noisy_optimization_BayBE_original_bounds.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ddfe121c-9cad-4d3f-b518-e42c72a1b70b",
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "97c15027-9275-43e5-9613-ecbfe31d4914",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import pandas as pd\n",
    "import sys\n",
    "sys.path.append('./src/baybe_utils')\n",
    "from run_grid_experiments_baybe import run_grid_experiments\n",
    "from run_grid_experiments_baybe_random import run_grid_experiments_random\n",
    "from src import visualization\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from src import schwefel"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5bcc2d3b-c62b-422c-bdee-15eee6f1452c",
   "metadata": {},
   "source": [
    "## Intro\n",
    "\n",
    "This work lightly explores the impact of a noisy oracle on Bayesian optimization performance. Here, the 2-dimensional Schwefel function with Gaussian noise is used as an optimization objective. However, this work is motivated by the need to deal with noisy data when applying BO to experimental optimization. \n",
    "\n",
    "This notebook looks at optimizing the schwefel function over the range [-50, 50]. See the analogous `noisy_optimization_BayBE_extendedBounds` notebook for optimization over the range [0, 500] which includes the global minimum of the objective. \n",
    "\n",
    "## Implementation Notes\n",
    "\n",
    "- This work was done as an entry in the 2024 Bayesian optimization for materials hackathon\n",
    "- Our team pursued multiple implementations in parallel\n",
    "- This notebook serves as an entry point to the BayBE implementation of the project\n",
    "- Individual optimization campaigns are run from the run_experiments() function in run_experiment_babye.py\n",
    "- Grid screening of parameters builds on this with funcitonality in the run_grid_experiments_babye.py\n",
    "- As this was a hackathon project, there are some hacks. Watch out for hard-coded gotchas throughout. Would not recommend directly re-using code. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02f05875-7a5c-4f50-8b55-fc0e30340897",
   "metadata": {},
   "source": [
    "## 1. Define grid search parameters \n",
    "\n",
    "This is to run a grid search over experiment parameters like number of BO trials to run, noise level, etc. These values were chosen by the team as 'reasonable' sounding values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "5319bdb1-b6a0-4dfa-b05c-bd854456278d",
   "metadata": {},
   "outputs": [],
   "source": [
    "seeds = list(range(5)) # run 5 replicates of each parameter set\n",
    "n_inits = [2, 4, 8, 10] # Number of initial randomly collected data points\n",
    "noise_levels = [1, 5, 10, 20] # variance (?) of gaussian noise. Bigger number -> more noise\n",
    "noise_bools = [True] # carryover from BoTorch side of project, ignore\n",
    "budget = 30 # Run 30 iterations of BO\n",
    "bounds = (-50,50)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abdf100a-84fe-466f-9b60-67d698e7578e",
   "metadata": {},
   "source": [
    "## 2. Run grid search\n",
    "\n",
    "Run the grid search over parameters. This will take a minute or 60. Results are written to disk so if you are just following along, skip this step and load below "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d8ffedfe-8639-479f-9cee-63f4a0f32e54",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 1 budget 30 seed 0, time: 9.03s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 5 budget 30 seed 0, time: 15.98s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 10 budget 30 seed 0, time: 23.97s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 20 budget 30 seed 0, time: 31.18s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 1 budget 30 seed 0, time: 37.90s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 5 budget 30 seed 0, time: 45.18s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 10 budget 30 seed 0, time: 54.21s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 20 budget 30 seed 0, time: 64.93s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 1 budget 30 seed 0, time: 73.17s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 5 budget 30 seed 0, time: 82.01s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 10 budget 30 seed 0, time: 88.16s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 20 budget 30 seed 0, time: 95.95s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 1 budget 30 seed 0, time: 104.06s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 5 budget 30 seed 0, time: 110.81s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 10 budget 30 seed 0, time: 118.32s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 20 budget 30 seed 0, time: 125.44s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 1 budget 30 seed 1, time: 134.62s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 5 budget 30 seed 1, time: 142.04s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 10 budget 30 seed 1, time: 152.19s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 20 budget 30 seed 1, time: 160.21s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 1 budget 30 seed 1, time: 167.79s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 5 budget 30 seed 1, time: 175.58s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 10 budget 30 seed 1, time: 185.98s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 20 budget 30 seed 1, time: 192.50s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 1 budget 30 seed 1, time: 201.80s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 5 budget 30 seed 1, time: 209.80s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 10 budget 30 seed 1, time: 217.41s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 20 budget 30 seed 1, time: 224.21s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 1 budget 30 seed 1, time: 232.65s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 5 budget 30 seed 1, time: 240.03s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 10 budget 30 seed 1, time: 246.48s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 20 budget 30 seed 1, time: 255.71s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 1 budget 30 seed 2, time: 263.26s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 5 budget 30 seed 2, time: 272.02s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 10 budget 30 seed 2, time: 279.79s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 20 budget 30 seed 2, time: 287.12s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 1 budget 30 seed 2, time: 295.66s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 5 budget 30 seed 2, time: 305.75s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 10 budget 30 seed 2, time: 317.32s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 20 budget 30 seed 2, time: 327.66s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 1 budget 30 seed 2, time: 336.58s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 5 budget 30 seed 2, time: 344.40s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 10 budget 30 seed 2, time: 354.19s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 20 budget 30 seed 2, time: 363.36s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 1 budget 30 seed 2, time: 372.11s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 5 budget 30 seed 2, time: 379.97s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 10 budget 30 seed 2, time: 388.45s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 20 budget 30 seed 2, time: 402.06s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 1 budget 30 seed 3, time: 411.24s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 5 budget 30 seed 3, time: 419.22s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 10 budget 30 seed 3, time: 428.12s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 20 budget 30 seed 3, time: 436.11s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 1 budget 30 seed 3, time: 444.58s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 5 budget 30 seed 3, time: 453.59s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 10 budget 30 seed 3, time: 462.30s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 20 budget 30 seed 3, time: 471.83s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 1 budget 30 seed 3, time: 480.32s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 5 budget 30 seed 3, time: 488.51s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 10 budget 30 seed 3, time: 496.07s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 20 budget 30 seed 3, time: 504.09s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 1 budget 30 seed 3, time: 512.36s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 5 budget 30 seed 3, time: 521.28s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 10 budget 30 seed 3, time: 531.76s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 20 budget 30 seed 3, time: 544.60s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 1 budget 30 seed 4, time: 554.02s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 5 budget 30 seed 4, time: 561.61s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 10 budget 30 seed 4, time: 569.97s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 2 noise 20 budget 30 seed 4, time: 579.05s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 1 budget 30 seed 4, time: 586.18s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 5 budget 30 seed 4, time: 593.57s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 10 budget 30 seed 4, time: 600.87s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 4 noise 20 budget 30 seed 4, time: 608.02s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 1 budget 30 seed 4, time: 615.86s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 5 budget 30 seed 4, time: 622.96s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 10 budget 30 seed 4, time: 628.90s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 8 noise 20 budget 30 seed 4, time: 635.66s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 1 budget 30 seed 4, time: 642.71s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 5 budget 30 seed 4, time: 650.22s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 10 budget 30 seed 4, time: 655.89s\n",
      "Collecting initial observations\n",
      "Beginning optimization campaign\n",
      "Started problem 10 noise 20 budget 30 seed 4, time: 663.50s\n",
      "all experiments done, time: 663.50s\n"
     ]
    }
   ],
   "source": [
    "run_grid_experiments(seeds, n_inits, noise_levels, noise_bools, budget, bounds, 'results_baybe_originalBounds/results_baybe')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "16aeaf0f-14b6-4827-b94a-cb87cd2a4d7c",
   "metadata": {},
   "source": [
    "### Run random search as well\n",
    "\n",
    "Generate some random baseline data to compare against\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5aac4ea7-cb9f-48a4-a874-e4396d8b9e22",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting random observations observations\n",
      "Started problem 2 noise 1 budget 30 seed 0, time: 1.24s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 5 budget 30 seed 0, time: 2.34s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 10 budget 30 seed 0, time: 3.42s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 20 budget 30 seed 0, time: 4.47s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 1 budget 30 seed 0, time: 5.72s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 5 budget 30 seed 0, time: 6.72s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 10 budget 30 seed 0, time: 7.76s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 20 budget 30 seed 0, time: 8.82s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 1 budget 30 seed 0, time: 10.01s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 5 budget 30 seed 0, time: 11.16s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 10 budget 30 seed 0, time: 12.33s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 20 budget 30 seed 0, time: 13.38s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 1 budget 30 seed 0, time: 14.50s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 5 budget 30 seed 0, time: 15.53s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 10 budget 30 seed 0, time: 16.56s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 20 budget 30 seed 0, time: 17.75s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 1 budget 30 seed 1, time: 18.83s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 5 budget 30 seed 1, time: 19.85s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 10 budget 30 seed 1, time: 21.08s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 20 budget 30 seed 1, time: 22.06s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 1 budget 30 seed 1, time: 23.14s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 5 budget 30 seed 1, time: 24.14s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 10 budget 30 seed 1, time: 25.17s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 20 budget 30 seed 1, time: 26.16s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 1 budget 30 seed 1, time: 27.23s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 5 budget 30 seed 1, time: 28.26s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 10 budget 30 seed 1, time: 29.34s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 20 budget 30 seed 1, time: 30.44s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 1 budget 30 seed 1, time: 31.49s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 5 budget 30 seed 1, time: 32.57s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 10 budget 30 seed 1, time: 33.72s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 20 budget 30 seed 1, time: 34.84s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 1 budget 30 seed 2, time: 35.90s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 5 budget 30 seed 2, time: 37.27s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 10 budget 30 seed 2, time: 38.41s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 20 budget 30 seed 2, time: 39.53s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 1 budget 30 seed 2, time: 40.74s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 5 budget 30 seed 2, time: 41.94s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 10 budget 30 seed 2, time: 43.17s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 20 budget 30 seed 2, time: 44.51s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 1 budget 30 seed 2, time: 45.78s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 5 budget 30 seed 2, time: 46.87s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 10 budget 30 seed 2, time: 47.91s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 20 budget 30 seed 2, time: 48.90s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 1 budget 30 seed 2, time: 50.06s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 5 budget 30 seed 2, time: 51.21s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 10 budget 30 seed 2, time: 52.31s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 20 budget 30 seed 2, time: 53.49s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 1 budget 30 seed 3, time: 54.92s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 5 budget 30 seed 3, time: 55.90s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 10 budget 30 seed 3, time: 57.02s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 20 budget 30 seed 3, time: 58.10s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 1 budget 30 seed 3, time: 59.16s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 5 budget 30 seed 3, time: 60.19s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 10 budget 30 seed 3, time: 61.83s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 20 budget 30 seed 3, time: 63.11s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 1 budget 30 seed 3, time: 64.16s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 5 budget 30 seed 3, time: 65.30s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 10 budget 30 seed 3, time: 66.58s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 20 budget 30 seed 3, time: 67.84s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 1 budget 30 seed 3, time: 69.13s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 5 budget 30 seed 3, time: 70.30s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 10 budget 30 seed 3, time: 71.39s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 20 budget 30 seed 3, time: 72.60s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 1 budget 30 seed 4, time: 73.87s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 5 budget 30 seed 4, time: 74.97s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 10 budget 30 seed 4, time: 76.17s\n",
      "Collecting random observations observations\n",
      "Started problem 2 noise 20 budget 30 seed 4, time: 77.46s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 1 budget 30 seed 4, time: 78.68s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 5 budget 30 seed 4, time: 79.76s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 10 budget 30 seed 4, time: 81.08s\n",
      "Collecting random observations observations\n",
      "Started problem 4 noise 20 budget 30 seed 4, time: 82.32s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 1 budget 30 seed 4, time: 83.38s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 5 budget 30 seed 4, time: 84.39s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 10 budget 30 seed 4, time: 85.45s\n",
      "Collecting random observations observations\n",
      "Started problem 8 noise 20 budget 30 seed 4, time: 86.59s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 1 budget 30 seed 4, time: 87.73s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 5 budget 30 seed 4, time: 88.80s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 10 budget 30 seed 4, time: 90.28s\n",
      "Collecting random observations observations\n",
      "Started problem 10 noise 20 budget 30 seed 4, time: 91.53s\n",
      "all experiments done, time: 91.53s\n"
     ]
    }
   ],
   "source": [
    "run_grid_experiments_random(seeds, n_inits, noise_levels, noise_bools, budget, bounds, 'results_baybe_originalBounds/results_random_baybe')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34ac9617-f371-40b1-b1a1-5203d84b1a65",
   "metadata": {},
   "source": [
    "## 3. Process Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ce441669-a569-468f-a482-fec91efef95f",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_24219/3826372340.py:17: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.\n",
      "  bo_results = pd.concat([bo_results, pd.DataFrame({\"n_init\": [n_init], \"noise_level\": [noise_level], \"seed\": [seed], \"noise_bool\": [noise_bool],\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n_init</th>\n",
       "      <th>noise_level</th>\n",
       "      <th>seed</th>\n",
       "      <th>noise_bool</th>\n",
       "      <th>best</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "      <td>767.079651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>767.079651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "      <td>767.079651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>767.079651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>767.079651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "      <td>767.079651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>767.079651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "      <td>767.079651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>779.453003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>767.079651</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>80 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   n_init noise_level seed noise_bool        best\n",
       "0       2           1    0       True  767.079651\n",
       "0       2           1    1       True  767.079651\n",
       "0       2           1    2       True  767.079651\n",
       "0       2           1    3       True  767.079651\n",
       "0       2           1    4       True  767.079651\n",
       "..    ...         ...  ...        ...         ...\n",
       "0      10          20    0       True  767.079651\n",
       "0      10          20    1       True  767.079651\n",
       "0      10          20    2       True  767.079651\n",
       "0      10          20    3       True  779.453003\n",
       "0      10          20    4       True  767.079651\n",
       "\n",
       "[80 rows x 5 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load BO results \n",
    "sm_list_bo = {}\n",
    "bo_results = pd.DataFrame(columns=[\"n_init\", \"noise_level\", \"seed\", \"noise_bool\", \"best\"])\n",
    "for noise_bool in noise_bools:\n",
    "    for n_init in n_inits:\n",
    "        for noise_level in noise_levels:\n",
    "            sm_agg = torch.zeros((len(seeds), n_init+budget))\n",
    "            for idx, seed in enumerate(seeds):\n",
    "                X, Y, Y_real, model = torch.load(f\"results_baybe_originalBounds/results_baybe/Schwe_n_init_{n_init}_noiselvl_{noise_level}_budget_{budget}_seed_{seed}_noise_{noise_bool}.pt\")\n",
    "                sliding_min = torch.zeros(Y.shape[0])\n",
    "                for i in range(Y_real.shape[0]):\n",
    "                    sliding_min[i] = Y_real[:i+1].min().item()\n",
    "                \n",
    "                sm_agg[idx] = sliding_min\n",
    "                sm = pd.Series(sliding_min.numpy())\n",
    "                \n",
    "                bo_results = pd.concat([bo_results, pd.DataFrame({\"n_init\": [n_init], \"noise_level\": [noise_level], \"seed\": [seed], \"noise_bool\": [noise_bool],\n",
    "                                                  \"best\": [sliding_min[-1].item()]})])\n",
    "                \n",
    "            sm_mean = sm_agg.mean(0)\n",
    "            sm_std = sm_agg.std(0)\n",
    "            sm_list_bo[(n_init, noise_level, noise_bool)] = (sm_mean, sm_std)\n",
    "bo_results  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "af01900d-a2bf-4c49-90a0-b3614a59c2fc",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_24219/2042071605.py:17: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.\n",
      "  random_results = pd.concat([random_results, pd.DataFrame({\"n_init\": [n_init], \"noise_level\": [noise_level], \"seed\": [seed], \"noise_bool\": [noise_bool],\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n_init</th>\n",
       "      <th>noise_level</th>\n",
       "      <th>seed</th>\n",
       "      <th>noise_bool</th>\n",
       "      <th>best</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "      <td>789.469910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>796.131165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "      <td>802.132568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>794.562561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>792.798035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "      <td>789.469910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>790.011475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>2</td>\n",
       "      <td>True</td>\n",
       "      <td>796.409363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>794.562561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10</td>\n",
       "      <td>20</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>792.798035</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>80 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   n_init noise_level seed noise_bool        best\n",
       "0       2           1    0       True  789.469910\n",
       "0       2           1    1       True  796.131165\n",
       "0       2           1    2       True  802.132568\n",
       "0       2           1    3       True  794.562561\n",
       "0       2           1    4       True  792.798035\n",
       "..    ...         ...  ...        ...         ...\n",
       "0      10          20    0       True  789.469910\n",
       "0      10          20    1       True  790.011475\n",
       "0      10          20    2       True  796.409363\n",
       "0      10          20    3       True  794.562561\n",
       "0      10          20    4       True  792.798035\n",
       "\n",
       "[80 rows x 5 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load random results \n",
    "sm_list_random = {}\n",
    "random_results = pd.DataFrame(columns=[\"n_init\", \"noise_level\", \"seed\", \"noise_bool\", \"best\"])\n",
    "for noise_bool in noise_bools:\n",
    "    for n_init in n_inits:\n",
    "        for noise_level in noise_levels:\n",
    "            sm_agg = torch.zeros((len(seeds), n_init+budget))\n",
    "            for idx, seed in enumerate(seeds):\n",
    "                X, Y, Y_real, model = torch.load(f\"results_baybe_originalBounds/results_random_baybe/Schwe_n_init_{n_init}_noiselvl_{noise_level}_budget_{budget}_seed_{seed}_noise_{noise_bool}.pt\")\n",
    "                sliding_min = torch.zeros(Y.shape[0])\n",
    "                for i in range(Y_real.shape[0]):\n",
    "                    sliding_min[i] = Y_real[:i+1].min().item()\n",
    "                \n",
    "                sm_agg[idx] = sliding_min\n",
    "                sm = pd.Series(sliding_min.numpy())\n",
    "                \n",
    "                random_results = pd.concat([random_results, pd.DataFrame({\"n_init\": [n_init], \"noise_level\": [noise_level], \"seed\": [seed], \"noise_bool\": [noise_bool],\n",
    "                                                  \"best\": [sliding_min[-1].item()]})])\n",
    "                \n",
    "            sm_mean = sm_agg.mean(0)\n",
    "            sm_std = sm_agg.std(0)\n",
    "            sm_list_random[(n_init, noise_level, noise_bool)] = (sm_mean, sm_std)\n",
    "random_results  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a62c64e2-b44c-4cb2-8e3c-4955abb334de",
   "metadata": {},
   "source": [
    "Calculate 'performance matrix' to generate iterations vs noise heat map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f9d3ca6c-22f6-459e-9107-cfa6cc663673",
   "metadata": {},
   "outputs": [],
   "source": [
    "performance_matrix_bo = np.zeros((len(n_inits), len(noise_levels)))\n",
    "\n",
    "for i, init in enumerate(n_inits):\n",
    "    for j, noise in enumerate(noise_levels):\n",
    "        y_vals = torch.load(f'results_baybe_originalBounds/results_baybe/Schwe_n_init_{init}_noiselvl_{noise}_budget_30_seed_0_noise_True.pt')[2]\n",
    "        best_y = torch.min(y_vals)\n",
    "        performance_matrix_bo[i,j] = best_y\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "cd2d4601-b8ee-4d96-8c54-9dad0285151c",
   "metadata": {},
   "outputs": [],
   "source": [
    "performance_matrix_random = np.zeros((len(n_inits), len(noise_levels)))\n",
    "\n",
    "for i, init in enumerate(n_inits):\n",
    "    for j, noise in enumerate(noise_levels):\n",
    "        y_vals = torch.load(f'results_baybe_originalBounds/results_random_baybe/Schwe_n_init_{init}_noiselvl_{noise}_budget_30_seed_0_noise_True.pt')[2]\n",
    "        best_y = torch.min(y_vals)\n",
    "        performance_matrix_random[i,j] = best_y"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9ba43cdf-7017-4853-a200-945af8e20e61",
   "metadata": {},
   "source": [
    "## 4. Plot\n",
    "\n",
    "### backtesting plots\n",
    "\n",
    "1. Fix n_init, compare noise level"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "8f2b0de9-e2b8-40a6-a910-a4945f06aaba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_init_val = 10\n",
    "#df_bo = bo_results.groupby([\"n_init\", \"noise_level\"]).agg({\"best\": [\"mean\", \"std\"]})\n",
    "#df_rand = random_results.groupby([\"n_init\", \"noise_level\"]).agg({\"best\": [\"mean\", \"std\"]})\n",
    "\n",
    "# we already got the statistics from all seeds above, but only want to plot one example for each so just pick first seed \n",
    "plot_bo = bo_results[bo_results['seed'] == 0]\n",
    "plot_rand = random_results[random_results['seed'] == 0]\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "for idx, row in plot_bo.iterrows():\n",
    "    if row['n_init'] == n_init_val:\n",
    "        mean = sm_list_bo[(n_init_val, row['noise_level'], True)][0][n_init_val:]\n",
    "        std = sm_list_bo[(n_init_val, row['noise_level'], True)][1][n_init_val:]\n",
    "        plt.plot(mean, label=f\"BO, noise_level={row['noise_level']}\")\n",
    "        #plt.fill_between(range(len(mean)), mean-std, mean+std, alpha=0.1)\n",
    "    \n",
    "for idx, row in plot_rand.iterrows():\n",
    "    if row['n_init'] == n_init_val:\n",
    "        mean = sm_list_random[(n_init_val, row['noise_level'], True)][0][n_init_val:]\n",
    "        std = sm_list_random[(n_init_val, row['noise_level'], True)][1][n_init_val:]\n",
    "        plt.plot(mean, label=f\"Random Baseline\", linestyle=\"--\")\n",
    "        break\n",
    "        #plt.fill_between(range(len(mean)), mean-std, mean+std, alpha=0.1)\n",
    "\n",
    "# aaawaaay\n",
    "plt.legend(loc=\"upper right\", bbox_to_anchor=(1.3, 1))\n",
    "plt.title(\"BayBE Optimization, 10 initial observations\")\n",
    "\n",
    "ax.set_xlabel('Campaign iteration')\n",
    "ax.set_ylabel('Schwefel function value')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8989570b-6b02-4e6a-99c7-f7ccafb0bb7f",
   "metadata": {},
   "source": [
    "2. Fix noise value, compare initial data number"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "409f03ea-c50e-461d-850d-f7723b031f98",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "noise_level_val = 10\n",
    "#df_bo = bo_results.groupby([\"n_init\", \"noise_level\"]).agg({\"best\": [\"mean\", \"std\"]})\n",
    "#df_rand = random_results.groupby([\"n_init\", \"noise_level\"]).agg({\"best\": [\"mean\", \"std\"]})\n",
    "\n",
    "# we already got the statistics from all seeds above, but only want to plot one example for each so just pick first seed \n",
    "plot_bo = bo_results[bo_results['seed'] == 0]\n",
    "plot_rand = random_results[random_results['seed'] == 0]\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "\n",
    "for idx, row in plot_bo.iterrows():\n",
    "    if row['noise_level'] == noise_level_val:\n",
    "        n_init_val = row['n_init']\n",
    "        mean = sm_list_bo[(row['n_init'], noise_level_val, True)][0][n_init_val:]\n",
    "        std = sm_list_bo[(row['n_init'], noise_level_val, True)][1][n_init_val:]\n",
    "        plt.plot(mean, label=f\"BO, n_init={row['n_init']}\")\n",
    "        #plt.fill_between(range(len(mean)), mean-std, mean+std, alpha=0.1)\n",
    "    \n",
    "for idx, row in plot_rand.iterrows():\n",
    "    if row['noise_level'] == noise_level_val:\n",
    "        n_init_val = row['n_init']\n",
    "        mean = sm_list_random[(row['n_init'], noise_level_val, True)][0][n_init_val:]\n",
    "        std = sm_list_random[(row['n_init'], noise_level_val, True)][1][n_init_val:]\n",
    "        plt.plot(mean, label=f\"Random Baseline, n_init={row['n_init']}\", linestyle=\"--\")\n",
    "        #plt.fill_between(range(len(mean)), mean-std, mean+std, alpha=0.1)\n",
    "        break\n",
    "\n",
    "# aaawaaay\n",
    "plt.legend(loc=\"upper right\", bbox_to_anchor=(1.6, 1))\n",
    "plt.title(\"BayBE Optimization, Noise level = 10\")\n",
    "\n",
    "ax.set_xlabel('Campaign iteration')\n",
    "ax.set_ylabel('Schwefel function value')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae3f1ea5-856d-47a3-9010-7a4d4ceb8b1f",
   "metadata": {},
   "source": [
    "### Heat map plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "e90c4e71-e671-415d-950b-377a463730b4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Bayesian Optimization')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "fig, ax = plt.subplots()\n",
    "visualization.grid_search_heatmap(n_inits, noise_levels, performance_matrix_bo)\n",
    "\n",
    "ax.set_title('Bayesian Optimization')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "5d63c7cd-755c-4fd3-8506-a317bf5bdd43",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Random')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "visualization.grid_search_heatmap(n_inits, noise_levels, performance_matrix_random)\n",
    "ax.set_title('Random')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "caba841d-9015-4097-a8f6-f046370d33d4",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}