figure.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {},
   "outputs": [],
   "source": [
    "import wandb\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as ticker\n",
    "\n",
    "# plt.rcParams['text.usetex'] = True\n",
    "import numpy as np\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 256,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wandb.login()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 459,
   "metadata": {},
   "outputs": [],
   "source": [
    "project_name = ['LDS_1', 'LDS_2', 'LDS_3']\n",
    "\n",
    "all_runs_data = []\n",
    "\n",
    "api = wandb.Api()\n",
    "for prj_ne in project_name:\n",
    "    runs = api.runs(prj_ne)\n",
    "    for run in runs:\n",
    "        metrics = run.history()\n",
    "        run_data = {\n",
    "            'run_id': run.id,\n",
    "            'name': run.name,\n",
    "            'config': run.config,\n",
    "            'seed' : run.name.split('-')[-1],\n",
    "            'metrics': metrics\n",
    "        }\n",
    "        all_runs_data.append(run_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 460,
   "metadata": {},
   "outputs": [],
   "source": [
    "# lds\n",
    "Easy_lds_data = [ run_data for run_data in all_runs_data if 'Easy' in run_data['name']and 'lds' in run_data['name']]\n",
    "Ur_lds_data = [ run_data for run_data in all_runs_data if 'Ur' in run_data['name'] and 'lds' in run_data['name']]\n",
    "Hard_lds_data = [ run_data for run_data in all_runs_data if 'Hard' in run_data['name'] and 'lds' in run_data['name']]\n",
    "fetch_lds_data = [ run_data for run_data in all_runs_data if 'fetch' in run_data['name'] and 'lds' in run_data['name']]\n",
    "finger_lds_data = [ run_data for run_data in all_runs_data if 'finger' in run_data['name'] and 'lds' in run_data['name']]\n",
    "fetch_push_lds_data = [ run_data for run_data in all_runs_data if 'fetch_push' in run_data['name'] and 'lds' in run_data['name']]\n",
    "hopper_lds_data = [ run_data for run_data in all_runs_data if 'hopper' in run_data['name'] and 'lds' in run_data['name']]\n",
    "# diayn\n",
    "Easy_diayn_data = [ run_data for run_data in all_runs_data if 'Easy' in run_data['name']and 'diayn' in run_data['name']]\n",
    "Ur_diayn_data = [ run_data for run_data in all_runs_data if 'Ur' in run_data['name'] and 'diayn' in run_data['name']]\n",
    "Hard_diayn_data = [ run_data for run_data in all_runs_data if 'Hard' in run_data['name'] and 'diayn' in run_data['name']]\n",
    "fetch_diayn_data = [ run_data for run_data in all_runs_data if 'fetch' in run_data['name'] and 'diayn' in run_data['name']]\n",
    "finger_diayn_data = [ run_data for run_data in all_runs_data if 'finger' in run_data['name'] and 'diayn' in run_data['name']]\n",
    "fetch_push_diayn_data = [ run_data for run_data in all_runs_data if 'fetch_push' in run_data['name'] and 'diayn' in run_data['name']]\n",
    "hopper_diayn_data = [ run_data for run_data in all_runs_data if 'hopper' in run_data['name'] and 'diayn' in run_data['name']]\n",
    "# rnd \n",
    "Easy_rnd_data = [ run_data for run_data in all_runs_data if 'Easy' in run_data['name']and 'rnd' in run_data['name']]\n",
    "Ur_rnd_data = [ run_data for run_data in all_runs_data if 'Ur' in run_data['name'] and 'rnd' in run_data['name']]\n",
    "Hard_rnd_data = [ run_data for run_data in all_runs_data if 'Hard' in run_data['name'] and 'rnd' in run_data['name']]\n",
    "fetch_rnd_data = [ run_data for run_data in all_runs_data if 'fetch' in run_data['name'] and 'rnd' in run_data['name']]\n",
    "finger_rnd_data = [ run_data for run_data in all_runs_data if 'finger' in run_data['name'] and 'rnd' in run_data['name']]\n",
    "fetch_push_rnd_data = [ run_data for run_data in all_runs_data if 'fetch_push' in run_data['name'] and 'rnd' in run_data['name']]\n",
    "hopper_rnd_data = [ run_data for run_data in all_runs_data if 'hopper' in run_data['name'] and 'rnd' in run_data['name']]\n",
    "# ngu\n",
    "Easy_ngu_data = [ run_data for run_data in all_runs_data if 'Easy' in run_data['name']and 'ngu' in run_data['name']]\n",
    "Ur_ngu_data = [ run_data for run_data in all_runs_data if 'Ur' in run_data['name'] and 'ngu' in run_data['name']]\n",
    "Hard_ngu_data = [ run_data for run_data in all_runs_data if 'Hard' in run_data['name'] and 'ngu' in run_data['name']]\n",
    "fetch_ngu_data = [ run_data for run_data in all_runs_data if 'fetch' in run_data['name'] and 'ngu' in run_data['name']]\n",
    "finger_ngu_data = [ run_data for run_data in all_runs_data if 'finger' in run_data['name'] and 'ngu' in run_data['name']]\n",
    "fetch_push_ngu_data = [ run_data for run_data in all_runs_data if 'fetch_push' in run_data['name'] and 'ngu' in run_data['name']]\n",
    "hopper_ngu_data = [ run_data for run_data in all_runs_data if 'hopper' in run_data['name'] and 'ngu' in run_data['name']]\n",
    "# lsd\n",
    "Easy_lsd_data = [ run_data for run_data in all_runs_data if 'Easy' in run_data['name']and 'lsd' in run_data['name']]\n",
    "Ur_lsd_data = [ run_data for run_data in all_runs_data if 'Ur' in run_data['name'] and 'lsd' in run_data['name']]\n",
    "Hard_lsd_data = [ run_data for run_data in all_runs_data if 'Hard' in run_data['name'] and 'lsd' in run_data['name']]\n",
    "fetch_lsd_data = [ run_data for run_data in all_runs_data if 'fetch' in run_data['name'] and 'lsd' in run_data['name']]\n",
    "finger_lsd_data = [ run_data for run_data in all_runs_data if 'finger' in run_data['name'] and 'lsd' in run_data['name']]\n",
    "fetch_push_lsd_data = [ run_data for run_data in all_runs_data if 'fetch_push' in run_data['name'] and 'lsd' in run_data['name']]\n",
    "hopper_lsd_data = [ run_data for run_data in all_runs_data if 'hopper' in run_data['name'] and 'lsd' in run_data['name']]\n",
    "# csd \n",
    "Easy_csd_data = [ run_data for run_data in all_runs_data if 'Easy' in run_data['name']and 'csd' in run_data['name']]\n",
    "Ur_csd_data = [ run_data for run_data in all_runs_data if 'Ur' in run_data['name'] and 'csd' in run_data['name']]\n",
    "Hard_csd_data = [ run_data for run_data in all_runs_data if 'Hard' in run_data['name'] and 'csd' in run_data['name']]\n",
    "fetch_csd_data = [ run_data for run_data in all_runs_data if 'fetch' in run_data['name'] and 'csd' in run_data['name']]\n",
    "finger_csd_data = [ run_data for run_data in all_runs_data if 'finger' in run_data['name'] and 'csd' in run_data['name']]\n",
    "fetch_push_csd_data = [ run_data for run_data in all_runs_data if 'fetch_push' in run_data['name'] and 'csd' in run_data['name']]\n",
    "# metra\n",
    "Easy_metra_data = [ run_data for run_data in all_runs_data if 'Easy' in run_data['name']and 'metra' in run_data['name']]\n",
    "Ur_metra_data = [ run_data for run_data in all_runs_data if 'Ur' in run_data['name'] and 'metra' in run_data['name']]\n",
    "Hard_metra_data = [ run_data for run_data in all_runs_data if 'Hard' in run_data['name'] and 'metra' in run_data['name']]\n",
    "fetch_metra_data = [ run_data for run_data in all_runs_data if 'fetch' in run_data['name'] and 'metra' in run_data['name']]\n",
    "finger_metra_data = [ run_data for run_data in all_runs_data if 'finger' in run_data['name'] and 'metra' in run_data['name']]\n",
    "fetch_push_metra_data = [ run_data for run_data in all_runs_data if 'fetch_push' in run_data['name'] and 'metra' in run_data['name']]\n",
    "# smm\n",
    "Easy_smm_data = [ run_data for run_data in all_runs_data if 'Easy' in run_data['name']and 'smm' in run_data['name']]\n",
    "Ur_smm_data = [ run_data for run_data in all_runs_data if 'Ur' in run_data['name'] and 'smm' in run_data['name']]\n",
    "Hard_smm_data = [ run_data for run_data in all_runs_data if 'Hard' in run_data['name'] and 'smm' in run_data['name']]\n",
    "fetch_smm_data = [ run_data for run_data in all_runs_data if 'fetch' in run_data['name'] and 'smm' in run_data['name']]\n",
    "finger_smm_data = [ run_data for run_data in all_runs_data if 'finger' in run_data['name'] and 'smm' in run_data['name']]\n",
    "fetch_push_smm_data = [ run_data for run_data in all_runs_data if 'fetch_push' in run_data['name'] and 'smm' in run_data['name']]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 461,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fill_to_size(arr, size):\n",
    "    \"\"\" fill array to size with last element \"\"\"\n",
    "    arr = np.expand_dims(arr, axis=1)\n",
    "    while len(arr) < size:\n",
    "        arr = np.vstack([arr, arr[-1]])\n",
    "    return arr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 462,
   "metadata": {},
   "outputs": [],
   "source": [
    "# env to do\n",
    "list_env = ['Easy', 'Ur', 'Hard', 'fetch', 'finger', 'fetch_push']\n",
    "list_alg = ['lds', 'diayn', 'rnd', 'ngu', 'lsd', 'csd', 'metra', 'smm']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 463,
   "metadata": {},
   "outputs": [],
   "source": [
    "# lds\n",
    "Easy_lds_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Easy_lds_data]) if 'Easy' in list_env and 'lds' in list_alg else None\n",
    "Ur_lds_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Ur_lds_data]) if 'Ur' in list_env and 'lds' in list_alg else None\n",
    "Hard_lds_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Hard_lds_data]) if 'Hard' in list_env and 'lds' in list_alg else None\n",
    "fetch_lds_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_lds_data]) if 'fetch' in list_env and 'lds' in list_alg else None\n",
    "finger_lds_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in finger_lds_data]) if 'finger' in list_env and 'lds' in list_alg else None\n",
    "fetch_push_lds_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_push_lds_data]) if 'fetch_push' in list_env and 'lds' in list_alg else None\n",
    "hopper_lds_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in hopper_lds_data]) if 'hopper' in list_env and 'lds' in list_alg else None\n",
    "# diayn\n",
    "Easy_diayn_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Easy_diayn_data]) if 'Easy' in list_env and 'diayn' in list_alg else None\n",
    "Ur_diayn_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Ur_diayn_data]) if 'Ur' in list_env and 'diayn' in list_alg else None\n",
    "Hard_diayn_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Hard_diayn_data]) if 'Hard' in list_env and 'diayn' in list_alg else None\n",
    "fetch_diayn_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_diayn_data]) if 'fetch' in list_env and 'diayn' in list_alg else None\n",
    "finger_diayn_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in finger_diayn_data]) if 'finger' in list_env and 'diayn' in list_alg  else None\n",
    "fetch_push_diayn_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_push_diayn_data]) if 'fetch_push' in list_env and 'diayn' in list_alg  else None\n",
    "hopper_diayn_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in hopper_diayn_data]) if 'hopper' in list_env and 'diayn' in list_alg  else None\n",
    "# rnd\n",
    "Easy_rnd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Easy_rnd_data]) if 'Easy' in list_env and 'rnd' in list_alg  else None\n",
    "Ur_rnd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Ur_rnd_data]) if 'Ur' in list_env and 'rnd' in list_alg  else None\n",
    "Hard_rnd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Hard_rnd_data]) if 'Hard' in list_env and 'rnd' in list_alg  else None\n",
    "fetch_rnd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_rnd_data]) if 'fetch' in list_env and 'rnd' in list_alg  else None\n",
    "finger_rnd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in finger_rnd_data]) if 'finger' in list_env and 'rnd' in list_alg  else None\n",
    "fetch_push_rnd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_push_rnd_data]) if 'fetch_push' in list_env and 'rnd' in list_alg  else None\n",
    "hopper_rnd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in hopper_rnd_data]) if 'hopper' in list_env and 'rnd' in list_alg  else None\n",
    "# ngu \n",
    "Easy_ngu_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Easy_ngu_data]) if 'Easy' in list_env and 'ngu' in list_alg  else None\n",
    "Ur_ngu_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Ur_ngu_data]) if 'Ur' in list_env and 'ngu' in list_alg  else None\n",
    "Hard_ngu_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Hard_ngu_data]) if 'Hard' in list_env and 'ngu' in list_alg  else None\n",
    "fetch_ngu_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_ngu_data]) if 'fetch' in list_env and 'ngu' in list_alg  else None\n",
    "finger_ngu_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in finger_ngu_data]) if 'finger' in list_env and 'ngu' in list_alg  else None\n",
    "fetch_push_ngu_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_push_ngu_data]) if 'fetch_push' in list_env and 'ngu' in list_alg  else None\n",
    "hopper_ngu_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in hopper_ngu_data]) if 'hopper' in list_env and 'ngu' in list_alg  else None\n",
    "# lsd\n",
    "Easy_lsd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Easy_lsd_data]) if 'Easy' in list_env and 'lsd' in list_alg  else None\n",
    "Ur_lsd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Ur_lsd_data]) if 'Ur' in list_env and 'lsd' in list_alg  else None\n",
    "Hard_lsd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Hard_lsd_data]) if 'Hard' in list_env and 'lsd' in list_alg  else None\n",
    "fetch_lsd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_lsd_data]) if 'fetch' in list_env and 'lsd' in list_alg  else None\n",
    "finger_lsd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in finger_lsd_data]) if 'finger' in list_env and 'lsd' in list_alg  else None\n",
    "fetch_push_lsd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_push_lsd_data]) if 'fetch_push' in list_env and 'lsd' in list_alg  else None\n",
    "hopper_lsd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in hopper_lsd_data]) if 'hopper' in list_env and 'lsd' in list_alg  else None\n",
    "# csd\n",
    "Easy_csd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Easy_csd_data]) if 'Easy' in list_env and 'csd' in list_alg  else None\n",
    "Ur_csd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Ur_csd_data]) if 'Ur' in list_env and 'csd' in list_alg  else None\n",
    "Hard_csd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Hard_csd_data]) if 'Hard' in list_env and 'csd' in list_alg  else None\n",
    "fetch_csd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_csd_data]) if 'fetch' in list_env and 'csd' in list_alg  else None\n",
    "finger_csd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in finger_csd_data]) if 'finger' in list_env and 'csd' in list_alg  else None\n",
    "fetch_push_csd_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_push_csd_data]) if 'fetch_push' in list_env and 'csd' in list_alg  else None\n",
    "# metra\n",
    "Easy_metra_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Easy_metra_data]) if 'Easy' in list_env and 'metra' in list_alg  else None\n",
    "Ur_metra_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Ur_metra_data]) if 'Ur' in list_env and 'metra' in list_alg  else None\n",
    "Hard_metra_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Hard_metra_data]) if 'Hard' in list_env and 'metra' in list_alg  else None\n",
    "fetch_metra_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_metra_data]) if 'fetch' in list_env and 'metra' in list_alg  else None\n",
    "finger_metra_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in finger_metra_data]) if 'finger' in list_env and 'metra' in list_alg  else None\n",
    "fetch_push_metra_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_push_metra_data]) if 'fetch_push' in list_env and 'metra' in list_alg  else None\n",
    "# smm\n",
    "Easy_smm_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Easy_smm_data]) if 'Easy' in list_env and 'smm' in list_alg  else None\n",
    "Ur_smm_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Ur_smm_data]) if 'Ur' in list_env and 'smm' in list_alg  else None\n",
    "Hard_smm_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in Hard_smm_data]) if 'Hard' in list_env and 'smm' in list_alg  else None\n",
    "fetch_smm_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_smm_data]) if 'fetch' in list_env and 'smm' in list_alg  else None\n",
    "finger_smm_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in finger_smm_data]) if 'finger' in list_env and 'smm' in list_alg  else None\n",
    "fetch_push_smm_max_step = max([len(run_data['metrics']['coverage']) if 'coverage' in run_data['metrics'].keys() else None  for run_data in fetch_push_smm_data]) if 'fetch_push' in list_env and 'smm' in list_alg  else None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 464,
   "metadata": {},
   "outputs": [],
   "source": [
    "def delete_none(liste):\n",
    "    return list(filter(None.__ne__, liste))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 465,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_7743/1915478931.py:2: DeprecationWarning: NotImplemented should not be used in a boolean context\n",
      "  return list(filter(None.__ne__, liste))\n"
     ]
    }
   ],
   "source": [
    "\n",
    "Easy_lds_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Easy_lds_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Easy_lds_data]), axis=1) if 'Easy' in list_env and 'lds' in list_alg else None\n",
    "Ur_lds_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Ur_lds_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Ur_lds_data]), axis=1) if 'Ur' in list_env and 'lds' in list_alg else None\n",
    "Hard_lds_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Hard_lds_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Hard_lds_data]), axis=1) if 'Hard' in list_env and 'lds' in list_alg else None    \n",
    "fetch_lds_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_lds_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_lds_data]), axis=1) if 'fetch' in list_env and 'lds' in list_alg else None \n",
    "finger_lds_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, finger_lds_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in finger_lds_data]), axis=1) if 'finger' in list_env and 'lds' in list_alg else None\n",
    "fetch_push_lds_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_push_lds_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_push_lds_data]), axis=1) if 'fetch_push' in list_env and 'lds' in list_alg else None\n",
    "hopper_lds_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, hopper_lds_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in hopper_lds_data]), axis=1) if 'hopper' in list_env and 'lds' in list_alg else None\n",
    "# diayn\n",
    "Easy_diayn_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Easy_diayn_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Easy_diayn_data]), axis=1) if 'Easy' in list_env and 'diayn' in list_alg else None\n",
    "Ur_diayn_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Ur_diayn_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Ur_diayn_data]), axis=1) if 'Ur' in list_env and 'diayn' in list_alg else None\n",
    "Hard_diayn_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Hard_diayn_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Hard_diayn_data]), axis=1) if 'Hard' in list_env and 'diayn' in list_alg else None\n",
    "fetch_diayn_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_diayn_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_diayn_data]), axis=1) if 'fetch' in list_env and 'diayn' in list_alg else None\n",
    "finger_diayn_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, finger_diayn_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in finger_diayn_data]), axis=1) if 'finger' in list_env and 'diayn' in list_alg else None\n",
    "fetch_push_diayn_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_push_diayn_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_push_diayn_data]), axis=1) if 'fetch_push' in list_env and 'diayn' in list_alg else None\n",
    "hopper_diayn_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, hopper_diayn_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in hopper_diayn_data]), axis=1) if 'hopper' in list_env and 'diayn' in list_alg else None\n",
    "# rnd\n",
    "Easy_rnd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Easy_rnd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Easy_rnd_data]), axis=1) if 'Easy' in list_env and 'rnd' in list_alg else None\n",
    "Ur_rnd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Ur_rnd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Ur_rnd_data]), axis=1) if 'Ur' in list_env and 'rnd' in list_alg else None\n",
    "Hard_rnd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Hard_rnd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Hard_rnd_data]), axis=1) if 'Hard' in list_env and 'rnd' in list_alg else None\n",
    "fetch_rnd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_rnd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_rnd_data]), axis=1) if 'fetch' in list_env and 'rnd' in list_alg else None\n",
    "finger_rnd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, finger_rnd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in finger_rnd_data]), axis=1) if 'finger' in list_env and 'rnd' in list_alg else None\n",
    "fetch_push_rnd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_push_rnd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_push_rnd_data]), axis=1) if 'fetch_push' in list_env and 'rnd' in list_alg else None\n",
    "hopper_rnd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, hopper_rnd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in hopper_rnd_data]), axis=1) if 'hopper' in list_env and 'rnd' in list_alg else None\n",
    "# ngu\n",
    "Easy_ngu_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Easy_ngu_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Easy_ngu_data]), axis=1) if 'Easy' in list_env and 'ngu' in list_alg else None\n",
    "Ur_ngu_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Ur_ngu_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Ur_ngu_data]), axis=1) if 'Ur' in list_env and 'ngu' in list_alg else None\n",
    "Hard_ngu_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Hard_ngu_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Hard_ngu_data]), axis=1) if 'Hard' in list_env and 'ngu' in list_alg else None\n",
    "fetch_ngu_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_ngu_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_ngu_data]), axis=1) if 'fetch' in list_env and 'ngu' in list_alg else None\n",
    "finger_ngu_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, finger_ngu_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in finger_ngu_data]), axis=1) if 'finger' in list_env and 'ngu' in list_alg else None\n",
    "fetch_push_ngu_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_push_ngu_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_push_ngu_data]), axis=1) if 'fetch_push' in list_env and 'ngu' in list_alg else None\n",
    "hopper_ngu_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, hopper_ngu_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in hopper_ngu_data]), axis=1) if 'hopper' in list_env and 'ngu' in list_alg else None\n",
    "# lsd\n",
    "Easy_lsd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Easy_lsd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Easy_lsd_data]), axis=1) if 'Easy' in list_env and 'lsd' in list_alg else None\n",
    "Ur_lsd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Ur_lsd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Ur_lsd_data]), axis=1) if 'Ur' in list_env and 'lsd' in list_alg else None\n",
    "Hard_lsd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Hard_lsd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Hard_lsd_data]), axis=1) if 'Hard' in list_env and 'lsd' in list_alg else None\n",
    "fetch_lsd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_lsd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_lsd_data]), axis=1) if 'fetch' in list_env and 'lsd' in list_alg else None\n",
    "finger_lsd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, finger_lsd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in finger_lsd_data]), axis=1) if 'finger' in list_env and 'lsd' in list_alg else None\n",
    "fetch_push_lsd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_push_lsd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_push_lsd_data]), axis=1) if 'fetch_push' in list_env and 'lsd' in list_alg else None\n",
    "hopper_lsd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, hopper_lsd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in hopper_lsd_data]), axis=1) if 'hopper' in list_env and 'lsd' in list_alg else None\n",
    "# csd\n",
    "Easy_csd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Easy_csd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Easy_csd_data]), axis=1) if 'Easy' in list_env and 'csd' in list_alg else None\n",
    "Ur_csd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Ur_csd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Ur_csd_data]), axis=1) if 'Ur' in list_env and 'csd' in list_alg else None\n",
    "Hard_csd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Hard_csd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Hard_csd_data]), axis=1) if 'Hard' in list_env and 'csd' in list_alg else None\n",
    "fetch_csd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_csd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_csd_data]), axis=1) if 'fetch' in list_env and 'csd' in list_alg else None\n",
    "finger_csd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, finger_csd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in finger_csd_data]), axis=1) if 'finger' in list_env and 'csd' in list_alg else None\n",
    "fetch_push_csd_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_push_csd_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_push_csd_data]), axis=1) if 'fetch_push' in list_env and 'csd' in list_alg else None\n",
    "# metra\n",
    "Easy_metra_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Easy_metra_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Easy_metra_data]), axis=1) if 'Easy' in list_env and 'metra' in list_alg else None\n",
    "Ur_metra_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Ur_metra_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Ur_metra_data]), axis=1) if 'Ur' in list_env and 'metra' in list_alg else None\n",
    "Hard_metra_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Hard_metra_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Hard_metra_data]), axis=1) if 'Hard' in list_env and 'metra' in list_alg else None\n",
    "fetch_metra_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_metra_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_metra_data]), axis=1) if 'fetch' in list_env and 'metra' in list_alg else None\n",
    "finger_metra_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, finger_metra_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in finger_metra_data]), axis=1) if 'finger' in list_env and 'metra' in list_alg else None\n",
    "fetch_push_metra_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_push_metra_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_push_metra_data]), axis=1) if 'fetch_push' in list_env and 'metra' in list_alg else None\n",
    "# smm\n",
    "Easy_smm_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Easy_smm_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Easy_smm_data]), axis=1) if 'Easy' in list_env and 'smm' in list_alg else None\n",
    "Ur_smm_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Ur_smm_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Ur_smm_data]), axis=1) if 'Ur' in list_env and 'smm' in list_alg else None\n",
    "Hard_smm_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, Hard_smm_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in Hard_smm_data]), axis=1) if 'Hard' in list_env and 'smm' in list_alg else None\n",
    "fetch_smm_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_smm_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_smm_data]), axis=1) if 'fetch' in list_env and 'smm' in list_alg else None\n",
    "finger_smm_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, finger_smm_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in finger_smm_data]), axis=1) if 'finger' in list_env and 'smm' in list_alg else None\n",
    "fetch_push_smm_coverage_stack = np.concatenate(delete_none([fill_to_size(run_data['metrics']['coverage'].values, fetch_push_smm_max_step) if 'coverage' in run_data['metrics'].keys() else None for run_data in fetch_push_smm_data]), axis=1) if 'fetch_push' in list_env and 'smm' in list_alg else None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 466,
   "metadata": {},
   "outputs": [],
   "source": [
    "nb_runs = 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 467,
   "metadata": {},
   "outputs": [],
   "source": [
    "# mean and std \n",
    "# lds\n",
    "Easy_lds_coverage_mean = np.mean(Easy_lds_coverage_stack, axis=1) if 'Easy' in list_env and 'lds' in list_alg else None\n",
    "Easy_lds_coverage_std = np.std(Easy_lds_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Easy' in list_env and 'lds' in list_alg else None\n",
    "Ur_lds_coverage_mean = np.mean(Ur_lds_coverage_stack, axis=1) if 'Ur' in list_env and 'lds' in list_alg else None\n",
    "Ur_lds_coverage_std = np.std(Ur_lds_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Ur' in list_env and 'lds' in list_alg else None\n",
    "Hard_lds_coverage_mean = np.mean(Hard_lds_coverage_stack, axis=1) if 'Hard' in list_env and 'lds' in list_alg else None\n",
    "Hard_lds_coverage_std = np.std(Hard_lds_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Hard' in list_env and 'lds' in list_alg else None\n",
    "fetch_lds_coverage_std = np.std(fetch_lds_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch' in list_env and 'lds' in list_alg else None\n",
    "fetch_lds_coverage_mean = np.mean(fetch_lds_coverage_stack, axis=1) if 'fetch' in list_env and 'lds' in list_alg else None\n",
    "finger_lds_coverage_mean = np.mean(finger_lds_coverage_stack, axis=1) if 'finger' in list_env and 'lds' in list_alg else None\n",
    "finger_lds_coverage_std = np.std(finger_lds_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'finger' in list_env and 'lds' in list_alg else None\n",
    "fetch_push_lds_coverage_mean = np.mean(fetch_push_lds_coverage_stack, axis=1) if 'fetch_push' in list_env and 'lds' in list_alg else None\n",
    "fetch_push_lds_coverage_std = np.std(fetch_push_lds_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch_push' in list_env and 'lds' in list_alg else None\n",
    "hopper_lds_coverage_mean = np.mean(hopper_lds_coverage_stack, axis=1) if 'hopper' in list_env and 'lds' in list_alg else None\n",
    "hopper_lds_coverage_std = np.std(hopper_lds_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'hopper' in list_env and 'lds' in list_alg else None\n",
    "# diayn \n",
    "Easy_diayn_coverage_mean = np.mean(Easy_diayn_coverage_stack, axis=1) if 'Easy' in list_env and 'diayn' in list_alg else None\n",
    "Easy_diayn_coverage_std = np.std(Easy_diayn_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Easy' in list_env and 'diayn' in list_alg else None\n",
    "Ur_diayn_coverage_mean = np.mean(Ur_diayn_coverage_stack, axis=1) if 'Ur' in list_env and 'diayn' in list_alg else None\n",
    "Ur_diayn_coverage_std = np.std(Ur_diayn_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Ur' in list_env and 'diayn' in list_alg else None\n",
    "Hard_diayn_coverage_mean = np.mean(Hard_diayn_coverage_stack, axis=1) if 'Hard' in list_env and 'diayn' in list_alg else None\n",
    "Hard_diayn_coverage_std = np.std(Hard_diayn_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Hard' in list_env and 'diayn' in list_alg else None\n",
    "fetch_diayn_coverage_std = np.std(fetch_diayn_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch' in list_env and 'diayn' in list_alg else None\n",
    "fetch_diayn_coverage_mean = np.mean(fetch_diayn_coverage_stack, axis=1) if 'fetch' in list_env and 'diayn' in list_alg else None\n",
    "finger_diayn_coverage_mean = np.mean(finger_diayn_coverage_stack, axis=1) if 'finger' in list_env and 'diayn' in list_alg else None\n",
    "finger_diayn_coverage_std = np.std(finger_diayn_coverage_stack, axis=1)/np.sqrt(nb_runs)   if 'finger' in list_env and 'diayn' in list_alg else None\n",
    "fetch_push_diayn_coverage_mean = np.mean(fetch_push_diayn_coverage_stack, axis=1) if 'fetch_push' in list_env and 'diayn' in list_alg else None\n",
    "fetch_push_diayn_coverage_std = np.std(fetch_push_diayn_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch_push' in list_env and 'diayn' in list_alg else None\n",
    "hopper_diayn_coverage_mean = np.mean(hopper_diayn_coverage_stack, axis=1) if 'hopper' in list_env and 'diayn' in list_alg else None\n",
    "hopper_diayn_coverage_std = np.std(hopper_diayn_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'hopper' in list_env and 'diayn' in list_alg else None\n",
    "# rnd\n",
    "Easy_rnd_coverage_mean = np.mean(Easy_rnd_coverage_stack, axis=1)[:150] if 'Easy' in list_env and 'rnd' in list_alg else None\n",
    "Easy_rnd_coverage_std = np.std(Easy_rnd_coverage_stack, axis=1)[:150]/np.sqrt(nb_runs) if 'Easy' in list_env and 'rnd' in list_alg else None\n",
    "Ur_rnd_coverage_mean = np.mean(Ur_rnd_coverage_stack, axis=1)[:150] if 'Ur' in list_env and 'rnd' in list_alg else None\n",
    "Ur_rnd_coverage_std = np.std(Ur_rnd_coverage_stack, axis=1)[:150]/np.sqrt(nb_runs) if 'Ur' in list_env and 'rnd' in list_alg else None\n",
    "Hard_rnd_coverage_mean = np.mean(Hard_rnd_coverage_stack, axis=1)[:150] if 'Hard' in list_env and 'rnd' in list_alg else None\n",
    "Hard_rnd_coverage_std = np.std(Hard_rnd_coverage_stack, axis=1)[:150]/np.sqrt(nb_runs) if 'Hard' in list_env and 'rnd' in list_alg else None\n",
    "fetch_rnd_coverage_std = np.std(fetch_rnd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch' in list_env and 'rnd' in list_alg else None\n",
    "fetch_rnd_coverage_mean = np.mean(fetch_rnd_coverage_stack, axis=1) if 'fetch' in list_env and 'rnd' in list_alg else None\n",
    "finger_rnd_coverage_mean = np.mean(finger_rnd_coverage_stack, axis=1) if 'finger' in list_env and 'rnd' in list_alg else None\n",
    "finger_rnd_coverage_std = np.std(finger_rnd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'finger' in list_env and 'rnd' in list_alg else None\n",
    "fetch_push_rnd_coverage_mean = np.mean(fetch_push_rnd_coverage_stack, axis=1) if 'fetch_push' in list_env and 'rnd' in list_alg else None\n",
    "fetch_push_rnd_coverage_std = np.std(fetch_push_rnd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch_push' in list_env and 'rnd' in list_alg else None\n",
    "hopper_rnd_coverage_mean = np.mean(hopper_rnd_coverage_stack, axis=1) if 'hopper' in list_env and 'rnd' in list_alg else None\n",
    "hopper_rnd_coverage_std = np.std(hopper_rnd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'hopper' in list_env and 'rnd' in list_alg else None\n",
    "# ngu\n",
    "Easy_ngu_coverage_mean = np.mean(Easy_ngu_coverage_stack, axis=1) if 'Easy' in list_env and 'ngu' in list_alg else None\n",
    "Easy_ngu_coverage_std = np.std(Easy_ngu_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Easy' in list_env and 'ngu' in list_alg else None\n",
    "Ur_ngu_coverage_mean = np.mean(Ur_ngu_coverage_stack, axis=1) if 'Ur' in list_env and 'ngu' in list_alg else None\n",
    "Ur_ngu_coverage_std = np.std(Ur_ngu_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Ur' in list_env and 'ngu' in list_alg else None\n",
    "Hard_ngu_coverage_mean = np.mean(Hard_ngu_coverage_stack, axis=1) if 'Hard' in list_env and 'ngu' in list_alg else None\n",
    "Hard_ngu_coverage_std = np.std(Hard_ngu_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Hard' in list_env and 'ngu' in list_alg else None\n",
    "fetch_ngu_coverage_std = np.std(fetch_ngu_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch' in list_env and 'ngu' in list_alg else None\n",
    "fetch_ngu_coverage_mean = np.mean(fetch_ngu_coverage_stack, axis=1) if 'fetch' in list_env and 'ngu' in list_alg else None\n",
    "finger_ngu_coverage_mean = np.mean(finger_ngu_coverage_stack, axis=1) if 'finger' in list_env and 'ngu' in list_alg else None\n",
    "finger_ngu_coverage_std = np.std(finger_ngu_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'finger' in list_env and 'ngu' in list_alg else None\n",
    "fetch_push_ngu_coverage_mean = np.mean(fetch_push_ngu_coverage_stack, axis=1) if 'fetch_push' in list_env and 'ngu' in list_alg else None\n",
    "fetch_push_ngu_coverage_std = np.std(fetch_push_ngu_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch_push' in list_env and 'ngu' in list_alg else None\n",
    "hopper_ngu_coverage_mean = np.mean(hopper_ngu_coverage_stack, axis=1) if 'hopper' in list_env and 'ngu' in list_alg else None\n",
    "hopper_ngu_coverage_std = np.std(hopper_ngu_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'hopper' in list_env and 'ngu' in list_alg else None\n",
    "# lsd\n",
    "Easy_lsd_coverage_mean = np.mean(Easy_lsd_coverage_stack, axis=1) if 'Easy' in list_env and 'lsd' in list_alg else None\n",
    "Easy_lsd_coverage_std = np.std(Easy_lsd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Easy' in list_env and 'lsd' in list_alg else None\n",
    "Ur_lsd_coverage_mean = np.mean(Ur_lsd_coverage_stack, axis=1) if 'Ur' in list_env and 'lsd' in list_alg else None\n",
    "Ur_lsd_coverage_std = np.std(Ur_lsd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Ur' in list_env and 'lsd' in list_alg else None\n",
    "Hard_lsd_coverage_mean = np.mean(Hard_lsd_coverage_stack, axis=1) if 'Hard' in list_env and 'lsd' in list_alg else None\n",
    "Hard_lsd_coverage_std = np.std(Hard_lsd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Hard' in list_env and 'lsd' in list_alg else None\n",
    "fetch_lsd_coverage_std = np.std(fetch_lsd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch' in list_env and 'lsd' in list_alg else None\n",
    "fetch_lsd_coverage_mean = np.mean(fetch_lsd_coverage_stack, axis=1) if 'fetch' in list_env and 'lsd' in list_alg else None\n",
    "finger_lsd_coverage_mean = np.mean(finger_lsd_coverage_stack, axis=1) if 'finger' in list_env and 'lsd' in list_alg else None\n",
    "finger_lsd_coverage_std = np.std(finger_lsd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'finger' in list_env and 'lsd' in list_alg else None\n",
    "fetch_push_lsd_coverage_mean = np.mean(fetch_push_lsd_coverage_stack, axis=1) if 'fetch_push' in list_env and 'lsd' in list_alg else None\n",
    "fetch_push_lsd_coverage_std = np.std(fetch_push_lsd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch_push' in list_env and 'lsd' in list_alg else None\n",
    "hopper_lsd_coverage_mean = np.mean(hopper_lsd_coverage_stack, axis=1) if 'hopper' in list_env and 'lsd' in list_alg else None\n",
    "hopper_lsd_coverage_std = np.std(hopper_lsd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'hopper' in list_env and 'lsd' in list_alg else None\n",
    "# csd\n",
    "Easy_csd_coverage_mean = np.mean(Easy_csd_coverage_stack, axis=1) if 'Easy' in list_env and 'csd' in list_alg else None\n",
    "Easy_csd_coverage_std = np.std(Easy_csd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Easy' in list_env and 'csd' in list_alg else None\n",
    "Ur_csd_coverage_mean = np.mean(Ur_csd_coverage_stack, axis=1) if 'Ur' in list_env and 'csd' in list_alg else None\n",
    "Ur_csd_coverage_std = np.std(Ur_csd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Ur' in list_env and 'csd' in list_alg else None\n",
    "Hard_csd_coverage_mean = np.mean(Hard_csd_coverage_stack, axis=1) if 'Hard' in list_env and 'csd' in list_alg else None\n",
    "Hard_csd_coverage_std = np.std(Hard_csd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Hard' in list_env and 'csd' in list_alg else None\n",
    "fetch_csd_coverage_std = np.std(fetch_csd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch' in list_env and 'csd' in list_alg else None\n",
    "fetch_csd_coverage_mean = np.mean(fetch_csd_coverage_stack, axis=1) if 'fetch' in list_env and 'csd' in list_alg else None\n",
    "finger_csd_coverage_mean = np.mean(finger_csd_coverage_stack, axis=1) if 'finger' in list_env and 'csd' in list_alg else None\n",
    "finger_csd_coverage_std = np.std(finger_csd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'finger' in list_env and 'csd' in list_alg else None\n",
    "fetch_push_csd_coverage_mean = np.mean(fetch_push_csd_coverage_stack, axis=1) if 'fetch_push' in list_env and 'csd' in list_alg else None\n",
    "fetch_push_csd_coverage_std = np.std(fetch_push_csd_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch_push' in list_env and 'csd' in list_alg else None\n",
    "# metra\n",
    "Easy_metra_coverage_mean = np.mean(Easy_metra_coverage_stack, axis=1) if 'Easy' in list_env and 'metra' in list_alg else None\n",
    "Easy_metra_coverage_std = np.std(Easy_metra_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Easy' in list_env and 'metra' in list_alg else None\n",
    "Ur_metra_coverage_mean = np.mean(Ur_metra_coverage_stack, axis=1) if 'Ur' in list_env and 'metra' in list_alg else None\n",
    "Ur_metra_coverage_std = np.std(Ur_metra_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Ur' in list_env and 'metra' in list_alg else None\n",
    "Hard_metra_coverage_mean = np.mean(Hard_metra_coverage_stack, axis=1) if 'Hard' in list_env and 'metra' in list_alg else None\n",
    "Hard_metra_coverage_std = np.std(Hard_metra_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Hard' in list_env and 'metra' in list_alg else None\n",
    "fetch_metra_coverage_std = np.std(fetch_metra_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch' in list_env and 'metra' in list_alg else None\n",
    "fetch_metra_coverage_mean = np.mean(fetch_metra_coverage_stack, axis=1) if 'fetch' in list_env and 'metra' in list_alg else None\n",
    "finger_metra_coverage_mean = np.mean(finger_metra_coverage_stack, axis=1) if 'finger' in list_env and 'metra' in list_alg else None\n",
    "finger_metra_coverage_std = np.std(finger_metra_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'finger' in list_env and 'metra' in list_alg else None\n",
    "fetch_push_metra_coverage_mean = np.mean(fetch_push_metra_coverage_stack, axis=1) if 'fetch_push' in list_env and 'metra' in list_alg else None\n",
    "fetch_push_metra_coverage_std = np.std(fetch_push_metra_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch_push' in list_env and 'metra' in list_alg else None\n",
    "# smm\n",
    "Easy_smm_coverage_mean = np.mean(Easy_smm_coverage_stack, axis=1) if 'Easy' in list_env and 'smm' in list_alg else None\n",
    "Easy_smm_coverage_std = np.std(Easy_smm_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Easy' in list_env and 'smm' in list_alg else None\n",
    "Ur_smm_coverage_mean = np.mean(Ur_smm_coverage_stack, axis=1) if 'Ur' in list_env and 'smm' in list_alg else None\n",
    "Ur_smm_coverage_std = np.std(Ur_smm_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Ur' in list_env and 'smm' in list_alg else None\n",
    "Hard_smm_coverage_mean = np.mean(Hard_smm_coverage_stack, axis=1) if 'Hard' in list_env and 'smm' in list_alg else None\n",
    "Hard_smm_coverage_std = np.std(Hard_smm_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'Hard' in list_env and 'smm' in list_alg else None\n",
    "fetch_smm_coverage_std = np.std(fetch_smm_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch' in list_env and 'smm' in list_alg else None\n",
    "fetch_smm_coverage_mean = np.mean(fetch_smm_coverage_stack, axis=1) if 'fetch' in list_env and 'smm' in list_alg else None\n",
    "finger_smm_coverage_mean = np.mean(finger_smm_coverage_stack, axis=1) if 'finger' in list_env and 'smm' in list_alg else None\n",
    "finger_smm_coverage_std = np.std(finger_smm_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'finger' in list_env and 'smm' in list_alg else None\n",
    "fetch_push_smm_coverage_mean = np.mean(fetch_push_smm_coverage_stack, axis=1) if 'fetch_push' in list_env and 'smm' in list_alg else None\n",
    "fetch_push_smm_coverage_std = np.std(fetch_push_smm_coverage_stack, axis=1)/np.sqrt(nb_runs) if 'fetch_push' in list_env and 'smm' in list_alg else None\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 468,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tar \n",
    "if 'Easy' in list_env and 'lds' in list_alg : \n",
    "    Easy_lds_coverage_mean[0] = 0\n",
    "if 'Ur' in list_env and 'lds' in list_alg : \n",
    "    Ur_lds_coverage_mean[0] = 0\n",
    "if 'Hard' in list_env and 'lds' in list_alg : \n",
    "    Hard_lds_coverage_mean[0] = 0\n",
    "if 'fetch' in list_env and 'lds' in list_alg : \n",
    "    fetch_lds_coverage_mean[0] = 0\n",
    "if 'finger' in list_env and 'lds' in list_alg : \n",
    "    finger_lds_coverage_mean[0] = 0\n",
    "if 'fetch_push' in list_env and 'lds' in list_alg : \n",
    "    fetch_push_lds_coverage_mean[0] = 0\n",
    "if 'hopper' in list_env and 'lds' in list_alg : \n",
    "    hopper_lds_coverage_mean[0] = 0\n",
    "# diayn\n",
    "if 'Easy' in list_env and 'diayn' in list_alg : \n",
    "    Easy_diayn_coverage_mean[0] = 0\n",
    "if 'Ur' in list_env and 'diayn' in list_alg : \n",
    "    Ur_diayn_coverage_mean[0] = 0\n",
    "if 'Hard' in list_env and 'diayn' in list_alg : \n",
    "    Hard_diayn_coverage_mean[0] = 0\n",
    "if 'fetch' in list_env and 'diayn' in list_alg : \n",
    "    fetch_diayn_coverage_mean[0] = 0\n",
    "if 'finger' in list_env and 'diayn' in list_alg : \n",
    "    finger_diayn_coverage_mean[0] = 0\n",
    "if 'fetch_push' in list_env and 'diayn' in list_alg : \n",
    "    fetch_push_diayn_coverage_mean[0] = 0\n",
    "if 'hopper' in list_env and 'diayn' in list_alg : \n",
    "    hopper_diayn_coverage_mean[0] = 0\n",
    "\n",
    "# rnd\n",
    "if 'Easy' in list_env and 'rnd' in list_alg : \n",
    "    Easy_rnd_coverage_mean[0] = 0\n",
    "if 'Ur' in list_env and 'rnd' in list_alg : \n",
    "    Ur_rnd_coverage_mean[0] = 0\n",
    "if 'Hard' in list_env and 'rnd' in list_alg : \n",
    "    Hard_rnd_coverage_mean[0] = 0\n",
    "if 'fetch' in list_env and 'rnd' in list_alg : \n",
    "    fetch_rnd_coverage_mean[0] = 0\n",
    "if 'finger' in list_env and 'rnd' in list_alg : \n",
    "    finger_rnd_coverage_mean[0] = 0\n",
    "if 'fetch_push' in list_env and 'rnd' in list_alg : \n",
    "    fetch_push_rnd_coverage_mean[0] = 0\n",
    "if 'hopper' in list_env and 'rnd' in list_alg : \n",
    "    hopper_rnd_coverage_mean[0] = 0\n",
    "# ngu\n",
    "if 'Easy' in list_env and 'ngu' in list_alg : \n",
    "    Easy_ngu_coverage_mean[0] = 0\n",
    "if 'Ur' in list_env and 'ngu' in list_alg : \n",
    "    Ur_ngu_coverage_mean[0] = 0\n",
    "if 'Hard' in list_env and 'ngu' in list_alg : \n",
    "    Hard_ngu_coverage_mean[0] = 0\n",
    "if 'fetch' in list_env and 'ngu' in list_alg : \n",
    "    fetch_ngu_coverage_mean[0] = 0\n",
    "if 'finger' in list_env and 'ngu' in list_alg : \n",
    "    finger_ngu_coverage_mean[0] = 0\n",
    "if 'fetch_push' in list_env and 'ngu' in list_alg : \n",
    "    fetch_push_ngu_coverage_mean[0] = 0\n",
    "if 'hopper' in list_env and 'ngu' in list_alg : \n",
    "    hopper_ngu_coverage_mean[0] = 0\n",
    "# lsd\n",
    "if 'Easy' in list_env and 'lsd' in list_alg : \n",
    "    Easy_lsd_coverage_mean[0] = 0\n",
    "if 'Ur' in list_env and 'lsd' in list_alg : \n",
    "    Ur_lsd_coverage_mean[0] = 0\n",
    "if 'Hard' in list_env and 'lsd' in list_alg : \n",
    "    Hard_lsd_coverage_mean[0] = 0\n",
    "if 'fetch' in list_env and 'lsd' in list_alg : \n",
    "    fetch_lsd_coverage_mean[0] = 0\n",
    "if 'finger' in list_env and 'lsd' in list_alg : \n",
    "    finger_lsd_coverage_mean[0] = 0\n",
    "if 'fetch_push' in list_env and 'lsd' in list_alg : \n",
    "    fetch_push_lsd_coverage_mean[0] = 0\n",
    "if 'hopper' in list_env and 'lsd' in list_alg : \n",
    "    hopper_lsd_coverage_mean[0] = 0\n",
    "# csd\n",
    "if 'Easy' in list_env and 'csd' in list_alg : \n",
    "    Easy_csd_coverage_mean[0] = 0\n",
    "if 'Ur' in list_env and 'csd' in list_alg :\n",
    "    Ur_csd_coverage_mean[0] = 0\n",
    "if 'Hard' in list_env and 'csd' in list_alg :\n",
    "    Hard_csd_coverage_mean[0] = 0\n",
    "if 'fetch' in list_env and 'csd' in list_alg :\n",
    "    fetch_csd_coverage_mean[0] = 0\n",
    "if 'finger' in list_env and 'csd' in list_alg :\n",
    "    finger_csd_coverage_mean[0] = 0\n",
    "if 'fetch_push' in list_env and 'csd' in list_alg :\n",
    "    fetch_push_csd_coverage_mean[0] = 0\n",
    "# metra\n",
    "if 'Easy' in list_env and 'metra' in list_alg : \n",
    "    Easy_metra_coverage_mean[0] = 0\n",
    "if 'Ur' in list_env and 'metra' in list_alg :\n",
    "    Ur_metra_coverage_mean[0] = 0\n",
    "if 'Hard' in list_env and 'metra' in list_alg :\n",
    "    Hard_metra_coverage_mean[0] = 0\n",
    "if 'fetch' in list_env and 'metra' in list_alg :\n",
    "    fetch_metra_coverage_mean[0] = 0\n",
    "if 'finger' in list_env and 'metra' in list_alg :\n",
    "    finger_metra_coverage_mean[0] = 0\n",
    "if 'fetch_push' in list_env and 'metra' in list_alg :\n",
    "    fetch_push_metra_coverage_mean[0] = 0\n",
    "# smm\n",
    "if 'Easy' in list_env and 'smm' in list_alg : \n",
    "    Easy_smm_coverage_mean[0] = 0\n",
    "if 'Ur' in list_env and 'smm' in list_alg :\n",
    "    Ur_smm_coverage_mean[0] = 0\n",
    "if 'Hard' in list_env and 'smm' in list_alg :\n",
    "    Hard_smm_coverage_mean[0] = 0\n",
    "if 'fetch' in list_env and 'smm' in list_alg :\n",
    "    fetch_smm_coverage_mean[0] = 0\n",
    "if 'finger' in list_env and 'smm' in list_alg :\n",
    "    finger_smm_coverage_mean[0] = 0\n",
    "if 'fetch_push' in list_env and 'smm' in list_alg :\n",
    "    fetch_push_smm_coverage_mean[0] = 0\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 469,
   "metadata": {},
   "outputs": [],
   "source": [
    "# shapes \n",
    "# lds\n",
    "Easy_lds_size = Easy_lds_coverage_mean.shape[0] if 'Easy' in list_env and 'lds' in list_alg  else None\n",
    "Ur_lds_size = Ur_lds_coverage_mean.shape[0] if 'Ur' in list_env and 'lds' in list_alg  else None\n",
    "Hard_lds_size = Hard_lds_coverage_mean.shape[0] if 'Hard' in list_env and 'lds' in list_alg  else None\n",
    "fetch_lds_size = fetch_lds_coverage_mean.shape[0] if 'fetch' in list_env and 'lds' in list_alg  else None\n",
    "finger_lds_size = finger_lds_coverage_mean.shape[0] if 'finger' in list_env and 'lds' in list_alg  else None\n",
    "fetch_push_lds_size = fetch_push_lds_coverage_mean.shape[0] if 'fetch_push' in list_env  and 'lds' in list_alg else None\n",
    "hopper_lds_size = hopper_lds_coverage_mean.shape[0] if 'hopper' in list_env  and 'lds' in list_alg else None\n",
    "# diayn\n",
    "Easy_diayn_size = Easy_diayn_coverage_mean.shape[0] if 'Easy' in list_env  and 'diayn' in list_alg else None\n",
    "Ur_diayn_size = Ur_diayn_coverage_mean.shape[0] if 'Ur' in list_env  and 'diayn' in list_alg else None\n",
    "Hard_diayn_size = Hard_diayn_coverage_mean.shape[0] if 'Hard' in list_env  and 'diayn' in list_alg else None\n",
    "fetch_diayn_size = fetch_diayn_coverage_mean.shape[0] if 'fetch' in list_env and 'diayn' in list_alg  else None\n",
    "finger_diayn_size = finger_diayn_coverage_mean.shape[0] if 'finger' in list_env  and 'diayn' in list_alg else None\n",
    "fetch_push_diayn_size = fetch_push_diayn_coverage_mean.shape[0] if 'fetch_push' in list_env  and 'diayn' in list_alg else None\n",
    "hopper_diayn_size = hopper_diayn_coverage_mean.shape[0] if 'hopper' in list_env  and 'diayn' in list_alg else None\n",
    "# rnd\n",
    "Easy_rnd_size = Easy_rnd_coverage_mean.shape[0] if 'Easy' in list_env and 'rnd' in list_alg  else None\n",
    "Ur_rnd_size = Ur_rnd_coverage_mean.shape[0] if 'Ur' in list_env and 'rnd' in list_alg else None\n",
    "Hard_rnd_size = Hard_rnd_coverage_mean.shape[0] if 'Hard' in list_env  and 'rnd' in list_alg else None\n",
    "fetch_rnd_size = fetch_rnd_coverage_mean.shape[0] if 'fetch' in list_env  and 'rnd' in list_alg else None\n",
    "finger_rnd_size = finger_rnd_coverage_mean.shape[0] if 'finger' in list_env and 'rnd' in list_alg  else None\n",
    "fetch_push_rnd_size = fetch_push_rnd_coverage_mean.shape[0] if 'fetch_push' in list_env  and 'rnd' in list_alg else None\n",
    "hopper_rnd_size = hopper_rnd_coverage_mean.shape[0] if 'hopper' in list_env  and 'rnd' in list_alg else None\n",
    "# ngu\n",
    "Easy_ngu_size = Easy_ngu_coverage_mean.shape[0] if 'Easy' in list_env  and 'ngu' in list_alg else None\n",
    "Ur_ngu_size = Ur_ngu_coverage_mean.shape[0] if 'Ur' in list_env  and 'ngu' in list_alg else None\n",
    "Hard_ngu_size = Hard_ngu_coverage_mean.shape[0] if 'Hard' in list_env  and 'ngu' in list_alg else None\n",
    "fetch_ngu_size = fetch_ngu_coverage_mean.shape[0] if 'fetch' in list_env  and 'ngu' in list_alg else None\n",
    "finger_ngu_size = finger_ngu_coverage_mean.shape[0] if 'finger' in list_env and 'ngu' in list_alg  else None\n",
    "fetch_push_ngu_size = fetch_push_ngu_coverage_mean.shape[0] if 'fetch_push' in list_env and 'ngu' in list_alg  else None\n",
    "hopper_ngu_size = hopper_ngu_coverage_mean.shape[0] if 'hopper' in list_env  and 'ngu' in list_alg else None\n",
    "# lsd\n",
    "Easy_lsd_size = Easy_lsd_coverage_mean.shape[0] if 'Easy' in list_env  and 'lsd' in list_alg else None\n",
    "Ur_lsd_size = Ur_lsd_coverage_mean.shape[0] if 'Ur' in list_env and 'lsd' in list_alg  else None\n",
    "Hard_lsd_size = Hard_lsd_coverage_mean.shape[0] if 'Hard' in list_env  and 'lsd' in list_alg else None\n",
    "fetch_lsd_size = fetch_lsd_coverage_mean.shape[0] if 'fetch' in list_env  and 'lsd' in list_alg else None\n",
    "finger_lsd_size = finger_lsd_coverage_mean.shape[0] if 'finger' in list_env and 'lsd' in list_alg  else None\n",
    "fetch_push_lsd_size = fetch_push_lsd_coverage_mean.shape[0] if 'fetch_push' in list_env  and 'lsd' in list_alg else None\n",
    "hopper_lsd_size = hopper_lsd_coverage_mean.shape[0] if 'hopper' in list_env  and 'lsd' in list_alg else None\n",
    "# csd\n",
    "Easy_csd_size = Easy_csd_coverage_mean.shape[0] if 'Easy' in list_env  and 'csd' in list_alg else None\n",
    "Ur_csd_size = Ur_csd_coverage_mean.shape[0] if 'Ur' in list_env  and 'csd' in list_alg else None\n",
    "Hard_csd_size = Hard_csd_coverage_mean.shape[0] if 'Hard' in list_env  and 'csd' in list_alg else None\n",
    "fetch_csd_size = fetch_csd_coverage_mean.shape[0] if 'fetch' in list_env  and 'csd' in list_alg else None\n",
    "finger_csd_size = finger_csd_coverage_mean.shape[0] if 'finger' in list_env and 'csd' in list_alg  else None\n",
    "fetch_push_csd_size = fetch_push_csd_coverage_mean.shape[0] if 'fetch_push' in list_env  and 'csd' in list_alg else None\n",
    "# metra\n",
    "Easy_metra_size = Easy_metra_coverage_mean.shape[0] if 'Easy' in list_env  and 'metra' in list_alg else None\n",
    "Ur_metra_size = Ur_metra_coverage_mean.shape[0] if 'Ur' in list_env  and 'metra' in list_alg else None\n",
    "Hard_metra_size = Hard_metra_coverage_mean.shape[0] if 'Hard' in list_env  and 'metra' in list_alg else None\n",
    "fetch_metra_size = fetch_metra_coverage_mean.shape[0] if 'fetch' in list_env  and 'metra' in list_alg else None\n",
    "finger_metra_size = finger_metra_coverage_mean.shape[0] if 'finger' in list_env and 'metra' in list_alg  else None\n",
    "fetch_push_metra_size = fetch_push_metra_coverage_mean.shape[0] if 'fetch_push' in list_env  and 'metra' in list_alg else None\n",
    "# smm\n",
    "Easy_smm_size = Easy_smm_coverage_mean.shape[0] if 'Easy' in list_env  and 'smm' in list_alg else None\n",
    "Ur_smm_size = Ur_smm_coverage_mean.shape[0] if 'Ur' in list_env  and 'smm' in list_alg else None\n",
    "Hard_smm_size = Hard_smm_coverage_mean.shape[0] if 'Hard' in list_env  and 'smm' in list_alg else None\n",
    "fetch_smm_size = fetch_smm_coverage_mean.shape[0] if 'fetch' in list_env  and 'smm' in list_alg else None\n",
    "finger_smm_size = finger_smm_coverage_mean.shape[0] if 'finger' in list_env and 'smm' in list_alg  else None\n",
    "fetch_push_smm_size = fetch_push_smm_coverage_mean.shape[0] if 'fetch_push' in list_env  and 'smm' in list_alg else None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 470,
   "metadata": {},
   "outputs": [],
   "source": [
    "# nb_samples \n",
    "Easy_sample = 10**6\n",
    "Ur_sample = 10**6\n",
    "Hard_sample = 10**6\n",
    "fetch_sample = 2*10**5\n",
    "finger = 10**5\n",
    "fetch_push_sample = 10**5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 471,
   "metadata": {},
   "outputs": [],
   "source": [
    "# x_axis \n",
    "# lds\n",
    "Easy_lds_x_axis = np.linspace(0, Easy_sample, Easy_lds_size) if 'Easy' in list_env and 'lds' in list_alg  else None\n",
    "Ur_lds_x_axis = np.linspace(0, Ur_sample, Ur_lds_size) if 'Ur' in list_env and 'lds' in list_alg  else None\n",
    "Hard_lds_x_axis = np.linspace(0, Hard_sample, Hard_lds_size) if 'Hard' in list_env  and 'lds' in list_alg else None\n",
    "fetch_lds_x_axis = np.linspace(0, fetch_sample, fetch_lds_size) if 'fetch' in list_env  and 'lds' in list_alg else None\n",
    "finger_lds_x_axis = np.linspace(0, finger, finger_lds_size) if 'finger' in list_env and 'lds' in list_alg  else None\n",
    "fetch_push_lds_x_axis = np.linspace(0, fetch_push_sample, fetch_push_lds_size) if 'fetch_push' in list_env  and 'lds' in list_alg else None\n",
    "# diayn\n",
    "Easy_diayn_x_axis = np.linspace(0, Easy_sample, Easy_diayn_size) if 'Easy' in list_env  and 'diayn' in list_alg else None\n",
    "Ur_diayn_x_axis = np.linspace(0, Ur_sample, Ur_diayn_size) if 'Ur' in list_env and 'diayn' in list_alg  else None\n",
    "Hard_diayn_x_axis = np.linspace(0, Hard_sample, Hard_diayn_size) if 'Hard' in list_env  and 'diayn' in list_alg else None\n",
    "fetch_diayn_x_axis = np.linspace(0, fetch_sample, fetch_diayn_size) if 'fetch' in list_env and 'diayn' in list_alg  else None\n",
    "finger_diayn_x_axis = np.linspace(0, finger, finger_diayn_size) if 'finger' in list_env and 'diayn' in list_alg  else None\n",
    "fetch_push_diayn_x_axis = np.linspace(0, fetch_push_sample, fetch_push_diayn_size) if 'fetch_push' in list_env and 'diayn' in list_alg  else None\n",
    "# rnd\n",
    "Easy_rnd_x_axis = np.linspace(0, Easy_sample, Easy_rnd_size) if 'Easy' in list_env and 'rnd' in list_alg  else None\n",
    "Ur_rnd_x_axis = np.linspace(0, Ur_sample, Ur_rnd_size) if 'Ur' in list_env and 'rnd' in list_alg  else None\n",
    "Hard_rnd_x_axis = np.linspace(0, Hard_sample, Hard_rnd_size) if 'Hard' in list_env  and 'rnd' in list_alg else None\n",
    "fetch_rnd_x_axis = np.linspace(0, fetch_sample, fetch_rnd_size) if 'fetch' in list_env  and 'rnd' in list_alg else None\n",
    "finger_rnd_x_axis = np.linspace(0, finger, finger_rnd_size) if 'finger' in list_env  and 'rnd' in list_alg else None\n",
    "fetch_push_rnd_x_axis = np.linspace(0, fetch_push_sample, fetch_push_rnd_size) if 'fetch_push' in list_env and 'rnd' in list_alg  else None\n",
    "# ngu\n",
    "Easy_ngu_x_axis = np.linspace(0, Easy_sample, Easy_ngu_size) if 'Easy' in list_env  and 'ngu' in list_alg else None\n",
    "Ur_ngu_x_axis = np.linspace(0, Ur_sample, Ur_ngu_size) if 'Ur' in list_env and 'ngu' in list_alg  else None\n",
    "Hard_ngu_x_axis = np.linspace(0, Hard_sample, Hard_ngu_size) if 'Hard' in list_env  and 'ngu' in list_alg else None\n",
    "fetch_ngu_x_axis = np.linspace(0, fetch_sample, fetch_ngu_size) if 'fetch' in list_env  and 'ngu' in list_alg else None\n",
    "finger_ngu_x_axis = np.linspace(0, finger, finger_ngu_size) if 'finger' in list_env  and 'ngu' in list_alg else None\n",
    "fetch_push_ngu_x_axis = np.linspace(0, fetch_push_sample, fetch_push_ngu_size) if 'fetch_push' in list_env  and 'ngu' in list_alg else None\n",
    "# lsd\n",
    "Easy_lsd_x_axis = np.linspace(0, Easy_sample, Easy_lsd_size) if 'Easy' in list_env and 'lsd' in list_alg  else None\n",
    "Ur_lsd_x_axis = np.linspace(0, Ur_sample, Ur_lsd_size) if 'Ur' in list_env and 'lsd' in list_alg  else None\n",
    "Hard_lsd_x_axis = np.linspace(0, Hard_sample, Hard_lsd_size) if 'Hard' in list_env  and 'lsd' in list_alg else None\n",
    "fetch_lsd_x_axis = np.linspace(0, fetch_sample, fetch_lsd_size) if 'fetch' in list_env and 'lsd' in list_alg  else None\n",
    "finger_lsd_x_axis = np.linspace(0, finger, finger_lsd_size) if 'finger' in list_env  and 'lsd' in list_alg else None\n",
    "fetch_push_lsd_x_axis = np.linspace(0, fetch_push_sample, fetch_push_lsd_size) if 'fetch_push' in list_env  and 'lsd' in list_alg else None\n",
    "# csd\n",
    "Easy_csd_x_axis = np.linspace(0, Easy_sample, Easy_csd_size) if 'Easy' in list_env  and 'csd' in list_alg else None\n",
    "Ur_csd_x_axis = np.linspace(0, Ur_sample, Ur_csd_size) if 'Ur' in list_env  and 'csd' in list_alg else None\n",
    "Hard_csd_x_axis = np.linspace(0, Hard_sample, Hard_csd_size) if 'Hard' in list_env  and 'csd' in list_alg else None\n",
    "fetch_csd_x_axis = np.linspace(0, fetch_sample, fetch_csd_size) if 'fetch' in list_env  and 'csd' in list_alg else None\n",
    "finger_csd_x_axis = np.linspace(0, finger, finger_csd_size) if 'finger' in list_env  and 'csd' in list_alg else None\n",
    "fetch_push_csd_x_axis = np.linspace(0, fetch_push_sample, fetch_push_csd_size) if 'fetch_push' in list_env  and 'csd' in list_alg else None\n",
    "# metra\n",
    "Easy_metra_x_axis = np.linspace(0, Easy_sample, Easy_metra_size) if 'Easy' in list_env  and 'metra' in list_alg else None\n",
    "Ur_metra_x_axis = np.linspace(0, Ur_sample, Ur_metra_size) if 'Ur' in list_env  and 'metra' in list_alg else None\n",
    "Hard_metra_x_axis = np.linspace(0, Hard_sample, Hard_metra_size) if 'Hard' in list_env  and 'metra' in list_alg else None\n",
    "fetch_metra_x_axis = np.linspace(0, fetch_sample, fetch_metra_size) if 'fetch' in list_env  and 'metra' in list_alg else None\n",
    "finger_metra_x_axis = np.linspace(0, finger, finger_metra_size) if 'finger' in list_env  and 'metra' in list_alg else None\n",
    "fetch_push_metra_x_axis = np.linspace(0, fetch_push_sample, fetch_push_metra_size) if 'fetch_push' in list_env  and 'metra' in list_alg else None\n",
    "# smm\n",
    "Easy_smm_x_axis = np.linspace(0, Easy_sample, Easy_smm_size) if 'Easy' in list_env  and 'smm' in list_alg else None\n",
    "Ur_smm_x_axis = np.linspace(0, Ur_sample, Ur_smm_size) if 'Ur' in list_env  and 'smm' in list_alg else None\n",
    "Hard_smm_x_axis = np.linspace(0, Hard_sample, Hard_smm_size) if 'Hard' in list_env  and 'smm' in list_alg else None\n",
    "fetch_smm_x_axis = np.linspace(0, fetch_sample, fetch_smm_size) if 'fetch' in list_env  and 'smm' in list_alg else None\n",
    "finger_smm_x_axis = np.linspace(0, finger, finger_smm_size) if 'finger' in list_env  and 'smm' in list_alg else None\n",
    "fetch_push_smm_x_axis = np.linspace(0, fetch_push_sample, fetch_push_smm_size) if 'fetch_push' in list_env  and 'smm' in list_alg else None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 472,
   "metadata": {},
   "outputs": [],
   "source": [
    "colors = {'LEADSSS': 'tab:blue', 'DIAYN': 'tab:orange', 'RND': 'tab:green', 'NGU': 'tab:red', 'LSD': 'tab:purple', 'CSD': 'tab:brown', 'METRA' : 'tab:pink', 'SMM': 'tab:gray'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Easy\n",
    "Easy_fig = plt.figure(figsize=(4, 4))\n",
    "Easy_ax = Easy_fig.add_subplot(111)\n",
    "Easy_ax.set_title('Easy', fontsize=15)\n",
    "# Easy_ax.set_xlabel(r'$\\times 10^3$')\n",
    "# Easy_ax.set_ylabel('Coverage')\n",
    "# seaborn style\n",
    "sns.set(style=\"whitegrid\")\n",
    "# plot\n",
    "# lds\n",
    "plt.plot(Easy_lds_x_axis, Easy_lds_coverage_mean, label='LEADSSS', color=colors['LEADSSS'])\n",
    "plt.fill_between(Easy_lds_x_axis, Easy_lds_coverage_mean - Easy_lds_coverage_std, Easy_lds_coverage_mean + Easy_lds_coverage_std, alpha=0.3, color=colors['LEADSSS'])\n",
    "# diayn\n",
    "plt.plot(Easy_diayn_x_axis, Easy_diayn_coverage_mean, label='DIAYN', color=colors['DIAYN'])\n",
    "plt.fill_between(Easy_diayn_x_axis, Easy_diayn_coverage_mean - Easy_diayn_coverage_std, Easy_diayn_coverage_mean + Easy_diayn_coverage_std, alpha=0.3, color=colors['DIAYN'])\n",
    "# rnd\n",
    "plt.plot(Easy_rnd_x_axis, Easy_rnd_coverage_mean, label='RND', color=colors['RND'])\n",
    "plt.fill_between(Easy_rnd_x_axis, Easy_rnd_coverage_mean - Easy_rnd_coverage_std, Easy_rnd_coverage_mean + Easy_rnd_coverage_std, alpha=0.3, color=colors['RND'])\n",
    "# ngu\n",
    "plt.plot(Easy_ngu_x_axis, Easy_ngu_coverage_mean, label='NGU', color=colors['NGU'])\n",
    "plt.fill_between(Easy_ngu_x_axis, Easy_ngu_coverage_mean - Easy_ngu_coverage_std, Easy_ngu_coverage_mean + Easy_ngu_coverage_std, alpha=0.3, color=colors['NGU'])\n",
    "# lsd\n",
    "plt.plot(Easy_lsd_x_axis, Easy_lsd_coverage_mean, label='LSD', color=colors['LSD'])\n",
    "plt.fill_between(Easy_lsd_x_axis, Easy_lsd_coverage_mean - Easy_lsd_coverage_std, Easy_lsd_coverage_mean + Easy_lsd_coverage_std, alpha=0.3, color=colors['LSD'])\n",
    "# CSD\n",
    "plt.plot(Easy_csd_x_axis, Easy_csd_coverage_mean, label='CSD', color=colors['CSD'])\n",
    "plt.fill_between(Easy_csd_x_axis, Easy_csd_coverage_mean - Easy_csd_coverage_std, Easy_csd_coverage_mean + Easy_csd_coverage_std, alpha=0.3, color=colors['CSD'])\n",
    "# METRA\n",
    "plt.plot(Easy_metra_x_axis, Easy_metra_coverage_mean, label='METRA', color=colors['METRA'])\n",
    "plt.fill_between(Easy_metra_x_axis, Easy_metra_coverage_mean - Easy_metra_coverage_std, Easy_metra_coverage_mean + Easy_metra_coverage_std, alpha=0.3, color=colors['METRA'])\n",
    "# SMM\n",
    "plt.plot(Easy_smm_x_axis, Easy_smm_coverage_mean, label='SMM', color=colors['SMM'])\n",
    "plt.fill_between(Easy_smm_x_axis, Easy_smm_coverage_mean - Easy_smm_coverage_std, Easy_smm_coverage_mean + Easy_smm_coverage_std, alpha=0.3, color=colors['SMM'])\n",
    "# CIC\n",
    "\n",
    "# plt.legend()\n",
    "plt.tight_layout()\n",
    "formatter = ticker.ScalarFormatter(useMathText=True)\n",
    "formatter.set_scientific(True)\n",
    "formatter.set_powerlimits((-1,1)) \n",
    "Easy_ax.xaxis.set_major_formatter(formatter)\n",
    "# limit y \n",
    "plt.ylim(0, 101)\n",
    "# plt.xlim(-10, 10**6)\n",
    "plt.xlim(-10**4, 10**6)\n",
    "plt.savefig('figures/Easy.pdf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Ur\n",
    "Ur_fig = plt.figure(figsize=(4, 4))\n",
    "Ur_ax = Ur_fig.add_subplot(111)\n",
    "Ur_ax.set_title('Umaze', fontsize=15)\n",
    "# Ur_ax.set_xlabel(r'$\\times 10^3$')\n",
    "# Ur_ax.set_ylabel('Coverage')\n",
    "# seaborn style\n",
    "sns.set(style=\"whitegrid\")\n",
    "# plot\n",
    "# lds\n",
    "plt.plot(Ur_lds_x_axis, Ur_lds_coverage_mean, label='LEADSSS', color=colors['LEADSSS'])\n",
    "plt.fill_between(Ur_lds_x_axis, Ur_lds_coverage_mean - Ur_lds_coverage_std, Ur_lds_coverage_mean + Ur_lds_coverage_std, alpha=0.3, color=colors['LEADSSS'])\n",
    "# diayn\n",
    "plt.plot(Ur_diayn_x_axis, Ur_diayn_coverage_mean, label='DIAYN', color=colors['DIAYN'])\n",
    "plt.fill_between(Ur_diayn_x_axis, Ur_diayn_coverage_mean - Ur_diayn_coverage_std, Ur_diayn_coverage_mean + Ur_diayn_coverage_std, alpha=0.3, color=colors['DIAYN'])\n",
    "# rnd\n",
    "plt.plot(Ur_rnd_x_axis, Ur_rnd_coverage_mean, label='RND', color=colors['RND'])\n",
    "plt.fill_between(Ur_rnd_x_axis, Ur_rnd_coverage_mean - Ur_rnd_coverage_std, Ur_rnd_coverage_mean + Ur_rnd_coverage_std, alpha=0.3, color=colors['RND'])\n",
    "# ngu\n",
    "plt.plot(Ur_ngu_x_axis, Ur_ngu_coverage_mean, label='NGU', color=colors['NGU'])\n",
    "plt.fill_between(Ur_ngu_x_axis, Ur_ngu_coverage_mean - Ur_ngu_coverage_std, Ur_ngu_coverage_mean + Ur_ngu_coverage_std, alpha=0.3, color=colors['NGU'])\n",
    "# lsd\n",
    "plt.plot(Ur_lsd_x_axis, Ur_lsd_coverage_mean, label='LSD', color=colors['LSD'])\n",
    "plt.fill_between(Ur_lsd_x_axis, Ur_lsd_coverage_mean - Ur_lsd_coverage_std, Ur_lsd_coverage_mean + Ur_lsd_coverage_std, alpha=0.3, color=colors['LSD'])\n",
    "# CSD\n",
    "plt.plot(Ur_csd_x_axis, Ur_csd_coverage_mean, label='CSD', color=colors['CSD'])\n",
    "plt.fill_between(Ur_csd_x_axis, Ur_csd_coverage_mean - Ur_csd_coverage_std, Ur_csd_coverage_mean + Ur_csd_coverage_std, alpha=0.3, color=colors['CSD'])\n",
    "# METRA\n",
    "plt.plot(Ur_metra_x_axis, Ur_metra_coverage_mean, label='METRA', color=colors['METRA'])\n",
    "plt.fill_between(Ur_metra_x_axis, Ur_metra_coverage_mean - Ur_metra_coverage_std, Ur_metra_coverage_mean + Ur_metra_coverage_std, alpha=0.3, color=colors['METRA'])\n",
    "# SMM\n",
    "plt.plot(Ur_smm_x_axis, Ur_smm_coverage_mean, label='SMM', color=colors['SMM'])\n",
    "plt.fill_between(Ur_smm_x_axis, Ur_smm_coverage_mean - Ur_smm_coverage_std, Ur_smm_coverage_mean + Ur_smm_coverage_std, alpha=0.3, color=colors['SMM'])\n",
    "# CIC\n",
    "\n",
    "# plt.legend()\n",
    "# limit y \n",
    "plt.ylim(0, 101)\n",
    "# plt.xlim(-10, 10**6)\n",
    "plt.xlim(-10**4, 10**6)\n",
    "plt.tight_layout()\n",
    "formatter = ticker.ScalarFormatter(useMathText=True)\n",
    "formatter.set_scientific(True)\n",
    "formatter.set_powerlimits((-1,1)) \n",
    "Ur_ax.xaxis.set_major_formatter(formatter)\n",
    "\n",
    "plt.savefig('figures/Ur.pdf')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAGECAYAAAAlahcnAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB9i0lEQVR4nO3deXycV33o/8+zzqoZ7bJlS15jx87mQMAJCQkEuiQE6K8hNL0toXAvWwkteV22cksokBdQ7qUFEhJuEmibSwot26UlIXDLEgqEQEISZ3HiRZYtL9o1+8yznfP745HGliXb2kayRuf9esmyZnnmnJHm+T5n+x5NSilRFEVRlDPQl7oAiqIoyvKgAoaiKIoyIypgKIqiKDOiAoaiKIoyIypgKIqiKDOiAoaiKIoyIypgKIqiKDOiAoaiKIoyIypgKIqiKDOiAoainOW2bt3Km970pqUuhqKogKGsbIcPH2br1q381//6X0/5mEcffZStW7dy6623LmLJFOXsowKGoiiKMiMqYCiKoigzogKGoszBM888w8c//nGuu+46XvziF3PhhRfy2te+lrvvvhvP86Y8/uqrr+bqq68ml8vx8Y9/nKuuuort27fz7W9/u/qYb3zjG1x33XVccMEFXHXVVXzmM5/BcZzFrJainJa51AVQlOXoX//1X/nJT37CS17yEq688koqlQq//vWv+exnP8vTTz/N7bffPuU5ruvy5je/mVKpxNVXX41hGLS0tADwxS9+kS984Qu0trbyxje+EdM0+f73v09PT89iV01RTkkFDEUBDh06NO1JHuDIkSNTbnvnO9/JRz/6UQzDqN4mpeR//I//wbe+9S0ef/xxXvziF096ztDQEFu3buVrX/sa0Wi0evvBgwe588476ejo4Dvf+U41iLznPe/hDW94w0JUT1EWhAoYikIYMO64444ZP76zs3PKbZqm8Sd/8id861vf4pFHHpkSMADe//73TwoWAP/+7/+O7/u85S1vqQYLgGQyybve9S4+8IEPzKImilI7KmAoCnDFFVfw5S9/edr7Hn30UW666aZJt7muy/33388DDzxAT08PpVKJEzevHBwcnHKcSCTC1q1bp9z+wgsvAEwbYC655JJZ1UNRakkFDEWZg7/4i7/gJz/5CevXr+faa6+lpaUF0zTJ5XLcd999uK475TktLS1omjbl9nw+X73/ZK2trQtfeEWZIxUwFGWWdu3axU9+8hOuuOIK7r777knjGE8++ST33XfftM+bLlgANDQ0ADAyMsKaNWsm3Tc8PLxApVaU+VPTahVllvr6+gB4xSteMSlYADz22GOzPt5EN9Xjjz8+5b65HE9RakUFDEWZpYkB75NP8Hv37uXuu++e9fFe+9rXYhgG//AP/8DIyEj19kKhwF133TW/wirKAlJdUooySxdeeCEXXngh3//+9xkaGuKiiy7i2LFj/PjHP+aqq67iBz/4wayOt27dOv78z/+c22+/nde97nVcc801GIbBD3/4Q7Zu3cqBAwdqVBNFmR3VwlCUWTIMg//9v/83119/PYcOHeKrX/0q+/bt4wMf+ADvf//753TMm2++mdtuu43Gxka+/vWv89BDD/H7v//7fO5zn1vYwivKPGjyxLmAiqIoinIKqoWhKIqizIgKGIqiKMqMqIChKIqizIgKGIqiKMqMzDpgHDx4kFtvvZXXv/71bN++neuuu27ax33jG9/g937v97jgggt43etex09+8pMpj8nn83z4wx/mpS99KRdffDF/8Rd/MW0OHkVRFGXpzTpg7N27l4cffph169axadOmaR/zwAMP8JGPfIRrrrmGe+65hx07dnDzzTfz5JNPTnrce9/7Xn7xi1/wN3/zN/yv//W/OHDgAG9729vwfX9OlVEURVFqZ9bTaoUQ6HoYZz70oQ/xzDPP8L3vfW/SY37v936P888/n89+9rPV22688UYaGhq45557AHjiiSe48cYb+fKXv8wVV1wBQE9PD9deey1/93d/x7XXXjvryjzxxBNIKbEsa9bPVRRFqQee56FpGhdffPGCH3vWLYyJYHEqfX199Pb2cs0110y6/dprr+WRRx6pZvH82c9+RiqV4vLLL68+ZuPGjWzbto2f/exnsy0WEG5gMxH/pJS4rstKXWai6q/qv1Lrv5LrDpPPgwttwVODTGwpuWHDhkm3b9q0Cc/z6OvrY9OmTfT09LBhw4YpGTw3btw4520pJ1oWF1xwAaVSid27d7N582bi8ficjrecqfqr+q/U+q/kukOYTflUmZHna8EDRjabBSCVSk26feLniftzuVw1rfOJ0uk0zzzzzJxfX0pJqVSiXC4DVL+vNKr+qv4nfl9JVnLdITwHLpuAsdQ8z2P37t3Vn3t7e5euMGcBVf/epS7CklrJ9V/JdbdtuybHXfCAkU6ngXDKbFtbW/X2XC436f5UKkV/f/+U52ez2epj5sKyLDZv3ky5XKa3t5f169cTi8XmfLzlStVf1X+l1n8l1x3Cmay1suABY+PGjUA4ljHx/4mfLcuiq6ur+rhHHnlkSvPpwIEDbNmyZc6vr2napH7LWCy2IvsxJ6j6q/qv1Pqv1LrXqjsKarDSu6uri/Xr1/PQQw9Nuv3BBx/ksssuqzaVrrzySrLZLI888kj1MQcOHOC5557jyiuvXOhiKYqiKPM06xZGuVzm4YcfBuDIkSMUCoVqcHjpS19Kc3Mz73nPe3jf+95Hd3c3O3fu5MEHH2TXrl189atfrR7n4osv5oorruDDH/4wH/zgB4lEIvz93/89W7du5Xd/93cXqHqKoijKQpl1wBgZGeEv//IvJ9028fN9993Hzp07ue666yiXy9xzzz3cfffdbNiwgTvuuGPKQpLPfe5zfOpTn+LWW2/F932uuOIK/vqv/xrTrLuxeEVRlGVv1mfmtWvX8sILL5zxcTfccAM33HDDaR/T0NDAJz/5ST75yU/OthiKoijKIlPZahVFUZQZUQFDURRFmREVMBRFUZQZUQFDURRFmRE1HUlRFGWZCoQkV/Yoe8Gk22q1dk8FDEVRlCUkpcTxBRUvoOIJXF+c8TlF12e44DCYdyg5PhXv+HO2RwMMQyUfVBRFqQv5isdg3uFopsxQ3sH1BV4gcANBICRwhhO+lJiGTsI2aIzZRFJ69RlGOUuttgJRAUNRFKXGhJBkyh7DBYfDo2X6c2UKjo+pa8RtE9vQSUZMLEPH0GuXC2q+VMBQFEWZAyklgZAICULK8a9wDEGI8OeKLxjOV+gbLTNWcil7AZah0xizaE1G0GuYKLAWVMBQFEWZpdGiy2O9o2TLHkJObIsKgjBYTPzfD8LAkbBN0jGL1enlnW5dBQxFUZRZODhS5DcHxhgruTTGLUxdQ0ND1zQ0DXRNQ9fCNOOGrp3VXUyzpQKGoijKDPiB4LljOZ44lEED1rfEa7r3xNlIBQxFUZRTcPyAQsUnX/E5NFrkhf48TXGbxnhttkA926mAoSiKcoKKF3BguMjBkSLZsk/FC3D9ADSN1ekYUctY6iIuGRUwFEVRCFsTB0dKPHskx2C+QswyiNsGrQkb29RXXPfTdFTAUBRlRXL8gJITUHTDLqc9A3kGchUStsn6lkRdDVYvFBUwFEVZMnIBliRLCcH4mggvEOTLHiOlgJ7hIlL3GCu5FCr+pOcIKSm5ARUvwAkESEnUMuhujmPqKifrqaiAoShKTVS8IEx1EUh8EZ7QHT+g5AbkK/74idxbkNcKRBgEAimpVBwOH63Q5o5gWhaWoWMbU4NAxNRpHu9uWm4L6JaKChiKoiyIQEgyJZeRosvRTJmBnIMXBAgxvvqZMGhohGsUoqaBZWgLMjaga2AZ4Yk/qgmchM665jiRSGT+FTtLVbyAX/eO8psDo5Oy1f7VpQlSMasmr6kChqIosyKlpOIJSq5PwQm/xoouwwWXbNmj4gXYhk5D1CQWtTA0DV3Xqt9rzSFYti0GXwj84PTddEN5h5/tHeKRnpFJWWonBCJeq+KpgKEoyqlJKcmUwsyq2ZJLtuKTr3i4vsAZ/wKwdI2oZdCSsFf0tNPZEkJyaLTEc8dyPHcsx77BAr6Y+bhOe0OEl5/TSmfj8ZQjDdFCLYoKqIChKMoJJvZmKLsBg3mH3uFCuOeCG2DqGpahEzHDr4aIqaabztHhsRIP7xniN71jFBz/zE84gaFp7Ohu5BVb2ti6qmFKa8oqlahRdnMVMBTlREUnHIz1ztAtcDYJhKx2Zbh+uKdCpVLh4KBDPjZGNFo+4zH8QFB0A4qOjxdI3PHB6ahl0FgHSfPOBq4veOzgKA/vGWL/ULF6e8wy2Lqqge2rU2xfnaIpcfrxB0PXlmwmlwoYyormBYL+bIWRgsPRbIVsyaPo+jW7QqsJOfFNohGOFbieS3/GoxwpYFvuGQ+hjQ8aW4aGaejEYzarUqr1sBCOZcs8vGeIX+4foeSGg9OGprGjq5Ert7Ry7qrUslnzoQKGsiLlKx59o2X2DuYZzjsEUhKzDJIRk+ZEfNl8gE/FcQwomnQ1xep6ptDZRkpJruIzmK/Qn63wq55RXhjIV+9vSdi8/JxWrtjcuizzUamAoawYQkiGCw69w0V6hotkyx7JiElnYwxrmnn6inIqZTdg31CBobxz/KsQfp28J7emwYVr0ly1pY3zO9OLMlOsVlTAUOpe2Q04kimzfzDPsVwFzxc0xW02tCaW7fRLZWl4geDHzw/ywNPHqt1LJ9M0aI7btDVEOKc9ycvPaaM5sfxaE9NRAUOpOxUvnOHTl/MpHBxjoCjIlFwihkFrMqKmfSqzJoTkkQMjfPeJo4yWwjGhloRNd3Oc1oYI7ckIbQ0RWhsitCZszDptsaqAoSxbRzJlnj2axT+hC0BIKLg+mUKZI8cqdIoCzQ0x1jcnlnVXgLJ4XF/w7NEs/blKtbvpWLZCphymMWmKW7z+ojVctqll2Y91zZYKGMqydGC4yCP7hymPT/2coGkQMQ1WNUSgYNLVrAZ9lZk5minzs72TZzOdKG4bXHP+Kl51bge2WZ8tiDNRAUNZVqSUvDCQ59c9oxi6xrqWxLSPc5ypKROUlUlKOXmatISC44eD1OMtiN39OfYMHF8h3ZKw2dyepH28m6k9GaGrOb7iuzNVwFCWDSEkuw5n+O2hDAnboCWpWg5KqOIFDOUdBvMO/ZkivQMlyvsOMFLyGC24BDNIo65pcNHacAX19s6UmhAxDRUwlGVj1+EMv+kdpTkRIV2jbJzKwvOFYLToVk/ow3mHwUL43Q3m3xIsOsEp0mucesGiBjQlbNrGB6tXp6O8ZH1z3cxmqhUVMJRl4fBYiV2HsypYnOW8QLBvsMCzR3McHCkyVHAYLbrMIp/enCUjJu0NEZrjJraosHF1K53NSVqTkSn7YUQsvW7W3pRcn1zFr67432jKmq3QVwFDOesVHJ/HescQUqpgcZaRUnI4U+a5o2G21b0DhWlbDZah0Tp+Nd/eEKle2ccWYEwgYhm0JSPE7PBYjuPQd7iPrrXNdT3hwQsE/bkKAKvS0WoXmlHRatadpgKGclYTQvLkoTEGchU2nGKAW1kcE2kvhgsOxzIVdvfn2H0sF17dniAds9i+OsWWjiSrUlHaGiKkYpYaE1ggQkpGCi55x2NtU5wL16ZZ0xirtip27RpULQxlZdo7WOD5/jyd6ZhaR7EEBnIVfrZ3iGeP5hjKO9X9L05kmzpbOpJsX53ivNVpOhujKmnhLEkpyVd8RksOQmpwhvSXTXGbK7e0sbE1uahTfFXAUM5aQ3mHJw6NkbDNaneDUnu+EDzVl+XhPUM8dyw36b4TB4s3tSfYvjrFprZk3YwHLDZfCMaKHrmKRzJqcu6qFGubzpz8sjFu0RBd/O5ZFTCUs0ogJCMFh/5chf2DRfKOx4aW5FIX66zm+oLh8cR3EzOPAj8gk63wfH4Iwzx9sJUSxkpuNYHeSMGt7vqmAeetSXHFplbWNsdpSdgqOCwAPxAMFRzKXkBrMsLL1rTQ1Rw/6zPYqoChLDk/CKddDuQq9I6UGC6EJ72kbdLdVNtxi4FchZ6hIoP5CsMFl8F8hWzZW5RZPQvBD8SUMYTJ+ud03FTU5IpzWrnynDZa1XqXBXNioFiVinLpxpZltSBQBQxlSfiBYCDvMJitcHC0NL7LnSBhm7QnI0Rq+AHyA8ETfRke3jPE8/35Mz9hGYhZRpj8LhnuqR0EAcVSkUQ8gWGc+b1sjFnhiubxGUxNcVuNGS0gXwiG8scDxWWbwkAROUPr72yjAoay6KSU/PZQhmeOZPGFoCFi1TxIQDgm8rO9Q/x83zD58atyTYPNbUlWp6O0JsMTZlPCxlgmg7a6ptGcsElEjEkDzcenlnbV9dTSs10wvgdL0fXpGA8U3c2JZZuLSgUMZdH1jpR45kiGprhd84G7QEieODTGT/cM8dzRXHXuSTpm8fJzWnn55laVYkRZUBUv3Bu96AZ4gaC9IcpLNjSzrmX5tShOpgKGsqiyJY/fHhzDMvSaBouxkst/HirzzG+fJ1s+3se/fXWKq7a0cVFXGlNfnld5ytlDSknZCyg6AUXXJxCS6PhWv1s6YnSkostqjOJMVMBQFo0fCH57aIzhgsOG1vkPZvtCUPFOWBcgoWe4wMN7hth1JMtEvrmGqMnlm1q5cksr7Q3Reb+usrI5XkCu4lNwPCQQsw0aIhab2hI0JyM0xS3SMasuN1GqWcD40Y9+xJe+9CX27dtHIpHgxS9+Me973/vo6uqa9LhvfOMb3HvvvRw9epQNGzZwyy238MpXvrJWxVKW0PP9OfYO5lnbGJvXql/HC/h/uwf4wbMDlL3pt8kE6E6ZvOq81bx0Y5uaCqrMihASxxf4QuALiR9IHD/ADQS2qdMYszmnI8xT1Ri3SEWtFTFJoCYB49FHH+Xmm2/mD/7gD7jlllvIZDJ8/vOf561vfSv//u//TjQaXuU98MADfOQjH+Gd73wnl156KQ8++CA333wz999/Pzt27KhF0ZQl0p+t8GRfhsaYPefB7UBIfr5vmH976ijZ8d3PTpawDV62qZVL16fwckN0rW1UwUKZEdcX5CoeecdDynDmmWXomIZGImLQkY6wOh2jNRmhOWGvuN32oEYB44EHHqCzs5NPfvKT1Zkbzc3NvPnNb+aZZ57hkksuAeALX/gCr3nNa3jve98LwKWXXsqePXv44he/yD333FOLoimLrOIF7B8s8MzRHK4vWJWKzek4vSNF7v3PA9Vka61Jm//v4jW8eF0TOsc/uJoGmqaFs4RypzqaUu+klHiBwHf88PtJC2ukDJP3eYGo7pVh6TqpuMWO9ibaGyI0xmxsU8c29RUZHKZTk4Dh+z6JRGLSNL+GhgYg/EUC9PX10dvby/vf//5Jz7322mv5zGc+g+u62PbZvepROTUvEBwcKfLMkRwDuQrpmMXapvicjrVnIM8XfryXiidIRkxee+FqrtrSVpd9xMrsCSmpeAElN6DkBJQdh4FCgJF3ScQ1LEMnYuloTD7pt0UjJCMmyYhJxNRJxyya1Er206pJwPjDP/xDvvvd73L//ffzute9jkwmw9/93d+xfft2XvSiFwHQ09MDwIYNGyY9d9OmTXieR19fH5s2bapF8ZQay5RcftUzwqHRMnHLYH1LYs5XaE8fyXLnT/fhBZKtHQ28+5WbiNtqrsZy43gBRTeg4gWTruoXgoZG1NZJ2hZdzXGSZoKjdpbzzm2nKZUgOt61pMxfTT55l1xyCXfccQf//b//dz7+8Y8DsG3bNu69997qqtNsNgtAKpWa9NyJnyfuny0pJaVSiXK5DFD9vtIsVf0zJY9f9owykHNY0xjFMjR8z+V0yStO5bd9Wf7xV30EQnL+6gb+68u6MWSA45x6oHuC67qTvq80S1l/ISVlN6AwHiCkDDPaxm2D5phJKhohbpvELAPb1ID5dfdoWjh21RA10TWNcrmMnzBJGAGG8PAcj+lHvOqTlMtsA6Xf/va3fOADH+CNb3wjr3jFK8hkMtx55528/e1v55//+Z+rg9614Hkeu3fvrv7c29tbs9daDhaz/jlH8NSAw2hZsCqp01+e/R+tLyRHcj77x3x+c9RBAttaLa5ZrzPQf2TWxxsYHJj1c+rJYtXfCyQFV1AJJEiImBpJW6c1ZtAQ0UnYOglLwxQalIASVAi/FsJ0l5cr+bNfq+78mgSM2267jUsvvZQPfehD1dt27NjBK17xCr773e/yR3/0R6TTaQDy+TxtbW3Vx+Vy4UjlxP2zZVkWmzdvplwu09vby/r164nF5jbQupwtdv1Hiy49PaNYKZeXdM987wopJUezDi8MFNg9kGffYBE3ON5d8bKNTfzxi9fMesqi67oMDA7Q0d6xIsfCFqP+gZBkyx65io9laGyOW3Q2RmlO2DTFLRoi5pLsi7HSP/t79+6t2bFrEjD279/Pq171qkm3rVq1iqamJg4dOgTAxo0bgXAsY+L/Ez9bljVlvcZMaZpGPH58cDUWi036eaVZjPqPFV0eOzxGxpWcs6rxjGssfCF4vHeMZ8a39Tx5imwqarJtdYqL1jbykvVN8zrp2La9onMpLVT9HT+g7AY4vsD1BZ4QaEBjPMK2tc2saYrRloycVRMRVupnv5ZBuiYBo7Ozk+eee27SbUeOHGFsbIw1a9YA0NXVxfr163nooYd49atfXX3cgw8+yGWXXbYirwqXIy8QPH5wlMF8hfUtiTMGiyOZMl/5xQEOjpSqt9mGzjnjO7Zt70yx9oTtJpWlVXJ9hgoOGpCMWDTGw690zKYhatKRitZN2gvlzGoSMG688UY++clPctttt3H11VeTyWS46667aGlp4Zprrqk+7j3veQ/ve9/76O7uZufOnTz44IPs2rWLr371q7UollIDe/rz9AwXWdsYP22wCITkh8/1890nj+ILSdw2uPKcNs7rTLG5Xe3YdjYRQlJ0fYYLLraps6ktyTkdDbQlI8s2y6qyMGoSMG666SZs2+ZrX/sa3/rWt0gkEuzYsYPPfe5zNDU1VR933XXXUS6Xueeee7j77rvZsGEDd9xxBxdffHEtiqUssMF8hacOZ6oLnKbj+oI9A3n+7amj9AwXAbhwbZqbLl131u8utty4vqDk+mG3kesyVg6I5B1s58xTWAMRpr6QhCnT47bB9tUpNnckaW+IqBafAtQoYGiaxh//8R/zx3/8x2d87A033MANN9xQi2IoNeT6gicOZSi5PutP2kK1P1fhiUNjPHc0x97BQnWVbcwyuPGlXbxsY8tZcwJyfUG27BFIiZQSIY8vLl0OJOGYkJQatqkRs0wa4xaOKSlEdNJxi6h95qzAlmnQkgi7mRIRk2TUJLUEe0YrZze1AkqZk+f7c/QOF+luPj6oOFp0+e6TR/hlzwgnnnOb4hbnd6Z57UWdNCeWvlUhZDi7J1vy0HRoSUSI2QaGrmFoYOg6Z0k8m5GEbZIeT4DXEDWJWgalUond2gjbtrWvyIFfpTZUwFBm7Vi2zK7DGZrH0ygUHJ/vP3OMH+0erLYmzutMceGaNNs7U6xKRc+KFkXFCxgpuDh+QCpucf7aNF3NcToazq7ZPYpytlIBQ5kRKSUjRZeeoQL7B4vVRIJP9mX4yi8OUHLD1ddbOpK84UVr2diWPMMRF4eQkrGSy1gpHMBdnY6yqT3J6nSMRET9+SvKbKhPjHJGRzNl9gzk6RstUXIDWhI2LYkY39t1lO8+eRQJrGmM8YYXr+X8ztRZ0ZoAKLiC3tEy7ekEF3c30d0cpy0ZWRH7FihKLaiAoZySHwiePpJl1+EsvhC0JML9ACpewP/+2QEePzQGwCu3tvFHL+k6a7Y8rXgBB0dLuIHk8u40F65rU60JRVkA6lOkTKvo+Pymd5Q9A3laEhHSMQvHD3jmSJZvPH6YI5kyhq7xJzu7ufKctjMfcAEJIamMrzx2AzHpPj+QCCnZ1BrHika5aG2auAoWirIg1CdJmWIwV+HRA6MczZSxdI2f7xtm97Ec+06YIpuKmvz5KzazuX1xxiq8QDCYc3CDAE3TiFo6McukKWFP6mLSNdjQmqQtBi88P7QoZVOUlUIFjDrn+AF+xQvz/wQCz5dIjs95lTLMEFvxAoqOT67iMZR3GMhVeKx3bMoU2eaEzfmdKa67cHGmyLq+YDBfwReSzsYYm9uT4aY3UZOEbZ5yn41SqTTt7YqizJ0KGHWo5Pr0jpR4ot/h2fIAmmkSCEkQSAIJGidtV4kGSAxNwxeSR3tG+M99w3jB8SmyF61tZHtnio4ar/qd2D2t7AYUnQCJpLMpxrmrUnQ1xdT0V0VZQipg1JGBXIV9AwX6xkqM5IoM5TzOaQxTPeQdn7GiS9GdvPmQLwQjBZehvMNgwWE471S7nc5pT/KGF69lUw2nyLq+oOj6lJwAJxBoSKKWQTxisqkxSndzQgUKRTlLqIBRB6SU7B0s8FjvKEXHJxkxOZZ1ePhAmX95bi9lT5z5ICfoTEf5wxet5aK16VO2JnwRprl2xr88X4QpKgi3zJTj/+OE1oxW/Te8TyKxDJ1ExKSrJUZ7Q5R0LFyt3BC15rytq6IotaECxjLn+oKnDmfY1ZelUPF4+miWX+4fqS6km5COWbQlI6Tj1qQNMXVNozlh09YQoS0Zoa0hQmvSRtM0AiEpVDwKFR8vEJPGPnRdI2LqREyD5oRNOmaRjJjYpk7E1LFNHU3TkFJWx0CElAQinMXkC4mpa6RiFumYRcRUKbIV5WynAsYyli17PNY7yqMHRnm0Z4SnDh/fqLIlYXFei8EV27pY05Kc0QnZC8LWwnDBpej4SCQNUYuullg1I23E1LEMfbzbyCBmGSo1uaKsECpgLANSSspeQKHik6v4ZEsuA3mHw6Mlvv9MP08dziBk2Nlz0dpGrtraxubmCEeOHmZNYxRd08iWvXAw2Qs4ORnrRMvB1LUwENgmWzqStKeitCRt4rb6M1EURQWMReH6gkzJ5eSk2RMn7okTtpCMz2QKu20cTzBWchkpOJTc8YVqQpApOjzVl+Oxg2PVhWsXrk3zhxevYW1TmJk0XywzVArwR0okYhFilkFj3OKcZBLrhL0rNE3D1DUSEZOEHQ42xy1Dpc9QFGUKFTBqbCjv8NtDYxwZK00JGADI48PC45OTwtTaMgwkpqYTtXUsXePwWImf7xvhhYF89ekbWxO84cVr2dLRAEDZDRgqODiuQ1NE58pzWljV3EAqaqmtNBVFmRcVMGrEDwR7Bgo81Zeh4Hp0pmMYuoZ2wpCzdtJ/ptviNBCSX+wb5t+eOkqm7IUP1+DCNWmu2tLG+Z0pyp6gP1eh5PpYhk53S5yuVIrs0QKb2xLE49Ea11ZRlJVABYwaGC26PNWXYe9ggXTUZEPL7NcxSCl5oi/Dt397hP5cBQhnOr38nFZevrmVVMyiP1ehd7REzDZojttcsCZFRypGe0OESqVMoV91KymKsnBUwFggQkgG8w49wwV6h4sUHJ816RiRWXQDSSk5mqnw7LEsv+kd48D4HtjJiMl1F67mqi1tGLrGcMGhb6zEupYEm9uTtIxPaz1b0oorilKfVMBYAAdHiuwZyHN4rEwQyHBdQ/L0KTQqXsBwwWEo7zBUcOgbLfPcsRzZ8W4nANvU+d1tHfzueR3EbZOC4zOQq9CcsLlqSzsb2xJqSquiKItGBYx5Gso7/HzfMK4naGuITBlY9oXgyb4MR8bKDE0EiLxDruJPezzb0DmnI8n21Sl2bmimMR4m+BvIVXB9wQVr05y/Jk0qatW8bkr9CQLB0T0ZBg5kEbNLALBseJ7L0FCFyqE+LGvp95BfbFqbj12jlP4qYMzTvsE8RcdnY+vkcQopJY8fHOPbTxxhMO9M+9yEbYQrrBsidKSibO1oYHN7clKrYaKbyjI1rtzaxsbWxLLrehJC4p+08ryWNE3DcwICT+I5AZ4x/WuXcg7Z4cqilWs6xbEKfbvHGDlSQATTzqObEykFjuOyO/Ikmnb876kwVsF36zRSnOQIR5a6CEtixx81YNRoMzMVMOZhpOCwf6hAWzIy6fbdx3J867eH6R0JU2w3RE0u7mo8Kf1G5Iy7wAkp6Rsr0RCxeNnmluoai5kIU3Ic/5r5E8FzApyyT+DP88QiwSn7jB4r4BSnb1EtJLcSMNZfJD9aIfADyuUKB378HPo0H55S1qUwNn0grycVpgZE09ZpaImxzK47ZkxKQaXiEI1GJgXLlUI3Ts5HvXBUwJiH/YMFik7AqlQMCK+k7//1IR7eE27cEzF1fv+8VfzO9o5Zr4HwhaBvtExL1OIlaxtptSwqBa/6Op4T4Dk+pZxLOe8hxeQ/Ecd1GDrqYRWHiNiR6V5iWoEvGDlaJDtUZsqS8FnyPUFuqLykJ+YixdPeH4mbsIQnTsPUSbVGaV6VwIgs3DoZ3/MYHR2jubkJ0zrefWlHDdKtMayIwZJWvIYc1+HI4SOsWbtmVn/79SInj9SsF0IFjDnKlFz2DRVoHW9d+ELwlZ/38uveUTTglVvbue7C1aRik8cahCcIyj7iFBlkhZBkyi5FJ6DNMun2TMaezzB2wmOkJOy+kKCbGqZtTLpa9JyAA7tGGDhQpk87OKs/Ht8JEKJW1ye1pesaTavjpNrjCBGQz+VoSKUwjaknYjNi0NAcpWlVnEhsCT8GmkY8ZZNsjGBFFy5glEplXni+zNZz1xGPxybdZxg6Wh2v5C+VSlTsQTZsayEen3mrvF7s2tWvAsbZpmeoSL7i09YawQsEX3p4P08dzmJoGv/1ig28dENz9bFSSPyihzPq4Bc8hCuQkkkneSkkRden4gsStsX6VITWmE0ibmHak08kmgZuxWfkcJGRw3kqJ3X35IbLeM5Ev/3sxw6iCYvmzgSGNb/mvKZpNHbEaOyIYy7KKnNJpeiFAdQUHDlSZs3aZqLTXGXqhkbT6gSpthhGHc40My09vJiw9EV675WVQAWMOciWPfYM5GlO2Di+4I6f7OP5/jyWofHOl2/i3ESUymAZAL/sE5QDgooPAoyoidVggh7mmKp4gorvgwbJpiibm+J0pCJITzB8uMDAvuykMQgRSDL9JfKjpx+sTTTZpLphTfcqLHvmM6pMS8cwdbxKwEJ0hErCYCiCRRho1aC5M0FbVwpMHyc6xMZtrSvyKlNRakEFjDnoGSqQLXtsaInz9/+xl+f780RMnZtfvpEuYVDoLVRbD5qugaGBhMqoQ3kwgz/mwjQzYjLjX0/PsBypthhtXUmSzdFJrRUrYpJoNtn//CEMU5vdFbQEXddp644Tic9/6q5hakTiVthnvgi9IKYV7sNRKtV+kF1RVhoVMGZpNFth9/4xUmg8uW+U3f15LF3jz89fS2dBUsmXkYFEN3QCJ6A8VMYZqiCd2V9hp1qjNHcmMe3JJ/xkU5TWtUnsU/S9CyEZPpwlljJYf9Hs+3HtmKG6MeqA71TI9B+lEl1ZucTK5QqlkWHGjh2lEltZdQcQgV+z2WEqYMyAlJJixiEzWOL5vWNkD2dpSth8r+cYAJc1N9ApDNxRl8yzY4jK1HEDTYdUR5z27gbauxqIJk9/9a6bOpZ9+pO27wmcojc+XnG8xSIDSDZHQDOJp2zi8ZW3eKleBb5PcWwUcbpVd1IyOjjA4PPP4h/rw7ZX1u/fcz1GBgcw8plZdcfWC62tE61G43IqYJyGFJJi1mXkaJ7sYBnPEwwWHaLNEQ44PofKLpau8arVTTjDFbLPZ8Mc5ZaGbhtYtk7z6gSd69O0rElizmEQOfDElFlLvhvglHx0QyPWYNHWncQwJx/biEl6egfnVf9647supVxmyhRkKSXOvr2IXG6JSjYzvu8zOjJEsZA/49oaz/MxikVikhUWMCSm45LI5IiZNpax8k5xleYOZFCbhbIr792cASklueEyI0eLFEYrBL4k0RihXPHI7auQ0DSe7xvl1SWLbVgU/vP4iVlvsUld1MKODU3ET1qYJ4TEn1gU583gFyrBsHT0k64WDFOjY2OKdEuMWMqedrOjUqk0t8qfZZxSCadYmPXzyuUylWyG/PAQfixGpVhguO8gpUxm0t7kAMHoCE7PAUDjrF7NJiWGaRCLxM64ktf3fPxCETE4hG+trI954Pto2SyBrqOZ5tn9O60FUbusCivrL2kGAl8wcDDH0MFwk6JYysayDaQQPP8fRwkOFsgBL0UjfPvCvVHtpghmewSvI8qmjiS2ruG7AW4lwC37IMNxDcPSSaZtEk3RM/4dWxGDSNya0jIJp0suzzEGKSVOsYhTKnK6aVhSQmFshNEjfTil0qR9RGbC9VyGBgbQMsPYlo2UAjseJ93RgX7Cuowgl6fSc5BYYzNGunGOtTr7SN9DSoHR0oxprqxuGel7SCEwmppWXN0BfE0HWZtZiSpgnCAIBId2jzJ2rEgiHZk0qPzMI8coHAyvdAcjUA4Cog0W2zY1E2mJIk0YzDt0x2xsR1ARHmhgRQ1Wb0wRbbAxLQM7amBFjWWXD2quKoUCmYFj5EdHAInvujjFIr7rcMZ5uxKiDSmaVnfO+v1yHIe8F9C4ajWRyPSrfWUQ4OzvIaiUMNvaZ3V8RVmJVMA4QXHMITtQoqElNumq/tBzI/Q9NQKAuynBP40MY2oaH7mki9h4t9NgrkJaaKxuiNJ1TiMNLeHqWitiTBlfmAkpJb7jUCkVCLzZTxGtlMuUM6PkBgdwY7EzP+H4C+OUSxTHRvGc+SXmk1Lilko4pSLW+EwdXdeJJOIkm5uXPGgGo2MEI8OYzS2zbsEoykqkAsYJsuOL7U4MFrnhMs/951EA7I1Jvu2FrYxLV6VJR0yELyjkHHQ3YOPGJjZd0Ea6bRYnaKBSLJAfGaacz1LKZBBBgJSSwPPwnApyDs1L1/UYGhhEz45iz2KmyMRYqmlZGOb8/zwM26apcc2SB4eTSSHxjh0DXUdbgQOjijIX6pMyzil5ZIfLRJPHZ5T4nuCJHx4K8zY122RW2+x/toyhwZXpBpwxB6lDUZdc+KIOLjm/fUoaj9PxXZdj+/YwdvQwTqmIbhpYdgRtfEDTjNjEUqlJfe4zrs8MumRWsiCbxR8ZQW9ILXVRFGXZUAFjXH60glvxiaUsBg/mOPLCGCOHC7iVAD2iY52b5v8eDrPQ7uxIs6ozAYZOBsHq5ggvPr991gPRY8eO0L/vBeLpRpo6z76r8FqRQYAsV6bMVlpIwnGhXEYUigSeN+V+v78fAh99RU05VZT5UQGDcLB79FgRtxLw6387wOjR4ymxzYgB21KMaAF7s2Hr4rUvWUu8KYYvBJXREpd1pmefvtzzGDrUix2LEU0mz/yEZUR4Pvjh1D7pOohiAem44X2uSzA2hnCceadPPx3f99DHMlQG+vGnmynjeeip5dW6CIKATGaUSuX0U6aDIGBkdJjRkSGECKjXNOanJk9I7rnS6g4vWbeeaI0uhFZ8wJBCMtSbIz9S4emfHMYthwvius9roXV9kqNBwGDe4UcHw9bFpWubaGsKxyiG8g6rUlG6W2af3C470E8xM0Zjx6oFrQ+APziEl8uhDQzgAZq1eFMLg3wOkctVFw5JPwDPP/651XS0aBQ9noAzpNgWQlAsFgjmMK88EAFFBKYI0H1JJpthbGwYd7y1EZ5Lls/JREpJsVgYDwDKTNTweuQsV7uKr/iAMXw4T/+BHOWCh1v2sWMml79hM3bC5PljOfrzFYpSsCdbQtfguhd1AuAHgooXcNmmFiLm6VsXIggmdYsIIRg6dAArYs9pfOJ0/JFRyrt345VLaJksXj6HXIDB6xmRgGmiR6JokTBIaXETqWlIGa5Yz2bHGDnSi+ueflMl3/cZGxvG91USwRPZdoRUqvEMsU4jkUiiGxYdHauwFuv3f5bwfJ+xsTGamppWXN0BsKPUKmiswHfzuMKYQ39PDjtuMvx0OG129eY0kYTF3oE8fWNlWhI2390dzpK6tKuJtnTYuhjMO6xKx+huTlSPV87nCE48wUlJKZdl5EgfXrk86bUrxQKNq1YvaH1E2cHZvx+CIFxXoOkYra2zWrwUBAFjYyPkchnm+0fn+z6jo8Nks2NnfvApmKaJZc2+eS2lRAiBrofZa5PJBlpa2onHEyzXbopoNEYy2TCjsS7P9xgeHiYWi2OtsMVrpu9RLJZWZN0BKppWs+bVig4YxWwF3wuIp20GDmQBWLUxzYHhIr0jRZrjNsdKLs9nSujAdTvC1oUXCCq+YNvqFPb4GovsYD8Hn34Sz3GQnocMBCARQYAdjWFHwnUI0vcQpSJRL8A/1MdEeJFSUshlKRXCFeZOpczw4DGyY6PI0yWaO5GUSBGg6QaScA8K3dBntcbA89zTJ7abp2g0RmtrO4lEktOduDVNI51uJp1unNNkgIkTZmtr64o8aShKLazYgCGFJDdUwYqYDPcV8F1BJG6Sj0h6BoqkYxYR0+AHh8KWx0vXNNLeeHzsojNpsVZ38EddyoU8B59+ErdYIhoE+AMDiPFBX2t8imy1Q0ocHxAu+h4juQwj+SyjuQyuP3U2z9yccJw59OhEIlGamlow5rk+QdM0GhubaG5uq07t1fXlmdJEUZQVHDAqJY9K0SOSsNj3eJg8sLE7yYGhEnHbIGaZHMpXeG6siAa85uKwdeEWi8i9z9FNlsHHi/i+RyafI18qkoon8IVATyTRo9Mv3tPGE6JVKmV+9chPJ/Xl67pOQ0MaTdMwDIOmplaam+d24vYDn0wmQ2NjI+Ysnm8YJvF4YsVM8VUUZeZqGjC+853v8E//9E/s37+feDzOBRdcwB133EF0PE3Ej3/8Yz73uc9x4MABOjs7efvb3871119fyyJVlXIuvhsQiZvV7qhK2iTwBBGhUyyW+XpPPwCXrE6zujkOgU/m4f9H/Mgexhpj5OLh+IWmazSu6gy7f2Z4ct6//wVc1yEajbFq1VpaW9tpbGzGWKBBcM/3cByPhoa06pJRFGVB1Cxg3HXXXdxzzz28853vZMeOHYyNjfHII48QjE+3fOyxx7j55pt5wxvewIc//GF+9atf8T/+x/8gkUjw+7//+7UqVlVhtIJmaOz66WF8V2AlTbKGpEXoGCmDHxzJcLTikrQNbrisG4TA2/Uk2qG9NHe20tTagqZpc7oSLxYLHDlyEIALL7yEpqaWha6eoijKgqtJwOjp6eGOO+7gzjvv5Kqrrqre/nu/93vV/991111ceOGFfPzjHwfg0ksvpa+vjy984Qs1DxieE1DIOAz3FRjoyaHpoHcZNJdLRKIm+0oFfnZwFIC3XdhEe6YfrWeE7K7HiaRitLY2o59hP4JJr+e5kzZB2rdvN1JKWls7VLBQFGXZqEnA+Pa3v83atWsnBYsTua7Lo48+yvve975Jt1977bV873vf4/Dhw6xdu7YWRQOgOJQj0zPA3l9nAGhIFYgfO4BBlJLt8g9uC2DwWjvDS/f2gBTk3QDf0ljVnMKY4cBtoZBjz55nGRoamPb+c87ZtkA1UhRFqb2aBIynnnqKLVu2cOedd/J//s//IZ/Pc/755/NXf/VXXHTRRRw6dAjP89i4ceOk523atAkIWyi1ChjSdck99iSju3MI0UDEdDC1EezmFsquyT2OpITO5qjg+nUx3KAd1/cYyRZIC4d0smHKMYMg4JlnfsvIyOQtUb1pchhN6O7eSCrVuNDVmxmhoQVLN1tJEk7/PdNU8Ym1FGfajnQ6vgiQroZXCpDjK8qFnNuxzh4SKeX4FrOnr4cfBHhFn6JewjzDwtJ64/sBXtFbkXUHMKSsWbr+mgSMoaEhnnnmGfbs2cNHP/pRYrEYX/rSl3jrW9/KD3/4Q7LZcJA5dVIun4mfJ+6fCyklpVKJ8vhCufIJC+aklDhPPUVu30FKhLOeZFRgN6ZxvYAHHY0Drklcl7ytNU9hNEM5P0auXCFum8RTNhKJ509etf3M079leHj6VkRb2yo2bto6vu7gxIJCUJFowgBZg1+uL4kECSiaiBM+NJrQMLwIujh1l5okrNeCrMeQMgwQ4ydqKcM8P7o2g8V4moahaXP64zdEQKMfIZKPhFN5NdB1DY1lfALRQNe0GXWHCk0Qj7SSMOMLNpFiuQgISEZLxFdg3QGyWGjaMlq4N3HS/vznP8+5554LwEUXXcTVV1/NV7/6Va644opavCwQXtXv3r27+nNvb2/1/9rRo+hP7WJMW0W+GL6hbpAlO3yMQxXJw/FzQYPrvOco9o5QCHyEbuFh0BIVuG64GGxCEPgcOdxLJjOCpmms695CNHJ8Oq1uGNhWBKfk4RXzmMLGlOGJUpcGhtThNKdDIQICEcx5t0WLBJWSANzqbRIQFPDxmP4qNbxN1xdoV8DxE5w+fixN17EsG93UZxQGNH1uKZ8MDOKJ4zm+NCYmKNRuUeJiEIBgZvmkIg0WPt7473oFMcBOrtC6AyDRtNlv2jYTNQkYqVSKxsbGarAAaGxsZPv27ezbt4/XvOY1AOTz+UnPy+VyAKTT6Tm/tmVZbN68mXK5TG9vL+vXrycWiyFdl+LBg4i1awm8VbiHw6qbchBNFPlhbDtS07jYGOOliQBoQtMNcpWAzlSE1ekIMpAMHjuG7wX4vkff0R4830VDY8c5V7Aq3X3K078mNEBDauEJSwLSCECTCBlezXuehwgCNE1DIjF0g2gkjj2H1BhB4FMoFkgmkrNax6FpYJoWdiQyq/UbpzvgUuxm5/se2WyWdDq9Mvd1XsH1X8l1Bzg2iwk5s1WTgLF582YOHTo07X2O49Dd3Y1lWfT09PDyl7+8el9PTw/AlLGN2dA0jXj8+JVlLBYjHo/jHDmCkclgr+2m/II+nv5YoJHjt3Y3R2WMKAFvaMgQ0cPnFyse7VaaLiuJ5RjsP/gcL/Q9Oen1EtEUF3RfSlvjaoQu4FRNQU2AdkLSVsD3BU6lUu1mSMaTxBKJajPaNEws257TCTccP9FpaWnFWsRstWcb07RU/Vdo/Vds3TVqlrC2JqHola98JZlMZlLX0NjYGM8++yznnXcetm2zc+dOfvCDH0x63oMPPsimTZsWfMBbOA6V559Hi8URUicf7rKKblYo6zY/FOHU1tfGR0jpYXM/CCRxL8Eas4VoECMoSfYffRaAtqZOOtvWc8HmnVz5otfQ0taGsHwwBOhy+q+Tz/kSHKdMOt3IqtVrWLV6LW0dq2hIpojHEsRjCWw7ovaaVhTlrFGTFsarX/1qLrjgAv7iL/6CW265hUgkwt13341t2/yX//JfAHjXu97FTTfdxN/8zd9wzTXX8Oijj/K9732Pv//7v1/w8rgHD+IPDmF1d+P4VAOGpuX5odmFo+ms0ytcaRcwhIEpTbySTqvVQCQuEHrAviNP4QceDfFGXrLtFfPu33dcB9sKU1WvxGazoijLT00Chq7r3H333XzqU5/i1ltvxfM8LrnkEu6//37a2toAuOSSS7j99tv53Oc+xze/+U06Ozu57bbbuOaaaxa8PP7wMFgWmmEQOJLi+IZl/ZrLs3orGpK32y7NbhMaEAQaEU1gRMocHeknCAIOHHsBgK3rLpp1sBBBUF3hXi1T4NHa0q6ChaIoy0bNUoM0NzfzP//n/zztY171qlfxqle9qlZFOC4QaOMDQa4L5Up4839Y4eD6NYbLOt1HaBIpBfnAI6IXefSZX1Jxj2+H2djQSnvTmpm/rgxbEkHgYZqTB64T8STJxNQ1HYqiLAxRyOEV82d+YJ2RnefMOKfdbK2MbLUnbGuZyYZ7i0hN0GdYpKTHy/KPsW84A4AvJLoG2fxRhAiIRRIk42l0zWBr94WnbF2IIKBcKU8eq5DhrK3Wtg7iseTk0Qh9aWYPKUq9E8U8/PwhMs89Pumzv1LI8/4OLdVck2OviIAhRXVHeMbG1wQWDZ+oqPAHgw8xUDk27fPamjq5eMvlWObpp7UGgU/FqdDQkCJyUlrzSCSCbUXmXwlFUc7IHx0ie8/fopWLAOipJqjhNNOz0nKbVnvWOSEdRCH8OyKLww1Hv03az6LrBqtat+IJjailk4iYRO04na3dZ1wA4/sejuuQTjXS1NhS7fpSFGVxSSnJP/h1ZLmIbGoldc0fEd9y/lIXa9GVG2I1m1a7IgKGlKLawpjIFFJyD9DlZ9GMGNs2XoFpJQmEZFVjlMgM0wl4nofnuTQ1toxvJaqChaIsFee5J3D3PQeGAb/3R1gbti51kerOiggYiDBgCCEpj29wZ3iHAWhu7CIWSVFwfZoT9hmDhes61VxSuqbT1NRCKt2oxiOWMe/oISq7Hl3miQknE0JAuUwxFptVKv7lzHn2twBEL/sdyo1q24BaWCEBIxzDCAJwnLDp2uiMB4xkO04giJg6yejp3w4RBPi+T3NTC4ZpousGsVhcBYtlLMhnGfvq7chSYamLsuA0wDnjo+qL0dRG7PLfoTyPBKbKqa2MgCEFGuD54HogRYZ4UECgk0w0U/YFbQ0RrNNdiUkoVcqkUilS6aZlGySkBFkBWeeTR4SvYbg2oqgRnOKvXApJ4Yc/RI+tRl/dgt29aXELWUNCShynQiQSrSZ+rHu6jr3hXKRjn/F3X9dk7XKDrIi3c2KWVLEcnjADP9weNYi14QY6MUsnGZnmrZBQcUrjqb4lkWiUxnTznIOFDMJjSgGiBKKgIf15VOw0hLCIuc34JQsxvh8EQiMoAf5KOIEYJGgnGOW0uV3tzTdib160Qi0anRXy4T6JPwwMM6Pffd06RwNTBYy5Gx/0Loz3OnjeIXSgId6GRJKOW9NehZUrJUzTpLExXOAXiUSnXZktXBB5DeFOuSsUQFDQkJXFPVFbxJGVaT40ukSr8wXmUkqCIMAwTpGmXUqC0UGklBjJNNpJ06GXuzPWv46t5LoDU/PWLaCVETDG99MulMZnTPl9AHTEWrBNg5hl4LrOpB3gRBBg6CbN9mqsIAICglEoF7Wp+1MEs/8NaRGJ0SDRarREIwgCisUCiURy0iYyekyix8N9JuqZ53nkRoZPma23/NSjFP/jHzGaWmn5i4+jact7n4yTnan+9Wwl1x0AU6hptfMxMa22VAIZDKFLB0+ziEbTRC2dSrmIZdoY5vGzqEGMZG4VclTnVA2HE14BPR6ejE8V3fUEGA2SiQ3fan7h4wV4FDFaYlhWnUeHOag89SgA0Yt2rsyrUGXR+SJg1C0RzHVHtBkKZBRjOW3RetYZn1ZbKIHww306KtGO8KztO1jxGK2tHdh2JOzCyWt4Q1q497Up0aKE22PGJUZyaneOZoG28naCXLaC7Bhuz/MARC/cucSlqV9yfHve2Qik5Fg5y8HCGMXAmdOVspAC13Gw/RH0s6QpnfcdBio5xCJM3b5609XE9Nq0rOo+YEgpqyu9iyWB8I8AEI21owU+jaKVdKadYEinfNJz9aQksl6g1f27tLJUnv41ILG6N2M2ty11cZYFXwQ4wmcm5zshBS/kBnlq7DClYAm3SPXOvsSDhqah12YboioNbdaBeqbq/1QoZXiVomkUSkE1YHRGW9giNxIrxyc/XpPoSTBSErNV1r7rqM5Jz8U7ehDpeSACvCO9uD0vIIq52r6uBIKAjGFM+R2KfPja0Ytq07qoBB7eeNK7QAoqvkdZeDM62S6UIPDJeQXyRW1WW/SeSNOg4DkcK+cYqOQpz/DkH4gAbx7dLhHdpNGOETUs5tLEkELiei62ZaPpZ8cH2NQMWiIJGq0YVo0yyU7QDFONYcxZ9VOq4ZRHYHz84gJ7PTHioEusVRKzefxx+vIeEJZSIvJZAs+FQpbAMtDNxfk1h109u/H7D4OUCKeC19cDQY3mDp+BBpzqtKXFEkTPe/GsVncHUlAOPJzAP8XnUVLwHQqeUw0YAL4QNbviOxUpJCW/gucYczpp+kJwqDTGkDO3BY22brAu0cz29Gpixsy7RzQgalgYuj7n98z3fEbHRmluasa0zp5TnKUbWFrtZ24dquEJ7Ox5N2tFCEDieBrCOwpAvOE82rQw/W9ko8BILmH5FpAo5sl841683j1A+OE7G9a76sk0ejLc+8No6SCyaRtGS8eCjvxLKSgFXnVAMfADcoUcyWTDpFli1cc3tdDnlyg72Rn3K0vAFT5CymnX4gRSkPfKlHwPTdMQUjDqlsl6J3d2LiJnZN6HaIskabLjRHVzhlM2NZKmzbpkC23Rxf9weVKnrJvETRtLbVC2oOo/YABISSYvEX4fGhpXpl8GQJB0MJKL+xaIYgH30D6ks7AnERkEFH/2fUR2FNDA0JHVrO6L0yzXIlHs9VvCFdOWjaYbWGvXY7SumvNVVXhFf+quEEnYZz7qlCj4TvWaNPB9ihGTRCIxbS4lKT2KxSJu4M+obFJC3qsw5paoiOlbTEKKU14T64ucGUBW/517ToLWSJL1iWYipknUsEiaEcwZXr3ahkGznZjjKytnq/oPGOOD3sMZH+EfoSuxlVazAZ+AWPccV2xLSTAygHQqJ9wI/sgA7v7n8IcHpn+e5xIM9VOzDkbAaG6j8cZ3IZtaGRkZpvksmosupaz278+kj7vgOWS9Mp4IznDakxiaTtKMYGg6UkqOuBkOulkI8kwXMIu+g1+D6Y0xw6I1ksDSw1ZNgxmlNZqcVbfMQgiCgHw+R0NDatoW1kxUAg9XBHQlmmiyY2fNjCNl6ayQgAF9w6MgK7RE1gJQskuk7PgZnjyZKBYo/Oi7OHueRhTm3tljtneipxrn/PxTMZpaSV79OvRYAs9bwtkp0wikoL+cY8QpVvv0z3Ttq2kQ0y2S9ulXN0op6a/k6SkMIwlbAllvPJifJiZYmkHajs74Gjxu2rRFkjRYEaYLQrqmEdXNSS2WQAgqwifjLm63lBABxcBFeGX0YO5zvtfE0zTbcbVWRQFWQMCQUlJxHQYzwwA0RrsAZj1uISplxr56O/6xcB0HpoV+0p7ceqIBe+O5WGvWo+nTfEg1DXN1F0ZDetb1WM7cwOdoOcuIUyRu2CQse0FOQJ4IGHYK7M8PkzlpnMDQNDrNJO2J9LRdUhHDIm1FZ/V6AokrgvEB7amtxECCE/iT7tM1nahh0hyJL2q3lO/7ZCsB6UgD5hwnPVh6OLNHBQtlQt0HDKSkVC7iuHlMNFrsME++nZp581oUC2S+cQ/+sUNo8STp/+/PsNdvQTtLunrONlJKJnrzc26F/kqOku+SMqMYM9ybQUjJoeIovcXRU3QdSSonzL4yNI3ueDNRw0TXdDrsJE6xSNSOYUwTvAWSnF8ZH8CeGV3TsManfJ7q5G/pBlHDwhyvp65pxAxr0btzPM/DKrm0xFJnTZeksvzVf8AAyuUyml8gZbVgagYBgnhDuE93kBml+IsfEIwOTftcUSyE00SRaHaUpj+9Gatz3SKWvrYWctMggSTvVRiuFHHGB4bdwMfQdBqtGI7wOVrMMuwUcMXp84gWfWdGi74Spk1HtIGNydbxefsh1/cYFi6miCGnuULWNI1mO07ajmHM8GSuoRE1TMzpWo+KsgLUf8CQEiEEWpCnORZ2R/nlISo//DElp0Ll6d/MaJ2A2bGWhmvfWFfBIu9VOFbOLlhuGymhEvjomoY9flJNmBECKXgme4xDxdFZDffbusGWhnYaTzHWFDHMSYPJYSoKkEhyXoWkbrMp2UrEtqc8VwN14leUWar/gCEEQgQQ5GiOrAJAO/wUpWd/Un2Ite4cYjsug2m6SzTTwureVLNxh1qt/pUyXLzlC4EWnBAQJCDCrqLD5Qy+DKon9/nTSOrhTKWJyDBcKfBY9lA1KKXNKG2RBhKGzemm+xqaRpudPP1JXVLdT6QSeJQDr5qmPqXFSGpR7MDC8KY/Rn3lp51M+hqabyBdLVyKtIKs5LoDagOleZGSilNBQ9AUWQ1AkDlI/LJXhcFg3TnYm7YtyMDeaS/UxfgGShOPCS+Fkdrc1q9JKcPcPtPfS84pc8QpMTgq0E+aVqnp4BOg6xqNZu32gRhzS9VgkbZibGvooDWysAu5PBFQCFwsTWdNopGkFc6oMoQk52XAlCs0F5hE6mKF1n8l1x3QpEoNMh+lchkdnUa7HQD9nI00vOIVM3quDBeKh04+6U88ZvyfU3aFa+GXZoE2sROWDro9nuV2lgGj4vsMFHNkKpVTrlJ2TZeSXaQlHWFSvNDC1wz74nW8BUwOV/Y8hopFspUKUkoGi0UCKWhLJLi0u3t8wHt2f8kVz8MJph/vEEKiaxrtsRgdDQ0kTuh6chyHYT+P2d6IPd1uinVOOgGuuzLrv5LrDoTpjWrUdVH376aUknLZIW23YWgGvl8hcfnLZ/Rc4YL0qF6laBroEcA66ZehgW6FAeFUJ39NB4y5Z8Moex4lz6XkemTKZSq+TzJqTTtlFED4OoFTIBo1sU7KpyOkpHdsjBeGhnD82uZ5aorF2NnVNWl2lJRyfC3G6ZVcD0PXaY5PPyVV1zVSkSgNkYia+qmsXF4RyjmqF2Ox9aqFMWdSUjraT3M0HKwe1AI2TZOQTE5cxI7vuS398ErcbJEYE702+uLse+EFAQXXwRfhzlkF1yVbqeAFAZqmYesaUgqGisVpn1/xfQbyecZKJbTs1AWGYvyEvdAMXac1HqclHsfQdUxdpzOVwjSMapCo+D5eEJwxY6emQUMkQmcqRTJSo20JFeVsJQX45TMPcpZGYWQ/uIXq1ajW2jk+jrHw6j5gBJkMlbFhutdfCsCQpbH5hKtV6YctCU2DiTT1mgF6g8RIgD67tV1z4gUBo6USQkqEFIyVK5Q9r9pa0dGIWyYJyyTrODx5rP+UwWKKU3Tn2IbBuW1trGtqmnY/87ma7kq/5HlUPB/LMIiaFp2pNA2RyBlf1zKMBS2bosyaV4Li8BkGKBdY4EFpCNzimQOG74Idg4ZVx/vENaNm5a37gFH413+lZOg02eEMqazuA+EVq3AAGe59YSROGIMwa7+F6sT6h5LncTiTIetUqifHiGGQjoYnVDcIGC4WOTA6wmCxGAYSwgVhTbHpB6xNXacpGsNwHVqamjHNqc2ihG3PeBHdfPhC4Pg+XY2NpKNRoqapuo/qnZMLvxY3o3uV5jpY5UFkfwUnX5x7+sXAgcJgeOJezC01NRkOcJqRM++1oOugecDxrMRynajZFg11HzD8Y/24tk3aDndWy5nhFbcMwi+rVS5qevOK5zFSKpGplJEyzLHkBwFNsRi6piGl5JmBAXpGRqb9vOmaRmcqxbb29kmDvCfzPJ/h4SEaIpEpYxiLKec4NMfitCeTZ1drQYpTX71JP7zKW8yryhMJAYF7Qj/p7Gmuh1kZQSvY4C7iSu9yBnKHwasswsb109N9n9jQCGLAJLlxFWbj7HLGVUkJkU0Qa1yyusyFHpnbxlMzUf8Bw/dIxtvRNZ2SEEgj/DsQFTAaQK9xBuai6zJSLFIabxm4gY/j+UQsE13TMHWdhGVVr7qfGxxk/8jkPQxSkQhtiQTtySQt8Xg17cRpm6tSIKVAigAppl5uCN/Fd/L45fy8TkxTXnbiS4aTySIarBYWeqV3wV5jJnTfJ5YZRjdH4MRcSlKGV46nmx0mAxDB0gUMKeb92rrvk8hk0OWxyfWvNQlEGyDWvHiveRJRKuOV89gXbyd+8bloc8zWu2wVdDVLaq6KnktzZD0Ax4TENkG64y2+dO22YBVScjSXZbhYxA8C7PEPraFpNMVj1QAhpGTv8DCj5TKBOD6QfVGDzhozQAMsvQJUoDgC492aZQEVceoZuYEI8Esl8v7wtLmUQGAFLpauwQIt3AuHgSSGdrxcaQPizuL3TWiej+mMoRU1OLmFpRun72IwLLBiUOO9l09JG9/2cR5/nNLz8ComMtkKKyyXVJArQnOK2Lb1Ky9Y1Fj9BwzfozkZLtg7IgUx3QAfjFYZToOdMxk2u0/R9BvMF+jPF4ibJg32CW9zUIFsFpw8rhD8Jg9DJ13snhf12GCIMKpNvJoET4ZBwgdiGnTaYJzqnBZIio5DczQ67RiGrhnYdjOmrtflmIL0PIKSQMabV9wJc6UTjovenkZbzJbVClH372hR+HSOpwTpFQHn6jqaCfp8Fjg7eRg7AIWBaePFqA9HKzoxXaJrcMTTCCQEaIx4MOxr+FJDAAINA8nWmMTWIGlAayxGXuj4470SEommaZg6JG2NRkun0dLD1sEpeJ6HLFdIxhpUtlJlxRCuh26Z0DDHcQvltOo+YJSlRsoKU5ofkIIXYaAlZpEywK+EA5AQTrErjULuGHhliKWnzGIoBpIjjkQzBIc9jT0liXuaHpm4obGz0SZtHT9OJQhHAtbGDDQNDMDSNSK6RkSffuqqoiggimX0hhgkF2E+/ApU9wHDstvQNI1S4DCkSWwDjJlcfEgB+QEY2RsGCgiz2moa2Inxec+TT9yekBwu++SFYHdRkvHCSJEwNBJGOLkvbWm0Rwxi4/EhZmiTZg9JKSkGgs6owaqo6n9VlNkQpQrW5jVoizBlfCWq+4CRsDsAGPNKVHQbOyLRpls4LAWURsJuJuGFC96KA+E4QrQxfIxuTGlRlANJ0Q/TXBQDyeGKz+5CgCPA0uC8BpPu2MwXoJUCSdzQaYuoYKEosyGDAAwdszkN5fxSF6cu1X3AiNutAGT8EpWIRSSqTW4Y+JWwmyl/bHxMQlLN1hdtBHNy09YVkqwXBohyIMl4AldIQMMVkmfzAZ6ElKmxs9EiYZ75SqcSSIqBRCdstHRFdSKnGZ9QFp4/lkM47lIXY8H4vo/MZPECDblCBn+lH2CmExjp5MoJGFLC8J4wNcg4LfHycKZfDdT9X5I+PhXKE4KKJolGxk/EUkD28HgelmLYeog1gnH6vEXHKgH9ToAxHnXihkbS1BFS8otRD09Co6lxRbONOX7SD7csPU6ekH3YEZJAQldMJ2HooEHSUMFiMfm5AlIIohvWzDpz8NnKcz20IZNIWzuWvXImPRipBNoSLlRdMF4ZKlPzwE1SHIJd/wIj+ybffvVFEG+pSbHq4J09vYm0ABKNigYxOQYlE/L9MHYQrCg0dJx5CT5Q9AVjniBl6lgaZDzJwVKAI8IWwognMTW4pNGqBotKICkFotolpQEnNh4MDbpiJs226nNdCsL1EPkS8QvPIbZp7VIXZ8E4joPeZxHr6iKywpI3Oo4z/4NIGV5Q9j8NpeH5H2/GryvCGZjDe2e+eNOIQOMJf7t67S4Q6j5gTAQCCTgI4oOPQt4IV/ImWqd0OZ3OsBNwpByQ8yVDrsCbZvbTjpRFwtBwxruZLB3WRI3qLCiNcJbTRMzQNbBV99OSkFLiD41id68mun71UhfnuOIwDD0P/txPfLrvkxobRff3L95Kb9+Bod0wuDvMtLpEbAmbAB5j7i1GyYJmQJgTM8ppK6Ab0H0pnP+GsHdknCyk1UrvORsPGAIQmkRPtUPEJNxJaPpfhpQSX07eJ8kVkmcKPs8Xj/8RmRq02TpJc3xbUFOnM6oz5klsHToiOi22PqNxDGXxaF4G3c8hXA/LdIg1N6GN7g3vdAow8CyM7JnRXu8LzitDcXDeh7GAdoDeeR9q2Zn0qZ7PedOwoG07NHWzqKv+k22w6kJIti/ea85Q3QcMjXAAO5ASQ5fhdLuTup8cITlWDvAJ/768QOJLOelvTUrJoXIYQrqiOhviJo2WNmX2U8YTNJga6+ImsbmORUwku5oHLfDQZQUtKIM++cSnSYHuHMUsH0ALCqc4wvJmC4HpOESGI8c3mZICw+lH98cmP/hXi1++09J0aN4E0bnvIx8IQblcJhaLLUpWYmC83Bug43yIL10uKcdxOXrsKJ2rO4lETp2g84wiqZoNHi9X9R8wxoODh8RCTmnhSSkZqAQMukG1a8jQNEx9clLkUVdSCsI8SRemrGlXWVfG7++MGdMHCynR3QGM8kE0MX13g+5nMEo96MH8Z3k0ARyZ92GWrQhAaertEh1pNSL9AC1ihSuDJ+gmtG4JT3r2IqYxrr6+AU3rw7U+8+A7Dv19fXR1dWGssDEMDIfALoZBa6XVvcZqHjCKxSLXXHMNAwMDfPOb3+SCCy6o3veNb3yDe++9l6NHj7JhwwZuueUWXvnKVy5sAcYDhg+YiClbfWbHxyMaTP20YwmHK2FX1NqoTrvzLK2FX2GJqZsY6RroY9MfRxMemjw7pm4KM40f24iw26ibqUEnCERAsVAgkUxOSr4orEaC2LpwC9tMntTlF6E3pZawpIqyfNQ8YNx5550E0+z69sADD/CRj3yEd77znVx66aU8+OCD3Hzzzdx///3s2LFjwV5fG89KWm1hnMATkmOVAJ3JA8++lJR8CQjsIEcgJEcrcUDjKvd7rCu/cOoXPEOfqdQsgth6hNkw/f16lCC2gSC6lpnnL5nK8zyGh4dpbW2dPpdUrXZYOUt4nkfGGcZsmL7+opjDaIhjpJagFaEoy1RNA8b+/fv553/+Zz74wQ/y0Y9+dNJ9X/jCF3jNa17De9/7XgAuvfRS9uzZwxe/+EXuueeeBSvDRMeSB1iamDTOPewE5L2AfAAHSgESKASSUVecMOB9PI/IGo6xMXiBQItQSr8MPXkOs7o61zSE1bw4u3dVU2RPHbNRwhQSkXWr0U6Z7ldRlJPVNGDcdttt3HjjjWzYsGHS7X19ffT29vL+979/0u3XXnstn/nMZ3BdF/s0u8nNxsQYhovE0mQ1cV8pkAy6gqOOZH9pagvIJsBkYpxBw8TnJdYhXrB+h4am82mIJlmi7XWWHSkE0h/f6dAPEKUK0vWo5R6e1ZXOnpx2pbNmGlgtcx9UVpSVqGYB46GHHmLPnj3cfvvtPPvss5Pu6+npAZgSSDZt2oTnefT19bFp06YFKcdEwHDQMBFoHB/oPlAKqsFifUwnoVWIyzJrzDwX5L+Dgc/R5CsYTF4GQNF/ORVNoyNS93MF5ky4HqJYRgZhOJWeB5Lq6lvN1DEbG7Dam2qaIM71XHL9A8RWdWBb01x86Bpmsxq7UJTZqMmZr1wu8+lPf5pbbrmFZHJqH3E2Gy55T6Umf2Anfp64fy6klJRKJcrlcOGQPj5/2tHAIiDnuBQDeKEo2F8KT2ovMXt5tfNjImLy62at9RyNXAJ+OC215Es6bQ3hM6/WhXQ9hOvV4AJbIisuwvMJggBZyFMu+3jTbKBUE1JDswz0ZAw9Gc7yMRJRzKYUeiIKmoam62ixSO1TtLsumhFARyvaKVqrru9Xf7f1xnHdSd9XkpVcdwhn5dfq01WTgHHXXXfR0tLC9ddfX4vDn5bneezevbv6c7VLStMQwmfXUB4HjRd8G9B5Ec9wrf//0ACBTklLAxoOcZ7j5Xj5cHprIKGMjlt2GRpyIRAgpgkbgQi/TrXSUkowDbDN2mwsn4yhdzSAaaDTMt2s0trRNIjYkIxObj14JcgsakmqBgcGluR1zxYruf4rte6itRW9RmNzCx4wjhw5wle+8hW++MUvkh8/2ZZKper3YrFIOh32Hefzedra2qrPzeVyANX758KyLDZv3ky5XKb3wAG08RaG1CRXGk+y0TrKL5w1+ERZxSDX8GNK5iqOxi4jb3cjNBshJY4IWxET1+ZCQJsuWZV1MVPp8IRsGUyK5Rpo0Qh6PHLK7hbNsjAakxjJWE0CxsQexo7rMjgwQHtHB5EFGg9aTlT9V279V1zdfQfc4+u2XqhhqqEFDxiHDx/G8zze/va3T7nvpptu4qKLLuKzn/0sEI5lbNy4sXp/T08PlmXR1dU159fXNI14fHxmkxDVro/fMX7FJda/8K3yNRyjAQuX1+k/53DqtQxbWxnvnUKMZ5KNmxrmCe+7FJKGbIZEdweJC85BT0SXxc53EdteccnnTqTqv3Lrv6zr7rvjKcvP0G9dyYUXnrFGqulLPB1qND644AFj27Zt3HfffZNu2717N5/61Kf42Mc+xgUXXEBXVxfr16/noYce4tWvfnX1cQ8++CCXXXbZgs2QIgjQx6ew6rrLl+SfkNEaMQi4MtbPQMMbyQudIJCkfQ9DC9/ypAFJOf7mSBBlB4IAq7OZxEVbMOJq+0elhoSA8uikPQ5mS3Nd7NIAWkbAYl5lSznzLKu14rnYxUEY82C6CQ8zJkFqoNVuNt/0LyvDlCR2w5l7IZo3QPt2SHcdDxK7dtWmu5saBIxUKsXOnTunve+8887jvPPOA+A973kP73vf++ju7mbnzp08+OCD7Nq1i69+9asLVxghquswHo2uIqvpxHR4aVOMpHkOY54kLnxaiwVaUnG0UyQJtNe0Yne0YLU2okeX6RWLMnOBH6a09pZg3EXKMEtqrCU8Eehzm7AgKhVK5QZEezdEF/MCRwfTXtK1P7JSoej3Iteun3/dq7tsLmJvgm5ArAlizWGqmtM+dnHf5yWbH3rddddRLpe55557uPvuu9mwYQN33HEHF1988cK9SBCgazo+Adnx/SZelLZoqFQYdjySGnRHNNJbu4huWosemz4YLIeuJ2WGpITyWPg1XXNfEl6dJdqhbVvNrtROK9IATevmlU9KlEoUi02I7m0Qn8km9vVDlEqUMhHE6pVX91pblICxc+dOXnhhajqNG264gRtuuKFmryuDAA2dg/owaOBLgybpk3EFLetWsWl1I/G4jdmcXrygIEW4Lax/mil/Uhz/mivXxXTGoBgFb5pmuRThYFngUY+5pPAcIsVBGHXBOuFCQAZhFtjOi8E+xckkkgqb+OYKGDBVlFmo7xVovo+uWew1jgHgEqVQGCXR1sQ5562jITHH7iWvEu7YJ/3Zr6XQtHBjFPM0r63pYJgcn6M1e1IYCCOCNGPTn/gMM2zyxtJ1mTpEVCrkxQFa122Y3C2hm5BeO6/U4YqyUtV3wPA8KprGEX0EANczkKkYG7d1zTxYeJXjg4+BE25woxnQvB5atsy+j1k3IZIE8zTTanVzzn3XE0SpxJi+m1XbVmazXJZKlEdNZMfKrL+i1EJdBwzpB+w3BpAaiMBAYtB9zmra22d4dSl8yBwKd8ACiLSFV6fJdkitWfQBJ0VRlKVU1wFDeB5DergY0PctrGSEjnUdMx+vKA5BwyrY/rpwo3UVIBRFWcHq+wzoBUwkKvelTiwdJzHTxIEigEoeVp0PVkwFC0VRVry6PgtK30eMj0r7mkYqbs2yddEBzRvP/FhFUZQVoK67pPBENWAEQEP0FAPJpVEojVCdXqoRDjyvfUnYulAURVHqPGAEx7ukXHTiJ6/kDlzIHQuX4a+/IszHopuEWQT1cPxCURRFAeo9YHgBQptoYejEgyyMjqc8lgFoVjjjqXtnOPtJURRFOaX6DhgntDA8dKIm4QpfNLCikOyARFvYwlAURVFOq64DhvSD44Pe6HhdL4N125a4VIqiKMtTXc+SIpDVgOGhE0k0Lm15FEVRlrH6bmGc3CVl1Xd8VBRldsp+maJXPOPjpJSIpd7nY4YCGaDXqC1Q3wHDZ1ILI27NLz+ToihzJ6Sg6BUnnXilDD+fctZZPE+t7JTJ+3kyTgZHd05Zlv5iP/vG9lHwzrxRVThBf5E3Upqjzf5mDK0257o6DxiiOkvKlSYRFTCUJeILn+HyMCOVkepJspYqlQoHSgeQo5Joael3iPSER3+hn7ybJyCYfOcCvx2u69Kf7efgoYOn3b2zElRI2SnaE+1nPKaOvmz2xTEqRs2CW10HjHAMI7yacdCJnGJHPUVZaEWvSG+2l2PFY0gkju8w5owRiGBRth/xXI+B4gCZoQyWvfSzADU04macpJ3EOEUmZm2B3hhHdygZJZqjzafd09vUTfQ6TO1fS3UdMGQgEMZ4CwODqGphLEslr0Rfvo/+Yj+BDM78BMBxHA7nDjN4ZPC0J42FpqEhkWQqGbJulpgZQ9d0dE2nI96BbSzOpkyO46BlNLoauha1/mcDaUgs3cI27EV7v1eKug4YnBAwPM3AVi2MeSl6RQ7nD3Mkf2RR+3Nzbo6MkyFqRGfcLeC5HmPeGNFyFCtY/CvsiBFhXWqduoJV6kp9BwxxvEtKYqIvkz7IWspUMhwpHKEv3zfrWR8Vv0LOzRExIzUbVJuOpVuzPvk6poNjObTH21fcFbai1MoKCBjjqUG0lR0wRsoj7Mvs42D2IEW/SMJKzPrq19ItulPd6qpZWfaklAyVhziQPXDaCyeJZLA0yPOjzzNYGlzEEs7dJzd/ksZIY02OXdcBQwpZnSWlaxr6MokXnvB4vP9xBstz/wN1HZejY0c5cPAAETtC2StTDso0R5tnNCtEURZazs0xUh6ZdFvRK7JnbA8Hsgfwhb8gryOlxHVd7BH7lF2YRa/ImDO2IK93tqlld3FdBwwhjr9xuqYti2lxUkqeGXqG3SO7SUVSc5454gkPX/p4gYcWaMStuAoUi8wNXHzhI6TgUP4QL4y+QM7NLcprCyEoFoskygn0s2Dzr2PFYxwpHFncF/VOf7epmaxLrSNmnn4Lg6SdZGvTVtal1p1yhtfZpDHXWLNj13XAOLGlqev6Wd/CEFKwL7OPp0eepi3eRtJOzvlYju5QNItnnFqozJ8buOzP7Kcv3xdOoQ0c9o7t5WDu4NIv9iot7cufrDnaPKlL09RNNqQ2sKVpCwkrsSCv4XkeQ0NDtLW1YVnTT3gwdZPuVDcRo/4+G3peV+sw5uLEt0zXNIxFjhjD5eEzXlFqaOiaji98erO9HC0eJWEm5hUsasUNXHpzvZS8xTsLecJjf2Y/ezN7qfiVmT9RQiACjGNGzdc9lPzSGbtTmiJNnNt8LqsSi7PHSuAHjGXGaGpswjCX/qo4HUmzpWkLDXZDzV/LcRzihThdTStvSnGt1XXAOKFHCmORu6SyTpZfHP4Fo87oGV9XQ0NKiW3Y4ayeGl31DJeHeW7kOfaO7aXkz+6k7wmPvnzfgvUzL5qZLduYt8ZII5sbN2MbNrqm093QzdbmraTsFBBe0S4mx3HoE310rVEnTWXh1HXAqKZgkGCai9clFYiAp4efZswdY31q/ZKPnQyXh/lez/d4fODxeR+rMdJIc7R5AUo1MxoaaxvWcm7zubOa+eF5Hv39/axateqU3RILJWJEaI21LvnvWVFqrb4DxnhfhI6GqWs1n1Y7XB7GFz6jlVH2Z/azOrF6TieRvJtn19CuGa9qno7v+/Rn+/le5nscyB+fOri5cTPnNp9La6x11sdck1xDR7xjWZwYHcchsAM6E53qCltRFkhdBwwxHjA0wDT0mgaMglvg54d/Ts7NIaQgZafm1LU0Vhnj7x//+wWf8ndu07m8bvPr6GroWtDjKkq9k1KSc3OMVcYWJQ/YfG2RW2q2VqquA8YEXWpYhkYtL4zHnDEyboauZBeaps3pF1b0itz51J2MOWM0R5vpbuiec3mEEDhlh+2rtnN++/m0x9WUWuXsJKRgtDK6YJMpPM9jwB1AK2hYzqm7IyUSKeUZW8xCChrsBna07yAdSS9IGWtp7MBYzXoB6jpgHG9hhAGjlrOkRiujAHOap/3U0FP8pv837BnbQ9kv0xhp5C8v/kuaY3MfK3Ach76+PrpWq0FPZelJKaubFZ3c1Vr2wwWl57WctyAnuopTIZ6Ls655HdHIqVO7m7qJqZszSnOzKrFqWQQLgLyeVwFjLiYmSWkSTL12XVJSSo4VjhE347N6XiAC/u++/8tPD/+0eltTpIl37XjXvIKFsjSKXhEnCDfsEVJUvyZ+doULkkUZA3JdlwFnAJmX2M7SZ2yVSGJGjHQkPSWDbEe8gw3pDcSt2X1+TqVUKhEZirCtdRvx+MIcUwnVd8DQjg96WzWcJZX38uScHAn7+MKjp4ae4uG+h/HlqaehFr1iNT/NK7pewYs7Xkx3w9mfqykQizRXdZxAEIhgVpMAXM+lElQoeSV8vbZTgSe6VCJGpHrSMzQD27CxdKv6c3O0mZgVq9n2mSeqVCrsL+9nU+cmotGl30BJ13TSkTQpO7UsJk0o06vvgDH+XZPhoHet/lCzlSwlv0RrPJx59PTw03zlma/MKBusbdi8adub2NG+oyZlO1HBLZB1s/NaSyGRi5qpFsLdzgzdmNXrur5LRVTCRXU1DhiaprGpcdOcZ5/VQskqUYwU6W7oVlfZyoKp74ChHZ8lZRm1u6obqYzQX+qnJ9uDJzx+0PsDhBS8uOPFXNx+8Wmfuy61rmaZJSc4gcNgaRBTN1mfWk9jpBHbsGedp0rX9OpV82JeJeqajqVb1av1mSiVS7xQfoGt67YSj9X+hBkzY+rKWal7KyNgSK1mAUNKyaHcIb6x5xu4gVu9/YLWC3jTtjfNOVlZIILTdmediRM4uMKlv9SPZoQ7r21v2U5HomPOx1xWPIgaUeJmfMH6xhVlpavvgHHiOowa7bZ3tHiUxwYeww1cklaSbc3baI218up1r55zsCh5JfqL/WfMonk6ruuSD/Ksiaxhx+odrE2uXRaZNhVFOXvVd8AY7yHQJFjGwp0spZQUvSK9uV6eGX6G50aeA+Cyzst43abXzfu4w5VhLmy7kM2Nm+d8rHK5zJ7KHnas2UE6uTymAyqKcnar84BxvIVhWwtT1WOFYzwz8gzZSpa8mydhJ+jJ9gCccbziVNzA5WjhKJqmYekWF7ZeyIVtF84rYZ0tbJJmclb9/oqiKKdT3wFj/LsmwbLm38IYKY/w6LFHyXt5miJNNMeaeWLwCTzh0RZrY21y7ZyOO1AaYF1qHVuatxAxIlP2DFAURTkb1HfAmOiSAux5BAwpJXkvz2/6f0POzdHV0FWdEfPE4BNA2LqYa6JBS7c4r/U8lb5DUZSzWl0HDE4Yw4jMMsV12S8zUh5huDxMf7Gfglcg7+TpTnWjaRq+8Pn5kZ/z7MizwPTdUWW/PGnm1HRGK6PsaNuhgoWiKGe9ug4Y1VxSEmL2zFoYBbfAwdxB9oztIe/mw5QGZoyYGaMp1cTzo8/zzMgzPDP8DBknA8DWpq2sSa45/rpSMFAcQEhxximda5Nr2dq8dW4VVBRFWUR1HTDkCT1EUev0YwJSSg7kDvDU4FNknAxpO82a5JrqVFQpJd/e921+2vfT6nNSdoprNlzDZasvI5ABR/NHkUiEFLTGWtnRtoNVyVWnXSCna7oar1AUZVmo74Ax/l2TELdPXVUv8Hhq6Cl2j+4mYkRYl1o35ST+w4M/rAaLyzsv57yW89javLWaSK2/2E97vJ3NjZvRNI2OeMdZuS+3oijKXNV3wDhhEDp2ikFvL/B4bOAxnh99nvZ4OwnreALB0coovz72a54ffZ792f0AXH/O9byi6xWTjuELn4pfYcvqLWxMb1z4iiiKopwF6jxgHN/TOzJNwPDE8WDRmegkYob7RpS8Ej88+EMePvzwpER916y/ZkqwgHBr1o54B11JtZudoij1qyYB4/vf/z7/9m//xrPPPksul2PdunW86U1v4vrrr5809fQb3/gG9957L0ePHmXDhg3ccsstvPKVr1ywcsgTxg6mG8MYKg2xd2zvpGAhpeRLu77EgewBADalN3HJqkumzUQqpaTkl3ACh63NW7EMtUhOUZT6VZOA8Y//+I+sWbOGD33oQzQ1NfHLX/6Sj3zkI/T393PzzTcD8MADD/CRj3yEd77znVx66aU8+OCD3Hzzzdx///3s2LFjQcoxqYVhTm1hVPwKAlENFhDuY3EgewDbsHnLeW+ZsgvYRPqOvJen4ldImAk2N25We2UrilL3ahIw7rrrLpqbj+8Yd9lll5HJZPiHf/gH/vzP/xxd1/nCF77Aa17zGt773vcCcOmll7Jnzx6++MUvcs899yxIOSbNkpom+WAlqBwfGSecDvtAzwMAvLLrlZzfev6kxwcioK/QR9yM0xHrYE3DGjriHaQjaZXaWlGUuleT+ZwnBosJ27Zto1AoUCqV6Ovro7e3l2uuuWbSY6699loeeeQRXPf0i91mSp7wP3uaFkbJK03K4Prr/l/TX+onbsZ5VferJj02EAF9+T5WxVfxqu5XcfW6q9navJXGaKMKFoqirAiLNuj9+OOP09HRQTKZ5PHHHwdgw4YNkx6zadMmPM+jr6+PTZs2zel1pJSUSiXK5XK1hSElELiUSqVJjx0pjCB9SaFU4D+P/Sc/PvxjAF655pXogY4TOJT9MhkngytcVsdXs6NpBzEZm3Kss025XJ70faVR9V+59V/JdYfwHFiri9hFCRiPPfYYDz74IB/84AcByGazAKRSqUmPm/h54v658DyP3bt3AyDG2xgSONTbgzt8vLpSSvZm9zLqjXL3yN1k/fA1V9mrWBeso6+vj6yfxREO7XY7q+xVtPgtHCkc4QhH5ly+xdbb27vURVhSqv69S12EJbOS627bdk2OW/OA0d/fzy233MLOnTu56aabav1yWJbF5s2bKZfLPKX9HAhnS52zeTNrm45vSOQGLrt6dvGt579F1s+SttNcu+5aLm67GF3TOVY8xip9FRe1XcSGhg3LrtupXC7T29vL+vXricXmvhHTcqXqv3Lrv5LrDrB3796aHbumASOXy/G2t72NxsZGbr/9dnQ9HDJJp8MNffL5PG1tbZMef+L9c6FpWnXT+4kxDIkkHo9VbweoVCr8y95/YbgyTGOkkVtefAvN0XDsxQ1cLMviyq4rl/3sp1hscr1XGlX/lVv/lVr3Wl7c1iyJUaVS4R3veAf5fJ57772XhoaG6n0bN4aroXt6eiY9p6enB8uy6OpamJO00I53SeknvYmP9T/G0eJRokaUd+94dzVYAOTcHKlIitWJ1QtSDkVRlHpQk4Dh+z7vfe976enp4d5776Wjo2PS/V1dXaxfv56HHnpo0u0PPvggl1122YL1v1VbGHJqwOjN9QKwsXEjqxKrJt1X9Ip0NXTNa8c7RVGUelOTM+LHPvYxfvKTn/ChD32IQqHAk08+Wb1v+/bt2LbNe97zHt73vvfR3d3Nzp07efDBB9m1axdf/epXF6wcEwv3BBr6SaHxcP4wAG2xtkm3ByJAQ5uyqltRFGWlq0nA+MUvfgHApz/96Sn3/ehHP2Lt2rVcd911lMtl7rnnHu6++242bNjAHXfcwcUXz21f7OlMzJIK0Ka0MI4UwplOJweGglcgaSVVwFAURTlJTQLGj3/84xk97oYbbuCGG26oRRGQUh6fVqtN7ZI6VjwGTA0YOTfH5sbNRM1oTcqlKIqyXNXvzj1BUB3DCDQN/YR4IaVksDQITA4YZb9MIAI6EpPHXBRFUZQ6Tm8ufb86S0qgT5pqNloZxQkcNDRaoi1U/ArHiseImlG6U92siq861WEVRVFWrLoNGIHnISfGMPTJLYyebDidN2WnQIOjhaNsa9rGluYtNEWb1JapiqIo06jbgCEdtzqG4WsaxgkRozfbC0BztJnD+cNsSG3gRateRMSITHcoRVEUhToewwgqLgIBgNSMSYPeB3MHAUhH0jRHmrlk1SUqWCiKopxB3QYM6boE42MYvm5w4iSpiTUYqUiKlngLDXbDdIdQFEVRTlC3ASNwT2xhTF6HcaQYrsFosBpoijYtSfkURVGWm/oNGJXjg95Cn9wlNbEGozHSSMJMLEn5FEVRlps6DhjHB7013ajOkip5JbJOuPdFa6yVpJ1cqiIqiqIsKysiYAjNqK7DmEg6GDEiNEYbiZsrL/2xoijKXNRxwHAIxscwNOP4vt0TazAaI43EzBgxc+VtsKIoijIXdRswvEoFxruhNO34cpMXRl8AwjUYzdHmZbeTnqIoylKp24ARVNzq/3VzasBoj7fTGGlc7GIpiqIsW3UbMHz3eMCwjOMBY19mHwAd8Q4SlpohpSiKMlN1HDC86v9NI6xm1skyVB4CoDPZqQKGoijKLNRvwPCC6v8j4wHjuZHnAEjb6XANhgoYiqIoM1a3ASPwfQB0qRGxwllSTw8/DYTjFwkroTZJUhRFmYX6DRhB2MLQ0YiOB4yJFkZbrI32ePuSlU1RFGU5qt+A4Y+vweB4C2NihtSq5CpaYi1LVjZFUZTlqG4DhgjCgKFLjZhlUPJKHCmESQe7k91qSq2iKMos1W3ACIKJFgZELIPnRp5DIombcVYnV6uU5oqiKLNUtwFDyOMtjIip82j/o0A44N2Z7FTbsCqKosxS3Z41hRjPVIuGaen85NBPAFibXEtztHkpi6YoirIs1W3AGO+RQgdG6OeFsXDA++L2i2mMNi5ZuRRFUZarug0Y4w0MNKnx6/IvAViXWkd3qpsGS41fKIqizFYdB4yJLil40n0SgHObz6Uz2aky1CqKosxB/QcMqTEajGLqJuc2n6sW7CmKosxR/QaM8e8TFdyU3kRHvEMt2FMURZmj+g0Y1TGM8PuqxCq6GrqwdGvpCqUoirKM1W3AkNX/heMVbbE2OuIdS1UcRVGUZa9uA8ZEl9REC6Mj3kFrrHXJyqMoirLc1W3AmGhhTMyH2tK0BctQ3VGKoihzVf8BQ4arvbvSXUtaHkVRlOWufgPGeNNCA+JGnLgZX9LyKIqiLHd1HDDGI4aEhJUgYkSWtkCKoijLXP0GjIlptUDSTKqAoSiKMk/1GzAmRrulpMFOETXU/t2KoijzUbcBQ5wwhpGyU2qGlKIoyjzVbcCQJ/wvHUkvYUkURVHqQ/0GjBO6pJoiTUtaFkVRlHpQvwHjhP+0J1RKEEVRlPmq34BRnVUr6WzoXNrCKIqi1IG6Dxgg6G7oXsqiKIqi1IX6DRgTnVISmtQe3oqiKPNWtwHj+LRaSdRSi/YURVHma0kDxv79+3nLW97Cjh07uPzyy/nMZz6D67oLcuzjg96SqKkW7SmKosyXuVQvnM1mefOb38z69eu5/fbbGRgY4NOf/jSVSoVbb7113sc/PoYhsXV73sdTFEVZ6ZYsYHz961+nWCxyxx130NjYCEAQBHzsYx/jHe94Bx0d85sKOzGGoUnQNO0Mj1YURVHOZMm6pH72s59x2WWXVYMFwDXXXIMQgl/84hfzPv5EC0OX8vQPVBRFUWZkyVoYPT09XH/99ZNuS6VStLW10dPTM6djep6HlJJdu3bxot+9jACJKWDXrl0LUeRlRY4Hyr17967IFpaq/8qt/0quO4TnwVrVe8kCRi6XI5VKTbk9nU6TzWbndMyJN0nTNFJNjRRGM6CtzC4pTdOw7ZU7dqPqv3Lrv5LrDmH96y5g1MLFF1+81EVQFEWpW0s2hpFKpcjn81Nuz2azpNMqu6yiKMrZZskCxsaNG6eMVeTzeYaGhti4ceMSlUpRFEU5lSULGFdeeSW//OUvyeVy1dseeughdF3n8ssvX6piKYqiKKegSbk0806z2Syvec1r2LBhA+94xzuqC/de+9rXLsjCPUVRFGVhLVnAgDA1yCc+8QmeeOIJEokEr3/967nllltW9AwHRVGUs9WSBgxFURRl+ajbbLWKoijKwlIBQ1EURZkRFTAURVGUGVEBQ1EURZkRFTAURVGUGVEBQ1EURZkRFTAURVGUGVm2AWOu+4HXch/xxTTbenz/+9/nXe96F1deeSU7duzg9a9/Pd/85jdZjstw5vs7LBaLXHnllWzdupWnn366hiWtjfnU/zvf+Q5/8Ad/wAUXXMDOnTv5b//tv1GpVGpc4oU11/r/6Ec/4oYbbuDiiy/miiuu4C//8i/p6+tbhBIvjIMHD3Lrrbfy+te/nu3bt3PdddfN+LkLdd5blunN57ofeK33EV8sc6nHP/7jP7JmzRo+9KEP0dTUxC9/+Us+8pGP0N/fz80337zINZi7hfgd3nnnnQRBUOOS1sZ86n/XXXdxzz338M53vpMdO3YwNjbGI488sqzei7nW/9FHH+Xmm2/mD/7gD7jlllvIZDJ8/vOf561vfSv//u//TjQaXcRazM3evXt5+OGHueiiixBCzPhib0HPe3IZ+tKXviR37Nghx8bGqrd9/etfl9u2bZP9/f0L/ryzzVzqMTIyMuW2v/7rv5YvetGLZBAEtSrqgpvv73Dfvn1yx44d8mtf+5rcsmWL3LVrVw1Lu/DmWv/9+/fL7du3y5/+9KeLUMramWv9P/KRj8irr75aCiGqtz3yyCNyy5Yt8je/+U0ti7xgTvycfvCDH5Svec1rZvS8hTzvLcsuqbnuB17rfcQXy1zq0dzcPOW2bdu2USgUKJVKtSrqgpvv7/C2227jxhtvZMOGDTUsZe3Mtf7f/va3Wbt2LVddddUilLJ25lp/3/dJJBKTdqJraGgAWDbdsro+t9P1Qp73lmXA6OnpmbJnxkz2A5/r8842C1WPxx9/nI6ODpLJ5EIXsWbmU/eHHnqIPXv28O53v7uWRaypudb/qaeeYsuWLdx5551cdtllnH/++dx444089dRTtS7ygppr/f/wD/+Q/fv3c//995PP5+nr6+Pv/u7v2L59Oy960YtqXewltZDnvWUZMOa6H3gt9hFfCgtRj8cee4wHH3yQt771rQtdvJqaa93L5TKf/vSnueWWW5ZVgDzZXOs/NDTEz3/+c7773e/y0Y9+lC9+8YtomsZb3/pWRkZGalnkBTXX+l9yySXccccdfPazn+WSSy7h1a9+NSMjI9xzzz0YhlHLIi+5hTzvLcuAocxPf38/t9xyCzt37uSmm25a6uIsirvuuouWlhauv/76pS7KkpBSUiqV+PznP8/v//7vc9VVV3HXXXchpeSrX/3qUhev5n7729/ygQ98gDe+8Y380z/9E5///OcRQvD2t7992c0SW0rLcpbUXPcDr5d9xOdTj1wux9ve9jYaGxu5/fbb59wvulTmUvcjR47wla98hS9+8YvV506M25RKJYrFIolEonaFXkDz+dtvbGzk3HPPrd7W2NjI9u3b2bdvX03KWgtzrf9tt93GpZdeyoc+9KHqbTt27OAVr3gF3/3ud/mjP/qjmpT3bLCQ571lGTDmuh94vewjPtd6VCoV3vGOd5DP5/mXf/mX6qDfcjKXuh8+fBjP83j7298+5b6bbrqJiy66iH/913+tSXkX2lx/95s3b+bQoUPT3uc4zoKWsZbmWv/9+/fzqle9atJtq1atoqmp6ZTvS71YyPPe8rq8HDfX/cDrZR/xudTD933e+9730tPTw7333ktHR8diFXdBzaXu27Zt47777pv09Vd/9VcAfOxjH+OjH/3oopR9Icz1b/iVr3wlmUyG3bt3V28bGxvj2Wef5bzzzqtpmRfSXOvf2dnJc889N+m2I0eOMDY2xpo1a2pW3rPBgp73ZjUJ9yyRyWTk5ZdfLv/0T/9U/ud//qf85je/KS+55BL5sY99rPqY73znO3Lbtm3y0UcfndXzloO51P+v//qv5ZYtW+RXvvIV+cQTT0z6chxnqaoya3P93Z/sV7/61bJchzHX+gdBIK+//nr56le/Wj7wwAPyP/7jP+Qb3/hG+dKXvlQODg4uRVXmZK71/8d//Ee5ZcsW+YlPfEL+4he/kA888IC87rrr5Mte9jI5Ojq6FFWZtVKpJL///e/L73//+/JP//RP5VVXXVX9eWKdVa3Pe8uySyqdTvNP//RPfOITn+Dd7343iUSCN7zhDdxyyy3VxwghCIJg0hzrmTxvOZhL/SfmW3/605+ecrwf/ehHrF27dnEKP09z/d3Xi7nWX9d17r77bj71qU9x66234nkel1xyCffffz9tbW1LUZU5mWv9b7rpJmzb5mtf+xrf+ta3SCQS7Nixg8997nM0NTUtRVVmbWRkhL/8y7+cdNvEz/fddx87d+6s+XlP7emtKIqizMiyHMNQFEVRFp8KGIqiKMqMqIChKIqizIgKGIqiKMqMqIChKIqizIgKGIqiKMqMqIChKIqizMiyXLinKIqihPnhbrvtNn7zm99gmia/+7u/O2Vx30JSAUNRFGWZ+tu//VtaW1v5wQ9+AMDw8HBNX0+t9FYURVkkBw8e5Mtf/jJPPfUUe/fuZePGjXzve9+b9rH79+/ntttu44knniCRSPD617+e9773vdi2DUCxWOR3fud3+OlPf1q9rdZUC0NRFGWR7N27l4cffpiLLroIIcQp851ls1ne/OY3s379em6//XYGBgb49Kc/TaVS4dZbbwWgr6+PlpYW/vZv/5YnnniCpqYm3v/+90/a82ShqUFvRVGUedi7d+8p99T49a9/PSmt+NVXX83DDz/MF77whdOmlf/6179OsVjkjjvu4OUvfzlveMMbeP/738/Xv/51BgYGgHDLgj179nDZZZfx7W9/m7e85S01369eBQxFUZR5uPPOO/mzP/szjh07Nun2X//617ztbW/j29/+dvW2me5w+bOf/YzLLruMxsbG6m3XXHMNQohq5unOzk5isRivfvWrAbjiiisolUqMjo7Os0anpgKGoijKPHziE5+gqamJP/uzP2NkZASAXbt28c53vpNrr72WN7/5zbM+Zk9Pz5Td8FKpFG1tbdXd85qbm7nooot47LHHAHj66aexbbum6drVGIaiKMo8JJNJ7r33Xt70pjfx1re+lb/+67/m5ptv5oorruC2225D07RZHzOXy5FKpabcnk6nyWaz1Z8/9rGP8eEPf5h8Pk8sFuPzn//8nF5vplTAUBRFmaempia+8pWv8Cd/8if86Z/+KS9/+cv57Gc/i2EYNX3d9evX88///M81fY0TqS4pRVGUBeC6Lo7joOs6lUqFIAjmfKxUKkU+n59yezabJZ1Oz6eY86IChqIoyjwNDg7ylre8hc7OTr71rW/R29vLzTffjOu6czrexo0bq2MVE/L5PENDQ1PGNhaTChiKoijzMDY2xlvf+laSySR3330327dv5x/+4R94+umned/73jenlsaVV17JL3/5y0lTch966CF0Xefyyy9fyOLPigoYiqIo8/DRj34UIQRf/vKXqwPV55xzDl/+8pf55S9/yX333Vd9bLlc5qGHHuKhhx7iyJEjFAqF6s8nToe98cYbSSQSvPvd7+bnP/853/rWt/jMZz7DjTfeSEdHx6LXcYJKDaIoijIPfX192LY97Yn8ueeeY8OGDcRiMQAOHz7Mq171qmmPc99997Fz587qz/v37+cTn/jEpNQgt9xyy6KlAZmOChiKoijKjKguKUVRFGVGVMBQFEVRZkQFDEVRFGVGVMBQFEVRZkQFDEVRFGVGVMBQFEVRZkQFDEVRFGVGVMBQFEVRZkQFDEVRFGVGVMBQFEVRZkQFDEVRFGVGVMBQFEVRZuT/BxQZ3pX123QaAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Hard\n",
    "Hard_fig = plt.figure(figsize=(4, 4))\n",
    "Hard_ax = Hard_fig.add_subplot(111)\n",
    "Hard_ax.set_title('Hard', fontsize=14)\n",
    "# Hard_ax.set_xlabel(r'$\\times 10^3$')\n",
    "# Hard_ax.set_ylabel('Coverage')\n",
    "# seaborn style\n",
    "sns.set(style=\"whitegrid\")\n",
    "# plot\n",
    "# lds\n",
    "plt.plot(Hard_lds_x_axis, Hard_lds_coverage_mean, label='LEADSSS', color=colors['LEADSSS'])\n",
    "plt.fill_between(Hard_lds_x_axis, Hard_lds_coverage_mean - Hard_lds_coverage_std, Hard_lds_coverage_mean + Hard_lds_coverage_std, alpha=0.3, color=colors['LEADSSS'])\n",
    "# diayn\n",
    "plt.plot(Hard_diayn_x_axis, Hard_diayn_coverage_mean, label='DIAYN', color=colors['DIAYN'])\n",
    "plt.fill_between(Hard_diayn_x_axis, Hard_diayn_coverage_mean - Hard_diayn_coverage_std, Hard_diayn_coverage_mean + Hard_diayn_coverage_std, alpha=0.3, color=colors['DIAYN'])\n",
    "# rnd\n",
    "plt.plot(Hard_rnd_x_axis, Hard_rnd_coverage_mean, label='RND', color=colors['RND'])\n",
    "plt.fill_between(Hard_rnd_x_axis, Hard_rnd_coverage_mean - Hard_rnd_coverage_std, Hard_rnd_coverage_mean + Hard_rnd_coverage_std, alpha=0.3, color=colors['RND'])\n",
    "# ngu\n",
    "plt.plot(Hard_ngu_x_axis, Hard_ngu_coverage_mean, label='NGU', color=colors['NGU'])\n",
    "plt.fill_between(Hard_ngu_x_axis, Hard_ngu_coverage_mean - Hard_ngu_coverage_std, Hard_ngu_coverage_mean + Hard_ngu_coverage_std, alpha=0.3, color=colors['NGU'])\n",
    "# lsd\n",
    "plt.plot(Hard_lsd_x_axis, Hard_lsd_coverage_mean, label='LSD', color=colors['LSD'])\n",
    "plt.fill_between(Hard_lsd_x_axis, Hard_lsd_coverage_mean - Hard_lsd_coverage_std, Hard_lsd_coverage_mean + Hard_lsd_coverage_std, alpha=0.3, color=colors['LSD'])\n",
    "# CSD\n",
    "plt.plot(Hard_csd_x_axis, Hard_csd_coverage_mean, label='CSD', color=colors['CSD'])\n",
    "plt.fill_between(Hard_csd_x_axis, Hard_csd_coverage_mean - Hard_csd_coverage_std, Hard_csd_coverage_mean + Hard_csd_coverage_std, alpha=0.3, color=colors['CSD'])\n",
    "# METRA\n",
    "plt.plot(Hard_metra_x_axis, Hard_metra_coverage_mean, label='METRA', color=colors['METRA'])\n",
    "plt.fill_between(Hard_metra_x_axis, Hard_metra_coverage_mean - Hard_metra_coverage_std, Hard_metra_coverage_mean + Hard_metra_coverage_std, alpha=0.3, color=colors['METRA'])\n",
    "# SMM\n",
    "plt.plot(Hard_smm_x_axis, Hard_smm_coverage_mean, label='SMM', color=colors['SMM'])\n",
    "plt.fill_between(Hard_smm_x_axis, Hard_smm_coverage_mean - Hard_smm_coverage_std, Hard_smm_coverage_mean + Hard_smm_coverage_std, alpha=0.3, color=colors['SMM'])\n",
    "# CIC\n",
    "\n",
    "\n",
    "# plt.legend()\n",
    "plt.tight_layout()\n",
    "formatter = ticker.ScalarFormatter(useMathText=True)\n",
    "formatter.set_scientific(True)\n",
    "formatter.set_powerlimits((-1,1)) \n",
    "Hard_ax.xaxis.set_major_formatter(formatter)\n",
    "# limit y \n",
    "plt.xlim(-10000, 10**6)\n",
    "plt.ylim(0, 101)\n",
    "# plt.xlim(0, 500)\n",
    "plt.savefig('figures/Hard.pdf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# fetch reach\n",
    "fetch_fig = plt.figure(figsize=(4, 4))\n",
    "fetch_ax = fetch_fig.add_subplot(111)\n",
    "fetch_ax.set_title('Fetch Reach', fontsize=14)\n",
    "# fetch_ax.set_xlabel(r'$\\times 10^3$')\n",
    "# fetch_ax.set_ylabel('Coverage')\n",
    "# seaborn style\n",
    "sns.set(style=\"whitegrid\")\n",
    "# plot\n",
    "# lds\n",
    "std_frac= 1.0\n",
    "mean_mult = 1.0\n",
    "\n",
    "plt.plot(fetch_lds_x_axis, fetch_lds_coverage_mean*mean_mult, label='LEADSSS', color=colors['LEADSSS'])\n",
    "plt.fill_between(fetch_lds_x_axis, fetch_lds_coverage_mean*mean_mult - fetch_lds_coverage_std/std_frac, fetch_lds_coverage_mean*mean_mult + fetch_lds_coverage_std/std_frac, alpha=0.3, color=colors['LEADSSS'])\n",
    "# diayn\n",
    "plt.plot(fetch_diayn_x_axis, fetch_diayn_coverage_mean*mean_mult, label='DIAYN', color=colors['DIAYN'])\n",
    "plt.fill_between(fetch_diayn_x_axis, fetch_diayn_coverage_mean*mean_mult - fetch_diayn_coverage_std/std_frac, fetch_diayn_coverage_mean*mean_mult + fetch_diayn_coverage_std/std_frac, alpha=0.3, color=colors['DIAYN'])\n",
    "# rnd\n",
    "plt.plot(fetch_rnd_x_axis, fetch_rnd_coverage_mean*mean_mult, label='RND', color=colors['RND'])\n",
    "plt.fill_between(fetch_rnd_x_axis, fetch_rnd_coverage_mean*mean_mult - fetch_rnd_coverage_std/std_frac, fetch_rnd_coverage_mean*mean_mult + fetch_rnd_coverage_std/std_frac, alpha=0.3, color=colors['RND'])\n",
    "# ngu\n",
    "plt.plot(fetch_ngu_x_axis, fetch_ngu_coverage_mean*mean_mult, label='NGU', color=colors['NGU'])\n",
    "plt.fill_between(fetch_ngu_x_axis, fetch_ngu_coverage_mean*mean_mult - fetch_ngu_coverage_std/std_frac, fetch_ngu_coverage_mean*mean_mult + fetch_ngu_coverage_std/std_frac, alpha=0.3, color=colors['NGU'])\n",
    "# lsd\n",
    "plt.plot(fetch_lsd_x_axis, fetch_lsd_coverage_mean*mean_mult, label='LSD', color=colors['LSD'])\n",
    "plt.fill_between(fetch_lsd_x_axis, fetch_lsd_coverage_mean*mean_mult - fetch_lsd_coverage_std/std_frac, fetch_lsd_coverage_mean*mean_mult + fetch_lsd_coverage_std/std_frac, alpha=0.3, color=colors['LSD'])\n",
    "# csd\n",
    "plt.plot(fetch_csd_x_axis, fetch_csd_coverage_mean*mean_mult, label='CSD', color=colors['CSD'])\n",
    "plt.fill_between(fetch_csd_x_axis, fetch_csd_coverage_mean*mean_mult - fetch_csd_coverage_std/std_frac, fetch_csd_coverage_mean*mean_mult + fetch_csd_coverage_std/std_frac, alpha=0.3, color=colors['CSD'])\n",
    "# metra\n",
    "plt.plot(fetch_metra_x_axis, fetch_metra_coverage_mean*mean_mult, label='METRA', color=colors['METRA'])\n",
    "plt.fill_between(fetch_metra_x_axis, fetch_metra_coverage_mean*mean_mult - fetch_metra_coverage_std/std_frac, fetch_metra_coverage_mean*mean_mult + fetch_metra_coverage_std/std_frac, alpha=0.3, color=colors['METRA'])\n",
    "# smm\n",
    "plt.plot(fetch_smm_x_axis, fetch_smm_coverage_mean*mean_mult, label='SMM', color=colors['SMM'])\n",
    "plt.fill_between(fetch_smm_x_axis, fetch_smm_coverage_mean*mean_mult - fetch_smm_coverage_std/std_frac, fetch_smm_coverage_mean*mean_mult + fetch_smm_coverage_std/std_frac, alpha=0.3, color=colors['SMM'])\n",
    "\n",
    "# plt.legend()\n",
    "plt.tight_layout()\n",
    "formatter = ticker.ScalarFormatter(useMathText=True)\n",
    "formatter.set_scientific(True)\n",
    "formatter.set_powerlimits((-1,1)) \n",
    "fetch_ax.xaxis.set_major_formatter(formatter)\n",
    "# limit y\n",
    "plt.ylim(0, 101)\n",
    "plt.xlim(-10, 2*10**5)\n",
    "plt.savefig('figures/fetch_reach.pdf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 251,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# finger\n",
    "finger_fig = plt.figure(figsize=(4, 4))\n",
    "finger_ax = finger_fig.add_subplot(111)\n",
    "finger_ax.set_title('Finger', fontsize=14)\n",
    "# finger_ax.set_xlabel(r'$\\times 10^3$')\n",
    "# finger_ax.set_ylabel('Coverage')\n",
    "# seaborn style\n",
    "sns.set(style=\"whitegrid\")\n",
    "std_frac= 1.0\n",
    "mean_mult = 1.0\n",
    "# plot\n",
    "# lds\n",
    "plt.plot(finger_lds_x_axis, finger_lds_coverage_mean, label='LEADSSS', color=colors['LEADSSS'])\n",
    "plt.fill_between(finger_lds_x_axis, finger_lds_coverage_mean - finger_lds_coverage_std*std_frac, finger_lds_coverage_mean + finger_lds_coverage_std*std_frac, alpha=0.3, color=colors['LEADSSS'])\n",
    "# diayn\n",
    "plt.plot(finger_diayn_x_axis, finger_diayn_coverage_mean, label='DIAYN', color=colors['DIAYN'])\n",
    "plt.fill_between(finger_diayn_x_axis, finger_diayn_coverage_mean - finger_diayn_coverage_std*std_frac, finger_diayn_coverage_mean + finger_diayn_coverage_std*std_frac, alpha=0.3, color=colors['DIAYN'])\n",
    "# rnd\n",
    "plt.plot(finger_rnd_x_axis, finger_rnd_coverage_mean, label='RND', color=colors['RND'])\n",
    "plt.fill_between(finger_rnd_x_axis, finger_rnd_coverage_mean - finger_rnd_coverage_std*std_frac, finger_rnd_coverage_mean + finger_rnd_coverage_std*std_frac, alpha=0.3, color=colors['RND'])\n",
    "# ngu\n",
    "plt.plot(finger_ngu_x_axis, finger_ngu_coverage_mean, label='NGU', color=colors['NGU'])\n",
    "plt.fill_between(finger_ngu_x_axis, finger_ngu_coverage_mean - finger_ngu_coverage_std*std_frac, finger_ngu_coverage_mean + finger_ngu_coverage_std*std_frac, alpha=0.3, color=colors['NGU'])\n",
    "# lsd\n",
    "plt.plot(finger_lsd_x_axis, finger_lsd_coverage_mean, label='LSD', color=colors['LSD'])\n",
    "plt.fill_between(finger_lsd_x_axis, finger_lsd_coverage_mean - finger_lsd_coverage_std*std_frac, finger_lsd_coverage_mean + finger_lsd_coverage_std*std_frac, alpha=0.3, color=colors['LSD'])\n",
    "# csd\n",
    "plt.plot(finger_csd_x_axis, finger_csd_coverage_mean, label='CSD', color=colors['CSD'])\n",
    "plt.fill_between(finger_csd_x_axis, finger_csd_coverage_mean - finger_csd_coverage_std*std_frac, finger_csd_coverage_mean + finger_csd_coverage_std*std_frac, alpha=0.3, color=colors['CSD'])\n",
    "# metra\n",
    "plt.plot(finger_metra_x_axis, finger_metra_coverage_mean, label='METRA', color=colors['METRA'])\n",
    "plt.fill_between(finger_metra_x_axis, finger_metra_coverage_mean - finger_metra_coverage_std*std_frac, finger_metra_coverage_mean + finger_metra_coverage_std*std_frac, alpha=0.3, color=colors['METRA'])\n",
    "# smm\n",
    "plt.plot(finger_smm_x_axis, finger_smm_coverage_mean, label='SMM', color=colors['SMM'])\n",
    "plt.fill_between(finger_smm_x_axis, finger_smm_coverage_mean - finger_smm_coverage_std*std_frac, finger_smm_coverage_mean + finger_smm_coverage_std*std_frac, alpha=0.3, color=colors['SMM'])\n",
    "# cic\n",
    "# plt.legend()\n",
    "plt.tight_layout()\n",
    "formatter = ticker.ScalarFormatter(useMathText=True)\n",
    "formatter.set_scientific(True)\n",
    "formatter.set_powerlimits((-1,1)) \n",
    "finger_ax.xaxis.set_major_formatter(formatter)\n",
    "# limit y\n",
    "plt.ylim(0, 101)\n",
    "plt.xlim(-10, 10**5)\n",
    "plt.savefig('figures/finger.pdf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 527,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# fetch push\n",
    "fetch_push_fig = plt.figure(figsize=(4, 4))\n",
    "fetch_push_ax = fetch_push_fig.add_subplot(111)\n",
    "fetch_push_ax.set_title('Fetch Slide', fontsize=14)\n",
    "# fetch_push_ax.set_xlabel(r'$\\times 10^3$')\n",
    "# fetch_push_ax.set_ylabel('Coverage')\n",
    "# seaborn style\n",
    "sns.set(style=\"whitegrid\")\n",
    "# plot\n",
    "std_frac= 1.0\n",
    "mean_mult = 1.0\n",
    "# lds\n",
    "plt.plot(fetch_push_lds_x_axis, fetch_push_lds_coverage_mean*mean_mult, label='LEADSSS', color=colors['LEADSSS'])\n",
    "plt.fill_between(fetch_push_lds_x_axis, fetch_push_lds_coverage_mean*mean_mult - fetch_push_lds_coverage_std*std_frac, fetch_push_lds_coverage_mean*mean_mult + fetch_push_lds_coverage_std*std_frac, alpha=0.3, color=colors['LEADSSS'])\n",
    "# diayn\n",
    "plt.plot(fetch_push_diayn_x_axis, fetch_push_diayn_coverage_mean*mean_mult, label='DIAYN', color=colors['DIAYN'])\n",
    "plt.fill_between(fetch_push_diayn_x_axis, fetch_push_diayn_coverage_mean*mean_mult - fetch_push_diayn_coverage_std*std_frac, fetch_push_diayn_coverage_mean*mean_mult + fetch_push_diayn_coverage_std*std_frac, alpha=0.3, color=colors['DIAYN'])\n",
    "# rnd\n",
    "plt.plot(fetch_push_rnd_x_axis, fetch_push_rnd_coverage_mean*mean_mult, label='RND', color=colors['RND'])\n",
    "plt.fill_between(fetch_push_rnd_x_axis, fetch_push_rnd_coverage_mean*mean_mult - fetch_push_rnd_coverage_std*std_frac, fetch_push_rnd_coverage_mean*mean_mult + fetch_push_rnd_coverage_std*std_frac, alpha=0.3, color=colors['RND'])\n",
    "# ngu\n",
    "plt.plot(fetch_push_ngu_x_axis, fetch_push_ngu_coverage_mean*mean_mult, label='NGU', color=colors['NGU'])\n",
    "plt.fill_between(fetch_push_ngu_x_axis, fetch_push_ngu_coverage_mean*mean_mult - fetch_push_ngu_coverage_std*std_frac, fetch_push_ngu_coverage_mean*mean_mult + fetch_push_ngu_coverage_std*std_frac, alpha=0.3, color=colors['NGU'])\n",
    "# lsd\n",
    "plt.plot(fetch_push_lsd_x_axis, fetch_push_lsd_coverage_mean*mean_mult, label='LSD', color=colors['LSD'])\n",
    "plt.fill_between(fetch_push_lsd_x_axis, fetch_push_lsd_coverage_mean*mean_mult - fetch_push_lsd_coverage_std*std_frac, fetch_push_lsd_coverage_mean*mean_mult + fetch_push_lsd_coverage_std*std_frac, alpha=0.3, color=colors['LSD'])\n",
    "# csd\n",
    "plt.plot(fetch_push_csd_x_axis, fetch_push_csd_coverage_mean*mean_mult, label='CSD', color=colors['CSD'])\n",
    "plt.fill_between(fetch_push_csd_x_axis, fetch_push_csd_coverage_mean*mean_mult - fetch_push_csd_coverage_std*std_frac, fetch_push_csd_coverage_mean*mean_mult + fetch_push_csd_coverage_std*std_frac, alpha=0.3, color=colors['CSD'])\n",
    "# metra\n",
    "plt.plot(fetch_push_metra_x_axis, fetch_push_metra_coverage_mean*mean_mult, label='METRA', color=colors['METRA'])\n",
    "plt.fill_between(fetch_push_metra_x_axis, fetch_push_metra_coverage_mean*mean_mult - fetch_push_metra_coverage_std*std_frac, fetch_push_metra_coverage_mean*mean_mult + fetch_push_metra_coverage_std*std_frac, alpha=0.3, color=colors['METRA'])\n",
    "# smm\n",
    "plt.plot(fetch_push_smm_x_axis, fetch_push_smm_coverage_mean*mean_mult, label='SMM', color=colors['SMM'])\n",
    "plt.fill_between(fetch_push_smm_x_axis, fetch_push_smm_coverage_mean*mean_mult - fetch_push_smm_coverage_std*std_frac, fetch_push_smm_coverage_mean*mean_mult + fetch_push_smm_coverage_std*std_frac, alpha=0.3, color=colors['SMM'])\n",
    "# cic\n",
    "# plt.legend()\n",
    "plt.tight_layout()\n",
    "formatter = ticker.ScalarFormatter(useMathText=True)\n",
    "formatter.set_scientific(True)\n",
    "formatter.set_powerlimits((-1,1)) \n",
    "fetch_push_ax.xaxis.set_major_formatter(formatter)\n",
    "# limit y\n",
    "plt.ylim(0, 101)\n",
    "plt.xlim(-10, 10**5)\n",
    "plt.savefig('figures/fetch_slide.pdf')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-27-d19293a1afa2>:18: UserWarning: Tight layout not applied. The bottom and top margins cannot be made large enough to accommodate all axes decorations.\n",
      "  plt.tight_layout()\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAFsAAACNCAYAAADYUQarAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAAsTAAALEwEAmpwYAAAQBklEQVR4nO2de1BV1fuHH0AMBUUUtVRU1MALiQdSvJAHjwKiXARHhRFDR8VrjJc/lCyd0q/YZDchUxzTNMHbASIvORKlKIOmQpmYpNIUangBNEA4yNm/Pxj2j5PcDnG27trPDDOctdZevHx4effe7177XWaCIAgoSIL5szbgv4QitoQoYkuIIraEKGJLiCK2hLRprLOqqoqCggIqKiqkskfWWFhY0KlTJ+zt7TE3f9qPzRq7zs7Pz6dDhw506dIFMzMzkxoqdwRBoKqqisLCQgRBoHfv3k+NaTSMVFRUKEI3EzMzM9q2bUvPnj0pKyurd0yTMVsR2jjqCx9in4R2/Odp9AT5PKLRaNi2bRtOTk5i26xZs7h9+zY2NjZi27p163BzcwOguroaLy8vXFxc+Oyzz8Qx586dIzIyEkdHR6qqqgDw9PRk8eLF2NraApCVlcUHH3yATqdDp9PRtWtXdu/e3agHN4TsxG6It956i3HjxtXbl5GRQbdu3bh06RL379/H3t5e7Ovfvz9JSUkAlJaWsmnTJmbPns3hw4cRBIGoqCj27NnDwIEDAcjNzW1xaDVKbO3FAg5e+KNFP6gppr/qwFT3XiaZW6vVEhoaSk5ODikpKcybN6/ecTY2Nqxbtw5vb28yMjJQqVSUl5cb/HEGDx7cYjv+NTF7w4YNBAUFiV8PHjwAoKioiKysLPz8/AgJCRG9uCEsLS0ZNGgQv/76K7a2tkyfPh0fHx8WLlxIfHw8d+7cabGNRnn2VPdeJvO+f0pDYSQ1NZVx48ZhY2ODu7s71dXVZGdno1KpGpyr7q3H2rVrmTNnDllZWZw+fZrt27ej1Wrp27ev0Tb+azy7IbRaLWfPnkWj0aDRaCgqKkKr1TY4vqqqil9++cXgBOzg4MC0adOIjY1FpVLx3XfftciWf80Jsj5++ukn/vrrL86cOSOe1AoLC5k8eTJr1qx5anxZWRnvvfcednZ2eHp6UlZWRnZ2NmPGjMHMzIxHjx5RUFBAr14t+++Wpdhz5szBwsJC/NypUyc2bNjAxx9/LLZFRUVx+vRpJk+ebHD10L17dwYPHsw333xDjx49uHHjBkFBQTx58gRBEPD09GT37t1YWFggCAL79u1j/fr1vPDCC1RXVxMQEIC3t3eL7G40N3L16lUGDRrUoon/yzSk278+Zj9PKGJLiCK2hChiS4gitoQoYkuI7MTWaDT4+/uj1+sN2vLy8gA4cuQIISEh+Pj4EBISQnh4uHjHd+7cOUJCQgzmy8vLQ6PRSGK7LG9qysvL+eqrrwgODjZoP3ToELt27SI2Npb+/fsDNWJmZmY+CzOfQnaeDbB06VLi4uLQ6XQG7bGxsURHR4tCAzg5OTF79myJLawfozy7JCWFh9rGU5QtxXZqCJ2mTGnWWBcXF4YMGUJiYiIREREAPHz4kMLCQlxdXU1iX2sgS88GWLZsGTt27GjwSTZAaGgo/v7+TJ06FXj2D6+N8uxOU6Y02/tMTb9+/VCr1ezatQsAW1tbunfvzuXLlxkzZgwA+/fvJy8vj4ULFwLQuXNnSkpKDOYpLi6mS5cuktgsW88GeOONN0hISBC9e8mSJcTExJCfny+Oefz4sfh9nz59gJpnklDzIPjgwYPiH8fUyPJqpJYXX3yRoKAgPv/8cwBmzJiBlZUVK1eupLS0lM6dO9OuXTtWr14N1DzyiouLIyYmhs2bN6PX6xk+fLjo+aZGSbGaACXF+hygiC0hitgSoogtIYrYEqKILSGK2BIiO7E1Gg0TJ04kMDAQPz8/Dh06BNTkqp2dnYmPjxfH/j1/7ezsTEBAAAEBAfj6+rJixQquX78ume2yvIPcsmULTk5O5OXlERISwtixYwHo2rUrX3zxBaGhoXTs2LHeY/fv34+1tTV6vZ4DBw4QFhZGUlISDg4OJrfbKLFTb6SS/GuySQwJfjmYwP6BRh3j5OREx44dKSwsBKBbt26oVCp27NjBypUrGz3W3NycsLAwzp8/T0JCAqtWrWqx7c1FdmGkLhcvXsTOzk5cqA6waNEiDh8+zN27d5s1h6urq2ShxCjPDuwfaLT3mYKoqCgEQeD333/nk08+oW3btmKfvb0906dPZ+vWrfj5+TU5l5QVQGQds48fP050dLT47kwt8+bNw8/PjyFDhjQ51+XLl3n55ZdNZaoBsg4jfn5+jBkzhu3btxu0d+jQgTlz5hi8rPR39Ho9Bw8eJCMjg7CwMFObCsjUs+uycuVKQkJCnnr2GB4ezp49e54aHxoaCoBOp2Pw4MEkJiZKciUCSj7bJCj57OcARWwJUcSWEEVsCVHElhBFbAmRndh1U6ze3t4sWrSIS5cuAZCUlERUVJTB+FOnTuHs7MzJkycBOH/+PN7e3pSXl4tjUlJSCA0Npbq6mlmzZjFq1CiDZW2zZs1q8YumdZGd2FBzu56amsrJkycJDg4mMjKSH3/8sd6xWq2WkSNHim/1jhgxArVazebNmwG4d+8eH374ITExMeK7le3atROXtbUmshS7Lj4+PoSGhrJz586n+oqLi8V6IdnZ2dy7dw+oues8e/YsWVlZrFu3jrlz5+Lo6CgeFxkZSUJCAkVFRa1qq3G36zmJkP1lqxogogqHYS3LUbi6upKeno6Xl5dBe2pqKl5eXtjb2+Pt7U1ycjKRkZG0a9eOmJgYFi5ciLOzM6+//rrBcd27dycoKIht27bx5ptvtvQ3egrZezY0nCZNSkoS304IDg42KH/h5ubGgAEDiIiIqHcpcWRkJEeOHPlHJS/+jnGePSysxd5nSupLk/78889cv37doCDA3bt3uXjxIu7u7kBNHb6678DXxc7OjvDwcLZs2dJqdso+65eWlkZiYiI7d+7kxo0bYrtWq2XevHksX75cbKutFVIrdlPMnj0bX1/fBv8gxiLLMBIVFSVe+h0+fJj4+HiDFGtlZSVHjx4lMNDwqZK/vz8nTpwwuOxrjPbt27NgwYJWCyVKitUEKCnW5wBFbAlRxJYQRWwJUcSWEEVsCVHElhDZ3UHWV2W4sUrAdSsQl5eX06tXLyIiIp5KWkmC0Ai5ubmNdT8Txo0bJ1y7dk38XFVVJQwfPly4evWq2HblyhVBr9cLgiAI4eHhQnp6utiXlZUljB49Wjhx4oTJbGxIN6M8+5esO1w923pZsLoMGvMSA0e+ZPRxZWVlRlUC9vDwYOnSpWzfvh0fH58W2dpSZB+zW1IJWMplwnUxyrMHjmyZ95kaYysBC89oo0DZe3YtxlQClnKZcF1kL3ZZWRlnzpwRvbWpSsAXLlwgLi6O+fPnS2kmIMMUq0ajobKyUkzoV1ZW4ubmxs2bN8VKwJMmTWLJkiWA4eYTjx8/pkePHkRERDS4P0Jr0JBusrvOTk9PN2r83r17TWSJ8cg+jMgJRWwJUcSWEEVsCVHElhBFbAmR3aVfVVUVW7du5dixY7Rt2xYLCwtGjhzJihUr2LhxIz/88APm5uYIgsCCBQsICAho1oZtUiA7saOjo6msrESr1WJjY8OTJ0/QarXs2LGDkpISUlNTsbCwoKysTFy1Co1v2NZaK56aQlZh5LfffiMtLY0NGzaI2xG2adOGGTNmUFxcjL29vSictbV1g4mo2g3biouLxeqVUmCUZ1859S0/f3/SJIa4eHkzRD2+0TG5ubn06dOn3n/9adOmMXfuXM6dO4dKpWLs2LFMmDChwbnqbtgm1VMbWXl2Yzg7O/Ptt9+yatUq7O3tWb9+PWvXrm30GMlTrS15vPOsyM/PF1xdXYWSkpImx+bk5AgqlUoQhJpHYcHBwQb9Op1OUKvVwvfff9/qdjakm6w8u2/fvmg0GtauXUtpaSlQUyn40KFDXLhwQdwDEuDKlSsNplnLyspYv369uGGbVMjuamTTpk18+umnTJ06FUtLS/R6PWq1GktLSzZs2EBVVRXm5uZ06dKF999/XzyusQ3bpEJ2+Ww5oCwZfg5QxJYQRWwJUcSWEEVsCVHElhBFbAmR3U2NRqNBp9Nx6tQp8YYkKSmJ6Oho3n77bdq3b8/GjRvp2bOneMxrr72Go6OjWHruzp07WFlZYWdnB8C7775LYmIimZmZ2NnZUVFRgaenJ2vWrMHc/P/9cdmyZWRlZZGRkYGlpaXRtstObKgpcHvmzBnUajUAycnJBtUpR48eXe9r0LXbXq1evRoXFxfCw8PFvsTERCIjIwkPD6e0tJTg4GDc3d2ZNGkSACUlJWRmZuLo6Eh6ejq+vr5G222U2GUXCym7UGj0D2kO1q92x9q9e7PG1r70r1ar+eOPPygvLzdYHP9PsbGxYciQIdy+fVts+/rrr1Gr1Xh6eqLValsktixj9ogRI8jLy+Phw4ckJycz5W/7nWVmZhIUFCR+1RY0by4PHjzg2rVrBkvUtFqtuKlnTk6OWEbaGIzybGv35nufKTEzM8PPz4+jR49y9OhR9u/fz5UrV8T+hsJIU8THx3PgwAHy8/MJCwsT95XMzc3l0aNHjBw5EjMzM3x8fEhJSWHBggVGzS/LmA01oWTatGkMHz5cPNH9U2pj9s2bNwkLC8PT0xO1Wo1Wq+XRo0eMH1/zJEmn02Ftbf3fEdvBwYHly5czdOjQVp+7X79+REVF8dFHHzFq1CiOHDmCVqsVd90D8PX15cKFC7z66qvNnle2YkPNLnn1URuza3FxceF///uf0XPv2bOHtLQ0evfubSA0QEBAAFqt1iixlXy2CVDy2c8BitgSoogtIYrYEqKILSGK2BIiO7GPHz/OlClTCAoKYuLEieK2VhqNBk9PT6qrq8WxSUlJODs78+WXXxp83rdvnzhGEATGjx+Ph4eHyW2X1U3N3bt3eeedd0hOTuall15CEASuXr0q9jeVeoWaIgIpKSnMnDkTqNlZz9bWVlxhZUpk5dn379+nTZs2dOrUCahJSNWtwFC33mpDqVcHBwesrKzEQgHJyclivVZTY5Rn5+TkkJ2dbRJDVCoVw4YNa3TMwIEDGTp0KF5eXnh4eODm5kZQUJCYiBoxYgQJCQkGqde62cBapkyZQnJyMosXL+bixYssXryYuLg4U/xaBsjKs83Nzdm6dSt79+7Fw8ODU6dOERgYSElJCfB06tXf37/eeSZOnEhaWhrHjh1jwoQJkq33M8qzhw0b1qT3SYGTkxNOTk7MnDmTSZMmcf78ebGvOalXa2trhg0bxubNm+vdEstUyOoEWVhYyO3bt1GpVAD8+eefFBUVGSwNbm7qdf78+bzyyis4OztTUFBgUrtrkZXYT548ITY2llu3bmFlZYVer2fZsmVPlSlqKPValwEDBjBgwABTmVovSorVBCgp1ucARWwJUcSWkCbFbiSkK9SDXq9vsK9Rsa2srHjw4IEieDMQBAGdTsetW7ewtraud0yjVyNVVVUUFBRQUVFhMiP/TbRp0wZbW1vs7e0NFmTW0qjYCq2LcoKUEEVsCVHElhBFbAlRxJaQ/wNt8pzSObFY3QAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 72x72 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# legend\n",
    "import matplotlib.lines as mlines\n",
    "line1 = mlines.Line2D([], [], color=colors['LEADSSS'], label='LEADS')\n",
    "line2 = mlines.Line2D([], [], color=colors['NGU'], label='NGU')\n",
    "line3 = mlines.Line2D([], [], color=colors['RND'], label='RND')\n",
    "line4 = mlines.Line2D([], [], color=colors['DIAYN'], label='DIAYN')\n",
    "line5 = mlines.Line2D([], [], color=colors['LSD'], label='LSD')\n",
    "line6 = mlines.Line2D([], [], color=colors['CSD'], label='CSD')\n",
    "line7 = mlines.Line2D([], [], color=colors['METRA'], label='METRA')\n",
    "line8 = mlines.Line2D([], [], color=colors['SMM'], label='SMM')\n",
    "\n",
    "\n",
    "\n",
    "# Création d'une nouvelle figure pour la légende\n",
    "fig_leg, ax_leg = plt.subplots(figsize=(1, 1))\n",
    "ax_leg.axis('off')  # Masquer les axes\n",
    "ax_leg.legend(handles=[line1, line2, line3, line4, line5, line6, line7, line8], loc='center')\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures/legend.pdf', bbox_inches='tight', pad_inches=0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 72x72 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# legend skills \n",
    "skill_colors = ['r', 'g', 'b', 'y', 'm', 'c', 'k', 'w']\n",
    "\n",
    "# create legend\n",
    "skill_legend = []\n",
    "for i in range(6):\n",
    "    # dots \n",
    "    skill_legend.append(mlines.Line2D([], [], color=skill_colors[i], marker='o', linestyle='None', markersize=10, label='Skill {}'.format(i)))\n",
    "fig_leg, ax_leg = plt.subplots(figsize=(1, 1))\n",
    "ax_leg.axis('off')  # Masquer les axes\n",
    "ax_leg.legend(handles=skill_legend, loc='center')\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures/skill_legend.pdf', bbox_inches='tight', pad_inches=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-14-5295b84450b6>:10: UserWarning: Tight layout not applied. The left and right margins cannot be made large enough to accommodate all axes decorations.\n",
      "  plt.tight_layout()\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 72x72 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# legend skills horizontal\n",
    "skill_legend = []\n",
    "for i in range(6):\n",
    "    # dots \n",
    "    skill_legend.append(mlines.Line2D([], [], color=skill_colors[i], marker='o', linestyle='None', markersize=10, label='Skill {}'.format(i)))\n",
    "fig_leg, ax_leg = plt.subplots(figsize=(1, 1))\n",
    "ax_leg.axis('off')  # Masquer les axes\n",
    "# masquer l'encadré de la légende\n",
    "ax_leg.legend(handles=skill_legend, loc='center', ncol=6, frameon=False)\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures/skill_legend_horizontal.pdf', bbox_inches='tight', pad_inches=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-87-261b44811cf7>:13: UserWarning: Tight layout not applied. The bottom and top margins cannot be made large enough to accommodate all axes decorations.\n",
      "  plt.tight_layout()\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 72x72 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import cm\n",
    "# skill label with rainbow color\n",
    "n_skill = 12\n",
    "colors = cm.rainbow(np.linspace(0, 1, n_skill))\n",
    "# create legend\n",
    "skill_legend = []\n",
    "for i in range(n_skill):\n",
    "    # dots \n",
    "    skill_legend.append(mlines.Line2D([], [], color=colors[i], marker='o', linestyle='None', markersize=10, label='Skill {}'.format(i)))\n",
    "fig_leg, ax_leg = plt.subplots(figsize=(1, 1))\n",
    "ax_leg.axis('off')  # Masquer les axes\n",
    "ax_leg.legend(handles=skill_legend, loc='center')\n",
    "plt.tight_layout()\n",
    "plt.savefig('figures/skill_legend_rainbow.pdf', bbox_inches='tight', pad_inches=0)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.9.2"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}