decision_tree.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyOg8Pq9F2m3X4UTebzDzMv5",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "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/ericyoc/ml_portfolio/blob/main/decision_tree.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "TbAWWeiUiCkw"
      },
      "outputs": [],
      "source": [
        "#decision trees\n",
        "\n",
        "import sys\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "import sklearn.tree as tree"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Surpress warnings:\n",
        "def warn(*args, **kwargs):\n",
        "    pass\n",
        "import warnings\n",
        "warnings.warn = warn"
      ],
      "metadata": {
        "id": "lh2GqlQViHi4"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def main():\n",
        "\n",
        "    # download dataset\n",
        "    my_data = pd.read_csv('ml_data/drug200.csv', delimiter=\",\")\n",
        "\n",
        "    my_data.head()\n",
        "    print(my_data.head())\n",
        "\n",
        "    my_data.shape\n",
        "    print(my_data.shape)\n",
        "\n",
        "    # pre-processing\n",
        "    # Remove the column containing the target name since it doesn't contain numeric values.\n",
        "    X = my_data[['Age', 'Sex', 'BP', 'Cholesterol', 'Na_to_K']].values\n",
        "    X[0:5]\n",
        "    print(X[0:5])\n",
        "\n",
        "    # some features in this dataset are categorical, such as Sex or BP. Unfortunately,\n",
        "    # Sklearn Decision Trees does not handle categorical variables.\n",
        "    # We can still convert these features to numerical values using the LabelEncoder() method to convert the categorical variable into dummy/indicator variables\n",
        "    from sklearn import preprocessing\n",
        "    le_sex = preprocessing.LabelEncoder()\n",
        "    le_sex.fit(['F','M'])\n",
        "    X[:,1] = le_sex.transform(X[:,1])\n",
        "\n",
        "\n",
        "    le_BP = preprocessing.LabelEncoder()\n",
        "    le_BP.fit([ 'LOW', 'NORMAL', 'HIGH'])\n",
        "    X[:,2] = le_BP.transform(X[:,2])\n",
        "\n",
        "\n",
        "    le_Chol = preprocessing.LabelEncoder()\n",
        "    le_Chol.fit([ 'NORMAL', 'HIGH'])\n",
        "    X[:,3] = le_Chol.transform(X[:,3])\n",
        "\n",
        "    print(X[0:5])\n",
        "\n",
        "    # fill the target variable.\n",
        "    y = my_data[\"Drug\"]\n",
        "    print(y[0:5])\n",
        "\n",
        "    # setting up decision tree\n",
        "    from sklearn.model_selection import train_test_split\n",
        "    X_trainset, X_testset, y_trainset, y_testset = train_test_split(X, y, test_size=0.3, random_state=3)\n",
        "    print('Shape of X training set {}'.format(X_trainset.shape),'&',' Size of Y training set {}'.format(y_trainset.shape))\n",
        "    print('Shape of X training set {}'.format(X_testset.shape),'&',' Size of Y training set {}'.format(y_testset.shape))\n",
        "\n",
        "\n",
        "    # modeling\n",
        "    drugTree = DecisionTreeClassifier(criterion=\"entropy\", max_depth = 4)\n",
        "    print(drugTree) # it shows the default parameters\n",
        "    drugTree.fit(X_trainset,y_trainset)\n",
        "\n",
        "    # prediction\n",
        "    predTree = drugTree.predict(X_testset)\n",
        "    print(predTree [0:5])\n",
        "    print(y_testset [0:5])\n",
        "\n",
        "    # evaluation\n",
        "    from sklearn import metrics\n",
        "    import matplotlib.pyplot as plt\n",
        "    print(\"DecisionTrees's Accuracy: \", metrics.accuracy_score(y_testset, predTree))\n",
        "\n",
        "    # visualization\n",
        "    from sklearn.tree import export_graphviz\n",
        "    from sklearn import tree\n",
        "    import matplotlib.pyplot as plt\n",
        "\n",
        "    export_graphviz(drugTree, out_file=None, filled=True, feature_names=['Age', 'Sex', 'BP', 'Cholesterol', 'Na_to_K'])\n",
        "\n",
        "    graph = tree.plot_tree(drugTree, feature_names=['Age', 'Sex', 'BP', 'Cholesterol', 'Na_to_K'])\n",
        "    plt.show()  # Display the tree\n",
        "\n",
        "    # Optional: Save as PNG image\n",
        "    plt.savefig('tree.png')"
      ],
      "metadata": {
        "id": "dkWY7hVSiJIK"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "if __name__ == \"__main__\":\n",
        "    main()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "sji0hBnViLXI",
        "outputId": "29e344fb-d73a-4936-9f3c-16cee92c3c77"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "   Age Sex      BP Cholesterol  Na_to_K   Drug\n",
            "0   23   F    HIGH        HIGH   25.355  drugY\n",
            "1   47   M     LOW        HIGH   13.093  drugC\n",
            "2   47   M     LOW        HIGH   10.114  drugC\n",
            "3   28   F  NORMAL        HIGH    7.798  drugX\n",
            "4   61   F     LOW        HIGH   18.043  drugY\n",
            "(200, 6)\n",
            "[[23 'F' 'HIGH' 'HIGH' 25.355]\n",
            " [47 'M' 'LOW' 'HIGH' 13.093]\n",
            " [47 'M' 'LOW' 'HIGH' 10.114]\n",
            " [28 'F' 'NORMAL' 'HIGH' 7.798]\n",
            " [61 'F' 'LOW' 'HIGH' 18.043]]\n",
            "[[23 0 0 0 25.355]\n",
            " [47 1 1 0 13.093]\n",
            " [47 1 1 0 10.114]\n",
            " [28 0 2 0 7.798]\n",
            " [61 0 1 0 18.043]]\n",
            "0    drugY\n",
            "1    drugC\n",
            "2    drugC\n",
            "3    drugX\n",
            "4    drugY\n",
            "Name: Drug, dtype: object\n",
            "Shape of X training set (140, 5) &  Size of Y training set (140,)\n",
            "Shape of X training set (60, 5) &  Size of Y training set (60,)\n",
            "DecisionTreeClassifier(criterion='entropy', max_depth=4)\n",
            "['drugY' 'drugX' 'drugX' 'drugX' 'drugX']\n",
            "40     drugY\n",
            "51     drugX\n",
            "139    drugX\n",
            "197    drugX\n",
            "170    drugX\n",
            "Name: Drug, dtype: object\n",
            "DecisionTrees's Accuracy:  0.9833333333333333\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 0 Axes>"
            ]
          },
          "metadata": {}
        }
      ]
    }
  ]
}