demo_BOIL_A2C_InvertedPendulum.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Logging before flag parsing goes to stderr.\n",
      "W1121 12:40:22.637740 17840 deprecation.py:323] From C:\\Users\\Vu\\Anaconda3\\envs\\tfgpu\\lib\\site-packages\\tensorflow\\python\\compat\\v2_compat.py:96: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "non-resource variables are not supported in the long term\n"
     ]
    }
   ],
   "source": [
    "from collections import OrderedDict\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import bayes_opt.visualization.visualization_utility_cost as viz\n",
    "from bayes_opt.test_functions import drl_experiments\n",
    "from bayes_opt.utility.basic_utility_functions import evaluating_final_util_MaxEpoch,set_seed\n",
    "from bayes_opt.curve_compression import return_logistic_curve\n",
    "\n",
    "from bayes_opt.sequentialBO.boil import BOIL\n",
    "\n",
    "from tqdm import tqdm\n",
    "import warnings\n",
    "\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "#set_seed(1337)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Specifying the search space"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "SearchSpace = OrderedDict([('gamma',(0.8,1)),(\"lr_actor\", (1e-6,1e-2)),(\"lr_critic\", (1e-6,1e-2)),\n",
    "                      ('maxEpisodes',(700, 1500))]) # 700 1500\n",
    "\n",
    "varParams = list(SearchSpace.keys())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# These are the fixed hyperparamters for DRL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "fixedParams = OrderedDict([\n",
    "        (\"lr_actor\", 5e-4),\n",
    "        (\"lr_critic\", 5e-4),\n",
    "        (\"ent_coef\", 0.01),\n",
    "        (\"maxSteps\", 200),\n",
    "        (\"val_architecture\", [32,32]),\n",
    "        (\"pol_architecture\", [32,32])])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Initializing the Advantage Actor Critic (A2C) on InvertedPendulum-v2 environment (require Mujoco license)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "myfunction=drl_experiments.DRL_experiment(\"A2C\", \"InvertedPendulum-v2\", varParams, fixedParams, SearchSpace)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Initializing BOIL class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "acq_name='ei_mu_max'\n",
    "\n",
    "func_params={}\n",
    "func_params['function']=myfunction\n",
    "\n",
    "boil=BOIL(myfunction.func,myfunction.bounds,acq_name=acq_name,verbose=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Initializing BOIL with 10 points"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "initialization 10 points - please wait...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W1121 12:40:24.489398 17840 deprecation.py:506] From C:\\Users\\Vu\\Anaconda3\\envs\\tfgpu\\lib\\site-packages\\tensorflow\\python\\ops\\linalg\\linear_operator_diag.py:166: calling LinearOperator.__init__ (from tensorflow.python.ops.linalg.linear_operator) with graph_parents is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Do not pass `graph_parents`.  They will  no longer be used.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Max #augmented points 10 \t #augmented points  8 \t Augmented points:  [0.    0.106 0.245 0.365 0.5   0.654 0.783 0.905]\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n"
     ]
    }
   ],
   "source": [
    "n_init_points=10\n",
    "print(\"initialization {} points - please wait...\".format(n_init_points))\n",
    "boil.init(n_init_points)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Running BOIL over multiple iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running BOIL for 40 iterations\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  0%|                                                                                           | 0/40 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.2\n",
      "selected point from acq func: [0.2155 0.8834 0.     0.9753] acq val=log(Utility/Cost)= -1.4024\n",
      "Max #augmented points 9 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8431 0.0088 0.    ] t=1480 current y=35.25, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  2%|██                                                                              | 1/40 [14:10<9:12:34, 850.12s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.2\n",
      "selected point from acq func: [0.2156 0.8161 0.0588 1.    ] acq val=log(Utility/Cost)= -1.0862\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8431 0.0082 0.0006] t=1500 current y=57.13, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  5%|████                                                                            | 2/40 [28:14<8:57:22, 848.48s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.2\n",
      "selected point from acq func: [0.1826 0.9096 0.0518 1.    ] acq val=log(Utility/Cost)= 0.7254\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8365 0.0091 0.0005] t=1500 current y=68.31, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      "  8%|█████▊                                                                        | 3/40 [54:10<10:54:06, 1060.71s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.2\n",
      "selected point from acq func: [0.4492 0.5481 0.1373 0.8854] acq val=log(Utility/Cost)= -1.5399\n",
      "Max #augmented points 8 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8898 0.0055 0.0014] t=1408 current y=97.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 10%|███████▌                                                                    | 4/40 [1:24:07<12:48:56, 1281.56s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.2\n",
      "selected point from acq func: [0.3757 0.537  0.1605 1.    ] acq val=log(Utility/Cost)= -1.5236\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8751 0.0054 0.0016] t=1500 current y=79.60, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 12%|█████████▌                                                                  | 5/40 [1:47:41<12:50:46, 1321.33s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.2\n",
      "selected point from acq func: [0.4556 0.6305 0.1474 1.    ] acq val=log(Utility/Cost)= -3.4403\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8911 0.0063 0.0015] t=1500 current y=2.35, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 15%|███████████▋                                                                  | 6/40 [1:50:39<9:14:22, 978.31s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.2\n",
      "==estimated lengthscale_x=0.0300   lengthscale_t=0.300   Logistic_m0=3.0   Logistic_g0=2.0\n",
      "selected point from acq func: [0.437  0.5073 0.0541 1.    ] acq val=log(Utility/Cost)= -0.8152\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8874 0.0051 0.0005] t=1500 current y=1.80, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 18%|█████████████▋                                                                | 7/40 [1:56:32<7:14:51, 790.65s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.4\n",
      "selected point from acq func: [0.4557 0.4991 0.6106 0.4336] acq val=log(Utility/Cost)= -4.6503\n",
      "Max #augmented points 4 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8911 0.005  0.0061] t=1047 current y=0.11, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 20%|███████████████▌                                                              | 8/40 [1:59:25<5:22:54, 605.45s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.4\n",
      "selected point from acq func: [0.1402 0.8433 0.0247 1.    ] acq val=log(Utility/Cost)= 0.1207\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.828  0.0084 0.0002] t=1500 current y=3.24, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 22%|█████████████████▌                                                            | 9/40 [2:18:39<6:37:44, 769.82s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.4\n",
      "selected point from acq func: [0.6668 0.913  0.0051 0.1565] acq val=log(Utility/Cost)= -12.9207\n",
      "Max #augmented points 1 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.9334 0.0091 0.0001] t=825 current y=0.01, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 25%|███████████████████▎                                                         | 10/40 [2:19:53<4:40:34, 561.15s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.4\n",
      "selected point from acq func: [0.2707 0.8908 0.0812 1.    ] acq val=log(Utility/Cost)= -1.2117\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8541 0.0089 0.0008] t=1500 current y=2.95, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 28%|█████████████████████▏                                                       | 11/40 [2:32:13<4:57:13, 614.96s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.4\n",
      "selected point from acq func: [0.4497 0.5289 0.1718 1.    ] acq val=log(Utility/Cost)= 0.7481\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8899 0.0053 0.0017] t=1500 current y=24.52, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 30%|███████████████████████                                                      | 12/40 [2:42:54<4:50:33, 622.62s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.4\n",
      "selected point from acq func: [0.3948 0.5826 0.0913 0.8227] acq val=log(Utility/Cost)= -1.1709\n",
      "Max #augmented points 8 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.879  0.0058 0.0009] t=1358 current y=4.56, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 32%|█████████████████████████                                                    | 13/40 [3:11:35<7:08:32, 952.30s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.5\n",
      "selected point from acq func: [0.5732 0.5138 0.1574 0.1863] acq val=log(Utility/Cost)= -1.6579\n",
      "Max #augmented points 1 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.9146 0.0051 0.0016] t=849 current y=0.01, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 35%|██████████████████████████▉                                                  | 14/40 [3:20:20<5:57:01, 823.89s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.5\n",
      "==estimated lengthscale_x=0.0300   lengthscale_t=0.300   Logistic_m0=3.0   Logistic_g0=2.0\n",
      "selected point from acq func: [0.4447 0.7411 0.     0.    ] acq val=log(Utility/Cost)= 1.4929\n",
      "[original scale] x=[0.8889 0.0074 0.    ] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 38%|████████████████████████████▉                                                | 15/40 [3:23:54<4:27:01, 640.88s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.5\n",
      "selected point from acq func: [0.3534 0.562  0.5141 1.    ] acq val=log(Utility/Cost)= -0.7618\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8707 0.0056 0.0051] t=1500 current y=107.80, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 40%|██████████████████████████████▍                                             | 16/40 [4:19:12<9:37:37, 1444.05s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.5\n",
      "selected point from acq func: [0.5291 0.7604 0.     0.7216] acq val=log(Utility/Cost)= -0.9255\n",
      "Max #augmented points 7 \t #augmented points  6 \t Augmented points:  [0.    0.124 0.285 0.379 0.526 0.602]\n",
      "[original scale] x=[0.9058 0.0076 0.    ] t=1181 current y=0.61, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 42%|████████████████████████████████▎                                           | 17/40 [4:33:11<8:04:02, 1262.73s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.8\n",
      "selected point from acq func: [0.2724 0.5118 0.4971 1.    ] acq val=log(Utility/Cost)= -4.0835\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8545 0.0051 0.005 ] t=1500 current y=2.66, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 45%|██████████████████████████████████▋                                          | 18/40 [4:35:04<5:36:30, 917.74s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.8\n",
      "selected point from acq func: [0.4148 0.6101 0.5313 1.    ] acq val=log(Utility/Cost)= -0.9427\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.883  0.0061 0.0053] t=1500 current y=109.32, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 48%|████████████████████████████████████                                        | 19/40 [5:16:32<8:06:06, 1388.87s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.8\n",
      "selected point from acq func: [0.2061 0.8662 0.0513 1.    ] acq val=log(Utility/Cost)= 0.0717\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8412 0.0087 0.0005] t=1500 current y=103.40, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 50%|██████████████████████████████████████                                      | 20/40 [5:55:55<9:20:20, 1681.04s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.8\n",
      "selected point from acq func: [0.0989 0.8761 0.2592 1.    ] acq val=log(Utility/Cost)= -0.7145\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8198 0.0088 0.0026] t=1500 current y=76.91, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 52%|███████████████████████████████████████▉                                    | 21/40 [6:19:54<8:29:21, 1608.50s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.8\n",
      "selected point from acq func: [0.4116 0.5422 0.5358 1.    ] acq val=log(Utility/Cost)= -1.7729\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8823 0.0054 0.0054] t=1500 current y=100.28, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 55%|█████████████████████████████████████████▊                                  | 22/40 [7:01:51<9:24:16, 1880.89s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.8\n",
      "selected point from acq func: [0.3613 0.6253 0.5039 0.8695] acq val=log(Utility/Cost)= -2.0676\n",
      "Max #augmented points 8 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8723 0.0063 0.005 ] t=1396 current y=73.11, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 57%|███████████████████████████████████████████▋                                | 23/40 [7:40:00<9:27:37, 2003.37s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.8\n",
      "selected point from acq func: [1. 0. 1. 0.] acq val=log(Utility/Cost)= -2.8905\n",
      "[original scale] x=[1.   0.   0.01] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 60%|█████████████████████████████████████████████▌                              | 24/40 [7:43:28<6:30:38, 1464.90s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.8\n",
      "==estimated lengthscale_x=0.0300   lengthscale_t=0.300   Logistic_m0=3.0   Logistic_g0=2.0\n",
      "selected point from acq func: [0.3322 0.637  0.     0.    ] acq val=log(Utility/Cost)= 0.2277\n",
      "[original scale] x=[0.8664 0.0064 0.    ] t=700 current y=0.04, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 62%|███████████████████████████████████████████████▌                            | 25/40 [8:01:00<5:35:16, 1341.07s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [0.7176 0.6845 0.     0.    ] acq val=log(Utility/Cost)= -2.2643\n",
      "[original scale] x=[0.9435 0.0068 0.    ] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 65%|█████████████████████████████████████████████████▍                          | 26/40 [8:08:55<4:12:17, 1081.26s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [1.     0.     0.0177 0.7918] acq val=log(Utility/Cost)= -3.0979\n",
      "Max #augmented points 7 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[1.     0.     0.0002] t=1333 current y=1.49, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 68%|███████████████████████████████████████████████████▉                         | 27/40 [8:13:09<3:00:27, 832.87s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [0.8121 0.0391 0.4744 0.    ] acq val=log(Utility/Cost)= -3.9587\n",
      "[original scale] x=[0.9624 0.0004 0.0047] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 70%|█████████████████████████████████████████████████████▉                       | 28/40 [8:23:36<2:34:15, 771.26s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [1.     0.8597 0.5415 0.    ] acq val=log(Utility/Cost)= -3.2401\n",
      "[original scale] x=[1.     0.0086 0.0054] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 72%|███████████████████████████████████████████████████████▊                     | 29/40 [8:29:41<1:59:02, 649.34s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [0.5656 0.     0.     0.    ] acq val=log(Utility/Cost)= -3.3067\n",
      "[original scale] x=[0.9131 0.     0.    ] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 75%|█████████████████████████████████████████████████████████▊                   | 30/40 [8:36:04<1:34:52, 569.27s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [1.     0.425  0.3237 0.    ] acq val=log(Utility/Cost)= -3.3812\n",
      "[original scale] x=[1.     0.0043 0.0032] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 78%|███████████████████████████████████████████████████████████▋                 | 31/40 [8:38:26<1:06:12, 441.37s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [0.9041 0.265  0.     0.    ] acq val=log(Utility/Cost)= -3.4959\n",
      "[original scale] x=[0.9808 0.0027 0.    ] t=700 current y=0.04, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 80%|█████████████████████████████████████████████████████████████▌               | 32/40 [8:53:44<1:17:54, 584.27s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "==estimated lengthscale_x=0.0300   lengthscale_t=0.300   Logistic_m0=3.0   Logistic_g0=2.0\n",
      "selected point from acq func: [0.4675 0.632  0.     0.    ] acq val=log(Utility/Cost)= 0.1328\n",
      "[original scale] x=[0.8935 0.0063 0.    ] t=700 current y=0.04, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 82%|███████████████████████████████████████████████████████████████▌             | 33/40 [9:09:13<1:20:13, 687.65s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [0.3689 0.5997 0.5599 1.    ] acq val=log(Utility/Cost)= -2.6681\n",
      "Max #augmented points 10 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[0.8738 0.006  0.0056] t=1500 current y=80.72, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 85%|█████████████████████████████████████████████████████████████████▍           | 34/40 [9:34:36<1:33:49, 938.31s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [1.     1.     0.1852 0.    ] acq val=log(Utility/Cost)= -4.3764\n",
      "[original scale] x=[1.     0.01   0.0019] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 88%|███████████████████████████████████████████████████████████████████▍         | 35/40 [9:43:22<1:07:53, 814.68s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [1.     0.575  0.     0.2486] acq val=log(Utility/Cost)= -4.6493\n",
      "Max #augmented points 2 \t #augmented points  0 \t Augmented points:  []\n",
      "[original scale] x=[1.     0.0058 0.    ] t=899 current y=0.02, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 90%|███████████████████████████████████████████████████████████████████████        | 36/40 [9:47:27<42:55, 643.76s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [1.     0.     0.2203 0.    ] acq val=log(Utility/Cost)= -4.3072\n",
      "[original scale] x=[1.     0.     0.0022] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 92%|█████████████████████████████████████████████████████████████████████████      | 37/40 [9:58:44<32:40, 653.60s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [1.     0.7088 0.2554 0.    ] acq val=log(Utility/Cost)= -5.241\n",
      "[original scale] x=[1.     0.0071 0.0026] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 95%|██████████████████████████████████████████████████████████████████████████    | 38/40 [10:01:00<16:36, 498.42s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [1.     0.3146 0.6599 0.    ] acq val=log(Utility/Cost)= -5.0712\n",
      "[original scale] x=[1.     0.0031 0.0066] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r",
      " 98%|████████████████████████████████████████████████████████████████████████████  | 39/40 [10:01:57<06:05, 365.78s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ln of conditioning number of GP covariance matrix 9.9\n",
      "selected point from acq func: [1.     1.     0.8646 0.    ] acq val=log(Utility/Cost)= -5.4989\n",
      "[original scale] x=[1.     0.01   0.0086] t=700 current y=0.00, ybest=127.79\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████████████████████████████████████████████████████████████████████████| 40/40 [10:03:01<00:00, 275.46s/it]\n"
     ]
    }
   ],
   "source": [
    "NN=10*myfunction.input_dim\n",
    "print(\"Running BOIL for {} iterations\".format(NN))\n",
    "strFolderOut=\"run_experiments\\plot\"\n",
    "\n",
    "for index in tqdm(range(0,NN)):\n",
    "    boil.suggest_nextpoint()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# We evaluate the final performance score util MaxEpoch. The final utility score is marginalizing with different choices for the Logisic curve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████| 49/49 [00:00<00:00, 49179.44it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAADgCAYAAAAZk/GlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAGPRJREFUeJzt3X+YXVV97/H3ZxIyEERCzDBCAhkGY6r4FOidUihei6AWKQrXiwo3V9KWa9o+3mqtvRTFp7Q+psUfLeittjdXKEEjwhNAFL2WSI2IheAEkB9iBCKENGEyFCI/ko4k871/7DXhzHBm5pwzc84+s/fn9TzzzNlr77P3WieT9T1rfc9ZWxGBmZnZWB15V8DMzNqTA4SZmVXlAGFmZlU5QJiZWVUOEGZmVpUDhJmZVeUAYdYmJK2X9D9m2rmtuBwgrG1IekzSbknPS3pG0rckHTFN533LBPtPkTScrjvy882pXne6SDovtUFjymdL2iHpzLzqZsXmAGHt5h0R8QrgMGAA+N8tuu62iHhFxc87WnTdWtwIzAN+a0z56UAA32l5jawUHCCsLUXEfwBrgdePlEnqlPRZSVskDUj6R0kHpH0LJN0saaekpyX9QFKHpC8DRwLfTCODC+upR7rm5ZK2pZ/LJXWmfb8r6fYxx4ek16THV0n6QhoJPSdpg6SjK459q6SfSvqFpL8HRo0QxrwW1wHnj9l1PrAmIvZIOiS1fzCNvm6WtGicNv2lpK9UbPekes9O2wdLukLSdkn/JumTkmbV87pZMThAWFuSNBd4L3BnRfGngNcCxwGvARYCf5H2fQTYCnQB3cDHgIiI9wFbSCOTiPh0nVW5GDgxXfNY4ATg43U8/zzgr4BDgEeAlal9C4Dr07kWAI8CJ09wntXAORUB8WDgHcDVaX8H8E/AYrKAuBv4+zrqOfZae8he4+OBtwHOX5SQA4S1m69L2gk8C7wV+AxAmn9/P/DhiHg6Ip4D/ho4Nz3vRbJpqcUR8WJE/CDqW2js8DT6GPl5TypfBnwiInZExCBZZ/++Os57Q0TcFRF7gDVkgQbgDOAnEbE2Il4ELgeeHO8kEfFDsim3/5KK3gP8LCLuTfv/PSKuj4hd6bVZycunpCYlqRt4O/AnEfFCROwALuOl19lKZHbeFTAb4+yI+G6a0jgL+L6k1wPDwFxgY0WuVsDI1MdngL8Ebkn7V0XEpXVcd1tEVJuSORx4vGL78VRWq8pOfxfwiorzPjGyIyJC0hNM7GqyaaWvkgWp1SM70ojrMrK8xCGp+CBJsyJibx31XQzsB2yveJ07Kutq5eERhLWliNgbETcAe4E3Ak+RTZscExHz0s/BKaFNRDwXER+JiF6yqZc/lXTayOmmUJVtZJ3miCNTGcALZEELAEmvruO824F9n9BKI6TJPrF1NXCapJPIpr2+WrHvI8BS4Dci4pXAm0ZOXeU8o+oNVNb7CWAIWFDxOr8yIo6poU1WMA4Q1paUOYvs3fBDETEM/F/gMkmHpmMWSvrt9PhMSa9JHe2zZIFl5J3zANDbYFWuAT4uqSvlDf4CGEnw/hg4RtJxkvYnG8HU6lvpue9KyeEPMrqjfpmIeBy4PdVpXURUjk4OIgugOyXNBy6Z4FT3Am+SdGTKZXy04hrbgVuAv5X0ypToP1pS3dNVNvM5QFi7+aak58k6+ZXA8oh4MO37c7JE752SngW+S/auGWBJ2n4euAP4YkSsT/v+hqyT3ynpz+qszyeBfuA+4H7g7lRGRPwM+ES67sNknXdNIuIp4N3ApcC/p/r/sIanriYb0Vw9pvxy4ACykdadTPDR14hYB1yb2rQRuHnMIecDc4CfAM+QfZrssBrqZgUj3zDIzMyq8QjCzMyqcoAwM7OqHCDMzKwqBwgzM6vKAcLMzKqa0d+kXrBgQfT09ORdDTOzGWXjxo1PRUTXZMfN6ADR09NDf39/3tUwM5tRJD0++VGeYjIzs3E4QJiZWVUOEFMwsGaAO3ruYH3Heu7ouYOBNQMNlU/nufK8hq/tf9ciXXumta8ZZvRSG319fZFXDmJgzQCbVmxieNfwvrKOuR28evmreXL1kzWXL12VLSU0HefK8xq+tv9di3Ttmda+pauW0r2sm1pJ2hgRfZMe5wDRmDt67mDo8aGpn2jkbgb1rNjfjtfwtVt/7aK3L89rz7D2dS7u5KTHTqr5+FoDRNOmmCRdKWmHpAcqyj6T7sF7n6QbJc2r2PdRSY9I2jSyhHM7G9oyDcEBRi9K3SytuIav3fprF719eV57hrVv2vqjMZqZg7iK7O5WldYBb4iIXwV+RlqHPt0x7FzgmPScL7b7TdI7j+ysvmO8Wo9TPufwOcw5fM60nCvPa/jarb920dvn17b254zbH01R0wJERNwGPD2m7JZ0b17I1qwfucXjWcDXImIoIn5Otub/Cc2q23ToXdlLxwGjX76OuR0cvuJwOubWXn70p4/m6E8fXddz2vEavnbrr1309vm1rf05vSsbvR/WxPL8FNPvA/8vPV7I6Hvebk1lLyNphaR+Sf2Dg4NNruL4upd10/vZl/5ROhd3snTVUl77xdeydNVSOhd3giYv717WTfey7rqe047X8LX971qka8+09tWToK5HU5PUknqAmyPiDWPKLwb6gHelm7V/AbgjIr6S9l8BfDsirp/o/HkmqQF2P7qbDa/ZwP5H7c+Jm0/MrR5mZvWoNUnd8qU2JC0HzgROi5ei01ZG37B9ES/dGL5t7X0hyyTNOrCt0yVmZg1p6RSTpNPJ7iv8zojYVbHrG8C5kjolHUV2f967Wlm3RowEiI4D/X1DMyuepo0gJF0DnAIskLQVuITsU0udwDpJAHdGxB9GxIOSriO7Sfoe4AMRkdeHzGrmEYSZFVnTAkREnFel+IoJjl8JrGxWfZph+IXs24wOEGZWRJ4bmQKPIMysyBwgpsA5CDMrMvdsU+ARhJkVmQPEFDgHYWZF5gAxBR5BmFmROUBMgXMQZlZk7tmmwCMIMysyB4gpcA7CzIrMAWIKPIIwsyJzgJgC5yDMrMjcs02BRxBmVmQOEFPgHISZFZkDxBR4BGFmReYAMQXOQZhZkblnmwKPIMysyBwgGhQRDO9KOYi5DhBmVjwOEA0a3j0MAR37d6BZyrs6ZmbTzgGiQc4/mFnRuXdrkPMPZlZ0DhAN8ncgzKzoHCAa5BGEmRVd0wKEpCsl7ZD0QEXZfEnrJD2cfh+SyiXp85IekXSfpF9rVr2mi3MQZlZ0zezdrgJOH1N2EXBrRCwBbk3bAG8HlqSfFcA/NLFe08IjCDMruqYFiIi4DXh6TPFZwOr0eDVwdkX51ZG5E5gn6bBm1W06OAdhZkXX6vmR7ojYDpB+H5rKFwJPVBy3NZW1LY8gzKzo2mUCvdo3zaLqgdIKSf2S+gcHB5tcrfE5B2FmRdfq3m1gZOoo/d6RyrcCR1QctwjYVu0EEbEqIvoioq+rq6uplZ2IRxBmVnStDhDfAJanx8uBmyrKz0+fZjoR+MXIVFS7cg7CzIpudrNOLOka4BRggaStwCXApcB1ki4AtgDvTod/GzgDeATYBfxes+o1XTyCMLOia1qAiIjzxtl1WpVjA/hAs+rSDM5BmFnRuXdrkEcQZlZ0DhANcg7CzIrOAaJBHkGYWdE5QDTIOQgzKzr3bg3yCMLMis4BokHOQZhZ0TlANMgjCDMrOgeIBjkHYWZF596tARHhEYSZFZ4DRAPilwF7QfuJjv38EppZMbl3a4BHD2ZWBg4QDXD+wczKwD1cAzyCMLMycIBogL8DYWZl4ADRAI8gzKwMHCAa4ByEmZWBe7gGeARhZmXgANEA5yDMrAwcIBrgEYSZlYEDRAOcgzCzMnAP1wCPIMysDBwgGuAchJmVQS4BQtKHJT0o6QFJ10jaX9JRkjZIeljStZLm5FG3WngEYWZl0PIAIWkh8EGgLyLeAMwCzgU+BVwWEUuAZ4ALWl23WjkHYWZlkFcPNxs4QNJsYC6wHTgVWJv2rwbOzqluk/IIwszKoOUBIiL+DfgssIUsMPwC2AjsjIg96bCtwMJqz5e0QlK/pP7BwcFWVPllnIMwszLIY4rpEOAs4CjgcOBA4O1VDo1qz4+IVRHRFxF9XV1dzavoBDyCMLMyyGOK6S3AzyNiMCJeBG4AfhOYl6acABYB23KoW02cgzCzMsijh9sCnChpriQBpwE/Ab4HnJOOWQ7clEPdauIRhJmVwezJDwFJhwInk00J7QYeAPojYrjeC0bEBklrgbuBPcA9wCrgW8DXJH0ylV1R77lbxTkIMyuDCQOEpDcDFwHzyTrtHcD+ZJ8wOjp19H8bEc/Wc9GIuAS4ZEzxZuCEes6TF48gzKwMJhtBnAG8PyK2jN2R8gVnAm8Frm9C3dqWcxBmVgYTBoiI+F8T7NsDfH3aa9TmhvcME78M6ICOTgcIMyuumno4SV+WdHDFdo+kW5tXrfZVmX/IcuxmZsVU61vg24ENks6Q9H7gFuDy5lWrfTn/YGZlUdOnmCLi/0h6kOyjqE8Bx0fEk02tWZty/sHMyqLWKab3AVcC5wNXAd+WdGwT69W2PIIws7KoaQQB/FfgjRGxA7hG0o1kgeL4ZlWsXfk7EGZWFrVOMZ09ZvsuSb/RnCq1N48gzKwsJpxikvRxSfOr7YuIX0o6VdKZzalae3IOwszKYrIRxP3ANyX9B9nSGINk36ReAhwHfBf466bWsM14BGFmZTFZgDgnIk6WdCHZMhuHAc8CXwFWRMTuZlew3TgHYWZlMVmA+E+SFgPLgDeP2XcA2cJ9peIRhJmVxWQB4h+B7wC9QH9Fuchu6NPbpHq1LecgzKwsJuzlIuLzEfE64MqI6K34OSoiShccwCMIMyuPmt4GR8QfNbsiM4VzEGZWFp4nqZNHEGZWFg4QdXIOwszKwr1cnTyCMLOycICok3MQZlYWDhB18gjCzMrCAaJOzkGYWVnk0stJmidpraSfSnpI0kmS5ktaJ+nh9PuQPOo2GY8gzKws8nob/DngOxHxK8CxwEPARcCtEbEEuDVttx3nIMysLFoeICS9EngTcAVky4ZHxE7gLGB1Omw1cHb1M+TLIwgzK4s8RhC9ZMuG/5OkeyR9SdKBQHdEbAdIvw+t9mRJKyT1S+ofHBxsXa2BGA6Gd2cjiI4DnIMws2LLo5ebDfwa8A8RcTzwAnVMJ0XEqojoi4i+rq6uZtWxqr27UoJ6bgfqUEuvbWbWankEiK3A1ojYkLbXkgWMAUmHAaTfO3Ko24T25R/menrJzIqv5QEiIp4EnpC0NBWdBvwE+AawPJUtB25qdd0m44+4mlmZTHY/iGb5Y2CNpDnAZuD3yILVdZIuALYA786pbuNygtrMyiSXABER9wJ9VXad1uq61MMBwszKxHMldfB3IMysTBwg6uAchJmViXu6OniKyczKxAGiDg4QZlYmDhB1cA7CzMrEAaIOzkGYWZm4p6uDp5jMrEwcIOrgAGFmZeIAUQfnIMysTBwg6rBvNVfnIMysBNzT1cFTTGZWJg4QdfAUk5mViQNEHTyCMLMycYCog78HYWZl4p6uDh5BmFmZOEDUwTkIMysTB4g6eARhZmXiAFGjiHAOwsxKxT1djYaHhmEYNEd0zPbLZmbF556uRs4/mFnZOEDUyPkHMyub3AKEpFmS7pF0c9o+StIGSQ9LulbSnLzqVo3zD2ZWNnn2dh8CHqrY/hRwWUQsAZ4BLsilVuPwCMLMyiaXACFpEfA7wJfStoBTgbXpkNXA2XnUbTzOQZhZ2eQ1grgcuBAYTtuvAnZGxJ60vRVYmEfFxuMRhJmVTcsDhKQzgR0RsbGyuMqhMc7zV0jql9Q/ODjYlDpW4xyEmZVNHr3dycA7JT0GfI1saulyYJ6k2emYRcC2ak+OiFUR0RcRfV1dXa2oL+ARhJmVT8sDRER8NCIWRUQPcC7wLxGxDPgecE46bDlwU6vrNhHnIMysbNppvuTPgT+V9AhZTuKKnOszikcQZlY2syc/pHkiYj2wPj3eDJyQZ30m4hyEmZWNe7saeQRhZmXjAFEj5yDMrGwcIGrkEYSZlY0DRI2cgzCzsnFvVyOPIMysbBwgauQchJmVjQNEjTyCMLOycYCokXMQZlY27u1q5BGEmZWNA0SNnIMws7JxgKiRRxBmVjYOEDUYfnGYeDFgFmhOtVtXmJkVjwNEDbZftT17sBfuPOpOBtYM5FshM7MWcICYxMCaAR790KP7toceH2LTik0OEmZWeA4Qk9h80WaGdw+PKhveNczmizfnVCMzs9bI9X4QeRhYM8DmizcztGWIziM76V3ZS/ey7pfvO6KTg3/rYIa2DlU9z9CW6uVmZkVRqgAxsGaATSs2MbwrGxGMTBeNGLVvyxA7vrxj3HN1HtnZ3MqameWsVAFi88Wb9wWAEcO7htn0B5v2PR6r46AO2Dt6X8fcDnpX9ja3smZmOStVDmK8aaHhF4b3fRHuZfueH2bpqqV0Lu4EQefiTpauWrpvWsrMrKhKNYLoPLKTocdfHiRmz89ehj1P76n6nO5l3Q4IZlY6pRpB9K7spWPu6CZ3zO1gyeeXsOTzS6ru81SSmZVVqUYQI6OA8T7FNNk+M7MyUUS09oLSEcDVwKuBYWBVRHxO0nzgWqAHeAx4T0Q8M9G5+vr6or+/v7kVNjMrGEkbI6JvsuPymGLaA3wkIl4HnAh8QNLrgYuAWyNiCXBr2jYzs5y0PEBExPaIuDs9fg54CFgInAWsToetBs5udd3MzOwluSapJfUAxwMbgO6I2A5ZEAEOHec5KyT1S+ofHBxsVVXNzEqn5TmIfReWXgF8H1gZETdI2hkR8yr2PxMRh0xyjkHg8QarsAB4qsHnznRlbbvbXS5u9/gWR0TXZCfK5VNMkvYDrgfWRMQNqXhA0mERsV3SYcD461wktTRwgjr015KkKaKytt3tLhe3e+paPsUkScAVwEMR8XcVu74BLE+PlwM3tbpuZmb2kjxGECcD7wPul3RvKvsYcClwnaQLgC3Au3Oom5mZJS0PEBFxOzDefTtPa2FVVrXwWu2mrG13u8vF7Z6i3JLUZmbW3kq1FpOZmdWulAFC0umSNkl6RFJhv7Et6UpJOyQ9UFE2X9I6SQ+n3xN+lHgmknSEpO9JekjSg5I+lMoL3XZJ+0u6S9KPU7v/KpUfJWlDave1kubkXddmkDRL0j2Sbk7bhW+3pMck3S/pXkn9qWza/s5LFyAkzQK+ALwdeD1wXlrqo4iuAk4fU1aGJU3KupzLEHBqRBwLHAecLulE4FPAZandzwAX5FjHZvoQ2coMI8rS7jdHxHEVH22dtr/z0gUI4ATgkYjYHBG/BL5GtsxH4UTEbcDTY4oLv6RJWZdziczzaXO/9BPAqcDaVF64dgNIWgT8DvCltC1K0O5xTNvfeRkDxELgiYrtramsLGpa0qQoGlnOZSZL0yz3kn3RdB3wKLAzIkbuhlXUv/fLgQvJVogGeBXlaHcAt0jaKGlFKpu2v/NS3Q8iqfYRW3+Uq4DSci7XA38SEc9mbyqLLSL2AsdJmgfcCLyu2mGtrVVzSToT2BERGyWdMlJc5dBCtTs5OSK2SToUWCfpp9N58jKOILYCR1RsLwK25VSXPAykpUyodUmTmWii5VzS/sK2HSAidgLryXIw8ySNvBks4t/7ycA7JT1GNmV8KtmIoujtJiK2pd87yN4QnMA0/p2XMUD8CFiSPuEwBziXbJmPsij8kiZlXc5FUlcaOSDpAOAtZPmX7wHnpMMK1+6I+GhELIqIHrL/z/8SEcsoeLslHSjpoJHHwNuAB5jGv/NSflFO0hlk7zBmAVdGxMqcq9QUkq4BTiFb3XEAuAT4OnAdcCRpSZOIGJvIntEkvRH4AXA/L81Jf4wsD1HYtkv6VbKk5CyyN3/XRcQnJPWSvbOeD9wD/PeIGMqvps2Tppj+LCLOLHq7U/tuTJuzga9GxEpJr2Ka/s5LGSDMzGxyZZxiMjOzGjhAmJlZVQ4QZmZWlQOEmZlV5QBhZmZVOUCYAZKeT797JP23aT73x8Zs/+t0nt+sWRwgzEbrAeoKEGmF4ImMChAR8Zt11sksFw4QZqNdCvzntL7+h9Pid5+R9CNJ90n6A8i+kJXuOfFVsi/kIenradG0B0cWTpN0KXBAOt+aVDYyWlE69wNpTf/3Vpx7vaS1kn4qaY3KsJCUtZ0yLtZnNpGLSN/EBUgd/S8i4tcldQI/lHRLOvYE4A0R8fO0/fsR8XRa5uJHkq6PiIsk/c+IOK7Ktd5Fdt+GY8m+7f4jSbelfccDx5CtH/RDsvWGbp/+5pqNzyMIs4m9DTg/LaG9gWwZ6SVp310VwQHgg5J+DNxJtiDkEib2RuCaiNgbEQPA94Ffrzj31ogYBu4lm/oyaymPIMwmJuCPI+KfRxVma/68MGb7LcBJEbFL0npg/xrOPZ7KNYP24v+rlgOPIMxGew44qGL7n4E/SsuHI+m1aeXMsQ4GnknB4VfIltke8eLI88e4DXhvynN0AW8C7pqWVphNA78rMRvtPmBPmiq6Cvgc2fTO3SlRPEj1Wzh+B/hDSfcBm8immUasAu6TdHdahnrEjcBJwI/JbmZzYUQ8mQKMWe68mquZmVXlKSYzM6vKAcLMzKpygDAzs6ocIMzMrCoHCDMzq8oBwszMqnKAMDOzqhwgzMysqv8PCcFQXC5LDu4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "boil.Y_original=evaluating_final_util_MaxEpoch(boil)\n",
    "viz.show_optimization_progress(boil)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot the training curve of the best found hyperparameter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best found hyperparameter = [0.84089045 0.0087813  0.00027485]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Training curve of best found hyperparameter')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# select the optimal hyperparameter\n",
    "idxMax=np.argmax(boil.Y_original)\n",
    "print(\"Best found hyperparameter =\",boil.X_original[idxMax])\n",
    "\n",
    "# play the best parameter to Max Iteration\n",
    "T_max=boil.SearchSpace[-1,1]\n",
    "\n",
    "curve,time=myfunction.func([boil.X_original[idxMax,:-1],T_max])\n",
    "curve=curve[0]\n",
    "average100_curve=[np.mean( curve[max(0,n-100):n+1] )     for n in range(len(curve)) ]\n",
    "\n",
    "# plot the found curve\n",
    "fig=plt.figure()\n",
    "plt.plot(average100_curve)\n",
    "plt.xlabel('Episode',fontsize=16)\n",
    "plt.ylabel('Average Reward',fontsize=16)\n",
    "plt.title(\"Training curve of best found hyperparameter\",fontsize=18)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Play the InvertedPendulum environment using the best found hyperparameter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.84089045, 0.0087813 , 0.00027485])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "boil.X_original[idxMax]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot the Conditioning number of the GP covariance matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Log of Cond Number of Cov Matrix')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=plt.figure(figsize=(5,5))\n",
    "plt.plot(np.log(boil.condition_number),linewidth=4)\n",
    "plt.xlabel('Iterations',fontsize=14)\n",
    "plt.ylabel('Log of Cond Number of Cov Matrix',fontsize=14)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot the estimated preference function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Estimated Logistic Curve $m_0$=3.0 $g_0$=2.0')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "estimated_curve=return_logistic_curve(midpoint=boil.gp.logistic_hyper['midpoint'], \n",
    "                                      growth=boil.gp.logistic_hyper['growth'],\n",
    "                                      MaxEpoch=SearchSpace['maxEpisodes'][1])\n",
    "\n",
    "fig=plt.figure(figsize=(7,5))\n",
    "plt.plot(estimated_curve,linewidth=4)\n",
    "plt.xlabel('Epochs',fontsize=14)\n",
    "plt.xlabel('Epochs',fontsize=14)\n",
    "\n",
    "strTitle=\"Estimated Logistic Curve $m_0$={:.1f} $g_0$={:.1f}\".format(boil.gp.logistic_hyper['midpoint'],boil.gp.logistic_hyper['growth'])\n",
    "plt.title(strTitle,fontsize=16)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}