{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "%matplotlib inline"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "\n# Default DL85Booster\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "from sklearn.model_selection import train_test_split\nfrom sklearn.metrics import accuracy_score\nfrom sklearn.ensemble import AdaBoostClassifier\nfrom sklearn.tree import DecisionTreeClassifier\nfrom dl85 import DL85Booster, MODEL_QP_MDBOOST, MODEL_LP_DEMIRIZ\nimport time\nimport numpy as np\nfrom sklearn.metrics import confusion_matrix\n\ndataset = np.genfromtxt(\"../datasets/tic-tac-toe.txt\", delimiter=' ')\nX, y = dataset[:, 1:], dataset[:, 0]\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n\ndepth = 1\n# params = {'model': MODEL_QP_MDBOOST, 'regulator': 100, 'name': 'MDBoost'}\nparams = {'model': MODEL_LP_DEMIRIZ, 'regulator': .9, 'name': 'LPBoost'}\n\nprint(\"######################################################################\\n\"\n      \"#                     DL8.5 boosting classifier                      #\\n\"\n      \"######################################################################\")\nprint(\"<<=== Optiboost ===>>\")\nclf = DL85Booster(max_depth=depth, model=params['model'], regulator=params['regulator'])\nstart = time.perf_counter()\nprint(\"Model building...\")\nclf.fit(X_train, y_train)\nduration = time.perf_counter() - start\nprint(\"Model built. Duration of building =\", round(duration, 4))\nprint(\"Number of trees =\", clf.n_estimators_)\ny_pred = clf.predict(X_test)\nprint(\"Confusion Matrix below\")\nprint(confusion_matrix(y_test, y_pred))\nprint(\"Accuracy DL85Booster +\", params['name'], \"on training set =\", round(accuracy_score(y_train, clf.predict(X_train)), 4))\nprint(\"Accuracy DL85Booster +\", params['name'], \"on test set =\", round(accuracy_score(y_test, y_pred), 4), \"\\n\")\n\nprint(\"<<=== AdaBoost + CART ===>>\")\nab = AdaBoostClassifier(base_estimator=DecisionTreeClassifier(max_depth=depth), n_estimators=clf.n_estimators_)\nstart = time.perf_counter()\nprint(\"Model building...\")\nab.fit(X, y)\nduration = time.perf_counter() - start\nprint(\"Model built. Duration of building =\", round(duration, 4))\nprint(\"Number of trees =\", clf.n_estimators_)\ny_pred = ab.predict(X_test)\nprint(\"Confusion Matrix below\")\nprint(confusion_matrix(y_test, y_pred))\nprint(\"Accuracy AdaBoost on training set =\", round(accuracy_score(y_train, ab.predict(X_train)), 4))\nprint(\"Accuracy AdaBoost on test set =\", round(accuracy_score(y_test, y_pred), 4))\nprint(\"\\n\\n\")"
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.6"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}