sleep_plot.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import xml.etree.ElementTree as ET\n",
    "\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.dates as mdates\n",
    "\n",
    "plt.rc('savefig', dpi=300)\n",
    "\n",
    "# References\n",
    "# https://www.python-engineer.com/posts/apple-health-data-python/\n",
    "# https://www.relevantmisc.com/r/python/2020/05/26/visualizing-baby-sleep/\n",
    "# https://stackoverflow.com/questions/72989080/how-to-make-plots-in-python-like-the-sleep-schedule-plot-in-ios-health-app"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Path to the XML file exported form Apple Health app\n",
    "XML_DATA = \"../apple_data/apple_health_export/export.xml\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Parse XML file exported from Apple Health app\n",
    "tree = ET.parse(XML_DATA)\n",
    "root = tree.getroot()\n",
    "\n",
    "# Store \"Record\" type data into Pandas.DataFrame\n",
    "records = [i.attrib for i in root.iter(\"Record\")]\n",
    "records_df = pd.DataFrame(records)\n",
    "\n",
    "# Convert datetime format\n",
    "date_col = ['creationDate', 'startDate', 'endDate']\n",
    "records_df[date_col] = records_df[date_col].apply(pd.to_datetime) \n",
    "\n",
    "# Select Sleep Analysis records\n",
    "sleeps_df = records_df.query(\"type == 'HKCategoryTypeIdentifierSleepAnalysis'\")\n",
    "\n",
    "# Cut overnight records into two separate records\n",
    "no_cross = sleeps_df[sleeps_df[\"startDate\"].dt.day == sleeps_df[\"endDate\"].dt.day]\n",
    "cross = sleeps_df[sleeps_df[\"startDate\"].dt.day != sleeps_df[\"endDate\"].dt.day]\n",
    "c1 = cross.copy()\n",
    "c2 = cross.copy()\n",
    "c1[\"endDate\"] = c1[\"startDate\"].apply(lambda x: x.replace(hour=23, minute=59, second=59))\n",
    "c2[\"startDate\"] = c2[\"endDate\"].apply(lambda x: x.replace(hour=0, minute=0, second=0))\n",
    "sleeps_splitted_df = pd.concat([no_cross, c1, c2]).sort_values(\"startDate\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Select plotting starting date\n",
    "# x = sleeps_splitted_df[sleeps_splitted_df[\"startDate\"] > pd.Timestamp(2022, 1, 1).tz_localize(tz=\"US/Central\") ]\n",
    "# x = sleeps_splitted_df[sleeps_splitted_df[\"startDate\"] > pd.Timestamp(2022, 6, 1).tz_localize(tz=\"US/Central\") ]\n",
    "x = sleeps_splitted_df[sleeps_splitted_df[\"startDate\"] > pd.Timestamp(2022, 4, 1).tz_localize(tz=\"US/Central\") ]\n",
    "\n",
    "# Plotting\n",
    "# fig = plt.figure(figsize=(20, 6))  # better for a year\n",
    "fig = plt.figure(figsize=(10, 6))  # better for a few months \n",
    "\n",
    "ax = fig.add_subplot()\n",
    "\n",
    "width = 0.4\n",
    "color_dict = {\"HKCategoryValueSleepAnalysisAsleep\": 'darkturquoise',\n",
    "              \"HKCategoryValueSleepAnalysisInBed\": 'turquoise',\n",
    "              \"HKCategoryValueSleepAnalysisAwake\": \"red\"}\n",
    "alpha_dict = {\"HKCategoryValueSleepAnalysisAsleep\": 1,\n",
    "              \"HKCategoryValueSleepAnalysisInBed\": 0.5,\n",
    "              \"HKCategoryValueSleepAnalysisAwake\": 0.2}\n",
    "\n",
    "for row in x.itertuples(index=False):\n",
    "    # Skip \"Awake\" for plotting\n",
    "    if row.value == \"HKCategoryValueSleepAnalysisAwake\":\n",
    "        continue\n",
    "    \n",
    "    # Timestemp at the begining of the startDate's day\n",
    "    row_date = row.startDate.replace(hour=0, minute=0, second=0)\n",
    "\n",
    "    # Plot rectangle patches ((x0, y0), width, hight)\n",
    "    rect = matplotlib.patches.Rectangle((mdates.date2num(row.startDate.date()), mdates.date2num(row.startDate-row_date)/1000),\n",
    "                                        width,\n",
    "                                        mdates.date2num(row.endDate-row.startDate)/1000,\n",
    "                                        color=color_dict[row.value],\n",
    "                                        alpha=alpha_dict[row.value])\n",
    "    ax.add_patch(rect)\n",
    "\n",
    "# Generate legend\n",
    "legend_patches = [\n",
    "    matplotlib.patches.Patch(color=color_dict[\"HKCategoryValueSleepAnalysisAsleep\"],\n",
    "                             alpha=alpha_dict[\"HKCategoryValueSleepAnalysisAsleep\"],\n",
    "                             label=\"Asleep\"),\n",
    "    matplotlib.patches.Patch(color=color_dict[\"HKCategoryValueSleepAnalysisInBed\"],\n",
    "                             alpha=alpha_dict[\"HKCategoryValueSleepAnalysisInBed\"],\n",
    "                             label=\"In Bed\")\n",
    "]\n",
    "plt.legend(loc=2, handles=legend_patches)\n",
    "\n",
    "# Plot range for current data range\n",
    "plt.xlim(mdates.date2num(x[\"startDate\"].min())-1, mdates.date2num(x[\"endDate\"].max())+1)\n",
    "plt.ylim(1,0)\n",
    "\n",
    "# Set Ticks\n",
    "# plt.gca().yaxis.set_major_formatter(mdates.DateFormatter('%H'))\n",
    "plt.gca().yaxis.set_major_formatter(mdates.DateFormatter('%I %p'))\n",
    "plt.gca().yaxis.set_major_locator(mdates.HourLocator())\n",
    "plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%m/%d/%Y'))\n",
    "plt.gca().xaxis.set_major_locator(mdates.DayLocator(interval=7))\n",
    "plt.gcf().autofmt_xdate()\n",
    "\n",
    "plt.tight_layout()\n",
    "# plt.savefig(\"sleep_plot.pdf\")\n",
    "plt.savefig(\"sleep_plot@2x.png\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.13 ('base')",
   "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": "3.9.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "0f4d23bbd019b7c7a09081ea7f9f7b35ba7f7fe21809859712127420437062c8"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}