Disease Outbreak Prediction.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Disease Outbreak Prediction \n",
    "* Building a Machine Learning model based on various parameters such as weather features. <br>\n",
    "* Feature engineering implementation is the heart of this model.\n",
    "* Dataset\n",
    "    * Kaggle: [Contagious disease outbreak prediction](https://www.kaggle.com/pitt/contagious-diseases)<br>\n",
    "    * Historic Data extracted for the past 60 years extracted from [NOAA](https://www.ncdc.noaa.gov/cag/time-series/us)<br>\n",
    "* Machine Learning implementation based on multiple classifier algorithm implementations.\n",
    "* References used:\n",
    "    * [Rowan Chakoumakos - Disease severity outbreak prediction](https://web.stanford.edu/group/journal/cgi-bin/wordpress/wp-content/uploads/2012/09/Chakoumakos_Eng_2010.pdf)<br>\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# load pre-requisite dependencies\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\") \n",
    "import pandas as pd\n",
    "import calendar\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# function to extract month and year from the dataset\n",
    "\n",
    "def week2month(hep):\n",
    "    hep['LastDayWeek'] = pd.to_datetime((hep['week']-1).astype(str) + \"6\", format=\"%Y%U%w\")\n",
    "    hep['MonthMax'] = pd.DatetimeIndex(hep['LastDayWeek']).month\n",
    "    hep['Year'] = pd.DatetimeIndex(hep['LastDayWeek']).year\n",
    "    hep['MonthName'] = [calendar.month_name[i] for i in hep.MonthMax]\n",
    "    return hep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# function to load the dataset\n",
    "\n",
    "def get_input_transform(file_):\n",
    "    return week2month(pd.read_csv(file_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the dataset from Kaggle\n",
    "hep = get_input_transform('hepatitis.csv')\n",
    "mea = get_input_transform('measles.csv')\n",
    "mum = get_input_transform('mumps.csv')\n",
    "per = get_input_transform('pertussis.csv')\n",
    "pol = get_input_transform('polio.csv')\n",
    "rub = get_input_transform('rubella.csv')\n",
    "sma = get_input_transform('smallpox.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>week</th>\n",
       "      <th>state</th>\n",
       "      <th>state_name</th>\n",
       "      <th>disease</th>\n",
       "      <th>cases</th>\n",
       "      <th>incidence_per_capita</th>\n",
       "      <th>LastDayWeek</th>\n",
       "      <th>MonthMax</th>\n",
       "      <th>Year</th>\n",
       "      <th>MonthName</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>ALABAMA</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>5</td>\n",
       "      <td>0.14</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>ARKANSAS</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>11</td>\n",
       "      <td>0.58</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>196601</td>\n",
       "      <td>AZ</td>\n",
       "      <td>ARIZONA</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>6</td>\n",
       "      <td>0.37</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>196601</td>\n",
       "      <td>CA</td>\n",
       "      <td>CALIFORNIA</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>89</td>\n",
       "      <td>0.47</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>196601</td>\n",
       "      <td>CO</td>\n",
       "      <td>COLORADO</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>1</td>\n",
       "      <td>0.05</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     week state  state_name      disease cases  incidence_per_capita  \\\n",
       "0  196601    AL     ALABAMA  HEPATITIS A     5                  0.14   \n",
       "1  196601    AR    ARKANSAS  HEPATITIS A    11                  0.58   \n",
       "2  196601    AZ     ARIZONA  HEPATITIS A     6                  0.37   \n",
       "3  196601    CA  CALIFORNIA  HEPATITIS A    89                  0.47   \n",
       "4  196601    CO    COLORADO  HEPATITIS A     1                  0.05   \n",
       "\n",
       "  LastDayWeek  MonthMax  Year MonthName  \n",
       "0  1966-01-01         1  1966   January  \n",
       "1  1966-01-01         1  1966   January  \n",
       "2  1966-01-01         1  1966   January  \n",
       "3  1966-01-01         1  1966   January  \n",
       "4  1966-01-01         1  1966   January  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# for our exploratory purpose we examine data for 1960 through to 2011 for US states\n",
    "# combining all the disease datasets is shown below\n",
    "\n",
    "train_data = hep\n",
    "for i in [mea,mum,per,pol,rub,sma]:\n",
    "    train_data = train_data.append(i)\n",
    "train_data = train_data.loc[(train_data['Year'] >= 1960) & (train_data['Year'] <=2011)]\n",
    "\n",
    "# examine the dataset \n",
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# some data discrepancies must be resolved (issues like \\\\N )\n",
    "train_data_bad = train_data[train_data.cases==train_data.cases.max()]\n",
    "ind = list(train_data_bad.index)\n",
    "train_data = train_data.drop(train_data.index[ind])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# will be using visulisation.csv for our visualization purposes later\n",
    "\n",
    "train_data.to_csv('for_visulisation.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# cases attribute must be in integer form but given in str format (Numerical data of tmost importance!)\n",
    "\n",
    "train_data.cases = train_data.cases.astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>week</th>\n",
       "      <th>cases</th>\n",
       "      <th>incidence_per_capita</th>\n",
       "      <th>MonthMax</th>\n",
       "      <th>Year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>361935.000000</td>\n",
       "      <td>361935.000000</td>\n",
       "      <td>361935.000000</td>\n",
       "      <td>361935.000000</td>\n",
       "      <td>361935.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>198461.036059</td>\n",
       "      <td>15.238026</td>\n",
       "      <td>0.429124</td>\n",
       "      <td>6.933999</td>\n",
       "      <td>1984.325221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1330.980200</td>\n",
       "      <td>84.493216</td>\n",
       "      <td>2.233045</td>\n",
       "      <td>3.279924</td>\n",
       "      <td>13.298776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>196001.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1960.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>197349.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1973.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>198342.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.030000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>1983.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>199524.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>0.190000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>1995.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>201152.000000</td>\n",
       "      <td>7101.000000</td>\n",
       "      <td>209.160000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>2011.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                week          cases  incidence_per_capita       MonthMax  \\\n",
       "count  361935.000000  361935.000000         361935.000000  361935.000000   \n",
       "mean   198461.036059      15.238026              0.429124       6.933999   \n",
       "std      1330.980200      84.493216              2.233045       3.279924   \n",
       "min    196001.000000       0.000000              0.000000       1.000000   \n",
       "25%    197349.000000       0.000000              0.000000       4.000000   \n",
       "50%    198342.000000       1.000000              0.030000       7.000000   \n",
       "75%    199524.000000       7.000000              0.190000      10.000000   \n",
       "max    201152.000000    7101.000000            209.160000      12.000000   \n",
       "\n",
       "                Year  \n",
       "count  361935.000000  \n",
       "mean     1984.325221  \n",
       "std        13.298776  \n",
       "min      1960.000000  \n",
       "25%      1973.000000  \n",
       "50%      1983.000000  \n",
       "75%      1995.000000  \n",
       "max      2011.000000  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining the characteristics of the dataset\n",
    "train_data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# converting non-numerical data to lowercase (to keep consistencies over all future datasets included)\n",
    "\n",
    "train_data.state_name = [i.lower() for i in train_data.state_name]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature additions\n",
    "* Adding features like\n",
    "    * Average temperature\n",
    "    * Average precipitation\n",
    "* These features have certain effect on the contagious disease spread "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Average temperature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the average temperature dataset (source: NOAA)\n",
    "'''\n",
    "    components of dataset:\n",
    "        Date   : yyyymm ,where y-year and m-month\n",
    "        Value  : average temperature value\n",
    "        Anomaly: deviation from mean temperature \n",
    "        state  : state for the corresponing data tuple (US states used for the data)\n",
    "'''\n",
    "\n",
    "join_tavg = pd.read_csv('tavg.csv', index_col= 0)\n",
    "\n",
    "# make sure non-nmerical data in lowercase format\n",
    "join_tavg.state = [i.lower() for i in join_tavg.state]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# convert the column names for tavg dataset to match column names to our primary dataset\n",
    "join_tavg.columns = ['week', 't_avg', 'tavg_anomaly', 'state_name']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(30576, 4)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>week</th>\n",
       "      <th>t_avg</th>\n",
       "      <th>tavg_anomaly</th>\n",
       "      <th>state_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>196001</td>\n",
       "      <td>44.5</td>\n",
       "      <td>-0.4</td>\n",
       "      <td>alabama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>196101</td>\n",
       "      <td>39.2</td>\n",
       "      <td>-5.7</td>\n",
       "      <td>alabama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>196201</td>\n",
       "      <td>42.6</td>\n",
       "      <td>-2.3</td>\n",
       "      <td>alabama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>196301</td>\n",
       "      <td>39.1</td>\n",
       "      <td>-5.8</td>\n",
       "      <td>alabama</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>196401</td>\n",
       "      <td>42.4</td>\n",
       "      <td>-2.5</td>\n",
       "      <td>alabama</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     week  t_avg  tavg_anomaly state_name\n",
       "0  196001   44.5          -0.4    alabama\n",
       "1  196101   39.2          -5.7    alabama\n",
       "2  196201   42.6          -2.3    alabama\n",
       "3  196301   39.1          -5.8    alabama\n",
       "4  196401   42.4          -2.5    alabama"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examine the dataset\n",
    "\n",
    "print join_tavg.shape\n",
    "join_tavg.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# join the avg temperature dataset with our priary dataset over common columns\n",
    "\n",
    "result = pd.merge(train_data, join_tavg, on=['week', 'state_name'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60610, 12)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>week</th>\n",
       "      <th>state</th>\n",
       "      <th>state_name</th>\n",
       "      <th>disease</th>\n",
       "      <th>cases</th>\n",
       "      <th>incidence_per_capita</th>\n",
       "      <th>LastDayWeek</th>\n",
       "      <th>MonthMax</th>\n",
       "      <th>Year</th>\n",
       "      <th>MonthName</th>\n",
       "      <th>t_avg</th>\n",
       "      <th>tavg_anomaly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>5</td>\n",
       "      <td>0.14</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>40.0</td>\n",
       "      <td>-4.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>RUBELLA</td>\n",
       "      <td>7</td>\n",
       "      <td>0.20</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>40.0</td>\n",
       "      <td>-4.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>11</td>\n",
       "      <td>0.58</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>34.7</td>\n",
       "      <td>-4.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>MEASLES</td>\n",
       "      <td>2</td>\n",
       "      <td>0.11</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>34.7</td>\n",
       "      <td>-4.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>196601</td>\n",
       "      <td>AZ</td>\n",
       "      <td>arizona</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>6</td>\n",
       "      <td>0.37</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>37.8</td>\n",
       "      <td>-2.9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     week state state_name      disease  cases  incidence_per_capita  \\\n",
       "0  196601    AL    alabama  HEPATITIS A      5                  0.14   \n",
       "1  196601    AL    alabama      RUBELLA      7                  0.20   \n",
       "2  196601    AR   arkansas  HEPATITIS A     11                  0.58   \n",
       "3  196601    AR   arkansas      MEASLES      2                  0.11   \n",
       "4  196601    AZ    arizona  HEPATITIS A      6                  0.37   \n",
       "\n",
       "  LastDayWeek  MonthMax  Year MonthName  t_avg  tavg_anomaly  \n",
       "0  1966-01-01         1  1966   January   40.0          -4.9  \n",
       "1  1966-01-01         1  1966   January   40.0          -4.9  \n",
       "2  1966-01-01         1  1966   January   34.7          -4.3  \n",
       "3  1966-01-01         1  1966   January   34.7          -4.3  \n",
       "4  1966-01-01         1  1966   January   37.8          -2.9  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#examining the primary dataset\n",
    "\n",
    "print result.shape\n",
    "train_data = result # make this your current working dataframe \n",
    "result.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>week</th>\n",
       "      <th>state</th>\n",
       "      <th>state_name</th>\n",
       "      <th>disease</th>\n",
       "      <th>cases</th>\n",
       "      <th>incidence_per_capita</th>\n",
       "      <th>LastDayWeek</th>\n",
       "      <th>MonthMax</th>\n",
       "      <th>Year</th>\n",
       "      <th>MonthName</th>\n",
       "      <th>t_avg</th>\n",
       "      <th>tavg_anomaly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>5</td>\n",
       "      <td>0.14</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.513353</td>\n",
       "      <td>0.395639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>RUBELLA</td>\n",
       "      <td>7</td>\n",
       "      <td>0.20</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.513353</td>\n",
       "      <td>0.395639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>11</td>\n",
       "      <td>0.58</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.460930</td>\n",
       "      <td>0.414330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>MEASLES</td>\n",
       "      <td>2</td>\n",
       "      <td>0.11</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.460930</td>\n",
       "      <td>0.414330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>196601</td>\n",
       "      <td>AZ</td>\n",
       "      <td>arizona</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>6</td>\n",
       "      <td>0.37</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.491592</td>\n",
       "      <td>0.457944</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     week state state_name      disease  cases  incidence_per_capita  \\\n",
       "0  196601    AL    alabama  HEPATITIS A      5                  0.14   \n",
       "1  196601    AL    alabama      RUBELLA      7                  0.20   \n",
       "2  196601    AR   arkansas  HEPATITIS A     11                  0.58   \n",
       "3  196601    AR   arkansas      MEASLES      2                  0.11   \n",
       "4  196601    AZ    arizona  HEPATITIS A      6                  0.37   \n",
       "\n",
       "  LastDayWeek  MonthMax  Year MonthName     t_avg  tavg_anomaly  \n",
       "0  1966-01-01         1  1966   January  0.513353      0.395639  \n",
       "1  1966-01-01         1  1966   January  0.513353      0.395639  \n",
       "2  1966-01-01         1  1966   January  0.460930      0.414330  \n",
       "3  1966-01-01         1  1966   January  0.460930      0.414330  \n",
       "4  1966-01-01         1  1966   January  0.491592      0.457944  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining the primary dataset now\n",
    "train_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Average Precipitation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the precipitation data (source: NOAA)\n",
    "\n",
    "join_pcp = pd.read_csv('pcp.csv', index_col= 0)\n",
    "\n",
    "join_pcp.state = [i.lower() for i in join_pcp.state] # lowercasing the non-numerical data\n",
    "\n",
    "join_pcp.columns = ['week', 'precipitate', 'pcp_anomaly', 'state_name'] # change colun names for merging purpose"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# joining the precipitation dataset to the primary dataset over common columns\n",
    "\n",
    "result = pd.merge(train_data, join_pcp, on=['week', 'state_name'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>week</th>\n",
       "      <th>state</th>\n",
       "      <th>state_name</th>\n",
       "      <th>disease</th>\n",
       "      <th>cases</th>\n",
       "      <th>incidence_per_capita</th>\n",
       "      <th>LastDayWeek</th>\n",
       "      <th>MonthMax</th>\n",
       "      <th>Year</th>\n",
       "      <th>MonthName</th>\n",
       "      <th>t_avg</th>\n",
       "      <th>tavg_anomaly</th>\n",
       "      <th>precipitate</th>\n",
       "      <th>pcp_anomaly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>5</td>\n",
       "      <td>0.14</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.513353</td>\n",
       "      <td>0.395639</td>\n",
       "      <td>5.96</td>\n",
       "      <td>0.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>RUBELLA</td>\n",
       "      <td>7</td>\n",
       "      <td>0.20</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.513353</td>\n",
       "      <td>0.395639</td>\n",
       "      <td>5.96</td>\n",
       "      <td>0.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>11</td>\n",
       "      <td>0.58</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.460930</td>\n",
       "      <td>0.414330</td>\n",
       "      <td>4.30</td>\n",
       "      <td>0.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>MEASLES</td>\n",
       "      <td>2</td>\n",
       "      <td>0.11</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.460930</td>\n",
       "      <td>0.414330</td>\n",
       "      <td>4.30</td>\n",
       "      <td>0.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>196601</td>\n",
       "      <td>AZ</td>\n",
       "      <td>arizona</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>6</td>\n",
       "      <td>0.37</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.491592</td>\n",
       "      <td>0.457944</td>\n",
       "      <td>0.76</td>\n",
       "      <td>-0.32</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     week state state_name      disease  cases  incidence_per_capita  \\\n",
       "0  196601    AL    alabama  HEPATITIS A      5                  0.14   \n",
       "1  196601    AL    alabama      RUBELLA      7                  0.20   \n",
       "2  196601    AR   arkansas  HEPATITIS A     11                  0.58   \n",
       "3  196601    AR   arkansas      MEASLES      2                  0.11   \n",
       "4  196601    AZ    arizona  HEPATITIS A      6                  0.37   \n",
       "\n",
       "  LastDayWeek  MonthMax  Year MonthName     t_avg  tavg_anomaly  precipitate  \\\n",
       "0  1966-01-01         1  1966   January  0.513353      0.395639         5.96   \n",
       "1  1966-01-01         1  1966   January  0.513353      0.395639         5.96   \n",
       "2  1966-01-01         1  1966   January  0.460930      0.414330         4.30   \n",
       "3  1966-01-01         1  1966   January  0.460930      0.414330         4.30   \n",
       "4  1966-01-01         1  1966   January  0.491592      0.457944         0.76   \n",
       "\n",
       "   pcp_anomaly  \n",
       "0         0.66  \n",
       "1         0.66  \n",
       "2         0.25  \n",
       "3         0.25  \n",
       "4        -0.32  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining the data after addition of the features\n",
    "\n",
    "result.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Normalize the statistical data\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler as mm\n",
    "\n",
    "scaler = mm()\n",
    "\n",
    "# avg temp deviations\n",
    "scaler.fit(train_data['tavg_anomaly'].reshape(-1,1))\n",
    "train_data['tavg_anomaly'] = scaler.transform(train_data['tavg_anomaly'].reshape(-1,1))\n",
    "\n",
    "# avg temp\n",
    "scaler.fit(train_data['t_avg'].reshape(-1,1))\n",
    "train_data['t_avg'] = scaler.transform(train_data['t_avg'].reshape(-1,1))\n",
    "\n",
    "# precipitation deviations\n",
    "pcp_anomaly = scaler.fit_transform(train_data['pcp_anomaly'].values.reshape(-1,1))\n",
    "train_data['pcp_anomaly'] = pcp_anomaly\n",
    "\n",
    "# precipitation\n",
    "scaler.fit(train_data['precipitate'].reshape(-1,1))\n",
    "train_data['precipitate'] = scaler.transform(train_data['precipitate'].reshape(-1,1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>week</th>\n",
       "      <th>state</th>\n",
       "      <th>state_name</th>\n",
       "      <th>disease</th>\n",
       "      <th>cases</th>\n",
       "      <th>incidence_per_capita</th>\n",
       "      <th>LastDayWeek</th>\n",
       "      <th>MonthMax</th>\n",
       "      <th>Year</th>\n",
       "      <th>MonthName</th>\n",
       "      <th>t_avg</th>\n",
       "      <th>tavg_anomaly</th>\n",
       "      <th>precipitate</th>\n",
       "      <th>pcp_anomaly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>5</td>\n",
       "      <td>0.14</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.513353</td>\n",
       "      <td>0.395639</td>\n",
       "      <td>0.374372</td>\n",
       "      <td>0.341324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>RUBELLA</td>\n",
       "      <td>7</td>\n",
       "      <td>0.20</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.513353</td>\n",
       "      <td>0.395639</td>\n",
       "      <td>0.374372</td>\n",
       "      <td>0.341324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>11</td>\n",
       "      <td>0.58</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.460930</td>\n",
       "      <td>0.414330</td>\n",
       "      <td>0.270101</td>\n",
       "      <td>0.317922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>MEASLES</td>\n",
       "      <td>2</td>\n",
       "      <td>0.11</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.460930</td>\n",
       "      <td>0.414330</td>\n",
       "      <td>0.270101</td>\n",
       "      <td>0.317922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>196601</td>\n",
       "      <td>AZ</td>\n",
       "      <td>arizona</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>6</td>\n",
       "      <td>0.37</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.491592</td>\n",
       "      <td>0.457944</td>\n",
       "      <td>0.047739</td>\n",
       "      <td>0.285388</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     week state state_name      disease  cases  incidence_per_capita  \\\n",
       "0  196601    AL    alabama  HEPATITIS A      5                  0.14   \n",
       "1  196601    AL    alabama      RUBELLA      7                  0.20   \n",
       "2  196601    AR   arkansas  HEPATITIS A     11                  0.58   \n",
       "3  196601    AR   arkansas      MEASLES      2                  0.11   \n",
       "4  196601    AZ    arizona  HEPATITIS A      6                  0.37   \n",
       "\n",
       "  LastDayWeek  MonthMax  Year MonthName     t_avg  tavg_anomaly  precipitate  \\\n",
       "0  1966-01-01         1  1966   January  0.513353      0.395639     0.374372   \n",
       "1  1966-01-01         1  1966   January  0.513353      0.395639     0.374372   \n",
       "2  1966-01-01         1  1966   January  0.460930      0.414330     0.270101   \n",
       "3  1966-01-01         1  1966   January  0.460930      0.414330     0.270101   \n",
       "4  1966-01-01         1  1966   January  0.491592      0.457944     0.047739   \n",
       "\n",
       "   pcp_anomaly  \n",
       "0     0.341324  \n",
       "1     0.341324  \n",
       "2     0.317922  \n",
       "3     0.317922  \n",
       "4     0.285388  "
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining the main dataset \n",
    "\n",
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# save this primary dataset for future purposes\n",
    "\n",
    "result.to_csv('result.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save the dataset in current working dataframe\n",
    "train_data = result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# load this dataset for future purposes\n",
    "'''\n",
    "uncomment the line below when needed,\n",
    "no need to perform above steps once you reach this step\n",
    "\n",
    "'''\n",
    "\n",
    "#train_data = pd.read_csv('result.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# load the working dataset for future purposes (no need to do the above steps)\n",
    "\n",
    "train_data = pd.read_csv('result.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>week</th>\n",
       "      <th>state</th>\n",
       "      <th>state_name</th>\n",
       "      <th>disease</th>\n",
       "      <th>cases</th>\n",
       "      <th>incidence_per_capita</th>\n",
       "      <th>LastDayWeek</th>\n",
       "      <th>MonthMax</th>\n",
       "      <th>Year</th>\n",
       "      <th>MonthName</th>\n",
       "      <th>t_avg</th>\n",
       "      <th>tavg_anomaly</th>\n",
       "      <th>precipitate</th>\n",
       "      <th>pcp_anomaly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>5</td>\n",
       "      <td>0.14</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.513353</td>\n",
       "      <td>0.395639</td>\n",
       "      <td>0.374372</td>\n",
       "      <td>0.341324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>196601</td>\n",
       "      <td>AL</td>\n",
       "      <td>alabama</td>\n",
       "      <td>RUBELLA</td>\n",
       "      <td>7</td>\n",
       "      <td>0.20</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.513353</td>\n",
       "      <td>0.395639</td>\n",
       "      <td>0.374372</td>\n",
       "      <td>0.341324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>11</td>\n",
       "      <td>0.58</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.460930</td>\n",
       "      <td>0.414330</td>\n",
       "      <td>0.270101</td>\n",
       "      <td>0.317922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>196601</td>\n",
       "      <td>AR</td>\n",
       "      <td>arkansas</td>\n",
       "      <td>MEASLES</td>\n",
       "      <td>2</td>\n",
       "      <td>0.11</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.460930</td>\n",
       "      <td>0.414330</td>\n",
       "      <td>0.270101</td>\n",
       "      <td>0.317922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>196601</td>\n",
       "      <td>AZ</td>\n",
       "      <td>arizona</td>\n",
       "      <td>HEPATITIS A</td>\n",
       "      <td>6</td>\n",
       "      <td>0.37</td>\n",
       "      <td>1966-01-01</td>\n",
       "      <td>1</td>\n",
       "      <td>1966</td>\n",
       "      <td>January</td>\n",
       "      <td>0.491592</td>\n",
       "      <td>0.457944</td>\n",
       "      <td>0.047739</td>\n",
       "      <td>0.285388</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     week state state_name      disease  cases  incidence_per_capita  \\\n",
       "0  196601    AL    alabama  HEPATITIS A      5                  0.14   \n",
       "1  196601    AL    alabama      RUBELLA      7                  0.20   \n",
       "2  196601    AR   arkansas  HEPATITIS A     11                  0.58   \n",
       "3  196601    AR   arkansas      MEASLES      2                  0.11   \n",
       "4  196601    AZ    arizona  HEPATITIS A      6                  0.37   \n",
       "\n",
       "  LastDayWeek  MonthMax  Year MonthName     t_avg  tavg_anomaly  precipitate  \\\n",
       "0  1966-01-01         1  1966   January  0.513353      0.395639     0.374372   \n",
       "1  1966-01-01         1  1966   January  0.513353      0.395639     0.374372   \n",
       "2  1966-01-01         1  1966   January  0.460930      0.414330     0.270101   \n",
       "3  1966-01-01         1  1966   January  0.460930      0.414330     0.270101   \n",
       "4  1966-01-01         1  1966   January  0.491592      0.457944     0.047739   \n",
       "\n",
       "   pcp_anomaly  \n",
       "0     0.341324  \n",
       "1     0.341324  \n",
       "2     0.317922  \n",
       "3     0.317922  \n",
       "4     0.285388  "
      ]
     },
     "execution_count": 185,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examining the dataset \n",
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 186,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check for any duplicacy in the dataset (0: no duplicates)\n",
    "\n",
    "train_data.duplicated(subset=None, keep='first').sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# remove the useless attributes from the dataset \n",
    "\n",
    "# store unique state_name and diseases for label encoding (dont drop them!)\n",
    "states = np.asarray(train_data.state_name.unique())\n",
    "#dis = np.unique(train_data['disease'].values)\n",
    "\n",
    "week       = train_data.pop('week')\n",
    "LOW        = train_data.pop('LastDayWeek')\n",
    "monthN     = train_data.pop('MonthName')\n",
    "state     = train_data.pop('state')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Create test_train splits\n",
    "\n",
    "from sklearn.model_selection import train_test_split as tts\n",
    "\n",
    "y = train_data.pop('disease')\n",
    "train, test, y_train, y_test = tts(train_data, y,test_size = 0.3, shuffle = True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(42427, 9)\n",
      "(42427,)\n",
      "(18183, 9)\n",
      "(18183,)\n"
     ]
    }
   ],
   "source": [
    "# check dimensions\n",
    "\n",
    "print train.shape\n",
    "print y_train.shape\n",
    "\n",
    "print test.shape\n",
    "print y_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Applying ML"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>cases</th>\n",
       "      <th>incidence_per_capita</th>\n",
       "      <th>MonthMax</th>\n",
       "      <th>Year</th>\n",
       "      <th>t_avg</th>\n",
       "      <th>tavg_anomaly</th>\n",
       "      <th>precipitate</th>\n",
       "      <th>pcp_anomaly</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>42427.000000</td>\n",
       "      <td>42427.000000</td>\n",
       "      <td>42427.000000</td>\n",
       "      <td>42427.000000</td>\n",
       "      <td>42427.000000</td>\n",
       "      <td>42427.000000</td>\n",
       "      <td>42427.000000</td>\n",
       "      <td>42427.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>25.680321</td>\n",
       "      <td>0.751023</td>\n",
       "      <td>2.065006</td>\n",
       "      <td>1982.141702</td>\n",
       "      <td>0.646158</td>\n",
       "      <td>0.548762</td>\n",
       "      <td>0.198425</td>\n",
       "      <td>0.308362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>99.047222</td>\n",
       "      <td>2.864421</td>\n",
       "      <td>0.793196</td>\n",
       "      <td>13.045616</td>\n",
       "      <td>0.171159</td>\n",
       "      <td>0.094952</td>\n",
       "      <td>0.124168</td>\n",
       "      <td>0.082830</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1960.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.010000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1972.000000</td>\n",
       "      <td>0.523244</td>\n",
       "      <td>0.492212</td>\n",
       "      <td>0.104899</td>\n",
       "      <td>0.258562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.090000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1980.000000</td>\n",
       "      <td>0.663699</td>\n",
       "      <td>0.548287</td>\n",
       "      <td>0.183417</td>\n",
       "      <td>0.299087</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>13.000000</td>\n",
       "      <td>0.390000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1991.000000</td>\n",
       "      <td>0.786350</td>\n",
       "      <td>0.607477</td>\n",
       "      <td>0.269472</td>\n",
       "      <td>0.347603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>4176.000000</td>\n",
       "      <td>143.950000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2011.000000</td>\n",
       "      <td>0.990109</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              cases  incidence_per_capita      MonthMax          Year  \\\n",
       "count  42427.000000          42427.000000  42427.000000  42427.000000   \n",
       "mean      25.680321              0.751023      2.065006   1982.141702   \n",
       "std       99.047222              2.864421      0.793196     13.045616   \n",
       "min        0.000000              0.000000      1.000000   1960.000000   \n",
       "25%        1.000000              0.010000      1.000000   1972.000000   \n",
       "50%        3.000000              0.090000      2.000000   1980.000000   \n",
       "75%       13.000000              0.390000      3.000000   1991.000000   \n",
       "max     4176.000000            143.950000      3.000000   2011.000000   \n",
       "\n",
       "              t_avg  tavg_anomaly   precipitate   pcp_anomaly  \n",
       "count  42427.000000  42427.000000  42427.000000  42427.000000  \n",
       "mean       0.646158      0.548762      0.198425      0.308362  \n",
       "std        0.171159      0.094952      0.124168      0.082830  \n",
       "min        0.000000      0.000000      0.000000      0.000000  \n",
       "25%        0.523244      0.492212      0.104899      0.258562  \n",
       "50%        0.663699      0.548287      0.183417      0.299087  \n",
       "75%        0.786350      0.607477      0.269472      0.347603  \n",
       "max        0.990109      1.000000      1.000000      1.000000  "
      ]
     },
     "execution_count": 190,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# examine the test/train data now (uncomment the one you want to examine)\n",
    "\n",
    "train.describe()\n",
    "#test.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# label encoding of useful non-numerical attributes\n",
    "\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "le = LabelEncoder()\n",
    "le.fit(y)\n",
    "\n",
    "y_train = le.transform(y_train)\n",
    "y_test = le.transform(y_test)\n",
    "\n",
    "le2 = LabelEncoder()\n",
    "le2.fit(states)\n",
    "\n",
    "train['state_name'] = le2.transform(train.state_name)\n",
    "test['state_name'] = le2.transform(test.state_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "state_name                int64\n",
       "cases                     int64\n",
       "incidence_per_capita    float64\n",
       "MonthMax                  int64\n",
       "Year                      int64\n",
       "t_avg                   float64\n",
       "tavg_anomaly            float64\n",
       "precipitate             float64\n",
       "pcp_anomaly             float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 194,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check dtypes of each attribute (numst be numerical)\n",
    "\n",
    "test.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training...%s LR\n",
      "LR\n",
      "testing_accuracy 34.0922840015\n",
      "training_accuracy 34.6618898343\n",
      "precision_test:  20.9610078099\n",
      "f1_score:  16.2646528436\n",
      "recall_score:  21.5504614559\n",
      "training...%s KNC\n",
      "KNC\n",
      "testing_accuracy 45.0695704779\n",
      "training_accuracy 92.2030782285\n",
      "precision_test:  51.0736642172\n",
      "f1_score:  51.2168910384\n",
      "recall_score:  51.4040419056\n",
      "training...%s SVC\n",
      "SVC\n",
      "testing_accuracy 49.3867898587\n",
      "training_accuracy 60.2493695053\n",
      "precision_test:  54.6867847223\n",
      "f1_score:  53.7997640592\n",
      "recall_score:  53.8458797092\n",
      "training...%s QDA\n",
      "QDA\n",
      "testing_accuracy 34.3617664852\n",
      "training_accuracy 34.025502628\n",
      "precision_test:  44.7915763401\n",
      "f1_score:  36.2026138485\n",
      "recall_score:  43.9296374291\n",
      "training...%s RFC\n",
      "RFC\n",
      "testing_accuracy 35.0932189408\n",
      "training_accuracy 90.645108068\n",
      "precision_test:  40.9571538163\n",
      "f1_score:  40.3022582389\n",
      "recall_score:  40.0889562749\n",
      "training...%s ADC\n",
      "ADC\n",
      "testing_accuracy 39.8504097234\n",
      "training_accuracy 39.4677917364\n",
      "precision_test:  47.5284800819\n",
      "f1_score:  44.9417314779\n",
      "recall_score:  44.888359898\n",
      "training...%s SGDC\n",
      "SGDC\n",
      "testing_accuracy 34.444261123\n",
      "training_accuracy 34.7891672756\n",
      "precision_test:  19.296605337\n",
      "f1_score:  19.6799181068\n",
      "recall_score:  25.6936528255\n",
      "training...%s DTC\n",
      "DTC\n",
      "testing_accuracy 47.9128856624\n",
      "training_accuracy 48.5445588894\n",
      "precision_test:  58.1908975894\n",
      "f1_score:  49.7755015505\n",
      "recall_score:  51.6967377034\n",
      "training...%s LDA\n",
      "LDA\n",
      "testing_accuracy 33.9327943684\n",
      "training_accuracy 34.2541306244\n",
      "precision_test:  26.6986628804\n",
      "f1_score:  24.1848539531\n",
      "recall_score:  26.6752161954\n",
      "training...%s MNB\n",
      "MNB\n",
      "testing_accuracy 28.4331518451\n",
      "training_accuracy 28.6468522403\n",
      "precision_test:  25.9270831835\n",
      "f1_score:  20.6208173559\n",
      "recall_score:  25.7209288624\n"
     ]
    }
   ],
   "source": [
    "#model training (try to load few classifiers for examining performance and chose the best)\n",
    "\n",
    "from sklearn.linear_model import SGDClassifier, LogisticRegression\n",
    "from sklearn.naive_bayes import MultinomialNB as MNB\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier\n",
    "from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis, LinearDiscriminantAnalysis\n",
    "\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n",
    "\n",
    "classifiers=[\n",
    "    (LogisticRegression(penalty='l1'),\"LR\"),\n",
    "    (KNeighborsClassifier(n_neighbors=3, weights='distance', algorithm = 'auto'),\"KNC\"),\n",
    "    (SVC(),'SVC'),\n",
    "    (QuadraticDiscriminantAnalysis(),'QDA'),\n",
    "    (RandomForestClassifier(max_depth=50, n_estimators=10, max_features=1),'RFC'),\n",
    "    (AdaBoostClassifier(),'ADC'),\n",
    "    (SGDClassifier(),'SGDC'),\n",
    "    (DecisionTreeClassifier(max_depth=5),'DTC'),\n",
    "    (LinearDiscriminantAnalysis(),'LDA'),\n",
    "    (MNB(), 'MNB')\n",
    "]\n",
    "\n",
    "eval_ =[]\n",
    "train_acc, test_acc, name = [], [], []\n",
    "f1_, rec_, prec_ = [], [], []\n",
    "for clf in classifiers:\n",
    "    print \"training...%s\",clf[1]\n",
    "    clf[0].fit(train, y_train)\n",
    "    y_preds = clf[0].predict(test)\n",
    "    y_preds_train = clf[0].predict(train)\n",
    "    name.append(clf[1])\n",
    "    train_acc.append( accuracy_score(y_train, y_preds_train)*100 )\n",
    "    test_acc.append( accuracy_score(y_test, y_preds)*100 )\n",
    "    f1_.append( f1_score(y_test, y_preds, average='macro')*100 )\n",
    "    rec_.append( recall_score(y_test, y_preds, average='macro')*100 )\n",
    "    prec_.append( precision_score(y_test, y_preds, average='macro')*100 )\n",
    "    print clf[1]\n",
    "    print \"testing_accuracy\", accuracy_score(y_test, y_preds)*100\n",
    "    print \"training_accuracy\", accuracy_score(y_train, y_preds_train)*100\n",
    "    print \"precision_test: \", precision_score(y_test, y_preds, average='macro')*100\n",
    "    print \"f1_score: \", f1_score(y_test, y_preds, average='macro')*100\n",
    "    print \"recall_score: \", recall_score(y_test, y_preds, average='macro')*100\n",
    "eval_.append(train_acc)\n",
    "eval_.append(test_acc)\n",
    "eval_.append(f1_)\n",
    "eval_.append(rec_)\n",
    "eval_.append(prec_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEICAYAAABVv+9nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xt809X9+PHXO+n9Tun9wr1AL7QBOoWpTBAdalW+m4ow\nGXNzmxfm5nQ6d/Gynzp1Y3NON8fmFKeC9xsOpyDgHeRSoCmXcimUpjcubXq/pOf3R5NSoKUXmqRJ\nzvPx4NEm+Xw+eSel75yey/uIUgpN0zTN+xjcHYCmaZrmHDrBa5qmeSmd4DVN07yUTvCapmleSid4\nTdM0L6UTvKZpmpfSCV4bskTkaRH5rbvj0DRPJXoevOYsIlIM3KiUWu3uWDTNF+kWvOYWIuLn7hhc\nwVdepzY06QSvOYWI/AcYAbwrInUicpeIKBH5gYgcAj6yH/eqiJSLSI2IfCwimV2u8ZyIPGj//kIR\nOSwid4hIpYiUicgNfYjjchHZKiJWESkRkftPefx8EflcRKrtj3/Pfn+wiCwRkYP22D6133ehiBw+\n5RrFIjLb/v39IvKaiLwgIlbgeyJyjoh8YX+OMhF5UkQCupyfKSIfisgxEakQkV+JSIKINIjI8C7H\nTRGRKhHx7+/PQ/NNOsFrTqGUWggcAq5QSoUBr9gf+gaQDnzTfnsVkAbEAVuAF89w2QQgEkgGfgA8\nJSLDegmlHvguEAVcDtwsInMBRGSk/fn/CsQCJiDfft4fganA14Fo4C6gvbfXbXcV8Jr9OV8EbMDt\nQAwwHbgIuMUeQziwGngfSALGAWuUUuXAOuDaLtddCKxQSrX2MQ7Nx+kEr7na/UqpeqVUI4BS6t9K\nqVqlVDNwP5AjIpE9nNsK/E4p1aqU+i9QB0w405MppdYppXYopdqVUtuB5XR8yAAsAFYrpZbbr3lU\nKZUvIgbg+8BPlVKlSimbUupze4x98YVS6i37czYqpTYrpb5USrUppYqBf3SJIQ8oV0otUUo12d+L\nDfbHlgHXA4iIEZgP/KePMWiaTvCay5U4vhERo4g8IiL77N0ZxfaHYno496hSqq3L7QYg7ExPJiLn\nishae9dGDXBTl+unAvu6OS0GCOrhsb4o6XpDRMaLyEp7V5QVeLgPMQC8DWSIyGjgYqBGKbVxgDFp\nPkgneM2Zupui1fW+BXR0Z8ymo+tllP1+GcQYXgLeAVKVUpHA012uXwKM7eacI0BTD4/VAyGOG/aW\ndewpx5z6uv8O7ALSlFIRwK9OiWFMd4ErpZro6Nq6no7uGd161/pFJ3jNmSroIXnZhQPNwFE6kubD\nToghHDimlGoSkXPo+FBxeBGYLSLXioifiAwXEZNSqh34N/AnEUmy/6UxXUQCgT1AkH3w1h/4DRDY\nhxisQJ2ITARu7vLYSiBRRH4mIoEiEi4i53Z5/Hnge8CV6ASv9ZNO8Joz/R74jYhUA1d38/jzwEGg\nFCgEvnRCDLcAvxORWuBeTgz2opQ6BFwG3AEco2OANcf+8J3ADuAr+2OPAgalVI39mv+yx10PnDSr\npht30vHBUgv8E3i5Swy1dHS/XAGUA0XAzC6Pf0bH4O4WpdTBfr96zafphU6aNsSJyEfAS0qpf7k7\nFs2z6ASvaUOYiHwN+JCOMYRad8ejeRbdRaN5PBEx2xdTnfrvO+6O7WyIyDI65sj/TCd3bSB0C17T\nNM1L6Ra8pmmal3JpIaSYmBg1atQoVz6lpmmax9u8efMRpdSp6y165dIEP2rUKDZt2uTKp9Q0TfN4\nIjKgKbK6i0bTNM1L6QQ/AGFhJ8qf/Pe//2X8+PEcPHiQ+++/n5CQECorK7s9try8nOuuu46xY8cy\ndepULrvsMvbs2ePS2DVN8x06wZ+FNWvWcNttt7Fq1SpGjhwJQExMDEuWLDntWKUU//d//8eFF17I\nvn372Lx5M7///e+pqKhwddiapvkIneAH6OOPP+aHP/whK1euZOzYEzWpvv/97/Pyyy9z7Nixk45f\nu3Yt/v7+3HTTTZ335eTkcMEFF7gsZk3TfItO8APQ3NzM3Llzeeutt5g4ceJJj4WFhfH973+fv/zl\nLyfdX1BQwNSpU10ZpqZpPk4n+AHw9/fn61//Os8880y3j992220sW7aM2lq9+FDTNPfRCX4ADAYD\nr7zyChs3buThh0+vcBsVFcWCBQt46qmnOu/LzMxk8+bNrgzTpxiNRkwmE1lZWVxxxRVUV1cDUFxc\nTHBwMCaTqfNfS0sLAKtWrSI3N5eMjAwmT57MHXfc4c6XoGmDTif4fth88DhbDx0HICQkhPfee48X\nX3yx25b8z3/+c/7xj3/Q1taxAdGsWbNobm5m6dKlncds376dTz75xDXBe7ng4GDy8/MpKCggOjr6\npA/XsWPHkp+f3/kvICCAgoICFi9ezAsvvEBhYSGbNm1i3LhxbnwFmjb4dILvh1+9sYOfrsjvvB0d\nHc3777/Pgw8+yDvvvHPSsTExMfzf//0fzc0d23iKCG+++SarV69m7NixZGZmcs8995CQkODS1+AL\npk+fTmlp6RmPeeyxx/j1r3/dOYZiNBq5+eabz3iOpnkalxYby83NVZ66krWp1Ubmff/D1q54Z/F5\nZKdEuTskrYuwsDDq6uqw2Wxcd911/OAHP2DOnDkUFxeTnp7OhAkde3Ofd955PPXUU0yZMoVnn32W\nnJycXq6sae4nIpuVUrn9Pc+lpQo82e7yWmztHR+GK7eX6QQ/xDQ2NmIymSgtLSU9PZ2LL7648zFH\nF42m+RrdRdNHhWVWACbEh7Nym4X2dl1meShx9MEfPHgQpdRJffDd0YPemi/oU4IXkZ+KSIF9Y4Wf\n2e+LFpEPRaTI/nWYc0N1r0KLlfBAP378jTFYaprYWnLc3SFpwLH6Fh7+787O2yEhITzxxBMsWbKk\nc4C7O7/4xS94+OGHO0tFtLe38/TTTzs9Xk1zpV4TvIhkAT8EzqFjQ+I8ERkH/BJYo5RKA9bYb3st\ns6WG9MQILs6IJ8DPwLvbytwdkgb8d0cZSz/ej63LWNLkyZPJzs5m+fLlPZ6XnZ3N448/zvz580lP\nTycrK4v9+/e7ImRNc5m+9MGnAxuUUg0AIrIe+BZwFXCh/ZhlwDrg7sEP0f1s7Ypd5bVcm5tKeJA/\nMyfE8t8dZfw2LwOjQdwdnk8zWzq6zh56c8tJ97/77rud3xcUFHR7bl5eHnl5ec4LTtPcrC9dNAXA\nBSIyXERCgMuAVCBeKeVoxpYD8d2dLCI/EpFNIrKpqqpqUIJ2tYNH62losZGRFAFAXnYSlbXNfFV8\nrJczNWcrtNQAkH+4xs2RaNrQ02uCV0rtBB4FPgDeB/IB2ynHKKDbUUel1FKlVK5SKjc2tt8bkgwJ\njlZipj3BX5QeR7C/kZXbLe4My+e12trZWd5RDmJbSTV6f2FNO1mfBlmVUs8opaYqpWYAx4E9QIWI\nJALYv1ae6RqerLDMir9RSIsLByAkwI9Z6XGs2lFOm63dzdH5rn1VdbS0tTNtTDQ1ja0UH21wd0ia\nNqT0dRZNnP3rCDr6318C3gEW2Q9ZBLztjACHArPFyri4cAL8TrxdV2QncrS+hS/3624adzGXdvxl\nteDcjlr820qq3RmOpg05fZ0H/7qIFALvArcqpaqBR4CLRaQImG2/7ZUKLdbO7hmHCyfEERqgu2nc\nyWyxEuRv4JuZ8QT7G8nXCV7TTtKnlaxKqdN2pVBKHQUuGvSIhpjK2iaO1DWTkXhygg/yN3JJZgKr\nCsr53VVZJ7XuNdcwW2qYmBBBoJ+RSSmROsFr2il0VuqFY4A145QWPEBediI1ja18tveIq8PyeUop\nCsusZCV3/FxMqVEUWqw0t9l6OVPTfIdO8L0oPEOCvyAtloggP97V3TQuV3KskdqmNjKTIgHISYmi\nxdbOrjK9yYqmOegE34vCMiup0cFEBPmf9liAn4FvZibwobmCplbdcnSlAvv8d8fYiGlER/G3bYd1\nN42mOegE34tCi5XMxMgeH8/LSaK2uY2P93jmIi5PZbbUYDQI4+M7pq4mRQYRExZI/iGd4DXNQSf4\nM6hrbqP4aH233TMOXx87nGEh/qzcrmvTuJLZYiUtLowgfyPQsaGKKTWKfN2C17ROOsGfwa4yK0px\n2gyarvyNBuZkJbJ6ZwWNLbqbxlXMFutpH7ym1Ej2V9VT09Dqpqg0bWjRCf4MHDXgM5N7TvDQseip\nocXG2t1eu5h3SKmsbaKqtpmspJO7znJSO/rht5fqVrymgU7wZ1RosTIsxJ+EiKAzHnfumOHEhAXy\n7jY9m8YVTq0N5ODYZUuvaNW0DjrBn4GjG0DkzCWBjQbh8kkJfLSrkrrmnjeZ0AaHubRjBs2pXTSR\nwf6MiQ3VC540zU4n+B602trZXVHbOc+6N3k5STS3tbNmZ4WTI9PMFisjh4cQ3s3UVVNqFPklNbqy\npKahE3yP9lfV09LWfsYB1q6mjhhGQkSQ3unJBczd1AZyMKVGcaSuGUtNk4uj0rShRyf4HphPWUjT\nG4NBuDw7kY/3VFHTqGdxOIu1qZVDxxp6/Msqx94Pr+fDa5pO8D0qtFgJ9DMwOia0z+fkZSfSYmvn\nw0LdTeMshT0MsDqkJ0YQYDToFa2ahk7wPTJbrExMCMfP2Pe3yJQaRcqwYF1C2IlOzKDpvgUf4Gcg\nIylCt+A1DZ3gu+WoVJjRxwFWB5GObppPi45wvL7FSdH5NrOlhrjwQGLDA3s8xpQaxY7SGr3blubz\ndILvhqWmiZrG1jOWKOjJFdlJtLUr/mcud0Jkmrm05wFWB1NqFI2tNooq61wUlaYNTTrBd6NznnUf\nZ9B0lZkUwajhIbqEsBM0tdrYW1XX69RVx4pWPR9e83U6wXejsMyKCKQnhvf7XBHhipwkvth3lKra\nZidE57t2l9dia1e9tuBHDQ8hMthfr2jVfJ5O8N0otFgZHRNKSECfdjQ8TV52Eu0K3i/Qc+IHk2OA\nNSv5zC14ESEnNUq34DWfpxN8NzoW0vRvgLWrCQnhpMWF8a4uITyozJYaIoL8SBkW3OuxptQo9lTU\nUq9LR2g+TCf4U9Q0tFJa3Tig/veu8rKT+Kr4GBVWvaJysPS1NhB0lA5uV1BgH0/RNF+kE/wpzGXd\nF7Lqr7ycRJSC93QrflC02drZWdb3v6w6V7TqbhrNh/UpwYvI7SJiFpECEVkuIkEiEi0iH4pIkf3r\nMGcH6wqdm2yfZQt+bGwY6YkRetHTINl/pJ7mtvY+l44YHhZIanSwXtGq+bReE7yIJAO3AblKqSzA\nCFwH/BJYo5RKA9bYb3u8wjJrrwtp+iovO5Eth6o5fLxhECLzbSdqA/V9bCQnJYptJbqLRvNdfe2i\n8QOCRcQPCAEswFXAMvvjy4C5gx+e6xV2sxXcQF2RnQTobprBYC7tqA00NrbvtYFMqVGUVjdSWavH\nQTTf1GuCV0qVAn8EDgFlQI1S6gMgXinlyFzlQHx354vIj0Rkk4hsqqqqGqSwnaOp1cbeyro+dwP0\nZsTwELJTIvWG3IPAbLEyMTGi37WBAN2K13xWX7pohtHRWh8NJAGhInJ912NUx+4K3e6woJRaqpTK\nVUrlxsbGDkLIzrO3so62dkVG4sCnSJ7qiuwkdpTWUHykftCu6WuUUpgtNf3+4M1MisRoEPJLjjsp\nMk0b2vrSHJoNHFBKVSmlWoE3gK8DFSKSCGD/6vE7Tve3BnxfXJ6dCMB7O3QrfqAOH2/E2tTW759L\ncICRiQnhugWv+ay+JPhDwDQRCZGOCcgXATuBd4BF9mMWAW87J0TXKbRYCQ0wMiI6ZNCumRQVzNSR\nw/SG3GdhIAOsDjmpUWw7XE17u97CT/M9femD3wC8BmwBdtjPWQo8AlwsIkV0tPIfcWKcLmG2WElP\njMBg6H0hTX/kZSeyq7yWvbq64YCYLVaMBmFiQv9rA5lSo6htamO/7iLTfFCfRqyUUvcppSYqpbKU\nUguVUs1KqaNKqYuUUmlKqdlKqWPODtaZ2tuVfSHN4HXPOFw2KRER9Jz4ATJbrIyNDSXI39jvc08M\ntOr58Jrv0StZ7Q4da6C+xTZoUyS7io8I4pxR0azcXkbHeLTWH2ZLDVkDrA00NjaM0ACjXtGq+SSd\n4O3MnStYB28GTVd5OUnsraxjd0WtU67vrapqm6mwNg/4g9doELJTovSKVs0n6QRvV1hWg59BSIsP\nc8r1L81KwCDowdZ+OpsBVoec1Ch2lllparUNVlia5hF0grcrtFgZFxc2oH7evogJC+TrY2N0N00/\ndf5ldRZdZ6bUKFptHfvsapov0QnezjyIJQp6kpedyMGjDRSU6kTTV4UWK6nRwUQG+w/4GnqgVfNV\nOsHT0c9bWdt81hUkezMnKwE/g+jZNP1gttSQeZbjIgmRQcRHBOoEr/kcneCh8093Z7fgo0ICuCBN\nd9P0VW1TK8VHG8hKPvufi0lv4af5IJ3gOVED/mxbin2Rl51EaXUjW3Wy6dXOso4ZR2czwOqQkxpF\n8dEGqhtazvpamuYpdIKnowWfHBVMZMjA+3n76uLMeAKMBlZu07VpejOYtYEc/fC6Fa/5Ep3g6Ugk\nzu6ecYgI8ucbE2J5b4dF10fpRUGplZiwQOIigs76WpOSIxHRpYM13+LzCb6hpY0DR+qdUqKgJ3nZ\niVRYm/mq2KOrOzjdQEoE9yQ8yJ9xsWF6wZPmU3w+we8qr0Wps9+DtT9mp8cT5G/QG4GcQXPb4G6+\nAicGWvUAt+YrfD7BOxbSZCY7f4DVITTQj1kT41hVUEabrd1lz+tJ9pR3bL6SNYg/l5zUKI7Vt3D4\neOOgXVPThjKfT/CFFiuRwf4kRZ59P29/5GUncaSuhQ0HdDdNd5yx+YpjoFXPYNJ8hU7wlhoyEiPo\n2MvEdWZOiCM0wKgXPfXAbLESHuhH6rDB23xlQkI4gX4GveBJ8xk+neDbbO3sKq916QCrQ3CAkdkZ\n8awqKKdVd9OcpsBSQ3rS4G6+4m80kJUcqRO85jN8OsEfOFJPc1u7y6ZIniovO4nqhlY+23vELc8/\nVNnaFbvKnPPBa0qNYkdpjf5Q1XyCTyf4wahUeDZmjI8hPMhPz6Y5xYEjdTS22gZlBeupclKjaG5r\nZ3e5rsuveT+fTvCFZVYC/AyMjXVODfjeBPoZuSQjgf+Zy2lu07XKHTpnNjmjBZ9iryyp58NrPsC3\nE7zFyoT4cPyN7nsb8nISqW1q4+M9upvGwWzp+OAdFzf4H7yp0cFEhwaQf0gneM37+WyCV0oN6krJ\ngTp/XAxRIf56Nk0XZksNExOc88ErIuSkROoWvOYTfDbBl1ubON7Q6rb+dwd/o4E5mQmsLqzQW8rR\n8cFbUGp16gevKXUYRZV11Da1Ou05NG0o6DXBi8gEEcnv8s8qIj8TkWgR+VBEiuxfh7ki4MFiLnVs\nsu3eBA8ds2nqW2ys3VXp7lDcrrS6kZrGVjKcMMDqkJMaiVKwo1QXHtO8W68JXim1WyllUkqZgKlA\nA/Am8EtgjVIqDVhjv+0xCsusiMDEIZDgp42JJiYsQM+mwbkDrA45joFWXVlS83L97aK5CNinlDoI\nXAUss9+/DJg7mIE5W6HFyqjhoYQF+rk7FPyMBi7NSmTNrgrqm9vcHY5bmS1WDALpCc5L8MNCAxg1\nPIT8kuNOew5NGwr6m+CvA5bbv49XSjmanOVAfHcniMiPRGSTiGyqqqoaYJiDz1xWMyS6ZxzyshNp\nam1njY930xRaahgbG0ZwgNGpz5OTGqVb8JrX63OCF5EA4Erg1VMfUx31V7utwaqUWqqUylVK5cbG\nxg440MFU09hKybFGtw+wdvW1UdHERwSycptvz6YxW5w7wOpgSo2i3NpEeU2T059L09ylPy34S4Et\nSqkK++0KEUkEsH/1mKbnLhdtst0fBoNw2aRE1u2uwuqjszuO1jVTVtPklBWsp8rRW/hpPqA/CX4+\nJ7pnAN4BFtm/XwS8PVhBOZsrBvIGIi87iRZbOx+aK3o/2Au58ueSkRiBv1H0fHjNq/UpwYtIKHAx\n8EaXux8BLhaRImC2/bZHKCyz7/UZ7toa8L2ZMiKK5Khgn1305MraQEH+RtITI/SKVs2r9SnBK6Xq\nlVLDlVI1Xe47qpS6SCmVppSarZTymJ0rzBbrkOqecRARLs9O5JOiI1Q3tLg7HJczW2pIjgomKiTA\nJc+Xk9JRWdKmNz/XvJTPrWRtaWtnb6V7asD3RV52Im3tiv+Zy90dissVWqxkJbvu52JKjaKuuY39\nVXUue05NcyWfS/BFlbW02tSQmiLZ1aTkSEYOD/G5RU/1zW0cOFrvkgFWhxy9hZ/m5XwuwQ/VAVYH\nESEvO5HP9x3laF2zu8NxmZ1lVpRy7c9lTEwo4UF+eocnzWv5XIIvtFgJCTAycniou0PpUV52ErZ2\nxaoC3+mmKSh1bLLtuha8wSDkpETpqZKa1/K9BF9mZWJCOMZB3OtzsE1MCGdsbKhPzaYxW6wMDw0g\nPiLQpc+bkxrJrvJaXclT80o+leDb2xU7LVaXthIHoqObJokNB45RYfWNlZaOmU0irv3gNaUOw9be\nsTeApnkbn0rwh483UtvcNiSnSJ7qipxElIL/7vD+wdaWtnaKKmvJSnb9B29OSsdzbtXz4TUv5FMJ\n3tFKG6ozaLoaFxfOxIRwn5hNs6eiY2aTOwa+4yKCSIoMYtth3YLXvI9PJfjCMitGgzAhIdzdofRJ\nXnYimw8ex1Ld6O5QnKqwc2aTe7rOTCOidOlgzSv5VoK3WBkbG0qQv3NL0Q6WvOwkAN7z8lZ8gaWG\nsEA/RkaHuOX5c1KiKDnW6FPTUjXf4FMJ3uwBA6xdjYoJZVJypNfPpjFbrKQnhmNw08wmk33B03bd\nTaN5GZ9J8Efrmim3NnlE/3tXedmJbDtcw6GjDe4OxSls7YqdZe794M1KjsQgekWr5n18JsHvLKsF\nhlYN+L64PDsRgJU7vLMVX3y0noYWm1t/LqGBfoyPD9crWjWv4zMJ3pNm0HSVMiyEySOieHebd/bD\nO0pHZLm568yUGsW2w9V0bE6mad7BZxJ8YZmVpMgghoW6phTtYMrLTmJnmZV9Xlj10GypIcBoIC0+\nzK1x5KRGUd3QykEv7QrTfJPPJPihWgO+Ly6flIgIrPTCVry51Mr4hDD8je79r+gYaNU7PGnexCcS\nfGOLjf1VdWR40AyarhIig/jayGivm02jVEeJgMxE9/9c0uLCCPY36hWtmlfxiQS/u6KWduV5/e9d\n5eUkUlRZx+7y2s77HnroITIzM8nOzsZkMvHAAw9wzz33nHRefn4+6enpANTV1fHjH/+YsWPHMnXq\nVC688EI2bNjg0tfRVVlNE8cbWsl04SYfPfEzGpiUHKlb8JpX8YkE7xhgHao14Pvi0qxEDEJnK/6L\nL75g5cqVbNmyhe3bt7N69WpmzpzJyy+/fNJ5K1asYP78+QDceOONREdHU1RUxObNm3n22Wc5cuSI\ny1+Lw1CrzW8aEYXZYqWlrd3doWjaoPCJBF9osRIe5EfKsGB3hzJgseGBTBsznJXby1BKUVZWRkxM\nDIGBHeV1Y2JimDFjBsOGDTupVf7KK68wf/589u3bx4YNG3jwwQcxGDp+7KNHj+byyy93y+uBjg9e\nEUgfIn9Z5aRE0dLWzq5yq7tD0bRB4RsJvsxKRqLrS9EOtitykjhwpB6zxcoll1xCSUkJ48eP55Zb\nbmH9+vUAzJ8/nxUrVgDw5ZdfEh0dTVpaGmazGZPJhNE4dMo0mC1WxsSEEhLg5+5QgI7a8ICeD695\nDa9P8LZ2xa6yWo8qUdCTOZkJ+BmEldvLCAsLY/PmzSxdupTY2FjmzZvHc889x7x583jttddob28/\nqXtmKCocYqUjkqOCiQkL1CtaNa/RpwQvIlEi8pqI7BKRnSIyXUSiReRDESmyfx3m7GAH4sCRehpb\n3btScrAMCw3gvHExrNxuQSmF0Wjkwgsv5IEHHuDJJ5/k9ddfJzU1ldGjR7N+/Xpef/115s2bB0Bm\nZibbtm3DZhsaOxcdr2+htLpxyPS/Q8dGK6bUSN2C17xGX1vwfwHeV0pNBHKAncAvgTVKqTRgjf32\nkOOpK1h7kpedyOHjjby17iuKioo678/Pz2fkyJFARzfN7bffzpgxY0hJSQFg7Nix5Obmct9993Wu\n1iwuLua9995z/Yug6wDr0GnBQ8d8+H1V9VibWt0diqadtV4TvIhEAjOAZwCUUi1KqWrgKmCZ/bBl\nwFxnBXk2CsusBBgNjItz70rJwXJJZgIBRgOrthazaNEiMjIyyM7OprCwkPvvvx+Aa665BrPZfFr3\nzL/+9S8qKioYN24cWVlZfO973yMuLs4Nr2LozmzKcVSWLNGVJTXP15fRrdFAFfCsiOQAm4GfAvFK\nKcfSynIgvruTReRHwI8ARowYcdYB91ehxUpafBgBft4x3BAZ7M+crATW7jawYd3H3Q5QxsTE0Np6\negs0IiKCf/7zn64Is1dmi5XkqOAhVzoiO+XEitbz02LcHI2mnZ2+ZD0/YArwd6XUZKCeU7pjVMff\n/N1WaVJKLVVK5SqlcmNjY8823n5RStkH8oZWK/FsLZw+ktqmNt7J99yVrWZLzZAcF4kM9mdMbCj5\nuh9e8wJ9SfCHgcNKKcfk6tfoSPgVIpIIYP9a6ZwQB66ytpmj9S1e0//ukDtyGBMTwnn+i4MeWf2w\noaWN/Ufqh+wHryklivwSXVlS83y9JnilVDlQIiIT7HddBBQC7wCL7PctAt52SoRnwbHXp6fWoOmJ\niLBw+kgKy6xs8cDaKTvLrCg19AZYHUwjoqiqbaaspsndoWjaWelrx/RPgBdFZDtgAh4GHgEuFpEi\nYLb99pDiGMhLT/SMTbb7Y64pmbBAP1748qC7Q+m3oVai4FQ59n543U2jebo+JXilVL69Hz1bKTVX\nKXVcKXVUKXWRUipNKTVbKXXM2cH2V2GZlZHDQwgP8nd3KIMuNNCPb09J5r3tZR63WbS51MqwEH8S\nI4PcHUq3JiaGE2A06PnwmsfzjqklPTBbrF7X/97V9dNG0mJr55VNh90dSr+Yy2rISo4csqUjAv2M\nZCRF6BZn63XfAAAgAElEQVS85vG8NsHXNnXszjNUuwEGQ1p8ONPGRPPihoPY2j1jQLDV1s6e8roh\nOYOmK1NqFDtKazzmfdW07nhtgt9V7pmbbPfXwmmjOHy8kfV7htwkpm4VVdTRYmsfsgOsDqbUKBpa\nbBRV1vZ+sKYNUV6b4M2ljpWSQzuRnK1LMuOJCw/k+S88Y7C1YIiuYD2VY0VrvgfOUtI0B69N8IVl\nVoaHBhAXHujuUJzK32hg/jkjWL+nioNH690dTq8KLVZCAoyMHh7q7lDOaNTwECKD/fUOT5pH8+oE\nn5Hk+TXg+2L+OSMwiPDShkPuDqVXZksN6YkRGAxD++ciIuSkRpGva9JoHswrE7ynDOQNloTIIC7J\niOflTSU0tQ6NcsDdaW/3rNIRppRIdpdbaWhpc3comjYgXpng91Z2DOR58xTJUy2cNpLqhlbe217W\n+8FucvBYA/UtNrI8ZFzENCKKdgUFpXoLP80zeWWCH+orJZ1h+tjhjIkN5T9DeGVrZ21+D/m5dFaW\n1PPhNQ/llQm+0GIlyN/A6BjvqAHfFyLCwmkjyS+pZsfhodlvbLZY8TcK4+M9o3RETFggKcOC9YIn\nzWN5Z4Ivq2FiQgTGIT6QN9i+NSWFYH8j//my2N2hdKugtIa0uHCPqs1vSo3SCV7zWJ7zm9ZH3loD\nvi8ig/2ZOzmZt/Mt1DQMrS3nPPXnYkqNorS6kapaz6r3o2nghQn+8PFGrE1tHtPPO9iunzaC5rZ2\nXt1c4u5QTlJh7ajN74kJHnQ/vOaZvC7BF5bZa8D70AyarjKTIpk6chgvbjhE+xCqo+IYYM1K9owZ\nNA6ZSZEYDaIXPGkeyesSvNlixSAwMcE3Ezx0TJk8cKSez/YdcXconcwWKyKQ7mEfvMEBRibEh+t+\neM0jeV2CL7RYGRMbRnCA0d2huM2lkxKIDg3gP0OoPo3ZUsPo4aGEBvZln/ehxTQiim0l1UPqLyJN\n6wsvTPA1Pts94xDoZ2Te11JZvbOC0upGd4cDdCwW8tRxEVNKFNamNg54QK0fTevKqxL88foWLDVN\nHjeQ5wwLzhmBApYPgfo01Q0tlFY3emxlT9MIPdCqeSavSvA7HQOsOsGTGh3CRRPjWPHVIVra2t0a\nS6GHryweGxtGaIBRJ3jN43hVgneUKPD1LhqH66eN5EhdC++by90ah6eXjjAahEkpkXqgVfM4XpXg\nC8usJEQEMTzMu2vA99WMtFhGRIfwgpsHW82WGhIjPfvnYkodRmGZlea2oVutU9NO5V0J3uK5A3nO\nYDAI108bwcbiY+wqd19FRLMHrmA9lSk1klab6uxu0jRP0KcELyLFIrJDRPJFZJP9vmgR+VBEiuxf\nhzk31DNrarWxt6rO4xPJYLtmaioBfgZecFOVycYWG/uq6sjw0AFWB1Nqx39v3Q+veZL+tOBnKqVM\nSqlc++1fAmuUUmnAGvttt9lTUYutXen+91MMCw3giuwk3txSSm2T6+vT7Cy30q48t//dISEyiPiI\nQLYN0UqdQ5XRaMRkMpGZmUlOTg5Lliyhvb2d//3vf5hMJkwmE2FhYUyYMAGTycR3v/tdADZu3MiM\nGTOYMGECkydP5sYbb6ShocHNr8bznM2qk6uAC+3fLwPWAXefZTwD1jnA6uGJxBkWTh/J61sO8+bW\nUr47fZRLn9vTB1i7yknRlSX7Kzg4mPz8fAAqKytZsGABVquVBx54gG9+85sAXHjhhfzxj38kN7ej\n7VhRUcE111zDihUrmD59OgCvvfYatbW1hISEuOeFeKi+tuAVsFpENovIj+z3xSulHNsHlQPxgx5d\nPxRarIQH+pE6TP8HOJUpNYrslEj+88VBlHLtasxCSw1RIf4kRwW79HmdwTQiigNH6qluaHF3KB4p\nLi6OpUuX8uSTT57x/+FTTz3FokWLOpM7wNVXX018vFtTjEfqa4I/XyllAi4FbhWRGV0fVB0/rW5/\nYiLyIxHZJCKbqqqqzi7aMygss3rEZs7ucv20kRRV1rHhwDGXPq9jgNUbNj832Xd42q67aQZszJgx\n2Gw2KisrezymoKCAqVOnujAq79WnBK+UKrV/rQTeBM4BKkQkEcD+tdufmFJqqVIqVymVGxsbOzhR\nn8LWrthZpmfQnMkV2UlEBvu7dEu/Vls7u8prPXYF66kmpUQigu6m0TxGrwleREJFJNzxPXAJUAC8\nAyyyH7YIeNtZQfbm4NF6GlpsOsGfQXCAkWumpvC/gnIqrU0uec69lXW0tLV7Rf87QHiQP+Niw/RM\nmrOwf/9+jEYjcXFxPR6TmZnJ5s2bXRiV9+pLCz4e+FREtgEbgfeUUu8DjwAXi0gRMNt+2y18vQZ8\nX31n2kja2hXLN7pmMxBvGmB1yLFv4efqsQxPopTir2uKeHeb5aT7q6qquOmmm1i8ePEZu+wWL17M\nsmXL2LBhQ+d9b7zxBhUVFU6L2Vv1OotGKbUfyOnm/qPARc4Iqr88bTNndxkdE8oFaTG8tPEgt8wc\ni7/RuevczJYagv2NXrX5uSk1itc2H+bw8UZSo/WAfnee/ayYJR/uwc8gNDY2YjKZaG1txc/Pj4UL\nF/Lzn//8jOfHx8ezYsUK7rzzTiorKzEYDMyYMYM5c+a46BV4D88rzt2NQouVcR62mbO7fHf6KH74\n/CbW7KxgTlaiU5/LbLEyMTHcqzY/79zC73C1TvDd+HL/UR76705mTYyjqLKWuAc/4L3bLmBYaECP\n56xbt+60+6ZPn84nn3zixEh9g1dkRG9YCu8qsybGkRwV7PTB1vZ2xU6LlSwvGWB1mJAQTqCfgfxD\nuh/+VOU1TSx+aQsjh4fwl+tM/G3BVI7UtXD7K/l6sxQ38fgEX1nbxJG6Zt3/3kdGg7Dg3BF8tvco\neyvrnPY8JccbqG1u87oPXn+jgazkSL1H6yma22zc9MJmGltsLF04lfAgfyalRPLbKzJYt7uKv6/f\n5+4QfZLHJ/hCvYK1367NTcXfKLy4wXmt+BMDrN7VgoeOFa07Smtotbm3zv5Q8sC7heSXVLPk2hzG\nxZ0YC7v+3BFckZPEkg928/kQ2iPYV3h8gtclCvovNjyQS7MSeW3zYRpa2pzyHAWlNfgZhPEJ3jPA\n6mAaEUVTazt7KmrdHcqQ8PJXh3hpwyFuuXDsaeM6IsLvvzWJUTGh3LY8n8pa10zR1Tp4fIIvLLOS\nGh1MRJC/u0PxKAunj6S2qY238y29HzwAZouVcXFhBPp53+bnjhWt20r0itb8kmp++5aZC9JiuOOS\nCd0eExbox9+/M5W65lZuW76VNv2Xj8t4fILfabHq/vcByB05jIkJ4U6rT9Mx8O193TMAqdHBRIcG\nkF9y3N2huNWRumZufmEzcRGBPHHd5DPOlpqQEM5Dcyfx5f5j/Hn1HhdG6ds8OsHXN3fsdO+ticSZ\nRISF00dSWGZlyyDPCKm0dgx8ZyV75weviJCTEunTLfg2WzuLX9rCsfoWnr5+6hmnQTp8e2oK130t\nlafW7mPtrp5r0WiDx6MT/K5yK0rpFawDNdeUTFig36BvBuLNA6wOOalR7Kmspa7ZOWMYQ92j7+/i\ny/3H+P23JpGV3Pef8/1XZpKeGMHtr+RTWt3oxAg18PAE35lIvLSl6GyhgX58e0oy720v42hd86Bd\n12zpaNmmJ3rvymJTahRKwQ4frCz5zjYL//zkAIumj+RbU1L6dW6Qv5G/fWcKbTbFrS9uoaVN98c7\nk0cn+EKLlWEh/iREBLk7FI91/bSRtNjaeWXT4UG7ZkGplVHDQwj34oHvnJQTK1p9ya5yK3e/tp2v\njRrGry/PGNA1RseE8oers8kvqeb3q3YOcoRaV56d4O0lgr2h1ri7pMWHM21MNC98eRDbIK02NJfV\neHX3DHRshThyeIhPrWitaWzlx//ZTHiQH08tmHJWpUEunZTIDeeN4tnPivnvjrLeT9AGxGMTvLfV\nGnenhdNGUVrdyLrdZz/wVdPYSsmxRp9Yl2BKjfKZFnx7u+JnK7ZiqW7k79dPIW4Q/mq+59J0TKlR\n3PXadg4cqR+EKLVTeWyC319VT0tbux5gHQSXZMYTFx44KPVpHCuL+zPw5qlyUqIoq2miwkX19d3p\nL2uKWLu7inuvyGTqyOhBuWaAn4GnvjMFP6Nw8wubaWq1Dcp1tRM8NsEXlnUMbvlCS9HZ/I0G5p8z\ngvV7qjh49OxaUo4BVm+rQdMd04iOfnhv3+FpdWEFf1lTxNVTU7j+3BGDeu3kqGD+PM/ErvJa7n/H\nPKjX1jw4wZtLrQT6GRgTE+ruULzC/HNGYBDhpQ2Hzuo6hRYr8RGBxIQFDlJkQ1dGYgR+BvHqHZ4O\nHKnn9pfzmZQcyYNzs5wy3jVzQhy3zhzLiq9KeH3z4A32ax6c4AvLrExMCMfPyZtW+IqEyCAuyYjn\n5U0lZ/WncoHF+wdYHYL8jaQnRnhtC76+uY0f/2cTfkbh79dPIcjfeWUnbp89nmljovn1WzvYXa5r\n/AwWj8yOSinMFisZPpJIXGXhtJFUN7Ty3vaBzWpoarWxr6reJ7pnHEypUWw/XON19c6VUtz1+nb2\nVtbx1/lTSBnm3M1N/IwGnpg/mbBAf25+cbPPLiAbbB6Z4C01TdQ0tur+90E2fexwxsSG8vwAB1t3\nlddia1c+leBzUqOoa25j/xHn1dZ3h39+sp/3tpdx95yJnJ8W45LnjAsP4q/zJ1N8pJ573tih970d\nBB6Z4DtrwOsZNINKRFg4bSTbSqrZPoDpfycGWH3nLyvHFn5bvWg+/Gd7j/DIql1cNimBH80Y49Ln\nnj52OHdcMoF3t1l44SzHgzQPTfBmSw0i3r0U3l2+NSWFYH/jgOrTmC1WIoP9SRkW7ITIhqYxMaGE\nB/qdNB/+rbfeQkTYtWsXAMXFxQQHBzN58mTS09M555xzeO655066zqpVq8jNzSUjI4PJkydzxx13\nuPJldDp8vIHFL21hbGwYj12d45ZFhDd/YywzJ8Ty/94tHFBDQzvBIxN8ocXK6JhQQgK8Ys/wISUy\n2J+5k5N5O99CTUNrv84120s3+9LKYoNByE49ubLk8uXLOf/881m+fHnnfWPHjmXr1q3s3LmTFStW\n8Pjjj/Pss88CUFBQwOLFi3nhhRcoLCxk06ZNjBs3zuWvpanVxs0vbKHNpvjHwqmEBbrn98tgEP50\nrYnY8EBueXFLv/8faif0OcGLiFFEtorISvvtaBH5UESK7F+HOS/MkxWW6RrwznT9tBE0t7Xz6uaS\nPp/TZmtnV5lvbn5uSo1iZ5mVplYbdXV1fPrppzzzzDOsWLGi2+PHjBnDn/70J5544gkAHnvsMX79\n618zceJEAIxGIzfffLPL4oeOQdXfvlXAjtIa/jTPxJhY9+7ENSw0gCcXTKbC2sQdr27T/fED1J8W\n/E+BrpWBfgmsUUqlAWvst52upqGVw8cbfaqf19UykyKZOnIYL2441OfZIfuq6mlua/fJyp45KVG0\ntXfM7Hr77beZM2cO48ePZ/jw4WzevLnbc6ZMmdLZhVNQUMDUqVNdGfJpXtxwiFc3H+a2WeO4OCPe\nrbE4TB4xjHsuTWf1zgqWfrzf3eF4pD4leBFJAS4H/tXl7quAZfbvlwFzBze07hWW6T1YXWHhtJEc\nOFLPZ33cKNkXB1gdHAOt+SXVLF++nOuuuw6A66677qRumq6GUot088HjPPCumZkTYvnZ7PHuDuck\nN5w3ikuzEnjsf7v5qviYu8PxOH1twT8O3AV0Ld4cr5RyTJguB1zyse9IJLqLxrkunZRAdGgAz3/R\nt8FWs8V3VxbHRQSRFBnEhp0H+eijj7jxxhsZNWoUf/jDH3jllVe6TeZbt24lPT0dgMzMzB5b+s5W\nWdvEzS9sJikqmMfnTcZwhm333EFEePTqbFKHBbP4pS0cGcR9C3xBrwleRPKASqVUj/8DVcf/4G6b\nJCLyIxHZJCKbqqqqBh6pXWGZlbjwQGLDvX8pvDsF+hmZ97VU1uys6NPOO2ZLDemJET67sjgnNYo1\nq95h3vzvcPDgQYqLiykpKWH06NGUlJw8llFcXMydd97JT37yEwB+8Ytf8PDDD7NnT8depe3t7Tz9\n9NNOj7nV1s6tL26htqmNp6+fSmTI0KzfHxHkz9++M5XqhlZ+tiJ/0Mpa+4K+/DaeB1wpIsXACmCW\niLwAVIhIIoD9a7e1ZpVSS5VSuUqp3NjY2LMOuNBi1d0zLrLgnBEoYHkv85GVUhRafHOA1eGbmQmU\nbl7N6saR/PrNHRRVdCy3//a3v83vf/979u3b1zlN8tprr+W2227jhhtuACA7O5vHH3+c+fPnk56e\nTlZWFvv3O7/P+aH3dvJV8XEe+fYk0of4X8QZSRH87qpMPt17hCfWFLk7HI/R6zwopdQ9wD0AInIh\ncKdS6noR+QOwCHjE/vVtJ8YJQHObjb2VdVyUHufsp9KA1OgQLpoYx4qvDnHbRWk9bvBw+Hgj1qY2\nn+x/d5g7OZktX3zKss+LeW3zYV7ccIjzx8XwvdnzuHXxTzD20vWRl5dHXl6ei6KFN7Yc5rnPi/nB\n+aO5ypTssuc9G9fmprLhwDGe+KiI3FHDuCDt7BuM3u5s/p5+BLhYRIqA2fbbTlVUUUdbuyIj0XcT\niatdP20kR+paeN9c3uMxBaW+UyL4TDKSInj06my+uOci7pozgX1Vddz4/CZm/nEdz3x6AGvT0JjP\nXVBawz1v7GDamGjuuXSiu8PpMxHhwblZpMWF8bMV+ZTXeH8d/rPVrwSvlFqnlMqzf39UKXWRUipN\nKTVbKeX0Ie7OEgU+nkhcaUZaLCOiQ3jhDIOtZosVo0GYkKBXFgNEhwZwy4Xj+PiumTy1YArxEYH8\nv5WFTHt4Dfe+XcDeSvfVrTle38JNL2xmWEgATy6Y4nFjJiEBfvztO1NparWx+KUttNr0pt1n4lE/\nXbOlhtAAIyOjnVvZTjvBYBCunzaCjcXH2FVu7fYYs6WGcbFhTi0n64n8jQYuz07k1Zu+zsqfnM9l\nkxJZsbGE2X9az3f/vZG1uypdWoXS1q64bcVWKq3NPL1wqsfW7B8XF8bvv53NpoPH+cP/drs7nCHN\noxJ8YZmV9MSIITeVy9tdMzWVAD8D/+mhFW+2WH1ygVN/ZCVH8sdrcvj8nlncecl4dpdbueG5r5i1\nZB3PfnaAWhd03yz5YDefFB3hd1dlds7d91RX5iSxcNpIln68nw/O0H3o6zwmwbe365ka7jIsNIAr\nspN4c2vpaYmoqraZytpmnx5g7Y+YsEAWz0rj07tn8cT8yUSHBvDAux3dN/e/Y3ba5tPvF5Txt3X7\nmH9OKtedM7jb7rnLb/LSmZQcyR2vbuPQ0QZ3hzMkeUyCP3SsgfoWm+5/d5OF00fS0GLjza2lJ93v\nS3uwDiZ/o4Erc5J445bzePvW8/hmZgIvbjjIzD+u44ZnN7J+T9Wgdd/srazjjle2kZMaxf1XZg7K\nNYeCQD8jf/vOFAS49aUtetPubnhMgi8ss3LoT1efNoPm/vvvJzk5GZPJREZGRo9Lw7WzY0qNIjsl\nkv98cfCklZlmPfB91nJSo/jTPBOf/XIWt88eT4HFyqJ/b2T2n9fz/BfFA97d6KGHHiI9IwOTKYcD\n/7yVH01sw4jiV7/6FWlpaZhMJkwmEw899FDnOUajEZPJRGZmJjk5OSxZsoT29hMDmRs3bmTGjBlM\nmDCByZMnc+ONN9LQ4L7Wc2p0CEuuNbGjtIYH3yt0WxxDlcckeEdLMS3+9Cp3t99+O/n5+bz99tv8\n+Mc/prV1aExH8zbXTxtJUWUdGw6cmDBlttQwIjqEiKChuQrSk8SFB/HT2Wl8dvcsHp9nIjzQj3vf\nNjP94TX87t1CDh7te/fNF198wcqVK/n6Hf8i/nt/5Y13VzE5PY3f/OY3WCwWduzYQX5+Pp988slJ\nvy/BwcHk5+djNpv58MMPWbVqFQ888AAAFRUVXHPNNTz66KPs3r2brVu3MmfOHGpr3buH6sUZ8fx4\nxhhe+PIQb+eXdntMWNjpeaNr4zAtLY1vfetbFBae/CFx5MgR/P39XbKy2Bk8JsEXWqwYRM44UyMt\nLY2QkBCOHz/uwsh8xxXZSUQG+/OfLpuBmC1WRgY1ctVVV5GWlsaYMWNYvHgxzc3NrFu3jsjISCZP\nnsyECROYMWMGK1euPO26JpOps0CXBgF+ho6a/IvP581bvs6s9Die/6KYC/+4jhuXfcWnRUd6LVZW\nVlZGHcGsKTrOry5LZ07ueKKiovjnP//JX//6V4KCggAIDw/n/vvv7/YacXFxLF26lCeffBKlFE89\n9RSLFi1i+vTpncdcffXVxMe7v/rknd+cwNdGDeOeN3awt7LvHziOxmFRURHz5s1j1qxZdC2p8uqr\nrzJt2jSP7RnwnARfZqW3yTNbtmwhLS2NuDi90tUZggOMXDM1hf8VlFNpbcLa1ErxkXo++ds9zJ07\nl6KiIoqKimhsbOSuu+4C4IILLmDr1q3s3r2bJ554gsWLF7NmzZrOa+7cuRObzcYnn3xCfb1zBhg9\n2eQRw/jLdZP57Jez+MnMcWw9VM31z2zgkj9/zAtfHqShpfvum+DRk9m9v5iaZbew6cXHWL9+PXv3\n7mXEiBGEh/d9vcKYMWOw2WxUVlYOibLGPfE3Gvjr/CkE+xu55cUtPb4vZzJv3jwuueQSXnrppc77\nli9fzpIlSygtLeXw4cODGbJLeESCP1LXTIW1ucfpkX/+85/JzMzk3HPP5de//rWLo/Mt35k2krZ2\nxfKNJey0WGk6uI2I8JDOuipGo5E///nPPP/889TVnbygx2Qyce+99/Lkk0923rd8+XIWLlzIJZdc\nwttvO73ahceKjwji55dM4PN7ZrHkmhwC/Q385q0Cpj28hofeK6Tk2Il+8JJjDfzynSJm3v0My5//\nN3FxccybN49169addM1nn30Wk8lEamrqaQXRPFFCZBB/uW4yRZV1/OatggGVZO5ap7+kpISysjLO\nOeccrr32Wl5++eXBDtnpPCLBO1aw9tSCv/322zGbzbz++uv84Ac/oKlJL2F2ltExoVyQFsNLGw+y\n7XA1rUcOMf2c3JOOiYiIYNSoUezdu/e087v+AgG8/PLLXHfddcyfP99j/wx2pUA/I9+emsK7i8/n\n9ZunM2N8LP/+rJhv/GEtP3p+E+v3VPHj/2xGKcXSRecw5+KLeOCBB3jyySd59913OXToUGef+Q03\n3EB+fj6RkZHYbN3PQNm/fz9Go5G4uDi3ljXuq/PTYvjpRWm8saWUl7/q/4dW1w+Fl19+mWuvvRY4\nc23/ocwjEry5M8GfuY/myiuvJDc3l2XLlp3xOO3sLJw2kgprM//65ABhQX792hu36y/Qpk2biImJ\nYcSIEVx00UVs3bqVY8f0pg59ISJMHRnNkwum8OndM7n5wrF8VXyMRf/eyM5yK3ecG07LMUvn8fn5\n+UyYMIEf/OAHLF68uLMRZLPZaGlp6fY5qqqquOmmm1i8eDEiwuLFi1m2bBkbNmzoPOaNN96goqLC\nuS+2n34yK40L0mK49x1z5+SMvupap3/58uU899xzjBo1iiuvvJLt27dTVORZlSw9YtfqqtpmUqOD\nOdTQQEpKSuf9P//5z0879t5772XBggX88Ic/xGDwiM8vj3NRejzJUcGUVjcycWI6mze+ctLjVquV\n8vJyJkyYwOrVq0967NRfoF27djFq1KjO815//XV++MMfuuR1eIvEyGB+8c2J/GRWGiu3lxHkbyCx\ntYxFixZRXV2Nn58f48aNY+nSpURGRvLb3/6WrKwswsPDCQ4OZtGiRSQlJQHQ2NiIyWSitbUVPz8/\nFi5c2Pl7Fh8fz4oVK7jzzjuprKzEYDAwY8YM5syZ486XfxqjQXh8nonLn/iUW17cwju3nt+n815/\n/XU++OADlixZwp49e6irq6O09MSsnPvuu4/ly5dz7733Oiv0waeUctm/qVOnqoFqbrUN+Fxt8D35\nUZEaefdK9eiqQjV16lS1bNkypZRSbW1t6sYbb1QPPvigWrt2rbr88ss7z9m2bZsaNWqUWr16tbLZ\nbColJUWVlpZ2Pv7RRx+pmTNnuvy1aN7pqwNH1dh73lOTf/eBEhGVlJysku3/lixZou677z6VlJSk\ncnJy1Lhx49TcuXOV2WxWSil1//33q7vvvvuk623btk1NnDjRHS9FAZvUAHKuKBfuDZmbm6s2bdrk\nsufTnOdIXTMLn9nIg3OziDPUceutt7Jz506qqqqYN28e//jHP1i3bh1XXXUVY8aMoaGhgbi4OO66\n6y6uuOIK1q9fz913382XX37ZeU2bzUZycjJbt24lMTHRja9O8xbbSqr5o70GT1x4IItnjWPe11IJ\n9POswngislkpldv7kaecpxO8Npg+//xz5s+fz5tvvsmUKVPcHY6mAfDl/qP86YM9bCw+RnJUMLdd\nNI5vT0nxmHLJOsFrmqadgVKKT4qOsOSD3Ww7XMOo4SHcfvF48rKTet1xy90GmuA94+NL0zTtLIkI\nM8bH8tat5/HP7+YS5G/kpyvyufQvH/N+QdmA5s0PdTrBa5rmU0SEizPi+e9tF/Dkgsm0tStuemEL\neX/9lLW7Kr0q0esEr2maTzIYhLzsJD742QyWXJODtamVG577im///XM+33vE3eENCt0Hr2maBrTa\n2nl102H++lERZTVNTB8znDsuGU/uqGh3h6YHWTVN0wZDU6uN5RsP8dTafRypa+bCCbHccfEEJqW4\nb9cyneA1TdMGUUNLG89/cZCn1++juqGVb2bGc/vF45mY4PrNbXSC1zRNc4Laplb+/Wkx//pkP3Ut\nbVyRncTPZqcxJvb0TUScxWnTJEUkSEQ2isg2ETGLyAP2+6NF5EMRKbJ/HTaQwDVN04ay8CB/fjo7\njU/unsnN3xjLh4UVzP7Ten7x6raTyjT3hYhw/fXXd95ua2sjNjaWvLw8AJ577jkMBgPbt2/vPCYr\nK7EkOiUAAAYoSURBVAsgwH5+sYjsEJF8+9erzvR8fZlF0wzMUkrlACZgjohMA34JrFFKpQFr7Lc1\nTdO8UlRIAHfNmcgnd8/khvNG8/Y2C7OWrOM3b+2gvKZvJcpDQ0MpKCigsbERgA8//JDk5OSTjklJ\nSTlpn9xuzFRKmYCrgSfOdGCvCd5e68axc4O//Z8CrgIcdXmXAXN7u5amaZqniwkL5Ld5GXz8i5nM\n+1oqL39Vwow/rOX/rSzkSF1zr+dfdtllvPfee0BHRdX58+ef9HheXh5ms5ndu3f3dqkI4Iz7k/Zp\nHryIGEUkH6gEPlRKbQDilVJl9kPKgW43ZhSRH4nIJhHZ1HWvQ03TNE+WEBnEg3Mn8dEdF3JVThLP\nfnaACx5dy2Pv76K6ofsa+9CxeciKFStoampi+/btnHvuuSc9bjAYuOuuu3j44Yd7usRaESkA1gO/\nOVOMfUrwSimb/U+CFOAcEck65XFFR6u+u3OXKqVylVK5sbGxfXk6TdM0j5EaHcIfrsnhw59/g4sz\n4vn7+n1c8Oha/rK6iNqm1tOOz87Opri4mOXLl3PZZZd1e80FCxbw5ZdfcuDAge4enqmUygImAU+K\nSI+jvf1ayaqUqgbWAnOAChFJBLB/rezPtTRN07zJ2Ngwnpg/mVU/vYCvjxvOn1fv4YLH1vL0+n2n\nbQJ+5ZVXcuedd57WPePg5+fHHXfcwaOPPtrj8yml9gEVQEZPx/RlFk2siETZvw8GLgZ2Ae8Ai+yH\nLQL0jsmapvm8iQkR/GNhLu8sPg9TahSPrNrFjMfWnVT+4Pvf/z733XcfkyZN6vE63/ve91i9ejU9\ndW2LSBwwGjjY0zX60oJPpKPPZzvwFR198CuBR4CLRaQImG2/rWmapgHZKVE8d8M5vHbTdDKSIhgd\nG9r5WEpKCrfddtsZzw8ICOC2226jsvK0zpG19jHRtcAvlVI9boqrFzppmqYNcboevKZpmnYSneA1\nTdO8lE7wmqZpXkoneE3TNC+lE7ymaZqX0gle0zTNS+kEr2ma5qV0gtc0TfNSLl3oJCJVnGFZbS9i\nAO/Y6nxw6PfjBP1enEy/HyfzhvdjpFKq39UaXZrgz4aIbBrISi5vpd+PE/R7cTL9fpzMl98P3UWj\naZrmpXSC1zRN81KelOCXujuAIUa/Hyfo9+Jk+v04mc++Hx7TB69pmqb1jye14DVN07R+0Ale0zTN\nS3lEgheROSKyW0T2isgv3R2Pu4hIqoisFZFCETGLyE/dHdNQICJGEdkqIivdHYu7iUiUiLwmIrtE\nZKeITHd3TO4iIrfbf08KRGS5iAS5OyZXG/IJXkSMwFPApXRsLjtfRHrcZNbLtQF3KKUygGnArT78\nXnT1U2Cnu4MYIv4CvK+Umgjk4KPvi4gkA7cBuUqpLMAIXOfeqFxvyCd44Bxgr1Jqv1KqBVgBXOXm\nmNxCKVWmlNpi/76Wjl/eZPdG5V4ikgJcDvzL3bG4m4hEAjOAZ+D/t3P/rFFEcRSG3wOrYGJjqxGy\nhVjHSkxnLMVPoIW9gpWgn0HEzka0MV1MYSFoYR/EPyCaTiXZqJhGBRsVj8VMINgELPa3O3Oebi4M\nnGLmMPfe4YLtn7a/1qYqNQAOSBoAM8DH4jxjNw0FfwTY3HU9ouelBiBpHlgA1mqTlLsFXAX+VAeZ\nAENgG7jXLlndkTS7101dZHsLuAFsAJ+Ab7af1KYav2ko+PiHpIPAA+CK7e/VeapIOgt8sf28OsuE\nGAAngNu2F4AfQC/3rCQdopnpD4HDwKyk87Wpxm8aCn4LOLrreq4d6yVJ+2jKfdn2anWeYovAOUkf\naJbuTku6Xxup1AgY2d6Z1a3QFH4fnQHe2962/QtYBU4VZxq7aSj4Z8AxSUNJ+2k2Sh4WZyohSTTr\nq+u2b1bnqWb7mu052/M0z8VT2737Stth+zOwKel4O7QEvC2MVGkDOClppn1vlujhhvOgOsBebP+W\ndAl4TLMTftf2m+JYVRaBC8BrSa/aseu2HxVmislyGVhuP4beAReL85SwvSZpBXhB8/fZS3p4ZEGO\nKoiI6KhpWKKJiIj/kIKPiOioFHxEREel4CMiOioFHxHRUSn4iIiOSsFHRHTUX4pw9pl4kEQRAAAA\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe1428ac190>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEICAYAAABVv+9nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe1428a0ad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEICAYAAABVv+9nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe142489a10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEICAYAAABVv+9nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe141ec89d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEICAYAAABVv+9nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe1280f1d10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "titles = ['train_accuracy', 'test_accuracy', 'f1_score', 'recall', 'precision']\n",
    "\n",
    "for i in xrange(len(eval_)):\n",
    "    fig = plt.figure()\n",
    "    ax = fig.add_subplot(111)\n",
    "    A = eval_[i][:]\n",
    "    plt.title(titles[i])\n",
    "    plt.plot(A)\n",
    "    for j, label in enumerate(name):\n",
    "        plt.text(j,A[j], label) \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualisations\n",
    "\n",
    "We try to visualize the number of cases of disease outbreak in every state for all states of U.S.for a particular year given as an input in range **1960 - 2011**<br>\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# using plotly for the beautiful plots \n",
    "\n",
    "import plotly.plotly as py\n",
    "import plotly\n",
    "import pandas as pd\n",
    "\n",
    "# login api for plotly (dont forget to sign up to plotly)\n",
    "plotly.tools.set_credentials_file(username= 'ab-bh', api_key ='*********')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# for UI\n",
    "\n",
    "import pandas as pd \n",
    "import numpy as np\n",
    "import plotly.offline as py\n",
    "\n",
    "def get_viz(the_yr_data, yr):\n",
    "    py.init_notebook_mode(connected=True)\n",
    "\n",
    "\n",
    "    for col in the_yr_data.columns:\n",
    "        the_yr_data[col] = the_yr_data[col].astype(str)\n",
    "\n",
    "    scl = [[0.0, 'rgb(242,240,247)'],[0.2, 'rgb(218,218,235)'],[0.4, 'rgb(188,189,220)'],\\\n",
    "                [0.6, 'rgb(158,154,200)'],[0.8, 'rgb(117,107,177)'],[1.0, 'rgb(84,39,143)']]\n",
    "    scl = [\n",
    "            # Let first 10% (0.1) of the values have color rgb(0, 0, 0)\n",
    "            [0, 'rgb(0, 0, 0)'],\n",
    "            [0.1, 'rgb(0, 0, 0)'],\n",
    "\n",
    "            # Let values between 10-20% of the min and max of z\n",
    "            # have color rgb(20, 20, 20)\n",
    "            [0.1, 'rgb(20, 20, 20)'],\n",
    "            [0.2, 'rgb(20, 20, 20)'],\n",
    "\n",
    "            # Values between 20-30% of the min and max of z\n",
    "            # have color rgb(40, 40, 40)\n",
    "            [0.2, 'rgb(40, 40, 40)'],\n",
    "            [0.3, 'rgb(40, 40, 40)'],\n",
    "\n",
    "            [0.3, 'rgb(60, 60, 60)'],\n",
    "            [0.4, 'rgb(60, 60, 60)'],\n",
    "\n",
    "            [0.4, 'rgb(80, 80, 80)'],\n",
    "            [0.5, 'rgb(80, 80, 80)'],\n",
    "\n",
    "            [0.5, 'rgb(100, 100, 100)'],\n",
    "            [0.6, 'rgb(100, 100, 100)'],\n",
    "\n",
    "            [0.6, 'rgb(120, 120, 120)'],\n",
    "            [0.7, 'rgb(120, 120, 120)'],\n",
    "\n",
    "            [0.7, 'rgb(140, 140, 140)'],\n",
    "            [0.8, 'rgb(140, 140, 140)'],\n",
    "\n",
    "            [0.8, 'rgb(160, 160, 160)'],\n",
    "            [0.9, 'rgb(160, 160, 160)'],\n",
    "\n",
    "            [0.9, 'rgb(180, 180, 180)'],\n",
    "            [1.0, 'rgb(180, 180, 180)']\n",
    "        ]\n",
    "    data = [ dict(\n",
    "            type='choropleth',\n",
    "            colorscale = scl,\n",
    "            autocolorscale = True,\n",
    "            locations = the_yr_data['state'],\n",
    "            z = the_yr_data['cases'].astype(float),\n",
    "            zmin=0,\n",
    "            zmax=500,\n",
    "            locationmode = 'USA-states',\n",
    "            text = the_yr_data['text'],\n",
    "            marker = dict(\n",
    "                line = dict (\n",
    "                    color = 'rgb(255,255,255)',\n",
    "                    width = 2\n",
    "                )\n",
    "            ),\n",
    "            colorbar = dict(\n",
    "                title = \"Disease outbreak - cases in %d\" %(yr)\n",
    "            )\n",
    "        ) ]\n",
    "\n",
    "    layout = dict(\n",
    "        title = '%d US Diseases Cases Found by State<br>(Hover for breakdown)' %(yr),\n",
    "        geo = dict(\n",
    "            scope='usa',\n",
    "            projection=dict( type='albers usa' ),\n",
    "            showlakes = True,\n",
    "            lakecolor = 'rgb(255, 255, 255)',\n",
    "            ),\n",
    "    )\n",
    "\n",
    "    fig = dict( data=data, layout=layout )\n",
    "\n",
    "    url = py.iplot( fig, validate=False)\n",
    "\n",
    "def get_1yr_viz(yr):\n",
    "    data = pd.read_csv('for_visulisation.csv')\n",
    "    the_yr_data = data.loc[data['Year'] ==yr]\n",
    "    from collections import defaultdict\n",
    "    har = defaultdict(set)\n",
    "    a = list(the_yr_data.state_name)\n",
    "    b = list(the_yr_data.disease)\n",
    "    #print len(a),len(b)\n",
    "    for i in xrange(len(a)):\n",
    "        har[a[i]].add(b[i])\n",
    "    \n",
    "    the_yr_data['disease_all'] = [' '.join(list(har[i])) for i in the_yr_data.state_name]\n",
    "    \n",
    "    the_yr_data['text'] = the_yr_data['state_name'] + '<br>' +\\\n",
    "    'Disease '+the_yr_data['disease_all']\n",
    "    \n",
    "    tf = the_yr_data.filter(['state_name','state','cases'], axis=1)\n",
    "    tf.cases = tf.cases.astype(int)\n",
    "    the_yr_data_2 = tf.groupby(['state_name','state']).sum().reset_index()\n",
    "    the_yr_data_2['disease_all'] = [' '.join(list(har[i])) for i in the_yr_data_2.state_name]\n",
    "    the_yr_data_2['text'] = the_yr_data_2['state_name'] + '<br>' +\\\n",
    "    'Disease '+the_yr_data_2['disease_all']\n",
    "    \n",
    "    get_viz(the_yr_data_2, yr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<script>requirejs.config({paths: { 'plotly': ['https://cdn.plot.ly/plotly-latest.min']},});if(!window.Plotly) {{require(['plotly'],function(plotly) {window.Plotly=plotly;});}}</script>"
      ],
      "text/vnd.plotly.v1+html": [
       "<script>requirejs.config({paths: { 'plotly': ['https://cdn.plot.ly/plotly-latest.min']},});if(!window.Plotly) {{require(['plotly'],function(plotly) {window.Plotly=plotly;});}}</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "data": [
        {
         "autocolorscale": true,
         "colorbar": {
          "title": "Disease outbreak - cases in 2011"
         },
         "colorscale": [
          [
           0,
           "rgb(0, 0, 0)"
          ],
          [
           0.1,
           "rgb(0, 0, 0)"
          ],
          [
           0.1,
           "rgb(20, 20, 20)"
          ],
          [
           0.2,
           "rgb(20, 20, 20)"
          ],
          [
           0.2,
           "rgb(40, 40, 40)"
          ],
          [
           0.3,
           "rgb(40, 40, 40)"
          ],
          [
           0.3,
           "rgb(60, 60, 60)"
          ],
          [
           0.4,
           "rgb(60, 60, 60)"
          ],
          [
           0.4,
           "rgb(80, 80, 80)"
          ],
          [
           0.5,
           "rgb(80, 80, 80)"
          ],
          [
           0.5,
           "rgb(100, 100, 100)"
          ],
          [
           0.6,
           "rgb(100, 100, 100)"
          ],
          [
           0.6,
           "rgb(120, 120, 120)"
          ],
          [
           0.7,
           "rgb(120, 120, 120)"
          ],
          [
           0.7,
           "rgb(140, 140, 140)"
          ],
          [
           0.8,
           "rgb(140, 140, 140)"
          ],
          [
           0.8,
           "rgb(160, 160, 160)"
          ],
          [
           0.9,
           "rgb(160, 160, 160)"
          ],
          [
           0.9,
           "rgb(180, 180, 180)"
          ],
          [
           1,
           "rgb(180, 180, 180)"
          ]
         ],
         "locationmode": "USA-states",
         "locations": [
          "AL",
          "AK",
          "AZ",
          "AR",
          "CA",
          "CO",
          "CT",
          "DE",
          "DC",
          "FL",
          "GA",
          "HI",
          "ID",
          "IL",
          "IN",
          "IA",
          "KS",
          "KY",
          "LA",
          "ME",
          "MD",
          "MA",
          "MI",
          "MN",
          "MS",
          "MO",
          "MT",
          "NE",
          "NV",
          "NH",
          "NJ",
          "NM",
          "NY",
          "NC",
          "ND",
          "OH",
          "OK",
          "OR",
          "PA",
          "RI",
          "SC",
          "SD",
          "TN",
          "TX",
          "UT",
          "VT",
          "VA",
          "WA",
          "WV",
          "WI",
          "WY"
         ],
         "marker": {
          "line": {
           "color": "rgb(255,255,255)",
           "width": 2
          }
         },
         "text": [
          "ALABAMA<br>Disease PERTUSSIS HEPATITIS A",
          "ALASKA<br>Disease PERTUSSIS HEPATITIS A",
          "ARIZONA<br>Disease PERTUSSIS HEPATITIS A",
          "ARKANSAS<br>Disease PERTUSSIS HEPATITIS A",
          "CALIFORNIA<br>Disease PERTUSSIS HEPATITIS A",
          "COLORADO<br>Disease PERTUSSIS HEPATITIS A",
          "CONNECTICUT<br>Disease PERTUSSIS HEPATITIS A",
          "DELAWARE<br>Disease PERTUSSIS HEPATITIS A",
          "DISTRICT OF COLUMBIA<br>Disease PERTUSSIS",
          "FLORIDA<br>Disease PERTUSSIS HEPATITIS A",
          "GEORGIA<br>Disease PERTUSSIS HEPATITIS A",
          "HAWAII<br>Disease PERTUSSIS HEPATITIS A",
          "IDAHO<br>Disease PERTUSSIS HEPATITIS A",
          "ILLINOIS<br>Disease PERTUSSIS HEPATITIS A",
          "INDIANA<br>Disease PERTUSSIS HEPATITIS A",
          "IOWA<br>Disease PERTUSSIS HEPATITIS A",
          "KANSAS<br>Disease PERTUSSIS HEPATITIS A",
          "KENTUCKY<br>Disease PERTUSSIS HEPATITIS A",
          "LOUISIANA<br>Disease PERTUSSIS HEPATITIS A",
          "MAINE<br>Disease PERTUSSIS HEPATITIS A",
          "MARYLAND<br>Disease PERTUSSIS HEPATITIS A",
          "MASSACHUSETTS<br>Disease PERTUSSIS HEPATITIS A",
          "MICHIGAN<br>Disease PERTUSSIS HEPATITIS A",
          "MINNESOTA<br>Disease PERTUSSIS HEPATITIS A",
          "MISSISSIPPI<br>Disease PERTUSSIS HEPATITIS A",
          "MISSOURI<br>Disease PERTUSSIS HEPATITIS A",
          "MONTANA<br>Disease PERTUSSIS HEPATITIS A",
          "NEBRASKA<br>Disease PERTUSSIS HEPATITIS A",
          "NEVADA<br>Disease PERTUSSIS HEPATITIS A",
          "NEW HAMPSHIRE<br>Disease PERTUSSIS",
          "NEW JERSEY<br>Disease PERTUSSIS HEPATITIS A",
          "NEW MEXICO<br>Disease PERTUSSIS HEPATITIS A",
          "NEW YORK<br>Disease PERTUSSIS HEPATITIS A",
          "NORTH CAROLINA<br>Disease PERTUSSIS HEPATITIS A",
          "NORTH DAKOTA<br>Disease PERTUSSIS",
          "OHIO<br>Disease PERTUSSIS HEPATITIS A",
          "OKLAHOMA<br>Disease PERTUSSIS HEPATITIS A",
          "OREGON<br>Disease PERTUSSIS HEPATITIS A",
          "PENNSYLVANIA<br>Disease PERTUSSIS HEPATITIS A",
          "RHODE ISLAND<br>Disease PERTUSSIS HEPATITIS A",
          "SOUTH CAROLINA<br>Disease PERTUSSIS HEPATITIS A",
          "SOUTH DAKOTA<br>Disease PERTUSSIS HEPATITIS A",
          "TENNESSEE<br>Disease PERTUSSIS HEPATITIS A",
          "TEXAS<br>Disease PERTUSSIS HEPATITIS A",
          "UTAH<br>Disease PERTUSSIS HEPATITIS A",
          "VERMONT<br>Disease PERTUSSIS HEPATITIS A",
          "VIRGINIA<br>Disease PERTUSSIS HEPATITIS A",
          "WASHINGTON<br>Disease PERTUSSIS HEPATITIS A",
          "WEST VIRGINIA<br>Disease PERTUSSIS HEPATITIS A",
          "WISCONSIN<br>Disease PERTUSSIS HEPATITIS A",
          "WYOMING<br>Disease PERTUSSIS HEPATITIS A"
         ],
         "type": "choropleth",
         "z": [
          113,
          28,
          285,
          50,
          1321,
          486,
          59,
          22,
          5,
          419,
          114,
          72,
          150,
          995,
          299,
          214,
          139,
          83,
          18,
          112,
          110,
          127,
          281,
          299,
          46,
          255,
          120,
          60,
          34,
          111,
          223,
          182,
          1022,
          189,
          59,
          790,
          46,
          186,
          511,
          20,
          105,
          30,
          105,
          613,
          286,
          47,
          197,
          759,
          41,
          356,
          8
         ],
         "zmax": 500,
         "zmin": 0
        }
       ],
       "layout": {
        "geo": {
         "lakecolor": "rgb(255, 255, 255)",
         "projection": {
          "type": "albers usa"
         },
         "scope": "usa",
         "showlakes": true
        },
        "title": "2011 US Diseases Cases Found by State<br>(Hover for breakdown)"
       }
      },
      "text/html": [
       "<div id=\"79fa5dd1-1f64-4b18-89df-ae120a591fff\" style=\"height: 525px; width: 100%;\" class=\"plotly-graph-div\"></div><script type=\"text/javascript\">require([\"plotly\"], function(Plotly) { window.PLOTLYENV=window.PLOTLYENV || {};window.PLOTLYENV.BASE_URL=\"https://plot.ly\";Plotly.newPlot(\"79fa5dd1-1f64-4b18-89df-ae120a591fff\", [{\"autocolorscale\": true, \"zmax\": 500, \"colorscale\": [[0, \"rgb(0, 0, 0)\"], [0.1, \"rgb(0, 0, 0)\"], [0.1, \"rgb(20, 20, 20)\"], [0.2, \"rgb(20, 20, 20)\"], [0.2, \"rgb(40, 40, 40)\"], [0.3, \"rgb(40, 40, 40)\"], [0.3, \"rgb(60, 60, 60)\"], [0.4, \"rgb(60, 60, 60)\"], [0.4, \"rgb(80, 80, 80)\"], [0.5, \"rgb(80, 80, 80)\"], [0.5, \"rgb(100, 100, 100)\"], [0.6, \"rgb(100, 100, 100)\"], [0.6, \"rgb(120, 120, 120)\"], [0.7, \"rgb(120, 120, 120)\"], [0.7, \"rgb(140, 140, 140)\"], [0.8, \"rgb(140, 140, 140)\"], [0.8, \"rgb(160, 160, 160)\"], [0.9, \"rgb(160, 160, 160)\"], [0.9, \"rgb(180, 180, 180)\"], [1.0, \"rgb(180, 180, 180)\"]], \"text\": [\"ALABAMA<br>Disease PERTUSSIS HEPATITIS A\", \"ALASKA<br>Disease PERTUSSIS HEPATITIS A\", \"ARIZONA<br>Disease PERTUSSIS HEPATITIS A\", \"ARKANSAS<br>Disease PERTUSSIS HEPATITIS A\", \"CALIFORNIA<br>Disease PERTUSSIS HEPATITIS A\", \"COLORADO<br>Disease PERTUSSIS HEPATITIS A\", \"CONNECTICUT<br>Disease PERTUSSIS HEPATITIS A\", \"DELAWARE<br>Disease PERTUSSIS HEPATITIS A\", \"DISTRICT OF COLUMBIA<br>Disease PERTUSSIS\", \"FLORIDA<br>Disease PERTUSSIS HEPATITIS A\", \"GEORGIA<br>Disease PERTUSSIS HEPATITIS A\", \"HAWAII<br>Disease PERTUSSIS HEPATITIS A\", \"IDAHO<br>Disease PERTUSSIS HEPATITIS A\", \"ILLINOIS<br>Disease PERTUSSIS HEPATITIS A\", \"INDIANA<br>Disease PERTUSSIS HEPATITIS A\", \"IOWA<br>Disease PERTUSSIS HEPATITIS A\", \"KANSAS<br>Disease PERTUSSIS HEPATITIS A\", \"KENTUCKY<br>Disease PERTUSSIS HEPATITIS A\", \"LOUISIANA<br>Disease PERTUSSIS HEPATITIS A\", \"MAINE<br>Disease PERTUSSIS HEPATITIS A\", \"MARYLAND<br>Disease PERTUSSIS HEPATITIS A\", \"MASSACHUSETTS<br>Disease PERTUSSIS HEPATITIS A\", \"MICHIGAN<br>Disease PERTUSSIS HEPATITIS A\", \"MINNESOTA<br>Disease PERTUSSIS HEPATITIS A\", \"MISSISSIPPI<br>Disease PERTUSSIS HEPATITIS A\", \"MISSOURI<br>Disease PERTUSSIS HEPATITIS A\", \"MONTANA<br>Disease PERTUSSIS HEPATITIS A\", \"NEBRASKA<br>Disease PERTUSSIS HEPATITIS A\", \"NEVADA<br>Disease PERTUSSIS HEPATITIS A\", \"NEW HAMPSHIRE<br>Disease PERTUSSIS\", \"NEW JERSEY<br>Disease PERTUSSIS HEPATITIS A\", \"NEW MEXICO<br>Disease PERTUSSIS HEPATITIS A\", \"NEW YORK<br>Disease PERTUSSIS HEPATITIS A\", \"NORTH CAROLINA<br>Disease PERTUSSIS HEPATITIS A\", \"NORTH DAKOTA<br>Disease PERTUSSIS\", \"OHIO<br>Disease PERTUSSIS HEPATITIS A\", \"OKLAHOMA<br>Disease PERTUSSIS HEPATITIS A\", \"OREGON<br>Disease PERTUSSIS HEPATITIS A\", \"PENNSYLVANIA<br>Disease PERTUSSIS HEPATITIS A\", \"RHODE ISLAND<br>Disease PERTUSSIS HEPATITIS A\", \"SOUTH CAROLINA<br>Disease PERTUSSIS HEPATITIS A\", \"SOUTH DAKOTA<br>Disease PERTUSSIS HEPATITIS A\", \"TENNESSEE<br>Disease PERTUSSIS HEPATITIS A\", \"TEXAS<br>Disease PERTUSSIS HEPATITIS A\", \"UTAH<br>Disease PERTUSSIS HEPATITIS A\", \"VERMONT<br>Disease PERTUSSIS HEPATITIS A\", \"VIRGINIA<br>Disease PERTUSSIS HEPATITIS A\", \"WASHINGTON<br>Disease PERTUSSIS HEPATITIS A\", \"WEST VIRGINIA<br>Disease PERTUSSIS HEPATITIS A\", \"WISCONSIN<br>Disease PERTUSSIS HEPATITIS A\", \"WYOMING<br>Disease PERTUSSIS HEPATITIS A\"], \"zmin\": 0, \"locations\": [\"AL\", \"AK\", \"AZ\", \"AR\", \"CA\", \"CO\", \"CT\", \"DE\", \"DC\", \"FL\", \"GA\", \"HI\", \"ID\", \"IL\", \"IN\", \"IA\", \"KS\", \"KY\", \"LA\", \"ME\", \"MD\", \"MA\", \"MI\", \"MN\", \"MS\", \"MO\", \"MT\", \"NE\", \"NV\", \"NH\", \"NJ\", \"NM\", \"NY\", \"NC\", \"ND\", \"OH\", \"OK\", \"OR\", \"PA\", \"RI\", \"SC\", \"SD\", \"TN\", \"TX\", \"UT\", \"VT\", \"VA\", \"WA\", \"WV\", \"WI\", \"WY\"], \"colorbar\": {\"title\": \"Disease outbreak - cases in 2011\"}, \"type\": \"choropleth\", \"marker\": {\"line\": {\"color\": \"rgb(255,255,255)\", \"width\": 2}}, \"z\": [113.0, 28.0, 285.0, 50.0, 1321.0, 486.0, 59.0, 22.0, 5.0, 419.0, 114.0, 72.0, 150.0, 995.0, 299.0, 214.0, 139.0, 83.0, 18.0, 112.0, 110.0, 127.0, 281.0, 299.0, 46.0, 255.0, 120.0, 60.0, 34.0, 111.0, 223.0, 182.0, 1022.0, 189.0, 59.0, 790.0, 46.0, 186.0, 511.0, 20.0, 105.0, 30.0, 105.0, 613.0, 286.0, 47.0, 197.0, 759.0, 41.0, 356.0, 8.0], \"locationmode\": \"USA-states\"}], {\"geo\": {\"showlakes\": true, \"scope\": \"usa\", \"projection\": {\"type\": \"albers usa\"}, \"lakecolor\": \"rgb(255, 255, 255)\"}, \"title\": \"2011 US Diseases Cases Found by State<br>(Hover for breakdown)\"}, {\"linkText\": \"Export to plot.ly\", \"showLink\": true})});</script>"
      ],
      "text/vnd.plotly.v1+html": [
       "<div id=\"79fa5dd1-1f64-4b18-89df-ae120a591fff\" style=\"height: 525px; width: 100%;\" class=\"plotly-graph-div\"></div><script type=\"text/javascript\">require([\"plotly\"], function(Plotly) { window.PLOTLYENV=window.PLOTLYENV || {};window.PLOTLYENV.BASE_URL=\"https://plot.ly\";Plotly.newPlot(\"79fa5dd1-1f64-4b18-89df-ae120a591fff\", [{\"autocolorscale\": true, \"zmax\": 500, \"colorscale\": [[0, \"rgb(0, 0, 0)\"], [0.1, \"rgb(0, 0, 0)\"], [0.1, \"rgb(20, 20, 20)\"], [0.2, \"rgb(20, 20, 20)\"], [0.2, \"rgb(40, 40, 40)\"], [0.3, \"rgb(40, 40, 40)\"], [0.3, \"rgb(60, 60, 60)\"], [0.4, \"rgb(60, 60, 60)\"], [0.4, \"rgb(80, 80, 80)\"], [0.5, \"rgb(80, 80, 80)\"], [0.5, \"rgb(100, 100, 100)\"], [0.6, \"rgb(100, 100, 100)\"], [0.6, \"rgb(120, 120, 120)\"], [0.7, \"rgb(120, 120, 120)\"], [0.7, \"rgb(140, 140, 140)\"], [0.8, \"rgb(140, 140, 140)\"], [0.8, \"rgb(160, 160, 160)\"], [0.9, \"rgb(160, 160, 160)\"], [0.9, \"rgb(180, 180, 180)\"], [1.0, \"rgb(180, 180, 180)\"]], \"text\": [\"ALABAMA<br>Disease PERTUSSIS HEPATITIS A\", \"ALASKA<br>Disease PERTUSSIS HEPATITIS A\", \"ARIZONA<br>Disease PERTUSSIS HEPATITIS A\", \"ARKANSAS<br>Disease PERTUSSIS HEPATITIS A\", \"CALIFORNIA<br>Disease PERTUSSIS HEPATITIS A\", \"COLORADO<br>Disease PERTUSSIS HEPATITIS A\", \"CONNECTICUT<br>Disease PERTUSSIS HEPATITIS A\", \"DELAWARE<br>Disease PERTUSSIS HEPATITIS A\", \"DISTRICT OF COLUMBIA<br>Disease PERTUSSIS\", \"FLORIDA<br>Disease PERTUSSIS HEPATITIS A\", \"GEORGIA<br>Disease PERTUSSIS HEPATITIS A\", \"HAWAII<br>Disease PERTUSSIS HEPATITIS A\", \"IDAHO<br>Disease PERTUSSIS HEPATITIS A\", \"ILLINOIS<br>Disease PERTUSSIS HEPATITIS A\", \"INDIANA<br>Disease PERTUSSIS HEPATITIS A\", \"IOWA<br>Disease PERTUSSIS HEPATITIS A\", \"KANSAS<br>Disease PERTUSSIS HEPATITIS A\", \"KENTUCKY<br>Disease PERTUSSIS HEPATITIS A\", \"LOUISIANA<br>Disease PERTUSSIS HEPATITIS A\", \"MAINE<br>Disease PERTUSSIS HEPATITIS A\", \"MARYLAND<br>Disease PERTUSSIS HEPATITIS A\", \"MASSACHUSETTS<br>Disease PERTUSSIS HEPATITIS A\", \"MICHIGAN<br>Disease PERTUSSIS HEPATITIS A\", \"MINNESOTA<br>Disease PERTUSSIS HEPATITIS A\", \"MISSISSIPPI<br>Disease PERTUSSIS HEPATITIS A\", \"MISSOURI<br>Disease PERTUSSIS HEPATITIS A\", \"MONTANA<br>Disease PERTUSSIS HEPATITIS A\", \"NEBRASKA<br>Disease PERTUSSIS HEPATITIS A\", \"NEVADA<br>Disease PERTUSSIS HEPATITIS A\", \"NEW HAMPSHIRE<br>Disease PERTUSSIS\", \"NEW JERSEY<br>Disease PERTUSSIS HEPATITIS A\", \"NEW MEXICO<br>Disease PERTUSSIS HEPATITIS A\", \"NEW YORK<br>Disease PERTUSSIS HEPATITIS A\", \"NORTH CAROLINA<br>Disease PERTUSSIS HEPATITIS A\", \"NORTH DAKOTA<br>Disease PERTUSSIS\", \"OHIO<br>Disease PERTUSSIS HEPATITIS A\", \"OKLAHOMA<br>Disease PERTUSSIS HEPATITIS A\", \"OREGON<br>Disease PERTUSSIS HEPATITIS A\", \"PENNSYLVANIA<br>Disease PERTUSSIS HEPATITIS A\", \"RHODE ISLAND<br>Disease PERTUSSIS HEPATITIS A\", \"SOUTH CAROLINA<br>Disease PERTUSSIS HEPATITIS A\", \"SOUTH DAKOTA<br>Disease PERTUSSIS HEPATITIS A\", \"TENNESSEE<br>Disease PERTUSSIS HEPATITIS A\", \"TEXAS<br>Disease PERTUSSIS HEPATITIS A\", \"UTAH<br>Disease PERTUSSIS HEPATITIS A\", \"VERMONT<br>Disease PERTUSSIS HEPATITIS A\", \"VIRGINIA<br>Disease PERTUSSIS HEPATITIS A\", \"WASHINGTON<br>Disease PERTUSSIS HEPATITIS A\", \"WEST VIRGINIA<br>Disease PERTUSSIS HEPATITIS A\", \"WISCONSIN<br>Disease PERTUSSIS HEPATITIS A\", \"WYOMING<br>Disease PERTUSSIS HEPATITIS A\"], \"zmin\": 0, \"locations\": [\"AL\", \"AK\", \"AZ\", \"AR\", \"CA\", \"CO\", \"CT\", \"DE\", \"DC\", \"FL\", \"GA\", \"HI\", \"ID\", \"IL\", \"IN\", \"IA\", \"KS\", \"KY\", \"LA\", \"ME\", \"MD\", \"MA\", \"MI\", \"MN\", \"MS\", \"MO\", \"MT\", \"NE\", \"NV\", \"NH\", \"NJ\", \"NM\", \"NY\", \"NC\", \"ND\", \"OH\", \"OK\", \"OR\", \"PA\", \"RI\", \"SC\", \"SD\", \"TN\", \"TX\", \"UT\", \"VT\", \"VA\", \"WA\", \"WV\", \"WI\", \"WY\"], \"colorbar\": {\"title\": \"Disease outbreak - cases in 2011\"}, \"type\": \"choropleth\", \"marker\": {\"line\": {\"color\": \"rgb(255,255,255)\", \"width\": 2}}, \"z\": [113.0, 28.0, 285.0, 50.0, 1321.0, 486.0, 59.0, 22.0, 5.0, 419.0, 114.0, 72.0, 150.0, 995.0, 299.0, 214.0, 139.0, 83.0, 18.0, 112.0, 110.0, 127.0, 281.0, 299.0, 46.0, 255.0, 120.0, 60.0, 34.0, 111.0, 223.0, 182.0, 1022.0, 189.0, 59.0, 790.0, 46.0, 186.0, 511.0, 20.0, 105.0, 30.0, 105.0, 613.0, 286.0, 47.0, 197.0, 759.0, 41.0, 356.0, 8.0], \"locationmode\": \"USA-states\"}], {\"geo\": {\"showlakes\": true, \"scope\": \"usa\", \"projection\": {\"type\": \"albers usa\"}, \"lakecolor\": \"rgb(255, 255, 255)\"}, \"title\": \"2011 US Diseases Cases Found by State<br>(Hover for breakdown)\"}, {\"linkText\": \"Export to plot.ly\", \"showLink\": true})});</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get visualization for year 2011\n",
    "\n",
    "get_1yr_viz(2011)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 219,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<script>requirejs.config({paths: { 'plotly': ['https://cdn.plot.ly/plotly-latest.min']},});if(!window.Plotly) {{require(['plotly'],function(plotly) {window.Plotly=plotly;});}}</script>"
      ],
      "text/vnd.plotly.v1+html": [
       "<script>requirejs.config({paths: { 'plotly': ['https://cdn.plot.ly/plotly-latest.min']},});if(!window.Plotly) {{require(['plotly'],function(plotly) {window.Plotly=plotly;});}}</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.plotly.v1+json": {
       "data": [
        {
         "autocolorscale": true,
         "colorbar": {
          "title": "Disease outbreak - cases in 1970"
         },
         "colorscale": [
          [
           0,
           "rgb(0, 0, 0)"
          ],
          [
           0.1,
           "rgb(0, 0, 0)"
          ],
          [
           0.1,
           "rgb(20, 20, 20)"
          ],
          [
           0.2,
           "rgb(20, 20, 20)"
          ],
          [
           0.2,
           "rgb(40, 40, 40)"
          ],
          [
           0.3,
           "rgb(40, 40, 40)"
          ],
          [
           0.3,
           "rgb(60, 60, 60)"
          ],
          [
           0.4,
           "rgb(60, 60, 60)"
          ],
          [
           0.4,
           "rgb(80, 80, 80)"
          ],
          [
           0.5,
           "rgb(80, 80, 80)"
          ],
          [
           0.5,
           "rgb(100, 100, 100)"
          ],
          [
           0.6,
           "rgb(100, 100, 100)"
          ],
          [
           0.6,
           "rgb(120, 120, 120)"
          ],
          [
           0.7,
           "rgb(120, 120, 120)"
          ],
          [
           0.7,
           "rgb(140, 140, 140)"
          ],
          [
           0.8,
           "rgb(140, 140, 140)"
          ],
          [
           0.8,
           "rgb(160, 160, 160)"
          ],
          [
           0.9,
           "rgb(160, 160, 160)"
          ],
          [
           0.9,
           "rgb(180, 180, 180)"
          ],
          [
           1,
           "rgb(180, 180, 180)"
          ]
         ],
         "locationmode": "USA-states",
         "locations": [
          "AL",
          "AK",
          "AZ",
          "AR",
          "CA",
          "CO",
          "CT",
          "DE",
          "DC",
          "FL",
          "GA",
          "HI",
          "ID",
          "IL",
          "IN",
          "IA",
          "KS",
          "KY",
          "LA",
          "ME",
          "MD",
          "MA",
          "MI",
          "MN",
          "MS",
          "MO",
          "MT",
          "NE",
          "NV",
          "NH",
          "NJ",
          "NM",
          "NY",
          "NC",
          "ND",
          "OH",
          "OK",
          "OR",
          "PA",
          "RI",
          "SC",
          "SD",
          "TN",
          "TX",
          "UT",
          "VT",
          "VA",
          "WA",
          "WV",
          "WI",
          "WY"
         ],
         "marker": {
          "line": {
           "color": "rgb(255,255,255)",
           "width": 2
          }
         },
         "text": [
          "ALABAMA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "ALASKA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "ARIZONA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "ARKANSAS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "CALIFORNIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "COLORADO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "CONNECTICUT<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "DELAWARE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "DISTRICT OF COLUMBIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "FLORIDA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "GEORGIA<br>Disease MEASLES HEPATITIS A MUMPS",
          "HAWAII<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "IDAHO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "ILLINOIS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "INDIANA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "IOWA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "KANSAS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "KENTUCKY<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "LOUISIANA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "MAINE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "MARYLAND<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "MASSACHUSETTS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "MICHIGAN<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "MINNESOTA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "MISSISSIPPI<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "MISSOURI<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "MONTANA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "NEBRASKA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "NEVADA<br>Disease MEASLES HEPATITIS A",
          "NEW HAMPSHIRE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "NEW JERSEY<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "NEW MEXICO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "NEW YORK<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "NORTH CAROLINA<br>Disease MEASLES RUBELLA HEPATITIS A",
          "NORTH DAKOTA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "OHIO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "OKLAHOMA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "OREGON<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "PENNSYLVANIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "RHODE ISLAND<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "SOUTH CAROLINA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "SOUTH DAKOTA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "TENNESSEE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "TEXAS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "UTAH<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "VERMONT<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "VIRGINIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "WASHINGTON<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "WEST VIRGINIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "WISCONSIN<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS",
          "WYOMING<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS"
         ],
         "type": "choropleth",
         "z": [
          1672,
          594,
          3229,
          439,
          22849,
          3122,
          5246,
          815,
          671,
          9495,
          784,
          2143,
          578,
          8977,
          5337,
          6626,
          848,
          5477,
          1077,
          2456,
          3773,
          7780,
          16123,
          1045,
          537,
          2255,
          1426,
          1873,
          158,
          851,
          8032,
          1671,
          8800,
          2019,
          917,
          13360,
          4424,
          3834,
          10405,
          3206,
          2716,
          189,
          6645,
          25363,
          651,
          1180,
          6816,
          12761,
          4684,
          15504,
          274
         ],
         "zmax": 500,
         "zmin": 0
        }
       ],
       "layout": {
        "geo": {
         "lakecolor": "rgb(255, 255, 255)",
         "projection": {
          "type": "albers usa"
         },
         "scope": "usa",
         "showlakes": true
        },
        "title": "1970 US Diseases Cases Found by State<br>(Hover for breakdown)"
       }
      },
      "text/html": [
       "<div id=\"ab4540e7-da42-4886-a8dc-398dc816b0cf\" style=\"height: 525px; width: 100%;\" class=\"plotly-graph-div\"></div><script type=\"text/javascript\">require([\"plotly\"], function(Plotly) { window.PLOTLYENV=window.PLOTLYENV || {};window.PLOTLYENV.BASE_URL=\"https://plot.ly\";Plotly.newPlot(\"ab4540e7-da42-4886-a8dc-398dc816b0cf\", [{\"autocolorscale\": true, \"zmax\": 500, \"colorscale\": [[0, \"rgb(0, 0, 0)\"], [0.1, \"rgb(0, 0, 0)\"], [0.1, \"rgb(20, 20, 20)\"], [0.2, \"rgb(20, 20, 20)\"], [0.2, \"rgb(40, 40, 40)\"], [0.3, \"rgb(40, 40, 40)\"], [0.3, \"rgb(60, 60, 60)\"], [0.4, \"rgb(60, 60, 60)\"], [0.4, \"rgb(80, 80, 80)\"], [0.5, \"rgb(80, 80, 80)\"], [0.5, \"rgb(100, 100, 100)\"], [0.6, \"rgb(100, 100, 100)\"], [0.6, \"rgb(120, 120, 120)\"], [0.7, \"rgb(120, 120, 120)\"], [0.7, \"rgb(140, 140, 140)\"], [0.8, \"rgb(140, 140, 140)\"], [0.8, \"rgb(160, 160, 160)\"], [0.9, \"rgb(160, 160, 160)\"], [0.9, \"rgb(180, 180, 180)\"], [1.0, \"rgb(180, 180, 180)\"]], \"text\": [\"ALABAMA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"ALASKA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"ARIZONA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"ARKANSAS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"CALIFORNIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"COLORADO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"CONNECTICUT<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"DELAWARE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"DISTRICT OF COLUMBIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"FLORIDA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"GEORGIA<br>Disease MEASLES HEPATITIS A MUMPS\", \"HAWAII<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"IDAHO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"ILLINOIS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"INDIANA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"IOWA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"KANSAS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"KENTUCKY<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"LOUISIANA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MAINE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MARYLAND<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MASSACHUSETTS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MICHIGAN<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MINNESOTA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MISSISSIPPI<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MISSOURI<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MONTANA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEBRASKA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEVADA<br>Disease MEASLES HEPATITIS A\", \"NEW HAMPSHIRE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEW JERSEY<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEW MEXICO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEW YORK<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NORTH CAROLINA<br>Disease MEASLES RUBELLA HEPATITIS A\", \"NORTH DAKOTA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"OHIO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"OKLAHOMA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"OREGON<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"PENNSYLVANIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"RHODE ISLAND<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"SOUTH CAROLINA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"SOUTH DAKOTA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"TENNESSEE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"TEXAS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"UTAH<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"VERMONT<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"VIRGINIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"WASHINGTON<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"WEST VIRGINIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"WISCONSIN<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"WYOMING<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\"], \"zmin\": 0, \"locations\": [\"AL\", \"AK\", \"AZ\", \"AR\", \"CA\", \"CO\", \"CT\", \"DE\", \"DC\", \"FL\", \"GA\", \"HI\", \"ID\", \"IL\", \"IN\", \"IA\", \"KS\", \"KY\", \"LA\", \"ME\", \"MD\", \"MA\", \"MI\", \"MN\", \"MS\", \"MO\", \"MT\", \"NE\", \"NV\", \"NH\", \"NJ\", \"NM\", \"NY\", \"NC\", \"ND\", \"OH\", \"OK\", \"OR\", \"PA\", \"RI\", \"SC\", \"SD\", \"TN\", \"TX\", \"UT\", \"VT\", \"VA\", \"WA\", \"WV\", \"WI\", \"WY\"], \"colorbar\": {\"title\": \"Disease outbreak - cases in 1970\"}, \"type\": \"choropleth\", \"marker\": {\"line\": {\"color\": \"rgb(255,255,255)\", \"width\": 2}}, \"z\": [1672.0, 594.0, 3229.0, 439.0, 22849.0, 3122.0, 5246.0, 815.0, 671.0, 9495.0, 784.0, 2143.0, 578.0, 8977.0, 5337.0, 6626.0, 848.0, 5477.0, 1077.0, 2456.0, 3773.0, 7780.0, 16123.0, 1045.0, 537.0, 2255.0, 1426.0, 1873.0, 158.0, 851.0, 8032.0, 1671.0, 8800.0, 2019.0, 917.0, 13360.0, 4424.0, 3834.0, 10405.0, 3206.0, 2716.0, 189.0, 6645.0, 25363.0, 651.0, 1180.0, 6816.0, 12761.0, 4684.0, 15504.0, 274.0], \"locationmode\": \"USA-states\"}], {\"geo\": {\"showlakes\": true, \"scope\": \"usa\", \"projection\": {\"type\": \"albers usa\"}, \"lakecolor\": \"rgb(255, 255, 255)\"}, \"title\": \"1970 US Diseases Cases Found by State<br>(Hover for breakdown)\"}, {\"linkText\": \"Export to plot.ly\", \"showLink\": true})});</script>"
      ],
      "text/vnd.plotly.v1+html": [
       "<div id=\"ab4540e7-da42-4886-a8dc-398dc816b0cf\" style=\"height: 525px; width: 100%;\" class=\"plotly-graph-div\"></div><script type=\"text/javascript\">require([\"plotly\"], function(Plotly) { window.PLOTLYENV=window.PLOTLYENV || {};window.PLOTLYENV.BASE_URL=\"https://plot.ly\";Plotly.newPlot(\"ab4540e7-da42-4886-a8dc-398dc816b0cf\", [{\"autocolorscale\": true, \"zmax\": 500, \"colorscale\": [[0, \"rgb(0, 0, 0)\"], [0.1, \"rgb(0, 0, 0)\"], [0.1, \"rgb(20, 20, 20)\"], [0.2, \"rgb(20, 20, 20)\"], [0.2, \"rgb(40, 40, 40)\"], [0.3, \"rgb(40, 40, 40)\"], [0.3, \"rgb(60, 60, 60)\"], [0.4, \"rgb(60, 60, 60)\"], [0.4, \"rgb(80, 80, 80)\"], [0.5, \"rgb(80, 80, 80)\"], [0.5, \"rgb(100, 100, 100)\"], [0.6, \"rgb(100, 100, 100)\"], [0.6, \"rgb(120, 120, 120)\"], [0.7, \"rgb(120, 120, 120)\"], [0.7, \"rgb(140, 140, 140)\"], [0.8, \"rgb(140, 140, 140)\"], [0.8, \"rgb(160, 160, 160)\"], [0.9, \"rgb(160, 160, 160)\"], [0.9, \"rgb(180, 180, 180)\"], [1.0, \"rgb(180, 180, 180)\"]], \"text\": [\"ALABAMA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"ALASKA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"ARIZONA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"ARKANSAS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"CALIFORNIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"COLORADO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"CONNECTICUT<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"DELAWARE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"DISTRICT OF COLUMBIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"FLORIDA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"GEORGIA<br>Disease MEASLES HEPATITIS A MUMPS\", \"HAWAII<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"IDAHO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"ILLINOIS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"INDIANA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"IOWA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"KANSAS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"KENTUCKY<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"LOUISIANA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MAINE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MARYLAND<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MASSACHUSETTS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MICHIGAN<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MINNESOTA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MISSISSIPPI<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MISSOURI<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"MONTANA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEBRASKA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEVADA<br>Disease MEASLES HEPATITIS A\", \"NEW HAMPSHIRE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEW JERSEY<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEW MEXICO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NEW YORK<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"NORTH CAROLINA<br>Disease MEASLES RUBELLA HEPATITIS A\", \"NORTH DAKOTA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"OHIO<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"OKLAHOMA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"OREGON<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"PENNSYLVANIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"RHODE ISLAND<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"SOUTH CAROLINA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"SOUTH DAKOTA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"TENNESSEE<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"TEXAS<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"UTAH<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"VERMONT<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"VIRGINIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"WASHINGTON<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"WEST VIRGINIA<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"WISCONSIN<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\", \"WYOMING<br>Disease MEASLES RUBELLA HEPATITIS A MUMPS\"], \"zmin\": 0, \"locations\": [\"AL\", \"AK\", \"AZ\", \"AR\", \"CA\", \"CO\", \"CT\", \"DE\", \"DC\", \"FL\", \"GA\", \"HI\", \"ID\", \"IL\", \"IN\", \"IA\", \"KS\", \"KY\", \"LA\", \"ME\", \"MD\", \"MA\", \"MI\", \"MN\", \"MS\", \"MO\", \"MT\", \"NE\", \"NV\", \"NH\", \"NJ\", \"NM\", \"NY\", \"NC\", \"ND\", \"OH\", \"OK\", \"OR\", \"PA\", \"RI\", \"SC\", \"SD\", \"TN\", \"TX\", \"UT\", \"VT\", \"VA\", \"WA\", \"WV\", \"WI\", \"WY\"], \"colorbar\": {\"title\": \"Disease outbreak - cases in 1970\"}, \"type\": \"choropleth\", \"marker\": {\"line\": {\"color\": \"rgb(255,255,255)\", \"width\": 2}}, \"z\": [1672.0, 594.0, 3229.0, 439.0, 22849.0, 3122.0, 5246.0, 815.0, 671.0, 9495.0, 784.0, 2143.0, 578.0, 8977.0, 5337.0, 6626.0, 848.0, 5477.0, 1077.0, 2456.0, 3773.0, 7780.0, 16123.0, 1045.0, 537.0, 2255.0, 1426.0, 1873.0, 158.0, 851.0, 8032.0, 1671.0, 8800.0, 2019.0, 917.0, 13360.0, 4424.0, 3834.0, 10405.0, 3206.0, 2716.0, 189.0, 6645.0, 25363.0, 651.0, 1180.0, 6816.0, 12761.0, 4684.0, 15504.0, 274.0], \"locationmode\": \"USA-states\"}], {\"geo\": {\"showlakes\": true, \"scope\": \"usa\", \"projection\": {\"type\": \"albers usa\"}, \"lakecolor\": \"rgb(255, 255, 255)\"}, \"title\": \"1970 US Diseases Cases Found by State<br>(Hover for breakdown)\"}, {\"linkText\": \"Export to plot.ly\", \"showLink\": true})});</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get visualization for year 1970 \n",
    "get_1yr_viz(1970)"
   ]
  }
 ],
 "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": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}