COVID19-dashboard.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests\n",
    "from io import StringIO\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy.stats as stats\n",
    "\n",
    "from scipy.optimize import curve_fit\n",
    "\n",
    "import re\n",
    "\n",
    "from ipywidgets import interact, interactive, fixed, interact_manual, interactive_output\n",
    "import ipywidgets as widgets\n",
    "import bqplot as bq\n",
    "\n",
    "from datetime import datetime\n",
    "import warnings\n",
    "\n",
    "warnings.filterwarnings('ignore')\n",
    "matplotlib.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_data():\n",
    "    confirmed = requests.get('https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_confirmed_global.csv').text\n",
    "    deaths = requests.get('https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_deaths_global.csv').text\n",
    "    recovered = requests.get('https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_recovered_global.csv').text\n",
    "\n",
    "    confirmed_df = pd.read_csv(StringIO(confirmed))\n",
    "    deaths_df = pd.read_csv(StringIO(deaths))\n",
    "    recovered_df = pd.read_csv(StringIO(recovered))\n",
    "    \n",
    "    c_df = filter_active_countries(confirmed_df)\n",
    "    d_df = filter_active_countries(deaths_df)\n",
    "    r_df = filter_active_countries(recovered_df)\n",
    "    \n",
    "    return c_df,d_df,r_df\n",
    "\n",
    "def filter_active_countries(df):\n",
    "    df1 = df.groupby(by='Country/Region').sum()\n",
    "    return df1.copy()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "confirmed_df,deaths_df,recovered_df = get_data()\n",
    "\n",
    "df = filter_active_countries(confirmed_df)\n",
    "dfs = [filter_active_countries(confirmed_df),\\\n",
    "       filter_active_countries(recovered_df),\\\n",
    "       filter_active_countries(deaths_df)]\n",
    "\n",
    "all_countries = df.index.tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# COVID19 Dashboard\n",
    "\n",
    "<!-- <div class=\"alert alert-danger\"> -->\n",
    "\n",
    "    \n",
    "This dashboard estimates growth rates of the number of infected people by the current COVID19 pandemic. It allows to compare these estimates between countries.\n",
    "\n",
    "* you can select several countries by pressing ctrl\n",
    "* you can change the number of days in which the estimate is based  using the top left dropdown.\n",
    "    \n",
    "    \n",
    "**Important**: \n",
    "* Testing coverage is uneven between countries.\n",
    "* The data is only updated once a day and is based on data from WHO. Individual countries do not report the data everyday, so there may be some fluctuations\n",
    "* Estimating exponential growth rates from just a few datapoints can lead to large uncertainties\n",
    "    \n",
    " \n",
    "<!-- </div> -->"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_timeline(countries,days=7,df_idx=0,range_day=0):\n",
    "    \n",
    "#     countries = country_select.value\n",
    "    fig,ax = plt.subplots(figsize=(12,4))\n",
    "\n",
    "    color_scheme = plt.get_cmap('tab20')\n",
    "\n",
    "#     df = confirmed_df.groupby(by='Country/Region').sum()\n",
    "\n",
    "    for idx,country in enumerate(countries):\n",
    "        timepoints = pd.to_datetime(df.columns[2+range_day:])\n",
    "\n",
    "        tp0 = timepoints[0]\n",
    "        tps = np.array([ (tp-tp0).days for tp in timepoints ])\n",
    "        \n",
    "        values = dfs[df_idx].loc[country].iloc[2+range_day:].to_numpy()\n",
    "        plt.plot( tps, np.log10(np.array(values.tolist())),label=country,marker='.',c=color_scheme(idx/len(countries)) )\n",
    "        \n",
    "        ## linear regresions\n",
    "\n",
    "\n",
    "        lr = stats.linregress(tps[-days:],np.log10(values[-days:].tolist()))\n",
    "        m = lr.slope\n",
    "        a = lr.intercept\n",
    "#         print(m)\n",
    "        if not np.isnan(m):\n",
    "            xx = np.array([tps[-days],tps[-1]+4])\n",
    "            xx2 = np.array([timepoints[-days],timepoints[-1]+pd.to_timedelta(5,unit='d')])\n",
    "\n",
    "            yy = np.array([ m*x+a for x in xx ])\n",
    "#             print(xx,yy,m)\n",
    "            plt.plot(xx,yy,ls='--',c=color_scheme(idx/len(countries)))\n",
    "    \n",
    "    if df_idx==0:\n",
    "        lab=\"infected\"\n",
    "    elif df_idx==1:\n",
    "        lab=\"deaths\"\n",
    "    else:\n",
    "        lab=\"recovered\"\n",
    "        \n",
    "    plt.title('Number of %s\\n Dashed line is prediction based on growth rate estimate from the last %d days'%(lab,days))\n",
    "    plt.legend(loc='upper left')\n",
    "#     plt.xlabel('time (days)')\n",
    "#     plt.xlim(tps[range_day],tps[-1]+5)\n",
    "\n",
    "    plt.ylabel('Number of confirmed cases (log)')\n",
    "    plt.yticks(range(6),[ \"$10^%d$\"%i for i in range(6)] );\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_change_R(countries,days,df_idx=0,range_day=0):\n",
    "    plt.subplots(figsize=(12,4))\n",
    "\n",
    "    color_scheme = plt.get_cmap('tab20')\n",
    "\n",
    "    rolling_slopes = []\n",
    "\n",
    "    for idx,country in enumerate(countries):\n",
    "        timepoints = pd.to_datetime(df.columns[2+range_day:])\n",
    "\n",
    "        values = dfs[df_idx].loc[country].iloc[2+range_day:].to_numpy()\n",
    "\n",
    "        ## linear regresions\n",
    "        tp0 = timepoints[0]\n",
    "        tps = np.array([ (tp-tp0).days for tp in timepoints ])\n",
    "\n",
    "        rolling_slopes = []\n",
    "        stderrs = []\n",
    "        for first_index in range(days,values.shape[0]):\n",
    "            lr = stats.linregress(tps[first_index-days:first_index],\\\n",
    "                                  np.log10(values[first_index-days:first_index]))\n",
    "            rolling_slopes.append(lr.slope)\n",
    "            stderrs.append(lr.stderr)\n",
    "        \n",
    "        rolling_slopes = np.array(rolling_slopes)\n",
    "        stderrs = np.array(stderrs)\n",
    "        xx = np.arange(days,days+len(rolling_slopes))\n",
    "        \n",
    "        plt.plot(xx,rolling_slopes,label=country,marker='.',c=color_scheme(idx/len(countries)))\n",
    "        plt.fill_between(xx,rolling_slopes+1.96*stderrs,rolling_slopes-1.96*stderrs,\\\n",
    "                         color=color_scheme(idx/len(countries)),alpha=.2)\n",
    "        \n",
    "        \n",
    "    plt.title('Change in growth rate estimate\\nbased on the previous %d days'%days)\n",
    "    plt.legend(loc = 'upper left')\n",
    "    plt.xlabel('Time (days)')\n",
    "#     plt.ylabel('Apparent $R$\\n%d day average'%days)\n",
    "    plt.ylabel('Apparent $R$ (%d day estimate)'%days)\n",
    "#     plt.xlim(tps[range_day],tps[-1]+5)\n",
    "    tics = plt.yticks()\n",
    "    plt.yticks(tics[0], [ \"%.2f\"%10**tic for tic in tics[0]] );"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "667d45d6d49749d285afc5582d797b84",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(VBox(children=(HBox(children=(Dropdown(description='days to estimate R', index=3, layout=Layout…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "layout_sidebar = widgets.Layout(width=\"25%\")\n",
    "\n",
    "country_select = widgets.SelectMultiple(\n",
    "    options=all_countries,\n",
    "    value=['Portugal'],\n",
    "    rows=25,\n",
    "#     description='Countries\\n(select several with Ctrl)',\n",
    "    disabled=False,\n",
    "    layout=widgets.Layout(width='auto')\n",
    ")\n",
    "\n",
    "df_select = widgets.Dropdown(value=0,options=[('Confirmed',0),(\"Deaths\",1),(\"Recovered\",2)],layout=widgets.Layout(width='auto'))\n",
    "\n",
    "# day_select = widgets.IntSlider(value=7,min=4,max=15,continuous_update=False,description='',layout=widgets.Layout(width='auto'))\n",
    "day_select = widgets.Dropdown(value=7,options=[i for i in range(4,16)] ,description='days to estimate R',layout=widgets.Layout(width='auto'))\n",
    "max_day = df.columns.shape[0]-2\n",
    "# range_day = widgets.Dropdown(value=0,options=[i for i in range(0,max_day)] ,description='',layout=widgets.Layout(width='auto'))\n",
    "range_day = widgets.IntSlider(value=0,min=0,max=max_day ,continuous_update=False,description='',layout=widgets.Layout(width='auto'))\n",
    "\n",
    "\n",
    "out_timeline = interactive_output(plot_timeline,{'countries':country_select,'days':day_select,'range_day':range_day,'df_idx':df_select})\n",
    "out_r = interactive_output(plot_change_R,{'countries':country_select,'days':day_select,'range_day':range_day,'df_idx':df_select})\n",
    "\n",
    "\n",
    "display(widgets.HBox([widgets.VBox([widgets.HBox([day_select,df_select]),country_select],layout=layout_sidebar),widgets.VBox([range_day,out_timeline,out_r])]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_timeline(all_countries[:4],)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}