Geolife/attack_geolife.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "# sys.stdout = open('logfile_geolife', 'w')\n",
    "sys.path.append('..')\n",
    "\n",
    "import attack\n",
    "import data_loader as dl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import geopandas as gp\n",
    "import pandas as pd\n",
    "from tqdm import tqdm\n",
    "import numpy as np\n",
    "import libpysal\n",
    "import itertools\n",
    "from joblib import Parallel, delayed\n",
    "import random\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Raw Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading raw geolife geojson file...\n",
      "Done.\n",
      "Number of trajectories in loaded dataset: 5400\n",
      "Number of users in loaded dataset: 73\n"
     ]
    }
   ],
   "source": [
    "# Load preprocessed data \n",
    "raw_full_trip_gdf, raw_trip_sp_gdf, raw_trip_ep_gdf, tesselation_gdf = dl.load_geolife(data_type='raw', \n",
    "                                                                                       min_n_trips_per_user=1,\n",
    "                                                                                       rand_n_week_period=None,\n",
    "                                                                                       tessellation_diameter=200,\n",
    "                                                                                       min_trip_length=200,\n",
    "                                                                                       min_nr_points=50,\n",
    "                                                                                       upper_quantile_length=0.95,\n",
    "                                                                                       only_2008=True)\n",
    "assert len(raw_full_trip_gdf) == len(raw_trip_sp_gdf) == len(raw_trip_ep_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Parameters\n",
    "LCSS_EPS = 200\n",
    "LCSS_FLIP = True\n",
    "\n",
    "HL_SP_START_TIME = '6:00'\n",
    "HL_SP_END_TIME = '10:00'\n",
    "HL_EP_START_TIME = '18:00'\n",
    "HL_EP_END_TIME = '0:00'\n",
    "\n",
    "CHAINING_INFLOW_HR_DIFF_THRESHOLD = 4\n",
    "CHAINING_HR_DIFF_THRESHOLD = 8\n",
    "\n",
    "HL_SP_OUTFLOW_THRESHOLD = 2\n",
    "HL_EP_OUTFLOW_THRESHOLD = 4\n",
    "\n",
    "RANDOMIZED_SIMULTANEOUS_SEARCH_ITERATIONS = 1000\n",
    "\n",
    "SIM_THRESH_FOR_NO_MATCH_TRIPS = 0.5\n",
    "\n",
    "# JSD Parameters\n",
    "GRID_RESOLUTION_JSD = 1000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1cAAAHWCAYAAACbsXOkAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABjl0lEQVR4nO3dd3hUZf7+8Xsy6aRAIIXQewkhFBUjrKB0FcSGorsUEV2FlWYBC01XEKWoRHFdFVfXVeQrWCmRKgoogUgRECKdEHoKgZSZ8/uDZH6OSSATJpmZ5P26rrnktOfc5zwzMZ+cc54xGYZhCAAAAABwVbxcHQAAAAAAKgOKKwAAAABwAoorAAAAAHACiisAAAAAcAKKKwAAAABwAoorAAAAAHACiisAAAAAcAKKKwAAAABwAoorAAAAAHACiisAlcaUKVNkMpkqZF/dunVTt27dbNNr1qyRyWTSokWLKmT/Q4cOVcOGDStkX2WVlZWlhx56SFFRUTKZTBozZoyrIzmkW7duatOmjatjlNqHH36oli1bysfHR9WrV3dq2wcOHJDJZNKCBQuc2i4AVDYUVwDc0oIFC2QymWwvf39/RUdHq3fv3nr99deVmZnplP0cO3ZMU6ZMUXJyslPacyZ3zlYaL730khYsWKBHH31UH374of72t7+5OlKltXv3bg0dOlRNmjTRO++8o3/9618lrvvtt99qypQpFRcOAKoQb1cHAIDLmTZtmho1aqS8vDwdP35ca9as0ZgxYzR79mx9+eWXatu2rW3d5557ThMmTHCo/WPHjmnq1Klq2LCh2rVrV+rtVqxY4dB+yuJy2d555x1ZrdZyz3A1Vq1apeuvv16TJ092dZRKb82aNbJarXrttdfUtGnTy6777bffKiEhwaECq0GDBrpw4YJ8fHyuMikAVG4UVwDcWt++fXXNNdfYpidOnKhVq1bptttuU//+/bVr1y4FBARIkry9veXtXb4/1rKzsxUYGChfX99y3c+VeMIvuSdOnFDr1q1dHcOtWa1W5ebmyt/f/6raOXHihCQ5/XbA/Px8Wa1W+fr6XnVGd3X+/HlVq1bN1TGuqPBnDwD3xm2BADzOzTffrOeff14HDx7URx99ZJtf3DNXiYmJ6tKli6pXr66goCC1aNFCzzzzjKRLf+2/9tprJUnDhg2z3YJY+FxJ4TM3SUlJuvHGGxUYGGjb9s/PXBWyWCx65plnFBUVpWrVqql///46fPiw3ToNGzbU0KFDi2z7xzavlK24Z67Onz+v8ePHq169evLz81OLFi306quvyjAMu/VMJpNGjRqlJUuWqE2bNvLz81NMTIyWLVtW/An/kxMnTmj48OGKjIyUv7+/4uLi9MEHH9iWFz5/tn//fn3zzTe27AcOHCixzdJmKulZs+L6vrDNzz77TK1bt1ZAQIDi4+O1fft2SdLbb7+tpk2byt/fX926dSsxX1JSkm644QYFBASoUaNGmj9/fpF1cnJyNHnyZDVt2lR+fn6qV6+ennrqKeXk5BSb6b///a9iYmLk5+d3xfP+5ptv2taNjo7WyJEjde7cOdvyhg0b2q4OhoeHy2QylXhVaujQoUpISLBlKXxJ//+5qldffVVz585VkyZN5Ofnp19//bXYZ66GDh2qoKAg/f777+rdu7eqVaum6OhoTZs2rch77pNPPlHHjh0VHByskJAQxcbG6rXXXrvscf8xz5w5c9SgQQMFBASoa9eu2rFjR5H1d+/erbvvvlthYWHy9/fXNddcoy+//NJuncLbjdeuXavHHntMERERqlu3bokZCtf/83uj8D2+Zs0a27y9e/fqrrvuUlRUlPz9/VW3bl3dd999Sk9Pt9v2o48+UseOHRUQEKCwsDDdd999RX5GXO5nDwD3xpUrAB7pb3/7m5555hmtWLFCI0aMKHadnTt36rbbblPbtm01bdo0+fn5ad++ffrhhx8kSa1atdK0adM0adIkPfzww/rLX/4iSbrhhhtsbZw+fVp9+/bVfffdp7/+9a+KjIy8bK5//vOfMplMevrpp3XixAnNnTtXPXr0UHJysu0KW2mUJtsfGYah/v37a/Xq1Ro+fLjatWun5cuX68knn9TRo0c1Z84cu/XXr1+vzz//XI899piCg4P1+uuv66677tKhQ4dUs2bNEnNduHBB3bp10759+zRq1Cg1atRIn332mYYOHapz585p9OjRatWqlT788EONHTtWdevW1fjx4yVd+sX/csqa6XK+//57ffnllxo5cqQkafr06brtttv01FNP6c0339Rjjz2ms2fPaubMmXrwwQe1atUqu+3Pnj2rW265RQMHDtSgQYO0cOFCPfroo/L19dWDDz4o6dLVp/79+2v9+vV6+OGH1apVK23fvl1z5szRb7/9piVLlti1uWrVKi1cuFCjRo1SrVq1LjswyZQpUzR16lT16NFDjz76qPbs2aO33npLP//8s3744Qf5+Pho7ty5+s9//qPFixfrrbfeUlBQkN3tsn/0yCOP6NixY0pMTNSHH35Y7Drvv/++Ll68qIcfflh+fn4KCwsr8RZUi8WiPn366Prrr9fMmTO1bNkyTZ48Wfn5+Zo2bZqkS3/gGDRokLp3766XX35ZkrRr1y798MMPGj16dInHXug///mPMjMzNXLkSF28eFGvvfaabr75Zm3fvt32edy5c6c6d+6sOnXqaMKECapWrZoWLlyoAQMG6P/+7/90xx132LX52GOPKTw8XJMmTdL58+evmOFKcnNz1bt3b+Xk5Ogf//iHoqKidPToUX399dc6d+6cQkNDJV36+fD8889r4MCBeuihh3Ty5Em98cYbuvHGG7V161a7K4+O/uwB4CYMAHBD77//viHJ+Pnnn0tcJzQ01Gjfvr1tevLkycYff6zNmTPHkGScPHmyxDZ+/vlnQ5Lx/vvvF1nWtWtXQ5Ixf/78Ypd17drVNr169WpDklGnTh0jIyPDNn/hwoWGJOO1116zzWvQoIExZMiQK7Z5uWxDhgwxGjRoYJtesmSJIcl48cUX7da7++67DZPJZOzbt882T5Lh6+trN++XX34xJBlvvPFGkX390dy5cw1JxkcffWSbl5uba8THxxtBQUF2x96gQQPj1ltvvWx7jmb683EX+nPfF7bp5+dn7N+/3zbv7bffNiQZUVFRdlknTpxoSLJbt7D/Z82aZZuXk5NjtGvXzoiIiDByc3MNwzCMDz/80PDy8jK+//57u/3Pnz/fkGT88MMPdpm8vLyMnTt3XvGcnDhxwvD19TV69eplWCwW2/x58+YZkoz33nuvyPFf7r1eaOTIkUXOlWEYxv79+w1JRkhIiHHixIlil/3xvThkyBBDkvGPf/zDNs9qtRq33nqr4evra8syevRoIyQkxMjPz79ituL2GRAQYBw5csQ2f9OmTYYkY+zYsbZ53bt3N2JjY42LFy/aZbnhhhuMZs2a2eYV/lzp0qVLqfIUrv/H94Vh/P/P++rVqw3DMIytW7cakozPPvusxLYOHDhgmM1m45///Kfd/O3btxve3t528y/3sweAe+O2QAAeKygo6LKjBhb+FfiLL74o8+APfn5+GjZsWKnXHzx4sIKDg23Td999t2rXrq1vv/22TPsvrW+//VZms1mPP/643fzx48fLMAwtXbrUbn6PHj3UpEkT23Tbtm0VEhKi33///Yr7iYqK0qBBg2zzfHx89PjjjysrK0tr164t8zGUNdPldO/e3e7KUKdOnSRJd911l10/Fc7/8768vb31yCOP2KZ9fX31yCOP6MSJE0pKSpIkffbZZ2rVqpVatmypU6dO2V4333yzJGn16tV2bXbt2rVUz6J99913ys3N1ZgxY+Tl9f//dz1ixAiFhITom2++Kc0pcNhdd911xauMfzRq1Cjbvwtve8zNzdV3330n6dLn8Pz580pMTCxTngEDBqhOnTq26euuu06dOnWyfabOnDmjVatWaeDAgcrMzLSd/9OnT6t3797au3evjh49atfmiBEjZDaby5SnOIVXppYvX67s7Oxi1/n8889ltVo1cOBAu/dJVFSUmjVrVuR94ujPHgDugeIKgMfKysqy+wX5z+6991517txZDz30kCIjI3Xfffdp4cKFDhVaderUcWjwimbNmtlNm0wmNW3a9LLPGznDwYMHFR0dXeR8tGrVyrb8j+rXr1+kjRo1aujs2bNX3E+zZs3sftm/3H4cUdZMjrRZ+EtwvXr1ip3/531FR0cXGeygefPmkmTr071792rnzp0KDw+3exWuVzjYRKFGjRqVKnvhuWzRooXdfF9fXzVu3PiqzvXllDafJHl5ealx48Z28/58fh577DE1b95cffv2Vd26dfXggw+W+vk+qehnqnAfhe3v27dPhmHo+eefL9IHhc+ilbUPSqtRo0YaN26c/v3vf6tWrVrq3bu3EhIS7J632rt3rwzDULNmzYrk3LVrV5GMjv7sAeAeeOYKgEc6cuSI0tPTLzvsdEBAgNatW6fVq1frm2++0bJly/Tpp5/q5ptv1ooVK0r1l2tHnpMqrZK+6NhisTj1r+mXU9J+jD8NRFCRSpPpcufOkTadefxWq1WxsbGaPXt2scv/XMiVx3vKmZydLyIiQsnJyVq+fLmWLl2qpUuX6v3339fgwYPtBkIpq8I/ljzxxBPq3bt3sev8+edEaY/RkffbrFmzNHToUH3xxRdasWKFHn/8cU2fPl0bN25U3bp1ZbVaZTKZtHTp0mLff0FBQWXKCMC9UFwB8EiFD+OX9MtUIS8vL3Xv3l3du3fX7Nmz9dJLL+nZZ5/V6tWr1aNHjxJ/eSqrvXv32k0bhqF9+/bZDTBQo0YNu9HeCh08eNDuKoAj2Ro0aKDvvvtOmZmZdlevdu/ebVvuDA0aNNC2bdtktVrtrl45ez8ludy5Kw/Hjh0rMlT3b7/9Jkm22w2bNGmiX375Rd27d3fq+6nwXO7Zs8fufZGbm6v9+/erR48eZWrXmRmtVqt+//1329Uqqej5kS5dbevXr5/69esnq9Wqxx57TG+//baef/75K34v158/U4X7KGy/8Nz4+PiU+ZyUpEaNGpJU5D1X0vstNjZWsbGxeu655/Tjjz+qc+fOmj9/vl588UU1adJEhmGoUaNGducLQOXCbYEAPM6qVav0wgsvqFGjRnrggQdKXO/MmTNF5hV+GW/hENmFvzQX9wt7WRSObFZo0aJFSk1NVd++fW3zmjRpoo0bNyo3N9c27+uvvy4yHLMj2W655RZZLBbNmzfPbv6cOXNkMpns9n81brnlFh0/flyffvqpbV5+fr7eeOMNBQUFqWvXrk7ZT0maNGmi9PR0bdu2zTYvNTVVixcvLpf95efn6+2337ZN5+bm6u2331Z4eLg6duwoSRo4cKCOHj2qd955p8j2Fy5cKPNodD169JCvr69ef/11uytq7777rtLT03XrrbeWqV1nv+f/+J4zDEPz5s2Tj4+PunfvLunSqHd/5OXlZftjw5+Hqi/OkiVL7J6Z+umnn7Rp0ybbezoiIkLdunXT22+/rdTU1CLbnzx50vGDKlD4DOC6dets8ywWi/71r3/ZrZeRkaH8/Hy7ebGxsfLy8rId45133imz2aypU6cWuUJqGEaR8wTAM3HlCoBbW7p0qXbv3q38/HylpaVp1apVSkxMVIMGDfTll19e9otNp02bpnXr1unWW29VgwYNdOLECb355puqW7euunTpIunSL0/Vq1fX/PnzFRwcrGrVqqlTp05lfiYjLCxMXbp00bBhw5SWlqa5c+eqadOmdsPFP/TQQ1q0aJH69OmjgQMHKiUlRR999JHdYA6OZuvXr59uuukmPfvsszpw4IDi4uK0YsUKffHFFxozZkyRtsvq4Ycf1ttvv62hQ4cqKSlJDRs21KJFi/TDDz9o7ty5l30Gzhnuu+8+Pf3007rjjjv0+OOPKzs7W2+99ZaaN2+uLVu2OH1/0dHRevnll3XgwAE1b95cn376qZKTk/Wvf/3L9kXOf/vb37Rw4UL9/e9/1+rVq9W5c2dZLBbt3r1bCxcu1PLly+2+CLu0wsPDNXHiRE2dOlV9+vRR//79tWfPHr355pu69tpr9de//rVMx1RYFD7++OPq3bu3zGaz7rvvvjK15e/vr2XLlmnIkCHq1KmTli5dqm+++UbPPPOMbVCMhx56SGfOnNHNN9+sunXr6uDBg3rjjTfUrl0727N6l9O0aVN16dJFjz76qHJycjR37lzVrFlTTz31lG2dhIQEdenSRbGxsRoxYoQaN26stLQ0bdiwQUeOHNEvv/xSpuOLiYnR9ddfr4kTJ+rMmTMKCwvTJ598UqSQWrVqlUaNGqV77rlHzZs3V35+vj788EOZzWbdddddki59nl988UVNnDhRBw4c0IABAxQcHKz9+/dr8eLFevjhh/XEE0+UKScAN+KaQQoB4PIKh0AufPn6+hpRUVFGz549jddee81uGO1Cfx6Oe+XKlcbtt99uREdHG76+vkZ0dLQxaNAg47fffrPb7osvvjBat25teHt72w033bVrVyMmJqbYfCUNxf6///3PmDhxohEREWEEBAQYt956q3Hw4MEi28+aNcuoU6eO4efnZ3Tu3NnYvHlzkTYvl624IckzMzONsWPHGtHR0YaPj4/RrFkz45VXXjGsVqvdepKMkSNHFslU0hDxf5aWlmYMGzbMqFWrluHr62vExsYWO1y8o0OxlzbTihUrjDZt2hi+vr5GixYtjI8++qjEodj/3Gbh8N6vvPKK3fzC/vvjUNqF/b9582YjPj7e8Pf3Nxo0aGDMmzevSM7c3Fzj5ZdfNmJiYgw/Pz+jRo0aRseOHY2pU6ca6enpVzzOy5k3b57RsmVLw8fHx4iMjDQeffRR4+zZs3brODIUe35+vvGPf/zDCA8PN0wmk+28lXRu/rjsz0OxV6tWzUhJSTF69eplBAYGGpGRkcbkyZPtho5ftGiR0atXLyMiIsLw9fU16tevbzzyyCNGamrqZXP+Mc+sWbOMevXqGX5+fsZf/vIX45dffimyfkpKijF48GAjKirK8PHxMerUqWPcdtttxqJFi2zrlOYrHoprt0ePHoafn58RGRlpPPPMM0ZiYqLdUOy///678eCDDxpNmjQx/P39jbCwMOOmm24yvvvuuyLt/d///Z/RpUsXo1q1aka1atWMli1bGiNHjjT27NljW+dyP3sAuDeTYbjw6WUAAOCRhg4dqkWLFikrK6tc2j9w4IAaNWqkV155hSs6ADwGz1wBAAAAgBNQXAEAAACAE1BcAQAAAIAT8MwVAAAAADgBV64AAAAAwAkorgAAAADACfgS4WJYrVYdO3ZMwcHBMplMro4DAAAAwEUMw1BmZqaio6Pl5XX5a1MUV8U4duyY6tWr5+oYAAAAANzE4cOHVbdu3cuuQ3FVjODgYEmXTmBISIjLcuTl5WnFihXq1auXfHx8XJajqkhOTlbXrl21du1atWvXrsL3T39XPfR51UJ/Vy30d9VCf1duGRkZqlevnq1GuByKq2IU3goYEhLi8uIqMDBQISEhfFArQO3atdWrVy/Vrl3bJf1Of1c99HnVQn9XLfR31UJ/Vw2leVyI4goo0KxZMy1fvtzVMQAAAOChGC0QKGCxWJSRkSGLxeLqKAAAAPBAFFdAgV9++UWhoaH65ZdfXB0FAAAAHojiCgAAAACcgOIKAAAAAJyA4goAAAAAnIDiCgAAAACcgKHYgQKxsbE6ceKEqlev7uooAAAA8EAUV0ABHx8fhYeHuzoGAAAAPBS3BQIFUlJS1L9/f6WkpLg6CgAAADwQxRVQID09XV999ZXS09NdHQUAAAAeiOIKAAAAAJyA4goAAAAAnIDiCgAAAACcgNECPUBGRoZyc3Ovup3AwECFhoY6IVHlVKdOHc2aNUt16tRxdRQAAAB4IIorD/Dq3ASdSM+66nZqBgfq2afGUmCVIDIyUuPGjXN1DAAAAHgoiisPcCbrgsJiuigoNKzMbWSln9HpneuVnZ1NcVWCs2fP6rvvvlOPHj1Uo0YNV8cBAACAh6G48hBBoWEKqRlxVW2ccVKWymr//v0aOHCgkpKSKK4AAADgMAa0AAAAAAAnoLgCAAAAACeguAIAAAAAJ6C4AgoEBASoffv2CggIcHUUAAAAeCAGtAAKtGrVSlu2bHF1DAAAAHgorlwBAAAAgBNQXAEFtm7dKj8/P23dutXVUQAAAOCBKK6AAoZhKDc3V4ZhuDoKAAAAPBDFFQAAAAA4AcUVAAAAADgBxRUAAAAAOAFDsQMFWrVqpR07dqhx48aujgIAAAAPRHEFFAgICFBMTIyrYwAAAMBDcVsgUODgwYN66KGHdPDgQVdHAQAAgAeiuAIKnD59Wu+++65Onz7t6igAAADwQBRXAAAAAOAEFFcAAAAA4AQUVwAAAADgBBRXQIHIyEhNmDBBkZGRro4CAAAAD8RQ7ECBOnXqaPr06a6OAQAAAA/FlSugQGZmptasWaPMzExXRwEAAIAHorgCCuzdu1c33XST9u7d6+ooAAAA8EAuLa7WrVunfv36KTo6WiaTSUuWLLFbbjKZin298sorJbY5ZcqUIuu3bNmynI8EAAAAQFXn0uLq/PnziouLU0JCQrHLU1NT7V7vvfeeTCaT7rrrrsu2GxMTY7fd+vXryyM+AAAAANi4dECLvn37qm/fviUuj4qKspv+4osvdNNNN6lx48aXbdfb27vItgAAAABQnjxmtMC0tDR98803+uCDD6647t69exUdHS1/f3/Fx8dr+vTpql+/fonr5+TkKCcnxzadkZEhScrLy1NeXt7Vhy+jwn17m83yklUmw1LmtrxklbfZLIvF4tJjcnd16tSRJJeco8J90j9VB31etdDfVQv9XbXQ35WbI/1qMgzDKMcspWYymbR48WINGDCg2OUzZ87UjBkzdOzYMfn7+5fYztKlS5WVlaUWLVooNTVVU6dO1dGjR7Vjxw4FBwcXu82UKVM0derUIvM//vhjBQYGlul4AAAAAHi+7Oxs3X///UpPT1dISMhl1/WY4qply5bq2bOn3njjDYfaPXfunBo0aKDZs2dr+PDhxa5T3JWrevXq6dSpU1c8geUpLy9PiYmJWvH9z4q6ppeCw8LL3FbmmZM6smmpnh8/ki/JdVOF/d2zZ0/5+Pi4Og4qAH1etdDfVQv9XbXQ35VbRkaGatWqVariyiNuC/z++++1Z88effrppw5vW716dTVv3lz79u0rcR0/Pz/5+fkVme/j4+MWH5B8i0VWeckwmcvchlVeyrdYZDab3eKY3NH27dvVt29fLV26VLGxsS7L4S7vO1Qc+rxqob+rFvq7aqG/KydH+tQjvufq3XffVceOHRUXF+fwtllZWUpJSVHt2rXLIRkqk7y8PB09epT7pQEAAFAmLi2usrKylJycrOTkZEnS/v37lZycrEOHDtnWycjI0GeffaaHHnqo2Da6d++uefPm2aafeOIJrV27VgcOHNCPP/6oO+64Q2azWYMGDSrXYwEAAABQtbn0tsDNmzfrpptusk2PGzdOkjRkyBAtWLBAkvTJJ5/IMIwSi6OUlBSdOnXKNn3kyBENGjRIp0+fVnh4uLp06aKNGzcqPLzszysBAAAAwJW4tLjq1q2brjSexsMPP6yHH364xOUHDhywm/7kk0+cEQ0AAAAAHOIRz1wBFaFZs2ZavXq1mjVr5uooAAAA8EAeMVogUBGCg4PVrVs3V8cAAACAh+LKFVDg6NGjmjhxoo4ePerqKAAAAPBAFFdAgbS0NM2YMUNpaWmujgIAAAAPRHEFAAAAAE5AcQUAAAAATkBxBQAAAABOQHEFFKhZs6aGDx+umjVrujoKAAAAPBBDsQMFGjRooH//+9+ujgEAAAAPxZUroMCFCxe0c+dOXbhwwdVRAAAA4IEoroACu3btUps2bbRr1y5XRwEAAIAHorgCAAAAACeguAIAAAAAJ6C4AgAAAAAnoLgCCphMJvn6+spkMrk6CgAAADwQQ7EDBdq3b6+cnBxXxwAAAICH4soVAAAAADgBxRVQYNeuXerQoQNDsQMAAKBMKK6AAhcuXNDWrVv5EmEAAACUCcUVAAAAADgBxRUAAAAAOAHFFQAAAAA4AcUVUKBRo0ZauHChGjVq5OooAAAA8EB8zxVQoEaNGrrnnntcHQMAAAAeiitXQIG0tDTNnj1baWlpro4CAAAAD0RxBRQ4evSoxo8fr6NHj7o6CgAAADwQxRUAAAAAOAHFFQAAAAA4AcUVAAAAADgBxRVQIDQ0VP369VNoaKirowAAAMADMRQ7UKBJkyb68ssvXR0DAAAAHoorV0CBvLw8nTx5Unl5ea6OAgAAAA9EcQUU2L59uyIiIrR9+3ZXRwEAAIAHorgCAAAAACeguAIAAAAAJ6C4AgAAAAAnoLgCAAAAACdwaXG1bt069evXT9HR0TKZTFqyZInd8qFDh8pkMtm9+vTpc8V2ExIS1LBhQ/n7+6tTp0766aefyukIUJnExcUpPT1dcXFxro4CAAAAD+TS4ur8+fOKi4tTQkJCiev06dNHqampttf//ve/y7b56aefaty4cZo8ebK2bNmiuLg49e7dWydOnHB2fFQyZrNZISEhMpvNro4CAAAAD+TS4qpv37568cUXdccdd5S4jp+fn6KiomyvGjVqXLbN2bNna8SIERo2bJhat26t+fPnKzAwUO+9956z46OS2bt3r3r37q29e/e6OgoAAAA8kLerA1zJmjVrFBERoRo1aujmm2/Wiy++qJo1axa7bm5urpKSkjRx4kTbPC8vL/Xo0UMbNmwocR85OTnKycmxTWdkZEi69KWyrvxC2cJ9e5vN8pJVJsNS5ra8ZJW32SyLxcKX5Jbg7NmzWrFihc6ePeuSc1S4T/qn6qDPqxb6u2qhv6sW+rtyc6Rf3bq46tOnj+688041atRIKSkpeuaZZ9S3b19t2LCh2Fu3Tp06JYvFosjISLv5kZGR2r17d4n7mT59uqZOnVpk/ooVKxQYGHj1B3KVev3lWklnpQtny95IgNS22/VKSkpyWq7KJiUlRZK0fv16paamuixHYmKiy/YN16DPqxb6u2qhv6sW+rtyys7OLvW6bl1c3XfffbZ/x8bGqm3btmrSpInWrFmj7t27O20/EydO1Lhx42zTGRkZqlevnnr16qWQkBCn7cdReXl5SkxM1Irvf1bUNb0UHBZe5rYyz5zUkU1L9fz4kUWKT1yydetWSVKXLl3Uvn37Ct9/YX/37NlTPj4+Fb5/VDz6vGqhv6sW+rtqob8rt8K72krDrYurP2vcuLFq1aqlffv2FVtc1apVS2azWWlpaXbz09LSFBUVVWK7fn5+8vPzKzLfx8fHLT4g+RaLrPKSYSr7QAtWeSnfYpHZbHaLY3JH3t7etv+68hy5y/sOFYc+r1ro76qF/q5a6O/KyZE+9ajvuTpy5IhOnz6t2rVrF7vc19dXHTt21MqVK23zrFarVq5cqfj4+IqKCQ9Vr149zZs3T/Xq1XN1FAAAAHgglxZXWVlZSk5OVnJysiRp//79Sk5O1qFDh5SVlaUnn3xSGzdu1IEDB7Ry5Urdfvvtatq0qXr37m1ro3v37po3b55tety4cXrnnXf0wQcfaNeuXXr00Ud1/vx5DRs2rKIPDx4mPDxcI0eOVHh42W+/BAAAQNXl0tsCN2/erJtuusk2Xfjc05AhQ/TWW29p27Zt+uCDD3Tu3DlFR0erV69eeuGFF+xu4UtJSdGpU6ds0/fee69OnjypSZMm6fjx42rXrp2WLVvGc0a4ojNnzujbb7/VLbfcorCwMFfHAQAAgIdxaXHVrVs3GYZR4vLly5dfsY0DBw4UmTdq1CiNGjXqaqKhCjpw4ID+9re/KSkpieIKAAAADvOoZ64AAAAAwF1RXAEAAACAE1BcAQAAAIATUFwBBapVq6brr79e1apVc3UUAAAAeCCP+hJhoDy1aNFCGzZscHUMAAAAeCiuXAEAAACAE1BcAQW2bNkik8mkLVu2uDoKAAAAPBDFFQAAAAA4AcUVAAAAADgBxRUAAAAAOAHFFQAAAAA4AUOxAwVat26tvXv3qm7duq6OAgAAAA9EcQUU8Pf3V9OmTV0dAwAAAB6K2wKBAvv379df//pX7d+/39VRAAAA4IEoroACZ8+e1X//+1+dPXvW1VEAAADggSiuAAAAAMAJKK4AAAAAwAkorgAAAADACSiugAK1a9fW5MmTVbt2bVdHAQAAgAdiKHagQO3atTVlyhRXxwAAAICH4soVUCAjI0PLly9XRkaGq6MAAADAA1FcAQX27dunPn36aN++fa6OAgAAAA9EcQUAAAAATkBxBQAAAABOQHEFAAAAAE5AcQUU8PPzU5MmTeTn5+fqKAAAAPBADMUOFIiJiWEwCwAAAJQZV64AAAAAwAkoroAC27ZtU3h4uLZt2+bqKAAAAPBAFFdAgfz8fJ06dUr5+fmujgIAAAAPRHEFAAAAAE5AcQUAAAAATkBxBQAAAABOQHEFFGjevLl+/PFHNW/e3NVRAAAA4IEcLq4OHz6sI0eO2KZ/+uknjRkzRv/617+cGgyoaEFBQYqPj1dQUJCrowAAAMADOVxc3X///Vq9erUk6fjx4+rZs6d++uknPfvss5o2bZrTAwIV5ciRIxo3bpzdHw8AAACA0nK4uNqxY4euu+46SdLChQvVpk0b/fjjj/rvf/+rBQsWODsfUGFOnDihOXPm6MSJE66OAgAAAA/kcHGVl5cnPz8/SdJ3332n/v37S5Jatmyp1NRU56YDAAAAAA/hcHEVExOj+fPn6/vvv1diYqL69OkjSTp27Jhq1qzpUFvr1q1Tv379FB0dLZPJpCVLltiW5eXl6emnn1ZsbKyqVaum6OhoDR48WMeOHbtsm1OmTJHJZLJ7tWzZ0tHDBAAAAACHOFxcvfzyy3r77bfVrVs3DRo0SHFxcZKkL7/80na7YGmdP39ecXFxSkhIKLIsOztbW7Zs0fPPP68tW7bo888/1549e2xXyi4nJiZGqampttf69esdygUAAAAAjvJ2ZGXDMNS4cWMdOnRI+fn5qlGjhm3Zww8/rMDAQId23rdvX/Xt27fYZaGhoUpMTLSbN2/ePF133XU6dOiQ6tevX2K73t7eioqKKnWOnJwc5eTk2KYzMjIkXbp6lpeXV+p2nK1w395ms7xklcmwlLktL1nlbTbLYrG49JjcWWhoqP7+978rNDTUJeeocJ/0T9VBn1ct9HfVQn9XLfR35eZIv5oMwzBKu7LVapW/v7927typZs2alSlciUFMJi1evFgDBgwocZ3vvvtOvXr10rlz5xQSElLsOlOmTNErr7yi0NBQ+fv7Kz4+XtOnT79sMTZlyhRNnTq1yPyPP/7Y4YIRAAAAQOWRnZ2t+++/X+np6SXWIIUcKq6kS7fcvfvuu7r++uuvKmSRIFcori5evKjOnTurZcuW+u9//1tiO0uXLlVWVpZatGih1NRUTZ06VUePHtWOHTsUHBxc7DbFXbmqV6+eTp06dcUTWJ7y8vKUmJioFd//rKhreik4LLzMbWWeOakjm5bq+fEjFRkZ6cSUlUd2drb27NmjFi1auKSoLuzvnj17ysfHp8L3j4pHn1ct9HfVQn9XLfR35ZaRkaFatWqVqrhy6LZASZoxY4aefPJJvfXWW2rTpk2ZQzoiLy9PAwcOlGEYeuutty677h9vM2zbtq06deqkBg0aaOHChRo+fHix2/j5+dlGQPwjHx8ft/iA5FsssspLhslc5jas8lK+xSKz2ewWx+SOUlJS1KlTJyUlJalDhw4uy+Eu7ztUHPq8aqG/qxb6u2qhvysnR/rU4eJq8ODBys7OVlxcnHx9fRUQEGC3/MyZM442eVmFhdXBgwe1atUqh68kVa9eXc2bN9e+ffucmgsAAAAA/sjh4mru3LnlEKN4hYXV3r17tXr1aoeHepekrKwspaSk6G9/+1s5JAQAAACASxwuroYMGeK0nWdlZdldUdq/f7+Sk5MVFham2rVr6+6779aWLVv09ddfy2Kx6Pjx45KksLAw+fr6SpK6d++uO+64Q6NGjZIkPfHEE+rXr58aNGigY8eOafLkyTKbzRo0aJDTcgMAAADAnzlcXEmXnk15//33lZKSotdee00RERFaunSp6tevr5iYmFK3s3nzZt1000226XHjxkm6VMBNmTJFX375pSSpXbt2dtutXr1a3bp1s2U5deqUbdmRI0c0aNAgnT59WuHh4erSpYs2btyo8PCyDwaBqsHLy0vBwcHy8nL4698AAAAAx4urtWvXqm/fvurcubPWrVunf/7zn4qIiNAvv/yid999V4sWLSp1W926ddPlBisszUCGBw4csJv+5JNPSr1/4I/atWtn+44zAAAAwFEO/4l+woQJevHFF5WYmGi7NU+Sbr75Zm3cuNGp4QAAAADAUzhcXG3fvl133HFHkfkRERF2t+cBnubXX39VTEyMfv31V1dHAQAAgAdyuLiqXr26UlNTi8zfunWr6tSp45RQgCtcvHhRv/76qy5evOjqKAAAAPBADhdX9913n55++mkdP35cJpNJVqtVP/zwg5544gkNHjy4PDICAAAAgNtzuLh66aWX1LJlS9WrV09ZWVlq3bq1brzxRt1www167rnnyiMjAAAAALg9h0cL9PX11TvvvKNJkyZp+/btysrKUvv27dWsWbPyyAcAAAAAHqFM33MlSfXq1VO9evVksVi0fft2nT17VjVq1HBmNqBCNW7cWF988YUaN27s6igAAADwQA7fFjhmzBi9++67kiSLxaKuXbuqQ4cOqlevntasWePsfECFqV69uvr376/q1au7OgoAAAA8kMPF1aJFixQXFydJ+uqrr/T7779r9+7dGjt2rJ599lmnBwQqyvHjxzV9+nQdP37c1VEAAADggRwurk6dOqWoqChJ0rfffquBAweqefPmevDBB7V9+3anBwQqyrFjx/TMM8/o2LFjro4CAAAAD+RwcRUZGalff/1VFotFy5YtU8+ePSVJ2dnZMpvNTg8IAAAAAJ7A4QEthg0bpoEDB6p27doymUzq0aOHJGnTpk1q2bKl0wMCAAAAgCdwuLiaMmWK2rRpo8OHD+uee+6Rn5+fJMlsNmvChAlODwgAAAAAnqBMQ7HffffdReYNGTLkqsMArlS9enXdfffdjBYIAACAMnG4uJo2bdpll0+aNKnMYQBXaty4sT777DNXxwAAAICHcri4Wrx4sd10Xl6e9u/fL29vbzVp0oTiCh4rNzdXJ06cUEREhHx9fV0dBwAAAB7G4eJq69atReZlZGRo6NChuuOOO5wSCnCFHTt2qGPHjkpKSlKHDh1cHQcAAAAexuGh2IsTEhKiqVOn6vnnn3dGcwAAAADgcZxSXElSenq60tPTndUcAAAAAHgUh28LfP311+2mDcNQamqqPvzwQ/Xt29dpwQAAAADAkzhcXM2ZM8du2svLS+Hh4RoyZIgmTpzotGAAAAAA4EkcLq72799fHjkAl2vXrp0uXrwoHx8fV0cBAACAByrTlwgDlZGXl5f8/PxcHQMAAAAeymkDWgCe7rffflO3bt3022+/uToKAAAAPBDFFVAgKytLa9euVVZWlqujAAAAwANRXAEAAACAE5SquOrQoYPOnj0rSZo2bZqys7PLNRQAAAAAeJpSFVe7du3S+fPnJUlTp07ltikAAAAA+JNSjRbYrl07DRs2TF26dJFhGHr11VcVFBRU7LqTJk1yakCgotSvX1/vvPOO6tev7+ooAAAA8EClKq4WLFigyZMn6+uvv5bJZNLSpUvl7V10U5PJRHEFj1WrVi099NBDro4BAAAAD1Wq4qpFixb65JNPJF36LqCVK1cqIiKiXIMBFe3UqVNasmSJBgwYoFq1ark6DgAAADyMw6MFWq1WCitUSocOHdKIESN06NAhV0cBAACAByrVlas/S0lJ0dy5c7Vr1y5JUuvWrTV69Gg1adLEqeEAAAAAwFM4fOVq+fLlat26tX766Se1bdtWbdu21aZNmxQTE6PExMTyyAgAAAAAbs/hK1cTJkzQ2LFjNWPGjCLzn376afXs2dNp4QAAAADAUzh85WrXrl0aPnx4kfkPPvigfv31V6eEAlwhKChIXbt2LfFrBgAAAIDLcfjKVXh4uJKTk9WsWTO7+cnJyQx0AY/WvHlzrVmzxtUxAAAA4KEcvnI1YsQIPfzww3r55Zf1/fff6/vvv9eMGTP0yCOPaMSIEQ61tW7dOvXr10/R0dEymUxasmSJ3XLDMDRp0iTVrl1bAQEB6tGjh/bu3XvFdhMSEtSwYUP5+/urU6dO+umnnxzKharJarUqJydHVqvV1VEAAADggRwurp5//nlNmjRJb7zxhrp27aquXbtq3rx5mjJlip577jmH2jp//rzi4uKUkJBQ7PKZM2fq9ddf1/z587Vp0yZVq1ZNvXv31sWLF0ts89NPP9W4ceM0efJkbdmyRXFxcerdu7dOnDjhUDZUPcnJyfL391dycrKrowAAAMADOXxboMlk0tixYzV27FhlZmZKkoKDg8u08759+6pv377FLjMMQ3PnztVzzz2n22+/XZL0n//8R5GRkVqyZInuu+++YrebPXu2RowYoWHDhkmS5s+fr2+++UbvvfeeJkyYUKacAAAAAHAlZfqeq0JlLapKY//+/Tp+/Lh69OhhmxcaGqpOnTppw4YNxRZXubm5SkpK0sSJE23zvLy81KNHD23YsKHEfeXk5CgnJ8c2nZGRIUnKy8tTXl6eMw6nTAr37W02y0tWmQxLmdvyklXeZrMsFotLj8md5efn2/7rinNUuE/6p+qgz6sW+rtqob+rFvq7cnOkX6+quCpPx48flyRFRkbazY+MjLQt+7NTp07JYrEUu83u3btL3Nf06dM1derUIvNXrFihwMBAR6M7Xa+/XCvprHThbNkbCZDadrteSUlJTstV2aSkpEiS1q9fr9TUVJfl4Pviqh76vGqhv6sW+rtqob8rp+zs7FKv67bFVUWaOHGixo0bZ5vOyMhQvXr11KtXL4WEhLgsV15enhITE7Xi+58VdU0vBYeFl7mtzDMndWTTUj0/fmSR4hOXbN26VZLUpUsXtW/fvsL3X9jfPXv2lI+PT4XvHxWPPq9a6O+qhf6uWujvyq3wrrbScNviKioqSpKUlpam2rVr2+anpaWpXbt2xW5Tq1Ytmc1mpaWl2c1PS0uztVccPz8/+fn5FZnv4+PjFh+QfItFVnnJMJnL3IZVXsq3WGQ2m93imNxRu3btdPjwYUVERLj0HLnL+w4Vhz6vWujvqoX+rlro78rJkT51aLTAvLw8de/evVTDoV+tRo0aKSoqSitXrrTNy8jI0KZNmxQfH1/sNr6+vurYsaPdNlarVStXrixxG6CQr6+v6tatK19fX1dHAQAAgAdyqLjy8fHRtm3bnLbzrKwsJScn24a+3r9/v5KTk3Xo0CGZTCaNGTNGL774or788ktt375dgwcPVnR0tAYMGGBro3v37po3b55tety4cXrnnXf0wQcfaNeuXXr00Ud1/vx52+iBQEl+//133XPPPfr9999dHQUAAAAeyOHvufrrX/+qd9991yk737x5s9q3b297vmXcuHFq3769Jk2aJEl66qmn9I9//EMPP/ywrr32WmVlZWnZsmXy9/e3tZGSkqJTp07Zpu+99169+uqrmjRpktq1a6fk5GQtW7aM54xwRefOndOiRYt07tw5V0cBAACAB3L4mav8/Hy99957+u6779SxY0dVq1bNbvns2bNL3Va3bt1kGEaJy00mk6ZNm6Zp06aVuM6BAweKzBs1apRGjRpV6hwAAAAAcLUcLq527NihDh06SJJ+++03u2Umk8k5qQAAAADAwzhcXK1evbo8cgAAAACAR3P4matC+/bt0/Lly3XhwgVJuuztfYAniI6O1ksvvaTo6GhXRwEAAIAHcri4On36tLp3767mzZvrlltuUWpqqiRp+PDhGj9+vNMDAhUlKipKEydOvOx3ogEAAAAlcbi4Gjt2rHx8fHTo0CEFBgba5t97771atmyZU8MBFencuXP68ssvGS0QAAAAZeJwcbVixQq9/PLLqlu3rt38Zs2a6eDBg04LBlS033//XbfffjvfcwUAAIAycbi4On/+vN0Vq0JnzpyRn5+fU0IBAAAAgKdxuLj6y1/+ov/85z+2aZPJJKvVqpkzZ+qmm25yajgAAAAA8BQOD8U+c+ZMde/eXZs3b1Zubq6eeuop7dy5U2fOnNEPP/xQHhkBAAAAwO05fOWqTZs2+u2339SlSxfdfvvtOn/+vO68805t3bpVTZo0KY+MQIXw9/dX69at5e/v7+ooAAAA8EAOX7mSpNDQUD377LPOzgK4VOvWrbVz505XxwAAAICHKlNxdfbsWb377rvatWuXpEu/lA4bNkxhYWFODQf3k56eruzs7KtuJzAwUKGhoU5IBAAAALgHh4urdevWqV+/fgoNDdU111wjSXr99dc1bdo0ffXVV7rxxhudHhLuIT09Xf+cOUenM6++uKoZHKhnnxrrVgVWcnKybrzxRq1bt07t2rVzdRwAAAB4GIeLq5EjR+ree+/VW2+9JbPZLEmyWCx67LHHNHLkSG3fvt3pIeEesrOzdTozW2ExXRQUWvarlFnpZ3R653plZ2e7VXFltVqVmZkpq9Xq6igAAADwQA4XV/v27dOiRYtshZUkmc1mjRs3zm6IdlReQaFhCqkZcVVtnHFSFgAAAMBdODxaYIcOHWzPWv3Rrl27FBcX55RQAAAAAOBpSnXlatu2bbZ/P/744xo9erT27dun66+/XpK0ceNGJSQkaMaMGeWTEgAAAADcXKmKq3bt2slkMskwDNu8p556qsh6999/v+69917npQMqUMuWLZWUlKSWLVu6OgoAAAA8UKmKq/3795d3DsDlAgMD1aFDB1fHAAAAgIcqVXHVoEGD8s4BuNyhQ4f08ssv6+mnn1b9+vVdHQcAAAAepkxfInzs2DGtX79eJ06cKDJs9eOPP+6UYEBFO3XqlN58800NHz6c4goAAAAOc7i4WrBggR555BH5+vqqZs2aMplMtmUmk4niCgAAAECV5HBx9fzzz2vSpEmaOHGivLwcHskdAAAAAColh6uj7Oxs3XfffRRWAAAAAPAHDldIw4cP12effVYeWQCXioiI0NixYxUREeHqKAAAAPBADt8WOH36dN12221atmyZYmNj5ePjY7d89uzZTgsHVKS6devy/gUAAECZlam4Wr58uVq0aCFJRQa0ADxVVlaWtm/frtjYWAUFBbk6DgAAADyMw8XVrFmz9N5772no0KHlEAdwnd9++0033HCDkpKS+DJhAAAAOMzhZ678/PzUuXPn8sgCAAAAAB7L4eJq9OjReuONN8ojCwAAAAB4LIdvC/zpp5+0atUqff3114qJiSkyoMXnn3/utHAAAAAA4CkcLq6qV6+uO++8szyyAC7l7e2tWrVqydvb4Y8FAAAA4Hhx9f7775dHDsDl2rZtq5MnT7o6BgAAADyUw89cAQAAAACKcvjKVaNGjS77fVa///77VQUCXGXnzp26/fbb9cUXXygmJsbVcQAAAOBhHC6uxowZYzedl5enrVu3atmyZXryySedlQuocDk5OUpJSVFOTo6rowAAAMADOVxcjR49utj5CQkJ2rx581UHAgAAAABP5LRnrvr27av/+7//c1ZzNg0bNpTJZCryGjlyZLHrL1iwoMi6/v7+Ts8FAAAAAH/ktDGnFy1apLCwMGc1Z/Pzzz/LYrHYpnfs2KGePXvqnnvuKXGbkJAQ7dmzxzZ9uWfEAAAAAMAZHC6u2rdvb1esGIah48eP6+TJk3rzzTedGk6SwsPD7aZnzJihJk2aqGvXriVuYzKZFBUV5fQsqNyaNm2qZcuWqWnTpq6OAgAAAA/kcHE1YMAAu2kvLy+Fh4erW7duatmypbNyFSs3N1cfffSRxo0bd9mrUVlZWWrQoIGsVqs6dOigl1566bKjv+Xk5NgNYpCRkSHp0mAdeXl5zjsABxXu29tslpesMhmWK2xRMi9Z5W02y2KxlPmYLBaL22QpDwEBAbr55pslySW5CvfpTucE5Ys+r1ro76qF/q5a6O/KzZF+NRmGYZRjFqdauHCh7r//fh06dEjR0dHFrrNhwwbt3btXbdu2VXp6ul599VWtW7dOO3fuVN26dYvdZsqUKZo6dWqR+R9//LECAwOdegxwX2fOnNHy5cvVu3fvcrnFFQAAAJ4nOztb999/v9LT0xUSEnLZdT2quOrdu7d8fX311VdflXqbvLw8tWrVSoMGDdILL7xQ7DrFXbmqV6+eTp06dcUTWJ7y8vKUmJioFd//rKhreik4LPzKG5Ug88xJHdm0VM+PH6nIyMgytZGWlqYXZiWobqe+Ls9SHrZu3apOnTpp06ZNat++fYXvv7C/e/bsKR8fnwrfPyoefV610N9VC/1dtdDflVtGRoZq1apVquKq1LcFenl5XXFgCJPJpPz8/NI26ZCDBw/qu+++0+eff+7Qdj4+Pmrfvr327dtX4jp+fn7y8/Mrdlt3+IDkWyyyykuGyVzmNqzyUr7FIrPZXOZjMpvNbpOlPHh7e9v+68pc7vK+Q8Whz6sW+rtqob+rFvq7cnKkT0tdXC1evLjEZRs2bNDrr78uq9Va6h076v3331dERIRuvfVWh7azWCzavn27brnllnJKBgAAAAAOFFe33357kXl79uzRhAkT9NVXX+mBBx7QtGnTnBqukNVq1fvvv68hQ4bYri4UGjx4sOrUqaPp06dLkqZNm6brr79eTZs21blz5/TKK6/o4MGDeuihh8olGwAAAABIZfwS4WPHjmnEiBGKjY1Vfn6+kpOT9cEHH6hBgwbOzidJ+u6773To0CE9+OCDRZYdOnRIqamptumzZ89qxIgRatWqlW655RZlZGToxx9/VOvWrcslGyqPGjVq6IEHHlCNGjVcHQUAAAAeyKGh2NPT0/XSSy/pjTfeULt27bRy5Ur95S9/Ka9sNr169VJJ426sWbPGbnrOnDmaM2dOuWdC5dOoUSN99NFHro4BAAAAD1XqK1czZ85U48aN9fXXX+t///uffvzxxwoprICKcvHiRe3bt08XL150dRQAAAB4oFJfuZowYYICAgLUtGlTffDBB/rggw+KXc/R0fwAd/Hrr7+qY8eOSkpKUocOHVwdBwAAAB6m1MXV4MGDrzgUO1BauTk5SktLu+p28vLyrnrI08DAQIWGhl51lvT0dGVnZ5d5e4vFIunSdynUrFnzqvMAAACgYpW6uFqwYEE5xkBVcjE7S9u2b9PMhHcVEBBQ5nZyc3L02+5f1aJ1jHx8fMvcTs3gQD371Ngyby9dKqz+OXOOTmeWvbjyNpt1S7fr9ercBE14YrRTCj4AAABUHIcGtACcIS/nonKtJtVo3VkRteuWuZ3jh/Yp45ftCm5+fZnbyUo/o9M711/VFSdJys7O1unMbIXFdFFQaFiZ2vCSVdJZncm6oOzsbIorAAAAD0NxBZepFlJDITUjyrx95tlTTmnnTJm3LCooNKzMWUyGRbpw1olpAAAAUJEoroACHTp0KHHIfwAAAOBKyvQlwgAAAAAAexRXQIE9e/YoPj5ee/bscXUUAAAAeCCKK6DA+fPntXHjRp0/f97VUQAAAOCBKK4AAAAAwAkorgAAAADACSiuAAAAAMAJKK6AAg0bNtSHH36ohg0bujoKAAAAPBDfcwUUCAsL01//+ldXxwAAAICH4soVUODkyZNKSEjQyZMnXR0FAAAAHojiCihw+PBhjRo1SocPH3Z1FAAAAHggiisAAAAAcAKKKwAAAABwAoorAAAAAHACiiugQHBwsHr16qXg4GBXRwEAAIAHYij2KiQ3J0dpaWll3j4tLU15eblOTORemjVrpuXLl7s6BgAAADwUxVUVcTE7S9u2b9PMhHcVEBBQpjayz2dp12/7VDc+x8np3IPFYtH58+dVrVo1mc1mV8cBAACAh6G4qiLyci4q12pSjdadFVG7bpnaOH5on3J27lZ+Xr6T07mHX375RR07dlRSUpI6dOjg6jgAAADwMBRXVUy1kBoKqRlRpm0zz55ychoAAACg8mBACwAAAABwAoorAAAAAHACiisAAAAAcAKeuQIKxMbG6sSJE6pevbqrowAAAMADUVwBBXx8fBQeHu7qGAAAAPBQ3BYIFEhJSVH//v2VkpLi6igAAADwQBRXQIH09HR99dVXSk9Pd3UUAAAAeCCKKwAAAABwAoorAAAAAHACiisAAAAAcAKKK6BAnTp1NGvWLNWpU8fVUQAAAOCBGIodKBAZGalx48a5OgYAAAA8lFtfuZoyZYpMJpPdq2XLlpfd5rPPPlPLli3l7++v2NhYffvttxWUFp7u7Nmz+uyzz3T27FlXRwEAAIAHcuviSpJiYmKUmppqe61fv77EdX/88UcNGjRIw4cP19atWzVgwAANGDBAO3bsqMDE8FT79+/XwIEDtX//fldHAQAAgAdy++LK29tbUVFRtletWrVKXPe1115Tnz599OSTT6pVq1Z64YUX1KFDB82bN68CEwMAAACoitz+mau9e/cqOjpa/v7+io+P1/Tp01W/fv1i192wYUORZ2Z69+6tJUuWXHYfOTk5ysnJsU1nZGRIkvLy8pSXl3d1B3AVCvftbTbLS1aZDEuZ2zKbJD9fX5lNRpnbcUYb7taOl6zyNptlsViUn58vScrPz3e43y0Wy1X3U+F2hXlc+d5DxSjsY/q6aqC/qxb6u2qhvys3R/rVZBiGUY5ZrsrSpUuVlZWlFi1aKDU1VVOnTtXRo0e1Y8cOBQcHF1nf19dXH3zwgQYNGmSb9+abb2rq1KlKS0srcT9TpkzR1KlTi8z/+OOPFRgY6JyDgdtLSUnR+PHjNWvWLDVp0sTVcQAAAOAGsrOzdf/99ys9PV0hISGXXdetr1z17dvX9u+2bduqU6dOatCggRYuXKjhw4c7bT8TJ060u+KVkZGhevXqqVevXlc8geUpLy9PiYmJWvH9z4q6ppeCw8LL3Fbq73u0/OP56jt0jCLrNnBZG+7WTuaZkzqyaameHz9SjRo1Urt27dSjRw+1atXKoXbS0tL0wqwE1e3Ut8z9ZDIsangxRSu+/1kTx/xdkZGRZWoHnqPwM96zZ0/5+Pi4Og7KGf1dtdDfVQv9XbkV3tVWGm5dXP1Z9erV1bx5c+3bt6/Y5VFRUUWuUKWlpSkqKuqy7fr5+cnPz6/IfB8fH7f4gORbLLLKS4bJXOY2LIaUk5sri2EqczvOaMPd2rHKS/kWi8xms9q2bautW7eWqR2z2eyUfpJky+MO7z1UDHf5WYOKQX9XLfR31UJ/V06O9KnbD2jxR1lZWUpJSVHt2rWLXR4fH6+VK1fazUtMTFR8fHxFxAMAAABQhbl1cfXEE09o7dq1OnDggH788UfdcccdMpvNtmeqBg8erIkTJ9rWHz16tJYtW6ZZs2Zp9+7dmjJlijZv3qxRo0a56hDgQbZu3So/P78yX70CAABA1ebWtwUeOXJEgwYN0unTpxUeHq4uXbpo48aNCg+/9EzLoUOH5OX1/+vDG264QR9//LGee+45PfPMM2rWrJmWLFmiNm3auOoQ4EEMw1Bubq7ceIwXAAAAuDG3Lq4++eSTyy5fs2ZNkXn33HOP7rnnnnJKBAAAAADFc+vbAgEAAADAU1BcAQAAAIATuPVtgUBFatWqlXbs2KHGjRu7OgoAAAA8EMUVUCAgIEAxMTGujgEAAAAPxW2BQIGDBw/qoYce0sGDB10dBQAAAB6I4goocPr0ab377rs6ffq0q6MAAADAA1FcAQAAAIATUFwBAAAAgBNQXAEAAACAE1BcAQUiIyM1YcIERUZGujoKAAAAPBBDsQMF6tSpo+nTp7s6BgAAADwUV66AApmZmVqzZo0yMzNdHQUAAAAeiCtXqPJyc3KUlpambdu2qU+fPlq2bJnatm3rUBtpaWnKy8stp4QAAADwBBRXqNIuZmdp2/Ztmpnwrs5nZkiS3nz/Y9WKWOlQO9nns7Trt32qG59THjEBAADgASiuUKXl5VxUrtWkGq07K/jieenT/ygq7kZFN2rhUDvHD+1Tzs7dys/LL6ekAAAAcHcUV4CkaiE1ZPHzkyQFhYYppGaEQ9tnnj1VHrEAAADgQRjQAijg5e2t0FqR8vLmbw4AAABwHL9FAgWiG7XQ5I/XuToGAAAAPBRXrgAAAADACSiugALH9u/R1Ptv1LH9e1wdBQAAAB6I4gooYM3PV/qpNFnzGfEPAAAAjqO4AgAAAAAnoLgCAAAAACeguAIAAAAAJ6C4AgrUqtNAj73yH9Wq08DVUQAAAOCB+J4roIB/YJCaxnVydQwAAAB4KK5cAQXOnUrT1+/O0rlTaa6OAgAAAA9EcQUUyDp7Sqs+/Zeyzp5ydRQAAAB4IIorAAAAAHACiisAAAAAcAKKKwAAAABwAkYLBAoEhlRXpz53KzCkuqujoBylp6crOzv7qtoIDAxUaGiokxIBAIDKguIKKBAWWUf3jvunq2OgHKWnp+ufM+fodObVFVc1gwP17FNjKbAAAIAdiiugQG7ORZ1JPayw2vXk6+fv6jgoB9nZ2Tqdma2wmC4KCg0rUxtZ6Wd0eud6ZWdnU1wBAAA7FFdAgROHUjR75J0al/C56jaLcXUclKOg0DCF1Iwo8/ZnnJgFAABUHgxoAQAAAABOQHEFAAAAAE5AcQUAAAAATuDWxdX06dN17bXXKjg4WBERERowYID27Nlz2W0WLFggk8lk9/L3Z3AClILJJLOPj2QyuToJAAAAPJBbD2ixdu1ajRw5Utdee63y8/P1zDPPqFevXvr1119VrVq1ErcLCQmxK8JM/LKMUqjbtLVe+WaHq2MAAADAQ7l1cbVs2TK76QULFigiIkJJSUm68cYbS9zOZDIpKiqq1PvJyclRTk6ObTojI0OSlJeXp7y8PAdTO0/hvr3NZnnJKpNhKXNbZpPk5+srs8koczvOaMPd2nGnLIXbeZvNslgsLn3vVVYWi+WqP09esjqtjwq3p6+rBvq7aqG/qxb6u3JzpF9NhmEY5ZjFqfbt26dmzZpp+/btatOmTbHrLFiwQA899JDq1Kkjq9WqDh066KWXXlJMTMlDa0+ZMkVTp04tMv/jjz9WYGCg0/LDvR0+fFhz5szR2LFjVa9ePVfHAQAAgBvIzs7W/fffr/T0dIWEhFx2XY8prqxWq/r3769z585p/fr1Ja63YcMG7d27V23btlV6erpeffVVrVu3Tjt37lTdunWL3aa4K1f16tXTqVOnrngCy1NeXp4SExO14vufFXVNLwWHhZe5rdTf92j5x/PVd+gYRdZt4LI23K2dP7aRe+G8Zo+6W+PmLVK9Zq0rPIvJsKjhxRSt+P5nTRzzd0VGRpapHZQsLS1NL8xKUN1Ofcv8eco8c1JHNi3V8+NHXnUfFX7Ge/bsKR8fn6tqC+6P/q5a6O+qhf6u3DIyMlSrVq1SFVdufVvgH40cOVI7duy4bGElSfHx8YqPj7dN33DDDWrVqpXefvttvfDCC8Vu4+fnJz8/vyLzfXx83OIDkm+xyCovGSZzmduwGFJObq4shqnM7TijDXdr549tyFQwvovJ8XPtrGOSLvW32Wx2i/deZWM2m6/682SVl9P7yF1+1qBi0N9VC/1dtdDflZMjfeoRxdWoUaP09ddfa926dSVefSqJj4+P2rdvr3379pVTOgAAAABw86HYDcPQqFGjtHjxYq1atUqNGjVyuA2LxaLt27erdu3a5ZAQAAAAAC5x6ytXI0eO1Mcff6wvvvhCwcHBOn78uCQpNDRUAQEBkqTBgwerTp06mj59uiRp2rRpuv7669W0aVOdO3dOr7zyig4ePKiHHnrIZccBzxAWVVeDn5ursCjHro4CAAAAkpsXV2+99ZYkqVu3bnbz33//fQ0dOlSSdOjQIXl5/f8LcGfPntWIESN0/Phx1ahRQx07dtSPP/6o1q0dG6AAVU9gcKja3djX1TEAAADgody6uCrNQIZr1qyxm54zZ47mzJlTTolQmWWePaWklV+pY/d+Cq5Ry9VxAAAA4GHc+pkroCKln0rTl/+aofRTaa6OAgAAAA/k1leuAHi+9PR0ZWdnX3U7gYGBCg0NdUIi95GRkSHp0vdvmc1lGxq+Mp4XAAA8FcUVgHKTnp6uf86co9OZV19c1QwO1LNPja00hUR6erpenZug66+J0wuzEpRvsZSpncp2XgAA8GQUVwDKTXZ2tk5nZisspouCQsPK3E5W+hmd3rle2dnZlaaIyM7O1pmsC5Kkup36ylqGu7Qr43kBAMCTUVwBBfyrBSvm+pvkXy3Y1VEqnaDQMIXUjLiqNs44KYs7Cg4Ll2Eq222Blfm8AADgaSiugAK1outr+LT5ro4BAAAAD8VogUABS36ess6dkSU/z9VRAAAA4IEoroACqft/06SB8Urd/5urowAAAMADUVwBAAAAgBNQXAEAAACAE1BcAQAAAIATUFwBAAAAgBMwFDtQILpxS720OEm+/gGujgIAAAAPRHEFFPAym+VfLcjVMQAAAOChuC0QKHDy6AG9PXG4Th494OooAAAA8EAUV0CBnOzz2pO0XjnZ510dBQAAAB6I4goAAAAAnIDiCgAAAACcgOIKAAAAAJyA4gooUD28tu4cNUnVw2u7OgoAAAA8EEOxAwWCqoepS/8HXB1Dubk5SktLu6o28vLy5OPjc9VZrradtLQ05eXlXnUOd5Obc/V9VFnPDYqXnp6uzMxMSZf63mw2l6mdwMBAhYaGOjMa3FR6erqys7Ovuh3eM3CFqvz+pbgCCpzPOKddP61Vq+u6qlpIdZfl2Llzp2YmvKuAgLJ9mXFuTo5+2/2rWrSOkY+Pb5lzOKOd7PNZ2vXbPtWNzylzDndzMTtL27Zvu6o+ki6dm98PHNTtPW90Yjq4o/T0dP1z5hylZ+folm7X64VZCcq3WMrUVs3gQD371FiP+2UDjil8z5zOvPpfTnnPoKJV9fcvxRVQ4GzaUX088ymNS/jcpcVVrtWkGq07K6J23TJtf/zQPmX8sl3Bza8vcxvOauf4oX3K2blb+Xn5Zc7hbvJyLl51H0kF52bv705MBneVnZ2t05nZCm/TWZJUt1NfWctwV35W+hmd3rle2dnZHvWLBhxX+J4Ji+mioNCwMrfDewauUNXfvxRXgBuqFlJDITUjyrRt5tlTV92Gs9opbKMyctb5RdVRLaSGpLMKDguXYSrbbYFnnBsJbi4oNOyqfs5IvGfgOlX1/cuAFgAAAADgBBRXAAAAAOAEFFdAAV//ADVo1U6+/mUfpAAAAABVF89cAQUi6jXW6Nc+dXUMAAAAeCiuXAEAAACAE1BcAQWO7N2pcb1a6Mjena6OAgAAAA9EcQUAAAAATkBxBQAAAABOQHEFAAAAAE5AcQUAAAAATsBQ7ECByAZNNfH9FaoeHuXqKAAAAPBAFFdAAR9fP4XXaeDqGAAAAPBQ3BYIFDidelgfzXhCp1MPuzoKAAAAPJBHFFcJCQlq2LCh/P391alTJ/3000+XXf+zzz5Ty5Yt5e/vr9jYWH377bcVlBSe7EJWhras+koXsjJcHQUAAAAeyO2Lq08//VTjxo3T5MmTtWXLFsXFxal37946ceJEsev/+OOPGjRokIYPH66tW7dqwIABGjBggHbs2FHByQEAAABUJW5fXM2ePVsjRozQsGHD1Lp1a82fP1+BgYF67733il3/tddeU58+ffTkk0+qVatWeuGFF9ShQwfNmzevgpMDAAAAqErcekCL3NxcJSUlaeLEibZ5Xl5e6tGjhzZs2FDsNhs2bNC4cePs5vXu3VtLliwpcT85OTnKycmxTaenp0uSzpw5o7y8vKs4gquTl5en7OxsWfPzdS7tsHKzM8vcVsbpVHl7eSnj5FH5mKwua8Pd2vljG3kXzkuSzqUdkb+/f4Vn8ZKhyIBseXup0vSTs7JkZ55TzsVs7d27V+fOnStzOydPnlRuzoWr+jw59/xK2dnZOn16v6wyOdyGs86LJBmGIZPJ8Qzl0Y47ZXFGO4Xvu/QTR5UdLpf3t7ucF2e2445ZLBaLsrOztWfPHpnNZofacMbPKon3TEVmuZr+dqfz4ox2nPn+tebn69y5c/L19S1zO86QmXnpOAzDuPLKhhs7evSoIcn48ccf7eY/+eSTxnXXXVfsNj4+PsbHH39sNy8hIcGIiIgocT+TJ082JPHixYsXL168ePHixYtXsa/Dhw9fsX5x6ytXFWXixIl2V7usVqvOnDmjmjVrOuUvAGWVkZGhevXq6fDhwwoJCXFZDlQM+rvqoc+rFvq7aqG/qxb6u3IzDEOZmZmKjo6+4rpuXVzVqlVLZrNZaWlpdvPT0tIUFVX8F71GRUU5tL4k+fn5yc/Pz25e9erVyxa6HISEhPBBrULo76qHPq9a6O+qhf6uWujvyis0NLRU67n1gBa+vr7q2LGjVq5caZtntVq1cuVKxcfHF7tNfHy83fqSlJiYWOL6AAAAAOAMbn3lSpLGjRunIUOG6JprrtF1112nuXPn6vz58xo2bJgkafDgwapTp46mT58uSRo9erS6du2qWbNm6dZbb9Unn3yizZs361//+pcrDwMAAABAJef2xdW9996rkydPatKkSTp+/LjatWunZcuWKTIyUpJ06NAheXn9/wtwN9xwgz7++GM999xzeuaZZ9SsWTMtWbJEbdq0cdUhlJmfn58mT55c5JZFVE70d9VDn1ct9HfVQn9XLfQ3CpkMozRjCgIAAAAALsetn7kCAAAAAE9BcQUAAAAATkBxBQAAAABOQHEFAAAAAE5AceXGEhIS1LBhQ/n7+6tTp0766aefXB0JZbBu3Tr169dP0dHRMplMWrJkid1ywzA0adIk1a5dWwEBAerRo4f27t1rt86ZM2f0wAMPKCQkRNWrV9fw4cOVlZVVgUeB0pg+fbquvfZaBQcHKyIiQgMGDNCePXvs1rl48aJGjhypmjVrKigoSHfddVeRLz4/dOiQbr31VgUGBioiIkJPPvmk8vPzK/JQUEpvvfWW2rZta/vi0Pj4eC1dutS2nP6u3GbMmCGTyaQxY8bY5tHnlceUKVNkMpnsXi1btrQtp69RHIorN/Xpp59q3Lhxmjx5srZs2aK4uDj17t1bJ06ccHU0OOj8+fOKi4tTQkJCsctnzpyp119/XfPnz9emTZtUrVo19e7dWxcvXrSt88ADD2jnzp1KTEzU119/rXXr1unhhx+uqENAKa1du1YjR47Uxo0blZiYqLy8PPXq1Uvnz5+3rTN27Fh99dVX+uyzz7R27VodO3ZMd955p225xWLRrbfeqtzcXP3444/64IMPtGDBAk2aNMkVh4QrqFu3rmbMmKGkpCRt3rxZN998s26//Xbt3LlTEv1dmf388896++231bZtW7v59HnlEhMTo9TUVNtr/fr1tmX0NYplwC1dd911xsiRI23TFovFiI6ONqZPn+7CVLhakozFixfbpq1WqxEVFWW88sortnnnzp0z/Pz8jP/973+GYRjGr7/+akgyfv75Z9s6S5cuNUwmk3H06NEKyw7HnThxwpBkrF271jCMS33r4+NjfPbZZ7Z1du3aZUgyNmzYYBiGYXz77beGl5eXcfz4cds6b731lhESEmLk5ORU7AGgTGrUqGH8+9//pr8rsczMTKNZs2ZGYmKi0bVrV2P06NGGYfAZr2wmT55sxMXFFbuMvkZJuHLlhnJzc5WUlKQePXrY5nl5ealHjx7asGGDC5PB2fbv36/jx4/b9XVoaKg6depk6+sNGzaoevXquuaaa2zr9OjRQ15eXtq0aVOFZ0bppaenS5LCwsIkSUlJScrLy7Pr75YtW6p+/fp2/R0bG2v7onRJ6t27tzIyMmxXQ+CeLBaLPvnkE50/f17x8fH0dyU2cuRI3XrrrXZ9K/EZr4z27t2r6OhoNW7cWA888IAOHTokib5GybxdHQBFnTp1ShaLxe7DKEmRkZHavXu3i1KhPBw/flySiu3rwmXHjx9XRESE3XJvb2+FhYXZ1oH7sVqtGjNmjDp37qw2bdpIutSXvr6+ql69ut26f+7v4t4PhcvgfrZv3674+HhdvHhRQUFBWrx4sVq3bq3k5GT6uxL65JNPtGXLFv38889FlvEZr1w6deqkBQsWqEWLFkpNTdXUqVP1l7/8RTt27KCvUSKKKwAoByNHjtSOHTvs7s9H5dSiRQslJycrPT1dixYt0pAhQ7R27VpXx0I5OHz4sEaPHq3ExET5+/u7Og7KWd++fW3/btu2rTp16qQGDRpo4cKFCggIcGEyuDNuC3RDtWrVktlsLjLiTFpamqKiolyUCuWhsD8v19dRUVFFBjLJz8/XmTNneD+4qVGjRunrr7/W6tWrVbduXdv8qKgo5ebm6ty5c3br/7m/i3s/FC6D+/H19VXTpk3VsWNHTZ8+XXFxcXrttdfo70ooKSlJJ06cUIcOHeTt7S1vb2+tXbtWr7/+ury9vRUZGUmfV2LVq1dX8+bNtW/fPj7fKBHFlRvy9fVVx44dtXLlSts8q9WqlStXKj4+3oXJ4GyNGjVSVFSUXV9nZGRo06ZNtr6Oj4/XuXPnlJSUZFtn1apVslqt6tSpU4VnRskMw9CoUaO0ePFirVq1So0aNbJb3rFjR/n4+Nj19549e3To0CG7/t6+fbtdQZ2YmKiQkBC1bt26Yg4EV8VqtSonJ4f+roS6d++u7du3Kzk52fa65ppr9MADD9j+TZ9XXllZWUpJSVHt2rX5fKNkrh5RA8X75JNPDD8/P2PBggXGr7/+ajz88MNG9erV7UacgWfIzMw0tm7damzdutWQZMyePdvYunWrcfDgQcMwDGPGjBlG9erVjS+++MLYtm2bcfvttxuNGjUyLly4YGujT58+Rvv27Y1NmzYZ69evN5o1a2YMGjTIVYeEEjz66KNGaGiosWbNGiM1NdX2ys7Otq3z97//3ahfv76xatUqY/PmzUZ8fLwRHx9vW56fn2+0adPG6NWrl5GcnGwsW7bMCA8PNyZOnOiKQ8IVTJgwwVi7dq2xf/9+Y9u2bcaECRMMk8lkrFixwjAM+rsq+ONogYZBn1cm48ePN9asWWPs37/f+OGHH4wePXoYtWrVMk6cOGEYBn2N4lFcubE33njDqF+/vuHr62tcd911xsaNG10dCWWwevVqQ1KR15AhQwzDuDQc+/PPP29ERkYafn5+Rvfu3Y09e/bYtXH69Glj0KBBRlBQkBESEmIMGzbMyMzMdMHR4HKK62dJxvvvv29b58KFC8Zjjz1m1KhRwwgMDDTuuOMOIzU11a6dAwcOGH379jUCAgKMWrVqGePHjzfy8vIq+GhQGg8++KDRoEEDw9fX1wgPDze6d+9uK6wMg/6uCv5cXNHnlce9995r1K5d2/D19TXq1Klj3Hvvvca+fftsy+lrFMdkGIbhmmtmAAAAAFB58MwVAAAAADgBxRUAAAAAOAHFFQAAAAA4AcUVAAAAADgBxRUAAAAAOAHFFQAAAAA4AcUVAAAAADgBxRUAAAAAOAHFFQCgUjhw4IBMJpOSk5NdHcVm9+7duv766+Xv76927dpddXtTpkxxSjsAgPJBcQUAcIqhQ4fKZDJpxowZdvOXLFkik8nkolSuNXnyZFWrVk179uzRypUri12nW7duGjNmTKnae+KJJ0psBwDgehRXAACn8ff318svv6yzZ8+6OorT5ObmlnnblJQUdenSRQ0aNFDNmjXL3I5hGMrPz1dQUNBVtQMAKF8UVwAAp+nRo4eioqI0ffr0Etcp7ta2uXPnqmHDhrbpoUOHasCAAXrppZcUGRmp6tWra9q0acrPz9eTTz6psLAw1a1bV++//36R9nfv3q0bbrhB/v7+atOmjdauXWu3fMeOHerbt6+CgoIUGRmpv/3tbzp16pRtebdu3TRq1CiNGTNGtWrVUu/evYs9DqvVqmnTpqlu3bry8/NTu3bttGzZMttyk8mkpKQkTZs2TSaTSVOmTCnSxtChQ7V27Vq99tprMplMMplMOnDggNasWSOTyaSlS5eqY8eO8vPz0/r164ucu8LzNHXqVIWHhyskJER///vf7QrCRYsWKTY2VgEBAapZs6Z69Oih8+fPF3tMAICrQ3EFAHAas9msl156SW+88YaOHDlyVW2tWrVKx44d07p16zR79mxNnjxZt912m2rUqKFNmzbp73//ux555JEi+3nyySc1fvx4bd26VfHx8erXr59Onz4tSTp37pxuvvlmtW/fXps3b9ayZcuUlpamgQMH2rXxwQcfyNfXVz/88IPmz59fbL7XXntNs2bN0quvvqpt27apd+/e6t+/v/bu3StJSk1NVUxMjMaPH6/U1FQ98cQTxbYRHx+vESNGKDU1VampqapXr55t+YQJEzRjxgzt2rVLbdu2LTbHypUrtWvXLq1Zs0b/+9//9Pnnn2vq1Km2DIMGDdKDDz5oW+fOO++UYRil7AUAgCMorgAATnXHHXeoXbt2mjx58lW1ExYWptdff10tWrTQgw8+qBYtWig7O1vPPPOMmjVrpokTJ8rX11fr16+3227UqFG666671KpVK7311lsKDQ3Vu+++K0maN2+e2rdvr5deekktW7ZU+/bt9d5772n16tX67bffbG00a9ZMM2fOVIsWLdSiRYti87366qt6+umndd9996lFixZ6+eWX1a5dO82dO1eSFBUVJW9vbwUFBSkqKkpBQUFF2ggNDZWvr68CAwMVFRWlqKgomc1m2/Jp06apZ8+eatKkicLCworN4evrq/fee08xMTG69dZbNW3aNL3++uuyWq1KTU1Vfn6+7rzzTjVs2FCxsbF67LHHis0CALh6FFcAAKd7+eWX9cEHH2jXrl1lbiMmJkZeXv//f1ORkZGKjY21TZvNZtWsWVMnTpyw2y4+Pt72b29vb11zzTW2HL/88otWr16toKAg26tly5aSLj0fVahjx46XzZaRkaFjx46pc+fOdvM7d+58Vcf8Z9dcc80V14mLi1NgYKBtOj4+XllZWTp8+LDi4uLUvXt3xcbG6p577tE777xTqZ6HAwB3Q3EFAHC6G2+8Ub1799bEiROLLPPy8ipyW1peXl6R9Xx8fOymTSZTsfOsVmupc2VlZalfv35KTk62e+3du1c33nijbb1q1aqVus3ydLU5zGazEhMTtXTpUrVu3VpvvPGGWrRoof379zspIQDgjyiuAADlYsaMGfrqq6+0YcMGu/nh4eE6fvy4XYHlzO+m2rhxo+3f+fn5SkpKUqtWrSRJHTp00M6dO9WwYUM1bdrU7uVIIRMSEqLo6Gj98MMPdvN/+OEHtW7d2qG8vr6+slgsDm3zR7/88osuXLhgm964caOCgoJsz26ZTCZ17txZU6dO1datW+Xr66vFixeXeX8AgJJRXAEAykVsbKweeOABvf7663bzu3XrppMnT2rmzJlKSUlRQkKCli5d6rT9JiQkaPHixdq9e7dGjhyps2fP6sEHH5QkjRw5UmfOnNGgQYP0888/KyUlRcuXL9ewYcMcLnCefPJJvfzyy/r000+1Z88eTZgwQcnJyRo9erRD7TRs2FCbNm3SgQMHdOrUKYeuxEmXhoofPny4fv31V3377beaPHmyRo0aJS8vL23atEkvvfSSNm/erEOHDunzzz/XyZMnbcUmAMC5KK4AAOVm2rRpRYqFVq1a6c0331RCQoLi4uL0008/FTuSXlnNmDFDM2bMUFxcnNavX68vv/xStWrVkiTb1SaLxaJevXopNjZWY8aMUfXq1e2e7yqNxx9/XOPGjdP48eMVGxurZcuW6csvv1SzZs0caueJJ56Q2WxW69atFR4erkOHDjm0fffu3dWsWTPdeOONuvfee9W/f3/bsO8hISFat26dbrnlFjVv3lzPPfecZs2apb59+zq0DwBA6ZgMxmMFAMAjDR06VOfOndOSJUtcHQUAIK5cAQAAAIBTUFwBAAAAgBNwWyAAAAAAOAFXrgAAAADACSiuAAAAAMAJKK4AAAAAwAkorgAAAADACSiuAAAAAMAJKK4AAAAAwAkorgAAAADACSiuAAAAAMAJ/h/Zn/YOz01EEwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Matching start and end points with tessellation...\n",
      "Done.\n",
      "\n",
      "Extracting trips that start and end within tessellation area...\n",
      "Number of trips that start and end wihin tessellation area: 5101\n",
      "Number of trips outside and therefore dropped: 299\n",
      "Done.\n",
      "\n",
      "Building mapping of trip chains...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████| 5101/5101 [01:21<00:00, 62.56it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Evaluating trip chaining...\n",
      "Number of edges (matched) between trips: 526\n",
      "Number of wrong matches: 153\n",
      "Done.\n",
      "\n",
      "Building trip chains...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████████████████████████████| 5101/5101 [00:00<00:00, 39506.00it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Merging trips...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████| 5101/5101 [00:59<00:00, 85.79it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "Number of trips that were matched at least once: 5046/5101\n",
      "Concatenating MERGED and UNMERGED trips...\n",
      "Done.\n",
      "\n",
      "Extracting data for concatenated trips...\n",
      "Done.\n",
      "\n",
      "Building clustering after concatenation...\n",
      "Number of unique clusters: 4630\n",
      "Done.\n",
      "\n",
      "Building HL from start points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 201 disconnected components.\n",
      " There are 126 islands with ids: 9, 10, 11, 12, 15, 16, 17, 27, 28, 29, 30, 31, 44, 67, 68, 89, 131, 146, 153, 157, 183, 184, 210, 216, 219, 220, 221, 229, 232, 233, 234, 235, 236, 237, 246, 295, 400, 563, 613, 713, 726, 727, 731, 732, 796, 810, 817, 840, 841, 845, 897, 904, 905, 914, 915, 927, 946, 947, 950, 951, 954, 956, 959, 960, 961, 964, 968, 1000, 1001, 1010, 1011, 1016, 1017, 1024, 1026, 1031, 1034, 1037, 1038, 1048, 1049, 1050, 1055, 1056, 1057, 1058, 1059, 1100, 1136, 1137, 1138, 1139, 1140, 1191, 1192, 1195, 1196, 1208, 1209, 1211, 1213, 1214, 1215, 1220, 1221, 1339, 1341, 1342, 1344, 1345, 1346, 1347, 1348, 1358, 1360, 1361, 1366, 1367, 1369, 1370, 1371, 1372, 1373, 1374, 1375, 1408.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Building HL from end points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 277 disconnected components.\n",
      " There are 182 islands with ids: 0, 1, 6, 13, 14, 15, 25, 27, 28, 31, 32, 35, 47, 48, 49, 50, 51, 52, 53, 56, 67, 68, 70, 82, 95, 107, 123, 125, 126, 148, 149, 150, 210, 239, 266, 342, 345, 354, 355, 368, 370, 372, 376, 379, 383, 393, 396, 397, 401, 407, 514, 542, 543, 602, 603, 604, 659, 660, 743, 744, 912, 1018, 1020, 1149, 1169, 1173, 1174, 1177, 1186, 1191, 1220, 1231, 1259, 1260, 1269, 1275, 1338, 1420, 1421, 1422, 1426, 1427, 1430, 1445, 1446, 1471, 1475, 1476, 1477, 1496, 1497, 1498, 1500, 1501, 1502, 1504, 1505, 1506, 1511, 1512, 1513, 1514, 1515, 1516, 1517, 1518, 1519, 1520, 1522, 1524, 1563, 1572, 1574, 1576, 1577, 1578, 1580, 1600, 1601, 1607, 1609, 1610, 1612, 1616, 1617, 1625, 1626, 1629, 1630, 1637, 1638, 1639, 1640, 1641, 1642, 1643, 1644, 1645, 1646, 1647, 1648, 1649, 1654, 1655, 1659, 1706, 1710, 1748, 1749, 1795, 1796, 1798, 1799, 1865, 1867, 1871, 1872, 1874, 1875, 1889, 1890, 1893, 1894, 1895, 1896, 1897, 1898, 1899, 1902, 1903, 2042, 2043, 2044, 2045, 2048, 2049, 2064, 2065, 2066, 2069, 2070, 2107.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Combining HL from start points and end points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 339 disconnected components.\n",
      " There are 220 islands with ids: 0, 1, 6, 14, 15, 25, 29, 30, 31, 32, 35, 47, 48, 49, 50, 51, 52, 53, 56, 67, 68, 191, 192, 205, 206, 209, 210, 211, 257, 258, 259, 260, 314, 347, 348, 361, 362, 369, 370, 536, 538, 544, 690, 691, 692, 1014, 1015, 1016, 1154, 1155, 1157, 1161, 1163, 1166, 1209, 1213, 1247, 1251, 1273, 1274, 1275, 1413, 1425, 1466, 1467, 1469, 1470, 1474, 1475, 1476, 1497, 1498, 1501, 1502, 1503, 1504, 1505, 1506, 1511, 1512, 1513, 1514, 1516, 1517, 1518, 1523, 1524, 1572, 1596, 1597, 1598, 1599, 1600, 1601, 1609, 1616, 1626, 1629, 1630, 1637, 1638, 1639, 1640, 1641, 1642, 1644, 1646, 1649, 1669, 1670, 1712, 1768, 1769, 1796, 1797, 1866, 1867, 1872, 1873, 1874, 1875, 1889, 1890, 1894, 1895, 1896, 1897, 1902, 2042, 2043, 2044, 2045, 2048, 2049, 2064, 2066, 2069, 2107, 2119, 2120, 2123, 2125, 2135, 2136, 2139, 2212, 2224, 2225, 2251, 2307, 2309, 2310, 2327, 2334, 2335, 2336, 2511, 2512, 2868, 2869, 2873, 2987, 2988, 2989, 2990, 3041, 3042, 3044, 3054, 3055, 3067, 3072, 3108, 3109, 3119, 3125, 3126, 3133, 3134, 3139, 3142, 3145, 3146, 3160, 3163, 3164, 3166, 3167, 3241, 3244, 3245, 3246, 3247, 3248, 3300, 3303, 3304, 3316, 3320, 3322, 3328, 3329, 3449, 3450, 3451, 3452, 3453, 3454, 3456, 3467, 3468, 3469, 3475, 3477, 3479, 3480, 3481, 3482, 3483, 3522.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of users for which at least on Home Location has been identified:  72\n",
      "Number of unique HL tiles: 339\n",
      "Done.\n",
      "\n",
      "Matching trips to HL...\n",
      "Number of unmatched trajectories (concatenated) that do neither start nor end in a HL tile: 172/4630\n",
      "Number of trajectories (concatenated) that start AND end in a HL tile: 2930/4630\n",
      "Number of trips that match different HL tiles with their SP and EP: 1700\n",
      "Done.\n",
      "\n",
      "Matching double assigned trips to unique HL...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 201 tasks      | elapsed:  2.3min\n",
      "[Parallel(n_jobs=-4)]: Done 551 tasks      | elapsed:  5.5min\n",
      "[Parallel(n_jobs=-4)]: Done 1001 tasks      | elapsed:  9.9min\n",
      "[Parallel(n_jobs=-4)]: Done 1551 tasks      | elapsed: 15.2min\n",
      "[Parallel(n_jobs=-4)]: Done 2201 tasks      | elapsed: 22.2min\n",
      "[Parallel(n_jobs=-4)]: Done 2951 tasks      | elapsed: 29.3min\n",
      "[Parallel(n_jobs=-4)]: Done 3400 out of 3400 | elapsed: 34.2min finished\n",
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "Done.\n",
      "\n",
      "Getting trajectories that happened during the same time...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:  1.8min\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:  4.8min\n",
      "[Parallel(n_jobs=-4)]: Done 1000 tasks      | elapsed:  8.8min\n",
      "[Parallel(n_jobs=-4)]: Done 1550 tasks      | elapsed: 13.6min\n",
      "[Parallel(n_jobs=-4)]: Done 2200 tasks      | elapsed: 19.2min\n",
      "[Parallel(n_jobs=-4)]: Done 2950 tasks      | elapsed: 25.7min\n",
      "[Parallel(n_jobs=-4)]: Done 3800 tasks      | elapsed: 33.1min\n",
      "[Parallel(n_jobs=-4)]: Done 4630 out of 4630 | elapsed: 40.3min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Building clustering after HL matching step...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                           | 0/189 [00:00<?, ?it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 34 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.7s finished\n",
      "  4%|███▌                                                                               | 8/189 [00:01<00:41,  4.41it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  34\n",
      "Running randomized subset search for 1000 iterations with 478 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 204 tasks      | elapsed:    2.4s\n",
      "[Parallel(n_jobs=-4)]: Done 751 out of 1000 | elapsed:    7.7s remaining:    2.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:   12.0s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  174\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  5%|███▉                                                                               | 9/189 [00:14<06:13,  2.08s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 110 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.9s finished\n",
      "  5%|████▎                                                                             | 10/189 [00:16<06:12,  2.08s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  88\n",
      "Running randomized subset search for 1000 iterations with 550 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    1.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    6.6s finished\n",
      "  7%|█████▋                                                                            | 13/189 [00:23<06:22,  2.17s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  83\n",
      "Running randomized subset search for 1000 iterations with 1515 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 205 tasks      | elapsed:    4.8s\n",
      "[Parallel(n_jobs=-4)]: Done 751 out of 1000 | elapsed:   15.9s remaining:    5.3s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:   23.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  114\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  8%|██████▌                                                                           | 15/189 [00:46<14:29,  5.00s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 43 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.7s finished\n",
      "  9%|███████▍                                                                          | 17/189 [00:48<10:52,  3.79s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  43\n",
      "Running randomized subset search for 1000 iterations with 59 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.6s finished\n",
      " 10%|███████▊                                                                          | 18/189 [00:50<09:48,  3.44s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  59\n",
      "Running randomized subset search for 1000 iterations with 65 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.8s finished\n",
      " 11%|████████▋                                                                         | 20/189 [00:52<07:23,  2.63s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  65\n",
      "Running randomized subset search for 1000 iterations with 85 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.9s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    2.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  85\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 13%|██████████▊                                                                       | 25/189 [00:55<04:07,  1.51s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 38 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.3s finished\n",
      " 14%|███████████▎                                                                      | 26/189 [00:56<04:03,  1.49s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  34\n",
      "Running randomized subset search for 1000 iterations with 64 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.4s finished\n",
      " 16%|█████████████▍                                                                    | 31/189 [00:58<02:26,  1.08it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  64\n",
      "Running randomized subset search for 1000 iterations with 44 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.4s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.3s finished\n",
      " 19%|███████████████▌                                                                  | 36/189 [00:59<01:41,  1.51it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  44\n",
      "Running randomized subset search for 1000 iterations with 33 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.2s finished\n",
      " 30%|████████████████████████▎                                                         | 56/189 [01:01<00:32,  4.05it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  33\n",
      "Running randomized subset search for 1000 iterations with 32 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.3s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.0s finished\n",
      " 31%|█████████████████████████▏                                                        | 58/189 [01:02<00:35,  3.67it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  32\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 33%|██████████████████████████▉                                                       | 62/189 [01:02<00:28,  4.49it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 212 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    4.6s\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:   11.2s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:   18.4s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  212\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 35%|█████████████████████████████                                                     | 67/189 [01:21<02:23,  1.18s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 47 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.7s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    2.0s finished\n",
      " 46%|█████████████████████████████████████▋                                            | 87/189 [01:23<00:54,  1.89it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  47\n",
      "Running randomized subset search for 1000 iterations with 53 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.5s finished\n",
      " 50%|█████████████████████████████████████████▏                                        | 95/189 [01:25<00:42,  2.23it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  53\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 74%|███████████████████████████████████████████████████████████▌                     | 139/189 [01:25<00:06,  7.25it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 115 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.7s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    2.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  106\n",
      "Running randomized subset search for 1000 iterations with 127 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    1.3s\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:    3.1s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    5.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  127\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 91%|█████████████████████████████████████████████████████████████████████████▋       | 172/189 [01:33<00:02,  5.73it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 212 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 201 tasks      | elapsed:    1.2s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    5.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  157\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 97%|██████████████████████████████████████████████████████████████████████████████▍  | 183/189 [01:39<00:01,  3.91it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 52 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.7s finished\n",
      "100%|█████████████████████████████████████████████████████████████████████████████████| 189/189 [01:41<00:00,  1.86it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  52\n",
      "Done.\n",
      "\n",
      "Clustering results after concatenation step:\n",
      "Number of unique clusters: 4630\n",
      "Homogeneity: 0.991\n",
      "Completeness: 0.436\n",
      "V-measure: 0.606\n",
      "Rand index: 0.965\n",
      "ARI: 0.002\n",
      "MI: 3.666\n",
      "NMI: 0.606\n",
      "AMI: 0.039\n",
      "Cluster accuracy: 0.025\n",
      "\n",
      "Clustering results after HL matching step:\n",
      "Number of unique clusters: 2626\n",
      "Homogeneity: 0.899\n",
      "Completeness: 0.507\n",
      "V-measure: 0.649\n",
      "Rand index: 0.968\n",
      "ARI: 0.261\n",
      "MI: 3.329\n",
      "NMI: 0.649\n",
      "AMI: 0.380\n",
      "Cluster accuracy: 0.295\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "attack.plot_hour_of_day_distribution(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_number_of_trips_per_user(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_trip_distances(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_trip_durations(raw_full_trip_gdf)\n",
    "\n",
    "# Merge Start Points (SP) and End Points (EP) with Tessellation\n",
    "print(\"\\nMatching start and end points with tessellation...\")\n",
    "gdf_sp, gdf_ep = attack.match_boundary_points_with_tessellation(raw_trip_sp_gdf, raw_trip_ep_gdf, tesselation_gdf)\n",
    "print(\"Done.\")\n",
    "\n",
    "# Extract Full Trips that Start and End within Tessellation Area\n",
    "print(\"\\nExtracting trips that start and end within tessellation area...\")\n",
    "full_trip_gdf, trip_sp_gdf, trip_ep_gdf, gdf_sp, gdf_ep = attack.extract_trips_that_start_end_in_tessellation(raw_full_trip_gdf, raw_trip_sp_gdf, raw_trip_ep_gdf, gdf_sp, gdf_ep)\n",
    "print(\"Done.\")\n",
    "\n",
    "# Build mapping of trip chains\n",
    "print('\\nBuilding mapping of trip chains...')\n",
    "mapping_cont_trips = attack.build_trip_chain_mapping(gdf_sp, gdf_ep)\n",
    "print('Done.')\n",
    "\n",
    "# Evaluate Trip Chaining\n",
    "print('\\nEvaluating trip chaining...')\n",
    "attack.evaluate_trip_chaining(mapping_cont_trips, full_trip_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Concatenate Trips\n",
    "full_trips_concat_gdf, trip_concat_dict = attack.merge_trips_from_matching(gdf_sp, mapping_cont_trips, full_trip_gdf)\n",
    "\n",
    "# Extract data for concatenated trips\n",
    "print('\\nExtracting data for concatenated trips...')\n",
    "gdf_sp_concat, trip_sp_gdf_concat, gdf_ep_concat, trip_ep_gdf_concat = attack.extract_concatenated_trips(full_trips_concat_gdf, gdf_sp, trip_sp_gdf, gdf_ep, trip_ep_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Clustering after Concatenation step\n",
    "print('\\nBuilding clustering after concatenation...')\n",
    "clustering_concat = attack.build_clustering_after_concatenation(full_trips_concat_gdf, trip_concat_dict, full_trip_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Home Locations (HL)\n",
    "# Build HL from Start Points\n",
    "print('\\nBuilding HL from start points...')\n",
    "gdf_hl_combined_sp = attack.build_hl_from_start_points(gdf_sp, gdf_ep, HL_SP_OUTFLOW_THRESHOLD=HL_SP_OUTFLOW_THRESHOLD)\n",
    "print('Done.')\n",
    "# Build HL from End Points\n",
    "print('\\nBuilding HL from end points...')\n",
    "gdf_hl_combined_ep = attack.build_hl_from_end_points(gdf_sp, gdf_ep, HL_EP_OUTFLOW_THRESHOLD=HL_EP_OUTFLOW_THRESHOLD)\n",
    "print('Done.')\n",
    "# Combine HL from Start Points and End Points\n",
    "print('\\nCombining HL from start points and end points...')\n",
    "gp_combined, HL_table = attack.concatenate_hl(gdf_hl_combined_sp, gdf_hl_combined_ep)\n",
    "print('Done.')\n",
    "\n",
    "print('\\nMatching trips to HL...')\n",
    "HL_table_se_concat, unmatched_trips, double_assigned_trips, nr_unmatched = attack.match_trips_to_HL(gp_combined, HL_table, trip_sp_gdf_concat, trip_ep_gdf_concat, full_trips_concat_gdf)\n",
    "print('Done.')\n",
    "\n",
    "print('\\nMatching double assigned trips to unique HL...')\n",
    "HL_table_trips_concat = attack.assign_double_matched_trips_to_unique_hl(HL_table_se_concat, full_trips_concat_gdf, unmatched_trips, double_assigned_trips, nr_unmatched)\n",
    "print('Done.')\n",
    "\n",
    "# Get trajectories that happened during the same time\n",
    "print('\\nGetting trajectories that happened during the same time...')\n",
    "full_trips_concat_gdf_overlap_dict = attack.getTripOverlaps(full_trips_concat_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Clustering after HL matching step\n",
    "print('\\nBuilding clustering after HL matching step...')\n",
    "clustering_after_HL, HL_table_dict = attack.build_clustering_after_HL_assignment(HL_table_trips_concat, full_trip_gdf, trip_concat_dict, full_trips_concat_gdf_overlap_dict)\n",
    "print('Done.')\n",
    "\n",
    "# Evaluate clustering Results\n",
    "print(\"\\nClustering results after concatenation step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_concat))}\")\n",
    "attack.evaluate(clustering_concat, full_trip_gdf)\n",
    "clustering_HL = list(dict(sorted(clustering_after_HL.items())).values())\n",
    "print(\"\\nClustering results after HL matching step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_HL))}\")\n",
    "attack.evaluate(clustering_HL, full_trip_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/geopandas/io/file.py:362: FutureWarning: pandas.Int64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n",
      "  pd.Int64Index,\n"
     ]
    }
   ],
   "source": [
    "full_trip_gdf = full_trip_gdf.copy()\n",
    "full_trip_gdf['clustering_HL'] = clustering_HL\n",
    "full_trip_gdf.to_file('../data/geolife/geolife_clustering_HL.geojson', driver=\"GeoJSON\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Obfuscated Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading private geolife geojson file...\n",
      "Done.\n",
      "Number of trajectories in loaded dataset: 4719\n",
      "Number of users in loaded dataset: 73\n"
     ]
    }
   ],
   "source": [
    "# Load preprocessed data \n",
    "raw_full_trip_gdf, raw_trip_sp_gdf, raw_trip_ep_gdf, tesselation_gdf = dl.load_geolife(data_type='private', \n",
    "                                                                                       min_n_trips_per_user=1,\n",
    "                                                                                       rand_n_week_period=None,\n",
    "                                                                                       tessellation_diameter=200,\n",
    "                                                                                       min_trip_length=200,\n",
    "                                                                                       min_nr_points=50,\n",
    "                                                                                       upper_quantile_length=0.95,\n",
    "                                                                                       only_2008=True)\n",
    "assert len(raw_full_trip_gdf) == len(raw_trip_sp_gdf) == len(raw_trip_ep_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Matching start and end points with tessellation...\n",
      "Done.\n",
      "\n",
      "Extracting trips that start and end within tessellation area...\n",
      "Number of trips that start and end wihin tessellation area: 4449\n",
      "Number of trips outside and therefore dropped: 270\n",
      "Done.\n",
      "\n",
      "Building mapping of trip chains...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████| 4449/4449 [01:10<00:00, 62.89it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Evaluating trip chaining...\n",
      "Number of edges (matched) between trips: 145\n",
      "Number of wrong matches: 75\n",
      "Done.\n",
      "\n",
      "Building trip chains...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████| 4449/4449 [00:00<00:00, 153359.35it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Merging trips...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████| 4449/4449 [00:53<00:00, 82.44it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "Number of trips that were matched at least once: 4443/4449\n",
      "Concatenating MERGED and UNMERGED trips...\n",
      "Done.\n",
      "\n",
      "Extracting data for concatenated trips...\n",
      "Done.\n",
      "\n",
      "Building clustering after concatenation...\n",
      "Number of unique clusters: 4310\n",
      "Done.\n",
      "\n",
      "Building HL from start points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 214 disconnected components.\n",
      " There are 136 islands with ids: 8, 9, 12, 13, 14, 24, 25, 26, 27, 28, 29, 38, 42, 52, 62, 93, 103, 108, 120, 121, 126, 127, 137, 149, 152, 171, 173, 175, 178, 179, 181, 183, 190, 193, 199, 208, 212, 239, 313, 358, 584, 622, 641, 661, 689, 690, 694, 736, 737, 738, 803, 804, 808, 816, 825, 846, 847, 851, 855, 857, 860, 861, 863, 864, 868, 872, 876, 906, 907, 908, 909, 913, 916, 917, 918, 919, 923, 924, 929, 930, 931, 938, 939, 940, 941, 942, 950, 951, 952, 953, 956, 957, 958, 967, 968, 1029, 1030, 1065, 1072, 1075, 1076, 1077, 1078, 1079, 1080, 1081, 1082, 1086, 1087, 1088, 1089, 1090, 1195, 1196, 1197, 1198, 1199, 1200, 1201, 1209, 1210, 1214, 1215, 1216, 1223, 1224, 1225, 1226, 1227, 1228, 1229, 1234, 1235, 1267, 1269, 1270.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Building HL from end points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 264 disconnected components.\n",
      " There are 186 islands with ids: 0, 1, 5, 9, 10, 11, 17, 25, 26, 27, 29, 31, 35, 36, 43, 44, 45, 46, 47, 48, 49, 50, 53, 60, 62, 73, 75, 82, 84, 91, 103, 107, 128, 136, 139, 161, 173, 181, 183, 267, 281, 302, 309, 314, 315, 316, 317, 318, 319, 320, 322, 329, 330, 331, 333, 356, 357, 358, 452, 489, 602, 875, 963, 971, 972, 991, 992, 1031, 1056, 1057, 1064, 1088, 1101, 1102, 1118, 1126, 1148, 1155, 1248, 1249, 1251, 1252, 1263, 1264, 1272, 1282, 1283, 1285, 1292, 1293, 1296, 1297, 1308, 1328, 1336, 1337, 1342, 1345, 1347, 1348, 1349, 1352, 1353, 1354, 1356, 1357, 1358, 1363, 1367, 1408, 1418, 1421, 1422, 1423, 1425, 1431, 1432, 1433, 1434, 1439, 1440, 1443, 1444, 1445, 1446, 1447, 1448, 1449, 1450, 1451, 1457, 1459, 1464, 1465, 1466, 1467, 1468, 1474, 1475, 1476, 1477, 1478, 1482, 1485, 1489, 1507, 1508, 1522, 1529, 1550, 1570, 1599, 1621, 1624, 1627, 1628, 1668, 1688, 1691, 1692, 1695, 1696, 1697, 1700, 1701, 1704, 1705, 1711, 1712, 1714, 1715, 1716, 1717, 1720, 1721, 1726, 1858, 1859, 1860, 1862, 1874, 1876, 1879, 1880, 1881, 1916.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Combining HL from start points and end points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 349 disconnected components.\n",
      " There are 234 islands with ids: 0, 1, 5, 10, 11, 20, 25, 26, 27, 30, 31, 41, 43, 44, 45, 46, 47, 48, 49, 54, 60, 71, 186, 228, 229, 231, 257, 262, 263, 264, 265, 268, 269, 270, 301, 308, 315, 325, 990, 991, 1128, 1129, 1130, 1133, 1134, 1139, 1140, 1141, 1142, 1143, 1144, 1182, 1183, 1186, 1208, 1210, 1211, 1244, 1245, 1318, 1321, 1324, 1325, 1326, 1327, 1328, 1331, 1332, 1334, 1339, 1345, 1346, 1348, 1358, 1359, 1360, 1361, 1362, 1363, 1364, 1369, 1415, 1433, 1437, 1438, 1439, 1440, 1441, 1443, 1444, 1445, 1446, 1447, 1448, 1450, 1451, 1458, 1461, 1464, 1465, 1466, 1467, 1468, 1475, 1477, 1531, 1534, 1535, 1538, 1539, 1587, 1617, 1618, 1621, 1624, 1685, 1686, 1687, 1688, 1691, 1692, 1695, 1696, 1697, 1700, 1704, 1705, 1712, 1714, 1715, 1716, 1717, 1720, 1721, 1857, 1859, 1860, 1863, 1874, 1878, 1879, 1880, 1916, 1926, 1929, 1931, 1941, 1942, 1945, 1946, 1996, 2001, 2040, 2072, 2073, 2078, 2084, 2085, 2101, 2279, 2287, 2530, 2534, 2537, 2543, 2616, 2622, 2625, 2725, 2726, 2727, 2759, 2760, 2763, 2769, 2775, 2777, 2778, 2813, 2826, 2827, 2828, 2829, 2837, 2838, 2839, 2841, 2842, 2846, 2847, 2848, 2855, 2857, 2869, 2870, 2871, 2872, 2873, 2874, 2940, 2946, 2947, 2982, 2992, 2993, 2994, 2997, 2998, 2999, 3003, 3004, 3005, 3006, 3112, 3113, 3114, 3115, 3116, 3118, 3129, 3131, 3132, 3133, 3140, 3141, 3142, 3143, 3144, 3145, 3146, 3151, 3152, 3185, 3187.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of users for which at least on Home Location has been identified:  71\n",
      "Number of unique HL tiles: 349\n",
      "Done.\n",
      "\n",
      "Matching trips to HL...\n",
      "Number of unmatched trajectories (concatenated) that do neither start nor end in a HL tile: 223/4310\n",
      "Number of trajectories (concatenated) that start AND end in a HL tile: 2524/4310\n",
      "Number of trips that match different HL tiles with their SP and EP: 1200\n",
      "Done.\n",
      "\n",
      "Matching double assigned trips to unique HL...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:  1.9min\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:  4.7min\n",
      "[Parallel(n_jobs=-4)]: Done 1000 tasks      | elapsed:  8.7min\n",
      "[Parallel(n_jobs=-4)]: Done 1550 tasks      | elapsed: 13.3min\n",
      "[Parallel(n_jobs=-4)]: Done 2400 out of 2400 | elapsed: 23.6min finished\n",
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "Done.\n",
      "\n",
      "Getting trajectories that happened during the same time...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:  1.4min\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:  3.9min\n",
      "[Parallel(n_jobs=-4)]: Done 1000 tasks      | elapsed:  7.1min\n",
      "[Parallel(n_jobs=-4)]: Done 1550 tasks      | elapsed: 10.9min\n",
      "[Parallel(n_jobs=-4)]: Done 2200 tasks      | elapsed: 15.5min\n",
      "[Parallel(n_jobs=-4)]: Done 2950 tasks      | elapsed: 20.7min\n",
      "[Parallel(n_jobs=-4)]: Done 3800 tasks      | elapsed: 26.7min\n",
      "[Parallel(n_jobs=-4)]: Done 4310 out of 4310 | elapsed: 30.3min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Building clustering after HL matching step...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                           | 0/210 [00:00<?, ?it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 36 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.7s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.8s finished\n",
      "  2%|█▉                                                                                 | 5/210 [00:01<01:18,  2.61it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  36\n",
      "Running randomized subset search for 1000 iterations with 94 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.6s finished\n",
      "  3%|██▊                                                                                | 7/210 [00:03<01:53,  1.79it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  75\n",
      "Running randomized subset search for 1000 iterations with 314 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    2.1s\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:    5.0s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    9.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  183\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  4%|███▌                                                                               | 9/210 [00:13<06:40,  1.99s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 2049 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 205 tasks      | elapsed:    5.2s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:   24.2s finished\n",
      "  8%|██████▏                                                                           | 16/210 [00:37<09:30,  2.94s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  92\n",
      "Running randomized subset search for 1000 iterations with 212 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    4.6s\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:   10.9s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:   17.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  212\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  8%|██████▋                                                                           | 17/210 [00:55<15:03,  4.68s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 58 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.7s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.9s finished\n",
      " 10%|███████▊                                                                          | 20/210 [00:57<10:38,  3.36s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  57\n",
      "Running randomized subset search for 1000 iterations with 46 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.5s finished\n",
      " 10%|████████▏                                                                         | 21/210 [00:59<09:49,  3.12s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  46\n",
      "Running randomized subset search for 1000 iterations with 62 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.5s finished\n",
      " 10%|████████▌                                                                         | 22/210 [01:01<09:01,  2.88s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  62\n",
      "Running randomized subset search for 1000 iterations with 44 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.4s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.4s finished\n",
      " 13%|██████████▉                                                                       | 28/210 [01:02<04:05,  1.35s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  40\n",
      "Running randomized subset search for 1000 iterations with 66 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.4s finished\n",
      " 14%|███████████▋                                                                      | 30/210 [01:04<03:39,  1.22s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  66\n",
      "Running randomized subset search for 1000 iterations with 103 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.7s finished\n",
      " 15%|████████████▍                                                                     | 32/210 [01:05<03:24,  1.15s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  90\n",
      "Running randomized subset search for 1000 iterations with 45 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.7s\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:    1.3s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    2.0s finished\n",
      " 18%|██████████████▊                                                                   | 38/210 [01:07<02:08,  1.33it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  45\n",
      "Running randomized subset search for 1000 iterations with 31 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.4s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.0s finished\n",
      " 29%|███████████████████████▍                                                          | 60/210 [01:09<00:36,  4.10it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  31\n",
      "Running randomized subset search for 1000 iterations with 40 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.4s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.1s finished\n",
      " 43%|███████████████████████████████████▌                                              | 91/210 [01:10<00:14,  8.08it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  40\n",
      "Running randomized subset search for 1000 iterations with 76 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.7s finished\n",
      " 54%|███████████████████████████████████████████▉                                     | 114/210 [01:12<00:10,  9.29it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  76\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 72%|██████████████████████████████████████████████████████████▋                      | 152/210 [01:12<00:03, 17.34it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 114 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.7s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    2.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  114\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 80%|████████████████████████████████████████████████████████████████▊                | 168/210 [01:15<00:03, 11.85it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 250 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    1.0s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    4.7s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  140\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████████| 210/210 [01:20<00:00,  2.62it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Clustering results after concatenation step:\n",
      "Number of unique clusters: 4310\n",
      "Homogeneity: 0.994\n",
      "Completeness: 0.440\n",
      "V-measure: 0.610\n",
      "Rand index: 0.967\n",
      "ARI: 0.000\n",
      "MI: 3.678\n",
      "NMI: 0.610\n",
      "AMI: 0.008\n",
      "Cluster accuracy: 0.022\n",
      "\n",
      "Clustering results after HL matching step:\n",
      "Number of unique clusters: 2675\n",
      "Homogeneity: 0.915\n",
      "Completeness: 0.502\n",
      "V-measure: 0.649\n",
      "Rand index: 0.970\n",
      "ARI: 0.262\n",
      "MI: 3.387\n",
      "NMI: 0.649\n",
      "AMI: 0.335\n",
      "Cluster accuracy: 0.276\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Run attack on private data\n",
    "attack.plot_hour_of_day_distribution(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_number_of_trips_per_user(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_trip_distances(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_trip_durations(raw_full_trip_gdf)\n",
    "\n",
    "# Merge Start Points (SP) and End Points (EP) with Tessellation\n",
    "print(\"\\nMatching start and end points with tessellation...\")\n",
    "gdf_sp, gdf_ep = attack.match_boundary_points_with_tessellation(raw_trip_sp_gdf, raw_trip_ep_gdf, tesselation_gdf)\n",
    "print(\"Done.\")\n",
    "\n",
    "# Extract Full Trips that Start and End within Tessellation Area\n",
    "print(\"\\nExtracting trips that start and end within tessellation area...\")\n",
    "full_trip_gdf, trip_sp_gdf, trip_ep_gdf, gdf_sp, gdf_ep = attack.extract_trips_that_start_end_in_tessellation(raw_full_trip_gdf, raw_trip_sp_gdf, raw_trip_ep_gdf, gdf_sp, gdf_ep)\n",
    "print(\"Done.\")\n",
    "\n",
    "# Build mapping of trip chains\n",
    "print('\\nBuilding mapping of trip chains...')\n",
    "mapping_cont_trips = attack.build_trip_chain_mapping(gdf_sp, gdf_ep)\n",
    "print('Done.')\n",
    "\n",
    "# Evaluate Trip Chaining\n",
    "print('\\nEvaluating trip chaining...')\n",
    "attack.evaluate_trip_chaining(mapping_cont_trips, full_trip_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Concatenate Trips\n",
    "full_trips_concat_gdf, trip_concat_dict = attack.merge_trips_from_matching(gdf_sp, mapping_cont_trips, full_trip_gdf)\n",
    "\n",
    "# Extract data for concatenated trips\n",
    "print('\\nExtracting data for concatenated trips...')\n",
    "gdf_sp_concat, trip_sp_gdf_concat, gdf_ep_concat, trip_ep_gdf_concat = attack.extract_concatenated_trips(full_trips_concat_gdf, gdf_sp, trip_sp_gdf, gdf_ep, trip_ep_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Clustering after Concatenation step\n",
    "print('\\nBuilding clustering after concatenation...')\n",
    "clustering_concat = attack.build_clustering_after_concatenation(full_trips_concat_gdf, trip_concat_dict, full_trip_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Home Locations (HL)\n",
    "# Build HL from Start Points\n",
    "print('\\nBuilding HL from start points...')\n",
    "gdf_hl_combined_sp = attack.build_hl_from_start_points(gdf_sp, gdf_ep, HL_SP_OUTFLOW_THRESHOLD=HL_SP_OUTFLOW_THRESHOLD)\n",
    "print('Done.')\n",
    "# Build HL from End Points\n",
    "print('\\nBuilding HL from end points...')\n",
    "gdf_hl_combined_ep = attack.build_hl_from_end_points(gdf_sp, gdf_ep, HL_EP_OUTFLOW_THRESHOLD=HL_EP_OUTFLOW_THRESHOLD)\n",
    "print('Done.')\n",
    "# Combine HL from Start Points and End Points\n",
    "print('\\nCombining HL from start points and end points...')\n",
    "gp_combined, HL_table = attack.concatenate_hl(gdf_hl_combined_sp, gdf_hl_combined_ep)\n",
    "print('Done.')\n",
    "\n",
    "print('\\nMatching trips to HL...')\n",
    "HL_table_se_concat, unmatched_trips, double_assigned_trips, nr_unmatched = attack.match_trips_to_HL(gp_combined, HL_table, trip_sp_gdf_concat, trip_ep_gdf_concat, full_trips_concat_gdf)\n",
    "print('Done.')\n",
    "\n",
    "print('\\nMatching double assigned trips to unique HL...')\n",
    "HL_table_trips_concat = attack.assign_double_matched_trips_to_unique_hl(HL_table_se_concat, full_trips_concat_gdf, unmatched_trips, double_assigned_trips, nr_unmatched)\n",
    "print('Done.')\n",
    "\n",
    "# Get trajectories that happened during the same time\n",
    "print('\\nGetting trajectories that happened during the same time...')\n",
    "full_trips_concat_gdf_overlap_dict = attack.getTripOverlaps(full_trips_concat_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Clustering after HL matching step\n",
    "print('\\nBuilding clustering after HL matching step...')\n",
    "clustering_after_HL, HL_table_dict = attack.build_clustering_after_HL_assignment(HL_table_trips_concat, full_trip_gdf, trip_concat_dict, full_trips_concat_gdf_overlap_dict)\n",
    "print('Done.')\n",
    "\n",
    "# Evaluate clustering Results\n",
    "print(\"\\nClustering results after concatenation step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_concat))}\")\n",
    "attack.evaluate(clustering_concat, full_trip_gdf)\n",
    "clustering_HL = list(dict(sorted(clustering_after_HL.items())).values())\n",
    "print(\"\\nClustering results after HL matching step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_HL))}\")\n",
    "attack.evaluate(clustering_HL, full_trip_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/geopandas/io/file.py:362: FutureWarning: pandas.Int64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n",
      "  pd.Int64Index,\n"
     ]
    }
   ],
   "source": [
    "full_trip_gdf = full_trip_gdf.copy()\n",
    "full_trip_gdf['clustering_HL'] = clustering_HL\n",
    "full_trip_gdf.to_file('../data/geolife/geolife_clustering_HL_private.geojson', driver=\"GeoJSON\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Raw (500m grid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading raw geolife geojson file...\n",
      "Done.\n",
      "Number of trajectories in loaded dataset: 5400\n",
      "Number of users in loaded dataset: 73\n"
     ]
    }
   ],
   "source": [
    "# Load preprocessed data \n",
    "raw_full_trip_gdf, raw_trip_sp_gdf, raw_trip_ep_gdf, tesselation_gdf = dl.load_geolife(data_type='raw', \n",
    "                                                                                       min_n_trips_per_user=1,\n",
    "                                                                                       rand_n_week_period=None,\n",
    "                                                                                       tessellation_diameter=500,\n",
    "                                                                                       min_trip_length=200,\n",
    "                                                                                       min_nr_points=50,\n",
    "                                                                                       upper_quantile_length=0.95,\n",
    "                                                                                       only_2008=True)\n",
    "assert len(raw_full_trip_gdf) == len(raw_trip_sp_gdf) == len(raw_trip_ep_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Matching start and end points with tessellation...\n",
      "Done.\n",
      "\n",
      "Extracting trips that start and end within tessellation area...\n",
      "Number of trips that start and end wihin tessellation area: 5101\n",
      "Number of trips outside and therefore dropped: 299\n",
      "Done.\n",
      "\n",
      "Building mapping of trip chains...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████| 5101/5101 [01:24<00:00, 60.52it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Evaluating trip chaining...\n",
      "Number of edges (matched) between trips: 738\n",
      "Number of wrong matches: 225\n",
      "Done.\n",
      "\n",
      "Building trip chains...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████████████████████████████| 5101/5101 [00:00<00:00, 26128.92it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Merging trips...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████| 5101/5101 [00:57<00:00, 88.34it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "Number of trips that were matched at least once: 5032/5101\n",
      "Concatenating MERGED and UNMERGED trips...\n",
      "Done.\n",
      "\n",
      "Extracting data for concatenated trips...\n",
      "Done.\n",
      "\n",
      "Building clustering after concatenation...\n",
      "Number of unique clusters: 4432\n",
      "Done.\n",
      "\n",
      "Building HL from start points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 96 disconnected components.\n",
      " There are 58 islands with ids: 9, 11, 13, 15, 16, 17, 28, 29, 30, 60, 121, 131, 177, 178, 220, 221, 231, 236, 347, 523, 661, 734, 817, 881, 943, 961, 962, 1020, 1026, 1040, 1041, 1053, 1054, 1055, 1056, 1057, 1063, 1136, 1137, 1195, 1206, 1207, 1208, 1213, 1214, 1220, 1345, 1346, 1347, 1360, 1361, 1362, 1369, 1370, 1371, 1376, 1377, 1415.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Building HL from end points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 136 disconnected components.\n",
      " There are 70 islands with ids: 0, 1, 6, 12, 14, 15, 27, 28, 31, 34, 35, 36, 48, 49, 52, 53, 68, 69, 70, 113, 115, 117, 322, 399, 541, 542, 546, 547, 780, 1148, 1152, 1258, 1259, 1270, 1357, 1464, 1506, 1519, 1521, 1560, 1578, 1579, 1613, 1619, 1629, 1637, 1639, 1640, 1641, 1642, 1647, 1648, 1749, 1751, 1752, 1868, 1869, 1873, 1889, 1892, 1894, 1898, 1902, 2042, 2043, 2049, 2064, 2069, 2070, 2107.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Combining HL from start points and end points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 144 disconnected components.\n",
      " There are 76 islands with ids: 0, 1, 6, 14, 15, 29, 30, 31, 34, 35, 36, 49, 52, 53, 1104, 1105, 1106, 1124, 1154, 1185, 1378, 1380, 1383, 1446, 1447, 1448, 1496, 1567, 1597, 1599, 1600, 1620, 1637, 1646, 1797, 1871, 1873, 1892, 1894, 1900, 1903, 2042, 2049, 2064, 2069, 2107, 2119, 2122, 2123, 2125, 2136, 2138, 2976, 2982, 3058, 3129, 3230, 3237, 3239, 3241, 3244, 3304, 3316, 3317, 3322, 3443, 3453, 3454, 3468, 3469, 3470, 3477, 3479, 3522, 3523, 3524.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of users for which at least on Home Location has been identified:  72\n",
      "Number of unique HL tiles: 144\n",
      "Done.\n",
      "\n",
      "Matching trips to HL...\n",
      "Number of unmatched trajectories (concatenated) that do neither start nor end in a HL tile: 69/4432\n",
      "Number of trajectories (concatenated) that start AND end in a HL tile: 3514/4432\n",
      "Number of trips that match different HL tiles with their SP and EP: 1248\n",
      "Done.\n",
      "\n",
      "Matching double assigned trips to unique HL...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 201 tasks      | elapsed:  3.9min\n",
      "[Parallel(n_jobs=-4)]: Done 551 tasks      | elapsed: 11.6min\n",
      "[Parallel(n_jobs=-4)]: Done 1001 tasks      | elapsed: 20.5min\n",
      "[Parallel(n_jobs=-4)]: Done 1551 tasks      | elapsed: 28.7min\n",
      "[Parallel(n_jobs=-4)]: Done 2201 tasks      | elapsed: 37.4min\n",
      "[Parallel(n_jobs=-4)]: Done 2496 out of 2496 | elapsed: 53.4min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "Done.\n",
      "\n",
      "Getting trajectories that happened during the same time...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 201 tasks      | elapsed:  1.8min\n",
      "[Parallel(n_jobs=-4)]: Done 551 tasks      | elapsed:  4.9min\n",
      "[Parallel(n_jobs=-4)]: Done 1001 tasks      | elapsed:  8.8min\n",
      "[Parallel(n_jobs=-4)]: Done 1551 tasks      | elapsed: 13.6min\n",
      "[Parallel(n_jobs=-4)]: Done 2201 tasks      | elapsed: 19.3min\n",
      "[Parallel(n_jobs=-4)]: Done 2951 tasks      | elapsed: 25.8min\n",
      "[Parallel(n_jobs=-4)]: Done 3801 tasks      | elapsed: 33.3min\n",
      "[Parallel(n_jobs=-4)]: Done 4432 out of 4432 | elapsed: 38.8min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Building clustering after HL matching step...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                            | 0/79 [00:00<?, ?it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 2909 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:   10.4s\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:   22.7s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:   42.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  125\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  6%|█████▎                                                                              | 5/79 [00:43<09:04,  7.35s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 218 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    2.5s\n",
      "[Parallel(n_jobs=-4)]: Done 751 out of 1000 | elapsed:    8.1s remaining:    2.7s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:   10.4s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  207\n",
      "Running randomized subset search for 1000 iterations with 32 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.5s finished\n",
      "  9%|███████▍                                                                            | 7/79 [00:55<08:15,  6.88s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  32\n",
      "Running randomized subset search for 1000 iterations with 56 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.5s finished\n",
      " 10%|████████▌                                                                           | 8/79 [00:57<06:44,  5.70s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  53\n",
      "Running randomized subset search for 1000 iterations with 53 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.5s finished\n",
      " 14%|███████████▌                                                                       | 11/79 [00:58<03:34,  3.16s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  53\n",
      "Running randomized subset search for 1000 iterations with 56 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.3s finished\n",
      " 15%|████████████▌                                                                      | 12/79 [01:00<03:09,  2.83s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  46\n",
      "Running randomized subset search for 1000 iterations with 221 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.7s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    3.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  96\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 18%|██████████████▋                                                                    | 14/79 [01:03<02:38,  2.44s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 130 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    1.4s\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:    3.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    6.2s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  130\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 23%|██████████████████▉                                                                | 18/79 [01:10<02:03,  2.03s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 275 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 201 tasks      | elapsed:    1.7s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    7.5s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  176\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 28%|███████████████████████                                                            | 22/79 [01:17<01:54,  2.01s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 63 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.6s finished\n",
      " 29%|████████████████████████▏                                                          | 23/79 [01:19<01:50,  1.98s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  63\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 47%|██████████████████████████████████████▊                                            | 37/79 [01:19<00:23,  1.80it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 41 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.4s finished\n",
      " 65%|█████████████████████████████████████████████████████▌                             | 51/79 [01:21<00:08,  3.46it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  41\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████████| 79/79 [01:21<00:00,  1.03s/it]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Clustering results after concatenation step:\n",
      "Number of unique clusters: 4432\n",
      "Homogeneity: 0.986\n",
      "Completeness: 0.437\n",
      "V-measure: 0.606\n",
      "Rand index: 0.965\n",
      "ARI: 0.003\n",
      "MI: 3.649\n",
      "NMI: 0.606\n",
      "AMI: 0.054\n",
      "Cluster accuracy: 0.027\n",
      "\n",
      "Clustering results after HL matching step:\n",
      "Number of unique clusters: 3183\n",
      "Homogeneity: 0.920\n",
      "Completeness: 0.475\n",
      "V-measure: 0.627\n",
      "Rand index: 0.967\n",
      "ARI: 0.188\n",
      "MI: 3.406\n",
      "NMI: 0.627\n",
      "AMI: 0.268\n",
      "Cluster accuracy: 0.204\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Run attack on private data with big tessellation\n",
    "attack.plot_hour_of_day_distribution(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_number_of_trips_per_user(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_trip_distances(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_trip_durations(raw_full_trip_gdf)\n",
    "\n",
    "# Merge Start Points (SP) and End Points (EP) with Tessellation\n",
    "print(\"\\nMatching start and end points with tessellation...\")\n",
    "gdf_sp, gdf_ep = attack.match_boundary_points_with_tessellation(raw_trip_sp_gdf, raw_trip_ep_gdf, tesselation_gdf)\n",
    "print(\"Done.\")\n",
    "\n",
    "# Extract Full Trips that Start and End within Tessellation Area\n",
    "print(\"\\nExtracting trips that start and end within tessellation area...\")\n",
    "full_trip_gdf, trip_sp_gdf, trip_ep_gdf, gdf_sp, gdf_ep = attack.extract_trips_that_start_end_in_tessellation(raw_full_trip_gdf, raw_trip_sp_gdf, raw_trip_ep_gdf, gdf_sp, gdf_ep)\n",
    "print(\"Done.\")\n",
    "\n",
    "# Build mapping of trip chains\n",
    "print('\\nBuilding mapping of trip chains...')\n",
    "mapping_cont_trips = attack.build_trip_chain_mapping(gdf_sp, gdf_ep)\n",
    "print('Done.')\n",
    "\n",
    "# Evaluate Trip Chaining\n",
    "print('\\nEvaluating trip chaining...')\n",
    "attack.evaluate_trip_chaining(mapping_cont_trips, full_trip_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Concatenate Trips\n",
    "full_trips_concat_gdf, trip_concat_dict = attack.merge_trips_from_matching(gdf_sp, mapping_cont_trips, full_trip_gdf)\n",
    "\n",
    "# Extract data for concatenated trips\n",
    "print('\\nExtracting data for concatenated trips...')\n",
    "gdf_sp_concat, trip_sp_gdf_concat, gdf_ep_concat, trip_ep_gdf_concat = attack.extract_concatenated_trips(full_trips_concat_gdf, gdf_sp, trip_sp_gdf, gdf_ep, trip_ep_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Clustering after Concatenation step\n",
    "print('\\nBuilding clustering after concatenation...')\n",
    "clustering_concat = attack.build_clustering_after_concatenation(full_trips_concat_gdf, trip_concat_dict, full_trip_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Home Locations (HL)\n",
    "# Build HL from Start Points\n",
    "print('\\nBuilding HL from start points...')\n",
    "gdf_hl_combined_sp = attack.build_hl_from_start_points(gdf_sp, gdf_ep, HL_SP_OUTFLOW_THRESHOLD=HL_SP_OUTFLOW_THRESHOLD)\n",
    "print('Done.')\n",
    "# Build HL from End Points\n",
    "print('\\nBuilding HL from end points...')\n",
    "gdf_hl_combined_ep = attack.build_hl_from_end_points(gdf_sp, gdf_ep, HL_EP_OUTFLOW_THRESHOLD=HL_EP_OUTFLOW_THRESHOLD)\n",
    "print('Done.')\n",
    "# Combine HL from Start Points and End Points\n",
    "print('\\nCombining HL from start points and end points...')\n",
    "gp_combined, HL_table = attack.concatenate_hl(gdf_hl_combined_sp, gdf_hl_combined_ep)\n",
    "print('Done.')\n",
    "\n",
    "print('\\nMatching trips to HL...')\n",
    "HL_table_se_concat, unmatched_trips, double_assigned_trips, nr_unmatched = attack.match_trips_to_HL(gp_combined, HL_table, trip_sp_gdf_concat, trip_ep_gdf_concat, full_trips_concat_gdf)\n",
    "print('Done.')\n",
    "\n",
    "print('\\nMatching double assigned trips to unique HL...')\n",
    "HL_table_trips_concat = attack.assign_double_matched_trips_to_unique_hl(HL_table_se_concat, full_trips_concat_gdf, unmatched_trips, double_assigned_trips, nr_unmatched)\n",
    "print('Done.')\n",
    "\n",
    "# Get trajectories that happened during the same time\n",
    "print('\\nGetting trajectories that happened during the same time...')\n",
    "full_trips_concat_gdf_overlap_dict = attack.getTripOverlaps(full_trips_concat_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Clustering after HL matching step\n",
    "print('\\nBuilding clustering after HL matching step...')\n",
    "clustering_after_HL, HL_table_dict = attack.build_clustering_after_HL_assignment(HL_table_trips_concat, full_trip_gdf, trip_concat_dict, full_trips_concat_gdf_overlap_dict)\n",
    "print('Done.')\n",
    "\n",
    "# Evaluate clustering Results\n",
    "print(\"\\nClustering results after concatenation step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_concat))}\")\n",
    "attack.evaluate(clustering_concat, full_trip_gdf)\n",
    "clustering_HL = list(dict(sorted(clustering_after_HL.items())).values())\n",
    "print(\"\\nClustering results after HL matching step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_HL))}\")\n",
    "attack.evaluate(clustering_HL, full_trip_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/geopandas/io/file.py:362: FutureWarning: pandas.Int64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n",
      "  pd.Int64Index,\n"
     ]
    }
   ],
   "source": [
    "full_trip_gdf = full_trip_gdf.copy()\n",
    "full_trip_gdf['clustering_HL'] = clustering_HL\n",
    "full_trip_gdf.to_file('../data/geolife/geolife_clustering_HL_500tessellation.geojson', driver=\"GeoJSON\")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Obfuscated (500m grid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading private geolife geojson file...\n",
      "Done.\n",
      "Number of trajectories in loaded dataset: 4719\n",
      "Number of users in loaded dataset: 73\n"
     ]
    }
   ],
   "source": [
    "# Load preprocessed data \n",
    "raw_full_trip_gdf, raw_trip_sp_gdf, raw_trip_ep_gdf, tesselation_gdf = dl.load_geolife(data_type='private', \n",
    "                                                                                       min_n_trips_per_user=1,\n",
    "                                                                                       rand_n_week_period=None,\n",
    "                                                                                       tessellation_diameter=500,\n",
    "                                                                                       min_trip_length=200,\n",
    "                                                                                       min_nr_points=50,\n",
    "                                                                                       upper_quantile_length=0.95,\n",
    "                                                                                       only_2008=True)\n",
    "assert len(raw_full_trip_gdf) == len(raw_trip_sp_gdf) == len(raw_trip_ep_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Matching start and end points with tessellation...\n",
      "Done.\n",
      "\n",
      "Extracting trips that start and end within tessellation area...\n",
      "Number of trips that start and end wihin tessellation area: 4449\n",
      "Number of trips outside and therefore dropped: 270\n",
      "Done.\n",
      "\n",
      "Building mapping of trip chains...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████| 4449/4449 [01:10<00:00, 63.01it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Evaluating trip chaining...\n",
      "Number of edges (matched) between trips: 394\n",
      "Number of wrong matches: 182\n",
      "Done.\n",
      "\n",
      "Building trip chains...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████████████████████████████| 4449/4449 [00:00<00:00, 56389.64it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Merging trips...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████| 4449/4449 [00:51<00:00, 86.91it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "Number of trips that were matched at least once: 4410/4449\n",
      "Concatenating MERGED and UNMERGED trips...\n",
      "Done.\n",
      "\n",
      "Extracting data for concatenated trips...\n",
      "Done.\n",
      "\n",
      "Building clustering after concatenation...\n",
      "Number of unique clusters: 4094\n",
      "Done.\n",
      "\n",
      "Building HL from start points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 96 disconnected components.\n",
      " There are 58 islands with ids: 5, 9, 12, 13, 14, 25, 26, 27, 34, 93, 106, 141, 175, 194, 197, 198, 313, 314, 572, 706, 856, 862, 903, 913, 914, 921, 922, 924, 929, 940, 942, 943, 954, 955, 956, 957, 1029, 1030, 1076, 1080, 1081, 1086, 1087, 1088, 1089, 1196, 1197, 1198, 1209, 1214, 1215, 1216, 1224, 1225, 1229, 1240, 1267, 1270.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Building HL from end points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 121 disconnected components.\n",
      " There are 60 islands with ids: 0, 1, 5, 6, 10, 11, 18, 25, 26, 27, 31, 43, 44, 47, 48, 62, 96, 163, 183, 324, 328, 441, 452, 1000, 1100, 1158, 1319, 1335, 1348, 1349, 1357, 1365, 1369, 1411, 1432, 1444, 1454, 1471, 1474, 1475, 1510, 1598, 1600, 1627, 1628, 1680, 1692, 1709, 1712, 1713, 1718, 1726, 1859, 1861, 1862, 1874, 1879, 1880, 1881, 1916.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Combining HL from start points and end points...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/libpysal/weights/weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 139 disconnected components.\n",
      " There are 82 islands with ids: 0, 1, 5, 10, 11, 20, 25, 26, 27, 31, 46, 47, 48, 79, 1057, 1089, 1268, 1309, 1311, 1333, 1336, 1345, 1353, 1362, 1363, 1364, 1386, 1387, 1457, 1581, 1582, 1623, 1624, 1685, 1686, 1694, 1713, 1715, 1718, 1858, 1859, 1862, 1863, 1874, 1879, 1916, 1926, 1929, 1931, 1942, 2671, 2672, 2674, 2769, 2779, 2838, 2839, 2840, 2932, 2933, 2939, 2940, 2941, 2942, 2946, 2997, 3003, 3004, 3005, 3006, 3113, 3114, 3115, 3130, 3131, 3132, 3133, 3142, 3146, 3185, 3186, 3187.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of users for which at least on Home Location has been identified:  71\n",
      "Number of unique HL tiles: 139\n",
      "Done.\n",
      "\n",
      "Matching trips to HL...\n",
      "Number of unmatched trajectories (concatenated) that do neither start nor end in a HL tile: 89/4094\n",
      "Number of trajectories (concatenated) that start AND end in a HL tile: 3094/4094\n",
      "Number of trips that match different HL tiles with their SP and EP: 1108\n",
      "Done.\n",
      "\n",
      "Matching double assigned trips to unique HL...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 201 tasks      | elapsed:  3.2min\n",
      "[Parallel(n_jobs=-4)]: Done 551 tasks      | elapsed:  7.7min\n",
      "[Parallel(n_jobs=-4)]: Done 1001 tasks      | elapsed: 13.1min\n",
      "[Parallel(n_jobs=-4)]: Done 1551 tasks      | elapsed: 18.7min\n",
      "[Parallel(n_jobs=-4)]: Done 2216 out of 2216 | elapsed: 30.6min finished\n",
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "All scores are 0.0, assigning -1 as best_hl_id!\n",
      "Done.\n",
      "\n",
      "Getting trajectories that happened during the same time...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 201 tasks      | elapsed:  1.5min\n",
      "[Parallel(n_jobs=-4)]: Done 551 tasks      | elapsed:  4.0min\n",
      "[Parallel(n_jobs=-4)]: Done 1001 tasks      | elapsed:  7.1min\n",
      "[Parallel(n_jobs=-4)]: Done 1551 tasks      | elapsed: 11.0min\n",
      "[Parallel(n_jobs=-4)]: Done 2201 tasks      | elapsed: 15.6min\n",
      "[Parallel(n_jobs=-4)]: Done 2951 tasks      | elapsed: 20.9min\n",
      "[Parallel(n_jobs=-4)]: Done 3801 tasks      | elapsed: 26.9min\n",
      "[Parallel(n_jobs=-4)]: Done 4094 out of 4094 | elapsed: 29.0min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Building clustering after HL matching step...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                            | 0/73 [00:00<?, ?it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 2686 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    7.7s\n",
      "[Parallel(n_jobs=-4)]: Done 550 tasks      | elapsed:   17.8s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:   34.4s finished\n",
      "  3%|██▎                                                                                 | 2/73 [00:34<20:26, 17.28s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  89\n",
      "Running randomized subset search for 1000 iterations with 32 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.4s finished\n",
      "  4%|███▍                                                                                | 3/73 [00:36<12:28, 10.70s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  32\n",
      "Running randomized subset search for 1000 iterations with 33 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.3s finished\n",
      "  8%|██████▉                                                                             | 6/73 [00:37<04:37,  4.14s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  33\n",
      "Running randomized subset search for 1000 iterations with 220 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 202 tasks      | elapsed:    2.6s\n",
      "[Parallel(n_jobs=-4)]: Done 751 out of 1000 | elapsed:    8.4s remaining:    2.8s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:   10.8s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  199\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 11%|█████████▏                                                                          | 8/73 [00:48<05:04,  4.69s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 57 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.6s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.5s finished\n",
      " 12%|██████████▎                                                                         | 9/73 [00:50<04:18,  4.04s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  46\n",
      "Running randomized subset search for 1000 iterations with 58 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.4s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.3s finished\n",
      " 15%|████████████▌                                                                      | 11/73 [00:51<02:52,  2.78s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  46\n",
      "Running randomized subset search for 1000 iterations with 274 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.8s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    3.5s finished\n",
      " 16%|█████████████▋                                                                     | 12/73 [00:55<03:00,  2.97s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  99\n",
      "Running randomized subset search for 1000 iterations with 119 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n",
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.9s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    3.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  119\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 25%|████████████████████▍                                                              | 18/73 [00:58<01:19,  1.44s/it][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 244 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    1.3s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    5.4s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  132\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 51%|██████████████████████████████████████████                                         | 37/73 [01:04<00:16,  2.17it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 43 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.5s finished\n",
      " 59%|████████████████████████████████████████████████▉                                  | 43/73 [01:06<00:12,  2.46it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  43\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 92%|████████████████████████████████████████████████████████████████████████████▏      | 67/73 [01:06<00:00,  6.15it/s][Parallel(n_jobs=-4)]: Using backend LokyBackend with 125 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 1000 iterations with 50 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-4)]: Done 200 tasks      | elapsed:    0.4s\n",
      "[Parallel(n_jobs=-4)]: Done 1000 out of 1000 | elapsed:    1.3s finished\n",
      "100%|███████████████████████████████████████████████████████████████████████████████████| 73/73 [01:07<00:00,  1.08it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  50\n",
      "Done.\n",
      "\n",
      "Clustering results after concatenation step:\n",
      "Number of unique clusters: 4094\n",
      "Homogeneity: 0.986\n",
      "Completeness: 0.441\n",
      "V-measure: 0.609\n",
      "Rand index: 0.967\n",
      "ARI: 0.001\n",
      "MI: 3.650\n",
      "NMI: 0.609\n",
      "AMI: 0.025\n",
      "Cluster accuracy: 0.026\n",
      "\n",
      "Clustering results after HL matching step:\n",
      "Number of unique clusters: 3092\n",
      "Homogeneity: 0.928\n",
      "Completeness: 0.475\n",
      "V-measure: 0.628\n",
      "Rand index: 0.969\n",
      "ARI: 0.195\n",
      "MI: 3.435\n",
      "NMI: 0.628\n",
      "AMI: 0.231\n",
      "Cluster accuracy: 0.190\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Run attack on private data with big tessellation\n",
    "attack.plot_hour_of_day_distribution(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_number_of_trips_per_user(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_trip_distances(raw_full_trip_gdf)\n",
    "attack.plot_distribution_of_trip_durations(raw_full_trip_gdf)\n",
    "\n",
    "# Merge Start Points (SP) and End Points (EP) with Tessellation\n",
    "print(\"\\nMatching start and end points with tessellation...\")\n",
    "gdf_sp, gdf_ep = attack.match_boundary_points_with_tessellation(raw_trip_sp_gdf, raw_trip_ep_gdf, tesselation_gdf)\n",
    "print(\"Done.\")\n",
    "\n",
    "# Extract Full Trips that Start and End within Tessellation Area\n",
    "print(\"\\nExtracting trips that start and end within tessellation area...\")\n",
    "full_trip_gdf, trip_sp_gdf, trip_ep_gdf, gdf_sp, gdf_ep = attack.extract_trips_that_start_end_in_tessellation(raw_full_trip_gdf, raw_trip_sp_gdf, raw_trip_ep_gdf, gdf_sp, gdf_ep)\n",
    "print(\"Done.\")\n",
    "\n",
    "# Build mapping of trip chains\n",
    "print('\\nBuilding mapping of trip chains...')\n",
    "mapping_cont_trips = attack.build_trip_chain_mapping(gdf_sp, gdf_ep)\n",
    "print('Done.')\n",
    "\n",
    "# Evaluate Trip Chaining\n",
    "print('\\nEvaluating trip chaining...')\n",
    "attack.evaluate_trip_chaining(mapping_cont_trips, full_trip_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Concatenate Trips\n",
    "full_trips_concat_gdf, trip_concat_dict = attack.merge_trips_from_matching(gdf_sp, mapping_cont_trips, full_trip_gdf)\n",
    "\n",
    "# Extract data for concatenated trips\n",
    "print('\\nExtracting data for concatenated trips...')\n",
    "gdf_sp_concat, trip_sp_gdf_concat, gdf_ep_concat, trip_ep_gdf_concat = attack.extract_concatenated_trips(full_trips_concat_gdf, gdf_sp, trip_sp_gdf, gdf_ep, trip_ep_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Clustering after Concatenation step\n",
    "print('\\nBuilding clustering after concatenation...')\n",
    "clustering_concat = attack.build_clustering_after_concatenation(full_trips_concat_gdf, trip_concat_dict, full_trip_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Home Locations (HL)\n",
    "# Build HL from Start Points\n",
    "print('\\nBuilding HL from start points...')\n",
    "gdf_hl_combined_sp = attack.build_hl_from_start_points(gdf_sp, gdf_ep, HL_SP_OUTFLOW_THRESHOLD=HL_SP_OUTFLOW_THRESHOLD)\n",
    "print('Done.')\n",
    "# Build HL from End Points\n",
    "print('\\nBuilding HL from end points...')\n",
    "gdf_hl_combined_ep = attack.build_hl_from_end_points(gdf_sp, gdf_ep, HL_EP_OUTFLOW_THRESHOLD=HL_EP_OUTFLOW_THRESHOLD)\n",
    "print('Done.')\n",
    "# Combine HL from Start Points and End Points\n",
    "print('\\nCombining HL from start points and end points...')\n",
    "gp_combined, HL_table = attack.concatenate_hl(gdf_hl_combined_sp, gdf_hl_combined_ep)\n",
    "print('Done.')\n",
    "\n",
    "print('\\nMatching trips to HL...')\n",
    "HL_table_se_concat, unmatched_trips, double_assigned_trips, nr_unmatched = attack.match_trips_to_HL(gp_combined, HL_table, trip_sp_gdf_concat, trip_ep_gdf_concat, full_trips_concat_gdf)\n",
    "print('Done.')\n",
    "\n",
    "print('\\nMatching double assigned trips to unique HL...')\n",
    "HL_table_trips_concat = attack.assign_double_matched_trips_to_unique_hl(HL_table_se_concat, full_trips_concat_gdf, unmatched_trips, double_assigned_trips, nr_unmatched)\n",
    "print('Done.')\n",
    "\n",
    "# Get trajectories that happened during the same time\n",
    "print('\\nGetting trajectories that happened during the same time...')\n",
    "full_trips_concat_gdf_overlap_dict = attack.getTripOverlaps(full_trips_concat_gdf)\n",
    "print('Done.')\n",
    "\n",
    "# Build Clustering after HL matching step\n",
    "print('\\nBuilding clustering after HL matching step...')\n",
    "clustering_after_HL, HL_table_dict = attack.build_clustering_after_HL_assignment(HL_table_trips_concat, full_trip_gdf, trip_concat_dict, full_trips_concat_gdf_overlap_dict)\n",
    "print('Done.')\n",
    "\n",
    "# Evaluate clustering Results\n",
    "print(\"\\nClustering results after concatenation step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_concat))}\")\n",
    "attack.evaluate(clustering_concat, full_trip_gdf)\n",
    "clustering_HL = list(dict(sorted(clustering_after_HL.items())).values())\n",
    "print(\"\\nClustering results after HL matching step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_HL))}\")\n",
    "attack.evaluate(clustering_HL, full_trip_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.8/dist-packages/geopandas/io/file.py:362: FutureWarning: pandas.Int64Index is deprecated and will be removed from pandas in a future version. Use pandas.Index with the appropriate dtype instead.\n",
      "  pd.Int64Index,\n"
     ]
    }
   ],
   "source": [
    "full_trip_gdf = full_trip_gdf.copy()\n",
    "full_trip_gdf['clustering_HL'] = clustering_HL\n",
    "full_trip_gdf.to_file('../data/geolife/geolife_clustering_HL_private_500tessellation.geojson', driver=\"GeoJSON\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Merge Start Points (SP) and End Points (EP) with Tessellation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "gdf_sp, gdf_ep = attack.match_boundary_points_with_tessellation(raw_trip_sp_gdf, raw_trip_ep_gdf, tesselation_gdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Extract Full Trips that Start and End within Tessellation Area"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of trips that start and end wihin tessellation area: 290\n",
      "Number of trips outside and therefore dropped: 22\n"
     ]
    }
   ],
   "source": [
    "full_trip_gdf, trip_sp_gdf, trip_ep_gdf, gdf_sp, gdf_ep = attack.extract_trips_that_start_end_in_tessellation(raw_full_trip_gdf, raw_trip_sp_gdf, raw_trip_ep_gdf, gdf_sp, gdf_ep)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build mapping of trip chains"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 290/290 [00:05<00:00, 57.81it/s]\n"
     ]
    }
   ],
   "source": [
    "mapping_cont_trips = attack.build_trip_chain_mapping(gdf_sp, gdf_ep)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of edges (matched) between trips: 73\n",
      "Number of wrong matches: 0\n"
     ]
    }
   ],
   "source": [
    "attack.evaluate_trip_chaining(mapping_cont_trips, full_trip_gdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Merge trips according to matching"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Building trip chains...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 290/290 [00:00<00:00, 96704.42it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "\n",
      "Merging trips...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 290/290 [00:04<00:00, 65.87it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done.\n",
      "Number of trips that were matched at least once: 289/290\n",
      "Concatenating MERGED and UNMERGED trips...\n",
      "Done.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "full_trips_concat_gdf, trip_concat_dict = attack.merge_trips_from_matching(gdf_sp, mapping_cont_trips, full_trip_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "gdf_sp_concat, trip_sp_gdf_concat, gdf_ep_concat, trip_ep_gdf_concat = attack.extract_concatenated_trips(full_trips_concat_gdf, gdf_sp, trip_sp_gdf, gdf_ep, trip_ep_gdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build Clustering after Concatenation Step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of unique clusters: 218\n"
     ]
    }
   ],
   "source": [
    "clustering_concat = attack.build_clustering_after_concatenation(full_trips_concat_gdf, trip_concat_dict, full_trip_gdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Build Home Locations (HL)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## From Start Points (SPs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We use the SP-tessellation matching that still contains all SP (and potential HL), and not just the SP and EP of the concatenated trips. We do this, because we do not want loose potential HL contributed of substrip concatenated in a chain."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\Bened\\Documents\\Git\\Master-Thesis\\env\\lib\\site-packages\\libpysal\\weights\\weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 16 disconnected components.\n",
      " There are 9 islands with ids: 12, 14, 15, 18, 20, 26, 27, 49, 51.\n",
      "  warnings.warn(message)\n"
     ]
    }
   ],
   "source": [
    "gdf_hl_combined_sp = attack.build_hl_from_start_points(gdf_sp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## From End Points (EPs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\Bened\\Documents\\Git\\Master-Thesis\\env\\lib\\site-packages\\libpysal\\weights\\weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 32 disconnected components.\n",
      " There are 24 islands with ids: 5, 28, 29, 30, 38, 39, 40, 41, 42, 49, 50, 53, 54, 58, 59, 60, 63, 64, 73, 74, 79, 110, 111, 112.\n",
      "  warnings.warn(message)\n"
     ]
    }
   ],
   "source": [
    "gdf_hl_combined_ep = attack.build_hl_from_end_points(gdf_ep)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Merge (concatenate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\Bened\\Documents\\Git\\Master-Thesis\\env\\lib\\site-packages\\libpysal\\weights\\weights.py:172: UserWarning: The weights matrix is not fully connected: \n",
      " There are 40 disconnected components.\n",
      " There are 29 islands with ids: 15, 55, 56, 58, 59, 60, 61, 62, 70, 71, 76, 77, 102, 103, 104, 105, 106, 107, 108, 109, 111, 112, 149, 151, 152, 153, 158, 159, 164.\n",
      "  warnings.warn(message)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of users for which at least on Home Location has been identified:  5\n",
      "Number of unique HL tiles: 40\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "gp_combined, HL_table = attack.concatenate_hl(gdf_hl_combined_sp, gdf_hl_combined_ep)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Match trips with Home Location tiles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Match concatenated trips"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of unmatched trajectories (concatenated) that do neither start nor end in a HL tile: 10/218\n",
      "Number of trajectories (concatenated) that start AND end in a HL tile: 74/218\n",
      "Number of trips that match different HL tiles with their SP and EP: 44\n"
     ]
    }
   ],
   "source": [
    "HL_table_se_concat, unmatched_trips, double_assigned_trips, nr_unmatched = attack.match_trips_to_HL(gp_combined, HL_table, trip_sp_gdf_concat, trip_ep_gdf_concat, full_trips_concat_gdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Assign double matched trips to one unique HL"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Iterate over all double matched trips and compare them to all other trips that are uniquely assigned in their respective potential HLs that they have been matched with. Then take the HL with the single maximum lcss score between the trip under question and any trip of the assigned HL tiles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "HL_table_trips_concat = attack.assign_double_matched_trips_to_unique_hl(HL_table_se_concat, full_trips_concat_gdf, unmatched_trips, double_assigned_trips, nr_unmatched)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get trajectories that happened during the same time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_trips_concat_gdf_overlap_dict = attack.getTripOverlaps(full_trips_concat_gdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create clustering after HL assignment step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/26 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running randomized subset search for 200 iterations with 37 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 12%|█▏        | 3/26 [00:00<00:05,  4.39it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  37\n",
      "Running randomized subset search for 200 iterations with 59 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 42%|████▏     | 11/26 [00:01<00:02,  5.87it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  59\n",
      "Running randomized subset search for 200 iterations with 45 trajectories...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 26/26 [00:02<00:00, 10.46it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done. Length of longest subset:  45\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "clustering_after_HL, HL_table_dict = attack.build_clustering_after_HL_assignment(HL_table_trips_concat, full_trip_gdf, trip_concat_dict, full_trips_concat_gdf_overlap_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Clustering results after concatenation step:\n",
      "Number of unique clusters: 218\n",
      "Homogeneity: 1.000\n",
      "Completeness: 0.182\n",
      "V-measure: 0.308\n",
      "Rand index: 0.534\n",
      "ARI: 0.005\n",
      "MI: 0.963\n",
      "NMI: 0.308\n",
      "AMI: 0.081\n",
      "\n",
      "Clustering results after HL matching step:\n",
      "Number of unique clusters: 34\n",
      "Homogeneity: 0.974\n",
      "Completeness: 0.414\n",
      "V-measure: 0.581\n",
      "Rand index: 0.697\n",
      "ARI: 0.368\n",
      "MI: 0.938\n",
      "NMI: 0.581\n",
      "AMI: 0.541\n"
     ]
    }
   ],
   "source": [
    "print(\"Clustering results after concatenation step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_concat))}\")\n",
    "attack.evaluate(clustering_concat, full_trip_gdf)\n",
    "\n",
    "clustering_HL = list(dict(sorted(clustering_after_HL.items())).values())\n",
    "print(\"\\nClustering results after HL matching step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_HL))}\")\n",
    "attack.evaluate(clustering_HL, full_trip_gdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Try to Assign Trips Without Match"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Comparing trips that were not assigned to any HL_ID with trips that were assigned to a HL_ID...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 10%|█         | 1/10 [00:05<00:46,  5.11s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "no match and assign new cluster id 35 to trips 11 100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 20%|██        | 2/10 [00:08<00:30,  3.82s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LCSS scores too low to match\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 30%|███       | 3/10 [00:10<00:23,  3.35s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "existing match and assign cluster id 2.0 to trip 262\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 40%|████      | 4/10 [00:15<00:22,  3.78s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "existing match and assign cluster id 2.0 to trip 255\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 50%|█████     | 5/10 [00:19<00:19,  4.00s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "existing match and assign cluster id 6.0 to trip 253\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 60%|██████    | 6/10 [00:20<00:11,  2.94s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "existing match and assign cluster id 2.0 to trip 1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 70%|███████   | 7/10 [00:23<00:08,  2.89s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "existing match and assign cluster id 5.0 to trip 51\n",
      "There are two or more HL_IDs with the same LCSS score. Looking for second highest LCSS score...\n",
      "Candidate HL_IDs: [-1.0, 15.0, 17.0]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 80%|████████  | 8/10 [00:29<00:07,  3.81s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There is a unique highest score\n",
      "The highest score is 1.0 and the best_hl_id is 15.0\n",
      "existing match and assign cluster id 15.0 to trip 64\n",
      "There are two or more HL_IDs with the same LCSS score. Looking for second highest LCSS score...\n",
      "Candidate HL_IDs: [15.0, 18.0]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10/10 [00:30<00:00,  3.04s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There is a unique highest score\n",
      "The highest score is 1.0 and the best_hl_id is 15.0\n",
      "existing match and assign cluster id 15.0 to trip 101\n",
      "Done.\n",
      "Assigning clustering IDs to all trips that are part of a new cluster...\n",
      "Done.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "clustering_after_double_assign_HL = attack.assign_trips_without_match(\n",
    "    clustering_after_HL, HL_table_dict, \n",
    "    full_trips_concat_gdf, \n",
    "    full_trips_concat_gdf_overlap_dict, \n",
    "    full_trip_gdf, \n",
    "    trip_concat_dict)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Clustering results after concatenation step:\n",
      "Number of unique clusters: 218\n",
      "Homogeneity: 1.000\n",
      "Completeness: 0.182\n",
      "V-measure: 0.308\n",
      "Rand index: 0.534\n",
      "ARI: 0.005\n",
      "MI: 0.963\n",
      "NMI: 0.308\n",
      "AMI: 0.081\n",
      "\n",
      "Clustering results after HL matching step:\n",
      "Number of unique clusters: 34\n",
      "Homogeneity: 0.974\n",
      "Completeness: 0.414\n",
      "V-measure: 0.581\n",
      "Rand index: 0.697\n",
      "ARI: 0.368\n",
      "MI: 0.938\n",
      "NMI: 0.581\n",
      "AMI: 0.541\n",
      "\n",
      "Clustering results after double assign HL step:\n",
      "Number of unique clusters: 27\n",
      "Homogeneity: 0.950\n",
      "Completeness: 0.422\n",
      "V-measure: 0.584\n",
      "Rand index: 0.702\n",
      "ARI: 0.378\n",
      "MI: 0.915\n",
      "NMI: 0.584\n",
      "AMI: 0.551\n"
     ]
    }
   ],
   "source": [
    "print(\"Clustering results after concatenation step:\")\n",
    "print(f\"Number of unique clusters: {len(set(clustering_concat))}\")\n",
    "attack.evaluate(clustering_concat, full_trip_gdf)\n",
    "\n",
    "print(\"\\nClustering results after HL matching step:\")\n",
    "print(f\"Number of unique clusters: {len(set(list(dict(sorted(clustering_after_HL.items())).values())))}\")\n",
    "attack.evaluate(list(dict(sorted(clustering_after_HL.items())).values()), full_trip_gdf)\n",
    "\n",
    "print(\"\\nClustering results after double assign HL step:\")\n",
    "print(f\"Number of unique clusters: {len(set(list(dict(sorted(clustering_after_double_assign_HL.items())).values())))}\")\n",
    "attack.evaluate(list(dict(sorted(clustering_after_double_assign_HL.items())).values()), full_trip_gdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "attack.store_results(clustering_concat, list(dict(sorted(clustering_after_HL.items())).values()), list(dict(sorted(clustering_after_double_assign_HL.items())).values()), full_trip_gdf)"
   ]
  }
 ],
 "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": "3.8.10"
  },
  "vscode": {
   "interpreter": {
    "hash": "543daf12f525df94f20bbdd448da69881f98a71c963c44c9c7818e0113666227"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}