titanic/titanic-solution-using-speedml.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Titanic Solution Using Speedml\n",
    "\n",
    "### This notebook uses the [Speedml](https://speedml.com)  Python package to speed start machine learning projects.\n",
    "\n",
    "Speedml integrates best ML packages and popular strategies used by top data scientists in an easy to use Python package.\n",
    "\n",
    "> Using Speedml on Titanic dataset we **quickly jumped from low 80% rank to top 20% rank**, within just a few feature engineering iterations.\n",
    "\n",
    "Speedml is under active development and Kaggle version of the API may not be the latest. For demonstrating the latest Speedml features we have also created the same notebook on GitHub.\n",
    "\n",
    "### Download the latest version of [this notebook from GitHub](https://github.com/Speedml/notebooks/blob/master/titanic/titanic-solution-using-speedml.ipynb).\n",
    "\n",
    "### Table of Contents\n",
    "\n",
    "- Loading the datasets.\n",
    "- Differentiating between numerical, categorical, high-cardinality, and continuous features.\n",
    "- Feature correlation heatmap matrix.\n",
    "- Feature distribution for outliers detection.\n",
    "- Plotting continuous features for outliers detection.\n",
    "- Plotting categorical features for outliers detection.\n",
    "- Fixing outliers with a single line of code.\n",
    "- Plotting continuous features against categorical features.\n",
    "- Feature engineering for high-cardinality.\n",
    "- New feature extraction from existing features.\n",
    "- Hyper-parameters tuning for model classifier.\n",
    "- Model evaluation and accuracy ranking.\n",
    "- Model prediction and feature selection.\n",
    "- Saving the results."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Getting Started\n",
    "\n",
    "To get started all you need to do is include one package in your project. Speedml includes pandas, sklearn, numpy, xgboost, by default so you do not need to import these."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:16.281555",
     "start_time": "2017-05-16T11:30:10.992468"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from speedml import Speedml"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "It takes one line of code to initialize train, test datasets, define the target and unique id variables. This also initializes wrapper components for EDA (sml.plot), XGBoost (sml.xgb), modeling (sml.model), feature engineering (sml.feature) and more..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:16.341609",
     "start_time": "2017-05-16T11:30:16.284311"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'train (891, 11) | test (418, 10)'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml = Speedml('../input/titanic/train.csv', \n",
    "              '../input/titanic/test.csv', \n",
    "              target = 'Survived',\n",
    "              uid = 'PassengerId')\n",
    "sml.shape()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Configure\n",
    "\n",
    "Speedml workflow and analytics are user configurable. Following method call configures the threshold for identifying if we may be over-fitting our model based on ratio of features compared with samples. The method returns default (call without arguments) or updated configuration dictionary.\n",
    "\n",
    "> Here we are (re)setting the overfit_threshold to $\\frac{\\sqrt{N}}{N}$ where $N$ is the number of samples in the train dataset. The rule of thumb is to have number of highly correlated features to be less than $\\sqrt{N}$ to avoid over-fitting. If features are not correlated then this number can go as high as $N - 1$.\n",
    "\n",
    "Also note how intuitively Speeml exposes the methods like sqrt of popular packages like numpy and pandas objects like the train dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:16.355417",
     "start_time": "2017-05-16T11:30:16.345225"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'high_cardinality': 10,\n",
       " 'outlier_threshold': 3,\n",
       " 'outpath': 'output/',\n",
       " 'overfit_threshold': 0.033501260508640406,\n",
       " 'unique_ratio': 80}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.configure('overfit_threshold', \n",
    "              sml.np.sqrt(sml.train.shape[0]) / sml.train.shape[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Speed EDA\n",
    "\n",
    "Then returning results from just one Speedml method performs quick EDA on the entire dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:16.409961",
     "start_time": "2017-05-16T11:30:16.358401"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Results</th>\n",
       "      <th>Observations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Speedml Release</th>\n",
       "      <td>v0.9.3</td>\n",
       "      <td>Visit https://speedml.com for release notes.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Nulls</th>\n",
       "      <td>1014</td>\n",
       "      <td>Use feature.impute.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Outliers Upper</th>\n",
       "      <td>[SibSp, Fare]</td>\n",
       "      <td>Positive skew (&gt; 3). Use feature.outliers(upper).</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Shape</th>\n",
       "      <td>train (891, 11) | test (418, 10)</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numerical Ratio</th>\n",
       "      <td>54%</td>\n",
       "      <td>Aim for 100% numerical.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numerical High-cardinality</th>\n",
       "      <td>[Age, Fare]</td>\n",
       "      <td>(&gt;10) categories. Use feature.density</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numerical Categorical</th>\n",
       "      <td>[Survived, Pclass, SibSp, Parch]</td>\n",
       "      <td>Use plot.ordinal.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Text High-cardinality</th>\n",
       "      <td>[Ticket, Cabin]</td>\n",
       "      <td>(&gt;10) categories. Use feature.labels.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Text Categorical</th>\n",
       "      <td>[Sex, Embarked]</td>\n",
       "      <td>Use feature.labels or feature.mapping.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Text Unique</th>\n",
       "      <td>[Name]</td>\n",
       "      <td>~80% unique. Use feature.extract or feature.drop.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Target Analysis (Survived)</th>\n",
       "      <td>Model ready.</td>\n",
       "      <td>Use classification models.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                     Results  \\\n",
       "Speedml Release                                       v0.9.3   \n",
       "Nulls                                                   1014   \n",
       "Outliers Upper                                 [SibSp, Fare]   \n",
       "Shape                       train (891, 11) | test (418, 10)   \n",
       "Numerical Ratio                                          54%   \n",
       "Numerical High-cardinality                       [Age, Fare]   \n",
       "Numerical Categorical       [Survived, Pclass, SibSp, Parch]   \n",
       "Text High-cardinality                        [Ticket, Cabin]   \n",
       "Text Categorical                             [Sex, Embarked]   \n",
       "Text Unique                                           [Name]   \n",
       "Target Analysis (Survived)                      Model ready.   \n",
       "\n",
       "                                                                 Observations  \n",
       "Speedml Release                  Visit https://speedml.com for release notes.  \n",
       "Nulls                                                     Use feature.impute.  \n",
       "Outliers Upper              Positive skew (> 3). Use feature.outliers(upper).  \n",
       "Shape                                                                          \n",
       "Numerical Ratio                                       Aim for 100% numerical.  \n",
       "Numerical High-cardinality              (>10) categories. Use feature.density  \n",
       "Numerical Categorical                                       Use plot.ordinal.  \n",
       "Text High-cardinality                   (>10) categories. Use feature.labels.  \n",
       "Text Categorical                       Use feature.labels or feature.mapping.  \n",
       "Text Unique                 ~80% unique. Use feature.extract or feature.drop.  \n",
       "Target Analysis (Survived)                         Use classification models.  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.eda()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "**Observations**\n",
    "\n",
    "- Most model classifiers cannot handle NaN or null values. Data wrangle these to remove all null values.\n",
    "- Models require numerical datasets so converting 100% of features to numerical is a data wrangling and feature engineering goal.\n",
    "- High-cardinality features or features with lots of categories need to be addressed in a special manner for better model performance. One of the feature engineering strategies is banding of ranges of values for reducing the number of categories. Another strategy is to determine density of repeating values.\n",
    "- Categorical numerical features can be plotted differently (plot bars, histograms) from high-cardinality or continuous features (use scatter plots).\n",
    "- Non-categorical text features cannot be directly converted to numeric. These are candidates for feature extraction by identifying repeating patterns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:16.442494",
     "start_time": "2017-05-16T11:30:16.412017"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></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>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>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>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>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>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": [
       "   Survived  Pclass                                               Name  \\\n",
       "0         0       3                            Braund, Mr. Owen Harris   \n",
       "1         1       1  Cumings, Mrs. John Bradley (Florence Briggs Th...   \n",
       "2         1       3                             Heikkinen, Miss. Laina   \n",
       "3         1       1       Futrelle, Mrs. Jacques Heath (Lily May Peel)   \n",
       "4         0       3                           Allen, Mr. William Henry   \n",
       "\n",
       "      Sex   Age  SibSp  Parch            Ticket     Fare Cabin Embarked  \n",
       "0    male  22.0      1      0         A/5 21171   7.2500   NaN        S  \n",
       "1  female  38.0      1      0          PC 17599  71.2833   C85        C  \n",
       "2  female  26.0      0      0  STON/O2. 3101282   7.9250   NaN        S  \n",
       "3  female  35.0      1      0            113803  53.1000  C123        S  \n",
       "4    male  35.0      0      0            373450   8.0500   NaN        S  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Feature Correlations\n",
    "\n",
    "You can quickly check feature correlations using a plot. Learn how to interpret this plot at https://speedml.com/plot-correlation-of-features/ "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:17.047929",
     "start_time": "2017-05-16T11:30:16.444715"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaIAAAGVCAYAAAC8dPV6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEKb6f+JlwrZ56ft0AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1126549b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.correlate()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Outliers Detection and Fix\n",
    "\n",
    "We can use distributions to understand skew (left/right) for determining outliers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:18.145472",
     "start_time": "2017-05-16T11:30:17.049768"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112b319b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZkAAAGTCAYAAAD6Pm+oAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11509bdd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.distribute()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Continuous or high-cardinality numerical features are better plotted using scatter plot for determining outliers.\n",
    "\n",
    "We do not expect outliers in case of Age feature as the distribution plot is fairly close to normal (rising in the middle of x-axis and falling on either sides evenly)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:18.571694",
     "start_time": "2017-05-16T11:30:18.149964"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe0AAAFyCAYAAADYhIJtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1129a8198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.continuous('Age')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "The method clearly shows some outliers in case of Fare feature. This coincides with our observation from the distribution plot for the Fare feature which is skewed towards left."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:18.868911",
     "start_time": "2017-05-16T11:30:18.574499"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfIAAAFyCAYAAAAOIVgIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxQxySZJS7P8D6My2LULuUEMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1128b7a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.continuous('Fare')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "To correct the outliers we fix only values in upper range of the 99th percentile. As the results show these constitute around 1% of overall samples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:18.889148",
     "start_time": "2017-05-16T11:30:18.871825"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Fixed 17 or 1.91% upper outliers. '"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.outliers('Fare', upper=98)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "While we impact only a few samples, the outliers fix is fairly significant as shown by the same plot after the fix."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:19.169028",
     "start_time": "2017-05-16T11:30:18.891249"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfIAAAFyCAYAAAAOIVgIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112b31400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.continuous('Fare')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "We throw-in another plot into the mix just to understand how the remaining (potential) outliers are spread across when correlating with seemingly related features. We know that the passenger class and fare paid may be significantly correlated. As the plot points in upper range mostly stick to Pclass=1 we can safely assume no further fixing is required for the Fare feature."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:19.573606",
     "start_time": "2017-05-16T11:30:19.171302"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfIAAAEFCAYAAAAYBmVxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n8DAJ5EIIIYSH/R9MEz8GlY8CLgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11548b7f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.strip('Pclass', 'Fare')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:20.371842",
     "start_time": "2017-05-16T11:30:19.575895"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAecAAAEFCAYAAADZgS/dAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xl4HPd95/l3HX3ivkgCBEnwLN3WffmUbTlK7Nix51kn\nm3U2Tp7Y652dJ3H8JHbsHJPZeZKMd2N7nqzHcRLFWcfnzviU5fiQbMmSJcuHJEoUJZZEEuCBg8R9\ndaOvqv2jGo2GCAIkuhvdaHxez9NPF6qru7+U0Pj0r+p3GL7vIyIiIrXDrHYBIiIispzCWUREpMYo\nnEVERGqMwllERKTGKJxFRERqjF3tAhaNjs6q27iIiGwZXV1NxsUeU8tZRESkxiicRUREaozCWURE\npMYonEVERGqMwllERKTGKJxFRERqjMJZRESkxiicRUREaozCWUqWTCZIp9PVLkNEpG4onKUkCwsL\nvP/9/56/+Is/qXYpIiJ1Q+EsJZmeniKTSXP+/Ai+rxlYRUTKQeEsJVlYWChs69S2iEh5KJylJMlk\nYsVtERFZP4WzlGR+fn7FbRERWT+Fs5Rkbm62sD0/P1fFSkRE6ofCWUoyOzuz4raIiKyfwllKMjMz\ns+K2iIisn8JZSlLcWp6Zma5iJSIi9UPhLCUpDmSd1hYRKQ+Fs5RELWcRkfJTOEtJ5ubmMOw4oKFU\nIiLlonCWkszPz2PYEQzT1lAqEZEysSv1wo7jhIDPAn1ADniP67rHKvV+svE8zyOTSWOFWvHN0LKp\nPEVEZP0q2XL+FcB2XfdO4P8E/qqC7yVVUJhL27DAsEilUtUtSESkTlSs5Qy8CNiO45hAM5BZ7eC2\ntji2bVWwHCm3uTkDAMMwMQwD8OnqaqpuUSIidaCS4TxHcEr7GNAJvGW1gycntWjCZrP8GrOB53mM\njs5e9HgREVmyWmOmkqe1/xD4nuu6h4BXAJ91HCdawfeTDWbbwXc73/fw/RyWXcnveiIiW0cl/5pO\nsnQqewIIATpvXUdCoXCw4efA9wgv/iwiIiWpZDh/AviM4ziPAmHgI67raiBsHTFNk3A4TNbLgJch\nFotVuyQRkbpQsXB2XXcOeGelXl9qQzzewPRcCt/LEovFq12OiEhd0CQkUpLGxib8bKKwLSIipVM4\nS0mampqKtpurWImISP1QOEtJmptbCtstLS2rHCkiIpdK4SwlKQ5ntZxFRMpD4SwlKT6t3dyscBYR\nKQeFs5SkuLWslrOISHkonKUkxT20Gxoaq1iJiEj9UDhLSRoblwK5oaGhipWIiNQPhbOUJBpdmhUs\nHlc4i4iUg8JZShKNLq1lEgqFqliJiEj9UDhLSSKRSGE7WNNZRERKpXCWkoTDWolKRKTcFM5SEtvW\nqWwRkXJTOEtJbLuSq46KiGxNCmcpiWnqV0hEpNz0l1VERKTGKJxFRERqjMJZRESkxiicRUREaozC\nWUREpMYonEVERGqMwllERKTGKJxFRERqjMJZRESkxiicRUREaozCWUREpMYonEVERGqMwllERKTG\nKJxFRERqjMJZRESkxiicRUREaozCWUREpMYonEVERGqMwllERKTG2JV8ccdxPgy8FQgDn3Jd958r\n+X4iIiL1oGItZ8dxXgfcCbwSeC2wq1LvJSIiUk8q2XL+JeAI8HWgGfjj1Q5ua4tj21YFy5FK6+pq\nqnYJIiJ1oZLh3AnsAd4C7AXucxznCtd1/ZUOnpxMVLAU2Qijo7PVLkFEZNNYrUFTyXAeB465rpsG\nXMdxFoAu4HwF31NERGTTq2Rv7R8D9ziOYziO0wM0EAS21BHfX/FEiIiIlKBi4ey67v3A08DPgG8B\n/4frurlKvZ9URy6n/6UiIuVW0aFUrut+sJKvL9WncBYRKT9NQiIlyWYz1S5BRKTuKJylJJmMwllE\npNwUzlKSdDpd7RJEROqOwllKkkqlCtvZbLaKlYiI1A+Fs5RkYSG54raIiKyfwllKkkwuBXIioVne\nRETKQeEsJUkmE0XbajmLiJSDwllKUtxaTiTmq1iJiEj9UDhLSZaHs05ri4iUg8JZSlLcWlbLWUSk\nPBTOUpLl15zVchYRKQeFs5REp7VFRMpP4SwlSaUWirZTqxwpIiKXSuEsJSkO5OKgFhGR9VM4S0ky\nmXTRthbBEBEpB4WzlCSTyRZtK5xFRMpB4Swl8bzcitsiIrJ+CmcpSS7nAUbRtoiIlErhLCXxfQ8M\nI7/tV7kaEZH6oHCWknieB4a5tC0iIiVTOEtJcrkchmHlt7NrHC0iIpdC4SwlyeVyYNpL2yIiUjKF\ns5QklUphmCEwTNLp9NpPEBGRNdmXeqDjOB3Aa4Ac8CPXdacrVpVsCr7vB7OCReIYps3CQrLaJYmI\n1IVLajk7jvN24Djwh8AHgeOO49xVycKk9iWTCTzPw7DCYEWYm5urdkkiInXhUlvOfwW8xnXdIwCO\n49wI3AvcWKnCpPbNzMwAYFgRTCvN7Ow4nudhmrpaIiJSikv9K5pYDGYA13WfAjSodYsbHx8DwAzF\nMUJxfN9namqyylWJbB1HjjzDQw89oDkG6tCltpwfchznQ8AnCa45/6/Ac47jtAGG67oTlSpQatdS\nODeAH4xxHhsbpb29o5pliWwZn/jERwE4ePAKent3VbkaKadLbTm/H/gbYBZIAJ8GfgsYB0YrU5rU\nunPnRgAwwo0YocZl+0SksopHR0xOqn1Uby6p5ey6bqjShcjmsxjEZripsG9kZLha5YhsKePjo0Xb\nY1WsRCphzXB2HMcBJl3XPe84zk3AbwNPuq772YpXJzVtZGQIwwxhWFHMcLBPLWeRjXHu3LmibX3u\n6s2qp7Udx3kr8Dhw0HGcHuCHQAz4Tcdx/nAD6pMa5Xke586dwwg3YRgGhhXFMEOMjAxVuzSRLWF4\neHDFbakPa11z/jDwKtd1HwN+Ezjsuu57gLcD765wbVLDpqYmyeWymOHgWrNhGBihBsbGRtVzVGQD\nDA0tBfLg0NkqViKVsFY4x13XfSG//SrgOwCu6yZYXMR3FY7jbHMc54zjOFeUVqbUmmU9tfPMUAPp\ndJrZ2dlqlSWyZQwPD4EBdkeU8bExTZ9bZ9YKZwPAcRwDuBN4pOixxtWe6DhOCPgHQHM61qHZ2aUJ\nSBYZdhSAuTmFs0iljY2NYsZtrMagv25xBzHZ/NbqEPa84zh/QHCdOQM8kQ/qPwB+scZz/5ZgyNWH\nS65Sas78/DxAMHVn3uL2/Lym8RSpJN/3mZ2dwWoLY0aDJVtnZ2fp7q5yYVI2a4Xz+wmm6ewG3uW6\nruc4zn8D7s7fVuQ4zruBUdd1v+c4ziWFc1tbHNu2Lq1qqbpYLP+rYxSdfMlvNzVF6OpqWuFZIlIO\nmUwG3/cxLBOs4ApjPG7rc1dHVg1n13VHgLe8bPdfAr/vuu5qi/f+LuA7jvNG4HrgXx3HeWv+9VY0\nOZm4tIqlJszOLuS3Lux6MDk5z+ioTm2LVIrv+xiGge/5kAs6YM7PZ/S522RW+zJ1SZOQOI5jAu8h\naC1ngH8DPnex413XfU3Rcx8G3rdaMMvmE4kE15p9L1vYt7gdjUarUpPIVmEYBvF4A8lUCi8dtJMa\nG9VqrieXOn3nx4BfJwjlB4H3Oo7znytWldS8WCwWbHiZpZ357Wg0VoWKRLaWzs5O/GQWbz74Uqw5\n7evLpS58cTdwveu6WQDHcb4APAn8+VpPdF33deuuTmpWU1MzAF52obBvcbu5ubkqNYlsJdu3d3Pq\n1ACZ80laWlqXvjBLXbjUlvMsy4PcA+bLX45sFq2tbQD42aWRcn52Adu2iccbLvY0ESmTnp6dK25L\nfVi15ew4zgfymyPAI47jfI5gycjfAF646BOl7q0czkna2toxjDXnpxGREnV39xS2d+zoWeVI2YzW\nOq19bf5+Kn+7Mf/ziYpVJJuCbds0NTUzlwrC2fe9fDjvqXJlIlvDtm3bC9vbt29f5UjZjNYaSvU7\nG1WIbD4tLS3MDQWd8P1cCoDm5pZqliSyZbS1dRRtt1exEqmEtU5r/3fXdd/pOM4R4ILVDFzXva5i\nlUnNa25u4ezZM/heDj8bhPNiRzERqayGhoai7VVnU5ZNaK3T2h/NT9f5ASANtAAhoANwK1yb1LjC\nkCkvG9yAeFw9RkU2gmUtzagYCoWqWIlUwlrhnAT6gf8A/Bh4iqAFHQfeW9nSpNaFw8Fc2r6fLUxA\nEg5HVnuKiJRJ8dKsnudVsRKphLWGUv3fwJ+6rns/QQ9tCDqJvRr4j5UsTGrfyr2y1VNbZCMkk0tT\nHicSGtlab9YK592u634hv30X8E3XdXOu654hOMUtW5jvL35bNwqZvLRPRCppYmJixW2pD2uFc/Hi\nFi9fz1kTKG9xqVSwuLthWhiGnd+XqmZJIlvGyMhw0fZQFSuRSljrmvOE4zivAJoIlo38EYDjOHcC\ngxWuTWpcOp0PYsMGczGcF1Z5hoiUy+nTA0Xbp6pXiFTEWuH8EYKFLlqAD7quO+84zh8Bfwr8WqWL\nk9qWSCTAMIOWsxX0Fk0mk2s8S0TKob8/mAvKjFqcOtVPNpvFti91uQSpdaue1nZd9wlgJ7DNdd3/\nmt/9OHCr67o/qnRxUtuSyQSGGfTYNsxQYZ+IVJbneZw8eRyzMURoR5x0Os3g4NlqlyVltObXLNd1\n0wRjnBd/fryiFcmmkUwmC6ezF+8XFnRaW6TSRkfPkUwmCe9qxG6LkBqY5dSpfvbs6at2aVIml7oq\nlcgFUqkFjHwoG4YJhqVwFtkAQ0NBBzC7OYzVEs7vU8u5niicZd2y2SwYS79ChmEG+0SkoqamgqFT\nZtzGitn5fZPVLEnKTOEs6+Z5/rJwxjDwvNzFnyAiZZFO54cxWgZYRn5fppolSZkpnGXdTNOA4klH\nfB/TtC7+BBEpi8Vpcv2sh5/z8/s0v3Y9UTjLutm2vSycfd/TUA6RDdDW1gaAl8ziJbL5fR2rPUU2\nGYWzrFs0GisseOH7Hvg5olFNHCdSaT09vQBkp9PkpoNT3N3dPdUsScpM4SzrFo1Gwctf58qHdGEZ\nSRGpmM7OLmKxGNmpNNmpYKa+vr69Va5KyknhLOsWi8Xx8+G8eB+LKZxFKs00TfbtO4A3lyFzLkko\nFGLnzl3VLkvKSOEs6xaPx8H38L0cfi6ztE9EKm7PnqCl7CWz7N69B8tSZ8x6onCWdVvsMYqXLZzW\njkR0zVlkI+zc2VvYXrwGLfVD4SzrFonkh3P4WXw/COdwOFzNkkS2jLa29hW3pT4onGXdDMO4pH0i\nUn6h0NIXYX0prj8KZ1k33/cvaZ+IlF/xCnBaDa7+KJxl3VKpYJELwwxhGHZ+X6qaJYlsGYODZ4q2\nB6tYiVSCwlnWLZlMBhumDZbWcxbZSM8883SwYRm88MJzWhGuziicZd2mp6cwrAiGYWJY0fy+6SpX\nJVL/Tp8e4Nix57E7o8QOtpBKpXj00YerXZaUkcJZ1m1qagrDDkLZsMJgmFq2TqTCfN/ny1/+PAAx\np5Xo/hYM2+Rb3/oas7MzVa5OykXhLOuSSCRIJhMYdjDpiGEYGHaM8fGxKlcmUt8efvhBjh17ntCO\nOKFtMcyIRezKVubm5vjCF/5fdcqsEwpnWZexsVEAzFBDYZ8ZamBmZrqw1qyIlNfQ0CBf/vLnMcMW\nDTd0FoYuRg+0YLdH+NnPnuDxxx+tcpVSDgpnWZfx8Xw4h4vDuTH/mFrPIuWWy+X4x3/6b2QyGeI3\ndGLFlpZnNQyDxlu2YYRMPv/5f9FnsA5UJJwdxwk5jvM5x3EedRznZ47jvLUS7yPVs9hyNopazkY+\nqBcfE5HyefjhBzl9aoDI7kYiOxsueNxqCBG/toNUKlW4Ji2bV6Vazu8Cxl3XfTVwD/DJCr2PVMn4\n+Djw8tPa8fxj+tYuUk6e5/Fv//YtDNskfm3HRY+L7GnEbovw5JM/Y2RkeAMrlHKrVDj/D+DP89sG\nkK3Q+0iVTEwEAZyZ6mfh3GEADDsIaoWzSHkNDQ0yOTlBqCeOGVlafWr+yDjzR8YLPxuGQWRvEwBH\njz674XVK+dhrH3L5XNedA3Acpwn4CvBnaz2nrS2ObWvJs81iZmYKDJPs3FDw9Wv79YWWcyIxQ1dX\nU3ULFKkjIyM5gGXXmQHSg/MANBS1ps14cIzvZ/Q53MQqEs4AjuPsAr4OfMp13S+udfzkpGaW2kyG\nh0fyw6i8wj4jFAMMBgeHGR2drVptIvXGzp+Vyk6vPRIiNxUcE4s163NY41b78lSpDmHbge8DH3Jd\n9zOVeA+pnkRintnZGczw8l8sw7AwQnFd6xIps87OLvbs2UtmJEF25uIB7ec8Fk7OYIdCXHvt9RtY\noZRbpa45fwRoA/7ccZyH87dYhd5LNtjZs8GE+2ak+YLHzHAzMzPTzMxoGk+RcnrrW98BwPzhsYtO\nNJJ0p/ASWV5/1900N1/4+ZTNo1LXnP8A+INKvLZUn+seA8CKdZKdPbPsMSveSW5+mBdfPMbNN99W\njfJE6tL119/IDTfcxNNPP0mqf5bovuXhm51OkXSnaWtr521v+3dVqlLKRZOQyGU7fPhJAKx41wWP\n2fHtADz99JMbWpNIvTMMg9/6rd8lGo2RODqBl8oVHvN9n/nD4+D7/PZv/x6xmE5UbnYKZ7ks/f0n\n6O8/gdXYg5lf9KKYGevADDfx858/wcyMJuEXKafW1jbe9rZ34Gc8FvqXPl/Z8QWy4wtcf/2NXHed\nrjXXA4WzXDLf9/n61/8HAOG2gyseYxgGobaDZLNZvvWtr29keSJbwmtecxeGYZA5nyzsW9x+7Wvf\nUK2ypMwUznLJHn/8UZ577lmshh1YDTsuelyodT9muIkf/vD7vPSSu4EVitQ/3/eDBS+K+4T5S49J\nfVA4yyUZGDjJ5z73LximTbT7lsJqOCsxTItI9634vs/f//3fMTExftFjReTSJZMJ7r337/E8j/CO\neGF/KL/9pS/9K+fOjVSrPCkjhbOs6fz5c3ziE/8X6XSKSPfty+bTvhg73kVk2yuYmprk4x//KPPz\ncxtQqUh9ymTSPProw/z5n3+Iw4efwu6KEj3YUng81BEl5rQyOnqev/iLD3HffV/TcMZNzqiV0yCj\no7O1UYgsMzh4lo997G+Ympoksv0mwu3LrzXPHb8PgMYDFy485vs+qXNPk5l8kV279vCBD/wJLS0t\nFxwnIhfKZrMcP/4ihw8/yWOPPRp8wTUg5rQSu6INwzSY/O5pANru2Y3v+6QH55k/PI6fzmHZNrfc\nfCs33ngrV199DbFYfI13lI3W1dV00VOQCme5qP7+E4VWb2Tb9YQ7rrjgmNXCGcD3PVIjT5KZOsG2\nbdv5oz/6CJ2dFw7BEtnqcrkcg4NnOH78JVz3BZ577hmSyaCjlxm2CPc1Ed3XjBVfmp6iOJwXeRmP\n1OlZUidmyM1lALAsC8e5kiuvvJoDBw7R17ePSCSygf86WYnCWS7b4cNP8ulP/z+k0xmi3bcQat23\n4nFrhTMELej06BHS48/T3NzC+9//x/T1rfx6IluB53mcP3+O06dPcebMACdPnuDEyeOkU6nCMWbc\nJrwjTqg7TqgzhmFd+Hd8pXBe5Ps+uak06eF50iOJwpzbAKZpsnt3H/v3H2D37j52795DT08voVCo\nAv9auRiFs1yWH/zg+3zxi58FwyTScwehpt6LHnsp4bwoPfEiqXNPEQ6Hed/7fp/rr7+xbDWL1CLf\n95mZmWFo6CxDQ2cZHDzLmTOnOHPmNOn08jmyraYQdkcUuz1KqCOC2RhateMlrB7OL+ctZMmMp8hO\nBGOis1Np8Jb+7JqmSXd3D7t399Hbu4vu7p3s3NlLR0cnpqnuSZWwWjhXbFUq2Xx83+eb3/wq9933\nNQw7Sqz31Vixiy/sfrnC7YcwQnFSQz/hk5/8OL/7u/8bd9756rK9vki1+L7P5OQEw8NDDA2dzd8P\nMjh4hvn5+eUHG2A1hwnvaMRuCWO3RLBaw5jhyi6Za0ZtIjttIjuDDp1+ziM7nSY3nSY7lSY3nWLo\n3CCDg2eXPS8UCtPT00NPTy89PTvp7u6hu3sn27Ztx7K0zG+lKJwFCP64fOlL/8qDD34PM9RAbPfr\nLlh1qhxCTb2Yu+8ieeYR7r3370kkErzxjb9U9vcRqQTf95mYGGdw8Axnz55haGiQ4eFBhoeHWFhY\nWH6wAWZDiFB3HLs5jNUUwmoOYzWFVzxFvdEMyyTUHiXUvjTTn+/7ePNZcjNpsjNpcrMZcjNpTp89\nxalTA8ueb9k227ftoKenJ9/K3kVv7y62b9+h0C4DhbMA8M1vfjUI5kgLsV2vwwxVbm5eK9ZJbPfr\nSZ55mC9+8bM0NDRwxx2vqtj7iaxHNptlcPAMJ08e58yZ05w9GwTywkJy+YGmgdVoE+5oCAK4KR/E\nTSEMa3OdDjYMA6sxhNUYItyzNGSyOLRzsxlys8H98PngTEExy7bp6e6ht3c3vb272Lt3P319+4hG\nL5zuVy5O4Sw89tgj3Hff1zBDjcR237XinNmrWU+/BSvaSmz3XSRPPchnPvOPtLd34DhXXvbriJRL\nLpfjhReO8vzzz3Hy5HH6+0+SyRRdFzYIQqujIWgBN4exm0OYDSEMs3ot4Y3oN1Qc2i9/by+ZC8J6\nJrhlpzOcHTrDmTOnlz2/t3cX+/cf5NChK3jFK27U4hxrUIewLW509Dx/+md/TM4ziO15I9YKazRf\nTG5hikT/9wAfI9xEbOcrsaKtl/X+2flzJM88TGtLK3/1V3+rD6xsuImJcb773fv52c9+srRYS/66\nsN0WwW6PYreGa64lnJ1OM/3Ds+CD2Rii6bbt2C3hapcFLLW0s9MpshPBLTeVws8Ff+ZDoRA33HAz\nd999D/v3rzxP/1ag3tpyUX/3d3/L4cNPEe25nVBL32U9d+7Et/HTs4WfzXATDfvffNk1pEaPkB47\nyj33vJl3vvN/uezni6xXNpvlP/2njzA4eBYjbBHe2UBkZwN2ewTDrp0gXsnk98/g5ccxQxDQbW/a\nVcWKVud7PrnpNOmRBKnTc3jzQe0f/eh/patrW5Wrq47Vwrm2f/ukol588RiHDz+FFd+G3bznsp7r\nZZPLghnAS8/iZZMXecbFhTuuxAw18P3vf5fJyYnLfr7Iep061V/onRy/uo34Ne2EtsVqPpi9heyy\nYAbw5jJ4C9kqVbQ2wzSw2yLEnFYab+gs7P/5z5+oYlW1q7Z/A6Wivve9fwMg3HXNmuMpL+DlLm//\nKgzTJtRxFZ6X4wc/+P5lP19kvfbtO8Db3/4/Yds280+PMXn/ANM/GiJxbJLMWBI/41W7xBUtnh6+\n1P3V5Hs+2akUCyemmfnJCJP3n2Lmx8MA3HHHq3jta19f5QprkzqEbVFnz57m8OEnMaPtWLHqT6cZ\naukjPfosP/zhA9xzz5tpbCz/MC6RlzMMg1/91bdz22138uMf/4jnnz9Cf/9JsuMLLJ4DMhtD2K1L\n45Gt5jBm1Lr8L7RbgJfOBb25p1Jkp9Jkp1PkZjLLJjvZvn0H11xzHbfccjuHDl04JbAEFM5b1Fe+\n8v/h+z7R9bSaK8AwLcKdV7Fw7mnuv/+b/MZvvKvaJckWsm3bdt7xjnfyjne8k7m5OY4dO8qJE8c5\nfXqAgVP9JM/Okz67NJmIETKXxi03h7GaQ9hNYYwtEtpexiv0zg56agfjob2F5WfObNtm75697Nmz\nl76+fVx11TWaW/8SKZy3oPvu+xrPPvs0VnwbVkN3tcspCLUeIDPxIg888B36+vZx++13Vrsk2YIa\nGxu5+ebbuPnm24Cg5/HY2CinTw9w+vSpYBrOoUHOnxshO5Fa9lwjZGI158c65ycesZs3b2gvC+Gi\nMc5e8sLLV23t7ew8sIudO3vp7d3Fnj197NjRg20rZtZD/9W2EM/z+Pa37+Mb3/gKZqiBaM9tNfUH\nwzAtojtfSfL0Q9x776cAn9tvf2W1y5ItzjAMurq20dW1jZtuurWwP5PJcO7cMIODwZzZw8ODS6E9\n/rLQjljBLGEtwc1uCcK7muOji/m+j5fIkp1KBdN55qf19BIXdjBrb++gZ38wlefOncF9T89OLUlZ\nZhpKtUW89JLLl778OQb6T2KG4sR2vx4z3Lju1/PSc8yfuP+C/Q3731LS6wLkkmMkTz+M72W59tpX\n8Ou//i56enaW9JoiGyWbzTIyMszQ0CBDQ2c5c+Y0Z86cYmxsdNlxhm0G46g7IvnFLqIYoUvro5ub\nzzD1vTMX7G/9pV1YDWuvLOV7PtnJVLAAxsQC2fEUXmp5a7ipqZndu/fQ27u7EMLd3Ts1F0EZaeGL\nLez06VPcf/83+MUvfgqA3bybyLbrMUO1+y3XinUS2/NGUuee4siRZzh69Aive90beNObfoVt27ZX\nuzyRVdm2TW9vMM90sWQykZ8C9DQDA/2cOPESQ0ODZEbzXc9Mg/COGOHeRsI74mUfzuX7PtmxBVJn\n58gMJvDSS2Hc2trGgWsPsXfvPnbt2sOuXbtpabm8CYWkvNRyrkOTkxM88cRjPP6THzN4Nvh2bUY7\niG6/ASveucazL00lW86LfN8nNzdE6vxhvPyY6gMHDnHnna/mlltuo6GhPO8jUi1zc3OcPHmcl15y\nefrpXzA0NAgEazk33bEduyWy4vMut+XspXPM/ew8mfPBF4Hm5mZuvPEWHOdKDhw4RHt7R01d4toq\nNEPYFjA5Oclzzz3DE088xrFjzwfz7RomdmMPdksfduPOsn74NiKcF/l+juz0aTLTA+QS5wCwLJvr\nr7+BW265g6uuuobGRgW1bH5nz57hkUd+yIMPfg8jZNL25j0rXpe+3HCeeXyEzEiCa665jl/+5V/F\nca7UGs01QKe161AqtYDrvsDRo89x9OizhW/cEJwWtlv6CDXvwrBW/ua9mRiGRah1L6HWvXiZBJmZ\nU2SnB3jyyZ/z5JM/xzAM+vr2ctVV13LNNdexf/9B9RCVTWdmZppjx57n5MnjQDC3N2X6Pm23hMmM\nJBgZGeaFF47S0tKqfhw1Ti3nTSKTyTAwcDIfyEc4fvxFcrn8NSPDCqbgbNiB3dRTkXWYX24jW84r\n8X0fLzXIifhFAAATJ0lEQVRFdm6I3PwIueQY5H+Xw+EIV1xxJVdddS2OcwW9vbu1vqzUlPn5OU6f\nPsWpU/0MDPRz6lQ/586NBA8aEOqK0XBjF1Z85S+Zl9ty9j2f+cNjpM/O42eDWc9aWlrp6wvGIO/Z\n08eePXtpa2vX6e0NpNPam1AiMc/x4y/y4osuL73k0t9/gmx2aViDGW3DbtiB1bADK9aJYW5s+FQ7\nnF/Oz2XIJc6TnT9Hbn4ELz1TeCwajbJ//0EOHnQ4eNBh374DRCKb/4yC1C7f95mZmWF09Bznz59j\ndPT8svuZmellxxshE6s1TLg7WHjDjK1+5me9vbX9rEd6OEF6cI7sZBovuXyoVDQaY9u27Wzbto2u\nru357e10dW2jvb1Dp8LLTKe1N4GJifF8EB/jpZdcBgfPFq3TamBGWwk1dWHFO7Hi2zFthUsxwwph\nN+3EbgpO1XmZBLn5c+SSo6QTYxw9eoSjR48AYJoWe/b0cfCgw6FDQWA3NV36UpmytXmex/T0FJOT\nE0xOTubvJ5iaWtqenJwgnU5f+GQj6OwV2hbDao1gt4WxWyOYcXtDWqyGbRLZ1UhkV/AF2lvIkZ1K\nBeObp1Jk5jKcGTzF6dMDFzzXsixaW9toa2sv3Ae34n1thEK1sWzlZqeWcxX4vs/58+d48cVjvPji\nMVz3heVjIA0LK9aBFe/CinUGLWNr7bGLG6nWWs5r8bIL5JJj5BJj5BKjeKmJwmlwgJ6eXhznCg4d\nCm5tbe1VrFY2mud5JBLzTE9PMzMzzfT0VP5+uuh+qvDzan83zYiFEbWwGmzMhhBWQwizwcZqDGHG\n7LJMPFLqOOfV+L6Pv5AjN58lN5/Bm8+Qm8/izWfwkrlg5atV/lo3NDTS0tJCS0srzc0ttLS05O+D\nnxf3NTU1b/nLTWo514Dx8TGeffZpXPcFXPcY09NThccMK4zV2IMd34YV78KMtmEYOn1UTqYdxWzq\nJdTUC4DvZcklx8klRsklRxkaGWZo6CwPPfQgAF1d2zh06AquuOIqrrvuerWsN6F0Os3MzDSzszPM\nzMws2w7ulz/meauvQGXYJkbExGoPY8ZszKiNGbOC7ZiNGbXKFr7VZBgGRv7fFOqMXvD4Ynh7Czm8\nZDZ/y5FbCLYXFlIkRoeXdVK92PssBnlTUzPNzc00NbXk74Ofm5uXHotGY1vqerjCuUJ83+fMmWDl\np6ee+sWy00SGHcVu2hXMbR3vwoy0bKlfulpgmDZ2w3bshmBSE9/38BYmg+vWiVHGJsYYfewRHnvs\nEQzD4OBBhxtuuIkbbrhZE6FUie/7LCwki1qyQagu3hZbtYu3VCq15msatokRNjFbQ1gRKwjYiIUZ\ntTGW/WzV/BrPG6U4vGm7+OU1P+fjpXL4qaC17aWCQPfz914qR2Jhgfnz8/j5NbVXY9t2IawXW+PF\nt6V9zTQ0NG766+M6rV1mQ0ODPPzwD3j66ScZH8+fqjbMoDd1407sxh0YocZNH8ab7bT25Sr0Bp8f\nITs7iJccKzzW09PLjTfezF13vVGnv8skk8kUrtlOTU0yMTHB1NRE/j7YPzMzTSaTWf2FjPxp5Yi1\n7H75tokRtTDDmzdwK3lauxp8zw9CO3+7cNsLtvOhXrwE5UpM06SpqXnFa+PFt2j0wjMDG6kqvbUd\nxzGBTwGvAFLA77mue/xix2/2cE6lUtx//9f5znfux/M8DDOE1didD+RuDKu+OknUezi/nJddIDs7\nSHZukNz8CPgekWiUd7z9nbzhDW/a9N/SN0I2m+X8+XMMDw8yNDTE8PAgw8NDTEyMMTs7e/EnGkvX\ncQut2Ii91KoteswIm5v+i++lqLdwvhy+7+Nn/aIWeLYQ2sta5/kw93MXj5ZYLEZbWwc7dnTT3d2T\nnz+8hx07ejYkuKt1zfnXgKjrunc4jnM78DHgbRV8v6px3Rf453/+B8bGzmOE4kS7b8Bu6sEw6r+z\nQzgcprOzk7GxsZV7p9YJ044SbttPuG0/vpchM32K9OizfOlL/8rjjz/Ke97z7zWpwwpGR8/z0EMP\n8MwzT3Pu3MiF13VNAzNuY3dFl13HtYqv4W7S5RY3wlb5/BUzDAMjZEDIxGpc/YuI7/v4Ga9wXbz4\nGrm3kCWdzDI8OsTQ0IWn1dvbOzh06Ape//q72b//4Ib/DlYynF8FfBfAdd0nHMe5uYLvVVVf/vLn\nGRs7T6jtIJFtr8Awt8al/HA4zPve9z7uvvtuHnjgAT796U9Xu6QNYZghwm0HsJt6SY38glOn+rn/\n/q/z3vf+h2qXVlM+85l/5LHHfoTv+xi2idUaItQUrHFs5e/Nho0ZQlSPturn73IYhoERDi5h0LLy\nMYXe6XOZojWrM0zNTvHEE4/xxBOPsW/fft7//g/S2Fj5CZ4WVTJFmoHikfY5x3Fs13UvXCAUaGuL\nY9ubs6V5++23cupUP2aoAbZAaxkA06Kzs5O7774bgLvvvpuvfOUrTG/wZCjVZFhhDCs49fXKV95B\nV9fGfXA3g7NnBwpDjkLbY9htkSCUW/JDihTK62ZYxoqfv6Sl/6aXq7iDm9UcDgJ6Jk12xiZ9Zg4/\n6zM8PERDg72hn/FKhvMMUPwvMS8WzACTk4kKllJZt976ar761a+SOn+YzHQ/obYDhJr7am5scjmZ\ndozxmSQPPPBA4Zv7xEySUEv9r/XqZRfITJ0gM3USPzNPS0sbhw5dx+joKtdNt6APf/gveeihH3D/\nt7/B3OAs6cH5wmOGbQYt58ZQYTiStTgkKRZ03lJ4X5wZtZlIjC///CWmaYhqJMHFLJ3iLj69nR8G\nlsjgzWQuWNPasm1e94Y38Ja3vA3DiJX9M75a2FeyQ9i/A37Vdd135685/0fXdX/5Ysdv9g5h/f0n\n+O53v82TT/4cz8sFQ3Wa+7Cbe/PTa9bfqe7cwhTe6E9pb44xMZPE7LoNK1qfa8D6uTS5xGiw6Mbs\nWfA9wuEId9zxSu65581s395d7RJrlud5jI6eY3DwLIODgwwOnmFo6CzDw0NL88O/nGkUrjmbsaLe\n1lELM98ZbPHnzT6ueL2y02lST07SHm9hIjFN5KY27Jb66nh6KXy/qKf3y4ZqeakcfjJHLhl0Gluc\nV/zlDMOgs7OLnTt72bmzl56e4L67u6eiM55Vu7f2dQRrq/yO67rHLnb8Zg/nRdPTUzzyyEM8/PAP\nmZwcD3YaJla0PRjX3LCt7sLaz6Xrrjf6YhhnE+fJJc7jLUwWHuvp2cldd72RO+54NfF4vIpVbm65\nXK4wTGpiYoKJifH81JfB/fjEONNTU6vOxgXBvNSFsC705r5wCJURsTBso+5a5F7GwwzV12gBP1c8\ndMq7yPCqfBinc6vOWAbQ2NhEe3sH7e1Lw6ja2zvy9+20tXUQDm/83zAtfFEFuVyO559/jhdeeI5j\nx17g1Kn+pT8yxWEd68SMtWPa1R1vt5UFQzMS5JIT+Sk+l4exZdns33+AK664iquvvpYDBw7V3R/4\nWpXL5VacZOTlE45MT08zN3cJpxxNIxjnvNL45xVCfau2ysvN9338dFHILuRW3k7n8FPeRVu4xaLR\n2LLJSFbabmlppaWlpWbn+1Y414BEIsFLL7m47vMXhjVg2HGsWDtmtD0I7lh73bVGa4WXSeItTJBb\nmCCXnMBbmMDPLc0mVRzGjnMl+/cfrMq3ark82WyW2dlZZmdXmp5zhtnZ6WVTd17K0CMjbGFGzWUB\nbkaXhnkt/mzUWcv1Uvk5P5j9ayE/C1h+iFJhes+iVu5arVvLsmhqbqZ52RSeLTQ1NRW2i/fXw2dS\n4VyDEokEx4+/SH//CQYGTtLff3KFZeQasWJBWJvRNqxomwL7MnnZhWBazoUJvGQQyH42ueyYjo5O\n9u7dR1/fPvbu3a8w3iJSqYVCcBe3wi9opc9Mk0ys3mHVsPOzjhWFthW3MeM2ZjyE1WBvutnI/JyP\nl8wGnaUSWbz5fAeqxeBdPKW8inAkQkvz0rSaK023uTgdZywW33JnpBTOm4Dv+0xOTtDff5KBgRMM\nDPTT33+SRGJ+2XFGuAkr2qbAXkEQxBPkFibzQTyJn13+R7W5uYV9+/bT17evcGtu1qIWsrpsNltY\nrWpqaoqpqcnCLdg3yeTkJLOzMxd9DSNsYcYXQzuE1RjCag7GfJuR6gxB9LNeflxvfnxvIou3GMbJ\niwdvLBajtbXtgltbW3C/GL6RiC7XrUbhvEn5vs/o6HkGBk4yMNDPqVP9DAz0k0wuDxwz3JQP6nbM\nWEcQ2HXU4Wwlfi4dXCNeGA+CODWJn7kwiPv69rJnz1727Olj7979mgtbKmoxxCcmxhkfH2NsbIzx\n8dGi+9EV5wc3IlZ+cpYQZrSyn10/4xXC2EtcOLrVNE3a2zvo7Oyio6OTjo7OwnZ7ewetra0K3TJR\nONeRpcBeDOuTDJzqX37azTAwI23BmtCxjqB3eKhh054y8n0fLz1DLjmGlxgPAjm1/BJAEMT7CmHc\n17eX1ta2Tftvlvrk+z4zMzOMjZ1nZCRYVnF4eJDBobOMjY6u2TO9nFpaWunp2ZmfTzqYU7qraxtt\nbe1bfp3ljaJwrnPFLewTJ45z8uRLDAwMkMstfSs27ChWtAMz1oHdsL2m14z2vVyh13QuOYaXnMD3\nllob4XCEffv2s3//AfbtO0hf3z7a2tqqWLFI6dLpNCMjw5fW67wEkUiE7u4e4vGGir6PrE3hvAVl\nMhlOnx7gxImXOHHiOMdPvMTkxHjhccMKY8W3YzVsx27YUdWVpAot4/kRsnMjeMlRfG/pi8X27d0c\nOHCQffsOsH//QXbu7NU3exHZ9BTOAsDk5AQvvfQizz9/hKNHjzA+vrRGsRluxIrvwG7aidWwveKt\nat/LkZ0byi/BeG5ZD+ru7h6uvvo6rrrqGg4cOLihk82LiGwUhbNcwPd9zp8f4bnnjvD880d44YWj\nLCwsAGDYMezmPYRa+so6Hafv++SSY2SnB8jOnC6cqm5qauKqq67l6quv5aqrrqG9vaNs7ykiUqsU\nzrKmbDbLyZPH+elPH+enP/1JYQiXGWnFburFKHG1KT+XJjtzBi8zB0Braxt33PEqbr31dnbt2oNp\n1ub1bxGRSlE4y2XJZDI8++zTPP74ozzzzGE8b/WJBi5VOBzmpptu5c47X82VV16tQBaRLU3hLOs2\nMzPDwMDJkod4WJbFgQOHiEY1PlJEBBTOIiIiNWe1cNZ5RRERkRqjcBYREakxCmcREZEao3AWERGp\nMQpnERGRGqNwFhERqTEKZxERkRpTM+OcRUREJKCWs4iISI1ROIuIiNQYhbOIiEiNUTiLiIjUGIWz\niIhIjVE4i4iI1BiFs4iISI2xq12AbF6O45jAp4BXACng91zXPV7dqkS2DsdxbgM+6rru66pdi5SX\nWs5Sil8Doq7r3gH8CfCxKtcjsmU4jvNB4F4gWu1apPwUzlKKVwHfBXBd9wng5uqWI7KlnADeUe0i\npDIUzlKKZmC66Oec4zi6VCKyAVzX/SqQqXYdUhkKZynFDNBU9LPpum62WsWIiNQLhbOU4jHgVwAc\nx7kdOFLdckRE6oNOQUopvg7c7TjO44AB/E6V6xERqQtaMlJERKTG6LS2iIhIjVE4i4iI1BiFs4iI\nSI1ROIuIiNQYhbOIiEiN0VAqkU0iP5b8b4AOgi/WZ4A/cl33aBle+31Aq+u6/6UMr3Uz8BXXdftK\nfS2RrUrhLLIJOI4TAe4H3uS67lP5fe8CvuM4zl7XdXOlvL7rup8uQ5kiUiYKZ5HNIQ60Ao1F+75A\nMIXqGxzH+bjrutcAOI7zOuCTrute4zjOXwJ3AN3Ac8Crgbe7rvuL/LFfBn4EbAc6gfuAj7mue23+\n8VagH9iXr+GTwG4gBHzZdd2/zh/3vwN/SDDXumaKEymRrjmLbAKu604CHwS+6zjOScdxPkcwI9uD\nQHqNp+8BbnRd9zeBzwDvBnAcpw24G/hi0bEPAI35U9MA/zPw7fz7fw74jOu6NwG3Am90HOedjuNc\nD/wl8BrXdW+5hHpEZA0KZ5FNwnXdjxO0cH8fGAY+BDwNtKzx1CeKFiT5DPBOx3HCBMH7Ldd1CyuL\nua7rA/9MPsAJvgDc6zhOA/Ba4D87jnMYeIKgBX098Abg+67rjuSf84+l/DtFROEssik4jvNKx3H+\n2HXdWdd173dd94PA1YBHEJBG0eHhlz19bnHDdd1TwFPAWwiC959WeLt/IQjw6wk6iT0MWPn3uNN1\n3etd170euB34a8B/2ftrZTKREimcRTaHUeDPHMd5VdG+bqAB+Aaw23GcbY7jGMCvrfFa/0TQ6o67\nrvvYyx90XXcQ+CnwD8C9+X0zBK3lD0DhWvRjwNsIToW/yXGc3vxLvHs9/0ARWaJwFtkEXNd9kSB0\n/zp/zfl54L8D73Vd9xmCIP0FQYAOr/Fy9wF9BKevL+afgBuAzxbt+03gdsdxjhCE95dc1/2C67pH\nCK6H/8BxnF8A0cv994nIclqVSkREpMao5SwiIlJjFM4iIiI1RuEsIiJSYxTOIiIiNUbhLCIiUmMU\nziIiIjVG4SwiIlJj/n9EIIPfGRynkQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1129fb4e0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fixed 7 or 0.79% upper outliers. \n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAecAAAEFCAYAAADZgS/dAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEEIIUWEknIUQQogKI+EshBBCVJj/H55Vt+hds1//AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11549cda0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe0AAAEFCAYAAADOo78UAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nrgAAABBJREFUIWKEBG0hhBAiRvx/sITRsnIhuEQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115472278>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.ordinal('SibSp')\n",
    "print(sml.feature.outliers('SibSp', upper=99))\n",
    "sml.plot.ordinal('SibSp')\n",
    "sml.plot.strip('SibSp', 'Age')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Feature Engineering For High-Cardinality\n",
    "\n",
    "High-cardinality features like Ticket and Age are candidates for feature engineering. We use the density method to create a new feature based on Age and Ticket and drop the Ticket feature in turn. This simple iteration improves our model significantly and helps us jump 100s of positions on the Kaggle leaderboard."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:20.418823",
     "start_time": "2017-05-16T11:30:20.374357"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Results</th>\n",
       "      <th>Observations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Speedml Release</th>\n",
       "      <td>v0.9.3</td>\n",
       "      <td>Visit https://speedml.com for release notes.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Nulls</th>\n",
       "      <td>1014</td>\n",
       "      <td>Use feature.impute.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Shape</th>\n",
       "      <td>train (891, 11) | test (418, 10)</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numerical Ratio</th>\n",
       "      <td>54%</td>\n",
       "      <td>Aim for 100% numerical.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numerical High-cardinality</th>\n",
       "      <td>[Age, Fare]</td>\n",
       "      <td>(&gt;10) categories. Use feature.density</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numerical Categorical</th>\n",
       "      <td>[Survived, Pclass, SibSp, Parch]</td>\n",
       "      <td>Use plot.ordinal.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Text High-cardinality</th>\n",
       "      <td>[Ticket, Cabin]</td>\n",
       "      <td>(&gt;10) categories. Use feature.labels.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Text Categorical</th>\n",
       "      <td>[Sex, Embarked]</td>\n",
       "      <td>Use feature.labels or feature.mapping.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Text Unique</th>\n",
       "      <td>[Name]</td>\n",
       "      <td>~80% unique. Use feature.extract or feature.drop.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Target Analysis (Survived)</th>\n",
       "      <td>Model ready.</td>\n",
       "      <td>Use classification models.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                     Results  \\\n",
       "Speedml Release                                       v0.9.3   \n",
       "Nulls                                                   1014   \n",
       "Shape                       train (891, 11) | test (418, 10)   \n",
       "Numerical Ratio                                          54%   \n",
       "Numerical High-cardinality                       [Age, Fare]   \n",
       "Numerical Categorical       [Survived, Pclass, SibSp, Parch]   \n",
       "Text High-cardinality                        [Ticket, Cabin]   \n",
       "Text Categorical                             [Sex, Embarked]   \n",
       "Text Unique                                           [Name]   \n",
       "Target Analysis (Survived)                      Model ready.   \n",
       "\n",
       "                                                                 Observations  \n",
       "Speedml Release                  Visit https://speedml.com for release notes.  \n",
       "Nulls                                                     Use feature.impute.  \n",
       "Shape                                                                          \n",
       "Numerical Ratio                                       Aim for 100% numerical.  \n",
       "Numerical High-cardinality              (>10) categories. Use feature.density  \n",
       "Numerical Categorical                                       Use plot.ordinal.  \n",
       "Text High-cardinality                   (>10) categories. Use feature.labels.  \n",
       "Text Categorical                       Use feature.labels or feature.mapping.  \n",
       "Text Unique                 ~80% unique. Use feature.extract or feature.drop.  \n",
       "Target Analysis (Survived)                         Use classification models.  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.eda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:20.608751",
     "start_time": "2017-05-16T11:30:20.421326"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Age_density</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>22.0</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38.0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>26.0</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>35.0</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35.0</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Age  Age_density\n",
       "0  22.0           27\n",
       "1  38.0           11\n",
       "2  26.0           18\n",
       "3  35.0           18\n",
       "4  35.0           18"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.density('Age')\n",
    "sml.train[['Age', 'Age_density']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:20.781179",
     "start_time": "2017-05-16T11:30:20.611452"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Ticket_density</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>PC 17599</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>113803</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>373450</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Ticket  Ticket_density\n",
       "0         A/5 21171               1\n",
       "1          PC 17599               1\n",
       "2  STON/O2. 3101282               1\n",
       "3            113803               2\n",
       "4            373450               1"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.density('Ticket')\n",
    "sml.train[['Ticket', 'Ticket_density']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:20.794339",
     "start_time": "2017-05-16T11:30:20.783356"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Dropped 1 features with 12 features available.'"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.drop(['Ticket'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Extracting New Features\n",
    "\n",
    "We will now extract new features Deck from Cabin and FamilySize from Parch and SibSp."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:20.827322",
     "start_time": "2017-05-16T11:30:20.797061"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>SibSp</th>\n",
       "      <th>0.0</th>\n",
       "      <th>1.0</th>\n",
       "      <th>2.0</th>\n",
       "      <th>3.0</th>\n",
       "      <th>4.0</th>\n",
       "      <th>5.0</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Survived</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>398</td>\n",
       "      <td>97</td>\n",
       "      <td>15</td>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>210</td>\n",
       "      <td>112</td>\n",
       "      <td>13</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "SibSp     0.0  1.0  2.0  3.0  4.0  5.0\n",
       "Survived                              \n",
       "0         398   97   15   12   15   12\n",
       "1         210  112   13    4    3    0"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.plot.crosstab('Survived', 'SibSp')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:20.859882",
     "start_time": "2017-05-16T11:30:20.829273"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Parch</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Survived</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>445</td>\n",
       "      <td>53</td>\n",
       "      <td>40</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>233</td>\n",
       "      <td>65</td>\n",
       "      <td>40</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Parch       0   1   2  3  4  5  6\n",
       "Survived                         \n",
       "0         445  53  40  2  4  4  1\n",
       "1         233  65  40  3  0  1  0"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.plot.crosstab('Survived', 'Parch')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:20.896571",
     "start_time": "2017-05-16T11:30:20.861707"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Filled 1014 null values across test and train datasets.'"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "'Dropped 1 features with 12 features available.'"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.fillna(a='Cabin', new='Z')\n",
    "sml.feature.extract(new='Deck', a='Cabin', regex='([A-Z]){1}')\n",
    "sml.feature.drop(['Cabin'])\n",
    "sml.feature.mapping('Sex', {'male': 0, 'female': 1})\n",
    "sml.feature.sum(new='FamilySize', a='Parch', b='SibSp')\n",
    "sml.feature.add('FamilySize', 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:21.554551",
     "start_time": "2017-05-16T11:30:20.898659"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAEFCAYAAADKVm8pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAFl5JREFUeJzt3X+UnVV97/H3JJNkCIE0QG7V2lar8qVWpbK4GCBSbAG7\nECVV1JBqS1KKFNG26nJx/dFal11WKtrayo+gI2jlrltRrMYSGMWWH5Fr0XtvY1f5KooXll5tGgMk\nIZNMknP/eJ4xZ8b5cYLnyck+836tlZV5zj7znO/Omczn7OfH3gOtVgtJklSOeb0uQJIkHRzDW5Kk\nwhjekiQVxvCWJKkwhrckSYUZ7HUBndqyZbuXxUuS5pTly48amOpxR96SJBXG8JYkqTCGtyRJhTG8\nJUkqjOEtSVJhDG9JkgpjeEuSVBjDW5KkwhjekiQVxvCWCjI8vJ7Vq1cxPLy+16VI6iHDWyrE6Ogu\nRkZuBWBkZCOjo7t6XJGkXjG8pUKMjY3RalVT/Lda+xkbG+txRZJ6xfCWJKkwhrckSYUxvCVJKozh\nLUlSYQxvSZIKY3hLklQYw1uSpMIY3pIkFcbwliSpMIa3JEmFGWxqxxExD7gaOBHYDVycmQ+0tf82\n8GZgHzCcmdc0VYskSf2kyZH3KmAoM08FrgCumtT+fuAs4HTgzRGxrMFaJEnqG02G90pgI0Bm3guc\nPKn9X4GlwBAwALQarEWSpL7R2GFz4Gjg0bbtfRExmJl76+1vAF8DdgKfycxHZtrZsmWLGRyc30yl\nUgEWLtw/YfvYY5ewdOlRPapGUi81Gd6PAe2/WeaNB3dEPA94CfB0YAfwdxHxysz81HQ727bt8QZL\nlQ5/27fvmLC9desO9uzxmlOpny1fPvUH9Cb/598DnAsQESuAzW1tjwK7gF2ZuQ/4D8Bz3pIkdaDJ\nkfctwNkRsYnqnPbaiFgDLMnM9RFxHXB3ROwBvg3c0GAtkiT1jcbCOzP3A5dOevj+tvZrgWuben1J\nkvqVJ8wkSSqM4S1JUmEMb0mSCmN4S5JUGMNbkqTCGN6SJBXG8JYkqTCGtyRJhTG8JUkqjOEtSVJh\nDG9JkgpjeEuSVBjDW5KkwjS5JKg0p/3uPe/q6v5au/ZO2H79/7ySgSO6+1/4xtPf1dX9SWqGI29J\nkgpjeEuSVBjDW5KkwhjekiQVxvCWJKkwhrckSYUxvCVJKozhLUlSYQxvSZIKY3ir7wwPr2f16lUM\nD6/vdSmS1AjDW31ldHQXIyO3AjAyspHR0V09rkiSus/wVl8ZGxuj1WoB0GrtZ2xsrMcVSVL3Gd6S\nJBXG8JYkqTCGtyRJhTG8JUkqjOEtSVJhDG9JkgpjeEuSVBjDW5KkwhjekiQVxvCeg5z7W5LKZnjP\nMc79LUnlM7znGOf+lqTyGd6SJBXG8JYkqTCDTe04IuYBVwMnAruBizPzgbb2/wp8ABgAfgC8JjNH\nm6pHkqR+0eTIexUwlJmnAlcAV403RMQAcD2wNjNXAhuBX2ywFkmS+kaT4T0eymTmvcDJbW3HA1uB\nP46IfwaOycxssBZJkvpGY4fNgaOBR9u290XEYGbuBY4DTgMuBx4ANkTEfZl5x3Q7W7ZsMYOD8xss\nd25YuHD/hO1jj13C0qVH9aia7uv3/jVt+XL/raQSNBnejwHtvwnm1cEN1aj7gcz8d4CI2Eg1Mp82\nvLdte7ypOueU7dt3TNjeunUHe/b0z3WL/d6/pm3Zsr3XJUhqM90H6iZ/q90DnAsQESuAzW1t3wGW\nRMQz6+0XAv/WYC2SJPWNJsP7FmA0IjYBH6Q6v70mIi7JzD3A7wE3RcS/AA9n5hcarEWS1DCnXj50\nGjtsnpn7gUsnPXx/W/sdwClNvb4k6dCZPPXymjWvZWjoiB5X1b+aPOctzegzGy/o+j53j7YmbG/4\n0loWDQ10bf8v/82bu7YvqZ9MNfWy4d0cr+SRJKkwhrckSYUxvCVJKozhLUlSYQxvSZIK49Xmh7nv\nfmZNV/e3c/fEq7Ef2vA6jlzUvauxn/bym7q2L00yv+19Gpi0LWlOceQtFWJg4XzmPfc4AOY95zgG\nFjrXvzRXOfKWCrLgzKfCmU/tdRmSesyRtyRJhZlx5B0RDwKt6doz85e6XpEkSZrRbIfNx+evvAzY\nA6wH9gJrgYUN1iVJkqYxY3hn5tcAIuI5mfmCtqY3RcRXG61MkiRNqdNz3j8TEcvHNyLiKcDRzZQk\nSZJm0unV5n8FbI6I26juMD0HeGtjVUmSpGl1NPLOzGuAFwP/B/hfwFmZ+fEmC5MkSVM7mFvFngUc\nQ3XR2nObKUeSJM2mo/COiCuAPwBeBQwBfxoR72yyMEmSNLVOR96rgXOBnZm5FVgBdHfSbUmS1JFO\nw3ssM3ePb2TmI8BYMyVJkqSZdHq1+cMR8RKgFRGLgLcA/7e5siRJ0nQ6De/LgU8AzwN2AvfiYXMd\nhua1LbQ1MDBxW5L6RafhvTMzfyMiFgPzM3N7k0VJT9SCBQM884QBHri/xTNigAULXPNaUv/pNLwf\njIgNwPWZeVeTBUk/rZNWzOekFb2uQpKa02l4Px24EHh/RCwFPgLcmJlbGqtMjRicV02R16I6rDzo\norCSVJxOZ1h7NDOvrRcneTXVamMPN1qZGrFowQArnll9ZlvxjEEWeVhZkorT6cibiDgJuIhqopav\nAq9sqCY17PyTFnL+Sa7oqsPP8PB6br/9HznnnHNZt+6SXpcjHbY6nWHtX4FPAT8ETsrMl2Xm5xut\nTNKcMjq6i5GRWwEYGdnI6OiuHlckHb46HXm/OTNHGq1E0pw2NjZGq9UCoNXaz9jYGENDR/S4Kunw\nNGN4R8RbM/NK4GUR8dLJ7Zn5xsYqkyRJU5pt5P1o/fd/Nl2IJEnqzIzhnZnX1V/+ALjJyVkkSeq9\nTu/yPRP4TkQMR8SpDdYjSZJm0el93hcCxwP3AX8dEZsj4g8brUySJE2p4/m1MnMbsB54L9XiJFc0\nVZQkSZpeR7eK1RO0rKWamOXrwJXA5xqsS5IkTaPT+7w/C3wUOCUzH2qwHkmSNItOw/uuzPyzRiuR\nJEkd6fSc969EhCtYSJJ0GOh05P0D4N8i4l5gx/iDzrAmSdKh12l4f6X+07GImAdcDZwI7AYuzswH\npnjeeuBHmenV65IkdaCj8H6C57tXAUOZeWpErACuAs5vf0JEvA54LvDPT2D/kiTNSZ3eKrYZaE1+\nPDOfN8O3rQQ21s+7NyJOnrTP04AXANcBJ3RasCRJc12nh80vb/t6IdWo+vuzfM/RHFjYBGBfRAxm\n5t6IeDLwp8BvAa/qpIBlyxYzODi/w3L7x3d7XcBBWr78qF6X0Cj715yFC/dP2D722CUsXdrf/979\nxPfv0Or0sPmEw9oR8UVgE/DnM3zbY0D7OzcvM/fWX78SOA74R+BJwOKIuD8zb5huZ9u2Pd5Jqeqx\nLVv6e+0a+9ec7dt3TNjeunUHe/Z0PAmkesz3rxnTfaDudOQ92bHAU2Z5zj3AS4G/r895bx5vyMwP\nAR8CiIiLgBNmCm5JknTAEznnPQD8AtW56pncApwdEZvq71kbEWuAJZm5/gnWK0nSnDdreNeTs7wJ\n2AMspbr167OZuXmm78vM/cClkx6+f4rn3dBpsZIkaZYZ1iLi2cCDwCLgq8D7gAuBz0fE2c2XJ0mS\nJpvtaoK/BN6emRuA1fVjzwFeCLyrwbokSdI0ZgvvX8jMT9Zfvwj4h8zcn5kPUx1ClyRJh9hs4b2v\n7evTgDvbtoe6X44kSZrNbBes/SgiTqS6X/vJ1NOY1rOjfa/h2iRJ0hRmC++3AV+kOkT+1szcGRFv\nAd5ONcuaJEk6xGYM73pO8p8DFmfmI/XDm4BTMvNbjVcnSZJ+wqz3eWfmHqp7vMe3NzVakSRJmpET\nz05heHg9q1evYnjYieAkSYcfw3uS0dFdjIzcCsDIyEZGR3f1uCJJkiYyvCcZGxuj1aqmcW+19jM2\nNtbjiiRJmuiJriomSSrYfXfs7ur+do3umbD9v+/awxFD3XuNk399Udf21Q8ceUuSVBjDW5IOES+G\nVbcY3pJ0CHgxrLrJ8JakQ8CLYdVNhrckSYUxvCVJKkz5t4rd/A/d3d/u0Ynbn7sVFnVx9dMLzu/e\nviRJc5Ijb0mSCmN4S5JUGMNbkqTCGN6SJBXG8JYkqTCGtyRJhTG8JUkqjOEtSVJhDO9JFsyfz0D9\n9QADLJg/v6f1SJI0meE9ydDgAs5+xgkAnP2MYGhwQY8rkiRpovKnR23Auuefyrrnn9rrMiRJmpIj\nb0mSCmN4S5JUGMNbkqTCGN6SJBXG8JYkqTCGtyRJhTG8JUkqjOEtSVJhDG9JkgpjeEuSVJjGpkeN\niHnA1cCJwG7g4sx8oK39QuCPgL3AZuCyzNzfVD2SJPWLJkfeq4ChzDwVuAK4arwhIo4A3gO8KDNP\nB5YC5zVYiyRJfaPJ8F4JbATIzHuBk9vadgOnZebj9fYgMNpgLZIk9Y0mVxU7Gni0bXtfRAxm5t76\n8PgPASLiDcASYGSmnS1btpjBwZ9cW3tL9+o9JJYvP+qgnv/dZspozMH2rzT274Dzbv5kV1+7NTrx\n8/sbNm1kYGioq6+x4YLf7ur+DsbChRPPCh577BKWLu3lz9PuHr72wev3/3sHq8nwfgxo/9eel5l7\nxzfqc+JXAscDr8jM1kw727bt8Zmai7Fly/Zel9Ao+1c2+9ec7dt3TNjeunUHe/Z4zXCn+v1nczrT\nfWhp8ifnHuBcgIhYQXVRWrvrgCFgVdvhc0mSNIsmR963AGdHxCZgAFgbEWuoDpHfB/wecBdwR0QA\n/HVm3tJgPZIk9YXGwrs+r33ppIfvb/va40WSJD0BBqgkSYUxvCVJKozhLUlSYQxvSZIKY3hLklQY\nw1uSpMIY3pIkFcbwliSpMIa3JEmFMbwlSSpMk3ObS1Kx/uDOb3V1f/tHd07YfstXvsO8oSO7tv9r\nznhW1/alw58jb0mSCmN4S5JUGMNbkqTCGN6SJBXG8JYkqTCGtyRJhTG8JUkqjOEtSVJhDG9Jkgpj\neEuSVBjDW5KkwhjekiQVxvCWJP3U5s9fAAwAMDAwUG+rKYa3JOmntnDBEL/67HMAOPGXz2HhgqEe\nV9TfXBJUktQVZ61cx1kr1/W6jDnBkbckSYUxvCVJKozhLUlSYQxvSZIKY3hLklQYw1uSpMIY3pIk\nFcbwliSpMIa3JEmFMbwlSSqM4S1JUmEMb0mSCmN4S5JUGMNbkqTCNLYkaETMA64GTgR2Axdn5gNt\n7S8F/gTYCwxn5vVN1SJJUj9pcuS9ChjKzFOBK4CrxhsiYgHwQeAc4NeASyLiZxusRZKkvtFkeK8E\nNgJk5r3AyW1tvww8kJnbMnMPcDdwRoO1SJLUNwZarVYjO46IjwCfzsxb6+2HgF/KzL0RsRJ4Q2a+\num57N/BQZn6kkWIkSeojTY68HwOOan+tzNw7TdtRwCMN1iJJUt9oMrzvAc4FiIgVwOa2tn8HnhUR\nx0TEQqpD5l9psBZJkvpGk4fNx682fx4wAKwFTgKWZOb6tqvN51Fdbf7hRgqRJKnPNBbekiSpGU7S\nIklSYQxvSZIKY3hLklSYxqZHLUlEvAB4X2aeOenxoqdwrWeyGwaeBiwC3pOZn2trL71/84HrgQBa\nwKWZ+Y229qL7Ny4i/gvwNeDszLy/7fHi+xcRX6e6dRTgwcxc29ZWdP8i4r8BLwMWAldn5kfb2krv\n20XARfXmEPCrwJMy85G6vfT+LQBupPrduQ/4/cPt/96cH3lHxFuBj1D9ALY/3g9TuL4G2JqZLwR+\nE/jb8YY+6d9LATLzdOAdwJ+PN/RJ/8b7cR2wa4rHi+5fRAwBA5l5Zv2nPbiL7l9EnAmcBpxOVf/P\nt7UV3TeAzLxh/H2j+mD5xrbgLr5/VLc5D2bmacC7OQx/t8z58Aa+Dbx8isf7YQrXTwHvrL8eoPqU\nOK74/mXmZ4FL6s1fZOJEP8X3r/Z+4Frg+5Me74f+nQgsjojbI+KOej6IcaX378VUc1vcAnwe2NDW\nVnrffiwiTgZ+JTPXtz3cD/37JjBY3/J8NDDW1nZY9G/Oh3dmfpqJb8y4o4FH27a3A0sPSVFdkpk7\nMnN7RBwF3Ew1Oh1XfP8A6ul2bwT+BvhkW1Px/asPTW7JzNumaC6+f8DjVB9OXgxcCnwyIsZP5ZXe\nv+Oo1nN4JQf6NlC3ld63dm8D/mzSY/3Qvx1Uh8zvpzo196G2tsOif3M+vGfQF1O4RsTPA18GPpGZ\nN7U19UX/ADLzd4Hjgesj4sj64X7o3zrg7Ij4J6pzih+PiCfVbf3Qv28Cf5eZrcz8JrAVeHLdVnr/\ntgK3ZeaezExgFFhet5XeNwAi4meAyMwvT2rqh/79MdX7dzzVEaIb69M8cJj0zwvWpvfjKVypPoWd\nQTVKKEZ9HuZ24PLM/NKk5n7o32uBp2bme6lGcfvrP9AH/cvMHx+KqwP80sz8Qf1Q8f2j+nDyXOCy\niHgK1Yjm/9VtpffvbuAPI+IDVB9IjqQKdCi/b+POACb/XoH+6N82DhyR/RGwAJhfbx8W/TO8J4mI\nNRyYwvVNwG0cmML1e72t7qC9DVgGvDMixs99Xw8c2Sf9+wzwsYi4k+o/1x8BvxUR/fL+/YQ++/n8\nKHBDRNxNdbfAOuBV/fD+ZeaGiDgD+CpV/a8HXt0PfWsTwHd+vNFfP5sfBIYj4i6quwXeBpx/OL1/\nTo8qSVJhPOctSVJhDG9JkgpjeEuSVBjDW5KkwhjekiQVxlvFpMJERAv4BtWCCePuy8yLu7DvlwFn\nZeYb63vL/zYzb57h+ScAV3Fg7u5twNsz8+763u2b6/mhJXWR4S2V6UWZ+Z/d3mm96tznZn3iAZ8G\n3pGZtwDU9zZ/ISKenpnfp1qcQ1KXGd5SH4mIdcDrqCaWOAb4i8y8pp4n/RXAEVRzNj8EfBi4nGpq\n2Q9k5lX18y7IzPPa9vl2qsUn1tTbp1ONyJ/PgdnDAMjMOyPiVcC+iHga8I3MXBIRHwOeXz9tIdXi\nDmdl5pfq/b+C6jTed4HL6uCXNA3DWyrTlyOi/bD5OVRTxP4+cG5mbq1X6RoBrqmf80Kq6Ui/R7Xi\n1WrgN+rH7o2ID07zWtcD34qIYzLzR1QfDq6t214PXB0RV1JNCXoncFNmPhoRy8Z3ML7cZ704x03A\nP9XB/Tv1659SLzJzCdUSvec+4X8ZaQ4wvKUyTXnYPCLOA14SEc+iWsxkSVvzv2Tmw/XzHgRuz8z9\nEfFtqvXsF0/1Qpn5HxGxAXhtRHycahWwy+q2/x4RtwArqeZ4Xge8Y9Lynu2uolrI4TX19nnAKcB9\nEQHV/NFT1iHpAMNb6hMR8VTgK8B6qlHwzVThOG73pG+Zainc6XyYagS/F/h0Zu6oL1a7KDOvAL5Y\n//mTiBgBLqhfv72+NwO/BpyRmeNHDeYD78vMa+rnLKKaj1/SDLxVTOofJwNbgPfUa4CfBxAR82f8\nrg5k5iaqFdvewoHD8D8ELomIC8afV6+09LPA19u/PyIupDrEfl5m7mxrug24OCKOrrffDXzip61X\n6neOvKX+cTvVYeuMiJ1UK1ptAZ7Zpf1/DHh1Zm4GyMxtEfHrwHsj4v3ATqrR/V9m5h31BWvjbqA6\n1/6FiBgfNFxLdZTg56jOubeoLqS7qEv1Sn3LVcUkzSoiBoHPAp/IzP/R63qkuc7D5pJmFBHPphrB\nPwp8qsflSMKRtyRJxXHkLUlSYQxvSZIKY3hLklQYw1uSpMIY3pIkFeb/A6Rk25Oi5B9fAAAAAElF\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1153f8470>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.bar('FamilySize', 'Survived')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.178035",
     "start_time": "2017-05-16T11:30:21.557229"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAEFCAYAAADKVm8pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115492f98>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.bar('Deck', 'Survived')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.191176",
     "start_time": "2017-05-16T11:30:22.180084"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Dropped 2 features with 11 features available.'"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.drop(['Parch', 'SibSp'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Here is a single line of code to impute all empty features values (numerical and text) with numerical median or most common text value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.243316",
     "start_time": "2017-05-16T11:30:22.194358"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Imputed 179 empty values to 0.'"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.impute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.266913",
     "start_time": "2017-05-16T11:30:22.245950"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 891 entries, 0 to 890\n",
      "Data columns (total 11 columns):\n",
      "Age               891 non-null float64\n",
      "Age_density       891 non-null int64\n",
      "Deck              891 non-null object\n",
      "Embarked          891 non-null object\n",
      "FamilySize        891 non-null float64\n",
      "Fare              891 non-null float64\n",
      "Name              891 non-null object\n",
      "Pclass            891 non-null int64\n",
      "Sex               891 non-null int64\n",
      "Survived          891 non-null int64\n",
      "Ticket_density    891 non-null int64\n",
      "dtypes: float64(3), int64(5), object(3)\n",
      "memory usage: 83.5+ KB\n",
      "----------------------------------------\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 418 entries, 0 to 417\n",
      "Data columns (total 10 columns):\n",
      "Age               418 non-null float64\n",
      "Age_density       418 non-null int64\n",
      "Deck              418 non-null object\n",
      "Embarked          418 non-null object\n",
      "FamilySize        418 non-null float64\n",
      "Fare              418 non-null float64\n",
      "Name              418 non-null object\n",
      "Pclass            418 non-null int64\n",
      "Sex               418 non-null int64\n",
      "Ticket_density    418 non-null int64\n",
      "dtypes: float64(3), int64(4), object(3)\n",
      "memory usage: 35.9+ KB\n"
     ]
    }
   ],
   "source": [
    "sml.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.658546",
     "start_time": "2017-05-16T11:30:22.269274"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAksAAADSCAYAAABaS4SgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSVIFw5IkSVKF/w8IDG0DsfM9ugAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x107d35208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.importance()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.687750",
     "start_time": "2017-05-16T11:30:22.660823"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Age_density</th>\n",
       "      <th>Deck</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>FamilySize</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Name</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Ticket_density</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>22.0</td>\n",
       "      <td>27</td>\n",
       "      <td>Z</td>\n",
       "      <td>S</td>\n",
       "      <td>2.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38.0</td>\n",
       "      <td>11</td>\n",
       "      <td>C</td>\n",
       "      <td>C</td>\n",
       "      <td>2.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>26.0</td>\n",
       "      <td>18</td>\n",
       "      <td>Z</td>\n",
       "      <td>S</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>35.0</td>\n",
       "      <td>18</td>\n",
       "      <td>C</td>\n",
       "      <td>S</td>\n",
       "      <td>2.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35.0</td>\n",
       "      <td>18</td>\n",
       "      <td>Z</td>\n",
       "      <td>S</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Age  Age_density Deck Embarked  FamilySize     Fare  \\\n",
       "0  22.0           27    Z        S         2.0   7.2500   \n",
       "1  38.0           11    C        C         2.0  71.2833   \n",
       "2  26.0           18    Z        S         1.0   7.9250   \n",
       "3  35.0           18    C        S         2.0  53.1000   \n",
       "4  35.0           18    Z        S         1.0   8.0500   \n",
       "\n",
       "                                                Name  Pclass  Sex  Survived  \\\n",
       "0                            Braund, Mr. Owen Harris       3    0         0   \n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...       1    1         1   \n",
       "2                             Heikkinen, Miss. Laina       3    1         1   \n",
       "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)       1    1         1   \n",
       "4                           Allen, Mr. William Henry       3    0         0   \n",
       "\n",
       "   Ticket_density  \n",
       "0               1  \n",
       "1               1  \n",
       "2               1  \n",
       "3               2  \n",
       "4               1  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.786595",
     "start_time": "2017-05-16T11:30:22.689733"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Sex</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Title</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Capt</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Col</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Countess</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Don</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dr</th>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Jonkheer</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lady</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Major</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Master</th>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Miss</th>\n",
       "      <td>0</td>\n",
       "      <td>182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mlle</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mme</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mr</th>\n",
       "      <td>517</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mrs</th>\n",
       "      <td>0</td>\n",
       "      <td>125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ms</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Rev</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sir</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Sex         0    1\n",
       "Title             \n",
       "Capt        1    0\n",
       "Col         2    0\n",
       "Countess    0    1\n",
       "Don         1    0\n",
       "Dr          6    1\n",
       "Jonkheer    1    0\n",
       "Lady        0    1\n",
       "Major       2    0\n",
       "Master     40    0\n",
       "Miss        0  182\n",
       "Mlle        0    2\n",
       "Mme         0    1\n",
       "Mr        517    0\n",
       "Mrs         0  125\n",
       "Ms          0    1\n",
       "Rev         6    0\n",
       "Sir         1    0"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.extract(new='Title', a='Name', regex=' ([A-Za-z]+)\\.')\n",
    "sml.plot.crosstab('Title', 'Sex')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.824341",
     "start_time": "2017-05-16T11:30:22.789753"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Replaced matching list of strings across train and test datasets.'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "'Replaced 2 matching values across train and test datasets.'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "'Replaced 2 matching values across train and test datasets.'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "'Replaced 1 matching values across train and test datasets.'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.replace(a='Title', match=['Lady', 'Countess','Capt', 'Col',\\\n",
    "'Don', 'Dr', 'Major', 'Rev', 'Sir', 'Jonkheer', 'Dona'], new='Rare')\n",
    "sml.feature.replace('Title', 'Mlle', 'Miss')\n",
    "sml.feature.replace('Title', 'Ms', 'Miss')\n",
    "sml.feature.replace('Title', 'Mme', 'Mrs')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.839513",
     "start_time": "2017-05-16T11:30:22.827661"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Name</th>\n",
       "      <th>Title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                Name Title\n",
       "0                            Braund, Mr. Owen Harris    Mr\n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...   Mrs\n",
       "2                             Heikkinen, Miss. Laina  Miss\n",
       "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)   Mrs\n",
       "4                           Allen, Mr. William Henry    Mr"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.train[['Name', 'Title']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:22.881495",
     "start_time": "2017-05-16T11:30:22.841764"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Dropped 1 features with 11 features available.'"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Age_density</th>\n",
       "      <th>Deck</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>FamilySize</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Ticket_density</th>\n",
       "      <th>Title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>22.0</td>\n",
       "      <td>27</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>38.0</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>26.0</td>\n",
       "      <td>18</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>35.0</td>\n",
       "      <td>18</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35.0</td>\n",
       "      <td>18</td>\n",
       "      <td>8</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Age  Age_density  Deck  Embarked  FamilySize     Fare  Pclass  Sex  \\\n",
       "0  22.0           27     8         2         2.0   7.2500       3    0   \n",
       "1  38.0           11     2         0         2.0  71.2833       1    1   \n",
       "2  26.0           18     8         2         1.0   7.9250       3    1   \n",
       "3  35.0           18     2         2         2.0  53.1000       1    1   \n",
       "4  35.0           18     8         2         1.0   8.0500       3    0   \n",
       "\n",
       "   Survived  Ticket_density  Title  \n",
       "0         0               1      2  \n",
       "1         1               1      3  \n",
       "2         1               1      1  \n",
       "3         1               2      3  \n",
       "4         0               1      2  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.feature.drop(['Name'])\n",
    "sml.feature.labels(['Title', 'Embarked', 'Deck'])\n",
    "sml.train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:23.205251",
     "start_time": "2017-05-16T11:30:22.883315"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAksAAAETCAYAAAAmiEIFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkioYliRJkioYliRJkioYliRJkir8HyPQ+IbBOiYnAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x107f3e6d8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.importance()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:23.705079",
     "start_time": "2017-05-16T11:30:23.207097"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYQAAAFmCAYAAABtMe1aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nvPD83Pc8sgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1151ad3c8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.correlate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:26.187390",
     "start_time": "2017-05-16T11:30:23.707312"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1152e4860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAskAAAK+CAYAAABQEoOUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqcIiWZIkSaqwSJYkSZIqLJIlSZKkiv8PC9gZeQS3eicAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1151d57f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.plot.distribute()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:26.228375",
     "start_time": "2017-05-16T11:30:26.189741"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Results</th>\n",
       "      <th>Observations</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Speedml Release</th>\n",
       "      <td>v0.9.3</td>\n",
       "      <td>Visit https://speedml.com for release notes.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Shape</th>\n",
       "      <td>train (891, 11) | test (418, 10)</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numerical High-cardinality</th>\n",
       "      <td>[Age, Age_density, Fare]</td>\n",
       "      <td>(&gt;10) categories. Use feature.density</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numerical Categorical</th>\n",
       "      <td>[Deck, Embarked, FamilySize, Pclass, Sex, Surv...</td>\n",
       "      <td>Use plot.ordinal.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Target Analysis (Survived)</th>\n",
       "      <td>Model ready.</td>\n",
       "      <td>Use classification models.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                                      Results  \\\n",
       "Speedml Release                                                        v0.9.3   \n",
       "Shape                                        train (891, 11) | test (418, 10)   \n",
       "Numerical High-cardinality                           [Age, Age_density, Fare]   \n",
       "Numerical Categorical       [Deck, Embarked, FamilySize, Pclass, Sex, Surv...   \n",
       "Target Analysis (Survived)                                       Model ready.   \n",
       "\n",
       "                                                            Observations  \n",
       "Speedml Release             Visit https://speedml.com for release notes.  \n",
       "Shape                                                                     \n",
       "Numerical High-cardinality         (>10) categories. Use feature.density  \n",
       "Numerical Categorical                                  Use plot.ordinal.  \n",
       "Target Analysis (Survived)                    Use classification models.  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.eda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:26.237202",
     "start_time": "2017-05-16T11:30:26.230764"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'train_X: (891, 10) train_y: (891,) test_X: (418, 10)'"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.model.data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Hyper-parameters Tuning\n",
    "\n",
    "This is an important step in the workflow. We optimize the hyper-parameters used for our XGBoost classifier to get the best results. We do this iteratively by setting certain hyper-parameters as fixed while checking permutations for two select parameters at a time.\n",
    "\n",
    "The resulting table ranks the permutations based on accuracy, highest rank = 1.\n",
    "\n",
    "We now substitute the highest ranking parameters as fixed in the next round of optimization, selecting an alternate pair of parameters to optimize this time. Repeat this process for as many parameters as required. Normally two rounds of tuning is sufficient."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:28.017156",
     "start_time": "2017-05-16T11:30:26.239115"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rank</th>\n",
       "      <th>params</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>{'max_depth': 3, 'min_child_weight': 1}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>{'max_depth': 3, 'min_child_weight': 3}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3</td>\n",
       "      <td>{'max_depth': 7, 'min_child_weight': 3}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>{'max_depth': 5, 'min_child_weight': 1}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>5</td>\n",
       "      <td>{'max_depth': 7, 'min_child_weight': 5}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6</td>\n",
       "      <td>{'max_depth': 3, 'min_child_weight': 5}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>{'max_depth': 5, 'min_child_weight': 5}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8</td>\n",
       "      <td>{'max_depth': 5, 'min_child_weight': 3}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>9</td>\n",
       "      <td>{'max_depth': 7, 'min_child_weight': 1}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   rank                                   params\n",
       "0     1  {'max_depth': 3, 'min_child_weight': 1}\n",
       "1     1  {'max_depth': 3, 'min_child_weight': 3}\n",
       "7     3  {'max_depth': 7, 'min_child_weight': 3}\n",
       "3     4  {'max_depth': 5, 'min_child_weight': 1}\n",
       "8     5  {'max_depth': 7, 'min_child_weight': 5}\n",
       "2     6  {'max_depth': 3, 'min_child_weight': 5}\n",
       "5     6  {'max_depth': 5, 'min_child_weight': 5}\n",
       "4     8  {'max_depth': 5, 'min_child_weight': 3}\n",
       "6     9  {'max_depth': 7, 'min_child_weight': 1}"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "select_params = {'max_depth': [3,5,7], 'min_child_weight': [1,3,5]}\n",
    "fixed_params = {'learning_rate': 0.1, 'subsample': 0.8, \n",
    "                'colsample_bytree': 0.8, 'seed':0, \n",
    "                'objective': 'binary:logistic'}\n",
    "\n",
    "sml.xgb.hyper(select_params, fixed_params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:29.318960",
     "start_time": "2017-05-16T11:30:28.020855"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>rank</th>\n",
       "      <th>params</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>{'learning_rate': 0.1, 'subsample': 0.8}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>{'learning_rate': 0.3, 'subsample': 0.7}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>{'learning_rate': 0.1, 'subsample': 0.9}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>{'learning_rate': 0.3, 'subsample': 0.9}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>{'learning_rate': 0.1, 'subsample': 0.7}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>{'learning_rate': 0.3, 'subsample': 0.8}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>{'learning_rate': 0.01, 'subsample': 0.8}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>7</td>\n",
       "      <td>{'learning_rate': 0.01, 'subsample': 0.9}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>9</td>\n",
       "      <td>{'learning_rate': 0.01, 'subsample': 0.7}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   rank                                     params\n",
       "4     1   {'learning_rate': 0.1, 'subsample': 0.8}\n",
       "0     2   {'learning_rate': 0.3, 'subsample': 0.7}\n",
       "5     3   {'learning_rate': 0.1, 'subsample': 0.9}\n",
       "2     4   {'learning_rate': 0.3, 'subsample': 0.9}\n",
       "3     4   {'learning_rate': 0.1, 'subsample': 0.7}\n",
       "1     6   {'learning_rate': 0.3, 'subsample': 0.8}\n",
       "7     7  {'learning_rate': 0.01, 'subsample': 0.8}\n",
       "8     7  {'learning_rate': 0.01, 'subsample': 0.9}\n",
       "6     9  {'learning_rate': 0.01, 'subsample': 0.7}"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "select_params = {'learning_rate': [0.3, 0.1, 0.01], 'subsample': [0.7,0.8,0.9]}\n",
    "fixed_params = {'max_depth': 3, 'min_child_weight': 1, \n",
    "                'colsample_bytree': 0.8, 'seed':0, \n",
    "                'objective': 'binary:logistic'}\n",
    "\n",
    "sml.xgb.hyper(select_params, fixed_params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:29.510968",
     "start_time": "2017-05-16T11:30:29.322225"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tuned_params = {'learning_rate': 0.1, 'subsample': 0.8, \n",
    "                'max_depth': 3, 'min_child_weight': 1,\n",
    "                'seed':0, 'colsample_bytree': 0.8, \n",
    "                'objective': 'binary:logistic'}\n",
    "sml.xgb.cv(tuned_params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:29.526590",
     "start_time": "2017-05-16T11:30:29.513254"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>test-error-mean</th>\n",
       "      <th>test-error-std</th>\n",
       "      <th>train-error-mean</th>\n",
       "      <th>train-error-std</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>0.171910</td>\n",
       "      <td>0.002752</td>\n",
       "      <td>0.130899</td>\n",
       "      <td>0.006115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>0.169663</td>\n",
       "      <td>0.004204</td>\n",
       "      <td>0.129775</td>\n",
       "      <td>0.006063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>0.170786</td>\n",
       "      <td>0.007621</td>\n",
       "      <td>0.129494</td>\n",
       "      <td>0.006306</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>0.170786</td>\n",
       "      <td>0.005729</td>\n",
       "      <td>0.129494</td>\n",
       "      <td>0.007073</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>0.167415</td>\n",
       "      <td>0.008257</td>\n",
       "      <td>0.126685</td>\n",
       "      <td>0.005646</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    test-error-mean  test-error-std  train-error-mean  train-error-std\n",
       "42         0.171910        0.002752          0.130899         0.006115\n",
       "43         0.169663        0.004204          0.129775         0.006063\n",
       "44         0.170786        0.007621          0.129494         0.006306\n",
       "45         0.170786        0.005729          0.129494         0.007073\n",
       "46         0.167415        0.008257          0.126685         0.005646"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.xgb.cv_results.tail(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:29.533516",
     "start_time": "2017-05-16T11:30:29.528628"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "tuned_params['n_estimators'] = sml.xgb.cv_results.shape[0] - 1\n",
    "sml.xgb.params(tuned_params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Model Evaluation and Prediction\n",
    "\n",
    "We can now evaluate various models against our datasets to see which one is best suited."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:29.540352",
     "start_time": "2017-05-16T11:30:29.536023"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "sml.xgb.classifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:33.369301",
     "start_time": "2017-05-16T11:30:29.543152"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAekAAAETCAYAAAASx/7oAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115434748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.model.evaluate()\n",
    "sml.plot.model_ranks()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:33.384836",
     "start_time": "2017-05-16T11:30:33.371393"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Classifier</th>\n",
       "      <th>Accuracy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>XGBClassifier</td>\n",
       "      <td>0.832222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GradientBoostingClassifier</td>\n",
       "      <td>0.832222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AdaBoostClassifier</td>\n",
       "      <td>0.821111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>RandomForestClassifier</td>\n",
       "      <td>0.814444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DecisionTreeClassifier</td>\n",
       "      <td>0.796667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LogisticRegression</td>\n",
       "      <td>0.786667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GaussianNB</td>\n",
       "      <td>0.777778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SVC</td>\n",
       "      <td>0.714444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KNeighborsClassifier</td>\n",
       "      <td>0.703333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Classifier  Accuracy\n",
       "0               XGBClassifier  0.832222\n",
       "0  GradientBoostingClassifier  0.832222\n",
       "0          AdaBoostClassifier  0.821111\n",
       "0      RandomForestClassifier  0.814444\n",
       "0      DecisionTreeClassifier  0.796667\n",
       "0          LogisticRegression  0.786667\n",
       "0                  GaussianNB  0.777778\n",
       "0                         SVC  0.714444\n",
       "0        KNeighborsClassifier  0.703333"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.model.ranks()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "## Model Prediction and Feature Selection\n",
    "\n",
    "This part of our notebook trains the XGBoost model, generates the model predictions and plots features based on model assigned importance (weights)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:33.661203",
     "start_time": "2017-05-16T11:30:33.386494"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAksAAAETCAYAAAAmiEIFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3XmcXXV9//HXEJaEMlDEiUCroIgfFtlEWfyBGSib2CFS\nRaECRogSWUQNIpXYVgQL1ai4AUUCimgEXEKUtQlhkUVAIRLko1C1CMiirLIl5P7+OCd6SWdOlrl3\n7p17X8/HYx5z71k/5wOP5J3v2XpqtRqSJEka3CqtLkCSJKmdGZYkSZIqGJYkSZIqGJYkSZIqGJYk\nSZIqGJYkSZIqrNrqAiR1noioAXcCL9ZNvjUzJ6/k9t4EHJ6ZUxpR3xD7qAF9mflos/YxxH4nA6tn\n5tdGcr+Slp9hSVKz7NbA4LEl8PcN2la72YUiWEpqU4YlSSMqIjYHTgfWA8YAX8rMGRGxCvAFYCeg\nF+gBJgP/C5wErBMR5wLfAL6Sma8vt9e/5HtE/DuwM7ABMD8zD46IE4F3UFx28FvgyMx8oKK+jYG5\n5c/OwGrAccARwGbArcBBwKuAa4B5wDZlvUdn5nURsRrweeAfKEbXbgY+kplPRcRvy+9bA58A9gP2\njIhngYuBs4BXAOsDvwPelZkPl+udV27zVcB3M/P4subDgKnlvh4F3puZ90XEADANWB14BjguM2+s\n/A8k6f/wmiVJzXJ1RNxe9zM+IlalCAQnZOb2wATguIjYCdgR2BDYOTO3oAhFJ2TmfcC/Atdl5vuW\nY78bAW8og9KhwFbADpm5LXAp8PXl2MargUsyc0tgDkW4O4hihGtXikAHRWi5otz2CcB3y6A0rTyW\nbcqfVYDP1m3/zszcPDN/AFwCfCEzvwocCNyYmTsDr6EIOIfUrbdWZu4KvBk4JiJeHRHbAKcB+2Tm\n1uX2ToyITYHPAPtm5nbAB4DvR8TfLMfxS6rjyJKkZvk/p+EiYgtgE2BGRCyZPA7YLjPPiIhpwBER\nsQnQDzy1Evu9KTMXlZ//EdgBuLXc3xhgzeXYxkJgdvn5XuCGzHyyPIYHgJcBDwCPZea3ATLzsoh4\nkWLE6K3AiZm5sFzny8AP67Z/3WA7zczTI2LXiPgosCnweopRqCVmlcvdHxEPl3VMoAhs95Xzvlju\n80iKEbY5db1eDLwWuGM5eiCpZFiSNJLGAI+XIzEARMQrgCci4m0UIzjTKULB3cDBg2yjRnHKa4nV\nl5r/9FL7Oy0zzyj3tQaw7nLU+UJm1r84c+EQyy1a6vsqFKfClh61X4XidN5gNf5FRJxGEe5mAFeX\n69Qf67N1n5f0YVH5eck2xlGMro0B5mTmu+vmvZIi5ElaAZ6GkzSSEnguIg6Gv/zlfSewPbAnMLsM\nNrcAb6f4Cx+KQLAkbDwCvKo8rddTLjeUK4DJEbF2+f0k4PwGHk9fROxTHssARaj6RbnfKRGxWnkt\n1lHAVUNso/7Y9ga+mJnnAw9T9GTMEOstcTWwR0RsUH4/AvhPimuu9oqIzcr69gXmA2NX+CilLmdY\nkjRiMvMFYCJFgJkPXAl8MjN/ApwJTCin30hx+uvVZdi4EdgsIn6QmXdRXAR9K3AT8GDFLr8O/Ai4\nKSIWUJwim9TAQ3oOOCQi7gBOBN6emS8CJwN/AG4HfkkRho4dYhuXAR+KiH+hCHOfi4jbgO8D11Oc\nNhtSZv4C+BhweVnHPsCUzFxAcZ3SzHL6p4H9MvPPwzlgqRv11Gq1ZS8lSXqJ8q65OzNzrVbXIqm5\nHFmSJEmq4MiSJElSBUeWJEmSKhiWJEmSKvicpWFYtOjF2mOPPdPqMjrWuuuuif1tLnvcXPa3+exx\nc3Vbf/v6ensGm+7I0jCsuuqyHn+i4bC/zWePm8v+Np89bi77WzAsSZIkVTAsSZIkVTAsSZIkVTAs\nSZIkVTAsSZIkVTAsSZIkVTAsSZIkVTAsSZIkVfAJ3sMwMHVWq0uQJKlrzDhh95bs15ElSZKkCoYl\nSZKkCoYlSZKkCoYlSZKkCoYlSZKkCh13N1xEbAzMB35WN3luZp7UmookSdJo1nFhqXRXZva3ughJ\nkjT6dWpYeomIGAOcBbwS2AC4JDOnRcR5wHrlz9uA44FdgTHA5zPzotZULEmSltbX19uS/XZqWNoi\nIubVfT8RuCkzJ0fEWOD3wLRy3tzM/EJEvBV4dWbuUi5zU0RclZmPj2zpkiRpMI888lRTtz9UGOvU\nsPSS03ARsTZwaETsBjwJrFG3bJa/twK2rwtZqwEbA7c3u1hJktS+uuVuuEnA45n5HmA6sGZE9JTz\nFpe/7wauLkPW7sCFwL0jXKckSWoz3RKW5gD7RMS1wBnAr4ENl1pmNvB0RFwH3AbUMrO5432SJKnt\n9dRqtVbXMGoNTJ1l8yRJGiHNfpFuX19vz2DTu2VkSZIkaaUYliRJkip4Gm54as2+jbGb9fX1Nv02\n0W5nj5vL/jafPW6ubuuvp+EkSZJWgmFJkiSpgmFJkiSpQqc+wXtEDEyd1eoSJEkdptm3x2vFObIk\nSZJUwbAkSZJUwbAkSZJUwbAkSZJUwbAkSZJUoSvCUkQcHxEPRsTYVtciSZJGl64IS8DBwEzgwFYX\nIkmSRpeOf85SRPQD9wJnAt8CzouIHYCvAk8BDwPPZeakiDgG+GegBszMzC+1pmpJUrfq6+ttdQkv\n0W71tELHhyVgMvD1zMyIeD4idgTOAA7JzAURcQrwdxGxBfBuYJdyvasi4orMzBbVLUnqQu304tou\nfJHuoNM7+jRcRKwL7AscGxGXA+sARwMbZuaCcrHryt+vBzYC5pQ/6wGbjmzFkiSp3XR0WKK4Vumc\nzNwrM/cBdgT2Ap4tR5IAdip/J7AA2C0z+4HzgPkjW64kSWo3nR6WJgPnL/mSmc8A36MIQjMi4r+B\nHYCFmXkHxYjS9RFxK8Wo0v0jXrEkSWorHX3NUmZuM8i0IyPiKGAgMx+JiJOBF8p5nwU+O8JlSpKk\nNtbRYanCQ8CVEfE08ATw3hbXI0mS2lRXhqXMvBi4uNV1SJKk9tfp1yxJkiQNS0+tVmt1DaNZrZue\nPzHSuu35Hq1gj5vL/jafPW6ubutvX19vz2DTHVmSJEmqYFiSJEmqYFiSJEmq0JV3wzXKwNRZrS5B\nkhpqxgm7t7oEqe04siRJklTBsCRJklTBsCRJklTBsCRJklShoWEpIo6PiAcjYmwjt1u3/ZkR0d+A\n7awfEV8rP78lIrYednGSJKkjNXpk6WBgJnBgg7fbUJn5h8w8svx6GLBhK+uRJEntq2GPDihHfO4F\nzgS+BZwXETsAXwWeAh4GnsvMSRFxDPDPQA2YmZlfqtjuUcBk4EFgfDlttXI/m1IEvmmZOS8i5gPX\nAFuX254IrA58t1xuLDAFeJwi1B0F7AO8ISLGAwOZeUC5j58AB2TmAw1pkCRJGpUa+ZylycDXMzMj\n4vmI2BE4AzgkMxdExCnA30XEFsC7gV3K9a6KiCsyM5feYES8AjgW2ApYDNxWt69HM/PwiFgPuBbY\nElgb+E5mHhMRFwBvpQhqfwQOBbYA/oYiLJGZt0XE5RTB6QrgExGxLsVI06MGJUndpq+vt9UlrLDR\nWPNoYn8bFJbKgLEvML4cNVoHOBrYMDMXlItdR3F67vXARsCccvq6FCNE/ycsAZsACzLz+XI/Py2n\nbwXsWgYygFUj4uXl55+Xv++jGEm6sNz+LGAhcPJgx5CZtYj4FnAQ8BrgnOVugCR1iNH20tRue9Hr\nSOu2/g4VDBt1zdLBwDmZuVdm7gPsCOwFPFuOJAHsVP5OYAGwW2b2A+cB84fY7q+BLSNiXESMAbYr\np99NMYLUTzF6dBHwp3Jebalt9AMPZuZeFEHpM0vNX8xf+3AucADwFuDSZR61JEnqeI0KS5OB85d8\nycxngO9RBKEZEfHfwA7Awsy8g2JU6fqIuJVi1Of+wTaamY8ApwI3AJcBfy5nnQVsFhHXlPN+l5mL\nh6jtDmByRMwDPgv8x1LzbwZOjYjNM/N+itN2czJz0fIfviRJ6lQ9tdrSAzGNU16cfWFmPhIRJwMv\nZOZJTdthA0TEj4APZ+Y9y1p2YOqs5jVPklpgtL0brttOE420butvX19vz2DTm/0i3YeAKyPiaeAJ\n4L2DLRQR+wEfHWTW6Zn5gybWV1/DOOB6YO7yBCVJktQdmhqWMvNi4OLlWO4S4JJm1rIcNTwLbN/K\nGiRJUvvxdSeSJEkVmn0arqPNnj6xq87ljrRuO1feCva4ueyv1BkcWZIkSapgWJIkSapgWJIkSarg\nNUvDMDB1VqtLkNQFRtuzj6RO48iSJElSBcOSJElSBcOSJElSBcOSJElShaZe4B0RGwPzgZ/VTZ67\nMi/TjYhtgf0y86SI+ENmrj/Ecn3AmUAvsBZwF3AMsA7wr5l55IruW5Ikda+RuBvurszsH+5GMvN2\n4PblWPRjwFWZeSZARHwRmJKZXwAMSpIkaYWM+KMDImIMcBbwSmAD4JLMnBYR5wELgY2ANYCZwADw\nKmBiufyUzDyw3M46FCNWr8vMFyPiNOA24CHgnRFxD/AT4DigVo5yzQR2Ay4ry1kV2Al4Xbn9U4AX\ngXuBIzJzYfM6IUmSRoORCEtbRMS8uu8nAjdl5uSIGAv8HphWzvttZr4/Is4EXp2Z+0bEpyhC00tG\nlTLziYi4Htg7Iq4A3gp8ElgEPEYxwnQRcD11I0qZ+SzQHxE9wPnAN4DfAJcDu2TmwxHxaWAScHYD\n+yBJK6Wvr7fVJbQ1+9Nc9rcFp+EiYm3g0IjYDXiSYhRpiSXXNj0O3F1+fgwYO8S2zwY+RHGh+n9n\n5gsRsQfwzcycERFrAMcDXwSmLrXul4HMzLMjYjzFKNeFEQEwDrhqZQ5WkhrNl/EOzZcVN1e39Xeo\nYNiKu+EmAY9n5nuA6cCa5SgPQG1FNpSZ1wObAIcD55STPwT8czn/eWAB8Hz9euXIUU9mfrqc9CjF\nCNfEMtidAsxdoaOSJEkdqRWvO5kDfDsidqYIMb8GNhzG9i4ADsjMBeX3KcDXIuIjwLPAI8AHgdUB\nIuJNwL8A8+pOD54EHAv8OCJWoRjxOnQYNUmSpA7RU6ut0GBO24mIjwF/zMwZI73vgamzRnfzJI0K\nvhtuaN12mmikdVt/+/p6ewabPqpfpFveQbchxQXgkiRJDTeqw1JmTmp1DZIkqbP5uhNJkqQKo3pk\nqdVmT5/YVedyR1q3nStvBXvcXPZX6gyOLEmSJFUwLEmSJFUwLEmSJFXwmqVhGJg6q9UlSFoJPrdI\n0opwZEmSJKmCYUmSJKmCYUmSJKmCYUmSJKlCx1/gHREnAHsAqwGLgeMy87bWViVJkkaLjh5Ziogt\ngP2APTNzAvARYEZrq5IkSaNJp48sPQG8CjgsIi7PzNsjYoeI2Ar4EtAD/BE4DNgV+DgwAfg3YFxm\nHt+iuiVJUpvoqdVqra6hqSLiDcDRFKfingFOBD4GHJaZd0XE4cBrMvPEiDgdeBnwSmCPzFxUte2B\nqbM6u3lSh5o9fWKrS5DUnnoGm9jRI0sR8Vrgycw8rPz+RuAyYCzwtYiA4lqmX5er/CfwO+BdywpK\nkkavkXq5rS/SbT573Fzd1t++vt5Bp3f0NUvA1sBXImL18vuvgMeBe4BDM7MfOB74UTn/TOBY4FMR\nse4I1ypJktpQR48sZeb3I2Jz4JaIeJoiHH4MuA/4ZkSsCtSAwyPiWOChzPxqRPwZ+DrwjlbVLkmS\n2kNHhyWAzDwFOGWQWf1Lff9V3TrnAec1rShJkjRqdPppOEmSpGExLEmSJFUwLEmSJFXo+GuWmmn2\n9IlddUvlSOu2W1ZbwR5L0rI5siRJklTBsCRJklTBsCRJklTBa5aGYWDqrFaXIGkpM07YvdUlSOow\njixJkiRVMCxJkiRVMCxJkiRVMCxJkiRVMCxJkiRVGFV3w0VEP3AhcBdQA8YBF2TmlwdZdh4wJTPv\nHskaJUlSZxmNI0tzM7M/M3cDJgBTI+JvW12UJEnqTKNqZGkQvcCLwDYRcSpF+LsfeM+SBSLi74Ez\ngLHABsC0zPxhRJwC7EbRg+9l5mkRcSTwXmAxcEtmfmhEj0bSsPX19ba6hJdot3o6kT1uLvs7OsPS\n7uUptsXAQuAY4HTgoMz8ZUQcDmxet/xmwPTMnBcRbwY+BfyQIlD1Aw8Ck8pl3wccmZm3RMQHI2LV\nzFw0AsckqUHa6cXAvqi4+exxc3Vbf4cKhqMxLM3NzAPrJ0TEjMz8JUBmnlNOWzL7QWBaGaJqwGrl\n9PcApwLrA5eV094HHBcRrwZuBHqaeBySJGkUGI3XLA3mgYjYFCAiPh4R+9fN+zTwzcw8BLga6ImI\nNYADgIMoTsVNioiNgPdTXBQ+AdgOePNIHoQkSWo/nRKWjgBmRMQ1FCHn0rp5FwGfi4hrgT2Bl2fm\n88CfgJsoAtSVwP8CvwCui4i5wMPAzSN3CJIkqR311Gq1Vtcwag1MnWXzpDbTTi/S7bbrPVrBHjdX\nt/W3r6930MtvOmVkSZIkqSkMS5IkSRVG491wbWP29IldNTw50rpt+LcV7LEkLZsjS5IkSRUMS5Ik\nSRUMS5IkSRW8ZmkYBqbOanUJUldpp8cCSOoejixJkiRVMCxJkiRVMCxJkiRVMCxJkiRVMCxJkiRV\n6Mi74SJiOrA9sD6wJvA/wELgusw8KSL2B24GVgdmZuZOLStWkiS1tY4MS5k5FSAiJgGbZeYJSy1y\nLDAFeG6ES5MkSaNMR4alwUREP0VAOh/YFvgmcHDd/AnAKcCLwL3AEZm5cOQrlTSUvr7eVpewwkZj\nzaONPW4u+9tFYWmJzPxxRNxOEZxeAIiIHuBsYJfMfDgiPg1MKqdJahOj7aW/vqi4+exxc3Vbf4cK\nhl0XlobQB2wAXBgRAOOAq1pakSRJagvdGpYW89I7AR8Ffg9MzMwnImI/4OmWVCZJktpKtz464AaK\na5ZeBpCZiyku+v5xRNwAHAnc2bryJElSu+ip1WqtrmHUGpg6y+ZJI2i0vUi32673aAV73Fzd1t++\nvt6ewaZ368iSJEnScjEsSZIkVejWC7wbYvb0iV01PDnSum34txXssSQtmyNLkiRJFQxLkiRJFQxL\nkiRJFbxmaRgGps5qdQlS04222/UlqdEcWZIkSapgWJIkSapgWJIkSapgWJIkSapgWJIkSapQeTdc\nREwHtgfWB9YE/gdYCFyXmScNsc4fMnP9Ze04Il4G7JOZ316eQiNiJnBmZs5bnuUrtrM+8K+ZeWRE\nvAV4PDPnD2ebkiSpc1WGpcycChARk4DNMvOEBu57a2A/YLnCUqNk5h+AI8uvhwEzAcOSJEka1Ao/\nZyki+oEpmXlgRBwOfBAYA1ySmf9Wt9xngHWAo4F3Ah8FXgSuL0PXicA2EfGBzPyvIfZ1FDAZeBAY\nX05bDTgT2JTiNOK0zJwXEfOBayhCWA2YCKwOfLdcbiwwBXicIiAdBewDvCEixgMDmXlAuY+fAAdk\n5gMr2h+p0/T19ba6hFHN/jWfPW4u+zuMh1KWAeMEinDyHPAfEbFWOe9zwOLMPKo83fYp4I2Z+UxE\nnB8RewKnUISuoYLSK4Bjga2AxcBt5azJwKOZeXhErAdcC2wJrA18JzOPiYgLgLcCTwF/BA4FtgD+\nhiIskZm3RcTlFMHpCuATEbEusGG5fYOSBL5odxh8UXHz2ePm6rb+DhUMh/ME79cAd2bms+X3E+Av\nIWdr4J5y+muBPuDSiADoBTYB7l7G9jcBFmTm8+V2f1pO3wrYNSJ2XHIMEfHy8vPPy9/3UYwkXUgx\nAjWL4lqrkwfbUWbWIuJbwEHlcZ2zjNokSVKXGM7dcPcCm0XEGgARcXFE/B3wELA3sGVE7AP8hiK8\n7JmZ/cCXgZsoRouq9v/rchvjImIMsF05/W6KEaR+itGji4A/lfNqS22jH3gwM/eiCEqfWWp+fQ3n\nAgcAbwEuXY7jlyRJXWClw1JmPgKcBlwTETcCP8vM+8t5NeBw4CsUgeTz5XI3UwScX1GEra0i4sMV\n2z8VuAG4DPhzOessipB2TTnvd5m5eIgy7wAmR8Q84LPAfyw1/2bg1IjYvKz9KWBOZi5aoWZIkqSO\n1VOrLT0Y070i4kfAhzPznmUuDAxMnWXz1PF8ke7K67brPVrBHjdXt/W3r6+3Z7Dpw7lmqSEiYj+K\nO+WWdnpm/mCEahgHXA/MXd6gJEmSukPLw1JmXgJc0uIanqV4+KYkSdJL+LoTSZKkCi0fWRrNZk+f\n2FXnckdat50rbwV7LEnL5siSJElSBcOSJElSBcOSJElSBa9ZGoaBqbNaXYI0LLOnT2x1CZLU9hxZ\nkiRJqmBYkiRJqmBYkiRJqmBYkiRJqtAxF3hHRD9wIXAX0AOsBnwxMy9cgW3MA6Zk5t3NqFGSJI0+\nnTayNDcz+zNzArAX8PGI2LbVRUmSpNGrY0aWlpaZT0fEWcA7I+LdwK7AGODzmXlRROwIfJEiMN4P\nvGfJuhExAHwU2D8zHx/56iVJUrvo2LBUegiYCvw8M3eJiLHATRFxFXAWcFBm/jIiDgc2L9f5J2AC\n8I+Z+eeWVC2NoL6+3laX0NHsb/PZ4+ayv50fljYCLgAOKa9HguJapo2B9TPzlwCZeQ5ARAD8A7A2\nsHCEa5VawhfpNo8vKm4+e9xc3dbfoYJhp12z9BcRsTbwfuAJ4OrM7Ad2p7gI/F7ggYjYtFz24xGx\nf7nqUcAVwEkjXrQkSWo7nRaWdo+IeRExB5gN/BvwJeDpiLgOuA2oZeZTwBHAjIi4BtgOuLRuOycB\n+0TELiNbviRJajcdcxouM+cB44eY/dFBlr+F4qLvev11n72LTpIkddzIkiRJUkMZliRJkioYliRJ\nkip0zDVLrTB7+sSuuqVypHXbLauSpPbkyJIkSVIFw5IkSVIFw5IkSVIFr1kahoGps1pdQkebPX1i\nq0uQJMmRJUmSpCqGJUmSpAqGJUmSpAqGJUmSpAojeoF3RPQDFwJ31U1+JDMPWMZ6k4DNMvOEldjn\nb8t1n1uBdcYCd2fmxiu6P0mS1FlacTfc3Mw8sAX7lSRJWmFt8eiAiJgH3AG8HngauA7YG/hbYK9y\nsZ0jYg6wNvDvmfnjiHgncBSwGlAD9i+3cRrwAvBfdfuYUm7rIGAn4BTgReBe4AhgDeACYF3gnuYd\nrSRJGk1aEZZ2L8PREj8uf/80M4+NiMuBZzJzz4j4BjChnP9n4G1AH3BzRFwGvA54W2Y+ExFnUQSs\n+4GxmbkjQER8GjgG2BY4AFgMnA3skpkPl/MnAesAd2bmiRGxI7B7k45fK6Cvr7fVJXQ8e9xc9rf5\n7HFz2d82OQ0XEW8DflZ+fZy/XtP0GDC2/Hx9ZtaAhyPiCWA94GHgGxHxNLAZcGO5bC61zz2ARZn5\nYkSMBzYALowIgHHAVcB4yuCWmTdHxMJGHKyGxxfpNpcvK24u+9t89ri5uq2/QwXDdrobrraM+W8C\niIj1gbUoTrN9CjgQmAw8C/SUyy5eat2JwGPlqbhHgd8DEzOzn+J03FyKgLZzuY/tKE7tSZKkLtcO\np+GgGN1ZlnERMZciKB0BPAn8hGI0aRHFKNSGwG+GWP9DwE+BOcCxwI8jYpVyO4cCNwDfjIjrgbuB\n51fgmCRJUofqqdWWNaCjoQxMnWXzmmj29IldNfzbCt02xD7S7G/z2ePm6rb+9vX19gw2vZ1Ow0mS\nJLUdw5IkSVIFw5IkSVKFtngo5WjlNTWSJHU+R5YkSZIqGJYkSZIqGJYkSZIqGJYkSZIqGJYkSZIq\nGJYkSZIqGJYkSZIqGJYkSZIqGJYkSZIq9NRqtVbXIEmS1LYcWZIkSapgWJIkSapgWJIkSapgWJIk\nSapgWJIkSapgWJIkSapgWJIkSaqwaqsLGG0iYhXga8A2wPPA5My8p7VVdY6I2BE4LTP7I+K1wHlA\nDbgTOCozF7eyvtEqIlYDZgAbA2sAJwN3YX8bJiLGAGcDQdHTKcBz2OOGiojxwG3AnsAi7G9DRcTP\ngCfLr78BTsEeO7K0Et4OjM3MnYETgOktrqdjRMTxwNeBseWkzwPTMnNXoAeY2KraOsDBwB/LXu4D\nfAX722gDAJn5/4BpFH/J2OMGKkP/WcCz5ST720ARMRboycz+8ud92GPAsLQydgEuB8jMm4A3trac\njnIv8E9137cHrik/XwbsMeIVdY6LgE+Wn3so/kVufxsoM38IfKD8uhHwOPa40T4HnAk8UH63v421\nDbBmRFwZEXMjYifsMWBYWhlrA0/UfX8xIjyd2QCZ+T1gYd2knsxc8j6ep4B1Rr6qzpCZT2fmUxHR\nC1xMMfJhfxssMxdFxDeALwMXYI8bJiImAY9k5hV1k+1vYz1DEUj3pjiN7P/DJcPSinsS6K37vkpm\nLmpVMR2u/rx4L8W/1LWSIuKVwNXA+Zn5bexvU2Tme4HXUVy/NK5ulj0ensOAPSNiHrAt8E1gfN18\n+zt8vwK+lZm1zPwV8EfgFXXzu7bHhqUV9xNgX4ByiPIXrS2no/08IvrLz28FrmthLaNaRLwCuBL4\neGbOKCfb3waKiEMi4l/Kr89QhNFb7XFjZOZbMnNCZvYDtwOHApfZ34Y6jPI63IjYkOJMypX22Lvh\nVsYPKP51cwPFtR/va3E9nWwqcHZErA78kuL0kVbOJ4B1gU9GxJJrl44FvmR/G+b7wLkRcS2wGvBh\nir76/3Dz+GdEY50DnBcR11Pc/XYY8Cj2mJ5arbbspSRJkrqUp+EkSZIqGJYkSZIqGJYkSZIqGJYk\nSZIqGJYkSZIqGJYkjUoRMaK38kbEuRGx0UjuU1J7MCxJ0vLZjeLZapK6jA+llDSqlU8XPpEiyGxC\n8dC8J4C3l9P2zcyHIuIR4EcULwZ9CnhPZv62fBL/6cBYigfwHZGZ95Sv1fgTsCVwLrAhcGlE7Ars\nTvFAxHHlz+TMvLZc56fArkAfcExmXlaOSJ1L8XqOZ8rl50fEoRQPr1wFuA04KjOfa1qzJK0UR5Yk\ndYIdKZ6mvyXwQYoXrr4RmA8cWC7zcmBeZm4NzOSvTy+fCRydmdtQvNH+O3XbnZ+ZkZmnUrzpfl/g\nMYqXjP7RuFwhAAABwElEQVRjuc6pwMfq1lk9M3cGPgKcXE77GvC9zHw98O/AtIjYEng/8ObM3BZ4\nGDiuUQ2R1DiGJUmd4M7MvC8zn6EYHZpTTv8dxWteAJ6jePkqwDcoRodeBzyWmbcAZOZFwGsjYsmb\n1W9eekeZuRjYH9g7Ik4CJgFr1S1y+ZKagJeVnycA55frX5qZ76I4rbcpcFNE3A5MBDZbqaOX1FSe\nhpPUCV5Y6vuiQZZZnJlLLgpfpVxmsH8w9gBjys/PLj0zItYCbqEIP9dSjF4dXbfIktNoNf56jdPC\nuvV7gM3LfVyYmR+q265/JkttyJElSd1izYgYKD+/D7gMSGC9iHgTQES8C/hdZv5pkPUXUYSZ1wGL\ngc8AcynexD5mkOXrXctfTwfuAfwXMA/YPyLGlwHqDIrrlyS1GcOSpG5yQETMB/YGPpyZzwPvBr4S\nEXdSjBC9e4h1fwRcSnHx+O3A3cDPgKeBZT1S4GjgHeXptk8BH8jMO8rPc4EFFH8enzqMY5PUJD21\n2og+qkSSWiIiapnprf+SVpgjS5IkSRUcWZIkSargyJIkSVIFw5IkSVIFw5IkSVIFw5IkSVIFw5Ik\nSVKF/w/N0Wc/mE7NvgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115aced68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sml.xgb.fit()\n",
    "sml.xgb.predict()\n",
    "\n",
    "sml.plot.xgb_importance()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:34.033419",
     "start_time": "2017-05-16T11:30:33.663479"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 79.480000%\n",
      "Thresh=0.032, n=10, Accuracy: 79.48%\n",
      "Thresh=0.044, n=9, Accuracy: 79.10%\n",
      "Thresh=0.066, n=8, Accuracy: 77.61%\n",
      "Thresh=0.069, n=7, Accuracy: 79.10%\n",
      "Thresh=0.071, n=6, Accuracy: 76.49%\n",
      "Thresh=0.076, n=5, Accuracy: 76.49%\n",
      "Thresh=0.083, n=4, Accuracy: 69.78%\n",
      "Thresh=0.099, n=3, Accuracy: 68.66%\n",
      "Thresh=0.169, n=2, Accuracy: 68.66%\n",
      "Thresh=0.290, n=1, Accuracy: 71.64%\n"
     ]
    }
   ],
   "source": [
    "sml.xgb.feature_selection()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:34.044125",
     "start_time": "2017-05-16T11:30:34.035379"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Accuracy = 86.76%. Found 773 correct of 891 total labels'"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.xgb.sample_accuracy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-05-16T11:30:34.062541",
     "start_time": "2017-05-16T11:30:34.047236"
    },
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Results saved.'"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "'e16.74-m83.22-s86.76-f79.48'"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sml.save_results(\n",
    "    columns={ 'PassengerId': sml.uid,\n",
    "             'Survived': sml.xgb.predictions }, \n",
    "    file_path='output/titanic-speedml-{}.csv'.format(sml.slug()))\n",
    "sml.slug()"
   ]
  }
 ],
 "metadata": {
  "hide_input": false,
  "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": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}