{
"cells": [
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"# Neuro-Fuzzy Network\n",
"The following notebook explains the theory behind Fuzzy Neural Networks. After that you can find a use case example where the model is applied. We create rules and implement them into the network which we build with the prosper_nn package.\n",
"\n",
"## Theory\n",
"\n",
"An MLP neural network gets numerical input data, e.g., sensor data and predicts a numerical forcast. The input data is processed inside the network's hidden layers. What the network does within this process can hardly be interpreted. The neural fuzzy architecture whitens this black box by giving each hidden neuron and its connections a human interpretable verbal meaning. Because of this verbal meaning, we can construct rules for the network to follow. A simple example would be: \"If input 1 increases, the output will fall\". If experts have a basic understanding of the relations between input and output of our problem, we can use those rules for initialization of our network. We infuse the domain knowledge into the network. The basic principles of the architectures for regression (left) and classification (right) are shown below: \n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"\n",
" \n",
"\n"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"The three steps **Fuzzification, Fuzzy-Inference** and **Defuzzification** are described in detail in the following chapters. \n",
"Regression (left) needs one more step to translate back from verbal to numerical space. Classification (right) happens in the verbal space and therefore no defuzzification is necessary. For *Fuzzification* membership functions are used. In *Fuzzy-Inference* the domain knowledge is infused by formulating rules. *Defuzzification* translates back to the numerical space."
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"### Fuzzification\n",
"\n",
"Fuzzification is the translation of numerical input values into a verbal description. This is achieved by membership functions. They divide the number domain of the input into groups. The functions define the membership of an input to these different groups. Within the package two membership function classes are defined: \n",
"* **gaussian membership function**\n",
" * zero-centered\n",
" * width of the cone is learned during training\n",
"* **normlog membership function**\n",
" * steepest tangent goes through zero\n",
" * slope is learned during training\n",
" * slope can not change its sign\n",
"\n",
"The user can create his or her own membership functions if necessary. \n",
"By using membership functions, the numerical value of an input is translated to the verbal space. E.g., an input can *decrease*, *increase* (corresponding normlog function) or stay *constant* (gaussian). By changing the parameters of the functions e.g. the width of the gaussian, the model learns what range of inputs corresponds with the membership term e.g. *constant* (for some measurements a fluctuation of 10 around 0 is considered stable, for others only a fluctuation of 0.1). \n",
" \n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Fuzzy Inference\n",
"\n",
"\n",
"\n",
"After translating the numerical input into a verbal description, we can apply the domain knowledge to the incoming data.\n",
"The domain knowledge about a given problem is infused into the network by IF/THEN rules. These rules can be modeled with the edge weights of layers. The relation \"*IF input_1 increases THEN rule_1 applies*\" can be modeled by setting the connection weight between node *input_1 increases* (output node of the corresponding membership function in `Fuzzification`) and node *rule_1* (`condition_fullfillment` in `Fuzzy-Inference`) to 1. **All weights that do not represent a rule are set to zero.** Example for a `conditions` matrix is shown on the right (2D input is flattened for visualization). \n",
"\n",
"An *AND* relation like \"*IF input_1 increases AND input_2 decreases THEN rule_1 applies*\" works after the same schema. More than one row is set to 1 in each column (see picture). For an *OR* relation create a new rule. Instead of: \"*IF input_1 increases OR input_2 decreases THEN rule_1 applies*\" we say: \"*IF input_1 increases THEN rule_1 applies*\" and \"*IF input_2 decreases THEN rule_2 applies*\". \n",
"\n",
"Knowing how much a condition of a rule is fulfilled, we have to classify based on the rules' consequences. \n",
"In the `FuzzyInference` the meaning of a rule is infused by setting the connection of the `consequences` matrix. The classification rule \"*IF rule_1 applies THEN class_1 should be predicted*\" sets the corresponding node connection to 1. All other connections are set to 0. The generation of the `consequences` matrix is described in the use case example below. You can create as many rules as you want.\n",
"\n",
"\n",
""
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"The layer has a constraint so that each column sums high to one. The entries of the matrix can be seen as a belief parameter of how much the network trusts the rule to predict the given class. When visualizing this matrix we can see which rules are strong and which ones are weak. On the left you can see the matrix before and on the right after training. \n",
"\n",
"In regression we classify the three classes *raising*, *stable* and *falling* in this layer. These three classes are the interpretable forecast of the Neuro Fuzzy Network. Depending on the values we see how likely the future bahaviour is."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Defuzzification\n",
"\n",
"In case of regression we have to calculate the numerical value corresponding to the verbal classification. This is done in the *Defuzzification* step of the network. We can use the `Defuzzification` layer that applies a Linear layer. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Fuzzy Neural Network Architecture\n",
"\n",
"Overall, the architecture for regression looks like this (for classification the last, blue layer is cut). The Inference matrix $P$ contains the rule premises. The matrix $\\kappa$ represents a belief value for each rule output, i.e. how much does the network trust the specific rules. The last matrix $W$ is responsible for Defuzzification.\n",
"\n",
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Neuro Fuzzy Example\n",
"First we import all packages that we will need:"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"tags": [
"hide_cell"
]
},
"outputs": [],
"source": [
"import sys, os\n",
"\n",
"sys.path.append(os.path.abspath(\"../../..\"))\n",
"sys.path.append(os.path.abspath(\"..\"))\n",
"sys.path.append(os.path.abspath(\".\"))"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import torch\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"import pandas as pd\n",
"from sklearn.model_selection import train_test_split\n",
"\n",
"torch.manual_seed(0)"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"# Initialization\n",
"# RuleParser\n",
"from prosper_nn.models.fuzzy.rule_manager import RuleManager\n",
"\n",
"# Layers\n",
"from prosper_nn.models.fuzzy.membership_functions import (\n",
" NormlogMembership,\n",
" GaussianMembership,\n",
")\n",
"from prosper_nn.models.fuzzy.fuzzification import Fuzzification\n",
"from prosper_nn.models.fuzzy.fuzzy_inference import FuzzyInference\n",
"from prosper_nn.models.fuzzy.defuzzification import Defuzzification\n",
"\n",
"# Utility\n",
"from prosper_nn.utils.visualization import plot_heatmap"
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"### Data Preparation\n",
"We now use Prosper_nn to predict wine quality. For explanation of the data processing see the [Regression Notebook](Regression.ipynb). "
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"Second, we load the dataset, split it into a training and test set and create a training and test loader. We use a batchsize of 200."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"batchsize = 16"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"df = pd.read_csv(\"data/winequality-red.csv\", sep=\";\").sample(frac=1)\n",
"\n",
"X = torch.tensor(df.drop(columns=[\"quality\"]).values).float()\n",
"y = torch.tensor(df[\"quality\"].values).float()\n",
"\n",
"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33)\n",
"\n",
"# Normalize data\n",
"normalize_data = lambda x: (x - x.mean(dim=0)) / x.std(dim=0)\n",
"X_train = normalize_data(X_train)\n",
"X_test = normalize_data(X_test)\n",
"y_train = normalize_data(y_train).reshape(-1, 1)\n",
"y_test = normalize_data(y_test).reshape(-1, 1)\n",
"\n",
"# Bring the data in batch format\n",
"X_train = X_train[: -(X_train.shape[0] % batchsize)]\n",
"y_train = y_train[: -(y_train.shape[0] % batchsize)]\n",
"X_train = X_train.reshape(int(X_train.shape[0] / batchsize), batchsize, -1)\n",
"y_train = y_train.reshape(int(y_train.shape[0] / batchsize), batchsize, -1)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Initialization\n",
"#### Rule Creation\n",
"\n",
"If you want to infuse domain knowledge in form of rules into the network, you have to create the corresponding weight matrices for the `FuzzyInference` layer. In the following, we propose a method for the matrix creation, but you can create those matrices however you want if your framework does not suite you. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### JSON file\n",
"Rules are in a language domain. \"*IF abc THEN xyz*\". We want to store the rules in this domain to keep the ease of understanding them. We use a JSON file to store the rules and use a parser which creates the matrices from the JSON file.\n",
"The file has four entries:\n",
"* `\"rules\"`, here the conditions of the `FuzzyInference` are defined in an understandable form. \n",
"* `\"classification_rules\"`, here it is defined which rule results in which class (consequence). \n",
"* `\"member_activations\"`, here the numerical translation of the member activations are stored. \n",
"* `\"input_names\"`, stores at what position in the network the corresponding parameter is put in. \n",
" "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For our dataset we created the `demonstrator_rules.json` file whose content is shown below. Under `\"rules\"` 4 dummy entries are listed. We create a rule we think will influence high quality, one for low quality, one where two conditions have to be fullfilled and a complex rule with five conditions. \n",
"Within the individual rules we define how each input must behave in order for the rule to apply.\n",
"In the `\"classification_rule\"` entry we define what rules result in which of our 3 error classes (low, average, high).\n",
"In `\"input_names\"` the names of the inputs are saved.\n",
"The dictionary \"member_activations\" stores what the verbal parameters of the rules mean numerically.\n",
"\n",
"All JSON code snippets together form the `demonstrator_rules.json` file."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Under `\"rules\"` you define the inference. The naming of the individual rules is up to you. The inputs' names must match the ones defined in `\"input_names\"`.\n",
"The following rule `and_rule` should be read like this: \"*and_rule applies IF citric_acid is average AND chlorids is low*\". The other inputs have no influence on `and_rule`."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"{\n",
" \"rules\": {\n",
" \"high_quality\": {\"free_sulfur_dioxide\": \"high\"},\n",
" \"low_quality\": {\"fixed_acidity\": \"low\"},\n",
" \"and_rule\": {\"citric_acid\": \"average\", \"chlorides\": \"low\"},\n",
" \"complex_rule\": {\n",
" \"fixed_acidity\": \"high\",\n",
" \"chlorides\": \"low\",\n",
" \"density\": \"average\",\n",
" \"citric_acid\": \"average\",\n",
" \"residual_sugar\": \"low\",\n",
" },\n",
" },\n",
"}\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
" What each rule means is defined in \"classification_rules\". In this demo: \"*IF 2_E1_r1 THEN output at index \\[1\\]*\""
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"{\n",
" \"classification_rules\": {\n",
" \"high_quality\": [2],\n",
" \"low_quality\": [2],\n",
" \"and_rule\": [1],\n",
" \"complex_rule\": [1],\n",
" },\n",
"}\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"It is important to know in which order the inputs are fed to the network. The order is defined in `\"input_names\"`. Each name is assigned at the index at which it is put into the network. The names can also be passed to the `Fuzzification` for a better overview during debugging (see [documentation](../api/fuzzy.rst#prosper_nn.models.fuzzy.fuzzification.Fuzzification) `Fuzzification`). "
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"{\n",
" \"input_names\": {\n",
" \"fixed_acidity\": 0,\n",
" \"volatile_acidity\": 1,\n",
" \"citric_acid\": 2,\n",
" \"residual_sugar\": 3,\n",
" \"chlorides\": 4,\n",
" \"free_sulfur_dioxide\": 5,\n",
" \"total_sulfur_dioxide\": 6,\n",
" \"density\": 7,\n",
" \"pH\": 8,\n",
" \"sulphates\": 9,\n",
" \"alcohol\": 10\n",
" },\n",
"};"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The meaning of the rule terms e.g. `\"high\", \"low\", \"average\"`, are stored in `\"member_activations\"`. Depending on the used membership functions you have to adjust this dictionary. The order of the membership functions in the `member_functions` dictionary of `Fuzzification` is important. Depending on the meaning of the membership functions, you can create the numerical description of corresponding relation."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"{\n",
" \"member_activations\": {\n",
" \"high\": [0, 0, 1],\n",
" \"low\": [1, 0, 0],\n",
" \"average\": [0, 1, 0],\n",
" \"high_low\": [1, 0, 1],\n",
" \"high_average\": [0, 1, 1],\n",
" \"low_average\": [1, 1, 0],\n",
" \"none\": [0, 0, 0],\n",
" }\n",
"}\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This is the JSON file for the following membership functions:"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"membership_fcts = {\n",
" \"decrease\": NormlogMembership(negative=True),\n",
" \"constant\": GaussianMembership(),\n",
" \"increase\": NormlogMembership(),\n",
"}"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Parser\n",
"The class `RuleManager` can be used to parse the JSON file and create the `rule_matrix` and the `classification_matrix`. The matrices are created when initializing a new `RuleManager` object and can be used by an attribute call. The class takes the path to the JSON file as an initialization parameter as well as the matrices' shapes. We will use the `RuleManager` in the Initialization part to parse the JSON file."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The data has eleven inputs and 3 different classes. We defined 4 rules in the `demonstrator_rules.json` file. We use three membership functions (*increase*, *stable*, *decrease*). The shape of the `rule_matrix` is defined by those parameters. They have to be passed to the `RuleManager` in order to parse the JSON rules correctly."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"n_features_input = 11\n",
"n_output_classes = 3\n",
"n_rules = 4\n",
"n_membership_fcts = 3\n",
"\n",
"n_epochs = 20\n",
"\n",
"rule_manager = RuleManager(\n",
" path=\"data/demonstrator_rules.json\",\n",
" rule_matrix_shape=(n_rules, n_features_input, n_membership_fcts),\n",
" classification_matrix_shape=(n_rules, n_output_classes),\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Membership Function Usage\n",
"For fuzzification use the `Fuzzification` class. It applies a set of membership functions to each input. The layer has a 2D output tensor of `shape=(n_features_inputs, n_memberships_fcts)`. For each input the membership to each function of the set is calculated. This set is a python dictionary containing a custom name for each membership function as keys and the membership functions themselves. The `Fuzzification` below has three membership functions. It will have `n_input = 11` input features and `11*3 = 33` output features of `shape=(11, 3)`."
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"membership_fcts = {\n",
" \"decrease\": NormlogMembership(negative=True),\n",
" \"constant\": GaussianMembership(),\n",
" \"increase\": NormlogMembership(),\n",
"}\n",
"\n",
"fuzzification = Fuzzification(\n",
" n_features_input=n_features_input, membership_fcts=membership_fcts\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### FuzzyInference Usage\n",
"For *Fuzzy-Inference* use the `FuzzyInference` class. It takes a 2D tensor as an input and outputs a 1D tensor. This fits since it follows the `Fuzzification`, which outputs a 2D tensor by default. As input parameters the `FuzzyInference` needs the number of inputs of the data set `n_features_input = 11`, the number of rules ` n_rules = 4`, the number of membership functions (`n_membership_fcts`) used in the previous `Fuzzification` and the number of output classes `n_output_classes`. The layer requires an optional `rule_matrix` parameter if rules should be used. Here, we use the `rule_matrix` of the previously created `rule_manager`. We set the parameter `learn_conditions` of the `FuzzyInference` to `True` and `prune_weights` to `False`, so all weights can change. If `prune_weights` is set to `True`, only the rule weights can change; all other stay zero. If `learn_conditions` is set to `False`, no weights change.\n",
"Similarly, we set the `classification_matrix` from the `rule_manager`."
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"fuzzy_inference = FuzzyInference(\n",
" n_features_input=n_features_input,\n",
" n_rules=n_rules,\n",
" n_output_classes=n_output_classes,\n",
" n_membership_fcts=n_membership_fcts,\n",
" rule_matrix=rule_manager.rule_matrix,\n",
" prune_weights=False,\n",
" learn_conditions=True,\n",
" classification_matrix=rule_manager.classification_matrix,\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Defuzzification Usage\n",
"For the *Defuzzification* in a regression task we use the `Defuzzification` class. As parameters this layer takes the number of classes ` n_output_classes` and the number of outputs `n_features_output`. \n",
"If we are in a classification task this layer is not needed. The consequence classes will directly be used as classification output."
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"defuzzification = Defuzzification(n_output_classes)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we can create our Fuzzy Neural Network by setting the layers in sequence. "
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"fuzzy = torch.nn.Sequential(fuzzification, fuzzy_inference, defuzzification)\n",
"fuzzy = fuzzy.double()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Training Loop\n",
"Now we can set our training parameters in the optimizer and the loss function."
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"optimizer = torch.optim.Adam(fuzzy.parameters())\n",
"loss_function = torch.nn.MSELoss()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"and write the training loop: "
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"losses = []\n",
"test_losses = []\n",
"for t in range(n_epochs):\n",
" for X_batch, y_batch in zip(X_train, y_train):\n",
" X_batch = X_batch.type(torch.DoubleTensor)\n",
" y_batch = y_batch.type(torch.DoubleTensor)\n",
" predictions = fuzzy(X_batch)\n",
" optimizer.zero_grad()\n",
" loss = loss_function(predictions, y_batch)\n",
" loss.backward()\n",
" optimizer.step()\n",
" losses.append(loss.item())\n",
"\n",
" # test set\n",
" for X_test_batch, y_test_batch in zip(X_train, y_train):\n",
" predictions = fuzzy(X_test_batch)\n",
" test_loss = loss_function(predictions, y_test_batch)\n",
" test_losses.append(test_loss.item())"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAkAAAAHHCAYAAABXx+fLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAAA9hAAAPYQGoP6dpAABphUlEQVR4nO3dd3hUVf7H8fedSTLpjZICoXcIRZRIUVCjgIqgrAKiFAEVsSKrskqxstblpyIoUnRdAWEtKIggK6iAgHTpPZQktJAKaXN/fwQGIwkkpNyUz+t57kPmzLlnvndmYz575tx7DdM0TUREREQqEZvVBYiIiIiUNgUgERERqXQUgERERKTSUQASERGRSkcBSERERCodBSARERGpdBSAREREpNJRABIREZFKRwFIREREKh0FIBEpcwYNGkSdOnXKzbhWW7ZsGYZhsGzZskLve+DAAQzDYObMmcVel0hZpgAkUkbMnDkTwzD4/fffrS5FSsAHH3ygkCFShrhZXYCISGmZOnUqTqfTktf+4IMPqFq1KoMGDSr2sa+//nrOnDmDh4dHofetXbs2Z86cwd3dvdjrEinLNAMkIhVeamoqAO7u7jgcDourubzz9RaUzWbD09MTm63w/0k3DANPT0/sdnuh9xUpzxSARMqZDRs20L17d/z9/fH19eWmm27it99+y9UnMzOTF198kYYNG+Lp6UmVKlXo1KkTS5YscfWJi4tj8ODB1KxZE4fDQVhYGD179uTAgQO5xvr++++57rrr8PHxwc/Pj9tuu42tW7fm6lPQsfLy9ddf06JFCzw9PWnRogVfffXVRX3yW+OS1/qVQYMG4evry969e7n11lvx8/Ojf//+ruf+vAbo/P5vvfUWH330EfXr18fhcHDNNdewdu3ai+qYO3cuzZo1y1VrQdYV1alTh61bt7J8+XIMw8AwDLp06QJc+Opz+fLlPPLII1SvXp2aNWsCcPDgQR555BEaN26Ml5cXVapU4e67777ofc3r/enSpQstWrRg27Zt3HDDDXh7e1OjRg3eeOONAr+HR44coVevXvj6+lKtWjVGjRpFdnZ2rv1PnjzJ/fffj7+/P4GBgQwcOJBNmzZpXZGUefoKTKQc2bp1K9dddx3+/v4888wzuLu78+GHH9KlSxeWL19OVFQUAOPHj2fChAkMHTqUdu3akZSUxO+//8769eu5+eabAejduzdbt27lscceo06dOhw7dowlS5YQExPj+oP+73//m4EDB9K1a1def/110tLSmDx5Mp06dWLDhg2ufgUZKy+LFy+md+/eNGvWjAkTJnDy5ElXkCqKrKwsunbtSqdOnXjrrbfw9va+ZP/PP/+c5ORkHnroIQzD4I033uCuu+5i3759rq+GFixYQJ8+fYiMjGTChAkkJCQwZMgQatSocdl6Jk6cyGOPPYavry/PP/88ACEhIbn6PPLII1SrVo2xY8e6ZoDWrl3LypUr6du3LzVr1uTAgQNMnjyZLl26sG3btsseV0JCAt26deOuu+7innvuYd68eTz77LNERkbSvXv3S+6bnZ1N165diYqK4q233uLHH3/k7bffpn79+gwfPhwAp9NJjx49WLNmDcOHD6dJkyZ88803DBw48LLviYjlTBEpE2bMmGEC5tq1a/Pt06tXL9PDw8Pcu3evq+3o0aOmn5+fef3117vaWrVqZd522235jpOQkGAC5ptvvplvn+TkZDMwMNAcNmxYrva4uDgzICDA1V6QsfLTunVrMywszDx9+rSrbfHixSZg1q5d29X2008/mYD5008/5dp///79JmDOmDHD1TZw4EATMJ977rmLXm/gwIG5xj2/f5UqVcxTp0652r/55hsTML/99ltXW2RkpFmzZk0zOTnZ1bZs2bKLas1P8+bNzc6dO1/Ufv5z79Spk5mVlZXrubS0tIv6r1q1ygTMTz/91NWW1/vTuXPni/qlp6eboaGhZu/evS96D/J6D1966aVcr92mTRuzbdu2rsf//e9/TcCcOHGiqy07O9u88cYbLxpTpKzRV2Ai5UR2djaLFy+mV69e1KtXz9UeFhbGvffey6+//kpSUhIAgYGBbN26ld27d+c5lpeXFx4eHixbtoyEhIQ8+yxZsoTTp0/Tr18/Tpw44drsdjtRUVH89NNPBR4rL7GxsWzcuJGBAwcSEBDgar/55ptp1qxZgcfJz/lZioLo06cPQUFBrsfXXXcdAPv27QPg6NGjbNmyhQEDBuDr6+vq17lzZyIjI4tcK8CwYcMuWofj5eXl+jkzM5OTJ0/SoEEDAgMDWb9+/WXH9PX15b777nM99vDwoF27dq7jupyHH3441+Prrrsu176LFi3C3d2dYcOGudpsNhsjRowo0PgiVlIAEiknjh8/TlpaGo0bN77ouaZNm+J0Ojl06BAAL730EqdPn6ZRo0ZERkby97//nc2bN7v6OxwOXn/9db7//ntCQkK4/vrreeONN4iLi3P1OR+ebrzxRqpVq5ZrW7x4MceOHSvwWHk5ePAgAA0bNrzoubyOsTDc3NwK9TVarVq1cj0+H4bOB7rztTZo0OCiffNquxJ169a9qO3MmTOMHTuWiIgIHA4HVatWpVq1apw+fZrExMTLjlmzZk0Mw8jVFhQUVKCg6unpSbVq1S6578GDBwkLC7voq7jiek9ESpICkEgFdP3117N3716mT59OixYt+Pjjj7nqqqv4+OOPXX2efPJJdu3axYQJE/D09GTMmDE0bdqUDRs2ALhOF//3v//NkiVLLtq++eabAo9VVH/9I37eXxfknudwOAp1RlR+Z0CZplngMYrqz7M95z322GO8+uqr3HPPPXzxxRcsXryYJUuWUKVKlQKdzl+U49JZYVLRaRG0SDlRrVo1vL292blz50XP7dixA5vNRkREhKstODiYwYMHM3jwYFJSUrj++usZP348Q4cOdfWpX78+Tz/9NE8//TS7d++mdevWvP3223z22WfUr18fgOrVqxMdHX3Z+i41Vl5q164NkOfXdH89xvMzMqdPn87Vfn5mpqSdr3XPnj0XPZdXW17yC3GXMm/ePAYOHMjbb7/tajt79uxF74NVateuzU8//URaWlquWaCCviciVtIMkEg5YbfbueWWW/jmm29ynQYdHx/P559/TqdOnfD39wdyTk3+M19fXxo0aEB6ejoAaWlpnD17Nlef+vXr4+fn5+rTtWtX/P39ee2118jMzLyonuPHjxd4rLyEhYXRunVrPvnkk1xf5yxZsoRt27bl6lu7dm3sdjs///xzrvYPPvgg3/GLU3h4OC1atODTTz8lJSXF1b58+XK2bNlSoDF8fHwKHVzsdvtFszXvvfdevjNfpa1r165kZmYydepUV5vT6WTSpEkWViVSMJoBEiljpk+fzqJFiy5qf+KJJ3jllVdYsmQJnTp14pFHHsHNzY0PP/yQ9PT0XNd3adasGV26dKFt27YEBwfz+++/M2/ePB599FEAdu3axU033cQ999xDs2bNcHNz46uvviI+Pp6+ffsC4O/vz+TJk7n//vu56qqr6Nu3L9WqVSMmJoYFCxbQsWNH3n///QKNlZ8JEyZw22230alTJx544AFOnTrFe++9R/PmzXMFjYCAAO6++27ee+89DMOgfv36fPfdd651SKXhtddeo2fPnnTs2JHBgweTkJDA+++/T4sWLXLVmp+2bdsyefJkXnnlFRo0aED16tW58cYbL7nP7bffzr///W8CAgJo1qwZq1at4scff6RKlSrFdVhF0qtXL9q1a8fTTz/Nnj17aNKkCfPnz+fUqVPAlc16iZQai89CE5Fzzp8Ond926NAh0zRNc/369WbXrl1NX19f09vb27zhhhvMlStX5hrrlVdeMdu1a2cGBgaaXl5eZpMmTcxXX33VzMjIME3TNE+cOGGOGDHCbNKkienj42MGBASYUVFR5hdffHFRXT/99JPZtWtXMyAgwPT09DTr169vDho0yPz9998LPVZe/vvf/5pNmzY1HQ6H2axZM/PLL7+86HR10zTN48ePm7179za9vb3NoKAg86GHHjL/+OOPPE/h9vHxyfO18jsNPq9T+AFz3Lhxudpmz55tNmnSxHQ4HGaLFi3M+fPnm7179zabNGly2eOMi4szb7vtNtPPz88EXKfEX+ryBwkJCebgwYPNqlWrmr6+vmbXrl3NHTt2mLVr1zYHDhzo6pffafDNmzcv8HtQkPdw3Lhx5l//bBw/fty89957TT8/PzMgIMAcNGiQuWLFChMwZ8+efdn3RcQqhmmW4io/EZEKpnXr1lSrVi3XVbYru6+//po777yTX3/9lY4dO1pdjkietAZIRKQAMjMzycrKytW2bNkyNm3a5LqtRWV05syZXI+zs7N577338Pf356qrrrKoKpHL0xogEZECOHLkCNHR0dx3332Eh4ezY8cOpkyZQmho6EUXDKxMHnvsMc6cOUP79u1JT0/nyy+/ZOXKlbz22mt5ntovUlboKzARkQJITEzkwQcfZMWKFRw/fhwfHx9uuukm/vnPf7ouGVAZff7557z99tvs2bOHs2fP0qBBA4YPH+5acC9SVikAiYiISKWjNUAiIiJS6SgAiYiISKWjRdB5cDqdHD16FD8/P13IS0REpJwwTZPk5GTCw8Mvez9ABaA8HD16NNc9lURERKT8OHToEDVr1rxkHwWgPPj5+QE5b+D5eyuJiIhI2ZaUlERERITr7/ilKADl4fzXXv7+/gpAIiIi5UxBlq9oEbSIiIhUOgpAIiIiUukoAImIiEilozVAIiJSqTidTjIyMqwuQ66Au7s7dru9WMZSABIRkUojIyOD/fv343Q6rS5FrlBgYCChoaFFvk6fApCIiFQKpmkSGxuL3W4nIiLishfKk7LFNE3S0tI4duwYAGFhYUUaTwFIREQqhaysLNLS0ggPD8fb29vqcuQKeHl5AXDs2DGqV69epK/DFH9FRKRSyM7OBsDDw8PiSqQozofXzMzMIo2jACQiIpWK7vFYvhXX56cAJCIiIpWOApCIiEglU6dOHSZOnGj5GFZSABIRESmjDMO45DZ+/PgrGnft2rU8+OCDxVtsOaOzwEqRaZrEnErDzW6jRqCX1eWIiEgZFxsb6/p5zpw5jB07lp07d7rafH19XT+bpkl2djZubpf/016tWrXiLbQc0gxQKXplwXY6v7mMT1cesLoUEREpB0JDQ11bQEAAhmG4Hu/YsQM/Pz++//572rZti8Ph4Ndff2Xv3r307NmTkJAQfH19ueaaa/jxxx9zjfvXr68Mw+Djjz/mzjvvxNvbm4YNGzJ//vxC1RoTE0PPnj3x9fXF39+fe+65h/j4eNfzmzZt4oYbbsDPzw9/f3/atm3L77//DsDBgwfp0aMHQUFB+Pj40Lx5cxYuXHjlb1wBaAaoFLWo4Q/Ab/tOWlyJiIiYpsmZzGxLXtvL3V5sZzM999xzvPXWW9SrV4+goCAOHTrErbfeyquvvorD4eDTTz+lR48e7Ny5k1q1auU7zosvvsgbb7zBm2++yXvvvUf//v05ePAgwcHBl63B6XS6ws/y5cvJyspixIgR9OnTh2XLlgHQv39/2rRpw+TJk7Hb7WzcuBF3d3cARowYQUZGBj///DM+Pj5s27Yt1+xWSVAAKkXX1qsCwB9HEkg6m4m/p7vFFYmIVF5nMrNpNvYHS15720td8fYonj/BL730EjfffLPrcXBwMK1atXI9fvnll/nqq6+YP38+jz76aL7jDBo0iH79+gHw2muv8e6777JmzRq6det22RqWLl3Kli1b2L9/PxEREQB8+umnNG/enLVr13LNNdcQExPD3//+d5o0aQJAw4YNXfvHxMTQu3dvIiMjAahXr14h3oErY/lXYJMmTaJOnTp4enoSFRXFmjVrLtl/4sSJNG7cGC8vLyIiInjqqac4e/as6/nx48dftEjs/JtttbDds1nu9QzDbN/x+4FTVpcjIiIVwNVXX53rcUpKCqNGjaJp06YEBgbi6+vL9u3biYmJueQ4LVu2dP3s4+ODv7+/67YTl7N9+3YiIiJc4QegWbNmBAYGsn37dgBGjhzJ0KFDiY6O5p///Cd79+519X388cd55ZVX6NixI+PGjWPz5s0Fet2isHQGaM6cOYwcOZIpU6YQFRXFxIkT6dq1Kzt37qR69eoX9f/888957rnnmD59Oh06dGDXrl0MGjQIwzB45513XP2aN2+e6/vOgiwIKxVZZ6ltHqaDbSu/7jvFjU1CrK5IRKTS8nK3s+2lrpa9dnHx8fHJ9XjUqFEsWbKEt956iwYNGuDl5cXf/vY3MjIyLjnO+a+jzjMMo1hvGjt+/HjuvfdeFixYwPfff8+4ceOYPXs2d955J0OHDqVr164sWLCAxYsXM2HCBN5++20ee+yxYnv9v7J0Buidd95h2LBhDB48mGbNmjFlyhS8vb2ZPn16nv1XrlxJx44duffee6lTpw633HIL/fr1u2jWyM3NLdfCsapVq5bG4Vxe3esBuNq2i9/3xllcjIhI5WYYBt4ebpZsJXk16hUrVjBo0CDuvPNOIiMjCQ0N5cCBAyX2egBNmzbl0KFDHDp0yNW2bds2Tp8+TbNmzVxtjRo14qmnnmLx4sXcddddzJgxw/VcREQEDz/8MF9++SVPP/00U6dOLdGaLQtAGRkZrFu3jujo6AvF2GxER0ezatWqPPfp0KED69atcwWeffv2sXDhQm699dZc/Xbv3k14eDj16tWjf//+l532S09PJykpKddWIqo1JdurCt5GOm6x60k6W7T7mIiIiPxVw4YN+fLLL9m4cSObNm3i3nvvLdaZnLxER0cTGRlJ//79Wb9+PWvWrGHAgAF07tyZq6++mjNnzvDoo4+ybNkyDh48yIoVK1i7di1NmzYF4Mknn+SHH35g//79rF+/np9++sn1XEmxLACdOHGC7OxsQkJyfw0UEhJCXFzesyP33nsvL730Ep06dcLd3Z369evTpUsX/vGPf7j6REVFMXPmTBYtWsTkyZPZv38/1113HcnJyfnWMmHCBAICAlzbn7/DLFY2G/Z6ObNA1xrbWLtf64BERKR4vfPOOwQFBdGhQwd69OhB165dueqqq0r0NQ3D4JtvviEoKIjrr7+e6Oho6tWrx5w5cwCw2+2cPHmSAQMG0KhRI+655x66d+/Oiy++COTcqHbEiBE0bdqUbt260ahRIz744IOSrdk0TbNEXyEfR48epUaNGqxcuZL27du72p955hmWL1/O6tWrL9pn2bJl9O3bl1deeYWoqCj27NnDE088wbBhwxgzZkyer3P69Glq167NO++8w5AhQ/Lsk56eTnp6uutxUlISERERJCYm4u/vX8Qj/Yvfp8N3T/GbsylLo6bz/G3NLr+PiIgU2dmzZ9m/fz9169bF09PT6nLkCl3qc0xKSiIgIKBAf78tWx1ctWpV7HZ7roskAcTHxxMaGprnPmPGjOH+++9n6NChAERGRpKamsqDDz7I888/j8128YRWYGAgjRo1Ys+ePfnW4nA4cDgcRTiaQqjbGYA2xm7e3HsEUAASEREpbZZ9Bebh4UHbtm1ZunSpq83pdLJ06dJcM0J/lpaWdlHIsdtzVtLnN5GVkpLC3r17CQsLK6bKiyi4Htm+4TiMLLzi1pF4RuuARERESpulZ4GNHDmSqVOn8sknn7B9+3aGDx9OamoqgwcPBmDAgAGMHj3a1b9Hjx5MnjyZ2bNns3//fpYsWcKYMWPo0aOHKwiNGjWK5cuXc+DAAVauXMmdd96J3W53XdzJcobhWgfU3rZV64BEREQsYOkFcvr06cPx48cZO3YscXFxtG7dmkWLFrkWRsfExOSa8XnhhRcwDIMXXniBI0eOUK1aNXr06MGrr77q6nP48GH69evHyZMnqVatGp06deK3334rWzd+q3s9bJ5NB9s2vtt3kuhmuh6QiIhIabJsEXRZVphFVFfkdAxMjCTLtNE3eDbznrDmQlwiIpWJFkFXDMW1CNryW2FUSoG1yAqog5vhxP/YWhLTtA5IRESkNCkAWcStfs7ZYO2Nrazer7vDi4iIlCYFIKucuy1GB9tWftunhdAiIiKlSQHIKucCUHPbQbbu2W9xMSIiIpWLApBVfKuTVaUJAMEn1nA67dJ36RURESltBw4cwDAMNm7caHUpxU4ByEJu9c9dD8jYympdD0hERP7CMIxLbuPHjy/S2F9//XWx1VreKABZKdc6IC2EFhGR3GJjY13bxIkT8ff3z9U2atQoq0sstxSArFS7IyYGDWxH2bl7t9XViIhIGRMaGuraAgICMAwjV9vs2bNp2rQpnp6eNGnSJNcd1DMyMnj00UcJCwvD09OT2rVrM2HCBADq1KkDwJ133olhGK7HBbF8+XLatWuHw+EgLCyM5557jqysLNfz8+bNIzIyEi8vL6pUqUJ0dDSpqalAzk3N27Vrh4+PD4GBgXTs2JGDBw8W/Y26ApZeCbrS8w4mOyQSt/jNVD2xhoTUWwny8bC6KhGRysE0ITPNmtd29wbDKNIQ//nPfxg7dizvv/8+bdq0YcOGDQwbNgwfHx8GDhzIu+++y/z58/niiy+oVasWhw4d4tChQwCsXbuW6tWrM2PGDLp16+a6ndTlHDlyhFtvvZVBgwbx6aefsmPHDoYNG4anpyfjx48nNjaWfv368cYbb3DnnXeSnJzML7/8gmmaZGVl0atXL4YNG8asWbPIyMhgzZo1GEV8H66UApDF3Op3hvjNdLDlrAPq1iLU6pJERCqHzDR4Ldya1/7HUfDwKdIQ48aN4+233+auu+4CoG7dumzbto0PP/yQgQMHEhMTQ8OGDenUqROGYVC7dm3XvudvDxUYGEhoaMH/7nzwwQdERETw/vvvYxgGTZo04ejRozz77LOMHTuW2NhYsrKyuOuuu1yvFxkZCcCpU6dITEzk9ttvp379+gA0bdq0SO9BUegrMKvVzbkgotYBiYhIQaWmprJ3716GDBmCr6+va3vllVfYu3cvAIMGDWLjxo00btyYxx9/nMWLFxf5dbdv30779u1zzdp07NiRlJQUDh8+TKtWrbjpppuIjIzk7rvvZurUqSQkJAAQHBzMoEGD6Nq1Kz169OD//u//iI2NLXJNV0ozQFardS1Ow41atuPs370VaG51RSIilYO7d85MjFWvXQQpKSkATJ06laioqFzPnf8666qrrmL//v18//33/Pjjj9xzzz1ER0czb968Ir32pdjtdpYsWcLKlStZvHgx7733Hs8//zyrV6+mbt26zJgxg8cff5xFixYxZ84cXnjhBZYsWcK1115bYjXlRzNAVnP4kR3WBoDqp9ZyKlXXAxIRKRWGkfM1lBVbEde9hISEEB4ezr59+2jQoEGurW7duq5+/v7+9OnTh6lTpzJnzhz++9//cupUzmVX3N3dyc7OLtTrNm3alFWrVvHn+6ivWLECPz8/atasee5tNejYsSMvvvgiGzZswMPDg6+++srVv02bNowePZqVK1fSokULPv/886K8FVdMM0BlgHv9znB0LR1sW1mz/yTdWoRZXZKIiJRxL774Io8//jgBAQF069aN9PR0fv/9dxISEhg5ciTvvPMOYWFhtGnTBpvNxty5cwkNDSUwMBDIORNs6dKldOzYEYfDQVBQ0GVf85FHHmHixIk89thjPProo+zcuZNx48YxcuRIbDYbq1evZunSpdxyyy1Ur16d1atXc/z4cZo2bcr+/fv56KOPuOOOOwgPD2fnzp3s3r2bAQMGlPA7lTfNAJUFf7oe0Ko9JywuRkREyoOhQ4fy8ccfM2PGDCIjI+ncuTMzZ850zQD5+fnxxhtvcPXVV3PNNddw4MABFi5ciM2W86f/7bffZsmSJURERNCmTZsCvWaNGjVYuHAha9asoVWrVjz88MMMGTKEF154AciZcfr555+59dZbadSoES+88AJvv/023bt3x9vbmx07dtC7d28aNWrEgw8+yIgRI3jooYdK5g26DMP88zyWAJCUlERAQACJiYn4+/uX/AtmniF7Qi3szgyG+E5m2qh7S/41RUQqmbNnz7J//37q1q2Lp6en1eXIFbrU51iYv9+aASoL3L1w1mwHQFjCGk6mpFtckIiISMWmAFRGuNfvAkB7m+4LJiIiUtIUgMqKc+uA2tu2sXrvcYuLERERqdgUgMqKGleRZfcm2Eghfs96q6sRERGp0BSAygq7O2at9gDUSFjLCa0DEhEpETr3p3wrrs9PAagMcW+Qc1uM9ratrN6ndUAiIsXp/BWSMzJ0wdnyLC0t5wa27u7uRRpHF0IsS86tA4qy7eCtvfHc1lIXRBQRKS5ubm54e3tz/Phx3N3dXdfDkfLBNE3S0tI4duwYgYGBBb6DfX4UgMqS0JZkuvvjl5nEiT1rgdZWVyQiUmEYhkFYWBj79+/n4MGDVpcjV6iwd7DPjwJQWWKzY9bpBLsXUuv07xxPHkA1P4fVVYmIVBgeHh40bNhQX4OVU+7u7kWe+TlPAaiM8WjQBXYvPHc9oJPc3jLc6pJERCoUm82mK0GLFkGXOefWAV1j28ma3bEWFyMiIlIxKQCVNdWakO6ogpeRQdKe36yuRkREpEJSACprDAOj7nUA1En+nWPJZy0uSEREpOJRACqDPBp0Ac7dFkPXAxIRESl2CkBl0bl1QG2M3fy+54jFxYiIiFQ8CkBlUXA9zniH4WFkk7ZnhdXViIiIVDgKQGWRYWCvl3NbjLrJ6ziWpHVAIiIixUkBqIw6vw6og20rq/adtLYYERGRCkYBqKw6dyZYpLGPDbtjLC5GRESkYlEAKqsCapLqWwe7YZK59xerqxEREalQFIDKMLf6OWeD1UtZT7zWAYmIiBQbBaAyzNGwC5CzDug3rQMSEREpNgpAZVmdnHVATW0xbN611+JiREREKg4FoLLMtzopAY0AyNI6IBERkWKjAFTGuZ87Hb5B6npiE89YW4yIiEgFoQBUxv15HZDuCyYiIlI8FIDKutodcWKjvi2WrTt2WF2NiIhIhaAAVNZ5BZIS3BwA88DPFhcjIiJSMSgAlQOeDXPuC9YodQNHT2sdkIiISFEpAJUDHg1uAKC9bRur9+t6QCIiIkVleQCaNGkSderUwdPTk6ioKNasWXPJ/hMnTqRx48Z4eXkRERHBU089xdmzua+SXNgxy7xa15Jt2ImwHWfH9i1WVyMiIlLuWRqA5syZw8iRIxk3bhzr16+nVatWdO3alWPHjuXZ//PPP+e5555j3LhxbN++nWnTpjFnzhz+8Y9/XPGY5YLDl+QqrQGwHdD1gERERIrK0gD0zjvvMGzYMAYPHkyzZs2YMmUK3t7eTJ8+Pc/+K1eupGPHjtx7773UqVOHW265hX79+uWa4SnsmOWFd+Ocr8Ean9nAEa0DEhERKRLLAlBGRgbr1q0jOjr6QjE2G9HR0axatSrPfTp06MC6detcgWffvn0sXLiQW2+99YrHBEhPTycpKSnXVtZ4nLsgYgfbNn7bc8LaYkRERMo5ywLQiRMnyM7OJiQkJFd7SEgIcXFxee5z77338tJLL9GpUyfc3d2pX78+Xbp0cX0FdiVjAkyYMIGAgADXFhERUcSjKwE1ryHTcFDdOM2+HeutrkZERKRcs3wRdGEsW7aM1157jQ8++ID169fz5ZdfsmDBAl5++eUijTt69GgSExNd26FDh4qp4mLk7klK9bYA2A9qHZCIiEhRuFn1wlWrVsVutxMfH5+rPT4+ntDQ0Dz3GTNmDPfffz9Dhw4FIDIyktTUVB588EGef/75KxoTwOFw4HA4inhEJc+ncReIX0nTsxs5nJBGzSBvq0sSEREplyybAfLw8KBt27YsXbrU1eZ0Olm6dCnt27fPc5+0tDRsttwl2+12AEzTvKIxyxOPhjkLoa+1bee3vVoHJCIicqUsmwECGDlyJAMHDuTqq6+mXbt2TJw4kdTUVAYPHgzAgAEDqFGjBhMmTACgR48evPPOO7Rp04aoqCj27NnDmDFj6NGjhysIXW7Mci28Dek2b4KcKcRsWwNX17K6IhERkXLJ0gDUp08fjh8/ztixY4mLi6N169YsWrTItYg5JiYm14zPCy+8gGEYvPDCCxw5coRq1arRo0cPXn311QKPWa7Z3UkJbYfj6DLcYn4B/mZ1RSIiIuWSYZqmaXURZU1SUhIBAQEkJibi7+9vdTm5pP/8fzj+N5al2W1o9NRCIoK1DkhERAQK9/e7XJ0FJuBo2AWAKNt2Vu/J/9R+ERERyZ8CUHkTEskZuz++xlmObMv/4o4iIiKSPwWg8sZmIzX8WgA8D69A32CKiIgUngJQOeTf9EYAmqdv5HCC7gsmIiJSWApA5ZBHg5zrAV1t28XqXbEWVyMiIlL+KACVR9Uak+oejKeRSfw23RZDRESksBSAyiPDIDW8IwCeR7QOSEREpLAUgMqpwOY3ARCZuYlDp7QOSEREpDAUgMopjwZdAGhj7GHNrhhrixERESlnFIDKq6A6JDrCcDeyObHtZ6urERERKVcUgMorw+BsjQ4A+BzVOiAREZHCUAAqx4KaRwMQmbmZgyfTLK5GRESk/FAAKsc8GnQGINLYz7qd+y2uRkREpPxQACrPAmpwyrMWdsPk1LZlVlcjIiJSbigAlXMZEZ0A8ItdqXVAIiIiBaQAVM4FR+asA2qVtZkDWgckIiJSIApA5ZxH/Zx1QE1th9iwfZfF1YiIiJQPCkDlnU9Vjns3BCBx208WFyMiIlI+KABVAFm1cu4LFhC/SuuARERECkABqAKoEnkzAK2ytrD/RKrF1YiIiJR9CkAVgEe9TjixUd8Wy8atW60uR0REpMxTAKoIvAI55tsEgJQdy6ytRUREpBxQAKogsmtfB0CQ1gGJiIhclgJQBVGtZc46oDbOLew7nmJxNSIiImWbAlAF4VG3A1m4UdM4waKfV1ldjoiISJmmAFRRePiQUq0NAOamWRw6patCi4iI5EcBqAIJ6PIoAINsC5mycLXF1YiIiJRdCkAViNGsJ2eqtMDXOEvtHVP540ii1SWJiIiUSQpAFYlh4NVtPAAD7IuZ/N2vOiNMREQkDwpAFU2DaNLDrsHTyOTaQ9P5efcJqysSEREpcxSAKhrDwNF1PAB97T8x87tlZDs1CyQiIvJnCkAVUZ1OZNbugruRze0Jn/D1hiNWVyQiIlKmKABVUO43jwWgl+1X5i1aytnMbIsrEhERKTsUgCqqmm3JbnQbdsPkvrOf8cnKA1ZXJCIiUmYoAFVg9ptewMTgNvsafvxpCQmpGVaXJCIiUiYoAFVkIc0wW/wNgOHZs5j00x6LCxIRESkbFIAqONsNozENOzfaN/LHb4t1iwwREREUgCq+KvWhdX8AnrTN5u0fdlhckIiIiPUUgCoBo/MzOG0eXGvbzokti3WLDBERqfQUgCqDwAhs1wwBYJTbHCYs3KZbZIiISKWmAFRZXDcSp5sXrW378N6/WLfIEBGRSk0BqLLwrY7t2uEAjHSbyz8XbNUtMkREpNJSAKpMOj6O0+FPU9shGh5fwle6RYaIiFRSCkCViVcQtg6PA/CU21z+74etukWGiIhUSgpAlc21D2N6V6GuLZ4OqUuYqVtkiIhIJaQAVNk4/DCuexqAx92+5KOftusWGSIiUumUiQA0adIk6tSpg6enJ1FRUaxZsybfvl26dMEwjIu22267zdVn0KBBFz3frVu30jiU8uHqIZh+4dQwTnJH5g+6RYaIiFQ6lgegOXPmMHLkSMaNG8f69etp1aoVXbt25dixY3n2//LLL4mNjXVtf/zxB3a7nbvvvjtXv27duuXqN2vWrNI4nPLB3ROj898BGOH2NXNX7dQtMkREpFKxPAC98847DBs2jMGDB9OsWTOmTJmCt7c306dPz7N/cHAwoaGhrm3JkiV4e3tfFIAcDkeufkFBQaVxOOVHm/sxg+pQzUjiXhbx1uKdVlckIiJSaiwNQBkZGaxbt47o6GhXm81mIzo6mlWrVhVojGnTptG3b198fHxytS9btozq1avTuHFjhg8fzsmTJ/MdIz09naSkpFxbhWd3x+gyGoCH3L7lp427dYsMERGpNCwNQCdOnCA7O5uQkJBc7SEhIcTFxV12/zVr1vDHH38wdOjQXO3dunXj008/ZenSpbz++ussX76c7t27k52d9ynfEyZMICAgwLVFRERc+UGVJ5F3Q7UmBBqpDHVbwGsLt+sWGSIiUilY/hVYUUybNo3IyEjatWuXq71v377ccccdREZG0qtXL7777jvWrl3LsmXL8hxn9OjRJCYmurZDhw6VQvVlgM0ONzwPwAP2RezYu5/lu45bXJSIiEjJszQAVa1aFbvdTnx8fK72+Ph4QkNDL7lvamoqs2fPZsiQIZd9nXr16lG1alX27Mn7bCeHw4G/v3+urdJo2gPCWuNrnGW423z++f0O3SJDREQqPEsDkIeHB23btmXp0qWuNqfTydKlS2nfvv0l9507dy7p6encd999l32dw4cPc/LkScLCwopcc4VjGHDjGAAGuC0hIe6gbpEhIiIVnuVfgY0cOZKpU6fyySefsH37doYPH05qaiqDBw8GYMCAAYwePfqi/aZNm0avXr2oUqVKrvaUlBT+/ve/89tvv3HgwAGWLl1Kz549adCgAV27di2VYyp3GtwEtdrjIJPH3L7i7cU7dYsMERGp0CwPQH369OGtt95i7NixtG7dmo0bN7Jo0SLXwuiYmBhiY2Nz7bNz505+/fXXPL/+stvtbN68mTvuuINGjRoxZMgQ2rZtyy+//ILD4SiVYyp3/jQL1NdtGW5JB3WLDBERqdAMU6f9XCQpKYmAgAASExMr13qgf98Je//Hf7M7Md7+OD///QaCfDysrkpERKRACvP32/IZIClDzs0C3WlfQUj6Ad7XLTJERKSCUgCSC2pcBU1ux4bJSLd5/HvVQd0iQ0REKiQFIMntxhcAg1vta2jo3KtbZIiISIWkACS5VW+ac4VoYJTbF3yz8ShbDusWGSIiUrEoAMnFujwHhp0b7Ju42tjBhO91iwwREalYFIDkYlXqQ5ucC0w+4z6XlXtP6BYZIiJSoSgASd46PwN2B+1s27nOtkW3yBARkQpFAUjyFlATrsm50OSzHnPZEZfEl+sPW1yUiIhI8VAAkvx1GgnuPrRgL7fYfuedJbt0iwwREakQFIAkf77V4NqHAXjW8V/iE9OYseKAtTWJiIgUAwUgubQOj4EjgPpmDLfbVjHppz1sPnza6qpERESKRAFILs0rCDo+BsA/vL7kbPpZ+k9dzbqDCRYXJiIicuUUgOTyooaDd1VCs2P5e8g6ktOzGDBtNav3nbS6MhERkSuiACSX5/CF654GYFj2HHrVySQ1I5tBM9ayYs8Ji4sTEREpPAUgKZirH4DgethS4vhX8igG1T7FmcxsHpi5VhdJFBGRckcBSArG3RMGLYTQSIzU44w7+XdG1dpDepaTYZ/8zo/b4q2uUEREpMAUgKTg/MNg8PfQ4GaMrDOMODaON2quJCPbycOfrWPRH7FWVygiIlIgCkBSOA4/6Dcb2g7GwOSeE+8zI+xLnM5sRny+gfmbjlpdoYiIyGUpAEnh2d3g9n9B9HgAbkiYx9fVPsTdeZYnZ2/gv+t0ywwRESnbFIDkyhgGdHoK/jYd7B60TP6FxcFvEmQmMmreJuasjbG6QhERkXwpAEnRtOgNA+aDVxC10raxxP9l6nKUZ/+7hX+vOmB1dSIiInlSAJKiq90ehiyBoDoEZxxlgfeLXGPsYMw3W5n2636rqxMREbmIApAUj6oNYciPUONqvLKTmeU5gR62lbz83TYmL9trdXUiIiK5KABJ8fGtBgO/hSa342Zm8p7H+wy3z+f1Rdt5d+luq6sTERFxUQCS4uXhDfd8CteOAOBZ99m85jaN/1uynbd+2IlpmhYXKCIiAm5WFyAVkM0O3V6DoNrw/bPc6/Y/wo2TjPjpcTKynYzu3gTDMKyuUkREKjHNAEnJiXoI+v4H3LzoYt/EXI+X+Obn33nx222lPhOUlpHF9tgkzUCJiAgAhqm/CBdJSkoiICCAxMRE/P39rS6n/DuyDj7vA6nHOWoG80DGM1zVrhOv9GyBzVYyM0EZWU42HjrNyr0n2LArBq+jK6lrHsHr2kE83qN9ibymiIhYqzB/vxWA8qAAVAISDsB/7oYTu0g2vXgk8wlC29zKP3u3xF4MIcjpNNkWm8TKvSdYs+sozpjfaOvcTEfbViKNfdiNnP+Z/+KMJPih72heI7DIrykiImWLAlARKQCVkDMJMPs+OPgrmaadf2QNITPyXt66uxVu9sJ9G2uaJvtOpLJy70l+2x1P0r41RGZsoqNtK1fbduEwMnP1zwysh5l4GA8zg3/5Ps3jI8cUS/ASEZGyo8QC0BtvvMFjjz2Gl5cXACtWrODqq6/G4XAAkJyczLPPPssHH3xQhPKtpwBUgrLS4ZtHYcsXALyX1YsdTR5nYr82uF8mBMUmnmHlnpOs2HOcuN0baHxmPR1sW4mybcffOJOrb6Z3CG4NumDU6wJ1r4eAmiT/+AZ+v77KKdOXH7rMp98NbUvqKEVExAIlFoDsdjuxsbFUr14dAH9/fzZu3Ei9evUAiI+PJzw8nOzs7CKUbz0FoBJmmvDTq/DzmwB8ld2RJQ1e4F/9o3C42V3dElIz+G3fSVbsPcHeXduISFxLR9tWOti2Us1IzDVklkcAtnrXY6vXGep2zrkw41/PNMvO5NS/OhCcsotvzeu4euRcwgK8SvxwRUSkdBTm73ehToP/a1bSt2dyRQwDbnwBAmvh/PZJ7rSvIHTP04yc+Qp/uy6SVXtP8seuPVQ5/hsdjK0Ms22ltu0YuF8YItvuCbXaY6/fBep1xi20Zc7p95didyewz2Sc06LpYfzCxNmf8ORDD5fooYqISNmk6wCJda4agM2/Blmz76c926gW8yjL/92KO21/8A/boVyBx2m44Qxvi1v9zlCvM/aa14Cbo9AvaYu4mlORDxC8ZRp3HXmLpZtu4aZW9YrxoEREpDxQABJrNbgJt6GLSf+0Nw3SjtLAdtT1VGa15rg3uAHqdsZWuz02h1+xvGTw7S+RuHMBtTLi+GX+eFKaTMfXoV8FEZHKpND/1f/444/x9fUFICsri5kzZ1K1alUgZxG0SKGFtsDx8E+w+AVw+OWs4al7Pe4+VUvm9Ry+ePb6P/iiD32z5vPx1/N5qM9dJfNaIiJSJhVqEXSdOnUKdAuD/fv3F6koq2kRdOVwbEZ/qh/8jj+cdXAOXUrLWiUUuEREpFToOkBFpABUSaQcI/Wdq/BxJvOx1wMMGvV2oa9HJCIiZUdh/n7rv/ZSeflWxxn9MgD90z7jv0t/tbggEREpLYUKQKtWreK7777L1fbpp59St25dqlevzoMPPkh6enqxFihSkvzaDyK+Sju8jAxqrHiew6dSrS5JRERKQaEC0EsvvcTWrVtdj7ds2cKQIUOIjo7mueee49tvv2XChAnFXqRIiTEMqvWdTAbudDI2s/Dzd3V9KxGRSqBQAWjjxo3cdNNNrsezZ88mKiqKqVOnMnLkSN59912++OKLYi9SpCTZqjUgOeppAHofn8SP67ZbXJGIiJS0QgWghIQEQkJCXI+XL19O9+7dXY+vueYaDh06VHzViZSSKreM4rh3A6oYyWQueI6ks5mX30lERMqtQgWgkJAQ1ynuGRkZrF+/nmuvvdb1fHJyMu7u7vntLlJ22d3xv2cyTgxuNZfz5RefWl2RiIiUoEIFoFtvvZXnnnuOX375hdGjR+Pt7c11113nen7z5s3Ur1+/2IsUKQ2OOu2IazIQgBv3TmDDniMWVyQiIiWlUAHo5Zdfxs3Njc6dOzN16lQ++ugjPDw8XM9Pnz6dW265pdiLFCkt4Xe+QoJbdWoZx9nzxfNkZjutLklEREpAoQJQ1apV+fnnn0lISCAhIYG77sp9+4C5c+cyfvz4QhcxadIk6tSpg6enJ1FRUaxZsybfvl26dMEwjIu22267zdXHNE3Gjh1LWFgYXl5eREdHs3v37kLXJZWQww+3O/4FwF3pX/PVwgUWFyQiIiWhUPcCe+CBBwrUb/r06QUec86cOYwcOZIpU6YQFRXFxIkT6dq1Kzt37qR69eoX9f/yyy/JyMhwPT558iStWrXi7rvvdrW98cYbvPvuu3zyySfUrVuXMWPG0LVrV7Zt24anp2eBa5PKya/l7Rz6rTsRR7+n2e9jiLm2M7Wq6YrgIiIVSaFuhWGz2ahduzZt2rS55LVSvvrqqwIXEBUVxTXXXMP7778PgNPpJCIigscee4znnnvusvtPnDiRsWPHEhsbi4+PD6ZpEh4eztNPP82oUaMASExMJCQkhJkzZ9K3b9/LjqlbYYiZHE/qO1fha6bwecCD9HvyjQLdB09ERKxTmL/fhZoBGj58OLNmzWL//v0MHjyY++67j+Dg4CsuNCMjg3Xr1jF69GhXm81mIzo6mlWrVhVojGnTptG3b198fHyAnBuxxsXFER0d7eoTEBBAVFQUq1atyjMApaen57qCdVJS0pUeklQQhl8IZ254Ed//PU2v05/w48q/cXPHKKvLEhGRYlKoNUCTJk0iNjaWZ555hm+//ZaIiAjuuecefvjhhyu6eu6JEyfIzs7OdW0hyDndPi4u7rL7r1mzhj/++IOhQ4e62s7vV5gxJ0yYQEBAgGuLiIgo7KFIBVTtuiEcCrgabyMd3x//zulU3eZFRKSiKPTNUB0OB/369WPJkiVs27aN5s2b88gjj1CnTh1SUlJKosZ8TZs2jcjISNq1a1ekcUaPHk1iYqJr08UcBQDDIKT/FDJwp725iUWz3rW6IhERKSZFuhu8zWbDMAxM0yQ7O7vQ+1etWhW73U58fHyu9vj4eEJDQy+5b2pqKrNnz2bIkCG52s/vV5gxHQ4H/v7+uTYRAI/qDTl21RMA3Hzo/1i/fY/FFYmISHEodABKT09n1qxZ3HzzzTRq1IgtW7bw/vvvExMTg6+vb6HG8vDwoG3btixdutTV5nQ6Wbp0Ke3bt7/kvnPnziU9PZ377rsvV3vdunUJDQ3NNWZSUhKrV6++7Jgieal523PEetajipHMif+OIj2r8GFfRETKlkIFoEceeYSwsDD++c9/cvvtt3Po0CHmzp3Lrbfeis12ZZNJI0eOZOrUqXzyySds376d4cOHk5qayuDBgwEYMGBArkXS502bNo1evXpRpUqVXO2GYfDkk0/yyiuvMH/+fLZs2cKAAQMIDw+nV69eV1SjVHJ2d/z+lnObjFuyfmLh159bXZGIiBRRoc4CmzJlCrVq1aJevXosX76c5cuX59nvyy+/LPCYffr04fjx44wdO5a4uDhat27NokWLXIuYY2JiLgpXO3fu5Ndff2Xx4sV5jvnMM8+QmprKgw8+yOnTp+nUqROLFi3SNYDkivk2uJa99fpTf99ntN3yEvs7dKNueDWryxIRkStUqOsADRo0qEDXQpkxY0aRirKargMkeTHPJnHqzauokn2cb33u5vZRU3VtIBGRMqQwf78LFYAqCwUgyc/xdfOp9u39ZJk2lnX+gugbb7a6JBEROacwf7+LdBaYSGVTre0d7K52C26Gk/Cf/86p5DSrSxIRkSugACRSSHXue5dkw4dm7OeXz162uhwREbkCCkAiheQeEMapjmMBuDluGus2brC4IhERKSwFIJErUPumh9jv2wZvI53sb5/ibEaW1SWJiEghKACJXAnDoNq9U0jHnXbZG1g6d5LVFYmISCEoAIlcId/wJhxoPgKA9rveZO+BgxZXJCIiBaUAJFIEje58nsPudQk2kjk0+ymcTl1VQkSkPFAAEikCw80Dj7sm4TQNupxdysIvPrS6JBERKQAFIJEiqt60I7vqDwLgxu1j+fmXZVaWIyIiBaAAJFIMmvR/i31+1+BtpFPvx6Fs37vf6pJEROQSFIBEioPdjVoPzSHeLZyaxnHSPruPYwnJVlclIiL5UAASKSZuvlXwGfgFaXjS1vyDtR8O52xmttVliYhIHhSARIqRb0QkybdNBuC2s9/y9bTX0P2GRUTKHgUgkWIWcs1dxLR6CoC7Yv/FV9/Ms7giERH5KwUgkRJQq9c4DoTegoeRzfUbRrJ8re4XJiJSligAiZQEw6DOAzOJ9WxAVSOJqt8NZkdMvNVViYjIOQpAIiXFw4eqw/5Lki2A5sZ+YmYO4UTyWaurEhERFIBESpR7lToY93xCFnZucf7Cog+fIz1LZ4aJiFhNAUikhPk1uYFT178MwL3JM/ns06k6M0xExGIKQCKloPqNIzjaoC82w+Tug+OZ98P/rC5JRKRSUwASKSXhfd8jLvAq/I0ztF05gl+27LG6JBGRSksBSKS0uHkQMnQOCe4h1LPFYswbwu7Y01ZXJSJSKSkAiZQiw7c6vgO/IB0HnYyN/D79CRJSM6wuS0Sk0lEAEill7jVbk3H7ewD0y/yaz6e+SWa20+KqREQqFwUgEQv4Xd2HE20eBWBowr+YOue/FlckIlK5KACJWKRqj5c5HnYDDiOTO3c+w9xla60uSUSk0lAAErGKzUa1gZ9yyrsuYcYp6v9vOCt2HLG6KhGRSkEBSMRKnv4EPTCPNJsvV9l2Ez97BPuPp1hdlYhIhacAJGIxo2oD3PrMJBsbd/ET3308jsQzmVaXJSJSoSkAiZQBHo1vJq3zOACGn53OB9OnkaUzw0RESowCkEgZ4dflCRIa9sbNcPLwsZeZ9NWPVpckIlJhKQCJlBWGQdA9H3A6qCVBRgrdNj/F3JXbra5KRKRCUgASKUvcPQkc/AUpHlVpbDtMwKJH+W3vcaurEhGpcBSARMoa/zB87p9FpuHOLbbf2fTv54g5mWZ1VSIiFYoCkEgZZES0w7x9IgAPMY+Z0/6P5LM6M0xEpLgoAImUUR5t7yP1qocAGJX6L9745L9kZOnMMBGR4qAAJFKG+dz2Gsnh1+FtpPPQ0ecZ+8lChSARkWKgACRSltnd8LvvU9L86lLTOMGjMU8wdua3pGdlW12ZiEi5pgAkUtZ5B+M97HvS/OtR0zjBY4eeYuyM7xSCRESKQAFIpDzwDzsXgupTwzjJk4efZOz0+ZzNVAgSEbkSCkAi5YVfKN4PLiItoCFhximeOvIUY6d/oxAkInIFFIBEyhPf6jkzQQENCTUSeProU4yZ9pVCkIhIISkAiZQ3vtXwfnARqYGNCTFO8/fYpxk37UuFIBGRQlAAEimPfKriM+x7UoOaUN04zajYpxn78TzOZCgEiYgUhAKQSHnlUyUnBAU3o5qRyDNxoxg7dS5pGVlWVyYiUuZZHoAmTZpEnTp18PT0JCoqijVr1lyy/+nTpxkxYgRhYWE4HA4aNWrEwoULXc+PHz8ewzBybU2aNCnpwxCxhncwPkMXkBrcnKpGEs8dG8W4qV8oBImIXIalAWjOnDmMHDmScePGsX79elq1akXXrl05duxYnv0zMjK4+eabOXDgAPPmzWPnzp1MnTqVGjVq5OrXvHlzYmNjXduvv/5aGocjYg3vYHyGLSC1SiRVjGRGH/s74z6aQ2q6QpCISH4sDUDvvPMOw4YNY/DgwTRr1owpU6bg7e3N9OnT8+w/ffp0Tp06xddff03Hjh2pU6cOnTt3plWrVrn6ubm5ERoa6tqqVq1aGocjYh2vIHyGfkdq1VYEGyn84/gzjP9otkKQiEg+LAtAGRkZrFu3jujo6AvF2GxER0ezatWqPPeZP38+7du3Z8SIEYSEhNCiRQtee+01srNzL/zcvXs34eHh1KtXj/79+xMTE3PJWtLT00lKSsq1iZQ7XoH4DJlParXWBBkpPH/iGV786HNSFIJERC5iWQA6ceIE2dnZhISE5GoPCQkhLi4uz3327dvHvHnzyM7OZuHChYwZM4a3336bV155xdUnKiqKmTNnsmjRIiZPnsz+/fu57rrrSE5OzreWCRMmEBAQ4NoiIiKK5yBFSpsrBLUh0Ejl+RPP8uKH/yH5bKbVlYmIlCmWL4IuDKfTSfXq1fnoo49o27Ytffr04fnnn2fKlCmuPt27d+fuu++mZcuWdO3alYULF3L69Gm++OKLfMcdPXo0iYmJru3QoUOlcTgiJcMzICcEVW9LgJHGmJPP8fKHnykEiYj8iWUBqGrVqtjtduLj43O1x8fHExoamuc+YWFhNGrUCLvd7mpr2rQpcXFxZGRk5LlPYGAgjRo1Ys+ePfnW4nA48Pf3z7WJlGue/vgM+YbUkGvwN9J44dRoXv7wU5IUgkREAAsDkIeHB23btmXp0qWuNqfTydKlS2nfvn2e+3Ts2JE9e/bgdDpdbbt27SIsLAwPD48890lJSWHv3r2EhYUV7wGIlHUOP3we+JrU0Hb4G2d44dTzvDLlExLPKASJiFj6FdjIkSOZOnUqn3zyCdu3b2f48OGkpqYyePBgAAYMGMDo0aNd/YcPH86pU6d44okn2LVrFwsWLOC1115jxIgRrj6jRo1i+fLlHDhwgJUrV3LnnXdit9vp169fqR+fiOUcvvgM/orUsGvxN84wJuEFXp0yg8Q0hSARqdzcrHzxPn36cPz4ccaOHUtcXBytW7dm0aJFroXRMTEx2GwXMlpERAQ//PADTz31FC1btqRGjRo88cQTPPvss64+hw8fpl+/fpw8eZJq1arRqVMnfvvtN6pVq1bqxydSJjh88Rn8Jakz/4bf0ZWMPT2Gl6eY/OPhBwjwdre6OhERSximaZpWF1HWJCUlERAQQGJiotYDScWRkUbqJ3fjc+RXUk0HLwW8yOiHhxDonffXxyIi5U1h/n6Xq7PARKQIPLzxGTSP1JrX42OkMy5xHK998DEJqXmfQCAiUpEpAIlUJu5e+Az8gtSIzngb6YxPHs8/J0/llEKQiFQyCkAilY27Fz4DviC11g2uEPTGBx8qBIlIpaIAJFIZuXviM2AOqbVvwsvIYHzKi7w56QOOJ6dbXZmISKlQABKprNwc+Nw/i9Q6N+NpZDI+9WX+9cG7xCWetboyEZESpwAkUpm5OfC573NS69+Kw8jixbQJTJr0NocT0qyuTESkRCkAiVR2bh743Ptv0hrdibuRzfj0N5k26Z8cOJFqdWUiIiVGAUhEwO6Gd99ppDXvi90wGZP5Lv+Z/DJ7jiVbXZmISIlQABKRHDY73r0nc6bVYGyGyfPZk/ly8li2HU2yujIRkWKnACQiF9hsePX6F2evfgSAZ8zp/PDRaDYfPm1tXSIixUwBSERyMww8b3uNsx2eBuApPuPXqU+z7sBJiwsTESk+CkAicjHDwPOWsaR3fgGAR4x5bJjxJKv2nLC4MBGR4qEAJCL5ctzwdzKiXwVgqDGf3Z+O4Oed8RZXJSJSdApAInJJHp0eJbP7OwAMsC0i7j8P8+MfRy2uSkSkaBSAROSy3KOGkNXjA5zYuMf2P1LmDOP7TYesLktE5IopAIlIgbi17Y9518dkY6eX/VeY9wDfrNtvdVkiIldEAUhECszesjf0+ZQsw53u9jX4fj2Yub/ttrosEZFCUwASkUKxN70dW7/ZZBoe3GTfQOiCQXz+yzaryxIRKRQFIBEpNFujaNwG/JcMmxfX2f+g/pLBzPzfFqvLEhEpMAUgEbkiRt3rcR/0Nel2H6JsO2i1bBAf/rDe6rJERApEAUhErphR61ocQxZwxi2ANrY9dFwxmEnfrcY0TatLExG5JAUgESma8DZ4DV1ImnswLWwHiF7zABO//lUhSETKNAUgESm60BZ4P/gDaY5qNLYdpueGobw19384nQpBIlI2KQCJSPGo1gjvhxaT6hlGPVscff94iNdnLyJbIUhEyiAFIBEpPsH18Hl4CSk+EUTYjjNo5yNM+OxbsrKdVlcmIpKLApCIFK/ACHwfWkKyX33CjFM8tPdRJsz8kowshSARKTsUgESk+PmH4ffQDyQHNqGakciImCcY/8EMTqSkW12ZiAigACQiJcW3Gn4Pfk9ScEuCjRTGnXyGDye+yB9HEq2uTEREAUhESpB3MP4PLSSlbjccRhbPZ01iw4cP8u2Gg1ZXJiKVnAKQiJQshx++98/ibMdnALjftogqX/bl3W9/0xliImIZBSARKXk2G543P0/2PZ+RbvOmg30bd669jxenzibpbKbV1YlIJaQAJCKlxt6sB46H/+c6Tf65o0/wr4mvs/d4itWliUglowAkIqWrelN8R/xMco3r8TbSGXf2Tf73/gh+2hZrdWUiUokoAIlI6fMOxu+Br0i7+hEAhhlf45zVj2k/btQ9xESkVCgAiYg17G543z6BrJ4fkml4cJN9A11+7ssrn87nTEa21dWJSAWnACQilnJr0xf3YYtJ9Qylvi2WJ/Y9xOvv/h9HTp+xujQRqcAUgETEeuFt8Hn0V5KqX4O/cYaxyS/x1f89xZp9J62uTEQqKAUgESkbfKvh/+BCUiIHYjNMHjVncWJGP2b9ut3qykSkAlIAEpGyw80D397vkt79X2Thxq321bRefA9vzv5BN1MVkWKlACQiZY4j6gHsg78jzb0KTW0xDNn+AK9N+lA3UxWRYqMAJCJlklG7Pd6P/kJScAuCjRReOPUP/v2v0fxx+LTVpYlIBaAAJCJlV0AN/If/SFKj3rgZTp7Knsaujwbw3fr9VlcmIuWcApCIlG3uXvj3m8aZG1/GiY27bMup8fXfmDT/F91MVUSumAKQiJR9hoHX9Y9j3vclZ+z+tLHt4W/r7uO1jz7RzVRF5IooAIlIuWFvcANeI5aT5N+QEOM0z8Q+zZSJL+pmqiJSaApAIlK+BNfDf8QyEut0x2Fk8czZ91jz/mAmL9pAanqW1dWJSDlheQCaNGkSderUwdPTk6ioKNasWXPJ/qdPn2bEiBGEhYXhcDho1KgRCxcuLNKYIlLOOHwJGPA5qR2eBaCfsZg+q27nw9dH8u9fduqaQSJyWZYGoDlz5jBy5EjGjRvH+vXradWqFV27duXYsWN59s/IyODmm2/mwIEDzJs3j507dzJ16lRq1KhxxWOKSDlls+Fzyz9w3juPZN96BBspjHR+wg0/3spbb4zlmw0xOLVIWkTyYZimadl/IaKiorjmmmt4//33AXA6nURERPDYY4/x3HPPXdR/ypQpvPnmm+zYsQN3d/diGTMvSUlJBAQEkJiYiL+//xUenYiUmuwssjb8h/Qlr+CTnvN/dnY6azLLbzCde9xPl8bVMQzD4iJFpKQV5u+3ZTNAGRkZrFu3jujo6AvF2GxER0ezatWqPPeZP38+7du3Z8SIEYSEhNCiRQtee+01srOzr3hMgPT0dJKSknJtIlKO2N1wu3ogPqM2k3Hji5x186ex7TDjU1/G7/PbGf/+x6yPSbC6ShEpQywLQCdOnCA7O5uQkJBc7SEhIcTFxeW5z759+5g3bx7Z2dksXLiQMWPG8Pbbb/PKK69c8ZgAEyZMICAgwLVFREQU8ehExBLuXnhc/ySeT2/hTNQTZNocXG3bxYsnR5Ew9U5e+ngOe47pjDERKQOLoAvD6XRSvXp1PvroI9q2bUufPn14/vnnmTJlSpHGHT16NImJia7t0KFDxVSxiFjCKxCv7i/h/tRmUiIHkI2dm+wbeOHQQ2x5rw///HwRsYlnrK5SRCzkZtULV61aFbvdTnx8fK72+Ph4QkND89wnLCwMd3d37Ha7q61p06bExcWRkZFxRWMCOBwOHA5HEY5GRMokv1B8e78HXZ4k6fvx+O+Zz532X8nYuYrZO24moe3jDLz5GgK9PayuVERKmWUzQB4eHrRt25alS5e62pxOJ0uXLqV9+/Z57tOxY0f27NmD03nhFNddu3YRFhaGh4fHFY0pIpVAlfr43/dveHAZieHX4WFkM8C2iKHr72TWG8P5aMlGzmRkW12liJQiS78CGzlyJFOnTuWTTz5h+/btDB8+nNTUVAYPHgzAgAEDGD16tKv/8OHDOXXqFE888QS7du1iwYIFvPbaa4wYMaLAY4pIJRbehoAHv8Mc8A1JwS3xMdIZzjx6/3o7k//5NLNW7iYzW9cQEqkMLPsKDKBPnz4cP36csWPHEhcXR+vWrVm0aJFrEXNMTAw224WMFhERwQ8//MBTTz1Fy5YtqVGjBk888QTPPvtsgccUETHqdcH/sZ9xbv2G1O/HUSX1ACOdMzj8w7e8tbw/LW99kFtb1tCp8yIVmKXXASqrdB0gkUokO4vMdZ+SsfQ1fNKPA7DDGcHcgMHceMcAOjasZnGBIlJQhfn7rQCUBwUgkUooI430VVMwf34Hz+xkANY6G7Eo9GE63ng7LWoEUN3P0+IiS1d6VjanUjM4mZLBqdSc7WRqBqdS012PA7zc6X1VTdrVDdaMmVhOAaiIFIBEKrEzCaT99Dbuaz/E3cwAYJOzHpud9Tjg0ZCM6i3xr9WSJjWCaRbuT90qPthsZf8Pv2mapGZkcyolg5Op6SSk5RVs/hRwUjJILcTC8AbVfenXrha9r6qhs+rEMgpARaQAJCIkHSVl8at4/TELO7mDQLrpznazFlucddlpq09alRb41oqkaY0qNAvzp3GoH57u9nwGLn6maXIyNYMjCWc4cvqM69/DCWeITTzjCjZXcpNYN5tBkI8HVXw8CPbxyPVzsI8HW48kMX/TUc5k5rxHHm42bo8Mo19ULa6uHaRZISlVCkBFpAAkIi6JRyBmFZmHN3D24O84TvyBR9bFV5POCUUR/OGsyx9mPU4HNsO7ZmTOTFFYAM3C/Qn2ubKZkWynSXzS2b+EmzQOn/v56OkznM28ONzYcOLLGWw4MTCxYeLpZhDs7UawtxtB3u45m5cbQV52grzcCfCyE+hlJ9DTjUAvd3wdBoZpAiaYznObeW5zgl8oyZ6hfL3xKJ+vjmF77IVbCTUKyZkVuqtNTQK8875/o0hxUgAqIgUgEcmX0wkJ++HoBpxHN3I2Zh3ux7bgnpl8Udd0043tZi3+cNZli1mPWK/GeNZoTuNzM0XNwv2JCPIm0+kk9vSFgHM4IY3Dfwo7cYlnyXKaOMggiGSqGMkEG0nnfk4i2Mj5N8wtlepuKVQxkvF3JuGVlYhBKfwnvl4XuHoIZqNubIpN4/PVB/l2U6xrVsjhZuO2lmH0j6rFVbU0KyQlRwGoiBSARKRQ/hSKzKMbyTy8HiNuc76haIcrFNVlj60uzuxMgkjOCTIkEWScCzkkEWwkEXzuOV/jbDEUa4BhA+P8v7Y/tZ1vN/Jp/0t/gMRDcD5k+YZC24Fw1QCSHCF8veEIn6+OYUfchfehcYgf90bVolebGgR4aVZIipcCUBEpAIlIkZ0PRbEb4ehGso5sgKMbccsjFBWKzQ28q4B3VfCp8qefq577ucq5n8899grM2ed8iCluCQdh3UzY8G9IzbmMAIYNGnXLmRWqfwMbDifx+eoYvtt81PVVnae7jdtbhnNvVC3aRARqVkiKhQJQESkAiUiJME3XTBFHN+I8uhEzfhu4e2HzrYpxqSBzvt0zoGSCTFFlZcCO7+D36XDglwvtgbWh7SBocz+J9kC+Wn+Yz9fEsCv+wjqqJqF+9I+qRc82NfD31KyQXDkFoCJSABIRKYLju2DdDNj4HzibmNNmc4dmd8DVD2DW6sD6Q6f5z+oYFmyOJf3c2Wle7nZ6tArj3qjatKoZoFkhKTQFoCJSABIRKQYZabD1q5xZoSO/X2iv2hiufgBa9SXR9OHLDYf5fHUMu49dmBVqFubPvVG1uKN1uGaFpMAUgIpIAUhEpJjFbsoJQpvnQmZqTpubF7ToDdc8gBnWht9jTjNrdQzfbYl1XbPIw81GdNPq3NGqBl0aVyvV6ytJ+aMAVEQKQCIiJeRsEmz5AtZOh2NbL7SHtcqZFYq8m9NZ7vx3/RFmr8k9K+Tn6Ua35qH0bF2D9vWrYC8HV+CW0qUAVEQKQCIiJcw04dCanFmhrV9BdnpOu8MfWvbJWStUvSnbYpOYv/Eo3246ytHEC5cBqObn4PaWYfRsXUPrhcRFAaiIFIBEREpR6knY9HlOGDq170J7xLU51xVqegdOdx9+P5jANxuPsGBLLKfTMl3dalfxpmercO5oHU6D6n4WHICUFQpARaQAJCJiAacT9i/PCUI7FoB57h5s7j45Z5C16gt1rifDCb/uOc43G4+yeGu864rTAM3D/enZOpwercIJC/Cy6EDEKgpARaQAJCJisaRY2PgZbJwFp/ZeaPevCS3vgVb9oFoj0jKyWLItnvkbj7J813GynDl/0gwD2tUJpmfrGnRvEUrQFd6HTcoXBaAiUgASESkjTBMOr4WNn8PWLy9cVwigRtucINSiN3gHk5CawcI/Yvlm41HW7D/l6uZuN7i+YTXuaB3Ozc1C8PZws+BApDQoABWRApCISBmUeRZ2fQ+bZsPuJRe+IrO5Q6Ou0PpeaHAzuHlw9PQZvt10lG82HmXbn+5Q7+Vu55bmIfRsHc51DavhbrdZdDBSEhSAikgBSESkjEs5DlvmwqZZELf5Qrt3FWjxt5z1QuFtwDDYHZ/M/HNhKOZUmqtrkLc7vdrUoF+7WjQK0eLpikABqIgUgEREypH4rTlBaPMXkBJ/ob1ak5wg1LIP+IdjmiYbD53mm41H+W5zLCdS0l1dr64dRL92tbg1MgwvD11ssbxSACoiBSARkXIoOwv2Lcs5pX7HAsg6f90gA+p1yfmKrMlt4OFDVraTX/acYPaaGH7cfozsc4un/TzduKtNDfpF1aJJqP77X94oABWRApCISDl3NhG2fp2zXihm5YV2D19o1jNn8XTtjmCzcSzpLHPXHWbWmhgOJ5xxdW1TK5B+7Wpxe8swLZwuJxSAikgBSESkAjm1L+frsU2zIOHAhfaAWudOqe8LVRvidJr8uucEs9bEsGRbvOuUej+HG73a1KBvuwiahwdYcwxSIApARaQAJCJSAZkmxPyW8xXZ1q8h/cLZYa5T6pvfBT5VOJ6czrx1h5m9NoaDJy8snG5VM4B+7WrRo1U4Pg7NCpU1CkBFpAAkIlLBZZ7JWSe0eQ7sWfqnU+rdoGFXaNUHGnXDafNg1b6TfL4mhsVb48jMzvmT6eNhp2ebGtzbrhYtamhWqKxQACoiBSARkUok5RhsmQebZ0PspgvtngE5M0Kt+kJEFCdSM/hy/WFmrTnE/hOprm6RNXJmhe5oHY6vZoUspQBURApAIiKV1LHtOQunt8yFpCMX2oPq5pxO36oPZlBdftt3illrYlj0RxwZ2U4AvD3s3NEqnH7tatFSd6i3hAJQESkAiYhUcs5sOPALbJoD276BzAszPkRE5cwKNb+TU06fc7NCMew9fqFPszB/7rqqBo1C/IgI9iY80BOHm64vVNIUgIpIAUhERFwyUnPWC22aDft+AjNnxge7BzTqBq36YjaIZu2hVGatiWHBllgyspy5hjAMCPHzpGaQFxHB3jn/Bnm7HocGeOq2HMVAAaiIFIBERCRPyXHnbsExG+L/uNDuFZxzU9ZWfTkdFMmXG46yYs8JDiWkcTjhDGkZ2Zcc1mZAWIBXvgEpxN8Tu01fqV2OAlARKQCJiMhlxf2Rs3B681xIibvQXqUBtOwLLe6CoLqYhsGp1AwOJ5xxBaJDp9JyPf7rjNFfudsNwgNzAlLNQG8igr0IDfAi2MedIG8Pgn08CPLxwM/hVqnXHikAFZECkIiIFJgz+9wtOGbDju8g88J1g3DzzFlAHVwPqtTL+ff85l8DbHacTpMTKekcSjjD4YS0i8LRkYQzrosyXo6bzSDIx4Pgc6EoJxi5E+ydE5CCfTxyBaZgb48Kde8zBaAiUgASEZErkp4M27/Luer0wRXgzMq/r90DgurkDkXnt4AIsOecUp/tNIlPOpsrGB06dYbjKekkpGZw6tx2JvPSX7Plx9PdlmdAcrjZsNkMbAbYDePczwZ2m4Fxri3nZwO7wYWfz+1jc/2cs6/dONd+rq1uVR8aVPe9oprzowBURApAIiJSZNmZkHgo51Ycp/af+/fclnAAsjPy39fmBoG1Lw5GVepDYC2wu1+0y5mMbBLScsKQ69/UDE6lZZ77N8MVmM4/f/7CjlYY3qU+z3ZrUqxjFubvt67YJCIiUhLs7heCy185s3OuM3RqH5zcmzskJezPuZP9qb05218ZdgiMAP+a4BUIXkHgFYiXVxBeXkGEewXltIUEgWcgeIWCwy/nVLQ/MU2T1IzsC7NIfwlIGVlOsp3gNE3Xlu0Ep/Pcz6aJaebMUOX8bOb87MwZO9s0cZo5/bOdfx4jpz080KtE3vaC0gxQHjQDJCIilnE6ITk294zRnwPSn69JVFCG/VxQCnIFptyPz4elPz12+IK7F7j7uL6OK+s0AyQiIlJe2WwQUCNnq3td7udME1Lic2aNUuLhTMKF7expOHM6d9uZhJzZJDMb0k7kbFfC7gHu3uDhc+5f75xg5O514WcP79x9XP3y2c87OGdmyiIKQCIiIuWFYYBfaM5WUJln8g5GZ/NoO5Nwru9pyEi+cNHH7Iyc7ezp4juWDo/BLa8U33iFpAAkIiJSkbl75Wz+YYXbzzQhKz3ntP7MNMhIy/n6LeP849Tc/2ae+VNbfn3TLozn7lMyx1tACkAiIiJyMcMAd8+cjeDiH9/iJci68YiIiIiUPouvWK0AJCIiIpWOApCIiIhUOgpAIiIiUukoAImIiEilowAkIiIilU6ZCECTJk2iTp06eHp6EhUVxZo1a/LtO3PmTAzDyLV5enrm6jNo0KCL+nTr1q2kD0NERETKCcuvAzRnzhxGjhzJlClTiIqKYuLEiXTt2pWdO3dSvXr1PPfx9/dn586drsdGHqfSdevWjRkzZrgeOxyO4i9eREREyiXLZ4Deeecdhg0bxuDBg2nWrBlTpkzB29ub6dOn57uPYRiEhoa6tpCQkIv6OByOXH2CgoJK8jBERESkHLE0AGVkZLBu3Tqio6NdbTabjejoaFatWpXvfikpKdSuXZuIiAh69uzJ1q1bL+qzbNkyqlevTuPGjRk+fDgnT57Md7z09HSSkpJybSIiIlJxWRqATpw4QXZ29kUzOCEhIcTFxeW5T+PGjZk+fTrffPMNn332GU6nkw4dOnD48GFXn27duvHpp5+ydOlSXn/9dZYvX0737t3Jzs7Oc8wJEyYQEBDg2iIiIorvIEVERKTMMUzTuptxHD16lBo1arBy5Urat2/van/mmWdYvnw5q1evvuwYmZmZNG3alH79+vHyyy/n2Wffvn3Ur1+fH3/8kZtuuumi59PT00lPT3c9TkpKIiIigsTERPz9/a/gyERERKS0JSUlERAQUKC/35bOAFWtWhW73U58fHyu9vj4eEJDQws0hru7O23atGHPnj359qlXrx5Vq1bNt4/D4cDf3z/XJiIiIhWXpQHIw8ODtm3bsnTpUleb0+lk6dKluWaELiU7O5stW7YQFhaWb5/Dhw9z8uTJS/YRERGRysPy0+BHjhzJwIEDufrqq2nXrh0TJ04kNTWVwYMHAzBgwABq1KjBhAkTAHjppZe49tpradCgAadPn+bNN9/k4MGDDB06FMhZIP3iiy/Su3dvQkND2bt3L8888wwNGjSga9euBarp/LeCWgwtIiJSfpz/u12Q1T2WB6A+ffpw/Phxxo4dS1xcHK1bt2bRokWuhdExMTHYbBcmqhISEhg2bBhxcXEEBQXRtm1bVq5cSbNmzQCw2+1s3ryZTz75hNOnTxMeHs4tt9zCyy+/XOBrASUnJwNoMbSIiEg5lJycTEBAwCX7WLoIuqxyOp0cPXoUPz+/PC+yWBTnF1gfOnSowq810rFWXJXpeHWsFVdlOt7KcqymaZKcnEx4eHiuyZO8WD4DVBbZbDZq1qxZoq9RmRZb61grrsp0vDrWiqsyHW9lONbLzfycZ/mVoEVERERKmwKQiIiIVDoKQKXM4XAwbty4SnFzVh1rxVWZjlfHWnFVpuOtTMdaUFoELSIiIpWOZoBERESk0lEAEhERkUpHAUhEREQqHQUgERERqXQUgErApEmTqFOnDp6enkRFRbFmzZpL9p87dy5NmjTB09OTyMhIFi5cWEqVXrkJEyZwzTXX4OfnR/Xq1enVqxc7d+685D4zZ87EMIxcm6enZylVfOXGjx9/Ud1NmjS55D7l8TM9r06dOhcdr2EYjBgxIs/+5elz/fnnn+nRowfh4eEYhsHXX3+d63nTNBk7dixhYWF4eXkRHR3N7t27LztuYX/nS8OljjUzM5Nnn32WyMhIfHx8CA8PZ8CAARw9evSSY17J70JpudxnO2jQoItq79at22XHLW+fLZDn769hGLz55pv5jlmWP9uSogBUzObMmcPIkSMZN24c69evp1WrVnTt2pVjx47l2X/lypX069ePIUOGsGHDBnr16kWvXr34448/Srnywlm+fDkjRozgt99+Y8mSJWRmZnLLLbeQmpp6yf38/f2JjY11bQcPHiylioumefPmuer+9ddf8+1bXj/T89auXZvrWJcsWQLA3Xffne8+5eVzTU1NpVWrVkyaNCnP59944w3effddpkyZwurVq/Hx8aFr166cPXs23zEL+ztfWi51rGlpaaxfv54xY8awfv16vvzyS3bu3Mkdd9xx2XEL87tQmi732QJ069YtV+2zZs265Jjl8bMFch1jbGws06dPxzAMevfufclxy+pnW2JMKVbt2rUzR4wY4XqcnZ1thoeHmxMmTMiz/z333GPedtttudqioqLMhx56qETrLG7Hjh0zAXP58uX59pkxY4YZEBBQekUVk3HjxpmtWrUqcP+K8pme98QTT5j169c3nU5nns+X188VML/66ivXY6fTaYaGhppvvvmmq+306dOmw+EwZ82ale84hf2dt8JfjzUva9asMQHz4MGD+fYp7O+CVfI63oEDB5o9e/Ys1DgV5bPt2bOneeONN16yT3n5bIuTZoCKUUZGBuvWrSM6OtrVZrPZiI6OZtWqVXnus2rVqlz9Abp27Zpv/7IqMTERgODg4Ev2S0lJoXbt2kRERNCzZ0+2bt1aGuUV2e7duwkPD6devXr079+fmJiYfPtWlM8Ucv43/dlnn/HAAw9c8sbA5fVz/bP9+/cTFxeX67MLCAggKioq38/uSn7ny6rExEQMwyAwMPCS/Qrzu1DWLFu2jOrVq9O4cWOGDx/OyZMn8+1bUT7b+Ph4FixYwJAhQy7btzx/tldCAagYnThxguzsbEJCQnK1h4SEEBcXl+c+cXFxhepfFjmdTp588kk6duxIixYt8u3XuHFjpk+fzjfffMNnn32G0+mkQ4cOHD58uBSrLbyoqChmzpzJokWLmDx5Mvv37+e6664jOTk5z/4V4TM97+uvv+b06dMMGjQo3z7l9XP9q/OfT2E+uyv5nS+Lzp49y7PPPku/fv0ueaPMwv4ulCXdunXj008/ZenSpbz++ussX76c7t27k52dnWf/ivLZfvLJJ/j5+XHXXXddsl95/myvlO4GL0U2YsQI/vjjj8t+X9y+fXvat2/vetyhQweaNm3Khx9+yMsvv1zSZV6x7t27u35u2bIlUVFR1K5dmy+++KJA/6+qPJs2bRrdu3cnPDw83z7l9XOVHJmZmdxzzz2YpsnkyZMv2bc8/y707dvX9XNkZCQtW7akfv36LFu2jJtuusnCykrW9OnT6d+//2VPTCjPn+2V0gxQMapatSp2u534+Phc7fHx8YSGhua5T2hoaKH6lzWPPvoo3333HT/99BM1a9Ys1L7u7u60adOGPXv2lFB1JSMwMJBGjRrlW3d5/0zPO3jwID/++CNDhw4t1H7l9XM9//kU5rO7kt/5suR8+Dl48CBLliy55OxPXi73u1CW1atXj6pVq+Zbe3n/bAF++eUXdu7cWejfYSjfn21BKQAVIw8PD9q2bcvSpUtdbU6nk6VLl+b6f8h/1r59+1z9AZYsWZJv/7LCNE0effRRvvrqK/73v/9Rt27dQo+RnZ3Nli1bCAsLK4EKS05KSgp79+7Nt+7y+pn+1YwZM6hevTq33XZbofYrr59r3bp1CQ0NzfXZJSUlsXr16nw/uyv5nS8rzoef3bt38+OPP1KlSpVCj3G534Wy7PDhw5w8eTLf2svzZ3vetGnTaNu2La1atSr0vuX5sy0wq1dhVzSzZ882HQ6HOXPmTHPbtm3mgw8+aAYGBppxcXGmaZrm/fffbz733HOu/itWrDDd3NzMt956y9y+fbs5btw4093d3dyyZYtVh1Agw4cPNwMCAsxly5aZsbGxri0tLc3V56/H+uKLL5o//PCDuXfvXnPdunVm3759TU9PT3Pr1q1WHEKBPf300+ayZcvM/fv3mytWrDCjo6PNqlWrmseOHTNNs+J8pn+WnZ1t1qpVy3z22Wcveq48f67Jycnmhg0bzA0bNpiA+c4775gbNmxwnfn0z3/+0wwMDDS/+eYbc/PmzWbPnj3NunXrmmfOnHGNceONN5rvvfee6/HlfuetcqljzcjIMO+44w6zZs2a5saNG3P9Dqenp7vG+OuxXu53wUqXOt7k5GRz1KhR5qpVq8z9+/ebP/74o3nVVVeZDRs2NM+ePesaoyJ8tuclJiaa3t7e5uTJk/Mcozx9tiVFAagEvPfee2atWrVMDw8Ps127duZvv/3meq5z587mwIEDc/X/4osvzEaNGpkeHh5m8+bNzQULFpRyxYUH5LnNmDHD1eevx/rkk0+63peQkBDz1ltvNdevX1/6xRdSnz59zLCwMNPDw8OsUaOG2adPH3PPnj2u5yvKZ/pnP/zwgwmYO3fuvOi58vy5/vTTT3n+7/b88TidTnPMmDFmSEiI6XA4zJtuuumi96B27drmuHHjcrVd6nfeKpc61v379+f7O/zTTz+5xvjrsV7ud8FKlzretLQ085ZbbjGrVatmuru7m7Vr1zaHDRt2UZCpCJ/teR9++KHp5eVlnj59Os8xytNnW1IM0zTNEp1iEhERESljtAZIREREKh0FIBEREal0FIBERESk0lEAEhERkUpHAUhEREQqHQUgERERqXQUgERERKTSUQASESkAwzD4+uuvrS5DRIqJApCIlHmDBg3CMIyLtm7dulldmoiUU25WFyAiUhDdunVjxowZudocDodF1YhIeacZIBEpFxwOB6Ghobm2oKAgIOfrqcmTJ9O9e3e8vLyoV68e8+bNy7X/li1buPHGG/Hy8qJKlSo8+OCDpKSk5Oozffp0mjdvjsPhICwsjEcffTTX8ydOnODOO+/E29ubhg0bMn/+/JI9aBEpMQpAIlIhjBkzht69e7Np0yb69+9P37592b59OwCpqal07dqVoKAg1q5dy9y5c/nxxx9zBZzJkyczYsQIHnzwQbZs2cL8+fNp0KBBrtd48cUXueeee9i8eTO33nor/fv359SpU6V6nCJSTKy+G6uIyOUMHDjQtNvtpo+PT67t1VdfNU3TNAHz4YcfzrVPVFSUOXz4cNM0TfOjjz4yg4KCzJSUFNfzCxYsMG02m+uO4OHh4ebzzz+fbw2A+cILL7gep6SkmID5/fffF9txikjp0RogESkXbrjhBiZPnpyrLTg42PVz+/btcz3Xvn17Nm7cCMD27dtp1aoVPj4+ruc7duyI0+lk586dGIbB0aNHuemmmy5ZQ8uWLV0/+/j44O/vz7Fjx670kETEQgpAIlIu+Pj4XPSVVHHx8vIqUD93d/dcjw3DwOl0lkRJIlLCtAZIRCqE33777aLHTZs2BaBp06Zs2rSJ1NRU1/MrVqzAZrPRuHFj/Pz8qFOnDkuXLi3VmkXEOpoBEpFyIT09nbi4uFxtbm5uVK1aFYC5c+dy9dVX06lTJ/7zn/+wZs0apk2bBkD//v0ZN24cAwcOZPz48Rw/fpzHHnuM+++/n5CQEADGjx/Pww8/TPXq1enevTvJycmsWLGCxx57rHQPVERKhQKQiJQLixYtIiwsLFdb48aN2bFjB5Bzhtbs2bN55JFHCAsLY9asWTRr1gwAb29vfvjhB5544gmuueYavL296d27N++8845rrIEDB3L27Fn+9a9/MWrUKKpWrcrf/va30jtAESlVhmmaptVFiIgUhWEYfPXVV/Tq1cvqUkSknNAaIBEREal0FIBERESk0tEaIBEp9/RNvogUlmaAREREpNJRABIREZFKRwFIREREKh0FIBEREal0FIBERESk0lEAEhERkUpHAUhEREQqHQUgERERqXQUgERERKTS+X8KfLWxZufUVQAAAABJRU5ErkJggg==",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.plot(range(n_epochs), losses, test_losses)\n",
"plt.title(\"Losses during training\")\n",
"plt.legend(['Train loss', 'Test loss'])\n",
"plt.xlabel(\"Epoch\")\n",
"plt.ylabel(\"MSE\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Evaluation\n",
"\n",
"For Fuzzy networks two weight matrices are of interest: The condition matrix and the consequence matrix. Both have prior knowledge infused into them. By visualizing those weights, we can gain knowledge about our problem and maybe change some rules and relations."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Normally, the rule matrix should not change or only alter the one-initialized weights. By turning off pruning and allowing all weights to be trained (`prune_weights=False`, `learn_conditions=True`), we can check if our predefined rules are good or bad."
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAxgAAANXCAYAAAC2c/ndAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/Z1A+gAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3RUxR7A8e9uSA/pkJCQkN5ooROqIF2Q3lGaSJcmvfcqIE0EpVelF2kiICiIUhVI6EhLIMluAgmpu++P4MLKBnyaZJPw+5xzz/POnZmdyRvu3dkpV6HVarUIIYQQQgghRBZQGrsAQgghhBBCiPxDOhhCCCGEEEKILCMdDCGEEEIIIUSWkQ6GEEIIIYQQIstIB0MIIYQQQgiRZaSDIYQQQgghhMgy0sEQQgghhBBCZBnpYAghhBBCCCGyjHQwhBBCCCGEEFlGOhhCCJEFFAoFEyZMMHYxst3bUk8hhBD/nnQwhBBvlVWrVqFQKHRHgQIFcHd3p0uXLty/f9/YxdPx8vJCoVBQp04dg9eXL1+uq8Nvv/32f+f/888/M2HCBNRq9X8sqRBCCKGvgLELIIQQxjBp0iS8vb1JSkri1KlTrFq1ihMnTvDHH39gYWFh7OIBYGFhwZEjR4iMjMTV1VXv2vr167GwsCApKelf5f3zzz8zceJEunTpgr29/T9O9+zZMwoUkEeHEEKIzMkIhhDirdSwYUM6derERx99xFdffcWnn37KjRs32LVrl7GLplO1alVsbGzYvHmzXvi9e/c4fvw47733Xo6UQ6PR6DoyFhYW0sEQQgjxWtLBEEIIoHr16gDcuHFDF/bOO+/wzjvvvBK3S5cueHl5vTHP+/fv061bN1xcXDA3N6d48eKsWLHiH5fJwsKCFi1asGHDBr3wjRs34uDgQP369V9Jc/HiRbp06YKPjw8WFha4urrSrVs3YmJidHEmTJjA0KFDAfD29tZNtbp9+zaQsc6iX79+rF+/nuLFi2Nubs7+/ft11/5ag/Hs2TOCgoIICgri2bNnuvxjY2MpUqQIVapUIT09/R/XVwghRP4gP0MJIQTovlw7ODhkSX5RUVFUrlxZ92W9UKFC7Nu3j+7duxMfH8/AgQP/UT4dOnSgXr163LhxA19fXwA2bNhAq1atMDU1fSX+oUOHuHnzJl27dsXV1ZVLly6xbNkyLl26xKlTp1AoFLRo0YKrV6+yceNG5s2bh7OzMwCFChXS5fPDDz/wzTff0K9fP5ydnQ12qCwtLVm9ejVVq1Zl9OjRzJ07F4C+ffsSFxfHqlWrMDEx+T//ckIIIfI66WAIId5KcXFxREdHk5SUxC+//MLEiRMxNzencePGWZL/6NGjSU9P5/fff8fJyQmAXr160b59eyZMmEDPnj2xtLR8Yz61a9fG1dWVjRs3MmbMGK5cucL58+f5/PPPuXnz5ivx+/Tpw5AhQ/TCKleuTPv27Tlx4gTVq1enVKlSlC1blo0bN9KsWTODnYeIiAh+//13QkJCXlu+SpUqMWzYMGbOnEnz5s2Jiopi06ZNzJ8/n4CAgDfWTwghRP4jU6SEEG+lOnXqUKhQITw8PGjVqhXW1tbs2rWLokWL/ue8tVotW7dupUmTJmi1WqKjo3VH/fr1iYuL4+zZs/8oLxMTE9q0acPGjRuBjMXdHh4euildf/dypyUpKYno6GgqV64M8I8/E6BmzZpv7Fz8ZcKECRQvXpzOnTvTp08fatasySeffPKPP0sIIUT+Ih0MIcRbafHixRw6dIgtW7bQqFEjoqOjMTc3z5K8Hz9+jFqtZtmyZRQqVEjv6Nq1KwCPHj36x/l16NCBy5cvc+HCBTZs2EC7du1QKBQG48bGxjJgwABcXFywtLSkUKFCeHt7AxmjNv/UX2n+CTMzM1asWMGtW7d48uQJK1euzLR8Qggh8j+ZIiWEeCtVrFiR8uXLA9CsWTOqVatGhw4diIiIwMbGBshY0KzVal9J+6aFyxqNBoBOnTrRuXNng3FKlSr1j8taqVIlfH19GThwILdu3aJDhw6Zxm3Tpg0///wzQ4cOJTQ0FBsbGzQaDQ0aNNCV65/4J9O3XnbgwAEgY9Tk2rVr/1cHRQghRP4iHQwhxFvPxMSE6dOnU6tWLRYtWsSIESOAjAXfhtY53Llz57X5FSpUiIIFC5Kenp7pi/L+X+3bt2fKlCkEBwcTGhpqMI5KpeLw4cNMnDiRcePG6cKvXbv2StysHGG4ePEikyZNomvXrpw/f56PPvqI33//HTs7uyz7DCGEEHmHTJESQggytqStWLEi8+fP173zwdfXl/DwcB4/fqyLd+HCBX766afX5mViYkLLli3ZunUrf/zxxyvXX87vn/roo48YP348n3322Ws/F3hl1GX+/PmvxLW2tgb4z2/yTk1NpUuXLri5ufH555+zatUqoqKiGDRo0H/KVwghRN4lIxhCCPHc0KFDad26NatWraJXr15069aNuXPnUr9+fbp3786jR49YunQpxYsXJz4+/rV5zZgxgyNHjlCpUiV69OhBSEgIsbGxnD17lu+//57Y2Nj/q2zFihXTvX8iM7a2ttSoUYNZs2aRmpqKu7s7Bw8e5NatW6/ELVeuHJCx21W7du0wNTWlSZMmuo7HPzVlyhTOnz/P4cOHKViwIKVKlWLcuHGMGTOGVq1a0ahRo/8rPyGEEHmfjGAIIcRzLVq0wNfXlzlz5pCenk5wcDBr1qwhLi6OwYMHs2vXLtauXUvZsmXfmJeLiwunT5+ma9eubNu2jX79+vH5558TGxvLzJkzs60OGzZsoH79+ixevJiRI0diamrKvn37XolXoUIFJk+ezIULF+jSpQvt27f/v0dWzp49y7Rp0+jXrx+1atXShY8YMYIKFSrQo0eP/zxCIoQQIu9RaA2tYBRCCCGEEEKIf0FGMIQQQgghhBBZRjoYQgghhBBCiCwjHQwhhBBCCCFElpEOhhBCCCGEECLLSAdDCCGEEEIIkWWkgyGEEEIIIYTIMtLBEEIIIYQQQmSZfPkm74gbd41dBPGWuF2xgbGLIN4STj+fMHYRxFvCTJlq7CKIt0Qp/8LGLkKm9poGGrsImXovNcLYRXgjGcEQQgghhBBCZBnpYAghhBBCCCGyjHQwhBBCCCGEEFlGOhhCCCGEEEKILCMdDCGEEEIIIUSWyZe7SL2NtFotG9at5uD+70hIeEpwSHF69x2Am3vR16bbu3sn27d+g0oVi7e3Lx/37kdAYBAAT57Es2Hdas6fPcPjx4+wtbOjclhVOn7QBWtrG10e7zeq80q+nw4fTY2atbK2kiJX8ujeHu9+XTEr7MyTSxGEj5hG3NnfM41frOcHeHRri4V7EVJiVUTtOsS1yfPQJKe8Etd7wEcEjBvEnaVrCR89IzurIfIIrVbL1g3LOXJwJwkJTwkILkm33sNwdfPMNM2VP86xd/s6bt2IQB0bzaBRMylfuaZenF9/PsL3+7dz+0Y4T5/EM3X+Grx8ArK7OiKX0Gq1bF7/NYcP7CYh4SlBwSXp0WcIRdw9Xptu/55t7Nq2EbUqlmLevnTrORD/wBAAHkU9pG/3NgbTDR4xibBqtTjy/XcsmT/dYJyv1u3Czt7hv1VM/GsKU4Wxi5CnSQcjn9i2ZTN7dm1nwOBhuLgWYf3alYwfO4LFS1dgZmZmMM3xY0f4evlS+vQbQEBQMLt2bGX82BF8sWwl9vYOxMbEEBsTQ9ePeuLhWYxHUVF8sWg+sTExjBg9Xi+vAYOGUrZcBd25tY3N3z9O5EOuzRoQNHkYlz6dSNyZ3ynW8wPKffslJyo1JiU69pX4RVq+h/+4QVz6ZCyq0+ew9vWixOKpoNUSMXaWXlzbMiUo2rk1T/7I/dvxiZyzZ9taDuz5hp4DxlHYpQjfrl/GjPEDmbV4I2Zm5gbTJCc/w9Pbn5p1mjB/+giDcZKSkwgMKU3lau/y1SLDX/hE/rVz6wb27d5Kv0GjKOxShE3rvmbKuCHM+2Jtpu3qpx8Ps/qrRXzcdwh+gSHs3fktU8cN4fMvN2Bn74CTc2GWrd2hl+b7/bvYtW0joeUqAVCl+ru6//7L4nnTSE1Jkc6FyNOMOkUqOjqaWbNm0bx5c8LCwggLC6N58+bMnj2bx48fG7NoeYpWq2XXjm20adeRymFV8fb2YdCQ4cTGxHDq5E+Zptu5fSv1GjSiTr0GeHoWo0+/gZibm/P9wf0AFPPyZuSYCVSsFEaRIm6UDi1Dp87dOP3LKdLT0/Xysra2wcHRUXdk1qkR+UuxPp25t3YLDzbsICHiBpeHTCT9WRLuHVsYjG9fMRT16XM83LqXpLsPiDn6M5Fbv8OubEm9eCbWVpRaOpNLg8aTqo7LiaqIPECr1bJ/12aatelK+co18PT2p/eg8ahjozlz6sdM04WWq0KbTr2oEPZOpnGq12pIi3bdKVG6QqZxRP6k1WrZu/MbWrb9kAqVq1PM249+g0ejio3h15PHM023Z8dm3q3fhFp138PD05uP+36KmbkFPxzaC4CJiQkODk56x+mTxwmrVhtLSysAzM3N9a4rlUr+uHiW2vXey5G6C5FdjNbB+PXXXwkICGDBggXY2dlRo0YNatSogZ2dHQsWLCAoKIjffvvNWMXLU6IiH6JSxVI6tKwuzNrahoDAYCKuXDaYJjU1levXrxL6UhqlUknp0LKEhxtOA5CYkICVlRUmJiZ64Uu/WEDHdi0YMrAvhw7uQ6vV/sdaidxOYWqKbekQYo6dfBGo1RJz7BT2FUobTKM+fR7b0iG6DoVlsaI4163O4+/1vxwGzxrD40M/EnvsVLaVX+Q9j6MeoFbFUPylToCVtQ2+AcW5FpH5tDwhXudR1EPUqlhKhpbXhVlb2+AXGExE+CWDaVJTU7l5/SqlQsvpwpRKJaVCy3M1kzQ3rkdw++Y13n1N5+HHwwcwN7egclWZYizyNqNNkerfvz+tW7dm6dKlKBT689y0Wi29evWif//+nDx5MpMcMiQnJ5OcnKwXlpKcjJm54SHN/EilUgFg76A/nGpvb49K9eo0FYD4+Dg0Go2BNA7cv2v4TejxcXFs3riO+g31b44dOnWhVOlQzC3MOX/2DEsXLyDpWRJNmjb/t1USeYCZkz3KAgVIfhSjF57yKAZrf2+DaR5u3Yupkz0V964FBShNTflz5SZuzVuui+PavCG2pYI5VadttpZf5D1qVUZbs7N31Au3s3fUXRPi//VX27G3//vz0BG12vAz9El8HBpNuoG26MD9e3cMpvnh4B7cPYoRGFzS4HWAw4f2UK1mHczfou8wIn8y2gjGhQsXGDRo0CudCwCFQsGgQYM4f/78G/OZPn06dnZ2eseXSxdnQ4lzj6NHDtOmRWPdkZ6elu2fmZiYwKTxo/HwLEb7jh/qXWvXoRMhxUvg6+tPy9btaNGqLdu3fpPtZRJ5j0PVCvgM/JjLQydzslZrzn34CYXq1sRnSC8ALNxcCZo2gos9hxtc9C3eLj8d3U+3NrV0R07c60T+d/zIQTq1qqc70tKyv10lJydz4tj3vFu3caZxIq78wf27d6hdL/M4QuQVRhvBcHV15fTp0wQFBRm8fvr0aVxcXN6Yz8iRIxk8eLBe2J17j7KkjLlVxUphup2eANJSUwFQq1Q4OjrpwtVqNT4+vgbzsLW1Q6lUon4++vEijQp7R/1fcRITE5kwdiSWVpaMGjuRAgVe32wCAoPYvHEdqakpmJrKWoz8KiVGjSYtDfPCTnrhZoWdSHkUbTCN/8j+PPhmF/fXbQXg6ZVrmFhZUnzuBG7O/RLb0BDMCzsTduRbXRplgQI4VCmPx0ftOVSkDGg02VcpkauUrVgd34DiuvO0tIx7XZw6FgdHZ114nDqWYj7+OV4+kTeVr1QNv+c7PcFLz1C1Sq9dqdWxeHkbblcFbe1QKk2I+9sIR5xahb2D0yvxT/10hOTkJGq8Wz/Tch0+uAcvH398/QL/r/qI7KEsILtI/RdG62B8+umnfPzxx5w5c4Z3331X15mIiori8OHDLF++nDlz5rwxH3Nz81eGEs3M8/eiUCsrK6ysrHTnWq0WBwdHLlw4h4+vH5Ax4nA14goN32tiMA9TU1P8/AK4cOEslatUBUCj0XDx/Dnea9JUFy8xMYHxY0ZgamrKmHGT/9Hi7Vs3b2BjU1A6F/mcNjWV+AuXcaxRmUff/ZARqFDgVKMSf3610WAapaUF/G19jjZdo0sb8+MpfqraVO96iUVTSbh2k1uffy2di7eMpZU1llbWunOtVou9gxOXLvyq20I2MTGBG1cvUaeh4Y0FhPg7SysrLP/2DLV3cOSP82fwft5RTUxM4HrEFeo3bGYwD1NTU3z8Avj9whkqhtUAMp6hv184Q4PGr7bFHw7upXzFqtjZGd4Z6tmzRE6e+IEOH/b8j7UTIncwWgejb9++ODs7M2/ePJYsWaLblcjExIRy5cqxatUq2rQxvH+00KdQKHi/WQu+2bQeNzd3XFxcWb92FY5OTlQOq6qLN2bkUCpXqUrjJs0AaNq8JfPnzsLPP5CAgEB27dxGUnIS79ZtAGTcYMeNHk5ycjKDh44kMTGRxMREAGzt7DAxMeH0LydRq1QEBgVjambG+XNn+HbzRpq3bJ3jfweR8+4sWU2JxdOIP3+JuLMZ29SaWFlyf8N2AEosmUbyw0dcmzwfgMcHjuLVpzPxF68Qd+YiVj6e+I/sz6MDR0GjIf1pIk/Dr+t9RnpCIqmxca+Ei7ePQqGgwftt2fHNKlzdPCjk4saW9cuwd3SmXOUaunjTxvSjfOWa1GuccR9KepZI5MN7uuuPox5w++ZVbAra4lzIFYCnT+KIfhyFOjZj9O3h/Yx59PYOTgZ/kRb5h0Kh4L2mbdi6eTWu7kUp7FKEzeu+wsHRiQph1XXxJo4aQMWwGjRs0hKAxs3asnjeNHz9g/ALCGbvzm9JTnpGrTqN9PJ/+OAeVy5dYOSE2ZmW4efjP5Cenk6NWvWyp5JC5DCjvgejbdu2tG3bltTUVKKjM27qzs7OmJqaGrNYeVKLVm1JSkpi8cJ5JDx9SkjxEkyYNENvxCHy4QPi416M7lSvWYu4+Dg2rF2FSqXCx8eXCZOm4/B84feN69e4GhEOQM/u+usulq9ch4uLKyYmJuzds5Ovl3+BVquliJs73Xv0ol4D/RusyJ8id+zHzNkRvxH9MC/sTPwf4Zxp05OUxxmLJi3di4DmxYjFzc++BK0W/1GfYF6kMCkxKh4fOMq1KZ8bqwoij2nc4gOSk5L4evEMEhOeEhBSiuET5uu9qyAq8h5P4tW685vXrzB1dF/d+bqvM9pb9dqN6DVwHABnTh9n2edTdHEWzR4LQIt23WnZoUd2VknkAk1bdiAp6RlfLpxNYsJTgkJKMnrSnL+1qwc8iX/xDK1a413i49RsXvc1alUsXj5+jJ40B3sH/YXfRw7txdG5EKXLZL4F8g8H91IprCbWNgWzvnJCGIFCmw/3E424YXgXJCGy2u2KDYxdBPGWcPr5hLGLIN4SZspUYxdBvCVK+Rc2dhEytd822NhFyFSD+CvGLsIbGfVFe0IIIYQQQoj8xahTpIQQQgghhMhtFKbyG/x/IX89IYQQQgghRJaRDoYQQgghhBAiy8gUKSGEEEIIIV4iL9r7b2QEQwghhBBCCJFl8uUIhlvMRWMXQbwllKe/M3YRxFvCOk223xY546G2qLGLIITI42QEQwghhBBCCJFlpIMhhBBCCCGEyDLSwRBCCCGEEEJkmXy5BkMIIYQQQoh/S2Equ0j9FzKCIYQQQgghhMgyMoKRT31z6ARrvztCTNwT/D3cGPphc0r4FjMY98a9SJZu3Uf47Xs8jFYxuGNTOjSoqRcn4VkSS7fu48hvf6CKf0JgsaIM+aAZxX08c6I6IhfRarWsX7eGA/v3kZDwlOCQ4vTp+wnu7u6vTbdn9y62bf0WlSoWb28fevbuS2BgkO76/n17OXr0CDeuX+fZs0Q2fbMNGxubV/L59fQvbNywjtu3b2FqZkbJEiUZM25iltdTGN/2vfvZvH0XsSo1vt7F+OTjbgQH+Gca/+iJk6xYv4nIR48p6ubKx507Ubl8Wd31GfMXceCHY3ppKpQpzayJY/TCTv56hjWbt3Dz9h3MTM0oXSKEKaOHZW3lRJ6j1WrZsn45Rw7uIiHhCQHBpejWZxhF3DwyTXPlj3Ps2baeWzciUMdGM2jUDCqE1cw0vhD5hYxg5EMHT51j3oad9Ghen3WTBxPg6Ub/WcuIjXtiMH5SSgpFCzvRr01jnOwKGowz5etv+OWPq0zq1YFN04dSqWQAfWYs5VGsOhtrInKjrVu+YfeuHfTt9wmfzVuAhYUF48aOJCUlJdM0Px47ylfLv6R9h058vnAJ3j4+jBs7CrVapYuTnJxMuXLladO2Xab5/HTiOJ/NmUWduvVZuGgps+fMo+Y7tbO0fiJ3+OH4T3zx9Wo6t2vNsnkz8fUqxrDxU1Gp4wzG/+NKBJPnzKdR3dosnz+LapUqMnbaLG7d+VMvXsWyoWxdvUx3jB06UO/6sZ9PMX3eQhq+W4uvPp/DwpmTebdmteyqpshDdm9dx4E939KtzzAmz/kaCwtLZowbSEpKcqZpkpOSKObtT9deQ3KwpEIYn3Qw8qH1+47R7J3KvF+jIj7urozs2goLc1N2/XjaYPziPp4MaP8+9cPKYGb66qBWUkoKP/x6kU/aNaFskC8eLoXo2aIBHi7ObDn8c3ZXR+QiWq2WnTu207ZdByqHVcHb24fBQ4YRGxPDyZM/ZZpux/at1G/QkLr16uPpWYy+/QZgbm7OoYMHdHGaNmtB6zbtCAwKNphHeno6y778gm7dP6LRe41xL1oUT89iVK8hvwbmR9/u3MN79d6lYZ1aeHl6MLjPx1iYm7Hv+x8Mxt+6ey8Vy4bSrkVTinkUpVundvj7+LB97369eKampjg6OOiOgi+NkqWnp7No+Up6dvmA9xvWw8PdDS9PD2pVq5KtdRW5n1arZf+uzTRr04XylWvg6e1H70HjUMdG89upHzNNF1o+jDYf9KRC2Ds5V1ghcgHpYOQzqWlphN++R6XiAbowpVJJxeIBXLx++1/lmZ6uIV2jeaXzYW5myvmrt/5LcUUeExUZiUoVS2joi2kn1tbWBAYGEX7lisE0qampXL9+jdDQMrowpVJJaGgZwsMNpzHk+vVrxMREo1Ao+aRfbz7o2I7xY0dx+7a0wfwmNTWVq9dvUi60lC5MqVRStnQpLoVfNZjmcvhVypUupRdWoWzpV+Kf/+MSzT/ozoe9P2HekmXExb8Y2b164ybRMbEolUp6DBhKy849GD5h6iujIOLt8yjqAWpVDCVCK+jCrKxt8A0I4Vr4H0YsmRC5U67uYNy9e5du3boZuxh5ivpJAukaDY5/m+rkaFuQGLXhKVJvYm1pQSk/L77acYjHqjjSNRq+++k3fr92m2h1fFYUW+QRKlUsAPYO9nrh9vYOqFUqAykgPj4ejUaDvYPDK2lUsbH/+LMjIx8CsGH9Wtq268D4CZOwsSnIqBFDefJE2mF+Ehf/BI1Gg4O9nV64g70dsWq1wTSxarWB+PaoVC/iVyxbhpED+/HZ5HF83LkTFy5dZsTEqaSnpwPwMPIRAKs3fkOnti2ZNnYENjbWDBw1gfgn/+7+KfKHOFUMAHb2jnrhdvaOumsif1EWUOTaIy/I1R2M2NhYVq9e/do4ycnJxMfH6x3JKak5VMK3x6ReHUCrpeEnE6nSdRibDh6nflgZlMq80dDFv3PkyGFatXhfd6Q9/yJmDFqNFoC27dpTtVp1/PwDGDh4CKDgxPHMpygI8ZfaNapStVIFfLyKUa1yRaaNHUn4tRuc/+MyABqtBoCOrVtQs0plAv18GT6gLwoFHP3plDGLLnLYiaMH6Nq6tu5IT0szdpGEyFOMuovUrl27Xnv95s2bb8xj+vTpTJyov4PMiI/aM6pHx/9UtrzKvqA1JkrlKwu6Y+Of4GRveAH3P1HUxZllY/rxLCmZhKRknO1tGbloDe6FnP5rkUUuVqlSmN5OT6mpGZ13tUqNo+OL/+/VahXePr4G87C1tUWpVL4ywqFWq3BwdDSYxhDH53E9PF/shmZqaoarqyuPHz/+x/mI3M/OtiBKpfKVBd0qdRyO9vYG0zja2xuIr8bBwXB8ADdXF+xsC3L/YSTlSpfE6fkom5dnUV0cM1NTiri68Eja2FulXMVq+AWE6M7Tnt/74tSxODg668Lj1LEU8wl4Jb0QbzujdjCaNWuGQqFAq9VmGkeheP0v5CNHjmTw4MF6YSkXDS8CfBuYFihAkFdRTl++xjvlSwKg0Wj49dI12tT97zuhWFqYY2lhTnxCIid/D+eTtk3+c54i97KyssLKykp3rtVqcXBw5PyFc/j4ZnQoEhMTiIgIp+F7jQ3mYWpqip+fPxcunCesSlUgo01eOH+exk3e/8dl8fP3x9TUlPv37lK8eAkA0tLSePQoisKFC//bKopcyNTUlAA/H85e+J1qlSsCGW3m7MXfaf5eA4NpQoICOHvxd1o1fU8Xdub8RYoHZf7l73F0DPFPnuL0vBMS4OeDqakpf957QMmQjM0G0tLSiIp6jEuhQllUO5EXWFpZY2llrTvXarXYOzhx6cJveD3vUCQmJnDj6mXqNGphrGIKkWsZtYNRpEgRlixZQtOmTQ1eP3/+POXKlXttHubm5pibm+uFPTEzzbIy5kUdG9ZkwrKNhHh7UNzHkw0HjvEsOYUmNTIe1OOWbqCwgy392mZ8IUxNS+Pm/ajn/53OY1UcEXfuY2VhhodLxkP15MVwtGgp5lqYu1HRLNi0G68ihXn/eZ7i7aBQKGjarDmbN23A3c0dFxdX1q1dhaOTE2FhVXXxRo0cRliVqjRpkvFvu1nzlsybOxt/f38CAoLYuXMbSclJ1KlbX5dGFRuLSqXi4YMHANy+fQsrSysKFS5EwYK2WFlZ07BRY9avW4tzoUIULuzCti3fAlCtWo0c/CuInNC6aWNmzF9MgJ8vwQF+bNm1l6SkZBq8WwuAafMWUsjRkR6dM0arWzZ5j4GjxvPN9t1UrlCWH378iYjrNxjStycAz549Y/Wmb6kRVhlHB3vuR0bx5aq1uBdxpULZUACsrax4v0FdVm38hsKFnHEp5Mzm7Rkj7e9UC8v5P4LINRQKBQ3eb8v2zatwdfOgkEsRvl23HHtHZ8pXfnH/mTq6H+XDalK/cWsAkp4lEvnwnu7646gH3L55FRsbW5wLu+Z4PYTIKUbtYJQrV44zZ85k2sF40+iGMKxe5TKonjxl6db9xMTFE+DpzsKhH+vecREZo0L50sjQY1U8Hcd8pjtf+91R1n53lLJBviwb3ReAp8+SWPTNXh7FqrG1tqJ2hVL0bd2IAgVMcrZywuhatmpDUlISCxfOJ+HpU0KKl2DSpGmYmZnp4kQ+fEh83IvpKjVqvkNcfBzr1q5BpVLh4+PDpElTcXhp4fd33+1h44Z1uvMRwzL2jR846FPq1K0HQLfuPTAxMWHunFkkJ6cQGBjI1OmzsCn476f/idypdvWqxMXFs2rD5owX7fl4MXPCaByfjzY8ehytdx8rERzImCEDWLF+I1+t3YC7WxEmjxqGd7GMl4EqlUpu3P6TAz8c42lCAk6OjpQPLUW3ju0wM33xo1Svrh9gYmLC9LkLSU5JITjAj8+mjtfbzla8nZq07ERy0jO+WjSDxISnBISUYsTEeZiZvfiRMyryPk/iX9z7bl4PZ8qovrrzdV8vAKBG7Ub0GjQ25wovRA5TaI34Df748eMkJCTQoIHhIe+EhAR+++03atb8//a5f3J6b1YUT4g3inQqYewiiLeEdZrhF8wJkdUeKoq+OZIQWaBcwD9fh5fTjviXNnYRMlXr2gVjF+GNjDqCUb169ddet7a2/r87F0IIIYQQQgjjydXb1AohhBBCCCHyFqOOYAghhBBCCJHb5JUX2uVWMoIhhBBCCCGEyDLSwRBCCCGEEEJkGelgCCGEEEIIkU8tXrwYLy8vLCwsqFSpEqdPn35t/Pnz5xMYGIilpSUeHh4MGjSIpKSk/+sz8+UaDLWDt7GLIN4SptpkYxdBvCWemcp7GETOiAyUlwqKHJIaYewS5HubN29m8ODBLF26lEqVKjF//nzq169PREQEhQsXfiX+hg0bGDFiBCtWrKBKlSpcvXqVLl26oFAomDt37j/+XBnBEEIIIYQQIh+aO3cuPXr0oGvXroSEhLB06VKsrKxYsWKFwfg///wzVatWpUOHDnh5eVGvXj3at2//xlGPv5MOhhBCCCGEEC9RmChy7ZGcnEx8fLzekZz86oyKlJQUzpw5Q506dXRhSqWSOnXqcPLkSYP1rlKlCmfOnNF1KG7evMl3331Ho0aN/q+/n3QwhBBCCCGEyCOmT5+OnZ2d3jF9+vRX4kVHR5Oeno6Li4teuIuLC5GRkQbz7tChA5MmTaJatWqYmpri6+vLO++8w6hRo/6vMkoHQwghhBBCiDxi5MiRxMXF6R0jR47MkryPHj3KtGnTWLJkCWfPnmXbtm3s3buXyZMn/1/55MtF3kIIIYQQQuRH5ubmmJubvzGes7MzJiYmREVF6YVHRUXh6upqMM3YsWP54IMP+OijjwAoWbIkCQkJfPzxx4wePRql8p+NTcgIhhBCCCGEEPmMmZkZ5cqV4/Dhw7owjUbD4cOHCQszvFtcYmLiK50IExMTALRa7T/+bBnByCd27vmOb7btIFalxtfbi349PyIoMCDT+MdO/MSqdRuJjHqEu1sRenT5kEoVyumur16/iaPHT/D4cTQFChTA38+Xbh92JPilPO/dv8+XK1Zz6Uo4aalpeHsXo2unDoSWKpmtdRW5y649e9mydRuxKhU+3t706dXztW3vx+MnWL1uHVFRj3B3c6N71y5UrFAegLS0NFatWcevv/3Gw8hIrK2tKRNamu5dOuPk5JRTVRJGsnv3brZu2YJKpcLbx4fevXsTGBiYafzjx4+zds0aoqKicHN3p1vXrlSoWFF3XavVsm7tWvbv309CQgIhISH07dcPd3d3IONXvI0bNnDhwgVUKhWOjo7Url2btu3aYWpqqpfPtq1b2bd/P4+iorCzs+O9996jXfv22ffHEEZVrHcHfAZ3x9y1EPEXw7k0cDJxv/5uMK6iQAF8h/ek6AfNsHB3IeHqLcJHzuHxweO6OL7DPsa1eT1sAn1If5aE6uQ5wkfNIeHqrZyqknhLDR48mM6dO1O+fHkqVqzI/PnzSUhIoGvXrgB8+OGHuLu769ZwNGnShLlz51KmTBkqVarE9evXGTt2LE2aNNF1NP4JGcHIB478eIKlX63kg/ZtWfr5Z/h4ezFi3CRUarXB+JeuhDN11lwa1H2XpQs+o2rlSoyfOoNbt+/o4hR1d6Nfrx4sWzyf+bOm4epSmOFjJ6KOi9PFGT1xKunpGuZMncSS+XPw9fZizMSpxKpU2V1lkUsc/fE4y5Z/RccO7Vm8YD4+3t6MHjsOdWZt7/IVps+aTYN69Viy4HOqhFVm4pSp3H7e9pKTk7l+4wYd2rdl8YL5jBs9knv37jN+0pQcrJUwhmPHjrF82TI6dOzIwoUL8fH2ZuyYMZm2pcuXLzNzxgzq1a/PwkWLCAsLY/Lkydy+fVsXZ8u337Jr1y769e/PvPnzsbCwYOyYMaSkpABw9+5dNFot/fv354ulS/m4Z0++++47Vq9apfdZXy5dyoEDB/joo49Ytnw548aPJ+A1HR+RtxVp3ZDg2SO5NmUxJyo258nFcCrt/RqzQo4G4wdOGkixHm25NHAyx0o14s6yTZTbsgjb0GBdHMcaFbnzxXp+qtaGXxp2RWlagIrffY2JlWVOVUv8n5Qmilx7/D/atm3LnDlzGDduHKGhoZw/f579+/frFn7/+eefPHz4UBd/zJgxDBkyhDFjxhASEkL37t2pX78+X3755f/1uQrt/zPekUfcvXbZ2EXIUf0GDyPQ34/+vT8GMoa/2nfpQbMmjWjfuuUr8SfPnENSUhJTx495kceQ4fh5ezGwX2+Dn5GQmEjTNh2ZNWUiZUNLERcXT8uOnZk3YyolS4QAkJj4jPfbdGDmlAmUCy2dDTXNfdIVb/cg4CeDhhAQ4E+/3r2AjLbXqUtXmjZuTNs2rV+JP3XGTJKSkpg8YbwubMDgT/Hx8WZAv74GPyPi6lU+GTSEtSu/NvhSoLdFujJ/t7WBAwcSEBBAnz59gIy21PnDD2ny/vu0adPmlfjTp08nKSmJiRMn6sIGDRyIj68v/fv3R6vV0qljR1q0aEHLVq0ASEhIoEP79gwePJia77xjsBxbtmzhu717WbFyJZDx8O3bpw9fLF1K0aJFs7jWuVN4YENjF8Goqvz0DXG//c6lAc8XtSoUvHvrGLcXr+XG7OWvxH/3znGuz/iCO19s0IWV3bwATVIy5zsPNfgZZs4O1H14ipO1OhJ74rdsqUde8F4uftHeidJljV2ETFW7cNbYRXgjo49gPHv2jBMnTnD58qudgqSkJNasWWOEUuUdqampXL1+g7IvfaFXKpWUDS3F5XDD/3Avh0foxQeoUDaUy+FXM/2MvfsPYm1tha+3FwC2tgXxKOrOwR+O8CwpifT0dPbsP4C9vR0Bfr5ZUzmRq6WmpnLt+vVX2l6Z0NBM296V8HDKhIbqhZUrW4Yr4eGZfk5CQiIKhQJrG3mTdX6VmprK9WvXCH2pbSiVSkJDQwm/csVgmvArV15tS+XK6eJHRkaiUqkILVNGd93a2prAwMA3tLcEbAoW1J3/8ssvuLq6cvqXX+japQtdOndm/vz5PHny5F/UVOR2ClNT7MoWJ/rwzy8CtVqif/gZ+8plDKZRmpuiSUrRC9MkJeNQJfMvqAXsMtpYiiou0zhC5GVG7WBcvXqV4OBgatSoQcmSJalZs6beME1cXJxujlhmDL5sJCXltWnyk7j4J2g0Ghzs7fTCHeztUanUBtOoVGoc7O31wuzt7YlV609tOnX6Vxq3ak+jFm3ZumM3MydPwM7OFgCFQsGsKRO4cfMW77fuQMPmbdi6YxfTJ46joHwRfCvEx8ej0Wiwt3fQC89oe4anyRlqe69rqykpKXy9chXv1KyBtZVVVhRb5EJ/tSUHB/22ZO/gkOmUS5VKhb2B+H+1vb/+11CembXPBw8esHvXLho1fPELfmRkJI8ePeL48eMM+fRTBg8ZwvVr15g6der/V0mRJ5g5O6AsUIDkRzF64clRMZi7OhtM8/jgCbwHdMHKrxgoFDi/WwXXZnUxL5LJiKtCQchno4j96QxPL13L6iqILKJQKnLtkRcYtYMxfPhwSpQowaNHj4iIiKBgwYJUrVqVP//88x/nYehlI4uXvjqEKf5/pUuV5MsFc/l89nQqlCvDlJlzdOs6tFotC75Yhr2dHfNmTmXx3FlUqVyJsZOmERMba9yCi3whLS2NqdNnAlr69+1j7OKIfC46OpqxY8ZQrXp1GrzUwdBqNKSmpjLk008pUaIEpUqVYuCgQVy8cIF79+4ZscQit7g8eCoJ1+/wzh/7aJj4B8U/H8fd1dtAozEYv8TC8RQs7s+5joNyuKRC5ByjdjB+/vlnpk+fjrOzM35+fuzevZv69etTvXp1bt68+Y/yMPSykb69emRzyXMPO9uCKJVKVGr9YVaVWo2Dg73BNA4O9q8sAFer1Tj+7ZdoSwsL3N2KEBIUyKcD+mGiNGHfwYytzs5d+J1ffj3D6OFDKBESjL+fLwP69MTczIyDh49kWf1E7mVra4tSqUT9t5GvjLbnYDCNobZnqK2mpaUxdcZMoh4/YvqUyTJ6kc/91Zb+PrKgVqlwzLQtOaA2EP+vtvfX/xrK8+/tMyYmhhEjRhAcEsInn3yid83R0RETExO99RceHh4APHr06J9WUeQRKdEqNGlpmBfW37XO3MWJ5MjoTNOcadWX/Xah/OBbi2MlGpD+NJHEm3dfiVv887EUbvQOp+p2Jul+lIHchMgfjNrBePbsGQUKvFi4qFAo+OKLL2jSpAk1a9bk6lXDawJeZm5ujq2trd5hbmaWncXOVUxNTQnw8+XshYu6MI1Gw7kLvxMSZHiXk5CgQM6dv6gXdubcBUKCMt9aFECjzfglDzKmpgEoFfpDdQql4v/aJ1nkXaampvj7+em1JY1Gw/nzFzJte8FBQZy/cEEv7Oy58wQHBenO/+pc3H/wgBlTp2Bra5s9FRC5hqmpKX7+/lw4f14XltGWzhMUHGwwTVBwMOdfig9w7tw5XXxXV1ccHBz08kxMSCAiIkKvvUVHRzN8+HD8/fwYNGjQK/u/h4SEkJ6ezsMHD3Rh9+/fB3irNx3Ir7SpqcSdvYRz7ZfeEaBQ4FQrDPWpc69Nq0lOIfnBIxQFCuDavB5Ruw/rXS/++Vhcm9blVL3OPLsto18ifzNqByMoKIjffnt194RFixbRtGlT3n//fSOUKu9p2ex9vjtwiIOHf+DO3bt8vuRLkpKSaFDnXQBmfPY5X61aq4vf4v3G/Hr2HN9u28mfd++xev0mrl6/QdPGjQB4lpTE16vXcTk8gqhHj7h6/Qaz5y8kOiaWmtWqABmdFBsba2bOW8CNm7eevxNjFZFRj6hUvtyrhRT5Uovmzdh34ACHvj/Mn3/eZeHiJSQlJVGvbh0AZn02lxWrVuviN3v/fX47c5Yt27bz5927rF2/gWvXr9O0cWMgo3MxedoMrl67zvBPP0WTriE2VkVsrErXuRX5U/Pmzdm/fz/fHzrEn3/+yeJFi0hOTqZu3boAzJkzh5XPd3YCaNq0KWfOnGHb1q3cvXuXdevWce3aNZo0aQJk/GDVrFkzNm3axKlTp7h16xZzPvsMJycnwqpk3Meio6MZMXw4hQsVovtHHxEXF0dsbCyxL03zDC1TBj8/P+bNm8eN69e5du0aCxcsoEyZMm/NrlJvm1vzV+LRvQ3uHzTDJsiHEosnUMDaMmPaE1B65UwCpwzWxbevWArXZnWx9C6KQ9VyVNz7FQqlkhtzvtLFKbFwPO4d3ufcB0NIf5KAuYsz5i7OKC3e/DZmIfIio+572Lx5czZu3MgHH3zwyrVFixah0WhYunSpEUqWt9SqUY24uHhWrduESqXC18eb6ZPG6aadPHr8GOVLi4KKBwcxauggVq7dwIo163B3K8LE0SPw9ioGgIlSyd179zh4+Ajx8fHY2hYkwN+PeTOn4lXMEwA7O1umTxzHijXr+XT0ONLT0inm6cGkMSPw9fHO8b+BMI53alQnLi6ONevWo1Kp8PHxYeqkibopKI8fP9Yb5SoeEsyIoZ+yeu06Vq1eg5u7G+PHjMbreduLjonh1C+/ANCnv/5UlVnTp1FaXuKYb9WsWZP4uDjWrluHKjYWH19fJk2e/KItPXqk15ZCQkIYNnw4a1avZtWqVbi7uzN27Fi8vLx0cVq1bk1SUhILFyzg6dOnFC9enEmTJ2P2fJT73LlzPHjwgAcPHvDh355D3+3bB2TsZjV+wgS++OILhg0bhoWFBeXKl6dHj7dnKu7b5uG3+zAr5EjA+E8yXrR34QqnG39EyvOF35YeRdC+tL5CaW5OwMSBWPl4kP40kUf7j3G+yzDS4l7sNFasVwcAwn5Yp/dZF7qP4N6a7TlQKyFylrwHQ4j/4G1/D4bIOfn9PRgi93jb34Mhck5ufg/Gz+UrGLsImary26/GLsIbGf09GEIIIYQQQoj8QzoYQgghhBBCiCwjHQwhhBBCCCFElpEOhhBCCCGEECLLSAdDCCGEEEIIkWVkWxIhhBBCCCFeojRRvDmSyFS+7GAkKgsauwjiLWGhTTR2EcRbwin+T2MXQbwl3t0xyNhFEELkcTJFSgghhBBCCJFlpIMhhBBCCCGEyDLSwRBCCCGEEEJkGelgCCGEEEIIIbJMvlzkLYQQQgghxL+lUMouUv+FdDDyCa1Wy4Z1qzm4/zsSEp4SHFKc3n0H4OZe9LXp9u7eyfat36BSxeLt7cvHvfsREBiku7544TwunDtLbGwMFhaWBIWE0KVrD4p6eOriLFu6iCuXL3Hn9m08PD35fNGX2VZPYXy79uzl263biVWp8PH2pm+vjwkKDMg0/o/HT7Bq3Xqioh7h7ubGR107U7FCeQDS0tJYtWYdp387w8PISKytrSkbWpruXT7EyclJl8cHXT8i6tEjvXy7df6Qdm1aZU8lRa707YEjrN99iBh1HP7FijKkazuK+3kbjHvz7gO+/GYXEbf+5OHjGAZ+2Jr279XJNO/VO/azZON22jaszeAubbOrCiKP2HTiPKuPnCH6SQIBboUY0bwWJYu5Goz7/cVrfP39ae5Gx5GqSaeYswMfvFOWJuVD9OJ8+/NFrtx7RFxiEpuHdCTIvXBOVUeIHCdTpPKJbVs2s2fXdnr3G8DseYswt7Bg/NgRpKSkZJrm+LEjfL18Ke06fMC8hUvx8vFh/NgRqNUqXRxfP38+GTSUxV+uYOKUGaCFcWOGk56erpdXnboNqF6jZrbVT+QOR388zpfLv6ZTh3YsWTAPH28vRo0dj0qtNhj/0uUrTJs1hwb16vLFgvlUCavEhCnTuHX7DgDJyclcu3GDju3bsmTBPMaPHsHde/cZN2nqK3l92KkDm9au1h1N32+cnVUVucyhn3/l8zVb6N7yPVbPGI1fsaIMmLaA2Lh4g/GTklNwd3GmT/vmONnbvjbvy9dvs/37H/HzfP0PMuLtsP9cBHN2/kjP+pXZNLgjgW7O9F62jZgnhrclt7Oy4KM6lVgzoC1bPv2AphVDGL/pID+F39bFeZaSShlvdwY2rpZDtRDCuKSDkQ9otVp27dhGm3YdqRxWFW9vHwYNGU5sTAynTv6Uabqd27dSr0Ej6tRrgKdnMfr0G4i5uTnfH9yvi9OgYWNKlCyFi4srvn7+dPywK9GPH/PoUZQuzse9+vFek6a4uBbJ1noK49u6fScNG9Sjft06FPP0ZEC/PphbmHPg4PcG4+/YtZsK5crSpmULPD096PJBJ/x8fdi1Zy8A1tbWzJw6mZrVq+FRtCjBQUH0692Ta9ev8+jRY728rCwtcXR00B2WFhbZXl+Re2zc+z1N361Gk1pV8SnqxoiPOmJhZsbuIz8bjB/i58UnnVpRr2oFzExNM803MSmJcYu+ZtTHH2BrY5VdxRd5yNpjZ2lRuQTNKhbH19WJMa3qYGFagB2n/zAYv4KfB++W8sPHxQkPZ3s61iiLf5FCnLv1QBenSfkQetWvTKUAT4N5iNxHaaLItUdeYPQOxpUrV1i5ciXh4eEAhIeH07t3b7p168YPP/xg5NLlDVGRD1GpYikdWlYXZm1tQ0BgMBFXLhtMk5qayvXrVwl9KY1SqaR0aFnCww2nSUp6xuFD+3FxdcXZuVDWVkLkeqmpqVy7fp0yoaG6MKVSSZnQ0lx5/u/37y6Hh1MmtLReWPmyZTOND5CQkIBCocDaxlovfPO3W2nZriO9+w/gm63bXhlFE/lXaloa4Tf/pGLJYF2YUqmkQskgfr928z/lPfvrjVQtU5KKpYLfHFnke6lp6Vy5F0XllzoCSqWCygGeXLz98I3ptVotv1z9k9uPYynn456dRRUiVzPqGoz9+/fTtGlTbGxsSExMZPv27Xz44YeULl0ajUZDvXr1OHjwILVr1840j+TkZJKTk/XCUpKTMTM3z+7i5xoqVcaUJnsHB71we3t7VKpYg2ni4+PQaDQG0jhw/+5dvbDv9uxk1YrlJCUl4V7Ug0lTZ2H6ml8ERf4UHx+PRqPBwd5eL9zB3p67d+8bTKNSqV+Jb29vT6xKZTB+SkoKX61czTs1a2Bt9eLX5KbvN8bf15eCBW24fCWcFavWEBuroleP7v+pTiJvUMc/JV2jwdGuoF64o50tdx5E/ut8D/70KxG3/mTltFH/tYgin1AlPCNdo8WpoP5ollNBK249MnzfAnjyLJm6E5eTmpaOUqlgVMvahAUWy+7iCpFrGXUEY9KkSQwdOpSYmBhWrlxJhw4d6NGjB4cOHeLw4cMMHTqUGTNmvDaP6dOnY2dnp3d8uXRxDtXAOI4eOUybFo11R3p6WrZ+Xs1a7zJ/4VKmzZyLu3tRZk2f/Nq1HUL8G2lpaUyZPgvQ8knf3nrXWjVvRulSJfHx9qZxo4Z8/FE3du7eQ0pqqnEKK/K8qOhY5q7ezMT+3TE3kx9MxH9jbW7GN0M6sX5Qe/o1qspnO3/k1+t335xQiHzKqCMYly5dYs2aNQC0adOGDz74gFatXuwK07FjR1auXPnaPEaOHMngwYP1wu7ce5RJ7PyhYqUwvZ2e0p5/yVKrVDg6vth5R61W4+PjazAPW1s7lEol6r/9kqxWq7B31B/VsLa2wdraBjf3ogQGBdOhTXNO/nyCmu9kPrIk8h9bW1uUSuUrC7pVajWODvYG0zg42L8SX61W4/i3kbO0tDSmzJjFo8ePmDVtit7ohSFBgYGkp6cTFRWFR1FZmJvf2dvaYKJUEhv3RC88Ni4eR3u7f5Vn+K0/UcU9ofOIFxsKpGs0nLtyjS0HjnJ8/WJMlEafRSxymIO1JSZKxSsLumOeJOJcMPP7klKpwLOQPQBB7oW5FRXL14d/pYKfR3YWV4hcy+h3T4UiY7GKUqnEwsICO7sXD4uCBQsSFxf32vTm5ubY2trqHfl9epSVlRVubu66w8OzGA4Ojly4cE4XJzExgasRVwgMDjGYh6mpKX5+AVy4cFYXptFouHj+HEFBhtNk0KJFq+vUiLeHqakp/n5+nD9/QRem0Wg4f/4iwUFBBtOEBAVx7sJFvbCz587rxf+rc3H/wQNmTJ2Mre3rd/wBuHHzJkqlEns7+39XGZGnmBYoQJCPJ7/+fkUXptFo+PWPcEr6+/yrPMuXCGLD7HGsnTlGdwT7FKN+tYqsnTlGOhdvKdMCJgQXdeGXay9GHzQaLb9cu0spr3++kYlGqyU1TdaJibeXUUcwvLy8uHbtGr6+Gb+ynzx5Ek/PFwur/vzzT4oUkZ2J3kShUPB+sxZ8s2k9bm7uuLi4sn7tKhydnKgcVlUXb8zIoVSuUpXGTZoB0LR5S+bPnYWffyABAYHs2rmNpOQk3q3bAIDIhw84/uNRypQtj52dHdHR0Wz9dhPmZmaUq1BRl++DB/dJevYMtUpFSnIyN29cB8DDs5is1chnWjZvyuy58/H39yMoIIBtO3eRlJRE/brvAjDrs3k4OTnSvUtnAJq934RPR4xiy7btVKxQgaM//sjV69cZ0L8vkNG5mDxtBtdu3GTy+LFo0jXExmaMqhUsaIOpqSmXr4QTHhFB6VKlsLK05HJ4OEuXf03tWjUpWNDGOH8IkePav1eHSUtWEezrRYivF5u+O0xScgqN36kCwIRFKynkaE/fDs2BjIXht+491P33Y5Waq7fvYmlhjodrYawtLfD11F+Ea2lhjp2N9Svh4u3yQc2yjN14gOIehSnh6cq6Y+d4lpJKs4rFARi9YT+FbW0Y8HzL2a+/P02IhwseznakpKVz/Mpt9v52hdGtXozyxyUk8VAdz+O4BABuP1/P4VzQGmdba0Tuo8gjuzXlVkbtYPTu3VtvJ5gSJUroXd+3b99rF3iLF1q0aktSUhKLF84j4elTQoqXYMKkGZiZmeniRD58QPxLI0LVa9YiLj6ODWtXoVKp8PHxZcKk6Tg8n75iambG5Ut/sGvnNhKePsXe3oHiJUoy87MF2Nu/mOKy6PPP+OP3F79SD+zfC4DlK9fh4mL4xUQib3qnRnXi4uJYs27D8zbjw9RJE3Rt5tHjx7pRSYDiIcGMHDqEVWvXs3L1Wtzc3ZgwZhTeXhmLH6NjYjj5y2kAevcfoPdZs6dPpXSpkpiamnL0x+Os3bCJ1NRUXF1caNHsfVo2b5YjdRa5Q90qFVDHP2XZN7uIUccT4FWU+SM/0b3jIiomFuVLb959HKvmg+FTdOfrdx9i/e5DlA0J4IvxQ3K8/CLvaFAmENXTZyzZf5Lo+EQC3Qux5OPmOBXM6AhEqp6gfOk+9ywllWlbfyBK/QRz0wJ4uzgytWMDGpQJ1MU5eukG4zYd1J0PX/sdAL3qVaZ3g7AcqpkQOUeh1Wq1xi5EVou4IQurRM6w0Bp+8ZIQWc3u6Zu3yBQiK1jcz3wbaSGyksV7vYxdhEz9VjP3dvzKHztp7CK8kUwyFUIIIYQQQmQZ6WAIIYQQQgghsox0MIQQQgghhBBZxqiLvIUQQgghhMhtFLJV9X8ifz0hhBBCCCFElpEOhhBCCCGEECLL5MspUh4PThm7COIt8cxBXsglckZ0wWLGLoJ4S8T7lzR2EcRboqyxC/AaCqW8aO+/kBEMIYQQQgghRJaRDoYQQgghhBAiy0gHQwghhBBCCJFlpIMhhBBCCCGEyDLSwRBCCCGEEEJkmXy5i5SAzT+cYvWBE8TEPSXAw5Xh7RtTwqeowbjbfvyVPSfPc/1+FADBxdzo37yeXvylOw9z4NffiYyNw7SACcHF3OjXvC4lfTxypD4i99q67zDrd+4jVh2Hn5cng7t3JMTfx2Dcm3/e56tN2wm/eZvIxzEM6Nqeto3r6cU5dymCDTv3EXHzDtEqNdOH9admpdy814jIKlqtlnXr1nJg/z4SEhIIDgmhb9/+uLu/fre2Pbt3sXXrFlQqFd7ePvTq3YfAwEDd9ZSUFL5avowffzxGamoqZcuWo0/ffjg4OOjlc+jQQXZs38b9+/exsrKiWrXq9OnbT3f9zJnfWL9uHX/+eQdTUzNKlCjBRz164OLimrV/CJHjtFotW9Z/xQ8Hd5GQ8ITA4FJ06zOUIm6vf8Yd3LuV3dvWE6eKxdPbjy49B+MXEKK7HvXwHutWLCLi8kXSUlMoVbYyXXoOxt7BURfn1vUINqxews1rV1AqlVSs8g4fdP8EC0urbKuveDOliewi9V/ICEY+dOD073z2zT56NqnFhnF9CPBwpc/8VcTGPzUY/7eIWzSoWIrln3Zn9cieuDrY0XveKh6p4nVxirk6M7xDY76d2J+Vw3vg5uRAn3mriH2SkFPVErnQ9z/9woJVm+jWpikrZ0/Ar5gHgyZ/RmxcvMH4SSnJuLkUonen1jjZ2xmOk5yMn5cHQ3p0ys6ii1xoy5Zv2b1rJ337fcLcefOxsLBg7NjRpKSkZJrmx2PHWL58OR06dGLBwkV4+/gwduxo1Gq1Ls7yZV9y+vQvjBw5mhkzZxMbG8PUKZP18tm+bStr16ymdeu2fLH0S6ZOm07ZcuV01yMjI5k8aSKlS5dm4aLFTJ4yhfj4+FfyEXnT7q3r2L/nW7r3GcrkOV9hbmHBjHGDSElJzjTNyePfs/arBbRs341p81dSzNuPGeMGEaeOBSAp6RnTxg1EoVAwZupCJsz6kvS0VOZMHopGowEgNuYxU8d+gmuRokyes5wRE+Zy789bfDF/So7UW4jskus6GFqt1thFyPPWHfqJFtXL07RaOXzdCjO60/tYmJmy48QZg/Gn9WhDm1qVCPQsgneRQozr0hytVssvV27o4jSsVJrKIX4ULeSIr7sLQ9o25OmzZK7di8ypaolcaNPug7xfpwaNa1fH28OdYT0/xNzcjD2HjxuMH+LnQ7/ObalbrRKmpoYHUMPKlqJnh5bUrFTO4HWRP2m1Wnbu2E7bdu0JCwvD29uHIUOGEhsTw8mTP2eabvv2bTRo0IC69erh6VmMfv36Y2FuzsGDBwBISEjg4MEDfNTjY0qHhuLv78/AQUO4cuUy4eFXAHjy5Alr165h8JBPeadWLYoUccPb24fKlcN0n3P9+jU0Gg0ffNiZIkXc8PPzp0XLlty8eZO0tLTs/eOIbKXVatm36xuat+lC+co1KObtR59B41DFRvPbqR8zTbd3xyZq13+fd+o0pqinN937DMPM3Jyjh/YAcPXyRR4/iqTXwDF4evni6eVL70FjuXk9nEsXM57H5379CZMCBejaawhuRYvhGxBC9z7DOP3zUSIf3MuR+guRHXJdB8Pc3JwrV64Yuxh5VmpaGlfuPKBSiK8uTKlUUinYl4s37/6jPJJSUklLT8fO2jLTz9j242/YWFoQUFSmBrytUlPTiLhxm/KliuvClEolFUqF8MfV60YsmciLIiMjUalUhIaW0YVZW1sTGBhEeCbPhNTUVK5fv6aXRqlUEhpaRtd5uH7tGmlpaXpxPDw8KFSosO5Zc/7cOTQaDTExMfTs2YMPP+jE9GlTefz4sS6Nn58/CoWSQ4cOkp6eTkJCAj8cPkxoaBkKFJDZxnnZo6gHqFUxlAgtrwuzsrbBNyCEa+F/GEyTlprKresRlCj9Io1SqaREaAWuRWSkSU1LRYECU1NTXRxTMzMUCiURly9kxElNpUABU5TKF1/HzMzMAXRxhMiLjHZXHDx4sMHw9PR0ZsyYgZOTEwBz587NyWLleaqniaRrNDja2uiFO9nacDsy+h/l8fmWAxSyL6jXSQH48UI4I5Z9Q1JKKs52Niwd3AWHgtZZVnaRt6ifPMloa/a2euGOdnbcuS8jW+L/o1KpAHBwsNcLt7e31137u/j4eDQaDfYG0ty9e1eXb4ECptjY6N8THRxe5Psw8iFarZZvNm/i4569sLa2Zs2a1YwZPZJFi7/A1NQUV1dXpkydyozp01i0cAEajYag4GAmTpQpUnldnCpjSpOdvaNeuJ29I+rn1/4uPl6NRpOOncOraR7cuwOAf2BxzC0s2LBqCe0+6IUWLRtXf4FGk446NgaA4qXKse7rBezetp6GTdqQlPyMjauXAKBSxWRpPYXISUbrYMyfP5/SpUtjb2+vF67Varly5QrW1tYoFG9eYJOcnExysv4cyfSUVMzNTDNJIV5nxXfHOHD6d5YP7Y65qf7fsEKQD5vG9UX9NJFtx39l2JebWDuq1yudGSGEeJMjR35g0cIFuvMJEycZrSxarZa0tDR69upN2bIZU/OGDx9Bp44duHjxAuXKlSc2NpYFn3/Ou+/WoeY77/As8Rnr1q1h2rQpTJ06/R89r0TucOLoAb5aPEt3PmzcnGz5HFs7BwYOn8LXX8zmwO5vUSiUVKlRB2/fQBTPRyw8ivnQe+BY1n69gE2rl6JUKmnQpDV29o4opU2JPMxoHYxp06axbNkyPvvsM2rXrq0LNzU1ZdWqVYSEhLwm9QvTp09n4sSJemGjurRidLc2WVrevMLBxgoTpfKVBd0x8U9xsnt9R2DNgROs3HecpUO6EuDx6tQnS3MzPF2c8HRxopSvB++Pmsf2E2fo3qhmltZB5A32BQtmtDW1/oLu2Li4V0Y1hPi7SpUqExgYpDtPTc1YyK1SqXF0dNKFq9VqfHwM70pma2uLUqlErVLrhavVahwcM3aIcnBwIC0tladPn+qNYqhUat0uUo7Pf4X29PTUXbezs8fW1lY3TWrvnt1YW1vRrftHujifDh1G5w8/ICIinKCg4P/7byCMo1zFavgFvJja+Vfbi1PH4uDorAuPU8fi5eNvMA9bW3uUShPd6MfLaV7eIapU2Up8vnwL8XFqTExMsLYpSK8PGhPm6qaLU/WdelR9px5qVSwWFhagULB35yYKu75+9zSRvRRK6eD9F0ZbgzFixAg2b95M7969+fTTT0lNTf1X+YwcOZK4uDi949NOzbO4tHmHaYECBBdz45crN3VhGo2G0+E3KfWaLWVX7TvO8j1HWDywM8W9/tlNTavVkJoqixvfVqamBQj09eLM75d1YRqNht8uXqFEgJ8RSybyAisrK9zc3HSHp2cxHBwcuHDhvC5OYmJCxpf3YMNf3k1NTfHz8+f8S2k0Gg3nz5/XfeH38/enQIECXDj/Is69e3d5/PgRwc/z/esHrXv3XiyqffLkCfHx8RQuXBjIGC1XKPQfmX/Nm9doZHOSvMTSyhpXt6K6o6inN/YOTvxx4TddnMTEBG5cvYx/UAmDeRQwNcXbL5A/Lr7YPEWj0XDpwm/4B76axtbOHmubgvxx4Tfi41SUq1jtlTj2Do5YWFpx8vhhzEzNKBlaIQtqK4RxGHVlWoUKFThz5gx9+/alfPnyrF+//v8eZjY3N8fc3FwvLPEtnx7VqW5Vxq3YSkgxN0p4F2XD9z/zLDmFplUzhv7HfL2Fwva2fNIy4/0DK/f9yBc7DzOtRxvcnO2JjnsCgJW5GVYW5jxLTuGrvUepWToYZ3sb1E8S+ebILzxSPaFuecM3X/F2aNekHlMWfkWQrxch/j5s3nOQpORkGtfOeHhOWrCcQo729O7UGshYGH7r3gMA0tLSeRyj4uqtP7GyMKdoERcAEp8lcS/yke4zHj56zNVbf2JrY41rISdE/qRQKGjarDmbNm3Ezc0NVxdX1q5dg6OTE2FhVXTxRo0cQViVKjRp8j4AzZu3YO7cOfj7+xMQEMjOndtJSk6ibt2M+5u1tTX16tVn+fJl2BQsiJWVFUuXLiEoOFjXCXEvWpTKlcNY9uVS+vUfgJWVFatXraRo0aKUKlUagAoVKrJjx3Y2bFhPzZrv8OxZIqtXr6Jw4cL4+voi8i6FQkHD99uwY/NqXN08KOzixrfrluHg6Ez5yjV08aaM7k+FsJrUb9wKgPeateOLeVPw8QvCLyCEfTs3k5yURM06jXVpjn6/B/eiXtja2XM1/A/WLJ9Pw6ZtcStaTBfnwJ4tBASVxMLSkt/P/8r6FYto37k31jYFc+6PIEQWM/rWFzY2NqxevZpNmzZRp04d0tPTjV2kPK9+xZKonibwxc7DxMQ/JdCjCIsHdtZNkYqMUevN7fz26GlS09IZ+sVGvXx6NqlFr6bvolQquP0wmt0/b0D9NBE7ayuKe7uzYvhH+Lq75GjdRO5Sp2ol1HFPWL5pB7HqOPy9PZk7ZjCOz99xERUdo9fWolVqunw6Xne+Ydd+NuzaT5nigSyeNAKA8Bu36Td+pi7OglWbAGj0TlXG9H8xPUXkP61atSYpKYmFCxeQ8PQpIcWLM3nSFMzMzHRxHj58QHxcnO68Rs2axMXHsW7tWlQqFT4+PkyaNEXvJXo9Pu6JQqFg2tTJGS/aK1eOPn366X32kE8/ZdmyL5kwYRxKhYISJUsyafJU3Q5RpUNDGTpsOFu3fMvWLd9ibm5OUHAwkyZPfeVHLpH3NGnZieSkJL5aNJPEhKcEhpRixMS5uh2dAKIi7/MkXq07D6teh/g4NVvWL0etiqWYjz8jJs7VmyL18N6fbFq9lKdP4ylUuAjN2nSmUdN2ep994+pltmz4iqRnz3ArWoyP+g6jeu2G2V5nIbKTQpuLXjxx7949zpw5Q506dbC2/ve7EyUe/zYLSyVE5p45yBxZkTNUlkWMXQTxlohPlzVUImeUDci9o9IXGtR4cyQjKb0/8/ez5BZGH8F4WdGiRSlatKixiyGEEEIIIYT4l3JVB0MIIYQQQghjUyhz3buo8xT56wkhhBBCCCGyjHQwhBBCCCGEEFlGpkgJIYQQQgjxEnnR3n8jIxhCCCGEEEKILJMvRzDuu1U0dhHEWyIdE2MXQbwl4lPlpVsiZ2iMXQAhRJ4nIxhCCCGEEEKILCMdDCGEEEIIIUSWkQ6GEEIIIYQQIsvkyzUYQgghhBBC/FtKE9lF6r+QEQwhhBBCCCFElpEOhhBCCCGEECLLyBSpfEKr1bJ+3RoO7N9HQsJTgkOK06fvJ7i7u7823Z7du9i29VtUqli8vX3o2bsvgYFBuuv79+3l6NEj3Lh+nWfPEtn0zTZsbGx01y9evMCoEUMN5j13/kICAgKzpoIi19BqtWxYt4pD+78jIeEpQSEl6N13AG7uRV+bbu/uHezY+g0qVSxe3r583Ls/AS+1tSUL53Lh3FliY2OwsLAkKKQ4nbv2oKiHpy7OtavhrFn5FTeuXwWFAv+AILp0+xhvH99sq6/IXbRaLds2LOPIoR0kJjwlIKgUXXoPx9XNM9M04ZfOsnf7Om5fD0etimbAyFmUr/yO7npaWhpb1n/BhTM/8yjyPlZWNhQvXYG2H/bDwalQDtRK5EZarZbtG5Zx9Hlb8w8qRed/0Nb2vdTWPhk5i3IvtTWA7RuX8cvxQ8RER1GggClevkG06tQb38AS2VwjIXKOjGDkE1u3fMPuXTvo2+8TPpu3AAsLC8aNHUlKSkqmaX48dpSvln9J+w6d+HzhErx9fBg3dhRqtUoXJzk5mXLlytOmbTuDeQQHh7B23Sa9o179hri4uuLvH5Dl9RTGt23LJvbu2k7vfgOZPW8RFhYWTBg74rVt7fixI6xYvpS2HT5k7sKlePv4MmHscL225usXwCeDhrHoy5VMmDIDrVbL+DHDSU9PB+DZs2dMHDsC50KFmTVvMTNmf46lpSUTxg4nLS0t2+stcoe929ZwcO9muvYewYTZKzC3sGTWhE9ISUnONE1yUhKeXv507mn4x5CU5CRu34igWZtuTJm7lgEjZ/Lw/p/Mmzoku6oh8oDvtq3h0N7NdOk9gnHP29qcf9DWPLz8+SCTtgbg6ubJBx8PZeqCjYyesQznwkWYPaE/8XGqTNMIkddIByMf0Gq17NyxnbbtOlA5rAre3j4MHjKM2JgYTp78KdN0O7ZvpX6DhtStVx9Pz2L07TcAc3NzDh08oIvTtFkLWrdpR2BQsME8TE1NcXB01B0FbW355dTP1KlTD4VCFkjlN1qtlt07ttG6XScqhVXFy9uXgUOGExsTzamTJzJNt3P7Fuo1aESdeg3w9PSid7+BmJub8/3B/bo49Rs2pnjJUri4uOLrF0CnD7sS/fgRjx5FAXDv7p88efKEDh90oWhRDzyLedGuw4eoVSoeP48j8jetVsv+3Zt4v3U3ylWqiaeXPz0HTkAdG82ZU8cyTVe6XBVad+pN+bBaBq9bWdswYtIiKlWrS5GixfALLEnnnkO5dSOc6MeR2VUdkYtptVoO7N5Ek9bdKPu8rX38vK2dfUNba/WatgYQVrMBxUMrUtjVnaKevnToPpBniQncvX0tO6oihFHkqg5GQkICK1euZPTo0SxatIiYmBhjFylPiIqMRKWKJTS0rC7M2tqawMAgwq9cMZgmNTWV69evERpaRhemVCoJDS1DeLjhNP/EL6dO8uTJE+rWq/+v8xC5V1TkQ1SqWErrtTUbAgKDibhy2WCa1NRUbly/qpdGqVRSOrQsEeGG0yQlPeP7QwdwcS2Cs3PGFBX3oh4UtLXl+wP7SE1NJTk5me8P7qOohyeFXVyzsJYit3oc9YA4VQwlSlfUhVlZ2+ATUJzrEb9n6WclJjxFoVBgbW3z5sgi3/mrrRXP5raWlprKkQM7sLK2wdNbRv1zE4VSkWuPvMCoazBCQkI4ceIEjo6O3L17lxo1aqBSqQgICODGjRtMnjyZU6dO4e3tnWkeycnJJCfrD1emJCdjZm6e3cXPNVSqWADsHez1wu3tHVCrDA+5xsfHo9FosHdweCXNvbt3/3VZDh7cT5my5XRfCkX+onrengy1G1WmbS3uH7e17/bsZPWKZSQlJeFe1IOJU2dhamoKgJWVFVNnzGXa5HF8s2kdAEXc3JkweSYmJiZZUj+Ru6lVGT862dk76oXb2TsSp8q6H6RSUpLZvGYRlavXw9JKOhhvo7hM2pptFrW1878eZ8mcMaQkJ2Hn4MzQiYsoaGv/n/MVIrcw6ghGeHi4bu70yJEjcXNz486dO5w+fZo7d+5QqlQpRo8e/do8pk+fjp2dnd6xdOmSnCi+0Rw5cphWLd7XHWnP56gbW3T0Y86dPUO9eg2MXRSRRY4e+Z62Ld7THenp2bvWoWatd5m38EumzZyHm3tRZk+fpFvbkZyczML5cwgOKc6suQuZMedzPIt5MXnCqFd+ZBD5w09H9/NR25q6I7vbH2Qs+F40axRarZauvYdn++eJ3OHno/v5uG1N3ZHdbS24ZHkmz1/HmJlfUapsZRbPGkm8OjZbP1OInJRrdpE6efIkS5cuxc7ODgAbGxsmTpxIu3aGFxf/ZeTIkQwePFgv7O69/D1ntlKlML2dnlJTUwFQq9Q4OjrpwtVqVaa769ja2qJUKl8Z4VCrVTg4OhpM8yaHDh6gYMGCVKoc9q/Si9ynYqUqBAa+WH/zoq2p/o+2ZveP25q1tQ3W1ja4uRclICiYjm2acernE9R4pzY/Hj3Mo0eRzJq7EKUy47eRIcNG07FNM3459RM1atbOkjqL3KNsxer4BRbXnaemZnQ249Sx2Ds668Lj1LEUy4LpJRmdi5FEP37IyMlLZPTiLVKmYnV8/0Fbi1fHZslUJnMLS1yKeOBSxAO/wJIM69WSY9/vokmrLv85b5E1FMpctYogzzH6X++vhcBJSUkUKVJE75q7uzuPHz9+bXpzc3NsbW31jvw+PcrKygo3N3fd4elZDAcHR85fOKeLk5iYQEREOEHBmS/O9vPz58KF87owjUbDhfPnCcpkQffraLVavv/+ILXfrUuBArmm3yr+IysrK4q4uesOj+dt7eKFs7o4iYkJXI24QmBwiME8TE1N8fUL4OJL7VOj0XDx/DkCgwynyaBFi1b3oE9OTkapUOptHqBUKlEoQKvR/reKilzJ0spa9yXMpYgH7h4+2Dk4cenir7o4zxKfcvPqJfwCS/6nz/qrcxH58C4jJi2W6Spvmcza2uVsaGuGaLQa0lIz34lPiLzG6N8E3333XQoUKEB8fDwRERGUKPFiH+g7d+7g5OT0mtQCMjppTZs1Z/OmDbi7uePi4sq6tatwdHIiLKyqLt6okcMIq1KVJk2aAtCseUvmzZ2Nv78/AQFB7Ny5jaTkJOrUfbFAWxUbi0ql4uGDBwDcvn0LK0srChUuRMGCtrp4Fy6cJyoyknr1ZXpUfqZQKGjSrAXfbFpPEbeiuLi4smHtShydnKkcVk0Xb+zIT6lcpRrvNWkGQNPmrfh87kz8/APwDwhi986tem0t8uEDTvx4lNCy5bGzsyM6Opqt327E3MyMchUqARBaphyrvv6SL5cs4L0mzdBqtWz9ZiMmJiaULB2a038KYQQKhYIGTdqx85sVuBbxoJCLG1s2LMXe0ZlylWvq4k0f24fyld+h7nttAEh6lkjUw3u664+jHnDn5lWsC9riXMiVtLQ0Fs4cwe0b4QweOxeNJh21KhoAGxs7CjxfByTeHgqFgvpN2rHrmxW4PG9r2563tbIvtbWZY/tQ9h+0NZuCtjgVciU56Rm7vl1JmYrVsXdw5km8msPfbUEd85gKVd/N8XoKkV2M2sEYP3683vnLL3AD2L17N9WrV8/JIuVZLVu1ISkpiYUL55Pw9CkhxUswadI0zMzMdHEiHz4kPi5Od16j5jvExcexbu0aVCoVPj4+TJo0FYeXFuN+990eNm5YpzsfMSxjX/iBgz6lTt16uvBDB/YTHByCh0fmLyAS+UOLVu1ISkpiycK5JDx9SnDxkoyfNP1vbe2BXlurXrMW8fFxbFi7CpUqYzrV+EkzsHfImCJlambG5Uu/s2vnVhKePsXO3oHiJUox47OF2NtntMeiHp6MGT+FTRvWMnxIfxQKJT6+foyfPENvupbI395r8SHJSUmsWDIt40V7waUZOv5zzMxejFw/irzPk3i17vzW9StMG9Nbd75hxXwAqtV+j54DxqOKecTZ0z8CMGZgJ73PGzXlC4JLlsu+Colcq9HztrbqeVvzDy7Npwba2tO/tbUZL7W1jS+1tR4DxqNQKnl47zYnftjL03g1NgXt8PYPYdT0ZRT1lBeGivxDodVq893cgms37hi7COItkY7sXiRyRnxqQWMXQbwlNMYugHhrVA6yM3YRMnWlZV1jFyFTwVsPGbsIb2T0NRhCCCGEEEKI/MPoazCEEEIIIYTITfLKC+1yKxnBEEIIIYQQQmQZ6WAIIYQQQgghsox0MIQQQgghhBBZRjoYQgghhBBCiCyTLxd5J2ktjF0E8ZYooEgzdhHEW0KpkM1DRc6wUKYauwhCiDwuX3YwhBBCCCGE+LdkF6n/RqZICSGEEEIIIbKMdDCEEEIIIYQQWUY6GEIIIYQQQogsIx0MIYQQQgghRJaRRd75mFarZfO6FXx/YDeJCU8JDC7Jx30HU8Td47Xp9u3Zxq6tm1CrYinm7Uv3XgPwDwwB4FHUQ/p0a2sw3eARE6lSvVaW10Pkflqtlo3rVnFo/14SEp4SFFKCXn0H4uZe9LXpvtu9g+1bN6NWxeLl7UuP3v0JCAw2mP/kcSM5e+Y0I8ZMonKVatlVFZHLabVatm5YzpGDO0lIeEpAcEm69R6Gq5tnpmmu/HGOvdvXcetGBOrYaAaNmkn5yjX14vz68xG+37+d2zfCefoknqnz1+DlE5Dd1RG5hFarZfP6rzl8YHfGPSy4JD36DHnj83L/nm3s2rZR97zs1nOg3vOyb/c2BtMNHjGJsGq1uH3zOju2rCP88u/Ex6spXLgIdRs25b2mrbO8jkLkJBnByMd2bNnAd7u38nHfIUyb+yXmFhZMHvspKSnJmab56cfDrF6+mNYdujBrwVd4efsxZeynxKlVADg5F2b52u16R9uO3bCwtKRM+Uo5VTWRy2zfsok9u7bRq98gZs1bjIWFBRPHDiclJSXTNCeOHWHF8i9o1+FD5i78Ei8fXyaOHY76eVt72e4dW0A29BDAnm1rObDnG7r2Hs6k2V9hbm7JjPEDX3tfS05+hqe3P116fpppnKTkJAJDStOuc9/sKLbI5XZu3cC+3Vv5uO+nTP/sS8wtLJkybsibn5dfLaJ1+y7M/Pwrinn7MXXcEL3n5bK1O/SONs+fl6HlMp6XN69HYGvnQP8hY5i3ZC0t2n7AhjVfsm/31hypt8icQqnItUdeIB2MfEqr1bJ357e0bPsBFcOq4+XtS/8ho1HFxnD65IlM0+3e/g11GjSmdt1GeHh68XG/IZhbWPDDwb0AmJiY4ODopHf8cvI4VarVwtLSKqeqJ3IRrVbL7h1badOuE5XCquLl7cuAISOIjYnml9e0tZ3bv6Veg0a8W68hHp5e9O43CHNzcw4f3KcX7+aN6+zc9i39Bw7L7qqIXE6r1bJ/12aatelK+co18PT2p/eg8ahjozlz6sdM04WWq0KbTr2oEPZOpnGq12pIi3bdKVG6QjaUXORmGc/Lb2jZ9kMqVK5OMW8/+g3OeF7+evJ4pun27NjMu/WbUKvue3h4evNx308xM7fgh0MvPS8dnPSO0yePE1attu55Wbvee3TrOYDiJcvg4upGjVr1qVWnEb+czLw9C5EXGLWDcfbsWW7duqU7X7t2LVWrVsXDw4Nq1aqxadMmI5Yub3sU+RC1KpZSoeV1YdbWNvgHBnM1/A+DaVJTU7l5/apeGqVSScnQckSEXzKY5sa1CG7fvEbteu9lbQVEnhEV+RCVKpZSoeV0YdbWNgQEBhNx5bLBNKmpqdy4flUvjVKppHRoOSLCX6RJTkpi7qypfNxnAA6OjtlXCZEnPI56gFoVQ/GXOgFW1jb4BhTnWsTvRiyZyMseRWU8L0v+7XnpFxic6bPvxfNS/x5WKrQ8VzN7Xl7PeF6++4bnZWLCU2xsCv6LmgiRexi1g9G1a1du3LgBwFdffUXPnj0pX748o0ePpkKFCvTo0YMVK1a8No/k5GTi4+P1jpTkzIc03xYqVQwA9g4OeuF29o6oVbEG0zyJj0OjScfOXj+N/WvS/HBwL0U9ihEUUjILSi3yor/axqttzQHVa9uaxnCa2Bdpvl6+hKDg4lQKq5rFpRZ5kfr5fc3OXr+zmXFfizFGkUQ+8FfbsTf07FO/6Xn597bokGlb/OHgHtw9ihEYnPnzMuLK7/x8/AfqNnj//6mCyAYKpTLXHnmBURd5X7t2DX9/fwCWLFnC559/To8ePXTXK1SowNSpU+nWrVumeUyfPp2JEyfqhfXqP4Q+nwzNnkLnUj8eOciyRZ/pzkdOmJntn5mcnMzxY9/Tqt2H2f5ZIvc4duR7vlg4V3c+ZuL0bPmc06d+4vcL55i7cFm25C9yv5+O7ufrJS/uZUPHffaa2EL8M8ePHOTLxXN05yPH58zz8sSx72nVtnOmcf68fZOZk0fSun1XSpetmO1lEiI7GbWDYWVlRXR0NMWKFeP+/ftUrKj/D6pSpUp6U6gMGTlyJIMHD9YLu3ZXndVFzfUqVKqm27kCIC01FQC1SoWDo7MuPE4di5ePn8E8CtraoVSa6Bao/UWtjsXe4dXpKad+OkpKchI1322QFVUQeUTFSlX0dnpKTc1YyK1WqXB0dNKFx6lVeL+2rSlRq/TbWpxapZsKdfHCOSIfPqBj6yZ6cWZNm0Bw8ZJMnTkvS+ojcq+yFavjG1Bcd56WlnFfi1PHvnJfK+bjn+PlE3lT+UrV8DP0vFTrPy/V6li8vA23qxfPS/0Rjji1CnsHp1fin/rpCMnJSdR4t77B/O7+eYtJYwZSp8H7tGyXeSdEiLzCqB2Mhg0b8sUXX/DVV19Rs2ZNtmzZQunSpXXXv/nmG/z8DH9B+Yu5uTnm5uZ6YWbmz7KlvLmZpZUVllYvFllrtVrsHRz5/cIZvH0zbpCJiQlci7hCvUbNDOZhamqKj18Av58/Q8Ww6gBoNBp+P3+Who2bvxL/8MG9lK9UFTs7+yyvj8i9DLU1BwdHLl44i49vxr/XxMQErkZcocF7hof5TU1N8fUL4OKFs7otZzUaDRfPn6VRk2YAtGzdgbr19ecqD+jTnW49+lChUlg21EzkNpZW1lhaWevOM+5rTly68KtuC9nExARuXL1EnYYtjFVMkcdk9rz84/wZvH1ePC+vR1yhfsNmBvPQPS8vnKFiWA3g+fPywhkaNH61Lf5wcC/lK1bFzs7hlWt379xi4ugB1KzdgA4ffpwFNRTC+IzawZg5cyZVq1alZs2alC9fns8++4yjR48SHBxMREQEp06dYvv27cYsYp6lUCh4r2lrtm5aQxG3ohR2LcKmtV/j4OhExbAX7xCYMGoglcKq07BJSwCaNG/DornT8fUPxC8gmL07vyU56Rm16jbSy//hg3tc+eMCoybMytF6idxHoVDQpFlLvt20Djc3dwq7FGHD2pU4OjlT6aW2NnbkECpXqcZ7TTI6q02bt+bzuTPw8w/EPyCI3Tu3kpScxLt1M0bEHBwdDS7sdi5UGBfXIjlTOZGrKBQKGrzflh3frMLVzYNCLm5sWb8Me0dnylWuoYs3bUw/yleuSb3GGe8SSHqWSOTDe7rrj6MecPvmVWwK2uJcyBWAp0/iiH4chTo2GoCH9+8AYO/gZPAXaZF/ZDwv27B182pc3YtS2KUIm9d9hYOjExWe/9gGMHHUACqG1dA9Lxs3a8viedPw9Q/Sf17WMfC8vHSBkRNmv/LZf96+ycTRAyhdtiKNm7fVrZ9UKpUGOyNC5BVG7WC4ublx7tw5ZsyYwe7du9FqtZw+fZq7d+9StWpVfvrpJ8qXL//mjIRBzVp1IDkpiS8Xznn+8rOSjJk8BzOzFyM+UQ8fEB8fpzuvWuNd4uPUbFq3IuPlZz5+jJ4055UpUj8c+g4n50KULitbOgpo3qodSUlJLFk4l4SnTwkuXpJxk2ZgZmamixP58AHxcS/aWrWatYiLV7Nx7UpUKhXePr6MnzTT4HQ8If7SuMUHJCcl8fXiGSQmPCUgpBTDJ8zXv69F3uNJvFp3fvP6FaaOfvF+i3Vffw5A9dqN6DVwHABnTh9n2edTdHEWzR4LQIt23WnZ4cXaQJE/NW3ZgaSkZ3y5cDaJz5+Xoyf97XkZ+YAnBp6Xm9d9/drn5ZFDe3F0LkTpMq8+L0/9dJT4ODXHjxzk+JGDuvBChV1ZsuLbbKipEDlDodVqtcYuRFb7/XqUsYsg3hIFFGnGLoJ4SySkyXtmRM4wU6YauwjiLVHKv7Cxi5CpGx/m3u33fdfsNXYR3ihv7HUlhBBCCCGEyBOkgyGEEEIIIYTIMtLBEEIIIYQQQmQZ6WAIIYQQQgghsox0MIQQQgghhBBZxqjb1AohhBBCCJHbKJQKYxchT8uXHYynqZbGLoJ4SxQ0TTR2EcRbQlWj+psjCZEFUmJlm1qRM0qlRhi7CCKbyBQpIYQQQgghRJbJlyMYQgghhBBC/FsKpfwG/1/IX08IIYQQQgiRZaSDIYQQQgghhMgy0sEQQgghhBBCZBlZg5GPabVatm/8kmOHdpCY8BT/oFJ82GsErm6emaaJuHSW77av5c6NcNSqaPqPmE25yu9kGn/VF9M5emAb7bsNov77HbKhFiK30Wq1bFq3gu8P7CEx4SmBwSX5uO9g3NyLvjbdvj3b2bl1E2pVLF7evnTvNQD/wGC9OBFX/mDDmq+4FnEFpVKJl48fYyfPwdzcHIBeXdvy+FGkXpqOnT+mRZuOWVtJkSsV7doOrz5dMSvszNPLEYSPmkb8uT8yje/5cSeKdm6LhXsRUmPVRO05yPWp89Ekp2Tk17ktRbu0xdLDDYCnEde5+dlSYn44kSP1EblXsd4d8BncHXPXQsRfDOfSwMnE/fq7wbiKAgXwHd6Toh80w8LdhYSrtwgfOYfHB48bjO87tAdB0z7l1oLVXB4yLTurIYTRyAhGPvbd9jUc2rOZzr1GMm7WSswtLPlsYn9SUpIzTZOc9AxP7wA+6DnsjfmfOXWEGxG/Y+9YKCuLLXK5HVs28t3ubfTsO4Tpc5diYWHB5LGfvrZd/fTjD6xavpg2HToze8Fyinn7Mnnsp8SpVbo4EVf+YMq4YZQuU4EZ85Yyc/6XNGzSHOXf9iJv16kbX63dpjsavd8i2+oqcg+Xpg0InDiMm599wS91W/PkUgRlN32JqbOjwfiuLRrhN3oQNz/7gp+rv8+lQeNwbdoAv1EDdHGSHkZyfco8fqnbhl/qtSX2xGlCVy/EOtA3p6olcqEirRsSPHsk16Ys5kTF5jy5GE6lvV9jVshwWwucNJBiPdpyaeBkjpVqxJ1lmyi3ZRG2ocGvxLUrXxLPHu2Ivxie3dUQwqikg5FPabVaDu7eyPttulG2Uk08vPzpMWAiqthozv5yLNN0pcpVpWXH3pSrXOu1+atiHrFu+Rx6DZ6MiYkMhL0ttFote3Z+S6u2H1AxrBpe3r70HzIKVWwMp09m/qvv7u3fUKdBY2rXbYSHpxc9+w3B3MKCwwe/08VZuXwxjd5vSYs2HfEs5o17UU+qVq+NqamZXl6WllY4ODrpDgsLee/N26BYrw+5t24LDzbtIOHqTa4MnUT6syTc2zc3GN+ufCjqX88Rue07ku4+IPbYz0Ru/w7bMiV1caIPHiP68HESb/1J4s073Ji+gPSEROzKlc6paolcyHtgV+5+/Q33Vm/j6ZUb/N5nPOmJSXh0aWkwvnvHplyfuZTH+3/k2a17/PnlRh7tO4bPoG568UysrQhdPZuLvcaQqorLiaqI/0ChVOTaIy+QDkY+9TjqPnGqGEJKVdSFWVnb4BtQnBsRF/9T3hqNhmXzx9OwWSfcPeWXvrdJVORD1KpYSoWW04VZW9vgHxhMRPglg2lSU1O5cf2qXhqlUkmp0HJcfZ4mTq3iWsRl7OzsGTWkD906NmPs8E+4cunVtrr92w10bteET/t3Z8fWjaSnp2VxLUVuozAtQMFSIcQeP/UiUKsl9sdT2JU33BmI++08tqVCsC1TAgDLYkVxercG0YcNT1tBqcSlWUNMrCyJ++18FtdA5BUKU1PsyhYn+vDPLwK1WqJ/+Bn7ymUMplGam6JJStEL0yQl41ClrF5YiYXjeLTvGDE/nMzycguR2xj1p+f+/fvTpk0bqleXN9RmtTh1DAB29k564bZ2TsSpYv5T3t9tW41SaULdxu3+Uz4i71GrYgGwd9CfKmBn76C79ndP4uPQaNKxt3d4Jc39u38CEBX5AIDNG1bRuXtvvHz8OHb4IBNGDWbeklW69R2N3m+Bj28ANgVtibjyB+tXLUMVG0PXHv2ytJ4idzFzdEBZoAApj/XvXSmPY7D29zaYJnLbd5g6OlBh11pQgNLUlLurNnP78+V68WyC/amwdz1KczPSExK50HUACVdvZltdRO5m5pzR1pIf6be15KgYrAN9DKZ5fPAE3gO6EHP8VxJv/Ilz7TBcm9UFExNdnCJtGmFbJoSfKrfK1vILkVsYtYOxePFilixZgq+vL927d6dz5864urr+X3kkJyeTnKw/9zslJRkzM/OsLGqu9/Oxfaz+YrrufNCYednyObevX+Hgnk1MnLsOhSJvDNOJf+/HI4f4ctFnuvNRE2Zky+doNFoA6jVsQu26jQDw8Q3g4oUz/HDoOzp1+RiA95u31aXx8valQIECfLnoMzp1+fiVqVTi7eZQpQLeA3oQPmIKcWcvYunlSeCUEXgP6smteV/q4iVcv8Wp2i0pYFsQlyb1KL5gKr817yKdDPGPXR48lZJLp/DOH/vQarUk3rjL3dXbdFOqLIq6UnzuaH5p2E23wYAQ+Z3RJ88fPHiQ3bt3M2fOHMaOHUvDhg3p0aMHjRo1QvkP3qI4ffp0Jk6cqBfWrc8IPuo3MruKnCuVqVgD34ASuvO01IybWJw6BntHZ114fFwMnt4B//pzIi6f40mciiEfNdGFaTTpbFr1OQd3b+Kz5bv+dd4i96lQqareTk+pqalAxkiGg+OL0bE4tQovHz+DeRS0tUOpNEH90oLuv9L8NRLyV15FPbz04hT1KEb046hMy+cfGEJ6ejqPoiJxL5r57mgib0uJVaFJS8OskP6IrFkhJ5IfRRtM4zu8Hw+/3c399VsBeHrlGiZWloTMGc+t+ctAm9Gp1aam8ez2XQCeXLyMbWhxPHt04srQSdlYI5FbpURntDXzwvptzdzFieRIw20tJVrFmVZ9UZqbYepkT/KDRwRN+5TEmxntyq5sccxdnKl2epsujbJAARyrV6BYn47ssy4JGk32VUoIIzB6B6NkyZK8++67zJ49m+3bt7NixQqaNWuGi4sLXbp0oWvXrvj5Gf7iAjBy5EgGDx6sF3buVua72eRXlpbWWFpa6861Wi12Dk5cvvgrxXwCAXiW+JQbVy9Rq8G/H6Kt+k4jipeuqBc2Z+InVHmnIdXfbZJJKpFXWVpZYWllpTvXarXYOzjy+4WzePv6A5CYmMC1iCvUb9TUYB6mpqb4+gXw+/kzVArLmA6p0Wi4eP4sDRtnLNAt7OKKo5MzD+7f1Uv78P5dypSvlGn5bt+8jlKpxM7OIdM4Iu/Tpqbx5OJlHKtX4vG+HzICFQocq1fi7oqNBtOYWFqg/fuXNk26Lu1fHYy/UyiVKM1kNOxtpU1NJe7sJZxrhxG163BGoEKBU60w7ixZ99q0muQUkh88QlGgAK7N6/Fwyz4Aon84xbHQxnpxS381nacRN7kxe7l0LkS+ZPQOxl9MTU1p06YNbdq04c8//2TFihWsWrWKGTNmkJ6enmk6c3Nz3R75fzEzi8/u4uZ6CoWCek3as/vbFbi6eeBc2J1tG5bi4OhM2Uo1dfFmjs3YMarOe20ASHqWSNTDF1/yoh894M7NCGwK2uFUyBUbW3tsbO31PsvEpAB29k4UcffKiaoJI1IoFDRu2potm9ZQxK0ohV1d2bh2BQ6OTlQMq6aLN2HUICqGVadRk4wtZJs0b8PCudPx9Q/CPyCIPTu3kJz0jNp1G+rybdqiHZvXr8TL2xcvHz+OHj7A/Xt/8umojF+SI678wbWIK5QoVQYLSyuuhl9i5fJF1KhVF5uCBXP+jyFy1J2layi+YCrx5y8Rf+4PPD/uhImVJQ827QCg+MJpJEc+4vrU+QA8PniMYr0+5Mkf4cSdvYiVlye+w/vz+NAx3Rc6v9EDiT58nKT7DylgY41ri/dwqFKBs217GqmWIje4NX8lpVfMRH3mD+J+vYjXJ50pYG3J3dUZIxClV84k6X4UEWPmAmBfsRQWbi7EXbiChZsLAeP6o1AquTHnKwDSnybw9NI1vc9IT0gkNUb9SrjIPfLKbk25Va7pYLzM09OTCRMmMH78eL7//ntjFyfPatT8Q5KTnrFyyTQSE54SEFyaIeMW6K1PeRR5nyfxat35retXmDm2l+5844qMtRxVa71HjwETcqroIhdr1qo9SUnPWLpwDgkJTwkKKcnYybP12lXkwwc8iX+xDWPVGrWJi1Ozad0K1KpYvH38GDNptt5i8cbNWpOSksLK5Yt4+uQJXt6+jJvyGa5F3AEwNTXjxI8/sHnDKtJSUyjsUoQmzVrTpHmbnKu8MJqonfsxc3LAd1g/zAs78+RSOGfb99It/LZwL6L3S/CteV+CVovfiP6YuxYmJUZF9MGjXJ++QBfHzNmREgunYe5SiLQnT3hy+Spn2/Yk9kfZ5edt9vDbfZgVciRg/CcZL9q7cIXTjT8i5fnCb0uPInqjY0pzcwImDsTKx4P0p4k82n+M812GkRb3xFhVEMLoFFptJuPEOcDb25vffvsNJyenN0f+P5y8IiMYImcUNE00dhHEW+Jh1TrGLoJ4S6TEphq7COIt8V5qhLGLkKk7HzczdhEyVWzZDmMX4Y2MOoJx69YtY368EEIIIYQQIovJi/aEEEIIIYQQWUY6GEIIIYQQQogskysXeQshhBBCCGEsin/wLjaROfnrCSGEEEIIIbKMdDCEEEIIIYQQWSZfTpEyURht513xllEgbU3kDMfjPxq7COItYaFMNnYRhDA6edHefyMjGEIIIYQQQogsIx0MIYQQQgghRJaRDoYQQgghhBAiy0gHQwghhBBCCJFlpIMhhBBCCCGEyDL5chcpIYQQQggh/i150d5/Ix2MfEyr1bJtwzKOHNpBYsJTAoJK0aX3cFzdPDNNE37pLHu3r+P29XDUqmgGjJxF+crv6MXZtnEZp44fIiY6igIFTPH2DaJVp974BZbI5hqJ3ECr1bJp3UoOHdhDYsJTgoJL8HHfwbi5F31tun17trNj6ybUqli8vP34qNcn+AcG666PHTGAS79f0EtTr2ETevUbAsCT+DjmzZ7Cnds3eRIfj529PRUrV6Vj5x5YWVlnfUVFnqDVatmyfjlHDu4iIeEJAcGl6NZnGEXcPDJNc+WPc+zZtp5bNyJQx0YzaNQMKoTVzMFSi9zGWPe1lz2Jj2NQv+7ExkSzdvNurG0KZk3lhDAC6Z7lY3u3reHg3s107T2CCbNXYG5hyawJn5CSkvke58lJSXh6+dO559BM47i6efLhx0OZvmAjY2csw7lwEWZN6E98nCo7qiFyme1bNrJ391Z69R3MjLlfYG5hyeSxQ1/brk78+AMrly+hTYcuzFmwHC9vXyaNHYpard9m6tZvzNdrt+qOD7v10l1TKJRUrFyNkeOmsmj5WvoPGsHF82f4ctHcbKuryP12b13HgT3f0q3PMCbP+RoLC0tmjBv4xvtcMW9/uvZ69UueeDsZ6772ssWfz8LL2zdL6yWEsUgHI5/SarXs372J91t3o1ylmnh6+dNz4ATUsdGcOXUs03Sly1WhdafelA+rlWmcKjUbUCK0IoVd3Snq6UvH7gN5lpjA3dvXsqMqIhfRarXs2bmFVm0/oGJYNby8fflkyEhiY6M5ffJEpul2b/+Wug3e4926DfHw9KJnv8GYW1jww8Hv9OKZWZjj4OikO14embApWJAG7zXFzz+IwoVdKRVajgbvNePypYvZVl+Ru2m1Wvbv2kyzNl0oX7kGnt5+9B40DnVsNL+dyvzFhKHlw2jzQU8qhL2Tc4UVuZYx72t/2b93JwkJT2naom2W108IYzB6B2PRokV8+OGHbNq0CYC1a9cSEhJCUFAQo0aNIi0tzcglzJseRz0gThVDidIVdWFW1jb4BBTnesTvWfY5aamp/HBgB1bWNnh6B2RZviJ3iop8iFoVS+nQcrowa2sb/ANDiAi/bDBNamoqN65HUOqlNEqlklKh5V5Jc/zI93Ru/z4D+nRh3aplJCclZVqW2JhoTv38I8VLlP6PtRJ51aOoB6hVMZQIraALs7K2wTcghGvhfxixZCIvMfZ97e6ft/lm42o+GTwKhULeHi3yB6OuwZgyZQqzZs2iXr16DBo0iDt37jB79mwGDRqEUqlk3rx5mJqaMnHixEzzSE5OJjlZfwgzJSUZMzPz7C5+rqZWxQBgZ++oF25n70jc82v/xblfj7N4zhhSkpOwd3Bm+MRFFLS1/8/5itxNrYoFwM5Bv13Z2zugen7t757Ex6HRaLC3fzXN/bt/6s6r16xDocIuODo5c/vWDdau/JL79+4yfMxkvXRzZ07i9C8/kZKcTPmKVegzIPPpfCJ/i8vm+5x4OxjzvpaamsLcWZPp3K0XhQq7EBX5ICurJoTRGLWDsWrVKlatWkWLFi24cOEC5cqVY/Xq1XTs2BGAoKAghg0b9toOxvTp01+5/lHf4fToNzJby57b/HR0Pyu/mK47HzJ2XrZ+XnDJ8kydv44n8WqOHNzBwlkjmTB75SsPepG3HTtyiC8XfaY7Hz1hRrZ9Vr2GTXT/XczLB0dHJ8aPGkzkw/u4FnHXXevaoy9tOnTmwf17rF+9nJXLl9Cz76BsK5fIPU4cPcDXi2fqzoeNm2PE0oi8Kjfd19atWk5RD09q1q6XbWUQ/5KMJv0nRu1gPHjwgPLlywNQunRplEoloaGhuutly5blwYPX9+ZHjhzJ4MGD9cIu3s58WkV+VbZidfwCi+vOU1NTAIhTx2Lv6KwLj1PHUiwLpjJZWFhiUcQDlyIe+AWW5NNeLTn2/S7eb9XlP+ctco+KlaoS8NKOKKmpqQDEqWJxdHTShavVKrx9/AzmUdDWDqVSiVqt/0ugWq3C3iHzDulfO7E8fKDfwfhrHnNRj2IULFiQ0cM+oXX7D/XKI/KnchWr4RcQojtP+6s9qmNx+Pt9zkembArDctN97fcLZ/nzzi1anaitF69z+6a0avsB7Tp1/f8qJ4QBixcvZvbs2URGRlK6dGkWLlxIxYoVM42vVqsZPXo027ZtIzY2lmLFijF//nwaNWr0jz/TqB0MV1dXLl++jKenJ9euXSM9PZ3Lly9TvHjGF+VLly5RuHDh1+Zhbm6Oubn+dCgzM222lTm3srSyxvKlhWNarRY7BycuXfxV96B9lviUm1cv8W6Dlln++VqthrTnnRqRf1haWWFpZaU712q12Ds4cvHCWbx9/QFITEzgWsRlGjR632Aepqam+PoFcvH8WSqFVQdAo9Fw8fwZGjVunuln37p5HcjoUGRGo834ty5t7+1g6D5n7+DEpQu/4fX8PpeYmMCNq5ep06iFsYopcrncdF8bNnqS3jTv69ciWDx/JlNnLcSliNt/q6gQwObNmxk8eDBLly6lUqVKzJ8/n/r16xMREWHwO3ZKSgp169alcOHCbNmyBXd3d+7cuYO9vf3/9blG7WB07NiRDz/8kKZNm3L48GGGDRvGp59+SkxMDAqFgqlTp9KqVStjFjHPUigUNGjSjp3frMC1iAeFXNzYsmEp9o7OlKv8Yr/36WP7UL7yO9R9rw0ASc8SiXp4T3f9cdQD7ty8inVBW5wLuZKU9Ixd366kbMXq2Ds48yRezfffbUEV85iKVd/N8XqKnKVQKGjctBVbNq2liFtRXFyLsHHt1zg6OlMxrJou3vhRg6kUVo1GTTK+5DVp3pqFc6fj5x+If0Awu3duITkpidp1GwIQ+fA+Px49TLnylShoa8vtWzdZuXwxISVK67ZtPPPrKdRqFX7+gVhaWvLnndusWbGUoJASFHYpkvN/DGF0CoWCBu+3ZfvmVbi6eVDIpQjfrluOvaMz5SvX0MWbOrof5cNqUr9xayDjPhf5t/vc7ZtXsbGxxbmwa47XQxiXMe9rL4/OQsbaDoCiHp7yHgwjUyjzxxSpuXPn0qNHD7p2zRgNW7p0KXv37mXFihWMGDHilfgrVqwgNjaWn3/+GVNTUwC8vLz+7881agdj4sSJWFpacvLkSXr06MGIESMoXbo0w4YNIzExkSZNmjB58uQ3ZyQMeq/FhyQnJbFiybSMF+0Fl2bo+M/1FsA/irzPk3i17vzW9StMG9Nbd75hxXwAqtV+j54DxqNUKnl47zYLftjLk3g1NgXt8PEPYcz0ZRT1lP273wbNW7UnOSmJpQvnkJDwlOCQkoydPEuvXUU+vE/88wclQLUatYmPU7Nx3UrUqli8ffwYO2mWbipBgQKmXDx/hj07t5Cc9AznQoUJq1qDVu0+0OVhZmbO9/v3sHL5ItJSU3FyLkzlKtVp0bpDzlVe5DpNWnYiOekZXy2akXGfCynFiInz9NpjVOR93Rc3gJvXw5kyqq/ufN3XCwCoUbsRvQaNzbnCi1zDWPc1If4NQxscGZrRk5KSwpkzZxg58sW6ZKVSSZ06dTh58qTBvHft2kVYWBh9+/Zl586dFCpUiA4dOjB8+HBMTEz+cRkVWq02380nOh0e9+ZIQmQB6wKJxi6CeEskad7unfFEzrFQZv5yOSGyUnG/3Dv6fH9A7n0nyXKH4Fc2OBo/fjwTJkzQC3vw4AHu7u78/PPPhIWF6cKHDRvGsWPH+OWXX17JOygoiNu3b9OxY0f69OnD9evX6dOnD5988gnjx4//x2U06giGEEIIIYQQ4p8ztMHR30cv/i2NRkPhwoVZtmwZJiYmlCtXjvv37zN79mzpYAghhBBCCJEfGZoOZYizszMmJiZERUXphUdFReHqani9WZEiRTA1NdWbDhUcHExkZCQpKSmYmZn9ozIa/U3eQgghhBBCiKxlZmZGuXLlOHz4sC5Mo9Fw+PBhvSlTL6tatSrXr19Ho9Howq5evUqRIkX+cecCpIMhhBBCCCGEHoVSmWuP/8fgwYP5H3v3HRXVsQBw+LdLFRR2QUVAlN5sKDY0aoy9Yy+xp9i7Ro29t9ij0Rg7scSOXWPsUZOnAkYFBbuChd0FBKm77w90dWVRUXRR5zvnnvd27szcGZzc3blT7rJly1i9ejWXL1+mV69eJCYmaneV6ty5s84i8F69eqFQKBgwYABXrlxh9+7dTJ06lT59+mR3Cb3EFClBEARBEARB+AS1bduWhw8fMnbsWGJiYvDz82Pfvn3Y2dkBcOvWLaQvdFqcnJzYv38/gwYNonTp0jg6OjJgwACGDx+eo+uKXaQE4R2IXaSED0XsIiV8KGIXKeFDycu7SN0b1N7QRciWw9z1hi7Ca32SIxhyE5WhiyB8JvJlJBi6CMJnQiK1MXQRhM/ErXLipanCh1Ei/rKhiyC8J2INhiAIgiAIgiAIuUZ0MARBEARBEARByDWf5BQpQRAEQRAEQXhbEqnE0EX4qIkRDEEQBEEQBEEQco3oYAiCIAiCIAiCkGvEFKlPhEaj4fegNezft5fExMf4+Jagd5/+ODo6vjLdrp3BbN2yCaVSgYuLKz169cHLy1t7ft/e3Rw5cpioyEiePEliwx9byZ8/v/Z8WFgoP44YpjfvOfMW4unplTsVFPKM7bv38sfWHSiUKtxcnOnX4xu8PT2yjX/0xN+sDFpPzIOHFHWw57uuHalU3l9v3LmLlrJr3wF6f9uNls0aa8NHT5pG1LUbKOPiKJDfknJlSvNd104UtBU7K33KNBoN64NW8uf+3SQmPsbbpyQ9+gzCwbHoK9Pt2bWN7Vs2olIqcHZx49ue/fH08tGb/6RxIzh/9h9GjJ5EpYAvtOfCQs6ybu1Kbt68hrmZOTVr1ePrLt9iZGSU6/UU8p5i33XApX93TO0KkvBfOJeHTSHu7IVs4xfv3Zli37TDvKg9qbFK7u84wJXxc1CnpGaJ6zLoW7wmDOHG4jWEj5j2PqshCAYjRjA+EVs2/8HO4O306duf2XMXYG5uztgxI0lNzXpze+bY0SP8tmwp7Tt0ZP7Cxbi4ujJ2zI+oVEptnJSUFPz9y9OmbTu9efj4+LI2aIPOUbdeA+yKFMHDwzPX6ykY1uHjJ1ny2yo6t2/DknmzcHMpzvCxk1Cq9L975uLlcCbPmkuDurVYOv8nqlauyNgpM7l+81aWuCdOneFyxBVsbbJ2GvxKlWTM8CGsXrKA8SOHcS/mPhOm/5Tr9RPylm2bN7B751Z69BnEjDmLMTM3Z+KYH155Xztx7C9WLvuFth26MHvBrzi7uDFxzA8697Vndm7fjESSdZ719WuRTBo3krL+FZizYBlDRozlnzN/s3blr7laPyFvKtKiAd5ThxM5fRF/V2tJwoUIym9dhmlB/Q807Fs3wnP8YCKnL+JEhUb813c0RVo0wHPcoCxxrcqVxKlbW+IvhL/vagiCQRm0gxEdHc3YsWP56quv8PHxoUSJEjRp0oTly5eTkZFhyKJ9VDQaDTu2b6Ntuw5UDqiCi4srg4f8gCI2llOnTmabbvu2LdSr34A6detRrFhx+vQdgJmZGQcP7NfGaRbYgtZt2uHlnfXpH4CJiQlyGxvtUcDKijOn/6Z27bp6v7iFj9vm7TtpWK829Wt/hXMxJwb27oGZmRn7Dh7SG39r8G4qlCtL2xaBFHcqSreO7fFwc2H7rr068R7GxrJw6W/8OGQAxsZZnxC3CmyCr7cndoULU8LHm/atmnM54grp6envpZ6C4Wk0Gnbt2Ezrtp2oFPAFzi5uDBgyEoXiEWdOncg2XfC2TdSp34hadRrgVMyZnn0HY2ZuzqEDum3uelQkwdv+oO+AH7LkcfL4YZxdXGnboQv2Do6ULOVHl+492Lt7O0+SxMs1P3XOfbtwe/Um7v6+jcSIKC4OHE/Gk2QcO7XQG19WqSyq0+eI3rSbJ7fuEfvX30Rv3o21fymdeEaWFpT5bRYX+48lXRX/IaoiCAZjsA7G//73P3x8fNizZw9paWlcvXoVf39/LC0tGTp0KNWrVychQbzE7E3cj4lBqVTg51dOG2ZpaYmXlzfhl/W/xCYtLY3IyKv4+ZXVhkmlUvz8yhIe/vYvvjlz+hQJCQnUqVvvrfMQ8qa0tDSuREZRrkxpbZhUKqWcX2kuRVzRm+ZS+BX8/UrrhJUv68el8AjtZ7VazfQ5C2jTohnOxYu9thzxCQkcOnKMEt5eGBuLWZ6fqvsx0SiVCsr4PZ9OZ2mZHw8vHyLCL+pNk5aWRlTkFZ00UqmU0n7ldNKkJCczZ9Zkvus1ALmeEbO0tDRMTE11wkxNzUhNTSUqUn9bFz4NEhMTrPxKEHv41PNAjYbYI6eQVfTTm0Z15jxWfiW0HYp8zkUpVLc6Dw8c04nnO3sMD/cfJfbIKX3ZCHmMRCrNs8fHwGClHDhwIIMGDeJ///sfx48fZ9WqVVy5coUNGzZw7do1kpKSGD169GvzSUlJIT4+XudITUn5ADXIO5RKBQAyuUwnXCaTo1JmnRYAEB8fj1qtRiaXZ0mjVCjeuiwHDuyjbDl/ChYs9NZ5CHlTXHwCarUa+UvtTC6zRqFU6U2jUKmQy6xfii9DoXoef8OW7RhJjWjRpNErr//rqrU0atWB5h26cv/hIyaOHvE21RA+Eqqn9zVrPfeoZ+delhAfh1qtxlr26jQrli3C26eEzpqLF5UtV4GIyxc5fuQQGRkZxD56yB/r1wCgVMS+dZ2EvM/UVobU2JjUh7r/zikPYjGzK6g3TfSm3UROXUil/UHUjQ2jRthBFMf/5drs51PqirRsiFUZX66Mn/Neyy8IeYXBOhjnzp2jU6dO2s8dOnTg3Llz3L9/H7lczsyZM9m8efNr85k2bRrW1tY6x5Ili99n0Q3u8OFDtGrRVHuk55HpZI8ePeT8ubPUrVvf0EURPhJXIqPYGrybHwb2fe2UurbNm7Fk/k/MmDgWI6mUGXMXoNFoPlBJhfft6OGDtG/ZQHukZ7yf6W//nD7JhbDzdP++b7Zx/MpVoHP3HixZNJc2gXXp831nypWvBIi98YWsbL6ogOuQ77k0eBJ/V2vJuQ79KFSvOm4/9ALA3LEIPjNGEvrtML2LvgXhU2Sw+QWFCxcmOjoaV1dXAO7fv096ejpWVlYAeHh4oHiDJ+kjR45k8ODBOmG378TkfoHzkEqVAnR2ekpLSwNApVRhY2OrDVeplLi4uunNw8rKCqlUmmWEQ6VS6p0y8CYOHthPgQIFqFQ54K3SC3mbtVUBpFIpypdGK5SqOGxeGtV4xkYmy7IAXKlSYSPLjH/h4mVUcXG0795De16tVrNkxWq2BO9i3fIlz69vbYW1tRVOjg4UdypKu27fcyniCiW8xU5ln4KKlari6eWr/ZyWlvlDLE6p1HNfc9ebRwEra6RSKXGqrPc1mTzzvnYh7Dwx0ffo2KaxTpyZU8fhU6IUk6fPA6BZ8zY0DWyNUhGLZf4CPLgfQ9DqZdgVcXjnugp5V2qsCnV6OqaFbHXCzQrbknL/kd407qP7c29DMHfWZD4UfXzpKkaW+Sg5fwJRs5Zg5VcCs8IFqXJ8izaN1NgYedXyFPu+AwcKlgG1+v1VSngr4mHCuzFYByMwMJCePXsya9YszMzMmDRpEjVq1CBfvnwAREREvHaLVQAzMzPMzMx0wkzN9E8L+lRYWFhgYWGh/azRaJDLbQgJPY+rW2aHIikpkYiIcBo0aqw3DxMTE9zdPQgNDSGgSlUg84ddaEgIjZs0zXGZNBoNf/55gK9q1RHz4j9RJiYmeLq7cT7sAl8EZD7NVavVnA8NI7BRA71pfL09ORcaprPl7NmQMHyfdgpq16xBuZfWaAwfO4k6NatTv/ZX2ZZF/fTL+FnnWvj45bOwIJ+e+1pY6Dlc3DI7FElJiVyNuEz9hs305mFiYoKbuydhIee005/UajUXQs7RoHFzAFq06kDturrT8Qb26U6373pToWIVnXCJRIKNbea0mONHD1GwUGFc3bLfkln4+GnS0ogPuYjtl5V5sPvp5hUSCbY1KnPz19/1pjHKlw+N+qXR1GczCyQSYo+e4kQl3e/VUr9M4fGV61yf+5voXAifJIP9Epw8eTLR0dE0adKEjIwMAgICCAoK0p6XSCRMmyb2h34TEomEZoHN2bhhHY4OjtjZFSFo7SpsbG0JCKiqjffjyB8IqFKVJk0yv5wDm7dk7pxZeHh44OnpzY4dW0lOSaZ2necLtJUKBUqlkuh79wC4ceM6FvksKFS4EAUKWGnjhYaGcD8mhrr1xPSoT1mrwCbMmLsQT3c3vD092LJjF8nJKdR72hmYPmcBBW1t+LZLRwBaNG3EoJFj+WNbMJXLl+Pw8ZNciYxicN+eQOaoiLVVAZ1rGBsbYSOX41Q08wHD5YgrRFyNpKSvDwXyW3Iv+j4rf1+Pg30RbUdF+PRIJBIaN2vFpg1rsXdwxK6IPevWrsDGpqDO2omxPw6mckA1GjbJ7EA0bd6aBXOm4+bhiYenD7t2bCY5OZladTLvTc92vHtZoUJ22BWx137etmUD5fwrIpFIOP33cbZtXs/QEePEezA+Azd+Xk2pJdOIO/8fcf+7gHPvzhhZ5ONu0DYASi2dTsq9+1yZMBeAh/sO49ynK/Fhl4n7XygWrsVxH92fB3uPgFpNxuMkHl++qnONjMQnpClUWcIF4VNhsA5G/vz52bhxI8nJyaSnp+u8vA2gbt26BirZx6llqzYkJyezcOE8Eh8/xrdESSZOnIrpCzuhxERHEx/3fLpK9RpfEhcfR9DaNSiVSlxdXZk4cQryFxZV7tmzi/Xrnnf8RvwwBICBg4ZSu87zf6OD+/fh4+OLk9PrdwESPl41q1UlLi6OVb9vQKlU4ebqwvQJo7VTpB48fKSzlqKEjzejhg5kRdB6Vqz5HUcHeyaO+gGXN9gt6hkzMzOOnzrDqnUbSU5OwVYup4K/H1+3bYWpiUluV1HIQ5q3akdy8hN+WTj76QtESzFm0oyX7mv3iI9/fl/7ovpXxMfFsSFoVeYLRF3dGDtxhnaK1Js6979/2LwxiPS0NJxd3BgxZjL+T9dhCJ+2mK17MS0ox+PH/pjZFST+wmX+1/J77cLvfEXtdUYdomYuQaPR4DGmP+b2dqQ+UvBg3xGuTpxnoBoIguFJNJ/gKsmrUTcNXQThM5EvQ2ylLHwY8VLx1nLhw7hVrpahiyB8JurHv/22+O/b/eGdXh/JQOxmrDV0EV7r49hMVxAEQRAEQRCEj4JYjSsIgiAIgiAILxC7SL0bMYIhCIIgCIIgCEKuER0MQRAEQRAEQRByjehgCIIgCIIgCIKQa0QHQxAEQRAEQRCEXPNJLvKWpdw3dBGEz4RaIl66JXwYRkYZhi6C8JlwOveXoYsgCMJH7pPsYAiCIAiCIAjCW5OKST7vQvz1BEEQBEEQBEHINaKDIQiCIAiCIAhCrhEdDEEQBEEQBEEQco3oYAiCIAiCIAiCkGvEIu9P1JY9B1m/fQ8KVRxuzk4M+rYzvp5u2cb/6+QZflu/hZgHjyhqb0evzm0J8PfTnleo4vhlzQb+CfmPx4lJlCnhxaBvO+PkUOQD1EbIS7buOcCGbbuetq1iDPiuC76e7tnGP3zyNMvXbSLmwSMc7YvQs3M7AsqX1Z6vHthBb7peXdrTvnkTAG7fjWbx6nX8dzmCtPQM3Jyd+KZDa8qVKpG7lRPyFI1Gw7qg1RzYt4fExMf4+JagV58BODgWfWW63Tt3sG3LHyiVClxc3Pi+V188vbwBSEiIZ13QakLOneXhwwdYWVtTOaAqX3fqiqVlfp18Dh3cz/Ztm7l39w4WFpZU/aI6Pfv0f2/1FQxHo9GwIWglB/fvIinxMd4+Jfm+z+DXtrW9u7axfcsGVEoFzi7ufNuzPx5ePtrzY0YM4OKFUJ00dRs0oWffIdrPYSFnWb92BTdvXsPczJwva9Xn6y7fYGQkfqIJHy+JRqPRGLoQue3hpX8MXQSDOnTiNJPnL2Voz274errxx859HP77H9b/PBO5zDpL/AvhV+g7ago9OrahSnk/Dh4/xe/bdrHip0m4FndCo9HQc8REjI2N6Nu1A5YW+dgQvJcz58MIWjCdfObmBqhl3vC5bVN76MQpps77hSG9uuPr6c6m4L0c/vsMvy+anW3b6v/jRL7v1JaA8uX489hJ1m3byW+zp+Ja3AmAWKVKJ82ZcyHM+HkZ63+Zg0MROwA69B5MUfsi9OjUDlNTEzbt3Me+v46xfslcbOWy913tPEFlWtjQRfjgtmzawOY/1jNg8A/YFbHn97UruXnjOouWrMDU1FRvmuNHDzN39kx69x2Ap7cPwdu3cPLEMX75dSUymZybN66zLmg1terUw6lYcR7cv88vP8/D2cWVEaPGafPZvnUz27dtolv37/H09iE5OZkH92OoVLnKh6q+waRrPr8ftls3rWPrpt/pP2gkhYvYs37tCm7duMb8JaswNTXTm+bEsb9YMHsaPfoOxtPLh13bN/P3iSMs/HUtMpkcyOxgODg40a5jN206M3NzLCwsAbh+LZLhg3rRqm1Hqn1ZC0XsI5b8PAf/CpXp+m3v919xAyvhbm/oImTr4ehur49kIIUmrzR0EV5LTJH6BG0I3kuTOl/SqFZ1XJwcGdazG+ZmZuw6dExv/E27DlCpbGk6NG+Es5Mj33VohaerM1v2/AnA7XsxXLwSyZAeXfHxcKWYoz1De3QlJSWVP4+f/pBVEwzsjx17aFy3Jg1rfYmzU1GG9PoGczMzdh86qjf+5p37qFiuDO2bN8HZyZFvv26Dp6sLW/cc0Maxlct0jhNnzlK2pK+2c6GKj+fOvRi+btEUN+diODnY07NzO5JTUrh+6/YHqbfw4Wk0GoK3b6VNu6+pHFAVFxdXBg0ZjiI2ltOnTmabbse2LdSt35DadetTrFhxevcdiJmZGX8e2AdAcWcXRo4eT8VKAdjbO1DGrywdu3TnnzOnycjIfNfI44QEgtauZNCQ4dSoWQt7ewdcXFw/i87F50ij0bBrx2Zate1ExYAvcHZxo/+QkSgUj/jn1Ils0+3ctok69RtRq04DnIo506PvYMzMzfnrwB6deKbmZshtbLXHs84FwMnjhynu4kqbDl2wdyhKiVJ+dO7ek327t/MkKem91VkQ3jfRwfjEpKWlcyXqBuXLPJ86IpVKKV+6BBcjIvWm+S8iUic+QCW/Uvx35WpmnunpAJiZmOjkaWpiQtjliNyugpBHZbat65QvXVIbJpVK8S9TkosRV/WmuRhxFf8X4gNULFs62/gKVRynzobQqPaX2jDrAgUo5mjP/iPHeZKcTHpGBjv2H0JubYWXm8u7V0zIk+7HRKNUKijjV04bZmmZH08vHyIuX9KbJi0tjcjIK/i9kEYqlVLGrxzh4frTACQlJmJhYYGRUeaIZMj5s2jUamJjH9G7R3e6dWrHjKkTefjwQS7VTshL7sdEo1IqKOPnrw2ztMyPh5cvEdm0m7S0NKIiIyj9QhqpVEppP/8saY4f/pMu7ZsyoHdXglb9Skpysk4+L4/GmZqakpqaSlSk+H4VPl4GHwdNTU1l+/btnDp1ipiYGACKFClClSpVaNasWbbD4IJ+cQkJZKjV2FjrTlexkVlx8+49vWkUKlWW6S1ymTUKZRwAxR3tsStky5KgPxjWqzv5zMzYuHMfD2IVxD6NI3z6nrWtl9uKjbU1t+5k37ZsXm5b1tYoXpoW9cy+v45hkc+c6gEVtGESiYQ5E35k1LQ51G//DVKJBJm1FbPGjaBA/vx68xE+fkqlEgCZXK4TLpPJUCoVetPEx8ehVqv1pJFz97b+0a74uDg2rg+iXoNG2rCYmGg0Gg2bNq7nux69sbC05Pc1Kxk7ajgLFv2KyQsPW4SPn+ppe7KW2+iEy2TybNtawrO2Jsua5u7tW9rP1WrUplBhO2xsC3LjehRrVy7l7p3bDB89CYCy5Sqwe8dmjh85RJVqX6JSKti0fg0ASoX+awsfhkS8aO+dGLSDERkZSb169bh37x6VKlXCzi5zSsT58+dZsmQJRYsWZe/evbi7Z7+ANCUlhZSUFN2w1FTMRMck1xgbGzNl+ACm//wbDTv1xEgqxb9MCSqXK82nt4JHMKQ9h45Qp3pVnf9+NRoNc39dhczaip+njsXU1JTdBw8zcspPLJ01iYI28lfkKHwsjhw+xOKFc7Wfx06Y8t6vmZSUyMRxo3AqVpz2X3fWhqs1GtLT0/m+Zx/KlisPwNDho+jydRsuhIVQzr9CdlkKH4Gjhw+y9OfZ2s+jxk9/b9eq26CJ9v8Xd3bFxsaWcT8OJib6LkXsHfErV4HO3XuydNEc5s+egomJKa3bdeLSxTAkUsl7K5cgvG8G7WD06tWLUqVKcf78eaysrHTOxcfH07lzZ/r06cP+/fuzzWPatGlMmDBBJ2xo72/5oc9376XMeZ11gQIYSaUo4nRHFhSqeGxlMr1pbGQylCrd+EpVHDby50+evd1cWDV3Co8Tk0hLT0dubcV3P4zDW0xR+Ww8a1svtxVFXBw22Sy0tpHJULzctrKJH3oxnFt3oxk/VHeXnnNhFzn1v3PsDlqGpYUFAF5uLvwbeoF9h4/TsWXTt6+UkGdUrBSg3ekJID0tDQCVUomNja02XKVS4eqqf0c8KytrpFIpqqejH8/TKJG91BFNSkpi/JiR5LPIx49jJmBs/Pzr0Obpk2ynYsW1YdbWMgpYWYlpUp+AipWq4vnCTk9pT9tanFLxUltT4uKq/wFngWdtTaU7yqBSKZG9NBLyomc7TEXfy+xgADRt3oYmga1RKmKxzF+Ah/djCFq9DLsiDm9XQUHIAww6/nPy5EkmT56cpXMBYGVlxaRJkzh+/Pgr8xg5ciRxcXE6x4DvuryvIud5JibGeLo5czbs+RxQtVrN2QsXKeGl/0ZZ0sud/4Vd1An7N/Q/Snp6ZImb39ICubUVt+/FEBF1nWqV/LPEET5NmW3LhbMvtBW1Ws25sIuU8MraVgBKeHlwLuw/nbB/Qy7ojb/7zyN4ubng7lJcJzz56QilRKJ7u5JKpGjU6reqi5D3WFhY4ODgqD2cihVHLrchNPS8Nk5SUiJXIi7j5eOrNw8TExPc3T0JDT2nDVOr1YSFnMfb+3mapKRExo0ejrGxMaPHTsoyFdfHN3Pd0N07z6dVJSTEkxAfT+HCdrlSX8Fw8llYYO9QVHs4FXNGJrch7IV2k5SUyNWIS3h5Z9/W3Ny9CAt5ua2dzTYNZO4aBSB/oSMDmVNBbWwLYmZmxvGjhyhYqDCubvrvq4LwMTBoB0Mmk3Hjxo1sz9+4cQNZNk/dnzEzM8PKykrn+NynR7Vr2oCdB4+w96/j3Lh9l5+WruJJcgqNalUHYNL8JSxZu1Ebv3Xjupw5f4H1O/Zw8849lm/YSnjUdVo2rK2N89fJM5z77zJ3Yx5w/MxZBo2fQbWK/lT0K/XB6ycYTptmDdl18DB7/zrGjdt3mb1kBU+Sk2lYqwYAU+YtZunaDdr4rZrU58z5MDZs383NO3dZsX4zEVHXaNGwrk6+iUlJHPn7DI3r1MxyzRLeHhSwtGTq/F+IvH4z850Yq34n+sEDnfdpCJ8WiURC08AW/LHhd86c/psb168x96cZ2NjaUjmgqjbe6JHD2LVzu/Zzs+YtObBvD4f+PMDtWzf5ZdF8klOSqVWnPpD5w3HsqOEkJyfTb+BQkpKSUCoUKBUK7S5SjkWLUqlyFZYtXczlSxe5eeM682bPxLGoE6VK+33IP4PwAUgkEho3a8XmDWv55/RJbt64xoLZU7GxKUjFgC+08cb9OJg9O7dqPzdp3po/9+/i8J/7uHPrJksXzSUlOZmv6jQAICb6Ln+sX0PU1Qge3I/mn9MnWTB7Gr4ly+Ds8nwUbvuWDdy8cY1bN6/zx/o1bNu8jm969NNuOiAIHyODTpH69ttv6dy5M2PGjKFWrVraNRj379/n0KFDTJ48mX79+hmyiB+lWl9URhWfwG8btqBQxuHuUozZY4dpF9vefxiLVPJ8bmcpb0/GDerFsnWb+TVoE0Xt7Zg2YqD2PQWQ+a6Cn1euQxEXh61cRv0vv6Br68APXTXBwGp9EYAqLp4V6zejUKpwdynOT+NG6LStF0caSnl7MnZwH377fRPLgjZS1KEIU0YM1mlbAIeOn0Kj0VCrWtZtQGVWmQu6lwVtZODYKaSnZ+BSzJGpI4dkGe0QPi0tWrUlOTmZRQvnkvj4Mb4lSjJ+4nSdEYeY6HvEvzAltFqNmsTFx7Fu7SqUSiWurm6MnzgN+dOF31GRV7kSEQ5Aj28661xv2cog7OwyXx46aOhwfvv1FyaOH4VUIqFEqTKMnzRNZyqV8Olo3qo9KcnJLFn409OXOpZizKSZOu/AiIm+S3z887b2RfWviI9TsT5oJSqlAhdXd8ZMnKmdImVsbEJYyFl27dhMSvITChYqTEDV6rRq10nn2uf+d4bNG9eSnpZGcRc3RoyZQrnylT5MxQXhPTH4i/ZmzJjB/PnziYmJQfL0R69Go6FIkSIMHDiQH374Icd5fu4v2hM+nM/tRXuC4XyOL9oTDONzfNGeYBh5+UV7seO/NXQRsmU7/jdDF+G1DH4XGT58OMOHD+f69es629S6uIjFw4IgCIIgCILwsckzm/y6uLgQEBBAQECAtnNx+/ZtunfvbuCSCYIgCIIgCILwpvJMB0MfhULB6tWrDV0MQRAEQRAEQRDekEGnSAUHB7/y/LVr1z5QSQRBEARBEARByA0G7WAEBgYikUh41TpziUS8yVIQBEEQBEEQPhYGnSJlb2/P1q1bUavVeo9z5869PhNBEARBEARByE1Sad49PgIGHcHw9/fn7NmzNGvWTO/5141uZCfJ1PpdiyYIbyRVYm7oIgifCXXeXjInfEIyNGL7bUEQ3o1BOxjDhg0jMTEx2/Pu7u4cPnz4A5ZIEARBEARBEIR3YdAORrVq1V553tLSkho1anyg0giCIAiCIAgCSKRiDfC7EGPugiAIgiAIgiDkGtHBEARBEARBEAQh14gOhiAIgiAIgiAIuUZ0MARBEARBEARByDUGXeQt5J7gXbvZtGUbCqUSVxcX+vT8Hm8vz2zjHzt+glVBv3P//gMcHRz4tlsXKlYoD0B6ejqr1gTxz//OEh0Tg6WlJeX8yvBN187Y2toCEBp2gWEjR+nNe+Hc2Xh5euR+JYU8QaPR8HvQGvbv20ti4mN8fEvQu09/HB0dX5lu185gtm7ZhFKpwMXFlR69+uDl5a09v2/vbo4cOUxUZCRPniSx4Y+t5M+fXyePiRPGcv1aFCqVivz5C+DnV5au3b/Vtkvh06LRaFgftIqD+3aTmPgYb9+S9OwzEAfHoq9Mt2fndrZt2YhKqcDZxY3vevXD08tHb/6Txo7k3Nl/GDF6IpWrfJElTnx8HIP6fEds7COC/gjO0iaFT4NGo2Hj78s5tH9nZlvzKcV3vYdg7+j0ynT7dm0leOt6VEoFxV3c6N5jIB5evgA8uB9Nn2/a6E03eMREAr6oCUDrxlk3vBk4bBxVa9R+x1oJguGIEYxPwJFjx1m6bDkdO7Rj8YK5uLo48+OYcShVKr3xL166zNSZP1G/bh1+WTCPKgGVGD95Ktdv3AQgJSWFq1FRfN2+LYsXzGXcqBHcvnOXsROnaPPw9fFmw9rVOkeDenUpYmeHp4f7h6i2YCBbNv/BzuDt9Onbn9lzF2Bubs7YMSNJTU3NNs2xo0f4bdlS2nfoyPyFi3FxdWXsmB9RqZTaOCkpKfj7l6dN23bZ5lO6dBmGjxzN0l9X8OOoMUTHRDNt6qRcrZ+Qd2zbvIFdwVvp2XcQM+cuwtzcnAljhr+yrZ04epgVy36hXYfOzFm4FGdXNyaMGa7T1p7ZuX0zvGajmJ/n/URxF9d3rYqQx+3Yso69O7fwfZ+hTJu9FDPzfEweO4TU1JRs05w8dojVv/1M6/ZdmTH/N4q7uDNl7BDinrY124KF+XXtdp2jzdfdMc+XDz//Sjp59R44UidehYBX77IpvH8SiTTPHh+Dj6OUwitt2baDBvXrUq9ObYoXK8aAvr0xMzdj/4E/9cbfHryTCv7laNOyBcWKOdG1U0fc3VwJ3rUbyNweeMaUSdSo9gVORYvi4+1N3149uBoZyYMHDwEwMTHBxkauPaysCvD36TPUrVMLiURs7fap0mg07Ni+jbbtOlA5oAouLq4MHvIDithYTp06mW267du2UK9+A+rUrUexYsXp03cAZmZmHDywXxunWWALWrdph5d31ifNzwQ2b4m3tw+F7ezw8S1B69ZtiQi/THp6eq7WUzA8jUbDzu1baNOuI5UCquLs4saAISNQxD7izKkT2abbsW0Tdes3pFbdBjgVc6ZX30GYmZlx6MBenXjXoiLZsXUT/Qb+kG1ee3fvIDHxMYEt9D+FFj4NGo2G3Tv+oGXbzlSoXI3iLu70HTwKpSKWf08dzzbdru0bqVWvCTXrNMKpmAvf9xmKqZk5fx3M/C41MjJCLrfVOf45dZyAL74iXz4LnbwsLfPrxDM1NXuvdRaE9010MD5yaWlpXI2MpKyfnzZMKpVS1q8Ml8PD9aa5FB5OWb8yOmHly5XLNj5AYmIiEokEy/yWes+fOvMPCQkJ1KsjhnQ/ZfdjYlAqFfj5ldOGWVpa4uXlTfjly3rTpKWlERl5FT+/stowqVSKn19ZwsP1p3kTCQnxHDn8Fz4+vhgbi9men5r7MdEolQpK+/lrwywt8+Pp5UPE5Ut606SlpREVeUUnjVQqpYyfPxHhz9OkJCczZ+YUvu89ALmNjd68bt+6wR/r1jJwyAgkUvFV+Sl7cD8alVJBKb/y2jBLy/y4e/kQEX5Rb5q0tDSu6Wlrpf3KcyWbNFGREdy4dpVadRtlOffbL3Pp3qExIwZ9z18HdqPRaN6xVoJgWHn6W/n+/fssXbqUsWPHZhsnJSWFlJSUl8JSMTMzfd/FyxPi4+NRq9XIZTKdcLlMxu3bd/WmUSpVWeLLZDIUyqxTCABSU1P5beVqvqxRHUsLC71x9h04iH+5shQqWDDHdRA+HkqlAgCZXKYTLpPJUWXTfp61UZlcniXNndu3c1yGlSt+Y9fOHaSkpODl7cO48WKK1KdIpW1ruu3GWibXtsOXJcTH6W1r1jI5d27f0n5evmwx3j4lqBRQVW8+aWmpzJ4xmS7f9KBQYTtiYqLfpSpCHqdSxgKZ96QXyWQ2qFSvamsZWMt0O6jWMjl379zUm+avA7twdCqOl08pnfC2X39DyTLlMDMzJ/T8v/z2yxySk5/QsGmrt62SIBhcnn4sExMTw4QJE14ZZ9q0aVhbW+sci5cu/UAl/PSlp6czedpMQEP/Pr30xnn46BFnz52nft06H7Zwwnt3+PAhWrVoqj3SMzIMXSRatGzNgoW/MGnyNIykUubMnime9n0Cjh7+k3YtGmqP99XW/jl9kguh5/mmR59s46xd+RtFnYrz5VfinvYpOn74AB1b1dUeH2KKZUpKCieO/kmtOo2znGvVvivevqVxcfMksNXXNG3ZnuCt6997mQThfTLoCEZYWNgrz0dERLw2j5EjRzJ48GCdsJjb+p8efIqsrKyQSqVZFnQrVSpsXnrK/IxcLssSX6VSYfPSU7/09HQmT5/Jg4cPmDl1crajF/sP/kmBAgUIqFTxbash5FGVKgXo7PSUlpYGgEqpwsbm+c5NKpUSF1c3vXk8a6Mvj3CoVMpsp6e8yrMHCY5Fi+JUrBhdO39NePhlfHx8c5yXkHdUrFRFZ6entLTMhdwqpVKnrcWplLi46t9IooCVtd62FvdCWwsLPU9M9D2+bt1EJ87MqePxKVGKKTPmEhZ2nls3rtOi8VGdOJ3bBdK6XUfad+z61vUUDK98pS9w93p+v0h/dl9TKZHbPB+FV6kUOLvo3xExs60ZEffSCEecSolMnnVXu9MnD5OSkkz1WvVeWz4PL1+2bFhNWloqJiafx2wM4dNj0A6Gn58fEolE79PHZ+GvWzBsZmaGmZnuYijlZzI9CjIXW3u4uxMSEkrVgMoAqNVqQkLCaNo46zxPAF9vb86HhtEisJk27Nz5EHy8n/+QfNa5uHvvHrOmTcHKykpvXhqNhgMHD1Hnq5piHvwnyMLCAosXOpYajQa53IaQ0PO4umV2KJKSEomICKdBo6xP5iCzjbq7exAaGkJAlcwpKWq1mtCQEBo3afpO5VOrM+8dzzo+wscrn4UF+fS0tbDQc7i6ZXYokpISuRJxmfqN9LcbExMT3Nw9CQs9p91yVq1WExZyjoZNAgFo2boDderp3hsH9P6G7t/1pkKlAACGjxpPasrznaoir4SzcN4sps6aTxF7h1yrs2AY+tqaTG7DfyFncXHN7FAkJSUSGXGZeg0C9eZhYmKCq7snF0LPUjGgOpDZ1i6EnqV+4xZZ4v91YDflK1bF2lqe5dzLblyLxDJ/AdG5MDSp2LDmXRj0F6GNjQ0zZ86kVq1aes9fvHiRJk2a6D0nPNeyeTNmzZmHh4c73p6ebN0RTHJyMvXqZP5dZ86ei62tDd907QJAYNMmDB3xI5u3bqNihQocOXaMK5GRDOiXOWUgPT2dSVOnczXqGpPGjUGdoUahyHwiWKBAfkxMTLTXDgkNI+b+ferXq/uBay0YgkQioVlgczZuWIejgyN2dkUIWrsKG1tbAl6Yz/7jyB8IqFKVJk0yO7GBzVsyd84sPDw88PT0ZseOrSSnJFO7zvOneUqFAqVSSfS9ewDcuHEdi3wWFCpciAIFrIgIv8yVq1co4VuS/PnzEx19j6C1q7G3d8DHJ/udp4SPk0QioUlgSzZtCMLBwZHCdvasW7sSG9uCVAp4/r6KMSOHULnKFzRq0hyAZs1bM3/OdNw9vPDw9Gbnji0kpyRTq059AOQ2NnpHzgoWKoxdEXsA7O113+kSHx8HQFGn4uI9GJ8giURCo2Zt2LJxNUUci1LYzp6NQb8ht7HV2S52wo8DqBhQnQZNWgLQOLAti+ZOxc3DG3dPH3bv2ERK8hNq1m6ok3/0vTtcvhjKyPGzslz7f2dOEqdS4OFVAhNTU8JC/mXbH2tp0iL77boF4WNg0A6Gv78/9+7do3jx4nrPq1QqMbf6DXxZvRpxcXGsCVqHUqnE1dWVKRPHI3865enBw4c6I0ElfH0YOWwIq9b+zsrVa3FwdGD86B9xcc78d3gUG8upM/8A0KvfAJ1rzZo2hTKlny9Q23fgIL4+3hRzevWLr4RPR8tWbUhOTmbhwnkkPn6Mb4mSTJw4FVPT50/bYqKjiY+L036uXuNL4uLjCFq7RttGJ06com2jAHv27GL9uiDt5xE/DAFg4KCh1K5TFzMzc06dPMG6oDUkJydjY2NDOf8KtG3XQTzp+0Q1b9WO5ORkFi+cQ+Ljx/iUKMXYidNfamv3dNraFzVqEhevYv3alSiVmVP3xk2cgUye8+l4wuejWcsOJCc/YenCWSQlPsbbtxSjJv6ks13s/Zh7JMQ/b2tVq9ciPk7FxqDlmS91dHVn1MSfsrS1wwd3Y1OwEGXKVshyXSNjY/bt3saq3xai0UARe0e6fNuXWvXEw1Xh4ybRGPAX/LZt20hMTKRjx456zyuVSoKDg+nSpUuO8r0Z+fq1G4KQG1Il5oYugvCZSM/bm/4Jn5A0tcnrIwlCLijtUdjQRciWclpvQxchW/KRiw1dhNcy6DdW8+bNX3leLpfnuHMhCIIgCIIgCILh5Oltam/fvk337t0NXQxBEARBEARBEN5Qnu5gKBQKVq9ebehiCIIgCIIgCJ8RiVSaZ4+PgUGnSAUHB7/y/LVr1z5QSQRBEARBEARByA0G7WAEBgZm+x6MZ173HgxBEARBEARBEPIOg46z2Nvbs3XrVtRqtd7j3LlzhiyeIAiCIAiC8BmSSCV59vgYGPw9GGfPnqVZs2Z6z79udCM799Ls37VogvBGbEzjXh9JEHJBgXmDDF0E4TMRtjzM0EUQPhOlk8INXQThPTFoB2PYsGEkJiZme97d3Z3Dhw9/wBIJgiAIgiAIgvAuDNrBqFat2ivPW1paUqNGjQ9UGkEQBEEQBEEQ3tXHsdeVIAiCIAiCIAgfBdHBEARBEARBEAQh1xh0ipQgCIIgCIIg5DkS8Qz+XYgOxidMo9Gwbf1Sjh7cTlLiYzy8S9O55wiKOBTLNk3ExXPs2baWm1HhqJSP6DdiFv6Vv8w2/qpfpnFk/1badx9EvaYd3kMthLxGo9GwLmg1B/btITHxMT6+JejVZwAOjkVfmW73zh1s2/IHSqUCFxc3vu/VF08vbwASEuJZF7SakHNnefjwAVbW1lQOqMrXnbpiaZlfJ59DB/ezfdtm7t29g4WFJVW/qE7PPv3fW32FvM2yWj0K1GqCkZWMtLs3UW5eQdrNqGzjS/JZYN24PfnKVERqkZ905UPitqwm+dL5D1hqIa8r3qMDLgO/wcyuIAkXwrk4ZDJx/7uQbXznPp0p9l178jnZkxqrJGbbfiLGzkGdkgqAx6i+eIzqq5PmccQ1jpVt+F7rIQiGIjoYn7A929ZwcNdGvhswnkJ2Dmxdt4TZE/oxZeEfmJqa6U2TkvyEYi6eVK/dlIXTf3hl/mdPHyYq4gIym0Lvo/hCHrV180Z2BW9jwOAfsCtiz+9rVzJuzAgWLVmBqamp3jTHjx5m+bIl9O47AE9vH4K3b2HcmBH88utKZDI5ithYFLGxdPu2B07FivPg/n1++XkeithYRowap81n+9bNbN+2iW7dv8fT24fk5GQe3I/5UFUX8ph85QKQNe+McuMyUm9eJf+XjSjUexQxkwaifhyfNYGREYX6jCbjcTyxy+eQEafAyKYgmidJH77wQp5l37IB3tNHcLH/eFT/huLctwsVd/zGUb8GpD5UZInv0KYxXpOGcKHnKJSnz2Pp4UzpX6eBBi6PmK6Nl3DxCmcad9d+1qSnf5D6CIIhiPGfT5RGo+HAzvU0bdOdcpVq4OTswXcDJqBUPOLcmaPZpivtX5WWX/fCv3LNV+avjH1A0LKf6Dl4EkZGop/6udBoNARv30qbdl9TOaAqLi6uDBoyHEVsLKdPncw23Y5tW6hbvyG169anWLHi9O47EDMzM/48sA+A4s4ujBw9noqVArC3d6CMX1k6dunOP2dOk5GRAcDjhASC1q5k0JDh1KhZC3t7B1xcXKlUucoHqbuQ9xSo2ZjEU4dIOnOE9Ji7qDYuQ5OaimWA/vuXZeWvkFrkJ/bXWaRejyBD8ZDUyMuk3b35gUsu5GUu/btye+Um7qzdyuPwKP7rN46MJ8kU7dxSb3xZ5bIoT53j3h+7eHLrLo8OneTeH7uxLl9KJ54mI4PU+4+0R1qs6gPURhAMQ3QwPlEP798lThmLb+mK2jALy/y4eZYgKuLdXqKkVqv5dd44GgR2xLGY27sWVfiI3I+JRqlUUMavnDbM0jI/nl4+RFy+pDdNWloakZFX8HshjVQqpYxfOcLD9acBSEpMxMLCAiMjIwBCzp9Fo1YTG/uI3j26061TO2ZMncjDhw9yqXbCR8XICBMnV5IjXpi2otGQHHEBU2dPvUnMS/mTcuMqsjbfYD/lV+xG/kSBus1B8nG8GVd4/yQmJliVLUHs4b+fB2o0PPrrFPJKfnrTqE6fx7psCW2HIp9zUQrXq87D/cd04lm4FeerqGN8efEgZVbMwryoeCmw8OnKE4+e79y5g0wmI39+3bnWaWlpnDp1iurVq2ebNiUlhZSUFJ2w1NSUbKcAfS7iVLEAWMtsdcKtrG2JU8a+U957tq5GKjWiTuN275SP8PFRKpUAyORynXCZTIZSmXXqAEB8fBxqtVpPGjl3b9/WnyYujo3rg6jXoJE2LCYmGo1Gw6aN6/muR28sLC35fc1Kxo4azoJFv2JiYvIuVRM+MlJLKyRGRqjjVTrh6gQVJnYOetMYF7TD2KYQSf87waMl0zAuVARZm2/ByIiEvZs/QKmFvM60oBypsTEp93W/J1MePCK/l4veNPf+2IWJrZyAP38HiQSpiQk3l60natZSbRzVv6GEfT+SxKvXMStSGI8f+xDwZxDHyjcl43H2LxwWhI+VQUcwoqOjqVixIsWLF0cmk9G5c2ceP36sPa9QKKhZ89VTdaZNm4a1tbXOsebXOe+76HnO30f30qNdde2R8Z7mdt6IvMyBXRv4dsA4JOKp3yfvyOFDtGnRWHtkZLz/OcNJSYlMHDcKp2LFaf91Z224WqMhPT2d73v2oZx/Bby9fRk6fBTR9+5yISzkvZdL+ARIJGQkxKNcv5S029d5cu4UCfu3kr9qHUOXTPiI2VSriPsP3/PfwImcrNKSs+36Urh+DdxH9NLGeXjgODHb9pPw3xUe/XmCf5t/j7G1FfYt6xuw5MKrSKSSPHt8DAw6gjFixAikUilnzpxBpVIxYsQIatasyYEDB5A/fdqp0WhemcfIkSMZPHiwTtj56ynZxP50la1YHTfPktrP6WmZO1fEqWKR2RTUhsfHxVLMRf/0gTcRcek8CXFKhnzbRBumVmewYdV8DuzcwOxlwW+dt5D3VKwUoN3pCSA9LQ0AlVKJjc3z0TGVSoWrq/7pclZW1kilUlRPRz+ep1Eis9Ed1UhKSmL8mJHks8jHj2MmYGz8/BZlI7cBwKlYcW2YtbWMAlZWYprUZ0idGI8mIwOplUwnXFpARsZLoxraNHEqNOp0eOF7Jf3+XYys5WBkBE/X+wifr9RHStTp6ZjZ6Y7+mxUuSMr9R3rTeI7tz911wdxZlTkKlnDxCkYW+Sj180QiZyzRaW/PpMclkBh5A0vX4lnOCcKnwKAdjD///JNt27ZRvnx5AE6ePEnr1q356quvOHToEMBrn5KbmZlhZqY7HcrUVM/uIZ+4fPksyZfPUvtZo9FgLbflUti/FHf1AuBJ0mOirlykZv1Wb32dql82pESZijphP03oT5UvG1CtVpNsUgkfKwsLCywsLLSfNRoNcrkNoaHncXVzBzJHHK5EXKZBI/3//iYmJri7exIaeo7KVaoCmet4wkLO06hJM228pKRExo0egYmJCaPHTsqyI5WPb2YH+u6d2xQsmLlzWUJCPAnx8RQubJd7lRY+DhkZpN2+hrlnSZLD/s0Mk0gw8yxJ4vF9epOkXI/Awr9q5pqLpz/6jAvZkxGnEJ0LAQBNWhrx5y9i+2UA93dm/g5BIsG2ZmVuLvldbxoji3xo1GrdfDLU2rT6OhhGlhZYuDhxN0Y8lBM+TQbtYMTFxWlHKiCzs7B161Zat25NzZo1CQoKMmDpPm4SiYS6Tdqzc9MKijg4UbCwI1vXLUFuU5BylWpo480Yk7ljVO1GbQBIfpLE/ejn8+IfPbjHzWsR5C9gjW2hIuS3kpH/pSeGRkbGWMtssXd0/hBVEwxIIpHQNLAFf2z4HQcHR+zsivD72lXY2NpSOaCqNt7okcOoXKUqjZsEAtCseUvmzZmJu4cXnp5eBO/YSnJKMrXqZE4PSEpKZOyo4aSkpDB42EiSkpJISsrcOtTK2hojIyMcixalUuUqLFu6mD79BmFhYcGaVctxLOpEqdJ+H/pPIeQBCYd3YdOxD6m3rpF6M5L8XzZEamZG4ukjAMg79SFDpSB+53oAEo8fIH+1eshaduXx0X0YFy5CgbrNeXx0rwFrIeQ11xesovSy6cSd+w/V/8Jw6dsFY4t83Fm7FYDSy6aTcu8BEeMyp2M/2HMY535diQ+9jOrfUCzdiuM5tj/39xyGpx0P76k/8GDPYZ7cuoeZfWE8R/dFk6EmetMug9VTeA2p2AfpXRi0g+Hq6kpYWBgeHh7aMGNjYzZt2kTr1q1p3LixAUv38WvYvDMpyU9YuXgqSYmP8fQpw5CxC3QWwD+IuUvCC9MJrkdeZsaYntrP61fMBaBqzUZ8N2D8hyq6kIe1aNWW5ORkFi2cS+Ljx/iWKMn4idN1Rhxiou8RHxen/VytRk3i4uNYt3YVSqUSV1c3xk+cpn3AEBV5lSsR4QD0+KazzvWWrQzCzq4IAIOGDue3X39h4vhRSCUSSpQqw/hJ03SmUgmfjyfnTqHKb4VVozYYFZCRdvcGjxZPRZ2Q2faM5QV1nh5nqGJ5tHgK1i26YDdyFhkqBY+P7iXh4HYD1UDIi6K37MW0kA2eY/phaleIhLDL/BP4HakPMhd+53NyAPXzdhU5/Rc0Gg2e4wZg7mBH6iMFD/YcJmL8PG0cc0c7/FbPxsRGRuojBcq/z3Lqy7akPlK+fHlB+CRINK9b5PAeDR8+nJCQEPbv35/lXHp6Oi1btmTnzp2oXxp6fJ1Tlz+/KVKCYdiYxr0+kiDkAst5QwxdBOEzEbb83bYyF4Q31TAp3NBFyFb8vMGvj2QgVgPz/mZGBn3sN2XKFO00iJcZGxuzZcsW7t69+4FLJQiCIAiCIAjC2zLoBDNjY2OsrKyyPR8dHc2ECRM+YIkEQRAEQRAEQXgXeXoFi0KhYPXq1YYuhiAIgiAIgiAIb8igU6SCg1+9Pdu1a9c+UEkEQRAEQRAEQcgNBu1gBAYGIpFIXvkyPfG2aEEQBEEQBOFDEr8/341Bp0jZ29uzdetW1Gq13uPcuXOGLJ4gCIIgCIIgCDlk0BEMf39/zp49S7NmzfSef93oRnbKKA6+a9EE4Y08tCtp6CIInwmT3sMNXQThM+E60M7QRRAE4SNn0A7GsGHDSExMzPa8u7s7hw8f/oAlEgRBEARBEAThXRi0g1GtWrVXnre0tKRGjRofqDSCIAiCIAiCILyrPL1NrSAIgiAIgiAIHxeDjmAIgiAIgiAIQp4jFc/g34X46wmCIAiCIAiCkGvECMYnauOhU6zed5zYuMd4OhVh+NdNKOnqpDfu1qP/suvvc0TevQ+AT3FH+rWsq42flp7B4m0HOREWwZ2HCvLnM6eSrzv9W9WjsNzqg9VJyBuCd+1m05ZtKJRKXF1c6NPze7y9PLONf+z4CVYF/c79+w9wdHDg225dqFihPADp6emsWhPEP/87S3RMDJaWlpTzK8M3XTtja2ubJa/UtDT6DxrKtevX+WXBPNzcXN9bPQXD27rnABu27UKhisPNuRgDvuuCr6d7tvEPnzzN8nWbiHnwCEf7IvTs3I6A8mW156sHdtCbrleX9rRv3gSAiKjrLF2znvCr15AaSalRuQJ9unfCIp957lZOyFM0Gg3rglZxcN8eEhMf4+1bkl59BuDgWPSV6Xbv3M72LX+gVCpwdnHj+1798PTy1p5fvHAOoefPoVDEYm6eD2/fEnTp9h1FnYoBcOjgPhbMnaU379XrNiOTyXOvkoLwAYkRjE/Q/n/CmL1xDz2a1mLduD54OtnTe85KFPGP9cb/X8Q16lcqw7IfvmX1qJ4UsbGm1+yVPFDGAZCcmsblm/f4rklN1o/ry+y+X3Mz5iEDF6z9kNUS8oAjx46zdNlyOnZox+IFc3F1cebHMeNQqlR641+8dJmpM3+ift06/LJgHlUCKjF+8lSu37gJQEpKClejovi6fVsWL5jLuFEjuH3nLmMnTtGb328rVmFra/O+qifkIYdOnGLRiiC6tmvBb3Om4O5cjKETpqNUxemNfyH8ChNn/0yj2l/y25ypVKvkz6jpc7h287Y2zraVi3WOEf2+RyKRUCOgIgCPFEoGj5uKo70dS2ZNZNbY4Vy/fZdpC5Z8kDoLhrN18wZ2B2+jV9+BzJr7M+bm5owfM4LU1NRs0xw/epgVy5bQtkNn5ixcgourG+PHDEelUmrjuLl70n/QD/y8dCXjJ09Ho9EwbvRwMjIyAPiiek1WBW3SOcr6V6BkqTKicyF81EQH4xMUtP8ELapXoFk1f9wc7RjVuRnmpqZsP35Wb/yp37elzVeV8SrmgIt9YcZ2a4FGo+HMpSgACliYs2Rod+pWLI2zfSFKuxVjRMemXL55l+hY1QesmWBoW7btoEH9utSrU5vixYoxoG9vzMzN2H/gT73xtwfvpIJ/Odq0bEGxYk507dQRdzdXgnftBjJ3ipsxZRI1qn2BU9Gi+Hh707dXD65GRvLgwUOdvP7531nOnjvP9990e+/1FAzvjx17aFy3Jg1rfYmzU1GG9PoGczMzdh86qjf+5p37qFiuDO2bN8HZyZFvv26Dp6sLW/cc0Maxlct0jhNnzlK2pC8ORTLf+/D3v+cwNjJi0PfdKObogI+HG0N6dufoqX+4Ex3zQeotfHgajYad27fSul1HKgVUxdnFjYFDhqOIfcTpUyeyTbdj22bq1m9I7br1KVbMmV59B2JmZsafB/Zp49Rr0JgSpUpjZ1cEN3dPOnbuxqOHD3jwIHPGgJmZGXIbG+0hNZJyIfQ8tes2eO/1FoT3SXQwPjFp6elcvnmPSr7PpxFIpVIq+boRFnXrjfJITkkjPSMDa0uLbOMkJCUjkUgoYCGmDXwu0tLSuBoZSVk/P22YVCqlrF8ZLoeH601zKTycsn5ldMLKlyuXbXyAxMREJBIJlvkttWFKpZJ5C35m+NBBmJmZvVtFhDwvLS2dK1HXKV/6+YsspVIp/mVKcjHiqt40FyOu4l9a98WXFcuWzja+QhXHqbMhNKr9pc51jY2Nkb6wuNPMzBSAC5ci3rY6Qh53PyYapVJBGb9y2jBLy/x4evkQcfmS3jRpaWlERV7RSSOVSinjV46IcP1pkpOf8OfB/dgVsadgwUJ64xw+dAAzMzOqfFH9HWokCIZn8A5GbGwshw8fRqFQAPDo0SNmzJjBxIkTuXz5soFL9/FRJiSRoVZjY5VfJ9zWKj+xcQlvlMf8zfsoJLOiUgk3vedT0tJYsHkf9SuVJr+Yl/zZiI+PR61WI5fJdMLlMhkKpUpvGqVSlSW+TCZDoVTqjZ+amspvK1fzZY3qWFpkdnA1Gg2z5s6nUcP6eHp4vGs1hI9AXEICGWo1cpm1TriNtXW2bU2hUmHzUnz5K+Lv++sYFvnMqR5QQRtWrnQJFKo41m/bSVpaOgmPH7N0zQYAYrPJR/j4KZ/ej2Ry3SlJMplce+5l8fFxqNVq/Wme/p55Zs+uHbRt0Yi2LRpz7n//MGHKTExMTPTme3D/Xqp/WUs8SMkDJFJJnj0+BgZd5P3PP/9Qt25d4uPjkclkHDx4kNatW2NsbIxarWb69OmcOHGCcuXKZZtHSkoKKSkpOmEZqWmYmer/j1d4tRW7j7L/nzCW/fAtZnpugGnpGfzwy3o0GvixUzMDlFD4VKWnpzN52kxAQ/8+vbTh23fu4smTJ7Rr3cpwhRM+OXsOHaFO9aqYmZpqw1yKFeXH/j1ZtDKIX9duRCqV0rJxPWxk1h/Nl7rwekcO/8kvC+dqP4+ZMPW9Xq9GzVr4lfVHqVCwbesfzJo2kek/LcD0hbYHEH75Indu32LQ0JHvtTyC8CEYtIMxatQoWrduzZw5c1i6dCmBgYHUr1+fZcuWAdC9e3cmTZrEtm3bss1j2rRpTJgwQSfsx26tGfVN2/da9rxKXsACI6k0y4Lu2PjH2FoXeGXaNfuOs3LPUZYM7Y6nk32W82npGQz/ZT3Rj1T8+sO3YvTiM2NlZYVUKs2yoFupUmEjl+lNI5fLssRXqVTYvPTULz09ncnTZ/Lg4QNmTp2sHb0ACAkN43J4BI0CW+qk6TNwMF/VrMEPgwe9dZ2EvMm6QAGMpNIsC7oVcXHZtjUbmQzFS/GV2cQPvRjOrbvRjB/aP8u5OjWqUqdGVRSqOMzNzJBI4I/gPTjYFX7r+gh5S8VKVfDy8tF+TktLA0ClVGJj83z3OpVKiYur/pF8KytrpFIpqpdGOFQqJXIb3Y0oLC3zY2mZHwfHonh6+/B1m0BO/32C6l9+pRPv4P49uLi64+6R/a58gvCxMOgUqbNnzzJ48GAKFCjAgAEDuHfvHt999532fN++ffn3339fmcfIkSOJi4vTOYZ2avG+i55nmRgb41PcgTOXI7VharWafy5HUdqtWLbpVu09xrKdf7FocFdKuGTdlu9Z5+LWg0csGdodWf7s12cInyYTExM83N0JCQnVhqnVakJCwvDx9tabxtfbm/OhYTph586H6MR/1rm4e+8e06dMwspKd+vjPj2+55eF87XH5AnjABg14ge6de6UW9UT8hATE2M83Vw4G3ZRG6ZWqzkXdpESXvqnyZXw8uBc2H86Yf+GXNAbf/efR/Byc8HdpXi2ZbCRWWORz5y/TpzG1MSU8mVKvWVthLzGwsICewdH7eFUrDhyuQ1hoee0cZKSErkScRkvH1+9eZiYmODm7klY6HltmFqtJizkPF7e+tNk0qBBQ1qa7u5UT5484cTxo9SpJxZ35xkSad49PgIGHcFITU0lX758QOZ/rBYWFhQsWFB7vmDBgsTGxr4yDzMzsyxzFZM+8+lRHet9wdjfNuPrXJSSLkVZd/AkT1JSafZF5lSz0cs2UVhuRf9W9QBYuecov2z/k6nft8WhoJxHT9dqWJiZYmFuRlp6BsMWryP85j3mD+iMWqPRxrG2zIeJsXidyueiZfNmzJozDw8Pd7w9Pdm6I5jk5GTq1akFwMzZc7G1teGbrl0ACGzahKEjfmTz1m1UrFCBI8eOcSUykgH9+gCZnYtJU6dzNeoak8aNQZ2hRqHIfCJYoEB+TExMKFxYdzFkvqcjZw5FilDohfuF8Glp06wh0+YvwcvdFR8PNzbt3MuT5GQa1qoBwJR5iyloa0OPTu0AaNWkPv1HTWLD9t0ElPfj0PFTRERdY1jvb3XyTUxK4sjfZ+jT7Wu9192yez8lvT2xMDfn39AL/LJqHT06t6PAC5sOCJ8WiURCk8AW/LHhd+wdimJnV4R1a1diY1uQygFfaOONGTmUylW+oFGTQACaNW/F/DkzcPfwxMPTm507tpCckkztOpnfrTHR9zhx7Ah+5cpjbW3No0eP2LJpPWampvhXqKRThhPHDqPOyKBGzdofrN6C8D4Z9Jehk5MT165dw9nZGYANGzZgb/98ak50dLROh0N4M/UqlkaZkMgv2/8kNi4BLyd7Fg3qpp0iFaNQIX1hPvGmw2e0nYgX9Wj6FT0Da/NQFc/RkMwF9+3GL9SJs+yHbynvLV529rn4sno14uLiWBO0DqVSiaurK1Mmjkf+dMrTg4cPkUiet60Svj6MHDaEVWt/Z+XqtTg4OjB+9I+4OGc+OX4UG8upM/8A0KvfAJ1rzZo2hTKlxVPjz1WtLwJQxcWzYv1mFEoV7i7F+WncCO1C7vsPY5G88CSvlLcnYwf34bffN7EsaCNFHYowZcRgXIvrvmD00PFTaDQaalWrove64VejWLlhC0+eJFOsqANDe31DvZrV3l9FhTyhRat2JCcns3jhHBIfP8anRCnGTZyms04iJvoe8XHPp+FVq1GT+Pg41q1dhVKZOZ1q3MTpyOSZU6RMTE25dPECwTu2kPj4MdYyOSVKlmb67IVZ3nFx8MBeKlepRv78uhu0CMLHSqLRaDSGuviECRPw8vKiXbt2es+PGjWK8PBwtmzZkqN8k07mLL4gvK2HdiVfH0kQcoF5mv4XZQpCblOa2hm6CMJnwtvt1W9KN6THi0cYugjZyt97uqGL8FoGHcEYN27cK8+PGjUKIyOjD1QaQRAEQRAEQRDeVZ5eKRIbG0uvXr1eH1EQBEEQBEEQhDwhT3cwFAoFq1evNnQxBEEQBEEQhM+JVJJ3j4+AQadIBQcHv/L8tWvXPlBJBEEQBEEQBEHIDQbtYAQGBiKRSHjVOvMXd6QRBEEQBEEQBCFvM+gUKXt7e7Zu3YpardZ7nDt37vWZCIIgCIIgCIKQZxh0BMPf35+zZ8/SrFkzvedfN7qRnQd2Yu984cOQYLBdnoXPTKKpzNBFED4TaZrP+2W1giC8O4N2MIYNG0ZiYmK2593d3Tl8+PAHLJEgCIIgCIIgCO/inTsYGRkZXLhwgeLFi2vf5vumqlV79dtRLS0tqVGjxrsUTxAEQRAEQRByRCLJ0xut5nk5/usNHDiQ5cuXA5mdixo1alCuXDmcnJw4cuRIbpdPEARBEARBEISPSI47GJs3b6ZMmTIA7Ny5k+vXrxMeHs6gQYMYNWpUrhdQEARBEARBEISPR447GI8ePaJIkSIA7Nmzh9atW+Pp6Un37t25cOFCrhdQEARBEARBED4oQ79M7yN/0V6OOxh2dnZcunSJjIwM9u3bR506dQBISkrCyMgo1wsoCIIgCIIgCMLHI8eLvLt160abNm2wt7dHIpFQu3ZtAM6cOYO3t3euF1B4M8G7drN5y1YUSiWuLi707tkDby/PbOMfO36C1UFB3L//AEcHB77p1pWKFcprz584+Te79+7lamQUCQkJLF4wHzc3V508FAolv61YwbnzISQ9eYJTUUfatW1DtapV31s9BcML3rWbTVu2adtan57fv7atrQr6XdvWvu3WRdvW0tPTWbUmiH/+d5bomBgsLS0p51eGb7p2xtbWVpvH1cgoflu5iitXI5FKpXxRJYCe331Dvnz53nt9hQ9n586dbN6yBeXTttWrVy+8vLyyjX/8+HHWrF3L/fv3cXRwoFv37lSsUEF7XqPRsDYoiH379pGYmIivry99+/TB0dExS16paWkMGjSIa9eu8fPChbi5uWWGp6ay8Oefibx6lVu3b1OpYkXGjh2b+5UXDEqj0bAxaAV/7t9JUuJjvHxK8X2fwdg7Or0y3d5dWwnesgGVUkFxFze+6TkADy9fnTgRl/9j/ZplXI24jFQqxdnVndGTZmNmZgbAvbu3WbN8MRGX/yM9LY3iLm606/gNJcuUe2/1FYT3LccjGOPHj+e3337j+++/5+TJk9r/QIyMjBgxYkSuF1B4vSPHjvPrst/4ukN7Fi2Yh6uLC6PGjEWlUumNf/HSZabNnEX9unVZvGA+VQIqM2HyFG7cuKmNk5ySTAlfX77p1iXb686aM4fbd+8yfuwYli76mapVqjB1+kwio6Jyu4pCHnHk2HGWLltOxw7tWLxgLq4uzvw4ZhzKV7S1qTN/on7dOvyyYB5VAioxfvJUrj9taykpKVyNiuLr9m1ZvGAu40aN4Padu4ydOEWbR2xsLCNGjcHRwZ4Fc2YxdeJ4bt66xay58z9AjYUP5ejRo/y6bBlfd+jAwoULcXF1ZfSYMdnexy5dusT0GTOoV7cuPy9cSEBAAJMmTeLGjRvaOJs2byY4OJh+ffsyb+5czM3NGT1mDKmpqVnyW7F8OTY2NlnC1Wo1pqamNG3WjLJly+ZWdYU8ZvvmdezZuYXv+wxh6pylmJmbM2nMUFJTU7JNc/LYIVYvW0TrDl2ZueA3nF3cmTxmKHEqpTZOxOX/mDJ2GGXKVmD63KVMn/crDZq0QPrCNJdp44ejzshg3NR5zJy/jOIubkybMAKlIva91lkQ3qccdzDWrFlDkyZNGDRoEEWLFtWGt2/fnri4uFwtnPBmtm7bTv369ahXpzbFixWjf9/emJmbsf/AQb3xtwcHU96/HK1btqBYMSe6dOqIu5sbO3bt0sap/dVXdOzQnrJ+ftle99LlcJo1aYy3lyf29kXo0K4tlpaWXI2MzO0qCnnElm07aFC/rratDdC2tT/1xt8evJMK/uVo87Stde3UEXc3V4J37QYyt6KeMWUSNap9gVPRovh4e9O3Vw+uRkby4MFDAE7/8y9Gxkb07dUTp6JF8fL0YEDf3pw4+Td37937YHUX3q9t27bRoH596tatS/FixejXty9mZmYcOHBAb/wdO3ZQ3t+fVq1aUaxYMTp37oybmxs7d+4EMp9Ib9++nXbt2hEQEICLiwtDhwwhNjaWv0+d0snr33//5dz583z77bdZrmNubk6/vn1pUL9+jrdiFz4OGo2G3Ts20bJtJyoGVMPZxY1+Q0ahVMTyz6kT2abbue0PatdvzFd1GuJUzJnv+w7BzNycvw7s1sZZtexnGjRtSfM2HXEq7oJj0WJUqfYVJiamAMTHqYi+d4fA1l/j7OKGvaMTHbv2JCUlmds3r7/3ugvC+5LjDka3bt30diQSEhLo1q1brhRKeHNpaWlcjYyknF8ZbZhUKqWsnx+XwiP0prkcHp6l4+BfriyXw8NzdG1fH2+OHjtOfEICarWaI0ePkZqaSulS4k3qn6Jnbe3FtpPZ1spk23YuhYdT9oW2CVC+XLlXtrXExEQkEgmW+S2fXjcdY2MTpNLntytT08wv54sXL79tdYQ85Fnb8nupbfn5+WXbVi6Hh+P30oiCv7+/Nn5MTAxKpVKnvVpaWuLl5UX45eftRqlUMn/BAoYOGYL50xF54fPyICYalVJBab/n04QtLfPj4eXDlfD/9KZJS0vjWuQVnTRSqZRSfv5EhF8EIE6l5GrEJayt5fw4pBfffN2MscP7cflimDZNAStrHIoW4+hf+0lOfkJGRjoH9u7AWibH1T376YGCkNfluIOh0WiQSLKuYL9z5w7W1ta5UihXV1euXr36RnFTUlKIj4/XOVJSsg5/f6ri4+NRq9XIZLpP1uQyGUqlUm8apVKFXCbTE1+Vo2uPGjGcjIwMWrfrQOPAFsz/eRHjRv+Io4NDjvIRPg7P2pq+tqPIpu3oa2symQxFNm0zNTWV31au5ssa1bG0sADAr0xplEolf2zZSlpaGgkJj1m+ag0AsUrFO9VJyBu0bUuu5z6m0P9vrFQqs7mPKbXnAf15Pj2n0WiYM2cOjRo2xNMz+3VEwqdNqcyciiR7qa1Yy2xQZXOPSYiPQ63OwPql717ZC2nux2SOsP6xbiW16zdh1MRZuLh5MuHHQUTfvQ2ARCJh3JQ5XI+6SqdW9WkfWIdd2/9g1MRZ5C9QIFfrKeSMRCrNs8fH4I0XeZctWxaJRIJEIqFWrVoYGz9PmpGRwfXr16lfv36OLr5gwQK94bdu3WLlypXa7XD79++fbR7Tpk1jwoQJOmED+vVlYP9+OSqLkHOr1/7O48eJTJ8yGSsrK06dPs2U6TOZPXM6Ls7Ohi6e8JFJT09n8rSZgIb+fXppw52LF2PY4IEsXbacFavWYCSV0qxpE+QyGVLxplXhHQQHB5P05Alt2rQxdFGED+jY4QP8+vNs7eeR42e8l+uo1WoA6jRoyld1GgLg6ubJhdCz/HVwD1937YFGo2HZ4rlYy2RMmvkzpqamHNq/m+kTRjJj3lLkNgXfS9kE4X174w5GYGAgACEhIdSrV4/8+fNrz5mamuLs7EzLli1zdPGBAwfi6Oio01mBzP8o16xZg4mJCRKJ5JUdjJEjRzJ48GCdsOjbt3JUjo+ZlZUVUqkUlUr3ibBSpcp2vrBcLsuyKDczvuyNr3svOprgXbtYuvhnnIsXB8DN1YUL/10keNduBvTtk6N6CHnfs7amr+3YZNN29LU1lUqFzUttMz09ncnTZ/Lg4QNmTp2sHb145qsva/DVlzVQKpWYm5uDRMLW7TuwL2L3rtUS8gBt21LquY/pWXgNmSMT+u9jcu15yBzJeHHxtlKlws01c0e80NBQwsPDadqsmU4+/QcMoGbNmgwdMuSd6iXkTRUqfaGz01N6WhoAKqVS5wd9nEqBs6u73jwKWFkjlRrpLOgGUKkUyOSZ7U1uk7kTnpOTs06cok7FefjwPgAXQs9x7t9TrNq4GwuLzGmhru5ehIb8y5E/99G8Tcd3qKkgGM4bdzDGjRsHgLOzM23bts38kn9H33//PWfOnGHdunX4+Phow01MTDhw4AC+vr6vSJ3JzMxMu5PVMwoz03cu28fCxMQED3d3zoeEUSUgAMjsoIWEhNK0cSO9aXy8vQkJDaVF4PMv1XPnQ/DJwTbDKSmZO2u8/ATZyEiKRq3JaTWEj8CzthYSEkrVgMrAs7YWlm1b8/X25nxo2Cvb2rPOxd1795g1bQpWVlbZluHZj8Z9Bw5iYmJCubJ+uVAzwdC0bSs0lCpVqgDP2lYITZs00ZvGx9ubkJAQmj99+AVw/vx5bdsqUqQIcrmckNBQ7ZaziUlJRERE0KhRZnvt2bMnnTt31qaPVSgYPXo0I0eMwEtsu/7JymdhQb4XHmJoNBpkchsuhJ7Fxc0DgKSkRK5GXKZuw0C9eZiYmODq7smFkLNUDKgGZLbZCyHnaNC4OQCF7eyxsS3I3bu6Dz3v3b1D2fKVAEhNSQbIMvVcKpGi1ojvUuHjleP5BV26dMmVzgXAkiVLGDt2LPXq1ePnn3/OlTw/Ry2aB7J3/34O/nmIW7dus3DRYpKTk6lbJ/MdJTNnz2HFqtXa+IFNm/K/s+fYvHUbt27fZu3v67gaGUmzxo21ceITEoiKusatW5nzRG/fvUtU1DUUisynNU5Fi+LgYM/8nxcRHnGFe9HRbN66jXPnQ6jy9Men8Olp2bwZe/Yf4MDTtrZg0S8kJydTr04tAGbOnstynbbW5IW2doc1v6/jSmSktkOSnp7OpKnTuXI1khFDh6DOUKNQKFEolKQ9faoIsGPnLq5GRnHn7l2Cd+1m0ZKldO/aWWckVfi4NW/enH379nHwzz+5desWPy9aREpKivZlrj/99BMrV67Uxm/WrBlnz55ly9at3L59m6CgIK5evUqTpx0SiURCYGAgGzZs4PTp01y/fp3ZP/2Era2t9mFM4cKFcXZ21h5Fn74fw97enkIFnz/JvnnrFlFRme8ESkxKIioqiiixHfcnQyKR0KhZa7ZsWMO/p09w80YUC2dPQW5jS8WAL7Txxv84kL07t2g/N2nehj/37+LIn3u5c+sGyxbNJiX5CTWfToeSSCQ0bdGOvcFbOHXiCNH37rB+7W/cu3OTWnUz74Ge3iWwzF+An+dM5ca1SO07MR7cj8a/QsCH/UMIQi7K8Yv2pFKp3kXez2RkZOQov+bNm1OxYkU6d+7M7t27db5AhDfzZfVqxMXFsSbo98wXVLm6MmXiBO3T3ocPHyJ94d+shK8PI4YNZfXaIFatXoODowPjRo/C2bm4Ns7p02eYPe/5ewamzZgJQMcO7en0dQeMjY2ZPH48y1etYtzESTx58gQHB3uGDh6o88I+4dPyvK2te6Gtjde2tQcPH+rcH0r4+jBy2BBWrf2dlavX4uDowPjRP+LytK09io3l1Jl/AOjVb4DOtWZNm0KZ0pk7kkVcucqa39eT/OQJTk5FGdC3D7W/qvkBaix8KDVq1CAuPp6gtWtRKJW4uboyaeJE3bb1wuJGX19fhv/wA6vXrGHVqlU4OjoyZswYnF9Y/9W6VSuSk5NZsHAhjx8/pkSJEkyaOFG7C9mbGjt2LA8ePNB+7tsvc43f3j173qHGQl4S2KoDKcnJLF34E4mJj/H2LcXoST9havp8hsT96HvExz/fRbNq9VrEx6nYELQClTJzOtWoiT9pp0gBNA5sQ1pqKquWLeRxQgLFXdwYM3kORewzO7NW1jJGTZzF+jXLGP/jQDLS03Eq7sIPY6ZmOz1LED4GEo0mZ2Nw27dv1/kBkZaWxvnz51m9ejUTJkzgm2++eauCaDQapk+fzoIFC3j48CFhYWFvNEVKnxuRV94qnSDklAQxhC18GBmSHD8PEoS3kqgRI4PCh1HKPe+uo0taMc7QRciWRfcJr49kYDn+xgp8Yb7rM61ataJEiRJs3LjxrTsYEomEkSNHUrduXU6cOIG9vf1b5SMIgiAIgiAIguHk2h6PlStX5tChQ++cj7+/PwMGDEAul3P79m26d++eC6UTBEEQBEEQBOFDyJUOxpMnT1iwYAGOTxfI5RaFQsHq1atfH1EQBEEQBEEQhDwhx1Ok5HK5zhoMjUZDQkICFhYWBAUF5Siv4ODgV56/du1aTosnCIIgCIIgCIIB5biDMW/ePJ3PUqmUQoUKUalSpWxf7JadwMBAJBIJr1pn/qodqwRBEARBEARByFty3MHo0qVLrl3c3t6exYsX0+ylt6g+ExISgr+/f65dTxAEQRAEQRBeS5pry5Q/S2+176FSqWT58uVcvnwZyNyPvFu3btjY2LwmpS5/f3/Onj2bbQfjdaMbgmBo+VITDF0E4TMRZ1bI0EUQPhNS1IYugiAIH7kcd8+OHTuGs7MzCxYsQKlUolQqWbBgAS4uLhw7dixHeQ0bNowqVapke97d3Z3Dhw/ntIiCIAiCIAiCIBhIjjsYffr0oW3btly/fp2tW7eydetWrl27Rrt27ejTp0+O8qpWrRr169fP9rylpSU1atTIaREFQRAEQRAE4e1JJHn3yKFFixbh7OyMubk5lSpV4p9//nmjdBs2bEAikeh9B97r5LiDERkZyZAhQzAyMtKGGRkZMXjwYCIjI3NcAEEQBEEQBEEQct/GjRsZPHgw48aN49y5c5QpU4Z69erx4MGDV6a7ceMGQ4cOpVq1am913Rx3MMqVK6dde/Giy5cvU6ZMmbcqhCAIgiAIgiAIuWvOnDl89913dOvWDV9fX5YsWYKFhQUrVqzINk1GRgZff/01EyZMwNXV9a2um+NF3v3792fAgAFERkZSuXJlAE6fPs2iRYuYPn06YWFh2rilS5d+q0IJgiAIgiAIgpBVSkoKKSkpOmFmZmaYmZnphKWmpnL27FlGjhypDZNKpdSuXZtTp05lm//EiRMpXLgw33zzDcePH3+rMua4g9G+fXsAfvjhB73nnu38JJFIyMjIeKtCCTkXvGs3m7dsRaFU4uriQu+ePfD28sw2/rHjJ1gdFMT9+w9wdHDgm25dqVihvPb8iZN/s3vvXq5GRpGQkMDiBfNxc9PtxSoUSn5bsYJz50NIevIEp6KOtGvbhmpVq763egqGt3XPAdZv341CFYebczEGftsFX0+3bOMfPnmG39ZvIubBI4ra29Gzc3sC/P2055OeJLN07QaO//M/4hIeY1+4EK0a1SOwfu0seWk0GoZNmsmZ82FMGTGI6pXKZ4kjfDo0Gg2/B61h/769JCY+xse3BL379MfR0fGV6XbtDGbrlk0olQpcXFzp0asPXl7e2vP79u7myJHDREVG8uRJEhv+2Er+/Pl18rh75w4rVizj8qWLpKWl4+LiQsdOXShdxu99VFUwMI1Gw4aglRzcv4ukxMd4+5Tk+z6DcXAs+sp0e3dtY/uWDaiUCpxd3Pm2Z388vHy058eMGMDFC6E6aeo2aELPvkO0n1s0+jJLvoN/GMMXNWq9W6WET9a0adOYMGGCTti4ceMYP368TtijR4/IyMjAzs5OJ9zOzo7w8HC9eZ84cYLly5cTEhLyTmXMcQfj+vXr73RBIfcdOXacX5f9Rr++ffD28mTb9mBGjRnL8l+XIJPJssS/eOky02bOonvXLlSqUIHDR48yYfIUFs2fh7NzcQCSU5Ip4etL9WpfMG/Bz3qvO2vOHB4nJjJ+7Bisraw4fPQoU6fPZOG8Obi7Zf+DU/h4HTpxip9X/s6Qnt3x9XRj0859DJk4nXU//4RcZp0l/oXwK0yY8zPfd2xLlfJl+fP43/w4fQ7Lf5qCa3EnAH5eGcS5C5cYM7A3RQoX4t+QC8xZupKCNnK+qKj7Hpw/du4TL9/8jGzZ/Ac7g7czaPAw7IoUIWjtasaOGckvS37D1NRUb5pjR4/w27Kl9OnbHy9vb3Zs38rYMT+y9NflyGSZL4NNSUnB3788/v7lWb1K/zSBCePH4ODoyJRpMzE1NSN4+1YmjB/Db8tXI8/hluxC3rdt83p279xC/0EjKVzEnvVrVzBpzDDmL1mFqamZ3jQnjv3FymWL6dF3MJ5ePuzavpmJY4ax8Ne12rYGUKdeY9p17Kb9bGZuniWvvgOHU9a/ovaz5UsdXkF40ciRIxk8eLBO2MujF28jISGBTp06sWzZMgoWLPhOeeV4DcbNmzdxdHSkePHiOoejoyM3b97UCRM+jK3btlO/fj3q1alN8WLF6N+3N2bmZuw/cFBv/O3BwZT3L0frli0oVsyJLp064u7mxo5du7Rxan/1FR07tKesn1+21710OZxmTRrj7eWJvX0ROrRri6WlJVfFYv9P1sbgvTSpU5NGtWrg4lSUoT27Y25mxu5DR/XG37xrHxXLlqZD88Y4OznybYfWeLo6s3XPAW2c/8KvUr9mNcqW9MW+cCGa1v0KN+diXL4apZPX1es32Bi8mxF9v3+vdRTyBo1Gw47t22jbrgOVA6rg4uLK4CE/oIiN5dSpk9mm275tC/XqN6BO3XoUK1acPn0HYGZmxsED+7VxmgW2oHWbdnh5++jNIy4ujnv37tKqdVtcXFxxdHSkS7dvSElJ4ebNG7ldVcHANBoNu3ZsplXbTlQM+AJnFzf6DxmJQvGIf06dyDbdzm2bqFO/EbXqNMCpmDM9+g7GzNycvw7s0Ylnam6G3MZWe1hYWGbJyzJ/fp042XVqhA9HIpXm2cPMzAwrKyudQ18Ho2DBghgZGXH//n2d8Pv371OkSJEs8aOiorhx4wZNmjTB2NgYY2Nj1qxZQ3BwMMbGxkRFRWVJk50cdzBq1qyJQqHIEh4XF0fNmjVzmp3wjtLS0rgaGUk5v+cL7KVSKWX9/LgUHqE3zeXw8CwdB/9yZbmczXBZdnx9vDl67DjxCQmo1WqOHD1GamoqpUuVynE9hLwvLS2dK1HX8S9TUhsmlUopX7okFyOu6k3zX0Qk5V+ID1DRrzT/XXneCS3p7cHJf8/xMFaBRqPh3IWL3L4XQwW/5+0oOSWFCXMWMei7rtjKZblbMSFPuh8Tg1KpwM+vnDbM0tISLy9vwvVsNAKZ98PIyKv4+ZXVhkmlUvz8yhIerj+NPlZWVhQtWpS/Dv1JcvITMjIy2Ld3NzKZDHd3j7evlJAn3Y+JRqVUUMbv+YippWV+PLx8iQi/pDdNWloaUZERlH4hjVQqpbSff5Y0xw//SZf2TRnQuytBq34lJTk5S37LfplPl/ZN+WFQTw4d2CNeMizkClNTU/z9/Tl06JA2TK1Wc+jQIQICArLE9/b25sKFC4SEhGiPpk2bUrNmTUJCQnBycnrja+d4itSz9RUvi42NxdIya688p3kfOXKEyMhI7O3tqVevHiYmJq9Mo2+hS0pKKmZm+ofPPzXx8fGo1Wqd4VgAuUzG7dt39KZRKlXIX5o6JZfJUCpVObr2qBHDmTpjJq3bdcDIyAgzMzPGjf4RRweHHOUjfBziEhLIUKuxsdadCiWXWXHz7j29aRQqFTYvTZ2ykVmjeKGtDfyuC7MWL6fFt/0wMjJCKpHwQ+9v8Svx/OnywhVBlPT2pJpYc/HZUCozH2TJXupQymRyVEql3jTa+6FcniXNndu33/jaEomEyVNnMHnieFq3DEQikSCTyZgwaSr5CxTIUT2EvE/1tK1Zy3Wnvslkcm07fFlCfNzT796sae7evqX9XK1GbQoVtsPGtiA3rkexduVS7t65zfDRk7Rx2nXsTqkyZTEzMyfk3L/8unguyclPaNS0ZW5VUfiMDR48mC5dulC+fHkqVqzIvHnzSExMpFu3zGl7nTt3xtHRkWnTpmFubk7JkroPBZ9NtX85/HXeuIPRokULIPPG27VrV52hmIyMDMLCwl75Vm59GjZsyPr167G2tkahUNCwYUP++ecfChYsSGxsLJ6enhw7doxChQplm4e+hS4D+vVlYP9+OSqLkHOr1/7O48eJTJ8yGSsrK06dPs2U6TOZPXM6Ls7Ohi6e8JHYsvsAF69EMv3HIdgVKkjopXDm/LqKgjZyypcpyYl/znLuwkWWz55q6KIK79Hhw4dYtHC+9vO4CZMNVhaNRsMvi3/GWiZjxsw5mJqZcmD/XiaOH8vc+QuxsbE1WNmEd3f08EGW/jxb+3nU+Onv7Vp1GzTR/v/izq7Y2Ngy7sfBxETfpYh95mYFbdp31sZxdfMgJTmZ7Vs2iA6GkCvatm3Lw4cPGTt2LDExMfj5+bFv3z7twu9bt24hleZ4QtNrvXEHw/rpU0uNRkOBAgXIly+f9pypqSmVK1fmu+++y9HF9+3bpx19GD16NAkJCURFReHi4sKdO3cIDAxk7Nix/PLLL9nmoW+hS/QLTw8+dVZWVkilUlQq3Sd6SpUK+UtP8Z6Ry2UoVSo98WVvfN170dEE79rF0sU/4/x0vY2bqwsX/rtI8K7dDOibs7e6C3mfdYECGEmlKOLidMKVqnhs9SzwBrCRyVCodOMrVHHYPG1rKSmp/Pr7RqYMH0SV8pnTWtydi3H1+k3W79hN+TIlOXfhEndjHtCwo+79ZczMeZT28Wbh5NG5VEPBkCpVCtDZ6SktLQ0AlVKl84NepVLi4qp/Ewnt/fClEQ6VSpmjhdmhoSH8+88ZNvyxRTtf3t3dg/Pnz3Hoz4O0btPujfMS8p6Klari+cJOT8/aWpxSoaetuevNo4CV9dPvXt0RDpVKiUyefVt7tsNU9L3nHQx9cTZtWENaWiomJp/HbAzh/erbty99+/bVe+7IkSOvTLtq1aq3uuYbdzBWrlwJgLOzM0OHDn3n6VAv++uvv5g5cyYuLi4AFC1alBkzZry206Jv31/FZzI9CsDExAQPd3fOh4RR5el8OrVaTUhIKE0bN9Kbxsfbm5DQUFoENtOGnTsfgo+3t974+jzrGEolur1eIyMpGrWYO/opMjExxtPNhbNhF7Xbw6rVas5e+I8WDerqTVPSy52zYRdp06SBNux/of9R0jPzSzs9I5309AykL027NJJK0ajVAHzdogmNa3+pc77LwBH069aRKhXKIXwaLCwssLCw0H7WaDTI5TaEhJ7H9emudElJiUREhNOgUWO9eZiYmODu7kFoaAgBVTK3y1ar1YSGhNC4SdM3LktKSuYceclL9zepRCrmxn8C8llYkO+ltiaT2xAWeg4Xt8w1NklJiVyNuET9hvrbjYmJCW7uXoSFnKNSQOabjtVqNWEhZ2nYuHm2175+LXP9mfwVo2A3rkWSP38B0bkQPmo5XoMxbty4XC3As/UcSqUSt5e2NnV3d+fePf1zu4XnWjQP5Kc5c/H0cMfL05NtO3aQnJxM3TqZ7xGYOXsOBW1t6d61CwCBTZsybMRINm/dRsUK5Tl67DhXIyMZ2O957zY+IYGHDx4S+3RB/+27dwGQy+XY2MhxKloUBwd75v+8iO++6Y6VVQH+PnWac+dDmDhu7Af+CwgfStumDZi6YCnebi74eLixadc+niSn0LBWDQAmz/+FgjZyenbKfMLbqnF9+o2ezIYduwnwL8uhE6cIj7rGsF7fAGBpYYFfCR8Wr16PmZkpdoUKEnLxMvuOHKdvt44A2Mplehd2Fy5UEAe7wh+m4sIHJ5FIaBbYnI0b1uHo4IidXRGC1q7CxtaWgIDn79r5ceQPBFSpSpMmmQ9MApu3ZO6cWXh4eODp6c2OHVtJTkmmdp162jRKhQKlUkn00++XGzeuY5HPgkKFC1GggBXe3r7kz5+fubNn0a7D15iZmrF//x7u34+hfIWKCJ8WiURC42at2LxhLfYORbErYs/6tcuxsSlIxYAvtPHG/TiYSgFf0LBJ5pTxJs1bs3DONNw9vPDw9GHnjs2kJCfzVZ3MByox0Xc5duQQ/uUrUcDKihvXr7Fy2SJ8S5bB2SXz986/Z/5GpVLg6eWLqakpoefPsuWP32nWou2H/0MIuiS5P23oc5LjDoaLi8sr96G/du1ajvJ7tp4jLS2N69evU6JECe25mJgYve9xEHR9Wb0acXFxrAn6HaVSiaurK1MmTtBOkXr48KHOE+ISvj6MGDaU1WuDWLV6DQ6ODowbPUr7DgyA06fPMHve8/nQ02bMBKBjh/Z0+roDxsbGTB4/nuWrVjFu4iSePHmCg4M9QwcP1Hlhn/BpqfVFAKr4BJZv2IxCGYe7S3F+Gjtcu5D7/sNYnftDKW9Pxg3qw7J1m/g16A+K2hdh6ojB2ndgAIwf0pelQRuZOHcx8Y8fU6RQQb7r0IbAeuIlU5+7lq3akJyczMKF80h8/BjfEiWZOHGqzjswYqKjiX9h2l71Gl8SFx9H0No12vvhxIlTdKaM7tmzi/XrgrSfR/yQ+dKzgYOGUrtOXaytrZkwcSpr1qxk1MgfSE/PoFjx4oweMx7XbKZnCR+35q3ak5KczJKFPz19qWMpxkyaqbNdbEz0XeLjn7e1L6p/RXycivVBK1EpFbi4ujNm4kztFCljYxPCQs6ya8dmUpKfULBQYQKqVqdVu07aPIyMjNi3azsrly0CjYYi9o50/a43derpH6UThI+FRJPD8d758+frfE5LS+P8+fPs27ePYcOGMWLEiDfO69kK9mcaNGhAmzZttJ9/+OEHwsLC2LdvX06KyI3IKzmKLwhvyyI13tBFED4TcWbZb3YhCLkpVSOm5ggfRgl3e0MXIVtPgvLuxiL5Ov5o6CK8Vo5HMAYMGKA3fNGiRfzvf//LUV7P1nVkZ9y4cRgZGeUoT0EQBEEQBEF4J9LsZ+sIr5drE8waNGjAli1bcis7ABQKBb17987VPAVBEARBEARBeH9yrYOxefNmbHKwDeCbUCgUrF69OlfzFARBEARBEATh/cnxFKmyZcvqLOLUaDTExMTw8OFDFi9enKO8goODX3k+pwvGBUEQBEEQBEEwrBx3MAIDA3U+S6VSChUqxJdffol3Dt6j8CwviUTyyn3FX7VjlSAIgiAIgiAIeYtB34Nhb2/P4sWLadasmd7zISEh+Pv759r1BEEQBEEQBEF4v3Lcwbh79y5btmzhypXMrWC9vLxo0aIFjo76X3n/Kv7+/pw9ezbbDsbrRjeyk0iBHKcRhLdhZJJu6CIInwlj0gxdBOEzIZVmGLoIgmBwEvGivXeSow7G4sWLGTx4MKmpqVhZWQEQHx/PsGHDmDNnTo53fBo2bBiJiYnZnnd3d+fw4cM5ylMQBEEQBEEQBMN54+7Z7t276d+/P3379uXu3buoVCpUKhV3796ld+/eDBgwgD179uTo4tWqVaN+/frZnre0tKRGjRo5ylMQBEEQBEEQBMN54xGMWbNmMWLECCZPnqwTbm9vz5w5c7CwsGDmzJk0bNgw1wspCIIgCIIgCMLH4Y1HMM6dO0enTp2yPd+pUyfOnTuXK4USBEEQBEEQBOHj9MYdjIyMDExMTLI9b2JiQkaGWBgmCIIgCIIgCJ+zN54iVaJECXbs2MGgQYP0nt++fTslSpTItYIJOaPRaNgQtJKD+3eRlPgYb5+SfN9nMA6ORV+Zbu+ubWzfsgGVUoGzizvf9uyPh5eP9vyYEQO4eCFUJ03dBk3o2XeI9nNYyFnWr13BzZvXMDcz58ta9fm6yzcYGeV4kzLhI7Bj1x7+2LodhVKFm4szfXt8i7eXZ7bxj544yaqg9cTcf4Cjgz3fde1MpQrPt59e/fsGjhw/wcOHjzA2NsbD3Y3unb/G54U8v+7+PfcfPNTJ95suHWnfumXuV1AwmOCdu9i8ZQtKpRJXFxd69+qJl5dXtvGPHT/OmrVB3L9/H0cHB7p370bFChW05zUaDWuDgti7bz+JiYn4+vrQr08f7a6HoWFhDB8xUm/e8+fNxctTt13fu3ePPv36I5VK2bLpj1yosWAoGo2GoLVr2bdv39O24Uufvn1fuyPmzp072bJ5M0qlEhdXV3r16qXTRlNTU1m2bBnHjh4lLS2Ncv7+9OnTB7lcniWv+Ph4+vTuTWxsLH9s2kT+/PkBOHnyJLt37+ZaVBRpaWkUL16crzt2FNv2f2hS8R62d/HGIxh9+vRh1KhRLF68mPT051tzpqens2jRIkaPHp3jXaSE3LNt83p279xCzz6DmT7nF8zM8zFpzDBSU1OyTXPi2F+sXLaYNh268tOCZTi7uDFxzDBUKqVOvDr1GrN87Rbt0bl7T+2569cimTxuBGX9KzJ7wTKGjBjHv2dOsnblr++troLhHD52giW/raRT+7YsmT8bVxdnRoydiFKl0hv/4uVwpsycQ/06tViyYDZVK1di3JTpXL9xUxunqKMDfXt+x6+L5jFv5lSK2BVm+JgJqOLidPLq+nV7/li7QnsENmn0PqsqfGBHjx5j2bJldOzQgZ8XLsDV1YVRY8agyqZtXbp0iekzZlKvbl0WLVxAQEAAEydN5saNG9o4mzZvZkfwTvr37cO8uXMwNzdn1JgxpKamAuDr48O6oLU6R/169ShSxA5PDw+d66WnpzNtxkxKigdpn4TNmzYRHBxM3379mDtvHubm5owZPVrbNvQ5evQoy379lQ5ff83ChQtxdXFhzOjROm3016VL+efMGUb++CMzZs5EERubZe3qM/PmzcPFxSVL+H8XLlC2bFkmTpzIgoULKV2mDBPGjycqMvKd6y0IH8obdzC6dOlC79696du3L7a2tpQrV46yZctia2tL//796dGjB127dn2PRRWyo9Fo2LVjM63adqJiwBc4u7jRf8hIFIpH/HPqRLbpdm7bRJ36jahVpwFOxZzp0XcwZubm/HVAdzcwU3Mz5Da22sPCwlJ77uTxwxR3caVNhy7YOxSlRCk/Onfvyb7d23mSlPTe6iwYxpbtwTSsV4f6dWpRvJgTA/v0xMzMjH0HD+mNvzV4FxX8y9K2ZXOKOznRrVMH3N1c2bHreRur9WV1/P3K4FCkCM7Fi9Hz224kJSVx7fpNnbzyWeTDRi7XHvnMzd9rXYUPa+u2bdSvX5+6detQvFgx+vXti5mZOfsPHNAbf/uOYMr7+9O6VUuKFStGl86dcHdzI3jnLiDzvrht+w7at2tLQEAAri4uDBsyhNhYBX+fOgVkTu21sbHRHlZWVpw6fZo6tesgkeg+vVy9Zg1ORYtSrdoX7/cPIbx3Go2G7du3065dOwICAnBxcWHI0KHExsZy6u+/s023bds26jdoQN26dSlWvDh9+/XDzMyMA0/baGJiIgcOHOC7777Dz88PDw8PBg0ezOVLlwi/fFknr927dpH4+DEtWmYdhe3RsyetW7fG08sLR0dHunbtioODA2fOnMndP4QgvEc5eovITz/9xN9//03Xrl0pUqQI9vb2dOvWjZMnTzJ37tz3VUbhNe7HRKNSKijj93z41NIyPx5evkSEX9KbJi0tjajICEq/kEYqlVLazz9LmuOH/6RL+6YM6N2VoFW/kpKcrJOPqampTnxTU1NSU1OJiozIjeoJeURaWhpXIqMo51dGGyaVSinnV5pL4fr/rS+FR+jEB6hQzo9L4VeyvcbufQewtLTAzcVZ59yGTVtp3r4TPfoPZuOWbWLN1yckLS2Nq5GRlPXz04ZJpVLK+vlxOTxcb5rL4eGULeunE+bvX04bPyYmBqVSqZOnpaUl3l5eXL6sP8/Tp8+QkJBA3bp1dMJDQkI5fuIEffqIUfpPwbO24Ve2rDbM0tISLy+vbNtbWloakVev4vdSG/Xz89N2Hq5evUp6erpOvk5OThQqXFgn31s3b7Ju3TqGDB2KVPr6n2FqtZonT55QoIB4ibDw8cjxJPnKlStTuXLl91EW4S2plAoArOU2OuEymRzl03MvS4iPQ61WI5NlTXP39i3t52o1alOosB02tgW5cT2KtSuXcvfObYaPngRA2XIV2L1jM8ePHKJKtS9RKRVsWr8GAKVC/7WFj1NcfAJqtRq5zFonXC6TcfvOXb1plEoVcplMJ0wmk6F4aRre6X/+ZfLMOaSkpGAjlzNj0nisra2055s3aYS7mxtWBfJz8XI4y1cHoVAo6fVd99ypnGBQ8fHxmfcjuUwnXCaTcfv2bb1plEolMj1tS6lUas8DyF6a+/5inJftP3AA/3LlKFSwoE7ZZs+dyw9Dh2JpYZGTagl51LN//5fXRcjk8mzbxrM2qi/N7Tt3tPkaGxtr11I8I5fJtN+HaampzJgxg2++/ZbChQsTExPz2vJu2bKFJ0+eUK169TeroCDkAQZdhXvnzh3Mzc0p+PRmfvz4cZYsWcKtW7coXrw4ffr0ISAg4JV5pKSkkJKiu84gNSUFUzOz91ZuQzt6+CBLf56t/Txq/PT3dq26DZpo/39xZ1dsbGwZ9+NgYqLvUsTeEb9yFejcvSdLF81h/uwpmJiY0rpdJy5dDEMiFkgJb6hM6VIsXTCHuPh49uw/yOQZP7Fw9gxt56RV82bauK4uzhgbGzNv0RK+6doJ01fsbicIb+rho0ecPXeOH0eM0Amft2ABNb+sQalSJQ1UMuFdHf7rLxYuXKj9PGHCBIOVZeWqVTg5OfHVV1+9UfzDhw+z7vffGTtuXJYOtSDkZQbtYLRs2ZIxY8bQuHFjduzYQYsWLWjcuDFVq1blypUr1KhRg61bt9K4ceNs85g2bVqWm0WvfoPp03/o+y6+wVSsVBXPF3Z6SktLAyBOqcDGxlYbrlIpcXF115tHAStrpFIpKpXuKINKpUT20kjIi57tMBV9L7ODAdC0eRuaBLZGqYjFMn8BHt6PIWj1MuyKOLxdBYU8ydqqAFKpFKVKd/G1UqVC/tKT52fkclmWBeAqlQobme5TwHzm5jg62OPoYI+vtxddvuvN3gOH6NBG/y5RPl6eZGRkcP/+A5yKvnrXFyHvs7KyyrwfKVU64SqVCrlN1t13IPPp88sLwFUqlfYJ87P/VSmV2NrY6MRxdXXNkt+BAwcpUKAAlStX0gkPDQ3j9OkzbN6yVRumVqtp2LgJA/r3o17dum9cT8EwKlWujJe3t/bzs+9MpVKJzYttQ6nE1c1Nbx7P2ujLIxwqpRKbF9pceno6jx8/1hnFUKpUyJ9eJyw0lBs3btC4ke4mFe3atqVdu3Z0fOF9Y0ePHGHB/PmM/PFHyr4w7Ur4QCQ5WkUgvMSgHYyLFy9qt7adNm0aU6dOZfjw4drzP//8M2PHjn1lB2PkyJEMHjxYJyzq9qc9NSefhQX5Xhiq12g0yOQ2hIWew8Utc+eTpKRErkZcon7DpnrzMDExwc3di7CQc1QKqAZkfmmGhZylYePm2V77+rXMXSzkL3RkACQSCTa2T0eijh6iYKHCuLp5ZEkvfLxMTEzwdHfjXGgYVQMyf4Sp1WrOh16gWeMGetP4entxPiSMls2ej4SdPR+Kr3f229oCqDVq7Y8AfaKuXUcqlSJ7abqW8HEyMTHBw92dkNAQqlTJHLVW3hm39AABAABJREFUq9WEhITQpIn++7+PtzchIaE0DwzUhp07fx6fpz8kixQpglwuJyQ0FLenPxoTk5IIj4igUaOGOnlpNBoO/nmQ2rW+wthY92tx7uyfUKvV2s+nTp9m06bNzJn9E7a2uvdBIW+ysLDA4qXvTLlcTmhIiLZtJCUmEhERQaNG+nenMzExwd3Dg9CQEKpUqQK80EabZn7Penh4YGxsTEhICF98kbkZwJ07d3j44IG2XY4aNYqUF3aqunLlCvPmzmXWTz9hb2+vDT9y5Ajz5s5l+IgRVKxYMRf/GoLwYeSog6HRaLh9+zaFCxfGPBd2cDE2NiYhIQGA69ev06CB7o+UBg0a6HQ49DEzM8PspelQpmaJ71y2j4lEIqFxs1Zs3rAWe4ei2BWxZ/3a5djYFKRiwPMdT8b9OJhKAV/QsEkLAJo0b83COdNw9/DCw9OHnTs2k5KczFd1Mv8dYqLvcuzIIfzLV6KAlRU3rl9j5bJF+JYsg7PL86c827dsoKx/RSQSCaf/Ps62zesYMmIcRkZGH/YPIbx3LQObMnPuArw83PDy9GDrjl0kJydTv3YtAKbPnk9BWxu+7Zr5FK5F08YMHjGaTVt3UKmCP4ePneBKZBSD+vYC4ElyMus2biagUgVsbeTExSewY9ceHsUqqPFF5pf4pcvhXL5yFb9SJclnkY/LlyP45bcV1PqyOgVemussfLxaNG/OT3Pm4OHhgZenJ9t27CA5JZm6dTIXXM/6aTa2trZ079YVgMBmTRk2fARbtm6lYoUKHDl6jKtXIxnQrx+QeV9sHtiM9Rs24ODgQBG7IqxZuxZbWxuqvDT1NiQ0lJiY+9SvVy9LuYoVK6bz+crVq0ikUpydnXP/jyB8EBKJhMDAQDZs2ICDoyN2dnasXbsWW1tbAp52HgBGjhhBlSpVtB2I5s2bM2f2bDw8PPD08mLH9u2kpKRQ52kbtbS0pG7duixbtowCBQpgYWHBkl9+wcfHB2+fzNF/ewfdkf34+HggczH4s1GPw4cPM2f2bHr0zHwPjOLp+g0zMzMsLS0RhI9BjjsY7u7uXLx4EQ+Pd386XaNGDdavX0/p0qUpW7YsR44coXTp0trzhw8ffu1Lb4RMzVu1JyU5mSULfyIx8TE+vqUYM2kmpqbPO18x0XeJj38+veWL6l8RH6difdBKVEoFLq7ujJk4UztFytjYhLCQs+zasZmU5CcULFSYgKrVadWuk861z/3vDJs3riU9LY3iLm6MGDOFcuV1pxkIn4aa1b8gLi6eVUEbUCqVuLm6MG3iWO0UqQcPHyJ9Ye1NCR9vfhw2iJVr17FiTRCODvZMGDUCF+fiABhJpdy+c4cDhw4THx+PlVUBPD3cmTtjCs7FM3/YmZiYcPjYCdas20BaWjpF7ArTollTWjXXPzonfJxq1KhOXHwca9cGZb5oz9WVyRMnaqc6PXj4UGddl6+vL8N/GMbqNWtZtWo1Do6OjB0zWueHf+tWrUhOTmbBwoU8fpxIiRK+TJ44KcvOd/v3H8DXxwcnJ6cPUlfB8Fq1bk1ycjILFyzg8ePHlChRgomTdNtGdHQ0cU87AJD5myU+Lo61QUEoFQpc3dyYOGmSzsLv73v0QCKVMmXyZNLS0vD396d3nz45Ktu+vXvJyMhg8aJFLF60SBteu3ZtBg8Z8oqUQq6SiHWk70Ki0Wg0OUlQokQJli9fnis7SV2+fJlq1arRqFEjPDw8mDFjBoGBgfj4+BAREcHGjRtZsmRJjt+vcTEy+p3LJghvwkqjf8cRQcht6VLT10cShFygFnPPhQ/ETc96qLwiedPs10cyEPPWeb+jmeO7yPTp0xk2bBj//fffO1/cx8eHM2fOkJqaysyZM0lMTOT3339n/PjxREZGsmHDBvHyPkEQBEEQBEH4iOR4BEMul5OUlER6ejqmpqbky5dP57ziLd99oNFoePDgAWq1moIFC2LyDltPihEM4UMRIxjChyJGMIQPRYxgCB+KGMF4Ox/DCEaOd5GaN2/eeyhG5qIrOzs7nbDbt28zbtw4VqxY8V6uKQiCIAiCIAhC7spxB6NLly7voxx6KRQKVq9eLToYgiAIgiAIgvCReKv3YERFRbFy5UqioqKYP38+hQsXZu/evRQrVkz7Xos3ERwc/Mrz165de5viCYIgCIIgCMLbk4qpgu8ixx2Mo0eP0qBBA6pWrcqxY8eYMmUKhQsXJjQ0lOXLl7N58+Y3ziswMBCJRMKrloFIxDZh/2fvrqOjOvYAjn93IUICMTQJcUcDQYuE4m7FtUALxYqVAsXdilsLFHeXUIoVLbR9eJEEgluQZDchbvv+CF1Y2AApSTaE3+ece87buTNzZ3jTu5kdE0IIIYQQ4qOR5g7G0KFDmTBhAgMHDiRPnjza8OrVqzN//vw05WVra8vChQtp0qSJ3vvnz5/Hz88vrUXEbuO4NKcR4r+433qMoYsgPhGaZPmxRWSOnIpEQxdBCPGRS/P4zz///EOzZs3eCC9QoADPnj1LU15+fn6cOXMm1fvvGt0QQgghhBBCZC1pHsGwsrLi0aNHuLi46ISfO3cuzaduDx48mKioqFTvu7u7c/jw4bQWUQghhBBCCGEgae5gtGnThiFDhrB582YUCgXJycn88ccffPfdd3Tq1ClNeVWpUuWt983NzfH3909rEYUQQgghhBAGkuYOxqRJk+jduzcODg4kJSVRpEgRkpKSaNeuHSNGjMiIMgohhBBCCJF55MDJD5LmDoaxsTFLlixh5MiRXLp0icjISEqVKoWHh0dGlE8IIYQQQgjxEflP52AAODo64uDgAMhWskIIIYQQQogU/6mD8csvvzBr1iyuX78OgIeHB/379+err75K18KJ/87Ezx+TCrVQ5rYg6fF9ovdvJOnhHb1xjUtUwLyR7gntmsQE1FO/TfmgVJLLvzFG7sVQWuVDExdDwq1AYg7vQBMZntFVEVmMRqNh45plHNy3m+ioSLx8itO990Bs7R3emm5vwDZ2bd2AWhWGk4sb3b7ph4dXEZ04QVcvsX7VEq4HXUWpVOLs6s6I8TMwMTEB4OGDe6z6ZSFBVy+RmJCAk4sbbTp0o1jJ0hlWX2E4Go2GjWt/4dC+3URFReLtU5yvew16Z1v7LWAbu7at17a1rj36a9vak8eP6N2tld50A4eOo2LlzwH45/xpNqz5hbt3bmBikotqNerSttPX5Mjxn3+XE1mYRqNh/ZrlHNy350VbK0aP3gOwsy/81nS/Bmxnx9aNqFVhOLu48dU33+Lp5aM3//Gjh3LuzN8MHTGe8hUra+9dPH+GdauXc+fOTUxNTPm8Rh3ad/6KHDlypHs9RRoo5cfzD5HmCWajRo2iX79+NGrUiM2bN7N582YaNWrEgAEDGDVqVEaUUaSRkY8fuWp+QezxPUT8MomkJ/fJ3eZbFGZ5Uk2jiY1BPXuI9gqfP/yVDI3JUciRmBO/EvHLZCK3LCZH3oLkbtUzE2ojspodW9bx6+6tdO89iEkzf8bE1JTxI78jPj4u1TR/HDvEyiULaNnuS6bNXYqzizsTRn5HuFqljRN09RITRw2mZKmyTJn1M1NmL6Zeo+YoX3nJTx4zhOSkJEZPms20OUtwcnFj8tihqMJCM7TOwjB2bl3H3t1b6d77OybP+BkT01xMGDXo3W1t6Xxatv2SqXOW4uTizsRRg7RtLW++AixevUPnatW+K6a5cuHrVx6A2zeDmTTme3z9yjFtzjIGDBnD6b/+YO2KnzOl3iLzbd+ygT27t9Gj9wCmzlyIiakp40Z+T3x8fKppThz7neVLFtG6XWdmzF2Ms4sb40Z+j/qV99q/du/Yone2x62bwYwfPYxSfmWZOXcJg4aO4u+/TrJ6+eJ0rZ8QmS3NHYxFixaxZMkSJk+eTOPGjWncuDGTJ09m8eLFLFy4MCPKKNLItHwN4s7/QfzFUyQ/CyH61/WQGI9xyYqpptGgQRMV8cr1/OXNuFgi188l4epZksMek/TwFtH7NpLT1gmFhXUm1EhkFRqNhj07N/NF646Uq1gFZxc3+g4ajioslL9PnUg13e7tm6hZtyHVa9XHwdGZ7n0GYWJqyu/792jjrFgyn3qNv6BZqw44OLlgX9iRz6pUx8jIGICIcDWPHt6nacv2OLu4YWvvQIcvvyEuLpZ7d25leN1F5kppa5v4onUnylaogpOLO30GprS1/506nmq6gB0bqVGnEZ/XaoCDowvde3+HsYkpvx9IaWs5cuTA2jqvzvX3qeNUrFydXLnMADh5/BBOLm60bNsFW7vCFC1eig5devLbnm3EREdnSv1F5tFoNATs3ELL1h0pX7Eyzi5u9Bs0jLCwZ/z1lvfaru2bqVW3ATVq1cPB0Zlv+gzExNSUQ/v36sS7dSOYXds30aff92/k8cfxwzi7uNK6XWds7ewpVtyXzl17sHfPDmlr4qOW5g5GQkICZcqUeSPcz8+PxEQ5/dPglDnIYetI4q3AVwI1JNwKJGdh11STKYxNsOgzAcu+EzFv+Q3KfLZvfYzCJBcaTTKa2Jh0Krj4GDwJeYRaFUYJ35fvAHPz3Hh4+XAt8JLeNAkJCdwMvqaTRqlUUtzXj6DAywCEq1VcD7qCpaU1PwzqSbf2TRg1pC9XL1/UpsljYYldYUeO/r6P2NgYkpIS2b93J5ZW1ri6e2VQjYWhPHmc0taKv9bW3L18tO3mdS/bmp82TKlUUsK3DNdSSXMjOIjbN69To3YDnXz+7dj+y9jEhIT4eG4GB31ItUQW9DjkESpVGCVfaTf/vtfe1tZuBF/TSZPS1krrpImLjWXm9Al83bMf1jY2evMxMn6trRmbEB8fz43gax9aNSEMJs0djI4dO7Jo0aI3whcvXkz79u3TlNeMGTO4c0f/uoD3FRcXR0REhM4Vl5j0QXl+zBRmuVEoc5AcFaETromKQGluoTdNUuhjogNWE7X5J6J2rgCFAovOg1HksdL/kBw5yVW9GfGXT0N8bPpWQGRpKlXKVCQra92RK0srG9SqML1pnkeEk5ychKWVbhqrV9I8DnkIwKZ1y6lZtxHDx03Hxc2TsT8M4NGDe0DKZhKjJ87k1o3rdGxRl7ZNaxGwYxPDx00nd57Up/+Jj5P637amr92o39XWdP+Qs7Sy1ub3ut/3B2Dv4ISXT3FtmG/pcgQFXuLE0YMkJSUR+uwpW9avAF7+NyCyj3/fQ5bWr7c163e815L1vNd00yxbsgBvn6I6ay5eVap0WYKuXub4kUPatrZp/SoAmfopPmrv1cEYOHCg9lIoFCxdupRixYrx1Vdf8dVXX1G8eHGWLFmCUpm2/srgwYNxc3OjVq1abNy48a1zHVMzefJkLC0tda5ZR8+mOZ9PWdKDW8T/8xdJj++TePc6UVt+Jjn6OSal9RyEqFRi3vxrUED03vWZX1iRqY4d3k+HL+por6SkjOm8JycnA1CrXmOq16qPq5snXbr3xa6wA78f+BVImcawZOEsLK2sGD9tPlNm/US5ClWYMnYYqrBnGVIukXmOH95Phxa1tVdmjIjHxcVx4uhBatRqqBNesnQ5OnbpyeIFP9KuWQ369WhHqTIVANk1MTs4evgAbb+op70SkzKmrf395x/8c/EcXbv3STWOb+mydOrag58WzKJV09r07t6J0mVS1gIpZJGx+Ii913YY586d0/ns55cyJHjjxg0A8uXLR758+bh8Wf9Q4tssXbqUHTt20LFjRywsLOjQoQNfffUVxYoVe6/0w4YNY+DAgTph0bO+S3M5sgtNdCSa5CSU5ha8+qegwtzijVGNVCUnk/T4Hjms8+uGv+hcKC1tiFw7W0YvPgFly1fW2ekpMSEBALVKhbVNPm14uDoMZ1d3vXnksbBEqcyhs6AbQK0Ow8o65Zdma5u8ADg4OOvEKezgxNOnjwH458JZzv7vFCs27sHMzBwAV3cvLpz/H0cO/kazVh0+oKbC0MqUr4y7vram1m1ranUYzi76z1162dZ0f3UOV6uwss77Rvw//zhMXFwsVWvUeeNeo2ZtaNi0NaqwUMxz5+Hpk0esW/kzBQvZ/af6iayjXPlKeL7S1hISUn7cDFepsLF52U7UahUub32vKfW811Ta99o/F88R8ughHVrpdmCnTRqNT9HiTJgyG4AmzVrRuGlLbVt78jiENSuXSFszNDlo74O8Vwfj8OHDGVaA+vXr8+WXX/LkyRNWrFjB8uXLmTdvHn5+fnz99de0adOGPG+Z/mBiYqLdwvJfSTk/4a3dkpNIenSXnM5eJFy78CJQgZGzF7Gnj7xfHgoFOfLbk3DjlTn1LzoXOawL8HztLDQxUeldcpEF5TIzI5eZmfazRqPBytqGfy6cwcUt5Y+86OgorgddpXb9pnrzMDIywtXdk3/On6FcxZRRseTkZP45f5Z6DZsBUKCgLTZ58/HgwV2dtA8f3KfUi1/z4uNSOrSv/4KsVChJ1mg+vLLCoFJra5fOn8HF9WVbCw66Sp16TfXmoW1rF85QrmJV4EVbu3CGug2bvxH/9/17KFOuEpaW+jerUCgU2ORN6dycOHqQvPkL4OLm+SHVFFmAvrZmbW3DxQtncXFL6VD8+16rW7+J3jyMjIxwc/fk4vmz2ulPr7/XmrdoR81X1vYA9O/dlS5f96Jsuc90wl9ta8ePHiJf/gK4uskBxuLjlWW6ZwUKFOD777/n6tWrHDlyhCJFijBgwABsbd++2Fi8KfavQ5iUqoxx8Qoo8xbCrF5bMDIh/uIpAMwadca02suXpmnl+uR08UFplY8chRwwb9IFpaUNcef/SImgVGL+RXdy2joStXMZKJQozC1QmFuA8hPuzH2CFAoFDZq0ZOuGVfzvzxPcuX2DeTMmYm2Tl3KvzDEe80N/9u7eqv3cqFkrDu4L4MjBvdy/e5slC2YQFxvD57Xqa/Nt3LwNe3dt5dSJIzx6eJ/1q5fy8P4d7eJbT++imOfOw/yZk7h9M1h7JsaTx4/wK5v6Dmni45TS1lqxdeNK/vdXSlubP3MC1jZ5KVvx5fTNsT/002lrDZu25tC+AI4c2sv9e7dZsvBFW6tZXyf/Rw/vc/XyBWrUaaT3+Tu3ruPO7Rvcu3OLLetXsGPLWrp27ydnE2RDCoWChk1asHnDav7+8w/u3L7JnBmTsbHJp7N2YtQPA/l193bt58bNWnJgXwC/H/yNe3fv8POCWcTGxlKjVl0ArG1scHJ20bkA8ucvSMFCL/+22b51A3du3+TunVtsWr+K7VvW81WPvtLWxEctzScGxcbGMm/ePA4fPsyTJ0+0c6f/dfbs+69/SG0ua5UqVahSpQpz585l48aNaS3iJy/h6hlizHNj6t8wZarU4/tEbpin3XpWaWkDr/ziqzA1w6xBe5TmFmhio0kKucvzldNJfhaSEj+PFcaeJQGw+HqEzrOer55J4t3rmVQzkRU0bdGOuNhYfp73Y8qBVEWKM2L8jxgbvxxJfPzoIRERLw9hrFS1BhHhajasWZZyIJWrO8PH/aidSgDQsGkrEuLjWbFkHpHPn+Pk4sbICTMpZGsPgIWlFcPHTWf9qiWM+aE/SYmJODi58P3ISalOzxIftyZftCM2Noaf500n+kVbGz7utbYW8pDnetraxjW/pNrWAA4f2INNvvyULFVW77PPnfmLbZtWk5AQj7OLO0NGTNauwxDZT7MWbYiNjWHRvBlERUXiU6Q4I8dPxfiVHZ5CXnuvVa5anYjwcDasWYFKFYaLqxujxk19o629y9nTf7Nl4xoSExJwdnFj6MgJ+L0YuRXiY6XQaNI2t6B9+/bs37+fFi1aULBgwTc6CaNHj37vvJRKJSEhIRQoUCAtRXgn1UQ5AE5kjvutxxi6COITodHIgk+ROXIqZMt5kTmKuGfddSaxO+cbugipMm2S+sYBWUWaRzACAgL49ddfqVSp0gc//PXRDyGEEEIIIcTHLc1rMOzt7d+66Do93bt3j65du2bKs4QQQgghhABAoci610cgzR2MGTNmMGTIkA8+IO99hIWFsXLlygx/jhBCCCGEECJ9pHmKVJkyZYiNjcXV1RUzMzOMjIx07oeF6T/1Up9du3a99f7NmzfTWjwhhBBCCCGEAaW5g9G2bVsePHjApEmT9C7yToumTZuiUCh42zpzOTVVCCGEEEKIj0eaOxgnT57k1KlTlCxZ8oMfbmtry8KFC2nSRP9BNufPn9eeGi6EEEIIIYTI+tLcwfD29iYmJiZdHu7n58eZM2dS7WC8a3QjNTEtZJtakTksNe8/JVCIDxGvNDV0EcQnIhGjd0cSQoi3SHMHY8qUKQwaNIiJEydSvHjxN9ZgWFhYvHdegwcPJioqKtX77u7uHD58OK1FFEIIIYQQ4r9TpnkfJPGKNB+0p3zxD/762giNRoNCoSApKSn9SvcfPQy6aOgiiE9EolJ+6ROZQ0YwRGaREQyRWbzdChu6CKmKDVhk6CKkyrRh1p+pk+YRDBlREEIIIYQQQqQmzR0Mf3//jCiHEEIIIYQQWYPsYvpB/tMEs+PHj9OhQwc+++wzHjx4AMDq1as5ceJEuhZOCCGEEEII8XFJ8wjG1q1b6dixI+3bt+fs2bPExcUBEB4ezqRJk/j111/TvZDi3bbv+Y2N23cRplLj5uLEt9274uPpkWr8IydOsWztBkKePKWwXSG6d+5AhTKltfenzJ7Pvt+P6qQpW6ok08aOAOD8P5cZMHyM3rwXzZiMt4f7h1dKZEk7A35l87btL9qaM717fI23l2eq8Y+e+IOVa9YR8vgJ9na2fPVlJ8qXLaO9v2rteo4cP8HTp8/ImTMnHu5udOnUAZ9X8ly7cTN//+80N27dImfOnOzYuC5D6ygMQ6PRsGbNavb9tpeoqCh8ihShd+++2NvbvzVdwO5dbN26BZVKhYuLK9/07IWXl5f2fnx8PEuXLObYsaMkJCRQurQfvXr3wdraWhunQf26b+T7/ZCh+PtX03nO7oDdPHn8mPz589O6TVtq1Kj54RUXWZJGo2HdmhUc+O1XoqIi8S5SjJ69+2Fn//Z1A3t272DH1k2oVGE4u7jRvWdfPL289eY/btQwzp75H8NGjKXCZ5UzqipCZLo0j2BMmDCBn376iSVLlujsIFWpUiXOnj2broUT7+f343+w6JeVdG7TksWzpuLm7MT3oyeiUofrjX/pahDjf5xN/VrVWTJ7GpXLl2PkpGncunNXJ1650r5sXblYe40c3F97r6i3p869rSsX06B2DWwLFsDL3S0jqysM6MixE/y8dBkd2rZh0ZyZuLo4M2zUWFRqtd74l68GMmnaDOrWqsmiuTOpVKE8YyZO4dbtO9o4he3t6PNNdxYvmMOsaZMpWLAAQ0eOQR3+sv0mJiZStXIlGtZ7849AkX1s2bKZ3bt20rvPt8ycNRtTU1NGjhxOfHx8qmmOHT3KkiVLaNeuA3PnzcfF1ZWRI4ejfqVNLln8M3///RfDhg1nytTphIWFMnHC+Dfy6j9gIKvXrNNeFSt+pr23Z08AK1asoH27Dixc9DPtO3Rk0cIF/PXXn+n6byCyjm1bNrBn13Z69unP9FnzMTU1ZczIoW9tj8ePHmbZkp9o3a4TM+f9hIurG2NGDkGtVr0Rd9eOrXKYsMi20tzBCAoKomrVqm+EW1pa6rzQRebZvDOABrVrUK/m5zg7OjCwV3dMTYzZe/B3vfG37t5DudK+tGneBCeHwnTt0AYPV1e27/lNJ56RkRE21tbaK0/u3Knes8iThz/++h91a3wuL8xsbOuOndSrU5u6tWrg5OhAv949MTExYd+BQ3rjb9+1m7J+pWn1RTOcHBz4smN73N1c2RnwcqSzejV/SvuWxLZQIZydHPnmq65ER0dz89ZtbZzO7dvyRdPGuDg7ZXQVhYFoNBp27thO6zZtqVixIi4urgwaNJiw0FBOnTqZarrt27dRt25datWujaOjE3369MXUxIT9+/cBEBUVxf79+/jq6+6U9PXFw8OD/gMGcfXqFQIDr+rklds8NzY2NtrL2NhYe+/33w9Rr349qvr7Y2tri79/NerWrceWzZsy5h9EGJRGo2H3jm20bNOB8hUr4eziRv9BQwgLfcafp1KfDr5z+xZq161Pzdp1cXR0pmef/piYmHBwv+73680bwezctpm+/QdndFWEMIg0dzAKFSpEcHDwG+EnTpzA1dU1XQol3l9CQgLXgm/i51tCG6ZUKildsgSXA6/pTXMl8Bp+JUvohJUtXfKN+OcvXaZZx2506vktsxYuJjziearl+OPv00Q8f069mp9/QG1EVpbS1m5Q+vW25luSK4FBetNcCQzSiQ9QpnQprqYSPyEhgV9/24+5uRluLi7pV3iR5YWEhKBSqfD1LaUNMzc3x8vLm8CrV/WmSUhIIDj4uk4apVKJr28pbech+Pp1EhMTdeI4ODiQP38Brr6W76JFC2jbphUD+n/L/v37dA56TUhIwNjIWCe+sbEx165dIzEx8b9XXGRJj0MeoVKFUdL35dRhc/PceHr5EHT1it40CQkJ3Ai+ppNGqVRS0rc0QYEv08TFxjJj2kR69PoWaxubjKuEEAaU5jUYX3/9Nf369WPZsmUoFAoePnzIqVOn+O677xg5cmSaCxAQEMDff/9NnTp1qFSpEr///js//vgjycnJNG/enO7du781fVxcnHYdiDYsPh4TY+NUUmQv4RHPSU5OxtrKUifc2sqSuy8W4L8uTK3WE98KlUqt/VyudCmqVCyPbcECPAx5zNLV6xg6diLzp00kR44cb+S598DvlC3lS/58eT+8UiJLetnWrHTCra0suXf/vt40KpUaKz3xw16bLvDn3/9j4rQZxMXFYWNtzdTxY7G0fP9DO8XHT6VKaRPW1lY64VZWVtp7r4uIiCA5ORkrPWnu3bunzTdnTiNyvzIC++9zXs23Q4eOlCzpi4mpCWfPnmXhgvnExsTQuElTAPxK+7Fv329UqPgZ7u7uBF+/zr79+0hMTCQiIhwbG3n3ZSf/tg2rV9bpAFhZWb+lPYa/aI9vprn/oj0C/LJkId4+RSlfsVI6l1qkK4UctPch0tzBGDp0KMnJydSoUYPo6GiqVq2KiYkJ3333HX379k1TXj///DN9+vShZMmSzJkzhwULFtCrVy9at25Njhw56N+/PzExMfTr1y/VPCZPnszYsWN1wgb2/oZBfbP+ISRZWfWqL198rs5OuDo70b57H85fuoJfyeI6cZ8+C+V/584z6vuBmV1MkU2ULFGcn+bOIjwigr379jNh6nTmzpj2RmdGZB+HD//O/HlztZ/HjB1nwNJA23bttf/bzc2d2NhYtm7dou1gtGnbDpVKxaCB/dFoNFhbW1OjRk22btmMQv4Q+egdOXyQRfNmaT+PHDspQ57z158nuXjhPLPm/Zwh+QuRVaS5g6FQKBg+fDiDBw8mODiYyMhIihQp8savQ+9j7ty5LFy4kK+//prDhw9Tv359ZsyYQa9evQCoUKEC06ZNe2sHY9iwYQwcqPuHbegd/VODsiNLizwolco3FnSr1OHYpPLHmY2VlZ746jd+OXyVXaGCWFrk4cGjkDc6GHsPHsYiTx4qlSuTSmqRHbxsa2qdcJU6XGc3nldZW1u9sTYrpW3qxs9laoq9nS32drYU8fai89c9+W3/Qdq2apGeVRBZSPnyFfB6ZWedhISUhbMqlVpnNECtVqc6/dbCwgKlUon6ldHXf9NY26S0MWtraxITE4iMjNT5nlKp1Km2WwAvLy82rF9HQkI8RkbGmJiY0H/AQPr0/Ra1WoW1tQ2//baXXLnMsLS0TDUf8XEoV/4zvLx8tJ8TEhIAUKtUr7VHFS6u+jcysbCwfNEedUc41GqVdirUPxfOEfLoIe1aNtaJM3XSWIoULc7EqTPTpT5CGNp//tnF2NiYIkWKUK5cuf/UuQC4desWderUAeDzzz8nKSlJZwF5tWrVuHPnTmrJATAxMcHCwkLn+lSmR0HKYmtPd1fOXvhHG5acnMzZi/9Q1Fv/1qFFvD05e/EfnbAz5y+mGh9SRikinkeS97VOiEaj4bdDh6n9uT85c6a5vyo+IiltzY1zFy5qw5KTkzl34SJFvL30pini7cW58xd1ws6eO49PKvH/pdEka7/gRfZkZmaGnZ2d9nJ0dMLa2poLF85r40RHRxEUFIi3j4/ePIyMjHB39+D8K2mSk5M5f/483t4padw9PMiZMycXzr+Mc//+PZ4+fYJPKvkC3Lx5k9y5c2P02rqLnDlzki9ffnLkyMGxo0cpV64cSqWMYHzszMzMsLWz114Ojk5YW9tw8cLL3TGjo6O4FnQVL58ievMwMjLCzd2TixfOacOSk5O5eP4cXt4pab5o2ZY5C5Ywe/5i7QXQ9euefDtAFnyL7OO9/yLs2rXre8VbtmzZez88b9683LlzB0dHRx4+fEhiYiJ3796lWLFiANy5cwcbWQD1Ti2bNGTK7AV4urvh4+nOll17iI2No26NlAXXk2bNI7+NDV93TpkC8EWjBvT/YTSbtu+mQtnS/H7sD4KCbzCodw8AYmJiWLlhM1UrVsDG2ooHIY/5ecVq7G0LUba0r86zz168xKPHT2hQu0am1lkYxhdNmzBt1hw8Pdzx8vRg+87dxMbGUqdmyv//U2fMJl/evHT7siMAzRo3YtDQ4WzetoPyZctw5NhxrgXfoH+flFHKmNhY1m3cTMXy5chrY014RAS7AvbyLDSMqpVfTtN78uQpEZHPefL0GcnJyQTfvAmAva0tuXLlyuR/BZERFAoFTZo2Y8OG9djZ2VGoYCFWr16FTd68OtvF/jBsKBU/+4xGjVJ+AW7WrDkzZ/6Ih4cHnp5e7Ny5ndi4WGrVqg2kLBSvXbsOS5YsJneePJiZmfHTTwvx9vHRdkL++utP1CoVXt4+GBsbc+7cWTZt3EDzL16OoD24f5+ga0F4eXkTGRnJju3buHPnNgMHDcrEfyWRWRQKBY2aNmfThrXY2hWmYMFCrFu9HJu8+ahQ8eV5FSOHfUeFzyrToFFTAJo0a8GcmVNx9/DEw9Ob3Tu3EhsXS81aKT+mWtvY6F3YnT9/AQoWss2UugmRGd67g7FixQqcnJwoVaqUzs4aH6JJkyZ069aNzp07s2vXLjp16sSgQYNQKpUoFAoGDx5M7dq10+VZ2Vn1KpUID49gxbqNKYefuTozdcxwbF6MNjx5+gzlK1vHFvPxYsSgfixbu56lq9dhb2fL+B++x8XJEUjZ9eLG7bvs+/0okVFR5LWxoYxvCbq2b4PxK2efAPx64BBFvb1wLPz2g7BE9lCtamXU4eGsXLMelUqFm6sLk8aN1k6ve/L0KQrly7ZW1MebYYMHsmL1WpavWoO9nR1jhg/VbjebQ6nk3v0HHDg0lYiICPJY5MHLw4NZUyfh/KI9AqxYu44Dhw5rP/f8NmVa5I+TxlOyhO6UPfHxatGiJbGxscybN5eoyEiKFC3K+HETdLaLffToIRGvnJFS1d+f8Ihw1qxejUqlwtXVlXHjJuhMf/q6ew8UCgWTJo5POWjPz49evfpo7+fIkZOAgACWLFmMRqPB1s6Or7/uTp269bRxkpKT2b5tGw8e3CdHjhyUKFGSH2fMpGDBQhn8ryIMpXmLNsTGxrJw3kyiIiPxKVqc0eMm67THkNfaYxX/z4mICGfd6hUpBz+6ujF63BSsrOXHUvFpUWjes7fQu3dv1q9fj5OTE126dKFDhw4fPLoQFRXFgAEDOHXqFJ999hnz5s1j7ty5DB8+nISEBPz9/dm4cSMFChRIU74Pgy6+O5IQ6SBRafTuSEKkg3ilqaGLID4Rich7TWQOb7e3n4puSLG/LTV0EVJlWvcrQxfhnd67gwEpW8Ju27aNZcuWcfLkSRo0aEC3bt2oXbt2uh6uFhsbS0JCAnny5PlP6aWDITKLdDBEZpEOhsgs0sEQmUU6GP/Nx9DBSNPKNBMTE9q2bcuBAwe4cuUKRYsWpVevXjg7OxMZGZluhTI1NSVPnjzcu3fvvdd+CCGEEEIIIQzvP2998e86CY1GQ1JSUnqWSSssLIyVK1dmSN5CCCGEEEKI9JemfUVfnSJ14sQJGjZsyPz586lbt+5/2qZv165db71/88VOMUIIIYQQQoiPw3t3MHr16sWGDRtwcHCga9eurF+/nnz58n3Qw5s2baodBUlNeq7tEEIIIYQQQmSs917krVQqcXR0pFSpUm/9o3/btm3v/XB7e3sWLlxIkyZN9N4/f/48fn5+aZ6CJYu8RWaRRd4is8gib5FZZJG3yCxZepH3vl8MXYRUmdbpZugivNN7j2B06tQp3UcT/Pz8OHPmTKodjHeNbqTGOCHqQ4smxHuJymVn6CKIT0ScxsTQRRCfiNxEGLoIQoiPXJoO2ktvgwcPJioq9c6Au7s7hw8fTvW+EEIIIYQQImtJ0yLv9FalSpW33jc3N8ff3z+TSiOEEEIIIQSg+M8brQo+YJtaIYQQQgghhHiddDCEEEIIIYQQ6UY6GEIIIYQQQoh0Y9A1GCJzbd17kHU79xKmDsfd2ZEB3TpQxMNVb9ybdx+wdMM2gm7eJuRpKN92aUvrhnUyucQiK9JoNKxds4p9v+0lKioSnyJF6dX7W+zt7d+aLmD3LrZt3YxKFYaLiys9evbGy8tbe/+3vXs4cuQwN4KDiYmJZsOmbeTOnVsnjwf377Ns2RKuXrlMQkIiLi4udOjYmRIlfTOiqsLANBoNG9Ys4+C+AKKjIvHyKU733gOxs3/71pZ7A7azc+sG1KownF3c6PZNPzy8fHTiBF29xLpVS7kedBWlUomzqzsjx/+IiUnKbl3fdGnN0ychOmnad+5O81bt07eSIkvYFbCHLVu3EaZS4eriQq9veuDt5Zlq/GPHT7ByzRoeP36CvZ0d3bp8SbmyZQBITExkxao1/O/0aR6FhGBubk4p35J0+7IzefPmBSDk8WPWrd/I+YsXUKnU5LWxofrn1WjbuhVGRrJNsPj4yQjGJ+LgH38xb8UGurZqyrLpY3F3cmDg+B9RhevfjjAuPg67gvnp2aElea0sM7m0IivbumUTu3ftoHefb5kxay6mpqaMGjmM+Pj4VNMcO3qEpUt+pm27DsyZtxAXV1dGjfwBtVqljRMXF4efXxlatW6Taj5jx4wkKSmJiZOnMXvuAlxcXBk7ZiSqsLB0raPIGnZsWc+vu7fRo/cgJs/8CVNTU8aP/I74+LhU0/xx7HdWLFlAq3admT53CU4ubowf+R3hr7S1oKuXmDDqe0qWKsuUWT8xdfbP1GvUDKVSdyv2Nh26snT1Nu1Vv3HzDKurMJwjx46zeMlS2rdry4K5s3F1cWH4yFGo1Wq98S9fucrkadOpW7s2C+fO4bOKFRg7YSK3b98BUt5lwTdu0K5taxbMnc2o4cO4f/8Bo8dN0OZx7959kjXJ9OvTm8ULF9Dj66/Ys/c3lq9clRlVFiLDSQfjE7Fx9z4a1fSnQfUquDjYM7hHZ0xMjAk4dExvfB93V/p0bkPNyhUwMpKBLpFCo9Gwc8d2WrdpR4WKn+Hi4srAQd8TFhrKqVN/pJpux/at1Klbj1q16+Do6ETvPv0wMTHhwP592jhNmjanZas2eHn76M0jPDychw8f0KJla1xcXLG3t6dzl27ExcVx587t9K6qMDCNRkPAzs20aN2RchUr4+ziRt9BP6AKC+XvUydSTbd7+yZq1m1I9Vr1cXB0pkefQZiYmnJo/6/aOMuXLKB+4y9o3qo9jk4u2Bd2pFKV6hgZGevklSuXGdY2ebWXqWmuDKuvMJxt23dQt24d6tSqiZOjI9/26YWJqQn79h/QG3/Hrl2U8StNyy+a4+joQOeOHXB3c2NnQACQsgPmlInj8a9SBYfChfHx9qZ3zx5cDw7myZMnAJQt48d3A/rjV7o0traFqFihPC2aN+OPk6cyrd7iHRSKrHt9BKSD8QlISEgk6MZtypYoog1TKpWUKVGUS9duGLBk4mPzOCQElSoMX9/S2jBzc3O8vLwJvHpVb5qEhASCg6/j61tKG6ZUKvH1LUVgoP40+lhYWFC4cGF+P3SQ2NgYkpKS+G3vHqysrHB39/jvlRJZ0uOQR6hVYZTw9dOGmZvnxsPLh6DAy3rTJCQkcCP4mk4apVJJCV8/rr1IE65WcT3oCpaWVvwwqBdd2zdl5JBvuXr54hv5bd+8js5tGvFd327s2LqepKTEdK6lMLSEhASuBwdT2rekNkypVFLK15crgUF601wNDKSUr69OmF/pUlwNDEz1OVFR0SgUCsxfm/apGyeKPHnypK0CQmRR8tP0J0D9/DlJycnYvDbVycbSgrsPHhmoVOJjpFKlTEWysrbSCbeyskatUulJARERESQnJ2Nlbf1Gmvv37r33sxUKBRMmTWXCuDG0/KIpCoUCKysrxo6fRG75Us521Nq2ZqMTbmllrb33uucR4SQnJ2FlZf1Gmgf37gLwOOQhABvXraBzt544u7pz9NB+xvwwkFkLV2jXd9Rv3BxXN09y57Eg6Ool1q5YjCoslC5f90nXegrD0r6fXmsz1lZW3Lt3X28alUqNtZXVG/FVKrXe+PHx8fyyfAXV/KtibmamN86Dhw/ZuTuAr7t1TXMdhMiKDN7BiImJYf369Zw4cYJHjx6hVCpxdXWladOm1KhR453p4+LiiIvTnY8bFx+PibFxKimEEO/r8OFDLJg3R/t59NgJb4mdsTQaDYsWzsfSyoqp02ZibGLM/n17GTdmFLPmzMPGJq/ByiY+3LHDB/h5/gzt5x/GTMmQ5yQnawCoXa8R1WvVB8DVzZOLF87w+4Ff6fBldwAaN2utTePs4kbOnDn5ef4MOnzZ/Y2pVEKkJjExkYmTpwIa+vbupTfOs2ehDB81hqqVK1G/rmymIrIHg3YwgoODqVmzJjExMZiYmHD//n3q16/P//73PxYtWkTz5s1Zt24dOXOmXszJkyczduxYnbDBPbvyfa+vMrr4Hw2rPHnIoVQSpg7XCQ8Lj3hjVEOIV5UvX1Fnp6eEhAQA1Cq1zh/0arUKF1c3vXlYWFigVCrfGOFQq1VY29joTaPPhQvn+d/ff7Fh01bMzMwBcHf34Ny5sxw6eICWrVJfHC6yvrLlK+ns9PSyrYVh/UpbC1ercHZ115tHHgtLlMocOpsH/Jvm35GQf/Mq7OCsE6ewgxPPnj5OtXweXkVISkriyeMQ7As7vn/FRJamfT+91mZUajXWr426/sva2grVawvAU+Jb6YQlJiYyccpUHj99wrRJE/WOXoSGhvL9sB8o4uNNv74yOiayD4Ouwfj222+pW7cuISEh3L17l8mTJ5OcnMyff/7J1atX+d///seECW//xXTYsGGEh4frXP2+6pRJNfg4GBnlxMvNmdP/XNGGJScnc+biFYp56v+jUAgAMzMz7OzstZejoxPW1jacv3BOGyc6OoqgoEC8ffQvzjYyMsLd3YMLF85rw5KTk7lw/jzeqSzo1icuLhYAhUL3taVUKNFoNGmolciKcpmZYWtXWHs5ODpjZW3DPxfOauNER0dxPegqXt5F9eZhZGSEm7sn/5w/ow1LTk7m4vmzeL5IU6BgIWzy5uPhA93peY8e3CN/gYKplu/2zWCUSiWWlvr/6BQfJyMjIzzc3Tl3/uUanOTkZM6fv0ARby+9aXy8vTl/4YJO2Nlz5/HxfvljzL+diwcPHzJl4gQsLCzeyOfZs1AGD/0BD3d3BvXvh1Ipy2JF9mHQ1nz06FEGDRqE4sWK+AEDBnDw4EFCQ0Px8PBg9uzZrFy58q15mJiYYGFhoXPJ9Kg3tW5Uh90Hj/Lr4RPcvv+QHxevIjYujgbVqwAwfu5iFq3ZrI2fkJDItVt3uHbrDgmJSTwNVXHt1h3uP0r9Fz6R/SkUCpo0bcbGDev4689T3L51i5k/TsMmb14qVqykjffDsO/ZvXun9nPTZl+w77dfOXRwP/fu3mXhgrnExsVSs9bL6QCqsDBu3rjBo4cpc+Rv377FzRs3eP48ZStlb+8i5M6dm1kzpnPz5o2UMzF+WczjxyGUKVsuk/4FRGZRKBQ0bNKSLRtW8b8//+DO7RvMnTEJa5u8lKtYWRtvzA8D+HX3Nu3nRs1acXDfHg4f/I37d2+zeMFM4mJjqF6rnjbfJs3b8OuurZw6cYRHD++zfvUvPLh/lxq1GwAp29gG7NjM7ZvBhDx6yLHDB1i+ZD5VP68l632yoebNmrJ33z4OHDzE3bv3mLdgIbGxsdSuVROAaTNmsmzFy79FmjZuzOkzZ9mybTt3791j9dp1XA8OpknDhkBK52L8pClcux7MkO++IzkpmbAwFWFhKu3I3LNnoQweNoz8+fPzdbeuhIdHaOOILEKpzLrXR8CgU6SsrKx4/vy59nN0dDSJiYkYv+gglChRgkePZBFyeqhZqTzq8Ocs3bCdMHU4Hi6OzBgxSDtF6vGzUG1HD+CZSkWX70ZrP6/f9Rvrd/1GqaJezB83LNPLL7KOL1q0IjY2lnnzZhMVGUmRosUYN26S9r9bgJBHj4gIfzklr6p/NcIjwlmzehUqlQpXV1fGjZuoMwXh118DWL9ujfbz0O8HAdB/wHfUrFUbS0tLxo6bxKpVyxk+7HsSE5NwdHJixMgxuKYyPUt83Jq2aEtsbAw/zfuRqKhIvIsUZ+T46Rgbm2jjhDx6yPOIl22tUtXqhIer2bBmGWpVGC6u7owYN11nsXjDpi2Jj49n+ZL5RD5/jrOLG6MmzKCQbcphkUZGxpw49jsb160gMSGeAgVtadS0JY2atcq8yotMU61qFcLDw1m1Zq32/TRx3Fjt++np06coX/l+LFrEh6GDv2Pl6jWsWLkKO3s7Ro8YjrOzEwDPQkP586+/AOjV91udZ02bPImSJYpz9tw5Hj58xMOHj2jf+UudOPv27M7A2gqRORQaA84t+PLLL7l9+zY//fQTJiYmDBs2jGvXrnH2bMqQ+NGjR+nYsSN3795NU77PLsk+0iJzqHLZGboI4hMRpzF5dyQh0kFu9B/AKkR6c3ZP/bR0Q4s9lHUPPTStkfWXAhh0BGPatGk0adKEIkWKoFAocHBwYPv27dr7T58+ZfDgwQYsoRBCCCGE+NRoPpID7bIqg3YwChQowKlTp7h+/TpxcXF4e3vr7BjVokULA5ZOCCGEEEIIkVZZYqWIh4cHxYoVe2M72nv37tG1qxw6I4QQQgghxMciS3QwUhMWFvbOXaSEEEIIIYQQWYdBp0jt2rXrrfdv3ryZSSURQgghhBBCpAeDdjCaNm2KQqF46yFZCllkI4QQQgghxEfDoB0MW1tbFi5cSJMmTfTeP3/+PH5+fmnO93oO/ae8CpHeTJPjDV0E8YlISDIydBHEJ0KR02C71wuRdSiy9CqCLM+g/3p+fn6cOXMm1fvvGt0QQgghhBBCZC0GHcEYPHgwUVFRqd53d3fn8OHDmVgiIYQQQgghxIcwaAejSpUqb71vbm6Ov79/JpVGCCGEEEII8aFkgpkQQgghhBAi3UgHQwghhBBCCJFuDDpFSgghhBBCiCxHdpH6INLByMY0Gg3b1//M0QM7iI6KxMO7BJ2+GUohO8dU0/y+dwu//7aVZ08eAWDv6EqTVt0o4VdJG2fFwklcvvA3atUzTE1z4e5dgpad+mJX2DmjqySyKI1Gw+a1S/l9326iop7j5VOCbr2+w9be4a3p9gVsZfe2dYSrwnB0cadLjwG4exXR3g95dJ+1vywg8MpFEhPiKelXgS97DMDK2iajqySyKI1Gw7Z1izn84r3m6V2CL3sOeet7LfDyWfZsX8Pt4EDUqmf0GzaNMhWqae8nJiayZe0iLpw5yZOQB5iZ5aZoybK07tQH67z5M6FWwtA0Gg0b1iznwL4AoqMi8fYpRvfeA7GzL/zWdHsDtrNj6wbUqjCcXdz56ptv8fDy0d4fObQfl/+5oJOmdr1GfNNnkPZz8wbV3sh34Pcjqexf48MqJYQBKTTZcB/YU1cjDF2ELGHPtpUEbFnB1/3GkL+gHdvW/cT9O8FMnLcJY2MTvWnO/X0MpTIHBe0cQKPhxOE97N2xmnEz12Dv6AbAkX3bsC3sjE2+QkRFRrBjw2Lu3rrGjz/vRJkjR2ZW0eBMc8o5GAA7t6xh5+bV9BowgvwFbdm0Zgn3bt/kx0VrUm1rJ48dZOHMCXzVezDuXkX4decm/jpxmJk/r8fSyprY2Bi+79MJJxd3Wrb/CoBNa5agCn3G+BmLUSo/rV+X5ByMFAFbV7J760q69xtN/oJ2bF37M/fuBDNl/sZU29qFMye5dvUCLm7ezJky5I0ORnRUJHOnDuXz2k1wdPYkKiqC1UtmoklOYtzMVZlUs6zDPGe0oYuQ6bZtXse2zWv5dsAwChSyZf3qZdy9fZM5P61ItV2dOPY7c2dMpkefgXh6+RCwYwsnTxxh3uLVWFlZAykdDDs7B9p06KJNZ2JqipmZufZz8wbV6NN/CKX8ymnDzHPnTvW52UlRd1tDFyFVMUfWG7oIqcpVra2hi/BOn9Y39CdEo9Gwf/d6GrfqSuny/jg4e/B1v7Gowp5x9q+jqaYrVa4qJctUopCdI4XsnWjRoRempmYEB13SxqlWpzleRUuTv6Adzm7efNG+J2HPHmtHPcSnRaPRsHfnJpq17kyZClVwcnGn98CRqMKecfrU8VTT7dmxkep1GlGtVgMKO7rwVe/BGJuYcORAAABBVy7y9EkIPQeMwNHZDUdnN3oNGMHN4EAuX0z9/ByRfWk0Gn7bvYHGLbviV94fR2cPevQfgzrsGWf+TP29VtLvM1p26EmZip/rvW9mnpuh4+ZTvnItbAs74e5VnM49BnPrRiDPnoZkVHVEFqHRaAjYuYUWrTtSrmJlnF3c+HbQMMLCnvH3qROpptu9fTO16jagRq16ODg606PPQExMTfl9/6868YxNTbC2yau9Xu1c/Ms8d26dOJ9C50Jkb1mig/H3338zZ84chg0bxrBhw5gzZw5///23oYv1UXv6+AHhqlCKlHj5i4iZeW7cPItyI+jie+WRnJTEn8f3Excbg7t3cb1x4mJjOH5oN/kL2mGTr2C6lF18XJ48fohaFUpx3zLaMDPz3Lh7FeFa4CW9aRITErgVHERx37LaMKVSSXHfMto0iQkJKFBgZPTyl3sjY2MUCiWBl9+vDYvs5enjh4SrQilWUve95upZlOCgf9L1WdFRkSgUCszNc6drviLreRzyCLUqjJK+ftowc/PceHgVISjwit40CQkJ3AgOosQraZRKJSV8/d5Ic/zwQTq3bUy/Xl+yZsVi4mJj38hvyaI5dG7bmO8HfMOh/b/KIcPio2fQNRhPnjzhiy++4I8//sDR0ZGCBVP+QH38+DEDBgygUqVKbN26lQIFCqSaR1xcHHFxcTph8fFxn3zvP1wdCoClVV6dcAvLvISrQt+a9t7tYCYM7UpCfDwmprnoO3Q69g6uOnEO/bqZTavmERcbQyF7JwaPWUBOI5nC8SlSq8IAsLTSXRdhaWWDWq2/rUVEqElOTtKb5sH9uwB4eBfFxNSUdcsX0qbTN2jQsH7FIpKTk1C/ow2L7Onf/9/1tZt3vdfSIj4+jo2r5lOhSm1ymUkHI7vTvsNeW9tlZWWN6sW91z2PCCc5ORkrqzfTPLh3V/u5in9N8hcoiE3efNy+dYPVy3/mwf17DBkxXhunTYeuFC9ZChMTU86f/R+LF84iNjaGBo2/SK8qCpHpDNrB6NWrF0lJSVy9ehUvLy+de0FBQXTt2pXevXuzefPmVPOYPHkyY8eO1Qnr2msoX/UZliFlzqpOHt3LykWTtZ8HjJj1n/OytXdi3Ky1xERF8r9Th1g6dwxDJ/6s08mo6F+Por7lCVc9Y++ONSyYPozhU5Z+8h27T8GJw/tYsmC69vOQ0dPfEvu/s7C0pv/Q8fyy8Ed+270FhULJZ/41cXHzQqFQZMgzRdbyx5HfWP7Ke23QyP/+XntfiYmJzJ/2AxqNhi49h2T480TmO3r4AD/Pn6H9PHzMlAx7Vu16jbT/28nZFRubvIz+YSAhjx5QyNYegFZtO2njuLp5EBcby46tG6SDYWAa+Z75IAbtYOzbt49jx4690bkA8PLyYu7cuVSrVu2teQwbNoyBAwfqhJ27FZdK7OyrVLmquHkW035OTEhZfByuDsXKJp82PCI8FEcXz7fmldPIiIK2Kbv/OLv7cOv6FQ7s3sCXvX7QxjEzz42ZeW4K2Tni5lmcXh2qc/bPI1SoWic9qyWyIL/ylXH3Kqr9nKBta2FYv9LWwtVhOLl46M3DwsIKpTIH4WrdXwfD1WE6O0SVLF2euUs3ExGuJkeOHJjnzkOPDo34rJDsrvIpKF2uSqptzeqNtvb299r7SOlcDOPZ00cMG79QRi+yqXLlK+H5yk5PCQkJAISrwrCxeTnqr1arcHF115tHHgtLlEol6tfeYWq16q273P27w9Sjhy87GPribN6wioSEeIyMjN+vUkJkMQZdg2FiYkJEROo7Pj1//hwTk7f/Im5iYoKFhYXO9Sn+ip4rlzkFbR20l52DK5bWebly8X/aODHRkdy4dhk3rxJpyluj0Wi/2PXeRwPviCOyj1xm5hSyK6y9Cju6YGWdl0vnXy68jo6OIjjoCp7exfTmkdPICBd3Ly5dOK0NS05O5tKFM3rTWFhaYZ47D5cunCEiXIVf+crpXzGR5eQy032v2b94r11+7b1289pl3L30rxN7X/92LkIe3WPouAXksbD6wNKLrCqXmRm2doW1l4OjM1bWNly8cFYbJzo6iutBV/DyLqI3DyMjI9zcvbh4/mWa5ORkLp4/k2oagFs3gwGwtsmbapzbN4PJnTuPdC7ER82gIxitW7emc+fOzJo1ixo1amBhYQFAREQEhw4dYuDAgbRtm/W34sqKFAoFtRu1ZffmZRSycyBfAXu2rfsJa5t8lC7vr403dWRP/Cp8Ts0GrQDYvHo+JUp/hk2+QsTGRPPn8d8IvHSGQaPnAfAk5D5/nzhAMd8K5LG0Jiz0MXu2rsTIxJSSr5yVIT4dCoWCek1asX3jSgrZF6ZAQTs2rVmCtU0+ylSsoo03/odvKVuxKnUbtQCgQdPWLJo1EVcPb9w9U7apjYuNxb9mA22aIwf2YO/gRB5LK64HXmbl4tnUb9Iau8JOmV5PYXgKhYK6jdqwc9MyCtk6kL+gHVvW/YSVTT78Krx8r00e2YsyFapR68V7LTYmmseP7mvvP338kDs3r2Gex4J8+QuRmJjIvKlDuX0jkIEjZ75Y5/MMgNy5LWV9WTanUCho2KQFWzasxtauMAUL2bJ+9S/Y2OSjXMWXP2aM/mEg5StWpn6j5gA0ataSeTMn4+7hhYenD7t3biEuNpbqteoBEPLoAceOHMKvTHnyWFhw+9ZNli9ZQJFiJXF2Sdn2/X9/nUStDsPTqwjGxsZcOHeGrZvW0qR568z/hxC65KC9D2LQDsbMmTNJTk6mTZs2JCYmYmyc0luPj48nZ86cdOvWjR9//NGQRfyo1W/WibjYGJYvnJRyIJVPSQaNmqszwvMk5AHPI9TazxFqFYtnjyFc9Yxc5rlxcHJn0Oh5FPMtD4CRsQnXrpxn/+4NREVFYGlpg2fRUoyYshQLKzn87FPV+Iv2xMXGsGTeNKKjIvEqUoKh42botLXHIQ94HhGu/fxZ1ZpEhKvZvGYpalUYTq4eDB03Q2d6wcMHd1m/8iciIyPIX8CWZq06U7+pfPF+yho070RcbCzLXnmvDR49563vtVvBV5k0oqf287plswGoXL0BPfqNRhX6hLN/HwNgRP8OOs/7YcIifIr7IbK3Zi3aEhcby0/zfiQqKhKfIsUZOX6aTrsKefSAiFfeYZWrViciXM36NctRq8JwcXVn5Lhp2ndYzpxGXDx/hoCdW4iLjSFf/gJUrFSVFm06avPIkSMHvwXsYPmSBaDRUMjWni+/7kWtOg0zr/JCZIAscdBeREQEZ86cISQkZb/xQoUK4efnpx3RSCs5aE9kFjloT2QWOWhPZJZP8aA9YRhZ+aC96GObDF2EVJlVbWXoIryTQUcw/mVhYcHnn+s/AEkIIYQQQgjx8TD4BLOYmBhOnDjBlStvHmYTGxvLqlWrDFAqIYQQQgghxH9h0A7GtWvX8PHxoWrVqhQvXhx/f38ePnyovR8eHk6XLl0MWEIhhBBCCCFEWhi0gzFkyBCKFSvGkydPCAoKIk+ePFSuXJm7d+++O7EQQgghhBAZQaHIutdHwKAdjJMnTzJ58mTy5cuHu7s7u3fvpk6dOlSpUoWbN28asmhCCCGEEEKI/8CgHYyYmBhy5ny5zlyhULBo0SIaNWqEv78/165dM2DphBBCCCGEEGll0F2kvL29OX36ND4+Pjrh8+fPB6Bx48b/KV8Lo6gPLpsQ7yM3siWyyByhynyGLoIQQgjxXgw6gtGsWTPWr1+v9978+fNp27YtWeCYDiGEEEIIIcR7yhIH7aW3y8GPDF0E8YmQEQyRWUKTZQRDZA4TpRwgKjJHlj5o7/hmQxchVWZVWhq6CO+UJQ7aE0IIIYQQIstQGvyouI+a/OsJIYQQQggh0o10MIQQQgghhBDpRqZIZRMajYYNa5ZzYF8A0VGRePsUo3vvgdjZF35rur0B29mxdQNqVRjOLu589c23eHi93NVr5NB+XP7ngk6a2vUa8U2fQTphvx/Yy+4dm3n44B65zMz5rHI1uvfqn271E1nHroA9bN66nTCVClcXF3p/0x1vL89U4x87foIVa9by+PET7O3s+KpLZ8qVLQNAYmIiK1at4e/TZ3gUEoK5uTmlfUvS7ctO5M2bV5tHxy5f8fjJE518u3buRJtWLTKmkiJL0Gg0bFm7lN/37yIq6jlePiXo2mswtnYOb023f89Wdm9bS7gqDEcXd77sMRB3zyLa+48f3WfNsvkEXblIYkI8JUpX4MseA7GytgHgyj9nGf9DH715T5ixFLdX8hLZgyG/Qy+eP8P61cu4c+cmpiamVKtRl/adu5Ejh/yJZkiaj+RAu6xKFnlnE9s2r2Pb5rV8O2AYBQrZsn71Mu7evsmcn1ZgbGyiN82JY78zd8ZkevQZiKeXDwE7tnDyxBHmLV6NlZU1kPJytLNzoE2HLtp0JqammJmZaz/v2r6JXds30anrN3h6+RAbG8uTxyGUq1ApYyudBXxqi7yPHDvO9Bmz+LZPL7y9PNm2YxfHT/zBL4sXYW1l9Ub8y1euMmjIMLp+2YkKZcvy+9GjbNqyjQVzZuHi7ERUVBTjJk2hft06uLo4ExkZycKfl5KcnMyCOTO1+XTs8hV1atekfp062rBcZrnIZWqaGdXOEj7FRd67tqxm55bV9Ow/gvwF7di8djH3bt9k+sK1qb7XTh0/yMKZ4+nWezDunkXZu2sjf504zIyf1mNpZUNsbAxD+nbEycWDFu2+AmDzmsWowp4x7sclKJVKEhMSiIzU/W9705rFXL5whtlLNqPI5n94fIqLvA31HXrrZjBDBvSkResOVKlWg7DQZ/w0fyZ+ZSvw5Ve9Mr7iBpaVF3lHndxm6CKkyvyz5oYuwjvJFKlsQKPRELBzCy1ad6Rcxco4u7jx7aBhhIU94+9TJ1JNt3v7ZmrVbUCNWvVwcHSmR5+BmJia8vv+X3XiGZuaYG2TV3u92rmIfP6cdat/4duBw6harSaFbO1xdnH7JDoXn6Kt23dSr25t6tSqiZOjI/369MLE1IR9+w/qjb9j127K+pWm1RfNcXR04MuOHXB3c2VXwB4AzM3NmTpxPP5VKuNQuDA+3t706dmD68HBPHnyVCcvs1y5sLGx1l6fUufiU6TRaNi7axPNWn1JmQpVcXJxp9eAUajCnnH6z2OpptuzYwPV6zSmWs2GFHZ0oVuv7zE2MeHIgQAArl25yNMnIXzTfwSOzm44OrvRc8BIbgYHcvniGQByGhlhZZ1Xe+XOY8mZv47jX7NBtu9cfIoM+R36x/HDOLm40qpdZ2ztClO0uC+dun7Db3t2EBMdnWF1FiKjZekOhkqlYtWqVYYuRpb3OOQRalUYJX39tGHm5rnx8CpCUOAVvWkSEhK4ERxEiVfSKJVKSvj6vZHm+OGDdG7bmH69vmTNisXExcZq7104fxpNcjKhoc/o26MTX3VqwY+Tx/Dsqe50FvHxS0hI4HpwMKV8fbVhSqWSUr4luRoYqDfNlcBASvmW1AkrU7p0qvEBoqKiUCgUmOc21wnfuHkrX7RpT8++/di0dRtJSUn/vTIiy3vy+CFqVSjFfMtow8zMc+PmWYTrgZf0pklMSOBWcBDFSr5Mo1QqKeZblutBKWkSEhNQoMDIyEgbx8jYGIVCSdCVC2/kCXDmr+M8fx6Bf80G6VE1kcUY8js0ISEBY2NjnfjGxsbEx8dzIzgoPaonhEFk6Ql+d+/epUuXLnTq1CnVOHFxccTFxemExcfFYWyif0gzO1KrwgCwfDF/+F9WVtaoXtx73fOIcJKTk7GyejPNg3t3tZ+r+Nckf4GC2OTNx+1bN1i9/Gce3L/HkBHjAXj86CEajYZtm9bQtXtfzMzNWb/qF8aOGMTM+ct0vsTFxy0iIoLk5OQ3pkJZW1lx794DvWlUKvUb8a2srAhTqfTGj4+PZ+nylVTzr4q5mZk2vEnjhni4uZEnT26uXA1k2YpVhIWp+Obrbh9UJ5F1hf/7XnvtHWVpZaN9570uIkJNcnLSG+9CSysbHt6/A4CHV1FMTE1Zt2IhbTp+gwYN61cuIjk5CXVYqN58jxwIoGSp8uTNV+BDqyWyIEN+h5YqXZY9O7dw/MghPqtSDbUqjM3rU35YVYXpf7YQHwODdjAiIt4+f/358+fvzGPy5MmMHTtWJ6xn34H0/va7DypbVnb08AF+nj9D+3n4mCkZ9qza9Rpp/7eTsys2NnkZ/cNAQh49oJCtPckaDYmJiXTr8S2+pcsCMGDIKLp1aM6li+co5Vcuw8omspfExEQmTJ4GaPi2d0+dey2aNdX+b1cXF3LmzMmc+Qvp+mUnjKUTmy2cOLKPpQumaT9/P+rHDHmOhaU1/YdM4JdF09m3ezMKhZLPqtbExc0LhZ5970OfPeHCub/o9/34DCmPyHxZ6TvUt3RZOnX9hp8XzGTOjIkYGRnTsk1Hrly+iEIp0/HEx8ugHQwrK6u3zmfVaDTvnO86bNgwBg4cqBN241727vWXK18Jz1d2qUhISABSfvGzsXm5845arcLF1V1vHnksLFEqlajVuv9WarVKu5OKPv/ujvHoYcrL0frF8wo7OmnjWFpakcfCUqZJZTMWFhYolUpUarVOuEqtxsbaSm8aa2urN+Kr1WpsrK11whITE5kwZRpPnj5h2qQJOqMX+nh7eZGUlMTjx49xKPz2XV7Ex8GvXGXcPYtqPyckpCw0DleHYW3zcoF7uDoMZ1cPvXlYWFihVObQjn68mubV91qJ0uWZs2QLEeFqcuTIgXnuPHzTsSEVC9m9kefRg3vIk8cCv/JVPqh+IuvISt+hAI2btaJR05aowkIxz52Hp49DWLNyCQX1tEeRiRRZehVBlmfQf708efIwefJkfv/9d73X4sWL35mHiYkJFhYWOld2nx6Vy8wMW7vC2svB0RkraxsuXjirjRMdHcX1oCt4eevfTtHIyAg3dy8unn+ZJjk5mYvnz6SaBlJ2vAC0HQufIsUAeHj/njbO8+cRPI8IJ3+Bgv+9kiLLMTIywsPdnfPnX85TT05O5vz5i/h4e+tNU8Tbm3MXLuqEnT13Xif+v52LBw8fMmXieCwsLN5Zlhs3b6JUKrGytPpvlRFZTi4zcwrZFdZehR1dsLLOy6ULp7VxoqOjuHHtCh7exfTmkdPICBd3Ly69WKwNKW308oXTeHi9mcbC0grz3Hm4dOE0EeEq/MpV1rmv0Wg4enAPVT6vR86cWXpGsUiDrPQd+i+FQoFN3nyYmJhw/Ogh8uUvgKub/o60EB8Dg74xS5cuDYC/v7/e+1ZWVmTDXXTTnUKhoGGTFmzZsBpbu8IULGTL+tW/YGOTj3IVX35hjv5hIOUrVqZ+o5TtzRo1a8m8mZNx9/DCw9OH3Tu3EBcbS/Va9QAIefSAY0cO4VemPHksLLh96ybLlyygSLGSOLu4AWBn70C5CpX4ZfE8evb5jlxmZqxduQT7wo4UK1Eq8/8xRIb6olkTps+cjYeHO96enmzbuYvY2Fjq1KoBwLQZs8ib14ZuX3YGoGnjRnw39Ae2bNtOubJlOXLsGNeCg+nXtzeQ0rkYP2kK12/cZPzokSQnJRMWlrI+I0+e3BgZGXHlaiCBQUGULFECs1y5uBIYyE9LfqH65/7kyZPbMP8QIsMpFArqNW7Fjo0rKWTnQIGCdmxesxhrm3yUqVBVG2/C8L6UrehPnYYpZ6I0aNqGRbMm4OrujbtnEfbu3EhcbCz+NRtq0xw5GIB9YWcsLK24FniJVUtmU69Ja+wKO+mU4fLFMzx5/JDPazdCZF+G/A4F2LF1A6X8yqFQKPjz5HG2b1nHoKGjyZEjR+b+QwiRjgzawWjXrh0xMTGp3i9UqBCjR4/OxBJ9vJq1aEtcbCw/zfuRqKhIfIoUZ+T4aTr7d4c8ekBERLj2c+Wq1YkIV7N+zXLUqjBcXN0ZOW6adng3Z04jLp4/Q8DOLcTFxpAvfwEqVqpKizYddZ797aAfWL54ARPHDEWhVFK0WElGjpsmv/hlQ9WqViE8PJxVa9ahUqlwdXVl4rgxWL+Y8vTk6VOdaY1Fi/gwbPAgVqxey/KVq7Gzt2PMiB9wcU75Q+5ZaCin/vobgJ59++k8a/rkiZQsURwjIyOOHDvO6nUbSEhIoFDBgjRv2pgvXlmXIbKnRl90IC42lqXzpxIdFYlXkRIMHTtT5732OOQBzyPU2s8Vq9QkIlzNlrVLUKvCcHL1YOjYmTrTVh7dv8uGlT8RGRlB/gK2NG3VmfpN2rzx/MP7d+PpUxx7B+eMrKbIAgz5HXr29F9s2biaxIQEnFzcGDpyIqXLlM+ciguRQeSgPSE+wKd20J4wnE/xoD1hGJ/iQXvCMLL0QXundhi6CKkyr9g0TfEXLFjA9OnTCQkJoWTJksybN49y5fRvwrNkyRJWrVrFpUspW3v7+fkxadKkVOOnRlawCCGEEEIIkQ1t3LiRgQMHMnr0aM6ePUvJkiWpU6cOT57o34jnyJEjtG3blsOHD3Pq1CkcHByoXbs2Dx7o344+NQYfwYiJieHMmTPY2NhQpIjuwqjY2Fg2bdr01nMw9JERDJFZZARDZBYZwRCZRUYwRGbJyiMYkX/uMnQRUpW7QuP3jlu+fHnKli3L/PnzgZTNCBwcHOjbty9Dhw59Z/qkpCSsra2ZP39+mv4eN+gIxrVr1/Dx8aFq1aoUL14cf39/Hj162TkIDw+nS5cuBiyhEEIIIYQQWUdcXBwRERE61+uHTkPK4bVnzpyhZs2a2jClUknNmjU5derUez0rOjqahIQEbGxS335ZH4N2MIYMGUKxYsV48uQJQUFB5MmTh0qVKnH37t13JxZCCCGEEOITM3nyZCwtLXWuyZMnvxHv2bNnJCUlUbCg7rEBBQsWJCQk5L2eNWTIEOzs7HQ6Ke/DoNv8nDx5koMHD5IvXz7y5cvH7t276dWrF1WqVOHw4cOYm5sbsnhCCCGEEEJkKfoOmTbJgDPgpkyZwoYNGzhy5AimpqZpSmvQEYyYmBidrUwVCgWLFi2iUaNG+Pv7c+3aNQOWTgghhBBCiKxF3yHT+joY+fLlI0eOHDx+/Fgn/PHjxxQqVOitz/jxxx+ZMmUK+/fvp0SJEmkuo0FHMLy9vTl9+jQ+Pj464f8uRGnc+P0XsbzKNjr4g8smxPt4aCYnrYrMkTtHlKGLID4R171rGboI4hNRNCHI0EXI1oyNjfHz8+PQoUM0bdoUSFnkfejQIfr06ZNqumnTpjFx4kT27dtHmTJl/tOzDTqC0axZM9avX6/33vz582nbtq2c5C2EEEIIITKXQpF1rzQYOHAgS5YsYeXKlVy9epWePXsSFRWl3USpU6dODBs2TBt/6tSpjBw5kmXLluHs7ExISAghISFERkam7Z/P0NvUZoSwi8cNXQTxiZARDJFZjBWydajIHDKCITJLgyw8ghH5125DFyFVucs3SlP8+fPnaw/a8/X1Ze7cuZQvn3JafLVq1XB2dmbFihUAODs7c+fOnTfyGD16NGPGjHnvZ0oHQ4gPIB0MkVmkgyEyi3QwRGaRDsZ/k9YOhiEYdA2GEEIIIYQQWY1GYdBVBB89+dcTQgghhBBCpBsZwcimtvz2O2t37SNMHY67kwMDu7alqIer3rg37z1gycadBN68Q8jTUPp92Zo2DXSHyFdu/5Wjf53lzoNHmBgbU9zLjV7tW+Bk//ZtzkT2o9Fo2LBmGQf3BRAdFYmXT3G69x6InX3ht6bbG7CdnVs3oFaF4eziRrdv+uHhpbuDXNDVS6xbtZTrQVdRKpU4u7ozcvyPmJiY8OTxIzavX8Wli2dRq8KwtslH1c9r8UXrjhgZGWVklYWBaDQa1q5Zyf7f9hIVFYlPkaL06v3tO9vant072bZ1MypVGC4ubvTo2RtPL28Anj+PYN2aVZw7e4anT59gYWlJhYqV6NDxS71nL0VERPBt7x6Ehj5j/abt5M6dO0PqKrIWp57tcB3YDZNC+Ym4GMjl/uMJ/98/qcZ3/rYzTt3bksvRlvhnKh5t20fQ8Bkkx6VMbcyR2xyvsf0o2KQmJgXyEnH+CpcHTiL8dOp5CvExkxGMbOjgH38zd+UmurVsxIqpo/BwcmDAxNmEhUfojR8bF49dgfz0av8Fea0s9cY5dzmIL+p8zpJJPzBn5EASE5PoP2EmMbFvHk0vsrcdW9bz6+5t9Og9iMkzf8LU1JTxI78jPj71tvDHsd9ZsWQBrdp1ZvrcJTi5uDF+5HeEq1XaOEFXLzFh1PeULFWWKbN+Yursn6nXqBlKZcqOGQ/u3UWjSaZHn++YtXAlXb7uw/69u1i3ckmG11kYxtYtGwnYtYNeffrx46x5mJqaMmrkMOLjU1+PcvzoEZYu+Zm27Towe94iXFxdGTVyGOoXbS0sNJTQ0FC6ftWd+YuW0H/AYM6e/h9zZ8/Qm9/c2TNwdnHJkPqJrMm2ZT18pg/j+oQFnCjXjOcXAym/5xeM89vojW/XpiHeEwdxfcJ8jhavz8Xuw7FrWR+vCS8PQivx8wTy1fiMC19+z7FSjXh64A/K/7YcE7sCmVUtITKVdDCyofUBB2hcowoNP6+Mi4Md33fvgImxMQG/n9Abv4i7C307taRWpXIYGekf1Jo9YgANPq+Eq4M9Hs4OjOjdlZBnYQTefHOnAZF9aTQaAnZupkXrjpSrWBlnFzf6DvoBVVgof5/S374Adm/fRM26Daleqz4Ojs706DMIE1NTDu3/VRtn+ZIF1G/8Bc1btcfRyQX7wo5UqlIdIyNjAEqVKU+fAcPwLV2WQrZ2lK1QicbNW/PnyWMZXm+R+TQaDbt2bKdVm/ZUqPgZLi6uDBg0hLDQUP489Ueq6XZs30qduvWoWbsujo5O9OrTDxMTEw7s3weAk7MLP4wYTbnyFbG1taOkbyk6du7C33/9SVJSkk5ev+7ZTVRUJM2at8zQuoqsxaV/F+79son7K7cRefUG//QaTVJ0LA5ffqE3vnXFUqhOnuXhhgBi7jzg2cE/eLgxAKuyKYeTKU1NKNS8NoHDphN24jTRN+5yffx8om/cwalHu8ysmhCZJkt0MJKTk1MNv3v3biaX5uOWkJBI0M07lC1RRBumVCopW8KHS9dupttzIqOjAbDI/eaUApF9PQ55hFoVRglfP22YuXluPLx8CAq8rDdNQkICN4Kv6aRRKpWU8PXj2os04WoV14OuYGlpxQ+DetG1fVNGDvmWq5cvvrU80VFR5MljkQ41E1nN45AQVKowfH1LacPMzc3x9PIm8OoVvWkSEhIIDr5GSd/S2jClUomvb2mCAvWnAYiKisLMzIwcOXJow+7evcOGdWsYMGgISmWW+KoUmUBhZIRl6aI8O3TyZaBGw7PfT2JVoZTeNKpT57AsXRTLssUByOVSmAJ1/Xmy92hKnjlzosyZk6TXRvyTYuKwqVT6jfyEyA4M+taMiIigVatWmJubU7BgQUaNGqXzC9LTp09xkaHpNFE/jyQpORkbS90/umwsLQhVh6fLM5KTk5m9YiMlvNxxc7RPlzzFx0GtCgPAylp3qoCllbX23uueR4STnJyElZV1qmkehzwEYOO6FdSs25AR46bh6ubJmB8G8vDBfb35Pnp4n727t1GrXtbfrk+knUrb1nTbjZWVNSqVSl8SIiLCSU5OxlpfmjD9acLDw9m4fi116tXXhiUkxDN96iS6dPuaAgVkCsunxDifNcqcOYl7EqoTHvc4FJNC+fSmebghgGtj5/LZkXXUi75E9WuHCD32Nzem/gxAUmQUqlNn8RjeCxPbAqBUYt+uMdYVfDEpJO0ryzL0YXrpdNCeoRi0gzFy5EguXLjA6tWrmThxIqtWraJJkyY682vfdUxHXFwcEREROlfcW+bnig/349K13Lz3gPEDuhu6KCKDHTt8gPZf1NVeSUmJGfKc5OSU/85r12tE9Vr1cXXzpEv3PtgVduD3A7++ET/02VMmjPqeipWrUauudDCygyOHD9GyeSPtlZhBbe1V0dFRjBs9AgdHJ9q176QNX7l8GQ4OjnxevWaGl0F8/GyqlsNtSA8u9R3LiXLNOd2iNwXq+eP+Qy9tnPNffg8KBTXvHqde1D849+nIw417IJUZHEJ87Ay6i9SOHTtYuXIl1apVA6Bp06Y0aNCARo0asWvXLgAU7+ipTZ48mbFjx+qEff/Nlwzp2TVDypzVWeXJTQ6l8o0F3WHhEaku4E6LH5eu5Y+zF1k09nsK5NW/4E1kH2XLV9LZ6SkhIQHgxS5OebXh4WoVzq7uevPIY2GJUplDu8j21TT/joT8m1dhB2edOIUdnHj29LFOWFjoM0YP64+XT1G+6fvdf6uYyHLKla+o3ekJXm1rKmxeaWtqtQpXVze9eVhYWKJUKt8Y4VCrVVjb6I5qREdHM3rkD+Qyy8XwkWPImfPl1+HFi+e4c/s2TRrqru9p3+YLWrVpR/sOnf9bJUWWF/9MRXJiIiYF8uqEmxTMS1zIM71pvMb248HaXdxbtgWA55eukdPcjOKLxhE8eRFoNETfvMefNTqSwywXOS1yExfylFJrZxF9616G10kIQzDoCMbTp09xcnLSfs6XLx8HDx7k+fPn1K9fn+gX8/zfZtiwYYSHh+tc/bt1yMhiZ2lGRjnxcnXi9D9XtWHJycmc/ieQYp76t6l9HxqNhh+XruXo3+eYP/o77ArmT4/iiiwul5kZtnaFtZeDozNW1jb8c+GsNk50dBTXg67i5V1Ubx5GRka4uXvyz/kz2rDk5GQunj+L54s0BQoWwiZvPh4+0P2yffTgHvkLFNR+Dn32lFFD++Hq7knv/kNlbnw2YmZmhp2dvfZydHTC2tqGCxfOaeNER0dxLSgQb58ievMwMjLC3d2Ti6+kSU5O5sL5c3h5v0wTHR3FqBFDyZkzJyNGjcPY2Fgnn2HDRzN3/k/aq++3AwCYOn0WDRo2Ts9qiyxGk5BA+NnL5Kte8WWgQkHezyui/vOc3jQ5zEzfGInQ/Dvd+7UfSZOiY4gLeUpOKwvy165MyO5D6Vp+IbIKg45gODo6cvXqVZ11Fnny5GH//v3Url2bZs2avTMPExMTTExMdMISX/uy+NS0bViL8QuW4e3mRFF3FzbsOUhsXBwNP68EwNh5v5Dfxope7VN2xEhISOTW/ZQ58ImJiTwNVXPt1l1ymZrgYJvyx92PS9ey/8RfTP2+D2ampoSqUtZzmJvlwtTk0/73/pQoFAoaNmnJlg2rsLUrTIFChVi/ehnWNnkpV7GyNt6YHwZQrmIV6jdqDkCjZq2YN3Mybh7eeHh6E7BzC3GxMVSvVU+bb5Pmbdi4djnOLm44u7pz5NA+Hty/y3c/jANedC6G9SN//kJ07taLiHC19nmvjqaI7EGhUNC4aTM2bliHnZ09BQvasmb1Cmzy5qVCxUraeMOHDabiZ5Vo2KgpAE2bfcGsmdNw9/DE09OLnTu3ExsXS81adYAXnYvhQ4mLi2PQ4KHEREcT8++mFZaW5MiRA1tbO52yRESkjAgXdnCUczA+AbdmL6fksqmoz1wi/H8Xcf62MznNc3Fv5TYASi6fSuyDxwSNmAnA44DDuPTvQvj5K6j/voi5myOeY/rxOOCwtuORr1ZlFAoFkdduYe7miPfU74kMusn9FdsMVk8hMpJBOxi1a9dm+fLl1K9fXyc8d+7c7Nu3j1q1aqWSUrxNzUrlUEVEsnTjTkLVEXg4OzBreH9sXkyRevwsFOUrv6o8U6np/P047ed1u/exbvc+ShXxZOHY7wHYtv8IAL3HTNd51oheXWjweSXEp6Npi7bExsbw07wfiYqKxLtIcUaOn46x8cuOfsijhzyPeLmpQKWq1QkPV7NhzTLUqjBcXN0ZMW66zmLxhk1bEh8fz/Il84l8/hxnFzdGTZhBIduUjQQunDtNyMMHhDx8QPfOLXTKtHXP0QyutTCEL1q0JjY2lvnzZhMVGUmRosUYO26yzohDyKNHRLwyJbSKfzXCI9SsXb0SlSplOtXYcZO0C79vBAcTFBQIQPduulOdli5fTcGCcnjop+7R5r0Y57fBc/S3KQftXbjK3w2/Iv7Fwu9cDrZoXhmxCJ6UMg3Ka2x/TO0LEv80jMd7DhM0cpY2jpFlHrwmDMS0cCESwtSEbN9P0MhZaBIzfq2REIag0LxrFXUGUqlUPHz4kKJF9U+teP78OWfPnsXf3z9N+YZdPJ4exRPinR6aeRi6COITYayQzStE5rjuLT/uiczRICHI0EVI1fPTvxm6CKnKU6auoYvwTgYdwbC2tn5jO8FX5cmTJ82dCyGEEEIIIYThGHyFZExMDCdOnODKlTcPQYqNjWXVqlUGKJUQQgghhBDivzBoB+PatWv4+PhQtWpVihcvjr+/P48ePdLeDw8Pp0uXLgYsoRBCCCGE+NRoFIose30MDNrBGDJkCMWKFePJkycEBQWRJ08eKlWqxN27dw1ZLCGEEEIIIcR/ZNAOxsmTJ5k8eTL58uXD3d2d3bt3U6dOHapUqcLNmzcNWTQhhBBCCCHEf2DQDkZMTIzO6akKhYJFixbRqFEj/P39uXbtmgFLJ4QQQgghhEgrg+4i5e3tzenTp/Hx8dEJnz9/PgCNG/+3E1MjzAq+O5IQ6SCnQvYwF5kjUWPQ17X4hBS89KehiyCE+MgZdASjWbNmrF+/Xu+9+fPn07ZtWwx4TIcQQgghhBAijQx60F5GuR0sU6tE5ohVmBm6COITkawx+K7i4hMRnZTL0EUQn4gyXqmfhWZoEWcPGLoIqbIonfUPw5RvLCGEEEIIIUS6kQ6GEEIIIYQQIt1IB0MIIYQQQgiRbqSDIYQQQgghhEg3su9hNrErYA9btm4jTKXC1cWFXt/0wNvLM9X4x46fYOWaNTx+/AR7Ozu6dfmScmXLaO+f+OMke/bu5XrwDZ4/f87CuXNwc3PVySMsTMXSZcs4e+480TExOBS2p03rVlSpVCnD6ikMT6PRsG7NCg789itRUZF4FylGz979sLMv/NZ0e3bvYMfWTahUYTi7uNG9Z188vby19xfOm8mFc2cJCwvF1DQX3kWK0rnL1xR2cHwjr4iIcPr37k5o6DPWbtpJ7ty5072ewvA0Gg3r1yzn4L49KW3Npxg9eg94Z1v7NWA7O7ZuRP2irX31zbd4evm8EU+j0TB+9FDOnfmboSPGU75iZe29i+fPsG71cu7cuYmpiSmf16hD+85fkSNHjnSvp8iaNBoNW9ct4fD+nURFReLpU5yuPb+nkN2b76R/Xb10jj3b13DrRhDqsGcM+GEqZSr468T538nDHPxtO7dvBBL5PIKJs1fh7Jr697UQHyMZwcgGjhw7zuIlS2nfri0L5s7G1cWF4SNHoVar9ca/fOUqk6dNp27t2iycO4fPKlZg7ISJ3L59RxsnNi6WokWK0K1L51SfO33mTO49eMCYUSP5ecF8Kn32GZOmTCP4xo30rqLIQrZt2cCeXdvp2ac/02fNx9TUlDEjhxIfH59qmuNHD7NsyU+0bteJmfN+wsXVjTEjh6BWq7Rx3Nw9+XbA98z/eTljJkxBo9EwesQQkpKS3shv/uwfcXZxfSNcZC/bt2xgz+5t9Og9gKkzF2Jiasq4kd+/ta2dOPY7y5csonW7zsyYuxhnFzfGjfxep639a/eOLSgUijfCb90MZvzoYZTyK8vMuUsYNHQUf/91ktXLF6dr/UTWFrBtNfsCNtGl5xDGTV+KiUkupozuT3x8XKpp4uJicHTx4Mse36UaJzYuFq8iJWnTuXdGFFukEw2KLHt9DKSDkQ1s276DunXrUKdWTZwcHfm2Ty9MTE3Yt1//Fms7du2ijF9pWn7RHEdHBzp37IC7mxs7AwK0cWpWr06Hdm0p5eub6nOvXA2kSaOGeHt5YmtbiHZtWmNubs714OD0rqLIIjQaDbt3bKNlmw6Ur1gJZxc3+g8aQljoM/48dSLVdDu3b6F23frUrF0XR0dnevbpj4mJCQf3/6aNU6deQ4oWL0HBgoVwc/ekQ6cuPHv6hCdPHuvktXfPLqKiomjavFWG1VMYnkajIWDnFlq27kj5ipVxdnGj36BhhIU946+3tLVd2zdTq24DatSqh4OjM9/0GYiJqSmH9u/ViXfrRjC7tm+iT7/v38jjj+OHcXZxpXW7ztja2VOsuC+du/Zg754dxERHp3tdRdaj0Wj4bddGmrbqQpkKVXF08aDngNGow55x5s9jqabz9fuMVh2+oWzFaqnGqfJ5PZq36UaxkmUzoORCZA0G72BoNBpu3bpFYmLKicjx8fFs3LiRVatW8ezZMwOXLutLSEjgenAwpX1LasOUSiWlfH25EhikN83VwMA3Og5+pUtxNTAwTc8u4uPN0WPHiXj+nOTkZI4cPUZ8fDwlihdPcz3Ex+FxyCNUqjBK+pbWhpmb58bTy4egq1f0pklISOBG8DWdNEqlkpK+pQkK1J8mNjaGgwf2UbCQLfny5deG3717m43rVtN/0BAUyo/jVxzx37xsa37aMHPz3Hh4+RAUeFlvmpdt7WUapVJJCd/SOmniYmOZOX0CX/fsh7WNjd58jIyNdcKMjU2Ij4/nhpyz9El4+vghalUoRV/pBJiZ58bNsyjXg/4xYMmE+DgYdA1GUFAQderU4d69e7i6urJ//35atmxJYGAgGo0GMzMzTp48iYeHR6p5xMXFERcX91pYPCYmxqmkyF4iIiJITk7Gykr3sBprKyvu3buvN41KpcbayuqN+CqVOk3PHj50CJOmTqNlm3bkyJEDExMTRo/4AXs7uzTlIz4eKlXKNBMra932ZmVlrb33uoiI8JQ2qifN/Xv3dMJ+DdjJymWLiY2Nxb6wA2MnTsPIyAiAhIR4ZkydyJfdupO/QEFCQh6lV7VEFqRWhQFgqafd/Hvvdc9ftDVLqzfTPLh3V/t52ZIFePsU1Vlz8apSpcsSsHMrx48c4rMq1VCrwti0fhUAqrDQ/1wn8fFQq1L+f7a00u2AWlrZaO8JIVJn0BGMIUOGULJkSc6fP0/Dhg1p0KABhQsXRqVSERYWRsWKFRk3btxb85g8eTKWlpY616Kff86kGnzaVq5eS2RkFFMmTmDe7Fl80awpE6dM49bt24YumkgnRw4fpHXzBtorKSkxQ5/n/3kNZs37mUlTZ2FnX5jpk8dp59uvWr6Uwg6OVKue9U8wFWl39PAB2n5RT3slZlBb+/vPP/jn4jm6du+Tahzf0mXp1LUHPy2YRaumtendvROly5QHkJGzbOqPI7/RtdXn2iuj33VCZHcGHcE4efIk+/fvp3jx4kyYMIE5c+awePFi7S+WQ4cOpW3btm/NY9iwYQwcOFAn7NErv1RldxYWFiiVyjcWMKrUaqxf++XvX9bWVqheWwCeEt/qvZ/78NEjdgUE8PPC+Tg7OQHg5urCP5cusytgD/36yOK17KBc+c/wemX3nYSEBADUKhU2Nnm14Wq1ChdXN715WFhYprTR10Y41GrVG9NTzM1zY26eGzv7wnh6+9C+VVP+PHmCqtWq88/F89y5fYtmDXU7GB3bNKNlm/a06/Dlh1RVGFi58pXw9Cqi/ZyQkNKxDNfb1tz15pHnRVsLV7/Z1qysU9raPxfPEfLoIR1aNdSJM23SaHyKFmfClNkANGnWisZNW6IKC8U8dx6ePA5hzcolFCwkI7TZUelyVXDzLKr9nJiY8q4LV4dhbZNPGx6uDsPJNfVZFUKIFAbtYERGRmLz4g8Mc3NzzM3NsbW11d53cHDg8ePHqSUHwMTEBBMTE52wsE9kehSAkZERHu7unDt/kc8qVgQgOTmZ8+cv0LhhA71pfLy9OX/hAs2bNtGGnT13Hh9vb73x9fl3WppSoTsIliOHEk2yJq3VEFmUmZkZZmZm2s8ajQZraxsuXjiLq1vKH3nR0VFcC7pK3QaN9OZhZGSEm7snFy+co8JnKVNSkpOTuXj+HPUbNX3L0zVo0Gj/0BwyfAzxr0yHvH4tiHmzpzN5+mwK2coffR+7XGZm5Eqlrbm80tauB12lbv0mevPQtrXzZ7XTn5KTk/nn/FnqNWwGQPMW7ahZW/fd2L93V7p83Yuy5T7TCVcoFNjkTfnj8vjRQ+TLXwBXN/njMjvKZWZOLjNz7WeNRoOVdV4uX/ifdgvZ6Ogobly7TM16zQ1VTJGJNAqDL1P+qBm0g2FnZ8fdu3dxdEzZU3ratGkUKFBAe//p06ep/govXmrerCk/zpyFp4c7Xp6ebN+5k9jYWGrXqgnAtBkzyZc3L12/TNlytmnjxgweOowt27ZTrmwZjh47zvXgYPr3fTllIOL5c54+eUpoWMpc53sPHgBgbW2NjY01DoULY2dny5z5C/i6W1csLPJw8tSfnD13nnGjR2Xyv4DILAqFgkZNm7Npw1ps7QpTsGAh1q1ejk3efFR4ZT77yGHfUeGzyjR40YFo0qwFc2ZOxd3DEw9Pb3bv3EpsXCw1a9UBIOTRQ04cO4Jv6TJYWlry7Nkztm5ej4mxMX5lU6am2L7WiYiICAegsIOTnIORDSkUCho2acHmDauxtbOnYCFb1q1eho1NPp21E6N+GEiFilWo3yilA9G4WUvmzpyCm4cnHp4+BOzcQmxsLDVq1QXA2sZG78Lu/PkLUrDQyx+4tm/dQGm/cigUCv48eZztW9bz3dDRcg7GJ0KhUFC3cWt2bFpBITsH8he0Y8vaxVjZ5MOvQlVtvEkj+lCmgj+1G7YEIDYmmpBHL9c/Pn38kNs3r5E7jwX58hcCIPJ5OM+ePkYdlrKRzaMHKVvEW1nnxcr65WidEB8zg3YwatasSWBgIJUrp3xZ9OzZU+f+/v37KV26tL6k4hXVqlYhPDycVWvWolKpcHV1ZeK4sdrO2dOnT1G+std70SI+DB38HStXr2HFylXY2dsxesRwnJ2dtHH+/PMvZsyeo/08eeo0ADq0a0vH9u3ImTMnE8aM4ZcVKxg9bjwxMTHY2dny3cD+Ogf2ieyneYs2xMbGsnDeTKIiI/EpWpzR4yZj/MquOyGPHhIRHq79XMX/cyIiwlm3egUqVcp0qtHjpminrRgZG3Pl8j/s2rmVqMhILK2sKVqsBFNmzHtjAwPx6WjWog2xsTEsmjeDqKhIfIoUZ+T4qW+2tYiXba1y1epEhIezYc0KVKowXFzdGDVuqratva+zp/9my8Y1JCYk4OzixtCRE/B7sQ5DfBoaNu9IXGwsvyyYQnRUJJ5FSjBkzGyMjV/Omngccp/nEWrt55vBV5k4/OUU4TW/pHyPVqlen2/6p/z4dubv4yyeM0EbZ/70kQA0b9ONL9p9nZFVEiLTKDQaTZadz3Lr1i1MTU11pk29j9uyjaDIJLEKs3dHEiIdJGtkuF5kjuikXIYugvhElPHKuj8gqc8fMXQRUmXlW83QRXgng45gvIuLi4uhiyCEEEIIIYRIA4P/JBYTE8OJEye4cuXNA7diY2NZtWqVAUolhBBCCCGE+C8M2sG4du0aPj4+VK1aleLFi+Pv78+jRy8PzwoPD6dLly4GLKEQQgghhBAiLQx+0F6xYsV48uQJQUFB5MmTh0qVKnH37qdzjoUQQgghhBDZiUE7GCdPnmTy5Mnky5cPd3d3du/eTZ06dahSpQo3b940ZNGEEEIIIYQQ/4FBOxgxMTHkzPlynblCoWDRokU0atQIf39/rl2T3aCEEEIIIUTm0igUWfb6GBh0Fylvb29Onz6Nj4+PTvj8+fMBaNy48X/KN15h+sFlE+J9WCU8NXQRxCciNqf5uyMJkQ7Ck+XgSiHEhzHoCEazZs1Yv3693nvz58+nbdu2ZOFjOoQQQgghhBCvydIH7f1X127IInGROSwSQg1dBPGJkBEMkVkeJRQydBHEJ6Kij4Whi5Aq1YWjhi5CqqxL+hu6CO9k8HMwhBBCCCGEENmHdDCEEEIIIYQQ6cagi7yFEEIIIYTIajQK+Q3+Q0gHI5vQaDSsXbOS/b/tJSoqEp8iRenV+1vs7Au/Nd2e3TvZtnUzKlUYLi5u9OjZG08vbwCeP49g3ZpVnDt7hqdPn2BhaUmFipXo0PFLzM1fzgf/+acFXL1ymTu3b+Pg6MDc+T9naF2FYW3fs48NO3YTpgrHzdmRft274OPpnmr8w3/8ybK1mwh58hR7u0J806kdFcqU0t73b9JGb7pvOrenbfNGAAybMJ3gW7dRh0eQO7c5fiWL8U2nduTLa5O+lRNZyq6APWzZuo0wlQpXFxd6fdMDby/PVOMfO36ClWvW8PjxE+zt7OjW5UvKlS2jvX/ij5Ps2buX68E3eP78OQvnzsHNzfWNfK5cDWTFqtUEBgWRQ6nE1dWVSePHYmJikiH1FFmTRqNh+/qfOXpgB9FRkXh4l6DTN0MpZOeYapqgy2f5dftq7twIRK16Rt+h0/GrUC3V+CsWTebIvm207TqAOo3bZUAthDAM6Z5lE1u3bCRg1w569enHj7PmYWpqyqiRw4iPj081zfGjR1i65GfatuvA7HmLcHF1ZdTIYajVKgDCQkMJDQ2l61fdmb9oCf0HDObs6f8xd/aMN/KqVasOVapm/UVH4sP8fvwkC5atpnPrFiyZORk3Fye+GzMZlTpcb/xLV4MY/+Nc6tf8nCWzplClfBmGT/6Rm3fuaeNsW/GTzjWk7zcoFAr8PyunjVOqeBHGfN+f1QtnMn7IAB4+esyoqbMyvL7CcI4cO87iJUtp364tC+bOxtXFheEjR6FWq/XGv3zlKpOnTadu7dosnDuHzypWYOyEidy+fUcbJzYulqJFitCtS+dUn3vlaiDDR43Gr5Qvc2fNYO7smTRu1ACFUr4uPzW/bl/FgYCNdP5mGKOmLcfENBczxvYlPj4u1TRxsTE4unjSscf378z/zJ+HuRH0D1Y2+dOz2EJkCfLGzAY0Gg27dmynVZv2VKj4GS4urgwYNISw0FD+PPVHqul2bN9Knbr1qFm7Lo6OTvTq0w8TExMO7N8HgJOzCz+MGE258hWxtbWjpG8pOnbuwt9//UlSUpI2nx7f9KZBoyYUKmSb4XUVhrVp5x4a1q5O/ZrVcHYszKCeX2FqYsyvB4/ojb9l917KlS5J2+aNcHawp1v71ni6urB9zz5tnLzWVjrXH3+fplTxItgVKqiN06pJA4p6eVCoQH6K+XjR/osmXLkWTGJiYkZXWRjItu07qFu3DnVq1cTJ0ZFv+/TCxNSEffsP6I2/Y9cuyviVpuUXzXF0dKBzxw64u7mxMyBAG6dm9ep0aNeWUr6+qT735yVLadq4Ea1btcTZyQmHwoXxr1IFYyOj9K6iyMI0Gg37d6+ncauulC7vj4OzB1/3G4sq7Bln/0p9d6ESfpX4on1P/Cp8/tb8VaFPWLPkR74ZOJ4cOWQySZakUGTd6yOQJTsY1atX586dO++OKAB4HBKCShWGr+/LaSfm5uZ4enkTePWK3jQJCQkEB1+jpG9pbZhSqcTXtzRBgfrTAERFRWFmZkaOHDnSrwLio5CQkMi1G7fwK1lcG6ZUKvErWZzLQdf0prkcdF0nPkDZUiVTjR+mVnPq9Dnq10z9yznieSQHjp6gmLcnOXPKF3N2lJCQwPXgYEr7ltSGKZVKSvn6ciUwSG+aq4GBb3Qc/EqX4mpg4Hs/V61WExgUhJWlJf0HDaZ1+458N2Qoly5f/k/1EB+vp48fEK4KpUiJlyOpZua5cfMsyo2gix+Ud3JyMotnj6Ze0w7YO7p9aFGFyJIM+u28a9cuveHHjh0jICAABwcH4O0nesfFxREXpztcGR8Xh/EnNFdWpQoDwMraWifcysoalUqlN01ERDjJyclY60lz/949vWnCw8PZuH4tderVT4dSi49NeEQEScnJWFtZ6oRbW1ly9/4DvWnC1Gq98cNU+qdU/fb7McxymVK1Yrk37v20ci3b9+wnNi6OIl4eTBnx7ikI4uMUERFBcnIyVla67ydrKyvu3buvN41KpcbayuqN+CqV+r2f+ygkBIDV69bzdbeuuLm6cPDQ7wz9YQQ/L1yAvb1dmuohPl7h6pQzjiyt8uqEW1jmJVz1Yecf/bptJUplDmo11L/+TIjswKAdjKZNm6JQKPSe1t23b18AFAqFznSc102ePJmxY8fqhPXp25++/Qakb2GzkCOHD7Fg3mzt51FjJ2T4M6Ojoxg3egQOjk60a98pw58nPk17Dx6hpn9lTIyN37jXplkjGtT8nJCnz1i5YSuTZi9kysjvUXwkw8Ui60tOTvkuql+vLnVq1QTA3c2N8xcusu/AAbp+mfraDfFxO3l0LysXTdZ+HjAiY9Z43Q6+yv6ADYyduUbeXSJbM2gHo06dOuTIkYNly5ZRoEABbbiRkREXLlygSJEi78xj2LBhDBw4UCfs7v3H6V7WrKRc+YranZ4gZToBgFqlwsbm5a8tarUKV1f9w68WFpYolco3RjjUahXWNrq/GkZHRzN65A/kMsvF8JFjZFrKJ8rSwoIcSuUbC7pV6nBsrK30prGxskolvuUbcS9cvsrdBw8ZPbif3rysLCywsrDAwd4Op8L2tOzWm8tB1ynmnfquQuLjZGFhgVKp1G448S+VWv3GqOu/rK2tUL22ADwlvtV7Pzfvi3ef04vR8385OBTmydOn752P+PiUKlcVN89i2s+JCSkbpISrQ7GyyacNjwgPxdHlv79zgq6c43m4ikFfNdKGJScnsWHFHPbv3sCMJfpndgjxsTHoGoy9e/dSo0YNypQpQ8ArC/HSwsTEBAsLC50ru0+PMjMzw87OXns5OjphbW3DhQvntHGio6O4FhSIt4/+TpqRkRHu7p5cfCVNcnIyF86fw8v7ZZro6ChGjRhKzpw5GTFqHMZ6flkWnwYjo5x4urlw5uIlbVhycjJnL16iaCpbhxb18tCJD3D6/EW98X89eBgvN1fcXZzeWZZ/Rz3/7VyL7MXIyAgPd3fOnX851z05OZnz5y9QxNtLbxofb2/OX7igE3b23Hl8vL31xtenYMGC5M1rw/0HulP+Hjx4qPMjmMh+cuUyp6Ctg/ayc3DF0jovVy7+TxsnJjqSG9cu4+ZV4j8/p1K1+oyfvY5xs9ZoLyub/NRr2oHvxsxNj6oIkSUY/KfoAQMG8Pnnn9O+fXt2797NrFmy9WRaKRQKGjdtxsYN67Czs6dgQVvWrF6BTd68VKhYSRtv+LDBVPysEg0bNQWgabMvmDVzGu4ennh6erFz53Zi42KpWasO8KJzMXwocXFxDBo8lJjoaGKiowGwsLTULvR++PABsTExqFRhxMfFc/NGMAAOjk4Yyc4r2UqrJg2YPGcR3u6ueHu4s2X3r8TExlGvZsoWxRNnLSB/Xhu6d2oLQItG9fh2+Dg27gigQplS/H78JEE3bvJd7+46+UZFR3Pkj7/o1aXDG8+8EnSdwOAbFPfxJk9ucx6GPOaXtZuwL1SQojJ6kW01b9aUH2fOwtPDHS9PT7bv3ElsbCy1X0xdmjZjJvny5tVOW2rauDGDhw5jy7btlCtbhqPHjnM9OJj+ffto84x4/pynT54SGpaybu3ei46EtbU1NjbWKBQKWjRvzuq163B1ccH1xRqMe/fvM+KHoZn8LyAMSaFQULtRW3ZvXkYhOwfyFbBn27qfsLbJR+nyL7dknzoyZceomg1aARAbE83jRy/XMT578pA7N4PInceSvPkLkdvCitwWVjrPypEjJ5ZWebG1d86Mqon3JAftfRiDdzAAfH19OX36NAMGDMDX11fvmgzxdl+0aE1sbCzz580mKjKSIkWLMXbcZJ0Rh5BHj4gIj9B+ruJfjfAINWtXr0SlSplONXbcJO0UhBvBwQQFpezA0r2b7tzjpctXU7BgIQDmzZnJpX9e/tLYr2/PN+KI7KF6lc9QR0SwbN1mwlRq3F2cmD56KDYvFtc+efYMpfLlvOJiPl6MHNSXX9ZsZMnqDRS2K8TEYd/h6qQ7BeXQ8ZNoNBpqVK3E60xMTDh26n8sX7+F2Ng4bKytKFe6JJ1aNZetQ7OxalWrEB4ezqo1a1+8n1yZOG6s9v309OlTlK/MYS9axIehg79j5eo1rFi5Cjt7O0aPGI6z88sRsT///IsZs+doP0+eOg2ADu3a0rF9yiFnzZs2ISE+np+WLOX58+e4urgwecI47GxlG+5PTf1mnYiLjWH5wklER0Xi6VOSQaPmYmz8cpbEk5AHPI9Qaz/fCr7K1JHfaD+vX5byo2mlzxvwdb8xmVV0IQxOoclif83v2rWLw4cPM2zYsP88JH3txt10LpUQ+lkkfNhuIkK8r9ic5oYugvhEPEqQH4ZE5qjoY2HoIqQq9NJJQxchVXmLfWboIrxTlhjBeFXjxo3fui2tEEIIIYQQIusy+ASzmJgYTpw4wZUrbx7uFhsby6pVqwxQKiGEEEIIIcR/YdAOxrVr1/Dx8aFq1aoUL14cf39/Hj16pL0fHh5Oly5dDFhCIYQQQgghRFoYtIMxZMgQihUrxpMnTwgKCiJPnjxUqlSJu3dlDYUQQgghhDAMDYose30MDNrBOHnyJJMnTyZfvny4u7uze/du6tSpQ5UqVbh586YhiyaEEEIIIYT4DwzawYiJidE5FVqhULBo0SIaNWqEv78/165dM2DphBBCCCGEEGll0F2kvL29OX36ND4+Pjrh8+fPB/jPu0nlTlR/aNGEeC8JObL3qfEi67CMevTuSEKkg7jcuQxdBPHJyLrb1IoPY9ARjGbNmrF+/Xq99+bPn0/btm3l0D0hhBBCCCE+IlnuoL308DDo4rsjCZEOkpRZ7igZkU3ljn5q6CKIT8ST3K6GLoL4RHi5ORi6CKl6dumUoYuQqnzFKhq6CO8kfx0JIYQQQgjxCo3C4EfFfdTkX08IIYQQQgiRbqSDIYQQQgghhEg3MkUqm9i+5zc2bt9FmEqNm4sT33bvio+nR6rxj5w4xbK1Gwh58pTCdoXo3rkDFcqU1t6fMns++34/qpOmbKmSTBs7Qvv52o2bLF6xhsDgG+RQKqlSsTy9u3UmVy7ZgSQ72xnwK5u27XjR1pzp0+MrvL08U41/9MQfrFiznpDHT7C3s+XrLztRvqyf9v7KtRs4cvwET58+I2fOnHi4u9G1U3t8Xsvzz/+dZs36Tdy8fQdjIyNKFC/KuBHDMqyeIuvZ8tth1uzeT5g6HHenwgzq2pai7i56496895DFG3cSeOsuIU9D6d+5FW0a1NSJs3X/EbbtP8qjp6EAuBa2o2uLBnxWqniG10VkLRqNhnVrVrL/t1+JiorEp0hRevbuh5194bem27N7J9u3bkKlCsPFxY3uPfvg6eUNwPPnEaxbs5LzZ8/w9OkTLCwtqVCxEu07fom5eW5tHhfOn2Xt6hXcuX0LE1NTqteoTcfOXcmRI0eG1lm8g+LjONAuq5IRjGzg9+N/sOiXlXRu05LFs6bi5uzE96MnolKH641/6WoQ43+cTf1a1VkyexqVy5dj5KRp3Lqje4J6udK+bF25WHuNHNxfe+9ZaBjfjRyHvW0hFk6fxNQxw7l99z5T5izIyKoKAzt87AQ/LV1Ox7at+WnODFxdnBk6ahwqtVpv/MtXA5k4bSZ1a9Xgp7kzqFShPKMnTuHW7TvaOIXt7ejzzdcsXjCb2dMmUahgAYaMHIs6/GX7PfbHKabOmEOdmtVZPG8mc6ZPprp/1YyurshCDpz8H3NWbearFg1ZOXUEHk4O9J84h7DwCL3xY+PisS+Yn97tmpHXSv9WmAVsrOndrjkrpgxnxeTh+BXz4vtpC7l572FGVkVkQdu2bCRg13Z69unH9FnzMTE1ZfTIocTHx6ea5vjRw/yy5CfatOvIrHk/4ezqyuiRQ1GrVQCEhYYSFhpKl696MG/RUvoN+J6zp//HvNkztHncunmDsaOGU9qvLLPm/cT3Q0fw91+nWLl8aYbXWYiMJB2MbGDzzgAa1K5BvZqf4+zowMBe3TE1MWbvwd/1xt+6ew/lSvvSpnkTnBwK07VDGzxcXdm+5zedeEZGRthYW2uvPLlf/uJy6n9nyJkjJ/2++QrHwvZ4e7gzsNfXHDv5Fw8eyn792dXWHbuoX6cWdWvVwMnRgf69v8HExITfDhzSG3/brgDK+pWi9RfNcHJwoEvHdri7ubIz4FdtnBrVquLnWxK7QoVwdnLkm6+6EB0dzc1bKZ2QpKQkFi7+he5dO9Oofl0K29vj5OhAtSqVMqXOImtYH3CAJjUq0/DzSrgUtmPI1+0xNTYm4PAfeuMXcXemb8cW1KpUDiMjI71xqpQpyWeli+NoWxBHu4L0bNsMM1MTLl2/mZFVEVmMRqNh145ttGrTngoVK+Hi4sqAQUMICw3lz1P62xfAzu1bqV23PjVr18XR0YleffpjYmLCwf0p36VOzi4MGzGGcuUrYmtrR0nfUnTo3JW///qTpKQkAI4fO4Kziwtt2nXEzs6eYsVL8mXXr/k1YCfR0dGZUn8hMkKW62DcunWLAwcOcOnSJUMX5aOQkJDAteCb+PmW0IYplUpKlyzB5UD9J6FfCbyGX8kSOmFlS5d8I/75S5dp1rEbnXp+y6yFiwmPeP7yuYmJ5DTKiVL5sgmZGBsD8M/VwA+ul8h6UtraDUr7ltSGKZVKSvuW4EpgkN40VwKDdOIDlC3ty5VU2mZCQgJ7ftuPubkZbi7OAFwPvsGz0FAUCgU9vh1Iq45dGTZ6nM4oiMjeEhITCbp5l7LFXx7KqlQqKVvch3+upU9nICk5mQN//E1MXDzFPWWb1k/J45BHqFRhlPR9OU3Y3Dw3nl4+BF29ojdNQkICwcHX8H0ljVKppKRvaQID9acBiI6KwszMTDv9KSEhAeMX353/MjY2Jj4+nhvB+t+TQnwMDNrB6NWrF5GRkQDExMTQokUL3N3dqVOnDiVLlqR69f+3d9/RURVtAId/u5AC6aGkkZDeaKETEaJIFelNRLqINEGU3ntTelFBepVOkN5BsNB7QugloSS7SUhPdr8/ggsrCXxIkk3C+5xzj+7cmdl34ri7c+fO3Fq68yJj0TGxaDQabKyt9NJtrK2IyuS2lSi1OoP81qhUz/NXqVCeIf1688O4kXzZ8XPOXbrM4DETdFddypctTZRKzdpNW0lJSSH26VN+Xr4KgMiojN9X5G2Z9zX9vvMilUqNjbW1Xpq1tTVRz24h+Mcff/3NJy3b8nHzNmzcEsyUcaOxskq/rSU84iEAy1evo12bVowfNQxzc3O+HTqCmNhYRP6njnlKmkaD7b9udbKxtiAyk1tB/19hd+7xYfs+1PysJ1MWrmLKdz1wK+H4VnWKvEWlSv88srax0Uu3trZGpYrKsExMTDQajSaDMjaoo1QZl4mOZt2aldRr0FCXVqFiJa5euczhQwdIS0sj8skT1q5eCUBUVMbvLUReYNABxk8//aSbAhw3bhx//vkn+/bt4+nTpxw5coQ7d+4wYcKEV9aRlJRETEyM3pH0insmxf+nVs3qVK9aGXfXkrxfrQoTRwzh6rXrnL2YfmXGzcWZwf168euWYOq3+pwWHbrhYFccG2srlEpZGCXeTLmyZfhpdvraisoVyzN+yve6dR2aZ88C/axNS2pWD8Tb04MB/fqgQMGRY8cNGLXID0o62rN82gh+mTiE5nWDGDtvCTfvyRqM/OzQwf20bv6J7khLS83294yPj2PsqGE4u5SkbbsOuvTyFSrRqcuXLJg7kxZNGvBVt05UqlwFAKUsMhZ5mEF3kXrxIeLBwcFMnTqVDz/8EIDq1aszffp0BgwYwKRJkzKtY9KkSYwZM0YvrX+vr/i2T4/sCTqXsbK0QKlUvrSgW6WOxvZfV47/YWttnUF+NTY2GecHcLS3w8rSgvvhEVQsl77DSu2gGtQOqkGUSk0hUxNQKFi/dTsOdnZv1SaRO2Xe1zLvOzY21i8tAFer1dha61/1K2RqipOjA06ODvj7+tCxW0927tnPZ61bUMQ2PW9J5+e7uRgbGeFgb8ejx/J063eBtaU5BZRKotT6C7pV6liK/GtG7U0ZFSyIs31xAHzdS3L5+i3W7djP4C/bv1W9IveqUjVQt9MTQGpKCgBqlQpb2yK6dLVajbu7R4Z1WFpaoVQqUav0ZyvUahXWtvqfb/Hx8YweMYRChQsxdMQYChbU/+nVtHlLmjRrQVRUJObmFjx6GMHypb9g7+DwVu0Ub0eb+1YR5CkG/+spno3QIyIiKFtWf11AuXLluHv37ivLDxkyhOjoaL2jd/eu2RZvbmNkZIS3pzunz13QpWk0Gk6fv0Ap34y3DvX39eb0+Qt6aafOns80P8DjJ5HExD6lSAY/JG1trClUqBAHjx7H2MiYSgFlX65A5Hnpfc2D0+fO69I0Gg1nzl3A39cnwzL+vj6cOXteL+3UmXP4v6KvAWi0GlKefel7eXpgZGTEvfvPryqnpqYS8egRxYsX/6/NEXmIUcGC+Li78PfF5+u7NBoNf1+8kuXrJbQaLckp2X9FWxhO4cKFcXR00h3OLiWxsbHl3Lkzujzx8XGEhlzBx88/wzqMjIzw9PTm3LnTujSNRsP5s2fw9X1eJj4+jlHDB1GwYEGGjxz30nqLfygUCooUKYqJiQlHDh+kaLFiuHtkvtW8ELmdwZ+DMWLECAoXLoxSqeTBgweUKlVKdy4yMhIzM7NXljcxMcHExEQv7Wkm/wPnV62afMLkmfPw9vTAz9uTDdt+IzExifofpc8GTZwxh2K2tnTr2A6AFo0a0m/oKH7dHEy1yhU4cOR3QsKu822v7kD6ephla9dTM7AatjbW3I94yE9LV+DkYE/lCgG69928fSel/HwoZGrKybPn+WnJCrp1bIe5+av/m4m8q0XTxkydMRsfLw98vL3YtHU7iYmJ1K/9EQCTf5hF0SK2fNEp/epv88af0H/wcNZv2krVyhU5eOQYoWHX+aZ3+gxjQmIiq9dtILBqZYrY2hAdE8vW7Tt4EhlF0PvvAWBWuDCNGtRj2aq1FCtaFLvixfh10xYAXR6R/7X9pA7j5i3Bz70k/p5urNuxj8SkZBp+kL6b2Ji5iylma03Pz5oD6QvDb95L39EuNTWVx1FqQm/dpZCpiW7GYv7qTQQGlMauqC3xiYnsOfYXpy+HMnNYX8M0UhiEQqGgcdPm/Lp2FY6OTtjZ2bNqxVJsixShWuDz3eqGDxlAtfeq80mjpgA0adaCmdOn4unlg7e3D9u2biIxKZGP6tQH0gcXI4cNIikpif4DhhAfH6+7LdzSykq30HvThnVUqFgZpVLJ8d+PsXH9WgYOHiHPwRB5mkEHGDVr1iQkJH33GX9/f27f1t8VZseOHXoDDpGxWjWqEx0dw9LV69IffubuypTRw7B9Ntvw6PETvXs5S/v5MPzbvixetYZFK1bj5OjAuKEDcSvpAqTvhHH91h12HzjM07g4itjaUimgLF3afYrxC9s9XrkWxtI1v5KQkIhzCSf69/qSuh8G5WjbRc76sOb76X1t5VpUKhUe7m5MGjtSd4vUo8eP9dbglPLzZeiAb1iyYjWLl6/EydGBMcMG4+ZaEoACSiV3791jz/6DxMTEYGlpgbeXJzOmTMD1WX8E+LJLRwoUKMDk6TNJTkrG18eb7yeM1ds6WeRvdd6rjDomloW/biNSHYOXawlmDP1a94yLiCdRuhlxgMdRajoMHKd7vSp4D6uC91De35sFo78DQBUdy5h5S4hURWNeuBAeJZ2YOawvVctmfNVa5F/NW7YhMTGReXNmEPf0Kf6lSjN67GS9GYeI8AfEvPB8nhpBHxIdE83qFUtRqVS4u3sweuwkbJ4t/L4edo3QkPRZt+5dO+i938IlK7Gzswfg1Mm/Wb9uNSkpKbi6uTNsxFgqPluHIURepdC+uBAil7lx4wbGxsaUKPHqJ2n+24OQ86/PJEQWSFMafBJQvCPM42W9icgZj8xlm16RM3w8nA0dQqYeXT5p6BAyVdy/kqFDeK1c/evI3V0+5IQQQgghhMhLDL7IOyEhgWPHjnH58ssPpklMTGT58uUGiEoIIYQQQryrtApFrj3yAoMOMEJDQ/Hz86NmzZqUKVOGoKAgwsPDdeejo6Pp3LmzASMUQgghhBBCvAmDDjAGDRpE6dKlefToESEhIVhYWFC9enXu3LljyLCEEEIIIYQQ/5FB12AcP36cffv2UbRoUYoWLUpwcDA9e/akRo0aHDx48LVb1AohhBBCCJHVtAqDryLI0wz610tISNB7oqVCoWDBggU0atSIoKAgQkNDDRidEEIIIYQQ4k0ZdAbD19eXkydP4ufnp5c+d+5cABo3bvyf6o0raPXWsQnx/3iSbGPoEMQ7wsbc0tAhiHdEmlYe8CaEeDsGncFo1qwZa9asyfDc3Llzadu2Lbn4MR1CCCGEEEKIf8nVD9r7r65dv/36TEJkAZnBEDnFxjjG0CGId4TMYIicUsrTwdAhZCri6hlDh5Ape9/yhg7htWQFixBCCCGEECLL5OoneQshhBBCCJHTtOSNB9rlVjKDIYQQQgghhMgyMsAQQgghhBBCZBm5RSqf0Gq1rFq5nN27dhIX9xQ//1L07PU1Tk5Oryy3PXgbmzauR6WKws3Nne49euHj46s7v2vnbxw6dJDrYWEkJMSz9tdNmJub69URFnaNpYsXce1aKEqlkveqv88X3b6iUKFC2dJWkftotVo2r/mJw3u3EB/3FC/fsnT4ajD2ji6Zlgm5dJodm1dw+/pV1Kon9Bk8jYrVPtDLs3DWaH4/+JteWuny1fhu1JzsaIbIZbRaLatXLmXvrh3ExT3F1780PXr1xdGpxCvL/Ra8hS0bf0WlisLVzYMve/TB+4XPtflzpnPuzGmioiIxNS2Er38pOnbuRgnn5/21yccfvVTvt4OGUTOoVtY1UORqWq2WtSuXsHf3duLjnuLrV5ove/V/bf/buX0zWzauRa2KwtXNky+++hovH/3t+EOuXGLV8kVcC7mCUqnEzd2TEeOmYWJikp1NEiLHyAxGPrFxw68Eb9tCr95f88OM2ZiamjJyxBCSk5MzLXPk8CEWLfyJtp99zqw583Fzd2fkiKGo1SpdnqSkJCpWrETrNp9mWEdkZCTDhw7GwdGJH2bMZsy4idy5fZsZ06dleRtF7rVj83L2bl9Hx6+GMHLqEkxMC/HDmD4kJydlWiYpMQEXN2/adx/4yrrLVAhk5pKduqPHtxOyOnyRS23asJbftm2mR+9+TJsxF1NTU0aPGPzKz7Wjhw+yeOGPtPmsA9Pn/IibuwejRwzS+1zz8PTm628GMvenJYwePxmtVsuo4YNIS0vTq+vrbwawdOV63VEt8P1sa6vIfTZvWMNvwRv5qld/Jk9fgIlpIcaNGPDKz7VjRw6wZOF8Wn/Wie9nL8TVzYOxIwbo9b+QK5cYN3IgAeUrMWXGAqbO/JEGjZqhVMo9/yL/kAFGPqDVatm6ZTNtPv2MaoHv4ebmTv9vBxIVGcmJE79nWm7L5o3Uq9+AOnXr4eJSkl69+2JiYsLePbt1eZo0bU6r1p/i4+uXYR1///UHBQsWoEfP3pQo4Yy3tw+9evfl+O/HePDgfpa3VeQ+Wq2WPcFraNy6CxWqBuHs6kW3vmNQRT3h9J+HMy1XtmJ1WrTrQcVqH76y/oIFjbG2Kao7zOSBc+8ErVZL8JZNtPr0c6oGVsfVzYN+3w4iKvIJf5w4lmm5rZs3ULf+x9SuWx8XF1d69O6HiYkJ+/bs0uWp1+ATSpUpi52dPR6e3nzeoTNPHj/i0aOHenWZmZljY2urO4yNjbOtvSJ30Wq1bN+6gZZt2lMl8H1c3Tz4+tshREU94a9X9L/gzeupU78hH9VpgLOLK91798fE1JQDe3bo8ixeOJePGzeneet2uJR0w6mEC9VrfIiRkfQvkX8YdICRlJRESkqK7vX169cZNmwY7du3Z/jw4dy8edOA0eUdDyMiUKmiCAiooEszMzPDx8eXq1euZFgmJSWFsLBrBAQ830tZqVQSEFCeq1czLpNZPQULFkSpfN6VjE3SPyQvX7r0pk0RedDjh/eJVkXiX7aKLq2wmTke3qW4HnL+reu/evEUfTrWZXDPFiz7cTJPY9RvXafI/R5GhKNSRVFO73PNHG8fP0KuXM6wTEpKCtfDQvXKKJVKygVUIORqxmUSExPYt3c3dvYOFC1aTO/cTwtm8/mnzfiuX0/27dkpD359hzyMCEetiqJcQEVdmpmZOV4+/pn2pfT+F0LZF8oolUrKBlTUlVGrVVwLuYKVlQ1Dvu1F53bNGD6oL1cuvf1npchaWoUy1x55gUGjrFevHlu3bgXg999/p1SpUmzfvp2UlBR27NhB6dKlOXHixCvrSEpKIiYmRu9ITsp8+jI/UqmiALC2sdZLt7a2Qa1SZVACYmJi0Gg0WNvYvFRGFRX1f7932XIBqFQqNm74lZSUFJ7GxrJ0yS8AREVFvkErRF4VrU7/72xlXUQv3dKqCNGqt+sDZSq8x5f9RjNwzHxadehDyMXT/DCuL5p/3coi8h/Vs8+uDD+jMv1ci/6/P9d2bN9Km+YNadP8E06f/IsxE6ZiZGSkO//Z550YOHgkYydMJbB6DX6cN4vt2zZnRdNEHqB+9r1qZWOrl57e/zL+joz9p/9Zv1zmn/oeRjwAYN3qpdSu/wkjxk7F3cOLUUO/5cH9e1ndDCEMxqCLvM+cOUO5cuUAGDZsGD179mT69Om68yNGjGDAgAEcO5b5dOSkSZMYM2aMXlrvPn35uu832RN0LnDw4H7mzZmlez1qzHiDxVKypCvf9B/AokU/sWzpYpTKAjRu0gRrGxu9WQ2Rfxw/vJNlCybpXn8zfEa2vVe1GnV1/+7s6omzqycDv2rG1Yun8C9X5RUlRV5z6OA+Fsx53pdGjJmYre8X9OFHBJSviCoqis2bfmXapLFM/n627jaoNp+11+V19/AiMTGRzRt/pVGT5tkalzCMwwf38tPcH3Svh42enC3vo9Wkz4LVbdCIj+o0ANL714Vzpzmwdwefd/oyW95XiJxm0AFGWlqablHd1atXmTVrlt75Tp06MXPmzFfWMWTIEPr376+XdvdeRJbGmdtUrRqot9PTP7eZqVVqbG2fX0VWq1W4uXtkWIelpSVKpfKlGQ61WoWNrW2GZTLzwYe1+ODDWqhUKkxNTVEoYMvmTdjbO7xRPSJvKF+lJh7epXWvU1PSF9xGqyOxti2qS4+JjsTFzTtL37u4fQksLK15GHFPBhj5TJWq7+Hzwk47zz/XVG/wuWb1f3+umZmZY2ZmjqNTCbx9/WjXuil/HD9GzQ8y3iXKx8ePX9esJCUlWe6Vz4eqVK2Odwb9L1oVlUH/88ywDot/+p9af4ZDrVZh/WwmxOZZXc7OJfXyODmX5PHjR2/fECFyCYNeYq5atSrBwcEAeHh4cO7cOb3zZ8+exfY1P3ZNTEywtLTUO4zz+TZvhQsXxtHRSXe4uJTExsaWs+fO6PLEx8cREnIVX7+MF2cbGRnh6enFuXNndWkajYZzZ8/im8mC7texsbGhUKFCHDlyGCMjIwLKV3h9IZHnFCpkhp2Ds+5wdHbHyqYIl8//rcuTEP+U66GX8PApm6XvHfXkIU9jo7G2KfL6zCJPKVy4MA6OTrrD+dnn2vlzp3V54uPjCA25go+ff4Z1GBkZ4eHpzfkXPgs1Gg3nz57BxzfjMum0aNGSkpL57lQ3blzH3NxCBhf5VKHChXFwLKE7nF1csc6g/10LuZxpX0rvfz6cP/u8THr/O6UrU9zOHtsiRbl//65e2fD7dylW3C4bWiaEYRh0BmP8+PE0aNCAuLg42rZty7fffsu1a9fw8/MjJCSE2bNnM2TIEEOGmCcoFAqaNG3GurWrcXJ0ws7OnpUrlmJbpAiBgdV1+YYOGUjge9Vp1KgJAE2btWDG9Gl4eXnh7e3L1q2bSExKpHaderoyqqgoVCoV4Q/S7xu9desmhQsVpljxYlhYpO/mExy8FT8/fwqZFuLMmdMsWbyQjp26vPS8DJE/KRQK6jZqS/D6xdg7OlO0uBObVv+IjW1RKlQN0uWbMiJ9x6jaDVsDkJgQz8Pw51+yTx494PaNEMwtrChSzJ7EhHi2rFtIpcBaWFkX4XHEPdYtm0NxB2dKlw/M8XaKnKVQKGjUtDm/rl2Fg2MJ7OzsWb1iCbZFiuptFztiyHdUe+99GjZqCkCTZi2ZNX0Knl7eeHn7Erx1o97nWkT4A44dOURAhUpYWVnx5MkTNq5fg4mxMRUrVwXgrz+Po1ap8PH1x9jYmLNnTrFh3WqatmiV438HYRgKhYJPmrRkw9oV6f3P3oE1K37B1rYoVV7of6OG9qdq4Pt83Cj91rlGzVoxZ/okPL188PL2I3jrBpISE6n17HYohUJBk+ZtWLdqKa5uHri5e3Jw/27u37vDgKFjMoxFiLzIoAOMwMBAdu7cSf/+/fnzzz8BmDAhfY97R0dHRo8eTd++fQ0ZYp7RomVrEhMTmTNnJnFPn+JfqjRjx07U21YxIjycmOho3euaQR8QHRPNyhXLUalUuLu7M3bsBGxeWCC5Y8d21qxeqXs9eOC3APT75jtq10m/Pz40JITVK5eTkJBICWdnevXuS62Pamd3k0Uu8nGzDiQlJrBk/kTi457i7VeOb0fOxtj4+Wzio4j7xL6wA9TNsCtMGfGV7vWaxen331f/sCHd+o5GqVRy71YYvx/8jfi4WKxtilE6oCrN230lV5HfEc1bfkpiYiLz50wn7ulT/EqVYdTYSf/6XHug97lWI+hDYmKiWb1iKSpV+u1Uo8ZO1t2iYmRszOVLF9i2dSNxT59iZW1DqdJlmfzDHKyt0z/7ChYoyI7t2/hl4QLQanFwdKJLt6+oW79hzv4BhEE1a9mWpMREfpzz/bMH2JZhxLipep9rEeH3iYl53v/er1mLmGg1a1YuQa2KSn+A3tipuv4H0KhpK1KSk1mycB5PY2NxdfNg1PjvsXd49YNxRc7SKuS5JG9Doc0l++49fvyYGzduoNFocHBwwNXV9T/Xde367awLTIhXeJJs8/pMQmQBG+MYQ4cg3hFp2gKGDkG8I0p55t61mvdDLxg6hEw5eZcxdAivZdAZjBcVK1aMYsWKvT6jEEIIIYQQItcy+D6iCQkJHDt2jMuXX35wTWJiIsuXLzdAVEIIIYQQ4l2lRZFrj7zAoAOM0NBQ/Pz8qFmzJmXKlCEoKIjw8HDd+ejoaDp37mzACIUQQgghhBBvwqADjEGDBlG6dGkePXpESEgIFhYWVK9enTt37hgyLCGEEEIIIcR/ZNABxvHjx5k0aRJFixbF09OT4OBg6tWrR40aNbhx44YhQxNCCCGEEEL8BwYdYCQkJFCw4PN15gqFggULFtCoUSOCgoIIDQ01YHRCCCGEEEKIN2XQXaR8fX05efIkfv962vTcuXMBaNy48X+qNw3ZYk8Ikb8kakxen0mILJBXFpEKIXIvg85gNGvWjDVr1mR4bu7cubRt25Zc8pgOIYQQQgjxjtAqlLn2yAtyzYP2stLV6/cMHYJ4R6iSLQ0dgnhHFCqYZOgQxDtCZjBETinvVdTQIWTq7rWXH5+QWzh7+Rs6hNfKG8MgIYQQQgghRJ4gAwwhhBBCCCFElpEBhhBCCCGEECLLGHQXKZF1tFotq1cuZe+uHcTFPcXXvzQ9evXF0anEK8v9FryFLRt/RaWKwtXNgy979MHbx1d3fv6c6Zw7c5qoqEhMTQvh61+Kjp27UcLZBYCYmGimT5vErZs3iI2JwcramqrV3qN9p64ULmyWrW0WuYdWq2Xzmp84vHcL8XFP8fItS4evBmPv6JJpmZBLp9mxeQW3r19FrXpCn8HTqFjtA708m9f8zJ/H9hD15CEFCxrh6uFLi8974uFdOptbJHIDrVbL+lW/sH93MHFxsfj4leGLnt/h4OT8ynK7t28keNMa1KooSrp50Ln7N3j6PL9nWa2KZOXi+Zw/8zeJCfE4lHCheesOVK3+gV49p/8+zsY1S7h96zrGRsb4lSnPgOGTsqOpwsDS+9oiDuj6Wlm6/t99bTXRqihc3Dwz6WvzuPBCX2vWugNVq3+oy/Pg/h1WLZ5H6JULpKak4OLmSevPv6BU2YrZ1l4hspvMYOQTmzas5bdtm+nRux/TZszF1NSU0SMGk5ycnGmZo4cPsnjhj7T5rAPT5/yIm7sHo0cMQq1W6fJ4eHrz9TcDmfvTEkaPn4xWq2XU8EGkpaUBoFQoqVrtPYaNHMeChcvo238g586eZsGcmdndZJGL7Ni8nL3b19HxqyGMnLoEE9NC/DCmD8nJmS9MTkpMwMXNm/bdB2aax97RhfZfDmD8rDUMm7SQosUd+X50b2KiVZmWEfnHto2r2Bm8gS96fceEH37G1LQQE0f2f2W/On5kP8sXzaVF285MnvULJd08mTiyP9EvfK7Nmz6eB/fuMHDEZKbNW0aVwJrMmDKSm9efP3vpz98PMfeHcXxQuyFT5yxl7LQFvB9UO1vbKwxn28ZV7ArewBe9BjD+h4WYmJoy6bV9bR8rFs2hZdsuTJq1mJJunkx6qa+NI/zeHQaMmMLUecupEhjEzH/1taljBqJJS2P4hNlMnJlez9QxA1GrIrO1zeLVtChy7ZEXyAAjH9BqtQRv2USrTz+namB1XN086PftIKIin/DHiWOZltu6eQN1639M7br1cXFxpUfvfpiYmLBvzy5dnnoNPqFUmbLY2dnj4enN5x068+TxIx49egiAuYUFDRo2xsvbh+J2dpQLqECDho25dOlCtrdb5A5arZY9wWto3LoLFaoG4ezqRbe+Y1BFPeH0n4czLVe2YnVatOtBxWofZponMKg+pcpVpbh9CZxcPGjbpR8J8XHcu3UtO5oichGtVsuOretp3qYDlavVoKSbJ736D0cVFcnfJ45mWu63LWv5qF4jPqzTkBIubnzRawDGJqYc3LtdlyfkykXqN2qBp48/dvZOtPi0E2Zm5twICwEgLS2VpT/P4vMuvajzcVMcnVwo4eJGYI2Psr3dIudptVp2bv2VZm06UknX10aginrCyVf2tXXUqteID/T6mgmHXuhroVcuUq9RS11fa/6sr90MuwpATLSaiAd3adzyc0q6eeLg5Ezbjl+RlJTI3ds3sr3tQmQXgw4wNm7cSHx8vCFDyBceRoSjUkVRLqCCLs3MzBxvHz9CrmS8zVpKSgrXw0L1yiiVSsoFVCDkasZlEhMT2Ld3N3b2DhQtWizDPJGRT/jj+DFKlyn7Fi0Secnjh/eJVkXiX7aKLq2wmTke3qW4HnI+y94nNSWFQ3s2U6iwOc5u3llWr8idHj18gFoVSZmAyrq0wmbmePr4c+3qxQzLpKakcCMslDIBlXRpSqWSMgGVuHb1ki7Nx680J44e4GlsDBqNht8P7yMlOZlSZcoDcDMslKjIxygUCgZ93Znu7ZswadS33LklP/jyo+d97Xm/+aevhb6ir90MC9Hrn//0tRfLePuV5sTR/bq+dvxZX/Mvk/7da2FphWMJF44e2EViYgJpaans27UVK2sb3Dx9sqnFQmQ/g67BaNWqFRYWFrRp04auXbtStWrVN64jKSmJpCT9KczkpCSMTd6dp96qVOnTsdY2Nnrp1tY2unP/FhMTjUajybDMvbt39dJ2bN/KssU/k5iYiFMJZ8ZMmIqRkZFenu+njOfPP46TnJRE5aqB9O773ds2S+QR0er0aXwr6yJ66ZZWRYjOgin+s38fZcEPw0hOSsTKpigDxszFwtL6resVuZtaFQWAlbX+Z5SVtQ1qdVSGZdI/19Kwsrb9VxlbHty7rXvdb9BYZk4ZRde2H1OgQAGMTUz5dthE7B3T16w9jHgAwIbVi+nwRR+K2dmzffNaxg7tw8yf1mBuIc+/yU+e97WX+41anfFnWEyMOtO+dv/eHd3rfoPGMWvKSL5o20DX1/q/0NcUCgXDxs/ih/GD6dyqDgqFEitrawaPmY65ufQzQ8orD7TLrQz+1/vuu+84efIkgYGBlC5dmpkzZxIZ+f//KJk0aRJWVlZ6x88/zsvGiA3v0MF9tGneUHekpaVm6/sFffgRM+b8xMQpM3B0KsG0SWNfWtvRtVtPZsz+kaEjxxER/oDFCxdka0zCcI4f3kn3T2vqjrTU7O1/fmUqMXbGKoZN/oUy5QOZP20oMZn8wBR519GDe+jQso7uyM5+tW7lIuLjYhk+fiYTZyyiYdM2zJwykju3rgOg1WoAaNYmfeG3u6cvPfoNBRScOHYg2+ISOePYwd10bFlbd2RnX/t15ULi4p4ybPwsJs74hYZNP2WWXl/TsnjBD1ha2TB6ynwmTF9IpWo1mTZ2IKqoJ9kWlxDZzeC7SHXv3p0RI0Zw6tQpfvnlF8aMGcPgwYNp3Lgx3bp1o06dOq8sP2TIEPr376+Xduve4+wM2eCqVH0PHx8/3euUlBQA1CoVtrbPryKr1Src3D0yrMPS0gqlUon6XzMcarUKG1v9KzJmZuaYmZnj6FQCb18/2rVuyh/Hj1Hzg1q6PDa2ttjY2lLC2QULCwuGDOhH67af68Uj8ofyVWrq7eKUmpI+2IxWR2Jt+/yprDHRkbhkwa1MJqaFsHNwxs7BGU+fMgzq0Zwj+7byScvOb123yD0qVX0frxd230nR9SsVNi/0q2i1Clc3zwzrSP9cK0D0vwag0eoorG3SP4siwu+ze/tGvp+3HOeS7gC4untx9dI5dm/fRLfeA3T9uISzq64OIyNj7OwdiHz88O0bKwyqYtX38fQppXv9vK9F/auvRVHSzSvDOiwtrV/R19K/QyPC77F7+0amzVuh62sln/W1Pds38kXvgVw8d4rTfx/nl7W7dDsvdvX04cKZvzmyfydNWrXPuoYLkYMMPoPxj4oVKzJ//nzCw8NZuHAhjx8/pn79+ri5ub2ynImJCZaWlnpHfr89qnDhwjg4OukOZ5eS2NjYcv7caV2e+Pg4QkOu4OOX8ePkjYyM8PD05vy5M7o0jUbD+bNn8PF91SPotWjR6j6QM8yh0QLPBz4ifylUyEz3g9/OwRlHZ3esbIpw+fzfujwJ8U+5HnoJD5+sX4uj0Wikb+VDhQoXxt6xhO4o4eKGtU0RLpw9qcsTHx9HWMhlvHwz3qa4oJER7p7eXDh3Spem0Wi4eO4UXr7pPyiTkxIBUCj1v/6UygK6mQt3Tx+MjIx5cP/57aKpqak8fhRB0eL2WdNgYTCFCptl2Ncunn3eb/7pa96v6Gtunj5cPPe8f/7T1/4pk/zs9m3lS31NiUarfZYnvT8qFfo7AymUCjTP+qMQeZFBZzAUipe32jI1NaV9+/a0b9+esLAwlixZYoDI8haFQkGjps35de0qHBxLYGdnz+oVS7AtUpRqge/r8o0Y8h3V3nufho2aAtCkWUtmTZ+Cp5c3Xt6+BG/dSGJSIrXr1AMgIvwBx44cIqBCJaysrHjy5Akb16/BxNiYipXT18uc/PtP1CoVXt4+mBYqxN3bt1jyy0/4+ZfGzk6+iN8FCoWCuo3aErx+MfaOzhQt7sSm1T9iY1uUClWDdPmmjEjfMap2w9YAJCbE8zD8+Q+4J48ecPtGCOYWVhQpZk9SYgLB6xcTUKUm1jZFeRqjZv/O9aiiHlOluuzmk98pFAo+btKKzeuW4eDkTHE7B9atXISNbREqB9bQ5Rs3tC+VA2tSv1ELABo2/ZT5Mybg4eWLh7cfO7b+SlJiAh/UbgiAY4mS2DuUYOHcabTv0gtzSyv+PnGEC2f/ZtDIqQAULmxG7QZNWL/qF4oULU6x4vZs27QagGrvZ77rmcibFAoFDZq0ZvO6Zdg7laC4nSO/rlyIjW1RKun1ta+f9bWWADRs2oYFMybg7uWLp7f/s76WSNBLfW0qn3fpjbmlJSdPHOXC2b8Z+KyvefmWxtzcgvkzxtPi084YmZhwYPc2Hj0Mp0Kl93L+jyFEFjHoAEP7bASfGU9PTyZMmJBD0eRtzVt+SmJiIvPnTCfu6VP8SpVh1NhJGBsb6/JEhD8gJjpa97pG0IfExESzesVSVKr026lGjZ2sm941Mjbm8qULbNu6kbinT7GytqFU6bJM/mEO1s8WXhobG7Nn928sXjiflJQUihYtRrXqNWjRqm3O/gGEQX3crANJiQksmT+R+LinePuV49uRszE2fj6b+CjiPrExat3rm2FXmDLiK93rNYtnAFD9w4Z06zsahVJJ+P1bHJvyG09j1JhbWOHm5c/QiT/j5JLxrX8if2ncoh1JiYn8PGcq8XFP8fEvw5CxP+j1q4f/6lfv1fyImGg1v65chFoVhau7J0PG/qD7XCtYsCCDR09j9bIfmTpuEIkJCdg5ONHzm2GUrxyoq+fzLr0oUKAA86aPIzkpCU8ff0ZMmCULb/Op9L6WwEJdXyvL4Az72vPv0Pdq1iYmWs36Z32tpLsXg//V1waN/p41yxYwbdzAZ32tBD2+GU75yumDB0srawaP+YF1y39m3LCvSUtNpYSLG98Nn0xJ94xvzxIiL1BoX/crPxvdvn0bFxeXDGcy3sbV6/eytD4hMqNKlh8bImcUKpj5A7+EyEp55UFeIu8r71X09ZkM5FZY6OszGYirZ+7fqt2gMxglS5Y05NsLIYQQQgghspjBF3knJCRw7NgxLl9++eFuiYmJLF++3ABRCSGEEEIIIf4Lgw4wQkND8fPzo2bNmpQpU4agoCDCw8N156Ojo+ncWbaiFEIIIYQQIq8w6ABj0KBBlC5dmkePHhESEoKFhQXVq1fnzp07ry8shBBCCCGEyHUMOsA4fvw4kyZNomjRonh6ehIcHEy9evWoUaMGN27cMGRoQgghhBBCiP/AoAOMhIQEChZ8vs5coVCwYMECGjVqRFBQEKGhuXcFvxBCCCGEyJ+0CmWuPfICg+4i5evry8mTJ/Hz89NLnzt3LgCNGzf+T/W6Xt/71rEJ8f8obu9p6BDEOyJRaWXoEMQ7IsHI3NAhiHdG7t2mVrwdgw6DmjVrxpo1azI8N3fuXNq2bfvah/EJIYQQQgghcg+DPmgvuyTuWWLoEMQ7Il5mMEQOSTSRGQyRM2QGQ+QUD3d3Q4eQqZvXwwwdQqbcPHL/b4+8cSOXEEIIIYQQIk+QAYYQQgghhBAiyxh0kbcQQgghhBC5jRaFoUPI02SAkU+tPXKKZfv/5ElMHN5OxRncsg5lXB0zzLvx97ME/3WRsPDHAPg729OnUZBe/vikZGZuPcTBC9eIjkvAqYgVbYMq0fr98jnSHpF7bdh1gFXbdhOljsazpDP9u7SllFfG99XeuHufheu2cvXGbSIeR9K3Uxs+bVhHL8+m3QfZtOcQ4Y8jAXAv4UiXVo0ILF8m29sicpfNv+1i3eZtRKnUeLiV5Osvu+Dn7ZVp/kPHTrB41VoiHj2mhKM9X3b8nGqVKujluX33Hj8vW8m5i5dJS9NQ0rkEY4Z8i12xYgAE79rL/iPHuHb9JvEJCQSvXoq5uVm2tlNkr+DgYDZu2IBKpcLN3Z0ePXrg4+OTaf6jR4+yYvlyHj58iKOTE106d6ZylSq681qtlpUrVrBr1y7i4uLw9/enV+/eODk56fLExsayYP58/vzzT5RKJdWrV6f7V19RqFAhAO7du8fcOXO4c+cOcXFxFClShKAPPqBdu3a67ft///131q1bR/iDB6SmpuLk5ESz5s356KOPsukvJUTWkVuk8qFdp67w/eYDdG/wPmsHdsbHqTg95q8jMjYuw/wnw+7QoKI/i77+jBX9O2BnY0mP+et4qI7V5fl+036OX7nBxA6fsHnYF7T7oDKT1+/h0IVrOdUskQvt+/0vZi/7la6tGrF0yki8SjrzzYSZREXHZJg/MSkZx+LF6NmuBUWsM160XKyIDT3btWDplBEsmTyciqV9GThlLjfu3s/Opohc5sDR31nwyzI6ftqKn2dMwcO1JANHTUCljs4w/8UrIYz7fiYf16nFwplTeb9qFUZMnMrN23d0ee6HR/D14BE4OzkxY8IYFs3+nvZtWmBsZKzLk5SUTJUKAbRr1Szb2yiy3+HDh1n488981q4dc+bMwd3NjRHDh6NWqzPMf/nyZaZMnkzdevWYM3cugYGBjBs3jlu3bunybFi/nm3bttG7Tx9mzJyJqakpI4YPJzk5WZdn6tSp3LlzhwkTJzJ69GguXrzI7NmzdecLFChArY8+YvyECfy8cCFfdu/O7l27WLlypS6PhYUFn7Zpww/TpzN//nxq16nDjOnTOXXqVJb/nUT+Nm/ePFxdXTE1NaVq1ar89ddfr8y/fv16fH19MTU1pUyZMuzYseON31MGGPnQioN/0TywHE2rlcXDoSjD29TH1NiILSfOZ5h/UsfGtKlZAd8SdrjZF2H0Zw3QaLX8FXJLl+fszfs0qlqGyl4lcSpiTcvqAXg7Fefi7fAcapXIjdZs30vjj2rwyYfv4+bsyMAvP8fE2JjtB45lmN/f040+HVpRp3oVjIwynkCtUSmA9yqUxdnBDhdHe776rDmFTE24GHojO5sicpn1W7fTsO5HNKj9Ia4uzvTv+SWmJsbs3Hcgw/wbg3+jSoUAPm3ehJLOJejy+ad4ubuz+bddujy/rFxD1Yrl+apze7w83HBysKd61crYvDDYbdmkIZ+1bIa/j3e2t1Fkv82bN1O/QQPq1q2LS8mS9O7TBxMTE/bs2ZNh/q1bt1KxUiVatmyJi4sLHTp0wMPDg+DgYCB99mLLli18+umnBAYG4ubmxrfffUdkZCQnjh8H4M6dO5w6eZKv+/bF19eXUqVL81WPHhw5fJjIyPSZWQcHB+rWrYu7uzt2dnZUq1aNDz78kEsXL+piKVu2LO9Vr46LiwsOjo40bdoUNzc3Ll26lM1/NQGgVShy7fEm1q1bR//+/Rk1ahSnT5+mXLly1KtXj0ePHmWY//jx47Rt25auXbty5swZmjZtStOmTbn4Qt/8fxh8gHHu3DkWL17MjRvpPx4uXbpEz549+eqrr9i9e7eBo8t7UlLTuHI3gmo+rro0pVJBNR9Xzt/6/64AJyankJqmwdKskC4twM2Jwxeu8VAdi1ar5a/Q29x+pCLQ1zXzikS+lpKSSsiN21Qu669LUyqVVC7rl2WDgbQ0DXt//4vEpGTKeHtkSZ0i90tJSSE07AYVA8rq0pRKJRXKleXS1dAMy1y+GkrFcmX10ipXKKfLr9Fo+OPkaUo4OjJg1Hiate9Kj++GcOyPV1/JE3lXSkoKYdeuERAQoEtTKpUEBARw9cqVDMtcvXKF8i/kB6hYsaIuf0REBCqVioDyz28PNjMzw8fHhytXr+rqMDc3x9v7+SC1fPnyKBQKQp7l+bcHDx5w6uRJSpfJ+FZQrVbL2TNnuHfvHqVLl35t24X4x/Tp0+nWrRudO3fG39+fH3/8kcKFC7N48eIM88+aNYv69eszYMAA/Pz8GDduHBUqVNA9BPv/ZdA1GJs2baJ169ZYW1uTlJTE5s2badWqFZUqVaJAgQI0bNiQ5cuX89lnnxkyzDxFFRdPmkZLEUv9e4aLWJhx82Hk/1XHzK2HKGZlrjdIGdyyDmPX7qLuiHkUVCpRKBWM+rQ+FT1dsjJ8kYeoY5+SptFga2Wpl25rZcnt+xFvVXfY7Xt8OWwSySkpFDI1YfKAnrg5Z7yGSOQ/0TGxaDQavZkFABtrK+7cz/hCSZRanUF+a1QqNQDq6GgSEhJZs3ELXT7/lO4d2/HX6bOMnPQ90yeMIqB0qWxpizCcmJiY9H5kY6OXbm1jw9179zIso1KpsM4gv0ql0p0HMqzzxTxWVvp9sUCBAlhYWOjy/OPb/v0JCwsjJSWFBg0a0L59e73zcXFxtP/8c1JSUlAqlfTq1YsKFfTXFYl3T1JSEklJSXppJiYmmJiY6KUlJydz6tQphgwZoktTKpXUrl2bEydOZFj3iRMn6N+/v15avXr12LJlyxvFaNABxoQJExgzZgzDhg1j7dq1tGrViv79+zNixAgAfvjhB6ZNm/bKAUZGf2RtcgomxkbZGnt+9cueE+w6fYVfvv4MkxduYVlz5BTnbz1g1pctcLS14lTYXSau30sxKwuqySyGyGIlHe1ZNm0kcfEJHPjjFOPmLmb+mIEyyBD/mUaT/kzZ96pWolWTTwDwdHfj0tUQgnfulQGGMIjBQ4aQEB/PjZs3+WXRIuw3bqRVq1a684UKFWLuvHkkJCRw7uxZFi5ciL2DA2XLln1FrSK/mzRpEmPGjNFLGzVqFKNHj9ZLe/LkCWlpadjZ2eml29nZcTWT2bSIiIgM80dEvNmFQ4PeIhUSEkK7du0AaNOmDXFxcTRt2lR3vlmzZoSFvfpJipMmTcLKykrvmLbut+wMO1ezMStMAaWCyBj9Bd2RsXEUtXz1TijL9v/Jkn1/8GOvNng7FdelJyanMDv4MN81q8UHZbzwdipO26CK1Kvgy7IDf2ZLO0TuZ21hTgGl8qUF3VHRMZku4P5/GRkVxNnBDl8PV3q2a4GnqzPrdux7qzpF3mFlaYFSqXxpQbdKHY2ttXWGZWytrTPIr8bGxlpXZ4ECBXB1dtbL41KiBA8fP8my2EXuYWlpmd6P/jVroFapsP3XDMQ/bGxsUGeQ/58Zi3/+mVGdL+aJjtbvi2lpacTGxr4081GsWDFcSpbkgw8+oHOXLqxetYq0tDTdeaVSiaOjIx4eHjRv0YLq77/Pr+vW/b9/ApFPDRkyhOjoaL3jxVmK3MCgAwwLCwvdgie1Wk1qaqruNUBkZCTm5uavrCOjP/KANg2zNe7czKhgAfyc7fkz9JYuTaPR8mfobcq6OmVabsm+P/h513Hm92hNKRcHvXOpaRpS0zQo/7WwSKlUotFqszR+kXcYGRXEx70kJy88v5dZo9Fw8sJVSntnvE3tf6XVaElJSc3SOkXuZWRkhLenO6fPXdClaTQaTp+/QCnfjBdf+/t6c/r8Bb20U2fP6/IbGRnh6+XB3X/dYnXvwQPsihfN4haI3MDIyAhPLy/OnT2rS9NoNJw9exZfP78My/j6+XH2hfwAZ86c0eW3t7fHxsZGr874uDhCQkLw8/XV1fH06VOuXXu+y+K5s2fRarX4PMuTEa1GQ2pqKtpXfK9qNRpSUlIyPS/eDSYmJlhaWuod/749CqBo0aIUKFCAhw8f6qU/fPgQe3v7DOu2t7d/o/yZMegAo3bt2vTq1YtVq1bRsWNH6taty5AhQ7h69SohISEMGDCA999//5V1ZPhHfsdvj2r/YRU2HT/Htj8vcCPiCeN/3U1CUjJNq6VPqQ5bHsysbYd0+Rfv/YN5vx1lTLsGOBax4knMU57EPCU+KX3LPfNCJlTydGb61oP8fe02956o2frHebb/dZGPyspOK++ytp/UYdv+I/x26Hdu3XvA1IUrSUxK4pMPqwMwZs4vzF+1UZc/JSWV0Jt3CL15h9TUVB5Hqgm9eYe74c8/zOav2siZy6GEP3pC2O17zF+1kdOXQ6hXo2qOt08YTqsmn7B9z3527T/E7bv3mLFgIYmJSdT/6EMAJs6Yw8Jlq3T5WzRqyF+nz/Lr5mDu3LvP0tW/EhJ2nWYN6+vytGnWmIPHjrN99z7uPwhn8/adHP/rFE0/rqfLE6VSEXbjJvfD028HuHH7DmE3bhIT+3zbbpF3NGvWjF27drFv717u3LnDvLlzSUpKok6d9OfvfP/99yxZskSXv0mTJpw6dYpNGzdy9+5dVq5cybVr12jUqBEACoWCpk2bsnbtWv744w9u3rzJ9z/8QJEiRQh87z0AXFxcqFipErNnzSIkJIRLly4xf8ECagYFUaRIEQAOHjjAkSNHuHPnDuHh4Rw5coSlS5dSs2ZN3XMw1q1bx+nTpwkPD+fOnTts2riRAwcO8GGtWjn5J3xnabWKXHv8v4yNjalYsSL79+/XpWk0Gvbv309gYGCGZQIDA/XyA+zduzfT/JlRaF81VM5mDx8+pH379pw4cYLq1auzbt06hg8fzrx581AoFHh4eLBz5048PN5s95jEPUtenymfW3P42YP2YuPwcSrOoJZ1KPvswXldZ63C0daKce3T70NuMGo+D6Jefm7BVw2q0+PjGgA8iXnKrG2HOXH1JjHxiTjYWNKiegDtP6yM4g23TMtP4u09DR2Cwa3feYDV23YRqY7By1X/QXs9R03FoVhRRvTuAkD4oyc07zX4pTrK+3szf8xAACbMX8rJi1eIVEVjXrgQHiVL0L5JfaqUe7fvkU80ebvbzvKizdt3Pn/Qnrsrfbp1wd8n/UF7/YaOwr54MQb3663Ln/6gvTVEPHyMk6MD3Tu9/KC9HXsPsHrDZh5HRuLs5Eintm14v1pl3fmlq39l2dr1L8UyqG9P3eAmv0swevWdA3lN8LZtbNi4EVVUFO4eHnz11Vf4PptJGDRwIHZ2dvT/9ltd/qNHj7J82TIePnyIk5MTXbp0yfRBe0+fPqVUqVL07NWLEiVK6PLExsYyf/58/vrzTxQKBdWrV+erHj10D9o7fPgwGzds4P79+2i1WooXL86HtWrRrFkzjI3Tn8uybNkyjh45wpMnTzA2NsbZ2ZnGTZoQFBSUE3+2HOHhnrWz3Vkp7PpNQ4eQKU8Pt/8777p16+jYsSM//fQTVapUYebMmfz6669cvXoVOzs7OnTogJOTE5MmTQLSt6kNCgpi8uTJNGzYkLVr1zJx4kROnz79RjuYGXSAkZkbN24QHx+Pr6+vbiT/JmSAIXKKDDBETnkXBxjCMPLbAEPkXjLA+G/eZIABMHfuXKZNm0ZERAQBAQHMnj2bqlXT7wr44IMPcHV1ZenSpbr869evZ/jw4dy6dQsvLy+mTp3Kxx9//EbvmSsHGG9LBhgip8gAQ+QUGWCInCIDDJFTZIDx37zpAMMQDP6gvYSEBI4dO8bly5dfOpeYmMjy5csNEJUQQgghhBDivzDoACM0NBQ/Pz9q1qxJmTJlCAoKIjw8XHc+Ojqazp07GzBCIYQQQgghxJsw6ABj0KBBlC5dmkePHhESEoKFhQXVq1fnzp07hgxLCCGEEEK8w7Qoc+2RFxg0yuPHjzNp0iSKFi2Kp6cnwcHB1KtXjxo1anDjxg1DhiaEEEIIIYT4Dww6wEhISNDbJUqhULBgwQIaNWpEUFAQoaGhBoxOCCGEEEII8abefA/YLOTr68vJkyfx+9cTNefOnQtA48aN/1O9T1wrvz6TEFnAOkpm2kTOKJCaZOgQxDsi2dLU0CEIIfI4g85gNGvWjDVr1mR4bu7cubRt25Z8uIuuEEIIIYQQ+Va+fA7GvdCLhg5BvCNkBkPklDTjwoYOQbwjVJYuhg5BvCNcPb0NHUKmrl2/begQMuXlUdLQIbyWQW+REkIIIYQQIrfRojB0CHla3tjrSgghhBBCCJEnyABDCCGEEEIIkWXkFikhhBBCCCFeILdIvR0ZYOQTW37bya+bthKlUuPh5kqf7l3x9fbKNP/hY8dZsnINEY8eU8LRgW6dPqdqpYoZ5p0x7ye279pDzy8606LJJwCcvXCRb4eOyjD/vB+m4Ovt+faNEnnWr/t+Z/nOw0RGx+Ll7MDAz5tS2iPjhaObDv3Jb7+f4vq9CAD8XJ3o1bJBpvnFu2v9nsOsDN5HZHQMXi5OfNepNaU8XTPMe/3uA37e8BtXb9wh/EkU37RvQduPa+nl+XnDbyzauEMvraSjHet/GJldTRC51Lbtv7Fh4yaiVCrc3dzo+VV3fH0yX4B85Ogxlq1cycOHj3BydKRr505UqVwJgNTUVJYuX8nfJ08SHhGBmZkZ5QPK0bVTR4oUKQJAxMOHrF6zjrPnz6FSqSlia0utDz+gbZvWGBkZ5USThchWMsDIBw4e/Z0fFy2lX6/u+Hp7sWnbdgaNHMfSH+dgY231Uv5LV64yftoMvujYjmqVK3Hg8FFGTpjKjzOn4VZS/0fdsRN/ciUklCK2tnrppXx9WL98kV7akpVrOXPuPD5eHlnfSJFn7PnzLNPXBDO0YwtKe7iwevdRen+/iE1TBmJraf5S/lNXr1OvWgDlPEtibGTEst8O0uv7hayf8B3FbV/uv+LdtPfEKWau2MTgrp9SytOVtTsP8vXkuaz/YRS2VhYv5U9KTsGpeBE+qlqeGSs2ZlqvewkH5g7ro3tdUFkgW+IXudehI0f5eeEi+vTuha+PN5u3bGPYiJH88vOPWFtbv5T/0uUrTJo6jS6dOlK1cmUOHj7MmPETmDdrJq6uJUlKSiLs+nU+a9sGdzc3nj59yoKfFjJq7HjmzpoBwN2799BoNfTt3QtHB0du3b7NzDlzSUxM5MsvuubwX0CIrCdrMPKBDVuC+bheberXroWrizP9enbHxMSEXXv3Z5h/07bfqFyhPG2aN6Wkcwk6f94WLw83tmzfqZfvcWQkc35axNBv+1KwoP6XrpGREbY2NrrD0sKC43/+Rb3atVAoZFrxXbZy1xGaBVWlcc3KuDvZMbRTc0yNjdh65K8M80/46jNaf/QePiWdcHMszoiurdBqtPx1+VoORy5ys9W/7adprfdo9EEg7iUcGNz1U0yNjQk+dCLD/P4eJfm6XXPqvlcJ44KZX0srUEBJUWsr3WGdwSBY5G+bNm+hfv161KtTm5IuLnzduycmpibs3rM3w/xbtm2jUsUKtGrRHBcXZzq2/xxPDw+2bt8OgJmZGZMnjCOoRg2cS5TAz9eXXj26cy0sjEePHgFQuVJFvvumHxUrVMDBwZ7AalVp2bwZvx/PuD8LkdfkihmMAwcOcOzYMcLDw1Eqlbi7u9O4cWO8vDK/xUekS0lJITTsOm1bNtOlKZVKKgSU5XJIaIZlLl8NpWXTRnpplcoH8Psfz38AajQaJk+fTevmTXAt+fpbVY7/+TcxsU+pX7vWa/OK/CslNZWrt+7T+ZPn/UCpVFKllBcXwv6/PcUTk5JJTUvD0lye+yDSpaSmcvXmXTo2qadLUyqVVC7ty4Vrb/csmrsRj/m4x1CMjQtSxsuNXp82wb6o7esLinwhJSWFa2FhfNq6pS5NqVRSPiCAy1dDMixz5epVmjdtqpdWsUJ5jv/xR6bvExcXj0KhwMw88wFsXFwcFhYvz8YJkRcZdIDx6NEjGjVqxMmTJ1EqlWg0GsqXL8+mTZsYNGgQ/fv3Z+rUqa+sIykpiaSkJP205GRMjI2zM/RcIzomFo1Gg42NtV66jbUVd+/dz7BMlFr90q1TNtbWRKnVutdrN26hgLIAzRs1/L/i2Ll3P5XKl6NY0SJvFL/IX9SxcaRpNBSx0v8SLWJlzq3wR/9XHbN/3UFRa0uq+ssFBpFOHfOUNI3mpVuhbK0suP0g4j/XW9rTlZFftaekgx1P1NEs2riDL8dMZ83U4ZgVMn3bsEUeEBMTg0ajwdraRi/dxtqau3fvZVhGpVJj869bp2ysrVGp1BnmT05O5pclS/kgqCZmhTO+cHL/wQO2Bm+nW9cub9wGIXIjg94i9fXXX+Po6IhKpeLp06f07NmTUqVKER4ezp49e1i8eDGzZs16ZR2TJk3CyspK75j306JXlhGvFhp2nU3bfmNgv97/1+1Oj59EcvLMORrU+SgHohP52ZLtB9jz51l++LojJsay0FFkr/cCSlG7WgW8SjoRWM6fmYN6EhuXwL4/Ths6NJFPpKamMmHSFEBLn149M8zz5Ekkw0aOpub71fm4fr0M8wiR1xh0BmPnzp0cP34cS0tLACZPnoyNjQ1z5syhVq1azJw5k/Hjx9O3b99M6xgyZAj9+/fXS3t8Jyxb485NrCwtUCqVL105Uamjsf3XrMY/bK2tUamj/5Vfje2zKzIXLl1BHR1N2y7ddec1Gg0/Ll7Gxm3bWf3Lj3pld+07gKWFOe9VrfzW7RF5m7WFGQWUSiKjn+qlR0Y/pWgGC3FftHzHIZb+dpAFA7/Ey8UxO8MUeYy1pTkFlEqiomP10qOiYylibZll72NhVhgXh+Lci3icZXWK3M3S0hKlUolardJLV6nV2NjYZFjGxsYa1Qsz/s/zW+ulpaamMmHyFB4+fsTUiRMynL2IjIxk4JCh+Pv50rdP77dqi8hask3t2zHoDIaJiYneFXKlUklaWhqpqakAvPfee9y6deu1dVhaWuod78rtUZC+2Nrb04Mz5y/o0jQaDWfOncc/ky32/H29OX3uvF7aqbPn8ff1AaD2h0EsnDOdn2f/oDuK2NrSulljpowZoVdOq9Wye98B6nz4AQVfsZBSvBuMChbE19WJvy8/H+RrNBr+vhxGGc+SmZZb9ttBFm3bz9xvv8DfzTknQhV5iFHBgvi6OfP3xef3xGs0Gk5eCqGMl3uWvU98YiL3Hz6hqE3WDVpE7mZkZISXpydnzj7/TtRoNJw9e073nfhvfr6+nD13Ti/t9Jmz+Pn66l7/M7i4/+ABkyeM111IfdGTJ5EMGDwUL09Pvu3XF6VS9t0R+YdBe/P777/PyJEjiYuLIyUlhaFDh+Lu7o7tsy1RHz9+nOkVBPFcy6aN+G33PnbvP8jtu/eYOf9nEhOTqPdswfXk6bNZtGylLn/zxg35+/RZft28jTt377Fs9TpCw67T9JMGQPqsiFtJF72jYMEC2NrY4FzCSe+9z5y/QPjDR3xcV26PEuk+r1+TzYf/JPjYSW4+eMikZZtISEqmcY30Ga6RP61hzq/Pnz2w9LeDLNi0m1FdW+FQ1IYn6hieqGOIT0zK7C3EO+izhh+x9eDvbD/8BzfvRzBl8VoSkpL4JKgaAKPmL2Pemq26/CmpqYTeukvorbukpKbxWKUm9NZd7kY8Xws0a+UmTl++xoPHkZwPvcHAHxaiVCqp+16lHG+fMJzmzZqyc/du9u7bz507d5kzbz6JiYnUrVMbgKk/TGfx0mW6/E0bN+bkqdNs2LSZO3fvsmLVaq6FhdHkk/TnRKWmpjJu4mRCr4Ux6Lvv0KRpiIpSERWlIiUlBXg2uBgyhGLFitGtaxeio2N0eYTIDwx6yfn777+nbt26WFtbp++uYGbG+vXrdeevXLlCp06dDBdgHvFhjepER0ezdNVaVCo1Hu5uTB4zXHeL1KPHT/Rmikr5+TLsu34sXrmGxctX4eTowNhhA196Bsb/Y+ee/ZTy88HFuURWNUfkcXWrBqCKiePHTbuJjI7F28WROd99QZFnt0hFRKlRKJ/3xw0HTpCSmsbAuSv06vmyaR26N6ubo7GL3KtOYEVUMbH8vGE7kepYvEs6MWtwL90tUg+fqFC+8Dn3WBXN50Mm616v3L6fldv3U8HPix9H9gPgUZSa4XOWEP00DhtLc8r5eLB43HfYWMpOPu+SD2rWIDo6muUrV6FSqXB3d2fC2DG6C5yPHz/W61ul/P0YPOA7lq1YydJly3F0cmTU8GG4uqbP0j6JjOSPP/8EoGefr/Xea+qkiZQrW4bTZ87w4EE4Dx6E065jJ708u38LzsbWCpEzFFqtVmvIAOLj4zl27BjJyclUq1aNokWLvnWd90IvZkFkQryeddTbbZEpxP8rzVi27RU5Q2X55hebhPgvXD0zf1q6oV29nvEuYrmBr0fuv6hr8JvmCxcuTN26cpVSCCGEEEKI/MDgK4oSEhI4duwYly9ffulcYmIiy5cvN0BUQgghhBDiXaVFkWuPvMCgA4zQ0FD8/PyoWbMmZcqUISgoiPDwcN356OhoOnfubMAIhRBCCCGEEG/CoAOMQYMGUbp0aR49ekRISAgWFhZUr16dO3fuGDIsIYQQQgghxH9k0DUYx48fZ9++fRQtWpSiRYsSHBxMz549qVGjBgcPHsTMzMyQ4QkhhBBCiHeQVps3bkXKrQw6g5GQkKD3cDaFQsGCBQto1KgRQUFBhIaGGjA6IYQQQgghxJsy6AyGr68vJ0+exM/PTy997ty5ADRu3Pg/1XtPI08CFjnjSRE7Q4cg3hEFFWmGDkG8IxLSTA0dgnhHuBo6AJFtDDqD0axZM9asWZPhublz59K2bVsM/JgOIYQQQgghxBsw+IP2ssMfV6MNHYJ4R5gWSDZ0COIdITMYIqfIDIbIKZV9rA0dQqYuhz0wdAiZ8vd0NHQIr2Xw52AIIYQQQggh8g+DP8lbCCGEEEKI3CSvPNAut5IZDCGEEEIIIUSWkRmMfEyr1bJ59c8c2ruF+LinePmWpWOPQdg7umRa5uql0+zcvJJbYVdRq57w9ZCpVKz2gV6ezWt+5s+je4l88pCCBY1w9fCl5ec98PApnc0tErmBVqtl/apf2L87mLi4WHz8yvBFz+9wcHr17m27t28keNMa1KooSrp50Ln7N3j6+OvOq1WRrFw8n/Nn/iYxIR6HEi40b92BqtU/0Kvn9N/H2bhmCbdvXcfYyBi/MuUZMHxSdjRVGJhWq2XtysXs272d+Lin+PiV4cte/XF0KvHKcju3b2brxrWoVVG4unnQ9au+ePno71YYcuUiq5cv4lrIFZRKJa7unowY9z0mJiY8ehjO+jXLuXj+NGpVFDa2Ran5YR1atGmPkZFRdjZZ5CJarZaNq3/m4J6txMc9xduvLJ17DHz1d+jFM/y2eSU3r19FHfWEfkOnUqlakO58amoqG1b+yNlTx3kccZ9CZuaULleZNh16YVOkWE40S4gcITMY+diOTcvZ+9s6OvUYzMhpizExLcT3o78mOTkp0zJJiYk4u3rRvvuATPPYO7rQ/ssBTJi9hmGTf6ZocQemje5DTLQqO5ohcpltG1exM3gDX/T6jgk//IypaSEmjuz/yn51/Mh+li+aS4u2nZk86xdKunkycWR/otXP+8y86eN5cO8OA0dMZtq8ZVQJrMmMKSO5ef3583D+/P0Qc38Yxwe1GzJ1zlLGTlvA+0G1s7W9wnC2bFjDjuBNdO/1LZOm/4ipqSnjRnz3yr72+5EDLF04j9afdWTa7IWUdPNg3Ijv9PpayJWLjB85kHLlKzN5xo9MmfkTDRo1Q6lMvyXi/t07aLUauvf+jhnzl9G5W2/27NzG6mULs73NIvfYvmkFe7b/Spcegxgz7RdMTEyZMqrvq79DkxJwcfOiYybfoclJidy6HkLTNl0YN2M5/QZPJvz+HaZP+C67miGEQcgAI5/SarXsDl5Lo1ZdqFA1CBdXL77sNxp11BNO/3E403LlKr5Hy897UCnww0zzBAbVp1RAFYrbO1HCxYPPuvYjIT6Ou7euZUdTRC6i1WrZsXU9zdt0oHK1GpR086RX/+GooiL5+8TRTMv9tmUtH9VrxId1GlLCxY0veg3A2MSUg3u36/KEXLlI/UYt8PTxx87eiRafdsLMzJwbYSEApKWlsvTnWXzepRd1Pm6Ko5MLJVzcCKzxUba3W+Q8rVbL9q3radmmPVUC38fVzYM+3w5FFRXJXyeOZVouePOv1K7/CbXqfIyziyvde3+Liakp+/fs0OVZsnAeHzduQfPW7XAp6YZTCReq16iFkZExAOUrVaX3N0MIqFAZewdHKlerTuPmbfjj+JFsb7fIHbRaLbu2raVJ685UrBaEi5sXX32T/h166jXfoa0+/4rKgR9keL6wmTmDx82h2vu1cSxREk/fMnTo/h03w67y5HFENrVGiJwnA4x86vHDB0SrIilVroourbCZOe7epQgLuZBl75OaksLB3VsobGaOi5t3ltUrcqdHDx+gVkVSJqCyLq2wmTmePv5cu3oxwzKpKSncCAulTEAlXZpSqaRMQCWuXb2kS/PxK82Jowd4GhuDRqPh98P7SElOplSZ8gDcDAslKvIxCoWCQV93pnv7Jkwa9S13bt3IptYKQ3oYEY5aFUXZgIq6NDMzc7x8/Ah5od+8KCUlhethoXpllEolZQMqEvqsTLRaxbWQy1hZWTP02550adeUEYO+5sql86+MJz4uDgsLyyxomcgL/vkOLf2v71AP71Jcy8LvUICEuKcoFAoKm5lnab1CGFKuWIPx119/ceLECSIi0kfv9vb2BAYGUqVKldeUhKSkJJKS9Kcrk5OTMDY2yZZY84poVSQAVta2eumW1ra6c2/j7N9Hmf/9cJKTErGyKcqAMXOxsLR+63pF7qZWRQFgZW2jl25lbYNaHZVhmZiYaDSatJf6opW1LQ/u3da97jdoLDOnjKJr248pUKAAxiamfDtsIvaO6ffbP4xI35N8w+rFdPiiD8Xs7Nm+eS1jh/Zh5k9rMJcff/nKP33N2ubf/cZGd+7fYp/1NesM+uf9u3eA5/1o3eqldOzaA1d3Tw7v38Poof2ZMX9phus7wh/cY2fwJjp07fHW7RJ5g/rZ96Rlht+hGfe//yI5OYm1y+YSWLMuhQvLACM3kV2k3o5BZzAePXpEjRo1qFatGjNmzODAgQMcOHCAGTNmUK1aNWrUqMGjR49eWcekSZOwsrLSO5b/PD2HWpB7HD+0iy/bBOmOtLTUbH0/vzKVGDdzJcOnLKJshWrMmzqEmEx+YIq86+jBPXRoWUd3pKVmX79at3IR8XGxDB8/k4kzFtGwaRtmThnJnVvXAdBqNQA0a5O+8Nvd05ce/YYCCk4cO5BtcYmcceTgXtq1qK87suszTKNJf7Zs3QaNqFXnY9w9vOn8ZW8cSzhzYO+Ol/JHPnnM+JEDCXz/A+rUb5QtMQnD+/3QLrq2/kB3ZPd3KKQv+J4zdRhaLXTqMTDb30+InGTQGYyePXuSlpbGlStX8PHx0TsXEhJCly5d6NWrF+vXr8+0jiFDhtC/f3+9tLO3ErMl3tysfJUaePiU0r1OSUl/wnS0Ogpr26K69Bh1VJbcymRiWgg7B2fsHJzx9CnDwK9acHjfNhq17PTWdYvco1LV9/F6Yaen5/1Khc0L/SparcLVzTPDOiwtrVAqCxD9rwFotDoKa5siAESE32f39o18P285ziXdAXB19+LqpXPs3r6Jbr0H6PpxCWdXXR1GRsbY2TsQ+fjh2zdWGFTlqtX1dnpKSUkBeLaLUxFderRahat7xn3N4llfU6v1N5yIVqt0MyH/1PViP0p/XZIn/+pHUZFPGDWkHz5+pfiqjyzCzc8qVKmBh/fz79DU1PT+F6OO0vusi1FH4eLu9dbvlz64GErko3CGjJ8vsxci3zHoDMbu3buZN2/eS4MLAB8fH2bPns2uXbteWYeJiQmWlpZ6x7t4e1Shwma6H/x2Ds44ObtjZVOEy+f/1uVJiH/KjdBLePqUyfL312g1pD778Snyj0KFC2PvWEJ3lHBxw9qmCBfOntTliY+PIyzkMl6+GW9TXNDICHdPby6cO6VL02g0XDx3Ci/f9C/05KT0iwIKpf5HklJZQDdz4e7pg5GRMQ/u39WdT01N5fGjCIoWt8+aBguDKVS4MA6OJXSHs4sr1ja2XDh3WpcnPj6OayFX8PEtlWEdRkZGeHh6c+Gsfl87f/Y03s/KFLezx7ZIUb1+BBB+/y7FitvpXkc+eczIwX1x9/SmV7/BKJWyZDE/K1TYDHtHZ93h5OyGlU0RLp17/h0aH/+U66GX8HrL79B/BhcPH9xl8Li5WFhavW34QuQ6Bp3BMDExISYmJtPzsbGxmJi8e4OFrKBQKKjX6FO2/boYOwdnitk5smn1j1jbFqXCC3tyTxnRkwrVPqBOw9YAJCbE8zD8nu7844cPuH0jFHMLS4oUsycpMYFt65dQvkoNrG2KEhujZv+ODagjH1O5uuzmk98pFAo+btKKzeuW4eDkTHE7B9atXISNbREqB9bQ5Rs3tC+VA2tSv1ELABo2/ZT5Mybg4eWLh7cfO7b+SlJiAh/UbgiAY4mS2DuUYOHcabTv0gtzSyv+PnGEC2f/ZtDIqQAULmxG7QZNWL/qF4oULU6x4vZs27QagGrvZ77rmcibFAoFnzRpxYa1y3FwLEFxe3vWrFiMjW0RqgS+r8s3eug3VAmswceNmgPQqFlr5kyfhIeXL17evmzfuoGkxARq1Wmgq7dJ809Zt2oJrm4euLp7cmj/bu7fu8N3Q8cCzwYXQ/pSrJg9Hbv2JCZarXu/F2dTRP6lUCio3/hTtvy6BDtHZ4rbObJh1U9Y2xal4gvfoROH96JStQ+o+0krIPPvUDMLS4oWsyc1NZXZkwdz60YI3474AY1Go1vvYW5uSUF5zorIJww6wGjTpg0dO3ZkxowZfPTRR1hapi/SjImJYf/+/fTv35+2bdsaMsQ87ePmHUhKTGTp/InpD9rzK8d3o2bpzfA8irjP0xi17vXNsCtMHv58IeOaxTMBeL9WQ7r1HYVCqST83i2OHfiNpzFqzC2scPPyZ+iknynh4pFTTRMG1LhFO5ISE/l5ztT0h5/5l2HI2B/0+tXDiPvEvtCv3qv5ETHRan5duSj94WfungwZ+4PutpWCBQsyePQ0Vi/7kanjBpGYkICdgxM9vxlG+cqBuno+79KLAgUKMG/6OJKTkvD08WfEhFmYm8sC7/yoacu2JCYm8OOc74mLe4qvfxlGjJum19ciwh8QGxOte129Zi2io9WsXbkYtSoKN3dPho+dprdY/JOmrUhOTmbJwrk8jY3F1c2DkeN/wN7BCYBzZ04S8eA+EQ/u82XHlnoxbfwt8y1KRf7ySfP2JCUmsHjepPQH7fmXY+Dol79DX/ysuxF2hYnDeuper/plJgA1ajWke7+RqCIfcfqv9C29h/Vtr/d+QyfMx79MRYTIDxRarVZrqDdPSkqiX79+LF68mNTUVIyN0/cgT05OpmDBgnTt2pUZM2a88SzGH1ejX59JiCxgWkBuCxM5o6AizdAhiHdEQpqpoUMQ74jKPtaGDiFTF8Jy79q+Mp52r89kYAYdYPwjJiaGU6dO6W1TW7FiRd2MxpuSAYbIKTLAEDlFBhgip8gAQ+QUGWD8N3lhgGHwVWtXrlxh48aNODg40LZtW8qXL8+vv/5Kv379OHBAtp4UQgghhBAiLzHoGoxdu3bRpEkTzM3NiY+PZ/PmzXTo0IFy5cqh0WioW7cue/bsoVatWoYMUwghhBBCvEO0WnnQ3tsw6AzG2LFjGTBgAJGRkSxZsoTPPvuMbt26sXfvXvbv38+AAQOYPHmyIUMUQgghhBBCvAGDDjAuXbpEp06dAGjdujWxsbG0bPl8x4527dpx/vx5A0UnhBBCCCGEeFMGX4OhUKRPQSmVSkxNTbGyev7AGQsLC6KjZcG2EEIIIYQQeYVB12C4urpy7do1PDzSn59w4sQJXFxcdOfv3LmDg4PDG9frlXopy2IU4lXUxvIEaZEzCmhSDR2CeEckK+Rhb0KIt2PQAUaPHj1IS3u+9WLp0qX1zu/cuVMWeAshhBBCCJGH5IrnYGS1yIvHDR2CeEeoC8sMhsgZMoMhckqUpoihQxDviAreubevnb322NAhZCrAq5ihQ3gtg6/BEEIIIYQQQuQfMsAQQgghhBBCZBkZYAghhBBCCCGyjEEXeYvss3HnflZt3UmUOhpPVxf6d22Hv5d7hnlv3LnPorWbuXrjFhGPI+nbuS1tPqmrl+fMpRBWb91JyI3bPFGpmTSwD0FVK+REU4SBBQcHs3HDBlQqFW7u7vTo0QMfH59M8x89epQVy5fz8OFDHJ2c6NK5M5WrVNGd12q1rFyxgl27dhEXF4e/vz+9evfGyckJgIcPH7Jm9WrOnTuHSqXC1taWWrVq0ebTTzEyer67zalTp1i5YgV37tzByMiI0mXK0K1bN+zs7LLvjyFy1Lbtv7Fh4yaiVCrc3dzo+VV3fH28M81/5Ogxlq1cycOHj3BydKRr505UqVwJgNTUVJYuX8nfJ08SHhGBmZkZ5QPK0bVTR4oUSb8PPOLhQ1avWcfZ8+dQqdQUsbWl1ocf0LZNa72+J/IfrVbLhlWLOLBnG3Fxsfj4laVLzwE4ODq/stye3zYSvGkV0aooXNw86dS9P57e/rrzD8PvsXLxXEIunyc1JZmyFarRqXt/rG1sdXn6dG3Ok0cRevV+2uErmrTqkLWNFCIHyQxGPrTv9z+ZvXQtXVo3Ycm00XiWdOabcT8QFR2TYf7E5CQc7YrR4/NWFLG2yjhPUhKers582+3z7Axd5DKHDx9m4c8/81m7dsyZMwd3NzdGDB+OWq3OMP/ly5eZMnkydevVY87cuQQGBjJu3Dhu3bqly7Nh/Xq2bdtG7z59mDFzJqampowYPpzk5GQA7t69i0arpU+fPiz48Ue+7N6dHTt2sGzpUl0dERERjB0zhnIBAcydO5fxEyYQEx3N+HHjsvGvIXLSoSNH+XnhItp91pZ5s2fi7ubGsBEjM+17ly5fYdLUadSvW5f5s2fxXmA1xoyfwK1btwFISkoi7Pp1PmvbhnmzZzJy2BDu3bvPqLHjdXXcvXsPjVZD3969+Hn+PLp3+4Lfdu5iybLlOdFkYUDBG1eya/t6uvYcwLjvF2Fiasrkkd+QnJyUaZkTR/exYtFsWrTtwsSZSyjp5snkkd8QrY4CIDExgYkj+6FQKBg+YQ6jp/5EWmoK348bgEaj0aurVbtuLFgerDvqNWqVre0VIrvJACMfWhu8h8a1a/JJrRq4OTsxsHsHTEyM2b7/aIb5/T3d6d2xDXXer4qRUcaTWoEVytL9sxYEVa2YnaGLXGbz5s3Ub9CAunXr4lKyJL379MHExIQ9e/ZkmH/r1q1UrFSJli1b4uLiQocOHfDw8CA4OBhIv0q4ZcsWPv30UwIDA3Fzc+Pb774jMjKSE8fTd3+rVKkS/fv3p0LFijg4OFCtWjWat2jB8ePPd4cLu3YNjUZDhw4dcHB0xNPTkxYtWnDjxg1SU2W3pfxg0+Yt1K9fj3p1alPSxYWve/fExNSE3Xv2Zph/y7ZtVKpYgVYtmuPi4kzH9p/j6eHB1u3bATAzM2PyhHEE1aiBc4kS+Pn60qtHd66FhfHo0SMAKleqyHff9KNihQo4ONgTWK0qLZs34/fjJ3Ks3SLnabVadm77lWatO1GpWk1KunnS85uRqKKecPKPI5mW+23LWmrVa8wHtT+hhIsbXXsOxNjEhEN70/tc6OXzPH4UwVf9huPi6oGLqwc9vhnBjbCrXDp/Sq8u00KFsbYpojtMTQtla5vF62lR5NojL5ABRj6TkpJKyPVbVCpbSpemVCqpXNafi6FhBoxM5DUpKSmEXbtGQECALk2pVBIQEMDVK1cyLHP1yhXKv5AfoGLFirr8ERERqFQqAsqX1503MzPDx8eHK1evZhpLXFwc5hYWuteeXl4oFAr27t1LWloacXFx7D9wgICAAAoWlDs/87qUlBSuhYVRIaCcLk2pVFI+IIDLV0MyLHPl6tWX+16F8q/pV/EoFArMzM1fkScOixf6nsh/Hj18gFoVSemASrq0wmbmeHj7c+3qxQzLpKakcDMshNLlnpdRKpWUDqjMtZD0MimpKShQ6N1eZ2RsjEKhJOTyOb36tm1YQbfP6jO4b0eCN60iLU0ulIi8LVcPMFQqFcuXy9T0m1DHxpKm0WBrbamXbmtlRZQ641ukhMhITEwMGo0GGxsbvXRrGxuiVKoMy6hUKqwzyK96lv+ff2ZUpyqTOh88eEDwtm183KCBLs3e3p4JEyawbOlSmjRuTKuWLYl88oQhQ4e+WSNFrvRP37O21u8nNtbWmfYTlUqNjbV1BvnVGeZPTk7mlyVL+SCoJmaFC2eY5/6DB2wN3s7HDeq/cRtE3hGtSr+lycraVi/dytoW9bNz/xYTo0ajScPKJvMyXj6lMDE1ZfXS+SQlJpKYmMDKxXPRaNJQR0XqytRv1IqvB45l+IS5fFS/KVt/Xc7qJfOysolC5Lhcfanvzp07dO7cmQ4dMl/olJSURFKS/j2SScnJmBgbZ3d4Qohs9uTJE0YMH877NWpQ/4UBRlRUFLNmz+aj2rX5ICiI+IQEVq5YwcQJE5gwcSIKRd6YQhaGkZqayoRJUwAtfXr1zDDPkyeRDBs5mprvV+fj+vVyNkCRrY4d2s2ieVN1rweO/D5b3sfSyoZ+g8bzy4Jp7A5ej0Kh5L2atXHz8EGhfH59t2HTtrp/L+nmScGCRvwybwqfduyBkZH8ljEUrVa+R96GQQcYMTGvvqIeGxv72jomTZrEmDFj9NIG9OjCoJ5d3yq2vMrawoICSuVLsxVR0dEvzWoI8SqWlpYolcqXrhirVSps/zUD8Q8bGxvUGeT/Z8bin3/+szvUi3ncPTz0ykVGRjJ48GD8/P35+uuv9c5t374ds8KF6dr1+f/nAwYMoEOHDoRcvYqvn98btlbkJv/0PbVavy+p1OqXZr/+YWNjjepfC8DT81vrpaWmpjJh8hQePn7E1IkTMpy9iIyMZOCQofj7+dK3T++3aovIfSpWeR9P7+e3EaekpG8wEa2Owsa2qC49Wh2Fq7tXhnVYWlqjVBbQzX68WObFHaLKVqjKrIUbiIlWU6BAAczMLfiq/ScE2jtmGp+ntz9paWk8fhiOY4mS/6mNQhiaQW+Rsra2xsbGJtOjZs2ar61jyJAhREdH6x39vmifA9HnTkZGBfHxcOXUhcu6NI1Gw8nzVyjt7WnAyEReY2RkhKeXF+fOntWlaTQazp49m+kPeF8/P86+kB/gzJkzuvz29vbY2Njo1RkfF0dISAh+vr66tCdPnjBo0CC8PD355ptvUCr1P6qSkpL0rgACKAsUSI9Rq33TpopcxsjICC9PT86cPa9LS+975/D3zXiLZD9fX86e07+v/fSZs3r96p/Bxf0HD5g8YTyWli9fdHnyJJIBg4fi5enJt/36vtT3RN5XqLAZ9o4ldEcJFzesbYpw8dxJXZ74+Diuh17Gy7d0hnUUNDLCzdOHiy8s1tZoNFw6dxIvn5fLWFpZY2ZuwcVzJ4mJVlGxyvuZxnf75jUUSiWW1hkPpoXICww6g2FhYcGwYcOoWrVqhuevXbtG9+7dX1mHiYkJJiYmemkp7/jtUZ82qsv4OYvw9XDF38udddv3kJiUxCe10j/Qxs5eSDFba3p8nr4NXkpKKjfvPQAgNTWNx5EqQm/eobCpCSUc0p8pEJ+QyL2IR7r3CH/0mNCbd7A0N8O+WJEcbqHIKc2aNWP6Dz/g5eWFt48PW7dsISkpiTp16gDw/fffU6RIETp37gxAkyZNGDRwIJs2bqRylSocPnyYa9eu0efZDIRCoaBp06asXbsWRycn7OzsWLFiBUWKFCHwvfeA9MHF4EGDKF68OF2/+ILo6GhdPP/MelSuXJktmzezetUqgj74gISEBJYtXUrx4sXx+NdMiMibmjdryvfTZ+Dt5YmPtzebt24lMTGRunVqAzD1h+kULVKELp06AtC0cWMGDB7Chk2bqVK5EoePHOVaWBj9ns1ApKamMm7iZMKuX2fsqJFo0jRERaXPkFhYmGNkZJQ+uBgyhOLFitOtaxeiX9ja29ZWfuzlVwqFggaNW7Nl3TLsHZ0pbufI+pU/Y2NblErVnl/oHD+sD5UDg6j3SUsAGjb9lAUzxuPu6Yuntz87t64jKTGRoNqf6Moc2rcdpxKuWFpZE3r1IssXzqRBkza6mYnQqxcIC7lMqbIVMC1UmGtXL7Ji0Sze/6Ae5uZy14HIuww6wKhQIf1BbUFBQRmet7a2RitXI99Y7epVUUfHsnDtFqLU0Xi5uTB9eH9snz3j4uGTSJQv3KP+RKWm03ejdK9Xb9vF6m27KF/Kh3ljBwNw9foteo+aossze+laAD7+oDrD+3yRE80SBhAUFERMdDQrVq5EFRWFu4cHY8eN092m8vjRI72+5O/vz8BBg1i+bBlLly7FycmJESNG4OrqqsvTslUrEhMTmTN7Nk+fPqVUqVKMHTcO42cXBs6cOcODBw948OABHdrrz0bu2LkTgICAAAYOHMiGDRvYsGEDJiYm+Pn5MW78+JcuOIi86YOaNYiOjmb5ylWoVCrc3d2ZMHbM8773+LFe3yvl78fgAd+xbMVKli5bjqOTI6OGD8PVNf2H3JPISP74808AevbRv+Vu6qSJlCtbhtNnzvDgQTgPHoTTrmMnvTy7fwvOxtYKQ2vU4nOSEhNZNHcK8XFP8fEvy+Ax0zE2fv558jDiPrExat3rwBq1iYlWs2HVQtSqKEq6ezF4zHS9W6TC791h7bIfefo0hmLFHWjauiMfN/lUd96ooDEnju5j45pfSElJpridIw2afErDps/zCJEXKbQG/AW/cOFC4uPj6du3b4bnHz58yI8//sioUaMyPJ+ZyIvHX59JiCygLmxv6BDEO6KARratFDkjSiOz0iJnVPDOvX3tdGjk6zMZSG7+u/3DoAOM7CIDDJFTZIAhcooMMEROkQGGyCm5+YfyqdCMtyjODSp6274+k4EZfPXalStXWLJkCVefPQzp6tWr9OjRgy5dunDgwAEDRyeEEEIIIYR4EwZdg7Fr1y6aNGmCubk58fHxbN68mQ4dOlCuXDk0Gg1169Zlz5491KpVy5BhCiGEEEIIIf5PBp3BGDt2LAMGDCAyMpIlS5bw2Wef0a1bN/bu3cv+/fsZMGAAkydPNmSIQgghhBBCiDdg0AHGpUuX6NSpEwCtW7cmNjaWli1b6s63a9eO8+fPZ1JaCCGEEEIIkdsYfA2G4tk2g0qlElNTU6ysrHTnLCws9PbAF0IIIYQQQuRuBl2D4erqyrVr13QPxjpx4gQuLi6683fu3MHBweGN6zWNffT6TEJkASsUr88kRBZIKVjI0CGId4SNkcGvPYp3Ru7dRUqrle/3t2HQAUaPHj1IS0vTvS5durTe+Z07d8oCbyGEEEIIIfKQfPkcjLgTWwwdgnhHJFjYGToE8Y6QGQyRUxKNzA0dgnhHuHl4GjqETJ0MURk6hExV8rExdAivJfOgQgghhBBCiCwjAwwhhBBCCCFElpEBhhBCCCGEECLLGHSRtxBCCCGEELmNVnaJfCsywMin1u07zvKdR4iMjsXbxYGBnzehtLtzhnk3HfqT7cdPc/3eQwD8XJ3o3bK+Xv79Jy+y8eAfXLl1n+i4eNaM6YtPScccaYvIezbu3MfqrTuJUkfj6erCN10/x9/LPcO8N+7cZ9HaTYTcuEXE40i+7tyWNp/Uy+GIRW60+bfdrN0STJRKjadrSb7+sjN+3pkvCj30+wl+WfUrEY8eU8LRnu4d2lGtUnm9PLfv3uOnZas5d+kyaWkaSjo7MXbwt9gVKwrA/fAIFixZyYUrV0lJSaVKhXJ8/WVnbK2ts7OpwsC2BW9nw8aNqFQq3N3c6NnjK3x8fDLNf+ToUZavWMnDhw9xcnSkS5fOVKlcWXf+2O+/s2PHTq6FhREbG8u8ObN1W/IDxMbGsmLlSk6dPsPjx4+xsrIiMLAaHdu3x8zMLFvbKkROkFuk8qHdf55j+trtfNn0I1aP+RovZwd6ff8LUTFPM8x/6uoN6lcN4OdBX7J0eE/sbK3oOW0Rj1TPH3KYkJRMgLcrX7dukFPNEHnUvt//ZM7StXRp3ZTF08bgWdKZ/uO+RxUdk2H+pOQkHO2K0ePzVhSxtsowj3j3HDh6nPmLl9OpTQsWTp+Mh1tJBoyeiEqd8cNXL14JYez3s2lY+0MWzZjM+1UrM3zSNG7cvqPLcz88gj5DRuFSwpGZE0bxy6ypdGjdAmMjIwASEhMZMHoiCgXMGDeSuZPHkpKaytDxU9FoNDnSbpHzDh8+wsKFC/n8s8+YO2c27u5uDBsxArVanWH+y5cvM3nKVOrVrcu8ObMJDAxk7Ljx3Lp1S5cnMTGJUqX86dK5c4Z1REZGEhkZRbcvuvLjgvl8+803nDp5ihkzZ2VDC4XIeTLAyIdW7T5Ks6AqNKlRGXcnO4Z1bIapsRFbj/ydYf4JX7Wl9UeB+JR0xM2xOCO7tESr1fLX5TBdnk+qV+DLJrWp6p97t5QTucO64N00qh1Ew1o1cHN2YkD3jpiYGLN9/5EM8/t5utO746fUfr8aRkYyqSrSrd/6Gw3rfkSD2h/i6lKC/j2+wNTEmB37DmaYf2PwTqpUCODT5o0p6VyCru3a4OXuxubfduvyLFq5lqoVy/NVp8/xcnfDycGe6lUrYfNsYHvxSggRjx4xuG9P3F1dcHd1YUjfXoSE3eD0+Ys50m6R8zZt3kz9+vWpW7cOJV1c6NO7NyYmpuzesyfD/Fu2bqNSxYq0atkCFxcXOnZoj6eHB9uCt+vy1P6oFu0++4zy5QMyrMPV1ZURw4dRrWpVHB0cCAgoR8eOHfjzzz/1ng8mDEerVeTaIy+QAUY+k5KaypVb96nq76VLUyqVVC3lyfnrd15R8rnEpBRS09KwNCucXWGKfColJZWQ67eoXNZfl6ZUKqlUthQXQ68bMDKRl6T3oxtULFdGl6ZUKqlYrgyXQ65lWOZSSCgVy+k/rLVK+XJcDgkFQKPR8MfJMzg7OjBg1ASaduhGj++GcfSP5xdeUlJSAQVGz2Y0AIyNjVAoFFy4EpKFLRS5RUpKCtfCwigfEKBLUyqVlA8I4MrVqxmWuXL16ksDh4oVK2Sa//8VFxdP4cKFKVCgwFvVI0RukCsGGJlNPWs0Gu7cefWP4qSkJGJiYvSOpOSU7AgzT1DHxpOm0WBrpf+gJFtLCyKjY/+vOmav30Exa0uZrRBvTB0bm97//nWrk62VJVGZ3NoixL9Fx8SgyaAf2VhbEaVSZ1gmSq1+aZ1Eev70fqeKjiEhMZHVG7dSpUIA00YP4/1qlRk5+QfOXrwMgL+PF4VMTfhp2SoSk5JISExkwZIVaDQaolS596Fb4r+LedbXrG2s9dKtra1RRWX831ylUmFtnUH+t+gj0dHRrFmzhgYN6v/nOoTITQw6wIiJiaF169aYmZlhZ2fHyJEj9aYGHz9+jJub2yvrmDRpElZWVnrH98s3Znfo+daS7QfZ/ec5vv+6AybGRq8vIIQQeYD22YWs6lUr0apJQ7zcXWnXsimBlSqwbddeAKytLBk98BtO/H2aBm060rBtZ57GxePt4YZCkSuux4l8KC4+npGjRuPi4sLn7doZOhwhsoRBb3geMWIE586dY8WKFajVasaPH8/p06fZtGkTxsbGAGi12lfWMWTIEPr376+Xlnpmdya58z9ri8IUUCqJitZf0B0VE0sRK4tXll2+8zBLfjvEjwO74e3skJ1hinzK2sIivf/9a7YiKjrmpavRQmTGytISZQb9SKWOxvZfV5r/YWttTdS/FuWm57fS1VmgQAFKOjvp5Snp7MSFy89vbalcvhyrf5qNOiaGAsoCWJib0azjl9R6v/jbN0zkOpbP+pr6XzNjarUaG1ubDMvY2Ni8tABcrVZjY5Nx/leJj49n+IgRFCpciJEjhlOwoKxDE/mDQS/JbNmyhZ9++omWLVvyxRdfcPLkSR4/fkyjRo1ISkoCQKF49WIWExMTLC0t9Y53+cq7UcGC+Lk66S3Q1mg0/HU5jLIeLpmWW7rjEIu27Wfut13wdyuRE6GKfMjIqCA+Hq6cvHBZl6bRaDh1/jKlvT1eUVKI59L7kTunz1/QpaX3o4v4+3hlWKaUj/dLC7FPnr2Av4+3rk5fTw/u3g/Xy3P3fjh2xYu9VJ+1pSUW5macPn8RdXQM71Wp9LbNErmQkZERXp6enD13Vpem0Wg4e/Ysfr6+GZbx8/Xl7Nlzemmnz5zJNH9m4uLjGTp8BAULGjF65EjdhVUh8gODDjAeP35MyZIlda+LFi3Kvn37iI2N5eOPPyY+Pt6A0eVd7erVYPPhvwg+doobDx4ycflmEpJSaFwj/QtyxM/rmLN+py7/0t8OsWDTHkZ1aYVjUVueqGN5oo4lPjFJlyf6aTwhtx9w48EjAG5FPCbk9gOeqP+/dR3i3dGmUT2C9x1mx8Fj3Lr3gO9/Xk5iUhINa9UAYNzsn1mwcr0uf0pKKqE3bxN68zYpqWk8jlQRevM298IfGqoJIhdo1aQh2/ccYNeBw9y+e48ZPy4iMTGJBrU/AGDijLn8vHy1Ln+LRg346/Q51m0J5va9+yxZs56Q69dp1vD5M1U+bdaIg8eOs33Pfu6FR7Dpt10c//sUTRrU1eXZue8gl0JCuR8ewZ5DRxk9dQatGn+MSwl57k9+1bxZM3bu2s3effu4c+cOc+bNIzEpkbp16gAw7fsfWLxkqS5/0yaNOXnqFBs3beLu3busWLmKa9fCaNzoE12e2NhYrl+/rltHeu/efa5fv05UVBSQPrgYNmw4iYmJfNOvL/Hx8URFRREVFSW7SOUSmlx85AUGnYtzcXHhypUreussLCws2LNnD3Xr1qVZs2YGjC7vqle1HKrYOBZs3kNkdCw+Lo7M/baL7hapiEg1yhdmhtYf+IOU1DQGzFupV8+XTWrzVbP0D9jDZy4z+pfnPwqHLFj9Uh4hAGpXr4o6OpZFazcTpY7Gy82FH4Z/q7tF6uGTSL2ZyScqFZ2/G6V7vWbbLtZs20X5Uj7MHTskx+MXuUOtGu+hjolhyepf0x+05+bK1FFDdAu5Hz6JRKF8fo2stJ8PI77twy8r17FoxVqcHO0ZP2QA7iWfz9zWCKxC/x7dWLVhC7MXLsHZyZGxg/tT1v/5lec798P5ecUaYp8+xb54cT5v1YxWjRvmWLtFzgsKqkl0TDQrVqxMf9Ceuzvjx47V3fL06PFjFMrnn1n+/v4MGjiAZctXsHTpMhydnBg5Yjiurq66PCf++IPpM2bqXk+aMgWAdp99RvvP2xEWFsbVkPSdybp0/UIvnqVLFmNvZ5dNrRUiZyi0r1vkkI2+/vprwsPDWb9+/UvnYmNjqVOnDn///fcbj+bjTmzJogiFeLUEC/kSEDkjpWAhQ4cg3hGJRuavzyREFnDzyL27Vf5xNffufFjNN/evaTToDMaYMWN48OBBhucsLCzYu3cvp0+fzuGohBBCCCGEEP+VQddg2NjYoFQqWbJkCVefPaDm6tWr9OjRgy5duvD3338TFBRkyBCFEEIIIYQQb8CgMxi7du2iSZMmmJubEx8fz+bNm+nQoQPlypVDo9FQt25d9uzZQ61atQwZphBCCCGEEMrSxn0AAAj8SURBVOL/ZNAZjLFjxzJgwAAiIyNZsmQJn332Gd26dWPv3r3s37+fAQMGMHnyZEOGKIQQQggh3jFarSLXHnmBQQcYly5dolOnTgC0bt2a2NhYWrZsqTvfrl07zp8/b6DohBBCCCGEEG/KoAMMeP4gPaVSiampKVZWz1fGW1hYEB2de1fxCyGEEEIIIfQZdA2Gq6sr165dw8Mj/Qm/J06cwMXl+Z7ld+7cwcHB4Y3rNQtsmlUhvjOSkpKYNGkSQ4YMwcTExNDh5Blmhg4gD5K+JnKK9DWRU6Sv5T+BfpaGDiFPM+hzMH788UecnZ1p2DDjhxgNHTqUR48esWjRohyO7N0TExODlZUV0dHRWFrK/1Qi+0hfEzlF+prIKdLXhNBn0BmMr7766pXnJ06cmEORCCGEEEIIIbKCwddgCCGEEEIIIfIPGWAIIYQQQgghsowMMAQAJiYmjBo1ShaniWwnfU3kFOlrIqdIXxNCn0EXeQshhBBCCCHyF5nBEEIIIYQQQmQZGWAIIYQQQgghsowMMIQQQgghhBBZRgYYQgghhBBCiCwjAwzBvHnzcHV1xdTUlKpVq/LXX38ZOiSRDx05coRGjRrh6OiIQqFgy5Ythg5J5EOTJk2icuXKWFhYULx4cZo2bUpISIihwxL50IIFCyhbtiyWlpZYWloSGBjIzp07DR2WELmCDDDecevWraN///6MGjWK06dPU65cOerVq8ejR48MHZrIZ+Li4ihXrhzz5s0zdCgiHzt8+DC9evXijz/+YO/evaSkpFC3bl3i4uIMHZrIZ0qUKMHkyZM5deoUJ0+epFatWjRp0oRLly4ZOjQhDE62qX3HVa1alcqVKzN37lwANBoNzs7O9OnTh8GDBxs4OpFfKRQKNm/eTNOmTQ0disjnHj9+TPHixTl8+DA1a9Y0dDgin7O1tWXatGl07drV0KEIYVAyg/EOS05O5tSpU9SuXVuXplQqqV27NidOnDBgZEIIkTWio6OB9B9+QmSXtLQ01q5dS1xcHIGBgYYORwiDK2joAIThPHnyhLS0NOzs7PTS7ezsuHr1qoGiEkKIrKHRaOjXrx/Vq1endOnShg5H5EMXLlwgMDCQxMREzM3N2bx5M/7+/oYOSwiDkwGGEEKIfKlXr15cvHiRY8eOGToUkU/5+Phw9uxZoqOj2bBhAx07duTw4cMyyBDvPBlgvMOKFi1KgQIFePjwoV76w4cPsbe3N1BUQgjx9nr37s327ds5cuQIJUqUMHQ4Ip8yNjbG09MTgIoVK/L3338za9YsfvrpJwNHJoRhyRqMd5ixsTEVK1Zk//79ujSNRsP+/fvlHlIhRJ6k1Wrp3bs3mzdv5sCBA7i5uRk6JPEO0Wg0JCUlGToMIQxOZjDecf3796djx45UqlSJKlWqMHPmTOLi4ujcubOhQxP5zNOnTwkLC9O9vnnzJmfPnsXW1hYXFxcDRibyk169erF69Wq2bt2KhYUFERERAFhZWVGoUCEDRyfykyFDhtCgQQNcXFyIjY1l9erVHDp0iN27dxs6NCEMTrapFcydO5dp06YRERFBQEAAs2fPpmrVqoYOS+Qzhw4d4sMPP3wpvWPHjixdujTnAxL5kkKhyDB9yZIldOrUKWeDEfla165d2b9/P+Hh4VhZWVG2bFkGDRpEnTp1DB2aEAYnAwwhhBBCCCFElpE1GEIIIYQQQogsIwMMIYQQQgghRJaRAYYQQgghhBAiy8gAQwghhBBCCJFlZIAhhBBCCCGEyDIywBBCCCGEEEJkGRlgCCGEEEIIIbKMDDCEEEIIIYQQWUYGGEIIkQcoFAq2bNli6DCEEEKI15IBhhBCZLNOnTqhUChQKBQYGRnh5ubGwIEDSUxMNHRoQgghRJYraOgAhBDiXVC/fn2WLFlCSkoKp06domPHjigUCqZMmWLo0IQQQogsJTMYQgiRA0xMTLC3t8fZ2ZmmTZtSu3Zt9u7dC4CrqyszZ87Uyx8QEMDo0aMzre/u3bu0bt0aa2trbG1tadKkCbdu3dKdP3ToEFWqVMHMzAxra2uqV6/O7du3s6FlQgghhD4ZYAghRA67ePEix48fx9jY+D+VT0lJoV69elhYWHD06FF+//13zM3NqV+/PsnJyaSmptK0aVOCgoI4f/48J06c4Msvv0ShUGRxS4QQQoiXyS1SQgiRA7Zv3465uTmpqakkJSWhVCqZO3fuf6pr3bp1aDQaFi1apBs0LFmyBGtraw4dOkSlSpWIjo7mk08+wcPDAwA/P78sa4sQQgjxKjLAEEKIHPDhhx+yYMEC4uLimDFjBgULFqRFixb/qa5z584RFhaGhYWFXnpiYiLXr1+nbt26dOrUiXr16lGnTh1q165N69atcXBwyIqmCCGEEK8kt0gJIUQOMDMzw9PTk3LlyrF48WL+/PNPfvnlFwCUSiVarVYvf0pKSqZ1PX36lIoVK3L27Fm9IzQ0lM8++wxIn9E4ceIE7733HuvWrcPb25s//vgj+xoohBBCPCMDDCGEyGFKpZKhQ4cyfPhwEhISKFasGOHh4brzMTEx3Lx5M9PyFSpU4Nq1axQvXhxPT0+9w8rKSpevfPnyDBkyhOPHj1O6dGlWr16dre0SQgghQAYYQghhEK1ataJAgQLMmzePWrVqsWLFCo4ePcqFCxfo2LEjBQoUyLRsu3btKFq0KE2aNOHo0aPcvHmTQ4cO8fXXX3Pv3j1u3rzJkCFDOHHiBLdv32bPnj1cu3ZN1mEIIYTIEbIGQwghDKBgwYL07t2bqVOncu3aNW7evMknn3yClZUV48aNe+UMRuHChTly5AiDBg2iefPmxMbG4uTkxEcffYSlpSUJCQlcvXqVZcuWERkZiYODA7169aJ79+452EIhhBDvKoX23zf+CiGEEEIIIcR/JLdICSGEEEIIIbKMDDCEEEIIIYQQWUYGGEIIIYQQQogsIwMMIYQQQgghRJaRAYYQQgghhBAiy8gAQwghhBBCCJFlZIAhhBBCCCGEyDIywBBCCCGEEEJkGRlgCCGEEEIIIbKMDDCEEEIIIYQQWUYGGEIIIYQQQogs8z/SGrvkSA4HagAAAABJRU5ErkJggg==",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"original_rule_matrix = np.reshape(\n",
" rule_manager.rule_matrix, (n_rules, n_features_input * n_membership_fcts)\n",
").T\n",
"plot_heatmap(\n",
" original_rule_matrix,\n",
" xlabel=\"Rules\",\n",
" ylabel=\"Member Outputs\",\n",
" title=\"Original Rule Matrix\",\n",
" annot=True,\n",
" fmt=\".2g\",\n",
" center=0.0,\n",
" figsize=(10, 10),\n",
")\n",
"\n",
"trained_rule_matrix = np.reshape(\n",
" fuzzy[1].conditions.weight.detach().numpy(),\n",
" (n_rules, n_features_input * n_membership_fcts),\n",
").T\n",
"\n",
"plot_heatmap(\n",
" trained_rule_matrix,\n",
" xlabel=\"Rules\",\n",
" ylabel=\"Member Outputs\",\n",
" title=\"Rule Matrix\",\n",
" annot=True,\n",
" fmt=\".2g\",\n",
" center=0.0,\n",
" figsize=(10, 10),\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"A good rule will keep the already high weights high or even increase them and lower all other, zero initialized weights. This would only strengthen the relation this rule already implied. The second rule is an example. \n",
"If on the other hand a rule is changing very much, hence loosing its original meaning, it probably is worth rechecking that rule, altering it or even discarding it all together. The first rule is an example for that. A similar approach can be taken when pruning the weights but allowing the remaining ones to change (`prune_weights=True`, `learn_conditions=True`). Here weak and strong rules can be discovered too."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The `consequences` matrix allows us to see which rules the network finds useful for classifying each class. Each column sums up to one. The entries of the matrix say how much the network trusts the rule for the classification of the column's class. "
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"belief_matrix = fuzzy[1].consequences.weight.detach().numpy().T\n",
"plot_heatmap(\n",
" belief_matrix, xlabel=\"Classes\", ylabel=\"Rules\", title=\"Belief Matrix\", center=0.0\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Here again weak rules can be found. In contrast to before, if a rule is weak in this matrix, it might be correct but does not contribute much for classifying the classes."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3.8.5 ('prosper')",
"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": "3.8.5"
},
"vscode": {
"interpreter": {
"hash": "a604604040b0261c277bc75aa34f15c6f86bb9bc8166d3b0f73ab3af3d1b81ef"
}
}
},
"nbformat": 4,
"nbformat_minor": 4
}