kaggle-titanic.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# kaggle-titanic\n",
    "kernel by Rasmus N\n",
    "\n",
    "* goal 1: start learning scikit-learn, pandas and numpy \n",
    "* goal 2: score ~ 80 (accuracy)\n",
    "* goal 3: have fun :)\n",
    "\n",
    "Inspiration:\n",
    "    https://www.kaggle.com/erikbruin/titanic-2nd-degree-families-and-majority-voting\n",
    "    https://www.kaggle.com/yassineghouzam/titanic-top-4-with-ensemble-modeling/notebook\n",
    "    https://ahmedbesbes.com/how-to-score-08134-in-titanic-kaggle-challenge.html\n",
    "\n",
    "TODO:\n",
    "    - Try more models (ADA, MLP etc)\n",
    "    - Feature selection based on more classifiers (select union of 15 best per classifier?)\n",
    "    - Build voting classifier based on iterations as well as different classifiers\n",
    "    - Plot learning curves"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. data prep & analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## // data reading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# import packages\n",
    "import re\n",
    "import sys\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import warnings\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "import seaborn as sns\n",
    "sns.set(style='white', context='notebook', palette='deep')\n",
    "\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.linear_model import LinearRegression, LogisticRegression\n",
    "from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier, GradientBoostingClassifier, VotingClassifier, AdaBoostClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier, ExtraTreeClassifier\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "from sklearn.model_selection import GridSearchCV, KFold, cross_val_predict, cross_val_score, learning_curve\n",
    "from sklearn.metrics import r2_score, accuracy_score\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.feature_selection import SelectFromModel\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>714.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>446.000000</td>\n",
       "      <td>0.383838</td>\n",
       "      <td>2.308642</td>\n",
       "      <td>29.699118</td>\n",
       "      <td>0.523008</td>\n",
       "      <td>0.381594</td>\n",
       "      <td>32.204208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>257.353842</td>\n",
       "      <td>0.486592</td>\n",
       "      <td>0.836071</td>\n",
       "      <td>14.526497</td>\n",
       "      <td>1.102743</td>\n",
       "      <td>0.806057</td>\n",
       "      <td>49.693429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.420000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>223.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>20.125000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.910400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>446.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.454200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>668.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>891.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>512.329200</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       PassengerId    Survived      Pclass         Age       SibSp  \\\n",
       "count   891.000000  891.000000  891.000000  714.000000  891.000000   \n",
       "mean    446.000000    0.383838    2.308642   29.699118    0.523008   \n",
       "std     257.353842    0.486592    0.836071   14.526497    1.102743   \n",
       "min       1.000000    0.000000    1.000000    0.420000    0.000000   \n",
       "25%     223.500000    0.000000    2.000000   20.125000    0.000000   \n",
       "50%     446.000000    0.000000    3.000000   28.000000    0.000000   \n",
       "75%     668.500000    1.000000    3.000000   38.000000    1.000000   \n",
       "max     891.000000    1.000000    3.000000   80.000000    8.000000   \n",
       "\n",
       "            Parch        Fare  \n",
       "count  891.000000  891.000000  \n",
       "mean     0.381594   32.204208  \n",
       "std      0.806057   49.693429  \n",
       "min      0.000000    0.000000  \n",
       "25%      0.000000    7.910400  \n",
       "50%      0.000000   14.454200  \n",
       "75%      0.000000   31.000000  \n",
       "max      6.000000  512.329200  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# read data\n",
    "train_data = pd.read_csv('~/.kaggle/competitions/titanic/train.csv')\n",
    "test_data = pd.read_csv('~/.kaggle/competitions/titanic/test.csv')\n",
    "\n",
    "# dataset info\n",
    "train_data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* 38 percent survived.\n",
    "* it looks like age is missing in 177 cases (891-714). we have to do something here, for example impute missing values or fill with mean/median etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "\n",
       "                                                Name     Sex   Age  SibSp  \\\n",
       "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
       "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
       "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
       "4                           Allen, Mr. William Henry    male  35.0      0   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  \n",
       "0      0         A/5 21171   7.2500   NaN        S  \n",
       "1      0          PC 17599  71.2833   C85        C  \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S  \n",
       "3      0            113803  53.1000  C123        S  \n",
       "4      0            373450   8.0500   NaN        S  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>PassengerId</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>418.000000</td>\n",
       "      <td>418.000000</td>\n",
       "      <td>332.000000</td>\n",
       "      <td>418.000000</td>\n",
       "      <td>418.000000</td>\n",
       "      <td>417.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1100.500000</td>\n",
       "      <td>2.265550</td>\n",
       "      <td>30.272590</td>\n",
       "      <td>0.447368</td>\n",
       "      <td>0.392344</td>\n",
       "      <td>35.627188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>120.810458</td>\n",
       "      <td>0.841838</td>\n",
       "      <td>14.181209</td>\n",
       "      <td>0.896760</td>\n",
       "      <td>0.981429</td>\n",
       "      <td>55.907576</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>892.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.170000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>996.250000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.895800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1100.500000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.454200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1204.750000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>39.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1309.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>76.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>512.329200</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       PassengerId      Pclass         Age       SibSp       Parch        Fare\n",
       "count   418.000000  418.000000  332.000000  418.000000  418.000000  417.000000\n",
       "mean   1100.500000    2.265550   30.272590    0.447368    0.392344   35.627188\n",
       "std     120.810458    0.841838   14.181209    0.896760    0.981429   55.907576\n",
       "min     892.000000    1.000000    0.170000    0.000000    0.000000    0.000000\n",
       "25%     996.250000    1.000000   21.000000    0.000000    0.000000    7.895800\n",
       "50%    1100.500000    3.000000   27.000000    0.000000    0.000000   14.454200\n",
       "75%    1204.750000    3.000000   39.000000    1.000000    0.000000   31.500000\n",
       "max    1309.000000    3.000000   76.000000    8.000000    9.000000  512.329200"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* age is missing here as well\n",
    "* also Fare is missing once."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>PassengerId</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892</td>\n",
       "      <td>3</td>\n",
       "      <td>Kelly, Mr. James</td>\n",
       "      <td>male</td>\n",
       "      <td>34.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>330911</td>\n",
       "      <td>7.8292</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>893</td>\n",
       "      <td>3</td>\n",
       "      <td>Wilkes, Mrs. James (Ellen Needs)</td>\n",
       "      <td>female</td>\n",
       "      <td>47.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>363272</td>\n",
       "      <td>7.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>894</td>\n",
       "      <td>2</td>\n",
       "      <td>Myles, Mr. Thomas Francis</td>\n",
       "      <td>male</td>\n",
       "      <td>62.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>240276</td>\n",
       "      <td>9.6875</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>895</td>\n",
       "      <td>3</td>\n",
       "      <td>Wirz, Mr. Albert</td>\n",
       "      <td>male</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>315154</td>\n",
       "      <td>8.6625</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>896</td>\n",
       "      <td>3</td>\n",
       "      <td>Hirvonen, Mrs. Alexander (Helga E Lindqvist)</td>\n",
       "      <td>female</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3101298</td>\n",
       "      <td>12.2875</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Pclass                                          Name     Sex  \\\n",
       "0          892       3                              Kelly, Mr. James    male   \n",
       "1          893       3              Wilkes, Mrs. James (Ellen Needs)  female   \n",
       "2          894       2                     Myles, Mr. Thomas Francis    male   \n",
       "3          895       3                              Wirz, Mr. Albert    male   \n",
       "4          896       3  Hirvonen, Mrs. Alexander (Helga E Lindqvist)  female   \n",
       "\n",
       "    Age  SibSp  Parch   Ticket     Fare Cabin Embarked  \n",
       "0  34.5      0      0   330911   7.8292   NaN        Q  \n",
       "1  47.0      1      0   363272   7.0000   NaN        S  \n",
       "2  62.0      0      0   240276   9.6875   NaN        Q  \n",
       "3  27.0      0      0   315154   8.6625   NaN        S  \n",
       "4  22.0      1      1  3101298  12.2875   NaN        S  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "# combine data\n",
    "combined_data = pd.concat([train_data, test_data])\n",
    "combined_data = combined_data.reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## // descripte stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7feb285cd4a8>"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = combined_data[(combined_data['Age'].notnull()) & (combined_data['Survived'].notnull())][['Age', 'Survived']]\n",
    "data['AgeGroup'] = (data['Age']/5).round()*5\n",
    "\n",
    "g = sns.factorplot(x=\"AgeGroup\", y=\"Survived\", data=data, kind=\"bar\", size = 6, palette = \"muted\")\n",
    "g.set_ylabels(\"Survival probability\")\n",
    "\n",
    "data = combined_data[(combined_data['Fare'].notnull()) & (combined_data['Survived'].notnull())][['Fare', 'Survived']]\n",
    "data['LogFare'] = data['Fare'].apply(np.log)\n",
    "data['LogFareGroup'] = data['LogFare'].round()\n",
    "\n",
    "g = sns.factorplot(x=\"LogFareGroup\", y=\"Survived\", data=data, kind=\"bar\", size = 6, palette = \"muted\")\n",
    "g.set_ylabels(\"Survival probability\")\n",
    "\n",
    "data = combined_data[combined_data['Survived'].notnull()]\n",
    "\n",
    "g = sns.factorplot(x=\"Pclass\", y=\"Survived\", data=data, kind=\"bar\", size = 6, palette = \"muted\")\n",
    "g.set_ylabels(\"Survival probability\")\n",
    "\n",
    "g = sns.factorplot(x=\"Embarked\", y=\"Survived\", data=data, kind=\"bar\", size = 6, palette = \"muted\")\n",
    "g.set_ylabels(\"Survival probability\")\n",
    "\n",
    "g = sns.factorplot(x=\"SibSp\", y=\"Survived\", data=data, kind=\"bar\", size = 6, palette = \"muted\")\n",
    "g.set_ylabels(\"Survival probability\")\n",
    "\n",
    "g = sns.factorplot(x=\"Parch\", y=\"Survived\", data=data, kind=\"bar\", size = 6, palette = \"muted\")\n",
    "g.set_ylabels(\"Survival probability\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## // imputation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      index   Age Cabin Embarked  Fare                Name  Parch  \\\n",
      "1043    152  60.5   NaN        S   NaN  Storey, Mr. Thomas      0   \n",
      "\n",
      "      PassengerId  Pclass   Sex  SibSp  Survived Ticket  \n",
      "1043         1044       3  male      0       NaN   3701  \n",
      "     index   Age Cabin Embarked  Fare  \\\n",
      "61      61  38.0   B28      NaN  80.0   \n",
      "829    829  62.0   B28      NaN  80.0   \n",
      "\n",
      "                                          Name  Parch  PassengerId  Pclass  \\\n",
      "61                         Icard, Miss. Amelie      0           62       1   \n",
      "829  Stone, Mrs. George Nelson (Martha Evelyn)      0          830       1   \n",
      "\n",
      "        Sex  SibSp  Survived  Ticket  \n",
      "61   female      0       1.0  113572  \n",
      "829  female      0       1.0  113572  \n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# replace fare\n",
    "print(combined_data[combined_data['Fare'].isnull()])\n",
    "combined_data.boxplot(column=['Fare'], by='Pclass')\n",
    "combined_data.loc[combined_data['PassengerId']==1044, 'Fare'] = combined_data[combined_data['Pclass']==3]['Fare'].median()\n",
    "\n",
    "# replace embarked\n",
    "print(combined_data[combined_data['Embarked'].isnull()])\n",
    "combined_data[combined_data['Cabin']=='B28']\n",
    "combined_data[combined_data['Ticket']==\"113572\"]\n",
    "combined_data[combined_data['Fare']==80]\n",
    "combined_data[combined_data['Name'].str.contains(\"Icard\")]\n",
    "combined_data[combined_data['Name'].str.contains(\"Stone\")]\n",
    "\n",
    "combined_data.Embarked.fillna('S', inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fare was easily replaced with some help from Pclass. Embarked was replaced by most common value though, got no help from other variables. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. feature engineering "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## // transform class variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Empty DataFrame\n",
      "Columns: [pclass_encoded]\n",
      "Index: []\n",
      "Empty DataFrame\n",
      "Columns: [embarked_encoded]\n",
      "Index: []\n",
      "Empty DataFrame\n",
      "Columns: [title_encoded]\n",
      "Index: []\n",
      "Empty DataFrame\n",
      "Columns: [cabin_letter_encoded]\n",
      "Index: []\n",
      "Empty DataFrame\n",
      "Columns: [family_id_encoded]\n",
      "Index: []\n",
      "    age  cabin_number  family_size     fare  id  log_fare  \\\n",
      "0  22.0        -999.0            2   7.2500   1  1.981001   \n",
      "1  38.0          85.0            2  71.2833   2  4.266662   \n",
      "2  26.0        -999.0            1   7.9250   3  2.070022   \n",
      "3  35.0         123.0            2  53.1000   4  3.972177   \n",
      "4  35.0        -999.0            1   8.0500   5  2.085672   \n",
      "\n",
      "   number_parents_children  number_siblings_spouses  sex_is_male  target  \\\n",
      "0                        0                        1            1       0   \n",
      "1                        0                        1            0       1   \n",
      "2                        0                        0            0       1   \n",
      "3                        0                        1            0       1   \n",
      "4                        0                        0            1       0   \n",
      "\n",
      "       ...       family_id_68  family_id_69  family_id_70  family_id_71  \\\n",
      "0      ...                  0             0             0             0   \n",
      "1      ...                  0             0             0             0   \n",
      "2      ...                  0             0             0             0   \n",
      "3      ...                  0             0             0             0   \n",
      "4      ...                  0             0             0             0   \n",
      "\n",
      "   family_id_72  family_id_73  family_id_74  family_id_75  family_id_76  \\\n",
      "0             0             0             0             0             0   \n",
      "1             0             0             0             0             0   \n",
      "2             0             0             0             0             0   \n",
      "3             0             0             0             0             0   \n",
      "4             0             0             0             0             0   \n",
      "\n",
      "   family_id_77  \n",
      "0             0  \n",
      "1             0  \n",
      "2             0  \n",
      "3             0  \n",
      "4             0  \n",
      "\n",
      "[5 rows x 114 columns]\n",
      "age\n",
      "cabin_number\n",
      "family_size\n",
      "fare\n",
      "id\n",
      "log_fare\n",
      "number_parents_children\n",
      "number_siblings_spouses\n",
      "sex_is_male\n",
      "target\n",
      "ticket_is_digit\n",
      "pclass_encoded\n",
      "pclass_0\n",
      "pclass_1\n",
      "pclass_2\n",
      "embarked_encoded\n",
      "embarked_0\n",
      "embarked_1\n",
      "embarked_2\n",
      "title_encoded\n",
      "title_0\n",
      "title_1\n",
      "title_2\n",
      "title_3\n",
      "title_4\n",
      "cabin_letter_encoded\n",
      "cabin_letter_0\n",
      "cabin_letter_1\n",
      "cabin_letter_2\n",
      "cabin_letter_3\n",
      "cabin_letter_4\n",
      "cabin_letter_5\n",
      "cabin_letter_6\n",
      "cabin_letter_7\n",
      "cabin_letter_8\n",
      "family_id_encoded\n",
      "family_id_0\n",
      "family_id_1\n",
      "family_id_2\n",
      "family_id_3\n",
      "family_id_4\n",
      "family_id_5\n",
      "family_id_6\n",
      "family_id_7\n",
      "family_id_8\n",
      "family_id_9\n",
      "family_id_10\n",
      "family_id_11\n",
      "family_id_12\n",
      "family_id_13\n",
      "family_id_14\n",
      "family_id_15\n",
      "family_id_16\n",
      "family_id_17\n",
      "family_id_18\n",
      "family_id_19\n",
      "family_id_20\n",
      "family_id_21\n",
      "family_id_22\n",
      "family_id_23\n",
      "family_id_24\n",
      "family_id_25\n",
      "family_id_26\n",
      "family_id_27\n",
      "family_id_28\n",
      "family_id_29\n",
      "family_id_30\n",
      "family_id_31\n",
      "family_id_32\n",
      "family_id_33\n",
      "family_id_34\n",
      "family_id_35\n",
      "family_id_36\n",
      "family_id_37\n",
      "family_id_38\n",
      "family_id_39\n",
      "family_id_40\n",
      "family_id_41\n",
      "family_id_42\n",
      "family_id_43\n",
      "family_id_44\n",
      "family_id_45\n",
      "family_id_46\n",
      "family_id_47\n",
      "family_id_48\n",
      "family_id_49\n",
      "family_id_50\n",
      "family_id_51\n",
      "family_id_52\n",
      "family_id_53\n",
      "family_id_54\n",
      "family_id_55\n",
      "family_id_56\n",
      "family_id_57\n",
      "family_id_58\n",
      "family_id_59\n",
      "family_id_60\n",
      "family_id_61\n",
      "family_id_62\n",
      "family_id_63\n",
      "family_id_64\n",
      "family_id_65\n",
      "family_id_66\n",
      "family_id_67\n",
      "family_id_68\n",
      "family_id_69\n",
      "family_id_70\n",
      "family_id_71\n",
      "family_id_72\n",
      "family_id_73\n",
      "family_id_74\n",
      "family_id_75\n",
      "family_id_76\n",
      "family_id_77\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEwCAYAAABMnTEqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3XtcVHXeB/DPAF4IuQiJ44NspVmUZjylKZnSMy6ggTIobLnqCmo+uiJhKhmuSiGU66Uo21zc1tAsS8LBCwqKCbve0pT1kpQWKPrIgIBc1ASG8/zBzlkQlDkzBwdPn/fr5Sv4cc6P78D04Zzf+f3OUQmCIICIiBTFxtoFEBGR/BjuREQKxHAnIlIghjsRkQIx3ImIFMjO2gX88ssvOH36NHr06AFbW1trl0NEdF8wGAwoLS3FgAED0LVr1xZft3q4nz59GhMnTrR2GURE96VNmzZh0KBBLdqtHu49evQA0FigWq22cjVERPeH4uJiTJw4UczQ21k93I1DMWq1Gr1797ZyNURE95c7DWfzgioRkQIx3ImIFIjhTkSkQAx3IiIFYrgTESkQw52ISIEY7kRECmT1ee53MmZeepvbbF8VfA8qISK6//DInYhIgRjuREQKxHAnIlIghjsRkQIx3ImIFIjhTkSkQAx3IiIFYrgTESkQw52ISIEY7kRECsRwJyJSoDbvLfPzzz9j7ty54udFRUWIioqCVqvF3LlzcfnyZXh4eOD999+Hs7MzBEFAQkICcnJy0LVrV7z77rvo379/u74IIiJqrs0j9z59+iA9PR3p6elIS0uDvb09/Pz8kJycDB8fH2RlZcHHxwfJyckAgNzcXBQWFiIrKwvx8fGIi4tr79dARES3kTQsc+jQIXh6esLDwwPZ2dnQarUAAK1Wi7179wKA2K5SqeDt7Y2qqiqUlJTIXzkREd2RpHDfuXMngoKCAABlZWVwd3cHALi7u6O8vBwAoNfroVarxX3UajX0er1c9RIRkQlMDvfa2lrs27cPo0aNuut2giC0aFOpVNIrIyIis5kc7rm5uejfvz8efPBBAICbm5s43FJSUgJXV1cAjUfqxcXF4n7FxcXiET4REd0bJof7zp07ERgYKH6u0Wig0+kAADqdDiNHjmzWLggC8vLy4OjoyHAnIrrHTAr3mzdv4uDBg/D39xfbZsyYgQMHDsDf3x8HDhzAjBkzAAC+vr7w9PSEn58fFi9ejKVLl7ZP5UREdEcmPUPV3t4eR44cadbWvXt3pKSktNhWpVIx0ImIrIwrVImIFIjhTkSkQAx3IiIFYrgTESkQw52ISIEY7kRECsRwJyJSIIY7EZECMdyJiBSI4U5EpEAMdyIiBWK4ExEpEMOdiEiBGO5ERArEcCciUiCGOxGRAjHciYgUiOFORKRADHciIgUyKdyrqqoQFRWFUaNGYfTo0Thx4gSuXbuGiIgI+Pv7IyIiApWVlQAAQRCwbNky+Pn5YcyYMThz5ky7vgAiImrJpHBPSEjA8OHDsXv3bqSnp6Nv375ITk6Gj48PsrKy4OPjg+TkZABAbm4uCgsLkZWVhfj4eMTFxbVn/URE1Io2w72mpgZHjx5FaGgoAKBz585wcnJCdnY2tFotAECr1WLv3r0AILarVCp4e3ujqqoKJSUl7fgSiIjodm2Ge1FREVxdXfHmm29Cq9Vi0aJFuHHjBsrKyuDu7g4AcHd3R3l5OQBAr9dDrVaL+6vVauj1+nYqn4iIWtNmuNfX1+P777/HhAkToNPpYG9vLw7BtEYQhBZtKpXKsiqJiEiSNsNdrVZDrVbj6aefBgCMGjUK33//Pdzc3MThlpKSEri6uorbFxcXi/sXFxeLR/hERHRvtBnuPXr0gFqtxs8//wwAOHToEPr27QuNRgOdTgcA0Ol0GDlyJACI7YIgIC8vD46Ojgx3IqJ7zM6UjRYvXoz58+ejrq4Onp6eeOedd9DQ0IDo6GikpqaiV69eSEpKAgD4+voiJycHfn5+sLe3R2JiYru+ACIiasmkcH/iiSeQlpbWoj0lJaVFm0qlwtKlSy2vjIiIzMYVqkRECsRwJyJSIIY7EZECMdyJiBSI4U5EpEAMdyIiBWK4ExEpEMOdiEiBGO5ERArEcCciUiCGOxGRAjHciYgUiOFORKRADHciIgViuBMRKRDDnYhIgRjuREQKxHAnIlIgkx6zp9Fo4ODgABsbG9ja2iItLQ3Xrl3D3LlzcfnyZXh4eOD999+Hs7MzBEFAQkICcnJy0LVrV7z77rvo379/e78OIiJqwuQj95SUFKSnp4vPUk1OToaPjw+ysrLg4+OD5ORkAEBubi4KCwuRlZWF+Ph4xMXFtUvhRER0Z2YPy2RnZ0Or1QIAtFot9u7d26xdpVLB29sbVVVVKCkpkadaIiIyicnhPm3aNIwbNw5ffvklAKCsrAzu7u4AAHd3d5SXlwMA9Ho91Gq1uJ9arYZer5ezZiIiaoNJY+5ffPEFevbsibKyMkRERKBPnz533FYQhBZtKpXK/AqJiEgyk47ce/bsCQBwc3ODn58fTp48CTc3N3G4paSkBK6urgAaj9SLi4vFfYuLi8UjfCIiujfaDPcbN26gpqZG/PjAgQPo168fNBoNdDodAECn02HkyJEAILYLgoC8vDw4Ojoy3ImI7rE2h2XKysowe/ZsAIDBYEBQUBBGjBiBp556CtHR0UhNTUWvXr2QlJQEAPD19UVOTg78/Pxgb2+PxMTE9n0FRETUQpvh7unpiW3btrVo7969O1JSUlq0q1QqLF26VJ7qiIjILFyhSkSkQAx3IiIFYrgTESkQw52ISIEY7kRECsRwJyJSIIY7EZECMdyJiBSI4U5EpEAMdyIiBWK4ExEpEMOdiEiBGO5ERArEcCciUiCGOxGRAjHciYgUiOFORKRADHciIgUyOdwNBgO0Wi3+93//FwBQVFSEsLAw+Pv7Izo6GrW1tQCA2tpaREdHw8/PD2FhYbh06VL7VE5ERHdkcrhv2LABffv2FT9fuXIlwsPDkZWVBScnJ6SmpgIAtmzZAicnJ+zZswfh4eFYuXKl/FUTEdFdmRTuxcXF2L9/P0JDQwEAgiDg8OHDCAgIAACEhIQgOzsbALBv3z6EhIQAAAICAnDo0CEIgtAetRMR0R2YFO6JiYlYsGABbGwaN6+oqICTkxPs7OwAAGq1Gnq9HgCg1+vRq1cvAICdnR0cHR1RUVHRHrUTEdEdtBnu33zzDVxdXTFgwIC7bqdSqQCg1aN049eIiOjesGtrg+PHj2Pfvn3Izc3FrVu3UFNTg4SEBFRVVaG+vh52dnYoLi6Gu7s7gMaj+CtXrkCtVqO+vh7V1dVwcXFp9xdCRET/0eaR+7x585Cbm4t9+/Zh9erVGDp0KFatWoUhQ4YgMzMTALB161ZoNBoAgEajwdatWwEAmZmZGDp0KI/ciYjuMbPnuS9YsADr16+Hn58frl27hrCwMABAaGgorl27Bj8/P6xfvx7z58+XrVgiIjJNm8MyTQ0ZMgRDhgwBAHh6eorTH5vq0qULPvjgA3mqIyIis3CFKhGRAjHciYgUiOFORKRADHciIgViuBMRKRDDnYhIgRjuREQKxHAnIlIghjsRkQIx3ImIFIjhTkSkQAx3IiIFYrgTESkQw52ISIEY7kRECsRwJyJSIIY7EZECMdyJiBSI4U5EpEBtPkP11q1bmDhxImpra2EwGBAQEICoqCgUFRXh9ddfR2VlJZ588kn8+c9/RufOnVFbW4uYmBicOXMGLi4ueO+999C7d+978VqIiOjf2jxy79y5M1JSUrBt2zbodDr84x//QF5eHlauXInw8HBkZWXByclJfFj2li1b4OTkhD179iA8PBwrV65s9xdBRETNtRnuKpUKDg4OAID6+nrU19dDpVLh8OHDCAgIAACEhIQgOzsbALBv3z6EhIQAAAICAnDo0CEIgtBe9RMRUStMGnM3GAwIDg7G888/j+effx6enp5wcnKCnV3jqI5arYZerwcA6PV69OrVCwBgZ2cHR0dHVFRUtFP5RETUGpPC3dbWFunp6cjJycHJkyfx888/t9hGpVIBQKtH6cavERHRvSFptoyTkxOGDBmCvLw8VFVVob6+HgBQXFwMd3d3AI1H8VeuXAHQOIxTXV0NFxcXmcsmIqK7aTPcy8vLUVVVBQD45ZdfcPDgQfTt2xdDhgxBZmYmAGDr1q3QaDQAAI1Gg61btwIAMjMzMXToUB65ExHdY21OhSwpKcHChQthMBggCAJGjRqF//mf/8Gjjz6KuXPn4v3338cTTzyBsLAwAEBoaCgWLFgAPz8/ODs747333mv3F0FERM21Ge5eXl7Q6XQt2j09PcXpj0116dIFH3zwgTzVERGRWbhClYhIgRjuREQKxHAnIlIghjsRkQIx3ImIFKjN2TL3u999Ocuk7b56+eN2roSI6N7hkTsRkQIx3ImIFIjhTkSkQAx3IiIFYrgTESkQw52ISIEY7kRECsRwJyJSIIY7EZECMdyJiBSI4U5EpEAMdyIiBWoz3K9cuYLJkydj9OjRCAwMREpKCgDg2rVriIiIgL+/PyIiIlBZWQkAEAQBy5Ytg5+fH8aMGYMzZ8607ysgIqIW2gx3W1tbLFy4ELt27cKXX36Jzz//HOfPn0dycjJ8fHyQlZUFHx8fJCcnAwByc3NRWFiIrKwsxMfHIy4urr1fAxER3abNcHd3d0f//v0BAN26dUOfPn2g1+uRnZ0NrVYLANBqtdi7dy8AiO0qlQre3t6oqqpCSUlJO74EIiK6naQx90uXLuHs2bN4+umnUVZWBnd3dwCNfwDKy8sBAHq9Hmq1WtxHrVZDr9fLWDIREbXF5Id1XL9+HVFRUYiNjUW3bt3uuJ0gCC3aVCqVedV1MAeCx7e5zbD0r+9BJUREd2fSkXtdXR2ioqIwZswY+Pv7AwDc3NzE4ZaSkhK4uroCaDxSLy4uFvctLi4Wj/CJiOjeaDPcBUHAokWL0KdPH0RERIjtGo0GOp0OAKDT6TBy5Mhm7YIgIC8vD46Ojgx3IqJ7rM1hme+++w7p6el47LHHEBwcDAB4/fXXMWPGDERHRyM1NRW9evVCUlISAMDX1xc5OTnw8/ODvb09EhMT2/cVEBFRC22G+6BBg/DDDz+0+jXjnPemVCoVli5danllRERkNq5QJSJSIIY7EZECMdyJiBSI4U5EpEAMdyIiBWK4ExEpEMOdiEiBGO5ERArEcCciUiCGOxGRAjHciYgUiOFORKRADHciIgViuBMRKRDDnYhIgRjuREQKxHAnIlIghjsRkQK1Ge5vvvkmfHx8EBQUJLZdu3YNERER8Pf3R0REBCorKwE0Pkx72bJl8PPzw5gxY3DmzJn2q5yIiO6ozXAfN24c/va3vzVrS05Oho+PD7KysuDj44Pk5GQAQG5uLgoLC5GVlYX4+HjExcW1S9FERHR3bYb74MGD4ezs3KwtOzsbWq0WAKDVarF3795m7SqVCt7e3qiqqkJJSUk7lE1ERHdj1ph7WVkZ3N3dAQDu7u4oLy8HAOj1eqjVanE7tVoNvV4vQ5lERCSFrBdUBUFo0aZSqeT8FkREZAKzwt3NzU0cbikpKYGrqyuAxiP14uJicbvi4mLxCJ+IiO4ds8Jdo9FAp9MBAHQ6HUaOHNmsXRAE5OXlwdHRkeFORGQFdm1t8Prrr+Pbb79FRUUFRowYgTlz5mDGjBmIjo5GamoqevXqhaSkJACAr68vcnJy4OfnB3t7eyQmJrb7CyAiopbaDPfVq1e32p6SktKiTaVSYenSpZZXRUREFuEKVSIiBWK4ExEpEMOdiEiBGO5ERArEcCciUqA2Z8tQ+3h73vY2t1myasw9qISIlIhH7kRECsRwJyJSIIY7EZECMdyJiBSI4U5EpEAMdyIiBWK4ExEpEMOdiEiBGO5ERArEcCciUiCGOxGRAjHciYgUiOFORKRA7RLuubm5CAgIgJ+fH5KTk9vjWxAR0V3Ifstfg8GAt99+G+vXr0fPnj0RGhoKjUaDRx99VO5vRQC+y1rQ5jbP+q8wqa9XM463uc26l54xqS8isi7Zw/3kyZN46KGH4OnpCQAIDAxEdnb2HcPdYDAAAIqLi5u1190ob/N7Xbp0qc1taitutrmNqX1dra2VpR8AqJHp9ZWW3ZCtpl/KS2Xpa3rCHpO+398W+Zm0HRG1ZMxMY4beTvZw1+v1UKvV4uc9e/bEyZMn77h9aWljoEycOFHy9xq5713pBd6pr/dGytSRTP0ASN+XKFNPe2XqBxgZL1tXsv7+iH6tSktL8dBDD7Volz3cBUFo0aZSqe64/YABA7Bp0yb06NEDtra2cpdDRKRIBoMBpaWlGDBgQKtflz3c1Wp1syEWvV4Pd3f3O27ftWtXDBo0SO4yiIgUr7UjdiPZZ8s89dRTKCwsRFFREWpra7Fz505oNBq5vw0REd2F7EfudnZ2WLJkCaZPnw6DwYDx48ejX79+cn8bIiK6C5XQ2iA5ERHd17hClYhIgRjuREQKxHAn6iAaGhpQU1MjeT+DwYD58+e3Q0V0P+uw4a70N6wgCLhy5Yq1yyAzyfX+nDdvHmpqanDjxg289NJLGDVqFP72t79J6sPW1hYVFRWoNWEFNXVMDQ0NOH687dt/SNFhw13uN+zFixfFvo4cOYINGzagqqrKan2pVCrMnj3brO9/O4PBgPDwcFn6+vTTT1vcCkKKMWPG3PWfuW7cuIGGhgYAQEFBAbKzs1FXVye5n4KCAkyZMgVBQUEAgPz8fPzlL3+R3I9c78/z58+jW7du2Lt3L3x9ffHNN98gPT1dcj8eHh6YMGECPvroI6xfv178Zym9Xo/jx4/j6NGj4j+pPvvsM1RWVlpcCwBcvXoVsbGxmD59OoDGn9+WLVvM6uvHH39ERkYGdDqd+E+qKVOmmNTWFhsbGyxfvlzyfncj+1RIORnfsBqNBg888IDYHhERIbmvOXPm4Ouvv8aFCxewaNEiaDQazJs3D+vWrbNaX08//TROnjyJgQMHSq6hKVtbW3Tt2hXV1dVwdHS0qK+amhpMmzYNzs7OCAwMREBAAB588EGT91+7dq1F3/9OJk2ahE2bNqGqqgrh4eEYMGAAMjIysGrVKkn9LF68GDExMViyZAkAwMvLC/Pnz8cf//hHyTXJ8f6sr69HXV0d9u7di0mTJqFTp053XdF9J+7u7nB3d4cgCLh+/brk/VuzYsUK7Nq1C3379m22enzw4MGS+iktLUVoaCiefPJJjB8/HsOHDzfrNQLAwoULMW7cOPF99vDDD2Pu3LkICwuT1M+aNWtw5MgR/PTTT/D19UVubi6effZZaLVak/a/desWbt68iYqKClRWVoor82tqalBSUiLtRf3bsGHDkJmZCX9/f7N/Pk116HCX8w1rY2MDOzs77NmzB1OmTMHkyZNN/kW2V19HjhzB5s2b4eHhAXt7e7F9+/btkvvq0qULxowZg+eff75Z0PzpT3+S1E9kZCQiIyORn5+PXbt2YdKkSVCr1fj0009N2t/Dw0P8+OrVqzh16hQAYODAgXBzc5NUS1OCIMDe3h6pqamYNGkSXn31VbN+5jdv3mzxx9Tc217I8f58+eWXodFo4OXlhcGDB+Py5cvo1q2b5H4iIyMBNJ7hNP39W2Lv3r3YvXs3OnfubFE/c+fORXR0NP75z38iLS0N8fHxGD16NEJDQ/Gb3/xGUl8VFRV46aWXxFuJ29nZwcZG+gBEZmYm0tPTodVq8c477+Dq1auS/l/ZvHkzUlJSUFJSgpCQELG9W7duZt0nCwDWr1+Pmzdvws7ODp07d4YgCFCpVGYP13TocDe+YeVgZ2eHHTt2QKfT4eOPPwbQeNRkzb7MOWu4kxdffBEvvviibP25ubnhwQcfhIuLC8rKyiTvn5GRgRUrVuC5556DIAiIj49HTEwMRo0aZVY9giDgxIkT2L59OxISEgDc+W54d9O9e3dcvHhRPDLavXs3evToYVZNxvdnTU0NVCoVHBwcJPfxhz/8AX/4wx/Ezz08PLBhwwbJ/Zw4cQKLFi3CjRs3sH//fuTn52Pz5s2Ii4uT3JeRp6cn6urqLA53oHEYskePHnjwwQdha2uLyspKREVF4fnnn0dMTIzJ/TzwwAOoqKgQf395eXlmna126dJFPEirqamBm5sbioqKTN5/ypQpmDJlCjZu3IjJkydL/v6tOXHihCz9GHXIcJ85c+Zdv27Oqf8777yDzZs3Y+bMmfD09ERRURHGjh1rVn1y9eXh4YFjx47hwoULGD9+PMrLy80+AgwJCUF5eeNthF1dXc3qAwA+//xz7Nq1C+Xl5QgICMCyZcvMuhf/2rVrkZqaKh6tl5eXIzw83Oxwj42NxV//+lf89re/Rb9+/VBUVIQhQ4ZI7mfp0qVYvHgxfv75ZwwfPhy9e/fGihWm3e/+dj/++CNiYmLE8eTu3btj+fLlklZkp6SkYPz48XBwcMCiRYtw9uxZzJs3Dy+88IKkWhITE/HJJ59g1qxZABqHm44dOyapj9vZ29tDq9XCx8enWcBLPRvcsGEDdDodunfvjtDQUMTExKBTp05oaGiAv7+/pHBfuHAhZs2ahYsXL+KVV15BRUUFkpKSJNUDNN6wsKqqCmFhYRg3bhweeOABs4ZHw8LCsH79enz33XdQqVR49tlnMWHCBHTp0sXkPn766Sf07dsXZ86cafXr/fv3l1wX0EFXqA4dOhS9evVCYGAgnn766RZ3mnzuuecs6r+yshJXrlyBl5eX5H0NBgPeeOMNrFy50qIagMZxv9OnT6OgoACZmZnQ6/V47bXXsHnzZpP7EAQBa9aswWeffQag8aq7ra0tJk2aZNaZz8qVKxEYGIgnnnhC8r5NjRkzptnwUkNDA4KDg80acmoPxgu05gyBGL3yyiuIjo7G0KFDATQOs7333nuSfn9jx47Ftm3b8I9//AObNm1CdHQ03nzzTWzdulVSLWFhYdiyZQu0Wq14YdDYt7nuVEPTYQhTJCUlITQ0tNmQnZEx2KSor69HQUEBBEHAI488gk6dOkna/3aXLl1CTU2NWXnw2muvwcHBQTy427lzJyorK/HBBx+Y3MfixYsRHx/f7Ayg6Zi7OWdyQAc9cj9w4AAOHDiAnTt3YseOHfD19UVQUJBF96iZPHkyPv74Y9TX10Or1cLV1RWDBw/Gm2++KamfprMkLD1d3bNnD3Q6nfg/S8+ePSUfuaekpOD48eNITU0VH5BSVFSEuLg4fPrpp5Jn0cyfPx/Hjh3D119/3exswti3qV544QVMmzYNgYGBABqHaUaMGCGpD0C+s7i2Zo6Yc5H+xo0bYrADwJAhQ3DjRtsPT2nKeOCSk5OD8ePHw8vLq9XbZrelV69eOH78OFQqFWpra7Fx40bJoXm7kJAQ1NbWorCwEAAkB+m1a9cA/Gf2iPFzIxcXF8k1GgwG5OTk4PLlyzAYDDhw4AAA835/er1e7AcAjh49KvlicUFBQbM/oEOHDpV8Fh8WFobS0lJs3LgRQOMf1czMTPTu3duioekOGe62trYYMWIERowYgdraWuzYsQOTJ0/G7NmzzR7fqq6uRrdu3bBlyxaMGzcOUVFRZk/Nk2sWj3FmhPGvtNRgAACdToe///3vzYZiPD09sWLFCkydOlVyuDc9mxg/fjzq6uqwYMECk49GL1y4gKtXr+KNN95AVlYWvvvuOwiCAG9vb7OGrvLy8u56FmcquWaQNOXp6YmPPvoIwcHBAIBt27ahd+/ekvoYMGAApk6dikuXLolz3s25QBgXF4eEhATo9Xr4+vpi2LBh4owgcx05cgQLFy6Eh4eHuC5j+fLlJgfguHHjoFKpxH2dnJwAAFVVVejVqxf27dsnuaaZM2eiS5cueOyxx8z6ORnJNRPoySefRF5eHry9vQEA//rXv/DMM9IeRbl06VLx4OPo0aNYtWoVFi9ejLNnz2LJkiWSzgKa6pDhDgC1tbXYv38/duzYgcuXL2Py5Mnw9/c3uz+DwYCSkhLs2rUL0dHRFtUm1yye0aNHY8mSJaiqqsJXX32Fr7/+WvKUrvr6+lbH2F1dXc26yGvp2URiYiLmzp0LAPD39xd/Z6dOnUJiYqLk6yVyncXJeXHeKDExER9++CHmzJkDQRAwaNAgvPPOO5L6SEhIwNmzZ+Hp6Ql7e3tUVFQgMVH6E7hcXV0lTwtty/Lly/HJJ5+gT58+ABqPUufNm4e0tDST9jeG95IlSzBy5Ej4+voCaDxLOXTokFk1FRcXyzK0J9dMoH/961/Q6XT4r//6LwDA//3f/6Fv377igaMptRoMBri4uABoPMN9+eWXERAQgICAAPHAwRwdMtzfeOMNnDt3DsOHD0dkZCQee+wxi/v84x//iGnTpuHZZ5/FwIEDUVRUhIcfftisvuQKimnTpuHAgQNwcHBAQUEBoqKiMGzYMEl93O002ZyxSEvPJi5fvtzq2OVTTz2Fy5cvS65HrrO4ZcuW3fXrUi8SAoCzs7NZ+zWlUqlw/vx5fPPNN4iMjMTNmzfNWhhVUFCAuLg4lJWVYceOHcjPz8e+ffvMmr9vVFdXJwY70DgsY87CsdOnT+Ptt98WP/f19TXrIigAjBgxAv/85z8lX3C+nVwzgaSuJm5NQ0MD6uvrYWdnh0OHDiE+/j/PsjRnRphRhwz39PR02Nvbo6CgQByHAmDRvM/Ro0dj9OjR4ueenp748MMPJfUh9yyeFStWYMGCBc0C3dhmqvz8/FZPAwVBMCskLD2buHXr1h2/9ssvv0iuB5DnLM7cGQetSUhIwKJFi1p9P6hUKri4uODll18WT9XvJi4uDjY2Njh8+DAiIyPh4OAgLpKTQs7FWUYDBgxAbGysePS4ffv2Oz7S7W66d++Ov/zlLxg7dixUKhW2bduG7t27m1WTt7c3IiMj0dDQADs7O7MzQa6ZQMaLxGVlZc3e+8YjeVMEBgZi0qRJ6N69e7Mn0124cMGiC/4dMtzz8/Nl62vdunV49dVXER8f3+qqLym/TLnGf40OHjzYoi03N1dSuJ89e9aiGm5n6dnEU089ha+++gq/+93vmrVv2bLFrICV6yxO6gyPuzGG3dSpU1uqcLHlAAAL7klEQVT9ekVFBWJjY5GRkdFmXydPnsTWrVvFBVnOzs5mHR3LuTjL6K233sKmTZuwceNGCIKAwYMH4/e//73kflatWoU1a9aIZ7yDBg0yewjp3XffxebNm/H4449btIpTo9HI8oS47OxsLF++HCUlJXB1dRWHZXbu3GlyH7NmzYKPjw9KS0sxbNgw8XU1NDRg8eLFZtfWIcNdTsar8eYccdxOrvHfzz//HF988QWKioqaXdS9fv265Isx7WHYsGFioBsMBmzbts3ki6GxsbGIjIzE9u3bxTA/ffo06urqsGbNGsm1yHUWJ+dZl/G9dLcpuaYOidnZ2cFgMIj/Q5eXl5t1oVDOxVlGnTt3RnBwMIKDgy1aO+Hi4mLx8JXRww8/jMcee8zi5fmWzgQySkpKwpdffomIiAjodDocPnxYUrAbtXaW98gjj0jup6kOOc/9fmAc//3zn/8sefy3uroalZWVWL16NebNmye2Ozg4iBdW7rWamhps2rQJer0eGo0Gw4YNw6ZNm/DJJ5/Ay8tLXIlrqsOHD+PcuXMAgEcffRQ+Pj7tUbbJ2mPtRGFhIVavXo3z5883OyXPzs42uY9t27YhIyMD33//PUJCQrB792689tpreOmllyTVUlRUhMWLF+PEiRNwcnISF2dJnb0DtFw7IQgCbGxszF47UV5ejnXr1rX4OZkzf3vhwoUoKirCiBEjmg2nSJ2pZulMIKNx48YhLS0NY8eOhU6ng42NDUJDQ5Gamiqpn/ag+HCXe5z89vFfjUaD0NBQ9OzZU3JtFy9ehFqtRufOnXHkyBH88MMP0Gq14pSxe2nWrFlwdnaGt7c3Dh06hKqqKtTV1WHRokUWL2jqCIxzonfu3IkffvhBlrUTEyZMQFRUlDgLKC0tDYIgICoqSlI/P/30Ew4fPgxBEODj42PR/HQ5Fmd9+umnyMnJwdtvv91i7cTw4cMlT6+dOnUqRo8ejb///e946623sHXrVri6ukoafjS609mf1D8648aNw8qVK82eCWQUHh6Ojz76CKtXr0ZFRQVcXV1x+vRpSQvZ2o2gcEOGDBG0Wq2wbt064dtvvxWOHDnS7J8UMTExQkhIiLB69Wrhhx9+sLi2sWPHCnV1dUJhYaEwcuRIISEhQZg+fbrF/ZojKChI/Li+vl4YNGiQUF1dbZVa2tutW7eEr7/+WhgyZIiwYcMGs/sJCQkRBKH5z27ChAmS+pg/f75JbW25deuWsG3bNuHjjz8WPvzwQ/GfOYKDg4WysrIW7WVlZUJwcLDk/lr7OU2cONGs2uTStJa7td1JYWGhcOzYMeH69euCwWAQ6urqhLS0NOHDDz8UTp06JWepZlP8mLucq13lnsVjvHFRVlaWxXeqtJSd3X/eCra2tujdu7dFR38dkdxrJzp37oyGhgY89NBD+Oyzz9CzZ0/JN1k7f/58s88NBsMd7zFyN7NmzYKjoyP69+9v8fQ+uddOGN9b7u7u2L9/P9zd3SU/M0DuM3BLZwIZ13MYFzHa2NggJCQEp06dwpo1a9rt1tdSKD7c5VztKucsHuA/d5dMT0+3+E6Vlmo6pVIQBNy6dQvPPPOMxbcd7SjaY+1EbGwsbt68iT/96U9ISkrC4cOHTX7gwl//+lesXbu22c8ZaPyDcftMI1Po9Xp88sknkvdrjdxrJ2bNmoXq6mq88cYbiI+Px/Xr1xEbGyupD7lnqlk6E0ju9RztQfFj7oC84+RyOn/+PDZv3gxvb28EBQWhqKgIu3btwowZM6xalxJ5eXmJ98xvOtPC2n+8Vq1a1eyiurkWL16MSZMm4fHHH7e4ryeeeKLZ8wWMhH+vnTDnzOJ2Uu971B7XTCy5i6qfnx/27Nkj+Wv3kuLDvekRW2BgoCxHbPTrJvcQwe03sAJMv8eJcSqtwWDAhQsX0Lt372bDMh3lLpy3e/HFF7F//36z9rVkppog00yg119/HUOHDm11PceBAwfw/vvvm9xXe1F8uHfUIzZAnql0dO/JOa1y5cqVyMjIaHEDK1P/QLQ1BNDabXY7Al9fX+Tk5EjaR44zcLlmAl29ehWRkZHo1KlTq+s5LF1jIIt7fQWX/uOVV14RDh48KAQFBQmXLl0SPvjgAyEpKcnaZVEb6uvrhZycHCEmJkYIDg4WVq9eLfz4449m9eXv7y/cunXL4pq++uqrFm0rVqywuN/24uvrK2l7uWaqyT0T6NChQ8KGDRuEDRs2CAcPHjS7rvag+AuqHdmtW7fExT0eHh6YM2cOfv/730ueJ033lpwX6eW6gVVmZia6dOkiriSOi4sz695Ccvrv//7vVleSCv++YC+FXDPV5J4JNHTo0Gb39O9IGO5WJMdUOrIOuaZVynUDqzVr1mDWrFmwsbFBbm4unJ2dLXp+qhzkfCaoXDPV5J4J1JEpfsy9Izt58iT69u2L6upqJCUlobq6GtOnTzfpboJkPXJepLf0UXZNn250/fp1zJ49G88884x49met21l0VPdiJlBHwXAnkqgjXaTXaDTi046M/zVSqVS8OP8rxnC3Armn0tH9yxjOt2Mok6U45m4Fcq+2o/tX04dy1NbWYteuXaisrDSrr+PHj7eYL2+t21mQ9fHI3QraY7UdKceECRPwxRdfSNpnwYIFKCoqgpeXlzhfXqVSyXYfdbr/8MjdCuScSkf3t6YX8BoaGnD69GmzHrp++vRpZGRkWPwQC1IOhruVyH2HQro/vfvuu+LHdnZ28PDwMGvper9+/VBaWgp3d3c5y6P7GIdlrID3uyG5TZ48Gfn5+Rg4cGCz+dq8OP/rxXC3go40lY6sq7q6GmvWrMHRo0cBNN6XZvbs2XB0dJTUz7fffttquzmPDyRlYLgTWdGcOXPQr18/cdFSeno68vPzzXqY+NWrV3Hq1CkAwMCBA+Hm5iZrrXR/kf6YdSKSzcWLFxEVFQVPT094enoiMjISRUVFkvvJyMhAWFgYdu/ejV27dokf068XL6gSWVHXrl1x7NgxDBo0CADw3XffoWvXrpL7Wbt2LVJTU8Wj9fLycoSHh2PUqFGy1kv3D4Y7kRW99dZbiImJQU1NDQDAycmp2QwaUwmC0GwYxsXFhYvjfuU45k7UARjD3dyHki9fvhw//vgjAgMDAQA7d+7E448/jpiYGNlqpPsLw53ICtavX49u3bohLCysWfvGjRthMBgkPV/UKCsrC8ePHxcf+Pzb3/5WpmrpfsRwJ7KCoKAgpKWltXhIR21tLcaPH2/ys0+bPhDj9v+Vu3Tpgt/85jeYO3eu+FAY+vXgmDuRFahUqlafviT1iUx3eyCGwWDAuXPnMH/+fOzYsUNyjXR/41RIIiu5evWqSW3msrW1hZeXFyZNmiRbn3T/YLgTWcG0adMwY8YMfPvtt6ipqUFNTQ2OHDmCmTNnYurUqbJ+r1deeUXW/uj+wDF3IivJycnBunXrcO7cOQCNN/969dVX4evra+XKSAkY7kRECsRhGSIiBWK4ExEpEMOdiEiBGO5EVnT16lXExsZi+vTpAIDz589jy5YtVq6KlIDhTmRFCxcuxAsvvICSkhIAwMMPP4wNGzZYuSpSAoY7kRVVVFTgpZdego1N4/+KdnZ24sdEluC7iMiKHnjgAVRUVIj3h8nLy5P8iD2i1nCeO5EVnTlzBvHx8Th37hz69euHiooKJCUlwcvLy9ql0X2O4U5kZfX19SgoKIAgCHjkkUfQqVMna5dECsBwJ7Ky48eP4/LlyzAYDGKbVqu1YkWkBLzlL5EVLViwAEVFRfDy8oKtrS2AxtsBM9zJUgx3Iis6ffo0MjIyxAuqRHLhbBkiK+rXrx9KS0utXQYpEMfciaxg5syZAIDr168jPz8fAwcObHYhde3atdYqjRSCwzJEViD3AzmIbsdhGSIreO655/Dcc88hJydH/LhpG5GlGO5EVnTw4MEWbbm5uVaohJSGwzJEVvD555/jiy++QFFREcaMGSO2X79+Hc8884wVKyOl4AVVIiuorq5GZWUlVq9ejXnz5ontDg4OcHFxsWJlpBQMdyIiBeKYOxGRAjHciYgUiOFORKRADHciIgViuBMRKdD/A9NB4ecjLgEFAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "le = LabelEncoder()\n",
    "\n",
    "# create features\n",
    "log_fare = combined_data['Fare'].apply(np.log)\n",
    "family_size = combined_data['SibSp'] + combined_data['Parch'] + 1\n",
    "cabin_number = combined_data['Cabin'].str.extract('(\\d+)', expand=False).fillna(-999)\n",
    "\n",
    "ticket_is_digit =  combined_data['Ticket'].apply(lambda x: 1 if x.isdigit() else 0)\n",
    "sex_is_male =  combined_data['Sex'].apply(lambda x: 1 if x == 'male' else 0)\n",
    "\n",
    "combined_data['title'] = combined_data['Name'].map(lambda name:name.split(',')[1].split('.')[0].strip())\n",
    "combined_data['title'].value_counts().plot(kind='bar', stacked=True)\n",
    "combined_data['title'] = combined_data['title'].apply(lambda x: x if x in ['Mr', 'Mrs', 'Miss', 'Master'] else 'Other')\n",
    "\n",
    "combined_data['family_id'] = combined_data['Name'].map(lambda name:name.split(',')[0].strip())\n",
    "combined_data['family_id'] = combined_data['family_id'].str.cat(family_size.astype(str), sep='')\n",
    "family_count = combined_data.groupby('family_id').PassengerId.count()\n",
    "family_count = family_count.rename('family_count')\n",
    "combined_data = combined_data.join(family_count, on=['family_id'])\n",
    "small_family_indices = (combined_data['family_count'] <= 2)\n",
    "combined_data.loc[small_family_indices, 'family_id'] = 'SMALL_FAMILY'\n",
    "combined_data = combined_data.drop('family_count', axis=1)\n",
    "\n",
    "combined_data['cabin_letter'] = combined_data['Cabin'].str.slice(0, 1)\n",
    "combined_data['cabin_letter'].fillna('Unknown', inplace=True)\n",
    "\n",
    "def encode_feature(df, feature, by):\n",
    "    \n",
    "    feature_encoded_name = feature.lower() + '_encoded'\n",
    "    \n",
    "    stats = combined_data[combined_data[by] != -999].groupby(feature)[by].mean()\n",
    "    stats = stats.sort_values(ascending=False).to_frame()\n",
    "    stats = stats.reset_index()\n",
    "    stats[feature_encoded_name] = stats.index\n",
    "    stats = stats.drop(by, axis=1)\n",
    "    stats = stats.set_index(feature)\n",
    "    \n",
    "    feature_encoded = df.join(stats, on=feature)[[feature_encoded_name]]\n",
    "    \n",
    "    print(feature_encoded[feature_encoded[feature_encoded_name].isnull()])\n",
    "    \n",
    "    feature_onehot = pd.get_dummies(feature_encoded[feature_encoded_name])\n",
    "    feature_onehot.columns = [feature.lower() + '_' + str(x).lower() for x in feature_onehot.columns]\n",
    "    \n",
    "    return feature_encoded, feature_onehot\n",
    "\n",
    "pclass_encoded, pclass_onehot = encode_feature(combined_data, 'Pclass', 'Survived')\n",
    "embarked_encoded, embarked_onehot = encode_feature(combined_data, 'Embarked', 'Survived')\n",
    "title_encoded, title_onehot = encode_feature(combined_data, 'title', 'Survived')\n",
    "cabin_encoded, cabin_onehot = encode_feature(combined_data, 'cabin_letter', 'Survived')\n",
    "family_encoded, family_onehot = encode_feature(combined_data, 'family_id', 'Survived')\n",
    "\n",
    "others = pd.DataFrame(data = {\n",
    "    'id': combined_data['PassengerId'],\n",
    "    'target': combined_data['Survived'].fillna(-999).astype(int),\n",
    "    'fare': combined_data['Fare'],\n",
    "    'log_fare': log_fare,\n",
    "    'age': combined_data['Age'],\n",
    "    'number_siblings_spouses': combined_data['SibSp'],\n",
    "    'number_parents_children': combined_data['Parch'],\n",
    "    'family_size': family_size,\n",
    "    'cabin_number': cabin_number.astype(float),\n",
    "    'sex_is_male': sex_is_male,\n",
    "    'ticket_is_digit': ticket_is_digit\n",
    "    })\n",
    "\n",
    "combined_data = pd.concat(\n",
    "    [others,\n",
    "     pclass_encoded, pclass_onehot, embarked_encoded, embarked_onehot,\n",
    "     title_encoded, title_onehot, cabin_encoded, cabin_onehot,\n",
    "     family_encoded, family_onehot]\n",
    "    , axis=1)\n",
    "\n",
    "print(combined_data.head())\n",
    "for x in combined_data.columns:\n",
    "    print(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## // binning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "def bin_variable(x, y, min_samples_leaf=20, max_leaf_nodes=2):\n",
    "\n",
    "    clf = DecisionTreeClassifier(criterion = 'entropy', min_samples_leaf=min_samples_leaf, max_leaf_nodes=max_leaf_nodes)\n",
    "    clf.fit(x.values.reshape(-1,1), y.values)\n",
    "\n",
    "    lower, higher = x.min(), x.max()\n",
    "    threshold = clf.tree_.threshold[clf.tree_.threshold>-2].tolist()\n",
    "    threshold.append(lower)\n",
    "    threshold.append(higher)\n",
    "    threshold.sort()\n",
    "\n",
    "    test = pd.concat([x, y], axis=1)\n",
    "    test['binned'] = pd.cut(x, threshold, include_lowest=True)\n",
    "\n",
    "    test.groupby('binned').target.mean().plot(kind='bar', stacked=True)\n",
    "    plt.show()\n",
    "    test.groupby('binned').target.count().plot(kind='bar', stacked=True)\n",
    "    plt.show()\n",
    "    \n",
    "    return threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAE3CAYAAAC6r7qRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAHk5JREFUeJzt3XtwlOXh9vFrIQRQSThIdlEyqUKkSKLYoRYqsroxRrOEmISTExHrpFRtbRygMLQa+EUFKlTNtI5MxKYWGKpEwcLiAUJLVA6KrQZE2wKuBmSXICFCOAQ2+/5B3ddtQjboJg/c+X5mnHkO9yRXZs3Fk/s52YLBYFAAgAteJ6sDAACig0IHAENQ6ABgCAodAAxBoQOAIWKs+KYnTpzQjh071LdvX3Xu3NmKCABwwQkEAqqpqVFKSoq6devWZL8lhb5jxw7l5+db8a0B4IK3bNkyDRs2rMl2Swq9b9++ks6EcjgcVkQAgAuOz+dTfn5+qEP/lyWF/vU0i8PhUP/+/a2IAAAXrLNNVXNSFAAMQaEDgCEodAAwBIUOAIZo1UnRyspKPf7442psbNS4ceM0ZcqUsP1z587V1q1bJZ25xvzLL7/Utm3bop8WAHBWEQs9EAiouLhYZWVlstvtGjt2rFwulwYOHBga8+tf/zq0vGTJEu3cubNt0gIAzirilEtVVZWSkpKUmJio2NhYud1uVVRUnHW8x+PR6NGjoxoSABBZxCN0v98fdvOP3W5XVVVVs2P37dunvXv3avjw4dFLGCVZ0161OkKbWv27bKsjALBYxCP05l5oZLPZmh3r8XiUkZHB81kAwAIRC93hcMjn84XW/X6/EhISmh27du1aud3u6KUDALRaxEJPTU2V1+tVdXW1Ghoa5PF45HK5mozbs2ePvvrqK1133XVtEhQA0LKIc+gxMTEqKipSQUGBAoGA8vLylJycrJKSEqWkpCgtLU3SmemWzMzMs07HAADaVquuQ3c6nXI6nWHbCgsLw9YffPDB6KUCAJwz7hQFAENQ6ABgCAodAAxBoQOAISh0ADAEhQ4AhqDQAcAQFDoAGIJCBwBDUOgAYAgKHQAMQaEDgCEodAAwBIUOAIag0AHAEBQ6ABiCQgcAQ1DoAGAICh0ADEGhA4AhKHQAMESrCr2yslIZGRlKT09XaWlps2PWrl2rzMxMud1uTZs2LaohAQCRxUQaEAgEVFxcrLKyMtntdo0dO1Yul0sDBw4MjfF6vSotLdXy5csVHx+vL7/8sk1DAwCainiEXlVVpaSkJCUmJio2NlZut1sVFRVhY1566SXl5+crPj5ektSnT5+2SQsAOKuIhe73++VwOELrdrtdfr8/bIzX69Wnn36qiRMnavz48aqsrIx+UgBAiyJOuQSDwSbbbDZb2HogENBnn32mJUuWyOfzKT8/X2vWrFFcXFz0kgIAWhTxCN3hcMjn84XW/X6/EhISwsbY7XalpaWpS5cuSkxM1BVXXCGv1xv1sACAs4tY6KmpqfJ6vaqurlZDQ4M8Ho9cLlfYmFtuuUVbt26VJB06dEher1eJiYltkxgA0KyIUy4xMTEqKipSQUGBAoGA8vLylJycrJKSEqWkpCgtLU033nij3nnnHWVmZqpz586aMWOGevXq1R75AQD/FbHQJcnpdMrpdIZtKywsDC3bbDbNmjVLs2bNim46AECrcacoABiCQgcAQ1DoAGAICh0ADEGhA4AhKHQAMASFDgCGoNABwBAUOgAYgkIHAENQ6ABgCAodAAxBoQOAISh0ADAEhQ4AhqDQAcAQFDoAGIJCBwBDUOgAYAgKHQAM0apCr6ysVEZGhtLT01VaWtpk/yuvvKLhw4crOztb2dnZWrFiRdSDAgBaFhNpQCAQUHFxscrKymS32zV27Fi5XC4NHDgwbFxmZqaKioraLCgAoGURj9CrqqqUlJSkxMRExcbGyu12q6Kioj2yAQDOQcRC9/v9cjgcoXW73S6/399k3JtvvqmsrCz98pe/1P79+6ObEgAQUcRCDwaDTbbZbLaw9ZtvvlkbNmzQ6tWrNWLECM2cOTN6CQEArRKx0B0Oh3w+X2jd7/crISEhbEyvXr0UGxsrSRo/frw++uijKMcEAEQSsdBTU1Pl9XpVXV2thoYGeTweuVyusDEHDhwILW/YsEEDBgyIflIAQIsiXuUSExOjoqIiFRQUKBAIKC8vT8nJySopKVFKSorS0tK0ZMkSbdiwQZ07d1Z8fLzmzZvXHtkBAN8QsdAlyel0yul0hm0rLCwMLU+bNk3Tpk2LbjIAwDnhTlEAMASFDgCGoNABwBAUOgAYgkIHAENQ6ABgCAodAAxBoQOAISh0ADAEhQ4AhqDQAcAQFDoAGIJCBwBDUOgAYAgKHQAMQaEDgCEodAAwBIUOAIag0AHAEBQ6ABiCQgcAQ7Sq0CsrK5WRkaH09HSVlpaeddzrr7+uQYMGafv27VELCABonYiFHggEVFxcrMWLF8vj8WjNmjXatWtXk3FHjx7VkiVLdO2117ZJUABAyyIWelVVlZKSkpSYmKjY2Fi53W5VVFQ0GVdSUqKCggJ17dq1TYICAFoWsdD9fr8cDkdo3W63y+/3h43ZuXOnfD6fbr755ugnBAC0SsRCDwaDTbbZbLbQcmNjo+bNm6eZM2dGNxkA4JxELHSHwyGfzxda9/v9SkhICK3X19fr3//+t+6++265XC598MEHuv/++zkxCgDtLCbSgNTUVHm9XlVXV8tut8vj8eh3v/tdaH+PHj20devW0PqkSZM0Y8YMpaamtk1idEjjX7zf6ght5qUJz1odAYaIWOgxMTEqKipSQUGBAoGA8vLylJycrJKSEqWkpCgtLa09cgIAIohY6JLkdDrldDrDthUWFjY7dsmSJd89FQDgnHGnKAAYgkIHAENQ6ABgCAodAAxBoQOAISh0ADAEhQ4AhqDQAcAQFDoAGIJCBwBDUOgAYAgKHQAMQaEDgCEodAAwBIUOAIag0AHAEBQ6ABiCQgcAQ1DoAGAICh0ADNGqQq+srFRGRobS09NVWlraZP/y5cuVlZWl7Oxs3Xnnndq1a1fUgwIAWhax0AOBgIqLi7V48WJ5PB6tWbOmSWFnZWVp9erVevXVV1VQUKB58+a1WWAAQPMiFnpVVZWSkpKUmJio2NhYud1uVVRUhI255JJLQsvHjx+XzWaLflIAQItiIg3w+/1yOByhdbvdrqqqqibjli1bprKyMp06dUovvPBCdFMCACKKeIQeDAabbGvuCDw/P1/r16/X9OnT9eyzz0YnHQCg1SIWusPhkM/nC637/X4lJCScdbzb7db69eujkw4A0GoRCz01NVVer1fV1dVqaGiQx+ORy+UKG+P1ekPLf//735WUlBT1oACAlkWcQ4+JiVFRUZEKCgoUCASUl5en5ORklZSUKCUlRWlpaVq6dKk2b96smJgYxcXF6be//W17ZAcAfEPEQpckp9Mpp9MZtq2wsDC0/PDDD0c3FQDgnHGnKAAYgkIHAENQ6ABgCAodAAxBoQOAISh0ADAEhQ4AhqDQAcAQFDoAGIJCBwBDUOgAYAgKHQAMQaEDgCEodAAwBIUOAIag0AHAEK16wQUAfFvvZOdZHaFN3fDqy1ZHCOEIHQAMQaEDgCEodAAwBIUOAIZoVaFXVlYqIyND6enpKi0tbbK/rKxMmZmZysrK0uTJk7Vv376oBwUAtCxioQcCARUXF2vx4sXyeDxas2aNdu3aFTZm8ODBevnll7V69WplZGRowYIFbRYYANC8iIVeVVWlpKQkJSYmKjY2Vm63WxUVFWFjhg8fru7du0uShg4dKp/P1zZpAQBnFbHQ/X6/HA5HaN1ut8vv9591fHl5uUaNGhWddACAVot4Y1EwGGyyzWazNTv21Vdf1Y4dO7R06dLvngwAcE4iFrrD4QibQvH7/UpISGgybtOmTVq0aJGWLl2q2NjY6KYEAEQUccolNTVVXq9X1dXVamhokMfjkcvlChuzc+dOFRUV6dlnn1WfPn3aLCwA4OwiHqHHxMSoqKhIBQUFCgQCysvLU3JyskpKSpSSkqK0tDQ98cQTOnbsmAoLCyVJ/fr106JFi9o8PADg/2vVw7mcTqecTmfYtq/LW5L+9Kc/RTUUAODccacoABiCQgcAQ1DoAGAICh0ADEGhA4AhKHQAMASFDgCGoNABwBAUOgAYgkIHAENQ6ABgCAodAAxBoQOAISh0ADAEhQ4AhqDQAcAQFDoAGIJCBwBDUOgAYAgKHQAM0apCr6ysVEZGhtLT01VaWtpk/3vvvaecnBxdffXVev3116MeEgAQWcRCDwQCKi4u1uLFi+XxeLRmzRrt2rUrbEy/fv00b948jR49us2CAgBaFhNpQFVVlZKSkpSYmChJcrvdqqio0MCBA0Nj+vfvL0nq1IkZHACwSsQG9vv9cjgcoXW73S6/39+moQAA5y5ioQeDwSbbbDZbm4QBAHx7EQvd4XDI5/OF1v1+vxISEto0FADg3EUs9NTUVHm9XlVXV6uhoUEej0cul6s9sgEAzkHEQo+JiVFRUZEKCgqUmZmp22+/XcnJySopKVFFRYWkMydOR40apddff12zZ8+W2+1u8+AAgHARr3KRJKfTKafTGbatsLAwtHzNNdeosrIyuskAAOeE6wwBwBAUOgAYgkIHAENQ6ABgCAodAAxBoQOAISh0ADAEhQ4AhqDQAcAQFDoAGIJCBwBDUOgAYAgKHQAMQaEDgCEodAAwBIUOAIag0AHAEBQ6ABiCQgcAQ1DoAGAICh0ADNGqQq+srFRGRobS09NVWlraZH9DQ4Meeughpaena9y4cdq7d2/UgwIAWhax0AOBgIqLi7V48WJ5PB6tWbNGu3btChuzYsUKxcXFad26dbrnnnu0cOHCNgsMAGheTKQBVVVVSkpKUmJioiTJ7XaroqJCAwcODI3ZsGGDfvGLX0iSMjIyVFxcrGAwKJvN1uzXDAQCkiSfz/edf4DWOnXsULt9LyuY/ldRQ+1xqyO0GdM/u4MNDVZHaFPt+fl93Zlfd+j/iljofr9fDocjtG6321VVVdVkTL9+/c58wZgY9ejRQ7W1terdu3ezX7OmpkaSlJ+f34ofAa2RtmG+1RHwLaU9lWZ1BHwXae3/+dXU1CgpKanJ9oiFHgwGm2z73yPv1oz5ppSUFC1btkx9+/ZV586dI0UAAOjMkXlNTY1SUlKa3R+x0B0OR9jUiN/vV0JCQpMx+/fvl8Ph0OnTp3XkyBH17NnzrF+zW7duGjZsWGt/BgDAfzV3ZP61iCdFU1NT5fV6VV1drYaGBnk8HrlcrrAxLpdLK1eulCS98cYbGj58eItH6ACA6LMFm5sv+R8bN27U3LlzFQgElJeXp/vvv18lJSVKSUlRWlqaTp48qV/96lf6+OOPFR8fr6eeeip0EhUA0D5aVegAgPMfd4oCgCEodAAwBIUOAIaIeNkiIrvvvvsijunZs6fmz+fmn/NNWVlZxDHdu3fXxIkT2yENzhWfXzhOikbBrbfeqscee+ys+4PBoIqLi+XxeNoxFVpj5MiREX/ZV69erTfeeKOdEuFc8PmF4wg9Ch566CFdf/31LY75+c9/3k5pcC7GjBkTeg7R2Rw7dqyd0uBc8fmF4wgdAAzBEXoUfPLJJ/r+978vSTp16pSee+45VVVV6aqrrtL999+v7t27W5wQrbVt2zZt375dycnJGjlypNVx0ApvvfWW1q9fL7/fL5vNpoSEBKWlpWnUqFFWR2t3HKFHQU5OTujRB/Pnz9fhw4eVm5ur9evX6/Dhw3riiScsToizGTt2rMrLyyVJL730kpYtW6b09HS9/fbbcrlcmjJlisUJ0ZLHH39cXq9Xd9xxh+x2u6Qzz5tatWqVkpKS9PDDD1ucsH1xhB4F3/w3cfPmzSovL1eXLl30wx/+UGPGjLEwGSI5ffp0aPnFF19UWVmZevfurXvvvVcTJkyg0M9zlZWVzZ7wzMzMVEZGhgWJrEWhR8GRI0e0bt06NTY2qqGhQV26dJF05hHCPKTs/NbY2Ki6ujo1NjYqGAyGnuF/0UUX8WjnC0BsbKyqqqp0zTXXhG3fvn27unbtalEq61DoUXD99ddrw4YNkqShQ4fq4MGDuvTSS1VTU6NevXpZnA4tOXr0qHJzc0Nv2KqpqVHfvn1VX1/f7HP+cX6ZP3++5syZo/r6+tCLePbv369LLrlE8+bNszhd+2MOHWjG8ePHdfDgQZ4aeoGoqamR3+9XMBiUw+FQ3759rY5kCQq9jX19xAcAbY1nubSx3/zmN1ZHwLf0s5/9zOoI+A5ycnKsjtDuOEIHzuLAgQNNXrcInM8o9Dbw2Wef6ZNPPtGAAQM0cOBAq+OglQ4fPiybzab4+Hiro+AcHTx4MOzGoksvvdTqSJag0KNg0qRJKikpUe/evbVq1So9++yzGjZsmKqqqjR+/HhNmjTJ6og4iy+++EILFizQ5s2bFRcXp2AwqKNHj2r48OGaNm2a+vfvb3VEtODjjz/W7NmzdeTIkdCNRT6fT3FxcZo9e7aGDBliccJ2FsR35na7Q8u5ubnBQ4cOBYPBYPDYsWPB0aNHWxULrTB+/Pigx+MJnj59OrTt9OnTwTVr1gTHjRtnYTK0xpgxY4IffPBBk+3//Oc/g1lZWRYkshYnRaMgJiZGfr9f0pkbUi666CJJZ256aGxstDIaIqitrVVmZmbYTUSdO3eW2+3W4cOHLUyG1jh+/LiuvfbaJtuHDh2q48ePW5DIWtxYFAWzZs3Svffeq1tvvVXJycmaPHmyRo4cqffff1+5ublWx0MLhgwZojlz5ignJyd0Y4rP59PKlSs1ePBgi9MhklGjRmnKlCm64447wj6/VatW6cYbb7Q4XftjDj1Kjhw5otWrV8vr9SoQCMhutystLU0DBgywOhpa0NDQoPLyclVUVOjAgQMKBoOy2+1yuVwaN26cYmNjrY6ICDZu3Njk80tLS5PT6bQ6Wruj0AHAEMyho0Nbt25daK780KFDmjlzprKysvTQQw/J5/NZnA6RVFVVadKkSZo+fbr279+vn/zkJxo2bJjy8vK0c+dOq+O1OwodHdpTTz2lnj17SpKKi4s1ePBgPffccxo1apRmzZplcTpEUlxcrIKCAt10002aOHGiJkyYoG3btmn69On6v//7P6vjtTsKHR1aIBAILX/++ee655575HA4lJubq0OHDlmYDK1x+vRpOZ1OjR49WjabTbfddpskacSIETp58qTF6dofhd6Gli1bprVr14a9RAHnlx/96EcqKSnRiRMndP3112v9+vWSpC1btqhHjx4Wp0MkXbt21dtvv63XXntNNpst9Pm9++676tSp49UbJ0Xb0LJly7Rnzx7t27dPixYtsjoOmnHq1CktWrRIL7/8sqQzl7x1795dLpdL06ZN02WXXWZxQrTkk08+0YIFC2Sz2TRr1iwtX75cq1atkt1u16OPPqof/OAHVkdsVxQ68F9HjhzR6dOneSkJLlgUepR8/vnnWrdunfbv36+YmBglJSVp9OjR/NkOoN10vEmmNvDnP/9Zs2fP1smTJ7Vjxw6dOHFCPp9PEyZM0NatW62OB6CD4Ag9CrKysrRq1Sp17txZx48f15QpU7RkyRJ98cUXeuCBB7Rq1SqrIwLoADhCj5KvL39raGhQfX29JOmyyy7jChfAAtu3bw89MK8j4eFcUTB27Fjl5eVp6NCheu+99/TTn/5U0pk7D3lZwoXp9ttvlyTl5+frrrvusjgNztXSpUv1r3/9S9/73vf09NNPWx2n3TDlEiX/+c9/tHv3biUnJ/NALkPU1tbqww8/1E033WR1FHxLR48e1SWXXGJ1jHZDobex+vp6XXzxxVbHADqE+vp6eb1eJSYmKi4uzuo47Y459DbmdrutjoAWlJeXh5Z9Pp8mT56sYcOGaeLEifr0008tTIbWmDNnTmh527Ztcrvdmj9/vrKysrRx40brglmEOfQoKCsra3Z7MBjUsWPH2jkNzsWyZcs0duxYSdK8efN0++23q6ysTBUVFZozZ45eeOEFixOiJR9++GFouaSkRM8884yGDBmi6upqFRYWdrhnonOEHgVPPvmk6urqVF9fH/bfsWPHeAXdBcTr9WrixInq1KmT0tPTVVdXZ3UknIP6+vrQS6ETExM75O8eR+hRMGTIEN1yyy1KSUlpsm/FihUWJEJr+Xw+PfbYYwoGgzp06JBOnTqlLl26SBKXnF4A9uzZo6ysLEnS3r17VVdXp/j4eDU2NnbIz4+TolGwZ88e9ezZU717926y7+DBg7r00kstSIXWWLlyZdi6y+VSfHy8ampqtGTJEk2dOtWiZGiNffv2ha0nJCSoS5cuOnTokLZt26Zbb73VomTWoNDbyJdffqk+ffpYHQNAB8IcehQcPnw47L/a2lqNGzdOdXV1odeb4cLzt7/9zeoI+A5+//vfWx2h3TGHHgXDhw9v8txsv9+vnJwc2Ww2VVRUWJQM38X27dt18803Wx0D39LXJ0g7EqZcouD555/Xpk2bNGPGDA0aNEjSmbnYDRs2WJwMQEdCoUeJz+fT3Llz1a9fPz344IPKzs7myPwC8MUXX6hPnz7q2rWrgsGgXnnlFe3cuVMDBgzQ+PHjFRPDH7Hns9OnT6u8vFzr1q3TgQMHZLPZlJCQoLS0NI0dOzZ0xVJHQaFH2YYNG7Ro0SLt27dP77zzjtVxEMHo0aO1YsUKde/eXQsWLFB1dbXS0tK0ZcsWSWduNsL5a+rUqerRo4dycnLkcDgknTm4Wrlyperq6jrUg7kk5tCjzuVy6cc//rE+//xzq6OgFRobG9W9e3dJ0ubNm1VeXq5OnTopOztbY8aMsTgdIvnoo4/0xhtvhG1zOBwaOnSoMjIyLEplHa5yaQPdunXTVVddJenM/3A4f/Xr10+bN2+WJF1++eXav3+/pDNPWsT5Lz4+Xq+99lrYXaGNjY1au3Zth3w4F1Mubezhhx/WY489ZnUMnMX+/fs1Y8YMNTY2qkePHnr//fc1ePBgffXVV5o5c6ZGjBhhdUS0YO/evVq4cKG2bNmi+Ph4BYNBffXVVxo+fLimTZumxMREqyO2KwodkLR79259+umnCgQCcjgcSk1NVadO/AF7IamtrVUwGGz2ju2OgkKPkiNHjuitt96S3+8PnWkfOXJkh/yz70ISDAZls9m+8xhYg6uUwnEIEgWrVq1STk6Otm7dquPHj+vYsWPasmWLcnNzeUH0ee7uu+8OvdD7mxoaGrR582bNnDmzyfNecP6YMmVKaP584cKF2rhxo6655hpt375djzzyiMXp2h9H6FGQkZGhFStWNDkar6ur0/jx45uchcf54+TJkyovL9fq1au1d+9excXF6eTJk2psbNQNN9yg/Px8DR482OqYOIvMzEytXbtWkpSbmxu6SkmSxowZo7/+9a9Wxmt3HevvkTbU3J/knTp1Ev9ent+6du2q/Px85efn69SpU6qtrVW3bt2YKrtAfH2V0ogRI0JXKV1++eUd9ioljtCjYOXKlXrmmWd0ww03qF+/fpLOzO1t2rRJDzzwgHJzcy1OCJiJq5TCUehRUldXp7ffflt+v1/BYFAOh0MjR45UfHy81dEA43GV0hkUehRwpQRgDX73wnWe883XZuNbufvuu1VfX6++ffuqR48eoe0NDQ1699139Yc//EFHjhzh5BoQZfzuheMIPQqau1LixIkTCgaDXCkBtCGuUgpHoUcZV0oA1uB3j0IHAGN0vNPAAGAoCh0ADEGhwxiDBg1SfX19s/uys7N14sSJ8yIL0Fa49R8dwquvvmp1BKDNUegwyh//+Ee98847qq2t1dSpU0OvIRs0aJD+8Y9/6OKLL5bL5VJ2drY2bdqkmpoa3XvvvbrrrrskqcV9e/bs0dy5c1VbW6tTp05p8uTJysvLkyS9+eabevLJJ9WzZ0+NGjXKmh8eHR6FDqPYbDb95S9/0Z49e3TnnXdq2LBh6tOnT5NxJ06c0Isvvqi9e/cqKytLOTk5uvjii8+6r2vXrpo+fboWLFigAQMG6OjRo8rLy9PQoUPVs2dPPfLII1q+fLmuvPJKPffcc+39YwOSKHQYZty4cZKkK6+8UldffbU++OADpaWlNRmXmZkpSerfv7/i4uLk8/k0YMCAs+4LBoPavXu3pk6dGvoap06d0p49e9SpUyddffXVuvLKKyVJEyZM0MKFC9v05wSaQ6HDWC09w6Nr166h5c6dOysQCLS4z2azqVevXs3Oxa9fvz6KqYFvj6tcYJSXX35ZkuT1evXxxx/r2muvjcrXveKKK9StW7ewN1Dt3r1bR48e1XXXXaedO3fK6/VKklasWBGV7wmcK47QYZTY2FhNnDhRtbW1Ki4ubnb+/NuIiYnRokWLNHfuXD3//PNqbGxUnz599PTTT6tPnz569NFHdd9996lnz5667bbbovI9gXPFrf8AYAimXADAEBQ6ABiCQgcAQ1DoAGAICh0ADEGhA4AhKHQAMASFDgCG+H8DSHiDZIxx4gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAFBCAYAAABn1FRNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3XtYVHXiBvB35CKmgoowo0V4LS9Q2qOuXVa2IZqVi4h4yVBxlTXtoolGWWk8ZNiWmGylRpj3JMXMEk0FfaTHzLZad8pLrRgB2QwYeAGFgZnz+4MfgxOXwRXmDN/zfp6n55k55wQv3868nfnOOWdUkiRJICKidq+D3AGIiKh1sNCJiATBQiciEgQLnYhIECx0IiJBuMrxSysrK/HDDz/Ax8cHLi4uckQgImp3zGYzSkpKEBAQAA8PjwbrZSn0H374ATExMXL8aiKidm/btm0YMWJEg+WyFLqPjw+A2lAajUaOCERE7Y7BYEBMTIy1Q/9IlkKvm2bRaDS444475IhARNRuNTVVzQ9FiYgEwUInIhIEC52ISBAsdCIiQdgt9CVLluD+++9HeHh4o+slScLy5csREhKCiIgInDp1qtVDEhGRfXYLfcKECUhPT29yfW5uLvLz83Hw4EG8+uqrSExMbM18RETUQnYLfeTIkfDy8mpyfU5ODsaPHw+VSoVhw4bhypUrKC4ubtWQRERk3y2fh240Gm0uDtJoNDAajfD19b3VH21XxKI9bf477PksJVLuCEREAFrhQ9HGvvBIpVLd6o8lIqKbdMuFrtFoYDAYrM8NBoNDjs6JiMjWLRe6VqvFJ598AkmScPLkSXTt2pWFTkQkA7tz6PHx8fj6669RVlaGMWPG4JlnnkFNTQ0AYOrUqQgKCsLRo0cREhKCTp06ITk5uc1DExFRQ3YLfdWqVc2uV6lUeOWVV1otEBER/W94pSgRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCSIFhV6bm4udDodQkJCkJaW1mD9hQsXMH36dIwfPx4RERE4evRoqwclIqLmudrbwGw2IykpCRs2bIBarcbEiROh1WoxYMAA6zZr167F2LFj8fjjj+PcuXOYM2cODh8+3KbBiYjIlt0jdL1eD39/f/j5+cHd3R1hYWHIycmx2UalUqG8vBwAcPXqVfj6+rZNWiIiapLdI3Sj0QiNRmN9rlarodfrbbZ5+umnMXv2bGzduhXXr1/Hhg0bWj8pERE1y+4RuiRJDZapVCqb51lZWYiKikJubi7S0tKQkJAAi8XSeimJiMguu4Wu0WhgMBisz41GY4MplczMTIwdOxYAMHz4cFRVVaGsrKyVoxIRUXPsFnpgYCDy8/NRWFgIk8mErKwsaLVam2169eqF48ePAwDy8vJQVVWFHj16tE1iIiJqlN05dFdXVyxbtgxxcXEwm82Ijo7GwIEDkZqaioCAAAQHB+OFF17Ayy+/jI0bN0KlUuH1119vMC1DRERty26hA0BQUBCCgoJsli1YsMD6eMCAAcjIyGjdZEREdFN4pSgRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBa9I1F5PwmfzRP7gjYMWWt3BGIFI1H6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJokWFnpubC51Oh5CQEKSlpTW6zb59+xAaGoqwsDAsWrSoVUMSEZF9du+HbjabkZSUhA0bNkCtVmPixInQarUYMGCAdZv8/HykpaVh+/bt8PLywu+//96moYmIqCG7R+h6vR7+/v7w8/ODu7s7wsLCkJOTY7PNjh07EBMTAy8vLwCAt7d326QlIqIm2S10o9EIjUZjfa5Wq2E0Gm22yc/Px88//4zHHnsMkydPRm5ubusnJSKiZtmdcpEkqcEylUpl89xsNuOXX37Bli1bYDAYEBMTg71798LT07P1khIRUbPsHqFrNBoYDAbrc6PRCF9fX5tt1Go1goOD4ebmBj8/P/Tt2xf5+fmtHpaIiJpmt9ADAwORn5+PwsJCmEwmZGVlQavV2mzzyCOP4MSJEwCA0tJS5Ofnw8/Pr20SExFRo+xOubi6umLZsmWIi4uD2WxGdHQ0Bg4ciNTUVAQEBCA4OBh//vOfcezYMYSGhsLFxQUJCQno3r27I/ITEdH/s1voABAUFISgoCCbZQsWLLA+VqlUWLJkCZYsWdK66YiIqMV4pSgRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBY6EREgmChExEJgoVORCQIFjoRkSBaVOi5ubnQ6XQICQlBWlpak9t9/vnnuPvuu/H999+3WkAiImoZu4VuNpuRlJSE9PR0ZGVlYe/evTh37lyD7crLy7Flyxbce++9bRKUiIiaZ7fQ9Xo9/P394efnB3d3d4SFhSEnJ6fBdqmpqYiLi0PHjh3bJCgRETXPbqEbjUZoNBrrc7VaDaPRaLPN6dOnYTAY8PDDD7d+QiIiahG7hS5JUoNlKpXK+thisWDFihV4/vnnWzcZERHdFLuFrtFoYDAYrM+NRiN8fX2tzysqKvDTTz9hxowZ0Gq1OHnyJObNm8cPRomIHMzV3gaBgYHIz89HYWEh1Go1srKykJKSYl3ftWtXnDhxwvp8+vTpSEhIQGBgYNskJiKiRtktdFdXVyxbtgxxcXEwm82Ijo7GwIEDkZqaioCAAAQHBzsiJxER2WG30AEgKCgIQUFBNssWLFjQ6LZbtmy59VREt+BYZLTcEfDgnl1yRyAF4pWiRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFzlDkBEbSdp0WdyR8CylAi5IygGj9CJiATBQiciEgQLnYhIEC0q9NzcXOh0OoSEhCAtLa3B+g0bNiA0NBQRERGIjY3Fr7/+2upBiYioeXYL3Ww2IykpCenp6cjKysLevXtx7tw5m20GDx6MXbt24bPPPoNOp8Obb77ZZoGJiKhxdgtdr9fD398ffn5+cHd3R1hYGHJycmy2GT16NDp16gQAGDZsGAwGQ9ukJSKiJtktdKPRCI1GY32uVqthNBqb3D4zMxNjxoxpnXRERNRids9DlySpwTKVStXotnv27MEPP/yArVu33noyIiK6KXYLXaPR2EyhGI1G+Pr6Ntjuyy+/xLp167B161a4u7u3bkoiIrLL7pRLYGAg8vPzUVhYCJPJhKysLGi1WpttTp8+jWXLlmHt2rXw9vZus7BERNQ0u0forq6uWLZsGeLi4mA2mxEdHY2BAwciNTUVAQEBCA4OxhtvvIFr165hwYIFAIBevXph3bp1bR6eiIjqteheLkFBQQgKCrJZVlfeALBx48ZWDUVERDePV4oSEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmChU5EJAgWOhGRIFjoRESCYKETEQmiRYWem5sLnU6HkJAQpKWlNVhvMpnw7LPPIiQkBJMmTUJRUVGrByUioubZLXSz2YykpCSkp6cjKysLe/fuxblz52y22blzJzw9PXHo0CHMnDkTK1eubLPARETUOFd7G+j1evj7+8PPzw8AEBYWhpycHAwYMMC6zeHDh/H0008DAHQ6HZKSkiBJElQqVaM/02w2AwAMBsMtha++VnpL/35rcJZ3I6ay63JHcJqxuGgyyR3BacainK8RodR1Zl2H/pHdQjcajdBoNNbnarUaer2+wTa9evWq/YGurujatSvKysrQo0ePRn9mSUkJACAmJqYFf4JzCz78utwRnEbwW8FyR3AewRyLOnsOJ8sdQTglJSXw9/dvsNxuoUuS1GDZH4+8W7LNjQICArBt2zb4+PjAxcXFXgQiIkLtkXlJSQkCAgIaXW+30DUajc3UiNFohK+vb4NtfvvtN2g0GtTU1ODq1avo1q1bkz/Tw8MDI0aMaOnfQERE/6+xI/M6dj8UDQwMRH5+PgoLC2EymZCVlQWtVmuzjVarxe7duwEABw4cwOjRo5s9QiciotankhqbL/mDo0ePIjk5GWazGdHR0Zg3bx5SU1MREBCA4OBgVFVV4bnnnsOZM2fg5eWFt956y/ohKhEROUaLCp2IiJwfrxQlIhIEC52ISBAsdCIiQdg9bVEUy5cvt7tN586dsXDhQgekkdfcuXPtbtOtWze8/rr4F01xv6jH/aJeex0LxRR6Tk4O5s+f3+w2aWlpinjhnj9/vtkikyQJSUlJDkwkH+4X9bhf1GuvY6GYQo+NjUVUVFSz21y+fNlBaeT17LPPYtSoUc1u89RTTzkojby4X9TjflGvvY4FT1skIhKEYo7QS0tLbW4WtmfPHnz//fcYOHAgJk+erKgrW8+ePYtBgwYBAKqrq/H+++9Dr9fjrrvuwrx589CpUyeZEzoO94t63C/qXb16Fe+99x6ys7NRVlYGAOjRoweCg4MxZ84ceHp6ypywcYo5y2X27NnWx2vWrMGnn36KoUOH4tixY1ixYoWMyRxvyZIl1scpKSkoKCjArFmzUFlZiVdeeUXGZI7H/aIe94t6zz77LDw9PbFlyxacOHECJ06cwObNm+Hp6YkFCxbIHa9JijlCv3Fm6dChQ9i2bRtuu+02hIeHY8KECTImc7wbx+L48ePIzMyEm5sbRo4ciXHjxsmYzPG4X9TjflGvqKgI69evt1nm4+ODOXPmYNeuXTKlsk8xhV5ZWYnTp0/DYrHAbDbjtttuAwC4ubmhQwfFvFEBUPt28tChQ7BYLDCZTHBzcwNQe8tjJU0xANwvbsT9ot7tt9+O999/H1FRUejZsycA4OLFi/j444+t3/3gjBRT6D4+Pta30F5eXiguLoavry/KysoUd0/2UaNG4fDhwwCAYcOG4eLFi+jZsydKSkrQvXt3mdM5FveLetwv6r311ltIS0vDtGnTUFpa+61P3t7e0Gq1WL16tczpmqb4s1zMZjNMJpOiPvAh+7hfUHukrPeUjXBxceGL9gZ1Xw+odNwvbHG/qHfq1Cm5IzRJ8YUOwO6FJUry0ksvyR3BaXC/qMf9ot727dvljtAkxU+5EBGJgkfoCmWxWGCxWAAAJpMJp06dwqVLl2RORXK6cuWK3BGchslksjmN86uvvsIHH3yAo0ePypjKPsUU+m+//YaFCxfi8ccfx7p161BdXW1d9+STT8qYzPGys7Px0EMPYcyYMcjOzkZMTAz+8Y9/YNy4cdazHJSC+0W90aNHY+bMmdi5c6fiy33ixInWMUhPT8fq1atRWVmJjRs3YuXKlTKna4akEDNnzpQ+/PBD6fTp01JSUpI0ZcoUqbS0VJIkSYqMjJQ5nWNFRkZKxcXFUkFBgTR8+HApLy9PkiRJKioqkqKiomRO51jcL+qFh4dLhw8fluLj46VRo0ZJc+fOlfbu3Stdv35d7mgOFxYWZn0cFRVlHYPq6mopPDxcrlh2KeYIvbS0FFOnTsXgwYOxdOlSTJ06FdOmTUNBQYHiLpoAas+/9vPzQ+/evdGvXz8AtRdTSAr7SIX7RT1XV1c8/PDDSElJwdGjRxEREYH9+/cjKCgIixYtkjueQ3Xp0gU//fQTAKB79+6oqqoCUHs6qzO/RhRzYVFNTQ2qqqrQsWNHAEBkZCR8fHwwe/ZsXL9+XeZ0jmexWNChQwckJydbl5nNZpspByXgflHvxqLy8PBAaGgoQkNDcfXqVWRnZ8uYzPESExOxePFiDBo0CN7e3oiOjsbIkSPx448/4oknnpA7XpNcEhMTE+UO4QjV1dWwWCy4/fbbrcv8/PwwYsQInD17FuPHj5cxnWMNGTIE3bt3h6urK9RqtXX5hQsX0KtXL+sd95SA+0W9iooK3HfffQ2Wd+zYEYMHD5YhkXx69uyJyZMnw8PDA+7u7ggICMDQoUMxf/583HPPPXLHaxJPWySiJv3+++/w9vaWOwa1kGLm0AHgiy++wM6dO1FUVGSzPDMzU6ZE8sjLy0NcXBzmzJmDgoICvPDCCxgxYgQmTpyIvLw8uePJTqfTyR1BFpcuXbL5p6ysDJMmTcLly5cVd0prVFQU1qxZg4KCArmj3BTFHKGnpKTgu+++w5AhQ3DkyBHExsZi+vTpAGr/4+3evVvmhI4TExOD2bNn49q1a0hJScHixYsRGhqKI0eOYNOmTdi0aZPcER1m+PDh1g8/614KlZWV8PDwgEqlwnfffSdnPIcaNGgQevfubbPMaDRCrVZDpVIhJydHpmSOp9VqodPpsH//fvTs2RPh4eEYO3aszRSlU5LvBBvHCg8Pl6qrqyVJkqTLly9LcXFx0muvvSZJkvJOT7vx733kkUds1o0fP97RcWSVlJQkPffcc1JJSYl12cMPPyxjIvmkp6dLs2bNks6ePWtdptSxuPF18K9//Ut65ZVXpAceeECaNm2alJGRIWOy5ilmyqWmpgaurrUn9Xh6emLdunUoLy/H/PnzFXdmh9lstj6eOXOmzTqljcXSpUsxY8YMxMfHY/PmzbBYLIo7XbHO7Nmz8dprr+Hdd9/FihUrUF5ertixuNGIESOQmJiI3Nxc/P3vf8fJkyfljtQkxRT6nXfeia+//tr63MXFBcnJyejbt6/i5o1jYmJQUVFhfVznl19+wf333y9XLNkEBARg48aNAIBp06ZZzzlWIo1Gg3/+85/405/+ZP36OSXq06dPg2UuLi4YM2aMU381oWLm0Ot2TA8Pjwbr6uYJiYqLi3HmzBkEBQXJHUV2lZWVKCgowF133SV3FGohxRQ6EZHoFDPlQkQkOhY6EZEgWOhklZ2djf/85z9yx3AKxcXFMJlMcsdwCt9//z2MRqPcMZyCs4+FYm7O1ZSxY8cCqD3bY9q0aTKnkZder8dPP/2EmpoapKenyx1HVgkJCSgoKIBOp8Pzzz8vdxxZbd26FT/++CP69Onj1N947wjOPhb8UBS1t1DV6/X4y1/+IncUciKSJOHcuXMYOHCg3FGcQnl5Obp06SJ3DKfgrGPBQicUFhbizJkz6N+/P/r37y93HNldunQJ3bp1kzuG7CoqKpCfnw8/Pz94enrKHUdW7WUsOIcOICIiQu4IDnXjV6tlZ2cjNjYWR44cwZNPPomPP/5YxmSOt2bNGuvjc+fOQafTYcKECdBqtU59RWBbuPFO2t988w3CwsLw+uuvIyIiwum/S7O1tdexUMwc+sGDBxtdLkkSSkpKHJxGXhcuXLA+Tk9Px6ZNm+Dn54fS0lLMnDkTEyZMkDGdYx06dMj6P7g33ngDL774IoKCgqDX65GcnIyMjAyZEzrOjR+Ip6am4t1338XQoUNRWFiIBQsWKOpiq/Y6Foop9IULFyIiIqLRe1Mo7VLvG8egpqYGfn5+AIAePXqgQwflvmkrLi62vlDvuecexV72DtROMQwdOhRA7Rd+WCwWmRPJpz2NhWIK/e6778asWbMavYz5yy+/lCGRfM6ePYv77rsPkiShuroaJSUl8PHxgclksrlxlxIUFhZi7ty5AACDwYDr16+jU6dOAGr/Z6ck58+ft04/FhUV4fLly/Dy8oLFYuFYtJOxUMyHot988w169+7d4H7PQO25pYGBgTKkci5XrlxBXl4ehg8fLncUh7nxhm0AMHToUHTu3BkXL17EgQMHbG5eJrpff/3V5rmvry/c3NxQWlqKb775Bo8++qhMyRyvvY6FYgqdiEh0ipkwrampQUZGBmbPno2IiAiMGzcOcXFx2L59u+LuAd6cpUuXyh3BaXz00UdyR3Aab7/9ttwRnIYzj4Vi5tATEhLQtWtXPPPMM9BoNABq50x3796N5557zimv+pLDlClT5I7gNPjmtV7dh4Lk3GOhmCkXnU6HAwcO3PQ6IqL2QjFTLl5eXti/f7/NKUcWiwX79u1z6iu/2oLZbEZGRgZWr16Nb7/91mbdjRfaKIEkSdi3bx/2798PSZJw/PhxLF++HNu2bXPq09PayhdffIGdO3eiqKjIZnlmZqZMieSxYsWKBq+N9kAxR+hFRUVYuXIlvvrqK3h5eUGSJFy5cgWjR4/GokWLrOdiK8FLL72EyspKBAYG4tNPP8XIkSOxZMkSAEBUVBR2794tc0LHSUxMRGlpKUwmE7p06QKTyQStVoujR4/C29sbL7/8stwRHWbVqlX49ttvMWTIEBw5cgSxsbGYPn06AOXtF6NHj0bv3r1RVlaGsWPHIjw8HEOGDJE7ln0O/EJqp1FaWir9/vvvcseQTXh4uPVxdXW19PLLL0tPPfWUVFVVJUVGRsqYzPHqxsJkMkmjRo2SqqqqJEmqHZcbx0kJwsPDperqakmSJOny5ctSXFyc9Nprr0mSJCluv6j7e3/++WfpnXfekUJDQyWdTie9/fbb0vnz52VO1zTFTLkAQF5eHtLS0vDuu+9izZo1SEtLU9wXRAOwOavH1dUVr776KgYNGoQZM2bg2rVrMiZzPBcXFwCAm5sbAgIC4O7uDqB2XJR21WxNTQ1cXWvPk/D09MS6detQXl6O+fPnK+5MsLqrqfv06YOnnnoKWVlZWL16NaqqqjBnzhyZ0zVNMXtsWloa4uPjAQCBgYHWC4ni4+ORlpYmZzSHCwgIQG5urs2yp59+GtHR0Q0uqBBdz549UVFRAQBYv369dXlJSQnc3NzkiiWLO++80+ZCKxcXFyQnJ6Nv376KO/CRGpmJHjRoEBYtWoRDhw7JkKhlFDOHrtPpsHfv3gYvUpPJhPDw8CZv3kXKdO3aNVy/fh3e3t5yR3GYunvXeHh4NFhnNBqhVqsdHUk2FRUV6Ny5s9wxbppijtBVKhWKi4sbLC8pKWn0hl2iKy8vR0FBQYPlZ8+elSGNvEpKSqx33CwtLcXBgwfx66+/KqrMgdoi/2OZr1q1CgAUVeYAGi3zurFwZoq5sOjFF1/EzJkz4e/vj169egGovY1sQUGB4q6O3LdvH5KTk+Ht7Y2amhqsWLEC99xzDwBgyZIlijqbISMjwzrlFhcXh927d2PAgAFISUlBXFwcJk2aJHNCx1m+fLnNc0mSsGfPHuvnKko646e9joViCn3MmDE4cOAA9Ho9jEYjJEmCRqNBYGCg9YMxpXjvvffw8ccfw9fXF3q9HgkJCYiPj8ejjz6quKsjt23bhqysLFRWVkKr1eLgwYPw8fHB5cuXMWPGDEUV+sGDBzFq1Cg89NBD1v0gKyvLqa+MbCvtdSwUU+gA0KFDBwwbNkzuGLKzWCzw9fUFUHvf782bN2Pu3LkwGAyKm35ydXVFp06d0KlTJ/j5+cHHxwdA7YVoShuLffv2ITU1FV988QUSEhKgVqvxzjvvICoqSu5oDtdex0Ixc+jNeeKJJ+SO4FCdO3e2mT/39fXF5s2bkZOTg//+978yJnM8lUplPSXvxrOdqqqqFHelaJcuXfDSSy/hb3/7GxYvXoz169cr7h1bnfY6Foo5y6U5xcXF1iNWJTh79iw8PDzQp08fm+XV1dXYv38/xo0bJ08wGVy4cAG+vr7W86/rGI1G5OXl4YEHHpApmbwkScKHH36If//731i5cqXccWTVnsZCkYV+6dIlqFQqeHl5yR1FFpIk2Z1OaMk2IuBY1ONY1GuvY+GSeOPXWwvswoULWLZsGRITE/Hpp59ix44dWLt2LU6dOoWAgABF3aBrxowZqKiogI+PD7p27WpdbjKZ8PXXX+Odd97B1atXMXjwYBlTOgbHoh7Hol57HQvFHKFPmTIFsbGx0Ol01rNazGYzPv/8c2zatAk7duyQOaHjVFVVITMzE5999hmKiorg6elpnTN+8MEHERMT43Q7alvhWNRrbCwqKyshSRLHop2MhWIK/dFHH23yatDm1omuuroaZWVl8PDwUNS7lMZwLOpxLOq1p7FQTKEvXLgQXl5eiIqKavCNRWVlZUhNTZU5IRHRrVFMoZtMJmRmZiInJwfFxcWQJAlqtRparRaTJk2y3mWPiKi9UkyhExGJTtEXFjn7VV9ERDdD0YXONydEJBJFF3pQUJDcEYiIWo1i5tDb65VfREQtpZgj9BkzZmDLli24cOGCzXKTyYTjx4/j+eefV9R9wIlIPIo5Qm+vV34REbWUYgr9Ru3pyi8iopZSZKETEYlIMXPoRESiY6ETEQmChU7CuPvuu1FRUdHousjISFRWVjpFFqK2oqgviSbl2rNnj9wRiNocC52E8sEHH+DYsWMoKytDfHw8dDodgNoj5u+++w6dO3eGVqtFZGQkvvzyS5SUlGDWrFmYNm0aADS77vz580hOTkZZWRmqq6sRGxuL6OhoAMDBgwexatUqdOvWDWPGjJHnjyfFY6GTUFQqFTIyMnD+/HlMnToVI0aMgLe3d4PtKisr8dFHH6GoqAgRERGIiopC586dm1zXsWMz9PQoAAABbUlEQVRHLF68GG+++Sb69++P8vJyREdHY9iwYejWrRuWLl2K7du3o1+/fnj//fcd/WcTAWChk2AmTZoEAOjXrx+GDBmCkydPIjg4uMF2oaGhAIA77rgDnp6eMBgM6N+/f5PrJElCXl4e4uPjrT+juroa58+fR4cOHTBkyBD069cPQO3XHTr7t8OTmFjoJKzm7s3TsWNH62MXFxeYzeZm16lUKnTv3r3Rufjs7OxWTE30v+NZLiSUXbt2AQDy8/Nx5swZ3Hvvva3yc/v27QsPDw988skn1mV5eXkoLy/H8OHDcfr0aeTn5wMAdu7c2Sq/k+hm8QidhOLu7o7HHnsMZWVlSEpKanT+/H/h6uqKdevWITk5GevXr4fFYoG3tzdWr14Nb29vvPrqq5g7dy66deuGv/71r63yO4luFi/9JyISBKdciIgEwUInIhIEC52ISBAsdCIiQbDQiYgEwUInIhIEC52ISBAsdCIiQfwfMuCI45uhdggAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAFBCAYAAACM4+9OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xt4VPWB//H3kABJgQQSEgYxVUEtV8X+EIjtwho2CZdEEi51XZCLsEjrlkIEKlJKHi7BbQHBqrUR24YuipJCoCIlJLCJD6JWLU0LYldYFiJmJpIbAcLkcn5/pEyICSRiMic55/N6nj7P5Dsn5DPfnnw8+c45ZxyGYRiIiIildDA7gIiItDyVu4iIBancRUQsSOUuImJBKncREQvyNztARUUFf/vb3wgLC8PPz8/sOCIi7UJ1dTWFhYUMHjyYgICABs+bXu5/+9vfmDZtmtkxRETapW3btjFs2LAG46aXe1hYGFAb0Ol0mpxGRKR9KCgoYNq0ad4O/TLTy/3qUozT6eTWW281OY2ISPtyveVsvaEqImJBKncREQtSuYuIWJDKXUTEglTuIiIWpHIXEbGgJk+FPHXqFIsWLfJ+ffbsWRYsWEBCQgKLFi3is88+o0+fPmzatIng4GAMw2Dt2rXk5OQQEBDAM888w6BBg1r1RYiISH1Nlnvfvn3ZvXs3UHu566hRo4iOjiY1NZXIyEjmzZtHamoqqampLFmyhNzcXE6fPk1mZiZ/+ctfSE5OZseOHa3+QkQac3jiZLMj8J3dvzc7gtjQV1qWOXLkCBEREfTp04fs7GwSEhIASEhIICsrC8A77nA4GDp0KGVlZbjd7pZPLiIi1/WVyn3v3r3ExcUBcP78ecLDwwEIDw+nqKgIAJfLVe82Ak6nE5fL1VJ5RUSkGZpd7h6Ph4MHDzJ27NgbbtfYR7I6HI6vnkxERG5as8s9NzeXQYMG0bNnTwBCQ0O9yy1ut5uQkBCg9ki9oKDA+30FBQXeI3wREfGNZpf73r17mTBhgvfrqKgoMjIyAMjIyGDMmDH1xg3D4OjRo3Tr1k3lLiLiY80q98uXL/POO+8QExPjHZs3bx6HDx8mJiaGw4cPM2/ePABGjx5NREQE0dHRrFixgpUrV7ZOchERua5m3fI3MDCQ9957r95Yjx49SEtLa7Ctw+FQoYuImExXqIqIWJDKXUTEglTuIiIWpHIXEbEglbuIiAWp3EVELEjlLiJiQSp3ERELUrmLiFiQyl1ExIJU7iIiFqRyFxGxIJW7iIgFqdxFRCxI5S4iYkEqdxERC1K5i4hYkMpdRMSCVO4iIhakchcRsaBmlXtZWRkLFixg7NixjBs3jj//+c+UlJQwe/ZsYmJimD17NqWlpQAYhsGaNWuIjo4mPj6eY8eOteoLEBGRhppV7mvXruWf/umf+OMf/8ju3bvp168fqampREZGkpmZSWRkJKmpqQDk5uZy+vRpMjMzWb16NcnJya2ZX0REGtFkuZeXl/OnP/2JKVOmANCpUyeCgoLIzs4mISEBgISEBLKysgC84w6Hg6FDh1JWVobb7W7FlyAiIl/WZLmfPXuWkJAQli1bRkJCAsuXL+fSpUucP3+e8PBwAMLDwykqKgLA5XLhdDq93+90OnG5XK0UX0REGtNkuVdVVXH8+HEeeeQRMjIyCAwM9C7BNMYwjAZjDofj66UUEZGvpMlydzqdOJ1O7r33XgDGjh3L8ePHCQ0N9S63uN1uQkJCvNsXFBR4v7+goMB7hC8iIr7RZLmHhYXhdDo5deoUAEeOHKFfv35ERUWRkZEBQEZGBmPGjAHwjhuGwdGjR+nWrZvKXUTEx/ybs9GKFStYvHgxlZWVREREsG7dOmpqali4cCHp6en07t2bzZs3AzB69GhycnKIjo4mMDCQlJSUVn0BIiLSULPKfcCAAezcubPBeFpaWoMxh8PBypUrv34yERG5abpCVUTEglTuIiIWpHIXEbEglbuIiAWp3EVELEjlLiJiQSp3ERELUrmLiFiQyl1ExIJU7iIiFqRyFxGxIJW7iIgFqdxFRCxI5S4iYkEqdxERC1K5i4hYkMpdRMSCVO4iIhakchcRsSCVu4iIBTXrA7KjoqLo0qULHTp0wM/Pj507d1JSUsKiRYv47LPP6NOnD5s2bSI4OBjDMFi7di05OTkEBATwzDPPMGjQoNZ+HSIico1mH7mnpaWxe/dudu7cCUBqaiqRkZFkZmYSGRlJamoqALm5uZw+fZrMzExWr15NcnJyqwQXEZHru+llmezsbBISEgBISEggKyur3rjD4WDo0KGUlZXhdrtbJq2IiDRLs8t9zpw5TJo0iddffx2A8+fPEx4eDkB4eDhFRUUAuFwunE6n9/ucTicul6slM4uISBOateb+2muv0atXL86fP8/s2bPp27fvdbc1DKPBmMPhuPmEIiLylTXryL1Xr14AhIaGEh0dTV5eHqGhod7lFrfbTUhICFB7pF5QUOD93oKCAu8RvoiI+EaT5X7p0iXKy8u9jw8fPsxdd91FVFQUGRkZAGRkZDBmzBgA77hhGBw9epRu3bqp3EVEfKzJZZnz58/zxBNPAFBdXU1cXByjRo1iyJAhLFy4kPT0dHr37s3mzZsBGD16NDk5OURHRxMYGEhKSkrrvgIREWmgyXKPiIhgz549DcZ79OhBWlpag3GHw8HKlStbJp2IiNwUXaEqImJBKncREQtSuYuIWJDKXUTEglTuIiIWpHIXEbEglbuIiAWp3EVELEjlLiJiQSp3ERELUrmLiFiQyl1ExIJU7iIiFqRyFxGxIJW7iIgFqdxFRCxI5S4iYkEqdxERC1K5i4hYkMpdRMSCml3u1dXVJCQk8PjjjwNw9uxZpk6dSkxMDAsXLsTj8QDg8XhYuHAh0dHRTJ06lfz8/NZJLiIi19Xsct+6dSv9+vXzfr1+/XpmzZpFZmYmQUFBpKenA7Bjxw6CgoI4cOAAs2bNYv369S2fWkREbqhZ5V5QUMB///d/M2XKFAAMw+Ddd98lNjYWgMTERLKzswE4ePAgiYmJAMTGxnLkyBEMw2iN7CIich3NKveUlBSWLFlChw61mxcXFxMUFIS/vz8ATqcTl8sFgMvlonfv3gD4+/vTrVs3iouLWyO7iIhcR5PlfujQIUJCQhg8ePANt3M4HACNHqVffU5ERHzDv6kNPvroIw4ePEhubi5XrlyhvLyctWvXUlZWRlVVFf7+/hQUFBAeHg7UHsV//vnnOJ1OqqqquHDhAt27d2/1FyIiInWaPHJ/8sknyc3N5eDBg2zcuJGRI0eyYcMGRowYwf79+wHYtWsXUVFRAERFRbFr1y4A9u/fz8iRI3XkLiLiYzd9nvuSJUv4zW9+Q3R0NCUlJUydOhWAKVOmUFJSQnR0NL/5zW9YvHhxi4UVEZHmaXJZ5lojRoxgxIgRAERERHhPf7xW586dee6551omnYiI3BRdoSoiYkEqdxERC1K5i4hYkMpdRMSCVO4iIhakchcRsSCVu4iIBancRUQsSOUuImJBKncREQtSuYuIWJDKXUTEglTuIiIWpHIXEbEglbuIiAWp3EVELEjlLiJiQSp3ERELUrmLiFiQyl1ExIJU7iIiFuTf1AZXrlxh2rRpeDweqquriY2NZcGCBZw9e5akpCRKS0sZOHAgP/vZz+jUqRMej4elS5dy7NgxunfvzrPPPsutt97qi9ciIiL/0OSRe6dOnUhLS2PPnj1kZGTw9ttvc/ToUdavX8+sWbPIzMwkKCiI9PR0AHbs2EFQUBAHDhxg1qxZrF+/vtVfhIiI1NdkuTscDrp06QJAVVUVVVVVOBwO3n33XWJjYwFITEwkOzsbgIMHD5KYmAhAbGwsR44cwTCM1sovIiKNaNaae3V1NRMnTuSBBx7ggQceICIigqCgIPz9a1d1nE4nLpcLAJfLRe/evQHw9/enW7duFBcXt1J8ERFpTLPK3c/Pj927d5OTk0NeXh6nTp1qsI3D4QBo9Cj96nMiIuIbX+lsmaCgIEaMGMHRo0cpKyujqqoKgIKCAsLDw4Hao/jPP/8cqF3GuXDhAt27d2/h2CIiciNNlntRURFlZWUAVFRU8M4779CvXz9GjBjB/v37Adi1axdRUVEAREVFsWvXLgD279/PyJEjdeQuIuJjTZ4K6Xa7eeqpp6iursYwDMaOHcuDDz7InXfeyaJFi9i0aRMDBgxg6tSpAEyZMoUlS5YQHR1NcHAwzz77bKu/CBERqa/Jcu/fvz8ZGRkNxiMiIrynP16rc+fOPPfccy2TTkREboquUBURsSCVu4iIBancRUQsSOUuImJBKncREQtSuYuIWJDKXUTEglTuIiIWpHIXEbEglbuIiAWp3EVELEjlLiJiQSp3ERELUrmLiFiQyl1ExIJU7iIiFqRyFxGxIJW7iIgFqdxFRCxI5S4iYkFNlvvnn3/Oo48+yrhx45gwYQJpaWkAlJSUMHv2bGJiYpg9ezalpaUAGIbBmjVriI6OJj4+nmPHjrXuKxARkQaaLHc/Pz+eeuop9u3bx+uvv86rr77Kp59+SmpqKpGRkWRmZhIZGUlqaioAubm5nD59mszMTFavXk1ycnJrvwYREfmSJss9PDycQYMGAdC1a1f69u2Ly+UiOzubhIQEABISEsjKygLwjjscDoYOHUpZWRlut7sVX4KIiHzZV1pzz8/P5+OPP+bee+/l/PnzhIeHA7X/ASgqKgLA5XLhdDq93+N0OnG5XC0YWUREmtLscr948SILFizg6aefpmvXrtfdzjCMBmMOh+Pm0omIyE1pVrlXVlayYMEC4uPjiYmJASA0NNS73OJ2uwkJCQFqj9QLCgq831tQUOA9whcREd9ostwNw2D58uX07duX2bNne8ejoqLIyMgAICMjgzFjxtQbNwyDo0eP0q1bN5W7iIiP+Te1wYcffsju3bu5++67mThxIgBJSUnMmzePhQsXkp6eTu/evdm8eTMAo0ePJicnh+joaAIDA0lJSWndVyAiIg00We7Dhg3jk08+afS5q+e8X8vhcLBy5cqvn0xERG6arlAVEbEglbuIiAWp3EVELEjlLiJiQSp3ERELUrmLiFiQyl1ExIJU7iIiFqRyFxGxIJW7iIgFqdxFRCxI5S4iYkEqdxERC1K5i4hYkMpdRMSCVO4iIhakchcRsSCVu4iIBancRUQsSOUuImJBTZb7smXLiIyMJC4uzjtWUlLC7NmziYmJYfbs2ZSWlgJgGAZr1qwhOjqa+Ph4jh071nrJRUTkupos90mTJrFly5Z6Y6mpqURGRpKZmUlkZCSpqakA5Obmcvr0aTIzM1m9ejXJycmtElpERG6syXK///77CQ4OrjeWnZ1NQkICAAkJCWRlZdUbdzgcDB06lLKyMtxudyvEFhGRG7mpNffz588THh4OQHh4OEVFRQC4XC6cTqd3O6fTicvlaoGYIiLyVbToG6qGYTQYczgcLfkjRESkGW6q3ENDQ73LLW63m5CQEKD2SL2goMC7XUFBgfcIX0REfOemyj0qKoqMjAwAMjIyGDNmTL1xwzA4evQo3bp1U7mLiJjAv6kNkpKSeP/99ykuLmbUqFH88Ic/ZN68eSxcuJD09HR69+7N5s2bARg9ejQ5OTlER0cTGBhISkpKq78AERFpqMly37hxY6PjaWlpDcYcDgcrV678+qlERORrabLcRUSs5sPMJWZH4P/F/LxV/33dfkBExIJU7iIiFqRyFxGxIJW7iIgFqdxFRCxI5S4iYkEqdxERC1K5i4hYkMpdRMSCVO4iIhakchcRsSDL3Fsm/sndZkfgDxsmmh1BRATQkbuIiCVZ5shd6nzv9e+bHYE3Hv6l2RFEbE1H7iIiFqRyFxGxIJW7iIgFac1dxCZWPfkHsyPw0w3xZkewDR25i4hYkMpdRMSCWqXcc3NziY2NJTo6mtTU1Nb4ESIicgMtXu7V1dWsWrWKLVu2sHfvXt58800+/fTTlv4xIiJyAy3+hmpeXh633XYbERERAEyYMIHs7GzuvPPORrevrq4GoKCg4Gv93MpLRV/r+1tCfn6+2REA8BRfNjtCm5mLLzwesyO0mbko1++IV+H5S2ZH+NpzcbUzr3bol7V4ubtcLpxOp/frXr16kZeXd93tCwsLAZg2bVpLR/G5MQefMTtCmzHm2TFmR2g7xmgurtp9MMXsCG1IVov8K4WFhdx2220Nxlu83A3DaDDmcDiuu/3gwYPZtm0bYWFh+Pn5tXQcERFLqq6uprCwkMGDBzf6fIuXu9PprLfE4nK5CA8Pv+72AQEBDBs2rKVjiIhYXmNH7Fe1+BuqQ4YM4fTp05w9exaPx8PevXuJiopq6R8jIiI30OJH7v7+/vz0pz9l7ty5VFdXM3nyZO66666W/jEiInIDDqOxRXIREWnXdIWqiIgFqdxFRCxI5S4iYkG2vOXvmjVrmtymS5cuLFq0yAdpzDV//vwmt+nevTvPPGP9C7S0X9TRflGnvc6FLcs9OzubBQsW3HCb1NRUW/wSnzp16oalZhgGq1at8mEi82i/qKP9ok57nQtblvvMmTNJTEy84TalpaU+SmOuhQsXMnz48Btu88QTT/gojbm0X9TRflGnvc6FToUUEbEgWx65FxUVERIS4v169+7d/PWvf+Wuu+7ie9/73g3vhWM1J06coH///gBUVlby8ssvk5eXx9133833v/99AgMDTU7oO9ov6mi/qHPhwgV+9atfkZWVRXFxMQAhISGMGTOGefPmERQUZHLCxtnybJk5c+Z4H7/44ovs2bOHQYMGcfjwYdatW2diMt9btmyZ9/GGDRs4c+YMjz32GBUVFaxcudLEZL6n/aKO9os6CxcuJCgoiN/97ne89957vPfee2zdupWgoCB+9KMfmR3vumx55H7tStSBAwfYtm0b3/jGN4iLi2PSpEkmJvO9a+fiyJEjpKen07FjR+6//34eeughE5P5nvaLOtov6uTn5/PKK6/UGwsLC2PevHn8/ve/NylV02xZ7hUVFRw/fpyamhqqq6v5xje+AUDHjh3p0MFef8xcuHCBAwcOUFNTg8fjoWPHjkDtbZrttAwB2i+upf2iTp8+fXj55ZdJTEykZ8+eAHzxxRfs3LmT3r17m5zu+mxZ7mFhYd4/s4ODg3G73YSHh1NcXGy7e8oPHz6cgwcPAjB06FC++OILevbsSWFhIT169DA5nW9pv6ij/aLOs88+S2pqKtOnT6eoqPbTrEJDQ4mKimLTpk0mp7s+nS1zjerqajwej63eLJKmab+Q9shef2s2wc/PT7/A17j6EYh2p/2iPu0XdY4dO2Z2hOtSuX9JUxex2Mny5cvNjtBmaL+oo/2izmuvvWZ2hOvSsoyIiAXpyF2oqamhpqYGAI/Hw7FjxygpKTE5lZiprKzM7AhthsfjqXdq6Lvvvsuvf/1rcnJyTEzVNFuW++eff86iRYv4t3/7N1566SUqKyu9z/3gBz8wMZnvZWVl8d3vfpdRo0aRlZXFtGnT+M///E8eeugh79kSdqH9os7IkSOZNWsWO3bssH3RT5kyxTsHW7ZsYdOmTVRUVPDb3/6W9evXm5zuBgwbmjVrlvHqq68ax48fN1atWmU8/PDDRlFRkWEYhjFx4kST0/nWxIkTDbfbbZw5c8a47777jJMnTxqGYRj5+flGYmKiyel8S/tFnbi4OOPgwYNGUlKSMXz4cGP+/PnGm2++aVy+fNnsaD43YcIE7+PExETvHFRWVhpxcXFmxWqSLY/ci4qKeOSRRxgwYAArVqzgkUceYfr06Zw5c8Z2F2hA7fndERER3HLLLfTt2xeovXDDsNnbMdov6vj7+/Pggw+yYcMGcnJyiI+PZ9++fYwePZonn3zS7Hg+1bVrV/7+978D0KNHD65cuQLUniLbln9HbHkRU1VVFVeuXKFz584ATJw4kbCwMObMmcPly5dNTud7NTU1dOjQgZSUFO9YdXV1vWUJO9B+Uefa0goICGD8+PGMHz+eCxcukJWVZWIy30tOTmbx4sX079+f0NBQJk+ezP33388nn3zC448/bna86/JLTk5ONjuEr1VWVlJTU0OfPn28YxEREQwbNowTJ06QkJBgYjrfGjhwID169MDf359evXp5x8+dO0fv3r29dwa0A+0XdS5evMi3v/3tBuOdO3dmwIABJiQyT8+ePfne975HQEAAnTp1YvDgwQwaNIgFCxZwzz33mB3vunQqpIg0y/nz5wkNDTU7hjSTLdfcAd5++2127NhBfn5+vfH09HSTEpnj5MmTzJ07l3nz5nHmzBmeeuophg0bxpQpUzh58qTZ8UwXGxtrdgRTlJSU1PtfcXExU6dOpbS01HanySYmJvLiiy9y5swZs6N8JbY8ct+wYQMfffQRAwcO5NChQ8ycOZNHH30UqP0/cteuXSYn9J1p06YxZ84cLl26xIYNG1i8eDHjx4/n0KFDpKWlkZaWZnZEn7nvvvu8b5xe/bWoqKggICAAh8PBRx99ZGY8n+rfvz+33HJLvTGXy0WvXr1wOBxkZ2eblMz3oqKiiI2NZd++ffTs2ZO4uDjGjRtXbxmzTTLvRB3zxMXFGZWVlYZhGEZpaakxd+5cY+3atYZh2O+Ut2tf77/8y7/Uey4hIcHXcUy1atUqY8mSJUZhYaF37MEHHzQxkXm2bNliPPbYY8aJEye8Y3adi2t/D/70pz8ZK1euNB544AFj+vTpxvbt201MdmO2XJapqqrC37/2RKGgoCBeeuklysvLWbBgge3OEKmurvY+njVrVr3n7DYXK1asYMaMGSQlJbF161ZqampsdwrkVXPmzGHt2rW88MILrFu3jvLyctvOxbWGDRtGcnIyubm5/Pu//ztHjx41O9J12bLcv/nNb/L+++97v/bz8yMlJYU77rjDduvM06ZN4+LFi97HV/3f//0fkZGRZsUyzeDBg/ntb38LwPTp073nNNuR0+nkueeeY8SIEd6P2LOj22+/vcGYn58fo0aNatMfv2jLNferO2lAQECD566uK4q43W4+/vhjRo8ebXYU01VUVHDmzBnuvvtus6NIM9my3EVErM6WyzIiIlanchcRsSCVuzQqKyuLv/zlL2bHaBPcbjcej8fsGG3CX//6V1wul9kx2oS2Phe2vHHY9YwbNw6oPWtk+vTpJqcxV15eHn//+9+pqqpiy5YtZscx1dKlSzlz5gyxsbH8+Mc/NjuOqf7rv/6LTz75hNtvv51NmzaZHcdUbX0u9IbqlxQVFZGXl8c///M/mx1F2hDDMPj000+56667zI7SJpSXl9O1a1ezY7QJbXUuVO5Sz9mzZ/n444/p168f/fr1MzuO6UpKSujevbvZMUx38eJFTp8+TUREBEFBQWbHMVV7mQutuX9JfHy82RF86tqPj8vKymLmzJkcOnSIH/zgB+zcudPEZL734osveh9/+umnxMbGMmnSJKKiotr0lYit4do7gX/wwQdMmDCBZ555hvj4+Db/2aEtrb3OhS3X3DMzMxsdNwyDwsJCH6cx17lz57yPt2zZQlpaGhERERQVFTFr1iwmTZpkYjrfOnDggPc/dj/72c94+umnGT16NHl5eaSkpLB9+3aTE/rOtW+mb968mRdeeIFBgwZx9uxZfvSjH9nqwq72Ohe2LPdFixYRHx/f6L0y7Ha5+bVzUFVVRUREBAAhISF06GDfP+zcbrf3l/aee+6x7aX3ULsMMWjQIKD2w0tqampMTmSe9jQXtiz3b33rWzz22GONXkr9zjvvmJDIPCdOnODb3/42hmFQWVlJYWEhYWFheDyeejcVs4OzZ88yf/58AAoKCrh8+TKBgYFA7X/47OTUqVPeJcr8/HxKS0sJDg6mpqZGc9FO5sKWb6h+8MEH3HLLLQ3uVw21564OGTLEhFRtS1lZGSdPnuS+++4zO4rPXHszOYBBgwbRpUsXvvjiC/bv31/vxmpW99lnn9X7Ojw8nI4dO1JUVMQHH3xATEyMScl8r73OhS3LXUTE6my5qFpVVcX27duZM2cO8fHxPPTQQ8ydO5fXXnvNdvcwv5EVK1aYHaHNeP31182O0Gb84he/MDtCm9GW58KWa+5Lly6lW7du/PCHP8TpdAK1a6y7du1iyZIlbfJqMzM8/PDDZkdoM/QHbp2rbyhK254LWy7LxMbGsn///q/8nIhIe2HLZZng4GD27dtX7zSmmpoa3nrrrTZ9xVlrqK6uZvv27WzatIkPP/yw3nPXXtRjB4Zh8NZbb7Fv3z4Mw+DIkSOsWbOGbdu2telT3lrL22+/zY4dO8jPz683np6eblIic6xbt67B70Z7YMsj9/z8fNavX8+7775LcHAwhmFQVlbGyJEjefLJJ73netvB8uXLqaioYMiQIezZs4f777+fZcuWAZCYmMiuXbtMTug7ycnJFBUV4fF46Nq1Kx6Ph6ioKHJycggNDeUnP/mJ2RF9ZuPGjXz44YcMHDiQQ4cOMXPmTB599FHAfvvFyJEjueWWWyguLmbcuHHExcUxcOBAs2M1zYcfxt0mFRUVGefPnzc7hmni4uK8jysrK42f/OQnxhNPPGFcuXLFmDhxoonJfO/qXHg8HmP48OHGlStXDMOonZdr58kO4uLijMrKSsMwDKO0tNSYO3eusXbtWsMwDNvtF1df7//+7/8azz//vDF+/HgjNjbW+MUvfmGcOnXK5HTXZ8tlGYCTJ0+SmprKCy+8wIsvvkhqaqrtPhwbqHd2kL+/P6tXr6Z///7MmDGDS5cumZjM9/z8/ADo2LEjgwcPplOnTkDtvNjtat2qqir8/WvPtwgKCuKll16ivLycBQsW2O6MsqtXcd9+++088cQT7N27l02bNnHlyhXmzZtncrrrs9ce+w+pqakkJSUBMGTIEO9FS0lJSaSmppoZzecGDx5Mbm5uvbH/+I//YPLkyQ0u3rC6nj17cvHiRQBeeeUV73hhYSEdO3Y0K5YpvvnNb9a7qMvPz4+UlBTuuOMO2x0EGY2sXPfv358nn3ySAwcOmJCoeWy55h4bG8ubb77Z4BfW4/EQFxd33RuLiT1dunSJy5cvExoaanYUn7l6L52AgIAGz7lcLnr16uXrSKa5ePEiXbpOGAg4AAAGGElEQVR0MTvGV2bLI3eHw4Hb7W4wXlhY2OjNxKyuvLycM2fONBg/ceKECWnMVVhY6L0zaFFREZmZmXz22We2KnaoLfUvF/vGjRsBbFXsQKPFfnUu2jJbXsT09NNPM2vWLG677TZ69+4N1N769syZM7a7KvOtt94iJSWF0NBQqqqqWLduHffccw8Ay5Yts9VZEdu3b/cuy82dO5ddu3Zx5513smHDBubOncvUqVNNTug7a9asqfe1YRjs3r3b+z6Mnc4caq9zYctyHzVqFPv37ycvLw+Xy4VhGDidToYMGeJ9U80ufvWrX7Fz507Cw8PJy8tj6dKlJCUlERMTY7urMrdt28bevXupqKggKiqKzMxMwsLCKC0tZcaMGbYq98zMTIYPH853v/td736wd+/eNn1FZmtpr3Nhy3IH6NChA0OHDjU7hulqamoIDw8Hau9bvnXrVubPn09BQYHtlqj8/f0JDAwkMDCQiIgIwsLCgNqL3uw2F2+99RabN2/m7bffZunSpfTq1Yvnn3+exMREs6P5XHudC1uuud/I448/bnYEn+rSpUu99fbw8HC2bt1KdnY2//M//2NiMt9zOBze0/yuPWvqypUrtrtCtWvXrixfvpzZs2ezePFiXnnlFdv9JXdVe50LW54tcyNut9t7JGsHJ06cICAggNtvv73eeGVlJfv27eOhhx4yJ5gJzp07R3h4uPf87qtcLhcnT57kgQceMCmZuQzD4NVXX+XPf/4z69evNzuOqdrTXNi+3EtKSnA4HAQHB5sdxRSGYTS55NCcbaxAc1FHc1Gnvc6FX/K1H+1tE+fOneOnP/0pycnJ7NmzhzfeeINf/vKXHDt2jMGDB9vq5mEzZszg4sWLhIWF0a1bN++4x+Ph/fff5/nnn+fChQsMGDDAxJS+obmoo7mo017nwpZH7g8//DAzZ84kNjbWe3ZMdXU1f/zjH0lLS+ONN94wOaHvXLlyhfT0dP7whz+Qn59PUFCQd435O9/5DtOmTWtzO21r0VzUaWwuKioqMAxDc9FO5sKW5R4TE3Pdq1Bv9JzVVVZWUlxcTEBAgK3+emmM5qKO5qJOe5oLW5b7okWLCA4OJjExscEnMRUXF7N582aTE4qIfD22LHePx0N6ejrZ2dm43W4Mw6BXr15ERUUxdepU790ARUTaK1uWu4iI1ekipn9o61ebiYh8FSr3f9AfMCJiJSr3fxg9erTZEUREWowt19zb6xVnIiLNZcsj9xkzZvC73/2Oc+fO1Rv3eDwcOXKEH//4x7a6j7mIWI8tj9zb6xVnIiLNZctyv1Z7uuJMRKS5bF/uIiJWZMs1dxERq1O5i4hYkMpdLOlb3/oWFy9ebPS5iRMnUlFR0SayiLQW235AttjX7t27zY4g0upU7mJZv/71rzl8+DDFxcUkJSURGxsL1B5Jf/TRR3Tp0oWoqCgmTpzIO++8Q2FhIY899hjTp08HuOFzp06dIiUlheLiYiorK5k5cyaTJ08GIDMzk40bN9K9e3dGjRplzosX21O5i2U5HA62b9/OqVOneOSRRxg2bBihoaENtquoqOD1118nPz+f+Ph4EhMT6dKly3Wf69y5M4sXL+bnP/85/fr1o7y8nMmTJzN06FC6d+/OihUreO211+jbty8vv/yyr1+2CKByFwubOnUqAH379mXgwIEcPXqUMWPGNNhu/PjxANx6660EBQVRUFBAv379rvucYRicPHmSpKQk779RWVnJqVOn6NChAwMHDqRv375A7Uc6rl+/vlVfp0hjVO5iCze6V1Dnzp29j/38/Kiurr7hcw6Hgx49ejS6dp+VldWCqUVuns6WEcv6/e9/D8Dp06f5+OOPuffee1vk373jjjsICAggIyPDO3by5EnKy8u57777OH78OKdPnwZgx44dLfIzRb4qHbmLZXXq1Il//dd/pbi4mFWrVjW63n4z/P39eemll0hJSeGVV16hpqaG0NBQNm3aRGhoKKtXr2b+/Pl0796dsWPHtsjPFPmqdPsBEREL0rKMiIgFqdxFRCxI5S4iYkEqdxERC1K5i4hYkMpdRMSCVO4iIhakchcRsaD/DzBRkDg4TrM9AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# binning\n",
    "x_siblings = combined_data[(combined_data['target'] != -999) & (combined_data['number_siblings_spouses'].notnull())]['number_siblings_spouses']\n",
    "y_siblings = combined_data[(combined_data['target'] != -999) & (combined_data['number_siblings_spouses'].notnull())]['target']\n",
    "\n",
    "siblings_bins = bin_variable(x_siblings, y_siblings, min_samples_leaf=10, max_leaf_nodes=4)\n",
    "\n",
    "x_parents = combined_data[(combined_data['target'] != -999) & (combined_data['number_parents_children'].notnull())]['number_parents_children']\n",
    "y_parents = combined_data[(combined_data['target'] != -999) & (combined_data['number_parents_children'].notnull())]['target']\n",
    "\n",
    "parents_bins = bin_variable(x_parents, y_parents, min_samples_leaf=10, max_leaf_nodes=4)\n",
    "\n",
    "x_cabin_number = combined_data[(combined_data['target'] != -999) & (combined_data['cabin_number'].notnull())]['cabin_number']\n",
    "y_cabin_number = combined_data[(combined_data['target'] != -999) & (combined_data['cabin_number'].notnull())]['target']\n",
    "\n",
    "cabin_number_bins = bin_variable(x_cabin_number, y_cabin_number, min_samples_leaf=10, max_leaf_nodes=4)\n",
    "\n",
    "x_age = combined_data[(combined_data['target'] != -999) & (combined_data['age'].notnull())]['age']\n",
    "y_age = combined_data[(combined_data['target'] != -999) & (combined_data['age'].notnull())]['target']\n",
    "\n",
    "age_bins = bin_variable(x_age, y_age, min_samples_leaf=20, max_leaf_nodes=3)\n",
    "\n",
    "x_fare = combined_data[(combined_data['target'] != -999) & (combined_data['fare'].notnull())]['fare']\n",
    "y_fare = combined_data[(combined_data['target'] != -999) & (combined_data['fare'].notnull())]['target']\n",
    "\n",
    "fare_bins = bin_variable(x_fare, y_fare, min_samples_leaf=20, max_leaf_nodes=4)\n",
    "\n",
    "x_family_size = combined_data[(combined_data['target'] != -999) & (combined_data['family_size'].notnull())]['family_size']\n",
    "y_family_size = combined_data[(combined_data['target'] != -999) & (combined_data['family_size'].notnull())]['target']\n",
    "\n",
    "family_size_bins = bin_variable(x_family_size, y_family_size, min_samples_leaf=20, max_leaf_nodes=4)\n",
    "\n",
    "x_family_id = combined_data[(combined_data['target'] != -999) & (combined_data['family_id_encoded'].notnull())]['family_id_encoded']\n",
    "y_family_id = combined_data[(combined_data['target'] != -999) & (combined_data['family_id_encoded'].notnull())]['target']\n",
    "\n",
    "family_id_bins = bin_variable(x_family_id, y_family_id, min_samples_leaf=20, max_leaf_nodes=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      family_id_bins_encoded\n",
      "1050                     NaN\n",
      "1172                     NaN\n",
      "1300                     NaN\n",
      "target   -999.0\n",
      "Name: 1050, dtype: float64\n"
     ]
    },
    {
     "ename": "SystemExit",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "An exception has occurred, use %tb to see the full traceback.\n",
      "\u001b[0;31mSystemExit\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/rasmus/miniconda3/envs/ml/lib/python3.6/site-packages/IPython/core/interactiveshell.py:2971: UserWarning: To exit: use 'exit', 'quit', or Ctrl-D.\n",
      "  warn(\"To exit: use 'exit', 'quit', or Ctrl-D.\", stacklevel=1)\n"
     ]
    }
   ],
   "source": [
    "# merge to master dataframe\n",
    "combined_data['fare_bins'] = pd.cut(combined_data['fare'], fare_bins, include_lowest=True, labels=False).astype(int)\n",
    "combined_data['family_size_bins'] = pd.cut(combined_data['family_size'], family_size_bins, include_lowest=True, labels=False).astype(int)\n",
    "combined_data['cabin_number_bins'] = pd.cut(combined_data['cabin_number'], cabin_number_bins, include_lowest=True, labels=False).astype(int)\n",
    "combined_data['siblings_bins'] = pd.cut(combined_data['number_siblings_spouses'], siblings_bins, include_lowest=True, labels=False).astype(int)\n",
    "combined_data['parents_bins'] = pd.cut(combined_data['number_parents_children'], parents_bins, include_lowest=True, labels=False).fillna(3).astype(int)\n",
    "combined_data['family_id_bins'] = pd.cut(combined_data['family_id_encoded'], family_id_bins, include_lowest=True, labels=False).fillna(5).astype(int)\n",
    "\n",
    "fare_bins_encoded, fare_bins_onehot = encode_feature(combined_data, 'fare_bins', 'target')\n",
    "family_size_bins_encoded, family_size_bins_onehot = encode_feature(combined_data, 'family_size_bins', 'target')\n",
    "cabin_number_bins_encoded, cabin_number_bins_onehot = encode_feature(combined_data, 'cabin_number_bins', 'target')\n",
    "siblings_bins_encoded, siblings_bins_onehot = encode_feature(combined_data, 'siblings_bins', 'target')\n",
    "parents_bins_encoded, parents_bins_onehot = encode_feature(combined_data, 'parents_bins', 'target')\n",
    "\n",
    "# >>> to solve: family id 77 has no target\n",
    "# peobably remove: combined_data[by] != -999] ^up\n",
    "family_id_bins_encoded, family_id_bins_onehot = encode_feature(combined_data, 'family_id_bins', 'target')\n",
    "\n",
    "sys.exit()\n",
    "### <<<\n",
    "\n",
    "\n",
    "\n",
    "# deal with age later on\n",
    "\n",
    "combined_data.drop(['fare_bins', 'family_size_bins', 'cabin_number_bins',\n",
    "                    'siblings_bins', 'parents_bins', 'family_id_bins'], axis=1)\n",
    "\n",
    "combined_data = pd.concat(\n",
    "    [combined_data,\n",
    "     fare_bins_encoded, fare_bins_onehot,\n",
    "     family_size_bins_encoded, family_size_bins_onehot,\n",
    "     cabin_number_bins_encoded, cabin_number_bins_onehot,\n",
    "     siblings_bins_encoded, siblings_bins_onehot,\n",
    "     parents_bins_encoded, parents_bins_onehot,\n",
    "     family_id_bins_encoded, family_id_bins_onehot]\n",
    "     , axis=1)\n",
    "\n",
    "print(combined_data.head())\n",
    "for x in combined_data.columns:\n",
    "    print(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## // imputation 2 (age)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['cabin_number', 'fare', 'number_parents_children', 'pclass_0',\n",
      "       'title_1', 'title_2'],\n",
      "      dtype='object')\n",
      "Linear regression score: 0.35\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# no one-hots here\n",
    "columns_age_regression = [\n",
    "    'cabin_number',\n",
    "    'family_size',\n",
    "    'fare',\n",
    "    'number_parents_children',\n",
    "    'number_siblings_spouses',\n",
    "    'sex_is_male',\n",
    "    'ticket_is_digit',\n",
    "    'pclass_0',\n",
    "    'pclass_1',\n",
    "    'pclass_2',\n",
    "    'embarked_0',\n",
    "    'embarked_1',\n",
    "    'embarked_2',\n",
    "    'title_0',\n",
    "    'title_1',\n",
    "    'title_2',\n",
    "    'title_3',\n",
    "    'title_4',\n",
    "    'cabin_letter_0',\n",
    "    'cabin_letter_1',\n",
    "    'cabin_letter_2',\n",
    "    'cabin_letter_3',\n",
    "    'cabin_letter_4',\n",
    "    'cabin_letter_5',\n",
    "    'cabin_letter_6',\n",
    "    'cabin_letter_7',\n",
    "    'cabin_letter_8'\n",
    "];\n",
    "\n",
    "# regression\n",
    "X_age_regression = combined_data[combined_data['age'].notnull()][columns_age_regression]\n",
    "y_age_regression = combined_data[combined_data['age'].notnull()]['age']\n",
    "\n",
    "# check mean age over columns\n",
    "for column in X_age_regression.columns:\n",
    "    if column != 'age':\n",
    "        combined_data[combined_data['age'].notnull()].drop(['id', 'target'], axis=1).groupby(column).age.mean().plot(kind='bar', stacked=True)\n",
    "        plt.show()\n",
    "        \n",
    "estimator_age_regression = RandomForestRegressor()\n",
    "\n",
    "estimator_age_regression.fit(X_age_regression, y_age_regression)\n",
    "feature_selection_age_regression = SelectFromModel(estimator_age_regression, prefit=True)\n",
    "feature_mask_age_regression = feature_selection_age_regression.get_support(indices=True)\n",
    "features_age_regression = X_age_regression.columns[feature_mask_age_regression]\n",
    "X_age_regression_reduced = feature_selection_age_regression.transform(X_age_regression)\n",
    "age_regression_selected_features = X_age_regression.columns[feature_mask_age_regression]\n",
    "\n",
    "print(age_regression_selected_features)\n",
    "\n",
    "cv_age_regression = KFold(n_splits=10, shuffle=True, random_state=0)\n",
    "cv_score_age_regression = cross_val_score(estimator_age_regression, X_age_regression_reduced, y_age_regression, cv=cv_age_regression, scoring='r2').mean()\n",
    "\n",
    "print(\"Linear regression score: %.2f\" % cv_score_age_regression)\n",
    "\n",
    "estimator_age_regression.fit(X_age_regression_reduced, y_age_regression)\n",
    "y_pred_age_regression = estimator_age_regression.predict(X_age_regression_reduced)\n",
    "\n",
    "plt.scatter(y_age_regression, y_pred_age_regression)\n",
    "plt.show()\n",
    "\n",
    "# replace missing age (age_bins)\n",
    "missing_age_indices = combined_data['age'].isnull()\n",
    "X_age_imputation = combined_data[missing_age_indices][age_regression_selected_features]\n",
    "y_pred_age_imputation = estimator_age_regression.predict(X_age_imputation)\n",
    "combined_data.loc[missing_age_indices, 'age'] = y_pred_age_imputation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Empty DataFrame\n",
      "Columns: [age_bins_encoded]\n",
      "Index: []\n",
      "    age  cabin_number  family_size     fare  id  number_parents_children  \\\n",
      "0  22.0        -999.0            2   7.2500   1                        0   \n",
      "1  38.0          85.0            2  71.2833   2                        0   \n",
      "2  26.0        -999.0            1   7.9250   3                        0   \n",
      "3  35.0         123.0            2  53.1000   4                        0   \n",
      "4  35.0        -999.0            1   8.0500   5                        0   \n",
      "\n",
      "   number_siblings_spouses  sex_is_male  target  ticket_is_digit     ...      \\\n",
      "0                        1            1       0                0     ...       \n",
      "1                        1            0       1                0     ...       \n",
      "2                        0            0       1                0     ...       \n",
      "3                        1            0       1                1     ...       \n",
      "4                        0            1       0                1     ...       \n",
      "\n",
      "   family_id_bins_0.0  family_id_bins_1.0  family_id_bins_2.0  \\\n",
      "0                   0                   0                   1   \n",
      "1                   0                   0                   1   \n",
      "2                   0                   0                   1   \n",
      "3                   0                   0                   1   \n",
      "4                   0                   0                   1   \n",
      "\n",
      "   family_id_bins_3.0  family_id_bins_4.0  age_bins  age_bins_encoded  \\\n",
      "0                   0                   0         1                 1   \n",
      "1                   0                   0         1                 1   \n",
      "2                   0                   0         1                 1   \n",
      "3                   0                   0         1                 1   \n",
      "4                   0                   0         1                 1   \n",
      "\n",
      "   age_bins_0  age_bins_1  age_bins_2  \n",
      "0           0           1           0  \n",
      "1           0           1           0  \n",
      "2           0           1           0  \n",
      "3           0           1           0  \n",
      "4           0           1           0  \n",
      "\n",
      "[5 rows x 154 columns]\n",
      "age\n",
      "cabin_number\n",
      "family_size\n",
      "fare\n",
      "id\n",
      "number_parents_children\n",
      "number_siblings_spouses\n",
      "sex_is_male\n",
      "target\n",
      "ticket_is_digit\n",
      "pclass_encoded\n",
      "pclass_0\n",
      "pclass_1\n",
      "pclass_2\n",
      "embarked_encoded\n",
      "embarked_0\n",
      "embarked_1\n",
      "embarked_2\n",
      "title_encoded\n",
      "title_0\n",
      "title_1\n",
      "title_2\n",
      "title_3\n",
      "title_4\n",
      "cabin_letter_encoded\n",
      "cabin_letter_0\n",
      "cabin_letter_1\n",
      "cabin_letter_2\n",
      "cabin_letter_3\n",
      "cabin_letter_4\n",
      "cabin_letter_5\n",
      "cabin_letter_6\n",
      "cabin_letter_7\n",
      "cabin_letter_8\n",
      "family_id_encoded\n",
      "family_id_0\n",
      "family_id_1\n",
      "family_id_2\n",
      "family_id_3\n",
      "family_id_4\n",
      "family_id_5\n",
      "family_id_6\n",
      "family_id_7\n",
      "family_id_8\n",
      "family_id_9\n",
      "family_id_10\n",
      "family_id_11\n",
      "family_id_12\n",
      "family_id_13\n",
      "family_id_14\n",
      "family_id_15\n",
      "family_id_16\n",
      "family_id_17\n",
      "family_id_18\n",
      "family_id_19\n",
      "family_id_20\n",
      "family_id_21\n",
      "family_id_22\n",
      "family_id_23\n",
      "family_id_24\n",
      "family_id_25\n",
      "family_id_26\n",
      "family_id_27\n",
      "family_id_28\n",
      "family_id_29\n",
      "family_id_30\n",
      "family_id_31\n",
      "family_id_32\n",
      "family_id_33\n",
      "family_id_34\n",
      "family_id_35\n",
      "family_id_36\n",
      "family_id_37\n",
      "family_id_38\n",
      "family_id_39\n",
      "family_id_40\n",
      "family_id_41\n",
      "family_id_42\n",
      "family_id_43\n",
      "family_id_44\n",
      "family_id_45\n",
      "family_id_46\n",
      "family_id_47\n",
      "family_id_48\n",
      "family_id_49\n",
      "family_id_50\n",
      "family_id_51\n",
      "family_id_52\n",
      "family_id_53\n",
      "family_id_54\n",
      "family_id_55\n",
      "family_id_56\n",
      "family_id_57\n",
      "family_id_58\n",
      "family_id_59\n",
      "family_id_60\n",
      "family_id_61\n",
      "family_id_62\n",
      "family_id_63\n",
      "family_id_64\n",
      "family_id_65\n",
      "family_id_66\n",
      "family_id_67\n",
      "family_id_68\n",
      "family_id_69\n",
      "family_id_70\n",
      "family_id_71\n",
      "family_id_72\n",
      "family_id_73\n",
      "family_id_74\n",
      "family_id_75\n",
      "family_id_76\n",
      "family_id_77\n",
      "fare_bins\n",
      "family_size_bins\n",
      "cabin_number_bins\n",
      "siblings_bins\n",
      "parents_bins\n",
      "family_id_bins\n",
      "fare_bins_encoded\n",
      "fare_bins_0\n",
      "fare_bins_1\n",
      "fare_bins_2\n",
      "fare_bins_3\n",
      "family_size_bins_encoded\n",
      "family_size_bins_0\n",
      "family_size_bins_1\n",
      "family_size_bins_2\n",
      "family_size_bins_3\n",
      "cabin_number_bins_encoded\n",
      "cabin_number_bins_0\n",
      "cabin_number_bins_1\n",
      "cabin_number_bins_2\n",
      "siblings_bins_encoded\n",
      "siblings_bins_0\n",
      "siblings_bins_1\n",
      "siblings_bins_2\n",
      "siblings_bins_3\n",
      "parents_bins_encoded\n",
      "parents_bins_0\n",
      "parents_bins_1\n",
      "parents_bins_2\n",
      "parents_bins_3\n",
      "family_id_bins_encoded\n",
      "family_id_bins_0.0\n",
      "family_id_bins_1.0\n",
      "family_id_bins_2.0\n",
      "family_id_bins_3.0\n",
      "family_id_bins_4.0\n",
      "age_bins\n",
      "age_bins_encoded\n",
      "age_bins_0\n",
      "age_bins_1\n",
      "age_bins_2\n"
     ]
    }
   ],
   "source": [
    "# two predictions were outside bins (younger than 1 years, set to 0 with fillna)\n",
    "combined_data['age_bins'] = pd.cut(combined_data['age'], age_bins, include_lowest=True, labels=False).fillna(0).astype(int)\n",
    "age_bins_encoded, age_bins_onehot = encode_feature(combined_data, 'age_bins', 'target')\n",
    "\n",
    "combined_data.drop('age_bins', axis=1)\n",
    "\n",
    "combined_data = pd.concat([combined_data, age_bins_encoded, age_bins_onehot], axis=1) \n",
    "\n",
    "print(combined_data.head())\n",
    "for x in combined_data.columns:\n",
    "    print(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. modelling"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## // prepare data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 331,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(891, 145)\n",
      "(891,)\n"
     ]
    }
   ],
   "source": [
    "train_observations = combined_data['target'] != -999\n",
    "test_observations = combined_data['target'] == -999\n",
    "\n",
    "X_train = combined_data[train_observations].drop(['id', 'target'], axis=1)\n",
    "y_train = combined_data[train_observations]['target']\n",
    "\n",
    "print(X_train.shape)\n",
    "print(y_train.shape)\n",
    "\n",
    "X_test = combined_data[test_observations].drop(['id', 'target'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## // feature selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 350,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x1800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['sex_is_male', 'ticket_is_digit', 'pclass_0', 'pclass_1', 'pclass_2',\n",
      "       'embarked_0', 'embarked_1', 'embarked_2', 'title_0', 'title_1',\n",
      "       'title_2', 'title_3', 'title_4', 'cabin_letter_0', 'cabin_letter_1',\n",
      "       'cabin_letter_2', 'cabin_letter_4', 'cabin_letter_7', 'family_id_42',\n",
      "       'fare_bins_1', 'fare_bins_2', 'family_size_bins_1',\n",
      "       'family_size_bins_3', 'cabin_number_bins_1', 'siblings_bins_0',\n",
      "       'siblings_bins_1', 'parents_bins_0', 'parents_bins_1', 'parents_bins_2',\n",
      "       'age_bins_0', 'age_bins_1'],\n",
      "      dtype='object')\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x1800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['age', 'fare', 'sex_is_male', 'title_encoded', 'family_id_encoded'], dtype='object')\n"
     ]
    }
   ],
   "source": [
    "def feature_selection(X, y):\n",
    "\n",
    "    estimator_feature_selection = RandomForestClassifier()\n",
    "    estimator_feature_selection.fit(X, y)\n",
    "    selection_feature_selection = SelectFromModel(estimator_feature_selection, prefit=True)\n",
    "    mask_feature_selection = selection_feature_selection.get_support(indices=True)\n",
    "\n",
    "    features = pd.DataFrame()\n",
    "    features['feature'] = X.columns\n",
    "    features['importance'] = estimator_feature_selection.feature_importances_\n",
    "    features.sort_values(by=['importance'], ascending=True, inplace=True)\n",
    "    features.set_index('feature', inplace=True)\n",
    "    features.plot(kind='barh', figsize=(25, 25))\n",
    "\n",
    "    plt.show()\n",
    "    \n",
    "    return X.columns[mask_feature_selection]\n",
    "   \n",
    "decision_tree_columns = [\n",
    "    'age',\n",
    "    'cabin_number',\n",
    "    'family_size',\n",
    "    'fare',\n",
    "    'number_parents_children',\n",
    "    'number_siblings_spouses',\n",
    "    'sex_is_male',\n",
    "    'ticket_is_digit',\n",
    "    'pclass_encoded',\n",
    "    'embarked_encoded',\n",
    "    'title_encoded',\n",
    "    'cabin_letter_encoded',\n",
    "    'family_id_encoded',\n",
    "]\n",
    "\n",
    "linear_columns = [\n",
    "    'sex_is_male',\n",
    "    'ticket_is_digit',\n",
    "    'pclass_0',\n",
    "    'pclass_1',\n",
    "    'pclass_2',\n",
    "    'embarked_0',\n",
    "    'embarked_1',\n",
    "    'embarked_2',\n",
    "    'title_0',\n",
    "    'title_1',\n",
    "    'title_2',\n",
    "    'title_3',\n",
    "    'title_4',\n",
    "    'cabin_letter_0',\n",
    "    'cabin_letter_1',\n",
    "    'cabin_letter_2',\n",
    "    'cabin_letter_3',\n",
    "    'cabin_letter_4',\n",
    "    'cabin_letter_5',\n",
    "    'cabin_letter_6',\n",
    "    'cabin_letter_7',\n",
    "    'cabin_letter_8',\n",
    "    'family_id_0',\n",
    "    'family_id_1',\n",
    "    'family_id_2',\n",
    "    'family_id_3',\n",
    "    'family_id_4',\n",
    "    'family_id_5',\n",
    "    'family_id_6',\n",
    "    'family_id_7',\n",
    "    'family_id_8',\n",
    "    'family_id_9',\n",
    "    'family_id_10',\n",
    "    'family_id_11',\n",
    "    'family_id_12',\n",
    "    'family_id_13',\n",
    "    'family_id_14',\n",
    "    'family_id_15',\n",
    "    'family_id_16',\n",
    "    'family_id_17',\n",
    "    'family_id_18',\n",
    "    'family_id_19',\n",
    "    'family_id_20',\n",
    "    'family_id_21',\n",
    "    'family_id_22',\n",
    "    'family_id_23',\n",
    "    'family_id_24',\n",
    "    'family_id_25',\n",
    "    'family_id_26',\n",
    "    'family_id_27',\n",
    "    'family_id_28',\n",
    "    'family_id_29',\n",
    "    'family_id_30',\n",
    "    'family_id_31',\n",
    "    'family_id_32',\n",
    "    'family_id_33',\n",
    "    'family_id_34',\n",
    "    'family_id_35',\n",
    "    'family_id_36',\n",
    "    'family_id_37',\n",
    "    'family_id_38',\n",
    "    'family_id_39',\n",
    "    'family_id_40',\n",
    "    'family_id_41',\n",
    "    'family_id_42',\n",
    "    'family_id_43',\n",
    "    'family_id_44',\n",
    "    'family_id_45',\n",
    "    'family_id_46',\n",
    "    'family_id_47',\n",
    "    'family_id_48',\n",
    "    'family_id_49',\n",
    "    'family_id_50',\n",
    "    'family_id_51',\n",
    "    'family_id_52',\n",
    "    'family_id_53',\n",
    "    'family_id_54',\n",
    "    'family_id_55',\n",
    "    'family_id_56',\n",
    "    'family_id_57',\n",
    "    'family_id_58',\n",
    "    'family_id_59',\n",
    "    'family_id_60',\n",
    "    'family_id_61',\n",
    "    'family_id_62',\n",
    "    'family_id_63',\n",
    "    'family_id_64',\n",
    "    'family_id_65',\n",
    "    'family_id_66',\n",
    "    'family_id_67',\n",
    "    'family_id_68',\n",
    "    'family_id_69',\n",
    "    'family_id_70',\n",
    "    'family_id_71',\n",
    "    'family_id_72',\n",
    "    'family_id_73',\n",
    "    'family_id_74',\n",
    "    'family_id_75',\n",
    "    'family_id_76',\n",
    "    'family_id_77',\n",
    "    'fare_bins_0',\n",
    "    'fare_bins_1',\n",
    "    'fare_bins_2',\n",
    "    'fare_bins_3',\n",
    "    'family_size_bins_0',\n",
    "    'family_size_bins_1',\n",
    "    'family_size_bins_2',\n",
    "    'family_size_bins_3',\n",
    "    'cabin_number_bins_0',\n",
    "    'cabin_number_bins_1',\n",
    "    'cabin_number_bins_2',\n",
    "    'siblings_bins_0',\n",
    "    'siblings_bins_1',\n",
    "    'siblings_bins_2',\n",
    "    'siblings_bins_3',\n",
    "    'parents_bins_0',\n",
    "    'parents_bins_1',\n",
    "    'parents_bins_2',\n",
    "    'parents_bins_3',\n",
    "    'age_bins_0',\n",
    "    'age_bins_1',\n",
    "    'age_bins_2',\n",
    "]\n",
    "\n",
    "linear_columns_reduced = feature_selection(X_train[linear_columns], y_train)\n",
    "print(linear_columns_reduced)\n",
    "decision_tree_columns_reduced = feature_selection(X_train[decision_tree_columns], y_train)\n",
    "print(decision_tree_columns_reduced)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## // model selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 374,
   "metadata": {},
   "outputs": [
    {
     "ename": "SystemExit",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "An exception has occurred, use %tb to see the full traceback.\n",
      "\u001b[0;31mSystemExit\u001b[0m\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/rasmus/miniconda3/envs/ml/lib/python3.6/site-packages/IPython/core/interactiveshell.py:2971: UserWarning: To exit: use 'exit', 'quit', or Ctrl-D.\n",
      "  warn(\"To exit: use 'exit', 'quit', or Ctrl-D.\", stacklevel=1)\n"
     ]
    }
   ],
   "source": [
    "X_train_decision_treee = X_train[decision_tree_columns_reduced]\n",
    "X_train_linear = X_train[linear_columns_reduced]\n",
    "\n",
    "test_estimators = [\n",
    "    {\n",
    "        'estimator': LogisticRegression(),\n",
    "        'params': {\n",
    "            'C' : [0.01, 0.1, 0.25, 0.5, 1]\n",
    "        },\n",
    "        'X': X_train_linear,\n",
    "        'y': y_train\n",
    "    },\n",
    "    {\n",
    "        'estimator': SVC(),\n",
    "        'params': {\n",
    "            'C' : [0.01, 0.1, 0.25, 0.5, 1]\n",
    "        },\n",
    "        'X': X_train_linear,\n",
    "        'y': y_train\n",
    "    },\n",
    "    {\n",
    "        'estimator': DecisionTreeClassifier(),\n",
    "        'params': {\n",
    "            'max_features' : range(1, len(decision_tree_columns_reduced), 1),\n",
    "            'max_depth': range(1, 10, 1)\n",
    "        },\n",
    "        'X': X_train_decision_treee,\n",
    "        'y': y_train\n",
    "    },\n",
    "    {\n",
    "        'estimator': RandomForestClassifier(),\n",
    "        'params': {\n",
    "            'max_features' : range(1, len(decision_tree_columns_reduced), 1),\n",
    "            'max_depth': range(1, 10, 1)\n",
    "        },\n",
    "        'X': X_train_decision_treee,\n",
    "        'y': y_train\n",
    "    },\n",
    "#    {\n",
    "#        'estimator': GradientBoostingClassifier(),\n",
    "#        'params': {\n",
    "#            'max_features' : range(1, len(decision_tree_columns_reduced), 1),\n",
    "#            'max_depth': range(1, 10, 1)\n",
    "#        }\n",
    "#    }\n",
    "]\n",
    "\n",
    "num_trials = 10\n",
    "\n",
    "results = pd.DataFrame()\n",
    "\n",
    "for test_estimator in test_estimators:\n",
    "\n",
    "    # Non_nested parameter search and scoring\n",
    "    estimator = test_estimator['estimator']\n",
    "    parameters = test_estimator['params']\n",
    "    X = test_estimator['X']\n",
    "    y = test_estimator['y']\n",
    "\n",
    "    for j in range(num_trials):\n",
    "\n",
    "        # Choose cross-validation techniques for the inner and outer loops\n",
    "        inner_cv = KFold(n_splits=10, shuffle=True, random_state=j)\n",
    "        outer_cv = KFold(n_splits=10, shuffle=True, random_state=j)\n",
    "\n",
    "        clf = GridSearchCV(estimator=estimator, param_grid=parameters, cv=inner_cv, scoring='accuracy')\n",
    "        clf.fit(X, y)\n",
    "\n",
    "        # Nested CV with parameter optimization\n",
    "        nested_score = cross_val_score(clf, X=X, y=y, cv=outer_cv, scoring='accuracy', n_jobs=12)\n",
    "\n",
    "        data = {\n",
    "            'estimator': clf.best_estimator_.__class__.__name__,\n",
    "            'avg_cv_score': nested_score.mean(),\n",
    "            'max_depth': clf.best_params_.get('max_depth', np.nan),\n",
    "            'max_features': clf.best_params_.get('max_features', np.nan),\n",
    "            'C': clf.best_params_.get('C', np.nan)\n",
    "        }\n",
    "\n",
    "        results = results.append(data, ignore_index=True)\n",
    "\n",
    "        print(str(j) + ' // ' + clf.best_estimator_.__class__.__name__ + ' // ' + str(nested_score.mean()) + ' // ' + str(clf.best_params_))\n",
    "\n",
    "results.groupby('estimator').mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 375,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "voting classifier cv score: 0.846350811485643\n"
     ]
    }
   ],
   "source": [
    "final_estimators = [\n",
    "    #('SVC', SVC(C=1)),\n",
    "    #('LogisticRegression', LogisticRegression(C=0.625)),\n",
    "    ('DecisionTreeClassifier', DecisionTreeClassifier(max_features=3, max_depth=7)),\n",
    "    ('RandomForestClassifier', RandomForestClassifier(max_features=3, max_depth=7)),\n",
    "    #('GradientBoostingClassifier', GradientBoostingClassifier(max_features=3, max_depth=7)),\n",
    "]\n",
    "\n",
    "warnings.filterwarnings(module='sklearn*', action='ignore', category=DeprecationWarning)\n",
    "\n",
    "voting_scores = []\n",
    "\n",
    "for j in range(num_trials):\n",
    "    cv = KFold(n_splits=10, shuffle=True, random_state=j)\n",
    "    voting_estimator = VotingClassifier(final_estimators, voting='soft')\n",
    "    score = cross_val_score(voting_estimator, X=X_train_decision_treee, y=y_train, cv=cv, scoring='accuracy', n_jobs=12)\n",
    "    voting_scores.append(score.mean())\n",
    "\n",
    "print('voting classifier cv score: {}'.format(np.asarray(voting_scores).mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 376,
   "metadata": {},
   "outputs": [],
   "source": [
    "# submission\n",
    "chosen_estimator = voting_estimator\n",
    "chosen_estimator.fit(X_train_decision_treee, y_train)\n",
    "\n",
    "X_test_decision_treee = X_test[decision_tree_columns_reduced]\n",
    "\n",
    "predictions = chosen_estimator.predict(X_test_decision_treee)\n",
    "output = pd.DataFrame(data={\n",
    "    'PassengerId': combined_data[test_observations]['id'],\n",
    "    'Survived': predictions.astype(int)\n",
    "    })\n",
    "    \n",
    "output.to_csv('submissions/kaggle_titanic_submission_aurany_24jun2018.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 324,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    age  cabin_number  family_size     fare  id  number_parents_children  \\\n",
      "0  22.0        -999.0            2   7.2500   1                        0   \n",
      "1  38.0          85.0            2  71.2833   2                        0   \n",
      "2  26.0        -999.0            1   7.9250   3                        0   \n",
      "3  35.0         123.0            2  53.1000   4                        0   \n",
      "4  35.0        -999.0            1   8.0500   5                        0   \n",
      "\n",
      "   number_siblings_spouses  sex_is_male  target  ticket_is_digit     ...      \\\n",
      "0                        1            1       0                0     ...       \n",
      "1                        1            0       1                0     ...       \n",
      "2                        0            0       1                0     ...       \n",
      "3                        1            0       1                1     ...       \n",
      "4                        0            1       0                1     ...       \n",
      "\n",
      "   parents_bins_encoded  parents_bins_0  parents_bins_1  parents_bins_2  \\\n",
      "0                     2               0               0               1   \n",
      "1                     2               0               0               1   \n",
      "2                     2               0               0               1   \n",
      "3                     2               0               0               1   \n",
      "4                     2               0               0               1   \n",
      "\n",
      "   parents_bins_3  age_bins  age_bins_encoded  age_bins_0  age_bins_1  \\\n",
      "0               0         1                 1           0           1   \n",
      "1               0         1                 1           0           1   \n",
      "2               0         1                 1           0           1   \n",
      "3               0         1                 1           0           1   \n",
      "4               0         1                 1           0           1   \n",
      "\n",
      "   age_bins_2  \n",
      "0           0  \n",
      "1           0  \n",
      "2           0  \n",
      "3           0  \n",
      "4           0  \n",
      "\n",
      "[5 rows x 147 columns]\n",
      "age\n",
      "cabin_number\n",
      "family_size\n",
      "fare\n",
      "id\n",
      "number_parents_children\n",
      "number_siblings_spouses\n",
      "sex_is_male\n",
      "target\n",
      "ticket_is_digit\n",
      "pclass_encoded\n",
      "pclass_0\n",
      "pclass_1\n",
      "pclass_2\n",
      "embarked_encoded\n",
      "embarked_0\n",
      "embarked_1\n",
      "embarked_2\n",
      "title_encoded\n",
      "title_0\n",
      "title_1\n",
      "title_2\n",
      "title_3\n",
      "title_4\n",
      "cabin_letter_encoded\n",
      "cabin_letter_0\n",
      "cabin_letter_1\n",
      "cabin_letter_2\n",
      "cabin_letter_3\n",
      "cabin_letter_4\n",
      "cabin_letter_5\n",
      "cabin_letter_6\n",
      "cabin_letter_7\n",
      "cabin_letter_8\n",
      "family_id_encoded\n",
      "family_id_0\n",
      "family_id_1\n",
      "family_id_2\n",
      "family_id_3\n",
      "family_id_4\n",
      "family_id_5\n",
      "family_id_6\n",
      "family_id_7\n",
      "family_id_8\n",
      "family_id_9\n",
      "family_id_10\n",
      "family_id_11\n",
      "family_id_12\n",
      "family_id_13\n",
      "family_id_14\n",
      "family_id_15\n",
      "family_id_16\n",
      "family_id_17\n",
      "family_id_18\n",
      "family_id_19\n",
      "family_id_20\n",
      "family_id_21\n",
      "family_id_22\n",
      "family_id_23\n",
      "family_id_24\n",
      "family_id_25\n",
      "family_id_26\n",
      "family_id_27\n",
      "family_id_28\n",
      "family_id_29\n",
      "family_id_30\n",
      "family_id_31\n",
      "family_id_32\n",
      "family_id_33\n",
      "family_id_34\n",
      "family_id_35\n",
      "family_id_36\n",
      "family_id_37\n",
      "family_id_38\n",
      "family_id_39\n",
      "family_id_40\n",
      "family_id_41\n",
      "family_id_42\n",
      "family_id_43\n",
      "family_id_44\n",
      "family_id_45\n",
      "family_id_46\n",
      "family_id_47\n",
      "family_id_48\n",
      "family_id_49\n",
      "family_id_50\n",
      "family_id_51\n",
      "family_id_52\n",
      "family_id_53\n",
      "family_id_54\n",
      "family_id_55\n",
      "family_id_56\n",
      "family_id_57\n",
      "family_id_58\n",
      "family_id_59\n",
      "family_id_60\n",
      "family_id_61\n",
      "family_id_62\n",
      "family_id_63\n",
      "family_id_64\n",
      "family_id_65\n",
      "family_id_66\n",
      "family_id_67\n",
      "family_id_68\n",
      "family_id_69\n",
      "family_id_70\n",
      "family_id_71\n",
      "family_id_72\n",
      "family_id_73\n",
      "family_id_74\n",
      "family_id_75\n",
      "family_id_76\n",
      "family_id_77\n",
      "fare_bins\n",
      "family_size_bins\n",
      "cabin_number_bins\n",
      "siblings_bins\n",
      "parents_bins\n",
      "fare_bins_encoded\n",
      "fare_bins_0\n",
      "fare_bins_1\n",
      "fare_bins_2\n",
      "fare_bins_3\n",
      "family_size_bins_encoded\n",
      "family_size_bins_0\n",
      "family_size_bins_1\n",
      "family_size_bins_2\n",
      "family_size_bins_3\n",
      "cabin_number_bins_encoded\n",
      "cabin_number_bins_0\n",
      "cabin_number_bins_1\n",
      "cabin_number_bins_2\n",
      "siblings_bins_encoded\n",
      "siblings_bins_0\n",
      "siblings_bins_1\n",
      "siblings_bins_2\n",
      "siblings_bins_3\n",
      "parents_bins_encoded\n",
      "parents_bins_0\n",
      "parents_bins_1\n",
      "parents_bins_2\n",
      "parents_bins_3\n",
      "age_bins\n",
      "age_bins_encoded\n",
      "age_bins_0\n",
      "age_bins_1\n",
      "age_bins_2\n"
     ]
    }
   ],
   "source": [
    "print(combined_data.head())\n",
    "for x in combined_data.columns:\n",
    "    print(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}