olympus_get_started.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "benchmarking_autonomous_experimentation_with_olympus.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cc6VYmj450Vm"
      },
      "source": [
        "# Benchmarking autonomous experimenation with Olympus\n",
        "\n",
        "In this notebook we show how ``Olympus`` allows to readily test the performance of various experiment planning algortihms against the supported chemistry benchmarks.\n",
        "\n",
        "Notes:\n",
        "- when executing the first cell, you will receive a warning (\"This notebook was not authored by Google\"). Please select \"Run Anyway\" to be able to run the cells of the notebook."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TTjOiS4t7Uu7",
        "outputId": "dcb8e2fb-6f3a-4385-cb42-398e2486d3a5"
      },
      "source": [
        "# we start by installing a couple of dependencies\n",
        "!pip install silence-tensorflow\n",
        "!pip install pyDOE\n",
        "!pip install sobol_seq"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: silence-tensorflow in /usr/local/lib/python3.6/dist-packages (1.1.1)\n",
            "Requirement already satisfied: pyDOE in /usr/local/lib/python3.6/dist-packages (0.3.8)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from pyDOE) (1.19.5)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.6/dist-packages (from pyDOE) (1.4.1)\n",
            "Requirement already satisfied: sobol_seq in /usr/local/lib/python3.6/dist-packages (0.2.0)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from sobol_seq) (1.19.5)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.6/dist-packages (from sobol_seq) (1.4.1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SM0rBQwy5mJx",
        "outputId": "67fa322e-18ea-4b84-8c8f-ab6ad15a90ac"
      },
      "source": [
        "# finally, we install olympus itself\n",
        "!pip install olymp[all]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: olymp[all] in /usr/local/lib/python3.6/dist-packages (0.0.1b0)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (1.1.5)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (1.19.5)\n",
            "Requirement already satisfied: SQSnobFit; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (0.4.5)\n",
            "Requirement already satisfied: deap; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (1.3.1)\n",
            "Requirement already satisfied: pyswarms; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (1.3.0)\n",
            "Requirement already satisfied: cma; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (3.0.3)\n",
            "Requirement already satisfied: phoenics; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (0.2.0)\n",
            "Requirement already satisfied: sqlalchemy; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (1.3.22)\n",
            "Requirement already satisfied: silence-tensorflow; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (1.1.1)\n",
            "Requirement already satisfied: tensorflow-probability==0.8; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (0.8.0)\n",
            "Requirement already satisfied: tensorflow==1.15; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (1.15.0)\n",
            "Requirement already satisfied: seaborn; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (0.11.1)\n",
            "Requirement already satisfied: gpyopt; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (1.2.6)\n",
            "Requirement already satisfied: hyperopt; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (0.1.2)\n",
            "Requirement already satisfied: matplotlib; extra == \"all\" in /usr/local/lib/python3.6/dist-packages (from olymp[all]) (3.2.2)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas->olymp[all]) (2018.9)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.6/dist-packages (from pandas->olymp[all]) (2.8.1)\n",
            "Requirement already satisfied: SQCommon in /usr/local/lib/python3.6/dist-packages (from SQSnobFit; extra == \"all\"->olymp[all]) (0.3.2)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.6/dist-packages (from pyswarms; extra == \"all\"->olymp[all]) (4.41.1)\n",
            "Requirement already satisfied: pyyaml in /usr/local/lib/python3.6/dist-packages (from pyswarms; extra == \"all\"->olymp[all]) (5.3.1)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.6/dist-packages (from pyswarms; extra == \"all\"->olymp[all]) (1.4.1)\n",
            "Requirement already satisfied: attrs in /usr/local/lib/python3.6/dist-packages (from pyswarms; extra == \"all\"->olymp[all]) (20.3.0)\n",
            "Requirement already satisfied: future in /usr/local/lib/python3.6/dist-packages (from pyswarms; extra == \"all\"->olymp[all]) (0.16.0)\n",
            "Requirement already satisfied: wheel>=0.33 in /usr/local/lib/python3.6/dist-packages (from phoenics; extra == \"all\"->olymp[all]) (0.36.2)\n",
            "Requirement already satisfied: watchdog>=0.9 in /usr/local/lib/python3.6/dist-packages (from phoenics; extra == \"all\"->olymp[all]) (1.0.2)\n",
            "Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow-probability==0.8; extra == \"all\"->olymp[all]) (1.15.0)\n",
            "Requirement already satisfied: cloudpickle==1.1.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow-probability==0.8; extra == \"all\"->olymp[all]) (1.1.1)\n",
            "Requirement already satisfied: gast<0.3,>=0.2 in /usr/local/lib/python3.6/dist-packages (from tensorflow-probability==0.8; extra == \"all\"->olymp[all]) (0.2.2)\n",
            "Requirement already satisfied: decorator in /usr/local/lib/python3.6/dist-packages (from tensorflow-probability==0.8; extra == \"all\"->olymp[all]) (4.4.2)\n",
            "Requirement already satisfied: keras-applications>=1.0.8 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (1.0.8)\n",
            "Requirement already satisfied: grpcio>=1.8.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (1.32.0)\n",
            "Requirement already satisfied: tensorflow-estimator==1.15.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (1.15.1)\n",
            "Requirement already satisfied: keras-preprocessing>=1.0.5 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (1.1.2)\n",
            "Requirement already satisfied: astor>=0.6.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (0.8.1)\n",
            "Requirement already satisfied: tensorboard<1.16.0,>=1.15.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (1.15.0)\n",
            "Requirement already satisfied: protobuf>=3.6.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (3.12.4)\n",
            "Requirement already satisfied: google-pasta>=0.1.6 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (0.2.0)\n",
            "Requirement already satisfied: absl-py>=0.7.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (0.10.0)\n",
            "Requirement already satisfied: wrapt>=1.11.1 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (1.12.1)\n",
            "Requirement already satisfied: opt-einsum>=2.3.2 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (3.3.0)\n",
            "Requirement already satisfied: termcolor>=1.1.0 in /usr/local/lib/python3.6/dist-packages (from tensorflow==1.15; extra == \"all\"->olymp[all]) (1.1.0)\n",
            "Requirement already satisfied: GPy>=1.8 in /usr/local/lib/python3.6/dist-packages (from gpyopt; extra == \"all\"->olymp[all]) (1.9.9)\n",
            "Requirement already satisfied: pymongo in /usr/local/lib/python3.6/dist-packages (from hyperopt; extra == \"all\"->olymp[all]) (3.11.2)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.6/dist-packages (from hyperopt; extra == \"all\"->olymp[all]) (2.5)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib; extra == \"all\"->olymp[all]) (1.3.1)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib; extra == \"all\"->olymp[all]) (2.4.7)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.6/dist-packages (from matplotlib; extra == \"all\"->olymp[all]) (0.10.0)\n",
            "Requirement already satisfied: h5py in /usr/local/lib/python3.6/dist-packages (from keras-applications>=1.0.8->tensorflow==1.15; extra == \"all\"->olymp[all]) (2.10.0)\n",
            "Requirement already satisfied: werkzeug>=0.11.15 in /usr/local/lib/python3.6/dist-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow==1.15; extra == \"all\"->olymp[all]) (1.0.1)\n",
            "Requirement already satisfied: setuptools>=41.0.0 in /usr/local/lib/python3.6/dist-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow==1.15; extra == \"all\"->olymp[all]) (51.1.1)\n",
            "Requirement already satisfied: markdown>=2.6.8 in /usr/local/lib/python3.6/dist-packages (from tensorboard<1.16.0,>=1.15.0->tensorflow==1.15; extra == \"all\"->olymp[all]) (3.3.3)\n",
            "Requirement already satisfied: paramz>=0.9.0 in /usr/local/lib/python3.6/dist-packages (from GPy>=1.8->gpyopt; extra == \"all\"->olymp[all]) (0.9.5)\n",
            "Requirement already satisfied: importlib-metadata; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from markdown>=2.6.8->tensorboard<1.16.0,>=1.15.0->tensorflow==1.15; extra == \"all\"->olymp[all]) (3.3.0)\n",
            "Requirement already satisfied: typing-extensions>=3.6.4; python_version < \"3.8\" in /usr/local/lib/python3.6/dist-packages (from importlib-metadata; python_version < \"3.8\"->markdown>=2.6.8->tensorboard<1.16.0,>=1.15.0->tensorflow==1.15; extra == \"all\"->olymp[all]) (3.7.4.3)\n",
            "Requirement already satisfied: zipp>=0.5 in /usr/local/lib/python3.6/dist-packages (from importlib-metadata; python_version < \"3.8\"->markdown>=2.6.8->tensorboard<1.16.0,>=1.15.0->tensorflow==1.15; extra == \"all\"->olymp[all]) (3.4.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IgA8sc3U6atA",
        "outputId": "ca92f17b-8388-473b-eb99-26358102350a"
      },
      "source": [
        "# get a list of available datasets \n",
        "from olympus import list_datasets\n",
        "list_datasets()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['alkox',\n",
              " 'benzylation',\n",
              " 'colors_bob',\n",
              " 'colors_n9',\n",
              " 'fullerenes',\n",
              " 'hplc',\n",
              " 'photo_pce10',\n",
              " 'photo_wf3',\n",
              " 'snar',\n",
              " 'suzuki']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fy9UMfrP6bga",
        "outputId": "2f675ecd-0282-49b4-f4d2-2f61165e2406"
      },
      "source": [
        "# get a list of available experiment planners \n",
        "from olympus import list_planners\n",
        "list_planners()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['BasinHopping',\n",
              " 'Cma',\n",
              " 'ConjugateGradient',\n",
              " 'DifferentialEvolution',\n",
              " 'Genetic',\n",
              " 'Gpyopt',\n",
              " 'Grid',\n",
              " 'Hyperopt',\n",
              " 'LatinHypercube',\n",
              " 'Lbfgs',\n",
              " 'ParticleSwarms',\n",
              " 'Phoenics',\n",
              " 'RandomSearch',\n",
              " 'Simplex',\n",
              " 'Slsqp',\n",
              " 'Snobfit',\n",
              " 'Sobol',\n",
              " 'SteepestDescent']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dZUXIFVU61LW"
      },
      "source": [
        "# initalize the Olympus orchestrator\n",
        "from olympus import Olympus\n",
        "olymp = Olympus()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VrdqaAuLYPR5"
      },
      "source": [
        "The `Olympus` class is the highest-level class in the package, which allows to run various algorithms with default settings against the desired benchmarks in a few lines of code. Lower-level usage of the package, which allow for more control by experienced users, is also possible as described in the documentation."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yDgyXYLW71U4"
      },
      "source": [
        "# we declare a local database to which we store our campaign results\n",
        "from olympus import Database\n",
        "database = Database()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MVWh8bEK7DpX",
        "outputId": "9c238a4b-7d97-47a7-c00a-4cd76222a360"
      },
      "source": [
        "from olympus import Campaign\n",
        "from time import time\n",
        "\n",
        "DATASET = 'hplc'\n",
        "NUM_REPETITIONS = 3\n",
        "PLANNERS = ['Gpyopt', 'Sobol', 'Genetic', 'RandomSearch']\n",
        "\n",
        "elapsed_times = {'planner': [], 'time': []}\n",
        "for PLANNER in PLANNERS:\n",
        "  for repetition in range(NUM_REPETITIONS):\n",
        "    print(f\"Algorithm: {PLANNER} [repetition {repetition+1}]\")\n",
        "\n",
        "    start_time = time()\n",
        "    olymp.run(\n",
        "        planner=PLANNER,      # run simulation with <PLANNER>,\n",
        "        dataset=DATASET,      # on emulator trained on dataset <DATASET>;\n",
        "        campaign=Campaign(),  # store results in a new campaign, \n",
        "        database=database,    # but use the same database to store campaign;\n",
        "        num_iter=100,         # run benchmark for 100 iterations\n",
        "    )\n",
        "    elapsed_time = time() - start_time\n",
        "    elapsed_times['planner'].append(PLANNER)\n",
        "    elapsed_times['time'].append(elapsed_time)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Algorithm: Gpyopt [repetition 1]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: Gpyopt [repetition 2]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: Gpyopt [repetition 3]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: Sobol [repetition 1]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: Sobol [repetition 2]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: Sobol [repetition 3]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: Genetic [repetition 1]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: Genetic [repetition 2]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: Genetic [repetition 3]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: RandomSearch [repetition 1]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: RandomSearch [repetition 2]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0mAlgorithm: RandomSearch [repetition 3]\n",
            "\u001b[0;37m[INFO] Loading emulator using a BayesNeuralNet model for the dataset hplc...\n",
            "\u001b[0m"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dnpd2WBD7SkM",
        "outputId": "af78190c-d329-4919-a0b1-8263c0e59ca1"
      },
      "source": [
        "# collect all campaigns\n",
        "campaigns = [campaign for campaign in database]\n",
        "for campaign in campaigns:\n",
        "  print(repr(campaign))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=Gpyopt, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=Gpyopt, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=Gpyopt, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=Sobol, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=Sobol, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=Sobol, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=Genetic, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=Genetic, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=Genetic, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=RandomSearch, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=RandomSearch, num_iter=100)>\n",
            "<Campaign (dataset=hplc, model=BayesNeuralNet, planner=RandomSearch, num_iter=100)>\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 360
        },
        "id": "RkrLJ8Ij9UU8",
        "outputId": "9a567b4e-588d-4dd4-924f-19aa23c2d99b"
      },
      "source": [
        "from olympus import Plotter \n",
        "plotter = Plotter() \n",
        "plotter.plot_from_db(database)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 288
        },
        "id": "W3rHwlITEBzM",
        "outputId": "79427462-f69c-44ce-f5ee-3f9f70c84347"
      },
      "source": [
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt \n",
        "sns.boxplot(\n",
        "    x='planner', \n",
        "    y='time',\n",
        "    data=elapsed_times\n",
        ")\n",
        "plt.xlabel('Planner') \n",
        "plt.ylabel('Elapsed time [s]')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RuTZ9ZKedGCE"
      },
      "source": [
        "Note that the times reported here reflect not only the algorithm implementation, but also the implementation of their corresponding wrappers in Olympus. In practice, the computational cost of most experiment planners is negligible for the vast majority of practical chemistry experiments. Only the cost of Bayesian optimization algortihms (e.g., ``Phoenics`` and ``GPyOpt``) might be become substantial (i.e. several seconds per iteration) after a few hundres iterations."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YNkKR4BAEPAg"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}