bayesian_optimization_skopt.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UCMUpDF_gCya",
        "outputId": "e139a566-b4d6-4f75-95ab-e2641f6e0578"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting scikit-optimize\n",
            "  Downloading scikit_optimize-0.10.2-py2.py3-none-any.whl (107 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m107.8/107.8 kB\u001b[0m \u001b[31m1.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.10/dist-packages (from scikit-optimize) (1.4.2)\n",
            "Collecting pyaml>=16.9 (from scikit-optimize)\n",
            "  Downloading pyaml-24.4.0-py3-none-any.whl (24 kB)\n",
            "Requirement already satisfied: numpy>=1.20.3 in /usr/local/lib/python3.10/dist-packages (from scikit-optimize) (1.25.2)\n",
            "Requirement already satisfied: scipy>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from scikit-optimize) (1.11.4)\n",
            "Requirement already satisfied: scikit-learn>=1.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-optimize) (1.2.2)\n",
            "Requirement already satisfied: packaging>=21.3 in /usr/local/lib/python3.10/dist-packages (from scikit-optimize) (24.1)\n",
            "Requirement already satisfied: PyYAML in /usr/local/lib/python3.10/dist-packages (from pyaml>=16.9->scikit-optimize) (6.0.1)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=1.0.0->scikit-optimize) (3.5.0)\n",
            "Installing collected packages: pyaml, scikit-optimize\n",
            "Successfully installed pyaml-24.4.0 scikit-optimize-0.10.2\n"
          ]
        }
      ],
      "source": [
        "%matplotlib inline\n",
        "!pip install scikit-optimize"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YL2C-s-zgCyi"
      },
      "source": [
        "\n",
        "# Bayesian optimization with `skopt`\n",
        "\n",
        "Gilles Louppe, Manoj Kumar July 2016.\n",
        "Reformatted by Holger Nahrstaedt 2020\n",
        "\n",
        ".. currentmodule:: skopt\n",
        "\n",
        "## Problem statement\n",
        "\n",
        "We are interested in solving\n",
        "\n",
        "\\begin{align}x^* = arg \\min_x f(x)\\end{align}\n",
        "\n",
        "under the constraints that\n",
        "\n",
        "- $f$ is a black box for which no closed form is known\n",
        "  (nor its gradients);\n",
        "- $f$ is expensive to evaluate;\n",
        "- and evaluations of $y = f(x)$ may be noisy.\n",
        "\n",
        "**Disclaimer.** If you do not have these constraints, then there\n",
        "is certainly a better optimization algorithm than Bayesian optimization.\n",
        "\n",
        "This example uses :class:`plots.plot_gaussian_process` which is available\n",
        "since version 0.8.\n",
        "\n",
        "## Bayesian optimization loop\n",
        "\n",
        "For $t=1:T$:\n",
        "\n",
        "1. Given observations $(x_i, y_i=f(x_i))$ for $i=1:t$, build a\n",
        "   probabilistic model for the objective $f$. Integrate out all\n",
        "   possible true functions, using Gaussian process regression.\n",
        "\n",
        "2. optimize a cheap acquisition/utility function $u$ based on the\n",
        "   posterior distribution for sampling the next point.\n",
        "   $x_{t+1} = arg \\min_x u(x)$\n",
        "   Exploit uncertainty to balance exploration against exploitation.\n",
        "\n",
        "3. Sample the next observation $y_{t+1}$ at $x_{t+1}$.\n",
        "\n",
        "\n",
        "## Acquisition functions\n",
        "\n",
        "Acquisition functions $u(x)$ specify which sample $x$: should be\n",
        "tried next:\n",
        "\n",
        "- Expected improvement (default):\n",
        "  $-EI(x) = -\\mathbb{E} [f(x) - f(x_t^+)]$\n",
        "- Lower confidence bound: $LCB(x) = \\mu_{GP}(x) + \\kappa \\sigma_{GP}(x)$\n",
        "- Probability of improvement: $-PI(x) = -P(f(x) \\geq f(x_t^+) + \\kappa)$\n",
        "\n",
        "where $x_t^+$ is the best point observed so far.\n",
        "\n",
        "In most cases, acquisition functions provide knobs (e.g., $\\kappa$) for\n",
        "controlling the exploration-exploitation trade-off.\n",
        "- Search in regions where $\\mu_{GP}(x)$ is high (exploitation)\n",
        "- Probe regions where uncertainty $\\sigma_{GP}(x)$ is high (exploration)\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9ITrUgaagCyt",
        "outputId": "38ba8042-5b3b-4b27-dbe3-5908316c7d69"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Automatically created module for IPython interactive environment\n"
          ]
        }
      ],
      "source": [
        "print(__doc__)\n",
        "\n",
        "import numpy as np\n",
        "np.random.seed(237)\n",
        "import matplotlib.pyplot as plt\n",
        "from skopt.plots import plot_gaussian_process"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xviosX16gCyx"
      },
      "source": [
        "## Toy example\n",
        "\n",
        "Let assume the following noisy function $f$:\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "ywPeKl9rgCyy"
      },
      "outputs": [],
      "source": [
        "noise_level = 0.1\n",
        "\n",
        "def f(x, noise_level=noise_level):\n",
        "    return np.sin(5 * x[0]) * (1 - np.tanh(x[0] ** 2))\\\n",
        "           + np.random.randn() * noise_level"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KqMoqK6kgCy0"
      },
      "source": [
        "**Note.** In `skopt`, functions $f$ are assumed to take as input a 1D\n",
        "vector $x$: represented as an array-like and to return a scalar\n",
        "$f(x)$:.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 430
        },
        "id": "vNnR-9wOgCy3",
        "outputId": "85ca80cd-4cd9-46ec-a031-509d2897565c"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAi8AAAGdCAYAAADaPpOnAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABuX0lEQVR4nO3dd5hTZfo38G8yyWR6AYYpMMDQi3QUwQIqzY7rZcVFXRcr6yK6CquC6Cp2+am4uLuvoqvYdhULthEFVxlRmtJEep+hDNNbJjnvHw9nJuWctDkn5eT7ua5cGZKTk3NIZnLnfu7nfkySJEkgIiIiihHmSB8AERERUTAYvBAREVFMYfBCREREMYXBCxEREcUUBi9EREQUUxi8EBERUUxh8EJEREQxhcELERERxRRLpA9Aa06nE4cOHUJ6ejpMJlOkD4eIiIgCIEkSqqurUVBQALPZd27FcMHLoUOHUFhYGOnDICIiohDs378fnTt39rmN4YKX9PR0AOLkMzIyNN233W7Hl19+iQkTJsBqtWq672hg9PMDjH+OPL/YZ/Rz5PnFPr3OsaqqCoWFhS2f474YLniRh4oyMjJ0CV5SUlKQkZFhyDel0c8PMP458vxin9HPkecX+/Q+x0BKPliwS0RERDGFwQsRERHFFAYvREREFFMMV/NCRETBcTgcsNvtYXkuu90Oi8WChoYGOByOsDxnOBn9/IC2naPVakVCQkKbj4HBCxFRHKupqcGBAwcgSVJYnk+SJOTl5WH//v2G7MVl9PMD2naOJpMJnTt3RlpaWpuOgcELEVGccjgcOHDgAFJSUpCTkxOWD1un04mamhqkpaX5bUQWi4x+fkDo5yhJEo4ePYoDBw6gV69ebcrAMHghIopTdrsdkiQhJycHycnJYXlOp9OJpqYmJCUlGfLD3ejnB7TtHHNycrBnzx7Y7fY2BS/G/J8lIqKAGXV4g6KPVu81Bi9EREQUUxi8EBERUUxh8EJERBSibdu2IS8vD9XV1Zrut1u3bliwYIGm+9Tb559/jiFDhsDpdOr+XAxeiIgoZphMJp+Xhx56KKzHM3v2bPzpT38KaDFBo5s0aRKsVivefPNN3Z+LwQsREcWMw4cPt1wWLFiAjIwMt9vuueeelm0lSUJzc7Nux7Jv3z588sknuOGGG3R7jlhzww034Pnnn9f9eRi8EBEFqq4OKC0FysqAhoZIH41+amvVL57n7Wvb+vrAtg1CXl5eyyUzMxMmk6nl37/++ivS09Px2WefYfjw4bDZbPjuu+9www03YPLkyW77mTFjBsaOHdvyb6fTifnz56OoqAjJyckYPHgw/vOf//g8lnfffReDBw9Gp06dWm6bN28ezjrrLLftFixYgG7durX8Wz6ep59+Gvn5+Wjfvj3uuOMOn12O//WvfyErKwvLly8HAIwdOxZ33nkn7r33XrRr1w55eXleWad9+/bh0ksvRVpaGjIyMnDllVeirKwMAFBZWYmEhASsWbOm5fzbtWuH008/veXxb7zxBgoLCwEAe/bsgclkwvvvv4/zzjsPBQUFGDp0KEpKStye8+KLL8aaNWuwc+dOn/93bcXghYjIH0kC9u0Dtm4FDh4EDhwANm8WPxtRWpr65fLL3bft2FF92/PPd9+2WzeYMzKQ1bkzzBkZrdtpbNasWXj88cexdetWDBo0KKDHzJ8/H6+//joWLVqEzZs346677sJ1112HlStXqj7mf//7H0aMGBHSMX7zzTfYuXMnvvnmG7z22mtYvHgxFi9erLjtk08+iVmzZuHLL7/Eeeed13L7a6+9htTUVKxevRpPPvkkHn74YRQXFwMQwcill16K8vJyrFy5EsXFxdi1axeuuuoqAEBmZiaGDBmCFStWAAA2btwIk8mE9evXo6amBgCwcuVKjBkzxu1Y7r//fsycORPffvstevXqhWuuucYtu9WlSxfk5ubif//7X0j/L4Fi8EJE5M/u3cDRo963l5YCe/eG/3jIp4cffhjjx49Hjx490K5dO7/bNzY24rHHHsMrr7yCiRMnonv37rjhhhtw3XXX4eWXX1Z93N69e1FQUBDSMWZnZ+PFF19E3759cdFFF+HCCy9syaq4uu+++7BgwQKsXLkSp512mtt9gwYNwty5c9GrVy9MnToVI0aMaNnH8uXLsXHjRixZsgTDhw/HyJEj8frrr2PlypX46aefAIjsjRy8rFixAuPHj0e/fv3w3XfftdzmGbzcc889uPDCC9GzZ0889NBD2Lt3L3bs2OG2TUFBAfbq/HvBDrtERL6UlgInTqjff+wYkJQE5OaG75j0dvKbtyLPrqhHjqhv69l9dc8eOJ1OVFVVISMjQ7cOtMFmQ3bs2IG6ujqMHz/e7fampiYMHTpU9XH19fVISkoK6RgHDBjg1mE2Pz8fGzdudNvmmWeeQW1tLdasWYPu3bt77cMzq5Sfn48jJ1+PrVu3orCwsGXYBwD69++PrKwsbN26FaeeeirGjBmD//f//h8cDgdWrlyJCRMmIC8vDytWrMCgQYOwY8cOt6E1z+fMz88HABw5cgR9+/ZtuT05ORl1dXVB/o8Eh8ELEZGa+nrg0CH/2x08CGRkAGFqsa+71FT9tnU6AYdD/KxT8JLqcUxms9lr4UnX+hJ5mGTZsmVu9SsAYLPZVJ+nQ4cOOOER2Pp7LpnVanX7t8lk8ppifNZZZ2HZsmV49913MWvWrJD24cvZZ5+N6upqrFu3Dt9++y0ee+wx5OXl4fHHH8fgwYNRUFCAXr16qT6n3C3X8znLy8uRk5MT8HGEgsELEZGafftEvYs/ck1Mnz76HxMFLScnB5s2bXK7bcOGDS0fxP3794fNZsO+ffu8hkl8GTp0KLZs2eJ2W4cOHXDkyBG3AGbDhg0hHfdpp52G6dOnY9KkSbBYLG4zqfzp168f9u/fj/3797dkX7Zs2YKKigr0798fAJCVlYVBgwbhxRdfhNVqRd++fdGxY0dcddVV+OSTT4L6v5A1NDRg586dPjNWWmDNCxGRkooK38MnnmpqgPJy3Q6HQnfuuedizZo1eP3117F9+3bMnTvXLZhJT0/HPffcg7vuuguvvfYadu7ciXXr1uGFF17Aa6+9prrfiRMnoqSkBA6Ho+W2sWPH4tixY3jqqaewc+dOLFy4EJ999lnIxz569Gh8+umnmDdvXlBN68aNG4eBAwdiypQpWLduHX788UdMnToVY8aMcRtWGzt2LN58882WQKVdu3bo168f3nnnnZCClx9++AE2mw2jRo0K+rHBYPBCRKTk8OHwPIZ0N3HiRDz44IO49957ceqpp6K6uhpTp0512+aRRx7Bgw8+iPnz56Nfv36YNGkSli1bhqKiItX9nn/++bBYLPjqq69abuvXrx+efvppvPTSSxg8eDB+/PHHoDImSs4880wsW7YMDzzwAF544YWAHmMymfDhhx8iOzsbZ599NsaNG4fu3bvjnXfecdtuzJgxcDgcbrUtY8eO9botUG+99RamTJmClJSUoB8bDJPkOTgX46qqqpCZmYnKykpkZGRoum+73Y5PP/0UF1xwgddYoxEY/fwA458jz08jlZWAxwyKgBUVAQHMcFETztewoaEBu3fvRlFRUciFp8EKR8FuOC1cuBAfffQRvvjiCwDGOz8laud47Ngx9OnTB2vWrFEN+ny954L5/GbNCxGRJ18zaPwpK2tT8EKx5ZZbbkFFRQWqq6vjfomAPXv24KWXXvKZrdIKgxciIleNjUBVVeiPr6sT9S86NF+j6GOxWHD//fdH+jCiwogRI0Ju2hcsY+a0iIhCdexY2/fRlswNEfmla/Dy7bff4uKLL0ZBQQFMJhOWLl3q9zErVqzAsGHDYLPZ0LNnT9V2yUREmpMk4Pjxtu+nogLQcUFAonina/BSW1uLwYMHY+HChQFtv3v3blx44YU455xzsGHDBsyYMQN//OMfWwqhiIh0VV0N+FgcL2BaBUFhYrB5GxTFtHqv6Vrzcv755+N8z4W5fFi0aBGKiorwzDPPAEDLGgvPPfccJk6cqNdhEhEJWvZpOX486pcMkNvTNzU1Idko3YEpqjU1NQGA29IIoYiqgt2SkhKMGzfO7baJEydixowZqo9pbGxEY2Njy7+rThba2e12n8uLh0Len9b7jRZGPz/A+OfI82sDSRLBSxDt1X2qrRWZnCCnIIfzNZQkCUlJSThy5AgSEhLCMrVXkiQ0NTWhvr6+pb28kRj9/IDQz9HpdOLIkSNISkqCJEle7/Fg3vNRFbyUlpYi1+ObSm5uLqqqqlBfX6/4zWD+/PmYN2+e1+1ffvmlbk1y5CXHjcro5wcY/xx5flEikHWRVITrHM1mM3Jyclq++BHpyW634+jRo/jll1+87gtmMceoCl5CMXv2bMycObPl31VVVSgsLMSECRN0aVJXXFyM8ePHG7YBmJHPDzD+OfL82mDfPu3b+ycmAifXkQlUJF5Dp9MJu90eltqX5uZmrFq1CqNHj4bFEvMfQV6Mfn5A6OdoMplgtVpVM3zBBNBR9T+bl5eHsrIyt9vKysqQkZGhOh5rs9kUV/20Wq26/eLrue9oYPTzA4x/jjy/ENTUaL/KcXOzKAAOIQsc7tfQ1+rJWrLb7WhubkZaWpoh36NGPz9Av3MMZl9R1edl1KhRWL58udttxcXFui/wRERxrrZWv6nNJ07os1+iOKZr8FJTU4MNGza0LAe+e/dubNiwAfv27QMghnxcF8e69dZbsWvXLtx777349ddf8dJLL+Hdd9/FXXfdpedhElG8q6iIzX0TxSldg5c1a9Zg6NChGDp0KABg5syZGDp0KObMmQMAOHz4cEsgAwBFRUVYtmwZiouLMXjwYDzzzDP417/+xWnSRKQvPYtVGxrEhYg0o2vNy9ixY30WgCl1zx07dizWr1+v41EREbmw28V6RHqqrAx6yjQRqYuqmhciorCrrtb/OTh0RKQpBi9EFN/C0d9Ez4JgojjE4IWI4ls4ghdJCk+GhyhOMHghovjV0KDNQoyBqKwMz/MQxQEGL0QUv8KZDWH7fSLNMHghovgVzuDFbgfq68P3fEQGxuCFiOJXTU3g2x45ArzzTtuej9kXIk0weCGi+BRMvcsXXwBXXAE89ZQIYmSffQZ8913gz8miXSJNMHghovgUaNblk0+ABx4Q050HDGgd+tm7F3j0UeDuu4EVKwLbV3W1mHlERG3C4IWI4lMgwcumTcDf/iYCjiuuAF55BejaVdyXnw+ccw7gcAB//SuwbZv//TmdIggiojZh8EJE8clf8NLQANx/v2gud+65wL33AgkJrfcnJgJz5wJnngk0NYltA1nDiHUvRG3G4IWI4o/dDjQ2+t5m8WLg4EEgNxeYMwcwmby3sViAhx4CcnKAPXvEY/wJpkiYiBQxeCGi+ONv6EaSgLIyEbDcfTeQlqa+bVYWcM894ufXXwcOHPC975oaMXxERCFj8EJE8cdf8GIyiSGhpUtFXYs/554LnHaaGD5atsz3tpLEuheiNrJE+gCIiMIu0KGbTp0C285kAmbMAEpLgbPO8r99dTWQnh7YvonICzMvRBRfJAmoq1O///vv/Q/9KOndGzj7bOXaGE+seyFqEwYvRBRf6uvVa06amkQB7mWXAevXh/4cDQ2+lwKorWW/F6I2YPBCRPHFV73J118DJ04AHTsCAweGtv///he48ELfSwmw3wtRmzB4IaL44ito+OgjcX3JJWIadCisVqCyUuzLV3aFQ0dEIWPwQkTxRa3epawM+Okn8fNFF4W+/3HjgORkYN8+YONG9e0YvBCFjMELEcUPp1O9C+5nn4lMybBhgc8yUpKSIqZOA2JdJDUMXohCxuCFiOJHXZ36UM5XX4nr889v+/NceKG4Li5W7+TrcPgu6iUiVQxeiCh+qNW7VFSIVaITEgJrSufPiBFiWYHqauB//1PfjtkXopAweCGi+KFW75KVJbIkf/+7+LmtzGbgggvEz76GjjjjiCgk7LBLRPHDV3O6pCRR76KViy8WDet8ZXKYeSEKCYMXIooPasW6khRYV9xgdekC3H67720aG8UK11ar9s9PZGAcNiKi+KCWdfnqK+Dqq4G33w7v8cg4dEQUNAYvRBQf1IKX778HduwQiypqzeEAVq4EHntMLD2ghENHREFj8EJE8UEpeHE6gZIS8fPo0do/p8kEzJ8PvP8+sHat8jbMvBAFjcELEcUHpeDlt9+A48dFR9whQ7R/TrNZrDQNACtWqB8XF2kkCgqDFyIyPklSLtb9/ntxfeqpQGKiPs89Zoy4/vZb5dWsnU42qyMKEoMXIjK++nrl7MaqVeJajyEj2amniiUDjh4Ftm1T3oZ1L0RBYfBCRManNGRUVwds2iR+HjVKv+dOTBQddwFg9WrlbVj3QhQUBi9EZHxKwzK1tcCECcDgwW1biDEQp58urn/4Qfl+Bi9EQWGTOiIyPqXMS04O8Mgj4Xl+OXg5cUJMn05IcL+/sRFobgYs/JNMFAj+phCR8UW6ILawUKxxlJenvk1tLZCZGb5jIophHDYiImNrahLZDs/btm9Xnv2jB5PJd+ACcOiIKAjMvBCRsSllXTZtAm6+GejeHXj33cD2Y7UCBQVARob494kTwOHD3oGRP2rDQwxeiALGzAsRGZtSvcu6deK6R4/A9pGcDPTvD3ToIGYPJSYCubniNpstsH04HMDddwPnnguUlQV2nESkiMELERmbUuZFDl6GDfP/+MREoHdv5WyJfF8gq0InJADHjokgRWnKdHOzKNwlIr8YvBCRsXkGL83NwC+/iJ+HDvX9WJMJKCryPQsoMVFsE4hTTxXXcvDkiUNHRAFh8EJExuV0emcztm0TSwVkZIiaF186dADS0vw/T3q6mHrtj5zpYfBC1CYMXojIuBoavJcFkLvqnnKKWDhRjcUSXPO6Tp3892kZPFgMHx06BJSWet/PuheigDB4ISLjUptpBIjgxZfcXO9mcr4kJAD5+b63SU0F+vQRP69d630/V5gmCgiDFyIyLqXg5bLLgGnTgDPOUH+cxRLYMJCnDh38F+8OHy6ulYaOnE7l1a+JyA37vBCRcSkFL8OG+Z9l1KFDcFkXmdksMjYHDqhvM3Ik8OuvQL9+yvdz6IjILwYvRGRcoSwLYDKFlnWRdeggalrUuveefnrrWkdKGLwQ+cVhIyIyJocDsNvdb/vpJ2DlSqCiQv1xmZli+nOoEhJEABMqBi9EfjF4ISJjUsq6vPGG6HL7+efqj2vfvu3PHUjwcuKEGD7yxJoXIr8YvBCRMXkGL5Lkf6aRxaLNys7JyWJmkZqSEmD8eGDuXO/7ONuIyC8GL0RkTJ4ZjIMHgcpKMRuod2/lx7RrJ2petOAr+9Krl7jetQuoqdHm+YjiCIMXIjImz8yLnHXp3Vu9piU7W7vnz85Wb4LXoYNoaueaDSKigDF4ISJj8sy8bN0qrvv3V97eag1sKYBAJSSIJQjUDBokrn/+WbvnJIoTDF6IyHiam71nGsnFsX37Kj9Gy6yLrF079fsGDhTXGzcq3+9waH88RAbB4IWIjMcz6yJJYkFGQD14ycrS/jgyM9WHjuTMy8aNyj1hQulRQxQn2KSOiIxH6YP/lVdEAKO0knRCgrZDRjKzWQQwJ05439ezp5iVVFsrCnd79nS/v65On2wQkQEweCEi4/HMvJhMImhRClwAEWBoNcvIU3a2cvBisQC33CIyPkodfZl5IVLF4IWIjCfYD34terv42rfJpNy/5brr1B/HTrtEqhi8EJHxeGZe/v1vEUCMHy8WTvTka1ZQW5nNQHo6UFUV3OMaG0XRbigLRBIZHAt2ichYlNY0euMNYMECoKzMe/uUFDGEoydfxcCbNwNLlogGep6YfSFSxMwLERmLZ9bl2DHg+HGRAZE727rSM+si8zUsNWcOsHcvUFgInHWW+321tSJrQ0RumHkhImPxrHeR+7t07Spm93gKR/CSmKj83EDrlOlffvG+j5kXIkUMXojIWDwzL776u5jN+kyRVqIWJPkKXmpr9TseohjG4IUoGE6n+DZcUyMKKin6qAUvffp4b5uaqt8UaU9qQ0dy8LJ5s+gM7Kqpyfs2ImLNC1FA6uqA0lJRVOnaDdVmAzp2FH06wvUhSL55Bi87dohrpZWkw1lPkpYmMj2e3XSLikQQ5atZXTiGtohiSFgyLwsXLkS3bt2QlJSEkSNH4scff1TddvHixTCZTG6XpKSkcBwmkTenUxRTbt0qGo15fvA0NgL794v7mYmJPKfT/XVoaAAOHhQ/ewYFQHiDF5NJ+fnM5tbFIjdv9r6fQ0dEXnQPXt555x3MnDkTc+fOxbp16zB48GBMnDgRR44cUX1MRkYGDh8+3HLZu3ev3odJ5K2xUQQlx47537a+XhSGsitqZHlmXZKSgBUrRJ8Xz0USzWaR8QgntWBJDl62bPG+j0W7RF50D16effZZTJs2DTfeeCP69++PRYsWISUlBa+88orqY0wmE/Ly8louuUpNpYj0VFcnghHPD0NfmpuB335jBiaSlF6v5GSgXz/v28NZ7yJTG/655BLgn/8EZs70vo/BC5EXXWtempqasHbtWsyePbvlNrPZjHHjxqGkpET1cTU1NejatSucTieGDRuGxx57DAMGDFDctrGxEY0uHxZVJ7tY2u122D0bVbWRvD+t9xstjH5+QIDnWF8v6iQcjuCfoKlJBDC9e6uvJqwjo7+Gfs+vtlZ5hWYlycnezez0ZrGIjrmez1tYKC4A7CePX75GQ4N4T+rdSC9M4v49agB6nWMw+zNJktKCG9o4dOgQOnXqhFWrVmHUqFEtt997771YuXIlVq9e7fWYkpISbN++HYMGDUJlZSWefvppfPvtt9i8eTM6d+7stf1DDz2EefPmed2+ZMkSpKSkaHtCRBRTTvnXv2Bubsauiy5CjcLfDyKKHnV1dbj22mtRWVmJDD9F6lEXyo8aNcot0Bk9ejT69euHl19+GY888ojX9rNnz8ZMl1RrVVUVCgsLMWHCBL8nHyy73Y7i4mKMHz8eVqtV031HA6OfH+DnHB0OYPv24IaKfOnZM3w9RE4y+mvo9/xch/okCZb//Q+mykp0vvZaoKCgdTuTCTjllMisG3T8uCjy9mBauxamFSvQPHgwPu/fH+Pz8mCVs3f5+cprMsWguH+PGoBe51gVxPpfugYvHTp0QEJCAso81hMpKytDXl5eQPuwWq0YOnQodsjTHT3YbDbYbDbFx+n1xtFz39HA6OcHKJyjJAF79ohhH62Gew4eBAYMiMgUaqO/hqrn19zc+vodPSqmtpvNsHbv7v66JieLYt5IyM5unQHl6uefgXfegamyEujfH1azuTV4aWgADPZ6xu171EC0Psdg9qXroHxiYiKGDx+O5cuXt9zmdDqxfPlyt+yKLw6HAxs3bkR+fr5eh0kEHDoU/Kq//jQ2Aj5m1ZHGGhvd613kLzxdungHKmHOiLmx2cRyAZ5OzjgyKU2XZtEukRvdKwpnzpyJf/7zn3jttdewdetW3HbbbaitrcWNN94IAJg6dapbQe/DDz+ML7/8Ert27cK6detw3XXXYe/evfjjH/+o96FSvKqsFA3o9HD4MDukhovncN/27eJaaTHGcE+R9qQUPMnBy/79sNTUuN9nt4e/uJgoiule83LVVVfh6NGjmDNnDkpLSzFkyBB8/vnnLdOf9+3bB7NLOvfEiROYNm0aSktLkZ2djeHDh2PVqlXoL/dBINKS3S6Gi/TicIjsi2u9BelDrbNujx7e20Y6eElPB8rL3W/LygI6dQIOHkT2zp3eHYFra8U2RBSegt3p06dj+vTpivetWLHC7d/PPfccnnvuuTAcFRFE4KJ3ZuTIEVFsGYni0HiiFrx4Zl4slsjVu8jUhq369wcOHkTW9u3A+ee731dXx+CF6CQuzEjx68gR7etclDgconiU9OUavDidrW31PZcFiHTWBRDBk1Jx4sl+VtnykJcr1r0QtYi6qdJEYdHUpDzjQy9y9oWLN+rHNXgxm4EPPwSqq72DlWjp/5SWJtbLcnUyeEnyHFICuMYRkQsGLxSf9u4NvBOrFux2UePQvn34njOeOBzKw39KawlFQ+YFUA1e7MuW4Vu7HRd4bt/cLIJupZlKRHGGw0YUnyLxLZZDR/oJprFgNAUvnhITgZwc9cdw6IgIAIMXijdNTZF77tparjqtF8/g5ZlngNtvBzyXIElMjJ41gpKTg2+IyKEjIgAMXije7NsX2edn9kUfnsHL2rXAjz963x4tWRdA1D8pZV+2bsXIv/0NCQ8+6H0fgxciAAxeKJ4cOwZ4Nv8Kt/JysRQBacs1SHE4RE0TAHTv7r5dtBTrylSCqbw1a2AqKfF+r3DYiAgAgxeKF3Y7cOBApI9CfLBWVET6KIzHNXg5fFgsFWCzeTcHjLbgRSnz0rMnHFYrTFVV3u9Zh0OcG1GcY/BC8WH/fvGHPxocPx7pIzAWSXL/QN+1S1x37erdGDDaghelzIvVisqiIvGz0jpHHDoiYvBCcaCy0ntKaiRVVXG9Iy01NbkPr8jBi+eQUTQV68oSEkThrocKuSswgxciRQxeyNgcjsgX6XqSJA4dacmzKFcOXuTshSzasi4yhaGjE3JX4K1bvbdn3QsRgxcyuEOHIjs9Wo1SB1UKjWfwYrMBmZnRX6wrUxg6qpCDl19/9R7urKtj0TfFPQYvZFy1taItfzSqqeHQkVY8C1jvvx9YvhwYM8b99mgNXhQyLzUFBZA6dBALNVZWut/pdAbXlI/IgKJsAJhII5LUOl02GslDRx06RPpIYp/aB7lnA7hoDV5sNlGL4xrMJiSgedkyWNVWIq+tVayVIYoXzLyQMZWWRn8322gqIo5lrsGL2nCKxaK8inO0UJp15GsRTxbtUpxj8ELGU18ven1Eu+rq6Jm+HascDtHDR/bmm8CllwKLF7tvF61ZF5mvzr9KmSUW7VKc47ARGYs8XKRXQaMkiQ/I+nqgXTtg0CCgZ0/f35J97auqCsjO1v4444VnvcvOncDBg+4BDRCbwUtFBTBtmsgirljhnjmqrxe1L8GujURkEAxeyFjKyrRPqUtSa3BiMgHvvitmMcl69wbuuAM444zg911RweClLTyzErt3i2vPadLRXh+iFLxkZoqhxcZGYMcOoF+/1vskSWRflDr0EsUBhu1kHA0N2g8Xffwx8Ic/uH+T/93vgMsuA04/XTQ+++034M9/Bv72t+BngVRWctprW3jWu8jBS6xMk5YlJABJSe63mUytAcuWLd6PYd0LxTFmXsgYJAnYs0ek0rXQ3Aw88wzw3nvi38uWAZMni59vuKF1u8pK4JVXgCVLgKVLxTfhGTMCfx6HQ3wI8Rt0aFyHjeSsW0ICUFjYervJJGb0RLvUVO/gt39/YPVqEbxcfrn7fQxeKI4xeCFjKC3V7o95fT1w993Ajz+Kf99yC3DxxcrbZmYCd90FjBoF/PvfwB//GPzzVVYyeAmV64e9nHXp0sW9PiQ5ObSapHBLTfVe96p/f3Gt1GmXwQvFMQYvFPvq6rQbLmpoaA1cUlKARx7xbnam5PTTgZEjQ/uQrKwEOnUK/nHknnnZs0dcd+vmvk2017vIlOpe5GGjnTvFe9N1aKmpSWQIo229JqIwYM0LxTanU3zj1qJuxG4H7ruvNXB58cXAAheZHLhIEvDf/4oZIoGor/eeHUP+2e3uU83T04FTTnEvbAViJ3hJTvaePZSbC7RvL87zt9+8H8PsC8UphuwU2/bv165V+p49wM8/i/qIBQvENOhQLFsGzJ8vhpSGDAGysvw/prKS3XaD5fm6X3SRuHiK9mJdmckkjrWqyv22ceNEdlHpPGprxfuMKM4weKHYVV4OHDum3f569RLNzUpLgWHDQt/PxInAG2+I6a0LFgAPPeT/MdXVDF6C5dnjRU2sZF4A7+AFAP7yF/Xta2r0PR6iKMVhI4pNDQ3arV3kOuTUrZuoX2kLqxX461/Ft+ZPPgE2bfL/GM8PLPLPNfPicCgvdGm1xlZNiK9Ou0rYaZfiFIMXij1OJ7BrlzbToisqgOuvB374oe37cjVoUOsQxoIF/mtympv5QRQs1+Bl+3bRJPDGG923iaWsC6AevDQ3A9u2eQ+VORzRv4YXkQ4YvFDs2bNHmz/YkgQ8+qjoofH008rf3Nvi1ltF/cyGDYEFR8y+BMd12GjfPvFB7jnbK9aCF5tN9KnxdOWVwJQpwObN3vexaJfiEIMXii2HDmm3GvP77wPffCOGFf72N+2HF3JzRTdeAHjtNf/bV1dr+/xGJknuwYs8hNi1q/t2sRa8AMqFuT16iGv2eyECwOCFYkl5uXb9XHbtAp59Vvw8fTrQt682+/U0ZQowerRYYsCfmhouFRAou939/2rfPnHtGbx4ttyPBUpDR3KzOqXMC4t2KQ7FUCUbxbXq6tYmZG1kam6GZc4c8c399NOBa6/VZL+K8vKA558PbFunU3wQpafrdzxG4Vn7YaTMi9Ix++q029AghsyUhpuIDIqZF4p+dXWiw6hGWYmeH3wA02+/if4YDz3k3Rgskjh0FBjXISNJag1eunRpvd1mi67XNlBKw0Zy470DB0RPIE/MvlCcicHfbIor9fViJolrJ9W2kCSkHTokfp45M3y9VUpLRcfef/7T93YMXgLjGrwcPy7qPsxm9wUZY3HICHBfl0mWkdF6bqx7IWLwQlGsrk60RNdyFpDJhPV//jOa//EP4IILtNuvP7t3iwZ4S5b47ghcW6vdythG5vp/2NwMTJoEnHUWkJjYenssDhn5Imdftmzxvo+ZF4ozrHmh6FRTIzrUapVx8SANGRLelYZPO03Uv5SWijWPJk1SOTBJBDCse/Gtqan157w8MVvMk9GCl0mTgJ49RQG4p9pa8d6JhdWziTTAzAtFn/JykXHRMnA5fBh48EFtlxMIRkICcPHF4ucPP/S9LYeO/HMNXtTE6rCRmrPPFrPWlGbGOZ1sVqc1SRKz2hoaRBZYvtTXi2HL5mbODowgZl4oekiS6ONSWqr9fufPB1atQkJ1NXDPPdruP1CXXAL861/ATz+JwsvOnZW34xBAcI4fF8XXnn16jBa8+FNTEzuLUEYTSRJBSW2tuG5oaA1OApGQIN57Vqu4yLO+KivF65GYGFtLVMQI/o9SdGhqEnUhenxwf/YZsGoVYLXCceed2u8/UPn5Ymp2SYnIvtxxh/J2HAIIzrRpIrO2aBEweLC4LVZnGrlKTPT+AC0rAzZuBLp3FxdXNTVAx47hO75YJkliaZATJ0Rn67ZkeR0OcZGLyOWatd27W9+DZrN4PeWLzeb+b6uVv+9BYvBCkXfsmMhE6FHfUl4OPPOM+HnaNLHwojzbKBIuuUQEL59/Dtx+u/IfLKdTBDBpaeE/vljT3AwcPCjeO3l5rbcbod5FaYXpl14Cli0T7+VbbnG/jxk7/+x24MgR8TdH6+VAfHE6RUbHV7G+nLnxvFgs3hcGOgxegiJH1OXl3mlAzzeT/G/Xa18Xs1lc5J/j4c1ZWwvs36/vNM+nnhLp2969galT9XueQJ11FtCpEzBypPhDpvYhW1PD4CUQcuCSlATk5LTeboQho+Rk7+Clf38RvCjNOLLbxbd/my08xxdLmptFdu7o0eitU7HbxSUQZrMYnpIv8r/lzxHXi+tnjOtnDuD9s+u1K8/bouD/kMFLMOTMwL59+qekTSb3N6XnxTMSlyN0pR4R0aamRtS1KDXb0tLKlUBxsfg/fPBB8f8T6WnISUnA0qX+g1N+iw6Ma3M6199JI2RefC0TsHWr8tBiTQ2DF09Hjohsq04zFyPC6RSXQIMdrUXBe4zBS7SSpNDSmiZTaxAjj6W6jq1Gany1uVmMLx87Jori9CZJrYshTpnS2iMjGgTyf8/gJTBqywIYJfPiqVcv8eWlvFzUv7gOlQHifdO+fXiOL9o1NoolRfi7ZEgMXoxGnt5nt6sHCSaTCGA8i8bkb65aZCccDvH8NTUi9S0XoYaLyQS88ALwxhvA9deH73kD5XQCv/wiOvwqzTpyOMSUTCNkELTkGdAbaUFGTwkJ4nfUtZtwUpJYYfq338TQkWfwwmn2wokTIrA1UraF3DB4iUeSJGb3ePbKkIOWX35RLhhzHVOVsweSJB7ncIgPlqYm8cc2kD4cektN9S5qjBZPPAH897/AddcBM2Yob1NTw+DFk+f7SinzYoSZRrLUVPfgBRBDR3Lwcu657vc1NoovLrEwfKyXAwdEVooMzSC/4aQ5h0MUlFZXi28xR4+KOpVDh8Qfh/37xeXAAXFbWZnot1FdHdnApb4e+OSTyNe2+HP66eL6q6/UM1JMd3vznK1x9tnA+PFiOEVmhKyLTKlvi1z3olS0C8Rv9sXpFAu4MnCJC8y8kLH8/e9i/aDVq4FHHon00agbNUpkCEpLxcKTvXt7b8PgxZtnFuK667y3MVLwolS0O3q0aLooBzGeamqAdu30Pa5o43CIwCVeA7c4xMwLGcfGjcBbb4mf1dYOihZJSWK9IwD47jvlbZqaIjebIFp5Bi9KjBS8KA0b5uWJbFOnTsqPibcPcIdDfAGIt/OOcwxeyBiamkSmRZLEatFnnKH9c1itopdI165AUZEotM3ICH3m1plnimu14AXQtwdOLHIdNjp2TPTu8BwiNFKdUEJC8MFYQ0P8BL1Op1jAlb8ncYfBCxnD3/8O7Nol0uUzZ2q7b4tFBCwDB4p+Ih06iOfJzRW1Fv37h7YKtBy8bNwo6oqUcOjInWs91X/+Ixa7fPxx922MlHkBlOte9u4FFi8GPvpI+THxkIWQJDFUxN+RuMTghWLf2rViSjQA3H8/kJWl3b4zM4EBA0TAopZhSUoSNSue01b9yc0Vj5MksWSAEv5hbtXU5J5lkWcaFRa23ua6MJ5RKNW9bN4MvPgi8MEHyo+Jh+Bl717vDsQUN1iwS7GtuRn4299EADB5MjBmjHb7zstTrytQ0qmTmKIbzNpJd94pgp+BA5Xvr6vjIo0yz3oXpWnSRsu6AMqZF/n98uuvIqhLTHS/3+jBy+HDYnYjxS1mXii2WSzAo4+KGpe77tJuv507Bxe4yPLzRZYmUKefDgwZop4tkKTwdCSOBa71Lk5na4O6Ll1abzdq8OIZvHbuLDKMdrvo+eIpWnot6eHEicgurkpRgcELxb7+/YH/+z/l9HooCgrEkE6ounRR/rYcKhYjCq7By9Gj4t8JCe4dio0YvJjN3udlMrVmXzZuVH6cEYdU6utFy3+KewxeKDbt26fepKstcnJE9qQtTCage/fAu7xu2iT6drz9tvL9rHsRXIaNTPKQUadO7iu8GzF4AZSD4VNOEde//KL8GKMFL3Ivl2hvQElhweCFYk9DA3DvvcBNNwErVmi334wM9+LPtrDZAg+Cdu0SSwV8/rny/cy8CC6ZF5OR1zRSohS8DBokrjdtUn6M0epe9uwJrM8PxQUGLxRbJElkKXbsEMGG/O2zrWw2kS3RsjA2NzewD1O5Wd2WLcofOGxW17oel/zPfv2AG28EzjuvdRuz2btw1SiUhkT79xfv1+PHgcpK7/ubm40T+B45AlRURPooKIpwthHFlqVLgWXLxAfVo48GVxyrxmwWgYvWU2xNJjGssXOn7+3y8kSdzL59Ytr32LHe29TWajsFPNY0NrqtASUNGOA9Q8uoWRdANN4zmdzXwUpNFUthdOumvhBjVZV2tWCRUl8v1lAjcsHMC8WO9euBJ58UP99+OzBihDb77dxZ2wJbV1lZgX14jBwprlevVr7fKN+gQ+W5IKMSIwcvSkW7gGiS6GsF6Vive3E6gd271RcvpbjF4IViw/79wD33iOGTc84Bpk7VZr9ZWaJIV08FBf63kYOXH39Uvj/egxfXYl27HaaffhJDCa4fakYOXoDQMii1taLQNVYdPCgyL0QeGLxQbPjgAzGu37+/WMMo0Jk8vlit3gWfesjI8P/BM3y4OKe9e8VK057ivdeLS+YlrbQUljvuAK64wn0bowcvStnB2loxfDp1qqhx8SRJsZt9qa4WASqRAta8UGyYPl206r/wQu0+pLp2dZ9mq6fcXDGrSE16uliGoKICKCvzXmrA4RDfQI206GAwXIOXgwfFD127uhdYx2PwkpwMFBeL6fQ7dgB9+3pvU1kJZGfrf3xacjjYz4V8YvBC0auhQQQXFovISlx/vXb7bt9eBEPhkpUlZsL46nq6cKHv2pva2vgNXlyGjdyCF1c2WxgPKALkTruuQ2Vms5hx98MPolmdWvASaw4cMG6HYNIEh40oOtXUAH/6EzB7tj5/xAKpQ9GSyQR07Oh7G39Fw/E6dORwuE0VT5Vbw7suC2CzaTOUGM1MJuXg1V+n3ebm2Gp0WFUFHDsW6aOgKGfw33aKSRUVwK23itlFP/3UuoaNliKx8nD79oH1kWluVg7Y4rVo12OmkWLmxehZF5lS7ZS/4AWInR4pTmfrgptEPnDYiKLLrl3AzJkibZydDbz4ItCzp3b7b9cucou6WSzinMrL1bd57jnRy+bee0V9j6v6evHH3egZBk8eXVVbghejL8ioRG2FaZNJzMg7dky591FlpfsaUNHq4MHoGS6qrhatC2prxe+ewyG+9FgsYgi4d+/WYTqnU1zCVUNH4QleFi5ciKeeegqlpaUYPHgwXnjhBZwmdxVV8N577+HBBx/Enj170KtXLzzxxBO44IILwnGoFEn/+x/wwAPij0VBAfD886IBl1YsFrFftXbq4dChg+/gxWwW579unXfwIknij2isNx0LlmvmpaICNrkLcTwGL0qvfXq6CPC3bwd+/tm967CsoSH6C75ra8M/u6i5WawNtX69yFydcw5w6aXivrIyYNYs9cdee21r8HLsmPh9bd8eCbm5ODU9HeZu3UShfm6u6MdTVKT76cQT3YOXd955BzNnzsSiRYswcuRILFiwABMnTsS2bdvQUaEGYNWqVbjmmmswf/58XHTRRViyZAkmT56MdevW4RStWsFT9HnjDWDBAvHzsGGiGZ3WHWULCyP/zSg93Xfh7vDhwL//LYIXJbW18R28WK1Yf8cdGGS3I8E1CxEvwUtSkghwPRcnHDJE/D/5WkaioiJ6gxdJCt/sIqcTWLMG+Phj4Pvv3aeS5+e3Bi8ZGeJvUXKyyHglJIjsS3OzyAb26tX6uKoqcQ7HjsF87BgKAFFELbvmGuDuu8XPFRUiu5yX1xrc5OW1XjIztV2mxKB0/0v+7LPPYtq0abjxxhsBAIsWLcKyZcvwyiuvYJZCVPt///d/mDRpEv7yl78AAB555BEUFxfjxRdfxKJFi/Q+XIqU004TfVd+9ztgxgzfXUNDkZkphoyiYY2g9u2Bw4eV7xsypHUI4OhR7wZ68Vj34jpslJqKfePH45SCArhVLcVL8GIyiQ9SzwLcu+/2H5ifONH2FdP1UloaWBfltjp6FPjjH8XwlCwzU/z9GTQIOPXU1ts7dgT+8Y/A9tu9O/Dll0BZGZoPH8aW7dsxoKEBCUePigyO69D34cMi26O2GviUKcBdd4mfy8uBF14Qw83Z2eILXXa2CKxSUkQmN06XDdE1eGlqasLatWsxe/bsltvMZjPGjRuHkpISxceUlJRg5syZbrdNnDgRS5cuVdy+sbERjS5/3KpORtF2ux12jT+o7EePYsgLL8CUkgKn2QzJbBZ/TE5eS4MGQZo4UWzc0ADzyy+73d/ys8UCqWdPSPIaNpIE05dfim/kiYnigzs5GVJamviWnZ4elm9M9pPf5uzhWHK+tBSmDRsgTZok/t2zp1hZWe5vouUxmM3ij7bLe0Lr90ZQMjLc/3i6SkmBpXdvmLZtQ/PatZAmTHC/v6bGZwAWFeentbq6lveD4ntULr42yDn7fQ1tNu/Gc0rZGE+1teL9E+HiZq/za2wUvw/hWAKgXTskZGfDVFEB56RJkCZMgDRwoHvgF+rfnqwsICsL9l69sLt3b/TMy4PVtT5N3m9eHkyPPQZTWRlQVuZ2bTp+HI4OHeCUtz18GNaPP1Z9Ssf118N5xx3iHwcPwvKHP4igJikJsFohWa3iM8VigXPcOEhyVqmyEgkLF0JS+nwymSANGQLp3HPFtnV1MC9eLH6W78/LA049VfvP2CD2p2vwcuzYMTgcDuTm5rrdnpubi19//VXxMaWlpYrblyp1HQUwf/58zJs3z+v2L7/8Eikar1eTdOwYJi5frnr/nspK/Hyy8t9aXY0L3nxTddv9Z5+Ndb17AxDtzi958EHVbUtHjMDqBx5o+feYmTPRnJSEpsxMNLpesrJQl5eHyu7dgz01N8Uq/9daSDt4ED2WLkWXb74BnE6syMhAtWtdi17FtB4LuxUXF+vzPBo4pXdv9Ni2Dfu/+w6/KA2V+mp2d1I0n19b5Pz8M9pbLPimvh7NrgH9/v2ROyidhPQaOhxIaGqCQ+3LTqSK1RWE4z3abvNm9Hn3Xfx0771oPjncmnznnWjKzIRDDuJ0qrPx+Xe0f39x8WC222FyOuE4+Tol2e0onDIFtqoqJJ682CorYa2thaW+HjsA7Di5bebu3Rh74oTIsJ3kOvi0s1MnbDmZWUo5fBjjVRICALCrqgobT9bz2E6cwCQ5eDmpunt34NRTNX8N64JoBxHzpdGzZ892y9RUVVWhsLAQEyZMQEZGhqbPZS8rw5bf/x4909KQIEnim4IkiYhaktC5Xz90kvuHNDTAcd11bve3/NzcjPz+/XGBy7bOESPEN8fGRpiamkRxXW0tUFuLju3bt27b3Ayrjw8v5+jRcMi1IwASpk0D0tIgnRxblU6Os0q5uSIt6jI8Y3c6UVxaivGe3xjaqrwcpu++g3nZMpjXr2891hEjcFZOjr49V1JSxKyAk+x2O4qLizF+/HhYtR6aCsaRI6ofJKYzzwQ+/hjdfvsNnZX+b3r2BNLSFB8bNeenlepqt1W5E+6+G+adO9Hw3HNIOOMMcWN2dniWeQgTv69hUxOwZYvXzebXX4f51VfhvPpqOG+5RXnnNhvQr5/GRxwct/OrrNRnxeimJphfegkJS5YAACZ98QWc06eL+3Tu8aTp39GCAjGcpaL3yQsAoF072JcsgamuTgzByS0X7HbAbke3oiJ0k889JQUO+T0ifz7JM6YkCV0GDkShvG1GBhxXX926DYCUTp0AQPO/M1VBLGWha/DSoUMHJCQkoKyszO32srIy5Hm2Pz8pLy8vqO1tNhtsCmlQq9Wq/R/vdu2w/fLL0augAAn+3pQpKaJ2IxApKYBaPY8kweRwwCw/X0IC8Pe/i6Kv8nIRZbtcm3v1at22vl7MPlAzapQYTz3J/OKL6GE2I7F3b1jy80UhWbt2wRW51tWJXxh5HHb9euDmm1tTwmYzcMYZwPXXwzxkiL6NhkwmMRat8D7Q5f0RjJwc5TWMAFH3MnIkTEOGwCqnc101Nvpt9x7x89OKw9E6NdzhgHQyw5LQrVvrB0NqqvY1UlFA9TW0WsXFc8HF9HSgthYJP/+s/vfJbhe/n1FQ9G0FYC0r037q/759YpbQb7+Jf196KRKmTvX/N1tjVrNZ2y+B/nh8UfOpXTtg2rTAts3IEIviunDabMDOnZr/nQlmX7oGL4mJiRg+fDiWL1+OyZMnAwCcTieWL1+O6XIU7GHUqFFYvnw5Zrh88BcXF2PUqFF6Hmr0Mpncg4eEBPeiMl8SEsR047Iy8UF5cmy15WfX4bn6eiS8/joU53OlpgITJgD33y/+3dws3swJCeKDtK5OXMrLxeWyy1q37ddPfNvr1k1MQ7z4Yv+dZrWSmxu9sysSE8UfG6U0afv2YqkANfHUade1iLOsDKamJjgsFve1n+KlWNdVaqp33cvQoeJ640YRpKh9EBw/HhXBCw4c0H7F69WrRVfuqirxBerBB4ExY7R9DooKug8bzZw5E9dffz1GjBiB0047DQsWLEBtbW3L7KOpU6eiU6dOmD9/PgDgz3/+M8aMGYNnnnkGF154Id5++22sWbMG/wi06ptaJSYCo0cr3ydJ7tN1HQ44rrsOh/fuRUFVFcxlZaIy3+EQw1euf2Sam4HvvlN/3uPHW39OSgI+/VRE7+GUlBS9Mytk2dmhBSLxNOPINXg52Xm1Nj8fya4dkuOlu66rlBTv4KWoSMycqawEtm5VH24oLxcN6yLd7LCyUttj+PJLEaw4HKJx31NPKTfsI0PQPXi56qqrcPToUcyZMwelpaUYMmQIPv/885ai3H379rUOcwAYPXo0lixZggceeAB//etf0atXLyxdupQ9XrRmMrn/0U9Lg/POO7H20CHkFhSI18ThEDUHVVUiEJJZLMCcOeJ+OYOQnCy+6XTqJNLXrsIduACiBiLSf5z9yc5Wn3UEiCBw/34xjOSqqUkEkJHuWRMOCsFLTadOcMunxWvmxZPJJPqSfPON6GOiFrw4HGLYuV07XQ9RldbZFtngwSKrO2wY8Ne/xmdQG0fC8tdv+vTpqsNEK1as8LrtiiuuwBVXXKHzUZFfCQkt0//cWCzAJZdE4ogCk5OjWtAaVWw28cGr1N9i507gqqvEh9TXX3uvxVRbG95VsSPB6XTPDsqZl4ICtHS/SUyM/iBVD2rDPiNGtAYvf/iD+uOPHo1c8OIrYG+L3FzR4JFN3uJCHP7Wk6ElJsbGGi4ytQZT3bqJjFZtrfLU6Hioe/FY08g189IiXr9dy0W7nuR6uJ9/9r1GUE2NKOgPt6oq38tjBOu110SwJsvKYuASJxi8kLHEwnCRK7XsSUICIA+VKnXijIe6F8+M1K23wnHffTg2YEDrbfE4ZCRTyr4UFYnC+Btv9L/AYbjXEXI4tF0C4O23xWzJWbMC6n1ExhJDf+WJ/MjJiUx9TVukpanXrsg1C0rBSzxkXjyDl0GD4Lz8ctTF+0wjmVrdy1NPiRb4/oZOy8tF7VS47N+vXRfkr74Cnn5a/PyHP4iWCBRXGLyQMdhssTVc5Eot4PIVvJxsPGVogax1E6/DRkDbpzs7neHLvlRUuM9CbItNm4C5c8XPV1wh+khR3GHwQsbQrVtsDRe5Uhs6koeN9u9XrhMw+tCRa/CyfTvw0UfAjh3u2zDzouzECZGd8BcAHjmi3+wfmd3eUq/UZqWlYhHKxkbgzDNFvynWuMSlGP1rT+QiLy82ZhepUcu8ZGS0psPjcejI9YN35Urg4YeR4LpemMnkPoU/3pjN6k0Yr7tO1IKorVwsczj0z77s3q3N8FRTE3DffSKD07s38Oij3rPwKG4weKHYlpqq+1olurNY1L9F33AD8NBDrVkYV0bOvDQ1ua/ue/Kbu9SlS+ttNhu/dau9b4YPF9c//uh/H2Vl+tW+HD4sekVppX9/kal85pno6BJMEcPghWJXQoKYXWGEDzC17MsFFwAXXaTcKdTImRfP4Y59+wAAkusCjPE8ZCRT+wCXl1NZtcr/PhwOfVabrqrSdr+JiSLz8s470d89m3TH4IViV9euxinYDGWWlLxqrBG5Bi+S1Jp5cQ1ejPLat4Va8HL66SKo/+034Ngx//s5elTbYLixUbvpy8eOuWeG2PKfwOCFYlXHjn5XVo4pqanq4/ebNwNLlojCXU9GHTpyDV7Ky0VTNZPJfUYZMy+i5kXpfZOdDfTtK37+4YfA9rV3b+vq723R3CwKrLUoBG5sBG6/HZg+XdvmdhTzGLxQ7ElLi91p0WpMJu81oWSLFgHPPqv8IWTUoSPX4OXkkBHy892zLQxehJQU5dvloaOSksD2U1fX9tb9TqeYEebZHTlUCxeKDM7u3drsjwyDwQvFlsREMQPHCHUuntSCF7mj7JYt3vfFQ/AiT7N1HTICOGwkU5tpJ68o/8MP7sXPvpSViWnWoXA6RcZFq2zgjz+KjCMgVouO1FpMFJXiYFlaMgyzGejRQ3lNFyNQq3uRg5fNm73vM+KwkcPh3oBv7FgxHd41WElIMO77IFhqdS+nnCI+9E8/PbgeSHv2iBlwasG0ErtdZFy0CqarqoB588TPl18ueroQuWDwQrHBZBIZF7UUuREkJYkPZM/Ouf37i+vdu0Ww4vph5XCIFL2RshCeM42yssQHMNCaQYjn/i6e1IIXiwW49NLg9ycP/RQVqS8c6qqmRgztaNnx+cknRRaosBCYMUO7/ZJhcNiIYkOXLuqdaI1E6dtu+/Yi8yBJwNat3vcbbegokGUBWO/SymLR/v/D6QR27hT1Rmo9YJqaxJDetm3aBi7LlwOffy6ya488ot6Ij+IagxeKfp07x8/0SH91L/EwdFRf3/pzczPw0kvAp5+6f4gaKdOkBbXsiyQB770nZuyE0kn36FFg40aRWSkrE9OWDx8WtS2bNgU2DTtY+flieHjqVOXmjETgsBFFu4ICIDc30kcRPmrBS//+4hupUvBi5MzLgQPAK6+Ib9/nn986lZfBi7u0NOWFD00m4LPPxDIBK1eKhQyD5XSKIt5QC3mD1b8/8MYb2kzbJsNi8ELRq3Pn+ApcAPGhnJjo3Xxu4kRg8ODW3h2ujBy87Nkjrrt2FR/E8gcah43c+WqVP3asCF5WrAgteAmX5mYxBAawGJv84rARRR+TSawSHW+Bi0xp6mteHjBkiPKHtsMRWJ1ILHA63XuEqE2TZsGuO7VmdYAIXgBgzRpt1xnSUm0tcOWVwP/7f/qts0SGwswLRReLRcwqCmaaptGkpwffTbSuzhjZCM/mZmrBC1cT9paWBlRWet/epYv4ndq1C/j+e2DSpPAfmz8LF4ri4A8/BK65pjUDE24JCeL3KDFRZH8sFnGb61RzSRJBtsMhLs3N4mK3i4tRl+yIMgxeKHqkpoo/svH+rVqt6djPPwNffAH07g1Mnux+X12dMZp4uRbrAurBC3lTC14AkX3ZtQv45pvoC142bBBFxQBw//3hbYdgsYhZjOnp4v9Pi1qqxkaxIGXPniLIaWoStzU1tV5Yz9NmDF4o8kwmMURUUGDMzrnBUuv38ttvwLvvirbvnsGLUWYceQ5/uda8kG9qQS8ggpdXXhGZl7q66OmX1NgopkNLEnDJJcDIkfo/p9ks1n5q316fDK+cpUlLU6/dkTM0nhc5e2O3M8Dxg8ELRVZqqkhrR8sf02iRluY9u8N1mQBJcg/0jFK065p5qahozSQwePEvJcW9qNlVv36ijqygQPy/Rsvv2yuviOxa+/b6N6NLSBALunbsGLlhKZnVKi6+Cq1dh6Pkn5ubW4eq5GErh6N1GMvpFJc4CHwYvARDfsP37q385vd8w0hS622u164X+Y3meu16cX2Dul5inc0m+jm0bx/pI4lOSsFLr17iD15lpVhAz3VxSqdTZC1ivRbENfOSkQEsXSrOlY3K/DObxYdhTY33fSYT8NZb0TWLZ/t2YPFi8fO996ovj9FWJhOQkyMCt1j6/bBYQm9AqPTZonaRt1e6VtqvbOfO4I9LQwxegiF/001JifwfATnydo3KXVOO0Zp+TE0V33yyszlE5IvSEIDVKgLnzZvFxXNl7dpa/T4AwkGS3At2zWZxjkZbQVxPaWnKwQsQ+b9ZHkw7dohg4qyzgHPP1edJUlNF1i7egl+TSd9ATcuOyiFi8BKr5KjcX4GZJKmPr8qFZIGuOBuq5GSxRkp2dvz9EQlVcrL48PZ8bQYMaA1eJk50v6+uLraDl4aG6Aq0Y5GvuhdZaaloaCcPQ0aIdP75ooNuaqr2X2RMJpHZzcvjlySDYvBidCaTmL3jawZPc7MIZOrqRJV8bq57VsduF//298Eir/Rrs4kP35QU30VrpM5kEv93VVXut8uLNBpxmQDPYt033hDvyUmTRF0UwM66/vgLXpYvB+67T9TA/Pvf4TkmX/SoZWK7hbjA4IVaszhykJGfrxxwuBaFyYGMySQyBJ69EKjtlIIXea2Xigrvot36+tjOXHhOk/7gA1HMOXhwa/AS79Po/UlIEF8a1Aq4hw0T22zdKlaO7tkzvMcnSTA/9RSyhw8XNShaS04W58T3ieHx04YCZza3ZlaSksTFZhO3MXDRntK36C5dxIq7//2vdzrcszttrPFckPHAAfGz67dzIzTi05uvjEN2NnD22eJnubdKOC1bhoT33sPouXNFAK6ltDSgTx8GLnGCnzhE0UqpFsBs9r3CdiwPHbkOGx08KLJ8SUmiwFvGYSP//A0dXXWVuF62LLzLBZSXA889BwDYdtVVog5OKxkZYjZeLM0mojZh8EIUrczm4AucPYdeYoXnTCO5OV2XLu5ZPQYv/vmr9Rg+HOjRQwSLH38cnmOSJODRR4HKSki9e2PnJZdot++MDDFUxOxvXOGrTRTNlJpY7d4N3HkncPvt3vfFaubFc6aRvCxAt27u2zF48U+ue1FjMolFEAExdKT3bEMA+OgjYOVKwGpF89y5kLRqEpeWJgIxziiKOwxeiKKZ0hBAcjKwahWwdq33DJ1YXV06kDWNzGbWMwTKX/bl/PPFe6u8vPX/Wi8HDgDPPCN+vu02MbyjBbk4lxmXuMRXnSiaKWVecnPFIowOh+hS6ipWZxt5Bi/794tr1+CFWZfA+QteUlKABQuATz8Fior0PZYPPxSzn4YOBaZM0WafiYmscYlznCpNFM1sNjGNvbm59TaTSfR7+e470e9l4MDIHZ9WPIOXl14S39hdl49g8BK49HT1dY5kQ4aE51huv11Miz7tNBFstHWYymwWQ0XsHxXXmHkhinZKQ0dys7qtW8N7LHrxDF4sFlHv4ppB4DTpwMnrHAVCkoCSEv3WTDOZgMsuAzp10mZ/3bpFz8KSFDEMXoiindKHkBy8bNkS3mPRg8MhOjz7w+AlOIEsFSFJovj7T38Ciou1e+5jx8TsIrV1lkKVlyd61VDcY/BCFO2Ugpd+/cT1nj3KM4xiaeVxzyLj774D5swRzfhccdgoOIG0xzeZWoePXnhBm9lqTU3ArFmiQ/K8eW3fnyw9XZ+uvBSTGLwQRTulZnXt24v1W4YNU+5UGkv9XjyPdf16UUj688/utzPzEpzU1MAKWqdMEUM6ZWXAokVte06nE3joIWDDBvH806e3bX8yi0UUFnNKNJ3E4IUo2pnNyh/c77wDvPyyci1BLPV7CWSadEKC+ACjwJlMgWVfkpJEpgQA3noLWL069Od84QXgyy/F6/XEE9otvFhUxAJdcsPghSgWKA0d+foWqrYwXzTyPFa5u65rgzpmXUKTmRnYdqNGAZdfLn5+8EHg8OHgnkeSgIULW1eqnjMHOP304PahJjc3sPodiisMXohiga+ZI0pFkbEUvHBBRv0E86F/111A796icd0jjwT3PIsWAa++Kn6eMQO48MLgHq8mOVm7WUpkKAxeiGKBUvDS1AT87nfAOed4173Y7YHN4Im0pib34uIDB0QAk5QkvnHLWKwbmsTEwNfHSkoCnn1WdK2dOTO45zn/fFGHdf/9wHXXBX+cSkwm1rmQKgYvRLEgOdm7+FJulS9JwK+/ej8mFrIvnvUuu3eL66Ii97bvzLyELpjsS14esGSJCGBkP/zgvmgmILJ9n3zS2gSvWzfg/fdFPxetFBQEvzApxQ1WwBHFipQUoLra/bZ+/YB9+0SnXc8ag9paICsrbIcXEs8A68gREbR4tqxn8BK6rCwxkyhQrkHjpk1ixlBaGjBokKihKSsDNm4U2b1OnUTbfyDwpniBSE0VgRSRCgYvRLEiNdU7eOnfH/jiC+VOu7Ew48gz83LVVcDkyd63c9godKmp3ktMBOrYMRFElJaKxUBdde/unZHRgtnsvZo4kQcGL0SxIthOu7W1Iq0fzTUDSkNbNpt7sJKYyJWD28JkEkNH5eXBP3bsWODss8X7a9s2EVRmZQEDBogAQ4/3Vn4+M23kF4MXolihtJ5L377ig/3IEfEtuV271vucTvFhE63rwDidgX1z5wdZ22VlhRa8AOL9dcop4qK3lBT3Qm0iFfw6QxQrEhO9G3UlJ7fWh6hlX6KVZ9blwAHgD38AnnzS/XYOGbVdZmb0Z69MJv2yOWQ4zLwQxZLUVO9p0eedJ74Vt2/vvX1tLZCTE5ZDC5pnXcvOncAvv3ivdcTMS9uZzWLoSGkpiWiRl8fZRRQwBi9EsSQlxfsD6OabW392Ot3vi6XMizxNunt399sZvGgjKyt6g5ekJFHrQhSgKM8jEpGbYKejNjSENsskHDyDl127xDWnSesjKyt6h2S6do3eY6OoxOCFKJaoBS92u2hU5zmVGojO7IskqTeoc828mM2tzfiobRISAl/rKJw6dBB9ZIiCwOCFKJYkJCgXsN58M3DddTAprQgcjcFLfX1rd1ZADHcpBS8s1tVWdnakj8Cd1Qp07hzpo6AYxOCFKNYoZV969QIAmJSa1Skt3BhpnkNGpaViiMtqdV+Ij0NG2srKiq5ZR4WF3steEAUgit7FRBQQpb4tJ5vVmXw1q4smnsHLiROiYLNbN9ENVsbgRVtmc/RkXzIzo+dYKOZwthFRrFHKvAwYAAAw/fqr94yjaGxW5xm8DBgAfPyxd3Exh4201749cPx4pI+Cw0XUJsy8EMUapSCkqAiw2WCqrUXa4cPe90fT0JFSsa7M4vF9ipkX7aWnR0dQ6NlwkSgIDF6IYo3Z7N3My2IB+vQBAGRt3+79mGgKXurrvbNDahi86KNDh8g9N2cWkQYYvBDFIh91L1k7dnjfF03Bi+fsJ0kSK0nfcov7cIbVymJOvXToEJm+KmYz0KVL+J+XDIc1L0SxKDXVu27h3HPhaN8eB7p1g9fHg90uFkGMhuECz3qXsjKxrtHhw6KFvYxZF/1YLKJYNtTFGkPVqRP79pAmGLwQxSKlzMuwYXAOGYKKQ4eUH1NTEx3Bi2fmRe7v0rWrex0Egxd9dewY3uAlLU08p90evuckw+KwEVEsSk4OPu0fDUNHTqf3wos7d4rrbt3cb4+GQMvIUlPDV39iNnu/vkRtwOCFKBYpFe0CQFkZCr7/Xrnfi9LSAeGm1HNGDl5ONtprwcyL/vLywvM8hYUMRklTDF6IYpXC0JH5nXdw6lNPwfTRR97bNzZGPmWvtFSBPDuqZ0/32xm86C8zU//+P9nZkZ3dRIaka/BSXl6OKVOmICMjA1lZWbjppptQ4yd1PXbsWJhMJrfLrbfequdhEsUmhWZ1kq9Ou0Dkh448g5fm5tbVpF0zL2Yzv6mHS0GBfvtOTBS1TEQa0zV4mTJlCjZv3ozi4mJ88skn+Pbbb3HzzTf7fdy0adNw+PDhlsuTTz6p52ESxSaFb8wtwcuOHUBTk/djIj105Bm81NQAQ4aIWSiuH6IMXMInM1M0rtOaySQW2eR0d9KBbrONtm7dis8//xw//fQTRowYAQB44YUXcMEFF+Dpp59GgY9oPyUlBXnhGoslilVy0a5rDUl+PhrT02GrrhbDMSeXDWgRyeBFadgqKwt46SXvbTlkFF6FhcDWrdqugdWli/JSFkQa0C14KSkpQVZWVkvgAgDjxo2D2WzG6tWrcdlll6k+9s0338Qbb7yBvLw8XHzxxXjwwQeRojIu29jYiMbGxpZ/V1VVAQDsdjvsGo/vy/vTer/RwujnBxjwHBMT3Vrt2yUJ1b16IXfdOjg2b4azXz/37evqxCUSrdkrKgLvrGu1KtbnGO71UxCRc7RYxJpHR45os7+cHJHRicPX0OjnB+h3jsHsT7fgpbS0FB07dnR/MosF7dq1Q2lpqerjrr32WnTt2hUFBQX45ZdfcN9992Hbtm14//33FbefP38+5s2b53X7l19+qRrwtFVxcbEu+40WRj8/wNjn2LdHD+SuW4eDa9Zg/ejR3huo9YGJAHNjI5xKQ0R+jtHIr58sps/x0CHg5599bhLT5xcAo58foP051nk2sPQh6OBl1qxZeOKJJ3xus3Xr1mB328K1JmbgwIHIz8/Heeedh507d6JHjx5e28+ePRszZ85s+XdVVRUKCwsxYcIEZLh269SA3W5HcXExxo8fD6sBFxUz+vkBBjzH48eB/ftb/ml3OrHxZOFr4d69yFcanm3fXgwThNuvv3r1eLFMngw4HGh+7jn32UZ9+ihOBTfc66cgoufY0AD89lvgGTJPaWmizsWsXk5p9NfQ6OcH6HeO8shJIIIOXu6++27ccMMNPrfp3r078vLycMQjBdnc3Izy8vKg6llGjhwJANixY4di8GKz2WBT+OZmtVp1e+Poue9oYPTzAwx0jpmZwMGDbjcd79cPzY89BsuAAbAqfYjU1oZ/2Ki5WRQQux5PbW1LhsXasaP7fWlpPj8ADfP6+RCRc7RagR49WnvvBCMjQzzWx+vm/lTGfg2Nfn6A9ucYzL6CDl5ycnKQk5Pjd7tRo0ahoqICa9euxfDhwwEAX3/9NZxOZ0tAEogNGzYAAPLz84M9VCLjS0oSHxYu35Tt6emQxo1T/xBpagr/OkdKU7TlD8icHFG4K0tMDPgDkHSQlSW64e7ZE/hjcnJENi8Siz1SXNLtL0S/fv0wadIkTJs2DT/++CO+//57TJ8+HVdffXXLTKODBw+ib9+++PHHHwEAO3fuxCOPPIK1a9diz549+OijjzB16lScffbZGDRokF6HShS7TCblTrv+BJGe1YTSLCe5OZ1nZ91Qzoe01b69eF0sfr7fWiximKhLFwYuFFa6Lsz45ptvYvr06TjvvPNgNptx+eWX4/nnn2+53263Y9u2bS1FOomJifjqq6+wYMEC1NbWorCwEJdffjkeeOABPQ+TKLalpHj3Tzl4EPjqK5HFmDLF+zFVVeLbcrgoZV527BDX7KwbnTIyxFT70lJRW9Xc3HqfzSYCnI4d2ceFIkLX4KVdu3ZYsmSJ6v3dunWD5NJXoLCwECtXrtTzkIiMJzUVOHrU7SbTgQPAwoWi8ZtS8FJdLXp6hOPbcnOzmJ7t6bffxDWDl+hlsQCdO4tLYyPgcIi6GIPXclD048AyUaxT6rQ7YIAITA4dAo4d836MwxG+pQKUnsfhaA1e+vZ1v4/BS3Sy2cR7jYELRQEGL0SxTi7adZWWJmZ+AMAvvyg/Llx1L0rP09AAXHghMHSo99o3DF6IyA8GL0SxzmRSXhlYLnLfuFH5cZWV+h2TK6Vi3dRUYNYs4J//dK+ZsFj8F4kSUdxj8EJkBL6CF7VOp/X1yos3aqmpyasxnU+caUREAWDwQmQESgvgycHL1q3qQUpFhW6HBEB9aGrPHuWghkNGRBQABi9ERqCUeSksFA3HzGa3JQTc6D10pLR/SQJuugk4+2zvTq7MvBBRADi4TGQESUmidsR1TRqTCXjlFSA/X32GSHW1mPmjR68OSVLOvJSViaAmIcF7jSVmXogoAMy8EBmFUvalSxffU1slSb+ho+pq5QX+fv1VXPfoIZrouWLwQkQBYPBCZBRKwUsg9Ape1Iaktm0T1336uN9usbCHCBEFhMELkVEoBS+SBDz+OHDllcDhw8qPq6wUQ0daUwuK5MyLZ/DCrAsRBYjBC5FRKM04MpmAzZuBXbvUm9VJEnDihLbHUlurPsNJzrx4dtZlsS4RBYjBC5FR2GzKhbf+mtUBQHm5tseiFgwdPQocOSJmQPXu7X4fMy9EFCAGL0RGojR0NHCguFbLvACiuFbLhnVqwYvVCtx1l1gs0vNYmXkhogBxqjSRkSgFL4MHi+tt20RjOLUMx/HjYlp1W/kKhLKylFe5Bhi8EFHAmHkhMhKl4CUvD+jYURTl+ho6On5cm2MIZT9c04iIgsDghchIlLIXJhMwbJj4ed069cc2Nra9467DoT5k5HAAn3wilgaQJPf7mHUhoiAweCEyEs+mb7Jhw0Q3W6UZSa6OHGnb8x8/rtyYDgB27wYeegiYOtV7GwYvRBQE5mmJ4sHkycDvfud/u6oqsdp0qMHE0aPq923aJK779fOeFcXghYiCwMwLUTwwB/GrXloa2nNUVCivFC3bvFlcn3KK930MXogoCAxeiOJJc7P/gtryct9BiBq1Dr4yOfMyYID3fezxQkRBYPBCFC++/RY47zzgwQf9b3vgQHD7rqgA6urU76+vB3buFD97Zl4SE/VZ1ZqIDIvBC5ER2Wzet3XqJNr2//wzYLf7fnxlZeAzj5xOYP9+39ts3iy269hRXFxxyIiIgsTghciIlPq9dO8umsQ1NgJbtvjfx969YpjJn0OH/HfnXb9eXA8Z4n0fgxciChKDFyIjUgpeXPu9/PST/33Y7WJ6s2dPFlcVFUBZmf99XXkl8MwzwFVXed/H4IWIgsTghciIlIIXABg5Ulz/8ENg+6mqEitSK/VuqawU9wUiMxMYM6Z1qQJXDF6IKEjs80JkRMnJItPimTU5/XRxvXEjUFMDpKX531dFBbB1q1hmIC1NDCUdOyYubWUycaYREQWNwQuREZnNIoDxnAHUqRPQubOYTbR2rciGBKKhQbT1D8UXX4jhp3PPBXr3dr9PDrKIiILA4IXIqFJTlacvX3mluL1Hj/Acx7JlwKpVYuhIKXghIgoSgxcio0pNVW7Xf+214TsGh0NMzQaAoUO972fwQkQhYMEukVH5W4QxHH77TfSWSUsDevXyvl+tsJiIyAcGL0RGlZSk3rm2ogIoLm7tv6KXtWvF9ZAhysfCzAsRhYDBC5GRqWVf3nsPmD0bWLJE3+cvKRHXI0Z435eYCFg4ck1EwWPwQmRkasHLWWeJ65KS0BZhDER9PbBunfj5jDO87+eQERGFiMELkZGp9XHp0wfIzRWBy48/6vPce/eKACU/H+jWzft+DhkRUYgYvBAZmVrmxWRq7fGycqU+z923L/Dll8A//qHcy4WZFyIKEYMXIiNLSFDvYDt2rLj+9lsxpVmv58/PV76PwQsRhYjBC5HRqQ0dDRsGpKcDJ06I5QK0ZLf7XtDRYhEFu0REIWDwQmR0akNHFgtw5pni502btH3O994DLrpIfTYTsy5E1Aacp0hkdL4WX5w2DbjjDrHoopa++w4oK1NejRpg8EJEbcLghcjokpJElqW52fu+Ll20f76KitbmdHJdjScGL0TUBhw2IooHgSwVoFW/lxUrRAFwnz5iBWslDF6IqA0YvBDFA19DRxUVwF13iRoVLQKYr74S1+edp3y/xQLYbG1/HiKKWwxeiOKBr+AlIwPYvl0EMd9+27bnOXastenduHHK2zDrQkRtxOCFKB6kpCg3igMAsxm44ALx87JlbXuezz4TRbqDBqnX00TDatdEFNMYvBDFA7PZd9AgBy8//AAcORL684weDVxzDXDllerbMHghojZi8EIUL3wNHXXrBgwdKgpt//Of0J+jRw/g7ruBSZPUt+GwERG1EYMXonjhK3gBgKuvFtfvv6/fStOJiYDVqs++iShuMHghihf+gpcxY8Q6RBUVwde+lJYC998PbNjgezsOGRGRBtikjiheJCSIIZu6OuX7LRbRbbeiQkybDsZrrwFffCFmG738svp2/gIoIqIAMHghiifp6erBC+C7VkVNWRmwdKn4edo039sy80JEGuCwEVE8CSbz0dQEVFf7327BArGK9LBhwPDh6tuZTCzWJSJNMHghiifp6YFtt2GDKOCdMweQJPXtSkqA4mIxFfvuu9V7yQC+e80QEQWBwQtRPJHrXvxJSQEOHwb+9z/gzTeVtzlyRAQ3gOjr0qeP732y3oWINMLghSjeBJJ96d0b+POfxc8LFojeL54ZmI8+Ak6cENtOn67N8xIRBYAFu0TxJiNDFNn6c9VVwMGDwFtvAY8/LtY9GjMGmDBBBCJ/+APQ2AhceimQlOR/fyzWJSKNMHghijdpaaL2xFctCyC2mTkTaN8e+PvfgVWrxGXkSBG8mM1ianUgkpLEVGwiIg3wrwlRvDGbRQATyEwikwm44QZg7FgxTLR7t8jcBItDRkSkIQYvRPEoIyOw4EXWrRtw552hPx+DFyLSEAt2ieJRKNmTtmDwQkQaYvBCFI9SUsJXg8J6FyLSGIMXongVruwLsy5EpDEGL0TxKjMzPM8T7iEqIjI8Bi9E8SozU/92/SYTMy9EpDkGL0TxKiFB/5b9qanieYiINMTghSie6T10xCEjItKBbsHLo48+itGjRyMlJQVZWVkBPUaSJMyZMwf5+flITk7GuHHjsH37dr0OkYgC/N0MWbjqaogorugWvDQ1NeGKK67AbbfdFvBjnnzySTz//PNYtGgRVq9ejdTUVEycOBENDQ16HSZRfLPZgORkffZttQa2gjURUZB0a74wb948AMDixYsD2l6SJCxYsAAPPPAALr30UgDA66+/jtzcXCxduhRXX321XodKFN+ys4H6eu33y6wLEekkajpH7d69G6WlpRg3blzLbZmZmRg5ciRKSkpUg5fGxkY0Nja2/LuqqgoAYLfbYbfbNT1GeX9a7zdaGP38AOOfY0jnl5YGOJ3aH0xaGsDfwaAZ/Rx5frFPr3MMZn8mSfK3tGzbLF68GDNmzEBFRYXP7VatWoUzzjgDhw4dQn5+fsvtV155JUwmE9555x3Fxz300EMtWR5XS5YsQQpT1kRERDGhrq4O1157LSorK5Hhp9g/qMzLrFmz8MQTT/jcZuvWrejbt28wu22T2bNnY+bMmS3/rqqqQmFhISZMmOD35INlt9tRXFyM8ePHw2q1arrvaGD08wOMf44hn19ZGXD4sHYHkp0NdO2q3f5OMvrrBxj/HHl+sU+vc5RHTgIRVPBy991344YbbvC5Tffu3YPZZYu8vDwAQFlZmVvmpaysDEOGDFF9nM1mg81m87rdarXq9sbRc9/RwOjnBxj/HIM+v44dRQCjlQ4dRMGuToz++gHGP0eeX+zT+hyD2VdQwUtOTg5ycnKCPqBAFBUVIS8vD8uXL28JVqqqqrB69eqgZiwRUQgSE0Un3Orqtu/LYmGxLhHpSrep0vv27cOGDRuwb98+OBwObNiwARs2bEBNTU3LNn379sUHH3wAADCZTJgxYwb+9re/4aOPPsLGjRsxdepUFBQUYPLkyXodJhHJOnTQZj/Z2fovO0BEcU232UZz5szBa6+91vLvoUOHAgC++eYbjB07FgCwbds2VFZWtmxz7733ora2FjfffDMqKipw5pln4vPPP0dSUpJeh0lEsuxsYP9+oLm5bftp316b4yEiUqFb8LJ48WK/PV48JzqZTCY8/PDDePjhh/U6LCJSYzKJ7Etpaej7SE4W6xkREemIaxsRUaucnLYN+XTsqN2xEBGpYPBCRK0SE8XwUSgsFqBdO22Ph4hIAYMXInKXnx9a9qVjR8DMPylEpD/+pSEid0lJwWdQEhI4ZEREYcPghYi8deokApJAFRQEtz0RURsweCEib1arCEgCkZwsCn2JiMKEwQsRKevY0X+nXLMZKCpiUzoiCisGL0SkrqgIUFud3WQCunUTmRciojBi8EJE6hISgF69vDMwFgvQvXvo06qJiNpAtw67RGQQFgvQsydQWwvU1Ih/Z2WxQJeIIobBCxEFJjWVrf+JKCpw2IiIiIhiCoMXIiIiiikMXoiIiCimMHghIiKimMLghYiIiGIKgxciIiKKKQxeiIiIKKYweCEiIqKYwuCFiIiIYgqDFyIiIoopDF6IiIgopjB4ISIiopjC4IWIiIhiCoMXIiIiiimWSB+A1iRJAgBUVVVpvm+73Y66ujpUVVXBarVqvv9IM/r5AcY/R55f7DP6OfL8Yp9e5yh/bsuf474YLniprq4GABQWFkb4SIiIiChY1dXVyMzM9LmNSQokxIkhTqcThw4dQnp6Okwmk6b7rqqqQmFhIfbv34+MjAxN9x0NjH5+gPHPkecX+4x+jjy/2KfXOUqShOrqahQUFMBs9l3VYrjMi9lsRufOnXV9joyMDMO+KQHjnx9g/HPk+cU+o58jzy/26XGO/jIuMhbsEhERUUxh8EJEREQxhcFLEGw2G+bOnQubzRbpQ9GF0c8PMP458vxin9HPkecX+6LhHA1XsEtERETGxswLERERxRQGL0RERBRTGLwQERFRTGHwQkRERDGFwYsPe/bswU033YSioiIkJyejR48emDt3Lpqamnw+rqGhAXfccQfat2+PtLQ0XH755SgrKwvTUQfn0UcfxejRo5GSkoKsrKyAHnPDDTfAZDK5XSZNmqTvgYYolPOTJAlz5sxBfn4+kpOTMW7cOGzfvl3fA22D8vJyTJkyBRkZGcjKysJNN92Empoan48ZO3as12t46623humIfVu4cCG6deuGpKQkjBw5Ej/++KPP7d977z307dsXSUlJGDhwID799NMwHWnogjnHxYsXe71WSUlJYTza4Hz77be4+OKLUVBQAJPJhKVLl/p9zIoVKzBs2DDYbDb07NkTixcv1v04QxXs+a1YscLr9TOZTCgtLQ3PAQdp/vz5OPXUU5Geno6OHTti8uTJ2LZtm9/Hhfv3kMGLD7/++iucTidefvllbN68Gc899xwWLVqEv/71rz4fd9ddd+Hjjz/Ge++9h5UrV+LQoUP43e9+F6ajDk5TUxOuuOIK3HbbbUE9btKkSTh8+HDL5a233tLpCNsmlPN78skn8fzzz2PRokVYvXo1UlNTMXHiRDQ0NOh4pKGbMmUKNm/ejOLiYnzyySf49ttvcfPNN/t93LRp09xewyeffDIMR+vbO++8g5kzZ2Lu3LlYt24dBg8ejIkTJ+LIkSOK269atQrXXHMNbrrpJqxfvx6TJ0/G5MmTsWnTpjAfeeCCPUdAdDJ1fa327t0bxiMOTm1tLQYPHoyFCxcGtP3u3btx4YUX4pxzzsGGDRswY8YM/PGPf8QXX3yh85GGJtjzk23bts3tNezYsaNOR9g2K1euxB133IEffvgBxcXFsNvtmDBhAmpra1UfE5HfQ4mC8uSTT0pFRUWq91dUVEhWq1V67733Wm7bunWrBEAqKSkJxyGG5NVXX5UyMzMD2vb666+XLr30Ul2PR2uBnp/T6ZTy8vKkp556quW2iooKyWazSW+99ZaORxiaLVu2SACkn376qeW2zz77TDKZTNLBgwdVHzdmzBjpz3/+cxiOMDinnXaadMcdd7T82+FwSAUFBdL8+fMVt7/yyiulCy+80O22kSNHSrfccouux9kWwZ5jML+b0QaA9MEHH/jc5t5775UGDBjgdttVV10lTZw4Uccj00Yg5/fNN99IAKQTJ06E5Zi0duTIEQmAtHLlStVtIvF7yMxLkCorK9GuXTvV+9euXQu73Y5x48a13Na3b1906dIFJSUl4TjEsFixYgU6duyIPn364LbbbsPx48cjfUia2L17N0pLS91ev8zMTIwcOTIqX7+SkhJkZWVhxIgRLbeNGzcOZrMZq1ev9vnYN998Ex06dMApp5yC2bNno66uTu/D9ampqQlr1651+783m80YN26c6v99SUmJ2/YAMHHixKh8rYDQzhEAampq0LVrVxQWFuLSSy/F5s2bw3G4YRFrr2GohgwZgvz8fIwfPx7ff/99pA8nYJWVlQDg83MvEq+h4RZm1NOOHTvwwgsv4Omnn1bdprS0FImJiV71Fbm5uVE7xhmsSZMm4Xe/+x2Kioqwc+dO/PWvf8X555+PkpISJCQkRPrw2kR+jXJzc91uj9bXr7S01Cv9bLFY0K5dO5/He+2116Jr164oKCjAL7/8gvvuuw/btm3D+++/r/chqzp27BgcDofi//2vv/6q+JjS0tKYea2A0M6xT58+eOWVVzBo0CBUVlbi6aefxujRo7F582bdF6ENB7XXsKqqCvX19UhOTo7QkWkjPz8fixYtwogRI9DY2Ih//etfGDt2LFavXo1hw4ZF+vB8cjqdmDFjBs444wyccsopqttF4vcwLjMvs2bNUiygcr14/iE5ePAgJk2ahCuuuALTpk2L0JEHJpTzC8bVV1+NSy65BAMHDsTkyZPxySef4KeffsKKFSu0Owkf9D6/aKD3Od58882YOHEiBg4ciClTpuD111/HBx98gJ07d2p4FqSFUaNGYerUqRgyZAjGjBmD999/Hzk5OXj55ZcjfWgUgD59+uCWW27B8OHDMXr0aLzyyisYPXo0nnvuuUgfml933HEHNm3ahLfffjvSh+IlLjMvd999N2644Qaf23Tv3r3l50OHDuGcc87B6NGj8Y9//MPn4/Ly8tDU1ISKigq37EtZWRny8vLactgBC/b82qp79+7o0KEDduzYgfPOO0+z/arR8/zk16isrAz5+fktt5eVlWHIkCEh7TMUgZ5jXl6eV6Fnc3MzysvLg3q/jRw5EoDILvbo0SPo49VChw4dkJCQ4DUzz9fvTl5eXlDbR1oo5+jJarVi6NCh2LFjhx6HGHZqr2FGRkbMZ13UnHbaafjuu+8ifRg+TZ8+vWUCgL8MXyR+D+MyeMnJyUFOTk5A2x48eBDnnHMOhg8fjldffRVms+9k1fDhw2G1WrF8+XJcfvnlAESV+b59+zBq1Kg2H3sggjk/LRw4cADHjx93+7DXk57nV1RUhLy8PCxfvrwlWKmqqsLq1auDnpHVFoGe46hRo1BRUYG1a9di+PDhAICvv/4aTqezJSAJxIYNGwAgbK+hksTERAwfPhzLly/H5MmTAYi09fLlyzF9+nTFx4waNQrLly/HjBkzWm4rLi4O2+9asEI5R08OhwMbN27EBRdcoOORhs+oUaO8ptVG82uohQ0bNkT0d80XSZLwpz/9CR988AFWrFiBoqIiv4+JyO+hbqXABnDgwAGpZ8+e0nnnnScdOHBAOnz4cMvFdZs+ffpIq1evbrnt1ltvlbp06SJ9/fXX0po1a6RRo0ZJo0aNisQp+LV3715p/fr10rx586S0tDRp/fr10vr166Xq6uqWbfr06SO9//77kiRJUnV1tXTPPfdIJSUl0u7du6WvvvpKGjZsmNSrVy+poaEhUqehKtjzkyRJevzxx6WsrCzpww8/lH755Rfp0ksvlYqKiqT6+vpInIJfkyZNkoYOHSqtXr1a+u6776RevXpJ11xzTcv9nu/RHTt2SA8//LC0Zs0aaffu3dKHH34ode/eXTr77LMjdQot3n77bclms0mLFy+WtmzZIt18881SVlaWVFpaKkmSJP3+97+XZs2a1bL9999/L1ksFunpp5+Wtm7dKs2dO1eyWq3Sxo0bI3UKfgV7jvPmzZO++OILaefOndLatWulq6++WkpKSpI2b94cqVPwqbq6uuX3DID07LPPSuvXr5f27t0rSZIkzZo1S/r973/fsv2uXbuklJQU6S9/+Yu0detWaeHChVJCQoL0+eefR+oUfAr2/J577jlp6dKl0vbt26WNGzdKf/7znyWz2Sx99dVXkToFn2677TYpMzNTWrFihdtnXl1dXcs20fB7yODFh1dffVUCoHiR7d69WwIgffPNNy231dfXS7fffruUnZ0tpaSkSJdddplbwBNNrr/+esXzcz0fANKrr74qSZIk1dXVSRMmTJBycnIkq9Uqde3aVZo2bVrLH95oE+z5SZKYLv3ggw9Kubm5ks1mk8477zxp27Zt4T/4AB0/fly65pprpLS0NCkjI0O68cYb3YIzz/fovn37pLPPPltq166dZLPZpJ49e0p/+ctfpMrKygidgbsXXnhB6tKli5SYmCiddtpp0g8//NBy35gxY6Trr7/ebft3331X6t27t5SYmCgNGDBAWrZsWZiPOHjBnOOMGTNats3NzZUuuOACad26dRE46sDIU4M9L/I5XX/99dKYMWO8HjNkyBApMTFR6t69u9vvY7QJ9vyeeOIJqUePHlJSUpLUrl07aezYsdLXX38dmYMPgNpnnutrEg2/h6aTB0tEREQUE+JythERERHFLgYvREREFFMYvBAREVFMYfBCREREMYXBCxEREcUUBi9EREQUUxi8EBERUUxh8EJEREQxhcELERERxRQGL0RERBRTGLwQERFRTGHwQkRERDHl/wONv9TU36sHvgAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Plot f(x) + contours\n",
        "x = np.linspace(-2, 2, 400).reshape(-1, 1)\n",
        "fx = [f(x_i, noise_level=0.0) for x_i in x]\n",
        "plt.plot(x, fx, \"r--\", label=\"True (unknown)\")\n",
        "plt.fill(np.concatenate([x, x[::-1]]),\n",
        "         np.concatenate(([fx_i - 1.9600 * noise_level for fx_i in fx],\n",
        "                         [fx_i + 1.9600 * noise_level for fx_i in fx[::-1]])),\n",
        "         alpha=.2, fc=\"r\", ec=\"None\")\n",
        "plt.legend()\n",
        "plt.grid()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-Mwt88C6gCy5"
      },
      "source": [
        "Bayesian optimization based on gaussian process regression is implemented in\n",
        ":class:`gp_minimize` and can be carried out as follows:\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "LbapI5UNgCy7"
      },
      "outputs": [],
      "source": [
        "from skopt import gp_minimize\n",
        "\n",
        "res = gp_minimize(f,                  # the function to minimize\n",
        "                  [(-2.0, 2.0)],      # the bounds on each dimension of x\n",
        "                  acq_func=\"EI\",      # the acquisition function\n",
        "                  n_calls=15,         # the number of evaluations of f\n",
        "                  n_random_starts=5,  # the number of random initialization points\n",
        "                  noise=0.1**2,       # the noise level (optional)\n",
        "                  random_state=1234)   # the random seed"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tvBBdpmKgCy_"
      },
      "source": [
        "Accordingly, the approximated minimum is found to be:\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "YMEigaq_gCzB",
        "outputId": "c10827fc-b6f1-4091-eb18-a604c0178b46"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'x^*=-0.3552, f(x^*)=-1.0079'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 7
        }
      ],
      "source": [
        "\"x^*=%.4f, f(x^*)=%.4f\" % (res.x[0], res.fun)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wDubnNTTgCzE"
      },
      "source": [
        "For further inspection of the results, attributes of the `res` named tuple\n",
        "provide the following information:\n",
        "\n",
        "- `x` [float]: location of the minimum.\n",
        "- `fun` [float]: function value at the minimum.\n",
        "- `models`: surrogate models used for each iteration.\n",
        "- `x_iters` [array]:\n",
        "  location of function evaluation for each iteration.\n",
        "- `func_vals` [array]: function value for each iteration.\n",
        "- `space` [Space]: the optimization space.\n",
        "- `specs` [dict]: parameters passed to the function.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eSOm1qiCgCzF",
        "outputId": "36f147e0-1993-481c-fa76-02568bb1cf39"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "          fun: -1.0079192739252805\n",
            "            x: [-0.355184142786249]\n",
            "    func_vals: [ 3.716e-02  6.739e-03 ...  8.157e-03 -7.976e-01]\n",
            "      x_iters: [[-0.009345334109402526], [1.2713537644662787], [0.4484475787090836], [1.0854396754496047], [1.4426790855107496], [0.957924846874036], [-0.4515808656843847], [-0.6859481178696925], [-0.355184142786249], [-0.2931537772945012], [-0.3209941516716812], [-2.0], [2.0], [-1.3373742108323974], [-0.2478422849815407]]\n",
            "       models: [GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775), GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5) + WhiteKernel(noise_level=0.01),\n",
            "                                        n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                        normalize_y=True, random_state=822569775)]\n",
            "        space: Space([Real(low=-2.0, high=2.0, prior='uniform', transform='normalize')])\n",
            " random_state: RandomState(MT19937)\n",
            "        specs:     args:                    func: <function f at 0x797e839e89d0>\n",
            "                                      dimensions: Space([Real(low=-2.0, high=2.0, prior='uniform', transform='normalize')])\n",
            "                                  base_estimator: GaussianProcessRegressor(kernel=1**2 * Matern(length_scale=1, nu=2.5),\n",
            "                                                                           n_restarts_optimizer=2, noise=0.010000000000000002,\n",
            "                                                                           normalize_y=True, random_state=822569775)\n",
            "                                         n_calls: 15\n",
            "                                 n_random_starts: 5\n",
            "                                n_initial_points: 10\n",
            "                         initial_point_generator: random\n",
            "                                        acq_func: EI\n",
            "                                   acq_optimizer: auto\n",
            "                                              x0: None\n",
            "                                              y0: None\n",
            "                                    random_state: RandomState(MT19937)\n",
            "                                         verbose: False\n",
            "                                        callback: None\n",
            "                                        n_points: 10000\n",
            "                            n_restarts_optimizer: 5\n",
            "                                              xi: 0.01\n",
            "                                           kappa: 1.96\n",
            "                                          n_jobs: 1\n",
            "                                model_queue_size: None\n",
            "                                space_constraint: None\n",
            "               function: base_minimize\n"
          ]
        }
      ],
      "source": [
        "print(res)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sFgixQtAgCzF"
      },
      "source": [
        "Together these attributes can be used to visually inspect the results of the\n",
        "minimization, such as the convergence trace or the acquisition function at\n",
        "the last iteration:\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 473
        },
        "id": "98nm0gtWgCzH",
        "outputId": "6c65487a-dfa6-4545-b52a-165062eb9f8f"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "from skopt.plots import plot_convergence\n",
        "plot_convergence(res);"
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.5"
    },
    "colab": {
      "provenance": []
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}