{ "cells": [ { "cell_type": "markdown", "metadata": { "toc": true }, "source": [ "

Table of Contents

\n", "
" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 1, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# code for loading the format for the notebook\n", "import os\n", "\n", "# path : store the current path to convert back to it later\n", "path = os.getcwd()\n", "os.chdir(os.path.join('..', 'notebook_format'))\n", "\n", "from formats import load_style\n", "load_style(plot_style = False)" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Ethen 2019-01-06 15:25:09 \n", "\n", "CPython 3.6.4\n", "IPython 6.4.0\n", "\n", "numpy 1.14.2\n", "pandas 0.23.0\n", "matplotlib 2.2.2\n", "sklearn 0.19.1\n" ] } ], "source": [ "os.chdir(path)\n", "\n", "# 1. magic for inline plot\n", "# 2. magic to print version\n", "# 3. magic so that the notebook will reload external python modules\n", "# 4. magic to enable retina (high resolution) plots\n", "# https://gist.github.com/minrk/3301035\n", "%matplotlib inline\n", "%load_ext watermark\n", "%load_ext autoreload\n", "%autoreload 2\n", "%config InlineBackend.figure_format = 'retina'\n", "\n", "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "from sklearn import metrics\n", "from sklearn.linear_model import LogisticRegression\n", "\n", "%watermark -a 'Ethen' -d -t -v -p numpy,pandas,matplotlib,sklearn" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Logistic Regression" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Logistic regression** is an excellent tool to know for classification problems, which are problems where the output value that we wish to predict only takes on only a small number of discrete values. Here we'll focus on the binary classification problem, where the output can take on only two distinct classes. To make our examples more concrete, we will consider the Glass dataset." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
rinamgalsikcabafeglass_typehousehold
id
221.5196614.773.750.2972.020.039.000.00.0010
1851.5111517.380.000.3475.410.006.650.00.0061
401.5221314.213.820.4771.770.119.570.00.0010
391.5221314.213.820.4771.770.119.570.00.0010
511.5232013.723.720.5171.750.0910.060.00.1610
\n", "
" ], "text/plain": [ " ri na mg al si k ca ba fe glass_type \\\n", "id \n", "22 1.51966 14.77 3.75 0.29 72.02 0.03 9.00 0.0 0.00 1 \n", "185 1.51115 17.38 0.00 0.34 75.41 0.00 6.65 0.0 0.00 6 \n", "40 1.52213 14.21 3.82 0.47 71.77 0.11 9.57 0.0 0.00 1 \n", "39 1.52213 14.21 3.82 0.47 71.77 0.11 9.57 0.0 0.00 1 \n", "51 1.52320 13.72 3.72 0.51 71.75 0.09 10.06 0.0 0.16 1 \n", "\n", " household \n", "id \n", "22 0 \n", "185 1 \n", "40 0 \n", "39 0 \n", "51 0 " ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "url = 'https://archive.ics.uci.edu/ml/machine-learning-databases/glass/glass.data'\n", "col_names = ['id', 'ri', 'na', 'mg', 'al', 'si', 'k', 'ca', 'ba', 'fe', 'glass_type']\n", "glass = pd.read_csv(url, names = col_names, index_col = 'id')\n", "glass.sort_values('al', inplace = True)\n", "\n", "# convert the glass type into binary outcome\n", "# types 1, 2, 3 are window glass\n", "# types 5, 6, 7 are household glass\n", "glass['household'] = glass['glass_type'].map({1: 0, 2: 0, 3: 0, 5: 1, 6: 1, 7: 1})\n", "glass.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Our task is to predict the `household` column using the `al` column. Let's visualize the relationship between the input and output and also train the logsitic regression to see the outcome that it produces." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "logreg = LogisticRegression(C = 1e9)\n", "X = glass['al'].values.reshape(-1, 1) # sklearn doesn't accept 1d-array, convert it to 2d\n", "y = np.array(glass['household'])\n", "logreg.fit(X, y)\n", "\n", "# predict the probability that each observation belongs to class 1\n", "# The first column indicates the predicted probability of class 0, \n", "# and the second column indicates the predicted probability of class 1\n", "glass['household_pred_prob'] = logreg.predict_proba(X)[:, 1]" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "scrolled": false }, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "image/png": { "height": 377, "width": 504 } }, "output_type": "display_data" } ], "source": [ "# plot the predicted probability (familiarize yourself with the S-shape)\n", "# change default figure and font size\n", "plt.rcParams['figure.figsize'] = 8, 6 \n", "plt.rcParams['font.size'] = 12\n", "\n", "plt.scatter(glass['al'], glass['household'])\n", "plt.plot(glass['al'], glass['household_pred_prob'])\n", "plt.xlabel('al')\n", "plt.ylabel('household')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "As we can see, logistic regression can output the probabilities of observation belonging to a specific class and these probabilities can be converted into class predictions by choosing a cutoff value (e.g. probability higher than 0.5 is classified as class 1)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Logistic Function" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "In **Logistic Regression**, the log-odds of a categorical response being \"true\" (1) is modeled as a linear combination of the features:\n", "\n", "\\begin{align*}\n", " \\log \\left({p\\over 1-p}\\right) &= w_0 + w_1x_1, ..., w_jx_j \\nonumber \\\\\n", " &= w^Tx \\nonumber\n", "\\end{align*}\n", "\n", "Where:\n", "\n", "- $w_{0}$ is the intercept term, and $w_1$ to $w_j$ represents the parameters for all the other features (a total of j features).\n", "- By convention of we can assume that $x_0 = 1$, so that we can re-write the whole thing using the matrix notation $w^Tx$.\n", "\n", "This is called the **logit function**. The equation can be re-arranged into the **logistic function**:\n", "\n", "$$p = \\frac{e^{w^Tx}} {1 + e^{w^Tx}}$$\n", "\n", "Or in the more commonly seen form:\n", "\n", "$$h_w(x) = \\frac{1}{ 1 + e^{-w^Tx} }$$ \n", "\n", "Let's take a look at the plot of the function:" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "image/png": { "height": 376, "width": 488 } }, "output_type": "display_data" } ], "source": [ "x_values = np.linspace(-5, 5, 100)\n", "y_values = [1 / (1 + np.exp(-x)) for x in x_values]\n", "plt.plot(x_values, y_values)\n", "plt.title('Logsitic Function')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The **logistic function** has some nice properties. The y-value represents the probability and it is always bounded between 0 and 1, which is want we wanted for probabilities. For an x value of 0 you get a 0.5 probability. Also as you get more positive x value you get a higher probability, on the other hand, a more negative x value results in a lower probability.\n", "\n", "Toy sample code of how to predict the probability given the data and the weight is provided below." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "def predict_probability(data, weights):\n", " \"\"\"probability predicted by the logistic regression\"\"\"\n", " score = np.dot(data, weights)\n", " predictions = 1 / (1 + np.exp(-score))\n", " return predictions" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Interpreting the Intercept" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can check logistic regression's coefficient does in fact generate the log-odds." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[0.65638445]\n" ] }, { "data": { "text/plain": [ "array([0.65638445])" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# compute predicted log-odds for al = 2 using the equation\n", "# convert log-odds to odds\n", "# convert odds to probability\n", "logodds = logreg.intercept_ + logreg.coef_[0] * 2\n", "odds = np.exp(logodds)\n", "prob = odds / (1 + odds)\n", "print(prob)\n", "\n", "logreg.predict_proba(2)[:, 1]" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "a1 [4.18040386]\n" ] } ], "source": [ "# examine the coefficient for al\n", "print('a1', logreg.coef_[0])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Interpretation:** 1 unit increase in `al` is associated with a 4.18 unit increase in the log-odds of the observation being classified as `household 1`. We can confirm that again by doing the calculation ourselves." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[0.99205808]\n" ] }, { "data": { "text/plain": [ "array([0.99205808])" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# increasing al by 1 (so that al now becomes 3)\n", "# increases the log-odds by 4.18\n", "logodds = logodds + logreg.coef_[0]\n", "odds = np.exp(logodds)\n", "prob = odds / (1 + odds)\n", "print(prob)\n", "\n", "logreg.predict_proba(3)[:, 1]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Defining The Cost Function" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "When utilizing logistic regression, we are trying to learn the $w$ values in order to maximize the probability of correctly classifying our glasses. Let's say someone did give us some $w$ values of the logistic regression model, how would we determine if they were good values or not? What we would hope is that for the household of class 1, the probability values are close to 1 and for the household of class 0 the probability is close to 0.\n", "\n", "But we don't care about getting the correct probability for just one observation, we want to correctly classify all our observations. If we assume our data are independent and identically distributed (think of it as all of them are treated equally), we can just take the product of all our individually calculated probabilities and that becomes the objective function we want to maximize. So in math: \n", "\n", "$$\\prod_{class1}h_w(x)\\prod_{class0}1 - h_w(x)$$ \n", "\n", "The $\\prod$ symbol means take the product of the $h_w(x)$ for the observations that are classified as that class. You will notice that for observations that are labeled as class 0, we are taking 1 minus the logistic function. That is because we are trying to find a value to maximize, and since observations that are labeled as class 0 should have a probability close to zero, 1 minus the probability should be close to 1. This procedure is also known as the **maximum likelihood estimation** and the following link contains a nice discussion of maximum likelihood using linear regression as an example. [Blog: The Principle of Maximum Likelihood](http://suriyadeepan.github.io/2017-01-22-mle-linear-regression/)\n", "\n", "Next we will re-write the original cost function as:\n", "\n", "$$\\ell(w) = \\sum_{i=1}^{N}y_{i}log(h_w(x_{i})) + (1-y_{i})log(1-h_w(x_{i}))$$\n", "\n", "Where:\n", "\n", "- We define $y_{i}$ to be 1 when the $i_{th}$ observation is labeled class 1 and 0 when labeled as class 0, then we only compute $h_w(x_{i})$ for observations that are labeled class 1 and $1 - h_w(x_{i})$ for observations that are labeled class 0, which is still the same idea as the original function.\n", "- Next we'll transform the original $h_w(x_{i})$ by taking the log. As we'll later see this logarithm transformation will make our cost function more convenient to work with, and because the logarithm is a monotonically increasing function, the logarithm of a function achieves its maximum value at the same points as the function itself. When we take the log, our product across all data points, it becomes a sum. See [log rules](http://www.mathwords.com/l/logarithm_rules.htm) for more details (Hint: log(ab) = log(a) + log(b)).\n", "- The $N$ simply represents the total number of the data.\n", "\n", "Often times you'll also see the notation above be simplified in the form of a maximum likelihood estimator:\n", "\n", "$$ \\ell(w) = \\sum_{i=1}^{N} log \\big( P( y_i \\mid x_i, w ) \\big) $$\n", "\n", "The equation above simply denotes the idea that , $\\mathbf{w}$ represents the parameters we would like to estimate the parameters $w$ by maximizing conditional probability of $y_i$ given $x_i$.\n", "\n", "Now by definition of probability in the logistic regression model: $h_w(x_{i}) = 1 \\big/ 1 + e^{-w^T x_i}$ and $1- h_w(x_{i}) = e^{ -w^T x_i } \\big/ ( 1 + e^{ -w^T x_i } )$. By substituting these expressions into our $\\ell(w)$ equation and simplifying it further we can obtain a simpler expression." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "$$\n", "\\begin{align}\n", "\\ell(w)\n", "&= \\sum_{i=1}^{N}y_{i}log(h_w(x_{i})) + (1-y_{i})log(1-h_w(x_{i})) \\nonumber \\\\\n", "&= \\sum_{i=1}^{N} y_{i} log( \\frac{1}{ 1 + e^{ -w^T x_i } } ) + ( 1 - y_{i} )\n", "log( \\frac{ e^{ -w^T x_i } }{ 1 + e^{ -w^T x_i } } ) \\nonumber \\\\\n", "&= \\sum_{i=1}^{N} -y_{i} log( 1 + e^{ -w^T x_i } ) + ( 1 - y_{i} )\n", "( -w^T x_i - log( 1 + e^{ -w^T x_i } ) ) \\nonumber \\\\\n", "&= \\sum_{i=1}^{N} ( y_{i} - 1 ) ( w^T x_i ) - log( 1 + e^{ -w^T x_i } ) \\nonumber\n", "\\end{align}\n", "$$" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We'll use the formula above to compute the log likelihood for the entire dataset, which is used to assess the convergence of the algorithm. Toy code provided below." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "def compute_avg_log_likelihood(data, label, weights):\n", " \"\"\"\n", " the function uses a simple check to prevent overflow problem, \n", " where numbers gets too large to represent and is converted to inf\n", " an example of overflow is provided below, when this problem occurs,\n", " simply use the original score (without taking the exponential)\n", " \n", " scores = np.array( [ -10000, 200, 300 ] )\n", " logexp = np.log( 1 + np.exp(-scores) )\n", " logexp \n", " \"\"\"\n", " scores = np.dot(data, weights)\n", " logexp = np.log(1 + np.exp(-scores))\n", " \n", " # simple check to prevent overflow\n", " mask = np.isinf(logexp)\n", " logexp[mask] = -scores[mask]\n", " \n", " log_likelihood = np.sum((label - 1) * scores - logexp) / data.shape[0] \n", " return log_likelihood" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "**Note:** We made one tiny modification to the log likelihood function We added a ${1/N}$ term which averages the log likelihood across all data points. The ${1/N}$ term will make it easier for us to compare stochastic gradient ascent with batch gradient ascent later." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Gradient" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now that we obtain the formula to assess our algorithm, we'll dive into the meat of the algorithm, which is to derive the gradient for the formula (the derivative of the formula with respect to each coefficient):\n", "\n", "$$\\ell(w) = \\sum_{i=1}^{N} ( y_{i} - 1 ) ( w^T x_i ) - log( 1 + e^{ -w^T x_i } )$$\n", "\n", "And it turns out the derivative of log likelihood with respect to to a single coefficient $w_j$ is as follows (the form is the same for all coefficients):\n", "\n", "$$\n", "\\frac{\\partial\\ell(w)}{\\partial w_j} = \\sum_{i=1}^N (x_{ij})\\left( y_i - \\frac{1}{ 1 + e^{-w^Tx_i} } \\right )\n", "$$\n", "\n", "To compute it, you simply need the following two terms:\n", "\n", "- $\\left( y_i - \\frac{1}{ 1 + e^{-w^Tx_i} } \\right )$ is the vector containing the difference between the predicted probability and the original label.\n", "- $x_{ij}$ is the vector containing the $j_{th}$ feature's value.\n", " \n", "For a step by step derivation, consider going through the following link. [Blog: Maximum likelihood and gradient descent demonstration](https://zlatankr.github.io/posts/2017/03/06/mle-gradient-descent), it uses a slightly different notation, but the walkthrough should still be pretty clear." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Stochastic/Mini-batch Gradient" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The problem with computing the gradient (or so called batched gradient) is the term $\\sum_{i=1}^{N}$. This means that we must sum the contributions over all the data points to calculate the gradient, and this can be problematic if the dataset we're studying is extremely large. Thus, in stochastic gradient, we can use a single point as an approximation to the gradient:\n", "\n", "$$\n", "\\frac{\\partial\\ell_i(w)}{\\partial w_j} = (x_{ij})\\left( y_i - \\frac{1}{ 1 + e^{-w^Tx_i} } \\right )\n", "$$\n", "\n", "**Note1:** Because the **Stochastic Gradient** algorithm uses each row of data in turn to update the gradient, if our data has some sort of implicit ordering, this will negatively affect the convergence of the algorithm. At an extreme, what if we had the data sorted so that all positive reviews came before negative reviews? In that case, even if most reviews are negative, we might converge on an answer of +1 because we never get to see the other data. To avoid this, one practical trick is to shuffle the data before we begin so the rows are in random order.\n", "\n", "**Note2:** Stochastic gradient compute the gradient using only 1 data point to update the the parameters, while batch gradient uses all $N$ data points. An alternative to these two extremes is a simple change that allows us to use a **mini-batch** of $B \\leq N$ data points to calculate the gradient. This simple approach is faster than batch gradient but less noisy than stochastic gradient that uses only 1 data point. Given a mini-batch (or a set of data points) $\\mathbf{x}_{i}, \\mathbf{x}_{i+1} \\ldots \\mathbf{x}_{i+B}$, the gradient function for this mini-batch of data points is given by:\n", "\n", "$$\n", "\\sum_{s = i}^{i+B} \\frac{\\partial\\ell_s(w)}{\\partial w_j} = \\frac{1}{B} \\sum_{s = i}^{i+B} (x_{sj})\\left( y_i - \\frac{1}{ 1 + e^{-w^Tx_i} } \\right )\n", "$$\n", "\n", "Here, the $\\frac{1}{B}$ means that we are normalizing the gradient update rule by the batch size $B$. In other words, we update the coefficients using the **average gradient over data points** (instead of using a pure summation). By using the average gradient, we ensure that the magnitude of the gradient is approximately the same for all batch sizes. This way, we can more easily compare various batch sizes and study the effect it has on the algorithm." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Implementation" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Recall our task is to find the optimal value for each individual weight to lower the cost. This requires taking the partial derivative of the cost/error function with respect to a single weight, and then running gradient descent for each individual weight to update them. Thus, for any individual weight $w_j$, we'll compute the following:\n", "\n", "$$ w_j^{(t + 1)} = w_j^{(t)} + \\alpha * \\sum_{s = i}^{i+B} \\frac{\\partial\\ell_s(w)}{\\partial w_j}$$ \n", "\n", "Where:\n", "\n", "- $\\alpha$ denotes the the learning rate or so called step size, in other places you'll see it denoted as $\\eta$.\n", "- $w_j^{(t)}$ denotes the weight of the $j_{th}$ feature at iteration $t$.\n", "\n", "And we'll do this iteratively for each weight, many times, until the whole network's cost function is minimized." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "# put the code together into one cell\n", "\n", "def predict_probability(data, weights):\n", " \"\"\"probability predicted by the logistic regression\"\"\"\n", " score = np.dot(data, weights)\n", " predictions = 1 / (1 + np.exp(-score))\n", " return predictions\n", "\n", "def compute_avg_log_likelihood(data, label, weights):\n", " \"\"\"\n", " the function uses a simple check to prevent overflow problem, \n", " where numbers gets too large to represent and is converted to inf\n", " an example of overflow is provided below, when this problem occurs,\n", " simply use the original score (without taking the exponential)\n", " \n", " scores = np.array([-10000, 200, 300])\n", " logexp = np.log(1 + np.exp(-scores))\n", " logexp \n", " \"\"\"\n", " scores = np.dot(data, weights)\n", " logexp = np.log(1 + np.exp(-scores))\n", " \n", " # simple check to prevent overflow\n", " mask = np.isinf(logexp)\n", " logexp[mask] = -scores[mask]\n", " \n", " log_likelihood = np.sum((label - 1) * scores - logexp) / data.shape[0] \n", " return log_likelihood" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "def logistic_regression(data, label, step_size, batch_size, max_iter):\n", " \n", " # weights of the model are initialized as zero\n", " data_num = data.shape[0]\n", " feature_num = data.shape[1] \n", " weights = np.zeros(data.shape[1])\n", " \n", " # `i` keeps track of the starting index of current batch\n", " # and shuffle the data before starting\n", " i = 0 \n", " permutation = np.random.permutation(data_num)\n", " data, label = data[permutation], label[permutation]\n", " \n", " # do a linear scan over data, for each iteration update the weight using \n", " # batches of data, and store the log likelihood record to visualize convergence\n", " log_likelihood_record = []\n", " for _ in range(max_iter):\n", " \n", " # extract the batched data and label use it to compute\n", " # the predicted probability using the current weight and the errors\n", " batch = slice(i, i + batch_size)\n", " batch_data, batch_label = data[batch], label[batch]\n", " predictions = predict_probability(batch_data, weights)\n", " errors = batch_label - predictions\n", " \n", " # loop over each coefficient to compute the derivative and update the weight\n", " for j in range(feature_num): \n", " derivative = np.dot(errors, batch_data[:, j]) \n", " weights[j] += step_size * derivative / batch_size\n", " \n", " # track whether log likelihood is increasing after\n", " # each weight update\n", " log_likelihood = compute_avg_log_likelihood(\n", " data = batch_data, \n", " label = batch_label,\n", " weights = weights\n", " )\n", " log_likelihood_record.append(log_likelihood)\n", " \n", " # update starting index of for the batches\n", " # and if we made a complete pass over data, shuffle again \n", " # and refresh the index that keeps track of the batch\n", " i += batch_size\n", " if i + batch_size > data_num:\n", " permutation = np.random.permutation(data_num)\n", " data, label = data[permutation], label[permutation]\n", " i = 0\n", " \n", " # We return the list of log likelihoods for plotting purposes.\n", " return weights, log_likelihood_record" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Comparing Result and Convergence Behavior" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We'll use the logistic regression code that we've implemented and compare the predicted auc score with scikit-learn's implementation. This only serves to check that the predicted results are similar and that our toy code is correctly implemented. Then we'll also explore the convergence difference between batch gradient descent and stochastic gradient descent." ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "# manually append the coefficient term,\n", "# every good open-source library does not\n", "# require this additional step from the user\n", "data = np.c_[np.ones(X.shape[0]), X]\n", "\n", "# using our logistic regression code\n", "weights_batch, log_likelihood_batch = logistic_regression(\n", " data = data,\n", " label = np.array(y),\n", " step_size = 5e-1, \n", " batch_size = X.shape[0], # batch gradient descent\n", " max_iter = 200\n", ")" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "auc 0.8699025622518947 0.8699025622518947\n" ] } ], "source": [ "# compare both logistic regression's auc score\n", "logreg = LogisticRegression(C = 1e9)\n", "logreg.fit(X, y)\n", "pred_prob = logreg.predict_proba(X)[:, 1]\n", "\n", "proba = predict_probability(data, weights_batch)\n", "\n", "# check that the auc score is similar\n", "auc1 = metrics.roc_auc_score(y, pred_prob)\n", "auc2 = metrics.roc_auc_score(y, proba)\n", "print('auc', auc1, auc2)" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [], "source": [ "weights_sgd, log_likelihood_sgd = logistic_regression(\n", " data = data,\n", " label = y,\n", " step_size = 5e-1, \n", " batch_size = 30, # stochastic gradient descent\n", " max_iter = 200\n", ")" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "weights_minibatch, log_likelihood_minibatch = logistic_regression(\n", " data = data,\n", " label = y,\n", " step_size = 5e-1, \n", " batch_size = 100, # mini-batch gradient descent\n", " max_iter = 200\n", ")" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABOQAAAN9CAYAAAAzD6YZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAWJQAAFiUBSVIk8AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xl4TVfbBvB7ZZ5HxJxE9a3W2KDU0JjHtvSjg1JjpV9p8eqEDoJWlQ4oraEqpC2qVb4O0peKMW0VpTS8VZKghqokSMj8fH/sc7Zzcoack0QS3L/r2lf22XuttZ89HOGx9lpKREBEREREREREREQVw6WyAyAiIiIiIiIiIrqVMCFHRERERERERERUgZiQIyIiIiIiIiIiqkBMyBEREREREREREVUgJuSIiIiIiIiIiIgqEBNyREREREREREREFYgJOSIiIiIiIiIiogrEhBwREREREREREVEFYkKOiIiIiIiIiIioAjEhR0REREREREREVIGYkCMiIiIiIiIiIqpATMgRERERERERERFVICbkiIiIiIiIiIiIKhATckREREREtxilVCellBiW4ZUdDxER0a2GCTkiIiK6ISil2imlZimlflJK/aWUylFKZSmlTiilvlVKTVZKRVZ2nETXm1Iq1iSZVny5opQ6afhOPKWU8q3seImIiMgSE3JERERUpSmlmiiltgDYBeAlAG0A1AbgCcAXQD0AfQDMBHBMKfWFUuq2yoqXqJJ5A6gL7TuxCMAhpVRUZQWjlIozJgsrKwYiIqKqyK2yAyAiIiKyRSn1AIDPAPgZNqUCWAMgCcA5w7ZaADoA6A/gNgADABwCEFuBoRJVllcAbDD57AsgCsAEAP8CEAHge6VUYxH5u+LDIyIiImuYkCMiIqIqSSnVCsBaaD3hBMBrAGaLSJ6V4uuVUi8CGAzgrYqLkqjS/SUih4pt+1kptQLANgCtAFQD8By0HqZERERUBfCVVSIiIqpylFLuAD6HlowDgGdF5HUbyTgAgIgUiUg8tN5BP1dAmERVlohcATDFZFOfyoqFiIiILDEhR0RERFXRUADGCRo2i8hCRyuKyFkR2Whrv1LqfqXU54bJIHKUUplKqd+UUm8rpcLtta2U2moYDyvV8NlfKfWKUuqAUuqyYdmjlJqolPKwUr+mUqrA0MZnJZ2L0qQZyh+xU66xUmq+UuqgUipDKZWrlDqllPpSKdVfKaWcOKdqSqnphnPKMOybYKVec6XUCsMEArlKqdNKqW+UUvcb9g83mWigUwnnWZ7xO3VPbLR5l1LqPaXUr0qpC0qpfKVUumFCkTlKqZYl1G+nlFqqlPqvUuqSUuqqUipFKfVpSdeinP1osh5R2kaUUh0M9/q4YdKIy0qpI0qpD5RSjW3UiTWMGzfMZJu1SSiGlzYuIiKiGxlfWSUiIqKq6EmT9bfLo0GllB+A1QD6FtvlCaCpYXlGKTVWRJY50N7tAL4D0LDYrpaG5UGlVE8RyTXuEJGzSqlNAHoB6K+U8heRy3YO0wlAfcP6SisxuAKYDW28sOL/0VoHwP8Ylu+VUo+KyMUSzqkVgK8B1Cyh3FgAc2H+d8la0K5tX6XUBwB+sdfGdYrf6XtSrL4bgHcBjLUSTzC0CUXaABgNIMhKfW8AS6G9Ol1chGF5XCm1EkCMrTjKUYHJutN/7zdcj0UARlnZfYdhiVFKTReR6aULkYiI6NbEHnJERERUpRgSZ60MH68A+KEc2lQA1uFaMi4ZWtLvHmhJr1kAcqAl5z5SSg0qoUkfAN9AS0LNAtAFWsLnCQB/GMpEA5hspa4xseYN4OESjvOE4acA+MTK/mUAJkL7O90eAE8D6G6IZQCA9YZyPQF8aUiA2eIH4CsAgQDmAOgG7T48CuCAsZBSqh+ABdASPLmGsp0BtIbWG+pXAGMAPFXCuZV3/GW5J0afAXjWEM95ANMN8URBe06eBZAAoLB4RUNs3+BaMu4HACOhXZtWhu3bDPuGQkt0XW/NTdb/KkV902TcKQDjALSFNonKFACZAFwBTFNKTSpW9wNoSW7TCSeaWlnWg4iI6FYkIly4cOHChQsXLlVmAXAvtASUAEgqpzZHmbS5FYCXlTJtAGQbymQCCLRSZqtJOxcBNLNSphqAvw1l/gbgWmy/t6GuANhqJ2ZvAJcM5bZY2T/IJJan7bQz3qTckBLOKRtASztteUBL7Ai0ZFwHK2XcoSWtxGTpVEHxl+qeGMqMNmnnZwChdmKqb2XbZEPdPAD97NR91+Q4FtfPwec51qSN4TbKKGi9HY3lFhfb38leGwC6muz/3dr1gDar8VlDmXwADa2UiTO2U5pz5cKFCxcuXG7WhT3kiIiIqKqpZrJ+rpzaHG/4mQfgCRHJKV5ARH4GMNPwMRDAiBLafE1EfrPSzj8Alhs+VgdwV7H9V6HNHgsA9ynb49b1B+BvWF9hZf+rhp9fiMiHtoIUkXnQep8BWtLJntkistfO/n4AahvW54vITivHy4eWALU5AYfB9Yi/VPfE0IPSOAFCNoD/EZELdmI6Uay+D7RZTAFgnohssKylmwTgjGG9pPNxmlLKRynVAVpvvfsNm/OgJQKdMd5kfaS16yEix3DtvN0APOPkMYiIiG5ZTMgRERFRVRNgsp5V1saUUjWhvRoHABtF5KSd4osBFBnWe5TQ9Kd29pmOn9bAyn7ja6sK115LLW6o4Wc2gC9NdyilGgG40/CxxMkhcO1VybYlvPZp7bVYU91M1pfbKiQifwH4j6391zH+0t6TZrg26cEqQ/zOuA9AqGHd7vmINlNwkuFjeyePY81y00kSoD0vO3BtVtWrAAaLyH8dbdBwjbsYPh4wJKtt+RxAumG9pO8MERERGXBSByIiIqpqTCc58CuH9pqarP9osxS0nlRKqaPQBqtvZqfoP4ZeV7akm6wHWNm/A0AKtJlknwDwuulOQxKxu+HjVyJSPDF5j8n6OjuTkBbnASAE2vhoxWUbejzZY7yWWQAOl1D2F1zroVXc9Yi/LPfEdNbUREeDMWF6PvucOJ9apTiWo05B6yX3joj86WTdBgB8DeslfWfylVK/QBvnr5FSyt3QS5KIiIjsYA85IiIiqmpMkyph5dBeqMn6WQfKG8uE2imTXUIbRSbrFj26REQAxBs+/ksp1bZYkcdN6lnMrgqgRgnHt8fHxvYMB+qGGH6eN5yDPfZeN74e8ZflnlQ3WT/tdESlPx9b5+KMV3BtgoQmAG4HECIi9UTk6VIk44DSf2cUrj0jREREZAd7yBEREVFVcxBAAbS/p7RQSrmJSEElx3Q9rATwmmH9CQA/mewzvq76F6zPMmv6d7gRuDbGmiNsJZwsZg69jq5H/JXJ9Hx64NoYcRXhLxE5VIHHIyIionLAhBwRERFVKSJyWSm1F9qsp97Qxi1LKEOTpoPR13SgvLGMzUH9y4OIHFNKJQFoB+BRpdS/RSRPKdUUQHNDsU9EpMhKddNXNrMrMCFjfO2zulJKldBLzl6vscqK3xbTeGrbLOVY/cwqcD5lVdrvjMD81WAiIiKyga+sEhERUVX0kcn6czZLOcZ01s3ir4aaUUpVg/bKHwAcKONxHWF8HTUUQF/D+lAr+4sznQn1vvIOyo6Dhp9+ABqVULa1nX2VFb8tpj30OpeiflU7n7I6jmsTqpT0nXHDtXt92Mr4cSW92kxERHRLYkKOiIiIqqKVAFIN692UUmMdraiUClNK9TZ+FpFzuJaU662Uqmun+mhc+/uRzVlCy9EaALmG9aFKKRdo48cBwF4RSbZR7wC0pImxXlnGZHPGZpP14bYKKaVqwf6Mm5UVvy0HcS2eQUqpOk7W/wHAJcP6WKVUeYwNV2lEpBDAFsPHFkope8nVh3Ft3Dhr35mrxhWllGf5REhERHTjY0KOiIiIqhwRyQPwCK4lq95XSr2slHK3VUcp5aKUehzAr9BedzU1z/DTE8AKa4kBpVQrAC8bPmYCiCv9GThGRDIB/J/hYx8Aj+LaK5Mr7NQTALGGjwEANiilqtsqDwBKqXuUUn3KFDCwAdfGcBuvlLrXynHcoPVwtJl8qcT47cUz0/DRF8CXSimbkxMopeoVq38ZwNuGj5EAVimlfC0qmrfRXSnVvvRRX3fzTNY/VkoFFy+glIoE8K7hYwGAhVbaMR3z71/lFx4REdGNjWPIERERUZUkIr8opR4G8Bm0VyRfB/CkUmo1gF0A/jYUrQVtHLb+sP0P/uUAHgPQHUAXAHuVUu9C6znnA6AngH9DG7MOAJ4WkYvlflLWrYTWy8gDwAeGbfkAVtmrJCLxSqn7ADwJ7bXCI0qpjwBshTbrpQe05F5LAA9Cm4XzDQDflTZQwxh3YwCsh5Zw26KUmgdgI7RXHO8CMAFAFLRJKoyvO1q8tlgZ8ZdwbsuUUj2h3Ys2hng+ALAD2phq/tBmMb3fsL9asSZmAmgP7Vl6EMB/lVJLACRBmznYG0A9aK93PgTgNmg9Mnddj/MpKxHZopRaBmAUtPPer5SaA+AXaLPU3gfgBVzrHfeqjRldd5isL1BKvQHgFK7NenumAr9rREREVQYTckRERFRlicjXhl5Y7wPoBCACwCR7VaC9Brq8WDuilPofAKuhjdXWGMAyK/VzAYwVkdVlDt5xCdCSizUABBm2bRSRfxyoGwPgJIBXoCVGXjQstpQ58SEiG5RSz0DrQeUF4CXDYuoDaOOqGRNyOTaaq/D4SzAYWvLsfwFUBzDV0ThEpFAp9SCA9wz16wCYZudYgmuvuVZV/wtAARgJoD6072FxRQCmicgsG21sB5AIbWy++2A5xt4IVEBvVCIioqqGr6wSERFRlSYih0SkM7TeR7MB/AzgDIA8AFegJXS+g5aoixCRQSKSZqWdLBG5H1rvpS+g9dLJhZYUOQjt1bs7RMRaou66EZECaL0ATdmazKF4XRGR6QAaAJiBaz0HC6CN3ZUG4HtoCa+mIjKnnGJeCK3nWjy065gHrVfbdwAeFJGxuJZcBGwk0iorfltEJF9ExgC4G1pSMRnAZQCF0GYP/QnALAAdbdTPM5x7Y2ivsO6B1ruuEEA2gGMAvoY2UUlDEfn8ep5PWYlIgYiMgna+8QBSoCVXswH8AWARgOaGe2irDYH2OvZkaN/dDGjXg4iI6Jam7M9WT0RERETkPKXUx9B6P+UD8BeR3BKqEBEREd0y2EOOiIiIiMqVUsoP2ph+gDZbLJNxRERERCaYkCMiIiIipyilbM6WaZgJ92MAxlk5P66QoIiIiIhuIHxllYiIiIicopRKB/ArgP8DcABAJrRZSKOgTdRwl6HoHgD3GsbJIyIiIiIDJuSIiIiIyClKqUwAgSUU+wXaBA9nKyAkIiIiohsKE3JERERE5BSlVDcAvQF0AFALQDUACsB5aL3i1gJYIyJFlRYkERERURXGhBwREREREREREVEF4qQOREREREREREREFYgJOSIiIiIiIiIiogrEhBwREREREREREVEFYkKOiIiIiIiIiIioArlVdgB0fSilUgAEAEit5FCIiIiIiIiIiG4WEQAuiUhkWRphQu7mFeDt7R1y5513hlR2IEREREREREREN4PDhw/j6tWrZW6HCbmbV+qdd94Zsnfv3sqOg4iIiIiIiIjoptCyZUvs27cvtaztcAw5IiIiIiIiIiKiCsSEHBERERERERERUQViQo6IiIiIiIiIiKgCMSFHRERERERERERUgZiQIyIiIiIiIiIiqkBMyBEREREREREREVUgJuSIiIiIiIiIiIgqEBNyREREREREREREFYgJOSIiIiIiIiIiogrkVtkBUNWSlZWFS5cu4cqVKygoKICIVHZIRER0C1JKwc3NDT4+PggICICfn19lh0REREREVG6YkCMAgIjg77//Rnp6emWHQkREBBFBfn4+Ll68iIsXLyIkJAQ1atSAUqqyQyMiIiIiKjMm5AgAcOnSJaSnp0MphdDQUPj7+8PDwwMuLnyrmYiIKl5RURHy8vJw+fJlXLhwAenp6fDy8kJgYGBlh0ZEREREVGZMyBEAICMjAwAQFhaG4ODgSo6GiIhudS4uLvDy8oKXlxfc3Nxw9uxZZGRkMCFHRERERDcFdn8iAEBOTg4AICAgoJIjISIiMmf83WT8XUVEREREdKNjQo4AQJ+8wdXVtZIjISIiMmccPoETDRERERHRzYIJOSIiIqrSOJEDEREREd1smJAjIiIiIiIiIiKqQEzIERERERERERERVSAm5IiIiIiIiIiIiCoQE3JEREREREREREQViAk5ItJFRERAKYWtW7dWdijl4mY7n+utU6dOUEohLi7ObPvWrVuhlEJERESlxFWZ+AwREREREdH14FbZARDdilJTUxEXF4egoCBMmDChssO54ezfvx/r169HREQEhg8fXtnhUCWIi4tDamoq+vfvjxYtWlR2OGQiNjYWADBhwgQEBQVVbjBERERERFUUE3JElSA1NRXTpk1DeHg4E3KlsH//fkybNg3R0dF2E3K33XYbvLy84OPjU3HB3YR8fHxwxx13oE6dOpUdii4uLg7btm1DREQEE3JVzLRp0wAAw4cPZ0KOiIiIiMgGJuSI6Kb1ww8/VHYIN4V77rkHR44cqewwiIiIiIiIbhocQ46IiIiIiIiIiKgCMSFHVA7y8vIwb948tGvXDkFBQXB3d0dYWBiaN2+OsWPH4scff9TLRkREoHPnzgCAtLQ0KKXMluID6gPAunXr0KtXL1SvXh2enp6oW7cuBg8ejH379pUYW0JCAgYOHIi6devC09MTNWvWRNu2bfH666/j5MmTNuulp6dj4sSJiIyMhKenJ+rUqYPRo0fjzJkzVstfvnwZcXFxeOSRR9CkSRMEBQXB29sbDRs2RExMDI4ePWrzWJcvX8aMGTPQsmVL+Pv7w8PDA7Vr10arVq3wwgsv4NChQ3pZpRRGjBgBANi2bZvF9TMdfL+kAfnz8/OxZMkSdO3aVb+24eHh6NGjB5YsWYLs7Gw7V9a6Xbt2oW/fvggJCYGvry+aN2+OuXPnoqioCMOHD4dSSh9jy6j4pAkbN25E7969UaNGDbi4uGDu3Ll62e3bt2P8+PFo06YNateuDQ8PD9SoUQO9evXCF198UWJ8CQkJ6NKlCwIDAxEQEIC2bdsiPj7ebh1HJnVITU3Fs88+izvuuAM+Pj7w9/dHy5Yt8dZbb9m8jsZ7lpqaihMnTmD06NH6cxoZGYnnn38ely5dMqsTFxcHpRS2bdsGABgxYoTZ/S/NxBOffvop2rZtCz8/P4SEhKBLly749ttvHaqblZWFmTNnonXr1ggMDISXlxduv/12jBs3zub3q6ioCHFxcejcuTNCQ0Ph7u6O6tWro3Hjxhg5ciQSEhKs1hMRrFmzBn379kXNmjX17+V9992H9957DxcuXLBab+fOnXjsscf0axsaGopu3bph1apVEBGL8sXv965du3D//fejWrVq8Pb2RvPmzbFgwQKLusbn2ygyMtLs3nC8RyIiIiIiEyLC5SZcAOyNiooSRyUnJ0tycrLD5ema/Px8iY6OFgACQJRSEhQUJK6urvq2Rx99VC/fqlUrCQ4OFgDi4uIiYWFhZsvq1av1soWFhTJ06FC9HVdXVwkKCtI/u7i4yAcffGA1rtzcXBkyZIheFoAEBgaKr6+v/nnq1KlmdcLDwwWAxMfH6+s+Pj7i6emp14mIiJD09HSL473//vtmcYaEhIiHh4e+zdfXVzZt2mRRLzMzU+666y6zcwoODhYXFxd920svvaSXDwsLk4CAAAEg7u7uFtdv165dFueTmJhocdxTp05JixYtzI5bPGZr9exZsWKFWdxBQUHi5uYmAOShhx6SYcOGWb3uiYmJAkDCw8Pl7bfftniO3nvvPRERuXz5stn99Pf316+FcYmJibEZ3+zZsy2eU2O8EydO1J/j5cuX24zPmi+//FK8vLz0tn18fMTd3V3/3LRpUzl79qxFPeP+9evXS0hIiH5OxmsGQFq1aiV5eXl6ndWrV0tYWJjefkBAgNn9b9WqlWM3y2Ds2LFmz0BQUJAopQSAzJs3z+4zlJycrO8HIG5ubmbfr+DgYNm5c6dFvccff9zie2n63LVp08aiTmZmpnTr1s3s/gUHB5td9+L3TUTkxRdfNDtWQECAfn4A5LHHHpPCwkKzOqb3e/ny5eLq6ipKKQkMDDRra/z48Wb1xo0bJ2FhYfr+atWqmd2bcePGOXVvrF1v/p4iIiIiosoWFRUlAPZKWfM2ZW2AS9VcmJCrOCtWrNCTEPHx8XL16lURESkoKJC0tDRZsGCBzJw506xOSQkOozfffFP/x/eMGTPk0qVLIqIlkx5++GE9ibBt2zaLumPGjNGTY1OnTjVLiBw/flzmzJkjixcvNqtjTC4EBQVJixYtJCkpSUS0pOOGDRv0ZOALL7xgcbxVq1bJyy+/LLt375bc3FwRESkqKpLDhw/L4MGDBYBUr15dsrKyzOpNmzZN3/fNN99Ifn6+iIjk5eXJH3/8IbNmzZIlS5aY1Vm+fLkAkOjoaLvXz1YyJScnR+6++249abBixQo9roKCAtm7d69MmDBBfvrpJ7vtmzp8+LCeVOnTp4+kpKSIiMiVK1dk/vz54ubmpl8/Wwk5Ly8vcXV1lTFjxuj36+rVq3Ly5EkREcnOzpaBAwfKV199JRcuXNDrZ2RkyIIFC8TPz08AyOeff24R344dO/REzJAhQ+TMmTN6XWPSxphwcSYht3v3bnF3dxc3Nzd5+eWX5dSpU/p1TEpKklatWgkA6dGjh0Vd08Rlly5d5ODBgyKi3Z9ly5bpieCFCxda1LWVPHTGJ598osfw/PPPS0ZGhoiInD17VoYOHSru7u7i4+Nj9RnKzMyUiIgIASAPP/ywHDhwQAoKCkRE5NixY3rSLSwsTG9XRGTbtm369/K9997Tv9NFRUVy+vRpiYuLk+eee84i1r59+woA8fb2lnnz5ultFhUVSXJysrz22muyfv16szpz587VY1iyZIlkZmaKiPZMrl69WmrWrCkAbP755OPjIx4eHvLMM8/oz2NGRoY8++yz+p9Lhw4dsojVeE2N34Hywt9TRERERFQVMCHHxf6NZUKuwjz99NMCQP73f//X4TqOJOQuX76s936aNGmSxf6CggLp0KGDAJCOHTua7Tt06JCefCmedLPHmMAKCwuTf/75x2K/sfdWZGSkw22KaEkDY++euLg4s329e/cWADJr1iyH2ytrQm7hwoUCQDw9PeXAgQMOH9ceY0/GJk2a6AlJU2+99ZaeqLCVkAMggwYNKnUMK1euFADSqVMni31dunQRANK5c2cpKiqy2D9q1CibPa3sPa/t27cXALJo0SKrMV24cEFq1aolAOSXX34x22c8XuPGjSUnJ8ei7jPPPKPHXFxZE3JFRUVy2223CQAZNmyY1f2mPdKKP0Mvv/xyiferV69eAkDmzJmjbzM+B7169XI41m+//VZPgG3cuNGhOhkZGeLn5ydeXl6yf/9+q2WSkpL0nnamz6zp8/jkk09ardu0aVMBINOmTbPYx4QcEREREd3Myishx1lWbVBKBQB4EcAAAOEArgLYD+BDESl5oCbbbfYD0ANAawD1ASgApwFsBzBfRH4te/TXT8Qkx8ZVuhGkzupbLu0EBAQAgM2x1Upr06ZNuHTpEjw8PPDiiy9a7Hd1dcWrr76Knj17YseOHTh79ixq1qwJAIiPj4eIoFGjRoiJiXH62DExMQgNDbXY3r9/fzz//PNISUlBdnY2fH19HWpPKYW+ffti8+bN2LVrF4YNG6bvu17Xz56VK1cC0MYfa9asWZnbKyoqwvr16wEAEyZMgIeHh0WZZ555BtOnTy9xXLoXXnih1HE88MADAICffvoJhYWFcHV1BaCNB5iYmAgAeOmll8zG+TKaMmUKli1b5tTxjh07hl27diEoKAijRo2yWiYkJAS9e/fGxx9/jE2bNqFVq1YWZSZOnAhPT0+L7f3798eCBQvMxhAsL/v378exY8cAAJMnT7bYr5TClClTsHnzZqv1V6xYAQB47rnnbB7j8ccfR0JCAjZt2oTnn38ewLXn/e+//0ZRURFcXEoeytX4vPbs2RO9evUqsTwAfPnll8jKysL999+P5s2bWy1z7733IjIyEsePH8fevXtx7733WpSxdm0AoF+/fjh48OB1uTdERERERLcCJuSsUErVhZYgizRsygIQAKALgC5KqQ9FZEwpmt4LoKHJ5yuGnw0MyxNKqZdE5J3SRU6VoXfv3njrrbewYcMGPPjggxg+fDiio6OtJrScYZywoXnz5ggODrZa5r777oOrqysKCwuxb98+9OnTB4CWkAGgf3ZW69atrW6vU6eOvp6ZmWmRkDt16hTef/99bN68GceOHcPly5dRVFRkVub06dNmn/v06YM1a9Zg/vz5uHDhAh5//HF06NAB/v7+pYq9JPn5+di7d69+7PJw/PhxffKBDh06WC3j4+ODli1bYvv27TbbMQ6Yb09BQQFWrFiBtWvX4sCBA0hPT0deXp5ZmZycHGRkZKBatWoAgF9//RUiAhcXF5vxNWjQAPXq1bM70UdxSUlJALSJDerWrWuzXFZWFgDYbLuk5y0jI8PhmBxl/H6FhYXhjjvusFqmXbt2cHNzQ0FBgdn2kydP4tSpUwC0Z8haghOAfl9Mz7tr167w8PDAvn370KlTJ8TExKBLly6oXbu2zVhL83023pstW7boiXpr0tPT9RiLJ+RCQkLQoEEDq/Wu570hIqIb0/nLuTideRXN6gba/N1IRETXMCFXjNJ+e3wBLRmXCmCwiCQppbwAPAtgFoCnlVK/ishSJ5t3B/ArgI8AfCsiaUopFwBNAMwF0BnA20qpwyLyXfmcEV1v0dHRmD59OqZPn46vv/4aX3/9NQCgUaNG6Nu3L5566incfvvtTrd7/vx5AOZJsOK8vLxQrVo1nDt3Ti8PAOfOnQMA1K9f3+njArCZDPPy8tLX8/PzzfZt27YN999/v558AaDPOgkAV69exaVLlyx6iA0dOhS7du3CkiVL8Mknn+CTTz6Bi4sLmjVrhgceeABPP/00atWqVarzsCY9PV1PsJT2+hT3zz//6Ovd3X39AAAgAElEQVT2YrWXdAGA0NBQuz2msrKy0LNnTz3ZAmhJvOrVq+v1jPc+OztbT8gZn43AwEC7vRrr1KnjVELO2KuxoKBAP649V65csbq9pOeteEKsPBivib174unpiWrVquHs2bNm2017c/79998lHsv0vG+//XZ8+OGHeOaZZ7Bjxw7s2LEDgDYjcK9evRATE4O7777brH5pvs/GGK9cuWLzutuK0cheUtx4b4r/OUBERLem9Ow8dJy9BTn5RZjRrzGeuDeiskMiIqrymJCz1A9AGwBFAB4Skf0AICI5AOYopWoDmABgulJqhYjk2W7KwhMissN0g4gUAfhNKdUXWg+6OwG8AKBKJuTK6zXPm82rr76KIUOGYM2aNdi6dSt+/PFHHDlyBEeOHMG8efOwbNkyDB06tFRt5+TklHO05S8/Px9DhgxBVlYWunXrhtdeew2tW7c2S+AtW7YMTz75pHGMQzOLFy/GuHHj8Pnnn2P79u34+eefsX//fuzfvx/vvvsuvvrqK3Tv3r0iT6lSGF8xtWXGjBlISkpCtWrV8M4776BXr16oUaOGvr+wsBBubtof69auc3kz9n5s3rw59u/ff92PV1WY9vrMyMhAUFCQU/VHjhyJvn37YvXq1UhMTMTOnTuRmpqKRYsWYfHixXj99dcxZcqUcolx/PjxmDt3bpnaIiIiKskvqenIydd+92z74x8m5IiIHFDy4DW3nsGGn5uNybhi3oY2YHVNaK+wOqx4Mq7YvqsA1hg+tnSmXaoaIiMjMWnSJCQkJOhjdt13330oKCjAmDFjHOpJY6p69eoAgBMnTtgsk5OTgwsXLpiVB7TX8AAgLS3N2dMolR9//BGnTp1CSEgINmzYgI4dO5ol4wCU2IOqcePGmDZtGhITE5GZmYmvv/4aTZs2RXZ2NoYNG1ZuPXFCQkL0pFV5XR9jTzTA/lh4ZR0nb+3atQCA999/H0OHDjVLxgG2r7Hx2bh48aLd3lLFXycuifE5c6ZXXVVhvCb2zjkvL8+s96OR8bwB+99Pe8LCwjB+/HisX78e58+fx+7du/HQQw9BRPDqq6/it99+szieM8+rsU5p4yMiInJGfuG1/6wqLDZcCRERWceEnKXOhp/fW9spIn8B+N3w0amEnAMuGH7a7yZDVZ6rqys6deqEb775Bu7u7sjOzsaePXv0/cbXC+31YoqKigIAHD16FH/99ZfVMtu3b9df5zOWB4C2bdsCADZu3Fi2E3GQcTytf/3rX/Dx8bFaxtbg+NZ4eHjg/vvv1xNQZ86cwdGjR/X9jlw/W9zd3dGypZbz/u678umI2qBBA32w/p07d1otc/XqVX3sutIyXufirzQa2brGd999N5RSKCoqshlfSkqK08kb45hj6enp+Pnnn52qW1ZleQaAa9+Xc+fO4Y8//rBaJikpyerrspGRkXrCqzy+Y0optG7dGmvXrkXdunUt7pPx++zM82q8N1u3bsXVq1fLHKMzjOMGVUQvTSIiqhoKCq/9mV/IP/6JiBzChJwJpVQNAMaR+H+3UzTZ8POucg4h2vCT09bdQIoPqG/Kw8NDfw0xNzdX325M3ly8eNFm3R49eiAgIAD5+fmYM2eOxf7CwkLMmDEDANCxY0ezgdufeOIJKKVw5MgRLF682LkTKoXAwEAAWvLQ2iu2//nPf/RZPouzd/28vb31dWvXLzMzs1TxGl8fjouLM+uJVFouLi7o168fAGDevHlWe/N98MEHZuPrlYbxOh88eNBiX1ZWFt544w2r9UJCQtCli/b/B7Nnz7aaKJk1a5bT8TRq1EhPFr344ot2ezFevXrV7B6WVVmfgRYtWqBhQ22Onbfeestiv4jYvSbDhw8HALz99ts2E+bGdkxjtPe8u7q6wt3dHYD58258Xv/zn/8gISHBZn1TDz/8MHx9fZGRkYHp06fbLVveEzOU9d4QEdGNp6Do2t8tioqYkSMicgQTcuZMR2O39+6WcV+5jTSvlIoC8JDh43In6u21tgBoVF6xkX1Dhw7FiBEj8P333+Py5cv69tTUVAwbNgw5OTnw9vZGx44d9X2333473N3dcfHiRXz55ZdW2/X19dXHkZo/fz7eeOMNPaHz119/YdCgQdi5cydcXFzw+uuvm9Vt3LgxnnrqKQDA2LFjERsba/bKbEpKCmJjY7Fo0aJyuQbt27eHj48PLly4gKFDh+qvZl69ehUff/wxBgwYYHPW2W7dumHcuHHYvn27WU+e33//XU961KpVC02bNjU7PwBITk4uVc+sUaNGoUWLFsjNzUXXrl0RHx+vv8pZWFiIPXv2YPTo0U61PXnyZHh4eODgwYMYMGCA/nphTk4OFi5ciEmTJjk91lhxxnH0Jk6ciG3btumJtV9++QVdu3bVX1+2JjY2Fkop/PDDDxg+fLj+euvFixcxZcoULFmyRE/4OWP+/Pnw9PTE9u3b0bVrV+zcuVMfv6ywsBAHDx7E9OnT0aBBgzK/smvK+AysW7fObmLbFqUUYmNjAQAff/wxXnrpJT2BdO7cOYwcORJbtmyx2eNz0qRJaNCgAf755x+0a9cOn3/+udnze+LECSxZsgRRUVFYv369vn3KlCkYOHAg1q9fr89wajzmuHHjkJKSAqWU2ZiJvXv3Ru/evSEiGDBgAN5//309VhFBcnIynnvuObPjhIaG4s033wSgJVtHjx5t1hPw6tWr2LFjB55++mm0a9fO6etnj/HerFy5EoWFheXaNhERVU0FZq+sMiFHROQQEeFiWAC0gzY+nABoaKfcG4Yy/y2n4/oDOGxocy8ANyfq7rWxZEdFRYmjkpOTJTk52eHydE2/fv2Mz4wopSQoKEh8fHz0ba6urrJy5UqLekOHDtXLBAYGSnh4uISHh8vatWv1MgUFBWblXF1dJTg4WJRSAkBcXFxk4cKFVuPKycmRRx55REyeaQkKChJfX1/989SpU83qhIeHCwBJTEy0eb7GuikpKWbb582bZ3aswMBAcXNzEwDSokULmT9/vgCQ6Ohos3rNmzfX67i4uEhwcLB4eXnp23x8fGTz5s0Wcdx33316mZCQEP36/fjjjw6dz4kTJ6RJkyZm1zY0NFQ8PDz0bfaugzUff/yxfm8ASHBwsLi7uwsAefjhh/V7OXPmTLN6iYmJAkDCw8Pttn/s2DGpVq2a3r6Xl5d+P729veX777+3eX9ERGbPnm32rAYHB4urq6sAkIkTJ0p0dLQAkOXLlzsV33fffSeBgYF6256enhIaGqqfu3FJTU01q2cvVhGRlJQUvUxxhw8f1u+Vm5ub1K5dW8LDw6V9+/Z2r2FxY8eOtfn9mjdvnt1n6OjRo3LnnXdaPEPe3t5m5x0XF6fXGT9+vNm+gIAA8ff3N9v2xhtvWBwrIyNDvz/G70pISIjZd6X4fRMRmTFjhtkz6evrK8HBweLi4qJvi4iIMKvjyPO4fPlyq99nEe17YPqM1q9fX8LDw+W5556z2Z4j+HuKiKjqiv8xVcJf+kbCX/pGHl6UVNnhEBFdV1FRUQJgr5QxF3RT9JBTSr2mlCoo5WL9Ha+Ki90NwGfQerRlAnhMRCwHLbJBRFpaWwAcuU4hUzGzZs3C7Nmz0atXLzRo0AB5eXkoLCzEbbfdhhEjRmDfvn144oknLOotWrQIkydPRqNGjZCbm4u0tDSkpaWZvdbo6uqKFStW4IsvvkCPHj0QFBSErKws1KpVC4MGDcLu3bsxZswYq3F5enpizZo12LBhAx544AGEhYUhOzsb/v7+aNu2Ld544w2MHj263K7DuHHjsG7dOr23XEFBARo1aoRp06YhKSkJ/v7+Vut99NFHmDZtGjp37oz69evrvYwaNWqEZ555BocOHULXrl0t6q1btw5jxoxBZGQksrKy9Ovn6Ky09erVw549ezB//nx06NAB/v7++rXt2bMnPvroI9xzzz1OXYMRI0Zg+/bt6NWrFwIDA5Gbm4u77roL8+fPx+rVq/WeXKXtKdegQQPs3r0bQ4YMQY0aNVBYWIigoCAMHjwYv/zyC3r06GG3/gsvvICNGzeic+fO8PPzQ0FBAVq1aoWVK1finXfeKVVMgNaD648//sArr7yCqKgoeHp6IjMzEwEBAWjXrh0mTZqEvXv3Ijw8vNTHKK5Ro0bYtGmTfq3Pnj2LtLQ0fZw9Ry1YsACffPIJ2rRpA09PT4gIoqOj8c0332DcuHF26zZs2BC//vorPvjgA3Tu3BnBwcG4ePEi3Nzc0KxZM8TExODbb7/FkCFD9Dr//ve/MX/+fPTr1w//+te/ICLIzc1FvXr18Oijj2L79u1WZ1gNCgrCli1bsGLFCnTr1g0hISG4fPkyQkNDER0djblz5+LBBx+0qPfKK6/gwIEDiImJwe23346ioiJkZ2frz/ns2bOxY4fN+YZKZcSIEVi6dCnuueceuLm54eTJk0hLS7M6QQYREd0cTHvI8ZVVIiLHKJEb/w9MpVQsgKmlrP6WiEwytNMcgHFm1WYiYjlQk1buPQATAOwzJL9KRSnlAmAltJldrwDoKSLWR1x3vu29UVFRUY4OIn/48GEAwJ133lkehyeiYkQE4eHhOHnyJBITE9GpU6fKDonohsLfU0REVddHO47j9W+1P6fvrh+Er8a0r+SIiIiun5YtW2Lfvn1lygcBgFt5BVSZRCQWQGw5NGU6blxtAFYTcoZ9AFDqAZGUNg3dh9CScXkAHiqvZBwRVT2rV6/GyZMnERAQgDZt2lR2OERERETlhpM6EBE576Z4ZbW8iMh5AMZ3ahrbKWqcXTXZTpmSzAUQA6AA2muq/ylDW0RUBcycORPvv/8+Tp48qU9skJGRgXnz5mHUqFEAgDFjxpjNHktERER0ozOdyKHwJngDi4ioItwUPeTKWSKAhwF0B/Bu8Z1KqTq4lqz7oTQHUErNAjAOQBGAYSLyVelCJaKqJDk5GZ9++inGjRsHDw8P+Pr6IjMzU58NtVu3bpg6tbRv1xMRERFVTflms6xWYiBERDcQJuQsfQYtIddDKdVcRA4U2z8RgIL2umqis40rpV4F8BK0GehiROSzMsZLRFXEmDFjEBAQgJ07d+LMmTPIzMxESEgImjVrhiFDhmDo0KFwc+Mfu0RERHRzKSjkK6tERM7ivwwtbQDwM4A2AL5SSj0uIj8ppTwBPANtMgcAmCoiecUrK6VSAYQDWCEiw4vtmwBguuHjsyKy7PqcAhFVhnbt2qFdu3aVHQYRERFRhSrgK6tERE5jQq4YERGl1EAA2wFEAvhRKZUFwAvXrtciEVlaiuaNr8AWAXjV0FvOltYicrIUxyAiIiIiIqowBSbvqbKHHBGRY5iQs0JETimlWkB7tfR/AEQAuAxgP4APRWRtKZtWhp8uAMJKKOtaymMQERERERFVGLNZVtlDjojIIUzI2SAilwC8bFicqRdhZ5+ytY+IiIiIiOhGVFBkMqkDE3JERA5xqewAiIiIiIiI6MZVaNpDjrOsEhE5hAk5IiIiIiIiKrV8k1lWCzmGHBGRQ5iQIyIiIiIiolIr5CyrREROY0KOiIiIiIiISi2fs6wSETmNCTkiIiIiIiIqNfaQIyJyHhNyREREREREVGocQ46IyHlMyBEREREREVGpFRTxlVUiImcxIUdERERERESlxldWiYicx4QcUTmIiIiAUgpbt26t7FDKRWpqKpRSUEpVdihV0vDhw6GUQmxsrNn2W/m6derUCUopxMXFVXYoREREVMHMJ3WoxECIiG4gbpUdABGVXlxcHFJTU9G/f3+0aNGissOhKmj9+vXYv38/OnXqhE6dOlV2OGRi7ty5yMzMxPDhwxEREVHZ4RAREZUae8gRETmPCTmiG1hcXBy2bduGiIgIJuSqAHd3d9xxxx2VHYaZ9evXY8WKFQDAhFwVM3fuXKSlpaFTp05MyBER0Q2NkzoQETmPCTkionJSp04dHDlypLLDICIiIqpQxZNwRUUCF5dbbwgPIiJncAw5IiIiIiIiKjXTMeQAvrZKROQIJuSIytmJEyfw5JNPol69evDy8kJkZCSef/55XLx40Wr53NxcrF27FkOHDkXz5s1RrVo1eHl5ITw8HIMHD8bevXst6sTFxUEphW3btgEARowYoU8moJSy+vpbfn4+lixZgq5du6J69erw9PREeHg4evTogSVLliA7O9vmOR06dAiPPfYYatasCS8vLzRq1AgzZsxAXl5e6S4SgOTkZDz66KOoUaMGvL290ahRI0ydOhU5OTmIjY2FUgrDhw83q1N80oSffvoJAwcORK1ateDq6ooJEyboZfft24dJkyahQ4cOqF+/Pjw9PREaGopOnTrho48+QmFhod34fv75ZzzwwAMICQmBn58fWrRogXnz5qHIzkjFjkzqcP78eUyePBlNmzaFn58ffH190aRJE7z88stIT0+3Wsd00pD09HRMnDgRkZGR8PT0RJ06dTB69GicOXPGrM7WrVuhlNJfV502bZrZM1KaiScSEhLQpUsXBAYGIiAgAG3btkV8fLxDdfPy8rBgwQJ07NgRISEh+vM3cuRIHD582Ga9DRs2oE+fPggLC4O7uztCQkJwxx13YNCgQVizZo3dWAcOHIi6devC09MTNWvWRNu2bfH666/j5MmTVuscOnQII0eORGRkJLy8vBAUFIT27dtj0aJFyM/Ptyhf/H47+j0xPt9paWkAgM6dO5vdF75aTEREN5riPeT42ioRkQNEhMtNuADYGxUVJY5KTk6W5ORkh8uTufDwcAEgS5culerVqwsA8fPzEy8vLwEgAKRhw4Zy+vRpi7pff/21XkYpJcHBwWb13NzcZOXKlWZ1Vq9eLWFhYeLu7i4AJCAgQMLCwvSlVatWZuVPnTolLVq00Nt0cXGRkJAQ8fDw0LclJibq5VNSUvTt33//vXh7ewsACQwMFBcXF31fv379SnW9Nm3aZHaOAQEBeixt27aVSZMmCQAZNmyYWT3TuFavXi1ubm56XO7u7jJ+/Hi9bGhoqF7Wx8dHgoKC9M8ApE+fPpKfn281vlWrVomrq6teNigoSD/WgAEDZNiwYQJApk6dajM+a3bs2CEhISF6GQ8PD7PrUK9ePTly5IhFPePzFR8fr6/7+PiIp6enXjciIkLS09P1Ort27ZKwsDC9fV9fX7NnJCwszMG7pZk9e7bZcxoUFKQ/CxMnTpTo6GgBIMuXL7eoe/r0aWnevLnZ8+fv769/9vLyki+//NKi3pQpU8zumb+/v9n1snYOubm5MmTIELN6gYGB4uvrq38uft9ERN5//32zZ9vPz8/sGejUqZNkZ2eb1Snt92TOnDkSFhamlwkODja7Lw899JBT9+ZWwd9TRERVV6c5iRL+0jf6kp1r/e9YREQ3g6ioKAGwV8qatylrA1yq5sKEXMUyJkkCAwOlYcOGsmPHDhERKSwslPXr10u1atUEgHTv3t2ibmJioowbN062b99u9g/+tLQ0mTBhgp6wSEtLs6hrLwlilJOTI3fffbcAkGrVqsmKFSskKytLREQKCgpk7969MmHCBPnpp5/0OqaJhqCgIHnkkUckJSVFRESysrLkzTffFKWUAJBvv/3WqWt1/vx5PVl2zz33yMGDB0VEJC8vTz799FPx8/PTk2f2EnJ+fn4yYMAAPa78/Hx9XURk0KBBsmrVKjlz5oy+LSsrS+Lj46VmzZoCQGbPnm0R359//qknfXr06CHHjh0TEZHs7Gx55513xNXVVQIDA51OyKWmpurn9fTTT8vRo0elsLBQCgsL5eDBg9KjRw8BIHfddZcUFBSY1TU+X0FBQdKiRQtJSkrSz3nDhg16uy+88ILFcW0lD52xY8cO/X4PGTJEv6YZGRny4osv6s++tWcxLy9PWrduLQCka9eukpSUJHl5eSKiJeqMz7iPj4/8+eefZtfSmLCaPHmynD9/Xt/3999/yxdffCEjR460iHXMmDECQFxdXWXq1Kly9uxZfd/x48dlzpw5snjxYrM6X331lZ7wmz17tn6s3NxcSUhIkNtvv10ASExMjFm9sn5PjPfVNBlOtvH3FBFR1dXhrR/MEnKXruZVdkhERNcNE3Jc7N9YJuQqlPEf1l5eXnL06FGL/Vu2bNH/4W5M1jlq5MiRAkBiY2Mt9jmSkFu4cKEAEE9PTzlw4IBDxzRNNHTv3l2Kioosytx///0CQEaMGOHwuYiIvPbaawJAatSoIRkZGRb716xZox/bXkKuffv2UlhY6NSxjbZv3673KivOeL3vuOMOuXr1qsX+GTNm2OxpZS8hN3jwYAEgkyZNshpTbm6uNGvWTADI2rVrzfYZn6+wsDD5559/LOq+/fbbAkAiIyMt9pVHQq5Lly4CQDp37mz1WRg1apR+3sWfxaVLlwoA6dixo56IK+6pp54SADJ27Fh9m/E5aNSokcNxHjp0SE+AFU+62VJQUKBf34SEBKtl/vzzT/Hx8RE3NzezXq5l/Z4wIecc/p4iIqq62s7cbJaQy8xmQo6Ibl7llZDjLKvknNjAyo6g/MRaH9OtLB555BE0bNjQYnvnzp3Rrl07JCUl4YsvvkCHDh0cbvOBBx7Axx9/jF27dpUqppUrVwLQxplr1qyZ0/UnTZpkdbyx/v3745tvvsGhQ4ecam/dunUAgJiYGAQFBVnsf+SRRzB58mQcP37cbjvPPfccXFxKNwxmx44dERQUhNTUVJw+fRq1a9cGoP0HhTG+f//73/Dy8rKoO2HCBLz55pu4cuWKw8e7cuUK1q5dCxcXF0ycONFqGQ8PDwwcOBC//fYbNm3ahIEDB1qUiYmJQWhoqMX2/v374/nnn0dKSgqys7Ph6+vrcGwlSU9PR2JiIgDgpZdesvosTJkyBcuWLbNa3ziG3fjx4+Hu7m61zODBg7F48WJs2rRJ3xYQEAAAuHjxIq5cuQIfH58SY42Pj4eIoFGjRoiJiSmxPKCNtZeWloYmTZqgZ8+eVsvcdtttaNu2LbZs2YKtW7di0KBBFmXK+3tCRER0I8kvFLPPnNSBiKhkTMgRlSN7g7FHR0cjKSkJ+/bts9iXnp6OhQsXYuPGjfjvf/+LixcvWkw6cPr0aafjyc/P1yeF6NOnj9P1AaB169ZWt9epUwcAkJGR4XBbubm5SE5OBgC7SckOHTqUmJC79957Szze2rVr8emnn2Lfvn04f/48cnJyLMqYJuSOHz+OzMxMANr9ssbPzw8tW7bEjh07Sjy+0d69e5GXlwelFJo2bWqz3NWrVwHA5qQDJd0LAMjMzCzXhNyvv/4KEYGLi4vNe9agQQPUq1fPIu6CggLs3r0bAPDUU09h7NixVusbn3XT+m3atEFISAjOnDmDe++9F2PHjkX37t0RGRlpM9affvoJgHPPelJSEgDg6NGjqFmzps1yxklZSntvnPmeEBER3WgKi016xUkdiIhKxoQcUTkyTYzY2nf+/Hmz7cnJyejSpQvOnTunb/P394e3tzeUUsjLy0NGRobdWVBtSU9PR0FBAQCgfv36Ttc3xmKNsfeYtdknbcnIyNBnKa1Vq5bNcsYEmT3Vq1e3ua+goACPPPIIvvrqK32bp6cnqlWrBldXVwDafSgqKjK7rqb3xl4M9u6zNcYZUEXE7D7bYqv3XUn3AnDufjjCeE0CAwPtJvrq1KljkaxKT0/XZxi9cOFCiccyJiQBIDg4GPHx8RgyZAh+++03PPXUUwCAmjVrokePHhg5cqRF0tR4bZ151o33Jjc397rem/K+L0RERFVJQbEeckXsIUdEVCIm5Mg51+E1z1vdiBEjcO7cOURFRWHmzJlo3749/Pz89P0//PADunXrZhwbkAyMiTVrli5diq+++go+Pj6YNWsWHnroIdStW9esTL169XDq1KkKua7GJGRgYKDeA+9WUGTyv+W//vorWrRo4VT9Pn36ICUlBZ9//jk2b96MnTt34vTp01i5ciVWrlyJ0aNHY8mSJeUSY79+/bB+/foytUVERHSrKijWI4495IiISla6AZiIyCp7r5Ua95n27Dpx4gR2794NV1dX/N///R969uxplowD4FCvHVtCQkLg5qbl3dPS0krdTnkJDg7Wx30z9kyyxt4+R6xduxYA8Oqrr+LZZ5+1SMYVFhbin3/+sahnem8cuZeOCgsLAwBcunRJf/XxRmG8Jsax3Gyxdk1CQ0P1xOmJEydKdfzAwECMHj0aa9aswV9//YXff/8do0ePBqAlXr/99lu9rPE6O/OsG+uUNj4iIiICCvjKKhGR05iQIypH27ZtK3FfVFSUvu3UqVMAtKSHrdcgN2/ebLNNY3LLVi8vd3d3tGzZEgDw3Xff2Ym8Ynh6euKuu+4CAOzcudNmOWfGZ7PGeF3vvvtuq/t37dpldTy5Bg0a6BNNbN++3Wrd7Oxs7Nmzx6l4WrVqBTc3N4gIEhISnKpbViU9IyW5++67oZRCUVGRzXuWkpJiNaHl7u6OVq1aAQA2btxYquMXd9ddd2HJkiVo27YtAPPvnHGbM8cyjkX422+/4a+//iqXGB1V1ntDRERUVRTvIcdfbUREJWNCjqgcrVmzxupkBNu3b9dnSX344Yf17YGB2qy1586dw99//21R7+DBg/jss89sHs84E6W91yCHDh0KAIiLi8Nvv/3mwFlcXw899BAArXeTtd5iX375ZYkTOpTEeF0PHjxosa+goACvvPKK1XpKKQwYMAAAMHfuXOTm5lqUmT9/vlMzrALa+GLGdl977TVcvnzZZtmCggJkZWU51b49jjwj9oSEhKBLly4AgNmzZ1tNHs2aNctm/eHDhwPQnr8DBw7YPZbpxAfGseds8fb2BgCze/TEE09AKYUjR45g8eLFdusbde3aFfXq1UNhYSFeeOEFh+MrD2W9N0RERFVBYZFYJOA4yyoRUcmYkCMqRx4eHujdu7c+c2NRURG+/vprDBw4EADQvXt3tG/fXi9/5513om7duicVOq8AACAASURBVBARPProo/jzzz8BaAPAr1u3Dt27d7d4hdVU48aNAQDr1q2z+SrkqFGj0KJFC+Tm5qJr166Ij4/XE0qFhYXYs2cPRo8ejZ9//rnsF8ABzz77LIKDg3Hu3Dn07t0bv//+OwAtEbV69WqMGDFC76VWWt27dwcAzJgxAxs2bNBn8Txy5AgeeOAB7N692+YEBZMnT4aXlxcOHz6M/v37IyUlBYA24cDcuXPx6quv6gk/Z8yaNQshISH4448/0K5dOyQkJOgD/YsIjh49infffReNGjVyugeePcZnJCEhodSvAsfGxkIphR9++AHDhw/XX6O+ePEipkyZgiVLlti8JqNGjULbtm2Rk5ODLl26YOnSpbh06ZK+/+zZs/j0008RHR2NefPm6ds//PBD9OzZE5999plZ3JmZmZg5cya2bt0KAOjZs6fZuRonfxg7dixiY2PNEt0pKSmIjY3FokWL9G3u7u5YsGABlFJYtWoV+vfvj/379+v78/PzsWfPHrz44ot2Z3gtDeO9WbVqldUem0RERDeC4q+rAnxllYjIISLC5SZcAOyNiooSRyUnJ0tycrLD5clceHi4AJClS5dK9erVBYD4+fmJt7e3ABAA0rBhQzl9+rRF3XXr1omLi4tezt/fXzw8PASA1K9fX+Lj4wWAhIeHW9Q9fPiwXtbNzU1q164t4eHh0r59e7NyJ06ckCZNmujHcHV1ldDQUL0uAElMTNTLp6Sk6NttSUxMtBlXSRISEsTT01M/RmBgoP65ffv2MmnSJAEgMTExZvUciUtE5MKFC3LbbbfpZd3d3SUgIEA/9+XLl+v3zPS8jVatWiWurq56/aCgIHFzcxMAMmDAABk6dKgAkKlTpzoV3+7du6V27dpmcRW/DwBk69atZvXsxWpkrJuSkmK2/fz58xISEiIAxMXFRWrWrCnh4eFO37fZs2frx1BKSXBwsH6NJk6cKNHR0QJAli9fblH33Llz0r59e72+i4uLhISEiK+vr9l5x8bG6nXee+89s32+/8/enYdJUZ5rA7+rl+lZmWGZAQRkMAuIGg1qgguCehKM0UQNcYlrFtxjotHEfGqUE81y1OhRT06MGvckgkkkMdEcg4IL0Qi4ISgaBQQBgYHZe3+/P2a6+q2qt3qtnqrquX/XxeXYPd1dXV09M3X387xPQ4NoaWkxXGY+PoQQIhqNipNPPtnwfS0tLYbHMr9uQgjxm9/8xvA61NXViVGjRhmOA/PrWu77ZMmSJfrta2pqxMSJE8XkyZPFKaecYv9CDGP8PUVE5E090YSY/IPHDf/e3trl9mYREVXMjBkzBICVoszchhVyRA76+Mc/jhUrVuAb3/gGmpubkUql0N7eju9973tYsWIFxo8fb7nNiSeeiKeffhqf+9zn0NTUhEQigcmTJ+Pyyy/HK6+8YhlIIJs2bRqeeuopHHPMMWhubsbWrVuxYcMGfQ21jEmTJmHFihW47bbbcPjhh6OpqQk9PT0YP3485s6di7vvvhuf+cxnHN8fdubOnYsVK1Zg3rx5GD16NGKxGKZMmYIFCxZgyZIl6O/vB4CSK+VGjRqFF198ERdccIG+/+rq6nDCCSdg2bJlehulnVNPPRUvvPACvvjFL6KlpQXxeBzTp0/HrbfeioULF0LTtJK26+CDD8Zbb72Fn//85zj00EPR2NiI3bt3o76+HgcddBAuueQSLFu2DLNnzy7p/lXGjBmDZ555BieddBJaW1uxfft2bNiwoeghH1dccQWeeOIJHHnkkWhsbEQymcRBBx2EBx54ADfffHPO27a1tWHZsmV4+OGHceyxx6K1tVVv2502bRrOOussLFy4EFdeeaV+m6997Wu46667cMopp2DvvfdGOBzWj9kvfelL+POf/6xsS41EInjkkUewePFiHH/88Rg7dix6e3vR1NSEmTNn4oYbbtCHQsi+/vWv4+2338Z3v/td7LPPPggGg+jq6sLo0aMxZ84cLFiwAG+//XZR+yyfo446Cn/6058we/Zs1NXVYfPmzdiwYQO2bt3q6OMQERFVknn9OIAVckREhdCE4A/LaqRp2soZM2bMWLlyZUHfv3btWgADLZREbps1axaef/553HvvvXnDMyIaHvh7iojIm3b2xHDg9cYhZI9/+3DsO6H4JT6IiPzgwAMPxKpVq1YJIQ4s535YIUdEnvLPf/4Tzz//PAKBAI4++mi3N4eIiIiIclBVyKVZ9EFElBcDOSIacr/+9a/xk5/8BP/+97/1gQs9PT144IEHcNxxxwEATj75ZEyaNMnNzSQiIiKiPNiySkRUmpDbG0BEw8/GjRtxww034KqrrkIwGERzczN2796N9OCUrgMOOAC33367y1tJRERERPkkU9Ypq6yQIyLKj4EcEQ25U089Ff39/Vi2bBk2bdqEjo4OjBgxAtOnT8e8efNw/vnno66uzu3NJCIiIqI81BVyLmwIEZHPMJAjoiG377775p3MSURERETel0yxZZWIqBRcQ46IiIiIiIhKkkyzZZWIqBQM5IiIiIiIiKgkrJAjIioNAzkiIiIiIiIqiXINOVbIERHlxUCOiIiIiIiISqKcssoKOSKivBjIERERERERUUnUU1YZyBER5cNAjoiIiIiIiEqiCuQ41IGIKD8GckRERERERFQSVcuq4iIiIjJhIEdEREREREQl4VAHIqLSMJAjIiIiIiKikiRTipZVriFHRJQXAzkiIiIiIiIqSTKtalllIEdElA8DOSIiIiIiIiqJqkKOLatERPkxkCPyuOuuuw6apuGcc85x9H7nzJkDTdNw3333FX3bc845B5qm4brrrnN0m9zU3t4OTdOwdOlStzfFc+677z5omoY5c+ZYrhuu+61S70siIiK/UVXDsWWViCi/kNsbQERUqKVLl2Lp0qU44IADcMIJJ7i9OeRBr776Kh577DG0t7czLPOYxx57DK+++irmzJmjDHeJiMifEqqWVVbIERHlxQo5Io8bM2YMpk6divHjxzt6v3vuuSemTp2K5uZmR++3kpYuXYoFCxbgsccec3tTaNDHPvYxTJ06FfX19W5vCoCBQG7BggUlVX5SZT322GNYsGDBsKumJCKqdsoKOeZxRER5sUKOyOMuvvhiXHzxxY7f7wMPPOD4fdLws2TJErc3gYiIiFyU4JRVIqKSsEKOiIiIiIiISpJMccoqEVEpGMgROUBe2H7Lli04//zzMWnSJNTV1WHvvffGLbfcgrS0vsaiRYswa9YstLS0YMSIEfjiF7+I1atXK+871+LxmqZB0zSsX78eGzduxPz58zFx4kREIhFMmTIFl19+Obq6upT3W85QB1k0GsW1116LadOmoa6uDm1tbTjttNOwbt0629s8++yz+M53voPPfvaz2GOPPVBTU4O2tjYcc8wxePTRRy3fv379emiahgULFgAA7r//fv25y/vA7Mknn8S8efP0fTJu3DjMnDkT119/PT744APb7evo6MBll12GKVOmIBKJYMKECZg/fz62bNlS/A4atGvXLlx66aVob29HJBLBpEmT8K1vfQsffPABli5dCk3T0N7ebrmdfGxt3rwZF154Ifbaay9EIhEccMAB+vdt2rQJN910E4455hh84hOfQH19PUaMGIFPf/rTuPbaa7F79+6c2/fhhx/i3HPPxYQJE1BbW4u99toLl112Wd7b5RvqEI/Hcccdd2DWrFkYNWoUIpEIJk+ejG984xtYu3at8jby0JBUKoVbb70V+++/P+rr6zFq1Cgcd9xxWLFiheV2mqbh61//OgBg2bJllmOk2FbJt99+G6eddhra2tpQV1eHadOmYcGCBYjFYgXd/i9/+Qu+/OUvY9y4cfoxfvzxx+Pvf/+77W1ee+01nHXWWfpx0tTUhL322gvHHHMMbr31VvT19Slvt3btWpx//vn45Cc/ifr6erS0tGC//fbDJZdcgpUrVypvs337dvzwhz/Efvvth8bGRjQ0NGDffffFVVddhY6ODuVt5Ne70PdJ5vi+//77AQALFiywvDZERORfSWXLKgM5IqK8hBD8V4X/AKycMWOGKNSaNWvEmjVrCv5+Mpo8ebIAIH7zm9+IcePGCQBixIgRIhgMCgACgLj44ouFEEL84Ac/EABEMBgUTU1N+vUtLS1i3bp1lvu+9tprBQBx9tlnW67L3Paxxx4To0aNEgBEU1OTCIVC+nUHHXSQiMfjltvOnj1bABD33ntv0c/37LPPFgDElVdeKWbOnCkAiJqaGjFixAj9cevr68WyZcsst+3u7ta/J7O98u0AiHPPPddwm40bN4qxY8eKhoYGAUDU1taKsWPHGv5t3LhR//5YLCbOOOMMw302Nzfrtwcgrr32WsNjZF7DBx98UP+6vr5eRCIR/Tbt7e2io6Oj6P31wQcfiPb2dv1+6urqRGNjowAgWltbxV133SUAiMmTJ1tum9mWO++8U4wZM0bfroaGBrH//vvr3/eVr3xFv/+amhoxatQoEQgE9Ms+9rGPiQ8++EC5fWvWrBGtra369zY0NIi6ujoBQHz84x8XN998swAgZs+ebbt9zzzzjOW6Dz/8UOy///76/QYCAcMxX1tbK/7whz9Ybpc5vq666ioxd+5cAUCEw2F9n2Vuu3z5csPtxo4dqx9L4XDYcoy88MILuV8oybJly0R9fb3+eCNGjBA1NTUCgDjkkEPED3/4Q9v3ZTweF6effrrh+DMf49///vctt/vrX/8qwuGw/j2RSMRyu7Vr11pud9tttxl+1jQ0NIiWlhb9/1Wv23PPPaf/zMgcM7W1tfr/T5o0Sbz11luW25XyPnnhhRfE2LFj9ftvaGiwvDaF4O8pIiJvuvWpdWLyDx43/PvV0nfd3iwiooqZMWOGALBSlJvblHsH/OfNfwzkhlbmxLS5uVkccsgh4rXXXhNCCNHb2yt+/OMfCwBC0zRxww03iHA4LG699VbR09MjhBDijTfeEFOnThUAxFe/+lXLfRcSyLW0tIijjjpKvPHGG0IIIaLRqLjnnnv0k+T/+Z//sdzWiUCuublZ1NfXiwceeEAP/V555ZXMDygxduxYS4DV29sr5s2bJ/70pz+JnTt36pfv2rVL3HHHHXrosnDhwqL2hezCCy/UQ89rr71WbN26Vb/uvffeEzfeeKO48847DbfJvIYtLS3igAMO0MOeRCIhFi9erAccV1xxRVH7Sojsvh47dqx4/PHHRSqVEkIMBBVTp07V7ztXINfY2Cj2228/Q6j0zjvv6F9fffXV4rbbbhPr1q3T7z8ej4ulS5eKgw8+WAAQxx57rOX+4/G4mD59ugAg9tprLz1ETaVS4s9//rNobW0Vzc3NRQdy8Xhcf9yjjz5aLF++XD9GPvzwQ/Hd735XD3Pefdf4R3vm+GppaRGjRo0SjzzyiIjFYkIIIV577TWx7777CgDi4IMPtmzPvffea7uthero6BBtbW0CgJgxY4Z49dVX9ed0//33i/r6en2fqI7FzHP7+Mc/LhYuXKi/17u6usQvf/lLPZT87W9/a7jdlClTBABx3HHHibffflu/vLOzUzz77LNi/vz54v333zfcZuHChfrPgXnz5hl+ju/cuVM89NBD4rLLLjPcZv369foxd8EFF4h33nlHpFIpkUqlxBtvvCE+//nPCwBi+vTpIplMGm5bzvsk87qaw/BC8fcUEZE33fz3tyyB3P88807+GxIR+RQDOf7L/cIykBtSmZPUkSNHil27dlmuP+qoo/ST5gULFliuf/bZZ/WKmEzwkFFIILfPPvuIaDRquf7iiy8WAMSRRx5puc6JQA6AeOihhyzXb9++XYwePVoAED/+8Y+Luu8HHnhAABBz5syxXFdIILd69WqhaZpeVVaozGs4duxYsWPHDsv1N910kwAgpkyZUvB9CiHE008/rQeyzz//vOX6999/X69GyxXItbS0GILFYuzcuVO0trYKTdMsgU5mf9fU1CgrojLHZrGBXKbqb9asWcoKTSGEOO+88wQAcdFFFxkul4+v5557znK7FStW6Ndv2LDBcJ0Tgdx//ud/CgBi9OjRYvv27ZbrH3zwQf3xzcfiunXrhKZporW11VC1Kfvd736nv28ztm3bpt9noa9zPB4XEyZMEADEaaedVvDzy1TvXXnllcrrY7GY+NSnPiUAiEWLFhmuK+d9wkCOiKg6/eyJtZZA7vYl1q4PIqJq4VQgxymrVJT97t/P7U1wzBtnv+H4fZ5//vloaWmxXP4f//EfePrpp1FTU4PLLrvMcv1hhx2G2tpaRKNRvPvuu5g+fXpRj3vZZZchEolYLj/hhBNwxx132K5PV67Jkyfja1/7muXyMWPG4LzzzsNPfvITPProo7j66qsLvs/jjz8eAPDiiy8ilUohGAwWtU0PPvgghBCYNm0azj333KJuCwDnnnsuRo8ebbn8hBNOwOWXX473338fvb29aGhoKOj+/vjHPwIYeI0PO+wwy/Xt7e049dRTce+99+a8n7POOgtjx44t6DHNRo0ahUMPPRSLFy/G8uXLDWvVZdbsO+mkkzB16lTLbWfNmoUjjjgCzz77bFGPmVkv7Dvf+Q7C4bDye04//XTceeedeOqpp5TXz5o1C4cffrjl8gMPPBATJ07Epk2bsHr1auy5555FbVs+mX0yf/58jBkzRrndV199NTZs2GC57oEHHoAQAqeccgomTZqkvP958+bhnHPOwZtvvoktW7Zg/PjxaGxsRCAQQDqdxpYtWwp6rZcsWYLNmzcjGAzixhtvLOi59fX1YdGiRQgEAsqfRQBQU1ODefPm4fXXX8dTTz2FefPmWb7H6fcJERH5l2qAg2LOAxERmXCoA5GD9ttPHVi2tbUBGAhfGhsbLdcHAgH9xH/Xrl1FP+7BBx+svHzChAkl32chZs+ebbsg++zZswEAq1evRjweN1yXTCZxzz334JhjjsH48eMRiUT0xd1HjhwJYGBYRCnb/eKLLwIAjj322KJvC+TflwDyDjqQvfLKKwCgDJYyZs2alfd+DjnkkLzf869//Qvf+MY3MG3aNDQ2NhoWzV+8eDGAgeENslWrVgHIvl4qua5TSSaT+Ne//gUAOO+88zBu3Djlv5NOOgkAbAds2L0WQOWO7Xg8jjfffBOA/fPWNA1HHHGE8rrly5cDGAgk7Z73xIkTkUgkAGSfe319vf54c+fOxfXXX49XX30VqVTKdlszx/r+++9vOD5zWblyJeLxOIQQ2G+//Wy38aabbjJsn5nT7xMiIvKvhGrKqrCGdEREZMQKOSIHjR8/Xnl5psrL7nr5ezIn6sVoampSXl5bWwtgICApxPLly/WQxOzll1+2VPzkCgEy16VSKezatUuv+Onp6cHcuXP14AIA6urq0NraikBg4DOCbdu2AQB6e3uVFUq5ZG5batVUvn0JFPca7dixA0Du136PPfbIez+tra05r7/pppvw/e9/P9OyjmAwiJEjR6KmpgYA0NnZiWg0it7eXsPttm/fnncbCg17Mjo6OvQQdufOnXm/v7+/X3m53WsBZF+PUt4vuXR0dOghWCn7JDNhtLu7G93d3XkfT56aevfdd+O4447D2rVrcc011+Caa65BY2MjjjjiCJx22mk49dRTEQplf22Xcqxntk8Iod++0O2TOf0+ISIi/1JVyKUVlxERkREDOSpKJdo8yTvi8bjtSXquSp1i/PjHP8by5csxZswY3HzzzTjmmGP0CsLM42RCB8FPV3W5WnfffPNN/OAHP4AQAhdffDEuuOACTJ061XCbM888Ew899NCQ7NN0OvtJ+SuvvIIDDjig4o/pFZnnfsstt+C73/1uUbfda6+98Prrr+Pxxx/HE088geeeew5r167F3/72N/ztb3/DLbfcgmXLlimrbIvdvubmZlawERGRIxIpRcsq/4YjIsqLLatEpJszZ47tgpPyumMZ5vZH1XWZSq2MRYsWAQBuv/12nHXWWYYwDkBBVTu5ZCrxVOt7uSFT4ZepTFLJdV0h/vCHPyCdTmPu3Lm4/fbbMX36dEuAZ7dfM5V3hbyWhRo9erT++Bs3bizqtm4bNWqUvu2l7JPM8Vfq8w6FQjjhhBNw5513Ys2aNdiyZQtuvPFG1NbWYtWqVViwYIHlsYo51jO36erqQmdnZ0nbSEREJEulrS2rrJAjIsqPgRwRlWzZsmV5r9t33331tkkA2LRpEwDg05/+tPJ2//jHP2zvM9PSmqvKa+bMmQCAJ554wvZ7hlLmeT7//PO23/Pcc8+V9Rj59mlvb6++3pjZjBkzACDn0IZcr7NKOBzGQQcdBGDoX4dCjpFcampqsM8++wCw3ydCCNvrMmv9PfnkkyU9vtm4ceNw+eWX69V28muROdZff/11bN68uaD7O+iggxAKhSCEcGwbC1Xua0NERN6UVFXIMZAjIsqLgRwRlWz9+vX43e9+Z7m8o6MDv/71rwEAX/3qVw3XNTc3AwDeeMPa/tzT04MbbrjB9vFGjBgBIPdi8WeeeSY0TcNbb72FO++8M/+TqLATTzwRAPDCCy/gn//8p+X6jRs34ve//31Zj5FrnwLADTfcYLueWeb1+eMf/4h33nnHcv3y5cuLnrAKAOeccw4A4L777sNrr72W83udHMxQyDGST2af3HXXXejo6LBc//vf/x7r169X3vass86CpmlYu3Zt3uNPft6JRCJnUFVXVwcAiMVi+mVHH300JkyYgFQqhSuuuCLnY2U0NTXhK1/5CgDgRz/6Uc517pLJJHp6egq630I48doQEZH3JFRTVvnhCxFRXgzkiKhkzc3NmD9/Ph5++GF9cMTrr7+OuXPnYvv27Whra8OFF15ouM3nPvc5AMBll12GZcuW6SHEyy+/jKOPPjrnEIBM5dLzzz+vDI8y33PeeecBAC666CJcd911+Oijj/Tr33//fVx33XX41a9+VeKzLs6RRx6JWbNmQQiBr3zlK3jiiSf05/ziiy/imGOOMVQQliKzT//617/ipz/9qb4Q//bt23HFFVfgpz/9KUaPHq287SmnnILp06cjFovh2GOP1Sv50uk0/vrXv+Kkk07Sg5RifPOb38TMmTMRjUZx1FFH4a677kJXV5d+/datW/Hwww9j9uzZ+O///u+i799O5hhZs2YNXnrppZLu46KLLkJbWxt27NiBuXPn4vXXXwcwEJo99NBDmD9/vh6Cmk2fPh2XXnopAODCCy/ED3/4Q72CERgY9vB///d/OOOMMwxh9Ztvvol9990Xt956K9atW6cfI4lEAn/4wx/wi1/8AsDABNaMcDiMm2++GQDwu9/9DieffDLeeust/fqOjg7cdddduOSSSwzb+LOf/QyjRo3CunXrcOihh+LJJ5/UBzAIIfDOO+/gF7/4BaZNm4YVK1aUtA9VMq/Nk08+WXabNhEReQdbVomISsNAjohKdsEFF2DffffFGWecgcbGRjQ3N2P//ffHihUrUF9fj0WLFhnWjwOA66+/HmPGjMEHH3yAOXPmoL6+Ho2NjfjMZz6DN954A7/97W9tH2/OnDn42Mc+ho6ODkydOhVtbW1ob29He3u7IfS49dZbcfLJJyOVSmHBggUYO3YsRo4cicbGRuy1115YsGABtm7dWrH9ItM0DQ899BD23HNPbNmyBcceeywaGhrQ1NSEQw45BB0dHbjpppsAAJFIpKTH+PznP69Px/1//+//obGxEaNGjcLYsWNx00034Zvf/CaOO+445W3D4TAWLVqE1tZWvPvuu5g1axaamprQ2NiI4447Dk1NTfjRj35U9DaFw2EsXrwYhx12GDo6OnDuuedi5MiRGD16NBobGzF+/HicccYZePbZZ6FpWknPW+UTn/gEjjjiCCSTScycOROjR4/WjxG7tl2zkSNHYuHChairq8OKFSuw//77o6WlBU1NTTjzzDPxqU99ChdccIHt7f/rv/4LF1xwAdLpNH72s59h0qRJaG5uRktLC5qbmzF37lw8/PDDlkEpa9aswaWXXoqpU6eirq4Oo0ePRm1tLebNm4fOzk4cdNBBuPrqqw23OeWUU3DzzTcjEAhg0aJF2HvvvdHU1KTv63PPPVcPFDPa29vx5JNPYo899sDq1avxhS98AQ0NDRgzZgxqa2vxyU9+Et/73vfw73//29HX5sQTT9SDwIkTJ2L8+PH6a0NERP7FoQ5ERKVhIEdEJYtEIli6dCl+9KMfYfLkyYjH42htbcWpp56KVatW4YgjjrDcZq+99sK//vUvnHHGGWhra0MqlUJLSwtOP/10vPzyy/j85z9v+3jhcBhLlizBmWeeiQkTJmDXrl3YsGEDNmzYoFfoZbbrkUceweLFi3H88cdj7Nix6O3tRVNTE2bOnIkbbrgB8+fPr8g+Udlzzz2xatUqXHLJJdhzzz315zx//nysXLlSr15raWkp+TEeeeQR/OxnP8Pee++NcDgMIQQOO+ww3H///bj77rtz3nb69Ol49dVX8a1vfQvjx49HIpHAuHHjcOmll+Lll1/GqFGjStqmtrY2LFu2DA8//DCOPfZYtLa26i2S06ZNw1lnnYWFCxfiyiuvLOn+7fzxj3/EhRdeiClTpqCnp0c/RqLRaMH3MXv2bLzyyis45ZRT0Nrailgshvb2dlx33XV4+umnc4anwWAQv/zlL/H888/jjDPOwOTJkxGLxRCNRrHnnnviS1/6Eu644w48+uij+m323ntvPProozj//PPx6U9/Gi0tLejq6kJzczMOP/xw3H777XjhhReU1YqXXXYZXnnlFXz9619He3s7EokENE3Dpz71KXznO9/BLbfcYrnNwQcfjLfeegs///nPceihh6KxsRG7d+9GfX09DjroIFxyySVYtmwZZs+eXfA+y2fMmDF45plncNJJJ6G1tRXbt2/XXxsiIvIv1XpxKWvRHBFVgV29cezqjbu9GVVD4+LK1UnTtJUzZsyYsXLlH7eRXgAAIABJREFUyoK+f+3atQAGTgqJaGhdc801uP7663H22Wfjvvvuc3tziDyJv6eIiLzpzHtewnPv7DBcdspBk/DzeZ9yaYuIqBLe/LATJ/5yOSCAP110KPbZQ72Ey3Bw4IEHYtWqVauEEAeWcz+skCMiclFHRwfuueceANm14IiIiIj8Qlkhx6IPoqrz1JptiCfTiKfS+Pub29zenKrAQI6IqMJeeuklfPvb38aKFSv0tslkMomnn34aRx55JLZs2YL29nZ9+iURERGRXyQVa8hxqANR9Ykls73o0UQqx3dSoUJubwARUbXr7u7GHXfcgTvuuAPAwNCA3t5exOMD6y+MGjUKjzzyCGpra93cTCIiIqKiJRVTVlkhR1R9ElIgF2Mg5whWyBERVdgBBxyA66+/HrNnz8akSZPQ19eHcDiMffbZB9/73vewevVqfOYzn3F7M4mIiIiKllQOdWAgR1RtEtK0FrlajkrHCjkiogobM2YMrrrqKlx11VVubwoRERGRo5Qtqz6vkNvRE8O2ruiwXrSeyCwuvdfZsuoMVsgRERERERFRSZQtqz6ukNvRE8PhP38aX7zteTzy8ka3N4fIM1gh5zwGckRERERERFQSdYWcCxvikJfe60A0MRA2PPPWdpe3hsg7GMg5j4EcEREReZrweesTEVE1U60h5+cpq3LFnxxAEA13xkCOLatOYCBHAABN0wAAaUXJORERkZsygVzmdxUREXlHUhFa+XnKqrz+XcLHwSKR0+LJ7PshlmBu4AQGcgQACIfDAIBoNOrylhARERllfjdlflcREZF3VNuUVTlfVIWNRMMVW1adx0COAABNTU0AgF27drE1iIiIPEMIgV27dgHI/q4iIiLvULas+vh8Qm63Va2PRzRcye3cnLLqjJDbG0DeMGLECHR0dKCrqwsAMHLkSNTW1kLTNLYIERHRkBJCQAiBaDSKXbt2oaurC5qmobm52e1NIyIiE2XLqp8r5Awtq6wCIspIyC2rrJBzBAM5AgDU1tZi4sSJ2LRpE7q6uvRgjoiIyG2apmHixImIRCJubwoREZmohzq4sCEOSbFCjkgpzqEOjmMgR7rGxkZMmTIFnZ2d6O7uRiKRYPsqERG5QtM0hMNhNDU1obm5mWEcEZFHKdeQ8/E5hGGoA9eQI9JxDTnnMZAjg0gkgra2NrS1tbm9KURERERE5HFV17IqV8j5+HkQOc0QyHHKqiM41IGIiIiIiIiKlk4LqDIrPw91MLasMnQgykik5DXkUuymcwADOSIiIiIiIiqaXQWZnyvkjC2r/n0eRE6LS22qacEKUicwkCMiIiIiIqKiJW2mN/g5kJOL4uyeH9FwZF5TMZrgYIdyMZAjIiIiIiKiotlVyPi5ZVXedk5ZJcoyB3Ic7FA+BnJERERERERUNLvAyt8VcpyySqRibuFmIFc+BnJERERERERUNLuWTh/nccYKOT8/ESKHxc0VcmxZLRsDOSIiIiIiIlLa3h3Djp6Y8rpqrJBLp9mySmQmhGDLagUwkCMiIiIiIiKLNzZ14pCfLsEhP12CNz/stFxvF7z5OZBLyVNWOdSBCMDAe9q8NCQDufIxkCMiIiIiIl9796MeLFzxAbqjCbc3pao8tWYrkmmBRErgH2s+slwvV8zUBLOnln4e6iAXAQnh73CRyCnm9eMAtqw6IeT2BhAREREREZWqP57CV/53OTr7E3j5/Q7c+NX93d6kqtEvnXBHk9aTbzmsioQC+hpTfg6xzGFiIpVGMBB0aWuIvMG8fhwARFkhVzZWyBERERERkW+9t6MHnf0DlXGvbdrt8tZUl2gie8IdV5x8y1UzkXC1VMgZt52DHYiApCKQY4Vc+RjIERERERGRb8nLfPm5MsuLYlJVnCqQk6esyi2rfn4dLIGcIoggGm6ULauskCsbAzkiIiIiIvItORTycQ7kSfIJd0zRsipXj0XC2bZOPwdy1pZV/z4XIqeYJ6wCDOScwECOiIiIiIh8Sw5Q/BwEeVEsT8tqUgqrjEMdKrtdlWRtWWXoQKRaQ04V0lNxGMgREREREZFvyaEQAzlnGVpWFSfkclglryHn59fBXCGXZIUckbpCLsGwulwM5IiIiIiIyLdSUoAifDxMwIvklrTiKuT8+zqYw0RVEEE03CSS1ve0avIyFYeBHBERERER+ZYcoKR8HAR5UVSaoqhaLyqVrsYpq8b/5xpyRDYtq6yQKxsDOSIiIiIi8i1DIMfzQ0flq5CTq8eqZcqqdagDDyoiDnWoDAZyRERERETkW3L44+fKLC8yBHKKE3J539eEjEMd/No+bB3q4M/nQeQkdSDHltVyMZAjIiIiIiLfSqY51KFSDEMdVBVy0v4OBwPQtOx1fn0pzG3PSVbIEbFCrkIYyBERERERkW+lWSFXMfIaUapALiVNWQ0FNASlRM6v4WjaMtTBn8+DyElxxVAHriFXPgZyRERERETkW3KFnDlMofLIQx1ULatyWBUKBhAIZAM5v4aj1pZVhg5EbFmtDAZyRERERETkW3LwwymrzpJb0lTVMMmU3LJaJRVylpZVfz4PIiepArkoK+TKxkCOiIiIiIh8Sw5MWMzkHCFEAUMdspcFAxqkAjnfhqPmIJFTVolYIVcpDOSIiIiIiMi3UqyQqwhzAKcc6iC3rAZMLas+rZAzF8RxyioREFdUinKoQ/kYyBERERERkW+lOGW1Iswn2+qhDnIgpyEYqIKWVVbI+d4vl76Lr/5qOV58b6fbm1I1VNOGGciVj4EcERERERH5lrmCSbBKzhHyQAdgoGLOvG8T8pTVYMC4hpxPXwfLUAeuIecrWzr7cePf38bL63fhF/+3zu3NqRrKltUEW1bLxUCOiIiIiIh8y1zR5NfKLK9RDXEwt7GmUsYKOWPLauW2rZLMQSKnrPrLls4oMi/h9p6YuxtTRRKKYFpVNUvFYSBHRERERES+Za6Q82tllteo2tHMJ+AJuWXVPGXVp6+DtWXVn89juOqNJfWv2W7sHFX4Zq6ipeIxkCMiIiIiIt8yBygsaHKGaoKi+aRcnrJqXkPOv0MdzC2rPKD8pCfKQK4S1FNWuX/LxUCOiIiIiIh8ixVylaGskDOdlMvrq4WCAQSks0u/tg6bg0ROWfWXHkOFHF87pzCQqwwGckRERERE5FtpUwBn/n8qjaodzbyuXNI8ZbUKWlbN281Qx1/YsloZqveBqoqWisNAjoiIiIiIfMs8BdOvrZJeU1iFnLFlNVAFLavmlme2rPpLDwO5ijC/94GBkM6vlbBewUCOiIiIiIh8y1zRxBNEZyinrOYc6hCoigo5c4VlgseTr/TEslVb5rCeSpewaU/lpNXyMJAjIiIiIiLfSplKmvwaBHmNqh3NXDWXSplaVqUKOb8Go+btZoWcv/TEEvrXybTwbaWm19hVG7JttTwM5IiIiIiIyLfMi+5zyqozlC2rlgo5qWU1GEBAk1tWK7dtlWSZsspAx1d6Y8aAKOHXA9Fj7NZSjCoqaalwDOSIiIiIiMi3zBUwrJBzRiFryKXMQx0CVdCymjYPdWDg4Cfd0aTh/9m26gzVGnIAK+TKxUCOiIiIiIh8y1ohxxNwJ8QUU1bNFXJy2BEKGoc6+LZl1Vwh58FARwiBl9d34I1NnW5viufIU1YBBqpOsW9Z5f4tBwM5IiIiIiLyLXMAZ16Un0pTSMtqMm2cshrM5nG+fR0sU1Y92PL41Jpt+Oqv/onj73geKzfscntzPKXHEsj58zj0GttAji2rZWEgR0REREREvmWukPNrZZbXKCvkUsbLDBVygYChZdWvlYrm48eLgc7z7+7Qv/7X+x0ubon3sEKuMuzeB2xZLQ8DOSIiIiIi8i1zJZZfK7O8RlUhZ66GkcPQYFAzDHXw6xpy1pZV7wU6nf3SJFEPbp+buhnIVYS8H2vD2RiJLavlYSBHRERERES+ZV7ji+ffzihkqIPczhm2VMhVbtsqyTLUwYOVfnIg58Xtc5O1Qo77xwlyINcYCetfRxWVtFQ4BnJERERERORb5oomVsg5Q3WibV5DTg47glUyZdUPFXK7+1ghp5JKC/TFjcctK+SckUhm3xdNtSH9a1bIlYeBHBERERER+ZZ5zS+uIecMZcuq6TJ5X4dNLavVsoacF6esdsktqz7dz5XQG09aLvPi6+dHxgo5OZBjhVw5GMgREREREZFvmQMJVsg5Q3WibZmyKp2kh4LGllW/BqN+a1k1vybDWU/UGsiZ26ypNHG7QI5TVsvCQI6IiIiIiHzLHKD4NQjyGtWJtnUNOXnKKoc6DAUhBHYbKuS8tX1uMq8fB7Bl1SmGCjm2rDqGgRwREREREfkWK+QqQznUwVIhJwVyQQ1B6ezSry2r5nzLay2PvfGUIXT22va5yTxhFeD+cYq8XmQTW1Ydw0COiIiIiIh8yxz8+DQH8pxChjrI1VmhKh3qkPBYBZrcrgpwiqiMFXKVk0jaVMixZbUsDOSIiIiIiMi3zBVybFl1RkEVcoaW1QA0zf9ryHl9qENnnzGQY8tqlmoNOQZyzrBbQy7KCrmyMJAjIiIiIiLfMreo+rVV0muUQx3Ma8hJYVUwoCEoT1n1YYWc6tjxWqCzuz9u+H+vBYZu6lFWyHH/OEF+HzTVhvWvWSFXHgZyNjRNG6Fp2vWapq3VNK1P07SdmqYt0TRtXhn3OUHTtCs0TVukadoaTdN2aJqWGPzvMk3Tvq1pWsTJ50FEREREVM3MgYRfWyW9RlUhZw7p5OqssGXKauW2rVJUx465AtNtXZaWVR/u6ApRB3LcP+VKpYW+FEBAA+prgvp1HOpQnlD+bxl+NE2bCOBZAFMGL+oBMALAUQCO0jTtf4UQF5Zw17MA/Jf0/zEAfQBGAzhi8N8FmqZ9TgixudTtJyIiIiIaLswhil9bJb1GOWXVdPIt7+ugacqqHysVVceO1wId8xpyXgsM3cQ15CpD3ofhYACRULaui0MdysMKORNtYOGDRzEQxq0HcJgQoglAE4DvA0hjIDSbX8LdbwSwAAPB3mghRK0QohlAM4BzMRD87Q3ggXKfBxERERHRcGAOUfzYKulFqrWhzNUwcjtg2DRl1Y+Viqpjx2stobv7WCFnRzVllS2r5ZNb1WuCAUTCciDH468crJCz+jKAz2IgeDtRCPEqAAghogBu1DRtDwDfBfCfmqbdL4SI29+VkRBiOYDlisu7ANylaVoawN0YqMKbJIT4oPynQ0RERERUvcwVQlzj3hmFVMglpRP1oHnKqg8rt3xZIcfASccKucqQJ6yGQwFEQlLLKteQKwsr5KxOH/zvPzJhnMlNAASAcRiodHPSy9LXezh830REREREVcfcGunHyiyvEUIUNtQhLVfIBYwtqz58HVRhrtdaQq0tqwxEMnpj1mOWgVz5jO9zzdCyyimr5WEgZ3Xk4H//rrpycG23Nwf/1+lA7lDp6/UO3zcREVFZ3t/Ri+sfX4MX3t3h9qYQEemsFXLeClD8KCkt4i6zVMhJ3xSqhgo5RYjotUDHHMixJTOrO8qW1UqQ3/fhYAC1YVbIOYUtqxJN09owMGAByIZuKmsA7AtgugOPWQNgAoATAfx48OJFQohtBd5+pc1V08rdNiIiItk1j63G8+/uwCMvf4CXrjoa9TX8M4KI3McKOefZrQsln5gLIXIOdfBlIKfYZq+1hLJCzp6qZTXpsUDVjzjUoXL4l7TReOnrD3N8X+a68Tm+JydN094F8DHTxQLAIgDfKPV+iYiIKuXf23sADCya/FFXDO1j+GcEEbnPHEj4MQjymlhCfZItt6yaq+M0zVgh58uWVdVQB48FXlxDzl4P15CrCPPwFsMachzqUBb+JW3UIH3dn+P7+gb/21jGY20fvH2DdD8LAVwthOgp9E6EEAeqLh+snJtRxvYREREZyH+Qee0EhYiGL3P+5sMcyHOi0km2pmX3qVwhZ66Ok/87cH2FN7IC1EMdBIQQ0KTqPzdZW1Z9uKMrRFUhF2dgWTZLhRynrDqmKtaQ0zTtR5qmJUv8d4Mb2yyEOEQIMU4I0YSBARFXATgOwBuapn3VjW0iIiLKRf6DjGuyEJFXsELOeXKFXKO0PIF88m0+SQfg/6EONtvspWNqd5+5ZdU72+a2brasVkScLasVUxWBHAaeR7CMfxm90td1OR6vfvC/BVey5SKE2CaE+AmA0wDUArhP07QJTtw3ERGRU+Q/atkiQ0RekUpxDTmnycFbU202kMtfIQfl9X5hV/ztldArnRboipoq5FihpFNVyLGCsHzyMVYTDBhbVjnUoSxVEcgJIa4TQmgl/rtSuit53bg9cjxk5rotDj+PvwDYgIHA71Qn75uIiKhcCemEJMGWVSLyCHMAxymr5TMGcmH963hSXSkdDg4Gcn4f6mAT5nol1OmOJS0t2Qkf7udKSKUF+uLWai3un/IZ3ushzVAhF7VZb5IKUxWBnFOEENsB7Bj8331yfGtmuuqaCmzG5sH/mgc+EBERuYoVckTkRebghxVy5ZNbVg0Vcqk0xOD+VVXIBXw+1MEuRPTK77wu0/pxAFsyM3rj1uo4gBWETjC3p9eGOdTBKQzkrJ4Z/O/nVFcOtpJmwrolFXj89sH/OtIOS0RE5IR0WhgWTucJABF5hTlEYYVc+eShDnU1Qb0CDsiuJyWfpIcCA6eVwSpdQ84rVeHm9eMA74SFbuuJ2gRy/HulbOY15GpCxqEOwofvda9gIGf128H/fl7TtP0V118GQMNAu+oziuttaZqWc6qtpmmnI9sO+1wx901ERFRJ5pMRtoAQkVeY1/fij6fyyRVykVAQNdLicJm2VXm/h4LWCjk/5iBer5AzT1gFvBMWuk21fhzAv1ecIIeaNcEAggFNGdJT8RjIWS0G8BIG9s2fNE2bCQCapkU0TfsegO8Oft+1Qoi4+caapq3XNE1omnaf4r6f1TTtSk3TpmuaFpRus6emadcC+M3gRSsB/NW5p0RERFQe88kIK+SIyCvMFXF+XLvMa+Q2tEjYWBGTCeRSablCLjPUwd8Vcn4M5LyybW5TTVgF2LLqBGPL6sB73DDYgfu4ZDkrtoYjIYTQNG0egGcBTAHwT03TejAw/TSzv34lhLirhLvfA8BPB/8lNE3rGrzfBul7XgbwJSEEj2oiIvIM8x/8CZ4AEJFHWCvk+POpXIZALmQK5PSWVXmog7Vl1Y/BqOdbVvst9SBIpgWEENCkfT8c2VXIeWVCrp8lktb3eiQUQE9s4LJYIj2QalDRWCGnIITYBOAAAD8B8BYGgrhuDLSoniyEuKDEuz4HwM8B/BPANgCNGHgN1gP4I4DTABwihNhaxuYTERE5znwykvTIyQkRkTlE8WMQ5DWxpKllVVkhl3uogx9fBz9WyAEMnQDjGnJ10tABriFXPsMacqFsIJfBSaulY4WcDSFEF4CrBv8Vc7v2HNctBbC0nO0iIiJyg/lkxI8nWkRUncxhBKesli+aMFXIKdaQMwx10Cvksvfhx0pF2wo5j4Q6toFcSkDKoIalHqlCbmR9GP2dAyFRnO2UZUua1pADwEmrDmGFHBEREeVlPhlhyyoReUE6LWDOUHyYA3mOXCFXGw6iRrFelPzBjGoNOT9+cGOXu3mlAq1TMWUV8E5LrZvkQK6lvkb/2iuvnZ8Z29MH3uPGSauskCsVAzkiIiLKy/wHLYc6EJEXqKrhKhkEvbW1C7t6ret4VZuYuULOcPJtXUMupGhZ9WOFnH3Lqjd+5+WqkBvu5DXkRjaE9a+9Ut3oZ3FFNWyEFXKOYCBHREREeZlPRhL8xJmIPEAVoFQqkPvtSxtxzK3P4fCfP131oZx5ympE0bJqqJAbrJqp2qEOHgm87AM5BiLdNhVyXnnt/Mw4ZdW6hpwc4FNxGMgRERFRXuY/aPnHPxF5gSr0qVRl1j/WbgMA9MZTeOn9nRV5DK8wD3WIhBVTVqU2yVBg4HrjUIdKb6XzbCvkPNISahfI8UMyU4Vcvfcr5GLJFHb3+SPYTxjWkBt4j0fYsuoIBnJERESUl/lkhO0xROQFqvWhKlWZJU9xrPYWrUKGOqRULauaz1tWbbbZK7/zdktryEnZJz8kg/H9OdJQIee9fbOzJ4ZDf/o0PvOTJXj+nR1ub05exjXkMhVybFl1AgM5IiIiystcIccFpInIC9LKCrnKPJbcElftkxuNFXLGNeQyz13+oEZvWZXOLn3ZsmqzzV4JdbqkCrlRDWzLlPXEssesYaiDB/fNP9Zuw87eOOLJNP7y2odub05e8s+7sD5lNftmjyZYIVcqBnJERESUl/lkxIt/4BLR8KOqkKtUZZbcEhf3SEBTKXLFy8CUVblldeDk2zjUYbBlVV5Dzo8VcrYtq+4/l2QqrYfCmmaeJFrdx2MhemLZsFJuWfXie1WudOzzQZhlWEMuxAo5JzGQIyIiorzMARzbY4jIC1ThW6Uqs3qHU4VcIS2rqqEO8pRVD4RYxbIf6uD+690ltWQ2RUKG1ySR9N++dlqvVCE30lAh5/5rZ9YVzQZyMZ8FcvoacmHr5GUqHgM5IiIiysvcosoFpInIC4Z0DTkpkPNCQFNJhpZVc4Xc4Mm3vA8yQZzfp6zavaxeqAqXBzq01NcgHMzuay8uI/HiezvxH79Yhh88+jrEEFRLyu/PkR5v5+322XqU6jXk5Cmr3g8VvYqBHBEREeXFCjki8qKU4mS7Ei2riVTacOJc9RVySVOFXMhaDSMHbmHFlFU3hzrs6Inh6sfewG1L3imqUs92qIMHAi85kGuuCyMkVch5ITA0+/Wz7+Hdj3rwyIoPsHpzV8Ufr8dmyqoXW1aNgZz3wyx5H3Kog7NCbm8AEREReZ85gPPiJ85ENPyoApRKVGbJ7aoAEK/yn4FyxYtdIJdQtax6pELuvhfW46EXNwIAZuw5Eod/YkxBt7Mf6uD+620O5OSQ0Isfku3qi+tf7+iJVfzx5Pdoi9dbVqXX0g/hfkIx1CGi+JlAxWOFHBEREeVlblH1QrUAEVFK8bOoEjlQjzmQq/ITUPNQh4hqDTkp6AgFrGvIuZlhbezo07/e0NFb8O1shzp4INTZLQVczfVhPRgBvLmMhNzS3BevbBVYKi0Mj9EUCSGTDaeF99qn/deyKq0hF1KsIceW1ZIxkCMiIqK8zCcjXmyPIaLhR5WTVGKYgLxgPDC8AjlzhVymfS1pqJBTtKy6GILIbYDRROGvlX3Lqvu/87rMLavSvvZCYGgmD5rojSdzfGf55PtvjIQQCGjGwLLM/RNNpPDX17fg39t7yrqfDMNQBx/8LFGtIVfLllVHMJAjIiKivCxryHng5ISISFWtaxeqlMNcIVf1Qx0SxqEO8npRmTDSEMh5bKiDHMJFi6je8VPLasgQOLm/fWbye6S/whVyPVLFWUNk4FitcTCQ+68n38ZFv12FE+54wVCpWKqqWEPOMGXV+8/BqxjIERERUV7mCW5e/DSeiIYfVfd8JSqzhnPLqqVCLhPIyS2rg2vIBaSzy0oEo4UyVsgVHhbYVsh54Hfe7j5jICdPWfXiMhJyiFPxCrmYsUIOyB6TQPmB5coNHQCA7lgSr2/qLOu+AFOFXBEVnG5J5hvq4IPn4FUM5IiIiCgvc4WcF9erIaLhZ6gq5MxDHaq9Qi6aY6hDJpCTQ47gYBInV8i527LqcIWcB37nyRVyLXVhhALenrJqWEMuVtkKqm5FIBc2TKEt7/0qH08fdZc3oCKdFoaA34tTYM2MLauDa8hxqIMjGMgRERFRXuaTTy9UCxARqdoih2KoQ6zKfwaahzrI7X+ZAEHe92HlUAcXAzlDy2oRa8h5eKiDtWVVrgBzf/vM5BCn0kMdDBVytYOBnHQslht6xQ2BXLSs++qNJyG/NfxQXZZQVsixZdUJDOSIiIgoL/OacV78NJ6Ihh9lIFeRoQ7Dp2U1mUrrP/MD2sD6cMoKOak6Mai3rHqlQq60llW7TfbCuqnmQC4sV8h5YPvMEtJ7pK/CLauGNeRqBgO5kHMVhHJAvb3MCrmuqCncT6YgXAyvC6FeQ45DHZzAQI6IiIjyMlcHeKF9h4hIFchVYphAT3T4tKzKJ9+RUBCaZgzkMmFXSm5jU7WsuvhrQq6K6y8qkMtutPRUPBHAyoHcCFOFnBcq+Mzk46jSFXI9eVpWy32/ys+l3JbVbmn9OGDgfeLFQFUm778afcpqdv8WE3qTEQM5IiIiysu8ILIX//gnouFH1RZZkSmr8eFTISe30GUmKdYo1ouSQ4SgqmXVM2vIldayKrfkeWFogmENufqwKXDyXqBjWEOu0hVyipbVUIVaVrd3lRvIWfeF13+eJJJS+B4aXEOOFXKOYCBHREREeZlPRtiySkReoKosGYqW1WqukIsmjQMdACASVExZTcttbIMtq4YKOW+0rBazvpUc5spTJL3wO8+yhlzAu1NWU2lhqJDsrfBQB/n92TBYIVdTQMvqP9Zswxl3v4S/vPZhzvs3BHI9zlbIAd4PtPKuIeeDdfC8KuT2BhAREZH3Waes8o8vInKfKnyrzJRVY6Dg9YqWchgq5AZDKcMacoMn5/LvhdDgSbo3K+RKm7IqBw5uV6DFk2m97TMY0NAYCen7HHB/+8zMgXVfhVsaVVNW5cDSLkC/ZvFqbOmM4o3NnfjCvuMM+1RmaFntKm+oQ1e/tULO60MRlGvIcaiDI1ghR0RERHlZW1a99cc/EQ1Pygq5oZiyWs2BnGHCqrVlNZ6zZTV7P25NWRVCGALT4lpWs1/XeKhl1Vwdp2maXpUIeO93srlFtC9W2ZbV3jxryKlaVoUQ2NI5EK519icQtXlPp9LCEC73xlOWitliKCvkPF5hplpDji2rzmAgR0RERHmZP12u5nYtIvIPVYVcJVpWh9NQh5ihZTVo+C8gBXKpPC2rLlXImcOBYoY6GFtWnZvSWS5zIAcAoYBzQwucljC9BhUf6hBEYcfDAAAgAElEQVS1BnL5WlbNIV3M5jhRVcOWM9jBPGUV8H6gJX8om3mvRxTrSlLxGMgRERFRXpY15Dw+EYyIhgfVz6JKtEr2moc6eCwAcZJ8cp056Va2rBoq5BQtqy5VyJmrjUpvWc2GkG4HXuYJqwAMU1a9toyEuaq+8kMdsq9xQ4Etq+YQye49rQrktpcVyFkr5LzcAi9XCGpa9j0e4ZRVRzCQIyIiorw4ZZWIvEgVvlVkymps+ExZlU+uVVNWsxVyUtVMQFUhV9HNtGVez6qollW5Qi4st6y6XSEX17/OVMjVBL1TwWdmDsB6K10hF8uGXE211pZV1Rp75uDW7j0dS1m3/aPu0teRU01Z9fIabOaBDtrge7yWLauOYCBHREREeZkDOLdPToiIAHUgNzRTVqv3Z6ByqEOeKavZNeTcH+pgDgfsWhFV7Ic6eKdCrkVRIee1D8nM+yueTFd0G3sVFXLGQM762JaWVbtAThHoftRVeoWcOpDz1usnU60fBwxUIGbe7qm08Nwx6BcM5IiIiCivRNpcIVe9J6NE5B/KQK4SFXLR4VMhV8pQh7BqyqpLLavm9rloEdVHKZuWVbd/53X2KdaQkwMnj31IpgqsKzlptUc51CFPy2qisMnJqlbW7T3lBHKKoQ6erpCzrh8HAJqmGd4jXg4VvYyBHBEREeVlrZDjH15E5D5V6ON0dpJOC0vLXXWvIaca6mBdwF0OqYLKllVvVMglUoVX79gOdXD5d95uxVCHcMA/FXIA0Bcb6kAud0uvpZLSJhRTDnUoo0Kuq99fU1bl11IOgQFjWzcDudIwkCMiIqK8zH/MVnO7FhH5h6p93ukgSFXZE0+mIVyqAKs05VCHoHGogxDCEFKF9Cmr2ftxbaiDIliJFhgWGFpWw/JQB7fXkJNaVuutFXJuV/CZqQLrXIMdyn0vGaasDq4hFwpaB5HIzEGbXaCknrLq7BpyXg747VpWAXNQ790qPy9jIEdERER5WVtWvfvHIxENH6rwzem1y8zrx2W4HdJUitzKlznhDgQ0w9TKeCptCIFCqimrblXIKaqNCp0C6dUKOdWUVUNLptdaVhUhVp/NYIcNXRvw5cVfxqmPn4rOWGfRj5VKC/RLr299OLPuYe4KQmuFXBEtq2VMWVWuIefpCjl1yypgbOv28nPwMgZyRERElJf5j1mv/fFPRMOTskLO4cos1Qk04P5C/5UiV5PJVWLmdeTkfa9XyAW817IKFBHISTetDXunAk25hlxA3j5vHYvKNeRsArnF7y7G+53v482db+LJ958s+rHM7aqZYzCUb8pqssA15BSXlxPIdfluDTnjlFWZ/B4pZq1GymIgR0RERHmZT0a89sf/cLZq4y58+3ev4MnVW9zeFKIh52aFXLUOdpArXWqlEC5iCeSkE/VMhZzmvaEOA5eV0LIakltWvVMh16yYsuq1ak3V/uq1aVntinfpX++M7iz6seT3Z0Mk+5qF87Ssmiu6ignkdvbGSzomkqm0Mpj08vpr8vM3B3KskCtfyO0NICIiIu9LmNp10mLgxEWuhiB3XPPYarz5YReWrN2Gwz/Rqi9oTTQcDEWFnH3LanWegBqGOthVyJlaVjOtqkFDhVwlt9JeWRVyti2r3ltDTm4fdLul1ky5hpzNUIdkOvv+6kn0FP1YvYqBDoC5ZdX6+pm30S4Us7t8Z08c45pri9rWHpufJV4O5AwVcqFca8h59zl4GSvkiIiIKC/VH7PmkI7csXl3P4CBdqCNO/tc3hqioZVS/BxyOjvx40l0OVRDHYDcLavhoHXKqqeGOhQYyNlVyLnesqqqkAt4p6XWTDll1aZCLpHOPreeePGBXLdNIGdsWVWtIVdgy6pN8F7KYAe79ncv/yyRqy9rzGvIhTnUoVwM5IiIiCgv1R+zXjsBGK7kxbO3dvW7uCVEQ091rux4y6pNkODlyYjlMFTIyYFc0BjIyftZVSHn2lAHZYVcYa+VoUIunDvQGUq787aseutYVAdy6sDGEMiVWyFXmw3k5PZK1QeI5hZLu0DJLqj7qKv4deTkYLWQx/aCXGvIsWW1fAzkiIiIKC9Vuw4DOW+QQ4EtncV/Yk/kZ6oKOaeDoJ7hNtQhIVfIyS2r0sl3Mq08UTevYuDGYAf1GnKFDnXwXstqNJHSQ6GaYAB1g23EcjjidkutWSJZ+FAHQ8tqCRVy8vuzoUYO5KTAUrE95uC2mDXkAGB7T/GBnG2FnIfDrHjOQI4tq+ViIEdE5EMbdvbiF0+tw+ubdru9KTRMqIY4sGXVfUIIQzvJlt0M5Gh4UbVFOr2GXI/N2ldVO9TBMGVV3bIaS6rXkNM0zRDKudG2qgo3Cp0AmRbqllU3w1e5qmpEXRjaYFtwKCCvkeatY1G5hlwBLau9id6iH8s8ZTUjnKdl1fz+tXs/21WvlVIh162YsDrwGN56/WSJHEMdaqU1JgsNvcmIgRwRkQ9dseh13LbkHXzz/hVV+wk9eYtqghsr5Nxnfl1YIUfDDYc6OC+WkFtWsyfckaD9GnJy+6TbbauqcKPfpjrLzLZCzsXfd7v75HZVu8DJW7+PlVNWCxjq0J3oLvqxegpoWVUNvTAHbXahmBzUyW3bTq4h5+Vw37CGXMi0hhwr5MrGQI6IyIfe3jbwB8v27pjhDzWiSlGuIccKOdeZX5ctnVxDjoYXVUuk0znZsB7qYFMhF0+lDb8D5AED8mAHp8PRQiiHOhT4WsnHjvzc3fx9pxroAPhvDbn+hPp9JAdyvfHiK+TkwLzBMNQhu3/ihbSs2uxDOSybMLJO/3p7d/EVcl1R67TcgW3xbnVZ7jXkONShXAzkiIh8SG5NEC5NMaPhhWvIeZP5pGcrK+RomBmKCjm7QM7LVS3lsB3qYJqymkr5p0IuVuiUVenYqTFVoLn195YcyLXU1+hfG6asem0NOcXfB3YVcnLLaikVcnZTVmvyVMiZ3792x4gc1E2UArmPSgjk5Aq51sZI9rE9/LMk5xpyYeO6klQ8BnJERD4k/+HlxvosNPywQs6bzJ/of9jZz5CehhVV4OP4lFXbltXqfK/ZDnUwtawmDBVyUiAnV8i58GvCqaEO4WDAuB6eS6GXXYVcTZ410txUzJRVuUKuP9mPVLq4SqveEteQK6VCbmKZFXLyGnJjfBLIFVwh5+HBFF7GQI6IyIcMgZzHPhWl6qSqhqvWk1E/Mb8G0UTacPJGVO1UvwOdnuwpV8jV12QDquqtkJMDuezpotzCGU+lDPve0LIacLtlVTHUocCwQN7eQEBDyAOTTHf3xfWv7VpWvVaxrg7k8g91AICeRHGTVuUpq41FtawWtoacfPmEFmMgV+wHYF392W0d0yQFch4eiGAc6pBrDTnvPgcvYyBHROQzQgjDH8EsUqKhoKqG89oJwHCUUJxAcLADDSfKQK6CQx1GSi2DXqtKcop8Yi1PUbRUyMktqwGbllWPTFntL6FCLqhpCAfcX6etq4A15LxWsa5sWS2gQg4oftKqPAW5oYiWVfNxYjvUQXrdm+tr0DAYysdTxX8A1h2TK+SyP0vsqvO8IGmqGpXJFbRervLzMgZyREQ+Yz75cOPTZxpehBDKP64THjsBGI5UJ4gc7EDDibJltYJryI1syAYiVVshl1BXyMlryMlTSwOasSrOMNTBlTXkSm9ZNVbIwVgh59KHUHYtq+GAcY07L1G9N/ptKuTMgVx3vLh15HqkkKupVl0hp2xZNV1m936WL48EA2gbUav/f7HryMlryBlaVj3c7mleQ06I7HqKtVLVbKHvMTJiIEdE5DPmlgmuIUeVZtcWzQo596k+kWaFHA0n6pZVZx9DXoxerpAzn9BXi0KmrPZJJ98hU9WM/L+uVMiV07IqfVtQ0wwtem59CNXRV0CFXIWOxQ86+vDgP9cXPTBIFYAVMtQBKL5CrtemQi4czB1YFlwhJ11eEwqgVWo1LXYduS7boQ7eDbMSUrtvErsw7y/zMO8v87C1dysr5BwQyv8tRETkJeZAzo1Pn2l4sVs3p1InAFQ4ZYXcbgZyNHyoAp9KVsiNapBaVqv0BFRez8ow1EEO5KQQRG5XBYxDHdxY51Y51KHAwEM+doIBzTjJ1KUPobbszlY9j2/OVmcZAqcK7Od0WuDMe17C+p19+PNrH2LR+YcWfNti1pAzV8gVvYZcyUMdjMdE3OYYyRXIfdRd3O/bbqnaUb4fL4dZ8r57L7oM67rWAQAef+9xtIa/oF/n5efgZayQIyLymVTK3LLq0obQsGG3bk4lTgCoOKpP/Vkhl1sylcbN//c2FvzlTQ7AqAKqDwycDIGEELZryHl53ady2A51kAKOXilcMQdyhqEOLuwiVTBQ6KL58rEzMNTB/cEJm6VAboI05VPe75X4gKwrmsD6nX0AgBUbdhW1hp7qd5PdlFXLUIe4U4Fc7pZVc4uqbcuqdNuaYABtciDXVXqFnB+nrEbFbv3rrliXacqqd6v8vIwVckREPmNumeCUVao0u5MQVsi5T3WSsbWLa8jl8sTqrbj96XcBAONG1OK82R9zeYuoHHZV4um0MARDpYol03roFw5qhhP+alxDLp0WhgDCbg05Q4WcpWXV5aEOqvXLSlhDbqBlVa5CG/rXO5FKY1vXwIcsmgaMkyrkgoZpts4d8xny8S3EQHvmHtKU0Zy3VVakpZFKC8N2Aw5UyMlTVmvVFXKqv2PMx4n9lNXssVMTCqCtKfsaFNuy2h3Nho+j5aEOHv5ZIr+WCZFtJ46lYmxZdQAr5IiIfIZDHWio2Z2EeG0R6eFIddLDltXc3tmWXTB8Y0efi1tCTrBrqXfqd6O5+iZfG5zfmcM4TWo/lQO5XBVybresqoc6FPZaGaasBjRTFdrQP5etnVG9E6K1MWIIQLQKr3FnDli2dhX+u8WunVvVtmqpkCsikEulhSFsrZemAsv7xi4glBU01MHSslp4IBdLpvTHDAY0tNSHDdd5lfxzLi6yr81AICdVyHn4OXgZAzkiIp+xDHVghRxVmG2FHKesuk510rOlM6pPQCOr7T1x/Wt+ou9/dr8DnarMkttVGyIhQyjl5aqWUtlNWAUG2vUy5CmrOVtWXfhZpArfCp0AKR9Pmmas/nMjgLVrV82o5Bp35p+PHxUTyNnsK1XbqqVCroiWVXNgLh97+SvkUqb/L2yoQ1uJa8jJE1ZH1IYs1WVe/b0tD3WIpU2BnBSAFhp6kxEDOSIinzG3CbJCjirNvmWVx57bVFWK/YkU10bLYUdPtqKBgZz/2QVyTn1eYD7hNwRyVVghJ4cU8sk2ANRIAYKhQs7csup2hZxqqEMpLasBYwWaXTVmJX0oBXKqdtFKrnFnDpyLmbRqV0EvB9wZ5bSsGgNz4/Gar5q10DXkYuZAbkRpU1blQK6pNmw4voTwbteBvO9iaXPLqlwhV30/D4dCSWvIaZr2a4ceXwghznPovoiIhgXLlFUGclRh9i2r/OPLbXavwZbOKFqkxecpa6ccyHERat+rfIVc9hhpiIRQI7fBVeEJaDRHhZz8/4VWyLnTsqqqkCuhZVUzt6y6UCG3KxvITVQEcpVc485cQbatiPDJLqw2V8ilRRopYbysnAo5WSjPUAdLy6rNNpvbuOXBLsW0rMrrxzUNrnVXEwwgkUrpj1MT8l69lPz8o6ke6fI4asNsWS1XqUMdvgVAAFCtGmn+qavluFwAYCBHRFQE8yegzESo0uw+dWe7tPvsTiC2dkax9/gRQ7w1/rCDLatVxTaQc+jnU08sexJtqZCrwuPHUCFnblk1rCEnD3UwrSEn3WyoPzQUQtgEcoVWyGW/HpiyKldZuVAh15mnQq6Ca9yZj+9tRVXIZW8b0LL71RzImavjgOIq5HIFcjV5XjvzBzJ2H9AYWlaDQYysr0EooCGZFuiOJhFNpFBrqiZV6eqXW1YH1o+LhIP6eymWSFmegxdkX0uB/lR2DdZoKmpsu2XLaklKfcVvsLk8jIGArRnAJgDPAdg8eN0eAI4AMBHAbgC/BsB+CiKiIpnX7WIoQpVmV4WV4LHnOrtAQD6JIyNjyyo/0fc7u9+BTq3H1CNVyA2HoQ5ymGUOGWpsK+SMwV3AxZZVuw8pSllDLmgemuDC671JqpCbkK9CzuHtM+/LbUWslyZvy4i6MHb3DZz2m4c6mAc6AEBvotdymR25ZbW+xlwhl3vfFFwhZ2pZDQQ0tDZFsGUwoNzeHcOkUfV5t1VVIeeHlk89zNQSSIns/o6n4r7Yfq8rKZATQlxjvkzTtBoA/8BAKHcOgAeF4jehpmlnAvhfADMBfK6UxyciGs7Mf9x6dRFYqh526+a40b5DRnYnYMWs9TOc9MWThgoNnkD4n11rqlNBkHmNqmG1hlzOCjl5DTlTy6rm3lAHuTW1NhzQ/z9a4HvdOmVVGgzgwiCjzUWsIed0IGeueCrm94pcrddiCOTyV8h1x7stl9mRn3MkbDxewznW1xNCWN6/iZRAOi0MLdeA8X2eeQ/IgdxH3dECAznjGnKAXwK5ge3SgsYP+qJJU4UcP+AqiZNNyt8HcBiAC4UQD6jCOAAQQjwI4CIMVMtd4eDjExENC+aye6fWySGyY/dHPoc6uE9+beRWlw93M5BT2Sm1qwJssakGdh8YVGzKalBuWa2+n4HGKavGCrmI9Nz7YvZryAUNa8g5vYW5yaFAYySsb0sqLQoKrOQAMRAwV8gNffvth3mmrBomiTpcjWgOrD7qKm0NuWZpzTXzUIdyK+Tk18RcqRkOGMNzOZ5IpARUPyJUIbu5Qg6AcdJqgfulS7WGXMj7a7Blnr85kIun4oYQ1KuBotc5Gch9DUAcwG8L+N6HB7/3dAcfn4hoWDB/6s+WVao0+5ZV/vHlNvlkZE/pE/qtXWxZVZHbVQHvngANlXc/6sbv/7XR11N50xWesmqoaomEEK7yCrmoYcqqfYWc/NzNQYibU1ZjpqEUtdI2F9K2ah3qIAVeQxzIdfTG9Qq/pkgIzXVhy/fIYajjLaumgKU7llROSVWRt0Xe7v6Es2vIJQzVa9bhInI4LAeWdj/7VR/SGNeQy1TI1eqXbe8pNJCT1pCry1TIZUNvr65Jma2Q6zNcbp6yGk+mbX8ekz0nA7l2AFEhRN536eD39A/ehoiIimBumeCUVao0u5MQVsi5T/4DXg7ktlSgZbU7msBHRawh5EU7TBVyhU5erEb98RTm/eqfuPKPb+Daxavd3pySDXWFXMRQIVd9ga450JLZTYC0tKy6ONRBrtKJhAOGdfDMYZCKsULO+NyGumVVrnRWtasCpgq5Cg91AICtXYX9DkgkjS2rGfLUYkBdIVfMlNVkjgo5wL5t1S78iqWM25dKC8PPmMz9tZZQISevITfCR2vIZZ6/FjB+0BdLxaBpWtW38Veak4FcD4ARmqZNz/eNmqbtg4HBD4W/24iICID1Dy4WKVGl2Z2EcA0598nVAZNHS4Hc7qij60tu2tWHz/5kCQ796dNY/u4Ox+53qFkr5IbvMbx+Z6++rtMrH+x2eWtKl7L5+eRUpYa8VlqDqULOjamblVboUAdZMGfL6lAHcvIaeEHDcyikRd061MG913vz7mxFkqpdFahsYKiqIttWaCBnVyFXwFCHeDqOeCpuuVxFDoDk10q/LKAOi+x+9puDOnO7qjZY/WloWS3wgyrjGnKDgZzc8unRD4j0fRC0BnKAKVT06HPwMicDuWcAaADu0TSt2e6bNE0bAeBuAGLwNkREVARLyyor5KjC7E5COGXVffJJz+jGGtTXDJx89idS6OovrLWoEE+/9RH64ikk0wJPvrnVsfsdajvZsqqTq4UKnUDpRXaBj1OVWeYpq8Y15Krv5DPnUAdF4AFYgxDDlFUXhzpEQgHUhstoWTWtITfUH0JtNlTI1Sq/Rw6cnA4MVaFVoYGcHH611EsVcgUMdQAKb1vN1bIKwBSg5w/kLJNXk8bjKUOukNvenbtCbld0F85+4my80HcttFAXAGBErbVl1au/j+yGOmQDOe8/By9zMpC7FkAUwGcAvK1p2nWaph2ladq0wX9HaZp2HYC3AHx28HuvdfDxiYiGBXN7DtdrIDud/Qm8+WFn2ZVS9i2r1Xcy6jfyCVhNMIBxzdmTtg87nVtHTm4z8nMIYW5ZHc4VclFDIOff/WAXyDlVmdUjtZk1RkKGKjGn1+zyAkPLZ6j8Crmh/htFDgRqTS2rhRzn8uYGAhpCcoXcED+XzbukgQ4t6imeoRyTRMulaj/cVmB7pl2FXF8BFXJA4W2rpbas2gVH5t9vcgurHMgZK+Ry75PH33scqz5ahR78G+HmlQCyU1blkNurv48yf2eo1pATQvii7dbLQvm/pTBCiLc0TfsigEcAtAG4xuZbNQA7AJwihHjbqccnsiOE0MuLiaqBOQRhHkcqvbEkZt/4DHb3JXDNcdPxzcOnlHxfdm0w1diu5TeGdp1QAHs01+G97QMT6rZ2RrH3+BGOPI5cTeX0JL+hZF58O55MD9u/E+RArpC1tbxKrsAKBzX955JTFXJyGN0QCRmqwfwcTtuRjwtzhZw5oMsIm9aQc3WogylQTISzjx8toHrH0rIacLNCLhuA2FXIGQNDh1tWFQHm1gLXJ5X/PjAGcpWrkFO1rIYC6gDddg25XC2r0v23jci+HvkCuR392WUetODA72dVy6pXf57oFXIB64d8iXTCNGnVv79L3OJkhRyEEM8AmAbgxwDWYqAtVRv8JwYvWwBg2uD3ElXMrt44TvifF/CF/34OH3T05b8BkU+YT4bZskoqr2zcra8P9dSa8loM7YK3oV7gmqzkP+DDpgo5Jwc7yCfpfq6MNLesAsP3E/3+uPHk1K/V1ilTlah+uUMva4801KExEqr6ahDDUIccU1ZlQVNlUkCukBvqoQ45Wlb74wUEctL2Bk0VckM9yEge6jDRZg05Y2BY+Qq5QtZLS6WFHmxqWrYaDLAOdbAN5AqskDN+KGX9YMWuorXUNeQyxjTW6F/v7InlDJ57E73Z/wkMPN/slFXvh1l2LasAEE1FURsqrgqV/j97bx4tyXWXCX43ltzeWrtUJck2khjc3hcOW7cBd8Npd2NsGAwMcGD6tIdhgD6MG5uZHhoYusHHw2rcYJu1AS9gkBsbbGzZMt73RZIlWVtJKqn25e0vX2bGduePyIj43Rv3RkRmxlsy637n1Kl8+fJlRkbcWO4X3yKiVkIOADjnq5zzX+GcPxvAPICbh//mOefP5pz/Kud8te7PNTCQ8b77LuKes+t46NIW7vjyuf1eHAOD2pBrWZ3SSZTB7oLeKe9NeIGkI2BMy+r+Q8jPsS2cFAi5+iyr/RlRyMmWVWA2SZUqkFVx07oeBIWcQwm53Sh1sKWQ/+lcZ0UosqzKirkErmxZFRRyNS5cBQgZeK4tkQUVWlbJuLEYEyyhdSvQynB+fVTLar3Lp2xZrXCjR1atzTWzbVDZslpRIUevQ1yFZdUhY5PeXNSVD+QIuVBNyDUdO83Giziw0tWr5Hb8TBjCWPx9U4WckL92MI8nyTpREXJe6BmF3ISonZCj4Jz3Oefnh//qu01rYFABm73sAE9bbQwMph0yCbLXdhCD6QCdVAwmtKPR3JzGPjbOGeQhT3xuWMpUFHUq5KiyZJqJWLllFbh+JxAyOTGtxQ70HEjJsvosq0Qh1xIz5FQKommHnMFGoSt1cGTLqrV/pQ6UaGk5FloNQshVIDxkhRwlefby2NfzQqx24xsIjsWEEgGK3cy4U7eslmfICecli6HTyFKy9tqyqiPQdcd9+XmdQg6QcuQK1stOQJxaEiHXmIKG0ixDLk/IDcKBaVmdELtKyDHGGoyxG4f/GuV/YWBQH+hJc6/l8gYGuwmZgDPj20AFqmKadKJN77rTCVpoLKv7Di8Q87NuXN4dhZyYITed290Po9TGTXG9TiByhNyUEpOB5oZBXedGelM3V+pwQBUtk6BIIWdZTFAcJSiyrE5aKjQqRIWcNZJCjnMOurgW210FWhGoOu7G5VauOCPBbmbcqRRyV7b6pc4MerPOdUoUcuFkpQ70s2RiOPn87LXlGXKFllWJ8KvatCpbVhuOle5bU2VZVWTIDYLBVKj8DjJqJ+QYY8uMsf/MGPsqgC6Ac8N/XcbYVxljv8gYW6r7cw0MZNCJolEQGcwS5DughpAzUCESCLlJLavZe7WJ2mCvG+cM8vClUocbdy1DLvucabWsJmoTGdfrBELO05rW7B96jdeo2bIahFE6PiwGtF1bKDCYRYVcUakDoM6Ry5c6ZI/3u9SB3kQqU4uHgl0VYIyJCqs9/C4XCCF3ckmdHwdgVzPuVKSVH3Ks7qiPpdlrRNVaxy1QyPH6FHIqBaers6yOU+ogjf1Fmo3n6d1YsmWV/h0lsw5iqUMU8eycX0Uhd0BJxYOMWgk5xtiLAdyHuLjh2QBsZKUO9vC5/wLgPsbYi+r8bAMDGfSkaULvDWYJoTQBmMH5gEENoMe9SRsUaW5O280uHqc53H9WIE9GblzMJm6XNvq1qVPEUofpPKeq7KrA9TuBkBVxVQLvDyLCXVTICQ2rDQeMsZxtf9ZyXAVCy61GyMnqLaqQ219CzkKLnLPKzoWyXRUQM8j2SyF3SlPoAIhkaO0tqxqC6PJm8c0eWVXWERRy4jbQKuTGsqwqFHICYVluWZVJsYGQIScqRh272g0AwbJq+VhsZQSlmL+mXt//3wcfwr/5vU/i06evKX+/m8jGFNdbVl2jkJsEtRFyjLHjAD4A4BSATQC/CeBlAJ4z/PcyAL81/N1NAD7AGDtW1+cbGMigB91Zu1gyuL5hWlYNqiCs1bKavVdLIOTM2NtvyEqExbaDzlDFuOOF2OzVk6E6C5ZVVaEDcP1OIGjLKjC9llUhQ86pt0xg2xPz44BENbV/Qf+7DbGl1M79XqlCkp6jpQ57reIXFX6iQq5MBUo3pTX8DquSHfYAACAASURBVJR0uXv7nfiJD/wE7rt6Xy3L+olzn8Ar3vMK/M6Xfyf3uwtCoUOBQm4XM+50iq0yQk4myTpEWU8zGYEChdxYllVFqYNG0ar7bgPpwFFkWdUVRsigllXGgjQ/DpAtq/llenKli7d87DF87eIm3njXI9rP2C2k38sagLH88g3CAVomQ24i1KmQ+wUARwDcD+CZnPP/i3N+J+f8geG/OznnvwDgnwH42vC1r6vx8w0MBFDSwlhWDWYJMiG31/ksBtMBOgkaBNFENyboDQ5jWT1Y8MgkoOFYYIzhBmpb3awnR24WSh2uaTJ+prXMYFLIaqFpXQ+7Weqw3acNq9kkmk7MD6LNbBIUlToA1RRyQqnDnresipmnVNVdNsYjhUIuIV+t5gU83H8vvnLlK3jrV99ay7L+/t2/j8c3Hsd/v/+/48L2BeF359cqEnKUHK47Q46834nFLC+trNghkPbJlmMj4WgHQSTss7TUwbUyK2ddllVdEZXWslrQPi1buB1hnOvXvZAhx3wstrPv2Sixe14h560rBTl1u4UkJ1OVHwckCjlCek/pjZ39RJ2E3L8FwAH8O875Jd2LOOcXAfw7xDbWl9f4+QYGAuiEwSiIDGYJMsFsCGcDFWTSZBIVECXeZtmyemmjj9+88yF89KEr+70olUFD5ZOJI80burheT44cvcie1gy5la7Osjpb47gqchPPKVU20Gs8gZCrYZxuDzSEnDPLhFyxQk6VK+cWWVb3vGVVzMBrCYRc8bYSLKuJQm6oQLMaV9PfXesVWwc/+vAV/Ow7v4LPPb5S+Lpz2+fSx2v9NfF3NEOugJDbzeZzeky4+VAnfXypJJ/UE85LFiyLoeOqix2oZfVQ81D6uKpCjl6HKEsdtJZVTalDgUIuR8jZ5Qo5zrmYIWfJCrliuyc9BlEL/V4hLXRQ2FWBJEOOfIcpPY/sJ+ok5J4GYItz/uWyF3LOvwhgC8AtNX6+gYEAOmEwllWDWYJMtBhCzkAFWR0yifqFkj7tGbasvv4fH8QffPQx/OTbvlTYmHaQ4CnUATfsQrFD35thy+p1OoGYBYVcFGWtmIyJGVJ1EEHUXjdPcrDcXSRB9htFiiAgn6MF5FtWBcvqfmbIubaQb1Wm3qHLmpCKCenC3I30d16oLzUII47X/s29eN9XL+IX7viqfjnDAba8rfTnXiASHhcqZsjtZgsstW/ecjgj5K5sjWBZHY6hdiMjoQTFNbGsLreW08eCqqzws9SEfAKdZZUSt2S45s4HRaUO1C6suxb3I1+05TIfC01a6lBsWaXHILmhdi+Q3JBl9o7y96bUYXLUSch5ABqMsTw1LYExZgFwAez9qDK4bkBPSjN2rWRwnUOeDJuWVQMV5OvySWwE9AZHS7CszhaR8eiV+I68H3I8uVJtMrDfkDPkAAhNq5c2arKszkKpg4ZkvV4nEDIhN2n5y35AVjRZjFrI6ibkrg+FnJzBJkNlWZWVSfY+ljrILbE036pfUlxClzVnWa1IyO14AVaGjc5Pre5ov7+ssuuHGckVRlxQoVXNkKs7RoKO7ZsPV1fIUZKsMVx/c4TQ7pLtoFPIbfkZWVm4jIpzIEVD00JLycZ5QhbmFXLZsspj37bK7cIyschYgMW2JkNOcXNoh6jidrxwzwnu1LKqUch5oVeq8jMoRp2E3MMAmgBeUeG1rwDQGv6NgcGuIDQKOYMZhWwXM8PbQAVZHTJJgyK90JxlhVwVO8tBg2BZdRJCLpu8XahLIUcmCtNqWb3WzSbRhzqZQmFatnXdkBVxZXa+gwiZQKET5DpuVm3pLKs0Q27GrPtlLatNlQpJtqzuY6mDbLltjaCQo+dNS7KsMmc9+4xQr6CWCVod0S0TclQhd2Wrnx5nj8w1hO8gw91NhRxZX1QhV5Yhp7pRRK8dKNGtVch51W6KBVKBhAxdxh4lv6iFVN5+KhW66vN0xGtO6Wf5WGhVz5Dblkowdvb4xkmZZbUf9Cs1xRroUSch97eIc+H+mDH2nboXMcb+DYA/Rpw3964aP9/AQAC9S2QsfQazBJMhZ1AFoXRhPslkmxJv9KK67gDp/Qb9PtNi3/MEu048ORAVcpMTcpxzSSE3ndudKuSoBex6nUD0fPkYMR1jniKQCTlBITf5+1+PCjmh1KGyQk5WDmWP9/oapbjUoXrLavIdMoXcZvo7P/Khg2xhlltFE+QUckF2rBYKHQrsqoC47uu+WULH9i1HKCFXkiGnIOQooU3PJ3RdLjczQq6qQq7Msiray9U33WjJgkyKFVlWqVVbt+53AsnqyQLB/l6mLpNtqjua8aTDIBzgCxe/UDmTT0a6LTWlDl7oSS2r03ce2W845S+pjP8G4McAPBfAXYyxzwH4KIDziNVwtwD4juHvGYB7h39jYLAroG03ptTBYJYgkyBGAWqggixem8SySq2ptGV1WpVSOtAL+2lRC6ka5m5czgi5CzVYVuVJwrRud1rqcGq5jfvPxxPs63UCIdv3prEdT1bIsTGVWT0vxEcfvoIX3LIsKEy7VUodppSg1oEqh1QKOSUhd5BKHQK95baMdN4NhZyWkNvRK+TO00KHpRJCroJtclzQ73JyuQ2Lxa6Mla6HQRAqLc2AXDYUr79OQ62Qo5ZVSsh1/S445yhLw1Kp8Sh0eY/0uxUq5AoIOVGdqCHkfJGQY4yjkxXWCvuYitzfloocuiVuh5XtAY7MZx/wK5/5Fbz/8ffj6YtPx3tf+V5YbDQ9VrLOqELOYhYiHi9rP+xj0alOehvkURshxzkfMMa+C8A7AfxLAN8C4JullyWj9i4AP8Y51xvwDQwmBD3oGsLCYJaQU8gZwtlAAfm4N4n6hV5otma4ZdWbQoWcMkNuMZvAXdroV5rUFEFeF9NoVY4ijhVS6kBbC69fhdwsWlaz341y7ffav70X77/vIk4tt/GRn//29DhHJ8NUIadT3cwCSksdSoLzgX0udZAIRbpkZcd1odQhIeRsBiAAczKFUVGGnBdKBIqmGfNav0AhV7HQAZBbRGtuWZUKnY4tNFO76tWtAW4izasUQoacE69HSsjpSh1aTgstu4V+2EfEI/SCHjqu+jOyzyq2rLo6yyohbqmFVD4f0Ky5olIHXdmRqpyi3cjWzyilDqqfKf6fv7sP7/z8U3jVi27Cb77qefBDHx868yEAwJnNM7jYvYhT86e0f6+CyrJ6pHUEV3tx67AXesJ6mbXj4V6gTssqOOdXOeffBeClAN4M4PMAHh/++/zwuZdyzr+bc36lzs82MJBBL9IMYWEwS5DtEIZvNlBBPu5NRMiRC82Wq77bPAvQ2VkOMlSE3GLbSSc/O16Izd5kHVoycTONCrmNnp8u90LTwWLBBOx6QT5DbjpIaAp6redIGXKjXPt99vEVADERcs/ZTAmltazaxaqWacY4pQ6u3LJq1WsdHgV9QSFnCTeRZJu2DHWpgwXmbIGx7Hd+5KcKIRleIFlWNc2YRRly1LJ6sqDQAZBaVmsuWvIkcvbEYqa+LrKtqs5Lc6Q4QVfq4Fou5hvz6c/bfrnNchTLKj13DcZRyEnvL677ihlyAJoN9T6mypCTx4+OkIsijnd98SwA4G+/fA49L8Tp9dOCJXgQjN4eryp1ONY5lr1nOBBIz4Eh5EZGnZbVFJzzjwH42G68t4FBVdCTgcnYMpglhHLLqhnfBgrIx71J1C++JkNu1o6t1GYzDeQE51yajMQXxYwx3LDUwuNX44nA+fUelkiJwaiQC0HqnvTtBahd9ehCUwqhPvjbejeQV8hN33oIJUXTOC2rUcSxvpMpnu45u45v/rojAMRA9evBsso5F4gKFfmmes4uKHXYc8uqT28g2aCLVmZPF1p7rcSyyoSG1QRxdlZL8XxFy2pBy+oFqpArIeQoGVr3TTJPUofFhFy8LoqKHVQZcjTuguaiBVH22LEczLvz6brZ9rZxvHO8cBnpfE9WasbPqclzHSEn36CRSUnhvS1qWVUfB3IZcgAarkgap5+tuE6Tx8+OxrLa9QLhmHdlq48HVx8UXkPHWFUk25KRDLljbZGQa7bJGJyxGxR7gVoVcgYGBwlCy6pRyBnMEOS7cEYBaqBCnpCbxLKqaVmdQmKmCEKG3BSQNIItyLYEW+ozjsyljx+7Ol6Yc4KcQm4KlZFXtzLC5chcQ1QlTKFVsw7shkLuy0+u4ZV/8Gn8+vu/NvF7VQE9/8kKuarXfpt9X1Ca3/NUppDbFhRy2ZhxZ1Qh50n2P5loA3SlDgUtq/tY6iAr5EazrMb/O7YF5q7nXutFattqLkNOQ6DIGXI6y+pNpaUOu9OyGoRReh3BWLx/3bBYrTBIZSOlhDYllaiCy7VczLujKuT0Crb4ObVl1dNYVuXtV0RQ0/1jJIWcqyb51KUOcoacmuCVrdGXNvr42op4HC7KPtQhy5DLiEVZITerNyj2CruikAMAxthhAC8EkGyxqwC+wjlf3a3PNDCgoBOGWVNxGFzfkCfDhnA2UEE+7smkyiigF5r0LvcsWVY558KF5DSQNPLkmeK24/P4yENxOsijVyYj5GR15TQSsde2iUJuvlk6CZp1BGGU23/ryJD7g4+exj1n13HP2XW88gWn8KyTSxO/ZxFC8h1sm0lEULX3WO2KxIresppN2psz2rIqNJRqAvtVuXJOzrKaPd7vUgdKWPVLtpVKIefaDMxRK+RUkDO0tAq5vtqyyjkf0bJK1Ek1znfo+aXpxDd8TixmZQGXtwoIuZJSB9oUShVyruVirpHdTKrSDLqnltVcqUN5fp9c6gAAjpN9Z3pzyFPcCNyumCEnv+7y1gAProgKOWqLrgpVhtzxdqZaHASDmc7U3AvUrpBjjH0zY+wuxATcnQDePvx3J4CrjLEPM8a+qe7PNTCQQScMUzh3MDDQQiZajGXVQAWZqJ3MsqpRyM3QhVdOUTgNCjk66ZEmCrcdz1QGj01MyE2/Qm6FEHJH5huCauZ6tKyqiIk6FHJXt7L1fHZ18obfMggEimxZrUgEre2IxMqlzX6q/hFbVqlCrr5mS875WMqV3UBZwypQUSFnTadCTrZAAzHZqLOsqlClZZVzrrWsbvaCVFXXdm0cKokbcCvYJseBKjtNyJArVMjlSTKBkNOUOjiWgwV3If25ikIuGNeySsa6mCkqjhHRsiqS1OMq5GyLEHJu8c2hfKmDegzLhNyljS4eXntYeG6sDDkFIXe0czR7z3Aw05mae4FaCTnG2KsBfBLAdyJuVOUAVof/+PC5fwngU4yxf1/nZxsYyKAHRmPpM5glyOqUGeJEDGpEvZZVjUJuhshgWS00DQo5VXB2gttPZJOaR69sTfQ5+Qw5Dj5l59VrpGE1p5Cbgm1dN+RtCpSrh6qATmZpbt9ugWaqjtuyutb1c8/dc3YNALClK3WoSSHnhR5+6H0/hJf89UvwiXOfGPt96kJZoQMANFUtq5K11R4jy68uCN/BtdAi26rnh4XHLnp5JSjkFJZVHYmaz5DL72sbgw1BGRYvW0x4ULvqyeVWaUO2s0stq6JVMx4LYqnDaBlynYqlDnMuUchVIOQ8KbpBRkNTeiG2rBZkyBVYYt0KdmFVhlyA7DvT9xwEUW58yuNnR2tZFZ9/dO10bowmP3cHAf7u7nM4XeFmXXx8iwArI2CPtgghFw2EG4KGkBsdtRFyjLHnAXgrABvA5wD8WwCLnPNjnPNjABYAfM/wdzaAtw7/xsBgV2AsqwazCmNZNagC+bhXFmZdBEq8zapCTp5ETYNqqmiiQBVyT1zrTqTiUakFp+28mit1ECyrB39b1w0VQd/XZF2NAjqZXd1WK4jqBL35ao/Zsrq6k1/Ou4e21W6FUodJ9q0vXvoiHlx9EDvBDv7m4b8Z+33qgqwuU0GtkCtoWd3DaxS5lCK2rFopYch5ccaVstTBtmA5m7nXVlXIqQgUWR0HZAo5SsidOtTRLmsCqgqr1bKqGAs3LI3espoQYp2KpQ4LDaKQq2BZpSSbPA7l53zSgEvHwWJBhpzOsso5x/0b/4TOM34PrVNvhxdqiDKFQo4q1SyLCcRejtCVW1Y1x2lZIffE1qO51ySW1df/44N4zbvuxSv/4NPY6OVvSFD4IQesftoyPO/OC6TpIJAUclOooN9v1KmQe+3w/d4N4F9wzj/AOU8pYc55j3P+jwD+OYC/Q0zK/XyNn29gIECwrBrCwmCGIMvizfg2UKHWDLlQnOAkiPjsWKblSXUdeVq7DVXDaoL5poOTw8mTH3I8uZK/S18VKjWVzp5zUEFLHY7ONdAULKsHf1vXDSUhVwMxSSevK93dJ+RCgZCzxioTWFcQckmxA1WnzBMVjWsX28yqgiqA1gd5FdZeg5LTKuJN97yskNuvUgc/5EguiWgpRVuwrRYQckrLqlohV5WQk4kSIJ8fB2SlDufXsmP1qeV8i6sM2rJa500yFTl7YoGUOmz2tWpDlXqbKuSKSh1GVciJeXV5NaEu34wqo6u2rCZj/1L3En7mIz+D95z/bditi3AX78fV8H7l8qky5OS2UzFHTiJ0JYWcNkOuLz5/ZXA695pEIfep08MW20GAhy7myWYKP4wEu+piYxENu5Etb+hJmZrX3w2uSVEnIfftiG2pP8c51x4Nhr/7ueGP31nj5xsYCBAsq1M2cTAwKII8ns34NlBBViVMQjBRVaZji3dz/RkJ6cwTcgf/orLIsgoAtxKV3OkJbKuqdTFthJxQ6iAr5KaAfK0bKoK+jjFPJ7N0ne8WREJOUmZVHKOrCsvqfec34IeRoE6Za2gsqxOQIJTUUSlp9hpCqYOrtqyqbIE5y+o+KeTkQof0MSXgC8Z5pFDIcQSwnDwxpLOsyucSuSUTAK7uXM09l6iXLpBstlMlhQ6A3LJap2U1T84uth20hplnO16oJBvl5UjsjDSDsbJCrlLLalmpg1pBKJY6EIWcrJaXypPueOQOfN97vw+fPP9J4XW9cEW5fCrLqkzm6kqGvCCqZIEG8kq6rehM7jWDcIAo4ri4no2x9VKFnETINRfRtLNyj0EolzpM17XBQUCdhNxxAOuc8wtlL+ScnwOwhqyB1cCgdhjLqsGsQr7YMwo5AxVkVcIkk21Kurk2Exr1pjHgXwVqZQGmQzXlBcWE3O3HSY7c5fGLHVTkzbTZlQXLaq5l9eCTr3VDRdDXoQqlZMfKHlhWCxVyFQ9NKoXcjhfi3rPrqdqq7doCydTQ2OBGBZ2Yb3mTZT3WAaHUQauQyxN1RaUOe3mo6GuWv0WC86sq5JIcvA0/r2YDAC/SKOQqtKyu9PLkTULIjdKwCojH/knIYRkqy2rctFpuW1VnyGXjhpJKcqmDoJArsaxyzqXrk+KWVV8iuxIstolCTjrfJa9j9hZ+66v/Eb/62V9VEoUeVxPqKqI9r5BTE3KqsaPLkBPJ0Qi+cy7/uUEf17YHwvbZ2Cm3rMoKuaYjEnJ1ZWper6iTkNsCsMAYKzW7D1+zMPwbA4NdAT2pGsLCYJZgFHIGVSCrEiazrBKFnGXt2h35/YQ8kZk6hZxi8nz7CaKQuzo+IaeawE6dQo5YVo/MNwT1zDSQr3VjLxRyq3ttWWWSVbJqhhxZTjoxTmxdgJgfB0DKTBp/vVGV1UFQyFHbclOnkFNaVqUMuX2yrIoKOUrIZd+l6FxIlzX5Sqv9K8rXVm5ZHTFDjt48OL5QwbKqKS2YFLrstCrFDuoMuWwf6hWUOsw3svNWmUIujDKLssVEZWb6nrT0Yrh+4qxBYkcn+7d8LZBYMJs3/D3uX/1y+vzTFp+G5yx/W/Y9RiDk5LHTEBTb2XKpxo42Q45YVq3GVTArT7T1w76QUQgA673i47QfRmCWRMjZBYTclN2sOwiok5D7CuJcuJ+t8Nr/AMAZ/o2Bwa6AngwMYWEwS5BDe83wNlAh37I6/kUSPZ7GllVyx3lGLKvyRGY6CLlsG6uaD28nltX6FXLTc+DpDoL0OzQcCwtNB01XrUi4XqDKBZyEtAfiYw4lavemZVW8WUB3g6rXfmtEIfcttx5JH3+aEHI0YwqQSx3qUch1/S4iferPnqCaQk4VnC9bVrPHe2tZVVtuRYWcfpyrSh2uaQi5SVpWr/byltUkQ44eW3U5fhS7pVgfaAi5Gwghd2lDrZCjqlGlQq7AsjrvEkKuRCFHjzcqdVz8PC1M4OnfJX/qWEzIGJSbTpPtaTUvp8/96DN/FHe8/A78T8svSJ/zuTqnVZkhF+gz5ESFXH7s6DLk6PNW67zyNf2gjwvr4mevlyrkRMvqUnOp2LJ6HZ5PJ0WdhNwfA2AAXs8Y+38ZYwvyCxhjxxljvwHg1xDnzf1RjZ9vYCBAVMjt44IYGNSMUCIOZiVU36Be5FpWJ7DlyRktNC9omoiZIkyjZVVUyOWVAbRp9bGr22PfnFKXOhz89ZOAWiePzTfBGJMy5A4++Vo3VKTEpFl6sjJotevt+g1R0bLKRKtkRSJojUxIX/oNx9PHdz+VBfnT/CtAsglOcKygtkcOrpy8FyHiEZ7YeEIbrj8qdAozCnWGnPjcOErFOkDHMCWRxFKHAkJOUeqwMqlCrqJlNSFp6E1XmehUQchI2zXLarb+TixmZMzlLQ0hV1Lq0CsodaCEXJlqVGWNlaEii+TCCse2ssxALhJ9yXqwnMzY95PP/Um0nBbm3cXsvTUKOVWGnEzm0htEdL0rFXK6UgdC3tmEkDvcOix87vl1cXnKMuQ8RamDTMgJ51OjkBsZtRFynPM7ALxz+J6/BOAyY+xTjLF3Mcbeyxi7G8CTiJtVbQDv5Jy/u67PNzCQYRRyBrMKmQDZy7vPBtMD+ZpIRapUBSVfHItpW8umGdNoWS3LkFvuNHBsIb5wHgQRzq2N17SqInOniYi9SsoFjszH7XCTWFbfeNcj+Ndv/AQ+8uDl8hcfUKjGtxdGE10vyeMk4up8tjoRSITcOFbJNWJZfcntx5BwevS9aaEDIFnMJiDk5Il5lRB7itd9/HX43vd8L1778deOvQzC8lQodaDkQYK8Qu4AWFYFhRwh5Aq2l6rU4WpPvZ9XJeRUpQ46y2rEIyGf07UqKOQES2Z969oTbKcay6pGIecp4hQoqV1VIbflF6dblTWsxs/nLasqO25DQ7IPgghgQUpK2czGcnMZAIRlDaA+v6pIxRwhN1KGnMayOsiINaudEXIvPP5C4XNlhVxphlzAwezsu5WVOniSwtCgHHUq5ADgxwH8MoAugBaAbwXwKgAvB/A8AM3h7/4zgJ+o+bMNDATQi0pDyBnMEuQLLjO+DVSQVQn9CRRyYsuqNVaT4UFHvmX14BONVdQBtx2b3LY67Qo5oWF1Pp5IjGtZvbzZx+995FE8dGkLv3vXI/Ut5B5DZ0+dREmrWo+7nSMnWwxHLXWIIi4oRE4dauPrT+RMPkLGFCCVOtTUsgqMliPXD/r40JMfAgB86MkPCcTGuJCVQyqo7PHFLasTL1Zl6EodKAFfrJDLHifk7qWdS+lzUZAVDuhKHeTxoGoivdZXF0X0g75wvlVlosnYLcW6qmUVGD9DriVsg4z8zynkSIZc1yveH4JIvDZRwVFYVlVtvLpzghdEYHZ27jzSOgKLDVtnG5lCLkB+Wf3QF75f+v45Qo7eICIZciOUOmT21gh2M+vYfP7x56ePe0FvrAw5SBlyruWCIV6v8XEnEsbqtGXM7jdqJeQ45xHn/NcAnALwgwDeAOBPh//eMHzuJOf89Zzvc0iCwczDFyyr5sBgMDuQCRAzvA1UkC+I6sqQc20mljpMETFTBHkSNQ3Nm75GwUBRR7GDMkNuii64qWX1yFyikBuvZfXSRj895q51i5UFBxk6Qm6S44TK8nptl5tWQ4m8sEa8WbDVD9LXLTQduLaFF9yynHtdrtShplZBeWI+StOqnEOWtHROAmrfbiraVAFdhlyBZXWfFHJjZcgJpQ7xd7jczRRy3DuaPq6ukBMJFC/0sDHYABCrrRYaGQHcD/u5VvMy7JZiXVfqcMMSyZDTtKzKMRdAvD5pjlxyDJpEIScso+YcSJ9P1IdCVqJbrJDzggiM2FWPtLOcycVmRshFyO9/OoJdzpATSx2KM+RUBC8AbA2fZ40VMDsem022iFsWbsneOxwILb7A6Blyi83FOPZBLnaoycZ/PcIpf8no4JxvAbhj+M/AYF8QGMuqwYxCJkDM+DZQQZ4ETWLBFO5CW5Zgo5kk0PwgIUfITYFCTiZKVaij2EHZsjpF211QyA0tvA3bAmPxDQ0/5AgjXkmNskHUVNNs19YRb5MUO6iIzd0udpAVcvaI2WWrxFJ7aEjWPv/mZfzVF84Kr5svKHWYKENuAoWcbHvsBT2B3BkHgkJOYU0FdC2rBQq5PSXk1Aq5VsUMOcGyOhxLl3cyQi7yjsDuPBl/lrbUQfy+fhg3eiYEJ82PO9w6DNuyUyJWVsjpVF8U4g2yGi2rmnV5YoEq5KpnyAFxsUNiudwZBJhvOjlCruN2wMDAwdELegijELalJofFUodyy2qyXKpx0lDcpImGRTU2IeSOtjNSdoEo5EKWt6yq8uOA/H6vtawq1HCJulA+XyVqOpof1+FPR8vJttcgGODCxmiEnCpDLl7mZtoMHNtWGZLT4zSfG/cDdVtWDQwOBCLSngMYhZzBbMFkyBlUQb5ldbyJNudceK+cQm6KiJkieFKpwyQW33HAOS/NcpGharKTcdvxbIJ++kp19Q3F9Cvk8pZVudihKqkyO4ScTiE3ZZbVSCRQRm1ZpQ2rhzouAOD5Nx/KvU62rNalSpIn5mWKIIorO2LZwKiFECqICrnxCTmqkNuvllWRkKMKOf32kktCBuEAq/1VAADnFiI/GxtVFXIAsEOUTpRIPdo+ipadESYxISdmtpZBvEFW3zFJ17J6nJQ6XNkaKBWQdB3Q6wVa7JAQc7Jl1WIW5tzMGlyUqyg2wJdbVpMbiCr1axHwfwAAIABJREFUn+p8kDasagi5JaqQUxByWoVcKLesks8Os7GiU8OpbKsqQg7eKYGQ6/o7OQJuo6TUwQ/FDLmlxlK8zJakkCOKWqOQGw2GkDOYSZiMLYNZhjy+TXiqgQryJGhcKxpVwDkWA2NMuPD1Z9Symqim9gJRxPHKN38GL/y1D+Mdn3+y8t+pgrNlUMvqo1e2xzpeqEiaYIrIKGqbPDosdQCq50pR0MnLNJPRupKXugm5XbesUgLFHr1llRY6JAq5247PY64hKnKKSh3kQphRIKusyjKzKFQKuUkhEloay2qFltX9KnXoayy3VVtW6Q18y2K40s1ITx4sAFF2/NAScorxQIkVajU+2j6KttNOf+4FPSFyR3ejhWK3bpDpCLmWa6cEdRjx1CpJoYtToJbVRP0lK+QAiDlyBarRsmIj+fMzhVx+nDQURT/J/0xDyHWcFng0/DsW5KyoumUvzJAj12o7CssqoC522O7H69EihFxv+0bBWrrt5RWN24OgkMj1gwjMyivkGra4LzTIOJyGlvqDhFoJOcaYxRh7NWPsg4yxc4yxLmPMK/i3uzp2g+sW8iTKEHIGswQzvg2qoC7LqtCwOrzgcncpRHo/ocrC26scuQcvbeLes+sII453fv6pyn9XJUPuyFwDy0Plz44X4oKmFa8IKvJmmqzKVxUKOUBvEyqCoJCbYjJar5CbgFxSWVa3q1/q+2GE37rzYbz+Hx9UhpmrEEbiDQNqWa3CPa8RtcjhTjzBtC2G594k5sjRhkhAtMfVaVkdpWU1p5DT2ONGgdiyOoJCLteymj3eN4WcK5JICYrGuGhZFQsduL8EzglxorGs+iqFnKdXyMmEnDCmK2TI7Vamq1dAzrYblEAqPj/IltUEOoUcILaXFuUq0hvUDc26EhVyesuq6nzgpYQcKXUgGXKubYFH2fbb9DaFz6b7JOeEsJLGjq61WaeQk4+PnPMhwclht7JCh/W142hamUJux1eT9kUqOVWGHABBedcP+8J3mGb1+H6gNkKOMTYP4JMA/hDAdwM4CaCNOKdO98+t6/MNDCjki2TDVxjMEuQT3RTNiw32ELKSMoj4WBdJwoX1UAUh3pGfjQsvP8jvSHvVtEo/Z20Ei59AyGkUcowxKUdudNuqyr47TTcCVJZVQG7Vq0a+bgqW1elZBzK0Las1K+RWRlDI/cknn8Dvf/Q0/ugTj+NvvnS2/A+Qt6wKVskqllWyvy13MsXH86VihwUpQ65Zk0JObuochZC7ulN/qQMl30cpdZDzrEbdDnWBjl/a6kkJuaKcRLopLYvhUpc2rC4DPJu66lpWVeOBZoHRDLmj7aM5YkPIX7PKp+pypmtdrgn6PWT7crtkfeoy5Gg5SkLIUYWcipArUshVsayK9vK8ZVWVISdbVpmTEW1UIefYDDzMCDmZPKQ2ch520seDQFbIqc9F5Y2qMXp+iIgDzF0Bs+ObblHQge8toecRu3agJpGLcuQGoQ9mx3/HwNKcypxCrqZj4vWIOhVyvwzgWwD4AN4K4F8DeA6A20v+GRjUjlDO2JqiiYOBQRnk8byXdhCD6YEqO3MclZyQZ5Mo5ATL6myMP9UF5CT2vVFA9+n1kjwXClGFoFdSiDlyoxc79Lz8upkmdRi1TR7RWFbHUciFEZ/a469WITeBKlRVhFI1Qy6MON722TPpz49VbASWM79GbVmlpQ6H5zKy5fk3ywo5ybJq15OXJCtltr0RCDmpZbUoQ45zjgdXHizNmaOZeomyVkbTzhN1sl1QsKweAIUcJTwKLasSwUsLHWSF3CgZct0Cy2o+Q240hZxlMVA+tK45D92fZQU2VU+qCDlPuFmULRwl8nYGATjnCHjesjrXqJghJ1hWRyl1yJZZlSGX/D7ZlpadLYNAyFkWEOoVcpRM5GH2nXKWVTdPBgLqllUgX/aQ2FWpOi7qnwLAsEl4+kGoVshv9PTH6X6Yffe2PQeLDdeX1LIqrGfFDU4DPepsWf2fAXAAP805/7Ma39fAYGTIEwUTem8wS5Atgqa0xEAF1UV5349ACtIqQWhYHV5wOdYMKuQU32OvclCozWjHC4VGviIMKuTnAGLT6jiEnGoCK9/4Oqjwgigl0SwGHOpQQo5MwCqqITf7ImHqRxGamgbAgwyqypxvOqk1SkW+VoVKZXitYsvqRx68LNipqxJ5gUTI0Tl5lXPj+o5aIfeCEkLOdfI2uHEwScvqKAq5N9/7Zrz13rfixrkb8f7vez9cW0220fV+eK6hfI2skGMsr5Cz90shV6FltUgNK7f2Cgo5f0lUyI2QIdctKHVou6JlNYiyZa1CyMWvs1IiJ4g4Kpw+SkHLBeTG3XaJBbiqQk7Ij2MO2HDcUIVcEUldJW/PLbWs2sP/FQq5kgw5x2aCZVVWyAmEXDAHDDmswgw5jWXVYpnjS1bOJa+j+XFh/xQAYIMcUnw+hkIuyt6g42Y39wRCLhhICrm9LcWadtSpkDsFIADw9hrf08BgLBgFkcEsQ84IMQpQAxXUhNzoF0mifSa+sHUUFpBph2pSvVcKOVlsVrVtVTfpkSEXO4wKZanDlCjkRIKhKRAHOptQEeSsnWnNUKSqlkNEGTbJmFcpg6paVt/2ObHMpLKyTiJQ7FEVchoC6vhiC6eWs4n2Qk4hN3pDrwo5hdwoGXK96hly//TUPwEALnYv4qHVh7Svo+vjSEVCTtUEagmlDtqPqx2DSqUO1VpWZcsqD5aAKBsHugy5MoVczrJKFHK9oKeMiSgDzXWtK79LaCLNKeSIZVWRMRpoMuTaQoZcIOTHJeo4QCLkKivkyi2ryXLRGzAqy6qYIcf1hJzFwMNs+20MNoTPFjLkwuw75Qk5dYYcJd6OkLgFWTmX/EwbVqP+SQDA6na2LSLuI9ZPAccWsvcrIuQ8QsgtuFmrbJFCzpQ6jIY6CblrAHY457tbp2RgUAHyBbJRyBnMCqKI5zIRjULOQIW6CDnRPhNfNri7FCK9n1ARi3tFyMnrcK0qIVcwYaK4nVhWH728NVLGkB9GuTxCIJ9ReFBxTciPEwmGSS2rwBQTcmQSTVWDE1lWFetwo+eXEgSPX93GJx8VG0PXutX2gZC8t2xZrXJupPuabNH84W+8GQBwarmN59y0JPxOlUs1DnKlDhUtq/2gn1PjFCnkqFKn6HUrFRRytsUEEk5uWE1ek2CSa/A7HrkD3/Gu78Abv/zGSq+vopCr2rIqW1YjfxmcZ6TRKJZVSqxQhdyx9jGh1IGG7jMmEptFoC3bdR2TdC2rgEisqdanViEntKyGykIHoDohR8+dOsuqoOgfxgwMFPmr9HyQbMNBEALWAMyKt1/baaPjZFlwjm0JGXK5UgchQy6zrMptrAIZSNbnNiHejgmEnKiQ2xrE65FaVhOF3NUtPyM7GQdY/J7/7MaMXCuKyvB4duyY1ynkwoGwv83Kjdq9Qp2E3J0AFhlj31DjexoYjAV5osA5ags5NTDYT6gmwUYhZ6CCahI0TkmBqmWVTsCmlZCQoZpE7dVdXnkfpja6IoiTHv3E7cRiE/NDhc9mP8DVrerNl7oA9GnZ7tc0hQ7AeKUOMiE3reHVdBK91KYKuUksq+q/LSsqeYeiWXilskIue2xbYqlDlXsFawUE1M++9DZ8+DUvwYf/40tyFvJxGnpVGNeyKufHAcUZcvR3fU2OlB9m9m7GRAuvDEogKBVyNVlW33rvW7HSX8Gf3v+nOYuuCnQ/piRcWeZZAkEhx5BXyFFCTlPqoCKgE2KFcy5suyPtI0KpAyXkqqrjAPGcvBsKOXn808IMZYac5mZRp1FgWaUKuUY1y6qnuGEogzEm2lajSCC9ImsDf/3QXyNgGVGaHNe9IMqp4xijZPSIltX0/cWxo7s5RInc44uEkPMUCjnmgznxfm7BBvcPAwAubw4EFSZY/NnPpIRcwTWHx7P1v9DQK+TqUg1fj6iTkPsvANYBvJExVmc2nYHByFBlGhnSwmAWoBrHZmgbqKCy6o+jfilrWZ2Vevv9tKzKRHvVYgc6GXE1LatAPCG57fh4ttW+wo4ETM921xU6AONlyMl24mlViNKxLSjkJmpZVf/ttQLbas8L8beKRtW1Ha/SjdSQrH+bMTG7bESF3GGJgGKM4fYTCwKJkEBsZJxgnUnWtS2/WguyipyaVCEnFDq03VwuHIVAyCluBtRV6rDaX00fP7r2aOnrVVZEoLpCjl5jRczD+mA9/oFb4MG8oJDTWlYVx8aEWNn0NlNVWMfpoON2BIVclxCnVfPjACknraaLQq+iQk5lWU2uG5iziT+6/7/hI09+BADQaYilDpMq5KqqxGXbKiW9Prv92/j1z/86vjT4TQBDZdxwjAzCCJbGrgokltUChZzGsiqT4nSsnh98GX9y359gY7AhKOGOE4vpjqSQ6w4CMDvbrzvOAoB4TFze7AukL7MCHJ1v4IbFapbVAMSySgg52rKaK3WYkuuDg4KxiDPG2EnF0z6A/xXAXwD4AmPstwF8CUDhmYVzfqHo9wYG40CpIuK81hYTA4P9gGryZyyrBiqoJqOqC+cy+IqWVdkCMgtQfY9J1EKjoB6FXPE91tuPz+Oes/Hk8vSVbXzbbUcLX59Atw6m5SZXoUJuRMtqFHFsSROhaVEKyhAy5Dr1ZMjpSM2VgmKHv7/3PDaHDYFPO9LBta0Bul6IMOLY7AeCek8FOu+zbcmyWjJGOedSq6heESZjtyyrXW8ChZwmQ86PfEHNpSXkiE1YZ1dNQMkPW6HkqqP10ws9gbA5vX4a33rqWwv/RteyShVyRcd1ej3l8bX0sYtDACxBIeeHahLDVxxLktB9mh93rHMsXjabKuQyokalPNSBknd1FS0VWVYFglNBSCfnpuaxD+IvH/wK3v6QhQ9+/wfRoaUOfjWFXJFqVFDwF6wvR8rYy8jGECvBaQDAdnQBzNkCD5ZEhZytJ+Rsi4ktq4MihVxmdfUjH2EUwh4WAiVjlbkruMf/XdzzFY6V3gq2B89N/+Y4aeSSFXJbgwDMyvbrhcYiLg4fX97so0nIPDAfJ5fbwvGu6CZgQCyrS83Muk9JvkEolToYhdxIGJefyN/KEvE8AH9Z4X34BMtgYKCF6gJ5Sm9iGxgIUI3taZkYG+wtVNfk45U6KFpW6d3mGRl/asvq3ijk8oTc6KUOReoAQC52qKbCAfT2rrpUGLsNSrjIJMOopQ5b/QAyzz2tllW6XZdrUsjp1oWuoIFzjr/8bFbm8GPf9DT85efOoLsaTyzXul4FQk6ckNPdoEwht9kP0n1vvunkSIciiI2CE1hWJdvjbijkZCurnF+VgBKnR+aaytckoN9fZZevw7IqEzGn10+X/k1fU+pQXSFH3otn6rwGDsUPqpQ6qBRyQ8uqYFdtHYmXzRFLHRKU3WShcK16CGIKT5PHB4glGfKNviji6XWB1YppoYhHeHLrSXTcm9LX7QwCPSFHFHKyDVRYxooq8YaUb5aSjZa4DS13BWGwlN5ckC2ryTZLwBgD4xkhlyt1oBlyURM8ctI8Oi/y0Lbiv03O33brIpLSha+tfA19/9nDzxEV3nKGXKyQyz7rUCsjzi5vDnDTLVQh5+PkUhtL5EZM0U3AkGXvu9SsppAzhNxoGNeyymr6V6dl1sAghUpFZIodDGYBJkPOoCpCxXGwP8ZFEr3bnjS50Ua3uu7G7zdUFouqNsZJkSPkKlpWfUUwtQ5isUN1y6qOkAunZLvTbUgnkYA4Sa+ikJPz44DpVMhxzgWV0HKnpgw5zd/qLKt3n13HAxdii1fTsfCqF9+Ew4QIqpIjJyjkGBuJCFoX1HHFxJ8MR2pzHec8zDnPkTq9oIcwKidF5YZVQJ8hJxN1OuJO1zirAj3eqKytdVhWZaviY+uPlf4N3Y9bgkKu2r5Ol7XHMzVbi8VETBXLqur9t4eWVVrokKitqGW1R8jSUSyrTk1FSxe2L+ADT3wAXb+rLD5I0G5QxaE4Xn3y+ZadfR8v9DDXrFjqUFUhV/GmlJyxl9yAYbZITrPG6nBZKSGXjUNZIQcAFs+Ub7JlVVj2qCkoLAdBNn6ayfhk2T64Psjeq+PaaQ4sAHQ9kZDb7gcAsawebi0jORSudAdo2KJC7tShNpbJzQ7VuS1BSCyry4SQo8pOL/Rqu0lxPWJcddpoZy0Dgz2GIS0MZhWqcWy4ZgMVlC2rY1hW6fE0taxKeSyzAGWG3B4p5HIZchUtq15A1AElagqaIXd6lAw5XanDlJxTi2xXVPVRRRm22c9PWqYxK0duouyUNCZWf1/SCLjQTMtDVrbVxMXbiTrue593EsudBg4TYqysDAKQMuQsSyCCys6NoxBQMhhjaDhWqgTxwyi1n1UFJSMoukEXiySrSYVrO9dyz02qkBPWx3x1y6rq2GNbkyvk5OU+vX4anHMhVF8GHYM6hVxRdANd1n6UKeRaLA7Ir2RZVSrk9IScTiGnaq/VoY6ipUE4wI+8/0ew0l/B93zd92Dgvyz9nUx2FZU6CAo9otoahAMhj7FXVOpQUSEnRGoUWFZdR7SsJsdAavMEAMsdEnLD33slGXIA4CAra9iUllXIkIua4JGbkoCU0E3ORczK9sFNQsjNNR3MUbvvQFzn25JCbrm1hCNzTVzbHoBzwGbZ/swshWW1QJXP2Q6StUeVd1Qh1w/7wvnUKORGw1gKNc55WNe/ur+QgQGgs6xOx+TBwKAIqgs9o/40UEF1yBuv1CGfU+ZIjWWzAJXNZ+8UcuLnVLWsesK2KVZTnFpup4qRla6nJUlkaFtWp+ScKk7QJUKOtqxW2NYqFcE0EnKUkGi5tpQHVU/L6snlTPWzolDIrWwP8L6vXkx//vFveToA4BAhxlYrENN0HNrWaFZJup8dGiE/LkGTkBTjNK3qFFZFrZIJlAo5TYacrDDqhWrijm6nIyUEZbNEISe03Y55qJCXeyfYwcXuRc2rY8hkc4IWJd8LzoN0zOxEmUKubQ2tilThFGksq4qx0B0SKJSQSzLkdAq5smM6hVtD0dKZjTNY6cff+bMXPiucX6jaEBBLHWRVrU/y2RixhMaEHFXIVSt1KFLI+RUtq3LmY7KNZIWcNVTIJeNI1bIqwwJtWdUr5HjUAHj2HdWEXPbclp+917xEyMkKObnUYam5hBOktAER0VIxH6eWW4JCrugmICek5aHWcrbMRHXnhV4tY/B6hbGMGswkjGXVYFahbFmdkomxwd5CqZAbQ/1Cb3A4qWV19hRyKovFfrWsrlUtdajYMAcAliU2rT52tVp4vE5VOS1W5aKmwFFLHZSW1Sk8/lJCou3awnoYp/glgUDILWWqH5X19O/uPp/uc8+7eRnPuSlWXtCm00oKOU4JOUsk5Equ+6gi7NCIllVAJADGmYDKhQ4JilolE4yUIScRdT2/3LJaRlAKLaslltVxFXKq9VCWIydkyGksq0XHdWpZ3Qkz8mxuSMhxkiGn2n5BGCkJyK7CsppmyNk0HJ9aVkdQyNWQ60rzz1b6KxgE2fdr2Hq7f14hpya7/NAXFHI7g2qlDoUtqxWLjdycZTVRyEmEnDs6IecShdy2rJCjKs+oKVieadNqegwWCMw+gHj9dJo25iiZKWXIbUuE3GJjETcsZuOKE0KOWT5OLXewSAg5mqdJwTkHt7LvcJgo5Cgh1w/6ptRhAhhCzmAmoToZGdLCYBZgyGaDqlATcmNkyNHQdIVCblqImTKomvH2yrIqn5/GKXUoUgckoBfoVUk/3TqoKzh8tyEqZsRJ5ailDkqF3BROPCjp1m7YgtplkiITj/ytoJBTtKze/dR6+vhVL8qC3qlVUlcGQRGGokLOHqFlle4Dh0a0rAIiCT7OBFRHyBUpghIoW1Y1GXLy+1EigIKu7yMlllW6LyktqzWUOqi+T1mOnG5/Fwm5CFxz3USXtUsUcnN2XiGn2n664+KOl1fIqTLk+oJldYQMOWtydZKcf9bn2T6ay5Cj61Mi8dObW7a4/WSF3I4vKeRstUKuSDEqnAMrWlaDkGOQkIi2SE6zVCEXpt+FZsgdaYulDgBgsxY4j9+/G3SFDEhRIdcUlGp0/DQUllUgIzXnGo5IZnoKy6olKuSOk/N9RIjkuGW1BdtiWGxlz28qFeBcVN5pCDkv9ATS1mTIjYaxMuQYY380fHiRc/4r0nOjgHPO//dxlsHAoAjKJkpDWhjMAJRksxnbBgqoJkE6+2ER6AQjmYAK9o8ZudmhzJDbI8uqvF8XBSxTVFUHJKiqEqHoeep1MC25rEJToCsr5EazHCoJuSlZDxQ9X7TxtkbM0tOBrsMbqUJOYVl95HKmJHn2qWySRxVylQg5SSE3SsuqQMiNYVltTKiQ01lWizKzgFiNonpNVYVcpQy5CUsdaPxZXaUOQLlCjlrPqc3SthgatpUSBYMgEo6HCeiY6YaEkHOOAtgEYIPBAkeEkMcKL6rs0hGz24oMucSySjPk6JgYpdTBrSHXVW4I9bEGDNVfst2/WCEXf76czzYIB+iQUoecQo5l67FpN+EwBwEP4EXekPDJj8n4u0ZonvgH/P3VVbx05VfxrCPPyr2OZux5RQo5Zxtgg3Q79nwfzCaEXCtPyLmWExc2DMmzLW8Ly61l+JGftShzBnAXnFhW6X6Yrl+JkIPdA8L5oWVVtPtSyC2ri41FwbK6MyCZj06Q7t/LnQY2+/F7rff83I0JP4xyyrtsmbP3H4QDgfRM9rNL3Us4v30eLzj+AljM6MB0GLfU4dXD/x8C8CvkOQ6g+tEjfr0h5Axqh6pdcFomDwYGRVDnI+7DghgceKgmo2NZVgWF3LDUYQZbVlVk9yRqoVEgn5+qqtc8mp9TYfI2FiGned20ZAcKGXISadmkzYtjZshN4/in277dsHPqoXFB1+EpIUNOJJ68IMIT1zLlyO3ESk0nhFX2A7rv2Gy0DLnVLsmQG0MhR/e5sRRy0XgKOaqOW2wspsqmqqUOtbSsCqUOu2NZVa2HUkJOU+oAxIR8QhT0/VBJyKWqSuZhEMVEjGM5mLeXEBNygM1cBDwe017oCYTcIKT5jFa6PxWVOggKuXDMUocaWlZlhVzA1gDE6tWillW9ZVUcZ17oYU5Seeksq4wxzDfmsT6IVXrb/jYO24dzy+yFEZzF+9A4/FlcGgB/eO8f4k0vfVPudQ1bJM91GXIAYDXW4AW3AAA2/XUwFo+JlrUgqPgSuDYDDzvpeyWEnLDf8SYABkTqlt7kZhHNkIt/7oED6FQpdWjIGXIZ0bu6zYHhLr3cYWkxynLHxVPD7pI4Ry6z3wLAjj/IVHvcEpSLTUsk5BoNUTG80lvBy//u5eiHfbz2xa/FTzzrJ2CgxriE3K8P/7+qeM7AYN+hkoxPydzBwKAQpkHYoCpUdq1xJtu+kCGXWFYt5e+nGaoJ9X4p5Pp+pJ0wUlBVTlmGHCDZjCp+N0reuDZLt3c4Jdu9ukKuumW1cfjjcBYewODad8EPX1jTku4d6LZvy6UONbWsHl1opuOl64XCeD6z0k3H/KnltjDRpGUCquw5GQIhZ4sZcmXCrPWdyTLkGiNmEMrQWVbLFHI0P+6WhVtw/8r9AGKiLeJRTolStWWVru8jc03laxKUKeTsXSh1AIAnNp5QfscEulIHIL4hsTVUA+mOf8khlbmZWuxE5wRcJxujMiHXcTvp7+j5cKntou/Hr+t6IQa+lxJMFrNwqHkoXi6ikPPC8UodHGvyc7KskAuY3rJadMzIyK68Qq4tKesGZB+gpQ4AMOfOZYSct43DrTwh54cRGoc+k/58fvu84pspLKvDY5Ws4gMAy11Jx9GWt5Ytj31I+d62xcCjbBsmxKacHwegoNTBHi6PRMgN1+F80xbIzO2BrJALwdqiki0gCrnVbQ53uPoWOuIYTbCuuOG00svGAKK20HBMLauDcJDLkPvy5S+n9vhPnvukIeQKMBYhxzn/pSrPGRjsF1QEhbGsGswCVOpPY1k1UKE2hZxCheXWcDf+oEFtWd0fhRwQkz+jEHLVLKujWxPp6+abDtaG+XbTUmYwEIov5Ay50UsdmL2FxvEPgjGO5rE74Yc/WN/C7hHyLavVGijLQNdhy7FxeK6By5vxBHOl66WquUcvZxaw20/MC+8hKORGJOQci42kzKKKsC5/Cr/06bfgO2/+Trz0lpeWfi4ANCZsFdRZVssUcrRh9cTcCZxeP51OfPtBXyCIgDjXikLVshpFXMrUKyYo6eRbdeyxaJbfmNcoqvXQC3o4v30eNy/cnPsd51wsdciRSOXHv2RZLScjIk50TsAhBjCbqUkVQLwB0HJttF07VZBd2M6I1EPNQ7Ct+PgjlDqQ5tZRFHLCOXlMQk5WyDEnJugsls+zE4k1qWW1QCFnWUxYJzt+NuaoQg4AFhoL6WNdscOK9zjszpPpz2v9NeXrHF2pg0Ihxxqr6XbcDlbT5+ddNSHn2hZ4mKkcU0KOWMX5kJDjJS2rsmU1WYedhoOWa4Gx+EbDIIgQhFF6c3Rbtqw2F+HSDLkw+9y5VjY+lolVf0ORXbvez8YE423hd7JlVVYh0m2WtPcaqGHMvAYzCdWFkVERGcwClJZVQ8gZKKBSMNVnWZ29llWVqmAc1cs4UJ2fqtj1hJbVCqUORbk/OlDyZp4EQI8bHL7XGNSYIbfZ88HcjdTCxJytqSSk6baXFXK6zMAqkNWIVGVFbas0P+7rT2STbmD0DLlAsKyykVpW6T7214//Lt5z+j14zcdeU1ockGDSVkEdIVfWsnpth+SQtY8Jlkc5Lw6oZlnd7PvpcWi+6eTsnjJGUcjVaVkFgNNrattqEPFUjWdbLNdSKiiENcRzsqw0yP9457jQ0mkjG6Oy7ZiOA9e2BPXn2c1L6eMkPw4QFXJ+NAAQv8coGXJiy+p4+7CskEtUgg3HEpRRAIQiGLnUIdknmSWOu2Rd0WKHrpftA7IddM6l7aXqfeLjlkuIAAAgAElEQVRx70PCz2v9NUQ8//1djWUVlsKy6q6mCrrtICP4FjSEnG0x8DCvkBMKHcLhmCGWVapUTcisXKnDcPnmmg4YY6Lld3gc55zHijla6tAQLatUmddpZutnmSrkFNcc62RMWFwk+ssUcnSbrfZXYaCHIeQMZhIm+N5gVmEsqwZVUZdCTm1ZZcrfTzP2UyGn2q+rNK2KGXKjljpUm7RR8mahmV28T8txxxMUcnKG3GiW1c2eL1qcmA8/mI71QEHHdUsi5AY1lTo0bEto6qTFDo9eyQg5mh8HxBaqhN/Z7AelxC+15tuSQq68ZTXbxy7uPBX/DY/w1nvfWvh3CcRSh9HHgR+q9/GiVklAVMgd6xwTFHEqsq1KqcPKCPlxgLgvqZRc1gjbQQdKaCT2TgB4bENNmIoKzfwyVTn+JedNZmefvdxcFs55FvRNq3KMAA3iv0gUcrSt02KWQG6AxXbEKsf0BK41+Tl5wxMJOcuJiSVVHELLKSh1KLCsAhCLHTyikGOSQs4tVsit99dxMfys8FzAA6Xl25WuWYoUclZjNc0a3AkzQm7RzVtmgZgM5VFeIUfHbxTF+xRVyNGxYw1LR1BgWQUgjKckR24QRAijUCxfaC7icKeRKhs5IQKbjWx7LXeKLasb/WxM2FzMl6MlG17oSaQnF7bZ+mBdaJ81EGEIOYOZhCEtDGYVZmwbVIVqXIyTiRYItkhFqcMUKoRU8BQT/ypB/3VAZUVX3a2W4RESqUqG3DhZYXTMiAq56TjuCCHvOYXc6KUOdNLDrGBqyi0oxFIHq0bLqriudXlwomVVVMhZFhMaT8uUokGOkMt+V6SQ45wTS2yEHWLrvPPMnXhk7ZHCzwVEwsQLR19v4yrkaIZcTiHn5xVystJMRciNUugAiOpS2c4ISKUONVhWn3vsueljXbEDJZObCru/QCJ5GstqqpDL1uOh1iFhW1tMTaoAEintWIKi6VI3225HW0eFv6PbMFFJqdarDmLL6pilDgPJsjpUyKnWpaCQkzPkCiyrACSVF8mQkxRyTTsjmi9u5q2o73703eBQ5J4p7JGyQi7LkFNYVt3MstqPMuvyciPfsAoMt1OYLWtCCNJ9MbGsMtqyGoqf3XQsbYZcZ7jOVDlysTrOS5XbHacD13JhWQzHF5LsumyfbrjZ9hIy5FSWVTImbIgKOWq17gd94QbFIIgEYjTiUY7wNcgwFiHHGPNq+qc+ExkYTAjVyciQFgazANXE3Yg/DVRQqYKr2hQp6IQ3scWI9pjZGIAqJc5etayOq5CjpBgNrdahPRYhRxVy2WRAdSw6iBBslHLr4oiWVZmQixVy07EeKOhxoOXYaNhxNhEQj6lxr5cG0ro+Mp+3rBY1rCYQc+SK9wNK9tgWE6x1RV9jexCk+12n5YMjezEHr6SSoyT4pJbVhpV951FaVo93jqPjFCvker74nOo1q0KhQwWFHCXkVC2rNVtWBUJOY1ntFxQ6ACIhX2pZJXlcS80lgRyzoM+QK1LIXSWEHLWsAqJtFSwe86Mo5JwJ8wwBVYbcJoBIrZAriD9Iz0u2SA4n64qSeT2fZOZJCrkrpE/gU4+LZQ1hFOJdD78r/Znz7PurcuTyhJyaNARiy2o/GJZ/RBmJtNzUE3KCZXWQz5BLSh2Kxk7TtXKW1cSGOj8894rqwiEh18/nxyU4sRQvF4+yz3XsrBBCyJBTKOQ2iWXVQbFCTjgeShlyALDSMzlyOoyrkHNq+jd6pZGBQQWoTv7GsmowC1CpUkxhiYEKaoXcOJZVopAbTkqoPWbcu/EHDSrr4V61rKosXSr7iIy9KHUQLKtUITclRKysWKEQrJol5CvnHJv9QFBUMMYx0NgODzKEltWGDcaYoB4a16pNVYZNR7KsDgmfoobVBKPkyNGsTNtiYrtnwRilRN/SXJD7/Yef/DAeWn2o8LNdmpk0hmKUqqsOtzM7XJlllSrkjraPlmbIyaUOe6GQoy7WOkodnnfseenjJzaeUNrfBgWFDgAqWbMzy2q2Hpeby8LxlRVYVj1ZIUfG97V+lv13tC0q5KjaiFnx2BwlQ062C46DnEKORWB2V7kui27u6EodEgJKVMhl+6Fc6tAfZDTBSk8k2T527mO42L0IAIiCOYTd29PfqfLKZMuqKkPOZknTaYjBUBk3QMYKHm5pCDmbCZbVRBkmZMhVIORcm2lLHZJx1CHrrju0rMaFDmJ+XIITC8NxxbO/syghV5IhR0lal4mEnFxGIlj4pQw5wOTIFWFcQu72Gv8ZGNQOJWkxJZMHA4MiKBuEzdg2kMA5V6pDxikpoKUNSoXclFgXy6Cy3k5i3xvts0cvdYgiLqoXK9ibxil1EFpWCSGnKg05iBgUqGboz2Xk6/YgiI+10iSzH5Rbiw8aelKGXPy/pfz9KKC276Zj4ahQ6hCvp6KG1QS04bNsP5AVclVbVun7LnTUpOqb73lz4Wc3J1TI0UKAwy1CyI1gWT3eOY62m5EBygw5ycYa8AB+JH5ngZCbH1Uhl59O1qGQo+vhxrkbcbx9HEC83s5unc29XsiQU9ksK2TIpZZVW8yvc221Qq6o1EG2rK4NMkKCZsgBomUVqWV1BIXchDESYRRiy89nrzF3Q1kYdKH7FJpH74LVvAg/5MLNoZSQs9SWVdHuqi918P2M8Hm4/178/WN/n/78Vw/9Vfa69W9E5C+nP6/2VIScWHoRjxUuZMg9Y/HW9PGAxfuYzzOF2BEdIWepW1YFpWtCyLFs3xoEskKOgzFx26WE3HCdzTXyCrnuIBAKNJaahJBbHLa7EoWcR4jAsgw5OiZkQo4q5AaBVOoQRrnxZAg5PfK3pSqAc16tfsjAYJ+gstIYhZzBLEBlRTBj20CGbgKky80pAh1zacsqvds8I4SwakK9dxly+XW4UWJZpdllseVwVMvq6KUO86TUYRqyA4MwyixoLE9ajlLqkNh5ZNVHX0GAHHTQ40BCXLRdG2vDPKax2pjJurZYTNIcFjLk4klgUcNqgsOUyCtTyBW0rBadG1cJIddpZ49PdE7g8s5lAMBHz34UD6w8gGcdeZbyPSZtWRUUchUJuV7QSye6juVgubk8coZc8j5ugyiQtke0rBKCw1VZVi3ROsw5r3SMoqDfpeN2cOvyrWmhxen103j60tOF1xeR70A1hXByn0GwrLaWBNKREbVRkWXVtZlgWd3wMsvesbZoWRUy5NjoGXLOhAo5VRECAFjOhnJdvu4Tr0Pj2ENwlr+I7ulfQN8PU9LLKyl1oLll3YJSB7v/bHDrvWB2HxF8/OKnfhEPXHsA33/79+PzFz8fv4gz+GvfDPfQ59O/UxE/9Jql54XxcYP5KQHWtJt4xtLX4fRGnB0ZsGvD/zOFmKxqTN/bYkpCjqpV+bDUwYaLRJ8mZ8i5bv5YR1tW6f+AmCEnFDo0Msvq8aRpVZNdRwk51TXH5d659HGTLQu/k1tWHUu0TRuFXHWYUgeDmYRaIbcPC2JgUDPUdux9WBCDAw2djXkcxZeQUza8Y+9aVCE3GwdX1XljPxVyZRlywnapaG1qjqGCouTNwpSVOsiKLZkQEEodSgiVlJCz1JPMaQIlH9upQm50slZ8z3xWn6pltahhNcFhqpAbhZCzR1DIkfdtN7PHzzn6HHz30747/blIJSfnUo0KOnYEQq7AsnptJ7M9HmsfA2OsNENOZWOVbaur3WxZaKmGDs86lalwnnnjYu73jLE0lxDMx5mNp0rfUwYlEufdedx26Lb0Z1Wxg2hZVZQ6VFAIR4oMueXmskSOZeNTbsqlx5yGYwsWwy0/IyRyllWaITeWZZXGSIx+bNYF7jN3M6eQ2/Q2Uzu35W6AOdvCMSM5N8iEXLKuji1kRM7WIBuHskJua+swumd+BuHgePrcOx96J378Az+e/jwXPh88WAYPMvWWivihBHJ3SGTR5VtoLODmhZvSn0P7GvpBHxGLX8O5hcNtkZBK4NgiIVdU6mAzMXeNwnXy5/zMspoo5Ijd19NYVgWFXD5DjirzltrZ8qgUcme3s/1s0Xqa8DvbslObMQeHZWX7lBdEuRsBJkNOj10h5BhjFmPsRYyxVzLGfmQ3PsPAoAjG1mcwqzAtqwZVoBMvjaV8iYoVcrNiWVVNqMfN0hoVKvtnmVWPlgm4CgWDCu0KGUoy+poMuWk47gj2MYWtTih1KCGhMoWcSGQMwsktq09tPoX/9Mn/hHc8+I6J36sKKMnabsTroDlG4QeFQMgNid+jilKHRwoaVhMcGiVDTrJt081cpJBbI4S328i26UJjAT/9/J8GQ3yM+8S5T+CrV7+qfI9JFXKUkKMWxqJSh0QhBmTFAFRdVaXUAcgTclSJeKSCZfXFTzuEt/zoC/H673sOfugbb1a+xmYMYB7mvu6N+N73fg/+9L4/LX3fBH7op3ZQm9lo2k3ctpwRco+t581afcUYpKhCOkdVMuQKFHID6ZgznyqaAmwFeoWcKkNupFIHS7Rkjgo5Py5dFmcjR26e2TgjvsbeFo4ZfhjF6jNLJHiSdXWMHBe2PGJZtURC7tr2ANw7hp0zP4PDeHH6PCWY5/ovAQDwMCP31Rly2frZHuQbVhcaC3jaYjaOubOCq4T85sE8Wq7aWOhYFhCRUocCy6rN1Eo1AHCUhFz8XecUpQ5dQSFHSh2IQu6GEoXckpQhR3M3V/ur2PDjdcAjB4vWydzyUZVcUkYCxMdDWXVpFHJ61E7IMcZ+HsAlAF8A8G4Ab5N+v8wYu48x9hBj7Ia6P9/AABCtPAmMrc9gFqAjP4rCqw2uP9ALctmmyEc8FvqKDDkhIHkKrItlkPPYEgyC0dfXOFApGlWNZxSjFjoA46mg6Ovmm1Qhd/C3u0gS5RUzYstqWLitN7WW1XxA/qh441feiPc9/j684QtvwOMbj0/8fmWQW1YB0c43Trsw/ZtkvcqlDl4Q4UxJwyoglgqUEXL0WGez6i2rVCHnONk2XGws4tblW/GyZ7wsfU6nknOlVsEycM7xts89id/58CPY6vuCumqpsQSLxe/XD/u5jLcEtGE1IXU6bqaQk9VwnPNcqQOQJ+7S9Wz1cdelP8fbvva2wv2BMYaXPedG/Mg33aJUowGAZTE48w/BasRE1D889g/a95NByYw5dw6MMdy6nGV8lSvkFISckBlZ0LLKvIwUs1x0nI6oVqtoWW04LCVQrOZVcMSfeWr+lLDN4mXLt6yOZlkVSwtGhU4hZzn5DLkzm2eEn5mzLRxT/DBStpcm6+r4YkbidAdqy2oQRpmtPGrilvCn8HMv/LmUKAeA25Zvg+3FcfQ8KCbk6PrZHgz3LXJzZcFdwM2EkLPcVVwirbg8WFCOqeS9VQo5VamDg+y7ywo5W0HIwRoAiDLLqkIh15Utq7RlVZEhR89Zcc5hPEYjDmx7WeHDQytZqU00uBGukyckRUIu+1tly2rfKOR0qJWQY4z9JYDfAHAUwFkAudoizvk6gE8jLnT44To/38AggUptMA138w0MyqDKRwRM06qBCDpMHJsJ6qBRix0CRcuqcDd+BhRylFR0bZYSjpxXm2hPCtX5qUwhJ1ijxiDkKltWhQy56VXIqSZTjm2lFseIqxXICTZ7Q5uTZFn1o8kVcg9ceyB9fG7rXMEr60GPkKyt4WSMtqz2vDHUXkLDavxenYaTEn2DIML9FzZKG1YBkZArLzfJHsstq1VLHVST2Z963k+lBNmnL3waT24+mXuPURVyH3/kKn7pPffjTR95FH/2qTMCmdO0m5hzM9td11Or5GihQ0LICQo5P2+pjnh+2WRCLiEoG4c+jX948u34jS/+Bj5+7uOl36kINmOw5x5Nf77QvVD5BgedzCfr5daljJA7s3kmR1qWEfAtWiagIZ0jznPqOMaYWLBA1EbyMsiq3IRAsZqX0udvX853GgoZckmpwwgKOdGyOplCbsHNlKvM3cidX3IKOWdbUN16YZQ7TsbPx9+LKuR2NKUOqzse6FDx/Aivfs6r8ZZ/9RYcbh2Gwxy85kWvSa8/RlHIdVOFnGhZvWk+s6yyRp6QU5VbAPFxh7asbg42wTlXZ8gRy6qskLPtfPwBYxywBugMxzM9ZmYKuRDQKORUGXIyibxMFMk0R+7B1QfTx2H/pFplTgg5juxvB4GX+xyjkNOjNkKOMfaDAH4MwGUA/5xz/nQAujX/DgAMwHfV9fkGBhRKW58hLAxmALoJo1GAGlDQ451jMcG+M6odTWjyTFtWxfDeaYeYx2YJio9x8rRGxTgZcnKbXxW0x7Al9gTLKp2IHvxjDlVt6daRqJLTb+tUsZgLKp+MkBuEA1zsXkx/VoXy1w267ZMxITYfjq6QE7OzsnV6hBQ0fPaxTCGha1gFJlDIjdmyyq18IPozlp6Bb7zhG9PnZRICoGMnwpM79+CBlQdyr6H4/BPZtOies2vChLVhNwQiRFfscLmbWVaPd+JsLZohJyvkdPZXSgZwzlPLqtXMSIgvXfqS/stUgG0BDiHkekEP64P1Sn8rK+QAYL4xjxvnbgQABFGApzbFXLrSUgeaGak5roeRSMgleVyU8OKcvI9EOtDjsmtbKYEiEHKHigm5pGW1ajYoIFtWJ1PI0eWznM2c/TenkJMtqwEvVMjRDLmer1bIXduS2muHx5dvO/VtuOsH7sKn/pdP4SU3vSR9nmbIrfXXcp/tCgq5JENOtKwe7xwHhtvWcrbxJMk9jMJ57Y0v17IA7qQqtIAH6AU98VieWFYtfcuqZauPda3mIL32oiUh3aGabbsfCOQizZBbbMU3RQSFnEQEirbV7LojyQkEgKh/UqnEp4RcRAg5n+fPY7n2WzNvSVGnQu7VADiA13DOP1Py2i8CiAA8p8bPrxWMsUXG2K8xxh5kjO0wxlYYYx9hjP3ALnzWexhjfPjvz+t+/+sRqvwEY+kzmAXoLat7vCAGBxpC0LnFxmrXTCC3xsX/T3bxf9DgS5OoSe17o0KlfB0EUSExMk6pQ2vEUoco4sIEk04GdGrdg4S+QrUlQ8yR06+TmJDjuQw5L5zMsnp+6zw4sm2pCuCvG3Rctdy8ZXWcMhNRIZe911FiW/3c4xkhp2tYBcQMudJSB3L4sS0GS2j3LGhZJe8bIiN/qLrk1Pyp9PGlbkaoJEj2O2fpbnxk/b/ih9/3w3hw5cHc6xI8cinLVDq71hMsa027iblGRiqoCLn/+91fxZ99LsuzS4oBijLkdOOJKul2vDAls6hK56G1h3J/NwqYew1WQyTgLmxfqPS3KkIOgGBbfXT9UeFvRNt0SamDpnE85ACzs88+1DoEQFKrEbWRbDv0JWI6sQParWz8fP2hr88vG7GsstSyOp5CbpybZFQhRwk55m6kyvgET2w8IfxsOd2cZZUqthKkCjlCyPWDbP1Rhdy1bSmbz4+E1yVjIlPIZaT0+mAdQSSa9MQMuUTtnB275xvzsC0bLMzKVe5fuTd9XKiQSxT1UtOqyrLqFpQ6MFt9E46WztCSkJ2B2rJKCTnGGF54yyExQ06KWaBNq+u97LNkhZzr5K8zBELOu4ZnsIt4IXsEL4o+l3vt6vYF4E/+FfCmFwBvuAV4x6uU3/d6hForPh5eiJiQe0/ZCznnfcbYBoBjZa/dDzDGbgLwCQDPGD61DWARwEsBvJQx9hbO+U/X9FmvBPCKOt7LIIOKtJgGe42BQRl05IdRgBpQ0ImoxdhYVsUEgaQeA8Rsm1loWRUtq9ZIYf91QEe0r+14uHGprfzdpBlyVb4XJWZariUSsVNgVdaptijiiXs8ESpVyDEfjIn7z6SWVdkKuRcKOaHUwc1bVsdrWVXndx0h9rQvP5kpV3T5cYCokFvpeuCc5xpyE4SyQq6iZZUqQaiag+Yv3dDJoq4v7eQJuUQx48w9kj73pctfwjOPPFP5mQ9fzgi5c2s7eL5EyM272TqRm1bXuh7++otn0b4lUzKlCjlX37KqG09UJUPJSdf1kWzJh1cfLlz3ZbA6pyGPpPPb5/Gso88q/VsdIXfb8m341PlPAcgXO/Q1pHCCKqRzFOUtqwAEUopHdippkUkVWblcVSFHSx3GaVl1Jjw2U4XcqflTcFkLPu+DWT4skrEYRmFOmagqdShSyM03Y9VW348Q8jBVByWNnYCCkNNsr+w8aGOxsYRNbwMcHOuDdaHJVrSsqltWAcAJj8J3YpXoAysZ+V1EyCVjg0ctADGxmSPkwiEhZ+ktq4ypG7tbhJCbaygUcp6+1AEAfutVz8P/+MoR/OHZ+Gc/8hFGIWwrfi+BkBseF7vdq9m5iTO83D+Nf3HtHHDX/wB2Vob/VtHEeWC4SM5fvRIfHS7rA46LH8aNwnLsIELv/JfQTq5PSWnG9Y46Cbl5AFuc86r97y6A3b/tPCJYfNa5AzEZdwbAj3LOP8MYawH4DwDeAOD/YIzdzTn/4wk/ax7AmxDvvRcAfMMk72eQQUVaGEufwSxAmyFnCGcDgkBSyLUmsqySllUrr5AbJ0D6oIF+h4Y9mcV3HOjOT+s7vpaQ88Yg5JqOBcaybLww4oLFT4ZM3FAidhqsyjrVFkVTUEMWE3KqSaanCd+viqe2xMltUcNmXaBkRELI1dqySsg9Sq7tkPGka1gFgE7DRtOxMAgiDIIIPT8UlCEUdBjaFgMdzkWnRUpCDaKM/KKT2RNzJ9LHl7uXc++RtBtbbqYC01lNtwcBzq1l46fvR9jysp9d2xUIOXkcJJNv5mSknkohJxNwOoUcVcnQhlXb9tLJ2fpgHZd3LuOGufE6+Hj7kdxzlRVygZ6QSyATcgIpXNqyqi91YE7eskoJLx7pSx086QbWXNMGrB1Ybkx4uZaLWxZvyS8bVciNZVmdrGiJKuSWmkuYd45gzT8PAAhZNr4vdC+k7bfp8kqlDtoMuSgj148tNHF2tQdKBdCW1Twhp/5O9Dx0uHkYm0NicbW/KhFyecsqLLHUAQBcfjQ1Xq57mcWSBwtay2oyNuRiB2Hf4/Fx0KGWVWnsxAUOeTSa2fNihly87rb7eoUcAJxcauFnv+1G/MXfNtAfEsiDu/8Cnd4msHMN//7aabzKvYDDbAu3fqAPvG8dD1shcDI+/t3mD/B77h8D5xD/I2jecBxox2PXI8R9V6PuXLMttJP9dMdkyiWok5C7CuAkY2yec64+Gw3BGLsVMYH3aNHr9gmvAPBNiC2138c5vweIVX0AfpMxdhLA/wngvzDG/oJzPsmt0f8K4GYAPwfg+2EIudqgUmxMwdzBwKAUOvJjL5ogDaYH1KJvMdmyOtpkW85XAyCQOKqIgGmDYFl1rInVQqNCp3wtypHzpfDwKmCMoelY6Xfq+6E2WB8Q1ZRt1xYmpdNwE4CSllpCTmpa1UFHyPlR1fvQauQUcntgWaWlDa1G/P0nIe0BPRlCm1YpihRyjDEcnmvg4kY8YV7tegWEXPZdPnf5Ljz86FdgNW5D5B3XEt2cc2Hf2gnUhFxVhRwjhJyujOERoo5LsNnPxpOskNvyxdcnx2HLyYiTRCFXZFnVEbx0nK12szHMpGD5h1cfHouQC6IAvJWf5p3fPl/p7+l61BFyj65JllVaVqKwrFaJbgg5Fy2rzcSySq3Q2VgsUsg1hwo5m6jjbl2+VSCe0mWjGXJjWVbrU8gtNhbRsQ5jDfG28gkhp8pSjEsdsu8dhOoMOSBWZzXsBo7NDwk5ojgWFXLietUTctl3PdQ6hDNbsZ1WzpFTK+TEDDkAaOIYVEdgO1rSKkVTMjQUix0oOZ4o5BpWphiWCTmuUci5RKEYx0ZwzKOHhZ2zwFkLz9z4FO5d3k6pzcU7fxnYWYsVaN2V+P+gj+Ytp9C3431g8P6fR2d47HwRkKrcMPyoBxezY9EzB/rrkAY5xvYb83iyt4xVvoBPowkgf8xbecXv4+QNLwY6R4D2svZ9rzfUSch9BsAPDP/9eclrX4fY3vqxGj+/Lvzo8P+7EjJOwm8hJtBuQGxh/eA4H8IYeyFixd09AP4AMSFnUBNMqYPBrEI3CZ6GybHB3iEUihiYoH4Z2bJKFXJJZpJdzRY2LZDtn3ufIZetw8TKAwDrBQ2TAlGqyHbRoe3alQk5Omlt5RRyB3+700w4PSFXzca70fOFzKEEwUT3ZZGzf+2JQk6RITcJaQ/k2yUTHCWlDgmKGlYTHOpkhNxa18dNh9SvS/Yd1riKN331dxEhQvOGW9F76n/THpu6XpiStW3XxpafkVzUslqmkIstbCEYIcl0CjmaH5e+1svGU9NuYr5BFHISseeHEcC8lERwmJPaKallVSZ0tZZVqpAj5IdMCjy89jC+/eZvV75HER5YeUBQICW40B09Q44Slc9Yekb6+OzWWXihh4Ydk75iy6oigL4C6SxbVtNSB0KO8TDbV2S1mBeKRTJzDQdWKxs7qoZVQNeyOopldbKbZLJCrmUdSX/2SEejXOgAVLesAjEJ1bAbaY4cjQD4/9l78zjLzrpM/HnPcvdbe3VXV++dkM5KCFmERMkyQGQXiR8EYRRGZITR8SeIP0VFhZlREREYhMgiCgIGZQkBDEtIQpLODiSkl3R3eqmuverW3bezvPPHe8853/csd6mqTqcr9/l8+tN193vPec/yPudZ5FKH6LIMCkkhl/S+s7/RUwvNkAtaVhNsU+jn6BgIvV+8t2NZ9dZhvpF3LakMzM1w0ykh55Q6cA7UC4hZnoWTcYC3VunPN78DfH4fUFnEc0uLOBRfQpyZojrzM8C7APzriFBdqpwj8/i/h37POCXPOtjQD8Y9xWasPopvWHuxY9sOXHb+OUBqRBBqqVHED34OWHgIANB83T/iZZ8xUW1a0JKPIImvBN43N7QVGA9mKD7bsZ6E3McB/AqADzDGHuSc7/c/gTGmA/gTAL8FoUD7v+v4+euF61v/3x72IOd8mjH2BICLsUpCjjGmALgZIoHgHZxza7X5DH2EI+zqUL/UoY+NgBUqdzYAACAASURBVCgrQp9w7oOCjgfVlyHXayYaJV6cK/Z0cnI2EDOd4Ld/Pu0tq2QZjqbjmM6LiUK+1kYhtwrLKuAQMOJ9O5GzfuLmbCNiZYVceKlDt8qwYs0IDSpfc4ZcSVbI+RVOpwNhLauJNRS/AH4yxHuvMIVcu4ZVB3KOXLQK0RmHevZnsFuJZWpyCoAded5HiyKGUipKTY8scybmACRl2Hx1PpCnFlMVMK0IxrzPiSLkDoUo5KqGR1jF1FgHhZwt2VUHYqPud1lNqQPNr6KNszZkEo02LfaCfTP73L+t+haoCdEkPF3qUiFHCDlKOKb0FLZltuFU+RQsbuFA7gAuHb8UgDyuO5U6RFpWeXiGnKSQs6MVcobpt6xqHfPjANmyulaF3GqOycUmIaVjA0gwjwFvwFObRSnkqk2vRKFp2dIypGhYDWSRxaZs6/cSQo6WOix2kSHHOZcEGGNJr5DBT8jF1ODFpDCFXCqKkGODofcDgKo4llVvHc4Ra3ac6Xitci9GWREXF2p4vPU16/UV4EPnA5UlwDZwztgIfpIV+4BNlol5TYyzEesocLRleQYAH2VQIOMka9v+hwW0BJKMjP8LXgFktgPpMTy0oODmR4rI8SyuuOg8/PFNP4+D3/1NoFXq8vnqm/E54xy8Z+9eXHbtudLbxqduc/9uWA3ENB3VphV68QoIrpc+BNaNkOOc380Y+zsAvw/gQcbY7RC2VDDG/gbADgiyyzF0/wXn/PH1+vz1AGNsEwCHXm/XXb4fgpC7cJUf9T8AXAHgM5zzfZ2e3EfvCFXInQWThz766AQr0rL6NH+RPp7RoPs7RWFIriVDLqRlVboavwHyAPyNpWu17/UKSqCOZT1CbqWNQm41GXJAd7YtB5JlNaZKE8SzwapMyef2pQ6t57fJkCvWIxRya7Cs1s16oL3zdCvkDMt2z5E0hbljZ82W1Yi8PkqsOWjXsBr2unbbgftbst5pO1OaYFoBFh8LfQ19v6E0R5mL757SUpKdMK2nkdWzKBklNKwG8o2827oJiDFF8+OANgq5EEKuZnpjJ6bEJGumfxwYFpfsqmnVIx9SGlHI+RRxUeOJtqx6GXI2TB8hdyh3KPT1nUAJOSN/FdSJbwAQCrluiiKiSh0A4Pmbn49TZRFmddfUXS4h1zDDx6CDbqIIIksdIgi5YIacrBRNxVSo8Vn3vrCGVUAudViNQk5f4zHZr5CLwRvndbuDQo7ZbnYbIMZqlEIu0LQaYVldLAUz5PzjRr5YyDCa8BRyy7Vl6fVhx0i6P3cIuawWJOS4FUNcSQJGDagsAuVFoLLQ+nsBNxw7jHP1E7gPC/hW6zXz9/4t0CLXskYVH459QnznAvCnLTVbExwoeWOjSn7bhGm5hFwxhJit8jjybACTk9tw56IFQOyHsvFh4DV/DKTGgPSYULKlx4BYBvFv3gSsiFzHxg3vBUZEUlbuiTl8/6FHAAAj1iYYegpHCkfcz7LqkwDCozFoy6pQP7aUj2o4IbdcXw69/9mO9VTIgXP+bsbYLIC/APBa8tC74PG5VQB/wjn/+/X87HUCrQNpp6l2HtvS5jmhYIxtBfABAMsA/rDX14e83yMRDz2r8+jCJgp9BVEfGwGRLat9wrkPAjoe/Aq53i2r1P7aUsjRq/EbYOwFLavdkTTrBbq+xggRUWiXIddFg2gYegnv9yupzjrLaocJuv/+qHXNOReW1YHgJNPiqy91mCpNBe473S2rtRC7qv/vqAbKdohqWR3LBC2r7fLjHFBCLleJXsa2zcG0PNSknDauxBdgN0dDX0MLHbJp772pXdXB5vRmlPKCTJurzEmEnK4qUn4cEJ0hdyjEstowG25bp9+y6m9ZFQo5jzSJMy9/qZ1Cjt5OaSlXMSe1rDqWVSW4nE+WTqJiVAKkWDtUjAoeW/QaKs3iJUht+z6qZgU1s4aVxgpGEiNt3kEmNqlyEABu2H4Dbj16KwDgh1M/xO8+/3cByGOQjmfvvs4tqxaXSx2GEi2FHCFFLJtYVju0rMZUBiXuWVZ3Zc8J/VwpQ661HnordVibQs6fIadzb5xXLI9EoQo5BRpsCGVcvuGp6AwzvNQB8AjMMEJOLnWQlyvn4nfFNHoMko/bdNtcacgZclrYRStCGmUWjwCLJ/CL5UdwMKWhrnmKv0mrjq81fw34X+G/6XkAnqcCeZ7Ft1pE5qzqjZE0Ob7rELZSizFYjMFo3YdYBgVCrBvGMJAQ43B/9hLgxt8B0qNoxEbwvA89ihoSiGkKnvytl+G97/8EYvgHAMDg4E7gsjeFfk9K+lLL+lBSblk9kj8C0xa/P6VsQqllxQ0jNf2EnPMcSnYOxYeQb4j9ZF8hF451JeQAgHP+IcbYZyHsq1dDkFYKgHkA+wDcwjlfXO/PXSfQo007z4Czp+58RhHERwFkAbyNc96niU8TwsiJvmW1j42AKFVKn5Drg0Ii5BS2ppIC/1Vo+j+wQRRypqwCpITC06GQo6Qntfi1K3WIyuzqhF6UULRlNaErZ59llUy6IxVyXeQF1gxLTAbDSh3WkCHnb1gF5HbJ04Gw/DigO/VQO0S1rIZbVjsr5IZTlJCLViGaNpfUcQ6U+Dzs+gWhr6HbVSrRRItTkOyqDjanN+NIXqhF5ipzuGDUe89uFXJL5UaAYABEwL1DyPktq/73MUzZsqpyz0InKeTMqqQkokqz0eQoqiUxhaFEnUNQsoiWxydXnsRlmy4LfSwMD889DJOLhWrVt4BbGYwnt+BESSzHmfJMR0KOEtPUsgoAL5x8IWJKDE27iSP5I5gqTWF7dntHAj4Z66ZlFZ0tq1YbQs5HEs1WZ92iDG6mkFLDwxClDLlVWFY1NZyo6gZ1s+4SZZqiIakloXKP8K1YgkSpGBUs1Bbc543H9mC2LhRXRYMQcpYNkH2lxjR3PLgKuYyTIUdb3AUtYdk8dJtvmJa0Hzd9ynY6pnLFU8DJB4DyHFBewJ6pY/g/2n6MszzGWQFjrIA3qQkstdoMsl99O2BZ+DUA34ptxuOaRzRtsZtIt6UFBLLk/HxO8ygWXU3iVuuFWOKD2LRlO2LqHai12rkb73wA+tBOQE/i1FfeCFSFefAnzcsRw48AAMvxIeC8lwIQJQoNth/g4hyg2jRhsiqcveVgyEUFB9QWTQn5IbKvzdcMyaYet7a7f2dCcj8lQs5seNsdIeR2DOxAfrFPyLXDuhNyAMA5XwHwj61/px2MsT8D8GerfPlfc87fu57fJwqMsVdClDfcD+Az6/GenPPLIz7rEQDPX4/POBsRdjDqK+T62AiImgRHtcn18ewEHQ+qsjYLphlijaQtqzYXFzwU5ezNQqUqP79CbjVqoV5BmyJHiaKonVXPCLESd4PVWlZFqQNVYTzzidjuFHKd8xULrSy/MNWHjTUQcsUgIXe6FXJ10oaYjHnLhBKTvapoAXlZxzpYVrtTyHmqjbYKOR5FyC3AjDguUoVcMt5wCTnasOqANq3OV+Vih5gWVMiFEXLUrnr+RBYHW2o5C4bDxwVaVgOEnM2h6J5CDqb3XXVVh6ZoMG0TNrfRtJvuRJmOp5HEiKvKpISca1mNIOQO5g72RMjtmyX5cRWRmTaemHAJuenyNC4eu7jte7RTyKX0FF4w+QLcfepuAMAPT/4Q//Wi/xppm3bQvWXVIzE9yypRyBFCLmBZ9e1znswd8F7XmEDVsBFGyUkZci3Lai/7dalltceLJf78OMYYVNsj5EqmKBugdtXt2e2I2aOYbfEuJcPbDpq+UoeR5AgWqgutxzpbVleqTfh/ggIbzcIssJIHygtAeQ7q8gzepz2McVbAFhTBv1VzpTK543cB+27xvi+AN/hYj7Kyzfvd5Bi8zTTxOLzj8KhzrFN0ID0uLKCZTUB6E5AZx4MLKr74RB1Hm0UAYkzOpgaA1jbWzF6A3zVEZ+SvbNqGhPUgai0FXyMzjowuyFiTeySWbXiEu0l6XxljSMc0lFrFFAvFBpgSLCEJg588czCUkhVyB8iYXcmPu39ftiPYihqpkCNtzTuzO13FrN9K3IfAuhFyjLE9nPOnenzNr3HO/3UdPl6BV9jbK+jr6GXJpP+JBM6lmvCgiBAwxtIQxRcWRJFDf/Z8GtFXyPWxURFVZ3+mh/dCqY7D82W8YM+oRNb0cWbgz5BLxFZPMFHixbkKzxiDrjJXPWfYNuLKag/DZx6GT222lhIMCs45PnvvceQqDbz92nMwkNBDnyeXOshXq6PQtGQSsVv0Yl+mvz15NpY6RJBEFN1YVl1CLiQXZy2W1RPFE4H7nlbLKiEoktKYXxshR5dpXFORjXsTyG4aVgFghLSz0hIGPwyUoaaOB+5XYwtoRIxR2l4cizXcs/8wQo42rfrz/mJqUCEXZlmlDauXbB3ESrWJ+WLDVUKJ94rJLav+DDlTtqwaTR9JpaVcYqVqVN2JsqSQIxlb1LLWSSHXa44czY8zKyIEfjzhpfzMlDs3rbbLkAOA67df7xFyUy1CjtqmQy2rnRVyJjdcMkGB4qomqSrcaqeQM+ULWPvzh93bdmMClYaJMFBCzs2Q60UhtwbVeqHh2VUdQsc20uBcBWMWalYJdbMu2VV3DexCpepNlSump5BrWhZYzCPkxpPjLiEnW1Y5FHjbgP7U3cDhu8Hmp/BB7XGMswLGWR6bWB4jKEL9hLw9ZwC8hexKnsprQEbkneU6HBMNAPXW8lU5kNx6OZDejAcXFcw25wF42/o9zRfgzWNvxuffeSMQkn14aN9xfP3xJ6CaR5FqEXKU8NaYt241VUGcxYHWpkYJXcP2XsNNQshxeV+QJvvT+WJdUnSG7cMcRCnkBolltVBr4sCyR8hVy+KCxFgmjt1jwe0wkCGnBS2rOwd2un/3FXLhWE+F3PcZY9dwzmc7PxVgjL0FQkG3ZkKOc/7nAP58re8DOTduEkBU6cRk6/+ufmsL74EotvgUgMOMMf/lQWfvrpHHKn3ibnUIy084GyYPffTRCc/EDLlKw8SLP3QXinUT/+P6c/HuG/eese/Sh4CcIedTBjR7JeQI8UMmCJqiwLDEe5kWRxdz7Gcs/Fk08S6yhrrBjw4v4f23idL5dFzDO647N/R5VNHoKgfQIUPODCoXu0Gyhwy5oEKOTPps3lU4+5lElI2SQibkwpeHux5CLKvWOltWo1ox1wv+og4Ha29ZpWSIPB5HMzF3AtlNwyoADFOFXBulqJ14wrW9Taa3YaYisuSU+AIAHqrepe+n6d7EcVUKOS2okPNvF7Rhde9EFseWKpgvNgDmkTMBy2pohpz3Po26vByTWtIl5GpmDcMtHRYdT6NJj5CjhMFKCCFHbYa9EHJzlTk8VWjpM7gGq7obADCW8JbjdLlz0yr93mGE3HXbr8Nf7vtLcHA8uvAo8vV8R4Wcn3wPGxsWIT8y+gAUJl6jSQo50rLqa1n2Z3s+2QrRB9oTclKGnGNZXaVCrtcMOb9CDgAMG+BGFiwmxvdCdQHHCsfc5+0e3I1ThIysWaTUwayCxcX+IMY0DNS9sdy84/1ArYYtpXnsj8/i9coIjosUNWjf/RPAMDEK4FdWcT4xQpb9iqoBWy8HMpuBzCYs8EF89IEiFvgQlvgg5lUdwM0AgExiEOxXfwAA+PatT2DfE19BEv/hvlfemEAzORhKxgG0ZTVcy6Mz735dZYKQa4ESYw1CyNnGELlfJuRScW9fPV9qSGrEbhVylJBP6CoSuoK6YcOwLBxa8bZ3u1XocNXu4dBjfZ+QWx+s5+nzLgDfY4xd2ykbjTH2dqCVPvgMAud8kTG2BNEEexGA2yOe6rSr7u/h7Z3R+LbWvyj8WusfAOwGcLyHz+ijhTBy4izIn+5jlfjC/Sdw609m8I7rz8F1e8MryzcKogi5M2lZ/dl0AcW6OMm89+gS3o0+IXemEciQkyyYvU22aW4hnSBoKoNzcTtKuXm2QMr90RSJvFkNOeHg6KI3ETm+FJ0NJmXIpXu3rPZW6tBDhpyPvGGMQVWYO74sm/c0aXy6QddrpGW1iwKP9pbV9VXIne6W1cgMuR6I2jDIZIhMfo5m4ji+LAiWbhpWAX+pQ/R2wFOPu61xrz33dfiX/Z9FySiBqXUwrQSLcyiQx+gKscAqGiHkIkodHPgVcnqIQo6Do2bWpNwzWuiwdyKLJ2aKePhEDkwhhJwiK+T8ltWmr9ShXJFz1ejnUbItipBzJuQN03LJUlX1lvOFoxfisSVhMzucPwzTNqUWzCjcP3u/+3fCOgclLsiWEULIdaOQo4RkGCE3lhzDJeOX4LHFx2BzG3dP342G6WWIhZU6KApDTFNcFVvDtCVSGgAM5n1uNuaRG7qU0abAoYvbZcjFNAWHVzyFnFWfQKURvm3JLatOqcPqMuRCi+1sC4wxl2CkCFPINU1bqLRahNx8dV5WyNWqUEsesXpe9Q7gUzcApTl8tLKIl6XF+h406ogvHADSYnw2Tj0AVGtgAFIMMAjJo3c4jzUTw9AGtgi7aGYzVpRhfPyhEhb4EPSBCXzwLS+GevsbYXELJYWh+db/REwV+5H8fAlfuO9u970YW3KD4LO6t0+Kawq4IecbcjPb9hjrLHtupcIfpwo5RZEIOTp+mra3L6IKubotk/zpmLcdLhTrXRNyVCHnt1oPJWOYM+pgsWV3/6FjANwUy+aqXeGZj87ydd7T3U6Imnz7gJdDt1Jfgc3t0HH4bMZ6EnLfAvAKAP/JGLuBcx6sEwLAGPsdAH8P0br6B+v4+euFH0IUUrwEwN/5H2y1pF7UuvmDp/F79dEDwrJt+pbVjYlyw8Rf3rYfTdNG4dvGhifkrGdgqQO9GttXoj4zYPky5JJrypCjLavk5FlqWn3m54m1g+ELh050EfTfDSiZ1y7Xh243UqlDzYhUoa02Q64X4kUudRCv0wghZ9ocEcKzZwSo9bIry2qnDLkQhdxqCbmaWXNtXBrTYMOGzW0YtgHDMqCr4fbmtSK6ZXVtqtB2eX3jJBexm/w4QCbkoiyrVaMKJD0F0g3bb8A9M3fip4s/BQAosfnQi1WU6ObMW6ehCrl0tELO4FUpK8lB2Si7BBnnHE/OeyTP3s1ZbB9JQSTYCGiKBlVRJeLJr5AzLQ6FEHKFkqzGoQoranumf0uW1ZYyh5KT6aTlxOlha2Yr5qpzWKguoGE1cKJ4AucMhTeEUlC7asryCjBG4x6x2Q0h10khBwjbqpNNdefUnWiYv+Q+FkXAJ3XVJeTqhhUg5CySRjRICDmpZdVUowk5sh1wNCXS3W5sRqXZjWXVAGBLimRAEPiDsUG3+ZVCzves4+G5h3Fo5RCeXHkSh3KHcCR/BApT8LEbPoaf2/Jz0muLDaKQa1SBn3wRL88/gFoiB2frmv+Pt+J4zABi4pfv/tHfI6WqwGaRMaZgEZgWWXOVmLfvGrBtxMg22PAdyyRC7twXA9nteHhZx78fMrDAh7DIh7DIB7GMQfz7f7sWl273fvv8XBGfvl8UH+yNZ6FsvhDDiWEs1cT3yNVz7vbrJzdp/AAtc4lpSqCd2baykeMJ8OzC3E6EPq5KllWGOA8q1TjnaFjUspoGtzUwxYTFTdSturuNp8iYXfAp5NpaViNaVgGRIzdXrENNeNumVZ8EWhczrtod3lhNx23TaiLWOiGgCrnRxCiyehYlowSLWyg2iqFj+NmM9STkboJQlL0IwG2MsRs559LaZoy9G8BfQ6zd/8k5/9g6fv564YsQhNxLGWOXcs5/6nv89yG+/ywEedcVOOe/AeA3oh5njN0J4FoA/9x6bh9rQNjEp1/qsDGRrzbdE6CFUjDfZ6MhyopwJhVylBjo1SrRx+kB5cf8Crlaz5ZVQvxIllWaWXN2r3f/b0x0EfTfDSjh1Y6spssvE9cQ1xQ0TBtN00bNsJCKBU/XVpsh10upAyVmkoSQcygIw7JDlSjPFKybZbVNhtxqCTla6LA1uxW5Wg4lQ1zLrppVDKrRSofVompUcYp8LiXq12pZ9YfZU/zqVdtx15OL2DwQx8su2eJ/aShoy+pKtRlqL7x35l5XZWbVN2P30C6cM3SOR8jFFxB2rYAq7izmKRI7WlYr8xJBvtJYCP3u5WYZm1Li4uB0voZyS4E2lNIxno1j+3AS8KnjALQtdagaHvnHuYpyLYa6YbnrjTatSgo5o71ldZm0WaYTJhytVEpP4fyR813S+GDuYEdCzua2pJBL2x4hNxIjCrnKTEe7eyeFHCAI2I88+hEAwD3T9yBlvsx9LIpASegKCq3FE0Y8W0QhNxAPV8iZdptSB3IsWayfhMXFZ9jNEYDHIy2rClMQV+Pe+zFTugD2nWPfwXvufg80RcM7Ln0H3nLxWyTFovh+NvTh+7E0ejvecnt4HuC/3fsB/Fz6IqA0C5TmgNIsCigCw4KUGjzyQ+DBr+KNAKZjQ3gSYpuYa+ZxIuVtH7sMEyZZfznVWyZFcp6Q0bKIpwcAY1Esn6veBuy8Echsxm9+9QTm8CdQWiSo9pp/AJJj+N53DuDL+4Ox9H4Fs2EGLxaOJEYiCDl5rFG1MyXk4poCbg64+XkAwM1MB4Vc6zE7DkERyMd6DVQhxxCHbPN0/ucQv4/bGgAV3E6CKeKYUGwUXUKOtp0u+DLkui518I1bJ0dOiXuEXK2VHzeQ0LB3IlzZTBVydauOmJMzTI6VaT2NkeSIe3zL1XN9Qs6HddMLcs4bAF4J4BEAPw/gq4wxd8Qwxt4LQcYBwG8/Q8k4APgGgAcgls3XGGMvAADGWJwx9i4Av9d63vs4DwaHMMaOM8Y4Y+xzT9cX7iOIUMtqX7mzIUEnAWc7KdANIltWz6BAiZ4kRSn4+nh6QS0rClurZZUQP+SkVM6sObvXu0TIaawnW2c70Alfu/2T32Lsbz0Lgz/3rlskemjTrEsKuZAspWf4cbW7UofOllXHkh9mWeWrJeRIftyO7A4k9XCF03qh0CjgVV97Ff5m/1ugD98DQCZnk2sg7QGZzPQv6+v2bsIjf/pi3PGu66TJZDvoqoJsQjzX5kCxHlzOPzjpGVXM8kVQGcOewT3ufUp8IfRiLN2mDE4C0UMsqyk95U7Ym3ZTykDKNeYDzwdkMo02rO7dnAVjDNtHUmAkP86ZKMfVuEuyGLYhqa8W615sNTcGATC5LZYq5Ii6jFqgRxKe7cwh5OTGWe87pfU09g578RPd5Mgdyh1yl89QfAgp7PB+o5J2CceaWcNKYyX0PQDAsAw3m01hiqTsodg9uNvNp6qZNVSUg97nRVwo6EQ824SgHYx5pIFj1wccwkSgnUJuunrc/dtqCHKj0mbb8hc7UNXb9058DwBg2iY++uOP4k3f/jUcPbUPOPUwsP9WrDzyITxn1weQmLg1si0XAKZzh4AHPgHs/zowdT+QP4EivO9M20Y3kW36sXjMLUEYsoGhnb+A+I4Xu48f1kaBt94O/M+f4v9LvMe9Pz5xJWK7XuQtny3PBXZdA4ydi8zAiLQd6Io47i2VwhWxTT8hZwePgcMJr8N2pb4SeNwBJYwoES72XYqbncZtHXZzBLE2x1jv4iSDhqBtVYGv1EELEmN0m+W2ILloJh3N+UuRfeh8sSEdl9op5OjnUtIe8JpWqULOWQZX7BqJLGuj26ZQyCkALLeYhIEJQo7se5br/aZVP9bVwMs5LwO4EcDB1v9fZALvB/B+CMr4NznnN6/n564nWiUKNwE4BpHhto8xVoJoVP1biGX2Sc75p87ct+yjE/qW1WcP6NXIs9021w2iiI8zqQCl3+nZQIqeDbB9ltW15ENJllVyUqb6Av7PZvib8RJdkDTdoCFZVqPfh26/msIwlCS21ShCro0iqR1WW+pAFXLud1jj9m7bHEcWSqft+NzORune34U9uVgzAFih9kTO1q6Q2zGwQ1IBnY4cuTtO3oGFmlA76cMPAJBLHbpZDu3QSY2YimkBhVsYpkpTblYVta0uV/zh+QbunvIyoazSRVAUJqm4lPh8KGlMLasNiyiiIiazm1Oe3ZLaVpfqnQm5Q3PErtpSmWwfSQUKHQBB+lByoNT0yLz5ujde7KawCS6Xvd/RVYZcSMsqJeRiOrGv6mmcP3K+e/tgziO7ovC1I19z/37h5AuhMm8c2AAmM5Pu7Xa2VX/DapSSjjGG67df795uxh5z/45UyGntiWeqmByMyyoed9/HvQsm7UodTpaPun/b9RYhF6GQA2RSdVKZxcD8A8BjXwHu+XsszTwsPfeJ5f34le+/DZ/+ymtx83ffibfO3IK5pLeutxomXl0q4w+WV/DBhSX3/hktSIgXCNk0OLQLuORX8G/6L+HextXu/Q8Pjrl/79r8PODXb0X8JR9276soBrDjBcDwLtSZt58ciA8GcsYcjGfjACMZtS0yerHsPYfuMvz7JbnYyFPIOaDkeYCQU8Itq86+qz77y7AKV6I+/auAnWqvkKPnRQiqOVVCyOkKQ1wJEnLSPr9FyIEQcnRfkCb77flSXSobaqeQS6re+4VlyAEcSsCyCly1Ozw/DghTyCkSIZzW01CYIu17+oRcEOveicY5zzHGXgLgHgCvg2gqvQAiLOGtnPPPr/dnrjc456cYY88D8IcAfhmisKIE4CcAPsE5/8oZ/Hp9dIHwUoeze8LYRzjoRPrZYJeMUqSc2Qy57nKy+nj6QHlbTZEz0Xol5MKuQgNyntzZroyk4zamKmsOuA97bbttlD6mBBRyEWqBVSvkyIS0k0KOEIoOeSOv97Vt72/7l4fxg4MLeMVzt+Djb3z+mt4rDI0uFHLd2JMLNUMKqaZYL4XcT7SfuLdPR9PqfTP3uX+r8UUwtSwRZ2tvWSWEnL666/1PLD+BX73tVwEAn3npZzCSjuFEqwxipdIExr3nPjT3kGuBsptDYM2tACAp5NTYAizfRSzOufRdy2awYdKPifQEjuSPABDFDheOim63pVo4IUcn11Qh5xRaSPczPwAAIABJREFUTAwkoGvetqcr3qQ2o2eQb+Td93FspkuNKfc5dkMsiCViN+0qQ85nWeWcS6SerhtA66afkDu0cqitzbTcLOMbR77h3r7pOTfh7w95y9m2BSHntI5Ol6dx8djFoe9VMWVCrh2u3349PvfE5wAAVvIJAK8GoERa6TtlJXLF++whH7mhqyJKAFy2rNLlQs9JTxSPuH/bjQkAHGYlDywcAIoz4l9pFihOA8UZJJoLrlTm84n3Y/dtHnm3tG0LoIvjAuMcnDEYjOEjIz7SkHP8Zr6It+ULiAGAloQ9MIE/5hYMBuRVFdX/8mdIDe4AshNAdguKj/8DcFIo8AZ+4Q+Ac16Fj/7VHZhrHEAKBwAAJbJP2jW4CwAwnhryrJ1KA3WzjoSWgAlKag5IVkmqKBzPxoFFuh04CjlvXE8MJDBTcApIfAq5kNgGSvxQQi5QPtQmQw4A7MYWVGdeF7g/DPS9VZ6Cr0MGCvcp5EIIObqdcjve+t97naSQIxEWi8UG2BBR+XapkAsQcikdTCtC0Vrrzo675RbtCDn/us2qiqw+bBXVSERprd+06se6E3IAwDmfYYy9GMCPIBpJDQBv5pzfcjo+73SAc14E8N7Wv15et2uVn3fdal7XRzjCiJm+Qm5joinZJXnHXJKzHVGEF3/GKOTObmJmo0AiePyW1R4IJrFNib8Zk1VxuhQifXbvXw2fQo6qK9aLkGu3jOh2o/kJudr6WlbjPRAvlLBzyBs5PHz123utaeEHB4Vi61uPzeKvftlANrG+RQZUUdGNQi7Kzl2oGZItiIGBOzlBLFrx0g407H3nwM7TqpCzbAv7ZvdJ96mp40jGLnVvJ7S1kdANo/Oy7oQ7p+50//7h1A8xkrrBve1vWg3YVVvjciI9IRQmShNMqyJXz2Ek4+XW+e3hVHkSZlkFohVyc1ViI+UKWEvtQ/PPDvoaVp3P3TSgunltjEzHoppWl5uEkGuKfDpJIReSIWfZllve4FjHdEWHYRvg4GjaTUktqGpNl5BL6Slsy25DSkuhaorluFRbwniKsKIEtx691SWSzxk8B1dOXAlVedB93OIcWzNb3dvdKuSoYjAMl45fiuH4MFYaK2BaCUpiGnZ9e1etymHjnFpWh/wKOZd4UaAxDSYX275pm9AVDajnscs8hucpi5hky7hvySPZP2v/My6P55B+sAE8iFAkJyeAuCBna77z2CWS0fbPs/P4q9Fh7I/HpeeM1dI4b/4y/Njeg9g7XwUMbAHiA1AYw5avvsK9CDBz8Wtw7vC57uuKZjCDrGnZsI1wtdWugV3i+8Y0cDMNpguyKFfPYTIzCQv0/QagqaTUgRBBY5mYm9MGeAq5JaKQ2zqcJIScTyEXcrEwyhrpt5xGZchFWVNjajjBK7639xqFJwOEHCOEnK4yJBTSdmqK3ypZSB1CLsKymol736XUMJChpQ4R+zCgfanDYEqHkvBac836FghiW8HFk93n0sU0RVIfOtvvSDJcudiHwKoIOcbYG7t86r9AqMy+DkALex3n/Iur+Q599NEOYWqNvkJuYyKQKWFxxLQNTMiRCTBjcMmSM6mQo+HyRp/4fkagvWW1exIlqtABkK8K+63KU8Up/M3Df4NdA7vw+5f//jOeJPeTW3R5rcWySpd1two5tUvLatOk6oDuly+1rDY6KuSIZTVEIbcWRay/bfDQXAlX7Iq+Er8aNHstdYhYHoWaIV31H0mMeJM9ZsCyeWTGThT8llVKqKy3Qu5A7gAKjYJ0n5o8Lo0FXWVQmMhrM20Ow7J7Inq7KdDoBJr5VGwWMUwsq5SQs7mNH055vWpm6SLEW8tfYQqYuRk8JgisY8WncO6YR8jRMaspTG6YbKOQczBXmQv9226MQ00Iss4h0kzLxtEFj1RzFHIAsHnQI+S47RHRUU2rK81T5LMcQi5CIdcaP3QcJbWkyGPTEjCaYp9SN+uSFVghluy0Jqxm5w2fh58sCmLpYO5gKCHHOceXDn7Jvf2G898AxhgUst+3bY7JtGdZnS5PIwqUkKNW3DCoiooXbXsRvnFUqPO07H6Yje2BhlIH8v4vZN+uRFhWjRr2sDkklRlMsmXcDbiNtI1PXA29OAM0y/iaAiAGrCgKXmRvAwDEbRvXWLPotJtOcO/7TLEhnDuxE/GRrahmNqO2+J8AgBjT8Ly3P4gvZDbhnw7+K27+6c1QFRVvv+QdeP8XR3EMrQtKm86X3nsyM+kRcpUZmZBrBreBpmmDm1lwzsCYvJ93FHJxTQG3MkCLkFuoLmEyMwmbVdxMrOH4EJokp5Eq5EbT5AIMV6AwBbbNpTE5OZQEIPYL/vUVZlldTYacZFmNUPe2U/3Ssabw4HhlPCY912/zBPwKuZAMuUZ4hhyUpnshQGdxiSDzg2YUOp/rYCgZC82Pe/6O4bbqQImQMxvQVSXUDhxlJe5DYLUKuS/AXyHSHje1/oWhT8j1se4Iy7HqC3c2JhqWn5Cz2x48znbQiXtcU9wJ/xnNkDO7Ix36ePpg+iyQveSGRb2P3/JBw/39+Y2f+dlnXLXLtduuxRUTV3T9mQCwVFtCxai4gd2nG1JjqbY2iy+FVOrQxtYrkwQKhshEZSXCskpJxF72eT2VOnTIkFuLVdmf33RgtrjuhFw3ltVuSh0EIeepEMZT4y4hxxQDhmVDVbonoapGFYs10TqoKRq2pLdIxMN6lzpQu6oDNXVcGguMif2EEzpfN6yeCLl2LavdwrFqAmICuoUScmQ7OJA74C4/20zDqu6CmiCTYmMzrBYhd7x4DMA17mMy+c5CyQg/ohRysxVPIWc3JgKE3PHlqmst3zKYcJsMAWA0qwKtIWWT1s6s7pEDzvvY3EbB9Agsy8mQiyh1cNQ2dBw54yupJV1VYM2sIUdUdpzJ2U8AsHdkr0vIHVo5hF/Y9gvw4/7Z+3G8eByAUMS86pxXAZAV1Za9OoVcWmtvWQWA63dc7xFymf1Q8y+PvAgUum+3DGEdLUxjSJ2DMwonHvgY8IP3A4VpoJbDVwGgNSRfZG5FtaWaauSOIOPbFx6Oeet6j2HCWcNNFkdseBswMOn9y4r/k0f+Fcj9DADwTvMduOcNv40tg0ksF6eArwlCbiy1CWxkN3QAv/Xc38KbLngTVEWFCh3vx3cAhF8oocveT4ZSst5RWAk1mgZupcE0ufF398BuAGKfwSxvvM6Xl4BNAGfeuBtKDKJA9vVUITecoftM8Xe+ZrjnkdmE5ha7AHJMg/93hinkKPGjKky6iB2m4gLaKeS6aFkFwHgy+ASfZdVfhADINm3eIvC43TlDjir9klp4E6oDP3lGMZTSocS9fVs3+XFA0AYbjyuSHdhVyPUJubZYLSF3H3oj5Pro42lF2MHI7ivkNiT8CrmNXirgz7pyCLkzObzpBOdsb9vcKKAWfZWxnkgYCr+VkkJXohVy9ISfTmC7wUx5Bq/++qvRsBr40LUfwkt3vbSn168GErmlKhJJs5o8Le+13bWs+hWNVCFXWGfL6tpLHdbHquxXyB0g1r71QlelDlQhF1Fm4Les0iwuMLNnpSDNj9uW2QZN0U4rIXfv9L2B+5TEDFRVJnsTEiFnIxtebhmKbuzBnZCvE0KuWcRFhJBbIeQTJXPs2g4Aihysbk3A+TbHi09Jn0G3Q1VrwuLimQk1ISlXKMIUcqZtYqG64H2Plo0UACpNMbkOy49zMJZVXELONL1tMh0jCrkWITdTnoHVyiq0zTRgiedQa1/Y+KEKOYdg8xN3VHlowZtIO+/XTbEDVce95tzXuK+lCjmLc+wgpQ5P5afwke8fxi8/f6souSCgVl1q4Y3CC7e8EDEljqbdEKToto/jq4eruHHXjUirSaC6BBSmgMI0biw9gMu1Y5hkS7jqjjrw3UWgPAe01GlDW7cgD0GmjZ7YBxjhlvQY2Wcbrd/J9RSeagximo/hVk0DIEj7seErceOxazDLR3DF3t347FuuCn3PxOz3gBZXwRTD3dcu1b1ShrHkmPQaZ1nT2JKw+BZaqDFb9ohkQFbIDcZaltXWvpMbgwAh5FSmYnt2u3ebZ11CYKG6BM45ONlXjiQHUTW97UAi5NJkP9HK5aNjejwTly+Y+BVyIcfAdllluqp4cwY1vJk0qp232ww52CGKTjsGtJpsdZVJJFaYQg5We8sqzZBjKtnGOxBykmU1oJDTwTRvPXFDKA2v6nCRLNyy6q3DMEJuudYvdfBjVYQc5/zn1/uL9NHHeiJMidBX7mxM+Ak5/xW0jQY6maAnCM+UUof+dvbMgOWzZa3eshq8Au2+r2RZld+TWq2oRaUb7JvZ5560f/Opbz7thFygBGMVjZPua6WW1TYZcj7L6jDJkFupdFbIna5SB6pic5bJepU6VBpBhdx6o9ljhlxbhVyGKOSSnm2PKSaahgXEuz+l9ufHAbISaD0tq+VmGY8tkvZJNoIGz4ExG0vGEYjeNYG1lJmsi2W14VnMCo0CRsaoZdWQHnPALTEBpmos1fQsqidKx6TPoGpeTWs4kWltw9ApITdfERcYlmpLLplnmxlwM5j9FpYf52A4w4AWn9c0vTFI1TrOfvSpgkcqOnZVQM6Q66iQa1mi/TlSy6QYwuTeGHcIPKnYIXcIfkyXp3HXqbvc26/f+3r3b7pbsm0ukULTpRl8+OFDuP+pZXzpt14gvWfY946EZSJVXsQLBvbi7nxrnCeO4333vQ9/de/78JJKDW8sFHBRUyyrXwa82e9K8O3ytHGUHtcUDbN8BCetEczwUWixZcAW37Pxxi8DW65ARcniv/z5dwEA6fhXobQIuc2br8a3+Q4AcAnvMFBLIZjh2jCXah4hR7O4KBhj0FXmHrP98S1b0t42QS+Y2dyWVaLxAZiWDWfXzs0BAN7zt2a2Qle945OOAXcbWqguwbC4pCYejA9isbro3jZsbzsm151g2wpqTUsqdBjLxNteMKHn/85xKarUARAXEZ3vGtWyGqWEa3eRgV4MYHZQIcd4HA77rilKqFKN7vM5b0/IpeOUkPOWdUaP3oeJ30CIwJAMOUrucSsFTWG4bMcw2sFPyOkRpQ70IlZfIRfEaSl16KOPM42wiU+fKNiYCCjkzvK2x06gv49Oes6kZZWug42uUDxbQMeD4iPkOuWGUdDx5res6pJlVV7vTvshELwS2wk03PjIypE2z1w/SOSW5suQWyeFXNQxyCbFGQCgMHRV6iDZbHvIkJMtW+1/G33cWSaSVXmNpQ4Uh+ZKsG0OpccstnbohiRqp8AAxDpsmjZiZAI3GBsUio5WIHnVrGME0dk9ftD8OEdtQhVO61nq8NDcQ274/Pkj56OwMolZ3AEAmKnvB/Aq97kyOSmvn1OlU/jeie/h2u3XSk2m3vPX3rJKFXKFZsGXIedN1GVCThBHUuGMPeGIz3DSR8jJCrnuwtD9llWb25JdlRtDUiOiQ8g9SQk5n0JuKOV933oznJBzxsGxgvcbqBKPkmlhGYR0kk8tqw78CrmmHSTkzh06V2R7cRsniidQNarSWL3l0C2wW+qyqyevxu7B3e5jkmWVcwzEBpDRM2L5KE0wtYJHTqiBDEZ6QSejJoClw0D+pFC55afkv0szALfxp6qKvx4dxp2pJMyWMqzGgFszSXwzncCXZuZwUTOqEZkBmc2wBidRUDzyK/NLnwJGdgGD24D0JvzGR+7BoZbq8cLUzUBZrJfmyG4gNYImzeOLe/mC5w4+B05bhl8ZTEHXDVOa7r6WEnJ+hRyFpigwLLHd+uNbouzCFaPirr+UloKu6Kg0vO+oWHKxhZMf5yDGPEJuuZaDYdkSSTQQG5CUp1QhZ4PuY1QslRtYJAq5sWysbaSA3zECyIQlJfgBcWxH67gTRhoB0fuutgo5mq9rBQk5sW9oEXIqC5BYgE8h18qQQ0TLapqUOlASbbDNPgyQCd9gy2osQMg9d9ugmx0bBX/LaqDUwSHk2hClffQJuT42ICzf5MZB37K6MRHIlNjghBCd1NMThDPZIkyJgY1OiJ4tsNbNsipnm1FoSrRCjuad9KqQowTeqfKpwATwdEAuSFC6sjF2AzqBiCKv/CHzjDEMUstqRKkDzW5sl2/jRy8qqNBSB2WdFHK+iWm1aeFkropdY50zo7pFs6sMufbruugQov4mO657hJzRCLyuHahl1VXI6adHIXfvjGdXfeHkC3HXsg4wQchNVZ+QnptoY9V+113vwv7l/bjl0C247bW3BTLz1tqyyjmXJtDFRhHDKW+akqtGKeTEBFgl9jzNHgW3NTDFxEpjGYVGwW2PpNuhooXb1vxI6SkMxAZQbBZh2AZy9Zxk+7ONIXDLm5i6hNx8tEIum/S+b63BXIshJQecCxuyQs5TZ+Y6KOSkcgRHIUcm5RWjJhH+dStocU1oCewe2I2jhaPg4DicP4xLx0U7b92s4z8O/4f7mjec/wbpN0qWVVv8vsnMJJ5ceRIAwPQVxOsKlo48gs18sUW2nUBlyashTT34j8DtH0QnTFgWPrywhGVFwW2ZNL6eTeNIrJXFxRi+uWknLkqdi0fzKXx/Wsc0H8V1V12G1157lchw02Io1fPg/yYy8riVQPyS14n2rBboRSmNefto5xjnjS0b0D1Cbu/IXgCPAwCqjeh9Ll2HUJruvrZrQk5laDmbA+fCVJ04U/EIubD8OLrfVPmwlFHlNKw6iCsDcOjT5dpyi5CTW1v9pI0DqpYDV7FQamCJjOnxTLytgllStrfWTUpLIabE0LSbqJk16fyBXkTsvWW1O8sqD1HIcYsoCn0Zcg4xRrfVjqUOsXCFnLOPi0I7y+pgQgsQcld2yI8DZEKubtXFcqLqw1YmZjaWdZuJy0ZZ5M21KaB4tqFPyPWx4RBFCPQVchsTzzbLKrUQ0hOEMzm86UmRzbHuKpc+egfd3ykKQ0xV3EBj0+YwLVtSOkXBH4BOISulfAq5ZnhmTDfwWymO5o/ikvFLenqPXkGPGzFVkXJk1itDLlIh58uPA4DhHksd9J5KHVaXIZcIKXVYS4acXyEHCNvqehJy3WXItS91cDL86MQnG8uCcR28lblVbfamAvU3rAK+lsx1yJCzbY6mZWPfzD73vmsmr8G+R5fdYPoTlQMwbAO6IsYbVUPQdV8369i/vB+AIMlXGisBYoAee1dTrFQ1q9IEvWk3kSFzW2rdLjRDLKtk/6QqKuzmONSEIM2eKjyFyzZdBkAesypRybQj5ABgc3qzq1KZr84HFHJUzVJpVlA3LBxfFpNshQHnbpKz0HTNW76GpWKlamAkHZMVcs32CrmlStMl8kIz5Ixwgs3BcqXsXsAeTOqRVtG9I3txtHAUAPCuO9+F1+99PV533utw19RdLqGzNbMVv7BVLnxQGTCEErazRUxMzwHVKiYreTzZevzvUv8Hv4R8oN6vOjwEDIn14S9LCEVmApXkFtwxF8e0OQ7b2o6vvug6fLt2Ev//Tz8KALhjaBR/+Lov487vPYl/OCmU17szzwGGd7lvkyOtnNxKBYoh6DFPY94+2jnGueejSgNQxHhNaSlsG/DWWbkRrZCj64Yxw93X0sytsUQ0ISer1uXlNp4ch6ZoMG0TuXoONbOGpJYMz4+jRBcfAr0k5FfIJZmnoFtprKBumjIhE8uGKsIAkcPogqtYLDWkDLmxTFw6zw2c70stq+J5jDGMJEfcrMdcPecRctRaGrHtRx0n2ivkSJuwJQdvJrUkTJNJzw1TDFJnALc7WVbDSx1Gku0JuXalDlAablsrt3WA6/i5Lgg5XdHBwMDBYdomVJWHqg8VpmA4MeyW8azUV6QogGc7VkXIMcacfekRzvnLfff1As4537ua79BHH1GIUkj1FXIbE8+2UocohdwzJUMOEIRPrE/InVHIJI/XoFh1AttNG5kuCDm5ZVV+PiXoKKHVsBrSxNpPsHWC//lH8kdOOyHnz2NLtLHu9YJuCDm/Qg6AW+qgJE5hSTuKmfJzJIUD4CNAVl3qED3Z5ZzLhJwWzJDrRRF721O34c6pO/EbF/0GLh67ONS6dWC2iJddsiXk1atDswsbZacMOZeQU2Q1FYPuKkcqRvdjvNa08GTOI1hCFXJGFZxzPHaqgE0DcWwZDGnua4Ni3cBr/u+9WKzNgO0Q5F9CTeCyTZfBNB6CjSEosTyadh0Hlw+621dUuzAtLwAEQUAJOc653Gjbw3h0EGZjUgkJSgk5am11CTlCnqiMwW5s8gi5vEfI0THL/KrHNphITeDwymEAothBalg1htxJNCAUckcWyu6Fsl2jaYkIBwTh6P0IDSdzVYykY9I4KBtlcM4jM+Sapo1yw0Q2oYcr5EhzY5hldalaAiBuj6RVLLVUMwpTpOddvvlyfPvYtwEIMvKjP/4oPvHTT0jf9fUjz4P60KeBlRPAynEgfwJ/vfgUEokWyfeQ+G/ryDAwKFQzOT1831om5w9pzoChHcDgDmBoOzC4Xfw/tEP8PbgN0OJ44lgOv3OzIJ+vzA7jHXuvxkttA//rwD+h1CxhtjKLA7kDUv6Z/9wlR8aVU5xBQQkdTSEKuda6dLYBfz5ZihAo1TYZcnGFkDmK4V6gkQi5tpbV6KIlVVExkZrAqfIpAKLYYc/QnvCGVXJciGFYJuR8CrmU5hFyhUYOhUYZjLU+245DV/RIyyol5DhXsViqyxly2bh0LuM/FodZVgFRIEAJuW3ZbQDohStbkKYt0HEcFW3QrWXV9llWU1oKZtN3fhFiHZVU0SGEnNyySugbomobS7bPe6Of61fI+S9yMAZcvrMzIceYsOA676cpVmSD7UhixCXkluvLfUKOYLUKuXNb/5sh9/WCjT1z7uOMICo4u6+Q25jwK+I2esunpOShltUzSDgHCTkbMawuQ6iP9YFc6iDWRYIScoaFTBch9FJoso9kpSeh9OSfnjgCa7OsAsDh/OGeXr8ayJZVWVFoWDyQcdQtKOHlVyw4sMiyc5SlQykdYAaS2/8JplbBn92bw6dv/LTvO5/eUoemZbvqmZiquISstN67PK7+bOln+KMf/REAkUX25Vd+OVwht45Nq4Ik8j6jm7DusHxFTyEnT7IZvIl9tQdC7o++/iDKplDi6IqOiZSYlFBFUsWs4CsPn8J7/uMx6CrD3e+5vidS7rtPzOPYUgX60EE4U7ArJq5ATI2h1rRgmbuhxH4MAHh04VGPkIuwrPqbkv0teYblRYVoCutKfesHJdkcWKwCVWGwbI5Sw0TDtBDXVN/ksWVZpaUOCpOIK0fdBcj7qiiVTBg2p0mOXGXenfADADdlQq5iVHBoLrphFfDtF20NU7kqnrd9SC51MMrI1XMuacLtGLgpq2CWy01kE3p4hlyI4o0SbblqBQ4hN5TmWCLPddVhloHXjl6G5R0vx5enf4icJcg+wzaQb4h1FrdtvPZHNwO+fVxYSe9W05s6zmga6lxHJTmJ0a3nuGRbZeURIC/s1KlXfxw499Uh7yRDbvkV41hXdFy77Vrc9tRtAIA7Tt4BXXuZ+zy/wjdXIwq5kLZMejFCJQo5v2VVUgjpGYlAqTTNQAOqA13xxpCqGu5zqGVVanj2v75DvufWzFaXkJsuTwtCjmxLnkKO7DfZMGiiJc0IBIAMIeSKxgpWat52rHCxDCkhF21ZVUIVcrRlPNCyGlLqAMiNnitE9eiewyhNlzRMaSloird+ooi3bi2rtiHvp9N6Wo7+8GfImcEMOdeyavdW6tBJIecvdKFwtmVAEHIXTAxgMKmjG8TUmHfephiB8e9+v37TaiRWS8i9pPV/JeS+Pvo4o/BnGTnY4DzNsxbtMiU2IqKuCJ5JwpmSGcDabGx9rA8ky2rrpD5BTjTDyJAw0PHWrmWVbnd+Qq7XUgf/iaKjSjmd8Ns/GWOIa4pLStQNSzoJ7gacc6mh1YrYLqTijNaEIaGrSCTzUDRxmuXYBaO+M23T64QoFZQfdXJVn76mnQojDJxz/O3Df+venipNAQi2rALr27Rq2txVKKltSKLVWFYHYgNQoLuR5LUeCLn7Tz4JtGLANie3ullsfoXcnU8KVZphcdx3ZBmvu3xb159xakVM7tS0Z165evJqAOI3WuYu6IMtQm7+Ufz6Rb8OINrOTMknAFiqL0m3G1202XaCP4AdAEpGEcMp3c2VylcNbB5QO7asKozBbnoE2lN5T2EmnSMo3iS4EyHnEKcAMFcNUcgRq1q5WcZTS14xwXM2y3ZVQFYKca5hqrXOaIZcuVkOyY+Tt/XlSgO7xtJI6h1aVvVgy2q+XgEwinHkcbG6AqdCJ2U2gc+9UqjditPQuYXfBvDfANyeSeFL2SweT3ikwisqVQxFXHCo8Dim+CZkJ/Zg6669mFQtYO57AIBblEvx6cbb8IvnbsEn33y595ofvBNo8QOZWJDMDAMlkOkYvGHHDR4hN3UHXjnyCvcxv8OCkhIsTCFH9iFqG8sqI+qrTCwDtdXcXTfERY6aYUk5YA5ihKhRFI+IottbO0JOVi8H981UZe2MX5pN5irkyHJJYBRJLYmaWcN4clwK5weArD7k5tZVzKJEajqEXFSGXMCy6i91yMSk/VDgfD/i/IQSP1R569paQ0oHHKzVsmpZci5aWk/7WtwVKL7cNSA8Qw523LWD1syaGy+QitFSB++4NBSXCzj8oC2r/iiRQl3ep17VhV3VQUJNoAgxjhizJMsyXb79ptVorIqQ45z/oJv7+ujjTKCbnJ4+Ng6CLasbez3TSf0zVSHXV6OeeUilDq1hIjWHdmnDNENCk93bSvjJP23IA9aukDuSP/1Nq0aI/TOhq+4kr2HaSPeYP0wVZkB36m2VqM+yqTqcJVE2yrC5DYWF5/b1pJDzEVBRmY9h+XGAPA6sLiyrd0zdgUfmH3Fvl5ol2NxG1QhaVk+t1FCsGxhIdHdlvh2aXVoo5VIHO6BeibKsKkQhV+uBdK7YHrk1nvSaD/0tq5yQ5u1u/FdAAAAgAElEQVSaGcMwm68DsKClvW3nmslrxHdtWrDqu9z7f7zwY/c3xyPI2k4KOblhtX0rXxQoGeKg2ChiJB1zCblcpYnNA4kIQo6QJe0UcmR740r3llW/Qi6YISdbVmdWvPfeNhxUN0r7Ra5hKiee729ZDcuPi2mKO76dZSMp5JwMOWKDS9dLwIFvIjnvkfsTB2/Gwfg0EszA0UUNX98mCJt0owJM/yjwnWMAXlWu4lXlKh6PxfCNbBpgCn7fGgD2PBcY2gkM72z9vwsfuK+CTz9aAsDwgedfjDe9YCe25g4C3xSEnKmXATA3a4/+bvd7691lSkqkMBnH10xe44b8H145jPKQN5b9DosVotJkdvBz6TEvTCHnvl9IhlYmrqFuiOeVG2Y4IUcsq4oq9jucc2l78xNiUd8vTJywJePFAUyXpwEgNEOObs9JPYnfe+Gf4xtHv4E3XfCmgLIvFYuB11JgWhUcNqbKXj6mwsUyjLKs+ksdFksNLJW87WIsE8cisbD6z1v8URMOJCVW3Vt2zjmz/+IKRaRCrh0hRz7bNIMZcmVbVsixEIIyzLIKMMSUNBq2OKcqNUsYSYwgrimucrjdb/FDV3SoTIXFLVjckvJD/Qq5F+zpnpCj6xeKEbBsO4giSvvolzr0sQFhdJHT08fGgZ+Q2+gKOeNssKxu8HVwNsAi40EhqisH3RYVSKSPv2WVnoS2Ucj1WurgDxteqi1hpb6C4UT7fJS1QL6C7SgKVTiX/juVH4SBKsyANoQcWVd0QpVM1lxCjoOj1CxJLWpRk5F2mCpO4WM//hiS40Bt8VoAYvJFw/wd1EIaVoH2ZR5+GLaBDz/yYek+57dEtQ0emivhyl3dTwai0OgiPw7wSk+cyXTTsiXVXCGiZVUi5LpsWbVsjgabhzPdGo15ihVKyNXMGkAIuXZB8GGYKdSgJE+BqeJ7jSc3u1azmmHBbo6Dm2ISvdJYwbHiMewZ3CPnC5LlN19pT8h1S362A7WWOSg0CxhOeSRCrlViEEbIaZJCDrCbo+BcAWM25ipzKDfLyMQy0nbDe1HIkbyjo/mj7n6OQQO30gAYdCUOw26Ag+NUwZvgTg51Q8iFKOSMskzItRpWtwwmcGJZPH+53AQ4R7LhkVi1Zhn46ttRLfwETnpE6r6PA6UyEkMDwHBLSWOuIMHE+K6Q/Xua+44P2S0+sm0nLhnaiUuGdwEDk4ASTsLW9MeBVgenc45CVVqKvgKA42SuKhHhEiEX65KQkxRy3vdJ6SlcPXk17jx1JwDgWPV+AM8BIFseAZ9CjocQcmRs0+3ftayGZMg5BKsg4FrkS8MCQoR/GrGsspZCzmn2BQQ52a5xXO+wb96a8S4AzJRF02qYQk7K3lQVvHzPy/HyPS8P/cykrsK20lA1MR5PlU94vwfdK+R4q2V1ueLtS8ezcRxdbNOyKsU2hFtWKfHj7COiCCOgXRt39IUGVVLIKUi1FIWAWGd53/mFHpLlJltWveWVUDIuIVdsFDGSGAFjDOmYimLdDDTadkJCS7jbV92sQ48FCbk9I5vwkgu7z3eT2lKZ4R53gGjLaq7WJ+Qo+oRcHxsOUbYgu0/IbUjQrAtg49slqZomLpU6nIlvI/BsUymeDbDtIMlDbYftssMoqJ1S99ki9QiFXMlYGyG3XC0H7juSP4IrJ67s6X16QZMSj63tKkot1C3q/gDqiI2U2j7piX08IatG1oOQ++Rjn8R3jn8H2higlM6BXd+GumGFEnL0N1OiRrLndNjWbzl0C04UTwTuLzaLUri5pjB3DB2YLa4LIdcLSRTXPEKuboQRcjyQIafCUwV0W1xSbphQYh6ZNUwJOU1WyDGy/Cu9EnL5GjRiV71w6AqX7BDrVYFZ2wk9ewCAsK3uGdwjk/bNNgq5ejuF3OoIuWiF3E73dq7SRN2quyH6uhIDuJhQUpWn+FuF3RyDGhfW32OFY7hk/BJpe+NsdZZVmmuZwCiKLRtpXEnBsMX+brZUAFqkTVj+n2RZtYll1ZchRy2rw80EzlF+huu0MmztGHaweVx5dx74wSzizRKUXdthMwYDNozHvozK+CiQEcSSQ7IlyTZbb42JPE9jNr0VaF0CSA+fA1z/cUG+De0A9LA0uM5QQ/YVA7EBxFgKTV4FUwwwtYJqM4PFcgObsuJzJEJO61YhF26xB4Rt1SHkniwRQs5qQ8iFKOR0KUMumIvWsKIJORp5EKV41VmQkOu20AHoXLgzmfb2Nw4hRzPknG1A2nd2sKAnYiq4mQHiIqz/VIUQckwsw6iWVcOSFXJHF8ruOXwmriGhq20jBaIiNaIy5NznkIsrdHsDxLFIYYD/0NbOiu8vuBqIDUiEnF/NLikGzWDLKmzv8ZSWRcEUqmp/jlyxbgaU250QV+Pu9tWwGsi2mGF6keMXL9jTU2YutcKCmX2F3Cqw2pbVq9frC3DO71uv9+qjD6BNcHafJNiQCLasbmx1lpQhRxVyZzJDLqCQ629rZxp0lTjtg1H5UO0ghRG3VchFlzr0SsgtVYKE3OGVw6eVkDNCyBu/tbNX+Jdx1DFItqySFj9NJuToyTiwOlUSVdsoeg52fRtqhoUw7SElbeMSIdc+OJx+30/+9JORj1XJpPSirYP46ZSYDK9XjlyUhS0McV2B44wSr/PUL8WaKa76M/F+cTWOuBqXLGt1s7sxXqwZEiGXVT31l5QhZ1ahUMtqhJowDJxzzOTr0CY90mhL7FIA4jjhjGWrulsi5G4676bIfEE/IUdD5oF1ypCLUsilvcnpSrUpTRzTWha5Fhkm2wnF33Zjs0vIHS0cFYQcOUe0CSHnV8r4QS2rNlGQJRWPJIkraZQhfsdCZQWAsJhODgUJLWm/yHXM5GuwmjVk8qfcu8uNIp6aedC9/SV8EntiJlCAN4Nr7SYYgCTnqLR+e01hqBJ7YSo9AUycg2RCBSrCtrpv4Do8d+YVKCKNtz23DJz4gHju6HOA56w9Glxh4eR9Rt2EnHlcfG99BdzK4MRyNZSQ82d8RSGs1MHBtduvhcIU2NzGVHU/mFoCt7KBC7iFBi0kCLOsUoWcN4UOZMiFhNqnyUWPqO1ZAyXkBMknFTq0sav6v19HhVwlWiEnEewdtuekroJb3jqaq3iWVR1By6qkkOOUmFRRIfu8sYx4TbsW7KYUqeE9j6rpO2XI+bd7xhhiJD/WQTtiUpWswhwD8QF3n5nSU3JOrMqkHEe3ZTVCIZfWMw5PLp1bOTlyVCHXyXYPRBc7UDK6UxadH5Rw5WhGNtj2M+SisVqF3D1Yn4ZUvobv0EcfoYic9PQz5DYknk2WVcv2muwYkycgzyjLahe5Un2cXoRZVpOrsqyGW0IAX6kDWefrnSEHnP4cOUkJ6GbIrVEh52+E6yJOgW7TTC+7YdlAkJCTrrp3WepAT4Kd7Jmo30YVUsmoUoc2FwI+/fin3ZP8rZmtGE+O4yeLPwEgrsZXm97p3+U7hgkhtz5Nq5Ltqo3dyP+4v8mvUDNCc3qoQqbe5Rgv1g2wmLcO0swjeWj7ZdWoQiGEZS+W1ULNQM0qIZMU5RmcMyTMveJ7EtJCa3ptiY8uPArA17JKnhuwrPoVchF2wV4QppArNAoYJYRcrtJEoUEme5o3AVWZXyEHKUfOKXag243Vg0IuqSUxGB+UCEEASBNCLqZ469DkYj82lNLlvDCjDqwcg5HzlG//Xfk23qB9Hur/zkEFEN+5DQ1FgQWOOYh1r3GO7SG5i96HZ5FiKiqt6Vn1JX+J6vxdQEEoJZOv+TgweTUSR28F7nkvACCHGIot0kRVvTHcbW5bJ6gR5yiCxDwOQNhW7fp2HF+quMpYSshR5Wg7RJU6AEKZc9mmy/DI/CPg4NCyB2DkrwpcTKRqMYUHiUB6zKMty45i0z1e+kodgO4UchrJkAMLIeTaFDoAPqVWyLnweGrczRBbqi2hbtbDW1Z7UMgl9ZZCroX5mkcox5i4P6Z0zpDjXN5vjGcFydOuBZv+xhj57ZS4lAm5lmVVjSbkxGeqQUKuzUUv3dc8ntW990xpqcCFzbBMvYpJSx08giuje/ulYNOqReyhrONFBSC62IGOg6HE6gm5ml1wG2wZj0sNtnS9+I8hz3aslgybwfoQcn30se6IImT6ltWNCf8J1Ua2rPrbGCU7yBkl5OTP7ltWzzxo2P6aFHJ2tEJOOgklY8BPHPXasto4A4ScbClp2c+03glMirUq5GwmE1N+5eFqLKsSIddSCUT9NkrIJCNKHaK29enyNL6w/wvu7d97/u/h9uO3u7eFQs6baFy+cxifvVeo9w7NlWDZvCfLTBgaPVpWw14HCFUbJeScSY/GeresLleqUDSxHjln0Ik2UVM0xNU4GpbIIBOKCfG9erGsTudriA0/4E6K7PpWLBbE6T5tV47bO6CrCdStOqbL05ivzIfmTBqWEZg8tS11WEeFXLFZxLaUTMjlG5Ss8SagaqhCjhByLesn3VdZ6E1dsjm1OUjIqR4hpzOPPJpUZ3CesoSfi+eB2/4TyB0Flo8ChVMAOBrETnohm8YkIQfTnMOvudxumMjzQZzgmzG09Tx882QcJ/hmDEw+B3/5llcDqVEkv/ZKoCQUSrULX4Vq7n7vPVskG1XImJyqlbwx3C0J1gkyIefdH4e3zESOHHCylaFn2IZLFChMkYjqduik0rxh+w1uuYyWfUIQcr5tnSrk1DDLqqSQCyl1CMmQc/YX6ThVyEUQcmSfgla2H932OlpWfcRQ8HENE+kJt9BhtjIbniFHomA6WlZ9CjmDjKm4Iu6PypDzlzpQjGUcQs67P3gBvouW1VqYQi64P6cI+83tloOiMDAG94J5lpD7aT0tXbDUQxRynHNJIQfbG1uDZL9E11U6pknEooaUVPoUhdOtkKuYNIdRbsLqZ8hFY7Utq913r/fRx9OMvkLu2YWm6SeDNq46y/KRI9QOciYJ52CpQ39bO9Ogq8RRi8TbZMgt15bxyPwjuGbrNZI6Qio7aKOQo1eqy8baFHKG3QB8XMyRlSOB9sv1RNMMkluyVWY1CrkgIRf2G6IIOQMysUlPxm2buxMuv1o2ClWjKufUtEimqDzBGimlkEodOjT5AcBHHv2IO9l67thzceOuG3H/rEcOFBtFVJveONsxksJYJo6lcgM1w8KJ5Qr2jLe3qs1V5vCpxz6FC0YvwE3n3RR4vBfLakyLXteFmtwa5yrklBjQ+vnd2rJPFWfcv7k5gLohP57W0+571a0a0FIv9aKQe3jmZ4iNf9+9bRQvxcmWrYkWNaT0OM4fvwQPzT0EQLStJvTz3ced8btQWwh8Rr6Rh2VbUFth/r0oaqKw0gixrDYKGByl9mFDIsRSajgh5/ASdtNTIDqkvneOwGHCI/e6yV+aSE/gyZUnpfv2GEVcqN6OXWwO9xeP4WTr674v/im8xKoBNQAPB9+rQfYDsdb5qQ0FytA2ZFQNOcjrfNZ+Hq5svBkA8Mlrno+PHBOqxvOMDJAWRA0N/K+a1VClGSW4LELCUUJuvRRyUZZVnXvEEouJ9X68VVJBiYm0lsbRxQq+8sgUXnLBZlzRJluyU9PvDTtuwAcf/iAAQE0dAZRG4NylSFRCGtor5MIsq877UaLEWZZUJRlVaKOAZnEFFXK9ZMj5f1uhZiBfbWIyM+kRcuXZUIVco43a0A+/Qo7CIeQ0RQMDAweHyU1330FLHcDlz/EIuTVaVhs597jrEnKkdCCUkAu5gNNpv6YrXg4pzaVL62lZIacqAZVa3aqDt3ROguj1xq9EyEkKORUgdlVnWXdCIqRQAgAKdTIOuiiHoKCKv6JBLvrZMpnutxKfznO6sw2rO2r20cczGFEKqX6G3MZEQCFnbtz1bEiydxYamHwmECx12Lik6NkCO6S5k6pfqPXD5jbeevtb8a673oU/vPsPpfcxI65A+29TO6bfstprhpzFafug+IySUQrkWK0n6OTFOfFeTSstRT0kdy5sO40i5BrcR8iRk/GmTx3XzUmtn/BwJo1+G5ADStRRKyOd+IT9nunyNL5z7Dvu7Xdf+W4wxiTCw1/qkIypuGCLNzE6ONfZtvqRRz+CW568BX+x7y9weOVw4PGeFHJ6dF5goWZIEx9nAqezcBtWO0yXZ92/bWMYNZ91jZIlDcsjT7tVyDWsBj53+H+7eXdWbTuM3NWu+ogq5BK6iv/H3nuHSXLV58LvqdRppifu7myc1WpXEhLKoABICLAJwmCMsQHDJRhzMcbp2mAD5tr4s0kXfMFcsAm+JlgYG2MbczFGgAiSACEQQmFRWIXd1ebJqUOl8/1RXXV+59Sp7uqZXm3q93n22Z6enu7qCqfqvPWGy9Zflvx8x9E7pH0+3v6qXRWIxgy6P1ES07EYdk/v7jojaL6ht6yqxyElEErUsiq1rLYUcu44jNZU5+DyQdS8mjiPMg8c0XI7hiNNVBOEITC/H3j428CP/g4bZtIFJS86eAP+3P4MXmvdiPGGIE+WjYx9jhnA8CTcirBvfcT9ZTyj+Vf48FW3AL9/DwbGzkn9WeCJvMENVbGsM8tivKT7T92ro+YTcitWyJHvScdal4v9rV2TZzeghx29aWgEYmIeK+T2zUTkoUQi2mX8zufvxMe/+whe/9kft70x0olE2jK4BeeORNZtZgSwKg/IjbucY4nsWybS64Ce8xjPVsiBKuRa9sUBYlnNItipDZ6vgpBzMnJdp5aaeMp7bsLT3/8dBK5QPx1cOZihkFtFqYMGBTN6njEmq+QSi28ehVw2IZdlWS1axYSA9kM/KZpKLL1GdqkDoL+B02k90PHnrKFdyeMdQzuUtnqWKrmg+7zN5HFopCjIMTlDzpKUfsWchJz02ST7lCrkYmI2L6jqbskl1xmh/F2KVjEZh3zup9wUZzL6+W19nHbIumN/IjO2+jh+cJULtKxSj9MB0sTdZLJC7gTu3uqd2NPZNnyqQEfyZGXIzTZmEztXrJaJodqkKbKUUqlSh5yB9zFCiEmiHW6GZ0ZZWHvm9mCiMpH1Z2uCzv5ZaKOaygOdLdgPOdSILd065pyj5ssEBV2vEoGY166qWETii/kshRxd/qKjt6zqjvW9C3uTx5esuwSXrr8UgGwJjBRyYlJaKZg4f2MVt+yJJp/3HV7E9RcKAkKHe6bvSR4/OPcgdo3skn7fSTFDIecUaTLkSkQh1/oeltSymG8fP7IiCDnuDUukJCArk5jRTLJh1Ndl4a9/8teYdiPLIg9t1A/9KgATh+br8IJQ3qYKIXfX1F24aFs6Sy+LCJ+pzyQEAV3Xx8yv4GX/+SWMFkfxues/hy2DnU01QRhIRFuMRXdRznL0A0khVyIKOanUIX7MLYwXN+NYIxpDHl14FH4Q/Y2cCzgIHPwJMPMQML0HmNnT+v9hgKhKJ4aqwKhs59pIxoZBMu4eZAP4XrAdw1vOwwUXXgqMng2MnQ2MbAesAtyvvRY4Gknn7g7PRcA3Yt9CdEzoSALuCvstJeRma25i8aZW05pfk9RmWoUcaJ6X+J49y5CjCjlyDc780UQSwuxoXNrXUshRhXXRLOPeVsnLXM3DscUmto7qyUJZEas/3p+57Zl4YO4BAJFt1Q2uTX634q0kJQM8tGGxQurv5XMgsVLGLasay2qcIVcmY2gtI0POJAq5mJCj9vCOpQ4ZLavffXAqKUyYnhPrb//i/oS0NZiR7HdyYVD7sbNkmwgDPRlUMsXzjukkiiw3cFGySlKpg5ohNz7YKnWQypWU6/02pVOjxVHUlqPvNlufRdWpJjeTqIJRp4zVKuQ6nGctkyWZry8865fh8zoG7AFcs+UaBOE3yesMGMyAbdiJipzejLANWVU2Sgg5VSFHxzBq328HSsjXyU0fOq52a1mlCrkFTxByYZC+yTFaHE0IyNnGbNdqvNMVfUKuj9MOeXJ6+jh9kG5ZPX23MyU9UpbVkyhDrn+snXjQbWAkGXJ6yyqdtNX9umQj8BSrBQX9mb4uvhsdoxuFXBCEgCHumpvuNnitcPqH5h/CNVuuyf1e3UD+nqvP3KPIIuRU6MjTml+Dx+X1Ri/GdZl3naAqluJJSWaGHFl+KUOug2WVfs7GAUGqpRRyxLZVti2cRxRynZpWvcDDgSURHj5Vm0q9xu0i1yz6fYDixi/ivXfdgA+OvQs7hnfA9UPUvQD2AMkcailebCmoPJ8te6pxJHkcesMpMpQSKnHLIpDPsnrb4dvwDz/7B7FMR58P7q6LPosDB+fq8jZ1TFwwfkHy86MLj8Ih5SCNNgo5QCYK6ER5DpGVcrYxi7fe8lZ8+rmfloK9dVhyl5Lm0jh0HoiIW/U4zCLkDE2GHABMlLYnhNyegz/AuqNF/Lp5GwadR/H3rddUF48An3xG22UEgIkgfUyHlYvw+Zkh7OUTODDcBHhEsn2YPx+u93P46ysvwQWXbE79HbXycx6tn5mV6DkdIReQPLyyY2K4bGO+5oFzYL7mYmygICvk/Lps/2yRbPQ19OaHG6Zfu1bQbUIVcqE3gph7ihRyYWKplNomA5kUW6h72JrxWe1KHWI8a9uz8Ld3/S0AwBq4H80l8f2pQogHZWnZY1gZCjlhWW3FCOhaViWFXJZllSrkPIQ87NKyqj8nzyyLcwn3RxMu8f7Z+5PnB53BJINMvpnRKUPOyFTIUUu5qgoDoFhWMxRybVpWpRtpapFHaRQHlqNzxGxjFtuHtudqWY0+M01CdiLkqHrSNop40yVvSn72ddcXZjEh5Kja2FEIufGKIMckQs6xJEKuYne23MefGyO+WUpVhCxnOYT0noTkW2iK8z8P9YTcY0vReDzbmMVZQ2elXnMmom9Z7eO0A7VO0QH0NBZOndFIlzqcvhtabWOk1wcnlWX1NN4Gpwp0JI/UoEgJOWJr4uASgeZLyjF5gmJn3I1XFXLdZMgdXRZ/y0MLcAWho7Ml9gquRnFWbDMRyANVaQUAgeaGgZoNCaRD8wE5Q26thQ6AUMhlkY2yvZG2rLYPDqefQ0OcqUJuobkgNQ1GllXx+05NqweWDySkDaDPOWt2kWtWsExYA/fDHr4Tjyzdi4/d/bFoOevRhElSU7W+ByXkvDAf6TzXFMvJvWFpHQMKEUKaGjtZVheaC3jHre9IfvaXz4U3f6V0jO6frck2ZNvAUGEoUUM0ggYaXOx3calHpkKOhM3Tfb3BBYFw19Rd+OTdn2y77IA8IZ2oTIC1QiSXvCXYJnlvL5QIuaJOIddYwA73AfyScQvebP0zLpq+L3nNw9/5CzzrjjfhT+1/wLW2UANX210glseBrVcBl7wCGy58ufSr0eIovvWk/4u3+a/Hx4MXYL54bvI71tp+m4b1pQTSjYowImtixX+sqqLwG0IdZZuG1D4bE3mUbFtoLiQKJMuwYJsRgUQnzyETY3MjOA6W1QyFnOsVEbZIHGb4YHZEhu2dkXPvmh4pOYA4HnXoVOoAAOeMnIP1peicwswmVgxBSNH9igcVadljSOdAnm5ZFaUOmpbVHAq5MGyd9+LvFDSl46xjyyrNICXniNkVsZ1DYlm9b1YcG9Sm2I3dv12GXMmSFXLJ+ycEJt2eekKOfr7rh+BkP2rXAk/PPXFhjJO0rBLLquZYK6TiOZiWoKVoFyEjlTq0zp90fdBzpkOadg0GjJX0CrlywQKjUQp2PhJNR4zSfb9aqCbZoHlBv8scIeRCP60ylZpWNdc5Zyr6Crk+TjvQdsGCJUI2+6UOpyfatS6dbpAm7soFwslkWe23rJ540PEusaw6essqnQABkbIinrS1a1m1MlpW1Qy5blpWDy4QZRS3ETTToezHA3rL6hoVchqbq85SryNPdflbUoacpoSiE9SmTLRUHJmlDnkUcprvk0nIEYXcfHMxGbNsk8GxDJy9bgC2yeAFHAfn61ioexgqiUkvBbXFAsCxWpqQ60ohZxswCuI9dk/vBkAIOYPaG1uEXEZzYDss+uIzQm8kZUWlRAhVyK1kKGpivPuH706IMxZW0Dj0EgAMF28Zxo/3iRbL9YNimeNtur26HT+d+ikAYM4TpRMNjWV1rDiW7EdUuZMQ2kYDPmkuBYCP3f0xXL3palyy/pLM5afqpLHSGBbdxYTY9yHWfcMLpNeuX9yH15l34Gx2GE8+OA28/wiwcgx/AiAWG904XcINGyKl4B6HFESQ46YacmDdecDYTmB8FzC2Cxg/BxjfCZRE3tnE4j7g378ufq5MKLliRBHSImU2Dmmy6aAQci1yJ75+URVqG8obsM93gJaJ2TIZxgYKeHgqGrunl5s4Z8OgtP/QyS59P6qQ4UQh1wzlMoVewJBICvF83fURGuMwrOhcYTgzCLxR7JtZQXFYnD+W6zIxMF9rR8iJDyhmWFYZY7h64un4j0f/KVoO+6fJ7ygpzP2ytuVZOgdyMYVOMuRiBaXRSSGnJ+S8kAOhAxjR71e8FWk87WxZ1Z+TZwgh5zVHgBZvKxExZHzupqSlaJsAd8BDG4yo2zlnKJt6Qi5eX9Syqirk1rUIOcNgyXkh+tswOTdntawC8rknHrOsnAo59TsX1JwJDbLIUCBDIUeIcWpZLZiCVK84lmTplFtWTYmQq+a0fupKHdZiV42WWZxXZsl1RhCkCbnREmla7TJn9HRGn5Dr47QDHaAdy0AckdG30Z2eOJPUWVLjpaFmyJ1AhVyKkDt9t8GpglBD8hQ0ge2AbFkFIkJuBNEktJ010pIuQIllVVHIeWFkvYntMO1waFFccPLQhltbnxSuPjz/sNTs2CtE7afRY4MRRSHNrlpNqYOG6NKdh3zNttLdOc7MkMvZahmrBGJ0UsjR70wnuNKkT/N96OfQVjWqwJinTZmt9kHbNLBz/WBiV73/8CKu3KGfgO5blMP1dZbVPIoZ+ntmi+Xev7Qfy+4yUcilM4ccopBzcyrkaqEgsbg3jFpby6p4TzcI4fqhdlvfdvg2fPXRr+q9elsAACAASURBVCY/Nw6/GLyVk3bFWaMJIffYbA1VQnDG23SyOpkQcrPNAwAiEjxW71HL6gXjF+DmAzcDUCyrrX3FsNI5cCEP8dZb3oovvuCLWjUKIE/MRgojmHVmxf6+sBvPNW7HLnYAlywfw2f3PpaIaa6472N4st1aT3VosdMTJMHDhSIODl2G784M4yZuAIgaU6sX/DJw7Xv1b0Cwvrxe+nljZaO0TRgXk2lmNmAwOe+NIrarAcKyGl/PqJbVHUM78JAUDG9gfIAo5JbTCjlKmNL9ir4GVCHnH4cMOUN/jVL3AnBjDCjvBQAYzjSClV3YN1PD1oo4Hy2uyGN9O4UcHcfaHe/XbHpmQsi5zl3wQx+WYaUtqxqFnNQ0riHkovMgly2rTpqQy2pZ9YMQnNvJOe/w8uHEyj1cGE5Ujlmg52h682eOEHL1+gCMESN53xiU+Olm7Ixu9DFwvwLmkNzTsAiHEFkFzQ0MqpDLypCLlsGEF/itZROEnN9GKU4JuXh8SV5jpks3KNTvnOcca5p6hRznPOVuARQSi6jKiqY4VisFKxX1EKNckEsdRvIScoSQb/jRepAKHVaR6Ua/C70BGwYFhCGXiHnddumjb1nt4zQEHQjpoNovdTg9kVbInb5kkKqkkewgJ5Bwbnc3sI8TA2lfiTPkpOB6vWUViAi5GFJuoUrIaSyrIQ+lUO7k83LmyB0jllWEDhrNItaVInWLG7pJ9kgvkWX/LLYJk84DHYmXN0MupWaDTGL1NkMuh0LOyVDIaY71PAo5eqefhp3nbVrdu7hX+llnq5RbF9uTuEXbhOHI6+fBuQdxZKE1uSAWp1hRQRUffphNFMQIeYgmJ3YebxiNNgo5alkFsm2rtx26LXn8jC3PgbsY5cKNVRycs0Gsz/2zNenzEoXc0PbkuThrDRAKzyM1kXt3/tj5yWPJstp6LSU1dw7vTCa7B5cP4j23v0e7/AAwvyKUg8Oz+zC0It574L9ejY85H8If2l/Es/ybsUCslVXd+d4s4JBzFr4aXIGP+L+I6Sf+OSwWfdcjJsPnn/hBvN3/DdwM0WSaV11SskqSgmRjZaNs6QvFxJQZDawfLGYqWGWFnELIKcTl9qrIWbKMSB0/VhGfFWeEUeKNEnKSQo62yRoegOgz60HvM+SyrlHqboiwlW8IAIYTEep7Z1ak80cQyJbV+Xq2EpWSddUMZS0AXLz+YoReNBZxcwU/bhVryJZVvULO7kDIuX4IMB+MRevUNuyErKCE3EqGZdULOHgovnOcgQZ0zo8DslXrVCFXczk2lDdABY0U6EZdHI8j8U2AGDwoS/u+zrIqK+TEayuOmdyoUZeBjutukCa6YugJuZZl1UiP5xQqAZenOMkm656en1UyLs7nlQi5OiXkBGFeLpjS8tGbcgNKqQNtY22HgpUmz3qpkKPgQTF1w75PyOnRV8j1cdohSznQV8idnkhlyHW5ne8+MI8PfP1BPG3nGP77tWf3ctF6Dnpit01DsYOcPKUOfcvqiYdU6qCzrPrtFXIxZFVmdqlDfPG/4q2AI73941a1Tji2RDLkuAU/5Ng5vBNT9WjCtmd+j0QgrAWHlg/hyw9/GZeMX5k8Ry+8C8dBIadT8OrunusIud5nyHkA83OVOhQtE37o40N3fAg/nj4ImFcDQUU77lCFHLVY0QnfspdByE1UARwE0L7YQSXkpmpTUhkJoOQC5lDIGba8fu6fvR9Lc9EkR2dZpVk/eTLkZhuz4KzV4uiXAV5AzZMn5nLLqkw+LDd9jFRkggKQ1YK7Bp6cPN44XJQaKdMZctF6P4uQPUdqBwA8CUA08fVDXyJ2njD6hOSxXOrQUsjZQmlx4fiFeMNFb8Bbbn4LAODLD38ZT1t/GZ5X2gZMPwBM3Q9MPQhM3Y85Ppe0l44cuhvVgg2UovFiQRl3FohC1q9eiBuOjuARvgmbzr4Qv/FLzwGGtuJ9X7gb//HTyH77wW0X46zHvpVkUE439wMwtbmAebChvCFRlExUJmD7YvloiDkzmtg4rFfHATIhF+eGxfusqpDbOjiZPI6P97GB9hly0w29Qs5gBgpmQXw+8wHuoO6L6IJeZchlXaPUXR8hEwST4UT70v6ZGs4hEQo8TJc6ZIHaWUfK6eMkRtGy4C89Ec7o9wEAX9/7dVy18SpFIVeB4bS3rHIuZ70B0XGQZYekGXJZ5LofhlI2HS2u6WRXBdQGbH2G3ErTx6aBTThMGp+BNVpWAXBfJnF5UJLOTZ0UcpSSGB+Ut3tW43k7pbguQ842DYB5YEb0HgYsLZm0KoVcxr6us6sC8vqg+16JHKsDBStFyMVug7JS6jBWzkek6UodJIWcszaFHAUPI0KOKuzpftwn5AT6Crk+TjvICjlT+/ypiLsem8eP985KgaZ9aFqXugxf/8DXH8TND07h3V+9Hwfmap3/4ASCnthNg4HeEDxR+0UQ8nSA7WmsUjxVIGfIRf/LpQ5iG7VTyHlhthLL1mSJqflxMfIq5KZXyN+3lALbq4Iof2iudzly7/jeO/DRn34Uf3jz7yTWLdrUtvaW1RBACGf8JhQ2fBkwavkVchrL6rK3lBzn3ZBNMXQXv8xotLGsygq5m/bfhM/87DPYvfRNOCPRZFZ3rFMykVpWB53BJKy/5q8AiN6fKkfkYodsQk61rLqhK93lB1SFXKeGPJ4Ey8e4f/Z+PNY6JzCNQq5gkVIH3jlD7tCyyGcL/WjSk8qQs9oo5DJUNZSctEKhOto0VMI2SsjN1LSqx8mqIHsOruxPHje8ANP16cTaNlocxURlIvk9JXzi8zBdhxs9F8+dPogXFkTD6F/c+qeY/dSzgf94E/D9/wPsuRGY34d5MnEfCQMMkf1qoTKO7wQX4+/85+Ed4euxYIvJ34+f9Am8w38d/j54Hh4auhoY2Q4YpqLMAnYO7RTL7UbfUdqmOQPRAWDL4BbpsVQeRjOTjEZmoQOg5A52sKxuqQjSNB6HaanDdMuympUhpxJslLiLiV9dI+taQU8ZsUuFc466FyB0KSEX7UtqqQNUQq5NhtxcTazP4TYKOds04C9emPx80/6bEISBdCOBB2XohMf0HEjLF+JSBy8IJTskXY+SZdXNsqxy8FAs+8Hlg8njToUO0fLp4wQoIecFHBvKG6FCtqzmVxfHCrkwkPdZHpRgk9ZmrUKOtKyWbfG940KHZBlsqlYXy9bOskrPPdSySglTxyhLN3GS51dByMkNt2K5qHWYErpUqUb3PXoOqDgWLMNK9iMOnihIK44FkAy59RXxfduBKmR1pQ5rtaxKCIspF5OU7dcvdUjQJ+T6OO3gqxlyLZzKltUf7Z3FL370e3jJx36Abz+QDq8+k5HKkOuSeD2yIC7Kjy3lIw1OFHzpxM4yG8weT+gm5Kc6+X06QCZ54tZQkiHn5lPIyZbVzgq5JU9vM4zvxHbCTI2oI1pKgW0DO5Ln9sz3pmk15CF+eizKzVp0FxK7Ip1wyZbVVSjk/ADmwAMorPsGnNHvwxn9nvbYUMtaAD155nM/2TarKXXQ3o0267lLHR5ZeCT5OS5A6MayajBDsuLFlllaGEEtqw8cXdKOL8vusqTaiqE2rSYB6+hMyHlsNrGZxbh/9n4cmKu3ljWtpioYYhLihzkIuRVCyHnR5KmdZVVVyOlUNSEPsX9RkGgeaeLcNFzC+ICTrN+lpi8suBDjwbbqtoQoPVo7DLCI9Gh4gWQF3lDeINnmZuozQGMReOx2PPHwv+PPrM/gaYVbk99v/PFngf/8A7ztgduwuaUEXDINfK+cVo3NmWIfGNnxcxiavCb5eeGZf4LXeH+Mv/T/Gz7nPwVey+ZWMAswmJjkU4WKoWSX7RwRhNyclybkulHIver8V2F9eT2u3Hglrt18rRyNQgg5ZjaxKaPQgXOuJeSaGZbVTRWdQi5tWc2TIQcotlXmwTQYVohC7nhkyMXjnBuECDkQumJfZfYsAB/Ty00sNGjTdj6FnB+EWGpE+wVj7S2rtmkgqE8i9KOxZrYxizuO3pHLskrPeWEo9llqWWWaQgegRaC0kFnqEITJjSgAUkRDPssqjROI9qWmH6Q+b7w4ARWrVcjF+7/atMqDkqw4N9IKOWpZHXDEPknzEYF002oMr41llSqx9i7uBecctsmk475o6jMtV2NZlcuOulPI0Qy5rSPDyb53ybZI9aazrZYVy+r6ijjXtgP93Ph6girkem1ZVc/hlFjuK+QEemZZZYxt6vJPmgAWOOftu9z76KNL+JJC7vSwrP7wkRnyeBbPPC+d/3AmIrqoXVuGHJ1od6uue7whT9xVy+qJWCL9+u5nyJ140BsQSYYctWAS24faskoVc+0ueCV7TGvfVAsdYuRVyM3VV8SVSUspsLkiCLleNa1O1aakUHVmLQDNCfkOtmRZXY1CLoDhCJLIKB7WHy+EaI+DxLPuHC+6iyjbZdmqk2OywDlvo5DTDx6UtC3YBmbnxd8zq9WAqZxX6349uci3DCul9Kk6VbGPGHUgqEjKkbGBAjYNFXFoIVqur+8+iudfJKs5VHVcjGO1YzhnROSCUYVcp0llE+lSiIfmH8LIXKRG0NnQ6CTEz6GQO7wsLGLciyY9eUsdAH3T6pGVI4k6Z7Q4itklsS43DRfBGMO20TIeOBqt8wePiuMzHg8c08GmgU04uHwQHByGM4OwOYGGH+LIisiP22CWMPLgN5Kf5+szCN67FSaAFwKABXzbXg8gmlhv9KPL+wHO8ZyVFfz9cKS8eGxoIzB5ETB+LrAu+jd/90eAQxGZN3zpq1GdvgeY+QkAYNFbgG0OR2MRsQ4PFYYQKEVHMehQFYYcZw8Lle2c/1hr/aZtyHlw2YbL8M2XfDNR1lBCnAeFpHCCtVHI+aGfWPtNZiLWR7itcZkeN1WnikFbTJITQq6StqzS/YeOuapCjtrWmOHCYWGiVrKYJRWWrAUqMQqQcYXbgD8MWPNgjIM5c+DuOhxbIcU+gUxoZhFy9Pmhkq0l02JEN10M+ItPhDP6AwDA1/d9PW1Z1ZU60O8T6jPkJEKOEKvlgiDwMhVyIU9uRAGyZTUXISeptKL1TdVxMUad9ByCKqPcLs4vhsFQtA0EKiEXljpmyFHL6mBRjKdphRy1rFJCLnuM3zG0A1WnikV3Ecdqx3DX1F2wzWGpAbdo6K3Zqiown0KOXo/rVXzS9YWpV8htGBjCF95wNR46toQXXBxRK1WnmozFi81FbB7YjIGCpRByOS2rGoXccSPkNAq59eX1eMUTXoHR4ig2VtJKzTMVvcyQW03SMmeMPQLg6wA+wjm/v4fL08cZCjq5kRVyJ2JpegM6KV6NUuN0hU/aEWN0Tch5+rttJyM8xbKa1WD2eEI92QL9DLmTAbK9OfpftmC2sax6RCGXMZ4CcohxfNGZaVnN2UK50KgBrev6eGKyvhgpeDg49i/uRzNoZlskcoKqlQCAWdEkkH7HQobFNy+aXihdMBvWQmeFnJFWyHFuJOqtJXcJE5UJJUOuc6nDkrck2YNiMLOebVklx3bJNqVlMhJCTl4vdGIxWhxN2YGGCkOJDYuZdXBPzjYEgJc8aSs+fFOkhPzU9x5NEXKPLj6qXV61abUb21WNp5XnXujh4MpeAONgZrT/MrCELCkSy1HAO5c67F8U9rMwJuSUibmkTMpR6kDtqpPVSRyeFvvbxqGIDNqaQchRZeL2oe3JdtlSeBBP9PbiXOMAjt4qjpMND38X9u1fwvC2zZg3TYSMYc40ME72xcMWIQRHzwHGzwfWnYst4QKw998BAI+d9xzgGrngYZ60B44URyRyYLG5iGKrZZGZ4ubBUGFIOtdQ8kdSZnGOXcO7xPsFEckhKeS6IOQASPs1HTN83wFi3sFoYiKjYZWSZXK2VrQut1W3wTIs+KGPy9ZfBtopE9sAOynkKFTFm/Q6w4NF1JhlW2/jWw10pQ5UeWv46xBaERlgOFMI3HWYrRGreuhgtOIkpNJ8hmV1jjzfzq4KRNvOMY1WjlxEyH1z3zcxWhIKo+xSB5oXmM6Q84IwGSsAmVgdKHRWyPkhlxRytFAlDyGni5HQEXJVS1PqQI4BqWXV7kxGFW0TyynLanelDtUCVci1yZDz9BlyqlLcNm383OTP4d/2/BsA4Ma9N2KL+TKpAbdk5VTI5SHkMlrns2I/KCk+3xCEWNkq4/LJEVw+KSyouqbVkm2AGeLabaSYj0ij443WspqzHCLrPSl4kCbkBp1BvPWKt3b9Gac7emlZZav4ZwDYCeCNAO5kjL2mh8vTxxkKOhE9XRRy9KSjKsLOZGjJoC5JtayA2JMR6sRdVQGcCOhITHWS3sfjD0rQGolCTp+J1t6ySvc51bKabttcdPW5X3ktq4tNsiwthRzjhSS3KeABHl3QEzLdgGbzAIBhR8stXTBLd+VXp5BjJOOF2YtastrX2IupQo5au+L16/pywUsn0BY3irYZcq6cN0YJuSyFXFahQww6sYgJkYpCyL3yym3J5ObH++ZwzwE5G+6OQw8mjzkX20ttWu3GdrUS6KMgfOuAlAk14AzAYC0LuEUVcp337wNLaYWc64fSuN7OsqqbxO9d2Js8nqxO4tC8WNZYnTU5Rpo3W1ljDCHGvcPAA/8F3PJXOOvYw8lrXlX6R/yN82H8nvVvODYr7pNPtI6BMWIFnrEKwPoL8KOBZ+K93ktwxBSkw4bXfQt48SeAa/4QW8+5nqwHofqJMdcU+81wYTg1AY0zpCiJNuQMSeOcrJCjSiaOzQObk0ljk88D5orUnNuNZVWFLY2DDIxHxANjHOMZ0XSUkHPIZDY+n46XxvHB6z6IV5//arztyrdJ133x8U5tfTOaDDkK1bIqZcgxF44tCK1e2VUBtdQh+p8qb+1QEENxscNCU9zU4WER150jchGzFXIkP65NoUPyuSZDUDsLYUvVNdOYSUo/gHaWVbpfaSyrQSgpsKjVsGAZyTWb64cZ7oJQypCL8xuBvKUO6RgJHSFXMdPknqSQo2NnjvNLyTa1llU7w6IZq9OpQm77qFhXl2yVyaVCRnyEl2EHjfGc7c9JHt+490aYBpcUjOUMQk6NOOgUeQDI10dyqYO+qV5q6eY0Sy99DOssq8WCm5RTMG7JNvQ2oMd+ryyr9LtICNMtq33o0UtCzgbwKwDmAdwL4HUAzkF0r3sAwC4Avw7gHgBzAF4MYB2A5wH4FoACgE8wxi7u4TL1cQYiS9FxKhNydNKjI6HOVOjWRbctq/Q9TvYThxQOaxrSpONkypDrW1ZPPHS5ZMUMC2bbUoc2F7y2ZFltKeS81Zc6eEGIFaLOixVyDT/AzmGRAUUnTqsFDdgHhEKOkluUwGyupmXVVwg5cxkNLz2hDBWi3Q3cJIuPwQD3hHIjblqVlAE5JgtZWS3MrEv2ZQo1Q04i5AwXMJqp1tisQocYMiEXrZuyI5s11leLkiru09/fK/3+tv0PJI/DpnhdWiGXP0Nu0RdkXomtTx6bhcOZ1sYCUTiEORRyhzUZcoC8nuWW1c4KOWrfnaxO4jDJRN3UavjcOejjSnYfXmN+De+1PoEvOf8T9xZehxd853nA518G3PT/YfLwz5K/e9QW2+MoyXbbUBwDzvsFjJNSg5lXfgH4re/j4+vejo+zZ6AVRYfx0rg0Sds6uDV5THOxYlAiV1XILTQXkrGLHk/DheF8CrmQwzRM7BgS1nfTObpqy6oKep3Z9MPIttrCYEV/bFG7vGPaiE/ltCTpuq3X4c1PfjM2DWySj/fWxL9atBMScqnpo+EFmQq5lGWVTt4ND/ZxIuSoQi4mT6kqtMgoIRdl3i0rLavPOE8cj1mE3NwKUciV2yvkgHjMjFRyOmRbVsmcIkiXOqiWVel4ZkzKkdPZVv2ASy2rFPlKHdIqLR0h52A0yY2MsdoMOaBFyCkKOQQlWXGuUWZR1fZTzl6PP3/hBXj/Sy7CdecKEhZQW1YzLKsa4vCKiSuSHNOp+hQON+6TSP0sQm5VGXIZDbf0Goq6CrIINN3xp96g4JzjA3e8N3luMxmTO0G3HdZKyGV9F51ltQ89eknIPRnA5wB8D8DlnPNPcc4f4pzXWv8e5px/GsDlAL4P4PMAtnPOb+Sc/zyALyGy0P5eD5epjzMQcoacuJg8lUsdJIVcf3BLoCPQus2Byzq5n4wIlDwvybJ6gghn3TboK+ROPCgnGk8sSqtRyJFtqVojLSN9Nz4rQ04KMM/AzLKbBMoDSBRyDS/ArhFhOetFjlxKIach5OgkIIu0aoe6qxByjGO2kc4qkxRyJpOIr5I5BB6ISXaskOs2Q44SHhTMaEiKFWn5yT5SVAg5ICIxVfI9q9AhBlUixROjspO2k772qaJV8v/ddQhTrcKdhhfgwLIgoYIVQbIcq6mlDvknlQueIOTG2OXJY6N4SLI40YlRibR9BuhMyE3Vhf0sVsgBQI20p7ZvWU1vJ0rIbS5vQXVlL55v3IY/sr+Aia+8GvjfF+Dl374G/1z4C7zT/ixeZn0HlxgPo8Lk995OiOKH7BK+HVyMj/m/gAPj4rjb8KJPAC/7HEY3ifUz40bqxaYfwrDFPrapIsdKbyhvgMUiMmKmMSONOV7gJUS+wQwMOoOpCWgxUciJvxsqDKXI7BiSQq71EkrqG8Wja7KsUtBxYq7mIiC5Z7alvxGhWlazQutjyAR89N0Mg2GU5MjNrrgpJVyMtgo5w4VpZufNrQW6Ugd67qkwUS5gONHYWCPlEuAFXLtrXUJYLjf91E0AAJgnRN1ILoVctL79xTQhx7kJhE6GZZVecxGSqLU93SCUxgs1Q7PQQXXthSF4qF/+rksdWufuWD1J0fQMrC+vl57LalnNQ8gVdQq5PBlyhJgu2QW8+inb8StP2pqyTGetN3pcqBm30XMWfn7y55Ofdy/erJRu6CWs6jm1W8uqpJCTbqTrFXIUOlKdnjcXm4v40kNfwlcf/Wry3G9f+psdly+GRMi1nAvxjT6gdwo5HloAt/pz1pzoJSH3dkQqud/mPPtWYavE4XcQKeL+hPzqT1v/P72Hy9THGYisltVTWiEX9BVyOujzy/KvHz8IpcnwyU7I+WE2IXdSKeRO4WPtdEEo2SB1ltU2GXKrsKzG+0FWhlwjaGifp5heboIZ4vKBt6xfTS+UJtO07XO1UAm5JEOulwo5L5QIBACYbqStkVIjLmOSXXXAGpYIuZjw7DZDjirXJJj1zHw8OnG2TC7dRQcAw1xqa1nVKeSGHDHpixVKOkLukq3DiW3JDUJ8/vaoGfM/7z4EbgtS068RQk5pWaXbrFOG3Jwr7KTVUBBOZvGQtA0lQo6oAkK0J5yX3WXU/OjY4KEFHggVBCVEc1tW6/PA3u9h79Q9yVNn/8uv41uFN+OjzofxW+aXwPbcCCym7aEx3MIIsP0a4Io3YPvT35E8f58ziNd6f4z3+r+GoxDrcEMlUjNR61zc5Nn0QzBbWIsnKnKLo2VY2DQgSLoDy2K5VHWGwYw2CjnZZuorx04MQ6PMok2rRvEgmOEny5alLMsDSjocXWwCpBmUNpdSpAg5qxMhp7eoyzlyLkp2vgw5tWXVssW+VrGOk2U1LnUg48qgKRSusWXVDcU2PmtkFENlG9WiUI0tNtJK0fkatax2VsjF43xQOwtDjjxO8aAMgOkVciadU4gxxQ99hDxMKeSo1RCQx6HMyJUwvfwmM3MRJVQt3U4ht9L0sXlgs/RclkKu09gJRJEGPChLEQKRZVWvkEtaVolCzjKyY+11hHUQ8oRsZwyZRR7Utnrv/K0AGc+zCLmCLX/nPJZV00iveyD7GopmyFF0sqzeeexOvPuH705+fvGuF+P5O57fcfmSzyXHfiNogHP53L6amxO678LD6LmTPZv7ZEEvCbmrAMxzzvX1VwSc872IrK1PJc/dC6AGoF+50ceakNWyeior5CjJdLLbKh9P6Aouuhn8Uw2t/sm9j6h32phGBfB4Q7fOTrRllXOO2x6ZwZ379aqgMwF0XzE1GXJ1LwCPLURtFHLtiB96sR2Pu7HVUkUehdzUUjNTIUfVAVkqvG6QsqzGGXIWzZAjBOZqMuT8QCIQAGCmk0LOYBJ5NuSMahVybsYEPQtUuUZVG8xskyFHnm+E6WxAZi2lboCsTiGnn4i99qnbk8c33LYPrh/ihh/dkxBVLCwjbGRbVun43m5CteguohZE+xQPLdj+tmRSwswGjIJQttGJUYlkyHWyrB5eUfPjxPJQgkJVyBkIsYMdwvXGbbjq0b8B/vGlwAefCLxvEu6nr8ehlkKNcY7Jpp784aaDe8Pt+Bf/WvyF90r8mvt2XN74W9z50juA13wFuP5/Yf1Vv52QUtyotcoTQsw2xTqNFTXUOheTxymF3ICskAOybas0P26kEJEjKULOSivkhgvDyo0HgzwWnxsT3pTUt0p7k8dVp7qmEgNVPRNPRIFsCz8dDx3TkS15gUY5lRFeL+XIrTSzFXIdWlYNQv72NENOk3NLrZpVe32rZRYw7AWAuZJV+7KtEbFLSTZKvsWYo4RcqbNCTmwzE1dtkHUg3I/WlW5YtZXcVJVk8oJQypxU16Vqb1bhB2FyI4pitDgK0+hMjOmKlmY162u56WPjgBg7C2ZBbt/0842dMSLC3JBuNKgZclqFHBk3bSObSNVlyKnHRNYxfNn6y7C+FI1dK/487MHdye8GCxkZcqtQyNlmlkJOX+qQVYSgO4YpSfatx76V3OA8e+jsrgsS6LHf8Buo+/XEcu0YzqpuTmjVfi2lsKsZz/pIo5ctqwMATMZYgfP26baMsQKACgB1K/kAVn9W7KMPyAGap4tCzusr5LToZO3o9u9PdrIzUCYfphJcfSKgtaye4PV44+6j+M0b7gAAfPE3r8aTtqeJgdMdlCeJVQqmEbXLxdus6Yco2iZWPHkiL1tW2yjkqD0mEC2gMUpWKXmvPBlyU0uK+Bn0xAAAIABJREFUQi4UGXL0IlUlELtFEAYSQQJE+W5AkG1ZzSCt2qGpUcjNNdOEXECJdkNWyI0WRyULU2JZ7dJSRImyyeokds9EkxJm6FtW/SBMzjsGA5bcNLnNrKXuLavaDDn9RPN5T9yIdw3eh2NLTRxbauIj334Idx97COXJ6PcTpS1Y9AVBNtOYgR/6idKCKuTaraODS7T9dAQuB84bPQ+3H7kdAGBVROEBJRSLlgPOGRjj4CxAEAaZk2a6v4W+rHRJCIr6PMqH7k6eLxg1/KjwOpRje+nR1r8WHrMt8NY5YJMfoMCBo3wY94WT8Nadj5+/7lnAxBPBxnbiN953M44syipV2m5rMAOT1Unc3ypxYM404I4i4NGyDRWGkokaJcdj8rjpBWC2UFmoCjkASTELIBc70IbB4VZTIN1PFtwFbNZkyA05Q5iWzonis3TKLMmyWphOHq/FrgqkCXGaIaeOrTEkQs5wurOskkn9WEUudsibS6W2rBqW3LLaK+haViXLqlPEluKWxHptFI4l5wDOGZ68LdqPhkhzqi5HjravjlRyZMiRdXjFhutw42NfSn6OFHJ6xRXd1l7A4RhOcm5rBs1IIVcU5zpVIddxO4d6hVye/DhAKVpqre9ZjWV1pelj04ggzdVjoBu7PyDiMILadhjVexH6ZYTuuPR9HUPsq4lCLsinkJMsq639R85VzKYOTMPEs7c/GzfcdwMA9djPUsh1T8iZGrswoJY66BWDFJ0y5GIUzSI+8PQPdE2gqQo52rA6XBhe1c0J3XeJb0y4J7nQ4WRBLxVyexARfK/L8drXIbK3JkEwjLEqgCoAfdVVH33kRCAp5Ezt86ca6IDeJ+QE9GRQ/u2s3qE86S2rUjgs06oAHm+cjJbVW/YI0uNHe89MlVyQ0T4oX9hG2659qYM+/yR637RFg1pW6cQ9FyG3rCjkWuHWTS+UrFh0+VaDqfqUZJUBonw3Zi3LlhI7fVe+G9T9ukQwAlmEnHhsmrJCbrw0prWsShOmLjPkJquTyWNmNtDQfDf6XMk2MdtMl0JECrnVE3Jxy2W5oJ+IOZaBV14llvXDN+1JcqYAYOvgJAAraUoMeSiRmXlLHah9mXujaPohzh09VyxmSbT6DhKLk2OZABfL3m4fP7xMCDlvGOswh+uMO/Hb5r9j2zfeAHzoIuB9kyj+w4tgtI5dzwBslvGepoO96wTBNLnuifjkld/Alc2/wWu8P8aPd/0ecNGvAOufAJg2to2mSZaSYsui+4XhTEkW1A1lEb5PLavx+nb9EAYh5NQMOaA7hVzRKiaTPD/04dit41VRyNFxTlLIaW5Wbaxs1KtP1tCwCmgm68SymmXh796yqlfISZbVlSa+88CU1NIZQ520qy2rVJV2vFpWY5cKtWiXbEPa78wiUS6HDi5v3UyjhNx8B0KOvjYLdB2eM3iJZAeNCTmdZZXePFhp+pIyyAu9lGW1XYac3rIaJmVGFPkJOXpObinkNJbVZVe2rFJFKiBILyCvQi5aL80jv4TG4Rehvu8NAHckC62uTIBmyLUl5DTKQul6uMMyUtsqRbWQM0POzKFO1DTcAkoxFr0Ws/Ir5NTtAwBvv/Ltkg0/L9TtQO2qQ8X053T7njF4axw82YUOJwt6Sch9BpG67YOMsT9kjKVu0zDGioyx/wHgfwPgAD5Nfn1V6/97e7hMfZyB8DIy5E5hPk76Ts3+4JZAd0HTzeCv5kKd9IQctSEaTBtc/XhDn+O39oX54h0HcMGffg3//bM/1oYft0McAA+c2lb1tYAStHRSpNpWgTTBlZUhpypB5LvxaYVc14TcUlPKzeLEstpLhZyaHxeDWQtwJMvqGhVyQXoyPudOp55TFXKU1NpQGQeoZbWZVsh1a1mVCDmjri11kCbNTrrQAYiKMNplyHW0rMYZcnb2ZOflV2yTJkdxEyMAbBmIvgcnKjla7JA3mJyqtUJ3FE0vwLkj54nlNMU+SZffNg2JkNPasjkHZh/B4X3fTZ56fXgrflR8Ez7tvB9vtv8F44/dCMxHCiEGoEzGrJrBcJQP47vBRfjK4K8CL/4k8MYfAG8/hH1PeUPyusmJS/FIXVx2bx6WCZhtY+lJXlFZ79ur25PHhjMNw8og5Ag5MN2gGXJiYkftcDG6UcgBct6gYUX7ipTnV6imio5iyGUC0f+MMUkll7zPGhVy6uR9NZZViZDTlVVljMO01OHAXB1/+h+7taUAbTPkDO+4EXI6hVxNIuRMbBvcJhalKMZmxovYMR4tCyXZFnWEXF2szzylDpJ7hht41rZnJT/HtksdITdYEMux1PBTNkw34NK6HHBkQo7uK5mRK5rtN17sXOgAyEqx+Hp2ZiV97l1p+hJBro7Vq1XI8aACb/4qhG40XmSVOugy5NZqWW2Hi9ddrL1JMKwhugBNy2qXpQ6SQi7UL2dWhpxO8aYqLZ+/4/l40c4XdVwmHWRrcnPNDatABiEXW1b7IpJc6CUh99cAvoFI+fa/AEwxxr7LGPun1r/vApgC8AEADoCbWn8TI1bW3djDZerjDASd3BROG8tqXyGnw1pLHVSi52QPH5Xsg6ZKyJ1ECrkerMdP3vwIVtwAX//ZUbzzy7s7/wHB1LK4AD3ReXYnCmpRQAyVZArCoD0hp5BFFJaSpwPIGXJUSRO3ebVDlkKu4YeSjUpV9HULNT8uhmEvKpZVuQSDd3mMNXl6Mr7gdsiQU0odJgbGpQm+rmV1LYQczAYafpD6bg21YbWuU8gtp+zp9HM6ljokCrlsQm7dYAEvuFhMpCghN9kikLgn3pMWO+QNJqcFA2FLIbehsEP7WjoxskwWNcm10PRqwNHdwE8/D/zXW4FPXQ+8dxvw4Utx6KGvJa/bGehtjDBsYMOFKJOJ2jPc9+DK5t/g1d5b8X9LrwEu+lVgw/mAaUsNq5PVSRycF8qcjUMKIadRyKUIuaHtYlFUhVylvUKu4QeSQm5jRUPIDRBCjqxzqr6MFXKATH6aZkzIibFpqDAkHTv0xkPWuVGnKFEnu91iNYTcWhRy9PNohtw/3f4YDszVJYVejE4tqzhehJxkHY7+p5mJJceSiGCTEHJFs5zY5ySFXC1NyM2trK5lFYjW7cvPe3nLUsngL52fWvYYg0VxvC81vJTayPUDKUOua4VcRstqnoZVQFbI+W1LHQJcvuFyXL3xagzag3jFE16R/I5zLi1bHgW2OpYky0MVYZpSB6qQa0fI6Y4Pj2aztbGsAhEZr1PJjRT1ZLx6vshFyGnswoBS6tChZbVoFrWxB9sGtyVZi5PVSfzPq/7nqnMv6XZoBI2eEHKmYaYVjkmpQ3/Omgc9y5DjnAeMsV8A8E4Av48oI+4azUvriIi4d3LO6Wz4lQBeBXSoquqjjw7wJMsqvQt26k7M6eDuriJc/HSFLix0LZbVk53sDJQ8L1kFcBIRcl2Qolmgd3U/f/tjuGjLMF5+xbY2fyFAFXJBD5blVIScNyj2k5JSVKCzf8qWVXoxKV+USgHScakDUchRJc2aMuS8QJpA1v06OOerviDNVsjJhJxpMNgmS9ZBnLmXB14Qghtp4nDRS7edqtmQ1LK6eXA9eCAIC22pg9V5PVCijE6AmVEH5+nvVlcJOY1CjlmL8F2xHJxz6XWUuInRTalDjNc8ZTv+9ScRgWMUBKG5Y3gSwEGEvnjPLIVcO9sVVWtxdxRNMwDzN4CHVtLEmSx/rKby6qhO34UB7iPeys2PXA409WTxYUt8x02+jxVewH18EveG23H+ZU/DFVdfB6x7AmA5KP/7C4DFvQCAeYPkkTXlZdm7sDd5vL26HYfnxXG7aVhWX2gtq04bhVxhCqErSACqkBstCTXNfHMeQRjADZcSdWvJKmtVZ1SRc3D5YJK5JynkyIRQzhusAxhIlzpwsb2tHITc2UNnp5ZrrQo5w2CwDJaMgVKGnJuRIReKqY5t2imCSEVWdMBYhZIc0Wt0ltVOLasg1uisYojVQLKsajLkSraJbVWikCuQtmMSuE9LHfQZcqtrWQWi675zR8/F1375a/jXOx/Be++LSH+dQm6AEHLLqmU18OAGIcw2LavS52quX4OQJzeiKPITckQhF3IEIddafJebPkzDxCee/YlU9qVP2ktNg6XO+zqoY0kMp4Nl1ec5M+Qky2orQ46qxHMQZs856zn41O5PSc9lEXIqAZfHtmsZaTIUUI7dDi2rWfmNY6UxfPiZH8YdR+/Ar533a2sizQ1mwDGcZAyi50ydNTYvCmZBUjyKDLkz8xq8W/Sy1AGccw/AnzDG3gfguQAuBRCPItMA7gTwNc55qq6r9bd99LFmUAuD1LJ6uijk+ncbEqzZsnqKZchRcsQ0mDa4+vGGqyFA12pZ5Zyn7oL/2X/sxnkTg7h0W1p1o/4tJeROdJ7diULI9YScRLq4AWq+HPQOZCvk1JZVWSEXvW4tGXLTS02grFHIeSFsw4Zt2PBCDwEP4IZuZihyJ1CF3FhxLCHAVEIOAIqWCa8VPN0NIdfwglShAxARciEPYTC9etsyGWZrgtRaVx5HxRSTuoVmWiHXScEQhIFkJd1aFcQIa6k5mp783dRJs56Qk1tW63492c4Fs6C13qRJFqCSMZmLceGWIVyzaxy37DkCRpo8d42eBeAgOCHkaNMqHd9nm4dhr1S0ZQOyQm4MTR7iyIKLsLkBZkkmbwd/cgNw47uBY/dhJw+wcfMEHkY0KW+GGZexpVEcLpYR95i9pfYWmAOX4pHZaF392YbzccXGs5KXS5MtolxaacoT+L0t0g6IFBSH5oWSeJOikNuqU8gpk0wpQ86ekRRvlJCzDRvDhWHMN+cR8hBzzTm4bBYxNTFRmdCS5WW7nBxvfujjSO0INg9sljPkiKpSmhwaNQAVudShMAQ/FC0XhmRZFX9Kjy+dQm6thBwQKa78sLV9urSsFsxCZysjCUaXM+Q09lSnDHVU79Syyo0m0PrY42VZjc9JkmXVMZQbBOJ3Y2WxXTqWOpDn8hBy9FwWj6XryuswbNcRTVezWlYNlGwTdS9AyAGTic9qBk14AYdl5GxZ9XTEK9cSqnkz5NSW1fmaC92lISX3VUVWt+o4IFshR/dV2yTrKmxlyAV5W1bTx4dMdHW+KXX+6PnYWN6CwzUx3o+WshRyaoZct6UOOVpWNRly7cjwa7dci2u3XNtxOfKgYBXgutEYdLQmxtDVKuSAaByjJTaxZfVkn1edLOilZTUB53yRc/4FzvnbOOevb/17W+u5FBnXRx+9hEcmCM7popDrt6xqoa+N70Yhp1pWT+51q2ZN0WuQbu10vYKn3QZrW481N0gRaW4Q4o03/EQi23RYbvrSfnEqW9XXgiyFXFGxYery2OqePkNObVmVVB0hhxd4aATRZMRkpnRxtzqFXIvoaB2nkm11DTlyVCF36fpLk8eGtQhHIR117W550NA0rAJACD9FbqnbSm1ZHSSqstVYVhfcBXBEnzFUGMKgPQiLRfdjmeEBzEdDGQvrSs4TVe3FMKyaZDmirxktjmpJmYpdScjISFEVZKorKP7mFZfh9543Csai772xshGjpUhBQwk5OrmI1eRm+SG88usvwnP/9bm4d1qOKQ7CQNofTHcQZ/sPYWj3DbjMTV+uVvd8Ezh6L9AyeDhkeHEZA6pbgHOvB57+VuBlnwf+x254b34AU6ylnuIM+7xzsG5ITNTryn5F93OaqbhMJtGL7mKyHzmGg4q5DiutbVayzRQpoSrkbDOtfBl0BhNVIzMCmOW9ye+oZRWQ1Y/TtWkETOzTmzVZTTF0OXJUIZdFyHGjBhhusv1LVgkFsyDdaJUUchk3q3QZcmtRhcSg15rUsprVstq1ZTUjh2p8QJ7UWwbDrvE0eaOS49LE3/DAkU0irQWGhhhVLasbKhu0N1c2DIjtMlwSxKN6s67pBwnJZxkMAxklMRRZisQg40YWBbWtWkwsV8NvIgiDJHOSgaXWJbVCZjbUr8myKsdI6OyqQFptSyERcjmUYUC6ICaGREBpLKtUUdW+ZZVkyHkxIZedb6sDYwzXbvq55GfOGUYyCLnVZMjZmhuU0XLK2c8xdPt8LxuO26FkivHg6Io4Z9JIiW6hfp94HFxNIdaZiONCyPXRx4kEndxIF0n8xJEWawVVIfQJOYGslqq8SCvkTu79Q7UP6gKTH28cj5ZVegd8sGAld8ePLDbwps/9pC1xqhJ2Z6pCjk4sqPWm6MiW1RU/PWGkCjlXIn7UllX5ApTmxw04A1qLShYaXoClpq/PkGtdgEvFDmvIkaMEzCXrL0ke6xRyao5cXmQp5ADZIgLI+yhDKGW6jJZGUS0OgPNouZpBA17gdUXI0fy3kcIIGGOSlYoZjVSxA500F2xDmyEHAB4XOWNUhafLjwMiu4yqkqt0sKwCwGDRxiVnicnbZHUSRduAwSBZVnUKOXvoTgBAwAN88cEvijcNPEzt+24yKawGwL3OG/EF9lZct+fduN4TGW0xquRc7A3vQC0UtrrGr30B+IPdwMs/DzzjbcB51wNDW3CsPoWQt+yE/gDALUwMCdJGXfd0P2eSQs5PrmH2L+5Pnt9W3YZjS2LivXG4mCJDxwccadKcpWiRcuRIhtxEWVYWUsXOkZWpjoUOMXRNq3R/lzLkyH4SshqYuZL6nZq/qHtMSbt1pXVgoTzp7YVCTrrWJJZVOiZSqKUOhU6EnKRaIpZVRSH3umvOwlhFKRIwipIiF5Atq4x5CJkg5HpJCugUcg2F7DeYIe0XMahltdpGIbdQk9VxeaIM5BINsX/QfUVnWQVkQo4Ro1nNawCKOk5d7x0VciFfU8uqpPwLQ8wQQo6qDJfbEHJ5rf4UJVv/Oqos69Syml8h17Ksdlk8AQDXbX22+CEoZ2aLphRyOd7f1ER4ANnFWFpCrod28Xag6jx6E2utllUJcctqf86aC8eFkGOMlRljL2aM/SVj7OOtf3/Zeu7x2dv6OGOhKjroTa5TVS3j9hVyWujuMOoslJl/r2bInfQKOVkNIKkATtCiH49SB3rBvXG4iA+//FLE18a3753Fu/7zvsy/VQm5U/WYXyuyFXKy4kurkMtoWVUVNfR9Qw4sNsTkc9Ae1N4Rz0K83XQZcrEyLW/T6t/d83f4xS/9Ir726NdSv/NDX7ojfPG6i5PHzF5MfUdagtFN22/TDwAjnc8HyHekAXlbeXw5IW+GCkORPbBkJ/YPIFJHuZKFjeHOY3fiD77zB/ja3vR3poq8uE1PyjYy6ymFHCUfVctqxRITZY8J0kb3OTpIBIhZb1vqQEFLDLZXt4MxhopjgXvpDDnOeTKxpM2Ntzx6I/hX/gD45DOBd2/GY//8MvGeXhMFJiaqT3DT+2z1mj8CXv0V4K37MfWa7+PhUORfNR29hZpapLkXEU4bqmJ71hRCzjGEeqHgeMkk2w/Fd6J21cnqJA7R/LihtFWYMSap5LIULdQ+SJFSyBGC4FhtGgaxEm8ayFbI6Qg5almVWlbJ5NDHSqrQAZAJFFOyrOobyBljMH2ZXOwJIUfHDaqQy8qQo4Sc0V3LKh2jyo6FCzZFyz85VsbvPWtXalJP96cYEiFnuAiOk0JOl3OrtqwC+v2OLodsWZWPyzmJkOtc6ACkM+TUZVSXnWKwKJaFQTxecZsd22olQu54KOSUHDOqkKPHfztCbjUKuTyWVTVvD8ivkNMdH91aVgFg1/BOuLNXgYcOgrlnZJK3KYVcDgWerSm5ArKLsXQZcr089tqBXpv10rJK0besdoeeZsgBAGPsjwC8DUDWGW6RMfYuzvkHev3ZffQBpAdp02AIW4NjwHnvd/rHAVT1dbKTRo8ntHbJrlpWFYXcSU52+srFou7u8+MNrW14jUUK1JIyXHLw9HPW4c3PPhfvv/EBAMCnv78X/+3qSZy9biD1t7RhtRfLcqoiu2VVDu7XlTr43IcXeLBNWxp71ItexuTSg7mmIGcGnUHpLmxsZc1Cst0MjUJOZ1nNUMgtuov4yJ0fQcADvOuH78Kztz9bUilM1aaSIOmx4pgcKK6zrK5aISdbVjk3ErtdWiEn3rcJsQ5jUqtatIFmCWjVByy6iyl1wJ9//8/x8MLDuPnAzbh649USmSEVLbSIlE4KObll1ZDeY8fQLtwz01KdMWHrzEvISU2lVj13ThEloWIlV7lgYnmFEHL1YwDn8Kb34heMH+ACcw8+VjgSx2PhmL+MPXd9Fud40X52oCQmQJv9aL/YF67Hg9ZO/GBlM4Afysv+1N8HWhMPy2sAtGU1o0n4yMqR5HHoRxOe9YPi2GhHyNm2B7tgJWPiStNH0TZTDauHSMOqWugQY+toGQ8cjUjzTIWchhgZsAdSE0VqWT22MiM1supy+mJ0tKwShRy1T/lYBjPkQgcgfU6Mwdqox5k3ATiPJD/TopHVQrasim2bu2U1gyBKnmujiP2H112J7zxwDNfsWoeyY6XsqTpCTnqN4SKAIEYqVi8tq+1bVsuOaI5UQRtK25U6SIUOpc75cUA7yypZ9jwKOU4Vcs0kkxPQt/d2UkL6mgw5x3AwaOdrAraUbLwZhZC752B0nKp5lBS0bCK3ZTUjdoCWLaxNIbd2yyoQEdjusRehefSFGB/Qj5OAptQhQwFIIWfIUcuqnkzXtaw+bpZVcvxTVTm9IdItsiyrfRFJPvRUIccY+3sA7wEwBMADcDuAf2v9u7313BCA97Ve20cfPYcckG3IbVun6LggN/bwU7qgopfQZ3B0kSGnZPec7HdyfMU+SPNZThQhp7P59lIhF1tVfuu6s/GkSTFZ+9khfRxpXyEXQVKOkIt0qWXVCzMzjmLCy8uIAIhB78gvNLItq7kVcky8TrSsaiyrGQq5+cY8gla+13xzXmqiBGS76uaBzRgpjMBAtE6Y2QCYvJxUIaeqyNpBtazSxkp6RxqQ1a2NUBAbMekxVLLBA3EBHSnkxB8ZLMAjCxHB0AyaeHDuQen9OynkmFlPkY100uzYfkKoFsyCRGIGRvcKuYolPrtYaOZuy5UIuRZxVHEsDAYcrGXpXXKXUH//TjgfvQQfcf4Pnlb6FkLl7b9XFhOxAwOCWLrbvwoXNz6Bp7sfwutrv42/938JoSt+XzAL0j5tG4ZkL4uDylXICrlhOJaBEaLkaXhtCDnLkyy98UR634KsFqQKuY0ahRwQKahiZCrkiGU1Bi10iEEVclP1KVkh1yZDTlXI1f16sm85hiNNFClR5vKVVKEDkK1oory6em7krmyp7U2pg/hASshlja+qZbVThhw9p6o3DUYrDl582Rasa5G86qTeYmniQbWs+vw4KeQ01mG1wRnQE3L0e7QrdViNQo42U9PrvizFJQUl5EAIubrfAGtT6AAoCjnN+cTTtKyOl8Zzj5GUmPJDjjlCyG0ZFcfWiutnRvjQc0GWpVNFtkJOLDcloJKW1bwZch1KHfIScoNFGy978lbYponXPvWszNep3zufQi7LsqqP/ZCajlvQFSEdD9DzWHy9BKzNsqoSjH2FXHfoGSHHGHsRgNe0fnw/gAnO+dWc819p/bsawASAWBn3asbYL/bq8/voI4an2PrMjHDfUwmewiT2VXIRet+yenLvH/LkQyabe0E8HZyv49hieyWTiuORIbdILrjjC3HGWGLNAYAjC/rlTGXIneTb9HhBCqeWFHKEYMqwrALCttpOIQfId+QXmtmW1U4Zcsl2owq5hJBrhdWTi1Wdsg9IT4DvmrpL+pkScpsGNoExhpIhyCOXzUuvL2rCpPNAVciFDUECpAk58b4SIdciPaoKIbfkLknHXS2cTUobAHRPyBkNjWWV/GwKlc9ocRTrSuvE91qFZbVsiePYcTqXfcSg5Orkw98D/vU3cEPtjbi7+AZMEIJjyhXb8L5CeoJ+66YnAK/8V+Atj+DA+dcnzx8NL8ACZNWt6QtFl0rc2JYhTcizSOfDK4eTx6E3jGrRTpRBAFBzZfsYJVAs25NC6ldar1Utq4fJeLh5WD+po5a1YoaiRUeMqHZVQLbQzdRn5Qy5Sr4MuQNLB6TcweHisEQ8UIWcGy5LltV4W6gxDjF0VskYYXO99HPvM+RIy6qb0bIaKi2rHS2r+ckHdVJvawg5GuoOw4XPxbrtaYacZjtIDc5tFHKU0KIKObXUQVLI5WhYBeR1KFlW85Q6FMRncKKQrXtNwBTj2YCTVvAXOigho/OtIb1v3vw4IJ3rSi2r6wYKyfmf87QyN1muVWSzZZY6GHqFnK7UgbawqtBlyLVrgG+H97z4ItzzzufgTc9IF7zoPg/ImyGnL3XwpTHq5MuQo1iTZVV9z1gh15+v5kIvFXJvAMAB/Cnn/I855/PqCzjn85zzPwLwTgCs9Td99NFTSE2UJjspgu/XCpVU6A9wEdZa6nCqZcip9elyTs7a9u0f7Z3FNe/7Fq56z0249+BC5z9o4XgQcvN1fRDxBFF+HM5JyJ2qx/xaQb83VVKqltUs66cg5LIz5AB5YrNAWilTpQ4Zdr4Y08tNAEFi64xUa2brb1vNirbY/lnL3YmQo2qlOOuqaAjlZYPPSa+nF+aqkqkdUgq5prDxqYQcPV5qPil0aJFaQyUbPCQKueaiNFYt+dPS++2Z2yP9TImyuGxBLVZotCl1UAm59WVCyJlim+cpdQCAoikmqo7d5gZAGAJTDwA//Ucs/b/fTVpcnZBj47feBdzzL9gURttzPSEUj1kmQqeKm4ML8S/2rtTb/sSdxsrk1UBlDAeWDyTPF5DOaRo2BVGgWtAsg0n2soav/y5pQs6SCDF1YmxAHDem6aJCMvbiYgfVsnqQKuQyLKvbiEJuMKOJcsvglkQxGkOrkKOW1doRGFZrH+EG1pH9Q/d3MWG05C1JxCK1qwKyWqMRLkvHk86ySu2RRptzY9iQv09PLKtShpxsWdUpkegNCsd0MgmiGBIh14EcUCf1JtorcZjhwjtOCjkoumZCAAAgAElEQVRDE6tB9/eYmKaq2xjUslqyzYR0afqhNBbPkxt4IzkJOUqw0Bux0nkzh2U1DMn51G9KCjmdzZS2hbZVQpIcuW4IOdmKyyXL6mjFkcn9jBw5ulyFnMoznUJOzTl2jLRCjlpW4+ZvHaSW1dbyyTmq3dEZWYq+ZFnN7gk5iQylubCBfN0eQ0fIPV4Zcrr8OmBtNycKRr9ldS3oJSF3OYAAwIdyvPaDrdc+qYef30cfABS/vmHIF2an6ORcJZn6nvwIulDckOcnYdIKuZN7vdL9QM2QWyvx9I2fHUXIo/X31XsOd/6DFvS24bWtR2pJoXe8aTbS4QW9QmpayZA7VVWxa0VWhlwhp2U1JuS8Dneh6UXoUlMQN1WnKhNyGXa+GFNLTUkdF9kfWGs585c6dKOQ2zywGQBQgLgr3AzlNlF64d6VZdUPJEVP0CQKuTalDrVQEHIx6VEtWqlSBzpWLfpyJl07Qi6vQo5mytGcuLHSGNaVBXHFTaGKzKuQKxhi0mFSQq42Czx4I3DTXwCfeSHwvkngo1cAX3oj9t3z+eRl23xPooxcbsLxxfo59rz3Yv/rf4ZXeW/Dbkc8H08I/dDHDw9H2XBxjhkAFJEmkjaXhIpCVQ/YpqyQy1KBqqUOgyUbZek4lNc9JVAM00WFTKKXmz6m6lMJIT1oD2K0OCqNh5syFHJPPXsc52+swjEN/OqT062WQJTjNFaUFW46hRwlCR5dekAsOx9uaz1jjCXHHQDcM3VP8ljNL6JEWT1Y0lpWKdkmKeTanBs9v4ygFhFA5wyf2xNVikwImCi0JrwcXKvmbWtZ1ZZV5Q+wVxVyZgfLKowmeCtlsWAW2m6/bqFTyNU1pQ5jxTGJgANkpR5jLNO2Oicp5NZW6iBbVvV/S0sdgkCsq0Yey6pJlV66ltVWGzOxreYtdABkxbofhphdEWPSaMVJjSU6SC2rObLTAL1CTiXJqKXRDVwEYZAouxkYTCObJNNtr9VYVvPCMJhMnnVLyNFSB3rskve0DRsM8rH8eGXI6eyyg87gmo797FKHM/MavFv0Mt++CmCJc66/uifgnC8zxhaRXfzQRx+rhpwhd7pYVhWFXJ+QA5C9HrwgbHtyj6FmeJz0hJxiz2HS3ee1vfdSQ1ycUbVFJ3h++oPXqpBb0FhWAWCiSgm5DIVcqtTh1Dzm14qsbKWSZMEMgG4Uckb6opRehC62UcjlyZBjTGz3ArmDmxBy5GI1r2X14fmHseQuJQTUoRVBjsTEgANBHtVCWSHXO8uqrJDjnCfHr6SQCwgh1yI9hso6y6r4mwVXhDIDwJ75PQh5mJRZ5GlZrbvyd6NWJ8MS63S0OIqJirD8MXMp+S5UIUcVVCpsJiaq2/l9wJd+C3jsh8DMQ5l/s9cWl6uTrABc9FJg8+X4wO4BfOLBCpj7X3DwfQDAlGUiilrzYRRFocKLdr4IX3jwCwCAWw/eiqs2XpWsG8uwUDLGAMgWwyeOPgkGuwj3z9yPl577Uul3psEkQq6hUYFyzuVSh5ZCrkxy4VIKOU4bMJuomHKG3L5FQSJOVifBuWzh17WsApHK4z9/92lYcQNJKaNiorQVUw3xGZ0Uco1AHIsOzyZiY2wd3IqH5qNtfc+0IORGC/LfUrVG3V+LQk7+fD/gCA68CmZlDz76y2/Inc/VDpRQYyxSdzVb2XjL3nJqoq2WOnS2MrbP8qRQP8vgaSVONDYzAByMiffutUKH8iTx9XdDkyHHGMNkdRK7Z3Ynv1MJuqGSjenlaFxaqHtJU/FCTX8Drx2ySx30N7IoqEIuCOkNG7dzqYPdfjsn5+1wdYSclGMWcMwsi3F8rFLQ5lGqkFpWcxJdJUennpfXn5ohl7fQAZDXW0wYrtaymheOacBrFVw4Zuf5hJWVIZdhWWWMoWAWpMKrx8uyqlPI0YiA1YBaVg0YSRai28WNzDMZvaSUpwAMM8aywyNaaL1muPU3ffTRU6iZR4YmVPZUg0oU9Qm5CFnrIS8Jk25ZPbn3D7WwxOyh+pPeLT0w1wUhd1wUcmJZKCFHlR9ZCrmUZfUMvTtHdwe6n9AMuXqHDDnOecomrYJehC6RvCQ1Qy7LzhdjallWyBUlQi5/qcOKLxNyHFxS4RxckjPkAMCGuBCtBapCbnWlDkvNFTAj2o95aIH7Q0kuUN2vS+2L9Nit+Rktq6Fc6kCPuzlXVsjV/bqkBNQRZXKpQyOl0qLHEVMz5IglkVmLyT4SW0oBjWW1Pgfs+QbwrXdhx/3/nDw96d8F/PRz2WRceRw457nYt+uZyVPbL/114MWfAK58A6aHLoQLG9wX5M3R2lE0/QBG4SgYi77XloEteO5Zz01ec+vBWyW76qbKJpTsNEm1bXQQNzzvBtz68ltx/Y7rU783ICaZdS+9j88155IJFw8KQFhCtWRLrYRqwy0ogWLICrmVpi/nxw1NYnq5mRC0I2U7s/EQiCaB7cg4ANhcke2DOkJutKQn3gqss72O5shRQk5VyA06g4mCpB6sgFlEgdtSz4XKTaoYWdd98ZjGgwH4i5diQyU/2dEOlLhYN1DAIMkP0zWtdlvqsJYMOUNjWWWMwURaTdbrUHnddtBZVoG0bVUlB+m1AM2Rowq5kbylDhmEXJhB8FJQQs7z6fmhCRCFnEooAqpCTlPq0DqOORffY7zYBSFnyN+LrpvRAdmymqWQkwi5nBlyOguo+reWYSW2VA4uXRN0UmZJGXKtc5W3BstqHlA3QdeWVSlDLps4VHPXHi+FnM4uu5b8OPU9i2YFscOhr5DLh14q5G4B8FJEpQ2v6PDav2r9f3MPP7+PPgCk70bo7tCdauhnyOmRqZDzQ0CfWdr270/29UpPbKbBerpvr0iEnJ7s0C9Tm7u8qwQNaaYX4eurYqMeW2rCC0LpQiwMeXIHPUZfIacScrJVLshQmtX9urTuTEWRGYNaMJZdUurgDKYsKu0wvawo5IilIp640IvVrAw5HVF319RdeMrmp8APfSm/LQ6fN7m4EF32Z6S/pW1rahNpO8w3BbEWqdsYuD8E5kTvf6x2LCHF6Hpe9gl5VsrXsjrXlAk5ICp2iIkPSpTpLav1FNlIlaaBIbbraHFUUmwwaxmuH8AymJxVt3QMeOAbwIHbgcd+BEwLW+OmcgnYEJF6i1R1aVjAxEXA1iuALU+O/g1vAxjD3u++BVi4G4AcAB8rzUJPEHJTtSm4fgizJEjJJ4w9AZesvwQD9gCWvWUcXjmMWw7ckvx+y+AWLE6nJ5VbRstR8UcGUWEyMT7VNQq5w8tyfhwQEaylNhlynORHcdbAAMmQW276mFUaVg/maFjtBlsGFEJOY1m1DRvDhWHMN+W46BLrTB5sGRRFGdI+o2TIGcxAtVDFQutYMmzx2ljNQSe7VNGUdW7MM6atBnTCvnG4hDIhY3TFDhIhZ3Qm5NwuCDlVZaNTyAERmRxA3md7r5CTHSqcc23LKiCak7OWhdpRZcsqUciVus+Qc7tWyInP8H2ioA5cMKN9qUPnNt2WZZUo5LoqdZAsq1xSOo+WnVQepQ6UKMxj1QTyWVaBiHz2/ehz6c2zdoUO0XKkM+Q8KS+894QcJU9zEXLk9fT6q10Or0qMnchSh6HiGhVy5LuULLHv9wUk+dBLQu4DAH4VwMtaCrh3A7iFc94EAMbYEIBnAPhjAFcgKoD4QMZ79dHHqiEPfqdHqUNfIadHFoGmttJm4VSzrEqFJT1Wfy4Ty+qxpSaafpCr8l5/N7+HLavEglKwTIwPFDC93ATn0XLSRsG5mps6xoOc+8Lphqxw6pKSIecSi2fBLCRWqrpfV+yq+skJvSO/7BOFnDMo5ZRQW4YKznmkyDLF5KFkrVIhp8nEi3PkjtaOIuDRMT9eGk+WzwzFheiyLyvkZKuMPF4cWTmCglnQFhgsEqLC4NFyh14VRouQO7pyFGcPnw1A3lbLXrrUoVPL6kxTzqQDohy5Z217FrzAw1KLKDWZmSiL5FKHRqrUgWYxNrmwIo8WR1EwC+BBCcysg7EQM/P7Ec7dlzTmlUKO0sefnlqmGFVyTB52RoCf/11gyxXApksAO00oHVo+hB8c/kHy8/ah7cnjmKyiCrljtWNo+iGMorAnnz92PmzDxlUbr8I3938TAPCFB76Q/H7LwBY8rMlL2jrSnuAyYCf9tjrLKi104F6rUKNoSRlydbUs5P9n773jJbnqO9HvqdD59s0zd+bOaDRBo5yfEiKIZCOwMMGAwxrbsGswxmlxgAXba3gPP1te1osxrHHAJnpBFjyBibKQAEsIISsgNELSzEiaGU28+fbtUOG8P6qr6ndOnequ6tvd05fp7+czn7mh+3Z1pT7ne77BCSc2nNUjCrmn6yIhR+37cflxabBDIkZUCjnAU1vKhFxRb0/IUYUchayQA7zz1CfkWIY0sjbVHPT2rsco5GInx22y2NKATthnx3JoZEIySaWQky2rQkaWsmW1tVKZQiaPGVcHuHvlISvCz7pNyIljFNGVkDFElb/ctNpKIUcJOdGymjRDLnxdoWWV7PokCrmGrQUz6bpTFyyrKoUcHVMp83eb56pb2woUnobBDFwweUGbdxOCEnJrUlZfPqOL95JGbxVyKkIuq2eDxTRKVLcqdADUllVLsNZ237J62fYxfPWHxzA9khXyi+NA7yc0Zshqcc+JEHJ9UsgJLctNdFMhVzDCa3c4X02GrhFynPP/YIz9JoAPAnhB85/DGFuAp1Xxl2MZABfAb3DOH+jW6w8xhA9bJi00cUCwESGrfIatNR5iLasJCSE5E2rQCTlLsqxqXcyQo/YFzoGjizVsGTfwJ3f/CVasFfzRtX+kbM9TDyq7V+owKq14bxnNBWTBsaWqQMjJ+XHetmxMEn69EFb6YyyrNcsBCLE1mZsMMtaqVlUqdFAPzOkEoEIUcqVMKbFCbrVuo2a50I3wuBfIpNInwuhEM2mGHAA8fPJhuNwVwvVpsLzmjvnuCixbokIuF6OQu+vQXXj7HW9HTs/h1p++NUIyLJN9oXFvYsbtkPijSr3wHOVYsaOlDqqW1XCQz3GqqibkAFGBNJYdC3Llogq58L0FBGkT1EY7mZsElp9FyTFQae6a6t9dC85dYLtnAZ5wFNZepgMzFwPbr8bjKyNA7XMAgGez08D1vxV9fBM1u4bf/uZvB6TMpvwmXDh5YfD7QnOCKRNyDduFngsVchdMeJPa62evDwg5mie4bWQbDismn7NjrSdIOjLw75xVhS2bnnOu3VTItbGsOkQd46AmBrE3bDxVeyr4fkd5B/79MC10aD9xbIfdYzvDb9xsbPPeZH4S+5f2Cz8rap0TcrJCDvCUcIdwCAAC+zEQljoI4z2dKuTULatJ7mmdgP6tLaN5zBEyRnVPilpW27VvplDISZN6qrik0BSW1W4TAnKpQ5xdFUhHyFEVvVjqsL4MOTfmc5OCEnJ1KyTkGo7YstpOIafKJPW3pX7ypfi166/CNdsuCqIVksBU5LwCXqEDgGSWVac7hJyhIMmoEo6q3Nsp5NSlDq0b4NeL//e1F+P5e6dx7a6JRIvT9P3ShWC7xT1nkBRy6yXk6HivQBVyAz6vGhR09QzmnH8YwIsAfAfe8NYAMA2vvIE1/30LwAs55x/p5msPMYQPWyItNnqpA+c82hI2vMEBaF3qkATRltXBPj8caaVN1WDWKeTB2eGFKr7+1NfxxQNfxJ2H7sTnHv+c8nnKDLkelToAHiHn49lFcQIs58cBG1cVu14kaVmtWo4wKKbWmIhCLmYFmg6EK7aYIWcwIyCAHO4ECioZwXEjGXJ5Iwd/sy3HuwcmsayqJr8r1goOLh0UctWESY4TEg7L1hw4+ZwQVuaJkumWx28B4Cn/7njmjshrrlqhqswvMXAtNSEXqFu1Ghzu7YO8kQ/ebzlnRlpWg3ufVkXVie6LJxY9Qm6hHqqKaO7XiElLHWoCKeQTpABQMBmW1kKV18StbwE+cD4ucsII4FOGhgVyHkw4DlCcBs59BfCS/w788peBdx0G3nIX8PKbcXjiJ4LH2ojvAeOc433ffR/2ze8D4Kkobn7BzcLA3yerXEJ2nlg7gTWrDi0bbvf5k+cDAJ47+1zla20b2RaZdE2VMi3z2LxtCu9P7RVyvmXVQNbQgvO74bgC4eLY4ftzeF2YRK/U6kIz7I7yDsGmv2kkQVZDG2wd2YTGwtXgXIO2/MJYW6fKSlc2NykeKf394tbgvkChUsj5xFvcz+OUwDpTL8Qmuad1giI5T7aO5QUySWlZdcNjltWzPc2Qg6s+JxiPEnJFo7eWVaoGlW2ONEPOYEbkfdCxgK+i55xjkYwXOsuQC8+JuM9NCmpZrTfI54Njtc+Qa9OmG5yfbgGv3/ufcN3W61q9jQg0jUHFI06WvP0iq21VoERhEiIK8I6zTN6pCiEoAZVKISftN9flPSPXfYwVMvj5a87CrunocVRBUMiRc8pqcc+RyxVOZ4Zc3L02Keh7oef+UCGXDN20rAIAOOffAvACxtg0gMuBoEf+JIAHOOfDIochegrZkrDRLasqkmh4g/MQa1lNqpCTLGiDvl/l/BtRIde9DDkAOLK4hnkznPidWIvmVAHqfZ1UoaiC6/LEhNwxqWn11BmkkDu2VIPluNg+ER3AUfsyY6L1Ji+1hlqEwKLNiR4hR9XG6gGvSf52RbKs+i1ivpqt7tSV4c0+IcdoqYORQ87QgwlczXJSW1Z1pgcW1YdOPiSQI1Qh5zoZcJYF0+uwuYXF+mJgQ80psms453jw5IPBz2mLZrAdTqgqyzCP/JKLB3z4K+g0tN63qwKeqtFAOFFeqi8FE3TNDBV1W4pbcKxyDBwcTy8/jbpTx3w12rAKKBRy/kTZrmP58bvxFv2LuEp7DFfrT+CmpVGg2TI3seiRmlNEBXfSMFAb2QbAOyYTO54L/OTfAzETWtcOr+EGjyfkPv3Yp3Hb/tuC79959TtxxeYrhMcERIibhY4cHNTQcBt4YumRoFQji8ngeM4UZ7BnbE/Q8uljW2kbsoao4tw23n5ypDOxOVAGPeeCDLm8CcYYCqaOSpMIXbMclJuTSssK/6bFq8Iker52HDb33tem/CYUzIJw35sqrZ+Qy5k66sdeg/rxV6Jgxv89VZPuWKY9IWfqJmYKM4JCEVAr5FTqPO5mAlKWLrLKDYY+hAy5BPe0TvDKy7bilvsPI5/RcdOlW/AP+8Lrq51lNaNnBBuqalzTSGFZlVU2boxCjvFMoAz20VvLKheIf5mQK2fKeNWeV+ELT34Br9372ghpKyjkmmOEquUE47asobUl0H2YcRlyKUsdag163CTLqkIhJ5QTqJSQMYrPNDB0LTKW9RVygto2rmW1A4Uc4B1P+rpxllUfNEOuXakDYx7h5//9huMKpQ69sKymBb2fiDZ5sk80mbQUr81+KeRUmajdVMgVM0OFXFp0nZDz0STevt6rvz/EEHGQP9C0GOvCRoHK/jfoxFG/EGfdTWqZlPfjoCsP5bYm+tm+3nN7RaGQs8ZCUiEuA6zbpQ6rDTuw3xYzujCom6/N4zBug14y4axegGelptUzRSG37+gyXvHBb4MD+PR/vhbX7RYnxq2CqYVSB9tBg1g/ZYUctUjHTQLppKFKBtj+ZEQm5FQTvkDhw0RCLmtqIiGXUiF3yfQleOCEl4zx0MmHBIUeVchZDoeLMnTdWy88sXYiJOSkEgwAeGr5KSE7i5Jr4faFltW8HrWsUoLbP0c10mZKjwVjDCOZEfh00XJjJRjkMkLI7RrdBVMz8czKM3C5i/2L+5WFDoBIyOn6Gl569G+AfzgIHLkfs04d72rOfR0AC1pIlIw7DmDk4TjjQHOL9l/3ezhraha4573e64xsiyXjAKBuGeBcA2MuHN5A3alHVuzvO3Yfbr7v5uD7V+95NV5/7usjf8svdQAAE2Nw4JGjP1z8bvDzUe1s4TnPm31ehJCbHZlF1nxa+Nm2NvlxAKBr4SSknULODTLkvJ2bzxgBIVdrOMHPLTskHixeQ4GoNOcaRB036ln8KCE32QVCLiANuIGa5YBzrlTJyQo5bhdQTKjw2D6yPULIqSaEfuah8DpOHrbjwtA1YfGHfh7GNZAnuad1git3TODed78EWUODqWuiQi5By2q2nUKuDdFBkZdyGF0nRjWmUsj1uNRBIOQU5Nn7rn8f3nHlO5RqSWpH9RfthEKHhHZVQFRv0X0rWFZjTg+RkAuNv5YrljoIKmT/ddsQckJWYIeEsakxyAERoWU1SakDzWZLvg05UwMdkqmuL0ra0EW1dpZVwLsv+ddG3XKluV73FXJpQcdCdFwsL6RT0JxdoPvXXxx60bJKsw7PHTsf/9r8etDnVYOC038GDzFElyG3C254hZwd3ebhioOH+JbVpAq5jUXIiee21rVzu2G7kX15ZKEqkA51xWTTf66M9exHGtAsq+P++oG/xvcWP4P8tk+AmfMRhZyKkPtxVMjd/uhxuNzL+vvmj6LKxVar/DS3Z6VmC4NiOsFes9dENUnM7CScHHJUJcsqgEQ5cidXvOOoUsj5qNmuqJBLQMhRu89DJx4SLKuzxVAhZzluLFlGJ8o+IffgiVAdB6gVclWHEnLevhAsq5VohhxVyMnqo3I2nNxVrFVYjjeZogq5meIMzhk/J/j+iYUnhAy5idwEsPws8INbkP/ae6A3TxOuObhh/lPAM3cDksprUdPAm/eZUS0L8z/fAbzrEO7SXh48Zs5awUIttMaqSi4oqg1XbI2tLwu/P7p6FL971+8G6saLpy7Gu699t5IUonZOWs7xo+V7w+0xdwrPuX72euH7cqaMcqYcsWclUciZjJ7f0fsPPTd8y6o/oS/ENK3WLYC7/vviMM1w4rzkhCSWn7k1RyyrU6Vklr1WMHQtsF+5PF5xLp+jrj2a2OJGm1Z9qM4blY2KO4Ug89CNUcjFtqwmuKd1ilLWCO6H1LJVaURVoJFShzZWxvVYVqkFWoAbJUG6niEnjVFaWVZ9qMg4QM6Qs5r/h+d+UrsqAGQMtSIxrp2cImvowfGynfA9WG5DsKyqyJV2xCs9P/V1KORkTCoUcolaVhVFN3GQj2dbhRz5rG5nWQXkplUnVfNwPxCnkLNa3HNOV6mDbJUFwubqTnHp9KX4yxv+En983R/jpl2vCX4+FJAkQ88UckMMcbogrzBpXczZOh1QNYYOb3AeGtJKnv8BnbxldWNlyNnCuS2rPzv/u6qB2eGFKiYmw8l+vxRy1K5algg5PxeLMQ4texzPLu0Sfq9WyP34XSt0H6nuBa1ycGgL4+GFNZQ3x1tWrRZWCx9BbgprwIX3+JyeC1a86cCvpgi9B0gZBwsnVzk9JxRQ1C0HhWx7yyol6q6euRofZR+FzW3sX9qP0Vo44JwdCQm5hu2Cs5DwOlkNkzVyUistAMGuCogqqGB73ZBcK5re67ZTyMVZVgFgNJfDScez1XJwWLwKIB+xrE4XpvFvz/wbAI+Q08l+mnjgM8CX/9R7LQDls2ax0LSirmgaJpvXynJ+O762shP38XORu/B8wPpL7/kjW4FtVwIATIzCF0PM1U6hXAuvVXnbZaxZDrhTAAzv3FtuLAuFMX9yz58EROJEbgIfuOEDyhV9ACgQxQdzRoNl5gUrJF8nTfE+ccWmK1AwCsG54pNDWcmelUQhZ2gkT8oVCWfHdQSiktseSePf1+gElhJyaw0b3M0GlltadrLqHgGaTzu72Ybabcsq4J33fq5ozXaU1rWpvFjg4FrjkX0YB5mQK5rFiH0LUE8SuVNAzXJQyhqS+iR8TFzLajcUSEmQRiHntay2vqenUfaZmgmDmbCbeZROjEKO8ygh13XLKlXxy4RcQnupD1XL6mKLBbxWiC91CB8TZ1kFvBzIU6sNgIfTaMttCJZVqkL2IVpWo5ZR4Th3qpBTnB/jaUodOlbIicdTdc+g1zgl5JIq5HzUbdGy2k21a6egZBsdqwvj9jalDiqirBdQlTqM5tZHyDHG8OIdLwYAnFgJr4NBFzoMCjq62hljjS79U0suhhhiHRAmozoTBmkb0rI6zJCLBV0hoxMzK+H+iWTIDfgHB5Xod1P9qRqYHV5YE5QrqnwkQE1irksh1yI/jhI6TKvhmGxZVWXI9Zhk5afhnkL3keq4C7lKpR/iww9+OCA4SlkjCH63HDsIF9eYJihUPEIuealD3ESEDsDjzqEwQy48D3NGLkKGUStWkpbVydwk9k7sDb73mzoBj7zyYTmuUApALagCKWirFXKnqqci6r8GIeTKpme743YJ4N7fW6gvBPsjIORiLKtAtGnVZd77p5bVLaUtOMcM1SVPPPRPWPiPjwXfT6yKEb4j5H5yW/EngNd/HHjH4/iby27B79lvxWedF6I+Gp4TlGjLIHyd+dqpqBKvBdbqNkAVco3wPlO1q7jn6D3eewPDB274AGaKM7F/q0gsqzSjj2Jzdo/wvambuGbLNcH320o+ISdOKFX5jDJMLZzYNKRFi+XGMjj8kPYcfCYttKyKBSvh164QxK9p4blVZc8EX583cR4455JCrluEXFQZKiNiWbXGEmdOyU2rcXYptWW1EGyTK6nGw6/VUSXdyOhKAnoPbNeyampm+1KHFJZVAMiSiX2cQo4riLpulzpoUpZftRHe4+MUcnGgltRlBSGXRiEnEHKE2HFbLGZR+MQWJ4RczamAsWa2JwwlwZzRW7fp0s/zjjPkFEReoJDLtFfI0e1KSrADUYJVdZ5mNDUhl0whJxJyrdpLTwdoqYMd07JqaPEKubyRh66luyY6hYr4W69llSKrUzXjYM+rBgWdnsFGl/4lX84YYoiEkFUdG96yqiA36gNOHPULdOBABxpJbYobL0NOHKx1q9RBRcgdW65hodaZZbVbCjk5E4aSMEyv4cRKXThm/c6Q++sH/xo3fPYGfOaxz/TsNVRYroX7SLIi8bkAACAASURBVJWX6E8qmDkHbP4nfOShj+BDD3wo+P2u6eaki0z0C0ZBsIRGWlbjSh2akwamqcOs6YAzzrIaZsiJCrmslHcnb58KdJBfNIu4aPKSyGM25TcJkyXL4eB2OIEWLasiKbhUX8KBpQORvynnyDV4aFkNVT4aTERVciqFXMSymjcFmyfTqwA48mZIsm358rux9/O/GXz/BLMxTwbGE44LGDng7OcBL/gDZEbODn736dxPARf8NDCyWbiOzEy4PynRRt/HQn0uHSHXcARykRKljy88Dpd75/TZo2fjys1XtvxbRbIQYzeiihTXGsFYJro9L9nxkuDri6YuAhC1ZyVRyJks/vymln9uh+duOR+1rNJsrWrDBidB/Ez3j4cNWw8tq+dNnIflmh0sJBUzemrVURwEe5il/lyMWFatsY4VcqpCB6CVQs7bJjuGQNGStKz2SSG3Yq0Iv+OcR0odumlZBYBzxi4EALhWGY4VvS4AwO2HZVUTj8N6FHJlRanDArWsFlNkyMWVOpBxVCuFXNC0Sgi5NSe83jOaej/Se4w8duKci4Rci9dvBRWRN1H07lOlHCXk1ER7XSDkkh+jqGU1uh1xllVTa3/sMpK6MO010WtQq7Adp8qVtpNmyPWr0EF+XR/dJORMagkfEnKJ0Kll9Zz2DxliiNMDOQNi45c6DBVycRAIOaqQS0isySs3nHvnT1x2yOmGOFjTInaQTqFaKXW5OKFMY1m1ukTIRRRyDlXI1cE5cGKljtmmDbOfhFzFquBvH/5bONzBRx78CH7uvJ/ryeuoQPeRsuW2+Z717DGAeV/vm9sX/H7nVAnfPTAPJhFyNHvIK3UgA96YiXYwqaUKORJmTQfgcedQXMuqnN9GB5FVuwrHdSIrytTK+qOjdXz1/iwgzUdniDoOaGbIkfyxk2vUskq2wXbw0MmHlO/hWOWYoPqxEU42RnNjALz3bvAxWMyzMR6vHMf2ke0kQy6cuE/kJctq3gBfCN//2zL/jJ/nB/GLZg4nmkO5maUj2Go7yLkuapqGk4aB/aRGceJlNwMXvB5o2lXyX3oCqHnWzpobbu8poriiqj1BIcfCwftCfQ7lWjiZaJch51ky1Qq5x+YeC74+f+L8ln8HEBdiGnUFIVebFYhdHz+166cwV53DqrWKN5z7BgBRNcjsWAJCjig+Gq54/6FEo+t4BI2hsWDiKlpWbfK1A9B206aZRMsdB5g3iZ4tzWI0O4r9JwmJ2yV1HJBMISefo9waU+5rFSIKubjcMGWGXF6tkNPVhBwlWcRJfO8+51tlyNmuHSgnDWbA0AzBGtgulzXJdv/OZX+IN3zyb2BXzkGjpL53u04fLKsRhVz4PtIq5GTLKudyI3uKDLk4y2pChVxQ7OAShZwb3scymno/0tdtFZliaEyZmZkEKnJqolPLagqFnGxZVSrkyEIYtXIbegKFnCmqrhq2uEB9uiEo5KhltYUql+6PfuXHAVGrrKEZXSUE466vIeLRESHHOd/f7Q0ZYohugHMuqogitr7TsVXrg63Y6CEh50GwrJKJWdIsONXKv+W4fZONpwV9X7rGIg1mnUJuWPXgYpWs7MfZDVWr+apzNimSWlb98ORjS1XMjuVhOa7QuBZsS48IubnqXBA6v1BfgMtdaKw/q7TL1fB4qfZ1MKnQQ3JqoR5mWe2a8iYLtBGuYEYJOTmzUAV/EMq0ULFG7VpJFHIBkSq1rNIBft3y9m/eyAfquJpTQ5FMfGzXJqQfwy/+3YOAsRkliaeZyokWSMvhcIndkSrk5G3wW1tlyMUODqPW2VGEhNw4gIPC64Qtq1KeH+fA3H7gqW/h5w99Baf0x/HvTWPBpcajmG5UcVL3iA3GOWZsB7pZwC4tj0fh7dPDROEwcfYLAjIOAMqEOK074WtTYtvRwnsAVURltAK4a4BpNhpuDUdWwsy2RAo5pi512DcfEsdJCDkaVVCrFSFTUk5tVqna0piGX7noV4Sf0WM9PZKNTDBVyLQ4vwWFnONNdkZyRjDRjrWsNhxwMlHjTUJOz4X72G+063ahgw9VdqIMUzMxlh0L3mcahVw5U8ZodjQgLeMUcuWM2rLq28fl8Z6PuJZVUWV+ejLkGq7YsAqgvWW1hcpGhc3FaVgLz4n9e4BaIdfTllWXC8RzWoVc1tCRN3VULQeOy7Fat7FQoaUOnWXI0f0jNJS3VMhFLascRFkWo5BrdZy7ZadWKeuUpQ6N7hJyMsGqyp+j44GqFY4ZTJYyQ05qWU2TddcrCBlyZNusFqpcah09nQq5sexYxwSwCoauQWPewr7LEbRiDxGP4d4Z4scKQrsg8yTnG73UQUV4DAk5D3EKuaSEkCpUd5Bz5BxpwNYtslmlkPMscSHiLIKq1S+Xd67YaxXSLGTINRVZzy56/9OJKUWvrnlKcAHxJQM9ee3aAvJnfRSFHR/GinMq8nt/UkGPISUHQstqO0Ku/QTB/znNkBMsq0ZrhZzr8iCUXlDI6TnkFA2nQtOqtM+pBYY7GXDOwK3xIEzfx0RWJuRcIX8stmXVdoT8uK3FrcHXQpsm53AJITdVDMkGzQ2VQL7N1R+8C5bV73wI+J8XAh+6EvjS7+CCua9jmiiwljUNJww9aECdNoow33w78M5ncM7ul0EFmSij+VwWVxNyFg/JMpoZZmqasM9omUYiQi4mQ04g5CbbE3JZQw8UQ5bKslrbmnhSSY/19gR2VQDIkDByi7cn5KjtLtayajngJEOu5q4hY2jQcqFd1ScraaFDdxVyol08DhdPXQwA4E4ObmNzKoubn90HpFXIhZZVccxHCbnw8XEKuU4tgUlA74FyhpxsV/X+T25ZTXI+Z9tYYAHAsaOajO4r5MTvaXlJWoUcEFXJ0UU4OeKiFajKkO5bej614idUllWKnK7ej62OS7cKR1TEx3hAyIX7PFHLag8VchW781KHhuMOnmU1TiHXQt1KCcp+KuTkDLlu2lV9iMUpG2/u3W+c/jN4iCG6CNXqp96lnK3TBWWpgxM/QD6TQAk5QSHXYYYckLwQ4nRAzr+hK1rrKRdYrSkGZrpIdqQpdQA6V6YJCjkS0uxyV7Ks+go5739KItCV8p4RcjWRkFM16fUKq4UvwSgegF54Bs/ad0Z+H5YEhIRcxaoECp6dgUJOsqyasmW1vSojKHXQ1KUO7RRyi1UrOFdMI77Uwbf30EErJYEAkaDzCY0bzt2E67eJOWRlY7PwvUfIjYBz73qar83Dcr3zUFQKWXjk1CPB9z+58yeDrykhV7WrgbWQuwbGcuHkTHOk8oi1eTzPugfvNT6GghGSq5OP3gYsh4ooQCxh+AKuwC+7bwm+n5nYA2y/CtBNnDMeTRXJaJnIZHssFxJqjSYhRwlSAFhzQtslJdoMnQm5ez5KZkkZZk4hW1Z9lZTlWHhi4Yng5+dNnNfy7/jw7/2qUoc4hZwKW0bDbTp3Rl0QISPDwomNJVlWF2sKQi5HCTmSPyW0rDoAyZCrWlUUM3qMQq77DatAMssqAPzxdX+Mbfy1WHvmzYCbTaWoobbVWIWcotQBTj4gMIUCGzIhFjPkkjUedhPUshpRyDmdKOTSkQ/t/h4A2E6UTOo2KcAYE0i5yjpKHQCRdFuqWliqNsjvOix1IOeEuw6FHEU+hpDzlUOA9zlNyZokC2BJkJGea+oM5eb2JrKspiR/feQz4mNpjpgPOh5YbRDLatpSB8tJVDrVT1D1Gx3/Oi1UuaeLkJNbVlULH+tFknvQECGGhNwQP1ZQ2RdkyfxGgyq0fXhz8wgoOnAoErVB8pbV1taQQUPEjt0lyyodmPmDZCYTcjGlDnH7utNrbTnGsioTgj4B9GyzafXkakgIzZCJda8sqzIhp2rS6wVOVE6Bl+4Lvq/zhchj/FuGrHL0t3n7RMG7P6ZRyMVMTnwrK0uQIacidSn5kzHJyryeVZICiRVybhZXnT2Oj/3yVbhm9grhcUVtWvjem+zq4M2cLw6Oueqctx10G7RnAlJ4a3ErLp26NPjdsbWQkKPZYdwpCJMg0w4naice+Szw57vwfuvP8TrzdtSa+9LgHGX/IGZGgL0vw76L/wC3NV4UPPdetgsHSSMibY3dOx42y/qYyE9ELCnjhJBz4J0rS4QgHckZWGyoyxoMTYMrKQ/lx8ShEqOQ27+0PyBC/Yy0JAju/dxAOROu9Lt2EdweTTypvGbnBN76gt14xSVb8PYX7Wn/BAAZMrGxXUtYGFEr5MJzgZK9EcsqUcit2WsoZBm07NHgZz5ZeZIog6e7aVk12ltWAWBzcTMm7Rvh1jxyLY2i5uotVwdfXzp9qfIxWT0r3JcAgDvFQLXnkM9rLc6ySj4C+tWySsnvilURzgt6H/Tvj+1LHdKRD2JWWZRQ5ZzDthWW1S63rALisaCqrE4KSKjCdGlNVMilaVmNU6oJGdQtM+Sa2+HGEHIt7Idxx7pbdmr5ueOFTGiTN/WAEKxZbttInDRW0JyRTiFHF9SSKeTEBTpBNToACjldUMjFWVYlhRz5/OirZVVSyKnKc9YLgUAdikjaotNShyGGGEioJpBal0iL0wUVQTQk5MSBjKExYZCjIjFl2I6rJGsGOYBULizpVoMwJeT2zozgoUOLEUKu5tTAOY9M6ulx0DUWbIflusgj/YA7LkNOyI8DghIBlUJuy2gO+456k3wnwbnQCehkG+ifQu5jj3wcTAuPVwNRIjCwQGoiIbdYX8Tm4maYuoazJgp4ptGi1MGqSpPA1go5JFDIqUhdetwMw4a/RbJCLiDkWijkqAUGbhZ7N4+AMRaZ7GfZVPA15zx4n9wuA03b6PG145gpzgiTgIZ+MKhIuGzTZZgphtbXo5WQLFlqhIQcnDw2LT6A39RvxfX6I+Arz+C/lLzXP26tAs1w93myfydggL3oPcDOG4CtlwO6gbknTuH4k/8T/jCa6VVoZngOUkJOpZBTEWWjRH3EtSosx8VJQpBOj2QxX1UTcqbOlIq0doUOtuOiYbvQnfA4+oQcLR5Jkh/ng+YiTWSnsdxoZprVZgGwxDZKTWN4543JVHk+sroB3tDAmAsOFza3gzwksWXVIzpEhVzUsmo7LhqOi4wrthHm8nNYbl73E9lNgX14rh+W1RYKOaBzi9ur97waOT2HolnEVTNXxT5uJDMiRCZ4pQ5Ny2oChRz9bGyV59RNGJoR5F263EXVrgb3LqqQUxJybRRySciHjKQAc10eiW/hfciQA/xj4e13qsbvhJAbkyyri2tUIdeFDDmy61u2rPr3nBiFXMGILlb4yBp6cP7WLRc+jyioINdhp5ZJH7/QAfAUi8WMEeQGVxoORvPi+SS0rKZQMcrHU3WeCqUOVCGnpVTISYTcICjk6DlF5xZ0PiKTlJQIi1MJ9wKmZkJjWtBoHhcZsK7XGFpWU+H0U8pDDNFFqCaQ9D69nibK0wW1ZXVwSaN+QQ6eNVLe/OP24SDvW9m20q2WVTpIPm+zR6bIhBwghlGrtqlABm9Ohx/Ai8SCkkwhFyXkZkbD1b++KeQavVfIrTZW8fknPyf8zOZRItANMuTEY0hz73ZOFYVSh6JZRE7PgTUpp4bbQMMJydG4Zr8wQy6cMAsZcm0UcjIh5yNv5KX8Njf4uQ8511BWyO2e9rbjwskLoTXLELhrQHfCga9wr7CjTatUpedkDgRfX7bpMmwuhtbXwLLqupg/cn/w8yv5Iez50s/gv5q34BrtMcwSUvK4oQNMww+wB/+tGKr4Nk9fADz/95oWVG+iMpo3BVUZ06pghJCj5OBUfipCwKmIMkqcMq2KquUIx2O6lMV8jRBypFVT19SW1bb5cT6540RLHWh+XFK7KgAUCCFXNsPXd2pexl8v1ROmrgkqGUo6y0pJgLQzQiTkfMtqoJSjCjlrDRqxq55VCgnXUz2yrGYTWlaBzifwhmbgpt034UVnvahloLislPQy5KKlDnqcZZWQdqJltbeT+LhiB0rImZp3X0rXstr+fNY01jKXruG44DyqKOuFbY4el9V69zLkFquWkDmbipAz6HiRtKzypAo5/zrWwRRT6UILpWGsQo6Wdq3j3JTPj0lJOSsUOyhsqx0r5BJkyNHxAF1QS2RZNSkhJ1pWByFDTlDIxdjkZeLwhu034NLpSzFbmsXrz3197zeyCcaYcCyGltXTj9N/Bg8xRBfh/BhaVi2FwkdltTzTIBNyaWu24z4gNpJCTpx0dP53aa7L3pl4Qk5WqdmOG7yuxsQBk+q8TQKqkKOr4TL54pNJx3zLKlXIlUNCrl+lDoI6q0e45fFbULFFAs5GlJDzT+E4yyrQLHYghFzeyIMxJrRvrZEWtDg1iR8+HZchR1fEVYQuPW5U+edZVhUKuRSWVb+8ImfkcFnhl+DWp1A/8XJUG+pAb+aEii+/cCGja81wbw4t/3Tw+8umL8NEbgIZzXt/K40VVD73y8Bf7MHKV94RPG7SFSc8m4h145SRgfO7T+LnzV/D/eMng5+/cvcrIaOcN4TctVYKOQA4Z0xUydGG1OBvkgZLptdQkwi5iVI4aTI0Q7AiG7oGtwNCzleCCRlyTUXhY/OPBT9LUujgg8YVbCuEdl2n4tlO6X2p2zB0Bs7VCwdKyypRyImWVe88CfcPCT+3KnDMQ8H3W3KhnfYUsazKE+/1QCx1aH0vp23laRRySSHbqTyFnLef3BhCLm7cJ1pWezsFisuRU1lWTV2czNP35bhc+JxtlW1G0coG27Bd4RwDvM+AXjSFU2KroojHSANKui2uWVgUxgtpMuTalzq0zpALt4MpzGatlIbZGKJCUFKtq9RBVsiJRH27YodutayqVGtCyyoZ0yWxrAo2bGvwSh2Ea5haVlsc14JZwCdf/kl85TVfwbkT5/Z+IwmobbX3pQ6DO68aFAwtq0P8WEHVoBW3UrpRoFTIDQk5MXhW12IbjuIQR2pa9uCeI3TVzdS7lyG3QhRyU6UMpkpZLCsIOVnhJK9QCqG2HSrklmJaVmUy0M8sO7FSh+W4wsR082lQyFH7RS/QcBr4xKOfiPzcVlhWVaUOgLjNO6dKYqlDUxnh26wAiZBL0bJKiRs66IvYjoHAOgMAHOH2eJbV8BpNUuog5PgRhRwAXDZ2I+663yMyVmrhOUY/MzQ3qpDzSEodNT4HzfSUXAWjgHPmD0H7/iex2bZwqHnaH3v8i9ht2VgaCV931HVhl7fjlvnduNu9EAdHrsR49n9hob4ABy6OulXwTf8MxrztuHz6CuVKeUQhp1cBFu6vCCE3fg7uPXZv8L2KKJMVcrWGKxByI8UasBQ+n6qYjA4Vcv4EkEsKOcd1BELOLy1IAqr4uHbqtdgxWcLn71vFE2tNQi5F82damLom2Nao+ilNy6qvkFsLCDlRSVLXQkJuOrMr+LpnpQ40r6mNQq7TEPikoMQxdzIADNRt18tBi8n8im9ZpU2WvVXICYRcozUhx5gXu+GP7RqOi5ymN7e5MxIxY2hA86XqlguQ2KiG4wKSZbUXdlVAtH6urjNDjo4JDi+sBZ91xYye6tyTLb1+HAedI7S0rBKlK+OmcC8GgJIZvTcGr22ISi8fYobceiyr4n6YkJSDNNN0RUHICYrXVKUO6TLkKJIp5Eh8hOP2Ve2aBJ0o5Hy0Ugj3ClkjG9wfekHItVP9DiHi9FPKQwzRRajabMSV0r5v0rqhWlkY3txaW1aT2E7rMUHVg2xZpatuEYXcOognukpayhqYHc9HyBwgmgEWIUV1tSohKRyXCwNEOnmlDauAr5Dj4Nwj5eQMufVsRxJEFHI9LnX40oEv4UT1BAAvrN6Hy6LEaUDIKTLkfHiW1XAS4U/GBEuoQ1aw4xRyipZValkVFHKKllU6IXEJIZfXJctqkxSg2ycr5ObWVoKvNZ7D1rHwsXQCRc8xIeaAh4PSE2sngq+zBsO2QmhDvXh1EcanXgfc8yHM1MN9dMzwXmMhG77/O/BinHrz9/FO+1dxm3s9TvBRbCpsCn7/vu++DyzjqfG4a+KPrnuvUqUykjMBJzyvmVaFZobnoEzIycUO7Qg56DXUbEewQOZy4XuTFXaGpnVmWfUJJ6nU4emVpwMieCo/han8lPL5KlCFHHeyeNtlb0O+/pzgZ70giXyYOhOIDXqfUivk1JbVasSyKmYtVRCqM8f0ncHXp4RSh95YVtsp8ilh1xOFHLFT+fuxZjmCKpwxkUDRhAby8HHdarJMgmJGbVn1i0sA8f6Y1dX7vNPw+paWVdsVlJ1A7wi52FKHdVpWn5kP7/9pGlYBjwARVXLeSZK81IEQSIocOfoZKCMTc5ztLuUbyuRUVCHX2rLaKSGXk5TIqnOVKuQokpU6DLpCTp0hN2htsD7oYmmvLatDV1d7nP4zeIghughVg9bGL3VQEHIDTBr1C1HLalqFnHrVf5Cl1WLLqga6gNutltVi1sC28bzasiqRYsKAyBBVip3sx5WaFUyeRrKGMJCXLatgLsC8ic3RxaoQRj8jWVZ5D677iEKuOeFq2C4ePrwYe351Ape7+NgjHwu+t+aeD+56kxnOrIjyzDsXXKCFQm63bFnVPYKEEl41O4FCzj9GSUodFBlylBi3OVGOGLJltb1C7vBi+P5GcyXh/KFWQaoIFdQnlJBbPQL84BbgC2/D1/lbcWPp1uB3l1fD/bLFDv/WsYtfDbzl2zh0yS+F749tltrXuJA9d/ezd4ePPXEjdpS3QwVdYygaRNFmLoLpHhmTN/KRAbVc7JBIISdZVjPZcP/S/DjAV8ilL3XwCTlwE6w5kbVcCw+eeDB4TJpCB0DMkKs0/36vbZQ+DF0DVyjkOOdtFXL5TPg8n4hTKeQem38MTlPK4NoloGmtrllOcO82dSY0uK4XYstqigy5HqgRBULODQk5VUSJj9hShy41WSYBVcjRBRt6H6SEnBljZbQ6VAJRUlVexLUcLpC+QO9aHumxWLdCjhBvT82F+zRNfpwPU0FY0nOlFSdGLatcQciNtCDkBKVXDPG6HsWXfF5PpM6QC6/39VhWVe/Bj3iQkb7UwRHmQQOnkBPac6lza3BoFxpP0muF3CDPqwYFXT8zmIdXMsb+ijH2BcbY16TfFxljz2GMXdft1x5iCFsxQNO7pCI6XRhaVtWIqrPStazGWlYH+IPDkSwNdADAOTomnmjQcskn5LQkllVxQESPQSfKNKFhVRpgq+yOvk3y6FJNIBI2jeR6nh25WBNbVitWBZxzvPWT9+OVH/p3/OLff69rr/XNZ76Jp5afAgBkWAGNxWuCiSkQbXx1XA5odTAmvm+q6pseycI0QnWN25yc0UmZmPHSumWVKiqpZTVrtCbk6GTf5sSyqostq/UEGXJHV8IQ/cmCqN4qUYWcwrJqwsZFTlhgcOLwvcC/vBl48FPYhHk8mAvfx2X1OjC1F7jmrdh83qvC199yEbDlEiySMP8MK4l2epdjcyEk5IKfV3bBWri2tSqDZr6RvL2Z4kzE8rJ7bHdQ0AGoiTI5Q67acARiWzPCCW9EIaczcKcIziU1RluFnL/dDDpCRc53j343+DpNoQMgWrD8CWavbZQ+TI0JChn/HF+z12A38wM1ngGaAfp0Ik8nsEGpQyNa6vBs5dnga7c2i7WG996omnGymO2q7SnXcalDry2rPiHniuSJ9N5jM+S61GSZBEKpQ4xllRJycRPYTpVArSxjXoZcvxRy4deUACqs07J6ZCH8zBlPqZADpIyr5v5JWupAla7cjZJJ5Uy8ZTVOCWl3iSyWz+vJorhv6P2Sjv1U25SGYI+UOijuu+uyrFIbvS1bVk8/0UU/510ezjeTWFZPB66ZuQaA1+6adhEsCYalDunQ1Qw5xthuAP8C4GIgGAnKM6E6gI8D2MkYex7n/G4MMUSXoJJ8b/RSBxW5NLy5ifsga4h2yUSW1Q1GyKnychhjYCy05LhcbBVOitV6SFCUsga2jeXBjPalDjRvz9Rlhdw6Cbl8AkJOq4GjjAMnK8HKe0bXUM576jr/erddjm4KNyzXwoq1IvysYlVw+74TuOMxz+r4vYPzWKpakfeRFpxz/P0jfx98v7fwE7jHzXmWP8PbhqX6ktCy6XKutBxTEpExhkLOgf+olTVvBwkKOacKP3hIVqD48Fem40od2irkguuQCwq5nJGTWlajllVZNXmqEhJhm4uiYkywrPoKufmDKDz4Jfyt+Xlcp/0QVrWB52MbAOAkOWHWGMOPMt5xZAAuedOdwLQ3gJ350WeBI7cDCJtWlwkhl9VGIlZualkFvAD/2tHXQmNay9yi0XwBy64hkHFA1K4KePvprPJZeHrZszpO56eVjwHXAOaCaRZWGiKxzbXwHJeJNm8CpIHbJTAz/nEyAoUcAJMVYTdD6r53NCSw0+THAVIWW/M+0GsbpQ9T1wRiwz/HKUmu8ZDoaGdZ9QlLOXDfh1PbGjymV4UOAJTq1DhQNXA/LatUFR5RyAmT47g8p95O4ul9kCrkqHWf3h/jJrD061SEXExWGdAc50gtqz0j5GipA7n+ZQInCWjREx3Oywt4SaAKne+k1MFxdGjSy5ezLSyrsUrIaAZ2J4iWOsgKufhSB855x4sZUYVcdy2r8vk8aJZVxhgMjQXjdNvlyGhM3M4BUsj99pW/jedsfQ7OGT+nJ+3KVLU4nLO2R9fODMbYGIDbAVwC4FEA7wWwLD+Oc24D+Ai8ce3PdOv1hxgCUIeiCtaFDWlZVSjkBpQ06idatayux7LaGNBSB1ENEE449BhrThpUqEIuZ2DbeEFpWZUJlW5nyC3GFDoAUbssgMAm+YMj4eR3eiQbDIzWsy0tt1NSxwHASmMVN3/tMeFnKjtIWnz/+Pfxg1M/AODZPWa1nwAQTkwBL4OLwnHVhJyce5fLhNu3sOrtL9GyGu7zuMmrt/DhBPZJBiYM7uiKeEuFHAu3JaNloDEttWV1vhqqULaWRQvGSM5EBhauTSDgZQAAIABJREFU136A1899BPir/wv44GWY/vZ78FL9fpRYDWOuC7P5GbGqaXhg++U4+tzfwtsmfx1u8zrbVtyFkelwNZmSYcfWPEJuxQqPR14vCTYVy3EjCrn6iRvBrcm2dpZyTmxaVW0DxZsuehNMzcS1W65Vqs4YYzAYUVpWlwXVlcXjiTZ/ssodUQmShpDLsJAAmKvNBV+naVgFgGImalntl0LOkEodAkKO3iOImlW0rNKWVTFDjlpWKdzabLD40KtCByC5Qs51uTBGSZNxlhRXzVwVqD395tya7cIhrysT2XGfi5Yi1qRXoATXSiO8lhIRco5aOZXGmtdKoVK3XYDr4Dx8TC8m5UB8OcJ6M+Qoxjsg5LKK/U2Hjq0WR3KmFt4DFQq5sVzUzq963bgMufUQTPJntayQKwoKOXGMYrs8WODVNZa40ReIEqyqe8H6FHLi+TyI2WwqAUi3yjq6DUMz8JzZ52C6EF2s6wbo/WdQhQ6DhG4q5N4BYAeAbwC4iXPeYIy9BYBKt3sbgJsBXN/F1x9iCDGw1ycsyGfCxrSsDhVyKsgTLrFl9cdPIRdnZ/AIZ+93nbYIr9bEUoc4y2rLDDldg06JhwS2YRmtFHKRDDmEltWHD4eKpKkRb4IT13jVDcjEFgDsn5vD48fFptVuEHJ3Hroz+PqVe16JU0+VACwLhNwSUWQBTUJOcfzk3DvdsIDmLj/ZnC9GFXIesRU3ETQ0JmTRlTIloZSABge3JOQ0sWEVEEkBn0AXLLWWeE4sN1aA5nj/rPGmRXPpMPDEN3DOD7+CB7J3osjqQA3ePwkntM3YpOdxxPV++UZjDjjyeWEUs3PkQuE5VJnoK+RWGuHxKBjlyCCdEk5XbroKd+7zrCPtJj+jeRO8FiojVdtA8ZpzXoOX73w5snq8ndFAATa883auuoS5SpNYZcCaE5JKMtHmTyzkYoexXOssmtCyCuT0EpakS6ScKWNrcWvLvyFDlYkkZsj1smWVCRlSKoWca8cQcoJl1dvu0LIao5Crzgbv8VRPCTlChrcYb8ifw71oC9w5uhO3vvJWfGXf4/gf+7z9014hF34dp5DrtVplUz5Uwh5ePRx8Te+DJpFWxVlMO1UCyQQGhfc3m4Ukurc9/Sh1oChk0k9B4wi5sXwnltWogseVXAhxYIxhJGdgcc1SZsiNZpO1rDYEy2p3yGLZsjouW1Yz8RlynRY6AMlaVtdV6iAVzXRadtJLmLoW7EPLdZHjWsusyx9nZMjn7lBE0h7dJOR+Gt6s8B2c82iVGgHn/AnGWAPAni6+/hBDRELvAWnFYgMq5FRkwpCQkxRyuiYGIidQyMXtww1ByNE2OU+gBKAzQo6G4xoaQ9bQsKlsBIon8bGSQk5SKZrrVKVRQk4OaY6zrAJey6oPv2mwlwo5mdgCgAOn5iI/k1efOwGd2F8ydQm+sK+5j0hLpSpDLk4hxzkPJs0aIdKOL3jHMm+Gf7dBCNg49Zahs9j8OKC9Qs4fvDKikAsJOYVCzlAr5ByXo2qtwc+M3nv4q8C9NwMnfggAXlqZPBY28licuQ4fOLgd33IvwcS283D+tn/BkWduV75XADirKNopZUKOc46KHRJmBb0cyZA7d/xcvP+578dTy0/hNbt/AdffdQ+A9oRcOW8ClVzk53EKOUAMblYhoxVRa17Gh5fmA3XERCGDhXqYqRch5Jrb6hJCrpwpCwSDClSNm9dHAOkSOX/i/NSkjmDB8ksdnM4nlmlg6prQsurfI+k16RBCbiTGsroWWFbjFXLcLoDbY0HuE7WsTnXZsppNWOrQr/KMPeN7cNFkGcB9wTZRAkO+doTs4NPUsrprbFfw9YHFA8HXSRRy9a5YVknmlqJlFQA4N8GahSFFo/eWVYpOzpdyHCG3bsuqomW1zf3YJ+TkllXOGUZzLUodhLFqeG1ZXWpZpQu2jEXz9VqVOshjujRIUuoQR8glK3WgmbKD17IKSPNNh0vHlPVkwWJQMbSspkM3CbldAGqc80cSPn4FQPd7doc4o6EKz9Q2eKmDamVhWCGtIoOoZfXHTyFH7Tn0Q3+9llU6QS5mDTDG0OCryse2K3Wg29XJfqSEnDzwVllWfYUcxfSIN/jsZXakSiHX4FFFWkURmJwWy/XQ/ljOloN91FIhF5MhZ7s2KlYFpWYDnMvC4/nsgrePqEKu7pJSByNGYaVrQn5cSWqXExRydkKFXPM5qqZHpWV15TgWHvpXnKU/DV+HMvrorUA9Sio/5W7GXbgcv/TG/wLsuB6PPlPBx5+8FwCw2dDxB1f/ATYVNuHp5acxV5vDfHUeJ6tz4HDgVGexqyD2UY1kRlAyS1i1VlF36lioL2DNDo9ZyRyBpjFoLCQGHJfjpt03AQAWKuE2JlLIKSyrW0vpFGUUWWIbPbw0D8BTFk6PZDFfI4RcpGXVu9/SptV2dlUAqBKFXMEoAdIpkbbQARAtq2t128tBkhZsegVDV5c6UELOsrxjxpioTqGKkppkWQU3wcDASQyzU9sKgPVJIZfMsirmx/VOiQiI94O65YKKsOVrJ65ltdHHIPidozuDrw8uHwwWQ+jnaJIMuU7bN+l5X7fUhBxVYvZKIaeyfubM1nmZcdA1T5lGm7KBLpQ6ONFSB7koRMZI1gRQjSrk3AxyZvz0WiBeacu40x0lFSWax/Jm5NpoVeqwnvvmekod2i3kANGW1UEsS6DXp+W6XVM9bkSoLOFDxKObhBxHwkw6xpgBoAyPlBtiiK5BtWK64UsdhhlySohWGV34sEtCBtVjJhlJ1HWnA9QCSgeSQnh1B6cFXSH1B2oyweNDJsUa0gqlmTLHT0baUgdoCkKuFLWsprnuOef4/VsexgOHFvF/v+oiXLtrMvIYlUKONdVmps6CVdFuKORoPlw5U8ZyzQsHFwi5hni83BiFHOCRiT5pRgm3w3MOHJcLhBxVyMXZu0yNCcToiNQu167UwVe+MRYee1/VlVWQAoJCbn4/8NEbgGcfwBSAzOwWAN55U/SPuZ4Bzn4u+J6X4iVfzGK/6ynafvbsFyJr6LCdkHzOGBpmijN41zXvErbxv9/2CP7xu48DbhbzinizmeIMnlx8EgCwf3E/XPgqJxPFjLc/DU0LrhdaMtLKdiejnDO9Mg/F63eKnF4MFLbHVhfgE3JTpSyOEEJOblk1FZbVJIQcDXUvmVFbV9r8OAAoEIXcat0WFlsyemcT/6QwdQ2ch/cqX/1E76H+tTqSNYRtyehaUD5jORyW44aWVTAYLA+LEP1ObRYAUOlzqYNM5lCsx+KWfpvEkhdhvCeRJ+LnYkzLao8nyJO5Se+e3VhGxarg+NpxzBRn0HDD40YJirgJrJgh16FlVRoT+WMkWkjSqww5lUKuE7uqj7GCGSHkOlLIqTLkUirkAEQVcm6uJZkVnxXYLctq+PflQgcguUIubWOybFlNlSGXQCEnK0iF8eeAlCXI406rjxb5QYMZY8EfQo1unh1PAcgyxna2eyCAF8EbNe/r4usPMYQyFFUsdej7Jq0bwww5NeRJl7DamYCAiVXIDei+jRsoCgrQDiyrK7UoISdbIH3ICichzNtYf6nDUqtSBwUhV8xFCa/pEd+ySsjBFEzl9w7O43P3H8aTJ1bx4Tv3Kx+jKnWAVsfsWB4vvSAM7Kd5WTIatouv/fAYnjpViX0MAKHNdSQzgmWFQo6q6IDmBE6LIeQImVgllk/LNnF4YU0k5Fxa6pBMIdfKskqtWj4ChY0WJeSoIsZt1IDHv47Cv38w+Nna6nHg2QeC7yvkuihe9DPAz/0z8PsHgV/8PNh1b8N8bnvwe/+8T2J72TqWB5oWwqNL0fNwczE85o8vPB58zZ1CQGzEEcRpJoCjeUNByLFISUQa5PRQ0UhbaqdKpnCuREsdvH3lNkKibrY02/b1aKnDSCZqkuiEkBMUcg2nb4UOQJPUIaHu/qIFvYfypmVVVv0yxqQcOUcsvdBEu7Fb85SQ/Sl1oBlySRVyvd3XsqJQuHak+1NcVIkq1qRXYIwJKrkDS55tld4H6f0xbgJrdWhZbZUhF1wj/PQo5DopdPChypEb60Ahl6VjRtsvdaD349bPD5pWpVIH7rQm5KiSVMiQ65J6kz53shi9L4gWf3GMMlcJ7ymeAjA5ctL1nyZDLq1llS7s6Brr6aJLGgjjTof31SI/aMgoFKhDxKObn0ZfhpfQ8tutHsQYKwD4c3iKutu6+PpDDCEMtlQKuY1oWVWRS0NCTmFZJR92SSyrGzlDjma1rTcjkQ7ISs0VX1lx5SNS6iBNFKjCZ72WVTmkWWVZLeatyM+mFaUOacjBfUdDcuu4gnwBIFj5fDCthv/60r3CxKBVqcMHvvE43vKJ+/GKD34bi2vxsau0na9klrBc9f4mb5Eh5ynkohZa+ljLsWC7zb/FvabIA6cqAiFn8XBwHtuyqjOghUKOWlZVx1CpkGs+J984hdfr38RHzf+BL9ffCHz6dcg/8i/B49b8Y6wZOFC8HPMsHOwXX/ZnwLk3AtmQcAomUIgj5NQD5s3l8D0cX46+B5rh9th82LTLnXygMjCE+1McQdA+Q062rBa0sVjVQRIUzHD/UDXmaMmGw71Jz4g5EnkNf185lT3YmX0hrth0Bd500Zvavh4lqUczYhNh3shjx8iO1O9BUHw07L7lmgHeBEylkKOkvU+el3PRCa7QtNpwULXC/ZPVRcVSoJAbKMtq/8hP0cLuSmHp4mvHtqz2USEHALtGwxy5g0sHASDeshozgRWUQCn2cauW1SBDjijkelfqEP1ZLqX6ikJNyHWikCMZV75lVWizb6dY9u47Ecsqz7UkiOKyAgWFXJcsq+PF6H4ptWhZPbQQLuRtG4+qsVu/rjgOV11f3bKs0rFVP67jpJCdOsI9akBy7vqFVvefIaLopmX1LwD8KoC3M8YWAPwl/WWTiPtJAO8DcAGAowA+3MXXH2II4QPNv0lv+FIHlUJuQEmjfqIhWWXogNxKIIXcaBly9DygaoD1ZiTKDatAvGW1XYYcQ2ckmI+0ltVsJp6Qk4P0k+KJE6GFcbGqJspUCkKmOXj5JdP40fGQQJPzWSjuOeCVQFQaDh4+vITn71VXz1P1W5aVgms/TYZcRssENilf9UQLEbwcIYaDJysobyKEnEstq+pBr6lpYESNJ2fIJVfIkQy5pSPA374YU0fux5+b4rErkGO5ZmSBn/kHYPeL8c6PP4qG9tbgDFTZr2igvn/eJ8mU2jIa7hOVQm6mEFpGfzT/o+Br7hQCEkE8H8PrhmZDtlvlH81HLaslQ33eJEXRCI8XJ8exkAu/lvPjAPq5quHy/K/iT268KNHrUQXYaFZUyJ07fi50Lb1yRlB81O0+K+Q0wbKmVMg5HtFBzz8ftNihajnEsgrk9PBYF40SVizvOPjZlHM9LHWQya84CJbVdaieEm2TGa+Qky8dys/RYZ+YO9X7CbKKkItTyMVnyBElepoMOSlziyL4HCEZkOtR2rZC1y2rikbVbmXIiQq51vu6FGNZZYqcTwoh2y/mOK/n3KQE1YRSIRdvWT28EI4Ltk+ktzDnTB2WYze3I0XLahJCjpC4dNw6KIUOQLRMzIpZSD8TICh+N6I9rc/o2lnMOT8Jr2l1FcAfAjgBYAoAGGMnACwBuAUeGbcA4FWcxySHDzFEhxDtPwrL6kZUyCluZI7LN+R76SbkSRddOU6UIRdjwxnUD444qw0di3RCOK8qMuSEySSnuT2tM+R0IdA2/bYspiTkTDNK8EyXPDVTpwq5Jwkht7BmgSv2qarUAQCqTkWwz7VSyK3Wwve6EKOQs107IM4YGCw7nHRQQm5RJuSkDLmzymeRx3rHtmKFVlm/0fHAqVVRISdYVlu1rHaeIedYDTxPexivN74R/Cx38nHgyPfBIO37yT0oXP2W4NuqxoCLXgvkx3BgbgGMeY/PaFnlAJ8SIivN/W8lCLGeIQq5YypCjmS4+VlygGxZDf82PR/TqCLKCkJu1FwnIUcsxvQ4ZrLh+aHKhjNSRgT4oAq58ZxIyHVS6ACIk/u1uiPkg/ZaIedZVhUKubpCIadQ9oiWVVsgLPNEIeftG++9VBo2bMfFPLlvqLKi1gOZ/IpDP9WIsmpPzF8UX1uMKiEKuS6pkJJCaFpVWFbTljqksdlmYiywQHjfa8zdgCnjAvzsuT+Lyzddnvhvp0G3Lauq60ilmmsH0SLsnSM03aJtqUMMIaehdbM1JZboOLRb+YZX7gjv19fviWbglgRCTry2D813rpADxOOqWgzJaJ1nyAmW1fqgEnKiMEC0rA7OdvYDQ4VcOnRTIQfO+bcYY5cB+FMArwHgXz1Tzf8dAF8A8Puc84PdfO0hhgAgBWj6Crnw9xvSshpDLjVsNxKieiZBboOiq09JCgXoymQxoweZFIOqkHNirG1ihlz6v6si5IRAcnsEzPS+jyrkxAw5en05HTRMLLcg5KpONBNNU7SsTilaVtMUTFBCrmG7qFnR60xV6gAAlUZFUOu0KnWgA2GqDKRYbYTbUsqUBMWdqJCTLKtctKzuKO8IiCLfbrtmhb/nzaa9g6cqeBEh5GxiWY2bIJg6a5khlzUUhFx9BXjydmDfl/At/CtGMlV8wSjiS/AmD7nmBJozHfc6e3G7cwX+zb0Ct//6m2EywPjE12BzG7Zrw3Is1CxgrrICX+tVjAknp5bVZYVlNS7jZVM5fA8nV+uwHVcYXFPLquWGx9Ij5LRgP/mg5LqbQpExmjcBV5zsjWbWR8iVCYFKlY6a0YaQo4R3iuuLEk5TBZGQu2DygsR/h6KYETORxKKB3n5GGromWNb8c1xV6pDMskoIOTO8Fi+YOh/3GBrqtgvOgWcXa4Hya7xgdn2y11nLaq8JOZpr5wr3dZn0ibOsdiunKymEDLlFj5Cjn6OUoBAUbY6akEtjWaXET1yGnFufwYtH/xjvvrazay8JVAq53DrGrrI9tZwz2t47VcgoFnHTlTp42yFbVjWeXCEnZMh1Kd/wirPGcOvbnoO1uoPn7I4ScsUWllVBITeeXiFXyhk4seKd3yrSlTEmKPZ9JFHI0eO12hh8y6pc6nAmZ8gNCbn26CohBwBNou1nGWNFAFcB2AJvWe84gPs452ov1BBDdAGOqmU1ZqV0oyCOTBgScqJCzojJX0ny/GLWCAm5Af3goO8pttShA0aOrjQWVYScNQrEEHIyKWoRRVMS27AMwbJaaK+Q40wk6UpZI1DLdFIwMbdax1xFHCguVhvIZ8QBtpgPlQ/UaBW7gmKWtIC2KHWgA+HFNTUhR/PjypmysH+oUmqpIZU6OFwgV1QKOdGy2lTInawgT1pMKSEXN0EwNK1lhhydcNadOvgnXwd28C6geS6NNA9TjZzHufGdwPP/HGzPi/HLf3ovas1zv2Y5KGYN5M18sG/W7DU8dZIDWritcVlII1mFQi5BqUPO1DFZzGCu0oDjcpxabWBmNCTG4lpO40od4jPkWk/CVC2r45n12czo8aIKOa6FZHA7Qs5KQb6vEVJ5Mj8u/K6TQgfAI8WyTbLK5eJ9pC+lDiRDzr9HqhVy6Syrk7lNwdeXTF2CUlZH3fbuT0/NhYRpt/PjAJHIrLX4TOxny6pcUtAqf1Em6FyXQ9NY15osk2JrcSuyehZ1p4652hyW6kvxltW4UgeBRExhWdVJeUCkZVVcTOslVEqzQhdLHcY7VIeq9ncay2qgkJNKHQy0JuTiyjY6Pc4yGGO44qzx2N/LFn/OOVjzGB2hGXIT6RVy/+maHfh/vrwPL9g7HWt5zeidEXJ0v9GpXK/LWdJA/lw0XZKpN0Db2Q+oCO8h4tF1Qs4H57wC4M5e/f0hhlBBlcFAB2YbUiEXM9mpOw68suIzExHLakq7JF3Zp6t6g/rBIYbDxpQ6dHB+Cy2ruahl1bVHA6mzTIpFyQySjZVyWyzHDUgqxkTyRPXaAGBDJOT8/DhAtAjGtaxW7Sp+447fwNHVo/iz5/8ZKstbIo9ZXLOEDDHOOebrYamD25iAnj8CwFO0FbNhJo9sBwme43KBkIuzrC5bIdFWzpQFBSF4BtzVwTQHDaeOml0L2kldKUPu7PLZ4Ws11X3UskobRDWMBT+2QUsd4lpWWXyG3PxB6I/9KwwO2M2nW09+HfL06ZA7je/gbHjrdkBu743AJa/zvjb1IMPKJ+QKRiEk5Kw1HDjlgOkJCDnBspo8Qw7wih18svbYck0g5GjLKgV38oHSKC5DrpXKR4YqQ24iuynm0ckwSs5XehwPV8MsPBUhR/dVmmt9jZQWbCqNB2RFTs9h9+juxH9HRjFrBGTVAiHVe29Z1cBdUSHXcBrEaq4FqkalQs4UG2KpgvCms9+AVfcIZkuzeMmOl+D92W8H5+DT8yGh3htCTiQNfEJLRqOPakTGWEC8AuKCh2rbdI0F56bDOTQwcYzYhwmyrunYUd4RtC8fXDoYW+oQT9S0t9WrEFceIH/fa6WgajevZzFZJuTGOrCrAiLp5e9jGlHR3rKqVsgZrB0hF753odShT42cWUOHqXvXgu1y1G0XOVOH63IcFkod0ivk3vTcnXjDVdsFFZ6MjJ4BpDXIZJZV9Xnaa0I5DQzpc9E+kxVyQ8tqKvSMkBtiiNMBla3vx1khdyYjYlklH4RJWlZp9g0lfzZahhydh3TUsipYVr2BomhZDSfsrUsdNMF+l+QYUFCyqZwzIxMsVUNnwxWbRGmwuRyuq8LdR+7GvUfvBQC85zvvwWs2fyDyGJksq1iVsJ3UNcGdkvC7UgLLakVSzi3FKORoocNIZkSytjJwtwCmecTUUn0pIOQshxJUDNtHtofvR1HqkDfy8L+bX6H3S1LqoBhMLjeW8dVDn4VeeCrczpWTwDffD+z7EnDihwCA3I5tWGXeOVvTGDIuBzZfjOrul+E135zAPn4WysZdAL4KQJykeuHy3vv2Jy+0sGHNXsP+EzZYEoWcomXVloj9OGwZzeHRZgPvsaUqsD0kLrN6FhO5iUj7rlDqEENgxVnRVciZGnSIk6TJdRJyY4SQ85WOZu4kvvHMV4IfX7f1usjzOrWEU4VcOZfDb17+m/jMY5/Br1z0KzD1zheYilkd802OeWGtfwo5Q2NChlTDaQgLGhk2AjSrRlSlDq0sq3vGd+EfX/aPwfd0kvv0qZBQn+xyoQPgEVwZQws+ZxuOi5yicKOfCjnAI+j916QLHqprR2cMTlO17bgcpt69nK402DW6SyDkqDooWalDZ0RNHMEnf9/ra0SlNMutQyEnE3BjHRQ6ADGlDqksq+oMuXYKudjj3CXLahIUs0agzK/UbeRMHSdX68FC93jBFLLm0v7tVlAVO6TNkKMYJMsqPWcsxxXuS/3IrBwkqK6vIeLRNUKOMbY15VPqAJY45/GeniGGSAnVwEUTJuZ936R1I07dc6YTcvWIZTW62pn0+SUyURrUDw5hpY1aVsnXqgKCdhAz5LyBrqCQs8KcJ5kUE0odDAaX0w/gdNvSqmEV8NRsMtbsivC9qJBr37I6V5sLvt6/tB93WF8BcLa4XRJZRvPjuFMEd0Kl1Kq1iqkEpQ4yUbcYkyFHLasjmRFRIYembdXwHrNYXwyUWhU7fJ6JAiZzYY5MYFklGXIjmRL8PXFiOTymLsKJI50gHFg8gE8/9mnctv82VO0q6Fh66//3W4Atvr8MOS8bL3w3cOHPABM7sbBYxb477gAA6LoF/1m0WEKVZUV/7ynkGoJlVdWwCkgtq3WVZTV+wEwVccqm1eJMhJADsawKVpYYy2q7CSBjDOVMGZQins6r7bJJMZEPr29fIVecuR0O9/bLdVuuw5Wbr4w8T8zES2FZJQqwYsbAGy98I9544RtTb7cMWqZCSfS+KOSIZbVm14T7p4GQHFaF0ReEUgdHUH3JGUyU7O+1Qg4AcoSQq1mOkkQRMuTMfhByGpaaHwX0/qrKKdM0eMnVCC1uwsJWn0LWadPqgaUDsZZVsZUw3K9Wh7l3rRRynaruOoHSstpFhdx4oVOFXLQFUiDk2ijkyjGEnKm1VpZlheOiLnXoNXlTzFBCzsFkCThE7imdqOOSQkXIJbGsmjoDY6JdFRgsy6opRaWcjvvNoCAuE3MINbqpkDvUwXM4Y+wAgK8D+BDn/LEubs8QZyCUCjlqWd2ACrk4YkPOBDnTIK/wigq59sdZyJDLDDYh9/DJh/GpJ/4FWm4r3No24ZwWw6vT/+1VIUPOGyRTZZagkLMlhZxNcmh0TQh3T2tZbUfIqSyrdaeOyZKOuVVvUDtNJqZJFHKCbRPAw5XPAux3hEwomSyjDavcLgYNpf7f21GKD0wOfl6TCLkYy6pMyC1VxefRYodlkiNXIVbXjFbEWC5Uc/nbTwm50VxIGhxbILZjhWX1n374T/iL7/9FZFvLjoNfXVzG2ZSM07PA7hcii8NAc5vql/88UJr1vibXoK7bASHnK/0Aue2xqZAzRIXcgZP1jhVySS2rQtPqsoKQK8zg0blHhZ9xJx+QFHHWcpcnV8gBQDlbwonGBLTMPNzGJEYzY22f0wrjeWJZ1WvQcofh5B8KfvZbV/yW8nmCJTwh+e66XCot6J7FkU7w54lltfcZcpqQIdVwGoLCWOOEkGtX6iBlyMmkBW2TfVrIkOu+Qg7wrj2//KRmqT9cxJbV3mfa0vsBvb+qyGxV06pAwPdJsbJzjBQ7LB2ItayKmUs0j3X9ltVBU8it59qX82U7VchlFRlXdKjQjufxFzC5lCGXYa3JrLjj4vSRvCkpih2oXXV7B/lxSUFJaB9JFHKMMWR0LUIupyk66TXkz0VL62+r8yBhWOqQDt0k5Do50xiAPQB2A3gzY+zXOOf/2MVtGuIMg2o1Iq5ta6Mgzvp3pt/gKCGZNTQhMDUJWSlnyPkYNELO5S7ecdc7cKxyDPmtU6gc+F2BOFhvhhyd1IzkDHDOxUDyFgqgVL0vAAAgAElEQVQ52bLa0Kn9It1+pISc3KIGiIScnzsFADPjwFwzfz6tQk4m5Cw2D3P8u7Dmnxf8TLasimHtRaH1smJVBLuGbE31sSIr5BKUOoxkRrC8ICvkaNNqSAKsOWEgv8lKKGfKYGDg4FiuL8N2bcGyOlkIg/2PzIfXBVXIZZgD7L8Dn/qPvxK2YU+jgV9YXsErVteQ5xzIjgJ7fxI47xXAnpcA2RKyn78pJOQIqUvbG3Ut3CdCrpLQrOg9nirgKo01HDxVAyuFf7dkkhw7gpIiQy5JqQMgKuSOKRRyW0rR/EHuFEOFHF0wiMmQS9IUOJrP4OnDvwCj/DDs5UuRWSepNZYtgXMNjLlgmoXspi8Hv3vpjpfiwqkLlc8zUmZ2AoiQce0y89KAXndihlxvSSJVqYOq0AFQlzqIllVb3EcZWSEXPv+ZfijkBDJcnYfZd8sqOZ5UIaeycqrGfrYiZ7jXEBRyiwfAQReyEpQ6dJj3FqfEApLf97oBldKsqxly3VDI+aUOXbCsZtoo5ARCTmjT7U6pQxIIxQ7NccqgK+QACBmSPjIDZFk1pXGn3udW50FCxkjnWjrT0c2zwwTwOgCLAB4B8GYAewGUmv/OAfAmAD8AsADgNQCmAdwI4A4AWQAfZYxd2sVtGuIMg8rWp62TsDjdiFXInemEHBlgZnQNJrn5J1Fs0A91IUPOHqxzZLG+iGOVYwAALXsKgCMMFBltWVUoQH80/yO86Wtvws333ay0tAotqxkDVbsaZNzozBQmlO1KHSgp6qzDsipbuxzXEXJ3qAVzshRuAyXkRIWc+lqRCTkAyE5+E9DC9ylbVqktkTtFcDecUK1aq8JAdy2m1CGikIuxrFLVm9yyCgCgTauEkKsSy2pWK8HQDJSbWWEcHEv1JYGQ21wKSden5sJt42jgBdoD+DPjo7js/1wN5xOvxglCyv7d0eO49cgx3LCs41brRfh17T3A7z0JvPZvgQtfBWQ9YowOwKk6hE7ymR6+N2pJpZNK//FUIXdkedG7llNaVpf9llUpizIOtNhDaVktRK2j3M2HGXIxBHGaCSDgXRtufRaNkzfCrW9d98StkDUE27VRPAAA0JiGt1/29tjnJbm+ZFCCej2WNRWoynm+75ZVsdSB2tpdmxByCoUctawu1+zg817XWOR8pPcWqljrHSFHrj1bfS/rp9JK3qYKUROqbJGqQi8hp6tPE/kd5R3QmhmaR1aPCAsttIU6zmK60TPkVMT7ehRysiKu81IHRYZcqlIH77qXSx2ybS2rpNSBXMdCA3CvLavtFHLjvVPIqci3JAo5QFygC587OEQXvT5tx+17q/MgQWh5PsPnq0nQzbP4KgCfAvDvAK7knH+Mc/4k53yt+W9/U/12JYC7AXwGwNmc869xzl8K4AvwFHtqf8QQQySA+IHmK+TC329My2qMQu4MX3GQB5RGglZNio2SITdflXKpNFsYrNE5m+r8/rsf/B3uO3YfPv7ox/HgyQcjv5dbVikJVDDKQj6SXOog5/gJgbZdtKzS183pOYxkQkXX887zSJ9S1sALzwsD7pOEzqsIOWasITN5V/C9rF5brBH1i10QyAyv1CGBZTWikGsoG6DpsYiWOojqG6rKqTqUkPMsc+PZceGx1LI6OxoSck8emUOGNd8D4/jf2b/AG4w7YdQXMafrcJqTlHGX45pLfgmrP/sFXF3/MN5tvxnfdi8BjKgdJWuoCTl6/mha+N6EUgcyAFeVOjyz4L3vJJbVchuFXKsB88xouE3HVZbVooKQcwoBgRB3PtJ7VTKFnHhtrJdwyhk64EYnXzftugm7xnYpnuGB3m+T5kUKdsxsdwk5+vcW+mhZNXQGuOI9kpLjttWakMvHWG3zpi4stgDxYem9KHUA1HZxGUKGXB8IOTohFxRyMS2rPnyiRSh16NNEPqtnMdu06XNw4b4eZ1mNC/vvNENOHi82EmZndgPdVsgVM7pwbMeLHZY6kEVcf3/Qz+F2hFwxY4AxRBRyeUP9+eMjiUKun5ZV/zo6tHD6FHKJCTnFPWaQLKviPIT3vdV5kKBqMR4iHt08O/4bPJXc2znn6uV+AM0Sh9+Ap4h7N/nVHzX/f0EXt2mIMwyWomJ6vZa+0w2qpqCqgjN9xYEOZLwMOTK4SrBvhAy57AATclJQPGMNgThoZ8k+vnY8+PrA4oHI78VSB0MgdkrGSGSySSFm2zAp0DalZXUtnpCjVtmckUMpE1oSLz0rh+/8wQtxz7tehE0jITmWJDty1QqtnXtL1wRfZya+A6Y3yxKqomV1vi4r5MLB5Wpj1bPiNV+6brtKy7lMyLkcWFXYW6mSopwpK0odiGW1QRRyxLKa1bx9NZ4LCbmF2oJARm7KF/DrM4/hf5kfwvfMt6BAbKVVcn6dKIek06bJ84CX3wyc/Vy4zaFE3P01iUIOLHxvQoYcVY0oFHJHl733LRByMRMiMUPOe73EGXKSQk5Wm8qEHHczADeCiacc9uwjdYac1NS5XhtMLqMJpDIA6DDwtsve1vJ5ce+nFWgrJlW0dQN0gikq5HptWY0q5Og9tNEI962qZZXmwgmEnIKwiGs97F2pQ1rLan8z5GgBhkqFpSnU47ZijNgPUNsqBbWsZuOImoQq3ujfjh8vNvpoNe52hhxjTBgjdJohR/eln4dLFXLtFkg0jXnXpJQhl9PaEHIx1uR+NgAXFYQcVcht66FCbr2W1ehzB0d5ZkhlR3YfbciDhlYZlkNE0c278LUAFjnnT7d7IOf8KXjW1uvJzx4BsAYgGsQyxBAJ4Sgk31qCifkgg35I08H7mX6Da0iDVKHUIcEEka7sjwwyIVeXFXINYaWt3flNlVBHK0cjv6+0IORGzNFIgyCFbFntJOjdh5AhJxNytkjIjZihQm7VWsW28YJAtgDRlUoV6L7ZzG6AU/NIFaZZyEz9GwBgoZVCTpEhxxgTyIaKwrYqW1YBYLESXceKljoQ0srUxFIHUsRBFXK5JiE3lg3D/z2FXEjI5b/+h/i9xffip/W7UWI1LwuuiQOYwt/Yr8Dh134Rx3/6g8HPNxU8NaJg+4nZz3TSKRJy5FrT1JZVlUqH/v74apOITN2yGlXItZrslrJGQIg0bDeinIwQck07sU9q0GvDis2Qaz8kk8nq9RJyGV0DlxRy10z/FLaWtrZ8nqj4S3bPFBpEu2xZpZ+NC+Ra6nmpg6a1zJCr11sTcvlMuH1zq+QcVuyfOIVcrwi5rKLhWEbfW1bJ8Vwl91YVmU1/5F9y9Nrr5wQ5CSEnEjW0ZbWzbU7astrrbCulZXWd1z8dI3RqWRVLNFxwzoUGzyQ8TzlnRiyr7RRy9Dqhx0XIwO6xmkpU8jtwXI5nFykh1zuFXKelDgCQUZD+g5TNJivhRcvq4GxnP9Dq/jNEFN08O0oASoyxtiOD5mOKAEakX9kAYtV1QwzRDirJ90YvdaDvqURsOUPLqmRZ1cUPwnaIs6w2UhJJvYZsWWWaJbxXUQUQfT7NCvOz6CjEllVDsFuVs2PC6m9UISeqi0xhdTDdflxsYVkVCDldVMhRlRtFEmUsfe7xxf+fvTcPk9yqz4XfI6n27uru6enpGXtmPB57vBtsbLDBGAzGxuw4BAcuEEgc3yyEJCxJWEII3Bu45AbuZYkDyQckAUICOEAAJ4AN4bIY2+AFr9jYHnvGnn3p7qruqtJyvj9Ukn7n6Eilqlap1dP1Ps88011VrVJJKumcV+/C0N5/hf97YeoWsMLBUIYczYeSFXKe6kzIZ1Eo31RWVlmJB4QVctRePF0r+aQPIGbItahCTg8r5A7f8rdYuv8b/u+1TnCMAIDhBEODq6y34wPWa+Acdz72tw74j3uEnJGAmCnrASkhWlapQo6omgTLqiJDjhBuB5vuNhq0ZVVQJRjxMzBa7CDnyM1UZqCz4PzskaV+qQM9HiPaiBMp5FIm5Bhj0HhwHHGniF/d/vqefyfmLyX7ri92hqeQq5EJPv1+DVv9Y+hMaFmULatehlzJ0JQTskoh+NuDDdGyKkNFyNWKeurkpodEllUzO6WVvE70ZpKK9FFbVlfGQnbixImhx3SmC0RElKKkM6CVMTZDTnIYDBMqDnG5DcvPO2MWAHD8ZAWnb6r3eLUacqmDLdhVEbKMqzBeNgTLKncMIaJBhcjyjgGzAgcBzaNstCzsnW/5Y7b1Y8WhnVOAISjkcmRZLcRYVvOk5MsCRUVG4wjRSPMofhBuBtzVCV57NVx76y+9BxhjdQB1APtTXKcR1hhUkxux1CHzVVo26B2WkUIugJxfJresqgoMhL8nEwmvvh4Q7SF5gGxZBTOFiUYvwpmqwGRCrm3Z/oChoDOUDE2YTE6W6gB0cO6+h81tmE5AUAk5NFKGXFLVjIe4DLklO7hzWzEqAuFCSSsKQ+tN0FLb5uOHHdjNU2E13YkTYw6K6/8rRJQdaRNCzgor5AAxz2pRQb4FhEFAUKiaVuMy5KbHipEZcm2HqN+0MeDRmzC1+/bgtbtvQZOTgH3HAaZOROcZb8brCh/CTntLsBJd5ZqhM+xfDC7Ps1V3QiRag6HMwotSyNHvIIfaslpS2OaoQu5oyyUfmd67ZbVa0OF9XRY7NizbCZHKcdhEm1bnl4TndE33SUogIOS8CURUqQP9eZAMuX4sbFEwnICs7Rx+Jk6aDufhyRgkCmKYCrko9djQFXK6JkzIO3ZHPEd0jwOVOg4QlXC9Lavhx9aPD0cdB8j5jRGlDhkSO+46kVKHHhlyws0qv2V1ZULWVXmMslJIIIjIeloJVbyh5ccQciYpr0rjHBKHYSjk3n7Fafi333sG/vOPLh74uJNLHfqxq3pwm+mDczJ3yj3XR1TIBd8rK8NGTrkNfndGDauAWiFHb2bFIe+WVV26KW2v5VKHmAzLEcJI8xv/jwAYgP/DGHsrY6wsv4AxVmaMvRnAhwFwAP9Anr6w+//dKa7TCGsMqkYbvUcLZd5BJ4z0rtZaJ+Q60iBV01hfk0T693S75u1OTihDTusIkw96k19FhlCFnGxZpdbJWskAY0wg5FxVFRMtWSRfTMy2YcLd+zRLHUKWVVLqoCpmAJIRBvRv980BAIN96DL/MaOyE0cWTYHcpcQXt2sYKwbkoKe461XssNCyYIzfjbFT3ofK1r8H4ODIYrxCrqqP+ctizM3NicqQaxHL6gse/ijwmSswtfs2/7EjuoZFsn2qL7sW+IPbUbz8L/Cal71IaI5lmrteMiHnkU+MidmBpiI7kN4R79jB56TNjQ6C46qiU8sqbXrsljqQDLklyyXGtASWVT/zp4tG25LCzeOHRLN1QsjNtUPPU9sqt6soGpo/GZWzZTz03bIqWbN7qfqSoLL0bFiNHTCPnovOwUsSkTy99rkKokIubUJOvbzht6wygZBr2+2wrR3RhCElJpaILVRlWa0qVIXTAwbaJwG1h0ZaVgWFXLYZcrS1V/XdUV0DBi1IWC5UCjlZKSRYKAlhZvah4o1anjwhbks304YJZanDMhVymsbwlK1ToaiKfiDkDtsO6GmsV6GDh7GSAW6Ng3ezdp32TE+Cs6RTontlWlbHpevgrozy4wAFEa0VEqkRAXXLap4sqwXppnSWRR15g6xAHSEeaR4dHwHwHbjKt78CcIAx9n3G2L90/30fwAEAfw2gCODG7t948JR130pxnUZYYxDtCMdGqQMdjAkKuZwRR1lDtqwC0SoUFdomzZAjCrmcbdeQQk4zhQu7ypbjweGOT1gArkLO4cHno/lmHlFBCad1Zbd9k5NiB1qwIOfQRNnykoAWFkxUe1hWiQKq0VFbVo0E9llKyHnW062106Axd/uy4iF07LZg16L7g9tVTJTEPDsAvTPk2haK098D09swag9BrzwaalAFJPUfsRWOlww3+4sScu05YO/dwA3vxbqFW/3HZ7sZhJNksH9kw2lYnDrB/7226Vx40rHnn7kR01WiMOsScgVNUxJygJTXp9jvdOJJ9yWd5DtEIUctP4JKxwy3rHrrVywGfx9lWQVEQmuhZSXOkAMkhdzcUuj5jVWRkKOEBs2Ho9ehfi2raWfIAUBV34ClXVejtefXUNLLwmQtClGfJw5NQshV0rasRqyzagKXJhhj0DUd3Ane58BSYO32vqNRhQxRxAS1snpQLWNY+XFAPltWhVIHcm5VkT6qYh9BIZehsqZerGN9Zb3wmExMCJlLEe2bfbWskte2JUJVzuAdJpSlDkO0RCYFPV4HV8gVAKeMpcf/GzpHLkB778t7KuQi23QzLByRSx12Z9SwCoSJ6KT5cYD6HJMn5Zl8XRRbnfOznllAzGhcfXPvrJHaWZhzbgN4MYAPAGjBzYi7GMBV3X8Xdx9rAfhfAF7c/RsPrwVQAfDJtNZphLUHVSiqFkNYrAZYI4WcEipCLsryoUJUhlzeLhzKllWqkIvJkKNkHAB0nI6wvIV2QGKoCLn11XXuD5ICxIM8UaDrlVQ14+FoXMuqpJCjhNyCqbasikR8eF0456K6znYHiafMrsPW8a0AAMY4tNI+37ZqOqZPknHOwO0qJsthtV6th0JuvrUErRTYh1lhPmRZ7dgdn/zUmY5OJ1hmvVJAQWdChtx8cz/wiYuAH34Yi6SxdMJxgMo6TJ34HP+xo9MnoglC9BPFGWMMZx+3gfwer5DznvPQi5CjCjmqrrGJQk5oWaWkgBVuWfUUfEwPlhtHyFHr4HzL7CvcPC5DDgA2jkmEHFl38c65mpBLpJCrpNuyCohtmjPjpURqhSSWcBlLRNGUukIuguArZaBKcFVywTmLnne972gUYahSwgFqwkK1jOkhEnKlJAq5rFtWhVIHYllVTMrpYaxuWc1WsSIXOxQ1iZCLUJT0o+KloNZIeTxkZmg1VlpWh0yUJ0FchpyK4FXBu57YjdPR3nslnM5sT4JTDrv3FPi2Yv4yLMiE3K7DwTlry7psFXL9EHKqYzVXCjmqHLe5OCfN0XpmgVHLan9I9ejgnJuc83fBbUp9FYAPAvhU998Hu49t4py/k3PeUfxtm/cKfhphhBiIJz+FZXUVKuSiMuTWemuNStlSSFjsYDvBhZIxcVKUN+WhSiGnRxFy0vFN8+M80Bw5lUKOtnXOVKfcSY2jtqzKpCgdGPWrRk2aIVc2pFKHKIVcj5bVtt2G3b0npKEANwIVOHnDOE6ePNl/nVba55Nl1M7rTrQ1TFdFQs7hjpD11FQQcoc7u8C0YNszvREi5Gh+XL1YxwJZznGlNi5euB5fKHwQhe4ls8WAVvdYmCef/bbT3gu87QFMPevt/mNH2kewZAbbVLZ4KhVyuqbMkPOe89DLsiq0rNL8HDIkoJZVFSkgKuS6y2MBQZaUkGu0LMEa1utO+0ZqWZ0PE3IvOvFFKGhFcMeANX+2QMgJ+Yq0ZVVxzYrDMDLkqC14JmEmmdhqneycSc83UUTUoIhaXhbNnwVNCzUtAkBZrwFw1ytKdaiyoQJu3qEMVYbczNgQLauJFHKEkMtgW1PFI7VAqyyGYr6q+/9KtawCYduqrBQSSxhotthg6xxljXSXn6FCTrFv0v7+DwK5mIaOn1Qkogoqy2wvglPXmHjzsjtWzbLUQY5uWEmFXNJCB0CtkMsiuzIp5BzlLFWPeUOxD4HECN4MJGVwzucBfLH7b4QRMoNoR+i2rJJz9eq0rNKW1VGpgwelZTVhq488GM1zG1CoZZWZwkAyzpJN8+M87G3uxVnrzwIANIhCrqZQyE2UJlApmEJoMbWsynfu5cr3pGhbtp+fpEs5X4CokKsYFSFDbtCWVfp3Gg8GiDs2jKHMd+CGx25wl1Pa6+e7yQ2rADBVLaFqVv1tvWQtoSoFJsuYcx4VfmdGA0elDDlqVx0vjmO+0cDztVvwcv1HeN7RO1CACWjAhH08DnaVKXPFKso7rsDh5u1wS8uBhZnnAHoBk+VJf3lHWkeEY4MqzgCxNIFpHZx1fB0O2v42K2pFTJQm/Nf0UktFEnL+JN8G7xZcaEwT7pirSAGVQs5GcIzIn4dCblrtR32yUbCshgm5U9edir+/5Ot4xd/+CHCqKE8Gy4uyUFP1ZpLcIjlDLo3JCFVjJbVADvJdFzLSEthi+0FkqUMWCjlDQ9sJv39FD85TkRlyUZbVhAq54ZY60PzGKIVc8HgW2zoqQ05lP1VdA1aqZRVQKORiLKudNCyrcaUOWWbIKfZNOW8KuWWUOsQtNwpFQ4PVJZQ7tuP+nmG+oaiQs4VCmS1DzpBbnmU1fNxkaT3vBUO4UcUB4kQoZHy+WWnIClTOeeKswLWIoRByI4ywUlA11mmrvNTBEjLk8qvkyhqqdrekxJqce0MHpHlqWe3YnbAlU+tEKuRkS7ZKIUeLHRpUIVcOE3KTpUlUi0ewSBVygmVVJDYLEcH1vSCr4+SL9kAZcj3yBOm2cWxCyM2OodQSFXJzXfWampArorZY8wmuRqchEIqqDLkmJEJOb+KolCEnEHKLR3DBdU/HM4vzkFF3HBzsqnDmfvN6TK4/HdbnznfXkWsod9VmU6WgSfPA4gFfHVjQCijoIslTKQQD8l85bwPeedGF2Le4239spjoj7CPxbn9yhZz/PdRIw6peFpZN74i3FS2rLiFn+xl0GtOE52XQfbPQNvvMkAuWq1LIAYCOKuC4hCCddBoRmWtWnxly9UoB1aKOxY6NetlIPHGMg2xZTQLRmpNUIUeafTNqWc1CtWVoDC0eVnmUtbr/c5JSBwrV9lFaVmsZtaxGKOQ6GSvkKElIL3cqRZM89uOcS6RHtpNDuWk1rtQhDcuqyhrpnVuzVMjJNxoYyyZvsBfEjCtHVMglJA7qCkIuyU2SkqH5Cs+2aWOsZGSab0jVtnNLJvaQTNTjJrMvdUgKZctqjqyghqSE58SImMa1ejVB7xbt+YU6NkcxhRKqYxX5OYpHGCEFqKT9q77UgaxzbaSQA+BaM+ldY29A2SvLyoNos9FDuQ95ASWAPIRaVsn1TXb8qxRyAiFHWlY9SxW1Sk6UJlAp6oIdi5Jjcvub3iPcPwrzMXZVQFTllYySYElMQyFnWu4AUWPAietr2DG1w39OK+31ybIjbULIWR4hVxDWp2k2pVKHsEKuo+0WfmfGgqiQ23cvFn5yrf/reGM/CqZIxu2unIr3ma/DI1Zgg5rjpmA55nYFeve7USvU/IGvxQkxomgkpZbRzdM6JiqFSLsqoLorLIIOwFUKOUYy72h+HCCRAl7LqmxZJQ2rNaMWexeWKhoWWpZwnPZqMJyqFvzJ1kLLUuYD0qwtSnTpCosSIGfI9R6S6RrDu150Ok7eMIZ3vPD0nq9PgjIhf2YGUMglva4udahlNWWFXASxVdSHr8Qp6JqQs+k/juDGgUpJA7iTd9XkW6WcU+XkrR+iZbWkyG+UQa2smSjkInLqkijk5Bu2Was1TqyLltWQQi4iQ66fnEuKKGskEG6pHybkxVcKei6UMvINXFEhl2wZg1hW5dd4+2JQJeQgoHOJXUcW/fzh2Xpp6OrFZVlWFaR/niyrshJ+ULv5sYI8u4/yhtQVcoyxTQDeAOAiAMfBLXKIOgo55/zUtNdhhLULpUJOKHXIfJWWDTNKIbeGCTl5MOkN7nopdTzQu/0lQxOk5B07P9LqUH4cADBTUNuIkw7xpb0z5EjIeskA51zISZssTaJaMITA8thShwTNpipQhVxdRchRy6qezLLay0apalg9YbqGkqFjy/gWaCjAgQmtMI89C4cAbMXRVqAe9BVytSLGFoliz2ygViK/S6SN7djghceFi6KmN8EaB4Cb/ga44wvAvrswX6sCG9xWvnp3xz7mzOCrzkUonvNr2FPYin+86VGU7X+Ct8Xm2nNSzl3Vz+9hjGGqNIX9S/tBobJ3UoWZF1AfVegAyER4+HtHSTaRkFMr5IS/FSyrilIH1gEjhJyKYKSQLav9THYZY9hYL+Oxw+73au9cCydvGBNeQwk5OnkQGoiJetTuM0MOAF5zwQl4zQUn9H5hQowRomdDvf8MOTPhd51aDI8lhVxBZ0LOpgeDEHJRpROAS1AsSOcJlXJO1xgqBV2w/g6z1KGsyG+UsdgRryHDRhRZoFTISS2rol01++v7huoG1Ao1/9oTUgpFKOSEmwZ9TupV1kh5+VmXOuQhPw4Qb8AMWuqgaj4uJthHgnqxOx6lroIsW1bpfdxh58cBYQKuL8uq4hqdK8uqNO7kZHXXWqkD4B7n3vWqYzkYoqB71SPVqydj7FUA/h5AFTEkHHluFdIjI+QZluIO02oudbAd7l8sGRMHoyNCzgUd2MghvdF/T3JvDA1a906yRyJZDs/F3SwVIcc0UxisxbUIKxVyjUAhRyeBYyUDDbPhWxmrRhUFvYBqSQe3IzLkJNuLLNdPirhCB0DRspqg1EGsnw+vi9Cw2iXkPHLF0AysK2zBQfNhAMCuhYcBnIvD7WB/cNsduE5Vi6gVRcXeWCnIa5MVcg8ffQxMbwuPbSjsxr82fwP4VrCeC2T9x6dPwec2/Cb+7GfupfVt63bA8EogSNPqXHtOUDjCrgiE7WR5MkTIqQoQVITcvsV9wfpKhBwltFXfuyiFnKd4i1fIhXOsBEuqZgrbM67QAQgr5PpVilBCbt+8ipALlidYViNUhNYANqm08bJzjsNX73gcYyUDzz9zY+8/wGAKucXO8EodorLYsrDGGbq61EHjwbE4FqGQA1zyTSbkohSEtZJIyCVVNA6CJJbVhnQNGTbKEQSralJOH3I4lwodsp8cM8awfWI77jp4FwCFZTUiBH1QhRzgjm+8713HcoCSapnDPe/I5FYe8uMAValD8FzyUodBLavh+Jks8w2jbhBsHnJ+HLBchVz42MmXZVWMpqAxSXkiDrNCYaSQS4zUjmLG2DkA/gmuIu6zAF7ZfeowgCsAvL77uAngIIDXAbg8rfcfYQRAtv+sfsuqMGjStFFrTVnzlM8AACAASURBVBdRd3eTNv+1BIWcHvrbvFw4lAq5kGW1v5ZValltSpMpudAB6E6cubplVZ4oGD2ImSjQhtHJHpbVslFGzRAtog4P7y86RlOp9QSFnO2SQDsIubKxss3/ec/STnc9FQq5yWpByLRbNBfFwGSp1OHOffeE1mVRt2CQ8F8YZcxvOtv/tX7y5bjPOB3evax6peCTsh4xCABzHZGQ445IyNEcOQ9KhVxhcIWc6hxLVW/0+AkUcoFdN86y6n1vdU33l8kYB9ODvL1ehFxdIOTMvie7tNhhj6LYgeZTihlyasWmoJBboQH7M05ej1ve+Tzc9I5LE5c6GNJ1VbbLq7A4RMuqpjElyZeFncnQmHCO9OEE3y1VQ6oH1XpHEZb03FLQGeqV4ZFgKnUqhWU7xHaejfIpinhV2b3lllUrB42HtGk1zrJq2sF3atAMOXmZ3rnJsh3foqix4St35OysqH2YNWSL8GClDuHvfaJSB4U9OcsGYE9tK2NLBgq55ZU6qDLk8kN0GVK2qkokspZQknIsR4hGmkfHW+Aq7j7KOX8D5/y67uMdzvm3Oeef5Zy/HsCFcGtH3gvg1hTff4QRlBe0uND7vEMOII5rzVpLiAokli+GURAy5LrbVMiRs/JxnCgVcswUBot0LCKXlqgUcodah9CxXQKEZsiNlY2QXRUAKgUDPEGpQ0FnPYmZKPRSyHmkEOCSO7qm+0QSBxfVbl0ICjnF915lWd0xGxBrW8eDAO6DHbeEQSh1sEipg5RpVyOTb7/UYWEv8KOP4LH/9+7w59M0LDIGa/OFwEs/DrztQSyc+nz/+fHieGgbeYQIJeSOto9KllWJkCuHCTlKvvmPUYWcGSbk4jLkVK2ygkLOIQo5VYacbFk11KQAtaYyY0H5uApjkkLO7HOSvokQcvsUxQ5ihpzaWi62rPY/CRwGJkg+XhIwxtCrOEXGMBVygNoyqWrlSxtFQwNXtKx65wgAGCtFq0AqCnIySkVElS3TtdJQoxV6taw2yf6sFY1MYh5UChlAbTHUJNJYDM1fmckxbVqViQlNY8I4JMgWGzzvjVq2vXFTlMNgWJCVv3mxrMqlDoNYVpUKuQT7iO6XgCil16Lh7xfV+TILhdyxXeogKuSytCHnEVHN0SOEkeZR/Cy4FtQPSY8LRyDn/A4Avw9gO4B3pPj+I4ygDMimk5zVZlkVBpC6FmrNWo1wHI7/vHsPrr9rz8CKxSQKuTiFltyyKi8nLxeOQ61D4QdlhVyMAlSlkAOAfU3XftiQ8n8omSMq5NSlDvJ+EFVAw7GsemQRta2qCDlBwdMjQ86zrO7YEGTTnTQRNK3OO7sASKUOtpqQo6UOJXTwpLnvAZ9/JfDh04Hv/DkecKTW3C4us9+DJ678CvCU1wHlutCyWi/WxZy9csEf+FFCbr49Lyrk7IowGZokVloPVG3ood8Mue3rg2X88MGDoeXRiadHBgMRLauSQo5OXCjZRddRI4QcVSuqMF6iGXKm0KqcZGA/W6cKuaXQ85GW1SQZcqvM0qJHqP6isDjEDDlAXeywkgo5ywqO0VqMQq6isGFGbR9qC10/PrxCB0CtTqVoCBmk2ZAskZZVxWRXl1pWTekG50rg0q2XwmDuPrzouItCz6uUU8tR9imVWFa2qh35RkMeLasd2xFuaCa1rNYHLXXQw2N5M+OMQ5Vqd8u6vCvk8m5ZFUtUhJzlFboJsJIYpI19rSLNo2MWQItzvos8ZgNQ0e1fBdAB8PIU33+EEZQXNHquXm0KOdOOUcit0pPbd+/fj9/53G34vc/fhu/cu7f3HygQnSGXbIKoIvRyaVldUivk6N1TedJBoVLIAYFtVW5ZjbKsRivkxIG9aBlOTyEnW1YBYLwQkGeUvPIQpUjyQMsguFMCY8BJMwGZc/p00DfUwuPgnIsKuQjLaqPTwPq5u/E/jE/jltLv4a1z7wce/DbQtdXeXwwm0QUe/LxHK+PoUkBWUWJtvDiOebKv6pWCf6zzmAw5bld7KuSULat9ZsjR3LHr79oTsi/SAbhoWe0q5OJKHcgAXFC2kibYYinYl8POkKMKub0Ky6qgkKOlDvS7Qb43QhHRKruDLhY79D5nUoXcMAoAVDbYlcyQMzvBMRqXr6Za72jLavD49JATsqPUqR7kyIMsEFnqoFA0CTdjuaSQW6Hv2raJbbj+V67Hl17yJVxx4hWh51UuiOVlyIXPn207fENymAhZVnOikBNdEQOWOgyokFPtZ3qjJgtCbqUUcsvJkFORnXlSnokuEScX55yVxMjVlRxpnombAOTbxfMAxhljwjecc24BaAPYmuL7jzCCeEFTWFZXnUJOuEBrQsNQR2EhWQ24Y1dA+vx891zMK6MRaVnVkpFqomV1dWXIMc2MzpCTDu8ohZxHyMktq5SQ8y2rUoacR45xzkPZNv0qZjzMkQy5iWrvUgcAoSIFGb3ss3TbcKeEzVMVYaKwfeo4P1uOa0vYt7hPVMhZNVQKOsoFXVTI3fJJnPS1l+J1xg2YYOL2P7Dt6TjUPd64U0QFp/jPMaMhZOlRknG8OI55gbQ0fIJHzpCLs6yqFHJJWlZtx8ahpUCtKRNyl5w64xMIDx9s4hf7RIJUIORoy6p3DiOW1ZIhDtYF2xxR6RRYQIxRQk71eSjiW1Z7D5hphtxepWW1t0JOtKwGr086CcwL9B4qVBmL7eEq5JSNhxkQDkVdAxSW1XY7OFZ6lTrIiLSsUoXcEAsd3HWIb1nNutDBXSf1dlERGPTrZDv5UatsGtuE09adpnxOddM1aryTBCpXhXwjbdiQT2t5sazKjgo6TkiqkNM1FlLm9lvq0FYqIbO3rDIGbJoYPiFX0JfRsqrYtv1+J4YJQ7hJJapy12LLah7nVXlFmkfHbgATjDE6QvhF9/+n0xcyxk4EMA634GGEEVKDqqVImDSsNoWcJVoVjoW7DbQhbtATdDuRZTUhIVdQZMjl5MJBFVk+tI5wTMdaVolCjpIxvkJOmlDNtwN1Vb1YB9AdPDvhUgdLyr7SpfybobWs6mGFnKpptR+FHJwSTp4RrY5TtSLsdqD8euDIAyGF3HTVAB76Lsbu+jf/8aZEou7GBuCSdwB/eCfue947g7dsbUJVD/aJpjdxdCkZIVePyZALlTqQ2dC68rrQdkiikDvcOuy3706VpkIZMOWCjueeFpB01/98j/B8FCHXVijkhAZViJlRbXLuMAghpxcGU8jNLZk+ic1Ysgy3jb0UchGlDlHlQvL3aDVBOGf2+L5zzrFoDq/UAQCqCgtWNgo5tWW11QqO+6hWQ0Adcr8aLKvyDZ0sUI7YnyoCJaSQWwV5TmqFHCHQjP7Wu6RYXlTkx7CQ15ZVuSTNEUodki9HJtuTEXIrW+oAhEn0TfVyJsfDsiyrCst6noguOSrFFm4C5POcM0yorNkjqJHmUfzz7vLOI499C26G3PsZYzMAwBibAvB3cPPmbknx/UcYQZjceCdGTWraWk0QL9DaMWFZFQm5/gnSz937OXz4zj+FVn4cQLRlNTZDzgxbNoQ8kZyXOtB1pWORkGWVqMBOmjzJ/3lv07UKy4ScSiFXLRrgCoWcSllEixTStKwu2YH4WpUhp1TIRWR2eZBbVuX3rRR0sE5AyN22906YTnc9HQNv1L6BL5u/B3z2SlQfvy1YF00DL9TwRevZuKr9blxq/V/gkrcDU9tw/+H7g3VqHYeaERByrkIusKwKhFxhPJQhV1Ap5No9FHJlhUIuASEXlx/n4YVnb/J/vv5u0YpOVW+9FHIhy2pEsLyGYJlcD0jI3i2rwX4+TLZ3QdcShdLPjJXgbdKDjY6QRwkASx01IRdF+DurOENODrCOQ8t04J2eSoY2FPJRRXploZ4oRFhWm4SQU4W/e1C3rKpfPzMeLHPz5HDVLKrgeQoaeZAZIdeHQi62ZTWneU7iOCRsWe13vUuKMeNyLLCDIK8tq5omFtNQwqAftbLctJqoZVWhXMz6+JS/s5szaFgFlmdZVWfI5ee6KSrhHekmQD7POcPEsSAiyQppHh3fgEu+vYI89jEABwA8FcBuxtijAPYDeC7cptW/TPH9RxhB6ddf3aUO4mTtWDi50Qlrv6Ti7oXd+Ktb/wo/P3ITSjPfAiAOOIWcphjFBn1fValDHhRynHMlIScr5MRJR7RC7qSJHoRc2cBch7Sslr2WVUkh1yVUVMHQYqlDihlyCssqzW1TZ8jFk4Nyy6o8iGaMocyP93+/ec9P/J83Oi38ceHL2Oi4uWpjhAhtbjobeOv9eLv927iFn462FRxPlJBzWsehXggy3WTLKlW6FbWa/xlKhoZyQffPbzRDTi51gF0VlCNTJUWGXALLalx+nIfnnLrBn2j9cn8DDxDbaqRltUuMU4WcbFkt6ppveRJsRU6gDrJYcNz2almlGVyDWMEMXcOG8YA03D/fFp5vW+oMOT2CvBIz5FbXgL0fi3pzyIUOQLhYoGgkI1mXi4LOQpbVsl5GsxXszzjCSmVZjcrZ+rWnbsE5Wybx9O3TeNm5xytfkxZKRrxCjl4/xlc6Q05ByMnq8X7t6SsBWVHCORfOEf2ut6rUYTkW2EGQ15ZVQCTPqC07qWUVCJPtgyjklrufB4Fc6rB53fDtqkC4ZfXYsqyKzox+W9yPNRQF19Lqmn9njTSP4q8BuBLAN70HOOeHATwPwB0ACgC2ANAB7AXwKs7591N8/xFGkBRy+besdiwHX7/zCSFXjUK+k6kaXK02UELO7PMz7FrYBQ53H7KCu83oNkmukMtvhtyuw4v48s92Y+/CvFBm4EHOkGMJM+S2T273f97TdIP3m1JLnlDqUCSlDjxsWVWF4RsDWlapVXOyV4acHibkeivkeresqu5ejutb/J/vP3Sn//MUPT4qUxg79SXBulSnwMp1Qa2z2HaP+fsO3ResU/s4TBQJIacHhBznXCDWHDsggDzC0s9A4kUwuMdwy27hwOIB/7XcrgjbQZkhpyCwKEmXVCFXKep4zmkz/u/fJLZVOgD3WlY554EigQVKtYouTgoYY8Ig3CO8uBMQdzaC46NXy6qha8oJYT8ToNmYHDkhQ44QGtEZcqtXIdePzd/7DgDDsauqlpuFXRVwxxpcsqzWSxP++bhkaLGqGdXxGKUi2jxVxVffeBG+8N8vVDY8poleGXIrYVmVoxE89FLIOVyeHOdnEk9Rkm4MysVe/RLMRcW5U8h+XYlSh5wo5ADxHEbP3ctRyPVf6mCHoguyuJEgK4pXg0JOXeqQn+8yVTZatmiTX4stq8eCiCQrpHZ0cM7bnPOvcc6/Kz1+F+f8PAAnAXg2gHMAbOGcX5fWe48wggeBkFsFpQ6f/tEjeNMXbseV1/4Ijx5qhp6X75gdCyc3alntx9YIQLDjMeZOBgTL6kClDuEMuZWyA7ctG1d98ia87Ut34u1f/ZH/uJD9JbesxihABYUcsazuae5By7T9wX5R11AydMy1gu3rtay6pQ7BwE1lWfX2gTwYSYL5lokDC23/s6yrhXORlAq5othsKqNXhlxYIUcGwOYScOe/4q8WPhUsg5xHpmwbP7TPxBdPeC/wlvtRe+abQ8ulE9RGx8JCZwG7G7vd9+ManPYspsh+ZUbDb1lt2S1YTvf41opY6gTbtd4l5ALyk6GAYFt46kf3c1WE85+qZbVmhC2eBb0Ag7nrb3Mbjzce95+LIuQA0bb6H3cHhBy1oXrHT8cOLIy6Hkzsvf1Locqysm31IL6XZRVQ2wf7GdRvqkfnyIktq4SQExqgg++OMBFbzaUOPc7ltEFYRbqnAVkhlxUhV9C1UIbceKHu/9yr8EAmKGQ1/EqBqlMtR2wpBYDmkFtzo1BW2NZUFujYltWckt/yGG+59tJeGXKlLBRy0rYu50ghR7f3Sink2pYTcsNkgbBldTUo5FaXZXWtK+TyInRYDUjtTMwYO6P7Tzka5pw/wjn/Aef855zz0V4ZYSgwFQOuPCvkbnvUDYnnHEqVnGjBPUYy5JZhWaUKLnRtbgIhZ6gnvTKorUyVIbdS0upHDy1iT3eCf+cTAQmysbYR4O5nY5oFjZE7uTHHN1XIzVZnfSvikrWEvY3ADutNZKll1SPkakUDnFhWPXKMDuqVltWEZOtdpGn3tI3jygEXVQp6n2G8SEodeinkFPtTJuQMTQMOPgj85zuAD50KfOW/44LOPdhgWaG/vdU8F68134Xdx78AKJQFVVZAyAWfo9m2RLtqexbgBtZXpv3HqEIuXOgQrMOET8gRqzaCy66nIAXCGXJFvRgirKIsntS2unN+p/9zHCH3nFM3+KqaB/Y18Mv97uegd8Atx4Lt2KIagRBy8t1zQJyAe5Mm01QP4ntZVoGwogHoz/YSV+wgEnJq4jxKIbf6Sh2SnzOpHXt4hJyskMtm4q+yrNaMicj1klGRlCpRdtWswRgTv3vSTcCFFs0gzW6dSwqFlYrMDrWsCjc4V57wVEEe4y2XkFOXRFCF3PDPOfK+qeZIIVeMsKz2c3OkLhNyfSvknFBedBaQbxRsyUghZzADDMH27S9DLrxt8vRdFmJzbPEmQJ6Iw6xwLIhIskKaR/HdcIsdsvlGjzCCAoL9p3tijMvYWmnQiZmqgaYjEYwlPVyVvtoglDr0+RkoIce6QfB08GNoySaIwh3i7uBQyDpYoW17sBHkUS3aAVG1rrwOGoK7ig6x+GkJM+RqhRo21QIF086jAeHntYSpSx10Qf3hZ8gpBhpRKqA43Lk7eM8nbwlbKi3H8ssUGJg/eKPE0nJbVp/H78avP/D7wMfPB35yLUCUgid3woRc03aJtKmumo+ui0fI0cFuQybkWscBADZU1/uPMaPplzoIbbelulTo4C6XNnbpCN8H404B4IVQW5xsW1VlyAEiIffo/KP+z3GEXK1k4JJTSNvqXa5ajzEmEG0dRyxD0I3ollVAbZ3rmOkq5PoZLFNCbk+IkCOWVaqQi8iQsxWq7tWC/hRyhJCrDKcdVLZgZaUyM3QWsqxW9OCGQS+FnGxZzVPGVpxttSmVAq3EOnlQfXdiFXI5/a7JsSSyZbXv5Slu4madISe/RV4IZ0C07FLCuR93YajUIcF5p0jG8h1bUshldGyulEJOHg8st2U1T4QcPeeYDpesyPlZz6wgWOZXqYgkK6R5dMwDmOOcH+j5yhFGGBJUsm96DsybZZWSGqq7B5Yt3tE9Fu42iC2r/X0Galn1mhmjW1aTWVa9AWkepNUHGwHRxvVAJbWuvA6MEHI2D4g7KqrhMQq5qlEVCLnH5p/wf64VDViOJSiz6kXXclUp6qJCjlgOPRQU2zCpQu5Oogx98uaJ0PO0BKBslP1slfFCD4WcTskCcX9yzrFIFHLXGp/AtoWfiQuY2obvbf493Lx0cWjZ3HJJrKlqmJDzyMGqlCEnN6wCwMYxqpBr4siS+1kXTFkhFy69EGzLPJyb5pU9yIHacrFDFIFVKQSD810Lu/yfZ6uzytd7eOGTSNvqXYFtVSh2sNpCjqOm9VDIKSyrrY56EK+y4MpQkQf9DOo3EUJun5whF1HqIIc9e1jNCjl6DJo9MiPnSKPtxJAUcjKRla1lVSp10Aa3rOYpY4t+9+SbgCuRISevkwf5PAeEW1bNVdCymrZllRI/3jlXIOQy+I7I+yaqmGMlQLdpm1pW+1DIyd/vJCSn2GDsrIidmqr4dY0J17Vhg9pWjyXLakEad5orYEXOE/IgdFgtSPNM/ACAMcZYdt/oEUaQIFRMrwLLKiXcVIo3sTKbhe52ygTMcvG1Ox7Hyz7+Q3zpp7t6v3hACKUOfVpDBYWcZgNwJEIuGSEkWFa7A6NCDuzABxcC8onpAWE0XZ4G45SQCwgaUaUSLMvhDpasJf/3ilFxra9dPNEIyJLxshGySeqaO/CpFg1RIWd5Crlg+3r7oJcqTYU7dwUkq0ohJ38GD0KGnIKQU7asOjbwwLfR+uerYHeTE4oOh//pmAac9mLgtdcBb7od9550NRrtbaFlc9slfTzrXUkv+ZlrHaeDjt0RM+TaFu47HBQ6OC23HXGiUka9W57BGMfRlnt8y/tCUMiFMuQAjYdVbh4hJ0885Rw5SrwJj5Nt7eXZAfEKOQB47mkb/OPh/r0LeOiAu2/kplUxryf4fKoMuZKgYnD/bqkVQcglUMipwvD7mezO1qlCbkl4rqUojAHkBmKaIZf/XKsoFPpRyFHLqqJJOQ3IE+MsCTl60wIAiiw4P40pFJkUsmJItrCuJEQyXFTINfKkkOvRsuo4UsB6jibxFEVDVE4tO0OuEB7XqG6mDRPyjYZhlboMAvr56di0n5sjsuI6ybElKyFVhXTDBv3ObpooZ1qOQMcDyy11yKtCLmxZzc96ZoVjIWYpK6R5dHwObpPqa1Nc5ggjJIbjcKFl0jsxyndJ8wQ6QKQkkQdTUsjpGvM/F+f9lyL0wvu+fi/u3D2H9/z7Pcr1SQNUIbesDDkAYJZwwjcSqtzEltVwqcNKZchRyyrTA5JpXXmdSMgxqpBTE84tq+XniVWMCnRNFxRyexeD8P9ayVDaVYFuyyrJR1KVOqgy5GyH9ySM9823/JbKalHHjg3jodeoGlYBqWVVZVkl26VmHgZ+8CHgo+cA//xKNB+6IVgOd7CXT+GWrdcAf3Q38KrPAyc/D9A0TFWLbt6bBI+Q8xRyjDHUiqJtlWYqHV1axMNHH+7+MYPddvfDeNkQCjvmO0fgOGLDar1Qx3xLoZDTKCGnsqx2FXLSVV4m5JJYVj0UtaKyqZVirGTgklOCttX/6KrkZEKO3oBgvQi5Qljl0Wiphy90P0RBaVntg7zZFJMh144odUiSIdePKiMPEOw5Pc7lvZqU00BVVqpk1rLKQgo5gxSt9FKP5dmyKpDhMYTcSpc6qEL45ZZV2XGQR9BxSDukkBvAskoVYIpShyy+I3luWS3SllVrsAw5all1i1ASEHJS++1KWFaPJxbVU2fDY69hYnCFXL4JOVkUoCoaXEsQyvJGCrlYpHkUfxzANwF8hDH2epZFZ/MIIxCId5iC2nBNyhHJEzpUIWcqFHKK/BD5zlpacByOQ03XVrTYsbFvrt3jLwaDqJBbhmUVADRTaAkrJJwgii2rigy5ITG3vQgqgZAzAoXcuso6sEDHBdsJSICoFmGaH+eRK5vGAkLuwFJAyI2VDGHbUtKlEpUhZ4UnCoyxvlRy1K561vETyrvSqoZVQFTIUUWZB50BF7D78LHCR/Hxva8FbnwfcPQxAMAiYamadh3PbH8Et530u8DE8cIyJqsFOO1ZcC6ul0zIAQgVO9AJ6mONh2Bz97hn1nrAccmpsVIB0xXSoKs3/EZWD+EMOY+QCz4Dc6IVcvI2DWXIJSh18DBTnUk02aBtq16OXJxCTiDk9B4tq5YNy3bQXFJP6qIIRgoVIVfsY7BMFXL7F9oCqRZV6kAH6vT1q3nALkw++il1GFKGnFwskF2pgwYuEXKUJO83Qy5PhIXKLu5ByJDroQIc1jp5UCvkgp9tzsXSr5x+1+RWVPmmbL9QlzoQdXsWLavSNaNSzA+BIras0hiFwRRySQlOuWV1JUodTttYxx9eugPPPmUGb3v+qZm8p4dBFXKGxiDvmjypXXVJCS8Q6jm1yQ8T1DI/almNR5pX0E8A2AfABvBpAP+LMXYrgAPdx1TgnPPfTnEdRljDkO2dHnJd6mBThVwPy2r3ZF40NF9l1rEc1MKRSwNBVqs9MbeErdPpdrRwzpeVIScr5JikkKMql7gJouoO8TAz5DqWgzd85hb8cn8DH3v1ubhg+7TydTRDTlbIwSkC3WubSTLk6NiNEs5LZmCl80iKjdXAsnq4vd//2SXkjvi/10tB/lFVypDzLKtRthdDY/73zLI54uaWQqGDIj8OEBtWBUJO0WzqrmADuPMLOOfHn8S/lh4ML7AyhcZZLwYOfh8AsOjUYcFQTugmK24xAu9Mg5UO+o9zq2tZrQXbRS52GBMIuWA97PZxwWcoG5gukxw5o4G5RVModRgvjuMxRcuqMAhVEHKw3cfkO/3LUcj1yo/zcOnprm21Yzm4d888dh5sCnfEO3ZHnNyzeIVcmU5eTBtHl0xwR5E1p5cT3W1Xtaz2MwkqF3RMVQs4smjCcjgONdrY0CXpaDB4pELOVivkVlvocz+lDnNLWWTIrYxCzm1ZlT4T+U72aiCVCaY8hd5TUrmdY8uqikCRb1ZZjJBbOf2uyTcGl21ZlZRYANAhSrCVUcjl07I6aMvqIIScTJSq8q+zwJsvOyWz96IYVCHHGEPJ0IV5RJZW214oSFEpK6F8zBOOhdzzrJDmWfG3AHDA7zKeBfDiHn/DAYwIuRFSQVQGg5wjkieIGXLxllXvZD4sT75MCMrZSGlAvsNuWoNnyAEAmClaVsm+jts2QoacgpBL+8Jx43378OOHDgEAPvXDRyIJuUOCQk4i5KhlFaRlNSIjsWkFJJWngqKW1bmOSMhFWVbLhqiQCyyrwXuVpH3gfQqXUI6eXPbKjwPEDDmqnhovBhaLBXMB/OAvwX76KeD2zwPtOYTopC0XAOf/JnDGy9E8dBfwLZeQ84gd1WRnsquAs9uz0CghZ9dgaAzjZBIqFDuYDVSLAam5Z+kh/2dzMdgH1YIuWFaZ3sCRxU6PDDn3PekgNE4hJ09UaalDSS9FDoZVhFyv/Dh/ncsFXLh9Gv/vAbfj6Z4n5gWirWW3hO8gJ63BSkJOUukcaXaE74OHKLVfeP3Cn7nfQf3GiQqOdFVfe+dbASEXYVkVMuTIjRbbWZmJWBpIWqIDZJMhJ7esrmSpA+zgfNDbsmpIv+eJkBPVqRTNdvB7lpbVUkKFXJgwzr9CLr7Uof91lhV3QDgKZdgIK+Tyc3wLGXLmYBlyNJM06T6i6l3ZmpwngmlYGJSQA9xcRLqvslB5JoWuizfeooQiawWCZXWkkItFmlfQv0xxkmEUZQAAIABJREFUWSOM0Dei7kTkudSBytRVJJBK7jwsy6q8rCeOtiJeOTiWpDvs/SjR5BZQAGCaJWyPpBYqlWW1YAwvQ+6BfQG59vjRaKJTUMhRy2p5nUCKWQ5RyEVZVqWGVQCYrQUKp6Z9BK54WcdYWbSsThQDtZqmMVSMQI1kOiZsxxaOl4Ku4ReHf4F7D90LvdQGOi5ZFrcPHIdLCjk1IUctq5QkKupFFLQCTMeE5Vho/81TUebi8dTgZXzVvgg3T78cH7v614PPThV1tkukqAZLXtaV094I4B4AAOcM3K5gcqwo2DdlhVyNEF8HO48Eb9dtWB0rGdA0hnXEssqMBo4umrEtq3VFhpxjKwg5pxJ6HQBMlkk+YIy9czmEHADMjgfHTKNtCgNw17IaoZBTWlZFFcOhZgfcCRNySQodALWapx/LKuDmyN23x1Uy7plr4UmbXQWwQMhF5FtGWVZXW8tqPwo5MUNuOJbVmqREyyxDTmFZtcwKAPfc1a9lNVeEnJHQspphUL8qQ0713RHzVQFnFWTIyYRcx1reOqsUKnRivNYz5MSW1cEsq+vHgmvdRMKbDTJRSq8DebJgDguDWlaBMAGXp+1VEG68cZH8zqkqd5hQ3RAYQY3UrqCc83entawRRhgEUW11+bas9mhZVWSeyNkTaUG+ezEMhZxMyPVzx4QG3ftgptBKJgaqJsyQK4SJzrQtq17TJADsm1dn83HOccBXyDlCy+pUeUogIMyIDDm62jRDzlMOFfUi1lfW4+DSQXA4YMYCuDWJsZKBh+ce9l8fyhkrGmg7BpjmTsDadlvYRlxr4NXfvAamYwKbgcriNlhzT8HR1gWYqq1Xft6dh5pYaLnLm64VsXlK3faptKy2F4A7/wXjlonD3d3WYEDZ+zpNn4w9p74Ol333eDRQxelaXVgmJeR8hZxisCQScl3YFQCuZZFCLpkIJuEcc3bQWuy0XIWc97xsWT26ZIYUcqpSB3qsc0ulkHMfk9UJVCEXpyhbLiEntszawgDctawG5wIH8Qo5WU1wpNlxLdzyeyYk5JZrWQWA4yaD9bzt0SN4/pkbYdpBsZChMYGEo9ckM9Kymp+JRRIIJTr9tKwOybIqq7SyU8iFLav9EHKyZVWVkbZSkMlwD5xzNDq01CG7dVZZVlUWQ/p94pwLN2TzqkalmUsd25GaYZdJyHktq9byltkv5LfIk0IuqrSkH45n40QZb3jGNlx/1x688TknJ/qbsGV19bZtDwJ6g65fQq4kff/zdN3UJSW8ag63ljBM59GxhrVH145wzELMYKCW1eA1eSt16JUhp7IWDMuTLy9rzzAUcmQAD/RHfIXsqnDD4MWWVdrokzBDTk+WIXek2cF7vnY3Pnrjgz2tz5xzOEStRQm5Q822cvkLbStYL30JjLk/jxXGUNJLQo4bzZCLKi1RKeQA0baqFdxtyrV5fP2hr/uPnzd7nrBuqmIHSqa22C6XjOvCqO5EedO/4RXXX463/NdbsLe5FzKoOu5JmyciywKEUgfbAv7j7cCHzwCufxvG7OA9G5oG7LgceM11OHL1t/F2815Yx18HaC3YEjmrIuRUg6VKQUdR13xVGwA4pktoTUkqH9my6pEDzJiH3Z2Y14xxcNsl7rwQdErIaXoTc4sdIUOuXowodSDr69hh8iyq1GFrfatvETmhfkLo7/zPvowMOUAkIRotSyDkXMsqsW1SQi6BQu7wYkeZIZek0AEA6qqW1T4nps89LSAnv3L747BsR7D0yaRKlJpsNRNyghog5lzOORcz5IZkWS0ZmhD4nW2pQ/CZdKaj1SaFLz0KD3KtkIsodVjs2PAuN+WClqnNTlnqoDh/y985MQIkn9Of9C2r4YZqeu3OgrQOWVZzRDjTbUrP3/0o5ADgL156Jm5+56V42TnH934xwtl+q+HYTBMlLbh+921ZJcd00lbbrCAo4W0uKR+P/f0qg57PRqUO8Rjq0cEYm2SMHdf7lSOMsHxETWzyrJATW1bDGXLi3dHBM+RueeQwPnbjg9g/H02yhSyrc8Mg5KQMuT6soaGGVSCUIVdMrJAjGXIFRYacYrv+40078Y83PYoPf+cBfO8X+0PPe5jvzOOqb1yFF1z3Ajxw5AE4DsfDBwgBxMU2VQ8HFwjJpkt2VUAk5ATLarAMgZBTKOQAYGMtUHuxLiF3+9zX0HHcCfPZ68/GUzc+VVi3ULFDSCG3CBVMx8R3Hv0Orvn2NQJhByTLjwOAJUIslu/7JnDz3wJdwqpGvs+NV38eeM2XgB3Pw6fu+QxuO3gTCvV7UKjfEWp7FSyrPiEXvhwyxjBZLYCb02jvvwKbSqehvf8KAGGVD1XILZqLPiGlFYNjZVN1K7yYVe952bJ6ZNEU1KD1Yl2wrHqB+FTRZ1vJCbn1lfX4wDM/gCtPvhJ/8tQ/Cf2dh+Uq5CgJ0exYMQo53lMhp86QM0Ltt7R5Nw5pKOSetWPGtyrtX2jjBw8ejGxYdZd/7GXI6VKAdRQWO8GEs1zQhqYAY4wJKrnMLKsaA5wy7MWtAICLN1+MZid5vlpB14TjQ86UW0lEKYiaK1ToAESUOigm5fQhm3Nla33eUJQcEMu2rCrGNaJCbvjbQRiPayxX214sdQi2Sz+lDh76IYbkcbxqrH8sY9NYcGOYjkmToBRxEz4PEHJVHUfKNs/XumaBYWWeH4tIfbTCGHsaY+zfGGNzAA4BeEx6fpIx9knG2CcYY2qP0ggjDICoO4m6oCBy79bnBfQi3EshZyiUXEkUcnOLJl7/6Vvwoe88gPd+/d7I14UUchlYVvtSyLUUCjlmCdlPhhSoGgVlhhy9kCrUdTsPBiTOL/YthJ738K2d38L9h+/HE80ncN0D12HvfCv0ufcqyM7YhlXEEHIRyhuqkKPkSkghpzdx86Fv+I/91tm/FRpYVouGWOxgtWCSbeiwYNsUWk9Ga++LYS8Fd4p3zu/Elx/4srDMO3aR/DgVIddeAG7+JNrf/0DwOWhG3PQOjE9t939dqAS5dzftuSn4jMXDISK+YQbb11NaRWWIecRb59AluKD8HtiLOwAkV8hppYCQmylt9X8eVyjkmN7NkCOW1Ype8yf3jAVZTcKxbhVCd5mjCDkAuOLEK/C+i96HkyZPUn5mIF3L6oKkkGvb7UCNwIKJfVErQmPhYYkcLH+o2QHAQrbVmpHUsqrIkDP6GywbuoZfeUpwjH/5Z7uFDCJZnRVFXq3mDLlCwvOtkB9XGU5+nAda7JBpqQOAxceuwZONd+BDz/6QQFiNJyCsqGooX5ZVdanDSjWsAuoMOUMROSDnq6pa6/MGeh3qyC2rAxzPKkcFXWYWgfh0P1QLeq4UTXSbLnUGK3UYBALxasrETT6PzTTx+jNejxee+EJcc/Y1OH/2/L7+lp7X86Y6k9vUV7MCPg2MLKvJkeqRzBj7bQA/BPByAONwpQDCEcg5PwrgOADXAHhFmu8/wtpG1ImPMSbcKc2TSK5Xy6pwR7f7mfrNOnvkUNMnhO7fq8hh66Jji+9/dNEUBihpYDFlyypky6qW7G6MOHEOW4FV67VItsWBBXUOHADsa+7zfz7cOizYVf3XKHLkxIbVsELOsQNCrEMIOTq4pcc2VchRskhWyBWnfoxON5Pu5MmTccmWS0Lr5irkgkmXq5AjpAKC9yo4MzCPPBOLO9+E157yO/7jn7jzE2h03G3RsRzcuyc4FoVCh7nHgW+/G/jwmcB//Alai4f8p8ocwClXAK/9N+CNt2Bs+hT/OU/1drh1GA8ceSD4jPpCiCygZKVvWY0YBFMCYeehYL9M1kSVlVzq4NnOtNIB//Gpwmb/Z18hV5ZKHZY6QqkD44G6sV4u+FYaSsjZDg/l/nmlDoPc6QfUhNxMdSbx31MSotmWCDmrHXwHtXh1HBBW6Rxpun8jFzssp2V1kIH9K54S7M/v3LsPe4kCWc5JMjQ1eUXt1KttwC7eDIg+3x5dDPbxsPLjPFRJlllWCjn/2OEFjPNTUdSLAmGVpIGUHi95sqyWIiyr/X6+NKEiLFWnb/H4VLfW5w2hbDGqnBrg/KAsdaCRHRnYuqn9s5yjYxsQx9PLsaz2C3rDpmM7UuROPo/NNLFpbBM++KwP4g+e8gd9E7TFHBNyhYgbbwWd5YqIzgrifDVHk+8cIrUjmTF2PoC/6f76ZwC2A9gX8fJPwyXqXpjW+48wQlyVe15tq5T46dWy6ink+s2QWyQD57gSCNVzT6SskmuFFHIpWFZ1OgFLlmmkahnrlSFHVW5UzSbjcOuw//NCZwEP7Q8TcvsXVAo5SsgRhVzX0ugIltFgvwgK0B4tq4CkkCvuQ3Hdj/3frzn7GqVCqSplyLXslrANbaKQ0xEQU1dsucp/v8Otw/jMPZ8BAPxi74J/7G5ZV8G6WhHW7p8B1/0W8JEnAT/+KNDd30t0MH/B7wL/7V+Bky8FNE2wKHqqsp/u/amw7sxoxCrkEJMhB4gEArUeywo5ui4Ns6G0rI7rQYKD93ytUIPB3PdgmolDi/OCQs62ApKqXgkmvnTgZ9pcaMZ1/7Bb6jDg4L5SEAm5ydKkQKr1gljqEK2QY0Qhp8qPA8KW1cNeQYCUI5e4ZTUlQu7UjeN40mZ3u3dsB1/6aVDeIdvqZALVw2q2tBQSDrbnFsOlJMMCVWtllSFH9623HZrt4HqRREFGbap5IuTocUxjNVaWkAt/V1U3VMSWVS6MCfJqC5RvDJrLtKzKWWWAOP7J2rKap/w4IKxU8zDojaxB3jdU6pDTYzMvKAlFbvnaVpomCkA8rAXVowq95lUjBEjzCHkrXJLtfZzz93POd8a89vvd/5+S4vunCsZYnTH2Pxlj9zHGFhljhxhjNzLGfnUZy9zGGOMJ/vWn3x0BQHw4dlTw/UqDTsaUllUnfJGWM0Z6gWbZxBF4qufSLnaQrZu2wxMTpOpSBytaIed0cOveW9G2w2o0OrEoKQg5lbpOVMhFbxeBkDMX8BAhcTzsU2T5HehhWbUjFHK6NOnwsGQFpB1VDlFCzqg9Aqa7r9syvgWXb7tc+ZkqRUPMkLPEDDkLxLJKCDkNRbzp3Df5v//TPf+Efc19uKNb6MDg4NWTd+P9//AMXPidX8cHHr8BcIiKcvpktLY/2/+1PB6sOyA1m3ZJtlv23iK8hhmN2Aw5brskUNRkhxJylKBeF2NZbZpNpWW1jGD9PVKIMYaJYqCSO9h6wi8EqRgVNNvBulMyQ7RnO5goiYSc97kGVV3JBQn92FUBqdShbQmtah27M7BCrm0RhRyP3gdxKBl6SD016CToV88LVHL/fucT/s+yrU5uX/PgCNet1TVoj7LLyxAsq8NWyBEyK9OW1S688+ICaUZOQshRoiJPpAU9jsUMuf4IxzRRUmwf1XlOblldDbZAoWXVckTybJmW1bZSIZdtqUOeyGZAKnUwM7Ss6uJ+oS3Va6HUYTnIs2UVEG+WelirJKvq2jiCGmkeyRd3//94rxdyzg8DWACwuddrVwKMsc0A7gDwLgCnAbAB1AE8F8CXGGPXpvA2+2L+mTF/N0IERPIq/wo5RyKjVOSaaFlVKOQSnOCoTTTu9SpCLm2F3KLCApv0JK20rLLoltWH2Efxm9/6TbzxxjeG/qxnhpxC7UHXvS+FnMKyundOUeoQYVmdKrmtnrZNLaMBoadFKOQo6UTJlagA3avPujqy7apaUCjkyDY0efBeBgtIEctx8KLtL8Jp607z/+7aO6/FvY/uxWv0G3B96Y9xv/URfIEtoK1p+EJ9DHMaA054JvDqfwHeeCta63f4y5MVVLIqDVATcolaViMG4JOEeKNcflypQ9NsupNUbRFaV+1Y1IrQ7IB4o5PYye4+BoAjZkDqjBfHhUKHOikjoBNK0+ECIeeSce5xnZZldTmEXLNtCWSb27LqKeSCzxdFyFGFXNt0cDjCspqUkAPCTauDZim99MnH+X9LLX2yrc5IkCG32hRyhhRgHYWjiyuTIZe5ZRXuOY9z3lepAyBaVmW780oiqmV1JUsdVISlikChD7ktq6tLISe3rA5yjiopxotZZ8jN1gMl8+apfEWH0+8uvWGsKglJE6WCuJ8F9eYquw5kDUrI55HoUp2LVtu1PS0UBMVvPubeeUWaZ+IZAPPdjLgksODNGHIE5pq8vwzgRAA7AVzEOR+Hm4n3JwAcAL/LGLtmOe/DOd8Y8+/OZX6MNYm4tjrhTn5OFHLyBEadIRdWyJX6DMmk1hIqyZehIutSV8gtg5BTWVYZM4UBZfCzjaZ+HwDg5j03w7SDCaHtBHfKGSPttVRardiuS4TYjMuQO7QUZJ41Og0lIae0rJJlUoXcWGESjsOlllNCyPXZsrquvA5FTZwUb6hswEtOeknkZ6oUdXAuZ8gRu7UTrG+RBcSU5XBoTMNbznuL/9hXH/wKrnrolXhn8TP40CYb36sF68YZw8FXfQ74jW8Cp74A0DS0rOCzyoTNeGHc/7nRaeDA4gE8MveI8BqmN2E54nGXtGUViLbYTdXiSx3KBQ0GyY87ob4Ni0TtRiex6ytBscOis9f/uV6sY25JbfeLVcg5waRn0Dv9MiE3W53t6++pLbTRtoRjzm1Z9RRyhJBLZFm1fUJuUMsqEG5aHfRO+2S1iMvOCG+bWMuqkCG3ekOfhbyc2FKH7DLk6HGXlULOkOzjbcvx92vR0BIRgxfvWA/AzV6Ma53OGvQ4phlbCznLkFNNeDVp3GcpSrLyhlD75jKzxWTFHf1ffr9hYfNUFX/xkjPwgrM24m3PP3Xo79cPxJZVqpAb7vsWdZmQy/+xmRfQ83oWhHK/UJGEa3WfitEqI4VcHNI8QuYBjDPGet7+ZIxNA5gEcDDF908LLwNwAVzi7UrO+Y8BgHPe4pz/bwAf7b7ufUk+6wjZQchbky2rlLTIiUJOnsCoyDJTCAUdNEOOWFb7VMil3bQqZ8gBye+aqEsdJMuqd9HTRJEpJWCEwaiu+UGrvbIOqEJubslUEqiAqJCb7ywoCxxUltWoDLkiq8NyuKAGonZUkWwm6xuRIccYC6nkfuOs3xAshTKqRR1wpJZV8mYdHqxvgRJy3dc83ZjERYarAnPA8TfrSrhm4wbcXAkTMIfrIrnRsqMJuVpRJMFkdRwAMOYIGXeAWiEXpZiQs+KCx2NKHTpNMMZQqQXk7NbxbWIzISEOZmvr/Z/bLCDkxovjmG9FEHJSe3S9WA9+t4PtlBYh169CrkbC9eVSh5bd8s8FyRRywXfz6JLpKxkYXw4hJ5IIy7G+UNuqB9lWR/cXvRmzmltWRRtuwgy5IRNyl5/hnttKhoZnnZK8hGQ5oNmlpu1godW/euwPnrsD/3zNBbjxrc8WlLArDVmd6kFUyGV7b12VIafKypRbVgUXRU6/a7KVcbmWVarE8pwBcXnLw8IbLjoRf/va83DaxnrvF2cISu7QU9jwSx1E4jXrXL/VjDyXOgDqc8taVT0WpGvjCNFI80i+E26G3DMTvPb13dfenOL7p4XXdP+/gXN+h+L5vwbAAWyEa2EdIScQFHLSBS1p1k2WCBFySstq+CItEnK9W1CbRNllOzyy7EBtWU1XIZe2ZZUxUxjYeBdCOskHgKYVEDCUSBOyKHpYgWV13yGFbbVtt4XCAFfJ5m7/aaKoUrasNoPlacSyavC6e8wSQo6qxjRp0uEhSiEHANNlQnrZNbzilPjC62pRB+dUodcWtlHbCda3pHmkCMfY4z8APvsrwLUX4s077wXrKvhurpRxVzkgU2hDKFUYAiL5KJNEVCG30FnArXtvVa6/zcR2YbVltXeGnPi4VOqgyLMrlAOF3KbqCSIhRxVy1UAhx4rB38gKuToh5BhjwsBvrBBMdLgd7O9Bx4FpWlYXWhZKRrC/XYVc9/uUIEOOkgJ7jgbHQ1GTCNo+CDmZKFnOJOjiHesxMy6Sg3EZcvQatKoVcpJKMwpZWlZf9KRN+O5bn42b3nEpZuvq4yltGJJScBA7p6YxPOOk9diQ0TonRZRCrrkKFHJyy6q1AkRUv4izrA6yzrISi/4vv99aRNQ2HXapA2NM2Dc0Wiav+YZ5QSkipiYvUKnh1qxCThevjSNEI80j5J/gkmzvZ4xVo17EGLsUwP+AS2p9JsX3TwvP6f7/LdWTnPPHAdzT/XVEyOUIorRfypBbtZbV8GcSBliJMuTE5Ub9jdqymq5CTi51AJKp/ICIllVNzJDzf9ZEsqzRISQZzY8jA/u48FHOORaldaeKNg9HWkdCjzHdJc/OO2HKnzTMLZkhtSC1rOqEkGP2WDeXSMxw86BF5CNGKeQAYFN1q/9zefE5IfJFRqVoCAq5tt0WbL0tO2gFLbEqXqDdjH8v/hnO/u7rgYduBACcapp4aUMuuGB494XvxvO3Pd9/5FBLJOQEy2pMhlzTbOLmPcE9HvqZuLYg/B0lTXsp5CYjLKvy41St5xF+jBQ6bChvEQg5qtDyijsAQCOE3HhxPLahkg5Gx4s0Q8797BqDrwDtF8sl5CoF3ScD25bjN8kC7vHjfQ9py2pUiysdgO8j35OyLq6jfJzHQVbILWdiaugafuXc44XHZBVPISJDLi5qIe+gJRTxpQ7ZWVYBYPvMGNbVsjMwGNK1YyUbSNNGVKlD1M2FrNfJgyrzS7hZxblQppLHiTwgKadCGXIDWFYVjor2Mkm+YwmRhFwG52K6b2hJSl6PzbxAbFnN3/Gruo6v1X1Kxz1J5qtrGWkeyZ8F8F8AngbgZsbYmwAUAYAx9gLG2O8wxr4Gl+iqAPg65/ybKb7/ssEY2wDAkyrcE/PSe7v/n7GM97qJMTbPGFtijD3CGPscYyyJulBezs9U/+CWUawpxIVjiyqizFYpFjLp07HcMGjhMUWGXL+WVTpwjvsbpULu6FJonZYDtWW192fgnEcq5ChB6SvkJMsqVYtR201UFoVso6WZQB5UOXIymQQA0FxCacfsGDYQFQ21rS51bD8EvKhzcM1dX84ZbKviErOSZdQDHY84ES2rlYJIXFx+/FWwGifDPPoUTNuXhddZQrUolTpYrWC/MQsmd7eFAQ3/e88b8bfFj+BJGs1yY8DpL8HvX34tNPeyAM41vOy4P8ZVp14lEFKyQi4uQ46q0n555JfY3djtfl6jggs3Xug/52gL/nHskqvB8QA7PkNOVsIBbiGA/PqaIRJynHM4xj7/saniFjQEK1uwPadJhhzTg+NqvDiOW3YGFmg5EJsOds5cd07wkRa3uZ9pGXfal5shxxgTyAi5pVelkIsihqkihn4PZeXnSmTIeXiFZFuVVTyaxuBdhjgPPsexopCLix4QFXL5sWOmhaIUd9BYQTtn2ihFlDo0BrDlpgWVZVWZISflq5oxN23zArntfbn2UlUJmGmpx0BrEVE34oZd6gCI254q5PJIMuUJqzFDTtW8uhYwsqwmR2pXUc45Z4y9HMDnAbwIwP8lT3+j+7+3Z74G4LVpvXeK2ER+fiLyVcFzm2Je0wsXAvAkP9u6/17DGPsIgDfzNFmQNQJVAYKHPCrkZPmuw11SUbQBpZEhl4yQU1lmmx0b8y0rMti+X6gtq733x5K1BNNRlA8zS50nIVlWqUKuYwfrEJVFIV84VGUUKoWcTCYBrkKOm8BJM2OYnSj7NuB9822cMF0LLWuq3oFHF3G7ioWWE5shJ6sAPFDSiZJFAFDTZ7G067cAAPVtvVvPqkUd3JFLHdz3qmmBKrBum9hkPe7/bmsl6Oe9Dnj6G4F127ERQPkbj+GodjOsufPw6stfCgCYLgeEFM3gA+Iz5Cght38pUKOdM3MONo4FOXnMaMDhgM7c5dm8uz+5Ae8yGN2yGj725UIHANA1HRWjgiVrCRwugWwx93jgnKGub0Kj/VCw7iW1Qk5YJq/gjl1Hu8tneNYOMROLnue2jp2Iz77gs9g1vwdv+v/c4385Y8CSXoLGNDjc/S70q5AD3IB6L0/LkY4fb3LPkpQ6KBQxAFArVHHAob+vTIYcAJwyO44nb57Anbvdy3oUaeB9byzHga7pompnlQ3aRUtg9LVIKCbJQCGXNSg5bzl8RcmqtCFYVskNNRqFkQfLqjJDTrKJqyJA8oY4y+ogtjdDY9CYO770tkFnpJDzEaWMzlwhR8aYq00pnTVoLmIelWeq63ge1zMLFGKEDiOISPVMzDmf55y/BMALAXwRwC4AJgAbwB4A1wF4Cef8Ss657F3KA+hoPs6r5810x2Jeo0ILwLUAngVgnHM+CaAK4DwAX+++5g8BvCPpAjnn56n+Abi/z3Vb9RAVcuKhHZWztZJQ3S2QSTFLEUJML+LtBHccmhKZpCLegGiiLs1ih0FbVpWFDkDIsuqd/JlkWaUZci1BIaeWvsvbQrarAmqFnEwmuevibr+TZsYwOx4QDlQhd4AQcpNjwbpzawxzSyZshwuW1chSh4QZcmL+T28FR6UgKeTsFortQ3ir8UV8vvJn/uP17r6c51V83HoZbrjiRuBFHwLWbXf/zrSxb/9mtPe+AlrnRJwy62bArasQhVwrJkNOsihSyyrF0zY9TSD5mN7wv0tiw2qwP6ImJipCTqWaA0RC6J5D9wCsq8oz16FjakLYOyWE6LpS7DmiweNYn7ptKkQEykTAORvOwSWbnwevwHw5OTiMMTx19qkAgDOnzxRy/pKCTtRtWyTkfIs+Ic9pzhyFitwCgPGSRDQvSyG3/AHzGy7a5v981nEToecNhcWTfmdXGR8nHH9mnGWVKuQivjurGUJhh+WsKFmVNoRSB3JdbBCLXdakYympQk4Tb1YJY8ScElF0PGPay7esMsZCKjlhmWtcIRelsMqakKM3zkeEXDzoPssjoay2rOZvPbNAr7K8EQIM5SrKOf9PAP85jGWrwBj7cwB/PuCff5Bz/q401ycKnPO9AN4oPcYB3AbgpYyxLwJ4JYB3Msau5ZxHsBAjqBCXD5LLUgfFerSEJ8wdAAAgAElEQVRNWxjcCnYFQ5Ehl0Qh1xEVcpGEXMTJcs/RVmrNWMoMuWUQckyzhIufv99lyypRi7Uj7BpyU56w3tI2BICDilIHJSHXzZDbPlPDbF20rH7s9o/h9v234xmTv+4/XqsGBBS3a5hfMt31SVDq4B0unHOBkJOtgGLbZ2/FSq1kiJbDX96ID+76AIpGB7eTdtYJpuNrM7+Dd+06Hw1U8X8KU8JyqEKyVjL8i3WcQq5tk8wwSSE3XhyHCk/b+DQ8cOQB/3fNWPC/9yIhF+yPqDuYlYKOoq4Jx6ncsOphrDCGg0tuefhdB+4K3qY9g2bbjsxdopZViof2BdvrsjM2hp6nzV3eMUuFSsttivvIcz+Cn+z5Cc6fPX+gLDraJGvbZGJPFHJaIfhuU8UjhdxY6mGiWBNunfVDyNVTVsgBwJXnbkataIADuPyMsMVXIG7sMCG32hRy9PiLLXWgGXLHomXVEInJqKzI1Qh6jRQUchGN0VlApZBTESi6lK8qFCTklPSQx3emtfwiiqKu+efbjuWMSh0IorZp1pZVQSG3RsmbpIjKfs4LVOeivJ5vho24bO4RRKzukUIADZ4koH/Qv6OqvTgflyc3acS8ZhD8KVxCrgbgUriKwhESIi4fJI+W1X4Vcl4GgRz62wv0Tnbc30Q9/sSQFXJJmncoIaczAzZ3JwOaZgpEga+QY3GlDhEtqzHSapXVVqmQW1Ip5FqYrZcwXi4IDXoPHn4I1z/8dwCA+w8+ArA/AHgR5fIS0F19bgcKOdde6aLjdGA7NnRNF0gXz+neslu+1bCkl2Bo4qm+34yjSlHHLF/wPfbtA/eiyN2VnNOCv6+fcDG+f+TVaOxybavyvqUTOmpDHDRDrmJUwMDAEbxPrVDDGdNnCMthRsMnwKMIuaiMD8YYJqsF7Cf7eyqBQu6ug4SQ62xAs2NFhr1Hqc8e3BO8XknwKBqs6PltuXf5a4UaLt166cB/L9xcsERCzvse6tXAxvukmScplxOlkJusjPvBDzrTI0sheq0bkN6d9svPDBOnHnRdJAgch4Pel1ltY3Z6fKluMAHud94jAwo6c/MojzEYEjFJLau14uoeZpcLEaUOK/gZVRZ21blOE27ESiVZOSU90m5ZdZepw2t7Dy9zlZ10UkYh0rI6/PcWSx2IQm6N75NeiBq35wWqdVqr+3RkWU2O1I5kxti/MsZeyhjL/PYn5/wvOOdswH9vJ4uiuXHHxbyl99yelD/HIwC8ir3taS57LSCurU4I982LQk5xcpIJOYFkHLDUIZQhF9WySpZF1SN7jrZULx8IKoVckrsmtGF1phLkaGma+NkKEQo5almNujscJ61WEnKqDDlVqYPewkkzrvJnIyHkdjeC082CdRDF6e+761EMlG3cqmFuyexOdpmoUusqx2QVABDfsAr0Gcj92E9wyg1X40+N4P7AUvc973K24WM8aEidKE9FtkkCEiFHSBaqEAtlyMUQchrTQqqq82bPg6EZYlGC0YBthwk5Tgk5I3rAJNtWo5oi6bpQQs5uz+DgQtvfP0VDE449QzNQ1sJqP9N01++0jePYsi68H+kgzyPv89TaSY+tjhlMojt2B23TAdMXoJfd4gtDM/CUDU9RLqeoa1AJFqYqwfauFqp9qfiGYVntBYG4cRyBPDU0NnAj7kqhoCCEZQj5cZXiqvuMSWBIk47msdSyGlXqsIItqwWdhchrlT2fPsY5F2zVeZ0gl6RIkjTWWVim5YguAf3YI8j7wcqWOgTbnirk1moBQFI8ffu0T8BfdNL6FV6bMJQKuRwSh1nAkNxpeZl/5xFpHiGvBPAVAHsZY3/HGHtOisvOBJzzAwAOdn89M+alXrvqvTGvGSFjrLZSBxURJRNsqhBiVWtWHGQyqa0gxeT33rY+UPqkqpBLwbK6vhKEyzOJkPOUkUwqdWh2AhJGtKyqM+QSlTr0kSG3fcbdnrOEkDu0eER4XXH6+2DGEWhGoObzFHIe2UKLHTxLKh27+YRcTH4cINmNSgpyiXPgkR8A//Bi4NPPx9hjN6JELqTtsRn8+cT/xEs6f4mfs+DeRb1YF1RAYUIu2K50ojdWGENBc9djyVryCUXTMWF11ZAGM/zXUMg5ck/b+DQAcnOpWiHn2MSyGjMInqyIirgkCjl6zDqdDdg3T9pTFRPYMcneCwDcdoXalynUcYA4cPfIe1rskcWkIg61CIVcy26hZdnQa4E67skzT1Yeq4CrUlS1Aa6vBSRmP3ZVIGwlzKKtjR5jls1XdcMqkCwKQsyPO/bsqkC4ZXXhGLKslqll1bJ9FbaYk5ctqcMYE64fjKnt+cK1kUulDjklPWTLaoco+gc9R8mEnKCQi7kRtRawohly5L1HCrnk2LKuiu+99RJc97vPwCvP39z7DzKGiuRd6ZujKwXGmGhbjSl/WutI84r0GbjmkSkAVwO4gTG2mzH214yx81J8n2Hje93/L1M9yRg7HgFZd2Oab8wYOxGAJ/95JM1lrwVYsQq5/GXIqeS71E4pv8abzBXJHc0kCrlmJ6FCjjy+bTqY3KapkGupSh0SfAZKbqwrkaZJiXgLFHLRpQ4CIUdUWnHZfEkVclEZcp5CjmbIHSWqP8AlF0uz18PRFvzHvFIHX31CFHKeckwTVADd9TWj8+MACBNGYTLFOfDQd4HPvAD4xxcDO3/gP1Ukm6R1/Ln4qXYOAAamB+81UZqIzZVqUbuwMKFiSpUcVcdFBf5HEnJlUSFndtt1G2ZAeFJCLk4hJRMJqpZVIJoUctobsJeUeKgylyaKCkLOcffd5Yr8OEBSyPntnfkheahypmXSGwkdmDaHXvul/9gFmy6IXZYqN2pDLci2jMqfi0KoZTWDLKVQ62OO9tUgSJIPc3Tx2M6PA2Sl6rGlkDN0zR9Pce6OEzjnK5ohB4jng6jJrjzusxSOg7xB05jweejYY3DLqkzy0aKIfBKTWSEqQy+Lm1mRpQ5rfJ8kwdbpKs47YSqXimt1y+ra3acj22oypHaEcM6vBjAL4OUAvgQ3avk4AG8GcAtj7H7G2J8zxnak9Z5Dwj93/7+cMfZkxfNvAcDg2lW/p3g+Eqz3meP93f+XAHy3n2WPEJ8PQic7eSHoLcWKhC2rYdVf/5bV/jPktk0HSpU0W1ZVbaVJTtDUsjoVQ8gx5g5mmWxZpQo5U50hZwiTSzlDLlzqsNCyBAsmEJ0h5xFyNENuvjMXem2hfhceadzu/87tGuZblj9x5zxc7KBSf9JmUpXqaEG2rHIOPPgd4FOXAZ+9EnjsJvIBdJhnvxrvNq/xH2pbbf/YZHrwXhOlCeiKJkl/nYUMOfE7KuTIda2/gl1VF+2q/voTIqZerOPUdad2l1/2W1QZc3wClJKVnmVV72EXDBFyEUofFSHnWOOAUxFadVUWL/r5/fWzy9g0UcZZx6tLVYSSAMcrdciPQo5+znaHKm1aADiMakDIXbjpwthlqXKjzp09A1vHtwIALjtBeQ8tEmHLagYKOYm4Wf0KuWh7uoejS8e+Qk625cwvrZydcxiQbauuysrd3wWdCUrzzNaJXD+iznPCuI9zQZ2R5+w0ei4SCLkBbxrQMeOSafm5lYytzvNOmogsdchgu0SVOqzVAoBjBSqyP8/nm2FDIOQSzFnXKlIdgXLOTc75v3POXwVgA4DXAvgPuGmipwB4D4D7GWO3Msb+iDG2Kc33TwlfA3Az3G3zFcbYhQDAGCsxxt4K4I+6r3sP5zxUs8gY28kY44yxf1As+78YY+9gjJ3FGNO7r2eMsXMZY18B8Kru6z7IOQ/P7EeIRZxCLo+WVWWGnCmerOhn+v/Ze+94WbK6XvS7KnXc4eyT05zJTAInkYcwg6iER2ZIivBEEVS4FxMq6ntP5Slc371eeaDyUT8XSSrRAN7nSBaQS2ZgYAYGZjhhwgn77L17d6qq9f6orqrfWrWqOu7uqu71/XzO5+zd3bu7qqtqVa1vfUM4qA1jWe24fuI1aYQcJQMvoAq5863IpjIuVNbPQTLkqEJu1SGZERIhB/ROhrJllWbIkc9LL3WQLKspNt/TRCXHOVcq5GC2cMm+gDRaLluo9CY3HiPlAoRoO9d5MH7YDUsdestDFXJeUiGnypCrWUmSKCaHOK7a/Azw9puBdz0POP6/4hcZNnD9TwO/9CWYz34bTvjxcN1yW0pCbtlZllQz6YRcRQp3F5pWm0mFnJwfF4IScjfuvxEGIxM1P7Y0hiUPVCEXljr0sxOsVgezrMpqPSBoWAXQl5Dbq2pa9cv40Sv3p5KFylKHHOUk1VIIuY7XAbPPwHCC47pqVXHNnmsy30tV7LBnqYIPPPMD+NAzP4RX/cirhlq2ZMvqlDPkPD9XeX+jQCYYVTi/LWbIzSNkWw7NzZsPQo5YHrteLhSAgyjk5HxVN6P4K09IC/sflaih1zm05CvI5izeuDNJpBFyqkzCSSO91CG/+6ZGf6jGozyPNzsNYW6VF0VMDrFjZ1LO+TYCtdm7GWO7EGTMvQjA4wDcAOB6AG8CkKsrNM45Z4w9D8CnAFwE4HOMsS0AZcTf159xzt8+wtsfQ6CCeyOALmNsA0FjK/WU/SmA/2vU5V9kZGXI5dOyqlLIicSPq2jYohYDWVEnQ6XsSvsbSlTtrjlYKlnYbLtouz7ONjrYXR+8wVAF3+fKzx42Q27Fpgq55PrZpoGObFntUoVcWoacOLn0fR7dJVVZVoGgafXIrkCBttHZiPLOKCyrhYPLoVKLYf9yCT84sy0QWWz9SfBXbkso+3y3jvPd2LLKeUzIhSo4WQUA9M+QO3WugR83voDXWB/E1Z++R3zSdIDrXwo89j8Bq0cBBHcnHCPe/k23HRG7smWVHnuedPIVMuQkVYVKIdf04u9IZb0FgF3l2Or5iIOPEJ4z+BL8Xk9O+J6qUod+6qiVymClDkqFXDvIPDxHiAlVrtT+uhhOzL0SACM1Pw5IEjyA1LI6a4UcWc9Gh8NkJjzuwYcPq35n9NyNB25U5gNSyJbVesnqHb8mLlm9ZKxlA6akkJMUZVQlbRbwgl0ocEmzrDaJZXVOFXJAsG1DW/w5YtOdhZ1z0qDnyZZ003BWLbI08iBNzWQI58bsa8Q8QSBqyDXcJBRytNBp0e2qQFaG3BQ+2xLPByEWWU01D1Cdy4t4w21SyLpJrxFjKqMx5/wc5/wvOOc3A3gogC8jsH3mst6Hc34cwLUIiLNvIyDiNhFYVG/lnA93Kz7GrwJ4O4CvATgLYBmAD+A7AP4KwKM456/hk5IjLRjEPJ4My2pOvl51hlxGy6oxvGW1oSCS0i2r5M6pZeDgaqxIOnV+/By5NJXZQC2rrdjeWbdiJRFn3YR6zzaNpEKum1bqEH+XWeGjclNtiNNb8cRL2bAKwHE6wsQgtK1SIquxtQ+dM09M/C336ui4fnxRTkodQkJOaBAOA7fJ+gotq74H/o33462br8GfO/8NVxuEjLPKwCN/Hnjt14Cn/XFExoWoEIVay22h09s3ZYWcqSgaiP4upWUVUDetDmJZfe5lz8VqaRWX7boMz7jkGcJzph9bPc9mEHL9JmeyIi5VIafIMfM7e5OvU6hKDiYIuQqWShYedbFCOdeD0HLZG/8Ey+rMM+RIi1zbhWPG35tV/3b08yMPZOfHAUiUOuyqjUfuVGxTOC9Mg5CTM+REy+qOf/zEQdcn1bJKSx3mNEMOECcddJ1nRVhNEnSsbrme0LA6q9IKukyDZMj5vtiymufWQzHsfwIZcuTvttrxvpmWn7ZImGWGXJrVe5HVVPMAZanDApOs2rI6GKZyJmWMOQCeDuDFAJ4KYDypzRTAOd8A8Fu9f8P83YUZz/09gnw9jR0AVRvI0n7ZupAHqDLkEi2rflIhVxqCkFMRSe20Ugca9GsZOLhSwZ33B/a+k+tNXHN4JfOz+iGVkBtggD7XjhtJy8YSuG9FDattry3YGYMMuQyFnJvMkOOc9wi5WOXQ9ThC3iRLIReC5sfVzFU0vEDVZ5oimblfQchxr4rOmcejtOuLgNVbV25GlsqQ+ON9Sh3CTZtQyHku8M0PAJ96M9jpO3E5OTy4XQW78X8HHvMaYCldjVU2SwjNni2PWFYNQsiVlmEb8Xcl7+NpLauApJBrKjLkUiyr1++/Hh+/9eOwjOTpzOLLCLdQSPLRfQFe8J79LoCTGXKDlzqECjkKlWpmX00i5PwKnnjFvsxJk2gZ9Hv/50ghRxp8t1ouyuVyrOys3h0916/QARAVMQCwlrINBgVjDEtlKyJPpqFKkEsQimKhS4OgKk5VyM1/hhwgTjoEhdxcWFbjY6/d9YVz+cwsq4TMSFOXUv5KblnNs2KllGZZHXGMoucQmh+rCbn073Qa2Xqq5vBpfbbGzkG1/RbZhmwrrlM1ktixPYQxZjDGfowx9tcAHkBARD0HgfXzHgB/CEBVmqChMRKEiahsWSV7up8TQm7ollVVqUMfddmWipBLIcboezmmgUOTVsilkFrDljrYrA5wkk3liW2ntmkAcqkDIWFk4vG2e27DE//uifi1T/2aWOxAXkfLKKgigGbI0fy4qhHnrXEmfncHek2rMiEHbmOt9ZzoMcPfhUBIDJxt9CZ4tNTBS5Y6+IoMueq5e4H/9xHAB34WOB3bBBu8hPeWngv22q8DP/4HmWQcAFTteH9oe3GpA6hl1VkRjj1ZNSMq5KQMOZVCzutPyAFQknEAYCFWyK23k4RcbFntkyFHlD1l20jk34VQEnKdJCGnmsTSDD0gKHTIsqsCIonTVWTIzVohRxt8tySFXEioG/4SLtvVv+tJ3l/Smm6HwXVHVwEA+5ZK2LeUvn9NCrJCjqq1izgJk9dHBSFDbkwSNc+g5w6qwp4Hy6pQ6uDmJUMuHvvS5rpyVEk3o/grT6DXeHRfGlUhR5VYWwLBl9/vYFpIswFPYzxOI0S1ZbXYUB1Xi1zUQb+PjpuP+XceMfEzKWPs0QiUcM8HsBfhjBJ4EAEp927O+Wcn/bkaGlRtYGdYVvNT6jBcy2q4To45hEJOZVkdQiEX4uQEmlbTFHL9SMWu342C+BkYTFTBuQ2GgKxpuS2slGL1nm0ysIEtqyb++pt/jbOts/jo9z8Ku3wN0AyIEfrdUzLxgrUqvnlyA4CkkCOEHO/sjka+Dt+Gz/2obCBSyBFlGbzguz5afhSecmULt91zG3D2qQhpyJCQ66uQCzPkyPpWvvNRYH0j+r1r1/G25o/ir9yn4NoLL8YL60lLpQpVJ94fOhEh54uW1dIybCMmT+XikhZVJ0qWVaqQC79L2habliGXBZsQcucyCLl+dgJa6pCmjgOSltWqVcOmm2xIXRqAkGO8gic+JHvbiAqlnmWVJ23uswJVB221XSWpWvevEIo40pBo5Z0AIfem5/0IPnr7KTzmkj1TUYskM+Tys61GgViEk9aySjLk5tqyqt5/KCldVAiW1a5oWa3PaP3EUof+pIrPuXg9lWPSI52oGT9DjpKpWiGXniE3Hcuq+rPzTBZr9IdWyInIKszTiDGxPYQx9kbG2N0APgPg1QhaVhsA3onApnqIc/6LmozT2CmIeTwFKHVQLEeiZbWfQq5fhpxCIZeeIScTckQht76TCrnsdaDquOXSMlyPCW2jHU+0pwYKOfGxttdG1w9IurZECoX2SACw7E3yvkQhR4KVj+2OM9moQo5myDW2a+C9vDcOX1CsBRlyPkCILN4j5PbWy3j9I16P255/G446NyU/x1eVOsTr6fs+8J2PYvvL/yN6rBruZ+UV4Im/gffd9FH8P+6tWMeSQLr2Q80mhJzfI+SMDhgL3r9qVWEbdqZqpkX2gYqskCOEVLhNqPoxLUMuC5SQO98j5FQtqzKBL+OSfTXsWwpe++hL0jPdao6okLto5SLE96RiqGxsa5U14fe91VUsl7MJDKFltWcFyJNCjq6nnCEXYoVdOdB7JSzOE1Bb7V0q4aWPvhCX7ktm/+0ExMITOUMuvwRBGlSWaRlChtyCWFZDOKaRmhNVJJSlUoeGQMjNvmU1bfg2hQw5SAR4fifIaSTRJCyrWiEnIrVldRoKubTtXMBzgUYMnSEnQo7q0FBjkmfS1/f+7wD4FwQNq//AOR9/Jq+hMQCy7n7msdRBrZCTWlbJJGcUy6pSITcIIWcaOLQaEzCndlAhNwwht6u0Cx3XByeWVWprBAKSQm4rBQIb50ppRWpZNbDRjtVjhhW/F1V8bAsKuZh0ScuQW98swXbKUZbdVncLdSeY9B9YLgNGKyKyHFZF2G+zpx6TDLTZM1LIcVWpAwPAcZNxO35z+33Ae+5Cc88a0FNrVQ0HeMLrgUe/Giiv4AcfuSN6j8Org5NcNccG5yYY630XzE00rALZJ98W2ccGsqwOkCGXBYfFysnzneA9KTnK/V6GXJ+LpZJl4sO/+Fh89d51PPEhSQtqCFkhd8nqxfgCCxr+hNcpiLaqVYUBBz6CbX3R2p7Ea2TQC3eVZXXW14CyQu6gkYyP3W0+dKD3SpY6FM/+aBrisUFvuBSSkBu61KF422xQqBSO86COAyTLqqSQy4NlNY1cM4x0AjzPCrk0kmjUVlT6dzpDToRpMJgGS9w8nIpCztYKuXmE6lze76bvPGMQJb3GZAm5TyAg4d7HOV+f4PtqaAwEUW0gWVYVwfezhoqIonZKzsXMk8iyOoRCTpkhl0bIeSJRRRVyJ3dQISfbGmWst+PhZKW0EiwnUYolM+RYomUVCKyK0d/3YJmiYsogqrU0y2qaQo5aVn233iN7ArJvo7OBA7UDAID9yyWByLJZTOLsrseEhYqQoy2rIVlVPfUf+Fvn9/BI49tBZzOAbbK/V3/i/waueEH0+4n1eB0p6doPFccEWhZghoRcN9GwCojHXkIhRzPkpMnAamkVDAwcHOvtdbi+K1hWRyHkSkgScnR7hyrGQdQSB1cqOPjQ7O9LzpC7aOUi1EqWMAkC1KoSxhjWyms43boPAHD9kUN9l8lShOrnSXVF87O22i5KlkjI+Z01rNTSCU6KhEKugISclUEQzHpbjQLBgpsyjp9v0gy5xVLIzUN+HCCS4UnL6mzWkSoP0w4dSqrQMivTYGAzLrzJwqQtq5T4odvOyTEpOU3YZpKQm6VCbpHVVPMA1fXkIm9TbVkdDBM7k3LObxn2bxhjy702Uw2NsaEqQAgh3ynNA1R3CijBJk/WwnUYLkNucMtqW7aslmPy4b6NFjyfj3WRMmqGHCXkVkur6Ho+OE9mqYWwTSPRsgrERAxVyHHWAkf8PVOCiX5P21KGXIjUDDmvFuXCAcBWJyaB9i2VBULO4DGJsyeFkDujalldvxf4m2fjwPc+hgP0/G+WsL12EdB5AABQq4nFACdHJOSqjgnetMF6vaXMcMGMpEJOKMaQM+QySh0sw8JqaRXn2uciUk5QyI1gWS2ZMSG32V0H51zMkOu1rE5KLSETchevXIx6iScIuaWUifqB2p6IkFur9G81pnfSQwu8l6OigIptwugpBFtdH44hkmhu41KUVgZTEZXlzMEiEnKCxZgL26qIGXL9Sh06rh8RAAZTZyfOC1RjSM2Zj/UtCaUO+bOspmfIxT/TMqu8H2tphNyok/oS+SJ0hlwStmkIDfBAelHIJOGk2NkXWU01D1CNL4tsD9eW1cEw9T2EBfgJxth7AJyc9udrzC88au+UBkQhSyQ3llVVy2q8DmmB3wIh5/ngGevTaA9uWaUDpWMFTZK7eooGz+cC+TQKJpEht1JaCZbfjycBcoacZagVcqFVkdqCfbYtvIYSZYJCjlzM718uR9ug0fEi0pNmyPFIIRdgsxNn01UcE7VK/F12OvHrqGV1mRJyjd7rKRF5xweB730sXl5u4gPGjwOv+Qq2145Fj1etmEAExMbcQ0NkyFUdS1Amygo5lWXVk3Kl6EWvTMgBom31TPOMYEcepdTBMcqRCs7lXWx0NgRCLsqQm9DFUtksw2Txel28crHS0pU2iaXrv+Qs9f08alkNFXL0K581IccYE9bfZCKJ5jUuSRBtaZg3hZzrceH4mPW2GgX97nwL6riKPfNMw52EUiE3JwQkPUbbXQ9b5LoiD5bVtP2KKuQ6E2grnRZ2stSB3hzK+/cwLaiUarMtdZjfcXIRoLIc5/0mwE5CW1YHw9RGY8bY1YyxNwE4DuCfAbwAwPAzLA2NFNCSBHlA7HcnfxZQhWBTsqjjqS8gDYMJpEeWwkylkJNz6qL3kTLkAEy0aXWbkFpCllKfGmxZIdd2fZGY8oZUyFHSk4uEHAxqWaUZcjQzxxSIs9ObwWfRDDnfq0fqKwDY7MaEHACs1OOJaqMZq+KoQo4Scq2uj0vZcfw0iwm40JbKmYG/dx+Pmzt/jD80fw5YOSyQTlU7JuS6no/7N4LvizFg/0oy0ysNFccUlInM6G9ZlYtLmhkKOUBsWj3TOjN2hpxlMHA3tgSf2DoBnwfb3zYchNl9k7oAZozhmj3XAACO1I/gyNIR1JzkeqZZ2W46HBR5WMzCww88vO/nCYorL6mQm8akoh/qWYTc9iXK/UAF+XVZbbd5BR33XF/MkMtzyHwaxPVJjuPnacNqAbfXMFCNIfNiWc3KkMuHQk49ztH9k5738054lFKIskkTcqNm0s0bVAToVCyrqcRrvvdPjWyoxqNFzgW0rewbdxoBdvRMyhjbDeDFAH4awHXhwwC6AD4G4P07+fkaiwXPS7f/0DuoeVHIqe4UUDulqmE1hGMa6HoBudFx/dQmty2VQk4xILqeHwXPGyw+eRxaLeNbpwJX+an1FnBB1hplgzZsLpctnOuFffcboGVCbmtdtKzKGXKWycGMJBEZklSUeOzwhvAan8W/0+WiltWqbWHvUgkne0qzB7faOLBqxaQbNwCvkqqQA4B6pYNQ9+e7Mem5dylpWT2MB/E6+314tvEZfJyV8bfYCwBoMQZc8zycff2qGo8AACAASURBVPh/xq++7R4AwJ7evk2LC6hC7v6NVrSd99ZLQzUAVm1ToZCLP2e5FBByKtVWCNGymrxAkZtWaYZcyRycPAxhGgzcXQKcgCy9d/Pe+P2M+HuZpFLgj5/wx7jt3ttw0+GbYBmWUkGSZt279SG34tLVS7G/uh+H6gNkyFFiO2pZzZfqik7YGTluvdZBcK+eqhKQIb+uiAo5up/JGXIF5OOEiaOqpIgWOlAL/jxCNYbMSj02aWS1rM5MIUfGg7RxTsiQc6llNd8HWxpBNOp4Ts/zDXJz0daWVQDqY9ecws2sVGtyzvdPjWyoCP9FJlnF8jFNyKVh4mdSxpgF4H9DQMI9pfcZQQ0g8E8A/h7AP3LOz6e+iYbGCHAzLavxz3lRyPUrdaATHPkE7VgGGp2YkEvDoBlyHcmuGoIq5MZtWqXqqOWKHRFy/TLkZMvqWU+0rMoZcoahVgCGhBz9jrsSIceJQi5cLs65QMhVHFNQsj242cbZVkwK+l4NgAGDx4QPzZADgFKpDfQ2DSdZc5Rk2M028dvW3+AnzX9FiQUvLhMyuXXBI4Gn/yX4VhtAQMiF+/a2Swg5opCj5RzD5McBoUKOkCuyZdUJLKtZatR2H8vqWiVWyJ1tnR3bsmoZBny3jvCT7t2ICbmyWSWvm9zF0v7afrzkypdEv6smrGmTWIMZuPHAjQN/lljqELasxs9PY1LRD3RdGQgh17gEgHo/UIHmWDFWTILHlC2rvNgKOTkTT4bQsDrHhQ6AelI/L5l59OZJXhRyFaI8HoyQo46D2Y+LWVARNeOco+j7bRGFXJoSb9Gg2h+mYa9PuxmVhxtpGqNDqZAr4Pl9UtClDoNhYmdSxtiNCEi4FwJYQ0zCfRrA43sv+yld4qCxUxAy1wpQ6qBSFNC7uNTuJ7dhCU2rGQOcKkNO1bKqsqsCImkzbtMqJbXoZLqvQq4lKuQ6brZCzjA7gIKTiwk5YgvmIlHmUYVc7zvpeH60z1gGg2MZgpLt9FYbZ1ukKMANgv13V5cjFZyskLPtJiHkAmJotWoHJ672FvD5t+K6z/wJbrTE5bvDvRhA8FjLDIZvMR8x+L/ZjYkySmRRUvXQ6nAW0ESGnOECKoVcRl5Ey52uZdU0GLgXW1YpIecY8feyk3YCecLKWFCQMQmILZf5a1kFxPW/fPkR+PjJD8OAhe75gHgcOEOOjHm7qk4u1m1YWJLF083ZthoW8vrIWCcZcqsFJFCHgWoSNjcKOaHUwctdqUPasVNUy6rKSjqOvdRJIc51qUMApUJuppZVvV2KDGWGXM7HnJ2EaFnNx/w7jxjrTMoYOwjgpxAQcVcgIOEA4BsA3gXgPZzzHzLGNCWqsePIyuPJY6mDnK8FZCjkpAFeIOSGVMj1JeSIvYGSNuMq5KhdkbZM9hugZcuqXOqQJOTcPoRcvK5tT8yQ80BLHYLlakrqOAAJhdyZJi10CML4jyyv4XwvQknOkDOs+HNCQu5AzQC+8Hbgk28CGg+AUjZf8i/DH3VfiC9iCTW8JViunp1TsGP7PFD0uWrL6gnasDpEoQPQI5FohhzrghnDKeSGsayebZ4du2VVzpATLKsmtazu3MVSrSSSb/WSBTYh5ZpgGex9136OWlYBccJ+cfXh+Idn/QPe8m/34v3tgFge1DZNJ+C7Cqq2sqTCk6yYhSJALKlQWVYXJ0NuUUodkpbVydxcGBZ03EhTAtNNIpQ65FytorKSjmMvLaXc9NDETwAVMTadUgf1sbPI5M08QN2yurjbVFtWB8PIVwuMsf8J4BYExRAMwL0A3gPgXZzz2yezeBoagyPTsiqQBFNbpEyoFXLETtknQy5EFiHXUDSbql5PP7eUYlk9eX48hVxTyJAbXCGXbFltZyrkmKJhFVBnyDU9kShzESvSwuUS8uN6hFxSIRcXOnAvUMgd27WGb94fPCYr5Dza7upV8Azjs/it7fcDHzklvO5O/zDe7L4A/+rfAIDB8B+IngvJKmHf5hwdvwOPB8tsGzZsM/6uTlJCbiTLKlXIqVtWrYyKc6FlVXExKivkaIbcKAo5I8yQ64Eq5GxGFHI7OEGrOeJpdpI2NpVl0BVyyWZ/EUhVQlttFxetXATmbSBUeo7SslrE/DhA3M+6XvEVcnSZfR7cEKD7nNyyOs9QTbjmh5DLX6nDMrmpl6Y4TiNV8k5EqdRw40zo09R1WiEXQPX9TGM8TrOs5p0w1siGipAzF3ibasvqYBjnTPpkBJbUdwP4c875ZyazSBoao0EsQRAHP0MiLfIAV1XqQBVyfvodXapiUyneQtA72SGGy5AjCrn1ybWs0slZFqEIKBRy3imxZVXOkDOzCbm2QMhJpQ7oAMwFuBV9JyIhFwyZyQw5Qsj1FFmX792Hj/QIOTlDrkussm823oFnO/eIqr7lw+BP/A08/X270PHjfZf7MRERKeTIud/nXCh0qNk14XPHyZCrOZagTEyUOvRaVinpICvkaI5gRTGJ2l0hCjnynQKjZsiJltUzrVjJKFpWd1IhJ55mJ9m8aCnuPPqU5MlBhhxVw4b5RfQYHDRD7qpDyzANBs/nuP7Yrsku5JQgq0fzZi8eFowxWAYTyGCHrMciZcipbErzYlktSaUOAiE3oybZR1y0husvWMVd92/hxY9Ut02lHVN5VyCpM+TGsKxqa2QmclfqkPP9UyMbqnOBXcDz+6SgLauDYRJn0mcCAGOsDuBfOefqRHUNjR1GNyNDTrCs5iRDTpX91iaERdfNUMgNmCG3rSDkaIZa9B4pGXIHVspgDOA8aBPtuP7Id1UFhdyAGXKcc0Eht1peRcc9Ae6nK+SQopDb6gYkGP2Ot93NxOuY0QT3lqLlostdKjXR9tp9FHIBAXTF/n1ATyssK+RanZgYup6fiJ8orwKP+2XgET8LZlew9M//ijON2PYlEJG9wgND2LeRalcFZIXccIozWSHHDLe/Qi7LstpPIdc8IxCKo1hWTcmySmERhdxO3pFOEHITnKTTiURc6pAvkoda2sIbBHQ/GLRl9fBqBe/9uUfhzvs38axrD092IacES7IYezmzF48CUyDkfDiIt6eQITfnhJxSITcjsmrSoCrW7Y4bKZ0NBlQGJNQnDcs08IFXPzbzmiRNIbeTmaGTgGpMHEfNlva3WiEXQGUHnoagKe3cpwm5YkNZ6pDzMWcnoRVyg2GcPeTZAD4MoATgJQD+GcApxtifMsYeM4mF09AYBt7AltV8EHIqhRwlxrp0faTBvDSOZVUxIIoZcvF726YRqcE4D8inUUEn4csDZsg1ug24PJjEV6wKSmYpUNdkWFbB1MsYKseoOmfb3Uq8LiSZwlKHMIfPrN6Fk0u/iZv/7mZw81z0+ge3xAw5363hgrUqDtRjFU+UIbdxEvjQL2Br+/7ouRXfR5M7+NLRlwOv/Rrw2NcAdkAWyTYvSkSmWVZDJSAgNqwC41lWq44pljpILauxQi49V4q2rKpybSghd7Z1duxSB8tg8FMIOcGyuoMXwHU5Q648OWJCJHh6pQ45I3nqpXh9t1SE3BAT+odfuIaXPPJYYZVH8rGRdc4qCuyUwHhAypCrFNNmPCjUGXKzIasmDapiPUtuENWcyeVhjoosUinNsp93tYpqncaxrKZllcllYYsK1fcw01KHBbY3zgN0qYMInSE3GEY+6jnnH+acPwfAIQCvBfAVAHsA/AKATzPG7maM/T5j7OrJLKqGRjaySh2MHJY6UEtqCLHUIV5O+QLSsQa746AqdRjGsgqIih5qORwW9G8HVchRu2qowOp42aUOzIhVGQ6LCamt7hY8qdlws5MsfQ5tmCFRGFpt7ZWvgjMXm51N/MeD/1/0+tObHcEKyb0lXHVwGXU7JoI22xvAx34f+O/Xw/3qO7HZO2EzzvFPncfjCe3/irse+jqgsiosy7Kcu8TF9fZ8T9i3PV+0rFKF3Gari42eZdCxDOweMoerKivkzBaYEUzOTGZGarY0y6rn82g/Y0x9d7hqVyNratfv4nTzdPTcaC2rBri3pHzOAlHI7eDdS5k8mmSGnC1lkgH5s6xSUiIk5ATL6oClDvMAmiPj+lwY44uaMSOQwtLNFSFDbs4VcipLISWjiwxKyJ3eIoRczonxtPEv75NjdYbcGKUOWiGXCdX3MBXLqmKbGiwf2a8ao0NZ6lDQ8/skoC2rg2HsPYRzfoZz/qec8xsBXAPgvwC4D8CFAH4DwNfJy9VBDxoaE4CbZVkle3peFHKqgYnaSam6SL4YG6Rlte16KZ/RRyEnq/HIZ7XGIORoFtvKgIScYFctrfaW1RNLHVyRkOMsnjBUzJjg2u5uC+tZsoxE+ykAoKf66kiWVWbFr/3GmS9H30uz6+F0k2bI1XDVoWUsOTERtNk8DXzqzYDbxCY9MftlvMH9OTyAXUIuXYhkELoBm8Wva3ttyOd+SshV7Jh0OkVKOQ6vVoZWNlQkhRyzYjJzpbQSvZ+dUupA9+2SZaR+Pm1a7fhkW46SIWcywHcEZWH0HIsJvp1UJ+2kZVUkQ4LvOu+lDoBoG09rAJxH2BkZckVVyAmqP+kmk5AhN++lDlZy+82qgXTSoJbVs434fJt3S27aHDjv2WnKDLkxlllnyGVD9T1M49ypUocvsrVxXqAi/PN+E2AnoS2rg2GiRz7n/Fuc818DcBTAUwD8PYDw7M0AfI0x9mXG2BsYY1dO8rM1NCiBJU9uclnq0Echl5WJN0jL6nY7nvTS70OpkEuxrALiRQNtyRwWomWVEnLp20OpkHN9QSkWZqmF8BGTOGUjJuQa3UaCFNpoZynkxFIHZsZW0K89+DXsWYqXgVpWuVvH1QeXUPneJ2D2Vq3NWLRU6/uviF7rkcKBPUuDEHKAZcSva7pNMMYEUm6rQyyrRCF3gthVaVnHoKg6lkCEGlZMloZ2VSDdHi40rGbYFNcqa8rHS2by++mHYFnUOXIGqGV1ei2rk1SV0OUOxwtfsKxO7KNGBiUg4wy5BVXICfmKfu7I01FAlWGyQk6wrFbn3LKqYH+W5kQhRy2P9H5mYRVyOT/WVATaOPZS3bKajZmVOujw/7mEUiG30ISctqwOgh0ZjTnnPuf8f3LOXwjgAICfB/BZBKTctQD+TwC3M8a+tROfr7GYEBRy0sVxHksd+qnXshRyVAKcVurQIHZVOhlSEXLtDEKuTH5vj2NZFUodaIbcYJbVSCHn+QOXOpTZSvTzVndLWM+SbSbKFoCg1IEuV7P3PTIrzptruk0sLd/X+41jvR1nyl3hn8FNn3052HtfhCU/XufNlcPAs/4M55/11ugx7sWE2Z56csKqIuRsQsiFZCQlwbaoZZVkyJ0ao2EV6IV3pyjklksxISfeDYv3caFhNYOQowq56D0NG5Yx/OQvvDCiTashTMGyupMKOTlDbpKW1aRCzssYB2eBukohR4jx8gIp5OiFuudxgTzNO0mQhjQC3vN5ZJEHxNzQeYRKATGPCjmKvGfkpbes5nvMURFEY1lWU7afVsgFUBJyUxiPVdcdech91RgPquuuPFyLzQrCnMDNx/w7j9jxPYRzvsE5/wvO+U0ALgfwBwDuRUDOPWSnP19jcSBkyCUsq3TSMLVFyoSKiOq4PnhvktbNUPzRO2sqCyogW0QthJykK1mlgOwMOapmSvusQUDz7Aa1rCoJOUkhJxNyPuLfHULIyZZV23QFS2SIqNShtz8FxRhcUMgBgFG9u/dDE16veMLxGP7Z+V2Ujn8WAFAnKsitl34QuPZFOE9y67gXk0KDWVYBx4jVbc1usKzU/tlIyZAbp9ABCI4hi8WkoWETy6qzIrwuhKiQoyRMhkKunFTIjZIfR5dFVexg8Pj73smJiWxRnWSGnCkQcsmW1bSWwWmCEpBbCoXcMKUORYeVmSE3+201CtLufm+Q/LilspV7EmRcqMaQvCvIBkWailVW/+YNabEIeVerqEsdxrCspvztoA3X8w6V+nAaimXGWGIbaJK0+DC1ZVWAtqwOhqke+Zzz73LOf5tzfhGAJwF4xzQ/X2O+4Wbk8QjB93mxrKZYNUPSi6qLRsmQCye/QEAKZNlchWy1HcqQo5NwalnNaomlGXLUsioo5FyZkIsngjaWwBBs+5bXwnYnfq3tqNtYQ0KuE7WseoDRBjPEgoy2+V0AQMmKScP9fgcGePhGWCLk0iYPlut8J16nUCG3VLKUJFU/Qi5SyJH9O63UgRJyh1dHI7gcMybkhIZVopBTNX8CUrNmxkRgdyWpkCuboy1vpJBzk8UOomV1ihlyk1TIKSyrlJDLw7V9TbCsBvtAiyrkFmhSKB8b85AhJ5DCZH3WCSG3OueFDkCS5LHN5GS7qEi7gZL3DLlUhVzO1SryNRgwJiGXVuqQhxNEDqD6fqY1HMufvcjEzbxAFV+wyESrcNMuJw61PGJmewjn/OOc85fP6vM15g908i/fjacXZvmxrKaVMYTh7HR9xJN0aQBCjmbIVR0rk8TLypCjF+N0Ij0MXM8XGjbpJH3QDDlRIRdP8JIZcjHRxuBE7Z8AcK4V205tO42QEzPkmh0XzNxKvO6cfyd+3PgP/Nfym6PH1sJtetmPAa/+HJb3XhU9F9pj11vxOoWEnCo/Dkgj5MQMOUDcvxturOSj6y5myA2vkAOAUgoxRhVysgooxMAZchNVyAXLosqQYz5RyO3gBE1WyO10qQO94ZCHXDKqCNxshaUOg+0L8wZZPer686CQU2fICflxlfnOjwOSJE+tZA1dnJNXpBGLkxzLdgKFbVlVKuTGyJDTpQ6ZmFWGHCDmMwL5J4s1+kN1Li/qDbdJQLSsaoVcGvSRrzE38DLsP8JEKC8KuRRiMMxXokSVfJJ2hsyQq5Us4aK67YnEWocQbUlCjirkRhtMaX5Y1TaFz5h0hpxHSh3gO0KO2vl2TKyZVhMqxJbVWCHHrEbidV2+jdfW3wqbZKmVPAf//qg/A17y98Deh6DuxERQSMhRhRzzg2W74kBSwQUAyypCjpBiLTcgI+nuvp2SIXfy/HiWVQAoWWriMFUhR/bhdnew3DClQm5EQi7OkEt+v4xYVndyglayDGH7TFJVogrUpzccpjWpyEJNKnXgnAvE/ryoiAaBLW0vsYBj9ttqFKQpYhdOISftx3knq4aBYTAlqZN3S25qy2rOSY9pWVZ1qUOAWbWsAsnzX97t1Br9odqGi0x+a8vqYFjcPURj7tAlkwF5QDRyWOrgpgxMoVqNkhmONULLqkDImcKduLZErAkZcqZMyNG/G00hJwT6O6Zwt8j1eeo2UVlW25JCTrasupwo5riDuk1IMYGQi19HlWRxqUOwTM2OB0OhkAOAL5bLOGPG389nOw/HnmufHv2+5MRE0FZ3K7FOz7n2crzsMRfiN5+qLp2m5RchSgrLKr14bHRiQq5iBcSb73Pcd56WOoxGcFVTiDFRIafOlKIkzLClDhVzNALRjCyrSYUc5/G67GS+FWNMmLhOMkPOVpAhdFjJw13ZqmNG+ZXNrodW10fIQ1kGm/tsMQrZ3inknuZgW40CM0URe347JuRUSt95g9yOOE+EHKC2lud9HedLITf6OMmYmlBdZJKAQvXdTOtmVtKyqrdJ0aFUyOV8zNlJaMvqYNBHvsbcwMuw/9BznFxoMCtQBZyqpEGwrGYp5FIIuUaWZdUb3LIqZMiNKDdudUSLGmNMGqTV7ztQqYOfrpDj3BbIts1OTKzRDLQj9SPxG4QZckQhZxEVnEP2ny+Wy7iNkW4afwWX7I0/TyADQ4UcIeQec9FR/B/PuBpH12IlG4VqIktVapFllWbIuUmF3OmtdrS/7araqI4Yxp2mVAvJUkDcV+mx1uwMZlNUEXIjK+TMdEIOXmyjkyfTkwaduE5UIWcmyZC8WVYZY6iT/e1MIz5eF8muCshqMg6PjHtmzlU7aRCbflMsqwugkJMn0nknq4aF6ljN+zqmldrknYhSKdrkm7LDQpVLpxVyAVSKplkp5Ip6Y0Yjhmp8WWQrsrasDobF3UM05gqcc7EEQRr88ljqQNVD1VJSvdbNaI0dyLJKSh1qjjlwqYN8MhEy5EZUyG1342WpOmbic9Jy5Ch5tVpaBec8aVmVFXKUoPOzCLlYMXZk6Qh5vJch1/tOLt34HF7mfCB6/jHNmMj7RHkXPm1cHP2+r7ZbmJgtO7GVM8qQU5CMaVARcmWiFossq+QirukmSx0mkR8HAFVbTYzR9ZTVj2Fr8CxaVsPj3vdEQq5kluD78TLs9F3phx0JCMuVio0LUsjXUSCqEcNSB0Ly5MCyCojWtjNbMVGTZV2eR4gZcr5YRFTQO+hi02+KZXUBMuTkSX3e7ZzDQjVm530d00iVvJMeKqJs3Am96j11qUMA1fcwrQgBebvknSzW6A/VvrPIVmRtWR0M+sjXmAtQ0RtjyQuxPJY60MlYjShI4gy5dJJsEMtqoxMTILWShZKd/jdtL10hJxJyI2bIdZJ2xUHumlDyaqW00iN4ICjkWl4rIn0AwOUxIcd9sdRhi+SrwVAr5ALlHMdq6zjw7hfiVx78LcCK/+5GawW7rOA9udmCVf1e9NyxlX3C8isz5KgNl1g9VVAq5EiG3KAKuZPr1K46OiFXsdV/SxVyhsGEzLRQJSc0a2YQMculZVhMnOhNumW1ZtcE6fxOXyz9/rMeijc87Uq8+2cfObI6UQUxUD9pWc2DQg4QVYFUIScHWs876KS663FBQZqm5sk7bIVKEwDWtxcsQ27uFXIqy2r+j1+1fSzfUx8VQTQuUaPK6tQKuQBy/iMwRcuqtF2LemNGI4bqerKoGbGTgOCG0oRcKvRorDEXEMgrxZ3EPJY60GWmF++RZZU8L9/RpRdS7dSWVTFDTmWLDUEJOvnCTSiDGLFltalQR/W7a9L1umh0gzIFgxlYcpbIchrgPJ4MdPxYdePy+GffExVyjW6skOOEkNtd2R3lrTHm45fs9+L3T74CuPOjAICzZFl33/K7uOHQo6PfjdLp6OfL9x4U1kGVIbfRie2v/RRy9ZKVOJFXrOxSh5CkA2KF3ClS6HB4xPw4AKinKeRIqQOgtlJSMjeLiDGYgV3lXcJjo7es9r4YvwQDMSlQs2vS8bWzp8K9SyW84nEX4+pD2QTssJAtkACEooC8KEGokub0Znx8lhZMIWcJCjmRkMvLthoWYi5efEydby5YhtzcE3LFU8gBamIl72oVZd7bmJbVSTe3zhPUpQ7T+Wz5HJj3whGN/lDFTyyy8nEQN5SGJuQ05gRZ+XGAZFnNCUFP83aogiQky0TL6ggZcp2MDLkMQi6RITdhhVxoWXX6BH3SNtIVZwUGM4TlZKTYISSmgveKf+a+JRFycVsqR6wkW3aWBfLsxc5H4CCYUPpguMOIc83Wqntx44Eblev50ANHhN+X7Pg9QyKOqv5kIksGYwzL5XS1WEi+DWNZHUchV3fUf0stq0DStgoMblkFkk2rIVk6LGLCisFBvIx1u55pCS8KbCNJavcbC2cBWmRxmmbILZhCzpQIVDeH22pY2CmtymKG3PxbVuUxpAhk1TBQHatFIB1V/Ebe85x2wl6qfE+tkAOQYlmdkUKuqOcBjRiqm2tFveE2CWjL6mDQo7HGXMDtM7nOo2WVDkz04r2jKHWQQ+dLw2bIlUwpd05UugmEnJwhRxVyE2pZBUSbgMqyut4iWWvlXqEDXVdCyHW8ePLXIRlynmRZpWSVx+Kfl5rrWGmci34/H34Hh2/Az1h/iO8ZsbJprbyGG/bfoFpN3HD0qPC7oJDrbKHrJ1V//SCrS8qEnApbVk2BkIvJt3DdT9IMuTEIuVoKISdbbwVCrrfNKCGX1bIKJHPkRrWs0juVNosJuapdFY+vghJyloIMyaMNskasbTRDbtEUcpRAdT1fVMgVdR9MaVkVMuQWwbIqkTyTLG/JA1THahEIORWxkvdjTUUQ6Qy5nYOyZXVqpQ7itVDe902N/lCp4RaZaNWW1cGgR2ONuYDYSKog5HJW6sA5lzLkSKlDzxZKSUY540LMkFOTZNsdWqRgidZTSenWGTRDblTLamd4y2pqw2oPgkLOowo5Qsi5okJu240Vci5VyP3LG7Dc2ox+v5ct4b/VXgv8zG34onsRmBX/3Vp5DZetXpYk0zjDoSVR2SVnyNH8uGVnGQbrPwTLhFzVVpQ6kP275WVnyI1lWXVK4Fwih40KbFNcRrVldbAMOSDZtDpyyyoZCywuKuQEEj/niok0iJbV/Crk6qV4/ziztcAKOWP+FHKy6i/EeZohtxCWVXH7FSFfbRgU1bKquimR9xswlmlAHg7GtayqYiK0Qi6ATKAwFrgTpgFd6jB/kM/ltsmmtj/lEdqyOhj0ka8xFxDb6lR5EPlSyMkTMXqxq2xZzciQS7WstmMCpF6y4JALMllVl2VZVS3bsKAKOVXLqkrld7oZZ7PtKgWZYjT7zkBsg6JNq21iWZUz5FpeqBTjMLv3RY8veV2sEFL3l/2X46PWjwKGgWanC2bGhNyu0i6Yhonr910vLK/NkgQbJe02O5vYaMf5cbQIIQvL0mSWZshFltVo9+BKyyrNkBvHslorWYIyMfiMpMpPVMglM+T6WVYTCrlxM+QAWBAVcpQELupdadtIXujQGw65KXWgCrnG4ras0v0syJDLXyPusLAValhAVMitLIBCTr7uoCT0PEBVClAEFaBqDCzCDRj5Omxsy6ri7zUhF0AmaKc5FsvbYJGtjfMCeX8qwnizk9CW1cGw2HuJxtzA7ROOTa9F8qCQcyWyraQoachsWR3Asioq5AYvdZAv3OiyTUIhF9oVhQw5xV2T41vHo58PLx3uvY4ScvGEp+0FhBznHB2ilnM9UyLkGriEncC77DeCde+PHl/2fSzX4obUlumh6/nouD481gBjwfItO8uRGuzG/WKO3IojFhEAYobcZncz0Ro7CGSFHM1TCxVyEfHE+9pePgAAIABJREFUPHg8+K4tw4Jt2mh1PZzu2QRNg2Hf0ugKuYpjgvvi8tTsbEKuq7Cs9iNi1ioiITdyhlyKQq5m14Qxo6h3pQV1Uu979nNYFEAn7qepZXXBFHKylZsO3YVVyAk23F6xiM+FDLnFKHWQM+Tma99WKuQm2Bi9U1AdV3lXyAHJ67Bxz1HqUodinvcmDfm7nuaNLJno1tuk+JDHnKLe8J0UtGV1MOT/bKqhMQDcPmoXsdRh9oRc1xfJNlWTqWDDldZJtKyqB7gtIUPOEjJgEoTcoJbVERRyd567Ew80Y/KrMqBl9fhmTMgdqR/pfX5M6ggKuR4h1/E74Ai2L/dNuB4TCLmlxh14v/O3cJiHTSMuYFh++p9gpXsf8K13AACY0UTH89HseII6jiq35By5i9f2J9aBWla3Ols4145z6uTctTRkWVZjhVxv/zBipWCsjosJygPL5bEm/lXHBLh42qipFHJk24bHW8sdXCGXsKyOnCEXr2uJHwF6v164fCG+nEF4FwWCQs5PZsjlRXVFrW2CZXXRFHJS3pqXMcYXBbbCsrrVcRHuhlXHXAjidf5bVsX1q9hmIUhklWVV5aLIGwJHQ3wNN+74IF/XMZafGzazhhwJM1OFXEHPAxox5HNBUa8vJwUhL1xbVlOx2HuJxtxAVMhlB2r6OVPI2SYTmkxVLatyYPQgltVtqdk0i8RrZ1pWiUJuyFKH2+65Dc/9h+fifff/EozSKQBApXdXXSDkFOsgEHJLAXlG18kyYkIuzJCjbavgNlyfC4RcqXMCDvPgAdgkn19/2IuxTBVrZhNdz8d21wWztqKHKSF35e4rI9ILAPZIzaDBMlqRuouD49TWqei5MBevHzIz5DwxQ46xDnldmB9H7aqjq+MAoGJbgKSQW1IQi6pss2FaVmXLaskqDb2sgDgWLHnX4dU/8mq87OqX4XmXPy/TEl4UWAqFnFDqkJP1ouTEWcGyOv9EDQU9D3mJDLliXo6pjvVFy48DVBlyc0bISaRqEfLjANEdEaII470j7U/jTupVSqxFzrWimGXTadKyWszzgEaMhEKuAOPNTsLRltWBUIwzqoZGH8gWUBlm3hRygqLPUBYuZKn+HIXFVQZtWa2XxFIHmZCjv8sXbmUFWTgoPnX8UwAAHx7slS+h/cDTUekRfHYf2y21rIaEHF0n20hmyNGGUe472NM9hfq/vTd6rNG72PkMLgMQ/M2SvQTTMAXFGjO30fU4tiWF3G5CulmGhev2XYd/P/nvAJIkUoglZylaLrpOo1pWaxmWVWYQQq5HFp5Yn0x+HBDYsLiUIbfsLCdeJ1jzlKUOfRRyErlZMUdbboEA8Qy86tpXkeUqvkLOEm40BFZBasnPy2pRcoKSUKpcqnmGrCbzcmgvHhaWwrJ6/0Z8Y2Sl6iT+Zh4hT6SLQlgNCnnMXipAfhyQVuqQ/3Fn4hly0vuVCvAdTAvy/jDNoVhWDxfBTq2RDflcXoTxZiehirDRSGKx9xKNuYFo78wudcjDeCDkwxlMVK95AXHRzci4KvUhszjnkkLOylTViRly4gWCkCE3pEKOtoqatbsABDlkweekZ8h1/S5ONQI1GQPD4XqQIUfXyTZi1VRoWaUKuRXewbs6r0X1nn+PHtsyTPxq9+fwcu+XosfC4oXlUkwsMbOJrtuzrBKFXFguEeJxRx4X/Xz5rstVX4FAWJ3YPBEv34iEXNWJVXmRZTXcv1WWVdKwOi4hV3XMhEJOTcglJ+kCIdeHiJlUqYMcok8hKOQKehHMGEuQn14OVVdp5MQiK+RczxfISRVxUASI+19wHvnIN+LCnMv31xN/M4+QbW9FIawGhWxZLUpGntqymv9jbdLtmwmF3ILdDMmCIzXYTlMhJ2+XIuybGtlgjImFYgu+TU2DIRyGfZ4PUUweMV9XDBoLi2EUcnmzrFqmIea7KRRy8l0zSpqpLKsdMtmzTQbHUufU0ddH7z3BDLnzHULIle8HszYiy6olNESK73vf1n3wefDYvuo+lMyAfKOEXMkoAb0/iyyrpNDhEN9AGR3U/Xj577dW8S3viTDsmBgLiThRIRdkyAUKOWJZlcoGbn3IrUFzKgOedvHTlN9B3Y4npBNRyNnJltVw9xAUckrL6niEXMW2wKUMueW+ltXhW1YTGXITaFl1E4QcJcWLOzmxTBatm+v7wviWlwy5pRRCrrRghJycIZfHAo5hYUo3Vjqujw99NR5fn3v9EdWfzR1safvNu0KuCIUOgJpcKYItMEnIjZkhJxeDLbhqh0ImO6dqWZU+uwj7pkZ/WAaLiKeintsnBcYYbNOI5qpdz4dpLNa13yAoxhlVQ6MPsgoQgGR2z6whLy+VrYe2UJFkHC5DrtEW1XH9/qaTlSFHl21IhdxGZ0P43azdhYp9CwA56FNcnh9u/TD6ObSrAmJzrGOWoszjjtcBmutofeIP4+XuERO1vVcBCIhBt2dTZUZMUikVckYvQ67jglnqUgcAsA1bsEGqEL4/IObijVrqUHcyMuSoQq5HyN1H7GMHl8fLkAtKHcTlWS0rCDlJCQQMZ1m1TRtLzhI2O5vB60dVyGUc9+4cKOSAgExsIc6dFBVys1oqEWnkxKJZVmXFppAhV9B9kJLZns/xsW/fH+UEHlop47GX7pnVok0V817qIJPnRVk/3bIaQN5+tpX/72BakK95p6lWLtkyIae3yzzAMhjCq/FFt6wCwQ2rUC7Q9fyFc0cMAr2XaMwF+inkDCNfCjlql3NSWlY7A2bIqSyrcn5c+Dlpf5NFyGW1s/YDtawCgFW7i7SsxuskE4SqhlVAJBpD1RwAtI5/AXjLw9G88yPRY7YPvNF7KSqv+Fj0mM86ADwwMybkQsulnCHn86CpVlDIpeTEZYE2rVIF36ClDssJQi62rIYW3diymsyQo0RYdUybUdUxwSXL6mqpj2U1almlhFz/U8+h2qHoZ9kqPCjohbWskOtH4hcFcrGDl0MbZD3FvrdoF2Vilsp8ZMjJNty/+2I8dj/3hiOFaOKcBOgNJtNgc0c2yzEDacd03qDa/YrRsjpZQk4r5NKRK4Wc3i5zAbodi3x9OSnoptX+0Ee+xlygb8tqrksdmJL0EptYh1PIyQ2rgHiHtC1ZTwXLap+8umG+v412UiFXslnic+QBWlXoAADb3ZhopKqp9tf/Fmg8gCbZ9l/xrsJfek+BYTlC0yqMjkDIpWXIAcD5Zje1ZXVQqDLWgDEy5EjLattrw+d+tH+rFHLtjMKOYVFRKOR2lZPEokgSJS2rlQGImFdd+yocrh/GS658CQ7VD/V9vQqiIknc57NajIsEeuGXzJDLx4VgmppmEGJ2niAqtX2BFM6LvXhY0BsrJ9Zb+MR3Hoh+f94Ni2FXBYCaY+Khh4Mx/XGX7Zm7BsuEZbXICrmcjItZcCYc9j9pgm+ekCx1mJ1CrgjqTY3+oDfYtA1ZPMZ0sYMaxTijamj0gUDIKU5odDzMByEnEoiCZTXMkKMKHukCUlC7qSyrxNpZHVMhx1hwtz8kdtquF9lgs9D22oIiDAAMq4EHWt8HsFsYoF0/QyFHCbmeQs6Eh2Nbd+ArvcdbvQuoViVWUrX8GrxeTlPNqqHRDaynzGgDRlIhV7frYGDg4GBmC4CP9e0uDNqyKmWbDQKaIUcxKCFHFXKMAbZpomyW49w8txVPOhQKuazCjmFRdayEQm6tkiTk6CSo29u2LZr/NwAh96QLnoQnXfCkURcVgLrtNfo9Q4FaJNiC6soH5bbzQsilW1YXTCEnkad02MvLthoWdJ0+evsphIfZoy5ew7HdtZS/mj8wxvDun30kvviDc3jUxcOfJ/IOmZArimVVXeqQ/wlywrI65s20hPNhzhSc40D+rqerkBOPK03ezAfoNWVR1e+ThKMJub7QR77GXECYXBei1CFeXs86jpa/Hv0ekmVdqfiBwpZC832JbKCW1VrYapqVIedlq6jKGeq6NMjquBB3nP8iAPGOycCW1Y6Lh7Hv4R+dN+Di03F7attgwCN+Dq0nvSF+k56Sq+v7qDnxxJAZ7R7hFiBUyJmGKdhLYTaxMQGFHM2QoxiYkCtbuHB3QK49ZH/wXlQd2HSbUYORSiEnbNsxFUmmwWBAJOR2V5LrQbet541mWZ0ETCnfiiKrxbhIEEgeTxwL8kLyVG0TKtHBoinkxGxFPhe2abpOVJn9/BuOzmJxZoqlso2br9gXNYnPExItqwUpdShuy6q4jOOquOXrOpmgW2TIqrSpEnK6ZXUuQYlVvU3F70BbVtXQI7LGXEAIx1ZZVnNX6tBrQF39An5Q/n288esvBbOCvLUwQ45O1uQ7eIyxzBw5mrUWqlPEnLr49ZxzSUWlIuTix1ruYMUOcn5ciK+f+QIAMVR4IMtqp4GnnnwLPuj8Dq4y7oFDiNX21c8CnvpmNMl5j/sOgGDiW7OoZbWtzJADpBw5o4lz282IvGMwBFvroFARciYzU5VzMhhjePtLb8Trnnw53vqS6wGIhFzLixVyrK9Cbvwh3zYc4fc1lWVVUKYFNudwGzM2vfya7FKHbBK/KBAbbfNpgzQMppzAlxdMISefh7w+560iQHXs1EsWnvLQAzNYGo2dQkIhV5AMObVlNf/HWlIhpy2rOwXLNISswWleDshEqbaszgfodZk+1rRldRAU44yqodEHYt5adqlDHsj5cEAy698GALT9Fqz6HeiuPyq2rPZpgSyRGum2K7bW0DZSlUKuTUi1rlSIYSiuRqi1rDWgQu58hxBy3V2AfQ4AcPuZr6LpNlMlzOfb56N2zYpVCWyid38S+MfX4EfXfwD0Fs8hw1e71/QZlhwAMSHX9SSFnNlStqwCvRy5rfB1TZxuno2eKxvLMNjwJ1YVIbdSWhkqY+iy/Uu4bH/8PhWLNK26rZh46UfITeCuuM1KUVsS54aoKuzBkhScQsOqZU4tXymNiPd9Dj+H1s5RIBcFUBuk6lieFeolC1tEuQuMr9gsGmSbvjsPpQ6KycbTH3ZwoFgDjeJAJs/rYxYETQuqMbAIipUdL3XQCjkBthnHssxUIVcAslijP+g+VNRz+yShCbn+0Ee+xlxAUIX0s6zmQCEXq4ViAsUo3wcgVq/RQUtFMmZZUBu01CHMkEt5vVDokHKRJijkusMr5Lz2AXit/QCArt/Fl+7/UuoATe2qh2sHwf7xNcA7ngGc+0H0+Ke8h+LB638tXqYwT41m1oWWVUkhxySFHLWOyk2r59oxIVezBmtFlaFSwqUVPQyKskkUcm4rIrj6WVYnQsiRdlv4FSW5JrSsehIhN0USJi1DruuLx1aRA9jl79rj+SR5aooJ/KK1rNLN4XNx3MtLI+6wUAXkP//GxbOrzjtk8rwwpQ6Kw6oIKiT5XD2uqlzObdWqHRH0+53mWJwkXvO/b2r0hy1YVvWxZmvLal/ovURjLpDVSArk0LLam4hRi6FRCgm5gLyQix9k2BZQ2v8PKB9+J05snRKeU2XIyW2p0c8DKKiEDDlFiYQKG504Q853K/Aal0a/f/bkZ8UMObI8P9z6YfTzkQe+C3z5HdHvm6yGX+m+Ei/tvh7llXjS13YDIkpUyIWEnC+1rLZhWTFxlVDI9cDMJs53zsWvs0cj5FQKudXSaO8Vgirktt1tRF8lVciFLauEDJsEIUfJQOZXla+RibAW2WcGaVidFNKOe7fPsVUkCBc6vtiCnCuFXNlOPLZollXGmHBs0LG3CKodFeQbYJfsreH6C8Yb3zTyB/lYLQohp8yQK8CYnwj7H7dlVSvkMkFLM6basprIkNPbZR4gWlaLeW6fJLRCrj/0ka8xF/D6BJkbOSt1iALliaLJLN0HgEeEV78WSFb9Jpy1z8Jevh3vu+vdwnPblJCLMuTUxQyDZIyVBcvq8Ao57lfgNi6Pfv/cyc+JRIIbb5PjZ74T/XykFRcq4Mpn4KfLb8H7vCcAYFguxaRU2w++x6YbK9/ASYacLSrkLDsm7lIz5Mwmtrrr5HVxg+swSLOsjgMhQ87tkyHXp7BjWDhEIZdKyFHLquej2aEKuemRMKJ6zCc/Z9vBi4REqQMZ3/KSIQeoLW6LZlkFxP2N3twoqm1avgH2/BuPFlpxqqGGrGxeKgohpy2rABSlDpr4EUC/j2mOxQlCrqDnAQ0RlmBZ1ceaON/ThJwKei/RmAvIFjQZRVDIMbMFZm1EBFnfFkjnvujHU42TwlPUshqGqaeVQAyikCuNaVnlXgXe9kUADybl313/LjqI1Wddzwc4B772Xhz/wlujx4+4LlDbB9z6DuAFf4MfujF5tkQJOTdJyEUKOd+XCLkWkJUhF71uG00vJuRWS/kh5IQMOa8VEc6yZdUnZQrAZC7Cy4SQM7makKMB9XKGnGyd2UmYpvq472YUphQNcoEGJRvzRPLUFRP4RVPIAeLFOT0uijoRo+SGaTA857rDM1wajZ2CPG4XRSGnuilRBLvmxDPkNCGXCVqaMU1lecmarBJSIx+gN0r1NpUUcjmYg+cRekTWmAv0U8jRaw8vBwq5aNJMCDkgsK0qFXKKdeJWTGg1ulvCc7TUodpTptALMDFDrr+lUVXq0Og24PP0Ox3Ussq9KsAdWJ1Losd+2PxavIzNU8C7ngd88JU4zuL3PHr0ccAvfgG46pnBehHln6CQ85KWVRDLqlA8YHTgs+34fTIUcm2+Gf2+q7SWuq5ZUGXITVohFxByPph9Jv4MZ0XMjzONiShXdtsXRz/b7oXK19iSQo6WiOQhQ26eFHJ2hkIuT7lkqgn8NPeFvICen+ZBIXdgOR6Lbn7IPuwjv2vMD+RjVUWw5xGq46oI5PekFW2JNs8xW1vnDfQ8Os1LAl3qMJ+g404RWp13GgIhpxVySui9RGMu0C9vzchbqYMfKuTawuNG6b4o80vIxVMQZb4ZFw7IhNxWO6mQoyo3Ssi1B7Gskr9tux7ed+f78Jj3PAYv+eeXwPVd5d9stCkhF5BnVe/K6LEfbH8FAMet5sfxn+76aeC7twEAjlvxhf6RJ78RqATKNM45tomiZKUcq95UpQ68V+rgejyybwKAYW3BR7DMjuEI5JacIWdY8fe6uzKaQq5iVWAxcfIyboYczXFruk2YBoNRPgHDCojGPZU9OLZ8bOKFDgCwu3QU2/e8Aq1Tz8ZK58nK15hyhhyxSE9TFZWmjO0KZHexT4Nio62YIZcnslE1gZfVAYsAe84sqzcc24XXPukyPPPaQ/iDZ18z68XR2CEEN3Ti34uikFNbVvM/5stOj7Ez5BIE3+KNvVmYlWVVvubWeWPzAbod83QdNiuIpQ6akFOhGGdUDY0+8PxsNZkwMc+BQi64Q+CBGSKZZZbuQ3s9tKwSG66CNPAYIeTcTeE5VYYcPfFTxRIl59IyxmjuV6vj4S/ueht87uP2M7fj9tO349p91yb+5nyHWlYDQmwZV2MDHwYAfH/rS/gr+6u4xfw60FuELhjus20AwTY6VD8Uf27XR7jpSpaBqh2TUqpSB/hBhpxc6sDs2IYq20mpQg7mNhjifWVvdXdiHQcBYwx1p471dvy5wueMAGpZbbpNGIzBqsfZe4859BgwxoSswEkRclXHhLd9KbztS+EcUqthBNWWZFmtOFMk5JhIDHLOwRjr22BcJFBCsSu1rOZJIack5BZcIddP2V0EMMbwn598ef8XahQajDFcuLuG759uYHfNwXK5GNOHwrasmrJyarKEnFbIiaDXLFMtdbC1Qm4eQWNbimCR32loy2p/6L1EYy4gKOQUg5+okJvKImXC9TlgdBOPG+X74PocrudL6yReIPjcR5cQck23ITzf6CRbVoUMOXfIDDny+L2Nu/DA9gPR7w82H1T+jZwhBwAr5jHs6mWxNbxNHKrcEf/B2iW470V/A69Hgu2r7BPUa8I6lSzhOZVlNVTIdT0u2EYNQshRRZz8OzObYEQht6+2R7megyBB/E281AGwandFj910+CYASFhWJwFKqKVdaAgKOc9Hszsby6phMNB5THgd4PrZ40WRINqDeW5JHpWiZhIlI0VD2oRLT8Q08o7/8vyH4bnXH8GfvPC6woybhW1ZJerhScRNyOf/UkG237RAr31nqZDTaqr5gC2UOuhtqi2r/VGMW1waGn0g2LT6KeRywM53PZ6wqwKA4TwIwEPb9TPX6UzzDDiLCaqm14jUPwCwTUodqqU+pQ4D2BqpQu47G/+RWBYVREIuUMgdNs7jgo6Hj/Ye/0ylgsvbLm5beQ5+7OffguOn41y5I0tHhPejTZ1Vx4RjOtHvISHX9EjLqh+2rPqolmLLKrPi5ZKJMponx4ymsI0O5oiQk0sdPGzDqNzbe4Th0QcfDUBSP06ICKva8WkjjeSzc2JZBYJj3++R267vwzRMybJa7IsloWVVsqzmqWV1SVLUlKzJZBoWDWkTrjyRpxoaKtxwbA03HBstS3VWUFlWC6GQs6jCZvzllUs5JqWYnxfYUjnNtGCZBgwW3yzUaqr5AN2HinLzYiehLav9ofcSjbmAUIDQr2U1B5ZV1/MThQ4AwAwXhnNGUINZBktMXE9KraoedyNSCgAabYVCjt6h8HiUpdcZMkPu7sb/Ep472wqUenfev4lb//xz+M0PfgO+z4VSB/hlPNf4FH7v5CvwuAfujh7+SHUZL+j8Nt65+vOAU8XxrePRczIhR7+TqmMKOWpKhVzUsspRNgkhR2zClIADRKKMmdtgVqw8PLi0F6NiyZ6wQs4UFXJn/NvBWLA9D5Yvw2o5yKgbZNsOiypVyKXYXgSSyJtdyyqgJuOFfMaCXyxR8pMe1wCQJyFImGUZojzl/SAvSJvsFZ0Y1tDII+SbEqrrqTyCEmaTmNAns8pydHLIAWZlWQXELFV9HpgP0P2pCDcAdhqCQk4TckpohZzGXMDto5DLX6kDB1MQckBQ7NAgpQwqgvHU1qnEY5udzcjKSP8+tIoxxuBYRkTSdDwfZcMcyLIaqpqYtYEHu98VngsVcn/+ybvxhe+fxRe+fxZPumIPNjtxrt2fGW/Fj5tfBXzg8dsMFudwGcNdZQNb5n48qrcMxzcJIVeXCLk2VchZKJml6PeW2wLnXMyQ470MOdfHdks91GUq5KytiOTivoVd5WRb6qCQP2fsUgdLLHU47X4j+v3C6vXRz4Ns22ExtGVVypCbdrNmYE/yo2UJ/s8m8IsES2q0pTcc8mTNqpdlQi4/yzZNpE24tEJOQ2PykI+rooz3jjChnwAhJ5c6aIWcgFmVOgDBtghjPbSaaj4gKORydB02K9iSIEQjCb2XaMwF+mVC5U0h1/V8MJa0rAIhIReruFQXY7JCDgA2uzEBJmbIxRNhmhvSJsRcCCfFThjaHa3adxLPnWkFhNz9GzEZ9uXjp8B7WXB1nwdkXA8rK8fwyLWro9+t5dujOyYCIScp5LaFDDkTpmHCMoJ14+Do+l003diyGirkXN/HVlNNyMkKuYpVgYGQwCT7iVeHMcZJte6IZN4kLatNt4kH3Njqe0GFEHJeTIRNstQhRNpEwZZIItomOW1llHDs9y4E6AVB0Svp6XjX9Tnozcc8rZpc6rCIDatA+sW5JuQ0NCYP2bJalPGe5ms6k7CsyqUOmvgRMEuF3KTtyRqzh6VbVgVoy2p/6BFZYy5AlUCqCw1qW8hDhlyaZRUAjNL92OpHyG0lCbmtTlBAwDkXMuSooonmiIVNq+2BLKvBe5hLdySeCy2rlAT8wanvRT+vEFLoi/ueB7zqs/jRK54fPWYtEUKOWFaPLh0VPkdYp16OmWxbFVtWA0Ku43Gc21KfEGVCjjEGm9USrzP46Oo4QFTIWYaFqlXNeHV/UIXc7advR5MH24B7ZexzLo2eE1pWJ3QBfsWB+Du7bJ/6e6ENU4mW1SkTcpaCjHczClOKBrp+rucLjdN5ypCTSx0WViGXliGXo22loTEvkHnuooz3AkkzgZtp8vlfK+REiKUO0/1sSpZqNdV8wBIUcsUYc3YS2rLaH/rI15gLUEVZvZSc8NNzXC4sq55oWaXkk1kWFXKqwVxFyIUWUVoI4ZiGcKFBL8oi6+qgllXWFZo8Q4SW1e2epfQJxtfw4hOvj55f9n380N+LF3begE9f9uuAU8MtF9wC1ht+zMq9aPpB8+kwCrlgfeJih0a3AZcHrwneO3iN6/k4E4sHBciEHACUjCTJZPHk64YBzZBbcVbGzrChCrl7N++NfnYbl4EO622PljpMhgi76tAy/vuLrsMvP/lyvPLxlyhfIyjkfF8o5Jg2EaPKkOsKltVinwbpBXyeW1ZlhZzOkIthMHX4vIaGxnhIZMgVZLy3J2xZNQwmnJcndYNuXmDP0LK6VLajn2uK+YtG8UDHmaKMOTsJbVntD72XaMwFhFZRJ2lPzJtl1fV9gDR4XrP7GoAHhyOzz+Ls9lb0nOpi7FRDkSHXs6xSMq9aSm/WUhFysq0hetw2YFbvBjO6AIA9lbhxNLSseu0t/J71V/gfzh8BRrz8hrmGn+j8IT7vXxXZHdfKa7h67dpgfRnHlvkVnG+fj4ogymYZu8u7hWWQM+TC14VYb6/HnwkHQLDNXY/j1Pk2uBeTdyHkbDdATcjZbExCjnzOuHZVQFxvCnfrcsGyuBOlDgDwjB85hF960mVYqdrK54UMOY+j5VJCbnYKOVdV6lBwIkSwAvg+6P2GPBNyaWPNvEN1g0WrIjQ0dgZJy2p+xsQsCKUOE1pmeg2gFXIiHFJQNW3L6ssfeyGqjomnPfQgju1OOjQ0iocL1qrKnxcV2rLaH7rUQWMu0FCopyjEUoepLFImXEkht1pehcP3ocPuA2Mc92x+P3pOtlhwzjMtq5SclJsNaW5TmB1HM+TS8itKlgmrHttVn3LRU/CeO94Dl7todBto/eDf8Zet1+GYFRCF58kE81zlGjQQKLqoffamg7fg9rNfBgA0na/gxNaJ6LkjS0cSKrJLYP25AAAgAElEQVSmQLqaveWKix0oIWex+PGO5+PkegvcL4GZok14uZQk2soKQq40JiFHM+TGLXQARIUchde4HD4hnAchW3cCtmRZpRmP5Slnh5lknw4z5Pq1MhcJlrR+uVXIJUodFlMJoCLfNB+nobEzKKpC7tJ9dSyVLGy2XVx3wfjXDEBAwjV611E6Q07ELBVyt954FM++7rDeJnOEFz3iApzZamOpbOOWK/bNenFmDm1Z7Q999GvMBQRVWAEUch3PFwi5qlVFFbFF80Tj7uhn+e7oRmcD2+524j1DQi6LnKR3RcN8sUEsqyWLwSL5cTcfvRlr5bXo97PvfCaOIVbtfZ5dGP1s+PHdIToJf9zhm6OfXfu7+MaDcVOo3LCaWK8eIZemkDMJIed6Pk6uN8H9pKpMpZCrmIrHjPFUbfuq8Ql5f3X/WO8FiBlyIbzWfnB3RSBkdqJldRCYUq4ZzZArzaRltbcsPTa+26cEpkig6xeUOsTrNu07/VmoSgTcwpY6KAhgrZDT0NgZyAq5otyAqZcsvP/Vj8Gbnvsw/OZTr5zIe9IxVyvkRAgtqzM4b2oybr6wUrHxW0+7Cq950mW5ujE6K2jLan/oEUBjLkDtjLIqDMhjqQMXLKtVu4q6EZNQp5qxQk4+UavUcQAiu6fK2hlCsKx6CkLOVE+Sz3TvgWGfBwAYvIJr912L3XYsrT/TO+Fs8gp+pftKvJc/OnqOEUKONnQeqO+Dt32s9yKOd97xzug5OT8OkGzJPftbyYyJt432RvSzTQk5n+PkehPw48dCqDLkKlZSIVc1xyPkHnngkXjKRU/BlWtX4mXXvGys9wLUCjmv8RAAEBRygxR27AQsIUOOo0XKJWbasuonFXJFz9KRG23pDYc8XQgaBhNsq9MmZvMC1TbJ03bS0JgnyIdWUVpWAeDy/Uu49eFHhYyxcaDbPNNBizN0nqeGxmRBjy+tkFNDW1Y15gKqwH8K+QTr+3ymJ13XTyrkVsyjON7jnE537gFwEwAFIddQE3Jb3Z5Crp2hkFOVOnj9VVTfXP989HPFvQr2F/4Sa/d/G6gEJNcZ08Tn/SvxK92fx3G+FyXzn6LXcy8mj2jDpmMa6G5eA7N6DwDgBxs/iJ5TE3JJhRwl5NIsq+vbXWy0XFQUhJxKIVe1ko/VrPEIOdMw8abHv2ms96BQKeTcrcsBiKUlHXc2yjS6z7oejxp9gdm2rKoy5IregEUVfq6kkMtbc2etZEYN0tO2LucFKjVc0fdBDY28Qia7i6KQ2wnQ67tFzfBMgz1jhZyGxjyDZndqQk4NPSJrzAUaNDetpOaZd9K2eq7R6f8igqBlVVTI7bKORb+vu3FzpnwBeWortoZyP75zGmfIUeJKypAjpExIkgxia/zamc9GP7944zvAv/w6drvd6LETVz4bL+r8Fo7zvQAAZsaWWj+FkLNMA+7m1crPU1lWt4nyr9Jbr7QMOduIH7/nbLAsfECFXM1KPrbk7FIu56wglzpYrASveSEAcd/uzEgJZkokmNiyOnuFXMebp5bVeP3oscxy2NxJFXLTbtvNC1TkW962k4bGvEC27Rd9vB8HyyTHc1Kqu3mBQ66z9XisoTFZaMtqfyzumUljrrAtZMipJ/w7ZVv9vX/6Fq77vX/F6/7uqwP/jev5gKSQW3UORARbm58HMwOCTbZY0PIDv30g+nmzE7asppOTKoVcuw8hd6Z5Bt89/y0AgME5Xrr1bQDAbkJq/HDv1eBkOGFmM/rZcwkhR7aNbTLw7hq8ZpJ8UynkVBlyVCF3vn0+fm+ikLv3TCP4QSLkGJhSIVdTKORW80bISQq5Q6VrAB5sa7przypDTrBR+r7UsjrtDDmVQq5/kUlRQC90qBIxj3f5BcvqgirkTGWGXP62lYbGPEBWyBWlZXUn8DM3XYzlsoWnXHMAl+1LRnMsMsRShxkuiIbGHEJbVvtDDzsahYfvc2x303PTQlBey5+gQu5d/xFYLj/w5ROCOi0Lri+2rFbtKsq2Bb8dB/4bpfsBKBRyjVgh57UORj+f7xFydBlkclIodXCTGXIlxZXIp+/9GDiC7+vadhurvWD83UcfFb3m9PYZ8Y8IIdftxuSRQMj1Noi7eU3iMw/VDyUeU2XIpZU6KBVynkjI1e06DJZcXxVJt1rOFyFnMEMgI49Vro9+9nNR6iBaVvOTIRcsB219LXqgPh0f2uR7zuNd/ppWyCkJAZ0hp6GxM9CW1RhPe9hBfOV3fgxv+8kbEi32iw6qGEybQ2hoaIwGqkDVhJwai3lFrDFXaHY9hPxa2TZSJzc7oZDrer5ANmy1ByPkOq4PsJiQq1gVOJYhEXL3AUhaLGipw5od21wf3AoIqSz7LlWlDJQhd9/t+ORn/iD69QnbTTyAXcBPfQhrP/KT0eNnWiIhRy2rAiFHyBjDYLAMhu6GSMjtrexVlhYIhFyokEuxrDqUqNsOrLWyZXW5lLSmAkBdsrFyr4SlUnJ5Zo2VUpxrd2Htuuhnum+3B8gH3AnYhqSQI4T5tLPDxMbX4LuhkvnCK+SMJMkO5F8hN21iNi8wdYachsbUIFtWF73NUpP/avzENQdwyd4aDiyX8ZzrD896cTQ05gr0xre2rKqx2GcmjblAIyMzjYIqRvwJEfSUJALEnLMsJBRyVhUly4RHLKghIScrKqhC7voDV0U/n2uFLavp34daIecln/d94PNvA3/7LfgiIQ6trYvxDPePgEtuxu7K7ujx8+1zwudQy2q7rVbIAcHFMe/ugdeK11tlV5XXq9rHsuoYybw4+KLNU6WEA4BlRyxw4F4d1RySB6982CuxWlrFS696Kdac+AKSZshRxdTMMuRkhZwz3dNOv5bVoismBIUcOZbzSPKIltXFvPxQbRc9SdbQ2BnIh1Yex0WN2WOt5uC21z0Bn/n1m3HFAfXNWg0NjdGgLav9oXW5GoXHdkZmGsVOlDo0JUKuMahl1fPBnLjUoWJXULJ8IRPOVFhWt7vbkRLMMizccsk1+NTXesvihS2r9PsQiSQ6Ce4oLKuOaQCb9wMfehXwvX/DadPAuhm8h+mb+J3GrwAwwTnH7jIh5LpnyXswMCNWyLVaDllPmZBjaHYD26pZDgjIo0tHoQIlP0OiMa1l1TGTLaTcd4TfVYUOALAiK+TcemSRzRNufciteP7lzwdjDH/1me9Hj9Ndm6ofS1MkFeXmz3Z3dqUO9M5caFXtzpVlNV7+Vs4tq3uX4+N1reZkvHJ+oSKANSGnobEzSFpWiz3ea+wcGGOFv0GnoZFH2Nqy2hf5m2VqaAyJRkZmGsVOWFblzLjGgAq5rscTpQ4laxt+S1bI+YLFgqrjDtYO4omXXQD0CDkPLZxrtKUMuXSFXEjWUNLmwP2fBP7pN4Dt0wCAO5140txtHwEQfL9t1xcUcpvdmAy78nAFdxu974FbaHbiz5QJuXB5uucejcsv/h42Out4wUNeABVUllWaIbfR3oh+LhkqQk5UzaUp5FYkhZzv1TL3q1kizIFRqcCA/vmAOwVLsqw2c2JZVSnkim5Zpd+1UOqQQ5LnJY84hs/ffRbLZQtPfejB/n8wh1Ar5DRJoKGxE0haVvM3LmpoaGjMM3TLan9oQk6j8NjOyEyjECyrE1LIyZbVgRVyvg9mxAq5ql1FyW6De3X4bg2G1QAzO2D2OmwzVozR/LhDtUPYXauCcQecdcAYxyfuOi5lyKUr5EI7Y8f1YcPFr1vvwTWf/Kjw+rsuuxnY/CYAwOgeFP52tbwKBgYOjqa3AcADYOLSAwx3B/0S8N0KfB58745pJO5Oh4M092r485vfi/0rDixDvQ0ForG3nWmGXFg8AQBlRQadbFlNU8iVbRvcc8DMgDDlbj23hFwII0X9OatSB7loIFSmGWz6EyJlyypVyBVcMSESckQhl8MMuQt2V/HhX3jsrBdjptAZchoa00NCIafJbw0NDY2pQiTktEJOBX1m0ig8VNliKuyEQo4qf4DBM+Q6ni8o5CpWpUeWsUSxA52sUULuYD0gyMpmLXrsM3f/MDtDzqQKuWBZ19on8H7nd/EKi5BxSweBl34Yd+69OHrI8uKcspbrwTIs7CLto8xqAABKxIrL/ZgYU7Uq2pK1MY2M830ukJ+h0o5aVinKiserVlX4PU0hZ1sGuB+/lns1VOx837ugc448tKzS7Ur3x7JtTr3dTaWQoxcERSdD6HdNMwMLzjPOLVSEdB7VjBoa/z97dx4nW17X9//9rb167763525zZ7n3zr13FmYGBhWQHYwEREAISAzGFU3ykxAMGoMSI4k/89OYnytGTTSZH4K4oKAGZFFDBEFhWGT2O3Nnu3P3rddaz++P6jrne059T3VtXevr+XjMY6qqa+uurr593v1ZxgFbVgFgsMItq1TIufArO0aea7aYS1xbX68eWwqHD82UKgUZU3sO6URG6UTar16rFoJKtNTUo6EKnlNr4Qo5SVqwtoV+/rFT4YAy624RlbbCmq/+vn7+0g/p9kQwg0xHXyH94F9Lh16shy4/5F+cre73T9e3Zi7llvzLTLI2w07WQgevEgRb0YUOUutzBTbL9gyyYJNubCDnqJDbPR1uRY2rkMskE/Iqwe1HoULODpvt6s/QBt0BLXVYiQRy/WYfgAUtq/aW1dH+ZzB+qcNof17jyhW+jXooDAyr6N9/0vxcBIC+ylAhty3+ZcLIW20SQNns38N6ViEXnSHXasuqZy102AqP6oFcefWo/7HU7NdkFzY9vWrNkNuqkNs9FQRNp65e0olza/75aEBZf4ycCnrFIz8t/cH3alq1AK3oJXXueT8pvfkD0vQulatlnbh8wr/tlAlaZ+utcfZiB5OqBXL2Qgc72IrOs5PCYUixHP+ahBZVWPcTH8g1zpDbOxMO4GIr5KKBXGX4A7lQy6r1b50d0GQdFYo7xT7osQPi6AzBfkiGljpUQ/+XRr9iwg7eQi2r/Os+lFzh2zAu4ADGQdJQIQcAg5QikNsWv7Jj5K03adG0hVpWd2iGXPR8nLIXVJHlt1ops1vD7ivrh+VVakFTInNR695T/nVdFXJzWStYSmzq/GoQ9kVnyGVSSR01T+jDmR/XXRc+4l9+srpHry/+e63f9Vb/T8qPXX1MpWpJkrRnao+mUjP+9f0KuXxjhVzVCuRkVci5qqMyLa7CDs+PC+7HFbxJUt5x+f65xdD5uay7Qi6dNKqs3yBJ8ryEKhsHndV9wyQRVyEX3aDbJ8mk/XyCy/sZCtalnC2rVoXciCdX6VCFnNWyOoQz5OCeWUiFHLAzohWpo14RDQCjJtQNVSaQcxnuwUhAC9Y6WeqwQy2rq622rFoVclN+ILf1i6KXUnn1mNLzX5EkPV35W0mvqJ12VMjZlV4muRl6nNDXw/N07Kk/0Icz/1E5U/Iv/ph5vn64+F1a1VQoIHvoUtCuetPiTTp3PgilNkvxFXIVBRV6dqVZfpsZcs0CuXYr5KZSeUnh+7t2YVFaCc7HtaymkwkVL7xE1eKyqsXd8kpLzuq+YZKMqf4c2Ay5mICh3xtWpfABWdm1ZTU12mFIKjRDLnifUHU1nFwtq8yQA3ZGdLkN4TcA9FeoZbVHx9/jhj8VYeTZ1VPTrS516FGF3Ea0Qq7FQK5iVchNpbcCOSuwKq/c6p9+qvC3kqRSpaRzG+ckSQmT0N6pvZKkmUxQuWYS4UDOD5I2r0i//9266ys/6YdxRZOVvvWX9aN6m1ZVew72D80HLz3onz66eDT0/OoVcrvyQSCX2ArkSnGBnOO1sX85LjYJ5DZKVtujdT9xgZyrQu76xYXQ+dgZcqmE5KVVvvpMVTdrbbpD37IaUyFXGFAgFxcwuBZ77DRnhZy9ZXXEK+Titqxy4DmcXK8LrxWwMxqXOoz2z3sAGDW0rG6Pf5kw8uzqqakmFXJ9WerQasuqgg2rU5GWVUkqrx2TV62dv1h+VE+tPqXTa6flqfa8l/PLSifTkqTZdHyF3FQmKZ36kh74jRfog49/XGtbwc391YN6z/5flZ71FhWt9r1mFXJ2dZNzhlyyVn5W8qxAztpW6tpUGm5Z7d0MuXy6canDtYsz/rw+qfkMOZsxVvXikLK/t+OWOvTzc4g76BnEUgdXhZxdMu/aejlKopuK66KVIRgOrvdGcsRDYWBYRSuFR/3nPQCMGlpWt8dvgRh5rVbIhaqIevTzYL0Urohbb3GpQ0VBcDa9VSEXqmCq5lRZP+Kf/YvH/0JPrQWz5PbPBBtP7Qo5WRVymZRR+p7f1sp//2a9Zaai9+xe0r/Ys6z/r/xSvab4Hj2WvE5SfFujvWH16OLRUJjiqpAzqVoQt1kJ+kK337JqBXJNfkiHZshltp8hV/+a2vbN57Rnao8kKWmS2p3f7bxt9Bf2qXRSZsjDDft72w6bCyV7hlz/wrC4ip+BbFm1g/hKfanDGFXIxRxg0gY5nFzvDYp2gJ0RfbuN+s97ABg14fFEtKy68C8TRp5dldZyhdwOtayuFravkKtUPclYFXLp8JbVuvLKLf7pTz7+yfD8uOl9/unwDLlaK+yUNvX/pn9F+pN/pYeTVW1s/RL6hXxOPzl9hwrKqFiuqFr1QuFEvWV1tbiqp1ZrAWDKpHTj3I3OltWlnLXUIVUL4sKB3HYz5OzqpWaBnPUat9CyOp0JV8gljLRnLqe3P+vtOrp4VG971ts0n5133jZaIdfse2pYhFtWg8tDFXJ9bBeNC4kGv2W1vtRhfLasxi2lIJAbTq7XhZAA2BlsWQWAwQoFctWqvB4dg4+T4T/SBLaxVmixQq4PLautzJArVaoyCWupQ7qxZVWqBXLe3j+SMZ6+ePaLOrxw2P9YqEIuHZ4hd8Q8qfemf0E3bW1nPRepjMpe879UXr1FxfJCKLDJJBN+JdjDlx/2L79h/galk+nQ89t0tqzWZsitld2BnGsxgv1DutisZTUmdM0l4yrkwoHcnrmc0smEXnb9y/Sy618W+zjR51R73sM9P06KtKzGLXXoYxlOXMAwiC2rroUXZXvL6oiXJ8UdYNKyOpxcLXOEp8DOoGUVAAYrmTBKmFrBgOfVfhfnjyNho30kAkhaL7S2ZdV+71d3qEKulRly5aonJVwz5MJvR68yq+pGra206lX1kRMf8T8WVyF3MHNaH878hG5KBO2t5254buh+E6l1ZZc/qkK5Gg7krMePLnSQwgP569scl/J2hdyapGokkAtaR13tiplWW1ZjQtdMMuO8/kykQm7fvDu4c0kmTOgAeRBVXe0KhU7e4LesJhNGrjxoMDPkGivk7GrMUR+oH/f8R/3zGleueXG8VsDOaKiQoxoVAPqOttXm+JcJI28tNEOu30sdwhVxa61UyJVjKuQc1UMla9vqenndP31g5oB/2g7k9prHNWVq972prPTaX9OFIy9puN/M4t9qVQ/HBjZ2IHfT4k2SwmFKfalDNpnVzNZSCWOqUnJDK8Wr/vWmU8FzcwVb6RY379iViHnrNY6bITcVqZDbv9C45KEZ+6/oo1AhZ2JmyMUFrv3gaqXMpQY8Q85vWbVmyI18hZz7+UcrQzAcXOEbrxWwM6LVp1TIAUD/ZSJtqwgb7SMRQJH5YtlWlzrsUMtqC0sdStWqjKNCztVSaM+Rs+2bCSrkZjau+KdXt375fLi6X+/Z98vSnW/WuY1z/sfzyaBi7crUB7RRDJ6H/fj2hlW/Qi7VuNRBkhZzi8Hnkl/VaqnWumpkdGQ5WJzgCrbSKWvzTtNAzl0h55ohl0/lG76WB9oO5ILbu1pth41dBVAvkCtXqn4AZUz/q3BcbXi5gbSsNm5ZtSvkRv0ALXapAy2rQ8n1elEhB+yM6I/BUf8DDACMohSbVpviXyaMvPAMuT4vdShFWlZbWOpQrrhbVlPJREOI4ZV265rcDQ334bes3vthzX7wu/zLVxIJfTr3Yr3Z/Ixe+sIXSZLOb5z3P/7dt/xzedW0JKmSPqU/ePgD/sfqFVSe5zkDOddSB0mazwRtq9n8Rf/0bGZWt+5b8M8vTje2l3Y7Qy6VSCllwq95LplrmAvWTsuqFA4nXdthh42r+jO00CGV6PumWFfwMIj233CF3NaWVbtCbsRbmFjqMFrcW1Z5rYCdEH1vEX4DQP/Rstrc8Jd+ANtotUJuZ1pWK0rNflmZXZ9W6fKztXH5OapUvaYHWOWK52xZlWrBSbTq7tb55+ns5kn//FJuSXmTlv78J6TP/KJmEkZSbWPoamZKL/jRP9Jnqp7/w88O5J6x+xkqnn+Zstd8VJL0vgd/Uyb1L+WV5/1A7sz6Ga2UanPgZjOz2jO1R5KUs1oeC9ZfN+bSQYVcOndO9Xh0Pjuvt77wkL70xGXN59N6xW17G74WmVZbVq3QdSoS6mRTWZVLwcdzqVxDGNR+y6pdITf8gZxd/VkPmwe10KHOdeAzkBlySft9X/u//b02rhVytEEOJ2bIAf0TrRQe9SU+ADCKWh1RNKn4lwkjzfO80Ay5aFhjC7Ws9qhCbr1YUG7fHyqZf1LZvX8skzm3bdtqqVp1VshJjYsdJOn2peeHzu/LL0t3v1b6zC9KkmaqnszWp7NWLariVUM/+M6tBy2re2euUfHC81UpXCNJ2qysK7vnTyQFoU1oftzCTX5llR2m2BVy06mgCk6Zs/7Jucycrt81rT992wv0O9//HM04Fm6kW13qULQXd0QCuUjbai7VWCHXdiBntdKORiAXnK63Y4fnAw6gMs1x4DOIllVnhVx1jGbIsdRhpLgCVFdIB6B70T9MsNkPAPrPnmNNINeI3wIx0jZKFX9mVjaVaHpwHa6Q683jr5sTMslatZsxnrK7/rKhwi2qVAnPkMtbCwiyjuDkhpmbQltV95++Xzr5af984ugrNJ2Z9s/XZ7hJUqVa0aXCJf/8vpndklIqPP1a/7L03FeVnDrh/7B0LXSQooFc8AXMJ4NArpo645+ez843fC5RnSx1iM50awjkHC2r3VTI5dPDX0hsf2/Xw2a7itEV9O40VyCUHZYtq6GW1dE+QDPGuBcFMENuKLlbVgfwRIAJwJZVABg8+3cfWlYb8S8TRpo9s81VgWVLxGyi7EYx/UDofGr+Hj1y6Ymmt2loWbUr5BwVROlUUi89GGxKvXZjZeuUkV7yLunb368Za9OqHchd3LyoqlcLZhazi5pK18KrysYhla7cGTzuno+onju55sdJ4VDHrpDLJYLgrWSsQC7TQiBnVaI1nyFnVUFmmlfI5VN5JRNGN10zI0k6tDytxan0ts/Flhm1ltXQfMTa/we5YVVyVyIMomXVvWXVblkd/X8G3VVXA3gi2JZrnAEVcsDOYMsqAAweLavN8VsgRprdHtpsfpwUPkDtRctqpepJ+YdDlxlT1Qcf+p9Nb1eqVCVjtaymm7esZlTUd516WAdKZe0pl/W61VUptyB9x+9JL/oRKZHQTHrGv/5KccU/bc+P2z21W8YYP2wqnP2HyiRqyw6SudNaTf+1pHCFnB3I2WGKXX2V0Zx/ump9XnPZ4PI4rc6Q22hSIZdLhRc25FO1arj/+pa79M5vPqbf+M5nt73QID1iSx1cG4QLpUFXyDlaVgfwPNxbVoP3/zgcoLkWO7AoYDi5AuBRr9IEhhVbVgFg8NK0rDbFv0wYaXaFXLMNq1LvlzqcX7+qRL6xGu4vT/2pzq6fddyiplz1Qi2rdoVctJLpWnNOt3/sTdr75d/X/3rylD72xCnduOtm6a1/Kd30Tf715jJB+GUHcuc2gvlxu3O7JQXhjFee1/OX3+h//HTiQ7qwcUEnr5z0LzuycMQ/HbdlNeUF1Xk2+znFCZcwx/+AtivkWpkhJ0mHlmf0L15yRIeXZ9QuO6SZHoFALumo/hx4hZwjZBhEuBmqkKs0VsiNwwEac8lGh7tCjkAO2AkNFXK81wCg79K0rDbFb+wYaetNWhmjer3U4XOn/lbG1A7sK5v7VNk4KEkqeyX9j6/9j9jblSqRpQ6hCrngc3hB4iv6SOZdmr7wVUmSkZS8483S9/y5tHRj6D5nMkHotFoMWlbtCrnlqWVJ4XDm9tlvVbVYmwFXNqv6sU//mMpe7Wt6YOZA6H5DM+SsCjlTdQdyLc2Qa/EvJuuFNmbIRSrmOhHesjr8M+QSjurPgW9ZdS51GMQMucYKuXHasiq5v9Zj8GmNJVdQTYUcsDMaZsiNwR9gAGDU0LLaHP8yYaSthbZv9rdC7vOnPxfc39phFc4Hc95+78Hf06XNS66bNcyQq7dYSvXqNU//PPnH+h/p/6RFsxWuJdLSK39Oeu17pcyUouyWVXuGXKhlNV+rkLMDuc1iUoWzr/TPf/bpz/qn7YUOwXOrKVgVcqbqroRrf6lD/GsSXuqwTYVcsvtAzv4ajULLqmupQ3jL6nBUyOUGsu3VsWU1tNRh9P8ZdFV9RLcLYji4quF4rYCdwZZVABg8uwCjSCDXYPSPRDDR1gqtV8i52vq6cc+5v/NPl9eOqLJ6syqbtW2oG+UN3X3v3c7bbZZLMomt5+2ZUIA0lyjoV9O/oB9J/64SZquaZ2qP9N1/Jn399zcORNkyay11uFq86p8+t261rObDLauStLJZVnnlGSqvhSvuJOmmhXAgFzdDrlhMyas2Lk1opWW1lRlylaqnDSsAzEeqrOJmyHVjYSrjn949k21yzeHgWlhSrARfs+FZ6jCIGXLhLaue543dDDl3hdzof17jiBlyQP80tqxy2AMA/Zaxftcu07LagH+ZMNLsQG67CrmEo4qoUxc2Lujx1dpCB89LqLJ+oySjolUl9/773x8Kx/znXNrwTydNNlg4cOmkfvzM2/XK5Of9j3+uelxPvvGj0sGvb/p87EAutmU139iyulooSTIqnHm1ak2xAXuhgxRpWbUCso1SVV55uuE5tV8h5w7k7DBuKpNs+Iv3TrSsfu/zb9SxPbN69R379Q03LnV9fzst3I5d++Y+RIkAACAASURBVP/AW1ZdFXKDaFmNhJV2GJdMmLYXfgwj19eauWTDiRlyQP9E31r2ZncAQH/Y3Si0rDYikMNIs1sZt13qEDow7+5xP386CM0qG9dJXi0UKq/cpvnUAUm11tEP3P+Bhtuul9aC56St8OiRv5J+/cW6tviI/7HfKn+zvqP4b5Wc3bvt82mlZXVXfpekSCC3WQs0q4X9uiEThImSI5CzW12tkGytUJZXaZwjN59pJZALXpNi2R2Srm9TBbkTLat3HlzQx/7VC/VLb37mSLSTuVpW7SrG7ACCMFcraHYgFXLhGXLhdtXhf21b4V7qMB6f27hhhhzQP4noDDkq5ACg79iy2hz/MmGk2ds3p7LbLHWwZ8h1WSH3uafD8+OsR9HN+df65+6+926VKqXQbddL6/7ppLLS3/yadPfrpI3azLmCl9I7S2/Vvy//U5WVamnmiV0hF7dl1a+QS9oVcsHX71mzb9ZsunY/S7klXTd3XegxsqEKueCH6XqxIq/cuMm0V0sdwvPjGkPXnaiQGzWuduzCoCvkHN+30XbjvjyPyOzIUtVe6DAe/wS6DjIJ5IaTs714TL4PgWHT0LI6BiMKAGDU2D972bLaiN8CMdLs7ZvbVsjZmyi7nCH3N0//jX+6snYk9LFrzHP88Oty4bJOXj0Z+vh6OQjklstXpI/+qOTVPo+rqV369uJP6PcqL/av02kg53meLmxc8C8PZsgFocjVzSCQm8ss6r3f9F694egb9PMv/nmlEuGvp10hVyhX5G2FmmvFsqqOQG4u25sZcmvbbNLNpsKBXC9myI0auwig/r09lEsdBtGymmxSITcmB2eug8xoZQiGg7O9mNcK2BENFXKE3wDQd2laVptqnmAAQ267sMbWq6UOT648qadWn5IkedW0KhsHlUoYfzbVRkm6bu46vzrtwuYF3aRgQcJGOZght7t8ObjjA8/Wf9v1bt3z+aDlVGptCLGrZXW1tKrNyqakWkg1na7NeXO1rEq1cOyO5Tt0x/IdzsdIJRP+51n1an/hyKSM1gsVeZVwIJdL5hoq11xamSHXbMNq/bFskxjIhVtWa/+3A7nsAAK5ZOT7NpkwA6lIC1fIVVW2vs/GpX3JWXVFhdxQcr0utKwCO6NxqQPvNQDoN3t+J4Fco/E4GsHEClXI9WmpQ6hddf1GSSntmgm2cq4XKn41mqRQlZokpa98zT89VU9P7vjH0nf9aW2jakS7FXL1pQ52u+ru/G5/eH14qYMVyLUQ2oQWO5RrX/u1YrmhZbWV6jgpMkMupoQ5NCfQ8RrTsho+6Ki3Y9trxQdRIRet2soN4DlIkRlyFU+lMduwKjGXbJS4/sBCeArsDCrkAGDw7D/I2wUDqOFfJoy01R5VyD1y5RF94rFPqFApbPuYdiBX3mpXXZ4NQqG1Ylm7crv88/ZiBd3zPt380C8Gz7nqSa/4Gem1vyqlc6F20rpWqorsCrn6ZldXu6oUrpZa2Qzm27US2mQdix1cM+RamR8nRSrkYn5AD2Kpw6gJbVmtz5ArDbZCLnrgM4h2VSlcBVf1vHCF3JgEcq6fEaOwjGQSJR3fc+PyfQgMm+iPRt5rANB/9oiicpdjo8YRLasYaXZYM7NNhZyrikiqBWZv+sibtFnZ1Hff9t16x13viL0Pz/P0udP2QodaILd7xgrkCuVwhdzmBalakT7+bumzv6zNmWn/Y0/Pfb30nH/mn3dtoWyl0iVUIbfVsnpuPVwhV9erCrl64FPbshqpkMu0ViGXaWGpw9p2Sx1SVMjZ3yJBhVzwdRvIUofI9+2gArnollV7mGwr7eCjwLlllblkQ8k5Q47wFNgR0Qq5cfmZDwCjxP49Na4AY5LxLxNG2nZhjS3Usmql8/ecvceftfa/n/jfTe/jocsP6eLmRUlSxsyqWtgrKRzIrRcr2pUPKuQurD4tvf/N0md/WZK0YT2PYvZg6P6jwYkxrR2suVpW7cq8uAo5O5xoJbSxA8NCuUmFXKaDCrmYQG5jmypIZshFZsgN6VKHnCNs7vfzqFQ9lavjVyHnmoVHhdxwYqkD0D/R35/G5Wc+AIySVo73JhmBHEbauhXWTGfbaVkNLj+1eso//djVx1SqlBTHblddStys+lvIblldjVbIPfxx6aGP+ecfnz7sn84mw+FRtEIunUj4s9+ayafySpra579Z2VSpUgoFcvWtr1J88NZShZzVUrtZqsrzvNqW1UpnLav2wWncGuy1dmfITWLL6jZLHQYSyEVnyA1DhVwlvGV1EEsmdoJrFh4hz3By/YGFCjlgZzTMkOO9BgB9FwrkaFltMB5HI5hY9lKHdirk7JbV+sZUSSp7ZT2+8njsffzN03/jn571bvFPN1TI2TPkKsFWVT3/HfrK0ov8s5lEODyKzpBr9a+5xhjNZIJQbKW00rDUwX+MmGCklTljdmC4Wapoo1SR50mq5OV5wcc6aVktxm1Ztdpq847nHm1RncyW1cb5iINe6hCt2hrYDLlk+GtTCs2QG49/Al2fB5Ugw8kYQ9UO0Cf2ey2VMC39gRMA0FtpWlabGo+jEUystQ4r5OyW1adXnw5d78TlE87bl6tl/d3pv/PP5yvH/NOhpQ6FsnY9+hn//IVkUkpmpNf+mvTyf6eit+l/rKFCLhKctPPXXHuxw2pxNbZltVcVcoVyVWt+IJqQsarkOlnqUG5hhpzrNc4kM6HzkxjIhao/t8LmQtle6tD/MCwaOgxDy2q5Wo3MkBuPgzPX5xGtDMHwiP5cTzLXCtgR9s9Bgm8AGAxaVpvjt0CMtLU2KuTsHCpUIbf2VOh6J664A7lHrzyq9fK6JOmaqWtUKQZVcEtTGRkjGVX1Q3q/dv3pv/Y/djGZUOUtfyTd+WZJUrESBHK5bQK5dlrq7DlyK6WVcMvqlNWyGhO8ZZLbhza5SIWc3TKc9ILHbz2Q275ldX2bOYHRFtWp1FRLjz1O7ON5zxHIDaJCLtpGmRtAKCiFw45KdTy3rLrbIAfwRNCShkCO8BTYEfa/jSx0AIDBsI9nizHHe5OMf50wsurzy+qmHQP/ba6lDp7nhWbISfEVcvdfvN8/fcvSLdooBQf2U9mkdmfK+pX0L+r/Sv2x0pLmt7ZcVo3R5T3H/euWvKCFNR8Jj6LtpO0EBnaF3EpxpaWlDrZ2t6xulqqhQDSlIISby7bWshr+AR3TsrpNFWR0hlz0/CRIulpW7UBuAOlMtOont837c6ekoltWq+M3Q87VskqF3PCKBqjMkAN2xlQm5b+/5vLpAT8bAJhMmRY6oibZeByNYCLVlgrUTmdTiW3nQblCi6vFq1orrYWu10ogd3zX8dD2z7niWd2d+Em9Mvl5/7LdVvWbHY4Vq1aFXGQjaDQ4cW1PjGNXyF3avKTLhcuSpIRJaDG7GDxGF4GcHeZFK+QWqs+WVJsf99x9z23pOUdLmD2v8a8mduiXTzuWOqSCAC6TyCiZGEzwM0h2+FL1akFzMdSyOrkVconollW7Qm5MghDXUodx+dzGUTQIHpdKTWDYzGRT+sEXHdLybFY/9NIjg346ADCRUqGOKAK5qOY9fsAQC8+P2/5bOelY6hCtjpOkk1dPqlwtK5UI3+cDFx/wTx9fPK73b7VS3mYe0Q0feodS3mn/45dv/17tmrqsE2f+VpJ0YeOC/7GyNUNuKhmtkAsfqLXTamgHcievnvRP78rtCoVU8S2r7VXIFcrV0Hy3fckX6r+/7k1ayi2FFkw0k0zUBpxXqp48rxaYRA9ON0rNK+TsltVJnB8nhUMnqRbKDXzLasNSh2GYIeeFWqPHZqmDI7in6mp4USEH9M87v/m4/vU/OMZCBwAYkHABBi2rUeNxNIKJFN6wun31TcKx1MEVyJWrZT2x8kToMs/zdN/F+/zztQq5il6cuEcfzLxHqbVaGFf2EnpX6Xv02Ne9W7umgjbRC5vuQC6f3pmlDo9eedQ/bber1h7D/bVqv2W1EtqAOpVJ6rq561oO4+q2myO33ZzApdySZtO1MPKG+RvaeuxxYh/UVz1v8FtWoxVyA9qymoxWyFWDr4ursmwUuT6PaEiL4RH9uU41I7CzCOMAYHBY6tAcgRxGVnh+XGcVck+tPuW8brRt9fTaaV0tXpVUq0TbP71fryr9uX4z/Z81ZQqSpFUzo39a+lG9r/JyrRXL2pULlj7YLauhQC46Qy4VnSHXYYXclZP+6WggFxfOtNLWaFfwbZYr4Q2oLbwGLtvNkbPbYl3Bay6V08+96Of0pmNv0ruf8+6OnsM4iLZkD7plNRoyDEWFXKWqsl0hNyZDvl0tjywKGF7Rn+vM+wMAAOMqTctqU7SsYmSFghpHK2NUeJZU7f9Prz3tX5ZOpFWqliTVArmXX/9y/2Oh6rjF49Kn/qN+KvHr/mXe/EH939M/qb9+pFbxtl6ohIKwUMuqCv7pmXQ0kItuWW39QC2uZbUhkIsJ+VqbIRde6tDua+B83G3+arLeQuj3vAPP0/MOPK+jxx8X9jF91fNUKAdft0FUyEXb8Aa3ZTUcVJbGcMuqK1ikQm54NVTIjcn3IQAAQBQtq82NR3kAJtJqob3qrKSjZdWukLtrz13+6RNXwhVy9vy4Y5eekvn0z/rnv+bdKPN9n9CVmUP+ZWvFsnblrQq5zaBCrmLPkGvYstqbltWNcrDJtbFltZsZcsF1CuVKqJ20FxVy2wVynYZ+kyDcsjr4LavRwfX5gW1ZtTY7VT2V7S2rY1Ihx1KH0RINq3mtAADAuNquG2rSjcfRCCaSPb/MNew/ys4HXEsdXnDgBf7pRy4/ErqtXSF381Nf9U//ZeUO/UDyp6TZvaFAai2mQs7zPFWtCrmpTCSQS3besho3u215ajn8GF1sWbWrnArRCrlOA7mUVcZcds2Qa96yippmLavDUCGXHYIZclUvsmV1TCqTXD8nWBQwvKKvV3JMgmEAAICoTCo8PgZh/BaIkdXu/LLtljq84NogkHv0yqOqVIP7f+DCvf7pY4WiJOl3yy/W95V+WN5WEGZvel2PmSFXqBQkU3tsr5pSLpUOPceGLasdzpCztTpDrpOlDqEKuQ6r15r91aRS9VTYCpaMGVzb4yiwWxSr1cEvdYhWbeUG8Byk5ltWo1V8o8pVYcVcsuHFUgcAADAp7G4VWlYbjcfRCCZSu/PLoksdrhavaqW0IknKJXO6Ye4GLedr1WTFalFPrj4pSbry5N/q1PoZSVLa83SoVNL5Z79DP1r+fpWV8qu27EAqWiF3cfNi7TmX1/3LvGqm4UAsGsC1U8FT3zQaVf+c/MdwBCPGtHZQaFfXbZYqodfADiTbkU7Et6yGXuN0krlYTdhfmooXBJlS/GbdnRSt+hnUltVEwvjz9TwvHPqOSxDiChapkBte0deGn2sAAGBcsWW1OQI5jKx255dFh7s/vRosdNg3s0/GGB1aCObAnbh8Qjr513rgg2/0LztSLCn9ml/RE7e/TVLt/uqBnN2yuVYsazG3KLN1nUubl1SqlrReCgI5VTON2/YSJhTKtbMFMq5l1Z5lJ7nDmUwyIdNCRY0dqhTKVa32oJ001LLaEMgFr3G+w5bYSRFtzSwMeMtqQ4XcgAI5KRy8bZaC76l2WsKHmXPLKiHP0Iq+N8YlGAYAAIjKNDnWA4EcRli788tCLaueF1rosH9mvyTpyMIR/7ITD3xYuvu1uk8l/7KbD75AeuY/0UYoKKoFDTOhCrmyUomUFnOLkiRPni5tXopUyGWdLal2eNLpllVbKy2rrbY02ksdahVyvV7qEC5jXmtzTuAkC7dkD36pQzQQyg8wkEuGAjn76zIeQYhrOUWSltWhFX1vEJ4CAIBxRctqcwRyGFmtzC/73NOf01898VeqetXGCrm1oELuwPQBSdKheatC7sEPS5WiHshk/MuO3fBSSZHNn1tBlB0K1j++lFvyL7uwcSFcIedlnJUt9hy5dip4XIHcTHpG+VQ+dJkrfGu1gio8Q64aXrjQgxlyzSrkOl0aMSnsQK5QtqrAEmYgLXHR6s5cenD/3NjPZRwr5FyBDiHP8Iq+N6iQAwAA4yqdomW1GY5wMbLC1VON38qfevxT+pd/8S8lST/9/J9W2nyd/7FKVaEKuX0z+yRJhxcO+5c9kq4tXLhvakZSLc2/edfNkqT1UmOFnB0K1ls5d+d36+HLD0uqLXZIJoLreNWsswLOrmZKt3Gglk1mlU6kVaoGFX3R6rjo/Te7zPkY9gy5ckVl668cnVbIZVoM5KbZsNqUHcBsWN+fg1joIA1Xy2oytmV1PIIQ1+fBXLLhFX29CE8BAMC4so8JCOQajUd5ACbSWjF+fpnnefr1r/y6f/7jj328YcaWvWH1wMwByfN0+Ct/5F/2SDqljf3P1KNbgYaR0dHFo5KkjciyAalxy6oUnt92YfOCNkobwXOsZpwz4rJWcNFuYBCtklueWm64TnSTq9ROy6o1Q65UDb0GnW9Zjf8hbd9/nkCuKftbyW6pHlQgFw0ZBlsh5w7kXK2eo8i51IGW1aEVrYgjkAMAAOMq02Q8EQjkMMKazS/74tkv6msXvuaf//K5Lythgh8AlWo4kNs/tU/66L/Rwl//onZttfsVEgn91UveropXO3/d3HWaTk9LCrfLOpc6bH18dy6oUDu/cT40Q6621MHRshqaIdfeWzQayNmPX+eskOtkhly5ovVC9y2l9udYLId/SG/0YEbdpLADGLtCbhALHaTG791BbHqtS8TMkBubCjlaVkcKM+QAAMCkSLFltSkCOYysZvPL7r737tD5i5sXdaEQtKhWPU+n1qwKub/+FelzvyZJOlwKWj7/9IlP+qePLR7zT2+UGrd/2hVi9Qo5u2U0OkOutaUO7b1FZ9LhTau7p3obyNmhSkOFXKeBXJO5Ar2YUTcp7Blym0PQstpYITckW1bL4zdDzlkhR8gztKLfd+1s0wYAABgltKw2x2+BGFlxFXJPXH1Cn3r8Uw3XP7kaVMwVq+u6UrgiScrIaOkrH/Q/dmjmoH/6/zz1f/zT9flxtcdubJeddlTIhVpWNy44KuRcgZzVstrmQfVMJhLIOWbIJRKmIZRreYZcky2rnbaUtjpDLtqWjDC7CmyjONgNq1LjDLlBthzHzZBrZ0bjMHNV+o1J1jiWoj/Xx+TbEAAAoEE6smXV82hbtfErO0ZW3Pyy993/PnlbSxiSJrj8kZUgkNv0zvun9xeLwRvhzn+iw3d9n/+xcjV4DLtCzhUU2TPk1mJmyNmBnBfXstrhllVJmk1HZsjlG2fISY1VU53MkLu6WVKlWvs6Z5KJjiuxWp0hR8tqc3Etq5kBtYomo1tWB1SpJ0VnyNktq+PxT6CrwirBDLmhZQfEyYSR4bUCAABjKpEwod99ylUCOdt4HI3sAGPMnDHmPxhj7jPGrBtjLhhjPmmMeUOP7n/BGPMTxpi/M8Zc2nqMR4wxf2iM+a5ePMa4C7UzboU1V4tX9aGHPuRf/r3P+F7/9ImVv/dPF+1Arrx1P3e8WfrWX9LhxZucj2dXyG04KsPsCq76bLVduSCQO79xPrLUIescKh/astrlUgdXhZzkCuRaC21y1vXsoZzdtJOGZshV4mfIdTqjblIkYqrABtWyaodgqYQZaPgVWyE3JjPkXJ8HLavDKxUJ5AAAAMYZbavxCOQcjDHXSvqSpHdJOi6pImlO0ksl/Z4x5le7vP8XSnpA0k9JuktSXlJJ0o2SXifpx7u5/0lhLxSoV6f94YN/6FehHVk4ou97xvcplah97PTG41JyTZJ0S+lj/m33l8vSra+TvvWXpURChxcONzzW7vzuULjlqpDLphL+wVWxUlWxXA3PkNt0taxuUyHX5myhVlpW68/V1mpbYzppnO1V3VSv2YFcqRydIUfLaqvs18UOMge11MH+3h7k/Dgp/D4qWN9j4zK7yxV2EvQML/v1ancsAQAAwKgJH+9RIWcbj6ORHjK13pHfVy0cOynpGz3Pm5U0K+lHJFUl/TNjzPd3eP/PkvRnkq6R9GFJd3mel/M8b17SgqRXSPqdbj+Pced5XqidcSqTVLla1vvuf59/2VtueYvyqbxuWbrFvyyZf0w/kPyI9lXv8S/bv3hE+rbfkJK1UGkpt6TF7GLo8Y4tHQudD81OS9duZ4zRdCa82GEhu6CEqb3NrhSu6HLhcvA5VDPOg7HQDLl2K+Q6bFltNbQxxji3ZU53USGXabLUITSrj6UOTdkBzDBsWbXDLns77yDYX5tCyV7qMB5hCFtWRwsVcgAAYJKEZoZXqZCzEcg1eo2kb1AteHud53mfkSTP8zY9z/tZSb+4db2fMsZk2rljY0xS0m9Jmpb0Pkmv9Tzvi/WPe553xfO8j3me9+4efB5jrVCuqt5+nkkllE4m9InHPqHTa6cl1UK1Vx16lSTpzmvu9G93fOoz+rH0+3UqFVR07f+GH5KS6dD9H1o4FDp/89LNofMbpcalDlJ0jlxFyURSS7kl/7InVp70TydN1jk7yA5Q2h3Ib1fIpRIpzWfnndeL3m87rXuucKWbdtJmJcxxizvQKBE3Q25AraL2+2I2l25yzZ1nB2+bVoXc+LSsOirkmEs2tJKRdm4AAIBxlqJlNRaBXKPv2Pr/JzzP+5Lj4z8nyZO0V7UW1nZ8i6TbJW1IepvHipGO2fPj6lVpd997t3/Zm469SdlkVpL0zGue6V++NHWvJOmUVeV1YP76hvs/snAkdL5ZhZwdPITnyG0tdrDmyD258oR/OunlGj8xSfP5ILyYzbUXQtkz5Hbnd8cOC+90qYPkbj/spkLOrqSKzpCzK+QGuaVzFMTNSRvUDLn9C3m94ta9SieNvvO5je+xfor72oxPy2rj+zxB0DO07ACVCjkAADDuaFmNR8lJo5ds/f9jrg96nveUMeZrkm5TLZD7aBv3XQ/7PuZ53sXOnyLWI8P+v3T2S/rK+a9IktKJtN547I3+x+0KuXuzaRUlPZEKihv3Te9ruP9D89tUyDmWOkjhCrnVrUBud363Hrj0QO2y0qr/8aTJOj+3N37dQf35vWc0k03pFbftdV4njt2yGteuKjlmyHUZyHVTIWc/djm6ZbVAhVyr7OP6YQjkJOnX3nKXNoqVgYepqbhAblwq5BzBIpVXwyu6ZRUAAGCc0bIajyNcizHmGkn1cqavNbnqvaoFcrc0uY7Lc7f+f48x5oCkfyfpVZJ2Szor6S8l/T+e5321zfudOPb8uOlsUh89GeSi33LoW0LLDHZn5nXQS+kJU1YxYfTp3Jyubh2IpxIpLU81Blf2Yoep1JQOzh4MfTwaCPrPxTpdv86ufFAhZ0sZd4Xc4eUZfeqHXxRb3dbMsaVjSpiEql5Vty/fHnu9hgq5ZOuBiWsm2XQXgUvTllUrPGGGXHOhltUhWOpQN+gwTgqHHvam9UG18/Za0lUhR8vq0ApvIB6P70EAAIA4oQo5WlZDCOTC7FKpU02uV/9YY2lVDGNMTtK1W2cXVdviultSQbUW1msl/RNJbzLGfKfneR9o8X6/EPOh460+t1EUalnNpnTfhfv88y+9LtJJ/Oc/oWeuXtYTs7X5au+ZeqakRyVJ+6f3+0sXbLftvk2L2UVdKlzSC699YcN14lpW7dbN+nOMC+SScgdykjoK4yTp2tlr9d6Xv1cPXXpIr7vpdbHXy0QWM7RTRZV1Vchle7RlNdqyWnDP6kOjuKUOg6yQGxZxoYdrO+koSrPUYaTY34+8TgAAYNyFZsjRshpCIBc2bZ3eaHK99a3/zzS5TtSCdfptklYlfbukP/A8r2yMuU3Sb6q2UOK3jDFf9DzvwTbuf6LYrYxTmYTfEipJx5esLPJLvyN97r26c3ZaH94K5M7PnFL9R8L+mf3O+59OT+vuV96tL5/7sl5y8CUNH98ouoMiu1quXsVnz5CzpRP5mM+uO8/b/zw9b//zml6nq5bVnlfI2TPkWOrQqfBSh+Dr2E7147iKCz3Gpa3TFSwS9Awv+5fScfkeBAAAiNPseG/SjcURrjHm3ZI63Uz6nzzPe1cvn0+MROT0OzzP+936BZ7n/b0x5jWSHlYt6Hu7pH++3Z16nneX6/KtyrlndfWMh5g97D+ZvqS10pokaSG7oD1Te2ofOP330kfeLkl65mbRv75JFvzTcYGcJF0/d72un2scRu95XriV0m5ZtbesboWGdvusLRMzQ64fogFcO22NPZ8hFxryGZkhFxN8opE9xH+zSIWcLS70cG0nHUXOpQ60rA4tOyxl+QYAABh39vFedGb4pBuPo5Ha55Hs4r+6Net0s/Klqa3/rza5TpR93SuSfjt6Bc/zzkj6na2zL2vjvieOXSFXST/lnz62dKzW7ul50p+9U6rUwrdDC4c1k2osaNw/HR/IxSmUq6rvx82kEqGDK7tSrB4aulpWPc8olRxcIJdNRmfItdGy6qqQ62K+WzrVZIacXSHXRVvsJLAzGVpWw+JCD5Y6YBDCM+R4nQAAwHgLtaxWaFm1jcWRmud5P+l5nunwv39j3ZU9N65ZUlP/2NNtPM0VBaHcCc/zKjHXq/deHoz5OBSukNs0T/injy9utat+7UPS45+pnU6klHjj/9TNS41LDppVyMU/tnt+nBSepbZar5DLOSrkqhnnQXS/NCx1GGCFXNwMuVKlquJWxZwxg19OMOySMZtE+bo1qZAbk4H6zgo5gp6hlWLLKgAAmCAsdYg3HkcjPeJ53jlJ57fO3trkqvXtqve2cd+emm9ubbhJG9edOGtWKLamIJA7tnRMKq5LH7c6mL/hB6XlY7p11x0N93Ng5kDbj22HgVORcCpUIddkqYNXzQ60Xa6rGXLpHlfIxcwUiM6P63TRxaSwvz7rQ7RldRjEzpAbkwo5V+BI0DO8ktbPPCrkAADAuCOQi8eRWqO/2Pr/N7k+aIw5oCCs+2Sb9/2Jrf8fNsbEJRj1jQQn27zviWJvNIe30gAAIABJREFU37xcOemfPr50XPrML0lXtkK6qd3SC98pSXrG7sZArpMKuQ0r7MhHKuRCM+S2rjefnVfKRCrIqhmlBxgGNFTItREO7ugMOesH9EaTSkQ0Shp3hRwtq/Ghx7gEcsaYhs8xSYA9tNJUyAEAgAmSSdGyGocjtUb1GW7/wBjTmOBI75BkVGtX/QvHx5t5n6SqpHlJ3x39oDFmj6R/vHX2z9q874lSbwdVYl1rlVpRYyaR0Q1KS//nvwRXfNlPSPnagttbl26T51mVCSal5fxy248dblkNB1F2pVi9ki5hElrKL4Wu51Wzzs2I/dJNy6qr4mqmi/lucX8xYaFDe+wD+9AMuTFZXNCNZExr6ri0rEqN4SJBz/BKEsgBAIAJkkpQIRdnfI5GeuePJX1Ota/Nh4wxz5EkY0zWGPPDqm0/laR/53leMXpjY8xJY4xnjPnt6Mc8z7tP0n/bOvufjTFvNKZWOmWMuVXSH0malnRJ0n+J3h6BetiVzAVj/I4sHlH6k++Ryhu1C/Y+Q3rmW/yPT2fyqm4GFXF7p/cqmWg/6FlvUiFnB3RrVhXfrly4bdXz0gOtkMumws+7+xly3bSsWn8xKQd/MVkvxAefaJSImSFHhdz4V8hJjeEiQc/wsr/veJ0AAMC4ixtRBImj3AjP8zxjzBsk/W9JN0r6rDFmVVJOwdfr1zzP+40OH+Jtkg5Leqmk35W0aYwpqFY1J9U2sL7e87xTMbeHgnbQRC74Mh1PL0l/9zvBlV7xnyQrcEskjCrr1yuZf1JSZ/PjJGmjFF+5ZVeK2Ztgd+cjix2q2dBfCvqt10sdutmAmk7FzZALvs7dzKibFPZxfdWqBCeQk5Ixwdsg5zj2WjRcJOcZXvbP/kH+OwAAANAPdstqmZbVEH4TdPA870lJd0r6aUn3qxbErajWovpGz/P+WRf3vanafLoflPRZSQXVwr6HJf2SpGd4ntduK+zEqc+QS2aDCrljT3whuMKtr5Nu+MbQbZLGqLx21D9/866bO3vsZltWM40tq1LjYgdv0DPkIkFEto1gwtWy2lWFXEwJ88pm8PXLUyG3rbiZYdFqyEkUWyE3RqmV3QKfMGIJyhBjyyoAAJgktKzG4yg3hud5VyW9a+u/dm53QwvXqUr6r1v/oQP1+WKhCrlzJ2snUjnpm36q4TbJhFFl7ag2T79aqexlfc9t39PRY4daVtPRGXLB+VWrZbWxQi4z0OqcrmbIuSrkugjM0jF/MXn84rp/+sBCruP7nxSJmAN7KuTiQ49xCkMIeUZH1tpUzfsTAACMO7asxiOQw0haL1YkU1Yie9a/7Ghxa6Tf894mLVzXcJuEMZKMSpe+UdWE0WJusaPHbrb90w7k7OCuYYbcgJc6RKvc2mpZdVXIddFSGvcD+tHza/7pG3dPd3z/kyKuQo6lDu5KuHTSjFUVGXPJRsdzDu3Svvmczq4U9Oo72t/0DQAAMErSbFmNRSCHkbRaKCuROSNjagHOtaWyZjxPkpGe7a58sw/KK9XOfxA0a1mdts6vNamQ86oZpQd40NzLGXKphOkq9MnEDPk8ecEO5GY6vv9JETeKigoc95bVcZofJ4Vbv+PCWQyHqUxKf/XOl+jKRknLs9lBPx0AAIAdlaFCLtZ4HZFgYqwXKkpYG1aP16vjrv9GaW6f8zbRlr5qh6HcRtGebRa/ZXW9WJHn1R4jOkNO1exANzw2VMi1EU5EA7mpTLKrSqO4CrlHztmB3FTH9z8pErEz5Pgx76qQG6f5cVK4Qi6ufRnDI5NKEMYBAICJwAy5eBypYSStFctKWvPjjtUDudte1/R2ditXxesskGtWIZdJJfxlDeWqp0K59gPHtdRhsC2rjc+79duGr9vNhlVJoeUW9RLmzVJFp65sSKoNqD+4RCC3nbg2RQI599dm3Crk7F90aFkFAADAsLBbVosEciHjdUSCieB5ntaLFSWsDavHCyXJJKSbX9P0tnYrV6dtq+sla6mDY5mBa45c4wy58WlZ7WbDqiSlrccubQWYj11YVz0vvXZxik2hLYirkKNlNaZCboAVqjvBDrbHrfoPAAAAo8vuxiozQy6EIzWMnEK5qkq1GqqQO14sSje+UJpZbnpbe5RUtcMKudBSh202jtbnyM1l5pROpK0Hzw7VltVssvXAK5fubYWca4bco+dX/ctY6NAaArl4rhbOVNzQvRFlV9zGfS8AAAAA/Wb/sZiW1bDxOiLBRFgvVmTSl2SSBUnSfKWiPZWKdOu3bXvbnlTIWTPkXNVh9mVrW9c1xoTaVgffsjpEFXKOGXKPnl/3LyOQa03ctxNbVuO3rI4T+3OkZRUAAADDItQRRSAXwpEaRs5aoRypjivJJFLSza/e9rZ2pUy1w58F9gy56FIHKVwxtlYIrntN/prgStXcQAOBrlpWI+2jM11WyCUTRvWXperVglIq5NoXN8ifCjl3QDXIQHwnpKmQAwAAwBCyf08tlmlZtY3XEQkmwlqxHJofd6xYlA69RJpa2va2vV/q4Johl7SuG1TTvf7o62WUUGVzryobBwfaMmdXTSUTpq2KmmykZdX1NWhXtEru0fP2hlUCuVbQshpv0rasUiEHAACAYRGaIddpVcyY6v5IGuiztUJFiej8uNu2b1eVetWyGr9ltXaZu0Lu2276Nn31wYP6rU+flpQY6FB5O6Rpt6UxWiFnB5AdP59kwt9IWySQ60gyLpAbs0qwTiQdX4Nx3rI6bmEjAAAARpd93EvLath4HZFgIqwXy8rmnvTPHytLOv6qlm4balnteKlDUPXmbFm1Z8gVyqGPJb1p1d92gwxK7Dlw0Yq37exIhZwVEF5cLer8alFSLTjcv5Dv+v4ngatlNZNKyNC+OHFbVuPalwEAAIB+o2U1HoEcRs65tcuqplckSWnP043XvUjKzbd0235UyNkz5OyWVUkqW485yEBg13RGd12/KEn65lv2tnXb6EKI6S6XOkjhMOHBMyv+6euXpmi/a5GrQi47ZlVgnXJ9D6XHeMtqXLUkAAAA0G+0rMajZRUj59GrD/mnjxRLSj/r9S3fNjRDrsNAbsMO5NKuGXJWy6p1XanWjlk3yKHyxhh94K3P0YNnVnTLvrm2b5tNBS2mU10udZDC7XYPnWWhQydcuSXz42omoULO/hypkAMAAMCwoGU1HkdrGDlnL/6Nf/posSwde0XLt7WLYjppWfU8T+ul5ltWp5q0rJatH0DpAR80p5MJ3bp/vqOWRrvltRcVcnZw9MDpoELuxmUCuVbFtaxiMras2oHcmH1qAAAAGGGhBX60rIbwaztGztXLn/dPLyUPStnZlm/bbctqsVL1b5dKGGfgMZN1L3WQpHLFblkd3bdfzpoj15stq+6W1UNUyLXMFTpF24snlWuj8aAD8V4LtayOWTsuAAAARldohhwVciH81o6Rc6V8zj+dmP26tm7b7VIHu13VVR0nhQOq6Ay5khUCpke4ZS5UIdeDLav2D+kT54KW1Rt2Eci1yjU3jAq5GucMuREOxF3snycj/KMFAAAAY4YZcvHG64gEE+GiKfinM/N3tHXbcIVc+4+93UIHKRxQRWfIhVpWRzgQsCuvelMhZ5UxW1WEtKy2ztV5TCBXMxkz5OwKufH63AAAADC6QjPkaFkN4WgNI8XbXNE567s2v9RmINflUgc7kJuOCaLsgCo6Q84eYukKCUaFvbhiNtd9IJdxhJMz2ZSWZ7Jd3/ekcIUwrq/rJEo6wrdRDsRd7Aq5BFtWAQAAMCTCxRdUyNnG64gEY+/i019QcSt4mKp4Wpxaauv29oHqTrWshirkGgI5u2V1dN9+b7jrWiWMdPO+Od1+7ULX95dONQYIN+6e7mjhxKRyBnJUyEmKqZAb4UDcxf7L47hV/wEAAGB02UUCJVpWQ7ovbQH66MzT9/in58ppHVyaauv23VfIBQFbbMtqaIZcpGXV+gE0ygfN3/EN1+sVt+7V4lTGud2zXa5w8kYWOrTFFV5mUt3P9xsHrvl6o7xUxcVuWaVCDgAAAMOCltV4BHIYKafO3+ufTpendHTPTFu3t8OjSgcVcuslu0LO/fZptULOtflxlOzqYTupK5C7gUCuLa7QiS2rNe6lDuMVWoWWOoxZ9R8AAABGFy2r8Thaw0h54tKj/umkt6TZXLqt29vH4NUOKuTsltWpdFzLqjVDLrJl1V7qkHG0aU4q16yzQwRybXEVfNGyWuOqRh31QDzKrvhzhbMAAADAINjHekUCuZDxOiLB2Du7fsY/nc9e1/bte7nUIa5l1V7qsF6ItqyOT4VcL7mqlWhZbY+rTTE7Zm2ZnUo63mvjViFnz8TrRRs5AAAA0Av2vPByhZZVG0drGB3lgi5U1/2zC/PH2r4LO7TopGV1w6p4i1vqYAd1a8WyPOtxiuXxmCHXa7Ssds8VyFEhV+Nc6jBm7z/7PTRuCysAAAAwuuxCFFpWwzhaw+i4cEJnrIDhyPKNbd+FXSHXyYKXVirk0smEH4RUPWmzFDyQXSE3yltWey0dCY52TWc0n2+vHXnSsWU1nutrM24VqnbASIUcAAAAhoXdmVKueqGClUk3XkckGG/nH9DpZNAOevu+69u+i2S3Sx2K2y91kKSZmDly9gw5qlgC6cjXgnbV9rlCGJY61Ljea+PcssoMOQAAAAwLY0zod+8Sbas+jtYwMgqnv6azqaAq7euubb9Czm7r62ipQ2n7CjkpvGn16kbJP23/8KFCLhD9WhDItc8VwlAhV+OskBuz99+u6WDr8dJ0ZoDPBAAAAAijbdVtvI5IMNaefOqrKm+FDtlKWrumZ9u+j+6XOgTVbs0CuWtmc/7p01c3/dNlq0+WQC4QbVm9cZlArl2ugstMMv57dJK42lPHrUL1RceW9Zo79+vrb1zSdz63/ephAAAAYKeEK+QI5Orie+6AIXPm8glpvnY6bxY6uo9ulzqEWlbT8WHH/oW8vvDYJUnSqctBIGdXyI3bUPluRMPJQ1TItc3VskqFXE3S8V4bt0A8nUzoF779mYN+GgAAAEAD+7ikSCDnG68jEoyvakVXCqf9s7PZ/R3djX0M3lHLamipQ3yevX8hqJA7dXnDP23/NSA9ZkPlu5GJBCZsWG0fLavx3DPk+NoAAAAA/WD/7l1mhpyPIxKMhksndc4qSNs7e7Cju+nlUodmLasHFvL+6aevBIFcmQo5p2g4csMuArl2ueaksdShxj1DjvcfAAAA0A8pWladOFrDaDj/oE5bCx0OLx3o6G5CLatdVsjlmwRy++aDQO6py8yQ2479tTiwkFeuSTsw3FyLNamQq5mELasAAADAsLKP9wjkAhytYSSUTt+n06mgRfSW5es6uh+7UqbaSYVcqbWlDq6WVc/zIltWCQTq7KUObFjtDBVy8ZwVcrSMAwAAAH2RCQVytKzWcUSCkbD61Nd0xtoYeXCuwxlyoQq59m/fUcvq5Q15nqeyVZGXTBgZV0nThDq4GHy9br92foDPZHQ5Z8hRhSnJHcgRiAMAAAD9QcuqG1tWMRK8sw/o9GwQgO2d3tvR/dibKLtd6pBvstRhPp9WPp3URqmitWJFVzfKofZBVwvdJHvBTcv64W86qgtrRf3ACw8P+umMJFfAS8tqDRVyAAAAwODQsupGIIfh53nKrpzQuYVdWxcY7Zna09FdhSrkul3q0GTOmTFG+xdyOnFuTZL01OUNXbsUVIExPy4smTD6oZfdNOinMdJcoROBXI0rfGOpAwAAANAf9vFvsUzLah1Haxh+V0/piinI2wrTZlILSifTHd2VXSHXyVKH9WIwQ67ZUgdJ2m+1rZ66vMGGVewoV8abTbEcQ5JcBamE4gAAAEB/2KN07EWHk44jEgy/8w/odDIo5twz1Vm7qhQOLdpd6lCqVP0BlAmz/cD80By5KxsqV9iwip2ToGU1ljGmoU2ctnEAAACgP5gh58bRGoZe8fT9Om1V+lw339lCBym61KG9QG5lM6iOm86ktl3KsG8+COSeurypkvV4acIA9JgzkCP49UVbelN8bQAAAIC+oGXVjSMSDL2VJ/4+FMgdmNnX8X1107L66Pm14DlYW0Hj7F/I+adPXd5QqRz8JYAwAL3GDLnmohVxbFkFAAAA+oOWVTeO1jD0qmfDLaudbliVwhVy7basnji36p8+cs3MttdvaFmt2oEcYQB6y1Uht11b9SRpqJBjyyoAAADQF7SsunFEgqE3dfVEqEJuz3RnG1al8EF5uz8HTpwNArnDy9sHcvtCSx02/flzkpQmDECPuSrkCOQC0apUKuQAAACA/rBbVku0rPo4WsNwW7ug6fIlnU5ZFXJdLHWwW1a7qZA73EKF3L75oGX19NVNFayW1XSKMAC95RpLSMtqgBlyAAAAwGCEZshRIefjiATD7fwDkqQzVoVcr1pW250hd+JcMEPu8PL0ttfPpZPaPZPxH+upSxv+x2iXQ68lmCHXFFtWAQAAgMHIWN0pZQI5H0drGGqFp+9TwUgXk7VALmmSWs4vd3x/nS51KJQrevziuiTJGOnQ7u0r5CRpv9W2+tjFINCjXQ69lmTLalPRCjnCSgAAAKA/7O4Ue5TTpOOIBEPtyhNf0xlrocPy1LKSiWSTWzTX6VKHxy6s+wHegYW88pnWnoPdtvr4hXX/NBVy6LVo4JQwtGXaqJADAAAABoOWVTeO1jDUqmfuCy106GZ+nCTZ+UQ7FXLtLnSosyvk6hV2EltW0XvRArlsqvPgehwxQw4AAAAYjHDLKhVydRyRYKjlr54IL3ToYn6cFGlZbaNCLrTQoY1A7oDdsmpVyNFKiF6jJbO56NeHtnEAAACgP0JbVqmQ83HEhqH273f/nH5FL/bPdxvIhVpW26mQsxY6HGlhw2rdvvkgkHv6irXUgTAAPRadIUcgF5aMtInTNg4AAAD0R4pAzim1/VWAwfnshWldzE8ps3W+60AutNSh9ds9HGpZ3X7Dat3+hWCGnJ3/0S6HXjPRQI7vsZDozDgq5AAAAID+sH/3ZoZcgCM2DLUP/sBzdeeNQZLV7Qy5RAdLHTzPC7estlEhZ7es2tIMlEePRVsys1TIhdhfn2TCNASYAAAAAHaG3b3DDLkAR2wYageXplQyl/zzva2Qa+0Hwemrm1ovViRJ8/m0dk1ntrlFYPdM1lmJQ4Uceo2W1ebsCjk2rAIAAAD9Y4+LoWU1wBEbht7ptdP+6T3Te7q6r06WOpw4G8yPO7w83VZlTSJhtHc+13B5mkAOPRYdiUaFXJgdxvP+AwAAAPqHllU3jkow1NZL61oprkiS0om0lnJLXd1fJ0sdHj674p9uZ6FD3f75xrZV5leh1xJUyDVlL1JhqQoAAADQP7SsunHEhqEWqo6b2qOE6e5b1i6MabVl1d6weni5/UDONUeODY/otegMOQK5MHvLKu8/AAAAoH/SbFl14qgEQ80O5LqdHyeFq4hablm1Fzp0EMjtW3C1rFKhg95qqJCjLTMkFWpZ5f0HAAAA9Iv9uziBXIAjNgy10+u9DeQ6WerQ6YbVuv2uCjkCAfQYFXLN2V8f3n8AAABA/6StY5MiLas+jtgw1HpdIbdkbUh98tLGtte/ulnSmasFSbWKo4OLjeHadpyBHC1z6LHo4tBsKjmYJzKkQhVyvP8AAACAvrG7d8pUyPk4KsFQO7N+xj+9d6r7QO743jn/9IOnV+Rt07b6iDU/7obdU0p10AboWupA9RJ6zRgju2uV77EwtqwCAAAAg0HLqhtHJRhqva6Q2zOX1Xw+LUlaKZR16spm0+ufONvd/DhJ2u+YIZeKljMBPWBvESaQC0vRsgoAAAAMBC2rbhyxYaj1OpAzxujYnln//AOnrza9frcLHSRpNpfWbC4VuqyTSjtgOwkrdGKpQ1hoyypfGwAAAKBvaFl1S21/FWBwfvZFP6tTq6d0eu20rp29tif3eWzvrD5/8qIk6f7TK3rp8T2x133YqpA70sFCh7oDC3ndf3rFP8+WR+wEu/AyS4VciJ3BpalQBQAAAPpmaTqjb3vWAWWSCR1wzFifVARyGGpHF4/q6OLRnt7nsb12hdxKk2v2pkJOkvbN50KBHEsdsBNoWY0XrpAjkAMAAAD6Zf9CXj//xjsH/TSGDkdsmDjHWwzkSpWqHruw7p8/tDzd8WNGN61SIYedYLesUiEXlmKpAwAAAIAhwlEJJs5RK5A7cW41dsvL4xfXVa7WBk7um89pOtt5QWljIMdbD72XoEIulr1llaUqAAAAAAaNIzZMnLlc2u9bL1U8PXJuzXm9Xs2Pkxo3rdIyh52QZKlDrPCWVb42AAAAAAaLoxJMJHuO3P0xm1Z7NT9OkvbPhyvkmCGHnRCukEsO8JkMn2TSblklEAcAAAAwWKQCmEitLHY4cTaonDvcxfw4iRly6A+78IuW1bBQhRyBOAAAAIAB46gEE8le7PDgmZhArocVcnvnc7KKl5ghhx1hV8ix1CGMLasAAAAAhglHbJhIR/fYLauNgZzneTphzZA73OUMuXQyoWtms/55AgHsBJY6xAttWaVCDgAAAMCAcVSCiXR4ecY/QH/y0oZWC+XQx09f3dTK1mWz2VQoTOvU9UtB2+tcLt31/QFRoaUOBHIhoS2rBOIAAAAABowjNkykTCqhQ9ZcuOgcuU/ed9Y/fcv+ORnT/QH8v3jpEe2dy+lVt+/Trfvnur4/IMoOnbK0RYeEKuT42gAAAAAYsNSgnwAwKMf2zunBM7W21AdOr+iu6xf9j33ky6f80//wtr09ebwXHV3WZ3/spT0J9wAX+1uLCrmwG3cHAfyhLpe0AAAAAEC3COQwsY7vndVHvlw7/cDpq/7lZ65u6vMnL0qqBRyvfMa+nj0mYRx2UjK01CE5wGcyfF528x79+Ktu1nqxojfcde2gnw4AAACACUcgh4l1LGaxw5999Wl5Xu30N9y4pGvmcv1+akBHDi5N6aGzq0oljPbO831rSyaMvu8Fhwb9NAAAAABAEoEcJtixvUEg98CZFXmeJ2OM/uQrT/uXf8vt+wfx1ICO/NtXHteu6Yyef9NuLfdgEQkAAAAAYGcQyGFiXbuY10w2pdVCWZfXSzq7UlCl6ukLj12SVKuo6dX8OKAfjlwzq5/9R3cM+mkAAAAAALbB1G9MLGOMju6Z8c/ff3pFf2pVxz3v8C7tmqHKCAAAAAAA9BaBHCaa3bb64OkV/clXgu2qr6ZdFQAAAAAA7AACOUw0e7HDx+89oy8/eUWSlE4affOttKsCAAAAAIDeI5DDRDu2d84//fmTF/3TL7hpWfNT6UE8JQAAAAAAMOYI5DDRjlstq7ZvuX1fn58JAAAAAACYFARymGiL0xldMxte3JBJJfRNt+wZ0DMCAAAAAADjjkAOE+9YpEruxUeXNZujXRUAAAAAAOwMAjlMvGjb6qvvYLsqAAAAAADYOQRymHj2Yod8OqmX3XzNAJ8NAAAAAAAYdwRymHjPObSkdNJIkl5z535NZVIDfkYAAAAAAGCckTxg4l27OKUPvPW5uv/0Vb32zgODfjoAAAAAAGDMEcgBku66flF3Xb846KcBAAAAAAAmAC2rAAAAAAAAQB8RyAEAAAAAAAB9RCAHAAAAAAAA9BGBHAAAAAAAANBHBHIAAAAAAABAHxHIAQAAAAAAAH1EIAcAAAAAAAD0EYEcAAAAAAAA0EcEcgAAAAAAAEAfEcgBAAAAAAAAfUQgBwAAAAAAAPQRgRwAAAAAAADQRwRyAAAAAAAAQB8RyAEAAAAAAAB9RCAHAAAAAAAA9BGBHAAAAAAAANBHBHIAAAAAAABAHxHIAQAAAAAAAH1EIAcAAAAAAAD0EYEcAAAAAAAA0EcEcgAAAAAAAEAfEcgBAAAAAAAAfUQgBwAAAAAAAPQRgRwAAAAAAADQRwRyAAAAAAAAQB8RyAEAAAAAAAB9RCAHAAAAAAAA9BGBHAAAAAAAANBHBHIAAAAAAABAHxHIAQAAAAAAAH1EIAcAAAAAAAD0kfE8b9DPATvAGHMhn88v3XzzzYN+KgAAAAAAAGPhvvvu08bGxkXP83Z1cz8EcmPKGPOopDlJJwf8VHrh+Nb/7x/os8Ag8NpPNl7/ycVrP9l4/ScXr/3k4rWfbLz+k2tUX/sbJF31PO/Gbu6EQA5DzxjzBUnyPO+uQT8X9Bev/WTj9Z9cvPaTjdd/cvHaTy5e+8nG6z+5Jv21Z4YcAAAAAAAA0EcEcgAAAAAAAEAfEcgBAAAAAAAAfUQgBwAAAAAAAPQRgRwAAAAAAADQR2xZBQAAAAAAAPqICjkAAAAAAACgjwjkAAAAAAAAgD4ikAMAAAAAAAD6iEAOAAAAAAAA6CMCOQAAAAAAAKCPCOQAAAAAAACAPiKQAwAAAAAAAPqIQA5Dyxiz1xjzC8aYE8aYTWPMGWPMR4wxLxv0c0PnjDHXGWPevvVaPm6MKRhjVowxXzbG/IwxZl/M7W4wxngt/Pfsfn9OaI0x5rtaeP1Wm9w+YYx5qzHms8aYy1vfN/cYY95pjMn083NBe1p879b/e1Hktrz3h5wxZtYY863GmPcYY/6XMea89bocb+H2Xb23jTHPNsZ8wBhzauv3hceNMb9pjDnSm88QzXT6+htjcsaY12+9Vl8xxqxu/U7wuDHmd40xL97mcf+yhZ8Lv9zzTxi+bt77Lf5cf8M298F7f4C6eO//dhu/E/yW4/a89wfMdHg8Z90+Y4z5EWPMl7Z+9l/e+h3grcYY08Ljv3zrsc9uvfdPmFpusKd3n2V/pAb9BAAXY8ztkj4ladfWRVcl7Zb0LZJeZYz5t57n/cygnh86Y4w5KOmkJPsH7VVJ05Ju3/rvrcaY13ue9xdN7upMk4+Vun2e2HElSRdjPrbmutAYk5b0R5JeuXVRUVJF0p1b//0jY8xLPc+LDfQwUM3es5I0Jymv2uv69x3eD+/9wXmZpA91csNu39s/XpoeAAAZd0lEQVTGmH8q6TdV+53WU+3flIOSvlfStxtjvtXzvE918tzQsk5f/49Ierl1vqDa+/jg1n9vNMb8gud5b9/mfq5K2mjyMeycjt/7lvOqveddNuNuxHt/KHT6+l9R83/P05KWtk5/scn1eO8PQLfHc8aYOdWO8+/aumhdtd8Bn7P136uNMa/zPK8c8/jvkvQfts5WJa1KOiTpbZLevPU7Q7PfJYcKFXIYOsaYvKQPqxbG3SPpNs/z5iUtSvrPqr35f9r8/+3debQkVZ3g8e+PrSgoKECQRUUYVwRFsFkGgUZ0aEShUZs+rdBaZ1wG7VZkRkVpPaAiIN2tzui0A3hYFFCEZnGhWxnbUloaZFBQUcGFkkXZlwKKTfjNH/cGL8nKzPdeZlZmvsf3c06cyIi4N+JGRt7MjBt3idhnfKlUn1av828CBwEb1Wu7DuVm7HrKdb4gIjbrtpPM3KzHdPWqPgkN7NIe1+85XeIcQ/mMPAQsoXxm1gX2pxTu7QScOIK0qw/T5NnNgOtq0G9k5p197se8P163ARcBHwXeMYt4feft+vDuZMoN+ZnAppm5AbAVcHHdzz9HxCazPhvNVj/Xf03gV8AHgG0yc+3MXAQ8FzinhjksIt41zX4O6/G9cGQf56LZ6TfvN3bqcf2+0SmCeX+izPr6Z2avPLsZU4UtjwBn9diVeX88Br2fO5lSGHcX5bd+UY27hPJf4LWUz9NKImI/pj4f/whsUI+9HXAVsAlwYUQsGOwURygznZwmagLeS3nSdR/wjA7bz6/brxx3Wp1mfW0XA9v32P5CypOuBI5q27ZVXZ/jPg+nvq//knoNl84y3maUH+gE3tNh+5/XbY8DLxn3eTrN+nPx0iZvAwd02G7en/AJWL3bNQNe2CPeQHmbUrMugSs6pGERcEPd/o/jfo/m8zTA9d+tPW7LtgC+U/fx2y5hltbtS8b9HjxVp36vfQ3bhNuqj+Oa9ydgGuT6T7Pfq+o+zuuy3bw/3us+yP3cDtP85zusblsBPH02nw3gmZTygwTePe73aaaTNeQ0iQ6u87My8+YO2/++zneMiBeMKE0agsy8N3vUYsnMXwKX1cWXdQunp5w3AAsoTRxOat+YmRdSalgF8KbRJk1D8JY6b56ya47JzG7NzabTd96OiA2Yaub6qfY0ZGni+n/q4htn0ieN+tPv9c/MS7vFzXJ39cW6uHVEbNQpnMZrgLzfN/P+5FgV1z8itge2r4unDXv/GtyA93PNb/m1mfm1DtFPovwnWAi8vnVDRGzL1Gfj79vikZk3AV+uiwe3b59UFshpokTEekxl3G91CXYZJaNC6btA80vTXG31nqH0VPKKOv9+ZnbrT+bbdb73CNKjIYmINZj6c3ZWdukvRPPWIHl7d0qTx9Yw7Zr/EZsD2/SVQo1Ta/N1/xOoYd6f35qHdLcD/zLOhGgg3e7nmt/9jnk3Mx8ELqmL7b/7Tdx7gcu7HLfJ+ztHxKKZJXW8LJDTpNmGqQ4ir+kUIDMfB66tiy8aRaI0GvXm/OV1sWtnnHUUnuUR8WBEXB8RZ0TE7qNJpYZg24i4pl6/+yLiZxHx6YjYukv4Jp93/E6ofl7n2/g0fE55NfD0+vr06QKb9+edQfJ2E/eW7N7v4M9bXvt/Ye5pRly+ldLxfzfvq6NsPhIRt0fEdyLinRGx9gjSqMF9NSLurqM03hQR/xwRr+kR3rw/T9X7gNaWUtMN1mTen0Dd7ufqb3gz+u5Mfvfb826z/ItaHtArbuuxJpoFcpo0rUMk/75HuGZbzyGVNef8DaVPocfpfXO+aw0Dpb+Kg4FLIuIzFsbMCRtTCt9XAGsD21L6jrwmIjo1OW3y+Uy+ExbVSXPDkjq/OjOvmkF48/78MkjenjZufdJ+T1t4zQER8Qzg0Lp4Wm3C2s22lBEZH6D8vuwN/BPww4jYcpUmVMOwE6UWzaPAMyjN1L4REV+NiLU6hDfvz1+tD+lOm0F48/5k6nY/tz5lwBXo7z5/Nv8ZOsWfSBbIadKs2/K62zDWUG7kwRvveaOOmHVcXfxcZv68LchDlB/ZPYH1soymtQ6lifPXa5jDgA+NILnqz++BoygjIa2dmU+j5OHXUJ5oLQROj4g92+I13wsz+U4AvxfmhNon1GvrYq8CePP+/DVI3p5J3Nb4fi/MEbV2xZlMdc5/XJegS4E3U266FmbmhsCmwJHAw8CLgYu6FOpo/E4H9gU2zMz1s4ywuw1wat1+EPC5DvHM+/NX01z1J9M8pFuKeX8iTXM/N+h9/ry8H7BATtLYRcTmlBGzFgJXAke0h8nMWzLzbzLzktphL1n8KDMPAM6pQY+sHf5qwmTmtzPzY5l5TWY+Utc9nJkXUUbb+zWwBnD8ONOpkXkjsBbwR8rNd0fmfekp57OU5qqPAG/KzHs7BcrMozPzS/U7Iuu62zLzOMqAIVBq0CwZQZo1S5m5JDO/lZn3tKz7ZWb+V6Y6bH+bA7g9NdSHdPvXxZ5dWJj3J9NM7ue0MgvkNGkeaHm9sEe4der8/lWYFo1A/QH+NrA18CvgNT069+6l+dJfFwf7mHPqDdexdXHXiNi4ZXPzvTCT7wTwe2GuaJ6E/0tm3jbAfsz7c9cgeXsmcVvj+70wB0TEsZSmqo8BB2fmD/rZT2Z+E/h+Xdy/V1hNpI9SasEEUzWpG+b9+emvmHpId0a/OzHvj8cM7+cGvc+fl/cDFshp0rS2+96iR7hm2x9WYVq0ikXEYspoONtRmqW8KjNv7WdfmXk9ZUQmgP80nBRqxJoRk4Lyg95ovhdm8p1wf2beN+yEabgiYhtKv0Ewg8EcejHvz2mD5O1p40bEQqCpNen/hQkXEX9HaXqewNsz89wBd9n8pvi9MMdk5gNMdQbffv3M+/NT85DuXwd8SAfm/ZGaxf3ccqYK1fq5z5/Nf4ZO8SeSBXKaNL+k/BGDUtV4JRGxGtBUX2/vZ0xzRESsC1wE/AlwC+XL+4bxpkoTqsnnHb8TqidGXlrFadFwLKnzu5jqB05PPYPk7SbuZhHxtGnitobXBIqIw4Fj6uJhmXlqr/B6SjPvzzP1Id3OdXGgh3Qardncz9Xmxc1v+Ux+99vzbuuo693KsZq4rceaaBbIaaLUp9//ry7+ly7BdgEW19ffWeWJ0tDVJ5dfp/Qbdifly/tXA+5za2CTunj9YCnUmOzS8npZy+vv1vkePYayb74v/E6YcBGxOnBIXfxy05/gAPsz789dg+Ttf6eMygjwqi5x96nz3zNH/pg/FUXEO4FP1cUPZuZnh7Tr5jfF74U5pt7kb1cX26+feX/+aWrH3QV8bQj7M++PQJ/3c83vfsf7/PpfYI+62P6738RdzFQri3ZN3r+81rSdeBbIaRKdVecH184h272vzq/MzGtHlCYNSR3x6DzgFZQh6ffJzGtmEC+mCdL0P/Yg8G8DJVJDN931i4j1gQ/WxR9m5u0tm8+jjJq1AfC2DnH3p9SaTeDLQ0mwVqVXMdWkYNon4eb9ea3vvF37nbyoLv739qfl9Yb+0Lr45abjb02WiHgL8L/r4scy85MzjDfdb8q+lJGZAb7Zfwq1Kszge/0jlH6ikql8Dpj355t6/ZqHdF+Z7iGdeX8y9Hs/x9Rv+Qsjor1/SIC3UwrcHgTOb91QR2y9ui6+v0OatqAMGAY9BgubOJnp5DRRE+UHeBnlR/hK4EV1/XrACXV9UjL+2NPrNKtruzpwbr1+y4FdZxH3e5S+ZbYDVq/rAtiB8oXdfC6OGvd5OnW8flsBlwFvBbZsWb8WsC/w03r9HgP27hD/k3X7CuCvWz4D+wF31G1njvs8nWb0WTirXq9rZhjevD8HJmDjlmmHluuya9u21dri9Z23ge0pI3Em8EVg47p+S0p/NgncDWwy7vdnvk/9XH/KaIh/rOFOmOXxPgScQqllsV7L+k2AD9TPU1K6Qlkw7vdnPk99XvtzgE9Qmrmt1bL+BcDJLfs4tcsxzfsTMvX73d8Sf5+WODvP4Hjm/fFf877v52r8s2vcO4D9Wvb55pbr94kucfdr+byc0HwGKE1Vr6zrfzOXrn3UE5AmSkRsT6mm2vQNsRxYRKnVmcCRmXn8mJKnPkXEnpSba4CHgHt7BL8xM5+ojhwRy4Bn18VHKZ+JdXjySDufpfQ94xfbhImIrXhy04GHKB27rg+sWdetAA7NzC91iL8mZSj1/eqqhymFd81oSlcAr0wHdJhotSbkLZR8e0RmnjCDOMsw70+8iJjpe791Zi5riTdQ3q41rL4ArMHUzUHTrcUDwAGZac3JVayf6x8Rv2VqAJ/pBnR6fWZe2nK8o4Gj6mJz3ZOpjvyhPOg5oPXzpuHr89ovBf60rn+M8n9wAWW07Ma5wCGZ+XCX45r3J0C/3/0t8c8E3gT8IjNftFKslcMfjXl/rAa5n6vx16e0aHhZXbWCUiC3oC5/A3hdZv6xy/E/DHy8Lj7G1P0ElEK+V2TmzzrFnURrjDsBUieZeXVEbEd5CvJa4BmUtuk/BD6dmfYTNTe1NitYu07dtA+V/X7K07Cdgc2AjShPR68FfgCclJmXo0l1K/AeYHfKk+1NKH+cH6AMj/4d4POZ+btOkTPz0dp87e2UAQFeRPnxvopS/f0zOWBfZBqJv6QUpD0OnDHDOOb9eWzQvJ2Zp0fENZTPyZ6Uz8eNwMXAcZn561V7BhpA63+CTacJu1bb8jmU+5jdgOdQHuCuSRlV7ypKYc6Z3QpzNHbHAj+h1KJ6JiXfPk55cHcZcFpmfrvXDsz7c18tmHldXZzpYA7m/fEb5H6OzFweEbsBh1OamD6X8jDux8CpwMm9HrBm5jERcRnwXkp/gesBv6UU5B2bnUd4nVjWkJMkSZIkSZJGyEEdJEmSJEmSpBGyQE6SJEmSJEkaIQvkJEmSJEmSpBGyQE6SJEmSJEkaIQvkJEmSJEmSpBGyQE6SJEmSJEkaIQvkJEmSJEmSpBGyQE6SJEmSJEkaIQvkJEmSJEmSpBGyQE6SJEmSJEkaIQvkJEmSJEmSpBGyQE6SJEmSJEkaIQvkJEmSngIiYqeI+HpE3BERj0dERsTRM4i3VQ2bI0jmRPI9kCRJw2aBnCRJ0hBFxHtq4c37W9ZtUdf9YExpeh6wFHgtsCFwB3ArcP+A+90rIo6OiAMHTuSYRMSB9Rz2GndaJEnSU8ca406AJEnSPLNHnV/SYd2/jzgtjXcA61DSdEBm3jOLuI8C13bZthdwFHA6cMEgCRyjA4G31NdLu4Tp9R5IkiTNmgVykiRJw/Vy4EHgypZ14y6Q27bOvzrLwjgy82bghcNP0tzheyBJkobNJquSJElDEhHPATYHLsvMR1s27QEkMJYmq8DCOh+oiaokSZKGwwI5SZKk4dm9zp9orhoRGwDbAb/IzLsG2XlErBYRb42I70XEXRHxUERcHxEnRcRzO4RfVgci2KuuOrUZnCAils3wmCsNaNCsozRXBXhLy36baasO+9o/Ii6MiFsi4pGIuK0ONPFnXY69pO5raV0+uJ77nXX9gXX96hHx6og4MSKujIhb6/5/HxHnR8TeHfa9Vz2HprnqUe3n0Os96LC/HSLijIi4MSIeroNnfCsi3tAjzrK6370iYqOI+FS9ng9HxM0RcXJEbN4l7mr1/flufT8ejYjbI+KaiDglIvbtdlxJkjR+kelgUZIkSf2IiK8Au7as2gBYTBk04YG6bi1KrbkHgdtawr4vM8+dxbHWAc4H9qmrHgVW1OMBPAT8VWZe2BLnCuBZwEbAmsDymg6AGzNzpxkcdyvgeoDMjLruWcAVwCJg3Xrse9ui7pSZN9bwawKnAge3bF8OrN+yfEJmHtF27CU13veAnwDvBh6vx1oMvCEzL4iI7YCftu179Zq2xpGZeVzLvncDzqv7WZtyvZ5UgzAzN+v2HrSl8x3A55l62H0PsF5NA8AZwJLMfKwt3jLg2cBfA8fU1ytqvAU12DJgx8y8uy3umcCbWlbdS6kJuVZdvjwzWz+bkiRpglhDTpIkqX+bUQpRmqkpHNu4ZV1Tw2lhW9hFszzWpyiFcQ8DhwLrZeYGwAsogxGsDZwVEc9vImTmTrVQ6dK66rDM3KxO0xbGdZOZN9b9/kNddXbLfpvpxpYoJ1AK434N/CWwKDMXUwrk3gXcB3wgIt7Y5ZAvA/6WUiPvaZm5EWW02Oa8HgFOAf4MWJyZizNzEbAp8BHgMeATEbFLyzlcWs/h7LrqH9rPYSbvRS3YawrjzgWelZkbUgpnP0xpqnwI8KEeu/kscDewW2auS/ls/DmlYG+r9rgRsSelMO4x4HBg/fpZWBvYAljC+PorlCRJM2CBnCRJUp8yc6/MjFprqinA+VGzrq7/j7p+y9b1mXnaTI9Ta2i9vS4elpknZubDNQ3XAa8BfkMZSfXDA5/YEEXE84DDgNuBvTPznMx8ACAz78vMz1NGgQX4uy67WQQcn5kfawalyMzlmXlbfX1dZr41M7+dmcubSJl5W2YeA3wUCEpB5rB9nPKf+geUGoo31WPfn5mfAI6v4Y6IiPW77ONh4FWZ+R817h8z82uUWnMAf9EWvqn5dnFmfiYz76vxMjP/kJmnZ+b7hnJ2kiRplbBATpIkaTj2rPPvNysiYiGldtfv2mqMzdbrKP/bbgG+0L4xM1dQaqEBvD4iVm8PM0ZvphSGnd3jPTiXUii1bZc+0x6j1BDs19fr/OUD7GMlEbER8Iq6eFx7k9Tqk5QmvYuA/brs6qTMvLPD+gvqfOuIaG1+2xQ6Pj0i/D8vSdIc5A+4JEnScOxR599vWbcLpU+vQZsP7ljnl3Qp9AH4tzpfl9KMdVLsVudvqYM5rDQBN1H6uIPS5127X2fmHb0OEhELI+LwiFhaB4t4tGUghh/XYFsM5Yym7EApbExKP3crycx7gSvr4o6dwlD64+vk5pbXG7S8/g6lme6OwNKIOCQihn1ukiRpFVpj3AmQJEmaJ5oacq2FbyuNutqnTer85h5hbuoQfhI0Nd7Wq9N01umw7vZeEWqtuqXA81tWP0Dpl+1xyiAJG/PkQR6GoXmf783M+3uEa65Nt+tyX6eVmflQxBNjSKzZsv5XEfFO4HOUguA94IlBIv6VUuPux0iSpIllDTlJkqRZiohndajl9RJKTamftqw7skY5ri1sv9YeNO1j0PzfPLytD71u09IO++hWK7DxGUph3G+BNwAbZeaizHx6HZxhVY82umD6IMOVmacAWwPvBS4E7qQMAHEocGVEHNk9tiRJGjcL5CRJkmZvdcoInq1T1Kl13cIafsO29bPV1BDbskeYZ3YIPwlurfNeae9bRKxFGZEU4ODMPC8z724L1s97PhPN+7wwInrVSmyuzVCvS2bempn/MzMPpNS+2xk4n/I5/HhEvGSYx5MkScNjgZwkSdIsZeaytpFUm9Ew39yy7k/qugvba4H1ccgf1fkuEdGpSSfA3nX+AHBtH8fox+N13uucmlFm911FadiYqRpq3ZppvqpH/JmcQzc/ptSKhKnBHZ4kIhZTBvaAqes4dHWE1SuAgyhNZFdjqsm0JEmaMBbISZIkDe5P67x1QIeVRl0dwHmUgqOnAe9o31gL6d7fhO0x8MOwNaN9btAjzBcphVbbRMR/67WziNiwjzTcx1Sh2Is77HNz4N094s/kHDrKzLuA79bFI7qMeHoEpanx/cBFsz1GJ7VWYLc0PQY8WhdH3pRWkiTNjAVykiRJA4iItSlNBW/IzN+1bGpGXe04+uZs1P2eVBePj4h3RMSCevznA98EngusYKq23ihcU+e7R8TzOgXIzJ8Dn66L/xQRx0XEE81rI2K9iNgnIs4AzpltAjLzPuCyunhKRLy07ne1iHgl5f3vVfutOYd9a+HdbH2EUli6I/CV5twiYlHtx+2DNdzxmbm8yz5m69iIODciDoyIjZqVEbFpRPwvSt9yCVw8pONJkqQhs0BOkiRpMLtQaiK1j6S6O6X21VVDOs7/oBSwLABOBO6LiLspzVP3Ah4G3pSZ1w3peDOxFPgNsBFwbUTcFhHL6tTap90HgM9T/nt+ELgxIu6NiHuAe4FvAQdT+ubrx+HAg5Qacj+OiPspNdL+L6VW4Vt7xD0fuIsyKMRNEfGH5hxmcuDMvBR4F6VQ7iDghoi4C7gH+ASlMPBM4Pg+zqubNSiDV5wP3Fnfy+XALUzVBvxwZv5siMeUJElDZIGcJEnSYFZqmhoR21A62b90WM1HM3MF8GrgbZTCvxXAOsDvgC8AL87MC4dxrFmk6VHglcCXgJspg1c8u05rtIR7LDPfRSmkPKOmeQGlKecNwNeAvwX+os90XA78Z+AC4G5gTeA2SsHlS4Gre8S9g9L/23mUQRc2aTmHmR7/RGAn4CzgD8AiSkHjxcBBmXnIkJsRfxp4D2V01esohX4LgBuBs4E9M/PYIR5PkiQNWWTm9KEkSZIkSZIkDYU15CRJkiRJkqQRskBOkiRJkiRJGiEL5CRJkiRJkqQRskBOkiRJkiRJGiEL5CRJkiRJkqQRskBOkiRJkiRJGiEL5CRJkiRJkqQRskBOkiRJkiRJGiEL5CRJkiRJkqQRskBOkiRJkiRJGiEL5CRJkiRJkqQRskBOkiRJkiRJGiEL5CRJkiRJkqQRskBOkiRJkiRJGiEL5CRJkiRJkqQRskBOkiRJkiRJGiEL5CRJkiRJkqQRskBOkiRJkiRJGqH/D175L87f73ezAAAAAElFTkSuQmCC\n", "text/plain": [ "
" ] }, "metadata": { "image/png": { "height": 446, "width": 626 } }, "output_type": "display_data" } ], "source": [ "plt.figure(figsize = (10, 7))\n", "plt.plot(log_likelihood_sgd, label = 'stochastic gradient descent')\n", "plt.plot(log_likelihood_batch, label = 'batch gradient descent')\n", "plt.plot(log_likelihood_minibatch, label = 'mini-batch gradient descent')\n", "plt.legend(loc = 'best')\n", "plt.xlabel('# of iterations')\n", "plt.ylabel('Average log likelihood')\n", "plt.title('Convergence Plot')\n", "plt.show()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Based on the convergence plot above, we can see that the it's a good idea to use mini-batch gradient descent since it strikes a good balance between batch gradient, which convergences steadily but can be computationly too expensive when the dataset is too large, and stochastic gradient, which is faster to train, but the result can be too noisy." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Pros and Cons of Logistic Regression" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We'll end this notebook listing out some pros and cons of this method.\n", "\n", "**Pros:**\n", "\n", "- Highly interpretable (if you remember how).\n", "- Model training and prediction are fast. Thus can be desirable in large-scale applications when we're dealing with millions of parameters.\n", "- Almost no parameter tuning is required (excluding regularization).\n", "- Outputs well-calibrated predicted probabilities.\n", "\n", "**Cons:**\n", "\n", "- Presumes a linear relationship between the features\n", "- Performance is (generally) not competitive with the best supervised learning methods.\n", "- Can't automatically learn feature interactions." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "# Reference" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "- [Notebook: Logistic Regression](http://nbviewer.jupyter.org/github/justmarkham/DAT8/blob/master/notebooks/12_logistic_regression.ipynb)\n", "- [Coursersa: Washington Classification](https://www.coursera.org/learn/ml-classification)\n", "- [Blog: Maximum likelihood and gradient descent demonstration](https://zlatankr.github.io/posts/2017/03/06/mle-gradient-descent)" ] } ], "metadata": { "anaconda-cloud": {}, "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.4" }, "toc": { "nav_menu": { "height": "277px", "width": "252px" }, "number_sections": true, "sideBar": true, "skip_h1_title": false, "title_cell": "Table of Contents", "title_sidebar": "Contents", "toc_cell": true, "toc_position": {}, "toc_section_display": "block", "toc_window_display": true }, "varInspector": { "cols": { "lenName": 16, "lenType": 16, "lenVar": 40 }, "kernels_config": { "python": { "delete_cmd_postfix": "", "delete_cmd_prefix": "del ", "library": "var_list.py", "varRefreshCmd": "print(var_dic_list())" }, "r": { "delete_cmd_postfix": ") ", "delete_cmd_prefix": "rm(", "library": "var_list.r", "varRefreshCmd": "cat(var_dic_list()) " } }, "types_to_exclude": [ "module", "function", "builtin_function_or_method", "instance", "_Feature" ], "window_display": false } }, "nbformat": 4, "nbformat_minor": 1 }