3. Hyperparameter Optimization.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here I test a number of options for the classifier. Since our classifier is an ensemble of Random Forests, we'll just try to test with Random forests, and assume the conclusions can be extended to the ensemble."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import Modules"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**General Modules:**\n",
    "- [Pandas](https://pandas.pydata.org/): Library for Python data analysis\n",
    "- [NumPy](http://www.numpy.org/): The fundamental package for scientific computing with Python\n",
    "- [MatplotLib](https://matplotlib.org/): plotting library which produces publication quality figures\n",
    "- [scikit-learn](https://scikit-learn.org/): Machine Learning in Python\n",
    "\n",
    "**Chemistry specific:**\n",
    "- [RDkit](https://www.rdkit.org/): An Open Source library for cheminformatics and machine learning\n",
    "- [MolVS](https://github.com/mcs07/MolVS): Molecular Validation and Standardization\n",
    "\n",
    "**Cosmetic**\n",
    "- [TQDM](https://github.com/tqdm/): For progress bars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "#%matplotlib notebook\n",
    "\n",
    "#Basic stuff\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import warnings\n",
    "\n",
    "# ML models\n",
    "from sklearn.model_selection import train_test_split, cross_val_score, cross_val_predict\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import confusion_matrix, precision_score, recall_score, f1_score\n",
    "from sklearn.metrics import precision_recall_curve, roc_curve, roc_auc_score\n",
    "from sklearn.metrics import make_scorer\n",
    "\n",
    "\n",
    "# Chemistry\n",
    "#import molvs\n",
    "#from rdkit import Chem\n",
    "#from rdkit.Chem import SaltRemover\n",
    "#from rdkit.Chem import Draw\n",
    "#from rdkit.Chem import PandasTools\n",
    "#from rdkit.Chem.rdmolops import RDKFingerprint\n",
    "#from rdkit.Chem.AllChem import GetMorganFingerprintAsBitVect\n",
    "\n",
    "# Initialize tqdm progress bar\n",
    "from tqdm import tqdm, tnrange, tqdm_notebook\n",
    "from tqdm.auto import tqdm\n",
    "tqdm.pandas(tqdm_notebook)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The averaged random forest model\n",
    "%reload_ext autoreload\n",
    "%autoreload 2\n",
    "from AverageRF import AverageRF\n",
    "from utils import train_test_split_data, evaluate_model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "code_folding": [],
    "hidden": true
   },
   "outputs": [],
   "source": [
    "def compare_data_sizes(model,X,y,n_actives,n_models=50,step=10):\n",
    "    \n",
    "    n_actives = 829\n",
    "    n_models = 50\n",
    "\n",
    "    rt_inac, fpr, tpr, auc, thr, f1sc, prec, rec = [],[],[],[],[],[],[],[]\n",
    "\n",
    "    for factor_inactives in tnrange(1,n_models+1,step, desc='Training models:'):\n",
    "        n_inactives = factor_inactives * n_actives\n",
    "        data_size = n_actives + n_inactives\n",
    "        \n",
    "        rt_inac.append(n_inactives/data_size)\n",
    "        \n",
    "        X_data = X[:data_size]\n",
    "        y_data = y[:data_size]\n",
    "\n",
    "        # Randomize data\n",
    "        shuffle_index = np.random.permutation(data_size)\n",
    "        X_data, y_data = X_data[shuffle_index], y_data[shuffle_index]\n",
    "\n",
    "        # split data into train/test sets\n",
    "        X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size=0.10, stratify=y_data)\n",
    "        \n",
    "        # Train models\n",
    "        model.fit(X_train,y_train)\n",
    "        \n",
    "        # Gather model predictions for test set\n",
    "        y_proba = model.predict_proba(X_test)[:,1]\n",
    "        y_pred = model.predict(X_test)\n",
    "        \n",
    "        # Calculate perfomance measures\n",
    "        # ROC\n",
    "        false_pr, true_pr, thresh_roc = roc_curve(y_test, y_proba)\n",
    "        fpr.append(false_pr)\n",
    "        tpr.append(true_pr)\n",
    "        thr.append(thresh_roc)\n",
    "        auc.append(roc_auc_score(y_test,y_proba))\n",
    "        \n",
    "        # Precision x Recall\n",
    "        f1sc.append(f1_score(y_test, y_pred))\n",
    "        prec.append(precision_score(y_test, y_pred))\n",
    "        rec.append(recall_score(y_test, y_pred))\n",
    "        \n",
    "        \n",
    "    return {\"Rate_inactives\":rt_inac,\n",
    "            \"fpr\":fpr, \"tpr\":tpr, \n",
    "            \"ROC_AUC\":auc, \"thr\":thr,\n",
    "            \"F1\":f1sc, \"Precision\":prec, \"Recall\":rec}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Prepare Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "active = pd.read_pickle(\"./active.pkl.bz2\")\n",
    "inactive = pd.read_pickle(\"./inactive.pkl.bz2\")\n",
    "inconclusive = pd.read_pickle(\"./inconclusive.pkl.bz2\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "active       =  (829, 10)\n",
      "inactive     =  (50000, 7)\n",
      "Inconclusive =  (1526, 6)\n"
     ]
    }
   ],
   "source": [
    "# Results *should* be:\n",
    "# active       =  (829, 10)\n",
    "# inactive     =  (50000, 7)\n",
    "# Inconclusive =  (1526, 6)\n",
    "\n",
    "print(\"active       = \", active.shape)\n",
    "print(\"inactive     = \", inactive.shape)\n",
    "print(\"Inconclusive = \", inconclusive.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Add Activity Column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add active bit for classification tasks\n",
    "active['ActiveBit']   = True\n",
    "inactive['ActiveBit'] = False"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Prepare a DataFrame with only the data we need\n",
    "columns = ['PUBCHEM_SID', 'RDKit Fingerprints', 'Morgan Fingerprints','ActiveBit']\n",
    "data = pd.concat([active.filter(columns,axis=1),inactive],\n",
    "                 axis=0,join='inner', ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['PUBCHEM_SID', 'RDKit Fingerprints', 'Morgan Fingerprints',\n",
       "       'ActiveBit'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test Data Size"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To test the data size, we will use the same parameters for both Fingerprinting methods"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_morgan = np.array(list(data['Morgan Fingerprints'])).astype(int)\n",
    "X_rdkit  = np.array(list(data['RDKit Fingerprints'])).astype(int)\n",
    "y = data['ActiveBit'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "testRF = RandomForestClassifier(n_estimators=250)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Morgan Fingerprints"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e742b8d86bf64e30b1953eabe45db754",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training models:', max=5, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "step=10\n",
    "morgan_data_sizes = compare_data_sizes(testRF,X_morgan,y,n_actives=829,n_models=50,step=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = pd.DataFrame(morgan_data_sizes)\n",
    "vals = _['Rate_inactives']\n",
    "ax = _.plot.bar(x='Rate_inactives',y=['ROC_AUC','F1','Precision','Recall'])\n",
    "leg = ax.legend(loc='lower left')\n",
    "#lab = ax.set_xticklabels(['{:,.2%}'.format(x) for x in vals])\n",
    "lab = ax.set_xticklabels(['{:.2%}'.format(x) for x in vals])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### RDKit Fingerprints"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fc30d1f98dc24be9a81931edff39152b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Training models:', max=5, style=ProgressStyle(description_wid…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "step=10\n",
    "rdkit_data_sizes = compare_data_sizes(testRF,X_rdkit,y,n_actives=829,n_models=50,step=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "_ = pd.DataFrame(rdkit_data_sizes)\n",
    "vals = _['Rate_inactives']\n",
    "ax = _.plot.bar(x='Rate_inactives',y=['ROC_AUC','F1','Precision','Recall'])\n",
    "leg = ax.legend(loc='lower left')\n",
    "#lab = ax.set_xticklabels(['{:,.2%}'.format(x) for x in vals])\n",
    "lab = ax.set_xticklabels(['{:.2%}'.format(x) for x in vals])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDAAAALICAYAAACJhQBYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3XmYHWWZN+DfAwHCGlb9lAhBwQFkEDDgriCoIArugBuufI67o46Z+RQZV9yXERdEB3fADZkBRXTUUUeUoIgsMgaN0IKCiIgKSOT9/jgVaJrupNPpTld37vu6ztXnvPVW1XtOpauf/E4t1VoLAAAAQJ+tM90DAAAAAFgZAQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3hNgAAAAAL0nwAAmXVVdWFX7rqTPdlX1p6padw0NCwBYC1TVvlU1NOz10qo6YDrHBEwOAQasZbo/4jd04cFvq+rfq2qTyVxHa+1erbVvraTPZa21TVprf5vMdQMA/TKi9vhNVZ042bUHsHYQYMDa6TGttU2S7JVk7ySvGT6xBuwfAIDJsrz22CPJnkn+eZrHA8xA/oMCa7HW2q+TfCXJblX1rap6U1V9L8lfkty9quZV1Uer6sqq+nVVvXH4KR9V9byquriqrq+qi6pqr6791kM1q2qfqlpcVX/sjvh4V9e+oKpaVc3pXt+1qk6rqt9X1ZKqet6w9RxTVadU1Se6dV1YVQvX3CcFAEyG1tpvkpyZQZCRqtqgqt5RVZd1dcKHqmrD5f2r6tCqOq+rIy6tqgO79mcNq0F+UVX/d3reEbAmCTBgLVZVd0vyqCQ/7pqenuSoJJsm+VWSjydZlmTHDL4teUSS53bzPinJMUmekWSzJIckuWaU1bw3yXtba5sluUeSU8YYzmeTDCW5a5InJnlzVe0/bPohSU5KsnmS05K8f1XfLwAwvapqfpKDkizpmt6a5J4ZBBo7Jtk2ydFd332SfCLJqzL4+/+QJEu7+a5K8ugMapBnJXn38i9SgNlLgAFrp1Or6g9Jvpvk20ne3LWf2Fq7sLW2LMmWGRQYL2ut/bm1dlWSdyc5vOv73CRva62d0waWtNZ+Ncq6bk6yY1Vt3Vr7U2vt7JEduiDlQUle3Vq7sbV2XpITMghUlvtua+2M7poZn0xy79X9EACANebUqro+yeUZhA+vq6pK8rwkL2+t/b61dn0GNcnyWuM5ST7WWjurtXZLa+3XrbWfJUlr7fTW2qVdDfLtJF9L8uA1/q6ANUqAAWunx7bWNm+tbd9ae0Fr7Yau/fJhfbZPsl6SK6vqD13g8eEkd+qm3y3JpeNY13My+GblZ1V1TlU9epQ+d02yvHBZ7lcZfAuz3G+GPf9LkrnLTz8BAHrvsa21TZPsm2TnJFsn2SbJRknOHVZrfLVrT1ZQa1TVQVV1dnfq6R8yOKJ06yl+D8A0U/wDw7Vhzy9PclOSrbsjMka6PINTQla8wNZ+nuSI7qKgj0/y+araakS3K5JsWVWbDgsxtkvy61V9AwBAf7XWvl1VJyZ5RwZ1wQ1J7tVdl2ukUWuNqtogyRcyOI31y621m6vq1CQ1ZQMHesERGMCoWmtXZnA45jurarOqWqeq7lFVD+26nJDklVV1n+6uJTtW1fYjl1NVT6uqbVprtyT5Q9d8u1unttYuT/I/Sd5SVXOravcMjtz49FS9PwBg2rwnycOT7J7kIxlcv+JOSVJV21bVI7t+H03yrKrav6tDtq2qnZOsn2SDJFcnWVZVB2VwnS5glhNgACvyjAyKhIuSXJvk80nukiSttc8leVOSzyS5PsmpGVw3Y6QDk1xYVX/K4IKeh7fWbhyl3xFJFmRwNMaXkryutXbWZL4ZAGD6tdauzuDinK9N8uoMLuh5dlX9McnXk/xd1++H6S7QmeS6DK7btX13tOZLMrgw+LVJnpLBBb6BWa5aayvvBQAAADCNHIEBAAAA9N5KA4yq+lhVXVVVF4wxvarqfVW1pKrOd/9lAGBF1BYAwESM5wiMEzM4h30sByXZqXscleSDqz8sAGAWOzFqCwBgFa00wGit/XeS36+gy6FJPtEGzk6yeVXdZbIGCADMLmoLAGAi5kzCMrbN4B7Nyw11bVeO7FhVR2XwTUo23njj++y8886TsHoAYE0799xzf9da22aKFq+2AIC1yHjriskIMGqUtlFvbdJaOz7J8UmycOHCtnjx4klYPQCwplXVr6Zy8aO0qS0AYJYab10xGXchGUpyt2Gv5ye5YhKWCwCsndQWAMAdTEaAcVqSZ3RXDL9fkutaa3c4xBMAYJzUFgDAHaz0FJKq+mySfZNsXVVDSV6XZL0kaa19KMkZSR6VZEmSvyR51lQNFgCY+dQWAMBErDTAaK0dsZLpLckLJ21EAMCsprYAACZiMk4hAQAAAJhSAgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpvznQPAIApdsy8NbCO66Z+HQAArNUcgQEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB74wowqurAqrqkqpZU1aJRpm9XVd+sqh9X1flV9ajJHyoAMFuoLQCAVTVnZR2qat0kxyV5eJKhJOdU1WmttYuGdXtNklNaax+sql2TnJFkwRSMF8Z2zLw1sI7rpn4dALOc2gIAmIjxHIGxT5IlrbVftNb+muSkJIeO6NOSbNY9n5fkiskbIgAwy6gtAIBVttIjMJJsm+TyYa+Hktx3RJ9jknytql6cZOMkB0zK6ACA2UhtAQCssvEEGDVKWxvx+ogkJ7bW3llV90/yyararbV2y+0WVHVUkqOSZLvttpvIePvHaQsAsKrUFgDAKhtPgDGU5G7DXs/PHQ/jfE6SA5Oktfb9qpqbZOskVw3v1Fo7PsnxSbJw4cKRhQrAWmfBotOnfB1L5075KmBVqS0AgFU2ngDjnCQ7VdUOSX6d5PAkTxnR57Ik+yc5sap2STI3ydWTOVBg8qyR/zQfe/CUrwOYsdQWAMAqW+lFPFtry5K8KMmZSS7O4IrgF1bV66vqkK7bK5I8r6p+kuSzSZ7ZWvMtCABwB2oLAGAixnMERlprZ2Rw+7LhbUcPe35RkgdO7tAAgNlKbQEArKrx3EYVAAAAYFoJMAAAAIDeG9cpJLC63GkBAACA1eEIDAAAAKD3BBgAAABA7wkwAAAAgN4TYAAAAAC9J8AAAAAAes9dSICpccy8NbCO66Z+HTDZ/G4AAEyIIzAAAACA3hNgAAAAAL0nwAAAAAB6T4ABAAAA9J4AAwAAAOg9dyEBgM6CRadP+TqWzp3yVTCFpvrfyNJjD57S5QPATOYIDAAAAKD3BBgAAABA7wkwAAAAgN4TYAAAAAC9J8AAAAAAek+AAQAAAPSeAAMAAADoPQEGAAAA0HsCDAAAAKD3BBgAAABA7wkwAAAAgN6bM90DmEoLFp0+5etYOnfKVwEAAABrPUdgAAAAAL0nwAAAAAB6T4ABAAAA9J4AAwAAAOg9AQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3psz3QMAAKBzzLw1sI7rpn4dADAFHIEBAAAA9J4AAwAAAOg9AQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3hNgAAAAAL0nwAAAAAB6T4ABAAAA9J4AAwAAAOg9AQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3hNgAAAAAL03rgCjqg6sqkuqaklVLRqjz5Or6qKqurCqPjO5wwQAZgt1BQAwEXNW1qGq1k1yXJKHJxlKck5VndZau2hYn52S/HOSB7bWrq2qO03VgAGAmUtdAQBM1HiOwNgnyZLW2i9aa39NclKSQ0f0eV6S41pr1yZJa+2qyR0mADBLqCsAgAkZT4CxbZLLh70e6tqGu2eSe1bV96rq7Ko6cLQFVdVRVbW4qhZfffXVExsxADCTTVpdkagtAGBtMp4Ao0ZpayNez0myU5J9kxyR5ISq2vwOM7V2fGttYWtt4TbbbLOqYwUAZr5JqysStQUArE3GE2AMJbnbsNfzk1wxSp8vt9Zubq39MsklGRQeAADDqSsAgAkZT4BxTpKdqmqHqlo/yeFJThvR59Qk+yVJVW2dwaGfv5jMgQIAs4K6AgCYkJUGGK21ZUlelOTMJBcnOaW1dmFVvb6qDum6nZnkmqq6KMk3k7yqtXbNVA0aAJiZ1BUAwESt9DaqSdJaOyPJGSPajh72vCX5x+4BADAmdQUAMBHjOYUEAAAAYFoJMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpvznQPAAAAVtWCRadP+TqWHnvwlK9jtpjq7WFbAIkjMAAAAIAZQIABAAAA9J4AAwAAAOg9AQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3hNgAAAAAL0nwAAAAAB6T4ABAAAA9N6c6R4AAAAAk2PBotOnfB1Ljz14ytcBo3EEBgAAANB7AgwAAACg98Z1CklVHZjkvUnWTXJCa+3YMfo9McnnkuzdWls8aaMEAGYVtQUzwjHz1sA6rpv6dcwGtgWQcRyBUVXrJjkuyUFJdk1yRFXtOkq/TZO8JMkPJnuQAMDsobYAACZiPKeQ7JNkSWvtF621vyY5Kcmho/R7Q5K3JblxEscHAMw+agsAYJWNJ8DYNsnlw14PdW23qqo9k9yttfafK1pQVR1VVYuravHVV1+9yoMFAGYFtQUAsMrGE2DUKG3t1olV6yR5d5JXrGxBrbXjW2sLW2sLt9lmm/GPEgCYTdQWAMAqG0+AMZTkbsNez09yxbDXmybZLcm3qmppkvslOa2qFk7WIAGAWUVtAQCssvEEGOck2amqdqiq9ZMcnuS05RNba9e11rZurS1orS1IcnaSQ1wpHAAYg9oCAFhlKw0wWmvLkrwoyZlJLk5ySmvtwqp6fVUdMtUDBABmF7UFADARc8bTqbV2RpIzRrQdPUbffVd/WADAbKa2AABW1XhOIQEAAACYVgIMAAAAoPfGdQoJAAAA0DPHzFsD67hu6tcxTo7AAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpvznQPAAAAgBnkmHlrYB3XTf06mHEcgQEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO/Nme4BAAAAwGyzYNHpU76OpXOnfBW94ggMAAAAoPcEGAAAAEDvCTAAAACA3hNgAAAAAL0nwAAAAAB6T4ABAAAA9J4AAwAAAOg9AQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3hNgAAAAAL0nwAAAAAB6T4ABAAAA9J4AAwAAAOi9cQUYVXVgVV1SVUuqatEo0/+xqi6qqvOr6htVtf3kDxUAmA3UFQDARKw0wKiqdZMcl+SgJLsmOaKqdh3R7cdJFrbWdk/y+SRvm+yBAgAzn7oCAJio8RyBsU+SJa21X7TW/prkpCSHDu/QWvtma+0v3cuzk8yf3GECALOEugIAmJDxBBjbJrl82Ouhrm0sz0nyldEmVNVRVbW4qhZfffXV4x8lADBbTFpdkagtAGBtMp4Ao0Zpa6N2rHpakoVJ3j7a9Nba8a21ha21hdtss834RwkAzBaTVlckagsAWJvMGUefoSR3G/Z6fpIrRnaqqgOS/L8kD22t3TQ5wwMAZhl1BQAwIeM5AuOcJDtV1Q5VtX6Sw5OcNrxDVe2Z5MNJDmmtXTX5wwQAZgl1BQAwISsNMFpry5K8KMmZSS5Ockpr7cKqen1VHdJ1e3uSTZJ8rqrOq6rTxlgcALAWU1cAABM1nlNI0lo7I8kZI9qOHvb8gEkeFwAwS6krAICJGM8pJAAAAADTSoABAAAA9J4AAwAAAOg9AQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3hNgAAAAAL0nwAAAAAB6T4ABAAAA9J4AAwAAAOg9AQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3psz3QMY7uabb87Q0FBuvPHGSVneRw65y6QsZ0UurlOmfB25+OJbn86dOzfz58/PeuutN/XrBYAZbqbVFuoKABhbrwKMoaGhbLrpplmwYEGqarWXd/PQHyZhVCu2yzqrP86VuusuSZLWWq655poMDQ1lhx12mPr1AsAMN9NqC3UFAIytV6eQ3Hjjjdlqq60mpcCYjaoqW2211aR9iwQAs53aYmzqCgBmml4FGEkUGCvh8wGAVeNv59h8NgDMJL0LMAAAAABG6tU1MEZasOj0SV3eaS964Er73PtuW+Tgxz85b37vh5Mky5YtywH32Tm77XmfvP/Ekyd1PADAmqW2AICZyxEYI2y40cZZcsnFufGGG5IkZ3/nm7nT/1m1K44vW7ZsKoYGAMxAagsAmBy9PgJjujxo3wPynf/6Wh5+8KH5ype/kAMPfUJ+9MPvJ0muu/bavO6VL8rQZUszd8ON8qm3vTK773rPHPPOD+WK316dpZdfma233DwnvOO1eebLjsnPLv1ldtlxhywdujLHvWlRFt571/zDojfnnJ9cmBtuvClPPHj//Osr/yFJsuC+B+fIJz06/3HWd3LzsmX53Iffmp13dFVwAJjpxltbbLnhOjn+ba+Z2trirntO50cBABPmCIxRHHjo4/PV076Ym268MT+/+ML8/Z4Lb532gXe9JTvvtns+f9b38uJXvzbPeOnRt0479/yL8+WPvSufOe7N+cDHP5ct5m2a879+Sl77sufl3PNvu+f6m179wiz+yqdz/tdPzrfP/lHOv+h/b5229ZZb5Ednfib/8PQn5h0f+uSaecMAwJQab23x5kUvUlsAwBgEGKO45y675YrLL8tXvvyFPGi/h99u2o/POTuPfvxhSZL7PvAhueba63LdH69PkhzyiIdmww3nJkm++8Pzcvihj0yS7Lbzjtl9l51uXcYp/3FW9nrkU7LnI4/IhZdcmot+/stbpz3+oIclSe6z+y5ZevkVU/cmAYA1Zry1xcMetI/aAgDGIMAYw0MfcVDe9cbX5qBDn3D7Ca3doe/yW5BtvNGGw7rdsV+S/PKyX+cdH/5EvnHyh3L+10/Jwfs/ODfeeNOt0zfYYL0kybrrrptlf/vb6r4NAKAn1BYAsHoEGGN43GFPzf992T9lp13udbv2ve77gJz+pc8lSc75/nez9ZabZ7NNN7nD/A/aZ4+c8h9nJUku+t9f5Kc/W5Ik+eP1f87GG26YeZttkt9efU2+8s3vTfE7AQD6YDy1xbf+Z7HaAgDG0OuLeC499uDVmv/8oT9MeN4732XbPPU5z79D+z+8fFGOfsUL88SHP3BwEc/3/Ouo87/gyCfnyJcend0PeHL2vNfO2X2XHTNv002y0923y5677Zx77ffE3H27+Xng3vee8BgBgFXT99piyw3XycfVFgAwql4HGNPh7EuG7tC29/0flL3v/6Akybwttsh7P/aZW6ftvs7gHNNjXnH7gmTuBuvnU//2xsydu0EuXXp59j/s+dl+/uCWaSeOUZgs/cFt96ZfeO9d863Pf2T13gwAMO1WpbZYXlckagsAGEmAMUX+csON2e9JR+Xmm5elpeWDb/mXrL/+etM9LABghlJbALC2E2BMkU032TiLv/Lp6R4GADBLqC0AWNu5iCcAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3uv3RTyPmbdas+8+4vX5z/3VSufZc/utstPOu2bZsmXZdrvt86b3fDibzZuXX19+WR63332zYMed8tebbspGG2+Sw498bnY/7H5JkhNPPi2Lz78o73/Totxyyy151suPybrrrpOPvvN1OfgZL8ln3v/mJMlnvvSVvOCZT16t9wUATFDPa4tXPfPQHPnkxyRRWwDASI7AGGGDuRvmlDO/ky9+4/uZt/kWOfnjt90vff72C3LKV/87p37zB3nbcR/Np074QP795C/fbv7WWp6/6E25edmynPCOo1NVOeOT/5bN522aP/zx+nzgE59b028JAJhGq1JbvPsjn1ZbAMAYBBgrsPtee+eq31w56rT52y/IK45+U9730ZNu1/7So9+ea669Lp947+uzzjqDj3fBfQ/O735/bRa9+X259FdD2ePhh+dVb3j3lI8fAOiXldUW73rdP6otAGAM/T6FZBr97W9/yw+/99953GFPG7PPLrvdOz+7dOmtrz9z6lezy4475FufPz5z5tzxoz32X16SCy65NOedddIdpgEAs9t4aou9/n4XtQUAjMERGCPcdOMNefIjH5yH7n73XPeHa3O/h+w3dufWbvdyr912zq+GrswPz7twikcJAMwUq1JbNLUFAIxJgDHC8vNUv/L983PzX/+akz5+wph9L77w/Oyy4w63vt55xwU55UNvzWHPX5QLL7l0TQwXAOi5VaktfnzBz9QWADAGAcYYNt1sXl79+rfmEx/+t9x88813mP7ryy/Lu9742rz42Yfdrv0Be987Hzr2X3LwM16Sy359+3NcN914o1z/pz9P6bgBgH4aT23xyje8R20BAGPo9zUwjrlutWY/f+gPqzX/Lrvtnnvuulu+etoXstc+D8jQr5bmyQc+5NZbnR3xzKPyrMMecIf5Hv3wh+Tq31+bA5/6onznix+9tX2rLTfPA/feI7s97Ek5aL8H5O2vfflqjQ8AWEU9ry1e8ezD8qzDDr3DfGoLAOh7gDENzr5k6Hav/+3fb7so1g+XjHbV8F8mSZ552CF55mGH3Nr6rMMOvbUAWfqD029t/8xxb57E0QIAfbcqtcXu6/zy1udqCwC4PaeQAAAAAL0nwAAAAAB6r3cBxsjbh3F7Ph8AWDX+do7NZwPATNKrAGPu3Lm55ppr/DEdQ2st11xzTebOnTvdQwGAGUFtMTZ1BQAzTa8u4jl//vwMDQ3l6quvnpTl/fbaGyZlOStycU3OWFfouotvfTp37tzMnz9/6tcJALPATKst1BUAMLZeBRjrrbdedthhh0lb3kGLTl95p9W0dO5Tpnwdq3vLNwBYW8202kJdAQBjG9cpJFV1YFVdUlVLqmrRKNM3qKqTu+k/qKoFkz1QAGD2UFsAAKtqpQFGVa2b5LgkByXZNckRVbXriG7PSXJta23HJO9O8tbJHigAMDuoLQCAiRjPERj7JFnSWvtFa+2vSU5KcuiIPocm+Xj3/PNJ9q+qmrxhAgCziNoCAFhltbKrclfVE5Mc2Fp7bvf66Unu21p70bA+F3R9hrrXl3Z9fjdiWUclOap7+XdJLpmsNzKNtk7yu5X2Yk2wLfrF9ugP26JfZsv22L61ts1EZlRbrNBs+fcxW9ge/WFb9Ivt0R+zZVuMq64Yz0U8R/u2Y2TqMZ4+aa0dn+T4caxzxqiqxa21hdM9DmyLvrE9+sO26BfbI4naYkz+ffSL7dEftkW/2B79sbZti/GcQjKU5G7DXs9PcsVYfapqTpJ5SX4/GQMEAGYdtQUAsMrGE2Cck2SnqtqhqtZPcniS00b0OS3Jkd3zJyb5r7ayc1MAgLWV2gIAWGUrPYWktbasql6U5Mwk6yb5WGvtwqp6fZLFrbXTknw0ySerakkG344cPpWD7plZc9jqLGBb9Ivt0R+2Rb+s9dtDbbFCa/2/j56xPfrDtugX26M/1qptsdKLeAIAAABMt/GcQgIAAAAwrQQYAAAAQO+t1QFGVS2tqp9W1XlVtbhr27Kqzqqqn3c/txhj3iO7Pj+vqiOHtd+nW+aSqnpfVVXX/taqOr+qPjGs79Or6qVT/T5ngqp6aVVdUFUXVtXLurYnda9vqaoxbw1UVZtX1eer6mdVdXFV3b9rP7nbtud12/q8rv2B3bY4p6p2HLaMM5dvr7XZGNvimKr69bDP81ErmH/dqvpxVf3nsLYTq+qXw+bfo2t/Qree71TVVl3bParqpKl+nzPFaNuja39xVV3Stb9tjHlf3k2/oKo+W1Vzu/b9q+pH3bb47rDfgxd3fc/oLqyYqnpQVb1rTbzXvhvjd2PU/cwo8461n3p713Z+VX2pqjbv2u2nZiB1RX+oK/pFbdEvaov+UFtMQGttrX0kWZpk6xFtb0uyqHu+KMlbR5lvyyS/6H5u0T3fopv2wyT3z+D+9V9JclAGt377Tjf900n+PsmGSb6RZL3p/hym+5FktyQXJNkogwvLfj3JTkl2SfJ3Sb6VZOEK5v94kud2z9dPsvkofd6Z5Oju+Re75T88yTuHTX/odH8W0/1YwbY4Jskrx7mMf0zymST/OaztxCRPHKXv/yTZNMnzkry4a/tskp2m+7Pow2MF22MkzaTwAAAgAElEQVS/7vkGXb87jTLvtkl+mWTD7vUpSZ7ZPf/fJLt0z1+Q5MTu+U8yCLbflOQx3X7szOX7t7X5Mda2GNHn1v3MKPOPup9K8ogkc7rnb033N8d+amY+oq7oxWMF+051Rb+2xzFRW/Rpe6gterItRvRRW4x4rNVHYIzh0Az+MaT7+dhR+jwyyVmttd+31q5NclaSA6vqLkk2a619vw3+RXyim/+WJOt3ydaGSW5O8qok72ut3Ty1b2dG2CXJ2a21v7TWliX5dpLHtdYubq1dsqIZq2qzJA/J4Gr1aa39tbX2hxF9KsmTM/jjlQw+/w0z2FncXFX3SLJta+3bk/mmZqhRt8V4Z66q+UkOTnLCOGe5JckGuW1bPDjJla21n6/asGetsbbHPyQ5trV2U5K01q4aY/45STasqjkZfMZXdO0tyWbd83nD2pNkva7vzUmenuSMbj+3tlvh78Yo+5kMmzbmfqq19rVueUlydpL53XP7qdlDXbHmqSv6RW3RL2qL/lBbTMDaHmC0JF+rqnOr6qiu7c6ttSuTpPt5p1Hm2zbJ5cNeD3Vt23bPb9feWrs+yReS/DiD1PK6JHu31r48mW9mBrsgyUOqaquq2ijJo5LcbZzz3j3J1Un+vTu08ISq2nhEnwcn+e2wP1xvyeB2Qy9L8v4MEuHXru6bmCVWtC1e1B129rEa4xDoJO9J8k8ZFA8jvamb/91VtUHX9q8ZpPAHZLBzfk2SN0zWm5kFxtoe90zy4Kr6QVV9u6r2Hjlja+3XSd6R5LIkVya5rrX2tW7yc5OcUVVDGRQSx3bt78jgD902Sb6X5MgkH5iydzezrGw/NXI/M9x49lNJ8uwMvmFP7KdmKnVFP6gr+kVt0S9qi/5QW0zA2h5gPLC1tlcGh2O+sKoeMs75RjtHqK2gPa21t7XW9mitvSKDnejRVfXcqjqlql4zkcHPFq21izM4vOmsJF/N4FCzZSuc6TZzkuyV5IOttT2T/DmDQ3SHOyLDksvW2nmttfu11vbL4Jf/igxCzpOr6lNVdefVekMz2Aq2xQeT3CPJHhn8wXrnyHmr6tFJrmqtnTvKov85yc5J9s7gEOlXd+s7q7V2n9baYzL4VvGMJH/Xnc/3kW5nvtZawfaYk8Fh5vfL4FvXU0aeu9gVgocm2SHJXZNsXFVP6ya/PMmjWmvzk/x7knd16/tka23P1trTMjhc931JDuq2x7uraq39mzGO/dTt9jMjrHQ/VVX/r1vep7v12U/NTOqKHlBX9Ivaol/UFv2htpiYtfYfTJK01q7ofl6V5EtJ9kny2+6QzXQ/Rzt8aii3T8fmZ/APYCi3HaIzvP1WVbVn9/R/kzyjtfbkJLtV1U6r/YZmsNbaR1tre7XWHpLk90nGe5jfUJKh1toPutefz+CXOUnSHd72+CQnj5yx2ykvT+Vf1z0+leQlE30fs8Fo26K19tvW2t9aa7ck+UgGvysjPTDJIVW1NMlJSR5WVZ/qlnllG7gpgz9qt5u/KyaWJ/JvySAtPjfJU6fkTc4gY/xuDCX5YveZ/jCDb6W2HjHrAUl+2Vq7ujuk/ItJHlBV2yS597DfmZOTPGD4jFV119z2be5rkhyW5KYk+0/Nu5wZxtpPrWg/01nZfurIJI9O8tTuNIEMm2Y/NYOoK/pDXdEvaot+UVv0h9pi1a21AUZVbVxVmy5/nsHFTi5IcloGO7t0P0c7HPPMJI+oqi26JPIRSc5sg0NDr6+q+3X/MJ4xyvxvSHJ0BueCrdu13ZLBuUhrraq6U/dzuwx+WcdKG2+ntfabJJdX1d91TfsnuWhYlwOS/Ky1NnSHmQfb9/Q2OAdvowy2g20xyrZYXnx3HpfB78rttNb+ubU2v7W2IMnhSf6rS9szrHivDL4NGTn/PyV5b/fHcMMMvmFc67dFMubvxqlJHta13zODCzf9bsSslyW5X1Vt1H3u+ye5OMm1SeZ18yWDCzldPGLeN+S2Qwptj84K9lMr2s+scD9VVQdm8K3hIa21v4wyu/3UDKGu6Bd1Rb+oLfpFbdEfaosJaD24kuh0PDI4bOYn3ePCJP+va98qg6t4/7z7uWXXvjDJCcPmf3aSJd3jWcPaF2awA700g3OLati0xyZ53bDX70jy0ySfnu7PY7ofSb6TwS/dT5Ls37U9LoN08aYkv82gmEsGh6ydMWzePZIsTnJ+BjvfLYZNOzHJ80dZ30ZJvpnuau0ZnGP20wyS+XtO9+fRw23xye7zOT+DYvwuo22LYcvYN7e/Uvh/dfNfkEHCu8mwaXcd0fdJ3e/k95JsM92fx3Q/xtge63ef4wVJfpTkYaNtjwzOA/5Z1++Tue3K4o/rtsdPMrga/92HzbNnko8Oe/2ybnt8dfn8a+tjtG3Rtd9hPzPe/VQGf0MuT3Je9/jQsHnsp2bQI+qKXj3G2HeqK/q1PdQW/doeaouebIuu/Q77mvHuqzLLa4vqBg4AAADQW2vtKSQAAADAzCHAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYMEtU1d+q6ryquqCq/qOqNl9J/82r6gWrsb7/mei8YyxvQVU9ZdjrhVX1vslcBwAwfmoLoG8EGDB73NBa26O1tluS3yd54Ur6b55kwkVGa+0BE513DAuS3FpktNYWt9ZeMsnrAADGT20B9IoAA2an7yfZNkmqapOq+kZV/aiqflpVh3Z9jk1yj+6blbd3fV9VVedU1flV9a8rWkFV/an7uW9VfauqPl9VP6uqT1dVddOO7pZ3QVUdP6x9x6r6elX9pBvXPbrxPLgbz8u75f5nVa1TVUuHf+tTVUuq6s5VtU1VfaFbxzlV9cBu+kO75ZxXVT+uqk0n9dMFgLWP2kJtAdNOgAGzTFWtm2T/JKd1TTcmeVxrba8k+yV5Z/fHflGSS7tvVl5VVY9IslOSfZLskeQ+VfWQca52zyQvS7JrkrsneWDX/v7W2t7dNzcbJnl01/7pJMe11u6d5AFJruzG851uPO9evuDW2i1Jvpzkcd37u2+Spa213yZ5b5J3t9b2TvKEJCd0s70yyQtba3skeXCSG8b5PgCAEdQWSdQW0AsCDJg9Nqyq85Jck2TLJGd17ZXkzVV1fpKvZ/DtyZ1Hmf8R3ePHSX6UZOcMio7x+GFrbagrCM7L4JDNJNmvqn5QVT9N8rAk9+q+sdi2tfalJGmt3dha+8tKln9yksO654d3r5PkgCTv7973aUk265b/vSTvqqqXJNm8tbZsnO8DALiN2kJtAb0yZ7oHAEyaG1pre1TVvCT/mcF5qu9L8tQk2yS5T2vt5qpammTuKPNXkre01j48gXXfNOz535LMqaq5ST6QZGFr7fKqOqZbb01g+d9PsmNVbZPksUne2LWvk+T+rbWR34IcW1WnJ3lUkrOr6oDW2s8msF4AWJupLW6jtoAecAQGzDKtteuSvCTJK6tqvSTzklzVFRj7Jdm+63p9kuHnb56Z5NlVtUmSVNW2VXWn1RjK8kLmd90yn9iN749Jhqrqsd16NqiqjUYZz/D31JJ8Kcm7klzcWrumm/S1JC9a3q+q9uh+3qO19tPW2luTLM7gGx8AYALUFmoL6AsBBsxCrbUfJ/lJBodEfjrJwqpanME3Jj/r+lyT5HvdRbDe3lr7WpLPJPl+d1jm5zPGH/1xjuEPST6S5KdJTk1yzrDJT0/yku7Q0/9J8n+SnJ9kWXfxrZePssiTkzwttx3imQyKqYXdhcEuSvL8rv1l3fv6SQbnqH5lou8DAFBbRG0BvVCD8BEAAACgvxyBAQAAAPSei3gCY6qqrZJ8Y5RJ+w87VxQAYFzUFsDqcAoJAAAA0HtOIQEAAAB6T4ABAAAA9J4AAwAAAOg9AQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3hNgAAAAAL0nwAAAAAB6T4ABAAAA9J4AA1gtVbW0qm6oqj8Ne9y1qo6vqkuq6paqeuZ0jxMAAJjZBBjAZHhMa22TYY8rkvwkyQuS/GiaxwYATIERX2L8pqpOrKpNhk1/QFX9V1VdX1XXVdV/VNWuI5axWVW9p6ou65azpHu99TjH8K2quraqNhil/bkj2vatqqFhr6uqXlJVF1TVn6tqqKo+V1V/P7FPBJhqAgxgSrTWjmutfSPJjdM9FgBgyjymtbZJkj2S7Jnkn5Okqu6f5GtJvpzkrkl2yODLje9V1d27Pusn+UaSeyU5MMlmSR6Q5Jok+6xsxVW1IMmDk7Qkh0xg7O9N8tIkL0myZZJ7Jjk1ycETWBawBsyZ7gEAAAAzW2vtN1V1ZgZBRpK8LcknWmvvHdbtNVV1nyTHJHlG99guyX6ttT91fa5K8oZxrvYZSc5O8oMkRyb53HjHW1U7JXlhkvu31n44bNKnx7sMYM1zBAYwGU6tqj90j1OnezAAwJpVVfOTHJRkSVVtlMGRFKMFCqckeXj3/IAkXx0WXqyqZ2QQOHw6ySOr6s6rMO/+SYZGhBdAzwkwgMnw2Nba5t3jsdM9GABgjTm1qq5PcnkGR0+8LoPTMdZJcuUo/a9Msvz6FluN0WelqupBSbZPckpr7dwklyZ5yiosYsLrBqaPAAMAAJiox7bWNk2yb5KdMwgnrk1yS5K7jNL/Lkl+1z2/Zow+43Fkkq+11pYv6zNd23LLkqw3Yp71ktw8CesGpokAA5gSVbV+Vc1NUknWq6q5VWWfAwCzUGvt20lOTPKO1tqfk3w/yZNG6frkDC7cmSRfz+DUj41XZV1VtWG3nId2dz/5TZKXJ7l3Vd2763ZZkgUjZt0hya+6599IMr+qFq7KuoHp5T8TwFT5WpIbMjgH9vju+UOmdUQAwFR6T5KHV9UeSRYlObK7TemmVbVFVb0xyf2T/GvX/5MZnHryharauarWqaqtqupfqupRK1jPY5P8LcmuGVw0dI8kuyT5TgbXxUiSk5M8q6r26W6Xes8MQo6TkqS19vMkH0jy2e72qut3X7YcXlWLJu8jASZTtdamewwAAMAMU1VLkzy3tfb1YW0fTHKn1toTuutUvDHJwgxOKflOkle31i4Y1n9eBoHGE5JskeS3Gdx69U2ttWvGWO9Xk1zYWnvFiPYnJ3lfkvmttWVV9ewkr0hytwyuz3FCkre11m7p+lcGt1A9KoOjM65N8t0kr2+tXbg6nw0wNQQYAAAAQO+t9BSSqvpYVV1VVReMMb2q6n1VtaSqzq+qvSZ/mADAbKG2AAAmYjzXwDgxyYErmH5Qkp26x1FJPrj6wwIAZrETo7YAVqCqtquqP43x2G66xwdMjzkr69Ba+++qWrCCLocm+UQbnItydlVtXlV3aa25rzIAcAdqC2BlWmuXJdlkuscB9MtKA4xx2DaDqwcvN9S13aHIqKqjMvgmJRtvvPF9dt5550lYPQCwpp177rm/a61tM0WLV1sAwFpkvHXFZAQYNUrbqFcGba0dn8HtFLNw4cK2ePHiSVg9ALCmVdWvpnLxo7SpLQBglhpvXTGea2CszFAGtyZabn6SKyZhuQDA2kltAQDcwWQEGKcleUZ3xfD7JbnOOaoAwGpQWwAAd7DSU0iq6rNJ9k2ydVUNJXldkvWSpLX2oSRnJHlUkiVJ/pLkWVM1WABg5lNbAAATMZ67kByxkuktyQsnbUQAwKymtgAAJmIyTiEBAAAAmFICDAAAAKD3JuM2qgAAAPTAgkWnT/k6lh578JSvA0bjCAwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHrPXUgAAIDVMtV3vnDXCyARYAAAAH13zLw1sI7rpn4dwGoRYAAAADB+AiWmiQADAAAAJtlUn1qVrH2nVwkwAACYcdbIfwzmPmXK1+FbZoDxE2AAAADATLSWnc7jNqoAAABA7wkwAAAAgN4TYAAAAAC9J8AAAAAAek+AAQAAAPSeAAMAAADoPQEGAAAA0HsCDAAAAKD3BBgAAABA7wkwAAAAgN4TYAAAAAC9J8AAAAAAek+AAQAAAPSeAAMAAADoPQEGAAAA0HsCDAAAAKD3BBgAAABA7wkwAAAAgN4TYAAAAAC9J8AAAAAAek+AAQAAAPSeAAMAAADoPQEGAAAA0HsCDAAAAKD3BBgAAABA7wkwAAAAgN4TYAAAAAC9J8AAAAAAek+AAQAAAPSeAAMAAADoPQEGAAAA0HsCDAAAAKD3BBgAAABA7wkwAAAAgN4TYAAAAAC9J8AAAAAAem9cAUZVHVhVl1TVkqpaNMr07arqm1X146o6v6oeNflDBQBmC7UFALCqVhpgVNW6SY5LclCSXZMcUVW7juj2miSntNb2THJ4kg9M9kABgNlBbQEATMSccfTZJ8mS1tovkqSqTkpyaJKLhvVpSTbrns9LcsVkDhIAmFVmbG2xYNHpU7r8pccePKXLB4CZbDynkGyb5PJhr4e6tuGOSfK0qhpKckaSF4+2oKo6qqoWV9Xiq6++egLDBQBmAbUFALDKxhNg1ChtbcTr/9/evcfZUdb3A/98uSZcDLdoK7EGBQuUImBABUURVBALxSJgvaDW+tOKqFXb9CJSrXfFasUqovUGBcRbKiCi1ZZaUKJCBAIVJIUVqohIbQVM5Pn9cYawWXaTzWaXzG7e79frvPacmWdmnjlPdvabz5mZ85wkH2+tzUvyjCSfqqr7rbu1dlprbUFrbcHcuXPXvrcAwEygtgAA1tp4AoyhJA8b9npe7n8a5x8lOSdJWmuXJJmVZIfJ6CAAMOOoLQCAtTaeAOOyJLtU1U5VtVkGN9JaNKLNjUkOTpKq2i2DIsN5nADAaNQWAMBaW2OA0VpbkeSEJBcmWZrBHcGvqqo3VdURXbPXJvnjqroiyT8leWFrbeSpoAAAagsAYELG8y0kaa2dn8ENtIZPO2nY86uTHDC5XQMAZiq1BQCwtsZzCQkAAADAejWuMzAAAHgAnDznAdjGHVO/DQCYAs7AAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDe22R9d2DaO3nOA7CNO6Z+GwAAANBjzsAAAAAAek+AAQAAAPTejL6EZP7C86Z8G8tmTfkmAAAAYIPnDAwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpvXAFGVR1aVddW1XVVtXCMNsdU1dVVdVVVnTm53QQAZgp1BQAwEZusqUFVbZzk1CRPTTKU5LKqWtRau3pYm12S/EWSA1prt1fVg6eqwwDA9KWuAAAmajxnYOyX5LrW2g9ba79KclaSI0e0+eMkp7bWbk+S1tpPJrebAMAMoa4AACZkPAHGjkluGvZ6qJs23KOSPKqqvllVl1bVoaOtqKpeWlWLq2rxrbfeOrEeAwDT2aTVFYnaAgA2JOMJMGqUaW3E602S7JLkyUmek+T0qtrmfgu1dlprbUFrbcHcuXPXtq8AwPQ3aXVForYAgA3JeAKMoSQPG/Z6XpKbR2nzxdba8tbaDUmuzaDwAAAYTl0BAEzIeAKMy5LsUlU7VdVmSY5LsmhEmy8kOShJqmqHDE79/OFkdhQAmBHUFQDAhKwxwGitrUhyQpILkyxNck5r7aqqelNVHdE1uzDJbVV1dZKvJ3l9a+22qeo0ADA9qSsAgIla49eoJklr7fwk54+YdtKw5y3Jn3YPAIAxqSsAgIkYV4ABzCzzF5435dtY9vbDp3wbAADAhmM898AAAAAAWK+cgcEDwif+AAAArAtnYAAAAAC9J8AAAAAAek+AAQAAAPSeAAMAAADoPQEGAAAA0HsCDAAAAKD3fI0qwHrkK4YBAGB8nIEBAAAA9J4zMJg5Tp7zAGzjjqnfBgAAAPfjDAwAAACg9wQYAAAAQO8JMAAAAIDecw8MYGq4JwkAADCJnIEBAAAA9J4AAwAAAOg9AQYAAADQewIMAAAAoPcEGAAAAEDvCTAAAACA3hNgAAAAAL0nwAAAAAB6b5P13QEA2KCcPOcB2MYdU78NAIAHmAADYKbzH2YAAGYAl5AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeG1eAUVWHVtW1VXVdVS1cTbujq6pV1YLJ6yIAMNOoLQCAtbXGAKOqNk5yapLDkuye5DlVtfso7bZOcmKSb012JwGAmUNtAQBMxHjOwNgvyXWttR+21n6V5KwkR47S7s1J3pnkrknsHwAw86gtAIC1Np4AY8ckNw17PdRNW6mq9k7ysNbalyaxbwDAzKS2AADW2ngCjBplWls5s2qjJO9N8to1rqjqpVW1uKoW33rrrePvJQAwk6gtAIC1Np4AYyjJw4a9npfk5mGvt06yR5JvVNWyJI9Lsmi0m2211k5rrS1orS2YO3fuxHsNAExnagsAYK2NJ8C4LMkuVbVTVW2W5Lgki+6d2Vq7o7W2Q2ttfmttfpJLkxzRWls8JT0GAKY7tQUAsNY2WVOD1tqKqjohyYVJNk7ysdbaVVX1piSLW2uLVr8GAJge5i88b8q3sWzWlG+i99QWAMBErDHASJLW2vlJzh8x7aQx2j553bsFAMxkagsAYG2N5xISAAAAgPVKgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYAAAAQO8JMAAAAIDeE2AAAAAAvSfAAAAAAHpPgAEAAAD0ngADAAAA6L1xBRhVdWhVXVtV11XVwlHm/2lVXV1VS6rqa1X18MnvKgAwE6grAICJWGOAUVUbJzk1yWFJdk/ynKrafUSz7yVZ0FrbM8m5Sd452R0FAKY/dQUAMFHjOQNjvyTXtdZ+2Fr7VZKzkhw5vEFr7euttV92Ly9NMm9yuwkAzBDqCgBgQsYTYOyY5KZhr4e6aWP5oyQXjDajql5aVYuravGtt946/l4CADPFpNUVidoCADYk4wkwapRpbdSGVc9LsiDJu0ab31o7rbW2oLW2YO7cuePvJQAwU0xaXZGoLQBgQ7LJONoMJXnYsNfzktw8slFVHZLkr5I8qbV29+R0DwCYYdQVAMCEjOcMjMuS7FJVO1XVZkmOS7JoeIOq2jvJh5Mc0Vr7yeR3EwCYIdQVAMCErDHAaK2tSHJCkguTLE1yTmvtqqp6U1Ud0TV7V5Ktknymqi6vqkVjrA4A2ICpKwCAiRrPJSRprZ2f5PwR004a9vyQSe4XADBDqSsAgIkYzyUkAAAAAOuVAAMAAADoPQEGAAAA0HsCDAAAAKD3BBgAAABA7wkwAAAAgN4TYAAAAAC9J8AAAAAAek+AAQAAAPSeAAMAAADoPQEGAAAA0HsCDAAAAKD3BBgAAABA7wkwAAAAgN4TYAAAAAC9t8n67sBwy5cvz9DQUO66665JWd9HjvjNSVnP6iytc6Z8G1m6dOXTWbNmZd68edl0002nfrsAMM1Nt9pCXQEAY+tVgDE0NJStt9468+fPT1Wt8/qWD/18Enq1ertttO79XKOH7pYkaa3ltttuy9DQUHbaaaep3y4ATHPTrbZQVwDA2Hp1Ccldd92V7bffflIKjJmoqrL99ttP2qdIADDTqS3Gpq4AYLrpVYCRRIGxBt4fAFg7/naOzXsDwHTSuwADAAAAYKRe3QNjpPkLz5vU9S064YA1tnn0w7bN4c86Jm9934eTJCtWrMghj9k1e+z9mHzg42dPan8AgAeW2gIApi9nYIwwe4stc921S3PXnXcmSS69+Ot58G+s3R3HV6xYMRVdAwCmIbUFAEyOXp+Bsb484cmH5OJ/+UqeeviRueCLn82hR/5BvvvtS5Ikd9x+e974uhMydOOyzJq9RT79ztdlz90flZPf86Hc/ONbs+ymW7LDdtvk9He/IS989cm55vobstvOO2XZ0C059S0Ls+DRu+flC9+ay664KnfedXeOPvzg/M3rXp4kmf/Yw3P8s5+Zf77o4ixfsSKf+fA7suvO7goOANPdeGuL7WZvlNPe+ddTW1s8dO/1+VYAwIQ5A2MUhx75rHx50edy91135QdLr8rv7r1g5bwPnvK27LrHnjn3om/mlX/+hrzgVSetnPedJUvzxY+dkjNPfWs++InPZNs5W2fJV8/JG179x/nOkvu+c/0tf/6KLDf1URYAABWfSURBVL7gjCz56tn510u/myVX/+fKeTtst22+e+GZefnzj867P/SpB2aHAYApNd7a4q0LT1BbAMAYBBijeNRue+Tmm27MBV/8bJ5w0FNXmfe9yy7NM591bJLksQccmNtuvyN3/M8vkiRHPO1JmT17VpLk3799eY478ulJkj123Tl77rbLynWc888XZZ+n/2H2fvpzctW11+fqH9ywct6zDntKkuQxe+6WZTfdPHU7CQA8YMZbWzzlCfupLQBgDAKMMTzpaYfllL99Qw478g9WndHa/dre+xVkW24xe1iz+7dLkhtu/FHe/eFP5mtnfyhLvnpODj/4ibnrrrtXzt98802TJBtvvHFW/PrX67obAEBPqC0AYN0IMMZw1LHPzf979Z9ll91+Z5Xp+zx2/5z3+c8kSS675N+zw3bb5EFbb3W/5Z+w3145558vSpJc/Z8/zPevuS5J8j+/+L9sOXt25jxoq/z41ttywde/OcV7AgD0wXhqi2/8x2K1BQCModc38Vz29sPXafklQz+f8LIP+c0d89w/etn9pr/8NQtz0mtfkaOfesDgJp5/9zejLv8nxx+T4191UvY85Jjs/Tu7Zs/dds6crbfKLo/4rey9x675nYOOziN+a14O2PfRE+4jALB2+l5bbDd7o3xCbQEAo+p1gLE+XHrt0P2m7fv4J2Tfxz8hSTJn223zvo+duXLenhsNrjE9+bWrFiSzNt8sn/77v82sWZvn+mU35eBjX5aHzxt8ZdrHxyhMln3rvu+mX/Do3fONcz+ybjsDAKx3a1Nb3FtXJGoLABhJgDFFfnnnXTno2S/N8uUr0tLyD2/7y2y22abru1sAwDSltgBgQyfAmCJbb7VlFl9wxvruBgAwQ6gtANjQuYknAAAA0HsCDAAAAKD3BBgAAABA7wkwAAAAgN7r9008T56zTovvOeL1kpf81xqX2fvh22eXXXfPihUrsuNvPTxv+bsP50Fz5uRHN92Yow56bObvvEt+dffd2WLLrXLc8S/Jnsc+Lkny8bMXZfGSq/OBtyzMPffckxe95uRsvPFG+eh73pjDX3BizvzAW5MkZ37+gvzJC49Zp/0CACao57XF6194ZI4/5veSqC0AYCRnYIyw+azZOefCi/O5r12SOdtsm7M/cd/3pc97+Pyc8+V/yxe+/q2889SP5tOnfzD/ePYXV1m+tZaXLXxLlq9YkdPffVKqKud/6u+zzZyt8/P/+UU++MnPPNC7BACsR2tTW7z3I2eoLQBgDAKM1dhzn33zk/++ZdR58x4+P6896S15/0fPWmX6q056V267/Y588n1vykYbDd7e+Y89PD/92e1Z+Nb35/r/GspeTz0ur3/ze6e8/wBAv6yptjjljX+qtgCAMfT7EpL16Ne//nW+/c1/y1HHPm/MNrvt8ehcc/2yla/P/MKXs9vOO+Ub556WTTa5/1v79r88MVdee30uv+is+80DAGa28dQW+/zubmoLABiDMzBGuPuuO3PM05+YJ+35iNzx89vzuAMPGrtxa6u83GePXfNfQ7fk25dfNcW9BACmi7WpLZraAgDGJMAY4d7rVC+4ZEmW/+pXOesTp4/ZdulVS7LbzjutfL3rzvNzzofekWNftjBXXXv9A9FdAKDn1qa2+N6V16gtAGAMAowxbP2gOfnzN70jn/zw32f58uX3m/+jm27MKX/7hrzyxceuMn3/fR+dD739L3P4C07MjT9a9RrXrbfcIr/43/+b0n4DAP00ntridW/+O7UFAIyh3/fAOPmOdVp8ydDP12n53fbYM4/afY98edFns89++2fov5blmEMPXPlVZ8954UvzomP3v99yz3zqgbn1Z7fn0OeekIs/99GV07ffbpscsO9e2eMpz85hB+2fd73hNevUPwBgLfW8tnjti4/Ni4498n7LqS0AoO8Bxnpw6bVDq7z++3+876ZY375utLuG35AkeeGxR+SFxx6xcuqLjj1yZQGy7FvnrZx+5qlvncTeAgB9tza1xZ4b3bDyudoCAFblEhIAAACg9wQYAAAAQO/1LsAY+fVhrMr7AwBrx9/OsXlvAJhOehVgzJo1K7fddps/pmNoreW2227LrFmz1ndXAGBaUFuMTV0BwHTTq5t4zps3L0NDQ7n11lsnZX0/vv3OSVnP6iytyenrat2xdOXTWbNmZd68eVO/TQCYAaZbbaGuAICx9SrA2HTTTbPTTjtN2voOW3jemhuto2Wz/nDKt7GuX/kGABuq6VZbqCsAYGzjuoSkqg6tqmur6rqqWjjK/M2r6uxu/reqav5kdxQAmDnUFgDA2lpjgFFVGyc5NclhSXZP8pyq2n1Esz9Kcntrbeck703yjsnuKAAwM6gtAICJGM8ZGPslua619sPW2q+SnJXkyBFtjkzyie75uUkOrqqavG4CADOI2gIAWGu1prtyV9XRSQ5trb2ke/38JI9trZ0wrM2VXZuh7vX1XZufjljXS5O8tHv520munawdWY92SPLTNbbigWAs+sV49Iex6JeZMh4Pb63NnciCaovVmin/PmYK49EfxqJfjEd/zJSxGFddMZ6beI72acfI1GM8bdJaOy3JaePY5rRRVYtbawvWdz8wFn1jPPrDWPSL8UiithiTfx/9Yjz6w1j0i/Hojw1tLMZzCclQkocNez0vyc1jtamqTZLMSfKzyeggADDjqC0AgLU2ngDjsiS7VNVOVbVZkuOSLBrRZlGS47vnRyf5l7ama1MAgA2V2gIAWGtrvISktbaiqk5IcmGSjZN8rLV2VVW9Kcni1tqiJB9N8qmqui6DT0eOm8pO98yMOW11BjAW/WI8+sNY9MsGPx5qi9Xa4P999Izx6A9j0S/Goz82qLFY4008AQAAANa38VxCAgAAALBeCTAAAACA3tugA4yqWlZV36+qy6tqcTdtu6q6qKp+0P3cdoxlj+/a/KCqjh82/THdOq+rqvdXVXXT31FVS6rqk8PaPr+qXjXV+zkdVNWrqurKqrqqql7dTXt29/qeqhrzq4GqapuqOreqrqmqpVX1+G762d3YXt6N9eXd9AO6sbisqnYeto4L7x2vDdkYY3FyVf1o2Pv5jNUsv3FVfa+qvjRs2ser6oZhy+/VTf+DbjsXV9X23bRHVtVZU72f08Vo49FNf2VVXdtNf+cYy76mm39lVf1TVc3qph9cVd/txuLfh/0evLJre353Y8VU1ROq6pQHYl/7bozfjVGPM6MsO9Zx6l3dtCVV9fmq2qab7jg1Dakr+kNd0S9qi35RW/SH2mICWmsb7CPJsiQ7jJj2ziQLu+cLk7xjlOW2S/LD7ue23fNtu3nfTvL4DL6//oIkh2Xw1W8Xd/PPSPK7SWYn+VqSTdf3+7C+H0n2SHJlki0yuLHsV5PskmS3JL+d5BtJFqxm+U8keUn3fLMk24zS5j1JTuqef65b/1OTvGfY/Cet7/difT9WMxYnJ3ndONfxp0nOTPKlYdM+nuToUdr+R5Ktk/xxkld20/4pyS7r+73ow2M143FQ93zzrt2DR1l2xyQ3JJndvT4nyQu75/+ZZLfu+Z8k+Xj3/IoMgu23JPm97jh24b3Htw35MdZYjGiz8jgzyvKjHqeSPC3JJt3zd6T7m+M4NT0fUVf04rGaY6e6ol/jcXLUFn0aD7VFT8ZiRBu1xYjHBn0GxhiOzOAfQ7qfvz9Km6cnuai19rPW2u1JLkpyaFX9ZpIHtdYuaYN/EZ/slr8nyWZdsjU7yfIkr0/y/tba8qndnWlhtySXttZ+2VpbkeRfkxzVWlvaWrt2dQtW1YOSHJjB3erTWvtVa+3nI9pUkmMy+OOVDN7/2RkcLJZX1SOT7Nha+9fJ3KlpatSxGO/CVTUvyeFJTh/nIvck2Tz3jcUTk9zSWvvB2nV7xhprPF6e5O2ttbuTpLX2kzGW3yTJ7KraJIP3+OZuekvyoO75nGHTk2TTru3yJM9Pcn53nNvQrfZ3Y5TjTIbNG/M41Vr7Sre+JLk0ybzuuePUzKGueOCpK/pFbdEvaov+UFtMwIYeYLQkX6mq71TVS7tpD2mt3ZIk3c8Hj7LcjkluGvZ6qJu2Y/d8lemttV8k+WyS72WQWt6RZN/W2hcnc2emsSuTHFhV21fVFkmekeRh41z2EUluTfKP3amFp1fVliPaPDHJj4f94XpbBl839OokH8ggEX7Duu7EDLG6sTihO+3sYzXGKdBJ/i7Jn2VQPIz0lm7591bV5t20v8kghT8kg4PzXyd582TtzAww1ng8KskTq+pbVfWvVbXvyAVbaz9K8u4kNya5JckdrbWvdLNfkuT8qhrKoJB4ezf93Rn8oZub5JtJjk/ywSnbu+llTcepkceZ4cZznEqSF2fwCXviODVdqSv6QV3RL2qLflFb9IfaYgI29ADjgNbaPhmcjvmKqjpwnMuNdo1QW830tNbe2Vrbq7X22gwOoidV1Uuq6pyq+uuJdH6maK0tzeD0pouSfDmDU81WrHah+2ySZJ8k/9Ba2zvJ/2Vwiu5wz8mw5LK1dnlr7XGttYMy+OW/OYOQ8+yq+nRVPWSddmgaW81Y/EOSRybZK4M/WO8ZuWxVPTPJT1pr3xll1X+RZNck+2ZwivSfd9u7qLX2mNba72XwqeL5SX67u57vI93BfIO1mvHYJIPTzB+Xwaeu54y8drErBI9MslOShybZsqqe181+TZJntNbmJfnHJKd02/tUa23v1trzMjhd9/1JDuvG471VtcH+zRjHcWqV48wIazxOVdVfdes7o9ue49T0pK7oAXVFv6gt+kVt0R9qi4nZYP/BJElr7ebu50+SfD7Jfkl+3J2yme7naKdPDWXVdGxeBv8AhnLfKTrDp69UVXt3T/8zyQtaa8ck2aOqdlnnHZrGWmsfba3t01o7MMnPkoz3NL+hJEOttW91r8/N4Jc5SdKd3vasJGePXLA7KN+byr+xe3w6yYkT3Y+ZYLSxaK39uLX269baPUk+ksHvykgHJDmiqpYlOSvJU6rq0906b2kDd2fwR22V5bti4t5E/m0ZpMXfSfLcKdnJaWSM342hJJ/r3tNvZ/Cp1A4jFj0kyQ2ttVu7U8o/l2T/qpqb5NHDfmfOTrL/8AWr6qG579Pcv05ybJK7kxw8NXs5PYx1nFrdcaazpuPU8UmemeS53WUCGTbPcWoaUVf0h7qiX9QW/aK26A+1xdrbYAOMqtqyqra+93kGNzu5MsmiDA526X6OdjrmhUmeVlXbdknk05Jc2Aanhv6iqh7X/cN4wSjLvznJSRlcC7ZxN+2eDK5F2mBV1YO7n7+VwS/rWGnjKlpr/53kpqr67W7SwUmuHtbkkCTXtNaG7rfwYHzPa4Nr8LbIYByMxShjcW/x3Tkqg9+VVbTW/qK1Nq+1Nj/JcUn+pUvbM6x4rww+DRm5/J8leV/3x3B2Bp8wbvBjkYz5u/GFJE/ppj8qgxs3/XTEojcmeVxVbdG97wcnWZrk9iRzuuWSwY2clo5Y9s2575RC49FZzXFqdceZ1R6nqurQDD41PKK19stRFnecmibUFf2irugXtUW/qC36Q20xAa0HdxJdH48MTpu5ontcleSvuunbZ3AX7x90P7frpi9Icvqw5V+c5Lru8aJh0xdkcAC9PoNri2rYvN9P8sZhr9+d5PtJzljf78f6fiS5OINfuiuSHNxNOyqDdPHuJD/OoJhLBqesnT9s2b2SLE6yJIOD77bD5n08yctG2d4WSb6e7m7tGVxj9v0MkvlHre/3o4dj8anu/VmSQTH+m6ONxbB1PDmr3in8X7rlr8wg4d1q2LyHjmj77O538ptJ5q7v92N9P8YYj8269/HKJN9N8pTRxiOD64Cv6dp9KvfdWfyobjyuyOBu/I8YtszeST467PWru/H48r3Lb6iP0caim36/48x4j1MZ/A25Kcnl3eNDw5ZxnJpGj6grevUY49iprujXeKgt+jUeaouejEU3/X7HmvEeqzLDa4vqOg4AAADQWxvsJSQAAADA9CHAAAAAAHpPgAEAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9wQYMENU1a+r6vKqurKq/rmqtllD+22q6k/WYXv/MdFlx1jf/Kr6w2GvF1TV+ydzGwDA+KktgL4RYMDMcWdrba/W2h5JfpbkFWtov02SCRcZrbX9J7rsGOYnWVlktNYWt9ZOnORtAADjp7YAekWAATPTJUl2TJKq2qqqvlZV362q71fVkV2btyd5ZPfJyru6tq+vqsuqaklV/c3qNlBV/9v9fHJVfaOqzq2qa6rqjKqqbt5J3fqurKrThk3fuaq+WlVXdP16ZNefJ3b9eU233i9V1UZVtWz4pz5VdV1VPaSq5lbVZ7ttXFZVB3Tzn9St5/Kq+l5VbT2p7y4AbHjUFmoLWO8EGDDDVNXGSQ5OsqibdFeSo1pr+yQ5KMl7uj/2C5Nc332y8vqqelqSXZLsl2SvJI+pqgPHudm9k7w6ye5JHpHkgG76B1pr+3af3MxO8sxu+hlJTm2tPTrJ/klu6fpzcdef99674tbaPUm+mOSobv8em2RZa+3HSd6X5L2ttX2T/EGS07vFXpfkFa21vZI8Mcmd49wPAGAEtUUStQX0ggADZo7ZVXV5ktuSbJfkom56JXlrVS1J8tUMPj15yCjLP617fC/Jd5PsmkHRMR7fbq0NdQXB5RmcspkkB1XVt6rq+0mekuR3uk8sdmytfT5JWmt3tdZ+uYb1n53k2O75cd3rJDkkyQe6/V6U5EHd+r+Z5JSqOjHJNq21FePcDwDgPmoLtQX0yibruwPApLmztbZXVc1J8qUMrlN9f5LnJpmb5DGtteVVtSzJrFGWryRva619eALbvnvY818n2aSqZiX5YJIFrbWbqurkbrs1gfVfkmTnqpqb5PeT/G03faMkj2+tjfwU5O1VdV6SZyS5tKoOaa1dM4HtAsCGTG1xH7UF9IAzMGCGaa3dkeTEJK+rqk2TzEnyk67AOCjJw7umv0gy/PrNC5O8uKq2SpKq2rGqHrwOXbm3kPlpt86ju/79T5Khqvr9bjubV9UWo/Rn+D61JJ9PckqSpa2127pZX0lywr3tqmqv7ucjW2vfb629I8niDD7xAQAmQG2htoC+EGDADNRa+16SKzI4JfKMJAuqanEGn5hc07W5Lck3u5tgvau19pUkZya5pDst89yM8Ud/nH34eZKPJPl+ki8kuWzY7OcnObE79fQ/kvxGkiVJVnQ333rNKKs8O8nzct8pnsmgmFrQ3Rjs6iQv66a/utuvKzK4RvWCie4HAKC2iNoCeqEG4SMAAABAfzkDAwAAAOg9N/EExlRV2yf52iizDh52rSgAwLioLYB14RISAAAAoPdcQgIAAAD0ngADAAAA6D0BBgAAANB7AgwAAACg9/4/KV2UX46eOR4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x720 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(nrows=2, ncols=2, figsize=(15,10))\n",
    "\n",
    "for axis,prop in zip([(0,0),(0,1),(1,0),(1,1)],['Precision','Recall','F1','ROC_AUC']):\n",
    "# Precisions\n",
    "    _ = {'Rate_inactives':morgan_data_sizes['Rate_inactives'],\n",
    "         'Morgan':morgan_data_sizes[prop],\n",
    "         'RDKit':rdkit_data_sizes[prop]}\n",
    "    a=pd.DataFrame(data=_).plot.bar(x='Rate_inactives', y=['Morgan','RDKit'], ax=ax[axis])\n",
    "    lab = a.set_xticklabels(['{:.2%}'.format(x) for x in vals], rotation=0)\n",
    "    a.legend(loc='lower left')\n",
    "    a.set_title(prop)\n",
    "    a.set_ylim([0,1])\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Conclusion**\n",
    "\n",
    "In all cases, the values are mostly converged with about 90% inactives.\n",
    "\n",
    "**From now on, we will use a rate of 90% inactives.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train / Test split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From now on, we want to find the best parameters in each case, and to compare the Morgan and RDKit results using the best fit parameters. So, let's make the train/test split here, and use the same sets for all tests from now on."
   ]
  },
  {
   "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>PUBCHEM_SID</th>\n",
       "      <th>RDKit Fingerprints</th>\n",
       "      <th>Morgan Fingerprints</th>\n",
       "      <th>ActiveBit</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>46501256</td>\n",
       "      <td>[0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 1, 0, ...</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>26671489</td>\n",
       "      <td>[1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 1, 1, 0, 0, 0, ...</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>50086104</td>\n",
       "      <td>[1, 1, 1, 0, 1, 1, 0, 1, 1, 1, 0, 1, 0, 0, 1, ...</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, ...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>99356103</td>\n",
       "      <td>[1, 0, 1, 1, 0, 1, 1, 0, 0, 1, 1, 1, 0, 0, 0, ...</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>92764075</td>\n",
       "      <td>[1, 0, 1, 1, 0, 1, 1, 0, 0, 1, 1, 1, 0, 0, 0, ...</td>\n",
       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PUBCHEM_SID                                 RDKit Fingerprints  \\\n",
       "0     46501256  [0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 1, 0, ...   \n",
       "1     26671489  [1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 1, 1, 0, 0, 0, ...   \n",
       "2     50086104  [1, 1, 1, 0, 1, 1, 0, 1, 1, 1, 0, 1, 0, 0, 1, ...   \n",
       "3     99356103  [1, 0, 1, 1, 0, 1, 1, 0, 0, 1, 1, 1, 0, 0, 0, ...   \n",
       "4     92764075  [1, 0, 1, 1, 0, 1, 1, 0, 0, 1, 1, 1, 0, 0, 0, ...   \n",
       "\n",
       "                                 Morgan Fingerprints  ActiveBit  \n",
       "0  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...       True  \n",
       "1  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...       True  \n",
       "2  [0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, ...       True  \n",
       "3  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...       True  \n",
       "4  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...       True  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train:  7461 \ty_train:  7461 \t( 746 actives and 6715 inactives)\n",
      "X_test:   829  \ty_test:   829  \t( 83 actives and 746 inactives)\n"
     ]
    }
   ],
   "source": [
    "# This split will return the indices for X, and the y values\n",
    "X_data = data.index\n",
    "y_data = data['ActiveBit'].values\n",
    "X_train_index, X_test_index, y_train, y_test = train_test_split_data(X_data, y_data, rate_inactives=0.90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7461, 829)"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(data.iloc[X_train_index]['ActiveBit'] == y_train), sum(data.iloc[X_test_index]['ActiveBit'] == y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Morgan Fingerprints"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Select Morgan Fingerprints\n",
    "X_train_morgan = np.array(list(data.iloc[X_train_index]['Morgan Fingerprints'])).astype(int)\n",
    "X_test_morgan  = np.array(list(data.iloc[X_test_index]['Morgan Fingerprints'])).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((7461, 2048), (7461,))"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_morgan.shape, y_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hyperparameter tuning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Thorough optimization using 5-fold cross validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "# This grid will have 96 different combinations. \n",
    "# Using 5-fold cross validation, it sums to 480 fittings.\n",
    "\n",
    "n_estimators=[150,250,350,450] # Number of estimators\n",
    "criterion=['gini','entropy']\n",
    "max_features=[None,0.33,'sqrt']\n",
    "bootstrap=[True,False]\n",
    "class_weight=['balanced','balanced_subsample']\n",
    "\n",
    "## Create the random grid\n",
    "param_grid = {'n_estimators': n_estimators,\n",
    "              'criterion': criterion,\n",
    "              'max_features': max_features,\n",
    "              'bootstrap':bootstrap,\n",
    "              'class_weight': class_weight}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 96 candidates, totalling 480 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.7min\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:  1.7min remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.7min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 2.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 2.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 2.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 2.8min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 2.7min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 3.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 3.4min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 3.4min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 3.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 3.4min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 4.2min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 4.1min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 4.3min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 4.3min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 4.2min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  27.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  25.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  27.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  27.5s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  26.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  42.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  41.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  44.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  43.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  44.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  58.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  57.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  59.5s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.3min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.3min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.3min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.3min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.5s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.5s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.5s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   7.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   7.5s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   7.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   7.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   8.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   9.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=  11.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   9.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=  11.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   9.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  57.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  58.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  59.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  57.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  55.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.2min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.2min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.3min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.2min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.2min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 2.8min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 2.8min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 2.9min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 2.8min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 2.8min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  20.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  19.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  22.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  20.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  20.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  33.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  33.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  34.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  33.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  33.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  48.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  44.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  48.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  47.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  46.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total=  58.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total=  59.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total=  59.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.5s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   5.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   5.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   5.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   5.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   8.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   7.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   7.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   7.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   7.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 2.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 2.2min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 2.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 2.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 2.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 3.2min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 3.1min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 3.2min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 3.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 3.2min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 4.1min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 4.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 4.1min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 4.2min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 4.1min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  25.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  24.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  27.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  25.9s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  25.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  42.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  41.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  45.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  44.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  42.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  58.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  57.4s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  59.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   7.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   7.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   8.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   8.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   7.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   9.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   9.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   9.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   9.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   9.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total=  58.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total=  57.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total=  55.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total=  56.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.6min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.2min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.2min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.3min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.2min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.2min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 2.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 2.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 2.9min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 2.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 2.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  20.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  19.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  21.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  20.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  20.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  34.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  31.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  35.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  33.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  34.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  46.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  44.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  49.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  46.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  47.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total=  58.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total=  59.9s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.4s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   6.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   5.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   5.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   7.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   7.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   7.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   7.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   7.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=  10.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 2.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 2.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 2.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 2.9min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 2.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 4.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 3.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 4.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 4.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 4.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 6.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 5.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 6.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 6.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 6.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 8.0min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 6.9min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 8.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 8.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 8.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  45.8s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  43.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  48.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  47.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  44.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total= 1.9min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 2.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 2.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 2.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 2.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.4s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   8.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   8.4s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   8.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=  12.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=  12.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=  13.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=  12.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=  12.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=  15.4s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=  14.9s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=  15.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=  14.9s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=  16.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.9min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 2.9min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 3.0min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 3.0min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 2.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 4.0min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 4.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 4.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 3.0min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 4.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 5.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 5.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 5.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 3.9min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 5.6min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  41.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  37.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  42.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  35.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  35.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  59.8s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  58.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  57.9s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  59.9s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   8.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   8.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   9.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   8.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   8.8s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=  11.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=  11.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=  11.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=  11.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=  11.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=  14.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=  14.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=  17.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=  14.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=  14.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 2.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 2.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 2.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 2.9min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 2.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 4.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 3.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 4.5min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 4.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 4.6min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 6.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 5.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 6.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 6.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 6.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 8.0min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 6.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 8.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 8.6min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 8.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  46.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  43.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  49.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  46.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  44.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total= 1.9min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 2.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 2.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 2.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 2.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.2s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   5.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   8.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   9.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   8.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   8.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=  13.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=  12.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=  12.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=  12.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=  12.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=  15.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=  15.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=  16.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=  14.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=  15.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.9min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 2.9min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 3.0min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 2.9min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 2.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 4.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 4.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 4.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 4.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 5.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 5.6min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 5.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 3.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 5.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  35.4s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  34.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  38.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  35.9s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  34.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  56.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  57.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  57.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  56.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total= 1.3min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   5.2s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   8.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   8.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=  12.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=  11.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=  14.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=  12.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=  11.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=  15.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=  16.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=  16.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=  15.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=  17.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done 480 out of 480 | elapsed: 744.0min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise-deprecating',\n",
       "       estimator=RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "            min_samples_leaf=1, min_samples_split=2,\n",
       "            min_weight_fraction_leaf=0.0, n_estimators='warn', n_jobs=4,\n",
       "            oob_score=False, random_state=42, verbose=0, warm_start=False),\n",
       "       fit_params=None, iid='warn', n_jobs=None,\n",
       "       param_grid={'n_estimators': [150, 250, 350, 450], 'criterion': ['gini', 'entropy'], 'max_features': [None, 0.33, 'sqrt'], 'bootstrap': [True, False], 'class_weight': ['balanced', 'balanced_subsample']},\n",
       "       pre_dispatch='2*n_jobs', refit='Precision',\n",
       "       return_train_score='warn',\n",
       "       scoring={'F1 Score': make_scorer(f1_score), 'Precision': make_scorer(precision_score), 'Recall': make_scorer(recall_score), 'AUC': 'roc_auc'},\n",
       "       verbose=2)"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = RandomForestClassifier(random_state=42, n_jobs=4)\n",
    "scoring = {'F1 Score': make_scorer(f1_score),\n",
    "           'Precision':make_scorer(precision_score),\n",
    "           'Recall':make_scorer(recall_score),\n",
    "           'AUC': 'roc_auc'}\n",
    "\n",
    "gs = GridSearchCV(estimator=model, param_grid=param_grid,\n",
    "                  scoring=scoring, refit='Precision',\n",
    "                  cv=5, verbose=2)\n",
    "\n",
    "gs.fit(X_train_morgan, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "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>std_fit_time</th>\n",
       "      <th>mean_score_time</th>\n",
       "      <th>std_score_time</th>\n",
       "      <th>param_bootstrap</th>\n",
       "      <th>param_class_weight</th>\n",
       "      <th>param_criterion</th>\n",
       "      <th>param_max_features</th>\n",
       "      <th>param_n_estimators</th>\n",
       "      <th>params</th>\n",
       "      <th>...</th>\n",
       "      <th>std_test_AUC</th>\n",
       "      <th>rank_test_AUC</th>\n",
       "      <th>split0_train_AUC</th>\n",
       "      <th>split1_train_AUC</th>\n",
       "      <th>split2_train_AUC</th>\n",
       "      <th>split3_train_AUC</th>\n",
       "      <th>split4_train_AUC</th>\n",
       "      <th>mean_train_AUC</th>\n",
       "      <th>std_train_AUC</th>\n",
       "      <th>Fingerprint</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>97.632134</td>\n",
       "      <td>2.680090</td>\n",
       "      <td>0.456473</td>\n",
       "      <td>0.007467</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>None</td>\n",
       "      <td>150</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.010647</td>\n",
       "      <td>75</td>\n",
       "      <td>0.999948</td>\n",
       "      <td>0.999964</td>\n",
       "      <td>0.999964</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999975</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>Morgan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>158.858815</td>\n",
       "      <td>3.343530</td>\n",
       "      <td>0.476449</td>\n",
       "      <td>0.045213</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>None</td>\n",
       "      <td>250</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.010407</td>\n",
       "      <td>70</td>\n",
       "      <td>0.999955</td>\n",
       "      <td>0.999968</td>\n",
       "      <td>0.999964</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999977</td>\n",
       "      <td>0.000019</td>\n",
       "      <td>Morgan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>208.584575</td>\n",
       "      <td>6.962614</td>\n",
       "      <td>0.514671</td>\n",
       "      <td>0.125014</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>None</td>\n",
       "      <td>350</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.010104</td>\n",
       "      <td>67</td>\n",
       "      <td>0.999960</td>\n",
       "      <td>0.999974</td>\n",
       "      <td>0.999972</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999981</td>\n",
       "      <td>0.000016</td>\n",
       "      <td>Morgan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>252.727259</td>\n",
       "      <td>5.606996</td>\n",
       "      <td>0.855076</td>\n",
       "      <td>0.000711</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>None</td>\n",
       "      <td>450</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.010396</td>\n",
       "      <td>68</td>\n",
       "      <td>0.999957</td>\n",
       "      <td>0.999969</td>\n",
       "      <td>0.999966</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999978</td>\n",
       "      <td>0.000018</td>\n",
       "      <td>Morgan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>26.515650</td>\n",
       "      <td>0.932962</td>\n",
       "      <td>0.451406</td>\n",
       "      <td>0.001214</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>0.33</td>\n",
       "      <td>150</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.011324</td>\n",
       "      <td>57</td>\n",
       "      <td>0.999962</td>\n",
       "      <td>0.999967</td>\n",
       "      <td>0.999971</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999980</td>\n",
       "      <td>0.000017</td>\n",
       "      <td>Morgan</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 71 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   mean_fit_time  std_fit_time  mean_score_time  std_score_time  \\\n",
       "0      97.632134      2.680090         0.456473        0.007467   \n",
       "1     158.858815      3.343530         0.476449        0.045213   \n",
       "2     208.584575      6.962614         0.514671        0.125014   \n",
       "3     252.727259      5.606996         0.855076        0.000711   \n",
       "4      26.515650      0.932962         0.451406        0.001214   \n",
       "\n",
       "  param_bootstrap param_class_weight param_criterion param_max_features  \\\n",
       "0            True           balanced            gini               None   \n",
       "1            True           balanced            gini               None   \n",
       "2            True           balanced            gini               None   \n",
       "3            True           balanced            gini               None   \n",
       "4            True           balanced            gini               0.33   \n",
       "\n",
       "  param_n_estimators                                             params  ...  \\\n",
       "0                150  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "1                250  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "2                350  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "3                450  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "4                150  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "\n",
       "   std_test_AUC  rank_test_AUC  split0_train_AUC  split1_train_AUC  \\\n",
       "0      0.010647             75          0.999948          0.999964   \n",
       "1      0.010407             70          0.999955          0.999968   \n",
       "2      0.010104             67          0.999960          0.999974   \n",
       "3      0.010396             68          0.999957          0.999969   \n",
       "4      0.011324             57          0.999962          0.999967   \n",
       "\n",
       "   split2_train_AUC  split3_train_AUC  split4_train_AUC  mean_train_AUC  \\\n",
       "0          0.999964               1.0               1.0        0.999975   \n",
       "1          0.999964               1.0               1.0        0.999977   \n",
       "2          0.999972               1.0               1.0        0.999981   \n",
       "3          0.999966               1.0               1.0        0.999978   \n",
       "4          0.999971               1.0               1.0        0.999980   \n",
       "\n",
       "   std_train_AUC  Fingerprint  \n",
       "0       0.000021       Morgan  \n",
       "1       0.000019       Morgan  \n",
       "2       0.000016       Morgan  \n",
       "3       0.000018       Morgan  \n",
       "4       0.000017       Morgan  \n",
       "\n",
       "[5 rows x 71 columns]"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with warnings.catch_warnings():\n",
    "    warnings.filterwarnings(\"ignore\")\n",
    "    df_morgan_opt = pd.DataFrame(gs.cv_results_)\n",
    "    \n",
    "df_morgan_opt['Fingerprint'] = \"Morgan\"\n",
    "df_morgan_opt.to_csv(path_or_buf='./morgan_opt.csv')\n",
    "df_morgan_opt.to_pickle('./morgan_opt.pkl.bz2')\n",
    "\n",
    "df_morgan_opt.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The best combination of parameters, according to the grid search:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight='balanced_subsample',\n",
       "            criterion='entropy', max_depth=None, max_features='sqrt',\n",
       "            max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "            min_impurity_split=None, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            n_estimators=350, n_jobs=4, oob_score=False, random_state=42,\n",
       "            verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gs.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x241cba5d2e8>"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Stats for the best estimator\n",
    "_ = (df_morgan_opt['param_bootstrap'] == True) & \\\n",
    "    (df_morgan_opt['param_class_weight'] == 'balanced_subsample') & \\\n",
    "    (df_morgan_opt['param_criterion'] == 'entropy') & \\\n",
    "    (df_morgan_opt['param_max_features'] == 'sqrt') & \\\n",
    "    (df_morgan_opt['param_n_estimators'] == 350)\n",
    "df_morgan_opt.loc[_].plot.bar(x='param_max_features', y=['mean_test_Precision', 'mean_test_F1 Score'], figsize=(15,6))\n",
    "#df_morgan_opt.loc[_]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight='balanced_subsample',\n",
       "            criterion='entropy', max_depth=None, max_features='sqrt',\n",
       "            max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "            min_impurity_split=None, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            n_estimators=350, n_jobs=4, oob_score=False, random_state=42,\n",
       "            verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chosen_rf_morgan = gs.best_estimator_\n",
    "chosen_rf_morgan"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RDKit Fingerprints"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Select RDKit Fingerprints\n",
    "X_train_rdkit = np.array(list(data.iloc[X_train_index]['RDKit Fingerprints'])).astype(int)\n",
    "X_test_rdkit  = np.array(list(data.iloc[X_test_index]['RDKit Fingerprints'])).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((7461, 2048), (7461,))"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_rdkit.shape, y_train.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hyperparameter tuning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Thorough optimization using 5-fold cross validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 96 candidates, totalling 480 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total=  40.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:   41.1s remaining:    0.0s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total=  38.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total=  38.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total=  37.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total=  38.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.1min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.1min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  14.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  14.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  14.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  14.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  13.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  23.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  22.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  23.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  23.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  22.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  31.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  32.5s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  32.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  31.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  31.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total=  40.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total=  40.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total=  41.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total=  41.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total=  40.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   3.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   2.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   3.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   2.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   2.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   3.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   4.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  36.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  37.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  36.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  36.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total=  37.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  14.6s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  14.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  13.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  13.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  14.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  23.7s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  23.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  23.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  22.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  23.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  31.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  31.5s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  31.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  30.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  31.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total=  40.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total=  39.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total=  40.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total=  40.8s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total=  40.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   2.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   2.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   3.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   2.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   3.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   4.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   4.4s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   3.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   3.9s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   6.2s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   5.0s\n",
      "[CV] bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total=  38.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total=  38.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total=  37.9s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total=  38.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total=  38.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.1min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.1min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.1min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 1.5min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 2.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 1.9min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  14.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  14.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  14.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  14.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  14.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  22.9s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  23.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  23.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  23.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  25.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  31.9s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  31.9s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  32.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  31.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  32.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total=  41.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total=  41.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total=  41.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total=  41.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total=  41.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   3.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   3.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   2.9s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   3.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   3.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   3.9s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   4.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   3.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.9s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.3s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   5.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total=  38.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total=  36.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total=  37.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total=  36.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total=  36.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.0min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 1.4min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 1.8min\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  13.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  13.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  14.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  13.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  14.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  22.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  23.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  22.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  23.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  22.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  31.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  32.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  31.7s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  31.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  31.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total=  40.9s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total=  40.5s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total=  40.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total=  40.8s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total=  40.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.1s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   3.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   3.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   3.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   3.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   3.1s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   4.0s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   4.2s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   5.6s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   5.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   5.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   5.4s\n",
      "[CV] bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=True, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   5.4s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=250, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 2.6min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 3.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 3.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 3.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 3.3min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=None, n_estimators=450, total= 3.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  24.9s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  24.4s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  23.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  24.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=150, total=  24.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  38.9s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  39.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  39.4s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  38.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=250, total=  38.8s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  54.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  54.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  54.9s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  54.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=350, total=  54.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   4.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   5.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   4.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   4.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=250, total=   4.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   5.9s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   6.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   6.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   6.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=350, total=   6.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   8.4s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   8.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   8.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=gini, max_features=sqrt, n_estimators=450, total=   8.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.9min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=250, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=350, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 3.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  23.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  23.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  23.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  24.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=150, total=  24.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  38.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  38.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  38.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  38.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=250, total=  38.8s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  53.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  54.4s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  53.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  53.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=350, total=  53.8s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.1s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.9s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.7s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.0s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.6s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   8.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   8.5s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   8.2s\n",
      "[CV] bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced, criterion=entropy, max_features=sqrt, n_estimators=450, total=   8.2s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=150, total= 1.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=250, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=350, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 3.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 3.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 3.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  23.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  23.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  23.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  23.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=150, total=  25.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  40.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  38.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  39.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  38.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=250, total=  39.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  54.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  53.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  55.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  54.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=350, total=  54.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   2.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=150, total=   3.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   4.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   4.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   4.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   4.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=250, total=   4.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   6.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   5.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   6.2s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   6.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=350, total=   6.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   8.2s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   8.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   8.2s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=gini, max_features=sqrt, n_estimators=450, total=   9.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.0min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.0min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.0min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.0min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=150, total= 1.0min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.8min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=250, total= 1.7min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.4min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=350, total= 2.5min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 3.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=None, n_estimators=450, total= 3.1min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  23.2s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  23.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  23.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  23.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=150, total=  23.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  38.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  38.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  38.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  41.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=250, total=  38.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  53.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  52.9s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  53.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  53.0s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=350, total=  53.9s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=0.33, n_estimators=450, total= 1.2min\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   2.9s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=150, total=   3.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.8s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.7s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=250, total=   4.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.6s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.2s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=350, total=   6.5s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   9.1s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   8.9s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   8.4s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   8.3s\n",
      "[CV] bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450 \n",
      "[CV]  bootstrap=False, class_weight=balanced_subsample, criterion=entropy, max_features=sqrt, n_estimators=450, total=   8.3s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Done 480 out of 480 | elapsed: 390.7min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise-deprecating',\n",
       "       estimator=RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "            min_samples_leaf=1, min_samples_split=2,\n",
       "            min_weight_fraction_leaf=0.0, n_estimators='warn', n_jobs=4,\n",
       "            oob_score=False, random_state=42, verbose=0, warm_start=False),\n",
       "       fit_params=None, iid='warn', n_jobs=None,\n",
       "       param_grid={'n_estimators': [150, 250, 350, 450], 'criterion': ['gini', 'entropy'], 'max_features': [None, 0.33, 'sqrt'], 'bootstrap': [True, False], 'class_weight': ['balanced', 'balanced_subsample']},\n",
       "       pre_dispatch='2*n_jobs', refit='Precision',\n",
       "       return_train_score='warn',\n",
       "       scoring={'F1 Score': make_scorer(f1_score), 'Precision': make_scorer(precision_score), 'Recall': make_scorer(recall_score), 'AUC': 'roc_auc'},\n",
       "       verbose=2)"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = RandomForestClassifier(random_state=42, n_jobs=4)\n",
    "scoring = {'F1 Score': make_scorer(f1_score),\n",
    "           'Precision':make_scorer(precision_score),\n",
    "           'Recall':make_scorer(recall_score),\n",
    "           'AUC': 'roc_auc'}\n",
    "\n",
    "gs = GridSearchCV(estimator=model, param_grid=param_grid,\n",
    "                  scoring=scoring, refit='Precision',\n",
    "                  cv=5, verbose=2)\n",
    "\n",
    "gs.fit(X_train_rdkit, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "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>std_fit_time</th>\n",
       "      <th>mean_score_time</th>\n",
       "      <th>std_score_time</th>\n",
       "      <th>param_bootstrap</th>\n",
       "      <th>param_class_weight</th>\n",
       "      <th>param_criterion</th>\n",
       "      <th>param_max_features</th>\n",
       "      <th>param_n_estimators</th>\n",
       "      <th>params</th>\n",
       "      <th>...</th>\n",
       "      <th>std_test_AUC</th>\n",
       "      <th>rank_test_AUC</th>\n",
       "      <th>split0_train_AUC</th>\n",
       "      <th>split1_train_AUC</th>\n",
       "      <th>split2_train_AUC</th>\n",
       "      <th>split3_train_AUC</th>\n",
       "      <th>split4_train_AUC</th>\n",
       "      <th>mean_train_AUC</th>\n",
       "      <th>std_train_AUC</th>\n",
       "      <th>Fingerprint</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>38.291902</td>\n",
       "      <td>0.834929</td>\n",
       "      <td>0.450875</td>\n",
       "      <td>0.001074</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>None</td>\n",
       "      <td>150</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.008432</td>\n",
       "      <td>48</td>\n",
       "      <td>0.999830</td>\n",
       "      <td>0.999970</td>\n",
       "      <td>0.999916</td>\n",
       "      <td>0.999870</td>\n",
       "      <td>0.999924</td>\n",
       "      <td>0.999902</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>RDKit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>63.063664</td>\n",
       "      <td>0.722598</td>\n",
       "      <td>0.451390</td>\n",
       "      <td>0.001023</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>None</td>\n",
       "      <td>250</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.010449</td>\n",
       "      <td>42</td>\n",
       "      <td>0.999817</td>\n",
       "      <td>0.999973</td>\n",
       "      <td>0.999925</td>\n",
       "      <td>0.999869</td>\n",
       "      <td>0.999934</td>\n",
       "      <td>0.999903</td>\n",
       "      <td>0.000055</td>\n",
       "      <td>RDKit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>87.258486</td>\n",
       "      <td>0.296383</td>\n",
       "      <td>0.471239</td>\n",
       "      <td>0.040750</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>None</td>\n",
       "      <td>350</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.010255</td>\n",
       "      <td>36</td>\n",
       "      <td>0.999831</td>\n",
       "      <td>0.999983</td>\n",
       "      <td>0.999932</td>\n",
       "      <td>0.999872</td>\n",
       "      <td>0.999942</td>\n",
       "      <td>0.999912</td>\n",
       "      <td>0.000054</td>\n",
       "      <td>RDKit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>112.010589</td>\n",
       "      <td>0.723698</td>\n",
       "      <td>0.633238</td>\n",
       "      <td>0.134951</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>None</td>\n",
       "      <td>450</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.010674</td>\n",
       "      <td>37</td>\n",
       "      <td>0.999823</td>\n",
       "      <td>0.999971</td>\n",
       "      <td>0.999921</td>\n",
       "      <td>0.999872</td>\n",
       "      <td>0.999930</td>\n",
       "      <td>0.999903</td>\n",
       "      <td>0.000051</td>\n",
       "      <td>RDKit</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13.863299</td>\n",
       "      <td>0.276404</td>\n",
       "      <td>0.450644</td>\n",
       "      <td>0.000695</td>\n",
       "      <td>True</td>\n",
       "      <td>balanced</td>\n",
       "      <td>gini</td>\n",
       "      <td>0.33</td>\n",
       "      <td>150</td>\n",
       "      <td>{'bootstrap': True, 'class_weight': 'balanced'...</td>\n",
       "      <td>...</td>\n",
       "      <td>0.009332</td>\n",
       "      <td>38</td>\n",
       "      <td>0.999827</td>\n",
       "      <td>0.999962</td>\n",
       "      <td>0.999930</td>\n",
       "      <td>0.999870</td>\n",
       "      <td>0.999922</td>\n",
       "      <td>0.999902</td>\n",
       "      <td>0.000048</td>\n",
       "      <td>RDKit</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 71 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   mean_fit_time  std_fit_time  mean_score_time  std_score_time  \\\n",
       "0      38.291902      0.834929         0.450875        0.001074   \n",
       "1      63.063664      0.722598         0.451390        0.001023   \n",
       "2      87.258486      0.296383         0.471239        0.040750   \n",
       "3     112.010589      0.723698         0.633238        0.134951   \n",
       "4      13.863299      0.276404         0.450644        0.000695   \n",
       "\n",
       "  param_bootstrap param_class_weight param_criterion param_max_features  \\\n",
       "0            True           balanced            gini               None   \n",
       "1            True           balanced            gini               None   \n",
       "2            True           balanced            gini               None   \n",
       "3            True           balanced            gini               None   \n",
       "4            True           balanced            gini               0.33   \n",
       "\n",
       "  param_n_estimators                                             params  ...  \\\n",
       "0                150  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "1                250  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "2                350  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "3                450  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "4                150  {'bootstrap': True, 'class_weight': 'balanced'...  ...   \n",
       "\n",
       "   std_test_AUC  rank_test_AUC  split0_train_AUC  split1_train_AUC  \\\n",
       "0      0.008432             48          0.999830          0.999970   \n",
       "1      0.010449             42          0.999817          0.999973   \n",
       "2      0.010255             36          0.999831          0.999983   \n",
       "3      0.010674             37          0.999823          0.999971   \n",
       "4      0.009332             38          0.999827          0.999962   \n",
       "\n",
       "   split2_train_AUC  split3_train_AUC  split4_train_AUC  mean_train_AUC  \\\n",
       "0          0.999916          0.999870          0.999924        0.999902   \n",
       "1          0.999925          0.999869          0.999934        0.999903   \n",
       "2          0.999932          0.999872          0.999942        0.999912   \n",
       "3          0.999921          0.999872          0.999930        0.999903   \n",
       "4          0.999930          0.999870          0.999922        0.999902   \n",
       "\n",
       "   std_train_AUC  Fingerprint  \n",
       "0       0.000048        RDKit  \n",
       "1       0.000055        RDKit  \n",
       "2       0.000054        RDKit  \n",
       "3       0.000051        RDKit  \n",
       "4       0.000048        RDKit  \n",
       "\n",
       "[5 rows x 71 columns]"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with warnings.catch_warnings():\n",
    "    warnings.filterwarnings(\"ignore\")\n",
    "    df_rdkit_opt = pd.DataFrame(gs.cv_results_)\n",
    "    \n",
    "df_rdkit_opt['Fingerprint'] = \"RDKit\"\n",
    "df_rdkit_opt.to_csv(path_or_buf='./rdkit_opt.csv')\n",
    "df_rdkit_opt.to_pickle('./rdkit_opt.pkl.bz2')\n",
    "\n",
    "df_rdkit_opt.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The best combination of parameters, according to the grid search:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight='balanced_subsample',\n",
       "            criterion='entropy', max_depth=None, max_features=0.33,\n",
       "            max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "            min_impurity_split=None, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            n_estimators=150, n_jobs=4, oob_score=False, random_state=42,\n",
       "            verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gs.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x241c052b5c0>"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Stats for the best estimator\n",
    "_ = (df_rdkit_opt['param_bootstrap'] == True) & \\\n",
    "    (df_rdkit_opt['param_class_weight'] == 'balanced_subsample') & \\\n",
    "    (df_rdkit_opt['param_criterion'] == 'entropy') & \\\n",
    "    (df_rdkit_opt['param_max_features'] == '0.33') & \\\n",
    "    (df_rdkit_opt['param_n_estimators'] == 150)\n",
    "df_rdkit_opt.loc[_].plot.bar(x='param_max_features', y=['mean_test_Precision', 'mean_test_F1 Score'], figsize=(15,6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight='balanced_subsample',\n",
       "            criterion='entropy', max_depth=None, max_features=0.33,\n",
       "            max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "            min_impurity_split=None, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            n_estimators=150, n_jobs=4, oob_score=False, random_state=42,\n",
       "            verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chosen_rf_rdkit = gs.best_estimator_\n",
    "chosen_rf_rdkit"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Morgan Fingerprints"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((829, 2048), (829,))"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test_morgan.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4AAAAIaCAYAAABxviPGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3Xl4TPf7//HnyR6JEHtksxMiyYRYa2nREoraIlWU0eqK0k9LdflQbb+q1pa2GFJqV7XUWqq2UsEEQSTWRBI7icieOb8/RvtrP9WSyuTMJPfjunKRZJx5TS7Jnfu8N0VVVYQQQgghhBBClHx2WgcQQgghhBBCCFE8pAEUQgghhBBCiFJCGkAhhBBCCCGEKCWkARRCCCGEEEKIUkIaQCGEEEIIIYQoJaQBFEIIIYQQQohSwmINoKIoCxRFuaooSuzffF5RFGWWoihnFEU5pihKqKWyCCGEENZEaqQQQgitWHIEMAro/A+f7wLUvff2IvClBbMIIYQQ1iQKqZFCCCE0YLEGUFXV3cDNf3hID2CRanYAKK8oipel8gghhBDWQmqkEEIIrWi5BtAbSPrD+5fufUwIIYQo7aRGCiGEsAgHDZ9buc/H1Ps+UFFexDwFBjc3tyYNGjT4109aoBZwM/Mm1zOvk5mXiaIoeLp4UqlMJco6l/3X1xVCCFH0Dh8+fF1V1cpa59DAv6qRoDQBFTuXsnhUqkoljzJ4uDii3O9qQgghbNaj1EctG8BLgO8f3vcBUu73QFVV5wJzAZo2baoeOnSoSAIYU40YjAaWHF9CfHY8NcvXZKhuKM+HPI+Ph0+RPIcQQoh/T1GUi1pn0Mi/qpGhoaFqq3Yd+Gr2TNKvJmMX+AJlW3XjGZ03/Zv5UqeK3OgUQoiS4FHqo5ZTQNcDg+7tdNYCSFNVNbU4A+i8dHwR/gUpo1NY0msJNT1r8t7O9/Cf4U/4knC+O/kduQW5xRlJCCGEgH9ZI+3s7Phi+hTiTp4k/MmOjOgcRPOaFVi47xwdp+2mz5e/sOpQEpm5+ZZ/BUIIIaySoqr3nVHy6BdWlGVAe6AScAX4AHAEUFX1K0VRFOALzLugZQJDVFV94NBeUY4A3s+5W+dYYFxAVEwUyXeSqVymMoOCB6HX6QmoHGCx5xVCCPFXiqIcVlW1qdY5ilpx1UhVVVEUhfEfTGTDtp8p03YolxVPyjo70D2kOpHN/Aj0Llfkr08IIYRlPUp9tFgDaCmWbgB/U2AqYOvZrRiMBtafXk++KZ+WPi3R6/REBEbg7uRu8QxCCFHaldQG0FL+rkbOnj2bcePGkZ2dTeTQlynboh8/nkknJ99Eo+oe9A/zpXuIN+VcHTVILYQQorCkAbSwq3evsvjoYgxGA6eun8LN0Y2IRhEMCx1GC58WKLK6XgghLEIawML5pxp5+fJl3nnnHRYuXIiXlxez5nxNVtXGLDuYxMnUdFwc7Qhv7EX/MD/CanhKbRNCCCsmDWAxUVWV/Zf2YzhiYMWJFdzNu0tApQD0Oj0DgwdSxa2KJrmEEKKkkgawcB6mRh48eJBRo0Yxc+ZMwsLCKCgo4GRqBsujE1kXk0JGTj61KrvRP8yXXqE+VHJ3Lqb0QgghHpY0gBq4k3OHlSdWYjAa2H9pPw52DnSv3x29Ts9TtZ/C3s5e64hCCGHzpAEsnIetkb+tDQQYPnw4+fn5fPzxx5T1rMjGY6msiE7i0MVbONordGpYlYgwP9rUqYSdnYwKCiGENXiU+qjlLqA2raxzWfShen7R/8KJV04wotkIdl/cTdelXfGf4c97P73HuVvntI4phBBC/MVvzZ+qqnh6erJ48WLq1avH17M/p2dwNVa/3Iof32jL4JY12H/2BoMXHKTNpzuZuT2BlNtZGqcXQgjxKGQEsAjlFuSy4fQGDEYDW89uxaSaeKLmE+h1enoF9MLFwUXriEIIYVNkBLBw/m2NjI+PZ9SoUWzevJmAgAC+/fZbQkNDAcjJL+DHk1dYEZ3EnoTr2CnQrl5lIsL86BBQBUd7uZcshBDFTaaAWqFL6ZeIioligXEB52+fp7xLeQY0HoBep0fnpdM6nhBC2ARpAAvnUWvkxo0bGT9+PBs2bMDX1/dPU0UBkm5msvJQEisPJXElPYdK7s70aeJDRJgvNSu5FcVLEEII8RCkAbRiJtXEzvM7MRgNrDm1hpyCHEK9QtHr9Dzb+FnKu5TXOqIQQlgtaQALpyhq5G9Nn6qqdOvWjZCQEMaNG4e7+/8//ii/wMSu+Gssj07ip7irFJhUmtesQGQzPzoHVsPFUdbBCyGEJUkDaCNuZt1k6fGlGIwGYi7H4OLgQu+A3uh1etrVaIedItNohBDij6QBLJyirJFZWVkMHz6cxYsXU716daZMmUJkZORfjoe4mp7NqsOXWBGdROLNTDxcHHhG503/Zn4EeHkUSRYhhBB/Jg2gDTqSeoT5R+az9PhS0nLSqOVZC71Oz+DgwXh7eGsdTwghrII0gIVjiRq5f/9+Xn/9dQ4fPkzr1q1ZsmQJ/v7+f3mcyaRy4NwNlkcnsSX2MrkFJoJ9ytG/mR9PB1fH3dmhSHMJIURpJg2gDcvMy2TNqTUYjAZ+vvAzdoodXep0Qa/T061eNxztHbWOKIQQmpEGsHAsVSNNJhNRUVF8/vnn7NmzB3d397+sD/yjW3dz+d6YzPLoROKvZFDGyZ5uQV70b+aHzre8HDIvhBCPSBrAEuLMzTMsMC4gKiaK1IxUqrhVYXDwYIbqhtKgUgOt4wkhRLGTBrBwLF0jf2v6cnNzadeuHZGRkbz88ss4Ot7/ZqWqqsQk3Wb5wSQ2HEshM7eAelXdiQjzo5fOG083J4tlFUKIkkwawBIm35TPljNbMBgN/BD/A/mmfFr7tkav09O3UV/cndwffBEhhCgBpAEsnOKqkVevXuW5557jxx9/pFGjRsycOZMOHTr847/JyMnnh6MpLItO4mjSbZzs7XgqsBr9w3xpWauiHDIvhBCFIA1gCXYl4wqLji7CYDRw+sZp3J3c6d+oP/pQPc29m8s0GiFEiSYNYOEUZ41UVZX169fzxhtvcP78eXr16oXBYKB8+Qfvbn0qNZ0V0UmsOXKJ9Ox8/CqUISLMlz5NfKjqIWfmCiHEg0gDWAqoqsq+pH0YjAZWnlhJZl4mjSo3Qq/TMzB4IJXKVNI6ohBCFDlpAAtHixqZnZ3NtGnT2LJlCzt37sTe3v4f1wf+6d/mFbD1xGWWHUzkwLmb2NspPF6/Cv3DfGlfvzIOcsi8EELclzSApUx6TjorYldgMBr4NflXHO0c6dGgB3qdnk61OmFvJ+cvCSFKBmkAC0fLGvlb03fjxg06duzI2LFj6dev30PPVDl//S4rDyWx6tAlrmfkUNXDmb5NfOnX1Be/imUsnF4IIWyLNIClWOzVWBYYF7Do6CJuZN3A18OX50OeZ6huKDXK19A6nhBCPBJpAAvHGmpkXFwckZGRxMTE0LZtW2bNmkVwcPBD//u8AhM/xV1lRXQSP5++ikmFx+pUIiLMlycbVcXZQW5yCiGENICCnPwc1p9ej8FoYNvZbQB0qNUBvU5PzwY9cXGQNRVCCNsjDWDhWEuNLCgoYP78+YwfP55bt24xfPhwZs2ahYND4c4CTLmdxep7h8wn387Cs4wjvUJ96B/mS92qZS2UXgghrJ80gOJPEtMSiYqJYoFxARfTLuLp4slzQc+h1+kJrvbwd2GFEEJr0gAWjrXVyFu3bvHBBx+QmprKqlWrAB56feAfmUwqe89cZ0V0EttOXiavQKWJvycRYb50C/KijJMcMi+EKF2kART3ZVJN/HT+J+Yfmc/3cd+TW5BL0+pN0ev0RAZGUs6lnNYRhRDiH0kDWDjWWiNNJhN2dnacPn2a5557js8++4x27dr9q2tdz8jh+yPmQ+bPXruLu7MD3UOq0z/Ml8be5WR3bCFEqfAo9VG21yrB7BQ7OtbqyPI+y0kZncLMzjPJLcjl5Y0v4zXVi0HfD2LXhV3Y2k0AIYQQtsXOzvzrxtWrV7l27Rrt27cnIiKCxMTEQl+rkrszL7StxfbR7Vj1UkuealSNNUcu0f2LfYTP2sui/RdIy8wr4lcghBAlh4wAljKqqnI49TCGIwaWxi4lPSedOhXqMDRkKINDBlO9bHWtIwohxO9kBLBwbKFGZmVlMWXKFP7v//4PgPfee49x48Y90jXTs/NYF5PCiuhEYpPTcXawI7yxF/3DfGlWs4KMCgohShyZAir+lcy8TFafXI3BaGD3xd3YK/aE1w1Hr9MTXjccR3tHrSMKIUo5aQALx5ZqZGJiIv/5z3/w8fFh6tSpwL9bH/i/YpPTWB6dyDpjCndy8qlVyY1+Yb70DvWhclnnooguhBCakwZQPLKEGwksMC4g6mgUlzMuU9WtKoODB6MP1VOvYj2t4wkhSilpAAvHFmvkb+sDf/rpJz7++GNmzJhBYGDgI183K7eAjcdTWRGdSPSFWzjYKXQMqEr/Zr60qVsZezsZFRRC2C5pAEWRyTflsylhEwajgY3xGylQC2jj1wa9Tk+fhn1wc3LTOqIQohSRBrBwbLlGrlq1iuHDh5Oens4rr7zChAkT8PT0LJJrn7mawYroRL47kszNu7lUL+dC36a+9Avzxbu8a5E8hxBCFCdpAIVFpN5JZdHRRRiMBhJuJlDWqSyRgZHoQ/WEVQ+TNRVCCIuTBrBwbL1G3rhxg/fff5+vvvoKT09Ppk+fzsCBA4vs+rn5JrafusKyg4nsPXMdgLZ1K9M/zJcOAVVxcpC98YQQtkEaQGFRqqqyJ3EPBqOBVSdWkZWfReMqjdHr9DwX9BwVy1TUOqIQooSSBrBwSkqNPHr0KCNGjKBv37689tprFnmOpJuZrDp8iVWHkkhNy6aSuxO9Q33oF+ZL7cruFnlOIYQoKtIAimKTlp3G8tjlGIwGolOicbJ3omeDnuh1ejrW6oidIndPhRBFRxrAwilJNVJVVVRVxc7OjgULFrB9+3Y+/fRTfHx8ivR5Ckwqu+OvsexgIjvirlJgUmlWswL9w3wJb+yFi6N9kT6fEEIUBWkAhSaOXzmOwWhg8bHF3My6iV85P4aEDGFIyBD8y/trHU8IUQJIA1g4JbVGTp8+nXHjxmFvb8/48eMZPXo0Li4uRf48V+9k893hZFZEJ3LhRiZlXRx4RudNRJgvjaqXK/LnE0KIf0saQKGpnPwc1satxWA0sP3cdgA61e6EXqenR/0eODvItttCiH9HGsDCKck18vz587z55pusWbOGWrVq8dVXX9GpUyeLPJeqqhw4d5MV0Ylsir1Mbr6JIJ9yRIT50j24OmVd5JgkIYS2pAEUVuPi7YssjFnIwpiFJKYlUtG1Is8FPYdep6dx1cZaxxNC2BhpAAunNNTIHTt2MHLkSCZPnkzXrl0t/nxpmXl8b7zE8ugk4i7fwdXRnm5BXvRv5kuon6dsiCaE0IQ0gMLqFJgK2HF+B/OPzGdt3FryTHmEVQ9jWOgw+gf2x8PZQ+uIQggbIA1g4ZSWGpmfn4+DgwMAEyZMIC0tjQ8++IBy5Sw3TVNVVY5dMh8yvz4mhbu5BdSt4k5EmC+9Qn2o4OZksecWQoj/9Sj1UXbsEBZhb2fPk7WfZGXflaSMSWH6U9PJys9i+A/D8ZrqxfNrn2fPxT3Y2g0IIYQQ2vut+QO4efMmM2bMoG7duhgMBkwmk0WeU1EUgn3L80mvIA6O78invYNwd3Fg0sZTtPh4B68tPcLehOuYTFLXhBDWTUYARbFRVZXolGgMRwwsi13Gndw71KtYj6EhQxkcMphq7tW0jiiEsDIyAlg4pbVGHjlyhBEjRrBv3z6aNm3KvHnzCAkJKZbnPn35DsujE/nemMztzDx8K7gS0dSXPk18qVau6DeqEUIIkCmgwgbdzb3LqpOrMBgN7E3ci71iT7d63dDr9HSp2wUHO4cHX0QIUeJJA1g4pblGqqrKsmXLGDduHBs2bCAoKKhYnz87r4BtJ6+w/GAiv5y9gZ0Cj9evQv9mfjxevzIO9jLpSghRdKQBFDbt9PXTLDAu4Juj33Dl7hW83L0YHDyYobqh1K1YV+t4QggNSQNYOFIj/7w+8KWXXqJmzZqMGjUKZ+fi25H64o27rDyUxKpDl7h6J4cqZZ3p08SHiDBf/Cu6FVsOIUTJJQ2gKBHyCvLYlLCJ+cb5bErYhEk10da/LcN0w+jdsDdlHMtoHVEIUcykASwcqZH/X15eHn379mXdunXUqVOHGTNmFMuuoX+UX2Bi5+lrrIhO5Ke4q5hUaFW7Iv2b+fFkw6pyyLwQ4l+z2k1gFEXprCjKaUVRziiKMvY+n/dTFGWnoihGRVGOKYoSbsk8wro52jvSo0EPNkRuIOmNJD5+4mOS05MZtHYQXlO9ePmHlzmUckg2jhFC2Dypj5bn6OjI2rVr2bp1Kw4ODnTr1o3w8HASExOLLYODvR2dGlZl/uAwfhnbgTefrEfSrUxGLDPS4pMdTNhwgtOX7xRbHiGEAAuOACqKYg/EA52AS0A0EKmq6sk/PGYuYFRV9UtFURoCm1RVrfFP15W7m6WLqqrsvrib+cb5rD65muz8bIKqBjFMN4wBQQOo4FpB64hCCAsqiSOAlqqPIDXy7+Tl5fHFF18wffp0oqOjqVq1qmZZTCaVX87eYHl0IttOXCG3wITOrzz9w3zpFlQdN2dZAy+EeDBrHQFsBpxRVfWcqqq5wHKgx/88RgV+OxCuHJBiwTzCBimKQrsa7Vj8zGJSx6QyJ3wOjnaOjNgygupTqxP5XSTbz23HpFpm228hhLAAqY/FzNHRkTfeeIOzZ89StWpVVFWld+/efPPNNxY7NuLv2NkpPFa3El88G8qBdzrwbtcA7mTn8/Z3x2n20XbGrTlGTNJtme0ihLAYS44A9gE6q6o67N77A4Hmqqq+9ofHeAHbAE/ADeioqurh+1zrReBFAD8/vyYXL160SGZhO45ePorBaODbY99yK/sWNcrXYEjIEIaEDMG3nK/W8YQQRaSEjgAWWX2891ipkYV048YNunXrxoEDB2jWrBmff/45zZo10yyPqqocSbzF8oNJ/HAslay8AhpUK0v/MF966rwpX0YOmRdC/Jm1jgAq9/nY/3abkUCUqqo+QDiwWFGUv2RSVXWuqqpNVVVtWrlyZQtEFbYmuFows7rMImVMCkt7LaW2Z20++PkD/Gf402VJF1afXE1uQa7WMYUQ4n6KrD6C1Mh/o2LFiuzbt49FixaRmJhI8+bNGTJkCLdu3dIkj6IoNPGvwJS+wRwc34GPngnEycGO/244SbOPdzBquZH9Z2/IqKAQokhYcqL5JeCPQzE+/HUKix7oDKCq6n5FUVyASsBVC+YSJYiLgwuRjSOJbBzJ+VvnWRizkIUxC+m7qi+VylRiYNBA9Do9jao00jqqEEL8RuqjFbCzs2PgwIH07NmTjz76iLVr1+Liov3B7WVdHBnQ3J8Bzf05kZLGiugkvjcmszYmhRoVyxAR5kfvJt5UKat9ViGEbbLkFFAHzIvcOwDJmBe5P6uq6ok/PGYzsEJV1ShFUQKAHYC3+g+hZIG7eJACUwE/nvsRg9HAurh15JnyaO7dnGGhw4hoFEFZ57JaRxRCPKQSOgXUIvURpEY+itzcXJycnMjKyuKZZ55h1KhRdO7cWetYgPmQ+U3HU1kencTB8zdxsFPoEFCF/mF+tK1XGXu7+w0qCyFKMqs9B/DettUzAHtggaqqHymKMhE4pKrq+ns7m80D3DFPf3lLVdVt/3RNKW6iMK7dvcbiY4sxGA2cvHYSN0c3+jXqh16np5VvKxRFiqYQ1qwkNoBgmfoIUiOLQnx8PN26dSMhIYGnn36aadOmUadOHa1j/e7stQxWRiex+vAlbtzNxaucC32b+tKvqQ8+nnJerhClhdU2gJYgxU38G6qq8mvyrxiOGFh+YjkZuRk0qNSAoSFDGRQ8iKru2m0JLoT4eyW1AbQUqZFFIzc3l5kzZzJx4kRyc3MZPXo0EyZMwMnJejZjyc038VPcFZYdTGJ3wjUAHqtTichmfnQMqIqTg0WPehZCaEwaQCEKISM3g5UnVmIwGvgl6Rcc7Bx4ut7T6HV6nqrzFA52cgaTENZCGsDCkRpZtFJTUxk3bhxnzpxhz549VjtrJPl2FqsOJbEyOomUtGwqujnRu4kP/Zr6UqeKu9bxhBAWIA2gEP/SqWunWGBcwKJji7h69yrVy1bn+eDnGaobSu0KtbWOJ0SpJw1g4UiNtIycnBycnZ25fPkyzz//PB999BFNmjTROtZfFJhU9iRcY/nBJLafukK+SSWshif9w/wIb+yFq5O91hGFEEVEGkAhHlFeQR4/xP+AwWhg85nNmFQT7Wu0R6/T0zugN66OrlpHFKJUkgawcKRGWtaePXvo06cP165dQ6/X89FHH1GlShWtY93XtTs5rDlyiRXRSZy7fpeyzg700FWnf5gfgd7ltI4nhHhE0gAKUYSS05OJioliQcwCzt06RznncgxoPAB9qJ5Qr1Ct4wlRqkgDWDhSIy0vLS2NDz/8kJkzZ+Lm5saECRMYMWKE1U4PVVWVg+dvsiI6iY3HU8nJNxHo7UFEmB89Qqrj4eKodUQhxL8gDaAQFmBSTey6sAuD0cDqk6vJKcghpFoIep2eAY0H4OnqqXVEIUo8aQALR2pk8YmLi2PUqFFUrFiRJUuWaB3noaRl5rHuaDLLDiZxKjUdF0c7ujauTmQzX5r4e1ptEyuE+CtpAIWwsFtZt1h6fCkGowHjZSPO9s70btgbvU5P+xrtsVNktzUhLEEawMKRGlm8VFUlJycHFxcXjh8/zn//+1+mTJlCrVq1tI72j1RV5XhyGsujk1gfk0JGTj61K7vRP8yPXqHeVHR31jqiEOIBpAEUohgZU40YjAaWHF/C7ezb1PKsxZCQITwf8jw+Hj5axxOiRJEGsHCkRmpn+fLl6PV6CgoKePPNNxk3bhxubm5ax3qguzn5bDyeyoroJA5fvIWjvcKTDasREebLY3UqYSeHzAthlaQBFEIDWXlZrDm1BoPRwM4LO7FT7OhcpzN6nZ5u9brhZG8950UJUdxyc6EojkyTBrBwpEZq69KlS7z99tssXboUHx8fpk2bRt++fbWO9dDir9xhRXQSa45c4lZmHt7lXYkI86VvUx+8yslmaEJYE2kAhdDY2ZtnWRizkKiYKJLvJFO5TGUGBQ9Cr9MTUDlA63hCFAtVhd274auv4OBBOH0aHB7xWE1pAAtHaqR12Lt3LyNGjCA8PJxJkyZpHafQcvIL2HbiCiuik9h75jp2CvQI8WZceAOqlHXROp4QAmkAhbAaBaYCtp7disFoYP3p9eSb8mnp0xK9Tk9EYATuTnIgryh5bt2CxYvNjd+pU1C+PAweDBMngofHo11bGsDCkRppPQoKCsjPz8fZ2ZlNmzaxfv16Jk2aRKVKlbSOViiJNzJZ8utFFu67gLOjHW89VZ9nm/tjL1NDhdDUo9RH2blCiCJkb2dPeN1wvuv3HZfeuMSUTlO4lX2LYRuG4TXVi2Hrh7E/aT+2duNFiP+lquZRvqFDwdsbRo6EsmVh4UJIToYZMx69+RPCltnb2+PsbN5M5cSJE8yfP5+6devyxRdfkJ+fr3G6h+dXsQzjwgPYMqoNQT7leG/dCXrN2cfxS2laRxNC/EsyAiiEhamqyv5L+zEcMbDixAru5t0loFIAep2egcEDqeJmnYcIC3E/GRmwbBl8+SUYjeDmBgMGwPDhEGqBYzJlBLBwpEZarxMnTjBy5Eh27NhBYGAgs2fPpm3btlrHKhRVVVl/NIUPfzjFzbs5DGpZg9FP1pOzBIXQgIwACmHFFEWhlW8rDD0MpI5JZf7T8ynnUo43f3wT72ne9FnZh80JmykwFWgdVYi/dfw4vPoqVK8OL74I+fkwZw6kpMDXX1um+ROiJGnUqBE//vgj3333HRkZGSQkJGgdqdAURaFHiDc7xrTjuRb+fLP/Ah2n7mLD0RSZ2SKEDZERQCE0cvLaSQxHDCw6tojrmdfx8fDh+eDnGaobSk3PmlrHE4LsbFi92ry2b98+cHaGfv3gpZegZUsojjOjZQSwcKRG2oasrCycnZ2xs7Nj3rx5JCcn89Zbb1GmTBmtoxXKsUu3Gf99LMeT02hTtxITewRSs5L1H30hREkgI4BC2KCGlRsy9ampJI9OZnXf1TSu0piP935MrVm16LCoA0uPLyU7P1vrmKIUSkiAN980r+0bOBCuXIHPPjOv7Vu0CFq1Kp7mT4iSytXVFTs7869gR44cYcKECQQEBLBq1SqbGkkL8inP2ldbM7FHI2ISb/PUjN3M2B5Pdp7MaBHCmskIoBBWJCktiaiYKBbELODC7Qt4ungyoPEA9KF6QqqFaB1PlGB5ebB+vXm0b/t28/ENPXuaR/sefxzsNLpdKCOAhSM10jbt2rWLESNGcOzYMdq3b8/s2bNp2LCh1rEK5Wp6NpM2nmL90RRqVCzDhz0DaVO3staxhCixZARQiBLCt5wv77V7j7MjzrJ94HY61+nMvCPz0H2to8ncJsyJnsPt7NtaxxQlSGIivP8++PtDnz7ms/s+/ND88VWroEMH7Zo/IUqLdu3acfjwYebMmcOxY8e4fPmy1pEKrYqHC7MidXyrb46iKAw0HOS1pUe4ki4zWYSwNjICKISVu5l1k6XHlzL/yHyOXjmKi4MLvQN6Myx0GO3826HIXDxRSAUFsHWrebRv40bzkQ7h4ebRvi5dwN5e64T/n4wAFo7USNuXkZGBu7v5zNgPP/yQSpUq8eKLL2JvTd+YD5CdV8DXu84x++czONnb8eaT9RjYsoacHShEEZKD4IUoBVRV5UjqEQxGA0uPLyUtJ43anrUZqhvK4ODBeHt4ax1RWLkrV2DBApg7Fy5cgCpVYNgweOEFqFFD63T3Jw1g4UiNLDlMJhOdO3fmxx9/JDg4mFmzZtncsREXrt/lvXWx7Em4TqC3Bx/1bEywb3mtYwlRIsgUUCFKAUVRaFK9CXO6ziFlTAqLn1mMj4cP438aj98MP7ot7cbauLXkFeRpHVVYEVWFnTshIgJ8fOCdd6C/JRm7AAAgAElEQVRmTVi5EpKS4KOPrLf5E6I0s7OzY+vWraxcuZKbN2/Srl07+vfvT3JystbRHlqNSm4sGtqML57VcTU9h55z9vHe2ljSsqROCaElGQEUwsaduXmGBcYFRMVEkZqRSlW3qgwKHoRep6d+pfpaxxMauXnTvGPnV1+Z1/V5esLzz5vP8GvQQOt0D09GAAtHamTJlJmZyaeffsr06dM5ePAg9evb3s/2O9l5TPsxnm9+uUAFN2fe7RpAj5DqsoxBiH9JpoAKIcg35bPlzBYMRgMbTm+gQC2gtW9r9Do9/Rr1w81JzmYq6VQVfv3V3PStWGE+x69FC3j5ZejbF1xdtU5YeNIAFo7UyJItPT0dDw8PAEaOHEn79u3p2bOnTTVRsclpjF8by9Gk27SqXZEPewZSu7K71rGEsDnSAAoh/uRyxmUWHV2EwWgg/kY8ZZ3K0j+wP3qdnmbezWzqlwXxYHfuwNKl5sYvJgbc3eG552D4cAix8dNDpAEsHKmRpUN6ejqtW7cmNjaWjh07MnPmTJs6NqLApLLsYCKfbokjO8/E8Ha1ePXxOrg42s5GN0JoTdYACiH+pJp7Nd5q/RZxr8axZ8geejfszZLjS2hhaEHjLxszff90rmde1zqmeERHj5pH96pXN+/gaTLBl19CSor5T1tv/oQQ9+fh4YHRaOTzzz/n0KFDBAUFMWrUKNLS0rSO9lDs7RSea+HPjjHt6Rrkxec/neHJ6bv5+fRVraMJUSrICKAQpUR6TjorYlcw3zifg8kHcbRzpEeDHgzTDaNjrY7Y28mdV1uQlWU+n++rr2D/fnBxMW/w8tJL0Lw5lLTBXRkBLBypkaXP9evXeffdd/nuu+84ffo0FSpU0DpSof1y5jrvrovl3LW7hDeuxvvdGlGtnIvWsYSwajIFVAhRKLFXYzEcMbD42GJuZN3A18OXISFDGKIbQo3yNbSOJ+4jPh6+/hqioswbvNSrZ276Bg8GG/x976FJA1g4UiNLr7S0NMqVK0dBQQHDhg1j2LBhtG7dWutYDy0nv4B5u8/x+U9ncLBTeKNTPZ5vVQMHe5msJsT9SAMohPhXcvJzWH96PQajgW1ntwHQoVYH9Do9PRv0xMVB7sBqKTcX1q0zj/b99BM4OECvXubGr337kjfadz/SABaO1Ehx9uxZ2rVrR3JyMgMGDGDy5Ml4e9vOObGJNzJ5f30sP5++RoCXBx89E0ion6fWsYSwOtIACiEeWWJaIguNC1kYs5CLaRep4FqB5xo/hz5UT1DVIK3jlSoXL8K8eTB/vvnwdn9/8/ENQ4dCtWpapyte0gAWjtRIAXD37l0++eQTPvvsMxwcHBg/fjxjxozByclJ62gPRVVVtsReZsKGk1y5k03/MD/e7lyf8mVsI78QxUEaQCFEkTGpJnac24HBaOD7uO/JLcilafWm6HV6IgMjKedSTuuIJVJBAWzebB7t27TJ/LGuXc2bvDz1FNiX0iWa0gAWjtRI8Ufnzp1jzJgxnDp1imPHjtlMA/ibjJx8ZvwYz8JfLlDe1ZF3wgPoFeotO1kLgTSAQggLuZF5gyXHlzD/yHyOXz2Oq4MrfRr2Qa/T09a/rRThInD5MhgMMHcuJCaaR/iGDTO/+ftrnU570gAWjtRIcT+3bt3C09OTjIwMXn75ZcaPH0+DBg20jvXQTqakM37tcYyJt2leswIfPRNInSpltY4lhKakARRCWJSqqhxKOYTBaGBZ7DLSc9KpW6EuQ3VDGRw8GK+yXlpHtCmqal7T99VXsHYt5OdDhw7m0b7u3cHRUeuE1kMawMKRGin+yb59+wgPDyczM5MRI0bw/vvvU66cbczqMJlUVhxK4v82x5GZm88LbWrx+hN1cXUqpdMjRKknDaAQothk5mWy+uRqDEYDuy/uxl6xJ7xuOHqdnvC64TjaS/fyd27cgG++Me/mGR9v3r1zyBDz+r569bROZ52kASwcqZHiQa5evcr48eMxGAxUrlyZTz75hCFDhtjMjI4bGTl8vCmO745cwsfTlQndG9EhoKrWsYQodtIACiE0EX8jngXGBXxz9BsuZ1ymmns1BgcPZqhuKPUqSkcD5tG+/fvNo30rV0JODrRqZR7t69PHfI6f+HvSABaO1EjxsA4fPszrr7+Oq6sr27dvt5kG8De/nrvBu2tjSbiawVONqvLB042oXt5V61hCFBtpAIUQmsoryGPzmc0YjAY2xm+kQC2gjV8b9Do9fRr2wc3JTeuIxS49HZYsMTd+x45B2bIwcCAMHw5BsqnqQ5MGsHCkRorCUFWV27dv4+npSVJSEhMnTmTixIl4ednGtP7cfBPz955j1o4E7BSFUR3rMqR1TRzl7EBRCjxKfZTvECHEI3O0d6R7/e6s67+OpDeS+KTDJ1zOuMzz657Ha6oXL/3wEtHJ0djaDad/IybGfE6ftze88grY2ZmnfCYnw+zZ0vwJIayHoih4eprP2Pvll1/45ptvqFevHlOmTCE3N1fjdA/m5GDHK+3r8OMb7WhVuyIfb4rj6c/3cujCTa2jCWHVLNoAKorSWVGU04qinFEUZezfPKafoignFUU5oSjKUkvmEUJYnldZL8Y+NpbTr51m1/O7eCbgGRYdXUSz+c0I/iqYmQdmciPzhtYxi1RmJkRFQYsWoNOZ1/n17Qu//gpHjpjX+JWVDevEH0h9FNYmIiKCEydO0L59e9566y0CAwPZ9NuZNFbOt0IZ5g8OY+7AJqRn5dHnq/28vfoYt+5afxMrhBYsNgVUURR7IB7oBFwCooFIVVVP/uExdYGVwBOqqt5SFKWKqqpX/+m6Mr1FCNuTlp3G8tjlGIwGolOicbJ3omeDngzTDaNDrQ7YKbY5GSEuzjy6FxUFt29Dgwbm0b9Bg+DeTXXxiEriFFBL1UeQGimKxubNmxk1ahRt2rRh/vz5WscplLs5+czakYBh73nKujgwLjyAPqE+2NnZ1hpHIR7EKtcAKorSEvivqqpP3Xt/HICqqp/84TGfAvGqqj70TxcpbkLYtmNXjmE4YuDb499yM+sm/uX8GRIyhCG6IfiV89M63gPl5sL335vX9v38s/nIht69zY1f27ZgY/soWL0S2gBapD6C1EhRdHJzc8nOzsbDw4Po6GhWr17Nu+++S1kbmc5w+vId3l17nOgLtwir4cmkno2pX802sgvxMKx1DaA3kPSH9y/d+9gf1QPqKYqyT1GUA4qidLZgHiGEFQiqGsTMLjNJHp3M8t7LqVexHhN2TaDGjBo89e1TrDyxkpz8HK1j/sX58/DOO+DrC/37w4UL8MknkJQEy5ZBu3bS/ImHJvVRWD0nJyc8PDwA+Omnn/j000+pV68eixYtwmQyaZzuwepXK8uKF1vyae8gzlzNoOusPXyy+RSZuflaRxNCc5ZsAO/3q9D/Djc6AHWB9kAkMF9RlPJ/uZCivKgoyiFFUQ5du3atyIMKIYqfi4MLEYERbBu4jXMjz/Fe2/c4de0UEasj8J7mzRtb3iD2aqymGQsKYP16CA+H2rVh8mTzOr/Nm+HsWRg7FqrK8VOi8IqsPoLUSGF5b7/9NgcOHMDX15fBgwfTunVroqOjtY71QHZ2Cv3CfNkxpj29Q334etc5Ok3bzbYTl7WOJoSmLNkAXgJ8//C+D5Byn8esU1U1T1XV88BpzAXvT1RVnauqalNVVZtWrlzZYoGFENqoUb4GEx6fwPmR59kyYAtP1HyC2dGzafxlY5rPb87cw3NJz0kvtjwpKfDhh1CzJvToYd7Z8733zKN+69ZB587m3T2F+JeKrD6C1EhRPJo3b86BAwdYuHAh58+fZ9u2bVpHemgV3JyY3CeI1S+1xN3ZgRcXH2bYN9FcupWpdTQhNGHJNYAOmBe5dwCSMS9yf1ZV1RN/eExnzAvfByuKUgkwAiGqqv7tFoGyvkGI0uF65nW+PfYtBqOB2KuxlHEsQ9+GfdHr9Dzm91iRH1psMsGOHea1fevWmUf/OnUyH9jerZt5rZ8ofrayBlBRFG/AH/PIHQCqqu7+m8dapD6C1EhRPNLT03FycsLFxYV169Zx9uxZXn/9dRxt4AdlXoGJhfvOM/3HBABGdKjLsDZydqCwPVa5BlBV1XzgNWArcApYqarqCUVRJiqK0v3ew7YCNxRFOQnsBP7zoOImhCgdKpWpxKgWozj20jEO6A8woPEA1pxaQ9uotjSY3YBP933K5YxHn8Zz/Tp89hnUrw9PPgm7dsHo0ZCQANu2wTPPSPMn/pmiKJOBfcC7wH/uvb35d4+X+ihsnYeHBy4uLgBs3LiRMWPGEBQUZBOjgo72drzYtjbbx7SjTd1KTN4SR/jMPfx6Tr69ROlhsRFAS5G7m0KUXndz77Lq5CoMRgN7E/dir9jTrV439Do9Xep2wcHO4cEXAVQVfvnFPNq3ahXk5MBjj5l38uzdG+79XiOsgC2MACqKchoIUlVV892LpEaK4qaqKhs3bmTUqFGcPXuW7t27M23aNGrXrq11tIey49QV3l93guTbWfRp4sO4Lg2o6O6sdSwhHsgqj4GwlPsVt7y8PC5dukR2drZGqWybi4sLPj4+NjF1Q4jfxF2PY4FxAd8c/Yard6/i5e7F4ODBDNUNpW7F+y6VIi0Nvv3W3PjFxoKHh/nMvuHDITCwmF+AeCg20gBuBvqqqpqhdRZpAIVWcnJymD59OpMmTWLu3Lk8++yzWkd6aFm5Bcz6KYF5u8/h5uzAuC4N6NfUV84OFFat1DeA58+fp2zZslSsWLHI1wWVdKqqcuPGDe7cuUPNmjW1jiNEoeUV5LExYSMGo4FNCZswqSba+bdDr9PTu2FvyjiW4cgRc9O3dCncvQuhoea1ff37g7u71q9A/BMbaQC/A4KBHcDvo4Cqqo4o7izSAAqtXblyhSpVqqAoCvPmzcPNzY3IyEib+P0s4codxq+N5eD5m4T6leejZxoT4OWhdSwh7qvUN4CnTp2iQYMGNvHDxRqpqkpcXBwBAQFaRxHikSSnJ/PN0W9YYFzA2VtncVU8cL/wLNe26nG51YQBzyq89BI0tep2QvyRjTSAg+/3cVVVvynuLNIACmuhqipPPPEEP//8M61bt+bzzz9Hp9NpHeuBVFVlzZFkPtp0irSsPIa0qsGoTvVwd364JQZCFBer3ASmuEnz9+/J106UFN4e3vSs+A7hZ+NxW7WTrJjuXPeOghfDqDNFR/ALn1Or0U2tY4oS5l6jtww4fO9tqRbNnxDWRFEUtm/fzrx584iPj6dJkya8+OKLWPtZlYqi0LuJDz+NaUe/pr7M33ueTtN2sSU2FVsbNBHi75SYBlAIUXrl5MCyZdCuHTRqBF99aUf3oPbsfmMxN95JZU74HJwdHRixZQTVp1Yn8rtItp/bjkk1aR1dlACKorQHEoDZwBwgXlGUtpqGEsIK2NvbM2zYMOLj4xk5ciRRUVEkJCRoHeuhlC/jxCe9GvPdy60oX8aJl749wtCoaJJuytmBwvZJA1hE7O3tCQkJITAwkL59+5KZ+eg/IA4dOsSIEX+/hCQlJYU+ffo88vMIYavOnYOxY8HXF559Fi5dgsmTzX8uXQpt2oCna3leDnuZQy8ewjjcyAuhL7D1zFY6Le5E7Vm1+XDXhySlJWn9UoRtmwo8qapqO1VV2wJPAdM1ziSE1ShfvjzTp0/nwoULtGrVCoDJkyezY8cOjZM9WBN/Tza81pp3uwZw8PxNOk7bxeydZ8jNlxuIwnaVmDWAWq9fc3d3JyPDvAHcgAEDaNKkCaNHj/7986qqoqoqdnbW2XNbw9dQiIeRnw8//GDe1GXrVrC3h6efNm/q0rEjPMy3WHZ+Nt+f+h6D0cCO8ztQUHiqzlPodXq61++Ok72T5V+IeCg2sgbwmKqqQQ/6WHGQNYDCFmRlZREUFMSZM2fo1asXU6dOpUaNGlrHeqDUtCwmbjjJ5tjL1K7sxqSejWlZu6LWsUQpJWsArUybNm04c+YMFy5cICAggFdeeYXQ0FCSkpLYtm0bLVu2JDQ0lL59+/7eNEZHR9OqVSuCg4Np1qwZd+7c4eeff6Zbt24A7Nq1i5CQEEJCQtDpdNy5c4cLFy4QeG/v+uzsbIYMGULjxo3R6XTs3LkTgKioKHr16kXnzp2pW7cub731ljZfFCEeUXIyTJgANWqYD2ePjYX//hcuXIDvvzcf4v6w91dcHFyIbBzJ9kHbOTfiHO+2fZfYq7H0XdUX72nejN46mhNXT1jw1YgS5pCiKAZFUdrfe5uHeS2gEOI+XF1dOX78OJMmTWLLli0EBATw/vvvF8nsKUvyKufKl881YeGQMHILTETOO8DoFTFcz9D8CFAhCqXEbWk0ahTExBTtNUNCYMaMh3tsfn4+mzdvpnPnzgCcPn2ahQsXMmfOHK5fv86kSZPYvn07bm5uTJ48mWnTpjF27FgiIiJYsWIFYWFhpKen4+rq+qfrfvbZZ8yePZvWrVuTkZGBy/+cVD179mwAjh8/TlxcHE8++STx8fEAxMTEYDQacXZ2pn79+rz++uv4+vo+4ldFCMszmWD7dvNo3/r1UFAATz0FX3wB3bqBQxH8BKvpWZOJj0/kg3YfsO3sNgxGA18c/ILpB6bTwqcFep2eiEYRlHUu++hPJkqql4FXgRGAAuzGvBZQCPE3XFxcGD9+PIMHD+att95iypQpDBkyxCaOpHq8fhV+fKMds3ee4atdZ9l+6gpvdW7As8385OxAYRNkBLCIZGVlERISQtOmTfHz80Ov1wPg7+9PixYtADhw4AAnT56kdevWhISE8M0333Dx4kVOnz6Nl5cXYWFhAHh4eODwP7/Ztm7dmtGjRzNr1ixu3779l8/v3buXgQMHAtCgQQP8/f1/bwA7dOhAuXLlcHFxoWHDhly8eNGiXwshHtW1a/Dpp1C3rrnh27MH3nwTzpyBLVugZ8+iaf7+yN7Oni51u7C632qSRycz9cmppOek88KGF/Ca6sXQdUPZl7hPdoETf6Gqao6qqtNUVe2lquozqqpOV1VVhgSEeAg+Pj4sXbqUhIQEatasiaqqvPPOOxw9elTraP/IxdGeMU/WZ/PItjSqXo5318byzJe/EJucpnU0IR6oxI0APuxIXVFzdXUl5j5Dj25ubr//XVVVOnXqxLJly/70mGPHjj3wKIaxY8fStWtXNm3aRIsWLdi+ffufRgH/6ZdSZ2fn3/9ub29Pfn7+A1+PEMVNVWHvXvNo3+rVkJsLbdvCpEnQqxf84b+xxVV2q8zolqN5o8Ub/Jr8K/OPzGfFiRUsjFlIg0oN0Ov0DAoeRBW3KsUXSlgdRVFWqqraT1GU48BffghrsQZQCFvl4+MDQHJyMnPnzmXy5MkMHz6cDz/8kIoVrXedXZ0q7ix9oTnrYlKYtPEk3b/Yy+BWNRjdqR5lXRy1jifEfckIYDFq0aIF+/bt48yZMwBkZmYSHx9PgwYNSElJITo6GoA7d+78pUk7e/YsjRs35u2336Zp06bExcX96fNt27ZlyZIlAMTHx5OYmEj9+vWL4VUJ8WjS0uDzzyEw0NzwbdwIL70EJ07Arl0QGVm8zd8fKYpCC58WzO8+n9QxqRi6G6jgWoH//PgfvKd502tFLzbGbyTfJDdVSqmR9/7sBjx9nzchRCH5+PgQHx/Pq6++yty5c6lbty5z5syx6pvXiqLQU+fNjtHteba5H1G/XKDjtF38cCxFZo0IqyQNYDGqXLkyUVFRREZGEhQURIsWLYiLi8PJyYkVK1bw+uuvExwcTKdOncjOzv7Tv50xYwaBgYEEBwfj6upKly5d/vT5V155hYKCAho3bkxERARRUVF/GvkTwtocOgTDhkH16jBiBLi5gcEAKSkwcyY0bKh1wj9zd3JnqG4o+4bu4+QrJxnVfBR7E/fSbVk3/Gf4M37HeM7ePKt1TFGMVFVNvffX60CSqqoXAWcgGEjRLJgQNq5ChQrMmjULo9FISEgIEydOtPoNYgDKlXFkUs/GfP9Kayq5O/PaUiODF0Zz8cZdraMJ8SdyDIQA5Gsoisfdu7B8OXz5JRw+DGXKmM/ve+klaNJE63SFl1uQyw/xP2AwGthyZgsm1cTjNR5Hr9PTK6AXro6uD76I+Ec2cgzEYaAN4AkcAA4BmaqqDijuLHIMhChpVFXl0qVL+Pr6kpeXx9ixYxk5ciR+fn5aR/tH+QUmFh+4yNRt8eQWmHi1fR1eal8LZwd7raOJEkKOgRBCWLXYWHj9dfNo37BhkJ1t3skzJQXmzbPN5g/Ayd6JXgG92PjsRi6OusikxydxMe0iz33/HNWnVefVja9yJPWI1jGF5SmqqmYCvYDPVVV9BrCyMWwhbJOiKL/vXG40GpkzZw4NGjRg4sSJZGVlaZzu7znY2zGkdU12jGnHkw2rMn17PF1m7GHfmetaRxNCGkAhhGXk5MDSpdCmDTRuDHPnmg9s37sXjh+HV1+FcuW0Tll0fDx8GN92PAmvJ7Bj0A7C64ZjMBpoMrcJuq91zD44m1tZt7SOKSxDURSlJTAA2HjvYyVukzUhtNasWTPi4uLo1q0bH3zwAQEBAXz33XdWvc6uqocLXzwbyqKhzShQVQbM/5URy4xcvZP94H8shIVIAyiEKFJnzsBbb4GPDwwYAKmpMGWK+SD3b7+F1q3hAZve2jQ7xY4naj7Bkl5LSB2TyhddvkBB4bXNr+E11YsBawbw0/mfMKkmraOKojMKGAd8r6rqCUVRagE7Nc4kRInk7+/PypUr2blzJx4eHkyYMIGCggKtYz1Q23qV2TqqLSM71GVL7GU6TN3Fov0XKDBZb/MqSi5ZAygA+RqKR5OfDxs2mI9w2LYN7O2hRw94+WV44gmwk1tNHEk9guGIgSXHl5CWk0Ytz1oMDRnK8yHP4+3hrXU8q2ULawCtiawBFKVJfn4+qamp+Pr6cvv2bT755BPGjh2Lp6en1tH+0blrGby/7gR7z1wnyKcck3oGEuRTXutYwsbIGkAhhCaSkuCDD8Df33xW38mTMHEiJCbCd99Bx47S/P0m1CuU2V1nkzomlW+f+Rb/cv68u/Nd/Gb40XVpV9acWkNuQa7WMUUhKIoy496fGxRFWf+/b1rnE6Kkc3Bw+H194Pbt2/nss8+oW7cuX3/9tVWPCtaq7M5ifTNmRepITcumx+x9fLAulvTsPK2jiVJCRgAFIF9D8fBMJvMo31dfmUf9VBU6dzbv5BkeDg6y8umhnb15loUxC1kYs5CUOylULlOZQcGD0Ov0BFSW70ew7hFARVGaqKp6WFGUdvf7vKqqu4o7k4wAitLs6NGjjBgxgt27dxMSEsLnn3/OY489pnWsf5SencfUradZfOAiFd2debdrAN2Dq6OU5LUSokjICKAVsLe3JyQkhMDAQJ5++mlu375dpNePioritddeA+C///0vn332WZFeX4gHuXoV/u//oE4d6NIF9u+Ht9+Gs2dh0ybo3l2av8KqXaE2k56YxMVRF/kh8gce83uMmb/OpOGchrRe0JoFxgVk5GZoHVP8DVVVD9/76yFgj6qqu+41fXuBaO2SCVE6BQcH8/PPP7N8+XKuX7/O5MmTtY70QB4ujkzoEci6Vx/Dq5wLI5fHMNBwkHPX5Ge/sBxpAIuIq6srMTExxMbGUqFCBWbPnq11JCEemarCrl0QGWne1GXcOPN0z+XLzdM/P/4YatbUOqXtc7BzoGu9rqyJWMOlNy4xpdMUbmbdRL9ej9dUL4atH8b+pP1WvdNdKbcDKPOH912B7RplEaJUUxSFiIgI4uLimDdvHgBnz57l448/JjvbenfebOxTju9fac3EHo04mnSbzjP2MO3HeLLzrHcqq7Bd0gBaQMuWLUlOTv79/SlTphAWFkZQUBAffPDB7x9ftGgRQUFBBAcHM3DgQAA2bNhA8+bN0el0dOzYkStXrhR7fiFu3YJZs6BRI2jfHrZsMR/bcOoU7NwJERHg5KR1ypKpqntV3mz1JidfOcm+ofvo27Avy2OX02pBKwK/DGTa/mlcu3tN65jiz1xUVf39dv29v5f5h8cLISzMzc2NatWqAbBmzRrGjx9Po0aNWLdundXeTLO3UxjUsgY73mxHl8bVmLUjgc4zdrM7Xn7mi6JV4iZsjdoyipjLMUV6zZBqIczoPOOhHltQUMCOHTvQ6/UAbNu2jYSEBA4ePIiqqnTv3p3du3dTsWJFPvroI/bt20elSpW4efMmAI899hgHDhxAURTmz5/Pp59+ytSpU4v09QhxP6oK0dHmtX3Ll0NWFjRvDgsXQr9+UEZ+nS1WiqLQyrcVrXxbMbPzTFacWIHBaGDMtjGM3T6W7vW7o9fpebL2k9jb2Wsdt7S7qyhKqKqqR8C8NhCw3hOqhShl/vOf/6DT6Rg5ciQ9e/akU6dOzJw502r3PqhS1oWZ/XX0a+rLe2tjGbTgIF2DvHi/W0OqerhoHU+UACWuAdRKVlYWISEhXLhwgSZNmtCpUyfA3ABu27YNnU4HQEZGBgkJCRw9epQ+ffpQqVIlACpUqADApUuXiIiIIDU1ldzcXGrK/DphYRkZsGwZfPklGI3g5gYDB5o3dbn331ZorKzz/2PvvsOjKt42jn8nIRB67x0hhNCrICUgRXpRQFBAmgIiHRuo+EMFFAFBFASDFJEqKiKIgILSewu9CkSldwgJmfePXXkRUBLI5myS+3NdudzdnOzeGUMmz54586Smc5nOdC7TmdCToUzaMomp26fy9e6vyZUmF+1Ltqdj6Y7kT6/fFw7pDcwxxoS572cHnnYwj4jcoVatWmzdupVx48bx1ltvMW7cOMaMGeN0rP9UuWAmFvWuymcrDjH2lwOs2HuKfnUCaFsxL0l8tYhPHlyCKwCje6Yutv19DeCFCxdo2LAhn3zyCT179sRay+uvv06XLl3+cfyYMWPuucNTjx496Nu3L40bN2b58uW8/coHHd8AACAASURBVPbbcfQdSGKzY4frbN+0aXDpEhQvDp9+6mreniaN0+nk3xTNUpQRT4xgaK2hzN87n5AtIbz323u8+9u71Mxfk06lO9GsSDP8k+hd4rhird1gjAkECgMG2GOt1X7uIl7Gz8+Pnj170rp1a/z8/ABYtWoVe/bsoUOHDvh4Yd+iZEl86VmzEE1K5eCt70L53/e7mLvpOO81K06p3OodKA/G+37S47m0adMyZswYPvzwQyIiInjiiSeYNGkSly+7Lg85ceIEJ0+epGbNmsyePZszZ84A3FoCeuHCBXLmdDWFnjJlijPfhCRY16/Dl19ClSpQogSEhEDTprB6NWzb5mrcruIvfkjqm5TmQc1Z9OwijvY+yuDqgzl47iDPzHuGHCNy0GNhj1hfDi/3ZoxJAbwK9LLW7gDyGWMaOhxLRP5F5syZSZfOVTxNnTqVzp078+ijj7JmzRqHk/27vBlTMrlDeT55pgynL4fT7NNVDPxmBxeu6r0miTkVgB5QunRpSpYsycyZM6lTpw7PPPMMlSpVonjx4jRv3pxLly5RtGhRBg4cSHBwMCVLlqRv376Aq8VDixYtqFq16q3loSIPa/9+6N8fcuZ0Le88eRJGjIATJ2DqVKhUCdRyKP7KnTY3bwa/ycGeB1nadil1C9Zl4uaJlP6sNGUnlGXchnGcvx67rWnkH74AbgCV3PePA+86F0dEomv8+PFMnz6dsLAwHnvsMdq1a0dYWNj9v9ABxhgalMjO0r7BtH8sHzPW/07Nkcv5Zstxr93YRryTGsELoDFMiCIiYP581zLPpUtdPfqaNnVd21ejBnjhSheJRWevnWX69umEbAlh21/b8E/iT/Og5nQq3YngvMHxpsmwNzeC/5sxZqO1tpwxZou1trT7sW3W2pJxnUWN4EUezOXLlxkyZAgjRozg/fffp3fv3k5Huq+dJy7wxrc72XrsPJUKZOSdpsUomCWV07EkjjzM/KgCUACNYULy++8wcSJ8/jn8+SfkyQMvvAAdO0L27E6nk7hmrWXzH5v5fPPnfLXzKy6GX+SR9I/QsXRH2pdqT47UOZyO+J/iSQG4GqgJrLLWljHGPALMsNZWiOssKgBFHs6hQ4fInTs3fn5+fPfdd/j6+tKgQQOvfdMsKsoyY8PvvL9oD9cibtKl2iO89HhB/P20O3RC9zDzo84BiCQAN2/CwoXQuLGrMft770HZsvD993DoEAwcqOIvsTLGUDZHWcY1HMcf/f5gatOp5EqTi4E/DyT3qNw0mtGIb/d8S8RNXUfyEAYBPwK5jTHTcTWGf8XZSCLyIAoUKHBrg5gxY8bQqFEjGjRowN69ex1Odm8+PoZnH83Lsn7VaVQiB2N/OUDtUSv4Zc9Jp6OJF0swBWB8O5PpTTR28ddff8HQoVCwIDRoAOvXw+uvw+HDsGABNGwIvnoTUNxS+KWgbcm2LG+/nP099vNq5VfZFLaJZrOakXtUbl5Z8gp7T3vnHzneyrhOC+wBngTaAzOActba5Q7GEpFYsGjRIkaMGMGqVasoVqwYL7/8MhcvXnQ61j1lTp2MkU+XYsbzFUnq60OHyRvo9uUm/riglqRytwSxBPTw4cOkTp2ajBkzeu0pem9lreXMmTNcunRJPQfjCWth+XLXtX3z5kFkJDz+uOvaviZNIGlSpxNKfBIZFcmi/YsI2RLCgn0LuGlvUiVPFTqV7kSLoBakTJrS0XzxZAnoJmttWadzgJaAinjCX3/9xYABA5g0aRLffvstTZo0cTrSf7oRGcXE3w4xZtl+kvgY+tQOoP1j+dQ7MIFJ9NcARkREcPz4ca5fv+5QqvjN39+fXLly3VryIN7p7FnXjp3jx8PevZA+PbRvD126QOHCTqeThODPy38yddtUQraEsO/MPlInTU2rYq3oVLoTFXJWcOQNtnhSAH4CTLbWbnA6iwpAEc/ZtWsXRYoUwRjDlClTCAoKonz58k7H+lfHzl7lre928sveUxTJnoZ3mxajbN70TseSWJLoC0CRhMpaWLfOVfTNmuXq41epkutsX4sWkDy50wklIbLWsurYKkK2hDA7dDZXI65SNHNROpfpTJsSbciUIu5a1MSTAnAXribwR4AruJrBW2ttibjOojlSxPMiIiIIDAzk0KFDdOjQgSFDhpAtWzanY92TtZbFoX/yv+938ceF67SukJtX6waSLoWWC8V3KgBFEphLl+Crr1yF39atkCoVtGnjKvxKxvnG8pKYXQy/yMydMwnZEsL6E+vx8/GjaWBTOpXuRK0CtfD18exFpvGkAMx7r8ettUfjOovmSJG4cenSJd59911GjRqFv78/gwYNokePHiT10uswroRH8tHSfUxadYS0yf0YUL8IT5XJqUun4jGvLQCNMXWB0YAv8Lm1dti/HNccmAOUt9b+58ylyU0Ssm3bXEXfl1/C5cuuYq9bN3jmGUid2ul0ktjtPLmTkM0hTNs+jTPXzpA7TW46lOpAh9IdyJcun0de05sLQGOMP9AVKAjsAEKstZHR/NpYnx9Bc6RIXNu3bx99+vRh0aJFbNmyhZJe/i7t7j8uMvCbHWz+/TwV8mfgvabFKJRVf2DER15ZABpjfIF9QG3gOLABaG2t3XXHcamBH4CkwEsqACWxuXYN5sxxFX5r1oC/Pzz9tOts36OPgt6cE28THhnOd3u/I2RLCEsOLgGgVoFadCrdiaaBTUmWJFmsvZaXF4CzgAjgN6AecNRa2ysaX+eR+RE0R4o4ZefOnRQrVgyAsWPHUrduXQoWLOhwqnuLirLM3niMYT/u4fL1SJ6vVoCejxcieVJtGx6feGsfwArAAWvtIWvtDWAmcK9tk94BPgC0g4skKvv2Qd++kDMnPPeca5OXUaPgxAmYPBkqVlTxJ94pWZJktCzaksVtFnO412EGBQ9i35l9tPq6FTlG5qDXol5s/2u70zHjQpC1to219jOgOVA1ml+n+VEkgfm7+Dt16hQDBw6kaNGivP7661y+fNnhZHfz8TG0qpCHZX2DaVo6J+OWH6TWyBUs3fWX09EkjniyAMwJHLvt/nH3Y7cYY0oDua21C/7riYwxLxhjNhpjNp46dSr2k4rEkRs3XGf7atZ07dz58cdQuzb8/DPs3g29e0OGDE6nFIm+vOnyMqj6IA71OsRPbX6idoHajN80nnITynHu2jmn43laxN83orv00y3W5kf3sZojRbxE5syZ2bNnD61atWLYsGEULlyY6dOne2XP5YypkvFhi5LM7lKJlMl86Tx1Iy9M3ciJ8+odmNB5sgC817mLWz/9xhgfYBTQ735PZK2dYK0tZ60tlzlz5liMKBI3jh6FN96APHmgZUs4eBCGDIFjx1y7e9aoobN9Er/5GB9qP1Kbmc1nEtY3jG+e/ob0yRP8duMljTEX3R+XgBJ/3zbG/Fe36FibH0FzpIi3yZ49O1OmTGHNmjXkyJGDLl26cPLkSadj/asK+TPwQ8+qvFYvkN/2n6bWiBV8tuIgETejnI4mHuLJAvA4kPu2+7mAsNvupwaKAcuNMUeAisB8Y4xXXushElM3b8KCBdCwIeTPD0OHQoUK8MMPrgLw9dfBS3eNFnkoGVNkpEFAA6djeJy11tdam8b9kdpam+S222n+40s1P4okAhUrVmTdunWsWbOGrFmzYq1l+PDhXlkM+vn60DX4EZb0rUblgpkYumgPDcesZMORs05HEw/wZAG4AShkjMlvjEkKtALm//1Ja+0Fa20ma20+a20+YC3QODoXuYt4sz//hPfegwIFoFEj2LQJBg6Ew4dh/nyoXx98dZ21SGKm+VEkkfDx8aF48eIAbN++nQEDBhAQEMDo0aOJiIi4z1fHvVzpU/D5c+WY0LYsl8MjaTF+Da/M3cbZKzecjiaxyGMFoPt6iJeAxcBuYLa1NtQYM9gY09hTryviBGth2TJXc/bcuV3LPQMCYO5c+P13eOcd1/JPERHNjyKJU8mSJdmxYwcVK1akd+/elCpViqVLlzod657qFM3Gkr7V6BJcgHmbT1BzxHJmbfidqCjvu5ZRYk6N4EUewpkzMGUKfPaZa1fPDBmgQwfo0gUKFXI6nUj8581tILyR5kgR72etZcGCBfTu3ZsbN25w4MABkiWLvfY5sW3vn5d449sdbDhyjnJ50/Nus2IEZvuvVe4SF7y1DYRIgmQtrF4N7dq5Wjj06weZM8O0aa4WDh9+qOJPRERE7s0YQ6NGjQgNDeXHH38kWbJkhIeHM3z4cK5cueJ0vLsUzpaaWS9U4oPmJTh46jINxqxkyMLdXAmPyebH4k1UAIpE08WLMG4clCoFlSvDt99Cp06wfTusXAlt2riauIuIiIjcj7+/P0WLFgVg8eLFvPLKKwQGBjJr1iyvaxvh42NoWS43P/erTouyuZjw6yFqj1zB4tA/vS6r3J8KQJH72LrVtaQzRw548UXXBi4TJkBYGHzyCbiv7RYRERF5II0bN2blypVkyZKFVq1aUb16dbZt2+Z0rLukT5mUYU+VYG7XSqRJ7keXaZvoPGUjx85edTqaxIAKQJF7uHoVJk+GihWhdGnX8s6WLWHdOteuns8/D6lSOZ1SREREEorKlSuzfv16JkyYwK5du+jatavXnl0rly8D3/eowsD6RVhz6Ay1R63g0+UHuBGp3oHxgQpAkdvs2QN9+riu7evQwbXsc/Ro17V9kya5+vipYbuIiIh4gq+vL88//zz79u1j2rRpGGM4deoU48aNIzLSu6658/P14flqBVjaN5jggMx88ONeGoz5jbWHzjgdTe5DBaAkejduwKxZUKMGFCniWtZZty4sXw6hodCzJ6RP73RKERERSSzSp09PwYIFAfjyyy958cUXKVOmDMuXL3c22D3kSJecz9qWI+S5clyLuEmrCWvpO3srpy+HOx1N/oUKQEm0Dh+GAQNcfftatYKjR2HYMDh+HGbMgOBgne0TERERZ/Xu3Zuvv/6aS5cuUaNGDVq2bMnRo0edjnWXmkWysqRPMC9Wf4Tvt4VRc8QKvlqn3oHeSAWgJCo3b8L8+VC/PjzyCLz/PlSqBIsWwYED8OqrkCWL0ylFREREXIwxPPnkk+zatYvBgwezYMECBgwY4HSse0qe1JdX6gayqFdVArOlZsA3O2g+fjW7wi46HU1uo0bwkiiEhUFICEycCMeOQfbsro1cOnd2nQEUEe+kRvAxozlSJOE7duwYPj4+5MyZk927dxMaGspTTz2F8bJlS9Za5m0+wZCFuzl/LYL2j+WjT+0AUiVL4nS0BEGN4EXuISoKliyBp56CPHngrbdc1/jNm+da7vm//6n4ExERkfgld+7c5MyZE4CxY8fSokULatasyY4dOxxO9k/GGJ4qm4tl/YJ5unxuJq06TK0RK1i04w+v3d00sVABKAnO6dPw4YdQuDDUqQMrVkDfvrB/PyxeDM2agZ+f0ylFREREHs7o0aP59NNP2bZtG6VLl6ZHjx6cPXvW6Vj/kC5FUoY0K87X3R4jfcqkdJu+mQ6TN/D7GfUOdIoKQEkQrIVVq6BtW8iVC15+GbJlg+nTXS0cPvgA3JtpiYiIiCQISZIkoVu3buzfv5+uXbvy6aefMmrUKKdj3VOZPOn5/qXKvNkwiA2Hz1J71Ao+Xraf8MibTkdLdHQNoMRrFy7Al1/C+PGwcyekSQPt2kGXLlCsmNPpRORh6RrAmNEcKZK4bd++nfz585M6dWpWrVpFVFQUVatWdTrWXf68cJ3BC0JZuONPCmROybtNi/HYI5mcjhWv6BpASXQ2b4YXXnA1bH/pJUiWDD7/3LXZy8cfq/gTERGRxKdEiRKkTp0agMGDB1OtWjVat27NsWPHHE72T9nS+vPps2WZ3KE8kTctz0xcR++ZWzh1Sb0D44IKQIk3rl6FSZOgQgUoW9Z15q9VK9iwATZuhE6dIGVKp1OKiIiIOO+bb75h0KBBfPvttwQGBvLuu+9y/fp1p2P9Q/XCWfipTzV6PF6QH3b8weMjljNt7VFuqnegR6kAFK+3axf06gU5criKvCtXXGf5wsJcZ/3KaXGYiIiIyD+kSJGCt99+m927d1OvXj3efPNNZsyY4XSsu/j7+dKvTmF+7F2N4jnT8ua3O3ny01XsPHHB6WgJlgpA8Urh4TBjBgQHQ9Girmv8GjSAX391Xev30kuQLp3TKUVERES8W758+Zg7dy4rV66kXbt2ACxcuJBdu3Y5nOyfHsmciumdH2V0q1KcOH+NxmNX8vb8UC5dj3A6WoKjAlC8yqFD8Nprrv58zzwDx4/D+++7/jt9OlStCl7W51RERETE61WuXBlfX1+ioqLo3bs3JUqUoE+fPpw/f97paLcYY2hSKifL+lXn2UfzMmXNEWqOWMH328LUOzAWqQAUx0VGwrffQt268Mgjrh5+Vaq4evbt3w+vvAKZMzudUkRERCT+8/HxYfXq1XTu3JnRo0cTEBDA559/zs2b3tOOIW1yP95pWoxvX6xMljTJ6DFjC+0mrefI6StOR0sQVACKY06cgP/9D/LlczVn37kT3n4bjh6FefNcTdx99BMqIiIiEqsyZcrE+PHj2bRpE4ULF+b5559n5cqVTse6S8nc6fiuexXebhTElt/PU+ejX/lo6T6uR3hPsRof6c9riVNRUfDTT/Dkk5A3r6sALF7cdQbwyBEYNMjV2kFEREREPKt06dL8+uuv/PLLLwQHBwMwZ84cwsLCHE72/3x9DO0r52dZv2DqBGXlo6X7qTf6N37bf8rpaPGWCkCJE6dOwQcfQKFC8MQTsHIl9O8PBw7AokXQpAkkSeJ0ShEREZHExRhD9erVAbhw4QKdOnUiICCAYcOGER7uPX35sqbxZ+wzZZjWqQLWWtqGrKfHjC2cvOhdrS3iAxWA4jHWwm+/wbPPQq5c8Oqrrv/OmAHHjsGwYVCggNMpRURERAQgbdq0bN26ldq1a/P6669TtGhRvv/+e6/agKVqocz82LsavWsVYvHOP6k5YgVTVh9R78AYUAEose78eVefvmLFoFo1+OEH6NoVQkNhxQpX8/ZkyZxOKSIiIiJ3KlCgAN988w0//fQTSZMmpVmzZhw5csTpWP/g7+dL71oBLO5TjVJ50jFofihNP1nF9uPes6OpN1MBKLFm40bo3Nl1DV/PnpAqFUya5GrYPno0BAU5nVBEREREoqN27dps27aNpUuXkj9/fgCmTZvGxYsXHU72//JnSsnUjhX4uHVp/rp4nSafrOKt73Zy4Zp6B/4XFYDyUK5cgZAQKFcOypd3Le989llXMbhuHXToAClSOJ1SRERERGLKz8/v1vWB+/bt47nnniMgIIAvvviCqKgoZ8O5GWNoVDIHS/sF81ylfHy59ig1R6zgu60nvGrpqjdRASgPZOdO6NEDcuRwnfULD4dPPnGd7ZswAcqWdTqhiIiIiMSWgIAANmzYQIECBejYsSOVKlVi3bp1Tse6JY2/H283Lsp33auQM50/vWZupU3IOg6euux0NK+jAlCiLTwcpk+HqlVdrRsmTIDGjV07em7fDi++CGnTOp1SRERERDyhbNmyrFq1imnTpnHs2DHq1avH5cveVWAVz5WWeS9W5p0mRdl+/AL1PvqNkT/tVe/A26gAlPs6cABeecW1g2ebNvDnnzB8uKuR+7RpULkyGON0ShERERHxNGMMbdq0Ye/evXz//fekSpWKqKgoJk+ezI0bN5yOB7h6B7atlI9l/YKpXzwbY34+wBMf/cryvSedjuYVVADKPUVGwjffuHr2FSoEI0dCcDAsWQJ797p6+GXK5HRKEREREXFC6tSpqVy5MgBLly6lQ4cOFC9enEWLFjmc7P9lSe3PR61KM73zo/gaQ/svNtB9+mb+vJC4eweqAJR/OHYMBg2CvHnhySdh1y4YPBh+/x3mzoVatcBHPzUiIiIi4lanTh0WLlwIQP369WnYsCH79+93ONX/q1wwE4t6V6Vf7QCW7v6LmiOWM2nlYSJvesdGNnFNf8oLUVHw44/QtCnkywfvvAOlSsH8+XD4MLz5pmuzFxERERGRe6lXrx47duzgww8/5Ndff6VJkyZetQtnsiS+9KhZiJ/6VKNcvgwMXrCLxmNXseX3c05Hi3MqABOxkydh2DAoWBDq1YM1a+DVV+HgQVfz9kaNIEkSp1OKiIiISHyQNGlS+vXrx759+5g2bRrGGK5evcrMmTO9pm1E3owpmdyhPJ8+W4YzV8J5ctxqBnyzgwtXE0/vQBWAiYy1sGIFtG7t2tTl9dddZ/1mzXIt/xwyBNy9PkVEREREYixbtmyUdfcEmzx5Mq1bt6ZKlSps3LjR4WQuxhjqF8/Osn7V6Vg5PzPX/87jI5Yzb/Nxrzpr6SkqABOJc+dgzBgoWhSqV3ct+ezeHXbvhp9/hpYtIWlSp1OKiIiISELStWtXvvjiCw4dOkSFChXo3LkzJ096x26cqZIl4c2GQXzfowq5M6Sg7+xttJ64lgMnLzkdzaM8WgAaY+oaY/YaYw4YY167x+f7GmN2GWO2G2OWGWPyejJPYmMtrF8PHTtCzpzQqxekSQOTJ7sato8aBYGBTqcUEUl8ND+KSGLh4+ND+/bt2bdvH/369WPq1Kl07NjR6Vj/UDRHWuZ1e4whzYqzK+wi9Ub/xgc/7uHajYTZO9B46jSnMcYX2AfUBo4DG4DW1tpdtx1TA1hnrb1qjOkGVLfWPv1fz1uuXDnrLaePvdXly/DVVzB+PGzZAilTuvr3dekCpUs7nU5EJPqMMZusteWczhGbPDU/guZIEfF+e/fuxVpLYGAgYWFh7Ny5kzp16jgd65bTl8MZsnA38zafIFf65AxuUpTHA7M6HesuDzM/evIMYAXggLX2kLX2BjATaHL7AdbaX6y1V9131wK5PJgnwduxw7WsM0cOV7EXGQmffuo62zd+vIo/EREvoflRRBKtwoULE+hegjZixAieeOIJmjZtyqFDhxxO5pIpVTJGtizFjOcr4u/nS8fJG+k6bRNh5685HS3WeLIAzAkcu+3+cfdj/6YTcM/OkcaYF4wxG40xG0+dOhWLEeO/69fhyy+hShUoUQJCQqBZM1i9GrZtg27dXMs+RUTEa8Ta/AiaI0Uk/hoyZAhDhw5l6dKlBAUFMXDgQC5fvux0LAAqPZKRhT2r8vIThVm+7yS1Rq5g4q+HiEgAvQM9WQCaezx2z/Wmxpg2QDlg+L0+b62dYK0tZ60tlzlz5liMGH/t3w/9+7uu7Wvb1tXSYcQIOHECpkyBSpXA3Ov/gIiIOC3W5kfQHCki8VeyZMl47bXX2LdvHy1atGDIkCEMHDjQ6Vi3JE3iQ/caBVnSJ5iKBTLy3sLdNPp4JZuOnnU62kPxZJe340Du2+7nAsLuPMgYUwsYCARba8M9mCfei4hwNWcfPx6WLnX16GvWDLp2hRo1VPCJiMQTmh9FRG6TI0cOpk2bRrdu3cjv7kcWGhrKjRs3KO0F1zDlzpCCkOfKsTj0L/73fShPjVtDq/K5ebVuIOlTxr9t9D15BnADUMgYk98YkxRoBcy//QBjTGngM6CxtdY79oP1Qr//Dm++CXnyQPPmsG8fvPuu6/HZs+Hxx1X8iYjEI5ofRUTu4bHHHiN79uwAvPHGG5QtW5YuXbrgDcvbjTHULZaNpX2DeaFaAeZsOk7NkSuYvfFYvOsd6LEC0FobCbwELAZ2A7OttaHGmMHGmMbuw4YDqYA5xpitxpj5//J0ic7Nm7BwITRu7GrM/t57UK4cLFgAhw7BwIHg/vchIiLxiOZHEZH7++KLL+jduzeTJk0iICCAjz/+mMjISKdjkTJZEgbUL8KCHlXInyklr8zdztOfrWXfX/Gnd6DH2kB4SkLf4vqvv2DSJJgwAY4cgaxZoXNneP55yKsuUCKSyCTENhCelNDnSBFJfHbt2kXv3r1ZsmQJo0ePpmfPnk5HuiUqyjJn0zGGLtrD5euRdKqan141C5EiqSevsnN5mPnR8+nkvqyF5ctd1/bNm+dq31CzJgwfDk2agJ+f0wlFREREROJeUFAQixcv5ocffuDxxx8HYO3atWTLlo18+fI5ms3Hx/B0+TzUDsrG0IW7+WzFIRZs+4O3GxeldpD39Q78myevAZT7OHsWRo2CIkVc1/EtWQI9e8KePa5NXpo3V/EnIiIiIombMYaGDRuSIkUKrLW88MILFClShEGDBnH16tX7P4GHZUiZlOEtSjKnayVSJvPl+akb6TxlI8fPOZ/tXlQAxjFrYe1aaN/e1cKhb1/IkMHVuuHECVcrh8KFnU4pIiIiIuJ9jDEsXLiQZs2aMXjwYAIDA5k9e7ZXbMRSPl8GfuhZldfqBbLqwGlqj/yV8SsOel3vQBWAceTSJfjsMyhTxtWj7+uvoUMH2LrV1bS9XTtIntzplCIiIiIi3i1Xrlx89dVX/Prrr2TMmJGnn36ahQsXOh0LAD9fH7oGP8KSvtWoUigTwxbtocGY31h/2Ht6B6oA9LBt26BbN8iRw9Wvz1rXtX5hYfDpp1CypNMJRURERETin6pVq7Jx40Zmz55NvXr1APj55585c+aMw8kgV/oUTGxXjontynEl/CYtP1vDy3O2cfbKDaejqQD0hGvXYOpUeOwxKFUKJk92Xc+3di1s2QJdukDq1E6nFBERERGJ33x9fWnRogU+Pj5cu3aNli1bEhAQwLhx47h586bT8agdlJUlfavRNfgRvtlygsdHLGfm+t+JinJuyaoKwFi0d6/rmr6cOeG55/5/k5ewMPjiC3j0UTVsFxERERHxhOTJk7N8+XJKlSrFiy++SJkyZVixYoXTsUiRNAmv1QtkYa+qBGRJzWvzdtDiszXs/uOiI3lUAD6kGzdgzhxX24bAQBg7FurUgV9+gd27oXdvSJ/e6ZQiIiIiIglfsWLFWLp0KXPnzuX8+fNUr16dXbt2OR0LgICsqZnVpSLDvJK74QAAIABJREFUm5fg8OkrNPx4Je/9sIsr4XHb4F4F4AM6ehTeeAPy5IGWLeHgQRgyBI4dg5kzoXp1ne0TEREREYlrxhieeuop9uzZw5w5cwgKCgLgxx9/5Nq1a45na1EuN8v6BtOibC4m/naYWiNX8OPOP+NsJ1MVgDFw8yYsWAANG0L+/DB0qGtZ58KFrgLw9dchq/f2fBQRERERSTSSJ09O8+bNATh27BgNGzYkKCiIefPmOd42In3KpAx7qgRfd6tE2uR+dP1yE52mbOTYWc/3DlQBGA1//AHvvgsFCkCjRrB5s+vs3+HD8N13UK8e+Po6nVJERERERO4ld+7cLF26lNSpU/PUU09Ru3ZtQkNDnY5F2bwZWNCjCm80KMLaQ2eoPWoFn/xygBuRnusdqALwX1gLy5ZBixauZZ5vvgkBAa7+fUePwuDBrsdFRERERMT7Va9enc2bNzN27Fg2b95MhQoVOHfunNOxSOLrQ+eqBVjaN5jqAVkYvngv9cf8xpqDnmlnoQLwDmfOwMiRrg1datVybebSuzfs2wdLlsCTT4Kfn9MpRUREREQkppIkSUL37t3Zv38/X331FenduzUuXLjQ8bYROdIlZ3zbsnzRvjzhkTdpPXEtfWdt5fTl8Fh9HRWAuM72rV4N7dq5Wjj06weZM8O0aXD8OAwfDoUKOZ1SRERERERiQ8aMGWnSpAkAq1evpkGDBpQvX56VK1c6nAxqBGbhp97BdK/xCN9vD+PxD5czfd3RWOsdmKgLwIsXYdw4V7P2ypXh22+hc2fYvh1WroQ2bcDf3+mUIiIiIiLiKZUqVWLmzJmcOnWKqlWr8uyzz3LixAlHMyVP6svLTwSyqFdVgnKkYeA3O3ly3GpCwy489HMnygJw61bo0gVy5IAXX3Rt4DJhgqth+9ixULy40wlFRERERCQuGGN4+umn2bNnD2+++SZff/01VapUcXxJKEDBLKmZ8XxFRrYsybGzV2n08UoGf/9wfQ2TxFK2eOOTT+CllyB5cmjdGrp2hXLl1LNPRERERCQxS5kyJYMHD6ZDhw4cOHAAX19fbt68ybJly6hduzbGoYLBGMOTZXJRMzArHyzewxerDz/U8yW6M4CNGsHo0XDiBISEQPnyKv5ERERERMQlf/781K5dG4CvvvqKJ554grp167J7925Hc6VN4cd7zYozr9tjD/U8ia4AzJMHevYE94Y/IiIiIiIi99SqVSs++ugj1q1bR4kSJejbty8XLjz8dXgPo3SehytkEl0BKCIiIiIiEh1+fn706tWLffv20b59ez766COaNm3qdKyHkuiuARQREREREYmJLFmyMHHiRLp27UpERAQAFy9eZNeuXVSsWNHhdDGjM4AiIiIiIiLRULZs2VsF38iRI6lUqRLt2rXjjz/+cDhZ9KkAFBERERERiaH+/fvz+uuvM2vWLAICAvjggw8IDw93OtZ9qQAUERERERGJoVSpUjFkyBBCQ0N5/PHHefXVV+nevbvTse5L1wCKiIiIiIg8oIIFC/Ldd9+xePFi8ubNC8Dx48e5evUqAQEBDqe7m84AioiIiIiIPKQnnniCwMBAAAYMGECxYsV45ZVXuHjxosPJ/kkFoIiIiIiISCwaPnw4bdu2Zfjw4RQuXJgpU6YQFRXldCxABaCIiIiIiEisypo1KyEhIaxbt468efPSvn17PvjgA6djAboGUERERERExCMqVKjA6tWr+fLLL6lfvz4Ae/bsIX369GTNmtWRTDoDKCIiIiIi4iE+Pj60a9eOTJkyAdC5c2cCAgIYMWIEN27ciPs8cf6KIiIiIiIiidSkSZOoXLky/fv3p0SJEvz4449x+voqAEVEREREROJIQEAACxcuZMGCBURFRVGvXj3mzJkTZ6+vAlBERERERCSONWjQgB07djB27FgaN24MwNatW7l8+bJHX1cFoIiIiIiIiAOSJUtG9+7dSZYsGRERETRu3JjChQszffp0rLUeeU0VgCIiIiIiIg7z8/Nj1qxZZM+enTZt2lClShU2bdoU66+jAlBERERERMQLVKpUifXr1xMSEsL+/fspX748GzdujNXX8GgBaIypa4zZa4w5YIx57R6fT2aMmeX+/DpjTD5P5hEREfEGmh9FROTf+Pj40LFjR/bv38+YMWMoW7YsABs2bCAiIuLhn/+hn+FfGGN8gU+AekAQ0NoYE3THYZ2Ac9bagsAo4H1P5REREfEGmh9FRCQ60qZNy0svvYQxhlOnTlGjRg1KlSrFkiVLHup5PXkGsAJwwFp7yFp7A5gJNLnjmCbAFPftuUBNY4zxYCYRERGnaX4UEZEYyZQpE9OnT+f69evUqVPnoZ4rSSxlupecwLHb7h8HHv23Y6y1kcaYC0BG4PTtBxljXgBecN+9bIzZ+5DZMt35GgJoXO5FY3I3jcndNCZ3i60xyRsLz+FtYm1+hLvmyHBjzM5YT5xw6d9uzGi8YkbjFTMar5gp/KBf6MkC8F7vVN65l2l0jsFaOwGYEBuhAIwxG6215WLr+RIKjcvdNCZ305jcTWNyN43Jf4q1+RH+OUdq3GNG4xUzGq+Y0XjFjMYrZowxD7wzjCeXgB4Hct92PxcQ9m/HGGOSAGmBsx7MJCIi4jTNjyIi4hhPFoAbgELGmPzGmKRAK2D+HcfMB55z324O/Gw91fFQRETEO2h+FBERx3hsCaj7moWXgMWALzDJWhtqjBkMbLTWzgdCgGnGmAO43tls5ak8d4i15aQJjMblbhqTu2lM7qYxuZvG5F94eH7UuMeMxitmNF4xo/GKGY1XzDzweBm9oSgiIiIiIpI4eLQRvIiIiIiIiHgPFYAiIiIiIiKJRIIuAI0xdY0xe40xB4wxr93j88mMMbPcn19njMkX9ynjVjTGpK8xZpcxZrsxZpkxJiH24PqH+43Jbcc1N8ZYY0yi2KI4OuNijGnp/nkJNcZ8FdcZ41o0/v3kMcb8YozZ4v43VN+JnHHFGDPJGHPy3/rOGZcx7vHabowpE9cZEyLNbTGjeS/mNC/GjObLmNFcGn0em2ettQnyA9eF9QeBAkBSYBsQdMcxLwLj3bdbAbOczu0FY1IDSOG+3U1jcuu41MCvwFqgnNO5vWFcgELAFiC9+34Wp3N7wZhMALq5bwcBR5zO7eExqQaUAXb+y+frA4tw9bSrCKxzOnN8/9Dc5pHxSlTzXmyMmfu4RDUvPsx4Jbb5MhbGK1HNpfcZL4/Mswn5DGAF4IC19pC19gYwE2hyxzFNgCnu23OBmsaYezXfTSjuOybW2l+stVfdd9fi6k+VkEXn5wTgHeAD4HpchnNQdMbleeATa+05AGvtyTjOGNeiMyYWSOO+nZa7e7slKNbaX/nv3nRNgKnWZS2QzhiTPW7SJVia22JG817MaV6MGc2XMaO5NAY8Nc8m5AIwJ3DstvvH3Y/d8xhrbSRwAcgYJ+mcEZ0xuV0nXO8qJGT3HRNjTGkgt7V2QVwGc1h0flYCgABjzCpjzFpjTN04S+eM6IzJ20AbY8xxYCHQI26iea2Y/s6R+9PcFjOa92JO82LMaL6MGc2lseuB5lmP9QH0Avd6t/POnhfROSYhifb3a4xpA5QDgj2ayHn/OSbGGB9gFNA+rgJ5iej8rCTBtaylOq53zH8zxhSz1p73cDanRGdMWgOTrbUjjDGVcPVxK2atjfJ8PK+U2H7HxgXNbTGjeS/mNC/GjObLmNFcGrse6Pd9Qj4DeBzIfdv9XNx9CvnWMcaYJLhOM//Xadb4LjpjgjGmFjAQaGytDY+jbE6535ikBooBy40xR3Ctr56fCC54j+6/n++stRHW2sPAXlwTXEIVnTHpBMwGsNauAfyBTHGSzjtF63eOxIjmtpjRvBdzmhdjRvNlzGgujV0PNM8m5AJwA1DIGJPfGJMU14Xw8+84Zj7wnPt2c+Bn676iMoG675i4l3V8hmsSTAxr1P9zTKy1F6y1may1+ay1+XBdH9LYWrvRmbhxJjr/fr7FtXkCxphMuJa4HIrTlHErOmPyO1ATwBhTBNekdSpOU3qX+UA79y5lFYEL1to/nA4Vz2luixnNezGneTFmNF/GjObS2PVA82yCXQJqrY00xrwELMa149Aka22oMWYwsNFaOx8IwXVa+QCud0dbOZfY86I5JsOBVMAc954Bv1trGzsW2sOiOSaJTjTHZTFQxxizC7gJvGytPeNcas+K5pj0AyYaY/rgWoLRPiH/4W2MmYFrSVMm97UagwA/AGvteFzXbtQHDgBXgQ7OJE04NLfFjOa9mNO8GDOaL2NGc2nMeGqeNYl0PEVERERERBKdhLwEVERERERERG6jAlBERERERCSRUAEoIiIiIiKSSKgAFBERERERSSRUAIqIiIiIiCQSKgBFHpIxJqMxZqv7409jzAn37fPuLZ9j+/WqG2MWxPBrlt+rSa8xpr0xZmzspRMREXkwxpib7vlzpzHme2NMulh+/ltznjHmbWNM/9h8fpH4QgWgyEOy1p6x1pay1pYCxgOj3LdLAVH3+3pjTILtxykiIhID19zzaTFcPSy7Ox1IJCFSASjiWb7GmInGmFBjzE/GmORw64zcEGPMCqCXMSazMeZrY8wG90dl93HBt51d3GKMSe1+3lTGmLnGmD3GmOnG3b3YGFPTfdwOY8wkY0yyOwMZYzoYY/a5X7tyHI2DiIhITKwBcv59xxjzsnt+3G6M+d9tj7dzP7bNGDPN/VgjY8w693y41BiT1YH8Il5LBaCIZxUCPrHWFgXOA0/d9rl01tpga+0IYDSuM4fl3cd87j6mP9DdfUaxKnDN/XhpoDcQBBQAKhtj/IHJwNPW2uJAEqDb7WGMMdmB/+Eq/Gq7v15ERMRrGGN8gZrAfPf9Orjm0wq4VteUNcZUM8YUBQYCj1trSwK93E+xEqhorS0NzAReieNvQcSraemZiGcdttZudd/eBOS77XOzbrtdCwhyn8gDSOM+27cKGGmMmQ7Ms9Yedx+z3lp7HMAYs9X9vJfcr7fP/RxTcC2f+ei213kUWG6tPeX+2llAQCx8nyIiIg8r+W1z2iZgifvxOu6PLe77qXAVhCWBudba0wDW2rPuz+cCZrnf9EwKHI6T9CLxhM4AinhW+G23b/LPN12u3HbbB6j097WE1tqc1tpL1tphQGcgObDWGBP4H89riB4bo+9AREQkblxzr3jJi6tw+/saQAMMvW2OLGitDXE/fq857WNgrHs1TBfAPw6yi8QbKgBFvMNPwEt/3zHGlHL/9xFr7Q5r7fvARiDwX74eYA+QzxhT0H2/LbDijmPWAdXdO5f6AS1i6xsQERGJDdbaC0BPoL97rloMdDTGpAIwxuQ0xmQBlgEtjTEZ3Y9ncD9FWuCE+/ZzcRpeJB5QASjiHXoC5dwXsu8Curof7+3eDnsbruv/Fv3bE1hrrwMdgDnGmB24diAdf8cxfwBv47q4fimwOba/ERERkYdlrd0CbANaWWt/Ar4C1rjnt7lAamttKPAesMI9T450f/nbuObC34DTcR5exMsZa7UaTEREREREJDHQGUAREREREZFEQgWgiIiIiIhIIqECUEREREREJJFQASgiIiIiIpJIqAAUERERERFJJFQAioiIiIiIJBIqAEVERERERBIJFYAiIiIiIiKJhApAERERERGRREIFoIiIiIiISCLhsQLQGDPJGHPSGLPzXz5vjDFjjDEHjDHbjTFlPJVFREREREREPHsGcDJQ9z8+Xw8o5P54ARjnwSwiIiIiIiKJnscKQGvtr8DZ/zikCTDVuqwF0hljsnsqj4iIiIiISGKXxMHXzgkcu+3+cfdjf9x5oDHmBVxnCUmZMmXZwMDAOAko4qSbUZbL4ZFcDo/k0vVIbkZZjHE6lYjnRYVf5caFU/gmTU7k1QunrbWZnc4kIiKSUDhZAN7rT1l7rwOttROACQDlypWzGzdu9GQuEUfcjLJsO36eX/ed4rf9p9n75yUey5ueqoUyERyQmYJZUmFUAUoC99prr/H+++9ToEABPvroIxo3bnzU6UwiIiIJiZMF4HEg9233cwFhDmWRB3QjMopFO//gcnik01HirRuRUWw8co5VB0+TLY0/1QIy07tWIcrny4C/n6/T8UQ87sqVK9y8eZM0adLQqFEj0qZNS58+ffD393c6moiISILjZAE4H3jJGDMTeBS4YK29a/mneK/le08yeMEusqb2J1+mlE7Hibd8faB64cy81SiIrGn0B68kHtZaZs2axcsvv0zTpk35+OOPqVy5MpUrV3Y6moiISILlsQLQGDMDqA5kMsYcBwYBfgDW2vHAQqA+cAC4CnTwVBaJXUdOX+GdBbs4eOoybzYM4vHALFqaKCIxsmXLFnr16sVvv/1GmTJlaNWqldORREREEgWPFYDW2tb3+bwFunvq9cUzpq87yvDFe+lS7RE+bVOGZEm0RFFEYmbChAl07dqVjBkzMmHCBDp27Iivr36XiIiIxAUnl4BKPLPxyFlGLdnHd90rkzejlnyKSPRFRkZy4cIFMmbMSK1atejVqxdvvfUW6dOndzqaiIhIouLJRvASj92Msly9EXnr448L1+g5YwvvP1VCxZ+IxMgvv/xC6dKladeuHQAFChRg1KhRKv5EREQcoDOA8g/WWhZs/4Nhi/Zw9sqNf3zuhWoFqFkkq0PJRCS+OXr0KP3792fu3Lnky5ePzp07Y63VNcMiIiIOUgEot+wKu8jb34dy6XokI1qWpGKBjE5HEpF4avHixTRt2hRjDO+88w79+vUjefLkTscSERFJ9FQAJiDHzl5lw5GzD/S1G4+e46fQP+ldK4DWFfLg66N36EUkZqy1nDlzhkyZMvHoo4/Stm1b3nzzTXLnzn3/LxYREZE4oQIwgbhwLYJnP19H0RxpHqh5eLa0/iztG0y6FEk9kE5EErodO3bQs2dPTp8+zZYtW0iXLh0TJkxwOpaIiIjcQQVgAmCt5eU526hRODP/a1LM6TgikoicPXuWQYMG8emnn5IuXTreffddXeMnIiLixVQAxjNfrj3Kr/tO/eOxi9cjuHbjJh8/U9qhVCKSGO3cuZPq1atz7tw5unXrxuDBg8mQIYPTsUREROQ/qACMRw6dusyIn/bybtPid12jVyF/BjVlF5E4cebMGTJmzEhgYCDNmjWjR48elChRwulYIiIiEg0qAB0WFWWjfex7P+yma/AjNCiR3YOJRETu7dixY7zyyiv88ssv7N27l7Rp0zJx4kSnY4mIiEgMqAB00JyNx3h57naie7lM4aypaV85n0cziYjc6fr163z44YcMHTqUqKgoXn31VZIm1YZRIiIi8ZEKQAdtPHKOd5oWo23FvE5HERG5p1OnTvHoo49y+PBhmjdvzocffkjevPqdJSIiEl+pAIxjpy6F3+rVt+HoWZ6uoP5YIuJ9zp07R/r06cmUKRONGzemSZMm1KhRw+lYIiIi8pB8nA6Q2Hz+2yH+930o328Lo2SudARlT+N0JBGRW86fP0/v3r3JkycPhw8fxhjDRx99pOJPREQkgdAZwDgWGnaRoU8W5/HArE5HERG55ebNm0yaNIkBAwZw5swZunTpQpo0eoNKREQkoVEBGIestYSGXaBojrRORxERuSUiIoIqVaqwfv16qlatypgxYyhVqpTTsURERMQDtAQ0Dv158Tq+PoYsqZM5HUVEhPPnzwPg5+dH/fr1mTFjBitWrFDxJyIikoCpAIxDoScuEpQjLSa6fR9ERDwgPDycoUOHkjt3blatWgXAoEGDaNWqlX4/iYiIJHBaAuohHy3dx0dL95M0yf/X2Dcio+hTK8DBVCKSmFlrWbBgAX369OHgwYM0bdqUHDlyOB1LRERE4pAKQA85dSmcNxoUoc0dPf6SJdFJVxFxRqtWrZg9ezZFihThp59+onbt2k5HEhERkTimAjCW/H7mKgdPXb51P+z8NQKzpcbfz9fBVCKS2F26dIlUqVJhjKF69eo89thjvPjii/j5+TkdTURERBygAjCWDPhmB1duRJI2+f//URWUQ1uoi4gzoqKimDJlCq+99hojR47k2WefpVu3bk7HEhEREYepAIwF1lp2hl3gp97VyJLG3+k4IpLIrVu3jh49erBhwwYqVapEUFCQ05FERETES+iCtFhw4vw1kvgYMqu9g4g4bMCAAVSsWJHjx4/z5ZdfsmrVKkqXLu10LBEREfESKgBjwY87/6RaQGZtny4ijrhx4wbh4eEAVKhQgddee429e/fy7LPP6veSiIiI/IMKwIdkrWX2xmO0LJfb6SgikggtXLiQ4sWLM3z4cACaNm3K0KFDSZ06tcPJRERExBvpGsAH1H/ONpbs+gtrLdnS+lMhXwanI4lIIrJ//3769OnDDz/8QOHChalQoYLTkURERCQeUAH4gI6fu8rw5iWokD8DyZP64uOjZVYiEjcmTpxI9+7d8ff358MPP6RHjx4kTZrU6VgiIiISD6gAjIHDp6/w54XrAFy4Fkkq/ySkS6E/ukTE86Kiorh+/TopUqSgTJkytGnThiFDhpAtWzano4mIiEg8ogIwBtp/sZ6MKZPi5+tD+hR+5EqXwulIIpIIbNy4kZ49e1KkSBFCQkIoW7YskyZNcjqWiIiIxEMqAKPp4vUITl8K5+d+1fHVck8RiQN//fUXAwYM4IsvviBLlix07drV6UgiIiISz6kAjKbdYRcpnC21ij8RiROLFi2iVatWXLt2jf79+/PGG2+QJk0ap2OJiIhIPKcCMJrOXY0gUyo1ehcRz7p69SopUqSgePHi1KxZk6FDh1K4cGGnY4mIiEgCoQJQRMQLHDp0iL59+3Lx4kWWLVtGrly5mDdvntOxREREJIFRAfgfVu4/Tf852wC4HnmTqoUyO5xIRBKay5cvM3ToUEaMGIGfnx9vvPEGUVFR+Pr6Oh1NREREEiAVgP/hjwvXKJM3HW82DAIgvVo+iEgs2rp1Kw0bNuTEiRO0bduWYcOGkSNHDqdjiYiISALm48knN8bUNcbsNcYcMMa8do/P5zHG/GKM2WKM2W6Mqe/JPA8iuV8SsqdNTva0yfH30zvyIvLwrl27BkDBggUpU6YMq1atYurUqSr+RERExOM8VgAaY3yBT4B6QBDQ2hgTdMdhbwCzrbWlgVbAp57KIyLitFOnTtGlSxfKlCnDjRs3SJUqFfPnz+exxx5zOpqIiIgkEp48A1gBOGCtPWStvQHMBJrccYwF/t7XPC0Q5sE8IiKOiIyM5OOPPyYgIIBJkyZRr149IiIinI4lIiIiiZAnrwHMCRy77f5x4NE7jnkb+MkY0wNICdS61xMZY14AXgDIkydPrAe905TVRwg7f429f10iY0q1fhCRB3fixAmeeOIJQkNDqV27NqNHj6ZIkSJOxxIREZFEypNnAO/VMd3ecb81MNlamwuoD0wzxtyVyVo7wVpbzlpbLnNmz+/EOWh+KACP5s/IsxU9X3CKSMJz/fp1ALJnz05QUBDffvstixcvVvEnIiIijvLkGcDjQO7b7ufi7iWenYC6ANbaNcYYfyATcNKDue4rXQo/ugY/QvqU2vVTRGLm6tWrDBs2jJCQELZv307GjBmZPXu207FEREREAM+eAdwAFDLG5DfGJMW1ycv8O475HagJYIwpAvgDpzyYSUTEI6y1zJ49m8DAQN555x2Cg4OJjIx0OpaIiIjIP3jsDKC1NtIY8xKwGPAFJllrQ40xg4GN1tr5QD9gojGmD67loe2ttXcuExUR8WrXrl2jXr16rFixglKlSjF9+nSqVq3qdCwRERGRu3i0Eby1diGw8P/Yu+84K6r7/+Ov9y6dBZSmgCCiNCFAkKhRsURjxRIswWBLothQ0fhTNBE7iUbsoMZY4Cu2WFAsWDBoMAFBFJEmSBdEpHeW5fP745yrw3UbsMvdXT7Px2Mee++cmTOfmXvF+dxTJm1dv8TrKcChpRmDc86Vlo0bN1K1alWqV69O69atOfvss7nwwgvJzvZnhjrnnHOubCrVB8E751xFtHnzZgYNGsTee+/NtGnTAHjssce4+OKLPflzzjnnXJnmCaBzzm2DUaNG0blzZy6//HLatWtHVpb/M+qcc8658sPvXJxzrhjMjHPOOYejjjqKVatW8dJLL/H+++/TqlWrTIfmnHPOOVdsngA651whNm3aBIAkWrRowa233srUqVM5/fTTkfJ73KlzzjnnXNnlCaBzzuXDzHj55Zdp1aoVI0eOBOC2226jX79+VK9ePcPROeecc85tn1KdBbQ8GfHlt9z//lcArN6wmexs/2XfuV3Vl19+yVVXXcUHH3zAz372M2rWrJnpkJxzzjnnSoQngNHXS9bQYa86XHDIPlSvkk3tapUzHZJzLgNuvvlm7rzzTurUqcPAgQPp1asXlSr5P5XOOeecqxh2mbuadZs2s2T1xgLLl6/dRP2cquzfuPZOjMo5Vxbk5eUhiaysLBo1asTFF1/MbbfdRr169TIdmnPOOedcidplEsB+r03mpU8X0KxujQK36f2r/XZiRM65smD06NFcccUVXH755Vx44YVccsklmQ7JOeecc67U7DIJYN4W496zOtK9816ZDsU5VwYsWLCA6667jueee46mTZvSoEGDTIfknHPOOVfqfBZQ59wu54knnqB169a88sor9OvXj2nTpnHqqadmOiznnHPOuVK3y7QAOud2bWZGXl4elSpVonHjxhx//PHcc8897LPPPpkOzTnnnHNup/EWQOdchTd16lSOP/54br75ZgBOOOEEXn75ZU/+nHPOObfL8QTQOVdhrVy5kmuuuYYOHTrwySef0LRp00yH5JxzzjmXUd4F1DlXIY0YMYLzzz+fJUuWcNFFF3HHHXf4RC/OOeec2+V5Auicq1Dy8vLIzs5mr732om3btrz99tt07tw502E555xzzpUJu0wCmJu3hewsZToM51wpWbhmoy/aAAAgAElEQVRwIX379iUvL4+hQ4fSvn17Ro0alemwnHPOOefKlGKNAZRURVK5fkr692s2Uj+naqbDcM6VsI0bN3LXXXfRunVrXnjhBZo3b46ZZTos55xzzrkyqcgEUNJJwCTgvfi+k6RXSzuwkrZk9UYa1PIE0LmKZMKECbRv356+ffty9NFHM2XKFO68804kb+13zjnnnMtPcVoAbwMOAlYAmNnnQLlrDfxu9UYaegLoXIWQl5cHQJMmTahXrx4jRoxg2LBh7LvvvhmOzDnnnHOubCvOGMBcM1uR9ot6uepftSE3j425W6hTvXKmQ3HO7YBVq1Zx++23M27cOD744AP22GMPxowZk+mwnHPOOefKjeK0AE6VdBaQJWkfSfcD5eqO6/GPZpGVhXcLc66c2rJlC4MHD6ZVq1bcc889tGjRgvXr12c6LOecc865cqc4CWBv4ABgC/AKsAG4qjSDKmlTFq3i4sO9a5hz5dG8efP45S9/yQUXXEDz5s0ZO3YsTz75JDVr1sx0aM4555xz5U5xuoAeZ2bXA9enVkjqTkgGy4XcPGPfhjmZDsM5tw22bNlCVlYWDRs2pFKlSgwePJhzzjmHrKxiTV7snHPOOefyUZw7qb/ks+7PJR1IaZm/bB3vT11M/ZpVMh2Kc64YNm3axIABA+jUqRPr16+nWrVqjB49mvPOO8+TP+ecc865HVRgC6Ck44DjgSaS7k0U1SZ0By0XNuTmsV/DHA7Zr36mQ3HOFWHEiBH06dOH6dOnc+KJJ7Jy5UqqV6/u43edc84550pIYV1AvwO+JIz5m5xYvxroW5pBOed2LatWreKcc85h+PDhtGzZkjfeeIOTTjop02E555xzzlU4BSaAZvYZ8JmkoWa2YSfG5JzbRaTG+dWqVYvc3FzuuusurrrqKqpW9Wd2Ouecc86VhuIMqGki6XlJX0j6KrWUemTOuQrLzHjmmWdo164d3377LZJ46623uO666zz5c84555wrRcVJAJ8GngIEnAC8CDxfijGVqGGff8OKdbmZDsM5F3366accdthhnHvuueTk5LB8+XLAn9PpnHPOObczFCcBrGFm7wCY2ddm9hfgqNINq+RM/3YNvzuoWabDcG6Xl5eXR69evfjFL37BzJkzeeKJJxg7dixt27bNdGjOOeecc7uM4jwHcKPCT/NfS7oE+AZoWLphlZxKWaLtnrUyHYZzuywzQxLZ2dnk5uZy9dVX069fP+rUqZPp0JxzzjnndjnFaQG8GsgBrgQOBS4C/lCaQTnnKob33nuPjh07MnHiRACefPJJBgwY4Mmfc84551yGFJkAmtlYM1ttZvPM7FwzOwWYuxNic86VU7NmzeI3v/kNxx57LGvXrmXVqlWAj/NzzjnnnMu0QhNASb+QdJqk+vF9O0lDgDE7JbodNP3b1SxauT7TYTi3S7nzzjvZf//9effdd+nfvz+TJ0+ma9eumQ7LOeecc85RSAIo6a/AUKAnMELSn4F/AxOBVsWpXNLxkqZLmikp34fHSzpL0hRJkyU9u+2nULC/vzOdqpWyae1jAJ0rVWaGmQGwceNGTj/9dKZPn84NN9xAtWrVMhydc84555xLKWwSmFOBjma2XlJdYGF8P704FUvKBgYCvwYWAOMkvW5mUxLbtARuAA41s+WSSnhyGeOiw1vQokFOyVbrnPvB559/zpVXXsl1111Ht27duPXWW72rp3POOedcGVVYF9ANZrYewMyWAdOKm/xFBwIzzWyWmW0iPDvw1LRtLgIGmtnyeJzvtqF+51wGff/991x66aUccMABTJ06lfXrQ3drT/6cc84558quwloAW0h6Jb4W0DzxHjPrXkTdTYD5ifcLgIPStmkFIOljIBu4xcxGpFckqRfQC6BZM3+mn3OZNnjwYPr06cPq1avp3bs3t9xyC7vvvnumw3LOOeecc0UoLAE8Pe39w9tYd37NAJbP8VsCRwJ7Af+R1N7MVmy1k9k/gH8AdOnSJb0O59xOknqm35YtWzjggAN44IEHaNeuXabDcs4555xzxVRgAmhmI3ew7gVA08T7vQjjCNO3GWNmucBsSdMJCeG4HTw2Uxet4v2p33Fh1xY7WpVzu7y5c+dy7bXX0rVrV6688kouuOACLrjgAu/u6ZxzzjlXzhTnQfDbaxzQUtI+kqoAPYDX07YZBhwFEB810QqYVRIH/3Tuck7vvBcHt6hXEtU5t0tat24dt9xyC23atOHNN98kLy8PCOP8PPlzzjnnnCt/CusCukPMbLOk3sA7hPF9T5rZZEm3AePN7PVYdqykKUAe8P/MbOmOHjs3bwvDPvuGbh0a7WhVzu2y3n33XS666CLmzZvHb3/7W+6++24fg+ucc845V84VOwGUVNXMNm5L5Wb2FvBW2rp+idcGXBOXEvPFgpWMn7ucu8/oUJLVOrdLSI3zq1y5MrvvvjtDhgzhiCOOyHRYzjnnnHOuBBTZBVTSgZImATPi+46SHir1yHaI0bnZbv78P+e2wbJly7jiiiu47rrrADjqqKOYMGGCJ3/OOeeccxVIccYAPgh0A5YCmNlE4ri9suqNLxaxblNepsNwrlzIy8vjkUceoVWrVgwaNIjc3FxC4zxkZZXmMGHnnHPOObezFefuLsvM5qatK9PZ1aQFKzm5Y+NMh+Fcmff5559zwAEHcNlll9G+fXs+++wz7r//fp/gxTnnnHOugirOGMD5kg4ETFI2cAXwVemGtWOqVs6i4167ZToM58qs1Di/mjVrsnbtWl544QXOPPNMT/ycc8455yq44iSAlxK6gTYDFgPvx3XOuXJmw4YN3HPPPUydOpWhQ4fSsmVLpk+f7l09nXPOOed2EcW569tsZj3MrH5cepjZ96UemXOuxJgZr776Kvvvvz833XQTGzduZOPGMKmvJ3/OOeecc7uO4tz5jZP0lqTzJdUq9YiccyVq3rx5HHvssXTv3p0aNWowcuRIXnrpJapWrZrp0Jxzzjnn3E5WZAJoZvsCdwAHAJMkDZPUo9Qjc86ViJycHObMmcNDDz3E559/zq9+9atMh+Scc8455zKkWH2/zOy/ZnYl0BlYBQwt1aicc9stLy+Pxx9/nOOOO468vDzq1q3LtGnT6N27N5UqFWfYr3POOeecq6iK8yD4HEk9JQ0HPgGWAIeUemTOuW328ccfc+CBB9KrVy/WrVvH0qVLAcjOzs5wZM4555xzriwoTgvgl8DBwN1mtp+Z/cnMxpZyXM65bbBixQrOOeccDjvsMBYvXsyzzz7LRx99RMOGDTMdmnPOOeecK0OK0x+shZltKfVInHPbrUaNGkyePJk///nP3HDDDdSsWTPTITnnnHPOuTKowARQ0gAz+xPwsiRLLzez7qUa2XaasXg1H89cytXHtMp0KM6VGjPjjTfe4O9//ztvvvkmtWrVYty4cT7GzznnnHPOFaqwu8UX4t+Hd0YgJWXp2k3s36g2XZrXzXQozpWKadOm0adPH9555x3atm3LN998Q5s2bTz5c84555xzRSpwDKCZfRJftjWzkckFaLtzwts+OdX8RthVPLm5uVx77bX87Gc/43//+x/33XcfEydOpE2bNpkOzTnnnHPOlRPFmQTmD/ms+2NJB+KcK1ylSpWYNGkS559/PjNmzKBPnz5Urlw502E555xzzrlypLAxgL8FegD7SHolUVQLWFHagTnnYMyYMfTt25chQ4bQrFkz3njjDU/6nHPOOefcdiusr+QnwFJgL2BgYv1q4LPSDGpHbLGfzFfjXLmzaNEibrjhBgYPHkyjRo2YM2cOzZo18+TPOeecc87tkAITQDObDcwG3t954ey4T2Yvo82etTIdhnPbbcCAAdxyyy1s3LiRvn37cuONN1Krln+nnXPOOefcjiusC+iHZnaEpOVAsllNgJlZmZtmc8sW46VPF/BIzwMyHYpz22369OkceeSR3HvvvbRs2TLT4TjnnHPOuQqksElgjop/6wMNEkvqfZkzYd5yalapRPsmtTMdinPFNmPGDE4++WTGjh0LwMCBAxk+fLgnf84555xzrsQV9hiILfFlUyDbzPKAXwIXAzV3Qmzb7PP5KzioRV0kZToU54q0evVqrr/+etq1a8eHH37I7NmzAXycn3POVUCSJks6sohtmklaIyl7J4XlnNsFFecxEMMAk7QvMITwDMBnSzWq7TRl4SraNfbWP1f2vfDCC7Rq1Yq7776bnj178tVXX9GjR49Mh+Wcc7scSXMkrY+J12JJT0nKKenjmFk7MxtVxDbzzCwn/uhe7im4S9LSuNytAn6ll3Rj/AxSy3pJWyTVj+V3S5ovaZWkuZL+nNi3vqSP4zFWSPqfpEMT5T0kTZe0UtJ3kgZLqp0of0bSolj3V5IuTJTtL2m8pOVxeV/S/onyPpJmxX0XSrpPUqVE+b8lLYnlEyWdmig7StKkGPNSSa9KapIor3Dn7MqG4iSAW8wsF+gO3G9mVwBNitgnI+YuW0fzemWycdK5rcyaNYumTZsyZswYnnrqKfbcc89Mh+Scc7uyk80sB+gM/AL4S/oGMZkpzn2T+1Ev4DSgI9AB6EboSfYTZtY/Jr858bO4CxhlZt/HTZ4A2phZbeAQ4HeSuseyNYTnVjcAdo/7Dk8kJR8Dh5pZHaAFYQ6MOxKH/yvQPNZ9CnCHpNSEEguBM4C6hGFQrwPPJ/YdDnSO+7aP53plovwqoFEs7wU8I6lRLJsCHGdmuwGNgRnAI4l9K+I5uzKgOP+QbZZ0JnAu8EZcVyb7qJkZlbK9+6crexYvXswf//hH/vWvfwFw7bXXMmbMGA466KAMR+accy7FzL4B3ibc1CJplKQ7JX0MrANaSKoj6YnYevKNpDuU6LIp6SJJUyWtljRFUue4fo6kY+LrA2MLyyqFVsd74/rmkix1Ey+psaTXJS2TNFPSRYnj3CLpRUlD4rEmS+qys65VMZ0PDDCzBfHaDgAuKGonSSLcdw5OrTOz6Wa2NrHZFmC/WLYhlm8hTFaYR0iK6sby+YlEkli+X6LuyWa2MfU2LvvGshVmNsfMLFF3ct+vzSz1fGwl44rlX5jZ5kTdlQnDqzCzxWa2sJC4Ktw5u7KhOAngHwgTwtxtZrMk7QM8V7phOVcx5Obmct9999GqVSuGDBmy1Ti/rCz/Idk558oSSU2BE9n6ecfnEloxagFzCUnJZsIN78+BY4EL4/5nArcA5wGplpWl+RzqAeCB2EKyL/BiASE9BywgtA6dAfSXdHSi/BRCy8xuhFaah7flfItL0u9iN8OClmYF7NoOmJh4PzGuK0pXYA/g5bQ4+kpaQ7gmNUkbkiTpC2AD4Vr808y+S5QdJmkl4XnWpwP3p+07SNI6YBqwCHgrrXxFrPshoH9a2e8krQK+J7SGPZZW/oakDcBYYBQwPlHWLNa9HrgWuLuin7PLvCLvQM3sS0Kz7nhJbYD5ZnZnqUfmXDk3atQoOnTowDXXXMMhhxzCpEmTuO666zIdlnPOuZ8aFm92RwMfsvXN7tOxtWQzoXXlBKCPma2NN9v3AalB3BcSfjAfZ8FMM5ubz/Fygf0k1TezNWY2Jn2DmIweBlwfW3s+B/5JSEhTRpvZW3HM4P8RbsRLnJk9a2a7FbLMK2DXHGBl4v1KICe28BXmfOAlM1uTFsffCIl4Z8L5rkwr70BIvH9H+CyTZaNjd8i9gL8Dc9LKL4t1dwVeATamle8G1AF6s/UPBKnrUxtoBTwKLE4r7xbrPhF4JzHRYmrc526ErpZ/ISRjFfqcXeYVmQBK6grMJPRDfhL4KjnI1DmXv8WLF5Obm8vw4cN56623aNOmTaZDcs45l7/TYiKzt5ldZmbrE2XzE6/3JnRnW5Rq/SK0fDSM5U2Br4txvD8SbpynSRonqVs+2zQGlpnZ6sS6uWw9D8O3idfrgGrJyTjKgDWE5CSlNrAmdi3Ml6TqwJkkun8mxcT6M0KL2a35lG8ws+eAvpJ+khDHrqgj2HpMW6osz8xGExKmS/MpX0tIdoZIaphP+QxgMjAon7JcM3sbOE7SKfmULyOc82vpn2FFPWeXOcXpg3YfcKKZHWpmhwAnEboulCnTvl3FhHkrqJzt3epcZqxZs4Y///nP3HfffQCcddZZTJ48mW7duvmjSZxzrvxKJivzCa0k9ROtX7XNrF2ifN8iKzSbYWZnExLHu4CXJKXPYrcQqCupVmJdM+Cb7T2R7SWpp7aeoTN9KagL6GS2bpXsGNcVpjuwjNBtsDCVKPxaVyZMfrI9+xZWngXUoOAJEXek7kqE70RBU9pXxHN2GVCcbKmKmU1JvTGzqUCV0gtp+yxauYE2e9biZ03qZDoUt4sxM5599llat25N//79mTFjBgCSqFq1aoajc845V1LMbBHwLjBAUm1JWZL2lXRE3OSfwLWSDgiThmo/SXun1yPpHEkNYre41GQaWz36wczmA/8F/iqpmqQOhJbDoaV1fgUxs6GWmKEzn6WgLqBDgGskNZHUGPgT8HQRhzsfGJJsJYzX+WJJu8freiBwOTAylh8cx7tVkVRd0vWEMYRjY3nPONZO8fO4M7FvQ4VHJuRIypZ0HHA28EEs/7Wkn8ey2sC9wHJgaiy/MNUypvCohBsSdbeRdEKMqbKkc4DDCd2MkdRdUut4fg1i3Z+Z2bKKes6ubChOAjhB0mPxS3aYpEdI6wdcFixfu4m6Nat4S4vbqb788ku6du1Kz549adSoER9//DGDBv2kF4RzzrmK4zzCD+FTCDfFLwGNAMzsX4Qb7WcJE28MI87KmOZ4YLLC5B4PAD3MbEM+250NNCe0Br4K3Gxm75XkyZSyxwiPDJgEfAm8SWKykNh62DXxvgnwK0LimO43hO61q4FnCBOTPBTLqgIDCRPufEMYd3aS/TjD5v6EZHoN4fEI04HUjKpG6Pq4gPB53kMY4/laLN+NMBnPynj8/YDjE5/XocAkSWsJk6i8BdyYOiXCpEDfAUsIj0f4rZlNiOVNCF0zV8drtCWeZ0U+Z1cGqJBu2GEDqRphEpjDCB/qR8BDBfxDVeq6dOli48f/dCKhg/q/T9eWDbjnzFIZ/+xcvsaMGcOpp55K//79+f3vf+8zezpXwiR9amZlbWp755xzrtwqdKCwpJ8R+uy+amZ3F7ZtpuVtMa47rnWmw3AVXG5uLo888gjffPMNd911FwcffDBz586lWrVqmQ7NOeecc865IhXYXCHpRkLXhZ7Ae5L+sNOi2kab87awYl0u9XJ8vJUrPSNHjuTnP/85V111FRMnTmTz5vCMU0/+nHPOOedceVFYf7WeQAczOxP4BflMDVtWLF27id1qVCE7y8f/uZK3YMECTj/9dI455hjWrVvHq6++yttvv02lSmVppm3nnHPOOeeKVlgCuDE++wMzW1LEtvmSdLyk6ZJmSupbyHZnSDJJ2zXOY8nqjTSo5a1/rnTk5eXx4YcfcscddzBlyhROO+00n2zIOeecc86VS4U1YbSQ9Ep8LWDfxHvMrHthFUvKJsxO9GvCLEPjJL2efKRE3K4WYZKZsdsRPwBL1mykfk6ZezKFK6fMjBdffJF33nmHJ554gr333pt58+ZRo0aNTIfmnHPOOefcDiksATw97f3D21j3gcBMM5sFIOl54FTCtMlJtwN3A9duY/0/2JxnVK3ksy+6HTdx4kSuvPJKPvroIzp27Mjy5cupW7euJ3/OOeecc65CKDBrMrORhS3FqLsJMD/xfkFc9wNJPweamtkb2xW9cyVk5cqVXH755XTu3JnJkyfz6KOP8umnn1K3bn6Pb3LOOee2n6QjJS1IvJ8j6ZhMxuSc23WUZrNZfoOkfnjooKQs4D7gT0VWJPWSNF7S+CVLlpRgiM4FWVlZvP7661x++eV89dVXXHzxxWRnZ2c6LOecc6UsJl/r40PJv5X0tKScTMdVnii4S9LSuNytAgbLSzpK0iRJK+K2r8YHwKfKm0h6TdIySQskXZIo6xo/p+Rikk6P5RdIyksrPzKx/yGSPpG0WtIXkg5Li+13kuZKWitpmKS6ibJnJC2StErSV5IuTNu3hqRBkr6XtFLSR2nn/O+4fk4h1/GIeD53FFD+QSyvlFiX/P6ukfRuouzRtGuxUdLqRHlzSW9JWh6/+w+n6pZUX9LH8TNaIel/kg5N7Nte0jvxfAt8qLiklpI2SHqmoG3czleaCeACoGni/V7AwsT7WkB7YFT8j+Fg4PX8JoIxs3+YWRcz69KgQYNSDNntSkaNGsVZZ51Fbm4utWrVYtq0aTz44IPe6uecc7uek80sB+gE/By4IcPxlDe9gNOAjkAHoBtwcQHbTgGOM7PdgMbADOCRRPkzwGxgD+AkoL+kowDM7D9mlpNa4nHWACMS+/8vuY2ZjQKIydzrwN+B3QjDj4ZL2j2WtwMeA86Nx14HDErU+1eguZnVBk4B7pB0QKL8H0BdoG38e3WibC3wJPD/CrgmSKoMPEABc2JI6knBQ7dOTpzvsamVZnZJ2vV6DvhXYr9BwHdAI8J3/wjgsli2BvgD0ADYHbiLcL1SMeQCLwJ/LOicooHAuCK2cTtZsRNASds6zeY4oKWkfSRVAXoQ/sMDwMxWmll9M2tuZs2BMcApZjZ+G4/j3DaZN28eZ511FkcddRRjx45l9uzZANSsWTPDkTnnnMskM/sWeIdwMwyE+x9J90iaJ2lxbFWpnig/VdLnsWXoa0nHx/W/lzQ1tjbNklRQQlQRnA8MMLMFZvYNMAC4IL8NzWyxmSUbBPKA/QAUWl6PBO40s1wzmwi8REhECjruS6lZ64twCLDYzP5lZnlm9gywBEhNatgTGG5mH5nZGuAmoLvCZIWY2WQz25g6jbjsG+NuTUgKe5nZklj/p4lz/sTM/g+YVUh8fwLeBaalF0iqA9wMXFeM88yXpJqE+T0GJ1bvA7xoZhvid38E0C7GvMHMppvZFkKvvjxCIlg3lk83syeAyYUcswewAijO0DG3ExWZAEo6UNIkwi80SOoo6aGi9jOzzUBvwj+kUwlfsMmSbpN0yg7G7dw227RpE7fddhtt2rRh+PDh3HLLLUydOpVWrVplOjTnnHNlgKS9gBOAmYnVdwGtCEnhfoT5DPrF7Q8EhhBadnYDDgfmxP2+I7RQ1QZ+D9wnqXOpn8QOiF0gVxSyNCtg13bAxMT7iXFdQcdpJmkFsJ4wCeDdqaK0v6nX7fOpowZwBlsnNAA/j90Sv5J0U6LFSmn1pte91TmY2dfAJsJnnzrmIEnrCEnaIuCtWHQQMBe4NR57kmK31OKQtDchyb2tgE36E1pJvy2gfKikJZLeldSxgG1OJyS8HyXWPQD0iN1XmxC++8nWVCR9AWwgNOL808y+K+Y51Y7nU+RQL7fzFacF8EHCP2BLAeKvMUcVp3Ize8vMWpnZvmZ2Z1zXz8xez2fbI731z5Wm7OxsXnnlFbp168a0adO4+eabfXZP55xzAMPi2Kj5hMTtZghj24CLgKvNbJmZrSbcjPeI+/0ReNLM3jOzLWb2jZlNAzCzN83saws+JLTudN3J57VNzOxZM9utkGVeAbvmACsT71cCOfH65XecebELaH3gL8RWr3h9PwZuklQtJsynA/n9z/p04Hvgw8S6jwgJXcNYfjY/drv8L9BY0tmSKks6n9CCl6o7/RxS51ErEfdl8X1X4BUg1SK4VzzuSkK31t7AYElt8zv/fDwI3BRbHrcSh0YdChTU+NITaA7sDfwbeEfSbvlsdz4wxMyS4/U+JCS+qwhDt8YDw5I7mVkHwo8YvwNGF/N8IMzy/4SZzS9yS7fTFScBzDKzuWnr8kojGOdK2pdffskZZ5zB8uXLyc7OZvTo0bz44ovsvffemQ7NOedc2XGamdUidD9sQ0hMIIx/qgF8mmoFI7SQpCYkaAp8nV+Fkk6QNEZhMpMVwImJeiuaNYQkIaU2sCYt2fgJM1tGaMF7LdFS15PQNXE+odVrKCE5SfeThMbMZpnZ7JiMTyK0QJ0Ry5YSHkd2DbAYOB54P1F3+jmkzmN1ckXs3jmakPRdGlevJ4yJu8PMNsWE/9/AsRRB0slALTN7IZ+yLMI4vatiz7qfMLOPzWy9ma0zs78Sulx2TaunKWF835C0ut8hJLI1Cd/N1Fi/9GNsMLPngL6FtDAmj9cJOIYw2aMrg4qTAM6PXRxMUrakPsBXpRyXcztk+fLlXHnllXTq1IkPPviASZMmAZCT4xO7Oeecy1+8cX8auCeu+p5wc98u0QpWJ06oASFJ2Te9njhvwsuxnj1ia9db5D9Depkhqad+OstmcimoC+hkwgQwKR0pZGxYmkqEFrvaAGY218y6mVkDMzsIqAd8khZnU0KyPoTCGYlrbmYfmtkvzKwuYbKX1om6tzoHSS2AqhR8z1uJHz/7L4qIozBHA10UZuH8Fvgt0EfSa4Rr0gV4IZalJlNZIKmg1uStzjk6D/ivxWdzR3UJP2A8bGYbY4L8FOGHioJUBloU45yOJLRKzotxXwucLmlCMfZ1O0FxEsBLCb+WNCP8YnIwP/7iUSZsyM1j85ZCf2Ryuwgz47HHHqNly5YMHDiQXr16MWPGDA4//PBMh+acc658uB/4taROcQKMxwnj9xrCD48pOC5u+wTwe0lHS8qKZW2AKoTkYQmwWdIJFKM1KNPMbKhtPYNm+lJQF9AhwDXx/BsTxn09nd+GkrpLah2vVwPgXuCz2BqIpLaSakmqIukcwnW7N62acwkJzddpdZ8gaY/4ug1hIpfXEuU/j90/axOS8wVm9k4sHgqcrPCoiZqE1sNXzGy1pIaSekjKiY0hxxG6l34Q9/0ImAfcIKmSwuMSjiS0sBHPtRohgVLs3lol7nsTP44x7UQYa/c4YdxoqktpqiyVnB0AjI1jKQ+N16qapP9HaMn7OO16nZf+eZjZ94TZVi+NMe9GaFWdGGM+WNJhse7qkq4nzI46NhQRwGsAACAASURBVJYrnlOV+L6afpww8h+E5DgV96PAm8BxuDKhyATQzL4zsx4WZuysH19/vzOCK65bh0+mZpWCZsZ1uxJJvPnmm7Rr144JEyYwaNAg6tWrl+mwnHPOlRNmtoSQ0NwUV11PmBRmjKRVhG6DreO2nxAneCHcrH8I7B3Hsl1JmCZ/OWH81E/mP6hAHgOGA5OALwk3+4+lCmPrYarFqgmhG+3quP0W4DeJuo4jzJa5HLgEOD5+Jknn8dPJXyC0pn0haS2hxfUVwpjNlOsIrbrzCY8++OG4ZjY5Hm8oYRxoLX58JIIRGj8WxLjuAfqY2Wtx31xC99ITCd+Dx4HzUuNBCZMDrY8xNYuv3437rjazb1NLLFsbx5xaWlnqOiw2s00xxkdiTN8QurWeEFvzAJD0S0J31eTjH1K6x32WEL7jm/nx8RVVCY9wWBrrPhE4yX6cwXXvGGuqpXc9MD2e07q0uNcAG/L5HF2GqIju2Uh6nMQD3FPMrFdpBVWYLl262PjxW88Vc/z9H3HvWZ3Yv3F61223K1iwYAE33ngj/fr1Y7/99mP16tXk5ORQwNhz51w5IulTM/vJ82Gdc845t32K0wX0fcLzO0YSmpQb8uOsRxm1fO0mnvtkHivW5WY6FJcBGzZsoH///rRu3ZoXX3yRceNC1/hatWp58uecc84551w+iuw3mT4rkaT/A94rtYi2wX9mfs8jo77m1/vvwV51qxe9g6swXn/9da6++mpmzZpF9+7dueeee9hnn30yHZZzzjnnnHNl2vYMnNuH0O+3TOiwVx1uP+0nzwd1FdzIkSOpVq0a7733Hsccc0ymw3HOOeecc65cKLILqKTl8Rk2qefYvAfcWPqhOfejlStX8qc//YmPPvoIgP79+/P555978uecc84559w2KLQFUGEgVUfC7D8AW4p6qKdzJWnLli08/fTT3HDDDSxZsoQGDRpw+OGHU7NmzUyH5pxzzjnnXLlTaAJoZibpVTM7YGcFtC02523B09GKa+zYsfTu3Zvx48dzyCGH8Pbbb9O5c+dMh+Wcc84551y5VZxZQD+RVCbvum94ZRJVKxXnFFx59L///Y+FCxcydOhQRo8e7cmfc84555xzO6jAFkBJlcxsM3AYcJGkr4G1gAiNgxm/G29UpxpXHN0y02G4ErJx40YeeOABmjZtytlnn83ll1/OhRdeSE5OTqZDc84555xzrkIorPnsk/j3NKA1cCJwJnBG/OtciXnzzTdp3749119/PR988AEAlStX9uTPOedcqZI0R9J6SWsSS+NY9g9J0yVtkXRBhkMttyRdLelbSSslPSmpagHbNZdkaZ/FTYnysyT9V9I6SaPy2d8krU3s+8+08s6SPopliyVdFdc3lPScpIUxxo8lHZTY78j4HUjGdX6ivK6kV+Ox50r6XdpxfxfXr5U0TFLdirqvpEaSXo/X0iQ1T/+cXOYVlgAKwMy+zm/ZSfG5Cm7GjBmcdNJJdOvWjezsbEaMGMHjjz+e6bCcc87tWk42s5zEsjCunwhcBkzIYGxA6JmV6Ri2h6TjgL7A0UBzoAVwaxG77Zb4LG5PrF8G3A/8rZB9Oyb2vTARR31gBPAYUA/YD3g3FucA44ADgLrAYOBNSclfoRemfUcGJ8oGApuAPYCewCOS2sXjtovHPDeWrwMGVdR9gS3xOp+OK7MKSwAbSLqmoGWnRegqtKlTpzJ69GgGDBjAF198wXHHHZfpkJxzzjkAzGygmY0ENhS1raRqkp6RtFTSCknjJO0Ry+pKeiq2iiyXNCyx30WSZio8but1xdbHWGaSLpc0A5gR17WR9F7cfrqks0r+zEvU+cATZjbZzJYDtwMXbE9FZva+mb0ILCxy45+6BnjHzIaa2UYzW21mU2O9s8zsXjNbZGZ5ZvYPoAqhB1yhJNUkJDs3mdkaMxsNvE5IniAkSMPN7CMzWwPcBHSXVKsi7mtmi81sECGhdmVUYQlgNuEXkVoFLBk1ddEq5ixdF5opXbmxZcsWBg8ezAMPPADAySefzOzZs7nmmmuoUqVKhqNzzjnnttv5QB2gKaGF6RJgfSz7P6AG0A5oCNwHIOlXwF+Bs4BGwFzg+bR6TwMOAvaPN9/vAc/Ges4GBiVaX0qNpMNiYlvQclgBu7YjtKSmTAT2kFSvkMPNlbQgJs31tzHUjxS6m76S1v3wYGCZQhfS7yQNl9QsvwokdSIkgDMTqxsqdBudLem++FkAtALyzOyrtHNMfSZbnX/sRbcp7lcR93XlQGHdCRaZ2W07LZJtNHXRKvZrmEOzujUyHYorpnHjxnHFFVcwduxYjjzySK644gqysrKoW7du0Ts755xzpWeYpM3x9SgzO2076sgldi00sy+ATyGMiQJOAOrFFjCAD+PfnsCTZjYhbnsDsFxSczObE7f5q5kti+W/BeaY2VOxbIKklwnzM0zejpiLLbb07LYdu+YAKxPvU69rAUvTtv0e+AXwOeFaDgSGAsXtHnQEMIaQbN8BvCGpU5zUcC+gM/BrYBJwN/AccGiyAkm1CQn7rWaWinUa0Cn+3ZvQRfRe4OJ8zi91jqnGksLK8yrgvq4cKHIMYFn2syZ1yMoq82Hu8hYvXswf/vAHDjzwQObOncvgwYMZOXIkWVn+CA/nnHNlwmlmtltcipX8aesJQZoRkoZ3gOdjV8+7JVUmtAguSyR/SY0JrX4AxO52S4EmiW3mJ17vDRyUbHkjJJF7bsvJ7mRrgNqJ96nXq9M3jF0Kx5vZZjNbDPQGjo1JWZFil8VNZrYCuArYB2gbi9cDr5rZODPbQBiHeIikOqn9JVUHhgNjzOyviXq/NbMpZrbFzGYD1xGS7vzOL3WOq4tRXhH3deVAYXfgR++0KFyFtmjRIp577jmuu+46pk+fznnnnefJn3POuXItbUKQeWaWa2a3mtn+wCFAN+A8QgJXV1J+rWcLCUkd8MN4snrAN8lDJV7PBz5MJKupyVIuLenzSyepa1rSm750LWDXyUDHxPuOwGIzS2/9y0/q3Lf3135L7PsFW1/LrepWmJl0GOHaX7wN9X4FVJKUfC5ZR35skd3q/CW1AKrG/Srivq4cKPAuPNXdwLntMWLECPr16wdAp06dmD9/PnfddRe1axfrRzznnHMu4yRVkVSNcLNfOU70ku+9k6SjJP1MUjawitAlNM/MFgFvE8bq7S6psqTD427PAr+X1CkmIP2BsYnun+neAFpJOjfWU1nSLyS1LWD7EmNm/0lLetOX/xSw6xDgj5L2l7Q78Bfg6fw2lHSQpNaSsuIYwQcJXXJXxvLs+HlUArLi51E5lrWL1zFbYfbOAYRkbmqs/ingN3GbyoSJTUab2Yr4/iVCK+F5ZrYlLa4jJTVT0JQwC+lr8bqsBV4BbpNUU9KhwKmEFmEIXVhPjgl0TeA24JU4CU1F3Jf4GaUe9VE1vndlSLlthvliwUrytljRG7qdaubMmZxyyimccMIJvPDCC6xZswaA+vW3dQy3c845l3HvEpKCQ4B/xNeHF7DtnoQkYhUh6fgQeCaWnUtICKcB3wF9AOIMozcBLwOLgH2BHgUFY2argWPjNguBb4G7+PFmu8wxsxGE8Xb/JnR3nQvcnCqXNFlSz/i2BeERAquBL4GNhIluUs4lfAaPAF3j69Szo/YAXiBc/1mER050M7PcGMcHwI3Am4TPYD8g9fy6VIvtscCKfFo1OwP/A9YC/42xXZmI6zKgeqz3OeBSM5scjzuZMCHQ0FheK25fIfeN1hO6ikL4zq/HlSkyK19JVJcuXWz8+PG0vWkEV/+6Jb0O3zfTITlgzZo19O/fnwEDBlClShVuuukmrrrqKqpWLbP/T3LOlQOSPjWzLpmOwznnnKsoyuVDRQHq1qzCCe0bZToMF61atYqHH36YHj168Le//Y1Gjfyzcc4555xzrqwplwngve9OZ9naTcgnAM2oCRMmMHjwYO6//34aN27MzJkzadiwYabDcs4555xzzhWgXI4BfGH8fPqdvD+N61TPdCi7pCVLltCrVy+6dOnC888/z5w5cwA8+XPOOeecc66MK5cJIMBRrRv6MwB3stzcXB544AFatmzJU089xdVXX81XX33FPvvsk+nQnHPOOeecc8VQLruAuszIzc1lwIABHHTQQdx///20bVvqs04755xzzjnnSlC5bQF0O8fs2bO57LLL2LBhAzVq1OCTTz5hxIgRnvw555xzzjlXDnkC6PK1du1a+vXrR9u2bRkyZAgTJkwAYM8990Q++45zzjnnnHPlkieAbitmxgsvvECbNm24/fbbOeOMM5g+fTqHHHJIpkNzzjnnnHPO7SBPAN1WzIz77ruPBg0aMHr0aJ555hmaNGmS6bCcc865Eiepp6Q1+Swmqd9OjKObpE8krZW0VNJQSXuV0rFukfRMadRdWiR1kvSppHXxb6dCtq0r6dV4LedK+l2i7Ma0z3m9pC2S6udTxxJJoxPr0r8r6+L35IBYXlXSo5IWS1omabikJon9m0t6S9JySd9KelhSpUT5yZK+jHX/V9L+aTFdHfdbKelJSVXj+oaSnpO0MJZ9LOmgxH4nSRotaUXc/3FJtRLlTSS9FmNeIOmStOP+Q9L0eJ0uSCurKum+eOzlkgZJqpwo7y1pvKSNkp5O27eo67mbpMGSvovLLYl9Cz1nVzRPAB1Lly6lT58+fPfdd2RlZfHaa68xbtw4Dj300EyH5pxzzpUaMxtqZjnJBegDLAYe39b6kjf027DPGcCzwANAfaAdsBEYLWn3ba2vopFUBXgNeAbYHRgMvBbX52cgsAnYA+gJPCKpHYCZ9U/7rO8CRpnZ92l13AVMTa5I/64AlwGzgAlxk6uAXwIdgMbACuChRBWDgO+ARkAn4IhYB5JaAkOBS4DdgOHA66nvk6TjgL7A0UBzoAVwa6w3BxgHHADUjdfnTUk5sbwOcEeMqS2wF/D3RFzPALPj9ToJ6C/pqET5xBjnBH6qL9AFaA+0AjoDf0mUL4zHfjJ9x2Jcz/uAGvF8DwTOlfT7Yp6zK4IngLuwzZs3M3DgQFq2bMnDDz/Mv//9bwD22GMPsrOzMxydc845t3NJ+jnhxrOHmS2K6+pIekLSIknfSLpDUnYsuyC2PtwnaRlwi6QsSX+JrU/fSRoiqU4BxxMwALgj3hCvN7NvgQuBNcDVieOMlnRPbGmZLemERD0FxriN599W0qjYWjRZ0imJshMlTZG0Oh7j2ri+vqQ34j7LJP1HUkneXx5JmLX+fjPbaGYPAgJ+lU/8NYHTgZvMbI2ZjQZeB87NZ1vF9YPT1v+SkNA8VURc5wNDzMzi+32Ad8xssZltAJ4nJPMkyl80sw3xMx6RKD8O+I+ZjTazzYQEtAkhSUwd6wkzm2xmy4HbgQsAzGyWmd1rZovMLM/M/gFUAVrH8mfNbISZrYv7Pg4cGs81h3B97zSzXDObCLwE/CEVtJkNNLORwIZ8rsHJwINmtszMlgAPpu37ipkNA5YWcS3zu54nA3fHuOcAT6TqLuqcXdE8AdxFjRo1is6dO9O7d286d+7MxIkT+e1vf5vpsJxzzrmMkLQb4eb3DjMblSgaDGwG9gN+DhxLSNBSDiK0XDQE7iTcmF8AHEVoqckBHi7gsK2BZsC/kivNbAvwMvDrtONMJ7QS3g08EZOY4sRYpNh1bzjwbjyXK4ChklI31U8AF5tZLUKC9EFc/ydgAdCA0Ip0I2DkQ9IXMVHMbxlUQGjtgC8SiQHAF2ydXKW0AvLM7KvEuokFbNs1xvtyIr5sQgti74LOIW63N3A4MCSx+gngUEmNJdUgtD6+nSh/AOghqYZC19ATCEkghIQ2OcNe6n37+L5dPI/kOe0hqV4+sXUiJEMzCwj/cGBy4jjJv6nX7Sme/OLeq6AfPAqsJP/rWey4inHOLo0ngLuoRx99lNWrV/Pyyy/z3nvv0a5dfv82OueccxVfTKQGA18SkqvU+j0IN+p9zGytmX1HbCFM7L7QzB4ys81mtp5w439vbKVYA9xAuPHPr3toauzZonzKFiXKAeaa2eNmlhdjbURIAooTY3EcTEhW/2Zmm8zsA+AN4OxYngvsL6m2mS03swmJ9Y2AvWMr0n/SkrUfmFkHM9utgOWyAuLKAVamrVsJ1NrBbc8HXoqfUcqVwFgz+7SAWFLOI7TYzU6s+wqYB3wDrCJ0t7wtUf4hIZFbRUiYxwPDYtl7wBGSjoxdW28kJDQ1Cjiv1OutzktSbeD/gFvNLP06IOnX8bz7AZjZauBj4CZJ1SR1JrSg1kjftwBvA1dJaiBpT8L1Yxv2T8nveo4A+kqqJWk/QuvfT+ot6pxd/ko1AZR0vMLA0ZmS+uZTfk3sTvCFpJHxFwBXCtavX8+tt97K5MnhR5+BAwcyZcoUunfv7o91cM45t6u7ntC6cH5a8rI3UBlYlGqpAh4jtJClzE+rqzEwN/F+LqEL4x4Kk4SkJr24EUiNPWuUT0yNEuUA36ZemNm6+DKnmDEWR2Ngfmx9TMaemsjkdOBEYK6kD2NXSQjjyWYC70qald/93g5aA9ROW1cbWL2920qqDpxJovunpMaEBObPxYjpPNK6jgKPANWAekBN4BViC2DsEvtOXFeTkNjvTujqiZlNIyRmD/Nj4j+FkCjmd16p1z+cVzyn4cAYM/tresCSDiaMNT0jrYW0J6F76vx4DkMTxy3KncBnwOfAfwkJbS5hrOO2yO96XgmsB2YQxoA+lx5XUefsClZqCWCiGf0EYH/gbKXNaET40nQxsw6Ebhd340qUmfHSSy/Rtm1bbrnlFt544w0A6tWrR/Xq1TMcnXPOOZdZko4k3PSfYWYr0ornEyZkqZ9oqaptZsluM+mtXQsJSVlKM0L3zMVmdkli4ov+hC6dCwjJSDKmLELCNbIYp1CcGItjIdA0bfxeM0KLFmY2zsxOJSSWw4AX4/rVZvYnM2tBGLd1jaSj8ztAHFeY36yrayQ9WkBck4EOie6uECZamZzPtl8BlRQmVUnpmM+23YFlwKjEugMJSfcUSd8SumweqDBz5g/jKSUdSkiWX0qrsyPwdBwPt5EwAcyBCjOM1gWaAg/HcYxLCWMMT0ztbGYvmVl7M6sH3Ez4Do1LXIOOacdaHOtBYUbQYYTP6uL0i6IwtvV14A9xPN8PzGyumXUzswZmdhAhgf0kvY78xDGrvc2sSfz8lwKfxlbqYinoesbr2NPM9ozf5axkXEWdsytcabYAHgjMjF0gNhEGw56a3MDM/p34FWsMYWYiV0ImTZrE0UcfzZlnnkmdOnX48MMPuf766zMdlnPOOVcmSGpEuD/pY2afpZdbmAjmXWCApNoKE7zsK+mI9G0TngOulrRPnGSjP/BCnNwjvX4DrgX+Iul3kqrHrnT/JLTy3FfUOWxnjFmxy19qqQqMBdYC10mqHBPjk4HnJVVRmLa/jpnlErox5sEPj7DYLyZoqfX5JgBm1s7SZl1NLJfktw8hScsDrlR47EDvuP6D9A3NbC2hle02STVjcnEqoYtgUvqEIxBa65oTZujsROgm+RnQKS2hOR94OXafTBoHnKcwIU9lwqyWC83sewuzjM4GLpVUSWG86fkkxvVJOkBStqQGhBbc4bFlEMLYuD9K2l9hZti/AE/H/SoTkqf1wHlpLbhIak/oTnmFmQ1Pv2YKE//Uip/xOYTxo/cmyqtIqkYYg1c5fl+yYlkThTGPii2MNxGS19S+leK+2UB23De9K3S+1zN+h+vFa3IC0Iswo2iR5+yKVpoJYBO27haxgB+7EeTnj2w9WPYHknopPEdk/JIlS0owxIrt2WefZeLEiQwaNIhPP/2Uww8/PNMhOeecc2XJRYSJQB4opEXqPMJ4rCnAcsKNZ35dNlOeJCQcHxFu+jcQJlTJl5m9QJiN8mpCl88pQHXg0FQLTzFsa4xnE26eU8vX8cf6Uwg9t74nPLbgvEQSci4wR9IqwuMKzonrWwLvE7op/g8YlDaJzg6JcZ1GOMcVhLFgp8X1qWf7Je8fLyNcv+8IyfilZvZDC6DCBCy/Im3Ckdgy921qIYyzy42vU/tWA87ip90VISTyGwhdFpcQWvd+kyjvDhwfy2YSWoWvTpQ/EM9vevx7USK2EYRecv8mdMudy4+J1iFAN0LitiLx/e0ay/9EmKDniURZskX0OMIkRssJn+vxFmb0THmX8B05BPhHfJ26odyX0PVzbbwmfc3s3cS+f4nb9yV8X9aTeExEEdfzAGASoZvrX4Geic+xqHN2RVAB43R3vGLpTOA4M7swvj8XONDMfvKPYPzFoTdwRGw2L1CXLl0su/vfeO3yw9izTrXSCL3cysvL45///Cf77rsvxxxzDGvWrGHTpk3UrVs306E559x2kfSpmXXJdBzOOedcRVGaLYALCP2dU/Yi9C/fiqRjCH3vTykq+XMF+89//kOXLl245JJLeP755wHIycnx5M8555xzzjn3g9JMAMcBLWMf+CqE6YhfT24QB6U+Rkj+tnXGIAfMnz+fs88+m8MPP5ylS5fy4osv8vjjj2c6LOecc84551wZVGoJYBzs3Jsw7e1U4EUzmyzpNkmnxM3+TpjC+F+SPpf0egHVbWVznpHlTy4A4O2332bYsGHcfPPNTJs2jTPPPNMf6+Ccc84555zLV6mNASwtnTofYHkn92fizceStQtmgWbGsGHD2LhxIz169CAvL49vvvmGZs2aZTo055wrcT4G0DnnnCtZpfog+NKwITePto1r75LJ35QpUzj22GPp3r07jz76KGZGdna2J3/OOeecc865Yil3CeCmzVtoUb9mpsPYqVasWMHVV19Nhw4dGD9+PA899BDvv/++d/V0zjnnnHPObZP0hzGWC7ta69/48eN58MEHueiii7jjjjuoX79+pkNyzjnnnHPOlUPlMgHcFfz3v/9l4sSJXHrppRxzzDHMmDGDFi1aZDos55xzzjnnXDlW7rqAVnQLFy7k3HPP5dBDD+Wee+5h48bwaERP/pxzzrmSJ2mOpPWS1kj6VtLTknLStjlE0geSVktaKWm4pP3Ttqkt6X5J82JdM+P7ArvtKJglaUoBcR2Ttu6C/9/evcdLVd13H/98uaoRMKK2gCAS7iCCoiFNXjYIVDFGjbfgJUhratJIKBoeL48GL1hF0mi0QYkYEE2Mt6pBKmIUTRsqCJSIQANS5Ik8eCGCoggInl//2OvEYZgDc5BzZjjzfb9e+8XZe62992+vmXOY36y115b0u5z1ZpKul/SapE1pnymSOu5pe9Q3Sc1TzBtT+1++m7q3S1oraYOkuyQ1zSl/UdKW1P4fSlqeU/Y1Sb+T9F46z2RJLYqJQ1JPSQvSOTdIei739Zc0Or2OG1Nst0tqklP+gqR1qfwVSafnlA2U9GqK611JT0hqV+DaD07H+F1+WSq/TlLkvmcknSvpPyV9JOnFAvv0lbQwlS+U1DenTJJuTTG9K2mC0r1HkrpK+nWKZ72kWZK65R27k6QZ6XfmT5Im5JSNTO25VdJ9ha7H6pYTwDKxdetWxo8fT9euXXn00Ue55pprWLx4Mc2bNy91aGZmZg3d1yPiQKAv0A+4urpA0peAZ4FfA22BI4FXgDmSOqU6zYDngV7AyUBL4K+Ad4Hjd3HeE4DDgE6SjtuDuB8DTgPOB1oBRwMLgUF7cKxSuR7oAhwBDASukHRyDXWvAvoDvYGuwDHAtXl1RkbEgWnJTUpaATeRvYY9gMPJHkdWTBxrgbOBg4FDyJ5r/VDOvk8Bx0REyxTb0cConPJ/BNqk8kuAX0hqk8qWASdFxEEptteAuwtc+61kj1XbiaQvpPjezCtaD/wEGF9gn2Zk7+lfAJ8HpgG/TttJcZ6RrqUPcCrwnVR2UGqDbsBfAC+nY+Ue+zfAbOAvydr6FzmnX0v2WkwpdD1W95wAlonXX3+dH/7whwwZMoRly5Zx00038bnPVdZkN2ZmZqUUEW+RPb+4b87mCcD9EXFHRHwQEesj4lpgLlnSADAc6AB8IyKWRURVRLwTEeMi4uldnPIisg/OT6efi5Z6eoYAp0fE/IjYHhHvR8TEiPh5bY5VYsOBcRGxISL+G5gMjKih7teBO9NrsA64E/i7Yk4SEQ9GxDMR8VFEbEjn+XIxcUTEexGxOrJnpwn4BOicc+z/iYj30qqAqrzyxen52AABNAXap7K3I2JtThw7HBv+/CVEb2BqDZf3U+BK4OO8a34uIh4hS7jyfZXsVrCfRMTWiLgzxX5iKr8I+HFErImI/w/8OKc9Xo6In6fXYRtwO9BNUuu07whgbUTcFhGbImJLRCzOievxiHiS7AsSKwEngCW0fPlybr31VgC6d+/OsmXLeOKJJzzc08zMrAQkHQ4MBVam9QPIevIeLVD9EbIEDGAw8ExEfFiLcx1A1mvzy7QMy+l9KcZg4OWIeKMW+9QZSVelYYwFlxr2+TxZr9crOZtfIetJLbhLWnLXD5fUKmfbLWnI4RxJX91FyCcAS2sTR7qOLcC/ADfnlZ0vaSPwJ7Jes5/llc+QtAWYB7wILMgp65COvRkYQ/alQ3VZY2AiMJIsedyBpHOAj3fzRUMhvYDFseMDwRfz6TX3ovjX5QTgrYioTugGAKslzUyvxYuSjqplfFaHnACWwMaNGxkzZgy9e/fmlltu4c03sx77Ll26lDgyMzOzivSkpA+AN4B3gOvS9oPJPivlD60jbau+v691DXV25UxgK9nw0hlkvTFfq8X+e3LOOhMR4yPioJqWGnarvtfy/Zxt7wMtCtQFmAn8o6RDJf0lnw6zPCD9eyXQCWgH3AM8lYZH7kDSELIerrG1iSNdRyuyZGxRXtmDaYhnV2AS8HZe+anpeKcAsyKiKqfsj+nYh5ANaf1Dzq6jgHkRsbDAdRxIloiOzi8rwoHseL2w4zXnl78PHFh9H2BODIeTJai5924eDgwj66FtfYXXMgAAEYlJREFUC/wbOw4vtRJzAliPqqqqmDp1Kl27duW2225jxIgRrFixgjZt2ux+ZzMzM6srZ0REC7Jhcd35NLHbQDacr9B/1G3IensgG8pW2//MLwIeSUM3twKPs+Mw0O1kQwVzNQW2fYZzlpvqHtOWOdtaAh/UUP+fyBKv3wP/CTxJ1h7vAETEvDRMd2tETAPmkCVcfyZpAPAgcHZErKhtHBGxiSzBu1/SYQXKXyPrWbyrQNm2iJgJnCTptALl6/n0XrwmktqSJYDX1NAeNwAPRMTrNZTvyofseL2w4zXnl7cEPsztMZR0KNkXGHdFxK9y6m4GfhcRMyPiY+Cfyb6w6LEHcVodcAJYjzZs2MDll19Op06dmD9/PpMnT+aww3b622FmZmYlEBG/Be4j+8Ba/WH/JeCcAtXPJZv4BeA5sg/1Rd28n3pNTgQuVDbj5Ftkw0FP0aezhv4R6Ji365HA/8s55/HpWCUn6f/q09k3d1oK7ZPuxXuTbMhktaNJQzML1N8cESMjol1EdCJLghdGxCc1hFV9z151jP3IJi/5u4h4/s+VahkH2efnA8h6GgtpAuzU81hkeROyiYFakk0g1AZYlt4jd5C95m+loaGDgFE576H2wCOSrtzFuastBfrk9ej14dNrXsou2iMNm30WmB4R/5R37MUUGK5q5cMJYB176623uOGGG6iqqqJ169bMmzePOXPmcOyxx5Y6NDMzM9vZT4AhOVPiXwVcJGmUpBaSPi/pJuBLZD0wAA+QDR/9V0ndJTWS1DolRafsfAq+Bawgm0Wxb1q6AmuA81Kdh4HR6XiS1J9swpOHIJvgg2ymxSckHZt6jFpI+q6koiZG2Zsi4uac2Td3Wnax6/3AtalduwN/T5aE70RSO0ltU3sMAH5IGq4r6SBJJ0naL7XFBWT3ps1K5b2BZ4DvR8RTtYlD0hBJ/SQ1ltQSuI2sd/i/U/m3q3sDlT0e4mrSlwPp9RsqaX9JTSVdmOL6bSo/U1K39J45NB17UeoNnEn2JUD1e2QsWQ9o35T0DiKbHKa6fC3ZTJ0T07EbS9qPLKlslNqmulf5RbIJZ0YpewTGyLR9dk57XF7d5sAPctqjZWrXORFxVYG2/AUwQNLglKiOJustr26vJimuxkDj6teswHGsrkTEPrUc2f2ouPrxxVHutm7dGj/60Y+iRYsW0axZs5g/f36pQzIz2+cAC6IM/u/x0nAXYDUwOG/b3cC/5qx/hewD84fARrJ7mnrn7dOKLHl8I9X7H7IP860LnPMPZIlI/vYrqt/zZF/SX0X2WICNZI8LuDivfjOyJHQlsImsd/BeoEOp27UW7d+c7HEAG8num7s8p6xDassOaf2E9Hp9BCwHLsipeygwn2wI43tks7QOySmfSjac98OcZWmRcZyTXrMPgXVks7b2yTv22+k1WE32eIn9UlkPsolfquOaTzZbbPW+3wdeT/u+RZbgH1FDW40gG1pZ1Hs51Y+85b6c8n5kjw3ZDPwX0C+nTGST0axPywRAqeyidKxNee3ZIWf/M9P7ciPZ706vnLLrC8R1fanfi5W0VL+Q+4xOPfrEsJt/yc3fKN/JhGbOnMno0aNZsWIFp556KrfddpsneDEz2wOSFkZE/1LHYWZm1lDsc92t72zcStNG2n3FEtm2bRujRo2iUaNGPP300wwdOrTUIZmZmZmZmQH74D2ABzRvzA9O6lbqMHbwwQcfMG7cODZt2kTTpk2ZOXMmr776qpM/MzMzMzMrK/tcAtikkWi5X/6syKVRVVXFAw88QLdu3Rg7diyzZs0CoHPnzjRr5kedmJmZmZlZednnEsBysWDBAr7yla8wfPhw2rdvz9y5cznzzDNLHZaZmZmZmVmN9rl7AMvFlVdeyapVq5g6dSrDhw+nUSPn0mZmZmZmVt6cABZp27ZtTJw4kXPOOYd27doxdepUDjroIFq2bFnq0MzMzMzMzIribqsiPPvssxx99NFcdtllPPjggwB06NDByZ+ZmZmZme1TnADuwqpVqzjjjDM46aST2LZtG0899RRjxowpdVhmZmZmZmZ7xENAd+Hmm2/mueeeY/z48YwePZrmzZuXOiQzMzMzM7M9pogodQy10rZzr1i7cmmdHDsieOihh+jevTv9+vVj3bp1bNu2jbZt29bJ+czMbNckLYyI/qWOw8zMrKHwENBk0aJFnHDCCZx//vlMmjQJgEMPPdTJn5mZmZmZNRgVnwCuW7eO73znOxx77LEsX76ce++9l7vvvrvUYZmZmZmZme11FZ8ATp48mSlTpjB69GhWrFjBxRdf7Gf6mZmZmZlZg1SR9wDOnj2bqqoqBg8ezObNm1m9ejU9evTYSxGamdne4nsAzczM9q6K6upavXo1Z511FoMGDeKWW24BYP/993fyZ2ZmZmZmFaEiEsCPPvqI6667jh49evDMM88wbtw4ZsyYUeqwzMzMzMzM6lVFPAdw+vTp3HjjjQwbNowJEybQvn37UodkZmZmZmZW7xpsArh48WJee+01zjrrLM4991y+8IUvcNxxx5U6LDMzMzMzs5JpcENA3333XS699FL69evHFVdcwfbt22nUqJGTPzMzMzMzq3gNJgHcvn07d911F127dmXSpEl873vfY/78+TRp0mA7Oc3MzMzMzGqlwWRHCxcu5NJLL2XgwIHccccdHHXUUaUOyczMzMzMrKzUaQ+gpJMlLZe0UtJVBcqbS3o4lc+T1LE2x3/jjTeYNm0aAF/84hd56aWXeP755538mZmZmZmZFVBnCaCkxsBEYCjQEzhPUs+8ahcDGyKiM3A7cGsxx968eTPjxo2jW7dujBw5kvXr1wMwYMAAJO21azAzMzMzM2tI6rIH8HhgZUSsioiPgYeA0/PqnA5MSz8/BgzSbjK4LZs+oGfPnowdO5ZTTz2VJUuWcPDBB+/14M3MzMzMzBqaurwHsB3wRs76GuCLNdWJiO2S3gdaA3+q6aAb3l5Du169mD17NgMHDtzLIZuZmZmZmTVcdZkAFurJiz2og6RLgEvS6tYlS5YsOfHEEz9jeBXjEHaRUNtO3F614/aqHbdX7XUrdQBmZmYNSV0mgGuA9jnrhwNra6izRlIToBWwPv9AEXEPcA+ApAUR0b9OIm6A3F614/aqHbdX7bi9ak/SglLHYGZm1pDU5T2A84Euko6U1AwYBkzPqzMduCj9fDYwOyJ26gE0MzMzMzOzz67OegDTPX0jgVlAY2BKRCyVdCOwICKmAz8HHpC0kqznb1hdxWNmZmZmZlbp6vRB8BHxNPB03raxOT9vAc6p5WHv2QuhVRK3V+24vWrH7VU7bq/ac5uZmZntRfKISzMzMzMzs8pQl/cAmpmZmZmZWRkp2wRQ0smSlktaKemqAuXNJT2cyudJ6lj/UZaPItrrcknLJC2W9LykI0oRZ7nYXXvl1DtbUkiq6Jkbi2kvSeem99hSSQ/Wd4zlpIjfxw6SXpC0KP1OnlKKOMuFpCmS3pG0pIZySboztediScfUd4xmZmYNRVkmgJIaAxOBoUBP4DxJPfOqXQxsiIjOwO3ArfUbZfkosr0WAf0jog/wGDChfqMsH0W2F5JaAKOAefUbYXkppr0kdQGuBr4cEb2A0fUeaJko8v11LfBIRPQjm/zqrvqNsuzcB5y8i/KhQJe0XALcXQ8xmZmZNUhlmQACxwMrI2JVRHwMPAScnlfndGBa+vkxYJCkQg+WrwS7ba+IeCEiPkqrc8mey1ipinl/AYwjS5S31GdwZaiY9vp7YGJEbACIiHfqOcZyUkx7BdAy/dyKnZ+RWlEi4t8p8AzYHKcD90dmLnCQpDb1E52ZmVnDUq4JYDvgjZz1NWlbwToRsR14H2hdL9GVn2LaK9fFwMw6jai87ba9JPUD2kfEjPoMrEwV8/7qCnSVNEfSXEm76s1p6Ippr+uBCyWtIZsp+fv1E9o+q7Z/48zMzKwGdfoYiM+gUE9e/nSlxdSpFEW3haQLgf7AX9dpROVtl+0lqRHZsOIR9RVQmSvm/dWEbHjeV8l6l/9DUu+IeK+OYytHxbTXecB9EfFjSV8iex5q74ioqvvw9kn+e29mZraXlGsP4Bqgfc764ew8ROrPdSQ1IRtGtashRA1ZMe2FpMHANcBpEbG1nmIrR7trrxZAb+BFSauBAcD0Cp4Iptjfx19HxLaIeB1YTpYQVqJi2uti4BGAiHgJ2A84pF6i2zcV9TfOzMzMdq9cE8D5QBdJR0pqRjZJwvS8OtOBi9LPZwOzo3Ifarjb9kpDGn9GlvxV8v1ZsJv2ioj3I+KQiOgYER3J7pk8LSIWlCbckivm9/FJYCCApEPIhoSuqtcoy0cx7fVHYBCApB5kCeC6eo1y3zIdGJ5mAx0AvB8Rb5Y6KDMzs31RWQ4BjYjtkkYCs4DGwJSIWCrpRmBBREwHfk42bGolWc/fsNJFXFpFttePgAOBR9NcOX+MiNNKFnQJFdlelhTZXrOAv5G0DPgE+D8R8W7poi6dItvrB8BkSZeRDWUcUcFfYCHpV2TDhw9J90VeBzQFiIhJZPdJngKsBD4C/rY0kZqZme37VMGfOczMzMzMzCpKuQ4BNTMzMzMzs73MCaCZmZmZmVmFcAJoZmZmZmZWIZwAmpmZmZmZVQgngGZmZmZmZhXCCaBVFEmfSPp9ztJxF3U7SlqyF875oqTlkl6RNEdStz04xnclDU8/j5DUNqfsXkk993Kc8yX1LWKf0ZIO+KznNjMzM7P64QTQKs3miOibs6yup/NeEBFHA9PInslYKxExKSLuT6sjgLY5Zd+OiGV7JcpP47yL4uIcDTgBNDMzM9tHOAG0ipd6+v5D0n+l5a8K1Okl6eXUa7hYUpe0/cKc7T+T1Hg3p/t3oHPad5CkRZJelTRFUvO0fbykZek8/5y2XS9pjKSzgf7AL9M59089d/0l/YOkCTkxj5D0L3sY50tAu5xj3S1pgaSlkm5I20aRJaIvSHohbfsbSS+ldnxU0oG7OY+ZmZmZ1SMngFZp9s8Z/vlE2vYOMCQijgG+CdxZYL/vAndERF+yBGyNpB6p/pfT9k+AC3Zz/q8Dr0raD7gP+GZEHAU0Af5B0sHAN4BeEdEHuCl354h4DFhA1lPXNyI25xQ/BpyZs/5N4OE9jPNk4Mmc9Wsioj/QB/hrSX0i4k5gLTAwIgZKOgS4Fhic2nIBcPluzmNmZmZm9ahJqQMwq2ebUxKUqynw03TP2ydA1wL7vQRcI+lw4PGIeE3SIOBYYL4kgP3JkslCfilpM7Aa+D7QDXg9Ilak8mnApcBPgS3AvZL+DZhR7IVFxDpJqyQNAF5L55iTjlubOD8HNAaOydl+rqRLyP5mtAF6Aovz9h2Qts9J52lG1m5mZmZmViacAJrBZcDbwNFkveJb8itExIOS5gFfA2ZJ+jYgYFpEXF3EOS6IiAXVK5JaF6oUEdslHQ8MAoYBI4ETa3EtDwPnAn8AnoiIUJaNFR0n8AowHpgInCnpSGAMcFxEbJB0H7BfgX0F/CYizqtFvGZmZmZWjzwE1AxaAW9GRBXwLbLerx1I6gSsSsMep5MNhXweOFvSYanOwZKOKPKcfwA6Suqc1r8F/DbdM9cqIp4mm2Cl0EycHwAtajju48AZwHlkySC1jTMitpEN5RyQho+2BDYB70v6C2BoDbHMBb5cfU2SDpBUqDfVzMzMzErECaBZNuPlRZLmkg3/3FSgzjeBJZJ+D3QH7k8zb14LPCtpMfAbsuGRuxURW4C/BR6V9CpQBUwiS6ZmpOP9lqx3Mt99wKTqSWDyjrsBWAYcEREvp221jjPdW/hjYExEvAIsApYCU8iGlVa7B5gp6YWIWEc2Q+mv0nnmkrWVmZmZmZUJRUSpYzAzMzMzM7N64B5AMzMzMzOzCuEE0MzMzMzMrEI4ATQzMzMzM6sQTgDNzMzMzMwqhBNAMzMzMzOzCuEE0MzMzMzMrEI4ATQzMzMzM6sQTgDNzMzMzMwqxP8CB6r6fSC+YzUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x648 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#chosen_rf_morgan.fit(X_train_morgan,y_train)\n",
    "morgan_stats = evaluate_model(chosen_rf_morgan, X_test_morgan, y_test)\n",
    "morgan_stats[\"Fingerprint\"] = \"Morgan\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## RDKit Fingerprints"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((829, 2048), (829,))"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test_rdkit.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x648 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#chosen_rf_rdkit.fit(X_train_rdkit,y_train)\n",
    "rdkit_stats = evaluate_model(chosen_rf_rdkit, X_test_rdkit, y_test)\n",
    "rdkit_stats[\"Fingerprint\"]=\"RDKit\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Comparison"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 256,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Test Set')"
      ]
     },
     "execution_count": 256,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x_col = ['Fingerprint']\n",
    "y_cols = ['mean_test_Precision', 'mean_test_Recall','mean_test_F1 Score', 'mean_test_AUC']\n",
    "stdev_cols = ['std_test_Precision', 'std_test_Recall','std_test_F1 Score', 'std_test_AUC']\n",
    "col_names = ['Precision','Recall','F1 Score','ROC AUC']\n",
    "\n",
    "# Stats for the best estimators\n",
    "\n",
    "# Training data (test split from 5-fold CV)\n",
    "# ------------------------------------------------------------------\n",
    "# Morgan\n",
    "_ = (df_morgan_opt['param_bootstrap'] == True) & \\\n",
    "    (df_morgan_opt['param_class_weight'] == 'balanced_subsample') & \\\n",
    "    (df_morgan_opt['param_criterion'] == 'entropy') & \\\n",
    "    (df_morgan_opt['param_max_features'] == 'sqrt') & \\\n",
    "    (df_morgan_opt['param_n_estimators'] == 350)\n",
    "training_stats = df_morgan_opt.loc[_].filter(x_col + y_cols)\n",
    "stdev = df_morgan_opt.loc[_].filter(stdev_cols)\n",
    "\n",
    "# RDKit\n",
    "_ = (df_rdkit_opt['param_bootstrap'] == True) & \\\n",
    "    (df_rdkit_opt['param_class_weight'] == 'balanced_subsample') & \\\n",
    "    (df_rdkit_opt['param_criterion'] == 'entropy') & \\\n",
    "    (df_rdkit_opt['param_max_features'] == 0.33) & \\\n",
    "    (df_rdkit_opt['param_n_estimators'] == 150)\n",
    "training_stats = training_stats.append(df_rdkit_opt[_].filter(x_col + y_cols))\n",
    "stdev = stdev.append(df_rdkit_opt[_].filter(stdev_cols))\n",
    "\n",
    "training_stats.columns = x_col + col_names\n",
    "stdev.columns = col_names\n",
    "\n",
    "# Now, plot the data\n",
    "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(15,6), sharey=True)\n",
    "\n",
    "# Training data\n",
    "training_stats.plot.bar(x='Fingerprint', y=col_names, yerr=stdev, ax=ax[0])\n",
    "ax[0].yaxis.grid(which=\"major\", color='k', linestyle='--', linewidth=0.3)\n",
    "ax[0].xaxis.set_tick_params(rotation=0)\n",
    "ax[0].legend(loc='best')\n",
    "ax[0].set_title(\"Training\")\n",
    "\n",
    "# Statistics obtained from the test set\n",
    "# Test data\n",
    "# ------------------------------------------------------------------\n",
    "pd.DataFrame(morgan_stats).append(pd.DataFrame(rdkit_stats)).drop('Zero-on', axis=1).plot.bar(x='Fingerprint', ax=ax[1])\n",
    "ax[1].xaxis.set_tick_params(rotation=0)\n",
    "ax[1].yaxis.grid(which=\"major\", color='k', linestyle='--', linewidth=0.3)\n",
    "ax[1].legend(loc='lower center')\n",
    "ax[1].set_title(\"Test Set\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Both fingerprints seem to give about the same result. However, the RDKit fingerprint seems a bit more robust on the testing set, so we should use this from now on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight='balanced_subsample',\n",
       "            criterion='entropy', max_depth=None, max_features=0.33,\n",
       "            max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "            min_impurity_split=None, min_samples_leaf=1,\n",
       "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "            n_estimators=150, n_jobs=4, oob_score=False, random_state=42,\n",
       "            verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 257,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chosen_rf_rdkit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 259,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
      "[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    5.0s finished\n"
     ]
    }
   ],
   "source": [
    "y_test_pred_rdkit = cross_val_predict(chosen_rf_rdkit,X_test_rdkit,y_test, verbose=1, cv=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "UNC-ML (Python 3.7.1)",
   "language": "python",
   "name": "unc"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "168px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}