in-depth-skewed-data-classif-93-recall-acc-now.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "a239913a-b27f-15e7-8911-53b62bf41c4e"
   },
   "source": [
    "# Credit card fraud detection\n",
    "\n",
    "#### This notebook will test different methods on skewed data. The idea is to compare if preprocessing techniques work better when there is an overwhelming majority class that can disrupt the efficiency of our predictive model.\n",
    "\n",
    "#### You will also be able to see how to apply cross validation for hyperparameter tuning on different classification models. My intention is to create models using:\n",
    "1. Logistic Regression\n",
    "2. SVMs\n",
    "3. Decision trees\n",
    "4. I also want to have a try at anomaly detection techniques, but I still have to investigate a bit on that, so any advise will be appreciated!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "029ecde6-086d-7a8e-de44-363a7a23dbd8"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "b4de5f93-d467-ad7d-4597-03d5f3e89f86"
   },
   "source": [
    "# Loading the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_cell_guid": "7e5ca1e3-3597-19d2-b4be-dffd335df630"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.225775</td>\n",
       "      <td>-0.638672</td>\n",
       "      <td>0.101288</td>\n",
       "      <td>-0.339846</td>\n",
       "      <td>0.167170</td>\n",
       "      <td>0.125895</td>\n",
       "      <td>-0.008983</td>\n",
       "      <td>0.014724</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>...</td>\n",
       "      <td>0.247998</td>\n",
       "      <td>0.771679</td>\n",
       "      <td>0.909412</td>\n",
       "      <td>-0.689281</td>\n",
       "      <td>-0.327642</td>\n",
       "      <td>-0.139097</td>\n",
       "      <td>-0.055353</td>\n",
       "      <td>-0.059752</td>\n",
       "      <td>378.66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108300</td>\n",
       "      <td>0.005274</td>\n",
       "      <td>-0.190321</td>\n",
       "      <td>-1.175575</td>\n",
       "      <td>0.647376</td>\n",
       "      <td>-0.221929</td>\n",
       "      <td>0.062723</td>\n",
       "      <td>0.061458</td>\n",
       "      <td>123.50</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.009431</td>\n",
       "      <td>0.798278</td>\n",
       "      <td>-0.137458</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>-0.206010</td>\n",
       "      <td>0.502292</td>\n",
       "      <td>0.219422</td>\n",
       "      <td>0.215153</td>\n",
       "      <td>69.99</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2   1.0 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3   1.0 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4   2.0 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9  ...         V21       V22       V23       V24  \\\n",
       "0  0.098698  0.363787  ...   -0.018307  0.277838 -0.110474  0.066928   \n",
       "1  0.085102 -0.255425  ...   -0.225775 -0.638672  0.101288 -0.339846   \n",
       "2  0.247676 -1.514654  ...    0.247998  0.771679  0.909412 -0.689281   \n",
       "3  0.377436 -1.387024  ...   -0.108300  0.005274 -0.190321 -1.175575   \n",
       "4 -0.270533  0.817739  ...   -0.009431  0.798278 -0.137458  0.141267   \n",
       "\n",
       "        V25       V26       V27       V28  Amount  Class  \n",
       "0  0.128539 -0.189115  0.133558 -0.021053  149.62      0  \n",
       "1  0.167170  0.125895 -0.008983  0.014724    2.69      0  \n",
       "2 -0.327642 -0.139097 -0.055353 -0.059752  378.66      0  \n",
       "3  0.647376 -0.221929  0.062723  0.061458  123.50      0  \n",
       "4 -0.206010  0.502292  0.219422  0.215153   69.99      0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv(\"../input/creditcard.csv\")\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "6268bbd8-6de5-2389-5693-ecd9a14872d4"
   },
   "source": [
    "# Checking the target classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "_cell_guid": "3f6e6674-12e9-6983-5788-5755f80c7ec2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7fc0e99cb550>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjQAAAGCCAYAAAD38Fn2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nFh6aXAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc0ebedf7f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "count_classes = pd.value_counts(data['Class'], sort = True).sort_index()\n",
    "count_classes.plot(kind = 'bar')\n",
    "plt.title(\"Fraud class histogram\")\n",
    "plt.xlabel(\"Class\")\n",
    "plt.ylabel(\"Frequency\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "801dd843-a90b-bb5f-97e7-2da84cc6cd6a"
   },
   "source": [
    "### Clearly the data is totally unbalanced!! \n",
    "\n",
    "#### This is a clear example where using a typical accuracy score to evaluate our classification algorithm. For example, if we just used a majority class to assign values to all records, we will still be having a high accuracy, BUT WE WOULD BE CLASSIFYING ALL \"1\" INCORRECTLY!!\n",
    "\n",
    "#### There are several ways to approach this classification problem taking into consideration this unbalance. \n",
    "\n",
    "- Collect more data? Nice strategy but not applicable in this case\n",
    "- Changing the performance metric:\n",
    "    - Use the confusio nmatrix to calculate Precision, Recall\n",
    "    - F1score (weighted average of precision recall)\n",
    "    - Use Kappa - which is a classification accuracy normalized by the imbalance of the classes in the data\n",
    "    - ROC curves - calculates sensitivity/specificity ratio.\n",
    "- Resampling the dataset\n",
    "    - Essentially this is a method that will process the data to have an approximate 50-50 ratio.\n",
    "    - One way to achieve this is by OVER-sampling, which is adding copies of the under-represented class (better when you have little data)\n",
    "    - Another is UNDER-sampling, which deletes instances from the over-represented class (better when he have lot's of data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "555427ea-a862-c64f-9d88-ddbb4f0366a3"
   },
   "source": [
    "# Approach\n",
    "\n",
    "1. We are not going to perform feature engineering in first instance. The dataset has been downgraded in order to contain 30 features (28 anonamised + time + amount). \n",
    "2. We will then compare what happens when using resampling and when not using it. We will test this approach using a simple logistic regression classifier.\n",
    "3. We will evaluate the models by using some of the performance metrics mentioned above.\n",
    "4. We will repeat the best resampling/not resampling method, by tuning the parameters in the logistic regression classifier.\n",
    "5. We will finally perform classifications model using other classification algorithms."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "fb559c9d-f4fe-5c63-7bdb-14cd3529f660"
   },
   "source": [
    "# Setting our input and target variables + resampling.\n",
    "\n",
    "#### 1. Normalising the amount column. The amount column is not in line with the anonimised features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "_cell_guid": "3fd30a6f-c0ad-5ece-943c-651cdf14d0d6"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>V10</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Class</th>\n",
       "      <th>normAmount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>0.090794</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>0</td>\n",
       "      <td>0.244964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>-0.166974</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.225775</td>\n",
       "      <td>-0.638672</td>\n",
       "      <td>0.101288</td>\n",
       "      <td>-0.339846</td>\n",
       "      <td>0.167170</td>\n",
       "      <td>0.125895</td>\n",
       "      <td>-0.008983</td>\n",
       "      <td>0.014724</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.342475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>0.207643</td>\n",
       "      <td>...</td>\n",
       "      <td>0.247998</td>\n",
       "      <td>0.771679</td>\n",
       "      <td>0.909412</td>\n",
       "      <td>-0.689281</td>\n",
       "      <td>-0.327642</td>\n",
       "      <td>-0.139097</td>\n",
       "      <td>-0.055353</td>\n",
       "      <td>-0.059752</td>\n",
       "      <td>0</td>\n",
       "      <td>1.160686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>-0.054952</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108300</td>\n",
       "      <td>0.005274</td>\n",
       "      <td>-0.190321</td>\n",
       "      <td>-1.175575</td>\n",
       "      <td>0.647376</td>\n",
       "      <td>-0.221929</td>\n",
       "      <td>0.062723</td>\n",
       "      <td>0.061458</td>\n",
       "      <td>0</td>\n",
       "      <td>0.140534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>0.753074</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.009431</td>\n",
       "      <td>0.798278</td>\n",
       "      <td>-0.137458</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>-0.206010</td>\n",
       "      <td>0.502292</td>\n",
       "      <td>0.219422</td>\n",
       "      <td>0.215153</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.073403</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9       V10     ...           V21       V22       V23  \\\n",
       "0  0.098698  0.363787  0.090794     ...     -0.018307  0.277838 -0.110474   \n",
       "1  0.085102 -0.255425 -0.166974     ...     -0.225775 -0.638672  0.101288   \n",
       "2  0.247676 -1.514654  0.207643     ...      0.247998  0.771679  0.909412   \n",
       "3  0.377436 -1.387024 -0.054952     ...     -0.108300  0.005274 -0.190321   \n",
       "4 -0.270533  0.817739  0.753074     ...     -0.009431  0.798278 -0.137458   \n",
       "\n",
       "        V24       V25       V26       V27       V28  Class  normAmount  \n",
       "0  0.066928  0.128539 -0.189115  0.133558 -0.021053      0    0.244964  \n",
       "1 -0.339846  0.167170  0.125895 -0.008983  0.014724      0   -0.342475  \n",
       "2 -0.689281 -0.327642 -0.139097 -0.055353 -0.059752      0    1.160686  \n",
       "3 -1.175575  0.647376 -0.221929  0.062723  0.061458      0    0.140534  \n",
       "4  0.141267 -0.206010  0.502292  0.219422  0.215153      0   -0.073403  \n",
       "\n",
       "[5 rows x 30 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "data['normAmount'] = StandardScaler().fit_transform(data['Amount'].reshape(-1, 1))\n",
    "data = data.drop(['Time','Amount'],axis=1)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "1e02844a-34ff-34ff-51f8-ef074ee2f1d5"
   },
   "source": [
    "#### 2. Assigning X and Y. No resampling."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "cfffc4c5-b621-250f-3b6b-6118cef52b9d"
   },
   "source": [
    "#### 3. Resampling.\n",
    "\n",
    "- As we mentioned earlier, there are several ways to resample skewed data. Apart from under and over sampling, there is a very popular approach called SMOTE (Synthetic Minority Over-Sampling Technique), which is a combination of oversampling and undersampling, but the oversampling approach is not by replicating minority class but constructing new minority class data instance via an algorithm.\n",
    "\n",
    "- In this notebook, we will use traditional UNDER-sampling. I will probably try to implement SMOTE in future versions of the code, but for now I will use traditional undersamplig.\n",
    "\n",
    "- The way we will under sample the dataset will be by creating a 50/50 ratio. This will be done by randomly selecting \"x\" amount of sample from the majority class, being \"x\" the total number of records with the minority class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "_cell_guid": "c1d874fa-5ea5-edbb-726c-ae98c84e6120"
   },
   "outputs": [],
   "source": [
    "X = data.ix[:, data.columns != 'Class']\n",
    "y = data.ix[:, data.columns == 'Class']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "_cell_guid": "2af7c203-44ed-66b6-6141-ac0d0637fcc6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Percentage of normal transactions:  0.5\n",
      "Percentage of fraud transactions:  0.5\n",
      "Total number of transactions in resampled data:  984\n"
     ]
    }
   ],
   "source": [
    "# Number of data points in the minority class\n",
    "number_records_fraud = len(data[data.Class == 1])\n",
    "fraud_indices = np.array(data[data.Class == 1].index)\n",
    "\n",
    "# Picking the indices of the normal classes\n",
    "normal_indices = data[data.Class == 0].index\n",
    "\n",
    "# Out of the indices we picked, randomly select \"x\" number (number_records_fraud)\n",
    "random_normal_indices = np.random.choice(normal_indices, number_records_fraud, replace = False)\n",
    "random_normal_indices = np.array(random_normal_indices)\n",
    "\n",
    "# Appending the 2 indices\n",
    "under_sample_indices = np.concatenate([fraud_indices,random_normal_indices])\n",
    "\n",
    "# Under sample dataset\n",
    "under_sample_data = data.iloc[under_sample_indices,:]\n",
    "\n",
    "X_undersample = under_sample_data.ix[:, under_sample_data.columns != 'Class']\n",
    "y_undersample = under_sample_data.ix[:, under_sample_data.columns == 'Class']\n",
    "\n",
    "# Showing ratio\n",
    "print(\"Percentage of normal transactions: \", len(under_sample_data[under_sample_data.Class == 0])/len(under_sample_data))\n",
    "print(\"Percentage of fraud transactions: \", len(under_sample_data[under_sample_data.Class == 1])/len(under_sample_data))\n",
    "print(\"Total number of transactions in resampled data: \", len(under_sample_data))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "6b74ba73-82a8-3585-b790-44fe486ff19d"
   },
   "source": [
    "# Splitting data into train and test set. Cross validation will be used when calculating accuracies."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "_cell_guid": "4a725b16-c14a-2be8-8240-617b7b2ed8cd"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.5/site-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number transactions train dataset:  199364\n",
      "Number transactions test dataset:  85443\n",
      "Total number of transactions:  284807\n",
      "\n",
      "Number transactions train dataset:  688\n",
      "Number transactions test dataset:  296\n",
      "Total number of transactions:  984\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "\n",
    "# Whole dataset\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,y,test_size = 0.3, random_state = 0)\n",
    "\n",
    "print(\"Number transactions train dataset: \", len(X_train))\n",
    "print(\"Number transactions test dataset: \", len(X_test))\n",
    "print(\"Total number of transactions: \", len(X_train)+len(X_test))\n",
    "\n",
    "# Undersampled dataset\n",
    "X_train_undersample, X_test_undersample, y_train_undersample, y_test_undersample = train_test_split(X_undersample\n",
    "                                                                                                   ,y_undersample\n",
    "                                                                                                   ,test_size = 0.3\n",
    "                                                                                                   ,random_state = 0)\n",
    "print(\"\")\n",
    "print(\"Number transactions train dataset: \", len(X_train_undersample))\n",
    "print(\"Number transactions test dataset: \", len(X_test_undersample))\n",
    "print(\"Total number of transactions: \", len(X_train_undersample)+len(X_test_undersample))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "6927cc69-57e6-4f34-4680-0b0016d414a0"
   },
   "source": [
    "# Logistic regression classifier - Undersampled data\n",
    "\n",
    "#### We are very interested in the recall score, because that is the metric that will help us try to capture the most fraudulent transactions. If you think how Accuracy, Precision and Recall work for a confusion matrix, recall would be the most interesting:\n",
    "\n",
    "- Accuracy = (TP+TN)/total\n",
    "- Precision = TP/(TP+FP)\n",
    "- Recall = TP/(TP+FN)\n",
    "\n",
    "#### As we know, due to the imbalacing of the data, many observations could be predicted as False Negatives, being, that we predict a normal transaction, but it is in fact a fraudulent one. Recall captures this.\n",
    "- Obviously, trying to increase recall, tends to come with a decrease of precision. However, in our case, if we predict that a transaction is fraudulent and turns out not to be, is not a massive problem compared to the opposite. \n",
    "- We could even apply a cost function when having FN and FP with different weights for each type of error, but let's leave that aside for now."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "_cell_guid": "9c7ec815-da54-993b-ef8d-b41b767cfacf"
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.cross_validation import KFold, cross_val_score\n",
    "from sklearn.metrics import confusion_matrix,precision_recall_curve,auc,roc_auc_score,roc_curve,recall_score,classification_report "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "88765ef8-cb56-860a-d249-9691d90afcb2"
   },
   "source": [
    "#### Very ad-hoc function to print K_fold_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "_cell_guid": "069bc837-cfd1-006e-c589-7085d5d29a8e"
   },
   "outputs": [],
   "source": [
    "def printing_Kfold_scores(x_train_data,y_train_data):\n",
    "    fold = KFold(len(y_train_data),5,shuffle=False) \n",
    "\n",
    "    # Different C parameters\n",
    "    c_param_range = [0.01,0.1,1,10,100]\n",
    "\n",
    "    results_table = pd.DataFrame(index = range(len(c_param_range),2), columns = ['C_parameter','Mean recall score'])\n",
    "    results_table['C_parameter'] = c_param_range\n",
    "\n",
    "    # the k-fold will give 2 lists: train_indices = indices[0], test_indices = indices[1]\n",
    "    j = 0\n",
    "    for c_param in c_param_range:\n",
    "        print('-------------------------------------------')\n",
    "        print('C parameter: ', c_param)\n",
    "        print('-------------------------------------------')\n",
    "        print('')\n",
    "\n",
    "        recall_accs = []\n",
    "        for iteration, indices in enumerate(fold,start=1):\n",
    "\n",
    "            # Call the logistic regression model with a certain C parameter\n",
    "            lr = LogisticRegression(C = c_param, penalty = 'l1')\n",
    "\n",
    "            # Use the training data to fit the model. In this case, we use the portion of the fold to train the model\n",
    "            # with indices[0]. We then predict on the portion assigned as the 'test cross validation' with indices[1]\n",
    "            lr.fit(x_train_data.iloc[indices[0],:],y_train_data.iloc[indices[0],:].values.ravel())\n",
    "\n",
    "            # Predict values using the test indices in the training data\n",
    "            y_pred_undersample = lr.predict(x_train_data.iloc[indices[1],:].values)\n",
    "\n",
    "            # Calculate the recall score and append it to a list for recall scores representing the current c_parameter\n",
    "            recall_acc = recall_score(y_train_data.iloc[indices[1],:].values,y_pred_undersample)\n",
    "            recall_accs.append(recall_acc)\n",
    "            print('Iteration ', iteration,': recall score = ', recall_acc)\n",
    "\n",
    "        # The mean value of those recall scores is the metric we want to save and get hold of.\n",
    "        results_table.ix[j,'Mean recall score'] = np.mean(recall_accs)\n",
    "        j += 1\n",
    "        print('')\n",
    "        print('Mean recall score ', np.mean(recall_accs))\n",
    "        print('')\n",
    "\n",
    "    best_c = results_table.loc[results_table['Mean recall score'].idxmax()]['C_parameter']\n",
    "    \n",
    "    # Finally, we can check which C parameter is the best amongst the chosen.\n",
    "    print('*********************************************************************************')\n",
    "    print('Best model to choose from cross validation is with C parameter = ', best_c)\n",
    "    print('*********************************************************************************')\n",
    "    \n",
    "    return best_c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "_cell_guid": "983c1c75-8092-9a8e-40ca-754fde9e2301"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-------------------------------------------\n",
      "C parameter:  0.01\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.931506849315\n",
      "Iteration  2 : recall score =  0.917808219178\n",
      "Iteration  3 : recall score =  0.983050847458\n",
      "Iteration  4 : recall score =  0.959459459459\n",
      "Iteration  5 : recall score =  0.954545454545\n",
      "\n",
      "Mean recall score  0.949274165991\n",
      "\n",
      "-------------------------------------------\n",
      "C parameter:  0.1\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.849315068493\n",
      "Iteration  2 : recall score =  0.86301369863\n",
      "Iteration  3 : recall score =  0.932203389831\n",
      "Iteration  4 : recall score =  0.945945945946\n",
      "Iteration  5 : recall score =  0.878787878788\n",
      "\n",
      "Mean recall score  0.893853196338\n",
      "\n",
      "-------------------------------------------\n",
      "C parameter:  1\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.849315068493\n",
      "Iteration  2 : recall score =  0.876712328767\n",
      "Iteration  3 : recall score =  0.983050847458\n",
      "Iteration  4 : recall score =  0.945945945946\n",
      "Iteration  5 : recall score =  0.909090909091\n",
      "\n",
      "Mean recall score  0.912823019951\n",
      "\n",
      "-------------------------------------------\n",
      "C parameter:  10\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.849315068493\n",
      "Iteration  2 : recall score =  0.876712328767\n",
      "Iteration  3 : recall score =  0.983050847458\n",
      "Iteration  4 : recall score =  0.945945945946\n",
      "Iteration  5 : recall score =  0.924242424242\n",
      "\n",
      "Mean recall score  0.915853322981\n",
      "\n",
      "-------------------------------------------\n",
      "C parameter:  100\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.849315068493\n",
      "Iteration  2 : recall score =  0.876712328767\n",
      "Iteration  3 : recall score =  0.983050847458\n",
      "Iteration  4 : recall score =  0.945945945946\n",
      "Iteration  5 : recall score =  0.924242424242\n",
      "\n",
      "Mean recall score  0.915853322981\n",
      "\n",
      "*********************************************************************************\n",
      "Best model to choose from cross validation is with C parameter =  0.01\n",
      "*********************************************************************************\n"
     ]
    }
   ],
   "source": [
    "best_c = printing_Kfold_scores(X_train_undersample,y_train_undersample)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "bdb1ff38-9d27-ae15-58df-9be4e4f76550"
   },
   "source": [
    "#### Create a function to plot a fancy confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "_cell_guid": "f5b049b3-4f9a-f5bb-db3d-6c48e9b1e1a5"
   },
   "outputs": [],
   "source": [
    "import itertools\n",
    "\n",
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=0)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        #print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        1#print('Confusion matrix, without normalization')\n",
    "\n",
    "    #print(cm)\n",
    "\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, cm[i, j],\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "e40b554a-5b88-2828-f655-d54560ad7480"
   },
   "source": [
    "### Predictions on test set and plotting confusion matrix\n",
    "\n",
    "#### We have been talking about using the recall metric as our proxy of how effective our predictive model is. Even though recall is still the recall we want to calculate, just bear mind in mind that the undersampled data hasn't got a skewness towards a certain class, which doesn't make recall metric as critical. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "_cell_guid": "5c8e4c0e-8cfd-7422-04a8-1b47b8531267"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall metric in the testing dataset:  0.925170068027\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAc4AAAGbCAYAAABNpXD0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n8P8BJmK3WYJFNukAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc0cb768b70>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use this C_parameter to build the final model with the whole training dataset and predict the classes in the test\n",
    "# dataset\n",
    "lr = LogisticRegression(C = best_c, penalty = 'l1')\n",
    "lr.fit(X_train_undersample,y_train_undersample.values.ravel())\n",
    "y_pred_undersample = lr.predict(X_test_undersample.values)\n",
    "\n",
    "# Compute confusion matrix\n",
    "cnf_matrix = confusion_matrix(y_test_undersample,y_pred_undersample)\n",
    "np.set_printoptions(precision=2)\n",
    "\n",
    "print(\"Recall metric in the testing dataset: \", cnf_matrix[1,1]/(cnf_matrix[1,0]+cnf_matrix[1,1]))\n",
    "\n",
    "# Plot non-normalized confusion matrix\n",
    "class_names = [0,1]\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix\n",
    "                      , classes=class_names\n",
    "                      , title='Confusion matrix')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "3aee694e-c838-434a-0cb6-dafaa3a0b224"
   },
   "source": [
    "#### So, the model is offering an 93.2% recall accuracy on the generalised unseen data (test set). Not a bad percentage to be the first try. However, recall this is a 93.2% recall accuracy measure on the undersampled test set.\n",
    "\n",
    "### Being happy with this result, let's apply the model we fitted and test it on the whole data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "_cell_guid": "2fac80a6-cc45-49e8-3fd6-2322e2461955"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall metric in the testing dataset:  0.91156462585\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAGbCAYAAAAcMUOXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmVkBDmAzM7MCHMBmZmYFOIDNzMwK+A8IryKVjXjPLQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc0cb7793c8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use this C_parameter to build the final model with the whole training dataset and predict the classes in the test\n",
    "# dataset\n",
    "lr = LogisticRegression(C = best_c, penalty = 'l1')\n",
    "lr.fit(X_train_undersample,y_train_undersample.values.ravel())\n",
    "y_pred = lr.predict(X_test.values)\n",
    "\n",
    "# Compute confusion matrix\n",
    "cnf_matrix = confusion_matrix(y_test,y_pred)\n",
    "np.set_printoptions(precision=2)\n",
    "\n",
    "print(\"Recall metric in the testing dataset: \", cnf_matrix[1,1]/(cnf_matrix[1,0]+cnf_matrix[1,1]))\n",
    "\n",
    "# Plot non-normalized confusion matrix\n",
    "class_names = [0,1]\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix\n",
    "                      , classes=class_names\n",
    "                      , title='Confusion matrix')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "e90966bc-2fb9-8442-4705-340d15a79b66"
   },
   "source": [
    "### Still a very decent recall accuracy when applying it to a much larger and skewed dataset! \n",
    "\n",
    "#### We can start to be happy with how initial approach is working."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "29bf4990-1b01-52fc-6c37-d4414b0aafa7"
   },
   "source": [
    "### Plotting ROC curve and Precision-Recall curve.\n",
    "\n",
    "- I find precision-recall curve much more convenient in this case as our problems relies on the \"positive\" class being more interesting than the negative class, but as we have calculated the recall precision, I am not going to plot the precision recall curves yet.\n",
    "\n",
    "- AUC and ROC curve are also interesting to check if the model is also predicting as a whole correctly and not making many errors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "_cell_guid": "4a0d3339-cb16-5899-0b0e-f86636284a63"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiAAAAGHCAYAAACJeOnXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nM7PSOQAxMzOz0v0f7iPicAffP/cAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc0cb7680b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ROC CURVE\n",
    "lr = LogisticRegression(C = best_c, penalty = 'l1')\n",
    "y_pred_undersample_score = lr.fit(X_train_undersample,y_train_undersample.values.ravel()).decision_function(X_test_undersample.values)\n",
    "\n",
    "fpr, tpr, thresholds = roc_curve(y_test_undersample.values.ravel(),y_pred_undersample_score)\n",
    "roc_auc = auc(fpr,tpr)\n",
    "\n",
    "# Plot ROC\n",
    "plt.title('Receiver Operating Characteristic')\n",
    "plt.plot(fpr, tpr, 'b',label='AUC = %0.2f'% roc_auc)\n",
    "plt.legend(loc='lower right')\n",
    "plt.plot([0,1],[0,1],'r--')\n",
    "plt.xlim([-0.1,1.0])\n",
    "plt.ylim([-0.1,1.01])\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "1f1e40a4-bca8-87ad-f368-d0b50e7d6158"
   },
   "source": [
    "#### An additional comment that would be interesting to do is to initialise multiple undersampled datasets and repeat the process in loop. Remember that, to create an undersample data, we randomly got records from the majority class. Even though this is a valid technique, is doesn't represent the real population, so it would be interesting to repeat the process with different undersample configurations and check if the previous chosen parameters are still the most effective. In the end, the idea is to use a wider random representation of the whole dataset and rely on the averaged best parameters."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "2e0e6bfc-37ac-2d2e-61af-7118619fdf27"
   },
   "source": [
    "# Logistic regression classifier - Skewed data\n",
    "\n",
    "#### Having tested our previous approach, I find really interesting to test the same process on the skewed data. Our intuition is that skewness will introduce issues difficult to capture, and therefore, provide a less effective algorithm.\n",
    "- To be fair, taking into account the fact that the train and test datasets are substantially bigger than the undersampled ones, I believe a K-fold cross validation is necessary. I guess that by splitting the data with 60% in training set, 20% cross validation and 20% test should be enough... but let's take the same approach as before (no harm on this, it's just that K-fold is computationally more expensive)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "_cell_guid": "2aaf245f-43cd-d543-b857-562fb696fc4e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-------------------------------------------\n",
      "C parameter:  0.01\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.492537313433\n",
      "Iteration  2 : recall score =  0.602739726027\n",
      "Iteration  3 : recall score =  0.683333333333\n",
      "Iteration  4 : recall score =  0.569230769231\n",
      "Iteration  5 : recall score =  0.45\n",
      "\n",
      "Mean recall score  0.559568228405\n",
      "\n",
      "-------------------------------------------\n",
      "C parameter:  0.1\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.567164179104\n",
      "Iteration  2 : recall score =  0.616438356164\n",
      "Iteration  3 : recall score =  0.683333333333\n",
      "Iteration  4 : recall score =  0.584615384615\n",
      "Iteration  5 : recall score =  0.525\n",
      "\n",
      "Mean recall score  0.595310250644\n",
      "\n",
      "-------------------------------------------\n",
      "C parameter:  1\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.55223880597\n",
      "Iteration  2 : recall score =  0.616438356164\n",
      "Iteration  3 : recall score =  0.716666666667\n",
      "Iteration  4 : recall score =  0.615384615385\n",
      "Iteration  5 : recall score =  0.5625\n",
      "\n",
      "Mean recall score  0.612645688837\n",
      "\n",
      "-------------------------------------------\n",
      "C parameter:  10\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.55223880597\n",
      "Iteration  2 : recall score =  0.616438356164\n",
      "Iteration  3 : recall score =  0.733333333333\n",
      "Iteration  4 : recall score =  0.615384615385\n",
      "Iteration  5 : recall score =  0.575\n",
      "\n",
      "Mean recall score  0.61847902217\n",
      "\n",
      "-------------------------------------------\n",
      "C parameter:  100\n",
      "-------------------------------------------\n",
      "\n",
      "Iteration  1 : recall score =  0.55223880597\n",
      "Iteration  2 : recall score =  0.616438356164\n",
      "Iteration  3 : recall score =  0.733333333333\n",
      "Iteration  4 : recall score =  0.615384615385\n",
      "Iteration  5 : recall score =  0.575\n",
      "\n",
      "Mean recall score  0.61847902217\n",
      "\n",
      "*********************************************************************************\n",
      "Best model to choose from cross validation is with C parameter =  10.0\n",
      "*********************************************************************************\n"
     ]
    }
   ],
   "source": [
    "best_c = printing_Kfold_scores(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "_cell_guid": "634c1907-a5c5-888c-c2e9-da73f81ee445"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall metric in the testing dataset:  0.619047619048\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeAAAAGbCAYAAAAcMUOXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc0cb768400>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use this C_parameter to build the final model with the whole training dataset and predict the classes in the test\n",
    "# dataset\n",
    "lr = LogisticRegression(C = best_c, penalty = 'l1')\n",
    "lr.fit(X_train,y_train.values.ravel())\n",
    "y_pred_undersample = lr.predict(X_test.values)\n",
    "\n",
    "# Compute confusion matrix\n",
    "cnf_matrix = confusion_matrix(y_test,y_pred_undersample)\n",
    "np.set_printoptions(precision=2)\n",
    "\n",
    "print(\"Recall metric in the testing dataset: \", cnf_matrix[1,1]/(cnf_matrix[1,0]+cnf_matrix[1,1]))\n",
    "\n",
    "# Plot non-normalized confusion matrix\n",
    "class_names = [0,1]\n",
    "plt.figure()\n",
    "plot_confusion_matrix(cnf_matrix\n",
    "                      , classes=class_names\n",
    "                      , title='Confusion matrix')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "f409ba42-8534-e960-0869-00b1e4e73d65"
   },
   "source": [
    "# Before continuing... changing classification threshold.\n",
    "\n",
    "#### We have seen that by undersampling the data, our algorithm does a much better job at detecting fraud. I wanted also to show how can we tweak our final classification by changing the thresold.\n",
    "- Initially, you build the classification model and then you predict unseen data using it.\n",
    "- We previously used the \"predict()\" method to decided whether a record should belong to \"1\" or \"0\". \n",
    "- There is another method \"predict_proba()\".\n",
    "    - This method returns the probabilities for each class. The idea is that by changing the threshold to assign a record to class 1, we can control precision and recall.\n",
    "    \n",
    "#### Let's check this using the undersampled data (best C_param = 0.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "_cell_guid": "a1275e7d-48fc-e0bd-c629-6ddf698b47b5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Recall metric in the testing dataset:  1.0\n",
      "Recall metric in the testing dataset:  1.0\n",
      "Recall metric in the testing dataset:  1.0\n",
      "Recall metric in the testing dataset:  0.972789115646\n",
      "Recall metric in the testing dataset:  0.918367346939\n",
      "Recall metric in the testing dataset:  0.877551020408\n",
      "Recall metric in the testing dataset:  0.816326530612\n",
      "Recall metric in the testing dataset:  0.768707482993\n",
      "Recall metric in the testing dataset:  0.571428571429\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA94AAAPeCAYAAAD6bcIrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAIABJREFUeJzs3XmYHVWd//H3tyO7IyBhEQRlkcWRLSyKC7uioogoAoMg\ng4oK/FyQwR0RHQVERBwZdFBBRRF0RpAtsokLqyyyCAEh7CQQCFsCJOmc3x91O3R3qu9at++t6vfL\npx7TdU/V/d5Af7jn1KlTkVJCkiRJkiR1x0CvC5AkSZIkqcrseEuSJEmS1EV2vCVJkiRJ6iI73pIk\nSZIkdZEdb0mSJEmSusiOtyRJkiRJXWTHW5IkSZKkLrLjLUmSJElSF72k1wVIqo6IWAuY3Obhs1JK\n9xdZjyT1EzNSkvJNhHyMlFKva5BUARGxFgMvuY+FC9o9xVxgozIEpyS1yoyUpHwTJR+94i2pKJNZ\nuIAl1tqZWPrlLR2Ynn+C+fdfsizZSGdfh6YktcmMlKR8EyIf7XhLKlQssxIDy67S0jELI7pUjST1\nFzNSkvJVPR/teEsqVkS2tXqMJE0EZqQk5at4PtrxllSsGMi2Vo+RpInAjJSkfBXPRzvekopXotFH\nSRp3ZqQk5atwPtrxllSsio9WSlJHzEhJylfxfCxPpZIkSZIklZBXvCUVq+ILY0hSR8xIScpX8Xy0\n4y2pWBFtTBMqT2hKUkfMSEnKV/F8tOMtqVgVH62UpI6YkZKUr+L5aMdbUsHaWBjD5SYkTRhmpCTl\nq3Y+2vGWVKyKj1ZKUkfMSEnKV/F8tOMtqVgVvz9HkjpiRkpSvornY3muzUuSJEmSVEJe8ZZUrIpP\nE5KkjpiRkpSv4vlox1tSsaKNhTFaXkhDkkrKjJSkfBXPRzvekgrWxv05lGe0UpI6Y0ZKUr5q56Md\nb0nFGohsa/UYSZoIzEhJylfxfCzPtXlJ5TA0TajVTZImAjNSkvJ1IR8j4i0RcW5EPBQRCyNit1Gv\nfzUibo+IZyPiiYi4OCK2HtXmj7Vjh7bBiDi51Y9nkksqVvDi4hhNb70uWpLGiRkpSfm6k4/LATcB\nBwMp5/VpwCHA64A3AfcCf4iIlYa1ScCPgFWB1YBXAEe0+vHseI+jiNiuNkqyR69rGRIR90bEuQWe\n71W1z7h/E21Pi4jpRb23pHIzIxdra0ZKAszHnLbmo5qSUroopXRkSukccrrpKaUzU0qXpZTuTSnd\nDhwGvAzYZFTTuSmlx1JKj9a2Z1utxY53h0ZNOxhrG4yIbWuH5I209FIv60k9fv+WRMTqEXFWRMyO\niKci4ncRsXaTx24VESdHxN8iYl5EDHa73t5pZ4pQeaYJqTVmZMfv3W9/H2NqNyMjc0BEnBMR99d+\nj2+JiC9FxFLjUfv4MiOVMR87fu9++/sYU4ffIT9S+/2cERHPR8Q9EfGTiHhVt+sef8XnYysiYgng\nY8CTwN9HvbxvRDxW++/TNyNimVbP7+JqnfvgqJ8/BOxc2z98VOV24LWj9qkkImI54I/AvwDfABaQ\njYj9MSI2SynNbnCKdwIHAjcDdwPrd6/aHuvOMxiHpgn9GPjfnNeHpgndAyxD9s/mDxGxbkrp8Vqb\noWlCX+HF38O5rRWqNpiRE0CHGbks8BPgKuC/gUeBbYCvATsCO3Wv8h4wI/Ui83ECKOA75OZkv7vn\nALOBtYGDgF0jYtOU0oxu1T7uevQc74jYFTiT7L9HDwNvTSk9MazJGcB9tdc2AY4j+y7//lbex453\nh1JKvxz+c0RsA+ycUvrV6LbR4b8YEbFMSum5jk5SQhGxAXB3SmlBD8s4BFgX2CqldEOtrouAW4HP\nAl9ucPzJwDEppRci4vtUvuPd6jMY6/9upJQuAi7Kmi7eOKV05sjTxWHAh8nC8fJhL81NKT3WWnHq\nhBnZfRXIyHnAG1NKVw/b9+OIuA84KiJ2TCld1qW6x58ZqRrzsfsqkI+klA4ZvS8izgH+BuxP1gms\nhi7kY5MuAzYFJgMfBc6OiK1TSrMAUkqnDmt7W0Q8AlwaEWunlJq+5cGp5uMvAQO1KXQPRMRzEXFJ\nRKw7vFFtSsnNETElIv4UEXOA/xz2+jtq+5+NiKcj4ryIeO2oc6waET+tvc/zEfFwbWrLWqOLiog3\nRcQ1tXrujoj9ctqsHRFnR8TjETEnIq6KiHc286EjYveIuLV2/psjYvdm/8KAzwMPRcS3I2LDFo4r\n0vuA64YCEyClNA24FPhAo4Nr94S80MX6+kfLi2K0MbpZ9+27O01IXWdGTrCMTCnNH9XpHvJ/ZFf4\nNiqy0J4zI9U+83GC5WMd99X+f4UOa+svDXJwcMZNzLvptBHb/Dt/3/HbppSeSyndk1K6NqX0UbJZ\nCR+uc8i1ZP99Wq+V9/GK9/gL4AvAIPBtYHngc8AvyKbWDUlkoy4XkE19+BkwE6AWaKeRjW4fQTYt\n4hPAnyNi85TS/bVz/C/ZF5aTyH5BVwHeCqwFDLUBeA1wNtkUtdPIpkT/NCL+VltkgIhYhWwa4NLA\n94AnyKZEnRsR76stWJD/gSPeBvyGbGTv88BKwE+BB5v6G8vqD+DjwGcj4sparWellObUed/lavU2\nMj+l9HSd8wTZVYEf57x8LfDWiFiuXi0TSjuPvingUTkxTtOE1HVmpBk55BW1/5/V4nH9zYxU+8zH\nCZyPEfFyYBLwKuBIsn/OlzZRY3k0yMdJq2/BpNW3GLFv4dMPMu+q7xZdyQBQb42Rzcn+/h9p5aR2\nvHtjKWDTlNIgQEQ8CZwYEa9NKf1jWLtVgY8Nn95QC4LvAT9KKX1i2P7TgTuBLwIfj4jlyUL48JTS\nCcPOeWxOPesDb0kpXVk719nAA8C/8+JS+V8AVgbenFK6qtbuVLJ7lk8gu+9kLMcCM2rHPls79grg\nYrIl++tKKd0IHBARhwJ7k4X6qcD3IuLXwI/HuFryX2TB3sgfye4jHMvLyf6Z5f1yDe1bHbirifdS\n94zLNCGNCzPSjITs7/Yp4MIWj1M+M7IazMeJm48P8WJncBbwyZRStTreXVD79349WLRGwjoRsSnZ\nANDjwJeAc8n+eUwGDiX7Z3J27fh1gH8jG8h6nCxHTwCuSCnd2kotdrx74ydDgVnzZ7J/GdYBhofm\nC2Sjh8O9lWyE88xY/Ply1wA71H5+juy+ue0j4icppSfr1POPocAESCnNiohptXqGvAO4digwa+3m\nRMSPgG/mBD4AEbEa2b+g3xy+7H5K6dKI+AfZSGtTasefCpwa2T07BwL7AR+uneu7KaXhI4rHAj9v\n4tSNFrUYmmqXN1X8+VFtRP1pkYMP38DgIzeO2JcWdH7bWe3etXtq27URcSfZNKG8LwowcpqQXyr7\nixk5wTMyIr5I9mX2E/WuJpWTGamOmI8TNx/fTnYVfiOyBfiWa/K4Emnn1pqG7bckW8tiaCX879T2\nn04222NDsnvlJ5N1rK8jG+i5vdZuHtmih58i+zt/gKxTvuj2jWbZ8e6NB0b9PPRLu+Ko/Q/lLAbx\nGrJ/wy5ncQl4GiClNC8iPgccD8yMiKuB84CfpZRmjjrufhY3e1Q9rwLyRgRvH/b6YqFZ2w/wz5zX\nppFN1WhZ7d6Yz0XED8mC8Q1ki1f8eFibO4A72jn/KEPfePKmnCw9qo0aLIwxaY0tmbTGliP2LXzq\nAeZd+Z0xjmhbV6YJaVyYkZkJmZERsRfwdeDUlNKPOqyt/5iR6oz5mJlw+ZhSuqL2x6mRPT/91oh4\nNqVUncf+dWfxySuov67Z+xoc/yCwfWtF5bPj3RtjPcN59L85eb+IA2Th+EFq9+uMsihkU0rfq/1i\n7g7sAhwNfCEidkgpDV9Qpdl6+kJkz3V9H9k0ph3J/p5+Dpwyqt3LaG4UcV6q/yiHJ8hGKl+R89rQ\nvoebeJ+JoQuPguinaUIaF2ZkB8qckRHxVrKrEL8nuxJRPWakOmM+dqDM+ThcSumeiLgR2JfsyTnV\n0KPHiY0XO97lczdZmD2Wmni8Su2+rO8C341s1cu/kz26YP8W3/c+YIOc/RsNe32s4yAbZR0t73xj\nioityYJyb+BlwI3AwcCvxpiK+D0KuD8npZQi4hayqSqjvR64x4XVhunOwkF9M01Ifc+MLGlGRsTr\nyRZ0uhbYK6W0sInayseMVO+YjyXNxzEsAyzZ5rH9qUeLT44XO97lM5VsKtAXI+KPo6cRRcTk2v01\nywAL08hHWE0HnqH+1LKxXAB8KiJen1K6pvZeywEHAdPz7s0BSCnNiIibgA9FxDEppWdqx74VeC1N\nLIwREXsAXwP+lWz60s/JpiDe3ODQou7PgWxFzW9FxJT04jMYNyAL2xHPT6xdOSCldE8T562eik8T\nUt8zI0uYkRGxEdlU1nuAd6cqP37RjFTvmI8ly8eImAT8y+j77GsDCRuTrWhfHb17jve4sONdMiml\nZyLiE2SPhrghIs4EHiN7vMOuwF+AT5KtMnlpRJxFdt/MAmAPssdB/KqNtz4G2Ae4KCJOIps6cwDZ\n/Td7NDj2C2RfqP4aET8hexTEoWSPhnhpE++9K9ln/CDw22a/kBV4fw5k03g+ClwQEceT/X1+hmza\n3gmj2l4GLGTYwiKRPfdy6LmWW9b2fan2830ppQoFZ1cWxpCaYkaWLyMj4qVkHYIVyL6EvitGZsjd\nKX/V4ZIyI9Ub5mP58pHsMz4Q2QrstwFzyB5PdgBZp/8bBdXYJ6qdj3a8uyO18Vre/ty2KaVfRcRD\nZM8zPJxs9PEhspUtf1pr9gDwS2AnsrBZQBYge6aUfjfqPRrWlFJ6NCK2IRsBPJRsQYibgXellC6q\nV3dKaWpE7EkWDt8km+p0ANl9Q9uO8d7DHVpbjbVnUkrPRsR2ZFOuvkR2ZeFy4LCU0uOjm7P43+na\nZIsFDd9/dO3/r6BqI5ZSfWbkyHonekauBKxR+/MxOac/nfyFmaQqMh9H1jvR83Eu8D9kK86/j2x6\n+cPAGcB/phefu64SiJTq/X5LUnMiYgpw/ZLbfpGBFdZq6diFT97PvD99E2CLoWlYklQlZqQk5Zso\n+egVb0nFqviKlJLUETNSkvJVPB/teEsqVsUXxpCkjpiRkpSv4vlox1tSsSo+WilJHTEjJSlfxfPR\njrekQgVBtBiCUaIVKSWpE2akJOWrej6W54njkiRJkiSVUOmueEfESsAuwL3A872tRqqspYFXA1Nz\nHnVRV0Qbo5UlmibU78xIqevazkcwI3vJfJS6znyso3Qdb7LAPKPXRUgTxL5kz/JsXtS2Vo9RUcxI\naXy0no9gRvaW+SiND/MxRxk73vcC/PT0X7DBhhv1uJTiHXH4Zzju+O/2uozCbb/fsb0uoSvmP/QX\nlljjzb0uo3ALn5/NgvsvhtrvW0uijdHHEoVmCdwL1czIquYjmJFl0lE+ghnZW/cC/M9Pf84GG2zY\n41KK9fkjDuOY407odRldsfOHq/m55j/wJ5ZYc9tel1Gohc89wYJ7p4L5mKuMHe/nATbYcCM2nzKl\n17UUbvnll6/k5xpYduVel9AVMWnJyn62mpan4lV9mlAJVDYjq5qPYEaWVFtTlc3InsrycYMN2Wzz\namXJ8ssvX7nPNGRg2VV6XUJXZPlYzc+G+ZirjB1vSX2s6itSSlInzEhJylf1fLTjLalQVR+tlKRO\nmJGSlK/q+ejjxCRJkiRJ6iKvePeZPffap9clqAWTVnxNr0voPxVfkVK9Yz6WjxmZw4xUF7x/z717\nXYJaNOnlG/S6hP5T8Xz0inef2Wtvv1iWyaQV1+91CX1naJpQq5vUiPlYPmbk4sxIdYMDk+Vjx3tx\n3cjHiHhLRJwbEQ9FxMKI2G3Yay+JiGMj4uaIeLbW5vSIeMWocywVET+IiFkR8UxE/CYiWl4Zz463\npGJF68FZptFKSeqIGSlJ+bqTj8sBNwEHA2nUa8sCmwFfAzYH3gtsAJwzqt2JwK7A+4BtgdWB37b6\n8ZxqLqlQVV+RUpI6YUZKUr5u5GNK6SLgIoAYdfKU0tPALiPOF3EocE1EvDKl9GBEvAw4ENg7pXRF\nrc2/A7dHxNYppWubrdWOt6RCVX1FSknqhBkpSfn6JB9XILsy/mTt5y3I+syXDjVIKU2LiPuBbYCm\nO95ONZdUrGhzq3fKPro/R5I6YkZKUr4u5GNLbx+xFHAM8MuU0rO13asB82pXx4ebWXutaXa8JZVB\n39yfI0l9yIyUpA5ExEuAs8ky9OBuvIdTzSUVqhvThPrp/hxJ6oQZKUn5GuXjC/f8lRemXzliX5o/\nt4j3Hep0rwnsOOxqN8AMYMmIeNmoq96r1l5rmh1vSYWq+v05ktQJM1KS8jXKx6XXfTNLr/vmEfsW\nPD6dJ3//hU7ec6jTvQ6wQ0pp9qgm1wMLgJ2A/6sdswGwFnBVK+9lx1tS4Xq5EFC378+RpE6ZkZKU\nr+h8jIjlgPV48W7wdSJiU+AJ4BGyW2o2A94FLBERq9baPZFSmp9SejoifgycEBGzgWeAk4C/tjob\nyI63pGK1s9BFQRk7HvfnSFJHzEhJytedfNwSuJws9xLwndr+08nWvnh3bf9Nw86YgB2AP9X2fQYY\nBH4DLEV2a88hLVZqx1tSsRpNE3r+7r/w/D1/HbEvzSvP/TmS1AkzUpLydWkNjCuov6B4w8XGU0ov\nAP+vtrXNjrekcZV3f878Wfcw+5zPt33O8bw/R5K6yYyUpGqy4y2pUN0Yreyn+3MkqRNmpCTl65PF\nJ7vGjrekQnUpNPvm/hxJ6oQZKUn57HhLUguqfn+OJHXCjJSkfHa8JalV5clASRp/ZqQk5atwPtrx\nllSsaGP0scIhK0kjmJGSlK/i+WjHW1Khqj5NSJI6YUZKUr6q52PDe34kSZIkSVL7vOItqVBVH62U\npE6YkZKUr+r5aMdbUrGC1u+3KU9mSlJnzEhJylfxfLTjLalQQRujlWVKTUnqgBkpSfmqno92vCUV\nqurThCSpE2akJOWrej7a8ZZUrIo/CkKSOmJGSlK+iuejq5pLkiRJktRFXvGWVKiqTxOSpE6YkZKU\nr+r5aMdbUrEqviKlJHXEjJSkfBXPRzvekgpV9RUpJakTZqQk5at6PtrxllSoqk8TkqROmJGSlK/q\n+WjHW1KhIrKt1WMkaSIwIyUpX9Xz0Y63pGK1MVpZqtSUpE6YkZKUr+L56OPEJEmSJEmVExFviYhz\nI+KhiFgYEbvltDk6Ih6OiLkRcXFErDfq9T/Wjh3aBiPi5FZrseMtqVBD04Ra3SRpIjAjJSlfl/Jx\nOeAm4GAgLf6e8TngUOAgYGtgDjA1IpYc1iwBPwJWBVYDXgEc0ernc6q5pEIFrS904XdKSROFGSlJ\n+bqRjymli4CLyM6d1/xTwNdTSufV2uwPzAR2B84a1m5uSumxloobxSvekgrVjdHKfpomJEmdMCMl\nKd94zwiKiLXJrmBfOrQvpfQ0cA2wzajm+0bEYxFxS0R8MyKWafX97HhLKlQMBAMtbjHQMDX7ZpqQ\nJHXCjJSkfF3Kx3pWI8u+maP2z6y9NuQM4IPA9sA3gf2An7f6Zk41l1SobjwKop+mCUlSJ8xIScrX\nr48TSymdOuzH2yLiEeDSiFg7pTS92fPY8ZZUqKD1R0FEB3cwjjVNKCKGpgkN/1K5b0TsB8wAfk/2\nRfS5tt9cklpkRkpSvkb5+NRtl/PUbZeP2Df4/JxO3nIG2W3iqzLyqveqwI11jru2dtx6QNMdb6ea\n94lTTv4BG75mbVb8l2XY9k1v4G/XXdfrkiasN22+Lmef+DHunvoN5lx/Ertut/GYbU/60t7Muf4k\nDt5nuxH7X73GSpx5/Ee479JvMeNP3+Znx/w7K6/40m6XPlGN6zQhjT/zsX+Yj6VkRk4QJ3z7WJZf\n9iV84YjP9rqUCelNm63D2Sd8hLsvOIo5157Artv+65htT/rCnsy59gQO3mvbEfun/vBQ5lx7wqLt\n2WtO4MTPvb/bpZfG8v+6A2t94OgR22pv/Xjb56tdrZ4B7DS0LyJeBrweuLLOoZuT5eojrbyfHe8+\ncPZZv+bzR3yWrxz5Na6+7kY22WRTdtt1F2bNmtXr0iakZZdZkpunPcinvvVr0mJ3yr1otx02YavX\nvYqHH31qxP5lll6C804+lIUpsctHv8cOB3yHpZZ4Cb/9XvvBUCb9+qiclNKpKaWLU0q3pZR+Rfal\n8r21q0HqU+ZjfzEfO2dGqhuu/9t1nPaT/2HjTTbtdSkTVpaPD/GpY39TPx+33zg3HwFSSvz4/67i\nVW/7Cq/e5UjWfvuRfOmkc7tYdX/p0uKTy0XEphGxWW3XOrWf16z9fCLw5Yh4d0RsDPwMeBA4p3b8\nOhHx5YiYEhGvqi1eeTpwRUrp1lY+n1PN+8D3v/ddPvzRj7HvfvtnP598ChdeeD6nn/YTPnu465qM\nt4uvvJ2Lr7wdGPuXefWVl+f4/3g/7z7kB/zu+58Y8do2m67LWq94OVvv9S3mPj8PgI8c+XMeueI4\ntttqfa647s6u1t9rEfWnCT1562U8eeuoaUIvlGeakMaX+dhfzMfOmZEq2rPPPstHD9yf7//3jzju\nW//Z63ImrIuvuoOLr7oDaJCPh+/Buw89hd9976DcNs89P59ZT3b0O19ajfJxrGMa2BK4nOwKdQK+\nU9t/OnBgSum4iFgW+CGwAvBn4B0ppXm1dvOAncnWylgOeAA4G2j5l82Od4/Nnz+fG2+4niM+/8VF\n+yKCHXfcmWuuvqqHlameU7++PyecfgnTpo+euQdLLfkSEol5CxYs2vfCvAUsXJh44+brVP+LZYPQ\nXHHjnVhx451G7Jv7yF38838+McYR9aWUpkfE0DShm7MSFk0T+kGdQ9uaJqTxYz6Wk/nYgBmpgn32\n04fyjne+i+2239GOd5879Wv7csLPLmXavYvn45C93r4F+7xzS2Y+/jQX/Ok2vvXjP/D8C/PHscoe\naqPj3eiSd0rpChrM8k4pHQUcNcZrD5LdgtOxvplqHhGHRMT0iHguIq6OiK16XdN4mDVrFoODg6yy\nyqoj9q+y6qrMnDGjR1WpnsMPfBvzFizglF//Kff1a2+Zzpzn5vHNT+3O0kstwbJLL8kxh72XgYFg\ntcnLj3O146/q04R6wXw0H8vCfGzMjCzWRM3HIb8560xuufnvHPX1b/a6FDVw+AE7M2/+IKec9Zcx\n25x54fUceOTP2eVj/8W3f3oJ//bOLfnJ0fuOY5W91a+34hSlL654R8ReZJf9DyKb2vQZsudLrp9S\n8kY+9Y3NN1qTg/fejm32OWbMNo8/OYd9j/gxJ31xLw7eZ3sGFy7krIuu56ZpD7JwYZ2bfioiaGra\nz2LHNNA304TGm/mosjAfm2NGFmei5+NDDz7I5484jHPP/wNLLLFEr8tRHZtv+EoO3vstbLPv8XXb\nnXbO1Yv+fPs9M3hk1tNcePLBvGr1l3Pfw090u8ye61I+9o2+6HiTBeUPU0o/A4iIjwO7AgcCx/Wy\nsG6bPHkykyZN4tFHR045eXTmTFZdbbUxjlKvvHGzdVl5xZdy14XfWLRv0qTg2MP24NB/24HXvvso\nAC6/Zhobv+doVnzZsiwYXMgzc57nnj/8J/c+VPnvAV3RT9OEesB8NB9LwXzsnQmckRM2HwFuuvF6\nZj32GG/ZZktSbTWvwcFB/vqXP/GjU37ArKeea33arrrijZutw8orvJS7zj9q0b5JA8Gxn3kPh+6z\nLa/d/Ru5x/3t1vuIgHXXnDwhOt5V1/OOd0QsAWxB9ugKAFJKKSIuIXu+ZKUtscQSbD5lCy6/7FLe\n9e7dgGxFw8svv5SDD/1kj6vTaGecdw2XXnPHiH3nnXwIZ5x3LT879+rF2s9+ei4A2221Piuv+FLO\nu+KWcamzl9qZ9uP3gnzmo/lYJuZjc8zIYkz0fATYfsedufpvfx+x7+MfPZANNtyQww7/nJ3uPnLG\n+ddx6TXTRuw7778+wRnnX8fPfn/NmMdttuErSQlmzHq62yX2harnY8873sBkYBL5z5fcYPzLGX+f\n/PRhHPThA5gyZQu23Gprvv+97/Lc3Lnst/8BvS5tQlp26SVZd62VF01dWXuNldh4/TWY/dQcHpz5\nJE8+89yI9vMXDDLz8ae5+/7HFu374Ltfz7TpM3hs9rO8YdN1+Pbh7+OkX1w2ok1VdWlFyonKfDQf\n+4r52DkzsjATPh+XW245NtzotYvte/nLV2KDDTfqUVUT17JLL8m6a05e9Pu69hqT2fg1qzP76bl1\n8vEZ7n4gm+3z6jVWYq+3T2HqX27niafmsPH6q3PsZ3bnzzf8k3/cPTHWNal6PvZDx7stRxz+GZZf\nfuRCLHvutQ977b1Pjypq3/v3/ACPz5rF0V87kkdnzmSTTTfj3POnsvLKK/e6tAlpyr+uxdQffZKU\nICU45rA9APjF76/h4187Y7H2ec9qXP/Vq3L0/9uNFV+2LPc9/DjH/M9F/OBXf+xy5e0ZnH0ng7Pv\nGrEvDc4bo3VjVR+tLIuqZKT52F/Mx87yEczIfvD5Iw5bLB/fv+fe7LlXufIxT5k6IVUz5bVrMvWU\nQ17Mx0+/B4BfnH8dHz/6V4u1T6MCcv78Bey49QYcsvd2LLfMkjw480n+95K/c+xP/jAu9bdq8Ilp\nDD4x8iq++VhfjP6HPu4FZFOF5gLvSymdO2z/acDyKaX3jmo/Bbj+ymuuZ/MpU8a1VrVvxa0O7XUJ\nasHCuY8x786zALZIKd3QzDFDv5sbHfxDllt9/Zbeb87Dd3L7yR9r6f0mglbzsfaaGVlCZmR5tJOP\nYEYWrZN8/NOV17HZ5uZjWazy5sN6XYKatHDuo8y7/VdgPubq+ePEUkrzgevJni8JQGTDdTsBV/aq\nLkntyVakbHHrddF9ynyUqseMLIb5KFVP1fOxX6aanwCcFhHX8+LjIJYFTutlUZJaV/X7c3rAfJQq\nxIwslPkoVUjV87EvOt4ppbMiYjJwNLAqcBOwS0ppYqy0IkljMB8lKZ/5KKlM+qLjDZBSOhk4udd1\nSOpM1RfG6AXzUaoOM7JY5qNUHVXPx77peEuqhqpPE5KkTpiRkpSv6vlox1tSsdoYrSzVyhiS1Akz\nUpLyVTzGggbYAAAgAElEQVQf7XhLKlS2ImWLo5XdKUWS+o4ZKUn5qp6PdrwlFarq9+dIUifMSEnK\nV/V87PlzvCVJkiRJqjKveEsqVNUXxpCkTpiRkpSv6vlox1tSoaoempLUCTNSkvJVPR/teEsqVsVX\npJSkjpiRkpSv4vlox1tSoYI2RivLlJqS1AEzUpLyVT0fXVxNUqGGVqRsdZOkicCMlKR83cjHiHhp\nRJwYEfdGxNyI+EtEbDmqzdER8XDt9YsjYr1ufD473pIKNXR/TqubJE0EZqQk5etSPv4Y2AnYF3gd\ncDFwSUS8ovaenwMOBQ4CtgbmAFMjYsmiP58db0ml0E8jlpLUT8xHSVpcRCwN7AH8R0rprymle1JK\nXwP+CXyi1uxTwNdTSuellG4F9gdWB3Yvuh473pIK1cVplH0zYilJ7epSRpqPkkqvC/n4EmAS8MKo\n/c8Bb46ItYHVgEuHXkgpPQ1cA2xT6IfDjrekgkUEAy1ujaYJ9duIpSS1q+iMNB8lVUXR+ZhSeha4\nCvhKRLwiIgYi4oNknepXkHW6EzBz1KEza68Vyo63pEJ16WpOX41YSlK7qn5FR5La1aXvkB8EAngI\neJ5s9s8vgYVd/TA5fJyYpEIFtLwQUKPWKaVnI2JoxPIOspHIfyP70ngX4zxiKUntKjojzUdJVdEo\nHx/52x+Y8bc/jNi34Lln654zpTQd2CEilgFellKaGRFnAvcAM2pvuyojM3JV4MY2PkJdTXW8I+Jt\nzZ4wpfSHxq0kVVUEDLS4AG+T30E/CPyEbMRyAXAD2YjlFq29W7HMR0mt6FJG9mU+ghkpqXmN8nGN\nrd7GGluNjJSn7r+Dq445oOG5U0rPAc9FxIrALsDhKaXpETGDbI2Mm7Ma4mXA64EftPcpxtbsFe+L\nmmyXyKY7SZqgGj3a4eHrpvLIdReP2Df/uWcanrefRixHMR8lNa0bGdnH+QhmpKQmtfP4xCbWCXob\nWQZOA14DHAf8Azit1uRE4MsR8U/gXuDrwIPAOS0V0oRmO97LFP3Gkiam1bfahdW32mXEvqfuv4Mr\nv/Whpo7vhxHLUcxHSYXpJCP7MB/BjJTUW8sD3wLWAJ4AfgN8OaU0CJBSOi4ilgV+CKwA/Bl4R0pp\nXtGFNNXxTimNXrADgIgYSCmN+43pkvpXkwtdLHZM4zb9M2I5nPkoqRXdyMh+zUcwIyU1rxv5mFI6\nGzi7QZujgKNae+fWtbyqeW0Z9v+IiLuB5yNindr+r0bE/oVXKKlUos3/NWF5sqszt5N9mfwT8Pbh\nI5bA98lGLK8hu8rSlRHLsZiPkhrpUkb2fT6CGSmpvi5+h+wL7TxO7HPAIcA3yRbwGHIn8PEiipJU\nXkMLY7SyNTO6mVI6O6W0XkppmZTSGimlT6WUnhnV5qiU0uoppWVTSruklP7Zrc85BvNRUl3dyMiS\n5COYkZLq6NZ3yH7RTsf734GDUko/BgaH7b8J2LCQqiSVVhCLFsdoeivRaGUD5qOkusxIM1JSvqrn\nYzvP8V6TbGQyz1Id1CKpArp1j3dJmI+S6jIjzUhJ+aqej+1c8Z4GbJOz/73UVsuUpAnKfJSksZmR\nkiasdq54fwP4YUSsQtZxf2dEbAB8lCw4JU1gAxEMtDj82Gr7PmY+SqrLjDQjJeWrej623PFOKf0m\nIp4Evkq2MMaJZPfm7JlSurDg+iSVTTsLXZQnM+syHyU1ZEaakZLyVTwf27niTUrpEuASgIiIlFIq\ntCpJpZXdn9NaCpZosLIh81FSPWakGSkpX9Xzsa2ON0BEvA7YqPbnf6SUbiusKkmlFbSxMEZXKukd\n81HSWMxIM1JSvqrnY8sd74hYDfg5sBPwXG330hFxObBfSumRAuuTVDJVvz+nHvNRUiNmpBkpKV/V\n87GdVc1PBVYENk8pLZdSWg6YAiwP/E+RxUkqp2hxqxDzUVJDZqQZKSlflfOxnanmOwFvTin9fWhH\nSunvEXEwcEVhlUlS+ZiPkjQ2M1LShNVOx/vhMfYnYEYHtUiqgIhoY2GMso1Zjsl8lFSXGZnLjJRU\n+XxsZ6r554Hv1xbGABYtknEi8LmiCpNUTgPR3lYR5qOkusxIM1JSvqrnY1NXvCPiEbLRyCErAn+P\niKGFMZYB5gHfA84utEJJpVL10crRzEdJrTAjzUhJ+aqej81ONT+qm0VIqpYSZWARjup1AZLKxYyU\npHxVzsemOt4ppR92uxBJ1VD10crRzEdJrTAjJSlf1fOxncXVFomIgdHnSCnN66giSaoA81GSxmZG\nSppoWl5cLSKWiYjjI+J+sntynhu1SZrAqr4wRj3mo6RGzEgzUlK+qudjO6uafwvYDfgCWWgeUts3\nEziwuNIklVHEi1OFmt96XXVhzEdJdZmRZqSkfFXPx3Y63u8FPpFSOgMYBC5JKX0Z+BLwviKLk1RO\n0eJWIeajpIbMSDNSUr6i8zEipkfEwpzt+7XXT8t57YJufLZ27vGeDNxV+/PTZI+FAPgjcFIBNUkq\nsYEIBlocfmy1fR8zHyXVZUaakZLydSkftwQmDft5Y+APwFm1nxNwIXAAL/blX2ipiCa1c8V7OrBW\n7c/TgD1qf96FLEQlTWDB0FShFrZG5+yj0coGzEdJdRWdkSXKRzAjJdXRje+QKaXHU0qPDm3Au4G7\nU0p/HtbshZTSY8PaPdWNz9fOFe+fA1sBfwG+DfwuIg4BliO7Z0fSBNalR0H0zWhlA+ajpLq6kJFl\nyUcwIyXV0e3HiUXEEsC+wPGjXto+ImYCs4HLgC+nlJ5oqZAmtNzxTikdO+zPF0bE68hC9J8ppWuL\nLE6SIButHP5zRIw5Wjm+lY1kPkoab2XJRzAjJfXce4HlgdOH7bsQ+C3ZjJx1yRZ8vCAitkkppSLf\nvKPneAOklO7ixft1JE1wQ1N/Wj2m+ba9Ha1shfkoabRuZmSZ8hHMSEkjdfs7JNnTEy5MKc0Y2pFS\nOmvY67dFxC3A3cD2wOWtVVNfUx3viDio2ROmlH7UfjmSyi7aWBijxWlFPR2tHM18lNSKLmdkX+Uj\nmJGSmtcoH+/6y/n88y8jl6iYN+eZZs+9FrAzsHu9diml6RExC1iPXnS8ga812S4BhqY0gVV9tDKH\n+SipaV3OyH7LRzAjJTWpUT6u/5ZdWf8tu47Y99g9/+A3/7FnM6c/EJgJ1F1cMiJeCawEPNLMSVvR\nVMc7pfSKot9YE8ykju9q0HiaNKlxmzEE9RfGuOvP53PXn0dm3gtzyzNamfNe5qM6Z0aWRwf5CN3L\nyH7Mx9r79V1GvmTSAEu8pJ0H+6gnBhf0ugI1a3Cwo8Mb5eNYxzRsk530AOC0lNLCYfuXA75KNito\nBlkuHgvcCUxtqZAm+F96SYUK6j+ncIO37MoGo0YrH737H5z1H+9v5vQ9H62UpE50MSPNR0ml1igf\nxzqmCTsDawI/HbV/ENgE2B9YAXiYrMN9ZEppfoulNGTHW1Ip9MtopST1G/NRksaWUrqYkY9dHNr/\nPPD28arDjrekQmX357S6cFBTzfpitFKSOtGljDQfJZVeF79D9gU73pIKNRDZ1uoxjfTLaKUkdaIb\nGWk+SqqCbn2H7Bd2vCUVKtoIzTKNVkpSJ8xIScpX9Xxsa0nHiNg6Ik6NiMsjYvXavr0j4g3Fliep\nbCKira0qzEdJ9ZiRZqSkfFXPx5Y73hGxG3AFsBSwDbB07aVVgC8XV5qkMhrgxalCTW+9Lrog5qOk\nRsxIM1JSvqrnYzu1fhU4NKW0HzB8YY6/AFsUUpWk0soWxmh9qwjzUVJdZqQZKSlf1fOxnY73hsCl\nOfufBFbsrBxJKjXzUZLGZkZKmrDaWVztUWBt4N5R+7cBpndakKRyiwgGWn4URImGK+szHyXVZUaa\nkZLyVT0f27ni/VPgxIjYFEjAShHxPuB44EdFFiepfAba3CrCfJRUlxlpRkrKV/V8bOeK9zeAJYCr\nyBbFuBpYAJyUUvpugbVJKqF27rcp0WBlI+ajpLrMSDNSUr6q52PLHe+U0kLgKxFxDLAB8FLglpTS\n7KKLk1Q+VZ8mVI/5KKkRM9KMlJSv6vnYzhVvAFJKc4AbCqxFUgUEbYxWdqWS3jEfJY3FjDQjJeWr\nej623PGOiAvqvZ5Semf75Ugqu6HnKrZ6TBWYj5IaMSPHZkZKE1vV87GdK973jfp5CWAzYD3gVx1X\nJEnlZT5K0tjMSEkTVjv3eH8ib39EfJNyXe2X1AVVvz+nHvNRUiNm5OLMSElQ/XwscgX2nwIfLfB8\nkkpoaEXKVreKMx8lAWbkGMxISZXPx7YXV8sxBZhf4PkklVDV789pk/koCTAjx2BGSqp8PrazuNov\nR+8CXgG8CTiuiKIklVtM0BmD5qOkZpiRL+7CjJQ0TJXzsZ0r3qP/NhYCNwEnpJTO7bwkSWU2QBuj\nlV2ppCfMR0l1mZEjmJGSFql6PrbU8Y6IScB3gWkppae6U5IklY/5KEljMyMlTXQtDRKklAaBPwMr\ndaccSWU3dH9Oq1vZmY+SmmFGmpGS8nUrHyNi9Yj4eUTMioi5EfH3iJgyqs3REfFw7fWLI2K9wj9f\nG8f8A1iz6EIkVUQE0eLWzJKU/RKaDZiPkurrQkaWJB/BjJRUT3fycQXgr8ALwC7ARsBngdnD2nwO\nOBQ4CNgamANMjYgli/x47XS8jwCOj4idI2LFiFhy+FZkcZLKpxujlf0Umg2Yj5LqKjojS5SPYEZK\nqqNLV7w/D9yfUvpISun6lNJ9KaVLUkrTh7X5FPD1lNJ5KaVbgf2B1YHdi/x87SyuNnXU/482qc1a\nJFVAO89UbKL9otActu++UW0WhWZ2ztgfmEkWmme1VlHbzEdJdXUhI8uSj2BGSqqjS98h3w1cFBFn\nAdsBDwEnp5ROzY6PtYHVgEuHDkgpPR0R1wDbUGBGttPxfkdRby6peoJgoMXUbOLREX0Tmg2Yj5Lq\n6kJGliUfwYyUVEeXvkOuA3wC+A7wn2Szfk6KiBdSSj8ny8dENhg53Mzaa4VpuuMdEUcCx6eUxhql\nlKS2FgJqon3fhGYe81FSs7qQkX2dj2BGSmpOl75DDgDXppS+Uvv57xHxOuDjwM9bLLEjrVzx/ipw\nCjC3S7VI0lj6JjTHYD5K6pV+z0cwIyUV4PpLzuWGS34/Yt9zzz7T6LBHgNtH7bsd2KP25xlAAKsy\ncoByVeDGdmvN00rHuwIPs5DUbY3uz7n+4nO5vsShOQbzUVJTupCR/Z6PYEZKakKjfNzyrbux5Vt3\nG7HvgWm38u0P7zbGEUC2+OQGo/ZtQG0tjJTS9IiYAewE3JzVES8DXg/8oLVPUF+r93inIt9cUvUM\nEAzU+Y611Vvfw1Zvfc+IfQ9Mu5VjP/zueqftm9Csw3yU1FAXMrIM+QhmpKQGGuXjWMc08F3grxHx\nBbI1LV4PfAT46LA2JwJfjoh/AvcCXwceBM5pqZgGWu143xkRdYMzpfTyDuqRVHZtrEjZRMb2TWjW\nYT5Kaqz4jCxDPoIZKamRLnyHTCn9LSLeCxwDfAWYDnwqpXTmsDbHRcSywA+BFYA/A+9IKc1rsZq6\nWu14fxV4qsgCJFVLNxbG6KfQrMN8lNRQ0RlZknwEM1JSA11aXI2U0gXABQ3aHAUc1dq7t6bVjveZ\nKaVHu1KJpEoYiNYfBdFM+34JzTrMR0kNdSMjS5CPYEZKaqBb3yH7xUALbb03R5LymY+SNDYzUtKE\n56rmkgpXosHHoky8TyypbWakJOWrcj423fFOKbVydVzSBJXdn9PqNKEuFTNOzEdJzTIjJSlf1fOx\n1Xu8JamuRs9gHOsYSZoIzEhJylf1fLTjLalQQWuLRwwdI0kTgRkpSfmqno92vCUVKiKIFocfW20v\nSWVlRkpSvqrnox1vSYUKWh99LE9kSlJnzEhJylf1fHSxC0mSJEmSusgr3pIKNRDRxoqUZRqvlKT2\nmZGSlK/q+WjHW1LhyhOBkjT+zEhJylflfHSqeZ845eQfsOFr1mbFf1mGbd/0Bv523XW9LmnCetNm\n63D2CR/h7guOYs61J7Drtv86ZtuTvrAnc649gYP32nbRvrVWW5E5157As9ecwJxrR26777jJeHyE\nngpefBxE01uvi1ZfMx/7h/nYOTNSRTMj+8ObNl+Xs0/8GHdP/QZzrj+JXbfbeMy2J31pb+ZcfxIH\n77Pdon1rvWJF5lx/Es/+7STmXD9y232nzcbjI/Rc1fPRK9594Oyzfs3nj/gsP/jvH7HlVlvz/e99\nl9123YWb/3EnkydP7nV5E86yyyzJzdMe4vRzrubM4w4cs91u22/MVq97FQ8/+tSI/ffPmM2rdzly\nxL4P7/FGPr3fDkz96+1dqbmfVH1FSo0v87G/mI+dMyNVJDOyf2T5+CCn/+5Kzjz+o2O2222HTfLz\n8ZHZvHrnL47Y9+H3v5lP77cTU/96W1dq7jdVz0evePeB73/vu3z4ox9j3/32Z4MNN+T7J5/CMssu\ny+mn/aTXpU1IF191B1//4YWcd8WtjPW7vPrKy3P84XtwwJd+zoLBhYu9/tjsZ0dsu+2wMb/5w408\n98L8LlffewNtblIe87G/mI+dMyNVJDOyf1x85e18/b/P57w/3lI/H//j/RzwxdNYMDi42OuL5+Om\n/OYPN/Dc8+ZjFfKxTLVW0vz587nxhuvZYcedFu2LCHbccWeuufqqHlamek792r6c8LNLmXbvzIZt\nN9/wlWy6/hqcfs4141BZH6iNVrayjflfKE1o5mM5mY8NmJEqiBlZPqd+fX9OOP0Spk1vIh83WpNN\nN1iD03935ThU1icqno990fGOiLdExLkR8VBELIyI3Xpd03iZNWsWg4ODrLLKqiP2r7LqqsycMaNH\nVameww/YmXnzBznlrL801f5D73kDt0+fyXW33dflylRF5qP5WCbmo8bTRM5HMCPL5vAD38a8BQs4\n5dd/aqr9h3bfhtvvmcF1t5qPVdEXHW9gOeAm4GAg9bgWaUybb/hKDt77LXzs6F821X6pJV/CB3aZ\nwmm/u7rLlfWPaHPTmMxHlYL52BwzslDmo0ph843W5OC9t+NjX/1FU+2XWvIlfODtW3La/02gq91U\nPx/7YnG1lNJFwEUAUaY75AswefJkJk2axKOPjpxy8ujMmay62mo9qkpjeeNm67DyCi/lrvOPWrRv\n0kBw7Gfew6H7bMtrd//GiPZ77LwZyyy1BL+8YOKsMJrN+ml1YYwuFVMB5qP5WBbmY3PMyOJM5HwE\nM7JM3rjZuqy84ku568IXc3DSpODYw/bg0H/bgde++6gR7fd46+ZZPp5vPjZzTFn0Rcd7IltiiSXY\nfMoWXH7Zpbzr3dkMqZQSl19+KQcf+skeV6fRzjj/Oi69ZtqIfef91yc44/zr+NnvF79H8UO7vZ7z\n/3QrTzw1d7xK7Ll2Frrol6k36i/mY7mYj80xI1UUM7I8zjjvGi695o4R+847+RDOOO9afnbu4rN+\nPvSebTj/ilt44qk541ViX6h6Ptrx7gOf/PRhHPThA5gyZYtFj4J4bu5c9tv/gF6XNiEtu/SSrLvm\n5EUjbmuvMZmNX7M6s5+ey4Mzn+TJZ54b0X7+gkFmPv4Mdz8wa8T+dV45mTdvvi67ffKUcau9L7Tx\nKIhSDVdqXJmP/cV8LIAZqQKZkf1j2aWXZN21Vl409XntNVZi4/XXYPZTc+rk49Pcff9jI/avs+Zk\n3jxlPXY75ORxqryPdDkfI+LzwDeBE1NKh9X2/RT40KimF6WU3tlaIY3Z8e4D79/zAzw+axZHf+1I\nHp05k0023Yxzz5/Kyiuv3OvSJqQpr12TqaccQkqQEhzz6fcA8Ivzr+PjR/9qsfYp5d9Wtv9uW/PA\njNlcds2dXa2337Rzv41fKTUW87G/mI+dMyNVJDOyf0z517WY+qNPvpiPh+0BwC9+fw0f/9oZi7Uf\nIx7Zf7dteGDGE1w26gr5RNDNfIyIrYCDgL/nvHwhcMCw073QYhnN1TDWfxR7JSIWArunlM4d4/Up\nwPVvfsu2LL/88iNe23Ovfdhr733GoUq1asU3fLrXJWgMg09MY/CJkdND0+A80rMPAWyRUrqhmfMM\n/W4ef+ZU1t1ok5ZquPv2mzl8712afr9ej1j2SqN8rLUxI0vIjOxPReUjjF9Gmo/mY9WsuNWhvS5B\nOQZn38ng7LtG7EuD80hzHoY+y8eIeClwPfAJ4CvAjaPycfmU0h4tvXEbSnvF+7jjv8vmU6b0ugyp\n9Ca9fAMmvXyDEfsWzn2UebcvfvWqGUHrsyJbad4PI5ZlYEZKnSs6H6G7GWk+Nsd8lDo3acX1mbTi\n+iP2LZz7GPPuPKvtc3YxH38A/D6ldFlEfCXn9e0jYiYwG7gM+HJK6YnWKmmsLzreEbEcsB4v/t2t\nExGbAk+klB7oXWWSWjVAMNDiRKFm29dGLH8BfIRsxHK0F1JKj+XsLy3zUaqWbmWk+QiYj1KpdSMf\nI2JvYDNgyzGaXAj8FpgOrAt8C7ggIrZJBU8N75eF4LYEbiSbApCA7wA3AF/rZVGS2hBDj4Nofmsh\nYxeNWI7x+vYRMTMi7oiIkyPi5UV8pB4zH6Uq6V5Gmo/mo1RuBedjRLwSOBHYN6U0P69NSumslNJ5\nKaXbareqvAvYGti+6I/XF1e8U0pX0D+DAJI6ELX/tXpMwzZ9NGI5nsxHqVq6kZHmo6QqaJSPV1zw\nv1xx4f+N2DfnmafrnXILYGXghnhxufRJwLYRcSiw1OgMTClNj4hZZLNpLm/5Q9TRFx1vSdWxaASy\nxWPqv75oxHLneiOWw368LSJuAe4mG7EsNDglqV1FZ6T5KKkqGuXj9rvuwfa7jlwD7Z//uJlPfeCt\nYx1yCbDxqH2nAbcDx+QNPNYydSXgkWbrbpYdb0mFanR/zh8v+F/+eEFLo5XQZyOWktSuLmSk+Sip\nEoq+xzulNAf4x/B9ETEHeDyldHttnYivks0ImkGWiccCdwJTWyqkCXa8JY2r7d+5B9u/c/HRyv/3\ngZ3rHdZXI5aS1C1tZKT5KEnNG56Jg8AmwP7ACsDDZB3uI8eaQdQJO96SitXGNMpGg5v9NmIpSW0r\nOCPNR0mV0YXvkKOllHYc9ufngbe3+I5ts+MtqVDduMd7DD0bsZSkdo1TRpqPkkpnHL9D9oQdb0mF\nyp7s0OqKva3r5YilJLVrPDLSfJRURuP1HbJX7HhLKtQAMNBiCvosGEkThRkpSfmqno92vCUVrPVn\n1JZrvFKSOmFGSlK+audjmQYJJEmSJEkqHa94SypU1RfGkKROmJGSlK/q+WjHW1Khoo1pQq1PK5Kk\ncjIjJSlf1fPRjrekQg1EGwtjlCczJakjZqQk5at6PtrxllSwai+MIUmdMSMlKV+189GOt6RCVf3+\nHEnqhBkpSfmqno92vCUVKmh97LFEmSlJHTEjJSlf1fPRx4lJkiRJktRFXvGWVKiBCAZanPfTantJ\nKiszUpLyVT0f7XhLKlTVpwlJUifMSEnKV/V8tOMtqXhlSkFJGm9mpCTlq3A+2vGWVLjWHwUhSROH\nGSlJ+aqcj3a8JRWq6o+CkKROmJGSlK/q+WjHW1Khqn5/jiR1woyUpHxVz0cfJyZJkiRJUhd5xVtS\nsao+XClJnTAjJSlfxfPRK96SChVt/k+SJgIzUpLydSMfI+LjEfH3iHiqtl0ZEW8f1eboiHg4IuZG\nxMURsV43Pp8db0mFGloYo9Wt/jn7JzQlqRNFZ6T5KKkquvEdEngA+BwwBdgCuAw4JyI2yt4zPgcc\nChwEbA3MAaZGxJJFfz473pIKFy1uTeib0JSkThWckeajpMoo+jtkSun8lNJFKaW7U0r/TCl9GXgW\neEOtyaeAr6eUzksp3QrsD6wO7F7gxwLseEsqWquJ2URy9lNoSlJHCs5I81FSZXThO+SI00cMRMTe\nwLLAlRGxNrAacOlQm5TS08A1wDYdf55R7HhLKpVeh6Yk9SvzUZIWFxGvi4hngBeAk4H3ppSmkeVj\nAmaOOmRm7bVCuaq5pEK1sxBQM+0j4nXAVcDSwDPUQjMitmEcQ1OSOtGNjDQfJVVBt75DAncAmwLL\nA+8HfhYR27ZcYIfseEsqVNDUQheLHdOEvghNSepElzLSfJRUeo3y8cJzzuaic38zYt+zTz/V8Lwp\npQXAPbUfb4yIrcluwzmu9rarMnKAclXgxuYrb44db0mFanS7TV5oPlOi0JSkTnQjI81HSVXQKB/f\n+Z49eed79hyx7/ZbbmKfd7U8zjgALJVSmh4RM4CdgJsBIuJlwOuBH7R60kbseEsqVoPUfMfue/KO\n3XNCc9dyhKYkdWR8MtJ8lFQ+LS6WtuiYei9HfBO4ELgf+BdgX2A74G21JicCX46IfwL3Al8HHgTO\nabGShux4SypUl+5f7JvQlKROFJ2R5qOkqujSPd6rAKcDrwCeIhuEfFtK6TKAlNJxEbEs8ENgBeDP\nwDtSSvNaq74xO96SihWt37/YRMb2TWhKUkeKz0jzUVI1dOE7ZErpI41OkVI6CjiqxXdumR1vSX2v\nn0JTkvqJ+ShJ5WDHW1KhunB7jiRVhhkpSfmqno92vCUVq+qpKUmdMCMlKV/F89GOt6RCdWlhDEmq\nBDNSkvJVPR/teEsqVLSxMEbLC2lIUkmZkZKUr+r5aMdbUqEqPktIkjpiRkpSvqrn40CvC5AkSZIk\nqcq84i2peGUafpSk8WZGSlK+CuejV7z7zK/P/FWvS1ALBp+Y1usS+lK0+D+pGeZj+ZiR+cxIFc18\nLJ/B2Xf2uoS+VOV8tOPdZ87+tcFZJn6pXNzQwhitblIj5mP5mJGLMyPVDeZj+QzOvqvXJfSdquej\nU80lFarqC2NIUifMSEnKV/V8tOMtqVhVT01J6oQZKUn5Kp6PdrwlFaqde27Kdo+OJLXLjJSkfFXP\nxzJ2vJcGmHbH7b2uoyueeuopbrzhhl6XUbiFcx/tdQldkQbnVfKzLXzuiaE/Lt3LOtSWymZkVfMR\nzBHSDVkAACAASURBVMgyMR9LzXwsoYVzH+t1CV2R5WO1PtvC52cP/dF8zFHGjverAf79Qx/scRnd\n88bXb9HrEtSCebdXekGTVwNXtnJAOwtdlGlhjBJ4NVQ3I83H8qlwRr6aFvMRzMgeezWYj+of8+48\nq9cldMurMR8XU8aO91RgX+Be4PneliJV1tJkoTm1nYNLlIFVZEZK3dVRPoIZ2UPmo9Rd5mMdpet4\np5QeB37Z6zqkCaDlkUqg8gtj9DszUhoX7eUjmJE9ZD5K48J8HEPpOt6S+lvVF8aQpE6YkZKUr+r5\naMdbUqGCNu7P6UolktR/zEhJylf1fLTjLalQFZ8lJEkdMSMlKV/V83Gg1wVIUiMR8ZaIODciHoqI\nhRGx26jXf1rbP3y7oFf1StJ4MiMlKV8/5aMd7z4REYdExPSIeC4iro6IrXpdk/I1+gWe8KLNrb7l\ngJuAg4E0RpsLgVWB1WrbPu1/CPUT87FczMgGzEgVzIwsD/OxgYrno1PN+0BE7AV8BzgIuBb4DDA1\nItZPKc3qaXHKM/QL/GPgf3tcS9/pxsIYKaWLgIsAIsa8++eFlNJjLb2x+p75WEpmZB1mpIpkRpaO\n+VhH1fPRK9794TPAD1NKP0sp3QF8HJgLHNjbspQnpXRRSunIlNI5lOvWkvER2cIYrWwF/S1uHxEz\nI+KOiDg5Il5eyFnVa+ZjyZiRDZiRKpYZWSLmYwMVz0c73j0WEUsAWwCXDu1LKSXgEmCbXtUltas7\ns4QauhDYH9gROALYDrigzsimSsB8VBWZkSqKGamqqXo+OtW89yYDk4CZo/bPBDYY/3KkziwagWzx\nmE6klM4a9uNtEXELcDewPXB5Z2dXD5mPqhwzUgUyI1UpVc9HO96SxtXvfvtrzvntWSP2PfP0U4W+\nR0ppekTMAtbDL5WSSsSMlKR8Zc9HO969NwsYJFtJb7hVgRnjX47UqfoTf3Z/397s/r69R+y75e83\n8s4d3lBcBRGvBFYCHinspOoF81EVZEaqMGakKqba+eg93j2WUpoPXA/sNLSvdk/BTsCVvapLaler\ni2I0M60oIpaLiE0jYrParnVqP69Ze+24iHh9RLwqInYCfgfcCUzt7qdVN5mPqiIzUkUxI1U1Vc9H\nr3j3hxOA0yLiel58FMSywGm9LEr5ImI5suknQ7/q60TEpsATKaUHeldZf2hnoYsm2m9JNt0n1bbv\n1PafTvZcxk3IFsZYAXiYLCyPrH0pUbmZjyVjRtZnRqpgZmSJmI/1VT0f7Xj3gZTSWRExGTiabHrQ\nTcAuPm+zb9X7BfbxHXS+0MVoKaUrqD9D5+3FvqP6hflYSmZkA2akimJGlo752ECV89GOd59IKZ0M\nnNzrOtRYE7/AE1rU/tfqMdJYzMdyMSPrMyNVNDOyPMzH+qqej3a8JRWrS/OEJKkSzEhJylfxfHTE\nRZIkSZKkLvKKt6RCVXywUpI6YkZKUr6q56Mdb0mFaubRDnnHSNJEYEZKUr6q56Mdb0mFykYrW10Y\n4/+zd9/xUpRnG8d/90E6gihKUTCgYAcRa+yxxRh9jSUaC7bEWGNsMVbsUaPYookKscUeE7siIhA1\nil1RkaCigPTe67nfP2YO7ln2nG0zZ3fnXN989iNn6r0ELuaZeeZ5REQaB2WkiEhmSc9HNbxFJFpJ\n7yckIlIMZaSISGYJz0c1vEUkchWUgSIiDU4ZKSKSWZLzUaOai4iIiIiIiMRIT7xFJFJJHxhDRKQY\nykgRkcySno964i0AmNnGZlZtZr3Dn/c0s1Vm1rYEtQw3s4H1rB9gZh/lecxqMzukyLruN7N/FXOM\nxsAK/J9IuVI+5nQM5WOOlJGSNMrInI6hjMxB0vNRDe8yFv4lrQ7Da5mZjTOzy80srv/fPOXXbwGd\n3X1+LjtmC7oYePZNpBSMH+5Y5vwpddFScZSP9VI+ljFlpDQEZWS9lJFlKun5qK7m5e9l4ESgBXAg\ncDewDLgpfcMwTN3dCw2U1X923X0lML3A44iINATlo4hI3ZSRImVET7zL3zJ3n+HuE939XuA14P8A\nzOxEM5tjZgeb2efAUqBruO7XZvaFmS0J/3t66kHNbEcz+zBc/y7Ql5Q7gGE3oerUbkJmtmt4V3KR\nmc02s5fNrJ2Z3Q/sCZyTcne1W7jP1mb2kpktMLOpZvaQma2XcsxW4bIFZva9mZ2X72+QmW1vZq+a\n2Qwzm2tmI8ysb4ZNu4S1LDazr83s8LTjbGRmT4S/p7PM7Bkz2zjfehq7vO9UFvA+j0hI+ZiF8rH8\nKCOlASkjs1BGlpek56Ma3pVnKdAs/LUDrYA/AKcAWwHTzexY4ErgYmBz4BLgajM7HsDMWgPPA58B\n24Xb3pzhXKkhui1BYH8G7AzsAjwLNAHOAd4G7gM6Ap2BiWbWDhgGfBCe5wBgA+DJlHPcDOwOHAzs\nD+wVbpuPtYEHgB8DOwH/A14Kv2eqq4GngN7AI8DjZrZZ+P3WAoYA84Bdw2MtAF4J10nOCnk3p4JS\nU8qZ8nFNyseyo4yUklFGrkkZWVaSnY/6w1BBzGxfguC5PWXxWsDp7v5ZynZXAue7+7Phou/MbCvg\nt8DDwLEEf0p/7e7LgTFm1pWgC1JdLgTec/ezU5aNTTnncmCxu89IWXYW8KG7X56y7NfABDPbFJgC\nnAwc4+4jwvUnAJNy+O1Yzd2Hp/5sZqcBRxHcQX0pZdWT7n5/+OsrzGw/4GzgLOBowNz91JTjnALM\nIQjy1/KpSUQalvIxM+WjiIAysi7KSGlIaniXv4PNbAHQlCDoHgGuSlm/PC0wWwGbAIPNbFDKdmsR\nBAAEdzA/DQOzxttZ6tiW2ncZc9EH+ElYfyoPa2xF8L3eXb3CfY6ZjSUPZrYBcB1BSG5AcAe1JdAt\nbdN30n5+O6wRgjuYPTPU2jysVaGZo6RPBSFlRfmYhfKx/CgjpQEpI7NQRpaXpOejGt7l73XgNGAF\nMNndq9PWL0n7uU3431+TEkahVUXUkX6eXLQBniPoxpT+12IK0LOIelI9BLQnuPM4gWDgkHf4oTtV\nLtoA7wPHsGatM9bcXOpSSKefCspMKS/Kx+yUj2VGGSkNSBmZnTKyjCQ9H/WOd/lb5O7j3X1ShsBc\ng7tPByYDm7j7N2mf78LNxgC9zSw1VHbJcuhPgX3qWb+c4C5hqg8J3hn6LkMtS4CvgZUE79QAYGbt\ngV7ZvmeaHwN3uPsQdx9D8A9Mhwzb7Zzh5zEptfYEZmSoNf0OptTHCvyI5E/5mJ3ysdwoI6XhKCOz\nU0aWk4TnoxreyTQAuNjMzjaznuGokCea2bnh+kcJuuoMMrMtzOxnwPkZjpP6R/lPwA5mdpeZbWNm\nm5vZaWa2brj+W2AnM9s4ZcTJu4B1CQag2N7MepjZAWb2dzMzd18EDAb+bGZ7m9nWwP3kf1d1HHB8\nWNNOwD+AxRm2O9LMTgp/T64CdgD+Eq57BJgJPGtmu5nZj8xsLzO73cy65FlPo5b/sBg1g2OINAjl\no/KxpJSRUuaUkcrIkkl6PqrhnUDuPpigm9BJBHcZRwAnAN+E6xcRjAC5NcFdumsIuvKscaiUY44j\nGDGyNzAKeAs4hOBuIwQjS64CviAYFbObu08hGN2ximC0x0+BgcCclHkiLwTeIOhO9Gr46w/y/Mon\nE3QT+gB4kGDgkPT5I53gH5OjgU+A44Cj3f3L8PstAfYg6Gb0dPg97iN4P2d+nvU0anFOBWFmZ5rZ\neAumMHnHzHaI99tI0igflY+lFldGKh8lCspIZWQpJT0f7Yc/uyIihTOz7YAP3nznffr2zW82j48+\n+pDddt4eoJ+7f1jH8Y8i+EfxVIJ3z84FjgR6ufvMYmoXEYlbnBmpfBSRStZY8lFPvEWkUpwL3OPu\nD4V3mU8j6A52cmnLEhEpOeWjiEhmZZOPaniLSLRiGBjDzJoC/YBhNcvCrmavkX1QFxGR8hFxRiof\nRSQxEp6Pmk5MRCJVyEAXOWzfgWDE02lpy6cBm+V1MhGREoohI5WPIpIISc9HNbxFJFJjvxyT9/iS\nY78ck30jEZEEUEaKiGSW9HxUw1tEojITWHzyCce1KnD/ZeEx6jr2KqBj2vKOwNQCzyci0pDiykjl\no4hUukaRj2p4i0gk3H2CmW1B0K2nEDPdfUIdx15hZh8A+xBMG4KZWfjzHQWeT0SkwcSVkcpHEal0\njSUf1fAWkciEoZex8RyBgcADYYDWTAfRCnggpvOJiEQqxoxUPopIRWsM+aiGt4hUBHd/0sw6AFcT\ndBH6GDjA3WeUtjIRkdJSPoqIZFZO+WjBiOoiIiIiIiIiEgfN4y0iIiIiIiISIzW8RURERERERGKk\nhreIiIiIiIhIjNTwFhEREREREYmRGt4iIiIiIiIiMVLDW0RERERERCRGaniLiIiIiIiIxEgNbxER\nEREREZEYqeEtIiIiIiIiEiM1vEVERERERERipIa3iIiIiIiISIzU8BYRERERERGJkRreIiIiIiIi\nIjFSw1tEREREREQkRmp4i4iIiIiIiMRIDW8RERERERGRGKnhLSIiIiIiIhIjNbxFREREREREYqSG\nt4iIiEgFM7Pdzew5M/vezKrN7JC09QPMbIyZLTSz2WY21Mx2TNtmRLhvzWeVmd3dsN9ERCS51PAW\nERERqWytgY+BMwDPsH4scCawNbAr8C3wqpmtl7KNA/cCHYFOQGfgD/GVLCLSuJh7pnwWERERkUpj\nZtXAoe7+XD3brA3MA/Zx9+HhsuHAR+5+XsNUKiLSuOiJt4iIiEgjYWZNgd8Cc4FP0lYfa2YzzGy0\nmV1vZi0bvkIRkWRaq9QFiIiIiEi8zOwg4HGgFTAZ2M/dZ6ds8gjwXbiuN3AT0As4ooFLFRFJJHU1\nF5HImFk3oEOBu8909wlR1iMiUk4aIiPr6moePr3uHJ7/N8A+wI7uPrOO4+wFDAM2dffxBdYsIpKT\nxnANqSfeIhIJM+tG1VrfUb2y0EMsNrMtKiE4RUTyFUFGLjOzXoVmpLsvAb4JP++a2f+AU4Ab69jl\nXcCATQE1vEUkNo3lGlINbxGJSgeqV9J04/2xFu3z2tGXzmHFd6+2IrjTWdahKSJSoGIzsjnRZmQV\n0Lye9X0JRjqfEtH5RETq0iiuIdXwFpFIWct1qWq1QV77VJvFVI2ISHmJIyPNrDXBk+maDXuYWR9g\nNjALuBR4jqAR3QE4C+gCPBXu3wM4Bngp3L4PMBAY6e6f5VWsiEiBkn4NqYa3iIiISGXbHhhO8ITa\ngVvC5Q8CpwObA/0JGt2zgPeA3dx9TLjdcmBf4ByCOcEnEjTKr2ug+kVEEk8NbxGJllUFn3z3ERFp\nDGLISHcfSf1TxB6eZf9JwF75FSUiErGEX0Oq4S0i0augbj8iIg1OGSkiklmC81ENbxGJVsLvVoqI\nFEUZKSKSWcLzUQ1vEYmWWf53KxN8d1NEpBZlpIhIZgnPRzW8RSRaZgXcrayc0BQRKYoyUkQks4Tn\nY+U8mxcRERERERHJkZntbmbPmdn3ZlZtZoekrR9gZmPMbKGZzTazoWa2Y9o2I8J9az6rzOzufGtR\nw1tEolXTTSjfj4hIY6CMFBHJLJ58bA18DJxBMN1iurHAmcDWwK7At8CrZrZeyjYO3At0BDoBnYE/\n5Pv11PAWkYhV/TA4Rq6fLFFUTncrRUSKE31GiogkQ/T56O6vuPsV7v4ssEYr3d0fd/fX3f1bdx8D\nnAe0BXqnbbrY3We4+/Tws7CAbyciEqGE360UESmKnniLiGRW4nw0s6bAb4G5wCdpq481sxlmNtrM\nrjezlvkeX4OriUi0YhgYw91fAV4JNl1zY3d/vPbh7DzgFIK7lcNTVi129xn5FSciEqGEDx4kIlKw\nEuWjmR0EPA60AiYD+7n77JRNHgG+C9f1Bm4CegFH5HMeNbxFJFolngoih7uVxwNTgeeBa9x9SWQn\nFxHJJuHT5YiIFKx0+fg60AfoAPwGeMrMdnT3mQDuPihl28/NbAowzMy6u/v4XE+ihreIRGv1Ozd5\n7lPsaRvobqWISFFKlJEiImUvSz6umj6aVTM+q7XMVy4t+rThQ5hvws+7ZvY/gp6TN9axy7sE74tv\nCqjhLSKNToPcrRQRERGRhtdkg21ossE2tZZVL5zC8o/ujfpUVUDzetb3JRg7aEo+B1XDW0QiVv/7\nOaumj2bV9NG1llXS3UoRkeIU8A7jmgPxiogkUPT5aGatCa71ajbsYWZ9gNnALOBS4DmCRnQH4Cyg\nC/BUuH8P4BjgpXD7PsBAYKS71378noUa3iISrSoLPnVo0qk3TTrVnqGhesFkln/wt8grIYa7lSIi\nRcmSkXXuIyKSdPHk4/YEA+16+LklXP4gcDqwOdCfoNE9C3gP2C2cWgxgObAvcA7BLDsTCRrl1+VX\nqBreIhK1GN5fLKe7lSIiRdE73iIimcWQj+4+kvqn0D48y/6TgL3yKyozNbxFJFpGASNSZt2ibO5W\niogUJZ6MFBGpfAnPR91CbUBmtqeZVZvZYaWupYaZfWtmz0V4vI3D79g/h20fMDO9W5s4VT/cscz1\nkyWK3H2ku1e5e5O0z8nuvszdD3f3ru7e0t03cvdfuPuHKftPcve93H19d2/l7pu5+8XuvjDu3w3J\nnTJyjW2VkYkUfUZK8ikf19hW+ZhIyc7Hyqm0TIUBke2zysz2CHfxkha8plLWU/P0siKYWRcze9LM\n5pjZPDN7xsy6F3Cctczsi/DPxnlx1FpSNXMw5vuRRFJGFn3ucvv9qFMxGWlm99fxZ+OLuOtucMpI\nCSkfiz53uf1+1KnYa0gLnG5mH5nZYjObaWbDzGyb7HtXkITno7qaF++4tJ9PIOjSehy1Oz+MAbZM\nWyYVInzHeASwNnAtsBI4DxhhZtu6+5w8Dvc7oCsV9A+GSBGUkY1ARBm5lGAmgtQ/A/MiLlWknCgf\nG4GI8vF+4FfAQ8CdBK/N9QU2iKNmiYca3kVy90dTfzazXYB93f2x9G2tyDsyZtYynDKpUTGzzYCv\n3X1lCcs4E9gE2KGmC7OZvQJ8BpwPXJbLQcxsA+By4AbgmnhKLTErYCqICrpbKflRRsYvQRm5MtOf\ni8RRRkpI+Ri/JOSjmf2SYBybQ909sq79ZSnh+aiu5g3PgSozu9TMJprZEjN7zcw2Sd3IzEaY2adm\ntp2Z/cfMFpEyEJSZHRguX2hm883sBTPbMu0YHcPuexPNbKmZTQ67tnRLL8rMdjWzUWE9X5vZ8Rm2\n6W5mT5nZLDNbZGZvm9nPcvnSZnaomX0WHv9TMzs0198w4I/A92b2ZzPbPI/9onQ48F7ae8NjgWHA\nL/M4zg0Ed64fiba8MpLwbkISO2VkI85IM6sys7VjqK98KCOlcMrHxpmP5wKj3P05C7SKqc7SS3g+\nquHd8Ay4GPg/4M/A9cDOwD/StnOCEZpfAj4kGI15OEAYaC8AC4A/AFcDWwBvpAXiv8LzDCYY+fl2\noA2QHpo9CUZ4fpWg68ts4H4z22J10cGT2reB/YC/AJcQzJH8nJn9X71f2Gx/4J8EXWv+CDxD0GVm\n+/r2S3EH8DJwGvCFmb1pZidZ0HWnvvO2NrP1cvi0zXIcA3oD72dY/S6wSbZawuPsSHDH8vckuZt5\nvoNiFDJ1hCSZMrKRZiTQCpgPzAsvzv+S436VRRkphVM+NrJ8DG9E7gi8Z2bXEbx+szC8wXFkvd+8\nEiU8H9XVvDSaA33cfRWAmc0FbjOzLd09dSCZjsBv3X1QzYLwL+ftwL3ufnrK8geB/xGE2Wlm1g7Y\nBbjA3QemHPPGDPX0AnZ39/+Gx3qKYLqlkwhCGYKgX59giqa3w+0GAZ8SzIf8bD3f90ZgarjvwnDf\nkcBQ4Nt69gPA3T8CTjSzs4CjgZOBQcDtZvYEMNjd38mw618I3pfKZgTwk3rWr0vw/9mUDOtqlnUB\nxmU5z53AY+7+rpltnENdFaqQu4+Vc7dSGoQysvFl5GTgJoJGQhXwU+AMoLeZ7eXu1TnUWSGUkVIU\n5WPjysdNCALgV8AK4AKCG5TnAI+b2Tx3fzWHOitEsvNRDe/S+HtNYIbeIPhT0wNIDc1lwANp++4H\ntCP4y7ZeynIHRgF7hz8vIZi7eC8z+7u7z62nni9qAhPA3Wea2diwnhoHAu/WBGa43SIzuxe4PkPg\nA2BmnYA+wPWpUze5+zALRqvNubtMuP8gYJAF7+ycDBwPnBIe61Z3H5yyy43AwzkcOtugFi3D/y7L\nsG5p2jYZmdlJwFbAL3Kop7Il/P0caRDKyEaWke5+adqiJ81sHMFAREcAT2YrsmIoI6U4ysfGlY9t\nwv+uC+zk7u8DmNnzwHiC98OT0/BOeD6q4V0aE9N+rvlL2z5t+fcZBoPoSRCwwzMc1wnuguHuy83s\nIuBmYJqZvUPQteghd5+Wtt+EDMeak1bPxkCmO4JjUtZnmval5snuVxnWjSUYkTFv4bsxF5nZPQTB\nuDPB4BWDU7b5EviykOOnqRmMpHmGdS3StllD2E3oeuAmd58cQT0iSaeMDDSKjKzHrQSDUO5Lkhre\nIsVRPgYaSz7WrBtf0+iG1TcungeONbOqZPUKSi41vEtjVR3L02/ZZPqLWEUQjscB6eEHwTswALj7\n7Wb2HHAocADBezwXm9ne7v5JAfWUBTNrTjBQxUkE3XuWEATn39K2a0uWpyyh5VmmcphNcKeyc4Z1\nNcvqa1BfCDQleIJT849I1/C/7cNlk919RQ61lr9CBrqooLuV0iCUkUWowIzMyN2Xmtksgic9yRFD\nRprZ7gT/1vQj+D1fPfqxma1FMLDWgQRPIecBrwF/dPcpKcdoTtDt9yiCRsIQ4Ax3n55fsRIz5WMR\nKjAfa9Zl+v9rOsH1ZWuCd/YrX8KvIdXwrjxfE4TZDHd/PdvG7j6e4KnBrRaMevkJwdQF/fM873fA\nZhmWb5Gyvq79ILjLmi7T8epkweBkJxG8o9MW+IjgHcDH3H1+hl1uJ4L3c9zdzWw0mQfy2An4xt0X\n1XP8rgR3ftPv5jpwKcE7VX0J3nWqfIUMdFFBA2NI2VNGVl5GZmRmbQgGiJqR775lLZ6MbA18TPDE\n7l9p61oB2wJXEfw7055gwKlnCQZtqnEbQeP8cIInn3cBTwO751eslDHlY4Xlo7tPMbOpwIYZVm8I\nLHX3ZDS6IfHXkGp4V54hBP8gXmJmI9K7EZlZh/D9mpZAtbunvlMynuCOWKbuLtm8BJxjZju5+6jw\nXK2BUwm6v2TqIoS7TzWzj4ETzOyGmnAws/2ALclhYAwzO4zggmErgu5LDwOD3D1bQzWq93MgGFHz\nT2a2nf8wB+NmBGF7U1q9PQDc/Ztw0e3Av9OOtwFwL8HInM8Q/H+TDAl/P0fKnjKywjIyfALVNPUd\nztAV4X9fzuH8lSOGjHT3V4BXgk1rbxw2Kg6ofTg7CxhlZhu5+6Tw6d7JwNHuPjLc5iRgjJnt6O7v\n5lewlCnlY4XlY+gJ4Hdmto+7Dwu36wAcQjAlWXIk/BpSDe8K4+4LzOx04CHgQzN7nOBpQDfgIOBN\n4HcEo0wOM7MnCZ60rgQOI2jwPVbAqW8gGFHxFTO7g6DrzIkE798clmXfiwneDXrLzP4OrAecBXzG\nD4NG1Ocggu94HPB02j8EdYrw/RyAu4HfAC+Z2c0Ev5/nEoxIOTBt29eBasKBRdz9Y4InEauldDn/\n3N2fj6jGMpHsESmlvCkjKy8jgU7AR2b2WEo9PyV4+vpSTZfp5CiLjFyHoNdVzaBZ/QiuCVdfxLv7\nWDObQDC6tRreCaB8rMh8BPgTwXzfT5vZrQQ3T35L8Hf2kohqLBNlkY+xUcM7HvXN0VzXukzLM27r\n7o+Z2fcE8xleQHD38XuCkS3vDzebCDwK7EMQNisJAuRId38m7RxZa3L36Wa2C8EdwLMIBoT4FPh5\neKe9zrrdfYgFcw1eSzDI2NcEgXsosEcd5051lrsXMjBPZNx9oZntSdDl6lKC96SGA+e5+6z0zclt\nnu5kzuWd8G5CEgllZO16G3tGzgWeJxhErT/QhGAwpT8Ct8RcesMrcUaGPQxuAB5N6WXQieBd1fQu\nt9PCddJwlI+1623s+Vjz+7crwWB3vyd4r/u/wDHu/lnc9TeohF9Dmnsyr/1FpGGZ2XbAB832vJSq\ndbrltW/13AksH3kdQL+ablgiIknSUBlpZtWkDK6Wtm4tgnfAOwN7+w/zIv+KYJqqlmnbjwJed/eL\n8ypYRCQPjeUaUk+8RURERMrEqknvsur792ot8xXFP7ALG91PEQz4+ZO0d+qnAs3MrG3aU++O4ToR\nESmSGt4iEq2ED4whIlKULBnZpOvONOm6c61l1XO/Y/mIa4s45epGdw+CJ93pA0J9QNCdeB/CwUDD\nwZ+6AW8XfGIRkXwk/BpSDW8RiVbC52AUESlKPPN4twY25YdRhnqYWR+CQaymEEwLti3wc6CpmXUM\nt5vt7ivcfb6ZDQYGmtkcgtGr7wDe0ojmItJgEn4NqYa3iETKMCzPELQKGpFSRKQYMWXk9gSDNdUM\nzFQzKN2DBFMpHRwur5lhw8Kf9wb+Ey47F1hFMPVRc4Lpyc7Mq1ARkSIk/RpSDW8RiZRZAaFZQXcr\nRUSKEUdGhnNv19c/M2vfzXCapbPDj4hIg0v6NWTFNbzNbD3gAOBbYGlpqxFJrBbAj4AhGaa6qJ+R\n/5SKlZOZZU8ZKRK7wvMRlJElpHwUiZ3ysR4V1/AmCMxHSl2ESCNxLMFcnlI5lJEiDUP5WHmUjyIN\nQ/mYQSU2vL8FOPdPf2Gj7j1LXEr0Bv95AKdceFWpy4jcX4d9U+oSYjF5yN10OeCMUpcRuaUzJzDp\n33+C8O9bXqyAbj8VdLeyAnwLsNaWv6Sq9folLiVaK8a9SNOeB5W6jFg89+fjS11CLAZccgFXXX9z\nqcuI1Lj/fcnZp54IheQjKCNL61uAtbrtR1WL9iUuJVorvn+TphvuVuoyYjHi4YtKXUIs/nDBA/xh\npgAAIABJREFUudx0862lLiNSY78cw0knHAdllI9mtjtwIdAP6Awc6u7PhevWAq4DDiSY9WEe8Brw\nR3efknKM5sBA4CiCMTCGAGe4+/R8Sq3EhvdSgI2692STLXqXupbItW7TNpHfq9UXTUpdQiyatGhD\nq87JuwGUIu+ueHG8n1NOoVkBlgJUtV6fqrU3LHUtkbK1WiTuO9XYpk/fUpcQi7Zt2yX2u1FgV+Wk\nv8NY5oJ8bNGeqlbJujFpTZol7jvV6LvddqUuIRbt2rVL7HejvPKxNcHAkoOBf6Wta0Uw48NVwKdA\ne4IZHZ4FdkzZ7jaC68zDgfnAXQSzReyeT62V2PAWkTIW04iUZROaIiLFSPqovSIihYojH939FYJZ\nGrC0g7v7fIJXUH44ntlZwCgz28jdJ5lZW+Bk4OhwIEvM7CRgjJntmM+Ui2p4i0ikYhqxt2xCU0Sk\nGHriLSKSWZnk4zoE0y3ODX/uR9BmHlazgbuPNbMJwC5AzteQWaeXEBHJixX4iVZOoQnUhKaISMMo\nj4wUESk/Jc7H8LXEG4BH3X1huLgTsDx80JNqWrguZ3riXWZ2P/DQUpcgeVhn671LXULZKfXdyrhD\nU0qnyQZ9Sl2C5OnQw48qdQllp9QZKcnUpH2ix5tJpCOP+lWpSyg7pczHcMygpwge3MQycrIa3mVm\njwN/UeoSJA/tt/5JqUuQFA0RmlI6TTqp4V1pDj1CDW+RhtCkfa9SlyB5OupoNbzztfzbt1nx3du1\nlvnyxUUfN+X6sSvwk5QHNwBTgWZm1jbtAU7HcF3O1PAWkWhlmQpi+bf/Zfm3aaG5onJCU0SkKJpO\nTEQksyz52Lz7j2ne/ce1lq2cPZ6FL19e+Cl/uH7sAezt7nPSNvkAWAnsA/w73GczoBvwNnlQw1tE\nIpVtRMrm3Xelefdday1bOWs8C16+rPBzNmBoiogUQ6Oai4hkFkc+mllrYFN+uIXZw8z6ALOBKQQz\n3GwL/BxoamYdw+1mu/sKd59vZoOBgWY2B1hAMHvOW/kOzquGt4hEKqZ5vMsmNEVEiqF3vEVEMosp\nH7cHhhO8hujALeHyBwmmoj04XP5xzSHDn/cG/hMuOxdYBfwTaE4w086ZeRWKGt4iErVCRpjMvn3Z\nhKaISFHiyUgRkcoXQz6G08jWN5NX1lm+3H0ZcHb4KZga3iISqZjm8S6b0BQRKYaeeIuIZJb0fFTD\nW0QilfTQFBEphjJSRCSzpOdj1qdEIiIiIiIiIlI4PfEWkchV0t1HEZGGpowUEcksyfmohreIREsD\nB4mI1E0ZKSKSWcLzUQ1vEYlU0t/PEREphjJSRCSzpOejGt4iEqmkh6aISDGUkSIimSU9H9XwFpFI\nJT00RUSKoYwUEcks6fmoUc1FREREREREYqQn3iISqaTfrRQRKYYyUkQks6TnoxreIhK9yslAEZGG\np4wUEckswfmohreIRMsKuPuY4JAVEalFGSkiklnC81ENbxGJVNK7CYmIFEMZKSKSWdLzUYOriUik\nakIz34+ISGMQR0aa2e5m9pyZfW9m1WZ2SIZtrjazyWa22MyGmtmmaetHhPvWfFaZ2d0Rf30RkTol\n/RpSDW8RiZYV+BERaQziycjWwMfAGYCvcUqzi4CzgFOBHYFFwBAza5aymQP3Ah2BTkBn4A/5f0ER\nkQIl/BpSXc1FREREKpi7vwK8AmCZH/+cA1zj7i+E2/QHpgGHAk+mbLfY3WfEXK6ISKOkJ94iEimj\ngG5ClXS7UkSkCA2dkWbWneAJ9rCaZe4+HxgF7JK2+bFmNsPMRpvZ9WbWsuATi4jkKenXkHriLSKR\nSvrAGCIixShBRnYi6EY+LW35tHBdjUeA74DJQG/gJqAXcEQxJxcRyVXSryHV8BaRaCV8KggRkaKU\naUa6+6CUHz83synAMDPr7u7j469ARBq9Ms3HqKjhLSKRSvrdShGRYmTLyIVjR7Jo7Mhay6qXLyrm\nlFMJLk07Uvupd0fgo3r2ezfcb1NADW8RiV3SryHV8BaRaBUywmTlZKaISHGyZGSbzfekzeZ71lq2\nbPpXTHn09wWdzt3Hm9lUYB/gUwAzawvsBNxVz659CbqoTynoxCIi+Ur4NaQa3iIiIiIVzMxaEzyZ\nrrkE7WFmfYDZ7j4RuA24zMy+Ar4FrgEmAc+G+/cAjgFeAmYBfYCBwEh3/6wBv4qISGJpVHMRiVQc\nI1Ka2e5m9pyZfW9m1WZ2SIZtrjazyWa22MyGmtmmaetHhPvWfFaZ2d0Rf30RkXrFNGrv9gTdxj8g\neEp9C/AhcBWAu98E3AncQzCaeUvgQHdfHu6/HNgXGAKMAf4MPAWskbUiInFJ+jWknniLSKRiej+n\nNfAxMBj4V4b9LwLOAvoTPM25FhhiZlukXFg6cC9wOT88FVqcV6EiIkWKIyPdfSRZHqa4+5XAlXWs\nmwTslVdRIiIRS/o1pBreIhIps+CT7z71cfdXgFeCbTNufQ5wjbu/EG7Tn2AQoUOBJ1O2W+zuM/Kr\nTkQkOnFkpIhIEiT9GlJdzUUkWvl2ESokZWudzroTzEU7rGaZu88n6E65S9rmx5rZDDMbbWbXm1nL\ngk8sIlKIBs5IEZGKkfBrSD3xFpFIleBpTieCLkDT0pZPC9fVeAT4DpgM9AZuAnoBRxR1dhGRPOiJ\nt4hIZkm/hlTDW0QiZeQ/p2JDXFO6+6CUHz83synAMDPr7u6ao1ZEGkS5ZqSISKmVaz5GdQ2phreI\nNKj5XwxnwZgRtZZVL1tUzCGnEuRuR2rfsexIMMpvXd4N99sUUMNbREREpIxV+jWkGt4iEqls3YTa\nbbU37bbau9aypVPH8d2DZxd0Pncfb2ZTgX2AT4MarC2wE3BXPbv2JeheNKWgE4uIFEBdzUVEMkv6\nNaQa3iISKasyqqry7CaUZXsza01wV7Fmwx5m1geY7e4TgduAy8zsK4KpIK4BJgHPhvv3AI4BXgJm\nAX2AgcBId/8sr2JFRIoQR0aKiCRB0q8h1fAWkUjF9DRne2A4wd1FB24Jlz8InOzuN5lZK+AeYB3g\nDeDAlPkXlwP7EkwZ0RqYCDwFXJdfpSIixdETbxGRzJJ+DamGd4k9PfhO/nHnnzj42N9w8oVXAbB0\n8WIeuu1aRo0YwoJ5s+nYpRs/P+YUDjiyf4mrbRz6dluHE3ftxpZd1qZDm+b8/vFPGTl2Zq1tzti7\nB7/Yrgtrt1iLjyfO5boXxjJx9pLV6w/r14WfbdOJzTuvTetmTdjthpEsWraqob9KSRhWwMAY9W/v\n7iPJMv2hu18JXFnHuknAXnkVJbHbdZuN+P0vd2a7Xp3otG4bfnnFP3nx7a8AaFJlXHXKnuy/4yZ0\n77wO8xct4/UPv+Xy+4YzdXbt97l22nJDBpy0Bzts0YVVq6r55KtpHPzHJ1i+onH8nSulOwfexCsv\nPstX48bSokVLtt9xZy658jo22bTX6m0G3ngtz/7rSSZ/P4lmTZuxzbZ9ueiyq+nbb4cSVl46cWSk\nJNOufTfh9yfsy3ZbdKVTh7b88rz7eHHk6Izb3nHp0Zxy2I+58OanufuxkQCss3ZLLj/9IPbZeXO6\ndlqXmXMX8vzwT7jq7hdZsGhpQ34VCW3eszsTvvtujeW/Pf1Mbr39zhJUVF6Sfg2pebxLaNxnH/Pq\n0//gR722qrV88M1X8PHbIznvhru465k3OOT4U7n3hkt5b+TQElXauLRsVsXYqQu57oWxeIb1J+22\nMUfvuBHXPD+G4+57jyXLq/nr8duyVpMf/uK3WKuKN8fNZNB/xmc8RpLV3K3M9yONT6sWzfj062mc\nc/sQ3D1tXVN6b9KR6x96k51/+3eOuuJpem20Lk9dU3vmjp223JBn/vRLhr73Dbuefj+7nvEAf3v2\nA6qrG9vfvNJ49+03Oek3Z/DC0Dd4/N8vs2LFCo457OcsWfLDjchNNu3FdX++ndf/+yHPvDKcrt1+\nxDGHHcTs2bNKWHnpKCMlV61aNuPTsZM4509P4PVE2iF792aHrTdm8vR5tZZ3Xr8dnTq05aKB/2K7\nI67j15c/zH4/3pK/XnFMzJVLXd56532+nTR19efFV4ZiZhx+5C9LXVpZSHo+6ol3iSxZvIhbLzmL\nMwfcwpP33lpr3dhPPmDvQ45kq+12BmC/w47llaceYtxnH7HDnvuVotxG5b9fzea/X80GMk9RcMxO\nXbn3P+P5z/+Ci8bL/v05r1+4Oz/ZfH1e/Xw6AI+OmgRAv43XaZCay4lZAXcrKyk1JTJD3/uGoe99\nA6z5Z2DB4uUc8scnai07985X+c9fTmDDDmvz/cwFANx4+j785en3uPXJUau3+/r7OTFXLjUefuq5\nWj/fdvcgevfciNEff8iOu+wKwP8dXvuCcsC1N/HYw/cz5vPR7Lr7Xg1VatlQRkquhv53DEP/Owao\nu3HRZf123HzhERx85l08c+fptdaN+WYqx/7h76t//m7yLK6863kGX9MfM1vjhqfEb7311qv1800v\nPE+PTTZht912L1FF5SXp+agn3iVy7/UXs8Oe+9F7p93WWLd5n+15d8SrzJ4+FYDR777FlAnj6fvj\nvRq4Skm34Tot6NCmGaO++eHCftGyVYyeNJ/eXduVsDKR5GvXpgUOzF0YdJHs0K4lO2zehVnzlvD6\n7ccz/qmzGXLLMeyy1YalLbQRmzdvLmbGOu3Xzbh+xYoV/OOB+2jXbh223Lp3A1cnkjyDrunPwAdf\nY+z4adk3Btqt3ZL5i5aq0V0GVqxYwROPPcIJJ51S6lKkgZRNw9vMzjSz8Wa2xMzeMbPEvvz1xsvP\nMH7s5xz/u0syrv/NxdexUY+enLL/dhzRrxvXnHUcp158PVv03bGBK5V067VphgOzFi6vtXz2ouV0\naNOsNEWVm/BuZT6fiuonVAKNKR/r0qxpE6759V48MexzFi1dAUD3zu0BuKT/bgx64SMOuegJPv5q\nGi/dfAzdOze+3ial5u4MuPgCdtx5V3ptvkWtda8NeYleXdejR6e2DPrbXTz275doX0fjPPGUkZFq\nzPl4wcn7s3zlSv72xH9y2n69dVrzx1//lMFPvxVzZZKLZ5/5N/PmzeO4408odSnlI+H5WBYNbzM7\nimCEuQEE86J9Agwxsw4lLSwGM6dNZvCfr+Dc6+9iraZNM27z4qODGTf6Qy6982FuefxVTjx/APdc\nfzGfjnqzgasVyV/S389paI0pH+vSpMp45IpfgMM5tw9ZvbxmypFBz3/Eo0M/Y/Q307nor8P438RZ\nnHCgnqY2tIvPP5txY8dw9+CH11i36x57M/SN93ju1f+w977789sTf8XsWTMzHCX5lJHRacz52HeL\nrpxx9J78dsA/ctq+Tavm/PuO0/n8qylcd89LMVcnuXjogb9zwE8PpFOnTqUupWwkPR/L5R3vc4F7\n3P0hADM7DTgIOBm4qZSFRe3rLz5l/pxZnH/0/ng47Fb1qlV88eE7vPT4/Tzy5pf8484buPi2v9Nv\nt30A2Ljn5oz/cjTPPPTXjF3TpeHMWrgcI3jyPXvRD0+9123djC+nLihdYWXEyP99mwrKzFJoNPmY\nSZMq49EBv2Cj9dfmwAseW/20G2DKrIUAjJ1QuwE3dsIsum6gVz8a0qUXnsPrQ4fw75eG0bFT5zXW\nt2zZko1/1IONf9SDvv12YLftt+Kxhx/gzN9fUIJqS0sZGalGm48/3nYT1m/fhnEvX7t6WZMmxo3n\nHcZZx+zNlgdfuXp565bNeP7uM5m7YAlHn3+fBp8sAxMmTOD1Ya/x5NPPlLqUspL0fCx5w9vMmgL9\ngOtrlrm7m9lrwC4lKywmfXbag9v/+XqtZXdc/ns26tGTw04+i+rqalatXEFVVZNa21RVNcGrqxuy\nVMng+7lLmblwOTv1aM+4acFFf+vmTdhmo7Y88e6kEldXHgq5+1hJdysbUmPLx3Q1je4fdV6Hn57/\n6Op3u2tMmDaPKbMW0LNr7cFqNt1oXYaM+rohS23ULr3wHIa89AJPvziUDbt2y2kfr65m+fJlMVdW\nnpSR0Wjs+fjIC6MYNurLWsteuPtMHnnhXR567p3Vy9q0as7zd5/JkqUrOOL397BipaZZLAcPPfB3\nNujYkZ8e+LNSl1JWkp6PJW94Ax2AJkD6qBDTgM0avpx4tWjViq6b1P5azVu2Yu127enaI5j3dKt+\nu/DAwKtp2qwZ63fuymfv/5fhLzzFKRdeXYqSG52WTavoum6r1X+RN2rfkl4d2zBvyQqmzV/GI+9M\n4Dd7dGfirCV8P3cJZ/5kE6bNX8bwsTNWH2Pd1s3o0KYZ3dZrhQG9OrZh0bJVTJm3lAVLV5bmizWQ\npI9I2cASnY+tWjRlky7tV/9d6965Pdv02IA5C5YwZdZCHrvyMPps2pHDLn2KpmtVsUH7VgDMnr+U\nlauCG5G3PjmKS/vvxmffTOeTr6Zx/AG96bXRuvzqyn+V6ms1KheffzbPPv0k9z/6NK1atWbG9OCP\n6tpt29GiRQuWLF7M7bfcwP4H/pyOHTsxe/Ys7r/vr0ydOoWf/99hJa6+NJSRkUl0PkIw5eIm3dZf\n/USv+4brsU2vDZkzbxGTps1l7oIltbZfsXIV02bN5+sJwfVIm1bNefGvZ9G8eVNOvORB1lm75ept\nZ8xZqAHWSsTdefihBzi+/4lUVZXFW79lI+n5WA4N70Yv/Q/MBTfdw8N3XM+tl5zNwnlzWL/LRhz/\nu0s44IjjS1Rh47Jll7YMOnE7HHDg/AN6AvD8x1MY8OwYHnhrAi2aNuGygzdn7RZr8dGEuZz5j49Z\nueqHf8CO3GFDTtuz++pjDD6pHwBXPPMFL3wytcG/k0g52q5XJ4bccizujrtzw2k/AeAfr47m+ofe\n5KBdeuLujLrnZIDV098ccP6jvDV6IgB3/et9mjdtwo2n70P7tVsw+uvpHPSHx/lu6rw6zyvRefj+\n+zAzjji49lSXA++6jyOPPo6qJk34etxYTn38EebMnsk6667Htn378czLw+m52RZ1HFVEALbbqhtD\n7v0d7uAON5wX3Kz6x/OjOO2qR9bYPr0dve0WXem31cYAfP7cACB4OugOW/x8ABOnaurFUnh92GtM\nmjiR/iecVOpSpIGVQ8N7JrAK6Ji2vCNQZwtl8J8H0LpN21rLdj/wUPY48BeRFxi3awb9s9bP66zX\ngbOvGliiauSD7+bS96rX693mbyPG87cR4+tcf8+I8dxTz/pyMuez15n72fBay1YtXVjw8ZLeTaiB\nFZSPACvGvYit1aLWsiYb9KFJpz6RFliMNz+dSOv9bqhzfX3rUg18YhQDnxiVfUOJ3KTZS+td37x5\nc+576Il6tylnz/zzCZ55unb98+cXd1NHGRmZwvPx+zexJrVnImnSvidN2veKtMBivfnBV7Tu97uc\nt099r7tm/zbb576/NIx99t2PRcsqv8v/E48/xlNPPFZr2bx5ysf6lLzh7e4rzOwDYB/gOQALHgHv\nA9xR136nXHgVm2yhUWtFitV+65/Qfuuf1Fq2eMo4vrrv9MIOWEA3oYpKzQZUaD4CNO15EFVraz5r\nkWIcesRRHHrEUbWWjf7kI366186FH1QZGYmi8nHD3ahqtX78RYok2FFH/4qjjv5VrWUfffghP96p\nX+EHTXg+lrzhHRoIPBAG6LsEo1S2Ah4oZVEikr9gRMr895E6KR9FEkQZGSnlo0iCJD0fy6Lh7e5P\nhnMuXk3QRehj4AB3n1H/niJSbpI+MEZDUz6KJIsyMjrKR5FkSXo+lkXDG8Dd7wbuLnUdIlKcpL+f\nUwrKR5HkUEZGS/kokhxJz8eyaXiLSDIk/W6liEgxlJEiIpklPR81eZyIiIiIiIhIjPTEW0SiVUA3\noYoaGUNEpBjKSBGRzBKej3riLSKRCkaktPw+pS5aRKSBxJGRZtbGzG4zs2/NbLGZvWlm26dtc7WZ\nTQ7XDzWzTeP7liIi+Uv6NaQa3iISqZqBMfL9iIg0BjFl5GCC+auPBbYGhgKvmVnn4Jx2EXAWcCqw\nI7AIGGJmzWL6miIieUv6NaQa3iISqbzvVBYwkIaISKWKOiPNrAVwGHChu7/l7t+4+1XAV8Dp4Wbn\nANe4+wvu/hnQH+gCHBrvtxURyV3SryH1jreIRCrpI1KKiBQjhoxcC2gCLEtbvgTYzcy6A52AYTUr\n3H2+mY0CdgGezKsYEZGYJP0aUk+8RURERCqUuy8E3gYuN7POZlZlZscRNKo7EzS6HZiWtuu0cJ2I\niDQANbxFJFqFvJuTw81KDR4kIokQT0YeF271PbCU4H3uR4HqOL+KiEikYrqGLBfqai4ikTIK6CaU\nW2oOBrYkGDxoCnA8weBBW7j7lJTBg/oD3wLXEgwetIW7L8+rIBGRmGTLyOkfDmX6h0NrLVu5ZFG9\nx3T38cDeZtYSaOvu08zsceAbYCrBpWlHaj/17gh8VMh3EBGJQxzXkGbWhuCa8FBgA+BD4Pfu/n7K\nNlcDvwbWAd4CTnf3r/IqJAdqeItIpAoZYTLb9imDBx3s7m+Fi68ys4MJBg+6gpTBg8J9+hNcZB6K\n3mEUkTKRLSM79tuPjv32q7VswcSxfHDLyVmP7e5LgCVm1h44ALjA3ceb2VSCUc8/DWqwtsBOwF0F\nfg0RkcjFcQ1JGT24UVdzEYlUTCNSFjR4EFAzeJCISFmIIyPNbH8zO8DMfmRm+wGvA18AD4Sb3AZc\nZmYHm9k2wEPAJODZ+L6piEh+kj7rg554i0ik4rhb6e4Lzaxm8KAvCZ5kH0PQqB6HBg8SkQoR0xOd\ndsCfgA2B2cA/gcvcfRWAu99kZq2Aewi6Ur4BHKjXcESknMSQj2U164Ma3iISKTOjKp6pII4D/k4w\neNBKgnd0HgX65VmiiEjJxJGR7v4U8FSWba4ErszrxCIiDSjqfCy3BzdqeItIg5r6/qtM/SB94KCF\nWffT4EEiIiIikqeyeXCjhreIRCpbN6HOO+xP5x32r7Vs/sSxjLrxxJyOr8GDRKSSxdTVXESk4mXL\nxynvv8rU99Me3iyt/+FNOT24yanhbWb7Z98q4O6vFl6OiFQ6I+eu47X2ybpNkEMGjAV6AjeRefCg\nrwhGpbyGBhg8SPkoIvmIKyPLlTJSRHKVLR+77HAAXXY4oNay+RO+5J0cHt6Uw4ObXJ94v5Ljdk7w\nAruINFJmUBXP05xyHTxI+SgiOYsxI8uVMlJEchJHPpbTg5tcG94toz6xiCRTjtODrbFPNmU8eJDy\nUURyFldGljFlpIjkJKZ8LJsHNzk1vN09fQh2AMysyt2roy1JRCpZY3t/UfkoIvlQRgaUkSKSLqYp\nacvmwU1VvjuYWZWZXWhmXwNLzaxHuHyAmfWPvEIRkQqhfBQRqZsyUkQas7wb3sBFwJnA9QRDstf4\nH3BaFEWJSOWyAv+XEMpHEamXMlIZKSKZJT0fC2l4nwSc6u6DgVUpyz8GNo+kKhGpWDUDY+TzqeRu\nlGmUjyJSL2WkMlJEMkt6PhYyj3dXgjuTmTQvohYRSQCjgIExKuhuZRbKRxGplzJSGSkimSU9Hwt5\n4j0W2CXD8l8Qzn8mIo1XzcAY+X4SQvkoIvVSRiojRSSzpOdjIU+8rwXuMbMNCBruPzOzzYDfEASn\niDRiVWZU5ZmC+W5fxpSPIlIvZaQyUkQyS3o+5t3wdvd/mtlcYADBwBi3Ebybc6S7vxxxfSJSaQq5\n+1g5mVkv5aOIZKWMVEaKSGYJz8dCnnjj7q8BrwGYmbm7R1qViEiFUj6KiNRNGSkijVVBDW8AM9sa\n2CL89Rfu/nlkVYlIxQret8lzYIwKuluZC+WjiNRFGamMFJHMkp6PeTe8zawT8DCwD7AkXNzCzIYD\nx7v7lAjrE5EKY+QfghWUmfVSPopINspIZaSIZJb0fCxkVPNBQHugr7u3dvfWwHZAO+C+KIsTkcpT\nMzBGvp+EUD6KSL2UkcpIEcks6flYSFfzfYDd3P2TmgXu/omZnQGMjKwyEalYlROBkVM+ikhWykhl\npIhkluR8LKThPbmO5Q5MLaIWEUkAMyvg/ZzExKzyUUTqpYzMSBkpIonPx0K6mv8RuDMcGANYPUjG\nbcBFURUmIpWpygr7JITyUUTqpYxURopIZknPx5yeeJvZFIK7kTXaA5+YWc3AGC2B5cDtwFORVigi\nUsaUjyIidVNGiogEcu1qfmWcRYhIciS9m1AGV5a6ABGpHMpIEZHMkp6POTW83f2euAsRkeSooAws\nmvJRRPKljBQRySzJ+VjI4GqrmVlV+jHcfXlRFYlIRUv63cpcKR9FJBNlZEAZKSLpkp6PeQ+uZmYt\nzexmM5tA8E7OkrSPiDRiSR8Yoz7KRxHJRhmpjBSRzJKej4WMav4n4BDgYoLQPDNcNg04ObrSRKQS\nmf1wxzL3T6mrjozyUUTqpYxURopIZknPx0Ia3r8ATnf3R4BVwGvufhlwKXB4lMWJiACY2Xgzq87w\nuTNc/0CGdS+VoFTlo4g0qArKR1BGikgjVsg73h2AceGv5xNMCwEwArgjgppEpMLFcPNxe6BJys/b\nAK8CT4Y/O/AycGLK6ZdFX0ZWykcRySrijKyUfARlpIhkUUEPsPNWyBPv8UC38NdjgcPCXx9AEKIi\n0ohVmRX0qY+7z3L36TUf4GDga3d/I2WzZe4+I2W7eXF+zzooH0WkXlFnZAXlIygjRaQecVxDllOv\noEIa3g8DO4S//jNwnpnNB+4Cbo+qMBGpTEbNOzp5fPI5vllT4FhgcNqqvcxsmpl9aWZ3m9m6kX2p\n3CkfRaRecWZkmecjKCNFpB4x5eP2QKeUz34EPYHSewV1TNnmV5F/OQroau7uN6b8+mUz25ogRL9y\n93ejLE5EKk8DTAXxC6Ad8GDKspeBpwmepmxCMFjPS2a2i7t7XsUUQfkoItnEnJFlm4+gjBSR+sWR\nj+4+K237OnsF5XXiAhQ1jzeAu4/jh/d1RKSRq7kDme8+eTgZeNndp9YscPcnU9Z/bmaGLWe3AAAg\nAElEQVSjga+BvYDh+VUTHeWjiKSLOSMrJh9BGSkitcV9DZnSK+jmtFV7mdk0YA7wOnCZu8/Or5Ls\ncmp4m9mpuR7Q3e8tvBwRqXSW5X2bcW++yFdv1n51ZvmiBbkeuxuwL3Bofdu5+3gzmwlsSswXlspH\nEclHXBlZjvkY1qWMFJGcZMvHuvbJQ0l7BeX6xPuqHLdzQKEpInXqudtB9NztoFrLZnzzBU//4chc\ndj+ZYL7Xege9MLONgPWAKQWWmQ/lo4hEpoiMLMd8BGWkiJSPkvYKyqnh7e6dozypiCRXXN2ELLil\neSLwgLtXpyxvDQwguFs5leApzo3A/4Ah+VWSP+WjiOQjjows13wEZaSI5C5bPo5740XGpfcIWlw5\nvSaLfse7VHbp3oG+W3YqdRmSo6NPuLbUJUgeqhcXPr6EUcDAGLmN2bsv0BW4P235KqA30B9YB5hM\ncEF5hbuvyKuQBBnxl5Ppu912pS5DctR+h7NKXYLkqJh8hNgyUvmYh+fvPY/e2/YtdRmSo/YH31bq\nEiRH1fMmFrV/tnzstcfP6bXHz2stm/H1Fzx14RG5HL7kvYIqtuEtIuXJyH+ewlwuQd19KNAkw/Kl\nwE/zPKWISEnEkZHKRxFJgriuIculV5Aa3iISqaCbUL4DY8RUjIhImVFGiohkFmM+lkWvIDW8RSRS\nVRZ88t1HRKQxUEaKiGQWVz6WS6+gfJ/mi4iIiIiIiEgeCmp4m9mOZjbIzIabWZdw2dFmtnO05YlI\npTH74Y5lrp8kdaNUPopIfZSRykgRySzp+Zh3w9vMDgFGAs2BXYAW4aoNgMuiK01EKpGZFfRJAuWj\niGSjjFRGikhmSc/HQp54DwDOcvfjgdSXzt8E+kVSlYhUrCryv1uZoHdelI8iUi9lpDJSRDJLej4W\nMrja5sCwDMvnAu2LK0dEKp0V0O2ngm5WZqN8FJF6KSOVkSKSWdLzsZCG93SgO/Bt2vJdgPHFFiQi\nlc3MqMp7KogKSs36KR9FpF7KSGWkiGSW9Hws5On8/cBtZtYHcGA9MzscuBm4N8riRKTyVBX4SQjl\no4jUSxmpjBSRzJKej4U88b4WaAq8TTAoxjvASuAOd781wtpERCqN8lFEpG7KSBFptPJueLt7NXC5\nmd0AbAa0AUa7+5yoixORypP093Pqo3wUkWyUkcpIEcks6flYyBNvANx9EfBhhLWISAIk/f2cXCgf\nRaQuykhlpIhklvR8zLvhbWYv1bfe3X9WeDkiUumMAu5WxlJJw1M+ikg2ysi6KSNFGrek52MhT7y/\nS/u5KbAtsCnwWNEViUhFq5lXMd99EkL5KCL1UkbWoowUkdWSno+FvON9eqblZnY9lXXTQURikPRu\nQvVRPopINsrINSkjRQSSn49RjsB+P/CbCI8nIhWoZmCMfD8Jp3wUEUAZWQdlpIgkPh+jbHhvB6yI\n8HgiIkmhfBQRqZsyUkQSr5DB1R5NXwR0BnYFboqiKBGpXEl/P6c+ykcRyUYZWXsRykgRCSU9HwsZ\nXC3961UDHwMD3f254ksSkUpnjfdVPeWjiGSljFxNGSkitSQ5H/NqeJtZE+BWYKy7z4unJBGpZFUU\ncLcylkoalvJRRHKhjFRGikhmSc/HvGp191XAG8B68ZQjIpWupptQvp9Kp3wUkVwoI5WRIpJZ0vOx\nkJsEXwBdoy5ERBLCDMvzU1FDUtZP+Sgi9VNGKiNFJLOE52MhDe8/ADeb2b5m1t7MmqV+oi5QRATA\nzLqY2cNmNtPMFpvZJ2a2Xdo2V5vZ5HD9UDPbtIHLVD6KiNRNGSkijVYhg6sNSftvuiYF1iIiCRDH\niJRmtg7wFjAMOACYCfQE5qRscxFwFtAf+Ba4FhhiZlu4+/L8KiqY8lFE6pX0UXuzUEaKSJ2Sno+F\nNLwPjLwKEUmMQnr95LD9H4EJ7v7rlGXfpW1zDnCNu78QHNP6A9OAQ4En86uoYMpHEalXTBlZKZSR\nIlKnpOdjzg1vM7sCuNnd67pLKSKCYVTlmYI5TB1xMPCKmT0J7Al8D9zt7oMAzKw70IngiTgA7j7f\nzEYBuxBzw1v5KCK5iikjy5oyUkRyEVc+mlkX4EaCm3+tgHHASe7+Yco2VwO/Bmp6WZ7u7l/lVUwW\n+bzjPQBoE+XJRSR5YhqRsgdwOjAW2J//Z+++4+Sq6j6Of34JJSQIhN6roqh0pCi9iIgghi6KgKiA\nKI+IgohUKSJdwQZSFGmCgLRIVzqEEkCaEGpCSCAkkBDSzvPHnQ27m9mdnZ07OzN3P++85gV725zZ\n7H5zz7mnwO+AcyLim6X9SwKJ7Al3e2NL++rNfJTUI/XIyBaYA8OMlFRRnfKxrSL9IdlwxdWAH1N+\nuOJ3gfWByWTDFXOde6Karuat3dwqqU/UqZvQAOChlNIvSl8/ERGfBQ4A/lJlEevBfJTUI3lnZIvM\ngWFGSqqo6MMVqx3jnfJ6Y0nFNIBgQDf3WI/cej2P3HZ9h21T33+v0mXHAM902vYMMKz0/2+S3dgt\nQcen3ksAj1UsdD7MR0kVVcrIrs7pRtPcVFZgRkrqVh3yEZpouGK1Fe/nI6Lb4EwpLVxDeSQV3Hrb\n7Mh62+zYYdtrzz3Fr769Q3en3Qt8stO2T1K6uUwpjYqIN4GtgJEAEbEAsAFwbj4lr8h8lNQITXNT\nWYEZKakR2oYrng6cSNaV/JyI+DCl9Bf6cLhitRXvY4CJeRZAUsH0optQDxo3zwTujYifkd0kbkA2\nAcZ32h1zFnBURPyPrCvlCcDrwHVVlqa3zEdJleWfkU1zU1mBGSmpe/W5h2ya4YrVVrwvTym9VZeS\nSCqEeqzBmFJ6JCK+BpwC/AIYBRySUrq83TGnRsRg4A9kM1L+B9iuD9fwNh8lVVSHjGyam8oKzEhJ\n3aqUjw//63oe7jRc8YMWGq5YTcXbsTmSKhoQ1S8F0ZPjU0o3ATdVOOZY4Niq3jwf5qOkHqmUkQ/9\n6zoevvWfHbZ98P6k7i7ZNDeV3TAjJVVUKR832ParbLDtVztse/W5pzhxn690d9mmGa7orOaScld1\nN6HW1/8+saRe6y4jy91YvvJstzeWTXNT2Q0zUlKP1OEesmmGK/a44p1SqmbNb0n9VNZNqNon3nUq\nTB8xHyX1VB0ysmluKrtiRkrqiXrcQzbTcMVqx3hLkiSpSTTTTaUkNaNmGa5oxVtSrqIXM1L2w67p\nkvqpemRks9xUSlItin4PacVbUq6CbIrdas+RpP7AjJSk8oqej1a8JeUqIogqmx+rPV6SWpUZKUnl\nFT0frXhLylVQfetj60SmJNXGjJSk8oqej1a8JeWqXut4S1IRmJGSVF7R89GKt6TctU4ESlLfMyMl\nqbwi56PrKkqSJEmSVEdWvJvMr089hcHzDOCnhx3a6KL0W19YexWuOut7vDj8l0wecQ7bb7Z6l8ee\n8/M9mDziHA7ac7M5tj913TG8fd8ZvHL7yVx5xnf4xAqL17voTSH4aDmIHr8aXWg1jXvu+Q+7fG1H\nVl5hGQbPM4Ab/nn97H0zZszg5z87nM+tvQaLLjQ/K6+wDPvv+y3GjBnTwBL3L5Xy8cjvbsdjVx/F\nuHtP5427fsUNvzuY9T6zQodjhv/pECaPOGf26/1HzuGsn+3Wlx+jocxI9dYD993Dt/YYxtqrrcTS\nQwcx/KZ/dtg/ZfJkjvzJIaz7mVVYaamF2GzDtbjkwj81qLT9zxc+szRXHbMDL/5lfybfeAjbb7Dy\n7H0DBwS/3G9jHjpvL8ZdcxAv/mV//vTjL7Lk0MEdrrHikgtw+VFf4ZXLvsubfz+QS47YjsUWnK+v\nP0rDFD0frXg3kUcefpg/n/9H1lhjzUYXpV8bPN88jHzudQ45+QpS6vq4HbdYg899dgVGvzVxjn2P\n/vcVvnvMX1hr2AnscOBvIYJ/nvf9Opa6ebTNSFntS4LsxnGNNdfi7N+cN8fPxZQpUxj5xOMc+Ytj\neODhx7jiqn/w/PPPseuwrzaotP1PpXx84ZW3+L9TrmTdXU5ky33P4JXRb/PP332fhRccMvuYlBIX\nXHMfK2x9JCtufSQrbXMkPz/r2j78FI1lRqq3pkyZzGdWX5OTTzu77M/E0Ucext133Ma5f7qY/zw0\nku8e9EN+/pP/49ZbbmxAafufwYPmZuRL4zjk3DtInQJy8KC5WWPlxTjp0gfZ8OC/sfsJ/2TVZYdy\n1TE7zj5mvnnn4oYThzErJbY9/O9scegVzDv3QK4+tv/8G1f0fHSMd5N4//332W+fb/C7P5zPySee\n0Oji9Gu33vcMt973DJC1pJWz9GILctpPdmGH75/Ltb85cI79F/3j/tn//9qbEzju3Bt48PIjWGHp\nRXhl9Nt1KXezGED1LXq2AKrNF7f9El/c9ksAc9y4LLDAAvzzpuEdtp159m/Z9Asb8Prrr7Psssv2\nWTn7q0r5eNXwER2+Pvz0a9hnp4347CeW5t+PvDB7+wdTpzF+wvt1LWuzMiPVW1tuvS1bbr0tMGc+\nAox4+EF23fObbPj5jQHYa+/9uOTPf+KxEY+wzZe279Oy9ke3jniFW0e8Asy5xNV7U6ax41H/6LDt\nR+fdyb/P3INlFp2fN8a/z0afXprlF/8Y6x/0V6Z8OAOA/U//F2OuPIDN1lyWu594vW8+SAMVPR9b\nqayF9n8/+D5f3n4HNt9iy0YXRT1w/gl7c8bFt/HcqLEVjx08aB6+9dUNGfXGeF4fO6EPStdgvWmp\nbKHWSjWXiRPfJSJYaKGFGl0UdTLXXAPYf+eNefe9D3jy+Tc67Nt9u/V49Y6TefjKIznu4B0YNO/c\nDSplA5iRqpP11t+Qf918A2+OGQ3Avf++i1Ev/Y/Nt9qmwSVTOQsOmZcEvPv+hwDMO/dAUoJpM2bN\nPubD6TOZlRKf/8wyDSplHyt4PjbFE++I2AT4CbAusBSwU0rp+u7PKo4rr7ickU88zr0PPtLooqgH\nDtvvi0ybMYPfX/Hvbo/7zq4bc+IhOzFkvnl47uWxfOXAc5k5c1a35xRB0ddg7Gv9PR+78+GHH/KL\nI49g9z2+zvzzz9/o4qjkSxt/hktO2ZfBg+ZhzLiJfOXA3zJh0pTZ+y+/+WFeHT2BMePeZfVVl+HE\nQ3biEyssztd/ckEDS913zMj8mI8dnXjqWfzkkINY59MrM9dcczFw4EB+ffbvWH/Dzze6aOpknrkH\ncsK+G3PFnc8yeep0AB56dgyTp07npG9vzNEX3cuAyMaFD4hgyYWHVLhiMRQ9H5ui4g0MAR4HLgCu\naXBZ+tTrr7/OT3/8f9x4y23MPXc/avFvUWuvthwH7bEZG+15SsVjL7vxYW67/1mWWnQB/m/vrbj0\n199mi2+dzvQZM/ugpI2TNT5WF4Mt1FjZCP02H7szY8YM9tpjV4jg7N+e1+jiqJ27Hn6e9Xc/mUUX\nmp99h32eS0/9Npt889e8/e5koONQnGdeepMx4yZx8x9+0C+G4oAZmTPzsZ0L/vBbHh3xEH+54h8s\ns+zyPHDff/jZYT9kySWXYuPNtmh08VQycEBw6ZFfBhKHnHvn7O1vT5rKXifdyDkHb8lBO67FzFmJ\nK+9+jsdfHMesWd1MOlQgRc/Hpqh4p5RuAW4BiFYaIZ+Dxx4dwbhx49ho/XVmj9eZOXMm9/zn3/z+\nvN8ycfKHLTVpQNF9fq1VWGzo/Lxw8y9nbxs4MPjVocM4+Otb8Okdjp29/f0pH/L+lA8Z9fp4Hnrq\nZcbc/Wu+uuWa/P1fjzag5GpV/TkfuzJjxgy+vseuvP7aa9x86x0+7W4yUz+czstvvM3Lb7zNI0+/\nwshrj2afnTbi9ItuK3v8I0+/TASsstxi/aLirfyYjx+ZOnUqp5xwDH++9Cq22iabJ+NTn/4MT458\ngt/99kwr3k1i4IDgbz/fnmUX+xjbHXH17Kfdbe58/DVW3/9ihs4/LzNmJd6bMo2XLv0OL98950S+\naj1NUfHuz7bcamseeezJDtu+8+19+NSnVuOwnx5hpbvJXHrDg9z+4LMdtt1w3ve59IaHuOT6B7o8\nb0AMIALmmaf4v3JFnxhDjdVW6X75pZe45bY7GTp0aKOLpAoGRHSbfWt9cjlSgjfH948bSzNS9TBj\n+nSmT5/OwIEDO2wfOHAAs2YVf5hbK2irdK+45IJ86fC/zx7bXc6E0r7N1lyWxRacjxseeKmvitlQ\nRc/H4tcCmtyQIUNY7dOfnmPbwosswqdWW61BperfBg+ah1WWX2z2mJGVllmE1VddhgkTJ/P62Hd5\n970POhw/fcZMxr49iRdfHQfACksvwq7brsNt9z/D+Anvs+ySQ/nxvl9kytRpDL/n6T7+NA3Qm6Ud\nbGBSyeTJk3nxf/+b3QNo1EsvMfKJJxi68MIstdRS7LnbzjzxxONcc+0NTJ8+nbFjswkOF154YYfr\n9IHu8vHtiZM5/Ntf4sa7R/Lm+EksstAQDth9M5ZabEGuufUxAFZcZhF23249ht/zNO9MnMzqqy7L\nrw4dxn9GvMB/X+wn67GbkeqlKZMnM+qlF2fn4ysvj+LpJ0ey0NChLLPscmz0hU05/qgjmOfUeVlu\nueW5755/c9Xll3L8yac1uOT9w+B552KVpRea/fu90lILsvpKizLhvamMeWcylx31FdZceTGGHXsd\nc881kMUXytbwfue9qcwozQH0ja0/zXOvvcO4iVPYcLWl+fX3NuOcax7lxdHvNuxz9amC56MV7ybk\nU+7GWuczyzP8jz8kJUgJTjl0GAB//eeDHHDcpXMc33lFjw+nTefza6/CQXtuztAFBvPWO+9xz4j/\nscU+Z8we41hkRZ8YQ/X16IhH2HbrLWbPVnrET38MwDe++S2O/MUx3HjDP4kINlhvLSBbUiciGH7b\nnWy8yaaNLHq/0F0+/vCkK/jkSkuw11f2Z5GFhvDOxMmMePpVttrvjNkrQEyfPpMtN/gU3//65gwZ\nNC+vj53ANbc9xq/Ov6WRH6tPmZHqrccfG8EuO3xxdj4ed9ThAOy25zc589w/8ocL/8pJx/2CH3xv\nXyZMeIdll1ueI48+gW/us3+DS94/rLPqEgw/ZRdSSqSUOGX/TQD46+3PcNKlD7D9BiuTUuLB3+4F\nZPf7KSW2PeJq7n0qW/lh1WWHcvy+X2Do/PPyythJnHLZg5x73eMN+0x9rej52LIV758e9iMWXHDB\nDtt23X1Pdt9jzwaVKD+33HpHo4vQr90z4n8MWfeHPT6+/bhugDfHT2LYD3+fc6nqZ+aE55k54YUO\n29LMab2+XlB942MrhWaraNWM3GTTzZgyretukd3tU/1Vysc9Dzu/2/PfeOtdtv3O2XkXq27yzkcw\nI5vBMUcexgILdMzHnXbZna/tsnuDStQzn994U0ZPmNrl/kUXW5wzfvuHPiyR2rvnyTcYsn3X+dbd\nvjZHX3QvR190b57FqpuZox9l5piO8xalGR90cXTPFD0fW7bifeppZ7L2Ous0uhhSyxs4dFUGDl21\nw7ZZU8Yx7fkre3W9AQQDqozBao+PiCOAk4CzUkqHlrZdCHyr06G3pJS+XNXFC8KMlGqXdz5C32Sk\nunfcSaexxlprN7oYUksbuPQ6DFy6433GrImvMe3+M3p9zaLnY1NUvCNiCPBxPmq0WDki1gTeSSm9\n1riSSWo2EfE54LvAE2V23wzsw0dZ0vXMJS3CfJSk8sxHSa2kKSrewHrAnUAqvU4vbb8Y2K9RhZLU\nC9GLeS56eHxEzA/8Fdgf+EWZQz5MKY2r8t2bnfkoFUkdM7IfMh+lIil4PjZFxTuldDetNRu8pC5E\n6U+15/TQucA/U0p3RES5ivfmETEWmADcARyVUnqnqsI0GfNRKpY6Z2S/Yj5KxVL0fDSsJOUqonev\nyteNPYC1gJ91ccjNwN7AlsBPgc2Am8JlAiQ1kXplpCS1unrnY0QcERGzIuKMdtsuLG1r/7qpHp+v\nKZ54SyqOekyMERHLAmcBW6eUppc7JqXUfrajpyPiSeBFYHOyroiS1HBFnzxIknqrnvnYDHMEWfGW\nlK8KrY933ngNd930jw7bJr83qdJV1wUWAx5t9wR7ILBpRBwMzJtSxxXVU0qjImI82cQ7VrwlNYeC\nj2GUpF6rUz42yxxBVrwl5apSt58tvzKMLb8yrMO2F/47koN32bq7y94GrN5p20XAM8ApnSvdWTli\nWWARYExPyi1JfaE3Xcftai6pP6hjPjbFHEGO8ZbU9FJKk1NK/23/AiYDb6eUnomIIRFxakRsEBEr\nRMRWwLXA88DwhhZekvpQo8cwSlIzaaY5gnziLSlXQfUzTPYy2do/5Z4JrEEWnAsBo8kq3Ed3NSZc\nkhqhnhnZDGMYJam38s7HZpsjyIq3pFwNAAZUWZPuTdeblNKW7f5/KvClXlxGkvpUvTKyWcYwSlJv\nVcrHO268hjtvvKbDtve7nyeoqeYIsuItKWfVr8HozEGS+o+6ZWRTjGGUpN7rPh+32n5nttp+5w7b\nnn/6CQ7cZauuTmmqOYKseEvKlRMHSVLX6pGR7cYwrtfFITcDVwOjgFWAk8nGMG5U7sZTkhoh73xM\nKU0G/tvx+OgwRxBwDFk+vkn2lPtX1GmOICveknIVvXiaU/3TH0lqTXlnZLONYZSk3uqje8iGzRFk\nxVuSJKlJ3H7D1dwxxxjGid2d0lRjGCWpmTVyjiAr3pJyNSB6MXGQD7wl9ROVMnKbHXZmmx3mHMP4\nnWFbdnFGc41hlKTeKvo9pBVvSTlzcjVJ6lq+GdlsYxglqfeKfQ9pxVtSrpxcTZK61kcZ2bAxjJLU\nW0W/h7TiLSlXQfVtjy2UmZJUk77IyEaOYZSk3ir6PaQVb0m5GhDBgCqbH6s9XpJalRkpSeUVPR+t\neEvKVdFbKyWpFmakJJVX9Hwc0OgCSJIkSZJUZD7xlpS/Vmp+lKS+ZkZKUnkFzkcr3pJyV/1SEJLU\nf5iRklRekfPRirekXBV9KQhJqoUZKUnlFT0frXhLylXRJ8aQpFqYkZJUXtHz0Yq3pHwVPTUlqRZm\npCSVV/B8tOItKVdR+lPtOZLUH5iRklRe0fPR5cQkSZIkSaojn3hLylXRJ8aQpFqYkZJUXtHz0Yq3\npNy1UAZKUp8zIyWpvCLnoxVvSfkq+MQYklQTM1KSyit4PlrxlpSrok+MIUm1MCMlqbyi56OTq0nK\nVfDRGJ0evypdM+KAiHgiIiaWXvdFxJc6HXN8RIyOiCkRcWtEfLx+n1KSeqceGSlJRVD0fLTiLakV\nvAYcDqwDrAvcAVwXEasBRMThwMHAd4H1gcnA8IiYpzHFlSRJkj5iV3NJuarH8JyU0o2dNh0VEQcC\nGwLPAIcAJ6SUbgCIiL2BscBOwJVVFkeS6qbgQxglqdeKno8+8ZaUr+jlq6eXjxgQEXsAg4H7ImIl\nYEng9rZjUkqTgAeBjWr+PJKUpzpnpCS1rILno0+8JeWqXhNjRMRngfuBQcB7wNdSSs9FxEZAInvC\n3d5Ysgq5JDWNok8eJEm9VfR89Im3pHxVOylGz1srnwXWJBvD/Tvgkoj4VP0+iCTVQf0yUpJaWx3y\nsZkm6PWJt6RcVcrAm669ipuvu6rDtvfem1TxuimlGcBLpS8fi4j1ycZ2n1p6yyXo+NR7CeCxHhdc\nkvpA0ccwSlJv1Skf2ybofaF0+D5kE/SulVJ6pt0EvXsDLwO/JJugd7WU0rQqi9MtK96S8lUhNb/8\ntV358td27bDtv08+zu7bbVLtOw0A5k0pjYqIN4GtgJEAEbEAsAFwbrUXlaS6suYtSeXVIR+baYJe\nK96Sml5EnATcDLwKfAzYC9gM+GLpkLPIgvR/ZK2VJwCvA9f1eWElSZLUdCJiALAbFSbojYi2CXqt\neEtqXnWaGGNx4GJgKWAi2ZPtL6aU7gBIKZ0aEYOBPwALAf8Btsu7i5Ak1arokwdJUm8VfYJeK96S\ncjV7sosqz+lOSmn/StdIKR0LHFvdO0tS36pHRkpSEdQxH9sm6F0Q2IVsgt5Nqyxezax4S8qVwxcl\nqWtmpCSVVykfb7z2Sm669u8dtr03aWLF6zbLBL1WvCXlz7tESepajhkZEQcABwIrljY9DRyfUrql\n3THHA/uTDcW5FzgwpfS//EohSTnpJh+3/9pubP+13Tps++/Ix9nlSxtX+y4NmaDXirek3DkeUZK6\nlnNGNs1SOZJUq7zvIZtpgt4BeV9Qtbni8ssaXQRVYeaE5xtdBKnfMB9bjxlZfymlG1NKt6SUXkwp\n/S+ldBTwPtlSOdBuqZyU0lNkFfClyZbKUUH84+9XNLoIqtLM0Y82ugj9RdsEvc8CtwHr0mmCXuA3\nZBP0PgjMR50m6LXi3WSuusIby1Yyc8ILjS5C02mbGKPal1SJ+dh6zMg51TMjI2JAROxBhaVyyG4u\nN8r/06lRrrXi3XJmjrHi3Vk98jGltH9KaeWU0nwppSVTSrMr3e2OOTaltHRKaXBKadt6DcWxq7mk\nXDlxkCR1rR4Z2SxL5UhSLYp+D2nFW1K+ip6aklSL+mRkUyyVI0k1Kfg9pBVvSbmK0p9qz5Gk/qBS\nRt7wjyu54R9XdthWabmcZlkqR5JqUfR7yFaseA8CeO7ZZxpdjrqYOHEijz1avDEfs6aMa3QR6iLN\nnFbIzzZr6oS2/x1U7bm9GbPtGO9cFTYji5qPYEa2klryESpn5A7DdmOHYR2Xy3l65GPstM0Xqnmb\nhiyV0wIGAbzw/LONLkfuJk2ayMjHi9mWMmvia40uQl2kGR8U7rPNen92+15d8rGrc1pFK1a8VwTY\n91vfaHAx6ufzG6zb6CKoCtOev7LyQa1rReC+ak9qoQwsohWhuBlpPraeAmfkivQiHyHfjGympXJa\nwIoAB393n8aWok623XzDygepqUy7/4xGF6FeVqQJ8rHZtGLFezjZPyovA1MbWxSpsAaRhebwBpdD\n1TMjpfpqtnxsWypnKWAi2ZPtDkvlRMRgsqVyFgL+Q52WymkB5qNUX82Wj02l5dK//vIAACAASURB\nVCreKaW3gb81uhxSP9CrlsqiT4zR7MxIqU/0Lh8h94xMKe1f6fSU0rHAsVW+a+GYj1KfaJp8bDYt\nV/GW1NyKPjGGJNXCjJSk8oqej1a8JeUq6MXEGHUpiSQ1HzNSksorej5a8ZaUq4L3EpKkmpiRklRe\n0fPRirekfBU9NSWpFmakJJVX8Hwc0OgCKBMR34+IURHxQUQ8EBGfa3SZVF5EbBIR10fEGxExKyJ2\nbHSZmkn08o/UFfOxtZiR3TMjlTczsnWYj90rej5a8W4CEbE7cDpwDLA28AQwPCIWbWjB1JUhwOPA\nQUBqcFmkQjMfW5IZKfURM7LlmI/9mF3Nm8OPgD+klC4BiIgDgO2B/YBTG1kwzSmldAtwC0BEtVNA\n9ANR/cQYLdRYqb5nPrYYM7ICM1L5MiNbiPlYQcHz0SfeDRYRcwPrAre3bUspJeA2YKNGlUvqrejl\nS+rMfFQRmZHKixmpoil6PvrEu/EWBQYCYzttHwt8su+LI9UmetFaaZuvumA+qnDMSOXIjFShFD0f\nrXhLylnBp6SUpJqYkZJUXrHz0a7mjTcemAks0Wn7EsCbfV8cqTZtrZXVvrq/ZvezgEbEhaXt7V83\n1fNzqk+YjyqcemSk+i0zUoVS9Hy04t1gKaXpwAhgq7ZtpckWtgLua1S5pCbTk1lAbya72Viy9Nqz\nb4qmejEfJalrZqTUWuxq3hzOAC6KiBHAQ2QzVA4GLmpkoVReRAwBPs5HfVtWjog1gXdSSq81rmTN\noR6dhHo4C+iHKaVxVb61mp/52GLMyO4VuyOlGsCMbCHmY/eKno9WvJtASunK0nqLx5M9sXsc2NZK\nRNNaD7iT7MlrIls/E+BisuU7+r0GdfvZPCLGAhOAO4CjUkrvNKQkyo352JLMyApaqWukmpsZ2XLM\nxwqKnI9WvJtESuk84LxGl0OVpZTuxmEaXYrSn2rPqdHNwNXAKGAV4GTgpojYqLS0ilqY+dhazMju\nNSgjVWBmZOswH7tX9Hy04i0pXxX6Cf3j75dz7d+v6LBt0sRJNb1lSunKdl8+HRFPAi8Cm5O1LEtS\ncyh6X0pJ6q2C56MVb0m5qpSZw3bZg2G77NFh28jHH+OLm22QWxlSSqMiYjzZOCor3pKaRsHvKyWp\n1+qRjxGxCfATYF1gKWCnlNL17fZfCHyr02m3pJS+XGVRKrLiLSlXvVnaIe/xPBGxLLAIMCbfK0tS\nbZohIyWpGdUpH9tWxrkAuKaLY24G9uGjevyH1ZWiZ6x4S2p63c0CWnodQzbG+83Scb8CngeG931p\nJUmS1AyaaWUcK96ScpV1E6p2YoyKupsF9CBgDWBvYCFgNFmF++jSGqeS1DTqlJGS1PIamI99sjKO\nFW9J+arDAJ0ezAL6pSrfUZIaw0HeklReY/Kxz1bGseItKXfeI0pS18xISSqvu3y85qrLueaqTivj\nTJpY0/v15co4Vrwl5cqJgySpa2akJJVXKR933m0Pdt6t48o4Tzz+KFtt3Bor41jxlpSrKP2p9hxJ\n6g/MSEkqrxnysZ4r43Q3ZlL9SESsEBGzImKN0tebRcTMiFigAWW5MyLO6Gb/MRHxWJXXnBURO9ZY\nrgsjoqtlCCQVlPnYo2uYj1I/ZUb26BpmZINExJCIWDMi1iptWrn09XKlfadGxAaln+OtgGup08o4\nVrybWOmXdFYpvD6MiBci4hcRUa+/t/YTCNwLLJVSmtSTEysFXR3kOtmB8hN81FWox69GF1otx3zs\nlvnYxMxI9QUzsltmZJOqUz6uBzwGjOCjlXEeBY4DZpKtjHMd8BzwJ+BhYNN6rIxjV/Pm17ag+yBg\nO+A8skXdT+18YClMUw0z8M3+2U0pzQDe6uV1JKkvmI8SEBGbAD8B1gWWAnZKKV3fbv+FwLc6nXZL\nSunLfVdKNYAZqX6vmVbG8Yl38/swpTQupfRaSumPwG3AVwEiYp+ImBARO0TE08BUYLnSvv0j4r8R\n8UHpvwe2v2hErB8Rj5b2PwSsTbsWwFI3oVntuwlFxBdKrZKTI+KdiLg5IhYs/YO+GXBIu9bV5Uvn\nfDYiboqI9yLizYi4JCIWaXfNwaVt70XEGxFxaLXfoIhYLyL+FRHjIuLdiLgrItYuc+jSpbJMiYgX\nI2LnTtdZNiKuKH1P346IayNihWrL099V3VLZi4mGpBLzsQLzsfnUKSOHAI8DB9H107ybgSWAJUuv\nPXP6SGpeZmQFZmRzKfo9pBXv1jMVmKf0/wkYDPwU+DbwGeCtiNgLOBb4GfAp4Ejg+Ij4JmRjHYB/\nAk8B65SOPa3Me7UP0bXIAvspYENgI7JuGQOBQ4D7ybpnLEHW2v5aRCwI3E7WtWMdYFtgcaD9tP2n\nAZsAOwBfJJu6f50qvycfAy4CPg9sQDYu46bS52zveOAqsi4llwKXR8QnS59vLrKxHBOBL5Su9R5w\nS2mfeiyq/mNHSuXEfJyT+dh08s/IlNItKaWjU0rXdXNwWyXsrdKrtjV41IrMyDmZkU2l2PeQ/jC0\nkIjYmix4zm63eS7gwJTSU+2OOxb4cekfYIBXIuIzwPeAvwB7kf2U7p9SmgY8ExHLkXVB6spPgIdT\nSj9ot+25du85DZiSUhrXbtvBwKMppV+027Y/8GpEfJxstsD9gK+nlO4q7f8W8HoPvh2zpZQ6TPUf\nEQcAu5O1oN7UbteVKaULS/9/dERsA/wAOBjYA4iU0nfbXefbwASyIL+tmjL1Zy6Vo0YwH8szH5tP\nAzNy84gYS/b3dgdwVErpnVyurKZnRpZnRjaXot9DWvFufjtExHvA3GRBdynZZABtpnUKzMHAKsAF\nEXF+u+PmIgsAyFowR5YCs839FcqxFh1bGXtiTWDLUvnbS6UyDib7XA/N3pHShIh4jipExOLAiWQh\nuThZC+p8wPKdDn2g09f3l8oIWQvmJ8qUdd5SWQ3NHupN22MLZaaai/lYgfnYfBqUkTcDVwOjyP7O\nTiZ7qrdRDWN61fzMyArMyOZS9HtIK97N7w7gAGA6MDqlNKvT/g86fT1/6b/70y6MSmbWUI7O79MT\n8wPXk3Vj6vx7MQb4RA3lae8SYChZy+OrZBOHPMBH3al6Yn7gEeDrzFnWcXMeLqkJmI+VmY8ipdS+\n0vN0RDwJvEj2NO7OsiepCMzIysxI9Rkr3s1vckppVE8PTim9FRGjgVVSSpd3cdgzwDciYp52LZYb\nVbj0SGArOraUtjeNrJWwvUeBYcArZcKeiHgRmEE2pub10rahwKrAXRXK097nybpKDS9dYzlg0TLH\nbQj8tdPXj7Yr627AuJTS+1W8tzorenOlmon5WJn52GwqZORVV1zG36/s+OM5cWK+w7FTSqMiYjzw\ncax4F5kZWZkZ2UwKfg/p5GrFdAzws4j4QUR8IrJZIfeJiB+V9v+NrKvO+RGxWkR8Gfhxmeu0/1E+\nGfhcRJwbEatHxKci4oCIWLi0/2WgbfH5thknzwUWJpuAYr2IWDkito2IP0dEpJQmAxcAv46ILSLi\ns8CFVN+q+gLwzVKZNiALxilljts1IvYtfU+OAz4H/La071JgPHBdRGwcEStGxOYRcXZELF1lefq1\n6qfFaJscQ+oT5qP52FCVsnC33b/OlVdf3+H1q1+fmW8ZIpYFFiF7cii1Z0aakQ1T9HtIK94FlFK6\ngKyb0L5krYx3ka3f+VJp/2SyGSA/S9ZKdwJZV545LtXumi+QzRi5BvAgcC+wI1lrI2QzS84E/ks2\nK+byKaUxZLM7DiCb7XEkcAYwod2Ysp8A/yHrTvSv0v+PqPIj70fWTWgEcDHZxCGd149MZP+Y7AE8\nAXwD2COl9Gzp830AbErWzejq0uf4E9n4nElVlqdfK/pSEGpt5qP52Gj1yMiIGBIRa0Y2ezTAyqWv\nlyvtOzUi2io2WwHXks3ePLy+n1atxow0Ixup6PeQ4ZwakvIQEesAI+554BHWXru61Twee+xRNt5w\nPYB1U0qPVjpeklpNPTMyIjYj6zLe+abuYrK1va8lm+BqIWA0WUXm6PazSEtSo/SXe0jHeEvKV8HH\n50hSTeqQkSmlu+m+F+OXqnxHSep7Bb+HtOItKVe9GW/TSuNzJKkWZqQklVf0fLTiLSlXzz3zTJ+c\nI0mtyIyUpPKKno+O8ZaUi4hYnmyZkcG9vMQUYLWU0qv5lUqSmoMZKUnl9Zd8tOItKTel4Cy3/mVP\njG/2wJSkWpiRklRef8hHK96SJEmSJNWR63hLkiRJklRHVrwlSZIkSaojK96SJEmSJNWRFW9JkiRJ\nkurIirckSZIkSXVkxVuSJEmSpDqy4i1JkiRJUh1Z8ZYkSZIkqY6seEuSJEmSVEdWvCVJkiRJqiMr\n3pIkSZIk1ZEVb0mSJEmS6siKtyRJkiRJdWTFW5IkSZKkOrLiLUmSJElSHVnxliRJkiSpjqx4S5Ik\nSZJUR1a8JUmSJEmqIyvekiRJkiTVkRVvSZIkSZLqyIq3JEmSJEl1ZMVbkiRJkqQ6suItSZIkSVId\nWfGWJEmSJKmOrHhLkiRJklRHczW6AJKKIyKWBxbt5enjU0qv5lkeSWomZqQkldcf8jFSSo0ug6QC\niIjlGTDXK8ya0dtLTAFWa4XglKRqmZGSVF5/yUefeEvKy6LMmsHcy29NDFq4qhPT1HeY/uptg8la\nOps6NCWpl8xISSqvX+SjFW9JuYr5FmHA4MWrOmdWRJ1KI0nNxYyUpPKKno9WvCXlKyJ7VXuOJPUH\nZqQklVfwfLTiLSlfMSB7VXuOJPUHZqQklVfwfLTiLSl/LdT6KEl9zoyUpPIKnI+t00QgSZIkSVIL\n8om3pHwVvJuQJNXEjJSk8gqej1a8JeWr4BNjSFJNzEhJKq/g+WjFW1K+InrRWtk6oSlJNTEjJam8\nguejFW9J+Sp4a6Uk1cSMlKTyCp6PVrwl5awX43Oc51FSv2FGSlJ5xc7H1impJEmSJEktyCfekvJV\n8G5CklQTM1KSyit4PlrxlpSvgk+MIUk1MSMlqbyC56MVb0n5KnhrpSTVxIyUpPIKno9WvCXlK3ox\nMUbVE2lIUosyIyWpvILnoxVvSTnrRTchWqe1UpJqY0ZKUnnFzkcr3pLyNSCyV7XnSFJ/YEZKUnkF\nz8fWeTYvSZIkSVIL8om3pHwVfHyOJNXEjJSk8gqej1a8JeUr6MWMlHUpiSQ1HzNSksoreD62ThNB\nAUTEZhExKyKGNbosbSLi5Yi4PsfrrVD6jHv34NiLImJUXu+tZjHgoxbLnr6MImFGljnWjCwkM1LV\nMx/nONZ8LKRi52PrlLRJlQKi0mtmRGxaOiU1tMBzamR5UoPfvyoRsXREXBkREyJiYkRcGxEr9fDc\n7n4+hte77H2qbQ3Gal/dXjI2iYjrI+KN0vdsx077j4mIZyLi/Yh4JyJujYj1Ox1zV5nfy/Pq8B1Q\nO2Zkze/dbN+PLtWSkaXzd4uI+0vnjy/9zn65nmVuCDNSJeZjze/dbN+PLuWQjwdHxH8jYmpEvB4R\np0fE4HqWuSEKno92Na/dNzp9/S1g69L29j8JzwCf7rRNLSIihgB3AR8DfgnMAA4F7oqItVJKEypc\novPPCcDngB8CBax4Vzs+p+KvxRDgceAC4Joy+58Dvg+8BMxH9nfzr4hYJaX0dumYBPwR+AUf/R5O\nqa6g6gUzsh+oNSMj4gfA2cA/gQuBQcA+wA0RMSyldG39St/HzEh9xHzsB3LIx18BPwGuBM4i+1n4\nQem/29Wv5A1Q8Hy04l2jlNLf2n8dERsBW6eULut8bFQ7ZmHO8+dLKX1Q00VaUER8EngxpTSjgcX4\nPrAK8LmU0qOlct0CPAX8GDiqu5M7/5yUzt+S7Bf58txL20g9aH0se043Ukq3ALdkh855cEqpw/cw\nIg4Fvg2sAdzZbteUlNK46gqnWpiR9VeEjAQOBh5KKX21bUNEXAi8QVYZKVjF24yU+dgXWj0fI2JJ\n4EfAxSmlfdttfwE4JyK2TyndWM/C96mC56NdzfteAgZExM8j4rWI+CAibouIVdofVOrSMDIi1omI\nf0fEZODEdvu3K21/PyImRcQNEfHpTtdYIiIuLL3P1IgYXerasnznQkXEFyLiwVJ5XoyIb5Y5ZqWI\nuCoi3o6IyZF1CexRN8CI2Ckinipdf2RE7NTTbxhwBPBGRPw6Ij5VxXl52hl4uC0wAVJKzwG3A7tV\ne7GImAcYBtyVUhqdWylFRMwNfA94F3ii0+69ImJcRDwZESdFxHx9X0JVYEb2z4xcAHir/YaU0nvA\n+0C/qyzUkxnZ0szH/pePGwEDgSs6bb+c7MnrHjmWs9+rdz76xLvvBfAzYCbwa2BB4HDgr2S/XG0S\nsChwE9kv1yXAWIBSoF1E1nrzU2AwcCDwn4hYO6X0auka1wCrAecArwCLA9sAywNtxwB8AriKrAvG\nRcB+wIUR8UhK6ZnSey4O3E/W/e9s4B2ypxDXR8TOKaXruvzAEV8E/k7WsncEsAhZV8LXe/Qdy8of\nwAHAjyPivlJZr0wpTe7mfYeUylvJ9JTSpG6uE2StXheU2f0QsE1EDOmuLGVsDywEXFrFOa2hQUtB\nRMT2ZL8rg4HRwDYppXfaHXIp2e/BaLK/z1OBVYFdan5z5cmM7J8ZeRewc0QcTNbdfBDZUJwFyLpW\nFocZqd4zH/tfPs5b+m/nBsi2bs7r9qCMraPg+WjFuzHmBdZMKc0EiIh3gbMi4tMppf+2O24J4Hsp\npfPbNpSC4GzgjymlA9ttvxh4HjgSOCAiFiQL4cNSSme0u+avypRnVWCTlNJ9pWtdBbwG7EsWypAF\n/WLAximl+0vHnQ+MBM4AugzN0nu+WTr3/dK5dwO3Ai93cx4AKaXHgH1KN2R7kIX6+cDZEXEFcEFK\n6YEyp/6WLNgruQvYspv9C5P9nY0ps69t29LACz14rzZ7AVOBq6s4p0X0optQPsPW7gDWJLvZ+A5w\nVUSsn1IaD9D+9wh4OiLGALdHxEopJWdGbS5mZP/LyB+Q/e6eU3oBjAO2Sik91IMythAzUjUxH/tX\nPj5HFgBfAO5ut71twr1lelDGFlLsfLTi3Rh/bgvMkv+Q/dSsDLQPzQ/JWg/b24ashfPyiFik3fYE\nPAhsUfr6A2AasHlE/Dml9G435flvW2ACpJTGR8RzpfK02Y5s/N397Y6bHBF/BE4qE/jA7LEpawIn\ntQVm6dzbI+K/ZC1LPVI6/3zg/MjG7OwHfBP4dulaZ6aU2rco/gr4Sw8uXWlitLauJB+W2Te10zEV\nRcTHgC8DN3bXStqyKkyMMXP0CGaOebTDtjS99p6kpbFrL5VeD0XE82RjdMrdKEDW0hzAxwFvKpuL\nGdn/MvIDshvM14AbyCYh+hHwj4jYOKX0Ug/K2RrMSNXGfOxH+ZhSeiwiHgQOj4jRZGOOPw2cB0zv\n7tyWVPB8tOLdGK91+rrtl3Zop+1vlJkM4hNkf9F3MqcETAJIKU2LiMOB04CxEfEA2c3MJSmlsZ3O\ne5U5TehUnhWAci2Cz7TbP0dolrYD/K/MvueAtctsr6g0NubwiPgDWTBuSDZ5xQXtjnkWeLY31++k\n7Td63jL7BnU6pid2KV2reN3MoeLEGAOXWY+By6zXYdusia8x7d7T8i7JAMr/nbVZm+x3plwrtBrL\njMz0p4z8OzAtdZxc7Xqyp0AnAnvWWsimYUaqNuZjpj/l4zCyMd4XkP39zSDrKbA5WY+D4ih4Plrx\nboyZXWzv/JNW7hdxANlf9DcojdfpZHbIppTOLt247ARsCxwP/CwitkgptZ8woKflaQoRMS/ZRBX7\nknXv+YAsOH/f6bgF6FlL4LTU/VIO75C1VC5VZl/btmomSNsLmAgUZxbK9uowPqfUPe7jfPQzuXJE\nrEn2d/M28HPgerIAXJRshuSlycadERErA18nG+/2NlkL+hnA3Smlp6orrPqAGVmDVsvIyNay3Zas\ne99sKaUJEXEPWRfL4jAjVRvzsQatlo8AKaUxwKaRTaK3JPBCSumtiHiDbIhAcRQ8H614t54XyX5w\nxqWU7qh0cGncwZnAmaVf2CfIli7Yu8r3fQX4ZJntq7Xb39V5kLWydlbuel2KbDH7fcnG6CwAPAYc\nBFzWRZfts8lhfE5KKUXEk8B6ZXZvALxUYdKg2UrdpjYn6yo2vSfnCMi+93eS3TAk4PTS9ovJJoX5\nFNnP9KJkofgw2Xiwttb0aWRrox5Ctp7ja2SBOnuWVxWGGdl6GblE6b8Dy+ybG+9VesKMVE+Yj62X\nj+2v9SLZ3yGRzUK/FPDnnpzbzzVNPvqPWesZTtYV6MiIuKtzN6KIWLQ0vmY+YFZKqf2YklHAe3Tf\ndaIrNwGHRMQGKaUHS+81BPguMKrc2ByAlNKbEfE48K2IOCVly8MQEduQjVF5udIbR8Qw4DjgM2Td\nl/4CnJ9SGlnh1LzG50DWDfLkiFgnfbQG4yfJwvbUTuVdGaCLMYl7kv2jV8xu5lBxfE6X53QjpXQ3\n3S9/uHOF818na/BQ8ZmRrZeR/wNmAbsDf2x33LLAJsC/e/D+rcOMVOOYj62Xj3OIiCidNxn4Qw/e\nv3UUPB+teLeYlNJ7EXEg2dIQj0bE5WQzvy5PtkTVPWRLsKxKNtvelWTjZmaQjRFZHLisF299Clml\n8ZaIOIese8Y+ZONvhlU492dkY4PujYg/ky0FcTDZ0hDz9+C9tyf7jN8Aru70D0GXchyfA9kkFt8B\nboqI08i+nz8i65ZyRqdj7yC7iVyZOe0FjC6FQEE1bEZKyYxswYws3ej/mWySo9vJljFagOxJxCDg\n5JzK2CTMSDWG+dh6+QgQEWeRZeHjZL2A9iJ7irt3qVJYIMXORyve9ZF6sa/c9rLHppQuK43rOAI4\njKz18Q2ymS0vLB32GvA3YCuysJlBFiC7ppSu7fQeFctUGkuyEVkL4MFkATAS+EpK6Zbuyp1SGh4R\nuwK/BE4i6yazD9m4oU2p7ODSbIMNk1J6PyI2I+ty9XOylrM7gUNTSm93Ppwy39OIWJVsMobTO+8r\nlAatwaiWYkZ2LK8Zma2x+zjZLLInlbY9BHwjpXRv3QreCGakumc+diyv+Zh1iz+EbJzxLLJs3DKl\nVKzeQFD4fIyUuvv9lqSeiYh1gBHzbPZzBiy0fFXnznr3VabdfSLAum3dsCSpSMxISSqvv+SjT7wl\n5asO43MkqTDMSEkqr+D5aMVbUr4qrMHY5TmS1B+YkZJUXsHzsXU6xUuSJEmS1IJ84i0pV0EQVbY+\nRgvNSClJtTAjJam8oudjy1W8I2IRYFuytfumNrY0UmENAlYEhpeZcbNbEb0IzRbqJtTszEip7nqd\nj2BGNpL5KNWd+diNlqt4kwXmpY0uhNRP7EW2pEjPBdUvqdg6mdkKzEipb1Sfj2BGNpb5KPUN87GM\nVqx4vwww1/LbMGDQ0AYXJX/T37iHuZfZuNHFyN1dfzm80UWoi58e9iNOPe3MRhcjd889+wz7fusb\nUPp9q0r0ovWxhUKzBbwMxczIouYjmJGtpKZ8BDOysV4G87HVmI+tw3zsXitWvKcCDBg0lAGDF2t0\nWXIXA+cp5Odae511Gl2EulhwwQUL+9lKqu6KV/RuQi2gsBlZ1HwEM7JF9aqrshnZUOZjCypqhpiP\ncyp6PjqruSRJkiRJddSKT7wlNbGiz0gpSbUwIyWpvKLnoxVvSbkqejchSaqFGSlJ5RU9H614N5mB\nQz/R6CKoCrvuvmeji9B8Cj4jpRrHfGw9ZmQZZqTqwHxsPeZjGQXPRyveTWbg0FUbXQRVYfc9DM3O\nit5aqcYxH1uPGTknM1L1YD62HvNxTkXPRyvekvJV8KUgJKkmZqQklVfwfLTiLSlXRZ8YQ5JqYUZK\nUnlFz0eXE5MkSZIkqY6seEvKVdv4nGpfFa65SURcHxFvRMSsiNix3b65IuJXETEyIt4vHXNxRCzV\n6RrzRsS5ETE+It6LiL9HxOJ1+jZIUllmpCSVV/R8tOItKV/Ry1f3hgCPAwcBqdO+wcBawHHA2sDX\ngE8C13U67ixge2BnYFNgaeDq6j6cJNXIjJSk8gqej47xlpSresxImVK6BbildGx02jcJ2LbT9Q4G\nHoyIZVNKr0fEAsB+wB4ppbtLx+wLPBMR66eUHqqqwJLUS2akJJVX9Hz0ibekXNWjm1AvLETWqvlu\n6et1yRoab287IKX0HPAqsFHeby5JXTEjJam8ouejT7wl5a4OIVjNe88LnAL8LaX0fmnzksC0Ustm\ne2NL+ySpz5iRklRekfPRirekfPVsvM2c5+Tx1hFzAVeRtVQelM9VJSlHZqQklVfwfLTiLalPTX3x\nXj586d4O22ZNm1LzddsF5nLAlu1aKgHeBOaJiAU6tVguUdonSU3BjJSk8lo9H614S8pVpfE28318\nY+b7+MYdtk0f/xITrjuilvdsC8yVgS1SShM6HTICmAFsBfyjdM4ngeWB+3v9xpJUJTNSksorej5a\n8ZaUq3rMSBkRQ4CP81GHopUjYk3gHWAM2ZIOawFfAeaOiCVKx72TUpqeUpoUERcAZ0TEBOA94Bzg\nXmfrldSXzEhJKq/o+WjFW1Ku6hGawHrAnWTjbhJwemn7xWRrL+5Q2v542yVLX28B/Lu07UfATODv\nwLxkS0t8v6qCSlKNzEhJKq/o+WjFW1Ku6rQG4910v/xhxaURU0ofAj8ovSSpIcxISSqv6PloxVtS\n/hq3EoQkNT8zUpLKK3A+VqzhS5IkSZKk3vOJt6R8RY/G28xxjiT1C2akJJVX8Hy04i0pV3WaGEOS\nCsGMlKTyip6PVrwl5arooSlJtTAjJam8ouejFW9J+Qqq7/bTOpkpSbUxIyWpvILnoxVvSbkKetFa\n2UqpKUk1MCMlqbyi56MVb0m5Kno3IUmqhRkpSeUVPR9dTkySJEmSpDryibekfBV8KQhJqokZKUnl\nFTwfrXhLylXRuwlJUi3MSEkqr+j5aMVbUr4KPiOlJNXEjJSk8gqej1a8g970kQAAIABJREFUJeWq\n6DNSSlItzEhJKq/o+WjFW1Kuit5NSJJqYUZKUnlFz0dnNZckSZIkqY584i0pVxHZq9pzJKk/MCMl\nqbyi56MVb0n56kU3oZZKTUmqhRkpSeUVPB+teEvKVdFbKyWpFmakJJVX9Hy04i0pV0H1E120UGZK\nUk3MSEkqr+j56ORqknLV1lpZ7av7a8YmEXF9RLwREbMiYscyxxwfEaMjYkpE3BoRH++0/67SuW2v\nmRFxXr6fXpK6Z0ZKUnlFz0cr3pJyFQOCAVW+YkDF9sohwOPAQUCa4z0jDgcOBr4LrA9MBoZHxDzt\nDkvAH4ElgCWBpYCf1vp5JakaZqQklVf0fLSruaSml1K6BbgFIMr3QToEOCGldEPpmL2BscBOwJXt\njpuSUhpX5+JKUp8yIyWpvGbKR594S8pVPboJdf9+sRJZ6+PtbdtSSpOAB4GNOh2+V0SMi4gnI+Kk\niJiv9+8sSdUzIyWpvKLnoxXvOvvC2qtw1Vnf48Xhv2TyiHPYfrPVuzz2nJ/vweQR53DQnpt12L7i\nMotw+Wn788rtJ/Pmv3/NJafsy2JD56930VXB7887l099YiWGfmw+Nv3Chjzy8MONLlJTCLKlIKp6\n1TY1xpJkXYDGdto+trSvzaXAN4DNgZOAbwJ/qeWNVRvzsbjMx66ZkeopM7KYzMeuFT0frXjX2eD5\n5mHkc69zyMlXkOYYVfCRHbdYg899dgVGvzWxw/b5Bs3NDecdzKyU2PY7Z7PFPqcz79xzcfXZB9S5\n5OrOVVdewRE//TG/OPo4Hnj4MdZYY0123H5bxo8f3+iiNVxft1b2VErp/JTSrSmlp1NKl5GF5tdK\nrZ1qAPOxmMzH7pmR6ikzsnjMx+4VPR8d411nt973DLfe9wzQ9Q/G0ostyGk/2YUdvn8u1/7mwA77\nNlpzFZZfamHW3/1kpkydBsD+R/+FMXefymafW5W7H36+ruVXeb85+0y+/Z3vsdc3986+Pu/33Hzz\njVx80Z/58WH9ey6athbIrrz71B28+9SdHbbN/HByLW/5JhBkE160b7FcAnism/MeKp33cWBULQVQ\n75iPxWQ+ds+MVE+ZkcVjPnav6PloxbsJnH/C3pxx8W08N6pzLweYd565SCSmzZgxe9uH02Ywa1bi\n82uvbGg2wPTp03ns0RH89IgjZ2+LCLbccmsefOD+BpasSVQIzaGrb8XQ1bfqsG3KmBf4358O7OKM\n7qWURkXEm8BWwMisCLEAsAFwbjenrk3WvWhMr95YfcJ8bC3mYw+YkcqRGdk6zMceKHg+Nk1X84j4\nfkSMiogPIuKBiPhco8vUFw7b74tMmzGD31/x77L7H3pyFJM/mMZJh+zEoHnnZvCgeTjl0K8xYECw\n5KIL9nFpBTB+/HhmzpzJ4osv0WH74ksswdg332xQqYotIoZExJoRsVZp08qlr5crfX0WcFRE7BAR\nqwOXAK8D15XOXzkijoqIdSJihdIajhcDd6eUnurrz1Mt89F8bBXmY2P054zsr/kIZmSrMR8bo5ny\nsSkq3hGxO3A6cAxZC8ITZOunLdrQgtXZ2qstx0F7bMb3jvlrl8e8/e5k9vrpBWy36WcZf+/pjPn3\nqXxsyCAef+51Zs3qZsCP1CB1Gp+zHlmXnxFkLYynA48CxwGklE4FfgP8gWwmyvmA7VJK00rnTwO2\nBoYDzwC/Bq4Cdszxo9eF+Wg+qljMyPz013wEM1LFVPR8bJau5j8C/pBSugQgIg4Atgf2A05tZMHq\n6fNrrcJiQ+fnhZt/OXvbwIHBrw4dxsFf34JP73AsAHc++Byrf/V4hi4wmBkzZ/He5Km89K8TefkN\nJ2JohEUXXZSBAwfy1lsdu3W9NXYsSyy5ZBdn9R8B3XYT6uqc7qSU7qZCQ2FK6Vjg2C72vU42E2Ur\nMh9LzMfmZz5WZkbmql/mI5iRrch8rKzo+djwindEzA2sSzY1OwAppRQRtzHn+mmFcukND3L7g892\n2HbDed/n0hse4pLrH5jj+AmTpgCw2edWZbGh83PD3U/2STnV0dxzz83a66zLnXfczld2yBq7Ukrc\neeftHHTwDxtcusbrzQyTfTEjZSsyH83HVmM+VmZG5qM/5yOYka3IfKys6PnY8Io3sCgwkPLrp32y\n74uTr8GD5mGV5Reb3Rqz0jKLsPqqyzBh4mReH/su7773QYfjp8+Yydi3J/Hiq+Nmb/vGDhvw3Kg3\nGTfhfTZcc2V+fdjOnPPXOzoco771w/87lO9+ex/WWWdd1vvc+vzm7DP5YMoUvrn3Po0uWsNVmpGy\nq3NUlvnYjvnYGszH7pmRuSl0PoIZWUTmY/eKno/NUPEutHU+szzD//hDUoKU4JRDhwHw138+yAHH\nXTrH8eXWaVx1xSU4/gc7MnSBwbwy+m1O+dMtnHvZXXUuubqzy6678fb48Rx/3NG8NXYsa6y5Ftff\nOJzFFlus0UVruKK3Vio/5mMxmY/dMyPVU2Zk8ZiP3St6PkYq91valwXIugpNAXZOKV3fbvtFwIIp\npa91On4dYEQMWZoYOE+Haw0c+gkGDl21/oVW1SY8/NtGF0FduOLyy7jqiss6bJs4cSL3/OffAOum\nlB7tyXXafjdXO+gPDFm6ut/DyaOf55nzvlfV+/UH1eZjaZ8Z2YLMyOaUVz6CGZk387H/MB+bk/lY\nvYY/8U4pTY+IEWTrp10PEFmfga2Ac7o6b+5lNmbAYFuHpFrtvsee7L7Hnh22Pfboo3x+g3UbVCK1\n6W0+ghkp5cF8bF7mo9RY5mP1Gl7xLjkDuKgUoA+RzVI5GLiokYWSVL1sRsrqz1GXzEepQMzIXJmP\nUoEUPR+bouKdUrqytObi8cASwOPAtiklZ36QWkzRJ8boa+ajVCxmZH7MR6lYip6PTVHxBkgpnQec\n1+hySKpN0SfGaATzUSoOMzJf5qNUHEXPx6apeEsqhqK3VkpSLcxISSqv6PloxVtSvnrRWtlSA3Qk\nqRZmpCSVV/B8tOItKVfZxBhVtlbWpyiS1HTMSEkqr+j5OKDRBZAkSZIkqch84i0pV0WfGEOSamFG\nSlJ5Rc9HK96SclX0iTEkqRZmpCSVV/R8tOItKVdFD01JqoUZKUnlFT0frXhLylfBZ6SUpJqYkZJU\nXsHz0Yq3pFwFvWitbKXUlKQamJGSVF7R89FZzSVJkiRJqiOfeEvKVdFnpJSkWpiRklRe0fPRirek\nXBV9YgxJqoUZKUnlFT0f7WouKVdtrZXVvipfN+aPiLMi4uWImBIR90TEep2OOT4iRpf23xoRH6/X\n55Sk3qhHRpqPkoqg6PloxVtSriKCAVW+ethaeQGwFbAX8FngVuC2iFiq9L6HAwcD3wXWByYDwyNi\nnnp8TknqjTplpPkoqeUVPR+teEvKVZ1aKwcBw4CfpJTuTSm9lFI6DvgfcGDpsEOAE1JKN6SUngL2\nBpYGdqrbh5WkKuWdkeaj/r+9+46Tq6r/P/76bBJIIySUhN67oQsYREFAEBQEBMGGiIiAoF9EVPyh\nVBuiIn4F+QpS5Uv5ohA0EFpARAGll9BCMCCkQUhCejm/P2YSdpfZnbJT776e+7gPdu/cufOZJXnn\nnjPnnCtlRdbz0Ya3pKoK3p2jU/JW/LR9gT7Agk775wG7RcSGwBrA3cseSCnNAh4CRlXprUlSj9Ug\nI81HSZmQ9XwsaXG1iNin1BOmlO6ovBxJeq+U0jsR8Q/g+xHxHDAF+Cy5UHyRXGim/P72puQfqxnz\nUVIjNXM+ghkpqXGaLR9LXdX89hKPS+R6FST1UhHQVptbQXwe+D3wH2Ax8ChwLbBjea9WdeajpJLV\nKCObNR/BjJRUoqznY6kN7wE1rUJSZhS7FcTr/xzLG/+8s8O+RfNmFz1vSmki8JGIGAAMSSlNiYjr\ngJeByeRGKI2gY6/lCOCxct9DmcxHSSWrRUY2cT6CGSmpRFnPx5Ia3imlzuPiAYiItpTS0uqWJKmV\nFVvoYu2d92XtnfftsG/mpOd44EdfLOn8KaV5wLyIGAbsC3wrpTQxIiaTW7XyyVwdMQTYBfhNBW+j\nZOajpHLUMiObLR/zNZmRkkqS9Xwse3G1iGiLiFMjYgIwPyI2yu8/IyKOrHaBklpLVPhV9LwR+0TE\nvhGxQUR8FLgHeBa4In/IBcDpEXFARGwNXAW8BtxSm3dasEbzUVK3apGRrZCP+TrNSEldyno+VrKq\n+XeArwE/IjdOfpkXgOOqUZSk1rVsfk45W4lzvFcm1/s4nlxY/hX4WEppCUBK6Tzg18Al5FajHADs\nl1JaWP132SXzUVK3apSRrZCPYEZK6kbW87HUOd7tfQk4NqV0R0Rc0G7/48AW1SlLkjpKKd0I3Fjk\nmDOBM+tRTxfMR0l11yL5CGakpDprpnyspOG9LrmeyUJW7EEtkjIg6H5hjK6ekxHmo6RumZFmpKTC\nsp6PlQw1f57CNxQ/mPykdEm917KFMcrdMsJ8lNQtM9KMlFRY1vOxkk+8zwUuiYjh5Bru+0fE5sBX\nyAWnpF6sLYK2MlOw3OObmPkoqVtmpBkpqbCs52PZDe+U0v9FxNvAGeQWxriA3Nycw1JKt1W5Pkmt\nppLex9bJzG6Zj5KKMiPNSEmFZTwfK/nEm5TSXcBdABERKaVU1aoktazcsJ8y5+e0UGgWYz5K6o4Z\naUZKKizr+VhRwxsgIkYCW+a/fzal9EzVqpLUsoLyQ7CFMrMk5qOkrpiRZqSkwrKej2U3vCNiDeBq\nYC9gXn53/4gYB3whpfRGFeuTpJZhPkpS18xISb1ZJauaXwoMA7ZPKQ1KKQ0CdiB3c/LfVbM4Sa1n\n2cIY5W4ZYT5K6pYZaUZKKizr+VjJUPO9gN1SSk8s25FSeiIiTgDuq1plklpW60Rg1ZmPkooyI81I\nSYVlOR8raXi/3sX+BEzuQS2SMiAiKlgYIzMxaz5K6pYZWZAZKSnz+VjJUPPvAr/OL4wBLF8k4wLg\nO9UqTFJraovKtowwHyV1y4w0IyUVlvV8LOkT74h4g1xv5DLDgCciYtnCGAOAhcCvgBurWqGklpL1\n3srOzEdJ5TAjzUhJhWU9H0sdan5mLYuQlC0tlIHVcGajC5DUWsxISSosy/lYUsM7pXRJrQuRpFZk\nPkpS18xIScqpZHG15SKirfM5UkoLe1SRpJaW9WFCpTIfJRViRuaYkZI6y3o+lr24WkQMiIjzI2IS\nuTk58zptknqxrC+M0R3zUVIxZqQZKamwrOdjJaua/xg4EDiNXGh+Lb9vCnB09UqT1Ioi3u2xLH1r\ndNVVYz5K6pYZaUZKKizr+VjJUPODgaNTSndHxG+Bu1JKL0XEBOBTwJVVrVBSy2mhDKw281FSUWak\nGSmpsCznYyWfeK8GvJj/fha520IA3At8pAo1SWphbREVbRlhPkrqlhlpRkoqLOv5WEnDeyKwXv77\n54FD8t/vSy5EJamqImJiRCwtsP06//gVBR4b04BSzUdJddVC+QhmpKQ6a6aMrGSo+dXATsDfgJ8B\nN0fE14BB5ObsSOrFgvLvwVjC4e8H+rT7eWvgDuCG/M8JuA04qt3pFpRXRVWYj5K6VYOMbJV8BDNS\nUjeyfg1ZdsM7pfTTdt/fFhEjyYXoSymlh6tZnKTWU4tbQaSU3ux0/AHAhJTS/e12L0gpTSvrhavM\nfJRUTLUzslXyEcxISd3L+jVkj+7jDZBSepF35+tI6uVyK1KW/5zSj41+wOeA8zs9tEdETAFmAPcA\np6eU3iqvkuoyHyV1VsuMbKV8BDNSUkdZv4YsqeEdEceWesKU0v9UXo6kVhcVLHRRZu/mwcDKdFz9\n9jbgJnLzBzcmd3uaMRExKqWUyiqmTOajpHLUOCObKh/BjJRUuqxfQ5b6ifdZJR6XAENT6sVq3VtJ\n7l6vt6WUJi/bkVK6od3jz0TEU8AEYA9gXHnVlM18lFSyGmdks+UjmJGSSpT1a8iSGt4ppTWr+aKS\nsivofn7Oi/f/hRfv77hY5IK5s0s7d8R6wN7AQd0dl1KaGBHTgU2o8YWl+SipHLXKyGbMx/zrmZGS\nSpL1a8gez/FulKsvOImttt6u0WWoROt99YbiB6lpLJr+cs3OvemHPs6mH/p4h33TJjzLjaceWsrT\njwamAN3e5iEi1gFWBd6osMyWd+nPTmCLkWZkq9j8lFsbXYJKtHDqhJqevwcZaT6W6LhTPss6m41s\ndBkq0ZHXPNroElSiGa+Mr+n5W/0aspL7eEtSl4JcsJSzlTJKKHJdoEcBV6SUlrbbPygizouIXSJi\n/YjYC7gZeAEYW633JUnVUIuMNB8lZUHWryFb9hNvSc0pNz+n3IUxSjpsb2Bd4PJO+5cA2wBHAkOB\n18mF5Q9SSovKKkSSaqxGGWk+Smp5Wb+GtOEtqaraIreV+5xiUkp3An0K7J8PfKy8V5SkxqhFRpqP\nkrIg69eQNrwlVVVUEJrlrmApSa3KjJSkwrKejxXN8Y6InSPi0ogYFxFr5fcdEREfqG55klpNRFS0\nZYX5KKk7ZqQZKamwrOdj2Q3viDgQuA9YERgF9M8/NBw4vXqlSVJrMR8lqWtmpKTerJJPvM8ATkwp\nfQFoP+n8b8COValKUstq4905OiVvjS66esxHSd0yI81ISYVlPR8rmeO9BXB3gf1vA8N6Vo6kVpdb\nkbL852SE+SipW2akGSmpsKznYyUN76nAhsArnfaPAib2tCBJrS0iaCv7VhAtlJrdMx8ldcuMNCMl\nFZb1fKzk0/nLgQsiYlsgAatGxKeA84H/qWZxklpPW4VbRpiPkrplRpqRkgrLej5W8on3uUA/4B/k\nFsV4EFgMXJhS+mUVa5PUgrI+TKgI81FSt8xIM1JSYVnPx7Ib3imlpcD3I+InwObAYOCplNKMahcn\nqfVkfZhQd8xHScWYkWakpMKyno+VfOINQEppDvBoFWuRpEwwHyWpa2akpN6o7IZ3RIzp7vGU0v6V\nlyOp1QUVDBOqSSX1Zz5KKsaM7JoZKfVuWc/HSj7x/nenn/sB2wGbAP/b44oktbRl91Us9zkZYT5K\n6pYZ2YEZKWm5rOdjJXO8jy+0PyJ+RGt1OkiqgazPz+mO+SipGDPyvcxISZD9fKzmCuyXA1+p4vkk\ntaBlK1KWu2Wc+SgJMCO7YEZKynw+Vry4WgE7AIuqeD5JLSjrw4QqZD5KAszILpiRkjKfj5UsrnZt\n513AmsAHgfOqUZSk1ha9dMSg+SipFGbku7swIyW1k+V8rOQT786/jaXA48AvUkqje16SJLUs81GS\numZGSuq1ymp4R0Qf4JfA8ymlmbUpSVIra6OCYUI1qaS+zEdJpTAjzUhJhWU9H8uqNaW0BLgfWLU2\n5Uhqdcvm55S7tTrzUVIpzEgzUlJhWc/HSjoJngXWrXYhkjIigihza6klKbtnPkrqnhlpRkoqLOP5\nWEnD+9vA+RGxd0QMi4gV2m/VLlBSa6lVb2VErBURV0fE9IiYGxFPRMQOnY45OyJezz9+Z0RsUqv3\n2QXzUVK3apGRLZKPYEZK6kbWryErWVxtbKf/dtanwlokZUAlnY/Fjo+IocADwN3AvsB0YFNgRrtj\nvgOcCBwJvAKcC4yNiC1TSgvLq6hi5qOkblU7I1soH8GMlNSNrF9DVtLw3q9aLy5JJfouMCmldEy7\nff/udMw3gHNSSn8GiIgjgSnAQcANdanSfJRUf62Sj2BGSqq/psnIkhveEfED4PyUUle9lJJEELSV\n2V1Zwj0bDwBuj4gbgN2B/wAXpZQuBYiIDYE1yPVmApBSmhURDwGjqPGFpfkoqVQ1yMimzsd8DWak\npKKyfg1ZzhzvM4DB1XphSdlUo/k5GwHHA88D+wAXAxdGxBfyj68BJHK9k+1NyT9Wa+ajpJLUICOb\nPR/BjJRUgqxfQ5Yz1Lx1loyT1DDF5uc8cudoHrnr1g775r0zu9hp24CHU0rfz//8RESMBI4Drq64\n2OoxHyWVpAYZ2ez5CGakpBJk/Rqy3DneqSZVSMqMNoK2bq6xdvroJ9npo5/ssO/V55/mp18+oLvT\nvgGM77RvPHBI/vvJ5C7sRtCxx3IE8FhJhfec+SipqBpkZCvkI5iRkorI+jVkuQ3vFyKi2+BMKa3S\ng3oktbpKbqlY/PgHgM077duc/OIYKaWJETEZ2At4EiAihgC7AL8ps5pKmY+Siqt+RrZCPoIZKamY\njF9DltvwPgOYWc0CJGVLqfdU7PycIn4JPBARp5Fb5GIX4BjgK+2OuQA4PSJeIncriHOA14Bbyqum\nYuajpKJqkJGtkI9gRkoqIuvXkOU2vK9LKU2tZgGSVExK6V8RcTDwE+D7wETgGyml69odc15EDAQu\nAYYC9wP71fEeteajpLprkXwEM1JSAzRTRpbT8HZujqSi2qL8W0GUcnxKaQwwpsgxZwJnlvXi1WE+\nSipJLTKyyfMRzEhJJcj6NaSrmkuqurLn57S+3veOJVXMjJSkwrKcjyU3vFNK5dzzW1IvlZufU25v\nZY2KqRPzUVKpzEhJKizr+VjuHG9J6laxezB29RxJ6g3MSEkqLOv5aMNbUlUFUO5HGy2UmZLUI2ak\nJBWW9Xx06I8kSZIkSTXkJ96SqioiiDLH/ZR7vCS1KjNSkgrLej7a8JZUVUH5w35aJzIlqWfMSEkq\nLOv5aMNbUlXV6h6MkpQFZqQkFZb1fLThLanqWicCJan+zEhJKizL+WjDW1JVBRXcCqImlUhS8zEj\nJamwrOejDe86u/S/z+fu2//MxAkvsGL//my34y6c/L2z2WCjTZcfs816Q4gIUkodnnvK/zuXL371\n6/UuudfZZdPV+NrHNmfb9YcxYuUBfPG/H2DsE68D0Kct+N4hW7PXyDVYf/VBzJq3iL8+O5VzbnqS\nqTPnLz/HCn3bOPvw7fjkTuuyYr82xj09me9c8yjTZy9o1Nuqm6wvjKHauena33PTHy7jjddeBWCj\nTbfgmJO+zajd937PsT8+/WRuvu4Kvnn6jzn8qOPqXWqvtdNGq/DVPTdm63VXZviQ/nzlsn9y19NT\nlj++7zZr8Lld12frdYcydGA/9vvZfTz3+uwO5/jhp7dmt81WZ8TK/ZmzYDGPTJzBT259lpenzqn3\n22kIM1KVmjl9Crf9z3k899BfWbRgHqutvQGHfecnrLPZSADuvPJCnrjnL7w99Q369OvHOpuNZN8v\nf5P1tty2wZX3TgF8evs12W2jVRg6oB8z5i7k3pfe4o9PTi54/FdGrctem63GlQ+/xm3jp9W32CaR\n9Xz0dmJ19sjD/+CzX/oq146+h99deyuLFy/iq587iPnz5i0/ZtyjE7jnkZcY9+gExj06gbPPv4i2\ntjY++vGDGlh57zFwxb48M+ltvnPNoyQ6dn4MWKEPI9cdyvm3PsOeZ93JUb/5OxuvsRJXnfjBDsed\ne8R27L3Nmhx90d858KfjWGPoAH5/wq71fBtSyxmx5tqc9O2zuHr0fVx1y728f9SH+dZxn2XiS893\nOG7c2Ft55ol/sfqItRpUae81cIU+PPufWZx+41Od0vHdx//58lv8ePSzBR8HeGrSTE659nH2/NE4\nvnDxg0TAVcd9oJZlSy1v3juzuOikw+nTbwWOOe9yvnXFWD5xwmkMXGnl5cesvu5GHPSNM/nm5WM4\n4dfXM2yNtbn01KOYM3NGAyvvvQ7aegR7b7Yalz74Kv/1p2e45l+vc+DIEey7xervOXan9VZmk9UG\n8dbcRQ2oVPXiJ951dvFVN3X4+dxf/Jbdt9uIZ596jB12zjXMVl2t41/Ie8b+mZ12/TBrrbNe3ers\nzcY9PZlxT+d6I6PTAJZ35i/m8F/+tcO+0/7wKLefvhdrDhvAGzPmMbh/Xz6z24Z89ZIH+ccLuR7L\nr//+nzxw7sfYfsNVeGziW/V5Iw3SRvk9evYACmC3j+zb4efjTzmdm669jKcf/ycbbrI5AFMnv84v\nzjmNC6+4if/68mGNKLNXu++5adz3XC7XCn3G8Kd//QeAtYcN6HL433UPTlr+/esz5nH+X57jtm/v\nzrqrDODVt+Z18azsMCNViXHX/pahw9fksFN/vHzfsDXW7nDMdnt+osPPnzjhe/xzzI288fJzbLL9\nqLrUqXdtNnww/3p1Jk/8ZxYAb855m91eH8Ymqw1kbLvjhg3sx5d2Xpcf3vkSp+29cWOKbRJZz8dW\nqjWTZs+cSUSw8tBVCj7+5vRp3D/uDj51xBfrXJlKtfLAfqQEs/K9lNuuP4y+fYK/jn93+OWEKbN5\n7a25vH/jVRtVZv3khwmVs5U9oUeZt3TpUu649SYWzJvH1tvvDEBKiTO/dRxfOPbryxviam0DVujD\npz+wHpPenMPrb88v/oQsMCNVgfF/v4d1Nt+aa848ibMP2YULjj2Qh/5yfZfHL1m8iIduvY7+g4ew\n1sZb1rFSLfP81HcYueZKrLHSigCsP2wAmw8fzGP5hvgyJ+62Prc8PYX/zOwlGdidjOdjU3ziHREf\nAk4FdgTWBA5KKY1ubFW1l1Lip2d9h+13GsXGm21R8JhbbryGwYNXYq+PHVDn6lSKFfq2cfqh2/DH\nhyYxZ8FiAIav3J+Fi5fyzvzFHY6dNms+w4f0b0SZdZX1ezDWW2/LxwnPP8uXD9uHBQvmM2jQYM67\n+Bo22HgzAK787S/p128FPn3ksQ2uUj31+Q+uz2kHbsXAFfowYeo7fOGiB1mytKvB6dliRlZPb8rH\nN994lQdHX8uHP/1l9vz8Cbz63BOM/vU59O23Ajvuc/Dy48b/Yxx/OOe/WLRgHkNWHc5XfnYlA4cM\nbWDlvdfNT01hQL8+XHDwVixNiYjgukdf5+8T3x36f9DWI1i8NDH2ud45p7uzrOdjUzS8gUHA48Bl\nwB8bXEvdnPu9k5nw4vNc/cc7uzzm5huu4ROHHEG/FVaoY2UqRZ+24NLjR5ESfOeaRxtdTtPIdT6W\nuzBGjYrJhl6Vj+tvvBl/+PPfeGf2TO6+bTRnfus4LrluDPPmzuUU3PHcAAAatElEQVT6Ky/h6lv/\nWvwkanp/+tdr/PW5aQxfuT/HfmRjLvrS+znkgr+xaEn2G99mZFX1mnxMaSnrbL4t+x59MgBrbbIl\nkye+yIOj/7dDw3uTHUZx8qW3MmfmDB7+y/Vcc9ZJnHTxTQxaufDIStXOrhsOY7eNhnHBfRN5beZ8\nNlhlAEftvC5vzV3E/S+/xYarDmC/LYfz7VvHN7rUppH1fGyKhndK6XbgdoBopaXpeuCHp5/C/ePu\n4MqbxrL6iDUKHvPIQw/w75df4ucXX13n6lRMn7bgsuNHsfYqAznkZ/cu/7QbYOrM+azQt43B/ft2\n+NR79SH9mTor+8OIsj4/p956Wz727duXtdfbAIDN37ctzz75CNdd8Vs22GhT3n5rOgfs9r7lxy5d\nsoQLfvT/+N8rLubme59oUMWqxJwFS5izYC6T3pzL46/M4MmffIx9t1mTPz/2eqNLqzkzsnp6Uz4O\nWWU4w9frOP93+Pob8/T9Yzvs67dif1Zdaz1WXWs91ttyW877wt48POZGPvKZr9azXAGf33Ft/vTU\nZB7899sAvPb2fFYfvCIHbzOC+19+iy2GD2ZI/75cfNjWy5/TFnDkTuuw/1bDOemmZxpVesNkPR+b\nouHd2/zw9FMYd8dfuPzG21hz7XW7PO6P113FVltvz6ZbbFXH6lTMskb3+qsP5uCf3cvMTitQPvHv\nGSxekvjwliMY81huoaGNR6zEOqsM5F8T3mxEyfVVwa0gWqq7UnW1dOlSFi1cwP6HfIadd/tIh8e+\nftQh7H/wEXziU59rUHXqTqmfXbdFbhnLFfq20uVTD5iRqsD6I3dk2qsvd9g37dWXGTZi7S6ekZOW\nLmXxwoW1LE1dWKFvG0s73Ro45YecA/x1wls82el2i6fvswl/nfAW417sBdeLhWQ8H21419m53zuZ\nMaP/j19fdh0DBg5i+rSpAKy00hBW7P/u/N93Zs/izjG38O0zftzVqVQjA1fow4bDBy//i7/+6oN4\n3zorM2POQqbMnM/vT9iVkesN5fO/+hv9+gSrD8ktmjFjzkIWL0m8M38x1/5tImcdvi1vz13IO/MX\n86PPbM/DL03P/IrmUk9cdP7ZjNp9b9ZYax3mvvMOt4++kccefoAvXfFHhqw8lCErd5yn2KdvX1Zd\nbTjrbdi7V4GtpwEr9GGD1QYtv85Zb9WBbLnWEN6eu5A33p7PkAH9WHvYANYY2p8ANh4+mCCYNns+\n02cvZN1VBvCJHdbm/uem8eY7C1hr6ACO33sT5i1awrhnpzb0vUnN7EOHfYmLTvo09/zhYrbdY38m\njX+Ch/9yI4d+64cALJw/j3uuuYitdt2LlVYdzpyZb/H3P13DrDenss0e+zW4+t7pkVdn8qlt1uSt\nOYt49e35bLjqAD6+1XDuyTeq5yxcwpyFSzo8Z8nSxNvzFjF59oJGlKwaa9mG93lnfZeVVhrSYd9+\nnzyM/Q9q7tvL3HDNZUQER396/w77z/n5xRx46GeX/3z7rbnbju134KF1rU+w7Qar8KdT9yDlv846\nfFsArv/7vzl/9DPsu+1aJBL3nPlRIHfLsUTi4J/dy4MvTAfg+9c9zpJPb8tlx+/Kiv3auOfpyXy3\nSeeBz5/wAAtefqDDvqUL51Z8vnosjBER3wV+BFyQUvpmft/lQOfl/29PKe3f+fm9wS/P/R6Dh3TM\nyH0+cSj7NnGmvPXmNM469XjenDqFQSsNYdMt3seFV/yRnXbdveDxGR9Z2pS2WXco1504ikTuE+3T\nD8oN/b/p4Vc59X+f4KMjR3D+Z7db/vivv7gjABfc/jwXjn2RBYuXsvNGq3D0hzdkyMB+TJ+9gIcn\nvMUhFzzAjDnN96ncnOfvY+4L93fYt3TBnB6ds9YZaT4Wd+tvfkj/wSt12Lfdngew/V7Nu5Dtuptv\nzZFnX8xtv/sZd1/9G1ZZYx0+eeLpy28h1tbWxtRJL/PIHX9izswZDBoylHW22IbjL7yOEetv0uDq\ne6ffP/Qqh2+/Jl/+wLoM6d+XGfMWccfz07npiTe6fE4rrXIx6cHbmfRQx6kOi+bO7uLo0mQ9HyOl\n5vpfHBFL6WZVyojYAXjk+jH3s9XW29W3OFVszzPGNLoElWHR9Jd5e/RpADumlErqMVj2d/Pn141l\n4622Kev1Jjz7JKccsW9JrxcROwHXAzOBcZ2CczhwFO/m8IKU0syyimlixfIxf8wOwCNX3XIvW4w0\nI1vFob9y0bhWsXDqBCZfdzKUkY9Qn4w0H0vLx69fcjPrbDayfsWpR/75Smb+mGbejFfGc9dZnwfz\nsaBeMqFKUr20ERVtpYiIwcA1wDHA2wUOWZBSmpZSmprf/NdaUlOpVUaaj5JaXdbzsSka3hExKCK2\njYhlH89slP+565XHJDWnWHY7iNK3MsYJ/Qa4NaV0TxeP7xERUyLiuYi4KCJa/v4p5qOUMbXLSPPR\nfJRaW8bzsVnmeL8fGAfLp4X9PL//SuDoRhUlqXyR/yr3OUWPiTgC2I5cXhRyG3ATMBHYGPgxMCYi\nRqVmm1NTHvNRypBaZKT5aD5KWZD1fGyKhndK6T6a5NN3Sc0nItYBLgD2TiktKnRMSumGdj8+ExFP\nAROAPchdmLUk81FSd8xH81FSYc2Wj03R8JaUHcuH/nTh3jF/5L4xf+qwb+7sWcVOuyOwOvBovLuU\ndR/gwxFxIrBi517JlNLEiJgObEILX1hKypYaZKT5KCkTsp6PNrwlVVWxhS723P9T7Ln/pzrse+nZ\nJznp03t3d9q7gK077bsCGA/8pNBQoHwv56pA1/ftkKQ6q0FGmo+SMiHr+WjDW1J1Femt7Oo53Ukp\nzQGe7fCUiDnAmyml8RExCDiD3BydyeR6KX8KvACMRZKaRZUz0nyUlBkZz0cb3pKqqtgwoa6eU4H2\nvZRLgG2AI4GhwOvkAvMHXc3pkaRGqFNGmo+SWk7W89GGt6Sqyt3ZodwVKcuXUtqz3ffzgY9VcBpJ\nqqt6ZKT5KKkVZT0fbXhLqqo2oK3MFHRJWkm9hRkpSYVlPR9bqVZJkiRJklqOn3hLqrIoe5hQZYPN\nJakVmZGSVFi289GGt6SqquPiapLUcsxISSos6/low1tSVUUFvZXl925KUmsyIyWpsKznow1vSVXV\nFhUsjNE6mSlJPWJGSlJhWc9HG96Sqizb83MkqWfMSEkqLNv56KrmkiRJkiTVkJ94S6qqrC+MIUk9\nYUZKUmFZz0cb3pKqKih/0E8LZaYk9YgZKUmFZT0fbXhLqqq2CNrK7H4s93hJalVmpCQVlvV8tOEt\nqaqy3lspST1hRkpSYVnPRxvekqqvlVJQkurNjJSkwjKcjza8JVVd+beCkKTew4yUpMKynI/eTkyS\nJEmSpBryE29JVZX1W0FIUk+YkZJUWNbz0Ya3pKrK+sIYktQTZqQkFZb1fLThLam6sp6aktQTZqQk\nFZbxfLThLamqIv9V7nMkqTcwIyWpsKzno4urSaqqZfNzyt26P2ccFxFPRMTM/Pb3iPhYp2POjojX\nI2JuRNwZEZvU8n1KUiWqnZHmo6SsyPo1pA1vSVUXZW4leBX4DrADsCNwD3BLRGwJEBHfAU4EjgV2\nBuYAYyNihSq9JUmqmipnpPkoKTOyfA1pw1tS00sp/SWldHtKaUJK6aWU0unAO8AH8od8AzgnpfTn\nlNLTwJHAWsBBDSpZkurCfJSkrjVTRtrwllRd5XZVlrmQRkS0RcQRwEDg7xGxIbAGcPeyY1JKs4CH\ngFE9fj+SVE01zEjzUVJLy/g1pIurSaqqWi2MEREjgX8A/YHZwMEppecjYhSQgCmdnjKFXJhKUtOo\nRUaaj5KyIOvXkDa8JVVVUHyhi0LPKcFzwLbAysChwFUR8eHyXkmSGqtGGWk+Smp5Wb+GtOEtqaqK\njfq57ZYbuX30/3XYN3vWzKLnTSktBl7O//hYROxMbl7OefmXHEHHHssRwGMlFy5JdVCLjDQfJWVB\n1q8hbXhLqq4iqbnfQYex30GHddg3/qnH+czHy+54bANWTClNjIjJwF7AkwARMQTYBfhNuSeVpJqq\nT0aaj5JaT8avIW14S2p6EfEj4DZgErAS8Dlgd2Cf/CEXAKdHxEvAK8A5wGvALXUvVpLqyHyUpK41\nU0ba8JZUVTVaGGM4cCWwJjCTXK/kPimlewBSSudFxEDgEmAocD+wX0ppYXnVS1Jt1SAjzUdJmZD1\na0gb3pKqK8pfGKNYZqaUjil2ipTSmcCZZb6yJNVXlTPSfJSUGRm/hrThLamqyryl4vLnSFJvYEZK\nUmFZz0cb3pKqK+upKUk9YUZKUmEZz0cb3pKqqkbzcyQpE8xISSos6/low1tSVUUF83PKns8jSS3K\njJSkwrKej22NLkCSJEmSpCzzE29JVZXx6TmS1CNmpCQVlvV8tOEtqfpaKQUlqd7MSEkqLMP5aMNb\nUtW10kIXklRvZqQkFZblfHSOd5MZc/ONjS5BZZg/4YFGl9B0li2MUe4mFTN29P81ugSVac7z9zW6\nhKZjRqoWHrv71kaXoDJNevD2RpfQdLKejza8m8xtt9jwbiULXrbh3VlUuEnF3PFnG96tZu4L9ze6\nhKZjRqoWHr/HhnermfTQ2EaX0HSyno82vCVJkiRJqiHneEuqrqwvSSlJPWFGSlJhGc9HG96Sqiry\nX+U+R5J6AzNSkgrLej62YsO7P8DLLz3f6DpqYvbsWTz71OONLqPqFk1/udEl1MTShXMz+d4Wv/2f\nZd/2L/e5lSx00UoLY7SA/gCvTHih0XVU3TuzZvHc09nLR4CFUyc0uoSaWLpgTube26K3Xl32bdn5\nCGZkg/UHmDopW38mAea/M5vXXni60WXUxIzX32l0CTWxaO5sZrwyvtFlVNWsNyYu+9Z8LKAVG94b\nAJz29WMaXEbtHL7/hxpdgsrw9ujTGl1CLW0A/L3cJ7VQBmbRBgA/+OaxDS6jNo785B6NLkFlmnzd\nyY0uoVY2oIJ8BDOygTYAuO6HpzS4jNq48KsHNboElemusz7f6BJqZQPMx/doxYb3WOBzwCvA/MaW\nImVWf3KhWf6Smxmfn9MCzEiptirPRzAjG8t8lGrLfOxGyzW8U0pvAtc2ug6pF6iwpzLb83OanRkp\n1UVF+QhmZCOZj1JdmI9d8HZikiRJkiTVUMt94i2puQUVLIxRk0okqfmYkZJUWNbz0Ya3pKrK+PQc\nSeoRM1KSCst6PjrUvElExNciYmJEzIuIByNip0bXpMIi4kMRMToi/hMRSyPiwEbX1FSiwq27Uxb5\nnUfE5fn97bcxVX5nahDzsbWYkUWYkaoyM7J1mI9FZDwfbXg3gYg4HPg5cAawPfAEMDYiVmtoYerK\nIOBx4AQgNbiWphMVfhVRyu/8NmAEsEZ++0y13pMax3xsSWZkN8xIVZMZ2XLMx25kPR8dat4cTgYu\nSSldBRARxwEfB44GzmtkYXqvlNLtwO0AEeXOROkFovz5OcUys8Tf+YKU0rQyX1nNz3xsMWZkEWak\nqsuMbCHmYxEZz0c/8W6wiOgH7AjcvWxfSikBdwGjGlWXVKkajBIq1R4RMSUinouIiyJileqcVo1i\nPiqLzEhVixmprMl6PtrwbrzVgD7AlE77p5Ab6iCpuNuAI4E9gW8DuwNj7E1ueeajVB1mZDaZkVLP\n1S0fHWouqaqigmFCPY22lNIN7X58JiKeAiYAewDjenZ2SaoeM1KSCst6PtrwbrzpwBJyE/rbGwFM\nrn85Uk91P/Dn5puu55abru+wb/asmVWtIKU0MSKmA5vgRWUrMx+VQWakqsaMVMZkOx9teDdYSmlR\nRDwC7AWMhuUT//cCLmxkbVIlivVWHnzo4Rx86OEd9j31xGPst8cHqlhDrAOsCrxRtZOq7sxHZZEZ\nqWoxI5U1Wc9HG97N4RfAFfnwfJjcCpUDgSsaWZQKi4hB5HrBlkXDRhGxLfBWSunVxlXWHCpZ6KLY\n8d39zvPbGcBN5Hr4NwF+CrwAjC2zFDUf87HFmJHdMyNVZWZkCzEfu5f1fLTh3QRSSjfk77d4Nrnh\nQY8D+3rbj6b1fnJDT1J++3l+/5Xkbt/R69VguZ7ufucnANuQWxhjKPA6ubD8QUppUdUrUV2Zjy3J\njCzCjFS1mJEtx3wsIsv5aMO7SaSULgIuanQdKi6ldB/eEaCuSvidf6xetaj+zMfWYkbWnxnZu5mR\nrcN8rL9mykcb3pKqKvJf5T5HknoDM1KSCst6PtrwllRdtZigI0lZYUZKUmEZz0cb3pKqKuOZKUk9\nYkZKUmFZz0cb3pKqqtitILp6jiT1BmakJBWW9Xy04S2pqnK9leXOz5Gk3sGMlKTCsp6PNrwlVVfW\nxwlJUk+YkZJUWMbz0eXsJUmSJEmqIT/xllR1LdT5KEl1Z0ZKUmFZzkcb3pKqKusLY0hST5iRklRY\n1vPRoeYCICLWj4ilEbFN/ufdI2JJRAxpQC3jIuIX3Tx+RkQ8VuY5l0bEgT2s6/KI+GNPztEbRIVf\nUrMyH0s6h/lYIjNSWWNGlnQOM7IEWc9HG95NLP+XdGk+vBZExIsR8f2IqNX/t9Tu+weANVNKs0p5\nYrGgq4FU/BA1QvBuj2XJW6OLVssxH7tlPjYxM1L1YEZ2y4xsUlnPR4eaN7/bgKOA/sB+wEXAAuC8\nzgfmwzSllCoNlOV/dlNKi4GpFZ5HkurBfJSkrpmRUhPxE+/mtyClNC2l9GpK6X+Au4BPAkTEUREx\nIyIOiIhngPnAuvnHjomIZyNiXv6/x7c/aUTsHBGP5h9/GNiedj2A+WFCS9sPE4qID+Z7JedExFsR\ncVtErBwRlwO7A99o17u6Xv45IyNiTETMjojJEXFVRKza7pwD8/tmR8R/IuKb5f6CIuL9EXFHREyL\niLcj4t6I2L7AoWvla5kbERMi4lOdzrNORFyf/52+GRE3R8T65dYjqW7MxyLMR6lXMyOLMCNVTza8\nW898YIX89wkYCHwb+DLwPmBqRHwOOBM4DdgC+B5wdkR8ASAiBgG3Ak8DO+SPPb/Aa7UP0e3IBfbT\nwAeAUcAtQB/gG8A/gN8BI4A1gVcjYmXgbuCR/OvsCwwHbmj3GucDHwIOAPYB9sgfW46VgCuAXYFd\ngBeAMfn32d7ZwI3ANsAfgOsiYvP8++sLjAVmAh/Mn2s2cHv+MZWo7CFCFSykIXXBfHwv87HJmJFq\nIDPyvczIJpL1fPQPQwuJiL3JBc+v2u3uCxyfUnq63XFnAqeklG7J7/p3RLwP+CpwNfA5ckOCjkkp\nLQTGR8S65IYgdeVU4J8ppZPa7Xu+3WsuBOamlKa123ci8GhK6fvt9h0DTIqITYA3gKOBz6aU7s0/\n/kXgtRJ+HcullMa1/zkijgMOJ9eDOqbdQzeklC7Pf/+DiPgocBJwInAEECmlY9ud58vADHJBflc5\nNfVulSx00UKpqaZkPhZmPjYjM1L1Z0YWZkY2m2znow3v5ndARMwG+pH7k/UH4Kx2jy/sFJgDgY2B\nyyLi0nbH9SUXAJDrwXwyH5jL/KNIHdvRsZexFNsCe+brby/laxxI7n09vPyBlGZExPOUISKGAz8k\nF5LDyfWgDgDW63Tog51+/ke+Rsj1YG5aoNYV87UamiXK+q0g1FTMxyLMx+ZjRqqOzMgizMjmkvV8\ntOHd/O4BjgMWAa+nlJZ2enxep58H5/97DO3CKG9JD+ro/DqlGAyMJjeMqfNfizeATXtQT3tXAcPI\n9TxOIrdwyIO8O5yqFIOBfwGf5b21Tnvv4epKUH7fYwtlppqL+Vic+dhkzEjVkRlZnBnZRLKej87x\nbn5zUkoTU0qvFQjM90gpTQVeBzZOKb3caft3/rDxwDYR0T5URhU59ZPAXt08vpBcL2F7j5KbM/Tv\nArXMAyYAi8nNqQEgIoYBmxV7n53sClyYUhqbUhpP7h+Y1Qoc94ECP49vV+umwLQCtXbuwVR3osJN\nKp/5WJz52GzMSNWPGVmcGdlMMp6PNryz6QzgtIg4KSI2za8KeVREnJx//FpyQ3UujYgtI2J/4JQC\n52n/R/nHwE4R8ZuI2DoitoiI4yJilfzjrwC7RMT67Vac/A2wCrkFKN4fERtFxL4R8fuIiJTSHOAy\n4GcR8ZGIGAlcTvm9qi8CX8jXtAtwDTC3wHGHRcSX8r+Ts4CdgP/OP/YHYDpwS0TsFhEbRMQeEfGr\niFirzHp6tajwS6oT89F8bCgzUk3OjDQjGybr+WjDO4NSSpeRGyb0JXK9jPcCXwRezj8+h9wKkCPJ\n9dKdQ24oz3tO1e6cL5JbMXIb4CHgAeBAcr2NkFtZcgnwLLlVMddLKb1BbnXHNnKrPT4J/AKY0e4+\nkacC95MbTnRH/vtHynzLR5MbJvQIcCW5hUM63z8ykfvH5AjgCeDzwBEppefy728e8GFyw4xuyr+P\n35GbnzOrzHpUIxHxtYiYGLlbmDwYETs1uia1FvPRfMwq81HVYEaakVnULPkY7/7ZlaTKRcQOwCN/\ne/BfbLd9eXfzePyxR9ntA+8H2DGl9GgX5z+c3D+Kx5Kbe3YycBiwWUppek9ql6Raq2VGmo+SWllv\nyUc/8ZZUdTWamnMycElK6ap8L/Nx5IaDHV3F0iWp5mqQkeajpEzIcj7a8JZUXTVYGCMi+gE7Ancv\n25cfanYXxRd1kaTmUeWMNB8lZUbG89HbiUmqqkoWuijh+NXIrXg6pdP+KcDmZb2YJDVQDTLSfJSU\nCVnPRxvekqrq+efGl72+5PPPjS9+kCRlgBkpSYVlPR9teEuqlunA3KO/+PmBFT5/Qf4cXZ17CTCi\n0/4RwOQKX0+S6qlWGWk+Smp1vSIfbXhLqoqU0qSI2JLcsJ5KTE8pTeri3Isi4hFgL3K3DSEiIv/z\nhRW+niTVTa0y0nyU1Op6Sz7a8JZUNfnQK9h4roJfAFfkA3TZ7SAGAlfU6PUkqapqmJHmo6SW1hvy\n0Ya3pJaQUrohIlYDziY3ROhxYN+U0rTGViZJjWU+SlJhzZSPkVtRXZIkSZIk1YL38ZYkSZIkqYZs\neEuSJEmSVEM2vCVJkiRJqiEb3pIkSZIk1ZANb0mSJEmSasiGtyRJkiRJNWTDW5IkSZKkGrLhLUmS\nJElSDdnwliRJkiSphmx4S5IkSZJUQza8JUmSJEmqIRvekiRJkiTV0P8HNcxPYA9M3wUAAAAASUVO\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc0c8170fd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "lr = LogisticRegression(C = 0.01, penalty = 'l1')\n",
    "lr.fit(X_train_undersample,y_train_undersample.values.ravel())\n",
    "y_pred_undersample_proba = lr.predict_proba(X_test_undersample.values)\n",
    "\n",
    "thresholds = [0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9]\n",
    "\n",
    "plt.figure(figsize=(10,10))\n",
    "\n",
    "j = 1\n",
    "for i in thresholds:\n",
    "    y_test_predictions_high_recall = y_pred_undersample_proba[:,1] > i\n",
    "    \n",
    "    plt.subplot(3,3,j)\n",
    "    j += 1\n",
    "    \n",
    "    # Compute confusion matrix\n",
    "    cnf_matrix = confusion_matrix(y_test_undersample,y_test_predictions_high_recall)\n",
    "    np.set_printoptions(precision=2)\n",
    "\n",
    "    print(\"Recall metric in the testing dataset: \", cnf_matrix[1,1]/(cnf_matrix[1,0]+cnf_matrix[1,1]))\n",
    "\n",
    "    # Plot non-normalized confusion matrix\n",
    "    class_names = [0,1]\n",
    "    plot_confusion_matrix(cnf_matrix\n",
    "                          , classes=class_names\n",
    "                          , title='Threshold >= %s'%i) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "cdf08467-7947-a71b-3d00-7b060905bbdb"
   },
   "source": [
    "### The pattern is very clear: the more you lower the required probability to put a certain in the class \"1\" category, more records will be put in that bucket.\n",
    "#### This implies an increase in recall (we want all the \"1\"s), but at the same time, a decrease in precision (we misclassify many of the other class).\n",
    "### Therefore, even though recall is our goal metric (do not miss a fraud transaction), we also want to keep the model being accurate as a whole.\n",
    "- There is an option I think could be quite interesting to tackle this. We could assing cost to misclassifications, but being interested in classifying \"1s\" correctly, the cost for misclassifying \"1s\" should be bigger than \"0\" misclassifications. After that, the algorithm would select the threshold which minimises the total cost. A drawback I see is that we have to manually select the weight of each cost... therefore, I will leave this know as a thought.\n",
    "- Going back to the threshold changing, there is an option which is the Precisio-Recall curve. By visually seeing the performance of the model depending on the threshold we choose, we can investigate a sweet spot where recall is high enough whilst keeping a high precision value.\n",
    "\n",
    "### Investigate Precision-Recall curve and area under this curve."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "_cell_guid": "c73b727b-6743-add2-a45f-a83ce5b8b375"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdMAAAHUCAYAAABh+8IVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nwWcH/3z5AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fc0da0b20f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from itertools import cycle\n",
    "\n",
    "lr = LogisticRegression(C = 0.01, penalty = 'l1')\n",
    "lr.fit(X_train_undersample,y_train_undersample.values.ravel())\n",
    "y_pred_undersample_proba = lr.predict_proba(X_test_undersample.values)\n",
    "\n",
    "thresholds = [0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9]\n",
    "colors = cycle(['navy', 'turquoise', 'darkorange', 'cornflowerblue', 'teal', 'red', 'yellow', 'green', 'blue','black'])\n",
    "\n",
    "plt.figure(figsize=(5,5))\n",
    "\n",
    "j = 1\n",
    "for i,color in zip(thresholds,colors):\n",
    "    y_test_predictions_prob = y_pred_undersample_proba[:,1] > i\n",
    "    \n",
    "    precision, recall, thresholds = precision_recall_curve(y_test_undersample,y_test_predictions_prob)\n",
    "    \n",
    "    # Plot Precision-Recall curve\n",
    "    plt.plot(recall, precision, color=color,\n",
    "                 label='Threshold: %s'%i)\n",
    "    plt.xlabel('Recall')\n",
    "    plt.ylabel('Precision')\n",
    "    plt.ylim([0.0, 1.05])\n",
    "    plt.xlim([0.0, 1.0])\n",
    "    plt.title('Precision-Recall example')\n",
    "    plt.legend(loc=\"lower left\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "96c3a75c-0333-e22f-958e-036c0e0d337d"
   },
   "source": [
    "# Upcoming updates: \n",
    "## testing SVMs\n",
    "## testing decision trees"
   ]
  }
 ],
 "metadata": {
  "_change_revision": 4,
  "_is_fork": false,
  "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": 1
}