{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Tutorial: Tree-based and ensemble models in Python (ANSWERS)\n",
    "\n",
    "This is not a standalone tutorial, as most of the theory and important questions are asked in the main tutorial in R. The second tutorial serves the purpose of helping you make your own tree-based and enseble models in Python.  I expect that you have already covered the video lectures and the main tutorial in R by now. So I assume that you already know what you are doing, you just don't know how to code it in Python yet. If R tutorial provides more assistance in digesting the material, this tutorial in Python pushes you to be more independent, with minimum hints. (I used these notebooks to prepare this tutorial: [1](https://github.com/ageron/handson-ml2/blob/master/06_decision_trees.ipynb), [2](https://github.com/ageron/handson-ml2/blob/master/07_ensemble_learning_and_random_forests.ipynb) [3](https://nbviewer.jupyter.org/github/JWarmenhoven/ISL-python/blob/master/Notebooks/Chapter%208.ipynb#8.3.4-Boosting))\n",
    "\n",
    "We keep on working with the same COMPAS dataset, and we will use `sklearn` library. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the usual libraries\n",
    "import sys\n",
    "import sklearn\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load data and split it in train and test sets\n",
    "Let's load the sample of COMPAS data we used in the previous tutorial. For your convenience, I already transformed all factor variables into dummy variables using `model.matrix()` function in R and saved it on github as a csv file. So you can directly access the data using the link below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {},
   "outputs": [
    {
     "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>Two_yr_Recidivismyes</th>\n",
       "      <th>Number_of_Priors</th>\n",
       "      <th>Age_Above_FourtyFiveyes</th>\n",
       "      <th>Age_Below_TwentyFiveyes</th>\n",
       "      <th>FemaleMale</th>\n",
       "      <th>Misdemeanoryes</th>\n",
       "      <th>ethnicityCaucasian</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.012530</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.400115</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0.701857</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>-0.930922</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>-0.545324</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Two_yr_Recidivismyes  Number_of_Priors  Age_Above_FourtyFiveyes  \\\n",
       "0                     1         -0.012530                        0   \n",
       "1                     1         -0.400115                        0   \n",
       "2                     1          0.701857                        1   \n",
       "3                     0         -0.930922                        0   \n",
       "4                     1         -0.545324                        0   \n",
       "\n",
       "   Age_Below_TwentyFiveyes  FemaleMale  Misdemeanoryes  ethnicityCaucasian  \n",
       "0                        1           0               1                   1  \n",
       "1                        1           1               1                   0  \n",
       "2                        0           1               1                   0  \n",
       "3                        0           1               0                   1  \n",
       "4                        1           1               1                   0  "
      ]
     },
     "execution_count": 190,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_location = \"https://raw.githubusercontent.com/madina-k/DSE2021_tutorials/main/tutorial_trees/data/compas_sample500matrix.csv\"\n",
    "data = pd.read_csv(data_location)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use `train_test_split` command from `sklearn.model_selection` to split the data into four objects: `X_train, X_test, y_train, y_test` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "\n",
    "y = data.pop('Two_yr_Recidivismyes').values\n",
    "X = data\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=1)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Decision Tree\n",
    "\n",
    "Train a decision tree using `DecisionTreeClassifier` class from `sklearn.tree`. Look at the documentation [here](https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html) and fit a decision tree using Gini node impurity measure, with minimum 25 observations in each leaf, and the maximum tree depth of 3. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=4,\n",
       "                       max_features=None, max_leaf_nodes=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=25, min_samples_split=2,\n",
       "                       min_weight_fraction_leaf=0.0, presort=False,\n",
       "                       random_state=None, splitter='best')"
      ]
     },
     "execution_count": 192,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "tree_clf = DecisionTreeClassifier(min_samples_leaf = 25, max_depth=4)\n",
    "tree_clf.fit(X,y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And visualize the resulting tree using `plot_tree` from `sklearn.tree`.  See documentation [here](https://scikit-learn.org/stable/modules/generated/sklearn.tree.plot_tree.html). Don't forget to tell the names of the features and class names, set the font size to 14."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x1296 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.tree import plot_tree\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.figure(figsize=(18,18))  # set plot size (denoted in inches)\n",
    "plot_tree(tree_clf, feature_names=X.columns, class_names=['No', 'Yes'],fontsize=14 )\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is some bushy tree but we do not have a way to prune it in an automated way with sklearn. If you really want to prune this tree, look how it can be done in this [blog](https://ranvir.xyz/blog/practical-approach-to-tree-pruning-using-sklearn/).\n",
    "\n",
    "Finally, let's check it's out-of-sample accuracy with `accuracy_score` function from `sklearn.metrics`. Read the documentation [here](https://scikit-learn.org/stable/modules/generated/sklearn.metrics.accuracy_score.html)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.728\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "y_pred_tree = tree_clf.predict(X_test)\n",
    "print(accuracy_score(y_test, y_pred_tree))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bagging\n",
    "\n",
    "Now it's time for some bagging! Use `BaggingClassifier` from `sklearn.ensemble`. See documentation [here](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.BaggingClassifier.html).\n",
    "Create a new tree object which allows to grow a tree of maximum depth of 2 and put that classifier inside the `BaggingClassifier`. Ask to bag over 500 trees with maximum samples of 100 observations with bootstrapping.  Print the accuracy score.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.72\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import BaggingClassifier\n",
    "\n",
    "tree_clf_forbagging = DecisionTreeClassifier(max_depth=2)\n",
    "\n",
    "bag_clf = BaggingClassifier(\n",
    "    tree_clf_forbagging, n_estimators=500,\n",
    "    max_samples=100, bootstrap=True)\n",
    "bag_clf.fit(X_train, y_train)\n",
    "y_pred_bag = bag_clf.predict(X_test)\n",
    "print(accuracy_score(y_test, y_pred_bag))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Forest\n",
    "\n",
    "Now, let's use `RandomForestClassifier` from `sklearn.ensemble`. See documentation [here](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html). Create a random forest of 500 trees, where each tree has a maximum depth of 2. Print the accuracy score for the test sample."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.696\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "rf_clf = RandomForestClassifier(n_estimators=500, max_depth=2)\n",
    "rf_clf.fit(X_train, y_train)\n",
    "\n",
    "y_pred_rf = rf_clf.predict(X_test)\n",
    "\n",
    "print(accuracy_score(y_test, y_pred_rf))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check for the importance of each feature, using `feature_importances_` property of the classifier. (For example, see this [page](https://scikit-learn.org/stable/auto_examples/ensemble/plot_forest_importances.html))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number_of_Priors 0.5951675288832771\n",
      "Age_Above_FourtyFiveyes 0.15327959016558776\n",
      "Age_Below_TwentyFiveyes 0.04963856624481278\n",
      "FemaleMale 0.06136793924146804\n",
      "Misdemeanoryes 0.07341448602922451\n",
      "ethnicityCaucasian 0.06713188943562985\n"
     ]
    }
   ],
   "source": [
    "for name, score in zip(list(data.columns), rnd_clf.feature_importances_):\n",
    "    print(name,score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Boosting with Gradient Boosting\n",
    "\n",
    "There are several different ways to boost trees (e.g., XGBoost library or AdaBoostClaasifier from sklearn). In this tutorial, I would ask you to train a `GradientBoostingClassifier` from `sklearn.ensemble`. See the documentation [here](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingClassifier.html). Make 300 trees with the maximum depth of 2. Set learning rate at 0.01. \n",
    "\n",
    "Then use the `staged_predict` method to predict the target variable at each stage of boosting (1 tree only, 2 trees together, 3 trees, and so on) using the test X. Get the mean squared error using the `mean_squared_error` wrapper from `sklearn.ensemble`. See the documentation [here](https://scikit-learn.org/stable/modules/generated/sklearn.metrics.mean_squared_error.html). And find the number of trees that minimize the MSE on the test sample. \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.304\n",
      "26\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "\n",
    "gbc_clf = GradientBoostingClassifier(max_depth=2, n_estimators=300, learning_rate=0.01)\n",
    "gbc_clf.fit(X_train, y_train)\n",
    "\n",
    "errors = [mean_squared_error(y_test, y_pred)\n",
    "          for y_pred in gbc_clf.staged_predict(X_test)]\n",
    "\n",
    "min_error = np.min(errors)\n",
    "print(min_error)\n",
    "\n",
    "\n",
    "optimal_n_trees = np.argmin(errors) + 1\n",
    "print(optimal_n_trees)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "Train the Gradient boosting classifier model with the **optimal number of trees** and and get the accuracy score.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.696\n"
     ]
    }
   ],
   "source": [
    "gbc_best = GradientBoostingClassifier(max_depth=2, n_estimators=optimal_n_trees, learning_rate=0.01)\n",
    "gbc_best.fit(X_train, y_train)\n",
    "\n",
    "\n",
    "\n",
    "y_pred_gbcbest = gbc_best.predict(X_test)\n",
    "print(accuracy_score(y_test, y_pred_gbrtbest))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Voting classifier\n",
    "\n",
    "Finally we will try to combine all the other models under one Ensemble model. \n",
    "\n",
    "Import `VotingClassifier` from `sklearn.ensemble`. \n",
    "\n",
    "We will also import Logistic Regression and Support Vector Machine."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import VotingClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "log_clf = LogisticRegression(solver='lbfgs')\n",
    "svm_clf = SVC(probability=True, gamma='scale')\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now put all different classifiers to be combined within the voting classifier and compare accuracy scores across different classifiers:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTreeClassifier 0.624\n",
      "LogisticRegression 0.656\n",
      "RandomForestClassifier 0.696\n",
      "SVC 0.688\n",
      "BaggingClassifier 0.728\n",
      "GradientBoostingClassifier 0.696\n",
      "VotingClassifier 0.72\n"
     ]
    }
   ],
   "source": [
    "voting_clf = VotingClassifier(\n",
    "    estimators = [('lr', log_clf), # Logistic classifier\n",
    "                  ('rf', rf_clf),  # Random forest classifier\n",
    "                  ('svm', svm_clf), # SVM classifier\n",
    "                  ('bag', bag_clf), # Bagging classifier\n",
    "                  ('gbrt', gbc_best)],  # Boosting classifier\n",
    "    voting ='hard')\n",
    "\n",
    "for clf in (tree_clf,log_clf, rf_clf, svm_clf, bag_clf, gbc_best, voting_clf):\n",
    "    clf.fit(X_train, y_train)\n",
    "    y_pred = clf.predict(X_test)\n",
    "    print(clf.__class__.__name__, accuracy_score(y_test, y_pred))"
   ]
  }
 ],
 "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}