heuristic.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: kaleido in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (0.2.1)\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "kaleido                   0.2.1\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "Requirement already satisfied: pydot in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (1.4.2)\n",
      "Requirement already satisfied: pyparsing>=2.1.4 in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (from pydot) (3.0.9)\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "Requirement already satisfied: graphviz in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (0.20.1)\n",
      "Note: you may need to restart the kernel to use updated packages.\n",
      "Requirement already satisfied: keras_tuner in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (1.3.5)\n",
      "Requirement already satisfied: packaging in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (from keras_tuner) (23.1)\n",
      "Requirement already satisfied: requests in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (from keras_tuner) (2.31.0)\n",
      "Requirement already satisfied: kt-legacy in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (from keras_tuner) (1.0.5)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (from requests->keras_tuner) (3.1.0)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (from requests->keras_tuner) (3.4)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (from requests->keras_tuner) (1.26.16)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /Library/Frameworks/Python.framework/Versions/3.11/lib/python3.11/site-packages (from requests->keras_tuner) (2023.5.7)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "%pip install -U kaleido\n",
    "%pip list | grep kaleido\n",
    "%pip install pydot\n",
    "%pip install graphviz\n",
    "%pip install keras_tuner\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "train_dir = \"/Users/lina/Desktop/ba-implementation/train\"\n",
    "filtered_dir = \"/Users/lina/Desktop/ba-implementation/train/5d27075f03f801723c2e360f/F1_filtered/\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Heuristics\n",
    "1. For each current AP, choose one random out of the five nearest APs and check if the next and the random one are identical\n",
    "2. same for next 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def aggregated_heuristic(size):\n",
    "    total_correct_guesses = 0\n",
    "    total_rows = 0 \n",
    "\n",
    "    for i in range(147):\n",
    "        file_name = filtered_dir + f\"floor_metric_interpolated_waypoints_acce_{i}_merged_data_acce.csv\"\n",
    "        df = pd.read_csv(file_name)\n",
    "        df = df.iloc[:, [0] + list(range(7, df.shape[1]))]\n",
    "        \n",
    "        for j in range(1, len(df)-1):\n",
    "            top_size_values = df.iloc[j, 1:].nlargest(size)\n",
    "            top_size_cols = set(top_size_values.index)\n",
    "            \n",
    "            # Check if the index of the max value of the next row is in the top x columns of the current row\n",
    "            if df.iloc[j+1, 1:].idxmax() in top_size_cols:\n",
    "                total_correct_guesses += 1\n",
    "            \n",
    "            total_rows += 1\n",
    "\n",
    "    aggregated_heuristic_value = total_correct_guesses / total_rows\n",
    "    return aggregated_heuristic_value\n",
    "\n",
    "sizes = list(range(1, 6))\n",
    "heuristic_values = [aggregated_heuristic(size) * 100 for size in sizes]\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "bars = plt.bar(sizes, heuristic_values, width=0.6)\n",
    "\n",
    "for bar in bars:\n",
    "    yval = bar.get_height()\n",
    "    plt.text(bar.get_x() + bar.get_width()/2, yval + 0.5, round(yval, 2), ha='center', va='bottom')\n",
    "\n",
    "plt.title(\"Probability of Correctly Predicting the Strongest Signal in Subsequent Timestamps\")\n",
    "plt.xlabel(\"Top X Access Points Considered\")\n",
    "plt.ylabel(\"Probability (%)\")\n",
    "plt.savefig(\"heuristic_plot.pdf\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot complete comparison with ml_values and heuristic_values\n",
    "\n",
    "ml_values = [51.56, 68.34, 76.09, 80.30, 82.72]\n",
    "\n",
    "bar_width = 0.35\n",
    "index = range(len(ml_values))\n",
    "plt.figure(figsize=(10, 6))\n",
    "bar1 = plt.bar(index, ml_values, bar_width, label='ML Values', color='C1')\n",
    "bar2 = plt.bar([i + bar_width for i in index], heuristic_values, bar_width, label='Heuristic Values', color='C0')\n",
    "plt.xlabel('Set size')\n",
    "plt.ylabel('Probability (%)')\n",
    "plt.xticks([i + bar_width/2 for i in index], ['1', '2', '3', '4', '5'])\n",
    "plt.legend()\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"comparison_ml_heuristic_1_to_5.pdf\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot top 3\n",
    "\n",
    "labels = ['ML Value', 'Heuristic Value']\n",
    "values = [ml_value_3, heuristic_value_3]\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.bar(labels, values, color=['C1', 'C0'])\n",
    "plt.ylabel('Probability (%)')\n",
    "plt.title('Comparison of Probability between ML Model and Heuristic')\n",
    "plt.ylim(70, 80) \n",
    "plt.savefig(\"comparison_ml_heuristic_3.pdf\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get the time for selecting the three maximums in recent APs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Duration: 0.07589983940124512 seconds\n",
      "Index(['27a2117236a35ee92c6eb925f0195b59f2fcc034',\n",
      "       'c89db743bbec414085f3db50bbf2ce70f0a13846',\n",
      "       'd091c86c2295f29e0c1ff1749d43d3b05340958e'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "\n",
    "file = filtered_dir + \"floor_metric_interpolated_waypoints_acce_1_merged_data_acce.csv\"\n",
    "\n",
    "start_time = time.time()\n",
    "\n",
    "df = pd.read_csv(file)\n",
    "df = df.iloc[:, list(range(7, df.shape[1]))]\n",
    "df = df.head(1)\n",
    "values = df.T[0].nlargest(3)\n",
    "bssids = values.index\n",
    "\n",
    "end_time = time.time()\n",
    "\n",
    "print(\"Duration:\", end_time - start_time, \"seconds\")\n",
    "print(bssids)"
   ]
  }
 ],
 "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": "3.11.3"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}