ADABOOST.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "local-reserve",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import sklearn\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.model_selection import train_test_split, StratifiedKFold, cross_val_score, learning_curve, ShuffleSplit, validation_curve\n",
    "from sklearn.metrics import plot_confusion_matrix\n",
    "import numpy as np\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.svm import SVC\n",
    "from imblearn.pipeline import Pipeline\n",
    "from category_encoders.one_hot import OneHotEncoder\n",
    "from sklearn.impute import SimpleImputer\n",
    "from imblearn.over_sampling import RandomOverSampler, SMOTE\n",
    "from mlxtend.evaluate import bias_variance_decomp\n",
    "from mlxtend.plotting import plot_decision_regions\n",
    "from sklearn.metrics import classification_report\n",
    "from imblearn.under_sampling import RandomUnderSampler\n",
    "\n",
    "from sklearn.datasets import make_classification\n",
    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "\n",
    "from sklearn import tree\n",
    "from sklearn.metrics import average_precision_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "fatal-christian",
   "metadata": {},
   "outputs": [],
   "source": [
    "parent_dir = '/mnt/c/users/17752/Documents/GeorgiaTech/machine_learning'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "equipped-nebraska",
   "metadata": {},
   "source": [
    "## Minimal Cost complexity pruning\n",
    "https://scikit-learn.org/stable/modules/tree.html#bre"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "temporal-filename",
   "metadata": {},
   "source": [
    "## Liver Data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "former-aspect",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 583 entries, 0 to 582\n",
      "Data columns (total 11 columns):\n",
      " #   Column                      Non-Null Count  Dtype  \n",
      "---  ------                      --------------  -----  \n",
      " 0   Age                         583 non-null    int64  \n",
      " 1   Gender                      583 non-null    object \n",
      " 2   Total_Bilirubin             583 non-null    float64\n",
      " 3   Direct_Bilirubin            583 non-null    float64\n",
      " 4   Alkaline_Phosphotase        583 non-null    int64  \n",
      " 5   Alamine_Aminotransferase    583 non-null    int64  \n",
      " 6   Aspartate_Aminotransferase  583 non-null    int64  \n",
      " 7   Total_Protiens              583 non-null    float64\n",
      " 8   Albumin                     583 non-null    float64\n",
      " 9   Albumin_and_Globulin_Ratio  579 non-null    float64\n",
      " 10  Dataset                     583 non-null    int64  \n",
      "dtypes: float64(5), int64(5), object(1)\n",
      "memory usage: 50.2+ KB\n",
      "0.6864309603440037\n"
     ]
    }
   ],
   "source": [
    "liver_df = pd.read_csv(f\"{parent_dir}/data/liver.csv\")\n",
    "liver_df = pd.read_csv(f\"{parent_dir}/data/liver.csv\")\n",
    "liver_df.info()\n",
    "liver_df.drop_duplicates(inplace=True)\n",
    "impute_Albumin_and_Globulin_Ratio = liver_df[~liver_df.Albumin_and_Globulin_Ratio.isna()].Albumin_and_Globulin_Ratio.mean()\n",
    "filt_ =  liver_df.Albumin_and_Globulin_Ratio.isna()\n",
    "liver_df.loc[filt_, \"Albumin_and_Globulin_Ratio\" ]= impute_Albumin_and_Globulin_Ratio\n",
    "assert sum(liver_df.Albumin_and_Globulin_Ratio.isna()) == 0 \n",
    "# format response \n",
    "liver_df.loc[(liver_df.Dataset ==2), \"Dataset\" ] = -1\n",
    "liver_df.head(10)\n",
    "liver_df.loc[:,\"Gender\"] = pd.get_dummies(liver_df.Gender, prefix='gender').drop(\"gender_Male\", axis =1  ).gender_Female\n",
    "X = liver_df.loc[:, \"Age\":\"Albumin_and_Globulin_Ratio\"]\n",
    "y = liver_df[\"Dataset\"]\n",
    "X_train, X_test, y_train, y_test = train_test_split( X, y, test_size=0.2, random_state=62)\n",
    "\n",
    "\n",
    "#inital shot \n",
    "\n",
    "liver_steps = [('model',AdaBoostClassifier(algorithm=\"SAMME\"))] \n",
    "liver_pipeline = Pipeline(steps=liver_steps)\n",
    "stratified_kfold = StratifiedKFold(n_splits = 5, random_state=65, shuffle=True)\n",
    "scores = cross_val_score(liver_pipeline, X_train, y_train, scoring='accuracy', cv=stratified_kfold, n_jobs=3)\n",
    "\n",
    "print(scores.mean())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "sudden-farming",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.712280701754386"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(liver_df.Dataset==1)/liver_df.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "underlying-render",
   "metadata": {},
   "source": [
    "## Adult Census Data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "compressed-pepper",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2409257153394597"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(census_df.income==1)/census_df.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "greatest-alcohol",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train data shape is (26029, 15)\n",
      "Test data shape is (6508, 15)\n",
      "Gradient Boosting  AUC imputing missing values: 0.906\n"
     ]
    }
   ],
   "source": [
    "census_df = pd.read_csv(f\"{parent_dir}/data/adult.csv\")\n",
    "census_df.drop_duplicates(inplace=True)\n",
    "# # Deleting duplicates \n",
    "census_df.duplicated().sum()\n",
    "#Replacing '?' for nans\n",
    "census_df = census_df.replace('?', np.NaN)\n",
    "# Mapping already our target variable in  1 and 0.\n",
    "census_df.income = census_df.income.map({'<=50K':-1, '>50K':1})\n",
    "\n",
    "train_data, test_data = train_test_split(census_df,\n",
    "                                         test_size=0.2,\n",
    "                                         stratify=census_df['income'],\n",
    "                                         random_state=65)\n",
    "print('Train data shape is {}'.format(train_data.shape))\n",
    "print('Test data shape is {}'.format(test_data.shape))\n",
    "X_c_train = train_data.drop('income', axis=1)\n",
    "y_c_train = train_data['income']\n",
    "\n",
    "#Transforming categorical columns\n",
    "categorical_columns= X_c_train.select_dtypes(object).columns\n",
    "\n",
    "# Using pipeline\n",
    "census_steps = [('encoding', OneHotEncoder(cols=categorical_columns)),\n",
    "         ('imputer',SimpleImputer(missing_values=np.NaN, strategy='most_frequent')),\n",
    "         ('model',AdaBoostClassifier(algorithm=\"SAMME\"))]\n",
    "census_pipeline = Pipeline(steps=census_steps)\n",
    "\n",
    "#inital shot \n",
    "stratified_kfold = StratifiedKFold(n_splits = 5, random_state=65, shuffle=True)\n",
    "scores = cross_val_score(census_pipeline, X_c_train, y_c_train, scoring='roc_auc', cv=stratified_kfold, n_jobs=3)\n",
    "print(f'Gradient Boosting  AUC imputing missing values: {round(np.mean(scores),3)}')\n",
    "X_c_test = test_data.drop('income', axis=1)\n",
    "y_c_test = test_data['income']\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "underlying-burner",
   "metadata": {},
   "source": [
    "## Load & process census Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "opened-jonathan",
   "metadata": {},
   "source": [
    "## Liver Pipeline "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "finite-bowling",
   "metadata": {},
   "source": [
    "# Census Pipleine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "nervous-length",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_learning_curve(estimator=None, title=\"title\", \n",
    "                        X=None,\n",
    "                        y=None,\n",
    "                        axes=None, \n",
    "                        ylim=None, \n",
    "                        cv=None,\n",
    "                        n_jobs=None, train_sizes=np.linspace(.1, 1.0, 5), \n",
    "                        param_name=None, \n",
    "                        param_range_value= None, \n",
    "                        param_range = None,\n",
    "                        validation_title = \"validation_title\"):\n",
    "    \"\"\"\n",
    "    Generate 3 plots: the test and training learning curve, the training\n",
    "    samples vs fit times curve, the fit times vs score curve.\n",
    "\n",
    "    Parameters\n",
    "    ----------\n",
    "    estimator : estimator instance\n",
    "        An estimator instance implementing `fit` and `predict` methods which\n",
    "        will be cloned for each validation.\n",
    "\n",
    "    title : str\n",
    "        Title for the chart.\n",
    "\n",
    "    X : array-like of shape (n_samples, n_features)\n",
    "        Training vector, where ``n_samples`` is the number of samples and\n",
    "        ``n_features`` is the number of features.\n",
    "\n",
    "    y : array-like of shape (n_samples) or (n_samples, n_features)\n",
    "        Target relative to ``X`` for classification or regression;\n",
    "        None for unsupervised learning.\n",
    "\n",
    "    axes : array-like of shape (3,), default=None\n",
    "        Axes to use for plotting the curves.\n",
    "\n",
    "    ylim : tuple of shape (2,), default=None\n",
    "        Defines minimum and maximum y-values plotted, e.g. (ymin, ymax).\n",
    "\n",
    "    cv : int, cross-validation generator or an iterable, default=None\n",
    "        Determines the cross-validation splitting strategy.\n",
    "        Possible inputs for cv are:\n",
    "\n",
    "          - None, to use the default 5-fold cross-validation,\n",
    "          - integer, to specify the number of folds.\n",
    "          - :term:`CV splitter`,\n",
    "          - An iterable yielding (train, test) splits as arrays of indices.\n",
    "\n",
    "        For integer/None inputs, if ``y`` is binary or multiclass,\n",
    "        :class:`StratifiedKFold` used. If the estimator is not a classifier\n",
    "        or if ``y`` is neither binary nor multiclass, :class:`KFold` is used.\n",
    "\n",
    "        Refer :ref:`User Guide <cross_validation>` for the various\n",
    "        cross-validators that can be used here.\n",
    "\n",
    "    n_jobs : int or None, default=None\n",
    "        Number of jobs to run in parallel.\n",
    "        ``None`` means 1 unless in a :obj:`joblib.parallel_backend` context.\n",
    "        ``-1`` means using all processors. See :term:`Glossary <n_jobs>`\n",
    "        for more details.\n",
    "\n",
    "    train_sizes : array-like of shape (n_ticks,)\n",
    "        Relative or absolute numbers of training examples that will be used to\n",
    "        generate the learning curve. If the ``dtype`` is float, it is regarded\n",
    "        as a fraction of the maximum size of the training set (that is\n",
    "        determined by the selected validation method), i.e. it has to be within\n",
    "        (0, 1]. Otherwise it is interpreted as absolute sizes of the training\n",
    "        sets. Note that for classification the number of samples usually have\n",
    "        to be big enough to contain at least one sample from each class.\n",
    "        (default: np.linspace(0.1, 1.0, 5))\n",
    "    \"\"\"\n",
    "    if axes is None:\n",
    "        _, axes = plt.subplots(1, 2, figsize=(20, 5))\n",
    "\n",
    "    axes[0].set_title(title)\n",
    "    if ylim is not None:\n",
    "        axes[0].set_ylim(*ylim)\n",
    "    axes[0].set_xlabel(\"Training examples\")\n",
    "    axes[0].set_ylabel(\"Score\")\n",
    "\n",
    "    train_sizes, train_scores, test_scores, fit_times, _ = \\\n",
    "        learning_curve(estimator, X, y, cv=cv, n_jobs=n_jobs,\n",
    "                       train_sizes=train_sizes,\n",
    "                       return_times=True)\n",
    "    train_scores_mean =  1.0 - np.mean(train_scores, axis=1)\n",
    "    train_scores_std = np.std(train_scores, axis=1)\n",
    "    test_scores_mean =  1.0 - np.mean(test_scores, axis=1)\n",
    "    test_scores_std = np.std(test_scores, axis=1)\n",
    "    fit_times_mean = np.mean(fit_times, axis=1)\n",
    "    fit_times_std = np.std(fit_times, axis=1)\n",
    "    \n",
    "\n",
    "    # Plot learning curve\n",
    "    axes[0].grid()\n",
    "    axes[0].fill_between(train_sizes,  train_scores_mean - train_scores_std,\n",
    "                         train_scores_mean + train_scores_std, alpha=0.1,\n",
    "                         color=\"r\")\n",
    "    axes[0].fill_between(train_sizes,  test_scores_mean - test_scores_std,\n",
    "                         test_scores_mean + test_scores_std, alpha=0.1,\n",
    "                         color=\"g\")\n",
    "    axes[0].plot(train_sizes, train_scores_mean, 'o-', color=\"r\",\n",
    "                 label=\"Training Error\")\n",
    "    axes[0].plot(train_sizes, test_scores_mean, 'o-', color=\"g\",\n",
    "                 label=\"Cross-validation Error\")\n",
    "    axes[0].legend(loc=\"best\")\n",
    "    axes[0].set_ylabel(\"Error\")\n",
    "    \n",
    "    v_train_scores, v_valid_scores = validation_curve(estimator,\n",
    "                                              X_train,\n",
    "                                              y_train, \n",
    "                                              param_name =  param_name,\n",
    "                                            param_range = param_range_value,\n",
    "                                            cv=cv)\n",
    "\n",
    "    param_range = param_range\n",
    "\n",
    "    v_train_scores_mean =  np.mean(v_train_scores, axis=1)\n",
    "    v_train_scores_std = np.std(v_train_scores, axis=1)\n",
    "    v_test_scores_mean = np.mean(v_valid_scores, axis=1)\n",
    "    v_test_scores_std = np.std(v_valid_scores, axis=1)\n",
    "\n",
    "    axes[1].set_title(\"Validation Curve with AdaBoost\")\n",
    "    axes[1].set_xlabel(validation_title)\n",
    "    axes[1].set_ylabel(\"Score\")\n",
    "    lw = 2\n",
    "    axes[1].semilogx(param_range, v_train_scores_mean, label=\"Training score\",\n",
    "                 color=\"darkorange\", lw=lw)\n",
    "    axes[1].fill_between(param_range, v_train_scores_mean - v_train_scores_std,\n",
    "                     v_train_scores_mean + v_train_scores_std, alpha=0.2,\n",
    "                     color=\"darkorange\", lw=lw)\n",
    "    axes[1].semilogx(param_range, v_test_scores_mean, label=\"Cross-validation score\",\n",
    "                 color=\"navy\", lw=lw)\n",
    "    axes[1].fill_between(param_range, v_test_scores_mean - v_test_scores_std,\n",
    "                     v_test_scores_mean + v_test_scores_std, alpha=0.2,\n",
    "                     color=\"navy\", lw=lw)\n",
    "    axes[1].legend(loc=\"best\")\n",
    "\n",
    "\n",
    "\n",
    "    #     # Plot n_samples vs fit_times\n",
    "    #     axes[1].grid()\n",
    "    #     axes[1].plot(train_sizes, fit_times_mean, 'o-')\n",
    "    #     axes[1].fill_between(train_sizes, fit_times_mean - fit_times_std,\n",
    "    #                          fit_times_mean + fit_times_std, alpha=0.1)\n",
    "    #     axes[1].set_xlabel(\"Training examples\")\n",
    "    #     axes[1].set_ylabel(\"fit_times\")\n",
    "    #     axes[1].set_title(\"Scalability of the model\")\n",
    "\n",
    "    # Plot fit_time vs score\n",
    "    #     axes[2].grid()\n",
    "    #     axes[2].plot(fit_times_mean, 1- test_scores_mean , 'o-')\n",
    "    #     axes[2].fill_between(fit_times_mean, ( (1-test_scores_mean) - test_scores_std) ,\n",
    "    #                          ((1.- test_scores_mean ) + test_scores_std) , alpha=0.1)\n",
    "    #     axes[2].set_xlabel(\"fit_times\")\n",
    "    #     axes[2].set_ylabel(\"Score\")\n",
    "    #     axes[2].set_title(\"Performance of the model\")\n",
    "    return plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "measured-science",
   "metadata": {},
   "outputs": [],
   "source": [
    "def my_learn_curve(title=\"some title\", \n",
    "                   estimator=None, \n",
    "                   cv = None ,\n",
    "                   X=None,\n",
    "                   y=None,\n",
    "                   label_suffix =  None,\n",
    "                   color_= [],\n",
    "                   shape='o-'):\n",
    "    plt.title(title)\n",
    "    plt.ylim(0.0, 1.01)\n",
    "    plt.xlabel(\"Training examples\")\n",
    "    plt.ylabel(\"Score\")\n",
    "\n",
    "    train_sizes, train_scores, test_scores, fit_times, _ = \\\n",
    "        learning_curve(estimator, X, y, cv=cv, n_jobs=4,\n",
    "                       train_sizes=np.linspace(.1, 1.0, 10),\n",
    "                       return_times=True)\n",
    "    train_scores_mean =  1.0 - np.mean(train_scores, axis=1)\n",
    "    train_scores_std = np.std(train_scores, axis=1)\n",
    "    test_scores_mean =  1.0 - np.mean(test_scores, axis=1)\n",
    "    test_scores_std = np.std(test_scores, axis=1)\n",
    "    fit_times_mean = np.mean(fit_times, axis=1)\n",
    "    fit_times_std = np.std(fit_times, axis=1)\n",
    "\n",
    "\n",
    "    # Plot learning curve\n",
    "    plt.grid()\n",
    "    plt.fill_between(train_sizes,  train_scores_mean - train_scores_std,\n",
    "                         train_scores_mean + train_scores_std, alpha=0.1,\n",
    "                         color=\"r\")\n",
    "    plt.fill_between(train_sizes,  test_scores_mean - test_scores_std,\n",
    "                         test_scores_mean + test_scores_std, alpha=0.1,\n",
    "                         color=\"g\")\n",
    "    plt.plot(train_sizes, train_scores_mean, shape, color=color_[0],\n",
    "                 label=\"Training Error - {}\".format(label_suffix))\n",
    "    plt.plot(train_sizes, test_scores_mean,shape, color=color_[1],\n",
    "                 label=\"Cross-validation Error - {}\".format(label_suffix))\n",
    "    plt.legend(loc=\"best\")\n",
    "    plt.ylabel(\"Error\")\n",
    "    return plt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "valid-header",
   "metadata": {},
   "source": [
    "# BASE LEARNING CURVE LIVER "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "equivalent-bidding",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "my_learn_curve(title=\"Adaboost\", \n",
    "               estimator=liver_pipeline, \n",
    "               cv=stratified_kfold, \n",
    "               X=X_train,\n",
    "               y=y_train,\n",
    "               label_suffix=\"No Tuning\", \n",
    "               color_=[\"orange\",\"blue\"],\n",
    "               shape = '^-'  )\n",
    "\n",
    "plt.savefig(\"ADABOOST_LIVER_TREE_BASE.pdf\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "enhanced-microwave",
   "metadata": {},
   "source": [
    "## Grid search booster Liver"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "romantic-badge",
   "metadata": {},
   "outputs": [],
   "source": [
    "AdaBoostClassifier?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "digital-yeast",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# cv = ShuffleSplit(n_splits=10, test_size=0.1, random_state=0)\n",
    "cv = RepeatedStratifiedKFold(n_splits=10, n_repeats=2, random_state=45)\n",
    "train_scores, valid_scores = validation_curve(liver_pipeline,\n",
    "                                              X_train,\n",
    "                                              y_train, \n",
    "                                              param_name = \"model__n_estimators\",\n",
    "                                            param_range =np.array(range(1,100, 10)),\n",
    "                                            cv=cv)\n",
    "\n",
    "param_range = np.array(range(1,100, 10))\n",
    "\n",
    "train_scores_mean =  np.mean(train_scores, axis=1)\n",
    "train_scores_std = np.std(train_scores, axis=1)\n",
    "test_scores_mean = np.mean(valid_scores, axis=1)\n",
    "test_scores_std = np.std(valid_scores, axis=1)\n",
    "\n",
    "plt.title(\"Validation Curve with Adaboost\")\n",
    "plt.xlabel(\"Number of Estimators\")\n",
    "plt.ylabel(\"Score\")\n",
    "plt.ylim(0.0, 1.1)\n",
    "lw = 2\n",
    "# semilogx\n",
    "plt.plot(param_range, train_scores_mean, label=\"Training score\",\n",
    "             color=\"darkorange\", lw=lw)\n",
    "plt.fill_between(param_range, train_scores_mean - train_scores_std,\n",
    "                 train_scores_mean + train_scores_std, alpha=0.2,\n",
    "                 color=\"darkorange\", lw=lw)\n",
    "plt.plot(param_range, test_scores_mean, label=\"Cross-validation score\",\n",
    "             color=\"navy\", lw=lw)\n",
    "plt.fill_between(param_range, test_scores_mean - test_scores_std,\n",
    "                 test_scores_mean + test_scores_std, alpha=0.2,\n",
    "                 color=\"navy\", lw=lw)\n",
    "plt.legend(loc=\"best\")\n",
    "\n",
    "\n",
    "plt.savefig(\"ADA_LIVER_num_estimators.pdf\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "cardiac-bidding",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEWCAYAAABrDZDcAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAw90lEQVR4nO3deZwcVbn/8c/TyyzZVwIkQBIMRCAbDGEnYbmERUGucCEgS1wAEVD5selViXq5ohcFUQQREUEkAVkEyRXlQtiXBE2AQAghBBiWmIXsmaW7n98fVd3p6emZ6YTpmczU9/16VbqWU9XnVE/OU3Wq6pS5OyIiEl2xzs6AiIh0LgUCEZGIUyAQEYk4BQIRkYhTIBARiTgFAhGRiFMgkGbMzM3sU+H4TWb23VLSbsX3nG5mf9vafHZ3ZrbezEa2snypmR1Zpu8eHv62iRLSTjaz2nLkQzqGAkE3ZGaPmNkPisw/wcw+KuU/d5a7n+fuP2yHPDWrWNz9Tnc/6pNuu4Xv62Nm15nZu2GFujicHlSO7ysHd+/l7ksAzOw2M/uvT7rNsNJ2M7vsk+ew85jZbDP7cmfno7tQIOiebgPOMDMrmH8GcKe7pzo+Sx3HzCqA/wP2BI4G+gAHAiuBiVuxvZIDZxdwFrAq/BQJuLuGbjYA1cAa4NC8ef2BOmAcQWX4HLAa+BD4JVCRl9aBT4XjtwH/lbfs0nCdD4AvFqQ9DvgnsBZ4D5iet967Ydr14XAAcDbwdF6aA4E5Yd7nAAfmLZsN/BB4BlgH/A0Y1EL5vwwsA3q1so9y+S4sJzAZqAUuBz4C7gBeBz6Tlz4BrAD2Dqf3B54N9+l8YHIL3zsNeChvejFwd970e8D4/DwC5wCNQEO47x4Kly8FLgFeDvfZTKCqlTL3CPfdqeG2avKWxYFrwjItAb4Wfn8iL9+vh+svAc7NWze7v74drr8UOD1veV/gdmA58A7wHSAWLouF0+8A/wrT9Q2XVQF/IAjgq8O/iSHAVUCa4O95PfDLzv4/19WHTs+AhjL9sPAb4Ja86XOBeeH4PmHFlQCGh//Bv5GXtmggIDi6XgbsBfQE/liQdjIwJvzPPTZM+7lw2fD8iiWcdzZhIAAGAB8TnLUkgKnh9MBw+WzgLWA3gkA3G7i6hbLPAH7fxv5pKxCkgB8DleH3fY/gbCqb/jhgYTg+NKysjg3L/m/h9OAi3zsyrNRiwA5hBfh+3rKP8yrJFgNyOG8p8CKwY7j/XgfOa6XMZxAE8TjwEHB93rLzgIXATuG2HqdpIDgO2BUwYBKwkc1BMLu/fhbur0nABmD3cPntwJ+B3uHfwSLgS+GyLxIEw5FAL+A+4I68v9mHCAJYnODvtk/e38OXO/v/WXcZ1DTUff0eONnMqsPpM8N5uPtL7v68u6fcfSnwa4L/vG35D+B37v6qu28ApucvdPfZ7v6Ku2fc/WXgrhK3C0FF86a73xHm6y6CiumzeWl+5+6L3H0TcDcwvoVtDSSo8D6JDHClu9eH3/dH4Hgz6xEuPy2cB/AFYJa7zwrL/ndgLkFgaMKDNv91Yd4nAY8A75vZ6HD6KXfPbEE+r3f3D9x9FUGlOb6VtGcBM909HeZ9qpklw2X/AVzn7u+F2/pRQb4fdve3PPAEwRnZIQXb/264v54AHgb+w8ziwCnAt9x9Xfj39lOCoARwOvAzd1/i7uuBbwGnhs1xjQS/5afcPR3+3a7dgn0jJVIg6Kbc/WmCU/ETwjtP9iWsuMxsNzP7S3jheC3w30ApF1F3JGi6yHonf6GZ7Wdmj5vZcjNbQ3CUWerF2R0LtxdOD82b/ihvfCPBEWQxKwmOtj+J5e5el51w98UER9yfDYPB8WwOBLsQBN3V2QE4uJU8PEFwFH1oOD6bIAhMCqe3REn7xMx2Ag4D7gxn/Zmg6eW4cLqt3/YYM3vezFaF5TuWpr/tx+HBQf76O4ZpKgq2l/+7Fv7u7xCcEQ4haJJ7BJhhZh+Y2U/yApe0IwWC7u12gjOBM4C/ufuycP6NBEfbo9y9D0HbbuGF5WI+JGg6yNq5YPkfgQeBndy9L3BT3nbb6ub2A4IKNd/OwPsl5KvQo8AUM+vZSpqNBE0OWdsXLC+W37sImqxOAF4LgwMEFegd7t4vb+jp7le38N3ZQHBIOP4EbQeCT9pN8BkE/98fMrOPCNr5qwj+PqCV39bMKoF7Ca4hDHH3fsAsmv7N9C/Y3zsT/KYrCI7sdylYlv1dC3/3nQmamZa5e6O7f9/d9yC4fvSZvPyq2+R2pEDQvd0OHAl8hbBZKNSb4ILu+rBJ4qslbu9u4Gwz2yM8Kr6yYHlvYJW715nZRILmk6zlBM0tLd0XPwvYzcxOM7OEmZ0C7AH8pcS85buDoHK+18xGm1nMzAaa2bfNLNtcMw84zcziZnY0pTVhzQCOIthff8yb/weCM4Up4faqwts0h7WwnScIjs6r3b0WeIrg+stAgovtxSyj5X1XijOB7xM0HWWHzwPHmdlAgt/2IjMbZmb9gSvy1q0gaPtfDqTM7BiC/VDo+2ZWYWaHEFTa94TNUHcDV5lZbzPbBbiYYJ9BEFy/aWYjzKwXwdnpTHdPmdlhZjYmbF5aSxBQ0u20PySPAkE3FrbHPktwYffBvEWXEFTS6wguKs8scXv/C1wHPEZwge+xgiTnAz8ws3UEF1fvzlt3I8HdHs+EzSf7F2x7JUHl8f8ImnYuI7hLZ0UpeSvYVj1BAFwI/J2gEnmRoJnihTDZ1wmuP6wmaKd+oITtfkhwt9WB5O0zd3+P4Czh2wSV5XsEd1cV/f/l7osI7nZ5KpxeS3CE/kxYcRbzW2CPcN+1mdd84b4eDtzg7h/lDQ8S/I5TCf4OHiG44+kfBBdts/ldB1xE8Ht+TPC38yBNfRQu+4Cg+ek8d18YLruQ4OLxEuBpgiB6a7jsVoLA/STwNsGdQBeGy7YH/kTw+71OEECzAeTnwElm9rGZXb8l+0OaM3edYYmIRJnOCEREIk6BQEQk4hQIREQiToFARCTiulxnWoMGDfLhw4d3djZERLqUl156aYW7Dy62rMsFguHDhzN37tzOzoaISJdiZoVP7ueoaUhEJOIUCEREIk6BQEQk4rrcNQKR7qyxsZHa2lrq6uraTixSRFVVFcOGDSOZLL2jVgUCkW1IbW0tvXv3Zvjw4TR/06hI69ydlStXUltby4gRI0peT01DItuQuro6Bg4cqCAgW8XMGDhw4BafUSoQiGxjFATkk9iavx8FAhGRiFMgEJGclStXMn78eMaPH8/222/P0KFDc9MNDQ2trjt37lwuuuiiNr/jwAMPbK/sSjvRxWIRyRk4cCDz5s0DYPr06fTq1YtLLrkktzyVSpFIFK82ampqqKmpafM7nn322XbJa3trrWzdnc4IRKRVZ599NhdffDGHHXYYl19+OS+++CIHHnggEyZM4MADD+SNN94AYPbs2XzmM58BgiDyxS9+kcmTJzNy5Eiuv37zS8R69eqVSz958mROOukkRo8ezemnn072RVmzZs1i9OjRHHzwwVx00UW57eZbsGABEydOZPz48YwdO5Y333wTgNtvv52xY8cybtw4zjjjDADeeecdjjjiCMaOHcsRRxzBu+++W7Rsb731FkcffTT77LMPhxxyCAsXLmz2vd1RNMOfSFfw0zJdNP5/W/5WwkWLFvHoo48Sj8dZu3YtTz75JIlEgkcffZRvf/vb3Hvvvc3WWbhwIY8//jjr1q1j991356tf/Wqze9v/+c9/smDBAnbccUcOOuggnnnmGWpqajj33HN58sknGTFiBFOnTi2ap5tuuomvf/3rnH766TQ0NJBOp1mwYAFXXXUVzzzzDIMGDWLVqlUAXHDBBZx55pmcddZZ3HrrrVx00UU88MADzcp2xBFHcNNNNzFq1CheeOEFzj//fB57rPCNrN2PAoGItOnkk08mHo8DsGbNGs466yzefPNNzIzGxsai6xx33HFUVlZSWVnJdtttx7Jlyxg2bFiTNBMnTszNGz9+PEuXLqVXr16MHDkydx/81KlTufnmm5tt/4ADDuCqq66itraWf//3f2fUqFE89thjnHTSSQwaNAiAAQMGAPDcc89x333Ba5jPOOMMLrvssmZlW79+Pc8++ywnn3xybll9ff1W7a+uRoFAZFu1FUfu5dKzZ8/c+He/+10OO+ww7r//fpYuXcrkyZOLrlNZWZkbj8fjpFKpktKU+h710047jf3224+HH36YKVOmcMstt+DuJd0+mZ8mW7ZMJkO/fv1y10iiRNcIRGSLrFmzhqFDhwJw2223tfv2R48ezZIlS1i6dCkAM2fOLJpuyZIljBw5kosuuojjjz+el19+mSOOOIK7776blStXAuSahg488EBmzJgBwJ133snBBx/cbHt9+vRhxIgR3HPPPUDwlO78+fPbu3jbpLIFAjO71cz+ZWavtrDczOx6M1tsZi+b2d7lyouItJ/LLruMb33rWxx00EGk0+l23351dTW/+tWvOProozn44IMZMmQIffv2bZZu5syZ7LXXXowfP56FCxdy5plnsueee/Kf//mfTJo0iXHjxnHxxRcDcP311/O73/2OsWPHcscdd/Dzn/+86Hffeeed/Pa3v2XcuHHsueee/PnPf2738m2LrNTTsC3esNmhwHrgdnffq8jyY4ELgWOB/YCfu/t+bW23pqbG9WIa6a5ef/11Pv3pT3d2Njrd+vXr6dWrF+7O1772NUaNGsU3v/nNzs5Wl1Hs78jMXnL3ovf3lu2MwN2fBFa1kuQEgiDh7v480M/MdihXfkSk6/jNb37D+PHj2XPPPVmzZg3nnntuZ2epW+vMi8VDgffypmvDeR8WJjSzc4BzAHbeeecOyZyIdJ5vfvObOgPoQJ15sbjYpf2i7VTufrO717h7zeDBRd+9LCIiW6kzA0EtsFPe9DDgg07Ki4hIZHVmIHgQODO8e2h/YI27N2sWEhGR8irbNQIzuwuYDAwys1rgSiAJ4O43AbMI7hhaDGwEppUrLyIi0rJy3jU01d13cPekuw9z99+6+01hECC8W+hr7r6ru49xd90TKrIN+Oijjzj11FPZdddd2WOPPTj22GNZtGhRZ2ermdtuu40LLrgACPoduv3225ulWbp0KXvt1ezu9WZp/vjHP+amS+1OuztRFxMikuPunHjiiZx11lm5J3HnzZvHsmXL2G233XLp0ul0ru+hbcF555231etmA8Fpp50GlN6ddkcr5z5XFxMikvP444+TTCabVKzjx4/nkEMOYfbs2Rx22GGcdtppjBkzhrq6OqZNm8aYMWOYMGECjz/+OFC8e+gNGzZw3HHHMW7cOPbaa69m3UZkMhmGDx/O6tWrc/M+9alPsWzZMh566CH2228/JkyYwJFHHsmyZcua5Xv69Olcc801ALz00kuMGzeOAw44gBtuuCGXZunSpRxyyCHsvffe7L333rn3IlxxxRU89dRTjB8/nmuvvbZJd9qrVq3ic5/7HGPHjmX//ffn5Zdfzn1fS91sZ6XTac4++2z22msvxowZw7XXXgvA4sWLOfLIIxk3bhx77703b731Fu7OpZdemkub3T+F+zydTnPppZey7777MnbsWH79619v2Q/cAp0RiGyjzL5flu26X9nisldffZV99tmnxeUvvvgir776KiNGjOCnP/0pAK+88goLFy7kqKOOYtGiRUW7h541axY77rgjDz/8MBD0V5QvFotxwgkncP/99zNt2jReeOEFhg8fzpAhQzj44IN5/vnnMTNuueUWfvKTn+S+u5hp06bxi1/8gkmTJnHppZfm5m+33Xb8/e9/p6qqijfffJOpU6cyd+5crr76aq655hr+8pe/AEHlm3XllVcyYcIEHnjgAR577DHOPPPMXKd0bXWzPW/ePN5//31efTXoZScb5E4//XSuuOIKTjzxROrq6shkMtx3333MmzeP+fPns2LFCvbdd18OPfTQZvv85ptvpm/fvsyZM4f6+noOOuggjjrqqFxPrVtLZwQiUrKJEyfmKp2nn3469+KX0aNHs8suu7Bo0SIOOOAA/vu//5sf//jHvPPOO1RXVzNmzBgeffRRLr/8cp566qmifQedcsopuSPhGTNmcMoppwBQW1vLlClTGDNmDP/zP//DggULWszfmjVrWL16NZMmTQLI5Q+gsbGRr3zlK4wZM4aTTz6Z1157rc3y5pfx8MMPZ+XKlbkglu1me9CgQblutvONHDmSJUuWcOGFF/LXv/6VPn36sG7dOt5//31OPPFEAKqqqujRowdPP/00U6dOJR6PM2TIECZNmsScOXOa7fO//e1v3H777YwfP5799tuPlStX5l7I80nojEBkG9XakXu57LnnnvzpT39qcXl+d9Qt9VNWrHvoww8/nJdeeolZs2bxrW99i6OOOoopU6bkuo74wQ9+wGc/+1kWL17M8uXLeeCBB/jOd74DwIUXXsjFF1/M8ccfz+zZs5k+fXqL+WutG+prr72WIUOGMH/+fDKZDFVVVW3tjqJlzG6/rW62+/fvz/z583nkkUe44YYbuPvuu7nuuutK/p6swn3+i1/8gilTprSZ9y2hMwIRyTn88MOpr6/nN7/5TW7enDlzeOKJJ5qlPfTQQ7nzzjuB4C1f7777LrvvvnvR7qE/+OADevTowRe+8AUuueQS/vGPf7Dffvsxb9485s2bx/HHH4+ZceKJJ3LxxRfz6U9/moEDBwJNu73+/e9/32r++/XrR9++fXn66acBcvnLbmeHHXYgFotxxx135HpO7d27N+vWrSu6vfwyzp49m0GDBtGnT5+S9uWKFSvIZDJ8/vOf54c//CH/+Mc/6NOnD8OGDcu9Ha2+vp6NGzdy6KGHMnPmTNLpNMuXL+fJJ59k4sSJzbY5ZcoUbrzxxtzLgBYtWsSGDRtKyk9rdEYgIjlmxv333883vvENrr76aqqqqhg+fDjXXXcd77//fpO0559/Pueddx5jxowhkUhw2223UVlZycyZM/nDH/5AMplk++2353vf+x5z5szh0ksvJRaLkUwmufHGG4t+/ymnnMK+++7b5D0H06dP5+STT2bo0KHsv//+vP32262W4Xe/+x1f/OIX6dGjR5Mj5/PPP5/Pf/7z3HPPPRx22GG5I+2xY8eSSCQYN24cZ599NhMmTGjy3dOmTWPs2LH06NGjzUCU7/3332fatGlkMhkAfvSjHwFwxx13cO655/K9732PZDLJPffcw4knnshzzz3HuHHjMDN+8pOfsP322zd7Z/KXv/xlli5dyt577427M3jw4FxQ+STK1g11uagbaunO1A21tIdtphtqERHpGhQIREQiToFAZBvT1ZprZduyNX8/CgQi25CqqipWrlypYCBbxd1ZuXJlSbfG5tNdQyLbkGHDhlFbW8vy5cs7OyvSRVVVVTFs2LAtWkeBQGQbkkwmP3F3ASJbSk1DIiIRp0AgIhJxCgQiIhGnQCAiEnEKBCIiEadAICIScQoEIiIRp0AgIhJxCgQiIhGnQCAiEnEKBCIiEadAICIScQoEIiIRp0AgIhJxCgQiIhGnQCAiEnEKBCIiEVfWQGBmR5vZG2a22MyuKLK8r5k9ZGbzzWyBmU0rZ35ERKS5sgUCM4sDNwDHAHsAU81sj4JkXwNec/dxwGTgp2ZWUa48iYhIc+U8I5gILHb3Je7eAMwATihI40BvMzOgF7AKSJUxTyIiUqCcgWAo8F7edG04L98vgU8DHwCvAF9390zhhszsHDOba2Zzly9fXq78iohEUjkDgRWZ5wXTU4B5wI7AeOCXZtan2UruN7t7jbvXDB48uL3zKSISaeUMBLXATnnTwwiO/PNNA+7zwGLgbWB0GfMkIiIFyhkI5gCjzGxEeAH4VODBgjTvAkcAmNkQYHdgSRnzJCIiBRLl2rC7p8zsAuARIA7c6u4LzOy8cPlNwA+B28zsFYKmpMvdfUW58iQiIs2VLRAAuPssYFbBvJvyxj8AjipnHkREpHV6slhEJOIUCEREIk6BQEQk4hQIREQirqwXi0VEujzPBEOLywufky2aqIQkJaSJJSAWL+H7towCgYh0Te6bK+lcZZ1pPq/YsMXpnFYr81JiQXsEg3gSBo2BRGUpX1gyBQIRaVu20sU3V4r58wqnW0uDN61c3fGw4nUvmM44Rgaz8JO8yh+HTDrvOzKbt53JQKYR0vWbh1Q9pBsgXRcuq4N0Nk1DMGTyPxuDz0wq74zAm+6TZtM0S+eeCbIMYfmCz83THhQhuwwPN+04+cszpHvvSt8TbiSmQCASQU0q0EzTys8LKsdi4+6kUynS6TTpVJp0KkMqFY6nM+FnMJ5qDD49E1bIgGeC7WUrK3A840AmrP+y055XueV9ZucTVniZDJauI57ZQDyzgVh6PfH0xmA6HQ6ZjcQzmzBPEacB80ZieYN5ihjhp6ew7DxvDMZLO0wvO6N4x2tbY23datavb6BPdTttMKRAINJZ3MMj1brw6DQczx6B5lXm7hkyeZV0OpUJKu6MB5V6dn7awwrdSWcypBozYfpgfibjpDNOJp0hk3FSqeAzk84E8zNOOp0hk3Yy2SNcd8hWqwbuBu4kqCNJHUk2kWQjSd9EBZtIsJEK6qjwjSQtO38jSTaGn8F0jHR5dy8x0iRJWwUZkmQsGNJU5MaDIZymArckaUviVkEmFix3S+LEgyP4jIWfQTzMuAVxGMdz48EJiXteKDLDzDALA4PFgvHgn83LwnEKps1g7cdrqe67I8PLsK8UCETKLR02Q6SCyj5Vt5FU/aZwqCfV2ECqoZ5UfQOphgbS6RSpVFjZZzyoxDNOOhNUPOk0pB0ymVj4aWQykApbSjIeI5123NNBUMk0EPMUiViKpKVIxhpz49WWHW8kbo0kEiniyUYS1JPwjSTSG0n4ehKZDSTS64lngul4ZmNpR9ytJMlYBel4L9LxnqRiPXPj6VhP0rEewWBB5RxUyImgwvZEWKEnyJAIKnkSpC2Yzs534kETTNi0QiaTa47xTLbZhbyKPO+sJnei5bkWHzOImRGLG7E4WCycts3jZhCLhfMM4jGwWLGbMy3bWFayD9Y1MqDHdgoEItuadDpDKpUh1ZgiVbcpqOQb6kjV14UVfV1QyTc0bB5SjaQb6rD6dZhvIpauJ+abiGeCIeaNkGkk5g0kvZGKTAMxGknQQIJG4hZ+0kCcRuLeSJx6Yh7Mi3lD0DxCwZ0urdz4slVlj/UIKvF4r6DyjvcKhljPoELPVvLxXqRjvUjHqkhbdThU4Rh4CjwdtvWHnxYDiwd3x8QSbG5YCRtZzDZP5qYNw4gTdGyWnZdb1/I/8xtqstuk6Wfh/MLGHSuYbrK9Ume3lLaF+es3QWUPSPZs4bu3ngKBSCjbVNLi0JgOjuAbwiP5+k2k69ZhdSuw+lXE6j/GGlcTb1hDIrWGeHotifRaqj1sFgk/K9lABRvK3oadsUTQxGEVeCxsAolVhs0eFeFRdna8Eo/lp60kHe+VV8nnHbHHe5GO9Qgqa6d4ZV5sXraajsUglgZLQLxyc6Vv8WBerGBoEgDyK+2CwEAblX/h+sUq/MJttljhd4K1a6BPD90++klsXLuWha99xIaNjZ2dFdkmZajMbKCSNcTqVxGrW06sfiXxhpUkGj8m0biaqvRqEul1JDPrqPB1VLCBJHVb/Y1Bc0hvUok+wWe8N+l4LzJWmVc5J8PKuSKsyCtwqwzbryuKVOTJXFpsKyuM8OIv2aYlz6vM02lo3ACZtUA6aAyPhRV8PBFW5nGIVxVU8PHmFXx2sES4bnZ6G6p8IyIygeCSL1zFjQ/16OxsSBdQmUhRnWykKpGiKpGgKtmf6mRvqhI7BPOSm5dXJtNUJo3KyhgVFXEqKhIkKxIkKyuorKwgUVlFsrKKRFU1FVXVJKp6Eq/qSaKqF5VVSSqT2fViVFbEsJhtroizbTm5O4TI+8wu33wXT3iFEqgH30Tuls1mt3MWzAM23y+ftyNiCYIKPt60gk/kVfDZ5ptiR/K5Cj4ZplcFv62KTCCoqozRu7K+s7Mh2ygHGtNx6lOJ3NC+MsCGcGhZZRL69YnTr0+MAX3i9OsbZ0DfOP37BMOAfgn69w2GAf0SVFfHad7MYXkVb3ihsnA811Yey97G0nSaWLiNgqP1wopeFXy3EJlA8JO7ruKUZ17j7YXvsmXX6iUqMhgZEsTiYZOKB7dQuoe3VaachoYM9Q1p6uvTRceDzwwNDenwM5iur0/T0Jg/HY5n59WnaWh06hth2co0y1amgbabMauq4vTvW0H//pX071fJgH6VwXj/Kvr3r2RA/yr696uk/4Bq+vevprIyEV6MhSAwZANBjFwQUcUeOZEJBIlEjEHDdqS6/+DOzopso9LpoFJubExRX5+msTGo0BsaUjQ0ZEgCfSvjJHPNOcGQTMZJJD55/42ZjFNfn2LNmnpWrdrEqlWbWL26jlWrNvHxx3V8/PEmVq0KPrPTdXVpPqzbxIfLNpX0HT17Junfv5oBA6ro1y/47N+/mv79qxgwoJp+/TZ/9uiRpEePJLGYAkN3F5lAALDrrgM6OwuyjWtsDI7w6+tT1NWlmowXBon16xtpaNhEQ0PQbl/5CYNELGZUVyeprk6y/fa92kzv7mzc2Fg0SBQPIpvYsKGRDRsaqa1dW/I+qa5O0KNHkp49k/ToUZE33nTIn5dNWzivqioRPDDVyTIZp7ExP9inm0w3NqbD3ztY5g7xuJFIxIjFjHg8Rjxu4RDM27wsf3nTz+y62bTbSpCNVCAQaUsyGVTevXpVNFvWmUGiGDOjZ88KevasYNiwPm2md3fWr29o8QwjmB+Mr15dx8aNjWzalMoNK1eWdtbRep5pIXg0Dxo9eiRJJuN5FXPTirqwEs+vuBsb88ebL0ul2vmhiq2UfQAtkYjlAkpLASeTgd12G8j995/S7vlQIBApUUcFicrKBJWV8XY/cjYzeveupHfvSnbZpbR1MpngrKNw2LAh+9nQbF6xdNm09fXp3FlJZ8vu82QyG5hjuc/N84I7usws7L4jk+uGI5Xa3CXH5u49nEwmGM9235FdL3/9VCoTPv1NrlsQSuhyo0+f9u1sLkuBQKQdtGeQWLFiE6lUJtckU1WVoLo6QTze8e+RisWMXr0qipZra6RSGTZtaj1o5AeXxsZMQSVdWGk3DaLZijv7e2THNy8P0iYSsW2miSo/SASBJBsomgaS2tq1DBzY/k8VgwKBSNltaZDIVop1dUFluGLFRurqUlRWxsNrCIlcs0lXk0jEcmclEvZLFIuTTLad1swYMKA8z0IpEIh0opaCRGNjmvXrG3JD0F7fyMaNKdata2DZsg3hxeUgKFRXJ8vSnCTRoEAgsg1KJuPhbZ1Bx/OZjLNhw+bAkG1G2bQpGFatqmvSnFRdnaC6Wrd+SmkUCES6gFjMmjSpuDt1damCs4ZUrjlp+fKN1NWlqayMUV2dzAWHrticJOWnQCDSBZltfuZg8ODgAmKx5qTsLaBr19bz0Ufr1ZwkRSkQiHQTak6SraVAINJNldqclA0Mhc1JVVUJEongKdhkMtYpt69Kx1AgEImILWlOygaI4And4ElcdyeZzAaGeJPxbMBojz6XtiX59/bndx/R3SgQiERYa81JwUNu6bxuHIKAEHymSaUy1NUFt7Nm57t77gwiGyDyxzsrWBR7cKutz0zGMcv2JRS87iF4yCvTav9CTbuGaJpuW212K2sgMLOjgZ8TvKPuFne/ukiaycB1QBJY4e6TypknEWlZYXNSvqB31mJ9/DTt32fzWcTmYLF+fWNuWX6waO3sohj35l02lPIJFHQAFyMeJ+wALk5lZWEHccHyRCKem5fdViqVJp0m7EoiQzq9OUAEy536+sbcE8L5TwcH+Wi9o7pigSQeL28AKVsgMLM4cAPwb0AtMMfMHnT31/LS9AN+BRzt7u+a2Xblyo+IfDLxeIzq6hjV1S2nSaczrQaK/A7jgrOIIFjU16dZv76RVCpYlsl4LiBsrvyDd0PkV5RNO2fbfC2jWO+f+Wck2XU3TxdfXuyOqmx+st1AZPsNKnW6WIDInoHU16cL+iTyXMABOv/JYjOrBnZ29zdKXGUisNjdl4TrzwBOAF7LS3MacJ+7vwvg7v8qNT8isu3JVsJVVS1XLZmMNwsUxYJHKhVUotkmlfzKv1gF3tqy9mySMTMSCduqJq78foS2NICk0048blRWdtLL683ss8A1QAUwwszGAz9w9+NbWW0o8F7edC2wX0Ga3YCkmc0GegM/d/fbi3z/OcA5ADvvvHMpWRaRbVQsZlRVJdoMFtmAkG1CyVbuXVksZlRUxAlay7dMOp3BrDzXGUo9I5hOcIQ/G8Dd55nZ8DbWKZbbwndEJoB9gCOAauA5M3ve3Rc1Wcn9ZuBmgJqaGr1nUqSbi8Us7I5b97NklTMIlrqXU+6+ZgufQKwFdsqbHgZ8UCTNCnffAGwwsyeBccAiRESkQ5QaYl41s9OAuJmNMrNfAM+2sc4cYJSZjTCzCuBU4MGCNH8GDjGzhJn1IGg6en0L8i8iIp9QqYHgQmBPoB74I7AG+EZrK7h7CrgAeISgcr/b3ReY2Xlmdl6Y5nXgr8DLwIsEt5i+uhXlEBGRrWTurTe5h7eBPuLuR3ZMllpXU1Pjc+fO7exsiIh0KWb2krvXFFvW5hmBu6eBjWbWt91zJiIina7Ui8V1wCtm9ndgQ3amu19UllyJiEiHKTUQPBwOIiLSzZQUCNz99+GdP7uFs95w98byZUtERDpKqU8WTwZ+DywleFBsJzM7y92fLFvORESkQ5TaNPRT4KhsP0NmthtwF8FTwSIi0oWV+hxBMr+zubALiGR5siQiIh2p1DOCuWb2W+COcPp04KXyZElERDpSqYHgq8DXgIsIrhE8SfAeARER6eJKDQQJgi6ifwa5p42bv8JIRES6nFKvEfwfQTfRWdXAo+2fHRER6WilBoIqd1+fnQjHy/PONBER6VClBoINZrZ3dsLMaoBN5cmSiIh0pFKvEXwDuMfMPiB4y9iOwCnlypSIiHScVs8IzGxfM9ve3ecAo4GZQIrgHQJvd0D+RESkzNpqGvo10BCOHwB8G7gB+JjwHcIiItK1tdU0FHf3VeH4KcDN7n4vcK+ZzStrzkREpEO0dUYQN7NssDgCeCxvWanXF0REZBvWVmV+F/CEma0guEvoKQAz+xTBe4tFRKSLazUQuPtVZvZ/wA7A33zzC45jBC+0FxGRLq7N5h13f77IvEXlyY6IiHS0Uh8oExGRbkqBQEQk4hQIREQiToFARCTiFAhERCJOgUBEJOIUCEREIk6BQEQk4hQIREQiToFARCTiyhoIzOxoM3vDzBab2RWtpNvXzNJmdlI58yMiIs2VLRCYWZzgJTbHAHsAU81sjxbS/Rh4pFx5ERGRlpXzjGAisNjdl7h7AzADOKFIuguBe4F/lTEvIiLSgnIGgqHAe3nTteG8HDMbCpwI3NTahszsHDOba2Zzly9f3u4ZFRGJsnIGAisyzwumrwMud/d0axty95vdvcbdawYPHtxe+RMREcr7uslaYKe86WHABwVpaoAZZgYwCDjWzFLu/kAZ8yUiInnKGQjmAKPMbATwPnAqcFp+AncfkR03s9uAvygIiIh0rLIFAndPmdkFBHcDxYFb3X2BmZ0XLm/1uoCIiHSMcp4R4O6zgFkF84oGAHc/u5x5ERGR4vRksYhIxCkQiIhEnAKBiEjEKRCIiEScAoGISMQpEIiIRJwCgYhIxCkQiIhEnAKBiEjEKRCIiEScAoGISMQpEIiIRJwCgYhIxCkQiIhEnAKBiEjEKRCIiEScAoGISMQpEIiIRJwCgYhIxCkQiIhEnAKBiEjEKRCIiEScAoGISMQpEIiIRJwCgYhIxCkQiIhEnAKBiEjEKRCIiEScAoGISMQpEIiIRFxZA4GZHW1mb5jZYjO7osjy083s5XB41szGlTM/IiLSXNkCgZnFgRuAY4A9gKlmtkdBsreBSe4+FvghcHO58iMiIsWV84xgIrDY3Ze4ewMwAzghP4G7P+vuH4eTzwPDypgfEREpopyBYCjwXt50bTivJV8C/rfYAjM7x8zmmtnc5cuXt2MWRUSknIHAiszzognNDiMIBJcXW+7uN7t7jbvXDB48uB2zKCIiiTJuuxbYKW96GPBBYSIzGwvcAhzj7ivLmB8RESminGcEc4BRZjbCzCqAU4EH8xOY2c7AfcAZ7r6ojHkREZEWlO2MwN1TZnYB8AgQB2519wVmdl64/Cbge8BA4FdmBpBy95py5UlERJoz96LN9tusmpoanzt3bmdnQ0SkSzGzl1o60NaTxSIiEadAICIScQoEIiIRp0AgIhJxCgQiIhGnQCAiEnEKBCIiEadAICIScQoEIiIRp0AgIhJxCgQiIhGnQCAiEnEKBCIiEadAICIScQoEIiIRp0AgIhJxCgQiIhGnQCAiEnEKBCIiEadAICIScQoEIiIRp0AgIhJxCgQiIhGnQCAiEnEKBCIiEadAICIScQoEIiIRp0AgIhJxCgQiIhGnQCAiEnEKBCIiEVfWQGBmR5vZG2a22MyuKLLczOz6cPnLZrZ3OfMjIiLNlS0QmFkcuAE4BtgDmGpmexQkOwYYFQ7nADeWKz8iIlJcOc8IJgKL3X2JuzcAM4ATCtKcANzugeeBfma2QxnzJCIiBRJl3PZQ4L286VpgvxLSDAU+zE9kZucQnDEArDezN7YgH4OAFVuQvruIYrmjWGaIZrmjWGb4ZOXepaUF5QwEVmSeb0Ua3P1m4OatyoTZXHev2Zp1u7IoljuKZYZoljuKZYbylbucTUO1wE5508OAD7YijYiIlFE5A8EcYJSZjTCzCuBU4MGCNA8CZ4Z3D+0PrHH3Dws3JCIi5VO2piF3T5nZBcAjQBy41d0XmNl54fKbgFnAscBiYCMwrQxZ2aompW4giuWOYpkhmuWOYpmhTOU292ZN8iIiEiF6slhEJOIUCEREIq7bBIIodmdRQplPD8v6spk9a2bjOiOf7a2tcuel29fM0mZ2UkfmrxxKKbOZTTazeWa2wMye6Og8lkMJf+N9zewhM5sflrsc1xk7lJndamb/MrNXW1je/nWZu3f5geBi9FvASKACmA/sUZDmWOB/CZ5d2B94obPz3QFlPhDoH44f09XLXGq589I9RnBDwkmdne8O+K37Aa8BO4fT23V2vjuo3N8GfhyODwZWARWdnfdPWO5Dgb2BV1tY3u51WXc5I4hidxZtltndn3X3j8PJ5wme0+jqSvmtAS4E7gX+1ZGZK5NSynwacJ+7vwvg7lEptwO9zcyAXgSBINWx2Wxf7v4kQTla0u51WXcJBC11VbGlabqSLS3PlwiOIrq6NsttZkOBE4GbOjBf5VTKb70b0N/MZpvZS2Z2ZoflrnxKKfcvgU8TPIj6CvB1d890TPY6TbvXZeXsYqIjtVt3Fl1IyeUxs8MIAsHBZc1Rxyil3NcBl7t7OjhQ7PJKKXMC2Ac4AqgGnjOz5919UbkzV0allHsKMA84HNgV+LuZPeXua8uct87U7nVZdwkEUezOoqTymNlY4BbgGHdf2UF5K6dSyl0DzAiDwCDgWDNLufsDHZLD9lfq3/cKd98AbDCzJ4FxQFcOBKWUexpwtQeN54vN7G1gNPBix2SxU7R7XdZdmoai2J1Fm2U2s52B+4AzuviRYb42y+3uI9x9uLsPB/4EnN+FgwCU9vf9Z+AQM0uYWQ+Cnn5f7+B8trdSyv0uwVkQZjYE2B1Y0qG57HjtXpd1izMC33a6s+gwJZb5e8BA4Ffh0XHKu3iPjSWWu1sppczu/rqZ/RV4GcgAt7h70dsPu4oSf+sfAreZ2SsETSaXu3uX7p7azO4CJgODzKwWuBJIQvnqMnUxISIScd2laUhERLaSAoGISMQpEIiIRJwCgYhIxCkQiIhEnAKBdAtmtr6Dv+/ZdtrOZDNbY2b/NLOFZnZNCet8zsz2aI/vFwEFApGizKzVZ2zc/cB2/Lqn3H0CMAH4jJkd1Eb6zwEKBNJuusUDZSLFmNmuwA0E3RNvBL7i7gvN7LPAdwi6Nl4JnO7uy8xsOrAjMBxYYWaLgJ0JukHeGbjO3a8Pt73e3XuZ2WRgOrAC2At4CfiCu7uZHQv8LFz2D2Cku3+mpfy6+yYzm0fYgZiZfQU4J8znYuAMYDxwPDDJzL4DfD5cvVk5t3a/SfTojEC6s5uBC919H+AS4Ffh/KeB/cOj8BnAZXnr7AOc4O6nhdOjCTo2mwhcaWbJIt8zAfgGwVH6SOAgM6sCfk3Qx9PBBJV0q8ysPzAKeDKcdZ+77+vu4wi6i/iSuz9L0MXApe4+3t3faqWcIiXRGYF0S2bWi+DFPPfk9UBaGX4OA2aGfbhXAG/nrfqgu2/Km37Y3euBejP7FzCEoNOvfC+6e234vfMIzijWA0vcPbvtuwiO7os5xMxeJugn52p3/yicv5eZ/RfBS2d6EXS1sCXlFCmJAoF0VzFgtbuPL7LsF8DP3P3BvKadrA0FaevzxtMU/z9TLM2W9H/9lLt/xsx2A542s/vdfR5wG/A5d59vZmcT9D9TqLVyipRETUPSLYX90b9tZidD7j2v2Xc29wXeD8fPKlMWFgIjzWx4OH1KWyuEPcT+CLg8nNUb+DBsjjo9L+m6cFlb5RQpiQKBdBc9zKw2b7iYoPL8kpnNBxaw+TWH0wmaUp4iuJDb7sLmpfOBv5rZ08AyYE0Jq94EHGpmI4DvAi8AfycILFkzgEvDW053peVyipREvY+KlImZ9XL39eH7dG8A3nT3azs7XyKFdEYgUj5fCS8eLyBojvp152ZHpDidEYiIRJzOCEREIk6BQEQk4hQIREQiToFARCTiFAhERCLu/wNvxLFvrtZ6ngAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# cv = ShuffleSplit(n_splits=10, test_size=0.1, random_state=0)\n",
    "cv = RepeatedStratifiedKFold(n_splits=10, n_repeats=2, random_state=45)\n",
    "train_scores, valid_scores = validation_curve(liver_pipeline,\n",
    "                                              X_train,\n",
    "                                              y_train, \n",
    "                                              param_name = \"model__learning_rate\",\n",
    "                                            param_range = np.linspace(.01, 1.0, 10),\n",
    "                                            cv=cv)\n",
    "\n",
    "param_range = np.linspace(.01, 1.0, 10)\n",
    "train_scores_mean =  np.mean(train_scores, axis=1)\n",
    "train_scores_std = np.std(train_scores, axis=1)\n",
    "test_scores_mean = np.mean(valid_scores, axis=1)\n",
    "test_scores_std = np.std(valid_scores, axis=1)\n",
    "\n",
    "plt.title(\"Validation Curve with Adaboost\")\n",
    "plt.xlabel(\"Learning Rate\")\n",
    "plt.ylabel(\"Score\")\n",
    "plt.ylim(0.0, 1.1)\n",
    "lw = 2\n",
    "# semilogx\n",
    "plt.plot(param_range, train_scores_mean, label=\"Training score\",\n",
    "             color=\"darkorange\", lw=lw)\n",
    "plt.fill_between(param_range, train_scores_mean - train_scores_std,\n",
    "                 train_scores_mean + train_scores_std, alpha=0.2,\n",
    "                 color=\"darkorange\", lw=lw)\n",
    "plt.plot(param_range, test_scores_mean, label=\"Cross-validation score\",\n",
    "             color=\"navy\", lw=lw)\n",
    "plt.fill_between(param_range, test_scores_mean - test_scores_std,\n",
    "                 test_scores_mean + test_scores_std, alpha=0.2,\n",
    "                 color=\"navy\", lw=lw)\n",
    "plt.legend(loc=\"best\")\n",
    "\n",
    "\n",
    "plt.savefig(\"ADA_LIVER_learning_rate.pdf\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "vietnamese-cologne",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "id": "written-preserve",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best: 0.814455 usin g {'model__base_estimator': DecisionTreeClassifier(max_depth=9, min_samples_leaf=22), 'model__learning_rate': 0.1, 'model__n_estimators': 49}\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# example of grid searching key hyperparameters for adaboost on a classification dataset\n",
    "from sklearn.datasets import make_classification\n",
    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "# define dataset\n",
    "# define the model with default hyperparameters\n",
    "\n",
    "model =  liver_pipeline\n",
    "# define the grid of values to search\n",
    "grid = dict()\n",
    "grid['model__n_estimators'] = [i for i in range(1,12,1)]\n",
    "grid['model__learning_rate'] = [0.01, .001, 0.1]\n",
    "grid['model__base_estimator'] = [DecisionTreeClassifier(max_depth=max_depth_,\n",
    "                                                 ccp_alpha=ccp, \n",
    "                                                 criterion=crit,\n",
    "                                                 min_samples_leaf = min_samples_split_)  \n",
    "                          for max_depth_ in range(1,4) \n",
    "                          for ccp in np.arange(0.0001, 0.051, 0.005)\n",
    "                          for crit in [\"entropy\"]\n",
    "                          for min_samples_split_ in range(6,16,2)\n",
    "                           ]\n",
    "\n",
    "# define the evaluation procedure\n",
    "cv = RepeatedStratifiedKFold(n_splits=10, n_repeats=2, random_state=45)\n",
    "# define the grid search procedure\n",
    "grid_search_l= GridSearchCV(estimator=liver_pipeline, param_grid=grid, n_jobs=-1, cv=cv, scoring='balanced_accuracy')\n",
    "# execute the grid search\n",
    "grid_result_l = grid_search_l.fit(X_train, y_train)\n",
    "# summarize the best score and configuration\n",
    "print(\"Best: %f usin g %s\" % (grid_result.best_score_, grid_result.best_params_))\n",
    "# summarize all scores that were evaluated\n",
    "means = grid_result_l.cv_results_['mean_test_score']\n",
    "stds = grid_result_l.cv_results_['std_test_score']\n",
    "params = grid_result_l.cv_results_['params']\n",
    "# for mean, stdev, param in zip(means, stds, params):\n",
    "#     print(\"%f (%f) with: %r\" % (mean, stdev, param))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "interested-jordan",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Best: 0.632695 usin g {'model__base_estimator': DecisionTreeClassifier(ccp_alpha=0.0101, criterion='entropy', max_depth=3,\n",
    "#                        min_samples_leaf=14), 'model__learning_rate': 0.01, 'model__n_estimators': 5}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "touched-trance",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "my_learn_curve(title=\"Adaboost\", \n",
    "               estimator=grid_result.best_estimator_, \n",
    "               cv=stratified_kfold, \n",
    "               X=X_train,\n",
    "               y=y_train,\n",
    "               label_suffix=\"No Tuning\", \n",
    "               color_=[\"orange\",\"blue\"],\n",
    "               shape = '^-'  )\n",
    "\n",
    "\n",
    "plt.savefig(\"ADABOOST_LIVER_TREE_final.pdf\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "greenhouse-gospel",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7f0bbd550730>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(estimator= ada, X=X_test, y_true= y_test  , \n",
    "                      cmap=plt.cm.Blues,\n",
    "                       normalize=\"true\",\n",
    "                      display_labels = [\"Not Patient\" ,\"Patient\"]\n",
    "                        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "liberal-tanzania",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8901518050476157"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "average_precision_score(y_train, liver_pipeline.fit(X_train,y_train).predict_proba(X_train)[:,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "realistic-orientation",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(456, 10)"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "id": "narrative-lingerie",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.795761915661005\n",
      "0.8313306510076846\n"
     ]
    }
   ],
   "source": [
    "print(average_precision_score(y_train, grid_result_l.best_estimator_.predict_proba(X_train)[:,1]))\n",
    "print(average_precision_score(y_test, grid_result_l.best_estimator_.predict_proba(X_test)[:,1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "incorporate-pixel",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'memory': None, 'steps': [('model', AdaBoostClassifier(algorithm='SAMME',\n",
      "                   base_estimator=DecisionTreeClassifier(ccp_alpha=0.0101,\n",
      "                                                         criterion='entropy',\n",
      "                                                         max_depth=3,\n",
      "                                                         min_samples_leaf=14),\n",
      "                   learning_rate=0.01, n_estimators=5))], 'verbose': False, 'model': AdaBoostClassifier(algorithm='SAMME',\n",
      "                   base_estimator=DecisionTreeClassifier(ccp_alpha=0.0101,\n",
      "                                                         criterion='entropy',\n",
      "                                                         max_depth=3,\n",
      "                                                         min_samples_leaf=14),\n",
      "                   learning_rate=0.01, n_estimators=5), 'model__algorithm': 'SAMME', 'model__base_estimator__ccp_alpha': 0.0101, 'model__base_estimator__class_weight': None, 'model__base_estimator__criterion': 'entropy', 'model__base_estimator__max_depth': 3, 'model__base_estimator__max_features': None, 'model__base_estimator__max_leaf_nodes': None, 'model__base_estimator__min_impurity_decrease': 0.0, 'model__base_estimator__min_impurity_split': None, 'model__base_estimator__min_samples_leaf': 14, 'model__base_estimator__min_samples_split': 2, 'model__base_estimator__min_weight_fraction_leaf': 0.0, 'model__base_estimator__presort': 'deprecated', 'model__base_estimator__random_state': None, 'model__base_estimator__splitter': 'best', 'model__base_estimator': DecisionTreeClassifier(ccp_alpha=0.0101, criterion='entropy', max_depth=3,\n",
      "                       min_samples_leaf=14), 'model__learning_rate': 0.01, 'model__n_estimators': 5, 'model__random_state': None}\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.50      0.68      0.58       138\n",
      "           1       0.84      0.71      0.77       318\n",
      "\n",
      "    accuracy                           0.70       456\n",
      "   macro avg       0.67      0.69      0.67       456\n",
      "weighted avg       0.74      0.70      0.71       456\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.35      0.65      0.46        26\n",
      "           1       0.86      0.65      0.74        88\n",
      "\n",
      "    accuracy                           0.65       114\n",
      "   macro avg       0.61      0.65      0.60       114\n",
      "weighted avg       0.75      0.65      0.68       114\n",
      "\n"
     ]
    }
   ],
   "source": [
    "\n",
    "print(grid_result.best_estimator_.get_params())\n",
    "print(classification_report(y_train, grid_result.best_estimator_.predict(X_train) ))\n",
    "print(classification_report(y_test, grid_result.best_estimator_.predict(X_test) ))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "downtown-friendly",
   "metadata": {},
   "outputs": [],
   "source": [
    "base = AdaBoostClassifier(algorithm=\"SAMME\", n_estimators=1)\n",
    "\n",
    "mid = AdaBoostClassifier(\n",
    "    base_estimator= DecisionTreeClassifier(ccp_alpha=0.0031000000000000003,criterion='entropy',\n",
    "                                                                 max_depth=2,\n",
    "                                                                min_samples_split=10),\n",
    "                                                                learning_rate=0.01, \n",
    "                                                                n_estimators=10, algorithm=\"SAMME\")\n",
    "\n",
    "overfit = AdaBoostClassifier(  n_estimators=200, algorithm=\"SAMME\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "handmade-harmony",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.model_selection import ShuffleSplit\n",
    "from sklearn.model_selection import learning_curve\n",
    "\n",
    "fig, axes = plt.subplots(2, 3, figsize=(15, 10))\n",
    "\n",
    "#X, y = load_digits(return_X_y=True)\n",
    "\n",
    "\n",
    "title = \"Adaboost\"\n",
    "# Cross validation with 100 iterations to get smoother mean test and train\n",
    "# score curves, each time with 20% data randomly selected as a validation set.\n",
    "cv = ShuffleSplit(n_splits=10, test_size=0.1, random_state=14)\n",
    "\n",
    "# def plot_learning_curve(estimator1, estimator2, title1, title2, X, y, axes=None, ylim=None, cv=None,\n",
    "#                         n_jobs=None, train_sizes=np.linspace(.1, 1.0, 5)):\n",
    "\n",
    "# estimator =  AdaBoostClassifier\n",
    "plot_learning_curve(estimator=base,\n",
    "                    title=\"Underfit High Bias Adaboost\",\n",
    "                    X=X_train, \n",
    "                    y=y_train,\n",
    "                    axes=axes[:,0], \n",
    "                    ylim=(0.0, 1.01),\n",
    "                    cv=cv, \n",
    "                    n_jobs=4,\n",
    "                    param_range = np.array(range(1,11)),\n",
    "                    param_range_value = [DecisionTreeClassifier(max_depth=max_depth_ ,criterion=\"entropy\") for max_depth_ in range(1,11)],\n",
    "                    param_name = \"base_estimator\", \n",
    "                    validation_title=\"max_depth_trees\")\n",
    "        \n",
    "\n",
    "\n",
    "plot_learning_curve(estimator=mid,\n",
    "                    title=\"Mid Complexity Adaboost\",\n",
    "                    X=X_train, \n",
    "                    y=y_train,\n",
    "                    axes=axes[:,1], \n",
    "                    ylim=(0.0, 1.01),\n",
    "                    cv=cv, \n",
    "                    n_jobs=4,\n",
    "                    param_range =  np.array(range(1,100, 10)),\n",
    "                    param_range_value = np.array(range(1,100, 10)),\n",
    "                    param_name = \"n_estimators\", \n",
    "                    validation_title=\"Number of Weak Learners\")\n",
    "        \n",
    "plot_learning_curve(estimator=overfit,\n",
    "                    title=\"Overfit High Variance Adaboost \",\n",
    "                    X=X_train, \n",
    "                    y=y_train,\n",
    "                    axes=axes[:,2], \n",
    "                    ylim=(0.0, 1.01),\n",
    "                    cv=cv, \n",
    "                    n_jobs=4,\n",
    "                    param_range = np.linspace(.001, .5, 10),\n",
    "                    param_range_value =np.linspace(.01, 1.0, 10),\n",
    "                    param_name = \"learning_rate\", \n",
    "                    validation_title=\"learning rate\")\n",
    "        \n",
    "\n",
    "# title = r\"Learning Curves (SVM, RBF kernel, $\\gamma=0.001$)\"\n",
    "# # SVC is more expensive so we do a lower number of CV iterations:\n",
    "# cv = ShuffleSplit(n_splits=10, test_size=0.2, random_state=0)\n",
    "# estimator = SVC(gamma=0.001)\n",
    "# plot_learning_curve(estimator, title, X, y, axes=axes[:, 1], ylim=(0.7, 1.01),\n",
    "#                     cv=cv, n_jobs=4)\n",
    "\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"adaboost-comparison-bias-mid-overfit.pdf\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "close-graduation",
   "metadata": {},
   "source": [
    "> **(large bias)**A model that is too simple can lead to an underadjustment .\n",
    "\n",
    " >**(high variance)** If the model is too complex for the underlying training data, over-fitting\n",
    " may occur"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "increasing-reserve",
   "metadata": {},
   "source": [
    "## Complexity Curves"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "caroline-public",
   "metadata": {},
   "source": [
    "## Complexity / Validation Curve "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "arabic-detective",
   "metadata": {},
   "outputs": [],
   "source": [
    "plot_confusion_matrix?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "synthetic-firmware",
   "metadata": {},
   "outputs": [],
   "source": [
    "print( f\"Train Accuracy {round(sum(clf.predict(X_train)== y_train )/len(y_train),4)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "injured-monthly",
   "metadata": {},
   "outputs": [],
   "source": [
    "print( f\"Test Accuracy {round(sum(clf.predict(X_test)== y_test )/len(y_test),4)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "connected-organ",
   "metadata": {},
   "outputs": [],
   "source": [
    "ada.base_estimator.cost_complexity_pruning_path(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "defensive-surfing",
   "metadata": {},
   "outputs": [],
   "source": [
    "# show accuracy has high false positives, use grid search on avg precision"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "pharmaceutical-mother",
   "metadata": {},
   "source": [
    ">"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "varied-foster",
   "metadata": {},
   "source": [
    "> high bias more data doesnt help \n",
    "> High variance more data helps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "pediatric-zealand",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "my_learn_curve(title=\"Adaboost\", \n",
    "               estimator=census_pipeline, \n",
    "               cv=stratified_kfold,\n",
    "               X=X_c_train,\n",
    "               y=y_c_train,\n",
    "               label_suffix=\"No Tuning\", \n",
    "               color_=[\"orange\",\"blue\"],\n",
    "               shape = '^-'  )\n",
    "\n",
    "plt.savefig(\"ADABOOST_CENSUS_BASE_LEARNING_curve.pdf\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "everyday-heath",
   "metadata": {},
   "source": [
    "# complexity Census"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "weighted-tolerance",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "dried-formation",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Using pipeline\n",
    "census_steps = [('encoding', OneHotEncoder(cols=categorical_columns)),\n",
    "         ('imputer',SimpleImputer(missing_values=np.NaN, strategy='most_frequent')),\n",
    "         ('model',AdaBoostClassifier(algorithm=\"SAMME\", n_estimators=1, learning_rate=.1, base_estimator=DecisionTreeClassifier(max_depth=10)))]\n",
    "census_pipeline = Pipeline(steps=census_steps)\n",
    "# cv = ShuffleSplit(n_splits=10, test_size=0.1, random_state=0)\n",
    "cv = RepeatedStratifiedKFold(n_splits=5, n_repeats=1, random_state=45)\n",
    "train_scores, valid_scores = validation_curve(census_pipeline,\n",
    "                                              X_c_train,\n",
    "                                              y_c_train, \n",
    "                                              param_name = \"model__n_estimators\",\n",
    "                                            param_range =np.array(range(1,2002, 500)),\n",
    "                                            cv=cv,\n",
    "                                             n_jobs=-1)\n",
    "\n",
    "param_range = np.array(range(1,2002, 500))\n",
    "\n",
    "train_scores_mean =  np.mean(train_scores, axis=1)\n",
    "train_scores_std = np.std(train_scores, axis=1)\n",
    "test_scores_mean = np.mean(valid_scores, axis=1)\n",
    "test_scores_std = np.std(valid_scores, axis=1)\n",
    "\n",
    "plt.title(\"Validation Curve with Adaboost\")\n",
    "plt.xlabel(\"Number of Estimators\")\n",
    "plt.ylabel(\"Score\")\n",
    "plt.ylim(0.0, 1.1)\n",
    "lw = 2\n",
    "# semilogx\n",
    "plt.plot(param_range, train_scores_mean, label=\"Training score\",\n",
    "             color=\"darkorange\", lw=lw)\n",
    "plt.fill_between(param_range, train_scores_mean - train_scores_std,\n",
    "                 train_scores_mean + train_scores_std, alpha=0.2,\n",
    "                 color=\"darkorange\", lw=lw)\n",
    "plt.plot(param_range, test_scores_mean, label=\"Cross-validation score\",\n",
    "             color=\"navy\", lw=lw)\n",
    "plt.fill_between(param_range, test_scores_mean - test_scores_std,\n",
    "                 test_scores_mean + test_scores_std, alpha=0.2,\n",
    "                 color=\"navy\", lw=lw)\n",
    "plt.legend(loc=\"best\")\n",
    "\n",
    "\n",
    "plt.savefig(\"ADA_census_num_estimators.pdf\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "cheap-choice",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'memory': None,\n",
       " 'steps': [('encoding',\n",
       "   OneHotEncoder(cols=Index(['workclass', 'education', 'marital.status', 'occupation',\n",
       "          'relationship', 'race', 'sex', 'native.country'],\n",
       "         dtype='object'))),\n",
       "  ('imputer', SimpleImputer(strategy='most_frequent')),\n",
       "  ('model',\n",
       "   AdaBoostClassifier(algorithm='SAMME', learning_rate=0.1, n_estimators=1))],\n",
       " 'verbose': False,\n",
       " 'encoding': OneHotEncoder(cols=Index(['workclass', 'education', 'marital.status', 'occupation',\n",
       "        'relationship', 'race', 'sex', 'native.country'],\n",
       "       dtype='object')),\n",
       " 'imputer': SimpleImputer(strategy='most_frequent'),\n",
       " 'model': AdaBoostClassifier(algorithm='SAMME', learning_rate=0.1, n_estimators=1),\n",
       " 'encoding__cols': Index(['workclass', 'education', 'marital.status', 'occupation',\n",
       "        'relationship', 'race', 'sex', 'native.country'],\n",
       "       dtype='object'),\n",
       " 'encoding__drop_invariant': False,\n",
       " 'encoding__handle_missing': 'value',\n",
       " 'encoding__handle_unknown': 'value',\n",
       " 'encoding__return_df': True,\n",
       " 'encoding__use_cat_names': False,\n",
       " 'encoding__verbose': 0,\n",
       " 'imputer__add_indicator': False,\n",
       " 'imputer__copy': True,\n",
       " 'imputer__fill_value': None,\n",
       " 'imputer__missing_values': nan,\n",
       " 'imputer__strategy': 'most_frequent',\n",
       " 'imputer__verbose': 0,\n",
       " 'model__algorithm': 'SAMME',\n",
       " 'model__base_estimator': None,\n",
       " 'model__learning_rate': 0.1,\n",
       " 'model__n_estimators': 1,\n",
       " 'model__random_state': None}"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "census_pipeline.get_params()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "daily-bearing",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best: 0.814455 usin g {'model__base_estimator': DecisionTreeClassifier(max_depth=9, min_samples_leaf=22), 'model__learning_rate': 0.1, 'model__n_estimators': 49}\n"
     ]
    }
   ],
   "source": [
    "     \n",
    "# example of grid searching key hyperparameters for adaboost on a classification dataset\n",
    "from sklearn.datasets import make_classification\n",
    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "# define dataset\n",
    "# define the model with default hyperparameters\n",
    "\n",
    "model =  census_pipeline\n",
    "# define the grid of values to search\n",
    "grid = dict()\n",
    "grid['model__n_estimators'] = [i for i in range(1,50,2)]\n",
    "grid['model__learning_rate'] = [0.01, 0.1]\n",
    "grid['model__base_estimator'] = [DecisionTreeClassifier(max_depth=max_depth_,\n",
    "                                                 criterion=crit,\n",
    "                                                 min_samples_leaf = min_samples_split_)  \n",
    "                          for max_depth_ in range(2,10) \n",
    "                          for crit in [\"gini\"]\n",
    "                          for min_samples_split_ in range(6,40,4)\n",
    "                           ]\n",
    "\n",
    "# define the evaluation procedure\n",
    "cv = RepeatedStratifiedKFold(n_splits=5, n_repeats=1, random_state=45)\n",
    "# define the grid search procedure\n",
    "grid_search = GridSearchCV(estimator=census_pipeline, param_grid=grid, n_jobs=-1, cv=cv, scoring='average_precision')\n",
    "# execute the grid search\n",
    "grid_result = grid_search.fit(X_c_train, y_c_train)\n",
    "# summarize the best score and configuration\n",
    "print(\"Best: %f usin g %s\" % (grid_result.best_score_, grid_result.best_params_))\n",
    "# summarize all scores that were evaluated\n",
    "means = grid_result.cv_results_['mean_test_score']\n",
    "stds = grid_result.cv_results_['std_test_score']\n",
    "params = grid_result.cv_results_['params']\n",
    "# for mean, stdev, param in zip(means, stds, params):\n",
    "#     print(\"%f (%f) with: %r\" % (mean, stdev, param))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "baking-anger",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Best: 0.814455 usin g {'model__base_estimator': \n",
    "#DecisionTreeClassifier(max_depth=9, min_samples_leaf=22), 'model__learning_rate': 0.1, 'model__n_estimators': 49}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "lightweight-square",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "my_learn_curve(title=\"Adaboost\", \n",
    "               estimator=grid_result.best_estimator_, \n",
    "               cv=stratified_kfold,\n",
    "               X=X_c_train,\n",
    "               y=y_c_train,\n",
    "               label_suffix=\"After Tuning\", \n",
    "               color_=[\"orange\",\"blue\"],\n",
    "               shape = '^-'  )\n",
    "\n",
    "plt.savefig(\"ADABOOST_CENSUS_final_LEARNING_curve.pdf\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "deadly-uncle",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5967312531627624"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# untuned precision training\n",
    "average_precision_score(y_c_train, census_pipeline.fit(X_c_train,y_c_train).predict_proba(X_c_train)[:,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "incomplete-galaxy",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8407571994770964\n",
      "0.8096953202493742\n"
     ]
    }
   ],
   "source": [
    "print(average_precision_score(y_c_train, grid_result.predict_proba(X_c_train)[:,1]))\n",
    "print(average_precision_score(y_c_test, grid_result.predict_proba(X_c_test)[:,1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "interpreted-calibration",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.89      0.95      0.92     19758\n",
      "           1       0.79      0.64      0.71      6271\n",
      "\n",
      "    accuracy                           0.87     26029\n",
      "   macro avg       0.84      0.80      0.82     26029\n",
      "weighted avg       0.87      0.87      0.87     26029\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "          -1       0.89      0.94      0.92      4940\n",
      "           1       0.78      0.63      0.70      1568\n",
      "\n",
      "    accuracy                           0.87      6508\n",
      "   macro avg       0.83      0.79      0.81      6508\n",
      "weighted avg       0.86      0.87      0.86      6508\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_c_train, grid_result.best_estimator_.predict(X_c_train) ))\n",
    "print(classification_report(y_c_test, grid_result.best_estimator_.predict(X_c_test) ))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "foster-anthropology",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7f0bbc8c0340>"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(estimator=  census_pipeline.fit(X_c_train,y_c_train) ,\n",
    "                      X=X_c_train,\n",
    "                      y_true= y_c_train  , \n",
    "                      cmap=plt.cm.Blues,\n",
    "                       normalize=\"true\",\n",
    "                      display_labels = [\"Not Patient\" ,\"Patient\"]\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": "3.9.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}