CDC vs CTP comparison.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as ticker\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "\n",
    "plt.rcParams['font.family'] = 'Helvetica'\n",
    "plt.rcParams['grid.color'] = '#505F69'\n",
    "sns.set_style(\"whitegrid\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read data\n",
    "cdc = pd.read_csv('cdc.csv', skiprows=2)\n",
    "\n",
    "# rename some CDC columns\n",
    "cdc.rename(columns={\n",
    "    'abbr': 'state',\n",
    "    'Total Cases': 'cdc_total_cases',\n",
    "    'Total Death': 'cdc_total_death',\n",
    "    'CasesInLast7Days': 'cdc_cases_last_7_days',\n",
    "    'RatePer100000': 'cdc_rate_per_100000',\n",
    "    'Total Tests Performed': 'cdc_total_tests',\n",
    "    'Percent Positive Range': 'cdc_percent_positive_range',\n",
    "    'cumulative_positive_tests': 'cdc_cumulative_positive_tests'\n",
    "}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th>jurisdiction</th>\n",
       "      <th>cdc_total_cases</th>\n",
       "      <th>cdc_total_death</th>\n",
       "      <th>cdc_total_tests</th>\n",
       "      <th>ctp_total_tests_for_analysis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AK</td>\n",
       "      <td>Alaska</td>\n",
       "      <td>388</td>\n",
       "      <td>10</td>\n",
       "      <td>46589.0</td>\n",
       "      <td>31762.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AL</td>\n",
       "      <td>Alabama</td>\n",
       "      <td>11642</td>\n",
       "      <td>484</td>\n",
       "      <td>168687.0</td>\n",
       "      <td>141985.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AR</td>\n",
       "      <td>Arkansas</td>\n",
       "      <td>4463</td>\n",
       "      <td>98</td>\n",
       "      <td>84496.0</td>\n",
       "      <td>75818.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AZ</td>\n",
       "      <td>Arizona</td>\n",
       "      <td>13169</td>\n",
       "      <td>651</td>\n",
       "      <td>210388.0</td>\n",
       "      <td>134338.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CA</td>\n",
       "      <td>California</td>\n",
       "      <td>74936</td>\n",
       "      <td>3108</td>\n",
       "      <td>924696.0</td>\n",
       "      <td>1104651.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  state jurisdiction  cdc_total_cases  cdc_total_death  cdc_total_tests  \\\n",
       "0    AK       Alaska              388               10          46589.0   \n",
       "1    AL      Alabama            11642              484         168687.0   \n",
       "2    AR     Arkansas             4463               98          84496.0   \n",
       "3    AZ      Arizona            13169              651         210388.0   \n",
       "4    CA   California            74936             3108         924696.0   \n",
       "\n",
       "   ctp_total_tests_for_analysis  \n",
       "0                       31762.0  \n",
       "1                      141985.0  \n",
       "2                       75818.0  \n",
       "3                      134338.0  \n",
       "4                     1104651.0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use CTP 5/14 for comparing test counts\n",
    "\n",
    "ctp_514_tests = pd.read_csv('ctp_514.csv')\n",
    "ctp_514_test_overrides = pd.read_csv('ctp_514_specimen_tests.csv')\n",
    "\n",
    "ctp_514_merged_df = pd.merge(\n",
    "    ctp_514_tests, ctp_514_test_overrides, on=['state', 'date'], how='inner', sort=True, copy=True)\n",
    "\n",
    "def pick_total_tests_for_analysis(row):\n",
    "    # prioritize if present: specimens, people, unclear source\n",
    "    if not pd.isnull(row.total_tests_specimens):\n",
    "        return row.total_tests_specimens\n",
    "    elif not pd.isnull(row.total_tests_people):\n",
    "        return row.total_tests_people\n",
    "    else:\n",
    "        return row.total_tests_unclear\n",
    "\n",
    "ctp_514_merged_df['ctp_total_tests_for_analysis'] = ctp_514_merged_df.apply(pick_total_tests_for_analysis, axis=1)\n",
    "\n",
    "# merge with CTP reported data - only keeping rows for which both CDC and CTP report data\n",
    "merged_df_test_counts = pd.merge(\n",
    "    cdc, ctp_514_merged_df, on=['state', 'fips'], how='inner', sort=True, copy=True)\n",
    "\n",
    "# of this, keep only some interesting columns:\n",
    "# state, jurisdiction, cdc_total_cases, cdc_total_death, cdc_total_tests\n",
    "cols_to_keep = ['state', 'jurisdiction', 'cdc_total_cases', 'cdc_total_death', 'cdc_total_tests',\n",
    "               'ctp_total_tests_for_analysis']\n",
    "merged_df_test_counts = merged_df_test_counts[cols_to_keep]\n",
    "merged_df_test_counts.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th>positive</th>\n",
       "      <th>death</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AK</td>\n",
       "      <td>392.0</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AL</td>\n",
       "      <td>11523.0</td>\n",
       "      <td>485.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AR</td>\n",
       "      <td>4578.0</td>\n",
       "      <td>98.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AZ</td>\n",
       "      <td>13631.0</td>\n",
       "      <td>679.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CA</td>\n",
       "      <td>76793.0</td>\n",
       "      <td>3204.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  state  positive   death\n",
       "0    AK     392.0    10.0\n",
       "1    AL   11523.0   485.0\n",
       "2    AR    4578.0    98.0\n",
       "3    AZ   13631.0   679.0\n",
       "4    CA   76793.0  3204.0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# will use CTP 5/16 for comparing positives/deaths\n",
    "ctp_516_positives_deaths = pd.read_csv('ctp_516.csv')\n",
    "cols_to_keep = ['state', 'positive', 'death']\n",
    "ctp_516_positives_deaths = ctp_516_positives_deaths[cols_to_keep]\n",
    "ctp_516_positives_deaths.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Make a minimal dataframe for analysis containing only final columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_merged_df = pd.merge(\n",
    "    merged_df_test_counts, ctp_516_positives_deaths, on=['state'], how='inner', sort=True, copy=True)\n",
    "# rename some CDC columns\n",
    "full_merged_df.rename(columns={\n",
    "    'positive': 'ctp_total_cases',\n",
    "    'death': 'ctp_total_death',\n",
    "}, inplace=True)\n",
    "full_merged_df.set_index('state', inplace=True)\n",
    "full_merged_df.to_csv('merged_data_for_analysis.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# How many nationwide tests according to CDC and CTP? What are the national positivity rates?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CTP:\n",
      "National positives: 1455954\n",
      "National test count: 10462017\n",
      "National positivity rate: 0.13916570772156076\n",
      "CDC:\n",
      "National positives: 1432265\n",
      "National test count: 10831698\n",
      "National positivity rate: 0.1322290374048464\n"
     ]
    }
   ],
   "source": [
    "# CTP\n",
    "print('CTP:')\n",
    "ctp_us_positives = sum(full_merged_df.ctp_total_cases)\n",
    "ctp_us_tests = sum(full_merged_df.ctp_total_tests_for_analysis)\n",
    "print('National positives: %d' % ctp_us_positives)\n",
    "print('National test count: %d' % ctp_us_tests)\n",
    "print('National positivity rate: %s' % float(ctp_us_positives/ctp_us_tests))\n",
    "\n",
    "# same for CDC\n",
    "print('CDC:')\n",
    "cdc_us_positives = sum(full_merged_df.cdc_total_cases)\n",
    "cdc_us_tests = sum(full_merged_df.cdc_total_tests)\n",
    "print('National positives: %d' % cdc_us_positives)\n",
    "print('National test count: %d' % cdc_us_tests)\n",
    "print('National positivity rate: %s' % float(cdc_us_positives/cdc_us_tests))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_merged_df['test_count_diff'] = \\\n",
    "    full_merged_df['cdc_total_tests'] - full_merged_df['ctp_total_tests_for_analysis']\n",
    "full_merged_df['abs_test_count_diff'] = \\\n",
    "    abs(full_merged_df['cdc_total_tests'] - full_merged_df['ctp_total_tests_for_analysis'])\n",
    "full_merged_df['test_count_percent_diff'] = \\\n",
    "    full_merged_df['test_count_diff']/full_merged_df['ctp_total_tests_for_analysis']\n",
    "full_merged_df['abs_test_count_percent_diff'] = \\\n",
    "    abs(full_merged_df['test_count_diff']/full_merged_df['ctp_total_tests_for_analysis'])\n",
    "# sort by test_count_diff\n",
    "full_merged_df.sort_values('test_count_diff', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>jurisdiction</th>\n",
       "      <th>cdc_total_cases</th>\n",
       "      <th>cdc_total_death</th>\n",
       "      <th>cdc_total_tests</th>\n",
       "      <th>ctp_total_tests_for_analysis</th>\n",
       "      <th>ctp_total_cases</th>\n",
       "      <th>ctp_total_death</th>\n",
       "      <th>test_count_diff</th>\n",
       "      <th>abs_test_count_diff</th>\n",
       "      <th>test_count_percent_diff</th>\n",
       "      <th>abs_test_count_percent_diff</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>state</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>CA</th>\n",
       "      <td>California</td>\n",
       "      <td>74936</td>\n",
       "      <td>3108</td>\n",
       "      <td>924696.0</td>\n",
       "      <td>1104651.0</td>\n",
       "      <td>76793.0</td>\n",
       "      <td>3204.0</td>\n",
       "      <td>-179955.0</td>\n",
       "      <td>179955.0</td>\n",
       "      <td>-0.162907</td>\n",
       "      <td>0.162907</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TX</th>\n",
       "      <td>Texas</td>\n",
       "      <td>45198</td>\n",
       "      <td>1272</td>\n",
       "      <td>454133.0</td>\n",
       "      <td>623284.0</td>\n",
       "      <td>46999.0</td>\n",
       "      <td>1305.0</td>\n",
       "      <td>-169151.0</td>\n",
       "      <td>169151.0</td>\n",
       "      <td>-0.271387</td>\n",
       "      <td>0.271387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC</th>\n",
       "      <td>North Carolina</td>\n",
       "      <td>17129</td>\n",
       "      <td>641</td>\n",
       "      <td>151449.0</td>\n",
       "      <td>219268.0</td>\n",
       "      <td>17982.0</td>\n",
       "      <td>652.0</td>\n",
       "      <td>-67819.0</td>\n",
       "      <td>67819.0</td>\n",
       "      <td>-0.309297</td>\n",
       "      <td>0.309297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NJ</th>\n",
       "      <td>New Jersey</td>\n",
       "      <td>143905</td>\n",
       "      <td>10138</td>\n",
       "      <td>409320.0</td>\n",
       "      <td>451696.0</td>\n",
       "      <td>145089.0</td>\n",
       "      <td>10249.0</td>\n",
       "      <td>-42376.0</td>\n",
       "      <td>42376.0</td>\n",
       "      <td>-0.093815</td>\n",
       "      <td>0.093815</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IL</th>\n",
       "      <td>Illinois</td>\n",
       "      <td>90369</td>\n",
       "      <td>4058</td>\n",
       "      <td>470698.0</td>\n",
       "      <td>512037.0</td>\n",
       "      <td>92457.0</td>\n",
       "      <td>4129.0</td>\n",
       "      <td>-41339.0</td>\n",
       "      <td>41339.0</td>\n",
       "      <td>-0.080734</td>\n",
       "      <td>0.080734</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         jurisdiction  cdc_total_cases  cdc_total_death  cdc_total_tests  \\\n",
       "state                                                                      \n",
       "CA         California            74936             3108         924696.0   \n",
       "TX              Texas            45198             1272         454133.0   \n",
       "NC     North Carolina            17129              641         151449.0   \n",
       "NJ         New Jersey           143905            10138         409320.0   \n",
       "IL           Illinois            90369             4058         470698.0   \n",
       "\n",
       "       ctp_total_tests_for_analysis  ctp_total_cases  ctp_total_death  \\\n",
       "state                                                                   \n",
       "CA                        1104651.0          76793.0           3204.0   \n",
       "TX                         623284.0          46999.0           1305.0   \n",
       "NC                         219268.0          17982.0            652.0   \n",
       "NJ                         451696.0         145089.0          10249.0   \n",
       "IL                         512037.0          92457.0           4129.0   \n",
       "\n",
       "       test_count_diff  abs_test_count_diff  test_count_percent_diff  \\\n",
       "state                                                                  \n",
       "CA           -179955.0             179955.0                -0.162907   \n",
       "TX           -169151.0             169151.0                -0.271387   \n",
       "NC            -67819.0              67819.0                -0.309297   \n",
       "NJ            -42376.0              42376.0                -0.093815   \n",
       "IL            -41339.0              41339.0                -0.080734   \n",
       "\n",
       "       abs_test_count_percent_diff  \n",
       "state                               \n",
       "CA                        0.162907  \n",
       "TX                        0.271387  \n",
       "NC                        0.309297  \n",
       "NJ                        0.093815  \n",
       "IL                        0.080734  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_merged_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sns.set_style(\"whitegrid\")\n",
    "\n",
    "# plt.figure(figsize=(18,7))\n",
    "\n",
    "# def get_color(row):\n",
    "#     if row.test_count_diff > 0:\n",
    "#         return '#ffad4a'\n",
    "#     else:\n",
    "#         return '#585bc1'\n",
    "    \n",
    "# palette = list(full_merged_df.apply(get_color, axis=1))\n",
    "# ax = sns.barplot(x=full_merged_df.index, y='test_count_diff', data=full_merged_df,\n",
    "#                  palette=palette)\n",
    "# ax.set_xticklabels(rotation=45, labels=full_merged_df.index, fontsize=12)\n",
    "# ax.set_ylabel('')\n",
    "# ax.set_xlabel('')\n",
    "# ax.set_title('Test count differences between CDC and CTP data', fontsize=20, pad=10)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test count comparison CTP vs CDC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x504 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# a bit too busy, let's pick a difference threshold - only show states with a >5% difference\n",
    "sns.set_style(\"whitegrid\")\n",
    "\n",
    "plt.figure(figsize=(18,7))\n",
    "\n",
    "def get_color(row):\n",
    "    if row.test_count_diff > 0:\n",
    "        return '#ffad4a'  # orange\n",
    "    else:\n",
    "        return '#585bc1'  # purple\n",
    "\n",
    "data = full_merged_df.loc[full_merged_df.abs_test_count_percent_diff > 0.05]\n",
    "palette = list(data.apply(get_color, axis=1))\n",
    "\n",
    "ax = sns.barplot(x=data.index, y='test_count_diff', data=data, palette=palette)\n",
    "sns.despine(ax=ax, left=True)\n",
    "ax.set_xticklabels(rotation=90, labels=data.jurisdiction, fontsize=14)\n",
    "ax.set_yticklabels(ax.get_yticks(), size=14)\n",
    "ax.set_ylabel('')\n",
    "ax.set_xlabel('')\n",
    "# ax.set_title('States for which CDC and CTP testing counts differ by more than 5%', fontsize=20, pad=10)\n",
    "ax.yaxis.set_major_locator(ticker.MultipleLocator(100000))\n",
    "ax.yaxis.set_major_formatter(ticker.FuncFormatter(lambda x, p: format(int(x), ',')))\n",
    "\n",
    "plt.savefig(\"test_count_differences_over_5_percent.png\", dpi=100, bbox_inches = \"tight\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# How close is CDC and CTP testing data numerically?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Within 1000: 5 states\n",
      "Within 10 percent: 29 states\n",
      "Within 20 percent: 35 states\n",
      "Within 30 percent: 41 states\n",
      "Over 25 percent: 13 states\n"
     ]
    }
   ],
   "source": [
    "# how close is our data to CDC?\n",
    "\n",
    "print('Within 1000: %d states' % sum(\n",
    "    full_merged_df.apply(lambda row: abs(row.test_count_diff) < 1000, axis=1)))\n",
    "print('Within 10 percent: %d states' % sum(\n",
    "    full_merged_df.apply(lambda row: abs(row.test_count_percent_diff) < 0.1, axis=1)))\n",
    "print('Within 20 percent: %d states' % sum(\n",
    "    full_merged_df.apply(lambda row: abs(row.test_count_percent_diff) < 0.2, axis=1)))\n",
    "print('Within 30 percent: %d states' % sum(\n",
    "    full_merged_df.apply(lambda row: abs(row.test_count_percent_diff) < 0.3, axis=1)))\n",
    "print('Over 25 percent: %d states' % sum(\n",
    "    full_merged_df.apply(lambda row: abs(row.test_count_percent_diff) > 0.25, axis=1)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# What are the most divergent states, by test counts?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>jurisdiction</th>\n",
       "      <th>ctp_total_tests_for_analysis</th>\n",
       "      <th>cdc_total_tests</th>\n",
       "      <th>test_count_diff</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>state</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>FL</th>\n",
       "      <td>Florida</td>\n",
       "      <td>691653.0</td>\n",
       "      <td>919109.0</td>\n",
       "      <td>227456.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CA</th>\n",
       "      <td>California</td>\n",
       "      <td>1104651.0</td>\n",
       "      <td>924696.0</td>\n",
       "      <td>-179955.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TX</th>\n",
       "      <td>Texas</td>\n",
       "      <td>623284.0</td>\n",
       "      <td>454133.0</td>\n",
       "      <td>-169151.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MA</th>\n",
       "      <td>Massachusetts</td>\n",
       "      <td>410032.0</td>\n",
       "      <td>574645.0</td>\n",
       "      <td>164613.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TN</th>\n",
       "      <td>Tennessee</td>\n",
       "      <td>302317.0</td>\n",
       "      <td>398173.0</td>\n",
       "      <td>95856.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IN</th>\n",
       "      <td>Indiana</td>\n",
       "      <td>160239.0</td>\n",
       "      <td>253619.0</td>\n",
       "      <td>93380.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AZ</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>134338.0</td>\n",
       "      <td>210388.0</td>\n",
       "      <td>76050.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC</th>\n",
       "      <td>North Carolina</td>\n",
       "      <td>219268.0</td>\n",
       "      <td>151449.0</td>\n",
       "      <td>-67819.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CO</th>\n",
       "      <td>Colorado</td>\n",
       "      <td>112505.0</td>\n",
       "      <td>173626.0</td>\n",
       "      <td>61121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MD</th>\n",
       "      <td>Maryland</td>\n",
       "      <td>178454.0</td>\n",
       "      <td>232086.0</td>\n",
       "      <td>53632.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         jurisdiction  ctp_total_tests_for_analysis  cdc_total_tests  \\\n",
       "state                                                                  \n",
       "FL            Florida                      691653.0         919109.0   \n",
       "CA         California                     1104651.0         924696.0   \n",
       "TX              Texas                      623284.0         454133.0   \n",
       "MA      Massachusetts                      410032.0         574645.0   \n",
       "TN          Tennessee                      302317.0         398173.0   \n",
       "IN            Indiana                      160239.0         253619.0   \n",
       "AZ            Arizona                      134338.0         210388.0   \n",
       "NC     North Carolina                      219268.0         151449.0   \n",
       "CO           Colorado                      112505.0         173626.0   \n",
       "MD           Maryland                      178454.0         232086.0   \n",
       "\n",
       "       test_count_diff  \n",
       "state                   \n",
       "FL            227456.0  \n",
       "CA           -179955.0  \n",
       "TX           -169151.0  \n",
       "MA            164613.0  \n",
       "TN             95856.0  \n",
       "IN             93380.0  \n",
       "AZ             76050.0  \n",
       "NC            -67819.0  \n",
       "CO             61121.0  \n",
       "MD             53632.0  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# top 10 most divergent states by test counts\n",
    "cols = ['jurisdiction', 'ctp_total_tests_for_analysis', 'cdc_total_tests', 'test_count_diff']\n",
    "full_merged_df.sort_values('abs_test_count_diff', ascending=False).head(10)[cols]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Same thing but by percentage difference?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>jurisdiction</th>\n",
       "      <th>ctp_total_tests_for_analysis</th>\n",
       "      <th>cdc_total_tests</th>\n",
       "      <th>test_count_percent_diff</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>state</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>IN</th>\n",
       "      <td>Indiana</td>\n",
       "      <td>160239.0</td>\n",
       "      <td>253619.0</td>\n",
       "      <td>0.582755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AZ</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>134338.0</td>\n",
       "      <td>210388.0</td>\n",
       "      <td>0.566109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CO</th>\n",
       "      <td>Colorado</td>\n",
       "      <td>112505.0</td>\n",
       "      <td>173626.0</td>\n",
       "      <td>0.543274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NH</th>\n",
       "      <td>New Hampshire</td>\n",
       "      <td>37739.0</td>\n",
       "      <td>19450.0</td>\n",
       "      <td>-0.484618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AK</th>\n",
       "      <td>Alaska</td>\n",
       "      <td>31762.0</td>\n",
       "      <td>46589.0</td>\n",
       "      <td>0.466816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MA</th>\n",
       "      <td>Massachusetts</td>\n",
       "      <td>410032.0</td>\n",
       "      <td>574645.0</td>\n",
       "      <td>0.401464</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FL</th>\n",
       "      <td>Florida</td>\n",
       "      <td>691653.0</td>\n",
       "      <td>919109.0</td>\n",
       "      <td>0.328859</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TN</th>\n",
       "      <td>Tennessee</td>\n",
       "      <td>302317.0</td>\n",
       "      <td>398173.0</td>\n",
       "      <td>0.317071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NC</th>\n",
       "      <td>North Carolina</td>\n",
       "      <td>219268.0</td>\n",
       "      <td>151449.0</td>\n",
       "      <td>-0.309297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MD</th>\n",
       "      <td>Maryland</td>\n",
       "      <td>178454.0</td>\n",
       "      <td>232086.0</td>\n",
       "      <td>0.300537</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         jurisdiction  ctp_total_tests_for_analysis  cdc_total_tests  \\\n",
       "state                                                                  \n",
       "IN            Indiana                      160239.0         253619.0   \n",
       "AZ            Arizona                      134338.0         210388.0   \n",
       "CO           Colorado                      112505.0         173626.0   \n",
       "NH      New Hampshire                       37739.0          19450.0   \n",
       "AK             Alaska                       31762.0          46589.0   \n",
       "MA      Massachusetts                      410032.0         574645.0   \n",
       "FL            Florida                      691653.0         919109.0   \n",
       "TN          Tennessee                      302317.0         398173.0   \n",
       "NC     North Carolina                      219268.0         151449.0   \n",
       "MD           Maryland                      178454.0         232086.0   \n",
       "\n",
       "       test_count_percent_diff  \n",
       "state                           \n",
       "IN                    0.582755  \n",
       "AZ                    0.566109  \n",
       "CO                    0.543274  \n",
       "NH                   -0.484618  \n",
       "AK                    0.466816  \n",
       "MA                    0.401464  \n",
       "FL                    0.328859  \n",
       "TN                    0.317071  \n",
       "NC                   -0.309297  \n",
       "MD                    0.300537  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Most divergent states by % difference in test counts\n",
    "\n",
    "cols = ['jurisdiction', 'ctp_total_tests_for_analysis', 'cdc_total_tests', 'test_count_percent_diff']\n",
    "full_merged_df.sort_values('abs_test_count_percent_diff', ascending=False).head(10)[cols]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Correlation between CTP and CDC data: tests, positives, deaths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot all 3 together\n",
    "\n",
    "fig, axes = plt.subplots(nrows=1, ncols=3, figsize=(12,4))\n",
    "fig.tight_layout(pad=3.0)\n",
    "\n",
    "def plot_gray_yx(ax):\n",
    "    lims = [\n",
    "        np.min([ax.get_xlim(), ax.get_ylim()]),  # min of both axes\n",
    "        np.max([ax.get_xlim(), ax.get_ylim()]),  # max of both axes\n",
    "    ]\n",
    "    # now plot both limits against eachother\n",
    "    ax.plot(lims, lims, 'k-', alpha=0.75, zorder=0, linestyle='--', color='gray')\n",
    "    \n",
    "    # remove gridlines!\n",
    "    ax.grid(False)\n",
    "\n",
    "# Case counts\n",
    "\n",
    "ax=axes[0]\n",
    "sns.scatterplot(x='ctp_total_cases', y='cdc_total_cases', data=full_merged_df, ax=ax, color='#585bc1')\n",
    "ax.set_ylabel('CDC', fontsize=14, rotation=0, labelpad=14)\n",
    "ax.set_xlabel('The COVID Tracking Project (CTP)', fontsize=14)\n",
    "ax.set_xticklabels('')\n",
    "ax.set_yticklabels('')\n",
    "ax.set_title('Cases', fontsize=18, pad=8)\n",
    "plot_gray_yx(ax)\n",
    "\n",
    "# Death counts\n",
    "\n",
    "ax = axes[1]\n",
    "sns.scatterplot(x='ctp_total_death', y='cdc_total_death', data=full_merged_df, ax=ax, color='#585bc1')\n",
    "# do a 2nd scatter plot on top, make the NY deaths dot in orange\n",
    "sns.scatterplot(x='ctp_total_death', y='cdc_total_death',\n",
    "                data=full_merged_df.loc[full_merged_df.index == 'NY'], ax=ax, color=\"#ffad4a\")\n",
    "ax.set_xticklabels('')\n",
    "ax.set_yticklabels('')\n",
    "ax.set_ylabel('CDC', fontsize=14, rotation=0, labelpad=14)\n",
    "ax.set_xlabel('CTP', fontsize=14)\n",
    "ax.set_title('Deaths', fontsize=18, pad=8)\n",
    "plot_gray_yx(ax)\n",
    "\n",
    "# Testing counts\n",
    "\n",
    "ax = axes[2]\n",
    "sns.scatterplot(x='ctp_total_tests_for_analysis', y='cdc_total_tests', data=full_merged_df, ax=ax,\n",
    "                color='#585bc1')\n",
    "ax.set_xticklabels('')\n",
    "ax.set_yticklabels('')\n",
    "ax.set_ylabel('CDC', fontsize=14, rotation=0, labelpad=14)\n",
    "ax.set_xlabel('CTP', fontsize=14)\n",
    "ax.set_title('Tests', fontsize=18, pad=8)\n",
    "plot_gray_yx(ax)\n",
    "\n",
    "plt.savefig(\"cases_deaths_testing_scatterplots.png\", dpi=200, bbox_inches = \"tight\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Make basic state-level summary tables, one per state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# make state-level tables\n",
    "for state in full_merged_df.index:\n",
    "    row = full_merged_df.loc[state]\n",
    "    positive_range = cdc.loc[cdc.state == state].iloc[0].cdc_percent_positive_range\n",
    "\n",
    "    data_types = [row.jurisdiction, 'Cases', 'Deaths', 'Tests']\n",
    "    cases_percent_diff_str = '%d%%' % round((row.cdc_total_cases-row.ctp_total_cases)/row.ctp_total_cases*100)\n",
    "    deaths_percent_diff_str = '%d%%' % round((row.cdc_total_death-row.ctp_total_death)/row.ctp_total_death*100)\n",
    "    tests_percent_diff_str = '%d%%' % round(row.test_count_percent_diff*100)\n",
    "    data = {\n",
    "        row.jurisdiction: ['State', 'CDC', '% Difference'],\n",
    "        'Cases': [int(row.ctp_total_cases), int(row.cdc_total_cases), cases_percent_diff_str\n",
    "                  ],\n",
    "        'Deaths': [int(row.ctp_total_death), int(row.cdc_total_death), deaths_percent_diff_str\n",
    "                  ],\n",
    "        'Tests': [int(row.ctp_total_tests_for_analysis), int(row.cdc_total_tests), tests_percent_diff_str],\n",
    "    }\n",
    "\n",
    "    pd.DataFrame(data).to_csv('states/%s.csv' % state, index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Specimens vs people counts ratios, for states that report both"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>state</th>\n",
       "      <th>total_tests_specimens</th>\n",
       "      <th>total_tests_people</th>\n",
       "      <th>total_tests_unclear</th>\n",
       "      <th>percent_more_specimens</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>20200514</td>\n",
       "      <td>FL</td>\n",
       "      <td>691653.0</td>\n",
       "      <td>609574.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>13.464977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>20200514</td>\n",
       "      <td>ME</td>\n",
       "      <td>33035.0</td>\n",
       "      <td>28357.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>16.496809</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>20200514</td>\n",
       "      <td>NV</td>\n",
       "      <td>82993.0</td>\n",
       "      <td>69484.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>19.441886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>20200514</td>\n",
       "      <td>VA</td>\n",
       "      <td>185551.0</td>\n",
       "      <td>165486.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12.124893</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        date state  total_tests_specimens  total_tests_people  \\\n",
       "9   20200514    FL               691653.0            609574.0   \n",
       "21  20200514    ME                33035.0             28357.0   \n",
       "33  20200514    NV                82993.0             69484.0   \n",
       "45  20200514    VA               185551.0            165486.0   \n",
       "\n",
       "    total_tests_unclear  percent_more_specimens  \n",
       "9                   NaN               13.464977  \n",
       "21                  NaN               16.496809  \n",
       "33                  NaN               19.441886  \n",
       "45                  NaN               12.124893  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# These states report both \n",
    "states = ['FL', 'ME', 'NV', 'VA']\n",
    "subset_df = ctp_514_test_overrides.loc[ctp_514_test_overrides.state.isin(states)].copy()\n",
    "subset_df['percent_more_specimens'] = (subset_df.total_tests_specimens/subset_df.total_tests_people - 1)*100\n",
    "subset_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>total_tests_specimens</th>\n",
       "      <th>total_tests_people</th>\n",
       "      <th>ratio</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>state</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>FL</th>\n",
       "      <td>691653.0</td>\n",
       "      <td>609574.0</td>\n",
       "      <td>1.134650</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ME</th>\n",
       "      <td>33035.0</td>\n",
       "      <td>28357.0</td>\n",
       "      <td>1.164968</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NV</th>\n",
       "      <td>82993.0</td>\n",
       "      <td>69484.0</td>\n",
       "      <td>1.194419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VA</th>\n",
       "      <td>185551.0</td>\n",
       "      <td>165486.0</td>\n",
       "      <td>1.121249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total</th>\n",
       "      <td>993232.0</td>\n",
       "      <td>872901.0</td>\n",
       "      <td>1.137852</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       total_tests_specimens  total_tests_people     ratio\n",
       "state                                                     \n",
       "FL                  691653.0            609574.0  1.134650\n",
       "ME                   33035.0             28357.0  1.164968\n",
       "NV                   82993.0             69484.0  1.194419\n",
       "VA                  185551.0            165486.0  1.121249\n",
       "Total               993232.0            872901.0  1.137852"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get some ratio counts\n",
    "specimens_people_df = ctp_514_test_overrides.loc[ctp_514_test_overrides.state.isin(states)].copy()[\n",
    "    ['state', 'total_tests_specimens', 'total_tests_people']\n",
    "]\n",
    "specimens_people_df['ratio'] = specimens_people_df.total_tests_specimens/specimens_people_df.total_tests_people\n",
    "specimens_people_df = specimens_people_df.set_index(['state'])\n",
    "specimens_people_df.loc['Total'] = [\n",
    "    sum(specimens_people_df.total_tests_specimens),\n",
    "    sum(specimens_people_df.total_tests_people),\n",
    "    np.mean(sum(specimens_people_df.total_tests_specimens)/sum(specimens_people_df.total_tests_people))\n",
    "]\n",
    "specimens_people_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.1506274946824986"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# alt ratio: average the individual ratios for each state\n",
    "alt_ratio = np.mean(specimens_people_df.ratio)\n",
    "alt_ratio"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plotting specimens vs people for the 4 states that report both"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(18,6))\n",
    "\n",
    "sns.barplot(x=specimens_people_df.index, y='total_tests_specimens', data=specimens_people_df, color=\"#ffad4a\")\n",
    "ax=sns.barplot(\n",
    "    x=specimens_people_df.index, y='total_tests_people', data=specimens_people_df,\n",
    "    color='#585bc1')\n",
    "\n",
    "sns.despine(ax=ax, left=True)\n",
    "labels = ['Florida', 'Maine', 'Nevada', 'Virginia', 'Four state total']\n",
    "ax.set_xticklabels(labels=labels, fontsize=14)\n",
    "ax.set_yticklabels(ax.get_yticks(), size=14)\n",
    "ax.set_xlabel('')\n",
    "ax.set_ylabel('')\n",
    "ax.yaxis.set_major_locator(ticker.MultipleLocator(250000))\n",
    "ax.yaxis.set_major_formatter(ticker.FuncFormatter(lambda x, p: format(int(x), ',')))\n",
    "plt.savefig(\"people_vs_specimens.png\", dpi=100, bbox_inches = \"tight\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}