dz6.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train = pd.read_csv('train.csv')\n",
    "test = pd.read_csv('test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train = train.Survived\n",
    "train.drop('Survived', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True], dtype=bool)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.columns == test.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train['is_test'] = 0\n",
    "test['is_test'] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pd.concat([train, test])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df[\"isMale\"] = df.Sex.replace({\"male\": 1, \"female\":0})\n",
    "df.drop([\"Sex\", \"Cabin\", \"Ticket\", \"Name\", \"PassengerId\"], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_dummies = pd.get_dummies(df, columns=['Pclass', 'Embarked'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = df_dummies[df_dummies.is_test==0].drop('is_test', axis=1)\n",
    "X_test = df_dummies[df_dummies.is_test==1].drop('is_test', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "columns = X_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import Imputer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "imputer = Imputer(missing_values='NaN', strategy='mean', axis=0, verbose=0, copy=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Imputer(axis=0, copy=True, missing_values='NaN', strategy='mean', verbose=0)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputer.fit(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_train_imputed = imputer.transform(X_train)\n",
    "X_train_imputed = pd.DataFrame(X_train_imputed, columns=columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "scaler = StandardScaler()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "StandardScaler(copy=True, with_mean=True, with_std=True)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaler.fit(X_train_imputed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_train_imputed_scaled = scaler.transform(X_train_imputed)\n",
    "X_train_imputed_scaled = pd.DataFrame(X_train_imputed_scaled, columns=columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test_imputed_scaled = scaler.transform(imputer.transform(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "X_train_fin, X_val, y_train_fin, y_val = train_test_split(X_train_imputed_scaled, y_train, test_size=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.tree import DecisionTreeClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "clf = DecisionTreeClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "depths = np.arange(1,10)\n",
    "#features_num = np.arange(5,15)\n",
    "grid = {'max_depth': depths}#, 'max_features': features_num}\n",
    "gridsearch = GridSearchCV(DecisionTreeClassifier(), grid, scoring='accuracy', cv=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 153 ms\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise',\n",
       "       estimator=DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=None,\n",
       "            max_features=None, max_leaf_nodes=None,\n",
       "            min_impurity_split=1e-07, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            presort=False, random_state=None, splitter='best'),\n",
       "       fit_params={}, iid=True, n_jobs=1,\n",
       "       param_grid={'max_depth': array([1, 2, 3, 4, 5, 6, 7, 8, 9])},\n",
       "       pre_dispatch='2*n_jobs', refit=True, return_train_score=True,\n",
       "       scoring='accuracy', verbose=0)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "gridsearch.fit(X_train_fin, y_train_fin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\adept\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:667: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[mean: 0.82444, std: 0.02699, params: {'max_depth': 9},\n",
       " mean: 0.81461, std: 0.01382, params: {'max_depth': 5},\n",
       " mean: 0.80758, std: 0.01408, params: {'max_depth': 6},\n",
       " mean: 0.80758, std: 0.03681, params: {'max_depth': 7},\n",
       " mean: 0.80618, std: 0.03527, params: {'max_depth': 8},\n",
       " mean: 0.79916, std: 0.01473, params: {'max_depth': 4},\n",
       " mean: 0.79635, std: 0.02025, params: {'max_depth': 3},\n",
       " mean: 0.78371, std: 0.02075, params: {'max_depth': 1},\n",
       " mean: 0.78090, std: 0.02054, params: {'max_depth': 2}]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(gridsearch.grid_scores_, key = lambda x: -x.mean_validation_score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\adept\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:667: DeprecationWarning: The grid_scores_ attribute was deprecated in version 0.18 in favor of the more elaborate cv_results_ attribute. The grid_scores_ attribute will not be available from 0.20\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x22f350c9f60>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe0AAAFJCAYAAAC2OXUDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n2iIiIi5CpS0iIuIiVNoiIiIu4v8HsV/3ADRf8pEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x22f351d8940>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "scores = [-x.mean_validation_score for x in gridsearch.grid_scores_]\n",
    "plt.plot(depths, scores)\n",
    "plt.scatter(depths, scores)\n",
    "best_point = np.argmin(scores)\n",
    "plt.scatter(depths[best_point], scores[best_point], c='g', s=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "clf_final = DecisionTreeClassifier(max_depth=9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=9,\n",
       "            max_features=None, max_leaf_nodes=None,\n",
       "            min_impurity_split=1e-07, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            presort=False, random_state=None, splitter='best')"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_final.fit(X_train_imputed_scaled, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 1, 1, 1,\n",
       "       1, 1, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 1, 0,\n",
       "       0, 0, 1, 1, 1, 0, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 1, 0, 0,\n",
       "       1, 1, 0, 0, 0, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0,\n",
       "       0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1,\n",
       "       0, 0, 1, 0, 1, 1, 0, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 1, 0, 0, 0, 1,\n",
       "       0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0,\n",
       "       1, 0, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 1, 0, 0, 1, 1, 0, 1, 0, 0, 0,\n",
       "       0, 1, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 0, 1, 0, 0,\n",
       "       0, 1, 0, 0, 0, 0, 1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0,\n",
       "       0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1,\n",
       "       0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0,\n",
       "       0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0,\n",
       "       0, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0,\n",
       "       1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 1, 1, 0, 1, 1, 0, 1,\n",
       "       1, 0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 1,\n",
       "       1, 1, 0, 0, 1, 0, 1, 0, 0, 1, 0, 1, 1, 0, 0, 0, 0, 1, 0, 1, 1, 0, 0,\n",
       "       1, 0, 0, 0], dtype=int64)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions = clf_final.predict(X_test_imputed_scaled)\n",
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "submussion = 'PassengerId,Survived\\n'\n",
    "submussion += \"\\n\".join([\"{},{}\".format(pid, prediction) for pid, prediction in zip(test.PassengerId, predictions)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "with open('submission.txt', 'w') as file:\n",
    "    file.write(submussion)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import export_graphviz\n",
    "\n",
    "def get_tree_dot_view(clf, feature_names=None, class_names=None):\n",
    "    print(export_graphviz(clf, out_file=None, filled=True, feature_names=feature_names, class_names=class_names))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "digraph Tree {\n",
      "node [shape=box, style=\"filled\", color=\"black\"] ;\n",
      "0 [label=\"isMale <= -0.3089\\ngini = 0.473\\nsamples = 891\\nvalue = [549, 342]\", fillcolor=\"#e5813960\"] ;\n",
      "1 [label=\"Pclass_3 <= -0.1027\\ngini = 0.3828\\nsamples = 314\\nvalue = [81, 233]\", fillcolor=\"#399de5a6\"] ;\n",
      "0 -> 1 [labeldistance=2.5, labelangle=45, headlabel=\"True\"] ;\n",
      "2 [label=\"Age <= -2.0931\\ngini = 0.1003\\nsamples = 170\\nvalue = [9, 161]\", fillcolor=\"#399de5f1\"] ;\n",
      "1 -> 2 ;\n",
      "3 [label=\"Pclass_1 <= 0.601\\ngini = 0.5\\nsamples = 2\\nvalue = [1, 1]\", fillcolor=\"#e5813900\"] ;\n",
      "2 -> 3 ;\n",
      "4 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "3 -> 4 ;\n",
      "5 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [1, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "3 -> 5 ;\n",
      "6 [label=\"Fare <= -0.0674\\ngini = 0.0907\\nsamples = 168\\nvalue = [8, 160]\", fillcolor=\"#399de5f2\"] ;\n",
      "2 -> 6 ;\n",
      "7 [label=\"Fare <= -0.08\\ngini = 0.1823\\nsamples = 69\\nvalue = [7, 62]\", fillcolor=\"#399de5e2\"] ;\n",
      "6 -> 7 ;\n",
      "8 [label=\"Age <= 2.024\\ngini = 0.1609\\nsamples = 68\\nvalue = [6, 62]\", fillcolor=\"#399de5e6\"] ;\n",
      "7 -> 8 ;\n",
      "9 [label=\"SibSp <= -0.0209\\ngini = 0.14\\nsamples = 66\\nvalue = [5, 61]\", fillcolor=\"#399de5ea\"] ;\n",
      "8 -> 9 ;\n",
      "10 [label=\"Fare <= -0.3816\\ngini = 0.0849\\nsamples = 45\\nvalue = [2, 43]\", fillcolor=\"#399de5f3\"] ;\n",
      "9 -> 10 ;\n",
      "11 [label=\"Fare <= -0.3902\\ngini = 0.1472\\nsamples = 25\\nvalue = [2, 23]\", fillcolor=\"#399de5e9\"] ;\n",
      "10 -> 11 ;\n",
      "12 [label=\"gini = 0.0\\nsamples = 11\\nvalue = [0, 11]\", fillcolor=\"#399de5ff\"] ;\n",
      "11 -> 12 ;\n",
      "13 [label=\"gini = 0.2449\\nsamples = 14\\nvalue = [2, 12]\", fillcolor=\"#399de5d4\"] ;\n",
      "11 -> 13 ;\n",
      "14 [label=\"gini = 0.0\\nsamples = 20\\nvalue = [0, 20]\", fillcolor=\"#399de5ff\"] ;\n",
      "10 -> 14 ;\n",
      "15 [label=\"Age <= -0.3616\\ngini = 0.2449\\nsamples = 21\\nvalue = [3, 18]\", fillcolor=\"#399de5d4\"] ;\n",
      "9 -> 15 ;\n",
      "16 [label=\"gini = 0.0\\nsamples = 6\\nvalue = [0, 6]\", fillcolor=\"#399de5ff\"] ;\n",
      "15 -> 16 ;\n",
      "17 [label=\"Age <= -0.1692\\ngini = 0.32\\nsamples = 15\\nvalue = [3, 12]\", fillcolor=\"#399de5bf\"] ;\n",
      "15 -> 17 ;\n",
      "18 [label=\"gini = 0.4444\\nsamples = 3\\nvalue = [2, 1]\", fillcolor=\"#e581397f\"] ;\n",
      "17 -> 18 ;\n",
      "19 [label=\"gini = 0.1528\\nsamples = 12\\nvalue = [1, 11]\", fillcolor=\"#399de5e8\"] ;\n",
      "17 -> 19 ;\n",
      "20 [label=\"Age <= 2.1394\\ngini = 0.5\\nsamples = 2\\nvalue = [1, 1]\", fillcolor=\"#e5813900\"] ;\n",
      "8 -> 20 ;\n",
      "21 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [1, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "20 -> 21 ;\n",
      "22 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "20 -> 22 ;\n",
      "23 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [1, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "7 -> 23 ;\n",
      "24 [label=\"Parch <= 1.3883\\ngini = 0.02\\nsamples = 99\\nvalue = [1, 98]\", fillcolor=\"#399de5fc\"] ;\n",
      "6 -> 24 ;\n",
      "25 [label=\"gini = 0.0\\nsamples = 84\\nvalue = [0, 84]\", fillcolor=\"#399de5ff\"] ;\n",
      "24 -> 25 ;\n",
      "26 [label=\"Age <= -0.4001\\ngini = 0.1244\\nsamples = 15\\nvalue = [1, 14]\", fillcolor=\"#399de5ed\"] ;\n",
      "24 -> 26 ;\n",
      "27 [label=\"gini = 0.0\\nsamples = 10\\nvalue = [0, 10]\", fillcolor=\"#399de5ff\"] ;\n",
      "26 -> 27 ;\n",
      "28 [label=\"Age <= -0.1308\\ngini = 0.32\\nsamples = 5\\nvalue = [1, 4]\", fillcolor=\"#399de5bf\"] ;\n",
      "26 -> 28 ;\n",
      "29 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [1, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "28 -> 29 ;\n",
      "30 [label=\"gini = 0.0\\nsamples = 4\\nvalue = [0, 4]\", fillcolor=\"#399de5ff\"] ;\n",
      "28 -> 30 ;\n",
      "31 [label=\"Fare <= -0.1783\\ngini = 0.5\\nsamples = 144\\nvalue = [72, 72]\", fillcolor=\"#e5813900\"] ;\n",
      "1 -> 31 ;\n",
      "32 [label=\"Age <= 0.5234\\ngini = 0.4839\\nsamples = 117\\nvalue = [48, 69]\", fillcolor=\"#399de54e\"] ;\n",
      "31 -> 32 ;\n",
      "33 [label=\"Embarked_S <= -0.4977\\ngini = 0.4721\\nsamples = 110\\nvalue = [42, 68]\", fillcolor=\"#399de561\"] ;\n",
      "32 -> 33 ;\n",
      "34 [label=\"Age <= 0.0308\\ngini = 0.4058\\nsamples = 53\\nvalue = [15, 38]\", fillcolor=\"#399de59a\"] ;\n",
      "33 -> 34 ;\n",
      "35 [label=\"Fare <= -0.3389\\ngini = 0.3799\\nsamples = 51\\nvalue = [13, 38]\", fillcolor=\"#399de5a8\"] ;\n",
      "34 -> 35 ;\n",
      "36 [label=\"Fare <= -0.3678\\ngini = 0.4388\\nsamples = 40\\nvalue = [13, 27]\", fillcolor=\"#399de584\"] ;\n",
      "35 -> 36 ;\n",
      "37 [label=\"Parch <= 1.3883\\ngini = 0.3122\\nsamples = 31\\nvalue = [6, 25]\", fillcolor=\"#399de5c2\"] ;\n",
      "36 -> 37 ;\n",
      "38 [label=\"gini = 0.2778\\nsamples = 30\\nvalue = [5, 25]\", fillcolor=\"#399de5cc\"] ;\n",
      "37 -> 38 ;\n",
      "39 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [1, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "37 -> 39 ;\n",
      "40 [label=\"Age <= -0.0269\\ngini = 0.3457\\nsamples = 9\\nvalue = [7, 2]\", fillcolor=\"#e58139b6\"] ;\n",
      "36 -> 40 ;\n",
      "41 [label=\"gini = 0.4444\\nsamples = 6\\nvalue = [4, 2]\", fillcolor=\"#e581397f\"] ;\n",
      "40 -> 41 ;\n",
      "42 [label=\"gini = 0.0\\nsamples = 3\\nvalue = [3, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "40 -> 42 ;\n",
      "43 [label=\"gini = 0.0\\nsamples = 11\\nvalue = [0, 11]\", fillcolor=\"#399de5ff\"] ;\n",
      "35 -> 43 ;\n",
      "44 [label=\"gini = 0.0\\nsamples = 2\\nvalue = [2, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "34 -> 44 ;\n",
      "45 [label=\"Fare <= -0.4305\\ngini = 0.4986\\nsamples = 57\\nvalue = [27, 30]\", fillcolor=\"#399de519\"] ;\n",
      "33 -> 45 ;\n",
      "46 [label=\"Fare <= -0.4921\\ngini = 0.4885\\nsamples = 33\\nvalue = [19, 14]\", fillcolor=\"#e5813943\"] ;\n",
      "45 -> 46 ;\n",
      "47 [label=\"Age <= -0.2578\\ngini = 0.2778\\nsamples = 6\\nvalue = [1, 5]\", fillcolor=\"#399de5cc\"] ;\n",
      "46 -> 47 ;\n",
      "48 [label=\"gini = 0.0\\nsamples = 5\\nvalue = [0, 5]\", fillcolor=\"#399de5ff\"] ;\n",
      "47 -> 48 ;\n",
      "49 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [1, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "47 -> 49 ;\n",
      "50 [label=\"Age <= -0.7849\\ngini = 0.4444\\nsamples = 27\\nvalue = [18, 9]\", fillcolor=\"#e581397f\"] ;\n",
      "46 -> 50 ;\n",
      "51 [label=\"Age <= -1.0927\\ngini = 0.4898\\nsamples = 7\\nvalue = [3, 4]\", fillcolor=\"#399de540\"] ;\n",
      "50 -> 51 ;\n",
      "52 [label=\"gini = 0.0\\nsamples = 2\\nvalue = [2, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "51 -> 52 ;\n",
      "53 [label=\"gini = 0.32\\nsamples = 5\\nvalue = [1, 4]\", fillcolor=\"#399de5bf\"] ;\n",
      "51 -> 53 ;\n",
      "54 [label=\"Fare <= -0.4876\\ngini = 0.375\\nsamples = 20\\nvalue = [15, 5]\", fillcolor=\"#e58139aa\"] ;\n",
      "50 -> 54 ;\n",
      "55 [label=\"gini = 0.4938\\nsamples = 9\\nvalue = [5, 4]\", fillcolor=\"#e5813933\"] ;\n",
      "54 -> 55 ;\n",
      "56 [label=\"gini = 0.1653\\nsamples = 11\\nvalue = [10, 1]\", fillcolor=\"#e58139e6\"] ;\n",
      "54 -> 56 ;\n",
      "57 [label=\"Fare <= -0.2941\\ngini = 0.4444\\nsamples = 24\\nvalue = [8, 16]\", fillcolor=\"#399de57f\"] ;\n",
      "45 -> 57 ;\n",
      "58 [label=\"Fare <= -0.3779\\ngini = 0.2311\\nsamples = 15\\nvalue = [2, 13]\", fillcolor=\"#399de5d8\"] ;\n",
      "57 -> 58 ;\n",
      "59 [label=\"gini = 0.0\\nsamples = 6\\nvalue = [0, 6]\", fillcolor=\"#399de5ff\"] ;\n",
      "58 -> 59 ;\n",
      "60 [label=\"Fare <= -0.3439\\ngini = 0.3457\\nsamples = 9\\nvalue = [2, 7]\", fillcolor=\"#399de5b6\"] ;\n",
      "58 -> 60 ;\n",
      "61 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [1, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "60 -> 61 ;\n",
      "62 [label=\"gini = 0.2188\\nsamples = 8\\nvalue = [1, 7]\", fillcolor=\"#399de5db\"] ;\n",
      "60 -> 62 ;\n",
      "63 [label=\"Fare <= -0.2633\\ngini = 0.4444\\nsamples = 9\\nvalue = [6, 3]\", fillcolor=\"#e581397f\"] ;\n",
      "57 -> 63 ;\n",
      "64 [label=\"gini = 0.0\\nsamples = 3\\nvalue = [3, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "63 -> 64 ;\n",
      "65 [label=\"Age <= 0.0232\\ngini = 0.5\\nsamples = 6\\nvalue = [3, 3]\", fillcolor=\"#e5813900\"] ;\n",
      "63 -> 65 ;\n",
      "66 [label=\"gini = 0.375\\nsamples = 4\\nvalue = [3, 1]\", fillcolor=\"#e58139aa\"] ;\n",
      "65 -> 66 ;\n",
      "67 [label=\"gini = 0.0\\nsamples = 2\\nvalue = [0, 2]\", fillcolor=\"#399de5ff\"] ;\n",
      "65 -> 67 ;\n",
      "68 [label=\"Age <= 1.947\\ngini = 0.2449\\nsamples = 7\\nvalue = [6, 1]\", fillcolor=\"#e58139d4\"] ;\n",
      "32 -> 68 ;\n",
      "69 [label=\"gini = 0.0\\nsamples = 6\\nvalue = [6, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "68 -> 69 ;\n",
      "70 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "68 -> 70 ;\n",
      "71 [label=\"Parch <= 0.147\\ngini = 0.1975\\nsamples = 27\\nvalue = [24, 3]\", fillcolor=\"#e58139df\"] ;\n",
      "31 -> 71 ;\n",
      "72 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "71 -> 72 ;\n",
      "73 [label=\"Age <= -1.8622\\ngini = 0.142\\nsamples = 26\\nvalue = [24, 2]\", fillcolor=\"#e58139ea\"] ;\n",
      "71 -> 73 ;\n",
      "74 [label=\"Age <= -2.0161\\ngini = 0.4444\\nsamples = 3\\nvalue = [2, 1]\", fillcolor=\"#e581397f\"] ;\n",
      "73 -> 74 ;\n",
      "75 [label=\"gini = 0.0\\nsamples = 2\\nvalue = [2, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "74 -> 75 ;\n",
      "76 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "74 -> 76 ;\n",
      "77 [label=\"Parch <= 5.1122\\ngini = 0.0832\\nsamples = 23\\nvalue = [22, 1]\", fillcolor=\"#e58139f3\"] ;\n",
      "73 -> 77 ;\n",
      "78 [label=\"gini = 0.0\\nsamples = 18\\nvalue = [18, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "77 -> 78 ;\n",
      "79 [label=\"Age <= 0.6773\\ngini = 0.32\\nsamples = 5\\nvalue = [4, 1]\", fillcolor=\"#e58139bf\"] ;\n",
      "77 -> 79 ;\n",
      "80 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "79 -> 80 ;\n",
      "81 [label=\"gini = 0.0\\nsamples = 4\\nvalue = [4, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "79 -> 81 ;\n",
      "82 [label=\"Age <= -1.7853\\ngini = 0.3064\\nsamples = 577\\nvalue = [468, 109]\", fillcolor=\"#e58139c4\"] ;\n",
      "0 -> 82 [labeldistance=2.5, labelangle=-45, headlabel=\"False\"] ;\n",
      "83 [label=\"SibSp <= 1.7938\\ngini = 0.4444\\nsamples = 24\\nvalue = [8, 16]\", fillcolor=\"#399de57f\"] ;\n",
      "82 -> 83 ;\n",
      "84 [label=\"gini = 0.0\\nsamples = 15\\nvalue = [0, 15]\", fillcolor=\"#399de5ff\"] ;\n",
      "83 -> 84 ;\n",
      "85 [label=\"Age <= -2.0931\\ngini = 0.1975\\nsamples = 9\\nvalue = [8, 1]\", fillcolor=\"#e58139df\"] ;\n",
      "83 -> 85 ;\n",
      "86 [label=\"gini = 0.0\\nsamples = 5\\nvalue = [5, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "85 -> 86 ;\n",
      "87 [label=\"Fare <= -0.0176\\ngini = 0.375\\nsamples = 4\\nvalue = [3, 1]\", fillcolor=\"#e58139aa\"] ;\n",
      "85 -> 87 ;\n",
      "88 [label=\"gini = 0.0\\nsamples = 3\\nvalue = [3, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "87 -> 88 ;\n",
      "89 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "87 -> 89 ;\n",
      "90 [label=\"Pclass_1 <= 0.601\\ngini = 0.2798\\nsamples = 553\\nvalue = [460, 93]\", fillcolor=\"#e58139cb\"] ;\n",
      "82 -> 90 ;\n",
      "91 [label=\"Fare <= 0.3925\\ngini = 0.2043\\nsamples = 433\\nvalue = [383, 50]\", fillcolor=\"#e58139de\"] ;\n",
      "90 -> 91 ;\n",
      "92 [label=\"Age <= -1.2851\\ngini = 0.1925\\nsamples = 417\\nvalue = [372, 45]\", fillcolor=\"#e58139e0\"] ;\n",
      "91 -> 92 ;\n",
      "93 [label=\"SibSp <= 1.3401\\ngini = 0.4444\\nsamples = 12\\nvalue = [8, 4]\", fillcolor=\"#e581397f\"] ;\n",
      "92 -> 93 ;\n",
      "94 [label=\"Embarked_S <= -0.4977\\ngini = 0.32\\nsamples = 5\\nvalue = [1, 4]\", fillcolor=\"#399de5bf\"] ;\n",
      "93 -> 94 ;\n",
      "95 [label=\"Age <= -1.4005\\ngini = 0.5\\nsamples = 2\\nvalue = [1, 1]\", fillcolor=\"#e5813900\"] ;\n",
      "94 -> 95 ;\n",
      "96 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [1, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "95 -> 96 ;\n",
      "97 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "95 -> 97 ;\n",
      "98 [label=\"gini = 0.0\\nsamples = 3\\nvalue = [0, 3]\", fillcolor=\"#399de5ff\"] ;\n",
      "94 -> 98 ;\n",
      "99 [label=\"gini = 0.0\\nsamples = 7\\nvalue = [7, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "93 -> 99 ;\n",
      "100 [label=\"Embarked_C <= 0.7962\\ngini = 0.182\\nsamples = 405\\nvalue = [364, 41]\", fillcolor=\"#e58139e2\"] ;\n",
      "92 -> 100 ;\n",
      "101 [label=\"Fare <= -0.3816\\ngini = 0.1636\\nsamples = 356\\nvalue = [324, 32]\", fillcolor=\"#e58139e6\"] ;\n",
      "100 -> 101 ;\n",
      "102 [label=\"Fare <= -0.4891\\ngini = 0.1884\\nsamples = 285\\nvalue = [255, 30]\", fillcolor=\"#e58139e1\"] ;\n",
      "101 -> 102 ;\n",
      "103 [label=\"Fare <= -0.4901\\ngini = 0.1292\\nsamples = 144\\nvalue = [134, 10]\", fillcolor=\"#e58139ec\"] ;\n",
      "102 -> 103 ;\n",
      "104 [label=\"gini = 0.1653\\nsamples = 110\\nvalue = [100, 10]\", fillcolor=\"#e58139e6\"] ;\n",
      "103 -> 104 ;\n",
      "105 [label=\"gini = 0.0\\nsamples = 34\\nvalue = [34, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "103 -> 105 ;\n",
      "106 [label=\"Fare <= -0.4876\\ngini = 0.2434\\nsamples = 141\\nvalue = [121, 20]\", fillcolor=\"#e58139d5\"] ;\n",
      "102 -> 106 ;\n",
      "107 [label=\"gini = 0.4734\\nsamples = 13\\nvalue = [8, 5]\", fillcolor=\"#e5813960\"] ;\n",
      "106 -> 107 ;\n",
      "108 [label=\"gini = 0.2069\\nsamples = 128\\nvalue = [113, 15]\", fillcolor=\"#e58139dd\"] ;\n",
      "106 -> 108 ;\n",
      "109 [label=\"Embarked_S <= -0.4977\\ngini = 0.0548\\nsamples = 71\\nvalue = [69, 2]\", fillcolor=\"#e58139f8\"] ;\n",
      "101 -> 109 ;\n",
      "110 [label=\"SibSp <= 0.8865\\ngini = 0.2449\\nsamples = 7\\nvalue = [6, 1]\", fillcolor=\"#e58139d4\"] ;\n",
      "109 -> 110 ;\n",
      "111 [label=\"gini = 0.0\\nsamples = 6\\nvalue = [6, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "110 -> 111 ;\n",
      "112 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "110 -> 112 ;\n",
      "113 [label=\"Fare <= -0.1303\\ngini = 0.0308\\nsamples = 64\\nvalue = [63, 1]\", fillcolor=\"#e58139fb\"] ;\n",
      "109 -> 113 ;\n",
      "114 [label=\"gini = 0.0\\nsamples = 38\\nvalue = [38, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "113 -> 114 ;\n",
      "115 [label=\"gini = 0.074\\nsamples = 26\\nvalue = [25, 1]\", fillcolor=\"#e58139f5\"] ;\n",
      "113 -> 115 ;\n",
      "116 [label=\"Fare <= -0.3434\\ngini = 0.2999\\nsamples = 49\\nvalue = [40, 9]\", fillcolor=\"#e58139c6\"] ;\n",
      "100 -> 116 ;\n",
      "117 [label=\"Age <= -0.554\\ngini = 0.2235\\nsamples = 39\\nvalue = [34, 5]\", fillcolor=\"#e58139da\"] ;\n",
      "116 -> 117 ;\n",
      "118 [label=\"Age <= -0.8618\\ngini = 0.4444\\nsamples = 6\\nvalue = [4, 2]\", fillcolor=\"#e581397f\"] ;\n",
      "117 -> 118 ;\n",
      "119 [label=\"gini = 0.0\\nsamples = 2\\nvalue = [2, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "118 -> 119 ;\n",
      "120 [label=\"gini = 0.5\\nsamples = 4\\nvalue = [2, 2]\", fillcolor=\"#e5813900\"] ;\n",
      "118 -> 120 ;\n",
      "121 [label=\"Fare <= -0.5029\\ngini = 0.1653\\nsamples = 33\\nvalue = [30, 3]\", fillcolor=\"#e58139e6\"] ;\n",
      "117 -> 121 ;\n",
      "122 [label=\"gini = 0.0\\nsamples = 10\\nvalue = [10, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "121 -> 122 ;\n",
      "123 [label=\"gini = 0.2268\\nsamples = 23\\nvalue = [20, 3]\", fillcolor=\"#e58139d9\"] ;\n",
      "121 -> 123 ;\n",
      "124 [label=\"Fare <= -0.241\\ngini = 0.48\\nsamples = 10\\nvalue = [6, 4]\", fillcolor=\"#e5813955\"] ;\n",
      "116 -> 124 ;\n",
      "125 [label=\"gini = 0.0\\nsamples = 4\\nvalue = [0, 4]\", fillcolor=\"#399de5ff\"] ;\n",
      "124 -> 125 ;\n",
      "126 [label=\"gini = 0.0\\nsamples = 6\\nvalue = [6, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "124 -> 126 ;\n",
      "127 [label=\"Fare <= 0.6205\\ngini = 0.4297\\nsamples = 16\\nvalue = [11, 5]\", fillcolor=\"#e581398b\"] ;\n",
      "91 -> 127 ;\n",
      "128 [label=\"Age <= 0.0885\\ngini = 0.4082\\nsamples = 7\\nvalue = [2, 5]\", fillcolor=\"#399de599\"] ;\n",
      "127 -> 128 ;\n",
      "129 [label=\"Age <= -0.2077\\ngini = 0.48\\nsamples = 5\\nvalue = [2, 3]\", fillcolor=\"#399de555\"] ;\n",
      "128 -> 129 ;\n",
      "130 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "129 -> 130 ;\n",
      "131 [label=\"Age <= -0.0654\\ngini = 0.5\\nsamples = 4\\nvalue = [2, 2]\", fillcolor=\"#e5813900\"] ;\n",
      "129 -> 131 ;\n",
      "132 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [1, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "131 -> 132 ;\n",
      "133 [label=\"gini = 0.4444\\nsamples = 3\\nvalue = [1, 2]\", fillcolor=\"#399de57f\"] ;\n",
      "131 -> 133 ;\n",
      "134 [label=\"gini = 0.0\\nsamples = 2\\nvalue = [0, 2]\", fillcolor=\"#399de5ff\"] ;\n",
      "128 -> 134 ;\n",
      "135 [label=\"gini = 0.0\\nsamples = 9\\nvalue = [9, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "127 -> 135 ;\n",
      "136 [label=\"Fare <= -0.122\\ngini = 0.4599\\nsamples = 120\\nvalue = [77, 43]\", fillcolor=\"#e5813971\"] ;\n",
      "90 -> 136 ;\n",
      "137 [label=\"gini = 0.0\\nsamples = 10\\nvalue = [10, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "136 -> 137 ;\n",
      "138 [label=\"Age <= 1.7931\\ngini = 0.4762\\nsamples = 110\\nvalue = [67, 43]\", fillcolor=\"#e581395b\"] ;\n",
      "136 -> 138 ;\n",
      "139 [label=\"Fare <= -0.1021\\ngini = 0.4949\\nsamples = 89\\nvalue = [49, 40]\", fillcolor=\"#e581392f\"] ;\n",
      "138 -> 139 ;\n",
      "140 [label=\"Fare <= -0.1155\\ngini = 0.2604\\nsamples = 13\\nvalue = [2, 11]\", fillcolor=\"#399de5d1\"] ;\n",
      "139 -> 140 ;\n",
      "141 [label=\"gini = 0.0\\nsamples = 4\\nvalue = [0, 4]\", fillcolor=\"#399de5ff\"] ;\n",
      "140 -> 141 ;\n",
      "142 [label=\"Age <= 1.2929\\ngini = 0.3457\\nsamples = 9\\nvalue = [2, 7]\", fillcolor=\"#399de5b6\"] ;\n",
      "140 -> 142 ;\n",
      "143 [label=\"Age <= 0.7927\\ngini = 0.4082\\nsamples = 7\\nvalue = [2, 5]\", fillcolor=\"#399de599\"] ;\n",
      "142 -> 143 ;\n",
      "144 [label=\"gini = 0.32\\nsamples = 5\\nvalue = [1, 4]\", fillcolor=\"#399de5bf\"] ;\n",
      "143 -> 144 ;\n",
      "145 [label=\"gini = 0.5\\nsamples = 2\\nvalue = [1, 1]\", fillcolor=\"#e5813900\"] ;\n",
      "143 -> 145 ;\n",
      "146 [label=\"gini = 0.0\\nsamples = 2\\nvalue = [0, 2]\", fillcolor=\"#399de5ff\"] ;\n",
      "142 -> 146 ;\n",
      "147 [label=\"Fare <= 7.1571\\ngini = 0.472\\nsamples = 76\\nvalue = [47, 29]\", fillcolor=\"#e5813962\"] ;\n",
      "139 -> 147 ;\n",
      "148 [label=\"Fare <= 2.0625\\ngini = 0.4635\\nsamples = 74\\nvalue = [47, 27]\", fillcolor=\"#e581396d\"] ;\n",
      "147 -> 148 ;\n",
      "149 [label=\"Age <= -0.1692\\ngini = 0.4812\\nsamples = 67\\nvalue = [40, 27]\", fillcolor=\"#e5813953\"] ;\n",
      "148 -> 149 ;\n",
      "150 [label=\"gini = 0.426\\nsamples = 13\\nvalue = [4, 9]\", fillcolor=\"#399de58e\"] ;\n",
      "149 -> 150 ;\n",
      "151 [label=\"gini = 0.4444\\nsamples = 54\\nvalue = [36, 18]\", fillcolor=\"#e581397f\"] ;\n",
      "149 -> 151 ;\n",
      "152 [label=\"gini = 0.0\\nsamples = 7\\nvalue = [7, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "148 -> 152 ;\n",
      "153 [label=\"gini = 0.0\\nsamples = 2\\nvalue = [0, 2]\", fillcolor=\"#399de5ff\"] ;\n",
      "147 -> 153 ;\n",
      "154 [label=\"Age <= 3.5246\\ngini = 0.2449\\nsamples = 21\\nvalue = [18, 3]\", fillcolor=\"#e58139d4\"] ;\n",
      "138 -> 154 ;\n",
      "155 [label=\"Embarked_C <= 0.7962\\ngini = 0.18\\nsamples = 20\\nvalue = [18, 2]\", fillcolor=\"#e58139e3\"] ;\n",
      "154 -> 155 ;\n",
      "156 [label=\"gini = 0.0\\nsamples = 12\\nvalue = [12, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "155 -> 156 ;\n",
      "157 [label=\"SibSp <= -0.0209\\ngini = 0.375\\nsamples = 8\\nvalue = [6, 2]\", fillcolor=\"#e58139aa\"] ;\n",
      "155 -> 157 ;\n",
      "158 [label=\"Age <= 2.1009\\ngini = 0.2449\\nsamples = 7\\nvalue = [6, 1]\", fillcolor=\"#e58139d4\"] ;\n",
      "157 -> 158 ;\n",
      "159 [label=\"gini = 0.5\\nsamples = 2\\nvalue = [1, 1]\", fillcolor=\"#e5813900\"] ;\n",
      "158 -> 159 ;\n",
      "160 [label=\"gini = 0.0\\nsamples = 5\\nvalue = [5, 0]\", fillcolor=\"#e58139ff\"] ;\n",
      "158 -> 160 ;\n",
      "161 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "157 -> 161 ;\n",
      "162 [label=\"gini = 0.0\\nsamples = 1\\nvalue = [0, 1]\", fillcolor=\"#399de5ff\"] ;\n",
      "154 -> 162 ;\n",
      "}\n"
     ]
    }
   ],
   "source": [
    "get_tree_dot_view(clf_final, list(X_train_fin.columns))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}