natality_dataset_exploration_sampling.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "<h1> Natality: Predict weight of a newborn</h1>\n",
    "<h2> Creating a sampled dataset </h2>\n",
    "\n",
    "This notebook illustrates:\n",
    "<ol>\n",
    "<li> Sampling a BigQuery dataset to create datasets for ML\n",
    "<li> Preprocessing with Pandas\n",
    "</ol>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# change these to try this notebook out\n",
    "BUCKET = 'GCP-bucket-name'\n",
    "PROJECT = 'GCP-project-Name'\n",
    "REGION = 'zone' #see gcloud compute zones list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ['BUCKET'] = BUCKET\n",
    "os.environ['PROJECT'] = PROJECT\n",
    "os.environ['REGION'] = REGION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "%%bash\n",
    "if ! gsutil ls | grep -q gs://${BUCKET}/; then\n",
    "  gsutil mb -l ${REGION} gs://${BUCKET}\n",
    "fi"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "<h2> Explore full Natality dataset using BigQuery </h2>\n",
    "<p>\n",
    "Let's evaluate the dataset before sample for ML\n",
    "</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from scipy.stats import zscore\n",
    "\n",
    "pd.set_option(\"display.max_rows\",101)\n",
    "pd.set_option(\"display.max_columns\",101)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "query = \"\"\"\n",
    "            SELECT\n",
    "              *\n",
    "            FROM\n",
    "              publicdata.samples.natality\n",
    "            WHERE\n",
    "              year > 2000\n",
    "\"\"\"\n",
    "\n",
    "import google.datalab.bigquery as bq\n",
    "natality = bq.Query(\"SELECT * FROM (\" + query + \") WHERE RAND() < 0.0001\").execute().result().to_dataframe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>source_year</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>wday</th>\n",
       "      <th>state</th>\n",
       "      <th>is_male</th>\n",
       "      <th>child_race</th>\n",
       "      <th>weight_pounds</th>\n",
       "      <th>plurality</th>\n",
       "      <th>apgar_1min</th>\n",
       "      <th>apgar_5min</th>\n",
       "      <th>mother_residence_state</th>\n",
       "      <th>mother_race</th>\n",
       "      <th>mother_age</th>\n",
       "      <th>gestation_weeks</th>\n",
       "      <th>lmp</th>\n",
       "      <th>mother_married</th>\n",
       "      <th>mother_birth_state</th>\n",
       "      <th>cigarette_use</th>\n",
       "      <th>cigarettes_per_day</th>\n",
       "      <th>alcohol_use</th>\n",
       "      <th>drinks_per_week</th>\n",
       "      <th>weight_gain_pounds</th>\n",
       "      <th>born_alive_alive</th>\n",
       "      <th>born_alive_dead</th>\n",
       "      <th>born_dead</th>\n",
       "      <th>ever_born</th>\n",
       "      <th>father_race</th>\n",
       "      <th>father_age</th>\n",
       "      <th>record_weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3281.000000</td>\n",
       "      <td>3281.000000</td>\n",
       "      <td>3281.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3281.000000</td>\n",
       "      <td>1605</td>\n",
       "      <td>3281</td>\n",
       "      <td>821.0</td>\n",
       "      <td>3279.000000</td>\n",
       "      <td>3281.000000</td>\n",
       "      <td>821.0</td>\n",
       "      <td>2983.000000</td>\n",
       "      <td>1605</td>\n",
       "      <td>2272.000000</td>\n",
       "      <td>3281.000000</td>\n",
       "      <td>3264.000000</td>\n",
       "      <td>3281</td>\n",
       "      <td>3281</td>\n",
       "      <td>1599</td>\n",
       "      <td>1329</td>\n",
       "      <td>134.000000</td>\n",
       "      <td>2040</td>\n",
       "      <td>115.000000</td>\n",
       "      <td>3227.000000</td>\n",
       "      <td>2017.000000</td>\n",
       "      <td>2014.000000</td>\n",
       "      <td>2016.000000</td>\n",
       "      <td>3269.000000</td>\n",
       "      <td>2272.000000</td>\n",
       "      <td>3281.000000</td>\n",
       "      <td>3281.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>50</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>50</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1934</td>\n",
       "      <td>2</td>\n",
       "      <td>58</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>CA</td>\n",
       "      <td>True</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>CA</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>99999999</td>\n",
       "      <td>True</td>\n",
       "      <td>CA</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>205</td>\n",
       "      <td>1679</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>204</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>158</td>\n",
       "      <td>2048</td>\n",
       "      <td>127</td>\n",
       "      <td>1195</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1901</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2004.520878</td>\n",
       "      <td>2004.520878</td>\n",
       "      <td>6.517220</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.999086</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.0</td>\n",
       "      <td>7.234834</td>\n",
       "      <td>1.032917</td>\n",
       "      <td>99.0</td>\n",
       "      <td>14.724438</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.941461</td>\n",
       "      <td>27.268820</td>\n",
       "      <td>38.630208</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>18.537313</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.686957</td>\n",
       "      <td>42.543849</td>\n",
       "      <td>1.070402</td>\n",
       "      <td>0.020854</td>\n",
       "      <td>0.389385</td>\n",
       "      <td>2.076476</td>\n",
       "      <td>14.102993</td>\n",
       "      <td>40.757086</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.292008</td>\n",
       "      <td>2.292008</td>\n",
       "      <td>3.436146</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.841344</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.311218</td>\n",
       "      <td>0.190030</td>\n",
       "      <td>0.0</td>\n",
       "      <td>22.244326</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10.422582</td>\n",
       "      <td>6.102801</td>\n",
       "      <td>2.581830</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>28.426837</td>\n",
       "      <td>NaN</td>\n",
       "      <td>15.844843</td>\n",
       "      <td>27.863945</td>\n",
       "      <td>1.227173</td>\n",
       "      <td>0.177083</td>\n",
       "      <td>0.820740</td>\n",
       "      <td>1.238272</td>\n",
       "      <td>31.743370</td>\n",
       "      <td>25.314927</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2001.000000</td>\n",
       "      <td>2001.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.562179</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>99.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2002.000000</td>\n",
       "      <td>2002.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.0</td>\n",
       "      <td>6.563162</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>99.0</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2005.000000</td>\n",
       "      <td>2005.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.0</td>\n",
       "      <td>7.312733</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>99.0</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>39.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.500000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>34.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2007.000000</td>\n",
       "      <td>2007.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.0</td>\n",
       "      <td>8.000575</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>99.0</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>49.500000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>39.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2008.000000</td>\n",
       "      <td>2008.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.0</td>\n",
       "      <td>13.933215</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>99.0</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>78.000000</td>\n",
       "      <td>46.000000</td>\n",
       "      <td>47.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        source_year         year        month  day         wday state is_male  \\\n",
       "count   3281.000000  3281.000000  3281.000000  0.0  3281.000000  1605    3281   \n",
       "unique          NaN          NaN          NaN  0.0          NaN    50       2   \n",
       "top             NaN          NaN          NaN  NaN          NaN    CA    True   \n",
       "freq            NaN          NaN          NaN  NaN          NaN   205    1679   \n",
       "mean    2004.520878  2004.520878     6.517220  NaN     3.999086   NaN     NaN   \n",
       "std        2.292008     2.292008     3.436146  NaN     1.841344   NaN     NaN   \n",
       "min     2001.000000  2001.000000     1.000000  NaN     1.000000   NaN     NaN   \n",
       "25%     2002.000000  2002.000000     4.000000  NaN     2.000000   NaN     NaN   \n",
       "50%     2005.000000  2005.000000     7.000000  NaN     4.000000   NaN     NaN   \n",
       "75%     2007.000000  2007.000000     9.000000  NaN     6.000000   NaN     NaN   \n",
       "max     2008.000000  2008.000000    12.000000  NaN     7.000000   NaN     NaN   \n",
       "\n",
       "        child_race  weight_pounds    plurality  apgar_1min   apgar_5min  \\\n",
       "count        821.0    3279.000000  3281.000000       821.0  2983.000000   \n",
       "unique         NaN            NaN          NaN         NaN          NaN   \n",
       "top            NaN            NaN          NaN         NaN          NaN   \n",
       "freq           NaN            NaN          NaN         NaN          NaN   \n",
       "mean           9.0       7.234834     1.032917        99.0    14.724438   \n",
       "std            0.0       1.311218     0.190030         0.0    22.244326   \n",
       "min            9.0       0.562179     1.000000        99.0     0.000000   \n",
       "25%            9.0       6.563162     1.000000        99.0     9.000000   \n",
       "50%            9.0       7.312733     1.000000        99.0     9.000000   \n",
       "75%            9.0       8.000575     1.000000        99.0     9.000000   \n",
       "max            9.0      13.933215     3.000000        99.0    99.000000   \n",
       "\n",
       "       mother_residence_state  mother_race   mother_age  gestation_weeks  \\\n",
       "count                    1605  2272.000000  3281.000000      3264.000000   \n",
       "unique                     50          NaN          NaN              NaN   \n",
       "top                        CA          NaN          NaN              NaN   \n",
       "freq                      204          NaN          NaN              NaN   \n",
       "mean                      NaN     2.941461    27.268820        38.630208   \n",
       "std                       NaN    10.422582     6.102801         2.581830   \n",
       "min                       NaN     1.000000    14.000000        18.000000   \n",
       "25%                       NaN     1.000000    22.000000        38.000000   \n",
       "50%                       NaN     1.000000    27.000000        39.000000   \n",
       "75%                       NaN     1.000000    32.000000        40.000000   \n",
       "max                       NaN    78.000000    46.000000        47.000000   \n",
       "\n",
       "             lmp mother_married mother_birth_state cigarette_use  \\\n",
       "count       3281           3281               1599          1329   \n",
       "unique      1934              2                 58             2   \n",
       "top     99999999           True                 CA         False   \n",
       "freq         158           2048                127          1195   \n",
       "mean         NaN            NaN                NaN           NaN   \n",
       "std          NaN            NaN                NaN           NaN   \n",
       "min          NaN            NaN                NaN           NaN   \n",
       "25%          NaN            NaN                NaN           NaN   \n",
       "50%          NaN            NaN                NaN           NaN   \n",
       "75%          NaN            NaN                NaN           NaN   \n",
       "max          NaN            NaN                NaN           NaN   \n",
       "\n",
       "        cigarettes_per_day alcohol_use  drinks_per_week  weight_gain_pounds  \\\n",
       "count           134.000000        2040       115.000000         3227.000000   \n",
       "unique                 NaN           2              NaN                 NaN   \n",
       "top                    NaN       False              NaN                 NaN   \n",
       "freq                   NaN        1901              NaN                 NaN   \n",
       "mean             18.537313         NaN         2.686957           42.543849   \n",
       "std              28.426837         NaN        15.844843           27.863945   \n",
       "min               1.000000         NaN         0.000000            1.000000   \n",
       "25%               5.000000         NaN         0.000000           25.000000   \n",
       "50%               8.500000         NaN         0.000000           34.000000   \n",
       "75%              20.000000         NaN         0.000000           49.500000   \n",
       "max              99.000000         NaN        99.000000           99.000000   \n",
       "\n",
       "        born_alive_alive  born_alive_dead    born_dead    ever_born  \\\n",
       "count        2017.000000      2014.000000  2016.000000  3269.000000   \n",
       "unique               NaN              NaN          NaN          NaN   \n",
       "top                  NaN              NaN          NaN          NaN   \n",
       "freq                 NaN              NaN          NaN          NaN   \n",
       "mean            1.070402         0.020854     0.389385     2.076476   \n",
       "std             1.227173         0.177083     0.820740     1.238272   \n",
       "min             0.000000         0.000000     0.000000     1.000000   \n",
       "25%             0.000000         0.000000     0.000000     1.000000   \n",
       "50%             1.000000         0.000000     0.000000     2.000000   \n",
       "75%             2.000000         0.000000     0.000000     3.000000   \n",
       "max             9.000000         3.000000     8.000000    10.000000   \n",
       "\n",
       "        father_race   father_age  record_weight  \n",
       "count   2272.000000  3281.000000         3281.0  \n",
       "unique          NaN          NaN            NaN  \n",
       "top             NaN          NaN            NaN  \n",
       "freq            NaN          NaN            NaN  \n",
       "mean      14.102993    40.757086            1.0  \n",
       "std       31.743370    25.314927            0.0  \n",
       "min        1.000000    15.000000            1.0  \n",
       "25%        1.000000    26.000000            1.0  \n",
       "50%        1.000000    32.000000            1.0  \n",
       "75%        2.000000    39.000000            1.0  \n",
       "max       99.000000    99.000000            1.0  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "natality.describe(include = 'all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are over 31 columns with a mix of categorical, boolean, and numerical (nominal and interval). Here are some notable insights from the description of the columns:\n",
    "*  The \"day\" column is empty\n",
    "*  Although cigarette use would be an interesting feature to dive deeper into, their lack of entries limits their utility, and it is also is an example of possible response bias. This is similar for both \"alcohol_use\" and \"drinks_per_week\"\n",
    "*  The \"father_race\", \"father_age\", \"drinks_per_week\", \"weight_gain_pounds\", \"cigarettes_per_day\", \"apgar_1min\", and \"apgar_5min\" columns seem to have imputed missing values with 99\n",
    "*  There seems to be an error in \"mother_race\" as the value 78 does not refer to any race in the codebook\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>source_year</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>wday</th>\n",
       "      <th>is_male</th>\n",
       "      <th>child_race</th>\n",
       "      <th>weight_pounds</th>\n",
       "      <th>plurality</th>\n",
       "      <th>apgar_1min</th>\n",
       "      <th>apgar_5min</th>\n",
       "      <th>mother_race</th>\n",
       "      <th>mother_age</th>\n",
       "      <th>gestation_weeks</th>\n",
       "      <th>mother_married</th>\n",
       "      <th>cigarettes_per_day</th>\n",
       "      <th>drinks_per_week</th>\n",
       "      <th>weight_gain_pounds</th>\n",
       "      <th>born_alive_alive</th>\n",
       "      <th>born_alive_dead</th>\n",
       "      <th>born_dead</th>\n",
       "      <th>ever_born</th>\n",
       "      <th>father_race</th>\n",
       "      <th>father_age</th>\n",
       "      <th>record_weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>weight_pounds</th>\n",
       "      <td>-0.031654</td>\n",
       "      <td>-0.031654</td>\n",
       "      <td>-0.018221</td>\n",
       "      <td>-0.007352</td>\n",
       "      <td>0.100693</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.295961</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.030203</td>\n",
       "      <td>-0.016242</td>\n",
       "      <td>0.094301</td>\n",
       "      <td>0.539412</td>\n",
       "      <td>0.128734</td>\n",
       "      <td>0.069094</td>\n",
       "      <td>0.057672</td>\n",
       "      <td>0.048080</td>\n",
       "      <td>0.061612</td>\n",
       "      <td>-0.041102</td>\n",
       "      <td>-0.024513</td>\n",
       "      <td>0.046624</td>\n",
       "      <td>-0.037104</td>\n",
       "      <td>-0.063675</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>gestation_weeks</th>\n",
       "      <td>-0.020197</td>\n",
       "      <td>-0.020197</td>\n",
       "      <td>-0.009340</td>\n",
       "      <td>-0.001679</td>\n",
       "      <td>-0.012215</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.539412</td>\n",
       "      <td>-0.262976</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.005245</td>\n",
       "      <td>0.024157</td>\n",
       "      <td>-0.001194</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.039806</td>\n",
       "      <td>0.042263</td>\n",
       "      <td>0.024803</td>\n",
       "      <td>0.003356</td>\n",
       "      <td>-0.047917</td>\n",
       "      <td>-0.037601</td>\n",
       "      <td>-0.067928</td>\n",
       "      <td>-0.047118</td>\n",
       "      <td>0.008888</td>\n",
       "      <td>-0.042219</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mother_married</th>\n",
       "      <td>-0.047714</td>\n",
       "      <td>-0.047714</td>\n",
       "      <td>-0.034851</td>\n",
       "      <td>0.026959</td>\n",
       "      <td>0.023880</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.128734</td>\n",
       "      <td>0.031751</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.006163</td>\n",
       "      <td>0.067297</td>\n",
       "      <td>0.453928</td>\n",
       "      <td>0.039806</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.012922</td>\n",
       "      <td>0.102231</td>\n",
       "      <td>-0.009328</td>\n",
       "      <td>0.065241</td>\n",
       "      <td>-0.033308</td>\n",
       "      <td>0.024112</td>\n",
       "      <td>0.052984</td>\n",
       "      <td>-0.385408</td>\n",
       "      <td>-0.412853</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>is_male</th>\n",
       "      <td>-0.015313</td>\n",
       "      <td>-0.015313</td>\n",
       "      <td>-0.004865</td>\n",
       "      <td>-0.016713</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.100693</td>\n",
       "      <td>-0.020113</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.002981</td>\n",
       "      <td>-0.007616</td>\n",
       "      <td>0.009958</td>\n",
       "      <td>-0.012215</td>\n",
       "      <td>0.023880</td>\n",
       "      <td>-0.089836</td>\n",
       "      <td>-0.046895</td>\n",
       "      <td>0.017601</td>\n",
       "      <td>0.001966</td>\n",
       "      <td>-0.060342</td>\n",
       "      <td>-0.028452</td>\n",
       "      <td>-0.005410</td>\n",
       "      <td>0.009375</td>\n",
       "      <td>0.009078</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mother_age</th>\n",
       "      <td>-0.010471</td>\n",
       "      <td>-0.010471</td>\n",
       "      <td>-0.035056</td>\n",
       "      <td>-0.002501</td>\n",
       "      <td>0.009958</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.094301</td>\n",
       "      <td>0.054673</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.005443</td>\n",
       "      <td>0.071221</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.001194</td>\n",
       "      <td>0.453928</td>\n",
       "      <td>0.084871</td>\n",
       "      <td>-0.084010</td>\n",
       "      <td>0.014590</td>\n",
       "      <td>0.311000</td>\n",
       "      <td>0.013893</td>\n",
       "      <td>0.194886</td>\n",
       "      <td>0.307869</td>\n",
       "      <td>-0.160923</td>\n",
       "      <td>-0.069648</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cigarettes_per_day</th>\n",
       "      <td>-0.232307</td>\n",
       "      <td>-0.232307</td>\n",
       "      <td>-0.135037</td>\n",
       "      <td>-0.072928</td>\n",
       "      <td>-0.089836</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.069094</td>\n",
       "      <td>0.178039</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.066922</td>\n",
       "      <td>-0.032548</td>\n",
       "      <td>0.084871</td>\n",
       "      <td>0.042263</td>\n",
       "      <td>0.012922</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.346794</td>\n",
       "      <td>0.050624</td>\n",
       "      <td>0.231472</td>\n",
       "      <td>-0.088741</td>\n",
       "      <td>0.189315</td>\n",
       "      <td>0.141352</td>\n",
       "      <td>-0.127913</td>\n",
       "      <td>-0.100863</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>born_alive_alive</th>\n",
       "      <td>-0.015249</td>\n",
       "      <td>-0.015249</td>\n",
       "      <td>-0.013077</td>\n",
       "      <td>0.016947</td>\n",
       "      <td>0.001966</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.061612</td>\n",
       "      <td>0.059460</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.033610</td>\n",
       "      <td>0.012707</td>\n",
       "      <td>0.311000</td>\n",
       "      <td>-0.047917</td>\n",
       "      <td>0.065241</td>\n",
       "      <td>0.231472</td>\n",
       "      <td>0.092146</td>\n",
       "      <td>0.027735</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.059407</td>\n",
       "      <td>0.155024</td>\n",
       "      <td>0.989962</td>\n",
       "      <td>0.008353</td>\n",
       "      <td>0.071979</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>drinks_per_week</th>\n",
       "      <td>-0.205771</td>\n",
       "      <td>-0.205771</td>\n",
       "      <td>-0.098143</td>\n",
       "      <td>-0.037423</td>\n",
       "      <td>-0.046895</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.057672</td>\n",
       "      <td>-0.015952</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.009906</td>\n",
       "      <td>-0.004914</td>\n",
       "      <td>-0.084010</td>\n",
       "      <td>0.024803</td>\n",
       "      <td>0.102231</td>\n",
       "      <td>0.346794</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.063128</td>\n",
       "      <td>0.092146</td>\n",
       "      <td>-0.023849</td>\n",
       "      <td>-0.100823</td>\n",
       "      <td>0.076289</td>\n",
       "      <td>-0.100344</td>\n",
       "      <td>-0.090691</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>weight_gain_pounds</th>\n",
       "      <td>-0.114866</td>\n",
       "      <td>-0.114866</td>\n",
       "      <td>0.001419</td>\n",
       "      <td>0.025886</td>\n",
       "      <td>0.017601</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.048080</td>\n",
       "      <td>0.068484</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.369079</td>\n",
       "      <td>-0.018349</td>\n",
       "      <td>0.014590</td>\n",
       "      <td>0.003356</td>\n",
       "      <td>-0.009328</td>\n",
       "      <td>0.050624</td>\n",
       "      <td>-0.063128</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.027735</td>\n",
       "      <td>0.001679</td>\n",
       "      <td>-0.028469</td>\n",
       "      <td>-0.001812</td>\n",
       "      <td>-0.001526</td>\n",
       "      <td>-0.009218</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ever_born</th>\n",
       "      <td>-0.023700</td>\n",
       "      <td>-0.023700</td>\n",
       "      <td>-0.004279</td>\n",
       "      <td>-0.010293</td>\n",
       "      <td>-0.005410</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.046624</td>\n",
       "      <td>0.050289</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.031905</td>\n",
       "      <td>-0.017265</td>\n",
       "      <td>0.307869</td>\n",
       "      <td>-0.047118</td>\n",
       "      <td>0.052984</td>\n",
       "      <td>0.141352</td>\n",
       "      <td>0.076289</td>\n",
       "      <td>-0.001812</td>\n",
       "      <td>0.989962</td>\n",
       "      <td>0.199898</td>\n",
       "      <td>0.162821</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.003989</td>\n",
       "      <td>0.091037</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>apgar_5min</th>\n",
       "      <td>-0.344763</td>\n",
       "      <td>-0.344763</td>\n",
       "      <td>0.031191</td>\n",
       "      <td>0.004766</td>\n",
       "      <td>0.002981</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.030203</td>\n",
       "      <td>-0.011544</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.044234</td>\n",
       "      <td>0.005443</td>\n",
       "      <td>-0.005245</td>\n",
       "      <td>0.006163</td>\n",
       "      <td>0.066922</td>\n",
       "      <td>0.009906</td>\n",
       "      <td>0.369079</td>\n",
       "      <td>0.033610</td>\n",
       "      <td>0.033392</td>\n",
       "      <td>-0.038717</td>\n",
       "      <td>0.031905</td>\n",
       "      <td>-0.116003</td>\n",
       "      <td>-0.014985</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>wday</th>\n",
       "      <td>0.001919</td>\n",
       "      <td>0.001919</td>\n",
       "      <td>0.015302</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.016713</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.007352</td>\n",
       "      <td>0.027096</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.004766</td>\n",
       "      <td>0.018900</td>\n",
       "      <td>-0.002501</td>\n",
       "      <td>-0.001679</td>\n",
       "      <td>0.026959</td>\n",
       "      <td>-0.072928</td>\n",
       "      <td>-0.037423</td>\n",
       "      <td>0.025886</td>\n",
       "      <td>0.016947</td>\n",
       "      <td>-0.008351</td>\n",
       "      <td>0.004912</td>\n",
       "      <td>-0.010293</td>\n",
       "      <td>-0.014901</td>\n",
       "      <td>-0.009096</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mother_race</th>\n",
       "      <td>0.066999</td>\n",
       "      <td>0.066999</td>\n",
       "      <td>-0.036935</td>\n",
       "      <td>0.018900</td>\n",
       "      <td>-0.007616</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.016242</td>\n",
       "      <td>-0.012940</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.044234</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.071221</td>\n",
       "      <td>0.024157</td>\n",
       "      <td>0.067297</td>\n",
       "      <td>-0.032548</td>\n",
       "      <td>-0.004914</td>\n",
       "      <td>-0.018349</td>\n",
       "      <td>0.012707</td>\n",
       "      <td>-0.014783</td>\n",
       "      <td>-0.016050</td>\n",
       "      <td>-0.017265</td>\n",
       "      <td>0.207414</td>\n",
       "      <td>-0.022473</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>month</th>\n",
       "      <td>0.005887</td>\n",
       "      <td>0.005887</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.015302</td>\n",
       "      <td>-0.004865</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.018221</td>\n",
       "      <td>0.021076</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.031191</td>\n",
       "      <td>-0.036935</td>\n",
       "      <td>-0.035056</td>\n",
       "      <td>-0.009340</td>\n",
       "      <td>-0.034851</td>\n",
       "      <td>-0.135037</td>\n",
       "      <td>-0.098143</td>\n",
       "      <td>0.001419</td>\n",
       "      <td>-0.013077</td>\n",
       "      <td>0.007400</td>\n",
       "      <td>-0.015157</td>\n",
       "      <td>-0.004279</td>\n",
       "      <td>-0.008461</td>\n",
       "      <td>0.004161</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>born_dead</th>\n",
       "      <td>-0.015537</td>\n",
       "      <td>-0.015537</td>\n",
       "      <td>-0.015157</td>\n",
       "      <td>0.004912</td>\n",
       "      <td>-0.028452</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.024513</td>\n",
       "      <td>0.016220</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.038717</td>\n",
       "      <td>-0.016050</td>\n",
       "      <td>0.194886</td>\n",
       "      <td>-0.067928</td>\n",
       "      <td>0.024112</td>\n",
       "      <td>0.189315</td>\n",
       "      <td>-0.100823</td>\n",
       "      <td>-0.028469</td>\n",
       "      <td>0.155024</td>\n",
       "      <td>0.063975</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.162821</td>\n",
       "      <td>-0.026771</td>\n",
       "      <td>0.041448</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>source_year</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.005887</td>\n",
       "      <td>0.001919</td>\n",
       "      <td>-0.015313</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.031654</td>\n",
       "      <td>0.003322</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.344763</td>\n",
       "      <td>0.066999</td>\n",
       "      <td>-0.010471</td>\n",
       "      <td>-0.020197</td>\n",
       "      <td>-0.047714</td>\n",
       "      <td>-0.232307</td>\n",
       "      <td>-0.205771</td>\n",
       "      <td>-0.114866</td>\n",
       "      <td>-0.015249</td>\n",
       "      <td>-0.016905</td>\n",
       "      <td>-0.015537</td>\n",
       "      <td>-0.023700</td>\n",
       "      <td>0.211613</td>\n",
       "      <td>-0.016073</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>year</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.005887</td>\n",
       "      <td>0.001919</td>\n",
       "      <td>-0.015313</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.031654</td>\n",
       "      <td>0.003322</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.344763</td>\n",
       "      <td>0.066999</td>\n",
       "      <td>-0.010471</td>\n",
       "      <td>-0.020197</td>\n",
       "      <td>-0.047714</td>\n",
       "      <td>-0.232307</td>\n",
       "      <td>-0.205771</td>\n",
       "      <td>-0.114866</td>\n",
       "      <td>-0.015249</td>\n",
       "      <td>-0.016905</td>\n",
       "      <td>-0.015537</td>\n",
       "      <td>-0.023700</td>\n",
       "      <td>0.211613</td>\n",
       "      <td>-0.016073</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>father_race</th>\n",
       "      <td>0.211613</td>\n",
       "      <td>0.211613</td>\n",
       "      <td>-0.008461</td>\n",
       "      <td>-0.014901</td>\n",
       "      <td>0.009375</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.037104</td>\n",
       "      <td>-0.030099</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.116003</td>\n",
       "      <td>0.207414</td>\n",
       "      <td>-0.160923</td>\n",
       "      <td>0.008888</td>\n",
       "      <td>-0.385408</td>\n",
       "      <td>-0.127913</td>\n",
       "      <td>-0.100344</td>\n",
       "      <td>-0.001526</td>\n",
       "      <td>0.008353</td>\n",
       "      <td>-0.018031</td>\n",
       "      <td>-0.026771</td>\n",
       "      <td>0.003989</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.670280</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>born_alive_dead</th>\n",
       "      <td>-0.016905</td>\n",
       "      <td>-0.016905</td>\n",
       "      <td>0.007400</td>\n",
       "      <td>-0.008351</td>\n",
       "      <td>-0.060342</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.041102</td>\n",
       "      <td>-0.004535</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.033392</td>\n",
       "      <td>-0.014783</td>\n",
       "      <td>0.013893</td>\n",
       "      <td>-0.037601</td>\n",
       "      <td>-0.033308</td>\n",
       "      <td>-0.088741</td>\n",
       "      <td>-0.023849</td>\n",
       "      <td>0.001679</td>\n",
       "      <td>0.059407</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.063975</td>\n",
       "      <td>0.199898</td>\n",
       "      <td>-0.018031</td>\n",
       "      <td>0.016342</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>father_age</th>\n",
       "      <td>-0.016073</td>\n",
       "      <td>-0.016073</td>\n",
       "      <td>0.004161</td>\n",
       "      <td>-0.009096</td>\n",
       "      <td>0.009078</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.063675</td>\n",
       "      <td>-0.003661</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.014985</td>\n",
       "      <td>-0.022473</td>\n",
       "      <td>-0.069648</td>\n",
       "      <td>-0.042219</td>\n",
       "      <td>-0.412853</td>\n",
       "      <td>-0.100863</td>\n",
       "      <td>-0.090691</td>\n",
       "      <td>-0.009218</td>\n",
       "      <td>0.071979</td>\n",
       "      <td>0.016342</td>\n",
       "      <td>0.041448</td>\n",
       "      <td>0.091037</td>\n",
       "      <td>0.670280</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>plurality</th>\n",
       "      <td>0.003322</td>\n",
       "      <td>0.003322</td>\n",
       "      <td>0.021076</td>\n",
       "      <td>0.027096</td>\n",
       "      <td>-0.020113</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.295961</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.011544</td>\n",
       "      <td>-0.012940</td>\n",
       "      <td>0.054673</td>\n",
       "      <td>-0.262976</td>\n",
       "      <td>0.031751</td>\n",
       "      <td>0.178039</td>\n",
       "      <td>-0.015952</td>\n",
       "      <td>0.068484</td>\n",
       "      <td>0.059460</td>\n",
       "      <td>-0.004535</td>\n",
       "      <td>0.016220</td>\n",
       "      <td>0.050289</td>\n",
       "      <td>-0.030099</td>\n",
       "      <td>-0.003661</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>child_race</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>apgar_1min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>record_weight</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    source_year      year     month      wday   is_male  \\\n",
       "weight_pounds         -0.031654 -0.031654 -0.018221 -0.007352  0.100693   \n",
       "gestation_weeks       -0.020197 -0.020197 -0.009340 -0.001679 -0.012215   \n",
       "mother_married        -0.047714 -0.047714 -0.034851  0.026959  0.023880   \n",
       "is_male               -0.015313 -0.015313 -0.004865 -0.016713  1.000000   \n",
       "mother_age            -0.010471 -0.010471 -0.035056 -0.002501  0.009958   \n",
       "cigarettes_per_day    -0.232307 -0.232307 -0.135037 -0.072928 -0.089836   \n",
       "born_alive_alive      -0.015249 -0.015249 -0.013077  0.016947  0.001966   \n",
       "drinks_per_week       -0.205771 -0.205771 -0.098143 -0.037423 -0.046895   \n",
       "weight_gain_pounds    -0.114866 -0.114866  0.001419  0.025886  0.017601   \n",
       "ever_born             -0.023700 -0.023700 -0.004279 -0.010293 -0.005410   \n",
       "apgar_5min            -0.344763 -0.344763  0.031191  0.004766  0.002981   \n",
       "wday                   0.001919  0.001919  0.015302  1.000000 -0.016713   \n",
       "mother_race            0.066999  0.066999 -0.036935  0.018900 -0.007616   \n",
       "month                  0.005887  0.005887  1.000000  0.015302 -0.004865   \n",
       "born_dead             -0.015537 -0.015537 -0.015157  0.004912 -0.028452   \n",
       "source_year            1.000000  1.000000  0.005887  0.001919 -0.015313   \n",
       "year                   1.000000  1.000000  0.005887  0.001919 -0.015313   \n",
       "father_race            0.211613  0.211613 -0.008461 -0.014901  0.009375   \n",
       "born_alive_dead       -0.016905 -0.016905  0.007400 -0.008351 -0.060342   \n",
       "father_age            -0.016073 -0.016073  0.004161 -0.009096  0.009078   \n",
       "plurality              0.003322  0.003322  0.021076  0.027096 -0.020113   \n",
       "child_race                  NaN       NaN       NaN       NaN       NaN   \n",
       "apgar_1min                  NaN       NaN       NaN       NaN       NaN   \n",
       "record_weight               NaN       NaN       NaN       NaN       NaN   \n",
       "\n",
       "                    child_race  weight_pounds  plurality  apgar_1min  \\\n",
       "weight_pounds              NaN       1.000000  -0.295961         NaN   \n",
       "gestation_weeks            NaN       0.539412  -0.262976         NaN   \n",
       "mother_married             NaN       0.128734   0.031751         NaN   \n",
       "is_male                    NaN       0.100693  -0.020113         NaN   \n",
       "mother_age                 NaN       0.094301   0.054673         NaN   \n",
       "cigarettes_per_day         NaN       0.069094   0.178039         NaN   \n",
       "born_alive_alive           NaN       0.061612   0.059460         NaN   \n",
       "drinks_per_week            NaN       0.057672  -0.015952         NaN   \n",
       "weight_gain_pounds         NaN       0.048080   0.068484         NaN   \n",
       "ever_born                  NaN       0.046624   0.050289         NaN   \n",
       "apgar_5min                 NaN       0.030203  -0.011544         NaN   \n",
       "wday                       NaN      -0.007352   0.027096         NaN   \n",
       "mother_race                NaN      -0.016242  -0.012940         NaN   \n",
       "month                      NaN      -0.018221   0.021076         NaN   \n",
       "born_dead                  NaN      -0.024513   0.016220         NaN   \n",
       "source_year                NaN      -0.031654   0.003322         NaN   \n",
       "year                       NaN      -0.031654   0.003322         NaN   \n",
       "father_race                NaN      -0.037104  -0.030099         NaN   \n",
       "born_alive_dead            NaN      -0.041102  -0.004535         NaN   \n",
       "father_age                 NaN      -0.063675  -0.003661         NaN   \n",
       "plurality                  NaN      -0.295961   1.000000         NaN   \n",
       "child_race                 NaN            NaN        NaN         NaN   \n",
       "apgar_1min                 NaN            NaN        NaN         NaN   \n",
       "record_weight              NaN            NaN        NaN         NaN   \n",
       "\n",
       "                    apgar_5min  mother_race  mother_age  gestation_weeks  \\\n",
       "weight_pounds         0.030203    -0.016242    0.094301         0.539412   \n",
       "gestation_weeks      -0.005245     0.024157   -0.001194         1.000000   \n",
       "mother_married        0.006163     0.067297    0.453928         0.039806   \n",
       "is_male               0.002981    -0.007616    0.009958        -0.012215   \n",
       "mother_age            0.005443     0.071221    1.000000        -0.001194   \n",
       "cigarettes_per_day    0.066922    -0.032548    0.084871         0.042263   \n",
       "born_alive_alive      0.033610     0.012707    0.311000        -0.047917   \n",
       "drinks_per_week       0.009906    -0.004914   -0.084010         0.024803   \n",
       "weight_gain_pounds    0.369079    -0.018349    0.014590         0.003356   \n",
       "ever_born             0.031905    -0.017265    0.307869        -0.047118   \n",
       "apgar_5min            1.000000    -0.044234    0.005443        -0.005245   \n",
       "wday                  0.004766     0.018900   -0.002501        -0.001679   \n",
       "mother_race          -0.044234     1.000000    0.071221         0.024157   \n",
       "month                 0.031191    -0.036935   -0.035056        -0.009340   \n",
       "born_dead            -0.038717    -0.016050    0.194886        -0.067928   \n",
       "source_year          -0.344763     0.066999   -0.010471        -0.020197   \n",
       "year                 -0.344763     0.066999   -0.010471        -0.020197   \n",
       "father_race          -0.116003     0.207414   -0.160923         0.008888   \n",
       "born_alive_dead       0.033392    -0.014783    0.013893        -0.037601   \n",
       "father_age           -0.014985    -0.022473   -0.069648        -0.042219   \n",
       "plurality            -0.011544    -0.012940    0.054673        -0.262976   \n",
       "child_race                 NaN          NaN         NaN              NaN   \n",
       "apgar_1min                 NaN          NaN         NaN              NaN   \n",
       "record_weight              NaN          NaN         NaN              NaN   \n",
       "\n",
       "                    mother_married  cigarettes_per_day  drinks_per_week  \\\n",
       "weight_pounds             0.128734            0.069094         0.057672   \n",
       "gestation_weeks           0.039806            0.042263         0.024803   \n",
       "mother_married            1.000000            0.012922         0.102231   \n",
       "is_male                   0.023880           -0.089836        -0.046895   \n",
       "mother_age                0.453928            0.084871        -0.084010   \n",
       "cigarettes_per_day        0.012922            1.000000         0.346794   \n",
       "born_alive_alive          0.065241            0.231472         0.092146   \n",
       "drinks_per_week           0.102231            0.346794         1.000000   \n",
       "weight_gain_pounds       -0.009328            0.050624        -0.063128   \n",
       "ever_born                 0.052984            0.141352         0.076289   \n",
       "apgar_5min                0.006163            0.066922         0.009906   \n",
       "wday                      0.026959           -0.072928        -0.037423   \n",
       "mother_race               0.067297           -0.032548        -0.004914   \n",
       "month                    -0.034851           -0.135037        -0.098143   \n",
       "born_dead                 0.024112            0.189315        -0.100823   \n",
       "source_year              -0.047714           -0.232307        -0.205771   \n",
       "year                     -0.047714           -0.232307        -0.205771   \n",
       "father_race              -0.385408           -0.127913        -0.100344   \n",
       "born_alive_dead          -0.033308           -0.088741        -0.023849   \n",
       "father_age               -0.412853           -0.100863        -0.090691   \n",
       "plurality                 0.031751            0.178039        -0.015952   \n",
       "child_race                     NaN                 NaN              NaN   \n",
       "apgar_1min                     NaN                 NaN              NaN   \n",
       "record_weight                  NaN                 NaN              NaN   \n",
       "\n",
       "                    weight_gain_pounds  born_alive_alive  born_alive_dead  \\\n",
       "weight_pounds                 0.048080          0.061612        -0.041102   \n",
       "gestation_weeks               0.003356         -0.047917        -0.037601   \n",
       "mother_married               -0.009328          0.065241        -0.033308   \n",
       "is_male                       0.017601          0.001966        -0.060342   \n",
       "mother_age                    0.014590          0.311000         0.013893   \n",
       "cigarettes_per_day            0.050624          0.231472        -0.088741   \n",
       "born_alive_alive              0.027735          1.000000         0.059407   \n",
       "drinks_per_week              -0.063128          0.092146        -0.023849   \n",
       "weight_gain_pounds            1.000000          0.027735         0.001679   \n",
       "ever_born                    -0.001812          0.989962         0.199898   \n",
       "apgar_5min                    0.369079          0.033610         0.033392   \n",
       "wday                          0.025886          0.016947        -0.008351   \n",
       "mother_race                  -0.018349          0.012707        -0.014783   \n",
       "month                         0.001419         -0.013077         0.007400   \n",
       "born_dead                    -0.028469          0.155024         0.063975   \n",
       "source_year                  -0.114866         -0.015249        -0.016905   \n",
       "year                         -0.114866         -0.015249        -0.016905   \n",
       "father_race                  -0.001526          0.008353        -0.018031   \n",
       "born_alive_dead               0.001679          0.059407         1.000000   \n",
       "father_age                   -0.009218          0.071979         0.016342   \n",
       "plurality                     0.068484          0.059460        -0.004535   \n",
       "child_race                         NaN               NaN              NaN   \n",
       "apgar_1min                         NaN               NaN              NaN   \n",
       "record_weight                      NaN               NaN              NaN   \n",
       "\n",
       "                    born_dead  ever_born  father_race  father_age  \\\n",
       "weight_pounds       -0.024513   0.046624    -0.037104   -0.063675   \n",
       "gestation_weeks     -0.067928  -0.047118     0.008888   -0.042219   \n",
       "mother_married       0.024112   0.052984    -0.385408   -0.412853   \n",
       "is_male             -0.028452  -0.005410     0.009375    0.009078   \n",
       "mother_age           0.194886   0.307869    -0.160923   -0.069648   \n",
       "cigarettes_per_day   0.189315   0.141352    -0.127913   -0.100863   \n",
       "born_alive_alive     0.155024   0.989962     0.008353    0.071979   \n",
       "drinks_per_week     -0.100823   0.076289    -0.100344   -0.090691   \n",
       "weight_gain_pounds  -0.028469  -0.001812    -0.001526   -0.009218   \n",
       "ever_born            0.162821   1.000000     0.003989    0.091037   \n",
       "apgar_5min          -0.038717   0.031905    -0.116003   -0.014985   \n",
       "wday                 0.004912  -0.010293    -0.014901   -0.009096   \n",
       "mother_race         -0.016050  -0.017265     0.207414   -0.022473   \n",
       "month               -0.015157  -0.004279    -0.008461    0.004161   \n",
       "born_dead            1.000000   0.162821    -0.026771    0.041448   \n",
       "source_year         -0.015537  -0.023700     0.211613   -0.016073   \n",
       "year                -0.015537  -0.023700     0.211613   -0.016073   \n",
       "father_race         -0.026771   0.003989     1.000000    0.670280   \n",
       "born_alive_dead      0.063975   0.199898    -0.018031    0.016342   \n",
       "father_age           0.041448   0.091037     0.670280    1.000000   \n",
       "plurality            0.016220   0.050289    -0.030099   -0.003661   \n",
       "child_race                NaN        NaN          NaN         NaN   \n",
       "apgar_1min                NaN        NaN          NaN         NaN   \n",
       "record_weight             NaN        NaN          NaN         NaN   \n",
       "\n",
       "                    record_weight  \n",
       "weight_pounds                 NaN  \n",
       "gestation_weeks               NaN  \n",
       "mother_married                NaN  \n",
       "is_male                       NaN  \n",
       "mother_age                    NaN  \n",
       "cigarettes_per_day            NaN  \n",
       "born_alive_alive              NaN  \n",
       "drinks_per_week               NaN  \n",
       "weight_gain_pounds            NaN  \n",
       "ever_born                     NaN  \n",
       "apgar_5min                    NaN  \n",
       "wday                          NaN  \n",
       "mother_race                   NaN  \n",
       "month                         NaN  \n",
       "born_dead                     NaN  \n",
       "source_year                   NaN  \n",
       "year                          NaN  \n",
       "father_race                   NaN  \n",
       "born_alive_dead               NaN  \n",
       "father_age                    NaN  \n",
       "plurality                     NaN  \n",
       "child_race                    NaN  \n",
       "apgar_1min                    NaN  \n",
       "record_weight                 NaN  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "natality_corr = natality.corr(min_periods = 50)\n",
    "natality_corr.sort_values(by = [\"weight_pounds\"], ascending=False, inplace=True)\n",
    "natality_corr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "According to the Pearson correlation coefficients, the best features are gestation_weeks, mother_married, is_male, and mother_age for the numerical columns. From domain knowledge of child births, plurality is also selected as a strong feature for predicting the weight of newborns.\n",
    "\n",
    "Let's continue exploring the data through simple, but effective visualizations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/envs/py2env/lib/python2.7/site-packages/matplotlib/font_manager.py:1320: UserWarning: findfont: Font family [u'sans-serif'] not found. Falling back to DejaVu Sans\n",
      "  (prop.get_family(), self.defaultFamily[fontext]))\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdoAAAFpCAYAAAA7uul0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noQUAwND/AVCHEuNZQAtiAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f553ed3ea50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def get_distinct_values(column_name):\n",
    "  sql = \"\"\"\n",
    "SELECT\n",
    "  {0},\n",
    "  COUNT(1) AS num_babies,\n",
    "  AVG(weight_pounds) AS avg_wt\n",
    "FROM\n",
    "  publicdata.samples.natality\n",
    "WHERE\n",
    "  year > 2000\n",
    "GROUP BY\n",
    "  {0}\n",
    "  \"\"\".format(column_name)\n",
    "  return bq.Query(sql).execute().result().to_dataframe()\n",
    "\n",
    "df = get_distinct_values('is_male')\n",
    "df.plot(x='is_male', y='avg_wt', kind='bar');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f5538c8e650>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeoAAAFZCAYAAABXM2zhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzs3Xl8W+WVP/7P1WZZiy1Z1uYlsZM4\nu7OvxCRDQtgStgClmYEyUNpCaWi/dLpAS6FMQ2fpl+m0tKVtOvwoDHxnWAMJlLBlcSBxEuI4e+zE\nuy1ZlmVbm63t/v6Qr2I7siVLV5t13n31Razl3sfXTo6e+5znHIZlWRaEEEIISUuCVA+AEEIIIWOj\nQE0IIYSkMQrUhBBCSBqjQE0IIYSkMQrUhBBCSBqjQE0IIYSkMVGqBxCOxWJP9RDiolbLYLO5Uj2M\ntEPXJTy6LuHRdQmPrkt4mX5dtFrlmM/RjDoBRCJhqoeQlui6hEfXJTy6LuHRdQlvMl8XCtSEEEJI\nGqNATQghhKQxCtSEEEJIGqNATQghhKQxCtSEEEJIGqNATQghhKQxCtSEEEJIGqNATQghhKSxiIG6\ns7MT9957L2688UZs2rQJL7300hWvYVkWv/jFL7Bx40bcfPPNOH36dOi5t99+G9dddx2uu+46vP32\n2/yOnhBCSNZ76KEHknq+zs4O3HvvV5J2voglRIVCIX784x9j3rx5cDgcuOOOO7BmzRrMmDEj9Jr9\n+/ejqakJe/bswYkTJ/D000/j9ddfR29vL55//nm8+eabYBgGW7Zswfr165Gfn5/Qb4oQQkj2eOGF\n/0r1EBIqYqDW6XTQ6XQAAIVCgWnTpsFsNo8I1J988gluu+02MAyDRYsWob+/H11dXaipqcGaNWug\nUqkAAGvWrMGBAwewefPmBH07qdfY14zTDjvs9oEJv3eWegbUUlUCRkUIIYn3v5824Mi5Ll6PuXy2\nDl9ZP2Pc12zceDVqa2vR3d2Np556HE6nE36/D//0T49j4cLFY77n9tvvwtGjNVAq8/Ctb30bv//9\nb2A2m/Hd7z6Gqqp16OzswD//888wMOAGAPyf//NDVFYuHHEcv9+PF154HsePH4PX68Htt9+F2267\ng59vfsiEmnK0tbXh7NmzWLhw5EDNZjMMBkPoa4PBALPZfMXjer0eZrM5ziGnrwHfAH795Qvwsf6Y\n3l+hmobvLXmI51ERQkh2+Oijv2HFilW4776vw+/3Y3Bw7AmT2+3G4sVL8e1vP4rHH/8n/PnPf8Cv\nf/17NDZewvbtT6Oqah3U6gL8x3/8Djk5OWhtbcHTT/8Ef/nLyyOOs2vXTsjlcuzY8Vd4PB48/PDX\nsWLFKhQVFfP2fUUdqJ1OJx599FE88cQTUCgUI55jWfaK1zMMM+bjkajVsowssH66qxM+1o9lRQuw\nvHhh5DcM8/6FT9HQ14gcJZAnHbuLSqYbr0NMNqPrEh5dl/DS9bo8cnf42WuicXFl9epleOKJJ5CT\nI8S1116LOXPmjPkesViMm2++HgzDoLJyLiQSCYxGNfT6xTCbO6HVKmG3A8888wzOnTsHgUCA5uYm\naLVKDA7KIRIJodUqUVd3DOfPn0d19V4AgMNhh93eDa12Nm/fX1SB2uv14tFHH8XNN9+M66677orn\nDQYDTCZT6GuTyQSdTgeDwYCamprQ42azGStWrIh4vkxtVVbXegEAUDV1BSpyZ07ovSZtD5r72rH3\n/BGsLlqeiOGlnFarzPgWpolA1yU8ui7h0XW5EjcpLCubjd/85o/4/PNqPPbY97F167248cbwS60i\nkQjd3Q4AgNvtBSAOXVefzweLxY6//OWPkMnysGPHKwgEAtiwYQ0sFjt6epzw+fywWOwYGPBg27bv\nY+XK1SOOP9GfUVxtLlmWxU9+8hNMmzYN999/f9jXrF+/Hu+88w5YlkVtbS2USiV0Oh2qqqpQXV2N\nvr4+9PX1obq6GlVVVRMafCZp7m8FAMwomDrh9y4onAsAqOs+w+uYCCEkW5hMnVCp1LjlltuxefOt\nuHDhfFzHczod0GgKIRAI8OGH78Pvv3JZc8WK1XjnnTfg8/kAAC0tzXC73XGdd7SIM+pjx45h586d\nmDlzJm699VYAwGOPPYaOjg4AwNatW7Fu3Trs27cPGzduRG5uLp599lkAgEqlwre//W3ceeedAIBH\nHnkklFg2GTX3t0EulkEr16Db7ZjQe3UyLQwyHc72XIDH74FEKEnQKAkhZHI6fvwYXn31rxCJRMjN\nleGnP/15XMe7/fa78NOf/hCfffYxlixZhtzc3Ctec/PNt8Fk6sQDD/wDWJaFSqXGL3/5f+M672gM\nG24hOcUy8baOw+PEj6p/jjkFM/Hzjf8npu9h58UPsKf5M3yr8j4s0M5LwChTi27ZhUfXJTy6LuHR\ndQkv069LXLe+SXRa7G0AgKl5pTEfg25/E0IIGW1C27PI2Jr7hwK1siTmY0zNK0WeRImT3WcQYAMQ\nMPQ5ihBC4vGNb9wHr9c74rEnn3wG06ePvzc7nVCg5kmzPZhINiUv9kAtYASoLJyDgx01uNTXjBmq\ncr6GRwghWenPf76y7HWmoSkbT1r625AvyYMqJ77yqAsKg2vTdd2nI7ySEEJINqBAzYPewT70efrj\nWp/mzFLPgEQowUnLmbAFYwghhGQXCtQ8CK1Px3HbmyMWijG3YCa63N0wu/itmUsIISTzUKDmQctQ\noZMpcSSSDRe6/W2h7G9CCInVd77zTZw7x8+/o19+eRQ//OH3AADV1fvw8sv/HwBg//69aGy8xMs5\nxkKBmgfNQ1uz4kkkG25e4WwIGAGtUxNCSBJx1cUiqapah3vv/UcAwIEDe9HURIE6rbEsi5b+NhRK\nC6AQy3k5pkIsx/T8MjT1t6JvMHM38BNCSDJ0dnbghhtuwC9+8RTuu++r+OlPf4iBgZGdszZuvDr0\n588++xjbtz8NANi+/Wn89rfPYdu2b+EPf/gtzpw5hYceegD33//3eOihB9DS0nTF+d5//z0899y/\n4uTJE6iu3o/f//43+Md//Hu0t7fhgQf+IfS61tYWPPDAPXF/f7Q9K07WgR44fS7MLqjg9bgLtPNQ\n33sJp7rPYE3xSl6PTQghifBWwy4c7zrJ6zEX6yqxZUb4xhrDNTY24gc/+AkWLFiEZ5/9Od566/Wo\nz9Ha2oJf//r3EAqFcDodeP75P0EkEuHIkcP44x9/h+3b/z3s+yorF6Kqai2uuqoK11xzLQBALleg\nvv48Kipm4f3338NNN0UeeyQ0o44Tl0jG121vDm3TIoSQ6BmNRixYsAgAcP31N+Hkydqo33vNNddC\nKAy2VnY4HHjyyR/j3nu/gt/+9rkJrz/ffPNt2L37Pfj9fnzyyR5s3HjDhN4fDs2o48QVOomnIlk4\nhbkFKJIbcM7WgAHfIKSiHF6PTwghfNsyY3NUs99E4HpSD3tkzK89Hs+IZ6RSaejPO3a8gCVLluGX\nv/wVOjs7sG3btyY0jnXr1uO//uvPWLp0GWbNmoP8/PgbUdGMOk4t/W1gwKBUWcz7sRdo58EX8OFc\nzwXej00IIZNJR0cHTp2qAwB8/PGHodk1p6CgAE1NjQgEAti//7Mxj+NwOKDVagEE16IjkclkcLlc\noa9zcnKwcuUq/OpX/4Kbbro5lm/lChSo4xBgA2ixt0Ev00IqkkZ+wwRRkw5CCInO9OnT8cEHu3Df\nfV9Ff38fbr/9zhHPP/TQd/DDH34Pjz76EDSawjGP8w//8DW88MLv8PDDDyAQCEQ874YN1+G1117G\n/fcHk8kAYOPGG8EwDFasWBXfNzWE2lzGweQ0458P/1+sNCzF1+beHXqcr3ZrLMvip58/C6/fi19W\nPQmhQBj3MVMp09vQJQpdl/DouoRH1+VKnZ0deOKJ7+PFF19L9VAAAK+++jKcTge+8Y2Ho34PtblM\nkEQlknEYhkFl4Vw4fS5c7GtKyDkIIYTw5/HH/wkffrgbd921lbdjUjJZHC4nksVf43ssCwrn4kD7\nF6jrPo2Z6ukJOw8hhGQqo7EIu3btSos7Db/85a94PybNqOPQ3N8GASNAicKYsHNUqKdDKsxBHTXp\nIISQrESBOkb+gB9tjg4Uyw0QC8UJO49YIMJczSxYB3rQ4TQl7DyEEELSEwXqGHU4TfAFfJjCQ2vL\nSKhJByGEZC8K1DFqHuqYxUdry0jmaahJByGEZCsK1DFq4TpmJTCRjCMT56JCNQ0t9jb0DvYl/HyE\nEELSBwXqGDX3t0EsEKFIrk/K+ej2NyGEZCcK1DHw+L3ocJpQoihOWhGSylCVMrr9TQgh2YQCdQza\nHB0IsIGEFToJR5OrRqmiCBdsF+H2DUR+AyGEkEkhYsGTxx9/HHv37oVGo8GuXbuueH7Hjh14771g\n4XK/34+LFy/iiy++gEqlwvr16yGXyyEQCCAUCvHWW2/x/x2kQMtQRTK+O2ZFUqmdh1ZHB85Yz2Op\nfmFSz00IISQ1Is6ot2zZgh07doz5/IMPPoidO3di586deOyxx7B8+XKoVJfber300kvYuXPnpAnS\nwLCKZEnYmjUc9agmhJDsEzFQL1++HPn5+VEdbPfu3di8OTW9SJOpub8NUmEOdLKxO7AkQonCiAKp\nGqet5+EP+JN6bkIIIanBW61vt9uNAwcO4Mknnxzx+Ne//nUwDIO7774bd9999xjvHkmtlkEkSs9O\nUS6vG10uC+bqKqDXjf0BZrxOKPFYUboQf6vfCwtMqNTOTsg5EilR1yXT0XUJj65LeHRdwpus14W3\nQP3ZZ59hyZIlI257v/baa9Dr9bBarbj//vsxbdo0LF++POKxbDZXxNekygXbRbBgYZQaxywAn8g2\ndBXyCvwNe7G/4QgMguKEnCNRqD1feHRdwqPrEh5dl/Ay/bokpc3l7t27sWnTphGP6fXBPcYajQYb\nN25EXV0dX6dLmcsVyZK7Ps2pUE1DriiXmnQQQkiW4CVQ2+12HDlyBBs2bAg95nK54HA4Qn8+ePAg\nKioq+DhdSjWHKpIlN+ObIxQIMU8zC7bBXrQ5OlIyBkIIIckT8db3Y489hpqaGthsNqxduxbbtm2D\nz+cDAGzdGmyM/dFHH2HNmjWQyWSh91mtVjzyyCMAgtu2Nm/ejLVr1ybie0iqlv42yMUyaKTqlI1h\nQeE8HDXXos5yGqXKzLr9TQghZGIiBurnnnsu4kG2bNmCLVu2jHistLQU7777buwjS0MOjxPWgR7M\nLZgFhmFSNo65mlkQMkLUdZ/BpmnXpWwchBBCEo8qk00Ad9s7GR2zxpMrkmKmejraHB2wum0pHQsh\nhJDEokA9AS1DiWSpWp8ejit+crKbmnQQQshkRoF6Ai7PqFOT8T3cAi016SCEkGxAgXoCWvpbocrJ\nR35OXqqHAlVOPqYoS1DfewkurzvVwyGEEJIgFKij1DvYhz6PPemNOMazoHAeAmwAp63nUj0UQggh\nCUKBOkrNQx2zktnaMpKFWlqnJoSQyY4CdZS4RLKpytSvT3OMcj2kQinanaZUD4UQQkiCUKCOUqgi\nWRrNqBmGgV6uhcXVTd20CCFkkqJAHQWWZdHS34ZCaQHkYlnkNySRXqaFn/XDOtCT6qEQQghJAArU\nUbAO9MDpc6XFtqzR9DIdAMDssqR4JIQQQhKBAnUUuI5Z6XTbm6OXaQFQoCaEkMmKAnUUuIzvdEok\n44QCtZMCNSGETEYUqKPQbG8FAwalyqJUD+UKWlkhGDAwu7pSPRRCCCEJQIE6ggAbQKu9HXq5DlKR\nNNXDuYJYIIImt4BufRNCyCRFgToCs8uCQb8nrSqSjaaXaeHwOuH0ulI9FEIIITyjQB1BOieScSih\njBBCJi8K1BG02NM3kYxzOaGM1qkJIWSyoUAdQXN/GwSMACUKY6qHMibaS00IIZMXBepx+AI+tDk6\nUKwwQiwUp3o4YzLIKVATQshkRYF6HB1OE3wBH6akcSIZACjEcuSKcilQE0LIJESBehwtXKGTNE4k\nA4LNOQwyLSxuas6RSTx+L/796PPYff6TVA+FEJLGKFCPI50rko2mk2kRYAPopuYcGeNSXxOa+ltQ\n3Xwk1UMhhKQxCtTjaLa3QiwQwSjXp3ooERm4hDLK/M4Y9b2XAAAt/R10J4QQMiYK1GPw+L3odJpR\noiiGUCBM9XAi0slpL3WmqbddBAB4/V5Y3N0pHg0hJF1RoB5Dm6MDATaQ9uvTHAMVPckoHr8HTUPF\ndACgzd6RwtEQQtJZxED9+OOPY/Xq1di8eXPY5w8fPoylS5fi1ltvxa233ornn38+9Nz+/ftx/fXX\nY+PGjfjTn/7E36iTgKtIlo49qMMpzNVAwAioOUeGuNTXDD/rD+0oaHN0pnhEhJB0FTFQb9myBTt2\n7Bj3NcuWLcPOnTuxc+dOfOc73wEA+P1+PPPMM9ixYwd2796NXbt2oaGhgZ9RJwGXSJbuW7M4IoEI\nhVJqzpEpuPXpa0qrAATv4BBCSDgRA/Xy5cuRn58/4QPX1dVh6tSpKC0thUQiwaZNm/DJJ5mzDaXF\n3gapUAqdrDDVQ4maTqaF0+uCw+NM9VBIBPW2S2DAoLJwDgplBWinGTUhZAwiPg5SW1uLW265BTqd\nDj/60Y9QUVEBs9kMg8EQeo1er0ddXV1Ux1OrZRCJUpfA5fK60eWyYK6uAnrdxD+kAIBWq+R5VJGV\nFxbjlPUsBiVOlGsNkd+QAqm4Lulm0OdBs70V09RTMMWow1RVMY51nIRYyUIlzUv18NIK/b6ER9cl\nvMl6XeIO1PPmzcOnn34KuVyOffv24ZFHHsGePXvAsuwVr2UYJqpj2mypbdd4wdYAFiyKpEWwWOwT\nfr9Wq4zpffFSMsEPFec7mqCBLunnjyRV1yXdnO9pgC/gQ5liKiwWO8pUpTjWcRJ1TfWYo5mZ6uGl\nDfp9CY+uS3iZfl3G+5ARd9a3QqGAXC4HAKxbtw4+nw89PT0wGAwwmUyh15nNZuh06Rc8wgmtT2dI\nxjeHmnNkhvre4LasCvU0AMBUVTEAWqcmhIQXd6C2WCyh2XNdXR0CgQDUajUqKyvR1NSE1tZWeDwe\n7N69G+vXr497wMnQHGptmVmBOlT0hDK/09qFofXp6fnlAIAydXBnAQVqQkg4EW99P/bYY6ipqYHN\nZsPatWuxbds2+Hw+AMDWrVvx4Ycf4rXXXoNQKIRUKsVzzz0HhmEgEonws5/9DA8++CD8fj/uuOMO\nVFRUJPwb4kNLfysUYjkKpOpUD2VCFBI55CIZzajTmMfvQXN/C0qVRZCJcwEAOrkGOUIJJZQRQsKK\nGKife+65cZ+/5557cM8994R9bt26dVi3bl1sI0sRu8cB64ANcwtmRb2mnk70ci2a+lvhC/ggEvCS\nK0h41NjXAh/rR4VqeugxASNAscKIpv5WeP3etG6pSghJPqpMNkqLvR1A+nfMGkuoOYebmnOko9Hr\n05wSRRECbACdTnMqhkUISWMUqEdpybCKZKPROnV6q+8duT7NKVYYAdA6NSHkShSoR2m2BwN1plQk\nG01HNb/TlsfvRVPfyPVpTomyCEB6lhJlWRbV7YdoDZ2QFKFAPQzLsmjub4MqJx/5OZlZeCLUnMNJ\ngTrdNPU3w8f6MUM17YrniuQGMGDSsjlHQ28jXjv/Fv777BupHgohWYkC9TB9nn70e+wZty1rOGrO\nkb4u2IL1vWeqp1/xnEQogU6mRbujM2yxoFQ62HEYQPBuU8tQjQFCSPJQoB6G65g1JUPXpwFAKBBC\nm6uB2WVJu3/ws11978Ww69OcEoURA/4BWAdsSR7Z2BxeJ45bTkIiCGaiV3ccSvGICMk+FKiH4SqS\nZfKMGgiuU7t8bji81JwjXXDr0yVh1qc5JYrgOnV7GiWU1XQegy/gw03lG6GRqnHEdBxunzvVwyIk\nq1CgHubyjDqzA7WBSommHW59uiLM+jSnWMllfqdH0hbLsqjuqIGIEWK1cTnWFK2EJ+BFjel4qodG\nSFahQD2EZVm02NtQmKuBXCxL9XDiog8llNE6dboYb32aE5pRp0lC2cW+JphdXVikq4RCIsfqouUQ\nMkJUtx+iZRVCkogC9ZBudw9cPnfG3/YGgtXJAJpRp5OGMfZPD5cnUUIhlqfNjLq6PZhEtqZoJYDg\n+BZp56PDacKlvuZUDo2QrEKBekho/3SG3/YGhu+lphl1OvD6vWjsH399Ggi2gS1RFME60JPydWCn\n14XjljroZIUjbtdXFa8CABxop6QyQpKFAvWQllAiWeZmfHMUYjkUYjnNqNNEY38LfAHfuOvTHG6d\nut1hivDKxKoxfQlfwIc1RStH1LyvUE2DXqbF8a4TcHgoWZGQZKBAPaTZ3goGDEqVxakeCi/0Mi26\n3T3wBnypHkrWq7cN1feOIlBz69SpLHwSTCI7DBEjxCrDshHPMQyDquJV8LF+HDIdTdEICckuFKgB\nBNgAWuztMMh1kIpyUj0cXuhlWrBg0e22pnooWY+r7z1DNfb6NCcdtmhd6muGyWnGQu18KCTyK55f\naVgKsUCE6vZDCLCBFIyQkOxCgRqAydkFj9+TsfW9w9HLh7ZoUeZ3SoXWpxVGyKLYTaCXaSFihClN\nKOMqkVUVrwz7vFwsw1LdIljcVlwYultACEkcCtQAWuxD69MZXJFsND0150gLofXpcbZlDScUCGFU\nGNDhNMEf8Cd4dFdyeV34susEtLmaET2zR6OkMkKShwI1hlUkmwQZ3xwK1OlhIuvTnGKFEb6AD13u\n7kQNa0w1puPwhkkiG60srxQliiLUdZ9G72BfEkdISPahQI1gIpmAEaBYbkz1UHijkRZAyAgpUKfY\nRNanOalKKGNZFgc7DkPICLHKuGzc13JJZQE2gC86jiRphIRkp6wP1L6AD+2OThQrjBALxakeDm8u\nN+fooipSKTLR9WlOiYLbopXcderG/hZ0OE1YqJ0HpUQR8fXL9YsgFeaguuNwSm7TE5Itsj5QdzhN\n8AV8k6Ii2Wh6mRZu3wD6PY5UDyUrNQ2tT89QR3/bGwCKuRl1kjO/q4fWm7lKZJFIRVIsNyxB72Af\nzvScT+TQCMlqWR+oL69PT55EMg6X+d1FFcpS4kJvsL73eElZ4cjEuSiQqpMaqLkkssJczbj1yEer\nGgrqlFRGSOJkfaBu4TpmTdIZNQCYaJ06JeptFye8Ps0pURTB7nGgb9CegJFdqcbMJZGtgICJ/p+F\nEmURyvOm4oz1PLrdPQkcISHZK+sDdbO9DWKBGEa5PtVD4R0XqLsoUCcdtz5drDDG1I2tOLROnfhZ\nNcuyONh+GAJGEDGJLJyri1eBBRvaf00I4VdWB2qP34NOpxmlyiIIBcJUD4d3l2fUdOs72ZpC+6cn\ntj7NSWZCWROXRFY4D3kS5YTfv1i3AHKRDF90HIGPStYSwrusDtRtjg4E2MCkvO0NADKxDEqxAl1O\nmlEnW6zr05wSZfISyqqHZsJrxqhEFolEKMZK41LYvQ6csJzmc2iEEACiSC94/PHHsXfvXmg0Guza\nteuK59999138+c9/BgDI5XI8/fTTmD17NgBg/fr1kMvlEAgEEAqFeOutt3gefnwmcyIZRy/X4mJv\nE7x+76TafpbuGmwT3z89XIFUDakwJ+GlRN0+N46ZT6BQWoBZ6hkxH6eqaCU+bT2A6vZDWKpfyOMI\nCSERZ9RbtmzBjh07xny+pKQEr7zyCt577z08/PDDePLJJ0c8/9JLL2Hnzp1pF6SBYYF6ks6ogcvN\nOSzUnCNpguvTzTGvTwMIFuBRGGF2dsHj9/I8wsuOmI7DG/BiTdHKCSWRjaaX6zBTPQMXei/CRPXl\nCeFVxL+Zy5cvR35+/pjPL1myJPT8okWLYDKlto/uRLTYWyEVSqGVFaZ6KAmjlwW3aPG1Tn3BdpEa\nfUTQ1N8Kb5T9p8dTrCgCCxadzsT8neLaWQoYAVbGkEQ22tVD9b+rO2irFiF8injreyLeeOMNrF27\ndsRjX//618EwDO6++27cfffdUR1HrZZBJEpscpfL44bZZcE83UzodWN/EImVVjvxpJxEmOmdAjQA\nTqY/7jG195vwn8f/CIlQjIeW34uqqcsnfIx0uS6JtK8reKdm6dR5UX+/4V43p78c+9s/Rx96oNXO\n5XWMANBgbUK7oxMrSxZjRklR3MfboFmFNxveRY3pGL6+4i5IRJK4j5kNvy+xoOsS3mS9LrwF6kOH\nDuGNN97Aq6++Gnrstddeg16vh9Vqxf33349p06Zh+fLI/7jbbC6+hjWmC7YGAECRtAgWC797VbVa\nJe/HjJXUG/zFvdjVBos2vjG9e+ETAIAv4MdvDv0XznZcwi3Tb4j6lmk6XZdEqm07CwYMdAJjVN/v\nWNclHwUAgHOmRizI43/dd9fZTwEAyzVLefu5rNIvw9+aP8WHZw7GtNVruGz5fZkoui7hZfp1Ge9D\nBi9Z3+fOncNPf/pT/P73v4darQ49rtcH9yZrNBps3LgRdXV1fJyOF9z69JRJ1DErHE2uGiJGGPde\n6kG/B4dMR5EnUeLHy78LnawQH7XsxR9OvAiXN/EfrDIFtz5dpDDEvD7NMcoNYMAkpDmH2zeAo+Za\naKRqzCqIPYlstKuKVoIBEypHSgiJX9yBuqOjA9u2bcO//du/obz8coary+WCw+EI/fngwYOoqKiI\n93S8aR6qSDaZE8mAYFKSVlYYd3OOY+ZauH0DuKpoBYoVRvxg6TbM1czCmZ7z+Lejv0Wn08zjqDMX\ntz49M8ZtWcNJhGLoZVq0O0y8N1Y5aj4OT8CLq+JMIhtNk6vGPM0sNPa3oDXJ3b8Imawi3vp+7LHH\nUFNTA5vNhrVr12Lbtm3w+YJFDbZu3Yrf/e536O3txc9//nMACG3DslqteOSRRwAAfr8fmzdvvmL9\nOpWa7W1QiOUokKojvzjD6WVadDrN6PfYkZ+TN+H3syyL/e1fgAETqu0sE+fi4QX3471LH2JP82f4\n1dHn8bW5X8VC7Ty+h59R6nuH+k/HWOhktBJlEUyuLlgHbCjMLeDlmCzLonqoEtlqHpLIRru6eDVO\nWc+huv0LbJ19B+/HJyTbRAzUzz333LjPb9++Hdu3b7/i8dLSUrz77ruxjyyB7B4HegZsmKuZBYZh\nUj2chOMyv82urpgCdbO9Fa36xdbEAAAgAElEQVT2diwonAe1VBV6XMAIcOv0G1GiMOLls6/jTydf\nwqbyjbihbAOvs7RMUt/bOLR/mp9AXaww4qi5Fm2ODt4CdYu9DW2ODizUzo/p9yGSuZpZUOeocMR8\nHLfP2ASpSMr7OQjJJln5r2mLnds/PXkLnQwXKiUaY4WyA23B9ca1xavDPr9UvwjfX/oICqRq7G78\nCH859QoGfAOxDXaIP+DHWesFvFH/Ls711Md1rGTxBnxo7GviZX2aw7W8bOfxNnJ1+1AlsijbWU6U\ngBGgqnglBv0eHDEfT8g5CMkmvG7PyhSh9elJnkjG0ctjb87h9LpwrKsWhbmacZOOSpVF+OGybfjL\nqVdQazkFs8uCb1X+I7QyTdTn8vq9ONtzAbWWU6jrPgO3zw0AOGquxTOrfwyJMP7tPonUzNP+6eFK\nQr2p+alQ5vYN4GhXLQqkaswpSFzOyGrjCuxu/AgH2g+hqmhVVty5IiRRsjJQczPqKdk2o46h6Mmh\nzqPwBny4unhVxNvZSokC2xZ9A2827MK+toP4t6O/wdfn34PZ4wSEAd8gzvScR23XSZyynsWg3wMA\nUOXkY4VhCQZ9gzhkOor97V/g2inrJjz+ZKq3cevT8SeScfJzlFCKFbx10TpqroXH78FVU65J6PJE\nfo4SC7XzcbyrDo39zZiWX5awcxEy2WVdoGZZFs39bVDl5CM/Z3Jujh8tV5SLPIlywjPqABvAgfYv\nIBKIot4TKxQI8ZWZt6JEUYT/Of8Wnq/dgdtnbML60qtDr3F53TjZfQa1llM423Me3qGOS4XSAlxd\nXIlF2kpMzSuBgBHA5XXhRPcp7Gn+DFVFK9N6vZNrxBFrfe+xlCiLcLbnAlxeN2Ti3LiOdXCoEtnq\nIv6TyEa7umgVjnfV4ddf/hHl+VMwW12BWQUzMFVZOim71RGSKFkXqHsH+9DvsWOhdn6qh5JUepkW\nDb2N8Pi9kETZnOO8rQEWtxUrDUuhEMsndL6ripbDKNfhzyf/ircadqHV3oEl9jk4cOkoztsa4Gf9\nAACDTIdFumBwLlEYr7hFKhPLsKF0HXY1fojPWg/ixvINExpHsnDr08UK44SvVSTFCiPO9lxAu6Mz\nrmzy09ZzoaRAVQ7/1fhGm6mejjsqbsZRUy0u9jahobcRuxr3QCrMwQzVNMwuqMAs9QwY5Xq6NU7I\nOLIuUDeHbntnx/o0Ry/Tor73EizubhQP9TqO5EDbFwCC221iUZ4/FT9c/ij+fPJlHDF/iSPmLwEA\npYqioeA8Hwa5PuJxrildg8/aDuCT1n1YV7IaMp4StfiUiPVpzuV16o6YA7Uv4MMb9e+CAYPN067j\nc3hjYhgG60uvxvrSq+H0ulBvu4jztgacs9XjlPUsTlnPAgDyJErMVE8PzbizYcskIRORdYG6JdTa\nMssCtXyoOYezK6pAbRvoRV33GZQqilAWRxtQVU4+vrf4W9jX/jnk8hxUyCpQmBt9ghkASEVSXDf1\nGrzdsBuftOzHzdNviHk8iZKI9WkO9/NqjyOhbG/bQXS5urG2+KqoP6jxSS6WBT+c6SoBBH+/ztka\ncL6nHudtDThqrsVRcy0AQJdbiFkFFbhRsBb5mNjvCiGTUdYF6mypSDYal1AW7Tr1wY7DYMHi6pLV\ncd+WFAvFuHbKurhq8a4tXo1PWvbj07Zq/F1pFZQSRVxj4lt9gtangeDPTiQQxZxQ1jdoxweNH0Mu\nkiVtNh2JWqrCauMyrDYuA8uy6HSacd7WgPO2etTbLuFA+xc40P4FluoW4pbpN/K2h5yQTJRVgZpl\nWbTY21CYq0nL26eJNJF2l/6AH5931CBXJMUy/eJEDy0qEqEE109dj9frd+Kjlr3YMmNzqocU4vK6\ncamvOSHr00AwQa9IrkeH0wx/wD/hRKx3L36AAf8g7p55O2/7u/nEMAyKFAYUKQy4prQK/oAfF2wX\n8UHLRzjWdQInLKewtuQq3FC2IS3HT0iiZVXBE4vbCpfPnXWzaQAokKogEoiimlGf6D6NPo8dKw1L\nkZNGe5fXFK+EOkeF/W2fo2+wP9XDARD88Pf/zr8Fb8CLpTr+O1xxShRF8AV8ME8wc7+pvwWHTEdR\nrDCiqjgxBU74JhQIMUczE9s3/hD3z92K/Jw8fNp6AE998a/4uGVfaJcAIdkiqwJ1qCJZHGuumUrA\nCKDLLYTZZYnY4OFyEtmqZAwtamKBCDeWbYA34MOHzZ+mejgAgEOmYzjWdQLleVMTus+7eFhCWbQC\nbACvXwiW8b2r4paMK+sqYARYZliMJ1f9ALfP2AQAeLthN/750L/jqOk4AmwgxSMkJDky629unC5X\nJMu+QA0EE8oG/R70ecaejZqcZlzovYiZqulRZWQn2yrjMhRKC1DdfhhWty2lY+lyWfC/F96BVCjF\n/fO2JnRvcIlyqJToBBLKakxfoqm/BUt0CxKS5JYsYoEI105Zh5+v/hHWl16NvsF+vHjmNfz70edD\nSXyETGZZFqjbwIAJbXfJNpdrfo+9Tn1gqI/w1SWxbclKNKFAiJvKN8LP+vG3pk9SNg5fwIcXT78K\nj9+DrbO3QJPgZKdihQEAou5NPeAbwM6LH0AsEIdmo5lOLpbhjoqb8eSqH2CpbiFa7G349fE/4oW6\nF2GiNqtkEsuaQB1gA2h1tMMg10Eqykn1cFIiUub3oN+Dw6ZjyJMosbAwfdtVLjcshl6mwyHTUXS5\nulMyhl2X9qDF3o6VhqVYpl+U8PPlinKhkarR5uiIqjf135o+Rb/Hjuum/t2k25dcmFuAB+b/A364\nbBtmqMpxsvssttf8B1479yb6BmPbVUBIOsuaQG1ydsHj92RNx6xwLtf8Dh+oj5qPw+0bwJqiFWld\n4lHACLCpfCMCbAAfNH2c9POf66nHRy17oc3V4Cszb03aeUsURXB4nej3jB+MzC4LPm09gAKpGtdO\n+bvkDC4FpuaV4nuLH8K3Ku+DNrcQ1R2H8fShf8Wr595AreUU3HF2cCMkXWTN9qxQRbIsK3QyHBeo\nzWFufbMsiwNtX4ABk7D2h3xarKtEcbMRR0zHcd3Ua2BM0nq63ePAX8/8PwgYAe6f9/dJrT1erDDi\nRPdptDk6x+0j/Vb9e/Czftw+Y1PU5WIzFcMwWKCdh3ma2fi8swbvN36Mgx01ONhRAwEjwPT8MszV\nzMI8zWwUyQ1UqpRkpKwJ1C1Z1toyHKlIinxJXtgtPk39rWh1dGBh4TyopaoUjG5iBIwAm8uvwx9P\nvoTdjR/hwfn3JPycLMviv8+9jj6PHbdNvynpSYmXE8o6ME8zK+xrTnWfxSnrOcxUTcdibWUyh5dS\nQoEQVxevxpqilWjqb8UZ63mcsZ5HQ28j6nsvYefFD5AvycNczSzM1czCbHVF3A1OCEmWrAnUzfY2\nCBlhaJtLttLLdbhga8Cg3zNij/SB9vjqeqdCZeFcTFWW4nhXHVrtHShVJvZnu6/9c5zsPotZ6hnY\nMGVtQs8VTmiL1hgJZb6AD282vAcGDO6ceUtWzh4FjADT8qdiWv5UbJ52HeweB872XMBp6zmc7bmA\nLzqP4IvOIxAwApTnTR2abc9CiaIoK68XyQxZEah9AR/a7R0oVhggFmTFtzwmvUyLC7YGdLm6Q4HN\n4XXiWNcJaHM1mFUwI8UjjB7DBBtM/O7EX7C78UM8tOD+hJ2r3dGJtxt2Qy6W4Wtz707JnmSNVA2p\nUDrmFi2unve6ktTU805HSokCKwxLsMKwBAE2gBZ7G04PzbYv9TXhYl8j3rv0N+RJlFhpWIoNU9am\nXXlaQrIianU4TPCx/qzrmBVOaJ3a1RUK1Ic6j8IX8KGqeFXGFcWYUzAT0/PLcLL7LJr6W1CWN4X3\nc3j8Xrx4+lX4Aj48OP+epLSIDIdhGBQrjLjU13RFu9JQPW+xDJvK06Oed7oRMAKU5U1BWd4UbCrf\nCIfXiXPWCzjdcx6nrefwUcte7G07iKrilbh2yrqU/ZwJGS2z/lWOUXMWVyQb7XKgDq5TB9gADrQf\nglggwirjslQOLSYMw+DmadcDCG6ZSoS3G3ah02nG2uKrUFk4NyHniFaJsggsWHQ6TSMe5+p53zzt\neqqHHSWFWI5lhsW4b+5Xsf2qn+ArM2+DQizHZ63VeOrzf8Fr59+C1d2T6mESkh0z6pYsr0g2HNec\ng8v8Pt/TgG63FSsNSxPSUCIZKoZ6GZ/tuYB626WYezaHc8JyGvvbv0CR3JAWhUNKhm5pt9k7Qr/P\nw+t5Z0LGfjoSC8VYV3IV1hStQI3pS3zY/Bmq2w/h844arDAswfVTr4Fu6EMuIcmWNTNqsUAMw1CQ\nymZqaT7EAnGo6EkmJpGFw7VvfO/Sh1EVBIlG72Af/vvs6xALRLh/3t+nxVanklDN7+A6dYAN4H8v\n7AQA3FVxa8YtXaQbkUCEq4pW4Gcr/wn3zf0qdLmFONR5FM8c+hVePP0qOhymyAchhGeTfkbt8XvQ\n6TSjLK80rYt4JIuAEUAnCzbn6Bmwoa77DEqVxSjL8LsN5flTMV8zG6es53DOVo85BTPjOl6ADeCl\nM/8Dp8+Fu2fehqKhEp6pZpTrIWAEoeYcNaYv0dzfiqW6hbzeSch2QoEQKwxLsEy/CLWWU/hb0yc4\naq7FUXMtFmnn44ayDShVFqd6mCRLTPpA3eboQIANZHVFstEMMh3aHZ3YfekjsGBxdfGqSbE1ZdO0\n63DKeg7vXfoQs9UVcX1PH7fswwVbAyoL56bV3QaxUAydTIsORyfcPvekq+edbgSMAEt0C7BYW4lT\n1rP4oPET1FpOodZyCvM1s3FD2QaU509N9TDJJBdVoH788cexd+9eaDQa7Nq164rnWZbF9u3bsW/f\nPkilUvzLv/wL5s0L1op+++238Yc//AEA8PDDD+P222/ncfiRNfdTRbLRuLW2w6ZjyBVJsUy/OMUj\n4scUZQkWaStRazmJU9azMSd+NfW34L1LHyJfkod7Zt+Vdh9iShRGmJxm/PfZN9DvsWNz+XUZUaQm\nkzEMg8rCuZivmYNzPfX4oOljnLKewynrOeRJlNDJCqHLLYROpoVOVghtbiG0uRqI02C5hGS+qAL1\nli1bcM899+BHP/pR2Of379+PpqYm7NmzBydOnMDTTz+N119/Hb29vXj++efx5ptvgmEYbNmyBevX\nr0d+fvK2PYRaW9LWrBDDUKBmwWKlYemIwieZblP5RpywnMKuS3swTzN7wmu2A74BvHj6NbAsi6/N\nvRsKSfol2JUoinDUXIvjlpPQSNXYkMA+2GQkhmEwRzMTczQzUW+7hM/aqtFm78DF3iY09DaOfC0Y\nFEhV0A4L4MGArkWBVAWhQAiWZeFj/fAFfKH/e4f92cde+ZjClYO+fjdYlkUALMCyYMGGvmaHfc39\nL8AG4Av44fV74Al44Rn6r9fvDfu1d9jXueJcqHPyocpRQZWTB1VO/uX/S4P/nUz/hqSjqAL18uXL\n0dbWNubzn3zyCW677TYwDINFixahv78fXV1dqKmpwZo1a6BSBT/tr1mzBgcOHMDmzZv5GX0Umu2t\nkAql0MoKk3bOWJxp6oEmTwp9QeK31ujkl7NX0+m2Lh+KFAYs1S/EUXMtXjr5JoryNBN6f0NvI7rd\nVmyc8neYXVCRoFHGZ3ib1ttnbE6LJLdsVKGeFsoL8AZ8sLqtMLu60eWywOLuRtfQn8/Z6nHOVj/i\nvQJGAAEjgC/gS8XQwxILRBALxJAIJcgR5kAhUUAkEMHldaHTaUaLvX3M9+aKckNBXJ2Tj/yc/KQX\nl5JbcuB0DiblXEKBEMv1i8etuc8nXq6k2WyGwXA52cZgMMBsNl/xuF6vh9kcuW+sWi2DSMRP4pfb\n58ZcfQX0uuQWL9BqlVG/dmDQh1+/fgLzpxfin791VQJHFSTIKQHrF0Ie0KGybHrCzzfcRK5LrG6Z\ndROOdJ7A0e4jQAxdMGcUlOH+lXdClMTkw4lcl5y8WRCfEmN24XRsnLs67W7N8ykZvy98KYIalbiy\nsp/bOwCTw4JOuxmd9i502rtgcljAsgGIhGKIBSKIhCJIBGKIhKJgwBQGg6ZYKIJIIIJEKIZIEHxO\nKBCAAQOGEYBBcJYvYLjHGAiY4H+5rxkEHxMLxcgRSiARSiARBf+cI5RAIpJAIhBDIBj77hPLsnB4\nnOhx98Lq6kWPuxc9btvlP7tssLp70ZlFfcHl8hzcUrIxKefiJVCH2w7DMMyYj0dis7n4GBYA4AdL\ntyFXJIXFkrw+tVqtckLnM/W44POzaOzoS8o4G9r7MHhyDSCQwtzVD0GS/qGf6HWJlak1gMFTayDI\nceOhW+cjNyf6X3OGYTAjvxw2K3+/g5HEcl2eWP495Ofko7vbkaBRpV6yfl+SQQEVKnJVqMidBcS5\nSzTu6+If+j+AQQCD8ADwRPVWGfIhE+ejVDwVCDOZHPANom+wD32efvgC/tjHGIP8fBn6+pLz91bI\nCDFNVcbr7+d4H0p5CdQGgwEm0+X9hSaTCTqdDgaDATU1NaHHzWYzVqxYwccpo6bJLUjq+WLRaw/e\nrulzeOAe9E0osMSi0+oE65FhAIDJ6kJRYfqtw8aj3eIAO6CAf0CBAWsBllZOvrrXVHyDpCOpKAdS\nkQ56efJrVmi1SlhEk+OD3Wi8VEdYv3493nnnHbAsi9raWiiVSuh0OlRVVaG6uhp9fX3o6+tDdXU1\nqqqq+DjlpGJzXF5XMfUk/hPh8HM0mfoTfr5ka+t2hv58vD6Ge9+EEJJGopq6PfbYY6ipqYHNZsPa\ntWuxbds2+HzBJIitW7di3bp12LdvHzZu3Ijc3Fw8++yzAACVSoVvf/vbuPPOOwEAjzzySCixjFzW\nOyxQd1qdKDcmNkHBNOy2bmOnHVfNn1wzznaLE0IBA02+FKcarfB4/ZCIqdgNISQzRRWon3vuuXGf\nZxgGTz31VNjn7rzzzlCgJuH12i+vD3UmYW200+pCbo4QHm8ATZ2Ta0YdYFl0dDth1MhQOU2DDw63\n4EyTDYsq0jvrnxBCxkKFgdNAMm99+/wBWHrdKNLIUVwoR0uXAz5/IKHnTCZr3wAGvX4UFcqxuCK4\njnu83pLiURFCSOwoUKeBXscgBAyDHIlwxG3pRLD0uuEPsDBoZCgz5sHrC6Bj2Jpupmu3BL+XYq0C\n04rykCcT40RDNwIBfhp1EEJIslGgTgO99kHkKyQwFshgtrkSGlS4GbuhQIZyY3A7QOMkuv3dPrRd\nqaRQDoGAwcIZheh3eXGpY/J8j4SQ7EKBOsVYlkWvwwOVQgKjRgafn0V3nzth5+Nm7EaNHGWGYNJa\nY+fk2dJweUYd3HK2eCbd/iaEZDYK1CnmHPDB5w9ApciBYah8aCLXqTuHzaiLtXKIRYJJlVDWZnFC\nIhKgUJULAJg7VQ2JWEDbtAghGYsCdYrZhoqdqJQ5MGqCs8BEZn6brC4IGAY6dS5EQgGm6BRo73bC\n401uFaFE8PkDMPU4UVQoD1Vbk4iFmF+uganHhU7r5FmLJ4RkDwrUKcbtoU7WjNrU44JWJYVIGPzR\nlxnz4A+waO3K/FKUXTY3fH42dNubs3hoa1YtzaoJIRmIAnWKceVD1Yoc6AtywSBxM2q7ywOH2xua\nuQNAmWHyJJRx2evFhYoRjy+YrgHDUJUyQkhmokCdYtweapVSArFIiEKVNGEz6uEZ3xyuClqTKfMT\nytosQxnfo2bUSpkEFSUqXGzvQ58zuuYDhBCSLihQp1ivIxg4VIocAIChQI5+pweuAS/v5+Jm6gbN\n5UBt0MgglQgnxYy6vfvyHurRFlcUggVwooFm1YSQzEKBOsVCt76VwUBtHAqinQmYVYebUQsYBmUG\nJUxWF9yD6dPEPhbtFidkOSKoFJIrnqN1akJIpqJAnWI2xyDEIgFkQ60tQwllCVinvryHWjbi8TJD\nHlgAzRl8+9vr88Nsc6FYKw/b81ynlqG4UI7TTT0Y9GR+hjshJHtQoE6xXscg1IqcUHAJzagTEKg7\nrU4ocsVQykbOOMuGKpRl8jp1p9UFlg1/25uzqKIQXl8Apxp7kjgyQgiJDwXqFPIHAuh3ekbcqk3U\nFq1gM46BEbe9OVxCWSavU4cqkhXKx3wN16SjlqqUEUIyCAXqFOp3esGywWInnDy5BLk5It6Lc3TZ\n3Aiw7IhEMk5hvhSKXHFGB+q27vAZ38OVGZXIV0hw4qIV/sDk6RhGCJncKFCn0PBiJxyGYWAokKHL\n5uY1mHAzdGOYGTUzlFDW3TcAuyszty9xM+qicWbUAobB4hmFcLi9aGjrS9bQCCEkLhSoU4jL+B4e\nqIHgOrU/wKK7d4C3c3Ez9HAzaiBYoQzI3ISydosTeXLJFevvoy0K9aim7G9CSGagQJ1Cw4udDMet\nI/O5RSvc1qzhMrnlpXvQB2v/wLjr05w5U9XIkQhRW98NlqUe1YSQ9EeBOoW4W9/qMDNqgN8tWiar\nC0IBA+1QV6nRMrnlZah06Djr0xyxSIDK8gJ09bpD7yOEkHRGgTqFhnfOGs4wVIvb1MNPIGFZdqgZ\nR26oGcdoamUOVAoJmkyZN6PmKpKVjLM1a7jLParp9jchJP1RoE6h0eVDOTpVLhiGv73UdpcXzgHf\nFYVORis35qHX4Ql9gMgUXI3vaG59A8EmHQKGoUBNCMkIFKhTqNcxCFmOCDli4YjHxSIBtKpc3vZS\nR1qf5nAJZU0Ztk7N3cIeL+N7OLlUjFlTVGjs7M+4DyWEkOxDgTqFeu2DV9z25hgKZLC7vHC442/O\nESnjm1POtbzMsNvf7RYnNHlS5A6VYY3GIq72NzXpIISkOQrUKeLx+uEc8EEdpoEEMCyhjIdZ9eU9\n1OPPOC/PqDMnoczu8qDP6YkqkWw4rknHcapSRghJcxSoU6TXGX59mhPaosVDhbJw7S3DUeSKoVVJ\n0djZnzFblyaS8T1cYX4uSnUKnGu2ZXzXMELI5BbVvcL9+/dj+/btCAQCuOuuu/DNb35zxPPPPvss\nDh8+DAAYGBiA1WrF0aNHAQBz5szBzJkzAQBGoxEvvPACn+PPWL1jZHxzjFzmNw8JZSarC0qZGIpc\nccTXlhvzUHO2C5a+AejG2MqVTtqGKpKVFEaX8T3c4opCtHY5cKqxB8tn6/geGiGE8CJioPb7/Xjm\nmWfw4osvQq/X484778T69esxY8aM0GueeOKJ0J9ffvllnDlzJvS1VCrFzp07eR525gtXPnQ4vppz\neH0BWPrcmFGcH9XrywzBQN3U2Z8Rgbo9xhk1EGzS8e7BJhyvt1CgJoSkrYi3vuvq6jB16lSUlpZC\nIpFg06ZN+OSTT8Z8/e7du7F582ZeBzkZjVU+lKOUiSGXiuLeotVlC7Z/jLQ1i8NVKMuUdep2iwMM\nE/33N9wUvQIFeTmoa7DC56cmHYSQ9BRxRm02m2EwGEJf6/V61NXVhX1te3s72trasGrVqtBjg4OD\n2LJlC0QiEb75zW/i2muvjTgotVoGkUgY8XXpTKtVjvv8gD+4BlxeqhrztSV6JRpae6EukI9ZqCSS\n+qGAO720IOKYAECulIJhjqPN6ozq9RPF5zFZlkWH1YWiQgWKjKqYjrG6sgi7Dzaiy+7BwqE64KmQ\niGs9GdB1CY+uS3iT9bpEDNThkooYhgn72t27d+P666+HUHg5yH722WfQ6/VobW3Ffffdh5kzZ2LK\nlCnjntNm47cXc7JptUpYLOPPSDuHinTA5x/ztYV5OTgfYHGmviu0Zj1R5xutAAClVBhxTBxDgQz1\nrb0wm/shEIT/WccimusyETb7IJxuL2ZPUcV83Nkl+dgNYO+RFhSppLyNbSL4vi6TBV2X8Oi6hJfp\n12W8DxkRp2kGgwEmkyn0tdlshk4Xfj3v/fffx6ZNm0Y8ptfrAQClpaVYsWLFiPXrbGazD4JBsP/0\nWEIJZXGsU4/X3nIs5cY8DHr8vBVcSZT2CVYkC2fWFBVyc4Q4Tk06CCFpKmKgrqysRFNTE1pbW+Hx\neLB7926sX7/+itddunQJ/f39WLx4ceixvr4+eDzBbUg9PT348ssvRyShZbNexyCUcsm4t7RDCWVx\nrFN3DjXjKJzAbLHcyDXoSO/CJ1zGd3GUNb7DEQkFqJymgbV/AK1dDr6GRgghvIl461skEuFnP/sZ\nHnzwQfj9ftxxxx2oqKjAf/7nf2L+/PnYsGEDgOBt75tuumnEbfGLFy/iqaeeAsMwYFkW3/jGNyhQ\nI7ic0OsYjFiAhEuQirXdJdeMQ6fOhVAQ/Rp32bCEsjWVxpjOnQzt3fHPqIFg9nfN2S7U1ndjin5y\nrnERQjJXVPuo161bh3Xr1o147Lvf/e6Ir7dt23bF+5YsWYL33nsvjuFNTu5BHzzeAFRjVCXjaFW5\nEDBMzDPqfqcH7kEf5kxVT+h9U3QKCAVM2pcSbbc4IRIy0Knj20ZWOU0DoYDBl/UW3FJVztPoCCGE\nH1SZLAVsXNesMYqdcERCAbTq2JtzRNuMYzSxSIhirRwtZkfablsKsCw6up0wxJERz5FJRZg9VY0W\nswPWvgGeRkgIIfygQJ0CXLET9Rh7qIczFsjgcHthd3kmfB5uD3Yse4zLjXnw+QNot/DTE5tv3b1u\neHwBlMRQ6CScxdSkgxCSpihQp0Ck8qHDGeJozhHrjBoYllCWpre/46lIFs6iGdSkgxCSnihQp0Ck\n8qHDXW7OMfFAHW0zjnDKDOldoYyb6RfHUOM7nII8KaYalDjf0gvXQPytRQkhhC8UqFOg1851zho/\nmQwY1u4ypkDtRJ5MDLk0cjOO0YoK5RCLBGjiaYuWa8ALr4+/9W6+Z9RA8Pa3P8Ci7pKVt2MSQhLP\n7vIgEJi8dRAoUKeAzTGBW98xNufw+vyw9g3AEGNFM5FQgCl6BdosTni8/piOwenudeMHf/gCv/mf\n43EdZ7h2iwM5YiE0+fxVE1s8VEK0tp7WqQnJFOYeF77/u4P40zsnUz2UhKFAnQK9jkEIBQyUUbSd\nVMokUOSKJ9yX2tzjBn3WBRoAACAASURBVIvYEsk4ZYY8BFgWLXEUAmFZFq98dAHuQR8O1Lajzznx\npLjRfP4AOq0uFBXKIRijnG0sSrRyqBQSnGmyIUBVygjJCJ+fMsHnZ7H7YCMa2vpSPZyEoECdAr2O\nQagUOWPWTB/NoJHB0jswoa1S8SSScS530or99veXFyyou2iFVCKEP8Dii1OmyG+KwGxzwx9geb3t\nDQRr2M8rL4DD7UWrmaqUEZLuWJbF4bNmiITBf0v/+uG5tN1SGg8K1EkWYFn0OTxQKSOvT3MMBTIE\nWBZdNnfU7+Fm4PHMqC+XEo0tocw96MOrH9dDJGTw/bsXQSwS4EBdR9w1tbka3yVxViQLZ15ZAQDg\ndFMP78cmhPCryWRHl82NJTO12LhiCtosTnx8tC3Vw+IdBeoks7u88AfYqPZQc4wxbNHiY0atL5BB\nKhGiKcYtWjurG2GzD+KmVVMxvTgfq+cb0Wl14WJHfAlq7TzU+B7LXC5QN1KgJiTdHT5jBgCsnKvH\nP26eB0WuGO9UX5p0hYsoUCdZaA/1BAL15S1a0a9Td1pdEAkZFObHXl5TwDAoMyhhsrrgHvRN6L3N\nJjs+OtoKnSoXm1ZPBQBcuyLY3vTAiY6YxwQkJuObkyeXoFSnQH1bLwbjTKIjhCROgGVx5FwXZDki\nzC/XIE8uwd3rZ8DjDeDVjy+keni8okCdZBPJ+OZMtN0l14xDr5bF3U+6zJgHFsHAG61AgMVfPzwP\nlgXuuX4mxKJgf/KFFVpo8qSoOdeFAc/EAv9w7RYH5FIR8sdpERqPeeUF8PlZ1Lf2JuT4hJD41bf2\nwmYfxNJZWohFwVB21XwDZpWqcLy+G8cvTJ7iRRSok2wi5UM5hflSCAXRN+fodXgw4PHHVOhktFgq\nlO070YHGzn6smKPD/HJN6HGBgEHVAiMGPX4cOdsV03g8Xj+6bG4UF8qjTsabKFqnJiT9Db/tzWEY\nBvdePwtCAYP//vhCXBOCdEKBOsku3/qOfjYoEgqgG2rOEU0iFh/r05zyoQpl0SaU9TkG8cbei8jN\nEeKrGyqueL6q0ggGwIG6zpjG02l1gUVi1qc5FSX5EAkFON1oS9g5CCGx8/kDOHregny5BLOnjOwO\nWFQoxw0rp6CnfxDvVjelZoA8o0CdZL0x3PoGgkHXOeCD3RW5vKWJh4xvjiZfCkWuOOotWv/zaQPc\ngz7csW562HV4Tb4Uc8sL0NDeh47uiTf8aBvK+E7E+jRHIhZiZmk+2iwO9A39vAgh6eNMUw8cbi+W\nz9aFXd7bfFUZCvOl2HOkFa1x1IFIFxSok6yXa3E5gVvfwOV63dEklIVqfBfEH8wYhkGZUYnuvoGI\nHbxON/Xg0Bkzyo1K/N2i4jFfd/UCIwCgOoZZdSiRLAFbs4bjbn+faaJZdaYIBFjUXbSioa0P/sDk\n20tLLgt323u4HLEQ91w3CwGWxV8/PJfxBYwoUCeZzT6IHIkQuTmiCb1vIqVE+bz1DQDlhuA6ddM4\nCWVenx+vfHgeDAN87frZ4yaxLa7QQi4V4fNTnRMuTtDRnbitWcPNK6d16kxyqaMfv/jrUfz69RN4\n9pVjePQ/q/G7t09iX237pNuqk878gQD21LTge785gI+OtCbkHINeP76s70ZhvhTTivLGfN2C6Ros\nm63Dxfb+uHeapNrEogWJW69jcEKJZBwu8zuaLlqdVhfy5RLIpPz8eMuM3Dp1PyqnacK+5v1DLTDb\n3Lh2WQmmDq1rj0UsEmD1PAM+PtaGuotWLJmpjXos7RYH8hXBsqqJVKJTQCkT43RTD1iWTVjiGomP\n3eXBm/su4cCJDrAAVszRQSYV49QlK46dt+DY+WDmr1Ejw7zyAswv12DWFBVyxMLUDnwSutDai1f2\nnEfbUJ2DndWNqFpgnPCkJJK6i1YMevy4dmlJxL+XWzdU4NQlK97YexGLK7TIS9BOkUSjQJ1EPn8A\ndpc3ptu20c6oB71+9PQPYNYUVUxjDIfL/B6r5aW5x4XdXzRBpZDg9qunRXXMqxcW4eNjbThwoiPq\nQO0e9MHaP4h5ZerIL46TgGEwt6wAh8+Y0dHtTPgMnkxMIMBi34kOvLXvIpwDPhRr5bhn40zMGkos\nYocq+Z1q7MGpS1aca+nFx0fb8PHRNoiEAswszcf8cg3mlxegWJu4HQTZoN/pweufNeDgUHngqxcY\nIc8V42+HW/Dpl23YtLqM1/OFbnvPCX/bezi1Mge3r52G1z6ux/982oBv3DyX17EkCwXqJIo1kQwA\nFLliKGXiiFu0zD3BrOhYu2aFo1LkQK3MCbtFi2VZvLznPHx+Fn9/7cyoPz2X6hQoMyhRd8kKm30Q\n6iiuSXuSbntz5papcfiMGaebbGkXqAMsy2tDkkxysaMPr+y5gGaTHVJJcHfB+iXFEAkvr+QxDAN9\ngQz6Ahk2LC2B1xdAQ3sfTl2y4lRjD8402XCmyYb//Sy4A2NeeQGqKo2hQE8iCwRY7Kttx5v7LsE1\n6MMUnQL3XD8LM4rz4RrwYV9tB/YcacW1y0p5u4PhGvCh7qIVxYVylOii+zu5fkkxPj9pwhenTaha\nYMScqZn3M6ZAnURcIlkst74BwFggQ317H7y+QGiD/2h8r09zygxKHK/vviKoHj5jxpkmGxZM12Dp\nrOhvYQP/f3v3HV9VfT9+/HVHbvYmuRmEEEiAECAB2UMkyFDACARrrYq1SvttrQO1FW2xWqvWjRNp\nf6V11IGAVFAcDAFjkB0IO5CQhCyy98299/z+uMllJWTd3HtJ3s/Hgwdw7zn3fu4nJ/d9Puv9sbSq\nM78+RsqhvDbddTfl+O7qiWRN4i5IJzp9VIRd3rMtKmsMPPXvXfQO8uLeOYM7tN/41aiixsDqrRnW\npX3j4vQsmBLdpomZLlo1sZH+xEb6s2CKZRnhodMlpJ8u4dDpEn44mM8PB/MZGOFH0sQoBl2FX+b2\ndDqvgve/PkZmfiXurhpuuz6GKSPC0agt30seblqmXtOb9SmZbDtwlmkjbfP7s/d4EUaTmdEtTCJr\njkat5s6ZA3nmP7t5/+tjPHX36Ba/P52VBGo76kj60AuFBHpwPKecwtKaFlt4TS1uWyzNulDfUB/2\nnTjH6bwK/L0tAbm6roGPN53ARavmF9MGtLv7cEysnk82nWB7Wh43jo1s9fyuzPHdnAAfN0IDPTiW\nXXrFmyN727Qnh5KKekoq6nnmP7u5P3mYdQ5Dd2Q2K2zdn8vabaeorjPSO8iT26cPZEBEx4d3fL1c\nmTA0lAlDQzErCidzytnwYxYHTxXzwkf7GBDhR9KEvgyK9Jdu8QtU1Taw5vsMvt9vmRMwLk7PLVOi\n8W3mO23ayN58s+sMG3eeYcrwi3s8OmrnkaZu7+B2nRcV6kPiiN5s2pvDxp1ZzJkQ1emy2JNzfPP0\nEB1JH3qhpuVWVxqn7qoWtXXLywu6v9d8f4qKmgZumtCXIL/25xT3cNNyzcBgCktrOd6GdJ1NXd9h\nvWz72a4krm8AhgYzGbnOsc9tvcHEpj05eLm7MGN0BAWltTzz3h4OnSp2dNG6REZuOX/9z24++OY4\nZkXh51NjePKXozoVpC+lVqkYEOHHQ7fE86c7RzKsfyDHs8t48eP9/P3DvRxunFDYk5kVhe0HzvL4\nilS27j9LaC9P/njbcO6dE9dskAbw9tBxXUI4pZX1pNhge9uKagNHMkuJCvUh2L/93wFzr+2Hr6eO\nL1KyKCht+wZHzkACtR11JH3ohc6vpW75IssrrsFFqybQx61D79GSviEXb3mZcbacrftyCQ30YEbj\nZhsdcW28ZU11WzKV5RZV0cvXDTed/TqCBjvZMq1taWeprjOSOCKcnyXGcO/swTQYzby66gDf7Mru\nNgGlotrAso/38bf395BVUMn4ISE8e+9Ypo2KsHavdoV+YT48uCCePy8cSXz/QI7nlPPSx/t5/sO9\n1hUAPc2Zgkqe/2AvK786SoPRzC1TovnLL0e1aTx/xug+aDUqvvwxq9Nr23cdLcSsKC2unW6Nh5uW\nn18fg9Fk5oNvjl9VP0vp+rajjqQPvVBr212e34zDvdObcVzKy92FYD93MvMqMJrMvL/xGApw54yB\nnerSGhDhR7C/O7uPFnLb9QNaXFJWUW2goqaBhGjfDr9XRwyM8EOjVnE4s4T5k/vb9b0vZTRZ1qjq\ntGqmXtMbgHFDQggOcOfN1Qf5eNMJcoqquGP6QKfppu+IY2dKeXPNwcZubi9unz7Api3otogK9eGB\nBfGczqvgix8y2X/yHC9/vJ/ocF+SJkYxuG/37xJvMJpZ/X0G3+7ORlFg5MAgbp0aQ0A7GgH+3q5M\nHBbG1n257DpSyNi4kA6XZ+eRAlTAqEHt6/a+0KhBwWxPyyP9dAm7jhYyug0zx51Bm36bt23bxowZ\nM5g2bRorVqy47Pk1a9YwduxYkpKSSEpKYtWqVdbn1q5dy/Tp05k+fTpr1661XcmvQk2TyVrqKmpN\n0+YcLbWoSyvrqW8w2XTG94X6hnpTXWfkk80nOVNYxYShIZ2eJatSqZg4NBSD0cxPjeNPzenKrS2v\nxN1VS/8wHzLzKqmqbT19a1fadbSQ4op6JsWH4e1x/mavf5gvf144ksgQb3ak5fHix/uoqL5yFjln\nlZqez8uf7KfOYOKepCE8+cuRdg/SF4oK9eH+5GE8edcohsf04mRuOS9/sp9nP9jDodPFNmmVmc0K\nNXUNlFTUkVtURUZuOYdOF7P7aCHb087y7e5svkjJZNWWk/z32+McscNe6VW1Dbz8yX6+2WXZqnbx\nz+L57dyh7QrSTW4Y0we1SsX6H7M6nCGsuLyOkznlDOzj16YVIi1RqVTcPn0AWo2aj747QU3d1bFp\nR6stapPJxNNPP83KlSvR6/UkJyeTmJhIdHT0RcfdeOONLF269KLHysrKePPNN1m9ejUqlYp58+aR\nmJiIr699W0XOoqyqHi93lw63djRqNfoAD+vmHJfe0XfV+HSTviE+/HSkkE17cvB003LLlOjWT2qD\nCUNDWbv9FNvTznLd8OZTj9p7xveF4qICOJ5TzpGs0k7dzXeGoih8lXoGtUrFjGZmoAf4uLHkFyP4\n15dH+OlIIX/9zy5+P38YffRXTj7jLBRFYf2PWazddgp3Vy33zR3CtaMiKSpq+/aqXSkyxJvfzx9G\nVn4l//vhNPtOnOOVTw7QP8yHviE+mMxmjCYFo9mMyaRgNJkxmRVMJsvjJvP5x4wmyzH1DSZqDUYM\nDe3rEt68L5efTYnm+pGtJ/zoiILSGl5blUZBSQ0jBwVzz6xYdJ1YXhXk587YOD0ph/LZd/xcu1eH\nANab+I52e19I7+/BnPGRrN1+mrXbTvGL6QM6/ZpdrdVAnZaWRmRkJBERli+HWbNmsWnTpssCdXN2\n7NjBhAkT8POz3BFPmDCB7du3M3v27E4Wu+32HCukl697q9my7KG0sp5evu2fdHWh0AAPzp6rpqLa\ncFnLPK+LZnw3aZpQBnDLlOiLWnWd4e/tytB+gaRlFJNTWNXs+sjzE8nsH6gHRwWwdvtp0k+XOCxQ\nHzxVQk5RFWMH6+nVwsQ9nYuGX98UR3iQF2u3neLZD/Zw7+y4Dn0x2pPRZOa9r4+xIy2PQB9XHlwQ\n73Tr1ps0BewzBZX874dM9h4vIuNs6xvWaNQqyx+NuvFvFW46DX7errjrNLjptLi5Wv62/F+Dm6sW\nN50G98bn3HVaKmsaeO+bY3y06QSn8ytYOHOQTbOsncgp443VB6mqbeDGsZHMm9zPJuv1Z42L5MdD\n+az/MZMRA3q1+wZj5+ECNGoV1wy0ze/fzDGR/JhewOa9OYwdoqd/mHM3HlsN1AUFBYSEnB9X0Ov1\npKWlXXbcN998w65du4iKimLJkiWEhoY2e25BQcvdm038/T3Qam1z8b33+g483LSsWHK9zcdtryQo\n6OIbg5q6BuoMJvSBHpc91x79IvzYc7yIWhNEX/I65Y07a8X279Wp92iJl487nm4HiQr35ebEAR2q\nz5bKNXtSP9Iyitl94hzD40Ive76wrA61WsXQgfpO3d13RECgF57uaRzNLqNXL68uacW09vPatMry\nO3fbDbGtHnt30lBi+wXyyn/38tbag/xi5iB+dn37l8/ZQ3VtA8+/t4v9x4uI7u3L0l+Nxf+C7tWu\nuI5tISjIm2uGhFFYWkN1bQNajRoXrRqNWo1Wq0KrUTf+UaFRq2363TMiLoTn/rOL1PQCCkvrWHLX\nKJsMd23bl8OrH+3HrCjctyCBGWMjbVBai6Agb8YPC+OHtLPklNQxoh03vNkFlZwprGJkrJ6oPgGt\nvk9b/f5nCTzxTgpvrD7IM78ZT5QTB+tWA3VzYzCX/sJPmTKF2bNno9Pp+Oijj/jjH//Ie++916Zz\nm1Nqw6nzCdGBbE/LY8tPWQzr33yealsLCvK+rMuuqVvaQ6fpVHeeT+Nkq6OnzhHie3GL+nSuZYmT\nq4ou6zL86z1j8HDVUlzc/q3jmquXJn2DPPHxcGHz7mxmjelz0fCAoihk5pWj93envMwxyyoGNd4g\npZ8oRN+BpSFXcqV6AcuGEwczzjEkKgAvF3WbfrbRId4suf0aXv8sjQ83HuV4Zgl3z4p1qhzXJRV1\nvLbqADlF1SRE9+LXN8VhrG+gqMhyw9lavTgDFeDl0nitms1gNmM0QleOfAYFebN4QTwfbTrB1n25\nPPjKVn59UxxDWsjD35qLhx00/PbmYcRFBdi87q8fEc4PaWf58KvDRAS2vWdx4w+nABgeHXjFMrX3\negn1dWPhzIG8t/EYS97aweKfJVjTJTvClW4yWh0sDQkJIT///Bq4goICgoMvvhvy9/dHp7N0g95y\nyy2kp6e3+dyuNmWEZcxz675cu77vpUo7meykyZWWaOWV1ODnpbN5EvwL+Xm5dkmLVqtRM35IKFW1\nDew/ee6i50or66mtNzlkfLpJ0zKtw3aYyHOpr1KzALihnS2ciGAv/rxwJAN6+7LraCHPf7CXkoqW\nd5JqMJoorawnu7CKI1ml7DpayJZ9uXyRksmnm0+y83ABDUbbbB+ZlV/JX9/bTU5RNYkjwrlv3lBc\ndc5zE+HsXLRq7pwxkF/eMIj6BhOvfnqADT9mtntym9Fk5l9fHmHttlME+rjy+O3XWHeOs7XIEG/L\nGvWc8jblTQDLTcTOI4XotGqGx/SyeZkmJ4Rz96xYauqNvPTxPk46Sb6ES7X6jT506FAyMzPJzs5G\nr9ezYcMGXn755YuOKSwstAbgzZs307+/ZRnLxIkTeeWVVygvt3z4HTt2sHjxYlt/hivqG+JDVKgP\nB06e41x5bafHiDvKuoa6EzMWwTJGDZcv0ao3mCipqL8q89g2mTgslI0/nWH7gbMXjQXbO8d3c5q+\nvA6dLmHKiN52e9/8khr2Hi+ib4g3gzqw0YqPp45Hfj6cD745xrYDeTz9n91cMzCI6toGKmsaqK5t\noKqugarahjZNavL2cGHSsDCuTQgjuANJbsCy+9E76w5hMJi4NTGaaaMinLJb/mowKT6M3sFevLnm\nIKu/P8XpvEp+NSu2TTfr1XUNvLXmIEfPlNE3xJsHkod1eEVKW80e15e0jGLWp2Sy+GcJrR5/pqCK\ngpIaRg0K7rL8CROGhqLVqPnHF4d5+ZP9PJg8zOlyvrf6ybVaLUuXLuWee+7BZDIxf/58YmJiWLZs\nGUOGDGHq1Km8//77bN68GY1Gg6+vL8899xwAfn5+/Pa3vyU5ORmA3/3ud9aJZfY0ZXg4/8qr4Pv9\nZx22Fta6IUcnfxE83Fzw8dSRV1x90eNdPePbHsJ6edI/3If00yUUl9cR6GsZq7SmDnVgizrYz50g\nPzeOninFZDZ3adKNC23ceQYF2pRitSVajZqFMwcRHuTFx5tOsGXv+d4lV50GLzcXQgM88XLX4unu\nglfjnwv/7eqiYd+JInak5fFlahZfpWYR1y+AKQnhDIsObHN9bN2XywffHEejUfHbuUNsNjmoJ4sK\n9eHJu0axfN0h9h4vIq+4mvvmDb1iWtnCslqWrTpAXnENIwYEce+cwXYZFonu7cugPn4cOl3C6byK\nVruarTtl2WC295WMGaxHq1GxfF06r646wP3zhzG4b9f0LHREm25RJk+ezOTJky967IEHHrD+++GH\nH+bhhx9u9tzk5GRroHaU0bHBfLL5BNsPnCVpYpRNcs62l7Xr27vzM6VDAzw4nl1Gg9GES+OkO2ug\n7qIZ3/YyaVgYGbkV/HAwj5smWvLxWpdm2XkN9aXi+gawdf9ZTudVEh3e9RNPyqrqSTmUR7C/e7v2\n7G6OSqVi2sgIrhkQRE290RKI3dq3VHBAhB/zru3H7qNFbNmXy6FTJRw6VYK/tyvXxodxbXxYiz1G\nZkVh9dYMvtp5Bi93Fx5IHkZ/O9RhT+HjqePhWxNYtSWDb3Zl89f/7Oae2YObvW4ycst5fXUalTUN\nzBgdwYLrou060XbW+L4cPbOfDT9mcd+8oS0eZ1YUdh4pwN1Vy9AOjr+3xzUDg/ndPDVvrz3Iss/S\nuG/eULu8b1tcvemL2kHnomHC0FAqahqsG8nbW2d3zrpQSKAHClBQUmt9rKmF3VVLs+xl1KBgXF00\n7DiYZ02OkHOuGq1GTbC/Y4YtmjTdYdtrnPq73TkYTQozR/ex2RdpgI8bvYO88PNy7dB6fhethnFD\nQnj8jmt46u7RTBkRTm29kXU7TvPo2ym8ueYgh04XX5TYosFoYvm6dL7aeQZ9gAd/uvMaCdJdQKNW\nc+vUGBbdNBizWeHNNQdZsy0Ds/n8z2L30UJe+GgfVbUN3DF9AD9LjLFrkAYYHOlPVKgPe48XWW/C\nm3Myp5zSynquGRBkt0x7CdG9uD95GABvrE5j3wnHxItL9YhADZbub4Ate3Mc8v5lVfWoVSqbrD1u\nbpy6O3R9gyUT2KjYYM6V13E0qxSzWSHvXDVhgR52625uSWxff1Qq++T9rq03smVfLj4eLkwY2vG0\ni10pItiLO6YP5JX7JrBw5kB6B3uy93gRr3xygCXv/shXqVnkFVfz4kf72X20kJjevjxxxzUd2lBB\ntN3YwSE8cedIgvzcWJ+SxWufHaC6roGvUrN4+/NDqNUqHkiOt+tciwupVCpmj7dMjNzQOFGyOfbq\n9r7UkKhAHlwQj1qt4u21h9h9tNCu79+cHhOo9QEexPX153hOOTlXuIvrKmWV9fh66Wxy93p+5vf5\ncer84hp0WnWHUvw5m2uHhQGWjTqKymsxGM0O7/YG8HRzISrUh4zcCmrruzb14Pf7z1Jbb+T6kRHW\n4Q1n5abTMjkhnCfvGsWf7hzJxKGhlFcZWLU1gyf+sZOTueWMjg3mkVsT8HLvGXtnO5plxv8ohvQL\n4NCpEv7wzo+s2pqBv7crS34xwm5LVVsSH92L3kGe7DxcQGEzy3GNJjO7jhbi4+HCoEj7z2uKjfRn\n8S0JuGjVLF+XTmp653f/6oweE6gBrhtuuYO091ItRVEoq6rv9ESyJk3JDZpa0WZFIb+0Bn2Ah02y\nCDla/3AfQgM92HOsyLqMw1kyVQ3uG4BZUTh6prTL3qPBaOabXWdw1WmsywuvBiqVin5hPtw9K5aX\n75vAz6+PISrUm5sm9GXRTXFOf8PR3Xi5u/Bgcjyzx0dSW2+kj96LP9050inSyqpVKmaP74uiwJep\nZy57/khWKVW1DYwapHdYT9qACD8evjUBV52Gf3xxmB1t2OGvq/SoQJ0QE4i/tysph/KpM9gvGXtV\nbQNGk9LhXbMu1cvHDa1GbV1LXVpRj6HBfNWPTzdRqVRMGhaG0WTmfzsyAcfO+L5QXF/Lso3Dp7su\nUKem51NWZeC6hDA83a7OFqinmwvTRkbw54WjuHmSbdJQivZTq1XMu7Y/z/9mHE/cMbLTy0NtaeTA\nYPQBHvxwMO+y9f2O6va+VP8wX/7w8+F4uGn515dH2LrfMfk4elSg1qjVTI4Po85gIjW99VSmtmKd\nSGajXxK1WoU+wN26OUd3GZ++0LghIWjUKoobf4GdoesboH+4L64umi4bpzYrCht/OoNGbZmlLYQt\nBPu5O93Wp2q1ihvH9sFkVti483yr2tBgYu/xIgJ93Ogf7rhMYU0iQ7z5w20j8PZw4b2Nx/h2d7bd\ny+BcPzk7mBQfhkatYvPeXLttHG6rNdQXCg3woM5goqzKYB2rvtqXZl3I11NnHUdz1WkIdJKxd61G\nzaA+fuSX1FBc3nKWr446cOIcecU1jI3Td4v5BkJcybi4EAJ9XNl24Kx1a9a0jGLqDCZGDw52mkQ4\nEcFe/PG2Efh66fjouxN8tbPlSXBdoccFan9vV4bH9CKnqIqM3NZ3vbEFW6UPvVBTUM4vriavsUUd\nGuAcrU5bmRRvmVTWu5en0/zCwvl0orZuVSuKwpeNXwAzx9huQwQhnJVWo2bmmEgMRjPf7LK0VHc2\nbWkZ69hu70uF9fLksdtG4O/tyqotGdbueXvocYEazi/V2rzPPku1bJU+9EJNQTm/pIb8xrFqfYBj\n1xnb2tB+AUwYEsLUaxyzjKQlcU3rqW0cqE/klJORW0FCdC+nGZMXoqtNGhaKj6eOzXtzKC6v48DJ\nYkIDPYhoZrtbR9MHePBY46x5H0/bzDlqix4ZqAdF+hMS4MHuo4VU1Bi6/P2axqhtNZkMLt6cI7+k\nBn9v1y7LhesoGrWaX80ezNg451pHHBrogb+3K4czSy9K7NFZ5zff6GOz1xTC2elcNMwYHUGdwcTr\nq9MwmsyMGax3ql60CwX5ufPggni77qvQIwO1SqViyvBwjCaFH+ww5b7Mmj7Uhl3fjRPHMvMrKa2s\n7zYzvq8GKpWKwX39qaptILvANmvyc4qqOJBRTHS4LzG97b9uVAhHui4hHE83LdmFlt8nR8/2djY9\nMlADTBgagk6rZsu+XJu2ippTWlWPTqvGw4bbT7q7avH10pHRuC1bd5rxfTWIs/E49deNs16lNS16\nIndXLdc3rnLo4mPd5QAAF3FJREFUG+Jt8z3fr3Y9NlB7uLkwZrCec+V1HDrVtSkhm5Kd2LorJzTA\nkvMbuOJOOcL2Bkc2Bmob5P0uKq0l9XABoYEexEfbfs9dIa4G00b2Zmi/QOZM6OvoojidHhuoARIb\nc912Zf5vk9lMRbXBpuPTTUIuCM7SorYvH08dfYK9OJFTRn2DqVOv9b/tGZjMCjPH9JHEIKLH8nBz\n4aFb4hke07md4rqjHh2oI0O8iQr1IS2jmHNlta2f0AEV1Q0oim3Hp5uEXhCcZYza/gZHBWA0KZxo\nTHPaEdV1DXydmomfl45xTjZpTgjhHHp0oAZIHBGOAnx/4GyXvH5XJDtp0hScXV00XXIjIK6saZlW\nZ8apN+/NpbbexPRRfRyyT7oQwvn1+G+GUYOC8XTTsv3AWRqMZpu/flOyk67IsdvU3a0PcJcuUweI\n6e2Li1ZNegfyfhtNZn44mMc3P53B003L5ISwLiihEKI76F4LbztA56JhwtBQvtmVzZ7jhYwdbNvu\nx65sUQf4ujFmsJ4BEbKcxxF0LhoG9PYlPbOU8qp6fNvwM66tN7L9wFm+3pVNaaVlj/JFc4fibsMV\nAUKI7kW+HbBkKvtmVzZb9+baPFCfTx9q+8lkapWKX98UZ/PXFW03OCqA9MxSDmeWMm5Iy9dOebWB\nTXuy2bwnl5p6IzoXNdeP7M30URHERgdTVFRpx1ILIa4mEqixpIWL6+tPemYpOUVV9Lbh3sddkT5U\nOI+4vgGsIoP0zJJmA3VBaQ1f7zzDjoP5GE1mvD1cmDspiikjeuPlfnVuYSmEsC8J1I2mjOhNemYp\nW/blcsf0gTZ73ab0oW3pFhVXn97BXnh7uJCeWYKiKNa18qfzKvgqNYs9x4pQgCA/N2aO7sOEoaHo\nXDSOLbQQ4qoigbpRfHQg/t6u/Hgon+TJ/W02ZlhWWY+HqxZX+XLultQqFXF9A0g9XEDuuWpKK+v5\nKjWLo2csS7Yi9d7cMLYP1wwMQqPu8XM3hRAdIIG6kUatZnJCGJ9vP03q4QLrDludVVZVL93e3dzg\nxkD99w/3Ul1nBCwpRm8Y04fYSH+n3VxACHF1kEB9gWvjw/jih0y27M3luoSwTn/BGhpMVNcZ6Rvi\nbaMSCmcUFxWARq2ipt7ImMF6Zo7uQ6T8zIUQNiKB+gJ+Xq4Mj+nF7mNFnMwt7/QuRl25NEs4D39v\nV/68cCQerlp6+XWvPcGFEI7XpkGzbdu2MWPGDKZNm8aKFSsue37lypXceOONzJkzh4ULF5Kbm2t9\nLjY2lqSkJJKSkvjNb35ju5J3kSlN+b/35bZyZOus+1BL13e310fvLUFaCNElWm1Rm0wmnn76aVau\nXIleryc5OZnExESio6Otx8TGxrJ69Wrc3d3573//y4svvshrr70GgJubG+vWreu6T2Bjg/r4ERro\nwe6jhdw6NQYfj46vf5YWtRBCiM5qtUWdlpZGZGQkERER6HQ6Zs2axaZNmy46ZuzYsbi7W1oTCQkJ\n5Ofnd01p7UClUnHd8HCMJoUdaXmdeq3zyU4kUAshhOiYVlvUBQUFhIScT+Sg1+tJS0tr8fjPPvuM\na6+91vr/+vp65s2bh1arZdGiRVx//fWtFsrf3wOt1nHLmW66LoY1206xPS2P22fFoVG3f1JZUJA3\n9SbLbtH9+vgTFCSTiwCphxZIvTRP6qV5Ui/N66710mqgVhTlssdamg29bt06Dh06xAcffGB9bMuW\nLej1erKzs1m4cCEDBgygT58+V3zP0tKa1orV5cbEBrPtQB5bf8pkWP9e7To3KMiboqJKzhZa0kIq\nDUZJEcn5ehEXk3ppntRL86Remne118uVbjJa7foOCQm5qCu7oKCA4ODgy45LSUlh+fLlvPPOO+h0\n58d19Xo9ABEREYwePZrDhw+3q/COMmV446SyvR2fVFZWZUAF+HZBnm8hhBA9Q6uBeujQoWRmZpKd\nnY3BYGDDhg0kJiZedMzhw4dZunQp77zzDoGBgdbHy8vLMRgsM59LSkrYu3fvRZPQnFlkiDdRoT6k\nZRRzrqy2Q69RVlWPj6dOMlIJIYTosFa7vrVaLUuXLuWee+7BZDIxf/58YmJiWLZsGUOGDGHq1Km8\n8MIL1NTU8MADDwAQGhrK8uXLycjI4Mknn0SlUqEoCvfee+9VE6gBEkeE8/82VPD9gbPMn9y/Xecq\nikJZZT2hgZ5dVDohhBA9gUppbhDawZxlnMHQYOLht35ArVbx0m8n4KJtW8s4KMibrOwS7nttOwnR\nvbg/eVgXl/TqcLWPIXUVqZfmSb00T+qleVd7vXRqjLon07lomDgslMqaBvYcL2zXuaVNyU5kfFoI\nIUQnSKBuxXUJls05trZzUlmZrKEWQghhAxKoW6EP8CAuKoDjOeXkFFa1+TxrVjJJHyqEEKITJFC3\nQdOWl1v2t71VLelDhRBC2IIE6jaIjw7E39uVlEP51NYb23TO+fShMkYthBCi4yRQt4FGrWZyQhj1\nBhOphwvadE7Tzln+0vUthBCiEyRQt9G18WFo1Cq27M1pNq3qpcqq6tFqVHi5u9ihdEIIIborCdRt\n5OflyvABQeQUVXMyt7zV40sr6/Hzcm0xL7oQQgjRFhKo28E6qayVpVpms0J5lUEmkgkhhOg0CdTt\nMKiPH6GBHuw+VkhFjaHF48qr6jErikwkE0II0WkSqNtBpVJx3fBwjCaFHWl5LR5XXFEHyBpqIYQQ\nnSeBup0mDAlB56Jm675czObmJ5WVNAZqf+n6FkII0UkSqNvJw82FsYP1nCuv49Dp4maPKS5vbFFL\noBZCCNFJEqg7YMrw3gBsbmFSWUm5dH0LIYSwDQnUHRAZ4k2/MB8OZhRzrqz2suebur5lMpkQQojO\nkkDdQVOGh6MAW/efvey54nJL8JaubyGEEJ0lgbqDRg0KxtNNy/a0szQYzRc9V1JRh5tOg7ur1kGl\nE0II0V1IoO4gnYuGicNCqaxpYM+xwoueK6mok9a0EEIIm5BA3QnXJTRmKtt3flJZg9HcmJVMxqeF\nEEJ0ngTqTtAHeBAXFcCJnHJyCqsAKK+2bG8pu2YJIYSwBQnUnWTN/93Yqm7a3lK6voUQQtiCBOpO\nio8OxN/blZT0fGrrjZRVWlrUsoZaCCGELUig7iSNWs3khDDqDSZS0/MprWrs+pYWtRBCCBuQQG0D\n18aHoVGr2LIv93yLWgK1EEIIG5CFvjbg5+XK8AFB7D5aiKFxTbWft8z6FkII0XltalFv27aNGTNm\nMG3aNFasWHHZ8waDgQcffJBp06axYMECcnJyrM+9++67TJs2jRkzZrB9+3bbldzJJDZOKisstWQl\n8/WUFrUQQojOazVQm0wmnn76af75z3+yYcMG1q9fz8mTJy86ZtWqVfj4+PDtt99y11138dJLLwFw\n8uRJNmzYwIYNG/jnP//JU089hclk6ppP4mAD+/gRGugBgI+nDhetjCoIIYTovFajSVpaGpGRkURE\nRKDT6Zg1axabNm266JjNmzczd+5cAGbMmMGPP/6Ioihs2rSJWbNmodPpiIiIIDIykrS0tK75JA6m\nUqm4rrFVHejr5uDSCCGE6C5aHaMuKCggJCTE+n+9Xn9ZsC0oKCA0NNTyglot3t7elJaWUlBQQHx8\n/EXnFhQUtFoof38PtFpNmz+Es7jpuhi+Sj1Dv3BfgoK8HV0cpyT10jypl+ZJvTRP6qV53bVeWg3U\niqJc9phKpWrTMW05tzmlpTWtHuOsnrlnDOFhvhQVVTq6KE4nKMhb6qUZUi/Nk3ppntRL8672ernS\nTUarXd8hISHk5+db/19QUEBwcPBlx+Tl5QFgNBqprKzEz8+vTed2Nx5uWrQaGZ8WQghhG61GlKFD\nh5KZmUl2djYGg4ENGzaQmJh40TGJiYmsXbsWgK+//pqxY8eiUqlITExkw4YNGAwGsrOzyczMZNiw\nYV3zSYQQQohuqNWub61Wy9KlS7nnnnswmUzMnz+fmJgYli1bxpAhQ5g6dSrJyck8+uijTJs2DV9f\nX1599VUAYmJiuOGGG7jxxhvRaDQsXboUjebqG3sWQgghHEWlNDeQ7GBX8zgDXP1jJV1F6qV5Ui/N\nk3ppntRL8672eunUGLUQQgghHEcCtRBCCOHEJFALIYQQTkwCtRBCCOHEJFALIYQQTkwCtRBCCOHE\nJFALIYQQTkwCtRBCCOHEnDLhiRBCCCEspEUthBBCODEJ1EIIIYQTk0AthBBCODEJ1EIIIYQTk0At\nhBBCODEJ1EIIIYQT0zq6AFezvLw8/vCHP3Du3DnUajW33HILCxcupKysjIceeojc3FzCw8N57bXX\n8PX1dXRx7aalennjjTf49NNPCQgIAGDx4sVMnjzZwaW1n/r6en7xi19gMBgwmUzMmDGD+++/n+zs\nbBYvXkx5eTmDBw/mhRdeQKfTObq4dtNSvTz22GP89NNPeHtb9ul9/vnniY2NdXBp7c9kMjF//nz0\nej3vvvtuj79emlxaL935epF11J1QWFhIUVERcXFxVFVVMX/+fN566y3WrFmDn58fixYtYsWKFZSX\nl/Poo486urh201K9fPXVV3h4ePCrX/3K0UV0CEVRqKmpwdPTk4aGBm677TaeeOIJVq5cyfTp05k1\naxZLly5l0KBB3HbbbY4urt20VC8ff/wx1113HTNnznR0ER1q5cqVHDp0iKqqKt59910eeOCBHn29\nNLm0Xh577LFue71I13cnBAcHExcXB4CXlxf9+vWjoKCATZs2cfPNNwNw880389133zmymHbXUr30\ndCqVCk9PTwCMRiNGoxGVSkVqaiozZswAYO7cuWzatMmRxbS7lupFQH5+Plu3biU5ORmw3NT09OsF\nLq+X7k4CtY3k5ORw5MgR4uPjKS4uJjg4GLAErZKSEgeXznEurBeADz/8kDlz5rBkyRLKy8sdXDr7\nM5lMJCUlMX78eMaPH09ERAQ+Pj5otZZRqJCQkB55U3NpvTRdL6+++ipz5szh2WefxWAwOLiU9vfs\ns8/y6KOPolZbvqpLS0vleuHyemnSXa8XCdQ2UF1dzf3338/jjz+Ol5eXo4vjNC6tl5///Od8++23\nrFu3juDgYJ5//nlHF9HuNBoN69at4/vvvyctLY1Tp05ddkxPbE1eWi/Hjx9n8eLFbNy4kdWrV1Ne\nXs6KFSscXUy72rJlCwEBAQwZMuSKx/W066WleunO14sE6k5qaGjg/vvvZ86cOUyfPh2AwMBACgsL\nAct4bdPkqZ6kuXrp1asXGo0GtVrNggULOHjwoINL6Tg+Pj6MGTOG/fv3U1FRgdFoBCxdek29MT1R\nU71s376d4OBgVCoVOp2OefPm9bjrZe/evWzevJnExEQWL15Mamoqf/vb33r89dJcvTzyyCPd+nqR\nQN0JiqLwxBNP0K9fP375y19aH09MTOTzzz8H4PPPP2fq1KmOKqJDtFQvTTcvAN999x0xMTGOKJ7D\nlJSUUFFRAUBdXR0pKSn079+fMWPG8PXXXwOwdu1aEhMTHVlMu2uuXvr162e9XhRF6ZHXy8MPP8y2\nbdvYvHkzr7zyCmPHjuXll1/u8ddLc/Xy0ksvdevrRZZndcKePXtYt24dAwYMICkpCbB0vyxatIgH\nH3yQzz77jNDQUJYtW+bgktpXS/Wyfv16jh49CkB4eDhPP/20I4tpd4WFhTz22GOYTCYURWHmzJlM\nmTKF6OhoHnroIV577TViY2NZsGCBo4tqVy3Vy5133klpaSmKojBo0CCeeuopRxfVKTz66KM9+npp\nySOPPNJtrxdZniWEEEI4Men6FkIIIZyYBGohhBDCiUmgFkIIIZyYBGohhBDCiUmgFkIIIZyYBGoh\nhBDCiUmgFsLJHTlyhC+//LJNx+bk5PDJJ59c9Ni9997LmTNnuqJoXeaOO+5gy5Ytji6GEE5BArUQ\nTu7IkSNs3LixTcfm5uZeFqj/8Y9/0KdPn64omhDCDiRQC9FFvv76a2bOnMnNN9/M8uXLGThwINXV\n1Rw4cIA77riDefPmMW/ePLZu3QpAcXExd911F3PmzLHuAFRaWsrrr79OSkoKSUlJPPPMM4AljeK8\nefOYM2cOv/vd76w7kT399NNkZGSQlJTE/fffD1hS2h4/fhyArKwsFi5cyJw5c5g7dy7btm2zlnfg\nwIEsX76c+fPnM3XqVGuaypZMmjSJ4uJiwNJqX7RokfVzTJo0CQCDwcDf//53kpOTSUpK4tFHH6W6\nuhqAqqoqnnjiCZKTk5kzZw7PPPMMJpPpsvfZsGED8+fPJz8/n7179zJ37lySkpKYNWsW69ev79DP\nRoiriiKEsLlz584po0ePVk6fPq0oiqKsXLlSGTBggJKXl6ckJSUpBQUFiqIoSkFBgTJp0iSlvLxc\nWblypbJkyRLra5SVlSmKoiirV69Wfv/731/0+sXFxdZ/v/LKK8qLL76oKIqipKamKnPnzr3o2ClT\npijHjh1TFEVRkpOTlU8//VRRFEU5ceKEMnr0aOtrDRgwQHn//fcVRVGU3bt3KxMnTrziZ3z44YeV\n9evXKwaDQbnhhhuUmTNnKgaDQfniiy+URx55RFEURXnrrbeUt956y3rOCy+8oLzyyiuKoijK448/\nrqxdu1ZRFEUxmUzKQw89pHzyySeKoijK7bffrmzevFlZsWKFsnDhQqWiokJRFEX5zW9+Yz3HbDYr\n5eXlVyyjEN2B5PoWogvs37+fwYMH07dvXwDmz5/Pc889R3p6Ojk5Odx7773WY1UqFVlZWcTHx7Ny\n5Ur+/ve/M3r0aCZOnNji669bt44vvviChoYGampqrO9zJVVVVRw5coT58+cDEB0dTWxsLPv377du\n7HDjjTcCkJCQQGFhIfX19bi6ujb7emPHjiUlJQW9Xk98fDyKonDgwAFSUlIYN24cAJs3b6aqqsra\nOjcYDAwaNMj6XFpaGitXrgQsG3Lo9Xrr67/xxhuEhYWxYsUKdDodAGPGjGHFihWcPXuWCRMmWPet\nFqI7k0AtRBdQFKXZfYIVRWHgwIF8+OGHzZ73+eefk5KSwrp161ixYgUfffTRZcfs3r2bjz76iI8/\n/piAgAC++OILPv300w6X9cJyNgVljUYDgNFobDFQjxs3jrfffpuQkBDGjh2LoiikpqaSmprKfffd\nZ/28Tz75pDVwX0hRFN5++20iIiKaff2EhAR++OEHzp49a70Rueuuu0hMTCQlJYW//vWvTJgwgYce\neqjDn12Iq4GMUQvRBRISEkhPTycrKwuANWvWABAXF0dWVhapqanWY9PS0lAUhezsbLy8vJg1axZL\nliwhPT0ds9mMl5cXlZWV1uMrKirw8vLCz88Pg8HA6tWrrc95eXlRVVXVbJm8vLyIjY1l7dq1AGRk\nZHD06NEOt0rDw8PRaDSsXbuWcePGMX78eNasWYNWqyUsLAywjI//+9//pq6uDrC06jMyMqzPrVix\nwjouXVJSQnZ2tvX1J02axF/+8hcWLVrEiRMnADh9+jR9+vTh1ltv5c477+xWew4L0RJpUQvRBXr1\n6mUNMv7+/iQmJuLi4oJer+ftt9/mxRdf5Nlnn6WhoYGIiAiWL1/OTz/9xMqVK9FoNJjNZp566inU\najXjxo3jX//6FzfddBOjR4/mscce43//+x833HADer2eIUOGWAPWwIEDiYqKYvbs2fTr14/XX3/9\nonK99NJLLF26lH//+99otVpeeOEFAgICOvw5x40bx549ewgODgbAzc2NkSNHWp9ftGgRb775JsnJ\nyahUKlQqFffddx/9+/fn8ccf58UXXyQpKQmVSoWLiwuPP/74RS3scePG8dxzz/F///d/LFu2jNWr\nV7Nz505cXFzQ6XT86U9/6nDZhbhayDaXQnSRqqoqvLy8AFi9ejWfffZZs13ZQghxJdKiFqKLvP/+\n+2zcuBGTyYSvr691aZUQQrSHtKiFEC0qLi7m7rvvvuzxadOmWSeMCSG6lgRqIYQQwonJrG8hhBDC\niUmgFkIIIZyYBGohhBDCiUmgFkIIIZyYBGohhBDCif1/zxWup6DApAoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f5538c94510>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gestation_weeks_vs_other = natality.groupby([\"gestation_weeks\"]).mean()[[\"is_male\",\"plurality\"]]\n",
    "gestation_weeks_vs_other.plot.line()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>source_year</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>wday</th>\n",
       "      <th>state</th>\n",
       "      <th>is_male</th>\n",
       "      <th>child_race</th>\n",
       "      <th>weight_pounds</th>\n",
       "      <th>plurality</th>\n",
       "      <th>apgar_1min</th>\n",
       "      <th>apgar_5min</th>\n",
       "      <th>mother_residence_state</th>\n",
       "      <th>mother_race</th>\n",
       "      <th>mother_age</th>\n",
       "      <th>gestation_weeks</th>\n",
       "      <th>lmp</th>\n",
       "      <th>mother_married</th>\n",
       "      <th>mother_birth_state</th>\n",
       "      <th>cigarette_use</th>\n",
       "      <th>cigarettes_per_day</th>\n",
       "      <th>alcohol_use</th>\n",
       "      <th>drinks_per_week</th>\n",
       "      <th>weight_gain_pounds</th>\n",
       "      <th>born_alive_alive</th>\n",
       "      <th>born_alive_dead</th>\n",
       "      <th>born_dead</th>\n",
       "      <th>ever_born</th>\n",
       "      <th>father_race</th>\n",
       "      <th>father_age</th>\n",
       "      <th>record_weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>05992006</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2006.0</td>\n",
       "      <td>2006.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.906100</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>26.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.043649</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.707107</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2006.0</td>\n",
       "      <td>2006.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.875235</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>26.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2006.0</td>\n",
       "      <td>2006.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.890668</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>26.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.250000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2006.0</td>\n",
       "      <td>2006.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.906100</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>26.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2006.0</td>\n",
       "      <td>2006.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.921532</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>26.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.750000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2006.0</td>\n",
       "      <td>2006.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.936965</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>26.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        source_year    year  month  day  wday  state is_male  child_race  \\\n",
       "count           2.0     2.0    2.0  0.0   2.0    0.0       2         0.0   \n",
       "unique          NaN     NaN    NaN  0.0   NaN    0.0       1         NaN   \n",
       "top             NaN     NaN    NaN  NaN   NaN    NaN   False         NaN   \n",
       "freq            NaN     NaN    NaN  NaN   NaN    NaN       2         NaN   \n",
       "mean         2006.0  2006.0    9.0  NaN   7.0    NaN     NaN         NaN   \n",
       "std             0.0     0.0    0.0  NaN   0.0    NaN     NaN         NaN   \n",
       "min          2006.0  2006.0    9.0  NaN   7.0    NaN     NaN         NaN   \n",
       "25%          2006.0  2006.0    9.0  NaN   7.0    NaN     NaN         NaN   \n",
       "50%          2006.0  2006.0    9.0  NaN   7.0    NaN     NaN         NaN   \n",
       "75%          2006.0  2006.0    9.0  NaN   7.0    NaN     NaN         NaN   \n",
       "max          2006.0  2006.0    9.0  NaN   7.0    NaN     NaN         NaN   \n",
       "\n",
       "        weight_pounds  plurality  apgar_1min  apgar_5min  \\\n",
       "count        2.000000        2.0         0.0         2.0   \n",
       "unique            NaN        NaN         NaN         NaN   \n",
       "top               NaN        NaN         NaN         NaN   \n",
       "freq              NaN        NaN         NaN         NaN   \n",
       "mean         0.906100        2.0         NaN         0.0   \n",
       "std          0.043649        0.0         NaN         0.0   \n",
       "min          0.875235        2.0         NaN         0.0   \n",
       "25%          0.890668        2.0         NaN         0.0   \n",
       "50%          0.906100        2.0         NaN         0.0   \n",
       "75%          0.921532        2.0         NaN         0.0   \n",
       "max          0.936965        2.0         NaN         0.0   \n",
       "\n",
       "        mother_residence_state  mother_race  mother_age  gestation_weeks  \\\n",
       "count                      0.0          2.0         2.0              2.0   \n",
       "unique                     0.0          NaN         NaN              NaN   \n",
       "top                        NaN          NaN         NaN              NaN   \n",
       "freq                       NaN          NaN         NaN              NaN   \n",
       "mean                       NaN          2.0        17.0             20.0   \n",
       "std                        NaN          0.0         0.0              0.0   \n",
       "min                        NaN          2.0        17.0             20.0   \n",
       "25%                        NaN          2.0        17.0             20.0   \n",
       "50%                        NaN          2.0        17.0             20.0   \n",
       "75%                        NaN          2.0        17.0             20.0   \n",
       "max                        NaN          2.0        17.0             20.0   \n",
       "\n",
       "             lmp mother_married  mother_birth_state cigarette_use  \\\n",
       "count          2              2                 0.0             2   \n",
       "unique         1              1                 0.0             1   \n",
       "top     05992006          False                 NaN         False   \n",
       "freq           2              2                 NaN             2   \n",
       "mean         NaN            NaN                 NaN           NaN   \n",
       "std          NaN            NaN                 NaN           NaN   \n",
       "min          NaN            NaN                 NaN           NaN   \n",
       "25%          NaN            NaN                 NaN           NaN   \n",
       "50%          NaN            NaN                 NaN           NaN   \n",
       "75%          NaN            NaN                 NaN           NaN   \n",
       "max          NaN            NaN                 NaN           NaN   \n",
       "\n",
       "        cigarettes_per_day alcohol_use  drinks_per_week  weight_gain_pounds  \\\n",
       "count                  0.0           2              0.0                 2.0   \n",
       "unique                 NaN           1              NaN                 NaN   \n",
       "top                    NaN       False              NaN                 NaN   \n",
       "freq                   NaN           2              NaN                 NaN   \n",
       "mean                   NaN         NaN              NaN                26.0   \n",
       "std                    NaN         NaN              NaN                 0.0   \n",
       "min                    NaN         NaN              NaN                26.0   \n",
       "25%                    NaN         NaN              NaN                26.0   \n",
       "50%                    NaN         NaN              NaN                26.0   \n",
       "75%                    NaN         NaN              NaN                26.0   \n",
       "max                    NaN         NaN              NaN                26.0   \n",
       "\n",
       "        born_alive_alive  born_alive_dead  born_dead  ever_born  father_race  \\\n",
       "count                0.0              0.0        0.0   2.000000          2.0   \n",
       "unique               NaN              NaN        NaN        NaN          NaN   \n",
       "top                  NaN              NaN        NaN        NaN          NaN   \n",
       "freq                 NaN              NaN        NaN        NaN          NaN   \n",
       "mean                 NaN              NaN        NaN   1.500000          2.0   \n",
       "std                  NaN              NaN        NaN   0.707107          0.0   \n",
       "min                  NaN              NaN        NaN   1.000000          2.0   \n",
       "25%                  NaN              NaN        NaN   1.250000          2.0   \n",
       "50%                  NaN              NaN        NaN   1.500000          2.0   \n",
       "75%                  NaN              NaN        NaN   1.750000          2.0   \n",
       "max                  NaN              NaN        NaN   2.000000          2.0   \n",
       "\n",
       "        father_age  record_weight  \n",
       "count          2.0            2.0  \n",
       "unique         NaN            NaN  \n",
       "top            NaN            NaN  \n",
       "freq           NaN            NaN  \n",
       "mean          18.0            1.0  \n",
       "std            0.0            0.0  \n",
       "min           18.0            1.0  \n",
       "25%           18.0            1.0  \n",
       "50%           18.0            1.0  \n",
       "75%           18.0            1.0  \n",
       "max           18.0            1.0  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "natality[natality[\"gestation_weeks\"]==20].describe(include=\"all\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is sparsity in the data as can be seen in the above line plot and table. At gestation weeks of 20, there are only two records, both of which are twin female newborn births."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x7f553ec09590>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgsAAAIQCAYAAAAPTi2WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n5OuVAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f553f843b10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.pairplot(natality[[\"weight_pounds\",\"gestation_weeks\",\"mother_age\"]].dropna())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdoAAAFUCAYAAACDXv0HAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAFkZJREFUeJzt3XtwlIW5x/FfstsIudEEVtJq8JCU\n+yX0CGQUSBUNF7nZQBiIWiFysBQmxxrlQBhpywClTNUpelAZyrUUaYRCLU6dFgyhQDlF0OAQSi0g\nWAkEyEhuJGTZ84fTKAWyCeyzy758P3+F3c3rk2dkvnnfvRDh8/l8AgAAJiJDPQAAAE5GaAEAMERo\nAQAwRGgBADBEaAEAMOS2OGh5eaXFYc0kJESroqIm1GM4GjsODvZsjx3bC8cdezxx172PM1pJbrcr\n1CM4HjsODvZsjx3bc9qOCS0AAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAAhkw+\nGQoA4Cy5i7YH9HgrZg0O6PECqbi4SMnJHdSxY0pAjscZLQAAX7FzZ5GOHz8asOMRWgDALWn27Hzl\n5j6uxx8fry1bNum3v31LS5f+ovH+d955Wy+/vFiStGrVcuXkjNUzz/xAP/pRgX7967XXPGZFxXnl\n5j4uSfr7349o4MC+KisrkySNHz9GBw9+qD//uVhLly7RpEk5+uc/P73pn4NLxwCAW9Ls2XMVH99G\ndXUXNWXK9/SLX7ymadOe0g9+8N+SpG3b/qjvfS9Xhw8fUlHRdq1cuU5er1e5uY+rS5du1zxmQkKi\n6uvrVV1dpZKSA+ratbtKSg5I6qOEhET16pWmgQMzdP/9A/Xggw8H5OcgtACAW1Jh4ZsqLi6SJJ05\nc1qfffaZvvnNu/TRRweVnJysEyc+Ue/eaSosXK9Bg76jO+5oJUkaMGBQk8ft1au3Sko+1AcfHNAT\nT0zW3r275fP5lJbWx+TnCKvQBvrJ+GC4lZ/wB4Bb1f79+7Rv3//pjTdWqlWrVpoxY6rq6+v00EOZ\neu+9P6pDh/9QRsYDioiIkM/na9Gxe/fuo5KSD1RWdkqDBn1H69atVkREhO6/v+lA3yieowUA3HKq\nq6sUFxevVq1a6ZNPjuvQoY8kSRkZg1VcvEN/+tO7euihTElfhHPXrmLV1dWppqZGu3f/uclj9+nz\nn3r33XeUnNxBkZGRio+P1549u9SrV5okKTo6WjU1gfv3cMPqjBYAEBrBvjqXnn6/Nm/epCefnKDk\n5HvUvXtPSVJ8fLw6duyoY8eONd7WrVsPDRiQoUmTJiop6Rvq2rW7YmNjr3vsb3zjm5KktLRvS/oi\n1OXlZxQfHy9JeuihIVq8eIHeeutNzZ+/WHfddfdN/SwRvpaeczdDeXlloA8piUvH4czjiTP7/wJf\nYs/22LG9G9lxTU2NoqOjdfHiRU2f/l+aOXOOunTpajTh1TyeuOvexxktACDsLV68QMePH1N9fZ2G\nDx8Z1Mj6Q2gBAGHvxz9ecNVtL774Mx08+OEVt2VnT9CIEaODNZYkQgsAcKj8/P8J9QiSeNUxAACm\n/J7RHj16VD/84Q8b/3zy5Enl5eVp0qRJlnMBAOAIfkObkpKiLVu2SJK8Xq8yMjKUmZlpPhgAAE7Q\nokvHe/bsUXJysu666y6reQAAcJQWhXbr1q0aOXKk1SwAADhOs191XF9fr+3btys/P9/vYxMSouV2\nu25qMKdo6k3Mtxt2ERzs2R47tuekHTc7tMXFxerRo4fatWvn97EVFYH7jMhwxyfIfIFP0wkO9myP\nHdsLxx039YtBsy8db926VSNGjAjIQAAA3C6aFdra2lrt3r1bQ4YMsZ4HAABHadal49atW2vv3r3W\nswAA4Dh8MhQAAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYI\nLQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0A\nAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGmhXaCxcuKC8vT8OGDdPw\n4cN14MAB67kAAHAEd3MetGDBAg0aNEhLlixRfX29Ll68aD0XAACO4PeMtqqqSn/96181btw4SVJU\nVJTi4+PNBwMAwAn8hvbkyZNKTEzU7Nmz9eijj2rOnDmqqakJxmwAAIQ9v5eOGxoadOjQIb3wwgtK\nS0vT/PnztWzZMj3zzDPX/Z6EhGi53a6ADhquPJ64UI9wy2AXwcGe7bFje07asd/QJiUlKSkpSWlp\naZKkYcOGadmyZU1+T0UFZ7z/Ul5eGeoRbgkeTxy7CAL2bI8d2wvHHTf1i4HfS8cej0dJSUk6evSo\nJGnPnj1KTU0N3HQAADhYs151/MILL+i5557TpUuXlJycrJ/+9KfWcwEA4AjNCm23bt20adMm61kA\nAHAcPhkKAABDhBYAAEOEFgAAQ4QWAABDhBYAAEOEFgAAQ4QWAABDhBYAAEOEFgAAQ4QWAABDhBYA\nAEOEFgAAQ4QWAABDhBYAAEOEFgAAQ4QWAABDhBYAAEOEFgAAQ4QWAABDhBYAAEOEFgAAQ4QWAABD\nhBYAAEOEFgAAQ4QWAABDhBYAAEOEFgAAQ4QWAABD7lAPgFvL9O0zQz1Ci/3v4MWhHgEAroszWgAA\nDDXrjHbw4MGKiYlRZGSkXC6XNm3aZD0XAACO0OxLx6tXr1ZiYqLlLAAAOA6XjgEAMNTs0D711FPK\nysrShg0bLOcBAMBRmnXpeP369Wrfvr3OnTunyZMnKyUlRf369bvu4xMSouV2uwI2ZDjzeOJCPYLj\nseMrsQ977Niek3bcrNC2b99ektS2bVtlZmaqpKSkydBWVNQEZjoHKC+vDPUIjseOv+TxxLEPY+zY\nXjjuuKlfDPxeOq6pqVFVVVXj17t27VKnTp0CNx0AAA7m94z23Llzmj59uiTJ6/Vq5MiRysjIMB8M\nAAAn8Bva5ORk/e53vwvGLAAAOA5v7wEAwBChBQDAEKEFAMAQoQUAwBChBQDAEKEFAMAQoQUAwBCh\nBQDAEKEFAMAQoQUAwBChBQDAEKEFAMAQoQUAwBChBQDAEKEFAMAQoQUAwBChBQDAEKEFAMAQoQUA\nwBChBQDAEKEFAMAQoQUAwBChBQDAEKEFAMAQoQUAwBChBQDAEKEFAMAQoQUAwBChBQDAEKEFAMBQ\ns0Pr9Xr16KOP6umnn7acBwAAR2l2aNesWaPU1FTLWQAAcJxmhbasrExFRUUaN26c9TwAADhKs0K7\ncOFCPf/884qM5CldAABawu3vAe+9954SExPVs2dP7d27t1kHTUiIltvtuunhnMDjiQv1CI7Hjq/E\nPuyxY3tO2rHf0O7fv1/bt29XcXGx6urqVFVVpeeee04///nPr/s9FRU1AR0ynJWXV4Z6BMdjx1/y\neOLYhzF2bC8cd9zULwZ+Q5ufn6/8/HxJ0t69e7VixYomIwsAAL7Ek64AABjye0b7Venp6UpPT7ea\nBQAAx+GMFgAAQ4QWAABDhBYAAEOEFgAAQ4QWAABDhBYAAEOEFgAAQy16Hy2Am3dkyiS7Yxsdt/Py\nVUZHBpyPM1oAAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoAAAwR\nWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoA\nAAy5/T2grq5Ojz32mOrr6+X1ejV06FDl5eUFYzYAAMKe39BGRUVp9erViomJ0aVLl5STk6OMjAz1\n6dMnGPMBABDW/F46joiIUExMjCSpoaFBDQ0NioiIMB8MAAAn8HtGK0ler1dZWVk6ceKEcnJylJaW\n1uTjExKi5Xa7AjJguPN44kI9guOF246PhHqAGxBuO7bGPuw5acfNCq3L5dKWLVt04cIFTZ8+XUeO\nHFHnzp2v+/iKipqADRjuyssrQz2C47Fje+z4Sx5PHPswFo47buoXgxa96jg+Pl7p6enauXPnTQ8F\nAMDtwG9oz58/rwsXLkiSLl68qN27dyslJcV8MAAAnMDvpeMzZ85o1qxZ8nq98vl8GjZsmB588MFg\nzAYAQNjzG9quXbtq8+bNwZgFAADHadaLoQAg3Ly2qCjUI7TItFkPhHoEGOEjGAEAMERoAQAwRGgB\nADBEaAEAMERoAQAwRGgBADBEaAEAMERoAQAwRGgBADBEaAEAMERoAQAwRGgBADDEPyoAALghJw7M\nszmuyVG/0OHbcw2Pfm2c0QIAYIjQAgBgiNACAGCI0AIAYIjQAgBgiNACAGCI0AIAYIjQAgBgiNAC\nAGCI0AIAYIjQAgBgiNACAGCI0AIAYIjQAgBgiNACAGCI0AIAYMjvP/x+6tQpzZw5U2fPnlVkZKTG\njx+vJ598MhizAQAQ9vyG1uVyadasWerRo4eqqqo0duxYDRgwQN/61reCMR8AAGHN76XjO++8Uz16\n9JAkxcbGKiUlRadPnzYfDAAAJ/B7RvtVn376qUpLS5WWltbk4xISouV2u25qMKfweOJCPYLjhduO\nj4R6gBsQbjsOR+G44xOhHuAGhGLPzQ5tdXW18vLyVFBQoNjY2CYfW1FRc9ODOUV5eWWoR3A8dmyP\nHdtjx8FhteemAt6sVx1funRJeXl5GjVqlIYMGRKwwQAAcDq/ofX5fJozZ45SUlI0efLkYMwEAIBj\n+A3t+++/ry1btugvf/mLxowZozFjxmjHjh3BmA0AgLDn9znavn376m9/+1swZgEAwHH4ZCgAAAwR\nWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoA\nAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAM\nEVoAAAwRWgAADBFaAAAMEVoAAAwRWgAADBFaAAAM+Q3t7Nmzdd9992nkyJHBmAcAAEfxG9qsrCwt\nX748GLMAAOA4fkPbr18/tWnTJhizAADgOG6LgyYkRMvtdlkcOux4PHGhHsHxwm3HR0I9wA0Itx2H\no3Dc8YlQD3ADQrFnk9BWVNRYHDYslZdXhnoEx2PH9tixPXYcHFZ7birgvOoYAABDhBYAAEN+Q/vs\ns89qwoQJOnbsmDIyMlRYWBiMuQAAcAS/z9G+9NJLwZgDAABH4tIxAACGCC0AAIYILQAAhggtAACG\nCC0AAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAAhggt\nAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAAhggtAACGCC0AAIYILQAA\nhggtAACGCC0AAIYILQAAhpoV2uLiYg0dOlSZmZlatmyZ9UwAADiG39B6vV7NmzdPy5cv19atW/X7\n3/9eH3/8cTBmAwAg7PkNbUlJie655x4lJycrKipKI0aM0LZt24IxGwAAYS/C5/P5mnrAH/7wB+3c\nuVMLFiyQJG3evFklJSWaO3duUAYEACCc+T2jvVaHIyIiTIYBAMBp/IY2KSlJZWVljX8+ffq07rzz\nTtOhAABwCr+h7dWrl44fP66TJ0+qvr5eW7du1eDBg4MxGwAAYc/t9wFut+bOnaspU6bI6/Vq7Nix\n6tSpUzBmAwAg7Pl9MRQAALhxfDIUAACGCC0AAIYILQAAhggtEMZKSkpUUlIiSfr444+1cuVK7dix\nI8RTOdfMmTNDPQLCEC+Ggol//OMfOnPmjHr37q2YmJjG24uLi5WRkRHCyZzj1VdfVXFxsRoaGjRg\nwAB9+OGH6t+/v/bs2aOBAwdq2rRpoR4xrH3/+9+/6ra9e/cqPT1dkvT6668He6Tbwr59+3Tw4EF1\n6tRJAwcODPU4AeH37T23k40bN2rs2LGhHiPsrVmzRuvWrVNqaqoOHz6sgoICPfzww5Kkl19+mdAG\nyLvvvqvNmzervr5eAwYMUHFxsWJjYzVlyhRlZ2cT2pt0+vRppaamKjs7WxEREfL5fProo4+Um5sb\n6tEcZdy4cXrrrbckSb/5zW+0bt06ZWZm6tVXX9WhQ4c0derUEE9487h0/BWvvPJKqEdwhMLCQm3a\ntElLly7VmjVrtHTpUq1evVrStT/SEzfG5XLJ5XKpdevW6tChg2JjYyVJrVq1UmQkf7Vv1saNG9Wz\nZ0+9/vrriouLU3p6uu644w71799f/fv3D/V4jtHQ0ND49YYNG7Ry5UrNmDFDK1as0Ntvvx3CyQLn\ntjujHTVq1HXvO3v2bBAncS6v19t4ufjuu+/W2rVrlZeXp88++4zQBtDXvvY11dbWqnXr1tq0aVPj\n7ZWVlYQ2ACIjIzVp0iQNGzZMCxcuVLt27eT1ekM9luNcvnxZn3/+uS5fviyfz6fExERJUnR0tFwu\nV4inC4zbLrTnzp3TL3/5S8XHx19xu8/n04QJE0I0lbO0a9dOpaWl6tatmyQpJiZGb7zxhgoKCnTk\nyJEQT+cc69atU1RUlCRdEdZLly5p0aJFoRrLcZKSkrRkyRIVFRU1XjVA4FRVVSkrK0s+n08REREq\nLy+Xx+NRdXW1Y34xv+1eDFVQUKCsrCz17dv3qvvy8/P14osvhmAqZykrK5PL5ZLH47nqvvfff1/3\n3ntvCKYCEE5qa2t19uxZJScnh3qUm3bbhRYAgGDiiRwAAAwRWgAADBFa4BbTpUsXVVdXB/SYs2bN\n0q9+9StJ0vr167Vq1SpJUmlpqd55552A/rcAXOm2e9Ux4EQNDQ1yu5v313nixImNX5eWlqqoqEiP\nPPKI1WjAbY/QAiHQpUsXzZgxQ7t27VJFRYWeffZZDR069JqP279/f+P7kr/65y5duuj555/Xjh07\ndO+992r48OH6yU9+otraWtXV1Wn8+PGaNGnSVcd85ZVXVFNTo6lTp2rJkiWqqqrSmDFj1K9fP7Vv\n316nTp3S3LlzJX3x3vLRo0dr27Ztat26telOAKcitECIRERE6M0339TRo0c1ceJE9e3bV23btm3R\nMS5fvqy1a9dK+uL9iKtWrVJUVJSqq6uVnZ2tQYMGKTU19Zrfm5CQoLy8PBUVFWnJkiWSpM8//1yP\nPPKI8vPzFRMTow0bNmjkyJFEFrgJPEcLhEh2drYkKSUlRd27d9cHH3zQ4mN897vfbfz64sWLKigo\n0KhRozRx4kSdOXNGhw8fbtHx2rRpo8GDB2vLli1qaGhQYWHhFZeaAbQcZ7TALeBfn4rz71wuV+On\n49TV1V11f3R0dOPXL730kjwejxYtWiS3263c3Nxrfo8/TzzxhPLz89W2bVulpqaqY8eOLT4GgC9x\nRguEyMaNGyVJx48fV2lpqdLS0q56THJysg4ePChJfj9gvbKyUklJSXK73Tpy5Ij27dvnd4bY2FhV\nVlZecVvnzp319a9/XQsXLlROTk5zfxwA10FogRCJiorShAkT9PTTT2vevHnXfH62oKBAc+fOVU5O\njs6fP9/k8aZNm6bCwkKNHTtWr732mvr16+d3hvvuu0+1tbUaPXq05s+f33h7dna2IiMj9cADD7T4\n5wJwJT6CEQiBf3818a1mzpw56tixo6ZMmRLqUYCwxxktgEanT5/W0KFD9cknn+ixxx4L9TiAI3BG\nCwCAIc5oAQAwRGgBADBEaAEAMERoAQAwRGgBADD0/0X7OIQYtp1cAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f553f4717d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = get_distinct_values('plurality')\n",
    "df = df.sort_values('plurality')\n",
    "df.plot(x='plurality', y='avg_wt', kind='bar');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdoAAAFpCAYAAAA7uul0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nzUg44+iJWAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f553f445d10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df = get_distinct_values('alcohol_use')\n",
    "df = df.sort_values('alcohol_use')\n",
    "df.plot(x='alcohol_use', y='avg_wt', kind='bar');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> Create ML dataset by sampling using BigQuery </h2>\n",
    "<p>\n",
    "Let's sample the BigQuery data to create smaller datasets.\n",
    "</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Create SQL query using natality data after the year 2000\n",
    "import google.datalab.bigquery as bq\n",
    "query = \"\"\"\n",
    "SELECT\n",
    "  weight_pounds,\n",
    "  is_male,\n",
    "  mother_age,\n",
    "  plurality,\n",
    "  gestation_weeks,\n",
    "  mother_married,\n",
    "  FARM_FINGERPRINT(CONCAT(CAST(YEAR AS STRING), CAST(month AS STRING), CAST(wday AS STRING))) AS hashmonth\n",
    "FROM\n",
    "  publicdata.samples.natality\n",
    "WHERE year > 2000\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "There are only a limited number of years and months in the dataset. Let's see what the hashmonths are."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 672 unique hashmonths.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hashmonth</th>\n",
       "      <th>num_babies</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3567934916849143973</td>\n",
       "      <td>50688</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8502900579502152792</td>\n",
       "      <td>52992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1671153752880571251</td>\n",
       "      <td>59392</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-6663129674444867382</td>\n",
       "      <td>60672</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6441112521155172543</td>\n",
       "      <td>47105</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             hashmonth  num_babies\n",
       "0  3567934916849143973       50688\n",
       "1  8502900579502152792       52992\n",
       "2  1671153752880571251       59392\n",
       "3 -6663129674444867382       60672\n",
       "4  6441112521155172543       47105"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Call BigQuery but GROUP BY the hashmonth and see number of records for each group to enable us to get the correct train and evaluation percentages\n",
    "df = bq.Query(\"SELECT hashmonth, COUNT(weight_pounds) AS num_babies FROM (\" + query + \") GROUP BY hashmonth\").execute().result().to_dataframe()\n",
    "print(\"There are {} unique hashmonths.\".format(len(df)))\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Here's a way to get a well distributed portion of the data in such a way that the test and train sets do not overlap:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 28644 examples in the train dataset and 7062 in the eval dataset\n"
     ]
    }
   ],
   "source": [
    "# Added the RAND() so that we can now subsample from each of the hashmonths to get approximately the record counts we want\n",
    "trainQuery = \"SELECT * FROM (\" + query + \") WHERE MOD(hashmonth, 4) < 3 AND RAND() < 0.001\"\n",
    "evalQuery = \"SELECT * FROM (\" + query + \") WHERE MOD(hashmonth, 4) = 3 AND RAND() < 0.0015\"\n",
    "traindf = bq.Query(trainQuery).execute().result().to_dataframe()\n",
    "evaldf = bq.Query(evalQuery).execute().result().to_dataframe()\n",
    "print(\"There are {} examples in the train dataset and {} in the eval dataset\".format(len(traindf), len(evaldf)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "<h2> Preprocess data using Pandas </h2>\n",
    "<p>\n",
    "Let's add extra rows to simulate the lack of ultrasound. In the process, we'll also change the plurality column to be a string."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>weight_pounds</th>\n",
       "      <th>is_male</th>\n",
       "      <th>mother_age</th>\n",
       "      <th>plurality</th>\n",
       "      <th>gestation_weeks</th>\n",
       "      <th>mother_married</th>\n",
       "      <th>hashmonth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6.294198</td>\n",
       "      <td>True</td>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>5196045867830167686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6.999677</td>\n",
       "      <td>False</td>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>True</td>\n",
       "      <td>-6718872612173142345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6.750554</td>\n",
       "      <td>True</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>8486568935975462149</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6.933538</td>\n",
       "      <td>False</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>-5725309102206228019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6.314039</td>\n",
       "      <td>False</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>-7901005608806392424</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   weight_pounds  is_male  mother_age  plurality  gestation_weeks  \\\n",
       "0       6.294198     True          15          1              NaN   \n",
       "1       6.999677    False          15          1              NaN   \n",
       "2       6.750554     True          16          1              NaN   \n",
       "3       6.933538    False          16          1              NaN   \n",
       "4       6.314039    False          16          1              NaN   \n",
       "\n",
       "   mother_married            hashmonth  \n",
       "0           False  5196045867830167686  \n",
       "1            True -6718872612173142345  \n",
       "2           False  8486568935975462149  \n",
       "3           False -5725309102206228019  \n",
       "4           False -7901005608806392424  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traindf.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Also notice that there are some very important numeric fields that are missing in some rows (the count in Pandas doesn't count missing data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>weight_pounds</th>\n",
       "      <th>mother_age</th>\n",
       "      <th>plurality</th>\n",
       "      <th>gestation_weeks</th>\n",
       "      <th>hashmonth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>28617.000000</td>\n",
       "      <td>28644.000000</td>\n",
       "      <td>28644.000000</td>\n",
       "      <td>28406.000000</td>\n",
       "      <td>2.864400e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.229129</td>\n",
       "      <td>27.290148</td>\n",
       "      <td>1.035051</td>\n",
       "      <td>38.628459</td>\n",
       "      <td>-5.161007e+17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.313827</td>\n",
       "      <td>6.132353</td>\n",
       "      <td>0.193170</td>\n",
       "      <td>2.578346</td>\n",
       "      <td>5.247425e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.500449</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>-9.153539e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.563162</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>-4.924562e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>7.312733</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>39.000000</td>\n",
       "      <td>-8.490120e+17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>8.062305</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>4.142761e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>17.749417</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>47.000000</td>\n",
       "      <td>9.203353e+18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       weight_pounds    mother_age     plurality  gestation_weeks  \\\n",
       "count   28617.000000  28644.000000  28644.000000     28406.000000   \n",
       "mean        7.229129     27.290148      1.035051        38.628459   \n",
       "std         1.313827      6.132353      0.193170         2.578346   \n",
       "min         0.500449     12.000000      1.000000        17.000000   \n",
       "25%         6.563162     22.000000      1.000000        38.000000   \n",
       "50%         7.312733     27.000000      1.000000        39.000000   \n",
       "75%         8.062305     32.000000      1.000000        40.000000   \n",
       "max        17.749417     53.000000      4.000000        47.000000   \n",
       "\n",
       "          hashmonth  \n",
       "count  2.864400e+04  \n",
       "mean  -5.161007e+17  \n",
       "std    5.247425e+18  \n",
       "min   -9.153539e+18  \n",
       "25%   -4.924562e+18  \n",
       "50%   -8.490120e+17  \n",
       "75%    4.142761e+18  \n",
       "max    9.203353e+18  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's look at a small sample of the training data\n",
    "traindf.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "traindf_corr = df.corr(min_periods = 50)\n",
    "natality_corr.sort_values(by = [\"weight_pounds\"], ascending=False, inplace=True)\n",
    "natality_corr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# It is always crucial to clean raw data before using in ML, so we have a preprocessing step\n",
    "import pandas as pd\n",
    "def preprocess(df):\n",
    "  # clean up data we don't want to train on\n",
    "  # in other words, users will have to tell us the mother's age\n",
    "  # otherwise, our ML service won't work.\n",
    "  # these were chosen because they are such good predictors\n",
    "  # and because these are easy enough to collect\n",
    "  df = df[df.weight_pounds > 0]\n",
    "  df = df[df.mother_age > 0]\n",
    "  df = df[df.gestation_weeks > 0]\n",
    "  df = df[df.plurality > 0]\n",
    "  \n",
    "  # modify plurality field to be a string\n",
    "  twins_etc = dict(zip([1,2,3,4,5],\n",
    "                   ['Single(1)', 'Twins(2)', 'Triplets(3)', 'Quadruplets(4)', 'Quintuplets(5)']))\n",
    "  df['plurality'].replace(twins_etc, inplace=True)\n",
    "  \n",
    "  # now create extra rows to simulate lack of ultrasound\n",
    "  nous = df.copy(deep=True)\n",
    "  nous.loc[nous['plurality'] != 'Single(1)', 'plurality'] = 'Multiple(2+)'\n",
    "  nous['is_male'] = 'Unknown'\n",
    "  \n",
    "  return pd.concat([df, nous], ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>weight_pounds</th>\n",
       "      <th>is_male</th>\n",
       "      <th>mother_age</th>\n",
       "      <th>plurality</th>\n",
       "      <th>gestation_weeks</th>\n",
       "      <th>mother_married</th>\n",
       "      <th>hashmonth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.654773</td>\n",
       "      <td>False</td>\n",
       "      <td>27</td>\n",
       "      <td>Twins(2)</td>\n",
       "      <td>17.0</td>\n",
       "      <td>False</td>\n",
       "      <td>8296688684370151053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.595248</td>\n",
       "      <td>True</td>\n",
       "      <td>22</td>\n",
       "      <td>Single(1)</td>\n",
       "      <td>19.0</td>\n",
       "      <td>False</td>\n",
       "      <td>-6917645359793053651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.714298</td>\n",
       "      <td>False</td>\n",
       "      <td>24</td>\n",
       "      <td>Single(1)</td>\n",
       "      <td>19.0</td>\n",
       "      <td>False</td>\n",
       "      <td>8452479967450223670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.687842</td>\n",
       "      <td>False</td>\n",
       "      <td>24</td>\n",
       "      <td>Twins(2)</td>\n",
       "      <td>19.0</td>\n",
       "      <td>False</td>\n",
       "      <td>-4474219042277703455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.859803</td>\n",
       "      <td>True</td>\n",
       "      <td>33</td>\n",
       "      <td>Twins(2)</td>\n",
       "      <td>19.0</td>\n",
       "      <td>True</td>\n",
       "      <td>-4044329270713657284</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   weight_pounds is_male  mother_age  plurality  gestation_weeks  \\\n",
       "0       0.654773   False          27   Twins(2)             17.0   \n",
       "1       0.595248    True          22  Single(1)             19.0   \n",
       "2       0.714298   False          24  Single(1)             19.0   \n",
       "3       0.687842   False          24   Twins(2)             19.0   \n",
       "4       0.859803    True          33   Twins(2)             19.0   \n",
       "\n",
       "   mother_married            hashmonth  \n",
       "0           False  8296688684370151053  \n",
       "1           False -6917645359793053651  \n",
       "2           False  8452479967450223670  \n",
       "3           False -4474219042277703455  \n",
       "4            True -4044329270713657284  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traindf.head()# Let's see a small sample of the training data now after our preprocessing\n",
    "traindf = preprocess(traindf)\n",
    "evaldf = preprocess(evaldf)\n",
    "traindf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>weight_pounds</th>\n",
       "      <th>is_male</th>\n",
       "      <th>mother_age</th>\n",
       "      <th>plurality</th>\n",
       "      <th>gestation_weeks</th>\n",
       "      <th>mother_married</th>\n",
       "      <th>hashmonth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>56763</th>\n",
       "      <td>6.790238</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>29</td>\n",
       "      <td>Single(1)</td>\n",
       "      <td>47.0</td>\n",
       "      <td>True</td>\n",
       "      <td>3514554563944531650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56764</th>\n",
       "      <td>5.654857</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>30</td>\n",
       "      <td>Single(1)</td>\n",
       "      <td>47.0</td>\n",
       "      <td>True</td>\n",
       "      <td>-6390112116658590704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56765</th>\n",
       "      <td>7.963097</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>31</td>\n",
       "      <td>Single(1)</td>\n",
       "      <td>47.0</td>\n",
       "      <td>False</td>\n",
       "      <td>8154469796392365364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56766</th>\n",
       "      <td>7.528786</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>36</td>\n",
       "      <td>Single(1)</td>\n",
       "      <td>47.0</td>\n",
       "      <td>True</td>\n",
       "      <td>-5984331063527788494</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56767</th>\n",
       "      <td>8.300404</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>37</td>\n",
       "      <td>Single(1)</td>\n",
       "      <td>47.0</td>\n",
       "      <td>True</td>\n",
       "      <td>-4892284310518763364</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       weight_pounds  is_male  mother_age  plurality  gestation_weeks  \\\n",
       "56763       6.790238  Unknown          29  Single(1)             47.0   \n",
       "56764       5.654857  Unknown          30  Single(1)             47.0   \n",
       "56765       7.963097  Unknown          31  Single(1)             47.0   \n",
       "56766       7.528786  Unknown          36  Single(1)             47.0   \n",
       "56767       8.300404  Unknown          37  Single(1)             47.0   \n",
       "\n",
       "       mother_married            hashmonth  \n",
       "56763            True  3514554563944531650  \n",
       "56764            True -6390112116658590704  \n",
       "56765           False  8154469796392365364  \n",
       "56766            True -5984331063527788494  \n",
       "56767            True -4892284310518763364  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traindf.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>weight_pounds</th>\n",
       "      <th>mother_age</th>\n",
       "      <th>gestation_weeks</th>\n",
       "      <th>hashmonth</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>56768.000000</td>\n",
       "      <td>56768.000000</td>\n",
       "      <td>56768.000000</td>\n",
       "      <td>5.676800e+04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.229158</td>\n",
       "      <td>27.295695</td>\n",
       "      <td>38.634371</td>\n",
       "      <td>-5.166335e+17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.313803</td>\n",
       "      <td>6.129348</td>\n",
       "      <td>2.558316</td>\n",
       "      <td>5.246233e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.500449</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>-9.153539e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.563162</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>-4.924562e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>7.312733</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>39.000000</td>\n",
       "      <td>-8.490120e+17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>8.062305</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>4.142761e+18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>17.749417</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>47.000000</td>\n",
       "      <td>9.203353e+18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       weight_pounds    mother_age  gestation_weeks     hashmonth\n",
       "count   56768.000000  56768.000000     56768.000000  5.676800e+04\n",
       "mean        7.229158     27.295695        38.634371 -5.166335e+17\n",
       "std         1.313803      6.129348         2.558316  5.246233e+18\n",
       "min         0.500449     12.000000        17.000000 -9.153539e+18\n",
       "25%         6.563162     22.000000        38.000000 -4.924562e+18\n",
       "50%         7.312733     27.000000        39.000000 -8.490120e+17\n",
       "75%         8.062305     32.000000        40.000000  4.142761e+18\n",
       "max        17.749417     53.000000        47.000000  9.203353e+18"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Describe only does numeric columns, so you won't see plurality\n",
    "traindf.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "<h2> Write out </h2>\n",
    "<p>\n",
    "In the final versions, we want to read from files, not Pandas dataframes. So, write the Pandas dataframes out as CSV files. \n",
    "Using CSV files gives us the advantage of shuffling during read. This is important for distributed training because some workers might be slower than others, and shuffling the data helps prevent the same data from being assigned to the slow workers.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "traindf.to_csv('train.csv', index=False, header=False)\n",
    "evaldf.to_csv('eval.csv', index=False, header=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  14012 eval.csv\n",
      "  56768 train.csv\n",
      "  70780 total\n",
      "==> eval.csv <==\n",
      "1.8739292269999999,True,19,Twins(2),22.0,False,4192359966907760603\n",
      "0.89066753848,True,32,Single(1),22.0,True,7817389469197718543\n",
      "1.1243575362,False,33,Twins(2),22.0,True,3080718970743272203\n",
      "1.28088574222,True,34,Single(1),22.0,True,3236745624095779459\n",
      "1.4770971553999999,True,36,Twins(2),22.0,True,4692594063779746439\n",
      "1.25002102554,True,18,Single(1),23.0,True,475224971195399155\n",
      "1.34041055296,True,18,Single(1),23.0,False,7090831592889798475\n",
      "2.20462262,True,19,Single(1),23.0,False,4931177248205345407\n",
      "1.06262810284,False,22,Twins(2),23.0,False,6441112521155172543\n",
      "1.1353806493,True,25,Twins(2),23.0,False,475224971195399155\n",
      "\n",
      "==> train.csv <==\n",
      "0.6547729181399999,False,27,Twins(2),17.0,False,8296688684370151053\n",
      "0.5952481074,True,22,Single(1),19.0,False,-6917645359793053651\n",
      "0.71429772888,False,24,Single(1),19.0,False,8452479967450223670\n",
      "0.6878422574399999,False,24,Twins(2),19.0,False,-4474219042277703455\n",
      "0.8598028218,True,33,Twins(2),19.0,True,-4044329270713657284\n",
      "0.50044933474,True,37,Single(1),19.0,True,5220925218139314588\n",
      "3.62439958728,False,17,Single(1),20.0,False,-1056251574422373543\n",
      "3.9572976028999998,False,17,Single(1),20.0,False,-8576655865514498436\n",
      "0.7870502753399999,False,19,Single(1),20.0,False,-3503095046632101015\n",
      "0.6834330122,False,21,Single(1),20.0,False,-4796714464186310809\n",
      "==> eval.csv <==\n",
      "6.6248909731,Unknown,22,Single(1),47.0,True,4692594063779746439\n",
      "7.50012615324,Unknown,24,Single(1),47.0,False,7943406415937454643\n",
      "7.25100379718,Unknown,24,Single(1),47.0,False,3782070227923997239\n",
      "7.87491199864,Unknown,24,Single(1),47.0,True,84083309734244571\n",
      "8.75014717878,Unknown,25,Single(1),47.0,True,5284681629343902631\n",
      "7.50012615324,Unknown,25,Single(1),47.0,True,4917403277273245239\n",
      "7.81318256528,Unknown,26,Single(1),47.0,True,1671153752880571251\n",
      "7.1870697412,Unknown,26,Single(1),47.0,True,5701501650722053367\n",
      "8.1130112416,Unknown,28,Single(1),47.0,True,3845742738988396339\n",
      "8.50102482272,Unknown,28,Single(1),47.0,True,2569165728043636699\n",
      "\n",
      "==> train.csv <==\n",
      "6.93794738514,Unknown,27,Single(1),47.0,True,1360100874594703713\n",
      "8.56275425608,Unknown,27,Single(1),47.0,True,-8662214199677122618\n",
      "7.2928916269599995,Unknown,28,Single(1),47.0,True,-6460253540370026418\n",
      "6.9996768185,Unknown,28,Single(1),47.0,False,4240703155864588030\n",
      "7.7492485093,Unknown,29,Single(1),47.0,True,5938032612605577757\n",
      "6.7902376696,Unknown,29,Single(1),47.0,True,3514554563944531650\n",
      "5.6548570203,Unknown,30,Single(1),47.0,True,-6390112116658590704\n",
      "7.963096903439999,Unknown,31,Single(1),47.0,False,8154469796392365364\n",
      "7.5287862473,Unknown,36,Single(1),47.0,True,-5984331063527788494\n",
      "8.3004041643,Unknown,37,Single(1),47.0,True,-4892284310518763364\n"
     ]
    }
   ],
   "source": [
    "%bash\n",
    "wc -l *.csv\n",
    "head *.csv\n",
    "tail *.csv"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Copyright 2017-2018 Google Inc. Licensed under the Apache License, Version 2.0 (the \"License\"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}