experiments/05_partial_coverage/08_comparison_for_full_bad.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from aemeasure import read_as_pd\n",
    "import os\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "from pcpptc import PolygonInstance, Solution\n",
    "from pcpptc.plot import setup_plot, plot_polygon_instance, plot_solution\n",
    "from pcpptc.plot.polgyon_instance import plot_coverage"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded dataframe ./solutions2/2.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:20:25.305058 and 2021-08-23 19:12:10.324509\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Index(['solution', 'coverage', 'touring_cost', 'length', 'turn_sum',\n",
       "       'instance', 'instance_path', 'runtime', 'timestamp', 'hostname', 'argv',\n",
       "       'git_revision', 'cwd', 'solver', 'i', 'turn_factor'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "read_as_pd(\"./solutions2/2.results.json\").columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded dataframe ./solutions2/75.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:15:25.537704 and 2021-08-23 21:18:47.477030\n",
      "Loaded dataframe ./solutions2/156.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 17:28:07.922508 and 2021-08-23 17:29:58.018417\n",
      "Loaded dataframe ./solutions2/119.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:52:15.380986 and 2021-08-23 22:01:16.874869\n",
      "Loaded dataframe ./solutions2/173.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:37:08.068632 and 2021-08-23 21:42:46.264003\n",
      "Loaded dataframe ./solutions2/144.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:24:14.480837 and 2021-08-23 19:20:55.900008\n",
      "Loaded dataframe ./solutions2/185.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:28:15.147155 and 2021-08-23 22:31:34.907581\n",
      "Loaded dataframe ./solutions2/154.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:02:03.554251 and 2021-08-23 20:07:16.064666\n",
      "Loaded dataframe ./solutions2/12.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:36:52.154269 and 2021-08-23 19:40:28.993302\n",
      "Loaded dataframe ./solutions2/197.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:34:45.942072 and 2021-08-23 20:42:25.333067\n",
      "Loaded dataframe ./solutions2/86.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:42:09.703926 and 2021-08-23 19:45:45.068338\n",
      "Loaded dataframe ./solutions2/35.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:37:00.667312 and 2021-08-23 19:40:09.452426\n",
      "Loaded dataframe ./solutions2/74.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:47:49.534665 and 2021-08-20 16:50:27.013633\n",
      "Loaded dataframe ./solutions2/57.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:56:16.011981 and 2021-08-23 20:17:07.604822\n",
      "Loaded dataframe ./solutions2/58.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:33:21.979894 and 2021-08-23 18:36:30.099678\n",
      "Loaded dataframe ./solutions2/14.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:19:27.086992 and 2021-08-23 21:20:50.540163\n",
      "Loaded dataframe ./solutions2/3.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:19:43.748533 and 2021-08-23 20:29:04.846946\n",
      "Loaded dataframe ./solutions2/36.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:11:22.124619 and 2021-08-23 22:57:51.655710\n",
      "Loaded dataframe ./solutions2/80.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 18:44:39.493872 and 2021-08-20 18:45:53.163592\n",
      "Loaded dataframe ./solutions2/70.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:25:49.512522 and 2021-08-23 21:27:20.240674\n",
      "Loaded dataframe ./solutions2/45.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:17:22.091808 and 2021-08-23 16:38:07.889739\n",
      "Loaded dataframe ./solutions2/21.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:18:08.357916 and 2021-08-23 18:25:17.330503\n",
      "Loaded dataframe ./solutions2/17.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:21:32.992932 and 2021-08-23 18:15:59.533005\n",
      "Loaded dataframe ./solutions2/125.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 19:02:16.988535 and 2021-08-23 19:36:36.891987\n",
      "Loaded dataframe ./solutions2/101.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 19:22:04.980766 and 2021-08-20 19:27:35.770493\n",
      "Loaded dataframe ./solutions2/16.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:50:55.021404 and 2021-08-23 18:54:48.679943\n",
      "Loaded dataframe ./solutions2/52.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:27:13.322992 and 2021-08-23 16:30:53.488865\n",
      "Loaded dataframe ./solutions2/167.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:38:15.076807 and 2021-08-23 21:40:29.709251\n",
      "Loaded dataframe ./solutions2/131.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 17:46:17.070324 and 2021-08-20 17:47:59.822763\n",
      "Loaded dataframe ./solutions2/149.results.json\n",
      "Executed on: ['algry01' 'algry03']\n",
      "During: 2021-08-20 16:43:15.272849 and 2021-08-20 17:40:04.669765\n",
      "Loaded dataframe ./solutions2/92.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:39:59.633943 and 2021-08-23 18:41:44.765395\n",
      "Loaded dataframe ./solutions2/84.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 18:52:27.440852 and 2021-08-20 18:55:05.782405\n",
      "Loaded dataframe ./solutions2/158.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 17:08:11.748723 and 2021-08-20 17:10:45.122446\n",
      "Loaded dataframe ./solutions2/41.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:15:50.957056 and 2021-08-23 20:18:24.425311\n",
      "Loaded dataframe ./solutions2/174.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:44:01.844831 and 2021-08-23 16:46:17.123745\n",
      "Loaded dataframe ./solutions2/191.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:09:03.719706 and 2021-08-23 18:13:47.844578\n",
      "Loaded dataframe ./solutions2/181.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:24:22.345111 and 2021-08-23 19:45:57.847490\n",
      "Loaded dataframe ./solutions2/29.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:57:59.424288 and 2021-08-23 17:27:17.642017\n",
      "Loaded dataframe ./solutions2/76.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:02:28.022335 and 2021-08-23 16:07:04.683651\n",
      "Loaded dataframe ./solutions2/130.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:21:25.513804 and 2021-08-23 21:23:08.647720\n",
      "Loaded dataframe ./solutions2/118.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:26:39.994349 and 2021-08-23 21:37:28.619093\n",
      "Loaded dataframe ./solutions2/38.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:44:34.613350 and 2021-08-23 18:49:21.084409\n",
      "Loaded dataframe ./solutions2/68.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:33:38.194874 and 2021-08-23 18:35:56.525867\n",
      "Loaded dataframe ./solutions2/143.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:35:41.159775 and 2021-08-23 20:49:17.004111\n",
      "Loaded dataframe ./solutions2/79.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:53:32.364195 and 2021-08-23 20:21:17.233596\n",
      "Loaded dataframe ./solutions2/5.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:16:23.118072 and 2021-08-23 19:17:15.220246\n",
      "Loaded dataframe ./solutions2/180.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:33:22.511448 and 2021-08-23 18:36:16.036021\n",
      "Loaded dataframe ./solutions2/99.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:47:50.346296 and 2021-08-23 19:52:11.979430\n",
      "Loaded dataframe ./solutions2/161.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:36:11.195643 and 2021-08-23 16:59:24.562796\n",
      "Loaded dataframe ./solutions2/136.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:17:20.138138 and 2021-08-23 19:05:22.555571\n",
      "Loaded dataframe ./solutions2/115.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:46:23.200512 and 2021-08-23 18:17:53.050861\n",
      "Loaded dataframe ./solutions2/153.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:09:50.321031 and 2021-08-23 22:15:08.339310\n",
      "Loaded dataframe ./solutions2/11.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:59:17.848925 and 2021-08-23 16:26:18.121020\n",
      "Loaded dataframe ./solutions2/32.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:23:07.595479 and 2021-08-23 20:26:41.700211\n",
      "Loaded dataframe ./solutions2/121.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:40:02.330201 and 2021-08-23 18:59:18.994265\n",
      "Loaded dataframe ./solutions2/189.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:45:16.631361 and 2021-08-23 20:51:09.860320\n",
      "Loaded dataframe ./solutions2/50.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:15:21.621057 and 2021-08-23 19:22:26.810482\n",
      "Loaded dataframe ./solutions2/1.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:50:26.470319 and 2021-08-23 16:15:52.415912\n",
      "Loaded dataframe ./solutions2/193.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:29:18.077547 and 2021-08-23 17:09:09.819558\n",
      "Loaded dataframe ./solutions2/96.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:06:00.695854 and 2021-08-23 18:07:02.930948\n",
      "Loaded dataframe ./solutions2/49.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:58:54.774254 and 2021-08-23 20:06:08.260390\n",
      "Loaded dataframe ./solutions2/9.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:38:01.778960 and 2021-08-23 18:42:59.932451\n",
      "Loaded dataframe ./solutions2/48.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:15:11.615832 and 2021-08-23 22:50:05.414743\n",
      "Loaded dataframe ./solutions2/129.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:52:27.176648 and 2021-08-23 22:04:58.217251\n",
      "Loaded dataframe ./solutions2/146.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:05:25.458466 and 2021-08-23 19:07:25.125888\n",
      "Loaded dataframe ./solutions2/127.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:35:34.227338 and 2021-08-23 19:39:36.358776\n",
      "Loaded dataframe ./solutions2/44.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:03:51.040406 and 2021-08-23 19:08:37.802046\n",
      "Loaded dataframe ./solutions2/194.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 18:46:56.286461 and 2021-08-20 18:49:42.706414\n",
      "Loaded dataframe ./solutions2/145.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:06:03.869846 and 2021-08-23 17:58:19.994666\n",
      "Loaded dataframe ./solutions2/53.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:20:38.705735 and 2021-08-23 19:04:39.145827\n",
      "Loaded dataframe ./solutions2/169.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:41:52.291701 and 2021-08-23 21:44:58.636043\n",
      "Loaded dataframe ./solutions2/132.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:17:49.329544 and 2021-08-23 18:39:08.586177\n",
      "Loaded dataframe ./solutions2/188.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 17:49:58.648823 and 2021-08-23 17:57:11.554844\n",
      "Loaded dataframe ./solutions2/46.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:02:37.529453 and 2021-08-23 18:32:23.816828\n",
      "Loaded dataframe ./solutions2/13.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:55:22.132204 and 2021-08-23 18:31:59.196441\n",
      "Loaded dataframe ./solutions2/114.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:42:01.802105 and 2021-08-23 18:28:35.344592\n",
      "Loaded dataframe ./solutions2/104.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:44:16.755613 and 2021-08-23 21:21:30.702696\n",
      "Loaded dataframe ./solutions2/59.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:18:49.187259 and 2021-08-23 19:21:35.776164\n",
      "Loaded dataframe ./solutions2/71.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:10:30.130406 and 2021-08-23 19:15:58.279210\n",
      "Loaded dataframe ./solutions2/7.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:29:13.151121 and 2021-08-23 17:18:24.305732\n",
      "Loaded dataframe ./solutions2/122.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:05:59.037825 and 2021-08-23 22:07:26.284368\n",
      "Loaded dataframe ./solutions2/170.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:27:56.811745 and 2021-08-23 19:33:29.918465\n",
      "Loaded dataframe ./solutions2/47.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:11:56.533895 and 2021-08-23 19:14:06.527951\n",
      "Loaded dataframe ./solutions2/179.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:11:58.801842 and 2021-08-23 18:25:34.107025\n",
      "Loaded dataframe ./solutions2/168.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:18:11.995701 and 2021-08-23 21:33:54.556038\n",
      "Loaded dataframe ./solutions2/126.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:56:10.058147 and 2021-08-23 19:02:01.983700\n",
      "Loaded dataframe ./solutions2/135.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:23:04.826596 and 2021-08-23 19:24:16.443474\n",
      "Loaded dataframe ./solutions2/134.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:16:33.965327 and 2021-08-23 22:20:02.527343\n",
      "Loaded dataframe ./solutions2/199.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:45:40.931323 and 2021-08-23 19:19:38.502295\n",
      "Loaded dataframe ./solutions2/62.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:44:32.297222 and 2021-08-23 19:52:46.809879\n",
      "Loaded dataframe ./solutions2/83.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:34:10.064331 and 2021-08-23 21:34:43.991960\n",
      "Loaded dataframe ./solutions2/6.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:00:59.407839 and 2021-08-20 16:56:30.452004\n",
      "Loaded dataframe ./solutions2/64.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:01:46.359671 and 2021-08-23 21:04:28.045494\n",
      "Loaded dataframe ./solutions2/22.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:23:51.345346 and 2021-08-23 19:01:57.951223\n",
      "Loaded dataframe ./solutions2/66.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:47:37.797617 and 2021-08-20 16:50:39.230309\n",
      "Loaded dataframe ./solutions2/94.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:44:13.271820 and 2021-08-23 21:47:47.056302\n",
      "Loaded dataframe ./solutions2/54.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:45:42.067282 and 2021-08-23 20:54:39.279994\n",
      "Loaded dataframe ./solutions2/163.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:37:33.119067 and 2021-08-23 22:48:22.171488\n",
      "Loaded dataframe ./solutions2/160.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:59:53.657485 and 2021-08-23 19:11:15.309437\n",
      "Loaded dataframe ./solutions2/186.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:05:15.712092 and 2021-08-23 16:15:43.498662\n",
      "Loaded dataframe ./solutions2/137.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:01:27.701786 and 2021-08-23 18:05:35.422878\n",
      "Loaded dataframe ./solutions2/0.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:39:46.889203 and 2021-08-23 22:06:29.244755\n",
      "Loaded dataframe ./solutions2/97.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 17:52:25.716469 and 2021-08-23 17:55:49.079121\n",
      "Loaded dataframe ./solutions2/40.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:10:01.043830 and 2021-08-23 16:13:30.412035\n",
      "Loaded dataframe ./solutions2/88.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 17:52:35.993928 and 2021-08-23 17:57:48.380286\n",
      "Loaded dataframe ./solutions2/61.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:20:25.176984 and 2021-08-23 18:32:05.630278\n",
      "Loaded dataframe ./solutions2/196.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:32:15.209348 and 2021-08-23 17:05:11.199271\n",
      "Loaded dataframe ./solutions2/93.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:20:47.001892 and 2021-08-23 16:25:35.112638\n",
      "Loaded dataframe ./solutions2/39.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:34:21.882336 and 2021-08-23 22:36:59.203438\n",
      "Loaded dataframe ./solutions2/73.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:51:13.893002 and 2021-08-23 20:57:11.383797\n",
      "Loaded dataframe ./solutions2/91.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 17:49:51.831761 and 2021-08-20 17:54:26.932920\n",
      "Loaded dataframe ./solutions2/112.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:26:06.610536 and 2021-08-23 19:42:18.861783\n",
      "Loaded dataframe ./solutions2/142.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:45:20.978224 and 2021-08-23 18:50:26.102059\n",
      "Loaded dataframe ./solutions2/110.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:28:33.224006 and 2021-08-20 17:40:00.374577\n",
      "Loaded dataframe ./solutions2/10.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:34:31.608811 and 2021-08-23 16:39:37.012711\n",
      "Loaded dataframe ./solutions2/28.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:31:06.418676 and 2021-08-23 19:35:43.964114\n",
      "Loaded dataframe ./solutions2/37.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:26:55.774879 and 2021-08-23 16:28:23.743005\n",
      "Loaded dataframe ./solutions2/65.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:18:40.768757 and 2021-08-23 20:20:52.366018\n",
      "Loaded dataframe ./solutions2/133.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:00:05.237401 and 2021-08-23 18:06:43.448023\n",
      "Loaded dataframe ./solutions2/77.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:31:29.848106 and 2021-08-23 18:24:47.305365\n",
      "Loaded dataframe ./solutions2/183.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:55:18.636756 and 2021-08-23 18:58:08.832780\n",
      "Loaded dataframe ./solutions2/98.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:47:44.619404 and 2021-08-23 19:29:12.985996\n",
      "Loaded dataframe ./solutions2/128.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:20:57.620133 and 2021-08-23 19:28:51.577672\n",
      "Loaded dataframe ./solutions2/107.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 17:58:38.700383 and 2021-08-23 18:00:11.489838\n",
      "Loaded dataframe ./solutions2/24.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 17:28:42.639737 and 2021-08-23 17:34:54.448892\n",
      "Loaded dataframe ./solutions2/166.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:14:10.200129 and 2021-08-23 22:28:16.307899\n",
      "Loaded dataframe ./solutions2/85.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:37:15.121567 and 2021-08-23 17:59:25.618313\n",
      "Loaded dataframe ./solutions2/81.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:33:22.196162 and 2021-08-23 16:50:29.847333\n",
      "Loaded dataframe ./solutions2/90.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:56:40.885052 and 2021-08-23 18:19:45.624510\n",
      "Loaded dataframe ./solutions2/139.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:02:13.912425 and 2021-08-23 16:06:35.978724\n",
      "Loaded dataframe ./solutions2/78.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:17:23.129133 and 2021-08-23 17:06:43.182533\n",
      "Loaded dataframe ./solutions2/165.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:27:34.222848 and 2021-08-23 21:31:02.461878\n",
      "Loaded dataframe ./solutions2/164.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:21:23.850885 and 2021-08-23 20:26:02.284407\n",
      "Loaded dataframe ./solutions2/87.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 18:04:24.467820 and 2021-08-20 18:44:01.956485\n",
      "Loaded dataframe ./solutions2/116.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 18:50:03.596539 and 2021-08-20 18:50:44.509964\n",
      "Loaded dataframe ./solutions2/141.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:07:22.470693 and 2021-08-23 17:14:05.021403\n",
      "Loaded dataframe ./solutions2/182.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:07:07.658722 and 2021-08-23 16:08:21.109020\n",
      "Loaded dataframe ./solutions2/60.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:53:39.939672 and 2021-08-23 19:55:49.059594\n",
      "Loaded dataframe ./solutions2/184.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:43:14.785108 and 2021-08-23 19:44:52.721466\n",
      "Loaded dataframe ./solutions2/171.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:14:53.621460 and 2021-08-23 20:17:44.996640\n",
      "Loaded dataframe ./solutions2/100.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:07:46.091829 and 2021-08-23 20:13:26.039107\n",
      "Loaded dataframe ./solutions2/2.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:20:25.305058 and 2021-08-23 19:12:10.324509\n",
      "Loaded dataframe ./solutions2/178.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:24:04.779643 and 2021-08-23 16:32:41.377550\n",
      "Loaded dataframe ./solutions2/69.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:39:55.096390 and 2021-08-23 16:18:40.936128\n",
      "Loaded dataframe ./solutions2/176.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 17:35:56.467811 and 2021-08-23 18:29:00.983390\n",
      "Loaded dataframe ./solutions2/102.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:25:30.490014 and 2021-08-23 18:43:07.685899\n",
      "Loaded dataframe ./solutions2/67.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:48:19.486673 and 2021-08-23 18:11:31.189733\n",
      "Loaded dataframe ./solutions2/148.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 15:55:47.608572 and 2021-08-23 18:37:43.839439\n",
      "Loaded dataframe ./solutions2/140.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:40:38.010294 and 2021-08-23 22:45:48.208942\n",
      "Loaded dataframe ./solutions2/151.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:27:21.684082 and 2021-08-23 19:30:40.861690\n",
      "Loaded dataframe ./solutions2/4.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:32:14.381140 and 2021-08-23 22:33:22.402613\n",
      "Loaded dataframe ./solutions2/123.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:29:28.465216 and 2021-08-23 19:24:01.389110\n",
      "Loaded dataframe ./solutions2/157.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:30:48.605371 and 2021-08-23 16:35:45.451485\n",
      "Loaded dataframe ./solutions2/55.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 17:15:00.704133 and 2021-08-23 17:17:05.432893\n",
      "Loaded dataframe ./solutions2/192.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:59:38.404351 and 2021-08-23 19:03:40.744534\n",
      "Loaded dataframe ./solutions2/195.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:36:27.687629 and 2021-08-23 20:19:34.061564\n",
      "Loaded dataframe ./solutions2/26.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:24:30.799419 and 2021-08-23 19:30:21.283666\n",
      "Loaded dataframe ./solutions2/124.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:05:43.263250 and 2021-08-23 18:11:13.511560\n",
      "Loaded dataframe ./solutions2/198.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:22:14.934079 and 2021-08-23 18:53:00.982210\n",
      "Loaded dataframe ./solutions2/147.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:23:22.328241 and 2021-08-23 20:28:50.873113\n",
      "Loaded dataframe ./solutions2/111.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:17:12.473838 and 2021-08-23 16:20:54.508401\n",
      "Loaded dataframe ./solutions2/63.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:28:28.265093 and 2021-08-23 20:33:50.957017\n",
      "Loaded dataframe ./solutions2/82.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 17:12:20.791099 and 2021-08-20 17:15:15.332001\n",
      "Loaded dataframe ./solutions2/150.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:39:21.685288 and 2021-08-23 19:40:22.616411\n",
      "Loaded dataframe ./solutions2/187.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:27:45.199901 and 2021-08-23 21:25:11.870245\n",
      "Loaded dataframe ./solutions2/72.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:52:44.989772 and 2021-08-23 16:57:13.030748\n",
      "Loaded dataframe ./solutions2/31.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 17:01:50.990330 and 2021-08-20 17:07:00.347967\n",
      "Loaded dataframe ./solutions2/25.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 17:41:53.747181 and 2021-08-20 17:45:33.840289\n",
      "Loaded dataframe ./solutions2/34.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:32:30.360091 and 2021-08-20 17:19:20.964770\n",
      "Loaded dataframe ./solutions2/108.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:37:09.076313 and 2021-08-23 18:38:38.347915\n",
      "Loaded dataframe ./solutions2/105.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:45:32.686242 and 2021-08-23 18:53:34.196936\n",
      "Loaded dataframe ./solutions2/42.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:30:23.045839 and 2021-08-23 17:25:40.668203\n",
      "Loaded dataframe ./solutions2/155.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 22:32:06.389609 and 2021-08-23 22:33:16.305571\n",
      "Loaded dataframe ./solutions2/138.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:23:39.852856 and 2021-08-23 19:38:43.156660\n",
      "Loaded dataframe ./solutions2/51.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:07:43.425187 and 2021-08-23 16:09:24.606398\n",
      "Loaded dataframe ./solutions2/95.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:12:34.662668 and 2021-08-23 22:31:44.436253\n",
      "Loaded dataframe ./solutions2/8.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 18:12:40.245905 and 2021-08-23 19:25:44.352784\n",
      "Loaded dataframe ./solutions2/103.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 21:00:35.706847 and 2021-08-23 21:20:36.935339\n",
      "Loaded dataframe ./solutions2/43.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:01:19.566129 and 2021-08-23 18:09:44.385057\n",
      "Loaded dataframe ./solutions2/177.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:33:46.310079 and 2021-08-23 20:29:29.036763\n",
      "Loaded dataframe ./solutions2/33.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:24:14.799366 and 2021-08-23 19:33:28.682057\n",
      "Loaded dataframe ./solutions2/15.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:40:48.961284 and 2021-08-23 16:43:13.976993\n",
      "Loaded dataframe ./solutions2/23.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 19:28:54.153733 and 2021-08-23 18:54:55.343603\n",
      "Loaded dataframe ./solutions2/109.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:57:05.383318 and 2021-08-20 17:29:03.637288\n",
      "Loaded dataframe ./solutions2/89.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:30:36.733103 and 2021-08-23 18:33:41.412023\n",
      "Loaded dataframe ./solutions2/56.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:51:19.689657 and 2021-08-20 16:54:02.499206\n",
      "Loaded dataframe ./solutions2/30.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 20:57:03.194743 and 2021-08-23 21:14:09.272694\n",
      "Loaded dataframe ./solutions2/117.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:55:44.470288 and 2021-08-20 16:59:19.043933\n",
      "Loaded dataframe ./solutions2/175.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 16:56:02.084711 and 2021-08-23 19:59:57.196119\n",
      "Loaded dataframe ./solutions2/120.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:42:39.119396 and 2021-08-20 16:46:33.033414\n",
      "Loaded dataframe ./solutions2/20.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 19:16:36.052821 and 2021-08-23 19:20:34.362989\n",
      "Loaded dataframe ./solutions2/113.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 19:01:17.234497 and 2021-08-20 19:18:46.029878\n",
      "Loaded dataframe ./solutions2/19.results.json\n",
      "Executed on: ['algry03' 'algry01']\n",
      "During: 2021-08-20 17:10:25.396870 and 2021-08-23 19:17:42.030101\n",
      "Loaded dataframe ./solutions2/106.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 16:14:46.160894 and 2021-08-23 16:18:39.451582\n",
      "Loaded dataframe ./solutions2/172.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 17:30:20.386364 and 2021-08-20 17:33:11.936330\n",
      "Loaded dataframe ./solutions2/18.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:42:07.856067 and 2021-08-23 18:43:00.442022\n",
      "Loaded dataframe ./solutions2/27.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:40:41.509864 and 2021-08-23 18:02:27.592707\n",
      "Loaded dataframe ./solutions2/152.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:21:35.402078 and 2021-08-23 21:00:34.048130\n",
      "Loaded dataframe ./solutions2/159.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-20 16:42:22.221293 and 2021-08-20 16:46:08.741734\n",
      "Loaded dataframe ./solutions2/190.results.json\n",
      "Executed on: ['algry01']\n",
      "During: 2021-08-23 18:39:43.430237 and 2021-08-23 18:50:57.915658\n"
     ]
    }
   ],
   "source": [
    "tables = []\n",
    "for f in os.listdir(\"./solutions2\"):\n",
    "    if not f.endswith(\".results.json\"):\n",
    "        continue\n",
    "    f = os.path.join(\"./solutions2\", f)\n",
    "    tables.append(read_as_pd(f))\n",
    "\n",
    "data = pd.concat(tables, ignore_index=True)\n",
    "data.groupby([\"solver\"])[\"instance\"].nunique()\n",
    "bad_instance = \"./instances2/122.instance.json\"\n",
    "data = data[data[\"instance_path\"] == bad_instance]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>solution</th>\n",
       "      <th>coverage</th>\n",
       "      <th>touring_cost</th>\n",
       "      <th>length</th>\n",
       "      <th>turn_sum</th>\n",
       "      <th>instance</th>\n",
       "      <th>instance_path</th>\n",
       "      <th>runtime</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>hostname</th>\n",
       "      <th>argv</th>\n",
       "      <th>git_revision</th>\n",
       "      <th>cwd</th>\n",
       "      <th>solver</th>\n",
       "      <th>i</th>\n",
       "      <th>turn_factor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [solution, coverage, touring_cost, length, turn_sum, instance, instance_path, runtime, timestamp, hostname, argv, git_revision, cwd, solver, i, turn_factor]\n",
       "Index: []"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data.isna().any(axis=1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "solver\n",
       "MeshAlgorithm(0.95, full_coverage=False)                                                                             1\n",
       "MeshAlgorithm(0.95, full_coverage=True)                                                                              1\n",
       "RotatingHexagonalAlgorithm(RotatingRegularHexagonal(full_coverage=False, point_based=False, with_boundary=False))    1\n",
       "RotatingHexagonalAlgorithm(RotatingRegularHexagonal(full_coverage=True, point_based=False, with_boundary=False))     1\n",
       "Name: instance, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.dropna(inplace=True)\n",
    "data = data[data[\"solver\"] != \"MeshAlgorithm(0.95)\"]\n",
    "data.groupby([\"solver\"])[\"instance\"].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "from optimizer.instance_converter.dmsh import GmshGrid\n",
    "\n",
    "\n",
    "def rename(s):\n",
    "    names = {\n",
    "        \"RotatingHexagonalAlgorithm(RotatingRegularHexagonal(full_coverage=False, point_based=False, with_boundary=False))\": \"Regular (Partial)\",\n",
    "        \"RotatingHexagonalAlgorithm(RotatingRegularHexagonal(full_coverage=True, point_based=False, with_boundary=False))\": \"Regular (Full)\",\n",
    "        \"MeshAlgorithm(0.95, full_coverage=False)\": \"Mesh (Partial)\",\n",
    "        \"MeshAlgorithm(0.95, full_coverage=True)\": \"Mesh (Full)\",\n",
    "    }\n",
    "    return names.get(s, s)\n",
    "\n",
    "\n",
    "data[\"solver\"] = data[\"solver\"].apply(rename)\n",
    "data = data[data[\"solver\"].apply(lambda s: \"Full\" in s)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "data[\"type\"] = data[\"solver\"].apply(lambda s: \"hex\" if \"Hex\" in s else \"square\")\n",
    "data[\"point_based\"] = data[\"solver\"].apply(lambda s: \"pb=True\" in s)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "instance_paths = list(data[\"instance_path\"].unique())\n",
    "instances = {f: PolygonInstance.from_json(file_path=f) for f in instance_paths}\n",
    "instance_areas = pd.DataFrame(\n",
    "    data={\"area\": [i.original_area.area for i in instances.values()]},\n",
    "    index=instance_paths,\n",
    ")\n",
    "data = data.merge(instance_areas, left_on=\"instance_path\", right_index=True)\n",
    "\n",
    "\n",
    "def compute_objective(row):\n",
    "    instance = instances[row[\"instance_path\"]]\n",
    "    solution = Solution.from_json(data=row[\"solution\"])\n",
    "    instance: PolygonInstance\n",
    "    return instance.compute_touring_cost(\n",
    "        solution.waypoints\n",
    "    ) + instance.compute_missed_covering_value(solution.waypoints)\n",
    "\n",
    "\n",
    "data[\"Obj\"] = data.apply(compute_objective, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "t_ = data.groupby(\"instance\")[[\"touring_cost\"]].min()\n",
    "data = data.merge(t_, left_on=\"instance\", right_index=True, suffixes=(\"\", \"_min\"))\n",
    "data[\"relative_cost\"] = 100 * ((data[\"touring_cost\"] / data[\"touring_cost_min\"]) - 1)\n",
    "data[\"relative_coverage\"] = 100 * data[\"coverage\"] / data[\"area\"]\n",
    "data[\"type\"] = data[\"solver\"].apply(lambda s: s[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "t_ = data.groupby(\"instance\")[[\"Obj\"]].min()\n",
    "data = data.merge(t_, left_on=\"instance\", right_index=True, suffixes=(\"\", \"_min\"))\n",
    "data[\"relative_obj\"] = 100 * ((data[\"Obj\"] / data[\"Obj_min\"]) - 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [],
   "source": [
    "data.sort_values([\"solver\"], inplace=True, ascending=True)\n",
    "data.sort_values([\"type\"], inplace=True, ascending=True, kind=\"stable\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "solver\n",
       "Mesh (Full)       1\n",
       "Regular (Full)    1\n",
       "Name: instance, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t_ = data.dropna()\n",
    "t_ = t_.groupby([\"solver\"])[\"instance\"].nunique()\n",
    "t_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mesh (Full) & 60.8 \\\\\n",
      "Regular (Full) & 13.2 \\\\\n"
     ]
    }
   ],
   "source": [
    "t = data.groupby([\"solver\"])[\"runtime\"].mean().reset_index()\n",
    "t[\"solver\"] = t[\"solver\"].apply(lambda s: s.replace(\"□\", \"$\\\\square$\"))\n",
    "t[\"solver\"] = t[\"solver\"].apply(lambda s: s.replace(\"△\", \"$\\\\Delta$\"))\n",
    "t[\"solver\"] = t[\"solver\"].apply(lambda s: s.replace(\"%\", \"\\\\%\"))\n",
    "for i, t in t.iterrows():\n",
    "    print(t[\"solver\"], \"&\", round(t[\"runtime\"], 1), \"\\\\\\\\\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "pycharm": {
     "name": "#%%\n"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Computing coverage polygon\n",
      "Plotting coverage polygon\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 252x180 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Computing coverage polygon\n",
      "Plotting coverage polygon\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 252x180 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot(row, path=None, ax=None):\n",
    "    instance_path = row[\"instance_path\"]\n",
    "    instance = PolygonInstance.from_json(file_path=instance_path)\n",
    "    solution = Solution.from_json(data=row[\"solution\"])\n",
    "    if not ax:\n",
    "        ax = setup_plot(figsize=(3.5, 2.5))\n",
    "    plot_polygon_instance(ax, instance, set_limits=True, plot_valuable_areas=False)\n",
    "    plot_solution(ax, solution, color=\"blue\")\n",
    "    plot_coverage(ax, instance, solution)\n",
    "    ax.set_title(f\"Coverage: {round(row['relative_coverage'],1)}%\")\n",
    "    if path:\n",
    "        plt.tight_layout()\n",
    "        plt.savefig(path)\n",
    "\n",
    "\n",
    "for i, row in data[data[\"instance_path\"] == bad_instance].iterrows():\n",
    "    plot(row)\n",
    "    plt.tight_layout()\n",
    "    plt.savefig(f\"08_bad_coverage_{i}.pdf\")\n",
    "    plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}