Final_experiment_VND.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Final Experiment for VND"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "from os import listdir\n",
    "from os.path import isfile, join\n",
    "import string\n",
    "\n",
    "import numpy as np\n",
    "from plot_utils import plot_two_solutions\n",
    "import matplotlib.pyplot as plt \n",
    "\n",
    "from ioutils import parse_input\n",
    "from ioutils import parse_solutions\n",
    "from utils import get_distance_matrix\n",
    "from utils import get_discount_matrix\n",
    "from utils import allocate_paths\n",
    "from utils import get_nodes\n",
    "from utils import get_total_cost\n",
    "from utils import get_flow_from_paths\n",
    "import cplex_utils\n",
    "import graph_utils as graph_utils\n",
    "\n",
    "from ioutils import get_comparison_table_file_name\n",
    "from utils import get_latest_commit_id\n",
    "from utils import get_comparison_table\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "DATASET = 'AP'\n",
    "INPUT_DIRECTORY = \"./data/AP/generated/\"\n",
    "SOLUTIONS_FILE ='/home/ubuntu/VNS_for_UMApHMP/data/AP/Solutions-UMApHMP.txt'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from vns import get_initial_solution_random\n",
    "from vns import get_initial_solution_robust\n",
    "from vns import get_initial_solution_circle\n",
    "\n",
    "from vns import local_search_best_improvement\n",
    "from vns import local_search_first_improvement\n",
    "\n",
    "\n",
    "from vns import basic_VNS\n",
    "from vns import deterministic_VNS\n",
    "\n",
    "from utils import NEIGHBOURHOOD_TYPES\n",
    "\n",
    "from ioutils import get_comparison_table_file_name\n",
    "from utils import get_latest_commit_id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "NUMBER_OF_PROBLEMS = 45\n",
    "OUTPUT_FILE_NAME = get_comparison_table_file_name('final_comparison_vnd', str(NUMBER_OF_PROBLEMS), get_latest_commit_id())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def best_ls(solution, neighbourhood_type, use_c):\n",
    "    return local_search_best_improvement(solution, neighbourhood_type, use_c)\n",
    "\n",
    "def first_ls(solution, neighbourhood_type, use_c):\n",
    "    return local_search_first_improvement(solution, neighbourhood_type, use_c)\n",
    "\n",
    "def random_init(n, p, distances, nodes_coor):\n",
    "    return get_initial_solution_random(n, p, distances, nodes_coor)\n",
    "\n",
    "def min_max_init(n, p, distances, nodes_coor):\n",
    "    return get_initial_solution_robust(n, p, distances, nodes_coor)\n",
    "\n",
    "def angle_init(n, p, distances, nodes_coor):\n",
    "    return get_initial_solution_circle(n, p, distances, nodes_coor)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ------- algorithm parameters --------\n",
    "# shaking_intensity = [0, 0.5]\n",
    "initializations = [random_init, min_max_init, angle_init]\n",
    "local_searches = [best_ls, first_ls]\n",
    "max_iters = [10]\n",
    "kmax_options = [1,2,3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "methods = []\n",
    "names = []\n",
    "\n",
    "for ls in local_searches:\n",
    "    for kmax in kmax_options:\n",
    "        for max_iter in max_iters:\n",
    "            methods.append(lambda problem, initial_solution: deterministic_VNS(problem, initial_solution=initial_solution, local_search=ls, neighbourhood_types=NEIGHBOURHOOD_TYPES[:kmax], max_iter=max_iter))\n",
    "            names.append(\" \".join([\"VND\", ls.__name__, \"kmax=\"+str(kmax), \"max_iter=\"+str(max_iter)]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['VND best_ls kmax=1 max_iter=10',\n",
       " 'VND best_ls kmax=2 max_iter=10',\n",
       " 'VND best_ls kmax=3 max_iter=10',\n",
       " 'VND first_ls kmax=1 max_iter=10',\n",
       " 'VND first_ls kmax=2 max_iter=10']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "names[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/26 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 26/26 [21:39<00:00, 49.96s/it] \n"
     ]
    }
   ],
   "source": [
    "comparison_table = get_comparison_table(\n",
    "    methods,\n",
    "    names,\n",
    "    DATASET, INPUT_DIRECTORY, SOLUTIONS_FILE,\n",
    "    from_problem=19,\n",
    "    to_problems=NUMBER_OF_PROBLEMS,\n",
    "    initializations=initializations,\n",
    "    display_nan=False)\n",
    "comparison_table.to_csv(OUTPUT_FILE_NAME, index_label='problem')\n",
    "# comparison_table = pd.read_csv(\"output/basic_VNS_final_comparison_vnd_45_d42a72836976d0d8abf612afa150b58ce819c49f.csv\", index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "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>VND best_ls kmax=1 max_iter=10 random_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=1 max_iter=10 random_init - time (s)</th>\n",
       "      <th>VND best_ls kmax=1 max_iter=10 min_max_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=1 max_iter=10 min_max_init - time (s)</th>\n",
       "      <th>VND best_ls kmax=1 max_iter=10 angle_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=1 max_iter=10 angle_init - time (s)</th>\n",
       "      <th>VND best_ls kmax=2 max_iter=10 random_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=2 max_iter=10 random_init - time (s)</th>\n",
       "      <th>VND best_ls kmax=2 max_iter=10 min_max_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=2 max_iter=10 min_max_init - time (s)</th>\n",
       "      <th>...</th>\n",
       "      <th>VND first_ls kmax=2 max_iter=10 min_max_init - deviation (%)</th>\n",
       "      <th>VND first_ls kmax=2 max_iter=10 min_max_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=2 max_iter=10 angle_init - deviation (%)</th>\n",
       "      <th>VND first_ls kmax=2 max_iter=10 angle_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 random_init - deviation (%)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 random_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 min_max_init - deviation (%)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 min_max_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 angle_init - deviation (%)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 angle_init - time (s)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40.2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.21</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.21000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.2000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.2000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.20000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.19000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.3</th>\n",
       "      <td>2.410000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>2.4100</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>2.410000</td>\n",
       "      <td>0.30</td>\n",
       "      <td>2.41000</td>\n",
       "      <td>0.31000</td>\n",
       "      <td>2.4100</td>\n",
       "      <td>0.3000</td>\n",
       "      <td>...</td>\n",
       "      <td>2.4100</td>\n",
       "      <td>0.3100</td>\n",
       "      <td>2.41000</td>\n",
       "      <td>0.32000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>2.4100</td>\n",
       "      <td>0.310000</td>\n",
       "      <td>2.41000</td>\n",
       "      <td>0.33000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.4</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.470000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.480000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.48</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.48000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.4800</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.4800</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.51000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.430000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.480000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.45000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.5</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.610000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.60000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.6000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.6000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.65000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.610000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.61000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.10</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.340000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.510000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.45</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>1.53000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.5100</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.5100</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>1.68000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.510000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>1.39000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.360000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.350000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.35000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.3500</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.3500</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.35000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.350000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.340000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.34000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.3</th>\n",
       "      <td>2.220000</td>\n",
       "      <td>0.530000</td>\n",
       "      <td>2.2200</td>\n",
       "      <td>0.530000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.66000</td>\n",
       "      <td>2.2200</td>\n",
       "      <td>0.5400</td>\n",
       "      <td>...</td>\n",
       "      <td>2.2200</td>\n",
       "      <td>0.5300</td>\n",
       "      <td>2.22000</td>\n",
       "      <td>0.54000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.640000</td>\n",
       "      <td>2.2200</td>\n",
       "      <td>0.540000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.63000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.4</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.860000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.970000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.83000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.9600</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.9600</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.89000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.820000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.960000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.82000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.5</th>\n",
       "      <td>0.070000</td>\n",
       "      <td>1.250000</td>\n",
       "      <td>0.0700</td>\n",
       "      <td>1.140000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.18</td>\n",
       "      <td>0.07000</td>\n",
       "      <td>1.12000</td>\n",
       "      <td>0.0700</td>\n",
       "      <td>1.1600</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0700</td>\n",
       "      <td>1.1500</td>\n",
       "      <td>0.07000</td>\n",
       "      <td>1.18000</td>\n",
       "      <td>0.070000</td>\n",
       "      <td>1.110000</td>\n",
       "      <td>0.0700</td>\n",
       "      <td>1.140000</td>\n",
       "      <td>0.07000</td>\n",
       "      <td>1.10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.10</th>\n",
       "      <td>0.830000</td>\n",
       "      <td>3.030000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.920000</td>\n",
       "      <td>0.140000</td>\n",
       "      <td>2.81</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>2.96000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.9200</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.8900</td>\n",
       "      <td>0.14000</td>\n",
       "      <td>2.56000</td>\n",
       "      <td>0.140000</td>\n",
       "      <td>2.780000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.900000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>2.88000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.210000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.260000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.27</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>2.12000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.2600</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.2600</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>2.08000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.020000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>2.260000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>2.10000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.3</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>3.680000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.07</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>3.90000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>3.6700</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>3.6600</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>3.80000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.090000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>3.660000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>3.70000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.4</th>\n",
       "      <td>0.060000</td>\n",
       "      <td>6.330000</td>\n",
       "      <td>0.0600</td>\n",
       "      <td>5.900000</td>\n",
       "      <td>0.060000</td>\n",
       "      <td>5.81</td>\n",
       "      <td>0.06000</td>\n",
       "      <td>5.58000</td>\n",
       "      <td>0.0600</td>\n",
       "      <td>5.9100</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0600</td>\n",
       "      <td>5.9000</td>\n",
       "      <td>0.06000</td>\n",
       "      <td>5.86000</td>\n",
       "      <td>0.060000</td>\n",
       "      <td>6.010000</td>\n",
       "      <td>0.0600</td>\n",
       "      <td>5.920000</td>\n",
       "      <td>0.06000</td>\n",
       "      <td>6.05000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.5</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.600000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>7.820000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.85</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>8.16000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>7.8100</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>7.8200</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>7.90000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.410000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>7.790000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>7.50000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200.2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>15.040000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>15.460000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.84</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>14.92000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>15.4600</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>15.4900</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>14.63000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>15.210000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>15.480000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>14.99000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200.3</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.230000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>28.290000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>27.13</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>29.81000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>28.3100</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>28.2500</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>27.87000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>26.470000</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>28.290000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>26.74000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.349375</td>\n",
       "      <td>4.680625</td>\n",
       "      <td>0.2975</td>\n",
       "      <td>4.525625</td>\n",
       "      <td>0.163125</td>\n",
       "      <td>4.42</td>\n",
       "      <td>0.15875</td>\n",
       "      <td>4.59625</td>\n",
       "      <td>0.2975</td>\n",
       "      <td>4.5275</td>\n",
       "      <td>...</td>\n",
       "      <td>0.2975</td>\n",
       "      <td>4.5225</td>\n",
       "      <td>0.30625</td>\n",
       "      <td>4.43875</td>\n",
       "      <td>0.016875</td>\n",
       "      <td>4.374375</td>\n",
       "      <td>0.2975</td>\n",
       "      <td>4.523125</td>\n",
       "      <td>0.15875</td>\n",
       "      <td>4.36375</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>17 rows × 36 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       VND best_ls kmax=1 max_iter=10 random_init - deviation (%)  \\\n",
       "40.2                                            0.000000            \n",
       "40.3                                            2.410000            \n",
       "40.4                                            0.000000            \n",
       "40.5                                            0.000000            \n",
       "40.10                                           0.000000            \n",
       "50.2                                            0.000000            \n",
       "50.3                                            2.220000            \n",
       "50.4                                            0.000000            \n",
       "50.5                                            0.070000            \n",
       "50.10                                           0.830000            \n",
       "100.2                                           0.000000            \n",
       "100.3                                           0.000000            \n",
       "100.4                                           0.060000            \n",
       "100.5                                           0.000000            \n",
       "200.2                                           0.000000            \n",
       "200.3                                           0.000000            \n",
       "mean                                            0.349375            \n",
       "\n",
       "       VND best_ls kmax=1 max_iter=10 random_init - time (s)  \\\n",
       "40.2                                            0.200000       \n",
       "40.3                                            0.330000       \n",
       "40.4                                            0.470000       \n",
       "40.5                                            0.610000       \n",
       "40.10                                           1.340000       \n",
       "50.2                                            0.360000       \n",
       "50.3                                            0.530000       \n",
       "50.4                                            0.860000       \n",
       "50.5                                            1.250000       \n",
       "50.10                                           3.030000       \n",
       "100.2                                           2.210000       \n",
       "100.3                                           3.500000       \n",
       "100.4                                           6.330000       \n",
       "100.5                                           7.600000       \n",
       "200.2                                          15.040000       \n",
       "200.3                                          31.230000       \n",
       "mean                                            4.680625       \n",
       "\n",
       "       VND best_ls kmax=1 max_iter=10 min_max_init - deviation (%)  \\\n",
       "40.2                                              0.0000             \n",
       "40.3                                              2.4100             \n",
       "40.4                                              0.0000             \n",
       "40.5                                              0.0000             \n",
       "40.10                                             0.0000             \n",
       "50.2                                              0.0000             \n",
       "50.3                                              2.2200             \n",
       "50.4                                              0.0000             \n",
       "50.5                                              0.0700             \n",
       "50.10                                             0.0000             \n",
       "100.2                                             0.0000             \n",
       "100.3                                             0.0000             \n",
       "100.4                                             0.0600             \n",
       "100.5                                             0.0000             \n",
       "200.2                                             0.0000             \n",
       "200.3                                             0.0000             \n",
       "mean                                              0.2975             \n",
       "\n",
       "       VND best_ls kmax=1 max_iter=10 min_max_init - time (s)  \\\n",
       "40.2                                            0.200000        \n",
       "40.3                                            0.300000        \n",
       "40.4                                            0.480000        \n",
       "40.5                                            0.600000        \n",
       "40.10                                           1.510000        \n",
       "50.2                                            0.350000        \n",
       "50.3                                            0.530000        \n",
       "50.4                                            0.970000        \n",
       "50.5                                            1.140000        \n",
       "50.10                                           2.920000        \n",
       "100.2                                           2.260000        \n",
       "100.3                                           3.680000        \n",
       "100.4                                           5.900000        \n",
       "100.5                                           7.820000        \n",
       "200.2                                          15.460000        \n",
       "200.3                                          28.290000        \n",
       "mean                                            4.525625        \n",
       "\n",
       "       VND best_ls kmax=1 max_iter=10 angle_init - deviation (%)  \\\n",
       "40.2                                            0.000000           \n",
       "40.3                                            2.410000           \n",
       "40.4                                            0.000000           \n",
       "40.5                                            0.000000           \n",
       "40.10                                           0.000000           \n",
       "50.2                                            0.000000           \n",
       "50.3                                            0.000000           \n",
       "50.4                                            0.000000           \n",
       "50.5                                            0.000000           \n",
       "50.10                                           0.140000           \n",
       "100.2                                           0.000000           \n",
       "100.3                                           0.000000           \n",
       "100.4                                           0.060000           \n",
       "100.5                                           0.000000           \n",
       "200.2                                           0.000000           \n",
       "200.3                                           0.000000           \n",
       "mean                                            0.163125           \n",
       "\n",
       "       VND best_ls kmax=1 max_iter=10 angle_init - time (s)  \\\n",
       "40.2                                                0.21      \n",
       "40.3                                                0.30      \n",
       "40.4                                                0.48      \n",
       "40.5                                                0.60      \n",
       "40.10                                               1.45      \n",
       "50.2                                                0.34      \n",
       "50.3                                                0.56      \n",
       "50.4                                                0.82      \n",
       "50.5                                                1.18      \n",
       "50.10                                               2.81      \n",
       "100.2                                               2.27      \n",
       "100.3                                               4.07      \n",
       "100.4                                               5.81      \n",
       "100.5                                               7.85      \n",
       "200.2                                              14.84      \n",
       "200.3                                              27.13      \n",
       "mean                                                4.42      \n",
       "\n",
       "       VND best_ls kmax=2 max_iter=10 random_init - deviation (%)  \\\n",
       "40.2                                             0.00000            \n",
       "40.3                                             2.41000            \n",
       "40.4                                             0.00000            \n",
       "40.5                                             0.00000            \n",
       "40.10                                            0.00000            \n",
       "50.2                                             0.00000            \n",
       "50.3                                             0.00000            \n",
       "50.4                                             0.00000            \n",
       "50.5                                             0.07000            \n",
       "50.10                                            0.00000            \n",
       "100.2                                            0.00000            \n",
       "100.3                                            0.00000            \n",
       "100.4                                            0.06000            \n",
       "100.5                                            0.00000            \n",
       "200.2                                            0.00000            \n",
       "200.3                                            0.00000            \n",
       "mean                                             0.15875            \n",
       "\n",
       "       VND best_ls kmax=2 max_iter=10 random_init - time (s)  \\\n",
       "40.2                                             0.21000       \n",
       "40.3                                             0.31000       \n",
       "40.4                                             0.48000       \n",
       "40.5                                             0.60000       \n",
       "40.10                                            1.53000       \n",
       "50.2                                             0.35000       \n",
       "50.3                                             0.66000       \n",
       "50.4                                             0.83000       \n",
       "50.5                                             1.12000       \n",
       "50.10                                            2.96000       \n",
       "100.2                                            2.12000       \n",
       "100.3                                            3.90000       \n",
       "100.4                                            5.58000       \n",
       "100.5                                            8.16000       \n",
       "200.2                                           14.92000       \n",
       "200.3                                           29.81000       \n",
       "mean                                             4.59625       \n",
       "\n",
       "       VND best_ls kmax=2 max_iter=10 min_max_init - deviation (%)  \\\n",
       "40.2                                              0.0000             \n",
       "40.3                                              2.4100             \n",
       "40.4                                              0.0000             \n",
       "40.5                                              0.0000             \n",
       "40.10                                             0.0000             \n",
       "50.2                                              0.0000             \n",
       "50.3                                              2.2200             \n",
       "50.4                                              0.0000             \n",
       "50.5                                              0.0700             \n",
       "50.10                                             0.0000             \n",
       "100.2                                             0.0000             \n",
       "100.3                                             0.0000             \n",
       "100.4                                             0.0600             \n",
       "100.5                                             0.0000             \n",
       "200.2                                             0.0000             \n",
       "200.3                                             0.0000             \n",
       "mean                                              0.2975             \n",
       "\n",
       "       VND best_ls kmax=2 max_iter=10 min_max_init - time (s)  ...  \\\n",
       "40.2                                              0.2000       ...   \n",
       "40.3                                              0.3000       ...   \n",
       "40.4                                              0.4800       ...   \n",
       "40.5                                              0.6000       ...   \n",
       "40.10                                             1.5100       ...   \n",
       "50.2                                              0.3500       ...   \n",
       "50.3                                              0.5400       ...   \n",
       "50.4                                              0.9600       ...   \n",
       "50.5                                              1.1600       ...   \n",
       "50.10                                             2.9200       ...   \n",
       "100.2                                             2.2600       ...   \n",
       "100.3                                             3.6700       ...   \n",
       "100.4                                             5.9100       ...   \n",
       "100.5                                             7.8100       ...   \n",
       "200.2                                            15.4600       ...   \n",
       "200.3                                            28.3100       ...   \n",
       "mean                                              4.5275       ...   \n",
       "\n",
       "       VND first_ls kmax=2 max_iter=10 min_max_init - deviation (%)  \\\n",
       "40.2                                              0.0000              \n",
       "40.3                                              2.4100              \n",
       "40.4                                              0.0000              \n",
       "40.5                                              0.0000              \n",
       "40.10                                             0.0000              \n",
       "50.2                                              0.0000              \n",
       "50.3                                              2.2200              \n",
       "50.4                                              0.0000              \n",
       "50.5                                              0.0700              \n",
       "50.10                                             0.0000              \n",
       "100.2                                             0.0000              \n",
       "100.3                                             0.0000              \n",
       "100.4                                             0.0600              \n",
       "100.5                                             0.0000              \n",
       "200.2                                             0.0000              \n",
       "200.3                                             0.0000              \n",
       "mean                                              0.2975              \n",
       "\n",
       "       VND first_ls kmax=2 max_iter=10 min_max_init - time (s)  \\\n",
       "40.2                                              0.2000         \n",
       "40.3                                              0.3100         \n",
       "40.4                                              0.4800         \n",
       "40.5                                              0.6000         \n",
       "40.10                                             1.5100         \n",
       "50.2                                              0.3500         \n",
       "50.3                                              0.5300         \n",
       "50.4                                              0.9600         \n",
       "50.5                                              1.1500         \n",
       "50.10                                             2.8900         \n",
       "100.2                                             2.2600         \n",
       "100.3                                             3.6600         \n",
       "100.4                                             5.9000         \n",
       "100.5                                             7.8200         \n",
       "200.2                                            15.4900         \n",
       "200.3                                            28.2500         \n",
       "mean                                              4.5225         \n",
       "\n",
       "       VND first_ls kmax=2 max_iter=10 angle_init - deviation (%)  \\\n",
       "40.2                                             0.00000            \n",
       "40.3                                             2.41000            \n",
       "40.4                                             0.00000            \n",
       "40.5                                             0.00000            \n",
       "40.10                                            0.00000            \n",
       "50.2                                             0.00000            \n",
       "50.3                                             2.22000            \n",
       "50.4                                             0.00000            \n",
       "50.5                                             0.07000            \n",
       "50.10                                            0.14000            \n",
       "100.2                                            0.00000            \n",
       "100.3                                            0.00000            \n",
       "100.4                                            0.06000            \n",
       "100.5                                            0.00000            \n",
       "200.2                                            0.00000            \n",
       "200.3                                            0.00000            \n",
       "mean                                             0.30625            \n",
       "\n",
       "       VND first_ls kmax=2 max_iter=10 angle_init - time (s)  \\\n",
       "40.2                                             0.20000       \n",
       "40.3                                             0.32000       \n",
       "40.4                                             0.51000       \n",
       "40.5                                             0.65000       \n",
       "40.10                                            1.68000       \n",
       "50.2                                             0.35000       \n",
       "50.3                                             0.54000       \n",
       "50.4                                             0.89000       \n",
       "50.5                                             1.18000       \n",
       "50.10                                            2.56000       \n",
       "100.2                                            2.08000       \n",
       "100.3                                            3.80000       \n",
       "100.4                                            5.86000       \n",
       "100.5                                            7.90000       \n",
       "200.2                                           14.63000       \n",
       "200.3                                           27.87000       \n",
       "mean                                             4.43875       \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 random_init - deviation (%)  \\\n",
       "40.2                                            0.000000             \n",
       "40.3                                            0.000000             \n",
       "40.4                                            0.000000             \n",
       "40.5                                            0.000000             \n",
       "40.10                                           0.000000             \n",
       "50.2                                            0.000000             \n",
       "50.3                                            0.000000             \n",
       "50.4                                            0.000000             \n",
       "50.5                                            0.070000             \n",
       "50.10                                           0.140000             \n",
       "100.2                                           0.000000             \n",
       "100.3                                           0.000000             \n",
       "100.4                                           0.060000             \n",
       "100.5                                           0.000000             \n",
       "200.2                                           0.000000             \n",
       "200.3                                           0.000000             \n",
       "mean                                            0.016875             \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 random_init - time (s)  \\\n",
       "40.2                                            0.200000        \n",
       "40.3                                            0.330000        \n",
       "40.4                                            0.430000        \n",
       "40.5                                            0.610000        \n",
       "40.10                                           1.510000        \n",
       "50.2                                            0.350000        \n",
       "50.3                                            0.640000        \n",
       "50.4                                            0.820000        \n",
       "50.5                                            1.110000        \n",
       "50.10                                           2.780000        \n",
       "100.2                                           2.020000        \n",
       "100.3                                           4.090000        \n",
       "100.4                                           6.010000        \n",
       "100.5                                           7.410000        \n",
       "200.2                                          15.210000        \n",
       "200.3                                          26.470000        \n",
       "mean                                            4.374375        \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 min_max_init - deviation (%)  \\\n",
       "40.2                                              0.0000              \n",
       "40.3                                              2.4100              \n",
       "40.4                                              0.0000              \n",
       "40.5                                              0.0000              \n",
       "40.10                                             0.0000              \n",
       "50.2                                              0.0000              \n",
       "50.3                                              2.2200              \n",
       "50.4                                              0.0000              \n",
       "50.5                                              0.0700              \n",
       "50.10                                             0.0000              \n",
       "100.2                                             0.0000              \n",
       "100.3                                             0.0000              \n",
       "100.4                                             0.0600              \n",
       "100.5                                             0.0000              \n",
       "200.2                                             0.0000              \n",
       "200.3                                             0.0000              \n",
       "mean                                              0.2975              \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 min_max_init - time (s)  \\\n",
       "40.2                                            0.200000         \n",
       "40.3                                            0.310000         \n",
       "40.4                                            0.480000         \n",
       "40.5                                            0.600000         \n",
       "40.10                                           1.500000         \n",
       "50.2                                            0.340000         \n",
       "50.3                                            0.540000         \n",
       "50.4                                            0.960000         \n",
       "50.5                                            1.140000         \n",
       "50.10                                           2.900000         \n",
       "100.2                                           2.260000         \n",
       "100.3                                           3.660000         \n",
       "100.4                                           5.920000         \n",
       "100.5                                           7.790000         \n",
       "200.2                                          15.480000         \n",
       "200.3                                          28.290000         \n",
       "mean                                            4.523125         \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 angle_init - deviation (%)  \\\n",
       "40.2                                             0.00000            \n",
       "40.3                                             2.41000            \n",
       "40.4                                             0.00000            \n",
       "40.5                                             0.00000            \n",
       "40.10                                            0.00000            \n",
       "50.2                                             0.00000            \n",
       "50.3                                             0.00000            \n",
       "50.4                                             0.00000            \n",
       "50.5                                             0.07000            \n",
       "50.10                                            0.00000            \n",
       "100.2                                            0.00000            \n",
       "100.3                                            0.00000            \n",
       "100.4                                            0.06000            \n",
       "100.5                                            0.00000            \n",
       "200.2                                            0.00000            \n",
       "200.3                                            0.00000            \n",
       "mean                                             0.15875            \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 angle_init - time (s)  \n",
       "40.2                                             0.19000      \n",
       "40.3                                             0.33000      \n",
       "40.4                                             0.45000      \n",
       "40.5                                             0.61000      \n",
       "40.10                                            1.39000      \n",
       "50.2                                             0.34000      \n",
       "50.3                                             0.63000      \n",
       "50.4                                             0.82000      \n",
       "50.5                                             1.10000      \n",
       "50.10                                            2.88000      \n",
       "100.2                                            2.10000      \n",
       "100.3                                            3.70000      \n",
       "100.4                                            6.05000      \n",
       "100.5                                            7.50000      \n",
       "200.2                                           14.99000      \n",
       "200.3                                           26.74000      \n",
       "mean                                             4.36375      \n",
       "\n",
       "[17 rows x 36 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comparison_table.head(NUMBER_OF_PROBLEMS+1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_all_columns_with_similar_value(df, value):\n",
    "    return df.loc[:, abs(df.iloc[-1] - value) < 0.02].columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "time_table = comparison_table[comparison_table.columns[1::2]]\n",
    "deviation_table = comparison_table[comparison_table.columns[::2]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('VND first_ls kmax=3 max_iter=10 random_init - deviation (%)', 0.016875)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# min deviation\n",
    "\n",
    "best_method_per_dev = deviation_table.idxmin(axis=1)[-1]\n",
    "worst_method_per_dev = deviation_table.idxmax(axis=1)[-1]\n",
    "\n",
    "best_dev_value = comparison_table[best_method_per_dev]['mean']\n",
    "worst_dev_value = comparison_table[worst_method_per_dev]['mean']\n",
    "\n",
    "\n",
    "best_method_per_dev, best_dev_value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('VND best_ls kmax=3 max_iter=10 angle_init - deviation (%)',\n",
       " 0.4256249999999999)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "worst_method_per_dev, worst_dev_value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('VND best_ls kmax=3 max_iter=10 random_init - time (s)', 4.344374999999999)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# min time\n",
    "\n",
    "best_method_per_time = time_table.idxmin(axis=1)[-1]\n",
    "worst_method_per_time = time_table.idxmax(axis=1)[-1]\n",
    "\n",
    "best_time_value = comparison_table[best_method_per_time]['mean']\n",
    "worst_time_value = comparison_table[worst_method_per_time]['mean']\n",
    "\n",
    "best_method_per_time, best_time_value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('VND first_ls kmax=1 max_iter=10 angle_init - time (s)', 4.7593749999999995)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "worst_method_per_time, worst_time_value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ------- sort all columns based on mean ----------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "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>VND best_ls kmax=3 max_iter=10 random_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 angle_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 random_init - time (s)</th>\n",
       "      <th>VND best_ls kmax=1 max_iter=10 angle_init - time (s)</th>\n",
       "      <th>VND best_ls kmax=2 max_iter=10 angle_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=2 max_iter=10 random_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=2 max_iter=10 angle_init - time (s)</th>\n",
       "      <th>VND best_ls kmax=3 max_iter=10 min_max_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=2 max_iter=10 min_max_init - time (s)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 min_max_init - time (s)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40.2</th>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.19000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.21</td>\n",
       "      <td>0.20000</td>\n",
       "      <td>0.2000</td>\n",
       "      <td>0.20000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.2000</td>\n",
       "      <td>0.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.3</th>\n",
       "      <td>0.310000</td>\n",
       "      <td>0.33000</td>\n",
       "      <td>0.330000</td>\n",
       "      <td>0.30</td>\n",
       "      <td>0.32000</td>\n",
       "      <td>0.3300</td>\n",
       "      <td>0.32000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>0.3100</td>\n",
       "      <td>0.310000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.4</th>\n",
       "      <td>0.500000</td>\n",
       "      <td>0.45000</td>\n",
       "      <td>0.430000</td>\n",
       "      <td>0.48</td>\n",
       "      <td>0.49000</td>\n",
       "      <td>0.4700</td>\n",
       "      <td>0.51000</td>\n",
       "      <td>0.480000</td>\n",
       "      <td>0.4800</td>\n",
       "      <td>0.480000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.5</th>\n",
       "      <td>0.580000</td>\n",
       "      <td>0.61000</td>\n",
       "      <td>0.610000</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.64000</td>\n",
       "      <td>0.6000</td>\n",
       "      <td>0.65000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.6000</td>\n",
       "      <td>0.600000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.10</th>\n",
       "      <td>1.480000</td>\n",
       "      <td>1.39000</td>\n",
       "      <td>1.510000</td>\n",
       "      <td>1.45</td>\n",
       "      <td>1.64000</td>\n",
       "      <td>1.4700</td>\n",
       "      <td>1.68000</td>\n",
       "      <td>1.510000</td>\n",
       "      <td>1.5100</td>\n",
       "      <td>1.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.2</th>\n",
       "      <td>0.350000</td>\n",
       "      <td>0.34000</td>\n",
       "      <td>0.350000</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.35000</td>\n",
       "      <td>0.3500</td>\n",
       "      <td>0.35000</td>\n",
       "      <td>0.340000</td>\n",
       "      <td>0.3500</td>\n",
       "      <td>0.340000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.3</th>\n",
       "      <td>0.580000</td>\n",
       "      <td>0.63000</td>\n",
       "      <td>0.640000</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.61000</td>\n",
       "      <td>0.6300</td>\n",
       "      <td>0.54000</td>\n",
       "      <td>0.540000</td>\n",
       "      <td>0.5300</td>\n",
       "      <td>0.540000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.4</th>\n",
       "      <td>0.790000</td>\n",
       "      <td>0.82000</td>\n",
       "      <td>0.820000</td>\n",
       "      <td>0.82</td>\n",
       "      <td>0.80000</td>\n",
       "      <td>0.8200</td>\n",
       "      <td>0.89000</td>\n",
       "      <td>0.970000</td>\n",
       "      <td>0.9600</td>\n",
       "      <td>0.960000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.5</th>\n",
       "      <td>1.140000</td>\n",
       "      <td>1.10000</td>\n",
       "      <td>1.110000</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.20000</td>\n",
       "      <td>1.0500</td>\n",
       "      <td>1.18000</td>\n",
       "      <td>1.140000</td>\n",
       "      <td>1.1500</td>\n",
       "      <td>1.140000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.10</th>\n",
       "      <td>3.040000</td>\n",
       "      <td>2.88000</td>\n",
       "      <td>2.780000</td>\n",
       "      <td>2.81</td>\n",
       "      <td>2.63000</td>\n",
       "      <td>2.8400</td>\n",
       "      <td>2.56000</td>\n",
       "      <td>2.910000</td>\n",
       "      <td>2.8900</td>\n",
       "      <td>2.900000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.2</th>\n",
       "      <td>2.070000</td>\n",
       "      <td>2.10000</td>\n",
       "      <td>2.020000</td>\n",
       "      <td>2.27</td>\n",
       "      <td>2.13000</td>\n",
       "      <td>2.1700</td>\n",
       "      <td>2.08000</td>\n",
       "      <td>2.260000</td>\n",
       "      <td>2.2600</td>\n",
       "      <td>2.260000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.3</th>\n",
       "      <td>4.060000</td>\n",
       "      <td>3.70000</td>\n",
       "      <td>4.090000</td>\n",
       "      <td>4.07</td>\n",
       "      <td>3.84000</td>\n",
       "      <td>3.6700</td>\n",
       "      <td>3.80000</td>\n",
       "      <td>3.660000</td>\n",
       "      <td>3.6600</td>\n",
       "      <td>3.660000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.4</th>\n",
       "      <td>5.290000</td>\n",
       "      <td>6.05000</td>\n",
       "      <td>6.010000</td>\n",
       "      <td>5.81</td>\n",
       "      <td>5.44000</td>\n",
       "      <td>5.4600</td>\n",
       "      <td>5.86000</td>\n",
       "      <td>5.900000</td>\n",
       "      <td>5.9000</td>\n",
       "      <td>5.920000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.5</th>\n",
       "      <td>7.400000</td>\n",
       "      <td>7.50000</td>\n",
       "      <td>7.410000</td>\n",
       "      <td>7.85</td>\n",
       "      <td>7.97000</td>\n",
       "      <td>7.2700</td>\n",
       "      <td>7.90000</td>\n",
       "      <td>7.830000</td>\n",
       "      <td>7.8200</td>\n",
       "      <td>7.790000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200.2</th>\n",
       "      <td>15.420000</td>\n",
       "      <td>14.99000</td>\n",
       "      <td>15.210000</td>\n",
       "      <td>14.84</td>\n",
       "      <td>14.96000</td>\n",
       "      <td>15.1000</td>\n",
       "      <td>14.63000</td>\n",
       "      <td>15.440000</td>\n",
       "      <td>15.4900</td>\n",
       "      <td>15.480000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200.3</th>\n",
       "      <td>26.300000</td>\n",
       "      <td>26.74000</td>\n",
       "      <td>26.470000</td>\n",
       "      <td>27.13</td>\n",
       "      <td>27.56000</td>\n",
       "      <td>28.4900</td>\n",
       "      <td>27.87000</td>\n",
       "      <td>28.270000</td>\n",
       "      <td>28.2500</td>\n",
       "      <td>28.290000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.344375</td>\n",
       "      <td>4.36375</td>\n",
       "      <td>4.374375</td>\n",
       "      <td>4.42</td>\n",
       "      <td>4.42375</td>\n",
       "      <td>4.4325</td>\n",
       "      <td>4.43875</td>\n",
       "      <td>4.521875</td>\n",
       "      <td>4.5225</td>\n",
       "      <td>4.523125</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       VND best_ls kmax=3 max_iter=10 random_init - time (s)  \\\n",
       "40.2                                            0.200000       \n",
       "40.3                                            0.310000       \n",
       "40.4                                            0.500000       \n",
       "40.5                                            0.580000       \n",
       "40.10                                           1.480000       \n",
       "50.2                                            0.350000       \n",
       "50.3                                            0.580000       \n",
       "50.4                                            0.790000       \n",
       "50.5                                            1.140000       \n",
       "50.10                                           3.040000       \n",
       "100.2                                           2.070000       \n",
       "100.3                                           4.060000       \n",
       "100.4                                           5.290000       \n",
       "100.5                                           7.400000       \n",
       "200.2                                          15.420000       \n",
       "200.3                                          26.300000       \n",
       "mean                                            4.344375       \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 angle_init - time (s)  \\\n",
       "40.2                                             0.19000       \n",
       "40.3                                             0.33000       \n",
       "40.4                                             0.45000       \n",
       "40.5                                             0.61000       \n",
       "40.10                                            1.39000       \n",
       "50.2                                             0.34000       \n",
       "50.3                                             0.63000       \n",
       "50.4                                             0.82000       \n",
       "50.5                                             1.10000       \n",
       "50.10                                            2.88000       \n",
       "100.2                                            2.10000       \n",
       "100.3                                            3.70000       \n",
       "100.4                                            6.05000       \n",
       "100.5                                            7.50000       \n",
       "200.2                                           14.99000       \n",
       "200.3                                           26.74000       \n",
       "mean                                             4.36375       \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 random_init - time (s)  \\\n",
       "40.2                                            0.200000        \n",
       "40.3                                            0.330000        \n",
       "40.4                                            0.430000        \n",
       "40.5                                            0.610000        \n",
       "40.10                                           1.510000        \n",
       "50.2                                            0.350000        \n",
       "50.3                                            0.640000        \n",
       "50.4                                            0.820000        \n",
       "50.5                                            1.110000        \n",
       "50.10                                           2.780000        \n",
       "100.2                                           2.020000        \n",
       "100.3                                           4.090000        \n",
       "100.4                                           6.010000        \n",
       "100.5                                           7.410000        \n",
       "200.2                                          15.210000        \n",
       "200.3                                          26.470000        \n",
       "mean                                            4.374375        \n",
       "\n",
       "       VND best_ls kmax=1 max_iter=10 angle_init - time (s)  \\\n",
       "40.2                                                0.21      \n",
       "40.3                                                0.30      \n",
       "40.4                                                0.48      \n",
       "40.5                                                0.60      \n",
       "40.10                                               1.45      \n",
       "50.2                                                0.34      \n",
       "50.3                                                0.56      \n",
       "50.4                                                0.82      \n",
       "50.5                                                1.18      \n",
       "50.10                                               2.81      \n",
       "100.2                                               2.27      \n",
       "100.3                                               4.07      \n",
       "100.4                                               5.81      \n",
       "100.5                                               7.85      \n",
       "200.2                                              14.84      \n",
       "200.3                                              27.13      \n",
       "mean                                                4.42      \n",
       "\n",
       "       VND best_ls kmax=2 max_iter=10 angle_init - time (s)  \\\n",
       "40.2                                             0.20000      \n",
       "40.3                                             0.32000      \n",
       "40.4                                             0.49000      \n",
       "40.5                                             0.64000      \n",
       "40.10                                            1.64000      \n",
       "50.2                                             0.35000      \n",
       "50.3                                             0.61000      \n",
       "50.4                                             0.80000      \n",
       "50.5                                             1.20000      \n",
       "50.10                                            2.63000      \n",
       "100.2                                            2.13000      \n",
       "100.3                                            3.84000      \n",
       "100.4                                            5.44000      \n",
       "100.5                                            7.97000      \n",
       "200.2                                           14.96000      \n",
       "200.3                                           27.56000      \n",
       "mean                                             4.42375      \n",
       "\n",
       "       VND first_ls kmax=2 max_iter=10 random_init - time (s)  \\\n",
       "40.2                                              0.2000        \n",
       "40.3                                              0.3300        \n",
       "40.4                                              0.4700        \n",
       "40.5                                              0.6000        \n",
       "40.10                                             1.4700        \n",
       "50.2                                              0.3500        \n",
       "50.3                                              0.6300        \n",
       "50.4                                              0.8200        \n",
       "50.5                                              1.0500        \n",
       "50.10                                             2.8400        \n",
       "100.2                                             2.1700        \n",
       "100.3                                             3.6700        \n",
       "100.4                                             5.4600        \n",
       "100.5                                             7.2700        \n",
       "200.2                                            15.1000        \n",
       "200.3                                            28.4900        \n",
       "mean                                              4.4325        \n",
       "\n",
       "       VND first_ls kmax=2 max_iter=10 angle_init - time (s)  \\\n",
       "40.2                                             0.20000       \n",
       "40.3                                             0.32000       \n",
       "40.4                                             0.51000       \n",
       "40.5                                             0.65000       \n",
       "40.10                                            1.68000       \n",
       "50.2                                             0.35000       \n",
       "50.3                                             0.54000       \n",
       "50.4                                             0.89000       \n",
       "50.5                                             1.18000       \n",
       "50.10                                            2.56000       \n",
       "100.2                                            2.08000       \n",
       "100.3                                            3.80000       \n",
       "100.4                                            5.86000       \n",
       "100.5                                            7.90000       \n",
       "200.2                                           14.63000       \n",
       "200.3                                           27.87000       \n",
       "mean                                             4.43875       \n",
       "\n",
       "       VND best_ls kmax=3 max_iter=10 min_max_init - time (s)  \\\n",
       "40.2                                            0.200000        \n",
       "40.3                                            0.300000        \n",
       "40.4                                            0.480000        \n",
       "40.5                                            0.600000        \n",
       "40.10                                           1.510000        \n",
       "50.2                                            0.340000        \n",
       "50.3                                            0.540000        \n",
       "50.4                                            0.970000        \n",
       "50.5                                            1.140000        \n",
       "50.10                                           2.910000        \n",
       "100.2                                           2.260000        \n",
       "100.3                                           3.660000        \n",
       "100.4                                           5.900000        \n",
       "100.5                                           7.830000        \n",
       "200.2                                          15.440000        \n",
       "200.3                                          28.270000        \n",
       "mean                                            4.521875        \n",
       "\n",
       "       VND first_ls kmax=2 max_iter=10 min_max_init - time (s)  \\\n",
       "40.2                                              0.2000         \n",
       "40.3                                              0.3100         \n",
       "40.4                                              0.4800         \n",
       "40.5                                              0.6000         \n",
       "40.10                                             1.5100         \n",
       "50.2                                              0.3500         \n",
       "50.3                                              0.5300         \n",
       "50.4                                              0.9600         \n",
       "50.5                                              1.1500         \n",
       "50.10                                             2.8900         \n",
       "100.2                                             2.2600         \n",
       "100.3                                             3.6600         \n",
       "100.4                                             5.9000         \n",
       "100.5                                             7.8200         \n",
       "200.2                                            15.4900         \n",
       "200.3                                            28.2500         \n",
       "mean                                              4.5225         \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 min_max_init - time (s)  \n",
       "40.2                                            0.200000        \n",
       "40.3                                            0.310000        \n",
       "40.4                                            0.480000        \n",
       "40.5                                            0.600000        \n",
       "40.10                                           1.500000        \n",
       "50.2                                            0.340000        \n",
       "50.3                                            0.540000        \n",
       "50.4                                            0.960000        \n",
       "50.5                                            1.140000        \n",
       "50.10                                           2.900000        \n",
       "100.2                                           2.260000        \n",
       "100.3                                           3.660000        \n",
       "100.4                                           5.920000        \n",
       "100.5                                           7.790000        \n",
       "200.2                                          15.480000        \n",
       "200.3                                          28.290000        \n",
       "mean                                            4.523125        "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ten_best_per_time = time_table.sort_values(by='mean', axis=1).iloc[:, : 10]\n",
    "ten_best_per_time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "column1= ten_best_per_time.iloc[:-1, 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f4db46947c0>]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 2000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "f = plt.figure()\n",
    "f.set_figwidth(20)\n",
    "f.set_figheight(10)\n",
    "\n",
    "\n",
    "plt.plot(ten_best_per_time.iloc[:-1, 0])\n",
    "plt.plot(ten_best_per_time.iloc[:-1, 1])\n",
    "plt.plot(ten_best_per_time.iloc[:-1, 2])\n",
    "plt.plot(ten_best_per_time.iloc[:-1, 3])\n",
    "plt.plot(ten_best_per_time.iloc[:-1, 4])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "time_table.sort_values(by='mean', axis=1).to_csv(\"vnd_best_per_time.csv\", index_label='problem')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "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>VND first_ls kmax=3 max_iter=10 random_init - deviation (%)</th>\n",
       "      <th>VND first_ls kmax=3 max_iter=10 angle_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=2 max_iter=10 random_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=3 max_iter=10 random_init - deviation (%)</th>\n",
       "      <th>VND first_ls kmax=1 max_iter=10 angle_init - deviation (%)</th>\n",
       "      <th>VND first_ls kmax=2 max_iter=10 random_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=2 max_iter=10 angle_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=1 max_iter=10 angle_init - deviation (%)</th>\n",
       "      <th>VND first_ls kmax=1 max_iter=10 random_init - deviation (%)</th>\n",
       "      <th>VND best_ls kmax=1 max_iter=10 min_max_init - deviation (%)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40.2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.3</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.41000</td>\n",
       "      <td>2.41000</td>\n",
       "      <td>2.41000</td>\n",
       "      <td>2.41000</td>\n",
       "      <td>2.41000</td>\n",
       "      <td>2.410000</td>\n",
       "      <td>2.410000</td>\n",
       "      <td>2.410000</td>\n",
       "      <td>2.4100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.4</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.5</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.10</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.3</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.220000</td>\n",
       "      <td>2.2200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.4</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.5</th>\n",
       "      <td>0.070000</td>\n",
       "      <td>0.07000</td>\n",
       "      <td>0.07000</td>\n",
       "      <td>0.07000</td>\n",
       "      <td>0.07000</td>\n",
       "      <td>0.07000</td>\n",
       "      <td>0.070000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.10</th>\n",
       "      <td>0.140000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.140000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.3</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.4</th>\n",
       "      <td>0.060000</td>\n",
       "      <td>0.06000</td>\n",
       "      <td>0.06000</td>\n",
       "      <td>0.06000</td>\n",
       "      <td>0.06000</td>\n",
       "      <td>0.06000</td>\n",
       "      <td>0.060000</td>\n",
       "      <td>0.060000</td>\n",
       "      <td>0.060000</td>\n",
       "      <td>0.0600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.5</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.010000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200.2</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200.3</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.016875</td>\n",
       "      <td>0.15875</td>\n",
       "      <td>0.15875</td>\n",
       "      <td>0.15875</td>\n",
       "      <td>0.15875</td>\n",
       "      <td>0.15875</td>\n",
       "      <td>0.159375</td>\n",
       "      <td>0.163125</td>\n",
       "      <td>0.293125</td>\n",
       "      <td>0.2975</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       VND first_ls kmax=3 max_iter=10 random_init - deviation (%)  \\\n",
       "40.2                                            0.000000             \n",
       "40.3                                            0.000000             \n",
       "40.4                                            0.000000             \n",
       "40.5                                            0.000000             \n",
       "40.10                                           0.000000             \n",
       "50.2                                            0.000000             \n",
       "50.3                                            0.000000             \n",
       "50.4                                            0.000000             \n",
       "50.5                                            0.070000             \n",
       "50.10                                           0.140000             \n",
       "100.2                                           0.000000             \n",
       "100.3                                           0.000000             \n",
       "100.4                                           0.060000             \n",
       "100.5                                           0.000000             \n",
       "200.2                                           0.000000             \n",
       "200.3                                           0.000000             \n",
       "mean                                            0.016875             \n",
       "\n",
       "       VND first_ls kmax=3 max_iter=10 angle_init - deviation (%)  \\\n",
       "40.2                                             0.00000            \n",
       "40.3                                             2.41000            \n",
       "40.4                                             0.00000            \n",
       "40.5                                             0.00000            \n",
       "40.10                                            0.00000            \n",
       "50.2                                             0.00000            \n",
       "50.3                                             0.00000            \n",
       "50.4                                             0.00000            \n",
       "50.5                                             0.07000            \n",
       "50.10                                            0.00000            \n",
       "100.2                                            0.00000            \n",
       "100.3                                            0.00000            \n",
       "100.4                                            0.06000            \n",
       "100.5                                            0.00000            \n",
       "200.2                                            0.00000            \n",
       "200.3                                            0.00000            \n",
       "mean                                             0.15875            \n",
       "\n",
       "       VND best_ls kmax=2 max_iter=10 random_init - deviation (%)  \\\n",
       "40.2                                             0.00000            \n",
       "40.3                                             2.41000            \n",
       "40.4                                             0.00000            \n",
       "40.5                                             0.00000            \n",
       "40.10                                            0.00000            \n",
       "50.2                                             0.00000            \n",
       "50.3                                             0.00000            \n",
       "50.4                                             0.00000            \n",
       "50.5                                             0.07000            \n",
       "50.10                                            0.00000            \n",
       "100.2                                            0.00000            \n",
       "100.3                                            0.00000            \n",
       "100.4                                            0.06000            \n",
       "100.5                                            0.00000            \n",
       "200.2                                            0.00000            \n",
       "200.3                                            0.00000            \n",
       "mean                                             0.15875            \n",
       "\n",
       "       VND best_ls kmax=3 max_iter=10 random_init - deviation (%)  \\\n",
       "40.2                                             0.00000            \n",
       "40.3                                             2.41000            \n",
       "40.4                                             0.00000            \n",
       "40.5                                             0.00000            \n",
       "40.10                                            0.00000            \n",
       "50.2                                             0.00000            \n",
       "50.3                                             0.00000            \n",
       "50.4                                             0.00000            \n",
       "50.5                                             0.07000            \n",
       "50.10                                            0.00000            \n",
       "100.2                                            0.00000            \n",
       "100.3                                            0.00000            \n",
       "100.4                                            0.06000            \n",
       "100.5                                            0.00000            \n",
       "200.2                                            0.00000            \n",
       "200.3                                            0.00000            \n",
       "mean                                             0.15875            \n",
       "\n",
       "       VND first_ls kmax=1 max_iter=10 angle_init - deviation (%)  \\\n",
       "40.2                                             0.00000            \n",
       "40.3                                             2.41000            \n",
       "40.4                                             0.00000            \n",
       "40.5                                             0.00000            \n",
       "40.10                                            0.00000            \n",
       "50.2                                             0.00000            \n",
       "50.3                                             0.00000            \n",
       "50.4                                             0.00000            \n",
       "50.5                                             0.07000            \n",
       "50.10                                            0.00000            \n",
       "100.2                                            0.00000            \n",
       "100.3                                            0.00000            \n",
       "100.4                                            0.06000            \n",
       "100.5                                            0.00000            \n",
       "200.2                                            0.00000            \n",
       "200.3                                            0.00000            \n",
       "mean                                             0.15875            \n",
       "\n",
       "       VND first_ls kmax=2 max_iter=10 random_init - deviation (%)  \\\n",
       "40.2                                             0.00000             \n",
       "40.3                                             2.41000             \n",
       "40.4                                             0.00000             \n",
       "40.5                                             0.00000             \n",
       "40.10                                            0.00000             \n",
       "50.2                                             0.00000             \n",
       "50.3                                             0.00000             \n",
       "50.4                                             0.00000             \n",
       "50.5                                             0.07000             \n",
       "50.10                                            0.00000             \n",
       "100.2                                            0.00000             \n",
       "100.3                                            0.00000             \n",
       "100.4                                            0.06000             \n",
       "100.5                                            0.00000             \n",
       "200.2                                            0.00000             \n",
       "200.3                                            0.00000             \n",
       "mean                                             0.15875             \n",
       "\n",
       "       VND best_ls kmax=2 max_iter=10 angle_init - deviation (%)  \\\n",
       "40.2                                            0.000000           \n",
       "40.3                                            2.410000           \n",
       "40.4                                            0.000000           \n",
       "40.5                                            0.000000           \n",
       "40.10                                           0.000000           \n",
       "50.2                                            0.000000           \n",
       "50.3                                            0.000000           \n",
       "50.4                                            0.000000           \n",
       "50.5                                            0.070000           \n",
       "50.10                                           0.000000           \n",
       "100.2                                           0.000000           \n",
       "100.3                                           0.000000           \n",
       "100.4                                           0.060000           \n",
       "100.5                                           0.010000           \n",
       "200.2                                           0.000000           \n",
       "200.3                                           0.000000           \n",
       "mean                                            0.159375           \n",
       "\n",
       "       VND best_ls kmax=1 max_iter=10 angle_init - deviation (%)  \\\n",
       "40.2                                            0.000000           \n",
       "40.3                                            2.410000           \n",
       "40.4                                            0.000000           \n",
       "40.5                                            0.000000           \n",
       "40.10                                           0.000000           \n",
       "50.2                                            0.000000           \n",
       "50.3                                            0.000000           \n",
       "50.4                                            0.000000           \n",
       "50.5                                            0.000000           \n",
       "50.10                                           0.140000           \n",
       "100.2                                           0.000000           \n",
       "100.3                                           0.000000           \n",
       "100.4                                           0.060000           \n",
       "100.5                                           0.000000           \n",
       "200.2                                           0.000000           \n",
       "200.3                                           0.000000           \n",
       "mean                                            0.163125           \n",
       "\n",
       "       VND first_ls kmax=1 max_iter=10 random_init - deviation (%)  \\\n",
       "40.2                                            0.000000             \n",
       "40.3                                            2.410000             \n",
       "40.4                                            0.000000             \n",
       "40.5                                            0.000000             \n",
       "40.10                                           0.000000             \n",
       "50.2                                            0.000000             \n",
       "50.3                                            2.220000             \n",
       "50.4                                            0.000000             \n",
       "50.5                                            0.000000             \n",
       "50.10                                           0.000000             \n",
       "100.2                                           0.000000             \n",
       "100.3                                           0.000000             \n",
       "100.4                                           0.060000             \n",
       "100.5                                           0.000000             \n",
       "200.2                                           0.000000             \n",
       "200.3                                           0.000000             \n",
       "mean                                            0.293125             \n",
       "\n",
       "       VND best_ls kmax=1 max_iter=10 min_max_init - deviation (%)  \n",
       "40.2                                              0.0000            \n",
       "40.3                                              2.4100            \n",
       "40.4                                              0.0000            \n",
       "40.5                                              0.0000            \n",
       "40.10                                             0.0000            \n",
       "50.2                                              0.0000            \n",
       "50.3                                              2.2200            \n",
       "50.4                                              0.0000            \n",
       "50.5                                              0.0700            \n",
       "50.10                                             0.0000            \n",
       "100.2                                             0.0000            \n",
       "100.3                                             0.0000            \n",
       "100.4                                             0.0600            \n",
       "100.5                                             0.0000            \n",
       "200.2                                             0.0000            \n",
       "200.3                                             0.0000            \n",
       "mean                                              0.2975            "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ten_best_dev = deviation_table.sort_values(by='mean', axis=1).iloc[:, : 10]\n",
    "ten_best_dev"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "VND-ls_best-kmax=3-max_iter=10 - angle_init - deviation (%)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "ten_best_dev.to_csv(\"vnd_ten_best_per_dev.csv\", index_label='problem')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}