{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Bayesian Optimization with ENTMOOT\n",
    "\n",
    "This notebook describes how to perform Bayesian optimization with a simple test function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from entmoot import Enting, ProblemConfig, PyomoOptimizer\n",
    "from entmoot.benchmarks import build_reals_only_problem, eval_reals_only_testfunc\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define problem\n",
    "problem_config = ProblemConfig(rnd_seed=73)\n",
    "build_reals_only_problem(problem_config)\n",
    "# sample data\n",
    "N_init = 10\n",
    "train_x = problem_config.get_rnd_sample_list(num_samples=N_init)\n",
    "train_y = eval_reals_only_testfunc(train_x)\n",
    "\n",
    "# create enting object\n",
    "# this defines the surrogate model\n",
    "params = {\"unc_params\": {\"dist_metric\": \"l1\", \"acq_sense\": \"exploration\"}}\n",
    "enting = Enting(problem_config, params=params)\n",
    "\n",
    "# create optimizer object\n",
    "params_pyomo = {\"solver_name\": \"gurobi\", \"verbose\": False}\n",
    "opt_pyo = PyomoOptimizer(problem_config, params=params_pyomo)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now run the BO loop `N_iterations` times. At each iteration, we fit the surrogate to the data, then use the optimizer to propose a new experiment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "N_iterations = 40\n",
    "for itr in range(N_iterations):\n",
    "    enting.fit(train_x, train_y)\n",
    "    result = opt_pyo.solve(enting)\n",
    "\n",
    "    next_x = np.reshape(result.opt_point, (1, -1))\n",
    "    next_y = eval_reals_only_testfunc(next_x)\n",
    "\n",
    "    # update the training data\n",
    "    train_x = np.concatenate([train_x, next_x], axis=0)\n",
    "    train_y = np.concatenate([train_y, next_y], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "iterations = np.arange(N_init, N_init + N_iterations + 1)\n",
    "plt.step(iterations, np.minimum.accumulate(train_y)[N_init-1:])"
   ]
  }
 ],
 "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": "3.10.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}