NIR_2_GridSearchCV.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.\n",
      "  from numpy.core.umath_tests import inner1d\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "from sklearn.model_selection import train_test_split \n",
    "\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn import ensemble\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "from sklearn import metrics\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import roc_curve"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  https://www.kaggle.com/omnamahshivai/surgical-dataset-binary-classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>bmi</th>\n",
       "      <th>Age</th>\n",
       "      <th>asa_status</th>\n",
       "      <th>baseline_cancer</th>\n",
       "      <th>baseline_charlson</th>\n",
       "      <th>baseline_cvd</th>\n",
       "      <th>baseline_dementia</th>\n",
       "      <th>baseline_diabetes</th>\n",
       "      <th>baseline_digestive</th>\n",
       "      <th>baseline_osteoart</th>\n",
       "      <th>...</th>\n",
       "      <th>complication_rsi</th>\n",
       "      <th>dow</th>\n",
       "      <th>gender</th>\n",
       "      <th>hour</th>\n",
       "      <th>month</th>\n",
       "      <th>moonphase</th>\n",
       "      <th>mort30</th>\n",
       "      <th>mortality_rsi</th>\n",
       "      <th>race</th>\n",
       "      <th>complication</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19.31</td>\n",
       "      <td>59.2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.57</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>7.63</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.43</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>18.73</td>\n",
       "      <td>59.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.21</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>12.93</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.41</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>21.85</td>\n",
       "      <td>59.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>7.68</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.08</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>18.49</td>\n",
       "      <td>59.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.65</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>7.58</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.32</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>19.70</td>\n",
       "      <td>59.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.88</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     bmi   Age  asa_status  baseline_cancer  baseline_charlson  baseline_cvd  \\\n",
       "0  19.31  59.2           1                1                  0             0   \n",
       "1  18.73  59.1           0                0                  0             0   \n",
       "2  21.85  59.0           0                0                  0             0   \n",
       "3  18.49  59.0           1                0                  1             0   \n",
       "4  19.70  59.0           1                0                  0             0   \n",
       "\n",
       "   baseline_dementia  baseline_diabetes  baseline_digestive  \\\n",
       "0                  0                  0                   0   \n",
       "1                  0                  0                   0   \n",
       "2                  0                  0                   0   \n",
       "3                  0                  1                   1   \n",
       "4                  0                  0                   0   \n",
       "\n",
       "   baseline_osteoart      ...       complication_rsi  dow  gender   hour  \\\n",
       "0                  0      ...                  -0.57    3       0   7.63   \n",
       "1                  0      ...                   0.21    0       0  12.93   \n",
       "2                  0      ...                   0.00    2       0   7.68   \n",
       "3                  0      ...                  -0.65    2       1   7.58   \n",
       "4                  0      ...                   0.00    0       0   7.88   \n",
       "\n",
       "   month  moonphase  mort30  mortality_rsi  race  complication  \n",
       "0      6          1       0          -0.43     1             0  \n",
       "1      0          1       0          -0.41     1             0  \n",
       "2      5          3       0           0.08     1             0  \n",
       "3      4          3       0          -0.32     1             0  \n",
       "4     11          0       0           0.00     1             0  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('Surgical-deepnet.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 14635 entries, 0 to 14634\n",
      "Data columns (total 25 columns):\n",
      "bmi                    14635 non-null float64\n",
      "Age                    14635 non-null float64\n",
      "asa_status             14635 non-null int64\n",
      "baseline_cancer        14635 non-null int64\n",
      "baseline_charlson      14635 non-null int64\n",
      "baseline_cvd           14635 non-null int64\n",
      "baseline_dementia      14635 non-null int64\n",
      "baseline_diabetes      14635 non-null int64\n",
      "baseline_digestive     14635 non-null int64\n",
      "baseline_osteoart      14635 non-null int64\n",
      "baseline_psych         14635 non-null int64\n",
      "baseline_pulmonary     14635 non-null int64\n",
      "ahrq_ccs               14635 non-null int64\n",
      "ccsComplicationRate    14635 non-null float64\n",
      "ccsMort30Rate          14635 non-null float64\n",
      "complication_rsi       14635 non-null float64\n",
      "dow                    14635 non-null int64\n",
      "gender                 14635 non-null int64\n",
      "hour                   14635 non-null float64\n",
      "month                  14635 non-null int64\n",
      "moonphase              14635 non-null int64\n",
      "mort30                 14635 non-null int64\n",
      "mortality_rsi          14635 non-null float64\n",
      "race                   14635 non-null int64\n",
      "complication           14635 non-null int64\n",
      "dtypes: float64(7), int64(18)\n",
      "memory usage: 2.8 MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((14635, 24), (14635,))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels = df['complication'] # y  - целевой признак\n",
    "features = df.drop('complication', 1) # X -  матрица признаков\n",
    "features.shape, labels.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((11708, 24), (2927, 24), (11708,), (2927,))"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split \n",
    "features_train, features_test, labels_train, labels_test = train_test_split(features, labels, test_size=0.2, random_state = 42)\n",
    "features_train.shape, features_test.shape, labels_train.shape, labels_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 3.84 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "feat_labels = df.columns[1:]\n",
    "\n",
    "forest = RandomForestClassifier(n_estimators=1000,\n",
    "                                random_state=0,\n",
    "                                n_jobs=-1)\n",
    "\n",
    "forest.fit(features_train, labels_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 1) asa_status                     0.215494\n",
      " 2) dow                            0.116529\n",
      " 3) race                           0.107071\n",
      " 4) Age                            0.099976\n",
      " 5) ccsMort30Rate                  0.071420\n",
      " 6) month                          0.069473\n",
      " 7) moonphase                      0.046906\n",
      " 8) complication_rsi               0.044517\n",
      " 9) ccsComplicationRate            0.035763\n",
      "10) gender                         0.032305\n",
      "11) mort30                         0.029112\n",
      "12) baseline_cvd                   0.025214\n",
      "13) baseline_cancer                0.015566\n",
      "14) baseline_psych                 0.013085\n",
      "15) complication                   0.011968\n",
      "16) baseline_dementia              0.011594\n",
      "17) baseline_osteoart              0.011075\n",
      "18) hour                           0.010977\n",
      "19) baseline_charlson              0.009001\n",
      "20) baseline_digestive             0.008028\n",
      "21) baseline_pulmonary             0.006916\n",
      "22) ahrq_ccs                       0.006304\n",
      "23) baseline_diabetes              0.001041\n",
      "24) mortality_rsi                  0.000666\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAEYCAYAAAAJeGK1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJztnXe4JFW1t9/fDFEyMiZgAAFRVJIICIgiGQUUiSLJwEUugpdrwCsKggkEUQwkQRBBBEQdEAUkI3GGjMjnMKQBFSXIoGTW98faPadOT5/uXVV9TnefWe/z9HNOVdeq2tVdXav22r+9lsyMIAiCIOg3JvS6AUEQBEHQinBQQRAEQV8SDioIgiDoS8JBBUEQBH1JOKggCIKgLwkHFQRBEPQl4aCCIAiCviQcVNBTJD0g6VlJz0h6UtJvJS3b63YFQdB7wkEF/cA2ZrYw8Hrg78D3e9yeIAj6gHBQQd9gZs8B5wGrNtZJer+kWyU9LelhSYcV3lsprdsgLe8l6dr0/4KSrpN0YFp+r6SZxeNJulbSXun/CZIOkfSgpMck/VTSYoVtN0z7eyodcy9JO6ee3zOSXpb0XGM52Rwm6Wc55562fbGwv2ckmaTl0/unSTpB0qWSZkm6StJyBXuTtFL6f3Lqlf4sLa8o6d5k93dJXyvYnda0vJIkKyzvLemeZDtD0n8V3hv2mUo6KrVrgbT8FklXps/sbknbNh33hXSeT0j6saR5cj6rYO4hHFTQN0h6FbAzcENh9b+BPYDFgfcDn5L0QQAzmw7sCJwtaeXCfiYAZwA3mdn3Mg+/V3ptDLwRWBj4QdrfZOB3eM9uErAGcJuZ/cLMFk69v2uA/QvLVSjub/EW7+8GHAEsBdwGnDnCfo4AHi8sPwZsDSwKrAd8QtLbM9v0GPCBZLs3cKyktZo3kvQFYFO8N/ycpHmBC4BLgNcAnwbOlLRKweyodK6r4t/tlpltCuYSwkEF/cCvJT0FPA1sBny78YaZXWlmd5rZK2Z2B/Bz4D2F928AvoI7kElp9dH4TfGgEm3YDfiOmc0ws2eALwK7pKf63YA/mNnPzexFM3vczG6rfLbV+a2ZXW1mzwNfAt7VPF4naTXgXcDpjXVmNsvM7jNPvCk8jPpozgHN7LcNWzO7Cnc472465ieAzwJbmtnTafV6uJP/lpm9YGaXAxcCu7Y4zMTUrsdbvBfMxYSDCvqBD5rZ4sD8wP7AVZJeByBpXUlXSPqHpH8B++I9iCKbAU8AhwPvADYBVsGdVJE3pHDTU8khrld8D3iwsPwgMA/wWmBZ4L6K57ZTOt4/U3jujRX3A/Bw45/kRJ/A213kSODLwIvFlSns9y9gOnAtMKvw9mcLn8ktTXZbSbohheGewntixc9/Ujref/CeZYM3AA+b2SuFdQ8CSzcfN53X9cDN7U4+mPsIBxX0DWb2spmdD7wMbJhWnwVMAZY1s8WAE/CnbQAkbQasDWyEh/Wewh3W8cB3mw7xqJkt3ngxPJT4KLBcYXky8BLe23gYWLHiaZ2TjvUG4CHgGxX3A+4oAZC0MLAkw3tC78OdxznNhmb2UPr8lsZ7oB8vvH104TOZHb6TND/wS7xH+tr0/kUUPn/8u9oK2Ac4SdIiaf2jwLIp3NpgMvBI83GBRYD5gM91/ASCuYpwUEHfIGc7YAngnrR6EeCJNK6xDvCRwvYLAD8C9ksCi+uA+8zsMeCbwBqScsc1fg78j6QV0s3/G/iY0Ev4WM+mknaSNI+kV0tao+3emjCzF4BnqPeb2zqJNebDx5luNLOHC+8fBnzOmmroSFpG0pJpcT48pPZsxvHmw3u1/wBekrQVsHnTNk+Y2Z/M7GLgMuCotP5GfPzw85LmlfReYBvg7BbHeRkwhkK0QQCEgwr6gwuS8u1p4OvAnmZ2d3pvP+BwSbPwsaZi7+AQ4AYz+0PzDtM4zb7ADyUtmNGGU/Ee2NXA/cBz+MA+ZvYQHtr6Xzysdhuweua5fUjSTEmP4L2TQzLtWnEWcGhqwzvwsbEit5rZlS3s3g7cmj7D6/Be0BmdDmZms4AD8M/8SfzhYEobk4OAD0h6b3LI2+K9q3/iDxJ7mNmfC9t/Pn3vf8PvRUd2alMwd6EoWBgE/Y+k04CZZlbHwQXBQBE9qCAIgqAvCQcVBEEQ9CUR4guCIAj6kuhBBUEQBH1J3+W+WmqppWz55ZfvdTOCIAiCUWLatGn/NLOO0wr6zkEtv/zyTJ06tdfNCIIgCEYJSQ923ipCfEEQBEGfEg4qCIIg6EvCQQVBEAR9STioIAiCoC8JBxUEQRD0JX2n4stFX1XnjQrYoTEhOQiCYJCIHlQQBEHQl4SDCoIgCPqScFBBEARBXxIOKgiCIOhLwkEFQRAEfUmWg5K0paR7JU2XdHCL9w+S9CdJd0i6TNJyhff2lPSX9Nqzm40PgiAIxi8dHZSkicAPga2AVYFdJa3atNmtwNpmthpwHnBUsl0SOBRYF1gHOFTSEt1rfhAEQTBeyelBrQNMN7MZZvYCcDawXXEDM7vCzP6TFm8Alkn/bwFcamZPmNmTwKXAlt1pehAEQTCeyXFQSwMPF5ZnpnUj8XHgdxVtgyAIggDIyyTRKmVDy7QMkj4KrA28p4ytpH2AfQAmT56c0aQgCIJgvJPTg5oJLFtYXgZ4tHkjSZsCXwK2NbPny9ia2UlmtraZrT1pUscii0EQBMFcQI6DuhlYWdIKkuYDdgGmFDeQtCZwIu6cHiu8dTGwuaQlkjhi87QuCIIgCNrSMcRnZi9J2h93LBOBU83sbkmHA1PNbArwbWBh4FxJAA+Z2bZm9oSkI3AnB3C4mT0xKmcSBEEQjCuyspmb2UXARU3rvlL4f9M2tqcCp1ZtYBAEQTB3EpkkgiAIgr4kHFQQBEHQl4SDCoIgCPqScFBBEARBXxIOKgiCIOhLwkEFQRAEfUk4qCAIgqAvCQcVBEEQ9CXhoIIgCIK+JBxUEARB0JeEgwqCIAj6knBQQRAEQV8SDioIgiDoS8JBBUEQBH1JOKggCIKgLwkHFQRBEPQl4aCCIAiCviQcVBAEQdCXhIMKgiAI+pJwUEEQBEFfEg4qCIIg6EvCQQVBEAR9SZaDkrSlpHslTZd0cIv3N5J0i6SXJO3Q9N7Lkm5LryndangQBEEwvpmn0waSJgI/BDYDZgI3S5piZn8qbPYQsBfw2Ra7eNbM1uhCW4MgCIK5iI4OClgHmG5mMwAknQ1sB8x2UGb2QHrvlVFoYxAEQTAXkhPiWxp4uLA8M63LZQFJUyXdIOmDrTaQtE/aZuo//vGPErsOgiAIxis5Dkot1lmJY0w2s7WBjwDflbTiHDszO8nM1jaztSdNmlRi10EQBMF4JcdBzQSWLSwvAzyaewAzezT9nQFcCaxZon1BEATBXEqOg7oZWFnSCpLmA3YBstR4kpaQNH/6fylgAwpjV0EQBEEwEh0dlJm9BOwPXAzcA5xjZndLOlzStgCS3ilpJrAjcKKku5P5W4Cpkm4HrgC+1aT+C4IgCIKW5Kj4MLOLgIua1n2l8P/NeOiv2e464O012xgEQRDMhUQmiSAIgqAvCQcVBEEQ9CXhoIIgCIK+JBxUEARB0JeEgwqCIAj6knBQQRAEQV8SDioIgiDoS8JBBUEQBH1JOKggCIKgLwkHFQRBEPQl4aCCIAiCviQcVBAEQdCXhIMKgiAI+pJwUEEQBEFfEg4qCIIg6Euy6kGNN/RVldreDrVRakkQBEEwEtGDCoIgCPqScFBBEARBXxIOKgiCIOhLwkEFQRAEfUk4qCAIgqAvmStVfHUIBWAQBMHYkNWDkrSlpHslTZd0cIv3N5J0i6SXJO3Q9N6ekv6SXnt2q+FBEATB+Kajg5I0EfghsBWwKrCrpFWbNnsI2As4q8l2SeBQYF1gHeBQSUvUb3YQBEEw3snpQa0DTDezGWb2AnA2sF1xAzN7wMzuAF5pst0CuNTMnjCzJ4FLgS270O4gCIJgnJMzBrU08HBheSbeI8qhle3SzRtJ2gfYB2Dy5MmZux48YvwqCIIgn5weVKu7au6dM8vWzE4ys7XNbO1JkyZl7joIgiAYz+Q4qJnAsoXlZYBHM/dfxzYIgiCYi8lxUDcDK0taQdJ8wC7AlMz9XwxsLmmJJI7YPK0LgiAIgrZ0dFBm9hKwP+5Y7gHOMbO7JR0uaVsASe+UNBPYEThR0t3J9gngCNzJ3QwcntYFQRAEQVuyJuqa2UXARU3rvlL4/2Y8fNfK9lTg1BptDIIgCOZCItVREARB0JeEgwqCIAj6knBQQRAEQV8SDioIgiDoS8JBBUEQBH1JOKggCIKgLwkHFQRBEPQl4aCCIAiCviQcVBAEQdCXhIMKgiAI+pJwUEEQBEFfkpWLL+g9UewwCIK5jehBBUEQBH1J9KDmAqL3FQTBIBI9qCAIgqAvCQcVBEEQ9CXhoIIgCIK+JBxUEARB0JeEgwqCIAj6knBQQRAEQV8SDioIgiDoS7IclKQtJd0rabqkg1u8P7+kX6T3b5S0fFq/vKRnJd2WXid0t/lBEATBeKXjRF1JE4EfApsBM4GbJU0xsz8VNvs48KSZrSRpF+BIYOf03n1mtkaX2x0EQRCMc3J6UOsA081shpm9AJwNbNe0zXbA6en/84BNJJVLXxAEQRAEBXIc1NLAw4XlmWldy23M7CXgX8Cr03srSLpV0lWS3l2zvUEQBMFcQk4uvlY9oeZkbSNt81dgspk9LukdwK8lvdXMnh5mLO0D7AMwefLkjCYFQRAE452cHtRMYNnC8jLAoyNtI2keYDHgCTN73sweBzCzacB9wJuaD2BmJ5nZ2ma29qRJk8qfRRAEQTDuyOlB3QysLGkF4BFgF+AjTdtMAfYErgd2AC43M5M0CXdUL0t6I7AyMKNrrQ9GnciEHgRBr+jooMzsJUn7AxcDE4FTzexuSYcDU81sCnAKcIak6cATuBMD2Ag4XNJLwMvAvmb2xGicSNB/hHMLgqAOWfWgzOwi4KKmdV8p/P8csGMLu18Cv6zZxiAIgmAuJDJJBEEQBH1JOKggCIKgLwkHFQRBEPQlWWNQQTDWhMAiCILoQQVBEAR9SfSggnFH2d4XRA8sCPqR6EEFQRAEfUn0oIKgiRj/CoL+IHpQQRAEQV8SDioIgiDoSyLEFwRdJMKDQdA9ogcVBEEQ9CXhoIIgCIK+JBxUEARB0JeEgwqCIAj6knBQQRAEQV8SKr4g6BNCARgEwwkHFQTjgHBuwXgkHFQQzOXUcW7hGIPRJBxUEAQ9IRxj0IlwUEEQzFXULccSznHsCAcVBEEwRkSvsRxZMnNJW0q6V9J0SQe3eH9+Sb9I798oafnCe19M6++VtEX3mh4EQRCMZzo6KEkTgR8CWwGrArtKWrVps48DT5rZSsCxwJHJdlVgF+CtwJbAj9L+giAIgqAtOT2odYDpZjbDzF4Azga2a9pmO+D09P95wCaSlNafbWbPm9n9wPS0vyAIgiBoi8zaxykl7QBsaWafSMu7A+ua2f6Fbe5K28xMy/cB6wKHATeY2c/S+lOA35nZeU3H2AfYJy2uAtxb45yWAv4ZtmHbZ8cO2/Ft28tjD6LtcmY2qdNGOSKJViNzzV5tpG1ybDGzk4CTMtrSEUlTzWztsA3bfjp22I5v214eexBtc8kJ8c0Eli0sLwM8OtI2kuYBFgOeyLQNgiAIgjnIcVA3AytLWkHSfLjoYUrTNlOAPdP/OwCXm8cOpwC7JJXfCsDKwE3daXoQBEEwnukY4jOzlyTtD1wMTARONbO7JR0OTDWzKcApwBmSpuM9p12S7d2SzgH+BLwE/LeZvTxK59KgTqgwbMe3bS+PHbbj27aXxx5E2yw6iiSCIAiCoBdEPaggCIKgLwkHFQRBEPQl4aCCIAiCviQcVBAEfUFS+nZcF8w9DLSDkrSjpEXS/4dIOl/SWiXsF+hCG5aTtGn6f8FGezrYvFbSKZJ+l5ZXlfTxzONVti3sY31JH5G0R+OVaSdJH5X0lbQ8WVLb1FWS3pz+rtXqVabdZZE0UdL/VLRdTNLOkg6S9D/p/8W73cYRjj1R0s9q7qPSd9xjftli3Xkt1s0mfVZ/GKX2ZJF+96tUsHtbjWNWvvdJWlHS/On/90o6YKyu7bIMtIpP0h1mtpqkDYFvAkcD/2dm62baTwf+DlwDXA380cz+VeL4n8RTNC1pZitKWhk4wcw26WD3O+AnwJfMbPU0uflWM3t7xjEr2yb7M4AVgduAhuTfzOyADNvjgVeA95nZWyQtAVxiZu9sY3OSme0j6YoWb5uZvS/juG8Cjgdea2Zvk7QasK2ZfS3D9koze2+n7Zps9gAOBS4BHkmrlwE2A75qZj/tYH8nLTKmNDCz1TLacDGwTcp/WYqa3/H2eLLn1+CZYJRsF82wXQBPHP1WYPbDn5l9rIPdm5PNUcDnCm8tCnzOzN7awX4KsHuZ326T/ZvScZejMPUm89rcBr/vzGdmK0haAzjczLbNsL0WmA84DTjLzJ4q0ebK9z5JtwFrA8vj04emAKuY2dZtbD5vZkdJ+j6tswF1vLaqMOj1oBo/vvcDx5vZbyQdlmtsZitJmgy8G/gAnm39KTNbI3MX/40nv70x7e8vkl6TYbeUmZ0j6YvJ7iVJufPD6tiCX5irWrUnk3XNbC1Jt6ZjP5kmb4+Ime2T/m5c4XgNTsZvICemfd0h6Sygo4MC/ijpB8AvgH8X2nVLG5svAe9ovmEkh3wj0NZB4dcS+PUBcEb6uxvwn4w2AzyAt30Kw9v9nQzbOt/xUbhjvKeC7RnAn4EtgMPx883Zzyr4Z7Y4sE1h/Szgkxn2zwF3SrqU4Z9V7k3zXOAE/DorO0/zMPwecGU65m0qlBtqh5ltmB5qPwZMlXQT8BMzuzTDvM6975V03/gQ8F0z+37jN92Gxvc4NfMYXWHQHdQjkk4ENgWOTN3W7LClpGWADXAHtTpwN3BtieM/b2YvSGrsbx7aPDkX+LekVze2lbQekPv0V8cW4C7gdcBfS9g0eFFeLqVx7El4j6ojknYEfm9msyQdAqwFHGFmnX4YAK8ys5san3Pipcw2r5/+Hl5YZ0C7p2PR+nt8hdb5JYdhZg8CSNrAzDYovHWwpD82tWUkHk2vCUDHsHETdb7jv1d0TgArmdmOkrYzs9PTQ8TFnYzSzfVC4Atm9o0Kx/1telXlJTM7vobtv5quzWzSQ+0h+I3/OGBN+c7+z8zOb2Na5973oqRd8ew/jQeCeTu084L0t1G1AkkTgIXN7OnM45Zm0B3UTnidqaPN7ClJr2d4iKATD+GpnL5hZvtWOP5Vkv4PWFDSZsB+wAUZdgfh3eoV0w1rEp4iKodWtjt2MpJ0AX7TXQT4U3pae77xfk5IAv8B/Qp4jaSvpzYfktnuL5vZuSkksQUekjgBz3rfiX9KWpEhx7gDmTffij23rwO3SLoEeDitm4yH+I4osZ+FJG1oZteCjwsBC+UYmtlXk81CZvbvTtunbbvxHU+V9Avg10227W6WDV5Mf59K4yt/w8NIHTGzl9NvqLSDKt40K3KBpP3wa7t4zk9k2N4l6SPAxNQbOgC4LuegKVS9N94LuhTvud4i6Q3A9UC7z7zOvW9vYF/g62Z2v1yIkjXmmR469sV7cNOAxSR9x8y+nXnsUgz6GNTkVuvN7KFM+9WBDYGN8BvQX4CrzOyUTPsJeMx9c/zJ+mLgxzmhldTbWiXZ3WtmL3YwadjNj18cs22BCWb2fAe797R738yuyjz+m4FN0rEvy33alnSrma0p6ZvAnWZ2VmNdhu0b8bQq6wNPAvcDuzV6Kh1sX4vf9N5gZlvJi2i+q9N3nMJ5WwBL4+c6E7jYzJ7sdMzCPt4BnIonTza8p/uxDuHFhu278BRiC5vZ5HSt/peZ7dfGpvZ3LOknrU3bjyMl20/gQofV8HHShYGvmNkJnWyT/dfxz6pMOBZJ99N6XOSNmce9v8Vqy7GX9Co8JLx5WnUx8DUzey7D9mo8rHiemT3b9N7uZnZGa8vZkZO7zWxWWl4ED+ve2Om4afsFgclmVqq0kaTbzGwNSbsB7wC+AEzLGVethJkN7Au4E7gj/f0LHva5u+Q+FsafRL4OPAg8UMJ2IWBiYXkiHo7qZLd9i9cmwGsybG/JWdfG/sicdSPYrgjMn/5/L/60uHim7YX4GNJ9+FjD/MDtmbYrFD7vRYrrMmx/hz9t3p6W58EdZNlrbckKNhPT30WBxUra3ohXAri1sO6u3M8LWKCwvCCwfNn2j/ULuKLF6/IMu1cXXksDn8GFCmPR5jUr2k0Ezqxx3FtJHYy0PCH3PoCH9e4F7k/LawBTMm3vxsOB5wLvSeuyfseVznMsvsSxeuHjGieW2H5qcm4nArvjRbTKHO8G/Am3sbwwcF2G3W/xpLq/TK/H07q/4GqkVjavw59Y7gHWTOe6VnIUfy7R5lYO7o5M29vSDX4lvDryscBFmbavwh3xymn59cDmNdo8LdP25vS3eKO/rYPNBulzvhsPQV4KzMDDfe8q8Vk/hPf8NineTDJtb2zR7lyHPhVXlTWW52t8Dhm2y+ChrsdwhesvgWU62Hw0/T2o1avMeXfrBVxbYtt58Yet89Jrf2DeTNsrcGHIEcBbS7bx98XvqaTtHNdwid/xNLynWry2sh7agE/jytaL8MjCcsA1o/U9DvoY1DDM47cjSp5bsJWZ/aPGIRcws2cKx38mdfk78QrwFjP7O8wOQx2P3wyvZkj1VWQLYC/8BlJUcs0C/q/TASV9Ch8je6OkOwpvLUJmzJwh9c/2wPcsT/3TCIXeZGaz532Y2V/pMI6kIfnxYumYDRalIGPuQBVRybF4r2th/MHhg2Z2rXyeyfdxB5bDKvjT6n8DpyQhwNmWxqQ68HAas7KklDyAPEUcwDxWkKebC3naqi0L/AQ4i6FxzY+mdZu1sWmMq7USc5QaQ5D0fuaUqbcVlWj4/J8JuIqxjLDkeNxJ/Sgt757WfaKToZltLOl1+PVykqRFgV9YxhQIPGJTVak5Q9IBqZ3gv+0ZGXbQWtiRMywxARfRLF1Y9xBQR6HbloF2UJIOKixOwHsUZRzOC5K+g49BAVyFhwbKKOrWshQjT2MOz3awAQ+3/L2w/BjwJjN7QlLLsSjzgeDTJX3YzFpNaOzEWXi465vAwYX1syxvMBiG1D97kKn+ATCzVyTdLmmyZY4PJurKj6GaIGVeM7sTQNI/Gg4lPQAtmNt483GFc4Bz0pjW9/BrbGKG+b5p+6Xx8a9LGJKtd+IfkrY1L4WDpO3IL809ycyK41CnSfpMOwMzOzH9+wcz+2PxPUm5zhxJJ+A97Y2BH+PfU079uGMK/7+ES/R3yj0u8E4zW72wfLmk23ONzexvwHHyuX6fB75C3hSIOkrNfXHR0iG4c7kMn5OZQyVhR/od749f0411Rr6itjSDLpI4tLDYuDB/aRkDlMn+l7gkt6EC2h1Y3cy2H9lqmP07gbMZqhL8emBnM5vWwe5HuCjj3LTqw/hN6HPAhdZBeVblKbPJfiLwWoZPSuzoOJLAYF/gejP7eVL/7Gxm38qwvRx4J37DKT4t5kxofJeZXd9puzb2pQQpkm5v3LAkfdDMfl14765iTzDj2O8Bdga2whWjv6j4gJGNpJVwVdYb0qqZeOj4vgzbP+ATR3+eVu0K7G0dJp8n21vMbK1O69rYNyafNv4uDJxvZpt3NK6BpFuAHRufTxLlnJfTbklvwb/fHfBQ/dn4PeixUWxyLUYQdhxhHYRWyfbL+EN4s5Al9yG3FIPuoHY0s3M7rWtjf5s1Tcptta7DPuZl6Ob35043v2QjfDxmw7TqceD1ZtbxCXmkp0wzy02VtD8+ufDvDM1hMhstFc7QcVsqzCxPWVY6Q0FTSLDVcUeU8EraFu8N/Kdp/YrAh83sqE5tTtvfj4/bnYMPQmfJxZPt6cCBliYLpx7YMe3OuWA70Vy2vTD+G59V4riTgR8A78KfzK9L7RhRMZkUh+vj4oRjC28tCnyoqXfS7tg3mtm6km7Afx+P48KQlTvYLYZn/qgUCZG0CR7GnMHQuMreZtYq+0mz7Q24Mz/XzB7ttH2T7SS8x9V8XedksKhzfVS+b9ZRPFZhoEN8wBcZ6oW0WzcSz2r4PJUNyAvRFVkFWBW/wNaUhHVIhWNmJuk+fMxpJ1w2nftUvX7hKfOrko6h/XyJZj6DpzV5vIQNACkc8E2GzhfIk/N2ckSSrjezd43wdpUMBY2Q4Gvwm+flaXljfNb/iJ9ZITQ27EdrZveN8AMdidWt+iTG1ayQycI8a0dHSX5iuqTz8OrXpSbdpp50znypIvPh43XzMDxU9TT58/sALpTnhPs2cAvuIE/OsDsVj4Q0wnq74w4nKxJiZpela7v4oNmxN5Fs18vZbgTOxHsiH8AjE3uSP0RR5/qofN80s7bJeyVtZnmZMPKwHihs6r7wcMn38V7AcYXXaXhvInc/qwO346HBB3Dp5mol7A/FVTx/x38Qf8NDAyNt/yY8Pn0PnrHi08CDJc+9oe66AQ/hzA/8pYT9FfggepXP/VpckXYH/pR5GJ6brhvf6a2d3iOplPBxr47y47TthXjvtLH8ejxslGNbV9J/OgUZPrAE7jRybG8HligsL0m+0moRfIzuunSd7AMs2sHm8+nv95t+U8cBx2Ued7luXAtpX/OTKc2ntaKtrVKzadsdGZq+cAj+8LJWB5tz0t/GVJfG605KqOnS3zsK664areuDLt03Oxwj+/eR8xrUHtSjuJR2W1wy2WAW0DF7dZO44qcMqZD+jacOuWMOo9bsgDu5W81s76TG+3Gb7f+MJ6bdxsymp7aUzbbd6imz3TGbmQFcKem3DJ81n6McWtD8aVPmIZ/DJF2DO+q6tIs1V85QgAtSimrBv+MPCiMiaStga2BpSccV3lqUcgPCdZ5yjwGuSz0h8Jvo13MMzUN6JwMnS9oID0Edm/Z1ROPaa6Ibudbml3QS/t2USroKs8Pln2IoVHelpBOtc9i8biSkVZaThqp2JA5Mfz/QZptONM7rr2lc+VFcpZtD8fowvPfY6fqodd/MpFrOpxEYSAdlZrcDt0s6K+PibUUjDLEKPnD58jdKAAAgAElEQVT/G/yD/Sgu887lOXNly0tyeeljQLtw14eBXYArJP0eH1At9YWaWSPVzi/lsuUFrFwW54fSa770KsNzcqnpX9JY1iN4CG20OSnF2A/BFXkL4z3RHK6UZwb/Of5D3gXvRbajWz/kCZKWsJR9QtKSZP7mzOynkqbhIUkB25vZn3Jskwjm/XhKm+Xxm9mZeM7Ji2jhoC3lWgP+Yy3GJ3KOy1DS1R9TPukqVJd7fwpXuC6Wlp/Ew2W5lE68Wnjo2c/MvlB8T9KReIaFTnwttfl/8Z7NomReX+n6mIrnlJzj+ihedwWb2fdN/DosnUkip2nd3VsXu2Nj/QJWxifW/QnvGcwAZpSwv4TUtU/Li+AJTXNshaeiWRyPH/8FDxH+JMN2IXwc5UI8u/XxZE5abbGvzYBLK9gtQmGScabNO3HnsAwe0jwfWK9L3+WIIb4u7Ht7fPD+WHzQPsem1kz/tI898J7JEfjY2Z8ZYSJ2mza8AVd8TsZvKDl2M9K1uX6L99qG66gR1iRz8nQb+zkmIrda12Kb+dPv6Sv4HMFD8RRLucetk+Wk8sT30Xy1+86okUmiznGrvAayB1XgJ/jFeCz+pLk35Xokk4FivZ0XyE9uaZLWMA/hnJB6RIuaWcfwoLma60zgzPRUvSM+N+mSkWwkvQ9/On0DnsjzG3h4UmSGftJ+3oaLDpZMy/8E9jCzuzPafXP69xn8s+4mu7damXoDS5jZP9PyfPjT8UFm9pacHZsr9soISTBXwb1a0nxWoSZT2kfpp9zCe5/Gr+2/40/4jQzrOWrL1awwgbypTS1LUHQprFkn6SrAy5JWtOFy75ye2G+Ap/CQ9yMdtm1F6cSrGpr4vqLmnPj+x9ZWc+xjEj5WuDzDQ6IdlXg5u2/z3mFULBGSwQNd2o/Ta09f01s3BhnvLKzLTruBzwW4Hf/CDsUlwV8sYf9DfJLfWJzrrXhao/mBD+IKqQMr7Oc6YOPC8nvJSM+Utn0TPrZxCa6Ku5x8scL2eC/zX6nts4CnO9jskrZ/FJcOb4zP6fkVHQax6xy3YHsiPnfpy4xC6h7aP+VOB15dcb9H4Y5lXnwC5z9J6Yja2KyOO/4H09/Ga3sKg/Ed9nF/i1eZiMYmePj5yvR9P1C8VtvYZeUobGG3ZLtXB9vFcMfyc1ww1Hhl52xMv8UjcQf54cZrDK6tVmm0coUdU/EJ41nXRN3XoM+D+iMeVz8Pv1k+AnzLzLLLL8vTpLw7LV5tefWJGrZ/wm/aD+ICi0b10a7PKVLThEdJ95nZihX2M3sSart1I9nivbhpFJ5srcPE5GQ7nZKF8CTdhacZmp6+p+uBXczsVyX2Ufq4BduW4g9LpTDqojbZ3OVZCTYzs9Kz9DWUcfpD+MPM/wBXZH7H81q1cd2uIM/WX0runYQZ37eU/aPEse7He6Uq/G1glpfNfEVgppk9L+m9eA/3p5ZRHVcl51yWofl+0fTeKfiDy8G4UzwAz57SseSQfBL43vjk5Kl4FOsSGyVHMugO6p14jH9xPM6/KHCUZaac78Lxl2u13jLKQFQ41gzgs4VVRxeXLa9eD5J+hYdCGvn+PgqsbWYfzLCdZmbvyG70cNs/2vDifTk2zU75z2b25tE+bot9LILfsFqGzWrst9NNZBU8F2AptaWku83srZJOxrMa/L7EQ0jluW7yDAUH4WNl+zTmFpnZhZ1sm+yXM7NPdrKXdCfuWObBx6Nn4J9V9oOiJAHLWrkUXEX70uXTC7Zfw6MXF1U5dod9t3v4KWaSaJQJOsIyM/CkfUzAFYzH4xP+T8Xzc3Y1o8SgO6hamSQGCbWu09PALDNundRwX8UTngpXLR7W7okvjZOBP2k9RoUxBknfwzOyZxfCkzST4YlxDyouZ96sSx+3YDtsvA4PlWWN1+XQwUFV7r1J+hbec3oWH2tYHE+h1bE4pKRrGRrX3YY0rmtmHacSyAsdTsM/o7fJ8xZen9tLKGs/0gNig9wHxZoPXreY2VqSPg88aymB8kjOocl2Fi6YeoEhybmZ2aIZtkt22qbT7zIpj81KZBpJdo1Ci1vjzu1MPCvO7t3uEQ66g6qV+2tuRNLa+NPT8gwNzLZ92mwKhTSTGwpp5WDbOtaRbtIF45ybdenjFmyvA75kKeVNCuF8w8zWb2uYSe6NrOK+l8DH2l5OT8yLmic17WQ3zczeIelOM3t7WneNmb07w3aqma1dPK/cnls37Ksi6YfAaTYkAipjeyPwXfw3tY15hdpS+RqrIOkBvF7Yk/jvcnF8/A46/CZT5OlUhqbbNApp5oTqp+GClFPw3vnzhffOt8w8prkMpIqvS4qjgUPSFviT8dK4w3gU+LWZXVxiN2fiocG7GMrF1xbrkN4kcx+lVX/dGOupctwCC1khH5uZXSkpq2R7A/nkz5XN7CdJtbWwmd2f3h4xAavq5WlbAH/C3VCS4RlAjm9vNZs6c91eSL0eS+1YkUKvdQzsq7Ix8F+SqowlVy6fDiDP+zh7YnJuOBSvJTWlER5M98RNzex/M2xPwedvXZNsN8THknLOd0czG1bWQ9IKZnZ/t50TDGgPSl7+eg18bklxwuYsfDA4uyz3oCDpu7gg46e4kg18PtIeeKqjA0eybdrPtWa2YectW9ougEtrN8RvItcAJ+TEriUtw1AtpcZN80Azm9nWkHpyXElvwm/Or01ho9WAbS2jXk+d8bpkfyg+PrGKmb1J0hvwpKIdx8QkXYLnafsshTxt1jQpdATbc/DfQuNGuSuuuuo44XaEcd1vm9kNGbab4ZOpV8WVnhsAe5nZlZ1sk/3meE+kaJ+VtLUOdceSVb18+rfwuYVnplW74srkg0e2mm07R1iy0QPNsJ1jXDZ3rHaEqFXlEGnH4w2ig2qgguIohTSWtYx5SIOIpP9nZnNkAEiDvP/POmR8Lmy/Cf5DuIzyYzJ1bnyX4jWpijf73cysXSG8hu11uDNsVg92TLAr6Sp8TsuJhbBRVgimynhdk/1tePXjWwrHviPnybwQapu9vaSrzOw9GbaVlZp1kReHXA//vG6wNH9trOzrIOk1DO+t5pSg2QYXLM1nZitIWgPPpJ5TRuYOYA0zeyUtT8Sl3znXx8X4b+Jn+APfR4GNzGyLNjYNx7I7XhGhkV1lZ+BJM/tSG9tG8dCjGD5HbFHgc2b21k5trsJAhvgKXJq6yPPgc5j+kX7EB3WwG0Sek7SOmTUXcHsnkK2+wUMSb8bnyMwut0HeRNZVmm5yVyi/sFvpQngFXpXTc2hje5OGVw/NDQOviMf5J+DX2Cb4pNvcaQQvmJmlMBslw4N18rTdKmm9Rq9H0rrkTx69FA/jFMs4nN3uxtfE0ngGjHmAjeTZ/XMVppeZ1536bYt1o0a6hxyDT4J/DJ/PdA9+Q+7EYcw56bVMSHxxoCFmWKzdhk3siotZfoX/fq9O69pxTNNycYy3U0+lG8VDSzPoDmoxM3ta0ifwFEOHavis7vHEXsDxcslzIyy2LD75dK8S+1m9Mfhdgco3PuCfkj7K8EJ4uSU/LpS0tVWT4/4zjWU0nMQOdCg1X6D0eF0T50g6EVhc0ieBj5FXPgJq5GnDk5zuIS/HDZ4x5R4lWXaHJ/SlbM4Et1ljUJJOxZ333ZR4+Emh41cBSyWH2HiaWJShooujyRF4r+0PZrampI3pfLNvUKl8euKb+G/qCvycN8LLXrQl9bS+mBvWn92oDoVQO9j+BviNahYPLcugh/juxLX8p+Nqq5tzQyiDiqTX4U+pwicIdlRmNdmfDBxrmYlHm2zvwZ+kht348JtR2xufKhTCK9jWkeO+ETgJrwn1JJ7dYLfM41YeryvsYzMK802sm7VyRj5mW/k1ru4bKcXSNDxf4UOFff2qedxhBNs/mdmqFdp7IF6n7A24KKNxt38aONnMflB2nyWP31AP3g6saZ4A+iYzWyfDtvKk12T/ejwKIjzDQ9bvWdLlOYKZNvalqnJL+ryZHSXp+7RwwDZCCq26DHoP6nBch39tck5vxNPajEuSc8LMpiXhwLvlk1fLOJsNgT3l0vFSkxrxfGWVsGqF8Bq2i3TeakQ+iGfwvgIP1f0b2DSN8dzWwfZQST+mwnhdYdtLgdJOSTUqpnZyvvIS5yM5nC8B16axO/Cn+n0ym329pFXLPvyY2feA70n6tJl9v4xtl3hKXn34ajw/5mPkh4E/jX9mz+PRgYvxHlkuk9LficD6JUKit0qagmeQL5ZezxlLblmVu4NZN8qxlGage1CdkPRFM/tmr9vRDST9F/6UJjx/1154KGUDPHvGKZn7qatYmgi8luFquhEHk7v15KWKclx5aYG18Rn+wksq3IyPw51rbcq3S/pZ2m5YyCrHSST77fHv6jXp2I2HgZye3xxzpFqtq0Kn/UhaiiGhwvW5QgV57akL8HpdZR9+kJf1+L2ZzZJ0CO5Ev2Zmt+TYVyWNDT6Ht3c3fCzoTKtQdbrkcVuGRHOuL9Wb33eHDVXlXi055/PNbPMy7R8LBr0H1Ykd8TjveGB/vEu+IJ77byUz+1t6sr4Cn9vQkVxH1AoNz7BdHGNodwOq/eSlOeW4B8oL1HWU4wKvxhPLPpP2dSieu3EjXBU4ooOi3ngdad+V8gBSo5ZUBnM8KEh6s5n9uaD0ejT9nSxpcqaTOBVXiN1JtTG7KoUDa2NeXaDB6WVsJV3AnJ/nv/Dr/URrPwVjvSohUag9v69RzPE/8qkPjwOdSrm3Os9ieypFRzox3h1UV6s79pgXzew/+EV1XyNWnQaxx6obfCCu5Mt+srTuFMLbmuFy3NPx7O45Dqq5pMqLeK63ZyV1mgR6Q5WQVYG/V3ROUK1iah3+F1djNSu9SMfPGe94yMym1GhD6cKB3aBOTxfP/zeJIfHPzgxVbT6ZEcrIJCqFRFOb65TqqFKV++iybewG491Bjaf45Ssamvf1/sbKpICaMEZteBh/OqzCF/F4ead1I1FVjnsW7mh+k5a3AX6ewjqdbgx1xusApsrzy5XOA2g1akllMMeDm5l9Mv2trPQC/pxCqhdQbczukaR63BQ4Up7ZfCyu7To93TXNbKPC8gWSrjazjSR1ytl4Ou6kqoREf4PPg/oDJasXW4Wq3GZ2Vbv3R4vx7qDGUw9qe5LDteHZF16NP/2OBTPwEurZGbbVnbRUleS4qW1HSLoIdzYC9jWzRrhxtw7mlUUhiUXxisnF2H7unDOSQxrJiV7GyEKHRvqa7BRLqRfRri05bV4Qvy4qnS8VCgd2iTo93UkpBNpQPU4GlkrvdSp0WSckWnpuoKT3mdnlrb7rXHGGamS7r8J4d1DjJqt54QdwZPHCNLNHJB2AP0mNNg+l13zplcOjeDx+W3zMp8EsMuf1mNnPJV3JkBz3C7ly3GQ/renYuXa1yqbUHCfoxIgPXyqkWMJzrM2LZxzYILWrVZbrbVqsa5DlZOqer5n9R9J9wBbyvJPXmNmIVabrUrhRV+7p4g+H16Z2Cx/L2S/10DuNZ9UJiVaZG/gevG5eq+8690HiJ9SrYl6KgVbxpfDWx5lTz9+Nksl9iVrnwhrTuV+qUB9JFQrhtRi4H8Zoq7uqMhZzRlpdB4X3KqdYqtmmynkPk/2B+LhK40b5IeCk0ZKej6CEa1BGrTk/rvZsFFnMyuwi6Ud46Do7JCqfE9ioLLBQsnuRcuNmlVGNbPdVGPQe1BnAn3HFz+F4yKZqV72vkfQpPFHrihqeLWMR8rM51G3DsPpIksrUR1peUtnQwEH4HJw6A/e9oCdzRgpUTrEkz4V3KEMJga/Fc8vlCGNOJuU9BDCzO9KYVJaDwh82122o6iQdiVdRHhUHVafH1yZc9sYSc5lKh0StMCcwKTtXpvB76tDmting2oXqC9TJdl+aQXdQK5nZjpK2M7PT04+hTOmJQeIs4Hd4/LeoXps1QshmNDgJOMiG10c6Gc/S0InSoQEza0wQ3ar5qTT1nvuShnLRzEpJlkvS7rOrk2LpbHzC6ofT8m54VvVNM2zr5D0EP6figP/LjME4sqpNiq4dLqvpID+Bq2qXwfOQrodnZ2mXt7Dh3FbBw+WN8OI2+Heew2fwSb4H4BOSN8YrKowKg+6gGiGjp9LT/d9w2eW4wzzf1yzg7XXHRmpQpz7SgmZ2mSSl9h8m6RqGJ6wcieuYUxDQal1f0K05I2WFDoX9Hy1PsfQ0fjP6iuWnWFqyoPICzwmYVV6EenkPwR9ibpSXOQHPApI1v68mq9mc+QfbToi2VGG4ipMZKfRb2HdOCPhA3MncYGYby7ONt62fZqm+mryUy1qWKukmKX/ueP3y5oUdn8EfMhvTRW7MtC/FoDuok9LTziH408DCDK8PNa4wzxF2e1E1NMbMkPRlhpfMuL/N9kVKhwY0lHdwwXTDKCYRfVXZxo8hteeMVBQ6zMYqpljCM9TvApyTlnegkF28A/+N97LfLOkR/Nr4aO6Bzew7SQzTUFzubWa3Nt6vKa1vR+lJ0TXDZd0I/T5nZs9JQtL8aax2lUzb5rmBL5D/YF93ukgpBlokMTci6XL8yekmhufgGpWZ3E3HbtRHatxAGvWROt40VKEQnqQ98ZROazP8Rz0LL9GdnROvV6h6Mbs6taTqpFhqJOZtyJ4b+QspsY+FgAmNJ/Ru0U4YUnO/e+A32WGTos3sjDY2jZ5/y3CZmX2i2+1sOv6v8B7MZ/Cx2CfxJLVbZ9h+CT/HRqmODwG/sDZp4TQ0XWQnPOTbYFFgVctIrFuFgXZQSfXzE/yGdTIe8jl4NKWpvUZSy4J1NoYT6SQtCrxSRsVX83gftozihP2G6hWzu8nM1mnclNNN//pMBzWd6hNPS9Olwfec43QlF+EI+16VoUnRl1nmpOgULvtwIVy2CJ7jseMcOvm8vlYqz1Lin3RPWAzPY9hp7lXDZi2goby7ulNPVT2qYj7oIb6Pmdn35HMmXoM/UfwELxc9LjGzqyS9Fn9qA7jJzB4bi2NLejtecr6o4tvTzO7KsK1cCM/MfqmS5QH6hMOYs5jd8pm2dYQOdSaekuThyzM8hU673mo3Bt9zGLWnaas+KbpOuOyzhf8XwIUpZUQlQLWHU/MpGiNN05jjfM3sdkl3AZuPsvhnGIPuoBpjElvjBQtvV5OEaLwhaSc8h9aV+Pl/X9LnzOy8MTj8icyp4mvUWupEnUJ4VcoD9AOtitllUVPoUHniqSoUHezS4Hs/0+4LPAO4KYXcGuGyrBu4+QTyIn/UUJmTXtLyfM3sZUmvljRfbk+tLoPuoKalH8UKwBdT97pKFuVB4kvAOxu9pqTu+gMePx9t6qj4XtHwlDDLkf9EvL4NlQf4qqRjyE+f00vukvQRYKI8RcwBuPowixpChzoplipn2KZebyKHXj18tlPcfV3S7xgKl2ULO5IYo8EE4B3A67rQ3rq0+10+iDvSKQwfA+9KGLeZQXdQH8fjovPiA+lLAaf1skFjwISmkN7jjF2y2DoqvjqF8EqXB+gTKhezqyN0qDO/hhoZtqnRm2hQVVrfS8qGywpMYygrxEv4b+njXW9gd3k0vSYwFNodNQZdJNFqstr1ZQcZBwlJ38ZDMMX0/ndYycSRFY/dUPFtwHAV31NtDYfsqxbC+zKeTWAT4Iek8gBm9uXSJ9Ej5IUeFzKzpzO3Ly10UBdSLKl+0cFSg+9NtrOl9Wb2pvQwcq6ZbZBz7NGijjhjNIUdo0VOm1Uh3Vmltgy4g7qToclqayhNVjOznXvctFFF0ocpOAkz+1UHk24dd228R7A8Q73vtjcvdTmfnjzvWcfyAP2APLPJvnhGhGm40uo7ZvbtDNs/lr0xS9rGzC5I8vw5yBncTo7xIJoybFsXJod3konXkdZ3oW0j9twkLWkVs7W0O2d5NpT9GJ5W6njLzOVXh6rnq6Z0Z0CZdGelGfQQX53JagNLklz3QnZ9Jq48uov8sb7ahfAk/TdegvspM3te0qsk7WdmP8psQ69Y1cyelrQbcBHwBdxRdXRQVBA6WHdSLNUtOtiOTmNIlXMI1kE1J0XX4Ke4TLuRa3BX/OafW8izEjXPt066s9IMuoOaKa8M+WvgUklPMlSqelyhoSzGc7zFGGQxTvzDhirkZmHdKYT3STP7YWGfTybpdb87qHklzYun7PmBmb2o/OrHpYUO6k6KpbpFB9vR6dzrSOvr8CFSzw3AzB5NIaxu0M4pr2JmqxeWr5B0e5eO244651tHKFWagXZQZvah9O9hadLbYsDve9ik0eQyXOFzPj7ruxf5+A6V9OPUltzyAN0ohDdBkizFo9N4Tm49ql5yIvAAcDtwdVIuZo1BVRQ6dKMsd92ig5WpKa2vQ62eWw1hx62S1rOUTUXSuoxNZYI651tHKFWagR6DmtuQtBheWXcXfGLfL/DJrmOSzVzSz/C6N8PmyFibrM/qQs2dJAxZHjgBv1nuCzxsZmNVSbhrSJrHzEacjNkNoUPaT6UUS6NJvwoGJH0WL1uxGV4t4GPAWZZRh6qOsEPSPbgjbuTVnIynA3uFEsKUstQ831pCqdJtDQc1eMiTru6Mx66/MVpzEFocd3aRsrEkne9/4U+jwjOF/NjMXm5r2AeoZAaMLgkd6qRYWga/rjZgaOD+QDOb2ck22VcWG9SR1tcl9dw2T8e8OLfnVkfYkXrU7Xi6neqxDjXOt7RQqlY7w0ENDpLWxwdS343fOH5hZteM4fFPBo61CnNkVK8Q3kCiETJgmNmoznWRNA0Xn1xZ4aZ5KV57rBjC2c3MNsuwrSUTryKt7zWqkTMxY9+jkhy3DpLupYVQarSGHAZ6DGpuQtIDwFN4Qbl9SDm7GvLtsnLtimwI7CnpfsrPkSldCE/SOWa2U5pO0CrcNWZl7itSOgNGl4QOlVMsAZPMrBiWPU3SZzJt64oNauUQrErNnttoCjtGJXNGzfMtLZSqQzioweEB/Ma1BUNd8wZjVf68Y4bmNlQphHdg+vuBGsftJVUyYHRD6FAnxdI/JX2UoYngu+LtzqGuTLxyDsGaHEXFntsoCztGK7xV+XypIJSqQzioAcHM3tsHbajTjS9dCM/M/tqF4/aSC9M0iKPw+U/gob4RsUJm6hpCh8oplvAewA+AY/Eb5HWkyqkZ1O1N1MkhWIdaPTernjOxV9Q5371xodS8ZCYTrkOMQQ0Y8vLKvzezWZIOwfN8HWGFlDL9iCoUwmsx90sM5S4bq7lflUkO5lP4mKEB15CZKaCO0KFpP2VTLJ0OfMaGV5c9OkdtmbavNPjeSyR9D5/CUSX7+6gJO0ZL9VjzfMdUKBUOasBoDHYntdQ38ZvY/5nZuj1uWtCEpHPwTAE/S6t2BRY3s50ybOsIHeqkWJrjpjja8vBuSetrHL/VVIjcKRC1hB11VI9VqXm+lYVSVYgQ3+DRkFa/H38a/4287k7fo/KF8Iq2a1FQAPZ7jzFRJ1NAHaFDnRRLE1RI6pp6UFn3iRq9icbNfWrOcbqN1cv+Xjlcph6lWKp5vnWEUqUJBzV4PJLi/JsCR8qTp45VuY3KqEIhvILtV/D8ZI1tT5N0rpl9bTTa2kXqZAqoI3Sok2LpGOA6Sefh389OwNczbSsNvlt3cgiWpks9tzrCjtFMsTQHXTrfOkKp0oSDGjx2wi+So83sKUmvBz7X4zblUKcQ3q7Amo2xG0nfwn/UfemgCrL4eYE9JD2Ulpdj5LLizdQROtRJsfRTSVPx8KKA7UuEcyr1Jrokra9CN3pudYQdY50ct/b5jrVgKcagBgxJ6wF321BZ7UXwkM6NvW1ZeySdAhxTJXYtr1i6q6V0KkkZ9zMz60v5eacMAWV/5GWFDiPso22KpW5QdfBd0nvavV9UNo4nVCPl0NxCOKgBQ9KtwFpmsxOnTgCmWp/NOG9GNQrhSfo1XvfrUvzpdDM8E8VjMPqD6L2gjtAh2ZdKsdQN6gy+F/YxZjkE6/TcuiXsGEvVYw97qpWJEN/gMTurN4CZvSJpEL7HU4HdaSqEl8mv0qvBlV1qUz9TWeigEVIsjWJbgdqD78Ok9UBlaX0J6kyK7oqwY4znUHVjEviYEj2oAUPS+fgN+vi0aj9gYzPrlJWhp0i63MzGItvFuEDS3cAaeF68H5jZVZJub1IFjmTbmIrQ+LswcL6Zbd7JtmJbu9WbqCytr8tY9twKx+xlcty+y3bfikF48g6Gsy9wHHAIfjO4DK9Y2+9ULoQn6QO4QGA5/JodiIm6NaksdKBaiqU6dEsmXkdaX5kqPbcuhcvqpByqTA96qpUJBzV4fBvYtyAYWAKXBmfH+XtEnUJ438XrYN1pc0mX38yOwx9EGjwoKbcqcekUS3Xooky8jrS+DocB65BCx2Z2m6TlO9h0I1zWk+S4VDvfnhAOavBYzQrFwczLn/ddEbhmao5PPAzcNbc4pwathA5AjtDhaIZSLF1PSrHU9QYmujj4XkdaX4fSPTfrTs7EXiXH7UlPtQrhoAaPyjP9e4nqFcL7PHCRpKsY/kMek0KNvaCm0OF0PMVSowe2K/BTfA7daNCVwXcz+w/uoL5UkNZ3zFvYBSr33GqGy3qVHLdXPdXShEhiwJC0B/BFYNhMfzM7o61hj1G9QniXAM/QpAA0s6+OQlP7gjpCh1ZiilyBRV3qDL7XldZXRdKrcMc4W+6NJ2DOSerbM2FHVeqc71jT90/ewXBqzvTvJXUK4S05Wgq0PqaO0KFOiqXKdGHwvU4OwcrU7LmVDpd1S/VYlR72VEsTDmoASQ5pEJxSkTqF8P4gaXMzu2R0mtaXlBY6dCnFUh0Oo97ge50cgpVp1XOTlNtzqxIu62ly3JrnO6ZEiC8YEyRNxgvhvYuhQngHmNlDGbaNWlIvAC+m1eNaZq4KtaTU5RRLZZF0o5mtq0J5jjLhLkkH4L2m2/Fs/ZPxlFbvHrVG+3FvM7M1Us/tHakN0zKznAxMuKxBnfMda6IHFYwVRwB7Nok7jqV3E+8AAATkSURBVCZDHm9mo5bhuY8pLXQYbQeUQa3B95rS+jpU7rlVCZd1UfVYlZ70VKsQDioYK1ZrOCfwWjdl5PGStgU2SotXmtmF3W5gn1GnllSvqC0TryGtr0PlSdEVw2W9TjlUZxL4mBIhvmBMSDfX9zb1oK6yjPLR8vIa7wTOTKt2xUMSB49We3uNpNOAE5qEDnua2X49bVgmqpCBfSRpvZl9fHRa2bYtWdnf64bL6qgeu0nu+Y410YMKxoo6hfC2BtYws1cAJJ0O3AqMOwfVB0KHynRh8H39grT+q5KOYfTnBAG1em6Vw2VdUD1Wpkc91dKEgwrGhC7I4xcHGiWwF+t2+/qIvqxxlUldmfhY5xAEak+KrhMuO4wepByqeb5jSjioYMyoIY//Jj635wrcuW2ET1Yed/SB0KEOdQffxzSHYIHKPbeawo5epRzqWU+1LOGggr7HzH4u6Up8HErAF8zsb71tVdCCuoPvY5pDsECtnluNcFmvUg71pKdahQm9bkAQjISkLSTtAGBmfzWzKWb2G2ATeSXSoI8ws+PMbGkz29qcB/EwUi6n4zf64/C8jW/BpfWjTXPP7QHg7BzDFC7bGVcwCtgRHy/M4dP4+TZUj08DudlV6lD5fMeaUPEFfYukG/B6Of9oWv864Fdm9q7etCwYiVa9CcssNa8e5RCsMim6YNuV4pBVVI9VqXO+Y030oIJ+5lXNzgkghfcW6kF7gjbU7E1AyiFY2N+Y5BCkXs+tOVz2IpnhMklnSVpU0kLA3cC9kj5XquXV6FVPtTQxBhX0Mwu0mp+RBuIX7FGbgpGpNPjeB9L6OpOi6wg7epIclwGaBB4OKuhnzgdOlrS/mf0bID1tHkefqo7mcqoOvvdaWl8n+3sdYUevUg71JNt9FcJBBf3MIcDXcOluQ349GTgF+HLPWhWMRKXeRK+k9V3qudUpDjmmKYf6oKdamhBJBH1PGtRdKS1ON7Nn220f9IZBGnyH7mR/77awYzRTDnXjfMea6EEFg8AHgN+b2SxJh0haC/iamd3S64YFwxjrUvO16NINuVa4bCxTDvWjA+pEOKhgEPiymZ0raUNgCzzufzywbm+bFTQxMIPvdelGuGyQUg71inBQwSDwcvr7fjxk9BtJh/WwPUFrBmbwvQt0Q9gxMCmHekU4qGAQeETSicCmwJGS5ifm8PUNgzj4XpcuhcsGJuVQrwgHFQwCOwFbAkeb2VOSXg+MxYTGII9ey8QHlV4lxx0YQsUX9D0pu8DdZjYrLS+CT3K8sbctC4LqDJrqsReEgwr6Hkm3AmtZulglTQCmmtlavW1ZEFRH0jm46vFnadWuwOJm1peqx14QIb5gEJAVnqTM7BVJce0Gg85co3qsSgw0B4PADEkHSJo3vQ4EZvS6UUFQk14lxx0YIsQX9D2SXoNP/nwfHqu/DDiwVabzIOh3mlSPqwDDVI9m9rYeNq+vCAcV9D2STscd0lNpeQngGDP7WG9bFgTlGcSUQ70i4vjBILBawzkBmNmTktbsZYOCoCrhgPKJMahgEJiQek0ASFqSeLgKgnFP/MiDQeAY4DpJ5+Gx+p2Ar/e2SUEQjDYxBhUMBJJWxUUSAi4zs3GZQicIgiHCQQVBEAR9SYxBBUEQBH1JOKggCIKgLwkHFQRBEPQl4aCCIAiCvuT/A3qgiCjEC0A8AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "importances = forest.feature_importances_\n",
    "\n",
    "indices = np.argsort(importances)[::-1]\n",
    "\n",
    "for f in range(features_train.shape[1]):\n",
    "    print(\"%2d) %-*s %f\" % (f + 1, 30, \n",
    "                            feat_labels[indices[f]], \n",
    "                            importances[indices[f]]))\n",
    "\n",
    "plt.title('Важность признаков')\n",
    "plt.bar(range(features_train.shape[1]), \n",
    "        importances[indices],\n",
    "        color='green', \n",
    "        align='center')\n",
    "\n",
    "plt.xticks(range(features_train.shape[1]), \n",
    "           feat_labels[indices], rotation=90)\n",
    "plt.xlim([-1, features_train.shape[1]])\n",
    "plt.tight_layout()\n",
    "plt.savefig('random_forest.png', dpi=300)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8701742398360096"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(forest.predict(features_test), labels_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.89      0.95      0.92      2237\n",
      "          1       0.79      0.61      0.69       690\n",
      "\n",
      "avg / total       0.86      0.87      0.86      2927\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(labels_test, forest.predict(features_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred_rf = forest.predict_proba(features_test)[:,1]\n",
    "fpr, tpr, thresholds = roc_curve(labels_test, y_pred_rf)\n",
    "roc_auc = metrics.auc(fpr, tpr)\n",
    "\n",
    "plt.title('Receiver Operating Characteristic')\n",
    "plt.plot(fpr, tpr, 'b', label = 'RandomForestClassifier - AUC = %0.2f' % roc_auc)\n",
    "plt.legend(loc = 'lower right')\n",
    "plt.plot([0, 1], [0, 1],'g--')\n",
    "plt.xlim([0, 1])\n",
    "plt.ylim([0, 1])\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_selection import SelectFromModel\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "select = SelectFromModel(\n",
    "    RandomForestClassifier(n_estimators=100, random_state=42),\n",
    "    threshold=\"median\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train.shape: (11708, 24)\n",
      "X_train_l1.shape: (11708, 12)\n"
     ]
    }
   ],
   "source": [
    "select.fit(features_train, labels_train)\n",
    "X_train_l1 = select.transform(features_train)\n",
    "print(\"X_train.shape: {}\".format(features_train.shape))\n",
    "print(\"X_train_l1.shape: {}\".format(X_train_l1.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Признаки, отобранные SelectFromModel с помощью RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([], <a list of 0 Text yticklabel objects>)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x144 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mask = select.get_support()\n",
    "# visualize the mask. black is True, white is False\n",
    "plt.matshow(mask.reshape(1, -1), cmap='gray_r')\n",
    "plt.xlabel(\"Sample index\")\n",
    "plt.yticks(())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Признаки, отобранные методом рекурсивного исключения признаков с помощью RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([], <a list of 0 Text yticklabel objects>)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x144 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.feature_selection import RFE\n",
    "select = RFE(RandomForestClassifier(n_estimators=100, random_state=42),\n",
    "             n_features_to_select=12)\n",
    "\n",
    "select.fit(features_train, labels_train)\n",
    "# visualize the selected features:\n",
    "mask = select.get_support()\n",
    "plt.matshow(mask.reshape(1, -1), cmap='gray_r')\n",
    "plt.xlabel(\"Sample index\")\n",
    "plt.yticks(())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 0.889\n"
     ]
    }
   ],
   "source": [
    "print(\"Test score: {:.3f}\".format(select.score(features_test, labels_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.848650495387769"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree = DecisionTreeClassifier(max_depth=5)\n",
    "tree.fit( features_train, labels_train );\n",
    "accuracy_score(tree.predict(features_test), labels_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.86      0.96      0.91      2237\n",
      "          1       0.78      0.50      0.61       690\n",
      "\n",
      "avg / total       0.84      0.85      0.84      2927\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(labels_test, tree.predict(features_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree_params = {'max_depth': np.arange(5,10), \n",
    "               'max_features': [0.5, 0.6, 0.7, 0.8, 0.9, 1]}\n",
    "tree_grid = GridSearchCV(tree, tree_params, \n",
    "                         cv=5, \n",
    "                         n_jobs=-1)\n",
    "tree_grid.fit(features_train, labels_train);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.8853775196446874, {'max_depth': 9, 'max_features': 0.9})"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tree_grid.best_score_, tree_grid.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#\n",
    "# forest_params = {'max_depth': np.arange(5,10), \n",
    "#                'max_features': [0.5, 0.6, 0.7, 0.8, 0.9, 1]}\n",
    "# forest_grid = GridSearchCV(forest, forest_params, \n",
    "#                          cv=5, \n",
    "#                          n_jobs=-1)\n",
    "# forest_grid.fit(features_train, labels_train);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# forest_grid.best_score_, forest_grid.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'mean_fit_time': array([0.01620102, 0.01840105, 0.02140121, 0.02960172, 0.03300185,\n",
      "       0.00700049, 0.02060118, 0.02560158, 0.0260015 , 0.03180189,\n",
      "       0.03860216, 0.00940056, 0.02840166, 0.03460193, 0.04320245,\n",
      "       0.04940281, 0.06560378, 0.01180072, 0.03680224, 0.0450026 ,\n",
      "       0.0470027 , 0.05420303, 0.0598033 , 0.01380081, 0.0470027 ,\n",
      "       0.04620275, 0.04680266, 0.04480262, 0.04420242, 0.01220059]), 'std_fit_time': array([0.00039995, 0.00135651, 0.00249806, 0.00407935, 0.00400028,\n",
      "       0.00126501, 0.0020591 , 0.00338238, 0.00451689, 0.00444543,\n",
      "       0.00320015, 0.0010199 , 0.00546303, 0.00403002, 0.00696892,\n",
      "       0.00293957, 0.02476122, 0.00097975, 0.00193917, 0.00424286,\n",
      "       0.00063249, 0.00172047, 0.00231525, 0.00614537, 0.00905588,\n",
      "       0.00552845, 0.0082322 , 0.00910884, 0.01068513, 0.00097986]), 'mean_score_time': array([0.00120006, 0.00180011, 0.00120015, 0.00100002, 0.00140014,\n",
      "       0.00140004, 0.00120006, 0.00140009, 0.00100007, 0.0012001 ,\n",
      "       0.00120015, 0.00160003, 0.00220013, 0.00160007, 0.00360017,\n",
      "       0.00200009, 0.00200009, 0.00180011, 0.00200005, 0.00200014,\n",
      "       0.00200009, 0.00180001, 0.00200009, 0.00160012, 0.00220017,\n",
      "       0.00180006, 0.00160017, 0.0012001 , 0.00160003, 0.00140014]), 'std_score_time': array([4.00137912e-04, 4.00090228e-04, 4.00090257e-04, 1.16800773e-07,\n",
      "       4.89959812e-04, 4.90037648e-04, 4.00018735e-04, 4.89901382e-04,\n",
      "       1.90734863e-07, 3.99994861e-04, 3.99971037e-04, 4.89881921e-04,\n",
      "       1.46984040e-03, 4.89920871e-04, 4.22393979e-03, 0.00000000e+00,\n",
      "       6.32409710e-04, 3.99971037e-04, 9.53674316e-08, 9.53674316e-08,\n",
      "       0.00000000e+00, 3.99923339e-04, 6.32560481e-04, 4.89959789e-04,\n",
      "       4.00042545e-04, 4.00066376e-04, 4.89901382e-04, 3.99994861e-04,\n",
      "       4.89881921e-04, 4.89959789e-04]), 'param_max_depth': masked_array(data=[5, 5, 5, 5, 5, 5, 6, 6, 6, 6, 6, 6, 7, 7, 7, 7, 7, 7,\n",
      "                   8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9],\n",
      "             mask=[False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False],\n",
      "       fill_value='?',\n",
      "            dtype=object), 'param_max_features': masked_array(data=[0.5, 0.6, 0.7, 0.8, 0.9, 1, 0.5, 0.6, 0.7, 0.8, 0.9, 1,\n",
      "                   0.5, 0.6, 0.7, 0.8, 0.9, 1, 0.5, 0.6, 0.7, 0.8, 0.9, 1,\n",
      "                   0.5, 0.6, 0.7, 0.8, 0.9, 1],\n",
      "             mask=[False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False],\n",
      "       fill_value='?',\n",
      "            dtype=object), 'params': [{'max_depth': 5, 'max_features': 0.5}, {'max_depth': 5, 'max_features': 0.6}, {'max_depth': 5, 'max_features': 0.7}, {'max_depth': 5, 'max_features': 0.8}, {'max_depth': 5, 'max_features': 0.9}, {'max_depth': 5, 'max_features': 1}, {'max_depth': 6, 'max_features': 0.5}, {'max_depth': 6, 'max_features': 0.6}, {'max_depth': 6, 'max_features': 0.7}, {'max_depth': 6, 'max_features': 0.8}, {'max_depth': 6, 'max_features': 0.9}, {'max_depth': 6, 'max_features': 1}, {'max_depth': 7, 'max_features': 0.5}, {'max_depth': 7, 'max_features': 0.6}, {'max_depth': 7, 'max_features': 0.7}, {'max_depth': 7, 'max_features': 0.8}, {'max_depth': 7, 'max_features': 0.9}, {'max_depth': 7, 'max_features': 1}, {'max_depth': 8, 'max_features': 0.5}, {'max_depth': 8, 'max_features': 0.6}, {'max_depth': 8, 'max_features': 0.7}, {'max_depth': 8, 'max_features': 0.8}, {'max_depth': 8, 'max_features': 0.9}, {'max_depth': 8, 'max_features': 1}, {'max_depth': 9, 'max_features': 0.5}, {'max_depth': 9, 'max_features': 0.6}, {'max_depth': 9, 'max_features': 0.7}, {'max_depth': 9, 'max_features': 0.8}, {'max_depth': 9, 'max_features': 0.9}, {'max_depth': 9, 'max_features': 1}], 'split0_test_score': array([0.81084543, 0.80572161, 0.81212639, 0.82322801, 0.83134073,\n",
      "       0.78095645, 0.79632792, 0.82194705, 0.82877882, 0.8206661 ,\n",
      "       0.82322801, 0.78650726, 0.82536294, 0.8206661 , 0.83091375,\n",
      "       0.82877882, 0.84158839, 0.77796755, 0.85610589, 0.83988044,\n",
      "       0.84500427, 0.87233134, 0.8616567 , 0.7912041 , 0.85226302,\n",
      "       0.83731853, 0.85311699, 0.87830914, 0.86122972, 0.76686593]), 'split1_test_score': array([0.82749787, 0.83091375, 0.83091375, 0.83048676, 0.82920581,\n",
      "       0.77369769, 0.82920581, 0.82963279, 0.82408198, 0.84628523,\n",
      "       0.83902647, 0.79974381, 0.82835184, 0.823655  , 0.84073442,\n",
      "       0.84329633, 0.86379163, 0.78437233, 0.84927412, 0.8616567 ,\n",
      "       0.84927412, 0.86421862, 0.84628523, 0.75960717, 0.85183604,\n",
      "       0.86549957, 0.84671221, 0.8766012 , 0.88684885, 0.79632792]), 'split2_test_score': array([0.80059778, 0.80273271, 0.79675491, 0.80657558, 0.80828352,\n",
      "       0.76601196, 0.80273271, 0.80999146, 0.80529462, 0.83347566,\n",
      "       0.81212639, 0.74679761, 0.83304868, 0.83988044, 0.83005978,\n",
      "       0.81554227, 0.81938514, 0.77156277, 0.81084543, 0.8501281 ,\n",
      "       0.8616567 , 0.84500427, 0.85610589, 0.76515798, 0.83390265,\n",
      "       0.85824082, 0.84500427, 0.85824082, 0.89069172, 0.75619129]), 'split3_test_score': array([0.81631781, 0.84408372, 0.8111918 , 0.82571551, 0.82357967,\n",
      "       0.765912  , 0.82785135, 0.84237505, 0.81802648, 0.82785135,\n",
      "       0.84280222, 0.74156343, 0.84109355, 0.83212302, 0.84023921,\n",
      "       0.83724904, 0.86715079, 0.7885519 , 0.83895771, 0.83938488,\n",
      "       0.86544212, 0.85049124, 0.88423751, 0.75950449, 0.88936352,\n",
      "       0.85946177, 0.86373345, 0.85390859, 0.89619821, 0.77744554]), 'split4_test_score': array([0.81033746, 0.821871  , 0.84451089, 0.82699701, 0.84322939,\n",
      "       0.76035882, 0.84792824, 0.84750107, 0.83810337, 0.83169586,\n",
      "       0.85946177, 0.78257155, 0.82656984, 0.84750107, 0.82016232,\n",
      "       0.86416062, 0.85732593, 0.78812473, 0.82827851, 0.85988894,\n",
      "       0.85476292, 0.85519009, 0.86544212, 0.77488253, 0.84536523,\n",
      "       0.8709953 , 0.87697565, 0.8816745 , 0.89192653, 0.79026057]), 'mean_test_score': array([0.81311923, 0.82106252, 0.81909805, 0.82259993, 0.82712675,\n",
      "       0.76938845, 0.82080629, 0.83028698, 0.82285617, 0.83199522,\n",
      "       0.83532627, 0.77143833, 0.83088487, 0.83276392, 0.83242228,\n",
      "       0.83780321, 0.84984626, 0.78211479, 0.83669286, 0.85018791,\n",
      "       0.8552272 , 0.8574479 , 0.86274342, 0.77007175, 0.8545439 ,\n",
      "       0.85830202, 0.85710625, 0.86974718, 0.88537752, 0.77741715]), 'std_test_score': array([0.00879465, 0.01548535, 0.01670741, 0.00834823, 0.01139315,\n",
      "       0.00717513, 0.01887643, 0.01359637, 0.01096161, 0.00839117,\n",
      "       0.0163503 , 0.02303004, 0.00573415, 0.00997558, 0.00759366,\n",
      "       0.01613759, 0.01758437, 0.00649923, 0.015996  , 0.00947188,\n",
      "       0.00756109, 0.00975453, 0.01253254, 0.01195913, 0.0186279 ,\n",
      "       0.01144222, 0.01190844, 0.01136438, 0.01243835, 0.01472788]), 'rank_test_score': array([25, 22, 24, 21, 19, 30, 23, 18, 20, 16, 13, 28, 17, 14, 15, 11, 10,\n",
      "       26, 12,  9,  7,  5,  3, 29,  8,  4,  6,  2,  1, 27]), 'split0_train_score': array([0.82543242, 0.81774504, 0.82596626, 0.83301303, 0.84689302,\n",
      "       0.78571429, 0.81913303, 0.84080717, 0.83162503, 0.8376041 ,\n",
      "       0.83258595, 0.79233397, 0.83610933, 0.85554132, 0.84539825,\n",
      "       0.8559684 , 0.86034593, 0.80354474, 0.87903054, 0.87678838,\n",
      "       0.87166346, 0.90326714, 0.88255392, 0.80674781, 0.87860346,\n",
      "       0.87102285, 0.88426223, 0.91212898, 0.89344437, 0.80365151]), 'split1_train_score': array([0.82639334, 0.82489857, 0.82553918, 0.82169549, 0.82457826,\n",
      "       0.77589152, 0.83119795, 0.82831518, 0.83098441, 0.8445441 ,\n",
      "       0.84497117, 0.80044843, 0.83568225, 0.83749733, 0.85650224,\n",
      "       0.84924194, 0.87091608, 0.78475336, 0.85361947, 0.87486654,\n",
      "       0.86717916, 0.87593423, 0.86547085, 0.78518044, 0.88041854,\n",
      "       0.89205637, 0.88276746, 0.89291053, 0.90903267, 0.81048473]), 'split2_train_score': array([0.82628657, 0.83162503, 0.8275678 , 0.83162503, 0.83301303,\n",
      "       0.77386291, 0.83023703, 0.83536195, 0.83792441, 0.85287209,\n",
      "       0.83728379, 0.7635063 , 0.86429639, 0.85159086, 0.85724963,\n",
      "       0.84411702, 0.84294256, 0.78261798, 0.84465086, 0.88426223,\n",
      "       0.88426223, 0.86814008, 0.88693145, 0.78966474, 0.88415546,\n",
      "       0.90049114, 0.87988469, 0.8930173 , 0.91202221, 0.79062567]), 'split3_train_score': array([0.82150101, 0.84317284, 0.8048468 , 0.82000641, 0.82427672,\n",
      "       0.76321127, 0.83271058, 0.84370663, 0.81904559, 0.83602007,\n",
      "       0.8501121 , 0.7525355 , 0.84221202, 0.84082417, 0.84701612,\n",
      "       0.84349311, 0.87498666, 0.79363724, 0.85214049, 0.84648233,\n",
      "       0.87947048, 0.85972029, 0.89516387, 0.7773033 , 0.89697876,\n",
      "       0.88011103, 0.88267321, 0.86431088, 0.91768976, 0.79235614]), 'split4_train_score': array([0.81915234, 0.82171453, 0.84071741, 0.82651863, 0.84317284,\n",
      "       0.75552471, 0.85523647, 0.85192698, 0.84189175, 0.83036191,\n",
      "       0.86762037, 0.78466958, 0.83559304, 0.86238924, 0.82982812,\n",
      "       0.8756272 , 0.86591225, 0.79417103, 0.84787018, 0.87808263,\n",
      "       0.87338529, 0.87893669, 0.88107185, 0.7876588 , 0.86270951,\n",
      "       0.90103555, 0.90541262, 0.91427351, 0.92057222, 0.80281841]), 'mean_train_score': array([0.82375314, 0.8278312 , 0.82492749, 0.82657172, 0.83438677,\n",
      "       0.77084094, 0.83370301, 0.84002358, 0.83229424, 0.84028045,\n",
      "       0.84651468, 0.77869876, 0.84277861, 0.84956858, 0.84719887,\n",
      "       0.85368953, 0.86302069, 0.79174487, 0.85546231, 0.87209642,\n",
      "       0.87519212, 0.87719969, 0.88223839, 0.78931102, 0.88057314,\n",
      "       0.88894339, 0.88700004, 0.89532824, 0.91055225, 0.79998729]), 'std_train_score': array([0.0029137 , 0.00891632, 0.01149765, 0.00517513, 0.00931567,\n",
      "       0.01047526, 0.01179192, 0.00793331, 0.00776532, 0.00775138,\n",
      "       0.01216547, 0.01794405, 0.0110436 , 0.00923377, 0.00992561,\n",
      "       0.01184766, 0.0111729 , 0.00749514, 0.0122018 , 0.01318796,\n",
      "       0.00600824, 0.0146406 , 0.00971292, 0.00967619, 0.01099991,\n",
      "       0.01173386, 0.00931432, 0.01796914, 0.00947094, 0.00745   ])}\n"
     ]
    }
   ],
   "source": [
    "print(tree_grid.cv_results_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:122: FutureWarning: You are accessing a training score ('mean_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
      "  warnings.warn(*warn_args, **warn_kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:122: FutureWarning: You are accessing a training score ('split0_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
      "  warnings.warn(*warn_args, **warn_kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:122: FutureWarning: You are accessing a training score ('split1_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
      "  warnings.warn(*warn_args, **warn_kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:122: FutureWarning: You are accessing a training score ('split2_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
      "  warnings.warn(*warn_args, **warn_kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:122: FutureWarning: You are accessing a training score ('split3_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
      "  warnings.warn(*warn_args, **warn_kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:122: FutureWarning: You are accessing a training score ('split4_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
      "  warnings.warn(*warn_args, **warn_kwargs)\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\utils\\deprecation.py:122: FutureWarning: You are accessing a training score ('std_train_score'), which will not be available by default any more in 0.21. If you need training scores, please set return_train_score=True\n",
      "  warnings.warn(*warn_args, **warn_kwargs)\n"
     ]
    },
    {
     "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>mean_fit_time</th>\n",
       "      <th>mean_score_time</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>mean_train_score</th>\n",
       "      <th>param_max_depth</th>\n",
       "      <th>param_max_features</th>\n",
       "      <th>params</th>\n",
       "      <th>rank_test_score</th>\n",
       "      <th>split0_test_score</th>\n",
       "      <th>split0_train_score</th>\n",
       "      <th>...</th>\n",
       "      <th>split2_test_score</th>\n",
       "      <th>split2_train_score</th>\n",
       "      <th>split3_test_score</th>\n",
       "      <th>split3_train_score</th>\n",
       "      <th>split4_test_score</th>\n",
       "      <th>split4_train_score</th>\n",
       "      <th>std_fit_time</th>\n",
       "      <th>std_score_time</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>std_train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.016201</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.813119</td>\n",
       "      <td>0.823753</td>\n",
       "      <td>5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.5}</td>\n",
       "      <td>25</td>\n",
       "      <td>0.810845</td>\n",
       "      <td>0.825432</td>\n",
       "      <td>...</td>\n",
       "      <td>0.800598</td>\n",
       "      <td>0.826287</td>\n",
       "      <td>0.816318</td>\n",
       "      <td>0.821501</td>\n",
       "      <td>0.810337</td>\n",
       "      <td>0.819152</td>\n",
       "      <td>0.000400</td>\n",
       "      <td>4.001379e-04</td>\n",
       "      <td>0.008795</td>\n",
       "      <td>0.002914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.018401</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>0.821063</td>\n",
       "      <td>0.827831</td>\n",
       "      <td>5</td>\n",
       "      <td>0.6</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.6}</td>\n",
       "      <td>22</td>\n",
       "      <td>0.805722</td>\n",
       "      <td>0.817745</td>\n",
       "      <td>...</td>\n",
       "      <td>0.802733</td>\n",
       "      <td>0.831625</td>\n",
       "      <td>0.844084</td>\n",
       "      <td>0.843173</td>\n",
       "      <td>0.821871</td>\n",
       "      <td>0.821715</td>\n",
       "      <td>0.001357</td>\n",
       "      <td>4.000902e-04</td>\n",
       "      <td>0.015485</td>\n",
       "      <td>0.008916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.021401</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.819098</td>\n",
       "      <td>0.824927</td>\n",
       "      <td>5</td>\n",
       "      <td>0.7</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.7}</td>\n",
       "      <td>24</td>\n",
       "      <td>0.812126</td>\n",
       "      <td>0.825966</td>\n",
       "      <td>...</td>\n",
       "      <td>0.796755</td>\n",
       "      <td>0.827568</td>\n",
       "      <td>0.811192</td>\n",
       "      <td>0.804847</td>\n",
       "      <td>0.844511</td>\n",
       "      <td>0.840717</td>\n",
       "      <td>0.002498</td>\n",
       "      <td>4.000903e-04</td>\n",
       "      <td>0.016707</td>\n",
       "      <td>0.011498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.029602</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>0.822600</td>\n",
       "      <td>0.826572</td>\n",
       "      <td>5</td>\n",
       "      <td>0.8</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.8}</td>\n",
       "      <td>21</td>\n",
       "      <td>0.823228</td>\n",
       "      <td>0.833013</td>\n",
       "      <td>...</td>\n",
       "      <td>0.806576</td>\n",
       "      <td>0.831625</td>\n",
       "      <td>0.825716</td>\n",
       "      <td>0.820006</td>\n",
       "      <td>0.826997</td>\n",
       "      <td>0.826519</td>\n",
       "      <td>0.004079</td>\n",
       "      <td>1.168008e-07</td>\n",
       "      <td>0.008348</td>\n",
       "      <td>0.005175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.033002</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>0.827127</td>\n",
       "      <td>0.834387</td>\n",
       "      <td>5</td>\n",
       "      <td>0.9</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.9}</td>\n",
       "      <td>19</td>\n",
       "      <td>0.831341</td>\n",
       "      <td>0.846893</td>\n",
       "      <td>...</td>\n",
       "      <td>0.808284</td>\n",
       "      <td>0.833013</td>\n",
       "      <td>0.823580</td>\n",
       "      <td>0.824277</td>\n",
       "      <td>0.843229</td>\n",
       "      <td>0.843173</td>\n",
       "      <td>0.004000</td>\n",
       "      <td>4.899598e-04</td>\n",
       "      <td>0.011393</td>\n",
       "      <td>0.009316</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   mean_fit_time  mean_score_time  mean_test_score  mean_train_score  \\\n",
       "0       0.016201           0.0012         0.813119          0.823753   \n",
       "1       0.018401           0.0018         0.821063          0.827831   \n",
       "2       0.021401           0.0012         0.819098          0.824927   \n",
       "3       0.029602           0.0010         0.822600          0.826572   \n",
       "4       0.033002           0.0014         0.827127          0.834387   \n",
       "\n",
       "  param_max_depth param_max_features                                 params  \\\n",
       "0               5                0.5  {'max_depth': 5, 'max_features': 0.5}   \n",
       "1               5                0.6  {'max_depth': 5, 'max_features': 0.6}   \n",
       "2               5                0.7  {'max_depth': 5, 'max_features': 0.7}   \n",
       "3               5                0.8  {'max_depth': 5, 'max_features': 0.8}   \n",
       "4               5                0.9  {'max_depth': 5, 'max_features': 0.9}   \n",
       "\n",
       "   rank_test_score  split0_test_score  split0_train_score       ...         \\\n",
       "0               25           0.810845            0.825432       ...          \n",
       "1               22           0.805722            0.817745       ...          \n",
       "2               24           0.812126            0.825966       ...          \n",
       "3               21           0.823228            0.833013       ...          \n",
       "4               19           0.831341            0.846893       ...          \n",
       "\n",
       "   split2_test_score  split2_train_score  split3_test_score  \\\n",
       "0           0.800598            0.826287           0.816318   \n",
       "1           0.802733            0.831625           0.844084   \n",
       "2           0.796755            0.827568           0.811192   \n",
       "3           0.806576            0.831625           0.825716   \n",
       "4           0.808284            0.833013           0.823580   \n",
       "\n",
       "   split3_train_score  split4_test_score  split4_train_score  std_fit_time  \\\n",
       "0            0.821501           0.810337            0.819152      0.000400   \n",
       "1            0.843173           0.821871            0.821715      0.001357   \n",
       "2            0.804847           0.844511            0.840717      0.002498   \n",
       "3            0.820006           0.826997            0.826519      0.004079   \n",
       "4            0.824277           0.843229            0.843173      0.004000   \n",
       "\n",
       "   std_score_time  std_test_score  std_train_score  \n",
       "0    4.001379e-04        0.008795         0.002914  \n",
       "1    4.000902e-04        0.015485         0.008916  \n",
       "2    4.000903e-04        0.016707         0.011498  \n",
       "3    1.168008e-07        0.008348         0.005175  \n",
       "4    4.899598e-04        0.011393         0.009316  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "# convert to Dataframe\n",
    "results = pd.DataFrame(tree_grid.cv_results_)\n",
    "# show the first 5 rows\n",
    "display(results.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean_fit_time</th>\n",
       "      <th>mean_score_time</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>mean_train_score</th>\n",
       "      <th>param_max_depth</th>\n",
       "      <th>param_max_features</th>\n",
       "      <th>params</th>\n",
       "      <th>rank_test_score</th>\n",
       "      <th>split0_test_score</th>\n",
       "      <th>split0_train_score</th>\n",
       "      <th>...</th>\n",
       "      <th>split2_test_score</th>\n",
       "      <th>split2_train_score</th>\n",
       "      <th>split3_test_score</th>\n",
       "      <th>split3_train_score</th>\n",
       "      <th>split4_test_score</th>\n",
       "      <th>split4_train_score</th>\n",
       "      <th>std_fit_time</th>\n",
       "      <th>std_score_time</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>std_train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.016201</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.813119</td>\n",
       "      <td>0.823753</td>\n",
       "      <td>5</td>\n",
       "      <td>0.5</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.5}</td>\n",
       "      <td>25</td>\n",
       "      <td>0.810845</td>\n",
       "      <td>0.825432</td>\n",
       "      <td>...</td>\n",
       "      <td>0.800598</td>\n",
       "      <td>0.826287</td>\n",
       "      <td>0.816318</td>\n",
       "      <td>0.821501</td>\n",
       "      <td>0.810337</td>\n",
       "      <td>0.819152</td>\n",
       "      <td>0.000400</td>\n",
       "      <td>4.001379e-04</td>\n",
       "      <td>0.008795</td>\n",
       "      <td>0.002914</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.018401</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>0.821063</td>\n",
       "      <td>0.827831</td>\n",
       "      <td>5</td>\n",
       "      <td>0.6</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.6}</td>\n",
       "      <td>22</td>\n",
       "      <td>0.805722</td>\n",
       "      <td>0.817745</td>\n",
       "      <td>...</td>\n",
       "      <td>0.802733</td>\n",
       "      <td>0.831625</td>\n",
       "      <td>0.844084</td>\n",
       "      <td>0.843173</td>\n",
       "      <td>0.821871</td>\n",
       "      <td>0.821715</td>\n",
       "      <td>0.001357</td>\n",
       "      <td>4.000902e-04</td>\n",
       "      <td>0.015485</td>\n",
       "      <td>0.008916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.021401</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.819098</td>\n",
       "      <td>0.824927</td>\n",
       "      <td>5</td>\n",
       "      <td>0.7</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.7}</td>\n",
       "      <td>24</td>\n",
       "      <td>0.812126</td>\n",
       "      <td>0.825966</td>\n",
       "      <td>...</td>\n",
       "      <td>0.796755</td>\n",
       "      <td>0.827568</td>\n",
       "      <td>0.811192</td>\n",
       "      <td>0.804847</td>\n",
       "      <td>0.844511</td>\n",
       "      <td>0.840717</td>\n",
       "      <td>0.002498</td>\n",
       "      <td>4.000903e-04</td>\n",
       "      <td>0.016707</td>\n",
       "      <td>0.011498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.029602</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>0.822600</td>\n",
       "      <td>0.826572</td>\n",
       "      <td>5</td>\n",
       "      <td>0.8</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.8}</td>\n",
       "      <td>21</td>\n",
       "      <td>0.823228</td>\n",
       "      <td>0.833013</td>\n",
       "      <td>...</td>\n",
       "      <td>0.806576</td>\n",
       "      <td>0.831625</td>\n",
       "      <td>0.825716</td>\n",
       "      <td>0.820006</td>\n",
       "      <td>0.826997</td>\n",
       "      <td>0.826519</td>\n",
       "      <td>0.004079</td>\n",
       "      <td>1.168008e-07</td>\n",
       "      <td>0.008348</td>\n",
       "      <td>0.005175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.033002</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>0.827127</td>\n",
       "      <td>0.834387</td>\n",
       "      <td>5</td>\n",
       "      <td>0.9</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 0.9}</td>\n",
       "      <td>19</td>\n",
       "      <td>0.831341</td>\n",
       "      <td>0.846893</td>\n",
       "      <td>...</td>\n",
       "      <td>0.808284</td>\n",
       "      <td>0.833013</td>\n",
       "      <td>0.823580</td>\n",
       "      <td>0.824277</td>\n",
       "      <td>0.843229</td>\n",
       "      <td>0.843173</td>\n",
       "      <td>0.004000</td>\n",
       "      <td>4.899598e-04</td>\n",
       "      <td>0.011393</td>\n",
       "      <td>0.009316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.007000</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>0.769388</td>\n",
       "      <td>0.770841</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>{'max_depth': 5, 'max_features': 1}</td>\n",
       "      <td>30</td>\n",
       "      <td>0.780956</td>\n",
       "      <td>0.785714</td>\n",
       "      <td>...</td>\n",
       "      <td>0.766012</td>\n",
       "      <td>0.773863</td>\n",
       "      <td>0.765912</td>\n",
       "      <td>0.763211</td>\n",
       "      <td>0.760359</td>\n",
       "      <td>0.755525</td>\n",
       "      <td>0.001265</td>\n",
       "      <td>4.900376e-04</td>\n",
       "      <td>0.007175</td>\n",
       "      <td>0.010475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.020601</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.820806</td>\n",
       "      <td>0.833703</td>\n",
       "      <td>6</td>\n",
       "      <td>0.5</td>\n",
       "      <td>{'max_depth': 6, 'max_features': 0.5}</td>\n",
       "      <td>23</td>\n",
       "      <td>0.796328</td>\n",
       "      <td>0.819133</td>\n",
       "      <td>...</td>\n",
       "      <td>0.802733</td>\n",
       "      <td>0.830237</td>\n",
       "      <td>0.827851</td>\n",
       "      <td>0.832711</td>\n",
       "      <td>0.847928</td>\n",
       "      <td>0.855236</td>\n",
       "      <td>0.002059</td>\n",
       "      <td>4.000187e-04</td>\n",
       "      <td>0.018876</td>\n",
       "      <td>0.011792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.025602</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>0.830287</td>\n",
       "      <td>0.840024</td>\n",
       "      <td>6</td>\n",
       "      <td>0.6</td>\n",
       "      <td>{'max_depth': 6, 'max_features': 0.6}</td>\n",
       "      <td>18</td>\n",
       "      <td>0.821947</td>\n",
       "      <td>0.840807</td>\n",
       "      <td>...</td>\n",
       "      <td>0.809991</td>\n",
       "      <td>0.835362</td>\n",
       "      <td>0.842375</td>\n",
       "      <td>0.843707</td>\n",
       "      <td>0.847501</td>\n",
       "      <td>0.851927</td>\n",
       "      <td>0.003382</td>\n",
       "      <td>4.899014e-04</td>\n",
       "      <td>0.013596</td>\n",
       "      <td>0.007933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.026002</td>\n",
       "      <td>0.0010</td>\n",
       "      <td>0.822856</td>\n",
       "      <td>0.832294</td>\n",
       "      <td>6</td>\n",
       "      <td>0.7</td>\n",
       "      <td>{'max_depth': 6, 'max_features': 0.7}</td>\n",
       "      <td>20</td>\n",
       "      <td>0.828779</td>\n",
       "      <td>0.831625</td>\n",
       "      <td>...</td>\n",
       "      <td>0.805295</td>\n",
       "      <td>0.837924</td>\n",
       "      <td>0.818026</td>\n",
       "      <td>0.819046</td>\n",
       "      <td>0.838103</td>\n",
       "      <td>0.841892</td>\n",
       "      <td>0.004517</td>\n",
       "      <td>1.907349e-07</td>\n",
       "      <td>0.010962</td>\n",
       "      <td>0.007765</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.031802</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.831995</td>\n",
       "      <td>0.840280</td>\n",
       "      <td>6</td>\n",
       "      <td>0.8</td>\n",
       "      <td>{'max_depth': 6, 'max_features': 0.8}</td>\n",
       "      <td>16</td>\n",
       "      <td>0.820666</td>\n",
       "      <td>0.837604</td>\n",
       "      <td>...</td>\n",
       "      <td>0.833476</td>\n",
       "      <td>0.852872</td>\n",
       "      <td>0.827851</td>\n",
       "      <td>0.836020</td>\n",
       "      <td>0.831696</td>\n",
       "      <td>0.830362</td>\n",
       "      <td>0.004445</td>\n",
       "      <td>3.999949e-04</td>\n",
       "      <td>0.008391</td>\n",
       "      <td>0.007751</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.038602</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.835326</td>\n",
       "      <td>0.846515</td>\n",
       "      <td>6</td>\n",
       "      <td>0.9</td>\n",
       "      <td>{'max_depth': 6, 'max_features': 0.9}</td>\n",
       "      <td>13</td>\n",
       "      <td>0.823228</td>\n",
       "      <td>0.832586</td>\n",
       "      <td>...</td>\n",
       "      <td>0.812126</td>\n",
       "      <td>0.837284</td>\n",
       "      <td>0.842802</td>\n",
       "      <td>0.850112</td>\n",
       "      <td>0.859462</td>\n",
       "      <td>0.867620</td>\n",
       "      <td>0.003200</td>\n",
       "      <td>3.999710e-04</td>\n",
       "      <td>0.016350</td>\n",
       "      <td>0.012165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.009401</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.771438</td>\n",
       "      <td>0.778699</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>{'max_depth': 6, 'max_features': 1}</td>\n",
       "      <td>28</td>\n",
       "      <td>0.786507</td>\n",
       "      <td>0.792334</td>\n",
       "      <td>...</td>\n",
       "      <td>0.746798</td>\n",
       "      <td>0.763506</td>\n",
       "      <td>0.741563</td>\n",
       "      <td>0.752535</td>\n",
       "      <td>0.782572</td>\n",
       "      <td>0.784670</td>\n",
       "      <td>0.001020</td>\n",
       "      <td>4.898819e-04</td>\n",
       "      <td>0.023030</td>\n",
       "      <td>0.017944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.028402</td>\n",
       "      <td>0.0022</td>\n",
       "      <td>0.830885</td>\n",
       "      <td>0.842779</td>\n",
       "      <td>7</td>\n",
       "      <td>0.5</td>\n",
       "      <td>{'max_depth': 7, 'max_features': 0.5}</td>\n",
       "      <td>17</td>\n",
       "      <td>0.825363</td>\n",
       "      <td>0.836109</td>\n",
       "      <td>...</td>\n",
       "      <td>0.833049</td>\n",
       "      <td>0.864296</td>\n",
       "      <td>0.841094</td>\n",
       "      <td>0.842212</td>\n",
       "      <td>0.826570</td>\n",
       "      <td>0.835593</td>\n",
       "      <td>0.005463</td>\n",
       "      <td>1.469840e-03</td>\n",
       "      <td>0.005734</td>\n",
       "      <td>0.011044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.034602</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.832764</td>\n",
       "      <td>0.849569</td>\n",
       "      <td>7</td>\n",
       "      <td>0.6</td>\n",
       "      <td>{'max_depth': 7, 'max_features': 0.6}</td>\n",
       "      <td>14</td>\n",
       "      <td>0.820666</td>\n",
       "      <td>0.855541</td>\n",
       "      <td>...</td>\n",
       "      <td>0.839880</td>\n",
       "      <td>0.851591</td>\n",
       "      <td>0.832123</td>\n",
       "      <td>0.840824</td>\n",
       "      <td>0.847501</td>\n",
       "      <td>0.862389</td>\n",
       "      <td>0.004030</td>\n",
       "      <td>4.899209e-04</td>\n",
       "      <td>0.009976</td>\n",
       "      <td>0.009234</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.043202</td>\n",
       "      <td>0.0036</td>\n",
       "      <td>0.832422</td>\n",
       "      <td>0.847199</td>\n",
       "      <td>7</td>\n",
       "      <td>0.7</td>\n",
       "      <td>{'max_depth': 7, 'max_features': 0.7}</td>\n",
       "      <td>15</td>\n",
       "      <td>0.830914</td>\n",
       "      <td>0.845398</td>\n",
       "      <td>...</td>\n",
       "      <td>0.830060</td>\n",
       "      <td>0.857250</td>\n",
       "      <td>0.840239</td>\n",
       "      <td>0.847016</td>\n",
       "      <td>0.820162</td>\n",
       "      <td>0.829828</td>\n",
       "      <td>0.006969</td>\n",
       "      <td>4.223940e-03</td>\n",
       "      <td>0.007594</td>\n",
       "      <td>0.009926</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.049403</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.837803</td>\n",
       "      <td>0.853690</td>\n",
       "      <td>7</td>\n",
       "      <td>0.8</td>\n",
       "      <td>{'max_depth': 7, 'max_features': 0.8}</td>\n",
       "      <td>11</td>\n",
       "      <td>0.828779</td>\n",
       "      <td>0.855968</td>\n",
       "      <td>...</td>\n",
       "      <td>0.815542</td>\n",
       "      <td>0.844117</td>\n",
       "      <td>0.837249</td>\n",
       "      <td>0.843493</td>\n",
       "      <td>0.864161</td>\n",
       "      <td>0.875627</td>\n",
       "      <td>0.002940</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.016138</td>\n",
       "      <td>0.011848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.065604</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.849846</td>\n",
       "      <td>0.863021</td>\n",
       "      <td>7</td>\n",
       "      <td>0.9</td>\n",
       "      <td>{'max_depth': 7, 'max_features': 0.9}</td>\n",
       "      <td>10</td>\n",
       "      <td>0.841588</td>\n",
       "      <td>0.860346</td>\n",
       "      <td>...</td>\n",
       "      <td>0.819385</td>\n",
       "      <td>0.842943</td>\n",
       "      <td>0.867151</td>\n",
       "      <td>0.874987</td>\n",
       "      <td>0.857326</td>\n",
       "      <td>0.865912</td>\n",
       "      <td>0.024761</td>\n",
       "      <td>6.324097e-04</td>\n",
       "      <td>0.017584</td>\n",
       "      <td>0.011173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.011801</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>0.782115</td>\n",
       "      <td>0.791745</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>{'max_depth': 7, 'max_features': 1}</td>\n",
       "      <td>26</td>\n",
       "      <td>0.777968</td>\n",
       "      <td>0.803545</td>\n",
       "      <td>...</td>\n",
       "      <td>0.771563</td>\n",
       "      <td>0.782618</td>\n",
       "      <td>0.788552</td>\n",
       "      <td>0.793637</td>\n",
       "      <td>0.788125</td>\n",
       "      <td>0.794171</td>\n",
       "      <td>0.000980</td>\n",
       "      <td>3.999710e-04</td>\n",
       "      <td>0.006499</td>\n",
       "      <td>0.007495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>0.036802</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.836693</td>\n",
       "      <td>0.855462</td>\n",
       "      <td>8</td>\n",
       "      <td>0.5</td>\n",
       "      <td>{'max_depth': 8, 'max_features': 0.5}</td>\n",
       "      <td>12</td>\n",
       "      <td>0.856106</td>\n",
       "      <td>0.879031</td>\n",
       "      <td>...</td>\n",
       "      <td>0.810845</td>\n",
       "      <td>0.844651</td>\n",
       "      <td>0.838958</td>\n",
       "      <td>0.852140</td>\n",
       "      <td>0.828279</td>\n",
       "      <td>0.847870</td>\n",
       "      <td>0.001939</td>\n",
       "      <td>9.536743e-08</td>\n",
       "      <td>0.015996</td>\n",
       "      <td>0.012202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.045003</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.850188</td>\n",
       "      <td>0.872096</td>\n",
       "      <td>8</td>\n",
       "      <td>0.6</td>\n",
       "      <td>{'max_depth': 8, 'max_features': 0.6}</td>\n",
       "      <td>9</td>\n",
       "      <td>0.839880</td>\n",
       "      <td>0.876788</td>\n",
       "      <td>...</td>\n",
       "      <td>0.850128</td>\n",
       "      <td>0.884262</td>\n",
       "      <td>0.839385</td>\n",
       "      <td>0.846482</td>\n",
       "      <td>0.859889</td>\n",
       "      <td>0.878083</td>\n",
       "      <td>0.004243</td>\n",
       "      <td>9.536743e-08</td>\n",
       "      <td>0.009472</td>\n",
       "      <td>0.013188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0.047003</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.855227</td>\n",
       "      <td>0.875192</td>\n",
       "      <td>8</td>\n",
       "      <td>0.7</td>\n",
       "      <td>{'max_depth': 8, 'max_features': 0.7}</td>\n",
       "      <td>7</td>\n",
       "      <td>0.845004</td>\n",
       "      <td>0.871663</td>\n",
       "      <td>...</td>\n",
       "      <td>0.861657</td>\n",
       "      <td>0.884262</td>\n",
       "      <td>0.865442</td>\n",
       "      <td>0.879470</td>\n",
       "      <td>0.854763</td>\n",
       "      <td>0.873385</td>\n",
       "      <td>0.000632</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.007561</td>\n",
       "      <td>0.006008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>0.054203</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>0.857448</td>\n",
       "      <td>0.877200</td>\n",
       "      <td>8</td>\n",
       "      <td>0.8</td>\n",
       "      <td>{'max_depth': 8, 'max_features': 0.8}</td>\n",
       "      <td>5</td>\n",
       "      <td>0.872331</td>\n",
       "      <td>0.903267</td>\n",
       "      <td>...</td>\n",
       "      <td>0.845004</td>\n",
       "      <td>0.868140</td>\n",
       "      <td>0.850491</td>\n",
       "      <td>0.859720</td>\n",
       "      <td>0.855190</td>\n",
       "      <td>0.878937</td>\n",
       "      <td>0.001720</td>\n",
       "      <td>3.999233e-04</td>\n",
       "      <td>0.009755</td>\n",
       "      <td>0.014641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>0.059803</td>\n",
       "      <td>0.0020</td>\n",
       "      <td>0.862743</td>\n",
       "      <td>0.882238</td>\n",
       "      <td>8</td>\n",
       "      <td>0.9</td>\n",
       "      <td>{'max_depth': 8, 'max_features': 0.9}</td>\n",
       "      <td>3</td>\n",
       "      <td>0.861657</td>\n",
       "      <td>0.882554</td>\n",
       "      <td>...</td>\n",
       "      <td>0.856106</td>\n",
       "      <td>0.886931</td>\n",
       "      <td>0.884238</td>\n",
       "      <td>0.895164</td>\n",
       "      <td>0.865442</td>\n",
       "      <td>0.881072</td>\n",
       "      <td>0.002315</td>\n",
       "      <td>6.325605e-04</td>\n",
       "      <td>0.012533</td>\n",
       "      <td>0.009713</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>0.013801</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.770072</td>\n",
       "      <td>0.789311</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>{'max_depth': 8, 'max_features': 1}</td>\n",
       "      <td>29</td>\n",
       "      <td>0.791204</td>\n",
       "      <td>0.806748</td>\n",
       "      <td>...</td>\n",
       "      <td>0.765158</td>\n",
       "      <td>0.789665</td>\n",
       "      <td>0.759504</td>\n",
       "      <td>0.777303</td>\n",
       "      <td>0.774883</td>\n",
       "      <td>0.787659</td>\n",
       "      <td>0.006145</td>\n",
       "      <td>4.899598e-04</td>\n",
       "      <td>0.011959</td>\n",
       "      <td>0.009676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>0.047003</td>\n",
       "      <td>0.0022</td>\n",
       "      <td>0.854544</td>\n",
       "      <td>0.880573</td>\n",
       "      <td>9</td>\n",
       "      <td>0.5</td>\n",
       "      <td>{'max_depth': 9, 'max_features': 0.5}</td>\n",
       "      <td>8</td>\n",
       "      <td>0.852263</td>\n",
       "      <td>0.878603</td>\n",
       "      <td>...</td>\n",
       "      <td>0.833903</td>\n",
       "      <td>0.884155</td>\n",
       "      <td>0.889364</td>\n",
       "      <td>0.896979</td>\n",
       "      <td>0.845365</td>\n",
       "      <td>0.862710</td>\n",
       "      <td>0.009056</td>\n",
       "      <td>4.000425e-04</td>\n",
       "      <td>0.018628</td>\n",
       "      <td>0.011000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>0.046203</td>\n",
       "      <td>0.0018</td>\n",
       "      <td>0.858302</td>\n",
       "      <td>0.888943</td>\n",
       "      <td>9</td>\n",
       "      <td>0.6</td>\n",
       "      <td>{'max_depth': 9, 'max_features': 0.6}</td>\n",
       "      <td>4</td>\n",
       "      <td>0.837319</td>\n",
       "      <td>0.871023</td>\n",
       "      <td>...</td>\n",
       "      <td>0.858241</td>\n",
       "      <td>0.900491</td>\n",
       "      <td>0.859462</td>\n",
       "      <td>0.880111</td>\n",
       "      <td>0.870995</td>\n",
       "      <td>0.901036</td>\n",
       "      <td>0.005528</td>\n",
       "      <td>4.000664e-04</td>\n",
       "      <td>0.011442</td>\n",
       "      <td>0.011734</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>0.046803</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.857106</td>\n",
       "      <td>0.887000</td>\n",
       "      <td>9</td>\n",
       "      <td>0.7</td>\n",
       "      <td>{'max_depth': 9, 'max_features': 0.7}</td>\n",
       "      <td>6</td>\n",
       "      <td>0.853117</td>\n",
       "      <td>0.884262</td>\n",
       "      <td>...</td>\n",
       "      <td>0.845004</td>\n",
       "      <td>0.879885</td>\n",
       "      <td>0.863733</td>\n",
       "      <td>0.882673</td>\n",
       "      <td>0.876976</td>\n",
       "      <td>0.905413</td>\n",
       "      <td>0.008232</td>\n",
       "      <td>4.899014e-04</td>\n",
       "      <td>0.011908</td>\n",
       "      <td>0.009314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>0.044803</td>\n",
       "      <td>0.0012</td>\n",
       "      <td>0.869747</td>\n",
       "      <td>0.895328</td>\n",
       "      <td>9</td>\n",
       "      <td>0.8</td>\n",
       "      <td>{'max_depth': 9, 'max_features': 0.8}</td>\n",
       "      <td>2</td>\n",
       "      <td>0.878309</td>\n",
       "      <td>0.912129</td>\n",
       "      <td>...</td>\n",
       "      <td>0.858241</td>\n",
       "      <td>0.893017</td>\n",
       "      <td>0.853909</td>\n",
       "      <td>0.864311</td>\n",
       "      <td>0.881674</td>\n",
       "      <td>0.914274</td>\n",
       "      <td>0.009109</td>\n",
       "      <td>3.999949e-04</td>\n",
       "      <td>0.011364</td>\n",
       "      <td>0.017969</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>0.044202</td>\n",
       "      <td>0.0016</td>\n",
       "      <td>0.885378</td>\n",
       "      <td>0.910552</td>\n",
       "      <td>9</td>\n",
       "      <td>0.9</td>\n",
       "      <td>{'max_depth': 9, 'max_features': 0.9}</td>\n",
       "      <td>1</td>\n",
       "      <td>0.861230</td>\n",
       "      <td>0.893444</td>\n",
       "      <td>...</td>\n",
       "      <td>0.890692</td>\n",
       "      <td>0.912022</td>\n",
       "      <td>0.896198</td>\n",
       "      <td>0.917690</td>\n",
       "      <td>0.891927</td>\n",
       "      <td>0.920572</td>\n",
       "      <td>0.010685</td>\n",
       "      <td>4.898819e-04</td>\n",
       "      <td>0.012438</td>\n",
       "      <td>0.009471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>0.012201</td>\n",
       "      <td>0.0014</td>\n",
       "      <td>0.777417</td>\n",
       "      <td>0.799987</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>{'max_depth': 9, 'max_features': 1}</td>\n",
       "      <td>27</td>\n",
       "      <td>0.766866</td>\n",
       "      <td>0.803652</td>\n",
       "      <td>...</td>\n",
       "      <td>0.756191</td>\n",
       "      <td>0.790626</td>\n",
       "      <td>0.777446</td>\n",
       "      <td>0.792356</td>\n",
       "      <td>0.790261</td>\n",
       "      <td>0.802818</td>\n",
       "      <td>0.000980</td>\n",
       "      <td>4.899598e-04</td>\n",
       "      <td>0.014728</td>\n",
       "      <td>0.007450</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>30 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    mean_fit_time  mean_score_time  mean_test_score  mean_train_score  \\\n",
       "0        0.016201           0.0012         0.813119          0.823753   \n",
       "1        0.018401           0.0018         0.821063          0.827831   \n",
       "2        0.021401           0.0012         0.819098          0.824927   \n",
       "3        0.029602           0.0010         0.822600          0.826572   \n",
       "4        0.033002           0.0014         0.827127          0.834387   \n",
       "5        0.007000           0.0014         0.769388          0.770841   \n",
       "6        0.020601           0.0012         0.820806          0.833703   \n",
       "7        0.025602           0.0014         0.830287          0.840024   \n",
       "8        0.026002           0.0010         0.822856          0.832294   \n",
       "9        0.031802           0.0012         0.831995          0.840280   \n",
       "10       0.038602           0.0012         0.835326          0.846515   \n",
       "11       0.009401           0.0016         0.771438          0.778699   \n",
       "12       0.028402           0.0022         0.830885          0.842779   \n",
       "13       0.034602           0.0016         0.832764          0.849569   \n",
       "14       0.043202           0.0036         0.832422          0.847199   \n",
       "15       0.049403           0.0020         0.837803          0.853690   \n",
       "16       0.065604           0.0020         0.849846          0.863021   \n",
       "17       0.011801           0.0018         0.782115          0.791745   \n",
       "18       0.036802           0.0020         0.836693          0.855462   \n",
       "19       0.045003           0.0020         0.850188          0.872096   \n",
       "20       0.047003           0.0020         0.855227          0.875192   \n",
       "21       0.054203           0.0018         0.857448          0.877200   \n",
       "22       0.059803           0.0020         0.862743          0.882238   \n",
       "23       0.013801           0.0016         0.770072          0.789311   \n",
       "24       0.047003           0.0022         0.854544          0.880573   \n",
       "25       0.046203           0.0018         0.858302          0.888943   \n",
       "26       0.046803           0.0016         0.857106          0.887000   \n",
       "27       0.044803           0.0012         0.869747          0.895328   \n",
       "28       0.044202           0.0016         0.885378          0.910552   \n",
       "29       0.012201           0.0014         0.777417          0.799987   \n",
       "\n",
       "   param_max_depth param_max_features                                 params  \\\n",
       "0                5                0.5  {'max_depth': 5, 'max_features': 0.5}   \n",
       "1                5                0.6  {'max_depth': 5, 'max_features': 0.6}   \n",
       "2                5                0.7  {'max_depth': 5, 'max_features': 0.7}   \n",
       "3                5                0.8  {'max_depth': 5, 'max_features': 0.8}   \n",
       "4                5                0.9  {'max_depth': 5, 'max_features': 0.9}   \n",
       "5                5                  1    {'max_depth': 5, 'max_features': 1}   \n",
       "6                6                0.5  {'max_depth': 6, 'max_features': 0.5}   \n",
       "7                6                0.6  {'max_depth': 6, 'max_features': 0.6}   \n",
       "8                6                0.7  {'max_depth': 6, 'max_features': 0.7}   \n",
       "9                6                0.8  {'max_depth': 6, 'max_features': 0.8}   \n",
       "10               6                0.9  {'max_depth': 6, 'max_features': 0.9}   \n",
       "11               6                  1    {'max_depth': 6, 'max_features': 1}   \n",
       "12               7                0.5  {'max_depth': 7, 'max_features': 0.5}   \n",
       "13               7                0.6  {'max_depth': 7, 'max_features': 0.6}   \n",
       "14               7                0.7  {'max_depth': 7, 'max_features': 0.7}   \n",
       "15               7                0.8  {'max_depth': 7, 'max_features': 0.8}   \n",
       "16               7                0.9  {'max_depth': 7, 'max_features': 0.9}   \n",
       "17               7                  1    {'max_depth': 7, 'max_features': 1}   \n",
       "18               8                0.5  {'max_depth': 8, 'max_features': 0.5}   \n",
       "19               8                0.6  {'max_depth': 8, 'max_features': 0.6}   \n",
       "20               8                0.7  {'max_depth': 8, 'max_features': 0.7}   \n",
       "21               8                0.8  {'max_depth': 8, 'max_features': 0.8}   \n",
       "22               8                0.9  {'max_depth': 8, 'max_features': 0.9}   \n",
       "23               8                  1    {'max_depth': 8, 'max_features': 1}   \n",
       "24               9                0.5  {'max_depth': 9, 'max_features': 0.5}   \n",
       "25               9                0.6  {'max_depth': 9, 'max_features': 0.6}   \n",
       "26               9                0.7  {'max_depth': 9, 'max_features': 0.7}   \n",
       "27               9                0.8  {'max_depth': 9, 'max_features': 0.8}   \n",
       "28               9                0.9  {'max_depth': 9, 'max_features': 0.9}   \n",
       "29               9                  1    {'max_depth': 9, 'max_features': 1}   \n",
       "\n",
       "    rank_test_score  split0_test_score  split0_train_score       ...         \\\n",
       "0                25           0.810845            0.825432       ...          \n",
       "1                22           0.805722            0.817745       ...          \n",
       "2                24           0.812126            0.825966       ...          \n",
       "3                21           0.823228            0.833013       ...          \n",
       "4                19           0.831341            0.846893       ...          \n",
       "5                30           0.780956            0.785714       ...          \n",
       "6                23           0.796328            0.819133       ...          \n",
       "7                18           0.821947            0.840807       ...          \n",
       "8                20           0.828779            0.831625       ...          \n",
       "9                16           0.820666            0.837604       ...          \n",
       "10               13           0.823228            0.832586       ...          \n",
       "11               28           0.786507            0.792334       ...          \n",
       "12               17           0.825363            0.836109       ...          \n",
       "13               14           0.820666            0.855541       ...          \n",
       "14               15           0.830914            0.845398       ...          \n",
       "15               11           0.828779            0.855968       ...          \n",
       "16               10           0.841588            0.860346       ...          \n",
       "17               26           0.777968            0.803545       ...          \n",
       "18               12           0.856106            0.879031       ...          \n",
       "19                9           0.839880            0.876788       ...          \n",
       "20                7           0.845004            0.871663       ...          \n",
       "21                5           0.872331            0.903267       ...          \n",
       "22                3           0.861657            0.882554       ...          \n",
       "23               29           0.791204            0.806748       ...          \n",
       "24                8           0.852263            0.878603       ...          \n",
       "25                4           0.837319            0.871023       ...          \n",
       "26                6           0.853117            0.884262       ...          \n",
       "27                2           0.878309            0.912129       ...          \n",
       "28                1           0.861230            0.893444       ...          \n",
       "29               27           0.766866            0.803652       ...          \n",
       "\n",
       "    split2_test_score  split2_train_score  split3_test_score  \\\n",
       "0            0.800598            0.826287           0.816318   \n",
       "1            0.802733            0.831625           0.844084   \n",
       "2            0.796755            0.827568           0.811192   \n",
       "3            0.806576            0.831625           0.825716   \n",
       "4            0.808284            0.833013           0.823580   \n",
       "5            0.766012            0.773863           0.765912   \n",
       "6            0.802733            0.830237           0.827851   \n",
       "7            0.809991            0.835362           0.842375   \n",
       "8            0.805295            0.837924           0.818026   \n",
       "9            0.833476            0.852872           0.827851   \n",
       "10           0.812126            0.837284           0.842802   \n",
       "11           0.746798            0.763506           0.741563   \n",
       "12           0.833049            0.864296           0.841094   \n",
       "13           0.839880            0.851591           0.832123   \n",
       "14           0.830060            0.857250           0.840239   \n",
       "15           0.815542            0.844117           0.837249   \n",
       "16           0.819385            0.842943           0.867151   \n",
       "17           0.771563            0.782618           0.788552   \n",
       "18           0.810845            0.844651           0.838958   \n",
       "19           0.850128            0.884262           0.839385   \n",
       "20           0.861657            0.884262           0.865442   \n",
       "21           0.845004            0.868140           0.850491   \n",
       "22           0.856106            0.886931           0.884238   \n",
       "23           0.765158            0.789665           0.759504   \n",
       "24           0.833903            0.884155           0.889364   \n",
       "25           0.858241            0.900491           0.859462   \n",
       "26           0.845004            0.879885           0.863733   \n",
       "27           0.858241            0.893017           0.853909   \n",
       "28           0.890692            0.912022           0.896198   \n",
       "29           0.756191            0.790626           0.777446   \n",
       "\n",
       "    split3_train_score  split4_test_score  split4_train_score  std_fit_time  \\\n",
       "0             0.821501           0.810337            0.819152      0.000400   \n",
       "1             0.843173           0.821871            0.821715      0.001357   \n",
       "2             0.804847           0.844511            0.840717      0.002498   \n",
       "3             0.820006           0.826997            0.826519      0.004079   \n",
       "4             0.824277           0.843229            0.843173      0.004000   \n",
       "5             0.763211           0.760359            0.755525      0.001265   \n",
       "6             0.832711           0.847928            0.855236      0.002059   \n",
       "7             0.843707           0.847501            0.851927      0.003382   \n",
       "8             0.819046           0.838103            0.841892      0.004517   \n",
       "9             0.836020           0.831696            0.830362      0.004445   \n",
       "10            0.850112           0.859462            0.867620      0.003200   \n",
       "11            0.752535           0.782572            0.784670      0.001020   \n",
       "12            0.842212           0.826570            0.835593      0.005463   \n",
       "13            0.840824           0.847501            0.862389      0.004030   \n",
       "14            0.847016           0.820162            0.829828      0.006969   \n",
       "15            0.843493           0.864161            0.875627      0.002940   \n",
       "16            0.874987           0.857326            0.865912      0.024761   \n",
       "17            0.793637           0.788125            0.794171      0.000980   \n",
       "18            0.852140           0.828279            0.847870      0.001939   \n",
       "19            0.846482           0.859889            0.878083      0.004243   \n",
       "20            0.879470           0.854763            0.873385      0.000632   \n",
       "21            0.859720           0.855190            0.878937      0.001720   \n",
       "22            0.895164           0.865442            0.881072      0.002315   \n",
       "23            0.777303           0.774883            0.787659      0.006145   \n",
       "24            0.896979           0.845365            0.862710      0.009056   \n",
       "25            0.880111           0.870995            0.901036      0.005528   \n",
       "26            0.882673           0.876976            0.905413      0.008232   \n",
       "27            0.864311           0.881674            0.914274      0.009109   \n",
       "28            0.917690           0.891927            0.920572      0.010685   \n",
       "29            0.792356           0.790261            0.802818      0.000980   \n",
       "\n",
       "    std_score_time  std_test_score  std_train_score  \n",
       "0     4.001379e-04        0.008795         0.002914  \n",
       "1     4.000902e-04        0.015485         0.008916  \n",
       "2     4.000903e-04        0.016707         0.011498  \n",
       "3     1.168008e-07        0.008348         0.005175  \n",
       "4     4.899598e-04        0.011393         0.009316  \n",
       "5     4.900376e-04        0.007175         0.010475  \n",
       "6     4.000187e-04        0.018876         0.011792  \n",
       "7     4.899014e-04        0.013596         0.007933  \n",
       "8     1.907349e-07        0.010962         0.007765  \n",
       "9     3.999949e-04        0.008391         0.007751  \n",
       "10    3.999710e-04        0.016350         0.012165  \n",
       "11    4.898819e-04        0.023030         0.017944  \n",
       "12    1.469840e-03        0.005734         0.011044  \n",
       "13    4.899209e-04        0.009976         0.009234  \n",
       "14    4.223940e-03        0.007594         0.009926  \n",
       "15    0.000000e+00        0.016138         0.011848  \n",
       "16    6.324097e-04        0.017584         0.011173  \n",
       "17    3.999710e-04        0.006499         0.007495  \n",
       "18    9.536743e-08        0.015996         0.012202  \n",
       "19    9.536743e-08        0.009472         0.013188  \n",
       "20    0.000000e+00        0.007561         0.006008  \n",
       "21    3.999233e-04        0.009755         0.014641  \n",
       "22    6.325605e-04        0.012533         0.009713  \n",
       "23    4.899598e-04        0.011959         0.009676  \n",
       "24    4.000425e-04        0.018628         0.011000  \n",
       "25    4.000664e-04        0.011442         0.011734  \n",
       "26    4.899014e-04        0.011908         0.009314  \n",
       "27    3.999949e-04        0.011364         0.017969  \n",
       "28    4.898819e-04        0.012438         0.009471  \n",
       "29    4.899598e-04        0.014728         0.007450  \n",
       "\n",
       "[30 rows x 22 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.head(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.81311923, 0.82106252, 0.81909805, 0.82259993, 0.82712675],\n",
       "       [0.76938845, 0.82080629, 0.83028698, 0.82285617, 0.83199522],\n",
       "       [0.83532627, 0.77143833, 0.83088487, 0.83276392, 0.83242228],\n",
       "       [0.83780321, 0.84984626, 0.78211479, 0.83669286, 0.85018791],\n",
       "       [0.8552272 , 0.8574479 , 0.86274342, 0.77007175, 0.8545439 ],\n",
       "       [0.85830202, 0.85710625, 0.86974718, 0.88537752, 0.77741715]])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores = np.array(results.mean_test_score).reshape(6, 5)\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1c2a8198>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWMAAAD8CAYAAACihcXDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xd8VFX6+PHPM5NkQnovJECA0EvoSO9FVGRxUbCirNi/q6uuu+4uuq5dsexaQdG1rIjYUECkKkhXaoCQAoSQngCphCRzfn/cMcmkIyEzye+8X6/7Ymbumbnn4dx55uTcc+8VpRSapmmaY5kcXQFN0zRNJ2NN0zSnoJOxpmmaE9DJWNM0zQnoZKxpmuYEdDLWNE1zAjoZa5qmOQGdjDVN05yATsaapmlOwOVSb6DPgy+3ulP8SvwdXYNLw1Tm6Bo0vXKLo2twaZR7tLqvFQBJf/yTXOxnWNO7Nvo/xxR29KK311R0z1jTNM0JXPKesaZpWnOyYm10WWfqjepkrGlaq1Kqyhtd1pkSoDPVRdM07aJdSM/YmehkrGlaq1LeQi8LrJOxpmmtipWWmYydafxa0zTtopWjGr00RESmikiciCSIyF9qWd9eRDaKyB4R2S8i02yvu4rIf0XkgIgcFpG/NrQtnYw1TWtVrKhGL/URETPwOnA50BOYIyI9qxX7O7BMKdUfmA28YXt9FmBRSvUBBgJ3iEhUfdvTwxSaprUqpU03ZjwESFBKJQGIyFLgauBQlTIK8LE99gVSq7zuKSIuQBvgPJBX38Z0z1jTtFblQoYpRGS+iOyussyv8lERwMkqz1Nsr1X1OHCjiKQAq4D7bK8vBwqBNCAZeFEplVtfvXXPWNO0VqX8AjrGSqlFwKI6Vtd2qnT1T58DvK+UWigiw4APRaQ3Rq+6HGgL+AObRWTdr73s2uiesaZprYr1ApYGpADtqjyPpHIY4lfzgGUASqltgDsQBFwPfKeUKlVKZQI/AYPq25hOxpqmtSrlSKOXBuwCuohIRxFxwzhAt6JamWRgAoCI9MBIxlm218eLwRO4DDhS38ZazDDFiG4deGTGWMwmE1/sOMi7G3bZrQ/z8+apOVPwbmPBLMIrK7ew+chxfD3ceemWK+ndLpSvdx3i6S83OiiCmkZ16sDfpozFLCY+23uQRVvtYwr38ea56VPwcbdgEmHhhi38kHic4R3b89D4kbiazZSWl/P8+s1sP36yjq00v5HRHfjb1LGYTCaW/3KQxVuqxeXrzbMzpuDtbsFsEhau28KP8ccZ3qk9D06sEtfazew45hxxjerUgb9PNPa/ZXsPsmh7zbZ6/sop+FgsmEzCi5uMthoR1Z6HxlbG9NzGzWw/4RwxAYzuEMWCMWMxiYllsQd4a7d9XG29vXlh0lR8LEZbPf/TFjYdP0bf0DCenjARAEF4dcc2vk9McEQINZSqprkQm1KqTETuBdYAZmCJUipWRJ4AdiulVgAPAotF5AGMIYy5SiklIq8D7wEHMYY73lNK7a9vey0iGZtE+NvM8cx/+wvSz+az9P7r2RibSFJG5Xj4HROHsmbvUZZt20+n0ADe+MMMpj61hPNlZbz23Vaiw4LoEhbowCjsmUR47PLx3PrxF6Tn5fP5vOtZfzSRxOzKmO4eOZTVh47yyS/76RwUwOLZMxj/2hJOFxVz56dfk1lQSJfgQJbMmcmofy92YDSVTCIsmDae2z78goy8fD67/Xo2xCWSmFUZ112jh7I69ihLd++nc3AAi26YwYRXjLju+uRrMvML6RISyDs3zmTMS46PyyTC45PHM3epra3mXs+G+EQScqq01fChrD58lP/t2U90YACLr53BuDeXcLq4mDuW29oqKJAls2cy6jXHxwRGXP8cO56bv/yc9IJ8vpp9A+uSEknIrYzrnsFDWRUfx8cH9hMdEMCSq3/H6Pfe5WhONld/8jHlShHs4cnKG25ifVKiU5z91ogeb6MppVZhHJir+tqCKo8PASNqeV8BxvS2RmsRwxR92oeRnHOGlNyzlJVbWb0njnG9OtuVUSi83N0A8Ha3kJVXCEDx+TL2HEvlfJlzXay3b9swTuSe4eSZs5RarayMjWNi11pisthisljIzDdiOpyRRWaB8Tg+Kwc3FzOuZnPzBlCHvhFhJOeeIeX0WUrLraw6GMeEbtXiUnXElZ5V8Tg+MweLk8TVt20YJ05XaavDcUyo1lZUaSsvd0tF+xyq2lbZRkxuThATQExoGCfOnuFknhHXt0ePMKlT9X0QvNyMi0J7u1nIsMVyrqysIvFaXMzUPK7lOFYljV6cSYvoGYf4epF+Jr/iecbZAvq2D7Mr88aa7Sy6YybXj+xHGzdXbn/78+au5gUJ9fYiPa8ypvT8AmLa2sf0nx+3s+T6mdw0uB9tXF2Z+3HNmKZ078LhjCxKyxt/papLKdTHi7SqceUVEBNpH9drm7bz7k0zuXGoEddtH9QSV88uHEp3jrjCvKrFVEtb/Xvzdt6bPZObBhox3bK0ZkxTuxkxnXeCmMAWV35lXGkFBfQLC7cr8+r2bXzwu2u4OaYfHq6u3PTl8op1MaFhPDdpMhHePjz4/XdO0SuGpu0ZN6d6e8YicluVx5Eisl5EzojIVhHpWs/7Kubu5e7fdtGVrH1+iX3DT+vfja92xTLxX+9w9ztf8fScqYgTt0ltdase05W9uvHlvlhG//sdbl/6FS9cPdXu/yI6KJCHJ4zkH6vWXdrKXiRV7Ut6RZ9ufLk3lrEvvcMdH3/FczPt2yo6OJAHJ47ksW+cJK7a2qpaTFf27MYXB2IZ9fo7/OGzr3jxqlraatxIFnznJDHVoXpc07t1Y/mhWEYsWcxtX3/JwsmXV8S1LyOdqR99wIyl/+OuQUOcpsdfjqnRizNpqDb3Vnn8EsYUjgDgBeDNut6klFqklBqklBoU0HfYRVcy42wBYX7eFc9Dfb3IPFtoV+Z3Q3uzZt9RAPadSMPi6oK/Z5uL3valkp5XQJhPZUxh3l4Vf6L/6vf9erPqsBHT3lNpWFxc8PcwYgr19uL1WVfx56/XcPL02eareAMy8goIrxqXT824runfm9WxtrhSqsXl48Vrs6/ikS+dJ670/GoxeXtVDD38alZMtbYyV8YU5u3FG9dcxcPfrCH5jHPEBJBeUEC4d2Vc4V5eZBYW2JWZ1as3q+LjANiTnobFxUxAG/vvVeLpXIpKS+kWGHTpK90ILXWY4kJ+Groqpd5WSlmVUl9iJOVmcfBkOh2C/IkI8MHFbOLy/t3YFGs/dzr9dB6XdWkPQMeQANxczOQWFDdXFS/YgdR0ogL8ifTzwdVk4ope3Vh/1D6mtLN5DIsyYuocaIupqBhvi4XFs2ewcOMWfkmpPu3RsQ6kptMh0J8IPx9czSam9e7Ghrha4upkxNUpKACLi5ncwmK83S28ff0MXlq3hT0nnSeuA6npRPn7E+lra6se3Vgfbx9Tal4ew+toq0WzZrBw0xZ+OeU8MQHsz0gnys+PSB8jriu7dmddUrW48vMZ3s4Wl38AFrMLOcXFRPr4YLb9OdPW25tO/v6k5DnHD815ZW704kyk+p8lditFMoGlGH+ozQSilFKltnUHlVK9G9pAU92QdFT3KP48YyxmEb7cGcvi9Tu5Z8owYlMy2BSbRKfQAB6fNQkPiytKKV76djPbjiYD8N3fbsPL3YKr2UR+cQnzF31hNxPjQjXVDUnHdI7i0cljMZuE5XtjeeunnfzfmGEcTM1gQ3wSnYMCePKKSXi6GTE9v2EzPyUlc9fIIdwxfAgnck9XfNat//uC3KKL+/FpqhuSju4SxaNTx2IS4fM9sby9eSf3jTPi2hiXROfgAP511SQ83FxRKF5cu5mfEpO5c/QQ5o+0j2veh1+QW/jb42qqG5KO6RzF3yYa+9/y/bG8uXUnfxw1jANpGWxISCI6MIAnp03Cw9UVUDy/cTNbjiVz9/Ah3DFsCCdOV8Y0d+nFt1VT3ZB0bFRH/jHaaKvPDh3kjV07uf+y4RzISGf9sSSiAwJ4esIkPF3dUCie3bKZLcknmNG9B3cOGkyZ1YpVKf6zYztrkxIvuj5NcUPSH443/oakY6Kc54akDSXjW6q9tEIpdVpEwoD/U0o92tAG9N2hWw59d+iWQ98dum4bjndr9H/O+Kg4p0nG9c6mUEr9t47X04EGE7GmaVpzK1fOdWCusX5zrUXkyqasiKZpWlOwIo1enMnFzDMeDHzbVBXRNE1rCudVizh9ooYGay0i3TEuqByBcZpNKsbY8WOXuG6apmkXzOpk84cbq6GTPh6hcjbFToyrGAnwSW33g9I0TXO0ciWNXpxJQz3jeUCvX6ez/UpEXgJigWcvVcU0TdN+C2c7s66xGkrGVowr1Z+o9no4jbo2s6ZpWvOyttDZFA0l4/uB9SIST+W9oNoD0difKq1pmuYUWmXPWCn1ne2CQEMwDuAJxq1IdimlnOPSU5qmaVWUOtlpzo3V4GwKpZQV2N4MddE0TbtoLfWkj5Y5IU/TNK0OznYyR2PpZKxpWquie8aapmlOoFUewGsKYS9vvdSbaHbJ/xzu6CpcEudCW98xWdezLfNgTkOsHq2vrZqKs100vrF0z1jTtFaltLVem0LTNK0laak3JNXJWNO0VqW1noGnaZrWouiesaZpmhPQPWNN0zQn0GpPh9Y0TWtJWupJHy2z1pqmaXWwKmn00hARmSoicSKSUNsNNUSkvYhsFJE9IrJfRKZVWddXRLaJSKyIHBAR9/q2pXvGmqa1Kk11Bp6ImIHXgUnYrlYpIiuUUoeqFPs7sEwp9aaI9ARWAVEi4gJ8BNyklNonIoFAKfXQyVjTtFalCc/AGwIkKKWSAERkKcb9QKsmYwX42B77YtwjFGAysF8ptQ9AKZXT0MZ0MtY0rVVpwhuSRlB5Uw0wesdDq5V5HPheRO4DPIGJtte7AkpE1gDBwFKl1PP1bUyPGWua1qqUWk2NXkRkvojsrrLMr/JRtXWxVbXnc4D3lVKRwDTgQxExYXR0RwI32P79nYhMqK/eumesaVqrciHzjJVSi4BFdaxOAdpVeR5J5TDEr+YBU22ftc12kC7I9t4flFLZACKyChgArK+rLi0iGQ+a0o+7X7kVk9nE6nfX8+lzX9mtv/OlW+g3tjcAFg83/EJ8+V3AXGLG9uKul+ZWlGvXvS1PzXmFrV/vas7q12lU5w78bcpYTGLisz0HWbzVvl7hPt48d/UUvN0tmEV4ccMWfkw4zvCO7XlwwkhczWZKy8t5Yd1mth8/WcdWmt+YdlEsGDkeswifHj7Am3t22q1v6+XNwvGX4+NmwWQy8dz2H9mUfIyYkDCeGTMZMLokr+zeyppjCQ6IoKZRHTvw94ljMZtMLNt3kEXba7bV81dMwcfdgkmEFzdt4Yek44yIas9DY0fiajJTai3nuY2b2X7CidoqMooFwyYYbRW3nzf3VWsrT28Wjp1mtJWYeG7XD2w6eYyY4DCeGTUFsLXVL1tZczzeARHU1IRn4O0CuohIR+AUMBu4vlqZZGAC8L6I9ADcgSxgDfBnEfEAzgNjgJfr25jTJ2OTycR9r83jkcn/Ijsll9d2PsO2FbtJPpxSUeatP/234vHV904lun9HAPZtiuXOAQ8D4O3vxfvx/+Hn7/c1bwB1MImwYOp4bv34CzLy8ln+h+vZcDSRxOzcijJ3jRrK6kNH+eTn/XQOCmDRnBlM+M8SThcXc9fSr8ksKKRLcCDvXj+T0a8udmA0lUwiPDFqIjd+8xnphfmsuOZG1h5PJOF05fGLewdexsrEOD6K3Ue0fyDvT5vJyI8XE5ebzVXLP6RcKYI9PFl97S2sO55Iuar+l2HzMonw+OTxzF36Ben5+Xw+93o2xCeSkFPZVncPH8rqI0f53579RAcGsPjaGYx702irO5bb2iookCXXzWTU607UViMmceOqZUZbzbiJtScSSThTpa36D2NlUhwfHd5LtF8g70+9hpFLFxlt9eUHRlu18WT1Nbew7kSCw9sKmu4AnlKqTETuxUisZmCJUipWRJ4AdiulVgAPAotF5AGMIYy5SikFnBaRlzASugJWKaVW1rc9p0/G3YZEk5qQTvqxTAA2ffoTw68eZJeMqxo3eyQfPL6sxuujfn8Zu1bvoaT4/CWtb2P1bRvGidNnSDlzFoCVsXFM6NbZLhkrpfCyuAHgbbGQmV8IwOH0rIoy8Vk5uLmYK3rJjtYvJIwTZ09zMt+I65uEI0yO6myXjFHg5WoBwMfNjYyiAgDOlZVVFLGYXVBO8MUG6BtutNXJs7a2OhTHhC6d7ZIxSuHlZrSVV5W2OpRRpa2yc7C4mHEzmznvDG0VHM6JvCptlXiEyR2i7ZIxUBGXj5ulsq3Kq7SViwtO0lRA054OrZRahTFdreprC6o8PgSMqOO9H2FMb2uUC0rGIuIDdAGSlFKnL+S9v1VQRABZKZU7R3ZKLt2Hdqm1bEj7IMI6hrB3w8Ea68ZeN4LPX/7mktXzQoX6eJGel1/xPCOvgL4RYXZlXvtxO+/eMJMbB/ejjasrt370eY3PmdKjC4fTs5wiEQOEenqTWlgZV1phAf1Cwu3KvLx7Kx9e+Xtu6dMfD1dXbljxWcW6fiFhPD9uKhHePvxp/Sqn6GmFeXuRll8ZU3p+ATFt7dvq31u28951M7lpYD/auLlyyyc122pqty4cyshyikQMEOrpRWpB1bbKr9lWP//Eh9NmcUvPAUZbrars6PQLDuf5MVOJ8PLhT5uco62g5d4Dr96fEBH5SESCbI+nALHAc8BeEZnVDPVDavl/ravHNG72CDZ/vh2r1Wr3ekCYHx37tGf3GucYooA6DtNWi+uKXt34cl8sY159h/mffMXzM6bavS86OJCHxo9kwap1l7SuF6L2w8/2cU2P7s7yuFiGffg2t678nJcnTKt4397MdCZ/+j7Tl3/EXf2HYjE753UGqsd0Zc9ufHEwllFvvMMfln3Fi1dVa6ugQB4eO5IF3zl7W9mbHt2D5UcPMuyTt7j1u895eWyVtspKY/Ly95j+1YfcFeM8bVVqNTd6cSYN9edjfj0aCDwGjFJKTQQGYpx5Uquq00VSjPnSv1lWSi7BkYEVz4MiA8hJza217NjrRrDxky01Xh9z7XB++mon5WXO0SMBSM8rIMzHu+J5qI8XmQWFdmV+3783qw8dBWDvqTQsLi74e7Qxynt78dqsq3jk6zWcPH22+SregPTCfNp6VsYV7ulFZmGBXZnrevRhZUIcAL9kpGFxMRPQxsOuTOKZXIrLSukaEHTpK92A9PwCwr0rYwrz9qoYhvjVrL69WXXY1lap9m0V5u3FGzOv4uFv15B8xpnaqoC2XlXbyrtmW3Xrw8okW1tlpmIxuxDgXkdb+Tu+raBpT4duTg0lY5NtaALAinHkEFuCrnOIQym1SCk1SCk1KFI6XVQF43YlENElnLCoEFxcXRh73Qi2rdhdo1xk17Z4+XtyaNvRGuvGza49STvSgdR0ogL8ifTzwdVk4ope3dhw1P6HK+1sHsOi2gPQKSgAi4uZ3KJivC0WFs2ZwUsbtvBLSvWZNo61LzOdKD9/Ir19cTWZuCq6O2uPJ9qVSS3IZ0SkEVdnvwAsZhdyiouI9PbFbPtTKMLLh05+AaTk5zV7DNUdSLO1la+trXp2Y32CfVul5uUx3NZWnQMDcDNXaatZM1j4wxZ+OeVkbZWVRpRPlbbq3J21yfazV1IL8hjRtlpbnaulrXydo63AGKZo7OJMGhoz/iewUUReB34CPhORr4HxwHeXunIA1nIrr933Ls989zdMZhNr3tvIiUMp3PLP6zi6O5Ft3xiJedycEWz6tObNT0M7BBPcLoj9Pxyqsc6RypXiie828M71MzGL8Pm+WBKycvi/McM4mJbBhqNJPLv2R568chJzLxuAUoq/rFgDwI2DY2jv78fdo4Zy9yjjhKDbPv6C3KJiR4YEGHEt2LyeD668BrOYWHbkAPGnc3hg8AgOZKWz7ngiT27dxLNjJjOv70AU8NCG1QAMDo/grv6/o8xqxaoU//hxHafPOUdM//x+A0uuM9pq+f5YErJz+OOoYRxIy2BDQhLPbviRJy+fxNzBA0Ap/rLSaKubBsbQwc+Pe4YP5Z7hRlvN/dSJ2mrrOj64/PdGW8XZ2mqgra2SE3ly+yaeHTWFeX0GGW31g62tQiO4a8rMyrb6aS2nSxwfE7TcG5JKQ0esRSQauB3j9D4XjMnMXyml1jRmA5NMs5xjVL8Jtda7Q5cEOc8wTlNprXeHLg0sa7hQC3T89ocvOpNet+3ORuecT4e95TSZu8HZFEqpBOCRZqiLpmnaRSv7/+16xiJyZVNWRNM0rSm01gN49RncZLXQNE1rIi01GTc4TCEi3TGu4RmBMQ0xFVihlHrsEtdN0zTtgjlbkm2shk76eARYijE/fCfGedYCfFLbLUg0TdMcrbX2jOcBvZRSdrcLsV0AIxZ49lJVTNM07bdwtvnDjdVQMrYCbYET1V4Pt63TNE1zKmXWljmboqFkfD+wXkTiqbz9SHsgGrj3UlZM0zTtt3C24YfGqjcZK6W+E5GuGDfmi8AYL04BdimlWt8ZApqmtXitMhkDKKWswPZmqIumadpFU601GWuaprUkrfUAnqZpWovSaocpNE3TWpLyVjqb4qLFvz/oUm+i2XWbX/N6yq1B3Cv9HV2FJnc+qHVe3Uyrmx4z1jRNcwJ6mELTNM0JOMl9US+YTsaaprUqejaFpmmaE9AH8DRN05yAHqbQNE1zAno2haZpmhPQyVjTNM0J6KltmqZpTqCljhm3zMOOmqZpdbBaTY1eGiIiU0UkTkQSarvVnIi0F5GNIrJHRPaLyLRa1heIyEMNbUsnY03TWhV1AUt9RMQMvA5cDvQE5ohIz2rF/g4sU0r1B2YDb1Rb/zKwujH11sMUmqa1Kk14AG8IkKCUSgIQkaXA1cChqpsDfGyPfYHUX1eIyAwgCShszMZ0z1jTtNblArrGIjJfRHZXWeZX+aQIKm83B8ZdjiKqbe1x4EYRSQFWAfcBiIgn8Ajwz8ZWu8X0jMe07ciCwRMwi4lPE/bx5sEdduvbenqzcMQV+Li5YxLhuV9+YNOpJLv1a6f/gVf2/cTiQzubu/q1GjSpL3cuvAmz2cTq9zax7MVv7Nbf8fyNxIwx/iqyeLjhF+zDNWHziRnTkzuev7GiXLtu4Tx902ts++bnZq1/Xca0i2LBiPGYRfj08AHe3Gv//93Wy5uF4y7Hx2LBJCae2/Ejm5KPERMSxjOjJwPG/b1e2b2VNccTHBBBTWMio3jssgmYRVgat58391eLydObl8ZMw8fNgslk4rmdP7Ax5RgxwWE8M3IKYIvpl62sORHvgAhq1xrjupCesVJqEbCojtW1fVD10Y05wPtKqYUiMgz4UER6YyThl5VSBSKNq0+LSMYmEZ4YOokb135KelE+K6bdwtqTCSSczakoc2+f4aw8foSPju4l2jeQ9yfMYuQXb1Ws/8egCXbJ2dFMJuGeV+fy1yueITsll//89C+2f/sLyUdOVZR5+88fVTyeftdkovt1AGDfD4e4e+ijAHj7e/Je7Ev8su5A8wZQB5MIT4ycyI3ffkZ6YT4rZt7I2hOJJJyu0lYDLmNlYhwfHdpHtH8g70+byciPFxOXm81Vn39IuVIEe3iyetYtrDuRSLmDD4+bRPjX8EncsHqZEdPVN7EuOZH4M5Ux3dd/GN8ei+Ojw3vp4hfIe1OuYeSni4yYvvqAcqUIaePJ6pm3sC45weExQeuNy2ptsmGKFKBdleeRVBmGsJkHTAVQSm0TEXcgCBgK/F5Engf8AKuInFNKvVbXxlrEMEW/wHBO5J/hZMFZSq1Wvjl+mMntutQo5+VqAcDH1UJGUUHF65PbdSG54AzxZ7Obrc4N6Ta4M6mJGaQfy6KstJxNn21n2FUD6yw/7tphbFq2rcbrI2cOYdf3+ygpPn8pq9to/ULCOJF3mpP5trZKPMLkqM41ynm52drKzY2MQqOtzpWVVXyZLWYXlBN8sQH6BYdzvGpMSUeY1CHaroxS4OXqBoC3m4VM2/53rrxaTM1b9Xq11rhQ0vilfruALiLSUUTcMA7QrahWJhmYACAiPQB3IEspNUopFaWUigJeAZ6uLxFDC+kZh3p4k1qYV/E8rSiffkHhdmVe3reFDydexy3dB+Lh4soNa5cC0MbFlTt7D+XGtZ8yv9eQZq13fQLbBpCVUtkDyT6VS/fBNZMWQEj7IEKjgtm7MbbGurGzhvHFq406WNssQj29SS3Ir3ieVlBAv9BqbbV7Kx9e8Xtu6d0fD1dXbvjms4p1/ULCeH7sVCK8ffjT+lVO0dMK8/AirbBKTIX59A+2j+mVX37iw8tnMbfXADxcXLl+1bKKdf2Cw3lh9FQivHx4YJNzxAStN66mqoZSqkxE7gXWAGZgiVIqVkSeAHYrpVYADwKLReQBjCGMueo39iLqTcYikgt8AXwCbPitG7lYtQ25VK/I9KieLE88wDuHdjEgqC0vj7ySySve5YGYkbx7aDdFZaXNUtfGqjWmOv57x866jC1f7sRqtV8fEOZHVK927F67/1JU8TepdZCtWlzTo7uzPC6Wd/bvZkBoOC+Pn8bkZe+hgL2Z6Uxe9j6d/QJYOP5yNp08Rkl5ebPUvU6N2f8692B5/EEWH9jNgJC2vDJ2GpM+t8WUlcakz98j2i+AhaOnsSklyfExQeuNqwmzlFJqFcaBuaqvLajy+BAwooHPeLwx22pomCIL2As8AaSIyKsicllDH1r1CGX+xh0NFW9QemE+bT19Kp6He3hX/Ln0q+u69GXl8SMA/JKdisXsQoC7B/2CwvnrwLFsmXknt/UYxD19LuPmbgMuuk4XK/tULsGRgRXPgyICyEk7U2vZMbNqH6IYfc1Qtq7YTXmZE3wBbNIL82nr5V3xPNzLq2Zbde/DysQ4AH7JSMPiYibA3cOuTOKZXIpLS+kaEHTpK92A9MICwj2rxOTpbTcMBnBdtz58m2SLKbNy/6sq4UwuxWWldPV3fEzQeuNSShq9OJOGknGhUuo1pdQIYBhwCnhDRJJE5Om63qSUWqSUGqTeQNqrAAAgAElEQVSUGuQ9buhFV3JfThpR3v5EevniajJxVVQP1p60P8qeWpjHiHDjAFdn30AsZjM554q4ds3/GPnFW4z84i2WHN7N6we280HcLxddp4sVtzuJiOgwQqOCcXE1M3bWZWz/tuZsiMgu4Xj5e3Joe80j1WOvHV5rknakfZnpRPn6E+lta6vO3Vl7PNGuTGpBPiMi2wPQ2S8Ai9mFnHNFRHr7Yrb9yRDh5UMnvwBS8vNqbKO57ctKo6OPP+1+3f86dWftiWr7X0EeI9oaMUVXiamdV7WYfJ0jJmi9cTXZWR/NrKEx44qfDqVUMvA88LyIdMMYzG4W5UqxYOdaPph4LWYRliUcIP5sNg/EjORATjrrUhJ4cvcGnh02lXk9BqNQPPTTqoY/2IGs5VZev/99nv7mEUxmE9//9wdOHD7FzQuu4ejPx9i+0vjBGHvdMH6oJeGGdggiODKA/T8ebu6q16tcKRZsWc8HV1yDWUwsiztA/OkcHhg0ggNZ6aw7kciT2zbx7JjJzOszEAU8tNEY8x4cFsFd/X9HmdWKVSn+sXkdp88VOzYgbDFtXccHl//eiOnoAeLP5PCnASPYn53OuuREntyxiWdHTWFe70Eo4MEfjZgGhUVwd8xMSq1WlFL8fetaTpc4PiZovXGppptN0aykvmFgEXlJKfWni9lA1AfPOdnvz8XrNt95xmibUmu8O7RyaXW7X6t24g8PX3QmvZCcc/zmR5wmc9fbM77YRKxpmtbsWujv72+eZywiVzZlRTRN05pECx0zvpiTPgY3WS00TdOaStOd9NGsGjzpQ0S6Y1ypKALjtyQVWKGUeuwS103TNO2COcm5Jxes3p6xiDwCLMWYVbET4/RAAT6p7ULLmqZpDmeVxi9OpKGe8Tygl1LK7vQ1EXkJiAWevVQV0zRN+y2kNfaMASvQtpbXw23rNE3TnEsLPYDXUM/4fmC9iMRTeZHl9kA0cO+lrJimadpv4mQH5hqroXnG34lIV4zbj0RgjBenALuUUs5zQQRN07RfOVmPt7EanE2hlLIC25uhLpqmaRevhQ6gtojrGWuapjVaaxym0DRNa2la6mwKnYw1TWtdWmgybhH3wNM0TWvtLnnP2OTS+iZdnF8V5ugqXBI97slpuFALc+RvPg0XaoE6RWQ5ugpOSw9TaJqmOQMnO825sXQy1jStddE9Y03TNMfTwxSapmnOQCdjTdM0J6CTsaZpmuPpYQpN0zRnoGdTaJqmOZ7uGWuapjmDFpqM9enQmqa1KqIavzT4WSJTRSRORBJqu++niLQXkY0iskdE9ovINNvrk0TkZxE5YPt3fEPb0j1jTdNalybqGYuIGXgdmITtphoiskIpdahKsb8Dy5RSb4pIT2AVEAVkA1cppVJFpDewBuMGHXXSyVjTtFZFmu7i8kOABKVUEoCILAWuBqomYwX8egEUXyAVQCm1p0qZWMBdRCxKqZK6NqaTsaZpWu0iqLz3Jxi946HVyjwOfC8i9wGewMRaPucaYE99iRhaUDIeHd6JBQMnYRJhWeI+3jq0zW59Ww8fXhh2FT6uFsxi4vl9G9mUmmi3fs0V83n1wGbeObKjuatfq8EB3bi3ywzMYmJl2g4+ObHBbn2IxY+/9JyDl0sbTCIsTlzJjpwjhLr789+hj3CyKBOAQ3kneDnuc0eEUKuBI7ty56NXYjKZ+G75Lj575we79fP/cgV9h3QCwNLGDb8AT2YNfQKA2x6aypAx3RER9mxN4K2nv2n2+tdmdHgnHhs8EZOY+DRhL2/F2t+JrK2HDy8OvxIfN2P/e27Pphr73/dX3c6r+zez+PDO5q5+nQYHdOeeLjMwYWJV2naWJtfcBx/pcT2eLu6YxcTixJXszD1MqLs/7w35S8U+eDjvBK8cXe6IEGq6gGEKEZkPzK/y0iKl1KJfVzfi0+cA7yulForIMOBDEeltu10dItILeA6Y3FBdWkQyNonwz0FTuHnDJ6QX5/HVlFtZlxJPQl52RZl7eo9g1YnDfJzwC9E+QSwZey2jV7xRsf7vAybyQ1pibR/vECaEP3abycN73iar5CxvDbqfrVmxnCjKqChzU9RENmXuZcWpbXTwCOXZmD8wZ9tTAKQWZ3P7rpccVf06mUzCPf+YzqPz3iU7I49Xl93Djo2HSU7MrCiz6NmVFY+n3zCMzj3aAtCjX3t69u/A3Ve/CsCLH99Jn8EdObDrWPMGUY1JhCeGTOam9UtJL8rj68vnGvvf2cpLjt7bZzgrTxzm4/g9RPsG8t64axn11ZsV6/8+aAI/pDrP/gfGPvh/XWfy571vkVVyljcGPcC2bPt98IaoSWzK3Ms3qVvp4BHK031v54btTwLGPnjH7oWOqn6dLmRqmy3xLqpjdQrQrsrzSGzDEFXMA6baPmubiLgDQUCmiEQCXwI3K6UabPwLnk0hIkEX+p6LFRPYlhMFpzlZeIZSq5VvTxxiUmQXuzJKgZerGwDebhYyigsq1k2K7EpywRniz2bjLLr7tCe1KIe0c7mUqXI2ZO5hRHAvuzIK8DC7A+Dp4k72+TwH1PTCdO3bjtTkHNJTTlNWWs4Pq/Zx2fgedZYfc0UMm1btA4x43SyuuLiacXVzwexi4kxOQZ3vbS4xgW05kX+akwXG/vfN8cNMiuxqV0YBXq4WALxd3avtf104WXCGo060/4GxD54qzq7YBzdm7GF4UG/7QsrY98D4N+f8WQfU9AKpC1jqtwvoIiIdRcQNmA2sqFYmGZgAICI9AHcgS0T8gJXAX5VSPzWm2vX2jEXkcuAN4BRwH/ARtoFo4Bal1PrGbORihbXxJq2wMhGlFeXTL6itXZlXD/zIB+PncHO3QXi4uHLT+k8AaGN25Y6el3Hzhk+4vcdlzVHdRgmy+JJZcqbieVbJWXr4tLcr8/6xNbzQ7w5mRo7E3ezGQ3vfrlgX1iaARYP/RFHZOd5NWs2Bs47tPf4qKMSHrPTKL2x2Rh7d+rartWxIWz/CIv3Zt93oNBzZm8z+HYl8/OOjiAjffLyNk0mOv4h6mIcXaUWV+196LfvfK/s388H42dzSbSAeLq7cuH4pYOx/d/Yaxk3rP+H2ntWHGx0ryOJL1rmq++AZevh0sCvz3+Pf8VzMncyIMPbBh/e+VbEurE0Abw36E0XlJbyXtMpp9sGmmk2hlCoTkXsxZkKYgSVKqVgReQLYrZRaATwILBaRB2xbnquUUrb3RQP/EJF/2D5yslIqs5ZNAQ0PUzwDTAP8gHXAFUqp7bZfgI+BAbW9qeo4TOC8q/EZP6RRwV8IVe0/fHpUL5Yn7efdIzvpHxTBwuHTmbpyEff3HcWSI7soKitt8jpcjFoHo6rFNCG0P9+l7eKzkz/Q06cDf+05h9t2vEhuSR6zf3qSvLIiunpH8q8+t3LrjucpKq/3+EDzaExgNmOm9WXLmoNYrcb68PaBtOscwk3jngXg6Xfn0funKA7uPn6JKts4UktQqlpM06N68nnSAd45bOx/Lw2/iinfLuaBmFEsObzT6fY/Qy1xVctk40MH8H36zir74PXM2/kCuSV5XL/1X+SVFdHFK5In+tzKvJ3OsQ824WwKlFKrMKarVX1tQZXHh4ARtbzvSeDJC9lWQ8MUVqXUYaXUNqBIKbXdtqHD9b1XKbVIKTVIKTWoKRJxenE+4Z6Vt88J9/AmszjfrsysTjGsSj4MwJ7sU1jMZgIsHvQLjOAv/cbx4/S7ubXbYO7uNZybug686DpdrKySs4RY/CqeB1t8a/wJOC18KJsyjT/hD+WdwM3kiq+rJ6WqnLyyIgCO5qeQWpxNpEdw81W+HtkZeQSH+VY8Dwr1ISez9uGVMZfHsGnlvornwyf25Mi+k5wrOs+5ovPs3hxH95j2tb63OaUV5RPuUbn/hXl42w1DAFzbOYaVJ2rZ/4La8pcB49g84y5u6z6Yu3sP52Yn2P8AskvOEOxedR/0I6fEvq0ur7YPutayD8YXpJBanOM0+2BTnvTRnBpKxmdE5A4ReRg4LSIPiEiEiNwCNNtg3v6cVKK8/Yn09MXVZOLKDj1ZdyrerkxqUR7DQ6MA6OwTiMXkQk5JEdet+5DRK95g9Io3eC9uF2/EbuXDoz83V9XrdCT/JBEeQYS5B+AiZsaH9GdrdqxdmYyS0wzwN8bG23uE4GZy4UxpAb6unphsvZpw9wAiPIJJK3aO+9cdPZBC2w5BhEb44+JqZsy0GLZvPFyjXERUEF6+bTi8N7nitay0M/QZ3BGT2YTZxUSfQR05mVjnX3XNpvr+d1VUD9alVNv/CvMYHhYF2PY/s7H/Xfv9R4z66k1GffUmS47s4o2DW/nACfY/sO2DbYIr9sFxof3Zmn3Qrkzmucbtg5EewaQV5zZ7DLVqujHjZtXQMMUtGGeYWDGmZszBGD85Adx+aatWqVwpHt/9Pf8dNxuTmPgsaR/xZ7O5v89oDuSmsf5UPE//sp6nh17Obd2HoICHt3/bXNX7TazKyr+PfsHz/eZjEmF16k6OF2Zwa8cpxOWnsDU7ljfjv+Gh7rOY1W40CsVzh41xyBi/TtzacSrlyko5Vl4+spz8smIHR2Swllt588kVPPnObZhNwvdf7CY5IZOb7pvI0YOn2GFLzGOviOGHVfvs3rtlzUFihnbmza//CEqxe0s8OzYdcUQYdsqV4rFda/lgwmxMInyWuJ/4s9k80HcUB3LTWJeSwFO/rOeZodOY12MwSsHD21Y2/MEOZlVW/nP0C56LmY9JTKxO28mJogzmdpxKXN5JtuXE8lbCCv7U/VquaTcGpRTPHzaOxfT168xc2z5oVVZeifuMfFtP2eGcLMk2llQf+2pqnf73dAv9r6lbhzDn6IU2Nfd73BxdhSan7w7dsqwf99JFX/+y119fbnTOiX3mAae53uZvvlCQiFzZlBXRNE1rEi10mOJirto2uMlqoWma1kTE2vjFmTR4Bp6IdMe4OEYExm9JKrBCKfXYJa6bpmnahXOyHm9j1dszFpFHgKUYExJ3YpyRIsAntV3bU9M0zdFa6tS2hnrG84BeSim7Gesi8hLGZeGevVQV0zRN+02cLMk2VoMnfQBta3k93LZO0zTNubTQA3gN9YzvB9aLSDyV1/Vsj3HO9b2XsmKapmm/hbMNPzRWvclYKfWdiHTFuOJ9BMZ4cQqwSylV3gz10zRNuyCtMhkD2C6SvL2hcpqmaU6htSZjTdO0FkUnY03TNMdrtcMUmqZpLYpOxpqmaY7nbKc5N9YlT8ZDOx2/1Jtodp4u5x1dhUsi8rPTjq5CkyuLcXd0FS4Jv80Bjq6C09LDFJqmac5AJ2NN0zQnoJOxpmma4+lhCk3TNCcg1paZjXUy1jStdWmZuVgnY03TWhc9TKFpmuYMdDLWNE1zPN0z1jRNcwY6GWuapjmePh1a0zTNCbTUYYqG7oGnaZrWsijV+KUBIjJVROJEJEFE/lLL+vYislFE9ojIfhGZVmXdX23vixORKQ1tS/eMNU1rVZqqZywiZuB1YBK2282JyAql1KEqxf4OLFNKvSkiPYFVQJTt8WygF8ZNndeJSNf6blfXYpLx2f25pHycAFZF4Jhwwq5sb7f+fM45ji86QnlROcqqiLi2I74xgQAUJRdw8v2jlBeXg0no/tgATG6O/6Mgd99pkj48jrIqwsaG0m56hN36c9klHH0rgbKiMpQVOs5uT0A/fwAKkwuJfzfJiEmg/7/6OkVMAKl7CvhlSSbKqug8wY+eMwPt1hdmlbL9P2mcLypHlUO/G4NpO9ALgNPHz7Hr7XRKi6yISZjyXAfMThDXoCkx3P3yXExmE6vf3cCnz39tt/7OhTfTb2wvACwebviF+PK7wNuIGduLuxbeXFGuXfe2PHX9q2z9enez1r8u+QdyOPW/eLAqAkaHE3JFlN368znnOPnOIcqLysCqCPt9Z3xigjifXUzcozuwhHkA4NHZh8hbujsgglo03TDFECBBKZUEICJLgauBqslYAT62x75Aqu3x1cBSpVQJcExEEmyft62ujbWIZKysipMfxNPlz31xDbAQ9/gv+PYPpE2EZ0WZtK+T8R8SQvCEthSfKiTxpQP4LgxElSuOv32EqDu649Hei7KCUsRFHBiNQVkVie8fo/dfe2IJcGPvPw4QMMAfz0iPijInv0oh6LJA2k4MozCliNgXjjDkVX9UueLIGwl0uysarw6elOY7R0wA1nLFz4szGLegHW0CXfn+keNEDPbCt52lokzs8mzaD/emy1R/zp4s4YenTjJ9YDTWcsW2V9MY9sdw/KPcKckvR8yOj8tkEu77z208MuUpslNyeG3HM2z7ZjfJh09VlHnrwQ8qHl99z1Si+0cBsG9TLHcOfAQAb39P3j/6b37+fn+z1r8uyqo49WEcHR/qj2uAhYQnduPTLxj3Kt+rzG+O4zc4hMDxkZw7Vcixl/fhExMEgFtIG7o+McRR1a/ThRzAE5H5wPwqLy1SSi2yPY4ATlZZlwIMrfYRjwPfi8h9gCcwscp7q947NMX2Wp0c3+VohMKkPCyhbbCEtMHkYsJ/aAhnf8mxKyMC5efKACgvLsPVz/jy5x3MpU07TzzaGz0vFy9XxOT4L3h+YgHuoe60CXHH5GIi+LIgcn+udj1hwej5Yvzr5u8KwOkDZ/Bs74FXB+NL4+rtHDEB5CacwyvMDa8wN8yuQvuRPqTsKrAvJEJpsfGNKS2y0ibAiCt9byF+URb8o4xrEFu8zZicIBl3GxJNamIG6ccyKSstZ9OnWxk+fXCd5cfNHs7GpT/VeH3U7y9j13d7KSl2juthFyXl4RbiUfG98hsSQt6erBrlKvfBMlz93Jq7mhdMrI1flFKLlFKDqiyLqn5ULR9fvd89B3hfKRUJTAM+FBFTI99rp9E9YxHxB8qUUvmNfU9TKT19HreAyp6Va4CFosQ8uzLhv+tA/AsHyFp7CmuJlS5/7gtASXoxIhD/wn7K8kvxHxpM2BX2QxyOUJJ7HktgZUxuAW7kJ9r/13aY2Y4Dzx4idU061pJyej/aE4DitHMIcODZQ5TmlxF8WSDtrqr3R7fZFOWW4hFUuVt5BLiQE19sV6bPdUFsfOIkR1edpqzEyvjHjPbISzuPABufOElJXhntR/rQc4b9EIcjBEUEkHWy8sc/+1QO3YdE11o2pH0QYR1D2LvhYI11Y68dzuevrLxk9bxQpadLcG3gexU6oyPHXtxLzvoUrCXldHy4X8W681nFHH1sJ+Y2LoTN7IRnV79mq3u9GnFgrpFSgHZVnkdSOQzxq3nAVGOzapuIuANBjXyvnXp7xiLSVkQ+EJGzQDYQKyLJIvK4iLjW8775IrJbRHYnfHW4vk00Tm3/t9V+d3K3ZxI4MpQ+rwyj84O9Ob7oCMqqUOWKgqN5dLyzB93+1o+zP2eTF+ukd7QQ+6Ayt2UTOjqEoa8NpNefexD3RoIRk1Vx9mg+3e/pQsyCXuTszuX0wbMOqnQ1jWirE5vz6DjOhxmLoxn7t3Zs+3dqRVtlHSlm+P3hTHyqAyk78knfX9gs1a6PSM1OTl3f93HXDWfz5zuwVrtyWECYHx37tGf3mn2XoopNp1qsZ3Zk4D8ynB4vjSDqgRhOLj6EsipcfC30WDiCrv8cQvjsaJLfiqW8uMxBlbYnqvFLA3YBXUSko4i4YRyQW1GtTDIwAUBEegDuQJat3GwRsYhIR6ALsLO+jTU0TPERsEQp5QvMAj4HemD0qF+v601Vu/7RM3o0sImGuQa4cT63pOJ5aW5JxTDEr3J+SMd/SDAAXtG+WEutlBWU4hpgwau7Ly7erpgsZnxiAik+Ue3PZgewBLhRklMZ0/nc81iq/QmYsSmT4MuMnqFPF29UqZXS/DLcAtzw7e6Dq7crZouZgH7+FB53fEwAHoGuFGVXfimLcssqhiF+lbj+DO2HG8c8grq1ofy8oiS/HI9AV0J6tsHi44KLxUTbAV6cTjrXrPWvTVZKDsHtKnvoQRGB5KTW/oM+9rrahyjGzBrGT1/tpLyszoPpzc7V30Jpje+V/T6Y+2MavoNDAPCM9kWVWikvKMXkasLFy2hXjygf3ELaUJJe1HyVr4+6gKW+j1GqDLgXWAMcxpg1ESsiT4jIdFuxB4HbRWQf8AkwVxligWUYB/u+A+6pbyYFNJyMA5VSm2wV+wIYrZQqVEr9HRjdwHubjGdHH0oyiinJKsZaZuX0jkx8+9v/+eoWaCH/0BkAilMLUaVWXLxd8enjT/HJQqwl5UYv+cgZ3Nt61LaZZuXdyYtz6ec4l3kOa5mVrO3ZBAz0tytjCbRwxtbjLTpVhLXUiquPC/59/Sg8WUS5Laazh/PwiHB8TAAB0e7kp52nIOM85aWK5C15RA7ysivjGexKhq3HezalBGupwuJjJryfJ2dOlFBWYsVarsiMLcKnnaW2zTSruF2JRESHERYVjIurmbHXDWfbNzVnQ0R2DcfL35ND247WWDdu9gg2Lt3aHNVtNI+O3pzPLOK87Xt1ZmcmPv2D7Mq4BVooOGz88JxLLcRaasXs7UpZ3nmUrfdfkllMSUYRbsFtmj2G2jRhzxil1CqlVFelVGel1FO21xYopVbYHh9SSo1QSsUopfoppb6v8t6nbO/rppRa3dC2GhozzhKRG4ENwDXAcQAx/m5rtoN/Yhba3RRNwgsHUFZF4Ogw2kR6kvrFMTyivPEbEETEnM4kLzlK5poUEOjwh26ICC6eroRMieTI47+AgE9MAL79HD8OKWah89yOHHzuMMqqCB0TgmekB8eXJ+Pd0YvAgQF0vKEDCe8kceq7NAC63hGNiODq6ULk5eHs/ccBEAiI8Segv38DW2weJrMw6A+hbPrXSZQVOo33xbe9hf2fZBEQ7U7kYG/63xLCzjfTifv2NAgMvTccEcHNy0y3qwJY8+fjiED4AC8iBno1vNFLzFpu5bX/W8Izqx/FZDax5r1NnDiUwi2Pz+Loz0ls++ZnwEi4mz6tmXBDOwQT3C6Q/T8cqrHOkcRsou0NXUlauBesCv9RbXGP8CL9yyTaRHnj2z+Y8Ou6kPL+EbK/NyYVtJvXAxGh8OgZ0r88Zsx2EYi8pXtFT9nRWurF5UXVM9gtIu2BF4GewF7gYaVUmogEAmOVUp83tIE52+e3zP+ZerTau0O7O+lY+kX4Sd8dukX5bPibFz19ZvT0Fxqdc35c8bDjp+vY1NszVkolA9fW8noOxvixpmmaU/n/7toUInJlU1ZE0zStSVhV4xcncjHjvnXPetc0TXOUJppN0dwaPOlDRLpjnGcdgVH9VGCFUuqxS1w3TdO0C9YqhylE5BFgKca0/Z0Yk6AF+KS2y8lpmqY5mlhVoxdn0lDPeB7QSylVWvVFEXkJiAWevVQV0zRN+02cK8c2WkNjxlaMa3FWF25bp2ma5lREqUYvzqShnvH9wHoRiafyUnLtgWiM0wQ1TdOcSwvtJjY0z/g7EemKcVHkCIzx4hRgV0PnWWuapjmCs/V4G6vB2RRKKSv2F0nWNE1zXi0zF7eMO31omqY1lrPNkmgsnYw1TWtdWuswhaZpWktyIffAcyaXPBn7uhY3XKiFKbG2zt+wru7pjq5Ck1twqvVdiQ7gioFTHF2FS+NUw0UapHvGmqZpTqBl5mKdjDVNa13E2jLHKXQy1jStdWmZuVgnY03TWpdWe9KHpmlai6KTsaZpmhPQyVjTNM0J6DFjTdM0x9OzKTRN05yBHqbQNE1zAjoZa5qmOYGWOUqhk7Gmaa1LS51n3NA98DRN01oWpRq/NEBEpopInIgkiMhfaln/sojstS1HReRMlXXPi0isiBwWkX+LiNS3rRbTM87Zd4b4D5LBqggfF0yH6fb3ST2XXcLht5IoKyxHWRWdZ7cjsL8fAAXJRcS9c4yyYitigoH/6oXZzfG/Q6f3nebYh8fACiFjQ4icHmm3viS7hPi34ykvMmLqcF0H/Pv5A1CYXEjikkTKi8sREfo+0ReTE8QEkLC7lO8WFWO1woDJboy81t1u/dlMK1+9VMS5QoXVqpg4tw1dBruyf+N5tn5+rqJcxnErd7zqRVhnx++mm3fA0/8BqxV+fwXcfoP9+tQM+OvTkF8A5Vb40x0w5jI4XwqPvwgH48BkgkfvgyH9HRNDbQaO7cGdT1yDyWTiu0+28dnra+3Wz398Jn2HdwHA0sYNv0AvZvV8BIDb/nY1Qyb0QkzCnh+P8NaCz5u9/rUqb5pxChExA68Dk7Ddbk5EViilDv1aRin1QJXy9wH9bY+HAyOAvrbVW4AxwKa6tuf4vbwRlFVx9L0T9PtrNyyBbuz+eyxBA/zxjGxTUeb4l6mEDA0gYlIohSnF7H8+jmH9+2EtVxx6PZGed3fGq4MHpfmlmFzq/YFqFsqqSPpvEr3+0gu3ADf2L9hPwMAAPCI8KsqkfJ1C0NAgwiaGUXSqiMMvHGbgKwNR5Yr4N+PpcmcXPDt4UppfijhBTADWcsWqN4u56UlPfIJMLH4gn26XuRLc3lxR5sel5+g5ypXBV1jISi7n48cKuP89X/qOc6PvODcAMo6Xs/SJQqdIxOXl8K9X4N2FEBoM194B40ZAdFRlmbc+gKnjYM4MSDgOdzwC6z+Fz7411q94H3JOw/w/w2dvG4nZ0Uwm4Z6nZvHonNfJTjvDq6seZsf3B0iOr7yU6qLHv6h4PP3W0XTubXQYegzqSM/Bnbh74jMAvPjVA/QZFs2BbQnNG0Rtmm6YYgiQoJRKAhCRpcDVwKE6ys8BHvu1FoA74IZx71BXIKO+jTnBLtGwvIQC2oRaaBPqjsnFROiwQLJ/tr9OrQiUFRv3SC0rKsPN3/hSn95/Fq/2Hnh1MJKcq7crYnJ84ipILKBNaBvcQ4yYgi4LIvfn3BrlyorLACgvKq+I6cyBM3i088CzgyfgPDEBnDpaTkBbE/7hZsyuQq/RbhzZXmpfSKCkyPjCnCtUeAYjFwUAAArfSURBVAfU3A0P/nCe3mNcm6PKDdp/GNpHQLu24OYK08bDhi32Zf5fe/ceHFV9BXD8e7KbBPIi2WwSSIwEMDwkyCOMgE6D46slraJ0OsWxtGWGARWd6tQ+aNFpO7VKX3/YaZlhoOrUV4VKKRYVWwWKApU3eQlGMCSQkIRAEvLa7J7+cZeQQF6YJXdz+/vM7GR/d3937/lld8/87rl394pAY5N1v6ERUpOt+6UnYHaudT85CRLirFlyOBg/fTSnTtRQWVZLu8/P9k37mP3lKT32n3tfLtv+vg8AVSUq2o07yk1klBuX28W56obBCr13oStTZAAnO7XLg8uuICKjgTHA+1YIugv4ADgdvL2rqsW9bewLTztEZLGqvvBF178arXU+hiVHd7SjPVHUf9rYpU/W1zM49NwnVGytwt8SYNpPJgLQVNkCAgefLcHX0E7qnGRG3zNqMMLuVWtdK1GeqI52lCeKxtKuY8pckEnRqiIqt1bib/UzecVkAJormxERilYV4av34Z3jJeNr3b5HBl1DbYAE76XkmuCNoOKT9i59bntwGC+vbOS/m1vxtcCiZ2KveJ7CHT4WPnXlcjucqYGRqZfaaSlWgu5s+WJY8n145U1oboY//95aPnGclbjzb4fKaig8CpVn4KZJgxd/T7wjE6k+dWlSU3P6HBOmZ3XbNzUjiZGZyRz68CgAJftOcPijY7yy/5eICJtf3MHJT3ud+A2eq7gGnogsBZZ2WrRGVddcfLibVXp68oXABlX1B5/3BmAScLH2+J6I5Knqjp5iGcg+4M+BQUnG3Q7/sn9T1Ue1jMzzcv1XR3H+aANFq0u5edUU1K+c/6TBqhNHR3DwmRLix8TgyRkxKKH3qB/vl5pdNaTkpZCRn0HDsQaOrT7GtOemoX6l/mh9R5248NlCYrNiScxJvPZx96E/e4gF29uYemcUtywYxsnidjb+rolH/hTfMbsvL2knMhpSs1x9PNPg6G5Ml39Kt/wL7p8Hi78JBwrgR89YpYkF+VBaBt9YBulpMG0yuMJjWD2kmu5fwLnzc9n5z4MEgoluVJaXzOyRLJr5FAC/ev1RcmaNo2BP6bWKtv+0/zXjYOJd08PD5UBmp/Z1wKke+i4Elndq3w/sVtVGABF5G5gN9JiMey1TiMjhHm5HgLRe1lsqIntFZG/Rm8d620S/RHsiaalt7Wi3nm0jOimqS5/T22pIne0BYMT4eAJtiq+hnWhPFImTEohKiMQV7SJ5WiKNx5sGHNNARXuiaTvb1tFuO9vWUYa4qGp7Fd5ZXgDis+MJ+AL4GnxEe6JJmJhAZLw1pqSpSVw4cWFQ4+9JgjeC+ppLH4b6mgDxyV3fZge2tjH5S9ZYMye5aW+DpvpLSaBgh4+cuV3/F3ZKS7FmsxdVVUOqt2ufDVusmjHA9BxobYO68+B2w4pHYeM6+GPwAN/orsdpbVNz+hwp6Ukdbe+oRGqrznfbd+78GWzbtK+jfctXplKy/zgtTW20NLWx9/0iJs7IutYh948/0P9b7z4GskVkjIhEYSXcf1zeSUQmAEnArk6Ly4C5IuIWkUisg3e9lin6qhmnAd8G7unmVtvTSqq6RlVnqurMGxdk97GJvsWPi6O5spXmM60E2gNU7arFm9t1FjjMG0VdQT0AFyqaCfgCRCa48dw0gsayJvytfgJ+5VxxAzGdDvzZJW5sHM2VzbScaSHQHqBmdw2eGZ4ufaKTozlfaH04miqagmOKJPGmRJqCY1K/Ul9Sz/AM+8cEkDHeRW1FgLpKP36fUrijjQmzutZ+R6REcPygVUeuLvPT7lNiRljTNA0oRTvbyMkLj3oxwJSJ8Hk5lJ+2zo7Y8r51AK+z9FTYHcxVpSesZOxJhOYWaApeBvLDj61ZcecDf3Y6erCM9DEppGUm4450MXd+Lru3HrmiX8a4VOJGxFC893jHsupTdUyZnU2EKwKXO4Ipc24InzJFiGrGqtoOPAq8i5VI31DVQhH5hYjc26nrA8Drql2ecANQChwBDgGHVHVzb9vrq0zxFhCnqgcvf0BEtvWxbshEuITx3x3NoedK0ACMui2F2Oti+Gx9OQljY/HmJnHDg9dTsvY4J9+uRESY9NBYRITIODeZ+SPZu7IIEfBMG4F3uv278+ISxn5nLEW/LkIDStrcNGKui6FsQxlxY+Lw5HrIejCL0rWlnHrH2jPKXpaNiOCOdZM+L53DTx8GgaSpSXime/rY4uCIcAn5Dw/n5acuoAGYdlcUqaNdfPCXZtKz3UyYHcndS4az+fkmdm+y9nbueyKGi6dgfl7QToLXOgAYLtxuWPk4LHnSOrVtQT5kj4Hn10HORLj9Vvjhcnj6N/DSeutg3rMrrL9n62DJDyBCIDUFVv3U7tFcEvAHWL1yPb989RFcEcLWv+6m7Ggli57M5+ihMva8VwDAbfNz2b5pf5d1d751gKm3ZrP63ytAlb3bijv62y6EX/pQ1S3AlsuWPX1Z+2fdrOcHll3NtkRDGHh3Htq3aGh+HaYXTr069F1JPZ2xM3QtjDNXhx5K3q74w4BPC5qX8Vi/c04othcqzswqhmH8/zI/oWkYhhEGhuhvU5hkbBiGs4To69CDzSRjwzAcRa/iPONwYpKxYRjOchXfwAsnJhkbhuEspmZsGIYRBszZFIZhGGHAzIwNwzDsp36/3SF8ISYZG4bhLOYAnmEYRhgwp7YZhmHYT83M2DAMIwyYmbFhGIb9huoBvGv+E5qDSUSWdrp+lWM4cVxOHBM4c1xOHFM4GhJXh74KS/vuMiQ5cVxOHBM4c1xOHFPYcVoyNgzDGJJMMjYMwwgDTkvGTq1rOXFcThwTOHNcThxT2HHUATzDMIyhymkzY8MwjCHJMclYRE6IyBEROSgie+2OJxREJFFENohIiYgUi8gcu2MaKBGZEHyNLt7qReRxu+MaKBF5QkQKRaRARF4TkWF2xxQKIvK94JgKnfA6hTPHlClE5AQwU1Vr7I4lVETkJeA/qrpWRKKAGFU9Z3dcoSIiLqACmKWqn9sdzxclIhnATuBGVW0WkTeALar6or2RDYyI5ACvAzcDbcA7wMOqeszWwBzKMTNjpxGRBCAPWAegqm1OSsRBdwClQzkRd+IGhouIG4gBTtkcTyhMAnarapOqtgPbgfttjsmxnJSMFdgqIvtExAknqY8FqoEXROSAiKwVkVi7gwqxhcBrdgcxUKpaAfwWKANOA+dVdau9UYVEAZAnIskiEgPkA5k2x+RYTkrGt6rqDGAesFxE8uwOaIDcwAxgtapOBy4AP7Y3pNAJll3uBdbbHctAiUgSMB8YA6QDsSLyLXujGjhVLQZWAe9hlSgOAe22BuVgjknGqnoq+PcMsBGrzjWUlQPlqron2N6AlZydYh6wX1Wr7A4kBO4Ejqtqtar6gDeBW2yOKSRUdZ2qzlDVPOAsYOrF14gjkrGIxIpI/MX7wN1Yu1hDlqpWAidFZEJw0R1AkY0hhdoDOKBEEVQGzBaRGBERrNeq2OaYQkJEUoN/rwcW4JzXLOw45Sc004CN1ucAN/Cqqr5jb0gh8RjwSnCX/jNgsc3xhESw/ngXsMzuWEJBVfeIyAZgP9Zu/AGc8621v4lIMuADlqtqnd0BOZVjTm0zDMMYyhxRpjAMwxjqTDI2DMMIAyYZG4ZhhAGTjA3DMMKAScaGYRhhwCRjwzCMMGCSsWEYRhgwydgwDCMM/A/Oy7xVs7oDWwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the mean cross-validation scores\n",
    "sns.heatmap(scores, \n",
    "#             x_label='max_features', \n",
    "            xticklabels=tree_params['max_depth'],\n",
    "#                       ylabel='max_depth', \n",
    "            yticklabels=tree_params['max_features'], \n",
    "            cmap=\"viridis\",\n",
    "#              square=True,\n",
    "            annot=True\n",
    "           )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}