{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%reload_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from datetime import datetime, timedelta"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Bedtimes Near Midnight\n",
    "\n",
    "Sleep times near midnight span the day boundary. Late bedtimes like 23:30 and early ones like 00:30 are actually close together — but since the 24-hour clock wraps at 0:00, those times get split across the boundary unless we treat time as circular."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "url = \"https://raw.githubusercontent.com/timpyrkov/circleclust/refs/heads/master/tests/sleep.csv\"\n",
    "\n",
    "# Read sleep-wake data csv from url (sleep.jpg from github repository)\n",
    "df = pd.read_csv(url, parse_dates=[\"sleep_start_datetime\", \"sleep_end_datetime\"])\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Show Sleep-Wake actogram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a raw  array of activity for 2025 (365 days x 24 hours x 60 minutes)\n",
    "sleep_array = np.ones(365 * 24 * 60, dtype=int)  # Initialize with ones (awake)\n",
    "\n",
    "# Fill in sleep intervals as zeros\n",
    "year_start = datetime(2025, 1, 1)\n",
    "year_end = datetime(2025, 12, 31, 23, 59)\n",
    "for _, row in df.iterrows():\n",
    "    # Get start-end of the sleep interval\n",
    "    start = row['sleep_start_datetime']\n",
    "    end = row['sleep_end_datetime']\n",
    "    # Calculate start-end indices for the sleep interval\n",
    "    start_idx = int((start - year_start).total_seconds() // 60)\n",
    "    end_idx = int((end - year_start).total_seconds() // 60)\n",
    "    # Mark sleep interval as zeros\n",
    "    sleep_array[start_idx:end_idx] = 0  \n",
    "# Reshape to daily samples (365 days x 1440 minutes)\n",
    "daily_sleep = sleep_array.reshape(365, 1440)\n",
    "\n",
    "# Plot as a heatmap\n",
    "plt.figure(figsize=(12, 6))\n",
    "plt.title('Sleep-Wake Actogram for 2025', fontsize=16)\n",
    "\n",
    "plt.imshow(daily_sleep.T, aspect='auto', cmap='cividis', origin='lower', alpha=0.8)\n",
    "plt.colorbar(label='State (0=Sleep, 1=Awake)')\n",
    "\n",
    "# Set xticks\n",
    "month_starts = [datetime(2025, m, 1) for m in range(1, 13)] + [datetime(2026, 1, 1)]\n",
    "xtick_days = [(m - datetime(2025, 1, 1)).days for m in month_starts]\n",
    "xtick_labels = [m.strftime('%b %d') for m in month_starts]\n",
    "plt.xticks(xtick_days, xtick_labels, rotation=45)\n",
    "\n",
    "# Set yticks\n",
    "ytick_minutes = [0, 360, 720, 1080, 1440]  # 12 AM, 6 AM, 12 PM, 6 PM, 12 AM\n",
    "ytick_labels = ['12 AM', '6 AM', '12 PM', '6 PM', '12 AM']\n",
    "plt.yticks(ytick_minutes, ytick_labels)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show histogram of go-to-sleep and wake-up time [minutes past midnight]\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.title('Histogram of go-to-sleep/wake-up times', fontsize=14)\n",
    "\n",
    "t = pd.concat([df[\"sleep_start_datetime\"], df[\"sleep_end_datetime\"]]).dt\n",
    "t = (t.hour * 60 + t.minute + t.second / 60.0).values\n",
    "plt.hist(t, bins=np.linspace(0, 1440, 50), width=25, alpha=.3)\n",
    "\n",
    "plt.xticks([0, 360, 720, 1080, 1440], ['12 AM', '6 AM', '12 PM', '6 PM', '12 AM'])\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Find Sleep-Wake Times\n",
    " \n",
    " Detect go-to-sleep and wake-up time from distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from circleclust import CircleClust\n",
    "\n",
    "# Initialize and run CircleClust.fit() to find pixel group centroids\n",
    "cc = CircleClust(verbose=False)\n",
    "cc.fit(t, period=1440) # Important: provide correct range of data values period!\n",
    "\n",
    "# Print detected centroids\n",
    "cc.get_centroids()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show detected centroids\n",
    "cc.show_centroids()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sleep-Wake Stats\n",
    "\n",
    "Note that the \"Go-to-sleep\" times cluster near the area of wrap at 12 AM (midnight). Both times clsoe to 1:00 AM and 11:00 PM are attributed to the same single \"Go-to-sleep\" cluster."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert minutes to HH:MM format\n",
    "def format_minutes_to_hhmm(minutes):\n",
    "    hours = int(minutes // 60) % 24\n",
    "    mins = int(minutes % 60)\n",
    "    return f\"{hours:02d}:{mins:02d}\"\n",
    "\n",
    "# Get detected centroids\n",
    "centroids = cc.get_centroids()\n",
    "labels = [\"Wake-up\", \"Go-to-sleep\"]\n",
    "\n",
    "# Put detected time ranges in a DataFrame\n",
    "data = []\n",
    "for i, label in enumerate(labels):\n",
    "    center_minutes = centroids['centroid'][i]\n",
    "    range_minutes = centroids['std'][i]\n",
    "    start_minutes = center_minutes - range_minutes\n",
    "    end_minutes = center_minutes + range_minutes\n",
    "\n",
    "    center_time = format_minutes_to_hhmm(center_minutes)\n",
    "    start_time = format_minutes_to_hhmm(start_minutes)\n",
    "    end_time = format_minutes_to_hhmm(end_minutes)\n",
    "\n",
    "    data.append({\n",
    "        \"Event\": label,\n",
    "        \"Time\": center_time,\n",
    "        \"± Std Dev (min)\": int(range_minutes),\n",
    "        \"Range\": f\"{start_time} - {end_time}\"\n",
    "    })\n",
    "\n",
    "# Create and display the DataFrame\n",
    "df = pd.DataFrame(data)\n",
    "df.set_index(\"Event\", inplace=True)\n",
    "df"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.12.11"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}