predicting_TM_identity_Yaklich.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Final Project - BIOF509\n",
    "## Spring 2020 - Emily Yaklich\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Project Goal: Create a classification model that will be able to predict if a protein will be alpha helical transmembrane or not based on sequence features\n",
    "Will be able to compare this model to already existing transmembrane prediction models and build a foundation for the creation of more complex models (predicting specific helix residues or protein-protein interface residues)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Code Overview\n",
    "1. Obtain FASTA files with the transmembrane (TM) and non-transmemrane (non-TM) protein sequences\n",
    "2. Using the [iFeature](https://github.com/Superzchen/iFeature) python package, acquire sequence features in terms of Amino Acid Compositon and Grouped Dipeptide Composition\n",
    "3. Read in the data as dataframes and create one feature table for both TM and non-TM feature data\n",
    "4. Create and train a Random Forest Classifier model in order to classify the sequence as TM or non-TM\n",
    "5. Plot an ROC curve and confusion matrix to better understand the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Links to data\n",
    "- Transmembrane sequence dataset from the [PDBTM database](http://pdbtm.enzim.hu/?_=/download/files): http://pdbtm.enzim.hu/data/pdbtm_alpha_nr.seq (The downloaded data is the non-redundant alpha in the sequences column on 05/05/2020 with 971 sequences)\n",
    "- Non-transmembrane sequence dataset from the [TMHMM negative testing set](http://www.cbs.dtu.dk/~krogh/TMHMM/): http://www.cbs.dtu.dk/~krogh/TMHMM/PDBseqs.fasta (downloaded on 05/05/2020 with 645 sequences)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Use iFeature to extract sequence features from the protein sequences\n",
    "iFeature can be downloaded [here](https://github.com/Superzchen/iFeature)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "data_path = os.getcwd() + '/data'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Descriptor type: AAC\n"
     ]
    }
   ],
   "source": [
    "# TM sequences getting Amino Acid Composition features: \n",
    "!python iFeature/iFeature.py --file {data_path}/pdbtm_alpha_nr.seq.txt --type AAC --out {data_path}/TM_sequence_features_aac.tsv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Descriptor type: GDPC\n"
     ]
    }
   ],
   "source": [
    "# TM sequences getting Grouped Dipeptide Composition features:\n",
    "!python iFeature/iFeature.py --file {data_path}/pdbtm_alpha_nr.seq.txt --type GDPC --out {data_path}/TM_sequence_features_gdpc.tsv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Descriptor type: AAC\n"
     ]
    }
   ],
   "source": [
    "# Non-TM sequences getting Amino Acid Composition features:\n",
    "!python iFeature/iFeature.py --file {data_path}/PDBseqs.fasta_nonTM.txt --type AAC --out {data_path}/non_TM_sequence_features_aac.tsv "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Descriptor type: GDPC\n"
     ]
    }
   ],
   "source": [
    "# Non-TM sequences getting Grouped Dipeptide Composition features:\n",
    "!python iFeature/iFeature.py --file {data_path}/PDBseqs.fasta_nonTM.txt --type GDPC --out {data_path}/non_TM_sequence_features_gdpc.tsv "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Read in the sequence feature data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "# read in the transmembrane data\n",
    "TM_aac = pd.read_csv(data_path +'/TM_sequence_features_aac.tsv', sep='\\t')\n",
    "TM_gdpc = pd.read_csv(data_path + '/TM_sequence_features_gdpc.tsv', sep='\\t')\n",
    "# merge the two dataframes \n",
    "TM_data = pd.merge(TM_aac, TM_gdpc, on='#')\n",
    "# add a 'tm identity' column that will indicate whether a protein is TM or not (2 means it is TM)\n",
    "TM_data['tm identity'] = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "non_TM_aac = pd.read_csv(data_path + '/non_TM_sequence_features_aac.tsv', sep='\\t')\n",
    "non_TM_gdpc = pd.read_csv(data_path +'/non_TM_sequence_features_gdpc.tsv', sep='\\t')\n",
    "# merge the two dataframes\n",
    "non_TM_data = pd.merge(non_TM_aac, non_TM_gdpc, on='#')\n",
    "# add a 'tm identity' column that will indicated whether a protein is TM or not (1 means it is non-TM)\n",
    "non_TM_data['tm identity'] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# combine the two dataframes to create the dataset containing all features\n",
    "all_seq_features = TM_data.append(non_TM_data)\n",
    "all_seq_features = all_seq_features.set_index('#')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize first five rows of data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>A</th>\n",
       "      <th>C</th>\n",
       "      <th>D</th>\n",
       "      <th>E</th>\n",
       "      <th>F</th>\n",
       "      <th>G</th>\n",
       "      <th>H</th>\n",
       "      <th>I</th>\n",
       "      <th>K</th>\n",
       "      <th>L</th>\n",
       "      <th>...</th>\n",
       "      <th>negativecharger.aromatic</th>\n",
       "      <th>negativecharger.postivecharger</th>\n",
       "      <th>negativecharger.negativecharger</th>\n",
       "      <th>negativecharger.uncharger</th>\n",
       "      <th>uncharger.alphaticr</th>\n",
       "      <th>uncharger.aromatic</th>\n",
       "      <th>uncharger.postivecharger</th>\n",
       "      <th>uncharger.negativecharger</th>\n",
       "      <th>uncharger.uncharger</th>\n",
       "      <th>tm identity</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>#</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3a0b_Z</th>\n",
       "      <td>0.096774</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.032258</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.080645</td>\n",
       "      <td>0.064516</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.080645</td>\n",
       "      <td>0.016129</td>\n",
       "      <td>0.177419</td>\n",
       "      <td>...</td>\n",
       "      <td>0.016393</td>\n",
       "      <td>0.016393</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.081967</td>\n",
       "      <td>0.032787</td>\n",
       "      <td>0.016393</td>\n",
       "      <td>0.016393</td>\n",
       "      <td>0.032787</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5a1s_A</th>\n",
       "      <td>0.107143</td>\n",
       "      <td>0.013393</td>\n",
       "      <td>0.026786</td>\n",
       "      <td>0.035714</td>\n",
       "      <td>0.055804</td>\n",
       "      <td>0.102679</td>\n",
       "      <td>0.008929</td>\n",
       "      <td>0.109375</td>\n",
       "      <td>0.040179</td>\n",
       "      <td>0.118304</td>\n",
       "      <td>...</td>\n",
       "      <td>0.004474</td>\n",
       "      <td>0.008949</td>\n",
       "      <td>0.008949</td>\n",
       "      <td>0.004474</td>\n",
       "      <td>0.098434</td>\n",
       "      <td>0.017897</td>\n",
       "      <td>0.022371</td>\n",
       "      <td>0.020134</td>\n",
       "      <td>0.038031</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4a2n_B</th>\n",
       "      <td>0.036082</td>\n",
       "      <td>0.005155</td>\n",
       "      <td>0.025773</td>\n",
       "      <td>0.056701</td>\n",
       "      <td>0.082474</td>\n",
       "      <td>0.056701</td>\n",
       "      <td>0.020619</td>\n",
       "      <td>0.118557</td>\n",
       "      <td>0.087629</td>\n",
       "      <td>0.113402</td>\n",
       "      <td>...</td>\n",
       "      <td>0.015544</td>\n",
       "      <td>0.005181</td>\n",
       "      <td>0.020725</td>\n",
       "      <td>0.015544</td>\n",
       "      <td>0.067358</td>\n",
       "      <td>0.041451</td>\n",
       "      <td>0.031088</td>\n",
       "      <td>0.010363</td>\n",
       "      <td>0.025907</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6a2j_A</th>\n",
       "      <td>0.090615</td>\n",
       "      <td>0.022654</td>\n",
       "      <td>0.009709</td>\n",
       "      <td>0.029126</td>\n",
       "      <td>0.077670</td>\n",
       "      <td>0.067961</td>\n",
       "      <td>0.032362</td>\n",
       "      <td>0.106796</td>\n",
       "      <td>0.032362</td>\n",
       "      <td>0.148867</td>\n",
       "      <td>...</td>\n",
       "      <td>0.006494</td>\n",
       "      <td>0.003247</td>\n",
       "      <td>0.003247</td>\n",
       "      <td>0.016234</td>\n",
       "      <td>0.113636</td>\n",
       "      <td>0.035714</td>\n",
       "      <td>0.016234</td>\n",
       "      <td>0.009740</td>\n",
       "      <td>0.048701</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6a2w_A</th>\n",
       "      <td>0.089820</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.065868</td>\n",
       "      <td>0.083832</td>\n",
       "      <td>0.083832</td>\n",
       "      <td>0.119760</td>\n",
       "      <td>0.011976</td>\n",
       "      <td>0.053892</td>\n",
       "      <td>0.035928</td>\n",
       "      <td>0.113772</td>\n",
       "      <td>...</td>\n",
       "      <td>0.024096</td>\n",
       "      <td>0.012048</td>\n",
       "      <td>0.024096</td>\n",
       "      <td>0.036145</td>\n",
       "      <td>0.096386</td>\n",
       "      <td>0.018072</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.012048</td>\n",
       "      <td>0.042169</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 46 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               A         C         D         E         F         G         H  \\\n",
       "#                                                                              \n",
       "3a0b_Z  0.096774  0.000000  0.032258  0.000000  0.080645  0.064516  0.000000   \n",
       "5a1s_A  0.107143  0.013393  0.026786  0.035714  0.055804  0.102679  0.008929   \n",
       "4a2n_B  0.036082  0.005155  0.025773  0.056701  0.082474  0.056701  0.020619   \n",
       "6a2j_A  0.090615  0.022654  0.009709  0.029126  0.077670  0.067961  0.032362   \n",
       "6a2w_A  0.089820  0.000000  0.065868  0.083832  0.083832  0.119760  0.011976   \n",
       "\n",
       "               I         K         L  ...  negativecharger.aromatic  \\\n",
       "#                                     ...                             \n",
       "3a0b_Z  0.080645  0.016129  0.177419  ...                  0.016393   \n",
       "5a1s_A  0.109375  0.040179  0.118304  ...                  0.004474   \n",
       "4a2n_B  0.118557  0.087629  0.113402  ...                  0.015544   \n",
       "6a2j_A  0.106796  0.032362  0.148867  ...                  0.006494   \n",
       "6a2w_A  0.053892  0.035928  0.113772  ...                  0.024096   \n",
       "\n",
       "        negativecharger.postivecharger  negativecharger.negativecharger  \\\n",
       "#                                                                         \n",
       "3a0b_Z                        0.016393                         0.000000   \n",
       "5a1s_A                        0.008949                         0.008949   \n",
       "4a2n_B                        0.005181                         0.020725   \n",
       "6a2j_A                        0.003247                         0.003247   \n",
       "6a2w_A                        0.012048                         0.024096   \n",
       "\n",
       "        negativecharger.uncharger  uncharger.alphaticr  uncharger.aromatic  \\\n",
       "#                                                                            \n",
       "3a0b_Z                   0.000000             0.081967            0.032787   \n",
       "5a1s_A                   0.004474             0.098434            0.017897   \n",
       "4a2n_B                   0.015544             0.067358            0.041451   \n",
       "6a2j_A                   0.016234             0.113636            0.035714   \n",
       "6a2w_A                   0.036145             0.096386            0.018072   \n",
       "\n",
       "        uncharger.postivecharger  uncharger.negativecharger  \\\n",
       "#                                                             \n",
       "3a0b_Z                  0.016393                   0.016393   \n",
       "5a1s_A                  0.022371                   0.020134   \n",
       "4a2n_B                  0.031088                   0.010363   \n",
       "6a2j_A                  0.016234                   0.009740   \n",
       "6a2w_A                  0.000000                   0.012048   \n",
       "\n",
       "        uncharger.uncharger  tm identity  \n",
       "#                                         \n",
       "3a0b_Z             0.032787            2  \n",
       "5a1s_A             0.038031            2  \n",
       "4a2n_B             0.025907            2  \n",
       "6a2j_A             0.048701            2  \n",
       "6a2w_A             0.042169            2  \n",
       "\n",
       "[5 rows x 46 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_seq_features.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualize shape of the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1616, 46)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_seq_features.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Look for missing observations in the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "A                                  0\n",
       "C                                  0\n",
       "D                                  0\n",
       "E                                  0\n",
       "F                                  0\n",
       "G                                  0\n",
       "H                                  0\n",
       "I                                  0\n",
       "K                                  0\n",
       "L                                  0\n",
       "M                                  0\n",
       "N                                  0\n",
       "P                                  0\n",
       "Q                                  0\n",
       "R                                  0\n",
       "S                                  0\n",
       "T                                  0\n",
       "V                                  0\n",
       "W                                  0\n",
       "Y                                  0\n",
       "alphaticr.alphaticr                0\n",
       "alphaticr.aromatic                 0\n",
       "alphaticr.postivecharger           0\n",
       "alphaticr.negativecharger          0\n",
       "alphaticr.uncharger                0\n",
       "aromatic.alphaticr                 0\n",
       "aromatic.aromatic                  0\n",
       "aromatic.postivecharger            0\n",
       "aromatic.negativecharger           0\n",
       "aromatic.uncharger                 0\n",
       "postivecharger.alphaticr           0\n",
       "postivecharger.aromatic            0\n",
       "postivecharger.postivecharger      0\n",
       "postivecharger.negativecharger     0\n",
       "postivecharger.uncharger           0\n",
       "negativecharger.alphaticr          0\n",
       "negativecharger.aromatic           0\n",
       "negativecharger.postivecharger     0\n",
       "negativecharger.negativecharger    0\n",
       "negativecharger.uncharger          0\n",
       "uncharger.alphaticr                0\n",
       "uncharger.aromatic                 0\n",
       "uncharger.postivecharger           0\n",
       "uncharger.negativecharger          0\n",
       "uncharger.uncharger                0\n",
       "tm identity                        0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_seq_features.isnull().sum()\n",
    "\n",
    "# there are no missing oberservations in the dataset!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Problem Statement\n",
    "Transmembrane proteins have different properties and functions compared to soluble proteins, so it is important to be able to identify transmembrane proteins from soluble proteins. Currently transmembrane prediction algorithms exist, but is it possible to create my own model that can reproduce these predictions? Can I build a model to differentiate a transmembrane protein from a soluble protein from sequence information alone?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### y-variable \n",
    "The y-variable or labels or target prediction is the transmembrane identity of a protein sequence. In this dataset, it will the the 'tm identity' column with either a 2 for a transmembrane identity or 1 for a non-transmembrane identity"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Build and test a Random Forest Classifier Model with tuned hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# separate the target labels and the rest of the features in the dataset\n",
    "import numpy as np\n",
    "# labels are the values we will to predict (TM or non-TM)\n",
    "labels = np.array(all_seq_features['tm identity'])\n",
    "# remove the labels from the features\n",
    "# axis 1 refers to the columns\n",
    "features= all_seq_features.drop('tm identity', axis = 1)\n",
    "# save feature names\n",
    "feature_list = list(features.columns)\n",
    "# convert df to numpy array\n",
    "features = np.array(features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Squared Error: 0.11\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "# split the data into training and testing sets\n",
    "# X_train = training features, y_train = training labels, X_test = testing features, y_test = testing labels\n",
    "X_train, X_test, y_train, y_test = train_test_split(features, \n",
    "                                            labels, test_size = 0.3, random_state = 42)\n",
    "\n",
    "rf = RandomForestClassifier(random_state=42)\n",
    "\n",
    "# hyperparameter grid\n",
    "param_grid = {'max_features': ['auto', None, 'log2'],'n_estimators': [100, 500, 1000]}\n",
    "\n",
    "# instantiate the grid search model\n",
    "grid_search = GridSearchCV(estimator = rf, param_grid = param_grid)\n",
    "\n",
    "grid_search.fit(X_train, y_train)\n",
    "# save the optimized \n",
    "best_tuned_model = grid_search.best_estimator_\n",
    "\n",
    "# fit \n",
    "best_tuned_model.fit(X_train, y_train)\n",
    "\n",
    "# predict \n",
    "model_predictions = best_tuned_model.predict(X_test)\n",
    "\n",
    "# calculate mean predicted class probabilities \n",
    "model_probs = best_tuned_model.predict_proba(X_test)[:, 1]\n",
    "\n",
    "# test\n",
    "mse = mean_squared_error(y_test, model_predictions)\n",
    "print('Mean Squared Error:', round(mse, 2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Accuracy of the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 93.09 %.\n"
     ]
    }
   ],
   "source": [
    "# calculate the absolute errors\n",
    "errors = abs(model_predictions - y_test)\n",
    "# calculate mean absolute percentage error (MAPE)\n",
    "mape = 100 * (errors / y_test)\n",
    "# calculate and display accuracy\n",
    "accuracy = 100 - np.mean(mape)\n",
    "print('Accuracy:', round(accuracy, 2), '%.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average number of nodes: 193\n",
      "Average maximum depth: 12\n"
     ]
    }
   ],
   "source": [
    "# find the average number of nodes and the depth for the decision trees \n",
    "n_nodes = []\n",
    "max_depths = []\n",
    "for ind_tree in best_tuned_model.estimators_:\n",
    "    n_nodes.append(ind_tree.tree_.node_count)\n",
    "    max_depths.append(ind_tree.tree_.max_depth)\n",
    "print(f'Average number of nodes: {int(np.mean(n_nodes))}')\n",
    "print(f'Average maximum depth: {int(np.mean(max_depths))}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ROC Curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The AUC is: 0.9\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, roc_auc_score\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# calculate false positive rates and true positive rates\n",
    "# the base rates are without the model\n",
    "base_fpr, base_tpr, _ = roc_curve(y_test, [1 for _ in range(len(y_test))],pos_label=2)\n",
    "model_fpr, model_tpr, _ = roc_curve(y_test, model_probs, pos_label=2)\n",
    "\n",
    "# plot the true postive and false positive rates as an ROC curve \n",
    "plt.figure(figsize = (8, 6))\n",
    "plt.rcParams['font.size'] = 16    \n",
    "plt.plot(base_fpr, base_tpr, 'green', label = 'baseline')\n",
    "plt.plot(model_fpr, model_tpr, 'b', label = 'model')\n",
    "plt.legend();\n",
    "plt.xlabel('False Positive Rate'); plt.ylabel('True Positive Rate'); plt.title('ROC Curve');\n",
    "roc_auc = roc_auc_score(y_test, model_predictions)\n",
    "print('The AUC is:', round(roc_auc,2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The true positive rate (TPR) is plotted against the false positive rate (FPR) for the baseline and the model. The baseline (green line) describes the rates with no predictive model, so where TPR=FDR. We can see that the predictive value increases when the model is applied to the data (blue line). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[177  15]\n",
      " [ 37 256]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, plot_confusion_matrix\n",
    "\n",
    "# create a confusion matrix\n",
    "conf_mat = confusion_matrix(y_test, model_predictions)\n",
    "print(conf_mat)\n",
    "\n",
    "# create a confusion matrix plot\n",
    "labels = ['non-TM', 'TM']\n",
    "plot = plot_confusion_matrix(best_tuned_model, X_test, y_test,\n",
    "display_labels=labels, values_format='.3g', cmap=plt.cm.Greens)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Confusion matrix illustrates the true positive (TP), false positive (FP), true negative (TN) and false negative (FN) predictions in from the test dataset. The top left corner shows the proteins predicted as non-TM and are truly non-TM and the bottom right corner shows the proteins predicted as TM and are truly TM. The bottom left corner displays the proteins predicted as non-TM but are truly TM and the top right corner shows the proteins that are predicted as TM but are truly non-TM. We can see that the true positive and true negative predictions are much greater than the false predictions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Importances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA9cAAAI0CAYAAAAX0NqxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOzde7ytU7348c+3vd0rNrZbYduR2orKJpeiVAiRWyq33IkcSRcdVDuqcxRSlEtJdKJEdKXc+uXQCeHYKgmHQrZLlDt7/P4YY7aeNfdca801n7nmXnPtz/v1mq+51jOfMZ7xzOc2vmOM55mRUkKSJEmSJHXuRfO7AJIkSZIk9TuDa0mSJEmSajK4liRJkiSpJoNrSZIkSZJqMriWJEmSJKkmg2tJkiRJkmqaPL8LMNEsu+yyadq0afO7GJIkSZKkLrvhhhseSilNbfWZwXWXTZs2jeuvv35+F0OSJEmS1GUR8X9DfeawcEmSJEmSajK4liRJkiSpJoNrSZIkSZJqMriWJEmSJKkmg2tJkiRJkmoyuJYkSZIkqSaDa0mSJEmSajK4liRJkiSpJoNrSZIkSZJqMriWJEmSJKkmg2tJkiRJkmoyuJYkSZIkqSaDa0mSJEmSapo8vwswUc352rkdp5160G5dLIkkSZIkaazZcy1JkiRJUk19FVxHxMoRcUFEPBYRj0fEhRGxSptpPxcRl0XEwxGRIuIDw8y7X0T8ISKeiYg/RsSBXVsJSZIkSdKE0zfBdUQsDlwBvArYE9gdWAO4MiKWaCOLDwGLAT8eYTn7AacBPwC2BL4PnBoRB3VeekmSJEnSRNZP91zvB0wH1kwp3QEQEbcAfwIOAE4YIf2SKaW5EbE6sEerGSJiMnAccE5K6d/L5CsjYiXgsxFxZkrpuS6siyRJkiRpAumbnmtgW+C6RmANkFK6C7gG2G6kxCmluW0sY0NgKtD8NLJzgGWAN7VdWkmSJEnSAqOfguu1gFtbTJ8NzOjiMmixnNnlvVvLkSRJkiRNIP0UXC8NPNpi+iPAlC4ugxbLeaTpc0mSJEmS/qWfgmuA1GJadDH/Rl6tljN0ooj9I+L6iLh+zpw5XSyOJEmSJKkf9FNw/Site46n0LpHuxND9VAv3fT5ICml01NKM1NKM6dOndqlokiSJEmS+kU/BdezGbgnumoGcFsXl0GL5TTute7WciRJkiRJE0g/BdeXABtExPTGhIiYBmxcPuuGa4GHgF2bpu9G7rW+pkvLkSRJkiRNIP30O9dnAIcAF0fEUeT7oj8L3Auc1pgpIlYF/gzMSinNqkzflPwzWyuUSTMj4p8AKaULyvtzEXE0cGpE/BX4JbAZsDfwoZTSs2O7ipIkSZKkftQ3wXVK6YmI2Aw4kfy70wFcDhyWUvpnZdYAJjFvr/xngE0r/x9cXo00jeV8PSIS8BHgo8A9wCEppVO7uDqSJEmSpAmkb4JrgJTSPcCOI8xzNy2eIJ5SessolnMald5wSZIkSZKG00/3XEuSJEmSNC4ZXEuSJEmSVJPBtSRJkiRJNRlcS5IkSZJUk8G1JEmSJEk1GVxLkiRJklSTwbUkSZIkSTUZXEuSJEmSVJPBtSRJkiRJNRlcS5IkSZJUk8G1JEmSJEk1GVxLkiRJklSTwbUkSZIkSTUZXEuSJEmSVJPBtSRJkiRJNRlcS5IkSZJUk8G1JEmSJEk1GVxLkiRJklSTwbUkSZIkSTVNnt8F0MjmfP20WumnHnhAl0oiSZIkSWrFnmtJkiRJkmoyuJYkSZIkqSaDa0mSJEmSajK4liRJkiSpJoNrSZIkSZJqMriWJEmSJKkmg2tJkiRJkmoyuJYkSZIkqSaDa0mSJEmSajK4liRJkiSpJoNrSZIkSZJqMriWJEmSJKkmg2tJkiRJkmoyuJYkSZIkqSaDa0mSJEmSajK4liRJkiSpJoNrSZIkSZJqMriWJEmSJKkmg2tJkiRJkmoyuJYkSZIkqSaDa0mSJEmSajK4liRJkiSpJoNrSZIkSZJqMriWJEmSJKkmg2tJkiRJkmoyuJYkSZIkqSaDa0mSJEmSajK4liRJkiSppr4KriNi5Yi4ICIei4jHI+LCiFilzbSLRsTxEXF/RDwVEddGxCYt5lsmIr4cEXeW+e6KiK9GxNTur5EkSZIkaSKYPL8L0K6IWBy4AngG2BNIwLHAlRGxdkrpiRGy+AawNfBR4E7gYODSiNgwpXRTWUYAlwCvBI4Bfg/MAD4LrBsRG6WUUtdXTpIkSZLU1/omuAb2A6YDa6aU7gCIiFuAPwEHACcMlTAi1gHeD+ydUjqrTLsamA3MArYts64BbAQckFI6vUy7KiLmAl8jB91/7PJ6SZIkSZL6XD8NC98WuK4RWAOklO4CrgG2ayPtc8D5lbTPA+cBW0TEImXywuX98ab0fy/v/fR9SZIkSZJ6pJ+CxbWAW1tMn00euj1S2rtSSk+2SLswsHrl/18BR0fEzIh4cUSsTx4i/rOU0u87Lr0kSZIkacLqp+B6aeDRFtMfAabUSNv4nHI/9Vbkod+/Bf4B/IZ8j/aOQ2UeEftHxPURcf2cOXNGKIokSZIkaaLpp+Aa8kPMmkUb6WIUac8ANgAOBDYt7zOBCyKi5feVUjo9pTQzpTRz6lQfKi5JkiRJC5p+eqDZo5Qe5iZTaN0rXfUI0Oonu6ZUPicitgbeB7w9pXR5+exXEXEncBnwLuDiUZZbkiRJkjTB9VPP9WzyvdPNZgC3tZF2tfJzXs1pnwUaD0l7bXn/bdN8/1PeX91eUSVJkiRJC5J+Cq4vATaIiOmNCRExDdi4fDZS2oWAnStpJwO7AJellJ4pkx8o7+s3pX9jef9rJwWXJEmSJE1s/RRcnwHcDVwcEdtFxLbkIdr3Aqc1ZoqIVSPi+Yg4pjEtpXQT+We4ToqIfSPibeSf4VoN+FRlGRcC9wHfjoiDIuKtEXEQ8O2ynIvGdA0lSZIkSX2pb4LrlNITwGbA7cA5wHeAu4DNUkr/rMwawCTmXbe9gLOAY4GfACsDW6aUbqws43Hyw8x+Bnys8v4jYMOm5UiSJEmSBPTXA81IKd3DMD+JVea5mxZPAU8pPQUcXl7Dpb8X2KfzUkqSJEmSFjR903MtSZIkSdJ4ZXAtSZIkSVJNBteSJEmSJNVkcC1JkiRJUk0G15IkSZIk1WRwLUmSJElSTQbXkiRJkiTVZHAtSZIkSVJNBteSJEmSJNVkcC1JkiRJUk0G15IkSZIk1WRwLUmSJElSTQbXkiRJkiTVZHAtSZIkSVJNBteSJEmSJNVkcC1JkiRJUk0G15IkSZIk1WRwLUmSJElSTQbXkiRJkiTVZHAtSZIkSVJNBteSJEmSJNVkcC1JkiRJUk0G15IkSZIk1WRwLUmSJElSTQbXkiRJkiTVZHAtSZIkSVJNBteSJEmSJNVkcC1JkiRJUk0G15IkSZIk1WRwLUmSJElSTT0LriPi9RFxYUQ8FBHPR8QbyvTPRcSWvSqHJEmSJEnd1pPgOiLeBFwLvAr4r6blzgUO7EU5JEmSJEkaC73quf4CcCmwFnB402c3Am/oUTkkSZIkSeq6yT1azhuAHVJKKSJS02cPAVN7VA5JkiRJkrquVz3XTwOLD/HZisBjPSqHJEmSJEld16vg+tfAYRExqTKt0YO9D3BFj8ohSZIkSVLX9WpY+NHANcDNwAXkwHrPiDgBWBdYr0flkCRJkiSp63rSc51SuhnYBPgb8O9AAIeUjzdNKf2xF+WQJEmSJGks9KrnmpTSjcDbImJRYGng7ymlJ3u1fA3429f+o1b65Q/6eJdKIkmSJEkTQ0+C64hYCFg4pfRESulp4L7KZ0sAz6aUnutFWSRJkiRJ6rZe9VyfCSwEvL/FZ6cBzwJ796gskiRJkiR1Va+C67cCHx3is0uA43tUDo2B+045vFb6lQ4+oUslkSRJkqT5o1c/xbUc8OAQn80Blu9ROSRJkiRJ6rpeBdcPAq8d4rPXAg/3qBySJEmSJHVdr4LrHwNHR8Ta1YkR8VryT3P9qEflkCRJkiSp63p1z/UxwDuAGyLit8BfgJcB6wN3AUf1qBySJEmSJHVdT3quU0oPAesBnwcCeF15Pw5Yr3wuSZIkSVJf6lXPNSmlv5N7sI/p1TIlSZIkSeqFXt1z3RURsXJEXBARj0XE4xFxYUSs0mbaRSPi+Ii4PyKeiohrI2KTIeZ9WUR8MyIeiIhnIuKuiPh8d9dGkiRJkjRR9KznOiL2BN4HrAIs2vRxSim9YoT0iwNXAM8AewIJOBa4MiLWTik9MUIRvgFsTf697TuBg4FLI2LDlNJNleVMA64h3wt+KPA3YBqw+ogrKUmSJElaIPUkuI6Io4HPALcCN5ED5NHaD5gOrJlSuqPkewvwJ+AA4IRhlr8O8H5g75TSWWXa1cBsYBawbWX2rwN/Bd6aUnquTLu6g/JKkiRJkhYQveq53gf4ckrpwzXy2Ba4rhFYA6SU7oqIa4DtGCa4LmmfA86vpH0+Is4DPhERi6SUnomIVwBbAHtUAmtJkiRJkobVq3uul6H+b1mvRe75bjYbmNFG2rtSSk+2SLswA0O+Ny7vT0XEL8r91o9GxLcjYplOCy5JkiRJmth6FVxfDaxTM4+lgUdbTH8EmFIjbeNzgJXK+zeB24F3Ah8n36t9aUS0/L4iYv+IuD4irp8zZ84IRZEkSZIkTTS9GhZ+GHBhRDwM/JSBoPZfUkpz28gntZgWbaSLNtM2guerUkoHl7+viIjHgPPIQ8Z/Nk+hUjodOB1g5syZrZYjSZIkSZrAehVc317ezxri89RGWR5loIe5agqte6WrHiE/pbxV2sbnAA+X9180zXdZeX89LYJrSZIkSdKCrVfB9Sxa9xyPxmzyvdPNZgC3tZF2+4hYvOm+6xnAs8Adlflg6LK207suSZIkSVrA9CS4Til9ugvZXAJ8MSKmp5TuhH/9JvXGwCfaSPsZYGfg7JJ2MrALcFlKqfHTYNcBDwBbAl+tpN+yvP+29lpIkiRJkiacXj3QrBvOAO4GLo6I7SJiW+Bi4F7gtMZMEbFqRDwfEcc0pqWUbiL/DNdJEbFvRLyNfA/1asCnKvM9Tw7Ut46Ir0fE5hHxQeBU4CrgijFeR0mSJElSH+rVsHAiYmHy07fXBBZt+jillD47XPqU0hMRsRlwInAO+WFklwOHpZT+WV0UMIl5Gw72Ao4DjgWWAm4Gtkwp3di0nLMjYi75KeF7ke/HPhc4MqXkw8okSZIkSfPoSXAdESsBvwamke9nbjyluxqsDhtcA6SU7gF2HGGeu2nxBPGU0lPA4eU10nLOIQfwkiRJkiSNqFfDwo8H5pCf2B3AG4Hp5J7kO8rfkiRJkiT1pV4NC38zcARwX/l/bulhPiYiJgEnA9v1qCySJEmSJHVVr3qulwHuSynNBZ5g4PelIT8k7C09KockSZIkSV3Xq+D6L8Cy5e8/A5tXPlsfeLpH5ZAkSZIkqet6NSz8SmBT4Ifkn806JSJeBzwHbEHlp7QkSZIkSeo3vQqujwKWBkgpfS0iJgO7AIsD/wnM6lE5JEmSJEnqup4E1ymlh4CHKv9/BfhKL5YtSZIkSdJY68k91xFxRUS8aojPXhkRV/SiHJIkSZIkjYVePdDsLcBLh/jsJeT7sSVJkiRJ6ku9uucaIA0x/RXAP3tYDo1zd33l3bXSr/ahH3apJJIkSZLUnjELriNiL2Cv8m8CTo+IfzTNthjwGuDysSqHJEmSJEljbSyHhc8FXiivaPq/8XoY+BqwzxiWQ5IkSZKkMTVmPdcppbOBswEi4krgoJTSH8ZqeZIkSZIkzS9j/kCziFgYWBJYZayXJUmSJEnS/DDmwXVK6VlgNeD5sV6WJEmSJEnzQ69+iusXwOY9WpYkSZIkST3Vq5/i+gpwbkRMBn4I3E/TT3OllO7sUVkkSZIkSeqqXgXXV5f3w4EPDzHPpB6VRZIkSZKkrupVcL3XyLNIkiRJktSfehJcl5/lkiRJkiRpQupVzzUAERHADGBp4GHg9ymlNHwqSZIkSZLGt149LZyI2Jf8ILNbgKuA/wXui4h9elUGSZIkSZLGQk96riNiV+B04HLgXOABYAVgV+D0iHgypfTdXpRFkiRJkqRu69Ww8I8B30kp7d40/eyIOAf4OGBwLUmSJEnqS70aFr4muce6lXPL55IkSZIk9aVeBdf/AF4+xGcvL59LkiRJktSXehVc/wz4XES8uToxIjYEji2fS5IkSZLUl3p5z/UGwFUR8VfyU8NXIPda31E+lyRJkiSpL/UkuE4pPRARrwP2Bt5M/p3ru4GrgW+llJ7sRTkkSZIkSRoLveq5pgTQXy0vSZIkSZImjJ4F1wARsTqwPvAy4C/Ab1NKd/SyDJIkSZIkdVtPguuIWBQ4FdgdmFT56IWIOBs4OKX0TC/KIkmSJElSt/XqaeFfBHYFPgWsDrykvH+aHHAf36NySJIkSZLUdb0aFv5e4DMppc9Vpt0JHBcRAB8GDu1RWSRJkiRJ6qpe9VwvAvzPEJ/9Bli4R+WQJEmSJKnrehVc/xLYfIjPNgeu6FE5JEmSJEnqul4NCz8BOCcilgC+D/wNWB54D7AVsFtETG/MnFK6s0flkiRJkiSptl4F11eX94OAAyvTo+nzhklIXXLL17atlX7tgy7pUkkkSZIkTVS9Cq736tFyJEmSJEnquZ4E1ymls3uxHEmSJEmS5odePdBMkiRJkqQJq1fDwomILYGdgZWBRZs+TimlTXtVFkmSJEmSuqknPdcR8THgp8A2wBLAC02vub0ohyRJkiRJY6FXPdeHAKcBh6SUXujRMiVJkiRJ6ole3XP9UuD7BtaSJEmSpImoV8H1pcAGPVqWJEmSJEk91cth4RdFRAIuAx5tniGldGePyiJJkiRJUlf1quc6Af8AjgP+B/hTi9eIImLliLggIh6LiMcj4sKIWKXNtItGxPERcX9EPBUR10bEJiOkeV9EpIj4SzvLkCRJkiQtmHrVc/0tYCPgROAPwLOjzSAiFgeuAJ4B9iQH7McCV0bE2imlJ0bI4hvA1sBHgTuBg4FLI2LDlNJNLZa3VCnvA6MtqyRJkiRpwdKr4Pot5CeFf6tGHvsB04E1U0p3AETELeRe7wOAE4ZKGBHrAO8H9k4pnVWmXQ3MBmYB27ZI9p/AzcD9wNtrlFuSJEmSNMH1alj4Q8DfauaxLXBdI7AGSCndBVwDbNdG2ueA8ytpnwfOA7aIiEWqM0fExsBu5N5tSZIkSZKG1avg+mTggxFRZ3lrAbe2mD4bmNFG2rtSSk+2SLswsHpjQkQsBJwOHF8N5CVJkiRJGkqvhoVPAV4D3BYRv2Dep4WnlNKnRshj6RbpAB4p+XeatvF5w8eBRYDPj5Dnv0TE/sD+AKus0tbz1SRJkiRJE0ivgut/r/z9yhafJ2Ck4LoxX7NoI120kzYiVieXdfuU0tNt5JsLldLp5N5uZs6c2Wo5kiRJkqQJrCfBdUqpG8PPH2VwD3PDFFr3Slc9ArTqUp5S+Rzy8PUrgOvK08IhDxuP8v8zKaWnRlVqSZIkSdKE16ue626YTb53utkM4LY20m4fEYs33Xc9g/yzYHdU/l+V1sH6o8CXgcNGU2hJkiRJ0sQ3Zg80i4gXImL98vfc8v9Qr+fbyPISYIOImF5ZxjRg4/LZSGkXAnaupJ0M7AJcllJ6pkx+L/DWptel5KedvxX4ahvllCRJkiQtYMay53oW8JfK33XvRT4DOAS4OCKOKvl9FrgXOK0xU0SsCvwZmJVSmgWQUropIs4HTipPA78LOAhYDdi1kTaldF3zQiPiA+Th4FfVLL8kSZIkaYIas+A6pfSZyt+f7kJ+T0TEZsCJwDnkh5FdDhyWUvpnZdYAJjFvr/xewHHAscBSwM3AlimlG+uWTQuWa0/fpuO0G+7/4y6WRJIkSdJ40U/3XJNSugfYcYR57qbFE8TLg8gOL6/RLPMDo5lfkiRJkrTgGbN7riVJkiRJWlAYXEuSJEmSVJPBtSRJkiRJNRlcS5IkSZJUk8G1JEmSJEk1GVxLkiRJklSTwbUkSZIkSTUZXEuSJEmSVJPBtSRJkiRJNRlcS5IkSZJUk8G1JEmSJEk1GVxLkiRJklTT5PldAGlBdsWZW9dKv9m+P+lSSSRJkiTVYc+1JEmSJEk1GVxLkiRJklSTwbUkSZIkSTUZXEuSJEmSVJPBtSRJkiRJNRlcS5IkSZJUk8G1JEmSJEk1GVxLkiRJklSTwbUkSZIkSTUZXEuSJEmSVJPBtSRJkiRJNRlcS5IkSZJUk8G1JEmSJEk1GVxLkiRJklSTwbUkSZIkSTUZXEuSJEmSVJPBtSRJkiRJNU2e3wWQ1D0/+cY7a6Xfep+fdakkkiRJ0oLFnmtJkiRJkmoyuJYkSZIkqSaDa0mSJEmSajK4liRJkiSpJoNrSZIkSZJqMriWJEmSJKkmg2tJkiRJkmryd64lDemCs7aslX6nvX7epZJIkiRJ45s915IkSZIk1WRwLUmSJElSTQbXkiRJkiTVZHAtSZIkSVJNPtBMUs+c860taqXf/QOXdqkkkiRJUnfZcy1JkiRJUk0G15IkSZIk1WRwLUmSJElSTQbXkiRJkiTV1FcPNIuIlYETgXcAAfwSOCyldE8baRcFPgvsBiwF3AR8PKX0q8o8rwQOBt4KTAf+AfwWODqldHN310ZSXWd8u94D0vbbY/AD0k7+Tr38Dt3VB65JkiQtqPqm5zoiFgeuAF4F7AnsDqwBXBkRS7SRxTeA/YBjgG2A+4FLI+J1lXk2JwfWZwPvAj4ITAV+ExHrdmlVJEmSJEkTTD/1XO9H7k1eM6V0B0BE3AL8CTgAOGGohBGxDvB+YO+U0lll2tXAbGAWsG2Z9TzglJRSqqS9Argb+Ddgj+6ukiRJkiRpIuin4Hpb4LpGYA2QUrorIq4BtmOY4LqkfQ44v5L2+Yg4D/hERCySUnompfRQc8KU0mMRcTvwsm6tiKSJ7/Pn1RtifuR7HWIuSZLUT/pmWDiwFnBri+mzgRltpL0rpfRki7QLA6sPlTAilgZeA/y+/aJKkiRJkhYk/RRcLw082mL6I8CUGmkbnw/lK+SHp5001AwRsX9EXB8R18+ZM2eEokiSJEmSJpp+Cq4BUotp0Ua66CRtRBxJvlf7kOpw9HkKldLpKaWZKaWZU6dObaM4kiRJkqSJpJ+C60dp3cM8hda90lWPDJO28fkgEXEg8DngqJTSN0dRTkmSJEnSAqafHmg2m3zvdLMZwG1tpN0+IhZvuu96BvAsMKhXOiJ2B04FvpRSOq7zIktSd3zi+1vWSv+FnX/epZJIkiSplX7qub4E2CAipjcmRMQ0YOPy2UhpFwJ2rqSdDOwCXJZSeqYyfXvgLODMlNIR3Sq8JEmSJGni6qee6zOAQ4CLI+Io8j3UnwXuBU5rzBQRqwJ/BmallGYBpJRuiojzgZMiYiHgLuAgYDVg10raTYDvArcA34qIDSrLfyal9LsxXD9JkiRJUp/qm+A6pfRERGwGnAicQ34Y2eXAYSmlf1ZmDWAS8/bK7wUcBxwLLAXcDGyZUrqxMs9mwCLA64FrmtL/HzCtKysjSfPZXhfVG2Z+1vYOM5ckSarqm+AaIKV0D7DjCPPcTYungKeUngIOL6+h0n4a+HSdMkqSJEmSFjz9dM+1JEmSJEnjUl/1XEuSxqd3Xrx7rfQ/2+6cLpVEkiRp/rDnWpIkSZKkmuy5liSNO1v98KO10v/03cd3qSSSJEntMbiWJE14W110bK30P93+qEH/b33hCR3n9ZMdhnyupiRJ6mMG15IkzUdbX3hqrfQ/2eGDXSqJJEmqw3uuJUmSJEmqyZ5rSZImkK1/cGat9D/Zcd8ulUSSpAWLPdeSJEmSJNVkcC1JkiRJUk0G15IkSZIk1eQ915IkaUjbXPDtWul/vNMeXSqJJEnjm8G1JEnqmW0uOK9W+h/v9N4ulUSSpO4yuJYkSX3rXRdcWCv9j3baoUslkSQt6LznWpIkSZKkmuy5liRJKra94Mcdp71kp226WBJJUr+x51qSJEmSpJrsuZYkSRoD777gl7XS/3Cnt3epJJKkXjC4liRJ6gPb/+DXtdJftOObulQSSVIrDguXJEmSJKkme64lSZIWQDv94IZa6S/Ycd0ulUSSJgaDa0mSJNX2nh/8sVb67+245qD/D73o3lr5nbz9yoP+/8+L7u84r49tv+Kg/7954YMd5wWw9w7L1UovaXxyWLgkSZIkSTUZXEuSJEmSVJPDwiVJkqT56MILHqqVfoedlh30/2XfrZff5u8bnN81355TK7+N95haK73ULwyuJUmSJPXM786sd8/66/f1nnWNTwbXkiRJkvrW7af8rVb6Vx68/L/+vu/4zh98B7DSR1cceSZNWAbXkiRJkjQGHvjiHbXSr3DE6oP+/9uJt9TKb/kPrz04vy//pl5+//bGwfmdfFW9/A59y6D/H/zqz2vlt9whW9ZKP1oG15IkSZKkCe/BUy7qOO1yB28/4jw+LVySJEmSpJoMriVJkiRJqsngWpIkSZKkmgyuJUmSJEmqyeBakiRJkqSaDK4lSZIkSarJ4FqSJEmSpJoMriVJkiRJqsngWpIkSZKkmgyuJUmSJEmqyeBakiRJkqSaDK4lSZIkSarJ4FqSJEmSpJoMriVJkiRJqsngWpIkSZKkmgyuJUmSJEmqyeBakiRJkqSaDK4lSZIkSaqpr4LriFg5Ii6IiMci4vGIuDAiVmkz7aIRcXxE3B8RT0XEtRGxSYv5XhQRR0bE3RHxdETcHBE7dn9tJEmSJEkTRd8E1xGxOHAF8CpgT2B3YA3gyohYoo0svgHsBxwDbAPcD1waEa9rmu+zwKeBrwLvBK4Dvh8RW3VhNSRJkiRJE9Dk+V2AUdgPmA6smVK6AyAibgH+BBwAnDBUwohYB3g/sHdK6awy7WpgNjAL2LZMWw44AvhCSumLJfmVEbE68AXgp2OwXpIkSZKkPtc3PdfkAPi6RmANkFK6C7gG2K6NtM8B51fSPg+cB2wREYuUyVsACwPnNqU/F3htRKxWaw0kSZIkSRNSPwXXawG3tpg+G5jRRtq7UkpPtki7MN1Wc/EAACAASURBVLB6Zb5ngDtazEcby5EkSZIkLYD6KbheGni0xfRHgCk10jY+b7z/PaWURphPkiRJkqR/iXnjyPEpIp4FvpRSOrJp+nHAx1NKQ94/HhG/AF6cUtqwafo7gMuATVJK/y8izgC2SSmt2DTfGsDtwB4ppXNa5L8/sH/5d03gj22s0rLAQ23M145u5mV+4yu/8Vw28xs/eZnf+MpvPJfN/MZXfuO5bOY3fvIyv/GV33gum/n1Jq9VU0pTW33QTw80e5TWPcdTaN0rXfUI0Oonu6ZUPm+8T4mIaOq9bp5vkJTS6cDpI5RhkIi4PqU0czRpepGX+Y2v/MZz2cxv/ORlfuMrv/FcNvMbX/mN57KZ3/jJy/zGV37juWzmN//z6qdh4bPJ90Q3mwHc1kba1crPeTWnfZaBe6xnA4sAr2gxH20sR5IkSZK0AOqn4PoSYIOImN6YEBHTgI3LZyOlXQjYuZJ2MrALcFlK6Zky+efkYHvXpvS7AbeWp5NLkiRJkjRIPw0LPwM4BLg4Io4CEvBZ4F7gtMZMEbEq8GdgVkppFkBK6aaIOB84KSIWAu4CDgJWoxJIp5QejIgTgSMj4h/AjeQAfDNG/rmv0RrVMPIe5mV+4yu/8Vw28xs/eZnf+MpvPJfN/MZXfuO5bOY3fvIyv/GV33gum/nN57z65oFmABGxCnAi8A4ggMuBw1JKd1fmmUYOnj+TUvp0ZfpiwHHA+4GlgJvJD0K7qmkZk4Ajgf2AFcgPJ5uVUrpgbNZKkiRJktTv+iq4liRJkiRpPOqne64lSZK6IiKmR8TG87sckjSUiDBW6zNuMKnHIiK6mNei5TkCkqQ2RcTCwFXkZ7loPouIJSNip/K3ddMJJiK+EhG7ze9y9JuIWBa4NCJWmN9lUfs8gfWBbgZjC4JuXpi7/d1HxMnAm7qY5a+ALbuRUalsdkVEvDQidijPMOgLdbd1RLyuW2Wp5Dnujv2IWCQiNoyI3cqFX6PUZ8fFuKondOs8lVJ6FjgPeHdEbNGNPFvpcmNqrW1RTV+eQzOefA34NEBKae78Lcq8un3MRsSKEfGaiFi7m/mORxGxHLAhcGJEvHF+l2esdbn+cyU5VnuqC3mNuW6d7xrf3Vh0HvXimjauLprKKjvVZIBU48b48VYxajYGweuklNLciJgcEdtFxIu7lO/eEfGymnl8kdxL8vculen1wEzgoS7ktQ7w1Yh4Ze2CZV8ELgCO6EYlrrqfdOtkWznOJkHt4+yXwFkRsXg3ytYoV0opRcRyEbFat/LtgjPK62PAit2qdDbnU+fc0HwOrZHP4qUR4byIOLBOXpU8lwV+ERGv6UZ+Jc9ufnfVY22x8RTolJ7ND3fxu/sO8FfyeWqZbmTY5fNKV7dFI31EHAbsFREvrVG2pSLi7RGxZUS8s4yiqnM9/w0wNSKWqpFH10XEOhERKaUXKtPqNnIcDHwfuB74TUSc2o3zaNP+Urtu1VjPiHhDnXxSSg8Cnyv/fjQiVq5btrHUhXpzV+o/EfETcpy2b0rpsZplGjOV/WQr4D11G0BL3eeFcq3crzTO1MlvcnlfumfXtJSSr3H6An4HnAJM7jD9iyp/bwFs3IUyvai8vx/4fBfymwFMB17Z5e/uavLvm6/Shby2BOYCB1S/01HmsR/5N9Tf1cV1XKOUa8ua+SxKDvjnApcC61Eedlgz3/8qeX4NWLZL67xbyW/1mvlE5e9fAQfXyOtLwIPAzG5t26b87wB+BLy4Rh5dOW6Bi4BbgPcAS3VxHRvlC+ATXcprSeAHwPQaeV0I3Ar8EngnsFgX1nVDYA75ZyQ37/K+cgqwbs08Gg86PQz4ad3tXNkeywJTa+RzJHA7cHE3zuuVfA8v56mdu5BX184rY7EtSl4blvXdC1ikxrb4n5JP43ULcDCwaId5zgCeB3Zv/i67sO91dN4r6/lP4NfAe4E3DLfN28xzn3K9PRk4Aji/fH8/rrPOwKTK9t20C99dY99bv5TvfZ2UD1io8vfHSl4frVm2SZW/O74uVvJ4FfBGYP3mfadGnrXqP+RRHE8Br667fsPsK4t2ery2yHOdsr4HAkvUyKd6Dr0N+DmwfJfKeDNwPLBwt7/TeZY11guYqK/qwd3FPKs71XvLjvXWTg7ypryOLxe+I+vs9JX8ViUHEycDS3aYx87lovJMeT1Grhiu0eEJvLq+awDXAm+ssY6Ty/vC5N+8+yLwkg7zehO50nBo87pRoyIMTAEeB/auuT1fSg4i7iD34txH/l33jvZxKo1BwKxywr0EWKPD/BoVpF3Lfvd1YL0661zJ+zDgTvJv2Xey3+0HPAds27wf1ihTtVFsz7Ivv7Zu3nWPW+DjwP3kytuLmsvape1xbtn/1upCXj8lBzhrdpj+h+TAemsqFcQurefryD8l+TRwQJfy3LscazvU3S5lf/tbOWfVadRpnEdfRa4kHQ2s3EE+e5KDnA8C01rte6NdZwaCh4XLtri9032lRd61zivd3hYMVKZfUs6jZ9H59ewT5TxyDPAGYF1y4HVN2f/OBpYe5TaYTL4O/Q74Wje2QWUZHZ33yNfXxjo9Wt4fIjcwbkNTh0A7+x8DdYEPAYuXaSsCJ5X8t6u5rkuR61PfoV6A09hfFiLXf74OLDfKPFoGMORRT88D7+6wbNXr49HkILTjhm1y49rvydfxF8jBcGO/7KROUK3/HFe268XAq0aRx54l3R3AjqPZx9otH7AM8OWy/h01elb2k0nAjnXOKy227WvInQqv61J+nwRupAt1i7aW3YuFTLQXuQX+CuC1Xc63cUBPJwfCZ1GzUgccRA7A3k+N3sPKQRTkytslwAod5nUk8L/kytZ+5ErDeaWcs8uFq9Me4ncBJ5CHW9Vq7QJeXA7I79NhDwSwNLkC/V8tPvsicANtVkSq26H8/VJysPPx8n/HJ15yxfwhYH/yhfkpcgWg056NRSp/71YuWr8F3tRhfjPIrf1HUrOlunKcLQUcS24AGHVDAsM3mnwLOKJmOd9WLnzfbi77aPeXusctsAK5ovmpTveJNpaxBjkY3pwORuswuIFtqbKubx/td1bSf7QcWxtU9pdqD8w3qNnTCawCfJtcifqPDvfBxvZdktw4+Sk67zlsNJgsRg6avs8oK9RD5PdS4P/IwyS36CCfV5N7+T/Sat8jB0GL19xfPkAOSo5o/my0+dU9r4zFtih5LQ78jHxtPK9R5tGsK7Ap8AiwOy2CJ+DEsj+f1eE2Obps6+U6/e5KPt06721NvnadTb4F67vAPWUd7wW+Sh7ZNuK2IdcFnqJ1XWADcuPRezpd18p+/CNg1Tr7SiW/rcpxu18HaW8iN1CuR6UuRu4Jv4FcB6zTAfIfZRv8O6OoQzXlsW/ZJieSr+PfLdv2vJrfW3WbvK/keTlt1H+AN5f5v0kOrn9Pvp2wERTXGd1QDTavJ4/w3KUL+8kZwP8Dzq1bxpL+QOB75AbPJZvL3kF+ryaPqjm27rq2vcxeLWgivRjbYX2vLgfWg9SoVJc0LyMHq8fRVCnpdOcvJ9ofAZd0khd5SNRj5KB6ucr0pctF6n+BB4B3dlC25cr6zgFurUzvtIKzMwND3k7oJB9gI+BP5B7hzSvT9yVftLcZRV6DGnXIldZ7gP9sbIfyavskxOAK5rnkYTMbkFtvXyCPepjSRj5rAe8AXj/E528q2+Uv5CHFw5axeZ8Cjiplm9Y0vdNGmDeSeyPuBI7pIP1wjSYnkBsCOm58A1av7Hu/pH4jW93jdjVypXr7NuZdop19pinN9uQeiF9RKoY1zlFHkIdxzqaDVm9yQ8L/IwdHizaXBfh8Zdt8rOZ2WYp8fn6OHECNetgv+ZpxcSnzQTXL8yLyiKn7yCMmavfYkytJ1wIv73Df25t8P+4rmqZvWLbF78q2/gBDNPyQrw1nljSrDjHPOcA/ah63tc4rY7ktgJXJ16Lny/vKlc9G3B7ka8vngMtoarhmcKX9U+XY2GeYvJYnB+BHlu37tlK+Q4GHqXlrQ2U5tc57Zf7TydfCdck9uSuQG6Ebgdj9wF3kIc87D5U/LeoClAaKsv5z6bDxueRxBHAqcH6n69qU32GlTM8CHxrlvvIGBs6R15ODrz0rn+9IPud9i3KrDaOrt2xHbozYsZJ+tNu12jjeOM+vQA60/0qlMYbKbUtD5DWp6f/q8XA+OVZ4hNzIOGT9B3h52f9PKv+vTW6I+Au5wW6ZLh0XZ5FHZa0JnffSl3SbkuvsDwEXDvWdtJlXkBsBbyDXrx6i0uPfSRkZuA3mBeCr3fj+2lpurxY00V4MHta3/wjzLg9sNoq8Z5UT2mPkXpxOA4gZ5CC9ZQ8LeSjHjFHkN5V8T+7ccpJ43Wh29nIyewH4NwYuKtWTUJBbh39byj3ivZLN3w353vJfljJ+q9VyRlHexYF3k1sOHyAHj6Meegi8vuwrz5QTa+N7OGSUeVUbdbYo066naRgduVK2GDkAXBrYhGFGLVAqpMBbSn67k4dKziJfAM9nmKGcZRmNC+mfy+tb5J6mDSj3R5IDiWvJlc+PMEQPGy0qkcBpwO+H2e67MIphcOSLyo9Lma+iKWhvI/1QjSaNe+u3Hu3+1pT/wsC25Psanya3fndUuS7H7c87PW5LHuuTA4+3lf/n6Vlm4CK9DnlURlv3NZV1/Q4DFbmOv7uS1yfJFd4XKPdvjjKPRkPCDi0++0gp50Hk3uu5wLdqbutFyC31j5KDyFE9T4Dc0/HPUpbvM8qGjaa8liYHrHeRzzVvrrluK5ErSp8cZp6XjpDHMeQK54sr095TtvFccg/ZneXvI4fI45Nlf3ia3Ej3BfKww6Ur82xUtvvp5AaiTipytc4rY7UtGAgOXszAaInvURki2c76km+1+OUIy1iu7Md/JF+H5smX3Nt4L3A38ETZf58v22YuOXCdPpqytVhG7fNeyedtwJPke2ir9ZWTyn55RmU5Lb+bxjowUBd4GvhgmT6JfM08q8b2XaWUcS65sal2EEY+lx5V8ryVyv3IbaY/nDxS4pvkxvp7yJ0DG5LrKB8ox+SH293O5TtciDxM/Sc0NUYycA0aNrBj+MbxjcnnlneQb8kY9jYW5u30qDbEnlD2kZeX4+Kmsq8fTutROJeRG5iXqqzLVPI55SHyuWmN5vXtYF/5Hfk2r5bf02jzJccpjVsoTmDgXNBp7LJw2caNHv91K5+Ntg6+Krlh4j5yY2VXn+805HJ7sZCJ+mLeYX2tKpxTgP8kB8ovb/F5dfhS9UK/c0lzHfm+62Er1uTW30Wbpq1dDuSDG8to+nwbciWj7eFm5IaCz5CDrnOBl7WZbsghUc3fBzlQeg44ulW5W3x3k8gVzMawmenkisMccm/Eyyrf8WhPGouSg+GbyAH2zq1Oim3uK+cwEISe2OGJsdqocyB52NXt5BbrC8kXnF8DfyBXKB4v70s05bErLe5bJl8I76ME4+QL4KPkE2fLXkBypfR3Zb1+WspyE/li/wS5F/eX5QS3JTkg/T/yhWLJpry+Qr6nrzl4/mRZ55mVaY0TeKMS9ZHRfKfkk+6XyBf4UxnFbQQM32jyoRbzL8Iw95i1KndJ88by3T8C7EHnQ35fTgfHbVP6x4CvDFfmMv0L5HubhhwW2mL7TiX3/DxFbsCr8xCyZcr+/UdyBX6DUabfgNyQ8Jbyf3U4+KzG9iX3chxf9vt3dFrektck8i0tfyJXQjdtdz8s768iD/F7jjz8raNhkiWv5cgPu7qHfB5Zu81081RCyfeTPkoJrpm3h2cNckPFkMEAuZHpcXJA9i7yCIcXyIHYe8o8rybfC/sILZ7tQK7Mr1C+myvI18W/kxv7DqL0ipMD+X9Q7x7Ojs8r3doWbeR7Yln/nwBvrX5PQ+1n5PuifwBcVflOhzoHHFXWf8UR9tvlyz6yHrkhe29yQ9sj5GvbRiOVbYT1rHXeq+RzTFmfncr/u5bj/gPl/4XJjXLDNhSVeVchDzOfSz6fXEHuUJha/W46KON0Bhp2jqKD2whafcdl/3ua/JNQb6LNW3bIjVcXk+/rnU7uzb6S3BD2XXLP+Gnk88OQHUBDTL8UuHqYZe9E00iXps+HG0Xw5nIOaDzk9emyvJPJz0JapSmveTo9yvTG81i2YXDAfVGZfhZN9yeTrwPVOKB6//bXSpkuave4oEWHA/AK8nW25bM+yCNed2eUt3WU7X0JuWHiTAbqkCM1mAxX/lllnS9r+m5HG2CvRD7PP0ium3T0/J9RLXOsFzDRXwwe1ve9pgNjMfLFfC6VITGtdpBy4H6JSkWwnMj+j9yquXOrA6XMdwC5Itl8oE4hX5BvoOkiR27BPpMc9LQ8iBgIXjYiX/RWL/8vSb4f94my04/4gIBWJ7Nh5g1yj90lzd/TEN/dd8m9BAdXyrwk+X6ovwK/YIjhyi3WdSVy8Lk1lYc8kHskfkI+0R7e/F2PYl85hnwxuYgOe5gYCNRfKPvWnLKdbyNXFn9Grkh8hVyxeG0lbbWX+b/Jgdu7G98POaD7FblVfuHyenvJ/0laVO7K9lqXfCF9mnxhW6rsf+8nt5BeVfblR8s2aTysY4lKHguTe8k3KtOWqSxjM3IL8A+oPLGV3PixN/mk2fJp+M37D4ODpSVL+Z4p5R9VqyZtNJqUdWv0eM7zwCQG35+1FHlI+FKVadPLej9HrjQN+1Ce5uU3reuhZTu2ddxWy0iuEP6dyn2BLdZ1WXIPzxkMUQlrWt+tgNXK3y8h3y7x97IPjtjT1GLbNirtS5Abcq4nNwq0fX902aaPA19utZymv/coZZ3WybHcYj9Zj1wBbfkrAMN9H+W7v5jc+/8ZWjTmtlOGyvf3LnKP1UOMMJqAPLLhOZqG8JEbOmYDl1emT658vje5MXDYB4kxEKw1znln0nS/Jnk48sPASiPktXQ5po5joLdlDrlh8GByI+p1jHCPbot9r2vnlTrboqRpNDy/jtxA+nWaGlTJ54I55GtGW/ddkp9FMJfKSDwGBw6NMn+4bIuXVLd5u/tz+e4eJt/qsO1ovq8W0+uc9xrrsxY5KLyCPIroWfKogsb1a7SV/aXIQ+wfK9/naBsAG9t3NfKw3JeX/6eQr0fPkxs5p430/TTltwS5brhm0+dblOPi9+TGrnnqjLS+B38TckPWsZVpu5OD6nvL+j9Nrou0apy7gcq5m3wdmkRugLmTgQaJ6rG3NrkONGxjNkOPIriDfN34KLmudxL5gcBPkzse5qn7MbjT4z3k3u9BDe0MDpRPIR97Iz4/hsHXy4+1e1yUffZGmhqUyI1/95LPofPcLko+T1xAuS4PsZ+sQB5l+l7yub9xX/SK5BFd95Fvxxj2WGtat9eTz1VvZvCImv3LvncjsNso8ptEpa5U9usdyJ1R9zGK0cSdvMYs4wXpxeBhff/DQGVxB/IJ/bgR0p9JDjzneRoouRLw3+SLasun+pErpY1Won3JvRiNi8LbStprycH6UsAryb2ITwIblvmaK8mNg2gm+X6PHzL4wrxIWb+/kC/6mzDMUBzmPZkN+4RccjB2yXDzVL67O8k/k/PiatnL358sB9NtQ52MKuu6LrkF+VFyBf9xcqVuevl8JfJJ8VlyoNFJgL0Yg4eAdjREhVxBPIZ8gh6xIlhJV+1lPru87iP3NH+PHMSeVb6HaZVtN5MRnuIKTCMH9XMp94E3rfeLyv64E3nY+JD3w5Z963IqPXgM/IzHr8mVr7czcH/zx4fa7yp/70K+UJ5ftuvypUyTyI0AD5ZtshGju/9r2EYT8v3EzwB7tUhbDdS+RL6APEhu/T2isu8tVz5/mlyxGKpHqBG4vKys7yfK/rZYZZ62j9umvNcnV4b+QK5gNp52W62YHUFu9R9qlEN1fb9D7n3ctzJtIXLAfSf5nPh22gvS30muDH2suuyy315MruiM+PMvDPTQnVO253talb38P4XciHAJXfiJrkq+a5AbxoY8v5B7S/Ygt+xvRGmMKftz46FSZzDMbT8MPu8dXY6Lb1LO8+W72JDca/MMw9zGQq7I7VL+br4f933kwPvbTdtsffJ5ua2nQ5dtuQ9NP2dYyrkIefTYfzOK3klyg8RryMHnDQz+aakhfwKHLp9XurwtGnmtRz6330luoH+cfH45vDLvNuRg4u+0Udks2+Ah8iiJ9ZuXWf5epOyDD5N7Ui+tLnOE/KsByM7kYeN/pMVooFbpGIPzXiX9v1X2jdOocftFyW9RcuDwN/I1vK3bQZrW9X+pNEg3tgW5fjeXfI4d9hkClfxWJI9+u5M8cuBy8rW6Ubd8ddl/HiD3Oldv01iW3FB1KPCapvx3LPvMh5rmX5fcaDwXOKdFuV5GbiBo3FZWPebeSO7cuZDB15Qp5Eazuxmhka3MP9QogpWa5ptEDkqHHIVCew3t1eOk0wfZtnVckEdgfa5yTFaPrRPI9di9mr6/V5TtOM8D3Rg4r7yc3NhwX1nP+8gdWI1z1aTyXTZuEWynLvBf5NE5T5Ov1b+mcosPud7YeJ7Sv7WR3yfJ550bqTxIrpRtU3KD+OPAgXWO4WG32VhlvKC9GBjWd0fZSQ6lPCF1hHS7kS9sm1Na35i3Erck+V66ucA6w+TVeAjSzZQLXzmodio75lxyr+GDpYx7DrG86lNP/0queEwbYp03IlccBrVmD1PGdobSr0luDPhQtTwt5tuCfGHaplLmf/20R2W+fcr6zvPwsMr8U8gXjXPILbPrku8bnUse3tPo2V4a+Czw8y7sK6MaAtoin8XIoxYeI1cqX9VYp/JqeV9sWbcfkU9kO5IbPfYiX+BvJrfaziUHc4PuiW+jTM0PaBrNBaTa+vwhcuPPVVR6Vcq+fDP5wjqXfPI8apg8G9v3yLKv/IZ8AX2UfHHavnwnjZPuH8nB4WiHTTc3mjRGeaxfpv3HCOlPLvvfv5N/bqjx0KxTGOj9eSm5B/wFWo8gaFz8liUHLX8iX/ieJPc47E4+H7yIUR63lWVsQD4nPEwOAF9dyrVeKfPfaTFKp0U+J5ArulvRNJSylG8m+QL4NPn4nWc0QOXv/yJfxB8mn3OfId+K0wg4X1m+37nAKW2u5xvJF9+hGhJeQg7mn6DFb9/WfTHMaB3yNeMecgNM48nFF1Duhy/zfJB8DP6GFg1vlfVYq3xv15ArlTeUdJ9h4FkMM8iB+lxg1xHKvWgp088r++1yJb8nyz55ErlSfRtwzXDrPMKyqj3gG5ErXm09ob95fyrTppKfO3E+IzwYji6eV8ZiW5DrDHeTh5LOKNMaD806nUrPI/l6cOoovvcPMvCgxa2aPluEPGKkcW/7VeSRGG3/1A+Drznrk89hHxxm/jE971W29eLkYO4hOnjg6lBlZxR1AQaf935fytPyidvk8+bj5ECo5QPiKusW5HPdL8gNxbuRO4qeKt9f41hblnxe/l1TWTYs2/u5sg2OJje2Ti6vL5a85+mhZ5hbahioE3+GXCdcurItGuffW8lP0z6CXLd8mjLqhzbOKQwxiqDy3bTdCEMboxMZOI+P9hbFUR0XTWmvJTc4Nc7Jy5PPc8+TGxf2I9c9/rvsi5Obl1nJaza5s+0dZTs0RuWdx+DbDz9CrmeOdP/78eQ6xQ4MjBy9v5RjRmW+NcnnxY1GyO8Ucl3qG+TY5YpSvgMr86zDQMPOqt04lucpx1hkuqC+GBjW13ig1u9GmHcy+UL3Y+btsW4c2ItU5n9ri3waTzlsnHQ2IV+U76c82Zd8UVm2nIA+Rw4CZjYvq0XenyD3aK7ZVO7miu4Mmi7YI3xPww2lfxH5onA7IwyVKvPdS9Pwx8p3N6Uybdh7OMmBwa+p3E9TTj63M+8Tqhdj4ILe6ZPIRxwC2mY+k8iNC6O9V3NVBnqZP1WZ/k5ywP5D4L0dlqn5AU1rNH3e3JgziYELzlRK4Egemn8X+eL5IQbuiVqRPKJjPdoYPUCu3D5e9pfG8KUdyrrv11SONwDvq7EtqhWl95EvRBeNkG4TcqC5PQMXtU1L+Q5qmndxRniaNrmR5BfkRpNlyK3RjYdd/etnLcg9dqcx+nurViGf454u+T5Q3q+mjZbgst3uJg8JnlRdFwZX2FYjD+8b8nfcycHNg2VfWZYcVFxKrjS8oanMn2YUPwXF4IaEk8kjgl5MrtgcX/bvPTrZVzrYt6oPi/snuSK0Vpl2aPn+92VwBWwnmkaQtDhObyZXSBu9QyuQg5KfVo8tcsNty5/jYd6hhR8v26Q6imuZcsz9iNwQci25d63xpN9R//RaSbdwOW5uBb7XYR6DKrvN6zNMuq6dV7q1LSrz7FK+k+oojrPJ56Zp5f830HTs0+b1jIH7cO8v+W5F7lE7iXxsf7Eyb0c/q1fZHm39NjVjfN4r6fcpeX6+uu/UedFBXYA8QueP5AduDfm0Z3JDxx8YYqRTZb5jmbeudyK54WjQg7rKMfevkTJN+exG7il8lhxkH04+j08hB+TVJ5kPeZw174fk2zUeLmX8/+ydZ7xU1dWHHy4dBKSIFCmCWCgqomDvvcWCPfbeorEbjb0bY2/RqDHRvHYN9kRjrxhL7L2AKAIWbKBM3g//tTlrzpwzM3DPNOTD/t17yuyzy9qrl2BE6IJcmP9t52Q6UuptP6d7Q74XwRwnlYz1lYl3YnPOBfmGinZI+ToDKbWDrDAAWXinITw2AYVY9rPnScaZXRAOWY6IHwuK8hBa6BMRBvkkLXZ+kMHJAe7dYThhGPGEXXw/SbBu9zdFStZN3LuhFNqesXcXx+SgtP6atVdZdzi//Q+kYTmdfIKYFtf8APB4yrPWqJj8krH77ZD7btDqjTTgDkhnKIoP/gG5MnnXqFQLUML3L0aIOW3sw4liXgIgl5vsIi6AhYQyoxFj9nv3bg8suVCsj2ORgO7jdgPiaULazr1jv0k7lNcjohAQxqmGcIIHwG40s7RKyndLuoCW0cdcCepEGtufVTAIZwAAIABJREFUkQY8MRnL3CAeytTIo5IaezlYnooSoAQhcxmkLPoYuRr1LHdcDhbOQBaUXnbdBlk4b7C160BC8hPmLoFO2IuHbC9eLuM329v8htt1X1zML9IyF8Be0vwRM/U2ilsKQsMpds5CCIgvMTJH5zb2rbWRVeNg2+uyrFNIgTN7PAn7tQDGDFIkAzxioJ9CQnOwLI9ETP9Rdt3Hvd/Wf6fMsQ4gSloXXOBmIUVC0dCWrJvB1nmIgfR7+LSNMYTFpFpLYvdWRS6gs6sgoHCNDzCGEGcNT4CZXrH7ZyAFRHvkERNybKwee6+3HxNzr6DsjmjHSzimvblrXO47ZIhXmrsXCfcPQsJGV7s+HeGUMaEvpBAviKucg7UagZRsnxNl+n+UfNrdck7OWzlrlfJeNfFe8BxIDRuYy7mWEw4SYO9KpExI42cGEgnAqQKO6+9mJIAFOhw8Xzax6/WIef4V+XYf5P33IMKVHyLL6JoGI6cWmV+b2PVo5G0QjAivIuFpw9j4RxBLKDencEdGHoVZ95XSf1p4STyp8a+REa8TUQ6IO3DWWqSkWNb2Lc9DK6H/wxGOC/xZHE5GIxpV4Mqf0t8whJc2t+t+dn2xG9vZuHNd4lwcg/jFoCDoi/DgeXbdi4QKInMKK2XtUdYdzm+zNyuuefsAZ+Www9cCuS5MIkpI4TVOIxHxHhvrawVExB5A2tlJSDj0Lhl9kUvELAP2VMYXCalj3XU4OH9CiLGDv2//90LEf66sfG4NSlr6kGbrioTfr2sH+0QKSzL0Q8j9pBJjCHO9E3jG/t+W/MygHZBle47cnOdgHZqt/bZ+5lhQJ1/J8TyFVubmMEZFhX6E9PexZ5cjz4HxCMnPrteNmOhxiKheQonkRwnjuA54zl0/Tb6L2W+Rdahs18UyvlmgYCvy7sHAN+76HZtvOHeb2d4UZHdP6GtzREyWsOtdYrA80ta57BJ8zViDbiQkIEFxqDOJtMatY88PtT0pVSGhIxJ2f2PXPQ1GriaKw73e1rdZxBMxl3sgS8wWzEUm3ozW9EbgCXd9m8FLYGYOQS6YJbPKo7Can4k8IY5DFqA17Ho4SpyTJNQNQHQlWPDOQUrR4ILcCjHTjyMlr6d9zaqrGhvH4SgL7Fxl0W/mtzPDK83Zi5T+xiImdUnEYHsc0AoxoY8Sy3w8F2vQCglyK6JQDu9ungldK3McVcN7KFnVfygjGdVc9B23BG9Mgtsq8jp7BfEmeQoMZOk7E/FH5SqgHwLutP/XRTT5ILtuhzx3rqNMLwLX/852BmbZeF9COLrA48fmsTNRlZgeyO37MPd8FeSd8BMViJclI4/CrPsq83u9kaI5eKweYjhlGbtujyz9XyB+YhRzqFxChpCvEa1dy+YVwjdbIY+K+5JgNqW/xW08u9n124j3CYriVZASacsy+7sceNddv4tyrgQPqd1pZjWSsteq0h+Y3/4HsjwdSqRFbueeDTPgepz8DMG9UBmPt+MIDVkcT0Iu0dOB592zpth7xyI3wgdJzsbY3ZBfUrKllRETfH3sfkhMNQHYuplrU9LShxi54BbyK6IEG03IQv8dircJGv6hSLM7kYjR8cQnyXKwpfVznSGkQ9yzDZHwXxUX0Gau59xYWyutZU0V+pGVchdELL/F1RYmFjduffxoyDJRaI3tc7BOHG/npDdSOH3iYGUhlDjoRsqsyzwn61rmvSEoVvliZJV8kah8XE/kVXBXke94j42N7ByNRJr8n4FjAmwgBdazVCBOOGFcF6L4vLi3TC/DHQU1YRFDdR/wlxJ9t0LMwnvA1XbvRcMjne16WRRvVVZcWj03IovStcBr9v9JiNEJSoouiGbcSBk135Fi9nukfNiU/NJCLZFVYjyxBEX2fCEUNvMFElpmEpUu82dwGeRq+DNwcoXWJnOrQ7FvVAKvNHMvghLSK9e7IsX4m7b2R9j9VkhRNIUSceX13mqJ98gwgWGRb3RDwkFB8jok5HgY8R57qxheTAzrgrzkWAF2jkE0eCdEi85376yMDBypSfRKzKO3wdw9RMm+dk1aU5TAbJrhlglIidCJfL52GMKDOeCsCq19sz0KK9FXie8siXj56bi8DOQb7Fohofh1ZIHehBIJQz3+QrlIZiCc/jWRd11LJKy/DxxXCu5i43nc8NTTiIYH/r4HUh4X8BAedmP3tkDK3J2QzON5qT4o+ej1SePLfD+q8ZH5LQ+J/RFZsQfYdTtkOfwcWWHOQLEu4+yQzM7Al9DnY0Suilu6+14gaY+0NVNIqe2GWWGQkPs3d787skh8g9y/lkXEaw/ESPw5w/VJtPTF5vJbm++pRFmUWyCXppwhi/8gwvJp0toREaGRwLnufn9DGN8h63k3ZBnbHDEpV9cahioMnxXVsiYhQvdsA/vml0SZXPMIgvv/95RIquRh2v5fGmVVnYQIQ3C/7mT9TXX3qmJtwbkUI6HofESsvsfcaFFs5SU2vr7x8RERPx/20RkJlO8b/rjAPVsWKVBS43AznuMgogQqx7j77ewsTzd4G2z3lkXeMn6+s5nGlG+cjJiw1xAjGr7XFQk2/2Eu3MHrtSEL4XfI9W0WZslEIUR7Izy/Vny+RPRngLu3oK3RRCQcn2b329gZnExKFn7Xx512dl/DJU2KndlBRPXAi5ZSqfeWBV6pxF4gJcaLKLFUgPd1kED+I7IIrok8Lz4ArnG/bahzQZ3jvQzn2YrIEt+P/MSeCxDxPUchQaQzEmJfAsal9NnC9beGu78Y8mT4CXgprDNy830JuDsLeEGK9N8W21vDFT8iIX9T98zzBCGHRo4SithmjDUzXiDLvkp8ZxminE/POVzj8XEL5Eb/gL1XLKHcAKQYWNrdOwgJsFMNx3RDYUAv4hL9psEJsi6vQxSusBRRksHDkEFhVaRAmUbk4l3Ax9v/vizg4kSZwKcR4d8+No/PicJ4KronFd/sX3ojphVCGpX3kCAY0sO3Q5bDvyMiOtH+XzcOVKFPRMD/iNzinkRM1eHx7xIxm7Otve6dphiQHk2UmChYf3ojBvZDpA2egQjYDb6fjNYqPs8kbdWfbYzX4GpXIwXCVUgpcSjG6JEsjLRCVrVZyIU0uN8uitxIP0IWmQ8Q03hj1nOt10aVtKx+fw3GlkeM4Qt2BnYk38VwICnKoYR+d7U+1nD3DrRzN8HOzFjk3fAlFtpQyb2NnbMtDIZvIoqL620w/bGdtVtQgqN3iFxDk5RE3RBh+pV7tgPSBH+JPFe6IcHrGeAx9161CP7aNt/xRIq87iir6gQ7hyE+9/2k+bq+LkLEORDmAUTZQG83+N3UcMHXRMzpXMX11rqhEJV/u+uuKPnQBIOV4UgpdgRiRoOFMgl3LmpwcSYRLhxu8PMtEu42QO7er+PKscT7I6IvDyJm5mPrZ9P4O/Z/B8p07avXliVeyWIv3O/aoJjEHMKfmyIhpTUSqMchpd03yDp0nuuroc4FDYT3spinu77A9tfHsy+Hksn9jIwJnxks+rCRAiu1/X85MWWXwc19SMB+AIUkvpDW3xzOp5ys3QGej7BzNgEpR703mxewF7DzN3puxjSvNuS99Q4SLu8i8t6Mhw/0Bk5J2wf7f0uDk+uRR0QIZ90RCaqfGW55Bbg9pQ//zV52Pt9AyR/bWn+rI6tyDvE/nyOL9sol+jsECflHuXsrIt5pElL+nokUDp8SGS4qjvdqDgjzakNWT4/MNiYK+l8TCcTfEwuuRxq5bqRkUowBVmBwVjPA/Aq4xD0fhQjzUEpoG5HmMmQ5nIQs0yFrcxtkWR6LhNhliTRimQOpzX9hd70bcJG7DkqA+3EZ1OPII9anP5z7oridkIjl30TufO2RVu0AZF1boZJzrcdGpTV66VbIJoOzu5Dm+ghDxn0Mjq8s1ScS2s4k8gTxzMNYxLR+hpRRtwM7VmlNA1O4PMoR8AFRze5wzlqjLL/n2Brsj7kxkqAksv+3REzIJFw2TCTA/wMJmNNReMn1RAn7KgrLRG7MPe1c7mVznkxU6qQVUqgcgoThPdxaFJQrQczAS4j4XkWUp6Iz0nK/iZR/XxhuWKMac63A2gUms7PBSg4xGgs7GA+VHGYQleI7zvWRV5LQ1npthNc/RR4Rob+FkFvzeGQVfxTHdBVbPyJlye7ojL5JvldGB2RZG1pOf/XWyBCvZLkX5OOAGxET+SbCm98g65J351za2oLxvhql0QB4L2u4s/9bI9x3DlJE/pX8yhmrImPBuShjewiha5XS3+Yo6d5MYq7VBiM7I6v4NShef8F4fxWee1/EB2+PhLZ3yA/Ta4HoqK94U/cKkyqsW8D5qyN+/gyksB5PvuW5FTLmJckSHk7ORAa86UTK8Y0c7HVG/MoOyGIc4poTjWT23hW2pznr9wjyPVDWMRhemQgnppUKHoC8Fz618/5HotDbRVAitHuRN+TpwGrVhJWaA8S82JAwug9y9WwywMtLwIAE1JsMyE5K6COx+DwS/joTy/KJBOirkbbqYVQy4i0SYjWRwLiV6/N+rM6ljX1bOwDfU8TaUAkgtfXaBWmeVkFuLjnEZHqt5RaIOLyMXFLKipdFmtlnkGZ7DGLKg1tpQamzSs71l9gczDUhpuBQxJzOzi5J5MLzs+3vK8BbZfQZYvefRcJIzpDuae7drkjj3S/WR8VcI4kY9M2QlvVapCC6GTHrbwObzUlf9v/ZqLbk+zbXGeSXwOmLiOxGiGlPJH4ZznNBO7OBwP3K9qM7ErC3QIlUZpKQsbPE2u2ChJAJ1nIohm+0e3eowcBiRAxmo7m8BianD2JEniZSxEwm8nZqQcRYr0a+u3KT/2v/32fw9iqybuVQ1thl3DudiRQWBTVe4ziQwkR0GyLc+hHK1NwPeT19RzMTZ9VoLzLDK1nvhXtvHLIubmhjWBfFqc5C8asLp8yt0c5F3eK9jOe5BPklBK8Oc0OKyhBO8wSuEkKx9Yrdv87g5S9IsHne1jA1p0ex/rLe24RnayDPmE+QsNcFuRF/Chxb6/2qdSt2jpFS5lAkB7wPrG/397c9H1bkt39GfMleKEY7yBOfIlrco5zxONy1MXIlPwkZF7dFSqJZiNfrVQ5OIqJtv0J08Q7ET72F6MxNRFWTwrernuzyf/+bL1xXZlEF1CcjQfefyA3iQuQC4ZmVQcjta5YBSGKSDCKGq4sBz1v2m5sMaIOVaGED3veQBv0fro+WbmwnIY3vCYZspyCNpSdgayOXvxz57uYVFzKRxf2fiGh+g4rChziy2a7syI1uoiHewyiRxAdpxL61w+33YS/EEH6OszhQJU3tL6XF4Ot2ZOn6wv7+B5cJ12B9R+Sa9idiZX8S+u6ChK7LiDIWr27f+QoRC2/JqSqjhZi8T5DCzSce2hnFbU7F1XOmhLIIZYSeiBRLvZGAdYfhnP+jAplsy5jjlohxOwdpoINSrK17Z1UkEORwCa5IyAzuzvmqyCJ3GHL5CiExE5Ewt0nabxu1ISvkrchVsz2qZf0iUjilKiaS5o0Y6TeQG2lHZLn4I6IBj5Mfs5ZYR5R8K8dYRLceRW6Z/dyzEUSWwy/s7G3s+26kljVeyWIv3PPRtsaeZrUwXBPciK8kqjXe8AriesR7Gc5tAcSTvYKEj6ORQLKme6cTsgB+iISLFdyzUt6JWyODybpE/NRAxA/ORMqj4InSRDPLqJU558Db9rQ5/wGFX3Qnwv+Lo8RWn9ucPwDuq/V+1UNzazQa8RLnIqv+Qu6dHRCd/Akpa7/FvIuS9hd5k02x33merR/ix75EiplF5gCun0RKP5/MuTvKTZFDgvaSaWOK9bcI4pdOJ3J5b4tkrGlIEbqKe79o1ZGK7U2tgWNebiib6AwDRp+MwltgF0ZukTkSanXGgPt5ZJXeETE4OcRw7U7kutPakNFwIveNeNx3Z0SIvkHM2t4pY1sWuQblcHHHVVq7xYlqy14aG1cLIsajE1I2vJV2iBwC2gMxfaE2t2eKDrRvTcElX5rfKrK3lyIFUEi8NB4xEZOBvWLvtiYlRCL23laGcFcmX3GyKHC37e295RKECsx5ScQIhzhMD8/b2Pi+A04vo68BSFA/nnzXvz7I0pEzAlNWjHqGc2yHiOTXNhev3PPjHIaY/pnAvSX6bIXcOp8Eusee7YMYhg+Afdz9Rhesd0E0Y4UYLK9pZyUH/M7dL+a23Q8JcyfF1wW5a4dkZEUrITgcuj9SRD6MBM4ccjFdmUjJ2w2FFx1GZC1pSMEuS7yS1V643wy1M/Rru24TexY8PP5FZL1t2LNRr3gv4zmui4TczxFvtovd90aFNihx2dOIhyvXC+g3yPKYlxwTeTyE9ZtIlWKYyffGfArxcI/bvG8yfBeUAJ2Rt+HVNHDOgAqt36bIdf4tZKiYhlzCfSjOKsjj4+/k8/tJwvUqiPcu8FpFiu5pyHvibKLcJ8Us6C2QAeH6+J7Z2X3UYO8xIg/GYt4MGyFeahW79rjgHCIX9rFpfVRlf2oNIPNic0B/FHLHeBcRVQ/UHiDaIAZ7sSJ9XoBi7ALwPYisRK8hRuwokkttxV35AkJd0YDwe8QkrO7eiWdlvAw4o0pr52NT/4pcRmYi7fRCsXf9OPvE5+v2ITB969uct/FrH36PkvI8YHtWYA2b3zLZ3w2RVW5Tuz4TMa/7ouyWeUlbSsGJu97OfhusNK0crPdGTMUnSPNaceaLwrijRRHBOtHd8/D7qM3/A1x8WdKcUXbtbzGBElkYfHKjlxHT9QLVY5T8uZtiZ/YRXCbS2Hz7IkL4IYWlBr2VtDWKPx/v7nkh4ngkzL9KBvWs66EhJeBPmDIhNt/NiMrZXFNGX4vYfni3Wb8PtyMh7GVM8VME7lYxuAp1xYNSaDqytm1CGSXy6rlVEq80dy+IrHxDEF3uioShv7p3vJfIOORB9xlSgDSqgqNu8V6F5rsZEjCnI8+IgAeayBdMQu3uv6b0MzuBq/09wGB5gOsvPOuP+NSQ06LipRrdOB+xeQxGgvZEG+ebyBsqkT7wyxasAw5aAgtRQfktAl8/C4WerOx+Ezey5WUSR94DbZFi7mvyS9mFdzsgBcjt9p3fpMFdgCu3x0+6+56mnYVkkDeQYqld0jjddajDvmm8P8MD7yGX9udxXh9V36NaA8m83JBVtQsS6kIs2kkx4FuXfAY0KbPoICT07W7XlyFBsCcqwfAqciu7AHOtKGNs/ZGF5FAk3DwNbOuet7Tx9yafca5WhuFWdlB6IAvVFEQ0l3LvbI+V08LVvPTjRPE5/0Va7z7I/fhJYu53KP76buRm9ikZlhmb36I9wbK62/VWiIEIbqPrIaIwHeUBKOmWT1Rbtw9iIv9MlBk/EOEhhrj/aO8cnuW8SozvUqI6xDchBmYz97wFsvTdg8ITHkEExhOfQPy6u3vPIE1vb9dPYKYeInK7v5YqMSFIEO5k3/wd0i6/gLOsuDF2sHPXPXbf45pQMuMopC33YQNhTX6LmOqnbI9r4pnQzHULuCqUHBmJlSVx7wTmoT/C95ci69Yesb5auj7C+t2MlLA+lrMJ4dhbkRfBq7is5Alj7GD9/Mn1/xOyho1GQuGbyDUxMR6vkVoWeKUSe4F4gY/DmSLyuLo89t4SWMgEcpf8gJhyut4bDYL3MpxvgKuxBlPXI17kH7j65ghvNtncu8R/H2DK/i6IXIUXRJ4THyBas1isv3WQgnc35EFWFe89JPC/ROQSfB+iGWsgvu1jRBeLhoX9Ehuit7cSWYT7Id7pRIcXHsCqB5SAuQVs/7e16yuQMWDz2PuL2TlbGSns3sS58MfevRoJzq2QtXkGrsqQvdPevnUakke+I5/Oe5j+HZKneth378USEbt3BiKvqn0NjlOT4FZ8f2oNIPNSiwFC3HK1jAHDpwZMA5BA9zVwcIl+2yAN+qLIejAJZXwMFtmLEaOTI8XimgT87tkmSPB/0w5la4S8/0YZ1pGM1s4z1QvG1rIz0mC+iawEu9pB/B44O22uiPjca8ggaNA2QNrvFxBjvgCKobsLeNbe+QdC8g1hbWmkhpj0kJTpfhR20Mk9+xB5ZVxYRl97GsyvZmfkSqQoOomoDnorxDg8jWIc70PCWGJ+g4znuoeNbyu7XhFp5V9DbrNNSBjdycbd237zHZGg1cLNYzoSNFsiF7lpiHgNdt/sjixWa6N4pp/CWlR4rknEdS2kTHgPxWgFAXET24feRfq7HllDO6EcDF8h99aViJjutrYOpyGhIwccWmsYn5t1QwlaPkehON2Rp85bSLE421KDmM6HUY6MN3DlT1yfCyHPgfPtenNkCbsdGOPe648YqrURA5bDxXDG+uxke7gZcv9/DQmcHRCePY/Ior53VutToz3JDK9ksRfkK+P/iOhTEPQXRkr1YLHdATgS0bzX7Z3DkNtogWdbvTYaBO9lOdeE+02G2z5GBoFQmrUF8vq5sIw+/omE1JAc71ikCLodM+oggeky4BG7fg64owrzbomU7WfY9YmIPw4VI3Yl8ow5lxolpqrXZjB/KVGo2XOoakAooxtCOt8Htk6CL/f/3wzGBtr1QKLcKGchBeoGyJv0Pdu7MxGNKvBWQh6KX9l5bIn47BORl+145F6+DgqJ/R7JNcNsr5NyqBxmY9nArvfChSTZvTZE+QjaIbnoY2qUh6HmADKvNCLt6WBUBuMGQ2SLkJ9w7Bqikh2fkeLS4/oNRCbUYt7HDksf984piCEYntKHP0SLo1iWVWLvjEBW3Ql2qG5CSYQqnuXVrd0yiJH5EAlexxNlHm6LGMtQ+P4znLberZMXype2ftYj3/1uOeQy8g1R5tcPba9aIMvhNcwXrpu7r8XiZtohC40vsbYE0qovWwasdEMKkWPI197/nyHrJ4FTEfGZjCmJUNmaeykzu3wzYLmLfftE8ut1L4mY8GmGAz6wc3aqPT8FWV76xs7tscgda4S7d4XB7912/rexc/sVUlBtY+ekYkwmkYWkk+3J0NjzxZAAMgklTzzQxndOkT5XRYq0rYhiutYgqo95Bso8fgaK1d/V1vtTGkiwc2u3gOGpi4kYo2VRPo33kaJhJVu7/2KZfZEAfidShnoB7AQkgPV397ZCSsV3kVXvHMSMfWhncUuDlUQaYn0MtL9rIbc7LxzuZ2fvBIpkMa7XRoZ4pVJ7gRI+XUdMgYQE+D2R98YMhFseJwqVuhcpZGqS2Gduz4X9X5d4L8O5en6lA8KXbWLv7I+EmNeQsH2Szb/A9T3W3wjEy61JfhjisQi/Bhj8BPFCY+z5s6gyQzWS1w5EhqbO9t3Dyfc6HA/cwvzM4Gnrtxjijdc1GFnJrd/vkLv9y/78JPQxBIWPxMsCj0AC9HfWfkD0yHvi/RsTXt13F0Ml3P5MfpmtrshTISTm/Al5tx1vz/dANHwlu/a81J0o7MvzUrsj3mmi4YiHbYwn2/M/I4XlfOG6UZsDgl5IYHjZEMIPiGna2AFgS+SycDpwdEIfvlRRkjVoJ6TdCdqqgYjgn+He8cTJI9tTbGxfICJ8KVIGhG92RRqslxGxDxqhimXNjq3dR0h43h+5x05DWSKXdO93QxqvDeN9xPo9FxGQNzEXKPJjtBZA7vp7IWtCyJJ5KiI6ZbnXz2/F99X+H4wUHXnJLxBynoK0oisClyPkXSrre3fEaP6TSJvvlScHIqT/JRJGLrH7iyCt63lZzLHI+JZCDPrjwP52r4lIydYOCSOhjuSv7P7SiLicG+tvaeAiIg2/X9sDkYX3J2Qhe4Mo0celSFCviJsukQV5SeSe9gGyvl6NLMnheQfDK5MR03uV66NF7O9AJDjfSZSkMZzZQYYXPkPM5btYDDvCsVMps6RZrRv51uiNDV7XoTA765+Q0JWz+T1gz7raXl8U63drhPvOCd9x6xeScP0XKVFvJ4ol/ovB0QKUUCoiq8RMB7etDJavI0bHGqmRMV7JYi9cX/sgfmIasGsYGxFD28L2YQRyZW+LLOpnIKVAqtKkXht1ivcqNNc9DLa+QMLu7lhstD3flChT80QsQW7aOUPhGjcZfIWQBB+PvwoyAt0IHAeMsvsHGJytWoE5FvOe7IEUsJ6PXR95K/Uop49fYnPnfwc7C8vZdRskMBfIGLHf/xrRlh+IEiP6jN4tkVfNrxF9CsnwDrTvrRvrrx+iy58CN7s+Ai/QAoWzrof47lCBYbTB9aWx/oYjWelfmEUbl8UeCfJXIQXiHURZ0JdBfMKJNdubWgNHozfyhdfxyPLWDWnipiDC9imyriTGPJHAkKBMfPcj4rKVu78Usrq+b39fsP8TMyo7IDwMaXb3JIrhziHBfDnyYzx7EjG21YqxfhQxGyGr6QREPCcjIWW1lN+l1XPchigxxhmYRcieFSgL7BBfjYjbVs2Zy/yWt64XI6L5NXJNXNM9G400q8F74AucNaxIn5fbb37EGHy77xnhBZDGs6NdL4nczV+uwpxXQwqaHFKydU0ao7vXAjFTzwH/jD3byPqZhlm3E+baFwmlyxEpMPZCTOcWFZpjIOpdEDN4CyLwv7PxPoUUJh6vjCI/Z0I8dKazwUAOedG0j3/P/l/c5rqwXS+DcGBRL6B6bEgp8TMSoAeHdSFiRloixeNoJJCFXBg32rp7Tf5qRK7Z18e+42lLZ6KKCX2Qh9APwIoBHu3v8ij8Zh3Mam33RyJ3u4eRFew4ZNnYqdbr2cy9yAyvZLUX7r0BSEE1EzGvSyT15+4tihQeH/m5NEqjTvFexnMMOHQ7xJv9AYWovYDox8XkGxYGomRTS5XR95UO/nZy9xMNJbaWh9l6FySpymCuAZ+1QXRgaYzHtPu9EC/wLKIjGyKB6sFa71M9NCIZoT8KJx2Mc8dG3gkfIo/LDRH/+zkpyVHd7zZBssB3wN/d/bTKOz2R0D4ZODDheXek1PncYGn5YrCHeJ8dEL2/O+F5cAfPobCkpjL624oEXqrqe1aECewhAAAgAElEQVRroJlXGorffY4otvcZxHyMQALsFHtnQBl93YkEw0cNQD9EcQ9B8FzKDtEDyK1swP9SgM3uj8KVq0FaqBmIkL+HrLsbUIU41JTxbW2INSS1eAZpadvZQQ4lx/aYw36XRoRqGkrUkZj8w64XRMQtMWPu/DZX+7q3IeHDkND1JGJ8DnXvDLb935EyrSuIwT3L4OLfwDD3LInRHGXw9Dyxck4VnPsSdn5/Qlrebu5ZPPtlO2TB/msCXLZBiracneGVSsx1QSJr55FVmOcdyNIXwlbeRUzR60j425wEt6z4Gti91ojRHI8S2+1GQmK32G/2NPxVtKRXvTakKHiRKH7Mh/skzXcVxLy8T4zJRh4C2yEr3hTEZBVUT3DXHRED/yGRJSww/JsgK1kInXkM2N79dn373be2zyfWei0z2IvM8EpGexGUHGFP2iK34CkIl/oKH0k4ZX0qYIGs0l7UNd7LcJ5DkALkOLteF9GMoHS7DQlTiYJEib53tvV7HSkr4vDkDUOroDJNR1Rgjv479yNlew7xxzu58YxBfO+PhndeocpGnnpsRIL1SNvLbxFf9RdgpHvvbFvbn+35teXAivV7g+3J34k87JLOWV/Eq/26yDibkMJ1KvKCXT/+jrtuhbwyfpM0RqR83AEZ295GCvuCxH3uehHkBXRPzfet1gOYFxqyJuyIlRBCjPK7WCY75GoWtC9XUSQxA7JQjCdyyR6AGNhJSHM90O6HAzDbwpHSXxMSpv+CCPjyhrwORARsbxvXBJRcoOpxWUgQucjGcwyycPrsmC8gBu4jEmLAiyFeZA14AHkQHEtCEqViiGd+m6v9DMTyDCz+xa6HIQYo1C6f63ADQ8q/RULYXQbXxQjIGKqc0AfFQt6BFFknUyKTdRwv4JghO6czkCC7Ytp5t/dXowpKIlSH+T6ixDj/RArGXoix/97w1qGUcPV3fbZBmUifRsR5B/LdGePKh05IwC5ZC73eGvnWnH/YuTgHFyuaMN92Nt+RKX22Q1aMlxGjugVF8gsgoSRuJe2OFBZnIgXlxoZ730OJpUJc+MKIMfOZhxuaCc4Sr8ztXpRaQ0Sn30Yuv17hMU/QsXrHe1nOE8WRjkPKmAHIg+t0xFMebzjhHlwC27T1Snm2FhJU30aCbKoBhQpnk0fKo9cMf+2MFCEf2jxD1YiF7NkGRBnhKxaW2CgN8e6vI8XqWBQTPwHRW+/ZuoE938jdK6jEkdB/P+B8JJQ/SpRQNUkJPtvFu8SY90HKxTewSkd+PLF3i42tA+In3kZK5Y2KwQTyLKpJnHXeOGo9gHmlISGul23s8zjXGuQ69iQSuvdK+K3XxAyxA+StrJ2RMPIpYmZLus7G+h9OxADfgrRUgUFqh4j/W8AzNVy/tvb3RmuBwPawtTsWyxQY+93sWo4oCdDOyI3Rx6t1JEqCchEJblWlEMX8VtYe+r1ogSzWe8Xe6YuS/PyIBIrUjNHxfbF+47UaxyKXuvEojqdt2u8rPPduyD15cfLdtYJVOijWhib8tsn/LfKNjZHb9H+RZbGm2VMRQd7LiN/ORvhWcM/HIetnDmcFLLUviLEMceszcAJdLfY2w/UqxYxcZmv1NxKE59hZ8K7FeZYootjboFQ8mGTvgUQLgOHLW8jPxjzI8PBk5OEzR0rOemtZ4pUs94IIh7ZBCvCrkDVpOPlKprHIlfJtnAKzEVuj4b2M574CMNb+vxXRxCBoDiPKL/EsyUKJh73eiAbFPVoGG5xMQxbCnlWeY4DpS8Nc7bovymo/BfFliyX8tqFwfMbr5nFMf+TF4BWZuyKB+y1SKg45fOThZD2k8N4H8dfBUNcF8dkf2llbvcxxjkHhQycTy1WErNLPIsVsWXgK0f/eMXzXGoV/PIoUUHviQqLqEV5qPoBGbMU2EGn9PwX+4O5tZEDRM96HQzxDkZb2fGS1KdDaIdfa9w34C5j0Msf+HFar1K6H2dhGE5XKqRiTlNY3YkJaIrehN9z9kDV3WXevRbwvovqW3yOry33kxys1oZiNHC7xzvyW2b76zI7XG5x9hSy3nWPvdkUxxjmKlHoj3x3qFOvzcpy1xp6PRszDR1TRiklEuLZCbrMzDQbvAZaJvXsAcvd7loSaneQTvy1QaaPjiMX/I0viq8iTY7/42tZg34OS7mykVAzZ/VsjAe1AEpIDujO8JrIUXoWYhUXcO4sgpmsmiiFsuCzUbi5B298bKSIuRPi8b+y9Q+1c/BOr/54GK0Tud/ejfBVHESUvaoEsYVcjRdaVuNj/IrA3mMjaGrJNB7rQDtVp/gIxeiVjP+uxZYlXstwL8unZg4ju3YPCpKbZWfJxqivZmXuRBEazERoNivcynH9rxPt0QPWefXz5UIOlEe4spinEzjM4no5o0Gmx77RFPFHIQVNxazARfeyE6hk/RlRHeTavifixL1FG/pUqPa5GaLG9bY3c9h8llqQPlbp6CkuoSEpOF/f/VYbTXkceDQ+hMpChHGrI//KenbXEc+b2dgdkmf4QeZjOQAocX5t+pMHeT7ga1in9bYZ4+ekGLycTlR1sQvTpeuRZ9Efq2GOt5gNopGabG5ikTkhjsyrOnQZpgh5C7i8HI5fs50muR+oz5E5CwkjQVJ7i3vMC9n7AxSnj84eoTfxgILezp5EFYhgi+Hk1MKmg5set3ULA0cj96YDYO1vZOvwHuRC/R4nEBCjl/tsoXqkdEq5/RBq99WPvjiUm+Mxvme7x8ygk4nak/cwhIaJX7L0OyMMgkSkkP6vuBMRsXo4YyS9xWUXtvX72vRurNM9ADFYweL0Qeaj42pJbxH6zNQklqGLn9kr77Xhby8k4RZ290wvFhP6IJfWqwnw9E5wU/3c+UiouY3sbSo7tEp+n29tf2W+eQYzh54iZ9PGV3VB5lhwWi9pozc13AYPR522+byEGZvXY+yGZ0w1pcGf/jzNcdzESwqYhy5ePw+uB8OxkXNx/yjgvsv0IpXl2dc+8FSHUVR5cau711rLEK5XaCxvPu5i3h43xOyKGsq97d1GiULGG8Ryw8dY93qv0vN29bki4vhUZaFojIecNt79ptaxPR3zjnkgIe93wx30UemRcS0Id4QrM0SuwHkE4/n2s7JI98zjl9zbmogm4fmkNKZ6fQnz7RxitIJ8eL2F7/TXFS26dYucq1Iq+3dZ8EjLseePfFliptzjcEfE+o1D89++R8iRkHs/Zee5PRPP743iBFFhZCYWCXY8s3kGpeBvOmIjCkS4hVimj3lrNB9CIDQnZTxJlth2PXCxCUp/BSLj+ETHejxGLeyCfMF+HLALDkcBxtfV7OZG7dFGXILsOwuteRIkjzkdKgADkq6NspJORtml6APr4IcpwvXym7iaUqOJ9OzhfI0YzZP5d0BDKY0hTfZ1bu6Q12AoJ1sHt/Q82t2OQYD4N2C1pD2sNR/NCi8HxUrZvQ5A23seSXozLcFtm3y2J6lyGbLDTEMPxlSFhny25o/u/4u5BKFxjPHClXfezs/V/iFH6Ctiv1Lq5e4cZ7G5q19fa+fwKEUKfdbwj5s1R6bk6HNQNJVS6G9XvXo4Inw1Dwsk7iJl6C3gqoa+Ah5a0vQw1KZcjIvQvkJ+tuRMJNV0brSGl66MohKg1yrgdssrHwycWw0KDkvYXJd76EFMUolCfKbZ+z5KfRMZnpE7zHNoH4eJjkRfCKzau/d07vkTLosX6q+eWNV7Jci9Q3PBLwOZ2fQFSOq2ILI45pExeinwGu27cIediP+oS72U4P2+UWQfxKLeiGGifX+EEpEB5EiWF/JoS9Z1RjO1EIovwAYZTLkNefC8R846p1tohejEZeeHcbucrh0uaFsMp61Z6TI3UgO2RJfhWRHOnIAVewDMtiOhvd2DtIn2tZL/d3q6PMDjZFMkKXyOeuaza8EiBfh9wtV0vhuSd05DXzizEBw0nZk1PwXtdkVLoMrte2OYekjo/Bqzh3u9YrL96aDUfQCM0lGTmPCLG8A6ikgHrIwvNpygBTLAAt0YWiOWIXCeTLD4roRjj7dy9RZBr1A8oVisInklAOSIgVrvexpDqXYjIf4OE1B2Jam0vgQj0qcDWFV67ixATGTKBn4c0caEe54FI0J5KVKeyBWJCuhK5DyWtXWvgSMySb3P8FljLrvciKudxea3haF5uqDb5uYgIdIo9O9f24WYsUV+Zfe6NmIwBdv1fZBVaEmkzc3Y+4nE+1Yqz3tjGMwgpEmaXg0JW2RxiEq8to6+lkJIp1GkMdSTHEsXiPkoRzXQF5rdW7PpFpD1/yeb1Joq1CkLgYFuPf6EcB6G6QVwJ2BrFFV9n14Ps3J6GMoTPQNbTpFIfdUlIy1jLPZFAN8KuH0DeOZsSlaM7qZz5Iu+Ap4Ft7PospNhYHNWuzSFGdv8SYwpKy85IiPM1UVey85oDLnD3w542hGCTMu/M8EpWe+H6G4MskZ0Nh3yNJSdCSstQXvJ9GjhMws237vBehnPrRX7pwdWQB8J/kTLmZyQ4reXeOdjwwT24rOcpsNcWJQO70K43sLXbBnnJXGrrNxnYsEpz9gqfXRCvHMIjlkeKqxxwlXsvnsjzFxuyR76X04WYcgV5ZG6BZI2pwGbuN3HvhCRYWQ8prRZEYT/fExnVVjKcNR0JxAVhawn9DbDxrYMMKa8jOaYdEowft31+D1eOq0h/+2A1za2PT7CQQWRxD0bMRGNFPbaaD6CeGxLyeiDi+SmRq8OVOC0R0uLcYgB6NQluxykAfyLS8LxDrI6zHYLdDcg+SOmzLbIwTELCTSdDqMcRaeSXQFakqRRJZpE0vgzWrxXSvr2JXPDGIO3YAe6d1ihJyTN2gHadw2/0AdawfkLSuOBB0Mv27Qngj7WGp3m1IWXQK4gRfNrd91rpoOh4E1czt0S/WwBn2//nIKYk1ALuiZQ2U5DnQudKwHCJ8XVFDHQTcBISPhd1zx+2dXmXEgQLMR5/N3he1s7r/vastRGWaUhZVvGwBsNN05FVqY2d46eQhrodckd9ycZzFPklpFqSkEjFPV8ACd972fUTSKgJtYOPtblOxkrUNHIzPLg3JjyjMImPMVc3xBQHd7qHy+ivN2KUBiNPpCmYcGfPbzF8+yUpMW7u3aVQ6cd/E/PwQQLiBTauf9R6HTPcj8zwSpZ7Ye93JKoycg1ScPh67/chJdTvar2OGe1FXeG9DOfVCikFXsN4RSRo/pEofvo3Bi9P4bKc4+Lq7bpYJuV1UZxqW8T/XECkAOuFeMuPqXKpQhTKczFm2XT3F0UKqJ+RRbtuY2ZrCDs9EP29H9jR3W+JKmncj5QoB81BnwsReXxcjwThYGxrQpbhe4FxZfbXBoUgtEQlB18jv3Th6TbOh3A8UZH+liTyYrsMyQPBUNnW4Hgi4jkWLGeMtW41H0AjNERw/2yb+yCy3gQLhBcgzjVCcA9luLhYv08h5uVOYjWwERO7qX0vMd4QxbQ8Z2M71RDWDvbMl3p5wBDaaRjxrtLatUIa28eQ28i0gBTI13KuiLS4OVIEYfJjtJLqVH8AHBWeI8I9DljavdeQlq96b8iD4w7bv8vc/baxd84roy+/z93s7yPInS54Mgyx/T4bCwmo0jzT4t6uBe734zZitRtlxkQSxVieYnihL5G3zMOIWF3QnPHPwTyXNAI3yXDGqYjoxbXkYc//QBlJFt182qE4rVHICreB29tDEDG9jYRkaI3YUKK/IYhReMbWM2Rp7YsUMFdQ2to8e/3s79HIi6qn4bwm5IJ5HXB4GeNajcg1/SoK83QsjLyDcsD1tV7HZu5BpniluXtRBj64AwlMQVhaFAls3mrV8Fa+esJ7Gc/rCGTRexUpl29EgrCHww2RFfsNZLVPLZVVDHaQovc94FB3bx2Dn7WJLKLVcAfvjyyjOTtfXWLPuyMc/x0yfrSvxrgapRkcBGVrQd1xlOguVL8pUD6XWksk8HpeZTCSG3zS4LLxCjLsTSVSFrdDvMLJDq+mlQouqLqAZKyL3f0ByBK+K8bLNwK81HwA9d5wMdIo5ul9lL3Wa5S8ALG/Af0+xYCJ/CRlfyOq/Rsvo9CS9DqaARi7IkY+WHs2de+0cf9faN/5Y7WBE2X7vBJp3O4gciX1MbtL2lqclvB7L4h3wykibG9aI8L8GrImLIuYxVfcAa/7A9kILW0dkav/tcjieSuRJbI1JbKTkh4L2hIpTj7G3KsR4zoGEe4BpcaV0ZwDHmgdg8WgwLoKWe67IwFqfeR1UpD9tMhcwzeuIT9jfhdkwdqYIhbhDOca8EoH28dgxTvLvePxylmGV+6lzFqpDg+G5F1Bq94KMV5nEAmfDaUQKzZeg4+J5Nd/XxEpP5eK91EEVgLcnQN86+73RKEZW7t7RRkb5N30OApD2o9Ysi3kabANdVA7NKu9mBu8kuVekE//V0Bu0H3IjyU8GXm2nYxiaW9Egn9ZNePrrdU73stwnl5g2AxZ399BsfPD7X5bB1dLIe/I6cjNOykUJC5Md4hd90X8X3Cl7Y2sn0/hLJRVXIMQIpRDnk0LxZ63Q6EyDVOXvMowtI2d/UkoPCQeWjUAeah6T7kW5PPTiyDvIy+ftEfete+jqhWrIPr9CQlhJr5PZP1eEnnXePy1AzLcnYgUOtshnmHHlLml8Y8tiZTP4939DZDHY1nx4PXSaj6ARmjkM9P7IDevieRnU/XMZpILd/xwxC2vgUG9k5RyBElAGRtbEJ5fxbmZx8Z2MLGas1Vcx8WRu8hM5AqWJGD7UiNJ8z3HDtonSMPV3z3bGFmAcijZ3FQipr2hGPR6bTF4Wwh5BwwhEoT62R59bkhysbS9dP2E3/ZCcVoHURgmcbzt62XI6vcOcFc154zyJ9yCrEe3xmBvGTt3M5GC5xtcybuU9RuFiNvysXe2tbmejYTPCxHjNaRG+3yOjecrLP7T7nu8cggJHidu7RZDbrNrk58sqq+t28soOWFwlSzQ2DdCc7DcDikUN6CwRvIdiGnaCiU1ehh4pMQebIcErF1i674SokUvIObmHqRgLcimXOIM9kDJB2cgYW6RlPfqXrhJ2Itm45Ws9oLCfBTXITqVs795nmVI4f4zUty9SWS5aZh9SFi/usV7Gc7X8zSjEK8SFJChXGFLIsVMZ3u2Wkp/QRDf3/DHc8gyHg+9+gEpYF5FnoKb+N9XaK5pwtKCSCH0M+JvB85pH7+UljR/pHSdgjxXdyem3HSw489WoLcnEGWNfwLYl8jY0R+FwnyPvAc+J8rr0JTUt53XV+03U5CXiS91G+SXkMT5kpR5hn5HII+hKwyOZyswkdA/zcZ/h/1/Ra33aI73tNYDqNdWjHgBmyMC+j7OLQNHaAOgxACqjQHlgyiZzYm4MizIhXQW0jaOpbTbWNKBPIwoQ/kWRcZWcctX0reRm+HBiFl4iijWLWkuY5EmLBzw4+xgn4eYoZAILWQZbYnc5A9EFpiGZETqtZHPMFzskPdUZO1Zwp4tYAjzDaTkWCulv44OqXZEAtYXCOHnUJhFOEOD7XoqYh5uqyQsI6vtqkReD71tbI8iYegjJED7+MrVEHP8F+A3ft0Szt/vECM+AzFE5zs474kYzG8NH3yC1X0thRMyXoPW7v+9Udzos+QnUEyqq9kq9nd5g4XvEKF83p9NpLD8j+35BBwhTcIL9dgoFKDvNVidZfC6KZH78FBkqc7Zmr5ElPSyiZhlEnlFTDI4+dLWaog9a48UG08h3Pg4sEKxc4GY/e0RPh2DKTTt2yHG+irKcPOvx0aGeCXLvSBK8BRgfz/ra1fEvF5k47sFl/gR1dsehik80va1HhsNiPcynLunl70RXZiKFLNegRJfowC7I5HSNlzvh2jOv1COhFnIZX4pe94J5a+5AykmNvf9VWiOAceH8oubIku8px1nGlzfQAMmpqsGjCCvlWEoTGBhorxBfZEgPB15ABTkbiHKQB/2Yk97/xyDmVeQ+/0fiJI4tkZ0d1tgTAzuOhouWtGuuxs+uxEpvy+x/RxHfgbvtQ2X+eoITQgveqNZfyTQv0qUj+IFohjrLkgRei/it05POlP13mo+gHps5GuCNkKCwq+BNd39MShOYaIhssQge/JdhB5FAsm/kYU6ZwdnTyICsxJRZrzEZA/uQHZFlsK+5Je72s7G9ZodoGrG2/i1O9bmeSdi2gID2RG5vbyPLM0FwhfS5L5qSGIHQzjjgD3dO6vYAfwZF2s0v1V8j89DdTVPRFkoL7B9/A4Xp4iYyf96BOye/R2FAIRMotchAX0F29ezkRZ0HBHz3x5Z2AZQJIt8RnO82M7hbvbdM1AMcA97vjJRvPHvYr/1TFVLYC2ba3+7twFikg5FlsvzkEb4IaLKAF2QtXcdogRHlS65FVeKxTXlmyGlwltIQRb2wM+3FVIermbXnZE18K/2+6ORsPMD+YqJpZAyY3TaeOq1IYvn4URxvJcb7trV2v0Gy0cTMU0dUdbWMUB3t3ZbIkZopN3bjsjKHTJQv4bw4iZuDL0Q05JY5omIZuyIhMhvkBI2hzLEruLeDRmbn6WMzLH11MgQr2S9F0iQ/BFZtNdBfMWxsfHvYXvyFBKUGoaZTNiLhsB7Gc95dtlQd9YDTLW1OU9AvODq8d+56y62Vs8jpW0vpJA7GBlpOgFbo7CGN3xf8TWr1Pq5ubZA1WkmIF7sZ5Q0aw337l42n1eAUbXep3pobv1WtX0Oyr8PkBFuiHv3LqRMuRbjQQJsIb7kW5QXamGUNPgoImG7HVJsTEUKnuVSxhOE691sHA8iYX91pPDzJfG2t31+ASWITFKyB7z3D6QoCxWDbkQKpkUMzvdH/OM08mWs9uR7uTUEPzB7vLUeQD03FJsw3QH9+8CfiIhyPwPWbwzA2xTp61RDhCOJ3NVGG1B9TH728d5EyQHi7uPhQI6yA/kNckW9BxHm8HwFJNh8hDTzHZq7HnO4dtcS1ad7EjFyHxG5KbVDxPdFW99uCX0MtXX9ATGv43BMoL2zlO1JDik5GuoANloz+J2MBOcAxy0MCT+LkqqMcO8n1dhsi7Jrf49iEocgi5AvR9cdEeTJBscFia3iZyPjeS5hRGAm0taejiXLc+8MIbI2XZl2/pHVN4eUcUsi75ILiZivrsgVagISREem9FPxmHL7/2w7s5NtDXYkqnU9AhHUjxGj2CPWz0jEHHyJhJFBSLm2mHtnLSRw5oDjS42nnpvB8j+JkkUOQwLeWH8GkGA1C7nBFcSl2xlqgSw/b9nar4cEvD+49W+B4vkftvX7bRljDEzTsgiXnoroxxDErM9Ewtxw95utgXNqvb5zsReZ4JUK7sU+yGrzhJ2jkNzTe3etimjnO8jyVMC4NkKjAfBexvMNwkxfhDffRMLk+eR7IhyG+L4XcMaChP58srOxSEjJUz4i5dxzyLK4HbGyVlWa9/02zz1QaN4xiKd7mXy+dgObS+9a71Wtm8PJ/YiqDO2KrL/3IiXc9eQL2H8iIZs3Cs+7FCkC70V8WMi47/HKmYZXxuHyMqWMbxfr7xkkmN9j95uIhOblDJd9gBSynRL6aYP4h9eQYnJFZKjY3r3TCuHapxCNTD0TjdRqPoB6bciyOtX+9kGE+lpkEb4ZExoMeC7BZfCM9dMCucneC/zF3Q+CSW8D4jvt2lt+02JZuiE3t9uR9ntfIiH9TCIL8cLIbe3WKq/dKFunEMfRAlkN/omUFau6+8uTEmNu7/RBLuCzEKHeNuGdRYhi506pNezMyw0xft9iRJ58969tbA8SE1nE+mln7082xPsWsEHsnfYoBONlYvG+VZprT8RM/2zzOj3hnV52BkNm1KR67G2R1fYTO4//AE5NeGdDez4lCc6rNOfLDLfcbHP/BDG/l2Jl/JC2+QHEQA2J/b6Fnf/biVxcH6bQvXZplMDoZ+CmWsN1M9esNVJI5JCl9Eli1hlkwT/Uzs59xBJXuveaEMN8v63N21j2ffKtUaMQ05MD/lbGGFsiwf5hzFLunoWM4Yn90CCKDhtrZngly70gPwHQpkhRn7NzFoT1lkTK8b6I/7ij1mvajL1oGLyXwVx9zOh7SNi8BlnoJhjs7OHe38rWpWiJLKSUeAl5knyBE67dO4sT1aTfvcrz3tLmt2bs/rKIB3ycfB4hhMbMN4JoHa5AQmXv2P1Q2/molN/lJfgzuDsFKam+wVUnID+h2Z7W7w0p/Xr5Yy3Dn98hBU6QWXyugK6GZ3+iSOJlRGMetfe+x3jEGF5cCbjJxndurfem2Xtb6wHUUyM/9ucEpPX2CSPaIa3Ll1iN1iJ9xS3OzwD/jN0L7kKnWZ+pmXZjQL8nElR7u3sLI4b2c2LCDVXIuktEXEYZwr2NwnqNQ5G29hliDF6sj7hLY2eiOO0nSWBMkUViW2qguZ2Xm0OiAZmvTIyIk088J5EghKb03RK5HD0RECqFCX9aIsXMC8DRVZqzZ5wXQMqrzxGDtHzC+12QlSaVMUTEb0U31zsptPq2QB4njyOiVhUGxJ274YjJ3czhjK7IpfsL5GrmCXU8Vsuv2xBkpfoaMV8hztRbyPsTJUIpYBobrSE34R9sPodRGEvZFlmEc8AxSTDi/h+KFBrfGc7rnfDOIGTNeLTM8d1q58i7c/oEOF/RoHHWsXk2G69kuRdEuNNn3l3Kxvcjio3s6d7x+LShK13UM97Lco7u/11tHgPcvc2RYPEq+ZVcRhGFGfk+4rxjd6S0yyE3+4EJY+iDvKsKPMUqNOeA8/cyvBHivj2Mb4GEqZ2S5vVLb7ZWd+PkAvItzdciw1tvIj7M09imhHt7I0vyt7iKRbF+1yQKw0gz4PmcFP8iSvrYN63PYnBi/y+DlAkzkTKog/+W/b8kUkrtXOv9afb+1noA9dKImIzBiIl+DjjfPfcE7xHEhLYkQWB1B2EpjOFGhHc6UZIAD3THoVjsnrH7LfzY7P8jEKG/xd2bXUsOlTPMRcsAACAASURBVHR4zo8j/r0M16xV7GD0sLWbgVzuWiWM4yhbh8Epffr+1rL9aIESZmyLYn1fREJeapmVWsPTvNAcgm2N4m8WQ4T+OQpdvlqiOqxvUyTTsztn/jwNR/HLM1BCoIKSPySUiajQnMP4+iCiFspDbI7CGt5AxCkeI1fgcZKCG5ZAwuos5G3RJeGdIaTEzlZif+3/Ecit7wmipCeeeN6KFCdJ4RsFwrVd90U1kr9EzGWfhO92p4FqV6bAisfZ6yIPog+RkqJ9/DckCLApsNLfYOQ7FPbTPf4u8mJq68eT8E4IMQrJuBa1a4+X90Aug4unzbeeW5Z4Jcu9iD3fCylggkDV2cY3FTGUvhRbm2LjqeeWsn51g/cymmOLBHy3MQoTutXDgP2/ClLQXlNsvWK4ZDEiRU5bosRg1xbDIVSQ/3FwHfZpNyRAr2PXHqcsiYS8ecLNNwuYcf+H8mi3IN6/NxEdDXxzSE7cs0Rf65NfxWM9ZMD6CDjJ3W9NgnCesLdLIF47JFAMMdtf2hlexv2mXZFxhf66uXuDiCoG3URyxaCGK/mYuN+1HkCtG3CYBwYkxL1oSGwi+fGjAegvRcJjqpUUWbw+NGTbFgnOryG3jVWJ3GMGIle9fyT04YltK0Tgv7exve4PHRFRPxpZi8qqN9vMtbsZaSe9W9vhSMDPq/VNhPi3Q0RmTIm+T0IxS8e4tQoWif8ijd4WFIlzn98y2+crkTZ0hF1vjix0L6CSNC2Qm+95iEFP1J6789MJ1VrfgSiEYRH7ziwUo1ZAUOy9amQ+HYgsVJditRUN9lZEiQan4OKQS40TWSvHEmmL+6AkcCEGN1FxUKm5ks+4t0JE9yWiEna+xEY42yPt+a9S+gyEdCgSZEISowVR/NZEZJVaLswtPr9K7m0F1tAzznshBieUOhmGlE+fIy+jril9FNSyRsqIE4kScnVD1ojJyNV0di6OctYLuav/n+3DUOvn3+QnimmN6MZbFCmXU68tS7xSqb2wfXgP0bW45fZCxETfSYoFqFEadYz3Mp5n4EmaiKqU5BB/dqd7b7YBBnnpfEWC114CHO6JFJKH4LwuUPKnmchzcSWqaEgg35p5B6L5CyKF88vEFK8oJvc9zHL9S29ubw9CycdaE5WfO8VgabaiGoURvY6rZZ3Q18GIT74WVz4RWYnvQXzyFfG9SejPGxffQjQjHk5zFhKw78eUKUX687zU3UihFhQKC9m4v0Lx4YP8nOaVVvMB1HTyqh2XIyGuiShR0Y24+DkkNN+FLDyp5a1Qps/HUMKCALjrIKvfD4iQ3mH9TCSmCUexyN8DZ8S+MRglJMghQt3LPWuBSl18jHNLqtDa9UOM+ATkCuW1U79CAvZPuEzKdqj+guLDOyb0GRDGHkjjuRuxeqs2x+FESR+OoUIZo+e3/4Gs0bej7JDxmq7vGiyHeLAJwLop/Xim9Qk7B/vG3ulq8Bvchpao4jzDuWuDGIJbgI0T3lvczu23KDFUopbV9bcHInBX4WKSkCv5b2397iYhsVIF53ofcGXs3gBEUHOI8PePPV8NEcMtEvrz1v53EN7b3D1vi6w6/0HEOVFAb5QWg+WbkbLvQPLx/8K2rz8iz6TEutGxfk823H0x+SE/7Q2nvmawVDT3gMOjI229t0LWhyYU4jDNYHwXe3a4wfNh5a5BvbRK4ZWs9sJ+s6Ot7zaYoIRzybfrQ5AF+3WM2WzEVs94L8M5DkX0Lp6he0sU75qzc+ZDaJqQ5fkdUgwfbu02Mng5meQQuI2QB99/kfKi4vyPG1sTUsTeRJR3aCvb64/tbAwB1kA82qu13q96aoiv/9bgIwibwSPhOuSROcT29TPg7IQ+An5fz87RQUQKQC+g90K0PMS+90gZU3i/tcHnTcQMe+7/fRGf90Ea7nOw0g7RmTuAX8feWcDw4bvWVyLf2Mit5gOo6eTFDB6NGJBXKEwqcIwB/YtI87IPcu/+lsgCk+QC9XsUw3Zt7H4L5HJxAtL0jUfJ0EJ9YA/EyyABfBISSD2i7kbkYnUJcp/ugJD7h8ClVVq/hQyB/oDKbnnNmS9V9Jj9f6PNZ5S9U6B1RZaGV1Ecerx2rNeK90EC0Hm1hqN5tSHt+ZOIiK8cYBTn7oNKqhyLrEUla1ja+XkbWfeSXCc7WJ8hg3FVY+hRXNFLOG0xjmDZdU+U3fNzkjNkBmK1IhKuDiFy7/P9hORLE3HuuhWeXx/DK58hIupL+HW2+eeQtjvs+WBk9fucFAJt771kffu184l+xiDFZA7YrdbwncFanomYyjWJ4sfi+3shkStn+yJ9bYzoyg7hvdi5CDHCITP10in9BMZmURvXDeSXUGmPrCX/QvRjFrI8+VqiDWdByBKvZLUX9m5HpMz6WxxXxNfZxvf7Wq9lM/agbvFexvNcD1kLfwB2jT0bhoTMycitN4QALIt4zBtL9N3D+r6IBAOEe284wsdXzu085nRf7f8gsK3k7rVFwuBjdh6+QfTlZSIB/BcbqufwUTvE/19Pft3n3shLZoqdmx+RwHmjeyfuwt0RlbD7C4UlMz2uao1o+iVljPNIJFwPdfeSPHo2R5WKUnkBe+8GZJnuV2RsaxJVNUr1BGzEVvMB1LoZkO5hwDwFVzLBnm9pwJ5DzOOpRIJ1Umbg4UhYzyGi2mStVey9BclPKJMkpC+C3Kgm2UEa6J61trHkiOplPwhc7N6peNySjeNSG8Ol5LuULo6Y9CkoznpLIs19orYVEafJuFT9dt8z6SF2s2P8+fyW6d6OJdLEX0C+e9oca8tRDNnLJNQkj+8fKtuxZhXnGuL6HwUuL/JeiBHqjNMWJ7zXGhHRcRRmZ44TynWAC6s414WRm+oEZO3zZ7YlsvLNsjP7qLVnMUsNyUqxnZFwM8bdKziTyOpzEQ3ofhybx2LIenkEsVJJCbB8HLB3EbhrhYTvccRiURNgZSQlFBPIPfp1FB883n/HvdMGCUFr4CylSbBc7y0rvFKhvVgAuVkmZr+1b26Z0HfD7YONu27xXsbzXJHIeBD3LlyQKDRuIlJkPUx+4qq0fDFjkKCxaey+F3CDl2PXpOcVnPOpiJebhpXtI9+brQfy2DwaWddDzO4vVrB2a9MKVQj42OAhjpdaIaX9b1D5vXXdsyR6uwCit79P239gWMK91L1APN5LaTCFDH4hTCw167vhtL5IJimZhBaFD8xzruE1H0BNJ58vsG2GkgD8TKE2cigSvj9Bgnhb//vYu00o1ijUcf2Ne5aU8S8tW18Y2wKI2Q3lHVaNvbcTUQ3pDYmSm1W1NibS0s7CajESKQ0WQgkMPkAxugEpp817FeRmvJVdx13vt0fMaqpWd37LdF+XRp4bX6Ewii7u2RwxgCj+5gdM0IgjZqTBXXtO+81wru0NTi+36zjsDUIJ+cqq04ms/qll8BBzXpDwqsJzDHilHXIHfgtZStaLvTcWWUZyKK5+iWLjQ1aqKaTH2y9BFAfbplhfjdCQEnA6KRnikYC7YhyWi+C9R7CyPEnwj6ypqWFIsfsdkFX9FZTMa2P3LFUplja2em9Z45UK7MVzSNGe5OUyGrmqF81B0kitHvFehnPzvNtSRJ4+Nye8e5U9ew5Z6oNXSbFcHWshHnTDpHVBuR32Jj9nQjWMKE3I4/Ilwyl7umepvGaj4pQKrF8/5G06EblCh6SSRfn0tL1FiUAnYaUB4+8iz9FbSBCwi3zrQuuzh7sX+IWu9nwHpIAvlWOiLzJK7mvXcXw5jEJX8XkKVmo+gJpNPNndYVUibeTpsfe7IleI7w3JDCjR12pEtV7PcPfLtviRb2nYEwk5HxNLEIFcKyYh974tSCDiWa9byrMtkBA2HlkIgktdWxQX8jpSAmxTpI+2yArxMi5piP3tgCzhDzGPZBSsl1ZiX/sgbet3KFSiLOEy3j9yL54KnOXu++y++9j56z+n/We4DjfbWQrZUENJqhZEsZYrl+ijJRJenwXuc/f9eV4OuWWn1niv4Bx9rOdOiPmbSMy6iqwo7yHhe/tieAUpHb7DLNLkx34tgKyBR9KgFrmE+Y5CwvWBYW/JZ7x3QcxI0cSSbo1uN/wdcJ6HlRHIlXnLOdzb3ZDy5BtcaRMaRJApcx8ywysV2IvQ34VIYbwzjm7ZuTjPcEpqkqtGafWO9yo05z4oDHA64nvi1vrjEQ94WzlzRkrIr4E/J8BRCDP5BwnZ1is4R4/XdkBKuxywX8oez1NCUobr2M/g4SvE34aQgYKqE6X2A3mI3Izc7zeg0CvpWKQIKSfXR4CvkIPqcPK9FEMi4XcpwrvH+mxrOPQBPy43/t2RPNVwORfK3u9aD6DqE44BMoVxvcOByw3Ibkr4/d/s2d5z2JcvnTVXDCayrj+M3KaPjz1bArkizURJBzJnYt18W6PM5SsQSw6DmM63kZV/B/ITnW1tB/SaYogEMfxf2+EbZvcGG8L4EXPrm4/EM9tXrxRa1JD1aPJrvHcg0sRfMCdIkXziHGJQ44LcAIPtAgtApeccu7+ana//EmVHb7L1eAW4YQ6+sa/N9cjY/bbIGvwRVaorXAwfoBjC+5HlOe7iOBC5hX9FfsbPuEV2ceRm/gguxsqejTSc8JvmzqNeGmIQHrI5D4o9646EubvTcFT8Popj+wZZGxZy9zsi5v1D5iJJJbKy/hu5cp7g7jekgB3DJf7/C+YWr2SxFwl9tCNfwL8PKebPRO7Qq6PwiB8xF9BiZ7TeG/k0ZL96wXtVmnsXw/WfIGFnVOz5TrbP/6U8pUzI9XMBkbdjB6Sw+4YUb5lK7WfCs7WBBwyneIXW/MSypdd1QaKkYJOB5ctZ85S+hqPEem8gj9oFkcX4IMM1O9t7cdxUzDX8b8gz4WSUfK0l4ttfBW5L+U0ajTscKdyvJj+3ywgk+F9d6/2o6F7XegBVnWy+Vv8sI7qfIbfl0e5ZPyOiXyJhsGuMeOw7F319jVxOE8uy2LtBSB+NyrtcjtyEfPKDMUhj9TlKLOHH0R0lGNt0TtZlLtbuZqRt/wFlAD4x9m4fQ745m4f/7Uj7G6zRKyLLwhXIdaofYlwPQDEqIdZ9AiJe+/nfz2/N3lfPoJ5DVJLpVSQcDI69f7Q9vwvnmpYGK3bd1/0/BOUG+BEl41gfhRQ8gLwyQmKoStZ3DgxLG0RAdkfMbk+7v7ud+xxisB+z9XjW9ZFYmzT2nUH2+0BgBqGwhyMQ8Tug2O8znK9Xio2y+a1AfrzcSERYp6Cspd7tsDMq6zEuYW+DNroVUn59jsJrNkUKsZ2Rlv7BUutVjy0Oh7HzshpKADMVMdBLIEXFXxDt6O/7iP22HUoIGGL42wN/QAzXi0gxuRtKWPkDlqU9bTyURzPC3jaqYO3pSEvyFbfLIQXRD5SBV7LeC9fvFoiWvWFrfpjd74gUi7OQ22+ga7sk7WsjtNga+r3piSy1P1BDvJfxXMP+9kO4cyeU3CvAVSskhLxg847njBltex8EnoCTByKcvL57tx8qKTeDKHvzEwi3nlHJecbGVg69CLWPE3mBX2IrE1a2QrzzTGCHMvZiJAq9OgV5qgar91CktMmhJIzTEZ98Qkp/gfdpj7xhr0PyS+CrWxIpKr9FfNon5NPvlgnja4VyRK1GlMSwMzLIfIG8345CHmwvYLlA/HrNa63mA6jaRPMJwa3ICnUmkbBwH2IIA/B1QckF3iEKtk8izOX2dTBysUwss+GAdD1k5XkTWYxmIaFnmHt3ECL0BVl3qQDBSpjvqza/lREDMgMxEt7lrSNi9Bb3axeb64bIBfd9m+/PyEV1O3u+AkIof7eDuVYl5/lLbojxmWIItzsq3ZAzWFw79u4uwNiUfvw+n4qYgi+tv73sLAxC2acnGex8gQS3xex3FdOAx8Y3DjHe3xJltQ/M7urAGch74gEkOHaOj8/B8vIood+NOGUTii26FCnXvkfuoW8AJ1cDlmPz/TtSGnxseOU8YFn3vB9wLiKot1BYdzPgs5ZIsz0OCdLHELmDH4KE6RxiHD7FWXBpIMGOfLy3p+3jbeE82JpsRKRI/NraSyQkfyNfALvX9uFjg7MQ034YUe6Pn5EL8u5JcMLc04w9ar22zdyL01Fc74vAOe7+SIRXPqMEXslyL1xfG6IzfjcS8EPelXHu3WH23lrkKx0bip452Btla/408tbYDuH4PnZevqIGeK9Ccx2K+JMJiGa8i/Cd38c1kdt2DnlANDn4yPP6QZbHt5BAlEMeJkvbsx7IU/FORIP+iuP1qJBAwtzRi59s7xtiP+sMVkJZ3YLylEQ0cwRS4H5ieOgr5PXiy2Xth8KuDiTfuFegALT/H0fyyIsGg1Nw5YUR730kou1r4BQD8bna/zdaf0HAvwDhulZIjnrQxv0qKm/YJd7fvNZqPoCqT1gC8IQAgIjh+ILISr272/jWRNaspKx4mfTlDtFAO0R/QMLpSnbwZiEGYEWiGNDulBn/kOHa7W8HaIxdX2rzv9oO1SMkKA9S1m4RW6uziZiZ9W2+p5YYxy8eiWe8rysgBUdICPRrQ+LnGOKdSizOv4w+z0PC88WGXO+zvQ3JwjqgZHfrICIdzklVhC8UmvAOym46CDE6OUR0fO34+Fn1BCWc20EGyy8jBnImIloBrrsgQrMXKoM0wvVRFa0tyuD7NhLEOtncc8jjZjWiJI0LGF5bMjZHL0w8jLTl9yKm7weUT2FNez4UWBd5+CyNJR+kQQkpqsv7JWIMgkfDOUSVD7ogBuQgW9+QrCbJsrmqrdefkRXiBOvvafe7PihZ0tLkW6ALXKKpc5pRgb04Gyl0b0LeFDNs7Qba847Icro2KXgl672w64UMNs4kytfQH7nx3om8ZNJCURqKnhEJEIshpvwxpIx7AuG+O5G3XydqjPcymGuAlU5IyLweeaiMIFKo/ZH8igvLA0cl9eOu2yGc+VfkGbE3Uo59SkzQokTW+grNe67oxS+5zSWsjCIho7brqyVSulwFLImE1csRPbqDIhVViuCbCwzWvEv6P22Mo9y9OI5L6+8Gg4+9EQ9/oPU1u4Qwkn8WIr9CUsMo2ucKHmo9gKpOVsTuIeBguz4SCYVLWptoCOUoYnGDlezLft9kiPZWu+6OYlrOR+7S3yDGf2MK62VWHEgRA3kJcJxdH2JjWhYRilvsQL0DbFJGfwchzd5i7t5dSCgJSojF7VA2BCFuxGZwt7cRglaIMfCZd0OSi68os346EjQmoyRYAbkG4TUto2/VGEyiGODtiASPVWx8O9p1N4pkZHUEogVydfo7KnHV1fp93dZg/SLjqJZgvT1i/IM19RKkFNsTuRr+ByVrW7DU+FDc9UcGJ2HtzkPa6tREb9Xc34zWLAiv/W199rS97Q+ciISI25mD5HuIEX0WuJL8MjoTDH+mJmlMWj/qnGZUYC8GIIt18GzqiAST92wNVykX7iqwF6NR2Nfa7t6DyAUy0LNdcYq7Rm6I6X8GeXL0dvenIjqemqCtWngv4/n+2fazI6KT7yBr3cWIbvzV4z8HswXeEvb/yshq6QWt4UShDQcVg8EKzzUzevFLbM2FFfKVqKOR8Wqt2P3fIoX+41hlnTLH1hUpD08kskZvjIwpOzg4XKzM/tZDOSg2JFIIrGvzDHg6nuzzFwEnNR9AVScra9nBSCAcjpjfvYmYxJB8LEeJxBNZ9mXvL4jcaH9t148Zou2KEoFcaX19EYC2Buu3IdL290LCyW+J3EhWMwT8ASUsz/b+yfZ+cJW6ArnzDbXrbYA/USLb7vyWyb4uglm0EGN0UyCcSHh8B7mm3V2in0Ak9kLCZbD+9EVM1xV23RXYEZfkosrzXRkpwoJ77wgsiYdd90aEsGQZC/vtxeRnY25t5+QhJITtmfUc5mCurRCjdpld72FzD0mUDiZy/z+EWIkc108TUqLdZbgtKE02QoQ5lNwYCgypNUw3c80CHHe3+d3t54Rw/87IavcMUtaUVB6gGPT3cKUere83MXdBFNtbMsOrvVv3NCPDPRmBysPdGduL1shK/aTB9Y5l9pf1Xmxsax1c0OP0bDRSxqxZ67XMcD9exzH2RJ57IQnpqqSU5mukhvidvwU8bjjweaJ607ehUJpHsBJaJfq7AIXRPZTwbBHE9/yIXH+7ZTWPMueaCb34pbYsYQU4zc7TJCJeqr17vgVSgnyA0d8yxtcTGUoOseulyed9Ohvc7U8ZnmaIj5vo8NwyiOc5wa4HIq+gqmW3r5dW8wFUfcKRK9/uyIVzgHt2MrI0l8WIZNxXC+RC1R5Z0d4CVnPP90MuOeOAFWq8hkORMmETN/adUDz2GPdegUupe3Y4ErjaoeRlMxwCb2eI5QHmC9eV3ss8jSLyJrjA3VscleFaC5c9vESfhyLBI2hG30BMf1Ck7Gx9VsWVjEIr+RrIqreqzfkzZLkPiqJ1EUFcr0S/wf1pOsakx9ZzOHI/zwFn13CPhxrRa4/iRk8hEo6HEtW6/lMZfY0HrnOw8R0W94oEuquQJbssWKnXZjjtBaQ4eI1YIkrEhK6D3ON/IMFqH8d7RLWxt7LrC5ClOSR5XAxZInadgzE2BM1o5l60QoJvDtGdUQnvLIusQzliyaSy3ovYGQ95GAYil97jgN+RT89ao6Sm/8Hc1xu92XrPwFyYUXy6z3y+FLLebVyrMWY83+VRHPSqKIRqXSJLY/DmyVGaZrRycDqdWD4Te6crUYjC8lnPpYy5ZkYvfoktQ1jZGdHbHHChu+8Tyy2HFB2JHnLuu7P/IiXszYbjJiHeJ5QdXMm+uXtKf3EPvv2QsN4OebN8jpKkBSPj1kgBvVyt96XqcFDrAdRs4hKIPTHogbRMp7t3yooRzLIve3dDpP3ZwK5bIqJ9OZHrRc1cKxBTPRM4164HIVeX2907SS6le2Cue0iDFlznc0QuKa2QhWIaDW5tabRmcPawEc/OKGbxSCMGC6f8JmmfN7Q93RVZa14nsgb1QJbxO6oBw0QKns7AGg7GnkMMwitI0A+xmX3telwZfQ5CMVQ/o/jPgvAP5Mp6PgkMfw32tzvyFjnJ3VvDxt4jYX7B9d0T8xsNProjrfkdRIQ51AAuiCFrxIYE1mcNlo+jUMBugZibW/36JfQTYnDbomRnDyDvjplEwl1I/PImLjZ1DsZa1zRjLuaTF1+OQq1CGb8/Ja03oktHUYTWZrkXSLnmE6oFoSkHHGH3OiJ69h0ppXEaocXHbGv9BVIarG5zDt4rTUjZ/ipO2d7IzZ2h7ZFQHKzzLVGyy4PT4C5+7gy+DkB04wEkjMXXtwMuYVWN5jxH9GJ+ax6s4DJvu3srIH5kmuG/2flL3PtFvQdQSOeZ7voUg71JiF777OYPAw+XmFcnzKOHKOfH/yGe6l8YnUReGP8C/l7rPakJHNR6ABWfYHrM1XKI6R+P3DhuRwkHErN5Z91XiTGviKxpf0WEeVs7CEfNTX8VWtfjUJbIzxCT/SXmrpe0TiiZwTuIEV0AMUw7IwZnOsqsuRZijj7DMS3zW1X3dS3kLplDSTl+pIxkZoZcL3XXVyKG9Sei5Fh9kYVoKublQfXiji9H2TbDd9dF8f1BCdAJWSJvQQxFsEgllZ3wNRsXRhk1ZxhsD074dl24ziGXtf+gRFcrIevSNUiADJaJOJPXFiWjW8muVzRc9APwintvoPX7nLvXMExXEdzeH3ldfIe8bQqUTERa+iSGaazhuH52vQPy6sgBx9u9LigMZjpzWaaIBqAZc7MXOIUVCtcIjOH1JMS7Q378YiX3AiW1m0W+t0Bwxb8LJTe7CSknzypn7vXcUHzlrQ4PHkukTPA11Nc0WDy5FuOs8BqsY3M7C5WVHIM8s7zwkpR4cElgo1hfWyOr3/MoEVSblG/WBI8yF/Rifpt7WLHrpZEQ7bOKD0A5XT5FFucB4beUkTvD8FkOVx+dSFE5zvDgQYjOf0gC7xPr705r7VGixqMNj/6IeRWhUKn/QzxkUAg0jHI3k/2v9QAyn9AcHHYjFv9CQt8jxMqnZNmXey+1bmrs/r4o1mUm0lr9bW7mWMF1XhjV6vsLivtbLmm+sd+cgJjyNe26PWIC77N5hpJNp6St1/xW8X1tiRj141Fow0Zl/KYL0ma/TGShHokYzR9ReaDr7VxMIPLwqFpSJVQ+bAZWhsiIwjbIdTGH3LcmoMycIX4oiUFf1OaxpbvX2fqfihiPAktEFeZXbmbPbWxPpiEXrs+Jyr8keSGsbetzuF23R0Lma9bHwagcy71IiEitrtAIDVkx78MlAEOKl+DefwFzoDRFcYk5zD3WzsrvkcdOKKnyBLKSnp+0n/F9SYMt6pxmzMVeHGNwtp6718Xm+Y2dw6XLnVsWexHrbxlE7/+ECRt2/0CUb+F1RB/3dM8akp4ha/Rptu5B0TbY1uwrZPHa3dbiVeC2Rptz7MylCRbtEC37wtZiIvBYMVhBYQF3I7y7b+zZGIOhj5GlsyqxqfFxpu0Rc0gvfimtgrCyn+Go25B3SJBFFkAecBNQZu/V52CsCyHZ5N+4HBIorvpd6/MjFNIVDGTFvH8us/kEPqmn4Yb3kVLyKeTd9hJFeKl5vdV8AJlPKNIUHkmKttgjBZQVuDMugVMl+rLrcFC6kuImGutvJBJg14r3UeG164vLAJo2voTfpTEhfk4PI4th/9g7I5AHgC930pAMer01B3fdkQa1IB62uYQSabWnA5e4ewNQVtG7EHPxO2DFLL43l2P8M2IOfK3OzsjSshtKOlRUOLTzOMkIx77ufjtkDX8LWf63rgX8kpAx2e57vDYSWQB/Q3lKsfOQUsznU9gCCZwf2lqcTeSh0JCEFClcLkJa+H8Di7tnLZHC6WdkvRs9B/3eaes02K6bkDL2HPvOxbi68TF8Wdc0o8L7cSgSPP5DfuKxDqiE3nvWSioAs9iLlPU+wmBiw9g77YlZImuB8zLej562dr5u90DDneORUuIh1fjFKQAAIABJREFU4MhGgz13zpqAPinv+H3fGylmDiYKiykmkGyI6l9PIeaVh7xjQimkvao878zpxbzeKgkryFPs/9k787i5pvuPv7NHIiSRhDRIIrZIECW2Sqy1C1pqj9pqVzu11FJbWxoNitJWUBVr7RT1Qy2lRClia0ISgoYQW0Lm9fvj8z3u957n3plJnlmeyfP8cV4z99w7Z+4553u++7Iryi3xHFK+dHL3j0VGqLsXBJ8gfqQAHBr197XWm8Ttu2jJQEQnn0fKzRBm0wm5sB+DlAA7kyRha5WwUvcXqMqklGxjFrKUdo/uhcD+rqS1zXmCYcXGcvcfR4R8QNTfvtgYC3KYFmLNAsIYjSyNpxGVDXHP9CFf+M50VSFxnfyxIYd9w5xy5tpwlpaW2ByM9jJk+HuiJCmkCen6zfiPM5DrUpMsmNWE3TLeL8Dexog59HGSCwxnyJr7JMrSeZabfydkzXoeWcMy3fyqML9A9Mba2RqV8cwCx926874KsiBOoGn5lcVJx4g19LlFQtGxtn+vAptH9/fAuRGXGCvAxbYkya4yFa7xPmWM1eJoRhX3wOOj7ZDb7JvAKa6/M2Lm3gBOr9ZehJYxXjgbSyLvl6eRkNSu0c9AifXbHVkxj8hYv0x+oZEaUg48Q9Pa0qUMCHnKWI8bhxsO/RC5+S7u7i2JPL2qXmmhWvSitbXmwkqR++1R9Z2XkFV5R9JlQX+Ecw3P2jd37eHvSmQ9Xyfn2XIqXnQwHPdTg+NdSv2mtba6v0DFJuKSAaBkGjeRU2sRMcF3IDemLtUcy57xzMIopOUdkYe4kNZnjRquXUCkg5Ei4SryNZodEbMzGZcdPefZMWSUqUFJkGayADVi29pC7auHu+cRE5hr5UGlVFKxOfF4JExWhwzkvJoh7xs8vMTvUuU5h7M7ADG7McN8EXKbXTvqL6ngIa1AG4osEdONaAUNbns7R8vljVvh+Yaz29vO5Yk0rWkciH0ve+/MucZrFf3258g9cHT83/H3RmkZsOETt+1j6zkD+HH0XJMkV5Sw1qNa6B/hskXjBLEMWGvRNKMGe+HP2loo7n0GsoqE5IMdyFDyVnov3DOnobCAATiGGimMv6aBE5YV24voXn9U9uwpcnINNNL8o3O2DQpt2ixvTsjDrklejRL/4fFKf+BCxGc9jFn3yl3/Csy3IvSiNbZqwYo91zOjfygyiMxGNcZ7R/dj/ivwPsNsX3tG77yFwd05Zc438G6ZoQpI2fsaSdWkdlnv1Vpb3V+gopMRsX0MWV5vdv0x4zIcMdgP1WIs9+xeyDXkryTuIfF4g5GA8zZRdtoqr1175Jp0LyUyQaKEbU9EfXtgGb/tepTNYyqyuuyMuePa4X8NMSqLtR3Gqu/tmch9cpjry9rXfZGFqElMKeYubd87Gaycg8tO7OBgLmXWXazQ/Lra/wbk3g0JvVMNVjfGFDnI4vUMihFfpszxtyMpI+azdPZGSTsKyFV4RPS7WikUeqGEMxMp4nlgZ7uAE8KIFIJIQTIOEeIlonuPopCOJlnRG7lhOQACfLjvW9l8v8ZZTd39DkQKRqS8mYiUrV4ACwkdbyKDkSrybi2WZlRhHzoZTunkroMw0J/EdfYGohr0SDiu2F4gl/Q73D4vi9w0p6MQiTtR7oFQbvBCpDAetrDzr3ejqZB1sMHe96L+dRCOXxSStHlBaB8U65p3zgYZ/N1d6pyh5HlvkoQZebrRHYUoFZB3TNlhDRWa80LTi9bcqgErKCStgJIDLh3da4+E99ko9OQyouSoiNe5jcSi3MH27QsUpnc1cJB7/kgbq0kJOLvfhXTYXC8USncv8vbpQ5LMbl27dxbC1W1KGL+W9X6Bik5GMUH3GLBOQZr+PCvNaiQp47OeqdhY1t8XCZQF+8y02qLkOQeRU2euims3CLl+HlXkmb7ue2AqOtgBfBPLKk3CEK2MtP0PIQZ1MmKMvmeH9XUWUAvc1hZoT9shoj4Rl2CmxL72yLh/M8p8G5iDwUjr/iZSLD1uiHc4EjgfRFkoM709qjDPmwOBcX0bofjJUErpGcztDsVXTiWnNmQ09mooeceTJHGa7UjHQQUL53/IsETUYP4TbI5fYMkCM57pjISMO0jqebdHpWDWd88di4jvTKSQORGL30WxvB8CY8Pv6w3jFVi7I2ztznB93mq6FWJuvgaujX57gMHX0nbdnSSxy3uI8bgAc51DYRPTyKiJnfNuLZpmVGEvDra53ktS6qW9g9elbf0+tvPuz2DF9gIxiidg8a8kDPSSKIPyMQjHzbFzciOK1Z+K8GTD1XhHmY3/jCnWDab+jhLkzbI9+TFWHgrh0ufJqDneaA3xel8a7P2T/KzdPW2fiyoVDGYPQvHVM3A15kkEk1UQfn2HGmdVZyHpRVurCqx0sf2Ya8+vGN1fHFmIbwYuzPj9bsirx9Os3kheOR1Z12cbbroIlYB7HCkG+maMNwlVVgmKhKURT/AEUix+ghRD+xuM3Gb/kRkm2ppb3V+gWS+fnXysD3Jh/NqIxcASY7Sv9Fh595A191o7mLmuGaSFhIozsMAqGX1DECE9Mn4Hux6BtF5BExtr6nrb56rAeSSWwrCWo1HSo8koWcNLtg6n1huOFpWGhNus+MDHgPtz9rUvKqUQvAqyFEhnYa6wOKs2ytS7AxK0pyACcSmyshWAw2s0734OLvfGKWxQwrKtUImiKUYcgtLsOaR8yNW4Is+K01Dm31eIEkXZ7x+xOZ9Zp33fEIVyzEeEL5O5t3ft4r4PQsJ1cOv6trwKElZ+jxQL76Isyjsihdh0yrT6t/SGYuSvsLXz4QxhLdawfb8Wl7Xa7o0kKde0DC5ZDbA9UuK8h1yQn0UCW4C7zsXgzsZpMTSjRnvREynp3kIM2/rR/TURk/ctPqrWXpBYz1dGwuWm0f2lkJJxnJ3/T2y8i+q9jgu59qcG+MLllkHWqYORcnImUq5dgBJcfQ38tt7vXoG5d0Sedc8aHvgx+SF+Xogpxu91QGFxz9o67eXutUPeUNfhlDul8EEF57tQ9KLe+9QSWjVgxe6faOPdjlNYISv5PR7fkQ7H6kJi6LoGV8XE+gYgY9YfkfEslB6cRoa7N6oUExSbG2KWcmQhH2l44j9IVphkMFwA/lRLGG6EVvcXaNbLJ4LbDiit/Cp23cOuP0OMf8lyHZUcy4+X0T8cMXMFpKmqqZYbxYrMJbKwIQHlDeAB1xcslR2Qm9zjZCgYSCdNuMTmdhWwZvRcGG9v5BJ+Xb1haFFpSEj+AtjT9QWYvszgd5jfL0TkN0GKjm0zxowZzv0MRprUvUZZLQ+xsYLiZGwN5u1hb0f73+tx1gK71wVpgQ9B3hNvZc05b/4o+cu/kGC5v+tfGgnW3tW66sJNxt4MBX5DwjDFyRdzY3rt+krk2joo6l8eKdXuR4LJXFvjsstRtfSGGJAzUC3rp3EeF6j+7HVkuOy578sipuUMIpd5u7cTckF+x9bujnL21PXXlWbUaA/i7Lr/RtaWPa2vo53vB0nqsDaJ8avEXpC2iK+PvAbm4TwDSCszlgXWJl1CsiGZTMTv3IcE6hhHbIsUjTOAF239nqv3Oy/EHLMUyF2QIuFJw3P74FxwF2Q/I3jcCIUQFIBzkZCyGlKY3ZkFw7WYMwtIL1prqzasROPugpR0LyCZYxdkYZ5EdllQz/ssi5TABeCInPFXQ/TsTsxrjzSP4+E2lC08mqaJCvvb/K9CAvsXuNCqtmbrVO8XWOgXTwS17ZBG9Xqsppr1d0KC8jsGdJt5olmtsfx49n0gSsiyluvrj4j/V8h9tt+CzL2Z6zYUy+YMrBgdqL2RlvX6CFlsgYSKMzPG84czWAvOMyRxF87tKEIGXUm8Btq0os3f155YZmP7PtjdG4g8Bt5G2sdA0L+LLJL3FxnXM5pbIsbqZVSWY7GM5xdHAntNkK2DoeA6eiayND2AxYNbv4fTJXEJyMoZ375vhZLCzUbC6CEG49OImJMqztfHlffECRBIQXYiEhIfxtW0LDaWfb8LEdPLgeE5z2+IYsNG2XXDZQMushZLIoZmOqIBh6E6y69jWvno+TiJzUNIAPstVpIs4zdrAT901x1yvrcomlHDPfBnbYyt6efINfUaxMj+KuN3FdmLGJ8Bu9rnSMTkFoDz3f08i1VD0TPSeQaGG36bjMqMZYUJdUPuqCeT5KJoCG+JCMa6IuWot9QPRsrSeah0ZNk5DCJ86r8PM1j8Eim5pyPvvZXiZ6sw34rRi9bWqgkrRf5zdaS4+hJ5jb2EuYrjjF3R3q5p+GtNEi+nS9yYeXgqTojm59sR8VBzkadKZvgm8lRdNWu81t7q/gIL9dIJUK1ghOBsMtwUkVvdSGSNKAAbVXOs8Jz7finSwM+zg3IbiXttT+R6OYMo2VSN1rCXvdP9JHW5+6I4jTlIEJuAGJUpwO1+zRBzso7r+xvpEkcHIwvCU7js0zQY49GIzeDsG2DjsObIlWmS9f8NuVe+iJRFcVKrDkg7GyzcK2NWXhSz/E/kKnwOzqKXtbfVQLjIurIyicvSVsgaFa73QQLSv4C9y4E9mmr2MzNhI+L3GxTD+YmtxfCwblXe10BU10cxWNNQ3PtNpD1t9kaKwMlEMaVujLC367p7P0O47S5cEqNo/g15fkkzu3FoRKABnZB1LoQ1vI3LVYBoQNfot4e57+fb724mHWeZdS4y3bhpwTSjgnuRWwImWosRiB6/hpSA51VrLxBtPwlzyUSJiB4mER5XQV4DhQgmcpXsLb0huuDXe0P7XA4lL3wfudMv657xytZwbhpCyRbB2dG2v28jHHo45rWD+KBxSLD4EzkJHCkhFEf/txRKInaR/Vc409XMDN5setFaW6VhZSH+fytkHQ7hlWEvl0AW7XBWN7O9G2nXK9iZnY88UMIZzazagnJT+JJw25AueRhw3nWklb1ZeLTN28GvR71foFkvL+b+KaJELxkAtDLwy1qNZc+di4TLQ5Fgs6cB/KMkVrNuyE1tClE91SqtV8yMHEMihAQtai87YDchJu8e4Dj3myB4/QwlltnVkMsniEHxTOOWKKHMZENIi5w7Y0toRAweitN8GGk/vQvzCKSpfgwlqDmJ7FI2eyDvgyUR8Z2H3P1D/NXiKDboI2TBXdX9thYu0dsixcBByCrwFZG7LHLHew0Rw+M8ASkxdl4JupSHBiJyq2IuxFRfsA4W+oF27m4xHHMKcp+djoVh2PncCjFTf3ZjLIYssiFB2fcQ4dzJPbOr7fe/kPWwo///RmwkjEl74Dtl7G87pERazs2/I1KsXBFgBFkzP8bVnkahEwWkvNra7VvJ9aMF0owq7kUvYPcy96ITLnM9okEV3QuD9Q8R3rwV4bY1SNOz/ojufWbno8+Czr+lNMNdh5EkabzM8Eh/u+6E8szMRl4DPuylIYTpPLhCvMhnKDTtLBIB4k8k+LwH4o8KOA+onPFOICdxFUUEjmL3KjDXZtOL1twqCSturAHkJP4qFxYM7z2LlPqHIJfsC0h7mvZFSchmo2z0y2b9B8Kt+xjOGwJ8x+ZwMel48cOt/2FclvFqwu+i0Or+As16eblAvJCH7JGw1y/qy7S8VHis1ZFQOZbEVXp95HJ9sl37ciOBoFUFWGmqMLgKKQm6IkZtMrJExklb4rp6ntEYgpIkzEXMuE/Y0cEhlFWQtf89JIQ3cSVuawu9r8v6PULEc3/7Phy59xdwFh+7l+kmZPfaG0zMRdliXzGkGgiJd+0PSpVbYtip8rwHkCRkmYXcRnu49w+MxfIoT8B04Hfx+XXjBVgdixiRUVnrUu/9tve4HVlW+7u+pwx3pbwIDAa+rd2L3NqC29hxtnbjDQ/4Or9r23l9ByVsyaxz2WjN4PiZeD6l8K5bl1WRYuprg/mPSYQ7LxB+DwlnrxtMlQwZoIXRjBrsxeMoEdiAqL+sc1aNvUBhYf9D3j3jXb/HeT2Q0D7NzlG/loIbFnD917Hz/TTwC6Sg3NnueTp/HmLS76IBlTkZ817F5v1tWBNSohWAc6NnF6OEdwhSWM9CiW/juOWgmOtCnTwcaAa9qPde1btVAlZIlImjkZL0NJrGMIdnlorv5bzXtvZeXyIlX6hU1N7tXzfkkfCqvW+TrP6I5o9ChorZyFPlejJCPZDy5TNkcBtLWwb50vBT7xco6yXlKvcDdx2YjAkowVKTjIZIY3cjziW50mMVed91EJHe2K772PXv7XoJJJiuWM54zVy7lbFkRXZ9PXIpHBTmiax8/4cIrLd0NsmkHo29qR3cAmLavftYe4c0FjfE0iRerq0t9L72RPHF9yAh+wAkEH/fPTMQaTW/Af5C0zqmefsaYGIeiscaS5rh8q6Bh5OU41qiUvMrY/5dSTJf3kU6i3k7B3udkPCdWQPYwXhvpBE+MW+dkLUtM4azRnPuhlz7z3DvPR7F7wU3w4Nwrt7RHNsZPggZPl8nrZzx67YUYsIKwCH1hveFXC8vZG2DGIPNaBqj6+uX5pZSM5w2CDE0BWBitHbt3FoPIEnst24Z79piaEYN9mKUreEIcgRTZLlfvch4FduL6HwXkDfXVGSlDkq7Dm689ii52kn1Xtdm7EcXpESbjvD3FaRjrz2OP9DW5L/Ad+v97s2c9/eRYsSHvbyOkjx1RbTvKGhi5fOCxrehNagO/U3klJ0koT/7UUSpXaW5NotetPbWXFhxn4ORAuYq8j3jOiL+YzI5VYkiHPocslq/iyzqXsD+NiYbeez8jUiJGY27BhLU5+PKwgUc6uYxxObxOpahvK0VgZ96v0DJF5Qr1h3A9hn3Nia7/mgn4EfI+rJpNcYq8r7t7FAWsKRoSDP0MEmW01FIGNmmBuu3NKpz/AHSVn5KkjXaH9Zg6fyayNKZNUf7HIFcGI+wcR+xg+oJUROBK0ZGbW2h9/YYxBxNRgL0PtbvBeE+tj8fo7jFsuqKo6RkHyLNZ1ZGTP8fY7AkeTWce3+UfOs0e7+ncMna7BmvIAuxS1kuob1QPNxEovI/0XP32rleoxJzWMD5tkfuXp8Ax1pfUGxsY9f9bA6nEWmWo3W5zM7rN4gxXCJ61u/tdTS4QIeE5n2Qu29Xf889O8j29m6KJKZBQu4/kAU8uAd2j/7Pu6EfXc570oJoRpX3ZC+Uff6vRfZiMEnMe5ZCLKzzEnbuK7kXG6BEn48gmvlbnOeX4Yoto980qpt0D8TXzEJ5Vn6CY5ojPLA1GYn9Gq2hRGxfkdTzfhgp3Zax61FIIC3K6yFly2PIaHBzDJvuejjC0Q/VeJ7NohdtrTKwYvvwIOId+rj+rCzktwFP5IwTw9WuyAByH1LCXkG6BGl7w2Pf5qYgX6m8ARLAbzY8ehlNQ0gDHv3WQp81h7bm1qzeL1DWSyZame1IJxNZCtVOnoMYjw2RNfVYxHCfW82x7HdNCCuy1D6Psu09jASUAeH/kRX8EWroKkQiGLxIuo5ebKG/wJ4bkzFGnqVzMVvPqSjOdTOkieuAyh5dVGqMtrbQ+xpifT4CjnL93vLQHfgB0k5eVWK8gHB7IoFjJUO6XyMi3Ms9uzq1dQf3oQZLkNSZ3gJpU6caHIbnvo88UppkvI3GnWBr+AUuu330TGdkfbiDOjIiiBl6CcVLzgcOdfd2Q943W5YY47vIUhqULv8kUroQuZFl4blGaIiB/NL29595e2fwPp6c2MkIDlZBIQfHI3rxKOksvEtgiq6s9ctaS1ogzajCXvQ1+lCwz+VznuuBLGr7ZdyLE/INtLbAe0HxOsWdkbL5Y0TDvmu45lok0DfqefBw2BkJCMMMB85HbuBeCOhGZK1ulLln7S/y8pqFcoj8FlnkQ+zx4tb/JJbktcjY/ZDXWAHlPxhBtsDUDpVACnxnTfkfKkAvWkOrFqwgHuplHG+W8Uxf9z3UrM5LurkxMmb4cpG/R27dt2JJhoHNbW/XK2PuixnOXQ7Vsf4KJTr2YQNbYkoa/z5trci61vsFSmx6e9IWl1MNmf2DRKPf3/rfQJaY+YgxGReNU7Gxct71RNJC63EG8F8Bh1vf6qiW7BySBGLVToQU3N4fNmIwBbmUjHHPeAG7CxmxVaSt0X2QS9mK0e/WQzFcHyJG9Tpbx4bMatuSGwmDeDqytj2FXJjOJWE8YwS9Zh5SzINru/cdg9v5SEO6KmLIZgB/rPF8N7MzeSEuNghlr3/UztxpwP4ohmhCGWNvSBLD/QdyEu8h5rpJ2EiV5uu9Q9Z3/bvbPs8HbrG+LkgROBP4dd7eoripY911N+SV8ybygNiAhOHOVLA1WiPJlP+srdmPyS9N4pO4lExEhoTekGn3TcN/yztYKuo6RwulGVXci/YGWyHu/5wizzbJph71XQac6q5D1uOQEbnoXpCmZ5uh6hanoCSO/t65iJ5NRbTtM2BoKdhoiS2a15GIYQ54ZimU1PUbpFRY3tb0bGTdLipstuSGxZLb9+4ok/xsg4sfWP8yKDHZlySx+369mpTZQnzQz5Hi+c/kuPNmrX8V5lgxetGaW3NgBcvAHo03xHDGkXYd82Qj7Cz2i+HMPRN4ny1tLy+J/wvxPP9DlvVr7MxeWQoOaVoppi8KMZxpY22BQqpmA3+o9/40Uqv7C5T1kgLyoSjO4UAUZ/AuiRapM9Jeb4dioVZ2v42BuSJjRch2NNJwP4Gr74tS+P/XAPUFFOv8H5IC7jVjkkgSUo1Fmsw3SJcu6Yms1j7zc6qenn3/GdLWFZCl5U9u7A7I0vlH5Hb6XFgPGrSET6M0pB2/w5DqNQ5Zd0BM/AFFfhuE8e5I2LocWcRHu2d6o/iiOQbrbwNPu/vVzHwa4HBlQ/IXkWFhRprmPxlsvgvc4O7llv+x66GoxFYQsOPkNDXT1JIIE0vYGXoFF8qCBIEZhlsmIrezl4GbMsbwCVWmIfevjdxznRGj9SRiEh8wfHVNvWF6IdcuiznpgsqaPEl2iENJhROyFpyNypzsRpLgphtymQ2uyS/b+Vg7Y4yGohnV2AvrH46UdAWiLP/ljIXyR7yA6JV3Y+6KGNAQh527F+435yBl3RRb6ynAD0mXp9nD/vM6Ema6YfYhY86/RPj7UpxXDxKmf2JrNhW5nH5JIlQ0nLUKKU5D0rsQPz8A8SjTkcfTJXYOpwIHZsEKifC6AwqF86WsDkUC1MMoLK7W1umK0YvW3JoDK0gAnUuUswN5OLwBPOD6vq1XjfD94+THWYe9XQm5+P+SnLwvhqeeQIrkS7P2lrTHz9mIbk0wGAmGuMURvXsO4dH3KRL60NZy4KneL5ADJJuQzj79HJYMC8VA/wBpVeYCPywyTrtKjhVdt3fvejiKzyogAn2ge24jpAkaj+rTDc8ar8LrFxOGWDu1JbLYv4M0cysjZucDMlxoSQhLILwnImHuSSSM/BNTTrjfrEDi1th2GCuzrzEM9oyu+yIXoXcRY7Q1IvwFLFFSMVhBSe1esjPyosGHV8B0Qx4LVyKLeYhvrLriBFkgn0AWAu+aPgrYnrS2fh1cxm8ylGI2l56kXUj7GWx/jhilZasxlxLz9MlyNkAKrNmIuPu92BhZ1e5DyoB9MsYIc+2PrG/jSJcq8h4rKyFh5e8412gaSIiIYLkryjfR3fUNRrG+85CVMje2Ohr3dMScPoUE3emI4RoY1gh5c5yIypiEGqSxYrfF0oxqwbF9H4g8S3yd1P4o0dJXKCFhZib/jHFPRszeRiQKjiWR8i8wh4NRmcHMvXDn4gAkFO2OeIGLbT++tL30rpdd3f41zD5kzPlAm/OOJOXdgjdOwBsbIgHsTmCPRpuzm8cKyB06hCI8RpLToCuwL6JlTyAly1ZujHYZ422HlF7Xh3GsvxMSut9Bgu1m1Ch8gwrSi9bYKgUrSDm/tV2vSJoW7Y2E9utJK3W3sH06s8Q7tkfx2HfieD4UOvNL4BLX14e0wqxJOBJyA38ZeeJcaDDzHuIdQzx5R4S392qDlYWErXq/QAYgdUFM3ocGxPchzc/gCFA2RnEBBeAE19+uGmNF7+iZpM8NQA9Dlr2pSCv887zfV3n9/KH+CbLkTUIuhz7hwTAUo/EpQsYfYLFVWQcIaWSnYklhkKaugBD1NMQgtvrYnSrvrdee/wlZXG5BJXtCaEN7g73/IoXRx7g4qyJjX4cE6zXsehJyM/qYbFfjDv6zBnMfZmc3WFG6IC3ydIPDN8lIqhadh/DO66M48uBCehNpS0Rw9Z2MMee13F/7/hiyIt+ISuV8gAS887G44ZxzmmWd+x1SnHjC/FNkjbsWp3gh7RrdMIQ0WrujkXLkbdvfw0mqI/RFSoa5doaWKzYe8uSYA+xt18cg19n3ET2JY9MbjmZUYS/8mbsUKa/nIVpzG0mm4p5IwJ2B6EepskfLIMWfdwcfbmO+gTx3vuvuFasxPARZv4+3612Qonh/xEQXEO+QGRfeiA2F97wAnOL6eiJvjFtRvLWvuuBrizcEXJLg+KFIyXwTipe9C3mtvEmUFDDGcxEuCXhgBcQnnU1GuSREd0cigaWA8w6q4lyrQi9aS6s0rNh1L8Nz92P0FtGc0xEdeRtZih9CXjK3543l+nsgQ9a5dr0sot0fIWXA57gEkWWMdy8KoQvKtcn2Lv+zcYYXW6+2ViZ81fsFcjZxccTsz0Ga7d3dPZ+oaXWkbSkA11d7rIxxJ6FkJ946MsIA9yvE/Pe2/poiMUOw/0V1Dn9p87oJWfoCsg0xgweQJELIRMAo+dGfkZZ2KNLuhyyUJ9n47+EId1ur6H567fn7iBk6AiXceBFZJHxijM1RrOn3yxh7e1zWS2SVm4bqKYZEeNdSx7g75F3xPooZ3tvmPxslGQtljHKTtZEINwORwuAWpMk/BQk300lCQ7qg+ORpwJ/rMNczEHO0Bokb12ooZ8IcFEcaNMyhXEYxQWKCzbcTiqW8wcb5O3I1e4oGL61BwgQfjixzF6Mu6NcsAAAgAElEQVQSJeNJMkmH8JUeJIkAtyoyZn+Ey0+16zH2m70Rfi0gF7ydynzHFk0zqrAn59q5PNRw0Z5IgPWMXTdgJ8TcFa2jjKzTbyG81xe5b89GXljjEJN5bZnvNhKFvwxGlqZPkFW8PeIF3rP9fRz4Tr3XskL70Q8x0gciy9S6tp6vIVf6b4Cf1fs9KzDPjkgJcytp74N9kaJnFrCv6y+ZoBIZZ54iUrbEeBfRqV/WeL4VpRetqVUCVmgq1B6DPBz+RZInoxcySN1k494DHOd+kyu42n5dj4wfl6PElu8hT4U+SHn+7zLheGekABph1/cjmrS0jVMw/LxDvfem0VvdXyBj84MQERiZzxCD70tqeaF4AGKkmpTZqORYGWMvi4TX0+za15dbDjELH6MYl9zaqVVaw50Rs7C7XR+BmJrZyK1xRxawJjESqPe377cZMgpMYEdkOXjFDn4b4q7Ovg5E2uhf2vU6iCGajBjz0xcG1hCTejUivHsjhngTuzcECbUf2f/0rtX+xv+DQhfmIaH3EZyG1eDxt8WIlD13O9LO9nd9TyGLjs+O2RFZxdplvUsV59wFKTIe8e9in71RCEao7V1WokAUS/4Zsmo8hxQJ69u9nWxvyyrR1pIbyuL9DrIGB5fh5Wy9zo2eXazU+tl5ux15OvS3dfuFu38TYqJmUkbyN1owzajCXqxucx1LUgpmfeQeebJdBzfuTuE8ljpniCmdjdwapyHLeFjDKxB9z2KAY1zSm8Rl/DxkSfLKyUDj7qPBlRwO1pZBgvSjto5v2mcQvO43vNLQ9NvO2dvAMXbtw192NXwwz5/lMsa8FtGITPqClDX9or5ahEtVnF60ptYcWKGpouUqpFzpiujqZGQR3zR6rnd0nWXQivHV5sjb7mWUzNEnE/454ocya61H46yNcvJ0QXTyHRLDWntkqAmeqIPqvT+N3Or+AkWAYCCynIXYtGew+B8HCH3suSZuPJUeKyawiCF4zohTYBLaOcR2M3LbfQe5ANZEW2gI9S4SASy4b+9mB2uGHfr9Y2JQbA2jtXoGp51FWtJHkVV1yaz1amvN3td2yFp2t8FeP8SIX4Rc+25Frq4XYlrJYnuY0Rfii/6KLJ0hE/eSiOm/GTipRnMNVuZOyLLlhd5NUHzrd9wz6xlcH1xi3G5IS3uG+4/xhhOCq+pBwLql1qsa83XXV9t8fHxW8DbZG2nT30CuYX3L/I+LkWvZ2SQu8B0RTnyREpluG6Gh0mvTsKRT1vc6ilXravM9KsZvDhaymJwgeOyHGJtVSfD9tcgK8GcyknJl7GuLpBlV2ovgTbKxXfex69/b9RLIQyazhnqxM4cs4r8gndm3l63rtcXWDOUWiHMSXI0Ep5DEKJRZ2r0YbLTklrcGSOnxGFLgnOj6eyBF0lU0WPLRjPPcHxkXfAlQb0S5FVn6piJvBc/nXYqFHvnfIZr4BhnVIhDfeCPwoxrNt+r0YlFtlYIVJERfCVxo19fjBFJEazZC4VhfYYYpv39ZZ9Thpr6Gq7YmMWJ1QElnA57qCnwPKXdPLgdWrC+EEN6AaFeA6T4o/vo3WBhUW2sGrNX7BYoBgbu3rQHu64g56oIY5QdwwfzVGCvj99eQuENfgCxCB2JCpfV3RVr03ZDw8zk1it1CTMuxKANwH6SVu8Ah20uQsF0gqv+ZMdavgMvddSeb23PAM9bXG2UafAvnelRveFrUGmK0x5B4D9yNLA3B1XUM0rAWEOOUp2EPxKEXaUtNOzsLk3Cu0MhyfR9pK3E1S4oEQWNFO0MvI2+LiTStxbwYEqj+Q4lslrZ+fREhDef3cFuzbey6n/3PadSoljUJMR0JnG3fd0NhFz/Dabltjw62dTkJeaNsljFWfxT+sSuwrV8v970LSlw0FTi/3vBdobXcDTEyfez6YcQYBbw0Chf+UGQcn707MEKnAp+5/m7IRW8vSuQfoIXTjCrsQzs7lwUSpd2Lth+BsRuFXK63yfi9j9ne0s7B6eRUPEAeZ+cjZWNww8zCASEXxbY4ZQiyXH+OFM6bIfr5EVGSzkZppIXFoSjsZ1eSEmIdos9eyLvtUxJrfsPRcOSOO9K+X4WUVDt6uELCyW0GU/chpUpIzNkfxe1vnzH2xsjr4tqovxPKy/BeKbxSoTlWjF605lYBWFka8QofWP+nmGdAdP6GI8H7a+C8Eu/kXfr/gXiVAvKYOh9XfgvRn0OQq7jP+p6VL2Al0knJOlr7P+BZ178qUsAPyxqvrS0gjNX7BWwDfXKxEcitebPomeHIYjoDxQo+aIAd16+s2FgZ7/kDpAX0gPo3xCydiWJeeiErxxzEQGyAGPi1yl2PCqxnyNI9BrmBreHuHY80bkeRbW0JB/I7SNNZIKpvh9zKP0TueS8hi2nJpFltrdn72hMJRf2RC77PMLwe8ljYGbPCZvzeM61hb0dEz5yLiPDByF38dqSI+tbSVsX5+ZIi7yCr4J+QRe89RGSCINwOWWOnkE4KUsotfKLB7GE2z0Pdvd2Qhr+mifmQkuANZC0Lnh83IEHxDBKivSpSAv7OrmdjMZIkTNcKSJD5n+1vyOYf16Q/2/rvzIKPlt6y3hW5+M1CCsTfIo1+iKNf3PqfpEjuABJXwF9E/esjget25KHzC+ALSjCrNAjNaOZeZFn8F0dWkAeQUP0qCV3qhSx9j1AkqzJSRHyAaM0c25cncEoHJHzfjTwWto/fh4SeLYuE5wJSsOxNWrlxh9373Pb5J/73jdTcnA81vPK1a+eQFsDWB36N8OvB1tcweMDNY3Pbv8Psej1EG14kTSc3NlgZgZQsBRy9xCoI2Bm/1fUvhXLXzEFKoQ2RASMoYs6t9hzduzSbXrTmVilYsWdCTpoXccktaerZcIE9lxk+5M5sDySX3IwUsJuhfClf2WdQkB2G5Jar3Rge733rcYM8hObjkjXbvbGIDt1lcPOt0aytVQDO6v4CaYb/ckSEZ6BY0stRHJ3XsP4BMUgPkWipO1Z6LDeOT19/AnIbWoy06+qViHB9hpDvHCzDMsrYPQMn4FZj7XLu72KHMpQI6IqSFlzhDrNHAmGum5K4yn1EkrgnJKHpgYjPFdZ8qbOGY0YarSFG8T1cFm/EnD+NWe2KwQpyV5pAEk+0q7s3HBHqgsHx61itd2rkGokE/8dJu1iNREz6bJKs9ssiJrlb/H4OvkeQLtO1OyKY84FbrK+LwfxMMjKjV2mOfj82R4RyZdIuaZfZPkxDypQpSMGwFIrRfYe0cqAT8iC5HSVkWxGVFHrD5hYywa+FXKV/737bUG6v7r29a3B3pOGfbfsbMssvg3D3lyR1iv36B1y+Akpw91+SSggBtpZEwvAUu/cOcKSHtXgtaYE0o8p7cSJpBvM424uvgMOtb3Xk8j6HhOZmCee7I0F3rK390khYfNv2YDl7bitEq3aM98Ltw9bIA+ZmJNBPtf8/AVduD9WJ3Q0Y7foaip6R4L0tkNL7LOSivzVSOM0DrnPPH4Vo+xGNNmcSvq8rUhqMJ53hfHXb9w/tTE9C1sAb7f6hBkurEIVgIE+VArIgBm+L/tb/BuIp5yMec5z7XVUUE1SBXrSmVklYsetgbHgYKTmmIOF0TPyf9r0LJRI22nPnIuX/ClH/iba357nxRpCRCT76vj1SAMyy31/h1qK/rcWrBif3kihrGpIfaEmt7i/ggOBXiAHcGTFDoaTB0yirpS8RswSJe0YWYa7YWNa/LIpTfj5CpB6It0Ca8ZNI4sxWQsxAbhbjZqyXJwSbISK5A2nXkXUM0T6KXN8uQoxOkwzSJER5CGJqLkCau76IQL9ra7pO9LsmTGpba/beBqbwu4hBPRFZs0KSphAfNxO5NI1Dlsozyxh7nJ2DB5BG80M7G2e5Z5ZBFrXRJNamWgnW3eycjY9gvB0SsKch6/Ni0e+yhKUlEMF7Befqh6zyMxARnYgYlZdJu1fVBJYNV/0BeND1eYZpe3vmZuTWOtD6L0aJ5nxitiB8DHVr0MVw06vIRbqPrWUvovq2jdaQ9ehrpNEPcWgDDK6nI8XQJcjaORWzSkSwEs7aJvbMXYgBeQkJvv+HWf1tLYfbs2u6MbJckFsczajC+vvzORq5ZT9B2q3+aDtnM5Gy9i0UxrFlHuwh5cOtyJrss6p3QV5T/0MKwI4Gy4tlvZNdL2fvdQEJLluRxJozDrMGZbxHQ9IzJEDcg7wDurn+JQ1HfMukhzPTqHNGtPD3Nt+zAgyQCD+Lk3jr3Uii5Olj8HhrNF53hD+7Isvhu9aCB0xnZInczmBxZffbquNRKkgvWlurNKzYvRCWNxbRjDdI1xfvabhn1XLgBBk9vKt2Z/f9N0hZuXL0m7z8CrcgD93LgR8j3PwxCidc1p7pivi9gSQ0tKFyLrTUVvcXsM38PtL8BEvDyUhrdDjKaPkacsvoXsux3JiDkYaqYMDqXSwzARE40p59tEprFoThAxET+LG935OkXRC/b3OeZZ8H+99njDsOMYX93X90RpaEWUjzvYt7vqGIcUtvbs1Xtz19x/b1fVzdVRIGdLrt12/KGHtnZL3bjkShtB6KCS0Af6z33I0YvY/F+uMyKtv1DTbfzBAOEmGpI1JIPI8I0nTSRG9jpCW+DzEr+8Rj1GC+SyLBvoDcX0e6/c9T9K2B3Dg/JXJfN1z3IYm11WvOj0LCxHr1hvFmrpl3fz/McFoBJWkK8b1dUSmVKxFDcQ6u5FaM+5CiYRoSxAOztDiK852GFC+j8mA2p7/F0Ywq7IWv3R2ShRaQAO3dKzdC5R7HI2+q4SXWrh1SAN7hrj0OuBoJPJleOtFY+yHBPg5/6Yy8uOYjQfu75cy5URpSKl4fw5vB+r2oTNCS0W8awmIdvfO6JDlk7iFtjcw7Z9sgJdpk5LHkPe+eA35l3zuhsI5/Iy+AHxZ5j1okq60ovWhtrQKwEte47hJdb4k8Hd5BHlQrI0XtB5jgWsY7Xo5ozhC3t8HSPBYZx1YqY5w9kJFsYxLlQW8UzvSRwfnIeu/Jotzq/wKKCdjPgLADYvznoti3Tihov2DE9HiKx2hVbKyMsZdDlt8Ccr9Zxt3LslzshYh2l3L/YyHWbnnEyByPNE+bIwXCf5F7Y3Dx6IMsIssWe2frvxJ4zV17rejpDjkdX2/YWVSbwe7dSMu6EiKqtxhS/LNHisgK06vMcX9u8LEMaevdEBQaUUBukwFu6sJsISHnc9LJuAKhuRjlSWgSN0vakvYYss7faATlA2StPp8MVyr3u1rXo/8OcjeejxQHy2XNx/Vtiwj4Xhn3QkypT3IWCOsQpFhpkkCqURqJYD3UcPhNyNX1Ljsbb8bzi/c4Z01HIsXhmPg3JO6h75JWwJQ8G7RAmlHBvfCC9efI1fswpMSZiqzvP19YHIJo+Me4uHMSJvMoXALNEuOE0KgNMvZ2PSR0FJDQEhSODSdkerhCCsl/4OInSdPxXyAeoaRyohEacut+3PbxSNIx5XFm7a7IAng/UoBdgBSS5yBF6xvW7+FkY3u+gItbrQecUEF60RpbM2DF884/QblgJiHPwiHu3jBk9PgUKfU/IAljK6m0J6nuczHy1PE8zVg7t5k5daJxjkb8TsghEPiA9ihMqYCMkDuXGqutLSSs1fsFbMOHIxe/DgbIl5K4wPZEyQImYZrsao5VDGGi2JWTkMB+O5F7RsbzZQvvC7BWXijqjdx//MEfjNzA37cDOmgB53igHbytXV8QRrZHGu8JyBK4Ub1hZ1FpGYj9Dzjh0vrOQwzn3/z+LMB/nIsErMAYdyQRWkcaMn4XJbqquvtYPGfXP8rO6WtYhmAjNGsacbm4xLhnGFFbg3QGzntQrOVVuMz21LnkEbImHYkSZN2Hq6+d83y/8O7R51CkYX8al7nWcOEOtiajqzGHGq5VR6R1vxVX1xNZqv+NhOR9XX/JrO8oBr0A7On6vDDyLxKXv10yft+iaUYV92JxREtvIO2+PQLRoK+QoiyUkslKQpen5B1juOhe0qXVuhlu/BdWLrDEO66JPLsuc30B/w1Brue72TMlvX9aaovXESlqPkdeST1cf1dkHSua2K/RGgpfu8PO2ZmUFjq9h8p1Rhe+AnZ3z3gcsDpSdhcwj4A6znWh6EVbax6s2PegmLodJbgrICXvKBI+uRfKBXMASQ3pWMnrlTcd/X8jI9Z8ZBgYjXj67ZBC8U9lznE/ooS17v1WQPz7U4jPamhvtpbaav+HxZFAZxSP9VvXtw6Ke2tSCqiSY8Xvh6wOo5Gg7uNqugN7orivfyLX01q5kfoDOQwVqn+OpPRMODxdEcPzAbKArlJkzJgoD7MxnwK28/+NLP83IPear2jLPFmpffUKk3UN5p4hSUTnYwoPInEFL0sb7WB6dRQicQVJfE07a2OQtfc4JKCMrfKcA0FZHIUcHIOIUUgesyVKtPMlYgSfQMLNP7LWzfV1QbVuH8n4r952ZgvI2jms0vMqMl/vrt4TMfa+FFo3lJ39XUTwRpHvqhbG6mzjeeI/FlliJiE38YFI8HwJl1W9URuKZX4bOMbvrX0Pmb7nEWX7zhkrnIvFkDD4DGmvkA7IyyOci1eAiUXGaXE0owZ78V/gNLv+NoTD1mIKUgZOBAbnrZt9H2Cw6nOGHIU8Ev6D8n7sgwScuVhcdxYOyPif4wwuLnG4oD0SQCfbWbzT45ZGaQ72OhvuCzxATyREf4h4nlGoJu4xiHbvFe9BS28kfF935LW3HFFtZ+TJUjA4WS1vDPseYHWM/eYzpMzxikkvYA9ABoujazDXitGL1tgqDSt2vTPin3a36yOQEDzbcNSOOFfzIu8WcFCIx34YCejHAkvbvaOQgWwu4oGmAPfmvVvGvIciD6LbSJfW6ohyNN2DlIpzgFPrvV+LYqvtn6URW3ecS4b19UEa6ceAtVEpgQsMcIN7Q7tKj+V+ExDadoawZpEkafExZF2Qq9ArBvxFrREVWjvPiEy0d5uJtJc/9u/mvo9DjGjfjPHCIeyBNG2eqdkBxX28hJJjbICsL7OAU+yZ14Ez6g3Ai1JDbpSzEJGfSzrBlt/XLRFRzcw+WQTxLokE63dR/GNI1NQPxWXdZddTicqvVXieHpafREqgT1ByqtdJrNUroZCH+5GS6CQSLxQvVMVE8GpkhU8RU/vc29b4DZSVvMnZqMJ8vVJsAkkM3wPAIf4dkWD2LMpGm1VvNeCo7siCNwnFlU/AMoyiDMpPIia6gDLL35n1Pi290RRH9zdYucj1eQb4VlvfqUi5UJbwgJQScwzWtgvjIlrwNsKRpyNGqp/7XYulGVXYi/icdUKK2HtwpfpImMebUSK9d5DbeJyROdDy/QyOPzVcMBFLAIQEnwdJSnE9iXkYkC0oDUT0al/kubKkvef5iFa+glz0J9r/nWO/u8H2vlu5MFPv5ua8GbLEvYR4noMR494J0ZTJJOFc/6UBmWk317VIwkBeAv5KJBghwWQuwqNNlDoZYw9E7tMhb8AzwB4e7hE/OTALfqs1V/veLHrRGls1YAUp5u8CfmnXwX17NyRfzLBztj9FPAUi+HkB4cZH7P3eR15nwdodPBSOQtn+Aw9TjPfxsHMMEv5vB3ayvtWQp8Zjdv0UVjWlrVUYDmv6Z4lAdyoikp8iZn8jEkI7GhHBTxBTOJsks2hWRuBmjxWNNwwxR+MQgTreAPRtVBs2MA4dUTbnEyq5RmWs4Sl2kA9BzN6/7JD7UhpeEAvur37twhxWRUR5OhLUHyWxCKyHGNWQwn8mcIXdG4msMPvXG4AXlUaSbfcYlNlxAiKY95K2VAbYzoyzJo28RyFL96kkJYg6o7jOabbvdyPh7CuU/K4dYnTPpwrMQ/R+v0NEZVMUO74uYqTnkdSgzHLTalJ2wmDybPu+G9L2/ox0TFU7xHhegQT1+ZSoU1zh+d6ClHtjETM3x/bg1yRKgw6IAN5KkcQlCOdNRkLBtSRZbYNWfVmEn3a1c97V+hvSumHnYqR9vwoxJTu6++2RwuE22/f7EPPi3ZUDrAxB3hInkmZWf2h47V3EYD+OFF2X2P1fk/YUagiaUYW9uAY41r5fYGt0IOm60V3tnO2G4s4/x9Wnds9tjnDPFbZu59jeTseS0CHlXx+DY+/inErmhNxjXyFJrPY+sgitQZI1/26kwHuKRFG8hj17er3XdiH2YhjicR40XHAPSQz5au6ZLZCiwivRGyIZqTtn/Qwubja4uZgkdnSH6Dc/IiO7c6l5I7z8lsHIUQY33ZBge0kN5loVetFaWqVhxT2zBLIsb2q46G3DfUHgvYREgbVPsXez7z9BvLs/j0fYfj9FTrIxMmQg+74P4qeuRTQn0Pt9bR0+RUqG9xBPP9Rg53l794ZQKDZSqwWwb0I6EcyRSNidYMDwCWISx5K4qg5BiQSOxVx0EPNUsbFy3rULSux0jV0PtDH/iJIozUNMQHDdaFJTs0prGBBGX+Tq5bMer0tSl/gS1981fsdorFCs/g4ksO+HXBa/ILHchPqio0mE9MFIy/f3egNvozcSprAbUpZcicURInfUUCf2VZIs4R3z4NfvN/BTRADet8+CwUmwVu+F8hFMQpbewMjuazBQsiZjM+feDVnPD8mA0QmI4S4rmyVy630DMZYhGdsNNsYZmFsUYs4fAH5n17OpUWgDYtSmAOva9W8RoXuAJDYylMdoR5R0jTQh3RIJET7Z00pIEfMJC5jZuqU3xBwVSBQu69lavkjaOrwxUhqNQMxoAUv+4tZxRfvtGyQeE69h+Q0MRn6NhLDbSGpZD7HfjY/erUXSjCruxQ8Qc+bp8N+QgH0mElR7IXoyByn4NrB1WMue95aXMxDeC+EgHQyWH0MK3RVz3iPGGYsjofwag5cONnaBKJaatOfBJkiR8ki913YB9iDAcjuEK28gHRoyBjHTfyMnhp8GEayjd77XWsCNweoXKkIcVGy9Mr6PQK6+m0XPDzeYmGFn+EHDFZkVKqo012bRi9beKgErGfdCKb8xiGas4e6FEl5HYbx3kXHWRcL+NYif8zC5l53dMxZgrhcZrnwaKYQ/RMk9R9n9oUiYH4dkppWt/ySDpUWqUkJLadUdXALcm4akjkRa/atRrcWgzV8BCcSzkQCcmQG0wmPlJVFZ25BpQGj/Bm6z70sjbeb/DIBrFq9p/z8axU49RVQSAjF+FyJL5z2UYZ1CyojHsTqI1jcRIfQhOb/ZDRHsf5XzH22trH0NtZj/gQl97l43pJF8y+CurFIaiGH80hD9iiTugQWcu5s96+N/xtr/VNWCg3IFfIwEoYNcf1D8LI8UAuPsOst67YXNzREDtDJpF+HLbM7TkDA6BWlxl0IxWO8Ah9Zgj/sgC9rJdn0sEjyGI8VAyF76ICWSBCJL9F7A9UTJumxO/8KSFeXhuUZoDi67ImvmeNKlU1ZHJbI+RK6uk5BQe6PdP9T221sGOhoc3Gy/7w6sb2t/PYliK/Zq2gZ59Tyd8Z4tlmZUcC+CMLcYcAKWJNSd106IsfwaCdlzrP3a7v8E0W3PjHZC9PvaLHyDXDpnIKtK+3hPMp7/KVK2+BrvVyGeYSW7HmSfPonjn2z/Go6eoRjPccBRGXu1FeIHFom8KChe/H4s+RJSvjyOPLF2tDP8FVbCMWcMTzMuR0rrGbZOl6NM0mH9eqGwmyeR8izAUNXhhArSi9bYmgsrZeCaUIEg5MTpivjpKxxuycuX8j37/xn+/0nzLdci3qSchI07Ibr3IxJvhiPtP/Yhm3daGXkmfgTsV+/9WlRb9f9A2TqfREztr+z79h6gEPG8D7nSpQquV2usIu87xg7Lgci6sTqJ9utGFMv3KhbfWLONknD9kR2aK4gyfSI3mGPt/rgi44TD/zdkXQlM7GlIixXch/eiaR3d5RDTWvU41dbSkMXlAdu3V3HJ9ux+exRv8ypiFItpVtshAeKPyKWstxvjDSRUdCAdNhDgYSdEkM6vwZy3RsqEAmJuu0Xv0gUJS9eWMdavEBP0oOvzhGp7e+ZmpGAYaP0XI6t+1bOi2//tgpQeQXFwJIl3yR52tj8GjisyxiCS0kFPI+EwJXigxE//q9W8qrxmPVCymXuAsxyMB1y/OInF4EbgcOvvg1zCb43G2xUJu2uTCGBXIkYmWLgHRb9ZBikd7yO/tnqLpBkV3otlkeL6eRx9IW112QLFHJ4EbGx9Kxm8XxWNtxOyZheAS8NYESz/H/Bwme/3K+BVd/1zRM82suvv29nwYSLdkct0ySRELa0h+hzcUM+1Pu8V0QV5mN1c73et0Hx7oxwK3VGo01ukMyHfhBRns4E1y4CVmchqvYzh0oBT1yVNH5cgKdFWM6swFaAXrbU1B1aiM7QZMlDsQFpJuw5S2D+KKrhchIXVlfl+v3Bn14c2dSAplfVvitTGJuGVLkMW+j5u7h9hHlaIFsZx5kNtDQ4r533b2kLCYVUHTxiYXigb5ywiCxxNE3B9SyyqNVY0bpMEaXZ9NjDZj420m7sD38n6TRXXMbzjEKSF+xIl31gqeq47LsN33p7YXP6DMTwkWXZDBtHFkVXhShaB2p8ttZGOpT7L9uA6IgHbnhlJGXVd7dkHMOuZXT+CrON97foXSLiMy0PUJMkSEpDWQ4JHAeUO8Gd3BSQEXVhinCWR9r6A3PZGEsVhZvxmDeT2+yllegJUYo/d9YbIvc9npf4JUhBsUmKsTsi74GFkbTk445lDkLDYBIYarSFGNzAh95C2XOdZBrZBiWcmZ+D0/RDNCAktjyWdeXp9W9uVo9/1wCkzaQCaUYW9GGxrU0Ausyu6e3l7caQ9+2jGvW4Gy88gxnTn6H5X5JZ/Hy5uvsj7nWJw35GEnu1t9zobzrsTc89v9IYU3ccjJn8yMDS639HOzB1E9XIbpblzljpvyGjyFAlv0t3meTJJTeE878TvI4+WH9j1ycjydzhSXr+GQkpKwlyV5tZFirkAACAASURBVFwxetGaWqVgxf3+QOSF8zFJIkUfCvN9g5VZ9nlwMbiL9xaF18xHOWd88rwVEK9WllIMhS/d765fRUaSMN8TEZ8VJ3yuWYhDa23V/4M0QF1kgPoiptm2fs9YH5BHACs8Vjh8S9qBiwXV/W38PZArzraG5H5UgzUrFlPbE2X/m4fqFjdJEhOvVc79c5Hm7nDE3Jzm7m2JhJuySj21tYrt++G2rw+hxHwLpCl3MH0H8Kx9v9Tgdqhd90OxPleSWNdqxnhFhGwVe7+CvfOhSFN8J7IAlnw/4Dso1nM+ij0sVbNyW2Q1rwpsx3vm8ZFdj7T5HmfXA4A/46z0Wec/WrfNEQENiokeiIFeDSnfHqk3LFdwPVchcYM8krTlMRZuu6L8Gvdj8XHR/Z1tnOEk7nmBKepk5+9R0mW0spJotjiaUaO9WI6E7l6GU/blnLW9kBdNfAY8LAeX+wISkJe3Pd8fVzKqjHcbZWP8BSmfT7f+DigWfwY1CAGp8X70RskvpyF6vSmytHa1dZ2D5QxopIaLKY/6g5ddsBQGZfGayBK5g3s2MxEmUrBdYd+3Q8q1Xe38H2Iw9C5SXFS95jxVohetpVUDVgwHvWUwMBDR2zdR+NEpJHld+iDPnGWLwV00tqcnI1FY0zeIbv8FGUZeIPFSKOWifp6NMRjJBa9gPBDyNroVKXirnuW+rUV7U/M/VGzULFSq5IeuP44fLClYLOxY7uCtjjTjrxlw7+eeGWLA+o39x0fAdTVYH38INgUOQ25MG+MUBSRMzh9xsWwZ4+Ulb9sAuY3Nw+p7Im33RoZISrrktrUF3tvAnK+DMkNehQSGlRxMboOssC8h18kF1jAiV+j/IQ3oXGBt6++CYibfx9ykaoVo85A7Eo5/Zu9bQILRT3EJ3MoYO5Ss+MLO8/Bi86JIqYxK7K99P8kI21+AfV1/T+TqPA8RwklIyRUSzRV7b79uI1Ayo4KN8YT9179xyaHqDfMVWte+SPkyFylSSilQYqE3WCOWN5z3PBLAQnb59oYPp2NxjhljtliaUYX1LgaDSxlsz7W5FvV2IT+hloflkSReLNMMhq/DQgGi33kFe6xcORFZmqbb+VgOeQ28AdxQzvwarSFBelvDAV8YHrgfWdqubrQ5u3O2NIpjvc3wXH/3zErIqDLH5vsu5YcPDEfW4A62TpeSxKr2tHEnAXfUYK5VpReLeqskrER70RuFkXmheTBSAr6PvP4GZYxR9l6QKAX62lktICXy3piCBSfH5I1NIvi/bbgvlONcEuVheheX+K7ee9aaWnUGTZiZ/iimaQVcnC7SFL5tyOJQSsSRVmIspJn8sbvujawK99mBDNaR37lnhqPsyeeTZqKqmRk8zPdY5Ob2kR3ogiGP0e7Zg0lihfpkjBWQz+IoGdlYXFwIck35J9J8PWjI49/Afe6ZVqsVrfC+BmQ6GllRpiL3tAJi1A8iEYqGIe1lAVi/jLG7IGYh7HcfpJ3/GBGWNREDexpydfpxtfd2QRA5Eo5/jBQKr5KUkMkqudURMRxDIjzQDZVSetfmPIoaJSlC7sSeOI+3M/sPlHRrHpZN2u6vhjLB3237FJJh5Wnh/dheKBlkuGk6EhZXdfeqbnWp8BoGxVN3JAQvR1SnHGXMLSBmc7UiY/hwi47ROPuThBQdizw5DjeYucM9144GoRlV2ItAg5ZD+GpD0tb87sCetob/RIqJPLobC8B5SraVkMJ4huEAz1h+e/bts6et71+R2+cqSEHSE/EAMxFN+wwJI9c34j5krFHuGtse3GLweC/Oo60R5ozyIHivw5cRPXiepIxqcOVuh+KRxyNL3/kONjzNKKYk6oxC437r+tZBcf7DXV/FaSQVphetrVUaViKYGYaMGr7kYvCg64q84z6wvVilmfPwVRMmIqXsb4gSCkc4YACShXzFg2NRmNtkg5M9EX3ypYdbJazUFU4rPmDC4Gxnmz3XEP5DyO0mAPQ6dhimGUA1SV9f4bEOs99eaofkABSTFGLhhtjB+wbFljVB1lnXVdkUuWXPRda4VexAHWxI9ylc1lmUTKdJUovoQIayJl8jxvKP7t4IW5u/IKZ1dxLNWcNlUG3JDWkpZyIt60qIGVwLWdM+xblAIm3sGUXGCrC/CVK6PIaSgwXBtBtyGXreYOlTRLxr4iZIwqCfAFxQ7Bn73h1Z3F9EjPFO8Vzt+wSkAJqLiKWvU9wZCQLP2jnevgbzHIbLNo2Shbwa3h+5lZ2NXNOeIikRGAh2SM6VZYH9XpkwdSwi+G+RZLVtGKWYg+W1ULz0R4hZ+itNk7EcZXv/LDC4yFjdkDvck4hxOdg9MwYJxu8jnPoSLuGWG6NhaEYV9mI7O4uzkJLuCdJlz7ogb6pXbD+aWLDdWL2w+usZz8ReLCcjK9Mr5FhcEOM7BSkgP0Iu0bu6M7UEckn/EWmvoIbZh2j92geYK7F+a5CE2fyWBqHfyHOpgJRSXZDn0uPIkNIF4fS/otCfk6LfdiZSqoU1c9+70zTmtA+qrPAYEtZWRYqa/5DkZKi4pY8q0ovW0CoNK9H5mYjw3UzkBfJjdy/O5/Q2OYl9SXifAeQkFs3aP+ShWkA0cKOM8caSGMMexhJ42r1dkMD/DlJ6/gUYE5+FtlZDWK3KoAL0z1FSrP0QoXsZCb+/JLHQ9UMJTX5X7bGQFv5nNtYtyA0nrn+5FImr+WRySlJVbTOsXiFyibuLKFsgihP8HJiQ8/uAODzCGI+EkU2Qy/c4pNF/GkvSk3PQ2w5k5ff3YMQIxsln2huynEEOE5Uz3mqIEf0bYhKmIG3lLu6ZlVHysA1w3g212F+kuJmFMvd2j+4FhrcLCePQHmn17zUis3YEy7cg5mcscoWcg6y2vyZx7etg63IrJmhWeY79DK98ZPuwIYqJ85plb1F7hyIlmUgT0o/JqFkd7x2K792TxOoxpt6wvgDrF3BWP9vLm5Gr28UGA1Nw8XH27I+IMoFnjPsEwuF3Ic+QmUj5FGBtgJ2NLWx/AtPqGfQWTzOqtBfDDPbG2docj5jVtxHjH85uR+C7wAklxn0cJVUckPV/UV935MXymu3/OqQFpX0R7VoRuT6ugnDfHOBo8nOsNKwwgmjDM1isZ7E5Ia+PM2w9HiMKkWiJDSlAbkIC5VEGbxdEz6yIyrKFaildSowZYPlUpGD71H63EQmOHY2EqE9ILJ5b+t9XYa4VpRetrVUDVuw3pyD+6xCUx+Rf9vsj3DNewF4mC05Ieyg+irwFl8l5ZqmMeyFvx95R/2bIO+0qJIQ/a/BxpXumD+IF+ke/bVjc18itcgMlBLcdsrj+FdMAWn8/xDi9D+yWM0b7So8V9fVEAvp0A+AbEIPgiXdXZL2dihiK3tUGznh8Q7r/8PfdmpyDGLmSZV2QoH4eYlYCQVkKWWTetXVY3f1Hm0Bd3X09GjHqwbro93VjJBjtXWJMz/xfgZKTBU371iSJrk4p970qPGfvvr0XIoSZDB5yu73DzqR33V0DGBs9exQStIJL3G8RQ/SArek1OEsXiRBVC0+THki4nY68Q17BmCV37joj7fIk2591MsYJ+K83Ip4n0lTBFsbrhaweHh6+j5QPZ9cb9hdiDe+1FvbtHSSQPY+Y3oNyfvetdc/1bWLrvKZdD0DM1kzk9TQwXnO/ttH4LZJmVHEfutgaXWPXA5Hw8UdUD3oewjtLx2tGtjVoFGJUR5BDX1CiuTUc/LdDrpkPkXZvXhcl4vwdaZfK7qg2bCi/2bAKjoz12wYpxzcjskQ7XLC6n7PhhkORBashqgYgD5zf2B7Ow8qzRc8sYzhxLrLqdo3ub0I6o/ORBrsTDGY+QQqwsSQW4SH2v8cCm1pfVfkgKkQvWmurBKz4fbbxLseVpjJcc4Ot/SWuPyQayzVIofjsWUgQzvQ+QzTkWYPHgaTpeJOa6sgDcLyD21UQHg4u6l4501AeOotqq8wgacA43wjdda4vMEwdkVvDkxnA06SETnPHcteeEVoMuZ8+jiy4u2b9zp7ZOW/OFduA9HxDkqkJSIs2KGNuRyImcbkS4+6EtKNvYqn+3eHvhNzoJhnyWSSy2bakRlMGsz2yuBUwBsD6PIP0ITnukxnjb4cUJz+J+tdEFroCUW3ZGs69A7KaPIorKZFxLodj2dFLjNcHCekn2/WxyDIz3M5ziH19EOdOVYN5erzSBdgKKTe+Ie2y5Z/bkuKeOr2Ql8FEisTbk1j314hgrYmrdEtvyCPnfmA9u37M9rQzsKPN8yvg8jLG2gXF3v+etKWhB7Ji/RcJ7EVd7mnBNKPKe7E2EqKDEuvfWFk/FKryOhLaHqeERQ0p14JyPJMpRYxoAVnFe0Xr2cN9D9ndpwO/ztmnn5OUbeu9IPNuSY00TdgHeeHkrd8gm/PdpDPpd2MBvKDqONfYffsQRAenGB6M57sk8gA7OervgXidGQZzPZG346EkCuwVkDAzG9GQzNKW8X9Waa4VoRetqVUKVqJnRqM4+6dwSZHt3hCUFOwbwynlJFdtj/iQb2tP58EUCud7wj+TMYdOiJZdSRT+iZS4ZxvufApXrq2t1b8178dOm4aE3eDCXUBxbD67dRCKz0HC4RLVGisaN2jTl8OYAcT8b4gS08zLQ2yur+pWCEyJgOJwhhvS+D/StV0XQ67wL1Ii47Eh7ids/X5NIlh7BLUpil8bX29AXFQbIvQvIMtNB1Rmag6wbfTcLkgLuXEZY+5FUv/3p9bnFTCDUF3XApYZsxYw7P6/nxGj4NY7Ioe4tEMu3CVj3Gx9NkFuj2/bugaGcw+kSPqYjPwDVZqjP0chs3k7lDjuVqRVPzHn+Y7+Mxo3ZAD/gpw6pkjo3A9Z/WteTq0Ka9kbxdp2R4n93sIUjXb/JmTdmY1Zo3PGGYTcPwvAvWHdccndUNz0C7a+q+bAZUPQjCruxxhkjT8QZdpe3cHZjchC/ypFvKeQNSi4dr9GfsnIHrbnPvlbXmbcgNM+AbbKWn/ETP+6nHm25IZw6Jc2338SVUBxz/VEFq3MvBYtublz1gMpUNsjYWJH5LnyH+TRFVvsPa3ze78mcgGfjlxnn8TybpCEfbS3Mxy8HFat1vyid64KvWgtrdKw4vpGI94huJD3jO73Q4qYAjCujPcchMKzjiryTCoRq59fxrM7Ie+GAvDzMCcHEx2QAmk6soRn4om2VgeYXegfygWxAJwf9S+PBIgCcBZOe2oH4gJEsJepxljR7wMA9jfAuw/HnCG3yj/Y+L+hTowqsmy+j4SHTja3g5CAPdUOzwHIJW4usGep9zTEvR5iTAs47V2EuAf739QbIBeFRqLI6ImY0J86JLoBsgyFrMeHIabxPcrUUCOm/3jk2j+JxMPB7+tSyJVo6xrMt0kGYEQAf26E4c84V9xia5bRHyeH2tDWaqTr+4md401qtL9+vmch9+UdXd9KqA5wINghHryku6HN7yrEbP2BDHc2e64jDZx40MFJ+Axn5gKkhe9u192REuFk4Lvx+kdjdkIun48Y3I1193z26V2B4/PW1T5bNM2o0l7E8YNnA5PddRfkPrk7SVK3vHPbHtG1a22tziny/7nltTLu/9DOxnOIsc7MsF/s3Rqh2fne2WBwPgrtyowdJe2h0RBzdvvWGXk2XE6iyAq5N55DCuc98vBgPG/k/XAnSUWALXPWaZzB5bk1mGvV6EVraJWGlXhfkIX6MaTMOpympRy7A9tl/L5JtnAb6zMscSxN+ZcRyDCQCgMo8o7dEE17zt5vW3fP07RdsFw+pcZsazWC24X+oYDkRiQU/tkDNMrWeY0hi6uQFfU7RiBmAadVayz3O4/QXkE1OTfNeG5ZQ3ifI0txr4VdkwVcv3CwRwA7YFZrd38xpLW6B1lK5iF3xlyNWM7cVyHJIHolibUvPvRtB7Ky+9ofWYBuJ7K22Z6fY8TgS6R1LakVjcYI5aumI1e3tcL/k8Ms12DOOyBF0Cp23cOuP0NJeVIuzDljxXDZJboeabB8nF0PMJxxrXum2jFzYb7HIwbuOCJ3bNv/nyEvhYeIkjlljeeuh5LElP2BpsngGvaskiMMkQi1FyEX8L52vabh7x3cs5leEO775igWv0DaGtREOCEtuLVomlGFvQgCyZKIgYyZypBcZw9Ej7ZFiq0mYUR5MIm8sK6wccZTJuOb9672fSRSPE9DSujFF2bMltRyYLoLiv18ElnX9iGdm6Ih8UB0zh5BHnZ7Z+CElUk8vc4gSuiWMa4/yxcZzL2I8wYjLWAfQE4CvGrMlwrRi9bUKgkrFOELkCHkdsRnn0u+p02gX9sgQ1e8j/2Qwe+BjN90QHl3HqeEsSFj7lsjBUCBtNdUzB81JE5YFFvzfizG9kIjuI+StoK2B84kcV99GWVuPSMHeCo2VvSOZ9rzQ6Px/H/3RNqkqdQwTgklXZiOmMnH3LvFri0jEJOZstxnjJdXQ/Q7hri/QJaYFh+L1cgNCbg3G6zOIrEsx4iwl+1rz4X8n66I4X0BuUSP8e9Qo7n60j0zUYmR1dz9TkjofgcJLJuRU4OZNAN9EnKX+wuwr+vviSz+82y8SchdeMUaz3stpBw5grTbsT93SyCtc4ogxs8i7XRPXB4FRKRPRALcw1iytkZuJAL00ijB0G3IDb6/e2YlxBDPQYzUu1hoQxnj+7UfQWI1vdrDYxnjnEkLpRlV2IvVjSa8hpg37549BDGc3xge+wiX/8Q954WagXY21nJ9/RHD+xWi20VDmkq8d8A3fVE8+Gzk+ZPpvdYILcJ7Xe18dHd9g5HFbh7KatyQypyMef8M0YW1yFEI2z5fZed44AKOH7L4P4GLpyVynaUGCaCoAL1oza25sBKt86bIY/BXyJXch5wGpcwfgTWKvM9QzCsQZSf3Z3hv5Dl1PWll2BaI3z9zAebt33skMkIWSNdnb0tg1gLbwv8wAfDFERP4JmJ2R0fP7WYE8A2UnCGVEbzSY2WMORFl386LWQrxnt9q7msFrIipPoukbu927l6uq6ebW9mCBGIGQ3mft4isYW2t4nu7M2IkC8AVrv/b2K8yxmiSiTfjmeAWdR+ygp9awzkGOFzBzuXZZDC59o4jUQmdAlHSMXt/T5zGIy+Wf5CUl7rG3V8Naf/vRlaxkHypFkxSmPMBKDlWXIc5dnPuSkacMImQsD5SxEwzvHcTacv/3oipmAxsWG+4Xsg1W5u0xehllEfjeZISOD8I64fCY8Yj6/P5uPgyv8YevuL1t++D7PdfIvfaUq6lLZ5mNHMfOpGu3drb1v8+pOQIiQF/554ZjspfnU9a8M7K0n4pomXzUOz7bSTVKHramZmB6M9ClxcizTuE3A4r1nt9F3IuHl6PRoq0tw0XHE6imO2LXJnnooSVRROatuRGEi97i521TC86EvfoHg4nxmffe4oNQ7TIx7Tuauv5CvKiquk5pUL0orW2SsGKe+5YRE8/QjxGwfDUaPfswdb/NC4pWc779TJcdz9Jadu+qJzXHIO9CcgbYQpwe/xOfv9z7vnvKyKvtq+BZ+q9P22tCGw068dpIrcvigt4F1cOwe6Nsv43UGxAlstGxcay59pZewL4m+vzzMAAVNJn9ZovfFrbv48h/89IW+mKEgKHME4gJ6FJdDC7IaGvoTPatuQW7et3keW1gKxoHeJnSo1jxCXTMhbt7VCk1XyLMqxzFZ7zOShOdvm897PrlYFfRn3DbH2uR8q1oShOfSe7PxAJ7V/Zf4RSFCHWtVPWf9VgzqciRdUSdh17mxxMU+Vg++hzoI1xC3JFOwVZQqeTlJEKWWWnAX+uN3wvxDr9zPZ3X5vLT5EQt4Jdb4iscvOBk6LfdnZr1eTMUCLjtz3TGylivqJETD4tnGZUYC8Os724FDHxByDhNMRPD0FC9DdIyEspNdw4WXtxLnJ5PRTRmD1tTx8lcfHvhkKdpgCbZ62/W+P+OXPIcp8emXevpTc358MR/b8YKdzH2179iUSB0wM4xvq3qtc7V3Duz2HCBhHNsnN7BDnuufZMwA3bIeXjXFubh1Am6UBD10GKvGlIMFmo0IRmznWh6UVbaz6s2HNbGowcicIlB9i6z0O8xTD37BhykqPG8GNnciYqORjKaPVCruM3IYXjPX480nxiB/ebzGoxpHm9pZG3zoP13pe2VgTeFujh0vGS26I09LNwLtt2bwiyRBVQnFRNxkKW8Plkx4ltiDTEP6j6Qpee79Y230+As1x/KQF7hK3Rz2kamxmYoy5kJN4p9U5trfn7jKxnF5Nkfx9g/WURTsTkPkOkRMraO5TorPeCjF+hOV+LXNPzrOsjiVxBHWz2QwLYR8hauCGKofZ1G3sipn0m0jovtNWrgnPe0vDK0a7P16m+BzHIuecXudz+lbRb9FO2lt5VrSOyIC6wx0q9G3LzvgkJt0ehuMMLomdWRHWoQ0KfzMRNfu4kXjijMp6JrQCLASuXu3YthWZUYS+Ws7P2OVLonAT8JnpmKRJ32smUUTMauZb/1/Yk5PRYH1lXQvm8oATrFOCdNC0KDOZoJJCfRuQF455ZinRC1IY7F9G8VjG8dhSJBW45MhJuGSzXHf81c75hv/6EXKVDzfRveRRkRPk3GYmkorFCVZmrUQWFvZBnzDRUWSUIs/0QHa1LWSsqQC9aY6sErNgznVFeo7twJf7s/vcMhibk/D7sU2w8uAoZDLoipeFkZPjbNHqud3Sdxyc9juLKB2T9f9TXza1BwyU0bQ1t4X4kRHECKrK+DWk3nJHW/wGyRnkNzZIojnK9So+VBYDWPwy5d7wNHGF9XRHxfx64q6YLXt58ZyGLZ94hDExGR0RMbiJKRuOe7YSy7e5PHTS2i3pzezECMa53IcvDEIf8eiLN9RTkodBEIHDjeYZzGyMam8UIlITorE4ZDHCF5nopTqggYZgnIE+SJtmrkXX2RorUUkcWmT2RxfZrW6NUNk1EHHdBMdYFXOm+Ks85L5dDPztX/wOOdf29DQ4+pnjZqG42lzNIiPd4w3XBlfYgzOXd/a7hrBnITS4kaJsHXJrxzDJIqJ2LFAxNcBVpZvRZez5mlAK89EJeELmJG/PWkhZGMyq8Fz2REDLdztENiI7Ecb+7Iy+K+bbexRTY69g52Niu+9j17+16CRTDmOm67fZ1MKJ9V5HjlWDv+ixiZAfWez0rtCffR8Lg91zf6yhBU1eb81HxHjQKLihyztZFwsh/cCWxkELuPiwPTRYM2Gc7kjrqvdz9fijU5n1gtwV5pwrMtSr0orW0SsJKxnn5G/APf9/B0jmGezLLCyIh+krgQru+HnkJDgowCWyEDChfAfvHc8p4H0+LRiHL94gia7AzSgrbEOe+tbfyH0wAZB+kpZ6EtG3fIMZpDffsYOQ6Nx9z76zWWPaMd6HdAVlHjsbiklD5o8cQ4/aitdeRJrOdH6Mqi7xw8y1QxDqCMg8+hrT8N7v++AAPRwztQ/UGtkWtObgbaoj2n0iD/qUh8jFYJluDzf2R9fWyImN6oXkfpEDq6u+5ZwcZnNxNlZPcoJi2O7C6odG9jZFQfG3U3wn4EYrrzMq67Bn64P58v52LzDrCSDlVc+sDsq6uD/wA8yJAyej+jtw5n0BJuv6OGKWx8bv7+SCB8xOM0UJuofOAbey6H4oxO40GrV0Z7Vt35Kr5IVIybZABz0siN72Ti4zZy347EVi/yHP32tnITEpDC6cZVd6LxYDtkaXkM2BXd88zfNtTIoQIMaihlOZq1vci8rgJVsNR9l/bFHs/5Ll1Ly7OMYYR67sNeKLea1rBvdkNMeR97PphpFRdxq3fJDJwaEtvpA0iw5ABIcyrk+GENxH9uBrRuyeRsqFvxhj++/nIa+o61/etNQ/R4yfDdTGYqsK8K0YvWkurJKxEY42wzwko58OgDJg4EvFmmbkMkCv2RKT8fgHFWYdyYB5nDkeC99fAeWXOey8SJVEerzcY4di3WUQSGi7qrbyHEmZiHRSkf6Jd72Ab/g0SJjZyyK03GXFBlRzL7nmm4W4kvM5ESXLeR/GanQ04D0aa+on2PbjQVs2tohnz3bLEuP1IkrlMQRqvzBI1KAFUL/8+ba1i+7oYsqLcgKwMyyAXo08MDvcn7eKca7WO9vZL29t/kp9YqSeydmbG21dhzgGGtgNudf1LIRe8OYiJ3hBl5DwWuXs3qSUandvlw5oignorUqadmPN8R/9ZxfkGQr0T0ph/ijLuz0CeCD2B5e37/yHCdy1OWCkx/kSU2Oswm++h7t5uWOLGesN6M9euB7JkhsQ0OyIX2P8gpUzskeEZnizFxAQ7F1+QE0ON8P1+SBnU5OzQwmlGFfdiORKGsIOd0/uQUidTmeX6chUKKFfC8ygB3cMoZ0IIf+mFPFceoUguCKQofJkipSZJe3l1y3ufltxy1nZZZDW7BCnWZ5LkXFjc+p9kIatK1HGuXugYjyyPXyAFm69Bv6Xdfw0JLuNIwjgCrl/HPd+RxB28gHCoD6MJfNQ5tpZL1Gi+VaUXi3KrJKxE416LlblFgm8Iz1vCPbMY4l9epEQlAxKl7YvA2h4m3feBwAX23JgS4/W1uRTsM68EWA/kybZfsfHaWstp5T8oALwduMquV0Va/ZMMmcxBcdBjYmRGdgxcRcZy/Rcjpm0TEmb9v8jFtEmx91LjVXSRKzDfCPkERqcPirX+GsWpDizxHq1WK1rFvT0deQ8Epvs1FM+4EtJO/w9Zxcp23UbMw87I/XE+qmedGYdKOgtztVzd4jJEpxox+AeJdaq/9b+BFEbzEZM9Ln6/aKyzEOO9o+tbCbiMJAZ3sWrOL2fO4YwNR0T+ApQHYoDtbwFzOXTzyoyLd58jcNZW5Ho7zdbqlrCfSDExE/h1veF7IdcuMJmdkTb+chKhLmS3fw5ZAfZgAcJVkEB4FUkN8Mzf2hlqEqYQPdNiaUYF9yIIJ/0Nn9yHcz9FXjd/MHj+CEM3AgAAIABJREFUDRm5OXLGPZE0c3kcUk58hQnqyPvmQkTfQqKfDllri8JoPgOO9DDk7o9A1p1+5bxfS284jwDk1XG+rd98ksz5y6AQsi8xl/FGgj03vyMRHTyCxEJXwNEGe24x0oqvgFeDZ8T50fPLI9f5AqIjqVKlCGe/QQ3KtFFBetGaW3NhJXpmFFKWboIUu+2RgPohCnk5FCV1PBfx43vm7QNJCNzDyKg1BdEwX/7UC9hdyEjamDFue3vPUDbynCLPdig237bWslr5D0rr9gtkeeiENHPXGWHoTEKg5+BqClZ7LBtvBaQ5P5SEOdgYEaqD7bofictFTQGzEvN1yHsHm6cv13MoYkweRjEZrR5J12hf2yOXnhPs+hqkTQ2upXuS1Ga/knwBOQuZd0FxRk8i6+8+pGsm1nyPDV6HIgv9gUi7/C6JlaUz0tpuZ7C+svutJwwBlo9HhPQ4XF17u9cfxaLNQdlfB1Rzbjnz7YBiRe8g7X3wqLVQeiPglSyrXmCklkDE+BWcaz2yhs5AQt1E5Br7MnBTPEYjNNKKk0eQgmnvGF5RDNudtr9nkFP1IR7TroeSxHD/gaaJHMtJWtaiaUYV9uIVpODNCs9YFgknnyOrThO3w2is0ciV9QlgC9d/NIkXwAsoVOY/mPfF/7N33mFWVVcffjcdQQRUUIoo2BCsoAYVa+xiLJ+xl6gRRcEaSzTBGo1d7DVWNGKNGmtiVNQ0NdgiQQXsoogIKqDM/v74rcPZ98y9MwMzc8vcvZ5nPXNPnd3OXn0t+552RIxs9nvvgQShJ4NzrYPnjkNeMf1KPa5NMC+bIIX4faTVD3rbXvMRCj+40sZ3GnB4pa3BYO46IvfnE0n3/X623haQiZfO10+kWLkHCUp3ESjU0L56K6KxN6LQol5IIT0TOKOYfaaR9KIasYnXSqjIHoFZrYPrHZFB6zF75wJEBwp6zGTen2TuPwh5TEwBRgXXuyLFyprZ/oXty/PewciQUIMs8zE/UoXj4t2shZ4QyHeorbl+GMUb5E0K0Izv6ouI/WF2vCpiFC60466IcS1Z+YrG9DfYfHZGjMudBPUSkcA+Allh3kYJsIpajqlaESk3uiH3nknAL4NrGyBm9WwKxM+Tq33tgGJ7OgXnVjGisgCVaipavA3S+O4fHL8SfFNtUTxZUqO9oBIsH0EB1keWy2PCtUouE9/FiFgNgdtqkef3r8D44PhuxPD2t+NjCZKXZJ4NEw8OQ0T8a8RAhwR5C6Q9fxwJiwdm31FpiBQjH9g859RyDe5ZHjHENeQRmoLnlrI9vG9wrQeynn6LlIp9FrN9ZU8zmnAuzkQKm4HBuaw3SldkKZpGntJ/pEqiLVF+gBk2b+9hwp9d3wxZg8ahBISDM3M5ENghGPNw/zsACZ13kqtI/Kl9M2eWeiwbMQfJXtAfhYEkrqDPk8aqd0Al665HgvW54fqjwpTmSGAYjxSG+2WudQeOQFbEf4Vrs8C7eiMviE+RoLpKcK2VrfFEkf0WSiw6tthjRyPoRTVjE6+VFWy/mIclOrM1kg1BWg/Fwed4POT7bcftM8fbIe+9D5DnyepIQJ5BJtGm3R8K2f0QbVw/OLciUjTPs/Vbp4t6xPLGJXsI9kIM9RZ23A7FLIQuoA1iCpfkXXmOV7cPb4wdT0NJTxJ30qEodnX/hrSpWQd8MftLypT0R4z52eRxc7LNY0OU5bYG2KzUfa0mRNryLwjii4E9kXDdvcAzIXN7HBISpqOkHUeTJt5YHsUWzUclKfIm3Wji/rRHGtgvEJP3ONLSrpJZn1ugJGQ1mAU/27d8fUYM+PsESqLM9YSZ70CRsqhm24ws9S9g8eVG+OYCm9hxNxQDfA21a1+Gc/s8ike9B3mwzEDW6vNJraYFrd6VhKSx1fch4TRvtu5gb16a1AsnX2mmn6DMvx/ad3EvuV47ByDm5p1kXgq0q2JpRmPXs83DUxTO3ZDkUuhEapmp5YKIBOtvkZAzCmWwnob2rN8W+uYL/U8Uk/oEqUVveRRmM8feeRvyWpmK1bjNrpNKwGAtD0SePvei2OpHkFfSu2SSvRX6bsodyRVO9kSCQg1SHmb3yKVQKNxXBElZ61jHnZFC7V1kRMjWhd4b8UhTkOCTrNtmyY+QnRMaQS+qEZtjrQT3nkWq+N85uFZwLZBLf8K2HYH4rteQIWxAcG0QyhHzja29GcAGdi2vGzequjIJGUy+QXQnqRDSFfFGHyPFZUWX3atmXLKHxGB/hjRNB9vi+5olCLZf0nchpjusGXgF0npPsQ0uiQXthZIQPFvqwW5kf89FJWqytfbyuVv+vtT9rDYkLXn2NoqlPxJZJ26u45mEaTjaiPDlRhTGGZH5AymzuzRwvJ0vijUNMTN3IGZ3HrBP2N/g99rADda2Oxv47tOR5TD5TrNa5ZHUZp6KImwSJBFDbsPJXPwI/F/SXuR6+CmwdR3vGosI7jqk8a9rIbe0Ochym2REdWQsipWKyMvhwexasePuyGOhUPKWhCnuZ2vkPsRw/RoJcx+RhiIkWeY/BO6qp00VSTMaMQfO8EXgqXCNBff0tnFYu553dUbM5XhyPWvWQ5bEeciVOck9kS8GMsswH49o4b9JY7K7Ie+uexED+hhwYvBMpXpytLFv4n6CspmIB5iEXJgPDs5XXIUAcoWTfvZ3c+SxMxN5OrXLPNMGKbI6Zt+RvS8zZq8gRcWBmfuG2/kpyHOiYLhJE/a7yehFtWBzrRVyBdoDET82N/NtNXgPQcrw91FIze9tbu+1dZYoxrshJe9hWOK9Qv8D0bEvbZ3sjkIHF9oemmQ8Xwq5rk+lAXHbEcsTl/xBJeX6AmmGZmD134r1LlSiZi6pxmcIsq59ixjW7shKnNTJThKhlJw4L2F/b0dxbIU+2g3JuJFQQRltWwIiAfsBW4NfAn9qwDNrIMvbmIBo9LVN/LzMvR0pkiaT1Nqyq7VlLhI4tgruCQXs3kg5cFwD37+dEZVQ2AnrGD+GlAxF/V5tH/mONLnJSki4m4vVyESZjQ9CSrGCMX1I8LudQEgjFbC7I8toDbJgtQgNNanC6A+2t/W040WJshBjMonAolDgXQ+ikIgVg3Mv2z4YZgdug9wKcxLI5XlfxdKMRs7Jyfat1ao1j+J/36WOso92Xx/EZJ5hx62CPaIvYgRnISt5Np46qxC+ESmBOyAm8h0kEG2Vua975rhi58HGbzpwfLJmg2t72T6wADin1G1tgr5egoTbre14DdISVMeQx2U2WVPB7/oS6u2EXIhnErh/27UByF23BiWwajZlJU1IL6oRG7tWGrBOdrB1Mhs4Kzhf716ChN/ZmFHB2rPQ5vFNlFemwZnokRHifVsLSdz9T5CC91Q7ThKntcXoXnOu34jNh417WG4ww4ENg3NLZF1a3HchF6uvgFuCc8NQzMM3SHj9HDGu69v1shE2C/UXuYzsEZxLPrbbbBOqlQEXWXjuIQ/zFLEkczsYWSfrtT6gTKgfYtlg7dz/ULKnDkhwGJPdYJf0O1uCvvRDjEwSZ/lPYN+wHShrfT9ytdH1CTo9kKX/S6zWs53vjuJ1Z1Ekd/BMu3oij4Ong3NDUF3NuUgR8jmy4l2ap7/ZOK2bkItXGEOaCJoHIOZwCorNW745+tTM41WoesNGSGB6k9zkLqshgfb5et67lI3xWFJGapytwUQ4/iWwUUPaY9cqmmY0Yi4GoXCh6cAxdq4DYuxeBR6p712I2XsFKb0SmuRIlUUTUHWAD5DbeCu7vjqKIb7Y7rsTuTuunIwvitP+G7J+H5ptQ6E9pJwx22YUTzkbuCQc0+D3/UjhNA04tRL7HPTlWGQxfIO0fnN3pGhciMKNejbwXduhjOnjkUdDWI5tQ1JF2J3kWi2XsTHduJn72ih6Ue3YVGulgetkJqKzDRGsuyM68Hs73hHL+G7z+zFSCuaUWq3nnUMRv7OFHS9nxzfYcReUEG/VUs9LxMZj076sCRn+ut4VEN1RiFkKmf3OSEu8C0qY0q2p29YsEyFmZEUkcOyS5/oWSMN1e+Z8W+DnyN2oVibYiOWNtlnPA5az478gJitxFR5uhLkoc1vPd7cTYoz/hwT+9kgIehK4sp73roqY+T0wVz2UTCTRUr+IYtH+igTrg+prT3P1HcWWziWwMhixXR+5sh5Cbjmj1pm/GwJnB/P7PVIYdA+eccj1/ToURrCQCnMXJJeZHWT9TtZtWxQa8a7tWzchZvclpExaPvuOcB5Q/O1sTPGClDsLsNhUpJz5I7JGN0SJ1eJoRr65sLEbgTLxH0dauWAYiv2fj2q0vm7f8T+pu4b1rcEcXGDfxeEE7rZIUL/O1volyBMgKW3W0+ZpBvI4+Ia0LFuojBuMBKQfgN+VejybcF6OxxToyGL/AbllB1shRfsDtkc8buPUqRTtbWRfw/ncASl0pgKnB3090/a6Jyjgsh18qwciS99rSKnzI6oSsE5w7yoorGEhsFuR+9soelHN2BRrZQnXSQ31eOnY/V2AE1BZzOWQYvICUsX4laQJ9A5sSH9JS8olCQxfR/xeEo40HIUn7Vjf+yKWP5a8AQ1qZGGN/ABkGamleQ/uqSgNISljtzOWGMOOl0UxH3PsA9zEPvwTELN4XinaG3Gx5jZfHGIfpFG90jb/z0jjSTvb+ZewpD/N3L5QWFoPuUVtnblnMGLIP0aC8NOIcV6q0PuQ6+ebiLH+zp49HSXvWMl+/80I2O3AXiWcj1akWuvXCFySG/IO5Lo/xYh8okQYjxQoY0kFizWRUuIaO/4aOK3Ua3Qxxi5kjsYhK/V3yPp7UHBtO7s+GQkNl2El2qjHKoyEsjeQQLwQOCq4treN83YFnq0mmhG60z5KWg7ra2Q1OxslzlwFKXTG29iOJI2RrjUXSBE2k9zs9U8hQeJMlEegG/ALRJeGIyF+AUEWXHvuz4ixfJ3cShlZD6wL7L5dGzsupUZgG+vLKDveGAkQr5ObYX0LpHBaDykwa6gnBr5ckNqJ19oHv9e1ffAzRNsSIWI0lpU/z/sSRc9QW1Mn2/EIG5cfbQ1uRirodKcIeUjy7Sk0kl5UEzblWmnEOslLLwq0t7f93RXRr1BYPwl55IwhT3K6bF/tXGfkKfQkEqr/G/yPbsj79FlipZ8WgSVvQMGGZdxKUdzKIGrHFe9sH1ODYj3LFaldGuV069fEYKNZ0c5Psc1joX2gYabxqt28KwWB3YPfnVDW6K9tPvew8ysgN6fvMZfx5pxbchn0a21dfWzr7FoUC5UIy91QyaiXUDbfJBlRyCgn3+1gJHRdgBjH3igmrQbYO/zfZLTTFEnIQQmsssmW1rF2j8s3RnWM3TZI4bA6uW6fV1ufP0RucFNRYq5lUczqBwTCY6UgYn6+RPFoo1GMdE24J9l9HTPjVCu+EgkXPwnO72PjtRC4z861R0rFz4CL8q0XqoRm5JmLy20dbUlqOX7f1tsadTwXzkVYc/ZXKExp0dwhj4TrkYV5LmJs5yRzgRJ0fowxoqQu5H9BzPNU5F6+a/A/w32jPRWcxIfUTb4DSkI6jiAuE8VdvoWUUIm1bTZwj10/ysao4HyVGyLL3pXkxosm62VFJOQkiaAGZp7NJ7B2RPkWbrTjNZHHxSlIUTsH8UW7kol5zfe+Zuhvo+hFNWNTrpWmWif1zRNKjDePtIRgB+Rldx0prcmrKEb5LrJld7+29x1t59ZGoTRzSHmpqvduqHQseQPyNkqC5mhSK24b5NJWg6y2pyP3vaSExwSUHKhWSZdKQyRsDbQP+HBkEfqE1JrZDmn4d0YJFVYPno0fZJkj8jj4AQmYS9u53ijW5iPkqnklcpOehlk4ikWogQuR4LI7EvCT0m5/R3G0oba5C+a+mG/tobrut6BQhx7B+ecMk++3Q6F3NEP/tiS3XvIY698E+90F6GzXzkbC3U8bMgc2djeTG38XCti72D0TUOmifnb+cmRhrNPqUS5IrgB2jTEMCZPRD2W8X4AE7W6ZZ7MMUsJYdUFC19sEYTHIuvoxEkT+iBQXbwH3hu+gimmGtb8/soocRWqp2QIpJkbacY/gW6sr/KMPiid8lcIlIn+K4g1PIY0hXA15n9yY551J1YODkDfCFHJrvXdFCrg18/2/SkJU2eEGpEw4K1lfpAJFZ1LL1z2kTPZyyLvj/lK0uxH9HWnf2ZOkYQitgnXYB9G2b5Drf31JqLqiLM0/Q8LXm6hqRSfE/9xs/28OsGcz921LmoleVCM25VppinVCrkFra5vTEQTKLWQd/xDxLL9DoS/zgG3red/mKMTtxWRN2PnjSL2L/oNC7d7ErOqVuu9FzKyFUjcgb6O0SU0i0A6i2K0jbFP7CjFcTyCGdX8srrHUbV+Cvm5JUEsVMZgX2u+2yDUvqddXkJDUR7AilnSOE2GkP3JxnWyb/vOk8TcdUImR620zPpfA1a0Y84tigqaSWs9PRVaVo1Hs7GRkfW5wPCByHR8fHN+NlAb97fhYgiRGzdy/DW3chwXnBqIST89ZX6chhcD2yBrxGea6Xc+7l0GCXw1yk9+QOmJZ7fw6wEWIkWiwu1o5IPJIGG993i9zrbvt1V8A/yJjgQjuS76LNsid+FWk1f+IXMFrC1TC5HHEMB2Y5x1VQzMKjGVfxMgdZserotjnhJZ0RcqJet1nkfv4X2wt/5MgwQ6FLTSj7d7nqK1AaZ853g5ZlD5AXjurIyvQDApkB64kRErIJB7zMXIt14XGb0fkWvxOdvzKHZFwdCipkDAsc30Ycnc9AejTwHf2Q8rZHW1Msta/h5Hw1b8Z+9Vs9KJasanXSmPXCSmNPhx54syyOX+JXDqzLeJ/ZtrfkeHzYf/s75akSWBrrL9hKMhmqHTXOGQZH5zvfRErF0vegIINSzXdxwDDg/PLIPeRc0ldSF62v98hJqIiFihygbsAMaHnIuZxCmJuQivBFogprAF+FZyviH5WM5Iy/wORB8K9KKboEcTwv0smgQW1Y5OKIVi3RrGT19nvnZFCZy/S5FQ11oeTyBMXlIfQdEJWw/vteKwRsE3suBvKgn8NeeKWmqmfm9nfNcjN0L4iEhivQkJejX2P0+z35g14dy/SJCzjybV45KvJuRMSMvZvTJ+KuJZDV7o9kfa+Bgm+WTfJpZA73lfAhLrWClIyPYmseOcghuRjJHglFo28ic8yxy2eZgR9yu4RqyM6MsaOp6EkWUl5v6HIUt+gtYaE9UtsjK7GEtXVsZb3R0JGqNw4ApVkew0xugOCa4NQgrtvkEJlBrBBobmuNLT95QUbv9HkJjPMrtsOKAHTE1gMZiUi8mZ4DSnHDrRzHZEw9RSWlyPb/3reuReiQ1vYcTuUeyavR0Uz9KnZ6EU1Y1OvlcasE5T35T3E1/RDoV3vIuvyr0lzpyyHvHP6BM+GdCwUrL9Frt6jkDV8GvLq+W2+/TNiy8OSN6BWg1LXqVZGgGsQUzAi+3Eg5n1tRPz/Coyw8xWzeJGL2B3IdWUeVlMvHAv7vTZyNasB7ix1uyMu1hy3QR4J92MCgJ0/GFnbZgIHB+frzX7cTO0cjNzWWyNG7ypS5rwrSsTzGvBwPe/ZLvh9lK3ZP6D47f8LxuQQlOW+2bNkk6usam//dzoSEttl7u1j43AvckEumPwqz//phpjp7xCztaj+coH7G1TGo9SYYSL62d/NEWM5E3nYZMexDRLqOmbfEdwzFglW65DGq66FLH5zUIblJAO5I5Obws5XFc0I+tKB3DrxV6CQkylIsEtydfRCCcWerWte81xbFrl9z0exjavX056QXp2DmNMHEZObxFEOJ1WYdEPl6A4Dhtq5ihesgzFYHlnO5iOlW33KtmWL1bZm7PNApCibh/JxPIwEjROW8H2rIGvweEQvj0DKmF80cz+KQi+qGZtyrSzuOiFXUdwdeTf1ybzvORSudTlWOjDzjnzfcGfEI40n8PBD+USes75eSZpIsqK8VCI2HEvegFoNSi19CWMwzBb4NMSMdw7uzflA7G/FMElBX3c15mMuYoK2Cu4JGZbe9qG3qEQ8LR2N+E4HjrfjMIHPXjb3C4Bzitimupjqdshl64rg3FCU0XtwcC5fMpohSLDcz45XQvHlc4GJdm5lFHv5NUH5kmbub5s87ZyOLPEjKVwWZnVS17FE+AvdmbuiDNRhXc2lEBP2CVJIDM/+/0pFZM2cgilEkEUnKad2DAXceguslfYoO/yz2XlCDM8/7Nt4BMuyXuDdVUMzMv0+w8Y9qfs9BCl0vkVKie62vyS1gHuE4xX2HVmqN0dCQpjHoxOwH0pa9w8b2zoFYJSvYTamKLZ1sdC+9zdRjGSXphiDUiOptaqT7XV9ydS0Jy0BdAMWBpTvHS0FkdLkVyie9FGM7oXrbTHftybyyphv6/jiIvShyehFxOKslYauk8z+NwglP3uFtBxqovzrEOydj9KAJIOI13sf42uQwjehT31R2N0sFKKzSqnHP2LzYckb4P0iwrQpFpuF3CqmYtoilIjldSzlPkFR+ZawgSFXlJ1IYzT+SW4d1lbIJaUfuRakimQKWzpm5wW5kM0GLgnOhUqT+5EFexqKc27WeSVXwOhE4LJo55YD/o3cdYcY0boAMcZJwqi8bURxrpPJTeo1BNU5noviLD9H2t1LC41ZE/a1P7lJ2B7AXPZQMpq/o9jb3xC4ZJJxeydlmELCfBtpPoQngSODa+2QoPIvZLGvVbu+EhHFyL+NklIltci7IyF5oa2TnnU8n3WJvQm5gOcIJPb3AGQVn4Ji9UMFRlXTDOvHQOR2f0twbhgK7fgGMZqfI+XE+tm+kzJ9O9tYzSRNwBPGB7ZHoUlv21ovaMEmLUv0ezveEasMYPvAxyg28lAqxGujjr4m47c+aZjPGyiR31qZe8fY2P2LKmGqkadJ+F0vsRLBvvfhWM3wxr6vjv/TpPQiYnHXSn3rhFz++Y+2532GDAKHBNfCNXAZUqwsn+f/ZelZWySoP0bqUeVIlcYTUCWWD5DbeC1PrIgtA0veAO89yCqWZPpbzRb8LUbUQ2bgYSPUlxEkWak0rGvjQEL2eyhr9Bgbg6UQ835lqdsecbHm+fhkk0eWpA+An4XrwIjBA8BpyOr0HxYjYdgStiuxtpyOEnd8gxjyzUiFyM2N4MxGLnFfk2azrLMkFRJ05hJYpRHTvb6NySFY9nu71iyuoEjQvx14wI6vsT4Nyuwr421fuZYCybfsvpAw32d71kH2zc5B8WMXkbpBt0YuzvdjJTYqFTN93wExmVOB04O1fCYSsJ8gj2WHVBjZEDjbfu+NJRYjNy7VIQvRdcg1eSFB+ABVRjPyjGXyrY1CQl2ojO2MLCi7oMRm3cJnMs8PQgL1ZcgL4yQb6+koSVzCFLYBNiDI+VGgXV1QMqKtkJJuOlK4JAqTK0mTfR3Y2HEog/HvYd/9BBSrebn1bSoWchA883MqLBN4uSLNI1g3Kb2IWHostE5QLPXHKJfMb5AxoQY4JrgnFLBXqOd9t2Lu7LbfzUVJ0pYJ7ulg9Gxv5AH2LVYyMWLLw5I3wHsPiu06BjH5s5GbYZgYJbTyXWwfwROYO1wlIbmWr/WQC93WmXsGI+v1xzYWTyOL9lKlbn/EBs/zNrZOR9nxxsZwvU6uVWgLVOZhPSSk1TTHukYCb5j9crR9a0lCsdnIonQQaYmwASjRzglYqAJ5GPTM/2lFar16jXrKSzUHkxS8uxOKvZptYz+P3ERX4b4y1sb+USwpSh3vHWPv28iOr0AKiCeNYN6KxW8hIbFgUq5yxmx7yWU21kXa+c+s/4lL9mgsQ3WBd3ZElujHSBPFjLe5GYu5fyNviSexzLtIuXNa8J6qoRnWh0JM3QCkZCgY50lhL5P2KNbxVjvuZ2N5i43nAsQM9sy+p661jFn0ULjTZKzmtZ1LSlCNoUhJDJt5Xv5smHzjH6Bsx0nm+18WeK6i9oJqwOaiFxHLA0kVYssjxUhYkWIjUqXJlcH5pHRhXWF0eyDlbshfPYUE7DNRPpFuKGHsHGRZH2b76/qlHpeIzbTeSt2ARQ0RU/6KLe4PqJ2iP9QcjkbWjorKrEmuYHItcg/5GLmNXoviFxPrTjdUduYlY4CS4vItwqWxJSKplacDykw8jtwyLGujhCdfoLic14yQ32PXjzKiXm9sz2K2a2mU/fJj+3a6Infco4I290fC9ddImF6hwLvyJfHolWWUjaB8B4wLzpUkthDVXa5BFr5t8s2Z/d7X7ju6jncthxLXnGrHJxjBHIwExyRD8NNYptlKRuvvlaQubm1JmZQVSbNv30vGihPcF+5729jYrE4us3q1vedD5H48FVkEl0Wxah8AR2XfTwumGaReJMnflZEVrUfmvp1tDBYrFwdy0/4rqZJoEqnVrifynvrS1nStmPf6vmdUYmYesIMdd0CKvOtoAXGpKCzhCWBjO37exqodiiuvsf5fW+q2RlyseW0yehGxvBB55P0NVYvYM3NtAFLE/oiUvwX3JlI+vSOKGb/Kfic0ry1SIv6AhOw5hhfZ9SMQP7ZOU/cxYnlgyRuwqCGydp2CXCn+bgtx38w9rbDkNEBX+1txGmDgQmTx2R1YwfpbY383ItdC1AVzE67EvlYbIkH2Btucz7JzjlQ46UxqvbknIcxIiPkPzeQ2iCyNLyGB5UL7vYtdC7MtP04aO7tmnvdsSW7W2zG2difY7y7BN3o2EpZ+mry/iPMQCnSnIeXV80jAOix7L6mSYUAD3v1/Ng4rIbfX0aQa7n0RUzYLOLHU67EJxnGkze+TybzbeCWWuj62pr5B9UTr0vBfiBSGYTx+KGDvYvdMQCVL+tn5y1Hs8IqZ97VYmmHtHk3q0t3G1m8NEuBORy7fSZ8moIRja9hxg+L4kHW5g43hFKQATOb2HpQH4r9k6sSSa8Xe2r79EQSKQeS6/yHKkvs75Ao5D9i21OPbRHPUHeWrGFM2AAAgAElEQVTI6AT8EoVzrRdcvxcpir4mCIOJWH7YnPQiYvkgEq6/sn30umT/DK73QArzGoIyXgXe1QcZJF6lQMkvVHLsUKNTW9i51RDfcGOpxyNiM661UjcgpzEW3I806g8gJv/04Ppw5FbbhxJZwZqgj9siq8wednwqsl4ejayLk5F7cLPG3UZstvndiDSm8DFyLdd5NaEo6c8jtlE3RyxZok3tBvwJuTB9SW7JrGwCjxrgvMx7NrTzw4JzA1H80nO2jqcht9LtkfX6M8y1t4hzkGiVW5PGSrVF1uU7k76RWs86oJj40AUwn8t71k16E+QOHiZPOQIJkFuWei021dox5uA95H6ctQ4PQ26wJxCUMsnznmWQxboGhbhsSJ4kcZln1kEx7N+EazVP+1oczUCKqUnkurr3tPU1ATGIHyPL6S6ozvT3BK7zBd6bjHk2Ec/ZwDvBcXskYOwD9Mo+E7zncGSZmWVz+xK57pHbIpo20/6ODJ+vNKS2N0Gyt16ArGGJIrwTivc/lbR+d0X2uaVjc9GLiOWFwfwNIFWcHE2mBJ59uzs34H2rAH+x9fFPgpweFOb1Rtu9z5V6PCI2L5buH+e6VfQyXC64vgaK5fvRmIkTkbanYhOC2Ob9C6Qxa43c+eajciltUXKFGlTu4SQyGSgjVgba2k3cg0eTm6gpy9R2QHHNT9CMLqvkalMvsba9ThAvRq6AfRh5sj5jrs7Wx02D8ysaM3IV0uTWICv4NPu9eZHGPhQAbkPWvI2Dc/2RdfRH4C6kyEpckntn3lUw5tiOE2XDiXbc2955e6H5rlREGvjXkJX6QDvXEQneT2H5IOrqr+3xZyIBeDz11/3dCZgI7B/OR7XQDIzpQ7HlISO/jH1v55K65b9sf79DYR/ZigWtgmc7UZuhPNSe39fGdyekOPp5He1bCSldTkLx2tsgBfH7SOGWxNQvh6w1YR3ZihI0g7WXHdfEgplY5Ze343WR1XpEcG9F9bkakCakFxHLC+uhRV2BB1Hc83kUSCxGPV5OKGQp4aeuJginK0DT9scSb5Z6fCI2L5bmn6YEqReypE1DcagPALsG9/VB7mafIfey8cG1snftK9D3wcji1RoJVFeRZhfuigSe14CHS93WiI2a5+WR5WI+EijqEySWLVbb7P8di6xJLxLEHmEuocFx6/Cv/W5vjPd0VM85+0wfW+P3ohjzgsmWmrF/tyB31kMx19rgWjek9JiJEpB9TKo0SPamkOk6BWX8vhs4ODjfFYUALECM9GvIBXTVQvNcyYi8FJ5EQsQztr6/xbKkNvAdydh/h5Qvg+saJ9LazFVDM8gN0xiEGLd/ILfrrMKnE3LlvhrFT4/Irr1g7Na2MZ+MLDe/CO4ZgJjNH+27+Aq4I0/bcuqEIyVGKDSvgrxYPkfu/CvneUdFfRfB+PVEMeMPIEFsxeCe1UhLv72IFOR/KXXbIzZ4jhtFLyKWF2b2v61QVYULUQLZsCxjIhjfQh3xz/XQqGURjzDf9tCCpQrt/mg0qwIs/j9MNejOiNEzqLTLCbY4JwOjg/vbIc33+tTjRlhuWM8H2Q65WV4RnBuKki0Mzo5XxPLDYC13QlacvmTq9SJLWg0SwtYq9I5mal/yvayImPT+5NYK3gsJyG+j5GZ1ZQBukzkeYs9+guJya5VesvtWD9pRFEYEOBi5Hm9NGj/aC5UaG0Iag7syyvS5djIXwE/IFSDGIUFhIlIULMCyK9v1tZCF9FHkkZIkh6qIPWoJxrYbSuDyH+vz8XnWW6KQaYMUEAMy624ppJT5xGjA8LrWRjXRjMz4JRnYh9kanIbK2HXOjo397h6MU1ty67Z2Rwqxx5FgmHjWXBPcM9i+nfPJFbzzKdgGAbuhhHLLJeNufzsgz4QZtkaaNEFjEedhCLnePG+hOtavkpYnTMK7HMrDMA4poM4nFcorZu1VI9IIelHqtkcsOKfJvn8CSnb5le2hNQS1y+2eJK/I3wk8oQq8ry+K296EQIhGPOB+KNzuH7Znx+++irF0/1jZiv9Bapl40YjWv1H81jnkcZ2olA0tw/R0InANtnPLWV+ftw18TRS39SZpEpuK0vBXEwYM5/ooXvorY7weIiNEI0vafOBfwCpFal8ikOyMYrnnGwF5BoUfJO0fat/dh8g9PZv1u3+GwVxEmFDysr+j2KXfELjJkdHOFmstIyb3BpuHsI9vGOFbaHtLPu+BxEp4J0o8NxBZM3az6/1QbOo85IablCxLGLLE4tjiv1sb51CRlKy3UAC7DcUNz0cCx5HBtXaIQfkXspbu0oD/2WJphtGITZP2I0FtKmb5RYl2EsvoyeRaX2opJpClpgZ5RnVAYR6PkcZPD0AC4I8objCvIEged2jgj8iK9xnyQDgkuJbN3TCdQLFSKYiSWtUgwas98vR5AdsPbe0+ZPvJKZln2+X7HiKWH9IIehGxvBHYzmjPaBQy1BsJ0guMfg8K7t2VAglIg3Wxs+3BM43evEhuWdX2yDL+tv3fOi3YEVs2luafytrzV8ydD8Wy/M8YjDWRReNrI+IFk+SUMwbE9XSU5OUbZNnajFQLtrkxJ7NJNeHbhc9HLD8M5rYHikGdgOINLzeGbCpBrJ3d+3OKHPtpjOC3SCj5BYr3eQsJ0r8ntYz1QEk2rsk83xO4nbQ8zzW2XgeRW5IkqQ95LZlyTCWan3uMeG4IHGff1UMobnicEb5aSg4bh9OQouQpxEDfRVD6CFlij0KCxQfkKVFUrUiuAHYfUhQehOIU59i3chFpGExrowX3Y6UG63h3i6YZiKF/E7gDuRjPxGLzMt/aw/atXUaQQCfP+/raWv7W5uIU4NLMPcuShoe8Q8My5f8aucUeiRRq/7b2HBPcEwrYSYKoiqJnNgf3IkXaGBRXfkHmnlVRmbok83CMo6xAXFJ6EbE8ESlM2tle+gimBA+ub2r74m0Fnm+V/Y14nlm273a1/WAhUh6eTaqcbANsAPyq1OMQsbRYun+sLH39UFbhT2wjSxbo5UjYnEUFle1A1oYwU+poJDjfhgST2cbEHERq9RqALIYnAFvZuYpiRKoVgT8bJpbLD1Dm5FeNQP+ywHPNZs0IviFn6+8hghgyJEBOQC5Sexd4R+jufoSt26nGaIaJlcIySmONyXyUIElaieZlE6TMmo8szxcF13ZBmuW8ggkqpbYfEgR/sHsTS2miFGuHynG9Zn0eWuq1WE6IhJGppC7yV9h+/qQxNbdiArCt0+T7qS95TIujGUHflkWJy76x7+2v5GYKD7+1i23dPYG5qBZ4Z1ekVPvI7h+PmL+QeeyAMoJPQ8xid6iVtCvZD5ZHCrRRwbWNSJVrV4bvTea31GPbiDlZHtHmhcjadVWee1ZAngTzkYDWodjtjNjoeV5iehGxfDDPvvUUMDG8HtCLc5FSsX8D3tseefzdasf9bI++xfbpBUi51jPbjvpoWsSWi8X5J6lbRb7yNkehmLnEFbq1MUpnEGQjLndETPm7SKs/2hibm6x/yQfdHwnXXyNheoUC76pYhqRaEGk/n8AyiyL3/heQ4PUzYzbnAdcWsU2hS+75yOp8R3AuEWLaIPfal5Lj4J587tJnWX++ArbJXAuf3dfuO7pI/c3RMKMsx0nW6hVQfNzA4J6uqL74P5L7CryrPRLgnkBus0cXuG87ilxmrNwRhbvcC5xqxycgq/Vgm58k1vdpLClQgfe0eJqRp1+tUAxzDVLUZcuehd/aaBSOUStjcWaNdkQCwguoZNZewbWQCdwF2L2Otm2O8oG8TJAA0a4NQAL/j8j1vKKTPFE7pOtIlDxvKoqlzDLxyyB301NL3faIDZ7XRtGLiOWF5PI+69nf2xA/vrIdZ/fPzwgSzdbx7iFIiE6UxZNIvfl6Ig+qL22PjZ5sEfG+CMI1qbWnN4rtWiNzPikBksRxrmmb2bHBOyrCkovKb7yELAUX2u9d7FqYAfZxpA2/AFiz1O2OuERz3R3VMO0E/BKVpFkvuH4v0nh/DazbzG0ZGvxuQ+oOXoNix8L4zETAPteIS5cC7wwZkdOQxSqpDXlY9l5SBVK9rqVN1OewfaOQa99/7NvaPbiW7DNrIev6t8CQ8B2Zd62UPIfcBO+3b/XkAv+7Tfg3ogdZ9bdESf6mGyOTWDL3RUqaWRSOcasampHpd3fkvn04ymUwB9g3c08r0gRLXe1vyFgmSom+GKOHlA+b2LexgALKouz4Z85tbvOWuEB3zVzvgRQpNcBlpR7LRsxBMn5LI0VRK5Qc7mdI4fEmiqvMJngMGfeKW3stHZuSXkQsX8QMCihnymCkFPsbAZ+DlCq/R/HTPRr43l2Rl8/hwBRUeSHhpe5Bnj//pQGW8IjVgc378lyifxhyvXmLwGKBSnc8g1xyXkHW3zdLPTBL0NeEUe+GSsXMRNqs7YJ7ssleaoDzSt32iA2eY5f5m8z5Bcii08mOO6HYyFOBDcJnmqFN29o6Oj9zfiVbhzXI8twruNbK2jyFPN4TAYPZmjRmsq0RqzuTdRuMQwfgRnJdxovCiKAyZ18bk/QwsmjWABcG96yLYkP/B/xfZu5C691ZyK3/Z8G51Ujrml5HGi8cGa0C6yY43gS5am8YnDsCuBnYsr53tHSaUaD/rZBiZwhKHrgQOD24PhwxhX2ya5BU0bMiShT3OIFiDyXou9nW8qWkzGGde1PwnQ8gVbAdTe1a2Z2AnUs9ho0Y+2Tfa4fCaa4lVVAklQReQZml9yW6gFccNpZeRCxftL3xc6TYbWvf7C+RgD0NeTwdhniX+cB+9lw+ZWIOjxecPxt4Jzhub/vEPqTJIuNaidh8wjW5msLfGKMwzTayL4FDg+vDUUbGx43oL2/nK8oaRC5jmNTPe50gBpVcAfswAqtixPJE8mTMteOEmb0EuYAn63ZdZLUeEdzbXML1ekhz+jlK8hTGaXZB8a01SPjdHpUYOQQpf87I877wu70NWQQ3Ds71R14ZP9r/24lU+KzlotrM87I5in3al1ToHYQUBzVYAifk3rcP8NM870iI6Em2L51IJnkNElZOQ5bEZ4rdz3JFagvT7TPHG9o8nGjHvW3N3F5gvVUVzQj2lY72XfYiKAWDMtxeYd/aBFub08mTGJFcJdHbqN7qVnnu64OUSN8ii063PPcUZA7tW3oQWcDPwzw96lsb5Y6Z8XsWZQI+gNp7/upIaTkHWTbzliCMWH7YFPQiYnlhQL/XA0ZgVuvgekdUMvAx27MWoJw4Ywq8L1G6L4OUhVkFYuI1ta+9eyekQP55qcciYnlh8/8DMfXvIoZ+Jfs7ESudEty3KB7U/lYUcS7Q9yQT64sEcWpJX4Pjiu9rS8VgPfZESekeQELnisE9q5GWyXkRJVv6SxHb2BvFPH4KPEcgHCLt7ZlGEGqQFfARYGxwTz7N7S3IzelQMgw48s4YbWv7WxTXtFk4XkXq91FI2BhALnPcC/gD0ljXqrGb7S8qpzYDJZRqm+8+pKg4iCLGlJczkisIn4Lc5+8GDg7Od0VlbhYgge81ZDVatdC6s/MtnmYEbe6FhLVptl4fwDKi2/U+KEHcZyjL//jgWq3+2rf+Frmxo60ya7mrfb/TCDxasnMCbIVcaC9ErtBheEmiPL4FWKfU49mE83Iacv8Oa6RnrVfL2xqtAfqVus0RGzy3TUIvIpYXonj5j5CB43k7tyhULbhvPWT4yPHiC34ne/LaSGk7GXnq/CK4ZwBSLv6I+J+vCPLaRIyYYPO+XAt5FnIFDK26G6AsyzUo6Vdoza2ojSwgwCsiLWh/grqewF62ob9tm3vFMIDVjMgtM1yXb6H45VdJy6btkawB5Io0DmVEPp8iMfzB+uuMsta+a2tt88x9e1ubp6BEXImGNl+N3IORsLk1qQDTC5WRG0Ia87kySgKzth0X1R0KWY4WkNYGDgXjDWx/qStJUzJ2hwHvU7s+edb9vwPNHD9f7ohcY0OGZBzympho38gCLKuqXV8LWVwfRW71SVKYvN9FldCMRSEJSCn3DLADilmej5i60cH97VD8byjwZb0GkvN/RFly2xX430kSuEVWmcw4J+85AQmZX9n81hDUuLd7Rtr5vxNY3CsRSWOr77MxLDS+icVzaTK5ACKWNzaWXkQsTwSWQt44k2z/3Dm4VlDZb/tvW+CQ4Fx3xN89jowoSQLOa4J7BiMe6XxyBe/I20dchM37cjHiP2IZhjOb2eqklq8/UyBzdjljwCTtjLKAz7cP8RmUYTRx+xuKhLIPkQtjjNUqY0TWixrbQNsjD4QXkOKkPYolfQjFQ56SebZdsC6KstmSm0znYBQX+AlBWTi7NtzOT0FJp2q5NBrBucH6F67fN5Br7kLkjltUhjLf/7P9ZT7yKEgYpmTshyJt9vYNePfpSKBL6n5nNd4jqa2sqLq4KqQ8rEFx951RDO9/gd3sej8UkzYP5SBIyg0mCpokqWPBtdPSaUamrzehsIuk1NuLRif+bevxHPLUTs639uy7dfaOp4JzoSKkN3IzL1i+y+7bzr6r0cg1vbd9AwtsXgcF9+5KgcR0lYi2dz6YXXt23B15t+R1hY9YPtic9CJi+SC5isEDkWFhLrkeVAX5MOSZUwNchZTnhyEX8iR+egASon9EyTXzGk3q+h8RqxNb0bwwFbna7Oeca++9/8E51wbAe/8/ZNF9AjFpRzVzW5ocvPc1zrn+KDP0RCRQH4is2KcDv3POdfHe/xtZJj5FgvW8UrU5QoPgPsPrEVPZFnjRe/++936+9/4lFKN7DXC+c+4651x7AO/9Au99jf1e2ByNc8658Nh7/2Pw+zYUr/oWcLlzbmxw7QXSpB/3Aj/P8y6PXKB7Ahs4545DyqL3UBza1chCvnKTd6wAOOdaWbtwzg1wzg1yzq2DrH4TEIP/O+dcb/sml0UKr3aI2NYH/0J9PhQ0ns65Vvb/utv7/8851zp5IJnjKoMv0L62E7JidkPC4EsA3vvpSHl4PEo69pZzbpD3foFd/8H++jr+R4umGQk459ZCyrrzvPcznHN3IXfj4SjW93sk3N7unOsTPptv7XkDlKRpG+fcz+1UeG8/9F2sVqBNzjnXDtGwp5AHwmTv/cfe++uRm/g66PtP/u+fvPeX2PPNzU80GwT74OvAps65nrb22gXXB6EESWuXqJkRGgBFoBcRSgR5+JWFwe87kMfNy8A459xZyT0h7c7AI4im/QIpjZcDJnvvP7Fn30Mhdycit/I3nXMDsrxdc/F6ESoYmkpKJ7+msAMSUD5B1r/uwbXeiEkaRprKviKSg5BqrxxigB4iiEtFZUkmICFm7wLvqDrLVyUhYnQvRZbaBcBVee5ZATGa89GGXlSPBGRh+hUwHtiR3HCEDe38DEQ0Qg3vMihGduMC790EZWmejyyTFwXXdkEMyKolmJPTkMvsPBTfniRpe8y+tako7vdFa3/BzOCZ9/ZAQsmXwAnB+e72P2dR5e7gwZgsDeyHrDw/2FpILK+J62w75BnxGrIKDC3wrqqhGQX6fzQSeLe3/v40oC2XI2XsLGDbPM/mpR9IAPw7UkIcE4zpT5Ai5JH65gEJ1hPD60G7zkXeAxVdcqaO8dvI5uJNgjKZSCHxOBbTGbH8sbH0ImL5Ig3jfWba/NZpVUY5KH5hNK3Gnm1DrtdPB2RcmIZ4wu756FfEiAk2zUtyGfdOSDBZxo67oALs36CkEZsjy8fdwGd2z/Eo5rHs47YyfT0fq6sXnFuUZAe5/L2UHAf3xI+yTDGzoXZC3ghfGCEelocRXQZZt08tZvuQdel9JMA8htyWLiVILoSsh1cYMditnr62Qtkvl7LjFVA8dZgYqSsSfP6R3FfEedkXWfN+h2pNnopiQl83hulwZI2fjGJ7E1flfALcqkjY2CPYp9a1fWouYrauseNZwEHZ8ao2zKyV9jbmT9i6y1s3GTFA1xR4X9XQjLC/+dYQssBPJo2Hbo2E6zOATet6F8qQexJwHNDXzg9DiXjm2/fxOior9E/IjdnOzMN69vc2lKRwZTsOaddolFytb6nHtLFzYb8HIWY8LDl4JMpd8QNy3b8feWd8SJqVPrqBljE2Jb2IWB7IkvE+NVhunELvs98dkeHgBcQD7BVcC5Pf7UKMy4/YAGz8C3IX3hmojMUHRsyPCa5dgbTpNUiT+D6wvl27x4hXp1IPSB39HBr8boNc+r61/rxBbibVRMA+1xiRLsVsa8QlnuOE4VwauQclSW5+Zmv6TZQ1NxuTGzKfzSaAkTLGQ5Em/mQ7HmHr8EdkddosWIPdyRNHliEso+wb/A+yzuye53+uhRLCfAsMae6+5hnXCwwX1Zk2AjrR9pI+dr5jgTFL5nY3m8dvgO+QEHE6UhysZL//ZnvV7SGRrVbMrJWVknFFQsn9SHlzcoH724R/wzmx3y2WZuRZg71R3N4amfNJeZfN7XhNpMA6NjummbF91MbkM5Ss8HMU897Ovo2RyArzR/vdPTsXwbtux8rYoIQ9X9h30CW4pyPweySg9Cj1uDZmLuz3OGSl/s76e1BwbTu7Phnti5cBqxcav4ilR5qQXkQsL2TJeZ/tCrwv4Qf6ktayb4289h5H3op5lcZxrURsCDbdi8R4z0JlO85GhdWzWfYGIber4aRWimNsIZdtTUFgW+vL+ZnzK6EyKjUoW2FOin/b3KdQ4Yl3qgGDzbYdcvO/Nth0WyFL5yvIzXpfSpSUzhjcB4Eb7XhNZKE6BQmOc4yB2JWMUqcAgTgTMeaPI9foybaeLwzuWRclWfofRXadQy7b99t3dkKe6xvYnFxk85SvPFFCmAcjRvoCZAntjWLra7DwDVIhZpl876g2JFcYOQsJwj8Lzq1GWuf8OgJmtgHvbrE0I2h/aCU9DCl13sJK19n5VVBeg/m2x7wLvFnPey9HCoktSRUe7yM3/VrlhILn8u0BSR6GLZEysRWKLf4CuUEeZW0/z9q4X3ZtVBoiC/yXtpaS0K4a4LLMfR3JePiUuu0R65zXRtOLiOWJNBHvQ6rwXRHlW3mcIOwL5fO42faDS0mF9Yrd7yIWH5vmJbJgfIISsSRZGDe1xfnrAs/0sgX8IWVeNxYlMrjHGJC7CAQr5MJ4q/U1ienphWqzzgTOKHX7I9Y7v6EA8SxyCz4gu5mibMV/sk18LCWI90QW1nOQNb0tssLegVxr2wVEYQ5BbfUC79ocmI2UBYlQNAgJnzXApcH/3IcSCDPWzzmZ9rTOEMt/kieWNPOe1qgm78MEVjdUF/w5oKsdd0juL/W6LAckVUychISREwnqqNu1FVF84xwkJPZuwHtbNM2w9oZr9DcoCdw06+OXwKHB9eH2XT+OGLrE/Tiflbk/ip8+ipTx2wJ5EIy04x7BWi5odUG0bYTtIZ2D6x0Rw/oYUmQssP85ptTj2oj5SBSoHVHYx4nBOPRDyqMFZHKoFBrDiOWHTUUvIpYf0gS8D7m83ttIWN8qz319bD/4FnnwdGuOPkVsudg0LxERno5pf2yxf4qEzlbIzfbULIEC9idTLqhcEVm5LrZ+PUfAYFofz7QPuwZZJh4Bxgb3RK1XmSMSED4gt55sds0uj5QoNUC/ErWznzEMO6IScEOCayciAfJO6kk6hJjz6ajcREh0eqFY1y/IYwUr5lpG7sc7ITfZeVmiaXvLU8gCXatsUebevwLjg+O7kbDT346PJRB4Ii4ap/WRtecYcktjhWumC3CQfRf1Cr7VQDOCNt+IrNGHIG+nQ5CFZRZwTnDfonwd9rdQLfC+9uxhdrwqYgIvtOOuyBW8zrJCKK/CR/ZdPW/nWlE77GU95L2S45lV6nFdwrkYjFzln8Ys8MG17qi++hfIojWwFG2M2Kj5bTJ6EbH8kKbjfc5EfHqYU6ZVhqZ1RV4t08K9L2LEhmDTvERZT+cFx6+ghCqJNWgHYyY2suOKEjRJBa3OKDv0u0jrla19uzdysZ2CYrYSF9MYo1XGSBpbfZ8xpdkahsn8J9bdpcnETZao3Xsht6gt7Lgdiom8LLinrhqPY5GlJrEchoLTBkhQKovkHcbg/9XaNBa5bg1BGUMXuQgHc5X1OuiEkpXcH/R9LrCJHXdDiZyuIdahT8YsGcvDkMvxWgWuJ/tcBxqYVb2l04ygX+siQfgIcl3EN0C1umtQ0qz2wbXs2s3uR6sjAXCMHU9DVvFkfxqKhIv962nbUsg6M8n2kZ2DawVpVqXNBblWyz2R0FWD3Nw7ZO5dCrmVfgVMKHXbIy7xnC8WvYhYWcgS8j4BzfojUrK0K/D+JLlkJ2DZQu+LGLEQLv4D+bPvrobcS89HGRjfw0r1GNN6B/AkFawpJDdZxsHGDH5CxoqC3Ps+QQL2/1HBpWKqDW1OH7TfbTPXuiPL3UqlbmfQplVQMqPxtiaPQMqdX+S5N993O8QI1DWkAnYiKA1FVq06rV9F7m9vpJWuQZbU15Cr96g6+rhd8Psoe/YPKAFKEj/eBlkTPwW2LnU/yw1RkrdZWBwbtS2bI6mtaAwFmqqkGdaXIbbWtrHjUIG1Ogod+hYJ2gVzcyDFxXHB8RUom/UUpDRK5qaXvevZetoVCvoHImXxXODgfPdUKpJriepnfzdHLu4zUcWAdpln2tj+1zH7joiVg0tCLyJWBrIYvE/mOWf4IvBUcC6kV71tf1271P2MWLm4eDfnLsABKPYrKf9xPdIIfwfsYuf6ohiJOZjVgwpxJ6tv40WuR08bgR6buTYAWV1qUGKYuImXMZJqM/9gRLinHS9KZIGUJpMIrDvlgCipxxdISJ4BXJznnux3OwhYB1nr70RC9CVYrCywLNL2z6DMSu4g9+PzUZbvScEc5cTV2bkhth8lCZhWQt4Jc7E6vsDKyJ35a2J+hEJjvh2K5w2Fu0QJ0x3F5Y4jv6WgamhGgbHrjpQ2N5MqsEJF7asoAdNU4Kw63nOGrdu17XgIis/+Frmdd0fWnKS2fV7GHDEAACAASURBVFJ7PIkzro+e7YDo2eywHfnmtBLR9rcpmPIMWIO09N4xwNIFnqvYtRdx8ehFxMpCGsD71PHsyUbTfp7n2ibIOzVvCa+IERuCS/aQFmZSzuZlpPlubQzELKQhvNmuTcdcSyuRUNOwYvUzkJASWgOWMaZp41L3IWKtOc1LVFFW4k9sba8ZnF/NGNnnS932Au3uhIT/DevqI4opn4wEmjmkCfgeQ8n6pqI45Bft2y5qZvDF6G875FY8G/gvaT3erDttT+vv08G5IcgNdy6Kr//c9qtLg3uqUhlWqN8oOdbDKAnXCcH57ramZlGPO3g10Ix844csztfbvnIsVg7LrvVGtcKHoYSZ0yjg6YTcWr8CbgnODUMZ2r9BTObnKNdHUq4sXzK0htCzmbYPVMzYN2BujkXW+TdI69Z3RyXIFqIkjj1L2caIzTb3DaIXESsPqYf3KcS7IAPD343WHGPnOqCqMK8SE95FbCQmFrsGg3Nub+RacxVi0LdHSUJu9N6f7Jzby84NREz7S977vznnnF/cf1YicM618t7XOOcORDFpsxFztD2y0NzqvX/d7l0FOA5pv/f03j9UomZHaAA451p77xfa70Eoxu5D7/1nzrm2KL70JJQ44zbkoroisqht4L3/InxHOUKyfjPn9kXf7WUofnY5VN/6a8Rs90WM97ooY/oT3vuHyvW7dc45lOX498gKNcR7/1pwPfmGt0T1gC/w3p9r17qj+d0SE+y895PsWlnPbTHAObcqWh+9gL9472c759ZFa2cjZAGahCwH66N6zLfnW3f2vmqgGeG+0gntKwts7LqgDNRDkcL1Dyh/x4EoU+0KzrnjUfKcjbz3X2benazlUcC5KGHc3XatM0q8sx5K8DPTez8rnIslpGejkXLtgeYZseJAuIacczugREY9gZu89+c551oBv0XZ3J9GJflml6q9EZoH6qMXEVoGZPa91t77hfaN74zm/UeUd+VD59wwtB42Rkp4kID9NTKK+cgPRFhiaIgETq4b228QgU7KfKyGmKZZKKlKkgggG8dUEVrCpJ0sfrH6solNjVj33NrvcYjB/A5ZfQ4Krm1n1ycD/0FCxep2rWKS02W+2wsMF9UhRnFLE5Gw3cfOdyw0ZuWIyHp3H5m6lkEfuyNr3mvAivW8q6ws9EUex8R9eDdSC/N3yJ3ydCTArWS//4Y0/rcDezVg7VUFzbDfZyDl1AeoLvwxwbUrbNxqkPfI+6RW5nuAl5AlppC1ZYDNTUGrSnbMGkHPtmto/8sNqZ0ALkwWty5S4Hxm85HEqo/Gsq1HbLkY0Iu8YQARWwaSa71+1Pbaz5Dg/DlwNvJoWAXlDBmPkpyNxLyLqCBeL2L5YcNvlGvgBMMj7VxCuLsh7e+HwL+BYaXuWKMGpYmK1UcsTzRG6kvkbTAaWZVqCDJNBuugXhejckb7bu9H9blPyHN9AxTWcBESRivOFZTcJFG9qJ0BeB0kKI6r5LlsxvFL9vHBNk4XoJwSvY0RrUEWvUXjRsZ9mTyCcDXRDOvTWKQwuNCYt2tt7K4J7hmErP/DkzG0fWgB8NNgfFa2e3tk/sfO9s7jFqNdVUfPkOfFlcnegPJLJGt3RaRQqEHJ9AZmnq3IPkds8NpoW+o2RCzaXF+OFJ1bYslokaD9NnnKjAbPxT0gYqOw4TeqcPtcI0in2rk2BDU5UfKuT1CikIqwOhToa6OL1UcsLyS1zHVE2bFPDBjZfsiytgAJ2t0yz1bsRmtreI6t10uTsSBXafBPKjDGyAhm3+B4jPVzgv3uAnS2a2cjQS4pwVKxc9pMY9kauW4/HAp0wHOGSYmsxPpcrxKmymjGhtaPA0gTl21qff91gWd6GS350IS8xILfBpUlq0GZwE9HtayTOZiASm01qBxgNdIzZIGqQRnnkwR6rUit831QIsdvUL6Uil17ESNGrI0oeearqEpI8t1vgXIsjLTjHgFNizxBxCbDVjQc/gT83BbrWOfcLt77H733Pzrn2nrvF3rvbzSitr/33lucS8WB9/5rlPToUWRNaANc7r3/1nu/AGm9HkGMaIzbqQDwir0ZjBjJ1YBPvfferk1HLoLHIIb4KefcwODZWnGkFQR/QvXX/wWMcs7tad9qEpe0NHKVmu+ca1/Cdi4WOOc2RAJZn+D008gttwcSJl4HxjnntkcW2Lao9E6lz2mTg1dc2crAt977GQDOubuR4ukX3vuvnXPHAvsF99cHVUMzkJX/B+AN7/18i7u+D7nOX+CcW9o5d6rF/wHgvU+UCv9G8YDf2/kfUebvI5Er44lIwXGPc24XpABch3Qt+7oaVqX07EbgcKSUeNw5N8x7X2P9BeWYmIJisE+tbwwjRIhQcfADcvte4L1fYLlE/gxc4r2/3jnXFXm3bAGRJ4jQtNBg4doL/owI1svAn4zZwnv/g3Ounf1+xHv/qSUWqFiC5b2fbgxkZ/SBdgawfvYA3vfeH+C9f98517qETY1QB4TMLGJg9wC2AQY55zokF7z3XyELxmEotvHsYrazuSD4bkeimM4JzrmxzrmBzrkhiIHfErjOhIKKEG689/9CtZVfds6t4Zzb1Hv/X+/974B9kJLkUZTo6XGUuGQecKRzbvOSNbxMIDvPJgy2Bdrb8VgUl7uf7XHdUAjB0PC7qQuqjGb0RtmmJ9nx80h4O86Ytk2BXVDs86Lx997f5b3fHZWImuecO8Y5N9x7/7n3/gY0doNQErRWSGFxDJqn3zjnujbkm602emaC9C1o3/sB7XsHAjjnOqLkeT+gfe+jDJ2IECFChUGefasjyinRyY6fQZ4sY+14VaRQXq4Y7YtQZbAk5m7ESNyJ3K5adBwjS1isPmLpkdxEQ/3s7+bIkjYTCdrZJEptEAPcMfuOSsfMdzsDWaluAUZVUl/JLXnXHtURng7smWc++6C6lfcCb1GB7u/NPJbbBb+PsrXxB8SUJKXY2gCH2DhvvYT/p8XQjHzfCfKGmY1q6t4LvAesate6ITfsJwmSa9m1JCa4FRKia5DL9whqJ+bqBKwNXI0s3iMKtaeOtlcdPUOC9JNIufYMstB/S54cFBEjRqxcRNm+jwuOr0BKtCkoxCZJYNgLWbGfLXWbI7ZMXPIHYWmU+OYH5NZWcUzSYvR1iYvVRyw9ApfY5rq1Ha9hzOlcZAXKmzm0Ja5pFId8PsoCPYk0Fql1pfSXTBZPVLt6Oop5HUnhWsGrk8bZV30mUBu375B1GpQN/D77LibauZWBg0wAO6OR/6/iaQa5uQoGoLi+JKb3ehPgvgN2sXN9UYjCHGCtPO9IckEkTN8wlM12GlJodC7wv5OMtoutEKtGeoYUHL9C1R8eBY4PrlWEUjFixIh1IwoLmwusbcdDkOfatyhUpDsKuRlve18Pu6/iErlGLG9c7DrXIVhd4OOAdb33ByzxiyoAzG1yA2Cel0tq3nrCEcoPzBV1JEpkcZFXTd7uKJPk/ihT9mXe+89L2MyigbmC/hL4HRJId/TeTyv3usLOuf7Ax977+Xb8AIodfd5qCT+FSu38DrjFe/+x3dfWe/9D8J6y7mexwDnXE7kvf+C939bODUEW7H2Ar5BnwCdIw3+C3bPE49dSaIZz7mSkdFgJeURcgxi2G5BHzDTkIbMWspIc571/0FwXO6BwhX97hWJsibwFtrLvsAeysK6CBPPbkr3JOdfGKya7se2vSnpmLvQdvPff23GL73OECNUClivnReAh7/2hdm4Y8tLZDykUa1AS1996719rqj01QoQQGiVcQy6jVU0F1yNRLn/IrM0dUPKansBN3vvzLM7ut6gO79Oo3NDsUrW3mGBM5ggUi7wGMMR7X7bJjEwQvAhZ8vZwzl2DLHsbApMT4uicG48Ew+uR+/F/S9TksoZk/zLB7lHgAu/9uXatO0pktiUqLfWat1jiptjjK51mOOf2RuEUVyGL9PaojNmN3vuTnXN72bmBqKbyS977vyX9ds4NBW5FYRlnA39HrspHAQuDtfww+kavAK723r/bjH2K9CxChAgVCwFNGwWcCxztvb/brnVGVRPWA94BZnrvZ8V9L0JzQaOF6wgRygmyzLpzrn1g6UysmkOAPwK/8d5/45wbjVw7Ty5Jo0sIptU9EcVczil1ewqBWdoSL4OvUK3abb33L9j1RdZpS8Y1FsVUXeS9f640rS4fyMdEmHKpK3Abik3fyXv/6eK8o1ogtG44536D4tDP90pCthpwLFqfzwKHGePWzqfZqQmE62WBfdFe5IFX0NjPs/vCtXwxcALyyviV9/6NYvU5QoQIEcoVCtEj59wApKyc6r0fUeDZ6L0WoVkhZsiM0KLAq+TWcs65K41Jne+ca2sb8SSUffd1YDRwk3NuoPf+ykSwrrassd77l4F9y1mwBvAqG3QDcuXvh+Ja2wXXf3DOtbHfZyFBZydkUax68GnptV5Jtm+vjMpfoTrKawCnJffn+w6qVbAGlcdyzvVwzk1AZbA+NcHaee+nIO+Xy5AnxdNOpZ8WZN6RlP6bidzIp6A49FWB9YP7wrV8EhLct0BKpQgRIkSoSkgqI9i+W+OcW9k5N8hCaQDw3r8HnALs7Jw7Lt97omAdobmhqgSJCFUDewJHA4865/oGVqB2Zpk7FMWR7gCcHpayqUYBIoxHLkfICHrzkMv3m2h+D0sumADUyqyMdwOree+vLnJzywacc1s65/oGx2OAj4A7nHNjnHNdnHOdvfevAxcDuzvnflqq9lYADAN2RPtLVzvX2tbbLOA85ObdCzgv3FfyQFeUVfwItBc95ZzbN7kYrOXO3vsrgRW99x+7FlAmK0KECBEWF4wPOMY51808gNoAtwNvAPc75053zq3qnOvqvX8MuB/Y1zm3hj1fEWVGI7QMiG7hEVoc2CZ8CLLIfQ/80iy0yfVhyCXzEeBe7/1HpWhnhPohcfM3oWJ57/1nTkmx1gBORUlKzkeZrL1ZZa8Ebg9cxqvOndk5tyEq6bRpsvYt2cvuKB54PRRP/VcUIvEpcj1+wHs/qiSNLnMw5mxHJEAPAvby3j9q10JX7hEoWdmndbkf2j7lUWKx04GfoSQ759n14ajs1k7AJ9W2hiNEiBAhAefc2WiP3DgIoelp57YFtkH83hsoH8YywE3A2d7780vS6AhVC1G4jtBiwaxwFwHLA6d57+9wznVE8Y77ALt577+rRuGrEiCcF+fcbaiE0Bjv/T/sXH/gSBST+kfgLmBnlBiqr7ds4dUKzrnNvPcTTXO/nPf+RTu/IrAsGrtNkKD9JErAtRKwpff++RI1u+zBObcecClK+Ha89/4KO5+NsQ7Xb6Ik6ojKQgEs8N5/adfXAEYhj5sHUZKzMUhI37M4PYsQIUKE8gXn3LLe+5nOuWOASYECfRlgKbR/bgT8FCmXN0bebr2A2dEdPEKxIArXEVo0mLXuchSzOBHVO/wpSmZ2aSnbFqFh4Jy7BbnkXgQ8aC64ybVuwAEoE3wHVI95bxMqq7LERpjUzznXHpWFWoCUEI9kBMA+SKA+Dllj3y+UBCZCCs653ijT/n7AVd77MXY+X+K4Nubm3Qu4DsVsdwJeAG713v/J7uuDynj9GtUCf8F7v59dq7is6hEiRIjQFJB4Bpm3z0Bknf4Xygr+50wS205AongfiMqsPhKTmEUoJkThOkKLBxPADkdJrj4C/uK9v8yuxQ23jME5dzASqvcBJnrvF5iQ0h+5gE323s91zq2M3GuneO/fqGZvhKxSwal29QNAW+As4B6fp+Scc251NH6+WhUTiwPOuaWRO/eJwCRgozyCdVIextk9M1B8+1oonGEaEs6vtPvbAV2AvsB/bC6iYB0hQoSqhcDzp4tXhZdhwEOIBzgTuM97P9fuDT2Gunvvv4p8XoRiQxSuI1QNGIPbwXv/vR1XrQBWCWDzdT3QA9jTiOtQ4A+oFFc3FDv/20g4F7nJf+zT0nMPAJd77593znVBMdVJObpbErf5MF7YjiMj0kCw+P/jgHW99wfUcd9NwNrACO/9DOfci0BHoAYYgGIEz03mLngu7lERIkSoOjAL9HooNGa+c25LRPu38t5Pc8oQ/gywCnAOcJv3/nN7NiqHI5QUYrbwCFUDXvB9cByZ1jIGE/C6AD2BDZzKajwDvIcs2VcDJwMrl6qN5QKW2OVM4G47vgZlw59pjMY33vufoHjes4AzLGSiVrb4KFg3HGzsLk4E63zZvJ1zayFPi/NMsL4L5YEYjkIavkelAW831/Dw/XGPihAhQjXCQKRcv8k5txrK/v0s8KnRtBne+3VQUs4LgFOdc6uCqi2UqtERIkAUriNEiFDeMA7FAk8ERgI3eu93894/gyyx7wGxPBHMRWO0jXNuKio3t733/i2L920LYDG8Z6GxvMg5t0XJWtxCIFRG5HPf9t6/jRjDSc657YGtUPKy+d77d1BJru+B7RBDGSFChAjVDlNRjoqfAf9GYTWjvPfzMzTtZyjB5LHAVc65tUvV4AgREojCdYQIEUoOLqhl7VTft6Nzbinv/UvA6ijD+x7e+1/ZPV2BEcAcVCe4qsF7/633/gaUvK8f8B3QLrj+g1NdULz3Z6H8AzsBg0vQ3BYNifU6XNPe+6u999ORBXsO8IoxiIli6GpgF+/900VvcIQIESKUGXjvZwLXAFOApYFVgfWD6yFNOwkJ11sAXxW/tREi5EIUriNEiFBSyCQgGQWMB14G7nfO7e69/8x7/wDwjt2zFiKkByBN9nehIFNtkOn7PORK9ybwqHPusOSCCXOtzKXubmA17/3VRW5uiwaLV19o2cSfthJbSf4AgPnAaij+Gvs9DJjj01JpVbuWI0SIECGArsiz5wikRH/KObdvcjGgaZ0tKeSK3vuP84XnRIhQTIgJzSJEiFAW4Jw7EyWHehmVjloTCR8Xe+9PtnvWBW5Gsdi/9t7fV81Jn4Isqq2B5b33n5m73BrAqahU1PnAGZZ5ugNwJXC7T2uEVu34NSVkSqAdBlwGfAiM9N5PtPOrADeieOs3gWWAed776EEQIUKECBkwZaNH1UBOR27iv/Xen2fXhyPPn52ATyIti1AOEIXrCBEilBycc5sDj6DalA957793zg0CDkRJyy733p9g7uA7AF9a3HXVQsbifxtSRozx3v/DziW1Pk8A/gjcBewMHAX0TbKFR2g8ZObiN8h9cQNUQ/wr4GTv/S12fTiKrx4K/Bc433v/RcxwGyFChGqGQFncEVUDAVjgvf/Srq+B8lUcjZJz/h0YgzKK71mKNkeIkA+icB0hQoSSg3PuKGRp3Rp4P0kSZTWtzwN2ATbz3k/OPFf1ZaOcc7cg1+KLgAe997OCa92Q+/yZQAfga2Bv7/3EKMw1PTjnbkQJy85FWWy3Bg5HSfmu8t7/xu5r51WzvU0Sex1rWUeIEKFaIdgLe6FEZusAnYAXgFu993+y+/oAewC/Bn4AXrBEncR9NEK5QBSuI0SIUHJwzo1FLl9LW03LRbWXnXMboGyhe3rvHyxlO8sNnHMHI6F6H2CiCWy9UOKs74HJ3vu5zrmVkSV1ivf+jegK3vRgIQt/A04Bbg5cxDdAwvYOwC3A0UEt8qpXDkWIEKG6IaFHlptiEjADuBhYC4U1TUPKySvt/nYoNKwv8B8LeYqCdYSygZg4JUKECEWFILlTCI+iuKrLnHPtLRNosj+1QslMvitWGysBbBw3BV4CnjPBeijwJPAQ8E/gFBPgpnnvH/DevwGxfnIzQRuU1fY9c21MSsW8inIJzEJZ7x90zq1g16JgHSFChKqGgB7diJTC+3nvnwD2BN5CFRbOds6dY/zBAu/9l97710ywbhUF6wjlBFG4jhAhQtHAiGDi8j3AOTfIObcO8DowAdgV+J1zrrdpspdFccLtgLdL1vAyBBvHLkBPYAPn3HHAM6j29z4oycvJwMqlamOVwVTgC2C/QEGUlIr5HzAdeALVsj6qdM2MECFChPICqwLSHzjPez/DOXcXsDxK/ngAErpHA7eba/giiMriCOUGUbiOECFC0SBI+nQa8GfgFeBFVM/yDuQSdgAw0Tl3N/An4ERUcuvDWKaoFoxD8bwTgZHAjd773SzZ21NI0I5lSZoYCnhffIfW647Akc657klMu5XmSlwd/wEc7JxbpljtjRAhQoRyBu/928D9wCTn3PYod8UoYL73/h1Ukut7lAxyYMkaGiFCA6BNqRsQIUKE6gKrU/lbVKrofWA5REQ3Bn6F4qi2QzHCzwIXee8fMvfmqtRQZ7JRtwLao5wZLznnVgc2Af7rvf+v3dMVGIHc6T4pUbNbJGRKbnUClkIZbWc7536FysedA6znnPsD0BllvV/Pe/+yc+4nwEZA29L0IEKECBFKC0Fm8EW0zXt/tV3bCdGuV5KEj/bY1cCz3vsXS9PqCBEaBlG4jhAhQrNDJjP1usAVwDlWcqsVKhV1B3AtsLn3/ibnXEfv/fclanLZQEawHgVsjspufeqcu8GSvD2QWFPNvW4v5AGwuff+u5jArGnAFDyJYH0GsA0wAJjnnBvnvb8K2No5dwWwG3Awqtn+CbJog5RInyErTIQIESJUFST7qHn03O6cG+W9nxwkeJyPlJRrA8/b72HA+ESwjjQtQjlDzBYeIUKEooBzrgcSntsCf/PeX5q5vgGKSb0NZVxeJMhEAOfcmSgx1stIYFsTMR0Xe+9PtnvWBW5Gsdi/9t7fF5mQpgfLbn8cSsAzD8UGjgSu896PsnsGoVIy7YHXzbJ9DHApsFO112mPECFC9UHG8+cw5MH2ITDSez/Rzq+C9tbhwJvAMsA87/3g0rQ6QoTFg2i5jhAhQrFgGHL37gS8CyK0KDdXjff+Vefc/7d3r7FyVlUYx/8PLQXBGgxasU3rh4ZwTYqUi00NQYjBKhADAoEaFVDCxQLFWEFoGozlFq0pFkRK8EIkhEpoCBhiuBoDRoFaTEkUhCoFQqyANgillMcPa4+MJzTRHs6ZOWee36fO+85M9pm3eWfW3muvtR7YO8Hgf5N0GLAQOBNY3Vb896PSjRe1zIALqKJZ3wE2doK3fJbvLkkHU4H0AmBVax03tx3b0Hme7XVdr5kqaRn1/39hAuuIGDTdVb0lLQY+CrxE7aFeLWmR7RttPyPpUup+eRDwINWSa2gWXERfysp1RIyKlrY8D1hCpYbPt31b1/nJVEGTf7Zzm3sy0D4k6SzgQuAI4OmuiutTgaXA0cDHbf9xyOvSR/ldJumz1LaGY22vbfuun6JaoJ1GTR6dA1zVPbEhaT6wg+2bejDsiIi+IGklVbDs28B91Pfal6ninCtsL27Pm9RaTE7s7L1ONluMBam8GxGjwuUX1ArfQ8AqSUsk7SNpNrUqeziVWrt5GxWZB9UU4MPAhtbXs9ND+Xng+8DuwL5DX5TAekRMAz5ke217/CvgSeD8FkzPpSY7DoK3K4vb/lkC64gYZG3r0ueAq4CbbP/V9o+Bc6ktTxdLuqHTz7q9bCtAAusYKxJcR8Sosv17Kp35ZmoV+0HgBio17Hzb9wzyius2JhXuBAx8r6uHcuf+vQNVMOtfozXGQbGNa/FLYLOkyyXdCuwGnGb7FUnvB+YDr1Jt5TLBERHxtonAZODPrahZZ6L4MaqOxcvAycDtkvZo53IPjTEle64jYtTZfq5Vvn4W+AIVIJ7ZUsAmUIHkwH2hDqkMPhPYmepT/TiwCjgWeE3SsvYZ7g58BpgEPNGjYY9L73AtDGyx/aSkW6g98G8BJ9p+StJ04AyqSvihLfsixeQiIt72DPA34BRJv273yYm237T9J0l/ac85EDiLmoCPGFOy5zoiekbSJOArwGW0dkW21w/yyjWApIuALwEfAbYAtwA/p1LnDqJWqX8DzKDalZyWyuAjQ9IiagJoBrAOuJbKurgeOA5YDzxGpeVPpbIvbs/+wIgYZO/0PS5pZ6pmxTHAlVRq+Evt3DSq28WlwHnAx4BZtv8xqgOPGKYE1xHRUy31tvNFuxcw2/aa3o6qdySdDNxItSh5GvgAcDbwCvB1YDpVRXUWcD9wt+3Vgz4hMRIknURdixXAJuAoYH9gpe1Fkk5ox/YB7gIesv1ArkVEDLIhLbd2BXYB3mgtCd8HrKYmim8DfgS8l9ou9gnbe0haSHVkOMT2xp78ERHbKcF1RPQFSXOArwGn2t7U6/GMpu72IpKuaIcvbS23dqBWsG+iVkYPs71B0ntsv9b1Hgno3gVDrsViavvU5bZfl7QntaIyn5rYON32y52qtl3vkWsREQOp+/4n6RLgSGAm8Dpwte0V7dxyahvNdOANKnvteNtr2tabGcAnbb/agz8jYrsluI6IviFpR9tbej2OXpA0BfgBsCPwgO1lQ84fCNwN/AT4BnX/TtrxCGjX4pr28F7b13V+MLaiZQuo7QwvAgtsP9yrsUZE9CNJS6giZSupwPqDVLeQ62yf3Z6zH9W+cCfg8bay/VVgGfBp2/f0ZPARw5CCZhHRNwY1sG7mUOnenb7JdIq72X7L9mOS1gN7Z1/1iJtD9WTfBXi0HZsgibZSvRR4gdobuFTSkVmpjogokg6mAukFwKpWuGxuO7ah8zzb67peM1XSMup7cGEC6xir0oorIqI/3AGcBPwOOFvS8ba3dlWsnkztu94saacejnMQ3AGcSBUqWyLp6FbN9s2WXbHV9krqh+L8tqKdvuwREWUaVYzzDy2w3pUqyvlT4ApJkyVd2NVSEtvPA/cB37R9zTu+a8QYkLTwiIg+IukAKiXucGpl9FZqBfUIYCktVS77ekfekGux0PbydnzoHutUaY+IaCSdA3zX9s7t8aPAq8Cxtl+R9CngEuAC27/N91mMJwmuIyL6TGtJciVwCrAReA5YAzxi+9r8EBk9Q67FCtvntuMJqCNi4G2j5daewCNU68KZwGzgKNtPtboVVwNTqGB782iPOWIkJbiOiOhDrV3JRVSP5Y3ANOMTJAAAAlNJREFUwbbf6N6H3dMBDpCWkn8xVc1+LdUeJp9/RAy07klGSTMBA1tsPyvph8AXgbeAE23fKWk6cAZV6OxQ209kojLGmwTXERF9StIkqir1ZVSbknm212flevRJ2pH6QTjL9ud7PZ6IiH4haRE1ETwDWEetWN8MXA8cB6ynaljsS7WUPN/27d39sCPGiwTXERF9rBXKOoZKTd4LmG17TW9HNZiG9G/Nj8KIGHiSTgJuBFYAm4CjgP2BlbYXSTqhHdsHuAt4yPYDmSSO8SrBdUTEGCBpDpWWfKrtTb0eT0REDCZJE22/2f69mGrte7nt19t+6/OA+cD9wOmtheHQQpAJrmNcSnAdETFGtDZQg9wLPCIi+oCkKUCnZda9tq/rBMytaNkCalvTi8AC2w/3aqwRoyl9riMixogE1hER0SfmAPOA44Hd2rEJbVX7Zap15LeoPdZL2xaniHEvK9cREREREfE/a8HyPCqA3g84wfad7dx/sqwkHUO1kXwhqeAxCBJcR0RERETE/03SAcAy4HBgoe3l7fjQPdZpuRUDIcF1RERERERsF0nTqI4WpwArbJ/bjiegjoGT4DoiIiIiIrabpMnAxVRXi7XAIQmsYxBN7PUAIiIiIiJi7LK9qbXl+jswK4F1DKqsXEdERERExLB1Fy2TNMH21l6PKWI0JbiOiIiIiIiIGKb0uY6IiIiIiIgYpgTXEREREREREcOU4DoiIiIiIiJimBJcR0RERERERAxTguuIiIiIiIiIYUpwHRERERERETFMCa4jIiIiIiIihunfD5IKA4KzPqcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "# identify which features contribute the most to the model and put into a dataframe\n",
    "feature_importances = pd.DataFrame(best_tuned_model.feature_importances_, index = feature_list,\n",
    "                      columns=['importance']).sort_values('importance', ascending=False)\n",
    "\n",
    "# plot the feature importances\n",
    "plt.figure(figsize=(16,6))\n",
    "ax = sns.barplot(x=feature_importances.index.values, y='importance' ,data=feature_importances)\n",
    "ax.set_xticklabels(ax.get_xticklabels(), rotation=50, ha=\"right\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Feature importances show how much each feature in the dataset contributes to the model. From this plot, we can see that the top five features are: Leucine resiudes, aliphatic-aliphatic residue pairs, aliphatic-aromatic residue pairs, aromatic-aliphatic residue pairs, and aspartic acid residues."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "\n",
    "Model that will be able to classify a protein sequence as transmembrane or not is able to predict with ~93% accuracy. The purpose of creating this model was to build a foundation and familiarize myself with machine learning protocols and now I can compare this model to already existing transmembrane prediction algorithms and then further improve this model to possibly classify specific TM helix regions and/or membrane protein interfaces."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### References:\n",
    "\n",
    "- Chen Z, Zhao P, Li F, Leier A, Marquez-Lago TT, Wang Y, Webb GI, Smith AI, Daly RJ* , Chou KC* , Song J* . iFeature: a python package and web server for features extraction and selection from protein and peptide sequences. Bioinformatics, 2018, doi: 10.1093/bioinformatics/bty140.\n",
    "- https://github.com/WillKoehrsen/Machine-Learning-Projects/blob/master/Random%20Forest%20Tutorial.ipynb\n",
    "- https://towardsdatascience.com/random-forest-in-python-24d0893d51c0\n",
    "- https://github.com/biof509/spring2020\n",
    "- https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html\n",
    "- https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}