Parameters_vs_Hyperparameters & Hyperparameter_Optimization.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Parameters vs Hyperparameters/Hyperparameter Optimization.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "W9SM_pqgR1zk"
      },
      "source": [
        "# Parameters vs Hyperparameters\n",
        "\n",
        "Parameters\n",
        "\n",
        "\n",
        "*   Intrinsic to model equation\n",
        "*   Optimized duringtraining\n",
        "\n",
        "\n",
        "Hyperparameters\n",
        "\n",
        "\n",
        "*   Defined beforetraining\n",
        "*   Constrain the algorithm\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6U53kqrwTvDE"
      },
      "source": [
        "EXAMPLES OF HYPERPARAMETERS\n",
        "\n",
        "\n",
        "*   Number of trees\n",
        "*   The depth of the tree\n",
        "\n",
        "\n",
        "*   Learning rate (GBMs)\n",
        "*   The metric of split quality\n",
        "\n",
        "\n",
        "*   The number of features to evaluate at each node\n",
        "*   The minimum number of samples to split the data further\n",
        "\n",
        "\n",
        "*   Regularization Parameters\n",
        "*   ...\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LbMZMAHIUWLS"
      },
      "source": [
        "# Hyperparameter Optimization\n",
        "\n",
        "The process of finding the best Hyperparameters for a given dataset is called Hyperparameter Optimization or Hyperparameter Tuning.\n",
        "\n",
        "Method to choose the hyperparameters that minimize the generalization error (not necessarily the loss)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XqQ5EBkzYWZj"
      },
      "source": [
        "## Challenges\n",
        "\n",
        "We can’t define a formula to find the hyperparameters\n",
        "\n",
        "Try different combinations of hyperparameter and evaluate model performance\n",
        "\n",
        "The critical step is to choose how many different hyperparameter combinations we are going to test.\n",
        "\n",
        "![a.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cTECstxZYxsq"
      },
      "source": [
        "## Search\n",
        "\n",
        "A search consist of:\n",
        "\n",
        "\n",
        "\n",
        "*   Hyperparameter space\n",
        "*   A method for sampling candidate hyperparameters\n",
        "*   A cross-validation scheme\n",
        "*   A performance metric to minimize (or maximize)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UhwmmrouZGEi"
      },
      "source": [
        "### Hyperparameter response surface\n",
        "\n",
        "![a.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NNxRT13iZbNz",
        "outputId": "13ef9426-74bf-4242-bdc0-7b9d64d24a6c"
      },
      "source": [
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "from mpl_toolkits import mplot3d\n",
        "\n",
        "from sklearn.datasets import load_diabetes\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.model_selection import train_test_split, GridSearchCV\n",
        "\n",
        "\n",
        "# if you want more information about the dataset for this demo:\n",
        "\n",
        "# scikit-learn dataset\n",
        "# https://scikit-learn.org/stable/datasets/toy_dataset.html#diabetes-dataset\n",
        "\n",
        "# in short, regression problem, trying to predict disease progression based on\n",
        "# a bunch of variables\n",
        "\n",
        "# load dataset\n",
        "diabetes_X, diabetes_y = load_diabetes(return_X_y=True)\n",
        "X = pd.DataFrame(diabetes_X)\n",
        "y = diabetes_y\n",
        "\n",
        "X.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "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>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.038076</td>\n",
              "      <td>0.050680</td>\n",
              "      <td>0.061696</td>\n",
              "      <td>0.021872</td>\n",
              "      <td>-0.044223</td>\n",
              "      <td>-0.034821</td>\n",
              "      <td>-0.043401</td>\n",
              "      <td>-0.002592</td>\n",
              "      <td>0.019908</td>\n",
              "      <td>-0.017646</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>-0.001882</td>\n",
              "      <td>-0.044642</td>\n",
              "      <td>-0.051474</td>\n",
              "      <td>-0.026328</td>\n",
              "      <td>-0.008449</td>\n",
              "      <td>-0.019163</td>\n",
              "      <td>0.074412</td>\n",
              "      <td>-0.039493</td>\n",
              "      <td>-0.068330</td>\n",
              "      <td>-0.092204</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.085299</td>\n",
              "      <td>0.050680</td>\n",
              "      <td>0.044451</td>\n",
              "      <td>-0.005671</td>\n",
              "      <td>-0.045599</td>\n",
              "      <td>-0.034194</td>\n",
              "      <td>-0.032356</td>\n",
              "      <td>-0.002592</td>\n",
              "      <td>0.002864</td>\n",
              "      <td>-0.025930</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-0.089063</td>\n",
              "      <td>-0.044642</td>\n",
              "      <td>-0.011595</td>\n",
              "      <td>-0.036656</td>\n",
              "      <td>0.012191</td>\n",
              "      <td>0.024991</td>\n",
              "      <td>-0.036038</td>\n",
              "      <td>0.034309</td>\n",
              "      <td>0.022692</td>\n",
              "      <td>-0.009362</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.005383</td>\n",
              "      <td>-0.044642</td>\n",
              "      <td>-0.036385</td>\n",
              "      <td>0.021872</td>\n",
              "      <td>0.003935</td>\n",
              "      <td>0.015596</td>\n",
              "      <td>0.008142</td>\n",
              "      <td>-0.002592</td>\n",
              "      <td>-0.031991</td>\n",
              "      <td>-0.046641</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          0         1         2         3         4         5         6  \\\n",
              "0  0.038076  0.050680  0.061696  0.021872 -0.044223 -0.034821 -0.043401   \n",
              "1 -0.001882 -0.044642 -0.051474 -0.026328 -0.008449 -0.019163  0.074412   \n",
              "2  0.085299  0.050680  0.044451 -0.005671 -0.045599 -0.034194 -0.032356   \n",
              "3 -0.089063 -0.044642 -0.011595 -0.036656  0.012191  0.024991 -0.036038   \n",
              "4  0.005383 -0.044642 -0.036385  0.021872  0.003935  0.015596  0.008142   \n",
              "\n",
              "          7         8         9  \n",
              "0 -0.002592  0.019908 -0.017646  \n",
              "1 -0.039493 -0.068330 -0.092204  \n",
              "2 -0.002592  0.002864 -0.025930  \n",
              "3  0.034309  0.022692 -0.009362  \n",
              "4 -0.002592 -0.031991 -0.046641  "
            ]
          },
          "execution_count": 2,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "693yQxfjZbN1",
        "outputId": "7a5a9729-b101-47e1-beb5-c31e59bf9427"
      },
      "source": [
        "# random forests\n",
        "rf_model = RandomForestRegressor(\n",
        "    n_estimators=100, max_depth=1, random_state=0, n_jobs=4)\n",
        "\n",
        "# hyperparameter space\n",
        "rf_param_grid = dict(\n",
        "    n_estimators=[10, 20, 50, 100, 200, 500, 1000, 2000],\n",
        "    max_depth=[1, 2, 3, 4],\n",
        ")\n",
        "\n",
        "# search\n",
        "reg = GridSearchCV(rf_model, rf_param_grid,\n",
        "                   scoring='neg_mean_squared_error', cv=5)\n",
        "\n",
        "search = reg.fit(X, y)\n",
        "\n",
        "# best hyperparameters\n",
        "search.best_params_"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'max_depth': 4, 'n_estimators': 2000}"
            ]
          },
          "execution_count": 3,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Qdeqf6YFZbN4",
        "outputId": "24d0fd27-3be6-4d25-fa2e-a779231082ee"
      },
      "source": [
        "results = pd.DataFrame(search.cv_results_)[['params', 'mean_test_score']]\n",
        "\n",
        "results.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "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>params</th>\n",
              "      <th>mean_test_score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>{'max_depth': 1, 'n_estimators': 10}</td>\n",
              "      <td>-3891.305592</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>{'max_depth': 1, 'n_estimators': 20}</td>\n",
              "      <td>-3893.071727</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>{'max_depth': 1, 'n_estimators': 50}</td>\n",
              "      <td>-3854.148676</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>{'max_depth': 1, 'n_estimators': 100}</td>\n",
              "      <td>-3874.187590</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>{'max_depth': 1, 'n_estimators': 200}</td>\n",
              "      <td>-3875.918730</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                  params  mean_test_score\n",
              "0   {'max_depth': 1, 'n_estimators': 10}     -3891.305592\n",
              "1   {'max_depth': 1, 'n_estimators': 20}     -3893.071727\n",
              "2   {'max_depth': 1, 'n_estimators': 50}     -3854.148676\n",
              "3  {'max_depth': 1, 'n_estimators': 100}     -3874.187590\n",
              "4  {'max_depth': 1, 'n_estimators': 200}     -3875.918730"
            ]
          },
          "execution_count": 4,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3vP3H08hZbN4",
        "outputId": "2f642a46-5dd6-4b60-e668-ceb5a0431d18"
      },
      "source": [
        "fig = plt.figure(figsize=(8, 8))\n",
        "\n",
        "# syntax for 3-D projection\n",
        "ax = plt.axes(projection='3d')\n",
        "\n",
        "# depth\n",
        "x = [1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2,\n",
        "     3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4]\n",
        "\n",
        "# number of trees\n",
        "y = [10, 20, 50, 100, 200, 500, 1000, 2000, 10, 20, 50, 100, 200, 500, 1000, 2000,\n",
        "     10, 20, 50, 100, 200, 500, 1000, 2000, 10, 20, 50, 100, 200, 500, 1000, 2000]\n",
        "\n",
        "# performance\n",
        "z = results['mean_test_score']\n",
        "\n",
        "# plotting\n",
        "ax.scatter(x, y, z,)\n",
        "ax.set_title('Response Surface')\n",
        "ax.set_xlabel('Tree depth')\n",
        "ax.set_ylabel('Number of trees')\n",
        "ax.set_zlabel('negative rmse')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0ejxTP6NZbN6"
      },
      "source": [
        "The more parameters we optimize, the more complex the Response surface "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rppOvecGaks8"
      },
      "source": [
        "### Low effective dimension\n",
        "\n",
        "Ψ(l) are more sensitive to changes in some dimensions\n",
        "\n",
        "Most parameters do not matter much"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bpuZhe1AZbdW",
        "outputId": "e295c52a-bc61-4f27-f4c3-29aad91ffe6d"
      },
      "source": [
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "from mpl_toolkits import mplot3d\n",
        "\n",
        "from sklearn.datasets import load_diabetes\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.model_selection import train_test_split, GridSearchCV\n",
        "\n",
        "\n",
        "# if you want more information about the dataset for this demo:\n",
        "\n",
        "# scikit-learn dataset\n",
        "# https://scikit-learn.org/stable/datasets/toy_dataset.html#diabetes-dataset\n",
        "\n",
        "# in short, regression problem, trying to predict disease progression based on\n",
        "# a bunch of variables\n",
        "\n",
        "# load dataset\n",
        "diabetes_X, diabetes_y = load_diabetes(return_X_y=True)\n",
        "X = pd.DataFrame(diabetes_X)\n",
        "y = diabetes_y\n",
        "\n",
        "X.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "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>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.038076</td>\n",
              "      <td>0.050680</td>\n",
              "      <td>0.061696</td>\n",
              "      <td>0.021872</td>\n",
              "      <td>-0.044223</td>\n",
              "      <td>-0.034821</td>\n",
              "      <td>-0.043401</td>\n",
              "      <td>-0.002592</td>\n",
              "      <td>0.019908</td>\n",
              "      <td>-0.017646</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>-0.001882</td>\n",
              "      <td>-0.044642</td>\n",
              "      <td>-0.051474</td>\n",
              "      <td>-0.026328</td>\n",
              "      <td>-0.008449</td>\n",
              "      <td>-0.019163</td>\n",
              "      <td>0.074412</td>\n",
              "      <td>-0.039493</td>\n",
              "      <td>-0.068330</td>\n",
              "      <td>-0.092204</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.085299</td>\n",
              "      <td>0.050680</td>\n",
              "      <td>0.044451</td>\n",
              "      <td>-0.005671</td>\n",
              "      <td>-0.045599</td>\n",
              "      <td>-0.034194</td>\n",
              "      <td>-0.032356</td>\n",
              "      <td>-0.002592</td>\n",
              "      <td>0.002864</td>\n",
              "      <td>-0.025930</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-0.089063</td>\n",
              "      <td>-0.044642</td>\n",
              "      <td>-0.011595</td>\n",
              "      <td>-0.036656</td>\n",
              "      <td>0.012191</td>\n",
              "      <td>0.024991</td>\n",
              "      <td>-0.036038</td>\n",
              "      <td>0.034309</td>\n",
              "      <td>0.022692</td>\n",
              "      <td>-0.009362</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.005383</td>\n",
              "      <td>-0.044642</td>\n",
              "      <td>-0.036385</td>\n",
              "      <td>0.021872</td>\n",
              "      <td>0.003935</td>\n",
              "      <td>0.015596</td>\n",
              "      <td>0.008142</td>\n",
              "      <td>-0.002592</td>\n",
              "      <td>-0.031991</td>\n",
              "      <td>-0.046641</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          0         1         2         3         4         5         6  \\\n",
              "0  0.038076  0.050680  0.061696  0.021872 -0.044223 -0.034821 -0.043401   \n",
              "1 -0.001882 -0.044642 -0.051474 -0.026328 -0.008449 -0.019163  0.074412   \n",
              "2  0.085299  0.050680  0.044451 -0.005671 -0.045599 -0.034194 -0.032356   \n",
              "3 -0.089063 -0.044642 -0.011595 -0.036656  0.012191  0.024991 -0.036038   \n",
              "4  0.005383 -0.044642 -0.036385  0.021872  0.003935  0.015596  0.008142   \n",
              "\n",
              "          7         8         9  \n",
              "0 -0.002592  0.019908 -0.017646  \n",
              "1 -0.039493 -0.068330 -0.092204  \n",
              "2 -0.002592  0.002864 -0.025930  \n",
              "3  0.034309  0.022692 -0.009362  \n",
              "4 -0.002592 -0.031991 -0.046641  "
            ]
          },
          "execution_count": 2,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6fNhxU4oZbdY",
        "outputId": "8e98c9ce-9ae2-44f7-a0fe-49d1ab2af649"
      },
      "source": [
        "# random forests\n",
        "rf_model = RandomForestRegressor(\n",
        "    n_estimators=100, max_depth=3, random_state=0, n_jobs=4)\n",
        "\n",
        "# hyperparameter space\n",
        "rf_param_grid = dict(\n",
        "    n_estimators=[2, 5, 10, 20, 50, 100, 200, 300],\n",
        "#     max_depth=[1, 2, 3, 4],\n",
        "#     min_samples_split=[2,4,8,12,24]\n",
        ")\n",
        "\n",
        "# search\n",
        "reg = GridSearchCV(rf_model, rf_param_grid,\n",
        "                   scoring='r2', cv=5)\n",
        "\n",
        "search = reg.fit(X, y)\n",
        "\n",
        "# best hyperparameters\n",
        "search.best_params_"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'n_estimators': 200}"
            ]
          },
          "execution_count": 3,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ri0Q1UEsZbdZ",
        "outputId": "bc561914-33c5-4e90-c16d-13e4b8659c85"
      },
      "source": [
        "# plot the resuls\n",
        "\n",
        "results = pd.DataFrame(search.cv_results_)[['params', 'mean_test_score', 'std_test_score']]\n",
        "results.index = rf_param_grid['n_estimators']\n",
        "\n",
        "results['mean_test_score'].plot(yerr=[results['std_test_score'], results['std_test_score']], subplots=True)\n",
        "\n",
        "plt.ylim(0.2, 0.6)\n",
        "plt.ylabel('Mean r2 score')\n",
        "plt.xlabel('Number of trees')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Number of trees')"
            ]
          },
          "execution_count": 4,
          "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"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rYXjk6KtZbda",
        "outputId": "475fb9d7-9e0f-4685-9fe3-4cb00c2dab6a"
      },
      "source": [
        "# random forests\n",
        "rf_model = RandomForestRegressor(\n",
        "    n_estimators=200, max_depth=3, random_state=0, n_jobs=4)\n",
        "\n",
        "# hyperparameter space\n",
        "rf_param_grid = dict(\n",
        "#     n_estimators=[2, 5, 10, 20, 50, 100, 200],\n",
        "    max_depth=[1, 2, 3, 4, 5, 6],\n",
        "#     min_samples_split=[2,4,8,12,24]\n",
        ")\n",
        "\n",
        "# search\n",
        "reg = GridSearchCV(rf_model, rf_param_grid,\n",
        "                   scoring='r2', cv=5)\n",
        "\n",
        "search = reg.fit(X, y)\n",
        "\n",
        "# best hyperparameters\n",
        "search.best_params_"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'max_depth': 3}"
            ]
          },
          "execution_count": 5,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-HhM-O9iZbda",
        "outputId": "1737f69f-6a2d-4cc3-c880-26da2ea5aaca"
      },
      "source": [
        "results = pd.DataFrame(search.cv_results_)[['params', 'mean_test_score', 'std_test_score']]\n",
        "results.index = rf_param_grid['max_depth']\n",
        "\n",
        "results['mean_test_score'].plot(yerr=[results['std_test_score'], results['std_test_score']], subplots=True)\n",
        "\n",
        "plt.ylim(0.2, 0.6)\n",
        "plt.ylabel('Mean r2 score')\n",
        "plt.xlabel('Tree depth')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Tree depth')"
            ]
          },
          "execution_count": 6,
          "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"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SX_uAv4bZbdb",
        "outputId": "00609175-bab3-47ba-ccd8-30303e489f38"
      },
      "source": [
        "# random forests\n",
        "rf_model = RandomForestRegressor(\n",
        "    n_estimators=200, max_depth=3, random_state=0, n_jobs=4)\n",
        "\n",
        "# hyperparameter space\n",
        "rf_param_grid = dict(\n",
        "#     n_estimators=[2, 5, 10, 20, 50, 100, 200],\n",
        "#     max_depth=[1, 2, 3, 4, 5, 6],\n",
        "    min_samples_split=[2,4,8,12,24,48]\n",
        ")\n",
        "\n",
        "# search\n",
        "reg = GridSearchCV(rf_model, rf_param_grid,\n",
        "                   scoring='r2', cv=5)\n",
        "\n",
        "search = reg.fit(X, y)\n",
        "\n",
        "# best hyperparameters\n",
        "search.best_params_"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'min_samples_split': 12}"
            ]
          },
          "execution_count": 7,
          "metadata": {},
          "output_type": "execute_result"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sOPjNgfJZbdb",
        "outputId": "09a141ee-894a-488e-e409-72dcf23f4cd1"
      },
      "source": [
        "results = pd.DataFrame(search.cv_results_)[['params', 'mean_test_score', 'std_test_score']]\n",
        "results.index = rf_param_grid['min_samples_split']\n",
        "\n",
        "results['mean_test_score'].plot(yerr=[results['std_test_score'], results['std_test_score']], subplots=True)\n",
        "\n",
        "plt.ylim(0.2, 0.6)\n",
        "plt.ylabel('Mean r2 score')\n",
        "plt.xlabel('Minimum samples at split')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Minimum samples at split')"
            ]
          },
          "execution_count": 8,
          "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"
        }
      ]
    }
  ]
}