Creating a Kaggle Workflow.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Creating a Kaggle Workflow"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data science, and particularly machine learning, contain many dimensions of complexity when compared with standard software development. In standard software development, code not working as you expect can be caused by a number of factors along two dimensions:\n",
    "\n",
    "* Bugs in implementation\n",
    "* Algorithm design\n",
    "\n",
    "Machine learning problems, have many more dimensions:\n",
    "\n",
    "* Bugs in implementation\n",
    "* Algorithm design\n",
    "* Model issues\n",
    "* Data quality\n",
    "\n",
    "The result of this is that there are exponentially more places that machine learning can go wrong.\n",
    "\n",
    "By defining a workflow for yourself, you can give yourself a framework with which to make iterating on ideas quicker and easier, allowing yourself to work more efficiently.\n",
    "\n",
    "In this mission, we're going to explore a workflow to make competing in the Kaggle Titanic competition easier, using a pipeline of functions to reduce the number of dimensions you need to focus on."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "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": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#read the data from Kaggle\n",
    "import pandas as pd\n",
    "\n",
    "train = pd.read_csv('train.csv')\n",
    "holdout = pd.read_csv('test.csv')\n",
    "holdout.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Processing the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# %load functions.py\n",
    "def process_missing(df):\n",
    "    \"\"\"Handle various missing values from the data set\n",
    "\n",
    "    Usage\n",
    "    ------\n",
    "\n",
    "    holdout = process_missing(holdout)\n",
    "    \"\"\"\n",
    "    df[\"Fare\"] = df[\"Fare\"].fillna(train[\"Fare\"].mean())\n",
    "    df[\"Embarked\"] = df[\"Embarked\"].fillna(\"S\")\n",
    "    return df\n",
    "\n",
    "def process_age(df):\n",
    "    \"\"\"Process the Age column into pre-defined 'bins' \n",
    "\n",
    "    Usage\n",
    "    ------\n",
    "\n",
    "    train = process_age(train)\n",
    "    \"\"\"\n",
    "    df[\"Age\"] = df[\"Age\"].fillna(-0.5)\n",
    "    cut_points = [-1,0,5,12,18,35,60,100]\n",
    "    label_names = [\"Missing\",\"Infant\",\"Child\",\"Teenager\",\"Young Adult\",\"Adult\",\"Senior\"]\n",
    "    df[\"Age_categories\"] = pd.cut(df[\"Age\"],cut_points,labels=label_names)\n",
    "    return df\n",
    "\n",
    "def process_fare(df):\n",
    "    \"\"\"Process the Fare column into pre-defined 'bins' \n",
    "\n",
    "    Usage\n",
    "    ------\n",
    "\n",
    "    train = process_fare(train)\n",
    "    \"\"\"\n",
    "    cut_points = [-1,12,50,100,1000]\n",
    "    label_names = [\"0-12\",\"12-50\",\"50-100\",\"100+\"]\n",
    "    df[\"Fare_categories\"] = pd.cut(df[\"Fare\"],cut_points,labels=label_names)\n",
    "    return df\n",
    "\n",
    "def process_cabin(df):\n",
    "    \"\"\"Process the Cabin column into pre-defined 'bins' \n",
    "\n",
    "    Usage\n",
    "    ------\n",
    "\n",
    "    train process_cabin(train)\n",
    "    \"\"\"\n",
    "    df[\"Cabin_type\"] = df[\"Cabin\"].str[0]\n",
    "    df[\"Cabin_type\"] = df[\"Cabin_type\"].fillna(\"Unknown\")\n",
    "    df = df.drop('Cabin',axis=1)\n",
    "    return df\n",
    "\n",
    "def process_titles(df):\n",
    "    \"\"\"Extract and categorize the title from the name column \n",
    "\n",
    "    Usage\n",
    "    ------\n",
    "\n",
    "    train = process_titles(train)\n",
    "    \"\"\"\n",
    "    titles = {\n",
    "        \"Mr\" :         \"Mr\",\n",
    "        \"Mme\":         \"Mrs\",\n",
    "        \"Ms\":          \"Mrs\",\n",
    "        \"Mrs\" :        \"Mrs\",\n",
    "        \"Master\" :     \"Master\",\n",
    "        \"Mlle\":        \"Miss\",\n",
    "        \"Miss\" :       \"Miss\",\n",
    "        \"Capt\":        \"Officer\",\n",
    "        \"Col\":         \"Officer\",\n",
    "        \"Major\":       \"Officer\",\n",
    "        \"Dr\":          \"Officer\",\n",
    "        \"Rev\":         \"Officer\",\n",
    "        \"Jonkheer\":    \"Royalty\",\n",
    "        \"Don\":         \"Royalty\",\n",
    "        \"Sir\" :        \"Royalty\",\n",
    "        \"Countess\":    \"Royalty\",\n",
    "        \"Dona\":        \"Royalty\",\n",
    "        \"Lady\" :       \"Royalty\"\n",
    "    }\n",
    "    extracted_titles = df[\"Name\"].str.extract(' ([A-Za-z]+)\\.',expand=False)\n",
    "    df[\"Title\"] = extracted_titles.map(titles)\n",
    "    return df\n",
    "\n",
    "def create_dummies(df,column_name):\n",
    "    \"\"\"Create Dummy Columns (One Hot Encoding) from a single Column\n",
    "\n",
    "    Usage\n",
    "    ------\n",
    "\n",
    "    train = create_dummies(train,\"Age\")\n",
    "    \"\"\"\n",
    "    dummies = pd.get_dummies(df[column_name],prefix=column_name)\n",
    "    df = pd.concat([df,dummies],axis=1)\n",
    "    return df\n",
    "\n",
    "def pre_process(df):\n",
    "    df = process_missing(df)\n",
    "    df = process_age(df)\n",
    "    df = process_fare(df)\n",
    "    df = process_titles(df)\n",
    "    df = process_cabin(df)\n",
    "\n",
    "    for col in [\"Age_categories\",\"Fare_categories\",\n",
    "                \"Title\",\"Cabin_type\",\"Sex\"]:\n",
    "        df = create_dummies(df,col)\n",
    "    \n",
    "    return df\n",
    "\n",
    "train = pre_process(train)\n",
    "holdout = pre_process(holdout)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "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>Embarked</th>\n",
       "      <th>...</th>\n",
       "      <th>Cabin_type_A</th>\n",
       "      <th>Cabin_type_B</th>\n",
       "      <th>Cabin_type_C</th>\n",
       "      <th>Cabin_type_D</th>\n",
       "      <th>Cabin_type_E</th>\n",
       "      <th>Cabin_type_F</th>\n",
       "      <th>Cabin_type_G</th>\n",
       "      <th>Cabin_type_Unknown</th>\n",
       "      <th>Sex_female</th>\n",
       "      <th>Sex_male</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>Q</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>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>S</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</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>Q</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>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>S</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>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>S</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 41 columns</p>\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 Embarked    ...    Cabin_type_A  \\\n",
       "0  34.5      0      0   330911   7.8292        Q    ...               0   \n",
       "1  47.0      1      0   363272   7.0000        S    ...               0   \n",
       "2  62.0      0      0   240276   9.6875        Q    ...               0   \n",
       "3  27.0      0      0   315154   8.6625        S    ...               0   \n",
       "4  22.0      1      1  3101298  12.2875        S    ...               0   \n",
       "\n",
       "  Cabin_type_B Cabin_type_C Cabin_type_D  Cabin_type_E  Cabin_type_F  \\\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",
       "   Cabin_type_G  Cabin_type_Unknown  Sex_female  Sex_male  \n",
       "0             0                   1           0         1  \n",
       "1             0                   1           1         0  \n",
       "2             0                   1           0         1  \n",
       "3             0                   1           0         1  \n",
       "4             0                   1           1         0  \n",
       "\n",
       "[5 rows x 41 columns]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "holdout.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploring the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 891 entries, 0 to 890\n",
      "Data columns (total 3 columns):\n",
      "SibSp       891 non-null int64\n",
      "Parch       891 non-null int64\n",
      "Survived    891 non-null int64\n",
      "dtypes: int64(3)\n",
      "memory usage: 21.0 KB\n"
     ]
    }
   ],
   "source": [
    "explore_cols = [\"SibSp\",\"Parch\",\"Survived\"]\n",
    "explore = train[explore_cols].copy()\n",
    "explore.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEACAYAAAC3adEgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n2DXqAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8f7bcb2390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "explore.drop(\"Survived\",axis=1).plot.hist(alpha=0.5,bins=8)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEACAYAAABGYoqtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\no/4PD1wl4qjEmy0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8f7bb0e7b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "explore[\"familysize\"] = explore[[\"SibSp\",\"Parch\"]].sum(axis=1)\n",
    "explore.drop(\"Survived\",axis=1).plot.hist(alpha=0.5,bins=10)\n",
    "plt.xticks(range(11))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXEAAAEGCAYAAACToKXdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAv7P3KHFkN7ETZsAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8f7bb36a58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXEAAAEGCAYAAACToKXdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYIQ4ABiMEAcAgxHiAGCw/wMPh9d1r+hjsQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8f7bd36240>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXEAAAENCAYAAAD5Z1UeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nGEOciEhiDHEiIokxxImIJMYQJyKS2P8AuxaGQM5ImnoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8f84f39908>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "for col in explore.columns.drop(\"Survived\"):\n",
    "    pivot = explore.pivot_table(index=col,values=\"Survived\")\n",
    "    pivot.plot.bar(ylim=(0,1),yticks=np.arange(0,1,.1))\n",
    "    plt.axhspan(.3, .6, alpha=0.2, color='red')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By combining these two columns, we can see that only 30% of passengers who had no familyon board survived."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false
   },
   "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>Embarked</th>\n",
       "      <th>...</th>\n",
       "      <th>Cabin_type_B</th>\n",
       "      <th>Cabin_type_C</th>\n",
       "      <th>Cabin_type_D</th>\n",
       "      <th>Cabin_type_E</th>\n",
       "      <th>Cabin_type_F</th>\n",
       "      <th>Cabin_type_G</th>\n",
       "      <th>Cabin_type_Unknown</th>\n",
       "      <th>Sex_female</th>\n",
       "      <th>Sex_male</th>\n",
       "      <th>isalone</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>Q</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>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>S</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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>Q</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>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>S</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>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>S</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 42 columns</p>\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 Embarked   ...    Cabin_type_B  \\\n",
       "0  34.5      0      0   330911   7.8292        Q   ...               0   \n",
       "1  47.0      1      0   363272   7.0000        S   ...               0   \n",
       "2  62.0      0      0   240276   9.6875        Q   ...               0   \n",
       "3  27.0      0      0   315154   8.6625        S   ...               0   \n",
       "4  22.0      1      1  3101298  12.2875        S   ...               0   \n",
       "\n",
       "  Cabin_type_C Cabin_type_D Cabin_type_E  Cabin_type_F  Cabin_type_G  \\\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",
       "   Cabin_type_Unknown  Sex_female  Sex_male  isalone  \n",
       "0                   1           0         1        1  \n",
       "1                   1           1         0        0  \n",
       "2                   1           0         1        1  \n",
       "3                   1           0         1        1  \n",
       "4                   1           1         0        0  \n",
       "\n",
       "[5 rows x 42 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#new feature \n",
    "def process_isalone(df):\n",
    "    df[\"familysize\"] = df[[\"SibSp\",\"Parch\"]].sum(axis=1)\n",
    "    df[\"isalone\"] = 0\n",
    "    df.loc[(df[\"familysize\"] == 0),\"isalone\"] = 1\n",
    "    df = df.drop(\"familysize\",axis=1)\n",
    "    return df\n",
    "\n",
    "train = process_isalone(train)\n",
    "holdout = process_isalone(holdout)\n",
    "holdout.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Columns \n",
      "------------\n",
      "['Pclass', 'Age', 'SibSp', 'Parch', 'Fare', 'Age_categories_Missing', 'Age_categories_Infant', 'Age_categories_Young Adult', 'Fare_categories_12-50', 'Title_Miss', 'Title_Mr', 'Title_Mrs', 'Cabin_type_D', 'Cabin_type_E', 'Cabin_type_Unknown', 'Sex_female', 'Sex_male', 'isalone']\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/dataquest/system/env/python3/lib/python3.4/site-packages/sklearn/utils/__init__.py:54: FutureWarning:\n",
      "\n",
      "Conversion of the second argument of issubdtype from `int` to `np.signedinteger` is deprecated. In future, it will be treated as `np.int64 == np.dtype(int).type`.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.feature_selection import RFECV\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "def select_features(df):\n",
    "    # Remove non-numeric columns, columns that have null values\n",
    "    df = df.select_dtypes([np.number]).dropna(axis=1)\n",
    "    all_X = df.drop([\"Survived\",\"PassengerId\"],axis=1)\n",
    "    all_y = df[\"Survived\"]\n",
    "    \n",
    "    clf = RandomForestClassifier(random_state=1)\n",
    "    selector = RFECV(clf,cv=10)\n",
    "    selector.fit(all_X,all_y)\n",
    "    \n",
    "    best_columns = list(all_X.columns[selector.support_])\n",
    "    print(\"Best Columns \\n\"+\"-\"*12+\"\\n{}\\n\".format(best_columns))\n",
    "    \n",
    "    return best_columns\n",
    "\n",
    "cols = select_features(train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticRegression\n",
      "Best Score: 0.8204264870931538\n",
      "Best Parameters: {'solver': 'liblinear'}\n",
      "\n",
      "KNeighborsClassifier\n",
      "Best Score: 0.7755331088664422\n",
      "Best Parameters: {'p': 1, 'algorithm': 'brute', 'weights': 'uniform', 'n_neighbors': 5}\n",
      "\n",
      "RandomForestClassifier\n",
      "Best Score: 0.8406285072951739\n",
      "Best Parameters: {'min_samples_split': 5, 'max_depth': 10, 'max_features': 'sqrt', 'n_estimators': 9, 'min_samples_leaf': 1, 'criterion': 'entropy'}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "def select_model(df, features):\n",
    "    all_X = df[features]\n",
    "    all_y = df['Survived']\n",
    "    \n",
    "    models = [{\n",
    "    \"name\": \"LogisticRegression\",\n",
    "    \"estimator\": LogisticRegression(),\n",
    "    \"hyperparameters\":\n",
    "        {\n",
    "            \"solver\": [\"newton-cg\", \"lbfgs\", \"liblinear\"],\n",
    "        }\n",
    "}, \n",
    "    {\n",
    "    \"name\": \"KNeighborsClassifier\",\n",
    "    \"estimator\": KNeighborsClassifier(),\n",
    "    \"hyperparameters\":\n",
    "        {\n",
    "            \"n_neighbors\": range(1,20,2),\n",
    "            \"weights\": [\"distance\", \"uniform\"],\n",
    "            \"algorithm\": [\"ball_tree\", \"kd_tree\", \"brute\"],\n",
    "            \"p\": [1,2]\n",
    "        }\n",
    "}, \n",
    "    {\n",
    "    \"name\": \"RandomForestClassifier\",\n",
    "    \"estimator\": RandomForestClassifier(),\n",
    "    \"hyperparameters\":\n",
    "        {\n",
    "            \"n_estimators\": [4, 6, 9],\n",
    "            \"criterion\": [\"entropy\", \"gini\"],\n",
    "            \"max_depth\": [2, 5, 10],\n",
    "            \"max_features\": [\"log2\",\"sqrt\"], \n",
    "            \"min_samples_leaf\": [1, 5, 8], \n",
    "            \"min_samples_split\": [2, 3, 5]\n",
    "        }\n",
    "}]\n",
    "    for model in models:\n",
    "        print(model[\"name\"])\n",
    "        grid = GridSearchCV(model[\"estimator\"], \n",
    "                            param_grid=model[\"hyperparameters\"], \n",
    "                            cv=10)\n",
    "        grid.fit(all_X, all_y)\n",
    "        model[\"best_params\"] = grid.best_params_\n",
    "        model[\"best_score\"] = grid.best_score_\n",
    "        model[\"best_model\"] = grid.best_estimator_\n",
    "\n",
    "        print(\"Best Score: {}\".format(model[\"best_score\"]))\n",
    "        print(\"Best Parameters: {}\\n\".format(model[\"best_params\"]))\n",
    "\n",
    "    return models\n",
    "\n",
    "result = select_model(train,cols)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Make submission to Kaggle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\n",
    "def save_submission_file(model,cols,filename=\"submission.csv\"):\n",
    "    holdout_data = holdout[cols]\n",
    "    predictions = model.predict(holdout_data)\n",
    "    \n",
    "    holdout_ids = holdout[\"PassengerId\"]\n",
    "    submission_df = {\"PassengerId\": holdout_ids,\n",
    "                 \"Survived\": predictions}\n",
    "    submission = pd.DataFrame(submission_df)\n",
    "\n",
    "    submission.to_csv(filename,index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "best_model = result[2][\"best_model\"]\n",
    "save_submission_file(best_model, cols)"
   ]
  }
 ],
 "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": "3.4.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}