Telemarketing_Prediction.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "TPU",
    "colab": {
      "name": "Telemarketing Prediction.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "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": "3.7.4"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "o-t61xaaNaaA"
      },
      "source": [
        "# **Telemarketing Prediction with Resampling and ML**\n",
        "\n",
        "Authors: Hongtianxu Hua, Bongju Yoo, Yonghao Li"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VZL2Fx1yi4NM"
      },
      "source": [
        "!pip install pycaret==2.0\n",
        "!pip install pandas-profiling\n",
        "!pip install -U imbalanced-learn"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3nTyc1OAi9YL",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        },
        "outputId": "189c4765-32aa-41cf-bfb4-644e9d704f8f"
      },
      "source": [
        "# Load libraries\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from matplotlib.legend_handler import HandlerLine2D\n",
        "import seaborn as sns\n",
        "from scipy import interp\n",
        "\n",
        "from collections import Counter\n",
        "from imblearn.over_sampling import ADASYN, SMOTE, RandomOverSampler\n",
        "from imblearn.pipeline import make_pipeline\n",
        "from imblearn.combine import SMOTEENN, SMOTETomek\n",
        "\n",
        "from sklearn.model_selection import train_test_split, KFold, StratifiedKFold, cross_val_score, GridSearchCV\n",
        "from sklearn.feature_selection import RFECV\n",
        "from sklearn.metrics import accuracy_score, roc_curve, auc, roc_auc_score, classification_report \n",
        "from sklearn.decomposition import PCA\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.naive_bayes import GaussianNB\n",
        "from sklearn.svm import SVC, LinearSVC\n",
        "\n",
        "import pandas_profiling as pp\n",
        "from pycaret.classification import *\n",
        "\n",
        "%matplotlib inline"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "84ei6u2HLE92"
      },
      "source": [
        "# Load dataset\n",
        "bm = pd.read_csv(\"https://raw.githubusercontent.com/georgehua/global-historial-crop-yield/master/bank_marketing_scikit_label_dataset.csv\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a0mWy3ZWzKx2"
      },
      "source": [
        "## 1. **Introduction**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Midl4nEFzO8T"
      },
      "source": [
        "A person’s financial decisions and banking choices depend on multiple variables. In this dataset collected using direct marketing campaigns (phone calls) from a Portuguese banking institution to its clients or potential clients, there are various features for exploration and analysis. Our classification goals include: to predict whether a client will subscribe to a term deposit (variable y); to predict whether a client has credit in default account (variable “default”); and to predict whether a client has any personal loan (variable “loan”).\n",
        "\n",
        "Our main objective is to analyze this dataset in order to help increase the effectiveness of the bank’s telemarketing campaign.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UwyJlFvjzWTQ"
      },
      "source": [
        "## 2. **Data Description**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bbm9oY0GzZG4"
      },
      "source": [
        "The dataset is composed of 17 attributes: Age of client (age), Type of job (job), Marital status (marital), Level of education (education), Existence of credit (default), Average yearly balance (balance), Housing loan (housing), Personal loan (loan), Communication type (contact), Day of the month (day), Month of the year (month), Last contact duration (duration), Number of contacts made (campaign), Number of days passed since a client was last contacted from a previous campaign (pdays), and Predictor class (y). All features have been converted into corresponding numeric values when we downloaded the dataset. The dataset is publicly available in the UCI Machine learning Repository, which can be retrieved from http://archive.ics.uci.edu/ml/datasets/Bank+Marketing#. All the detailed information about features and description of the one-hot encoding process can be found on https://www.mldata.io/dataset-details/bank_marketing/."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4LZEnIfZjBbr",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "outputId": "bcfc3ea9-20cd-487f-832d-e1e7ea334cb2"
      },
      "source": [
        "bm.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>age</th>\n",
              "      <th>job</th>\n",
              "      <th>marital</th>\n",
              "      <th>education</th>\n",
              "      <th>default</th>\n",
              "      <th>balance</th>\n",
              "      <th>housing</th>\n",
              "      <th>loan</th>\n",
              "      <th>contact</th>\n",
              "      <th>day</th>\n",
              "      <th>month</th>\n",
              "      <th>duration</th>\n",
              "      <th>campaign</th>\n",
              "      <th>pdays</th>\n",
              "      <th>previous</th>\n",
              "      <th>poutcome</th>\n",
              "      <th>y</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>30.0</td>\n",
              "      <td>10.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1787.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>19.0</td>\n",
              "      <td>10.0</td>\n",
              "      <td>79.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>33.0</td>\n",
              "      <td>7.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>4789.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>11.0</td>\n",
              "      <td>8.0</td>\n",
              "      <td>220.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>339.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>35.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1350.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>16.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>185.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>330.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>30.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1476.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>6.0</td>\n",
              "      <td>199.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>-1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>59.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>8.0</td>\n",
              "      <td>226.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>-1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    age   job  marital  education  ...  pdays  previous  poutcome    y\n",
              "0  30.0  10.0      1.0        0.0  ...   -1.0       0.0       3.0  0.0\n",
              "1  33.0   7.0      1.0        1.0  ...  339.0       4.0       0.0  0.0\n",
              "2  35.0   4.0      2.0        2.0  ...  330.0       1.0       0.0  0.0\n",
              "3  30.0   4.0      1.0        2.0  ...   -1.0       0.0       3.0  0.0\n",
              "4  59.0   1.0      1.0        1.0  ...   -1.0       0.0       3.0  0.0\n",
              "\n",
              "[5 rows x 17 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qMyXMpIOYJff",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 606
        },
        "outputId": "a3e3cbb0-ef9b-4d0d-d5a3-e15b35f1d22a"
      },
      "source": [
        "# Feature plots\n",
        "plt.style.use('seaborn-whitegrid')\n",
        "bm.hist(bins=20, figsize=(14, 10), color='skyblue')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzkAAAJNCAYAAAAS1rt6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzde1xUdf4/8NfMwEAoiJhja5nrdtMCJclcvINyU3fFC3JZLFcrXa2kdVMitjQrEcX1Emari/DVTNbRNdZ1xSwsTSSFfiZ2UWvbEEFBhovcwc/vDx6clWAGGGY4M8Pr+Xj4eMhnODPvz5l5c+Zzzue8PwohhAAREREREZGNUModABERERERkSlxkENERERERDaFgxwiIiIiIrIpHOQQEREREZFN4SCHiIiIiIhsCgc5RERERERkUzjIISKrlJWVBT8/v05t4+vri3PnzpkpIiIy5Pr165g+fbrcYRBZlPPnz+Pbb781evsffvgBZ8+eNWFEtoODHCIiIjK7AQMG4PDhw3KHQWRRDhw4gO+++87o7Y8fP85Bjh52cgdAnbN//34kJSWhsbER/fv3R3x8PO6++26sWLECOTk5eOihh/Doo4+iuLgYcXFxKCwsxKpVq/Cf//wHABATE4OJEyfK3Asi01m3bh0++eQTKBQKvP322xg2bBheeeUVfPPNN6ivr0dAQABWrlzZaru2cunee+/FwYMHceLECfTu3RvZ2dlQqVTYvHkzHnroIZSUlCAmJgaXL1+Gk5MTVq5ciXHjxqG8vBxr1qzBV199hYaGBixZsgSzZ8+WYW8QtXbo0CG8++67AIDhw4fjrbfewocffmjw829vb4/s7GwMGTIES5cuxYYNG/DTTz9h2bJlCA0NxdatW5Gfnw+dTodLly5hwIABSExMRL9+/fDDDz/g1VdfRWlpKRoaGrBs2TJMnz4dV69ehb+/P77++mvU1tbqPW7NmzcPvr6+OHbsGK5evYpRo0YhISEBCoVC5j1J9D9t5dXHH3+MxMRENDQ0QKPR4M0338T999+PrVu3QqfT4fr16/j222/Rt29fbNu2DR9//DE+/PBDfPLJJygpKcHTTz+NNWvW4PTp06ivr4eXlxfefvtt2Nvbt3n8qaurw3vvvQd7e3uUl5cjOjpa5r1iYQRZjeLiYuHu7i4KCgqEEEJER0eLmJgYsXv3bhEWFibq6+vF1atXhbe3t1i5cqUQQoinnnpK/OUvfxFCCPHjjz+KJ598UpSUlMjWByJTOXPmjBg2bJg4fPiwEEKI1NRUMWPGDPG3v/1NPPPMM+L27duitLRUPPnkk+Ls2bNCCCF8fHzE2bNn9eaSEEIcOHBAjBgxQly4cEEIIcSqVavEq6++KoQQIiYmRsTHxwshhLh48aJ48sknRW1trXjllVfEihUrRGNjo7h586aYOHGi+O6777p1fxC1JS8vT/z6178WhYWF4vbt22Lp0qXivffeM/j59/T0FD/88IOora0V48ePF4sWLRINDQ3ik08+ERMmTBBCCLFlyxbx+OOPi59++kkIIcSf/vQn8dZbbwkhhFi0aJF47733hBBCfPHFF2L48OGirq5O5OXliWHDhgkhhMHjVmRkpIiMjBTV1dWisrJSeHt7i3PnznXfTiNqR1t5tXXrVuHl5SV+/PFHIYQQf/vb38TTTz8thGjKF29vb3H16lVx+/Zt8dxzz4lt27YJIZo+74cOHRJCCHH06FExffp0UVdXJ2pqakRQUJD0mL7jz8qVK0ViYmI37wHrwOlqVqRfv37Izs7GPffcAwB44oknkJeXh3PnziEgIAB2dna49957pSs1VVVVyMrKwvz58wEAgwcPhpeXFz799FO5ukBkUg4ODggKCgIABAUF4ZtvvsHvfvc7bNu2DQqFAn369MFDDz2Eq1evtthOXy41e+CBB+Du7g4AePTRR1FQUAAA+PTTT6V7Ch599FF8/PHHUKvVyMjIwFNPPQWlUgk3Nzf4+fnh2LFjZu8/UXs+//xzPP744xgwYAAUCgUSEhKwYMECg5//Bx98EEOGDIFarcbgwYMxbtw4qFQqPPzww7hx44b0e6NHj8agQYMAAP7+/vjyyy8BANu2bcPChQsBAF5eXqitrUVRUVGLuPQdt5oFBgbC0dERTk5O+OUvfynlIJElaCuv7r77bowePRqDBw8GAISEhCArKwsNDQ0AmvLs3nvvhUKhwLBhw9r8TAcEBODAgQOwt7eHg4MDPDw8pNzUd/wh/ThdzYo0NjZiy5Yt+OSTT9DY2IjKykoMGTIE5eXlcHV1lX5vwIABKCwsREVFBYQQCAsLkx6rqqrCr3/9aznCJzI5V1dXKJVN52p69+4NALhw4QJ27tyJH374AUqlEoWFhZg1a1aL7fTlUjNnZ2fp/yqVCo2NjQCA0tLSFo81v2ZFRQWioqKgUqkAALW1tQgMDDRDj4k6R6fTwcXFRfrZwcGh3c9/r169pP+rVCo4OTlJ/799+7b02J3HHRcXF5SXlwMATp48iXfffRc6nQ4KhQJCiBbbAdB73GrWnFvNr9ucg0SWoK28Ki8vb9Hm7OwMIQR0Op30czN9n+mSkhKsWbMGX3/9NRQKBYqLi/H0008D0H/8If04yLEiR44cwSeffII9e/bAzc0Nf//73/HPf/4TvXv3RmVlpfR7zWfM+vXrB5VKhQMHDrQ4aBHZirKyMun/zV+wNm/eDE9PTyQmJkKlUrUY5DfTl0vtcXV1hU6nw3333QcAuHr1KgYMGACNRoPExEQ8/PDDJuoZkWn07dtXusICALdu3cJHH31k1Of/55q/vAFNudinTx/U19cjKioKmzZtwsSJE1FXV4fhw4e32lbfcYvIGrSVV0DTQKRZWVkZlEol+vbt2+Hn/ctf/gI7Ozv885//hFqtxvLly6XH9B1/SD9OV7MiN2/exL333gs3NzfodDr8+9//RmVlJTw8PHDs2DHcvn0bBQUF+OyzzwAAdnZ2mDhxIvbt2wcAqK6uxiuvvMLL/mQzampq8NFHHwEA0tPT4eHhgfLycgwbNgwqlQqff/45/vvf/6KqqqrFdvpyqT2+vr74xz/+AQC4cuUKZs2ahcbGRvj6+kp51tDQgLfffhsXL140cW+JOm/ixInIycnB1atXIYTA66+/juvXrxv1+f+57Oxs6XiSnp4OLy8vVFdXo6qqSprumZKSAnt7+1Y5qO+4RWQN2sqruro6nDt3Tppetm/fPowdOxZ2doavJ9jZ2aGiogJA07Hp4Ycfhlqtxrfffosvv/xSyh19x587t6eWOMixItOnT0dpaSn8/PywfPlyREVFobCwEEVFRXBwcMCUKVOwevVqTJs2TapCs2rVKpw9exaBgYGYOXMmBg0ahF/84hcy94TINH71q1/hyy+/RGBgIJKTk/Haa6/hD3/4A9atW4fp06fjiy++wPPPP4+tW7ciOztb2k5fLsXFxRl8vZdffhmFhYXw9fXFSy+9hA0bNsDR0RFRUVGoqKhAQEAApk2bhtu3b+ORRx4xd/eJ2nXPPffgjTfewNNPP42AgAAAwG9+8xujPv8/N2bMGKxevRoTJ07EtWvX8Oyzz8LFxQXPPPMMgoODERwcjPvvvx9TpkzB4sWLUV1dLW0bHh6u97hFZOnayqtnn30Wb775JpYsWYLAwECcPXsWb7zxRrvPNWXKFGzYsAFr167FggULsG/fPgQFBeH999/HypUrsX//fvz73//We/zx8fHBvn378OKLL5q721ZHIYQQcgdBXSeEkA4Q69atQ2NjI2JiYmSOioiIbNHWrVtRWFiIt956y+jn4HGLiMyJV3JswMcff4zZs2ejrq4OlZWV+PTTT+Hp6Sl3WERERG3icYuIzI2FB2zApEmT8OmnnyIoKAhKpRKTJk1iZSciIrJYPG4RkblxuhoRERGRDaiurkZ0dDRu3ryJ2tpaLFmyBEOHDsWKFSvQ2NiI/v37Y/369VCr1UhLS0NKSgqUSiXmzp2LkJAQ1NfXIzo6GteuXYNKpcLatWultZCIrA0HOUREREQ24MiRI8jPz8ezzz6L/Px8LFiwACNHjsSECRMQFBSEjRs34p577kFwcDBmzpwJrVYLe3t7zJkzB3v27EFGRga++uorvP766zh16hS0Wi02bdokd7eIjGKR09XurIJE1N28vLzkDsFozB2SE3OHyHimyJ+pU6dK/y8oKMCAAQOQlZWF1atXAwB8fHyQlJSEIUOGwMPDQ1pccuTIkcjJyUFmZiaCg4MBNFXP60ghCOYOyU1f7ljkIAdoP9mzs7Ot5oBqLbFaS5yA+WK1hT/WxuwXS37vLTU2S40LkCc2W8wdS36PTY19lZep8ycsLAyFhYXYvn07fv/730OtVgNoWiS8qKgIxcXFcHNzk37fzc2tVbtSqYRCoUBdXZ20PZE1sdhBDhERERF13r59+/DNN9/g5Zdfxp13Jei7Q6Gz7T9naNBoaYNKxmOYpcUDGI7J0AkClpAmIiIisgG5ubkoKCgAAAwbNgyNjY3o1asXampqAADXr1+HRqOBRqNBcXGxtN2NGzek9qKiIgBAfX09hBBdvorj4ODQpe2JjMUrOV2UeqUMpXWNrdpd1SqEPthHhoiI2mbos/oYD0JEeunLHYB/68mynDt3Dvn5+Xj11VdRXFyMqqoqjB8/Hunp6ZgxYwaOHTuG8ePHY8SIEYiNjUV5eTlUKhVycnIQExODW7du4ejRoxg/fjwyMjIwevToLsd00XEQPv+6pFU7c4fMjYOcLiqta4Su9rbcYRC1i59VIuMwd8hahIWF4dVXX0VERARqamrw2muvwd3dHStXrkRqaioGDhyI4OBg2NvbY/ny5Vi4cCEUCgWWLl0KZ2dnTJ06FadPn0Z4eDjUajXi4uK6HBPzh+Ri9CAnLS0NO3fuhJ2dHV588UU88sgjHa7DTkRERESm5ejoiISEhFbtu3btatUWGBjYagHW5rVxiGyBUYMcnU6HxMREHDhwAFVVVdi6dSvS09MREREh1WHXarUIDg5GYmJiizrsfn5+cHV1NXU/iIiIiIiIABhZeCAzMxPe3t7o3bs3NBoN1qxZg6ysLEyePBlAUx32zMxMnD9/XqrD7ujoKNVhJ+op4uPjERoaitmzZ+PYsWMoKCjAvHnzEBERgWXLlqGurg5A05XR2bNnIyQkBPv37wfQdNPn8uXLER4ejsjISOTl5cnZFSIiIiKrYdSVnKtXr6KmpgaLFy9GeXk5XnjhBVRXV3e4DntHdKRmfHety6CvMkhTf+/Wu11ubi5qa2sBWM8aEtYSJ2D5sZ45cwaXL19GamoqdDodZs6cCW9v7w5f8czIyICLiwsSEhJw6tQpJCQkcOVpIiIiog4w+p6c0tJSvPPOO7h27RqeeuqpLtVhb4slLQaqr7LOfQ72QEO93u3c3d0BWGbN8bZYS5yAdSwGOmrUKAwfPhwA4OLigurqarOvPE1ERERERg5y+vXrh8cffxx2dna4//770atXL6hUKtTU1MDR0dFgHXZPT0+TBd9d9FUG6aNuu6QoEdB0A6eTkxMAQKvVYsKECTh16pTZV55ua6Dm4OAAKAd2ejtLYamxWWpcgGXHRkREZG5GDXLGjRuH6OhoPPvssygrK0NVVRXGjRvX4TrsRD3J8ePHodVqkZSUBH9/f6m9u1eebmudAgBQoGnqZfOVR0tjqVcYLTUuQJ7YOKgiIiJLYtQgZ8CAAQgICMDcuXMBALGxsfDw8OhwHXZLpHdKWi97GaIhW3Hy5Els374dO3fuhLOzM5ycnDp8xbN55emhQ4eabOXptriolfgSA5ChZxDEBduIiIjI2hh9T05YWBjCwsJatHW0Drsl4pQ0MrWKigrEx8cjOTlZKps+ZswYWVee1oeLtZElqa6uRnR0NG7evIna2losWbIEQ4cO7fBabPX19YiOjsa1a9ekdT8GDRokd7eIiKgbGT3IISLDjhw5Ap1Oh6ioKKktLi4OsbGxsq08TWQNMjIy4O7ujmeffRb5+flYsGABRo4cycqERETUYRzkEJlJaGgoQkNDW7Vz5Wkiw6ZOnSr9v6CgAAMGDGBlQiIi6hQOcoiIyCKFhYWhsLAQ27dvx+9//3tZKhPm5ua2W5nwzjXRrFlPKh7Rk/pK1FNxkENERBZp3759+Oabb/Dyyy93aS02YysTZmdnw93dXW9lwmaWWpmwMyy5WqCpWWJfOegiMj2l3AEQERHdKTc3FwUFBQCAYcOGobGxEb169UJNTQ0AGKxM2NxeVFQEAGatTEhERJaLgxwiIrIo586dQ1JSEgCguLgYVVVVUmVCAC0qE164cAHl5eWorKxETk4OnnjiCYwdOxZHjx4FALNXJiQiIsvE6WpERGRRwsLC8OqrryIiIgI1NTV47bXX4O7u3uG12FiZkIiIOMghIiKL4ujoiISEhFbtrExIREQdxelqRERERERkUzjIISIiIiIim8JBDhERERER2RQOcoiIiIiIyKaw8AARERGRDYmPj0d2djYaGhqwaNEieHh4YMWKFWhsbET//v2xfv16qNVqpKWlISUlBUqlEnPnzkVISAjq6+sRHR2Na9euSUU8Bg0aJHeXiDqNgxwiIiIiG3HmzBlcvnwZqamp0Ol0mDlzJry9vREREYGgoCBs3LgRWq0WwcHBSExMhFarhb29PebMmQM/Pz9kZGTAxcUFCQkJOHXqFBISErBp0ya5u0XUaRzkyCj1ShlK6xpbtbuqVQh9sI8MEREREZE1GzVqFIYPHw4AcHFxQXV1NbKysrB69WoAgI+PD5KSkjBkyBB4eHjA2dkZADBy5Ejk5OQgMzMTwcHBAIAxY8YgJiZGno4QdREHOTIqrWuErva23GEQERGRjVCpVHBycgIAaLVaTJgwAadOnYJarQYA9OvXD0VFRSguLoabm5u0nZubW6t2pVIJhUKBuro6afu2ZGdnt9nu4OAAKAfq3S43Nxe1tbWd7mNX6YtXLoynfcbExEEOERERkY05fvw4tFotkpKS4O/vL7ULIdr8/c6238nLy0vvY59/XaL3MXd393af29Sys7MNxtvdGE/7DMVkaPDD6mpmorjj/w4ODrLFQURERD3LyZMnsX37duzYsQPOzs5wcnJCTU0NAOD69evQaDTQaDQoLi6Wtrlx44bUXlRUBACor6+HEMLgVRwiS9WlQU5NTQ2mTJmCgwcPoqCgAPPmzUNERASWLVuGuro6AEBaWhpmz56NkJAQ7N+/3yRBWwMXtRKpV8rw3tcl+Fw5EO99XdLi37/+WyF3iERERGRjKioqEB8fj/feew+urq4Amu6tSU9PBwAcO3YM48ePx4gRI3DhwgWUl5ejsrISOTk5eOKJJzB27FgcPXoUAJCRkYHRo0fL1heirujSdLV3330Xffo03SC/ZcuWDlfuaE46W2fonps+6tYFB4iIiIi64siRI9DpdIiKipLa4uLiEBsbi9TUVAwcOBDBwcGwt7fH8uXLsXDhQigUCixduhTOzs6YOnUqTp8+jfDwcKjVasTFxcnYGyLjGT3I+f7773HlyhVMmjQJADpVucPX17frkRMRERFRC6GhoQgNDW3VvmvXrlZtgYGBCAwMbNHWvDYOkbUzepCzbt06/PnPf8ahQ4cAANXV1R2u3NERHamiYKrqD+1V/5ADK460z5piJSIiIqLuY9Qg59ChQ/D09NS7Am5XKnQ0a6+yg6mrPxiq/iEHVhwxzFyxcuBEREREZP2MGuScOHECeXl5OHHiBAoLC6FWq6XKHY6OjgYrd3h6eposeCIisk3x8fHIzs5GQ0MDFi1aBA8PD6xYsQKNjY3o378/1q9fD7VajbS0NKSkpECpVGLu3LkICQlBfX09oqOjce3aNWnqjb6TckREZJuMGuRs2rRJ+v/WrVtx77334ssvv0R6ejpmzJjRonJHbGwsysvLoVKpkJOTI9vKualXylBa1/bN/vf1su/maIiISJ8zZ87g8uXLSE1NhU6nw8yZM+Ht7d3h4jYZGRlwcXFBQkICTp06hYSEhBbHLSIisn0mWyfnhRdewKFDhxAREYHS0lIEBwfD0dFRqtzx+9//XqrcIYfmSmdt/auoZ6UzIiJLMWrUKGzevBkA4OLigurqamRlZWHy5MkAmorbZGZm4vz581JxG0dHR6m4TWZmJvz8/AA0lc7NycmRrS9ERCSPLpWQBpoGN806WrmDiIhIH5VKBScnJwCAVqvFhAkTcOrUqQ4Xt7mzXalUQqFQoK6urt0FDdu6Jy83N7fdwjRyFYoxtZ50T2JP6itRT9XlQQ4REZE5HD9+HFqtFklJSfD395faO1vcpqNFb35ezCQ7Oxvu7u7tFqaRo1CMqVlT4ZmussS+ctBFZHomm65GpqOQOwAiIpmdPHkS27dvx44dO+Ds7CwVtwFgsLhNc3vzcgX19fUQQrR7FYeIiGwLr+RYIBe1Um+hBFe1CqEP9pEhKiKi7lFRUYH4+HgkJyfD1dUVQNO9NR0tbnPr1i0cPXoU48ePR0ZGBkaPHi1zj4iIqLtxkGOhmgsldJahKnIcIBGRNThy5Ah0Oh2ioqKktri4OMTGxiI1NRUDBw5EcHAw7O3tpeI2CoVCKm4zdepUnD59GuHh4VCr1YiLi5OxN0REJAcOcmyMsYMjMo9Lly5hyZIlmD9/PiIjI1FQUMC1PojaERoaitDQ0FbtHS1u05wvRETUc/GeHCIzqaqqwpo1a+Dt7S21bdmyBREREdi7dy8GDx4MrVaLqqoqJCYmIjk5Gbt370ZKSgpKS0tx+PBhuLi44IMPPsDixYuRkJAgY2+IiIiIrAcHOURmolarsWPHDmg0GqmNa30QERERmR+nqxGZiZ2dHezsWqZYdXW1LGt9ODg4tLvWhyFyrwNiqeVVLTUuwLJjIyIiMjcOcqwMy0vbju5a66NZe2t9GCLnOiCWuKYFYLlxAfLExkEVEREZo72iWY85OBj1vBzkWBlD5aXv62UvQ0TUGc1rfTg6Ohpc68PT01Na62Po0KFc64OIiIhskrmKZvGeHCvU/GH4+b+K+rZHwWQ5mtf6ANBirY8LFy6gvLwclZWVyMnJwRNPPIGxY8fi6NGjAMC1PoiIiIg6gVdyiMwkNzcX69atQ35+Puzs7JCeno4NGzYgOjqaa30QERERmREHOURm4u7ujt27d7dq51ofRERERObF6Wo9CIsWEBEREVFPwCs5PYihogVdqV5BREREluPSpUtYsmQJ5s+fj8jISBQUFGDFihVobGxE//79sX79eqjVaqSlpSElJQVKpRJz585FSEgI6uvrER0djWvXrkkzCgYNGiR3l4g6jVdyehh9RQv0le4jIiIi61FVVYU1a9bA29tbatuyZQsiIiKwd+9eDB48GFqtFlVVVUhMTERycjJ2796NlJQUlJaW4vDhw3BxccEHH3yAxYsXIyEhQcbeEBmPgxwiIiIiG6FWq7Fjxw5oNBqpLSsrC5MnTwYA+Pj4IDMzE+fPn4eHhwecnZ3h6OiIkSNHIicnB5mZmfDz8wPQVBE0JydHln4QdZXR09Xi4+ORnZ2NhoYGLFq0CB4eHh2+FEpEREREpmdnZwc7u5Zf76qrq6V11vr164eioiIUFxfDzc1N+h03N7dW7UqlEgqFAnV1dQbXadO3GLCDgwOgHKh3u9zcXNTW1na4b6ZiaYsX9+R42vuMNDMmJqMGOWfOnMHly5eRmpoKnU6HmTNnwtvbGxEREQgKCsLGjRuh1WoRHByMxMREaLVa2NvbY86cOfDz84Orq6sxL0tEREREXSCEMEn7nby8vPQ+9vnXJXofc3d3b/e5TS07O9tgvN2N8Rj+jDTTF5OhwY9R09VGjRqFzZs3AwBcXFxQXV3dqUuhREREhly6dAlTpkzBnj17AAAFBQWYN28eIiIisGzZMtTV1QEA0tLSMHv2bISEhGD//v0AgPr6eixfvhzh4eGIjIxEXl6ebP0gsgROTk6oqakBAFy/fh0ajQYajQbFxcXS79y4cUNqLyoqAtCUS0IIg1dxiCyVUVdyVCoVnJycAABarRYTJkzAqVOnOnwptCM6clmqo5euOnopjCzvkqkh1hQrEXWcoRunOzJbICMjAy4uLkhISMCpU6eQkJCATZs2ydgjInmNGTMG6enpmDFjBo4dO4bx48djxIgRiI2NRXl5OVQqFXJychATE4Nbt27h6NGjGD9+PDIyMjB69Gi5wycySpdKSB8/fhxarRZJSUnw9/eX2rtyybNZe5fKOns5rSOXwqj9/W4pzHU5lQMnIvk13zi9Y8cOqS0rKwurV68G0DRbICkpCUOGDJFmCwBoceN0cHAwgKYvdzExMd3fCSKZ5ObmYt26dcjPz4ednR3S09OxYcMGREdHIzU1FQMHDkRwcDDs7e2xfPlyLFy4EAqFAkuXLoWzszOmTp2K06dPIzw8HGq1GnFxcXJ3icgoRg9yTp48ie3bt2Pnzp1wdnaWLoU6OjoavBTq6elpksCJiMg2yXHjNND2SY7c3Nx2ZwLIdfO0qfWkkzy23Fd3d3fs3r27VfuuXbtatQUGBiIwMLBFW/PaOETWzqhBTkVFBeLj45GcnCwVEejMpVAiIiJjmePGaaD1lezs7Gy4u7u3OxNAjpunTc3Sbn42J0vsqy0PuojkYtQg58iRI9DpdIiKipLa4uLiEBsb26FLoURERJ3RmdkCzTdODx06lDdOExH1UEYNckJDQxEaGtqqvaOXQomIiDqDN04TEVFndKnwABERkanxxmkiIuoqDnKIiMii8MZpIiLqKqMWAyUiIiIiIrJUHOQQEREREZFN4SCHiIiIiIhsCgc5RERERERkUzjIISIiIiIim8JBDhERkZEUcgdARERtsqkS0qlXylBa19iq/b5e9jJEQ0REts5FrdR77HFVqxD6YB8ZoiIiIpsa5JTWNUJXe7tVex9164MPtaQAoFar5Q6DLBDPVBMZpu/YQ0RE8rGpQQ4Zz0WtxJcYgIyvS1o9xrORPRvPVBMREZmHg4OD3CHYLA5ySMKzkaQPPxtERESmd9FxED7nCWaz4CCHiIjIDDjVk4jaw5OI5sNBDhERkRlwqicRkXw4yCEiIjITnqUlIstpzJsAACAASURBVJIH18khIiIiIiKbYrWDHFaj6D6cV0768LNBRERElqhbpqu9/fbbOH/+PBQKBWJiYjB8+HCjn0ua36wc2KIaBRf8NB9D88r7qlWY2868cmO3NTSX/bEeNMg1Zf6YGu85IEtmyblDZMmYO2QLzD7I+eKLL/Df//4Xqamp+P777xETE4PU1FSjn48LfspD/37X/yUXaBp8GrOtoe160sKlps4fczD0PpkCr9qSMSw9d9rLD0N/VztyconIWJaeO0QdZfZBTmZmJqZMmQIAeOCBB1BWVoZbt26hd+/eRj2fq1rVZruzvQoKPYcNQ491ZVu+ZtNjt+oN31Rr6D0ztK2+7e7tZY8fFANx/vuyNh8fcJcdqhoEKupbf0Fwtlch8H7jPntyMGX+GJM77T1u6LF7e9nj6E+32nwfDL1HrR6/axC+ueO9tpT30JIHX5YcW3fpjtwBzJcfhvSyVxqVW4Zypyd9ZnpSX43RXd/bDOWVOVna+8/9Y7ivrmoVUGPc85p9kFNcXIzHHntM+tnNzQ1FRUXtJkt2dnab7Q/q26DawJMZeqwr2/I1ZX3N/voeLwV6GXg8u8jAc1sYY/LHpLnT3uPGvk/tvEfW8h7q29eWwJJj6w6myp3s7Gz9uQOYLT868prG5Jah3OlJn5me1NfOMuVxBzB87JHrbbCk95/7p/2/d7UwLqZuLyEthGj3d7y8vLohEiLr017+MHeI2sbcITIOc4esldmrq2k0GhQXF0s/37hxA/376z1/S0R3YP4QGYe5Q2Qc5g7ZCrMPcsaOHYv09HQAwMWLF6HRaIye10nU0zB/iIzD3CEyDnOHbIXZp6uNHDkSjz32GMLCwqBQKPD666+b+yWJbAbzh8g4zB0i4zB3yFYoREdukiEiIiIiIrISZp+uRkRERERE1J04yCEiIiIiIpvS7SWkjRUfH4/s7Gw0NDRg0aJF8PDwwIoVK9DY2Ij+/ftj/fr1UKvVcocJAKipqcH06dOxZMkSeHt7W2ycaWlp2LlzJ+zs7PDiiy/ikUceschYKysrsXLlSpSVlaG+vh5Lly5F//79sWrVKgDAI488gtWrV8sbpJV6++23cf78eSgUCsTExGD48OFmfb2O5nFaWhpSUlKgVCoxd+5chISEoL6+HtHR0bh27RpUKhXWrl2LQYMG4dtvvzXZZ6EjuStHbB3NVbn2m63p7ryQgzUdU7vKWo7J1s5Q3pw+fRobN26ESqXChAkTsHTpUlnj8fX1xT333AOVqmkRyg0bNmDAgAFmj+nSpUtYsmQJ5s+fj8jIyBaPybGPDMUjxz76+d8lf39/6TGj9o+wApmZmeKZZ54RQghRUlIiJk6cKKKjo8WRI0eEEEIkJCSI999/X84QW9i4caOYNWuWOHDggMXGWVJSIvz9/UVFRYW4fv26iI2NtdhYd+/eLTZs2CCEEKKwsFAEBASIyMhIcf78eSGEEH/84x/FiRMn5AzRKmVlZYnnnntOCCHElStXxNy5c836eh3N48rKSuHv7y/Ky8tFdXW1mDZtmtDpdOLgwYNi1apVQgghTp48KZYtWyaEECb9LLSXu3LE1tFclXO/2ZLuzgs5WNsxtaus4Zhs7drLm6CgIHHt2jXR2NgowsPDxeXLl2WNx8fHR9y6dcusMfxcZWWliIyMFLGxsWL37t2tHu/ufdRePN29j9r6u3QnY/aPVUxXGzVqFDZv3gwAcHFxQXV1NbKysjB58mQAgI+PDzIzM+UMUfL999/jypUrmDRpEgBYbJyZmZnw9vZG7969odFosGbNGouNtW/fvigtLQUAlJeXw9XVFfn5+dJZGUuK1ZpkZmZiypQpAIAHHngAZWVluHXrltler6N5fP78eXh4eMDZ2RmOjo4YOXIkcnJykJmZCT8/PwDAmDFjkJOTg7q6OpN9FjqSu3LE1tFclWu/2Zruzgs5WNMxtaus5Zhs7QzlTV5eHvr06YNf/OIXUCqVmDhxotn3uyXmsVqtxo4dO6DRaFo9Jsc+MhSPHNr6u9TY2AjA+P1jFYMclUoFJycnAIBWq8WECRNQXV0tXWLu168fioqK5AxRsm7dOkRHR0s/W2qcV69eRU1NDRYvXoyIiAhkZmZabKzTpk3DtWvX4Ofnh8jISKxYsQIuLi7S45YUqzUpLi5G3759pZ/d3NzMuh87msfFxcVwc3NrFded7UqlEgqFAsXFxSb7LHQkd+WIraO5Ktd+szXdnRdysKZjaldZyzHZ2hnKm6Kiojb/NskVT7PXX38d4eHh2LBhA0Q3FBq2s7ODo6Njm4/JsY8MxdOsO/dRW3+XmqfKGbt/rGKQ0+z48ePQarV47bXXWrR3x4ezIw4dOgRPT08MGjSozcdNHee5c+fg6+tr9PalpaV45513EBcXh1deeaVFfJayTwHgww8/xMCBA/HRRx8hJSUFL7/8covHLSlWa9Zd+7GzedyZ9ua2juTG+++/j7Fjx+Ldd981KnezsrLw/vvvAwDq6uqQlZXVodiMYUyuGrPfqDVb3jfNufjb3/5WuloOdL3PR44ckc6ar1ixAp988kmXns9Y3X1Mpv+xtH3783hefPFFvPLKK9i9ezcuX74sLX5qrFdffRVbt27t0nO0pbi4GB9//DEA4KuvvsLChQtN/hr6mHofdZS+7wjGsJrCAydPnsT27duxc+dOODs7w8nJCTU1NXB0dMT169ct4nLbiRMnkJeXhxMnTqCwsBBqtdoi4wSazmA9/vjjsLOzw/33349evXpBpVJZZKw5OTkYN24cAGDo0KGora1FQ0OD9LglxWpNNBoNiouLpZ9v3LiB/v37m/U1O5LHbcXl6ekJjUaDoqIiDB06FPX19RBCoH///i2+nHXms3Ds2DFERUUhJCQEUVFRHcrdn8fW2Ngovd5nn32Gp59+2iSx3amjudpd+83WyZEXcrgzF69cuQKFQmGyv/9btmzByJEj0bt3b8THx5sw6s6xpmOytTOUNz9/rDv2e3t5HBwcLP1/woQJuHTpEgIDA80akyH69lFWVhZOnz6NyZMnY/jw4fjb3/7WbTHJsY9+/h2hmbGfIau4klNRUYH4+Hi89957cHV1BdA0t7x5VHns2DGMHz9ezhABAJs2bcKBAwfw97//HSEhIViyZInJ49y2bRsmTpyI4OBgnD59GkDT5feoqCgEBATA19cX69atA9B0pnrRokXStrdv38aYMWPwzTffYNy4cThz5gxu374NnU6Hqqoqi9ynADB48GCcP38eAJCfn49evXrhgQcewLlz5wBYVqzWZOzYsdL7ffHiRWg0GvTu3dtsr9fRPB4xYgQuXLiA8vJyVFZWIicnB0888QTGjh2Lo0ePAgAyMjIwevRo2Nvb41e/+hXOnTuHbdu24fXXX8fx48el3Kirq8Obb74p5cb27dsBNFVw+X//7/9h8+bN2Lp1q1RxrKysDFVVVfjlL38pxebr64s9e/ZIsR09ehSfffYZampqUFtbi969e+PixYvIysrCSy+91GZsd/avszqaq8but67EZou6Oy/kUFFRgZUrV+L69euYP38+Tp8+DQcHByxYsADbtm2TPg/R0dHYtm0bgKZKS++88w4CAgJw7do1/PDDDwgPD0dQUBD8/Pxw+PBhAMArr7yC//znP5g3bx7OnTuHefPm4cMPPwTQdPVz5syZCAwMREhICC5cuAAAOHjwIF588UXExMQgICAAU6dOxeXLl7vcz+44JlMTQ3lz33334datW7h69SoaGhqQkZGBsWPHyhZPRUUFFi5ciLq6OgDA2bNn8dBDD+l9ruPHj+M3v/kNJk+ejAULFqCkpAQ6nQ4LFiyAr68vnnvuOVRUVEi//8gjj6CwsLDNn//617/i//7v/5CYmIi1a9dKV5g+/PBDXLx4EZMmTcJzzz2H48ePQ6PR4I033kB6ejpeeuklZGVlSfdX1tbW4rXXXkNAQACCgoIQFxcn3b/i6+uLffv2Yc6cORg3bhzi4uI6vf86u49Moa3vCM2M/QxZxZWcI0eOQKfTISoqSmqLi4tDbGwsUlNTMXDgwBYjTkvywgsvYOXKlSaJ88qVK0hOTsaRI0fQt29fvPjiiwCADz74AJWVlTh69CjKy8vh7++PyZMnIzAwEPHx8dDpdOjbty9ycnLg4uKCYcOGAQACAgIwd+5cAEBsbCw8PDxMFqsphYaGIiYmBpGRkWhoaMCqVavQv39/vPbaa7h9+zZGjBiBMWPGyB2m1Rk5ciQee+wxhIWFQaFQ4PXXXzfr63U0j+3t7bF8+XIsXLgQCoUCS5cuhbOzM6ZOnYrTp08jPDwcarVa+sMdExODl19+Gd999x3mzJmD119/XcqNHTt24MqVK/jnP/+JhoYG/O53v5PKL1+4cAFz5szBjBkzkJSUJOXQhg0bsHfvXqxfvx7vv/8+ioqKUFFRIcXWp08f6bX79OmDUaNGYf78+fjXv/6FwsJCvP/++y1i6+rndMCAAR3KVWP2G3Oote7OCzkkJyejpKQEnp6esLOzw969e+Hs7IyffvoJ+/fvh5eXF4KDg5Gdnd1iu+vXr0tfHBcvXgwfHx8899xzOHv2LJ555hkEBARg7dq1OHjwIHbv3o177rlH2rayshLLli1DYmIivLy8kJ6ejj/96U/497//DaDpSuiePXvw9ttvY/Xq1UhJScGbb75p8r6b8phM/9NW3hw8eBDOzs7w8/PDqlWrsHz5cgDA1KlTMWTIEFnjmTBhAkJDQ+Hg4IBHH31U7xWKvLw8rFixAvv27cPDDz+M9957D6tWrcJ9992Hvn37IikpCVevXsVvf/vbdgcB+/fvxzvvvAM3NzfY29tj//79KC0txcMPP4z3338fiYmJSExMRHZ2Njw9PeHv74/vvvsOhYWFeOutt1pMiU5JSUFhYSH+9a9/oaGhAZGRkTh8+DBmzJgBoGlQkpqaiuLiYkyePBnz589vkY8AkJubi3Xr1iE/Px92dnbSSb377ruvU/vIVNr6jjB69Gg88sgjxn+GjKz0RjLYu3evWLJkifTz0aNHhY+PjxBCiLq6Oqn9d7/7nfjHP/4hhBBiwYIF4oMPPhBCCPH222+LzZs3d2PERN1DX27Mnj1bpKenS+27du0S0dHRQoimEsqHDh2SHtOXQz4+PuLs2bPSY80/nzlzRkyZMkUIIcSWLVtETEyMeTpHZGL68mXlypUiMTFRar/zZx8fH3H8+HHpscbGRtHQ0CD9/5FHHhH5+flCCCEefvhhUVBQIIT4X56dPn1azJgxo0Uco0aNEj/99JM4cOCAmDVrltT+97//XSxYsMDEvSbqvD179ohFixZJP5eWlorHHntMzJo1S3z00UdS+6JFi8SWLVuEEC0//3f+nJCQIN58802pvaqqSsqh2tpaqX3z5s3ScerOY8udx5zw8PAWr5+cnCxefvllIURTrmZkZEiPTZgwQeTk5HRtR1gpq7iSQ03KyspazFFsro70448/Ii4uDj/88AOUSiUKCwsxa9YsAE2VyQ4ePIiwsDB8/PHH0nQdIluiLzcqKiqwdu1abNy4EUDT9LW2FnY0lENEtkZfvrSnT58+0v9PnjyJd999FzqdDgqFAkII3L59W++2JSUlrV7H2dkZN2/elP7fTKVSSVNviORUUVGBc+fOtbiK0bt3b5SWlnY6h3Q6XYv7SO666y4ATbccrF27VrpSU1ZWJpU816ekpKRFPvbp00fKpeYYm/XkfOIgx4q4uLi0mPep0+kAAG+88QYee+wxJCYmQqVSISwsTPodPz8/vPHGG/j0009x11134cEHH+z2uInMTV9uaDQaLFiwAD4+Pga3N5RDSqWyxZe3srIyE0dP1L305UtHP+v19fWIiorCpk2bMHHiRL0nD+7Ur1+/VhXcysrK0K9fP/zwww9d6Q6R2Wg0GowZMwZbtmxp0R4SEtIih0pKSqQqfkqlUhpU3JlDffv2lXIN+F/epaam4scff8TBgwfRq1cv/OUvf8H169cNxnX33Xe3yKfS0lLcfffdRvbSdllF4QFq8vjjjyM7OxslJSVobGxEWloaAODmzZsYNmwYVCoVPv/8c/z3v/9FVVUVgKazY+PHj8fq1asRFBQkZ/hEZqMvNyZPnoz9+/ejsbERQghs27YNn332WavtDeVQ//798e233wJomjNcW1vbans7O7sWBzwiS6YvX+78rOfl5SEnJ6fN7aurq1FVVQV3d3cATfcH2NvbSzljZ2eH8vLyFtsMHz4cxcXF+PLLLwEA//rXv3DPPffgvvvuM0sfiUxh3LhxOHfuHPLy8gA0lXF+88034enpiePHjwMAfvrppxb3r92ZRwcOHIBS2fRV29fXF5988gnKysrQ0NCApUuX4tSpU7h58yZ+9atfoVevXsjPz8enn37aIpfaOrZMmjQJWq0WjY2NqKqqwocffoiJEyeadV9YIw5yrMiwYcMQFhaGmTNnYtasWRg5ciQA4A9/+APWrVuH6dOn44svvsDzzz+PrVu3Skk3bdo05OfnY+rUqXKGT2Q2+nIjIiICAwcOxLRp0xAYGIjvv/8eXl5erbY3lENLlixBcnIypk+fju+//77Nq6Fjx47FmTNnMHv2bLP3lair9OXL3LlzkZ+fD39/fyQkJCAgIKDN7V1cXPDMM88gODgYwcHBuP/++zFlyhQsXrwYVVVVCAwMRFhYGI4cOSJt4+TkhE2bNmHNmjUIDAzE3r17sXHjRigUim7pM5ExNBoN1qxZg6VLlyIoKAhvvPEGpk6dikWLFiE/Px++vr5Ys2YN/P39pW1eeuklrFq1CjNmzMBdd90lTR3z9PTEwoULERwcjGnTpuHRRx/F9OnTERYWhrNnzyIgIEBavDYzMxPJycl6jy3z5s3DPffcg2nTpmH27NmYNGkST2S3QSGEha3YRCb31Vdf4Y033oBWq5U7FCIiIiIis+OVHBvX0NCAxMREzJs3T+5QiIiIiIi6BQc5Nuzrr7+Gn58fNBoNfvvb38odDhERERFRt+B0NSIiIiIisim8kkNERERERDbFItfJubMUH1F3a6v6lrVg7pCcmDtExrPW/GHukNz05Y5FDnKApoCzs7OtNukBWH38gPX3obPx28Ifa339tbT30tLiASwvJmuKh7ljuRi/vDoSv7Xnj6H+9YT3z5JZe/yA8cceTlcjIiIiIiKbwkEOGeTg4CB3CGQifC+JjMPcIeq5mP/Wy2Knq1H3S71ShtK6xpaNyoG4eKUMoQ/2kScoMpmLjoPw+dclrdpd1Sq+v0QG6MsdgPlDZEv4Pci2cJBDktK6Ruhqb8sdBpkJ318i4zB3iHoG5rpt4XQ1IiIiIiKyKRzkEBERERGRTeEgh4iIiIiIbAoHOUREREREZFM4yCEys5qaGkyZMgUHDx5EQUEB5s2bh4iICCxbtgx1dXUAgLS0NMyePRshISHYv38/AKC+vh7Lly9HeHg4IiMjkZeXJ2c3iIiIiKwGBzlEZvbuu++iT5+m0pNbtmxBREQE9u7di8GDB0Or1aKqqgqJiYlITk7G7t27kZKSgtLSUhw+fBguLi744IMPsHjxYiQkJMjcEyIiIiLr0KVBjrFnqIl6iu+//x5XrlzBpEmTAABZWVmYPHkyAMDHxweZmZk4f/48PDw84OzsDEdHR4wcORI5OTnIzMyEn58fAGDMmDHIycmRqxtERGRF4uPjERoaitmzZ+PYsWOcRUA9UpfWyWnrDHVQUBA2btwIrVaL4OBgJCYmQqvVwt7eHnPmzIGfnx9cXV1NEjyRpVu3bh3+/Oc/49ChQwCA6upqqNVqAEC/fv1QVFSE4uJiuLm5Sdu4ubm1alcqlVAoFKirq5O21yc7O7tVm4ODA6AcqHeb3Nxc1NbWdrp/XdVWrHKztJgYDxF1xpkzZ3D58mWkpqZCp9Nh5syZ8Pb27vB3tIyMDLi4uCAhIQGnTp1CQkICNm3aJHe3iDrN6EFOW2eoV69eDaDpDHVSUhKGDBkinaEGIJ2h9vX17XrkRBbu0KFD8PT0xKBBg9p8XAhhkvaf8/LyarNd34rtAODu7t6h5zal7OxsvbHKxdJisqZ4OPghsgyjRo3C8OHDAQAuLi6orq7u1He0zMxMBAcHA2iaRRATEyNPR4i6yOhBTlfOUHdE8wHT2g+c1hK/pZ7pNwW53oMTJ04gLy8PJ06cQGFhIdRqNZycnFBTUwNHR0dcv34dGo0GGo0GxcXF0nY3btyAp6cnNBoNioqKMHToUNTX10MI0e5VHCIi6tlUKhWcnJwAAFqtFhMmTMCpU6fMOougveOsNXwX4vcgy2ZMH4wa5Jj6DHVbvLy8LO4sZmdZW/yWdqbfFDr7HpjyD8Gdl/e3bt2Ke++9F19++SXS09MxY8YMHDt2DOPHj8eIESMQGxuL8vJyqFQq5OTkICYmBrdu3cLRo0cxfvx4ZGRkYPTo0SaLjYiIbNvx48eh1WqRlJQEf39/qd0cswgMHWet6bsQvwdZJmNnERg1yOnqGWqinuqFF17AypUrkZqaioEDByI4OBj29vZYvnw5Fi5cCIVCgaVLl8LZ2RlTp07F6dOnER4eDrVajbi4OLnDJyIiK3Dy5Els374dO3fuhLOzM2cRUI9k1CCnq2eoiXqaF154Qfr/rl27Wj0eGBiIwMDAFm0qlQpr1641e2xERGQ7KioqEB8fj+TkZKnQ05gxYziLgHqcLlVXu1NnzlATEREZcunSJSxZsgTz589HZGQkoqOjcfHiRelL28KFCzFp0iSkpaUhJSUFSqUSc+fORUhICOrr6xEdHY1r165JJwv0Ta8msjVHjhyBTqdDVFSU1BYXF4fY2FjOIqAepcuDHGPOUFub1CtlKK1rbNXuqlYh9ME+MkRERGS7qqqqsGbNGnh7e7do/+Mf/wgfH58Wv8cSuEQthYaGIjQ0tFU7ZxFQT9OlxUB7itK6Ruhqb7f619bAh4iIukatVmPHjh3QaDQGf48L6RIRkT4mm65GpsWrR0TUU9nZ2cHOrvXhac+ePdi1axf69euHP//5zxaxkC5gHaVlrb2ELOMnos7iIMdCNV896ix9gyPAtgZIPaWfRNRkxowZcHV1xbBhw/DXv/4V77zzDh5//PEWvyPHQrqA5ZeWtfYSsj0hfg6CiEyP09VsjL6pdbY2va6n9JOImnh7e2PYsGEAAF9fX1y6dKnNErjNpXGbF55mCVwiop6JgxwiIrJ4L7zwAvLy8gAAWVlZeOihhzBixAhcuHAB5eXlqKysRE5ODp544gmMHTsWR48eBQCWwCUi6qE4XY2IiCxKbm4u1q1bh/z8fNjZ2SE9PR2RkZGIiorCXXfdBScnJ6xduxaOjo6yl8BVmO2ZiYioKzjIISIii+Lu7o7du3e3ag8ICGjVJncJXBe1koViiIgsEAc5REREXWBsoRgiIjIf3pNDREREREQ2pcdcyWHJYSIiIiKinqHHDHI4nYCIiIiIqGfgdDUiIiIiIrIpHOQQEREREZFN4SCHiIiIiIhsCgc5RERERERkUzjIISIiIiIim8JBDhERERER2RQOcoiIiIiIbJCDg4PcIcimx6yTQ0RERETUE6ReKUNpXSOgHIjPvy6R2l3VKoQ+2EfGyLoPBzlERERERDaktK4RutrbcochK05XIyIii3Pp0iVMmTIFe/bsAQAUFBRg3rx5iIiIwLJly1BXVwcASEtLw+zZsxESEoL9+/cDAOrr67F8+XKEh4cjMjISeXl5svWDiIjkwUEOERFZlKqqKqxZswbe3t5S25YtWxAREYG9e/di8ODB0Gq1qKqqQmJiIpKTk7F7926kpKSgtLQUhw8fhouLCz744AMsXrwYCQkJMvaGiIjkwEEOERFZFLVajR07dkCj0UhtWVlZmDx5MgDAx8cHmZmZOH/+PDw8PODs7AxHR0eMHDkSOTk5yMzMhJ+fHwBgzJgxyMnJkaUfREQkH96TQ0REFsXOzg52di0PT9XV1VCr1QCAfv36oaioCMXFxXBzc5N+x83NrVW7UqmEQqFAXV2dtL0+2dnZrdocHBwA5UCj+5Kbm4va2lqjtzeVtvpmTRg/EXUWBzlkcxR3/L8nl04kslVCCJO0/5yXl1eb7XdWJuosd3d3o7c1lezsbL19swY9IX4OgohMj4OcHkTR/q/YBBe1kqUTiWyMk5MTampq4OjoiOvXr0Oj0UCj0aC4uFj6nRs3bsDT0xMajQZFRUUYOnQo6uvrIYRo9yoOERHZFg5yepAWX/5/5r5e9jJEZD6WUjoxPj4e2dnZaGhowKJFi+Dh4YEVK1agsbER/fv3x/r166FWq5GWloaUlBQolUrMnTsXISEhqK+vR3R0NK5duwaVSoW1a9di0KBBcneJSBZjxoxBeno6ZsyYgWPHjmH8+PEYMWIEYmNjUV5eDpVKhZycHMTExODWrVs4evQoxo8fj4yMDIwePVru8ImIqJsZPcjpypc3ko++L/991K0HPtQ1Z86cweXLl5GamgqdToeZM2fC29sbERERCAoKwsaNG6HVahEcHIzExERotVrY29tjzpw58PPzQ0ZGBlxcXJCQkIBTp04hISEBmzZtkrtbRGaXm5uLdevWIT8/H3Z2dkhPT8eGDRsQHR2N1NRUDBw4EMHBwbC3t8fy5cuxcOFCKBQKLF26FM7Ozpg6dSpOnz6N8PBwqNVqxMXFyd0lIiLqZkYNcrr65c3V1dXU/SCyOKNGjcLw4cMBAC4uLqiurkZWVhZWr14NoKlCVFJSEoYMGSJViALQokJUcHAwgKaz2DExMfJ0hKibubu7Y/fu3a3ad+3a1aotMDAQgYGBLdqar3wS9VSXLl3CkiVLMH/+fERGRqKgoICzCKjHMWqQ09Uvb76+viYK37LxpveeTaVSwcnJCQCg1WoxYcIEnDp1yiIrRMlVAcoSb7a1tJgYDxF1hqF1pjiLgHoSowY5VL24VAAAIABJREFUXf3y1hHNB1JTHFA7UgJU35c8Y78cOjg44KLjIL1VeVzVKjxWk2fy1+xKqVN9FGg6K9S8wvjPdfeXY2sr6Xr8+HFotVokJSXB399farekClFyVICyxIpJlhaTNcXDwQ+RZWheZ2rHjh1SG2cRUE/UpcIDpvry1hYvLy+THuDbKwFq6EuesV8OP/+6xODN7+Z6TVNzUSvxJQagVNn6vh1XtQqh7t1frcxcJV1N/UXt5MmT2L59O3bu3AlnZ2dWiCIiIrOSY52p9o6d1nASxFJnPBjDlvrSzJjPkNGDnK58ebM0PaW0clfoK1jQ1X2nr9qbLZR6rqioQHx8PJKTk6X70FghioiI5GSOWQSGTkhb2hVpQyxtxkNX2FJfjJ1FYNQgp6tf3ixNTyqtbGqG9h3Q/mDFUko9m8ORI0eg0+kQFRUltcXFxSE2NpYVooiIqNtwFgH1REYNcrr65c0SWUtpZUu86mTLA5WuCA0NRWhoaKt2VogiIqLuxFkE1BMZNcjp6pc3Mh6vOhEREZE+XGeKqEmXCg/0dHJdVbGWq05ERETUvbjOFFETDnK6gFdViIiIiIgsDwc5XWTsVRVLvLeGiIiIiMgWcJAjE14FIiIiIiIyDw5yZMR7a4iIiIiITM/qBjm2vHgkERHZDk5LJgIcHBzkDoF6KKsb5HBNFiKinikrKwvLli3DQw89BAB4+OGH8cwzz2DFihVobGxE//79sX79eqjVaqSlpSElJQVKpRJz585FSEhIt8draFoyT8xRT3HRcRA+/7qkVTtzgMzN6gY5RETUcz355JPYsmWL9PMrr7yCiIgIBAUFYePGjdBqtQgODkZiYiK0Wi3s7e0xZ84c+Pn5wdXVtdvj5Yk56umYAyQXpdwBEBERGSsrKwuTJ08GAPj4+CAzMxPnz5+Hh4cHnJ2d4ejoiJEjRyInJ0fmSImIqDvxSg5ZJM5lJ6K2XLlyBYsXL0ZZWRmef/55VFdXQ61WAwD69euHoqIiFBcXw83NTdrGzc0NRUVF7T53dnZ2qzYHBwdAOdB0HbhDbm4uamtrzfLcP9dW36wJ4yeizuIghywSS2wT0c/98pe/xPPPP4+goCDk5eXhqaeeQmPj//5GCCHa3E5f+895eXm12d7W/QSm4O7ubpbn/bns7Gy9fbMGPSF+DoKITM9mpqvxzL/taZ7H+/N/FfUssU3UEw0YMABTp06FQqHA/fffj7vvvhtlZWWoqakBAFy/fh0ajQYajQbFxcXSdjdu3IBGo5ErbCIikoHNXMnhmX8iItuWlpaGoqIiLFy4EEVFRbh58yZmzZqF9PR0zJgxA8eOHcP48eMxYsQIxMbGory8HCqVCjk5OYiJiZE7fCIi6kY2M8gBuLgmEZEt8/X1xZ/+9Cd8/PHHqK+vx6pVqzBs2DCsXLkSqampGDhwIIKDg2Fvb4/ly5dj4cKFUCgUWLp0KZydneUOn4iIupFNDXKIiMh29e7dG9u3b2/VvmvXrlZtgYGBCAwM7I6wiIjIAtnMPTlEREREREQABzlERPT/2bvzuKrq/H/gr3uBC4MCBnI1c8kstRGXXHKUQEFZtZHKBRitZqwkt2wwJUKxxcQFEx1mLEoxbSGvZmQGWqGTimRgjNj0LetR4YYgi7LJBT+/P/hxRoS7cLm7r+fj0SPv59xz7/sczvt8zueez+dziIiI7AwbOUREREREZFfYyCEiIjIzPvaAiMi0OPEAERGRmWl77EE3hQNm3ethgaiIiOwHGzlEREQWoOmxB0RE1HnsrkZERERERHaFjRwiIiIrwvE6RESdx+5qZFKsrImIOkbbeB2AY3aIiPTBRg6ZlLbKuncXJwtERERk/Theh4huF7p+1Bni7GzQ57KRQyanqbL2ULR/QBMRERHR7cFUP+pwTA4REZENYTdgIiLdeCeHiIjIhmjrBnyHwgEzbxmv42xgVw8iIltmlkbO66+/jsLCQshkMsTHx2PYsGHm+Foiu8D8ITKMPeeO5m7A7TSA5L1w7PvydhtARO2x59yh24fJGznffPMNfvvtN2RkZODnn39GfHw8MjIyTP21RHaB+UPmpukOQWcGf1rC7Zw7HWoA/X9sAFGL2zl3yL6YvJGTm5uLyZMnAwAGDBiAqqoqVFdXo2vXrgZ9XjeFQ7vlbk4OkGnoqaxtWWfW5Xfa3ndqOn6slTHzR9O2W2qfWGMXGmuLydrisSXmyB3A+s5xupZVq9sf3NvFSY6s36txTd3+hDA9/uCI2kbR7nJty9ycHBDa17D6voWt54GtxW+u6zZrrI9tKVZdbGlbtMXUTeEA1Bv2uSZv5JSVlWHIkCHSa09PT5SWlupMlvz8/Fb/b3GvphXqtHyYtmWdWZffaZPfecshZdUMyZ9bc6aFttyx1D7RFKslWVtM5o5H23FyHda3fzQxS+4AVnmO68wyb03LK4EumpZrWwYgv1TL9+rJVo47TWwpfmPmDmCddY8mthSrLra0LbrOsYbWPWafeEAIofM9o0aNMkMkRLZHV/4wd4jax9whMgxzh2yVyaeQViqVKCsrk15fvnwZ3t4afysiopswf4gMw9whMgxzh+yFyRs5vr6+yM7OBgCcOXMGSqXS4H6dRLcb5g+RYZg7RIZh7pC9MHl3tZEjR2LIkCGIjIyETCZDYmKiqb+SyG4wf4gMw9whMgxzh+yFTOgzSIaIiIiIiMhGmLy7GhERERERkTmxkUNERERERHbF7FNI6+v1119HYWEhZDIZ4uPjMWzYMEuHpJcff/wR8+fPx5NPPonZs2fj4sWLWLZsGZqamuDt7Y3169dDoVBYOkyN1q1bh/z8fDQ2NmLevHkYOnSozcRfV1eHuLg4XLlyBdevX8f8+fMxePBgm4m/s7TlzPHjx7Fx40Y4ODjA398fCxYs0LmOKeM5ceIENm7cCLlcjv79+2P16tU4efIknnvuOdx3330AgIEDB2LFihVmiScwMBA9e/aEg0PzA8k2bNiAHj16mPw8pOnzS0pKsHTpUul9xcXFiI2NhVqtRkpKCvr27QsAGD9+PJ599lmjxXPr+etmljiGrIG1b5++dU5mZiZ27NgBuVyOmTNnYsaMGVCr1YiLi8OFCxfg4OCANWvWoE+fPmaNX986x9ri70h9Y22xm4u1544+tJ0TbcGt+RUcHGzpkPTWXo4FBAR07EOEFcrLyxPPPPOMEEKIs2fPipkzZ1o4Iv3U1NSI2bNni4SEBLFz504hhBBxcXHiwIEDQgghkpOTxXvvvWfJELXKzc0VTz31lBBCiPLycjFhwgSbiv+zzz4Tb731lhBCiHPnzong4GCbir8zdOVMWFiYuHDhgmhqahJRUVHip59+Mmme6frsoKAgcfHiRSGEEIsWLRKHDx8WJ06cEIsWLTJaDB2JJyAgQFRXV3doHVPH1EKtVovIyEhRXV0t9uzZI5KSkowaR4v2zl83M/cxZA2sffv0rXNqampEcHCwuHr1qqirqxNTpkwRFRUVYu/evWLVqlVCCCG+/vpr8dxzz5k1fn3rHGuMX9/6xhpjNwdrzx196DonWrv28suWtJdjHWWV3dVyc3MxefJkAMCAAQNQVVWF6upqC0elm0KhQFpaGpRKpVSWl5eHSZMmAQACAgKQm5trqfB0GjNmDFJSUgAA7u7uqKurs6n4w8PD8fTTTwMALl68iB49ethU/J2hLWeKi4vh4eGBO++8E3K5HBMmTEBubq5J80zXZ+/duxc9e/YE0Pw07YqKCqN8r6HxGGsdU8T08ccfIyQkBF26dDHad7envfNXC0scQ9bA2rdP3zqnsLAQQ4cOhZubG1xcXDBy5EgUFBQgNzcXQUFBAJrvChYUFJg1fn3rHGuMX9/6xhpjNwdrzx19aDsn2oL28qupqcnCUemvvRzrKKts5JSVleGOO+6QXnt6eqK0tNSCEenH0dERLi4urcrq6uqk7lFeXl5WvR0ODg5wdXUFAKhUKvj7+9tU/C0iIyOxdOlSxMfH22T8htCWM6WlpfD09GyzzJR5puuzW565cPnyZRw7dgwTJkwAAJw9exYxMTGIiorCsWPHjBKLPvEAQGJiIqKiorBhwwYIIUx+HtL383fv3o3p06dLr7/55hvMnTsXTzzxBL7//nujxdPe+auFJY4ha2Dt26dvnVNWVqbx79dSLpfLIZPJ0NDQYLb49a1zrDV+QHd9Y82xm5K1544+tJ0TbUF7+dXSJduW3JxjHWW1Y3JuJuxkluv2tuPcuXMIDg426sXKrf7zn/8gJSUF77zzjl7v/+KLL6BSqbBt27ZW/Tdt5e/w4Ycf4r///S9eeOGFVjHbSvzGYMi2mnL/tPfZV65cQUxMDBITE3HHHXfg7rvvxsKFCxEWFobi4mI8/vjjOHjwoFHHULXkwq3xLF68GH5+fvDw8MCQIUPQu3dvvbbBmNr7/FOnTuGee+6RGoXDhw+Hp6cnJk6ciFOnTmH58uX49NNPTRpXR9h7jtna9mmKV9/yvLw8JCQk4NChQ0aP7WYdrXM6Wm5KHa1vrCl2c7L37TOWQ4cO4auvvsKaNWvwyy+/4MqVKxgzZozWdeLi4tC3b1/Mnz+/3eU355ctujnHMjMzIZPJ9F7XKu/kKJVKlJWVSa8vX74Mb29vC0ZkOFdXV9TX1wNoHkxsiduew4YN07uB8/XXX2Pr1q1IS0uDm5ubVcSvr6KiIly8eBEAcP/996OpqQldunSxmfg7Q1vO3LqsZT+YMs90fXZ1dTWefvppLFmyBA899BAAoEePHggPD4dMJkPfvn3RvXt3lJSUGDWelly4NZ6IiAh4eXnB0bH5d59ffvnF5OchfT7/8OHDGDdunPR6wIABmDhxIgDggQceQHl5uVm6H1jiGLIGtrh97Z2z29uOlvKWX9fVajWEEGafmEWfOsca49e3vrHG2M3BFnPHGgQFBWHNmjUAmhsnJ0+e7NTn3ZpftqS9HCsvL+/QZ1hlI8fX1xfZ2dkAgDNnzkCpVEq/ZNqa8ePHS9ty8OBB+Pn5tfs+lUqFhx9+GBMmTMD+/ftx48YNvPHGGwgNDUVoaCji4uJQW1sLoHkmqG+//VZat+V1Y2MjXnrpJYSEhCAoKAgLFy5EdXU18vLypL6/W7ZswSuvvIIFCxZg0qRJmD59Oi5fvgwAOHnyJJ599llUVFRg165dePjhh3HvvffqFb81+Pbbb6VfKsrKylBbW6v3/rd12nKmd+/eqK6uxrlz59DY2IicnBz4+vqaNM90fXZSUhKeeOIJ+Pv7S2WZmZlSY7y0tBRXrlwxqA+utnjy8vIwYcIEdO/eHevWrUNISAhCQkIQEBCAuro66f0VFRX45JNPsHr1arz33nsmOQ/ps/9Pnz6NwYMHS6/T0tKwf/9+AM2z/nh6epql+4EljiFrYIvb1945b/jw4Th9+jSuXr2KmpoaFBQUYPTo0fD19UVWVhYAICcnB2PHjm33M69fv46VK1ciJCQEYWFhSEpKkhrXp06dwqOPPorQ0FCEh4fj+PHjAJp7KTz00EN499138fDDD8PPzw8HDhxo9bnXrl3DunXr8Oabb6Jbt24mi98U9K1vrDF2c7DF3DFEy3GelpYm1SffffcdnnnmGfj5+eHFF18E0NztOCwsDMHBwfjLX/6C8+fPA2gen7pw4UI88cQTWLduHfbu3Ysnn3wSX331Fd588028++67SEpKAgCkpqYiJCQEkydPxrx583D16lWtsbWXX7akvRy7uQukPqyyu9rIkSMxZMgQREZGQiaTITEx0dIh6aWoqAhr167F+fPn4ejoiOzsbGzYsAFxcXHIyMhAr169EBER0Wa9GzduQK1W49NPP0VWVhY2bNgAmUyGf//739i7dy+cnZ2xaNEipKena7wdCQBHjx7FuXPnpBNnSkoKTp061ebXoaysLOzevRu9evVCTEwM9uzZg2effRbLli2Ds7MzevXqhU8++QTFxcWIiYmBSqXSGr+1iIyMxEsvvYTo6GjU19dj5cqV8PHxwfLly20i/s5oL2f27t0LNzc3BAUFYdWqVYiNjQXQPJivf//+6N+/v8nyTFs8Dz30EPbt24fffvsNKpUKADB16lRMmTIFS5cuxZdffgm1Wo1Vq1YZ7ZfNlnheeeUVVFRUICgoCN988w2ef/55TJw4UbpAa2lUde3aFV988QUSEhLw6quvwsfHB6+99ppRYrk1Jk1/M6C5sefl5SWt8/DDD+OFF17Ahx9+iMbGRqxevdpo8bR3/goMDETv3r0tcgxZA2uvi/Stc5ycnBAbG4u5c+dCJpNhwYIFcHNzkxolUVFRUCgU0sXUrXbs2IFLly7hs88+Q2NjI2bPno39+/dj2rRpWLlyJWJiYjBlyhTs27cPiYmJUve2iooKyOVyfPrpp/j888/xxhtvIDw8XPrcAwcOoKKiAkuWLJHKkpKSkJCQYNT4TUHf+sYaYzcHa88dfbSXX1u2bGnTYKioqIC3tzeys7OxePFiPP/889izZw9kMhn8/f0RGRmJV155BYcOHULPnj3x4osv4p///Kd0/j527Bg+/vhj3H333di7dy+A5h+vg4KCpG5oRUVFeO+993Dw4EG4urpi7ty52LVrl9Zrwvbya+3atejVq5cJ9pbxtZdjcnkH780YMq0bGU9xcbEYOHCgqK2tFUII8dtvv4khQ4aIpUuXivT0dOl9hw4dEtHR0UKI5uluT548KS1reV1YWCh8fX1Fdna2qKmpkZafOHFCTJ48WQghxObNm1tN0/vGG2+IFStWiLq6OjFw4EBRVVUlhBCioaFBDBkyRJw4ccJ0G09kZi25EBUVJQ4dOiSVp6enixdeeEEIIcTAgQNFYWGhtGzatGniyJEjZo+VyNL0yZfGxkbR1NQkhBDi999/F4MGDRJC/K9ua6mLWuo2InvScpzX1dUJIYTYsmVLq2usluuz69evS2UqlUrMmTNHCCHEnj17REREhLRsz5494oknnhBCCLF8+XKRmpoqLbv5M1JSUkRcXFy776P/sco7ObcbBwcH/OEPfwDQPMvKjRs3UF5eDg8PD+k9Hh4euHLlitbPGTZsGBISErBz504sX74cgYGB7f56cnO/TAcHBzQ1NaGqqgoymQzu7u4AACcnp1a/IBPZE135dfMtcTc3N53dAojsmbZ8+fTTT/Huu++ipqYGN27caDXA/ObZnVrqNiJ74+DgIM3CJpfLpWO+ZZlarcbmzZvx1VdfoampCTU1Nejfv7/0nptzS5O6ujqsWbMGeXl5AICqqippfCZpZpVjcgjo3r07KisrpdeVlZXo3r07gLaVRVVVlfTv0NBQ7Ny5Ezk5Oairq9N7woGuXbtCCCGNS2hsbOzwAC8iW6Etv4DWOVVVVaVXJURkrzTlS0lJCRISErB69WpkZ2cjLS3NglESWaeysjJ89dVX2LVrl9SlraN27NiBX3/9FXv37kV2djZmzZplgkjtDxs5VmrixInIzMxEXV0dGhsboVKppGeJeHt744cffgDQ3Ofy+vXrAIA9e/YgNTUVANCtWzfcc889en9fly5dMGDAAHz++ecAgIyMjA5N00dkSyZOnAiVSoWmpibU1tbik08+kfILgDS4/+eff8bvv/+OoUOHWipUIovTlC/l5eVwdXXFPffcg8bGRmRkZAAAampqLBwxkfW4cuUK7rrrLunB159//rleOeLo6Ihr165Jn3HPPfegS5cuOH/+PI4cOSJNRkWasZFjpUJDQ+Hv749HH30UU6dORc+ePfH4448DAObPn4/09HRMnToVP//8M+69914AwKRJk3DmzBkEBwcjLCwMZ8+exV//+le9vzMxMRFbt27FlClTUFtbix49erChQ3Zpzpw56NmzJ6ZMmYLHHnsMEydORFhYmLT8rrvuwrRp0/DUU0/hpZdessmZaYiMRVO+DB48GP7+/ggJCcGsWbMQGBiIESNGYM6cOZYOmchqKJVKVFZWIigoCLGxsViyZAkuXbqkc7KJgIAAfPjhh1i8eDEiIyNx8uRJhISEYO3atYiLi0Nubi7S09PNsxE2SiYEn9BE/yOEkBo2f/rTn5Cent5qClsiW2auhxsSERGRZfFODkkWL14s9anOzc2FEAJ33323ZYMiMqJr165JA0SJiIjIfnF2NZI899xzePHFF7Fnzx44OTlh3bp1vCAku3HkyBGsWLECMTExlg6FiIiITIzd1YiIyCrV19dj6tSpmD9/PsaNG4dly5ahqakJ3t7eWL9+PRQKBTIzM7Fjxw7I5XLMnDkTM2bMgFqtRlxcHC5cuAAHBwesWbMGffr0sfTmEBGRGbG7GpEJ/fjjj5g8eTJ27doFALh48SLmzJmD6OhoPPfcc2hoaAAAZGZm4rHHHsOMGTOwe/duAIBarUZsbCyioqIwe/ZsFBcXW2w7iCzhX//6lzR99+bNmxEdHY33338f/fr1g0qlQm1tLVJTU5Geno6dO3dix44dqKysxP79++Hu7o4PPvgAMTExSE5OtvCWEBGRuVlld7X8/HxLh0C3sVGjRhnlc2pra/Hqq69i3LhxUlnLhVpYWBg2btwIlUqFiIgIpKamQqVSwcnJCdOnT0dQUBBycnLg7u6O5ORkHD16FMnJydi0aZPW72TukCUZK3eA5um7z549Kz3wLi8vDy+//DKA5lmHtm3bhv79+2Po0KHSA45HjhyJgoIC5ObmIiIiAgAwfvx4xMfH6/w+5g5ZmjHzx5yYO2RpmnLHKhs5gOaA8/PzrepEwHh0s7aYtMVjzJO1QqFAWlpaqwfkmfpCDbDu3GEM9huDsS901q5dixUrVmDfvn0Amp/4rVAoAABeXl4oLS1FWVkZPD09pXU8PT3blMvlcshkMjQ0NEjrE5FxaTuXWMP57maMRztriwcw/LrNahs5RLbO0dERjo6tU8wcF2raEt4afnFjDIxBl3379mHEiBEax9FoGkra0fJbWfMPBLeytpgYj3a64rHWXCSyZWzkEFnI7XihxhjsNwZjXqQdPnwYxcXFOHz4MC5dugSFQgFXV1fU19fDxcUFJSUlUCqVUCqVKCsrk9a7fPkyRowYAaVSidLSUgwePBhqtRpCiE7dxXF2djbGZhERtcHzi+mwkUN2KeNsFSobmtqUd1M4YIgFTyiWvFCzJ9r+vrPu9bBARGRMN48927JlC+666y6cOnUK2dnZmDZtGg4ePAg/Pz8MHz4cCQkJuHr1KhwcHFBQUID4+HhUV1cjKysLfn5+yMnJwdixYzsVzxmXPjj2fXm7y3jMEVFnaDq/8NzSeWzkkF2qbGhCxfUblg6jjfHjx1vsQs2eWOvfl0xn0aJFWL58OTIyMtCrVy9ERETAyckJsbGxmDt3LmQyGRYsWAA3NzeEh4fj+PHjiIqKgkKhQFJSUqe+m8cbEZkKzy+mw0YOkYkUFRVh7dq1OH/+PBwdHZGdnY0NGzYgLi7OIhdq2mi6MwLw1ySyrEWLFkn/3r59e5vloaGhCA0NbVXW8mwcIiK6fbGRQ2QiPj4+2LlzZ5tya7xQ4y9JREREZE/4MFAiIiIiIrIrbOQQEREREZFdYSOHiIiIiIjsChs5RERERERkV9jIISIiIiIiu8JGDhERERER2RU2coiIiIiIyK6wkUNERERERHaFjRwiIiIiIrIrbOQQEREREZFdYSOHiIiIiIjsChs5RERERERkVxz1edO6deuQn5+PxsZGzJs3D0OHDsWyZcvQ1NQEb29vrF+/HgqFApmZmdixYwfkcjlmzpyJGTNmQK1WIy4uDhcuXICDgwPWrFmDPn36mHq7iIiIiIjoNqWzkXPixAn89NNPyMjIQEVFBR555BGMGzcO0dHRCAsLw8aNG6FSqRAREYHU1FSoVCo4OTlh+vTpCAoKQk5ODtzd3ZGcnIyjR48iOTkZmzZtMse2ERERERHRbUhnI2fMmDEYNmwYAMDd3R11dXXIy8vDyy+/DAAICAjAtm3b0L9/fwwdOhRubm4AgJEjR6KgoAC5ubmIiIgAAIwfPx7x8fGm2hYiottGxtkqVDY0tSnvpnDArHs9LBARERGR9dDZyHFwcICrqysAQKVSwd/fH0ePHoVCoQAAeHl5obS0FGVlZfD09JTW8/T0bFMul8shk8nQ0NAgra9Jfn5+u+XOzs4oKipqd9n169d1bY5JaIrVUqwtHsC8MTk7OwPyXlrfY437iKgjKhuaUHH9hqXDICIiskp6jckBgC+++AIqlQrbtm1DcHCwVC6EaPf9HS2/1ahRo9ot1/rrpY/5f73Mz8/XGKslWFs8gGViOvZ9udblmuJh44eIiIjI9unVyPn666+xdetWvP3223Bzc4Orqyvq6+vh4uKCkpISKJVKKJVKlJWVSetcvnwZI0aMgFKpRGlpKQYPHgy1Wg0hhM67ONrw10siIiIizThhFJEeU0hfu3YN69atw5tvvolu3boBaB5bk52dDQA4ePAg/Pz8MHz4cJw+fRpXr15FTU0NCgoKMHr0aPj6+iIrKwsAkJOTg7Fjx5pwc4iIiIhuXzdPGPX222/j9ddfx+bNmxEdHY33338f/fr1g0qlQm1tLVJTU5Geno6dO3dix44dqKysxP79++Hu7o4PPvgAMTExSE5OtvQmERlE552cAwcOoKKiAkuWLJHKkpKSkJCQgIyMDPTq1QsRERFwcnJCbGws5s6dC5lMhgULFsDNzQ3h4eE4fvw4oqKioFAokJSUZNINIiIiIrpdccIoomY6GzmzZs3CrFmz2pRv3769TVloaChCQ0NblbXc6iQiIiIi07LEhFG6xrNa23hXa4lH10RJRUVFFplUy1r2z80MiUnviQeIiIiIyDaYc8IobZMLWduESNYWj7aJknx8fMwYSTNr2z+A9pi0NX50jskhIiIiItvRMmFUWlpaqwmjAGidMKqlvLS0FACMMmEUkaWwkUNERERkJzhhFFEzdlcjIiKoK1m6AAAgAElEQVQishOcMIqoGRs5RERERHaCE0YRNWN3NSIiIiIisits5BARERERkV1hI4eIiIiIiOwKGzlERERERGRX2MghIiIiIiK7wkYOERERERHZFTZyiIiIiIjIrvA5OUREZHXWrVuH/Px8NDY2Yt68eRg6dCiWLVuGpqYmeHt7Y/369VAoFMjMzMSOHTsgl8sxc+ZMzJgxA2q1GnFxcbhw4YL0zI8+ffpYepOIiMiM2MghMrO8vDw899xzuO+++wAAAwcOxFNPPaX3BRyRvTtx4gR++uknZGRkoKKiAo888gjGjRuH6OhohIWFYePGjVCpVIiIiEBqaipUKhWcnJwwffp0BAUFIScnB+7u7khOTsbRo0eRnJyMTZs2WXqziIjIjNhdjcgCHnzwQezcuRM7d+7EihUrsHnzZkRHR+P9999Hv379oFKpUFtbi9TUVKSnp2Pnzp3YsWMHKisrLR06kcmNGTMGKSkpAAB3d3fU1dUhLy8PkyZNAgAEBAQgNzcXhYWFGDp0KNzc3ODi4oKRI0eioKAAubm5CAoKAgCMHz8eBQUFFtsWIiKyDN7JIbICeXl5ePnllwE0X8Bt27YN/fv3ly7gAEgXcIGBgZYMlcjkHBwc4OrqCgBQqVTw9/fH0aNHoVAoAABeXl4oLS1FWVkZPD09pfU8PT3blMvlcshkMjQ0NEjra5Kfn9+mzNnZGZD30rpeUVERrl+/3qFtNIb24rUkxqOdtcVDZO/YyCGygLNnzyImJgZVVVVYuHAh6urq9L6A00VbRWoNF3Gdreh1xatPrNZwsdGZGIyxDzobgzl88cUXUKlU2LZtG4KDg6VyIUS77+9o+a1GjRrVbvmx78u1rufj46PX5xtTfn6+xngtgfFopysea89FIlvERg6Rmd19991YuHAhwsLCUFxcjMcffxxNTU3SclNdqBUVFWm8GDPXRZyxLjy0xasrVmu4+DFGDJ3ZB8aK4dbPM6avv/4aW7duxdtvvw03Nze4urqivr4eLi4uKCkpgVKphFKpRFlZmbTO5cuXMWLECCiVSpSWlmLw4MFQq9UQQui8i0NERPaFY3KIzKxHjx4IDw+HTCZD37590b17d1RVVaG+vh4AtF7AKZVKg7/3jEsfvPl9eZv/PvvtWqe3iciYrl27hnXr1uHNN99Et27dADSPrcnOzgYAHDx4EH5+fhg+fDhOnz6Nq1evoqamBgUFBRg9ejR8fX2RlZUFAMjJycHYsWMtti1ERGQZvJNDZGaZmZkoLS3F3LlzUVpaiitXruDRRx9FdnY2pk2b1uoCLiEhAVevXoWDgwMKCgoQHx9v8PdWNjSh4vqNNuUeiqZ23k1kOQcOHEBFRQWWLFkilSUlJSEhIQEZGRno1asXIiIi4OTkhNjYWMydOxcymQwLFiyAm5sbwsPDcfz4cURFRUGhUCApKcmCW0NERJbARg6RmQUGBmLp0qX48ssvoVarsWrVKtx///1Yvny5XhdwRPZu1qxZmDVrVpvy7du3tykLDQ1FaGhoq7KWZ+MQEdHti40cIjPr2rUrtm7d2qZc3ws4IiIiItKOY3KIiIiIiMiu6HUn58cff8T8+fPx5JNPYvbs2bh48aLeT2dXq9WIi4vDhQsXpC4Effr0MfV2ERGZTcbZKlQ2tB3b1E3hgFn3elggIiIiItugqQ4FmuvRIc7OBn2uzjs5tbW1ePXVVzFu3DiprCNPZ9+/fz/c3d3xwQcfICYmBsnJyQYFSkRkrVomdbj1P00nbSIiImqmqQ7tbD2qs5GjUCiQlpbWauravLw8TJo0CUDz09lzc3NRWFgoPZ3dxcVFejp7bm4ugoKCADRPAVpQUGBwsERERERERLro7K7m6OgIR8fWb+vI09lvLpfL5ZDJZGhoaND5YDZDnsxuzKeyd4S1PanY2uIBzBuTruMEsM59REREZAwcZkBkhNnVOvp09s4+tb2zT/k2Nmt4evrNrC0ewDIxaTtOAM3HFxs/RERky7QNMwgLC8PGjRuhUqkQERGB1NRUqFQqODk5Yfr06QgKCkJOTg7c3d2RnJyMo0ePIjk5GZs2bbLgFhEZxqDZ1VxdXfV+OrtSqURpaSkAQK1WQwih8y4OEREREXUchxkQNTPoTs748eP1fjp7dXU1srKy4Ofnh5ycHIwdO9bY20BEREREsMwwA129IKytl4S1xMNhGPoNMQAMi0lnI6eoqAhr167F+fPn4ejoiOzsbGzYsAFxcXF6PZ09PDwcx48fR1RUFBQKBZKSkjocJBERERF1nimGGWjrkm5t3eitLR4Ow9A9xAAwbJiBzkaOj48Pdu7c2aZc36eztwxaIyIiIiLzaxlm4OLionWYwYgRI6RhBoMHD+YwA7JpBo3JISIiIiLb0DLMAECrYQanT5/G1atXUVNTg4KCAowePRq+vr7IysoCAA4zIJvW6dnViIiIiMg6cJgBUTM2coiIiIjsBIcZEDVjdzUiIiIiIrIrbOQQEREREZFdYSOHiIiIiIjsChs5RHRbcXZ2tnQIREREZGKceICIbhsZZ6tQKe/V7oPHuikcMOteDwtERURERMbGRg4R3TYqG5pQcf2GpcMgIiIiE2N3NSIiIiIisits5BARAZBZOgAiIjvEcZBkKeyuRkQEwF0hbx6z09DUZhnH6xARGeaMSx+OgySLYCOHiDrFnhoGHLNDRIbSdi4cchvfzeB5lSyFjRwi6hRWYEREPBcSWRuOySEiIiIiIrvCRg4R0W2Ig4GJiMiesbsaEdFtRtNDUW1xHBUREVF72MghIovQNEgX4MW2qXHsABER2Ts2cojIInihTURERKbCMTlERERERGRXeCeHiMiC7Ok5Q0RERNaCjRwiIgtitz0iIiLjY3c1IiIiIiKyK2zkEBERAEBm6QCIiIiMxCzd1V5//XUUFhZCJpMhPj4ew4YNM8fXEtkF5g+Zi7tCbldjhJg7RIZh7pA9MHkj55tvvsFvv/2GjIwM/Pzzz4iPj0dGRoapv5bILjB/yNzsZYwQc4fIMMwdshcmb+Tk5uZi8uTJAIABAwagqqoK1dXV6Nq1q0Gf103h0KFyU3N2drbI92pibfEAlolJ63FSb+ZgOsGY+aNpn7g5OUCmpaOSrtzqSE7efCxo+1xjfqe+y7XtB2PHo89+MNU+0PaezuwDa2OO3NG1zJSs7VzPeFjvaMLrNu24f/Q4xxqYPyZv5JSVlWHIkCHSa09PT5SWlupMlvz8/HbL79W0Qh2gYRWT0xSrpVhbPID5Y9J2nFyHde6j9hiSP4bkjlY6cqujOdkSn8b1TPCdOtf7/+uaMx6d+8FE+0DXuhpZ8BxrCLPkDsC65ya3ezysd9rH6zbtuH90n2MNzR+zTyEthND5nlGjRpkhEiLboyt/mDtE7WPuEBmGuUO2yuSzqymVSpSVlUmvL1++DG9vb1N/LZFdYP4QGYa5Q2QY5g7ZC5M3cnx9fZGdnQ0AOHPmDJRKpcH9OoluN8wfIsMwd4gMw9whe2Hy7mojR47EkCFDEBkZCZlMhsTERFN/JZHdYP4QGYa5Q2QY5g7ZC5nQZ5AMERERERGRjTB5dzUiIiIiIiJzYiOHiIiIiIjsitmnkNbl9ddfR2FhIWQyGeLj4zFs2DBp2fHjx7Fx40Y4ODjA398fCxYs0LmOKeM5ceIENm7cCLlcjv79+2P16tU4efIknnvuOdx3330AgIEDB2LFihVmiScwMBA9e/aEg0PzQ5U2bNiAHj16WGT/lJSUYOnSpdL7iouLERsbC7VajZSUFPTt2xcAMH78eDz77LNGiwcAfvzxR8yfPx9PPvkkZs+e3WqZJY4hSzAkj4xt3bp1yM/PR2NjI+bNm4fg4GBpmaZj1Zjy8vK05qI59sPu3buRmZkpvS4qKsKpU6ek10OGDMHIkSOl1+np6dI+6axb8+DixYtYtmwZmpqa4O3tjfXr10OhULRax97yQF+sdwyPxxL1jraYLFX3sN6xjnpH33jMUQe1x5DjxFLxWGIfabtuMGj/CCuSl5cnnnnmGSGEEGfPnhUzZ85stTwsLExcuHBBNDU1iaioKPHTTz/pXMeU8QQFBYmLFy8KIYRYtGiROHz4sDhx4oRYtGiR0WLoSDwBAQGiurq6Q+uYMp4WarVaREZGiurqarFnzx6RlJRktBhuVVNTI2bPni0SEhLEzp072yw39zFkCYbkkbHl5uaKp556SgghRHl5uZgwYUKr5e0dq8amKxfNsR9ulpeXJ1atWtWq7MEHHzTJd7WXB3FxceLAgQNCCCGSk5PFe++91yY+e8oDfbHe6Vw85q53OvL55qp7WO9YR73TkXjMUQfdypDjxJLxmHsf6bpuMGT/WFV3tdzcXEyePBkAMGDAAFRVVaG6uhpA868xHh4euPPOOyGXyzFhwgTk5uZqXceU8QDA3r170bNnTwDNTwSuqKgwyvcaGo+x1jF2PB9//DFCQkLQpUsXo3yvNgqFAmlpaVAqlW2WWeIYsgRD8sjYxowZg5SUFACAu7s76urq0NTUZPTvMZS59sPNUlNTMX/+fJN+R4v28iAvLw+TJk0CAAQEBLTZXnvLA32x3ulcPMZaxxQxmavuYb1jHfWOvvFYiiHHiaXisQRt1w2G7h+rauSUlZXhjjvukF57enqitLQUAFBaWgpPT882y7StY8p4AEjzxl++fBnHjh3DhAkTAABnz55FTEwMoqKicOzYMaPEok88AJCYmIioqChs2LABQgiL7p8Wu3fvxvTp06XX33zzDebOnYsnnngC33//vVFiaeHo6AgXF5d2l1niGLIEQ/LI2BwcHODq6goAUKlU8Pf3b9MN69Zj1RQ05aK59kOL//znP7jzzjvbPFCvoaEBsbGxiIyMxPbt2432fe3lQV1dndQ9zcvLq8322lse6Iv1TufiAcxb7+gbE2C+uof1jnXUO/rG08IcddDNDDlOLBVPC3PuI23XDYbuH6sbk3MzQ3aoKf8I7X32lStXEBMTg8TERNxxxx24++67sXDhQoSFhaG4uBiPP/44Dh482KbvuyniWbx4Mfz8/ODh4YEFCxZID/O6WXV1NX755Rf0798fe/fuRWZmJtLT000SDwCcOnUK99xzj1QxDx8+HJ6enpg4cSJOnTqF5cuX49NPPzXK9xuLOU525mTJ7fniiy+gUqmwbdu2VuXtHauhoaFG/W5tufjWW2/h119/Ner3aaNSqfDII4+0KV+2bBn+/Oc/QyaTYfbs2Rg9ejSGDh1q8nj0OSbsLQ/0xXpHu1svLPSpd0x9LNlD3WNv+WZt26PP9ZKx66COeO2118zS26UjLLWPNF03GMKq7uQolUqUlZVJry9fviz98nnrspKSEiiVSq3rmDIeoLnB8PTTT2PJkiV46KGHAAA9evRAeHg4ZDIZ+vbti+7du6OkpMQs8URERMDLywuOjo7w9/fHjz/+2Gadc+fO4cqVK2aJBwAOHz6McePGSa8HDBiAiRMnAgAeeOABlJeXm60bkyWOIUswJI9M4euvv8bWrVuRlpYGNze3VsvaO1aNTVsuuri4oLa2VnqvKfcD0NxV7IEHHmhTHhUVhS5dusDV1RV/+tOfTLIfWri6uqK+vh5A+9trb3mgL9Y7+sdTWlqKCxcudLjeMfaxZEt1D+sd89Y7+sQDmKcO6ojNmzejoaFBem2OfaSLJfaRpusGQ48hq2rk+Pr6Sr8CnTlzBkqlUvoVpnfv3qiursa5c+fQ2NiInJwc+Pr6al3HlPEAQFJSEp544gn4+/tLZZmZmXjnnXcANFcIV65c6fBsFOfOncNDDz2EtLQ0hISEICQkBN999x2OHj2K1atX48UXX8SZM2fg5OSEyMhIhIaG4i9/+Quio6PR0NCALVu2ID09Hf/+97/x0UcfYf369bh8+TI2btyI2tpabN68uVXXmFdeeQXBwcGYMmVKhw5iffb96dOnMXjwYOl1Wloa9u/fD6B5Vg9PT0+jzSaliyWOIUswJI+M7dq1a1i3bh3efPNNdOvWrc2yuXPnSif0kydPSrNCdVZeXh4efvhhJCUl4aGHHsKDDz6I7777DmfPnsWPP/6IJ554As888wxu3LiBhoYGnDt3Dt9++y3S0tKwfft2hIeH4/jx4wCAxx57DFlZWdJn5+TkYNq0aWhsbMRLL72EkJAQBAUFYeHChVr7epeUlKBLly5tflX/5ZdfEBsbCyEEGhsbUVBQYLT90J7x48dLx8XBgwfh5+fXarm95YG+7L3euTknQkJCEBgYiO+++w7Xr1/HypUrERISgrCwMCQlJUkX/YMGDcKlS5ekeBITE3Hp0iU89thjaGpqwvTp09HQ0IBvvvkGI0eORHBwMGbPno0jR47gvvvuQ+/evbFhwwaEhoYiJCQELi4u6Nq1K/Ly8jBr1iysXr0akyZNwqOPPorCwkLMmTMHvr6+2Lx5s7RNGRkZCA0NRWBgIP7+979LDXR99hFgPXUP6x3z1Tv6xmNoHaQpl7Zs2YKEhARMnz4d6enpEELgH//4B0JCQhAQEIDXXnsNTU1NeO+99xATEyN9XlNTE8aOHYuff/4Zjz/+OEpLS3Hu3Dl89tlneOutt7B9+3Y8/vjj+P333wEAcXFx+Oc//ymtf/PrXbt2ISwsDKGhoZg+fTp++umnTu0/U9bT2r5T03WDoceQVXVXGzlyJIYMGYLIyEjIZDIkJiZi7969cHNzQ1BQEFatWoXY2FgAQHh4OPr374/+/fu3Wccc8Tz00EPYt28ffvvtN6hUKgDA1KlTMWXKFCxduhRffvkl1Go1Vq1aZVCXgYqKCnh7eyM7OxuLFy/G888/jz179uCf//wndu3ahdOnT+Py5cuIiYlB//79UVxcjIyMDMyaNQvl5eWorKzE9u3bcddddyEoKAgzZ87EnXfeiSFDhuDxxx/HtGnTsHfvXhQWFmLp0qVYuXIlXnnlFaSnp+P111/v9P4JCgoC0Fzhenl5Ses8/PDDeOGFF/Dhhx+isbERq1ev7vC+0aaoqAhr167F+fPn4ejoiOzsbAQGBqJ3794WOYYswZA8MrYDBw6goqICS5YskcrGjh2LQYMGISgoCP7+/pg1axacnZ3xxz/+0ai3wH/++Wc8++yzWLhwIaKjo/Hkk0+ia9euGD58OB555BE0NDRg3bp1CA4ORmxsLH744QcEBQVhw4YN2LdvHxITE3Ho0CFMnToV+/fvl2I7dOgQpkyZgqNHj+LcuXNSAyglJQWnTp1q02hocWtf4rfeegtjxozBAw88gJ49e2L69OmQy+UIDAw02hSy7eXBhg0bEBcXh4yMDPTq1QsREREAgOeffx5r1qxp97i5HdwO9U5LTsTFxWH37t1YtWoVwsPDcenSJXz22WdobGzE7NmzsX//fkybNq1NPAAwf/58uLu7QyaT4ZlnnsGRI0ewfv16TJ06FadPn8aFCxcghEBwcDAefvhhjB49GpWVlaitrcX58+elHwLOnDmDv//974iPj8eMGTPwyiuvYNeuXSguLsYjjzyCefPm4fTp00hJScHHH3+MHj16YOXKlUhJScHy5cv1+psB5q17WO9YR73TkXgMrYPay6VJkybhyJEj+OSTT+Dp6Yl9+/YhKysLKpUKf/jDH7BgwQJ88MEH6NevH44cOYIffvgBCoUCu3fvhpOTE3755RcAwN/+9jcsWrQIP/zwA+bOnYulS5di27ZtWLlypdZhBdXV1UhJSUFOTg66du2Kzz//HIcPH9bZKNF13Jqynm6PrusGg46hTsz2RiZSXFwsBg4cKOrq6oQQQmzZsqXV9KABAQFiy5YtYv78+VLZ1atXxeDBg4VarRabN29u9f433nhDrFixQgghxOzZs8W+ffuEEELs2bNHPPbYY9L7PvroI/G3v/3NpNtGZEonTpwQo0aNEjdu3BBCCFFZWSkGDhwoHnvsMXHo0CHpffPmzRObN28WQgjR2NgompqahBBC/P7772LQoEFCCCEuXbokhg0bJq5evSoaGxvFn/70J/H777+LwsJC4evrK7Kzs0VNTY2Zt5CoYzTlxKOPPtoqJ9LT08ULL7wghBBi4MCB0jTVN78+ceKEmDx5shBCiF9++UWMHj1aeo9arRb19fXit99+E2PGjJG+Twghpk2bJo4fPy5OnDghxo0bJ5XHxcVJ0zrfuHFDDBo0SFy4cEGsXbtWvPrqq9L7vv/+exEYGGjM3ULUYZpyae3atWLBggXS+xYvXiy2bdsmvT506JCYM2eOEEKI6OhoKe9effVVkZqaKoRovq47efKk+OijjzRe2y1fvlx6vxBCet3Q0CDGjBkjtm/fLkpLS023A2yQVd3Jof9xcHCQZr2Qy+XSjBMtyxQKBdzd3aUyNzc3CCGk6URv7svo4OCgse/xzbfHtb2PyFa0/Nrc8m+g+Tb4zTlxc+58+umnePfdd1FTU4MbN25IA1R79OiBYcOG4eDBg+jbty/uuusu9OnTB3369EFCQgJ27tyJ5cuXIzAwEImJia0+k8iatJcT1dXV8PDwkN7j4eHRofGaFRUVrXLK0dERjo6OKC8vh5ubm/R9Ld9ZXl6O7t27txpcfXPdJpPJIJfL0dTUhGvXruHQoUM4evQogOZB42q12oAtJzIuTfXLzbl07do1vPPOO8jIyADQ3C2t5W5+SEgIvvrqK0yePBlffvllm1k1KyoqtF7btcfJyQnp6enYunUrtmzZgkGDBiExMRGDBg0yzkbbMDZybJSXl1erJ6dXVVVBLpe3mjKR6HZUWVkp/buqqgpAc2V07do1qby8vBx9+vRBSUkJEhISsHv3btx///349ddfERISIr1vypQpyMrKQr9+/RAeHi6Vh4aGIjQ0FJWVlYiPj8c777yD559/3gxbR9RxmnLi5vLKykp0794dAKTGxs3vv9Udd9yByspK3LhxA3K5HGq1GiUlJfDy8kJVVRWEENLFYGVlZauuY7oolUo88sgjUvc0ImvRXi55eHi0aoQolUoEBgZi9uzZbdYPCQnBm2++idOnT8PDwwN33313q+Xaru3kcjlu3LjR5vsB4I9//KM0ecHbb7+NxMREfPjhh53eXltnVRMPkP4cHR3x7bffori4GADw4YcfwtfXF46O2tutjo6OrS72iOxNfX09vvjiCwBAdnY2fHx8MGLECKns999/R35+PoDmxo6rqyvuueceNDY2Sr+81dTUAGhuzOTn5yMrKwthYWEAgD179iA1NRUA0K1bN9xzzz1m3T6ijmovJ0JCQqBSqdDU1ITa2lp88skn0jN3vL298cMPPwBoPt7l8uZLBUdHR9TW1qKxsRF33303evbsiYMHDwJoniZ95cqV6N27N3r27IkDBw4AAAoKClBWVtah8WaBgYE4ePAgysvLATRPKfvWW28ZZ2cQdUJ7ueTs7NzqPZMmTcInn3yCuro6AM3XZx9//DGA5h4Cffr0wdatW6U65Wa+vr4ar+1uzsvi4mIUFBQAAP7v//4PixcvRkNDAxQKBXx8fFrdSb2d8U6OjerZsydee+01zJ8/H2q1Gr1798arr76qc73Jkydj/fr1KC4u5q1Mskt33XUX8vPzsX79eqjVamzatAm9evXC888/j8DAQAwYMADBwcEAgMGDB8Pf3x8hISHw8vJCXFwcCgoKMGfOHOzduxfdunXDmDFjUFVVhTvvvBNAcwUWHx+P4OBgODg4oF+/fkhKSrLkJhNp1V5ODBo0CMXFxZgyZQpkMhlCQ0Oli67nn38eq1atwubNmxEZGSl1ax40aBA8PDzg6+uLjz/+GCkpKXjhhRewceNGeHt7Y82aNZDJZNi4cSMSExPxj3/8A3/4wx+QkpLSqsu1LkOGDEFMTAzmzJmDGzduwMvLCy+//LJJ9g1RR7SXS0eOHGn1nsmTJ+Onn36Sno3Wt2/fVhNdhISEICkpqd07ldqu7WbOnImFCxciODgYf/zjH6VeBwMHDkTv3r0xdepUODk5oUuXLli5cqWpdoFNkQlhZU9sIiIyUF5eHhISEnDo0CGjfeaqVatw33334S9/+YvRPpPIXEyRE0S3I+aS7WF3NSIiDX799Vf8+9//xp///GdLh0JEREQdwO5qRETtSElJwSeffIIVK1a0mkWKiIiIrB+7qxERERERkV1hdzUiIiIiIrIrVtldrWV6VyJLGDVqlKVDMBhzhyyJuUNkOFvNH+YOWZqm3LHKRg6gOeD8/HyLnwgYg/3GYA8na2vOHWuJgzEYPwbmjnWz9W2w9/iNnT/19fWYOnUq5s+fj3HjxmHZsmVoamqCt7c31q9fD4VCgczMTOzYsQNyuRwzZ87EjBkzoFarERcXhwsXLsDBwQFr1qxBnz59dH6frm2z5b+dvridlqEtd9hdjYiIiMiO/Otf/4KHhwcAYPPmzYiOjsb777+Pfv36QaVSoba2FqmpqUhPT8fOnTuxY8cOVFZWYv/+/XB3d8cHH3yAmJgYJCcnW3hLiAzHRg7RbeLWpzITEZH9+fnnn3H27FlMnDgRQPPzXSZNmgQACAgIQG5uLgoLCzF06FC4ubnBxcUFI0eOREFBAXJzcxEUFAQAGD9+PAoKCjodD+seshSr7a5GtiPjbBUqG5raXdZN4YBZ93qYOSJqzxmXPjj2fXmbcv6NiIg6T1Nd2E3hgCFmvNBfu3YtVqxYgX379gEA6urqoFAoAABeXl4oLS1FWVkZPD09pXU8PT3blMvlcshkMjQ0NEjra6Kpy5Czs7PWumdIfTGuX79u0HZaI3votqsPW9lONnKo0yobmlBx/YalwyAd+HciIjIdazjH7tu3DyNGjNA4jkbTU0M6Wn4rbWM0jn1frnG/+Pj46PX5tsDaxqqYirVtp7YGFxs5RERERHbg8OHDKC4uxuHDh3Hp0iUoFAq4urqivr4eLi4uKCkpgVKphFKpRFlZmbTe5cuXMWLECCiVSpSWlmLw4ADnUl4AACAASURBVMFQq9UQQui8i0NkrTgmh4iIiMgObNq0CXv27MFHH32EGTNmYP78+Rg/fjyys7MBAAcPHoSfnx+GDx+O06dP4+rVq6ipqUFBQQFGjx4NX19fZGVlAQBycnIwduxYS24OUafwTg4RERGRnVq0aBGWL1+OjIwM9OrVCxEREXByckJsbCzmzp0LmUyGBQsWwM3NDeHh4Th+/DiioqKgUCiQlJRk6fCJDMZGDhEREZGdWbRokfTv7du3t1keGhqK0NDQVmUtz8YhsgfsrkZERERERHaFjRwiIiIiIrIrbOQQEREREZFdYSOHiIiIiIjsChs5RERERERkV9jIISIiIiIiu8JGDhERERER2RU+J4fIROrq6hAXF4crV67g+vXrmD9/PgYPHoxly5ahqakJ3t7eWL9+PRQKBTIzM7Fjxw7I5XLMnDkTM2bMgFqtRlxcHC5cuCA9u6BPnz6W3iwik2PuEBFRZ/FODpGJ5OTkwMfHB7t27cKmTZuQlJSEzZs3Izo6Gu+//z769esHlUqF2tpapKamIj09HTt37sSOHTtQWVmJ/fv3w93dHR988AFiYmKQnJxs6U0iMgvmDhERdRYbOUQmEh4ejqeffhoAcPHiRfTo0QN5eXmYNGkSACAgIAC5ubkoLCzE0KFD4ebmBhcXF4wcORIFBQXIzc1FUFAQAGD8+PEoKCiw2LYQmRNzh4iIOovd1YhMLDIyEpcuXcLWrVvx17/+FQqFAgDg5eWF0tJSlJWVwdPTU3q/p6dnm3K5XA6ZTIaGhgZpfU3y8/PblDk7OwPyXhrXKSoqwvXr1w3ZPIO0F6O5MQbriUETa8gdfZbZClvfBmuOX9c5FrDu+InsERs5RCb24Ycf4r///S9eeOEFCCGk8pv/fbOOlt9q1KhR7ZYf+75c4zo+Pj56fbYx5Ofna4yRMdhuDKa4gLOW3LGGv1Vn2fo22EL82s6xgObjC2ADiMgU2F2NyESKiopw8eJFAMD999+PpqYmdOnSBfX19QCAkpISKJVKKJVKlJWVSetdvnxZKi8tLQUAqNVqCCF0/hJNZA+YO0RE1Fls5BCZyLfffott27YBAMrKylBbW4vx48cjOzsbAHDw4EH4+flh+PDhOH36NK5evYqamhoUFBRg9OjR8PX1RVZWFoDmgdhjx4612LYQmRNzh4iIOovd1YhMJDIyEi+99BKio6NRX1+PlStXwsfHB8uXL0dGRgZ69eqFiIgIODk5ITY2FnPnzoVMJsOCBQvg5uaG8PBwHD9+HFFRUVAoFEhKSrL0JhGZBXOHiIg6S69Gzo8//oj58+fjySefxOzZs3Hx4kU+r4BIBxcXl3anrt2+fXubstDQUISGhrYqa8kXotsNc4eIiDpLZ3e12tpavPrqqxg3bpxUxucVEBERERGRtdLZyFEoFEhLS4NSqZTK+LwCIiIiIiKyVjq7qzk6OsLRsfXb6urqbvvnFTCGZkVFRTqfDWDqZ7BYw34gIiIiIuvR6YkHbsfnFTCG/8Xg4+Oj89kApnwGi7H3AxtMRERERLbPoCmkXV1d+bwCIiIiIiKySgY1cvi8AiIiIiIislY6u6sVFRVh7dq1OH/+PBwdHZGdnY0NGzYgLi6OzysgIiIiIiKro7OR4+Pjg507d7Yp5/MKiIiIiIjIGhnUXY2IiIiIiMhasZFDRERERER2hY0cIiIiIiKyK2zkEBERERGRXWEjh4iIiIiI7AobOUREREREZFd0TiFNRIZbt24d8vPz0djYiHnz5mHo0KFYtmwZmpqa4O3tjfXr10OhUCAzMxM7duyAXC7HzJkzMWPGDKjVasTFxeHChQvSVOx9+vSx9CYRmQVzh4iIOoONHCITOXHiBH766SdkZGSgoqICjzzyCMaNG4fo6GiEhYVh48aNUKlUiIiIQGpqKlQqFZycnDB9+nQEBQUhJycH7u7uSE5OxtGjR5GcnIxNmzZZerOITI65Q0REncXuakQmMmbMGKSkpAAA3N3dUVdXh7y8PEyaNAkAEBAQgNzcXBQWFmLo0KFwc3ODi4sLRo4ciYKCAuTm5iIoKAgAMH78eBQUFFhsW4jMiblDRESdxTs5RCbi4OAAV1dXAIBKpYK/vz+OHj0KhUIBAPDy8kJpaSnKysrg6ekprefp6dmmXC6XQyaToaGhQVpfk/z8/DZlzs7OgLyXxnWKiopw/fr1Dm+jodqL0dwYg/XEcCtryh19ltkKW98Ga45f1zkWMG/87O5JxEYOkcl98cUXUKlU2LZtG4KDg6VyIUS77+9o+a1GjRrVbvmx78s1ruPj46PXZxtDfn6+xhgZg+3GYIoLOGvJHWv4W3WWrW+DLcSv7RwLaD6+AOPmD7t7EjVjdzUb5ezsbOkQSA9ff/01tm7dirS0NLi5ucHV1RX19fUAgJKSEiiVSiiVSpSVlUnrXL58WSovLS0FAKjVagghdP4STWQvmDtEhmF3T6JmvJNjgzLOVqFS3qvNr0bdFA6Yda+HhaKiW127dg3r1q1Deno6unXrBqC5wsjOzsa0adNw8OBB+Pn5Yfjw4UhISMDVq1fh4OCAgoICxMfHo7q6GllZWfDz80NOTg7Gjh1r4S0iMg/mDpHhLNHdU9OdKGvrKm1q1tyl0phsZTvZyLFBlQ1NqLh+w9JhkA4HDhxARUUFlixZIpUlJSUhISEBGRkZ6NWrFyIiIuDk5ITY2FjMnTsXMpkMCxYsgJubG8LDw3H8+HFERUVBoVAgKSnJgltDZD7MHaLOM2d3T21d8aylq7Sp2UKXSmOwtu3U1uBiI4fIRGbNmoVZs2a1Kd++fXubstDQUISGhrYqaxnwSXS7Ye4QdU5Ld8+33367VXdPFxcXrd09R4wYIXX3HDx4MLt7kk3jmBwiIiIiO9HS3fPNN99s090TQKvunqdPn8bVq1dRU1ODgoICjB49Gr6+vsjKygIAdvckm2ZXd3IyzlahsqGpTTnHqhAREdHtgN09iZrZVSOHY1WIiIjodsbunkTN2F2NiIiIiIjsChs5RERERERkV9jIISIiIiIiu2LQmJy8vDw899xzuO+++wAAAwcOxFNPPYVly5ahqakJ3t7eWL9+PRQKBTIzM7Fjxw7I5XLMnDkTM2bMMOoG0P/ILB0AEREREZEVMHjigQcffBCbN2+WXr/44ouIjo5GWFgYNm7cCJVKhYiICKSmpkKlUsHJyQnTp09HUFCQNKUhGZe7Qq5xhjmAs8wRERER0e3BaN3V8vLyMGnSJABAQEAAcnNzUVhYiKFDh8LNzQ0uLi4YOXIkCgoKjPWV1I6WGeba+09T44eIiIiIyJ78v/buPTiq+v7/+Gtz2fBNCJeFbAAFtXjnJrciCQRBwiVYi44EwmTQmVSlCahtEDIpCB2n3AwM2NqqiMgwvUTAYmodQnXCVCBETJyUYDsYOrXcTLIQAgkJuXh+f1j3ByabLMleT56PGWfcz+Gz530+ez7nnHfO53xOp+/klJeXa/HixaqpqdGSJUtUX1/vfCNuv379VFVVJYfDIZvN5qxjs9lUVVXl1vcXFxff1LKIiAgpZJDLOmVlZbp27Zpb63ZHe/F5U0fb2RFPt0NZWVmH8Xh6nd/nr98CAAAAgalTSc7tt9+uJUuWaPbs2Tp9+rQWLVqklpb/f5fAMIw267kqb8vYsWPbLC8uLna57PAXF11+3/Dhw91ed0fai8EX2tvOjni6HYYPH95hPJ5cZ1sxePK3IGECAAAIfp0arhYbG6ukpCRZLBYNGTJE/fv3V01NjRoaGiRJFRUVstvtstvtcjgcznqVlZWy2+2eiRwAAAAA2tCpJCcvL0/bt2+XJFVVVenChQt6/PHHlZ+fL0k6cOCAJk+erFGjRun48eO6fPmy6urqVFJSonHjxnkuegAAAAD4nk4NV5s2bZqWLVumjz/+WE1NTVqzZo3uu+8+rVixQrm5uRo0aJDmzp2r8PBwZWZmKi0tTRaLRRkZGYqOjvb0NgAB6+TJk0pPT9dTTz2l1NRUnT9/3u2p1puampSVlaVz584pNDRU69at0+DBg/29SYBP0HcAAF3RqSSnZ8+eev3111uV79ixo1XZrFmzNGvWrM6sBghqV69e1csvv6yJEyc6y1599VW3p1ovKChQr169tGnTJh06dEibNm3Sli1b/LhFgG/QdwAAXeWxKaQB3MhqtWrbtm03PId2M1OtFxYWKjExUZIUFxfH9OvoNug7AICu6vQU0oAnuHp5qRleXBoWFqawsBu72M1MtX59eUhIiCwWixobG531XQmEKdY7Egiz2BFD4MTwfYHUd9xZFiyCfRsCOX53Xu8QyPEDZkSSA7/67uWl3dHNTrXu7hTs/p5ivSP+noKdGLwTgy8v4HzddwLht+qqYN+GYIi/o9cptBc/CRDgeQxXA3woMjLS7anW7Xa78+W5TU1NMgyjw79EA2ZF3wEA3AySHMCH4uLi3J5qPT4+Xvv375ckFRQUaMKECf4MHfAr+g4A4GYE3XC1iIgIf4cAuKWsrEwbNmzQ2bNnFRYWpvz8fOXk5CgrK8utqdaTkpJ05MgRpaSkyGq1av369f7eJMAn6DsAgK4KuiTnRI/BbY57vTUq3A/RAK4NHz5cu3btalXu7lTr373fA+hu6DsAgK4KuiTH1YPqva2tZ+gCAAAA0P3wTA4AAAAAUyHJAQAAAGAqJDkAAAAATCXonskJNLnlNbrU2Pp5oD7WUM2/s7cfIgIAAAC6N5KcLnI1EQIAwBx4dQEABB+SHAAA2uHq1QUSd+0BIFCR5AAA0A7u2ANA8CHJ8SOe5wEAAAA8jyTHj/jrIAAAAOB5TCENAAAAwFRIchDUmPUIADrGsRJAd8NwNQQl5/NMIYNazXrEM02BjYstwLvafN4zZJBOlNdwbETAsPg7AJgeSQ6CUnvPM3HgDBxcbAG+Z8bnPfnjiPn0soa4nICprzVUyZwj0EU+SXLWrl2r0tJSWSwWZWdna+TIkb5YLW6SWWZ7a+/AGWzbIgVG/3HVnrdGhau26RuXy/xxsWWm3x5dEwh9B13j6q49/dm7fNV3XJ0jepMAwQO8nuR8+umn+uqrr5Sbm6tTp04pOztbubm53l4tOsFMf/0zy7b4ov+4c+fL9YmoRTWN37hc1pV1BlNihcDDuccc6M++Fyh9pzMJkNT+OYIEqXvxepJTWFio6dOnS5KGDh2qmpoa1dbWqmfPnp36vj7W0DbLo8NDZXFx6eSqzvX2/7dWV5pad4jo8FDNGnJjrNffNnf13e6ss7N1O9MGXfleV+XftUN73+vpdbqzvKv7QiDxZP9xte23RIW73P8lKfb/wjq1z7W3zJ11dpY7+1QgDH0hBu/yRd/paJm/dOW8FGiCaVvajbXBx8F0QSBct3W0rLapc4lvVHiIy3NP7P+F6Wqz4fa14PeZ+Xh6vWDaTq8nOQ6HQ8OGDXN+ttlsqqqq6rCzFBcXt1l+p6sK9e18Wb3k4uucYv73X5uxVLVR9r8vbC+ejtbZ2bqdaoMufq+resXFxa7reWmd7db7X93OxhNoOtN/Ott3XO3/utT5ftfRMl+v8/qmcdVOvkQM3uOTviMF5DGlK+elQBNM29JerNcUPH3Nk31H8t75oyt12zz3XJKiXC1T29eCrf5NkPzGXRUs2+nziQcMw+jw34wdO9YHkQDBp6P+Q98B2kbfATqHvoNg5fX35NjtdjkcDufnyspKxcS4/PstgOvQf4DOoe8AnUPfgVl4PcmJj49Xfn6+JOnEiROy2+2dHtcJdDf0H6Bz6DtA59B3YBZeH642ZswYDRs2TAsWLJDFYtHq1au9vUrANOg/QOfQd4DOoe/ALCyGOw/JAAAAAECQ8PpwNQAAAADwJZIcAAAAAKbi8ymk3bV27VqVlpbKYrEoOztbI0eOdC47cuSINm/erNDQUCUkJCgjI8MrMWzcuFHFxcVqbm7Ws88+qxkzZjiXTZs2TQMGDFBo6LcvucrJyVFsbKxH119UVKTnn39ed911lyTp7rvv1qpVq5zLfdEOu3fvVl5envNzWVmZPv/8c+fnYcOGacyYMc7P77zzjrNNPOHkyZNKT0/XU089pdTUVJ0/f17Lly9XS0uLYmJi9Morr8hqtd5Qp719p7vydZt8v++MGDGizd8tLy9PO3fuVEhIiJKTkzVv3jyPxtHQ0KBHHnlE6enpmjhxos9jyMvL01tvvaWwsDA999xzuueee3waQ11dnVasWKGamho1NTUpIyNDMTExWrNmjSTpnnvu0S9/+UtJ0ltvvaX9+/fLYrFoyZIlmjJlikdiCHZmOJ58/zgabNo7Fwe6+vp6ZWVl6cKFC7p27ZrS09M1depUf4flE2boO+3pzPVJMHL3fB6QjABUVFRkPPPMM4ZhGEZ5ebmRnJx8w/LZs2cb586dM1paWoyUlBTjyy+/9HgMhYWFxk9+8hPDMAzj4sWLxpQpU25YPnXqVKO2ttbj673e0aNHjaVLl7pc7ot2uF5RUZGxZs2aG8p++MMfem19dXV1RmpqqrFy5Upj165dhmEYRlZWlvHhhx8ahmEYmzZtMn7/+9+3irG9fac78nWbtNV32vrd6urqjBkzZhiXL1826uvrjTlz5hjV1dUejWXz5s3G448/buzdu9fnMVy8eNGYMWOGceXKFaOiosJYuXKlz2PYtWuXkZOTYxiGYXz99dfGzJkzjdTUVKO0tNQwDMP4+c9/bhw8eND473//azz22GPGtWvXjAsXLhgzZ840mpubPRJDMDPD8aSt42gw6ehcHOj++te/Gm+++aZhGIZx5swZY8aMGX6OyDfM0Hfa05nrk2Dk7vk8UAXkcLXCwkJNnz5dkjR06FDV1NSotrZWknT69Gn17t1bAwcOVEhIiKZMmaLCwkKPxzB+/Hht3bpVktSrVy/V19erpaXF4+vpLF+1w/Vee+01paene3Ud17Nardq2bZvsdruzrKioSA8//LAkaerUqa22ub19p7vydZu01Xfa+t1KS0s1YsQIRUdHq0ePHhozZoxKSko8FsepU6dUXl6uhx56SFLb+443YygsLNTEiRPVs2dP2e12vfzyyz6PoW/fvrp06ZIk6fLly+rTp4/Onj3r/IvqdzEUFRVp8uTJslqtstlsuuWWW1ReXu6RGIKZGY4nbR1Hg0mgn4s7kpSUpKefflqSdP78eY+P+AhUZug77enM9Ukwcvd8HqgCMslxOBzq27ev87PNZlNVVZUkqaqqSjabrc1lnhQaGqrIyEhJ0p49e5SQkNBqGNbq1auVkpKinJycDt8I3Fnl5eVavHixUlJSdPjwYWe5r9rhO//4xz80cODAVi8Ea2xsVGZmphYsWKAdO3Z4dJ1hYWHq0aPHDWX19fXO26L9+vVrtc3t7Tvdla/bpK2+09bv5nA4vLoPb9iwQVlZWc7Pvo7hzJkzamho0OLFi7Vw4UIVFhb6PIY5c+bo3LlzSkxMVGpqqpYvX65evXo5l/vqtwhWZjietHUcDSbunIuDwYIFC7Rs2TJlZ2f7OxSfMEPfaU9nrk+Ckbvn80AVsM/kXM9bCYQ7PvroI+3Zs0dvv/32DeXPPfecJk+erN69eysjI0P5+fmaNWuWR9d9++23a8mSJZo9e7ZOnz6tRYsW6cCBA34Z+7hnzx499thjrcqXL1+uRx99VBaLRampqRo3bpxGjBjhk5jc2S/8ue8EKl+1yfV95/ox9K7W78m49u3bpwceeECDBw++qXV5um0uXbqk3/zmNzp37pwWLVp0w/f7Iob3339fgwYN0vbt2/Wvf/1LGRkZio6O9mkMZkK7+I+rc3Gw+NOf/qR//vOfevHFF5WXlyeLxeLvkHyqu/Uds23vzZ7PA0VA3smx2+1yOBzOz5WVlc47CN9fVlFR4bXb8J988olef/11bdu27YYLA0maO3eu+vXrp7CwMCUkJOjkyZMeX39sbKySkpJksVg0ZMgQ9e/fXxUVFZJ82w7St7dhR48e3ao8JSVFUVFRioyM1IMPPuiVdrheZGSkGhoaJLW9ze3tO92VP9rk+32nrd+trbg8tQ8fPHhQH3/8sZKTk7V792799re/9XkM/fr10+jRoxUWFqYhQ4YoKipKUVFRPo2hpKREkyZNkiTde++9unbtmqqrq53LXcXg7eNJsOB4EhjaOxcHurKyMp0/f16SdN9996mlpUUXL170c1Te1x37TkfXJ8HKnfN5oArIJCc+Pl75+fmSpBMnTshut6tnz56SpFtvvVW1tbU6c+aMmpubVVBQoPj4eI/HcOXKFW3cuFFvvPGG+vTp02pZWlqaGhsbJUnHjh1zzoDmSXl5edq+fbukb4enXbhwwTme9/p2ePzxx7Vv3z5nOxQUFOjHP/6xx+KoqKhQVFRUqztI//73v5WZmSnDMNTc3KySkhKvtMP14uLinPvGgQMHNHny5BuWt7fvdFe+bpO2+k5bv9uoUaN0/PhxXb58WXV1dSopKdG4ceM8EsOWLVu0d+9evfvuu5o3b57S09N9HsOkSZN09OhRffPNN6qurtbVq1d9HsNtt92m0tJSSdLZs2cVFRWloUOH6rPPPrshhgcffFAHDx5UY2OjKioqVFlZqTvvvNMjMQQzjif+1965OBh89tlnzrtPDodDV69evWEYl1l1x77T0fVJMHL3fB6oAnK42pgxYzRs2DAtWLBAFotFq1ev1nvvvafo6GglJiZqzZo1yszMlPTtQ3133HGHx2P48MMPVV1drRdeeMFZNmHCBN1zzz1KTExUQkKC5s+fr4iICN1///0eH6omfTtN9bJly/Txxx+rqalJa9as0QcffNCqHaqrqxUVFeVsh7/97W+aM2eOx+L4/vM/b775psaPH6/Ro0drwIABeuKJJxQSEqJp06Z5dIrIsrIybdiwQWfPnlVYWJjy8/OVk5OjrKws5ebmatCgQZo7d64k6Wc/+5nWrVvX5r7T3fm6TdrqO+vXr9fKlStv+N3Cw8OVmZmptLQ0WSyWVkOpPG3p0qVasWKFz2KIjY3VzJkzlZycLElauXKlRowY4dMY5s+fr+zsbKWmpqq5uVlr1qxRTEyMXnrpJX3zzTcaNWqU4uLiJEnJyclKTU2VxWLRmjVrFBISkH8D8ykzHE/aOo7++te/DpqEoa3jyYYNGzRo0CA/RuW+BQsW6Be/+IUWLlyohoYGvfTSS92ib5mh77TnZq5Pgpm75/NAZTECfUAdOlRRUaEZM2bo0KFDioyM1KRJk/Tuu++6fB4BAAAAMLOAvJODmxMbG6uRI0fqwIEDGjJkiG655RYSHAAAAHRbJDkmMWfOHO3fv1+33XabkpKS/B0OAAAA4DcMVzOJS5cuadq0aYqMjNTu3bs1cOBAf4cEAAAA+AV3ckyiT58+Gj9+vGpqakhwAAAA0K2R5JjIwIEDlZCQ4O8wAAAAAL8y/zyG3cR//vMf/f3vf9ejjz7q71AAAAAAv+JOjgls3bpV77//vlatWhV0b4MGAAAAPI2JBwAAAACYCsPVAAAAAJhKQA5XKy4u9ncI6MbGjh3r7xAAAADQBW4lOQ0NDXrkkUeUnp6uiRMnavny5WppaVFMTIxeeeUVWa1W5eXlaefOnQoJCVFycrLmzZunpqYmZWVl6dy5cwoNDdW6des0ePBgtwJzdaFZXFwcUBehxNOxQIupvXhIsAEAAIKfW8PVfve736l3796SpFdffVULFy7UH/7wB912223as2ePrl69qtdee03vvPOOdu3apZ07d+rSpUv64IMP1KtXL/3xj3/U4sWLtWnTJq9uDAAAAAB0mOScOnVK5eXleuihhyRJRUVFevjhhyVJU6dOVWFhoUpLSzVixAhFR0erR48eGjNmjEpKSlRYWKjExERJUlxcnEpKSroccERERJe/AwAAAIB5dThcbcOGDVq1apX27dsnSaqvr5fVapUk9evXT1VVVXI4HLLZbM46NputVXlISIgsFosaGxud9dvT1rChiIgInegxWIe/uNhqWR9rqIY1nNa1a9c6/G5PC7QhToEWjxR4MQVaPAAAAPCcdpOcffv26YEHHnD5HI2r2advtrwtrp6ZOPzFRVVf+6bNZcOHD3f7+z0lmJ438ZdAi4lncgAAAMyt3STn4MGDOn36tA4ePKivv/5aVqtVkZGRamhoUI8ePVRRUSG73S673S6Hw+GsV1lZqQceeEB2u11VVVW699571dTUJMMw3LqLAwAAAACd1e4zOVu2bNHevXv17rvvat68eUpPT1dcXJzy8/MlSQcOHNDkyZM1atQoHT9+XJcvX1ZdXZ1KSko0btw4xcfHa//+/ZKkgoICTZgwwftbBAAAAKBbu+n35CxdulQrVqxQbm6uBg0apLlz5yo8PFyZmZlKS0uTxWJRRkaGoqOjlZSUpCNHjiglJUVWq1Xr16/3xjYAAAAAgJPbSc7SpUud/79jx45Wy2fNmqVZs2bdUPbdu3EAAAAAwFfcek8OAAAAAAQLkhwAAAAApkKSAwAAAMBUSHIAAAAAmApJDgAAAABTIckBAAAAYCokOQAAAABMhSQHAAAAgKmQ5AAAAAAwFZIcAAAAAKZCkgMAAADAVEhyAAAAAJgKSQ4AAAAAUyHJAQAAAGAqJDkAAAAATIUkBwAAAICpkOQAAAAAMBWSHAAAAACmQpIDAAAAwFRIcgAAAACYCkkOAAAAAFMhyQEAAABgKiQ5AAAAAEyFJAcAAACAqZDkAAAAADAVkhwAAAAApkKSAwAAAMBUSHIAAAAAmApJDgAAAABTIckBAAAAYCokOQAAAABMhSQHAAAAgKmQ5AAAAAAwlTB3/tHGjRtVXFys5uZmPfvssxoxYoSWL1+ulpYWxcTE6JVXXpHValVeXp527typkJAQJScna968eWpqalJWVpbOnTun0NBQrVu3ToMHD/b2dgEAAADopjpMco4ePaovv/xSubm5qq6uFyPkwQAACXlJREFU1mOPPaaJEydq4cKFmj17tjZv3qw9e/Zo7ty5eu2117Rnzx6Fh4friSeeUGJiogoKCtSrVy9t2rRJhw4d0qZNm7RlyxZfbBsAAACAbqjD4Wrjx4/X1q1bJUm9evVSfX29ioqK9PDDD0uSpk6dqsLCQpWWlmrEiBGKjo5Wjx49NGbMGJWUlKiwsFCJiYmSpLi4OJWUlHhxcwAAAAB0dx3eyQkNDVVkZKQkac+ePUpISNChQ4dktVolSf369VNVVZUcDodsNpuzns1ma1UeEhIii8WixsZGZ31XiouLW5VFRERIIYNc1ikrK9O1a9c62iSPaytWfwq0eKTAiynQ4gEAAIDnuPVMjiR99NFH2rNnj95++23NmDHDWW4YRpv//mbLv2/s2LFtlh/+4qLLOsOHD3fruz2puLjYZaz+EGjxSIEXU3vxkPwAAAAEP7dmV/vkk0/0+uuva9u2bYqOjlZkZKQaGhokSRUVFbLb7bLb7XI4HM46lZWVzvKqqipJUlNTkwzD6PAuDgAAAAB0VodJzpUrV7Rx40a98cYb6tOnj6Rvn63Jz8+XJB04cECTJ0/WqFGjdPz4cV2+fFl1dXUqKSnRuHHjFB8fr/3790uSCgoKNGHCBC9uDgAAAIDursPhah9++KGqq6v1wgsvOMvWr1+vlStXKjc3V4MGDdLcuXMVHh6uzMxMpaWlyWKxKCMjQ9HR0UpKStKRI0eUkpIiq9Wq9evXe3WDAAAAAHRvHSY58+fP1/z581uV79ixo1XZrFmzNGvWrBvKvns3DgAAAAD4glvP5AAAAABAsCDJAQAAAGAqJDkAAAAATIUkBwAAAICpkOQAAAAAMBWSHAAAAACmQpIDAAAAwFRIcgAAAACYCkkOAAAAAFMhyQEAAABgKiQ5AAAAAEyFJAcAAACAqZDkAAAAADAVkhwAAAAApkKSAwAAAMBUSHIAAAAAmApJDgAAAABTIckBAAAAYCokOQAAAABMhSQHAAAAgKmQ5AAAAAAwFZIcAAAAAKZCkgMAAADAVEhyAAAAAJgKSQ4AAAAAUwnzdwCAN+SW1+hSY0ur8j7WUA2LiPBDRAAAAPAVkhyY0qXGFlVf+8bfYQAAAMAPGK4GAAAAwFRIcgAAAACYCkkOAAAAAFMhyQEAAABgKiQ5AAAAAEyFJAcAAACAqfhkCum1a9eqtLRUFotF2dnZGjlypC9WCwAAAKAb8nqS8+mnn+qrr75Sbm6uTp06pezsbOXm5np7tQAAAAC6Ka8nOYWFhZo+fbokaejQoaqpqVFtba169uzZqe/rYw29qXJvi4iI8Mt6XQm0eCT/xNTuftLg42AAAADgU15PchwOh4YNG+b8bLPZVFVV1WGSU1xc3Gb5na4q1Esuqnidq1j9JdDikXwfU3v7yTUFZhsBAADAM3zyTM71DMPo8N+MHTvWB5EAAAAAMCOvz65mt9vlcDicnysrKxUTE+Pt1QIAAADoprye5MTHxys/P1+SdOLECdnt9k4/jwMAAAAAHfH6cLUxY8Zo2LBhWrBggSwWi1avXu3tVQIAAADoxiyGOw/JAAAAAECQ8PpwNQAAAADwJZIcAAAAAKbi8ymkO7J27VqVlpbKYrEoOztbI0eOdC47cuSINm/erNDQUCUkJCgjI6PDOt6M5+jRo9q8ebNCQkJ0xx136Fe/+pWOHTum559/XnfddZck6e6779aqVat8Es+0adM0YMAAhYZ++yLMnJwcxcbG+qV9KioqtGzZMue/O336tDIzM9XU1KStW7dqyJAhkqS4uDj99Kc/9Vg8knTy5Emlp6frqaeeUmpq6g3L/LEPAQAAwMeMAFJUVGQ888wzhmEYRnl5uZGcnHzD8tmzZxvnzp0zWlpajJSUFOPLL7/ssI4340lMTDTOnz9vGIZhLF261Dh48KBx9OhRY+nSpR6L4WbimTp1qlFbW3tTdbwZz3eampqMBQsWGLW1tcbevXuN9evXeyyG76urqzNSU1ONlStXGrt27Wq13Nf7EAAAAHwvoIarFRYWavr06ZKkoUOHqqamRrW1tZK+vRPQu3dvDRw4UCEhIZoyZYoKCwvbrePNeCTpvffe04ABAyRJNptN1dXVHllvZ+PxVB1Px/PnP/9ZM2fOVFRUlEfW2x6r1apt27bJbre3WuaPfQgAAAC+F1BJjsPhUN++fZ2fbTabqqqqJElVVVWy2WytlrVXx5vxSHK+76eyslKHDx/WlClTJEnl5eVavHixUlJSdPjwYY/E4k48krR69WqlpKQoJydHhmH4tX2+s3v3bj3xxBPOz59++qnS0tL05JNP6osvvvBILN8JCwtTjx492lzmj30IAAAAvhdwz+Rcz+jE7NadqdOV775w4YIWL16s1atXq2/fvrr99tu1ZMkSzZ49W6dPn9aiRYt04MABWa1Wr8fz3HPPafLkyerdu7cyMjKcL2HtaBu8FY8kff755/rBD37gTAhHjRolm82mhx56SJ9//rlWrFihv/zlL16LqTO82UYAAADwvoBKcux2uxwOh/NzZWWlYmJi2lxWUVEhu92u8PBwl3W8GY8k1dbW6umnn9YLL7ygSZMmSZJiY2OVlJQkSRoyZIj69++viooKDR482OvxzJ071/n/CQkJOnnyZId1vBmPJB08eFATJ050fh46dKiGDh0qSRo9erQuXryolpYW52QJ3uSPfQgAAAC+F1DD1eLj4513H06cOCG73e68A3DrrbeqtrZWZ86cUXNzswoKChQfH99uHW/GI0nr16/Xk08+qYSEBGdZXl6etm/fLunb4VEXLlxQbGys1+O5cuWK0tLS1NjYKEk6duyY7rrrLr+2jyQdP35c9957r/Pztm3b9MEHH0j6dhY0m83mkwRH8s8+BAAAAN+zGAE2NicnJ0efffaZLBaLVq9erS+++ELR0dFKTEzUsWPHlJOTI0maMWOG0tLS2qxz/UW1t+KZNGmSxo8fr9GjRzv/7SOPPKI5c+Zo2bJlunz5spqamrRkyRLnszrejCcxMVE7d+7Uvn37FBERofvvv1+rVq2SxWLxS/skJiZKkn70ox9px44d6t+/vyTp66+/1osvvijDMNTc3Ozx6ZrLysq0YcMGnT17VmFhYYqNjdW0adN06623+m0fAgAAgG8FXJIDAAAAAF0RUMPVAAAAAKCrSHIAAAAAmApJDgAAAABTIckBAAAAYCokOQAAAABMhSQHAAAAgKmQ5AAAAAAwlf8HgOSycz0jF4IAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 1008x720 with 20 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2xIAlmhwa1HI",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 290
        },
        "outputId": "106ba73e-c360-48d9-9832-1e1998097758"
      },
      "source": [
        "bm_predict_yes = bm[bm['y'] == 1.0]\n",
        "bm_predict_no = bm[bm['y'] == 0.0]\n",
        "\n",
        "sortbyAgeYes = bm_predict_yes.groupby('age').groups\n",
        "agesYes = []\n",
        "for x in list(sortbyAgeYes):\n",
        "    agesYes.append(len(sortbyAgeYes[x]))\n",
        "\n",
        "# Number of deposits by age group\n",
        "xAxisNumb = len(sortbyAgeYes)\n",
        "ind = np.arange(xAxisNumb) \n",
        "plt.figure(figsize=(15,4))\n",
        "plt.bar(ind, agesYes, width=0.5)\n",
        "plt.ylabel('Number of Deposits')\n",
        "plt.title('Number of Deposits by Age Group')\n",
        "labelAges = sortbyAgeYes.keys()\n",
        "plt.xticks(ind, labelAges, rotation='vertical')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AD-3meeFYXjQ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 333
        },
        "outputId": "bd7068c2-d860-4b8f-a930-dbe513de4ca3"
      },
      "source": [
        "# The average balance in account by job occupation\n",
        "groupJob = bm_predict_yes.groupby('job').groups\n",
        "sumBalance = bm_predict_yes.groupby(['job'])['balance'].sum()\n",
        "arrNumbBalance = []\n",
        "for x in list(groupJob):\n",
        "    arrNumbBalance.append(len(groupJob[x]))\n",
        "\n",
        "arrAmountBalance = []\n",
        "for i in range(len(sumBalance)):\n",
        "    arrAmountBalance.append(sumBalance[i])\n",
        "\n",
        "arrBalance = []\n",
        "for i in range(len(sumBalance)):\n",
        "    arrBalance.append(sumBalance[i])\n",
        "\n",
        "arrAvgBalance = [x/y for x, y in zip(arrAmountBalance, arrNumbBalance)]\n",
        "    \n",
        "xAxisNumb = len(sumBalance)\n",
        "ind = np.arange(xAxisNumb) \n",
        "plt.figure(figsize=(15,4))\n",
        "plt.bar(ind, arrAvgBalance, width=0.5)\n",
        "plt.ylabel('Balance Amounts in EURO')\n",
        "plt.title('The Average Balance Amounts by Job Occupation')\n",
        "labelAges = ['admin.', 'blue-collar', 'entrepreneur', 'housemaid', 'management', 'retired', 'self-employed', 'services', 'student', 'technician', 'unemployed', 'unknown']\n",
        "plt.xticks(ind, labelAges, rotation='vertical')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GplwNAevbTQT",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 316
        },
        "outputId": "fbea3269-34f4-4a2c-f52b-a0a14074c6ee"
      },
      "source": [
        "# The average balance by educational group\n",
        "groupEdu = bm_predict_yes.groupby('job').groups\n",
        "arrNumbEdu = []\n",
        "for x in list(groupEdu):\n",
        "    arrNumbEdu.append(len(groupEdu[x]))\n",
        "\n",
        "sumEdu = bm_predict_yes.groupby(['education'])['balance'].sum()\n",
        "arrAmountEdu = []\n",
        "for i in range(len(sumEdu)):\n",
        "    arrAmountEdu.append(sumEdu[i])\n",
        "\n",
        "arrAvgEdu = [x/y for x, y in zip(arrAmountEdu, arrNumbEdu)]\n",
        "    \n",
        "xAxisNumb = len(sumEdu)\n",
        "ind = np.arange(xAxisNumb) \n",
        "plt.figure(figsize=(5,4))\n",
        "plt.bar(ind, arrAvgEdu, width=0.5)\n",
        "plt.ylabel('Balance Amounts in EURO')\n",
        "plt.title('The Average Balance Amounts by Educational Group')\n",
        "labelAges = ['primary', 'secondary', 'tertiary', 'unknown']\n",
        "plt.xticks(ind, labelAges, rotation='vertical')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AX6UDcVybmUy",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 280
        },
        "outputId": "7956a91b-8b10-4c09-da7d-456150e585c2"
      },
      "source": [
        "# The average balance by educational group\n",
        "\n",
        "# Get average balance  \n",
        "def getAvgBalanceEdu(pred): \n",
        "    groupEdu = pred.groupby('education').groups\n",
        "    arrNumbEdu = []\n",
        "    for x in list(groupEdu):\n",
        "        arrNumbEdu.append(len(groupEdu[x]))\n",
        "\n",
        "    sumEdu = pred.groupby(['education'])['balance'].sum()\n",
        "    arrAmountEdu = []\n",
        "    for i in range(len(sumEdu)):\n",
        "        arrAmountEdu.append(sumEdu[i])\n",
        "\n",
        "    arrAvgEdu = [x/y for x, y in zip(arrAmountEdu, arrNumbEdu)]\n",
        "    arrNumbEdu = []\n",
        "    arrAmountEdu = []\n",
        "    return arrAvgEdu\n",
        "\n",
        "# Bar graph\n",
        "xAxisNumb = len(sumEdu)\n",
        "ind = np.arange(xAxisNumb) \n",
        "plt.figure(figsize=(10,4))\n",
        "w = 0.2\n",
        "plt.bar(ind, getAvgBalanceEdu(bm_predict_yes), width=w)\n",
        "plt.bar(ind + w, getAvgBalanceEdu(bm_predict_no), width=w)\n",
        "plt.ylabel('Balance Amounts in EURO')\n",
        "plt.title('The Average Balance Amounts by Educational Group')\n",
        "labelAges = ['primary', 'secondary', 'tertiary', 'unknown']\n",
        "plt.xticks(ind + w/2, labelAges)\n",
        "plt.legend(labels=['Yes', 'No'])\n",
        "\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Wjt3hVMwvtKp"
      },
      "source": [
        "## 3. **Preprocessing & Exploratory Data Analysis**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zTHN7LVx16Fh"
      },
      "source": [
        "Before diving into our analytical and machine learning research questions, we took some time to explore our dataset. Here is a list of important preprocessing steps that we took:\n",
        "\n",
        "- No Missing Values;\n",
        "- Change the unit of \"duration\" from seconds to minutes;\n",
        "- Drop rows with \"duration\" < 5s because those interactions do not contribute much to telemarketing;\n",
        "- Drop rows with \"job\" or \"education\" == \"unknown\";\n",
        "- Change the target classifier columns to “category” types.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pV7TQpSsLXl1"
      },
      "source": [
        "# Change the unit of \"duration\" from seconds to minutes\n",
        "bm['duration'] = bm['duration'].apply(lambda n: n / 60).round(2)\n",
        "\n",
        "# Drop rows with \"duration\" < 5s because those interactions do not contribute much\n",
        "duration_condition = (bm['duration'] < 5 / 60)\n",
        "bm = bm.drop(bm[duration_condition].index, axis=0, inplace=False)\n",
        "\n",
        "# Drop rows with \"job\" or \"education\" == \"unknown\"\n",
        "job_condition = (bm['job'] == 11.0)\n",
        "bm = bm.drop(bm[job_condition].index, axis=0, inplace=False)\n",
        "edu_condition = (bm['education'] == 3.0)\n",
        "bm = bm.drop(bm[edu_condition].index, axis=0, inplace=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UHFWl6nhLXl4"
      },
      "source": [
        "# Change the types of the target classifier columns\n",
        "\n",
        "bm[\"response\"] = bm['y'].astype('category')\n",
        "bm[\"response\"].describe()\n",
        "bm = bm.drop('y', axis=1)\n",
        "\n",
        "bm[\"credit_default\"] = bm['default'].astype('category')\n",
        "bm[\"credit_default\"].describe()\n",
        "bm = bm.drop('default', axis=1)\n",
        "\n",
        "bm[\"personal_loan\"] = bm['loan'].astype('category')\n",
        "bm[\"personal_loan\"].describe()\n",
        "bm = bm.drop('loan', axis=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wZSUEn6aLXl_",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "outputId": "c337d85e-9635-4cc1-970f-b04fce2550f8"
      },
      "source": [
        "bm[\"month\"].value_counts()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "8.0     1336\n",
              "5.0      675\n",
              "1.0      611\n",
              "6.0      491\n",
              "9.0      375\n",
              "0.0      286\n",
              "3.0      212\n",
              "4.0      134\n",
              "10.0      72\n",
              "7.0       46\n",
              "11.0      45\n",
              "2.0       18\n",
              "Name: month, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TtPi1LUfLXmD"
      },
      "source": [
        "# Visualize summary stats and correlations\n",
        "pp.ProfileReport(bm)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I4QcqiJIxLZx"
      },
      "source": [
        "From the above profile report and visualizations, we can clearly see that there is a class imbalance issue within the original dataset. Next we will explore this further."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8Gyi3Tsp1nS7"
      },
      "source": [
        "## 4. **Research Questions**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dmxSP6OWjU3Z"
      },
      "source": [
        "Part I. Analytical Questions\n",
        "1. What are the more influential factors when predicting a client’s subscription outcome?\n",
        "2. Establish the profile of the most responsive customer groups. (Shown in the EDA section)\n",
        "3. Check for seasonal effects (existence of most responsive months). (Shown in the EDA section)\n",
        "\n",
        "Part II. Machine Learning Questions\n",
        "4. Build classification models using “y” as target variable (whether the client has subscribed to a term deposit).\n",
        "5. Build classification models using “default” as target variable. **(codes in a separate notebook file)**\n",
        "6. Build classification models using “loan” as target variable. **(codes in a separate notebook file)**\n",
        "7. Which classification models or ensemble can be best used to question 4. predict whether a client would subscribe to a term deposit (best accuracies)?\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "w1kxOuozf1B8"
      },
      "source": [
        "**(Note: We applied all techniques and modeling to the three machine learning questions with the same workflow. However, due to the lengthy codes, we've decided to show only codes for the first question starting in the following section, and codes for the other two questions will be submitted as separate notebooks.)**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_IY8bNeM16Fv"
      },
      "source": [
        "According to our EDA results and the pandas profiling report, we can answer our 2nd and 3rd analytical questions as such:\n",
        "\n",
        "- Age group of the most responsive clients is 25 < age < 45;\n",
        "- Job categories of the most responsive clients are blue-collar, management, and unemployed;\n",
        "- Most responsive months are April, May, June, July, August, and November (this fact suggests that the company should focus more on the Summer and Fall seasons)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xVUr1L2MvMdZ"
      },
      "source": [
        "## 5. **Techniques Used** (modeling, resampling, and feature selection)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yxsHqPCw2L_y"
      },
      "source": [
        "### 5.1 **Balancing Data**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_Kh7FP09AA4K"
      },
      "source": [
        "#### 5.1.1 **Check Imbalance Issue**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7bQqn4FRAPez",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 301
        },
        "outputId": "36d451a4-c1ad-4c8c-97da-e524f46821f8"
      },
      "source": [
        "# Check the imbalance problem in \"response\"\n",
        "bm.response.value_counts().plot(kind='bar', title='Count(response)')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f04c0df06d8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3OT_tr3tARg8"
      },
      "source": [
        "From the above bar graph of the \"y\" or \"response\" column, the \"1\" or \"yes\" class is significantly less than the \"0\" or \"no\" class at a ratio of approximately 1:9."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "agRI_sBdCJvp"
      },
      "source": [
        "#### 5.1.2 **Testing Stratified Sampling**\n",
        "\n",
        "To address and try to solve the class imbalance issue, we first test the effectiveness of using stratified sampling alone. The testing algorithm in this case is the Gaussian NB model.\n",
        "\n",
        "Before the actual testing, we need to set up the different sets of X (predictor sets) and y (target) since our three research questions are targeting different predictors."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r92x2OA_CSfG"
      },
      "source": [
        "# Set up X and y for the first question\n",
        "X = bm.drop((['response']), axis=1)\n",
        "y = bm['response']\n",
        "X = X.astype('int')\n",
        "y = y.astype('category')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4DxqHTGMDrqu"
      },
      "source": [
        "# Specify metrics used for performance evaluation\n",
        "metric_names = ['f1', 'roc_auc', 'average_precision', 'accuracy', 'precision', 'recall']\n",
        "\n",
        "scores_df = pd.DataFrame(index=metric_names, columns=['Random-CV', 'Stratified-CV'])\n",
        "\n",
        "# Set up a 10-fold CV to compare with the stratified approach\n",
        "cv = KFold(n_splits=10)\n",
        "scv = StratifiedKFold(n_splits=10)\n",
        "clf = GaussianNB() # trial run with basic algorithm"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i5kU7HlnDtVb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 234
        },
        "outputId": "bdef2c4e-f76f-4039-d866-38274910c374"
      },
      "source": [
        "# Trial testing performance comparison for \"response\" target\n",
        "for metric in metric_names:\n",
        "    score1 = cross_val_score(clf, X, y, scoring=metric, cv=cv).mean()\n",
        "    score2 = cross_val_score(clf, X, y, scoring=metric, cv=scv).mean()\n",
        "    scores_df.loc[metric] = [score1, score2]\n",
        "\n",
        "scores_df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Random-CV</th>\n",
              "      <th>Stratified-CV</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>f1</th>\n",
              "      <td>0.382771</td>\n",
              "      <td>0.386351</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>roc_auc</th>\n",
              "      <td>0.809286</td>\n",
              "      <td>0.807362</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>average_precision</th>\n",
              "      <td>0.351555</td>\n",
              "      <td>0.350992</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>accuracy</th>\n",
              "      <td>0.837717</td>\n",
              "      <td>0.837716</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>precision</th>\n",
              "      <td>0.341861</td>\n",
              "      <td>0.344575</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>recall</th>\n",
              "      <td>0.438774</td>\n",
              "      <td>0.441347</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  Random-CV Stratified-CV\n",
              "f1                 0.382771      0.386351\n",
              "roc_auc            0.809286      0.807362\n",
              "average_precision  0.351555      0.350992\n",
              "accuracy           0.837717      0.837716\n",
              "precision          0.341861      0.344575\n",
              "recall             0.438774      0.441347"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UHk4F1fADzgj"
      },
      "source": [
        "As we can see from the above results, although this is only a testing NB classifier, most of the F1, precision, and recall scores are extremely low due to the class imbalance problem.\n",
        "\n",
        "Notice that the accuracy can seem very high even with just a basic tester model because of the highly unbalance nature of the dataset. For example, since the \"response\" column has a 1:9 ratio between two classes, even a 100% \"no\" blind classification would still have an accuracy of around 90%.\n",
        "\n",
        "Even after we implemented stratified sampling, the imbalance issue is still severely affecting the performance metrics (with little to no improvement or even decrease in performance). At this point, it becomes critical for us to find ways to balance the binary classes for each research question."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XIfpFHwdDmVr"
      },
      "source": [
        "#### 5.1.3 **Resampling**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z4yx8R_ND8ks"
      },
      "source": [
        "After extensive research, we decided to use resampling methods to counter-act the imbalance in our dataset. Resampling is a widely recognized and implemented technique for dealing with highly unbalanced datasets such as fraud detection or facial recognition. Fundamentally, resampling includes the following two ways of sampling:\n",
        "\n",
        "1. Undersampling: only sampling a small amount of observations from the majority class in order to minimize the difference in amount between the majority and the minority class;\n",
        "2. Oversampling: either copying or synthesizing data points that belong to the minority/under-represented class so that its amount can be similar to that of the majority class.\n",
        "\n",
        "Since our dataset only has about 4000 observations, undersampling would lead to a major loss of information and we would have only a very small chunk of data to work with.\n",
        "\n",
        "However, some oversampling techniques are more prone to introducing overfitting into the dataset. For example, the most basic and naive strategy of oversampling is the Random Over Sampling method, which functions by randomly duplicating existing samples until the amount of majority and minority classees are equal.\n",
        "\n",
        "To control for the possibility of overfitting, we take the following steps:\n",
        "\n",
        "1. We are not using the Random Over Sampler method;\n",
        "2. We ensure the validity of the oversampling process by implementing stratified 10-fold cross validation during each model evaluation;\n",
        "3. Instead of only looking at accuracy, we are also checking other metrics such as F1, AUC, precision, and recall scores.\n",
        "\n",
        "This way we can get more meaningful results from our dataset.\n",
        "\n",
        "In addition to scikit-learn, we will also be using the Python library imbalanced-learn which is compatible with scikit-learn and is part of scikit-learn-contrib projects.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "np5d-fq7EAb-"
      },
      "source": [
        "##### 5.1.3.1 **Setting Up Before Oversampling**\n",
        "\n",
        "Here we set up the stratified 10-fold CV and a list of models (not yet tuned) that we will use for testing the effectiveness of several oversampling methods."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "G1k6xjGAEF_B"
      },
      "source": [
        "RANSTA = 50\n",
        "\n",
        "scv = StratifiedKFold(n_splits=10, random_state=RANSTA, shuffle=True)\n",
        "\n",
        "models = []\n",
        "models.append((LogisticRegression(solver='lbfgs', max_iter=7000)))\n",
        "models.append((KNeighborsClassifier(n_neighbors=5)))\n",
        "models.append((DecisionTreeClassifier()))\n",
        "models.append((RandomForestClassifier()))\n",
        "models.append((GaussianNB()))\n",
        "models.append((SVC()))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jksq0MkAEICz"
      },
      "source": [
        "To make the oversampling method testing process more succinct and easy to read, we first create a helper function using for loops to evaluate the list of models and print out the metrics table. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o4CLfC9QEJor"
      },
      "source": [
        "def oversampleReport(models, metric_names, X_resampled, y_resampled):\n",
        "  \"\"\"\n",
        "  Print model and metrics report table\n",
        "  \"\"\"\n",
        "  scores_cols = ['LR', 'KNN', 'DT', 'RF', 'NB', 'SVC']\n",
        "  scores = pd.DataFrame(index=metric_names, columns=scores_cols)\n",
        "\n",
        "  for key, model in enumerate(models):\n",
        "      for metric in metric_names:\n",
        "          score = cross_val_score(model, X_resampled, y_resampled, scoring=metric, cv=scv).mean()\n",
        "          scores.at[metric, scores_cols[key]] = score \n",
        "  \n",
        "  return scores"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yy_mRStPEL3L"
      },
      "source": [
        "##### 5.1.3.2 **Oversampling Only (SMOTE and ADASYN)**\n",
        "\n",
        "To perform oversampling, we start by using one of the most known techniques: SMOTE (Synthetic Minority Over-sampling Technique). This method create synthetic datapoints in the following steps:\n",
        "\n",
        "1. For each real/existing data point, seek for its K nearest neighbors (default K = 5)\n",
        "2. Randomly choose one of the neighbors and build a vector between the two points\n",
        "3. Multiplying the vector with a random number between one and zero to get the new data point.\n",
        "\n",
        "The Adaptive Synthetic (ADASYN) is an adaptive approach of SMOTE.\n",
        "\n",
        "SMOTE: https://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-14-106\n",
        "\n",
        "ADASYN: https://www.researchgate.net/publication/224330873_ADASYN_Adaptive_Synthetic_Sampling_Approach_for_Imbalanced_Learning"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-vX1rsMKEdE7"
      },
      "source": [
        "##### **SMOTE**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HvYbP_kgEfAs",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "outputId": "88f3e63f-a250-4da4-ddb6-7a41796d29c2"
      },
      "source": [
        "# SMOTE oversampling for the first question\n",
        "smote = SMOTE(random_state=RANSTA)\n",
        "X_resampled_s1, y_resampled_s1 = smote.fit_resample(X, y)\n",
        "\n",
        "# Check the number of observations in each class after resampling\n",
        "print(sorted(Counter(y_resampled_s1).items()))\n",
        "\n",
        "oversampleReport(models, metric_names, X_resampled_s1, y_resampled_s1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[(0.0, 3803), (1.0, 3803)]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "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>LR</th>\n",
              "      <th>KNN</th>\n",
              "      <th>DT</th>\n",
              "      <th>RF</th>\n",
              "      <th>NB</th>\n",
              "      <th>SVC</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>f1</th>\n",
              "      <td>0.855145</td>\n",
              "      <td>0.845146</td>\n",
              "      <td>0.885778</td>\n",
              "      <td>0.928486</td>\n",
              "      <td>0.794455</td>\n",
              "      <td>0.504118</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>roc_auc</th>\n",
              "      <td>0.918126</td>\n",
              "      <td>0.905572</td>\n",
              "      <td>0.881932</td>\n",
              "      <td>0.980662</td>\n",
              "      <td>0.867242</td>\n",
              "      <td>0.62695</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>average_precision</th>\n",
              "      <td>0.894122</td>\n",
              "      <td>0.865935</td>\n",
              "      <td>0.833388</td>\n",
              "      <td>0.978691</td>\n",
              "      <td>0.841094</td>\n",
              "      <td>0.618768</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>accuracy</th>\n",
              "      <td>0.853275</td>\n",
              "      <td>0.831711</td>\n",
              "      <td>0.88154</td>\n",
              "      <td>0.925323</td>\n",
              "      <td>0.759932</td>\n",
              "      <td>0.568634</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>precision</th>\n",
              "      <td>0.844151</td>\n",
              "      <td>0.78323</td>\n",
              "      <td>0.869918</td>\n",
              "      <td>0.914583</td>\n",
              "      <td>0.694734</td>\n",
              "      <td>0.592367</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>recall</th>\n",
              "      <td>0.866685</td>\n",
              "      <td>0.917958</td>\n",
              "      <td>0.895077</td>\n",
              "      <td>0.942939</td>\n",
              "      <td>0.92769</td>\n",
              "      <td>0.439121</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                         LR       KNN        DT        RF        NB       SVC\n",
              "f1                 0.855145  0.845146  0.885778  0.928486  0.794455  0.504118\n",
              "roc_auc            0.918126  0.905572  0.881932  0.980662  0.867242   0.62695\n",
              "average_precision  0.894122  0.865935  0.833388  0.978691  0.841094  0.618768\n",
              "accuracy           0.853275  0.831711   0.88154  0.925323  0.759932  0.568634\n",
              "precision          0.844151   0.78323  0.869918  0.914583  0.694734  0.592367\n",
              "recall             0.866685  0.917958  0.895077  0.942939   0.92769  0.439121"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WPaaAUhVEkwL"
      },
      "source": [
        "##### **ADASYN**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ML1-sW0zElYm",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "outputId": "100959d3-36ef-463f-a216-800b32d60e8b"
      },
      "source": [
        "# ADASYN oversampling for the first question\n",
        "ada = ADASYN(random_state=RANSTA)\n",
        "X_resampled_a1, y_resampled_a1 = ada.fit_resample(X, y)\n",
        "\n",
        "# Check the number of observations in each class after resampling\n",
        "print(sorted(Counter(y_resampled_a1).items()))\n",
        "\n",
        "oversampleReport(models, metric_names, X_resampled_a1, y_resampled_a1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[(0.0, 3803), (1.0, 3760)]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "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>LR</th>\n",
              "      <th>KNN</th>\n",
              "      <th>DT</th>\n",
              "      <th>RF</th>\n",
              "      <th>NB</th>\n",
              "      <th>SVC</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>f1</th>\n",
              "      <td>0.849048</td>\n",
              "      <td>0.83503</td>\n",
              "      <td>0.881194</td>\n",
              "      <td>0.927511</td>\n",
              "      <td>0.789457</td>\n",
              "      <td>0.480433</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>roc_auc</th>\n",
              "      <td>0.915679</td>\n",
              "      <td>0.899661</td>\n",
              "      <td>0.88357</td>\n",
              "      <td>0.981492</td>\n",
              "      <td>0.867891</td>\n",
              "      <td>0.611678</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>average_precision</th>\n",
              "      <td>0.889896</td>\n",
              "      <td>0.855351</td>\n",
              "      <td>0.83021</td>\n",
              "      <td>0.980309</td>\n",
              "      <td>0.842721</td>\n",
              "      <td>0.601281</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>accuracy</th>\n",
              "      <td>0.848474</td>\n",
              "      <td>0.819247</td>\n",
              "      <td>0.879809</td>\n",
              "      <td>0.927408</td>\n",
              "      <td>0.755258</td>\n",
              "      <td>0.556658</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>precision</th>\n",
              "      <td>0.840975</td>\n",
              "      <td>0.764927</td>\n",
              "      <td>0.869149</td>\n",
              "      <td>0.913724</td>\n",
              "      <td>0.689867</td>\n",
              "      <td>0.575991</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>recall</th>\n",
              "      <td>0.857447</td>\n",
              "      <td>0.919681</td>\n",
              "      <td>0.891489</td>\n",
              "      <td>0.944681</td>\n",
              "      <td>0.922872</td>\n",
              "      <td>0.412234</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                         LR       KNN        DT        RF        NB       SVC\n",
              "f1                 0.849048   0.83503  0.881194  0.927511  0.789457  0.480433\n",
              "roc_auc            0.915679  0.899661   0.88357  0.981492  0.867891  0.611678\n",
              "average_precision  0.889896  0.855351   0.83021  0.980309  0.842721  0.601281\n",
              "accuracy           0.848474  0.819247  0.879809  0.927408  0.755258  0.556658\n",
              "precision          0.840975  0.764927  0.869149  0.913724  0.689867  0.575991\n",
              "recall             0.857447  0.919681  0.891489  0.944681  0.922872  0.412234"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8wzpnQlREqyQ"
      },
      "source": [
        "##### 5.1.3.3 **Combination of Over- and Undersampling (SMOTETomek and SMOTEENN)**\n",
        "\n",
        "In addition to using oversampling by itself, we also decided to test the combination methods which consist of both oversampling and undersampling techniques. Since SMOTE generates samples without regards to the underlying distribution, it might create some noisy samples where the two classes have no clear distinction. In this case, implementing undersampling algorithms such as Tomek's link and Edited Nearest Neighbors can be good for cleaning noisy samples.\n",
        "\n",
        "Tomek's link: https://dl.acm.org/doi/10.1145/1007730.1007735\n",
        "\n",
        "ENN: https://pdfs.semanticscholar.org/c1a9/5197e15fa99f55cd0cb2ee14d2f02699a919.pdf\n",
        "\n",
        "The imbalanced-learn library has implemented these two algorithms in combination with SMOTE to create two useful classes: SMOTETomek and SMOTEENN."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "A0ONVFWDExbR"
      },
      "source": [
        "##### **SMOTETomek**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7Cowx9itEusR",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "751fee99-abe7-4293-ece6-1f1f53cc3408"
      },
      "source": [
        "# SMOTETomek combo for the first question\n",
        "smo_tmk = SMOTETomek(random_state=RANSTA)\n",
        "X_resampled_st1, y_resampled_st1 = smo_tmk.fit_resample(X, y)\n",
        "\n",
        "# Check the number of observations in each class after resampling\n",
        "print(sorted(Counter(y_resampled_st1).items()))\n",
        "\n",
        "oversampleReport(models, metric_names, X_resampled_st1, y_resampled_st1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[(0.0, 3678), (1.0, 3678)]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AF0OX07HE5Up"
      },
      "source": [
        "##### **SMOTEENN**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KpjWY7fkE5sg",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 252
        },
        "outputId": "9ea04782-0ca5-4b63-ec24-1ec02c528054"
      },
      "source": [
        "# SMOTE-ENN combo for the first question\n",
        "smo_enn = SMOTEENN(random_state=RANSTA)\n",
        "X_resampled_se1, y_resampled_se1 = smo_enn.fit_resample(X, y)\n",
        "\n",
        "# Check the number of observations in each class after resampling\n",
        "print(sorted(Counter(y_resampled_se1).items()))\n",
        "\n",
        "oversampleReport(models, metric_names, X_resampled_se1, y_resampled_se1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[(0.0, 2180), (1.0, 3044)]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "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>LR</th>\n",
              "      <th>KNN</th>\n",
              "      <th>DT</th>\n",
              "      <th>RF</th>\n",
              "      <th>NB</th>\n",
              "      <th>SVC</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>f1</th>\n",
              "      <td>0.904723</td>\n",
              "      <td>0.965724</td>\n",
              "      <td>0.923638</td>\n",
              "      <td>0.952533</td>\n",
              "      <td>0.859778</td>\n",
              "      <td>0.700816</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>roc_auc</th>\n",
              "      <td>0.945166</td>\n",
              "      <td>0.989878</td>\n",
              "      <td>0.905882</td>\n",
              "      <td>0.987742</td>\n",
              "      <td>0.896995</td>\n",
              "      <td>0.678393</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>average_precision</th>\n",
              "      <td>0.950599</td>\n",
              "      <td>0.987347</td>\n",
              "      <td>0.8913</td>\n",
              "      <td>0.990052</td>\n",
              "      <td>0.899628</td>\n",
              "      <td>0.732697</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>accuracy</th>\n",
              "      <td>0.887452</td>\n",
              "      <td>0.959032</td>\n",
              "      <td>0.912137</td>\n",
              "      <td>0.943341</td>\n",
              "      <td>0.819685</td>\n",
              "      <td>0.649111</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>precision</th>\n",
              "      <td>0.89317</td>\n",
              "      <td>0.943181</td>\n",
              "      <td>0.910678</td>\n",
              "      <td>0.937182</td>\n",
              "      <td>0.786571</td>\n",
              "      <td>0.696014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>recall</th>\n",
              "      <td>0.916893</td>\n",
              "      <td>0.989487</td>\n",
              "      <td>0.937252</td>\n",
              "      <td>0.967473</td>\n",
              "      <td>0.948422</td>\n",
              "      <td>0.706293</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                         LR       KNN        DT        RF        NB       SVC\n",
              "f1                 0.904723  0.965724  0.923638  0.952533  0.859778  0.700816\n",
              "roc_auc            0.945166  0.989878  0.905882  0.987742  0.896995  0.678393\n",
              "average_precision  0.950599  0.987347    0.8913  0.990052  0.899628  0.732697\n",
              "accuracy           0.887452  0.959032  0.912137  0.943341  0.819685  0.649111\n",
              "precision           0.89317  0.943181  0.910678  0.937182  0.786571  0.696014\n",
              "recall             0.916893  0.989487  0.937252  0.967473  0.948422  0.706293"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V6Qhnhx216GR"
      },
      "source": [
        "##### SMOTEENN Wins"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VV3eEt4kE-QZ"
      },
      "source": [
        "From the above results, we can clearly see that the combination of SMOTE (oversampling) and ENN (undersampling/cleaning) works best for all three research questions with different target variables. After resampling using SMOTEENN, the metrics for each algorithm have significantly increased compared to those before resampling.\n",
        "\n",
        "After stratified CV and resampling, the resampled dataset can now be used for feature selection, further modeling, hyperparameter tuning, and model evaluation."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vxYsh_dO2ImL"
      },
      "source": [
        "### 5.2 **Feature Selection Using RFECV**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ylNYG-jIFGjx"
      },
      "source": [
        "# train-test 80/20 split for \"response\" classification\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_resampled_se1, y_resampled_se1, test_size=0.2, random_state=RANSTA)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Djk24TTW16GU"
      },
      "source": [
        "#### Plotting out the feature importance (answering anaytical question 1)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-icr9ffWLXmX",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 555
        },
        "outputId": "9755ea16-28c5-4bf7-a933-9790b56321b9"
      },
      "source": [
        "cv_estimator = RandomForestClassifier(n_jobs=4, random_state =RANSTA)\n",
        "cv_estimator.fit(X_train, y_train)\n",
        "cv_selector = RFECV(cv_estimator,cv= 10, step=1,scoring='accuracy')\n",
        "cv_selector = cv_selector.fit(X_train, y_train)\n",
        "rfecv_mask = cv_selector.get_support()\n",
        "\n",
        "rfecv_features = []\n",
        "\n",
        "for bool, feature in zip(rfecv_mask, X_train.columns):\n",
        "    if bool:\n",
        "        rfecv_features.append(feature)\n",
        "\n",
        "print('Optimal number of features :', cv_selector.n_features_)\n",
        "print('Best features :', rfecv_features)\n",
        "n_features = X_train.shape[1]\n",
        "plt.figure(figsize=(8,8))\n",
        "plt.barh(range(n_features), cv_estimator.feature_importances_, align='center') \n",
        "plt.yticks(np.arange(n_features), X_train.columns.values) \n",
        "plt.xlabel('Feature importance')\n",
        "plt.ylabel('Feature')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Optimal number of features : 15\n",
            "Best features : ['age', 'job', 'marital', 'education', 'balance', 'housing', 'contact', 'day', 'month', 'duration', 'campaign', 'pdays', 'previous', 'poutcome', 'personal_loan']\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "h1NUg_Zm16GX"
      },
      "source": [
        "From the RFECV feature selection plot above, we can answer the last analytical question regarding most influential factor. When using term deposit response (Yes/No) as the classification target, the msot influential factor is \"duration\"."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AzjdxKjZLXmZ"
      },
      "source": [
        "# Apply feature selection to do train-test split again\n",
        "X_resampled_se1 = X_resampled_se1.loc[: , ['age', 'job', 'marital', 'education', 'balance', 'housing', 'contact', 'day', 'month', 'duration', 'campaign', 'pdays', 'previous', 'poutcome', 'personal_loan']]\n",
        "X_train, X_test, y_train, y_test = train_test_split(X_resampled_se1, y_resampled_se1, test_size=0.2, random_state=RANSTA)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zlG5yFoDvSUu"
      },
      "source": [
        "## 6. **Further Modeling and Hyperparameter Tuning**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RSl7g0iG3nsU"
      },
      "source": [
        "To apply hyperparameter tuning and check if we can further improve our models, we select KNN, Logistic Regression, and Random Forest to do some more testing.\n",
        "\n",
        "Here we create a helper function which we will use to streamline the process of hyperparameter tuning using GridSearchCV and return the best parameters for each model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rgghgnKUQnue"
      },
      "source": [
        "def hyperTuning(model, param_grid, X_train, y_train, X_test, y_test):\n",
        "  \"\"\"\n",
        "  @params model: ML model, scikit-learn object\n",
        "  @params param_grid: the hyperparameter for the model\n",
        "  @return best hyper params (cross-validated)\n",
        "  \"\"\"\n",
        "  grid_search = GridSearchCV(estimator=model, param_grid=param_grid, n_jobs=-1, cv=10, scoring='accuracy',error_score=0)\n",
        "  grid_result = grid_search.fit(X_train, y_train)\n",
        "  # summarize results\n",
        "  \n",
        "  means = grid_result.cv_results_['mean_test_score']\n",
        "  stds = grid_result.cv_results_['std_test_score']\n",
        "  params = grid_result.cv_results_['params']\n",
        "  for mean, stdev, param in zip(means, stds, params):\n",
        "      print(\"%f (%f) with: %r\" % (mean, stdev, param))\n",
        "\n",
        "  print(\"-----------------------------------------------------------\")\n",
        "  print()\n",
        "  print(\"Best Accuracy: %f using %s\" % (grid_result.best_score_, grid_result.best_params_))\n",
        "  print()\n",
        "\n",
        "  y_true, y_pred = y_test, grid_result.predict(X_test)\n",
        "  print(\"Detailed classification report:\")\n",
        "  print()\n",
        "  print(classification_report(y_true, y_pred))\n",
        "\n",
        "  return grid_result.best_params_"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VRmS4WG0tiwr"
      },
      "source": [
        "Before actually using GridSearchCV to tune the hyperparameters of the classifiers, we want to look at **ROC curves** since there are roughly equal numbers of observations for each class after the resampling process.\n",
        "\n",
        "The area under the **ROC curve (AUC)** is a measure of how well a parameter can distinguish between two diagnostic groups, which makes it perfect for analyzing binary classification (\"0\" or \"1\" in our case) -- An area of 1 represents a perfect test."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1CoMDjXIIuNE",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 576
        },
        "outputId": "46a8d368-307c-47f9-f319-46506e05872f"
      },
      "source": [
        "classifiers = [LogisticRegression(random_state=RANSTA), \n",
        "               KNeighborsClassifier(), \n",
        "               RandomForestClassifier(random_state=RANSTA)]\n",
        "\n",
        "# Define a result table as a DataFrame\n",
        "result_table = pd.DataFrame(columns=['classifiers', 'fpr','tpr','auc'])\n",
        "\n",
        "# Train the models and record the results\n",
        "for cls in classifiers:\n",
        "    model = cls.fit(X_train, y_train)\n",
        "    yproba = model.predict_proba(X_test)[::,1]\n",
        "    \n",
        "    fpr, tpr, _ = roc_curve(y_test, yproba)\n",
        "    auc = roc_auc_score(y_test, yproba)\n",
        "    \n",
        "    result_table = result_table.append({'classifiers':cls.__class__.__name__,\n",
        "                                        'fpr':fpr, \n",
        "                                        'tpr':tpr, \n",
        "                                        'auc':auc}, ignore_index=True)\n",
        "\n",
        "# Set name of the classifiers as index labels\n",
        "result_table.set_index('classifiers', inplace=True)\n",
        "\n",
        "fig = plt.figure(figsize=(8,6))\n",
        "\n",
        "for i in result_table.index:\n",
        "    plt.plot(result_table.loc[i]['fpr'], \n",
        "             result_table.loc[i]['tpr'], \n",
        "             label=\"{}, AUC={:.3f}\".format(i, result_table.loc[i]['auc']))\n",
        "    \n",
        "plt.plot([0,1], [0,1], color='orange', linestyle='--')\n",
        "\n",
        "plt.xticks(np.arange(0.0, 1.1, step=0.1))\n",
        "plt.xlabel(\"False Positive Rate\", fontsize=15)\n",
        "\n",
        "plt.yticks(np.arange(0.0, 1.1, step=0.1))\n",
        "plt.ylabel(\"True Positive Rate\", fontsize=15)\n",
        "\n",
        "plt.title('ROC Curve Analysis', fontweight='bold', fontsize=15)\n",
        "plt.legend(prop={'size':13}, loc='lower right')\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_logistic.py:764: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
            "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "Please also refer to the documentation for alternative solver options:\n",
            "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
            "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iU4JeBNmqUkp"
      },
      "source": [
        "We also want to use **Hyperparameter Tuning with GridSearchCV** to get the best parameters for each of the following models.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AKtNv9YbvZTP"
      },
      "source": [
        "### 6.1. **KNN**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QqE3MmXwqEnz",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "fb7400ed-553a-4114-a9f2-68a483de0c4c"
      },
      "source": [
        "# GridSearchCV\n",
        "best_knn = hyperTuning(KNeighborsClassifier(), \n",
        "            dict(leaf_size=list(range(1,50)), n_neighbors=list(range(1,30)), p=[1,2]),\n",
        "            X_train, y_train, X_test, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.983008 (0.006468) with: {'leaf_size': 1, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 1, 'n_neighbors': 1, 'p': 2}\n",
            "0.980377 (0.007923) with: {'leaf_size': 1, 'n_neighbors': 2, 'p': 1}\n",
            "0.981095 (0.008251) with: {'leaf_size': 1, 'n_neighbors': 2, 'p': 2}\n",
            "0.968174 (0.010594) with: {'leaf_size': 1, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 1, 'n_neighbors': 3, 'p': 2}\n",
            "0.962669 (0.011590) with: {'leaf_size': 1, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 1, 'n_neighbors': 4, 'p': 2}\n",
            "0.946398 (0.013073) with: {'leaf_size': 1, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 1, 'n_neighbors': 5, 'p': 2}\n",
            "0.944722 (0.010459) with: {'leaf_size': 1, 'n_neighbors': 6, 'p': 1}\n",
            "0.937544 (0.013850) with: {'leaf_size': 1, 'n_neighbors': 6, 'p': 2}\n",
            "0.930842 (0.012649) with: {'leaf_size': 1, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 1, 'n_neighbors': 7, 'p': 2}\n",
            "0.926775 (0.012264) with: {'leaf_size': 1, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 1, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.013942) with: {'leaf_size': 1, 'n_neighbors': 9, 'p': 1}\n",
            "0.904759 (0.015407) with: {'leaf_size': 1, 'n_neighbors': 9, 'p': 2}\n",
            "0.916246 (0.011235) with: {'leaf_size': 1, 'n_neighbors': 10, 'p': 1}\n",
            "0.902368 (0.014628) with: {'leaf_size': 1, 'n_neighbors': 10, 'p': 2}\n",
            "0.905237 (0.013524) with: {'leaf_size': 1, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 1, 'n_neighbors': 11, 'p': 2}\n",
            "0.901889 (0.012446) with: {'leaf_size': 1, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013764) with: {'leaf_size': 1, 'n_neighbors': 12, 'p': 2}\n",
            "0.895907 (0.014807) with: {'leaf_size': 1, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 1, 'n_neighbors': 13, 'p': 2}\n",
            "0.890401 (0.015033) with: {'leaf_size': 1, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 1, 'n_neighbors': 14, 'p': 2}\n",
            "0.884899 (0.014347) with: {'leaf_size': 1, 'n_neighbors': 15, 'p': 1}\n",
            "0.865996 (0.017388) with: {'leaf_size': 1, 'n_neighbors': 15, 'p': 2}\n",
            "0.877722 (0.013210) with: {'leaf_size': 1, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 1, 'n_neighbors': 16, 'p': 2}\n",
            "0.872936 (0.016898) with: {'leaf_size': 1, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 1, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.012959) with: {'leaf_size': 1, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 1, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.018105) with: {'leaf_size': 1, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 1, 'n_neighbors': 19, 'p': 2}\n",
            "0.861927 (0.017401) with: {'leaf_size': 1, 'n_neighbors': 20, 'p': 1}\n",
            "0.842545 (0.019616) with: {'leaf_size': 1, 'n_neighbors': 20, 'p': 2}\n",
            "0.857620 (0.015711) with: {'leaf_size': 1, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017903) with: {'leaf_size': 1, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016165) with: {'leaf_size': 1, 'n_neighbors': 22, 'p': 1}\n",
            "0.835847 (0.014629) with: {'leaf_size': 1, 'n_neighbors': 22, 'p': 2}\n",
            "0.850681 (0.013468) with: {'leaf_size': 1, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015464) with: {'leaf_size': 1, 'n_neighbors': 23, 'p': 2}\n",
            "0.850204 (0.013411) with: {'leaf_size': 1, 'n_neighbors': 24, 'p': 1}\n",
            "0.827474 (0.015454) with: {'leaf_size': 1, 'n_neighbors': 24, 'p': 2}\n",
            "0.846854 (0.011000) with: {'leaf_size': 1, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 1, 'n_neighbors': 25, 'p': 2}\n",
            "0.843744 (0.012736) with: {'leaf_size': 1, 'n_neighbors': 26, 'p': 1}\n",
            "0.819578 (0.010753) with: {'leaf_size': 1, 'n_neighbors': 26, 'p': 2}\n",
            "0.844222 (0.011831) with: {'leaf_size': 1, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 1, 'n_neighbors': 27, 'p': 2}\n",
            "0.840393 (0.014856) with: {'leaf_size': 1, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 1, 'n_neighbors': 28, 'p': 2}\n",
            "0.838479 (0.013868) with: {'leaf_size': 1, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 1, 'n_neighbors': 29, 'p': 2}\n",
            "0.983008 (0.006468) with: {'leaf_size': 2, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 2, 'n_neighbors': 1, 'p': 2}\n",
            "0.980377 (0.007923) with: {'leaf_size': 2, 'n_neighbors': 2, 'p': 1}\n",
            "0.981095 (0.008251) with: {'leaf_size': 2, 'n_neighbors': 2, 'p': 2}\n",
            "0.968174 (0.010594) with: {'leaf_size': 2, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 2, 'n_neighbors': 3, 'p': 2}\n",
            "0.962908 (0.011088) with: {'leaf_size': 2, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 2, 'n_neighbors': 4, 'p': 2}\n",
            "0.947115 (0.012952) with: {'leaf_size': 2, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 2, 'n_neighbors': 5, 'p': 2}\n",
            "0.944723 (0.010619) with: {'leaf_size': 2, 'n_neighbors': 6, 'p': 1}\n",
            "0.937544 (0.013850) with: {'leaf_size': 2, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.013074) with: {'leaf_size': 2, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 2, 'n_neighbors': 7, 'p': 2}\n",
            "0.927254 (0.011785) with: {'leaf_size': 2, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 2, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.015062) with: {'leaf_size': 2, 'n_neighbors': 9, 'p': 1}\n",
            "0.904759 (0.015144) with: {'leaf_size': 2, 'n_neighbors': 9, 'p': 2}\n",
            "0.915049 (0.011700) with: {'leaf_size': 2, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 2, 'n_neighbors': 10, 'p': 2}\n",
            "0.903562 (0.014421) with: {'leaf_size': 2, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 2, 'n_neighbors': 11, 'p': 2}\n",
            "0.902367 (0.012345) with: {'leaf_size': 2, 'n_neighbors': 12, 'p': 1}\n",
            "0.889206 (0.013924) with: {'leaf_size': 2, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014531) with: {'leaf_size': 2, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 2, 'n_neighbors': 13, 'p': 2}\n",
            "0.889444 (0.015487) with: {'leaf_size': 2, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 2, 'n_neighbors': 14, 'p': 2}\n",
            "0.885139 (0.013919) with: {'leaf_size': 2, 'n_neighbors': 15, 'p': 1}\n",
            "0.865996 (0.017388) with: {'leaf_size': 2, 'n_neighbors': 15, 'p': 2}\n",
            "0.878439 (0.012657) with: {'leaf_size': 2, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 2, 'n_neighbors': 16, 'p': 2}\n",
            "0.871739 (0.015587) with: {'leaf_size': 2, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 2, 'n_neighbors': 17, 'p': 2}\n",
            "0.867672 (0.013358) with: {'leaf_size': 2, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 2, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.018073) with: {'leaf_size': 2, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 2, 'n_neighbors': 19, 'p': 2}\n",
            "0.861687 (0.017529) with: {'leaf_size': 2, 'n_neighbors': 20, 'p': 1}\n",
            "0.842545 (0.019616) with: {'leaf_size': 2, 'n_neighbors': 20, 'p': 2}\n",
            "0.858338 (0.017292) with: {'leaf_size': 2, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 2, 'n_neighbors': 21, 'p': 2}\n",
            "0.856183 (0.016507) with: {'leaf_size': 2, 'n_neighbors': 22, 'p': 1}\n",
            "0.835847 (0.014629) with: {'leaf_size': 2, 'n_neighbors': 22, 'p': 2}\n",
            "0.851878 (0.013560) with: {'leaf_size': 2, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015464) with: {'leaf_size': 2, 'n_neighbors': 23, 'p': 2}\n",
            "0.850204 (0.013532) with: {'leaf_size': 2, 'n_neighbors': 24, 'p': 1}\n",
            "0.826995 (0.014756) with: {'leaf_size': 2, 'n_neighbors': 24, 'p': 2}\n",
            "0.846855 (0.010836) with: {'leaf_size': 2, 'n_neighbors': 25, 'p': 1}\n",
            "0.824124 (0.012759) with: {'leaf_size': 2, 'n_neighbors': 25, 'p': 2}\n",
            "0.843265 (0.012426) with: {'leaf_size': 2, 'n_neighbors': 26, 'p': 1}\n",
            "0.819578 (0.010753) with: {'leaf_size': 2, 'n_neighbors': 26, 'p': 2}\n",
            "0.845179 (0.010635) with: {'leaf_size': 2, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 2, 'n_neighbors': 27, 'p': 2}\n",
            "0.839915 (0.013791) with: {'leaf_size': 2, 'n_neighbors': 28, 'p': 1}\n",
            "0.816947 (0.015756) with: {'leaf_size': 2, 'n_neighbors': 28, 'p': 2}\n",
            "0.838479 (0.013821) with: {'leaf_size': 2, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018520) with: {'leaf_size': 2, 'n_neighbors': 29, 'p': 2}\n",
            "0.983008 (0.006468) with: {'leaf_size': 3, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 3, 'n_neighbors': 1, 'p': 2}\n",
            "0.980377 (0.007923) with: {'leaf_size': 3, 'n_neighbors': 2, 'p': 1}\n",
            "0.981095 (0.008251) with: {'leaf_size': 3, 'n_neighbors': 2, 'p': 2}\n",
            "0.968174 (0.010594) with: {'leaf_size': 3, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 3, 'n_neighbors': 3, 'p': 2}\n",
            "0.962908 (0.011088) with: {'leaf_size': 3, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 3, 'n_neighbors': 4, 'p': 2}\n",
            "0.947115 (0.012952) with: {'leaf_size': 3, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 3, 'n_neighbors': 5, 'p': 2}\n",
            "0.944723 (0.010619) with: {'leaf_size': 3, 'n_neighbors': 6, 'p': 1}\n",
            "0.937544 (0.013850) with: {'leaf_size': 3, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.013074) with: {'leaf_size': 3, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 3, 'n_neighbors': 7, 'p': 2}\n",
            "0.927254 (0.011785) with: {'leaf_size': 3, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 3, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.015062) with: {'leaf_size': 3, 'n_neighbors': 9, 'p': 1}\n",
            "0.904759 (0.015144) with: {'leaf_size': 3, 'n_neighbors': 9, 'p': 2}\n",
            "0.915049 (0.011700) with: {'leaf_size': 3, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 3, 'n_neighbors': 10, 'p': 2}\n",
            "0.903562 (0.014421) with: {'leaf_size': 3, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 3, 'n_neighbors': 11, 'p': 2}\n",
            "0.902367 (0.012345) with: {'leaf_size': 3, 'n_neighbors': 12, 'p': 1}\n",
            "0.889206 (0.013924) with: {'leaf_size': 3, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014531) with: {'leaf_size': 3, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 3, 'n_neighbors': 13, 'p': 2}\n",
            "0.889444 (0.015487) with: {'leaf_size': 3, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 3, 'n_neighbors': 14, 'p': 2}\n",
            "0.885139 (0.013919) with: {'leaf_size': 3, 'n_neighbors': 15, 'p': 1}\n",
            "0.865996 (0.017388) with: {'leaf_size': 3, 'n_neighbors': 15, 'p': 2}\n",
            "0.878439 (0.012657) with: {'leaf_size': 3, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 3, 'n_neighbors': 16, 'p': 2}\n",
            "0.871739 (0.015587) with: {'leaf_size': 3, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 3, 'n_neighbors': 17, 'p': 2}\n",
            "0.867672 (0.013358) with: {'leaf_size': 3, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 3, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.018073) with: {'leaf_size': 3, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 3, 'n_neighbors': 19, 'p': 2}\n",
            "0.861687 (0.017529) with: {'leaf_size': 3, 'n_neighbors': 20, 'p': 1}\n",
            "0.842545 (0.019616) with: {'leaf_size': 3, 'n_neighbors': 20, 'p': 2}\n",
            "0.858338 (0.017292) with: {'leaf_size': 3, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 3, 'n_neighbors': 21, 'p': 2}\n",
            "0.856183 (0.016507) with: {'leaf_size': 3, 'n_neighbors': 22, 'p': 1}\n",
            "0.835847 (0.014629) with: {'leaf_size': 3, 'n_neighbors': 22, 'p': 2}\n",
            "0.851878 (0.013560) with: {'leaf_size': 3, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015464) with: {'leaf_size': 3, 'n_neighbors': 23, 'p': 2}\n",
            "0.850204 (0.013532) with: {'leaf_size': 3, 'n_neighbors': 24, 'p': 1}\n",
            "0.826995 (0.014756) with: {'leaf_size': 3, 'n_neighbors': 24, 'p': 2}\n",
            "0.846855 (0.010836) with: {'leaf_size': 3, 'n_neighbors': 25, 'p': 1}\n",
            "0.824124 (0.012759) with: {'leaf_size': 3, 'n_neighbors': 25, 'p': 2}\n",
            "0.843265 (0.012426) with: {'leaf_size': 3, 'n_neighbors': 26, 'p': 1}\n",
            "0.819578 (0.010753) with: {'leaf_size': 3, 'n_neighbors': 26, 'p': 2}\n",
            "0.845179 (0.010635) with: {'leaf_size': 3, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 3, 'n_neighbors': 27, 'p': 2}\n",
            "0.839915 (0.013791) with: {'leaf_size': 3, 'n_neighbors': 28, 'p': 1}\n",
            "0.816947 (0.015756) with: {'leaf_size': 3, 'n_neighbors': 28, 'p': 2}\n",
            "0.838479 (0.013821) with: {'leaf_size': 3, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018520) with: {'leaf_size': 3, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 4, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 4, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 4, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 4, 'n_neighbors': 2, 'p': 2}\n",
            "0.968174 (0.010594) with: {'leaf_size': 4, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 4, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.011738) with: {'leaf_size': 4, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 4, 'n_neighbors': 4, 'p': 2}\n",
            "0.946876 (0.012970) with: {'leaf_size': 4, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 4, 'n_neighbors': 5, 'p': 2}\n",
            "0.944723 (0.011247) with: {'leaf_size': 4, 'n_neighbors': 6, 'p': 1}\n",
            "0.937544 (0.013850) with: {'leaf_size': 4, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.013074) with: {'leaf_size': 4, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 4, 'n_neighbors': 7, 'p': 2}\n",
            "0.927733 (0.010904) with: {'leaf_size': 4, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 4, 'n_neighbors': 8, 'p': 2}\n",
            "0.916246 (0.015069) with: {'leaf_size': 4, 'n_neighbors': 9, 'p': 1}\n",
            "0.904759 (0.014988) with: {'leaf_size': 4, 'n_neighbors': 9, 'p': 2}\n",
            "0.916007 (0.012091) with: {'leaf_size': 4, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 4, 'n_neighbors': 10, 'p': 2}\n",
            "0.904280 (0.014094) with: {'leaf_size': 4, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 4, 'n_neighbors': 11, 'p': 2}\n",
            "0.902367 (0.011580) with: {'leaf_size': 4, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013764) with: {'leaf_size': 4, 'n_neighbors': 12, 'p': 2}\n",
            "0.897103 (0.013840) with: {'leaf_size': 4, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 4, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.014061) with: {'leaf_size': 4, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 4, 'n_neighbors': 14, 'p': 2}\n",
            "0.885377 (0.014025) with: {'leaf_size': 4, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 4, 'n_neighbors': 15, 'p': 2}\n",
            "0.877961 (0.013196) with: {'leaf_size': 4, 'n_neighbors': 16, 'p': 1}\n",
            "0.860729 (0.017203) with: {'leaf_size': 4, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016953) with: {'leaf_size': 4, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 4, 'n_neighbors': 17, 'p': 2}\n",
            "0.867912 (0.013262) with: {'leaf_size': 4, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 4, 'n_neighbors': 18, 'p': 2}\n",
            "0.865995 (0.017332) with: {'leaf_size': 4, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 4, 'n_neighbors': 19, 'p': 2}\n",
            "0.862405 (0.017144) with: {'leaf_size': 4, 'n_neighbors': 20, 'p': 1}\n",
            "0.842545 (0.019616) with: {'leaf_size': 4, 'n_neighbors': 20, 'p': 2}\n",
            "0.857621 (0.015111) with: {'leaf_size': 4, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 4, 'n_neighbors': 21, 'p': 2}\n",
            "0.855706 (0.016131) with: {'leaf_size': 4, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.015017) with: {'leaf_size': 4, 'n_neighbors': 22, 'p': 2}\n",
            "0.853074 (0.012859) with: {'leaf_size': 4, 'n_neighbors': 23, 'p': 1}\n",
            "0.830824 (0.015437) with: {'leaf_size': 4, 'n_neighbors': 23, 'p': 2}\n",
            "0.849964 (0.012664) with: {'leaf_size': 4, 'n_neighbors': 24, 'p': 1}\n",
            "0.827474 (0.015230) with: {'leaf_size': 4, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.012386) with: {'leaf_size': 4, 'n_neighbors': 25, 'p': 1}\n",
            "0.824124 (0.012759) with: {'leaf_size': 4, 'n_neighbors': 25, 'p': 2}\n",
            "0.843744 (0.012232) with: {'leaf_size': 4, 'n_neighbors': 26, 'p': 1}\n",
            "0.819578 (0.010753) with: {'leaf_size': 4, 'n_neighbors': 26, 'p': 2}\n",
            "0.845658 (0.010647) with: {'leaf_size': 4, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014423) with: {'leaf_size': 4, 'n_neighbors': 27, 'p': 2}\n",
            "0.840154 (0.014960) with: {'leaf_size': 4, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 4, 'n_neighbors': 28, 'p': 2}\n",
            "0.838479 (0.014349) with: {'leaf_size': 4, 'n_neighbors': 29, 'p': 1}\n",
            "0.811203 (0.018693) with: {'leaf_size': 4, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 5, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 5, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 5, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 5, 'n_neighbors': 2, 'p': 2}\n",
            "0.968174 (0.010594) with: {'leaf_size': 5, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 5, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.011738) with: {'leaf_size': 5, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 5, 'n_neighbors': 4, 'p': 2}\n",
            "0.946876 (0.012970) with: {'leaf_size': 5, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 5, 'n_neighbors': 5, 'p': 2}\n",
            "0.944723 (0.011247) with: {'leaf_size': 5, 'n_neighbors': 6, 'p': 1}\n",
            "0.937544 (0.013850) with: {'leaf_size': 5, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.013074) with: {'leaf_size': 5, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 5, 'n_neighbors': 7, 'p': 2}\n",
            "0.927733 (0.010904) with: {'leaf_size': 5, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 5, 'n_neighbors': 8, 'p': 2}\n",
            "0.916246 (0.015069) with: {'leaf_size': 5, 'n_neighbors': 9, 'p': 1}\n",
            "0.904759 (0.014988) with: {'leaf_size': 5, 'n_neighbors': 9, 'p': 2}\n",
            "0.916007 (0.012091) with: {'leaf_size': 5, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 5, 'n_neighbors': 10, 'p': 2}\n",
            "0.904280 (0.014094) with: {'leaf_size': 5, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 5, 'n_neighbors': 11, 'p': 2}\n",
            "0.902367 (0.011580) with: {'leaf_size': 5, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013764) with: {'leaf_size': 5, 'n_neighbors': 12, 'p': 2}\n",
            "0.897103 (0.013840) with: {'leaf_size': 5, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 5, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.014061) with: {'leaf_size': 5, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 5, 'n_neighbors': 14, 'p': 2}\n",
            "0.885377 (0.014025) with: {'leaf_size': 5, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 5, 'n_neighbors': 15, 'p': 2}\n",
            "0.877961 (0.013196) with: {'leaf_size': 5, 'n_neighbors': 16, 'p': 1}\n",
            "0.860729 (0.017203) with: {'leaf_size': 5, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016953) with: {'leaf_size': 5, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 5, 'n_neighbors': 17, 'p': 2}\n",
            "0.867912 (0.013262) with: {'leaf_size': 5, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 5, 'n_neighbors': 18, 'p': 2}\n",
            "0.865995 (0.017332) with: {'leaf_size': 5, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 5, 'n_neighbors': 19, 'p': 2}\n",
            "0.862405 (0.017144) with: {'leaf_size': 5, 'n_neighbors': 20, 'p': 1}\n",
            "0.842545 (0.019616) with: {'leaf_size': 5, 'n_neighbors': 20, 'p': 2}\n",
            "0.857621 (0.015111) with: {'leaf_size': 5, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 5, 'n_neighbors': 21, 'p': 2}\n",
            "0.855706 (0.016131) with: {'leaf_size': 5, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.015017) with: {'leaf_size': 5, 'n_neighbors': 22, 'p': 2}\n",
            "0.853074 (0.012859) with: {'leaf_size': 5, 'n_neighbors': 23, 'p': 1}\n",
            "0.830824 (0.015437) with: {'leaf_size': 5, 'n_neighbors': 23, 'p': 2}\n",
            "0.849964 (0.012664) with: {'leaf_size': 5, 'n_neighbors': 24, 'p': 1}\n",
            "0.827474 (0.015230) with: {'leaf_size': 5, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.012386) with: {'leaf_size': 5, 'n_neighbors': 25, 'p': 1}\n",
            "0.824124 (0.012759) with: {'leaf_size': 5, 'n_neighbors': 25, 'p': 2}\n",
            "0.843744 (0.012232) with: {'leaf_size': 5, 'n_neighbors': 26, 'p': 1}\n",
            "0.819578 (0.010753) with: {'leaf_size': 5, 'n_neighbors': 26, 'p': 2}\n",
            "0.845658 (0.010647) with: {'leaf_size': 5, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014423) with: {'leaf_size': 5, 'n_neighbors': 27, 'p': 2}\n",
            "0.840154 (0.014960) with: {'leaf_size': 5, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 5, 'n_neighbors': 28, 'p': 2}\n",
            "0.838479 (0.014349) with: {'leaf_size': 5, 'n_neighbors': 29, 'p': 1}\n",
            "0.811203 (0.018693) with: {'leaf_size': 5, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 6, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 6, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 6, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 6, 'n_neighbors': 2, 'p': 2}\n",
            "0.968174 (0.010594) with: {'leaf_size': 6, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 6, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.011738) with: {'leaf_size': 6, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 6, 'n_neighbors': 4, 'p': 2}\n",
            "0.946876 (0.012970) with: {'leaf_size': 6, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 6, 'n_neighbors': 5, 'p': 2}\n",
            "0.944723 (0.011247) with: {'leaf_size': 6, 'n_neighbors': 6, 'p': 1}\n",
            "0.937544 (0.013850) with: {'leaf_size': 6, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.013074) with: {'leaf_size': 6, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 6, 'n_neighbors': 7, 'p': 2}\n",
            "0.927733 (0.010904) with: {'leaf_size': 6, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 6, 'n_neighbors': 8, 'p': 2}\n",
            "0.916246 (0.015069) with: {'leaf_size': 6, 'n_neighbors': 9, 'p': 1}\n",
            "0.904759 (0.014988) with: {'leaf_size': 6, 'n_neighbors': 9, 'p': 2}\n",
            "0.916007 (0.012091) with: {'leaf_size': 6, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 6, 'n_neighbors': 10, 'p': 2}\n",
            "0.904280 (0.014094) with: {'leaf_size': 6, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 6, 'n_neighbors': 11, 'p': 2}\n",
            "0.902367 (0.011580) with: {'leaf_size': 6, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013764) with: {'leaf_size': 6, 'n_neighbors': 12, 'p': 2}\n",
            "0.897103 (0.013840) with: {'leaf_size': 6, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 6, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.014061) with: {'leaf_size': 6, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 6, 'n_neighbors': 14, 'p': 2}\n",
            "0.885377 (0.014025) with: {'leaf_size': 6, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 6, 'n_neighbors': 15, 'p': 2}\n",
            "0.877961 (0.013196) with: {'leaf_size': 6, 'n_neighbors': 16, 'p': 1}\n",
            "0.860729 (0.017203) with: {'leaf_size': 6, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016953) with: {'leaf_size': 6, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 6, 'n_neighbors': 17, 'p': 2}\n",
            "0.867912 (0.013262) with: {'leaf_size': 6, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 6, 'n_neighbors': 18, 'p': 2}\n",
            "0.865995 (0.017332) with: {'leaf_size': 6, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 6, 'n_neighbors': 19, 'p': 2}\n",
            "0.862405 (0.017144) with: {'leaf_size': 6, 'n_neighbors': 20, 'p': 1}\n",
            "0.842545 (0.019616) with: {'leaf_size': 6, 'n_neighbors': 20, 'p': 2}\n",
            "0.857621 (0.015111) with: {'leaf_size': 6, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 6, 'n_neighbors': 21, 'p': 2}\n",
            "0.855706 (0.016131) with: {'leaf_size': 6, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.015017) with: {'leaf_size': 6, 'n_neighbors': 22, 'p': 2}\n",
            "0.853074 (0.012859) with: {'leaf_size': 6, 'n_neighbors': 23, 'p': 1}\n",
            "0.830824 (0.015437) with: {'leaf_size': 6, 'n_neighbors': 23, 'p': 2}\n",
            "0.849964 (0.012664) with: {'leaf_size': 6, 'n_neighbors': 24, 'p': 1}\n",
            "0.827474 (0.015230) with: {'leaf_size': 6, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.012386) with: {'leaf_size': 6, 'n_neighbors': 25, 'p': 1}\n",
            "0.824124 (0.012759) with: {'leaf_size': 6, 'n_neighbors': 25, 'p': 2}\n",
            "0.843744 (0.012232) with: {'leaf_size': 6, 'n_neighbors': 26, 'p': 1}\n",
            "0.819578 (0.010753) with: {'leaf_size': 6, 'n_neighbors': 26, 'p': 2}\n",
            "0.845658 (0.010647) with: {'leaf_size': 6, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014423) with: {'leaf_size': 6, 'n_neighbors': 27, 'p': 2}\n",
            "0.840154 (0.014960) with: {'leaf_size': 6, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 6, 'n_neighbors': 28, 'p': 2}\n",
            "0.838479 (0.014349) with: {'leaf_size': 6, 'n_neighbors': 29, 'p': 1}\n",
            "0.811203 (0.018693) with: {'leaf_size': 6, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 7, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 7, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 7, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 7, 'n_neighbors': 2, 'p': 2}\n",
            "0.968174 (0.010594) with: {'leaf_size': 7, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 7, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.011738) with: {'leaf_size': 7, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 7, 'n_neighbors': 4, 'p': 2}\n",
            "0.946876 (0.012970) with: {'leaf_size': 7, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 7, 'n_neighbors': 5, 'p': 2}\n",
            "0.944723 (0.011247) with: {'leaf_size': 7, 'n_neighbors': 6, 'p': 1}\n",
            "0.937544 (0.013850) with: {'leaf_size': 7, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.013074) with: {'leaf_size': 7, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 7, 'n_neighbors': 7, 'p': 2}\n",
            "0.927733 (0.010904) with: {'leaf_size': 7, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 7, 'n_neighbors': 8, 'p': 2}\n",
            "0.916246 (0.015069) with: {'leaf_size': 7, 'n_neighbors': 9, 'p': 1}\n",
            "0.904759 (0.014988) with: {'leaf_size': 7, 'n_neighbors': 9, 'p': 2}\n",
            "0.916007 (0.012091) with: {'leaf_size': 7, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 7, 'n_neighbors': 10, 'p': 2}\n",
            "0.904280 (0.014094) with: {'leaf_size': 7, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 7, 'n_neighbors': 11, 'p': 2}\n",
            "0.902367 (0.011580) with: {'leaf_size': 7, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013764) with: {'leaf_size': 7, 'n_neighbors': 12, 'p': 2}\n",
            "0.897103 (0.013840) with: {'leaf_size': 7, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 7, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.014061) with: {'leaf_size': 7, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 7, 'n_neighbors': 14, 'p': 2}\n",
            "0.885377 (0.014025) with: {'leaf_size': 7, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 7, 'n_neighbors': 15, 'p': 2}\n",
            "0.877961 (0.013196) with: {'leaf_size': 7, 'n_neighbors': 16, 'p': 1}\n",
            "0.860729 (0.017203) with: {'leaf_size': 7, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016953) with: {'leaf_size': 7, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 7, 'n_neighbors': 17, 'p': 2}\n",
            "0.867912 (0.013262) with: {'leaf_size': 7, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 7, 'n_neighbors': 18, 'p': 2}\n",
            "0.865995 (0.017332) with: {'leaf_size': 7, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 7, 'n_neighbors': 19, 'p': 2}\n",
            "0.862405 (0.017144) with: {'leaf_size': 7, 'n_neighbors': 20, 'p': 1}\n",
            "0.842545 (0.019616) with: {'leaf_size': 7, 'n_neighbors': 20, 'p': 2}\n",
            "0.857621 (0.015111) with: {'leaf_size': 7, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 7, 'n_neighbors': 21, 'p': 2}\n",
            "0.855706 (0.016131) with: {'leaf_size': 7, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.015017) with: {'leaf_size': 7, 'n_neighbors': 22, 'p': 2}\n",
            "0.853074 (0.012859) with: {'leaf_size': 7, 'n_neighbors': 23, 'p': 1}\n",
            "0.830824 (0.015437) with: {'leaf_size': 7, 'n_neighbors': 23, 'p': 2}\n",
            "0.849964 (0.012664) with: {'leaf_size': 7, 'n_neighbors': 24, 'p': 1}\n",
            "0.827474 (0.015230) with: {'leaf_size': 7, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.012386) with: {'leaf_size': 7, 'n_neighbors': 25, 'p': 1}\n",
            "0.824124 (0.012759) with: {'leaf_size': 7, 'n_neighbors': 25, 'p': 2}\n",
            "0.843744 (0.012232) with: {'leaf_size': 7, 'n_neighbors': 26, 'p': 1}\n",
            "0.819578 (0.010753) with: {'leaf_size': 7, 'n_neighbors': 26, 'p': 2}\n",
            "0.845658 (0.010647) with: {'leaf_size': 7, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014423) with: {'leaf_size': 7, 'n_neighbors': 27, 'p': 2}\n",
            "0.840154 (0.014960) with: {'leaf_size': 7, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 7, 'n_neighbors': 28, 'p': 2}\n",
            "0.838479 (0.014349) with: {'leaf_size': 7, 'n_neighbors': 29, 'p': 1}\n",
            "0.811203 (0.018693) with: {'leaf_size': 7, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 8, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 8, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 8, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 8, 'n_neighbors': 2, 'p': 2}\n",
            "0.968414 (0.010362) with: {'leaf_size': 8, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 8, 'n_neighbors': 3, 'p': 2}\n",
            "0.962669 (0.011391) with: {'leaf_size': 8, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 8, 'n_neighbors': 4, 'p': 2}\n",
            "0.946637 (0.013110) with: {'leaf_size': 8, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 8, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010863) with: {'leaf_size': 8, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 8, 'n_neighbors': 6, 'p': 2}\n",
            "0.930842 (0.012604) with: {'leaf_size': 8, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 8, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.012396) with: {'leaf_size': 8, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 8, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014227) with: {'leaf_size': 8, 'n_neighbors': 9, 'p': 1}\n",
            "0.905238 (0.015042) with: {'leaf_size': 8, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.010548) with: {'leaf_size': 8, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 8, 'n_neighbors': 10, 'p': 2}\n",
            "0.904758 (0.014213) with: {'leaf_size': 8, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 8, 'n_neighbors': 11, 'p': 2}\n",
            "0.903084 (0.011310) with: {'leaf_size': 8, 'n_neighbors': 12, 'p': 1}\n",
            "0.889206 (0.013924) with: {'leaf_size': 8, 'n_neighbors': 12, 'p': 2}\n",
            "0.895428 (0.013630) with: {'leaf_size': 8, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 8, 'n_neighbors': 13, 'p': 2}\n",
            "0.890640 (0.014972) with: {'leaf_size': 8, 'n_neighbors': 14, 'p': 1}\n",
            "0.873890 (0.015198) with: {'leaf_size': 8, 'n_neighbors': 14, 'p': 2}\n",
            "0.885617 (0.013787) with: {'leaf_size': 8, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 8, 'n_neighbors': 15, 'p': 2}\n",
            "0.877242 (0.012671) with: {'leaf_size': 8, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 8, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016123) with: {'leaf_size': 8, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 8, 'n_neighbors': 17, 'p': 2}\n",
            "0.868629 (0.013716) with: {'leaf_size': 8, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 8, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017130) with: {'leaf_size': 8, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 8, 'n_neighbors': 19, 'p': 2}\n",
            "0.863123 (0.016725) with: {'leaf_size': 8, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 8, 'n_neighbors': 20, 'p': 2}\n",
            "0.858098 (0.016830) with: {'leaf_size': 8, 'n_neighbors': 21, 'p': 1}\n",
            "0.836085 (0.018257) with: {'leaf_size': 8, 'n_neighbors': 21, 'p': 2}\n",
            "0.855705 (0.016417) with: {'leaf_size': 8, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 8, 'n_neighbors': 22, 'p': 2}\n",
            "0.851638 (0.012808) with: {'leaf_size': 8, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 8, 'n_neighbors': 23, 'p': 2}\n",
            "0.849726 (0.012639) with: {'leaf_size': 8, 'n_neighbors': 24, 'p': 1}\n",
            "0.826995 (0.014756) with: {'leaf_size': 8, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010689) with: {'leaf_size': 8, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 8, 'n_neighbors': 25, 'p': 2}\n",
            "0.844461 (0.011257) with: {'leaf_size': 8, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 8, 'n_neighbors': 26, 'p': 2}\n",
            "0.845418 (0.011065) with: {'leaf_size': 8, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014815) with: {'leaf_size': 8, 'n_neighbors': 27, 'p': 2}\n",
            "0.839436 (0.013707) with: {'leaf_size': 8, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 8, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012581) with: {'leaf_size': 8, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 8, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 9, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 9, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 9, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 9, 'n_neighbors': 2, 'p': 2}\n",
            "0.968414 (0.010362) with: {'leaf_size': 9, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 9, 'n_neighbors': 3, 'p': 2}\n",
            "0.962669 (0.011391) with: {'leaf_size': 9, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 9, 'n_neighbors': 4, 'p': 2}\n",
            "0.946637 (0.013110) with: {'leaf_size': 9, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 9, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010863) with: {'leaf_size': 9, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 9, 'n_neighbors': 6, 'p': 2}\n",
            "0.930842 (0.012604) with: {'leaf_size': 9, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 9, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.012396) with: {'leaf_size': 9, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 9, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014227) with: {'leaf_size': 9, 'n_neighbors': 9, 'p': 1}\n",
            "0.905238 (0.015042) with: {'leaf_size': 9, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.010548) with: {'leaf_size': 9, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 9, 'n_neighbors': 10, 'p': 2}\n",
            "0.904758 (0.014213) with: {'leaf_size': 9, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 9, 'n_neighbors': 11, 'p': 2}\n",
            "0.903084 (0.011310) with: {'leaf_size': 9, 'n_neighbors': 12, 'p': 1}\n",
            "0.889206 (0.013924) with: {'leaf_size': 9, 'n_neighbors': 12, 'p': 2}\n",
            "0.895428 (0.013630) with: {'leaf_size': 9, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 9, 'n_neighbors': 13, 'p': 2}\n",
            "0.890640 (0.014972) with: {'leaf_size': 9, 'n_neighbors': 14, 'p': 1}\n",
            "0.873890 (0.015198) with: {'leaf_size': 9, 'n_neighbors': 14, 'p': 2}\n",
            "0.885617 (0.013787) with: {'leaf_size': 9, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 9, 'n_neighbors': 15, 'p': 2}\n",
            "0.877242 (0.012671) with: {'leaf_size': 9, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 9, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016123) with: {'leaf_size': 9, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 9, 'n_neighbors': 17, 'p': 2}\n",
            "0.868629 (0.013716) with: {'leaf_size': 9, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 9, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017130) with: {'leaf_size': 9, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 9, 'n_neighbors': 19, 'p': 2}\n",
            "0.863123 (0.016725) with: {'leaf_size': 9, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 9, 'n_neighbors': 20, 'p': 2}\n",
            "0.858098 (0.016830) with: {'leaf_size': 9, 'n_neighbors': 21, 'p': 1}\n",
            "0.836085 (0.018257) with: {'leaf_size': 9, 'n_neighbors': 21, 'p': 2}\n",
            "0.855705 (0.016417) with: {'leaf_size': 9, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 9, 'n_neighbors': 22, 'p': 2}\n",
            "0.851638 (0.012808) with: {'leaf_size': 9, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 9, 'n_neighbors': 23, 'p': 2}\n",
            "0.849726 (0.012639) with: {'leaf_size': 9, 'n_neighbors': 24, 'p': 1}\n",
            "0.826995 (0.014756) with: {'leaf_size': 9, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010689) with: {'leaf_size': 9, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 9, 'n_neighbors': 25, 'p': 2}\n",
            "0.844461 (0.011257) with: {'leaf_size': 9, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 9, 'n_neighbors': 26, 'p': 2}\n",
            "0.845418 (0.011065) with: {'leaf_size': 9, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014815) with: {'leaf_size': 9, 'n_neighbors': 27, 'p': 2}\n",
            "0.839436 (0.013707) with: {'leaf_size': 9, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 9, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012581) with: {'leaf_size': 9, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 9, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 10, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 10, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 10, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 10, 'n_neighbors': 2, 'p': 2}\n",
            "0.968414 (0.010362) with: {'leaf_size': 10, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 10, 'n_neighbors': 3, 'p': 2}\n",
            "0.962669 (0.011391) with: {'leaf_size': 10, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 10, 'n_neighbors': 4, 'p': 2}\n",
            "0.946637 (0.013110) with: {'leaf_size': 10, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 10, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010863) with: {'leaf_size': 10, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 10, 'n_neighbors': 6, 'p': 2}\n",
            "0.930842 (0.012604) with: {'leaf_size': 10, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 10, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.012396) with: {'leaf_size': 10, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 10, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014227) with: {'leaf_size': 10, 'n_neighbors': 9, 'p': 1}\n",
            "0.905238 (0.015042) with: {'leaf_size': 10, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.010548) with: {'leaf_size': 10, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 10, 'n_neighbors': 10, 'p': 2}\n",
            "0.904758 (0.014213) with: {'leaf_size': 10, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 10, 'n_neighbors': 11, 'p': 2}\n",
            "0.903084 (0.011310) with: {'leaf_size': 10, 'n_neighbors': 12, 'p': 1}\n",
            "0.889206 (0.013924) with: {'leaf_size': 10, 'n_neighbors': 12, 'p': 2}\n",
            "0.895428 (0.013630) with: {'leaf_size': 10, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 10, 'n_neighbors': 13, 'p': 2}\n",
            "0.890640 (0.014972) with: {'leaf_size': 10, 'n_neighbors': 14, 'p': 1}\n",
            "0.873890 (0.015198) with: {'leaf_size': 10, 'n_neighbors': 14, 'p': 2}\n",
            "0.885617 (0.013787) with: {'leaf_size': 10, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 10, 'n_neighbors': 15, 'p': 2}\n",
            "0.877242 (0.012671) with: {'leaf_size': 10, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 10, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016123) with: {'leaf_size': 10, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 10, 'n_neighbors': 17, 'p': 2}\n",
            "0.868629 (0.013716) with: {'leaf_size': 10, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 10, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017130) with: {'leaf_size': 10, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 10, 'n_neighbors': 19, 'p': 2}\n",
            "0.863123 (0.016725) with: {'leaf_size': 10, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 10, 'n_neighbors': 20, 'p': 2}\n",
            "0.858098 (0.016830) with: {'leaf_size': 10, 'n_neighbors': 21, 'p': 1}\n",
            "0.836085 (0.018257) with: {'leaf_size': 10, 'n_neighbors': 21, 'p': 2}\n",
            "0.855705 (0.016417) with: {'leaf_size': 10, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 10, 'n_neighbors': 22, 'p': 2}\n",
            "0.851638 (0.012808) with: {'leaf_size': 10, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 10, 'n_neighbors': 23, 'p': 2}\n",
            "0.849726 (0.012639) with: {'leaf_size': 10, 'n_neighbors': 24, 'p': 1}\n",
            "0.826995 (0.014756) with: {'leaf_size': 10, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010689) with: {'leaf_size': 10, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 10, 'n_neighbors': 25, 'p': 2}\n",
            "0.844461 (0.011257) with: {'leaf_size': 10, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 10, 'n_neighbors': 26, 'p': 2}\n",
            "0.845418 (0.011065) with: {'leaf_size': 10, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014815) with: {'leaf_size': 10, 'n_neighbors': 27, 'p': 2}\n",
            "0.839436 (0.013707) with: {'leaf_size': 10, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 10, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012581) with: {'leaf_size': 10, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 10, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 11, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 11, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 11, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 11, 'n_neighbors': 2, 'p': 2}\n",
            "0.968414 (0.010362) with: {'leaf_size': 11, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 11, 'n_neighbors': 3, 'p': 2}\n",
            "0.962669 (0.011391) with: {'leaf_size': 11, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 11, 'n_neighbors': 4, 'p': 2}\n",
            "0.946637 (0.013110) with: {'leaf_size': 11, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 11, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010863) with: {'leaf_size': 11, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 11, 'n_neighbors': 6, 'p': 2}\n",
            "0.930842 (0.012604) with: {'leaf_size': 11, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 11, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.012396) with: {'leaf_size': 11, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 11, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014227) with: {'leaf_size': 11, 'n_neighbors': 9, 'p': 1}\n",
            "0.905238 (0.015042) with: {'leaf_size': 11, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.010548) with: {'leaf_size': 11, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 11, 'n_neighbors': 10, 'p': 2}\n",
            "0.904758 (0.014213) with: {'leaf_size': 11, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 11, 'n_neighbors': 11, 'p': 2}\n",
            "0.903084 (0.011310) with: {'leaf_size': 11, 'n_neighbors': 12, 'p': 1}\n",
            "0.889206 (0.013924) with: {'leaf_size': 11, 'n_neighbors': 12, 'p': 2}\n",
            "0.895428 (0.013630) with: {'leaf_size': 11, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 11, 'n_neighbors': 13, 'p': 2}\n",
            "0.890640 (0.014972) with: {'leaf_size': 11, 'n_neighbors': 14, 'p': 1}\n",
            "0.873890 (0.015198) with: {'leaf_size': 11, 'n_neighbors': 14, 'p': 2}\n",
            "0.885617 (0.013787) with: {'leaf_size': 11, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 11, 'n_neighbors': 15, 'p': 2}\n",
            "0.877242 (0.012671) with: {'leaf_size': 11, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 11, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016123) with: {'leaf_size': 11, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 11, 'n_neighbors': 17, 'p': 2}\n",
            "0.868629 (0.013716) with: {'leaf_size': 11, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 11, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017130) with: {'leaf_size': 11, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 11, 'n_neighbors': 19, 'p': 2}\n",
            "0.863123 (0.016725) with: {'leaf_size': 11, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 11, 'n_neighbors': 20, 'p': 2}\n",
            "0.858098 (0.016830) with: {'leaf_size': 11, 'n_neighbors': 21, 'p': 1}\n",
            "0.836085 (0.018257) with: {'leaf_size': 11, 'n_neighbors': 21, 'p': 2}\n",
            "0.855705 (0.016417) with: {'leaf_size': 11, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 11, 'n_neighbors': 22, 'p': 2}\n",
            "0.851638 (0.012808) with: {'leaf_size': 11, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 11, 'n_neighbors': 23, 'p': 2}\n",
            "0.849726 (0.012639) with: {'leaf_size': 11, 'n_neighbors': 24, 'p': 1}\n",
            "0.826995 (0.014756) with: {'leaf_size': 11, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010689) with: {'leaf_size': 11, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 11, 'n_neighbors': 25, 'p': 2}\n",
            "0.844461 (0.011257) with: {'leaf_size': 11, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 11, 'n_neighbors': 26, 'p': 2}\n",
            "0.845418 (0.011065) with: {'leaf_size': 11, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014815) with: {'leaf_size': 11, 'n_neighbors': 27, 'p': 2}\n",
            "0.839436 (0.013707) with: {'leaf_size': 11, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 11, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012581) with: {'leaf_size': 11, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 11, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 12, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 12, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 12, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 12, 'n_neighbors': 2, 'p': 2}\n",
            "0.968414 (0.010362) with: {'leaf_size': 12, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 12, 'n_neighbors': 3, 'p': 2}\n",
            "0.962669 (0.011391) with: {'leaf_size': 12, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 12, 'n_neighbors': 4, 'p': 2}\n",
            "0.946637 (0.013110) with: {'leaf_size': 12, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 12, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010863) with: {'leaf_size': 12, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 12, 'n_neighbors': 6, 'p': 2}\n",
            "0.930842 (0.012604) with: {'leaf_size': 12, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 12, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.012396) with: {'leaf_size': 12, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 12, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014227) with: {'leaf_size': 12, 'n_neighbors': 9, 'p': 1}\n",
            "0.905238 (0.015042) with: {'leaf_size': 12, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.010548) with: {'leaf_size': 12, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 12, 'n_neighbors': 10, 'p': 2}\n",
            "0.904758 (0.014213) with: {'leaf_size': 12, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 12, 'n_neighbors': 11, 'p': 2}\n",
            "0.903084 (0.011310) with: {'leaf_size': 12, 'n_neighbors': 12, 'p': 1}\n",
            "0.889206 (0.013924) with: {'leaf_size': 12, 'n_neighbors': 12, 'p': 2}\n",
            "0.895428 (0.013630) with: {'leaf_size': 12, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 12, 'n_neighbors': 13, 'p': 2}\n",
            "0.890640 (0.014972) with: {'leaf_size': 12, 'n_neighbors': 14, 'p': 1}\n",
            "0.873890 (0.015198) with: {'leaf_size': 12, 'n_neighbors': 14, 'p': 2}\n",
            "0.885617 (0.013787) with: {'leaf_size': 12, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 12, 'n_neighbors': 15, 'p': 2}\n",
            "0.877242 (0.012671) with: {'leaf_size': 12, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 12, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016123) with: {'leaf_size': 12, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 12, 'n_neighbors': 17, 'p': 2}\n",
            "0.868629 (0.013716) with: {'leaf_size': 12, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 12, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017130) with: {'leaf_size': 12, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 12, 'n_neighbors': 19, 'p': 2}\n",
            "0.863123 (0.016725) with: {'leaf_size': 12, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 12, 'n_neighbors': 20, 'p': 2}\n",
            "0.858098 (0.016830) with: {'leaf_size': 12, 'n_neighbors': 21, 'p': 1}\n",
            "0.836085 (0.018257) with: {'leaf_size': 12, 'n_neighbors': 21, 'p': 2}\n",
            "0.855705 (0.016417) with: {'leaf_size': 12, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 12, 'n_neighbors': 22, 'p': 2}\n",
            "0.851638 (0.012808) with: {'leaf_size': 12, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 12, 'n_neighbors': 23, 'p': 2}\n",
            "0.849726 (0.012639) with: {'leaf_size': 12, 'n_neighbors': 24, 'p': 1}\n",
            "0.826995 (0.014756) with: {'leaf_size': 12, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010689) with: {'leaf_size': 12, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 12, 'n_neighbors': 25, 'p': 2}\n",
            "0.844461 (0.011257) with: {'leaf_size': 12, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 12, 'n_neighbors': 26, 'p': 2}\n",
            "0.845418 (0.011065) with: {'leaf_size': 12, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014815) with: {'leaf_size': 12, 'n_neighbors': 27, 'p': 2}\n",
            "0.839436 (0.013707) with: {'leaf_size': 12, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 12, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012581) with: {'leaf_size': 12, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 12, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 13, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 13, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 13, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 13, 'n_neighbors': 2, 'p': 2}\n",
            "0.968414 (0.010362) with: {'leaf_size': 13, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 13, 'n_neighbors': 3, 'p': 2}\n",
            "0.962669 (0.011391) with: {'leaf_size': 13, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 13, 'n_neighbors': 4, 'p': 2}\n",
            "0.946637 (0.013110) with: {'leaf_size': 13, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 13, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010863) with: {'leaf_size': 13, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 13, 'n_neighbors': 6, 'p': 2}\n",
            "0.930842 (0.012604) with: {'leaf_size': 13, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 13, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.012396) with: {'leaf_size': 13, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 13, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014227) with: {'leaf_size': 13, 'n_neighbors': 9, 'p': 1}\n",
            "0.905238 (0.015042) with: {'leaf_size': 13, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.010548) with: {'leaf_size': 13, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 13, 'n_neighbors': 10, 'p': 2}\n",
            "0.904758 (0.014213) with: {'leaf_size': 13, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 13, 'n_neighbors': 11, 'p': 2}\n",
            "0.903084 (0.011310) with: {'leaf_size': 13, 'n_neighbors': 12, 'p': 1}\n",
            "0.889206 (0.013924) with: {'leaf_size': 13, 'n_neighbors': 12, 'p': 2}\n",
            "0.895428 (0.013630) with: {'leaf_size': 13, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 13, 'n_neighbors': 13, 'p': 2}\n",
            "0.890640 (0.014972) with: {'leaf_size': 13, 'n_neighbors': 14, 'p': 1}\n",
            "0.873890 (0.015198) with: {'leaf_size': 13, 'n_neighbors': 14, 'p': 2}\n",
            "0.885617 (0.013787) with: {'leaf_size': 13, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 13, 'n_neighbors': 15, 'p': 2}\n",
            "0.877242 (0.012671) with: {'leaf_size': 13, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 13, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016123) with: {'leaf_size': 13, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 13, 'n_neighbors': 17, 'p': 2}\n",
            "0.868629 (0.013716) with: {'leaf_size': 13, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 13, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017130) with: {'leaf_size': 13, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 13, 'n_neighbors': 19, 'p': 2}\n",
            "0.863123 (0.016725) with: {'leaf_size': 13, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 13, 'n_neighbors': 20, 'p': 2}\n",
            "0.858098 (0.016830) with: {'leaf_size': 13, 'n_neighbors': 21, 'p': 1}\n",
            "0.836085 (0.018257) with: {'leaf_size': 13, 'n_neighbors': 21, 'p': 2}\n",
            "0.855705 (0.016417) with: {'leaf_size': 13, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 13, 'n_neighbors': 22, 'p': 2}\n",
            "0.851638 (0.012808) with: {'leaf_size': 13, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 13, 'n_neighbors': 23, 'p': 2}\n",
            "0.849726 (0.012639) with: {'leaf_size': 13, 'n_neighbors': 24, 'p': 1}\n",
            "0.826995 (0.014756) with: {'leaf_size': 13, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010689) with: {'leaf_size': 13, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 13, 'n_neighbors': 25, 'p': 2}\n",
            "0.844461 (0.011257) with: {'leaf_size': 13, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 13, 'n_neighbors': 26, 'p': 2}\n",
            "0.845418 (0.011065) with: {'leaf_size': 13, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014815) with: {'leaf_size': 13, 'n_neighbors': 27, 'p': 2}\n",
            "0.839436 (0.013707) with: {'leaf_size': 13, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 13, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012581) with: {'leaf_size': 13, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 13, 'n_neighbors': 29, 'p': 2}\n",
            "0.982290 (0.006707) with: {'leaf_size': 14, 'n_neighbors': 1, 'p': 1}\n",
            "0.984924 (0.004795) with: {'leaf_size': 14, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.007647) with: {'leaf_size': 14, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 14, 'n_neighbors': 2, 'p': 2}\n",
            "0.968414 (0.010362) with: {'leaf_size': 14, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 14, 'n_neighbors': 3, 'p': 2}\n",
            "0.962669 (0.011391) with: {'leaf_size': 14, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 14, 'n_neighbors': 4, 'p': 2}\n",
            "0.946637 (0.013110) with: {'leaf_size': 14, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 14, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010863) with: {'leaf_size': 14, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 14, 'n_neighbors': 6, 'p': 2}\n",
            "0.930842 (0.012604) with: {'leaf_size': 14, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 14, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.012396) with: {'leaf_size': 14, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 14, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014227) with: {'leaf_size': 14, 'n_neighbors': 9, 'p': 1}\n",
            "0.905238 (0.015042) with: {'leaf_size': 14, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.010548) with: {'leaf_size': 14, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 14, 'n_neighbors': 10, 'p': 2}\n",
            "0.904758 (0.014213) with: {'leaf_size': 14, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 14, 'n_neighbors': 11, 'p': 2}\n",
            "0.903084 (0.011310) with: {'leaf_size': 14, 'n_neighbors': 12, 'p': 1}\n",
            "0.889206 (0.013924) with: {'leaf_size': 14, 'n_neighbors': 12, 'p': 2}\n",
            "0.895428 (0.013630) with: {'leaf_size': 14, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 14, 'n_neighbors': 13, 'p': 2}\n",
            "0.890640 (0.014972) with: {'leaf_size': 14, 'n_neighbors': 14, 'p': 1}\n",
            "0.873890 (0.015198) with: {'leaf_size': 14, 'n_neighbors': 14, 'p': 2}\n",
            "0.885617 (0.013787) with: {'leaf_size': 14, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 14, 'n_neighbors': 15, 'p': 2}\n",
            "0.877242 (0.012671) with: {'leaf_size': 14, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 14, 'n_neighbors': 16, 'p': 2}\n",
            "0.871500 (0.016123) with: {'leaf_size': 14, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 14, 'n_neighbors': 17, 'p': 2}\n",
            "0.868629 (0.013716) with: {'leaf_size': 14, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 14, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017130) with: {'leaf_size': 14, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 14, 'n_neighbors': 19, 'p': 2}\n",
            "0.863123 (0.016725) with: {'leaf_size': 14, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 14, 'n_neighbors': 20, 'p': 2}\n",
            "0.858098 (0.016830) with: {'leaf_size': 14, 'n_neighbors': 21, 'p': 1}\n",
            "0.836085 (0.018257) with: {'leaf_size': 14, 'n_neighbors': 21, 'p': 2}\n",
            "0.855705 (0.016417) with: {'leaf_size': 14, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 14, 'n_neighbors': 22, 'p': 2}\n",
            "0.851638 (0.012808) with: {'leaf_size': 14, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 14, 'n_neighbors': 23, 'p': 2}\n",
            "0.849726 (0.012639) with: {'leaf_size': 14, 'n_neighbors': 24, 'p': 1}\n",
            "0.826995 (0.014756) with: {'leaf_size': 14, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010689) with: {'leaf_size': 14, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 14, 'n_neighbors': 25, 'p': 2}\n",
            "0.844461 (0.011257) with: {'leaf_size': 14, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 14, 'n_neighbors': 26, 'p': 2}\n",
            "0.845418 (0.011065) with: {'leaf_size': 14, 'n_neighbors': 27, 'p': 1}\n",
            "0.816947 (0.014815) with: {'leaf_size': 14, 'n_neighbors': 27, 'p': 2}\n",
            "0.839436 (0.013707) with: {'leaf_size': 14, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 14, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012581) with: {'leaf_size': 14, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 14, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 15, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 15, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 15, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 15, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 15, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 15, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 15, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 15, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 15, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 15, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 15, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 15, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 15, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 15, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 15, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 15, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 15, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 15, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 15, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 15, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 15, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 15, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 15, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 15, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 15, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 15, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 15, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 15, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 15, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 15, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 15, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 15, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 15, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 15, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 15, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 15, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 15, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 15, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 15, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 15, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 15, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 15, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 15, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 15, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 15, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 15, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 15, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 15, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 15, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 15, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 15, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 15, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 15, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 15, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 15, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 15, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 15, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 15, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 16, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 16, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 16, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 16, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 16, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 16, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 16, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 16, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 16, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 16, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 16, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 16, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 16, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 16, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 16, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 16, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 16, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 16, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 16, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 16, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 16, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 16, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 16, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 16, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 16, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 16, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 16, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 16, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 16, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 16, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 16, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 16, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 16, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 16, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 16, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 16, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 16, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 16, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 16, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 16, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 16, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 16, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 16, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 16, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 16, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 16, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 16, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 16, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 16, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 16, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 16, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 16, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 16, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 16, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 16, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 16, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 16, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 16, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 17, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 17, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 17, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 17, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 17, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 17, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 17, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 17, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 17, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 17, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 17, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 17, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 17, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 17, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 17, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 17, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 17, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 17, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 17, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 17, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 17, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 17, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 17, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 17, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 17, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 17, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 17, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 17, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 17, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 17, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 17, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 17, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 17, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 17, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 17, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 17, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 17, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 17, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 17, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 17, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 17, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 17, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 17, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 17, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 17, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 17, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 17, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 17, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 17, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 17, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 17, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 17, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 17, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 17, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 17, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 17, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 17, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 17, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 18, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 18, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 18, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 18, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 18, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 18, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 18, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 18, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 18, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 18, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 18, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 18, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 18, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 18, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 18, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 18, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 18, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 18, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 18, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 18, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 18, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 18, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 18, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 18, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 18, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 18, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 18, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 18, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 18, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 18, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 18, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 18, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 18, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 18, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 18, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 18, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 18, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 18, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 18, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 18, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 18, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 18, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 18, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 18, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 18, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 18, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 18, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 18, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 18, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 18, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 18, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 18, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 18, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 18, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 18, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 18, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 18, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 18, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 19, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 19, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 19, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 19, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 19, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 19, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 19, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 19, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 19, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 19, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 19, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 19, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 19, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 19, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 19, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 19, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 19, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 19, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 19, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 19, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 19, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 19, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 19, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 19, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 19, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 19, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 19, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 19, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 19, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 19, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 19, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 19, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 19, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 19, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 19, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 19, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 19, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 19, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 19, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 19, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 19, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 19, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 19, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 19, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 19, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 19, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 19, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 19, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 19, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 19, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 19, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 19, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 19, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 19, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 19, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 19, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 19, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 19, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 20, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 20, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 20, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 20, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 20, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 20, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 20, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 20, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 20, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 20, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 20, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 20, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 20, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 20, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 20, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 20, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 20, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 20, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 20, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 20, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 20, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 20, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 20, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 20, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 20, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 20, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 20, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 20, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 20, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 20, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 20, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 20, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 20, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 20, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 20, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 20, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 20, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 20, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 20, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 20, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 20, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 20, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 20, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 20, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 20, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 20, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 20, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 20, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 20, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 20, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 20, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 20, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 20, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 20, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 20, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 20, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 20, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 20, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 21, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 21, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 21, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 21, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 21, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 21, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 21, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 21, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 21, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 21, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 21, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 21, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 21, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 21, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 21, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 21, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 21, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 21, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 21, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 21, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 21, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 21, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 21, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 21, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 21, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 21, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 21, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 21, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 21, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 21, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 21, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 21, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 21, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 21, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 21, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 21, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 21, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 21, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 21, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 21, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 21, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 21, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 21, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 21, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 21, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 21, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 21, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 21, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 21, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 21, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 21, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 21, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 21, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 21, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 21, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 21, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 21, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 21, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 22, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 22, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 22, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 22, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 22, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 22, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 22, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 22, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 22, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 22, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 22, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 22, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 22, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 22, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 22, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 22, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 22, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 22, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 22, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 22, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 22, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 22, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 22, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 22, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 22, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 22, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 22, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 22, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 22, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 22, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 22, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 22, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 22, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 22, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 22, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 22, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 22, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 22, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 22, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 22, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 22, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 22, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 22, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 22, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 22, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 22, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 22, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 22, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 22, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 22, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 22, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 22, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 22, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 22, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 22, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 22, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 22, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 22, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 23, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 23, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 23, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 23, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 23, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 23, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 23, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 23, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 23, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 23, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 23, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 23, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 23, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 23, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 23, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 23, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 23, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 23, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 23, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 23, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 23, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 23, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 23, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 23, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 23, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 23, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 23, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 23, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 23, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 23, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 23, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 23, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 23, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 23, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 23, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 23, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 23, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 23, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 23, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 23, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 23, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 23, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 23, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 23, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 23, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 23, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 23, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 23, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 23, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 23, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 23, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 23, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 23, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 23, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 23, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 23, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 23, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 23, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 24, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 24, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 24, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 24, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 24, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 24, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 24, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 24, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 24, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 24, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 24, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 24, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 24, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 24, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 24, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 24, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 24, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 24, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 24, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 24, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 24, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 24, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 24, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 24, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 24, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 24, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 24, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 24, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 24, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 24, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 24, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 24, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 24, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 24, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 24, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 24, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 24, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 24, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 24, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 24, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 24, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 24, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 24, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 24, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 24, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 24, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 24, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 24, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 24, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 24, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 24, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 24, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 24, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 24, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 24, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 24, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 24, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 24, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 25, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 25, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 25, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 25, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 25, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 25, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 25, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 25, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 25, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 25, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 25, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 25, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 25, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 25, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 25, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 25, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 25, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 25, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 25, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 25, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 25, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 25, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 25, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 25, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 25, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 25, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 25, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 25, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 25, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 25, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 25, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 25, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 25, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 25, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 25, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 25, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 25, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 25, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 25, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 25, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 25, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 25, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 25, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 25, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 25, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 25, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 25, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 25, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 25, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 25, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 25, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 25, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 25, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 25, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 25, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 25, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 25, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 25, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 26, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 26, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 26, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 26, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 26, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 26, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 26, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 26, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 26, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 26, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 26, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 26, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 26, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 26, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 26, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 26, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 26, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 26, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 26, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 26, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 26, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 26, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 26, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 26, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 26, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 26, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 26, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 26, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 26, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 26, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 26, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 26, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 26, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 26, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 26, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 26, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 26, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 26, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 26, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 26, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 26, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 26, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 26, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 26, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 26, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 26, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 26, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 26, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 26, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 26, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 26, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 26, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 26, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 26, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 26, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 26, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 26, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 26, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 27, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 27, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 27, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 27, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 27, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 27, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 27, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 27, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 27, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 27, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 27, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 27, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 27, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 27, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 27, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 27, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 27, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 27, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 27, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 27, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 27, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 27, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 27, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 27, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 27, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 27, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 27, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 27, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 27, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 27, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 27, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 27, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 27, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 27, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 27, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 27, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 27, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 27, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 27, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 27, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 27, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 27, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 27, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 27, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 27, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 27, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 27, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 27, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 27, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 27, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 27, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 27, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 27, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 27, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 27, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 27, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 27, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 27, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 28, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 28, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 28, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 28, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 28, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 28, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 28, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 28, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 28, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 28, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 28, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 28, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 28, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 28, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 28, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 28, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 28, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 28, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 28, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 28, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 28, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 28, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 28, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 28, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 28, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 28, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 28, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 28, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 28, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 28, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 28, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 28, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 28, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 28, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 28, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 28, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 28, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 28, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 28, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 28, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 28, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 28, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 28, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 28, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 28, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 28, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 28, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 28, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 28, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 28, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 28, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 28, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 28, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 28, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 28, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 28, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 28, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 28, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.006521) with: {'leaf_size': 29, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 29, 'n_neighbors': 1, 'p': 2}\n",
            "0.980138 (0.008294) with: {'leaf_size': 29, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 29, 'n_neighbors': 2, 'p': 2}\n",
            "0.967935 (0.011077) with: {'leaf_size': 29, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 29, 'n_neighbors': 3, 'p': 2}\n",
            "0.963147 (0.010930) with: {'leaf_size': 29, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 29, 'n_neighbors': 4, 'p': 2}\n",
            "0.946397 (0.013799) with: {'leaf_size': 29, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 29, 'n_neighbors': 5, 'p': 2}\n",
            "0.943765 (0.011444) with: {'leaf_size': 29, 'n_neighbors': 6, 'p': 1}\n",
            "0.937543 (0.013518) with: {'leaf_size': 29, 'n_neighbors': 6, 'p': 2}\n",
            "0.931321 (0.012449) with: {'leaf_size': 29, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 29, 'n_neighbors': 7, 'p': 2}\n",
            "0.926776 (0.011435) with: {'leaf_size': 29, 'n_neighbors': 8, 'p': 1}\n",
            "0.916487 (0.011988) with: {'leaf_size': 29, 'n_neighbors': 8, 'p': 2}\n",
            "0.916006 (0.014150) with: {'leaf_size': 29, 'n_neighbors': 9, 'p': 1}\n",
            "0.904280 (0.015639) with: {'leaf_size': 29, 'n_neighbors': 9, 'p': 2}\n",
            "0.915529 (0.011232) with: {'leaf_size': 29, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 29, 'n_neighbors': 10, 'p': 2}\n",
            "0.904759 (0.014448) with: {'leaf_size': 29, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 29, 'n_neighbors': 11, 'p': 2}\n",
            "0.901887 (0.013171) with: {'leaf_size': 29, 'n_neighbors': 12, 'p': 1}\n",
            "0.889685 (0.014059) with: {'leaf_size': 29, 'n_neighbors': 12, 'p': 2}\n",
            "0.895667 (0.014536) with: {'leaf_size': 29, 'n_neighbors': 13, 'p': 1}\n",
            "0.878679 (0.016439) with: {'leaf_size': 29, 'n_neighbors': 13, 'p': 2}\n",
            "0.890880 (0.013862) with: {'leaf_size': 29, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 29, 'n_neighbors': 14, 'p': 2}\n",
            "0.885857 (0.013069) with: {'leaf_size': 29, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 29, 'n_neighbors': 15, 'p': 2}\n",
            "0.877721 (0.011699) with: {'leaf_size': 29, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 29, 'n_neighbors': 16, 'p': 2}\n",
            "0.871740 (0.015946) with: {'leaf_size': 29, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 29, 'n_neighbors': 17, 'p': 2}\n",
            "0.869108 (0.012970) with: {'leaf_size': 29, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 29, 'n_neighbors': 18, 'p': 2}\n",
            "0.865516 (0.017559) with: {'leaf_size': 29, 'n_neighbors': 19, 'p': 1}\n",
            "0.843981 (0.017513) with: {'leaf_size': 29, 'n_neighbors': 19, 'p': 2}\n",
            "0.862167 (0.017631) with: {'leaf_size': 29, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.019129) with: {'leaf_size': 29, 'n_neighbors': 20, 'p': 2}\n",
            "0.858337 (0.015629) with: {'leaf_size': 29, 'n_neighbors': 21, 'p': 1}\n",
            "0.836325 (0.018066) with: {'leaf_size': 29, 'n_neighbors': 21, 'p': 2}\n",
            "0.856902 (0.016271) with: {'leaf_size': 29, 'n_neighbors': 22, 'p': 1}\n",
            "0.836325 (0.015054) with: {'leaf_size': 29, 'n_neighbors': 22, 'p': 2}\n",
            "0.852356 (0.013245) with: {'leaf_size': 29, 'n_neighbors': 23, 'p': 1}\n",
            "0.830345 (0.014999) with: {'leaf_size': 29, 'n_neighbors': 23, 'p': 2}\n",
            "0.849486 (0.013161) with: {'leaf_size': 29, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015130) with: {'leaf_size': 29, 'n_neighbors': 24, 'p': 2}\n",
            "0.847572 (0.010527) with: {'leaf_size': 29, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 29, 'n_neighbors': 25, 'p': 2}\n",
            "0.844223 (0.012431) with: {'leaf_size': 29, 'n_neighbors': 26, 'p': 1}\n",
            "0.820057 (0.011246) with: {'leaf_size': 29, 'n_neighbors': 26, 'p': 2}\n",
            "0.844700 (0.010055) with: {'leaf_size': 29, 'n_neighbors': 27, 'p': 1}\n",
            "0.816708 (0.014578) with: {'leaf_size': 29, 'n_neighbors': 27, 'p': 2}\n",
            "0.840394 (0.013853) with: {'leaf_size': 29, 'n_neighbors': 28, 'p': 1}\n",
            "0.816707 (0.015460) with: {'leaf_size': 29, 'n_neighbors': 28, 'p': 2}\n",
            "0.838239 (0.012063) with: {'leaf_size': 29, 'n_neighbors': 29, 'p': 1}\n",
            "0.811442 (0.018735) with: {'leaf_size': 29, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 30, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 30, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 30, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 30, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 30, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 30, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 30, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 30, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 30, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 30, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 30, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 30, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 30, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 30, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 30, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 30, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 30, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 30, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 30, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 30, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 30, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 30, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 30, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 30, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 30, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 30, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 30, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 30, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 30, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 30, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 30, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 30, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 30, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 30, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 30, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 30, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 30, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 30, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 30, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 30, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 30, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 30, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 30, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 30, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 30, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 30, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 30, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 30, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 30, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 30, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 30, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 30, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 30, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 30, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 30, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 30, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 30, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 30, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 31, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 31, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 31, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 31, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 31, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 31, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 31, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 31, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 31, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 31, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 31, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 31, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 31, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 31, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 31, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 31, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 31, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 31, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 31, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 31, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 31, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 31, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 31, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 31, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 31, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 31, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 31, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 31, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 31, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 31, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 31, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 31, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 31, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 31, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 31, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 31, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 31, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 31, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 31, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 31, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 31, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 31, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 31, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 31, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 31, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 31, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 31, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 31, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 31, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 31, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 31, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 31, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 31, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 31, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 31, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 31, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 31, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 31, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 32, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 32, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 32, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 32, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 32, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 32, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 32, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 32, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 32, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 32, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 32, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 32, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 32, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 32, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 32, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 32, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 32, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 32, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 32, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 32, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 32, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 32, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 32, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 32, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 32, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 32, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 32, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 32, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 32, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 32, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 32, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 32, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 32, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 32, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 32, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 32, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 32, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 32, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 32, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 32, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 32, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 32, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 32, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 32, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 32, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 32, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 32, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 32, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 32, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 32, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 32, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 32, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 32, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 32, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 32, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 32, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 32, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 32, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 33, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 33, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 33, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 33, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 33, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 33, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 33, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 33, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 33, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 33, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 33, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 33, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 33, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 33, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 33, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 33, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 33, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 33, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 33, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 33, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 33, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 33, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 33, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 33, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 33, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 33, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 33, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 33, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 33, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 33, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 33, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 33, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 33, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 33, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 33, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 33, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 33, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 33, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 33, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 33, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 33, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 33, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 33, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 33, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 33, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 33, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 33, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 33, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 33, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 33, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 33, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 33, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 33, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 33, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 33, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 33, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 33, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 33, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 34, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 34, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 34, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 34, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 34, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 34, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 34, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 34, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 34, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 34, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 34, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 34, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 34, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 34, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 34, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 34, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 34, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 34, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 34, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 34, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 34, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 34, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 34, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 34, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 34, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 34, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 34, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 34, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 34, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 34, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 34, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 34, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 34, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 34, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 34, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 34, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 34, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 34, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 34, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 34, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 34, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 34, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 34, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 34, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 34, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 34, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 34, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 34, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 34, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 34, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 34, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 34, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 34, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 34, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 34, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 34, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 34, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 34, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 35, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 35, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 35, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 35, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 35, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 35, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 35, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 35, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 35, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 35, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 35, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 35, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 35, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 35, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 35, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 35, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 35, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 35, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 35, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 35, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 35, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 35, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 35, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 35, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 35, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 35, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 35, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 35, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 35, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 35, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 35, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 35, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 35, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 35, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 35, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 35, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 35, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 35, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 35, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 35, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 35, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 35, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 35, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 35, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 35, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 35, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 35, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 35, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 35, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 35, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 35, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 35, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 35, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 35, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 35, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 35, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 35, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 35, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 36, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 36, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 36, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 36, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 36, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 36, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 36, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 36, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 36, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 36, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 36, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 36, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 36, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 36, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 36, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 36, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 36, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 36, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 36, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 36, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 36, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 36, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 36, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 36, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 36, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 36, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 36, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 36, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 36, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 36, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 36, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 36, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 36, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 36, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 36, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 36, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 36, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 36, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 36, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 36, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 36, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 36, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 36, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 36, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 36, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 36, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 36, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 36, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 36, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 36, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 36, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 36, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 36, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 36, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 36, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 36, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 36, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 36, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 37, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 37, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 37, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 37, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 37, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 37, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 37, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 37, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 37, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 37, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 37, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 37, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 37, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 37, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 37, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 37, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 37, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 37, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 37, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 37, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 37, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 37, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 37, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 37, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 37, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 37, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 37, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 37, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 37, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 37, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 37, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 37, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 37, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 37, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 37, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 37, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 37, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 37, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 37, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 37, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 37, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 37, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 37, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 37, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 37, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 37, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 37, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 37, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 37, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 37, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 37, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 37, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 37, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 37, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 37, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 37, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 37, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 37, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 38, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 38, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 38, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 38, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 38, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 38, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 38, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 38, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 38, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 38, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 38, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 38, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 38, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 38, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 38, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 38, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 38, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 38, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 38, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 38, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 38, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 38, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 38, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 38, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 38, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 38, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 38, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 38, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 38, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 38, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 38, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 38, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 38, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 38, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 38, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 38, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 38, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 38, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 38, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 38, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 38, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 38, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 38, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 38, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 38, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 38, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 38, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 38, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 38, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 38, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 38, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 38, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 38, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 38, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 38, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 38, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 38, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 38, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 39, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 39, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 39, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 39, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 39, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 39, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 39, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 39, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 39, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 39, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 39, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 39, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 39, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 39, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 39, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 39, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 39, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 39, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 39, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 39, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 39, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 39, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 39, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 39, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 39, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 39, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 39, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 39, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 39, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 39, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 39, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 39, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 39, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 39, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 39, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 39, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 39, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 39, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 39, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 39, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 39, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 39, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 39, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 39, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 39, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 39, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 39, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 39, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 39, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 39, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 39, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 39, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 39, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 39, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 39, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 39, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 39, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 39, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 40, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 40, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 40, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 40, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 40, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 40, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 40, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 40, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 40, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 40, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 40, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 40, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 40, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 40, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 40, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 40, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 40, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 40, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 40, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 40, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 40, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 40, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 40, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 40, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 40, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 40, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 40, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 40, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 40, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 40, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 40, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 40, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 40, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 40, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 40, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 40, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 40, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 40, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 40, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 40, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 40, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 40, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 40, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 40, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 40, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 40, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 40, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 40, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 40, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 40, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 40, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 40, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 40, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 40, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 40, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 40, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 40, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 40, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 41, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 41, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 41, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 41, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 41, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 41, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 41, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 41, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 41, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 41, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 41, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 41, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 41, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 41, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 41, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 41, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 41, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 41, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 41, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 41, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 41, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 41, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 41, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 41, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 41, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 41, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 41, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 41, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 41, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 41, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 41, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 41, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 41, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 41, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 41, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 41, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 41, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 41, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 41, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 41, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 41, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 41, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 41, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 41, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 41, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 41, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 41, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 41, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 41, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 41, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 41, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 41, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 41, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 41, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 41, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 41, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 41, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 41, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 42, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 42, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 42, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 42, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 42, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 42, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 42, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 42, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 42, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 42, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 42, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 42, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 42, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 42, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 42, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 42, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 42, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 42, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 42, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 42, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 42, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 42, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 42, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 42, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 42, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 42, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 42, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 42, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 42, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 42, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 42, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 42, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 42, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 42, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 42, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 42, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 42, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 42, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 42, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 42, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 42, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 42, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 42, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 42, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 42, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 42, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 42, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 42, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 42, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 42, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 42, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 42, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 42, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 42, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 42, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 42, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 42, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 42, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 43, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 43, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 43, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 43, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 43, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 43, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 43, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 43, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 43, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 43, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 43, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 43, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 43, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 43, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 43, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 43, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 43, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 43, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 43, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 43, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 43, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 43, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 43, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 43, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 43, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 43, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 43, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 43, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 43, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 43, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 43, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 43, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 43, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 43, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 43, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 43, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 43, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 43, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 43, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 43, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 43, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 43, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 43, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 43, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 43, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 43, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 43, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 43, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 43, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 43, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 43, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 43, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 43, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 43, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 43, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 43, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 43, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 43, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 44, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 44, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 44, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 44, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 44, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 44, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 44, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 44, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 44, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 44, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 44, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 44, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 44, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 44, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 44, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 44, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 44, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 44, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 44, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 44, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 44, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 44, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 44, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 44, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 44, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 44, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 44, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 44, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 44, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 44, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 44, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 44, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 44, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 44, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 44, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 44, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 44, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 44, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 44, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 44, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 44, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 44, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 44, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 44, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 44, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 44, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 44, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 44, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 44, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 44, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 44, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 44, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 44, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 44, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 44, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 44, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 44, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 44, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 45, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 45, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 45, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 45, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 45, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 45, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 45, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 45, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 45, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 45, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 45, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 45, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 45, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 45, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 45, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 45, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 45, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 45, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 45, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 45, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 45, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 45, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 45, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 45, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 45, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 45, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 45, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 45, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 45, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 45, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 45, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 45, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 45, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 45, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 45, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 45, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 45, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 45, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 45, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 45, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 45, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 45, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 45, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 45, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 45, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 45, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 45, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 45, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 45, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 45, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 45, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 45, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 45, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 45, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 45, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 45, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 45, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 45, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 46, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 46, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 46, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 46, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 46, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 46, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 46, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 46, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 46, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 46, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 46, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 46, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 46, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 46, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 46, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 46, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 46, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 46, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 46, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 46, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 46, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 46, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 46, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 46, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 46, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 46, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 46, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 46, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 46, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 46, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 46, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 46, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 46, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 46, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 46, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 46, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 46, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 46, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 46, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 46, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 46, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 46, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 46, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 46, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 46, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 46, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 46, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 46, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 46, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 46, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 46, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 46, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 46, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 46, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 46, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 46, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 46, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 46, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 47, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 47, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 47, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 47, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 47, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 47, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 47, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 47, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 47, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 47, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 47, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 47, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 47, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 47, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 47, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 47, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 47, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 47, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 47, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 47, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 47, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 47, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 47, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 47, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 47, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 47, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 47, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 47, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 47, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 47, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 47, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 47, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 47, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 47, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 47, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 47, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 47, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 47, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 47, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 47, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 47, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 47, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 47, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 47, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 47, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 47, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 47, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 47, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 47, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 47, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 47, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 47, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 47, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 47, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 47, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 47, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 47, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 47, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 48, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 48, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 48, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 48, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 48, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 48, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 48, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 48, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 48, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 48, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 48, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 48, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 48, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 48, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 48, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 48, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 48, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 48, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 48, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 48, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 48, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 48, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 48, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 48, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 48, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 48, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 48, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 48, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 48, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 48, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 48, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 48, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 48, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 48, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 48, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 48, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 48, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 48, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 48, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 48, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 48, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 48, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 48, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 48, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 48, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 48, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 48, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 48, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 48, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 48, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 48, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 48, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 48, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 48, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 48, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 48, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 48, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 48, 'n_neighbors': 29, 'p': 2}\n",
            "0.982530 (0.007028) with: {'leaf_size': 49, 'n_neighbors': 1, 'p': 1}\n",
            "0.984684 (0.004812) with: {'leaf_size': 49, 'n_neighbors': 1, 'p': 2}\n",
            "0.979899 (0.008094) with: {'leaf_size': 49, 'n_neighbors': 2, 'p': 1}\n",
            "0.981335 (0.008206) with: {'leaf_size': 49, 'n_neighbors': 2, 'p': 2}\n",
            "0.967695 (0.011081) with: {'leaf_size': 49, 'n_neighbors': 3, 'p': 1}\n",
            "0.963869 (0.011930) with: {'leaf_size': 49, 'n_neighbors': 3, 'p': 2}\n",
            "0.963148 (0.011083) with: {'leaf_size': 49, 'n_neighbors': 4, 'p': 1}\n",
            "0.956929 (0.012101) with: {'leaf_size': 49, 'n_neighbors': 4, 'p': 2}\n",
            "0.945919 (0.014046) with: {'leaf_size': 49, 'n_neighbors': 5, 'p': 1}\n",
            "0.939698 (0.012562) with: {'leaf_size': 49, 'n_neighbors': 5, 'p': 2}\n",
            "0.944245 (0.010323) with: {'leaf_size': 49, 'n_neighbors': 6, 'p': 1}\n",
            "0.937783 (0.013411) with: {'leaf_size': 49, 'n_neighbors': 6, 'p': 2}\n",
            "0.930603 (0.011833) with: {'leaf_size': 49, 'n_neighbors': 7, 'p': 1}\n",
            "0.920553 (0.015022) with: {'leaf_size': 49, 'n_neighbors': 7, 'p': 2}\n",
            "0.927015 (0.010453) with: {'leaf_size': 49, 'n_neighbors': 8, 'p': 1}\n",
            "0.916247 (0.011870) with: {'leaf_size': 49, 'n_neighbors': 8, 'p': 2}\n",
            "0.916724 (0.014202) with: {'leaf_size': 49, 'n_neighbors': 9, 'p': 1}\n",
            "0.905237 (0.015085) with: {'leaf_size': 49, 'n_neighbors': 9, 'p': 2}\n",
            "0.915290 (0.011239) with: {'leaf_size': 49, 'n_neighbors': 10, 'p': 1}\n",
            "0.902128 (0.014696) with: {'leaf_size': 49, 'n_neighbors': 10, 'p': 2}\n",
            "0.905477 (0.013984) with: {'leaf_size': 49, 'n_neighbors': 11, 'p': 1}\n",
            "0.890880 (0.015196) with: {'leaf_size': 49, 'n_neighbors': 11, 'p': 2}\n",
            "0.902606 (0.011593) with: {'leaf_size': 49, 'n_neighbors': 12, 'p': 1}\n",
            "0.888967 (0.013847) with: {'leaf_size': 49, 'n_neighbors': 12, 'p': 2}\n",
            "0.895905 (0.015235) with: {'leaf_size': 49, 'n_neighbors': 13, 'p': 1}\n",
            "0.878918 (0.016022) with: {'leaf_size': 49, 'n_neighbors': 13, 'p': 2}\n",
            "0.891120 (0.014346) with: {'leaf_size': 49, 'n_neighbors': 14, 'p': 1}\n",
            "0.873651 (0.015338) with: {'leaf_size': 49, 'n_neighbors': 14, 'p': 2}\n",
            "0.885379 (0.012319) with: {'leaf_size': 49, 'n_neighbors': 15, 'p': 1}\n",
            "0.865756 (0.017440) with: {'leaf_size': 49, 'n_neighbors': 15, 'p': 2}\n",
            "0.879156 (0.012084) with: {'leaf_size': 49, 'n_neighbors': 16, 'p': 1}\n",
            "0.860490 (0.017475) with: {'leaf_size': 49, 'n_neighbors': 16, 'p': 2}\n",
            "0.871261 (0.015682) with: {'leaf_size': 49, 'n_neighbors': 17, 'p': 1}\n",
            "0.854749 (0.018232) with: {'leaf_size': 49, 'n_neighbors': 17, 'p': 2}\n",
            "0.868869 (0.013978) with: {'leaf_size': 49, 'n_neighbors': 18, 'p': 1}\n",
            "0.848529 (0.019194) with: {'leaf_size': 49, 'n_neighbors': 18, 'p': 2}\n",
            "0.865994 (0.017825) with: {'leaf_size': 49, 'n_neighbors': 19, 'p': 1}\n",
            "0.843742 (0.017422) with: {'leaf_size': 49, 'n_neighbors': 19, 'p': 2}\n",
            "0.861449 (0.017633) with: {'leaf_size': 49, 'n_neighbors': 20, 'p': 1}\n",
            "0.843262 (0.018949) with: {'leaf_size': 49, 'n_neighbors': 20, 'p': 2}\n",
            "0.857381 (0.015817) with: {'leaf_size': 49, 'n_neighbors': 21, 'p': 1}\n",
            "0.836564 (0.017515) with: {'leaf_size': 49, 'n_neighbors': 21, 'p': 2}\n",
            "0.855945 (0.016122) with: {'leaf_size': 49, 'n_neighbors': 22, 'p': 1}\n",
            "0.836086 (0.014670) with: {'leaf_size': 49, 'n_neighbors': 22, 'p': 2}\n",
            "0.852117 (0.012883) with: {'leaf_size': 49, 'n_neighbors': 23, 'p': 1}\n",
            "0.830585 (0.015051) with: {'leaf_size': 49, 'n_neighbors': 23, 'p': 2}\n",
            "0.847811 (0.013189) with: {'leaf_size': 49, 'n_neighbors': 24, 'p': 1}\n",
            "0.827234 (0.015092) with: {'leaf_size': 49, 'n_neighbors': 24, 'p': 2}\n",
            "0.847094 (0.011680) with: {'leaf_size': 49, 'n_neighbors': 25, 'p': 1}\n",
            "0.823885 (0.012934) with: {'leaf_size': 49, 'n_neighbors': 25, 'p': 2}\n",
            "0.844940 (0.012818) with: {'leaf_size': 49, 'n_neighbors': 26, 'p': 1}\n",
            "0.819817 (0.010851) with: {'leaf_size': 49, 'n_neighbors': 26, 'p': 2}\n",
            "0.843743 (0.010029) with: {'leaf_size': 49, 'n_neighbors': 27, 'p': 1}\n",
            "0.817186 (0.014658) with: {'leaf_size': 49, 'n_neighbors': 27, 'p': 2}\n",
            "0.840872 (0.014114) with: {'leaf_size': 49, 'n_neighbors': 28, 'p': 1}\n",
            "0.816468 (0.015820) with: {'leaf_size': 49, 'n_neighbors': 28, 'p': 2}\n",
            "0.838000 (0.013439) with: {'leaf_size': 49, 'n_neighbors': 29, 'p': 1}\n",
            "0.811681 (0.018651) with: {'leaf_size': 49, 'n_neighbors': 29, 'p': 2}\n",
            "-----------------------------------------------------------\n",
            "\n",
            "Best Accuracy: 0.984924 using {'leaf_size': 1, 'n_neighbors': 1, 'p': 2}\n",
            "\n",
            "Detailed classification report:\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       1.00      0.98      0.99       420\n",
            "         1.0       0.99      1.00      0.99       625\n",
            "\n",
            "    accuracy                           0.99      1045\n",
            "   macro avg       0.99      0.99      0.99      1045\n",
            "weighted avg       0.99      0.99      0.99      1045\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sS-YbEiAq5ii"
      },
      "source": [
        "After hyperparameter tuning, we’ve found that the best parameters for the K Nearest Neighbors classifier’s 'leaf_size', 'n_neighbors', and ‘p’ are '1', ‘1’, and ‘2’ with the accuracy of 0.984. All other metrics are almost perfect.\n",
        "\n",
        "(Note: \"Best hyperparamters\" might change at each runtime, due to cross-validation.)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YBx5EKr8yO4F"
      },
      "source": [
        "### 6.2 **Logistic Regression**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0srqWsgi5FvC",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "67268b31-1d65-4c54-89b6-f16d5f32995d"
      },
      "source": [
        "best_lr = hyperTuning(LogisticRegression(), \n",
        "            dict(C = np.logspace(-4, 4, 20), max_iter = [1000, 2500, 5000]),\n",
        "            X_train, y_train, X_test, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.830590 (0.017765) with: {'C': 0.0001, 'max_iter': 1000}\n",
            "0.830590 (0.017765) with: {'C': 0.0001, 'max_iter': 2500}\n",
            "0.830590 (0.017765) with: {'C': 0.0001, 'max_iter': 5000}\n",
            "0.829873 (0.017457) with: {'C': 0.00026366508987303583, 'max_iter': 1000}\n",
            "0.829873 (0.017457) with: {'C': 0.00026366508987303583, 'max_iter': 2500}\n",
            "0.829873 (0.017457) with: {'C': 0.00026366508987303583, 'max_iter': 5000}\n",
            "0.833941 (0.017355) with: {'C': 0.0006951927961775605, 'max_iter': 1000}\n",
            "0.833941 (0.017355) with: {'C': 0.0006951927961775605, 'max_iter': 2500}\n",
            "0.833941 (0.017355) with: {'C': 0.0006951927961775605, 'max_iter': 5000}\n",
            "0.844946 (0.017225) with: {'C': 0.0018329807108324356, 'max_iter': 1000}\n",
            "0.844946 (0.017225) with: {'C': 0.0018329807108324356, 'max_iter': 2500}\n",
            "0.844946 (0.017225) with: {'C': 0.0018329807108324356, 'max_iter': 5000}\n",
            "0.857392 (0.018360) with: {'C': 0.004832930238571752, 'max_iter': 1000}\n",
            "0.857392 (0.018360) with: {'C': 0.004832930238571752, 'max_iter': 2500}\n",
            "0.857392 (0.018360) with: {'C': 0.004832930238571752, 'max_iter': 5000}\n",
            "0.870792 (0.018047) with: {'C': 0.012742749857031334, 'max_iter': 1000}\n",
            "0.870792 (0.018299) with: {'C': 0.012742749857031334, 'max_iter': 2500}\n",
            "0.870792 (0.018299) with: {'C': 0.012742749857031334, 'max_iter': 5000}\n",
            "0.875817 (0.019348) with: {'C': 0.03359818286283781, 'max_iter': 1000}\n",
            "0.877731 (0.018627) with: {'C': 0.03359818286283781, 'max_iter': 2500}\n",
            "0.876295 (0.019368) with: {'C': 0.03359818286283781, 'max_iter': 5000}\n",
            "0.879887 (0.019120) with: {'C': 0.08858667904100823, 'max_iter': 1000}\n",
            "0.880843 (0.019507) with: {'C': 0.08858667904100823, 'max_iter': 2500}\n",
            "0.879886 (0.019769) with: {'C': 0.08858667904100823, 'max_iter': 5000}\n",
            "0.881799 (0.019107) with: {'C': 0.23357214690901212, 'max_iter': 1000}\n",
            "0.883234 (0.016141) with: {'C': 0.23357214690901212, 'max_iter': 2500}\n",
            "0.882278 (0.018484) with: {'C': 0.23357214690901212, 'max_iter': 5000}\n",
            "0.879645 (0.019251) with: {'C': 0.615848211066026, 'max_iter': 1000}\n",
            "0.883234 (0.015600) with: {'C': 0.615848211066026, 'max_iter': 2500}\n",
            "0.883713 (0.015725) with: {'C': 0.615848211066026, 'max_iter': 5000}\n",
            "0.879165 (0.018284) with: {'C': 1.623776739188721, 'max_iter': 1000}\n",
            "0.883236 (0.018607) with: {'C': 1.623776739188721, 'max_iter': 2500}\n",
            "0.882518 (0.017355) with: {'C': 1.623776739188721, 'max_iter': 5000}\n",
            "0.879409 (0.022075) with: {'C': 4.281332398719396, 'max_iter': 1000}\n",
            "0.885390 (0.018963) with: {'C': 4.281332398719396, 'max_iter': 2500}\n",
            "0.884434 (0.020093) with: {'C': 4.281332398719396, 'max_iter': 5000}\n",
            "0.879167 (0.019610) with: {'C': 11.288378916846883, 'max_iter': 1000}\n",
            "0.884911 (0.020077) with: {'C': 11.288378916846883, 'max_iter': 2500}\n",
            "0.884911 (0.019381) with: {'C': 11.288378916846883, 'max_iter': 5000}\n",
            "0.880845 (0.022117) with: {'C': 29.763514416313132, 'max_iter': 1000}\n",
            "0.883713 (0.017448) with: {'C': 29.763514416313132, 'max_iter': 2500}\n",
            "0.882517 (0.018067) with: {'C': 29.763514416313132, 'max_iter': 5000}\n",
            "0.882519 (0.020442) with: {'C': 78.47599703514607, 'max_iter': 1000}\n",
            "0.884193 (0.018404) with: {'C': 78.47599703514607, 'max_iter': 2500}\n",
            "0.883953 (0.018746) with: {'C': 78.47599703514607, 'max_iter': 5000}\n",
            "0.880363 (0.019189) with: {'C': 206.913808111479, 'max_iter': 1000}\n",
            "0.884194 (0.019868) with: {'C': 206.913808111479, 'max_iter': 2500}\n",
            "0.884194 (0.020575) with: {'C': 206.913808111479, 'max_iter': 5000}\n",
            "0.879886 (0.020311) with: {'C': 545.5594781168514, 'max_iter': 1000}\n",
            "0.881083 (0.019455) with: {'C': 545.5594781168514, 'max_iter': 2500}\n",
            "0.882279 (0.020171) with: {'C': 545.5594781168514, 'max_iter': 5000}\n",
            "0.880604 (0.021850) with: {'C': 1438.44988828766, 'max_iter': 1000}\n",
            "0.884912 (0.020162) with: {'C': 1438.44988828766, 'max_iter': 2500}\n",
            "0.884194 (0.020210) with: {'C': 1438.44988828766, 'max_iter': 5000}\n",
            "0.879644 (0.016605) with: {'C': 3792.690190732246, 'max_iter': 1000}\n",
            "0.883476 (0.021918) with: {'C': 3792.690190732246, 'max_iter': 2500}\n",
            "0.883475 (0.019691) with: {'C': 3792.690190732246, 'max_iter': 5000}\n",
            "0.879409 (0.021202) with: {'C': 10000.0, 'max_iter': 1000}\n",
            "0.882758 (0.019601) with: {'C': 10000.0, 'max_iter': 2500}\n",
            "0.881561 (0.020041) with: {'C': 10000.0, 'max_iter': 5000}\n",
            "-----------------------------------------------------------\n",
            "\n",
            "Best Accuracy: 0.885390 using {'C': 4.281332398719396, 'max_iter': 2500}\n",
            "\n",
            "Detailed classification report:\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       0.88      0.87      0.88       420\n",
            "         1.0       0.92      0.92      0.92       625\n",
            "\n",
            "    accuracy                           0.90      1045\n",
            "   macro avg       0.90      0.90      0.90      1045\n",
            "weighted avg       0.90      0.90      0.90      1045\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_logistic.py:764: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
            "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "Please also refer to the documentation for alternative solver options:\n",
            "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
            "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Snjbf7eHqWZn"
      },
      "source": [
        "After hyperparameter tuning, we’ve found that the best parameters for the Logistic Regression classifier’s 'C' and 'max_iter' are '4.281' and ‘2500’ with the accuracy of 0.885. Other metrics range from 0.87 to 0.92.\n",
        "\n",
        "(Note: \"Best hyperparamters\" might change at each runtime, due to cross-validation.)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6XyhXJG0FX3X"
      },
      "source": [
        "### 6.3 **Random Forest**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XibWH8wYPvg6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 386
        },
        "outputId": "95c83c3e-ad83-4661-99db-8aae76202460"
      },
      "source": [
        "best_rf = hyperTuning(RandomForestClassifier(), dict(n_estimators = [10, 100, 1000], max_features = ['sqrt', 'log2']),\n",
        "            X_train, y_train, X_test, y_test)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.928214 (0.013141) with: {'max_features': 'sqrt', 'n_estimators': 10}\n",
            "0.938026 (0.011234) with: {'max_features': 'sqrt', 'n_estimators': 100}\n",
            "0.939702 (0.012637) with: {'max_features': 'sqrt', 'n_estimators': 1000}\n",
            "0.928931 (0.010037) with: {'max_features': 'log2', 'n_estimators': 10}\n",
            "0.940419 (0.009701) with: {'max_features': 'log2', 'n_estimators': 100}\n",
            "0.938983 (0.012840) with: {'max_features': 'log2', 'n_estimators': 1000}\n",
            "-----------------------------------------------------------\n",
            "\n",
            "Best Accuracy: 0.940419 using {'max_features': 'log2', 'n_estimators': 100}\n",
            "\n",
            "Detailed classification report:\n",
            "\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       0.96      0.92      0.94       420\n",
            "         1.0       0.95      0.97      0.96       625\n",
            "\n",
            "    accuracy                           0.95      1045\n",
            "   macro avg       0.95      0.95      0.95      1045\n",
            "weighted avg       0.95      0.95      0.95      1045\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y881XGW5lIiV"
      },
      "source": [
        "After hyperparameter tuning, we’ve found that the best parameters for the Random Forest classifier’s 'max_features' and 'n_estimators' are 'sqrt' and ‘1000’ with the accuracy of 0.941. All other metrics are above 0.90. \n",
        "\n",
        "(Note: \"Best hyperparamters\" might change at each runtime, due to cross-validation.)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZobewJBsMK0D"
      },
      "source": [
        "After GridSearchCV tuning, here are the ROC curves with the best hyperparameters:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8ET3HJccMJx3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 576
        },
        "outputId": "78c52862-8ad1-4317-c124-b5b4ffe01426"
      },
      "source": [
        "classifiers = [LogisticRegression(C = best_lr['C'], max_iter = best_lr['max_iter'], random_state=RANSTA), \n",
        "               KNeighborsClassifier(leaf_size = best_knn['leaf_size'], n_neighbors = best_knn['n_neighbors'], p = best_knn['p']), \n",
        "               RandomForestClassifier(max_features = best_rf['max_features'], n_estimators = best_rf['n_estimators'], random_state=RANSTA)]\n",
        "\n",
        "# Define a result table as a DataFrame\n",
        "result_table = pd.DataFrame(columns=['classifiers', 'fpr','tpr','auc'])\n",
        "\n",
        "# Train the models and record the results\n",
        "for cls in classifiers:\n",
        "    model = cls.fit(X_train, y_train)\n",
        "    yproba = model.predict_proba(X_test)[::,1]\n",
        "    \n",
        "    fpr, tpr, _ = roc_curve(y_test, yproba)\n",
        "    auc = roc_auc_score(y_test, yproba)\n",
        "    \n",
        "    result_table = result_table.append({'classifiers':cls.__class__.__name__,\n",
        "                                        'fpr':fpr, \n",
        "                                        'tpr':tpr, \n",
        "                                        'auc':auc}, ignore_index=True)\n",
        "\n",
        "# Set name of the classifiers as index labels\n",
        "result_table.set_index('classifiers', inplace=True)\n",
        "\n",
        "fig = plt.figure(figsize=(8,6))\n",
        "\n",
        "for i in result_table.index:\n",
        "    plt.plot(result_table.loc[i]['fpr'], \n",
        "             result_table.loc[i]['tpr'], \n",
        "             label=\"{}, AUC={:.3f}\".format(i, result_table.loc[i]['auc']))\n",
        "    \n",
        "plt.plot([0,1], [0,1], color='orange', linestyle='--')\n",
        "\n",
        "plt.xticks(np.arange(0.0, 1.1, step=0.1))\n",
        "plt.xlabel(\"False Positive Rate\", fontsize=15)\n",
        "\n",
        "plt.yticks(np.arange(0.0, 1.1, step=0.1))\n",
        "plt.ylabel(\"True Positive Rate\", fontsize=15)\n",
        "\n",
        "plt.title('ROC Curve Analysis', fontweight='bold', fontsize=15)\n",
        "plt.legend(prop={'size':13}, loc='lower right')\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/sklearn/linear_model/_logistic.py:764: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
            "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "Please also refer to the documentation for alternative solver options:\n",
            "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
            "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfoAAAGOCAYAAAB/psGeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd5hU1f3H8fdspywdpKqoeBQVEZEiVhRFTSSmiS0xtmg0aqKxpKAhdvOzxJigYsGKRk2ikYhGVBSUJooiHkVQ2UVEelm2zOz8/jh3lmHZcrfM3Cmf1/Pss3PPLfNdyn7n9FA0GkVEREQyU07QAYiIiEjiKNGLiIhkMCV6ERGRDKZELyIiksGU6EVERDKYEr2IiEgGyws6AJFUYIx5BPhpreLNwEfAPdbap+q4pyPwa+AUYE+veBnwT+D/rLUb67inO/Ar4DvAHkA18BXwX+Cv1tovG4mzADgPOA3YHygCSoHZwN+tte/4+HFTkjFmP9yfd8wt1tprk/C+j+D93VtrQwl4/hvAkcCX1trdW/v5Io1RohepXzEwEhhpjKm21j4dO2GM2RWYwfYEH7O/93W6MWa0tfaruHsOBl4Cdql1z37eVylwV33BGGO64D4QDKt1ak/vqwPwPd8/Xeo5tdbxj4CEJ3qRTKeme5GdHY2rKccnmQtqXfME25P8bUBX7+t2r2xP7xoAjDHFwAtsT/J/Bvp673MgcA9Q1UhcU9ie5F/GfaAoBHYHrgTWNvaD+WWMadNaz2qCH9U63tP7cJTWrLVHWWtDqs1LUFSjF6mDtbbCGHMvcLNX1C92zhgzEjjMO5xjrb067tarjDFHAMOBw4wxI73m9POA3t41T1trfxN3zyLgUmNMbn3xGGMOwjX3g2vq/561tsI7/hL4v9j9xpijgNe9cz+z1j7ilT9CrSZqY8z1wHXetccCV3s/2xRjzHigE/CYtfYncbHcD5wPlAG7WGu3GGP6eM85AfdhZg0wDfi9tXZVfT9X3DMHAft4h48DZ3qvTwUWxF0X/7P9wrvnDO/4BeCX1tqt3rXfBS4DBuI+hIWBT4D7rbX3NRDLGV4MAIdba9/2youBb3Efrv5irb3M+yDyR2Ao0BlYDywBnrDWTvbue4NaTffGmL2AG4DDgW7AJuAz4EVrbezfnEirUI1epH7x/bWr416Pjnv9BDuLLzva+z42rqzO5nlrbaSBWOLv/3tckvd7vx//AMYAbYAK4Fmv/Lve2AC8DxOneOX/jkvy83DJvy+QD/QCzgXeNcZ08/He8c32dwEfe69r1/Lj3QxcyvbWlJ8Bf4g7Pwo4xoulAGgLDAEmGWMuauC5z+I+qACcE1d+Ei7JAzxqjGkHTPfKd/HeYxfgKBrvQnkR9zP39u7rhusmOqOhm0SaQ4lepA7GmEJcjTHm2bjXfeNef1HH7fED6mLX7hpX9kkzQmrp/X58AwzCjU34C9trtZ1wtX1wSSyWuGPnJ+KS6RpgBC4ZHo77sLAbrluhMbGEXmqtXYBLhAC7G2Nqj0mICeO6MvoDsVaDH8ad/5d3vivuw8duwHveuXoTvfch6qFYXMaY9t7rH3jfP/Zi3Md7duxcAe7ve1xc/DsxxnRle+vFr3F/Xj2B44BH67tPpLnUdC+ys9fjXodx/ed/aeEzW7p7VDJ2n5pgrf3Qe73FGLMcWIHrtvghrik+lkhXA694r0/wvncD3q3juUfXUVbD65YY4B2+GPc91iVyKjC3jlsftNbO854xE/gxcV0sQAlwE64FZhd2/H23d0MxAfcBvwHa45L9VLb/nI9530uBCJALXAzsBSwGZllrGxovsQHXVN8BOB1o5933rrX21UbiEmky1ehFGhbC/bKPVxL3erc67okvK/W+fxVXZpoRR0vvB5eQGvJh/IG1Ngo86R2O81o5vu8dP22tDXuvuzfy3C6NnI9vtl9ijNkfN7Vxi1f2Q2NMXdPePot7Xe59j3Ux5AD/Ac4C+rBzpaaQBlhrl+Ga5cE13x+PS8jVeC0Z3tiDS4GNuA8Tt3rvucoYc13tZ8Y9O4LrZvgG17f/J+B5oNQY80BDcYk0hxK9yM6OxiWvf+GS4/m4ue8x8TX+0+u4P75shvd9elzZZXW9aUOD8Wrdf2Gsz7ye++P774viXu/ewPNhe7KMFxtv0AW4HujhHT8ed02sP/t9b3T5Dl80XnuO74e/G/eB4wO2f8DaFdclUFs47nXtFo8BuNkM4GrgnbxYnmsklnh/974fxvbuhzestTUf9Ky1f8P9WxmG61//L+5DxXXGmPgunh1Ya5/H9c8PxrVEPIH7UHmeMWZUE2IUaZSa7kXqYK1dY4w5D1dT6wD83hgz2Vq70Vo72xgzCzfYa6Qx5mbcFDuAq9ielN6OW8BmMi5Z9AZOM8Z8hesOWIPrrz0f1/d+bz3xLDTGvAh8F5ew/2mM+Q2uVtsblyz3xQ2Ai29xOM4Ycx/uw0uTE4i19kNjzCJc330s2X1qrY1vSn8ZOBsYbIy5CtfsHcHVVs/BNfHHfzCoYYwZils4qDGnAk1ZDCj+g9A2oNIYMwY4sQnPeIntXRexP7uaPnRjzC64D4D/BJbiZk/sjmviD+G6MuL/Loi79x7cuI8l3lch2wfiNdZCItIkqtGL1MPrZ73HO+yEGzgVczrwuff6GmCd93WNV/Y5cSOorbWbcYO0YqP3r8Y161fgaq+X4AaMNeRs3Oh2cAlrMVCJGxB4O97AMGvtCrb3aZ+C6w9+jbpr7H7EavV5tY5jrsM1Q4Nrvt6Aa3p/Hdd03lCFIr7Z/nu1WgLy2d5aUF/zfX0+wa1SCG4NhDJcq0ijU/1ivCb2++OKytixRaAN7u/xXS/OcuBG79wKts8cqMslwBu4P7cKtvf7b6TucQ4izaZEL9KwO3CJEuByY0xnAG/Fu1j/6ke4WuM2XPL9EzA0flU87575wAHALd51Zbh+6MXA/+FqhvWy1q7DNSNfAszCJYUKXEJ7nO2tCuCWyH0N2IpLQr8CnmnqD+95Etc3HbND7Tzuz+IBXIKrwn2geQf4Ha7GX59Ys/16XLN3/HPDbE+sfWhCi4S1tgr3weot3N/L57g1BGb6fYZnMtu7CP5prd0Sd24trlVmoRd/Fe7D25PAsdbaygaeeyswB/d3U4X7APKCd5/vDyMifoSi0WQM5hURST/GmANwLS4h4Bhr7YxGbhFJOUr0IiK1eHP3H8cNBCzErYBY14BAkZSnpnsRkZ21xY3cr8Z1gdTecEckbahGLyIiksFUoxcREclgSvQiIiIZLFMXzFF/hIiIZJN615lQjV5ERCSDKdGLiIhkMCV6ERGRDKZELyIiksGU6EVERDKYEr2IiEgGU6IXERHJYEr0IiIiGUyJXkREJIMp0YuIiGSwpC6Ba4x5CPgOsNpau38d50PA3cCJQBlwtrX2vWTGKCIikkmSXaN/BBjbwPkTcHtADwAuAP6ehJhEREQyVlITvbV2JrCugUvGAY9aa6PW2neBTsaYXsmJTkREJPOk2u51fYAVccclXtnXwYSTJNHojt+J7vi65lx9r/3eU/s6mnFPS+Op4319vs/KstW8+e1CItFw3P6E8RsVRtlctY2vy9ewsnwNK7etYWu4nPq11iaHzXxOnbclYuNFn89s9LLWjq2R5zX57RL899msxyfm7zMciVId9ffserc0q+fZTdG0ZzftfZr/bH/v17rPr/3s+n++g3O20D5aTWHe6fz+rFsTGMV2qZbo08/az+HhE6F8g7+EVzuRpZktoRALiwqxBQVEQrA0P5+ZbdtQnYT3rgiFiIb8//fsGo7Qobo6of+hRaT1NDcd1//bdOf7/f/mbe69oR3ujb+nPRFuKCwhTIj7yPUdSUulWqIvBfrFHff1ylLXqkWwZRV03BXad/cKQ1CTkGq9Bu84/nXt62jGPfVd18j71LqugigVsX+aoRCVRPkwUsb86s3Mr97MJ9VlOyX1fqFCOoby2Ok/RminF42o47q4ojahPMYW9qRrTuHOP7f3uk0ol1557eiV24aiUG4r/lk1dg9NuKeO61Lk38gLH6xk4YoNNWXRuO/bf2GF4sqJu27nsh2vq/3M7a+poyz+XP3PrBVPqO54YudXb64EoFtxoc94Qk38GXf+8/H3M+5cFiVU8/PU9Wc+et9duHT03t7ppvy7ohn3tMLvniZ8SM840ej2n3/FP6FtX67qekjS3j7VEv0LwCXGmKnAcGCjtTa1m+3DFe77kb+BIT8JNhYfqiJVfLr+Uz7f+Dmfb/icLZVbiBJl1dZVLNu4jJVbVhKt53NrXk4eB/Y4iKG7DGVQ90EU5hbSPr89+3fbn1A2/ydOMzdNW8JLi+r+b1W6oRiAPp3aJDOk5OkIJw3qxW9P3DfoSCRbfPkMLLkdjpkB+cXQ75Skh5Ds6XVPAUcB3YwxJcB1QD6AtXYSMA03tW4pbnrdz5IZX7PEEn1uYbBx1GH2ytn85/P/UB1XB39p2Uv1Xt+1qCtDew6lfX77mrKcUA57d967JrkX5RUlNOZs0VCyTbTSDduAupN5n05tlAhFWkO4DBZcBp9Phty2sH4h9DgikFCSmuittac1cj4KXJykcFpHLNHnpVain/7FdK6eeTWRaKTO838Y8Qf26LgHXYq6ANC1TVc6FnZMZogZr+Gac/3JNtGUzEUSbP0imDUeNi2BzoNh1FToYAILJ9Wa7tNPJLUS/cySmTz+8ePMXTWXorwi7jjqDvbouMcO13Rt05X8nPyAIkxvTamJq+YskoWWPwFzzoXqCtj7UjjoVsgNtiVUib6lYtO3EpDoF65eyMySmby/+n22hbc1en2UKHadJRKN0L1Nd+46+i4GdR/U6nFlssYSeVNq4krmIlmoeE8o6AzD7oe+3w06GkCJvuUS0EdfHi7n9nm388ynzwAQIuS7b7xnu57cfPjNDO4+WAPk4vitiTeWyJW8RWQnq9+Ctn2hfX/oNgJOXgZ5qTOgVYm+pWr66JvfNBONRnmz5E0e/uhhSjaXsC2yjc2VmxnQeQCXD7mcg3c5mHb57Vop4OxQO7H7rYkrkYuIb9URWHwjfPRH6DYSjn3LTaNLoSQPSvQtV5PoC5r9iItfu5i3St8CoF9xP4ryihi35zguG3KZRrk3USzB107sSuAi0qrKSmD2mbD6TWjbDw68JWXXClCib6lI82v0sSb6t0rfojC3kKdOeooBnQe0coDpranT0OITvBK7iCREyYsw52dQsRb6ngLDJ0Nhl6CjqpcSfUvV9NE3rUa/bMMyrpx5JZ+t/4y9O+/Nn4/8M/079k9AgOmjrqTe1GloSvAiklBVm1ySr9oCh/wN9rowZWvyMUr0LdWEPvpwdZgF3yxgztdzeHzJ42wLb+NUcypXDr0y65ro/SZ1JW4RSQnVYcjJg/wObl58YQ/onB6zmpToW6oJffS3zL2Fp+3TABQXFHPHUXcwZrcxiYwu5dTXhx57raQuIiklGoXlU+DjW2DMLCjsCj2PDTqqJlGibymfffT//OyfNUn+5sNv5si+R1JcUJzo6ALVWK1dSV1EUlrVJph7EXz5pKvJb1wc2DK2LaFE31KNzKOvqq7i9nm389QnT1GcX8z5g87nO3t8J4kBJld8cletXUTS1tp5MOs02PI5dB0Oo55y8+TTkBJ9S4UrICcfcnJ2OrW+fD1Xvnklc1fNZa9Oe3HP6HvoW9w3gCATr64meSV1EUlLyx51y9hGIzDwGhg00f2eT1NK9C0VLq+z2X7+qvlc/dbVrC5bzeh+o7np8JvSetGbpiwNq+QuImmtyxA3N374/WnXH18XJfqWilTuNBBv3qp5nPfKeYQIcelBl3LuAeeSE9q5xp8OGho8F08JXkTS2qr/QVEv6LQfdNofvvupG2WfATLjpwhSHTX6hasXUh2t5oZRNzBur3EBBdY4P4vRqKYuIhmtugoWTYCPb4XOB8HY+W5efIYkeVCib7lwZb0713Vv2z3JwTSsOeu/K8GLSMbashxmnQ5r34X2e8Kw+1J+8ZvmUKJvqXA5FHXcoeijNR8B0LGgY113BOalRV+zalM5PTu4FgglcRHJWl8+A3PPd1Podj/DrXKX3yHoqBJCib6lwhU79NEvWbuE11e8zuDugxnYdWCAgdWtZ4ciZl0zOugwRESCE94K7/3ajaofMQX2+EnQESWUEn1LRSp26KOfu2ouAKftc1rK7Acfa7KPr82LiGSdSDnkFkFeOzj8WSjoAh32DjqqhFOib4nqiFv/OLeAOV/P4cEPH+SrzV8B0Da/bcDBOTdNW8L9M5cB25vqRUSySjQKn/0NFt8Mx8+Btn2g24igo0oaJfqWCFcQBf6as4UHXjmfKFFyQ7l0LuzMHh33CDq6HZL8BUfsob54Eck+Fevc4jcl/3Lr1G9Z5hJ9Fkl6ojfGjAXuBnKBydbaW2qd3w14COgOrAPOtNaWJDtOX8LlLCgq5P7oWvq078NtR9zGoO6psZuRkryIZL3Vb8Hs06GsBHocBYc+nnVJHiCpq7gYY3KBe4ETgIHAacaY2iPW/gw8aq0dBEwEbk5mjE0SqWSjt/TtGfueoSQvIpIqlj0Crx0F21bCoD/B6P9lZZKHJCd6YBiw1Fq7zFpbCUwFaq8oMxCY4b1+vY7zqSNcHnQEO1GSFxHB7TLXcT84dibs/3vIyQ06osAku+m+D7Ai7rgEGF7rmg+A7+Oa908Bio0xXa21a5MTYhOEK4OOAKh7xzgleRHJOiUvQNEu0G04tN8DTngf0nT58daUin8CVwJHGmMWAkcCpUAk2JDqkSI1+tjUOXAj65XkRSSrRMph/qUwcxy8+zOIVrtyJXkg+TX6UqBf3HFfr6yGtXYlrkaPMaY98ANr7YakRdgUkUqWFLjFcoLYtKb2/HgthCMiWWfjJzBrPGz4ADoOhFFTleBrSXainwcMMMb0xyX48cDp8RcYY7oB66y11cC1uBH4KenTDZ9zX+eOdM8p4phdj0n6+8cnec2PF5GsEo26AXfzL4FIGex1AQy5E/JSYw2TVJLURG+tDRtjLgGm46bXPWStXWyMmQjMt9a+ABwF3GyMiQIzgYuTGWNTfLvtWwBO67gvPdv1DCQG1eRFJCtFymHxTZCTDyOfgV1/FHREKSvp8+ittdOAabXKJsS9fhZ4NtlxNUc0UgVAKCc/4EhERLJE1Sa3+UxeGzjiecgrhva7Bx1VSlNHRgss2PgZAL0KOyf9vW+atqRmhL2ISMaLVsOSP8O/d4dN7ncvnQ5QkvdBib6ZqiJVPL/mPTpGIhzTObkj3OPnyqtvXkQy3rZv4I0TYeFvIKcQKr4NOqK0orXum2nKx1NYF97KT7ZspSi/fVLfOzZnXtPoRCTjff0qvHMWlH8DvU6AkY9AUY+go0orqtE3w+qy1dz7/r10y23LBRs2QV5h0mPo06mNkryIZLZlj8Drx0HlOjjo/+Co/yjJN4Nq9M3wzdZvCFeHOanTvnSs/iSQRC8ikvF6HQ/dDoWD74auQ4OOJm0p0bdATmz1JSV6EZHW8eXTUNgNeh4DbXrBmLchFAo6qrSmRN8S1d7KvLnJSfS1V8ITEckY4a1uGdtlD0H7veA7n7iNaJTkW0yJviWqw+57XnKSrlbCE5GMtH4RzDoVNn0CnQ9yy9hm8W5zrU2JviViNfq8gqS9pVbCE5GMEY3CZ3+D966A6gowl8PgW5LWSpotlOhbIhpL9Imt0avJXkQyUnUlLH0A8othxLPQ5ztBR5SRlOibIUrUvajpo09cjT5+cZw+ndqoyV5E0l/5ajdNLrcQDn8WcttC295BR5WxlOibwa63APSMen1ICarRxyd5LY4jImmvOgKLb4CPb4PjZkPnA6F4r6CjynhK9M0ws2QmAIfTxhUkqI9eK+CJSMYoK4HZZ8LqN6HtrhCpCDqirKFE30SVkUrmfD2H/h3702+L14SfgBp9bNMarYAnImmv5AV492duhbt+P4DhD0BB8jcDy1ZaAreJ5q+az7bwNg7vcziEvU+krTxCVJvWiEjGWPYIzBwHkTI4ZBIc9g8l+SRTom+it0rfAuCIvke4RJ+TDzmt+8eoJnsRyRh9x7nNaI6fBwN+rgVwAqCm+yaaWTKTdvntGNJjCITLW63ZPjaFDmDVpnI12YtIeopGYdnDrtbe7xT3/ehpQUeV1VSjb4JVW1fx1eavGNZzGPm5+RCpbLWBeLF58oBWvhOR9FS5EWafDnPO9RbBqQo6IkE1+iap8EaJdinq4gpasUYPWvVORNLYmrkwazxsXQ7dRsKhT7quTQmcEn1LhCtanOi16p2IpLVoNSz5M3zwO7da6H6/hQOuV5JPIUlP9MaYscDdQC4w2Vp7S63zuwJTgE7eNddYa1OzgydcAUWdWvQIbVQjImktWg2lL0JRdxj5OPRUq2SqSWqiN8bkAvcCY4ASYJ4x5gVr7cdxl/0eeMZa+3djzEBgGrB7MuP0LVzRKn30arIXkbSz5Qtovzvk5Hm7zRW4ZC8pJ9mD8YYBS621y6y1lcBUYFyta6JAB+91R2BlEuNrmkjLm+5FRNJKpBIWXgUvDoDVb7uytn2U5FNYspvu+wAr4o5LgOG1rrkeeMUY80ugHXBsckJrouqI24++BRvaxK9+JyKS8rYsg1mnwdq50H4vyGsbdETiQypOrzsNeMRa2xc4EXjMGJN6ccZWxWtBjT42b1598yKS8r54CqYNdkl+97PghPegy5CgoxIfkp1AS4F+ccd9vbJ45wLPAFhr3wGKgG5Jia4pwm7OO3nNW/5Wa9mLSNpYNsXNj6caRkyBQx91e8hLWkh20/08YIAxpj8uwY8HTq91zVfAMcAjxph9cYn+26RG6Uek0n1vRqLXWvYiklZ2/SF8/TIc8EfosHfQ0UgTJbVGb60NA5cA04EluNH1i40xE40xJ3uXXQGcb4z5AHgKONtaG01mnL7EavTN2NBGa9mLSEqLRsH+1dXkAfLawainlOTTVNLn0Xtz4qfVKpsQ9/pjYFSy42qycPNr9ICa7EUkNVWsdUvYlvwb2u0Gu53WokHHEjytjNdcLeyjFxFJOatnwuwzoKwEdjnaLYCjJJ/2Um80e7poZh99bBCeiEjKqI7AouvhtaNh29cw6AY4+lVo2zvoyKQVqEbfXM3oo9cgPBFJSaEQrJ0Dbfq6vvjuhwYdkbQiJfrmqplH7z/RaxCeiKSUDYuh034QyoGRj0FOrts/XjKKmu6bq5kL5mgQnogELlIO8y6BaftDqTc2uqibknyGUo2+uSJNq9FruVsRSQkbl7h94zcsgo77Qbtdg45IEkyJvrma2HSv5W5FJFDRKCx7GOb/EiJlsNeFMOQOyFPlI9Mp0TdXMwbjqdleRALzxRNufnx+Jxj5KOz6g6AjkiRRom+uFi6YIyKSVLv+GL59G/a71i2EI1lDib65tGCOiKSyaDUs+TPkFMI+l7mFb4ZNCjoqCYASfXM1cTCeiEjSbFsF7/wEVr3qau8Dfg65zd9SW9Kbptc1V2wwno8+eq2GJyJJs3I6/PdAl+R7nwjHz1OSz3Kq0TeXz3n0Wg1PRJKiOgIfXAtLboecfBhyJ5jL3Kp3ktWU6Jug3OuXz8/Jh3CZK8yrf8OH+CSv1fBEJKFCOVC2AooHwKip0GVI0BFJilCib4LSLaUA9G7fG75d7QobqNFryVsRSbg170K3Ea7mPuw+IAT5xUFHJSlEffRNsHLLSsBL9DV99A1v4ai58yKSEOGt8O458MpI+GKqK8vvoCQvO1GNvgliNfq+7fs22kevJW9FJGHWv++Wsd1kofMQ6HJw0BFJClOib4Idmu5rEn3dNXoteSsirS4ahU//CguvhOpK2OfXcOBNTVqhU7JPkxK9MSYE9AX6AR9Ya7cmJKoUVbqllLZ5belU2CluHn39ffRqtheRVvXl07DgUijsBiOmQJ8Tg45I0oDvPnpjzC+AUuBL4C3AeOXPG2MuT0x4qSMajbJyy0p6t+9NKBRq0jx6EZFWseuPYODVcMIHSvLim69Eb4z5DXAH8AAwGoifmPkGcGqrR5ZiNlVuYkvVFvq07+MKwhVurmqOxjOKSIJUh2HRdbDoeneckwuDb4G2vQMNS9KL36b7i4EJ1trbjDG5tc5ZYG+/b2iMGQvcDeQCk621t9Q6fydwtHfYFuhhre3k9/mJsi3sVrZrX9DeFYTLG10sR0Sk2baugNlnwLdvQbv+MPAqyGsbdFSShvwm+p7AgnrOVQO+Mp73IeFeYAxQAswzxrxgrf04do219ldx1/8SOMhnjMkVrqh3IJ5G3ItIi6z4F8w5ByrXQ78fwvAHlOSl2fy2Oy8Fjqzn3BHAx/Wcq20YsNRau8xaWwlMBcY1cP1pwFM+n51ckYp6a/QacS8izRKthnkXw1unQGSbWwDnsGegIPBGTUljfmv0dwF/M8ZUAs96ZT2MMecCvwbO9/mcPsCKuOMSYHhdFxpjdgP6AzN8Pju5wvUnetCIexFphlCOS/Yd94NRT0On/YKOSDKAr0RvrZ1sjOkMTAD+6BVPA8qA6621TyYgtvHAs9baSAKe3XLhCijSp2wRaaFoFFb9D3oe65axHXKHK89T15+0Dt9Dxq21twO9gROAM4ETgT5euV+luDn4MX29srqMJ1Wb7aHBPnoREV8qN8Ks0+D14+DzB1xZXhsleWlVvmr0xpifAC9Za9cCr9Q61wX4jrX2UR+PmgcMMMb0xyX48cDpdbzfPkBn4B0/8QWigT56EZFGrZnjkvzW5dDtUOh1fNARSYbyW6N/GNiznnP9vfONstaGgUuA6cAS4Blr7WJjzERjzMlxl44Hplproz7jS67qiJvf2siGNiIiO4lWw8e3wquHwdYvYL/fw7FvQrvdgo5MMpTfwXihBs51BTb5fUNr7TRc/3582YRax9f7fV4gGtnQRkSkXiX/gvevgTa94dDHYZejG79HpAXqTfTGmHHsOPXtD8aYb2tdVgQcjmuSzx7hcvc9T8vfiohP0agbbNf3FDjwZtjzXCjqHnRUkgUaqtH3AA6IO94Tt3BOvEpcn/0NrRxXaotUuu9K9CLSmEglLPq9G9cz9G6X7Pe7JuioJIvUm+ittQ/g1rbHGPM6cJkce5kAACAASURBVJG19pNkBZbSYjV6bWgjIg3Z/LkbcLduHhTvDVWbIb846Kgky/idR69OpHhh1ehFpBFfPAVzfw7hzdD/pzD0r5DfPuioJAv53o/eGFOM67PfmzrWtrfWXtWKcaU29dGLSH2iUZh7Pnz+IOS1h5GPQ/8zgo5KspjfefR7ArOBNkA74Fugi3f/emAjkD2JXn30IlKfUAgKu0OXg2HUVCjeK+iIJMv5nUd/J25k/S64qXYn4pL+mcAWsmA/+h2oj15E4kWj8NU/3Bx5gEETYcxsJXlJCX4T/TBgEuBNIKfAWhvx1rj/P9z+8tmjZh69Er1I1itfAzPHwds/hk/udGU5+VpQS1KG30RfBGyy1lYD63Br3sd8BBzY2oGltAYWzIntRS8iWeCbN+C/B0Lpi7DLMbD7Tit6iwTO72C8T4HY+owLgQuNMdOACHAusDIBsaWuBgbjaS96kSxQHYaPJsJHN7itZQ+8Cfa9CnJyg45MZCd+E/1UYDDwGPAH3Fr1m4BqIBc4OxHBpaxGBuNpL3qRDPf1K/DRn9z69Ic+Bd1HBh2RSL38zqO/I+71u8aY/XHb1RYBM6y1HyUovtSkwXgi2ak64mrtfU6EQ/4Ou42Hgk5BRyXSIN/z6ONZa1cA9wMYY0LGmFOttU+3amSprJ4Fc2L98306aS9pkYwS3gYLr3Ar2x36mCsbcGGwMYn45HcefXdgTfy2scaYNrj++V8BuwNZlOjr7qNX/7xIBtr4McwaDxs+hI77Q+VGKOgYdFQivtU76t4Y09YYc78xZiuwClhvjLnSO/dz4AvgL8BS4KjEh5pCIvVPr1P/vEiGiEZh6WR4eahL8gMuguPnKslL2mmoRj8B+CnwEPABbtT9b40xw4EfADOAa6212bVFLWyfXqc+epHMFI3COz+BLx6H/E5u3/h+3w86KpFmaSjRfx+YaK29MVZgjHkTmAY8ZK09L9HBpawG5tGLSAYIhaDTAdB9FBz6hBtdL5KmGlowZzfgzVplseMpiQknTdQkeq18JZIxotXw+UNQXeWO970SjnlDSV7SXkOJPh+orFUWO96amHDSRGTnGr1WxBNJY9tWwevHw5xzYfEtriyUAznNmpgkklIa+1f8S2PM13HHIe/7ZcaYb+LKo9baq1s3tBRW00e/vUavEfciaWrly64/vuJb6P0dN+hOJIM0lOi/Ag6ro/xL4IhaZVEg+xJ9rT56jbgXSSORSlj0O1jyZ8gpgCF3gbnU9c+LZJB6E721dvdEvKExZixut7tcYLK19pY6rvkxcD3uA8QH1trU2ilCffQi6e/bt12SLx7g9o3vMiToiEQSwu/uda3CGJML3ItbPncgcJoxZmCtawYA1wKjrLX7AZcnM0Zf6uijF5E0EduroudoOPRJGPuekrxktKQmety+9kuttcustZW4zXLG1brmfOBea+16AGvt6iTH2DjNoxdJP1Vb4J2z4a0fuHnyALufBvntAw1LJNGSPaS0D7Ai7rgEGF7rmr0BjDGzcM3711trX05OeD6FyyEnH3KS/TlJRJpl3UK3jO3mT6HLUKhcD4Vdgo5KJClSMVPlAQNwy+qeBjxgjEmt7aHCFWq2F0kH0Sh8cje8MsIl+X2ugDGzlOQlqyS7Rl8K9Is77uuVxSsB5lhrq4DlxphPcYk/dZbaDVdoIJ5IqotGXS3+q2egsDuMnAK9Twg6KpGka3KiN8aEgF7AamttuIm3zwMGGGP64xL8eKD2iPp/4WryDxtjuuGa8pc1Nc6EilTUuViOtqcVSSGhEPQ8BirXwchHoY3WuJDs5Lvp3hhzojFmDlCOm2M/yCu/3xhzpp9neB8MLgGmA0uAZ6y1i40xE40xJ3uXTQfWGmM+Bl4HfmOtXev7J0qGcIUWyxFJRdVh11QfLnPHe54PR09Xkpes5nc/+p/gdrF7Avgb8HDc6c9w+9I/7udZ1tppuI1x4ssmxL2OAr/2vlJTuAKKdhw2oMVyRAK29SuYfYabH1/+DQy+yVv8RgvgSHbzW6P/HXC7tfan7JzQF+PmxGcP9dGLpJYV/4T/DnZJftcfw8Crgo5IJGX47aPfDXi1nnPlQIfWCSdNRDTqXiQlhLfBwivgs79DbhsY9gDsea6WsRWJ47dGvwI4qJ5zQ4GlrRNOGohGXT9grmr0IoFb/z4svc/tHT92Pux1npK8SC1+a/QPAtd5O9b9yysLGWOOAa4CJiYiuJRUXe2+q0YvEoxoFMJb3Yp23UfCES/ALqMhT7NeROriN9Hfipv/PgWIeGWzcSvX3Wet/UsCYktNUe/Hz9PytyJJV7kB5l7gBtuNngE5udDnpKCjEklpvhK9NxL+YmPMncBooBuwDphhrf00gfGlnmisRq9EL5JU374Ds0+DrV9C91FQtQEKuwYdlUjK8zu9rp21dqu1dinZ1B9fl2qvRq8NbUSSI1oNH98Ki/7gXu//B9h/AuQke2FPkfTk93/KamPMf3C7zU2z1lYkMKbUphq9SHK9/WNY8Ry06Q2HPgG7HBV0RCJpxW+ivwr4MfAssMUY8wIu6U9vxjK46U199CLJteuPoLoShj8ERd2CjkYk7fiaXmetvddaeyRuQN51wJ7Ai8A3xpgHjTFjEhhjalGiF0msSCV8+Cc38A5gt1PhiH8ryYs0U5M6uay1K4G7gLuMMbviavm/An7a1Gelm2g0CkAoNr1OffQirW/zUrfj3LoFbs/4g+9w5ZobL9JszUrOxpi9gFO9r164BXUy2pptawDoktfWFahGL9K6lj8O8y6C8BbY42wYlD3Lc4gkku9Eb4zZje3JfTCwGvgHcJG1dlZiwksdX291O9T1yi92BVowR6R1VG2B+ZfA8imQVwwjH4f+ZwQdlUjG8Du9bi5wMG7u/PPAlcCb1trqBMaWUmoSfW47V+DV6LUXvUgLbVkKXz4JXYbCqKlQvGfQEYlkFL81+sXABOBVa22ksYsz0aqtqwDoGdd0f9O0Jdw/cxmgvehFmiQadX3whV2g82AY/T/oOkJ7SIgkgN+V8X6W6EBSXU2iz/H65nMLeWmRq+VfcMQe2otexK/yNfDuz2DbSjhuthvY2uOIoKMSyVj1JnpjzInA29baTd7rBllrp7VqZCnm661fU5BTQJdorivwmu77dGqjJC/i1zevw+wzXZLveazbnEYzWEQSqqEa/X+AEcBc73VDorgNbjLW11u/plf7XoQi3qKAGnUv4l91GD78Iyy+EUK5MPgW2Pc3EPK7U7aINFdDib4/8HXc66xVEalgXfk6BnQeAEr0Ik0361RY8Ty02x1GPQXdRgQdkUjWqDfRW2u/jDuMAl9ba6tqX2eMyQN6JyC2lPHN1m8A6Nm2J1R6iT63EMjeJf9FmmTPCyCUD8MmQUGnoKMRySp+R90vB0bimvFrO9Ar99V0b4wZC9ztXT/ZWntLrfNnA7cDpV7RX621k33GmRA1U+va94LVblCem0evRC9Sp/A2+PA62OcKaLML9D7efYlI0vlN9A2tP+k74xljcoF7gTFACTDPGPOCtfbjWpc+ba29xGdsCVeT6Nv1grDX0JGnaUAiddqw2C1ju/EjiGyDofcEHZFIVmto1P0g3Ap4MScaY/apdVkRbr37T32+3zBgqbV2mfceU4FxQO1En1Jiib5nu55xffRaGU9kB9EofP4ALLjcJfgBF8NBtwcdlUjWa6hGfwpupzpwffQT6rluOfBzn+/Xhx3XxS8Bhtdx3Q+MMUfgPkD8ylob6Fr6sT56V6OP9dGrRi9So3IDzL0AvvoHFHSGQ5+Eft8LOioRoeFtam8CioEOuKb70d5x/FehtXZPa+3/WjGmF4HdrbWDgFeBKa347GbZoUYfVo1eZCfl38DKadD9cDjhAyV5kRTS0Kj7KiA2yr61JruW4va0j+nL9kF3sfddG3c4Gbitld672b7e+jWdCjvRJq9NXKJXjV6yXHXEJfi2vaGDgWPfgk4HQE5G71gtknYa6qMfCHxura3wXjeojgF1dZkHDDDG9Mcl+PHA6bXet5e1NjZ//2RgiY/nJkw0GmXV1lXs3mF3VxAud99Vo5dsVrYS3jkLykpg7ALIbw9dDgo6KhGpQ0MfvT9i+8p4H+H66esSwufKeNbasDHmEmC6d/1D1trFxpiJwHxr7QvApcaYk4Ewbre8s33+LAlRHilnW3gbXdp0cQWRSgBuemWZdq2T7FQ6Dd79KVSsgT7fhehOy2uISAppKNEfzfbR8Ee31ht6a+JPq1U2Ie71tcC1rfV+LRWNus83ObHei3A55OTz0odugJ52rZOsEamA968FeyfkFMDBf4G9L4FQQ7NvRSRoDfXRv1nX66wXrqhptteGNpJVZp/ulrHtYNy+8Z0HN36PiATO16gZY0wPoJ21drl3HALOBwYCr1lrX0xciCkmXEFZdS6lm9RsL1lmnysgvxMcfLfrkxeRtOB3NP0jwK/ijicCfwPGAv/0lq3NDpEKNofdcAQ120tGq9oMcy+CLV+44+6HwogHleRF0ozfRD8EmAFgjMkBLgR+a63dB7gRuDwx4aWgcAVVoXw120tmW/cevHwwLJ0EH9/S+PUikrL8JvqOQGx++8FAF+AJ73gGsFcrx5W6whVUkh90FCKJEY3CJ3fBKyNg82duz/iD/xJ0VCLSAn5XtijB9ce/BZwEfGKtjS100xEoT0BsqSlcQSXaZlMyUPkaePdsWPkSFPWAEY9qxzmRDOA30T8E3GaMORaX6OOnv40g4EVtkipSQSVaFU8yUKQM1syGnmNg5KPQpmfQEYlIK/CV6K21NxtjSoFDgF/iEn9MF9xStZmvOgLVYaq0xKdkiuowbP0SiveEdrvCce9A8QAItdaq1yISNN8Zy1r7KPBoHeUXtmpEqcxb5141eskIW7+EWadD2Qo44X0o7OLmyItIRvGd6I0xecAPgMNwtfh1uD7756214cSEl2K8de41GE/S3lfPwZzzoGoD7HoqhBpdwVpE0pSv9jlvwZz5wFO4Pvo9vO9TgXnGmO4JizCVeOvcK9FL2gpvg7kXwts/hOoKGD4ZRj0FBR2DjkxEEsRvjf4OoCswwlo7N1ZojDkEeM47f1brh5divBp9lRK9pKt3fgIrnoVOg9wyth21FoRIpvM74uZE4Or4JA9grZ2HG4F/UmsHlpJiffQhJXpJU/v/Afa+FI6foyQvkiX8JvpCYHM95zZDloxOqxmMp0QvaaJyPcw+CzZ6M2A7D4Khd0NuUbBxiUjS+E307wJXG2PaxRd6x1d75zNSFLdNbSgU0qh7SS/fzoZpg+GLx8HeFXQ0IhIQv330VwCvAyuMMa8A3wA9gOOBEHBUQqJLAZXeALzC3EK3HzdQ5X+ygkjyVUdgya2waAIQhQOuh/1+H3RUIhIQXzV6a+37wADgfqA7MAaX6CcBA6y1HyQswoBVeMm9ILdg+/S6kGr0kqK2rYLXj4MPfgdFPeGY1+GA6yBH0+dEslWjVVNjTFdgd2CVtfaahEeUYmKJvjC3EMKudl+hPnpJVaFc2LQE+o6D4Q9CYdegIxKRgNWb6I0xxcCDuEVyYmXzgDOstZ8nIbaUUFOjzynQ9DpJTZEKt9Ncp/2hqDscNwfa9oVQKOjIRCQFNNR0/0fgBGACbvrcJUAfdlznPuPF+uiL8oriFsxR072kiE2fwSuHwmtHQ9lKV9aun5K8iNRoqOn+ZOD31tq7YwXGmI+AN4wxHa21GxMeXQqou49eg/EkBSx/DOb9AsJbYI9ztLqdiNSpoYy1GzCvVtkc3Cj73YBFzXlDY8xY4G4gF5hsrb2lnut+ADwLHGKtnd+c92oNO/TRV7oa/aqtQIegIpKsV7UZ5l0MXzwGecVw6FOw+/igoxKRFNVQ030uUFWrLBJ3rsmMMbnAvbgugYHAacaYgXVcVwxchvtgEagdptfVbGqTx0mDegUZlmSzuRe4JN91GJz4vpK8iDSosTbom40x6+KOYx1/txlj1seVR621p/p4v2HAUmvtMgBjzFRgHPBxrev+BNwK/MbHMxNqh6Z773Vxu/b89kQtHyoBGXQDtN8T9p8AuRovIiINa6hGPxNXc+8e99UNeBP3ASG+vIfP9+sDrIg7LvHKahhjhgD9rLUv+XxmQu04vU5r3UsAyr+FmafAWq8nrXhPOPAGJXkR8aXeGr219qgkxgGAMSYHtxPe2cl+7/rsOBhPa91Lkq2aAe+cCdu+hsLu0PWQoCMSkTTjd6371lIK9Is77uuVxRQD++NG9n8BjABeMMYMTVaAte3YRx9bAleJXhKsusqtbjfjWFejH3wrDJsUdFQikoaSPU9sHjDAGNMfl+DHA6fHTnpT9rrFjo0xbwBXpsyo+4g2tZEkKFsJb/8Q1rwD7frDqKeg2/CgoxKRNJXUGr21NoxbeGc6sAR4xlq72Bgz0RhzcjJj8avOPnrV6CWR8ttD+Tew23g4YaGSvIi0SNJXfrHWTgOm1SqbUM+1RyUjpobUOb1Og/GktYXLYMMi6DYC8ju4ZWwLu2qFOxFpsWT30aedHQfjuaSvPnppVRs+gunDYMZxsGWZKyvqpiQvIq2iSTV6Y0wIN4CuH/CBtXZrQqJKIRXh+KZ7V6PX7nXSKqJRWHofvPcriJTD3pdAm95BRyUiGcZ3jd4Y8wvcALovgbcA45U/b4y5PDHhBW/HBXNUo5dWUrke3v4RzLsIctvCEf+CofdAblHQkYlIhvGV6I0xv8HNb38AGM32FfIA3gD8rIqXlmr30VeRRzSkHg9pofmXwornoMcRcOIHbv94EZEE8Nt0fzEwwVp7m7defTwL7N26YaWO2qPuNeJemi0a3d7vPvgW6DgQ9r0Kcpq1dYSIiC9+q6Y9gQX1nKsGMra9saJaiV5aQdlKmDEGvnndHbftA/tdqyQvIgnnN9EvBY6s59wR7LwpTcaINd3HNrVRopcmK30J/nsgfPMafDk16GhEJMv4bbq/C/ibMaYSt0c8QA9jzLnAr4HzExFcKqiIVJCfk09OKAfCFVRpDr34FamA968BexfkFMLB98DeFwcdlYhkGV+J3lo72RjTGZgA/NErngaUAddba59MUHyBq4xUumZ78Jru2wUbkKSHslJ487uwfiF0MDDqaeh8YNBRiUgW8j2P3lp7uzFmEnAo0BVYB7zjrU+fscrD5a7ZHrxE3ynYgCQ9FHSBaBj2OAeG/gXy9AFRRILRpAVzrLWbcevUZ40davSRCm1oI/Wr2gxr50DPYyGvDYyZBfnFQUclIlnOV6L3FstpkLX2by0PJ/VURCooLiiG6ghUh6nKSfr2AJIO1i2At8dD2Qq3EU3HfZXkRSQl+M1af23gXNT7npGJvqZGH9YWtVKHaDV8chd8cI3bQ37g1VC8V9BRiYjU8DsYb6dpeMaYTsDxwNXAaa0cV8qoiFTsuHOdptdJTPm38O7ZsHIaFO0CIx+DXmOCjkpEZAfNboe21m4AnjbGdATuA45qraBSRTQapbK6kvzcfO1FLzv74Lcuyfc6HkZMgTa7BB2RiMhOWqPDeTkwtBWek3KiXq9EbijXzYlGG9pkvWg1xPY6GHwrdDoQ9v7F9jIRkRTTot9OxphewBW4ZJ/ZYjV6LZiTvbZ8Aa8eBiv+5Y4Lu4C5REleRFKa31H337J90F1MAVAMlAPfb+W4Uo+a7rPbV/+AOedD1UZY9Sr0+17QEYmI+NKSUfflQAnwsrV2beuFlKI06j47hctgweXw+QNu3/gRD0P/nwYdlYiIb40memNMPvA/YLm1dmXiQ0pRNX30mkefNcpK4fXjYOPHri9+1FTouE/QUYmINImfzsUIMAPI7t9wsel1IdXos0ZRD8jvBHv/Eo5/V0leRNJSo9VTa221MeYz3J70LWaMGQvcDeQCk621t9Q6fyFwMe4DxhbgAmtt8Nvght12tRXqo89slevhmzeg3ymQkw/HzIDYEsgiImnI73Dh3wETjDEHtOTNjDG5wL3ACcBA4DRjzMBalz1prT3AWjsYuA24oyXv2Wq8Gr2m12Ww1W/DtAPh7R/B+vddmZK8iKS5emv0xpgjgPestVuA3+N2rHvfGFMKfEOtUfjW2mE+3m8YsNRau8x7j6nAOKCmxm6t3RR3fbva7xOYiKvRazBeBqqOwMc3w4fXueP9r4OOLfpMKyKSMhpqun8dGAnMBT7yvlqqD7Ai7rgEGF77ImPMxcCvcVP4RrfC+7ZcTR+9BuNllLJSmH0mrH4D2vaFQ5+EHocHHZWISKtpKGuFYi+stT9LQiw1rLX3AvcaY07HtSYEP58prBp9Rvr4Fpfk+34Phj/oFsEREckgya6elgL94o77emX1mQr8PaER+aU++sxRXQWhPAiF4MCbocvBbm58KNT4vSIiaaaxRH+iMcbXnCJr7aM+LpsHDDDG9Mcl+PHA6fEXGGMGWGs/8w5PAj4jFUS0Ml5G2PQpzBoP5jLY46eQ3x72ODvoqEREEqaxRD/B53OiQKOJ3lobNsZcAkzHTa97yFq72BgzEZhvrX0BuMQYcyxQBawnFZrtQWvdp7toFJY/CvMvhvBWWPeeS/QiIhmusUR/NDC/Nd/QWjsNmFarbELc68ta8/1ajda6T19Vm2HeRfDFE5Dfwa1wt9upQUclIpIUjSX6bdbarUmJJNWFtU1tWipbCf87ArZ8Dl2HwainoP0eQUclIpI0mivmV2x6nUbdp5c2PaF4AOz6Qxj0J7fanYhIFlGi96tmwRwlipRXvhpW/tf1wYdy4Mj/QE5u0FGJiASi3kRvrfW7PG52qFkwR4k+pa36H8w+C8pXQYd9oNtwJXkRyWpK5n55C+aojz5FVVfB+9fCjOOgYg0Mvg26HhJ0VCIigVPTvV9ejV6716WgLV/ArNNg7btuoN2oqUryIiIeJXq/IqrRp6zP/u6S/G6nw7C/uyl0IiICKNH7Fy6HnHyiIfV2pITwNsgtcsvWDpro+uL7nqJlbEVEalHW8itcAXlFQUchABs+hOlDwf7FHecWQr/vK8mLiNRBNXq/whWQVwDVQQeSxaJRWDoJ3vs1RMphW0nQEYmIpDwler8iFWwK51K6aRt9OrUJOprsU7EO5pwHJf+Ewq4w6hno+92goxIRSXlK9H6FK9hU5Xo6ThrUK+Bgssy2VTB9GJStgB5HwqFPQNs+QUclIpIWlOj9CldQSTv6dGrDb0/cN+hoskvRLtD9MOiwL+z3Wy2AIyLSBEr0foUrqKRT0FFkj7JS+OpZ2OcyN8ju0Cc02E5EpBmU6P2KVGhDm2QpeRHm/Awq1kKXIdDjcCV5EZFm0vQ6P6JRqA5TFdLnooSKVMD8y2DmyVC1BQ75m2uyFxGRZlPm8iPq5tSpRp9AmyzMGg/r33d98aOmQudBQUclIpL2lOj9qEn0Wv42Yb76h0vye54PB98Jee2CjkhEJCMo0fuhRJ8YVZshtw3k5MHAa6HrMOh1XNBRiYhkFPXR+xIFtKFNq1o7D/57ECy+0R3n5CrJi4gkQNJr9MaYscDdQC4w2Vp7S63zvwbOA8LAt8A51tovkx3nDmI1+pASfYtFq+GTO9ze8dGI20deREQSJqk1emNMLnAvcAIwEDjNGDOw1mULgaHW2kHAs8BtyYyxTtVqum8V5avhjZNg4W+gsBuMfgUOvCHoqEREMlqya/TDgKXW2mUAxpipwDjg49gF1trX465/FzgzqRHWRaPuW658DUw7EMpXQa+xMHIKFPUIOioRkYyX7ETfB1gRd1wCDG/g+nOB/yY0Il9col9bAarUN1NRN+j3A2jfH/b5FYQ0PEREJBlSdtS9MeZMYChwZNCxxGr0FdF8bWjTFFuWw/JHYf8JbmW7Q/4adEQiIlkn2Ym+FOgXd9zXK9uBMeZY4HfAkdbaiiTFVr+oG3Vf1KYtl2pDG3++fAbmng9Vm9zqdj2PCToiEZGslOxEPw8YYIzpj0vw44HT4y8wxhwE3AeMtdauTnJ8dfNq9Jpe50O4DBZcBp9Phty2MOJh2GV00FGJiGStpHaUWmvDwCXAdGAJ8Iy1drExZqIx5mTvstuB9sA/jDHvG2NeSGaMddJgPH/WL4KXh7ok33kwnPAe7HG2NqQREQlQ0vvorbXTgGm1yibEvT422TE1qmYefcoOaUgN37wOm5bA3pfCQbdCblHQESVFNBolEolQ7U3DFBFpTbm5ueTm5jb7fmUuP7w+etXo61Cxzq1Ln1sI5lLoegh0PzToqJImEomwefNmCgoKyMvTfycRaX3btm0jJyeHtm3bNut+/WbyIxoB1Ee/k9Vvw+zTod/34eC7XBN9FiX5aDTK5s2b6dixIyF1T4hIghQUFLBp0yai0WizftdoMrMfNTV6JXoAqiPw4UR47UjYVupWufP+jLJJJBKhoKBASV5EEi4/P59wONyse1Wj90Nr3W9XVgKzz4TVb0LbXWHUk9B9VNBRBaK6ulrN9SKSFDk5OUSbWaHSbyk/tE2tU7ne7ThXscY11w+fDAWdg45KRCTjtaTlUIneD82jdwo6w4CLoE1v2OvnmjYnIpIG1EfvRzbPo9/4CSy4vObPgEETYcCFSvKygwkTJjBx4sRm33/SSScxbdq0xi/MMJMmTeLCCy8MOgzJcKrR+5GNg/GiUVj2CMy/BCJlbse53mODjkpa4KyzzmLkyJH84he/aPVn+03yJSUlHHPMMbz55pv07Nmzpvyll17y/V7GGIqKisjJyaGgoIB9992Xq6++mn33Tb/lqYNK8tu2bePwww+nU6dOvPrqqzs0C99zzz0sWLCARx55ZId76iqfPXs2kydPZtGiRUSjUXr27Mlxxx3HOeecQ3FxcZNjmjhxIq+++ioAxx13HBMmTKCoqO71OCKRCJMmTeK5555j/fr17LvvvkyYMIF99tkH2P5vrU2bNjU/X3FxMTNnzqx5xrnnnou1lrKyMoqLixk7dixXXHEFBQWZn0ZMiAAAIABJREFUValTjd6PbBuMV7UJZp8Bc86BnHw47BkleUkpDz74IAsXLuS1116jS5cuXHzxxQl7r6qqqoQ9OyixD1YrV65k9uzZzXrG888/z4UXXsioUaN4+eWXWbBgAZMmTWLr1q1Ya5v8vBtvvJHly5fz8ssvM336dJYtW8bNN99c7/UPP/wwL7zwAlOmTGHOnDkMHTqU8847jy1btuxw3csvv8zChQtZuHDhDkke4Morr2TGjBm89957PPfccyxevJi//jXzNt9Sovcjm/ro185zA+6+fAq6joAT3oddfxR0VJJgpaWlXHTRRQwfPpwjjzySG2+8kfLy8przy5cv58wzz2TIkCGcfPLJTJkyBWNMzflrrrmG3/3ud4BbX+DOO+/ksMMO46CDDmL06NE89thjAIwbNw6AsWPHctBBB3HvvfcCMHr0aP7973/XPO+TTz7h3HPPZcSIEQwbNoyzzz67zrjbt2/PySefTGlpKevWrQMgHA4zadIkjj/+eIYOHcr48eP58MMPa+6pqqripptuYuTIkYwaNYoHHniAMWPG8PzzzwMugY0ZM4bJkydzxBFH8L3vfQ+ATz/99P/bO+/4nq7/jz+zIxEhdojxNY5RI8OIvdVKUErrF1SINCiKGNUaVauUVBHEbo1Qu2pVVatGKbVv1cwyQxBJPlm/P+4nt5l8QnwS6Xk+Hp+H3M89957Xvffjvs95n/c5b01TixYtmDt3rtYI0Ol0fPrpp7i7u+Pi4kK7du348Uc1w3ZoaCje3t64ublRr149unXrxrVr1wC1l5z62h4+fIi/vz+NGzemcePGjB07lkePHmn7W7VqRWBgIP369cPZ2ZnOnTvz559/GvSMU7Nx40Y8PDxo1qwZGzduzPbx0dHRTJ8+HR8fH7y9vSlWrBgATk5OTJgwATc3t2ydLzY2lh07djB8+HCKFStG0aJFGT58ONu2bSMuLvO8Znv27OH999/HyckJS0tLhg0bxqNHjzhw4IDB9VavXj1N793ExITr169nS/ubgHTdG0JKmtr/gqGPuqiml60xHmpPUXv0EoOZvvsSP5yNMEpdnWqXZkIOZFNMSEhg8ODBuLi48PPPP/PkyRP8/PyYNWsWkyZNIiEhAV9fX5o2bUpQUBD37t3jww8/zPJ8R44cYevWrWzatInSpUvz4MED7ty5A8D27dtp3bo1e/bsSeO6T83du3fx8vJi4MCBLFiwAHNzc06ePJlp2cePH7N161aKFi1KoUKFANV4priUHR0d2bJlCwMHDmTfvn3Y29uzZMkSDh8+THBwMMWLF2fOnDmEh4enOW9YWBh3795l3759JCcn8+DBA7y8vBg5ciSLFy/m4cOHfPjhh1hZWTF06FC2bt3KuXPn2L17N0WKFCEiIoLo6GgA5s2bR+nSpVm8eDFmZmZcuXIFe3v7TK9n9OjRWFhYaPEKY8aMwd/fn6VLl2plvv/+exYtWsT//vc/Zs2axbhx49i3b1+WzyM9ly9f5uzZs0yePJnw8HBGjhzJ/fv3NWNtCKdPn+bJkyd06dLlueUmT57Mrl27stzv4+ODj48P169fJy4ujpo1a2r7atSoQWxsLNevX9fc8alJTk7OMN0sOTmZS5cuaY0zgJ49exIfH0+VKlUYOnQoDRo0yKBx27ZtxMTEYG9vT2Bg4HOv6U1E9ugNIb/36GPuQLze3VWxL3T8C+pOl0b+P8LZs2e5ceMG48aNw8bGhpIlSzJixAi+//57kpOTOXPmDGFhYYwePRpra2ucnJzo169fluezsLAgLi6OK1euEBcXR9GiRalRo4bBerZv3065cuUYPHgwNjY2WFpa0qhR2hUXBw0ahIuLC/Xq1eOvv/5i4cKFmJubk5yczJo1a/D398fJyQkzMzN69uxJiRIlOHTokHb+gQMH4uTkhLW1NaNHj84wdcnc3Fy73gIFCrBt2zaEEPTu3RtLS0tKlizJ4MGDNS+EhYUFz5494+rVqyQkJFC6dGkqV66s7bt//z4hISGYmZlRrVo1ihYtmuG679y5w2+//ca4ceOwt7fH3t6ecePG8csvv3D37r+JPHv16kWVKlW0a7t58yZPnjwx+P5u2LCBatWqUbNmTVq0aEGhQoX4/vvvDT4e0LwnJUuWfG65yZMnc/LkySw/Pj4+AFqjKPW4fsrf6V3xKbRs2ZJ169Zx48YN4uLimD9/PomJiVr5IkWKsHHjRn766ScOHjxIu3btGDRoEJcvX86g8fTp0+zcuZPevXtn2QB9k5E9ekNITiIec5JN8mG7KGI/HPVSg+3cV6nR9IVr5baqN5YJHavnSC/bmNy+fRsHB4c062iXK1eOuLg4IiMjuXv3LkWLFk0TFFWmTJksz9egQQM+/vhjFi9ezIgRI6hbty4jR46kVi3DfldhYWFUqFDhuWWWLVuGm5sbN27cYPDgwVy5cgVnZ2cePnzIs2fP8PX1TWO8ExISNK/CnTt30ui3trbGwcEhzfmLFy+exqUbGhrKn3/+mcYlnZycrCUy8vDw4P79+0yfPp2bN2/i7u7OmDFjKF++PP7+/ixatAhfX19iYmJo3749H3/8Mba2tmnqvH37NgBly5bVvitXrpy2r0SJEpq2FAoUKACohtKQ4Ldnz56xc+dOhg8fDqiNEE9PT4KDg/Hx8cHExARzc/NMV2BLSEjQFohKuV937tyhfPnyL6z3RaTciydPnmiemZTGS8GCBTM9xsfHh2fPnjFgwABiYmJ45513qFSpEkWKFNHOWbduXUBdQtbLy4uDBw+yZ8+eDB4CExMTqlatSrVq1Rg5cuRLDWfkZfKh5XoNJCfnv4j7pHg4Mw5+bge6SNW4/weXsZVAqVKliIyMJCYmRvsuJCQEKysrHBwcKFGiBJGRkWnG7NO7utPTq1cv1q9fz5EjR6hWrRrDhg0D1NW9XkSZMmW4efOmQdorVKjAlClTmDFjBnfu3KFIkSLY2NiwcuXKND3HM2fOaL3HkiVLEhYWpp0jNjZW66GmkF6no6MjjRo1SnPOU6dOcfr0aUD1APj4+LBlyxZ+/vlnrK2tmTBhAqAaxYkTJ7J//37Wr1/PiRMnCAoKynAtKT3J1NpCQkLS7HtVfvjhB54+fcrChQu1OIDNmzcTGhrKb7/9Bqj3PyQkJINb/ObNmzg5OQHg7OyMnZ3dc93yoE67dHZ2zvKT4iavWLEiVlZWXLhwQTv24sWLWFtbU7FixUzPbWlpib+/PwcPHuTo0aMMGDCAkJAQ6tevn6WeF60ul5iYyI0bN557TW8i0tAbQnJS/jL0T6/D/qZwcRYUrARtf4fqo+Tc+P8AiYmJxMXFpfnUrl2b8uXLM3PmTGJiYrhz5w4BAQF0794dExMT6tati6OjI3PnziUuLo6QkBBWr16dZR1nz57l5MmT6HQ6LC0tsbW11Qyng4MDpqamz32Zenh4cP36dZYuXUpMTAw6ne65keENGzakdu3aLFy4EBMTE/r27cvs2bO1OqKjo/n111+1Hr2npyfLly8nJCSEuLg45s6d+8KlRbt27cr58+fZvHkzcXFxJCUlERISokVxHz16lPPnzxMfH4+1tTU2NjbaNe/evVsznAULFsTCwiLTlKMlS5akSZMmzJw5k8ePHxMVFcWsWbNo1qyZ1pt/EcePH0cIQWhoaKb7N27cSJcuXdi1axfbtm1j27Zt7N69m0aNGhEcHAxA8+bN0el0LFy4kGfPnhEfH8+BAwc4ePCgNvZta2vL+PHjWbp0KatWrdIaSmFhYcyaNUuLqZg6daoW8Z7ZJ2V6obW1NR4eHnz99dc8ePCABw8e8PXXX+Pp6YmVlVWm13Lv3j3tOiMiIhg/fjx169aladOmAJw5c4a///6bhIQE4uLi2LhxIydOnKBt27YAXL16lQMHDhAdHU1SUhIXL15k4cKFNGvWzKB7/SYhXfeGkJ8Mffxj2FsP4h5AhT5QbxFYFMptVRIj8c0332SYPvTbb78RGBjIF198QYsWLbCysqJdu3aMGjUKUHurixYt4rPPPqNhw4Y4OTnh6enJ/PnzM60jOjpaM7RmZmZUrVpVK2ttbc3w4cMZNWoUcXFxeHt7ZwjsK1myJGvXrmX27Nlaz7dWrVoZxulTM2zYMPr164e3tzfDhg1j7dq1+Pn5cfv2bWxsbKhTpw6ffvopoLp8Hz16RM+ePTEzM6Nfv36UKFHiuXOnixcvzpo1a5gzZw7z5s0jNjaWMmXK0KtXLwAePHjA559/TkREBBYWFtSqVYvPP/8cUHums2fP5tGjR9ja2tKyZUu8vb0zrefLL79k+vTpvP22Op21cePGjB8/Pktd6YmIiKB8+fKZjp1funSJc+fOMW3atDTuf1Dnkw8ePJi7d+9SokQJVq5cyVdffUWbNm2Ij4+nQoUKBAQEUKdOHe2Yd955h1KlShEUFMSCBQsA1fPQvn37TIPnXsSECRP4/PPPad++PaDOo0/xioC6uNDOnTu1qYG3b99m1KhR3LlzB1tbW20OfMqQTWhoKAEBAdy7dw8rKysqVapEYGAgb731lnbOoKAgxo0bR2JiIsWKFaNt27YMHTo029rzOiYvu0h+HidHLiopOYk6a+pQX5fElPAEelkt5si4Vjlx6tzl8jywdFAD72Qv/qXR6XQA+W5xDUPYsGEDK1euZO/evbkt5ZWJjo6mfv36rF27FhcXl9yW80r4+/vTsmVLOnTokNtSJDmMAe+bLF/mskdvCG96j/7hWbj8FTRYpkbSVxuZ24okbxgnT56kRIkSODk5oSgKQUFBeHh45Lasl+LRo0ecPXsWd3d3YmNjmT59OmXKlDE4WDAvM3v27NyWIMmDSENvCG9qMF5yMlxZBH+OgqQ4deGbMp1yW5XkDeT27duMHj2ahw8f4uDgwNtvv83gwYNzW9ZLkZSUxPz58xk+fDgWFha89dZbLF68GAuLN/D/uERiANLQG0Jy0puX0CYuEo57Q+g2sCoKDTdLIy95aTp37kznzp1zW0aO4ODgoK2CJ5H8F5CG3iCSiTd5g27VvSNw5D14FgIlWkCjb8Em63nPEolEIsm/GN16CSHeBgIAMyBIUZSZ6fY3A+YDtYHeiqJsNrbGzHijevRx9yEmHGp/ri5la5pxKo9EIpFI/hsYdR69EMIMWAh0AGoA7wkh0q+NeQvoD6wzprYXkefH6J+Fqu56gLKe0OVveGuiNPISiUTyH8fYC+bUB/5RFOWaoig6YAPgmbqAoig3FEU5CyQZWdtzydOGPnQH7K6jppVNmS5Z8H+5q0kikUgkeQJjG/oyQEiq7VD9d3mePJnQJjEWTn4Ehz0h8Zm6Xr1EIpFIJKmQS+AaiM4kjxn6qMuwtyH8vQDsa0D7P6CKr1wAR5LrBAYGakubGkLqXPaZkZIfPq/h5eXFokWLXmsdn332GVOnTtW2Q0JC6N27Ny4uLgwZMoSTJ09mO/e75L+HsQ19GOCUarus/rs8T55y3cc/hQNN4NFfUNlHNfKF33rxcZL/NOkNU1RUFH369OH9998nKipKWye9b9++aY7bvn07rVoZviKkr69vvsjp/fvvvzNgwADc3NxwdXWlU6dOBAQEZCsl7KsydepUPvvsM2176dKllCpVilOnTrFw4ULc3Ny0deWNweLFixFCsHXr1gz7hBCZakn//dOnT5k9ezbt2rXT1qb38fHh6NGjL6Xp8OHDdOrUidq1a9O5c2ctOU9WXL58mX79+lGvXj2aNGlCQEBAmlwH9+7dY8SIETRs2JB69erRt2/fNKltExMT+fLLL2nevDnOzs506dKFPXv2vJR2Y2FsQ/8HUEUIUVEIYQn0BnYYWcNLkaei7i0KQt0voUkw1F8C5jYvPkYiSUVERAR9+vTBwcGBlStXYm9vD6jZvS5fvszPP/+cywpfH4mJiVp62azYsmULvr6+NG7cmD179nDq1CkCAwOJjo5GURQjKc1ISEgIVatWTZOC91WIj483uGxSUhKbNm2icOHCL53GNTo6mvfee49Tp04xZ84cTpw4wYEDB3j33XdfyliGhIQwbNgwfHx8tPz2Q4cOzTKpz5MnTxg4cCBNmjTh6NGjrF69mq1bt7JixQqtzJQpU4iKimLv3r0cOXKEt956i8GDB2uNge+++47t27ezcuVK/vzzT4YPH87o0aO5evXqS90TY2BUQ68oSgIwFNgLXAKCFUW5IISYKoTwABBC1BNChAI9gSVCiAtZn9F4xOf2kgMP/oDDXSFBn0q00gfqSncSSTb5+++/6dWrF/Xr1ycgICBNdjATExM+/PBDvvzySxITEzM9PiEhgcDAQNq3b4+bmxu9e/fm3Llz2v4FCxbQv39/bfvevXv4+vri6upK+/bt2bRpU4YMazqdjokTJ+Lm5kbTpk3ZsGFDhnqXLl1KkyZNcHd3Z+bMmWmM1OXLl+nbty/16tWjdevWLFq0SNMfGhqKEIJNmzbRsWNH6tSpw4MHD/jhhx/o0KEDzs7ONGrUiLFjxwKqMZo+fTo+Pj54e3tTrFgxAJycnJgwYUKWrvLx48drvbyOHTuyc+dObV9UVBQfffQRDRo00LwDKb3cixcv8t577+Hq6kr9+vXp3bs3UVFRQNphDQ8PD44fP86iRYtwdnZm06ZNHD9+nBo10k5cCg4OpnPnzri6utK1a9c0PdwFCxbQt29fZs2aRaNGjTIkFHoeKRkAZ82axenTp/n7778NPjaF1atXc/fuXZYuXUrt2rWxtLTEysqKNm3aMGXKlGyfb+vWrdSsWRNPT08sLS3x8PCgRo0abNu2LdPyf/75JzqdjkGDBmFubk6lSpXo0aMH69b9O8nr5s2bvP3229jb22NpaUmPHj24ffs2Dx8+BODWrVs0aNCA//3vf5iYmNCmTRsKFy7MlStXsq3fWBjdeimKshvYne67z1L9/QeqSz9PoTPJpR59chJcmgt/TYDkRLi9H8q+mWuMS3KfM2fOsHLlSry9vbMcR+/Tpw/ffvstwcHBvPfeexn2L1iwgN9//52goCAcHR3ZsmULAwcOZN++fZpnIDWjR4+mUKFC/PLLL8TFxTFixIgMZfbu3cu8efOYOnUqBw4cYOTIkTRt2pQyZdRY3fDwcCIiIjhw4AB3795l0KBBFC5cGF9fX548ecKAAQPo06cPQUFBhISE4OPjg6WlJQMHDtTq2LVrF6tXr8be3p6EhAT8/f0JCgrC3d2dZ8+ecfHiRQBOnz7NkydP6NKlS7burYuLC/7+/hQqVIg9e/YwduxYqlevTuXKlVm+fDmxsbEcPHgQGxsbbty4gbm5+vqdOnUqTZo04dtvvyUpKYkLFy5kuhzvjh078PLywt3dHT8/P0BNS5ua4OBgli1bxoIFC6hatSq//vorw4YNY9u2bZQvXx5Q8xa0aNGCQ4cOZdmYy4zg4GCaNm1KixYtEEKwceNGLSOgoRw+fJimTZtm+jtJITw8/IV5FFIaSZcvX6ZmzZpp9tWoUSONqz01ycnJ2ifFK5KUlERoaChPnz6lYMGCeHt7s3PnTtq2bYutrS3BwcG4urri4OAAQM+ePRk9ejT//PMPFStWZP/+/SQkJFCvXj2D74OxeYOWe8td4nJjjD7mDhzrBxF7wbqUusJdqdbG1yExnH0T4cJ249RV0xPaTcvWIadOncLKyuq5y9laWloyatQovvjiiwwv3OTkZNasWcPSpUtxclLDbXr27MmaNWs4dOgQnp5pZsty+/Ztjh07xoEDByhYsCAFCxbEz8+PEydOpCnXsGFDWrdWf9vt2rXDzs6OS5cuaYbexMQEf39/rK2tKVeuHAMHDiQoKAhfX18OHTqEhYUFfn5+mJiYUKlSJQYNGsTKlSvTGPqhQ4dq6VkTExMxNzfn2rVrVK9encKFC2s99ZTc6pmlen0ePXv+62Hr1KkTy5cv5/jx41SuXBkLCwsePXrE9evXqVGjBhUrVtTKWlhYEBERQUREBGXLlqVu3brZqjc1a9asYciQIVqa2ObNm9OgQQN++OEHrXHg6OjIgAEDsnXeO3fucOjQIQICAgDo0aMHCxYsYMyYMVhbWxt8nsjISFxdXZ9bxtHR0eC4g+joaOzs7NJ8V6hQIf75559My9etWxdTU1OWLFnCgAEDuHHjBt9//z2AZuhdXV3Ztm0b7u7umJmZUapUKZYtW6adw8nJCTc3Nzp37oypqSmWlpbMnj2bokWLGqQ5N5BR9wZi9Ol1tw/Cj7VVI+/YETqelUZe8sp88MEHNG3alD59+nDt2rUsy3Xs2JEyZcqkecEBPHz4kGfPnuHr64ubm5v2CQkJ4c6dOxnOk/Jd6dKlte8cHR0zlEufH93Gxobo6Ghtu2jRohQoUEDbLlOmDLdv3wbUeANHR8c049blypXT9qc+JoUCBQqwdOlSfv31V9q2bUv37t01V3tKzy2z68mKpKQkAgICaN++Pa6urri5uaEoiubu9fb2xt3dnXHjxuHu7s7YsWO5f/8+ANOnTycpKYn333+fVq1aMX/+fBISEgyuOzWhoaFMmTIlzbM5fvx4mmvJ7P6/iM2bN2Nvb0+LFi0AdRghNjaW3bv/dc6am5tn0J0yvJLioXBwcMjWfX0Rtra2GYIjHz9+TMGCBTMtX7hwYZYsWcKvv/5K06ZNGTt2LO+88w6mpqYUKlSIpKQk+vfvT4UKFTh16hRnzpzhww8/pE+fPtrzmjJlCpcuXeKnn37i/PnzrFixgkmTJr0wCDA3kT16AzF6MJ6JCcQ/BpevQAwHE9kmeyNoNy3bvWxjYmpqysyZM5kyZQr/93//x4oVK7TeX3rGjh2Lt7e31hMEKFKkCDY2NqxcuZLatWu/sL6UXnFERITmAQgPD8+27gcPHhATE6MZ+7CwMEqVKgWojYjw8PA07tiQkBBtfwqmpmn/DzVo0IAGDRqQmJjIwYMHGTZsGHXq1MHZ2Rk7Ozt27drFkCFDDNK3a9cuNm3axIoVK6hcuTKmpqZ0795dC+CysbFh5MiRjBw5knv37jFmzBhmz57N7NmzcXJyYsaMGQAoioK3tzdly5alR48e2b5Pjo6ODBs27Ln56NPfhxeRlJTE999/z+PHj2nevHma7zdu3Ej37t0BtSF169YtGjZsqJW5desWgPbsmzVrxurVq4mKisrSfR8eHk6nTs9PwHX69GkAqlWrlmH44tKlS7i7u2d5bN26dfnuu++07VmzZlGrVi1sbGyIjIwkNDQULy8vrbHQs2dP5syZw5kzZ2jTpg0XLlzg/fff1xqOLi4uuLm58csvv9CkSZPn6s4tpPUwEJ0xkto8vQbP9C/Bki3B44aaO14aeUkOYmJiwuTJk/H09KRv376cPXs203Kurq40bdo0TUSyiYkJffv2Zfbs2dy4cQNQ3acpgVrpKVWqFPXr12fOnDk8ffqUyMhIFi9enG3NycnJzJkzh9jYWEJCQli+fDldu3YFoEWLFuh0OgIDA9HpdFy7do1ly5Y911Dev3+fvXv38uTJE8zMzDT3r6mpKba2towfP56lS5eyatUqzZUfFhbGrFmzMnUrP336FHNzcxwcHEhKSmLz5s1povMPHjzI1atXSUxMxMbGBisrK8zM1OWpt27dqt27QoUKYWZmpu3LLv379+ebb77h0qVLJCcnExsby8mTJ18YEe7l5cW4ceMy3ffrr78SERHBhg0b2LZtm/YJDAzkzJkz2nV269aN5cuXc/nyZZKTk7l79y6zZs2iefPmmpekb9++lChRgsGDB3Pu3Dni4+PR6XQcOnSIyZMnA2pj5fTp08/9pNC1a1fOnz/Prl27iI+PZ9euXVy4cEH7bWTGhQsXiIuLQ6fT8eOPPxIcHMzIkSMB1eNQoUIF1q1bx7Nnz0hISGDz5s1ER0cjhABUw75z507tmf3111+cOHEiQ6xAXkL26A3ktffob6yHE4OhqBu0OqAa9wLZGyOUSLLD2LFjsbW1pX///ixZsiTTMqNHj6ZTp07Y2Pw7hXPYsGGsXbsWPz8/bt++jY2NDXXq1MkyMGvu3Ll8+umnNG/enGLFitGvXz+OHz+OpaXh/6ccHR0pWbIkrVu3JjExkS5dumjj73Z2dqxYsYLp06ezYsUK7Ozs6N69Ox988EGW50tKSuK7775j4sSJJCQkULp0aWbOnEnZsmoc8DvvvEOpUqUICgpiwYIFgNpoad++faYekG7dunHs2DHatm1LgQIF8PT0TDMWfevWLWbMmMG9e/ewsrKiQYMGjB49GoBjx44xd+5cnj59SqFChejSpUuGWAdDeffdd7GwsGD8+PGEhoZibm5OzZo18ff3f+5x4eHhdOvWLdN9GzZsoE2bNrz1Vtq1OooXL46zszMbN27ks88+Y9CgQZiZmTFixAju3r2Lvb09zZo104woQMGCBVm/fj2LFi3SvBuFChWiRo0a2Y4bAHWIZsGCBcyaNYsJEybg5OTEN998oz1HAGdnZ6ZMmaLFm2zYsIE9e/YQHx9PlSpVCAgISOMBWLRoEbNnz6Zly5YkJCRQvnx55s+fr3kl/P39mT17Nj169ODp06cUK1aMDz744LmNi9zGJPVCAfmIHLmopOQk6qypQ/2YWIo+GMgRs3ocGWf4wiEGkRCtLmN7bQWY24LbIvhf3xcfJ8l1dDodQLYMlkTtIfr5+XH27Nkcmw8ueXlCQkLw8/Nj+/bt2XbrS4yHAe+bLP8zyR69gbyWlfEe/gVHesPjy1DEGRpvgEJVc74eiSQXuXTpEiYmJtrc+fnz59OxY0dp5PMITk5Oaeb8S/If0tAbSI6vdZ/wDA62hbh7IEZA3ZlgZvXi4ySSN4yoqCg+/fRT7t27R8GCBWnWrFmW48ESiSTnkYbeQHK8R29uA/UWg5k1lHl+hKlE8ibTsGFD9u/fn9syJJL/LHJAxkByZB793cNwoIU6bQ6g3DvSyEskEonktSINvYG8UtR9UgKcnQzVdSJ4AAAgAElEQVQ/tYR7v6mL4UgkEolEYgSk695AQp4kQtbLM2dNdAj83gfu/Qo25aDxOijeOMf1SSQSiUSSGdLQG0hcsgXdapd+ccHUhO+F398HXSQ4vQMNloFlkdcjUCKRSCSSTJCG3kCK2tsxoWP17B1k5QDJCVAvECr7qMvaSiQSiURiRKShNxCDs9dFXQYzSyj4PyhaDzxvgmXh1ytOIpFIJJIskMF4BvLCqPvkZLi6Ava4wm+9IEmf51kaecl/mJMnT2prhEuyZty4cXzyySevtY7AwEB8fX217UePHuHt7Y2rqyvdu3cnPDwcZ2fnHM0uJ8kbSENvAEmYkPy8xDK6KHUs/rg3mFpAjbFg+nJJKSSS14WXlxdvvfUWzs7OuLq64unpyY8//pjbsl6ZVq1aUatWLZydnbXPwoULc0WLECLTpDfnzp3Dz8+Phg0b4uLiQvv27fniiy+4e/eu0bT5+voSGBiobW/YsIHo6GiOHz/Oli1btGQyKRkHXzc7duxACME333yTYV+rVq3Yvn37C7/X6XQsXryYTp06UbduXRo3boyXlxd79ux5KU3nzp2jR48e1KlThzZt2mSqITVhYWH4+flpmRAnT56sLVUL8OzZMz799FMaN26Mq6srPXv25NixY2nOsX79etq3b4+zszNdu3bNkI0vJ5CG3gCSsl5CGO6fgB+d4eYGKOYOHc5Aueynl5RIjIGfnx+nT5/m+PHjdOvWjdGjR3Pz5s3clvXKTJs2LU12M0PTy6YmMTGRpKSkHNd25MgR3n//fSpWrMj27dv5888/Wbt2LYULF+aPP/7I8foMJSQkhEqVKmFunjMjuCm55w1l48aNFC5cmM2bN5OYmJjt+hITExk8eDA7duxg4sSJHDt2jF9++QU/Pz/27duX7fM9efKEQYMG0a5dO/744w+mTJnC5MmT02TLS1+/r68vpUuX5vDhw2zfvp0zZ84wc+ZMrUxAQABnzpxhy5Yt/PHHH3h4ePDhhx8SFRUFwI8//khAQADz58/n5MmT9OrVi8GDB79UKufnIQ29ASRjQqfMIu4T4+DXbhB9A2pOgDa/QMEKxpYnkWQbc3Nz3n33XRISErh06ZL2/fjx42nevDnOzs507NgxzRrox48fp0aNGuzevZs2bdrg6urK8OHDefr0qVbmxo0beHl54ezsjIeHB+fPn09Tb0xMDNOmTaN58+Y0aNAAPz+/NC81Ly8vZsyYwZAhQ3B2dqZNmzYcPXqU33//nc6dO+Pi4sKQIUPS1Pk8Ll++TN++falXrx6tW7dm0aJFmlEJDQ1FCMGmTZvo2LEjderU4cGDBzx8+JAJEybQvHlzGjZsyPDhw7l//752zjVr1tCqVSucnZ1p2rQpX331FYCWHc3b2xtnZ2fNFT9lyhQ6d+7MmDFjtN5yiRIlGDJkSJZ517/66itat26t3YNVq1Zp+3Q6HZ9++inu7u64uLjQrl07zTMTGhqKt7c3bm5u1KtXj27dunHt2jUAFixYQP/+/QG1d5+SbtbZ2Zmvv/5aux+3b9/W6jpw4ADdu3fHzc2NDh06sGPHDm3fli1baNu2LUFBQTRr1ixb2duuXr3KyZMnmTlzJvfu3ePw4cMGH5vCrl27+OOPP1i8eDHu7u5YW1tjbm6Ou7u79kyyw759+yhQoACDBg3C0tKSxo0b06ZNG4KDgzMtf/36df7++29GjBiBlZUVpUqVol+/fmzZsoW4uDgAbt68ScuWLSlZsiSmpqa8++67PHv2jFu3bgGwZ88ePDw8qF69OmZmZrz33ns4ODiwZcuWbOt/HjIYzwCSMU0bcZ+crEbQm1lBw5VgYg6lcjirneSNZO7Juey7kf3exMvQrkI7RrmNeqljdTod69evB6BChQra9y4uLvj7+1OoUCH27NnD2LFjqV69OpUrVwbUXsxvv/3G9u3biYmJ4f3332ft2rV8+OGHJCQkMHjwYBo1asSyZcu4fft2mjFhgBkzZnDp0iWCg4Oxs7Pjiy++wNfXl61bt2o52Ldv386SJUv4+uuvCQgIYMyYMbi6uvLtt98CpKnzeTx58oQBAwbQp08fgoKCCAkJwcfHB0tLSy3FLagGY/Xq1djb22NmZoaXlxcVK1Zk165dmJub8/nnnzNq1ChWr17N9evXmTt3Lps3b6ZKlSo8fvxYM6Qprujly5fj5uYGqMbg5s2bWq51Q6lUqRLr1q2jRIkSHDt2jMGDB1OpUiWaNm3K1q1bOXfuHLt376ZIkSJEREQQHR0NwLx58yhdujSLFy/GzMyMK1euYG+fcQGQwMBAxo0bh5mZGV988QWgNhJSc+TIET755BMWLlyIi4sL58+fx9vbm9KlS1OvXj1AdV3fvXuXffv2kZ1MqBs3bkQIQcuWLWnWrBkbN26kZcuW2bpHhw8fplatWml+v5mR8iyyYseOHTg6OnL58mWqV6+eJtlSzZo1s3Tfp1xv6utOSkoiJiaG69evU61aNfr27cu8efMIDw+nRIkSrF+/nvLly1O1alXt2Mzu2+XLl5+rObvIHr0BJKd23Ufsgz0uEHtP3S7dThp5yRtDYGAgbm5u1KlTh/nz5zNt2rQ0+dV79uxJkSJFMDMzo1OnTlStWjXDmOHo0aOxtbWlWLFitG7dWuu1//XXX4SFheHv74+1tTUVKlRIk2M8KSmJrVu3MmLECEqWLImNjQ0TJkzg2rVrnD17VivXoUMH6tSpg5mZGR4eHty7dw9vb28KFy5M4cKFadGiRQZPwaRJk3Bzc9M+d+7c4dChQ1hYWODn54elpSWVKlVi0KBBbNq0Kc2xQ4cOpXjx4lhaWnLx4kUuXLjApEmTsLOzo0CBAowZM4Zjx45x+/ZtzM3NSU5O5sqVK0RHR1OoUCHq1q2b5f2OjIwEyPa4t6enJyVLlsTExAR3d3datGjB0aNHAbCwsODZs2dcvXqVhIQESpcurTXELCwsuH//PiEhIZiZmVGtWjWKFi2arbpTWLNmDV5eXri5uWFqakrt2rXx8PBg27ZtWhlzc3NGjx6NtbU1BQoUMOi8cXFxbN++ne7duwPQo0cPDh8+nMaTYAiRkZEG3deTJ08+9+Po6AhAdHQ0dnZ2aY61s7PL0ntUsWJFypcvz1dffUVMTAxhYWGsWbMGQDumWrVqlClThpYtW1K7dm0WL17MjBkzsLJSE5i1bNmSHTt2cO7cOeLj4/n2228JDw832GNlKEbv0Qsh3gYCADMgSFGUmen2WwFrAFfgAdBLUZQbxtaZmiRMIFEHZyfCpS/VgLv7v0NZz9yUJcmDjHIb9dK9bGPg6+uLn58fUVFRfPLJJxw/fpyePXsCqiFesGABu3fv5v79+5iYmBATE8PDhw+1483MzHBwcNC2bWxstN7knTt3KFq0aJoXftmyZbW/IyMj0el0ab6ztbXFwcGBiIgInJ2dAShevLi239raGlBd3SkUKFBAqzOFKVOm4OmZ9v9jREQEjo6OaXpo5cqVy2BQypQpo/0dGhqKTqejUaNGacpYWVkRHh6Oi4sLc+bMYf369UycOBEhBEOGDKFJkyZkRsq9unPnDpUqVcq0TGasWbOGTZs2cfv2bZKTk4mNjaVLly6AOkRw//59pk+fzs2bN3F3d2fMmDGUL18ef39/Fi1ahK+vLzExMbRv356PP/4YW1tbg+tOfS+OHz+eZtggMTExTQ85pYGUHX788Ueio6O1oY7mzZvj4ODApk2bGDZsGKA2IBISEjIcGx8fj4WFOgPKwcEhR8eybW1tCQsLS/PdkydPKFiwYKblzc3NNcPdqlUr7O3t6dGjB5cvX6ZIEXVhtI8++gh7e3t+//137O3tOXz4MIMHD2b9+vVUqVKFrl27cu/ePUaPHs2jR49o3bo1jRo1ytQL8yoY1dALIcyAhUBbIBT4QwixQ1GUi6mKeQMPFUWpLIToDcwCehlTZ3qKmSTAgabw4ATYVVHzxju45KYkieSVsLe3Z9q0abRt25YDBw7Qpk0bdu3axaZNm1ixYgWVK1fG1NSU7t27G+ySLVGiBA8ePCAmJkYz9qlfnA4ODlhaWhIWFkb58uUBtRcVGRlJ6dLZXHXSAEqXLk14eDjJycmasQ8JCaFUqVJpypma/uvYLFOmDAUKFODEiRNpvk9Nu3btaNeuHTqdjg0bNuDn58fx48cpUKBAmkYF/Nvr27VrV4bGQ1acOnWKOXPmsGrVKs2z8dFHH2nPwdzcHB8fH3x8fHj8+DFTp05lwoQJfPfddzg4ODBx4kQmTpxISEgIfn5+BAUFMXz4cIPvWwqOjo5069YtzTBHerK6R88jODiYpKQkreEC8PjxYzZv3oyfnx9mZmaUKVNGG8dOITo6mvv372sNxWbNmvHJJ59w8+ZN7feUGSkNyKz44YcfcHR0pFq1avz0009p9l28eDGNxys9lSpVIigoSNv+7rvvKFGiBBUrVgTgwoULzJ8/X/OqtGrVinLlynHkyBGqVKmCiYmJ9ixBHVJr3bo1fn5+z9WcXYztuq8P/KMoyjVFUXTABiB9t9gTWK3/ezPQWgiRa0vKNTF5wsyCV1QjX8EL3j4ljbwkX1C4cGH69+/PvHnzSEpK4unTp5ibm+Pg4EBSUhKbN29GURSDz1e3bl3KlCnDnDlziI2N5datW6xcuVLbb2pqSteuXQkICODOnTvExMQwa9YsKlasSO3atXP8+lq0aIFOpyMwMBCdTse1a9dYtmwZPXpkPSvmrbfeolq1akybNk3zZERGRvLDDz8AcO3aNQ4fPkxMTAwWFhbY2dlhYmKiGbxixYplmMUwadIkdu7cyVdffaXNUb9//z5LlizRzpuap0+fap4TExMTDh06lCZY7ejRo5w/f574+Hisra2xsbHR6t+9ezchISEkJydTsGBBLCwstNiH7NKvXz9WrVrFyZMnSUxMRKfTcf78ec6dO/fc48aNG4eXl1em+/755x9OnTrFN998owUDbtu2jU2bNnH//n1++eUXALp168bGjRu1uqOiovjiiy+oWrUqNWrUAKBz5864ublpDa24uDgSExM5ceIEo0b961VLPRsjs0+K675t27Y8e/aMoKAgdDodR48eZf/+/bz77rtZXquiKERHR5OQkMDvv//OwoULGTlypPY8XFxc2Lx5M48ePSIpKYlDhw5x5coVatasCageg6tXr5KcnExkZCSTJ0/Gzs6Obt26GfiUDMPYrvsyQEiq7VCgQVZlFEVJEEJEAUWB+xgZk2SITDIn3tQMGq2Cipn/eCWSN5V+/fqxevVqtm3bRrdu3Th27Bht27alQIECeHp64urqavC5zM3NWbRoEZMmTcLd3R0nJyfeffddZsyYoZUZP348c+fOpUePHuh0OpydnbXAsZzGzs6OFStWMH36dFasWIGdnR3du3fngw8+yPIYU1NTFi1aREBAAO+88w4PHz6kaNGiNGrUiE6dOhEfH8+iRYu4cuUKAOXLl+frr7/WxlxHjhzJ119/zcyZM+nQoQNTp06lcePGrFu3jsWLF+Ph4YFOp6N48eK0aNFCG6dOTdOmTfH09KRnz56YmJjQunVr2rRpo+1/8OABn3/+OREREVhYWFCrVi0+//xzQO2Bzp49m0ePHmFra0vLli3x9vZ+qfvXpEkTpk2bxuzZs7l+/TqmpqZUrlyZjz766LnHhYeHU79+/Uz3bdiwgZo1a9KqVdq4puLFi/P222+zceNGWrVqhYeHB7GxsUyZMoXw8HBsbGyoX78+gYGB2nRAMzMzli5dyvLly5kyZQphYWEULFiQSpUqZdnQeB6FChVi6dKlTJ06la+//prixYszefLkNB6BTp060aVLFy3IdP/+/axdu5bY2FjKlSvH+PHj03gqpk+fzqxZs+jYsSOxsbGULl2azz77TAtmfPr0KcOHDycsLAwLCwuaN2/OmjVrtCGrnMIkO5GSr4oQogfwtqIoA/XbXkADRVGGpipzXl8mVL99VV8mO4Y+xy5K+XsXDnalKV7a8Bee5L9BysIY2R2jlEjyK7GxsXTo0IFdu3a9VFyAJGsMeN9k6fk2do8+DHBKtV1W/11mZUKFEOaoyWEfGEdeRkTVzrlVtUQikbxRWFtb8/PPP+e2DEk6jG3o/wCqCCEqohr03sD76crsAPoBR4EewEFFUYzndpBIJBKJJB9h1GA8RVESgKHAXuASEKwoygUhxFQhhIe+2HKgqBDiH+BjYJwxNUokEolEkp8w6hi9EcmXFyXJW8gxeolEYizi4uIwMTF5qTF6uTKeRPKSmJqaZrqoh0QikeQ0SUlJGdZpMBRp6CWSl8TMzAydTpetNb4lEonkZYiPj3/pTIPSdS+RvAKJiYk8efIES0tLzMzMXrrFLZFIJJmRnJxMfHw8pqam2NjYPK9oli8faeglklckOTn5teUyl0gkEjMzM0MWlZKGXiKRSCSSfIwMxpNIJBKJ5L+INPQSiUQikeRjpKGXSCQSiSQfIw29RCKRSCT5GGOvdW8s5BwniUQikUiQPXqJRCKRSPI10tBLJBKJRJKPkYZeIpFIJJJ8jDT0EolEIpHkY6Shl0gkEokkHyMNvUQikUgk+Rhp6CUSiUQiycdIQy+RSCQSST5GGnqJRCKRSPIx0tBLJBKJRJKPya9L4L4UQoi3gQDADAhSFGVmuv1WwBrAFXgA9FIU5UYe0tcMmA/UBnorirLZWNqyofFjYCCQANwDBiiKcjMP6fMFhgCJwFPAR1GUi8bSZ4jGVOXeATYD9RRFOZlX9Akh+gNfAmH6r75RFCXIWPoM0agv8y4wGUgG/lIU5f28pFEIMQ9oqd+0AUooilI4D+krB6wGCuvLjFMUZbex9BmosTywAigORAL/pyhKqBH1rQA6A3cVRXkrk/0mqPo7As+A/oqi/JnTOmSPXo8QwgxYCHQAagDvCSFqpCvmDTxUFKUyMA+Ylcf03QL6A+uMpSs1Bmo8DbgpilIb1UjNzmP61imKUktRlLp6bV8ZS182NCKEsAOGA8fzoj5go6IodfUfYxv5F2oUQlQBxgONFUWpCYzIaxoVRRmZcg+BBcCWvKQPmAgEK4riDPQGFhlLXzY0zgHW6N83U4EZxtQIrALefs7+DkAV/ccHWPw6REhD/y/1gX8URbmmKIoO2AB4pivjidqCBdVItda3yPKEPkVRbiiKchZIMpKm9Bii8WdFUZ7pN48BZfOYvsepNm1Re3vGxJDfIcDnqA3NWGOKw3B9uYkhGgcBCxVFeQigKMrdPKgxNe8B642iTMUQfclAIf3f9kC4EfWBYRprAAf1f/+cyf7XiqIoh1E9CVnhidoQSVYU5RhQWAhROqd1SEP/L2WAkFTbofrvMi2jKEoCEAUUNYo6w/TlNtnV6A38+FoVpcUgfUKIIUKIq6g9+o+MpC2FF2oUQrgAToqi/GBMYXoMfcbvCCHOCiE2CyGcjCNNwxCNVYGqQogjQohjehewMTH4/4re/VyRfw2WMTBE32Tg/4QQocBuYJhxpGkYovEvoLv+726AnRDCWO9sQzDKe10aekmuIIT4P8ANdSw3T6EoykJFUSoBY1Hdk3kGIYQp6nDCqNzW8hx2AhX07tL9/OsFy0uYo7pLW6D2lpcJIYw2/p1NegObFUVJzG0h6XgPWKUoSlnUMea1+t9nXmI00FwIcRpojho3ktfu42snrz2U3CQMSN3zKMu/wUQZygghzFHdVQ+Mos4wfbmNQRqFEG2ATwAPRVHijKQNsn8PNwBdX6uijLxIox3wFnBICHEDaAjsEEK45RF9KIryINVzDUINXjUmhjznUGCHoijxiqJcB/5GNfzGIju/xd4Y120PhunzBoIBFEU5ClgDxYyiTsWQ32K4oijd9XEEn+i/e2Q8iS/EKO91GXX/L38AVYQQFVFvdG8gfRTuDqAfcBToARxUFMVYY7iG6MttXqhRCOEMLAHezoVxUUP0VVEU5Yp+sxNwBePyXI2KokSR6mUqhDgEjDZi1L0h97C0oigR+k0P4JKRtKVgyP+Vbag90pVCiGKorvxreUwjQohqQBHUd44xMUTfLaA1sEoIUR3V0N/LSxr1zzZSUZQk1ODLFUbUZwg7gKFCiA1AAyAq1f+dHEP26PXox9yHAntRX0zBiqJcEEJMFUJ46IstB4oKIf4BPgbG5SV9Qoh6+vGynsASIcQFY+kzVCOqq74gsEkIcUYIsSOP6RsqhLgghDiD+oz7GUtfNjTmGgbq+0h/D/9CjXHonwc17gUeCCEuogZpjVEUxVjeuew8597ABiN2KLKjbxQwSP+c16NODTOaTgM1tgAUIcTfQEngC2PpAxBCrEdtpAkhRKgQwlsI4aufxgtqbMM14B9gGeD3OnSYJCcbO6hYIpFIJBKJsZA9eolEIpFI8jHS0EskEolEko+Rhl4ikUgkknyMNPQSiUQikeRjpKGXSCQSiSQfI+fRSyQviRBiMjApk10/KYrSxoDjKwDXgS6KouzKWXUZ6roBlNdvxqNO6fkOmKVfJzwn6ugPrATsFEV5KoQogTpdaFXqLI9CiBaoU9pqKYpyPifqfoGu1FOLYoEbqMlG5uqnaGXnXP7ACUVRDuWUPonkdSMNvUTyakSRMTtVVG4IMYB1qFnQrFDTn05CXd1xdA6d/wfAHTXdJkAJfR2HUI1rCn/qy13NoXoNYS5qIqoCqGlDZwIWwLRsnscf+Ab1miSSNwJp6CWSVyNBn3XqTSAildZfhBBlAV8hxJicWOhEUZR7GLAymj5DoLHv2Y1U1/6zEKIm0JfsG3qJ5I1DGnqJ5DWgTzX5BerKXKVRM1QFA1Of5yrXr+g1CagG6FDXYPdXFOUX/X5T1F7lQNQ1sm8CXyiK8jKJY06hrldeDLgnhGiFmq+7DqpX4nt93U/1dVvo97+LusrYA+A40EtRFF1q173+nOf09fwshABAURST9K57/TK+9xRF6ZnuXnwJ9ALKK4qSLISwRs0p/h6qt+AyMF5RlN0vce1/oSY5SV3fTNRljysCj4BfgFGKotzW77+Bmq1ykhAiZcimpaIoh3L4uUgkOYo09BLJK6JPcJSaRFRDF4m6jO5D1LXUJwPFgcFZnKcSqns5ABiDuna4K+CQqtgC1GV5p6K6wNsCK4QQD15inL8CamMiUt/D3YOabe4dVGM1E/gf/w5NjAf6oC79fB0ohZq1zCyTc0foy34HDNFrzYqNwBwhhK2iKNEAQggT1AZFcCpvw2bUHOSTUN3+76JP6KMoyplsXns5/TWkpgQwHTWvenHUJV4PCiHe0q+V3g21gbIZNVkPwEX9vzn5XCSSHEUaeonk1SiKGtyWmraKohwg1di3EOIIEI368h+WRa/eGXiiKMqYVN9pvVUhRGXgQ+CDVD3FA3rvwSTgRQbFRN8osQRaAb7ATkVREoUQn6L2Qj1S0qEKISKBjUIId312svrAunS91ODMKlIUJU4IcVa/efEFwxubUQ1lF9SMgaBm5SuXsi2EaI3a226R4t0A9gkhqqJmJUvjDcgEU/21p4zRdyddHgNFUQak/C2EMENdozwUaAIcVhTltBAiAQhNfT058FwkkteKNPQSyasRBaSPsFf0PdLhgA+qK9g61f5yqEks0nMOsBdCrEbtCR9J6eHqaQ0kAVvTeRF+At4TQpi9IGf5x/pPCrtQe9ugGvH0Oc+/BxJQDd1R4AzwoRDiDmrv/1xOje0LIQ6iuulTDH0v4GqqrHxtgNvAkUyuvb8B1QToPynMUxRlQ+oCQogOwKdATaBQql1VgcPPOferPheJ5LUiDb1E8mokZJYiVggxEjVT3yzUsd6HQD1gIWmNvoaiKIoQwhPVNb4biBdCbAWG6wPdiqG6ybOK6i+N2gPNim9RjV0canDak3TH3kmnJ1EI8YB/hw6moRo0P/11hQkhvlQUJbUBfVk2AIuEEIWAp6g99FWp9hdDHSpI7z0BdajkRXyJ6n2wB0YAI4UQB1LG94UQ9VBThm5FHbK4CySjBg1m+rzSaXuV5yKRvFakoZdIXg89UXvIn6R8IYSo8aKDFEX5AfhBCGGP6qqej+rW7o065p8ANEY1uOm5+4LT33lO3voI1DFqDb37uqi+XhRFiQU+Az4TQlRBdf3PF0IoiqLsedG1vYCtwGLAE3UIwRF17D6FSNSc411f8vy3Uq5dCHEY1XvypRDiR71XohvqjIFeKV4KIUT5LM+Wlld9LhLJa0Uaeonk9VAAteecmj6GHqwoShSwTgjRHHXOOcBB1J6jvaIo+3NE5b8cB7oJISakcjN3R31H/JaJvitCiNGorv8aqK789KTEIbyoR4yiKA+FEPtQXfY3gUuKopxNVeQn1OC4p4qiXDbwmrKqK14fkxCMGhewA/V5xacbisjseenIeD2v87lIJK+MNPQSyethP/CREOI4aoR4H6Dy8w4QQgxGNep7UCO/q6B6BtaA5toPBDYIIWYDJ1GNTk2gqqIoA19B7zTgNLBNCLEYKIvqnt+rD8RDP4xwSl8uBuiB+g7Javz6lr5cPyFEFKohzcqjAGoPfgWqC/ybdPv2A3uB/UKIWcAF1HH0uoC1oijjs3e5fI86PW8MqqHfD4wQQswHdgKNgP/L5LjLQCchxB7UIQblNT8XieSVkWvdSySvh6nAelQDuh61J/jRC445izqt6ytgHzARWAaMTVVmCPA56mIvu1HHsTvx/GCxF6IoygWgA6r7fksq3T1SFfsd1XW+DtiOOvXvnayMt97VP0hf7hfgjxfI2I7qAi/Gv0F5KedKRvUwrEAdY98LLEFtGGXwOLwI/XS5GUATIURD/Vj9WNSphTtQ59h3zuTQMaizJ37QX4+r/vvX8lwkkpzAJDn5lYNmJRKJRCKR5FFkj14ikUgkknyMNPQSiUQikeRjpKGXSCQSiSQfIw29RCKRSCT5GGnoJRKJRCLJx0hDL5FIJBJJPkYaeolEIpFI8jHS0HIEkLsAAAAXSURBVEskEolEko+Rhl4ikUgkknzM/wMXqjQvNpzhJAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "h6dcF-m816Gw"
      },
      "source": [
        "##### Random Forest Wins (For Now...)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "omHe7ix8NYAK"
      },
      "source": [
        "Compared to the previous ROC curve analysis, the resulting AUC scores for the LR and RF classifiers have improved slightly. Particularly, the AUC of Logistic Regression model has grown 0.032 and the curve has become closer to (0,1), which means after hyperparameter tuning the performance of the LR model has improved the most among the three compared classifiers."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EU4u-wVb16Gx"
      },
      "source": [
        "#### Conclusions and Suggestions\n",
        "\n",
        "From our analysis, we can conclude that the bank should consider targeting the following client profile when conducting telemarketing campaign:\n",
        "\n",
        "- Aged between 25 and 45 and;\n",
        "- Working as blue-collar, management, or unemployed.\n",
        "\n",
        "We also suggest that the bank should focus more of its energy on the Summer and Fall seasons because the target group is more likely to say \"Yes\" during those months. However, more data collected from different years should be analyzed to check if this seasonal effect is consistent across time.\n",
        "\n",
        "After stratified sampling and resampling, we populated the imbalanced dataset so the classification algorithms would work better with higher accuracy. We've looked at several algorithms, but the Random Forest and KNN seem to outperform the others. The bank should consider using our models with the suggested parameters to predict the likelihood of a client submitting a term deposit, while informing future telemarketing decisions.\n",
        "\n",
        "Based on the results of the ROC curve analysis, the best classifier with the highest AUC score is RF, followed by KNN. But, the RF’s score is just 0.001 higher than the KNN’s. Meanwhile, the KNN’s F1 score is 0.04 higher than the RF’s, which is also not a huge difference between these two classifiers. Consequently, other factors should be used to determine the best model between the classifiers. In the 7. Discussion section, we will discuss further about the potential factors that may help us to choose the best classifier model. \n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "szxJSGGR3Wzj"
      },
      "source": [
        "## 7. **Discussion**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vwKYrZFskAmI"
      },
      "source": [
        "In section 6.1, the GridSearchCV renders the hyperparaemter of K = 1, for our KNN model. We believe this is the behavior of either overfitting or the resampling method perfectly sperate the two predicting classes. To dig out the reason behind the unnormal K value, we tried to use PCA to visualize the distribution of two predicting classes."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XiIoI-mG-WYY",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 526
        },
        "outputId": "03d04d5d-3baf-4f8c-eceb-b40203ab5f36"
      },
      "source": [
        "bm_x_pca = StandardScaler().fit_transform(bm.drop(['response'], axis=1))\n",
        "pca = PCA(n_components=2)\n",
        "principalComponents = pca.fit_transform(bm_x_pca)\n",
        "principalDf = pd.DataFrame(data = principalComponents\n",
        "             , columns = ['principal component 1', 'principal component 2'])\n",
        "final_pca_df = pd.concat([principalDf, bm[[\"response\"]]], axis = 1)\n",
        "\n",
        "fig = plt.figure(figsize = (8,8))\n",
        "ax = fig.add_subplot(1,1,1) \n",
        "ax.set_xlabel('Principal Component 1', fontsize = 15)\n",
        "ax.set_ylabel('Principal Component 2', fontsize = 15)\n",
        "ax.set_title('PCA Before Resampling', fontsize = 20)\n",
        "\n",
        "\n",
        "targets = [0, 1]\n",
        "colors = ['r', 'g']\n",
        "for target, color in zip(targets,colors):\n",
        "    indicesToKeep = final_pca_df['response'] == target\n",
        "    ax.scatter(final_pca_df.loc[indicesToKeep, 'principal component 1']\n",
        "               , final_pca_df.loc[indicesToKeep, 'principal component 2']\n",
        "               , c = color\n",
        "               , s = 50\n",
        "               , alpha = 0.3)\n",
        "ax.legend(targets)\n",
        "ax.grid()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "X2SHPICn_wKX",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a7d135c9-51dd-48f1-a383-597a591f78ac"
      },
      "source": [
        "pca.explained_variance_ratio_"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.15896604, 0.10772808])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zP84JQYrSA0L"
      },
      "source": [
        "Together, the first two principal components contain around 26% of the information in the dataset."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZYeXSIF_y0BM",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 526
        },
        "outputId": "a1690c66-bb0e-4de1-a75d-e0fe29f0f5a3"
      },
      "source": [
        "train_df = pd.concat([X_train, y_train], axis = 1)\n",
        "train_df.reset_index(drop=True, inplace=True)\n",
        "X_train_pca = StandardScaler().fit_transform(train_df.drop(['response'], axis=1))\n",
        "\n",
        "pca = PCA(n_components=2)\n",
        "principalComponents = pca.fit_transform(X_train_pca)\n",
        "principalDf = pd.DataFrame(data = principalComponents\n",
        "             , columns = ['principal component 1', 'principal component 2'])\n",
        "final_pca_df = pd.concat([principalDf, train_df[[\"response\"]]], axis = 1)\n",
        "\n",
        "fig = plt.figure(figsize = (8,8))\n",
        "ax = fig.add_subplot(1,1,1) \n",
        "ax.set_xlabel('Principal Component 1', fontsize = 15)\n",
        "ax.set_ylabel('Principal Component 2', fontsize = 15)\n",
        "ax.set_title('PCA After Resampling', fontsize = 20)\n",
        "\n",
        "\n",
        "targets = [0, 1]\n",
        "colors = ['r', 'g']\n",
        "for target, color in zip(targets,colors):\n",
        "    indicesToKeep = final_pca_df['response'] == target\n",
        "    ax.scatter(final_pca_df.loc[indicesToKeep, 'principal component 1']\n",
        "               , final_pca_df.loc[indicesToKeep, 'principal component 2']\n",
        "               , c = color\n",
        "               , s = 50\n",
        "               , alpha = 0.3)\n",
        "ax.legend(targets)\n",
        "ax.grid()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DItml9DH1KMO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "89aa07c5-168e-4166-f0af-0a1db9efe58e"
      },
      "source": [
        "pca.explained_variance_ratio_"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.17372207, 0.11289552])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EsdLDT1fSJZ3"
      },
      "source": [
        "Together, the two principal components for after-resampling dataset contain around 28% of the information in the dataset."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xfMvnxN9SNKA"
      },
      "source": [
        "From the PCA plots, we can find resampling indeed created more true values for our predicting variables, but this process also makes it harder to differentiate the two classes.\n",
        "\n",
        "We obtained a group of better models, for KNN, LR & RF after resampling, but this process may lead to overfitting the dataset, the examination on a new fresh dataset is necessary to validate our models."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6JcXSFSe3Z4L"
      },
      "source": [
        "## 8. **Self assessment**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HHFpOoGujwFh"
      },
      "source": [
        "**Bongju Yoo**\n",
        "\n",
        "\n",
        "1.   Hyperparameter Tuning (part)\n",
        "2.   Modeling (part)\n",
        "3.   EDA (part)\n",
        "4.   Data description (part)\n",
        "5.   Formulate the report (part)\n",
        "6.   Clean up codes (part)\n",
        "\n",
        "**Hongtianxu Hua**\n",
        "\n",
        "\n",
        "1.   Hyperparameter Tuning (part)\n",
        "2.   Modeling (part)\n",
        "3.   Discussion\n",
        "4.   Data description\n",
        "5.   Formulate the report (part)\n",
        "6.   Clean up codes (part)\n",
        "\n",
        "\n",
        "**Yonghao Li**\n",
        "\n",
        "1.  Preprocessing & EDA\n",
        "2.  Check Imbalance Issue\n",
        "3.  Stratified Sampling and Resampling (Oversampling and Combo)\n",
        "4.  RFECV Feature Selection\n",
        "5.  kNN Hyperparameter Tuning (Part)\n",
        "6.  ROC Curve Analysis\n",
        "7.  Formulate the report (Part)\n",
        "8.  PyCaret (Excluded in Final Version)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ayQ9CSbN3fmK"
      },
      "source": [
        "## 9. **Appendix**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KWZL4pvkYu6d"
      },
      "source": [
        "## Citation"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HJssf-xTRr5r"
      },
      "source": [
        "Bank Marketing Data Set: http://archive.ics.uci.edu/ml/datasets/Bank+Marketing#\n",
        "\n",
        "Bank Marketing Data Set Description and Encoding: https://www.mldata.io/dataset-details/bank_marketing/\n",
        "\n",
        "PCA using Python (scikit-learn): https://towardsdatascience.com/pca-using-python-scikit-learn-e653f8989e60\n",
        "\n",
        "PyCaret: https://towardsdatascience.com/announcing-pycaret-2-0-39c11014540e\n",
        "\n",
        "Imbalanced-learn Python Library: https://imbalanced-learn.org/stable/install.html\n",
        "\n",
        "RFECV Feature Selection: https://medium.com/analytics-vidhya/feature-selection-using-scikit-learn-5b4362e0c19b\n",
        "\n",
        "KNN Hyperparameter Tuning: https://medium.com/datadriveninvestor/k-nearest-neighbors-in-python-hyperparameters-tuning-716734bc557f\n",
        "\n",
        "GridSearchCV Hyperparameter Tuning: https://towardsdatascience.com/logistic-regression-model-tuning-with-scikit-learn-part-1-425142e01af5\n",
        "\n"
      ]
    }
  ]
}