxbDvql7IKqehETTxy9rPtzf0npvrDX5rzkk70HeA5wHvAw8JeLW878SPI04KPA71fVt3v7Wj3eU+zzrI51y0G/ZB+/UFWHut+HgZtZOk8M/Xo3t3lsjvPwItezIKrq61X1WFU9DvwdDR7vJKcyEXj/WFUf65qbPt5T7fNsj3XLQT/Ioxuak+T07uINSU4HNgL3zbxWM3ofxXEJ8IlFrGXBHAu7zmtp7Hh3jzx/P/DFqvqrnq5mj/d0+zzbY93sXTcA3a1H7+RHj274s0Uuad4leTYTZ/Ew8YiLD7W430luADYw8djWrwM7gI8DHwGeBXwNeH1VNXXhcpr93sDEn/IFHAR+u2fu+qSX5ALgc8C9wONd858wMWfd5PGeYZ+3Motj3XTQS5LanrqRJGHQS1LzDHpJapxBL0mNM+glqXEGvSQ1zqCXpMb9P9HLVd2rhsxZAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            " The number of important features with threshold as 0.2 : 15\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "caJ31ANmiROR",
        "outputId": "7a583c9d-ffaf-4707-f69a-8488cd84979a"
      },
      "source": [
        "def select_features(X_train, y_train, X_test):\n",
        "\tfs = SelectKBest(score_func=mutual_info_classif,k=k1)\n",
        "\tfs.fit(X_train, y_train)\n",
        "\tX_train_fs = fs.transform(X_train)\n",
        "\tX_test_fs = fs.transform(X_test)\n",
        "\treturn X_train_fs, X_test_fs, fs\n",
        "\n",
        "def select_features2(X_train, y_train, X_test):\n",
        "\tfs = SelectKBest(score_func=mutual_info_classif,k=k3)\n",
        "\tfs.fit(X_train, y_train)\n",
        "\tX_train_fs = fs.transform(X_train)\n",
        "\tX_test_fs = fs.transform(X_test)\n",
        "\treturn X_train_fs, X_test_fs, fs\n",
        "\n",
        "def select_features3(X_train, y_train, X_test):\n",
        "\tfs = SelectKBest(score_func=mutual_info_classif,k=k5)\n",
        "\tfs.fit(X_train, y_train)\n",
        "\tX_train_fs = fs.transform(X_train)\n",
        "\tX_test_fs = fs.transform(X_test)\n",
        "\treturn X_train_fs, X_test_fs, fs\n",
        "\n",
        "\n",
        "X_train1, X_test1, fs = select_features(X_train, y_train, X_test)\n",
        "X_train3, X_test3, fs = select_features2(X_train2, y_train2, X_test2)\n",
        "X_train5, X_test5, fs = select_features3(X_train6, y_train6, X_test6)\n",
        "\n",
        " \n",
        "#---------------SVM------------------#\n",
        "#\n",
        "print(\"Without feature Selection : \")\n",
        "acc = svm(X_train6,y_train6,X_test6,y_test6)\n",
        "\n",
        "print(\"With feature Selection : \")\n",
        "acc1 = svm(X_train5,y_train6,X_test5,y_test6)\n",
        "\n",
        "#-------------Decision Tree-----------------#\n",
        "print(\"Without feature Selection : \")\n",
        "acc1,clf1 = Dec_tree(X_train1,y_train,X_test1,y_test)\n",
        "\n",
        "\n",
        "print(\"With feature Selection : \")\n",
        "acc,clf = Dec_tree(X_train,y_train,X_test,y_test)\n",
        "\n",
        "print(' ')\n",
        "\n",
        "#-------------XGBoost-----------------#\n",
        "print(\"Boosting the Decision Tree  \")\n",
        "acc = xgboost(X_train2,y_train2,X_test2,y_test2,clf)\n"
      ],
      "execution_count": 150,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Without feature Selection : \n",
            "Model accuracy score with default hyperparameters: 60.0000\n",
            "With feature Selection : \n",
            "Model accuracy score with default hyperparameters: 70.0000\n",
            "Without feature Selection : \n",
            "Model accuracy score with Decision Tree 40.0\n",
            "With feature Selection : \n",
            "Model accuracy score with Decision Tree 50.0\n",
            " \n",
            "Boosting the Decision Tree  \n",
            "accuracy= 93.33333333333333\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/utils/validation.py:760: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/preprocessing/_label.py:235: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n",
            "/usr/local/lib/python3.6/dist-packages/sklearn/preprocessing/_label.py:268: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
            "  y = column_or_1d(y, warn=True)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "udZl3BhOubrS"
      },
      "source": [
        "Accuracies vs Techniques"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 334
        },
        "id": "WKBN8OCcuUKB",
        "outputId": "06bd190c-9b55-4726-b1e6-1deab9ea9bc0"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "#fig = plt.figure()\n",
        "fig = plt.figure(figsize =(10, 5)) \n",
        "#ax = fig.add_axes([0,0,1,1])\n",
        "Name = ['Decision Tree(with feature selection)','Decision Tree(without feature selection)', 'XGBoost', 'SVM(with feature selction)','SVM(without feature selction)']\n",
        "accuracies = [50,40,93.333,70,60]\n",
        "plt.bar(Name,accuracies)\n",
        "plt.ylabel('Accuracies')\n",
        "plt.xlabel('Techniques')\n",
        "plt.show()"
      ],
      "execution_count": 172,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zyc7XYm7uuaQ"
      },
      "source": [
        "Prediction for new input"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O7c6prGGwigQ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "20ecd102-f806-43f8-d59e-8921ffffc8ae"
      },
      "source": [
        "# x_new = ['3','5','4','3','5','4','3','5','2','2','5','2','5','4','2','5','5','3','2','5','5','4','2','3','4']\n",
        "x_new = []\n",
        "feat = list(X.columns) \n",
        "for i in feat:\n",
        "  print(i)\n",
        "  val = input(\"Enter: \") \n",
        "  x_new.append(val)\n",
        "x_new"
      ],
      "execution_count": 170,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "sslc\n",
            "Enter: 3\n",
            "hsc\n",
            "Enter: 5\n",
            "cgpa\n",
            "Enter: 4\n",
            "school_type\n",
            "Enter: 3\n",
            "no_of_miniprojects\n",
            "Enter: 5\n",
            "no_of_projects\n",
            "Enter: 4\n",
            "coresub_skill\n",
            "Enter: 3\n",
            "aptitude_skill\n",
            "Enter: 5\n",
            "problemsolving_skill\n",
            "Enter: 2\n",
            "programming_skill\n",
            "Enter: 2\n",
            "abstractthink_skill\n",
            "Enter: 5\n",
            "design_skill\n",
            "Enter: 2\n",
            "first_computer\n",
            "Enter: 5\n",
            "first_program\n",
            "Enter: 4\n",
            "lab_programs\n",
            "Enter: 2\n",
            "ds_coding\n",
            "Enter: 5\n",
            "technology_used\n",
            "Enter: 5\n",
            "sympos_attend\n",
            "Enter: 3\n",
            "sympos_won\n",
            "Enter: 2\n",
            "extracurricular\n",
            "Enter: 5\n",
            "learning_style\n",
            "Enter: 5\n",
            "college_bench\n",
            "Enter: 4\n",
            "clg_teachers_know\n",
            "Enter: 2\n",
            "college_performence\n",
            "Enter: 3\n",
            "college_skills\n",
            "Enter: 4\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['3',\n",
              " '5',\n",
              " '4',\n",
              " '3',\n",
              " '5',\n",
              " '4',\n",
              " '3',\n",
              " '5',\n",
              " '2',\n",
              " '2',\n",
              " '5',\n",
              " '2',\n",
              " '5',\n",
              " '4',\n",
              " '2',\n",
              " '5',\n",
              " '5',\n",
              " '3',\n",
              " '2',\n",
              " '5',\n",
              " '5',\n",
              " '4',\n",
              " '2',\n",
              " '3',\n",
              " '4']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 170
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z2kjQBMywiy1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "aef6d03b-287b-4e2d-8c2b-15658ac044c8"
      },
      "source": [
        "#x_new = ['3','5','4','3','5','4','3','5','2','2','5','2','5','4','2','5','5','3','2','5','5','4','2','3','4']\n",
        "new_pred  = clf.predict([x_new])\n",
        "print(\"Prediction : {}\".format(y1[y1['Associated Number']==new_pred[0]]['ROLE']))"
      ],
      "execution_count": 182,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Prediction : 4    Technical Writer\n",
            "Name: ROLE, dtype: object\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jHdvspMzzaKs"
      },
      "source": [
        "Algorithms : SVM, Decision Tree and XGBoost"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gm5sjkrDwiQ_"
      },
      "source": [
        "#Support Vector machine\n",
        "from sklearn.metrics import confusion_matrix,accuracy_score\n",
        "def svm(X_train,y_train,X_test,y_test):\n",
        "  from sklearn.svm import SVC\n",
        "  from pandas import read_csv\n",
        "  from sklearn.model_selection import train_test_split\n",
        "\n",
        "  # import metrics to compute accuracy\n",
        "  from sklearn.metrics import accuracy_score\n",
        "  svc=SVC() \n",
        "  # fit classifier to training set\n",
        "  svc.fit(X_train,y_train)\n",
        "  # make predictions on test set\n",
        "  y_pred=svc.predict(X_test)\n",
        "  # compute and print accuracy score\n",
        "  print('Model accuracy score with default hyperparameters: {0:0.4f}'. format(accuracy_score(y_test, y_pred)*100))\n",
        "  return accuracy_score(y_test, y_pred)*100\n",
        "\n",
        "def Dec_tree(X_train,y_train,X_test,y_test):\n",
        "  from sklearn import tree\n",
        "  clf = tree.DecisionTreeClassifier()\n",
        "  clf = clf.fit(X_train, y_train)\n",
        "  # Prediction\n",
        "  y_pred = clf.predict(X_test)\n",
        "  y_test_arr=y_test['role']\n",
        "  from sklearn.metrics import confusion_matrix,accuracy_score\n",
        "  accuracy = accuracy_score(y_test,y_pred)\n",
        "  print('Model accuracy score with Decision Tree', accuracy_score(y_test, y_pred)*100)\n",
        "  return accuracy*100,clf\n",
        "\n",
        "def xgboost(X_train,y_train,X_test,y_test,clf):\n",
        "  #X_train,X_test,y_train,y_test=train_test_split(X1,y,test_size=0.3,random_state=10)\n",
        "  shape = X_train.shape\n",
        "  X_train=pd.to_numeric(X_train.values.flatten())\n",
        "  X_train=X_train.reshape(shape)\n",
        "  from xgboost.sklearn import XGBClassifier\n",
        "  model = XGBClassifier()\n",
        "  model.fit(X_train, y_train)\n",
        "  xgb_y_pred  = clf.predict(X_test)\n",
        "  xgb_accuracy = accuracy_score(y_test,xgb_y_pred)\n",
        "  print(\"accuracy=\",xgb_accuracy*100)\n",
        "  return xgb_accuracy*100"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}