Created
November 27, 2018 18:14
-
-
Save Katba-Caroline/2c98d1b264ca3f6e396cd29dd5610d9f to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| { | |
| "cells": [ | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "# Developing an AI application\n", | |
| "\n", | |
| "Going forward, AI algorithms will be incorporated into more and more everyday applications. For example, you might want to include an image classifier in a smart phone app. To do this, you'd use a deep learning model trained on hundreds of thousands of images as part of the overall application architecture. A large part of software development in the future will be using these types of models as common parts of applications. \n", | |
| "\n", | |
| "In this project, you'll train an image classifier to recognize different species of flowers. You can imagine using something like this in a phone app that tells you the name of the flower your camera is looking at. In practice you'd train this classifier, then export it for use in your application. We'll be using [this dataset](http://www.robots.ox.ac.uk/~vgg/data/flowers/102/index.html) of 102 flower categories, you can see a few examples below. \n", | |
| "\n", | |
| "<img src='assets/Flowers.png' width=500px>\n", | |
| "\n", | |
| "The project is broken down into multiple steps:\n", | |
| "\n", | |
| "* Load and preprocess the image dataset\n", | |
| "* Train the image classifier on your dataset\n", | |
| "* Use the trained classifier to predict image content\n", | |
| "\n", | |
| "We'll lead you through each part which you'll implement in Python.\n", | |
| "\n", | |
| "When you've completed this project, you'll have an application that can be trained on any set of labeled images. Here your network will be learning about flowers and end up as a command line application. But, what you do with your new skills depends on your imagination and effort in building a dataset. For example, imagine an app where you take a picture of a car, it tells you what the make and model is, then looks up information about it. Go build your own dataset and make something new.\n", | |
| "\n", | |
| "First up is importing the packages you'll need. It's good practice to keep all the imports at the beginning of your code. As you work through this notebook and find you need to import a package, make sure to add the import up here.\n", | |
| "\n", | |
| "Please make sure if you are running this notebook in the workspace that you have chosen GPU rather than CPU mode." | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "##### Most of the code in this notebook was based on various code examples available online including the pytorch documentation tutorials. Here are some links: [link 1 ](https://pytorch.org/tutorials/beginner/transfer_learning_tutorial.html#load-data), [link 2](https://github.com/fotisk07/Image-Classifier/blob/master/Image%20Classifier%20Project.ipynb), [link 3](https://medium.com/@josh_2774/deep-learning-with-pytorch-9574e74d17ad)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 30, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "#Imports here\n", | |
| "\n", | |
| "import torch\n", | |
| "import torch.nn as nn\n", | |
| "import torch.optim as optim\n", | |
| "import torch.nn.functional as F\n", | |
| "from torch.autograd import Variable\n", | |
| "from torch.optim import lr_scheduler\n", | |
| "from collections import OrderedDict\n", | |
| "import numpy as np\n", | |
| "import torchvision\n", | |
| "from torchvision import datasets, models, transforms\n", | |
| "import matplotlib.pyplot as plt\n", | |
| "import time\n", | |
| "import os\n", | |
| "import copy\n", | |
| "\n", | |
| "import json\n", | |
| "from workspace_utils import active_session\n", | |
| "\n", | |
| "from PIL import Image\n", | |
| "from __future__ import print_function, division\n", | |
| "plt.ion() # interactive mode" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Load the data\n", | |
| "\n", | |
| "Here you'll use `torchvision` to load the data ([documentation](http://pytorch.org/docs/0.3.0/torchvision/index.html)). The data should be included alongside this notebook, otherwise you can [download it here](https://s3.amazonaws.com/content.udacity-data.com/nd089/flower_data.tar.gz). The dataset is split into three parts, training, validation, and testing. **For the training, you'll want to apply transformations such as random scaling, cropping, and flipping.** This will help the network generalize leading to better performance. You'll also need to **make sure the input data is resized to 224x224 pixels as required by the pre-trained networks.**\n", | |
| "\n", | |
| "**The validation and testing sets are used to measure the model's performance on data it hasn't seen yet. For this you don't want any scaling or rotation transformations, but you'll need to resize then crop the images to the appropriate size.**\n", | |
| "\n", | |
| "The pre-trained networks you'll use were trained on the ImageNet dataset where each color channel was normalized separately. For all three sets you'll need to normalize the means and standard deviations of the images to what the network expects. For the means, it's `[0.485, 0.456, 0.406]` and for the standard deviations `[0.229, 0.224, 0.225]`, calculated from the ImageNet images. These values will shift each color channel to be centered at 0 and range from -1 to 1.\n", | |
| " " | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 2, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "data_dir = 'flowers'\n", | |
| "train_dir = data_dir + '/train'\n", | |
| "valid_dir = data_dir + '/valid'\n", | |
| "test_dir = data_dir + '/test'" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 3, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "cuda\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "# TODO: Define your transforms for the training, validation, and testing sets\n", | |
| "### data_transforms = \n", | |
| "# For the TRAINING, Apply transformations such as random scaling, cropping, and flipping.\n", | |
| "## Make sure the input data is resized to 224x224 pixels.\n", | |
| "# **NOTE** this class does both scaling and cropping at the same time\n", | |
| "## transforms.RandomResizedCrop(size, scale=(0.08, 1.0), ratio=(0.75, 1.333), interpolation=2)\n", | |
| "\n", | |
| "data_transforms = {\n", | |
| " 'train': transforms.Compose([\n", | |
| " transforms.RandomRotation(35),\n", | |
| " transforms.RandomResizedCrop(224),\n", | |
| " transforms.RandomHorizontalFlip(),\n", | |
| " transforms.ToTensor(),\n", | |
| " transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n", | |
| " ]),\n", | |
| "\n", | |
| " # The validation and testing sets are used to measure the model's performance\n", | |
| " ## on data it hasn't seen yet. For this you don't want any scaling or\n", | |
| " ## rotation transformations,but you'll need to resize then crop the images to the appropriate size.\n", | |
| " \n", | |
| " 'valid': transforms.Compose([\n", | |
| " transforms.Resize(256),\n", | |
| " transforms.CenterCrop(224),\n", | |
| " transforms.ToTensor(),\n", | |
| " transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n", | |
| " ]),\n", | |
| " \n", | |
| " 'test': transforms.Compose([\n", | |
| " transforms.Resize(256),\n", | |
| " transforms.CenterCrop(224),\n", | |
| " transforms.ToTensor(),\n", | |
| " transforms.Normalize([0.485, 0.456, 0.406], \n", | |
| " [0.229, 0.224, 0.225])\n", | |
| " ]),\n", | |
| "}\n", | |
| "\n", | |
| "# TODO: Load the datasets with ImageFolder\n", | |
| "data_dir = 'flowers'\n", | |
| "image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x),\n", | |
| " data_transforms[x])\n", | |
| " for x in ['train', 'valid', 'test']}\n", | |
| "\n", | |
| "# TODO: Using the image datasets and the trainforms, define the dataloaders\n", | |
| "\n", | |
| "dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=4,\n", | |
| " shuffle=True, num_workers=4)\n", | |
| " for x in ['train', 'valid','test']}\n", | |
| "\n", | |
| "\n", | |
| "dataset_sizes = {x: len(image_datasets[x])\n", | |
| " for x in ['train', 'valid', 'test']}\n", | |
| "\n", | |
| "class_names = image_datasets['train'].classes\n", | |
| "\n", | |
| "#can use this code to change to CPU if Cuda is not available\n", | |
| "##device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", | |
| "\n", | |
| "#Check cuda\n", | |
| "device= torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n", | |
| " \n", | |
| "print(device)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "# Explore the data a little" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 4, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "Classes: \n", | |
| "['1', '10', '100', '101', '102', '11', '12', '13', '14', '15', '16', '17', '18', '19', '2', '20', '21', '22', '23', '24', '25', '26', '27', '28', '29', '3', '30', '31', '32', '33', '34', '35', '36', '37', '38', '39', '4', '40', '41', '42', '43', '44', '45', '46', '47', '48', '49', '5', '50', '51', '52', '53', '54', '55', '56', '57', '58', '59', '6', '60', '61', '62', '63', '64', '65', '66', '67', '68', '69', '7', '70', '71', '72', '73', '74', '75', '76', '77', '78', '79', '8', '80', '81', '82', '83', '84', '85', '86', '87', '88', '89', '9', '90', '91', '92', '93', '94', '95', '96', '97', '98', '99']\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "print(\"Classes: \")\n", | |
| "class_names = image_datasets['train'].classes\n", | |
| "print(image_datasets['train'].classes)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 10, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<matplotlib.figure.Figure at 0x7fcbc2ffe208>" | |
| ] | |
| }, | |
| "metadata": { | |
| "needs_background": "light" | |
| }, | |
| "output_type": "display_data" | |
| } | |
| ], | |
| "source": [ | |
| "def imshow(inp, title=None):\n", | |
| " \"\"\"Imshow for Tensor.\"\"\"\n", | |
| " inp = inp.numpy().transpose((1, 2, 0))\n", | |
| " mean = np.array([0.485, 0.456, 0.406])\n", | |
| " std = np.array([0.229, 0.224, 0.225])\n", | |
| " inp = std * inp + mean\n", | |
| " inp = np.clip(inp, 0, 1)\n", | |
| " plt.imshow(inp)\n", | |
| " if title is not None:\n", | |
| " plt.title(title)\n", | |
| " plt.pause(0.001) # pause a bit so that plots are updated\n", | |
| "\n", | |
| "\n", | |
| "# Get a batch of training data\n", | |
| "inputs, classes = next(iter(dataloaders['train']))\n", | |
| "\n", | |
| "# Make a grid from batch\n", | |
| "out = torchvision.utils.make_grid(inputs)\n", | |
| "\n", | |
| "flower_labels = list(cat_to_name.values())\n", | |
| "\n", | |
| "imshow(out, title=[labels[x] for x in classes])\n", | |
| "\n", | |
| "#imshow(out, title=[class_names[x] for x in classes])" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "### Label mapping\n", | |
| "\n", | |
| "You'll also need to load in a mapping from category label to category name. You can find this in the file `cat_to_name.json`. It's a JSON object which you can read in with the [`json` module](https://docs.python.org/2/library/json.html). This will give you a dictionary mapping the integer encoded categories to the actual names of the flowers." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 6, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "import json\n", | |
| "\n", | |
| "with open('cat_to_name.json', 'r') as f:\n", | |
| " cat_to_name = json.load(f)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "# Building and training the classifier\n", | |
| "\n", | |
| "Now that the data is ready, it's time to build and train the classifier. As usual, you should use one of the pretrained models from `torchvision.models` to get the image features. Build and train a new feed-forward classifier using those features.\n", | |
| "\n", | |
| "We're going to leave this part up to you. Refer to [the rubric](https://review.udacity.com/#!/rubrics/1663/view) for guidance on successfully completing this section. Things you'll need to do:\n", | |
| "\n", | |
| "* Load a [pre-trained network](http://pytorch.org/docs/master/torchvision/models.html) (If you need a starting point, the VGG networks work great and are straightforward to use)\n", | |
| "* Define a new, untrained feed-forward network as a classifier, using ReLU activations and dropout\n", | |
| "* Train the classifier layers using backpropagation using the pre-trained network to get the features\n", | |
| "* Track the loss and accuracy on the validation set to determine the best hyperparameters\n", | |
| "\n", | |
| "We've left a cell open for you below, but use as many as you need. Our advice is to break the problem up into smaller parts you can run separately. Check that each part is doing what you expect, then move on to the next. You'll likely find that as you work through each part, you'll need to go back and modify your previous code. This is totally normal!\n", | |
| "\n", | |
| "When training make sure you're updating only the weights of the feed-forward network. You should be able to get the validation accuracy above 70% if you build everything right. Make sure to try different hyperparameters (learning rate, units in the classifier, epochs, etc) to find the best model. Save those hyperparameters to use as default values in the next part of the project.\n", | |
| "\n", | |
| "One last important tip if you're using the workspace to run your code: To avoid having your workspace disconnect during the long-running tasks in this notebook, please read in the earlier page in this lesson called Intro to GPU Workspaces about Keeping Your Session Active. You'll want to include code from the workspace_utils.py module." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 8, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stderr", | |
| "output_type": "stream", | |
| "text": [ | |
| "Downloading: \"https://download.pytorch.org/models/vgg16-397923af.pth\" to /root/.torch/models/vgg16-397923af.pth\n", | |
| "100%|██████████| 553433881/553433881 [00:05<00:00, 94281079.32it/s] \n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "structures = {\"vgg16\":25088,\n", | |
| " \"densenet161\": 2208,\n", | |
| " \"SqueezeNet\":512}\n", | |
| "\n", | |
| "\n", | |
| "def my_model(structure='vgg16', dropout=0.2,\n", | |
| " hidden_layer=1024,lr = 0.001):\n", | |
| " \n", | |
| " \n", | |
| " if structure == \"vgg16\":\n", | |
| " model = models.vgg16(pretrained=True)\n", | |
| " elif structure == \"densenet161\":\n", | |
| " model = models.densenet161(pretrained=True)\n", | |
| " elif structure == \"SqueezeNet\":\n", | |
| " model = models.SqueezNet1_0(pretrained=True)\n", | |
| " else:\n", | |
| " print(\"wrong model, try vgg16, densenet161, or SqueezeNet\")\n", | |
| " \n", | |
| " model = models.__dict__[structure](pretrained=True)\n", | |
| " \n", | |
| " for param in model.parameters():\n", | |
| " param.requires_grad = False\n", | |
| " \n", | |
| " classifier = nn.Sequential(OrderedDict([\n", | |
| " ('fc1', nn.Linear(structures[structure], hidden_layer)),\n", | |
| " ('relu', nn.ReLU()),\n", | |
| " ('dropout', nn.Dropout(dropout)),\n", | |
| " ('fc2', nn.Linear(hidden_layer, 102)),\n", | |
| " ('output', nn.LogSoftmax(dim=1))\n", | |
| " ]))\n", | |
| " model.classifier = classifier\n", | |
| " \n", | |
| " criterion = nn.NLLLoss()\n", | |
| " \n", | |
| " optimizer = optim.Adam(model.classifier.parameters(), lr)\n", | |
| " \n", | |
| " exp_lr_scheduler = lr_scheduler.StepLR(optimizer, step_size=7, gamma=0.1)\n", | |
| " \n", | |
| " model = model.to(device)\n", | |
| " \n", | |
| " return model, optimizer, criterion, exp_lr_scheduler\n", | |
| "model,optimizer,criterion,exp_lr_scheduler = my_model('vgg16')" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 9, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "def train_model(model, criterion, optimizer, scheduler,\n", | |
| " num_epochs=15, device = 'cuda'):\n", | |
| " since = time.time()\n", | |
| "\n", | |
| " best_model_wts = copy.deepcopy(model.state_dict())\n", | |
| " best_acc = 0.0\n", | |
| "\n", | |
| " for epoch in range(num_epochs):\n", | |
| " print('Epoch {}/{}'.format(epoch, num_epochs - 1))\n", | |
| " print('-' * 10)\n", | |
| "\n", | |
| " # Each epoch has a training and validation phase\n", | |
| " for phase in ['train', 'valid']:\n", | |
| " if phase == 'train':\n", | |
| " scheduler.step()\n", | |
| " model.train() # Set model to training mode\n", | |
| " else:\n", | |
| " model.eval() # Set model to evaluate mode\n", | |
| "\n", | |
| " running_loss = 0.0\n", | |
| " running_corrects = 0\n", | |
| "\n", | |
| " # Iterate over data.\n", | |
| " for inputs, labels in dataloaders[phase]:\n", | |
| " inputs = inputs.to(device)\n", | |
| " labels = labels.to(device)\n", | |
| "\n", | |
| " # zero the parameter gradients\n", | |
| " optimizer.zero_grad()\n", | |
| "\n", | |
| " # forward\n", | |
| " # track history if only in train\n", | |
| " with torch.set_grad_enabled(phase == 'train'):\n", | |
| " outputs = model(inputs)\n", | |
| " _, preds = torch.max(outputs, 1)\n", | |
| " loss = criterion(outputs, labels)\n", | |
| "\n", | |
| " # backward + optimize only if in training phase\n", | |
| " if phase == 'train':\n", | |
| " loss.backward()\n", | |
| " optimizer.step()\n", | |
| "\n", | |
| " # statistics\n", | |
| " running_loss += loss.item() * inputs.size(0)\n", | |
| " running_corrects += torch.sum(preds == labels.data)\n", | |
| "\n", | |
| " epoch_loss = running_loss / dataset_sizes[phase]\n", | |
| " epoch_acc = running_corrects.double() / dataset_sizes[phase]\n", | |
| "\n", | |
| " print('{} Loss: {:.4f} Acc: {:.4f}'.format(\n", | |
| " phase, epoch_loss, epoch_acc))\n", | |
| "\n", | |
| " # deep copy the model\n", | |
| " if phase == 'valid' and epoch_acc > best_acc:\n", | |
| " best_acc = epoch_acc\n", | |
| " best_model_wts = copy.deepcopy(model.state_dict())\n", | |
| "\n", | |
| " print()\n", | |
| "\n", | |
| " time_elapsed = time.time() - since\n", | |
| " print('Training complete in {:.0f}m {:.0f}s'.format(\n", | |
| " time_elapsed // 60, time_elapsed % 60))\n", | |
| " print('Best val Acc: {:4f}'.format(best_acc))\n", | |
| "\n", | |
| " # load best model weights\n", | |
| " model.load_state_dict(best_model_wts)\n", | |
| " return model" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 16, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "#Visualizing the model predictions\n", | |
| "##Generic function to display predictions for a few images\n", | |
| "\n", | |
| "def visualize_model(model, num_images=6):\n", | |
| " was_training = model.training\n", | |
| " model.eval()\n", | |
| " images_so_far = 0\n", | |
| " fig = plt.figure()\n", | |
| "\n", | |
| " with torch.no_grad():\n", | |
| " for i, (inputs, labels) in enumerate(dataloaders['test']):\n", | |
| " inputs = inputs.to(device)\n", | |
| " labels = labels.to(device)\n", | |
| "\n", | |
| " outputs = model(inputs)\n", | |
| " _, preds = torch.max(outputs, 1)\n", | |
| "\n", | |
| " for j in range(inputs.size()[0]):\n", | |
| " images_so_far += 1\n", | |
| " ax = plt.subplot(num_images//2, 2, images_so_far)\n", | |
| " ax.axis('off')\n", | |
| " ax.set_title('predicted: {}'.format(class_names[preds[j]]))\n", | |
| " imshow(inputs.cpu().data[j])\n", | |
| "\n", | |
| " if images_so_far == num_images:\n", | |
| " model.train(mode=was_training)\n", | |
| " return model.train(mode=was_training)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 12, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "Epoch 0/14\n", | |
| "----------\n", | |
| "train Loss: 3.4426 Acc: 0.2297\n", | |
| "valid Loss: 2.0760 Acc: 0.4731\n", | |
| "\n", | |
| "Epoch 1/14\n", | |
| "----------\n", | |
| "train Loss: 2.5934 Acc: 0.3723\n", | |
| "valid Loss: 1.6763 Acc: 0.5856\n", | |
| "\n", | |
| "Epoch 2/14\n", | |
| "----------\n", | |
| "train Loss: 2.4234 Acc: 0.4136\n", | |
| "valid Loss: 1.5836 Acc: 0.5770\n", | |
| "\n", | |
| "Epoch 3/14\n", | |
| "----------\n", | |
| "train Loss: 2.2979 Acc: 0.4461\n", | |
| "valid Loss: 1.5401 Acc: 0.6516\n", | |
| "\n", | |
| "Epoch 4/14\n", | |
| "----------\n", | |
| "train Loss: 2.2976 Acc: 0.4476\n", | |
| "valid Loss: 1.6078 Acc: 0.6296\n", | |
| "\n", | |
| "Epoch 5/14\n", | |
| "----------\n", | |
| "train Loss: 2.2053 Acc: 0.4675\n", | |
| "valid Loss: 1.5239 Acc: 0.6589\n", | |
| "\n", | |
| "Epoch 6/14\n", | |
| "----------\n", | |
| "train Loss: 2.2088 Acc: 0.4728\n", | |
| "valid Loss: 1.3586 Acc: 0.6809\n", | |
| "\n", | |
| "Epoch 7/14\n", | |
| "----------\n", | |
| "train Loss: 1.8875 Acc: 0.5357\n", | |
| "valid Loss: 1.1614 Acc: 0.7433\n", | |
| "\n", | |
| "Epoch 8/14\n", | |
| "----------\n", | |
| "train Loss: 1.7376 Acc: 0.5653\n", | |
| "valid Loss: 1.1375 Acc: 0.7592\n", | |
| "\n", | |
| "Epoch 9/14\n", | |
| "----------\n", | |
| "train Loss: 1.6614 Acc: 0.5859\n", | |
| "valid Loss: 1.1010 Acc: 0.7787\n", | |
| "\n", | |
| "Epoch 10/14\n", | |
| "----------\n", | |
| "train Loss: 1.6617 Acc: 0.5900\n", | |
| "valid Loss: 1.0654 Acc: 0.7775\n", | |
| "\n", | |
| "Epoch 11/14\n", | |
| "----------\n", | |
| "train Loss: 1.5164 Acc: 0.6216\n", | |
| "valid Loss: 1.0254 Acc: 0.7873\n", | |
| "\n", | |
| "Epoch 12/14\n", | |
| "----------\n", | |
| "train Loss: 1.4734 Acc: 0.6316\n", | |
| "valid Loss: 0.9859 Acc: 0.7958\n", | |
| "\n", | |
| "Epoch 13/14\n", | |
| "----------\n", | |
| "train Loss: 1.4067 Acc: 0.6439\n", | |
| "valid Loss: 0.9503 Acc: 0.8081\n", | |
| "\n", | |
| "Epoch 14/14\n", | |
| "----------\n", | |
| "train Loss: 1.3679 Acc: 0.6532\n", | |
| "valid Loss: 0.9396 Acc: 0.8081\n", | |
| "\n", | |
| "Training complete in 123m 26s\n", | |
| "Best val Acc: 0.808068\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "from workspace_utils import active_session\n", | |
| "\n", | |
| "with active_session():\n", | |
| " model = train_model(model, criterion, optimizer, \n", | |
| " exp_lr_scheduler, num_epochs=15, device='cuda')" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 17, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<matplotlib.figure.Figure at 0x7fcbc2b5a400>" | |
| ] | |
| }, | |
| "metadata": { | |
| "needs_background": "light" | |
| }, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<matplotlib.figure.Figure at 0x7fcbc2818c18>" | |
| ] | |
| }, | |
| "metadata": { | |
| "needs_background": "light" | |
| }, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<matplotlib.figure.Figure at 0x7fcbc2f1f6a0>" | |
| ] | |
| }, | |
| "metadata": { | |
| "needs_background": "light" | |
| }, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<matplotlib.figure.Figure at 0x7fcbc2cf1be0>" | |
| ] | |
| }, | |
| "metadata": { | |
| "needs_background": "light" | |
| }, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<matplotlib.figure.Figure at 0x7fcbc2b80da0>" | |
| ] | |
| }, | |
| "metadata": { | |
| "needs_background": "light" | |
| }, | |
| "output_type": "display_data" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "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\n", | |
| "text/plain": [ | |
| "<matplotlib.figure.Figure at 0x7fcbc2851860>" | |
| ] | |
| }, | |
| "metadata": { | |
| "needs_background": "light" | |
| }, | |
| "output_type": "display_data" | |
| } | |
| ], | |
| "source": [ | |
| "visualize_model(model)\n", | |
| "\n", | |
| "plt.ioff()\n", | |
| "plt.show()" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": null, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Testing your network\n", | |
| "\n", | |
| "It's good practice to test your trained network on test data, images the network has never seen either in training or validation. This will give you a good estimate for the model's performance on completely new images. Run the test images through the network and measure the accuracy, the same way you did validation. You should be able to reach around 70% accuracy on the test set if the model has been trained well." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 18, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "Accuracy of the network on the test images: 78 %\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "# TODO: Do validation on the test set\n", | |
| "def check_accuracy_on_test(model, data, cuda=False):\n", | |
| " model.eval()\n", | |
| " model.to(device='cuda') \n", | |
| " \n", | |
| " correct = 0\n", | |
| " total = 0\n", | |
| " \n", | |
| " with torch.no_grad():\n", | |
| " for data in (dataloaders[data]):\n", | |
| " images, labels = data\n", | |
| " #images, labels = images.to('cuda'), labels.to('cuda')\n", | |
| " images, labels = images.to(device), labels.to(device)\n", | |
| " outputs = model(images)\n", | |
| " _, predicted = torch.max(outputs.data, 1)\n", | |
| " total += labels.size(0)\n", | |
| " correct += (predicted == labels).sum().item()\n", | |
| "\n", | |
| " print('Accuracy of the network on the test images: %d %%' % (100 * correct / total))\n", | |
| " \n", | |
| "check_accuracy_on_test(model, 'test', True)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Save the checkpoint\n", | |
| "\n", | |
| "Now that your network is trained, save the model so you can load it later for making predictions. You probably want to save other things such as the mapping of classes to indices which you get from one of the image datasets: `image_datasets['train'].class_to_idx`. You can attach this to the model as an attribute which makes inference easier later on.\n", | |
| "\n", | |
| "```model.class_to_idx = image_datasets['train'].class_to_idx```\n", | |
| "\n", | |
| "Remember that you'll want to completely rebuild the model later so you can use it for inference. Make sure to include any information you need in the checkpoint. If you want to load the model and keep training, you'll want to save the number of epochs as well as the optimizer state, `optimizer.state_dict`. You'll likely want to use this trained model in the next part of the project, so best to save it now." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 19, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "# TODO: Save the checkpoint\n", | |
| "\n", | |
| "model.class_to_idx = image_datasets['train'].class_to_idx\n", | |
| "model.cpu()\n", | |
| "torch.save({'structure': 'vgg16',\n", | |
| " 'hidden_layer': 1024,\n", | |
| " 'state_dict': model.state_dict(), \n", | |
| " 'class_to_idx': model.class_to_idx}, \n", | |
| " 'classifier.pth')" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## Loading the checkpoint\n", | |
| "\n", | |
| "At this point it's good to write a function that can load a checkpoint and rebuild the model. That way you can come back to this project and keep working on it without having to retrain the network." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 20, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "# TODO: Write a function that loads a checkpoint and rebuilds the model\n", | |
| "\n", | |
| "def load_model(checkpoint_path):\n", | |
| " chkpt = torch.load(checkpoint_path)\n", | |
| " structure = chkpt['structure']\n", | |
| " hidden_layer = chkpt['hidden_layer']\n", | |
| " model,_,_,_ = my_model(structure, 0.2, hidden_layer, 0.001)\n", | |
| " model.class_to_idx = chkpt['class_to_idx']\n", | |
| " model.load_state_dict(chkpt['state_dict'])\n", | |
| " \n", | |
| " \n", | |
| " return model" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 21, | |
| "metadata": { | |
| "scrolled": true | |
| }, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "VGG(\n", | |
| " (features): Sequential(\n", | |
| " (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (1): ReLU(inplace)\n", | |
| " (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (3): ReLU(inplace)\n", | |
| " (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n", | |
| " (5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (6): ReLU(inplace)\n", | |
| " (7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (8): ReLU(inplace)\n", | |
| " (9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n", | |
| " (10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (11): ReLU(inplace)\n", | |
| " (12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (13): ReLU(inplace)\n", | |
| " (14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (15): ReLU(inplace)\n", | |
| " (16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n", | |
| " (17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (18): ReLU(inplace)\n", | |
| " (19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (20): ReLU(inplace)\n", | |
| " (21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (22): ReLU(inplace)\n", | |
| " (23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n", | |
| " (24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (25): ReLU(inplace)\n", | |
| " (26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (27): ReLU(inplace)\n", | |
| " (28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n", | |
| " (29): ReLU(inplace)\n", | |
| " (30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n", | |
| " )\n", | |
| " (classifier): Sequential(\n", | |
| " (fc1): Linear(in_features=25088, out_features=1024, bias=True)\n", | |
| " (relu): ReLU()\n", | |
| " (dropout): Dropout(p=0.2)\n", | |
| " (fc2): Linear(in_features=1024, out_features=102, bias=True)\n", | |
| " (output): LogSoftmax()\n", | |
| " )\n", | |
| ")\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "model = load_model('classifier.pth')\n", | |
| "print(model)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 22, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "Accuracy of the network on the test images: 78 %\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "check_accuracy_on_test(model, 'test', True)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "# Inference for classification\n", | |
| "\n", | |
| "Now you'll write a function to use a trained network for inference. That is, you'll pass an image into the network and predict the class of the flower in the image. Write a function called `predict` that takes an image and a model, then returns the top $K$ most likely classes along with the probabilities. It should look like \n", | |
| "\n", | |
| "```python\n", | |
| "probs, classes = predict(image_path, model)\n", | |
| "print(probs)\n", | |
| "print(classes)\n", | |
| "> [ 0.01558163 0.01541934 0.01452626 0.01443549 0.01407339]\n", | |
| "> ['70', '3', '45', '62', '55']\n", | |
| "```\n", | |
| "\n", | |
| "First you'll need to handle processing the input image such that it can be used in your network. \n", | |
| "\n", | |
| "## Image Preprocessing\n", | |
| "\n", | |
| "You'll want to use `PIL` to load the image ([documentation](https://pillow.readthedocs.io/en/latest/reference/Image.html)). It's best to write a function that preprocesses the image so it can be used as input for the model. This function should process the images in the same manner used for training. \n", | |
| "\n", | |
| "First, resize the images where the shortest side is 256 pixels, keeping the aspect ratio. This can be done with the [`thumbnail`](http://pillow.readthedocs.io/en/3.1.x/reference/Image.html#PIL.Image.Image.thumbnail) or [`resize`](http://pillow.readthedocs.io/en/3.1.x/reference/Image.html#PIL.Image.Image.thumbnail) methods. Then you'll need to crop out the center 224x224 portion of the image.\n", | |
| "\n", | |
| "Color channels of images are typically encoded as integers 0-255, but the model expected floats 0-1. You'll need to convert the values. It's easiest with a Numpy array, which you can get from a PIL image like so `np_image = np.array(pil_image)`.\n", | |
| "\n", | |
| "As before, the network expects the images to be normalized in a specific way. For the means, it's `[0.485, 0.456, 0.406]` and for the standard deviations `[0.229, 0.224, 0.225]`. You'll want to subtract the means from each color channel, then divide by the standard deviation. \n", | |
| "\n", | |
| "And finally, PyTorch expects the color channel to be the first dimension but it's the third dimension in the PIL image and Numpy array. You can reorder dimensions using [`ndarray.transpose`](https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.ndarray.transpose.html). The color channel needs to be first and retain the order of the other two dimensions." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 23, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "def process_image2(image):\n", | |
| " ''' \n", | |
| " Scales, crops, and normalizes a PIL image for a PyTorch \n", | |
| " model, returns an Numpy array\n", | |
| " '''\n", | |
| " # Open the image & resize using thumbnail\n", | |
| " img = Image.open(image)\n", | |
| " img.load()\n", | |
| " \n", | |
| "\n", | |
| " #Resizing\n", | |
| " if img.size[0] > img.size[1]:\n", | |
| " img.thumbnail((10000, 256))\n", | |
| " else:\n", | |
| " img.thumbnail((256, 10000))\n", | |
| " \n", | |
| "\n", | |
| " #Cropping\n", | |
| " size = img.size\n", | |
| " img = img.crop((size[0]//2 -(224/2),\n", | |
| " size[1]//2 - (224/2),\n", | |
| " size[0]//2 +(224/2),\n", | |
| " size[1]//2 + (224/2) \n", | |
| " ))\n", | |
| " \n", | |
| " #Normalizing\n", | |
| " img = np.array(img)/255\n", | |
| " mean = np.array([0.485, 0.456, 0.406]) \n", | |
| " std = np.array([0.229, 0.224, 0.225]) \n", | |
| " img = (img - mean)/std\n", | |
| " \n", | |
| " img = img.transpose((2, 0, 1))\n", | |
| " \n", | |
| " return img\n", | |
| " " | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "To check your work, the function below converts a PyTorch tensor and displays it in the notebook. If your `process_image` function works, running the output through this function should return the original image (except for the cropped out portions)." | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 24, | |
| "metadata": {}, | |
| "outputs": [], | |
| "source": [ | |
| "def imshow(image, ax=None, title=None):\n", | |
| " if ax is None:\n", | |
| " fig, ax = plt.subplots()\n", | |
| " \n", | |
| " # PyTorch tensors assume the color channel is first\n", | |
| " # but matplotlib assumes is the third dimension\n", | |
| " image = image.transpose((1, 2, 0))\n", | |
| " \n", | |
| " # Undo preprocessing\n", | |
| " mean = np.array([0.485, 0.456, 0.406])\n", | |
| " std = np.array([0.229, 0.224, 0.225])\n", | |
| " image = std * image + mean\n", | |
| " \n", | |
| " # Image needs to be clipped between 0 and 1\n", | |
| " image = np.clip(image, 0, 1)\n", | |
| " \n", | |
| " ax.imshow(image)\n", | |
| " \n", | |
| " return ax\n", | |
| "#imshow(process_image(\"flowers/test/13/image_05775.jpg\"))" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 25, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "data": { | |
| "text/plain": [ | |
| "<matplotlib.axes._subplots.AxesSubplot at 0x7fcbc2a987b8>" | |
| ] | |
| }, | |
| "execution_count": 25, | |
| "metadata": {}, | |
| "output_type": "execute_result" | |
| }, | |
| { | |
| "data": { | |
| "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQUAAAD8CAYAAAB+fLH0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzsvUusbVt63/X7xmPOudbar3POPo/7qLq3bJeLMo5J7EAs8hARBFl0InpJGjSISCsSRHRoIqULQaIBUgKhA4IG0EIGRINISFh5OLGJXeW6LlfdqrqPc89rP9ZrzjnGNz4aY6x1zrWrqBu7rriRzpCW9t5rzb3WnGOO8Y3/9///v7HEzHjdXrfX7XU7NPf/9wm8bq/b6/bFaq+Dwuv2ur1un2qvg8Lr9rq9bp9qr4PC6/a6vW6faq+Dwuv2ur1un2qvg8Lr9rq9bp9qn1tQEJFfEZFvici3ReQ/+rw+53V73V63n2yTz8OnICIeeA/4C8AHwD8E/rKZfeMn/mGv2+v2uv1E2+eFFP4V4Ntm9h0zm4H/AfiLn9NnvW6v2+v2E2zhc3rft4AfvPL3B8Cf+lEHOycWOggBQvBgoCYghlEwMxBAwHvwzhOc4Vx9shRDtZBzfb9iUIohDpwXnBNEDKOiImsPBMS1n+0FoT0HYIIVoaiQ50JORtF2zu1/nHeE4HAOnHNYKageHvDDgZi08zDM6nt55/BeKKWQFbSAyOFYq/8hL9/M2isi0t6N4/Udzl1an9nxgMPr8spx7b1EEKSeE4aZHboc7wTnXX0/V49CBCtQtGDFiCHS9x3eOVQzaZ4we/mRIu18gGyFXApm7RNEjn0hCE7seI4VyR7v2PHsQwhEHxEcOSs5ZbJmfHD4IIiv763t/DQXDKEYODH6PtD1kVKMeVRSUpwUlmeeVBTnPE4c8z4RxdPFyHY/oVb7wLDWv4d7cxivn+rdl+csh35uD/fy98NYPN6LV48/DM7WkXY4oN1UK2DF2svy8p5afe3wuwB5b8/M7P4POb1Ptc8rKPywfvnU9BCRvwb8NQDxcPkVeHh5wslixbzPrG9GRp3oB2M4BzqwUH/GXjlZwrILBAJlFva7xHoP4iNzUUpJdH3PYtkRukLsDVyiWCKbQwFiQTrAQXSeKBFR0JTIyZi2gc2VsHnuef54Iu0zDogeFgsPIiyHFct+oOsc4grT7pb1es/NDYxjveq+93gnFDIhAH4FQJ4yeZ7wYpyeOB5cnjNOW37wFG52mUU3oDkjASDjXK7nXQATutjho7wcC1Zq35ZCLFBiAA85QMkZMDrvMY0Uq5OlFEg508WeGDrMjG3ZgSYW3tO7GoglAsEgQOwhCDA79lvl7vKMP/PLv8yX33qEF+ODxx/yj3/zN1lvRiAQfEfsO7yHcR7ZbLbk2VADdQGRnskKWQ3nPJ5yvJ6iGS0KBoWCFQUHq2HF6fIOQWFzc8NuvOX8Ts/lG+dcnDu2OfPxiz3Pno3MN0aXhWweXQR+8Ze+yq/8yr/Mu+9e8vzxDX/vf/sGv/b3vkGan/Pn/+J9vn97DUPgwek5m/c+4a1widrA33//A8bFAE4BI0TBRyiu4IPhfKGYYVJXjgKY1v6L0ZOL4hz0HYTB0/WCeEAyeMHMyLkFvOBIeSbG2N6p9kEuBtIRxJEzTLvMuMuYCkIk4CnZk1JmnjJ5cmgGU8f1Pxm/91km7+cVFD4AvvTK328DH716gJn9beBvA7go5uQcy0vGnSftJuK4ZbmCe/cHLh6sSF5Zjzv2aUbw7FVJU6ZDCebxvefuSY8abEbYb5VxTKSU6YdIN0VC9MRuSXB1dS5aINVVzIcIFtiv9+xuO7a3yu3Vhmlfb+z5mefOly9Y9gMpZXbbzGazZxz3dSUqHVhms02sty8DghNBxCMOvAjeO5arAec8czcxbgtpnLl354Kf+/rPEoLR/e5jfvu9H6C6R4LgPIgozoM/dGABHw3vpE4eq9cBUhHHYZl3dQkqZnjAIajpcYUpalhWVJTgC1ogpAJW0Y+IJ0aBAJlMzgrOGDPYXIPDxf1THr59l3jicGQ2U+ZqDdMYWKxWZOfBHMkKm5SZiqIGxQpaBJOJXCAbMGfanMLEEEpFfQIijuAjWGS3zow3T4m94rvMxUPh4ZcHzu4VuqJsrrbsyshcqIE0A13mrTcu+GO/8LP89E/9DPfuLvF6Qtd9lyIFNY/IQN/3bKYtujzh3uVdzuSM73znCTiHDx5Ej8i1Is3DsgzOCXOGEAQPZKmIVZwQfcA78KGukKqGoPjDLLR6r6wYWQsgZNUjworBI1hDqQItsDsnYBXZePFY6z8RT4hyXEQ+a/u8gsI/BL4qIl8BPgT+EvBXftTBgkMmYd7MhKXn/NRz91Hgzv1TTi7P8IvIeh5Jz0bEd8xWGINHgGRQUBzK+dkCESjryDiNjDsjzeAl433COc8wDAxLRxGtqYkooesRYNxuuL7asn4KaQeLpfDg/or79wfeeHSf0+Up0z7z9OkNH+gNbJQpZ1SFlAArjKORJsgJxHtA0FJXcxccBRDJOIHoIEnBdXB+dsKDy0v6TriaHO9/8DE3NyN9X/uoHwKLVcSJknIhjYaRQHyb+66upmaoGf4wksQhUvAiiNUgUApQDM2KlVL/30BVsaSQlRA8UsB7YVie0C0Ccx7ZTHvMjKSJgLHsO3px7J6/oJ8W3L54wie//nvE7685HRYsstINHt8bJomLVNgHx7UrlFIn/pRniqtIYcpKTIFS9DDPKGaEWFNLVzJlX0gpEXrj8mTB6f2Bs0dCf6cQTkamXWLyE/HUM0zCNCkUuHy04u0vX9D3jpubW1La89433ufb3/0OKScU4fp2pFtF8t7YbDdcri6JKbKfE2pgh0mtNbXNVmpQ8B7nBUPr6t/6Pvg6Q60YsYs4B1g+wnonHqhBuF0tWhTnKsJKecI5IcaIcw5XElBTzJwVpBCCp+BeyS7rIoAY3jnM2Y9IY394+1yCgpllEfnrwP9OXdz+rpn99o86Xsw4cZkvvTHw1tsXDCeFEgp+0TPKyHq+5sXtlm3O9P2AFYf4mmeqFSwbYrDZ34CDVHq6IbIy2G8K49bYbjKWM0ImdBXuaYuoIewBWN/UiH5554KLL6+4/2DJ5eWCk9PI0C/Ik7C9veH2dsv1es2YlTnDbJkuu8prpJYnuYBZTRxLEXAFzYWCMU47vPdYrsleMdhtd+z3e4TInPcgBR/r+TkPi0XPnYtTchnZbyfmcSRnI4SMSEfwrubNbZLjwkviQ0CcR4piOWEWK2lhdRV2zgOGpYwWw5tDSgs01Hz6/OSMVAa4Mq7GTJCOwRtuUrie2HznMauTE+Juz1sl8uYbj7i4d4eT0xW+c8y6YzvfMGfDWFJKQwo4UsqoGarKmIzbZEzTxJzqDRKpfBMYWTOGkpyxPA28cX/J+cMlZRgZpw373ZabZKQinJ2fIrljvVkz9J6L8xMQ5f33f48Pf/B9Nps1H3zvCR99/5rsHHTw/PkN909P6WJk3I9YV5HGbppQwEVHjJ6kBagBQRx4746czzEgA957VLVOYAo+eKQhIHGOUFkbVBURB7gaFBRidBQFeYVjqWPWoGiFsCKIq6mWNb7kgF6iPyxCDS19xvZ5IQXM7FeBX/0sx0Yv/MLPX/JTXz/HL2aupluepz0p79hr5nZSRoXiYbKJ4Dqs5Dr5VCvkdbCZjRCkkn7R0xWPzkpJVif0mAlSODtZcffuOYs+EIMwDD0YbMeZe/ce8s47D3CyZz9uyDqh2Xh6fc0nH6/56MMrnjzfsNe6+tQZ55nUoAhmQhFP6Dw5GyaCBI8PkDOMM8BM8ILH471n3Gc+ePyc+P/8U1argY83G1KeCb6ujsEHQoiUAnkujJMxTW2lcZ7lskMA1bnmpxLqSHCuEo2NrDssYB45DhJXgSilHK7HIw3ZFC/krEz7mTHu6YLjJPRMONKYkJyJKP3KOPc9b57eZXF2n/sPHrBxiZQzu3HPerxhV9bsZI3GTO9gyEIxQ3wgLBzRCWawMOGORYzKQTjnMCsNQhvFCmUsaFa6PnJ2sWA4HZid53qfGG9Gist0XYeEBXNU9rGQp5HV6T3Oz09Yr9d89OEnPH1yixVwfUcQBZQXN5k7eaDrAtM0MeXEfhqZNROHBdJFxM84KiciDnzwGAU9phCeYhXmQyW6SzFKqZPZO6BYS/s83ipatgbxvXhKUbIGzKj3XRUzRynaiEqP8x7JpWUGhkhpJHkNWOIcqIEHp58dKnxuQeGfpS1XgS//i0v09IaP1o/ZaGbjYDJlL8ZOQDvoPejscQbRlCA1jy4YWQUlkrMQpMI1nKPvAyVlZp8YBvjpd9/hT/7in+Ar73yJe+cndAGcN6Z5Yj9PzMXIuuPq6ilPn488/vgFz57c8vTJmt0mk3ONutaB4DECVgQxQbzDOY9YxtLL6B6C4H1ALUPOVJGkjigfIsNKmKeRb3/vMWcnPXOMLBYDaS5Im8rr25HNeo/aTJpAswcU857gB4wMzI3BtpqPN1b/oKo473HmyEVqjt0GihWlaA0iXozK09f8Nk8K1zfYPLKIEZsT/W7i/mrgp776Lm88uODO2YrLi3O248T3Hz/mI7fme7tnbLZ7smX2eSILEMD3EHpQV1OsnGv88gG8g07hrQ2ELuC9B6uB4BDjnIewiDjxmCSeyxV+vAXvyB5O7p4yTYmrMbHdX1E00w+Zd75yj5/52hvMwdjuR1QheF+DUueYfAKB3Qi77Ug47ckh1qA9TxQgDD3ZgQulrvSqOOcIQerELDUAi5MabOWgIDm8L7wCII7qwGGS+65vHENNS1QNbXJaUbBgWCnH/xVKRRZieHGYL4gETJWs5UAlVd1Gf5Qm8sPbFyIoxN6x71/w8eYZa6eo6ympMKEkB0TovdCliM8eGxMxVgXAERmnwrSb2YugPjC4CctWI7DrGJYBTYHlMvDOO5f81Ltf5itffoc7ZwNSRmKEze6Wp9fPefz8E37wg6d861sf8jvf/D7Pn41oDohrD4FiE5qsDtp6e3Dim4xUSSgJib4bKrxp8M25wDB4/FCJIiHgZMAtmi5pSugXhG7AR9iUHdOkaFK0ZMoBHRHAekTqZFaFEAOx95iBWkGBIA6kgFXJ0xmIKZWMrKvRPBtaAAxvig8R847ZEoN3eIFxnsm7TMzw5uUJ//q/9HX+ha9+idXlCc/3z3n/xYf89nd+i49vbtmmBJ1nlxPmPYgneU/pgE6JvaChqzyMZBKZ4GrgVO9QhCedNcg8H0lG18i6eqaBYJlgGVcyNjfUFKEbAufDXYwdty82dB28+/Wen333PsMpPNvvmPKO9e0OLULfRXxUOmeIh0DAitX+jB0mQi75KA8rBRNrErSvq3GjA5x3lFJqrLWXJiBpZHNduJXSXq9SegscueC8wxRyUYpCEsNJRVSmoA3qHYR1OUaZgjiPFyGnptBQJVK1mmJjR4r6x7YvRFDIKB+Oz9hapsRASrmlS55QFC/VmwCCibHPGTNP3wv94MkFMM/V8wnfCxoU74yTfoH3HdOUEFGWQySlDc+uvsuwGBnTiugULTM3t7d8+Pgx3/idb/E77z3hk09m5j0432F0R9LLOUXV4wExV2+QGaUksHoDMWXl/XGSTiljpa4gsRuIUdCcUc1MsqdzHqLD4cjiERM0Kzkp876mLzXQ+0Zi9VjDEK4oIY+shp7edWw1sZ0zRsH7UtedRmplhTwb4h2d84gTSp4gW00neqlEmjMQj/OBhQSWOZLGLZeXPX/h3/xT/OLXHnJ1+4R/8sGv872bT3g6b7hWmAZIHQyNGJuzgleiDyiKqcAMPhlDFtQcVgRXwJnHF4dKYdsdfAL+lZy9rrxmlQ6hFDxGrKwfaPUIrMlInhhWPY+6kZNTx9sPLll0kfV24rvffsrvfusx8wyrs4FuoUgwhhhwHpa9Z1gZi6g4n9FpQ8mlyeEF8b6u2FKDgLS0Rxp543AIqS4O1ricxu20IXT0kxSrQoY5wbRgUpd3bw6zSgLj6uWZ17qQWPOluJpKGFTJVqAg5KJMjcSWDEOBIoH02WPCFyMozKUwS2Twik0ZNwu3GMF5Ai9NGBklCYwRyiTIpMQhIx5SKmyfg4+JcmKcrgKEyDjOjPuJ6KDrHZvtC/7pN/8vfvf7nrPzE5bLBaUY45h4/mzN82c3hKFw8dCT9oU8OdIMeYKc6810uHpOJR/dFwYUczUqKwgBTTP4Qgh1QkKocuAIJc8km6tENQRMHMkcZVLctGfOiZynmnIAdXWPCEJ2FXo4y3SaWVG40/ckAiVHxpSxOCPiEBOaPoGJMJkRzYMEIhCcMZaJCSjZmF1mNWT8csC5wEnpeeiFyzcCv/Tnv8ajnz/hHzz+Db7/9COudeJ6mNn3MBeICCEZVjJTocqigPcZV2puLAnUJfSAYgJ4geIMFQUrxIZo3GGSiEesyZJOUBlpHG7NO/AUzUcD0OQE8crdsxMe3jth6M549snEN7/xhPe+dU3OcO9RT39R6JaFWZUYe0I0TpcTiyCcDhGdPcMIG1egA/zUln9XDXG+/ql11kJxOGoadEjhpfk81BpvU2/lEXmYUXkFHKaKk8ZJuFIjYKkuJivUAFlCjS7eMWetfewMZ4bOmUkz6gM9AZlaX8aATvNnno9fiKBQcsGNsFyt8IvIPhqb/Q1qBe88s+Z60y3UHDBU4884wX6fIUcsRabdjAvKYgj4MjCNI+v1nujh8tEZ9+7eQUyYbM1mHNnnmcUu4rxUp2KEu49OOS+B23Xi9sXE+iaTLUNqcm8BcHWE46AGc8ohR2gDI+XqBOy6gRA9KSvTlNht9jUwUCfDsKhwWKWSiLu0wUwpbVSJPzjeKpPsDktOmwDFQ1j0XFzexQ8d7vqG/PyK3ZRfdrAzgguA0JuBZdRy9Wd0QpCOkpV9qZKkRgia6ToIpnTR8fO/8NP8zM99md95/E2++d332Jkhy44iQimwGgaW/YLt7ZrdvgbLI7ptvx8clnWUV9wcvcc511BARTYIzYVaA8eRXGuuvSKH95OG0pTS5FPnHMqO/VxYLU/Zj5GnH93wu996zu++NzKNnnuPBk4uBhbn4GLCqxK9B524POtZhAWUwCSZLi7wMoKB5tr/VdiRGtQOlK3oMb0RqWThQflxzh9thUUV8Qe+oV13O/+Dk1ZE8M5jptQyolwRkr70qZiVI3oVqAR3MVyGhXME70mhsM0FdKa4z17R8IUICpqN9VVitVwRw4KpZELYVPtqzk2+Apr11juHj9UBth+VvIdpF3C5RlXdDWwLxG5mWMDlZeD+oxXBw4vn1+zzhJqSi8J1YegjIQh9PxBCJKWJuIDTu5FsmXEe63wvINZVYw+uWl0BEKSAloNTRkiASGC5PKFfdGx2W8ZxJiVFBUIE33l8FymuEUTArArFjvmoEyjV9lZNKi3/r3GpMstxOTCcLVmcrkjB2JeMXu9JOVfTSlZ8FMR1LILHbMTnGnBwEL3HsiNNQrJKLhaf8SGxzwl3suDOgzNu9ld88Mn32HljUxTLM/t23qvYs1ismHcjjoz56otABN9Gu28j+nAJxwDRgHWdKB5x5ZiDg6OYvmKZNkw4egFKeflalVcL6upEHZPy9Ome93/nBe9/e2baQQgdWjxTLshYyUIB5mkiuszF8pzAwPp25PZ6h1lAcwtIGdBcST1XpcDSoDscbNzVF+LEN/SolJKrUiCOfOBHjmihqhJaFC+uksPFqiriPT4EzDJCTU2dVRL4ECScr5+N1EDlRYi5fq46QYeAZaEvERg/03z8YgQFhX0ubDWz2e3ZjCOJjJXGTtMmxxGSOaTzoFUN2G+U3a0QrSfPE7sXiTRkzu8VHr4xcHl/QbdwXN/c8PT6ms0IRWpQ0QLBz/Q9+DDRdR39UN18BE9/IpxqDzazNsOSICrgtA4UL20pr+xwzc49hUhWJWmmJzQoXG+i6wLDoqfvI955tGiNJd7wXUfzpWLartlLRQoNnECVtawUzMOoypPrG+K4ZUwVWrsu4g6wNitqheBKtTmL4MxBKGQUpAqTXh2WtQa2FMizZ5wSt2XmZj2zezHxYj2yHYQ9IFLIAppgt91iU5UvqzZux9X+QIgdOAFK0+Tl5aT23rd6BY7HHppzjcQ1RURQXn3vw/HaCD2HxUrUjangJtjvhHELlICjZ3+byT7RbxI+VlTmDN581BOI5Kzsdonrm5lZtwxuBQQ8njkbxZcW4V4GMyetFsKaGcw15r/U865Bw+Edre7BtXSombisBgaoQaFY5VmwqtI476oxCgihYtODyGiAwyOhKm8+F0o9KfCBPCfc+M9Z+oDAjHG135Ks5l/hmI/WFdqJgAlqNYKWtiqnJOhcDf6LGJjnxHY7MXSBO2fnnCwHnIPtfmI3jlUMCAOmdTJayRTnmWZwWZmzY8qZ2GeCD0jnWZ4Jmjp2u4mcKuzVMuOk+tSrLg0kQ7JByWjxUGb2k2BkpjkxTTPRebphIPgA6smpWoudC3Te49EasBBKI4vAVY269UloEl4G1AnX25FN/gQJHkMpeIqFNvEEH1yDptoynro6S+eqjd9DEUfQhBUlqAADJXUUdWy2mY8f33DSeaYcyJKZDEQNadZczcY+jZg2CzXUXJcKgaUF4VIq++4d7Tna9StOQgusjT/AUSy9EhTq9Tt38F5U1r0+V3+KE5JVog1xkIWLu6c8P5m5ea615mMM6BWk3pBgjKNyvvIsuzukBE+vr7ndTowZpk3i3tLhCUR8HYOlFXsp4AR3OF+pn+kaknPOYVpNWWZG0VQvwPwxfTikSgc/iZZKIJhVniw12daVgnMBpZqahLpWIGDqXvJuqriGM3PK5JIZJKKbz4YS4AsSFLyDKFRdNoCT8Kk8VKRqtaWyMjXn5qXWmzOULHQC5pSTBdw5X3L37B4lT3zy8TXmFd93+C5DqkYeEU9R34icGl1LKqSiBIOhy5XFjZ5+KQy9Z72rg9r1EDxAprScjzZZAWyeydnYjXumeaQkwyEshgUhVBQxTYk81+W/HwKxF3wIqNiniCorFTy0Z8AXJEAXe3AwmsdNUKZcB6PUwptDwJLS4JaUmjK5UieO9/hloCsB7xNYBjJD8mTn0MaTj9PM7XYk5IGpwDRnikDvPM4JgUq+itVA5qW+7uCIEszsWLBVU+963aX1m/OCd5DNCD7UWhQMtMl+VldUmhx47I1XgsWhFTzZhERiERPndwKn557xNldYOtViIkzIViXEO2/fY7E44cX6A27mkeQ7Sq9M15mz4oh+YLsfCcMCJVBKTW8cDmsohcP1/hif0HEPk8YxOFelX21pUh2QgrYCMQFSUmJ0BC+UbGRqvQuFFoilVr553ywobQwTKLczd8KSx2x/3FQEviBBIYhw7qvEkg1m6uA+rC5HT7e1wVVq+a2TQCkwTpl5b5z2EQuwWMH5qUezsn0+89HzDcvzwL03VzhfCHGDlVqz4KTak2vkN0Rqzm/aVumQ6cQRoiN0grhqcz1/4Oj6jv1uZBwrsTY4oa0bTPtKCk1jRieji4EwDATfk6eJcRyZxjr5QwjksRqV+s5Tmq4dQmOn9dXB70FGBCX2A+IDJp5ipVbQteofXyEG0kwzBxts5UNrgZYLVRcXBy4aMYKYMAQjxZlJCqFA1pntuGalwmw1APbBE8RXEc4BSCtX9zgz3KGqqTntat78EiEc6FmEo3PTe9+Ch1VC7hBQSlsGWysHgk3q/XpZbHD4xBXiBbWJ4rYMqxX3H/TsrzKbq6nKhzZjkzIX487dgQcP73FyFvhkM7IT8EOHxswmKUlh6BeU3QZmJSyGmtGJ4L2hWtFStSy9PIuDs/AQkI/Ry0q1vlNwXo6l90Vf1ntAk1+11kGQ6zjQZpE+BJDaRZ6sCsnIpYCPDD4imwm7Hfnq5dv8uT/+y/zWb/yPn20+fqajPue26Du++uiUfdiyyYnrMbNvhSJKzatNPCbVL3+YwK4xZXnKjKNy2gs+wHLhcQFevFjz9GrN9Q6KVxa7mX5JtSAn5bATgTTfPVbLol0MFO8wSVS7kNbR7A0fYHUSefBmz+nqlJubNZv1loNprLR7f3rWs+pPuLm6YXO9hVwHTJq1BoQpk1PNpZ2B5oJMBQ9YfNk31e3nauVsG1MhBPwhAJWW2bqOzkutJ1BDbUZKrmy5860nfYPbQgyhVkHlRNIR7wqxV4KHXpXiIIgRBGKGfdoxjqF6Ixy4EAiNYBWrWnnRQi61xPkgylcUXxGHCZVkdTSyzI4koVZWrt4HAZ0V7z3eC6q5sf11IubmSfAH05B7WSNRTdyHMlhImnCy4ezewPllT9obMQayKWOeORng3XfOuP8wEJYj0wYk9syp8k5JYTeOnC3O6aTHZkE1kC0Tgmvop1rRjmaiI4l4FCCOv0NzxGrGI8Suw/kaIBqX3moYHM4VUjJ8CEe0jHMN3RiFWjlZmiKVsiLOo8Ex7if6fearj97mX/uTf5bxdv+Z5+MXIigEbzy4CHByzjbt+O4nt2y2VbbLgIoH8+SXtg8QhSToviPtKirIAsOqZ3VaN17YbzK3LwqpwLgx9ustISomgraCpVIc5g6KkVZmm5rPmoPsqWXFRQhLoVsYzid0yswIth9xqcLzUg18FMsYIxJO8Z2HaGSdIQvzCPNYB3EIlawTOVTSeaaxKQbSyLdGsOGU4Kgrq+sQYiXwi+EtgSbEOTzgUXKrvHO4xtR68J5kxkCPWMHmiS5kzs4Dp6sOx8y0h+0+YNNI7wtd7tDgmcShKXBKzyfUNE/M8DbXjVVSvV94j5VIKYUGICpBGDxItXibM2IxEI/42ufaqv5yQxPVDCmID835p9UqXuu4GqnXUitTnLMjmx+4hhzIMlCoJd/9aebu28I8T7jUoXNHGEcePer5qbcLZxefcDXfYDFguRovXDFCB+s0crq64ERWjJuEDp7iFFs6SuAokfqDGuW1me3quGraUeuLQpb2pDOUCSdd5ciot94dRqDzWMl1z5BSS/19CGStVZAxVuNXMcN7R/EKvgbNmODn3/wK/8Yv/mm++50f8L/+2j/47PPxDzeNf7Jtm2e+cfOCE+kIvSOeBropM6cKLQ/xym5kAAAgAElEQVQ69EGnBkADcwqMOeFP4OwkcPlw4O7lKef9is1aefzkCSUZseuZN8r+KjD4U+R8Q+wybS7jg+DwrWSoQIYwVejsSyX2eue5c29g60ZevJj51jcdOj4jSo8vvq3IkC2RCyQHYo/xUmG5955xVDRL3V2KmgpVCOp5da/MPNe8v+ih+KVCzOAjcqAJzY4/D/m5B8R7+i6w6EKdPF4IIaKlMM91o5XQ7RgWEc2JbmHce/uEh2/coVgtAltvjN0cmGaYnxuajMcvrulP4Pz+gpAmymwUDy5G1DliH9E5M8/V/+AEpBJB7RqMw95Oh3so0hALBxWipYalFXFaxnmrmrsq81zlaaH6EYhSyTY7rNg1ah42jglOa4ojhUTi9M4p7t0FL36Q6IfIeLrk5F7E+sjz3Zqnu4kJT7ItOQlIoO8D83Vmt5s4XZ2ye/GccZPplx5Niu88LoSWa9YBJS40rbDtXnWwJzdOrIvgYvUiqGbGcT7auJ0zKBx9KsF55nlEJKCqTOOMCw4XpAZ5DAmClhp4w0b5iq74V//sn+Hxkyf8rf/+v61elrPuM8/HL0RQUOBFSaTkWIiQ8QSnlGCo87USUrUqdamuDs4EPPQriKEWSy0vEquLxDJkCrBaKeuVggv4LqDJmPaZ4dThotAD1drQ5DGTI8tfDLK4JicBJRPchDtVOvMM2jOWkTIqUQaieERnxGpmmdXjSg8+kZNiRRHXgYfm0yTn/KlgcPjdqJIapeACrdBGKistFZ4XqYYYOxTGALnkGtZSwUli6Hv6rifGgPMBzYXHjz9idbHk4Rt32ewdV+vnfPziGSnuKDKjJYF5tAf1njkUxpLIE9yuJy4uVwwWGFPt49zy5dgNRAdqhuQ6qK00Pkjq/QrOv+SIjlLip4nCV3k6awSwinwq19bGsMmhGIBSGX+0jguhpZYeE4eSEZTsEovzJeebyPrJFhpncLXLmOxRB/NemWdIIyzcwOJk4PrFmqurNat7Pb4Typyg7yjJKFp5pgYHKhqg7splJlSQb8fXquW55gj1/vMKN8BRRYADB1SlaN/Ko7MKnXdgoS4KzSWZx4xLcCIdf+7rf5y0m/i1f/SbjD3EVUDDP2e1DwqM4ugMLBVE67qXU66boeSqhZe5rhIBxxBBOtAeLCuDF4ZFZrGKDPGEgnF+N7Afpe33CClNjNuZOHoWXY8EYSJVx1zw4GoxjPZCPgzIxjJnwEImrgKLHhDhhsI+GSntcS4SKDjRuimIBSi+1kR0zYQoB9xb10zf3DzaGGh4BT20qSFWVZJwtDZac7m9bPbKjNGSKXPGWW4auRBjYLHoWQ4D83yLsqMwcXLWcb33PLvKlHDLyXktuS5lZkwwq2csgopnnDObTWK/bWmMZrrQoyipKFISOSuegncdU5mPDry6K1Q1NAHN2KNUqfUV+y+VCnh1gpSDvVl8LfSxw4SvgcJamuWkIosqewLU/RjR0phaIwSlWMIP1C3nrBKzm8kTl9WaKqX2vYsdS3dCGoXQjWyuNkxyQr+MlPUezbUOZ9wLoRTC4ClScK4QYqhl0VTi1aQRQlIdusXpS1mpVDWllAYupAZUeQVNOe/aiKkSZ2mSjR72ZpyVXgXdG197913unJ3xq7/2f7MeZ4Y3Bqzv+FRU/THts3sff18TkS+JyP8pIt8Ukd8WkX+/Pf8fi8iHIvIb7fFv/fg3A5VASsY8ZTQZmKdkmGfIc1PL8Hjpa1jIIKoMnePOxcDpeWyEbx0YzgkhBnyr9KMNlrQzxptMXoON4BOIGgGlc4UuQB+NPmY6B52rtQN1Q1HAw7A0Tu9kzu86+hUUUbJNZFfQQ7WdM7AZHwpDB8PC42NdtQ4lsTHW0lz3+yyoXpo7rdEBodlWvTuw2y/zz9IeOHf8vZgwTTDuEuNuZrfbo2mk7+BLb99jcRoZ04asytnpgr6D3Qb2G5j3gTJDmmC7Vba7zDiDj4H9zri9zcQkDBkkGU6qH2KeJqRJuTpPOIx33nrI248eEKXafr15PP5lRamVo2/h1Yd96lE7/WWlYYUS2tKNKl27+iAgbZ3TUhFFMcB5TArFZSxkSkzkaIwCJYJ11f+SJvAl0pdAR8e0y9yu9+xzZhTY2Ibh3NH7wrjd1JmcDQhY8WR1qLr62dTheARBjZEt8jIA+rYnwsHhCC+vu86ll7xECJHYxebEraqVqdbdsTLIxnh7ecbPvfsz/Obvvcf3nj5nuAhYFMyXlxs7fYb2R0EKGfgPzewfi8gp8Osi8n+01/4zM/tPPusbCVV7nTUBhvMeNFDy1Nh8qaSU+bpD8qT4SQm95+T0jIdv3GO17Hjy3e+RtoWbvbK+zTx7suf2VgkSQByeARcK801iWxLDUvCDETtaWXGVj6QoooZTcOpBhWJat+YzA1/wAyxOPOOpkPZGnn3tTh9wZAJKxohBWKwCw3Jgs85sNzPOdwz9gA91x539fv+pNOLokFSay68yakItfz6YmF52YGMgaQPKfK2s9EZOynazwdmIl5E3376HO7vTEEmlxk4Xmdvbkds500VYnVTJMs/KPFUovSqRMc3c3iYe3KuM6jgrbiH4WPP46A6OvYJ0ngeXl4z7iY8+fEI3RKTrGNNM1gzuZfXm72+vrmkOw8qrNmfXFCgjU+U6LTWFKgfLazM3lVJ9wC5I3ZJMDHGGXwrxxJNNkUWtOynZyDN0paubnm4zm2fXTHtQyazuBlaXS+5dXjBOiQ+ePa9ORucwFTS3zR5crXCtgNCabHq8qfW+GmCKihC8xySgmo8b6VoLdDWlrddfrJKMlTMph4p4osECz36d+eo7b7C/XvOb3/893JmnREgpkTUdU8zP0v7QQcHMPgY+br+vReSb1K3d/xBvBpYLCauuYakUTd3+rpKAqpXhTqMxrpX9BhYnhfuXS+4sH3J+tmJ4o0PHme0WrvZXrG8y+w04r3ipyEG0FjdZVizBiQMf22QvNU/2BQatPzsTDE8uVlMbKbjBoVGQLtB3jq4rjDmQG1dAMVZ+AuquPMMislr1jJMd5XkO+wPoH5wZerTLAQgqoP5lVV7rsuPGT/WwVzawF8FLR3AF5yHPI+sykdKOy/tLTu6s6IcB1cKL5zcEN+Ass7/OsHBEV3CDhySUGVCH4UlF2Gwm3swDxMA+z5SWI9RS8np+zoMV43s/eJ+S6rZyxQxSlS3LgUeQP8gp/P5xUWikm7xMk4yGhl9JMzLlaCfmmH/XIB9bPxlWi8OGntWdiKVCXATUJkg0e7cjrTObq8z6tqYqw6LnwaNzHr1zn8vzO9genq1fkKaRGAZwASeGi74iJ1fTlVrDUXmfVw1NB8enNBXCOVc3sC32qeOqCivMOeNdJvpQ58KBl8lVtp33mfunA5eLE97/zne5lkxcdPhFRDQh1eH0Y6fhof1EOAUReRf4E8DfB/408NdF5N8B/hEVTVz9f/4/EFp+mKQWc8wyU1xAt4ZXY9CZPEO+hXwDmz1cJNAnEX+6YNiv2D6tq/W5DoSLgeFrCz66uuXpsxt2NyPpulbAdg76JdxfBB50A95lRApd7/Ax0JdMnw91EQ7vezRXL0EIHd4HXDfwXDd8Z3WNrjJTyrU7te7lgAPxA91grM4zvrumSMEIzMx0xSE5kMdcYaAI5mtFaGhanhUjYZScMa90rkOdUUqF4sFXhGCmlKKEBlhzAV8EyeB8wEnE1DGO8MH7a97arVg86BjOO6bTjptbwxMoW8fmxvFsnLm8d8IwBSzNjDaiNgOO/T6gN4FheYfH5QWWSy3fFiAK276qAiHBbW4OulWocnKzB9O4AawqP9BIVqmcgjOpezlW1RKRQ2Wkb7n43BSNqu5L1SVrXUSx6lIVcDLjTbDkEanl6ZmZFKC/K7hsOD8SVKqVWHrWLxLji4Ht7ZZSjG4hDGeR5Z1IPAMG5ez8nOWw5OnNFtclfAAXPVFqzm+x7oolWP2eEoPUCHLvQ63TwI6cSOgc4jpSySB1e4BcefSKZgw0ZbwPXITI85KYg+PiyoirwHOX+aWHl6zM8d31U9K5EAYh5R0+1sjq7LNP9T9yUBCRE+B/Av4DM7sVkf8S+Jv1lvM3gf8U+Hd/yP8dv/ehOwGzukddbsSSV4VZ8Qk6F+hjTxYh+4RbBu6cVsPG9z/6kNvbDZdn57x554QYHFimc55HZxd0WbhfIutyS3bK4AOrYFzcjTx86w7n93qkTyTb1RJpgTlm8mAEAyvCXKiwtxSC5rqx6TzhgC89vCAyMpSR7Q6yq7ZXFfDR0y8Dw1IRlxl6o49C1vrFJZbrjspFD14bpepRrXyRum+iNePKnLTq9xKrhn+ECc3/LxUyC4BUlCKtkEZzwZnx4uoW5g27PHGWT0miiEU2tzeMe2EIp/z8O+9y9eIFm5sJHQs51SIfcZ4xJ65vlQcP7+K8Z55n6Brg0QbpaeRiI2Ec1WBTS6P1uKO0OMjJjiXD3gtCIxRbq2lDTaUqXi94PEhNp8RaiXVzd5VXyEdpkMKKYuJR2q5IJGLncL4ja6pOS+kpFnj6fEu6raXvMQDmSPvM5mrH8yhsZMv0WCgZ+sPmu8kIi7rlHk5RO6QIoJbbTk21EM3MVYTQHI6lybAiB3WmKilatKbKjZNU1brfQvDouK9l3LEqSh3C+fk5z2+u2Y4T4aQjxEjWudWNCP8spMIfKSiISKQGhP/OzP7nehPtk1de/zvA//LD/vfV731Y3nemuSnYAlGEWKAvnhgGLpanLPpawnoejKIR3wX2c+bDj57x3kef8M3f+4Q/9uiS036gW8DgPSfdwJs+cv7oLeThW80Np2x1je+BUPjk5gVZZsYyMeaEiTJHyBEO3z5U2vZYxZpvIgXGWzhdKF/76Yc8urjkVF7w7fdvyd3APgmjy/gF9KtGEJWRGDIxJupWjZmcMprAu0ChbdByzP2E0r4l67Cjb9v3szrgFFzd8bPt9tQqP9W3kVQLbZzU7xfIOSMOogqbrTJ9cs31NNINHfNOEQ3M68xf/at/ib/xN/49/qv/5u/yn/8XfwfxPUXqty45EXLOXO8yp2OmW0bSvK8ASapNV5zU4HDYZkCqO/LloPw0d2LNjQcv7deUSrRqCwaZajijbWN/qFsBqrX66Cqs/SHW3tMEK5XFKy2A1NmqlVsQRymeYh4tke1tTT0DEDvwrgN1bG8TOmf2t3vyPrF/Wq+pXwyUuZrbdJJqRotU7knqdzSYZoqTtnVbPa8QKk+WNFO0ljiLuLYXpatjrPEmNRusZdOaq0QRRBhzxkKPJOXC9xQ1vvfkCSXAMHTVNu4chVJ5iz9ARP3o9ocOClI9nf818E0z+1uvPP9G4xsA/m3gtz7L+82z1i2vgyAW8HOmz5GzeMEbJ/cZOo+WQl5GUg7s5xHY0y+WmB+ZJfOt954RJri4A1//6Uvuv33JG/cuQRNXuxuebK+53q95bFtmLWRTZpuxABPU/C5CJNDNDqMW4YgILni6LqAexr3n1heyZLQb/1/q3ixWs+w8z3vWtId/OmPN1dVVXd3NpiiKbComGdLWYMOOLowoQGIZDpzJCWLAQK5yEcNAbnKluwBBgjgIcpEIiWMpsmLZQSwphB1RkSgOEummKPZYPdR8pn/ew5py8e1zqq3AUgtUkva+qQFV5/z/f/Za+1vf977Py+G05OqNggdP4LQN2PGYwiwxMZG1H6AjAaUiZRHRKPpW0oCMNhhlRVWpkEZGRG7exLlhXp70CYFlJCkxVdBDV/vDYzzJs1BaFoqsN5HgRp9JhSEoaBeeVdtRjytULtCdYWds+dRLn+Th77zNdbtP1VtanTDKEfBYqwg50oTMdtvhCoNNRgoaK40zgxNTmE6DnkBhhumKwFSk2+69LFBj3OA9SRC1HH1yHiyA54ImRT73IiML5FziHIezstFcLDxpD0ViiCIH106elud6hyzWdLLF4CAaNsvA4qinduBKSVvKvcF3mdBGQm/wHvo2ETvRJihjMU6qu7jt8TrgKFBODyAVRcYSoyfEcDFBCXGoCBHHL0hDMKWMsQZrrVQHRHwOGAxZR8KwgZRWsw6R1kbqrLlczMgRzvwGW1k84lcpKkcaIgzThzfjP+L6QSqFLwP/FvCaUurbw9/9LeCvKKU+i/wI3wX++h/9pRRk2VkVmhwMqYG0NTS953Q1ZzquMINZSLsKgqGZN6xOVqQmYJOiKgtu37rEy3f3uHntEqnvef3BOxyvjjjLDce5E5TbSPpK2hpMUeI1eCJYjSk0o1yQukQfpEsuCy7ikjQkm5RYlho3gqM0Z+wM1eWKvjSs1zB2lnFZ4kwB9LTbVsaJOVFUoKyMTNtz4vN5020IpxEdzhDocn4W0I5hGE8KGVS4cNQpBdZpIOO9JxIF4pIGazQZreRYkyP4kPGDSSkET2UdNjtqV3Lr8jXu/da3efr+e0yiiJT0WJEjdNFTOENPYLNNHOyO6KMjp8DQzpJehh4WjRq2pKFJKA++CEpi0SQFCYyRUZ4kVkWRRSsr94QSlE0IAa3t+fT+Iv9A7MvyTXJWqCyhKAqZ3OSYiUp+PxQPQ+MtoykodMVq41k9aelXcHl3QiKwXnb0m57kFTZbrBeFYpEdPkZ8GwnbNXuXxriqoI8tsQtDgpToDzDSrTZGDGvD8OQi6EZroXNpbchZ1JomJ9k8BnWjNTLGjSGTYyQpS+EUqoeu1Iyi5rIZk7NirSOmLMmDuSyEjqykAZvOg1Y/wvWDTB9+g2e97w9fHynr4f/xQuxw1kKRgqLfWE4ebGDRsdAbCpsoJiW6rCnGO/Q5cfx0jj/tcS1YMl/+0o9w9/YtfJzz5qMPePj0IdvY0JeRrlbEsUHPKnayI2w7fMioaNB9xCIbgsmiu/dIbl9OUQwpODovZ8WtT2SX0VPoa2gqiN4QCyVKv+WG/UO4fnlKDJ5mq2mbDcbCzi6EUNKshEeZvSUFI2dxFQUgokBbQcbnDOcmggQSXHuRC8aFTFshYhqrhl6DlvlEyvLEVUphCzuQgjO9EoGMVUZweBEu7e1x/eCQ9+8/4hu/9hVsEJ2FjxllEoWxjAqH9p6uS+QGRnXFJm4JAxRGR/GgZC0gFREjSTM0pqGZqOV4Y5U8Ic/L/GdTCEv0UlmgkIAbZaVZOGwuErgSBTuSMnzo/5sB7xZRshEmsYqrwTChFQJ8jYbQQTvv6U7BdnD95j7YSGMDceqwZsSoqhnVJYpA17fM52tWyyWpD+zt72Jrw9PNE4mdS2ngSYrOBJXIOl9s7kn+ahAqDbFwHyJWpyRHjvOlee4UPl9qkUCpM4WGJif2omYnlXQx0VtwTmGdGtovSviP5lwF+tEmEB8LRaM84VqMNThrCV2mXyjmxxmzQn6woUVVDck15GKDKh304NaJHVvy4u3nuHltjwdH7/C9h++waJfk0pDHhlxZks1knSh9IqlMzoNXIFlyEJ6DiWJqiakfytZezsLD7DgliEmhY2ZUBupSEVRm1UeIGuMMNgRmquBw5NipLUqX9HXNYmFJdKADiopRCTp1bOcI4UkZgsr0uaPEUtkKZQy+l9zFjFChUxArsjZaouBEwjHcgJALjdMaiiRlqIIc7UD7GazoGgrBueB7aWCpTWZyueRwd8rN3Smq3VJbjTOJTRY2hAXwmqQszbbHrzvqyrIOEJHjlhugLaAxH2IWosQdaZ3FKEvfZ7yP+BBlU2DgLSRpOJ+DYNBa/o3R4rY87wsoBR+CrwgvVzYiez7lzQoQ7LoaVqIYjzUWh4oWvMEFw7XphKvXD/ncJz5FdpG2kd6VUqUY80JH26/p+oKyNMxqR+ojo3GNnRQE03PazSEbGR86gy0ganGsGmNIShNjBm2xw5EqxHihaxCmhLnI2XwWgiM/3xDBx8DYKpyFeY7kaKiDZtl2eGsErXfeYEWyRuOzAfZHuj42m4LSz5RdfYg0jcW34GIpngHAtx1dH+mbFl0lWLdcqWZ8/oc/yUsv3eI33vh1Pjh7wlENeeIw44JoM0lHTM64NlJ0ms2op7NgrENFRVYGawzJRgLdYFyRLrrW0uOPaShVs6EkMgbKmOm3sFEeth46uLoz4ye+9OOctG+z2jxhPJ0x2dljMhnT9HN8WDFfdPiYkdxBI2w+I6o1nQJFNJTOitEGCK2IuIyCfM4xOBc1wdBXSINNGXShoTDi1MuyWIjCKXBlSe0cgUzTtzSrDUSYFCUvPP88hVYcP3pAt1kzuTxhbC3b1KOSjP189EQtQJzQ1xSmJGcpgaOICcTlN3S9UwpDNqJUYHXlqMox261nvV5jzdCDDFKZxRQxWQlKHWlenjMMMwqrzycOzzQsGSmt8yAljlENuob8rGF3cRSTJl/MGpUUVlmuHV7n+u0r3JhdRnUty9WCzSqwXLU0XaLrPW3Y0oUNPkSSihQorHZsVnNUbQkFlK6kjZ4YO6EoAxBxhUFZ4V6EGMnh3Got1VAaSGLxQ7Tn85HtQHm/mLD4gCSva4hkUohon2hTR9SgrL3A0Gul5AF47r/4iNfHYlNQKApt5aYPQUaQMwf7lnS6JcQOX1mCcqAULgVo1xzakj/zysvcffE2ry8e8tp2gZvOYKZJZBwK4zMGUa9lFFtncR2MjGjKvcoElfBKE7QjYil9xHQDu9+K4y1FGTSbGCgtVAWQCvFkpJrQdKjcQw+bJ8fkkyUru8Jqx7gYsTNyzPbGLLoNT48UR0eJdlGJI7NosDZIcy5qFCW996S4JVsPdRb9aLCYaInGkM259G1IT3LgbIGxFm07Ai0mzeh7C3TYsiWmgFUTtAlkFEUHdgW6rpldv8Ltm89TfHDK07cfME2w7yyzg4oxLe8er+SGDBaP0I19tChKRlpB7IgWmVagyCEJsn2AjAyYBMKAkY8xStWlLX6YxuQhNyOqTJEjNmmxrCdB06MC0VrKpPE5o4ZcDZXEcNZZaIrAKGdcUheGIQGeSuozgErS30k60tsGbWBuVsyfrnjvrUfMQ0Pbebq+l3GhU0OVkUlEkgObFYYChSKeBcoaDi9PmVQ1vQYfEjhzYcu3AaxKmJDxLmKVRic9ZM1qvNFEndkiOHydnonTsrEi4hpUoDGXTG2gWgdU1mydoc+aWhnK6OnSEChs3RAWXDLcQB9pPX4sNoVzoU7M0mibjqZcf/EO6/GC43dOOD5qZK8byiO/7dmtKr706c9x57mb/N57b/K1h6/j98TCbAYjTpE1+pwErCAaTUIJ90BDUkmAIFZ+n2MiG31RbJ3HuSujBk7kIKYpNJSJZBJRi9y3bTw+ZFabjqePj6nONrgrhvnpmvUmcOXqIVduzvB5i3GBtmnYbBIGjU09NoOO4If+QttsocxMLykOxhUxGFYnHW0MQ0l4Pr0UY1VROuqqQg/8AVIrQFIVGc8sZV3StpH1vMUnj1ca3QmPsa7HmKLk+du36VctZ09P0AEOxxNu3nmOKm9pecDj+3OstlSuwsRA6BN9J3P4FLhwCl4Iev8ZQxP0ETKtfFY+EUPGmErwbWbITczyxA8xo3KUzAplcAaaGMQNmUTZE0U4iA7DjZyhiyDBd8Mizs+OfgzKyJQSwWisFvBs26w5ebKSZuNG0epBFVqVsqEN05DBxgY6ErDk7C6ozU3fMl9uuD45IKlA61t0McTQ54zSmhCkmjtH9iU45+rKW9IKN0xXPnylHElKo0UbR85QWEtlJbkqa9HTWCOEaT1oOIxS5GEsLVOOf4E2hcx5tJWU7MWoZG93n4P6gNQUnGwe0Hcea0uC96hk+MwnXuKVH36F33/7db75wessq4AtalIfcb1CG4Ue3HFiLz7P3gOXM066dqScBASbIzEmAYZo6G0moUlaIt+D8kLR1YpcApNAMCLoWTee+WnLagOtB1TBtXqfpBY83bQ8fjrnbOGJ8Qrj6YjDg8Dj2YZmGWCYjsWoyJRAzbpdUZawd3XM4bUpRW3ZrFthOqaenCwqiz1XD9RkrQZoTMz4CCE5Ymypxo79SzX1qOBsDqs5xKxIWmNKgxs5ktY0bcfe/j7tdku/2jAyFhsMNhoKLDvjKcuypeodrRGhUtO0dBuDrR05dheBJbKTDiPEbC6SjACUB6XThdcjpYBSGicPNUo9VA1JFnJCXITKOBF7ZUXCymQhyP3iEhhlsSljY4KB0iX9B3kdeSBSKaUuotqClsXoEbFZMSnwPlNp8TnlODw0tAS1DBPRgSGqGTK6MKYk5cx63bFat+zsVZCHrBKTZXPSWcJ/jJN+lYr0OZK1lZh5Jape+jiAaOSjzMjR1SjRgsQEXciMJjWjGDBngzgrxyHqTuNVGizwwpmIQ3/io14fi00BxKxS6EJcXxT0IdJtOo5Wc9axJ1lLmwL0nrtXrvLZT77C8eKIb737PeYjiLMKFFRO0GNKyRHiwklojZy1jRJUGUMcmRK6TY55oBY9A5gonS7UeVrLiElbi6oUVMM9ny2bBubzSNeLzVjZES9ev852+w4nbWRx1vDw8YLYRT718g2muxUvvXxApVYcP+7oW0idJmlDzBpr4bkXLnN4fRdTZXzu6Nctrk6UCfptQmeLUkYi0NHEEGlCK3CWNMzpTMvOXsFkBtokjE4CXLEKtEI7iynAp4TfbjG24PT+MbHp0D7SnK2xq57Fdk6zaiiVpSpLupxJObBZtqwWhsMrFZYNXcwY92zzRSuMlgSnPwgfVINrsPdBgl61QaU4uEMt0WY6hKGhUsTFkkLX+DbQxIiqC7yLOC2DlhxFDqyVE4v2+Zl+UHjKhsmAdTP0KaNzkImGFfOQqi1pvpUmtDxeB30mhIsGp3AtU5QHhFYCuFXGQu7pu8S4rCgrw9n25MIQ1QdPNsKUTCrBIGlX2pB1kHQtwKT8rEodLqWkWjh/T+fCt8JqrI3E0IIJ2MJRVIZoAn2QJiYqY40WEtdHvD4mmwIoO2AvkyH6xPx0ydmTJafLleQrFg7TZxivFVwAACAASURBVJyL/MgnP0lZGL76vdc4SivCbEJnElUSvLcfbn5yJGkp3bJOpJQonaMpDF30ImxJkRgC58JirTLGR4yXM4g2Rj4lI6ozZy1uOH7YVJA6x+qkpVmB0iNwiU3fc/PmHb7zW99jRInpFF0Pjx9tuDJdcrBTcuP6ZSozoipOuX9vTdMkEgGf1tx5acTVWwfUOxXzzQlnyxNCDuhCSmUTEyqI9DX44aGcpHHJYCk2Bnb2Kvb2Ha6IbNae5bzFGAfKia5AKaH39IGoLIUtePrwMaHrqXRJ6CIP7z3kzacP6KYV2o1o25YWj46SS9mvxWDmRoZspXEKAnwBUWuKaMkNzbRMzgE4H1fGwciUhpGcptAOCLLYAZIh9wndGUapoqgq5t0WnCMqT7SGvoloMUNc8E5g2IuUxmotHIYMChlT9jGitRjezmmcuipI7ZALfgG9gQ+7tvSQs8mAfJd/JJ+lLUpm0ym6gG1csw1LXClHDgUoLerMECOFszIOThqVJTdE6UGQaoS6FPs4BA2dLxRDCND3HlsrykKhhilF1nlIPs9onUUEpoZ3oP8FOz6cd4pRRkQy68DJZsnTx8esGg+ukrFK6tgfldy5eZV1O+f+4gm+0nTKk6M89ROQnaXNHms01jkJqM0RQkCHnsnKM0ninKyLinE9E89EGuLbLeQqCyrLb+m2rXD1isjYVVTaiBjE16yOM/MHnm5lKJLCFiUn2wXjq5dpHs6Z7M3YLSY0ekvT9ZzNN4zWBeWB4vBggqOAeMa9t0/Z9h0HB2Nu3tmnmjl87jhbzTmerykLi60cJmZCC1mJ2EUCYwcc+sVn2TOaGG7evMTlqxNWqxVnJxuaRmOJmCEJO3mRP3ddYGeyw7Qc8dYHjyAlSlfQtYHcZVIHW+2xpYTR9gRcVLhk8U0gtlCMSzxbieAjQQ6i1jTn4q9zOhKAlZ+3fiZqApnTkyQOLWmNtqJQdKmCTuN6w1/7N/9DPvPq5/jbf+e/4yvf/ieUM8UmdhRGEVGkpKkvcEbnH8u5LfXZ+PJZrPuzz83ngLaSNO39EAN3EfbzoWsQjQ28O3na47HOMt0ZM5ntMt0b0ZmOB6ctWQVKZwmD2a3LUpXWCFtUvLUZT6RXCLAb6RGYbOiCZEwmDVmJFmYaFZWrKK3oW5Qe+heGAbuvhveayQZ0+oNv4p9/fWw2hTTo9UMfOVusySeJ1bwlJ0e2EgdukmdsR5Qm8iisiBWocQEqUGFxSgv63hrRmCuD8ZHYtLiUmBnHwWzKC3aHS6MJ4/GYyXjC7myGRuG9R+XMpjKsS0UOnq7fsO02nG3OONmcsuq2+GWHrSsII5onK1YPIyZVYDW2spyujllVhu5sw/7OHq9cvwHNCfPumGVYkT9IbLcNOxPHdDrjpRdvEnzi3QdnPHen5uBKTdYdxydHnC0WKGsISTISfBrmzjEQszROTZL3nIdOfwJ29kuu3zhktlNzdtLRrA1WjyC14EUXEL0idRB6z6WdAy5P9/jGw6eiZUARPdR2zKgac9Su6VMLKROtwlFitcK3DX3jKbQ9FwcAaQhWTeLPODcIDbN4UR7KE1YUgHIbaDXYwIKIk7I1EDUllrD1/Myf/zf4G3/1b8DG84+rf8g/DRV3n7/DBw/fZrlZo4pn0BmhXUtvI6dECHk4noDE8ulBdjwIipQiETFGGAfee3LSorb8Z3aFPCg8FHkwraUkk5bJuGJnfw9VOGxVcOvObbzd8uDJfZRSlFa4Ia6GUlnUvONqtYPTjqPNnOCS3L8oAaco+XyMkZwTCfOSCAEfIqWBqjDoGFFKoK0ZGf1mNSSPDc2Zj74lfEw2hQznLhe6PrA68cQngRASWEffJwIBlQJ1YSktLPotrU0i6YxQJ7BGEbR08XVW4D0mJ2aq4pUb1/j0c7e5ubsPRcHWd2y3W063DR88us92u6VtWinrlJhOSmeZ7Yw53Jvw3I19OvM8R6sTnixOONqumZ9lVk964goZJ44Sbbcm+MBxakk99CdzZtcvsb87JfRrYr/m7KRltVizu2e4dT0xmVznhbvXmexp9q86qrHieDFntV2hrWI8GtN3itgz6DnaIY26EKRbUtLGHiTSxsBsx2GsZ7HsOD5a0jYOoyoiC6kilRzV1CCp3tvdp1aGzekCY6wIzpWlKEusdaSOARUfhlzNRA5ifOq2HZWqgAH1PoTRoDMp9UM7QYREIE7BC4OUeva0Fmu5FtmvjgN5SrE4WfDFVz7Pv/dX/hrr956y+v59xqeBF+sDPn/nkzRPn9D5NSEBhSLpINWyHr5fShd6p6yNCNWQCU4MUhGiniHarbVYZy8CeT58qUFAoLPMOHIWBF41sezu7jKdTbFVSTaG6d4Ot91d2rjm7Hh+IeDyWkHI1CjuHl7DZsVqMactDbmMqGhQ8Rx/H4d+i/RgjNbgSlJs8b2nKCyFt5gh7NJHL/4ZeCY1fybR+EjXx2JTUBlMKDCxxGw1/XZLS8KQsNnLeNFaFAVOVRg1QedIFRQ6GdoIMURiIZWCTT226bikKz515Tk+decuu7t7zJs133z7de4tTzhar9huGzrv6XyPj4MI1ECVwXnZp3YnFbO6Qme4cukSr7zyCT7/4svMT8/43Qdv8PDRETtLoHAopQkVeJ85OT6jMrBZrDGLxxSXK/wyQqpQKdF6y6NVgNDy/POBw6sjbt65jg9nnLYbjk/nkBKzyQhrNU5nvI3k3BPWkJQdZqTIhjQUoClLXmTXW46PFMt5x9HjjtBr0B2oMW0RKPrBe1WI0WenqoirNd3qDJuDOBtTZjQp2Q1T9OaMFAI2KLR1pABrlXEKthvFpX6Cs1sa1Q/MBFnQOZuLTAw5L8jP3A5n+JyiAF5hgNVGbNGxtoY+WcpF4vOTF/lP//J/xPafvsnP/dzf4eXJFeYn73N49SbzJrOaByZqzCZ3NLHF5WcCJwmX0kPjWIvtGlG1RhIhRkKQfyPOigClE4ZBN8iNExicVB5Bdl2BR4t0uCwj00nN3k6BMT1ojbUFLsFL12+zWzi+9c2vsV6tqWto9IhpCz+x9xwvVNd4bfmIdRnpXWYaESZpVhcKWowkUIcgIT/aCfJNNZlsDOsiQt9S+4ByFi/Ia4yO5EI0D73//xiy8gNfGWzSYlVsFanL+HOSThzGM3EI+cCRciEdXwuZSNIRXULhpFln1y0v7B7yxU98muvjXZanC776xvf57tP7HKVWPPdBBD/KGpJTmFo4iDEn2hBoHRTaEHPgbLEmrgOvvzPn9def8OlP3uHVT7zCn/3CTzAJl/g/f/1brFpP9BC1ottGzo4X7O8e8PDshPXpGU09YXPWyz1XK5RzdB7eemeFqxe8OLvEeDqm2a45fXpK23jK0lK7ShKLcyT6CNmjTCnJUCkMPohz85TwKJXRnJ41bNcnNKue7TpJvFnMRGUILlJlg03QDgCTO9eusHz4mGa5oLSWHkixQxtFVRcygUFcfiYaEplgpA/UtRmXKioKNrmXp3+UCLicIxk9NCBl6qFyliduFtpCJF7AVo3SKIaufK+omoIf+6FXUd/9gK//5m/zZ37kR3jywQek1PHqp/4UH3RzvAdbOEa1YtOtyT6SjR2OKkgaeEYmT0AyQ6q30gMBC9AaY8SmjXbSQyRitBPc/NCji8PaOmdoZpUlMWx3zGw6wugIqROhUs5Y5bjz3Auktue7r/0Om/Wa0AVuVld59fJLbNdLjubHhLFBV5lyLWHDyWpS78nnqVPKoGzG+CzUbS3ms2hhayP4QBkjURn6PJitrMEo6H0W4thHvD4emwLgQ4cKRmK0BmYGSZpNCbHsOiVWWZ8CuXJ0JXQjTa8CRcoUfU+lFJ+99Qlefv4Fuq7jK9/8Om89esJaB/yOpXUW0wbs+YaQxPDjEHcaMaCiNHhU0viQyH3EFVKivXu85OFXv8Obb97niz/6ee5+5i6n2yVf/+b3aLtIjhW0hqOjM67tHaDnJ/h1i9rU7Nqa+WpNrmuULsFp5usNb7xzH1snfJ6SY6Jve9pNGDwZaUDPW/CK0Hr6NuH7MMA5DFpJmTgMs8gp0m4DbYqENpOCQlsph0MYlF1Y8tDIcgGuznY5e3Af3zYUZIgJkzWVsdRlLQKg4WdhcsaghHGQFV0bUGhG1YhlXpO09A5SkhI7iRBDTE1IZRjOE5az0LJSDJKpaKXjbjuYbuFaHuEfLfjVr/4it2/eYhwt3z59j53DHb5w6wUeffUr2JQ57huqqmRqK3zweB9IIeDcAHIdmo+KJKW5YhAkyRQEpUS+nRJOKYwuBj0FIhXPwvo0ZsC2DxOfqCOTsmJvd4e6rlAq47SRaYe1Qy6p5s6LL1MUBd/42te4va74V179IosHj1mrljO/Qc8sxgZcIccpnTwmiXybHoHeagsm46NMcSKRkNJAA1coJVuIc3aQiQRiCuScBh/KR+ssfGw2hUlVsT+7RHCwOnnM6SpgMhcBpHnQqXXB03kvYhajSGUCm9Fdz/N2zA/deYkro+d4+/0P+Pab3+fRuiGMgFkB9QDrLCRYI+uBN5BFJJPCkOWXNTnJIspBnsAhQvSe4CB4xVsPT8jud/mJP/1lPvulT3Lv8X3mr5/hrKXMY+69d59PXb+B++AdVBtQC8/lnV06eoaKEEXC1dDpwPHyFEzLdOzY3zkg+jknx2vmbUPOlphEw9A2Gt9FdBa0F8pKulBmuCmGpvuACTemgEJjlB3UfR0kWRTZOOgTNY4bs32O3nyP3G/QVuGUpiBhYmZvMhtMTZpoNTacm4sMOXr6LuK3PeWOo8oQ9PAGh8V/fjZPgxeBnNHDB59zhGzEFq1EM5GdwkZLPuv53K1P8qVbn2Ph32UzX/DO917nze0T/vLnf4r3f/MbbN+8R20V876j2EYum5JicIhKMK84T7MeUP9ZC51bqUHUJEnPBoVRz/QUdvCkSNNetCNGiR5EKCrDjTN4GyaTEfWoQpuIwxC9p/ca4zUpR3bqEbevvcj0CzM+0x2wvPeANsKbiwfMc0tUhhSDJIwlSc3SVox6yYtpTGPIRpOshMh0MRJUAF1KVL0XVJNSFq25GP+eNyw/6kjyozOa/l+8FIrajbm0d4nZeCpEMo+w9YOoU6VIhrZv2bYbRpSUUeNioEqBcc68+twdDoPjG999ja9+9zUetA3dDPraQCGz7lHQ0inXAsDMiJXWeyEUxSCSUfH/yI3kgyLiCKbEYwnGEIqKcn+XPMqMDgtmhwVuJKNPFROPj55y5dYtlLZMsejTnlEHl+sZtZWjjrGenQO4eXvCeMfR+i0KxdXDS7x8+w7jsmZ+1LI+a1kvezaLnuSlJHSFyGU1SQ7jOWKVoigchS0xWiS4yjiss1IdaI9xHrLHp46kZTO5MtvlE9dvsnj4CEuicAZnFQ6I64bdcsTEVZDNRRq0UufafEuKsJ5vcclQRjPIejPaSnCsswbr3GDrNRhrLwJllRb3I0rm+ElDYzLtesOVco9/+y/+DC/fvIs1JdV0Qq8i5eEud1+4w71vfpsb1ZhRYaG2dGTaTUfoA6UrKctSNsAkT/XkgzQVs3xkMZ5HvwNotNGSDyk3pQh/tNiRx+OCstQCVTEdMXcoE0EF+n5L227JRJyzFNpAEoSd7z2pi+RV4lDt8lOv/jn0m2c8/M5bnG3W3N/M6S2kIGEuMKRrDQ1jo2WD0mYw0A1jR7Qcz4KSJrwxCpUyhCB5oikIgVyfy/M/+lL/mFQKCpUsoVWsznrajfjnB6K1NImUhGismg2L1Zq9/Rm7dkIOC/oYuTyuaRY9333tTX7Xt4TaEp2lzQGTEzYAfcSljDdCUlLIopevL2TcEAMpDU+ywYCSlIRxxKiISbr9DYlOJU62C1btCeNDx3MvTNicWWJsWbc9syuHKK2YmRLftcTTNQe7Oyz7ObbwjExgdHnM7RduQIwcPXhEu2ow+4nd0Zg7N2+ymb/HYtVT1KXIX3PEmoTVgu42RtJocxYPhFHPnggxSyMvRgi5xVWB3R3NdhlpNg19NuhomLkJZt1wdO9dKiNjuNB7KhRhuWG/HHNt94A3jo9IVtgVOQfZPDN0bWZ5tuFaHFFh6VIUZ+bAUkRpFAmttfQlBuVizll6dVkR4gBbGhal6eGLL76CPlnze9+8hy0so0uX+M5rv82f/ukvcvbwiOXxKVdv7JHWnqq0ZAXBB0KKZC/07qIo0encTxFRPoArpUpMYgzLgwuTpDBGk0MackOURAYWlul0SmkdffC0fkvfeXzoiVimszFKyxQjxkzUQnSKwWO1QneBWVlwze3w7f/lN7j397/CK//ypzg9cDQd+JRxvSxerPSPVBRpM8oQrHTToooi5FKSc+ojRCVNU61FhZmyQGsy6UKLkYgf9eQA/MmAW98FVsjDPOSc/yWl1D7wd4HbCH3pZ/5QonNWrBaB7cMHHN9bsJ57jB2JGUaLPl5CQBS+jyxWS67duM3eeJfFdk7hYH+6y9d/9x7LdU+YFQSjCSFQlRWKBH3CDQYply1dA2iwusTHgDV28LprwaKngIRNWyxmiKvLmKhQVtH6nvefPGZn37I7cVSziqvP7ZIOLdoe8eh+y9U7N6nrCtP07GI5nq+pJjN2ihGnJ0uKQ8WsqDFR06wbFk861l3H/qjm8tUrPH/9OuuF5/dffw9DBAut7yhdZjwSc47vEHR8GuLLUJxHzJEGiKiOGNMx2dFce86xfKJ5ECLb1RrbKA4v3cSfLlgdHTF2lh6RShsUm6MTriTD8weXeeP+Q1wlupCUFUk5DJYQFetVS2w840nJmo40YN9zBpUjOQ8p00rJTH3AI+msUNpIeZtFjj0JcHtnh1tUfOXnfoFRW/Dpz32WZW6pJgU/9eqP8Qv/xX+NqkrUuCAtO8oUaC2kCnIPyz4w1jCqKgocoffg5T1J9qIeYLJSLaosC0poTUGs6Vo216IeMZ44JvVI9CB6nxAibduRDUynNbO9Ec6JCSsmeUIX1mEjuCZxY7TD27/2u/xff/cf8dmbl6gOprz0hZdZbyf89u99g9ViTtO0JCdPdJfEVh2GSQIOMFp4GmnQV4SMT/liCuWMIeQ0SKsNKp0DWv44KoU/uUrhJ3POxx/6898EvpJz/lml1N8c/vyf/PP/u2J1tuH43WP8KeRgwQxk55xIGvq+pVKaPvU8fPSI51+6zXTvgLx+D6stypTca3pCVQ3uQKi1hTZIGVYacIZF2xPOIuOipM9ZphgayrGQfc/PvdpACtLLIAZIStKPMuicMQ426xVniyVlMQMHbgLJeoqJJ8aewxtXGU8q2uWKsbGs+ky7WDK5ssui7wjrSL+OnD6as1009PNIWsGj94+YTWsOLx1y5/lbHJ+ueHR0AgXszCquX51wuD+ma+H46ZqjozW+z+icUGkQ2+hzOEskJo8tM5M9xeG1gokuWbZHbJuID5mrh4dsTubQBXkaGo+zQiqer7ecPnzKJ269wNd+/022XYBSYB+mqnCUuOTpNh3tesN0xwzmnXPmAXITDxtWAhQaNRwXpJN3brEEnRL7GV6oJzz42nco3wrcfvlzXN7Z4fe//30+8+qnmfQlH7zxLtduXGYZW4L3FEWW6IZKAQr6TBsCtu+ZjkfU1uGSomtbejzBWNAZsTCcqx2lWhzuCgonhKbxyGFdIuYWqwqMLRiNJuwf7JNUBhVwhRXTHaCHakdrhfKJW4fXmb/5iN/6+V/hVpixs7fL+No+xaUZz5UvMDucsHr8lEdPT3i43fL06Qnzs1OMA1NYutyBNQQVUdlgnZX+mu8JKeJDoNJGjj7Zy+jSnX/+5/ro//8VjT8N/MTw+/8e+Cf8YZtCVqzXhvVKUSVh9Iv/XqSxWkGB8BaCs7w7P+Ll01N2JzsUtqLPDU9WWxpjcK6SSCOfsYWg0LMV4OV2HTFhh5/5i/8qP/6FL/DO977PG2+8wW9/51vMTzeUdQHW0NGASWhlSJ2U3yT5qDRyFrXG4vvM0fGauqzYGRUUReDo5AnonsPrhp3LuxxeucnDD04plWJiFcfbNdNmh2uTPe5tH/Po/hyrCmyvML5Ak3n8Qc/OzppxucUmuDTb4/TpAqUDd1+YcPulS9jC8vTxmnjW4VUgYmXHIqBzvBDiZJ3QNlNPFdO9kqLKuKliPHZYCz5EDsqSk/ffBr+ByuAw2LIihECZPUdvvM5epfmha5f4+jsfSMYBsjk6B9o4mthxtgrsqcnQKZemnSYPnXGhWulBSZPPc3PSoBZWA8NRa0xQpJBwDrp+TVUldGg4vvceN29c5du/8n/QPF1SvXybD8qWdZ0xBUysYhkyzjp0jngfafseUmI6GjPdmWCNwm96VJTKywxOV5UtOisZkRqpEKrSMp1OODjYwTlLDJEQO2KzIeQxs3JX+JYUOKWIJHCKWERSjrTHW+5OrrF3VvCVv/fLhPmS259+EQ5KZjcusYqS0j3bv8TVvWu88rJhtWp46613+c2vfZ0HR08Z70kuCT4Png6JuLNK0yNT/IbISFtcMjjvaVuDUpqsAlHrC0/MR73+JDaFDPyqEsDifzOg26+cE51zzo+UUpf/4H/6cO6DGxkWy0zXDcgvHcSMpKQjrPUgKlUQK8NJanj68D5X715jWk84CZ5F73GFBe9JKoupJksNu00tOhsqPeU//ut/i//gL/1VTu/dw//OQ37yx/91/t0/99P8D7/yS/zqN36D7CzlCKId8hGzFRRWGEaDMjTGporY98yfbpmNWkwAkwKFK7h0KfH8845yt+L6C3d5/M3vYWOkMgYVO+x8yeFzU06qipPQonyk9iUmQ3CettO89/ac2L9FirCaJ3SASW24cc0xO/CcrJYs/Yp1bOiHjrOkNAXUuVHfSE9BlTDeLajGTkr/5FHOYh2UNjIKgeN3Xqe0npg9Ok3osyFby9hawnpOfPIe1+qCSkdismitMDEI2NBZgjKsuowzFWWqhTA0JCNpLdLgQoHNhgj0QzNuoJhLYzfK1KBNBacq8tIrN3j47jHfe+vb9Lkhzzf8+j/4FYrDXe4cHKKS4URnVjONyj1ltnK/KANaUVjIMbJtA75fMp2OGY/HHBhH223ZhkAfh1WgNCoXAx5+UF8Zgysss9mE8WRMCIFms6LvVnha+rhlXO4wdgUJL+RlY4mqx2/gkCvczTf45v/4v3P22vtcv7qH3+3RV2b0BXSbFj3SbFUmGs1uKrg92eETf+plpmmXX/zffpnNqUdVDqKnVJZkMoEOIx4rkoZWZYJWlBiKYDG9wzhFLAqSDhKk88dY0H8Sm8KXc84Ph4X/a0qp73+U//Th3Idq1+a+az/0ajJFEXBK49Rge0vDfF1DioEHjx8xvjZhOpsyXy3xyTPbr1mdbjE+YZWTbmwUp2OIkWoy4ks/+SUq5RhFh9p4/vbP/ud84c9/mX/tL/wUuYavfutrrHwgW9mVxbaqh1Ga2JQB0CIYatuOs9M5/QZ2xyWX9/exKjG9XLAMa2aXdsRgNYh/SmXZrrbUvePui5eZplOaTUN7tmGzgWwKtDWs1g33HzZYY9msLT5EtCoI3rFd98xP52zX/uI1yqc2DMmyjBwjkWgyZQ3VGEyZaUNLs9Zst57QRSbOUhvDer6gUhZiwlYFXQe2EA9JUTgWJ6eUV68x0Y7TGNDOCA4sZkoHhXOEtsO3gfFkzKZfYgo3YBIzSiuiEnlzTHkgGhtS1gIOSZLBmHPGEHjQzTm7+hzVrQMW3z3h3dd+j1xYTpslu0Zz9+aLLIxh3W7pYqA0kH2mzpqgMsYqsnYCb42BtsvEuIYZTMtKGp19C6HDZyFjh5xBVyQjfANXloxGNZDRBAodqWY1MTuW24aYezwdnR0i/ZQj64xpNJNc88krt3nzq9/m7d/5PSptmezu0hnNjU88x9r19GgKZTHaUOoSHxIrHTi4dMiXv/Rn+d7b7/O173wLnSzZa1JMqCIymhoZPcVMCJHYBXF5aktWaiBiG7RSlEaa7fwxdAo/8Egy5/xw+PUp8EvA54EnSqlrAMOvT/+IL4IiUlYGU2lspShKsC6hbMToOMSHcbHlHS/OePD4kWi9kenBdL9ielBgTSQHLxLwAWJSliVPzo74+X/wC5ycHBE2HRM74mC0wy///N/jv/0v/ysWpyfs708lpCNLWZwH8Oe5u47hJaR4XqIrFouG07M1nU/sH1ymnk3Y5hVHm2OqwwlUMhUojWVsKkzOrJ6esl9q7j6/y9XbjulNYAdak/AqUU8N42nFZDqhrGq0KujayGK+ZTnfEn1EDVBZCU9XJD3QfLR4QDwBCqh3wI1Bl5Ftmzlb9qw2PaEPzMoK1fXQtBRYalOgB2OX1gptoC4sJvTExYLDwokCsBCPfu89IUDhatomsF60VMahY8IpjdVyNEg6E3XC64RXkZAzPkdCHvyTSoMZRqk6clr1/PbiPdyLh9ixxfaBfr1GOc3myRknD58wqkY03tPGIInzfaTMCjcU2mlwShpnKWrRaZwu1sznK3LIQmmuKgqtyCkQUkfQELGEBOPplJs3b3KwN6MuNE4F6koznlSMK4vREa861nlLqz2mtNS24CDN+NHrnyLcm/PNX/rHTBrYr6fUOxOqq3v4KwVHak2bPSpmSq+xQWNsQW8sy03DpJ7w/HPPU9kKEx0m1HQLDcvEQTnh8niXg3pCkZBcihDEEKU1WmScwpcYsjDMkB/yUa4fNCFqDOghYHYM/AXgPwN+Gfh3gJ8dfv37f+gXyllgpKWDEIaFP7y0rCRBGUiDiSVlCYN578FDxrEkjRKmNNgqsXupwq9b/LynMiUhZFxhJSvQKH7+f/2f+eG9F/ixF17ldLOhj1DVY957/IiT+T3akULXCo1QhFM8zxjQMCQPZQWoREyanI0o+jIkDMoWGF2waBY8Xhyzd/0yunLQRwplqZBSsd30bB6dsnd1RLcLqTI0StE8sKQu1WJ9QAAAIABJREFUMJqO2d2fkqNhtWzxPtNsA20TmKYCjcX7Ft8FMTYNtmMQf1Qg4xXUNYx3DdVYpMaLRWK+CIQWbLTsV2Piak1er5kgqrk2ZgrrBnk5oBQjLMvtmktVxYOoaBierlGi2LQuydmyXgemaAoiuY+YMpKVkIaGopysnz2NcupR1qBw6BRRMaAipFnBw6ZhVTquXNolrhbUdUVVZHbGjvZshVmsqA9qjLfkAbSrpJMpATNxGMUpK9/QSExf0wR0VFTKUhg7VHIdTRR4TIxawlu0YW9vl91JCbGlXSuiSuiqorCGum3wztKoSJ96ql5RU/LK7DbTleFXfvHX2GkNe7qiHo+JpWF25wrrKtPqjHMag6bSJQqRlkejycoT/ApNx7hSbFsBLPjzjS8pdkYzDLCZL1FJ0qN8Bu2sMB5zFoFeSjKm/mOcH37Q48MV4JeGppYF/qec8z9SSn0D+Hml1L8PvA/8pT/8y4jfNfs0uP+U9EUCFMaJFDYpMJakDTkGogmcLRoa17Jbz0BB0B2HV/ehhYfrY/rekzQSzVYoKBRn82O++p2v85mXfhg1rciFw9UTatOjXaYrO7Q26It46PNXqC/Kc5Uy6EROSlKCjCXElvWm5/HRKQeHFUYVPDk54Ydf/FHGu1PUWiqaIiuis/TBsL6/YnxLk/Y9yUaKSUU1qlE604fEpuloNkHoz13E1OBbT+jAdwE/NEHToNDLg1lSxDeJqoK9S5bd/RFlaUi9Zv40s14E6Cymi0ymjho5Rjtr0bagdo42Q1VWBN9iTEEE6j6wax27tWG1PiXGgKIgRuSG1CVHZ2sO2opZPWK+WaJlkERIF1YDBmmCJLcjVZ41VvoQfcAqRcqR4ArePH7Mzk5FMQ04U1LmjsJV+BjZnC6pDyzFsAh0NrRB2JnPMg7km6YA5y+m7YMg8BSUpaWwFl0qyJ4uRqyVh8hyuWAxnzOt93FGsbu/I2IhY7FVzbiraJ1mEXr6Vcv/Td2bB1ua3vV9n2d7l/Oe5a59e++Z6dlntFkjJBACBGKRJRaBkQGBDQ5gCEn+IKHKSdkuKqlKJQ6OKylSJk7FmApgsziSQEKA1tGANDOMNCPNPj3dM73dvvu9Z3/f99nyx3O6RSUOjBPKNbxVXbf71j2nzj19nt/7PL/f9/v9lLPAbd0Bp1ni07/7CSaXdzmpKqSL9Ja7uKWc7sllRrlDyYwoJFJmSJIorK7nRB8QtMRgiXFC0fE03uK9pTMwqFikUXkUZEKTS0OILT4EauvpdDqI8SQFCusk9ZY6uUP/g4SsxBgvAW/6d3x/H/i21/o8EsVA9vEaykJgwwSrAsJLXG0JThKCTPQjYdJdXBZ4WkaTFjFsGJQF03ZMGXJuf+MdKFWx9cI+WSyQWOp6StUp6C8ZPvqZf8ub7r2bb/66N9FtPM88/xx7l74ETpBrTeN8EoUEgORQCzYgUYuC4RG45GTzKYKNmLG5O2I4eY53veNB3nbvO3j6yZf4wIPv49jx0xzuTtBeYm2LE5IVvcbB9Sk3nqjpvrNHnQ2xraMrDFMvORgO2Tsapfayz+l2S6KaMptPaeclwYKdRZqahVEsCVa8AKSjyDy9dc36anJlbl2ZcbgVGV4vydsaLJQ1LPclcTKmYzQ6L/A6o4ekkJpaSGoFEMgIGBzCTgi1Q9cuqepEYN5EmDpWehWhl7F5PbC8dIxudox2XhODTQG63qOEwSuP07MkGpRJ/egW5GxIPZFBK7BjS2d5lW/6Wz/K7/93v0r2ypBV0WWiHKXJ2bu6TX9jjTOyy9SPaQTMS4O0LhnstEzpBx6iCASZYI+x0NRtpJk0mOmcXAuKXFOpHCUEViSL7OHuEV/68p+xfWqDbjenU+Xk3QpnNP060hGCot9hsHKc284e563qJE//1qf47Q//j2SqoCtznJIsn1nGn8i571veQLMh6IkCnVfIrEifJJ1RVgUbgxzXTChajfI5g3KFIu8woWGOxQlBERUz5+iFSJbnGK2Yz0AXBuc9TZvS0CfTMdIskH7Cp+PZa7xeN4pGrQzNbMbcBoqeZnlNk5uCyXDOfO6o5x7nWqT0GG3SH5NhuanaSsEazWyKXNVsnDjO1sVD6mmNlCRxknWJVxAdf/y5P+IbfuQeVFbQVR1OFmvsTkZYF5mJiMxUKqxNRLQRtejRpBNE0s/LuNhL+LjII1Q4G7AelvvH+MqV5xg3jlNnzzN+5gW09GgjMF4itSFXOePdCWYzo1itOBjVCOWRyiBdjnVzQgCtPNWgC9rhasd8GvBC0LYqaSliOlgl56FITEbrsFPP0c4cR2B8FDnaBd3k6ed8AO/IrEV4mcxhIpmF8syQlx2Cs9gFG8FEMAS0B+ok7NJa41HI6JlNG4pJTrVScngY2Ly8xdzWyDxBTm/GNuZaozsgqog2kryjkTpBS5RYMBQjSKfwY897vu+9nDl3ntZHTEzhJLXRyezTtKhpS7dXcGSH2Exhg8d4t9gniEXUm0oJ1yL1X7wQoAUyqmSuaxKVyWSKXqfCSZgBQ+uYW0feLTl24hhKBOrG4mzNbLE7UvuObL9mcOcGzz/5OI/90cMUQqZtPIqOkehc0L99jfvf9RAjU3NEi0XiYiKRJZdK4oWCIQhN2SnJOxVCGaROWQo+CLzwHM1m+J1dlgYDlvvrhHaHpglYFYi+Js9yNCo9X0y7tIXZ5DVdr4uiEHxkOp4xGY3RhaK7NKC/gHHOZKTxNTYk0q+Rjm5hyDopT3FqHSFGrMrJAqhJgxvVDAYrLJ3osXV1j+hBt8kwFKWj7Goee+FLfPrlr/LgyXPs/fHjlLuR+9dO83Kxz+5kh2AMuZdkMRBtICDxGhpS2m4atC9yHW9i3GKy22/tzRjVgna8SSM0b3rbN3Pw8KepJ3uMii7diaJxU4yZ03cC/Zxn/cwye+0O28U+JqRxmxQaWTqy0pH1ZsRW0B7mzLqCUHj8rACblHdR1CkwxZUQM2JoaYYKPwVQuKlD1wodBUJ0sWbMQMCKs+Qh4rTAacWSAykNXgi0kRhfoNDgAhVQx5qu1Oxbh41ppRssOjjqo8C2MmihmBw6JlOXjvKCxLaUoEVE5JAVAVMoOmtJCJapQK7BFCVW92it58z6Cd754Ldw+Y+eYK3OCI3AFRGCIkiBwJKNZ/SrgpE3OKlQLqYpgEjEcNdGonOkeLOUjRBvBh5mhuhFWuitoIyGUuSJCZJLmuCYtAmysrGxTtcoYt0yrRuOhMdGSefQ84Bfpv/VAz754U9hm4bSqJQZ2sByqykySzxX8aI/ou9zYhYoS0WV5UgEjQ24aGjR1F7glafRM2LhIHe4mQUX0E5BnlHHljBsCfMp6ysrbAwyDqZ7eGFprGNVVziZMbEtUmSoxVjyr1VGY4ieyWyEi7DU1fSqgugVk1nNeByoJ4K6VkQnUDpH0EMXapG72GBbjwope6Dwnv3DHfq3rXHq1AZHO0dMhzMCikLrxBkwgvlozp8++gW+9UffwbE7zrC/d4PT/WXe9rav49cf+yTP7mzR6/aYuzkotRj7paQiTSTiFvU9LsYR6atrI7u7e4wnY3yMzOcT2vmYtqlTDDkp9VWqnHYx0puMa/SkRoaACqCFReXpzqkzSdXNKPMOk3bGvJ6ijgyqZ7BWEEM6mAebHIAyZsmyLBTz2qHFwkADFLlBtCqFkCwmvbnWxHlALLrTi6QDII1ylZZokeGERTZpipJJ0JGFFgEg8Qa8C0yGI7QS+NBSFKm7lUK1Ftp8wNcu8ULtzRCoSAwWXEQYDz2PrC1/9yc+SNyd8fhnH2HQr9jV2/g8gJYYoRFNJNqIDBplJU0t04IPecpR8CkW3tmkBLRhwZAU4NqaqBZZBV6ClFjvmLkxNkqMLhhkPdQMZtszhJN0litsppgpMHVLHM7ZED3OdFZ47Pc/yWhrTFl0kDLcEnCJQUn3zDpNIfnTxx9huegxWK5YGvRYGfQpjEErQ5lX9IuMIDTalTRNg46ggqSepaIl0anwkxqT48kMguPsuRP010/R+Cm7WzdoaBAa5nVL7sBkKgWvvMbrdVEUpIR77jtBv+piFIQQaaaO4cGMg8MJUWi0KIhSEa0iNDkhLuLDoyG0ETcL3HXHXah5wzObVxhPD1lZ67O2vkw9bRajSUVwgohFacVTLz7NnpvwbX/r/YT5jP0bl2meusZ33/N2Dg8+y/BwisgzWhUJNqBJwSQEh5c+uRD/XJqQRJAbRT2r2d7dAwmjwz2aw31c3bLa7TAkJy6Skko0YAiNwx9MWVupiNrjcg0EahepuiVLSz1E1ExFi20j41GLEZIQBD46vHVoJcEnk48UisAiPSgkm3Um0mJAOUTMEwRGC0qhkbbGdAxKJfYlyWaSJN4RfEzyXSWTlj4n8Rq1T/Fiqfm6cBQ7AE9uUrKx1gbnHe3c3po+AAhnQCrcNEdrAc5Rj+c00TG/MeHejeO8/y3fyud+7beotw84tbLGK35OrnNcsGilMFJgG4eYCfzEYet0KnIxhaEE61L0m095jEYV2BDwTZOKVCbRWQbSE6Ol9hbHlFyCqTuUqssKy0xvzLj07CWKt9xGK1sO5nPc0Zz1tuCucoXnPvko289dZq0YMK5rhFGEGMm7FeXpNczZZYa5pdUtIzUnTC2TZsTR0S69sqDKc8qyQ6/qU3YqjFgm+EBHGk4fO4X1PW60QxoHduZonYdMkilFay2j8ZBTy2us9SoINaP9I9AKEyF6i4oS838DzPxF1+ujKCjBN3/r2zl/9hy7NzaZjqdsbx+RFYKim7O7P2Ze14QgiFbROoloMzCJ1CCiILSRk8un2D26zGw+Z+9wl1OrJcsrA3Zv7FHPAs6lyiuUwBjFwd4en/ziw/zCh36O+972Zl74zJC9516kI0/zY+9+L7/x+x9md+6g1DglUT6iFmIAL/2iwcetFB+xwIg3TcvW7i7dqmA8PKRSAkkgU5ICQ9QB6y2dqMhkjpSOo2HLsY1V8o7kcnNAzCDLBEoJQBEiicjUAuklUVSelpZgQZtF7oObIVWBdwIlk4MytJ46JrpSoQoEHi1VwoyFSBZBk4EwCGFQWUbUKt1FhcS7NsXbK40WEoNLRKt40wFJyiawHmlSglVVGoqyi4+RuganYwpWkQpcxDmbWqNBInVGJgoKoYmuZj6f8N73fytr85InPvYZ7jq+TtnJaaWjFo6yKJEBCmWYzR1yqmFM4oYGQRTulqZELTgSWhiCdQyWlrj7jXfy4H0PsH58ncbWjGdDDkc7HAy3uXz9WabDQ2Z7Nc24oWcKikrzwjMv4tWUtXNrQMGG6nJ/d53Nzz/DpS8+x2o+oGkDmcwISpI7we13nac8VyLO9mg7E6pBhakqQhNovAXZEKIjREtbz2n9hJ7rYYoMYQ1+3pBjyEROM60ZHc7BC+q2xeaOwaBAKcHu7g5VV6L1gDIvOFh8Jrvdgmlo0DEm+ORrvF4XRQERWVrtcMcdJzm+3GE0HHLm3JT752c5GE24fG2b3d0jdm8csLt9gLP7xFkBWqOFw/mIiorl7iqH7hpFp2Le1EymY6puSbfXpZ4e4mrwMiBlQGcZZWV45PE/5cd+4EPc+/V/g/radczWPl999Et8030f5O995/fyyx/5XSZEdJZ0Ch6QITGWJQl8cpOBqBeBr9Z7DodH3H7qDPP5mB6pqSc9FJnByYboPVmIqBhQMqexDWE0Z9BbQkVPVIay06fIS6yFdl4zHM6wFqoMeksanUE+bnFBEaygGXtqH3G+JYosSYxj8p+31oHzKFmTdypoGyqjKVHkIUXrWxlwMoFJhUoimIR8k+nYJDVGKbS1GOBrLASZALAqxaslcpJOEfTO47xbNPYSucu6AMFhg8P4SCBQO4mKEunhdG+VH33P9/CFf/NRwvYe2cYqUYAwyRPjS4GKGoOkg6ZHQVZL8iJfZOwlMpMQabCslSB6S21rvvHr3s3P/Ec/x9133ocuDPO6ZjQ7ZGv/GkeTHS688iW2Nq9w7cVNNl+6wcHWHrPZGKXgwguXsMFxz+l7eGDpFO6pa7z8yFP0RA6kDA6hFcFHjq+ucea20+ysz2nWM2ShECoybKZIL1EiEDKNKgRCJFCMFIpMJlCRiikj8tULl7jw6i7TUYMbg9I50cLcN+meKDRVR3P51W22N/fo9jqURR8bW7Q0NM4SWkeWFa95Ob4uikKMgaPxHlWVsVys0esoJm1GVBIXBHfee5qjowl7O3tsXd9ia2uXo5kn1wrnLDf29/HtjKroItB4Fzg8GqOc5ljvBKvLA6ZHU8ZHc2IU6MykzreSXLj0Ml9+6Rl+4F3fzsZ9t7H11adZ3erw3Mc+xdu//32887bzfOHqSzhV4FRKCgoZt6TFcFMXkoIfJAKtMiZ1zdLygNHokOUWVAQjFFlUYG61IMBLIpHW5FzfO4AcZKUou0v0+gM6VZfNzS3294ZA4t0KkVDkGycr+g76veNIv8JTj7/IbDxcPHEKJVVBouJCEtt4YhaTGSh4OrKgkgYTmjQaEwvKjVqYliIIeTO7MmnttdTkWi+8/hAJBJUUlUh1c6hBUwdsM6MNLdY7TKnpdCuqbom3lmY4ZDypCXhaH8mEQiqDHU34we94LxsePvJ7H6UrJMF7UlqhQcccj0YITaELpClpTAfTCBQBYRRRebz3CQYbA42b4RpLcJGNpRVuv+002kR826KkREvNcq9PpwNK3sXpY+s8cOf97L9ll2ee+govPv88R+MhkwPP1aevcn6+ju91ePoTX6BodUIHeguFSbFvInDP+TuolipG2YiZtzT1mLaBIwGZyqiKHAvUPiK1pGNypCqZ156pm1MqjYiRw9093vGWNzLcH3Ph2Zc5Gnuk0AQH01mN1DlF1iFYzXh3jtqQ6CripaAsMnSssba+Ne59LdfroigAHB7tMRrts9bpU5UFurNoSgkoqoyl5YqTJ5e5++5TjIdjJjNPlXc4PDzkjz7/eQ7rmrLsoHVOFBpnG3Z396lkj6I05LlhpuZp6x8lTR0orMRbzyc//xm+733v5fw73sCzjzxM73KGGjc8/+nPc+fSEhdQHLSWYCStlmmUFxcJ4gAinbFTApIEEambhqpbpjGqMBgFRiXysTcRFW5GpwXwMChy9p3laDgkFAWEjBgM+3sjbmzuMh3NKfOCqkoBo611IDy9fsH5288TmnWeeWKLTLapOSoCyKSXxwsIChkChPR4RaRTGDQC6WKSShtDZkrEAsQbRbIVS63wrYMYUSqlO2mVMOf+JiRJydThlomL6QDnWxwOnWuWVpZYP7FOWRW0s5ojZajDHr5NAzljcvzcsTZY4kPf9R08+vHfo9nbpZNL5q5Ouv6oKShxOiO0AR00HVVQyQJR+5TLqB0uxORjWGgj0ssSNDUc7oy4fv0K68ePIWRGWGRPFCbHtiBsQEdFd1CQ99apTryVs/dv8MJTL3DphcvYnQZT1Fzc+gpxe05lCobRpXh24VAIlns97r37Dq7UR9Qrgb1mAn5MjAK5tERHZPRUgUGRy9Qg76gOo/0RLz79PG+/N+POcwNmszmEyN//8R9jPpry6U98gg//wWNMjhymY/BY6iaytzujkAU//qGf5Vve/S4Oxq/wkY//W67uXkSXGW1sbhG7Xsv1uigKQgr2Dvd5+eLLzPtrdDsVolCJ9qxAikCnNJS5pio1x1aX8B7KrMP1zTIt+MMRq6srCCTWBpTIGB4O2WaXY4NVijxDG5U+sK1LLXmhqMqSLz/1JK/ubHJuqUtnYxWp4FR3wPZwQmXgbGfAaLyfTCUmw7nk+FtsDVJwh0yx2yGGWzyATlWhpml0GgIE5xHSoZW4hSlPvoXEEuzlhnlooImMhlPmtWf/4JDhYY3RiiovUB1F7SdMjmomQ00Rc9p5pF8sUege7XwfoTTaxAXMVNDaQLBhsa1Pi9gg6CiTLLlagTbJCKQVxiSXoY+CmEFdNwgpMdoQ6mkqKlqlLcEilk0JiSKm6HGRmnvRCdCSPM/o9/p0qw4uttRNjQsyqfrU4ggmBY21/OD73ktRT/mzz3+GKtc44bAxIrUmuohBA5qAA6npVV2OQgSbJMzeCwL5wnopb/3O0khCrLnyyg2uXr+O6WUY3UHJLHk8ZEriyk0nmdD8EBsiKycHVP2C5e4SfdXHXhnSmwgOLu/Q0TmNdYRC4YRAaUVTN5y/83aW1pZ44dp1GlrqYFE+oNAYYah0TiVzZIQCRUcXjA8bHvnjx7h+6TJff+830lrHcDQixsjtt51BWsex7vfi6PAbv/cnjOcTqkGBdYHZ1PMjf+9D/MxP/edkWnDlsuSNd7+BSX3AiH3qIPn36DO+PooCMrI12eYLTz7Jul7n1MpJTp/Z4OTpNQplcSLlzblIuisrTSdryLKC0QwOZ4HaQ4+cfp2zHAyjxqGbjNlBy1RMKQqJMYKm9lShojEwK6ZUuWa+d8RnP/cYf+c7v4dzy7expzKiaLgtz7Ee6v4S2+2EsW1QUqLIQVsWcZ8LoUxAKIgypn5DKxDdLuV8Bo2gjhU1OR0xpG4kOYaAwgpBqyW5EqzaSGcCdcdwZa9m27Z84Ps/yB2rfT78G/+KmW1pqgozm+F3YEtAf6XC9Srue/PtPN19ghvWEdB46Yg2IoJGC4VVHh8cTsOUyHodWSszGqk5rHIKlU4NQYMwkkJKhI0EkRp4diGxMTHS8YJSpmIWtSKLksprgnW0cdFrWSxH4SS0MB+17LpDJvWMST0mhgblIKAJuSA2LSe6Be9/8AGe/v0PY4dHtKZgHgRd1WGl7JErj3X7LNerTLuGulSc6HbpzWeUMcO5GhNztMhool3YHhKZGwm50Tz/0os8/+IlVk+fpOqAkBOEjGhtkFkH0z+FpGFp0sXHOQfNAZNmn7wseNcb78FxjeHHnkEFy7woaFyGVIay9mRI1pY3eOC+N/DCdJPD/hytLGHaMsoU3U5Fz0psXmNjhy6GzNTM3JjnLsz41Oevc/5ExckTZxGi5OCo5tpuy7XhHsf6PdZv2+BH/+77KJZzfvNff4LJuKYUXX7+J/9T/uOf/s/Ye/Y6n/7d3+apL36Uw6Uh3Qe71ER0nZLA/1oFt4LABcXmzh6Xrl7lsaee4vEvPc323hFR5GRZhRAGQmpWaSlR0hAjbN7YYjQcIxQMhzOcE2TKUM9qtJTUdcP+4RFISb/fI8/yRAYKHusDSmtCcHzucw8TIiwtLVGYDGk90TpoLGfXj3NqaRlpF8lYRt8KHY3RE3yEuEDeL36f2bTB5Bl1O2c+nyFUsjhrIdNdValUYFQiRxtjKEyOBPpe0Jt43n7yPP/gx3+Wv/+BH+Yb7rgPeVQjJjUEhQ+Co72aKxe2OLx6yInuGudWT3K8t0I7rmknDtf4xQwwyX1vdj+cdSghyLRJW2uR4LlKKzJjQIpb5G2pFFovILEIMp2KTG40xiRV481AFxYWFecdEZ/gvBKsbTk8PGRra5u97R0mR0Nmo5q6tgQNCo9ua77zobdhpjOeeuxxekVBptKxRUpJv99PMFiZPBoawerSMoNuLzU3pVyE+wacmiFNjTYRKUSyZzci4e9uHPDslx5huv0CfnQFMbtBGUdUckrftKz1yqQdyDMm9YzxbIpwnsLDqirYv3iVo8PD9D6xSKhe0G7auua2c2fQAq5uXkV0cmoNjYxJbCVSoYw4QnSoRR7m3t4hjz36JFdePeL4sTOsrx/HxsC0dmwfwG/+zie4vHOAL3LWj2/wI9/7ffzQ+97DMd3lH/3MT/NzP/gDfOHXf51/9Y//IZ/9nd9kZ+86g0HBwcENpHSYTOAWbuLXcr0udgpJc1DhhMRqzXzSYq9sc/eDDWfP98nyNGFAtYmN5yNRGawN7O7tYaMjLzOU7pBlAzqmi5/vIihRCibjCVpCkXfJcs+4tikmO7a01mJUycWLF7lxYxOjkjGrUAbhArG2FE1gVRX0BIyFJIqbmue4CC9N4pibBeFmzJh1SfMvtEJnC6gqYsEbSB8SKSUaDUJRZBlNntMfW+7tDjhfbTB++EmsnfGmcp3NaoXnhxNmKhLyCiM0w50h4SDQdyVLouKhux9g89oOI+tRGcgy+SJEivRFogm+QQmZFrTjFjpeSoXKDEqnCQoxoJUm8zmuaZJM2RiMcxRGkwWN1uImZxUQyWNAROK5CU3w3jOdOLJco41ChwKSuTM1K5uak6bgb77pjVz47OeRtccYsC5lZ6bmp0zHCKnQUiFjpFuW5CZLRUtrCJIgFFHUKe8iyOS+9GByjSISXMPjn/803/7OU7zx/HeRl5rR5JDpuGF7Z583ve3dBDw3hju00mMyg5jB6XyJ7We+xObTlzne7eGUXng5FM41GBTLa8vcfu4sR3t7CKOYSMduO6MtUsHNdYaRGqkdYbH7nc0bXrl0jUsXrqMRnD1xF0JpHDCaR2yE/+Pf/BFza/nQD72X8yvHOLt2nA9+53v5lrveQr015X/4qZ/i6NVNOlGyHlrs6SVWNnpsza6RZ5oqy5iPm9e8Hl8fRcFFtOpRz6Y07ZRKVeyN5rxyZZs7772TtbJDWSabbVPPcc4RRc7u/pDNrW1sgCKTzOYt8ybQMRXCJ727MRnzKBhNppQrPaqiw3g8IaoE/qjrGu8FRzdu8NLFi7x10Cf4QJYZRIwJBz5uWI8Zy2jmzlNrUN4vRpHpkx1DuKUcVDrtCqICUygcjiADzjkyDIvScOu6ScvOlKIwhs5wzko34z7TZ/OzjzO5cY1K13z9qbuIO1d5dH8LLytyXaJtTe4LlvUSamw5N1jjZG+J0cF+4g4uyGdCaqQGqTVygXlPScopB/OmAEtIic4Nvm0hpqOQMhrpPIh28XeLEYsociJeLJKGVVzwIJPgKfhkTBTia45TFSW4gCFjJj0RR2E9f+PsCeqXL7Hz9DN0lUFFyIqCyXx6KyNCmeRy1FpjdExLxTlkAAAgAElEQVS7rpgCSzOVFJNJTaUINkKUaK0oSknTzhlOa+699xg//MPfwDe+6+3s7l7ji48/ztPPvsCsbtnbH/Lzv7DC6smzBAFl1YVxy1K1irq2ycXPP0nfRmShE1QHhTEGgUI6z8bqKioG9va28Sdzrg0POViyEHPKqJExHTcb24KtyaOBpuba1S329qYoClYGJxA6w/nIdBpRqsvhUeQjH/kMuY789Pd/gOXlAetln73RJV78whe4/KUnuXNjjcK11D3FcLXL4XQPlSXDWWkMavLXDEUfXGRykLT7zXRC1smY15YLr7zKHZfOkue3MRh00DJPxhZpmc7h4qvXePXqFihBb6m8BakVXhJbgTFZyrPTGd43NI1Fq4IiL5iHOZnJEI0DJKPphAuXX+FdD3wdMssAECogI1ROsOwkPQfap1UWYwBSug2IJAkWgIgJdNtCY+cYXzOcTRLRWBeEEJOUWKS0Xpk6c4SQItC10HRdoJxb1nRGFUTiATgwa8tMVmGrdlyaTRbb6Iy2hW5vGeaWSuW89Z77uPzo4wgWiVGQ/qcDiwj4RYhVAGLEZDlKpzIVYsTFQJRJq0AICwJVig5XXqFixKikbLwVJ48EIRbZiykaPxKSJVsm7oPSCa5iUOA8wYDxsNQq3nbqLFtPPUVRzxG6h0hzGRweYSTSKHSuF2SkiNIaKSTBOnKpyIRKjIcYyMIAaRJCdj4bU9ctp093+Ymf/C5+9Me+F5TkNz7yh3zyU5/l4PAI61PK1HTq+ee/8i953/vfz+1vvJvoA5nocFIInnnyIvXlA9Z1hY0qKTgX7+9NoOHa8jKHB/uMpxOOpoFD1YIsoBEQJI13qNLRRI+NDuMtWnmmkxpnIVcFa0trZCpjOqvxNqK9phAZey/v8Ie/+SnONBlvPnMXmy9dpt484ORgjeV3vo3Z0QHteEj3zAZf2rpOrgPqhEZriTYZyv81Ey/FALtXd+l1M04cX6Gb95jub3Px4kWeXO+jTODc2Q2qKnX+nfVc2zzkK08/T2OndFcyOl1Db6lkbWOZVy8EsElIAiBV6laPJxO0d8ioCa1HKtAL2rAPjhdeuUB86BtxmaTxLZky+HlNoQvWMKyJgkt2muTVf0lHV0hoXEMvE4mYrVPcWJ6bRHH2aWutVSowQUgCyYsQCsPwaJ9rN65z28lzVEs95sOasD9hI+9w39Ixxls17SJtpxGRol8xWF5i+9ULPHDuDr544RU297cQxWJkKEGY5FaUQVJoSZFl5AGkjyilUVohlUSYxWsiuT+FkreOOkoqjBBJoK0ST5EFszAiEHoRUQ5Jq6/ELc4lCJRgISjSSOXQo4YHV46RHU4Yb25RSEEtWCQzOVoibQzMmia9BiWIWiIyjSly8pBRZSW50sk74sFg8K1l7oasb3T47u/7dn7iJ/82WSn4xB9+jN/66Ge5snmAEFCUOSqX1Baikjz8yBPs7h7ygz/xI7zxDfcy0Br36iu8/MhXWNEDtMhoAyAlwqcwHiEj3aqiX1XMJxNilnHp2ibXlx15dYyyNZhS0djAtJ0zj1P6pqAIApNFmsYTLHTKnF6/k7gjUeHncwqf6GLve8tDfOs738SbT50hHM7ozA3bVw4YrB3nuz/wQT728f8TszZgV1ief+mIO44vsZSXtK2lk+dI+x+gKAgh7iGxHW5edwD/GFgCfgrYXXz/v4ox/sFf+FwR/NSyenydB++8C4lgsjdhc/MKFy9eoOwohuN9+ksdfGiw1jE7Mrz00hUaD+v9DF0Eajvi7O3H+fKLYhE+kRiFnsTb8zaQBbUIqYgQ0nzee48sC168/ApTYRFVyXh/xnKvQ6gbhLX0s5z1oiJrJ8wXi4UFzfjfdakgGE/HHM8HhAUO3C9M10orBDE16CQ4m1BqeFAodJExmg95/KWvMjOS+48dQzQZYjajUwfOqoKtQnMVi+xIQs9AP2ft9AYvPPGn3HX/Azxw1z1c3bqOjgpt0pjRNQnFJ6XEaJ2OF94lDaLWaG1QSiG1RGEwpIajC0m+LaRMgBWjMSGh5TQxfV9IghIo7dPuIgAi7aAkiSuJ84vmQyQKhwiOJa94cP0k0yubiNaRFQXTAM6l5CiXnOCLoJGAMhkiM5Dp1BNpEghYeZAukueaUI+JTPn297yZn/8HP87q8T4P/8kjfOwPPsNXnrqIy7uIoociEmOkqX16bTGxNZ999kVmv/rr/MLP/gz3nb2Hz/3J7zO7ckSvXCIqRaBN6kUhkVIRg6PICqJ3tE1Nmymubu1wddTQDZ6zJ48jlwoOmHLDj7B+xoleh47o0e1KirxDCIcUuaTfyzDK4BuH8g1+MuGXf/Gf8O43vZl2b5frzzzN4bVNXnziSbaubvKWv/kA25tHjL2mc2yZL//Z4xy2MIuaHoa2toiOYalcArZe09r+/1wUYowvAm8GEEIo4Dopo/EngH8WY/yl1/5k0Csqbj91mjMnNmjrmuVBh+1NONjf4+rVq9jQUA1KZvWY+bzmxNK9NG1Aa40uNCaHWTPi9ttup57NEQtoaQTEAh0uSDkDwYmkFQgehSLK1P1/5eqr1AKOnT3N5vVrLK+uYIqCUDs6nZKVXp/ueMiERI1isahvRev/uSsA48kYvb4KIi66zUU63iBQRiLdgh/IghoNSShUGESnYLs+oti/zsmNZZaqgtw5XAvrRcapXsXObEhUkUZ7Wh3p9CuGwyP2tra55/x5Hn3qcfabKYNjBSIzOAHUEemTJNtog1QhOSYXQBYpkzrR5BlKG5xztNamuzRJU2K0QNmEg5M+IlTaQC9ofEgk2iUHpRBhEQfl8YH0nkuYtjWym7NeFHQcTHeO6ARovccpjVMQtEAUmqLqYPKMENOOJhiVkpbzDBM1om4odIpWm40nbKwYfvpn/w4/9lPfw+NPPcw/+Z9+m6ee2cY7jclOELWiaafoTOOdxTWWjjZED0JonPRcevFl/vijf8iJtze88OhX6OkeShXU3qYo98XEA8B7R69b4RZ9mJYUUjsZ1YwubpJPPRxvuOaGvDDfRsSIXTaUpGOqVomG1qlyyjI5cdu6QThLEaEYj/i1//afcnjxCmE0pgCmkwnr3SU6seTZJ1+g6Cxz8cY2z13bwy1phlPLyWyF+f4hm9MbRGte83L8qzo+fBtwMcZ4+dZ46t/jChLMhqR3rkOdt4zHM1A6ufnagrwWqOmMJkw5nEy5sXvIcCB467n7EbVlxD5bbofd+YyH1u/AzBL4dKobhMkonKbjB7QeJjrdeYLzZLGgXfAYV5uG9mDKhWvXKc6dwT7yMB0ctWqZmMiAnNvkEifUDfZFTSMUWYhILxKYwzucTNvmEBTRVNj9SLORgfJ0CBStw3UypBZoJME5CJ5MJ+BtUCnJaDUUTJRlP2u4PLrK1vQUPTMgz3MaN2Eg4JTu8bKfEectzgpckAyqAX40Y+fCRR54z3287bY38KmnH8W0GTGDWBjmc8hsyovsqYJcBqK0WBnQMqKix3tJNBKRS2Id0EpiosKLnBqITNECpLNI0VJLB0YjpECGJOlebKSIhKQZCCBk2jU4qXBZRmEb3tqrMFcuUdYtMuvSoDEShMqxRmCUp1dGtB2RKY80miA7KJmjqmXmyqKaGaeAQeM4+4YT/Je/+EO89aE38yu/9qv8+m99nKlTqLxCdAXTcIhuFRUKO2sJzqdiLUuCjMxFg5YGM5lw7bOP8/ilI9g7wChFoyNjIPiCTlA4JZnJSIbmpCpRrWXYDVxbOSBOIyuzku1ZzcWtLSaqQUrH6mjOjhNcHl4lhMCkOU4965MJ6HQgiA5tk+Ec1E3NsQIe/V9+BdWCl5oQA2PXQEdw6oFTXN99ieF0yPlv/jq+8vwmewa0icxnDdobulSMr+9T7/3Fa/DPX39VOoUfAv71n/v3fyKE+KoQ4l8KIZb/sgcLoMgls8mQSy9d5Pq160wmo4XzLmLrBhEjwQV86zk8HPPq5jb93hJh1HDf6u0co0c7tVS9ZbpllzLPFuTgCPFr+oGFwydRgeLXQq+1TK6/Rx9/jKUTx0Eb6tkMHywmM+ACy3nFuqkoSX6FmwaItFFIWoWbT6i0Zng0Zt5avIipdxFZoMFZjAQlUsrF4yJSpDO7FppcG4iB2XzClWtXcSKCkugyQ8TIsd4KFTlu0hBtanqW3S5SZxzu7CNmDW+89356VZ/JtMY6hw823YVswpMnn4ZCSpms0ov3SSzepyDjraNPlmm0SUkpUqpFUtLNPIn0c0n3n4hbqWpEvHR46QjK44UnyohnofVoI+uZhsmUIsvQRZF8FDHSesdsVjOfzmibOaPRId6n44UpJFkGUbVM/QinZjTuiG979338i9/4JZZPHeO/+If/Df/rr30ckZVUvR7CKDweoQNKeWzTJpOXyfE+0jSOpnHYxuEbR249qyLH7xyRu2Qbn1tHe/PzkmYOxBDolAXdskQYQasDQ2Up1vqsnt5geX0J2TG0wVN2Ck6uLbFUZMyGU65evMLFF65y/cohbQ15XlKUHWyItM5hp3PuOXOMKnpOH1vH5Jpg0qRnUJWsLw/YOdzlxH2389bv+Ea+5Tu+ifW1Ajf2uFlAOEmZF0gB9fS1L+b/30VBCJEB3wP8zuJb/xw4Tzpa3AD+6f/L435aCPGEEOIJAhQKQtugSECYTENZapp2yuHRYZIWZwVVp4fEcPH6NqbToaMKdr7yCveywdGr+0SdOtRZVGgHOghiDKTOQmABUE+cygVxOBAJErJM8ZVnn6Vc6pEvD8BolNQoJQnO0pGKk8WAXisSzk4JGg1OxIVk+WvBpN459g8Pado29RB0Mhmx6GWoBfxUqYSST2LNJGwKAnSu6WQZCs31rS3q4AhaInMDUVLpkrXOADH3UDuEkHSXlsmqLvs7++y+eoXT68e47cxp5vMpbVsjI2iXjgsxRmIIhBDQ2qTJAUl8lPAWC8eWXABwFgQk732aekidNAQx2ZOlCLCIXUPGFCuXgy4UptBIk1KvHA0IR2waSg/5zFEIjclyrIRWLnYUUqK0RmnNbDrn6GiY0rxjANHimRNUjSoc03DEe//2t/Ff//I/4ukrT/ELv/jf8ydPvIzpajAZrU9BK4RItB7vItJkKY26Tb0EZy3WNqlQty1LZNzeWyObWvKokFLTeIfHo0yyqmsCSgaqqkTlkgaHzyVjAbHXweUGq6EVgVk9p3WBvOpyarXHWqVpRzU3Lu+ydfUQO4c8yynLHFFK6nbGwGjOdjtkQlLXMzr9khpL4x23b5xitnuIGlScf8/Xk7/hDu55wz284/4HyOcaPQkUMTlEZa7Jitce8f5XsVN4L/DlGOM2QIxxO8boY5rZ/W8kDsT/44ox/osY40Mxxod0JshywbQ+IOiG7nLG2lqX06fX0AYOh0fUtqZTVfQ6vURR1o5We37wg9/P0YV99r54Bb/TYL3H6IxS5lReknsBzuNIdzMRfcpFEEAMhOBQOh0BvIxs7mxx1NaU6+t4KYjREVyL0RITIrdVK6wFTebTFtmrpFhLv9TXvsQgORqNmTY1Os9uwTiUJEXaS5Vm/kalKYm6tZchKoHRGZUu0B5m8xlT3+AzhSNgVE6sA2cHx1gRBu0CIQaKbpdOb0Bja66+eAE3nnDfHecxQoKzlFoTmgZvISbQZBo3Svk1bUEIKUSVkHYFWqGMROiUtYgWqSgIjZEpS0EqecuZiQAvHF6m9KhON6PqF5S9HJ0BOhUH7SIbMsMMZ5QiiXVaQGSGoBYKzJj6L9PplJ3d3UUKdNrhaW0oOhVFv6LodXjzNzzEp77wML/4S/+MV3eO6Kx20VWXiU0dgOg9KkC2eN7gA8ELpFBolazhwoOJ0PGBO3rLrDYCPWrIhcETiUoQ1aLZqjOM1mRSs7a2BplibOe0mcB3C2yZM8NRx4D1LdPWsnc05Gg8pV9q7ji5Tj/PmexMmB9Zykyw1O0SvSWKSK4Ud66t0ashQzCejBiNx9TOcvz0CY6vrOEaz/pdd1Ded462C0sby7zrrW/hRDVgWSg2ypxKe0wOefbaj/V/FUXhh/lzR4ebEJjF9QHgmb/sCYQQFL2CmHlmHKD7jtvPn+DBt9zNgw/dw5l7TtFf6zNYGYCS1HVLd1Bwef8iB/GI8/ecY/rqmHySzurvePvXc9vxM/RFQZzWSQKrwIuICBFBQCzUdgLwMaU36zJjVI+5MRpSnTjBrLGp8UfARQutZ4BhQ5QUdUp4DlojjUkKRR/xMSSMG5KmbTgYHSEzg1CJ6ScFGJG2nUJKdGZQmUEu5MKKNELNjKFA0kFga8u4rpFljlzcPQ2K1azkRNGh9B4pkohIZzlaGHY2r3Lj1Vc5s7LKW+48T26hnc4IPuUmVmVJkRconWLD0khSL3YGi52PFkgtkJlGLApnkGCEplgErmhx8xTlF0apiFQ3lVEBoSHPE34tLzLyQiO9p2o8d/T6FHaxW5MQtVoYrNLiE0KhpaatW3Z395BGgzGEYJC6Q1YMEKpCFgM+9vFP8z//8v/O0dyRdfs0XjO3EU/SBwSgLLsMqj4STWsjkKTqyY6bxFCyaVgVijNFDzNsKXyGEBq70JYYqRKsTEiidfS7PZZXlolGMVOBttDoXp9oDEW3w2Cpx2BpiW6viw2Ro9GM8dGYTChObqyzNOhC8BgFnTJDiEgeJT2vOJn1sVuHtO7/ou69gyxLz/O+35dOuOfc1Ll78sxO2IBNCItFjiQAQSQIgARFFklRDDbAZNKmRNCQqVJJqrIou2RZKlG0LZuuoi1SJKtISyIJEAQRCWIBbJjZNLuzk2e6p8O9t2864Qv+4/RSLMoy1yXbtTpVXT19qvtWz8z93vN97/s8v8cik5h5UbC4tMT7P/BBppWld/QwT928xqe/+ScMfU3ay7nn3tPcf2aZI23BqW6b9VZCIgJSvXLr9H9QURBCtID3Ar/1Z27/fSHEeSHEU8A7gZ/+i14nb7V49LUP8Mgb7uPMuWMsrGT0Vgyd5ZijZze497XnOHr2KP31RfaLMTaUrK0oynTM5y9/hdWHTpB2u9x85gqyqPnIt38Hpw4dI7YCVTf698bi6zHeo0NTGF5OcvDeo4xCpxGD8YDnb1ynu7EBkSZNU5xyzP0cpKOtIw4lXXrWoKumk25ig5aiAWMET30wXpwUNTuDATWOVjulqArqek6wHl83CVM6idFGN9t35wnW47xFB0EbRV+3kAT2RkNMGhPHMco01OM0BM6ub3BibZVIN1bl6f6IRBmKquDW5Sv00Lz97gd46PhxTOkQiWKhn9POMpIoQh30E4RqOIVeHhwfVFMcnDggMWtxUBg8ColRjaxbhIOnsGiSzKRqdkMaCN5hqwpbVyghSJKIyBgyJOsm4miSEyuB1zTjTAQqSIKUICRSNOE0DXZPYOIEHcc4BNqkxElGbSWCFrduDplOodfpEbx8+WSIkh4tm+i42XjCfFo0yVpoQhC4A+Wjc03SrSwqjudd1nSCKmqk1pRGY41uJNZBoIxptARes9RbJoky5mWNbndoLS+T5T1aaUYapRihyNOEbrtNlmY4JINRwe5wTNxKOHx0kSNHO7RzCb5EiYApPBuqgxnWiInF5Dnt/iL9To/v+ssfYX31MPMkZi+J+M0vfJXf+f3PcWdnF5O2WDmyzAMPHuLkSpdVH9GeaEwhMP9/JUSFEGbA4p+7933/T1+nFWseOLJGaFmubt7i9p1NZnYbH3KkysjbbWSumNT73N7dREaB1UWPbQ355nBI7+ibWDh3gitf32Rwc5OjD9zb+OyFpiU0VQgE0WCyxcvdxYOmWMAjtWrQ7QqCEJx/4QXe+ug7mjh13di3K9sgtLRI2Mj7bEx2GdVjimY6hUCi8NQc9BgClKVjc3ubfdFG1haJbUxRLzcYBYBARhGyrvHWNb+WkMRC09UJVawYqSm7g13KuiZWGm88oq5JpeT+u05yz7e8iyRpDEpZFDETnmAU0+GA4ZVr5Ost7uotM9+Y8SdXt0mTNq0kIYQmCar0ZbOdDv5AMQrKKCzhT48UQTQ9D89Bxqc0B70ECMEh40b85HxASIVC4VxT/JwXaK0QQeLqmrh2nGz16XmBUApnNF5qjJcIL6l0Q/KWeJTTzZMrNDsopQ1eVDhRog0YI4mMJo1SYmXY3rfN0/9lriSNylGrhgRVzGogbXokrtkxqgMWhpSKfhxzLGuTls1xImQRU1FTakEcDAqNjjJk6ei2Oyz1V7A1TAuLzTW70xm70wE7u7uMx2MGe0OEgOWlRbQ0jIopVd3g/LwZkncz+v0u2zfmjEd3ml7VoGTr8ReZXN0m1xmm36e3ssaRpMO5I2f45lNPUuQZj12/zI61jL7xLOe/+hTH3/d2VD/jntecpl/P4ZYjbEnCdkyoBDB5RevxVaFoFLaivHWdoZ9z6+YdRvUEOjXezimrAX4Irb0WxdyxM9plZblHJ56wnXh2g+OJwSXe/ZrXce0rtxlcu878zCnaaYYJChUE3jvqA41/7A+gI6LZ3nrx8jGicQWaVsyLVy5RPfK2P3X5Je0MEUXYsUd4R6+VsZLkXJ7OwDlq54hs8zoImhCSCpCwtz9klI5YUJAmCUiFVk3zUokDoZC3SNnECCua/AgdaeoQSApHYiJGoxE7u7vc1V/GWotJDMJ5EhXYvHGFajpkNhlSzsa44MDETEf73Hr+Rc61z5IQ6OcdarnNbDyliixpmmAKS+1qpG6yH6RSBCkO/u6uKQpYpJZEicGIgAESGaOdaRSGPmCkQimBkwFlD6AzNA5M5x2+EvgK6soSC81G1iVMpkgjqQ/OINoLVBBMtEA6hXQSKQ/+/yzIuNnF6MQiTYGKHVEMeRaRZzGRlKgQkNI3gBXfNFOFa9yp1jfCLQ40LN45jNQE32DkJJKVhT4tD242I48TpnhmylEQaJuksbzrlEg5lvsrtJKc4d4OHsnmzoDPvfAiL/gJWluSJMGXNbVzlFlxMKUS1MEwnFYMqynLYsxqL6fTFhjj8c7xzc/8EZ/7zd9FjyqiOKG1sEjWX+S+4+eY3h7w0vMvcehdb+LGegcuPslgc8of/Pbv8eZHHqDT73DirlO0tubMLt9h8pJlvOvYV698qb8qrNNB1OxuPs+tJ75JdPMGD7UWePfht/DokYc52T+OrBM2twbsjQckfUuyUjCUDlU7luvAcPsqO9kewwjm2w4bJP3lJe45cYyjnRwzn2Nqi0UwlQYtMiIbI52iaY7VKOXwrkDjuDMacHW+j+ovE0rNRr6AyiXzlod5zVqVsqEX6YcINavB1VRYwKM9pEVonkAo5CAQColf7DPJdJPW07TsCPJAgnywfX95QTZIZIkyGVJFpFriyzn7k11Er4VLIuI4JVMZdhS49Nx15pWgsp5QFLQj1RCIEhjOZ4gSDusenUKQrfW4MZkzG0/oOk2XhEgmiCRB6iaa1UpPwKNDg3LzIsLpCJmkRCaiTiOcVnR8QsfGqKCwxlBIRQgCG0MhLLFX+D2Y3IHZPKGyGlUJ1pRkWZVE1MxDgiWjPJj716LRSngZCKZpdCbSYJwn1hppBESKtN0mStPGgWoaJ2VpLcIIrDQEIoRvuImVkRRaUbgANsbXNb4o0IUjAkoRqAxgK+62GYtjjRSamZEUwmOA9CBTUyYGgqNeitGrbcSsIBlVaDJ2Jo6rt2ZMx5aWSDi9eoJ7jhxhuRdRVSNSVbOWpsQyZ75ZYbYl9Y5ga2AZIMl6XWbXb/DUv/k0fmuXXEpCUZEnOf2lNdLlVc5fe4H+kT5nXnuGmXQUKoI44Y+/+jTPPPUsWZLQWTlE3O4TnMFPHWFgqYfFK16Pr4qiYGVgmlhMP2JhI2dhKedof53TG8c5e/gYd584xWvO3cNrXnOO1UM5JAWFMggLmRfEwvPCnUscvv8MNzd3QJpm1GYdD544xcm8D0NLhgEMVoimwy0PSAOiQZQ1qkPB1JY8f+MaraUlrBW00xxlNF4L8J7UKhZaGT2TkAKR0k0QgmnkviY053EVBLoO1KWjTiIK1XQxfGiKgTjwXbxsUrLeHUwu7IH/XaKEJtYKKBjNhhTKo7IUHUXEJiVPOqz01zAYivG8USv6gKuaIjieV1x54Qr1cEZcB1ZWFlAtMMYQqea8bkxEOBhJailR+mD77QPCC5SKCELhg0AEQU2zlY9R6DKQqMbV52i4DEVpsUFRjjwnVk7xiR/8cRa7y8wGBWZuOdrpYOoC4RzdxXWWDx3Hy5iCQDC6AcVK2YhHoGEvetvYsYUnCIWKYpRuQl+CbPohSNH8Hj4cqCcB17hha+8IaETQhNJy/71nuffsXdi6bPwgSmCcZyFIcp2go4hSNcxH6SDVESZOCKqxbi9srFIrx61rl1HBU7qKzeEQd2DuEGgWOz1Wlnp0Ow1MtpMlLOQdsjghRtA1KYnOKZ1hUsP+3piv/f5n2XnpCnkUIXUzlcnTNsfWD2Ormt3JiGP3nOLLX/8yv/6b/5LSgVQJ2zv7fOkLX6KczWl1eiRZjjCKbqeNkU3Wxiu9XhVFoXSWp/e3uBTGXAwTvrl3haduPcPubAulSpYzw8nVRTaWe0gtqARNIGdpqT1k7Q5bd3Z503vexovb19FS04nbjG7uks8E3/G6t/LOu86gh1NMVVJQM9cBp5qzq5EK4ZoMQ9fsLnnp2mXibpt9W1J6RxTFREbj8FRUdLKYhTQhl5r4YHEjZfOGFgqkR0tN7Ty7+/tUvpEwB98wA9WBaMgYQyvPMKbRK1hrm+9xDolvMhFwSC24vX2HaT0n6+XEWQqxxiSKdpaQEJjvDfCumcNHxCQ6RUvD7c0tnjl/gUxr0qqi02p+zkcKEo2MdGOh9k3Qa6QN8oCzJmjQa0LIJmTGN1QlrQydds7a8jKhrDEHbk5tFTokaNciT/r8rU/9LT7+Qz/Mat4mL+G//JHv58PveTfT4QSjI9aOHOX0/SfNtZIAACAASURBVK8h6raZOovVAqVTpIoOZMQBqUBIC9IjdGj8FFqBbrBxyugGgecD3jtCXaJsBXXTPFQohNMkJmU6nBBpzcc+/BH6C33KqomDTzzkNNATkQiCaoq0ExCExiQZSauNFBGri6ucOHSE0fYuL127xL4bMxITbk+2KSjxFcznFi8rko4g6yvSNuSdjCjOEOWcWENnqUuUZ026+cSyf32Hy9+4CGXD15wFR6E1qytr9JOM0fWbSAR7dcXvfPozbA8nGC2RQiARfPlLX+eFF68glSJOEsqq4YWAJIlfOc35VVEUnIRBItlLAltRzQvzbR576Zucf+lJ5rMd2pEnCSXz0W6TzKugtg7nFTJojGo60jf3d2htdJmMJyz2FulEbfYvbZJuznjHobO88fApFpygqOdY4wiieYNLGsCpc42eQSWKy7euUuhAaRRlgDRJyVothJYUtiBTgYXI0HYBU9UQQjO6pJH2IjjI84M74wmjqkTGMfhGG9DAQ8QBmVljjEGrZiwpRcN11KIJvJVC0O62ecObXo+PJDNb40yzoKNWTBYZlA3MdgcI54i0IcKABUmEQDPZmyLmFXlp6SUtvIRxqHBpM3VJ4ojERERGIzkY03lQyqCUwcQxcZKgTYRQCmst7Szj5LGjDcS2tigHqUzITY/5qOI973gXb370ET79f/wGbniLf/CpH+FHP/qXmF2/fJAurailgjTFt2JKBVZJpEpQJm40FDiMPoDkqoCMJMqog01AaHBwrZQA1M7inUfYGlHVCO+QTuKrgPQSUQXc3PL93/FdLHW6PP/sc2gFMZAUgQUZk2iFVQ6rPFZB5QMqTlA6xQuDiTOOHzlOIiOGuzuMixETVTCUcwZhDi2FjhLKyjKajrByTpR72guaKI2pSoedjOjmMflCBycF9dzRV236vkU69qTB4ICJD+SH1ui1uww2txjcvEXUSvnyhfM8e+UKSSfCi8bqniQtrlzd4Qtf+CrVrMToJilsPJ1RVzXilU8kXx1FIQBWQqUFLlHYlmLHT7i8e5Xt/S208UQaJqPhQeWDqmiAJe2oQz21TEdTvvLE11g6ucZgOGJhaZV2p4+bWeZXd9h7/CL3tZd4aO0weXCYYNHBE5tGlVfbRqZr8QijGddzbk4GhHaLcVURRylZmqFSgxOOyMFK1KKPJKotSjisaLa3Pji8clA5ApK9yZzdYo6KU1RojgpCikYZffD0VbKZ42ulCQfxao21WmGiiMhEPPyG1xFlLeJ2TtTOEHmETg3S1YTBkHJ3gC2rg9EZTYIUkrJuJLbLrS7LwtCLGoitbxnIY2QaEScxcRKhoiZJynlwNDkXRVURpEQnMTKOsQjyXoe1tTXGg0FjlLKNZVk6zd7WDr12zg/8wPfy4otPcePqM/zcz/wwD51a41f//t/m6uPfoJ930dJA3KJWhlooZKyRRhO0odmjeITwSNHsFlTUEJiEVk0kvAjI2KCSGEfA1Q7hPbJ2/xbc6gU6aFQF050hH3nPe/mR7/xupps7TTqF0mjrSa2nc5D2VQqoRMCr5pgpjEGnCV5IOr0FFheWmU+mVPM5TgR23YQbxR516sn7bbJWn7qC23e22drbwutA3svweGaTMThHt99BxZq6soRJxYrIWRFtxKAilA7rIe53OX3/AxAEW5ubVFimGr70zNMMbIGTjtoWDbDWwmTq+MZj59m+cQtfWZz1WOtwQTKvXzmO7VVRFPjTWbdEIppAjVwxcHMubV3j1vAOhauZFyXVzGHnIIuAqQWxlbhxRTWe8/QLz/LUC0+TdDuQ5lRRjEwyTNBEo5KbXzvPUR3x4PIyybAgtYFIaoJUVKHptGsl0FowsQUvbd8mWVlgVBQooYi1QUcGlCLFsJp0OZR06ciosSZrh1UViAqLa+KTkYyqihkCFceN/x4a9PqBRqKua6SSxHGCUhIbmqTKygaE1LSSFlVZ86u/+r9z8fkXMGmCaWeYdosoM8xHA+48fxExKYiEbo4fIlC7mqA1cdZFiYhqXLIsW3TQRFJBqqkiAbHCBovFYVVAmAgvRJNs7Rx3BgOmdYWPDDKNEWnC0to6WafNeLxPrCTdLGd/b8ra8iG+88PfyYe+7ds4e+YU165c4q1vexRsxS//d/+Elx5/mswZjKOB4aiIWe0pa4uREUkUNz4MFUD4gwiKRiuhohihm7O2B7wUTZ/H/NusRCUkoFBCY1SEIEYHTZhXvPb0GT75Yz/O5Notxjc2SbVC+SbNuu0lPRVjvWfuPYVv3I5Ba1QcN70FrVhYWUYnCfujEVVZQCyYCs/tyQiRtBAmxbnArHBs707ZHc6xwpB3utSuYjge4EIgShNG+2OmexP8dklrv6ZTgrbNI72wlsOn7mJx7RC2tEzKOUVb8di1izy7dQubKIgFQTYaESEUSmuef+4yzzz2BPPhBGygKmsQgjj9jywMBhouv5ENPcd5TzhIBL4928ffuMIh55mUFpzCTh1UAm0Cs90RW9fnTEdQVFO++Cdfofhh6KytkB7dYGfzNuN5yWqnz/beFrsXLnL02BrxwgrXZiOm3lGqQClogkEQjU9CCy5t3uT1961TVDXeN9r0KEuoa4exmgWTcyhb4Mp8xiAqmyagbSTUFTU4gxMSpxRTZ4naLYwaYq1Fe0ekYsRBLyEoj5fNCK4WEoEkBIXWKZFNMTLh2ks30XEfFwRx3sKkMbUKFPsVV59+ht2r14mEwSvQEfja0V9d5vTGBuPrFxluD/BB05dxA2LSAqII6STT4ZzaKNIoIxxkFEqtafW6MJ4g0wRhImxVEpmcrN9le7RNmqWstFbYV4o3vuEefvI/+Sle++hbKGczoijhze94Hzc3X+LX/uDzbG8W5LOMRCQk2oCO8CgIGuEEERrlQcUCHSTlaIz0JT40CDutE1SUNoRpAio2mDQFLyjrsuk9IEDGlKVHu4ZPWRUW7TR/4xOfQE8n3Hj6OcJwAlWNlgJdW3KZ0I2TRkJPk13hReO/0EmEx7G0uMjhY4exTjAYDimKApXElEZxZ7fAJymhNpTlmNoFRiNLa6zpihypW5TVABsaZeo8eCb7JfXunIXScER10NM5CktpA2uHj3Dk6HGCC4ggKQycH93g8y9cYB4ZQuwoad6XWkmcBW0StreGXH3mBY7d9Ro0sskBjTST4j8yRmMAUKrpgHtLEAIXFCZtuIIv7exybXuKCAJbCt72+jdz912nWU96VFtTXrp4i6+dv8JXnrjEpZeu8s9++X/iv/r5T3Lkwfu4ceEpvA90WhkrrTbFaJvJtW1OnNwgdCXP1iNmzmK1whyYqErnkInh1s4u1YEHwFlPkiQkeUpVWNRY0Yo0i60OmYuIdEW7myCrGlEFbG3QBY3+z0iG8zmTeE5HmWbygDgAjhp88I1sonZYbXBWEZAs9BfIuwtU2yU78z06rQ6T/Tnj/Rk6TsnShLSuUApmwzHDre1mx2UEzVsGFteWOXryJHeKfW5fH7B3Y4f2+iKtLEYnBq1jVC0bNkFs8JHASwFKMbc1WhtMkhxs0Q+Qj0mMSmPkTLCytkrWNtz1yBv5ju/6AY6unsQSUSiLxJPnOesbCR//+Kd4cuUhvvgP/2fAYrJAFSAyaTObdY3GX3uPExXzWUHeyei3MsrxVmNikhFCxPhQY50D1QBpwrxmPp83JCzvCCrF49FIbB0YD6Z86qd/grtPHOcL/+I3OJotMr+zg92fYiKNVtDWKZlOQErqoBsytwJpDrIwkogjJw6T9zKGt8eU1lL7Gmk0+2XBYL9g7hs3ZF3N0VJhncfaDO8zyrqZjiwsJtShy6Ssmc9L6knJsc4qiypBlvvMywmtfIlTZ07TbneIVIyvC7YnY7546VmuzgrIEyrhcAS0ifDzRovhRJPn4cZzjJNEwqCQ7O8XmMUeTIavaD2+Oo4PQBAO6y3O+4ZdoizOB7TJGQwFF5+bsn1zwkLU4y+97b3cd/gUiYg4dPgoH/0rH+Of/dP/hh/8nvdzbD3mG4//EaPhDTYWl0nTPlMXsV8IOlmfJRWzUDnsldscKgUnfczyPNCzEIlAJSwtoUh1yi1mXJvtY4RuCEyxxESSNJIE7TDOs2RylkwXHE0zrp0QtwRpFqhjT22aKUcxswy3p0QhI1IpJjIEDaUomfoZxAqVpogoJyZqpFSLbZIjG8SL66TJIpFK2N/f5tadK9wZb2JbQFtjUkMrNmRJhPU1QYvGa+okRkXErRYijbFI/LxifmeX2DagFec8Ske0D8ZYXmica7iNQiqq2hK1MmSS4qKEUmlU0qL0ApKcc695HT/2iZ/lJ/7Tn+Xo6inGs5rd7SGD3X0G+zO2RlPuDOdcubnHe77nB3jnJz7O5ThmLgxGCGpTIowH5ykDDEWgtIZZPWb9eIezDxxlWs+oaXBpaE/AEYLF2ZLZZIwAirrGC4UVGjG1xEIxV7Czv88H3/ct/PB3fz+f+61/RT/PqbxgVFoKGahDRa4lS2lGHsd4bZAHAnihDSrWBB1o9zIWl3sEambTIWU1p7QepxM2RzP2rWNmA0VRMkXgI4OT4HxJNZ8Q6po0yui1V2ilC+Ak1aggqmG51wfrURhS1eWuo2fpLa4ikhatuMWeG3NpfoOR36W1oChtjfAe7RUiKLzQze7OVmgBptWipmZYDtiaDjl87gR/9+/+nVe8Fl8VOwVootxwDa5MBJCqRPqYcqqo98E1uz3OPHiC/Zub/OP/8bd58cU7xLGh3+1z792n+Zuf+iTvessb+Ye/+D/wxGN/yNmFu9hY2mDr5oRdPGvLq0ynu4iiQFvL8Po2xxZyFBHXqzmDRDA3kFYa4QN7wvPszWvclx9lPp2hRRsyg5ob6rJAFo6eTthIFohGm4wKi8k1UZaQe0dRB8rKIWpNWZXMfYWIFrCuJriApcJKTyUqpIwaBLtMWesvcrucMPAzkpbCLK6h7+wQhZJ6usOVGxfxbcXRe0+CaII+2u0WyyuLXIxh7moCBuEEoXJoraiEZ06FEpJiMMWuVKQ6IRGSRBukNthgiUxCRCN51lVFVRd4L5mWBaa9QEgtJBnX94bEnYx3vut9nLv/EVAtysKyMxwSu4gsiri2eYPHzz/OPWdPc/rue6iAY9/yZu4dbbH7r38fOxmTmjnCFRihcXHERAsMrYYbYCakvR5lqJAiQhiJ0A7pA1oLbDFj984W+cZx5mVFbcGLmMQBUjGcDDlx+hR/86//HE//0R9z+9kXecOH3s/Xn7zI5b09xnVJpBX9KGI5jkmFZkTAKAmKhuiVRqhI4XEMBjvUScH21nU2t25jlcAZw+64QfRZLIX1BGXwVLSiQCyLRqpdGoxrkZge++OCye4cP6pZT3MWs5x6b4KfWU701llbOozKe9RxArVgs9hhmlccPb1IXhouXB5QOQtS4YQCo/BVTaI97SzGJBlDO2fPzzh+/1k+9CM/wdn7H37Fa/FVUhQk3mu8c2jRpCl7LEYZtvYmjEcWHMzGkJiUZ555jq+f36QuAkks2NnZ5uLzt9nbKfnkJ/8Gv/ALf5NrT1/m/IVv0OlGyCNL7F5+kXa7TevQKuX2NmZWENcBPy5Y62cgDcV0HxdDGmlq58ikYmdrk6FZhgBGgcgN1iaoAqSraemIDdVjcS9isDOkFglRKyKNNFFb4KYFKpLNhMQX2LZHjAPeWYxXJDpCCnVgkJJoLYiSGGmnzPcnuLKi3W4TZymTUhHHGTev3qKzuk5kYlRbU5cC4yoWlpbJ0py98ajZtUgwWhJFEVVdESUJotCEIjAvA0JEtPIWisbbYL1Dxs2YV9iaLBbgK1RkGIcCKwO1MZx63etZWlsh77c5du4MnoCdzSgqy0qnT6oTPveZz/LZL3ye6WzCOx99MxERu7fv8Ozjz3HPfQ8zmpZc+OxnkUnW9EhU3OClVcO+aOUtpJbknfYBIi4gREDKQECSxjG2KplOxk0knJRY74mEoYoCs+mctk74hZ/8z1nXCb/0S7/E+9/5NmxdMK3nbA93SbxkUbY41OrSMQlR0BCaRp82piFQK0UUx1RlzfPPPo93nv3NAcPZEBFpChwlFpMnhESjrCW1DlmXtBNFL27RNlnz2hVU+wWDnSF2UqBmjpPHl+g6QT0tSGNNvtShtbiA6S1iOxmtIjAsxgznE0rjOH7yOLLT5evnn2ZcOqLYIwrIIomYORJr6CQ5stWhffg4733knRxZP8Hn/uWvv+LV+KooCiF4Kl9iTKMqtB40mq2bE66/YClHKYlLqO2QrVv7LK90GBOjk4iJp0n0UZYnzz/Pe9/zYb790Yd48Mgp3vLAo2wcPc1nL/8LRAyj3dv0V5bor6wSD8YsFiV1UTHdLukoy8mVNbbCnPPTEdOWIjGaK+MxF+oRD2Z91maWrNNikFaIxRatlsQUipOl4t0L9+DunOfF6T6jTkEiA71uRncxpg4WJOzMJ9xZmHLItVAItJNEIsLowLwu0C0JsSLNIPYaN5owGwzJF1dYOXWKUb1Pe1qxNdpmOtjHBU/Ua0OQlMUuK0cOc+bee3n8639CWRd/GmqnVEMtFpHmhz7+CT73e18iFoKst8SsmJC3DHE/pyocdeQpXITRHiUDC3lKpWOev3gdlff5rv/sp1m56wx3dnfpt3N2iwl1MeOZJy9w7vQ52h3Fpz/3ab785S/z0Q9/lDxv80u/+E947WseZuvqLW5fu8ny2gInjp9m/Z2KQaSIaoVspcja4v0ckRseuP8hZvPrtDqLRGnKZNIYm4p6jjEJrSgiTGckkWDkpvSPruGvXUVO4PZ4wofe8R7++d/5RS7+3h/y0+/8Vt748APcfWKDF29c4vmtl7g93GRDG167cIiTSZcuEdPa4kKNMpK4daDLiCOCVIQAk3nB4M4diuEec+nw3YRBKlH9JVbzlHnl2N8f4ao5cVAsdjVHjq7TWUoY7o25fWWL7YtTxsOCR86c5ns/9G6G5y+y8/wllvKMQ6eP0d1YQh5ah+4ivV7C7OpVXhhf58LgBtlyzuzWJc7ec4r3fPBjXN/a4WtfeYaXvnkbsQ+n+11ef/Iuti7d4cS9D/Ku17+NC597jN/8738F0sErXo+viqKAECgRDkYrjV+/GGh2bhaUY8AqtDbUtWJ/WrEgTDMb13Uz9/clQQSmZUneUzzx9fOUL22zOBGI7hrVnS2iYs50MiZZXgalUBLiKKElIlQhqIoJ070xG+2IWTfnQjVB1YLaBO7omjoxlLsjVjotQhQxTT2ubizXuchY9X16NiUTM2ZOYuuKqrDIWEAkSToJQmrII+Tw5YyCgK0so/mEpJViUkNRlETdnGyWkoyn1EWBDZ728iLJjS6yLunnBdP9AePpPr1+RtJv4ffHyJlhdW2JJDVMJ+Om1aZebhs1ir+yKHjk0TewefUKlStptSIwEOcJhSrRrQRFhBBNsvNgNuHW9pC1M/fx9o98P4dO3sWorAgCZuWM7c3bbKyuct8995ImObcuX+fzn/siF56+wOryBm99y9swRJx/8mkWu4usrh9mVM64PXHcfeIcKim48dxlRNaimkwJUoNWCB1ze3uPnb0ROmoRJ552p4uMInwAvCdPYo4dPcLzmze5duc2CEE1tZxaXufnf/JnKK/e4rf+8T9FTqcs9Np4PFVVM9zbhdqy2l+kayIioQhKYl0jOXfSE5RARE2DUaNxRUkxK6nLA5ye1syF4sZwyFZ8EGIlGkNXoQXSNdqbeVUgJpbx/j63bw5xM8G3PnAP737odUyv32C8uUWqDVm7g25lqHYH0+lgshSjYfPGVSaupBACXzpSHMPBLkfmayxHMRsmZreC9V6f0yurZDrmW77tI6yeXOcPP/37PPe1p4hbgfaRPnztlS3HV0VREF7gS4lXEMcCWbWYbkrGtwuk1UjVnL+dgN39kqMippVAXdRU5RwnNYnWBKGY1iWqUiSLqwzHM85feYLFdk6+vMidvYy6rsjjFlPGiEjRahmEOfD0z4cI5zgR53ht2RwWDFP4+s2XuDvtseFSpHUIA1FsCImn8gITp3SxHF5Y59LWkKqUzENCObHIWCJD01vwwlAFiPOUYjxvWIWqQaGhFU4HKmUpQ0mUaLSEajZlXtXk3RydpriBoZPmzOsZW9s36B5bpgqeYAIyhrVDyyytLnBrPCAIjYokcWyIo7hhUH7pKzzw0P0cPbFG1o2x8xmFcwynY/bGIxZbEbRb7O6PsRjmpkv37Gne9R3fTXvlKEVV0tKaVicjBEfodsjTFg6DkhF1WXPu3H0gI5599kWSKCdNWty4dpOqsBxaP0wURQyLilmi2beeuRDUviaoQB0rkjhBmoTxtOSp8xeZzTxp0iZN2sydowo1aZIgvOPpC09y09WM5jNqG4gQ/L2f/XnuXT3EL//4z3D7uedYWmhjWjG7+2Oq0jEdjWkJOLywSLeVAZLZwWhaCwOxRycaHUeNnNoLfIDSOkSQCOsaX4LQbO1N2ExA1JbEGOIAwTQejHpeU03m+KpkcmdMOyje+Ojreajd4cJnPktft+johDo2xN0utHLS7iJxp02rl0MxYmfzOiESDOcVuobVPOHq1du0dUQWFMXzN7i/v8Ravo4oLR/69o9y8vQ5Pvs7v8aFP/4j0iyhc3KRnXjvFa/HVzR9OACw3hFCXPgz9xaEEJ8RQrxw8Ll/cF8IIf6REOLFA3jrX9jhsBYoc6LQwfge0x3BzpWaagzSAb7GhTleBgbTEhl10EFTz+FNj7yJo+vHCCGicgYlM0Y64rk7Oxx/zX0kvQ7jcp/SFdx1912sHz2EQKKVobQV1kDczej0clbaPTIbyG4NuXuqOSMSDLATLFdnQ0oVqGTA6SbiTcQan0b4xCCVZn1xnZ5pwzwAKcXUMR7MmU5LytISPFSlJc1aJHnaqBaVpNNfAC2x0kMimdo5QgY67QwlFEVZMprN6C0uYXTCQq/L+sYSO3t3qEKJFZa00wBN252EjcNr6EQj4uZJ54InTRKM0GzeuIFjxvbeLa5cfYG8k2BiifU1O4Ndrt+6yc7+Nslil5MPP8I7PvbX+MBf/QmyleMM9mdooZjubnH+a19huHmT+WiALeY8c/48zz/9LMW85HWve4T3feBDHD5ygqtXb/Ge976fw0ePUXvPYDJkWpT4EDGZWabWc+zuc3TXVxnYGXMZSNM2lRPYYNifVJQVhGAIQR+oNAWtNKUYj9navM1wss9wPmUwKPjYd34vH3z7u/iNX/xveepzn2cpTTCxZmFjFZUmTKYl050Rh3o9lrodWlmKM5K5htpI4jQiasWYpBnR6gOzWFXbRgiEJFEJ3fYSthKUFhyaeWmZj+eUkwI7LpEzS9dkZE4TRnMWkHzra+9jwcL5T/8h3bpkrZNTljNa/R7Z6hJxp0e7u0yrndPu5dy+foV6OqHdX2A08RRzSStfJI7bXH3+MtvPXOVo1GWDBL895q9+z/dx5uxZvvj5z/P457/AkrWcOtzn4u6L3FSz/3eLAvC/AO/7c/d+DvhsCOE08NmDr6FhNp4++PhRGpDr/+1VF57bl+YMNyXFMGa07ZgPHJE4iGLDEw5QOrauyNKIltAs5F0+8UM/xsPnHoRRSaQTapOg220u3LjBv3rsayyevQsXtxiPp1y79BLldESex/SX+sxtzbioKXwgSjL6eY/FrEvfSfrDitMq50TI0BOLWuzQvvcUlwe7xEmGTDQVFhFJ0iwnSXOWe0ucOXKUDEHiI5w1VBNNuefQhUHTYjQtkVmHVqffgENtTZwYvGzyKeIoQUQRUd6ikyWkxmNtwf50TKvTJkoivK1ZXV7injN3EWyJEI4oi5jbkuFwQJIk6CShCXqH+WRMHEfEaUIxL/jt3/03vHTrOs9fv8KTL76AS1Nsq0XI23QOHeGeN72bt374e7nvHe9n9fRrUFkfJyNUnOAJ3Lp9i8WFBUJtuXj+WYZbu3zlC1/iiSeeYDgZ8/gTTzDd3+evfOxjxEZz9eol8DVLC52Ge6ACrirZHu8zG5ccu+sMH/7BH4BuhzuTCcQwHg0Jk4KlJCMGmvxYD0qRG0mvnVFWNZUPVDaws7XP29/8Bn7mp36Cx37jt/jMr/8a/XZKFGu6eU4/y5mNp+yORgRvuevIMdppC20M3iisaZKnIh2hdZN5IY1EqIAQvinS3YxTx4/w4Lm7OXv3aXarCTMsUmmMTNDEBCup5hW+dvQiyUqsubu3xF2tBSYv3eLK4+dpa8PG6jqT2ZwgNL12nzxt011YJF5exGQJtii4+PR54kyxsdhmNW+hZYKJexiRsSQXONVaZ6EyqHnNX/7ABzh39h6++vkv89hn/4DEe04cP8KwmPD4S1eIVv5CqPqfXq+oKIQQvgD8+f3HtwO/cvDnXwE+9Gfu/6+hub4K9P4ct/HfubyFnRdrLj854vJTQ7av1TjfWHlrayE0mYmi9mg7pRuX9FPNZDjij373C1S3hjzYW8bNxkxlDaEiWcz5nScf44qStI+fQouUem/M7uVLBFGzfGgVLwRF6XHOQIgxUUqWd+h3OmRIliaWN0wT7pkopoUnfvgc+zqiHFsKAyFqFqxSmsX+Gt2kzZmlBQ6LQDyfI02GoYUaKtRQUU08Nwf7TKMUnfca+7IEbZr0qKqwqDqmDBFBa6rZgEzVqFA1nAWt6C93sVhu377D9vWb1Pv7aBwYmNZzdvcGmCim0+4hykBkm6IwrgoKY7BJxqlH3s63/LWP86bv+j520y7pvQ9z8r0f5E3f98O8/Ud+klPv/G7E0t1UqkdhBda7xlZuJHNb0lldZ/3EGdrdVY5v3MVocx/tDY++6U2oVDMabdNNNXY+5vDaAkvdFtRThneuI4oxC5Gn04KtcozbD3zx97/AVy88zff8+E/zlm/9IKHtmU9HdJxkeOky1WiHLPOo1DOrxrTCnG4n5sbOFj7KkGR0Vcrf++Rfp7h1lX/9QApcjwAAIABJREFUS79MFw7+TUp6psUR3SaalQzGe6RJxNrCIrGKkdKg48bXEERDooqjplggAsI4qjBlMttjfX2Rhx+4h+NH1hnYMVfGO5BqIgGJBV8HShQGj6wskR1xMo04p3Py2zOmFzeJa8/iyiEqk7FXBZLWAiv5MkumQ7ffo1w0qF7C4M4me6NtFs8usZjMeLCXwbhk986YvIi4v3MGLk8ZX9vmzW99lLMPneP8M0/z3Ne+Tmdvk7uOHKZMEz7zxHOMgd29VyZcgv8w8dJqCOE2wMHnlYP7h4Drf+b7bhzc+/deAtBIyrllZ3PKbFJh/cu2LgVCYF/GsXsHSGSqGU8tl69c4p5z5ziycZh2Adn2FO1ASsHcVnzt4rN0z51iLxLYPGPuFFt3RtSlxEQ5Zd2kSLkDHFeUxLT7XXr9LrFWtCPDqaVFbjx1kcneiO7GGsOyIDcx3XaHSjh2ZiOSTossTdnoLHNicZW6qnC2bihIWjbGoqJiMpkx9RUhNVgtsSJg4ojYmEafEUJDjBKaeVkhtEYgKKdzbOVot3uoNOf27h6/8r/9OhcuPNOkP/saJR21n5LEivVen0gJ0iRhNC/Yns+Qi33+i//6H/Cpf/TLvPEDH+XEQ2/mQz/6U2w8+CjHHnoLZ970blR7jdFwyGwyQYjQpGeHQG1r6oOkqMWFRYxU3Nm8jVCC0XjE4SNH2Dh8iGtXr9Hudjh26jjPvXCRF69cYTie0u4tMS0cjojBcEKSJigpkUqidcT5p54DF/G+b/12Hn79W/Fxm7GOKfKc/smjtBYXcZVFVg6X5Yxqx3OXLrPUWyQJgR///u/hVKfPP/+Fv834xoAkpLgS8m6frN9ja3ub6WBEtTPg5PI6C0lOruOmyWg9RhniOCGODJExRJEmSg4w8LUlSVNOHDuOyduM2m1e2t7DFZaejMmtIgOCqLAUbJTwyNoS7zr7IPlMcfvZG+zf2Cd1isWoy0K7QzmZoj30+l2ibot0uUtvdYFer0ulA0/fusSertGHF0kWcg6vL9KxBevWcjZps/nCJa7d2OTeh1/Pax95hFCXPP3Yl5nv3Wbj8CrdXofHv3mBzdv7RFpz69ori4yD/28Ujf9XLOl/J3Lxz+Y+BAs4QXKgfcc1GC+pGmtxCAF8k2wsVQO7sKrR7Tz77LP/J3VvFmPZeh3mfXuezzzV2FXVXdXDndm37728I3kvKZISB3MQNFuKpBiCHEMv9luCQMhLHvKSAAkCIY5hWZAjybIlkZQpkbzinceeu6vn7pqrzjzus+e983Cagp1IMYMEBvkXUCicOgVsoM6/8K9/rfV9pEmKoar84ssv8rRVhLFHEkbEgsB7tzYZF23ipXk6koRoFfE8gb39NqZZRJI1wjhiGgRkZMiqgqIp2DmDnKFjhjHVRKAYZ1z+4CKF+Xm6YQB+jG5o2GUHXw7xUh/bMCgIOmfqq1TzOQQRVEVB09UZyEORCaOQfuYjVxyUvEkkpDM0mzQzRmVxjJiJjMcubhgy8aeoiow/ntA6OEKQFcxcgUp5DhWFcbuPJchoWTqzPWU+ohDTKFdxLIf+eMIkSdh49hl+63d/l0985avgVHFR0RvLmHMrJFqOYSLhZTq+aCAIEoauoagySZoycSdoqjq7ABRF7t+7y9HRIVvb2+wdHrK1s02uVGD7wRa7O7tMfQ8vChlNJ2zt7nLp2nVkzSITdaJs5vgMfI8kSbDtPLZVwLHK3Li+zc52j3x5ic9/45fJKnM0RY2kUiHSDOIoI/ETjLlFpqrBOMywzRyvPvciv/CFL/Fv/4f/keYHH1OUHbJQJV+ZY+n0KdSSw/nrlznY38dAplEooSDMHBsAikIsiSSSOAPlSLMBIkVTmQQehm3z3LOf5NjSMcIEWqnA0SQgEiR8P0KIYgwy9CSmQMKzlTrPVBYJ73fYv7pF78EE0ZcxBYOSmaNgOBAm5EybXLGAlDORiiairSFJ0Bt32B0echAOeePGRXRbZ96xeHX9JM/lyqR3tujc2+H0o4/x1PMvkCBw7dJ5Bvv3mS/p5IoW7X6XVm+IpAoIgk7s/33W0//7+v9SfWgKgjCXZdnhw/Sg9fD1PWDpP3jfInDwf/3jLMt+D/g9AMUQM9JZ5BBnPvkZVDXNSNKEJAFNkCEFVTZQFJWJH2HYMp12h8tXr3GqVqaaBnz5ibNkD25zeXcHwza4d7jPrUGHc6+9zPvNNtI4IEtERuOAWqOCoquM+m0U6SEGPgNV00ER0dOZfDUMI9ZMg3vnL3LuE59AzdlEXoBq61gVm244otk/whBUbEFnMVfj5MKIUWcPKW9CHCBOPBRNwUtTWsGENKejlmz87hRRlVFVGV8IEMjQBJmRF4Iq4sY+RUND7I7pdnssFRwMzSTwPdaqc3j7HdLOCMu20AURWRQJogjFsshklZ7n8blPfYUnP/cS9lyZUQCaAIppQCYQkRGnICkacZLhBxGmohJGEe50iiRJM4K0LDH1pjPfgqajaBqmbWMoGoPhCCvn0Op2OXH8OJ94+mky4KlzZ+kNhkzHPmGSIqsGrU6fRq1Mt9vDMCxAYH//iI0T69y8eQt3EiCpPueefYrf/Gf/NZ32Dqnb58JbbzKNUorFOo99+nPMFaucfXGbpUqdldV1vvn7/wcffOd1FlUbNwRnfp7yagVnzsHUJIYPtjE1FcPJoVoGkTjT2ftRyjhJaccBLzz/PHtXr1GybERVZuAOsQt5Hts4w4m5VaZHA0RBoXs4ZL8zwNUVUkNGiWOYTDmmqJxszLGSyTRv3WUymmDpoCCjmzZColOuVgiCgGgaUFuqYxVykNNJCwaxKpDP22iqzBc+91k2Hj/On/3Fv+Ql5yTrxRq2PeDo+j2mrRbHGoucfeZZavMLXL16kc2L51nIW9RrBVJFpDnoEoqQaSqypqNI/3lkMH8B/OrDn38V+PP/4PV/+LAK8Rww/GGa8f+0ZnIWESGbtdykWUqSJg/PHeLDY5yApjloqokoCw+lIDrnN2/Q8qe0Dw4Rx1OeWlrhkUoDO4gRgoSPrlyhsLaCvjBPoBmzerAiE8UxlVoZP/ZIpZQ4iYniBD9MiDMBSVawdZ2aqlFNJdSxx71r16lWq4ynU2IhQVQzzJxKTIAf+BRLFRw9x2PzSxR1HccwsE1rNhatKESZwL32PlMxRslboIgIioiqq6i6SkxK6HkIJKi2RiJGRLFLwdIJphM6wx6OZaKFMXXdxttvzz6smYZuFEk0m26SMVIVHv/0K3z1H/0mL37t69hz8wSCSKbIiFlEEkzxvAFZ4qPKCXIWMukekbhDXHc8M0EpKtJDB8WdO3dpNluoisra2trs5GDozC0s8OKnX+HY6ip3795lMpmQZSnvvvseC8sL/OzP/Sxnz56l2+vg5BwURcHzfKZTH9M0mU7HpGnCwcEhpmnTPOoxGAZ88zt/gxsrPPvaF3nhG7/C2Vc/z1TSGWUynaHPnQf7nHvpNQqVBf7dn/wZf/Fn30LXTOIYcvMNTr/wNOQNKitzLJ85jmCpTMWU1NLJLI1Yk/AViQEpR4HHoy8+zxf+4S+RPETZx0JKab7Ok8+d4/gjp0gl8eEAl05vv0VnMCS2NUZahktARVH59MIy55Q842aHqetj2BaKXUS0S0xTAadaplDN406nWKZJo1bDcCyUso1YNMnPl7l59TLf/Of/gvf/9Fs4fsYzjzxN1W6wdXObq+9doHXYQZANnjh7lupcncOdbfZu3aZkWdTnG/QCn83tPTqRjyvHpEaCqCdI4o9OWfmRTgqCIPxr4FNARRCEPeC/Bf574I8FQfgNYAf42Ydv/0vgp4G7wJSZhfo/sTKyh+2lAvJDJWMyO3IjzTBnmUycJBi6QZZJs/wyEdEMh7Hrc2Fri6Kt4boujqjzlXNPc+HWXd7a3OT2pSu0h0PWHn2U3UlMPhRImj6immIXDAxLe/gU4EczorLKzPWo6Bk5AdzpiIaq0Ny8wcmFBSRNxY8CNEOkXLRIfJFB0yWUZaZJhjKYkA1GRFJKImQYzAZsgjhjq91kEE7ZeDi5NpyMaXe7SKJGHCVIYcrUGyPrIkkQzS7u7HkUIWUw6jFfrWGoAoKkcnjgs7O1z8L6GfbciK7m0Dj3OKeefp7y/GmM+iqJmDFxJ8RijGZoSGGC57lE6YxJKcXQb7dQRImcY+GlIrKskKYJkijh+R5RFFGv10mShPfefYdet8vC3BzFeoXpyEWWVUzTQFU1ms0Wm9c3eeG5F2geNtncvIqiCOimzkRImU7HWIaN4zgM4w5z81Vu3bjDqY0zjCcuecch7Ld5/XvvgiSzsjLP2qPP4BQbDPo9coUa1z68QFvWqTl5zjzzArqR5zt/8MfUFhdZP/skg2DANBiRBFN27h6SxhEJKpJpkskyyCJ+FDGceiytHufLP/fzbG9vI6oiiqWxsLrA4sYa1XqDcOwzCcaolklzv8vt/QfEmY8iqMiuy7Ju8+mlVZbdlOnWNiig23kkRSeVZZI4QVIUFhbnkaUUP/BZbCzg5HKgyejlPGatgKDJ3Lt+ne/94Z9gizpH23sIVYO7pgJGiRNnqyzVj83awKt5Jt6EzY/fQ55MmFtZZJBkfPf8VcZeRHmxyEgOSHIxouH9nWb0v2/9SEEhy7Jf+Ht+9drf8d4M+Mc/+iPMNmMoZEgkCMLskYS/TYHSHxpcIc2QHyrDvXGKIhsg6mSGwtXDA7746ecY3D0gN47QRIWXV09gyio/uLXJm6//gJ//3JcYPTjAf7CNIM8YepKSUJ8vs333PppdQhAkskwmzgRiCRBiBCmjYhhE/oTu0QG9vV3qS/P4UYAtGghJQs42abcmNCdjrGoFc+cOtihzMBiRauKMEyiLaLZClIV0hz2kUgHdNhh2JiRChmkZ6KZE2OsTRzNVu5hluG6famkB29SYBi6DaY98UZuBRXSZC3duc+JTr2JunOaFF19jcWMdxbJJY4cggijNyEQJU5ER/Yg4TlE1E11VERSJyWjMcORybGGJJEzQTI12r4MAFItlZFlhY32DLMs4PDokTmJKpRJ2LofnB9y8c5taucZrn/kMqqpy+dJVkjBmPBxy9/Ytjg52+dmvfp2cbfPWm29y6cI1vFQkiiIgplItMBpW0U2FONUggbWlNVrdJs3tLlkMuZLDoy9/hlbniDAYcfr0Y0hBQuD6mA2LpxdXqTZW6d3fYTzosH/wgCDsceZzL9K659PzQ6aCgGLkUEQRMUqI3JScbvH805/Ebfb5t//779PQDKxijuOnN7AbZdwgwo9CUGXEROL2wRbn+/eJlQxrMmVBNvjM8kk2NJODO1cgDdD0MrIgkwgqSSaSxgGL9QrVnMPw6BAr71Cdb6CZBplloBdsZMcAReLsk08SnXsOf+Bz++Y2RzemLD7583zjv/odTKWCYdYJ3Qk7Nz/m/W/+McKkz7FiGVGS+esr17jSHJA380wGA+SqzPJGnmEyhOl/JkPU/68re8jcy2bRQHyYA6UZxHFKmmSIooSmqrNutkwgjjKCNCNRRTAV+kBUsHCnU/zBkKQ35HS1wdnV49y4cJ7Ad3n8mbNkWso0GONOh/iBz7GVFZJkNmiTZjP8WJhleElCIs4kKIogUNQN5KnP7u07hElCmMSkWYph6BiqTD7v0PcmKAWHvG6xWq0TDl3CqY8fBsQCKLrOyJ3SbLWQJRlN09Btk1y5gKTPymOSLiNpIIgpui4ThRNkMUZRYepPmIQekSQQSDJCLk/l+Aa59dOcee3zLJ79JJ7lMElgGEeM0/hhp6iGkILvuiDrIOsIqobnxyAqrK6dAGRkzSCMQsajMZZpocoyqiQzHA4YDQeMBgNWlld4+aVXOH78OFEYcuHiBZqt5kyAMhrjTT2iIELIJCrFMq3DI95842/otlsYmoqmanQ6fcIg5vDwiCgKMU2d3b0tDENlOByRJimhF7J3fw9dtjGtAg/2jtg5arN9b5eD3Ra1+WW0XJ79bp/SyiovfOlLHD/3DFEacbR7n3HzkN1r17hz4TJ+d4QlqWgP/RJxEGOqOk8//hRaKvD7//P/yuiohaHI+EmIH0dIqvbwfwxIMr3RkCvbdzgUfRIhphhnvFRf5lgsz4KR7zNRBJIEVHV29xUFIZqksFRrIIYRaeDTWJgjVyqiWxbFen3Wwu6YEAYEgwGmbtLr9vDDkN/4x/+EDz6+Sladxzh5iqGq0Edgt9en3e5QzOcoFgvc29mi57s45TJ+InA0nDC33uDMJ06QK2oPNYk/2vqxaHOe1QTlHwqTyMRsRtcRJLJEABTIZheNmiKhqRmqFOAKOpIGidDBEH3+4sIdPvXqE0ixz8iLWCPGmIw4l3foE9O+f43G8jGsZYudlkuKxfbWEU+cepSVhQ0m4yGCFDMQXE6sL5F0Rkz2O+RllSTLkBSBvKIx2TsgGXhoi0UCzZ6BVU2Rki3S8lP2JhPE0gJPWQL3tra5NwmxjjcQchbSuE80TWn32kQhGHaBjt9jkvioqYSeyEytBNcLKRoGeibTl328rIPV0Ol1XYSuTWzNcWzjFF/7+hMcP/cMSr6ImwmEiYCYGijJbJ4EKUISJdI4JU4SNMcgY1bxaO7vcrB/wPHVVYaTCFGQKBaKdA6PqBYL2JpO4E25de06nuexfvwE026f2kqFUX/CrZubCEKMLotsnDjO/dvbvPuDDzhs7TOeuLjjBE0poUo5HtzbQU41bL2Eo89xo32B/d0B/V5MvxciCBoHB00W5paQZYX9vUPiNGYyHHH0YIebFy9SqZZQVYV8oYJYlXiw18I0DVaOn+bC+5ewVBundoznPvNZRuMDDh5cJUwCjsYTdLuCJ0lE6ZRMKIDhMDe/RETGO3/zfUYHeywslInEBPyARBBBVNEEHTUTyUSJu702H9+7RSwLqFHCqVKBOgEHW7cYTgNipwiaTgUZJJHIdzGJOFGtUNV1xp5Lmiuil8vEpo6UzyGUy5AvkCo66XjK0d0tPrpwjREpX/2nv8lTX3iBN3/wBtZUBzOPccwhX8zR3FxANBbo51Q2W/tc37/PJ86e5kQ/5p0P7uLpUDlu4pQVtFsSXqQyU/j+p9ePR1Dgh9ZzkYdArYf3Cik8rEakD78kSUTTNERBQswgS9OHnH2Bfm/Ig509nlteonP1Ns3RkLrtoGg6RU2iu39I2S5SK1fYsU0SNyEMPZqtQ/IFh1ZrH8vUIYFSscbi6hk+GL/JUauDoymIcUbZsDns99m6cp1HKucgiYkVEDURxVGRRxnjQRvd96hXaqwurdLpNFFljXK5youvvIBim+xcuEV7MmKlbKMFGnIKUpoiEKLFKppmI+g2aDrDoz7+cIRklllaP8PJZ57mhZ/5PLW5eVTLAVkhTFJUTUMRRJLsIQYd/lY3DzOcXJqmiIKI57rcuL7JXL2Bqev02i6VWh13PEYURWzbJs1SOq0W/cGAjY0Ncvk8tVqNQrHA+QsXuPDxRyws1Fk/sUG+UKLV+oj79+9Ra1Q4cWKV6XSMLClIMkSRT5KGeMEUURKolssIpCiizGQ0oVYpUS6W8DyPQi7P3v4etmNiWgbNZovxZDjTwMsS8eE+n/3pzzMYdhkMejRqdWzH4Wi3SU/sM5dz+Nqv/yMuffB9do4O2OqP2Zg7TmYZeMkURxOpl0pE6ZQPz1+h024zP1dFUVWiOEaRREzbIU6Sh9j9CIGMO1v3aB11sLKMY40CBd2m1eshRBGabSFrGhESkqQSxylZCk+ffYaNxjyd/UN8L6BYL6PnyzPvZKOANeegOTKylLF7b5vv/JtvMxmP+PKv/RJPP32Wb//hH7JWrfDxn/85Z1UZc7GObCs88vxTdJtbtPfucu2jt5AFlbXqEv1sRL14iFb0sTWVQa9P4Ick/y+Sgh+boCAKs80vPHz25D867syUbKIISRKjKDKqOksvkihDUATIJGRZ58LFa5z6TI35x07TunIDRVGwJFAUieaDPaKOy6nTC6ytHefmpRuEUcruzn0W5+bJiAnjKcRw7/Yuxz9zipPPfpJ3v/99SGKcVMAWZBY1m/3bd9gu66zZT6DOVYjIEPMKBc/A3e/Qah4yt3iaE8dOcLXfZ2PtJL/zO79N1ZD4F//6X/HOx+epfFJnZe1xtEhFCTKkNCbJPDTJIIwkfD/FKBUpP1GkMr/Go48+y/z8cbRjC4yzgCkCiaSgKhph6DHYPyAKI6q1OrJuEcfxDCGvKgg/dCci4I5G3L1zhxMrqywvHSMJQ2q1Or7nMR6OWFheIo1jfN+j2+2yuLjI4sIivW6HhYUF+oMBR4eHvPKpV1hZXcE0TeIoYm9vj2KxyG/85q8znUy4d2uPna1dSqU87fYYUYIgmjL1XObqdRr1Kv5kgCLK6JqBoRrEfoRsijimiSzNhqOOjprYeZt8sYIki1y/do33336XQqVAGHlI4ox4feaR02RxxtHhPkdHHc489yrhqEtq1Tk46hH7MblGmUalTOZ53H+wjRdMmFsso2gSmZSRyTJziws4+RxBEBCEAbal0x93ee/8hwhJzGM5m7psEU09kjRF1hRUeUYFt1UNTTYo5Czm5+ZYnVsgGLgImUSxWKNUbyCXa+i1ImlJhoKAUhAZ7bf5N//8X8Eg5ed+4Zc4+1Ov8tbrr3Pjo48oj0U+fP0N7t+4yrNf/zKpoeOoKp/6+hdxO4f0/SG3PvwQ2YO6avLMk8d5kGYcPNgmMiEKJSbuTxijcXZO+GFAeKhySwSQ/uPoJggSvh8gSgKSPJOTkKaIWQZxiqzruL7ID65c4pc//0XEwybN/oCikOEIAras07q7TeR3ePTJMwwOu4xaXSLPYzzuo2kiaZxiqxb79/a42LjBqccfp7y+TvvWbZQopACUVQPXS+jd3aG8Vqe4VCPRZRAjrKJKbizRTjw0RSZv2nz+0z/FL/+T36JiKfzJ7/1P/OV3v8POnUMePb7G89KjqHkDdaoipCKRBNlYx0s08o0VnvrCT1M/+QimU0UWHfzOhKP9Dnq1iJMvkIbRbFb/cJ9Rf0C+kJuJaQBVU4Ef+hNny51OuHP9Bo35ORbmF0njGFU3GPV7XL54iUqlgmJpRHFCHMUEQcCx5WNMfY+333obURSpVOvYjsnxjQ3IMva393AnU6IopFDM0Tra5+rlK1y5eJu11eP82m/8GqNBG03VeeP77xHHXUIvJkkKhGFAFEUzCnMKnhfge1OSOEY3dQzdQFFVOt0+lVqNKElYWjjGytIqQ3dIo7GArqmkUUK9XGfUH+EFGe50SmcUEk/gE8++ytxRmxv3b1Io5gijCZ3mLhNvgGapiGqKYWsgiYiqyurGSTJRIPBDVFXBkg0ubd3nxq271As2p8wSnjslyWJERUFAIsoybEXj2OIS86U6Zj5HzrFREuhPu8QplGpVJMsk0g0i20Cr6Wh1G4SQb//BH9C7v8OvfPUXWX/hES5e+IB3X38db9DlVH4Bq1qiefUy37l/j2GazmCypko/mpIrWBTLVSRBwlIETh1rYBDxfvMa/V5CmhlM3J+0OwVmF4okPPw285pnWTQzLknCQ1z6jHOfZQmaLkAWP0SAZ4hkpFmG5ug86HZ44/Ilnl8/wd7lqySxR+SmiErGXLlCd9Dl/r0dHMchnrh4QcDUGyNKIu7EJefYaILA+Y8voFcqNI6fIJy4TO/cIRJAijNysk6YCox2m0RnPARTIdMFdFumWDLIV02Ggw55M8eLP/MlFupLfPdP/iWXPviQ1kGLJIPbe7u0phPmHJtyrcF0GhIICeOBxLnXvsijX/wC6vISQy9kuz8mdQfkJJNKoYYhGmxfucu1zWssH19h/ZGTSBsySTa7O4jjmDSaXdxqqkqWZQRhQOAHLCwtUqvXZ+W9XA7P97hw4QKqrLC0tEQgpuimhZBBPpcjZztsXr/O9c3rrB8/Qa1e4dSZk8iCwL17W7z79jtM3SmlQoGXXnmFctGm02pzdHhEvdKgUa+jGzL7O7tMA5+EFN2cdUxm2WwDR3GCYTmEgY+iKZiZhef5xHGKadj0B0Nu3bxDuVxGEQTcsYem6yRRgk9AFESMJhOu39jEUoqsLq7R2ruFLBhcvXiLWq3Bc08+R3t4yOaVq7jjLrpuIqsCkiqQiDGmaVOdn8cuF4mzBEVTUIKMcOpx/v33IQqoVfKok3CmphNFojgjJaPg2Jx97Enma/PEQUomCmiyQhL6pEKGUcqhlnKIRQuhmkOt5bDn8yi2xVt/9Me8/62/4udf/RqLSwtcvXaJD6+8QzIaUhRkHEVCjyQqtQrNoz72OKRxap237mzSwSdaKGFnGaIuk3ketqqwUlnizuSAXqeHIFtks7bhH2kv/vhUH4AfnhiAh2VIgZlPUUIUBAQBfH8KJKiaAiTIUor80AuQzrIIZNvm3UtXaPsec6c3GBIzyiJGkY8bBmi6zf5eE5BxnByKqhKGIVk6g5imUUjBVpgMOly5fBlJ0agtLSPncgyyBE8WUHQNRzQZbbcY7rYQ4mxmkNYljJxGbs7iwdYdcrbJxvo6F15/nW/+0R9RVHUatk2loHPYadOdTJE0B9upU184RaoVaacyi594Dqk0z927++w3h2hqjmK5SqFexZ+M2b26id8bsjw3z7HFJSajCcPBEEVVH5qcFADkhzang8MDoijCzjnMLS6SZglmziaTRL7/ve8ycV0ef/ppzFwO3TRxLIdOp0McxwwGA27fvs3i4hKf+5mfYXFpAUmWiaKYXm/Ajc1bTF2Pf/C1r3FsbYWD3QMCL4JM4NKlK/z1X32Pv/r33+G99z6k1e5iOQ4vvfoy5XoZM2cjyBK90RBJVQnThFQSMRybhBTP8xEFCcfMEUwCxt0xJAKGZtJr97h69RoP7m9x9dpVdvf3ECSRjz98j6uXLpIlCTmnQCFf4srFa2zd3mK+UKNRXUSWDHTDRFFVBEVAVEUyKWPt9AZHHIV2AAAgAElEQVSiqZFKIiIgZxmdnV1uX7jMYsFCzVLCzCdOZ1YwMZ2h6Y8vrVLKFxkNhrijIbKQkUQh48kIQZMpLTVQ6wW0hRJqWUMrmVilEtdff59v/d4f8flzr3B64wTn71zgzXe+RzDs4WRwauEYzaMmnueiSxLKxGdBsTjTWKIgqqwvLBN7EWmWIBszD0TkhexvHzHq+GSeijcSCPy/a/rg714/FkFhdskozSzNSH97v/DDx0uyWSOTqkoE4RRByLAddQb5SH3IIkRiMiEhISWRZVJN4bvvvkNl7Rj19eM0pyMmWcTReEAYZaiKxdT1kSUN07AQRZE0FZEkjSicookxeUNh684tDnZ3KZSrlJeXGQjQjkPGfjAjOLsZhze2kaOZH1I1FayiSa5u0R00GXSabL7/PneuXiUejckrCk8cP85yo46Tz3Fve5841VGkIuOJzNU7XfZDEEpVup0xWaqxXFumYhfRFYWd7fs0e0c4S2XWzp5h9bGTjKejGTzGcoimAa39I4bDAaqikGUZw9GQXC6HYzvIkjxjMSoqumly89ZNmt0On3z5RTRD5/z5j9jf3cX1XA4OZoGk1+uxvr7OV77yFQzLpN1u8eYPXuf+vXtEYUSvNyCXK5DGCVcuXuCtN99iOvYIpjGaYnKwf8R46BJFCdNpiGk7IAkMJmP8OCKVBAaTCaKqIagqo8Bl5LmEacLEnSCLMqqoQJBiyjq24fDg/hZpBhvrpzhz5hGef+FFLMeiWq/x3Evn2Dm8xxvv/YAb926xvLbKT33hC/T6A65tbrK0uMrpU4/PJiTVmYBHViUSUor1CrEoEiYxcRyjixKX3vsAt9OhrOhIScJEiEjl2WdWTmC5Os/68hqDTo/A96lXyziWSZpFRGmIaqtoZYssJ5MUNdSSSr7q4HWHfPCtN3i8dpIXzjzNzs4DPrj7EWE6xRJgrT5HQTUYeC5mvsA0COjHU8xjDXwJRE0jl88z6g1wLJMs84kTn253xK3NexzuDBj3YybdiGj6E5c+CLOagwACMz6+LED8sG8hSzIESUCUFCZjF8/3KRRsBOEIsgRZEhDT2UkBadb+rGo67cGIv/7BG7zyzFnGvR79wx6qZKKFEbIokikaYRQgyyq6ZjJ1fQQkhDQhDibUygWa201uXLlGyclRaNQY9mqYpEgjD88LCOKA7s275G4fY+HcKVJBRjFVlk4s8MZ0RBR6HO3tcOXjj3np2WcxoxFmkhBLCncP97l3bxv/6WcZ9id87+0LeGqRT/7MT5OfX0aQNAqZwHA4Yuq5CEJMtVZCtQyCNKE16ZOEEaW5KoZiEIcRk8EYGZFcLo8oiSRhgmVZaKr2t9OOqqLiBz6iJLF4bJm5+TnK5QrXr1+jNxqydvoksiRz+vTp2UCXolJcXePq1atkSUKY+Ny/f4+TG4/i+00kWebFl15iOBzy8Ycfce/ufbJIRhIlvvFz32A46NLpHbK7d0CcpLQ6bS5dvTQzg4kCkqKSCSKdfhdZkfCiAFGQaHbaWEvH6PUG2LqBLEq4gzGOk+OgeYCR16jOVSgUi8RJhG1a9Ds9lhYbLCz/In/zvW9z6cIVuv0xr730WT79uc/ywUdvcPv2A5aX5lhalDjs7KHqCpquESYpiq4iyDKJl2EbBuP9Lh+/9x5iFKOkKX4aEUgJtqIixwpCHLNQm8PRLVx1iu3Y1CpFwizlwW4LWZOozFdRHB2lVkCZK6PkNVRT5eoHFwm6Hp954gWEacqDBw8YJWMapklZNpCTjMCdki+WSWSZfrtL49Q6xUeOc3fvANEymHgeU2+CbagozPaM54X0Wi7TEDwlxk8jouAnLiiAJM+gmaKYzgSnmUCSxsiyTJRkIM2kHJMgpDcMKNplNOEmkBLGBUIJBHz0xAA/JkNAsMq8eeUeklnk6dOPcd99l2E8JiCloBRRwhQj1QGFTFVJx22MNEBQBLzIJYfDiqWze+cOWzmb9ccfQTm2gpXGBBcvMAqjmQGqFzO+cJfC6jFGVehqAblUpJ8OaLXuUC5W6V2/gbO+hJorIVzZ5cVjZ6Dn03nQ453379HsuNRPPM/Xv/INFtdXQIJm84iDgwMM02BhaQlFURBEmLhT/CDE0E2ckkOapIyn7sw4LUM+VyIMAgLfx7EdAKIoQlc1yDKarSYApVKJKI7J5fMcdo7wkoizzz1DuVplZ3uH5sEBq8fWKOVLNJstrl3Z5LVXX+P65fOsLa5g2AZ3Dh7w5Z//Gk+cfZydW1sMdocU7Rr1+Tke3N9BUkSGrstgNMXK2fzKb36aQW/CB++eR5dk5ktFNFlEVHUe7OwwN79MFEgUixa6NSCRIUwC3HFMY+00vc6IQT8in6/THza5fvEawWRMlsX0O11816OayxMnMWsrj7B4bIOPP/qY/+1Pf59HT5zkibMvMNndY3/rPnrJYXnpEaaZyzgckogJiQ6pDJZuYssmH15+g537DzAUHS3NEB+SlixFQ0BAsx0KjRLjZIKWl3FKJq4c4U2npGJMvtHAXKwS5w30ahHB1DGcPFkq09xu8uST5xANh/c/epf24X1W5IyqoSAGAkpm02v2mS+pTEY9is4cK2fOEaoxzTs3qVkqTc9DUxRMrYKTlkgQaAohcqShTjI8RSPWRALxR08KfizShwyBONYRBQeBHFlqIKQiUgpiFiPKIZISoeoxYRoyGO9RzlvIgEqMmMSIWYpGiBx5yFKCIMzAJXrO4r0LF9hrtlk6foJJEjOJIrw4YuL7+FFMnIGgKTN9tx+hKApJEhO4E/KGTt5S2L57m0GnhanreIFPqEpMSUgFETkTaN3bZuv6TZQ0wzQ0VMuk2MjTHrSRpIyVpQUufvwxVy9eZtQ5RAl9njx2gtX6MrGo87mf+wV++bd/m8W1NaLQ5+bmNa5du4Lt2BxfX0eWJSbTCaPxGFEQKeQKOJYzk6oiYOgGmmFgWCaiJBKFIZqmESUxnu+hKipRFLGzs4PrupTLFUBA0zXSLCOKEzY2NiiVy/gTl9ubmwz7A8IgQJQVjg6bNObmsXI5HuzssLyyQq/b48z6Bq+98jLXr93gqNmmsbjIr/6X/wVf+PIXyVdKvP/BB2xubrJ/sM/LL73Icy++gONYSKKM54cszC+RJCCJMkkCaQqW4TAajFheXCZLQZVVjg6PaLba1OoNBAGCIABBoNcfMnSnnH7sCUwnR3sw5O2338ed+IDMeDjl8See5LVXPs2w2+PCex+i6QYnTp1Gk3UMSadk5jEVA0mQkDR1NhkqK3ijEW++/jpimuHYBhJgyAqGqqEqKilQqdUwbWs2nyPNKNBCluFOJpg5h/LCHHo5T2G+hpIzUW0T3chx/o13cAyHUyfWuXHtMjc2LyMmIY1cDiWJiTwPEZF2q0OGiB9mlGrzVPNVvPYAWRFwKiaKklDQJHKGSCj6jBnixk2CeEwY+EThkNQfIUTej7wffyxOClmWkaYSUqSAICFkKVIakjGdVRoUCUWUQUgJY+j2uhRyNooyG5yShRiRDJUAMYVAVpAUkSgOAYU0Tfnum2/xa1/9Io0Tp2ne32fsB0hZiqKIWKoCooBVyNFpdwjDCE3RiAIfzdTIWxpH3QG3Ll/gsWfPEqkizFWR2zLhcIqBiNsd8+D6DZwzNayVMokCdtXBbcdYRZOFhQaXLnxEPFVYr+YJ2geUrDonXvwsp7/xNczlRZJMoL9zn+2DfVJJ4oknn6JcLOEGHoHvk5KhKAq6qqNpOkEwm8wURZHRaEShUMB2HEI/wDJMFE1lOp1iGiauOwFmidri4iKSKBInCZZp4fkexWIRQRAQBZG9rR1MVWdlZQXTMBlPRrS6bZ546kkO200E02Bu5RgKEo+efpRJb8w3//TPqNUb/NQXPk9leY6/+vZfc9A+ouAUaLVazDWqnHnkEfZu36W5e4AsKgzHQ+RlG0EaMxhOsZ0Cvu+jSgHtoxZiVqRaKeK6LguNOTzXgyTCsg3a/SZ6TkXQ4MHuHn6aEAUh648/zr044972Llo+h2nlSeOIWinPxud/mp1bd7h4/RqqrnDq+ApJMGE6cWk4RXaGHrph4scZsizx4PZtrl+4SN600ASRNMmQlJkZWxRFZEUhXy6iWSZOziGKglkD1MjD9TwWT6yiFh2kgo2QN0lViXytytb5TQ5u3uP5M09x/rtvcvHdt6kYEmXTwhRlDg52KebnaB7uISsKzZ6PnW+Qry+AHzPZP6I4X6J6qs5kP6LhShjKFNeQiOdT5moFzs2fYs3PcAUJNV8EXeO/+V+++yPtxx+LoCCSojIg8ROSDKREg8QglSCSRCzJQpfySFJCGB9ysB2wevYYdkGmP8qIsxT5IYxFVBUEAWICBEkizUJkQ2IwnvKdN9/hcy+/jDWKmBy2URAQMw8/8nAMDVkUsUtF0nCETDZTkSUBhiKQN2S6B7tsXdM4dnKVjphg1soodoGoPYAwYHDQZn/zHserDqLjUJirMJkcsX20RaVemlmrUwHiMXKgoiQOcW+CpmiMooC7t26StbosrJ6gurCEKEmMJiMmrku+WECRVRRdJQ5jev3eTEybZQyHQwBq1RrT6ZTJaEQpXyCNYlRZYTQaEoUh5XKF1ZUVDNtiNBwS+AESAqIoMej1ZhduokSv1Wa+3qDT7nCq1uDw8JBStUKxUiYVJb74tX+AgkRra49ioUb/qM323bs4hTyRkrJzcER3NGTiTcnlipw8fYpXP/Uyg3aXv/zWt1EVE8fMMVantLsjnGKF3s42tpPHc6ekYUSpUEJEQJNVAsEnlTICIaTfbVGr5pFkgSRNyVt53GDCjZu36XU6vPj88zz17DNcuXqF9y6cZ35xgeXFRdrDIXdv3ODE4hIvfPY1rly5xO0791iplLBlncDzKdsFDvb2sKwCOdXkvde/TzSeYDhFhCRC12aiW1FR8YOIYqXK6sYJ1k6uI0syo2GPKAgIo5BSo0pxeQ6tXiLLmwQiOKUiRwcHbL79Ec+deowrb77FxTdep6hJlHSVnCITDicoqUjBsmnu7lKozJHpFvOrj2BXG0w7PQatI8ynq+RPVKgqQ9KkxOrJOuqyg6+VkLOUFSPHcJoycCNK1QaaYf1kBYXFhTL/3e/+FIPBCNeLaTUntFohh+0D9ltdxpMO04mPqZsIUUK32QdNQXVs/JEPikKUJSRIaIICUgxJhiBms2NpEKPZBje3dylu3uLpkxu4Yxd3GiEKMbEAWZiBKJNzcpiCQRJPCLs9vKmLk8+TJhFh5HL44C6mraKUTDrNHkYko4kCUQziNObo7i7VtUVyjxRZXFtgNHJpT1qceupR8qUiiecyGg/IWyZ5TaC7s8U7/+7PWH71BZQ0ZuPJRxG1HNPplIOjQyRZplKtoKk6iqYQBiGdTpcwjCiXy2iqhmHMhm9kWSYKYvK5PIqu024ecdRsYls2i0vLBKGPLMlEQcSdW7cZjcY899xzaJKMruvkbIej5hGGPiMsFfJ5FFGiUCxSrtYYjcdYlolhGrzxvdf5+M13+K1fr3D18iWqtSpPnX2STJzp4F1vih8ETFyXU+sblMol3n33Xe4/uM+ZjcdxRyNyTo7hxEO38miGxWTqEgQ+pqxTLpeZuEPG4ymiIBEELqPRgOlkTLlqUSwVOWgegJBgGCrPnz3H2uoylVIR1VJZONagO3bZ3trhyo0bHF9apnpsmZvbDxA1lTOPP4EwGnN48yZRFiA5IrZlIscZUhCwdes+F956B0dWZtOlIiiKQpDMNHxBnNJYWGB1Y51ivUqv0yXKEoI4JJJFGmvLmPM1AkdBMGQ0y0KIMz7699+nLsocXL/OpTd+gJkF5BUdXUxRRImjTp96vQ5JSq5oIWkic8cfxSxU0fMOH3/0DpGREslTKlbM0uMr1BbKyEWL+0GbrU4Lz41IUpHBOKTdGZNEMBn8hKUPmipgqj1C06NYdlg/uYKgqQjyGSZTl1Zzyv7OmL0HB2zdHXGwfZtOe4VGw2Zrb4wkgahoZBH4qYCcJsiKhCQIBGGAqmkEboBesHnv0gUajTrzpzbYv3aDNAwwRZFEzJAFEdswcd2QWqmG64V40ymiAJomkzNMBmMXOYmYrxQZd/tEZNSPzdHajshIcFsTDjf30ObnmFuqs7O/j9cPEGyJxuoSd65cwxBk2u4UpxBiiBHbb77B8kKDJz7zCkEcMJ6M6fZHAMzPz+OHAQf7e1g5hyzLME2TatUhjiKm3hTLsoFZWqUoCnapxLDXZ/P6JjnHYWF9gclkzHtvv81nPvtZDvf3CHyf06dPIssyWZpimSZhFKKpCqvr63hTl0KpTEpGznbY3d/nm9/6Nl/68pcIJjrn3/twFqB6HY56Hb741S/z2Nmn8NOYOEwxVJ0ojJhMxvzgzTcwdZVUADufp1gu0NofkpIx9UP6kzH5cpHp0YRavYLXnzKajCGDyTQg51hkZBRKORRF5uBgn2KphKmo5FWdr3zlS0wHHW6+/n3CyRhrrsDpxx9ntVpnsVqnUixx/qOPmBgqx09t0O/2eeetd/nEyZOceuwxrn78PuIkwTQlVC9GygKuv/8RyWhCo1BARyLOYjIFQETMIJfPs7C8RKFWJhQypqFPCow9D62Sw1qeg4pDqoJuGtimzYP3L6AdjcgbDn/1539OXkopFkz0NEVXFUajCZmgU6nMsfN/UvemMZJl6Xnec/cl9i33zMrKpTIra++u3ruru6enp2fYnI2cIWmKEk2aBMegDVmyDVmULdiQf8iSYZiyDUsiIAwp0RzNwuHM9Gy979W171VZVVmVWblnZOwR98bdj39k0yAsyTMUaHj8AQFEnLjnBuLH+e5Zvvd51x+g6Ao9v4edKzE+NcXa5j3uNR+QH09z/cENlIMGJ554mCoB59aW2BUOjdAlSjT8IKHe6NJs9wndhOZ296cejz8TSaHXc/jOD15nczem44CkSqiKgmEbFEsWhw7OcuLIAi8+9RSJ32Xx5hUGMgmzwynO+zGeX0dLWZi6BrKCEst7numSQJVlJGJM00BBAkPjjY8+4Nd+4RcojI9QW7lPEHsoRgY3EXiyRBJD10sYHp+k1+vS9z1M3UBT+6iSRMGwiWpV8rZG043YDnpkx0eJPZdmbZfd+1UK6ztU5guk8tYeyttSOPDQEa5cvUY6NmnJKhtek3FZYipWuPu1bzJQqcChaWrtPfns4MgwSBL1Rp0wDEmShFQqjW2n6HX3nqqWvadxqO1WMS0L3TBp1utcuXCJSqXMwYVDSJJEbWWH4ZERnF6PwA9YOHSIXDaH4zpsrK3h+T6DA4MUSiWcXpft2i5WJo2hWyRJzJ2biwRun2wqzc79deqbVZ574Xn0bIaTzz7J/MJB3nr9TRCCU898At9xsXSD2ZlZlpaW0CyTUAQcOXmCpx95Cr8Z8O7p82i5YZqdNnakUakU0FSVJJAI4xgQ9H0PwzIpDpTp9boUCzna9SZbG+vkUjYHJya59dZ7XHjrRwykdZ568hG+/Z1vkXE66PkhNtsu+cExjh9YYH3tAbdv3mFqfJKHjz3MlXNneGBrFAoZQr+H23IIah3SgyVunb9AybIo6inkKMINYzxiIhIMYGR0lOGREWIhCPw+QpHxwgDDMinNTWKPD+BYMpEskEwTZ7dB49oSY5HJ9bffxUoCCqZGRgVD1ZAlmXq7S740TJgoeypLfIrD4+QKRRJJcOH8+2RmS3gDEN6LCeWQs/ducm2rSjNM0HQTX4CkaXi+SyeIcZKEMJEJNIO9QfGT4ycmBUmS/gXw80BVCHH447Z/DHyWPS3mPeA3hBAtSZImgVvA7Y+7fySE+MpP+g1F07m32Uc18wg7S7+nQQ8SucfS9ToX3jtLSjlHOZvj0OwkRw8PM54J+eVPP8rMxALvX3jA+es38QIXQzFRFZ0YiESMLBKESNiTxOtYqRStvsP333yLL556hsBt0drZxol9dEmmF4bkrDSNdhvTLFEeGKK2tU4xXyAMIvr9AKfZRO66dACRLbDrdHGEx+TAELIs0d1us7G0wuiJo5QGilRr28SGwuSxw6QrRfxqH08Dwi7lvmDIs9B7Due+/QqTI3+DofEJUqqJ4zo0m02cvsvQ8DC5fB5N12i32iSJoJAv0Pc9dut79H3Ltum7Dos3bjIzNcXI2Dih5yFJEtP7p4mTGKfXY3hkBF3TIU7Y3tik1WxhGHvHlZqqcfbCeVJ2ima7xdCgTb/nsLa6xlOPP0HKslhdXEL0Q2ZmDxApEiPT+2j1Oty9c5fO7i6HZg9h6Dq/85XfJpvN861vfptsIUdluEQ2nyKMQiQpwbZNAhKcvsfJxx5nZmqY733nFVQlDYqEDOimQa3ZIJ9PESUhXuiRyWbweg45O83a4h22F69ycHgQzW9z9tvfpHrjMtLBBR7cesBOX5A+maLW7TI9MU6nm2ZzeZWRoSGOHj/Gnds3WKtWGR8qQuAiOi7n7txg494yI2YaKYxQVQXbSOFHLrphooU66VSKKIlotltY6RRBGBJHMZMz06Tnp0kyJj4+0sf8iBsX3sfbqtG4tUpna5NCRkcVAXEUoepZdut1QiGh2TnqrQ5+HBHKIcOjgwyPT/DRR+/Sae4gjxbw8iYnnzxJIvmcv3OTDdNGpIpoHmihihx59HsOSRSThDF+GOKHP11CgJ/uSPKr/JtGMK8Bh4UQR4E7wN/9C9/dE0Ic//j1ExMCgKVnkFwTvw2SL6MoGroto8gaqUwJ0yoTyXlW6j7f+fAK//CrP+Jv/sM3+KPvXiWTtvnKrz7L//L3f53RvIrv9fB8hQiDxNAIbIFvBsSGRxh3UURANmXxYGOVM4vXGDk4R2IZdH2HTtCjJbq0CfBQWV3fppwrkTEsIqdPzk5h6jpu30WTDXbXa7R2mtipDOutOku7W1jlIqZl4CyuYDoWY6VhVD2gLxrkR4qMzhyir0uEUUTcidjabbMmfPplhe2lS7Rf+yH5WNDY3Ka6u4usm0xMTFEuD4Gk0gsCAk3GKmSRJYmw45CKJbKSSne3zvVr1xibnGBkfHwPtKqodNttrl25woXz50nlcshIyAIcx+XG5atk0xkcxyGVzbBV3abZaDI9PY2sqghgeXWV8elJFh46RixDlLJ5/NOfZGRmAjOtYdk6QkTsn57k1AsvIMkabten3/N47613KeSz7N83QT6XJ5vJEQuZtuth59OoRszwQJ7xUgXTUynLORI/RJMUNE1FVWRCr09zu0retFG9GKWZUNSySJJC37TRJ+dY6SWsLu+w9NF5pDCm7/joUoJOn9FKCl2NWbx9i34EmUyRKIyIwpAjRxdIZ1Osbm6hmBkWb6/S2mmQrwwSqSqSrmIZOpaqkbFMhCETl7IYuTyWmiKnZpACaIUO6nQR68QYYUGjZwmEbZK3CrSur9G+eBfRdlitrZNO+eTVkKykICcmXUy2QxienCKVwO5OA7enUSkeJ8xM4m6v01m8hDqosW1UGT6Uwx1N80GtzmakImEiun2E00OWBEGsEycGSagjRSYGNspfovrgJ175bzOCEUK8KvYUFgAfsUds/vcOVdU5MncMp95GFxHCa+PFNULZA0UgFEg0UFIaatokVAwe9DW+9dYV/sE/+WP+0e//AadPX+C/+lu/xd/48ieYHPUQ3jZ4beiDQQFFzoOZwlNCvChAtQzOXbvG8k6V/fMHiSQVL4pwY59eEpJYBj0R8mB7h+LAEE7o48kxIq0R6jKj+6cRQqO5WyeXNhgZG2Sjvs3tjQcYpQKJBE7bY27/AQrZFHeXbqAqEiP7Z7BLGVRVRgQhrudT81w6iUfWVrn1/Ve4+t1XUBJBIZ8HVUY2DHzPx3P7ICTSdhpNVmg06qyurNBpt2m3O6w9WOXQwQX27ZtEkiUsy6JWr/Huu+9x7tw5ypXKXpVjFOL5HqvLy1TKZYYHBhkZHsG0bOqNGseOHGVkeJjRwSF63S7ZTIZnnnl6bxmxeIfByXE+8fJL1GtVrpz5iKDVot9scerpp3nsqafQLRvLsnF6LgJBtbrD0tISXr+P67js7FTpuR5xEjM3N42uSbz/9rt8+M6H5FJZJCHodTvEYYjvuYRBQC6bJez7xH5It91DVRQ8P6Da7hDpKdKDo7TCBFmzOPzQw8wfPYRQBF2nzc3b15idm6YyOMja+jqdroOqGgxUBjF1k7nZeVKpNA/WNzGsFHNzC3zuS79AoqnEsoQs7+lvTNNGN1I88vSzPPbUU+TyWeLIZ2tnA8nSGDowhVrJ48tiD5CLhBYKFt85jbO+w/qd22gqZFImmiJQE7D1DJ2uh2GnGRsZw201CR2f8YkDVCYOUB6b5p2P3kQasojHs+w7PoedTnFrbYP1rksoGyS+QIsTFBETJwFeEOMHCYEv8NyYXtcj8MKfejz+VRQv/Sbww7/web8kSZckSXpHkqRn/l2d/qLvw/Z2jc9/8XPYlkG/1YUgIJYFQhIIeU/kJGT2HHsk0AwVo2CilS1E3uRmtccf/dl53n/7LI8fOcr/8Pd/mf/2P/8kj83kKcU+Sb1D0k+QlRSBlgHTBkXDS+CD8xfJDgwzMLGfSDPoRgmxqdEmILY1trw2XVVgDQ/QiH3UcoamiOlGCWgKEOG7uwwNpcmW02z1Oqx0OvTzOneX70OYoEcytc0qsmowPjtNtpDDi9w9mJQscPtdgn6flKZjABdef4O0LCP8PkIWOJHPbruJYRrYmooZCsJGlysXLtHxHJRsivXGLhMze6BRKYxJhGBtbZUP3/+AdrvNM6dOMTM3j+M4vPnWW5w9fZpcJsPCoUPcvHmToeFh2u0WjuMyOz1Lp95gc32DrfV1SoUCIk64v3iXd954C83WsU2VxUuX0VyPy2+8Q3tjk0Z1l2qzxmtvvk69tcv4vlEOLsxz7PgxCqUC2VwGy7KQAE2RyBcKjI4N8ejjJzFsjX7oYKQM+p5DImIs28T1HKJKuRIAACAASURBVCzbYHzfOKqu0XEcPEOwvLOOZmjEfkgYBCSKglzI8nNf+S0ePvU4q1urrFc38UXMR+fPc/rMGaanpzhx9BiaonJvaYVr124hSyqZTI6DCwsYhobn96m3Wozum+ATn34RI2ODbqJZKSJhMD4+y8zkATzfw406dOIGSl5mdHaS4tgkIRah0FATlaKZ4/75Czy4eoWwVSNs1SgqYMgaSSztGQArKlKQsK88SrvaYHtjh5HKOKPTB2Agz+j+CWpWzJvdVd5YuUWkpbhyeZVrV5aJPJkklImjhDDcQwNKsoIQMVGckMQxXuDTd1zC+K92+fDvDEmS/h57uxd//HHTFjAhhDgB/G3g/5AkKftv6yuE+OdCiJNCiJOe32dgsMJnPvUCsesgJxFJEiGIEWoCHyeFRNqjM2myhBzGkMh4wiAycpBP8/23zvKP/8k/44//5dtoicXv/vbn+W/+y8/wyafy2KKBW6si9U0MNYVqaqSzOWrdLm+fu8DM8YcglUGYNs2+S6DExJqGkcmwur1NKEuoaZtEkRGaQisIGd9fYXjYREQNkrhDYSCDVSlwv1Fnw++xs7vL5bPn6Tfa5O00jtMjVcxTrJTQbYtIFgRSTD/06Dldes0W5XQWvdXm9uuvU8pmgJiu1yNbyO9pSP0Qr97g+tkLiDBicGyUer/LzJGFPfJwr0cYhtxduku30+WJp57ic5/7HFPT0/iuw+3bt+l2OgwODTGyb5Lr12+wvraGosjcW1qiWCiiqRrbG1v84HuvsLm2QTaTIwpCcukshWyOkbFBao0qM5MTHJue5uIbr6N5PpahU92tIekyuqXx3ofvksmleeHF55k9OENloIIqKUiyRCqbRtcVTp48ysTEMIMjFR598hEUS0I3VY4/dISHHz3B3PwsI6MjrK6uohs6rt+HtIYvx2xWt8lm03S7Xeq9LjMPHePA809zd3WJ7//ou2zubLJvdj/7Z2d49bXXeO2118hnM4yPT2CZNv2+z5kzF9jZ2aVQKHL4yGHWtzbw4oT16jbCkNCyFrKtIdsp9FSO2fmjex4OuoKa1fANj+L+MgMz42CmQM+hmRlMxYamw/tf+zP6mxtErSolUyUlAhI/II7AsDNEkUQ+lSetWKwvr5HJ5Zk6cIgwnaY4N43T7XB3Z5U3bt9kdHaKXivi9Jk7tDuCKFQIvJg4SkiSmD0RN4hE3it5F+wZDIk/hxj9dPHvffogSdKvs7cB+cLHBGeEED7gf/z+giRJ94ADwPn/p3t5fsiffvc7fPbTz3PtzHXubWxg51J79t9BRCT2BFEISGJQ4oScJONLCl0PwkRB6CZmMWLH7/PaR3Ve//AHHJxL86kX5vmd3/4Ev1gXfP1b57l8eZN8fgw9PUySyCBk2v0+ka5z9MknWbl5k/72NlYcIj5Ga4FCs9pgdGKItZ1VdFkiVhKstInWT5CkEFWJsG2dkm4RtCTqnQbEguXb99AVlbF9+/GDkJ1ah0QIrEyK3q6LIkOSRGh+H8/zSBs2A5LO/dMfMPX4CdJjw5ilCrqsEXW6rCzfY+POCr2Ow2OffA4ll6Kia3t+Cr6LEidcv3yZptPl0ZOPkM8WSKKQXreLbpnYtsXCwiGmpqdxnB6Lt27y1KlTrK6u4gcBo2PjeH2HO7dvU9vd5djxh+g5Duura6w+WCWVSqOqCo7T4cDcFBe/9wr1tRX8bpvh8TGuf3CGTD7DwoFZwiBgemEaEUY47Q63b92mulvlyMHDzByY5vrNG9xZvI0sSTQ7NR579ATVxg7prMXG1hqtbp1nT51ClSX+9F9/nX6/T7FcZK1dQ7U07i4v0XEcMtkcQhFkBwucPv02t65fJPZdFDuFZug8/fgTSGaa1197F0kIThw9yczsNKsP7uN7gpXlNbrtJvsnhynki7Q7LpqlkC1myAzn8WsdwlgwODbF2OwMiiyjGjotv4WWT5GbHEIp5vCFQoACQkGXDRbfe5PGzVtUggg9jkibCkoS0o9AVnQkw0AkgsH8IJtbu3Q6Ho88/iiimCc3tZ/UYIUbP36T7dUVTp5YIBfrXP/oFq0eKGYO35ORLT72P1GAmCQRxElCHAl8zyOMBJKkfMwr+n+RpyBJ0qeBvwN8Tgjh/oX2iiTtpSRJkqbYc56+/5PuZ1gKDzbv0HUa/MZvfIly2iRq9ZADH1kkKArExKCArOwZz1ohpCKZuaEx9pUHUMIYDwhNFT9rIooDXF/z+d//8Az/4//8fbrNLr/3n36Bf/C3T/HEQo6iEVHJKIwMFsgVsiyvr3Ho4YcZnZ5BS9k4oUsQh4R+iKVbhD0PyQ0YsLKYcYLnNIgjF0GC43oosoRGQr9RJS0iiopF0HEJnD5+32dnawfbtInDCFlWyeSKKKaFJ0v4skIgyTi+T98PSSURnaU7VBcXKadyKI6Ps9vk6oWL/PAHP2CrtsPjz56iVC6RsWx0WUGKE0gEt27epNPr8ehjj2HbKZI4otfrYdk2iqJQ3d3FME3MTIYrFy6Qy+dJpVK88frrDFQqqLLE8r37zM3P8/kvfIHpmRnazSa6rtPtdRkcHkKWBFISQthnY/Uujteh02lCHKGIPbu/kclhDhyaZW1thXPnz9LvO2x8XB8xun8cSZHY3NognbJQNZmO0+TyrUs89cknOXRkntt3b7G5tc6DtRV2dndIZ9N0nA5JEmNoGi++9CJ/77/7r6mMlVmvPuCJ5x7j+OPHWF++RdBpUErbVPI5Hty/z/1793ju+ef5uc++zOLiDa5cPk8UeczM7sc0bUwjRbfrcfPGHeZmFijlBuh0uvgiYHh6iOL+CqGdMP/YUSpTI2An9OWAJGUwODdLamSYUJEJpb1Zra4bhLtN3vqXX0Oq1igoMoYhEUshQhHIqBh2mk4YItspZN2i0e6yb+Eg2tAA3YxBZXofq7dvs3TtKiPFAQqRycbVB2yvN2m7Ma6fEMcfIwckBSEgSSRikewtJ6KIvucTBn8Oa/0r5Cl8bARzGpiTJGn9Y/OX/xXIAK9JknRZkqR/+vHlp4CrkiRdAb4JfEUI8X93q/43IpNNMTU/ykfn36MykOdLX3yZdCyheCEGCXISfwwhFcQIZFUhVhUEMTMTwzw+P8WhwTK0PfRIRbe6JIaLViwRZUa5vurz1T95k29+63uYqs9/9OUX+cVPPkpeD/A6VWxbZWn5HtVmg5lDh9ALGbpSSF9JCDWJSEjoms725iaVXIGULhH1quhKgusI6q0Iz43IKDq23yPn9Sj3JeR+SBImuJ5Hq91BVlTkBGRJo1QeQtItZCtFYpm0wwA3ErhBjOS7jGVsrrz6GsHqJlLL5b1Xfszpd9/nyPGjvPjFz6LmbPwogjBGi0GNEi6dPc/Wzg6PPf0UKTtFnMR0ul2SJMHt97l58yblcpmRkWGqGxtcunSJhx96iAvnzlHIF5ic3I/juuiWycT0FMtrD4hEjGlbDI+PMTkzzfzhg/hOD1tXCf0uG7U1fMknN5AHBLqAbquJF7oIWdDutHD7DpquoaoKQ4ODqAlsbW+CJBgaqhAnAYVihnp7F9VWmJga25OJD5a4s3QbP/AYGBpAN3RK5RLjg8MU0hk0U+XEEw/zH/7Or3Py2Ue4f/kj1q6fx4z62BKkVJX7d+7y6quvceP6DV544Xkee/xRLl0+x41bl2m1moyNjWEYNnEEvp/QaboM5Pc2XmvNHRr9GgNTZSaP7mff0WlCK8KhSWhElPfvIz00TihbBB/DhWU5xkTiwYXL1G7dIRvEpGSBrEIgJ4S6TCIkEkXFLpYojo6xurNDoukMzk7TMmSKh2fw+i6333iXxvYm+VIJ4Qqs3CDtSKa62yQMI6I4IokiZEmQxAlRnBBFEAQBURgjYgFCRghpj2z2U4b05z4L/1/GwEhGPPeFMp36DmqscHj6IF944fP88LW3+MZ33iBVyBPKKj6CmJhQRKi6BlGMEgdUrDRlPcVvPfcF9Fjm1aULvHf9CnU69G1BrOlYkcXjw0d5ceFZDuQ9us4marFIMxG8ce4iN26vkc0O8ju/8zdxmru884PvENaa5BOZrB9h9F3c3V0quTTHj8xwf/UCxYkhXEXh/uYuvifYNzhCRpPZvv8AwyuQOpYjKsMjn3mRmZPPIYshblxeY+XiZZzmLuv37tKqV/e0HwkkHY80BpO2wlB+gF5gEA/v49iv/grazAQ7nQZb95YYnZ7CLuSxZQ1DVrlw9hzICo+/8ByqbfLqa6+yfPcOuxs7TAyOsHT7LrKq8Lf+7t+hMD5MdWmVf/Wv/ogv/cqXmZydod6oIckypm6xW91FhIIP3nkXU9d57MknqXXblAcG6DRa9FsdyimTsaECP/7hN/kX/9v/xFhlkP3js3z6c7/Ces1lpdHir/3Hv0ksCW4s3mR+dprY8fnW17/Fk08+zZEjx/n2V7/B1uoGn/nCizT9Nvc3V/jEC59Cx+KjVz9k8c4iDz/6MMdPHOPyuUs47R61zSrjI2Posszpy2eQ0jJ//bd+DcXv8f1/9vvc/dGfMSFFqEkEqTwjR56glxrgx2eu0nRdDh9aYHpyjAMHZrh27SYb6xsMVUaY2rePbCbN7s462xurDA0NomUE41NZet4qi8uXeebFT5AZGGW30ceSC2SGR9GKg0SJujfwghhN2wPIdi7f4vd/9z+j2HSYzWZRRABagqRJRCJE1tKUxiZJD4xwa3EZEassHH+YuJJn4umHWbl2mwtf/wFetYlnBazk66y6fc7Xe7T7EZYkyA9YDIxmyFdsVANCP0BBRpZUHDek1XTY2W7Qa3nsWSRI1O75F4QQJ3/SePyZkE47jse5C+vcXu5zfbXHj89c5NX33uDTL7/Epz7xFGGnixqEKGGIIimosorme2iqhK9JrHtdtvo97py7iLpV4/mRg/zeZ3+V33jkJR7LzVLqayRNB+E7NGsPWL62SX8DfvCHr/D+N1/h5OgQpw4dQHZ7vPrK95kYHGdieBKnHxCrEn05JlBlrHSaXtuj0eiRHRpEsjXywxkUG9zYp9ppE2oKejGLYqUpTwxgFTP4IuHc6bN4CcSKTC8IcZ2QwAmRQxARBAh8WeAT01Jlur6P2QsoeDHFUpHvv/Njvvf6jxgtj+Js1rh59hy76xv84LvfQdVUnn7xBYQsUWs00E0dU1d58vFHkYWgtrPLgZk5drZ28Hp9zrz7IfPz86TzGXY7NZr9NrVmnbu37/LWa29y9fwVPnr7A4pWjsWrN0nZKSRVodtqM1woMZDK0qu32FhZQ01kipqF2nFYOneesLFNe22bjdsbhEGCZhv4cUCn2+LYkSOMT07QbDb2ZjJuzOl332fm4DTPPPc0umqyem+TTqNH6Me8/fpbbG9uMTm1D8/tkUmn2N7a5P7mPcpDRU498xR52+DOh++wdvUiWuCDkFB1DSEpdPshVqbI8YceIZst0mo28NwO95aWmNi3j2PHjuO4XVZXV3CcDuPjo7iBS7PbxesIGps+g8X9jI1PcmPxKm5/F9tWsIujKOkigaQSCBCxhKGY6IqFHKssvfUuQb1BKp+lFydIio4aq0hBjBwLIkunUC4RtrrEQUh2/yjS5CD7Hj2O1+1x/Z338JKIpiWz7jYxRxUeRNt0hYupm0hIe4ZFcUgoIqI4Jo4ikhCSCKIwRCQxItmjoEPyl1k9/GyUOQd+RLeuoshpgiBGTctcvHAJy0rzG7/5y8RSwFvvnUNLpYnDGFM1iEWIiOI9Gzk1wQ187rd22d2tU5HTFLNZFso5psaPszk0w1J1G02x6DTraJKJ6yVsr9UQksfFts/xJ57iwHOHeFBt47faLBycZ/nWDXq9LtlUmthvkzJN3FaXldUVJh4eBl2lMjrOqAeS2iRtZLAzWSqlIo1aTJRRUew0DddjePwgppGj1V6m1myRVjWsTAY/6BMnHpquI8cKUT/eg6iYJlkrR6NR47VvfI3SiTkeOnmSxso2aw+W0TM2f/jVr/Lwww8zPTPDg+VlNNvCzmfJpzJIg+PEQnDr3l0ef/4ZHnv6KXzf58qH53D7fX7u05/l3to9rFwKO5VCqAkfXPwAIkGz2SQIQxKRUG+0OJrLoVkGC4cP4tfbrG+sYuiCnVqVgeEhBgcHyEgm9a1NKppJdXON0x+8x4tDLzM2MooIAiIvIm2n6Xccrl29xZ3lewRJxNtvv8PzX/wEwxNjbK/V2V5dxzLTGJpFu9Oi3eqgFHJICsRRSLaYZq3eZmJsjIXpKZbPnufa+6eJOi6mZhLLMbKqkclmyeWzZEsFjMEx0gNl7ty5zm6vS2VkGEkSTE9P0Wo02VhdQ1EkSsVDTExMsLG5harKGC6srLWZPDDHpcULbKw2mNw/TDZvIzQZREKEjFBUkHRIZHq1Xa6eOben1A1DUBRiGeIk3nMx0XXsfAlby9BubDA4OErq4Az5YwfQDYtzX/suUatFN2zRClpEmRizUMZxInRMJASxFOMFEX4YE0eQJDFxJIiSCEWBKEwIwz3fCRQV4p/+5AF+RpICCSRORCz2cGi1tkcjY/Dh6few0gpf/pWX6bhtPjp3h2y2jOfvgVv7UYgAUBR8Yla9LoNWFqvbR3cjtGoXyzKZyJpkjAFGD83R9FzadW+v6MhIk9fzeNUe5779GqPzh/jk53+RgXweMZDjzswsl89+hKuHpA2dKOljpE2avTZ2t8/o0Ah6Ns/4jEkk7VBMFVAjn0wuRcNsI1fSHDx0ADM/QmVgHllLUSgOs39mFq+2S94yaBVydJwOiqIQdlxE1yesV3GiEF+NCUMX2XV4+bnnSPJ5vKEJZg7M8Ad/8E/Zv38/zz77LBevXAVN49GnniSdylBMZ9ls3uHq4k2y5SIvffFlbMOiXW1y8Z0zHHzoKM1Wk83NLQakIcrFCnbWZmb/NHNTB/ijP/xjNFWjUW+g2wbpTAozY5H0far36hi2xvLKbVLZFKNjw+iKStB2SBIfMwpI+j3cXgvL1NCsFGHiUO+HdJodSsVBsrkc/SggjPeeaJfOX2R4cpLY9Vi9t4Ktl8jnSmztbLK1uUmpmKXRqqMiUyzlKZplZuf2s33vDu9/93usXb6K6nTQFAXF0NBtk8HhEXKDg6iZFIliUdEqxPIBlu7eYnVrnfmpWVqdBnMH5yBJ2NpYI5sxmTswg6pr1OtNXD8iciK2d7rs37dAo+3QrvdIZZvoskDVM2iKikAmjiWUWKH2oErS9zl+6AhZWSYjQXtnmzAC1TBQUxbD5TE67T6BplGem0WdmiAzPMD1H79H4/YKO80NGpqLZwXoYxbLjV08wNJ1XDckREAQ0XcDgkAAEUki0FSZMIyJIrF3TClARv6/rFh/2viZWD4MVgp85a+/zC+9/AiffnaGR0+OYqTyWJk0X/vG9zh99jQ//9lPsX9iAL/bwVb2Nh11VUPTFGIhEIpCNXHZkTw27Zht2acdBGiBhLTbJ9eD/soO1DpU8iU0xSCjZTB9menCCA+Pz9JYvM+bX/8Wt86cRYsSxoaGyOYLNByXxNCJNQXZ0hCqwmatTSTp+ImEmsoiTItAUfAkiUbYJyzK2EMF1mpVEllGUXWQZAzT3CMu2yaFcgnNNNFtG3QNzbLJVwYwiyW6UYwT9kGEJJ0WYa2Omgh2mw2+9aff5vD8As+eOsXlK1dottvMH5ynVCpjmQateovLFy7Td10+9ZkXKZTziCTCbXcJ2y5TU9M4XYfAj5CEtMdcaLQZGxlD13Q2N9YZGRnF8zzG9o2TzqTwPY9adYsw8kFKaDRrxMkepcpQVXy3i9dq4O5uoQqfSilHPpPBUjRUVNZX1qlv1/ccqhWVVrdDGMdk0hluXLlK3OnQ2q0jRwmbq5tkU1nSqTQ3b90CSWJmdppUxmZrZ4O5hRnGxoY4986b3D13Ft11sQVYukEmX2RkfIJUJksURTSbddxeF0WWyGRzFCuDbG5t02jXkZU9SM/RE8coD1RYW98gQTAyNsLM3DzpbJFsboA40clmBigXhtna2OHGpTOs3r6O36ljmQqJlBBGAXE/YPXabVKqQT6TIZNNYaQNhCkRqhKxrqFkMqSEwepuDevYAawjU4zNzdO8u8rt0+epdhvU4z5RSqZJi4bUYa1WIxR7rmlIIEsSSSzodVy6TRffEwShhB9EdDoOgR/j9yPiSCBJf/kh/jMxU8ikLX7hs0+j6AJJS+gHMfWuhqbC1/71H/O9H73Br/5ymc9/7jN862vfodXsIqVNYpGgSAqGuvfHu16I5zbx8wLPsAkafbodn3I6z77yIKpl0eg26VRrrCzeJhULHj5ygsjt0e71eHh+gfTAANff/5BsNsXR+UNcuHCRdqNBJwoIRUTRVlESi07TY2e7SX5ijFS5hJUtcOPmbfqtKpISUTdCFkSdsX2TIGJcp4VpjiLj0+u1qK49QPL6uG6XSAhUQ8c2DYQPei5Pr+vSiXxSmqC1uUp7eZluz+FPvvsDPvn8J9g/NMg3/+zbdByHk088wdi+fciygkgkrl+/wZ3lZX7pr32Z40eP0O10SfyQVrXK7PQBBDJBsGf0ousG3Y6DFCcMVQa5eek6nhcgFxQcx2FoaIgkjun2mjhuj8pAidbWA/pOF5Ek1Os1mtUGatsln85B0GNssEC5tDfll+KEVq3Fxsomy3fucfXydSoTo1iWRTGdQRF7NRB7prQBlmni9BrUa1WKhTzVxhbdbpe5+YO0W108z0dxXV77xtc488aryJ0GKcsgDkJsI8tAZQjd0okkqFW3Wbu/QS2U8BUNK53C9VxWVu7Ta9V46VM/R6lQRpBw7MQxTn/wPo1Gk0KpyNDIENu1HbykgYpCq90llyvQqNVZX1km7PuEsUxWyNjFEUQcEzfbrF+4TFrWsDUNQwWSmF7k0fZ65FIWtmawuVlFHS6Te3SBwvR+4laPG9/9MdXl+2xGdUQlj0hHKJFGfsimc30XSVMJ4hhZVyFQSBKVyBO4bQ8psVENBacf4nRdCMDr+yRRDGiAQJZ++k2Fn4mksLa+xX/xe/89qbyGmVaxs3nSuXHKlQp6ZgAz3eaVH77Nlz//83z6M0/ztT/5PpGQUGSFOElIRIwky8QaeEGPMBRoGRVdsejXI7pyQFaVKWsmo6UxykrMiqGyvbXG5WsBM/snGd83QTcIsHSDw5P7Wbl6ndnpWfZPTrG2uUYvjpBU6MYhCQl4Mkkg4bkhcdcjW6rghNdZq1YJooDiTJndrW3Ghkf44M03OHzIR4ttZHpEQY9LF88i9fvEUUi11SBXKDG/f5aRTIUkZZOkbVzHx0xA7gdcePU1VhSN51/+HCcffZwff/1rLC/d5/DJE1QGBzFSKQC2V1bZrdU5eeoZnnzmFMuLd5gYHWF7e5f1+yscPfgoH3z4IdcXrzM5N4Wu6XTbbUqZItsPNrh0/iLZbBbH6aHJUC6Xcd0uzWaDrG1iKiqdWo3VpSXu3V9k98F97F6fiVSOgm0ji4CJ0TEKWQuSmPtLy5x++ywfffARUd8nimMUQ6dYLKI4CZZlsXZ7jU63i6KpRHGEKgt6nRaTB/aBGnHnzl0sw6LZ7OB2Xa6/+z4fvvYjWhurFEkItBhLVymXShTSOSIlpNnpcOPeJrVQQaRL9IWMl85QKBUZHChz48YlgjDmF7/4JUItopDJcvjIYe6v3GNSlkgXhsgVywQdB0WNSOKQdqPGg5UlNCsi6nepri1TC3zG5iUqVpGNq5cIV1exsjpECSEhkhQRSDGBslfV140iUhMVph49jjlYwlA1rr79BtW1BzTdBlFJQSmXiBUXPJnJyVG803cw9DTdMEaWAR80ZNRIod/0UGKZiJhe6NDve6R1kwQFJIijj6cXf4n4mUgKYSJzbzsm2gmJVIA6qrREJCCfybIwN0/sdfjBj37E5z/9JC++tMC3311ElgWqooIQew44qrS3Nvc67EYxmlYml8ujGBm25YSw51CWDCwt4NRDR/GPzLG5uclWu04n8pmdmUdCYXiogtlpsru5xbOnnuPKnVs0axvkCim8rrNHz1U0uo0uqlCRJQ3DtEgXiuzTZjm4MEWqYLOzu0RGNxmvVJgcKUHYJJdREYlPHPaxVTBsm67fo9Nu0mzUGMuWCBWN2NBp1moM5EoYSUz13jIv/e5/QuXIcd5/4w0erKxgmgaqrOw9zT2fRr3O6Q8/JFfI8/KXfhG326a+tc1UZYDbl67g9xw2q1UuXLpIIieMjIygyArtdo+0luLixYskYUw+X2B9eYW5mWnSKZut9g52yiafTbN+6w4Xzp3hyoXz1JpVbCIsy0TTNWQpwfcchsp5yqUCXqvJ9avX2NzYRMSCsdFxdqpV6rt1xicmyFgm951dHMdhfXOd0YEp3P5lVEVQLuUxTY1yuczi4h001WR1Y4eBQhF3Y4mk3cXSFFQZvMijXC5jWzZBP8RXQ9a3t+gHfVLpQXa7XUb2zSBrOrHj89CJE2QLFmdPX+TcuXO89OJLuH2XdDpDHMesr28wPH6AXKFEIHfoeT0kWRB4LvfvL2LnVBRPYBd8jJTJ4u0r5GaOce/CGSoyqEjEYYhuyHhxRKxAKIMTBgzn8mSfOkRxfho7U+DeB+fYWr7Pcm0DLw1m3uLm5iZbzW00fYfjj05jJBKxohCqEUoQkzYMRBgThz5eP6Tf9RieHGR0fIQrN27R63lougFCQpYUEvHTVTL+efxM7Ckgy8jpPGq6gGoV0KwcurWHCOt0A1bXd8kXBvH8iI/Onmd+bpa5iQGUICClSGgJKEjEIgQ5xlAUXK/PltOirvowmKZjJDQIqIsQPxHYRopCrsShuUPMz8xjqjrLS0usrtxn6eYtpH7Axp37jA8Mc2juIEkCiZDQNRNN1VAljXatR7fl4bkBURxx8rGTvPTZn+Pnv/AFDh85xBNPPcngUAlZCVhdXuT2tcsErTa2pJFPZTA0E9tOM1AoYmsajPxQkgAAIABJREFUcgJCAl3XSBSZ9cYuDb+LhkTZsJnIlXjr+69w98YtUpqNUGXmTxxGNRUCp8vGnSV2lh9weO4gtm2zvHSXw0eOUN3cZH1lhUw2y8Ub11i8dZPZmRmKxTJRkCDHEl7XpdNsImsylq4TRQEz87P0PIcgCqgUSvQ7Xc58+CEfvPseftcln0pjKAaaIqPrEoqIEb02itMjpSqgSLheQOTF5IwM5XQRyzD3REytPsVcAUnIGLLB5nqV4uAAQonRLUG+kCGTzmDqNmfPnee1H/+QgazNWM7Cb2whex3SSoKtgm1oDFbKGIZGImLa/R6bG+tYssq+ygAH9+8n6nsszM4hRxB2+zz20EkePXmClQd32dxaRTMVwjikUh6m13Op1jbw4xb5is3K6j36/T6FbIG5qQVatYB+pPHIE8/Sr/fYvnyL2pXrdNZXsEyo1Wo0O20kRUIFspoJ/QgjnWXyoeOkjhzAGB2lefM+tat3uHfnFtthk0OnHubgiQWcyOPuvS1koZFPZxC6QSTLmIaJnMSEnovT7RL0AiZHJvm1X/oP+P1/9Ps8dPgksQcSOiKW9ujYIt47lJT/f+b7IEREIrVAUlBFgkgEiWQSRzKyLFHbbdAeKJEvDLO6sYptrfH4/D5623XaDQfFMEDRiRTvY/GHimJrtEVEQovRVExfDpEljSDoE8QyRdkgi0ZGkhmwylgljV6/hxN5NBu79JwekWrQvPuAUyce49qlc3iOS1ooKDHEqowiyeystTB9QXo4R36kiGwo3Fq+ix+12ao9QFU9vF4bNUzTfAC/8rkFzEAlcvaQcX0nQIpkRkpDDJYHCRMwRIIkxfQ1QSsJKcoqcq/P5VffYD1yGBnYR7fRYuHkCcyhHDu1TXQ9x9bNRUYzRY4dPMTd24tMTe4jn8vx7ve+i5fEDO0b559/43scmBlnfHIfSDKRH1FJ5di8vwpC4IU+cd8ll0szMj1O3WlhmybC91i+cYvL587j9T0ydhqJEM0EmwhTVUjJAst1iDc2CLa3yM/OUaoMcbV+g4KUpqxnqKZSSJpKZ71By8xhm2lsNUd1eRtFgtyAjec2sW0d4cfcX1rFdfpcPneaFxb2Uwpi7nY2sGUPM45JxYJiyiJvmWiagqQobK/soEiQVzQKksr8/CHeu3qTWrXKkYXDLN6+wIHJcY4cmKNW3eT2/asUKxkKxRLtWofRkXH8sIkTCiQlwE9cnG5Ea2ODofwMQ08dpzIzwfkbq5x98yyPj8/gtW+jui6+prJbr1GxyghVRvMVirLNhi+RGxohf+QwxsA+oq0G2zeXuH3tCluNdY6/+Aizzx7CS3yObtf/T+reNMay9D7v+73L2e5+b+3V1V29d89MT89whrNwRiSHHIpaSJuMLcRWDBtBYjMBIiQGbGQDggRQ/M2OkTiwESeCjcSyZRmSA5qiRFIUyVk4w9mn96W6uvbtVt393rO/bz6cGkqxFGuk+AP9ArdQ59S5955z6rz/9789z8ODBw9YOtUkLTt0PPAyReNwTDJKKDWaPPX8i7z04k/xwvMvMndikd/8xjf45r/8NjYGXQqweV7oq5JjlEGqf8dyClBkVkVRyS1AQlkOArRySOKQB+trPPXEBfx6nXdu3OSFT1zlypOP8sob1wEQyoAUKGUhs2hZYCSkhOFkgO9rHF0QuGwdddkf5MyV69SUpuH4ZGUfI3PyiUF5CUJkZKllc22Vq5//DGfOnOfW+z+iWgpQDogclFTIKCNp98l8h914gjMVYB2LNDGH+216/R0qgSbs7HNh8Vlac1P4JZ/zF85jTEqahJg8JwgquNIlz3PAMB6NMDbH5DlxEiNUIde2cOkc1ekmjdkp0rrD1soa5+YW2Vhb53DQ49lPf5bOsMdr3/sej1+6wLf/xa/ze7/921y+/DjbW1sYm/O5L77M/OI8vfGQ+UaLuD+mvb2DNIbAcxgOEh698hhaKLI4xq+WiEcj7t6+w7A/wJUSabJCaxGJPPbUSq5LPB6zvbGOf/cuzSuPszTVolYJkFGMdgu6/TSKEIGg3T7AcRVaCm7fvkVqcp548hP8YK9DGI4xmeJwd4dLy6e5fbSDThKSJKS3e0BJO7hCoPIM3/MwWMIkQbkO3YMOvnQJ3ACtNN3OEZcvXeD+2iZnl04z02zx9ptv8dJPf46N7XV6vS5rG6tUHi2zub/K1SeuEjJiOJzgCs3hYcjh7n1e+tTnOHfqIrkR/F+/+U948/3XefzsWUwWcdCbYI3LYBDTbu8xtzxDZiyu55OXKsSuz9KFS9Sm53AGMas/epf33v0ha6NtnvuFzzH/3Hn28kOUD48+e47XflRm/mSDTvuQyjjFjxIemzvDy7/4BZ56+XNcOn0WhOD6zVv83b/3P/M7r3yfUZ7il3ywOQKwGCQFsvhPMn5ijIIQhf6DwIIQBRJNKjIsaMXRqMf9rQ0WZxuEVvBgZ58nn3iKM0d9bt97iFAaP/DIbAoZGGvQnkelXiV3LZ10iHEMJeVza/UWk8MeNaGZ9ss0lEc+mmCTDAuIdMTlC2fwqy3u3rzGxec+wXPPPcfKvdv00wSExpUxOhUEmcKMDdnhiNuHqwxLGaWZCjUlMXaCSTLCNKYcNLn6+BUG/UOq1Qrnz59jPO4THis7pWkGucAaycHePu2DfWyWkZviZfKcRlBiamqaxvkzBI0a3/7utzh7epl0EPLB+9eot1p0J0O21laolQLuXLvOt77+r4hHE564epXbDx7whS9+nieffgrrKKSUzNSavPbaO5g4wSQZ1akGqcm5cOlCIQlvU7JyCRsniDRDGkvgeogcbJajrEGkGco16Nyickt7e4tbb/+I2UuXaTRPsjw/y3Z/kxyDVpr2/j7L00tIKSj5Ps16nd2dHcajPtNzswUF/XDI8uIMFdfFa9QJLpxnxvPoPriLnUzwXIULeMcaDEprko8MgxFoBI4QSEEhBCvg9MklDvd3Obd8hq1XH7C+ssrLn3mJd66/S69/SC/scDDYJsnOMbs0w8b+FtkoJ889KuUay+ceR0mPa2+9w6s/eJX5kzWySR+HRdIww5EVNnYOkNriOEWJUJVKJEGKnpnl5OXHcOot+m9f58M33mTtaIduxdKvWRamfCpJQh6NWDg5zc9+8TkWiel/sMafPfcMz37pqzz9+Zc5VWkSrm/z1re+zbs3b/CjG9f4cPUesU3RvsKQI438sWEopJYKhPHHHT8RRkGI4oewkOcGpEQLQZbGJIB0NHEKO+0jStUSbm2K3aM+3uo6iydPctjr0e52EVZjcoHCkqYZXqCpNapIVxHHGZ18wu7giO3xEeXAYZwbXBuTT1KarounHaLRBK0VuUlwFRy021x/5x0e+6nnObl8htXbdymXXUQe46Nx0Jgoo6nLuCms3LqPntLMlXxcbWk0HKo1n7OnTtEfHDE4ukHJmSYMJyRpUqhlZwVJqVKavYN9Vu7eJY7H1H0PKQoPypqcaDTiRKPJ8tVHeePtNxn1+gQ43HjvGlGc8PJnfoowT9g52OHsySXeeriKiVP+07/6NS4/eoXf/v6r/Jd/7Ws0Wg2Oeh3yKKK7t8/B+gZTjSaOkigleeSxR4jjiN2dLaanGqg8Z39nl6ofYKII8hwNCClxpMJzNI4Q2DRFZDkV5THY2uDG97/HpU99gamyw7BRIsoiatUq2zu7ZCYjTmKc2NCs11g/2GLY71JrNlGOy7BXUIudPXWSvc0HnH/kMnVp2dzeoGxBJQlagOe7iLyADAtHoZRCA77QaKlwtWIQh8QCmlNzWJURhyEvPP8ir/7oVU6dXeaZpz/Jw711NrYfUmoF9CY95rxFfL8CSvPUEy8gUPzglTeIhmMe3L7Jp55/lnpD8fC9d5mt1TnaPWTUj+j1Izxf4GiFlA6JcOjklqXHHmPuzFn2Hzzke7/zbR7s7bJpYt5bWefb/8sDXr5xlb/0lS9xdvYMo62UrY5mc3uTi5ef4pd+6b/Cby6wvXXI9775Da797m9xe3OTiSvY6rUxOsfxFKkAYSyC4xbnj3BN9t/B6oMQRdVASEmhFJuDlUitUBiMzXAcTThJ6XSHLC0scDQK2dja4ZFLFzl/7gydd7oF30JWSJBleY52HFw/IMstwlVEacb60S69eEi5MUXgBRBahBQIqajogOlqgzjuMRyPmW0J5mo1Vj68xpVnnubqo4+zem+VyFqsyJGui5CaPEuJRwlnF5fpiiE926NaUpRch4qn8JUmTyJW7t8lcBc4O18jTmJqtSr9XhcBhOGE1fsP2dttY6KYasXH0YpChdUgrCELI2ZbU/R6HfrdDmcWTzI66nHrxm1e/vmfZfHkEmkc4mYpK7duc+O993ju6Wf44hd/jl/9Z79GnqWcOnuKeDxmb2OTelBhbWWFZDgimJ2jl05YOnOKE2dO8aM33qBS8jk5P0s8GGLDiEBKTszMsjHoIMgRWYLMMxQgTI5NUzTQCjw8R3N45zZTM0vYUNFoVnCCGvv9DkmckJoMIT2sOXb/04QoSZgOPBrNBuv7XTqHh5xcmMfLR4w27nK40aW3u4ljc8gLQlUHQZ4kpEmM6/vk1qKFPoaoC6qNOrGQ7Pf7+JUKVTdgPOxz4uQCy8unef0Hr/Nn/sJXeOxyhQ/vfcg4mzBMQpLc8olPvkA4mvDWj97mwb2brNy/h+9oZqYqnD61gMgnTF15krgfEg9DtjY3MWT4nkMex0grSI3Ab01x6vIl2jub/PC1V7mzt8X6qM8HD9fohBE1V7Pz+gZvdF/lnueze3eX6XMX+PJ/+Dd45AsvkB0O2PrmG7z9a9/jrbs/pNMaEGvNVq/NURoSa4s57tWReREyGwuYQiAJzDF12ccbPxnVBwTCFnVVJYsYyGIAw0d+jwAUioPtDmkIjeY8jlOme3REvVzi5Pw0IrG4aHIjQWiq5QbCaqw5ho46Du1OlzEZAzPClBV6ykc1PfJAEsqURGXIwCXNIA1TWkEFMxhx8823ePTceWaaLcIkwnoOiRDIShnjuuy3j9h+uMm0W+FUbYpGQ9CsOTSrAa6AeDLB5BmNZp3RaMi1ax+ws72FtTkIi1ISz/Ow5AS+i+drtBIgChUorEUL2Hu4xuoHH1LWDr39Nq987xXKzSZPPv8MO5ubqDBlrlLn9ocfkIwn/Ad/6S+zcn+FD97/gGeefRatBf3DQzq7e8zV6mzef0DZcZgMB/gln4uPP4JwHTKTkSUJ4XDExsoKJozp7h0w02iwfOokfjkgSiOSJC4o8rQmTeMihk1jqjajFEVsXb9GNOwQlBya0y0atQZ5mjEzN82JpRM0azXKgY/Ugm6/W2hhSsloOGQyHqEVnFqaJ/A1uztrRGEPTIYrwdMCaXOSMCQZh0gL0kpmpmfQjktuLcNoglP20YHDQbdNmExwSx7d0YBHH7sCVrG/1aZWbtJqTLPXPqI5v8iZC48grUt754hb71/jxntvs7d+j3i8R3NGs7ezymyjxcWTFzna7tA/6hGGfSpVTeAXzUtlNyDPLRcfexTtKV75zjd58M6PaG8+YO/hCtN+wNW5E1ytL7HEHNd/uMLBoeILf/Ov81f+3t/mmS/+OQYPJ3zv7/xv/JP/8Ze5+f4PcaYChr7lXneLnaxPGFgSD6y2KGGR0pI7IH0H4UnQxaSy4uPHDz8RRsFaUEqjlUJrhdICqRRSglICrQQFf4WEGHbWdvG8Kr5fIppMiEcjLp5eJosSXOGSG4XrVahWm2jlFtUI4TAeTgijGN1QTPyY3eiQDmMGXsbYzwkDGOgEEQTUKlOEowiRZLT8Equ3blFSkiuXLyGMIccUFjrwUPUaMbC5ts7koMNcEOB5MbWyg8giNDlZHDPdatGo11FK0e/3uHX7Bru7u4UuodacPrPMqeVllJS42sF1NUjI8hStBJ7jcO/mTR7cus3qjVvceOc9bt+8xed/9udIlaTdbpMPi9zIsNflpc98hunWFN//9ncI/IAvfPGLjMMx1959j6W5BXbXNuns7hNojyiKeOTxK9Rnp+n3ewDs7+1is4yt9U1ef+VVekcdlBU0Wk0Wl04wMzeHEZYki7FYlFZYLDLL0FFEVVjGB/vsbjwkjCdYCTMzs1RKFbzA58LF8+RphskN5aDE/t4urudRrzeI4pit7S2iJCROI0pll3LFw/MUNk8Rogg7hS2SnPEkROQGX2kW5ucpV8okeUpQLTFJI6YX5wiqZVKbgSMYJyHjSchnPvMSezttugd9ttf3uHLlE5w8d4Fub8yND25y/9Y9Nu6vEHYOcLIRrRoY0aHX32W60aK3PyAdWY52D3Acg+tnuI6iGpRwjCRwfZbPneXevRvc+/Bt6O0zO+hw1dd8qtVisR9RbU/wM81f/ut/k//iH/8fXP3yTzMcj7nzyru8/4//FW/+xj+l77c5esbjHW+X250d2joirDqMHEPiGHJR0JIJk2M05NpitQJXYsTHVXwoxp9W9+F/AP4a0D4+7L+11n7z+G//DfAfU3A//efW2m/9cd9hjCXLcoS0fEQZpWRxMRow1iJEsV97LsNBn363T6NauO+HR4fMz8+jxXHySwhKQQmtHLI0x+QWKeFgb48syRCuIfccJiKhb0f4nibWCp1ZbC4YpSkYKDkOQkjKrscgmnD3+g0evXyRD6+/Q5oNiLKMcRqhpKRcrqCspr87YH6+hFtTSGGIshQcByXASkN3cIQJS1SrVZIoZGN9DSEMrdYUBlheOkXWH6FEjsKSxxlW2IJdJ8lQxtLZ2iayloO9fV7+mS9y6swy/V4fkWWYfp80jWmWKzx99Qneee1Vrl+7xhf//Fc58+glbj24y87mFo+dv8Rr3/w2ZcelP+yxfOUyjz79BMPxkDxPOdjbhXBCPfBRWcq1d9+iVgo4eXIJYw2u7zEzO0fTc/GiETJOsHkh3hJ4AhtGSCbUm1UmaUqWTBAYyr5PqeQTJRMcV1Ou+FQqAeVKie3dLXKTUS6XKJd8djbWmam4tEoCk8YIJVg6fwYTTRgPBiSTSeFVGsNkNCIaTQhqHqWgxMVLl9npjJhEMW6zwSiJWT59iqOtfQbjEUGlRJTFzC4usN/vcP32HTa3Dvi5P/dlTGZ5+HCdQbdXKEj3OuTxiNnZgJMnmwRNB0fOkYxjJv2EQWfIUbdLc7aG6ytKpSpGKiZxyJnLFyFLWL91m+jggMBY5q2hG6ccbe3w1GNP89KXf4HFZ56nurzAWjzCvPcBN771CrdvrNPZ2qL2yCwHzZy3JrfYHY8p+S65ckgwRchtC09aa8gyS5KnqDxDoRAIjChC5H9rRoFC9+F/Bf7Pf23/37XW/u0/uEMI8SjwF4HHgEXgd4UQF639N7dUWWMZ9SOUo3C9gqGneIMp4mlypBIYmZKRkcUZ++ubzFy9jPLqDCaWvhVMtap0OwN816dZ8XG0IM0NQmnyLGcyGOJjEdIBBNI15CpiLA8h93G9aURaAgmpiEnzMSVRp1GqE6QJGyurXJxpsXjuLHduX6MUeIBBm5zAcSkZn8POkN56Smt6mlilqErAJM1wAw8ZQGq7jJNC7alEQHfQY+3+PZzHLlB1p6ipGpfOXmB/dx07GuGaItciLOg8o5qkzDuazUmfE+dO8Omf+TzhcERFQD4c0uu2kRb8Scb2jbvsHLWZO7PA81/+HKqi2Flfoz7XIN4/wGzvo0zOuKo5/eITpI5hf+UhrWqZ7t42l+fn8IYdsr0NxOiQjb0Bo8kB84vLaKVwpEO9OU1DzEC3i+30MJ5iICzSCJoJ6EHI7HQFV0KgclKbUKooDoc79MMDpmarZM4CB9E+w1GH3ETUp2pMz1QYrI3o3r9OlEXM1oMCKuz4TNV8ys0a3e090uGELMnJogndvS6+1yBXCdKvsXDqMkfGQAa1kgtxRqu5yEZ7k3q1TOC7dNMu5TMtXn3lJotLp9m9vUel6ZEKSW9wyPbKXfK0iytDLj16jhNnW6zsPeTSzCcJ90Kibsxhp4vyNL5fQSKxjkvoetSmK0yfavHgvbdwdw6YijTNoELfy1AzDV76yp/lU3/xF2ktnCTcbzN68JD4R+/y1nd/nbt33iNOoTo1TTy9yP3te6Q2oea7pOS4mUVkxSyRQG5AOWAkBBkoAVpaTJozCeFPAp7+Y42CtfaVY+WnjzO+AvzaMYHrQyHECvAsBZ3b//d3AHGaoa3BWIOjHRxHIgu2R4rLtkihQVqEzumNRhz1+5w+d4pRNCKJU5YWF0jCiEkC1kA8iQmqdYyV7B926HR6OMrBcUBri9SCzGZERmCzjKjXoSxz5qoNXB3QPRqw2+syPTNPQ0muPbjP3X/+L2idmMPDoVqqkklJamPKWuOUfURX0NvvUWvXCEoK4cJg2KdcnyFOMjxRwpMernRRGOqVJkfjfe7cus2lc1dpegFeqcL8iZOM2/vY3pAMg/AUYTrBtTHNuRlW1/s8/fyzJCYlySKmGk3u7Gyze9RmrjlFOol47/67iHLAF/79r3LizDnC4Zj+TpvFqVk2H64xSmO64ZiLn3iG0+fPc/36NZSxeFKydv8+j8zOMBmOaO/v42pNyfPZ3dmlczRkfm6OhekWju/hSIgs9EdDpoMynqeplKrUnArWq+LPTTPWAmML/EicxDxcWUV8+iWyLEcimJub4zAaEg7HhecV+MzPTKO6XW5f/5BVUk4vLzA1UwepCo8xDJn0ezSqdUgN7fYOKDhz4RS5lHjVEhWl6RMzvTBDdzCm0WhxrnmOOBowXakSdQY8WL3HubkFZmbnGex3mKSWnd19Bt0uKotQCmYW5qlM18mxiKEhkiPk2GXS6WKiCTPNKr5MyZKcSqypGclSc47tlYfcW12jcmKJ9YcPsWWXM8+9wAtf+SozFy6TZpaD997l4O49HrzxQzY++JCDvVVQPsvnlrHlCu/vHaCoMB1UiFWOkeZY17JYULNj5S+RgxAac6xVIWWRm2s1BH7gs7+y/7Em8f+f6sMvCSH+CgVT89+w1naBExTiMB+NreN9f2gIIb4GfA1AavFjDEOW5JBZsC5KC5T6CM4hCo5DAUJBFljurD6kOdtCao/RKOTcwhT96hGD3S5b6xtIr83swinqjSnCMCTJMgLfA2KssORGIGROLBycks/mXo/hQYdLS4ucm58laNY5Gifc3nzIsNPl4dpDDkcjnnRc5kpNHKFJhcAGHqGx5NrB1x6BcUg3U2Yv1RjIAda1ZCIjihMkFmUsgXZJohH1cgXlG7YOt9jZ26Z2qoVQkkma0x2PcdKIkiPIVY5yDIaY/njI1OIC1ek6u4c7NKZr5HlImkSYMGJAn6g/4HB/hyufeo4nn3kGa3Ie3LmHClPqueJ2+5DdYZfK/DTPvfwZoiTmcGeXJx55jFvXPmBy1KVVKhMOB4yGPUycUikFIKHbHbE+GDLYK3P+ZEFgYpQkEWADl/rsDNWghK98cEpMAg2+puQ73HuwytrmGihD56BNK/BpNBrYgy2G3R5H+wfo3KMS+GSVCl6esJ6lHO5tEfbanD53ioXFeZrlMmiXxOYYclxXE09C2jtrhOmY2eWLTM8vsz/sMHZy6s4MpakK48mIqVYDT6aM2x1+8N3fZad7yOVnP4kp1QijEQfr22STiKrnIhxYXJphnHV56/o1PqkfYdrWaN/dpiVyOltbuCam5gVIkeJrqCJZrLYIu302hl38ap1kOMZdXOSzf+YrPPazn8MJSnQfrLJ7e4WtW3fYXbnL6q33IZ3QmF2kuXSaoyTlzev3uRf2kb5DkmXoAFAUnprjFFoPQmDCgsXZcyWu8nA8F8/zQYPBkEUfH//wpzUK/wD4ZYpF/peBv0MhCvNHBS5/ZNrTWvsPgX8IoD1pP2KC+Aj/nSc5mKJ/XqmiJq6QRXKJHOVLosmYlbV1Hrl0kVEvY9gfMjPVot0P6QzHJFFKp3+bwK+S2Ryli/DEyhhJXsRZWmG1YpxZhhL2ooxofZNBPOHq2QsgFXf3d8l6A6r1Kp6nGe7scvmJp5hYUwiLmoTMGHIEjnbw8TD7CTrTdEddcCHNU7q9HvvdIWWmcASkxiJSg5aCZqXKqNfjYX4fhMf29gY6j5j1HCIpidIQkScc7e8yIuXpL/08a/u7jE3KOJrl2hvXuPbGK/jDkIYOONzeIxsNEWlCtVJmsLfPys1bnGhOE6Sw9mAVW3L53Fe/xOLpZV5/7TXaO3vs+VU2b9/jZGuKmutx+9qHHO7soBWYLEVbS833SOOYSbfDyqBPf7NK3XeQ5IySmDmtiPOMzIbkeUqYJ3RGIwY7ba6v7RHFITONOuFwzO21h3gljzs3bzFMJww7XZYXz9GolMkcwXjQxcew0GiQmYj99U3299vMTzXReYIUmjRNETqnGijG45BOu83eMGEWhwNp2Ys7qKZipjFDPMwJB0f0dnZ559vfZ9g+IjQZ29s7PPHic9y5f59GrcJMqcwwj/BrmvmTU+wPJdfvXOOFJx1mKXFnZ5P9cYIMI+aaDZQMsTbGdTxSaVnZeMj+ZEReDsh8n1q5wYuf/jyPP/8ig86Y1bsfsPPedXbev05nbZVJ2KNUlbTOzaPnznFru8sbd++yOhyTlF3oR5DmuGNFajKUBO1AloLjKqw15LlFiQxBCMfJegPk2ONO2Y83/lRGwVr7Yz9ECPG/A9843twCTv6BQ5eAnT/+AwueQiFEEQwZiREGUrCmCB88pRBYrAFHeyRkuJ7H/n6HEwsh1UoDGOP5Ho1GlVE4QSCwqWE47iG0wAkERhl8RyMlCGExeY5REMUhozhFBDBMLbe3DhlGOVfPnGZqronvaUrG0syrTHoDBnttqAY82N5gEE4QcYoej2lqKAc+jGByMMHxXQaTAd2kS3//kMPtCS1nimnVolpykXkOUpHIIjTqtg8ZhQkWg6cAY1FC0DvssNvpsR9OqDx+haOdLfZ31nj8U09xePiQb30up9o5AAAgAElEQVTn/ybaPGRWlXiwuUPVaipCcu37P+AHv/qrPPJTL1DWkqlGk50PV9jvHHL+had58sXn2drc5Hu//W1mKnVeXd/mnd/9DrO1KulgQHtrizQcF8ZYgTQZPjk13ysUriYR48M2VkuqvkcsFZsP10iimDCNsFKSCI+xrFA6fQlfe9QqZQLtEI0ntPf3SWzG7tYWqTB09g549NwVSo7GMRmDvR18k+B5Cu02yGxGL8pob+/iiZya7yKlwVUKm1mqrofnlTiIDe2DA8TCFKsP73JwtM7lU+eo6zo769vsrKwiRgkV7ROUfEIfVm58iCMEo90+mUmxNiSvwoXnHmPp9CwHOyVatoY9jAgSF5NLZmYX8L2IKJqABddVDJyczZXbRFmOW61hpUZqjzd3jrjx/TdpzkzT39nh4OYdvMkIn4xgqkHlwgkmZcV3b93nzTubTDzJuOYRy4JbxFOQkJLLnExIYiMQDmSooiNYW1IMomhtITeAACsKNap/a6rTf9QQQixYa3ePN/894Mbx71+nUIX6nygSjReAtz7GJ1I0phZ1Znt8MQWrnSQzBR5AK4WQoLVAGIurHMIkZXNzlwvnTuHUAiajLnNzLfqjIb3xhKDkFgjK/LiyoVKEAZlrhDQYaUijBJMrKmWHUS9HeRotNRvdLtp3eGTpBK1WgMigjKRSChh3OmhqBSfCaEQWhlQRWCExyuJ6Dr2tIfWzDVILaQJeSaOrBpOPGccGbao4wivcq9TgIZEIrFIoR+FIiQe4SuFJBUbhpoq008eZjLm0NIdIevzW13+D3nCbz33+RS7OLPPr/+BX0KOQelDBZIJX/+mvMRkNufCpZ9G+z9r2FgQuV5/7JE5Q4vb716hrj0tnzvEbv/EbhAeHnD5/DhtGpOMxCsjzjDzLUBYUGT4WrUC5Gs9qHFU8TCIKieOMMIoY5xHK0TjaUA9KTJfLuEGN4TCk7gRMBkPqtTpuyWXnYJ73b37I3uY2JomZn2qxm4T0DrZx84yyq3FcTWY0roY805BFuKrAoRhhUBY810Nrj7lyDdNooKenePzSJe7e+oDQ2SEb7XCwsYuPplqtYZGImo8RGYmJqEmJdAQeklgJ7GyZ+1sPeGHxU3zi1KOU+pJ+e8JMbYbq3BSCCGO6WEpYq6nUm6y3uxwN+5xcOkUpKHO4d1iQ/5gD+lLQQ2DjCJ3FBLUy1AKCM0usJiHfff0W1/YGZL4m0ZJYSHIJMrfkqpADTFFFGH3sVcvjOfTjlzjmUBCAFAgBUmtg9LHm98cpSf4z4CVgWgixBfz3wEtCiCcpQoM14D8BsNbeFEL8OnCLwiz9Z39c5eGj61FCUlyjOOZHKDwEc8xEm8US3BwtJBxn4wWSUlDh8LDL9FSTRrlEhiXQlmYjYBSNwRVorYmStOCs0waMRJsCi4+SRHHK5fOXaDw+z7/8598jSi0R0JifZi8cEG1MeGR2Dq/WYJIZmvWAQGr6k5DpUgnXGBIpC5Vjk5M5KWWnRP9wQKlWZnp6lpFMyNwJzQWfqvEJN8bEWY4rKtjcouIUR3u42sOl0AlQWNzc4KEJhKZcquH6TXpGoAZ9PD9g88E6R+015k80+dzPf5ZSXqI2U8OPDiibjIoK2Nja4eDGTV782Zd5f2+btcM9Lj/9BE88/yy9vT127q3ymU88yyiOCKOQmWqdkzOzaJOThBPIMoQoOhYtAocEH4mHwlVQEgJXKQoW0RwpHLQX4DsuQgk8fAySqgG30aDTH+Pj4EqFkQ6zM7NcPHeBldUVkkmIF/joVoNR94io38PLU/yghkYgpMUzIBy/4M/QFoVFFasIKEHNK+P6JVSthqpWeXbuKvnRLvHuPraTM68DhO8DirHNSYUhVUXrdkNaSo7EWoO7OM22G7N/1CdPMq7MnmdyaxebCmbn5qlVp+gP20zCIblQlMtNxmHO2p1Vzp49zXSlztHhESqOaAUeSEEYhrhJjvTAtkp4S3PEjSo/WNvmtbvr7I1yqFdQjsRMouPFIEOaDEGGoEgiCgQcx9zmOBH/+0MiBUgpEEhsZv5Eq//HqT784h+x+1f+Dcf/LeBv/QnOoaiz/riOWli7gj2qQE2CRUlRlHuALCuwANp1jo2HYO+gzZnT55DGJc9CZqYbtHs9RlmKIxVSFQARoXK08ArjkheEl1mUUyuV+MJnP8P1t29z49YeaMU4TpFasD8eo/e2qShw/BK+sUyVfMrWEIYxTdcjl4IkjtCOJDYpqZMTOFXGBxHNaplaxaU7GaDK4CEoz1cZdIcY6yC1g6MUeXq82h7jHSQGYQwitygBnnAIanWSpE/74UPKtk51zmVutsGFi5c5d/URVt+9g19yCFxB03WZUmUcpRns7zPudVg72MKpl/n0T79MpVbj9R++icoMZ5ZO8o3vfAshJSXtUC+VMElKNB6TZxnSAazB5hYpUpRWaGvRWXFftbFIjgWAtSLXGqUzXE9T0xXi1KWiFeVymXq5DKnA93zCOKXk+5w5vczczAy9zhFpnhGNx+xubSLyDGlNET4KSZ7luByX30wh8KuEwFUeQgoc6aAMBNplemaO/SSmVNZM+yU2O32mvWl0pc7+YEzuWHA1oU1BQNV3qWQpjs1pzrWI65JuZ5fM5shc4EaSaGCYmV1gZmYeIRyGY01mLKVanSDwef/1t5kv11luzXF42CYdDPGVJMsjhsMRytEE9Qpu1cdZnGLLxLx64wZv3jsk1h4EDaIkRIscDwVJhiIDkR2Xyi3CmMIw/BjT8P+mcBeiWDSVkYWxLDrgPvZ0/InBPhQc9YVqLghsLhDy90/P5EUjxjEmtMgXZCBURuBBv9smV5fBr1POXRQJc1WfbG+EsjnYEolMQKdYaYgtaFs0TGlHEMYj3Iri8acvcP/BHgaBNdCPEzIsUkbMukeIcsTQCMKsxpwqUzaSURoxUJDouJAfRxLZiGl/il7YY3IUIxwBZETZiHymwicvPM77669DZ4yhRi40RhlUPkImxcOaKIklJUojtFPBFTmxmdAoaapuzCC8w+b6AKsVjz/xBfKxz+7N9zhXdZgMPPx6oZY802gRmpBuZ4+D3hrLLz3KuSvLdFZXuPX225y9dJGRL7FK0lQe81OSgCFhd0I2HpFbSDJbdF6mMb5RuFYiycltRmpBo9BIHDyk1Mg0x7OGQEFFw1BZSg74gUO1ojkYj4jyEBWBk0CzVOL06Tl60R6jaJ/u7iqdjQ2CDCqVMoFbLhrTMpC+RSlBbvIiD2XBEQ6pyRHKxfXnGWkHt+qx6Gq+//VvMFzdYlY1EI7LtrEMq2WUyJE2omFS6jg0rIOXg6n6VK6e5frudXpJxJIqMx/7TI5i3NmTeKeWUFWfpHeIkD0aVQe/XuXayi1kQ/KJhcts7+wyjse4JU2YRkyikKCkKVfKqPo0WaXJzXbIt26ssjEKyfyAOAvxdY6fgcgVkcxJhS3CB9RxV2+BbQCwH0EAxEeLqEQBjrUYAcZm5EKAq8j/BJion5g2549GnueFHJaFPLfYvJDZTuKYNMuOAR6Sj7zF7HhbaU0SW8qVJsr1QTs0phoEZUWaZKhjwIgxBksEcgKOJZFglGQ8iYjHY568fIH5wKM8yvirX/wyX/v8l1jOPU6UyqSuYK+WsxOkdCYDhnmELLvkDhgt8Mo+mcixEnKT4Qcevucz7o2JBhENr4pOFCXf5/FPXKE+VSW1BdV5nhpMXnhGMs+RaY4yYFIKBKnUuNoncHw87VENypyYm0ESMdXyWFio8v7r3+HWzXcpNT3yJgzFkGHUxcYjnHiMGHRJ+20+9enncTyfOx/eoFZp8uwzzzIcdIijDvNTLq1mhYO9Xba3NkmTFCkFucmwWYbmuBEmL5JZApBSYawly0yBVrE5xuTFApbnhVR8mjLsdXCkpV4JGHQPGY8GDEcDOt0jXF+zdGqROBqTdXscrq2TjCYE2qNWqROUApTWeJ6H67hooY972wrG4txkRd5ISjIslUqFWuBzuLHN1v2HOKrQ2QgFJDIvdB2zkADLjOfSdBx8IbCeYmZ5ntbCNP3uiNFhxoX5U5RlgJWa6ZOLzJyYxa26hERIXzA9N0UaThi3O0yXK2w8XCFNQhxPE2cpRoJXCig3mtSaLcblgNfv3+d33vwh7V6XQMN0ucSVM6d5+vIFAl+TJDEfhQRWfBQo/HEI6OL+Z9YW/w8smc0I04gwCT/2fPyJ8BT+4KUW1k/8eMvY4nYICdYWIhe5ydHeR22cBSJMaY+trQPqjVlSii7AxvQUle6A3niAKyxIMAaEPhbNEJIMRZQa9vf7HG4ccGVpmaevPMLXv/4Ou9fv8R/9wlf4+YvnqZwoszLZ4NW1dxi0j+gPegibICsSr16FfhebGTzPgTQFIIwn+K4mzTRRL6JcKTMlq2hrybKYylSNHXGIJ3IUkiyTCK1xdE6apTieQy4UYSboxxlGx1SCasH/P4yZ0j7TtRIXzp4gjdb55tf/EY8uzWHHMYNGikosMkvQ0ZhpX7H34Yc89enLLJ9ZZrjdZntzl8cffxLP89h6eBOZHTDXSpGhIE8z4igiSzJwweY55DkOxxgUa47zWIUHZ63B5JBLQ5ZloCzSCqyFPEkRWtM5aBO096mUfOqBS+9wD5FJpvMGSIMQBpKYaHuPg3urVITGc6BSqWAA7TlkcYKwhdCwchykLMBuaW6xSBzPw6n4lCslNu/c47Xf/RY+EjfwGdqEoTBkwqJMRkBOSzo0lEJkWUGFN1NB1BwG/S6MMhbcCldOXUQkgtrsDDPLS/gNvzj3sqJZbSGSkKODA8pGEO0cIrTCL1cYZxHK8TDKwfF8qs0Wh0ddfu/mKlu9Ac1mjblKQKneRAUl5pZOsrq+ThpnRUevMaCKUODHVkEeV+n+wPiDm/ZYoAZMwUUiCmFm7X382fgTYRSEEAgpyLIcJRXa0cfVgsKFN/Bjn8bYvIARm+y4iUMXWVhr2d/pMD/Xp1VzUSojKLnUWlV0e0CWRceZWIUiK4yJkWTCRztlOrtj7n+wwmdPPcaf/+pX2OqEfPcHb5Md7PKLX/oCyfCQJT/laxd/itGpCd35Ab3tDgerB6AkZe3SH/VRGKTNyTUkaUSlVCIQDvFkTHIQUq765KOYlZX7TEhwmz5mYpHWYFNLXuT6SWSOSWBqfg6NR3PpLMkopZvFtNwqNo3IQ8PphQVkYHn1O7/G2fM1mhWXgYnQDRenJ3FTSSlXaFti4/4Gn/+lv4BUOb3eIa4jmJ9pcbDzkPFgBxPt4KuELAehLGkcY02GsQW7kgM4BqQxCHmMXBUfwXMFQqmCWUkalLUYVGHApSZwXIZ5yrDbpto8yamFBqNBiOuWkZ5B6gxjQnSeMFhdo7u2gWcs9XodPyiT5DF59hEQSiLEccJNCURuCxZo7eL6AZWpGlE84Y3Xvk/S6dFsNUnIyTyXWEhSE1KS0FSahhSUkpQ4izG+R9CqMTATwq0+06Uyp5ZOUFdVklxQW5zFnasRpX1iFTO1NE1JGDZv3WF3a4usN6JWrpBVXUZxBL7HJMsIyg0qzWnWd/a5dWeFCZonL1xgYX6GUTgkyg2xFmyur7OxtUeY53Csri7h9w1CMVsQfygUKIzARx53DmAhP/ZatQdOxQWSf/2Nf+T4iTAK8JEbWAi7mCRDqaJN0wiJ8+M7UzT6CFVs5Xl+XLU41n3o9dla22DqE+dxfIkRY2qNMuWKZhzmuNojSjMyq9FocjyU8PCVxzAacf/aLX4rlTz3qZ/i7/93/zUPf+E+b3z393j9R6/x5PnT6GiMvSdozs8S1JtcePw0N7M73L92j9lKE6nKHO1t4wearKyxAqRWBPgkxiceJXiuT7fX4dadO0QmxW2WyS2YKEFRME+FyhDbjMzkPPnYIyjPZ2ppmZLwONrcQ0aG1CRsrR1QdR2i+JBG1eXco+dIhxOGh31ma9OIowFO7pJPFMZx8WYaNE8sQzJkf+seSk3odB7y+uvfoeyOKHkhw/1Nsm6N6cUGcTgp6t9pis1SHCXwDCh7TMN2DG2XCIQoyqlQhE5SC4QqjnIQaCmRWYyNx9RLCs+JsFWoTZUwYoyxI2anSvQqDvboCNvv0yhVaDVbhUozkGQpgVI4jia3OQKFNSC1hAyk61Kp19Ge4v7Na3T39piaahRcGq4mERYrBL4U1DQ0tIMfZ5CmKHJyI9CBpt3d5ajb49TCKc6cOIcSHvOn5mmdWiJ1JWEvpjxdZbpS5uD+Cjdv3qA36DFbbeL4JULXkFhNKqC2sEBiFe/dvsf2dhvP97lQmcKPYLy1TypzUkdx0B9xv91nZzQi1QUIECgShEoUrrIs7sMfRkEX8Hp53EyipCQzBgv4JR8diCJR/DHHT4RRsNaitKRaKhPFCZPR+BjnII6bjIqHTcqCv9FiC4JWA3mWo5VBCaj6mu7+DlF0hunZFqPhkGazytxci/W1NibNca0C6ROnoIVAyJh82MdEEX1jWd9/wMGv3OSTpx/l0aef4hc+/1lu3bnN4LDNY+cuICdjBu0ed969zVs31pCOx0KzRTiKcDNFya8wSkYYCVGQURUFLNz3fGwGMpNk45yOGSCFj/QE2gPlKogyLIZMFAmmwSjCa9UpL8zyj37z13nuypM8dfoCQawZT8rEScLgMOSZn3mR/4e6Nw22NL/r+z7/7dnOfvfb62w9+2gZaUYL2gAhkDFlV+xA5BQxRXBsp1zJG+eFE1dloVKVSsAuJy6CKRMoo0ACFmAUkBgJpNFIM4xmn+llumd6vd1993vPfp7lv+TFc7pnNGAjYXCJf1Xf2+ec5yz3Oc/ze37Ld3n6hSfZPdih16rn/vfcuc7O3iV0lSJcjGl0CFIRJw2GO5fZvPIKrazNq689wXMv/h4f+8AjCHLGg30SmzEeDpiMRwhqD0fhLEZq4tv9bFHjA6izO/AEWYeFeiBUY+mEMLWAjrOU4xFq3Cc2lsF4A4thrbtGkU+QYoIoByQ+Z7S9RTke02tmZM0muctxft7o1JLIRJS2xAqJmjuEKR3R7nbJ2i2ubW5y4fxZep0m2mhmCgodyL1DounIhJ4SyMqRlzmm7lQidEAEz6g/ZTwsOfK+OxE6JWq26B1fJTvSZmhn4AyL3QWGVzd48st/wPbGDRaX1wgm4cA6+s7hooTWwiJbwwmvnbvIdFax0OsRmwQ7nTHOZyRLPWZoruwf8sbhAbvOYrUCXWeLhLpUC77uMP677N+EqDMLZ0ukjjBGE2lDHEd12fwXjWj8i1hSKuI4BiGpigJkAFH3FNScOy8l8wIq1IQiIXCuIFg/byRWTGaO869fYm39PTV9WgiWeh22Nw6YTS1pnFAQg/bgxqgi5+FTa3z0M99PNaxQwVFdPOT0089w4cIrrHRWOXXXfchSsnd1m/sevJ+19TvIOmtcG1ieP3eJUe45kjRZazaxScLM5/iJJW9YRtMpzVaCiiIUDm8txsdI3WA0q4iVJMsi9EwQCkvuS1JliNCgYfv6dY4td9ibHHLl5hu0yzHH2kdJ28vcc+K9NE60Weh0idSrvPzqqzSbCe+++2EWZj2un72ItQIfAo1mzEIrxownPP/NJ+nvXUR121y/eYHlVUF3QbO9XVLknthZ9rZ3mIxHeKPxziJCqLUPa9fSmpIbHMF5goYwnxSJ+Y8gavQpAaSXaAGNJGHUP2Dz5iXKcEiUdVBJSTvRdBqC8/sbFIMdDl1CVczI2j2EMUjpweUkSUI0f4P6QlqXEB7wlaPd7WE9vH7uNInSdNpNJljKWDARDueh4QS9OCazgrIYUTqHjyUqUiijiYKgmzTp9lYxaZu0vcji+jGytS65KXFxYLV3B+XWHl/94pd4/eXz3LG+hkgzdquKKtEUJiHJMs7d2Obsm1eRKHq9Hk4ZZlUgxBWdlVX6UcaZm5uc39on1wLi2gzn1n5UHv4kFEIIf7zdKKXA+1DrnHqHVjFaa7BgrWPepfi21ndJUKgzgcFwRHCuzpSUqjMEX3+ZUNeNStUGF0YqppMZOM/q6hIL3RQRcja3R1x68yqr6w3uv7eDNNDKEjrNJsVkXJ94ukRGHulyfvhT9/OTn/mr2OmE869fYjxyLD/8KH6lxeYzZwmTMfvPn0YUgn1b8vXXr7F4fJ3O0XU++qnvo/fQA3z5d79GubdXoy0zTZHEiKmdOwNXGKtBlthQIYIk0glBJwgDVWExRpJEhlLnzKqKhBiJIEqa3HzzEsunjrOy3GUy3KPvKtzBhKg9xW1nqM0G3Z0GR4/fz43d66Aq1u4+xhvPnkWtGkyaoiKFJaehDK994fd4/sxXue/O41y+/AqBCSvrMbNywNLiEuPWAX7fU+ZTlFRUtxqKUKNEXJ2WCiERbx+P+1sbKYSoM55buDWtJd55tAJhLXt7m5BWBF0ymh7Qy5qMhnukOrC22MbeLEgbMUsrK8RxjM0D0moSo4i8Jy8cUiiiKEZqibUVWTMjbWScPXeOne1tlnvdemISKUahIheCWGjSoEmsIhQ1mE1EMTYKCCMwkYS85NjyUbK1dbJ2j+UTJ2gvLSKaEV4XZM0GvlQ88cQTPPPk0zy8tkan0WGvskyNZhJrAhFvXrnBpSs30MaQNTIKF5BKkaUJxaJhK/a8evUylw+m2GYTcChytACv5kEvgPCSb80RAt7/8ZzB+7qvo1WdydmiwlqLxeEEVN+BIcx3RVAQQLCCogChU7ydYZIRSswLISvgVoo6LzWqnYITRyN+/Mc/wCe//1E6zTYbN24yq0a8eH6LzZ3rtExFZXMiE+i1E/o7gWBbiKxENEqKgSCoLrMKfvXXfo2Ll/oM+tBZeo17TtzFbCliSzlCJmjaGF3GxFFKd2GJg1AyunHAA80mSx95L6++eIbt/QOiok2cNcgSz25eYsuYvCxoNjRSKmSwJLMELwUNIzhkxl4M3aUMO8tolDDWHseMVHrCbEzXKx576DGe/vpXuKmnrMUZaXlAVOSoQcHO61vkazGnFk+xdt8ibbHM4f6I7rFlis0ZylUkNrDcXuJLv/0FVh/LaHWnzNihtXoXOwcHuO1LnEqPkYYuh5MRRVVRBYfzAeU8iRSoIHDCYHwFlcULC4S5p6eoZ+lVASKgjUYLiZYgMAifYLRjYTFDNSU38x2i1LHYg2p0yMbWBqvtZfJpzOTGAQsrq6hujzQy2HIEriTOEpAG7SZUPhDFDSwSgWX92FEmec6ly2/SjgxCwMhDiYBC0FAQYzGmQFiPrwqCtsSRRtiSWGliDNIkrC+toTtdmuuLxCtt3EILiyDNFtBB8MV/9Vm++P98gSOtVYrmMjeDYA/I04SBLdm+dJWd7T3SJCFNGgRl8GlMmaWIRpNBZDl/9QpXN/v1fgoFznmkViAlys17NQKCgjDXV5T+1mQu+tYTaB43xLzUvi3LaG8916O9wv5Fch/+vFcIAusShM6RaoQ0FaESBFVrNgbF7SYKBCpbkXUV/80//ru86z0dvvaVL3L18ibbByXvfvQU9586xfvf+yBSFRTljMm4QJSnufLGeSaTGciIWGQI2+bpP3id733/x4ijU8yK81Q+4uz5MRcvvsZS0mZBJky1oS0Vi80mbZPyzTfOccd77uWjH36MMM0p11f58HvfzWtnL3Lu4hU2NrcZ+4rBeIyNJFHSIY0C7SyjmgwpbIXU4Kqc4CvGQGJSok6Km1RIHSirCT6URFqRj0c89v7Hef7lFxi4kqiYsahyImlIhSMTktmNAw53xowP9ti5skmYGporC9jujESBGAfQgrUjx9DRkN2D6/RWOogk46B/laQlMBLspGIymlCzTUAEh3QefWv8e1sEdJ7mzpGogQB+XjLUrW+QnoDEB0lZAaaucaNGynBvwuLqOgtLGWcunWa6OWH9nndzMJyhlWZ1fZ2o2USE2shEaYFOIoKO0UZBUSJ1gq8CC90FeovLfOMbXyMvc9ppUn+GSCIjjSkc3jqU8PXMP3hCKIF6xC3KgswYlIzQ7TZmZYGFE0dpnzyO7nVwQpMmLbSK+dy/+CV+/ed+mV6ziep0OERQCpiZmO2DEW9sXMMOC5qNBsYkyCTBCYVIY6LFBQbFjFfOX2KvP6SRRjjAWQtCEhz4IFHKcwvgO6+S3rHEH791C9AUahKhkJKatyuR/tb2f4mCgsdRMSZKAtaNwWsSvQCUICqEcAhpUULjvGCaV/zg3/oQJx+9g5/+33+WM69dYTaFiYWvX9xnPXqWbjPh+Il17r33OPfec4If/iuf4MH77+Tpp1/mwvk+h9s7NKKUve2Sb37tNGtrDzKd3GRWNFFxxsTvMyvG7ImCaiVjimN7sMmdrVXiYx1+83f+P65dvsRDd9zJaHePxXaX9zx4B8fXeuwPR9wYTfmjP3qWne0+UfA09QK5c6QmYTodMz4cI9qCSCtmeUGhC9JWgu1YdOFoRoaGMqS9Lv1hn1a7zcPveZRvPvskSkeEMMRFDseYjokxyhFXjvzGiKJfMFGe68M9Th29k8VWi+0rNxiWlqTXZvPgIq3FgqMnTnD1ZsloNEA02hgEFAVVUeBlnZw57+fcgnn5EETd+IJvgb6FUAcMFRRC1WgbXys2EqTGATrS6MjQaC1QVhlBxly9dh7np0zHAw53+uxt9TmSLNNbXUOZDFFMidpNkBU6yyi8xBsLkcREDYyX9LoL7Oxuce3qFbIkJk50PT/F432Fp0Jg57M6h/cWsAghsGVJK8lIsxYVErO0iDi+THx8HbPcI2q0MDqlmFZ89l/8Sz77f/4yx3pNOssrDAHrC5zQ7B0M2djcoqg8adpAJgkqa+CkRiQxUbvJ9nTAxasb7I2G6DTCKol39ei2Fi/+zhSSbu/7d1QTnoCcf0c1Uvg7W98VQWFhOeIHP3OcZ796nRuXmnTbR4hUQWkdIRQI6Ql4KmcpS1g/0uVv/8T38TP/5J/x7HNXEDJGZxmdBcVo2sDa9L4AACAASURBVOfCzCKmOa/uzfidZy9igqwRZhLWj7Z55CNHeNeD93Lq5F1EoWJwcIPYHPD9H2nw4nM32Nu29McwqAr2S8WbF95g0Szw+D2PsLXXw4iChz78PWxc3+Tq115gMU7oRYdcPH+NWEo6jSb3Z4usP/Y9VDqwsb/JucvnyZZSessdVpOENE7ol2MyEyNjxTAf4QzExzPafYMsJUQVC0cX2R31eeL3nuBv/Zd/n4uDPtcuX2ToRwwZ09OBSaVYlE0avkG7aFLkEWNpOXCWp6+co6kjHnngfrqLPTrHMnZfP4N1BXfceTd/+PQTjEY5tpMwsQMYWbI4xgaLtRbhKoxSpFITUfsZeOuQQtZ8FCVwziOkIDhwAYyruSpR0kA3WgSR0ep0KBqQy5K01WFSRAQZMxjvcs/dd6F6p7jy/DZ3nniQpZVVRNoCabBTj0kXaKseMmmyO5oxLfdoxw0Wkg6qDLz67HO8/NzzNDNDmhiECeR2SjAKX2liPMJbogCNoMCVBBxSxzSyjFbawgqJ6Szy+I/8Nfxag+l4yuXzV3j1xdd46omvcPGV83SiBiebi1Sx4Npkgm80uHHY53BvSCwTFpqLGAu6ZdBJwkwFVBpzOJtw5cKrHI5zTKIIiaoh2W/DFogaEnabzxCoMxop5Dt+/8lhw99qRc5TC/+21qSQ4jsKDt8VQWE0LJgdGr7vEx/m+eZpLpy7RKoNiICJonlXFYT05IXlnnuOMt4Zs3Fhi1SkdbMozJgOc4QHqxTOCmxQIBW5BZ1EVK7g/LUR5zeG/JsvXWC50+CutR733bHC3cc6fM+HPs6nPtKh3x+ztb/Pm9ducv7NLTY3BuxePeC1l7/Ow8fuYWk5I1TQWGgxcZ7DcYGQmqBixrOCvBhSbR7SXj/C1HnuOHKU3fEhZ65dI5eBXqtBq5miK1vL0BmDkDCuZqh2jGxEeF1iWjGyqSn2BmxeeIN8WPDQw+9nc3OLvJqwNRpBKyGIgBQFzklSF7AeKm9xMiCM5GCwzze+8RR3n7qbDz7wKXqNReJ2zHiYs7M9wDlFUcBtIzgBkVAEX0KAVGlSrWvtBwJS1ErbnlCPjUOYH6wKQs1klFLgnKPykKUJKs6QkUXHoI0gS6iFbSeO5FiHPDhMnLCwskzUzhBZhhMClabIvCSONBZDHBlsXtFudOiYFhvn3uTC+deR0tcu3JEgryxVcEhp0ErgqoC0kBhNLBSeEg8kkaGRNUAITBRRCs/v/87vUBjY3t6l3x9x5eIGl85dIlMxnaRJCIJR6VCNjIPDIYeHfXSS0M6ahBKazYxSSYbFlEavzc29HW7s7jKzlsTUAkHWzZOsee+vHr+7eb2m5pheUYv1zk9uKQVS1Nqibx8+hHlG9i138p0FgXeu74qgILzkma++yN33nuAjHzvJ8krOy8/t4T1UFhQxgrqEQFoe+8DDbFzbZuvmmCjrUBaWENla3DRotIjBuXpWPgfoh6AQKsGkyRz5BDv7gc0bWzz3wnUWWrDSeY47j53ixF0pC2sF971vmcc/cYyGaVCNSt44vcErL5xhf7eP3M9IowbNXodKaw7zKaWtwVVFsMQuJ9+5jlMSOUu5b20dFUr2Bn02hmM6sybtlS6SihA8UjpGsxlhNqYVJTgBLgrIZkSaxhTXB3z+Vz9Heuok3d4S+dAxnI4Jo5IpkkI5ZsKxqOvU2AYYzaaEElpxhBzPuHzmDB/9q5/ksXc9jo22KWODkobKwmRqqZRA65Q4KgGHKwXSS1IVESuJdQ5vSyQaIepZugCEUAip5ijD2vKPEJgVFSqyRItNsnabYGY4E1DK0W1DsDnjsac/sMz6OatHjtBe6uAjTUg0lbckWUSINCqOqKwmaWREhWCpvUA4nHHu9Bmm4wm9douAw92eeEQ4FGVRZzyRqt28tANPBCoQmQQpJLYsUUkDi+PMyy+iVUyQips3trn0xhUCEDUb5KrWJfDaUOQFB/t9ElGDs6bDPo00YViUzCrori7RnwwZjIZoDS1dY2wEAifC7YBZg0Et85EOyDD3PJFzCL+vg8N8PDkXLv0TTiIxN2SdixcqMfcR+ktaPhw7tsx/8p8/xHMvnOPpb3yTBx68kw9/osvzf3SF4UFFYpoIDDbvE0dw4uQS33j6KcYFZHEApfA+oH0DbJdEeFyY4ZmBLHEhIBQYk9T23LZCoxGxIcQGJPSDpd8fc+HwVaoXBVoFUulZ6yScXOtyfH2JBx46xY/85F/HjWeM3rzJ6+evsXNzD6s1TitmtiSWmpbRDGxO7AIhBzs8JGs1eGhhnaKzwNXJDsN8RnU4ImqZOQotQqmK8XTMrNEkbkZMGJNT0mg0SHXO5pWbTEdj5uY/xM0u/cM+VfB1ZzlyiDDDFoEpKSRz9ypXstZsYmzMb//iL/Pez3yEBz90Ny9fvoA2Cc4LJjNLmYKQCcaMIQiUEAgEkart4TzVW8QcF0DXBsBBhttpqxQKJWpBFpFoojRFGUOSNXBaMmaKFBWNODAZHVKVMTc3+yyoLovLK8SNGJ8YCqnQApJYoZIGkkBlNcJr2lGLRrPNhdfe5MbNGwgCriprQVMhkTpBS03pAra0NegNjSss1tWCvUkjJTJJ7TmiNUJCaXOkEpgguXLtOqfPXcMpaPfalEZTSYEygnFesXn9BlUeaGSCZpzQXV+ls9xBKM27H/0Id9xzF//Lz/6vJKmhm6UEaxEuECmJmBPICmcpKkduLYW1VDZQOUcVmJsLg9IgdY3RCXMC2NtbCBKBVhKEvA0qc/Nxpve1qth34iMJf3bfh/8XuG++SRfohxDeM1d9Pgecnz/2RyGEv/envcdgMOSO40f43o+/m52dTZ766pPIkPPBh3qcPjPi4KBCxJqgFVEGLoULFzfRTUFpKgiByBksBis9sarA1loJci5VhYMgQ51N1AYRBGqPrSBARBFWO5wF0gZ5MJTOMh1aLu/vIl7bJPryaRaXOzx0z3E+8egJPv34pxlsb3Pl9Te5fmGTGxdHzApBZRPKKOGYVzSCIyks6nBQS7YrwXrUpIgbzKwln3m8AGcEcWbwhWWzGtOJmyQBJtv7xEWC1lOq4pDZfoFVFVWu0DolSgLFJKdfFWgXsFUNVhmFMXGlacQNxrZk241Zb7aZXbnK13/hN4mGP4BbjIjjJrmX5FWCCAmxGVI2I6TS+GKG0RGRjhHegdFUwiHy+gDXQiKlIczTYR8qhAqgdQ2xNQ2SqEWIUkSnSZQZTDXAJx7fhHw0ok2GLhyNXpOQxMwijUlSFB5tQDYEqA5VVcHEEZfUCt3Tiv7mTVrUJ7XzIE2ENAaLorCOshbqQAlZQyikxrsKLyCNYkLlyKclppkxrnJ2vecwQH9jk6ubW0gh6Cx0kGnKzAVK55gVfUIO9z/yMO9618Mcv+MEDzz4ACfvPEl7oUsUx+zvTfnsr/4qb16+iTYSPTNUVTHH3TmEjsmSmEaW0G4aehrS2NRoVmcJo4pqVjLJCyZFydharJpnZMGjXI0aDfMhD+ga8SjE/O+cM1VxcwZg+I740H8m34cQwo/d+r8Q4meBwdu2vxhCeM+3/xGg35/xP//0/81dd67wwQ+e5NOf/CSz/UPOv7FDObvGa+UWEytqcxhlOHtmm9GgpokW5Yw0jsBK0BarZ0hh8doSrMCjcfP2rKLmIzhbQ0hRoKVEMJ97eomSHkxJFXKCBpkIhNMopygdbPbHbHzjNZ56/jT3nuzygYdO8uh9J3nXe97FztUBr754kauX99ieBZo4lIJmopGFhbIi4EmKhBAqfAJax0zyksJaTMMQIs/WeEQwkhNZhi+Lmg4rx1Q+RpVgjEaGiNkkx3mwwVEUOaGQ5NKAFpRK40uQrqy1LbUlzaecaPW4cW2T3/rnv8nxT72P1cYCp4tLWKlJsg5VNkR5DcogEkPsDJHU5OMCKyrKyGN0QAhPoJbFEyGghaDCEagzMh8UsU7QUUzUbBItdAjplKTM8AZkoqj2R7RSQzdOSZMUk9W9BN1uYYQnijyhAeNQUFmD0hVp0KRxg4uvvMDm5cskDhwSZ2IwmoCmnF+Bvaw5F1rWAjBKgok1wtRu1LaqiJstXCNhc3jIQSJ4Y2eP/esTKgvtVkxA0R+MKLSivdTj/e9+lI9/6Ht56KGHOHrsGKbZBOfwRc7hYMDXvvIVfuU3PsfXnnqWJEkoXWCWgxMajMfW3GZGwxlyOEZJiDW0M0MWa5IkZrHRoNlbRGhJYQNbgwEHgzGD4QTvLN7XgudSg1SaQMA6BwicBD+3XRQ4wu1exbd/Pv57+T6ImpTwo8D3fftv+Se9kCR3GWfPDzh99gV+9/Nn+dhjp/joJx7nnkceZPnJJ/mj5y6zvQv5juBL//oVCp/jrSJODbhQu0ApYG7drpRAmJqlh6sppSF4lK4dM7yrd5Yta0UbCbWykKgVa3Tw+NIijEELhTYCoWtWtDYJOS1eujjilTMvs9h4jYfuOsoH3/de3v1XPk5y9iIv/+EF9kvPlIqlWJMJRZw7hK2QlKRaYirPZHtIFHuqxGN9gswSMiUY94cMpMRkDaRJ0J0G01EBdoazkiA8VTXDeQsmUIgSX5bkASIdEUVtbIBZYUniCKEktijxzrO+vsZoZ5NXvvwMax98gOUQYayk0Wmz30jIPFgvaDUbpJUgcRJ0DXcuSofwoZZGE76GQHtuk58UtWN4CNQGsLEhbiXoNMJnnkbWxVZ9/MRRjSyd5UVa7QUWFpdodjuYVhubaYyASEOILMo5KhXAaOK0SbEx4tmvPMloY4Msl0RJjEhirBTMigIbAlrWKl2+qohjRTQndimtibUh2AKyhNBpculgj+2y5DAPXL05JihoJ7rWaJjlfN/HP8YP/ejfYOXeu+l2e3QaK7UqagXj8xtM+0NOnzvLl776FV45d4YXNi/hg0KriKkt0LFBm4jcTnHB1lz/UA9s/VyXoiodux4kY94w+wilSbSmlTU4ubzKem8ZX1WMq4pBlTPsDxmMx+S5xWPRsgb4BSFxRoIXt4HNEkWNS/0Pg1P4KLAdQnjjbffdKYR4CRgC/ziE8NSf+ipCIuIuVV4SJ4brWyN+47de4pmXLvPJTz/Ip37oI9z/wH187tef4fKFCVuXNoh7JRZHkjawZQEOYkMtsYZHBIkSoKRCCkFZebwLKA1C3mL41UIuOHDCzptl4K1FB0AqZICqKrBKoZWmCg4bDAWKZnMJmXbYn434g1dv8PXT17j7riUWWl36kxH7QtIUiomE5ciwKCOSyiDcDMqK1HtaWjOeWXatZzuf4JdadJpN+rOKzf1DKgmrWRPRTnGjMdrOCJXBaYmSCV4UuAAyywhGMp3klM6RzHKkMEhpKIoSX1iaSYIPGhUsdy4tM7mxwe7zF1hKFVJI4kaCb0YY66imOZ2FFlml0aMSVVWIwlJZReVLjJ6j6vzcqicE4sQQmxgjDUGCiSPSboOo20A0I3QT4pCxO9thOrEUFZisQ6O3TGN1BZO20O0GubbzgzhghQNjIBKgFKQxG6e/yWhzn6ZJ0NaikoRKClxVEcq6JJRCEFyFco7YC0QoEcHe6uUjTEyVRlwd7HNhOGCC4WBYN2aRtXLyye4SD516gA88/CiNmSUuKjomoty+yXBrl2tnL/Ly099kMByzsbfHuZtXuH5wQCVqdGIeLCiJJ1D6Cjtvgioj8HP4skTeihEIKXDBkQcFQjMpLPvjfQ6392mohFYS01rt0em2aDYbrDlPmZd1gDgck+cFVoENoi7j5ihTpKjtEv4DBYXPAL/2ttubwIkQwr4Q4n3AbwshHgohDN/5xLebwQgjyJlB5Cmkx7QUIk+4cqPPL/3S07zyylV++Ie+l5/6yR/jt37jd3n5pRvYSpE0DN7WfH1VWQSuDgRK4mzAzRl8UqpaeDQEgvc1goz6SuZxtzu0StbRNdhawEVqOe8Oz1Wb8GA0zpZoOcR6RSDCpBmxluDGnL+2B9UenSoit57FrMs0FkyKitwr1qKMlvC19mAViJylZyK809zsH7Lb36e12qLZa7Mz3OXG3gCddfHBgQqIqsQXkspKMAYXAtYHgmrivaYyqvbLnFvESwXBK1ywTEOOjxvI4Mh8wtG4xcbumEYvYWGpQ6+zwHYjQVUVRkUcWT1OPAtMruyQFzllpUmEJEoSpKzfhxAQSiGkIE4SojgmlglCS+JeE93LMIsZZrEBkULamMObM/bGJUJnqM4SurcI7RaFMfi0VuwmSIKswT1CSYx2xLGhmJVcfuFVGigaWlHUUiJU3s+1Nurjy1mHcpY0FshgcbZAmogQGWwcYbVmczLlXH/EjgvsDSeM+yXLWrHSSrlv5TgPrJ+kEyWc/72vcelwC99J+MD3foxmM+KlZ17g0tkLxEkDlyS8vrfLhcMDJsLWKTye0nqkqPsd1lUgAtKoOU/B1w1BBErI2xoItbd6TQUXQSONJrcVM5ezO8phNEAqaKYJi70uK4uLHD+yhgyC3e0drt7c5cZ0QigsOtForbD21izj21t/5qAghNDAfwS879Z9c7u4Yv7/F4QQF4F7qV2kvmV9ixlMQ4WgxkBZS6RHGl8ZtO4SrOepJ2+ycfnz/Md/82P86Gc+Ttp+gqee2cM7gStzpIxJ44S8HEMskVGEsxVS1Dp+AHEU4b3DVnVAuKVzJ4W8Pcaaf7D5eCgQvJhPimqapqfO/Iz0KDEEIQguovKGyBgQGZqcECoKYGQdrpzhPSAN3lnyomIlkiy1MrwuceMCKkcjTjjS7JGPh+xf7VMUlrjZYjyZcfPmHovNHlYKlPbgA74KWFdSUVIGj0QRnCb4BESF1Y4qSGbO40LAeMsMmFIQK09ZVbRkQk94ru8NUaslDTSdNGN/vE+60ubofScpd0aU/SGiaoIOUEiKsqibWLoGPgchkEahk5g4jUmTNkSKaLFNvNZBLWeIXor3FmkTtg4O2elPWescJW2vkPWWIU4QSQRaESFruTXqhpqKPFp6Mil588xZhjd20FVdFmodMXN1I66yteYGQuGsQxLw1lJSZxMiazJTmr6DvcMhV6czNgpL30JZOmKhObW4yv0rK6y3e5hhzvb+DfrDIal2HO4d8vsXfwWS+nhJWm1oSs5vbnDuYMiBgVzNH57rG/i53oQU83Peu7dO0LnAqg9+fkDWB6Vxc80EIQgqYONaFsgJR+wh8pJZkXP1xhY3bm6TxTELvQ7H19d49H3rrB30uXLtGnt7A2Qi0Np8RwOIf59M4ZPA6yGE67fuEEIsAwchBCeEuIva9+HSn/ZCgkAkLE4EhHR4LyBJ8VbjHSwsr7Kze8jP/8vP8yM/8hCf+vQnQL/Gl79+HpUKtHTk44q0lzB2M/JZXpcCIWCtRQmFSTRVBd7ZtzKH+jOjtKonAK7uKYg5Hfdb8aOihqDOxz1GMNfBK3CioCKZbydBRogkIK0hLy3DKqeZKRrthK3ZhOEkZxInHE+Tep6c14y2TpSwoiwzJ9m/PibuxsSdhNlhycjOUAKqUMuyYaeMxiOskgShiCKDllDhsd5RGkGEpAo1q846Sw51tiBrgowUCZGO0LnmxvkNzj//MsuLbTYNLKz3iFfaDPpDZtIRlMer+uSPVEyc1TN+72rauoli4jgiS1OStI2PNWKxQbLWhV5GaGoMGaqcMRxMcDm0jyyy3FljsbOCbmSozODmrk9KR3hbIaOMc6dfZLXXodHtcv3pV2Fa0E4bWJdTqkDuLBZR1+8erJNUtkIpV+tb6hhnYsYeBkXBXlmxO8vZHlsGDrxQpCpisRGzECeEUcFGfwNfWaoqp3IV1gWcLUgTzVRDe2mRIZIXL17j+rRgrEXtLg7gytvExhqlOJewE4Lb80KgRi/VylVirg2g5iKscv7oraaim2NtrJ+DzJQgUI8vR2XOcCtnY2ePhU6LU3fcwQfe/SA3d7Z4480N8rxEqm//VP8z+T6EEH6R2l36196x+ceA/0kIYed/098LIRz8qe8RQIUEERyhnr1Q4QmqRGeC0WxE1hS4KuI3fv0Me9uGRx+/m0p5vvH8G+Rlzlp3if64j2kacl/UYxlXAz18cFTzDGE+O5s3YRzylsITHudqd2uUwvlACB4C9egtzCc7IYBTBJcQlMXJOoGtfI5QAhlqhp71JYoYHEzzGbtlSaO5TG+tS3kw4Wq/j/cl69LQjKibmqWk4UB7TaIbzA4nVLOSrNtgYsdkzSbbB4dUrkDGkDRS4maXwXjG4HCfVESkSuFloBACZcH5gA4C6z0Tn1OGhEBOLBK0jrEWTGTIixmnn36Bj37ovahOzPJdR2iu9hjvj0gXOpSziqoq0SomMYJmq0kUxTWwJ45qCT0fSJMYE7UIzRTXS9C9DNGKKLUgNjHjScHW5jZiBuutFY6011CVId8esN/fYu9glyJ3DKYzrh/ucVBNOXP2Vf7BT/xtootb7Lx4gThI0ihhZixVkWMJ5NbXRrxe1qrd0mOMQmmJiDV969gcFxzawI4PHJaQo1FRQiY1DQFtBcV0xJ4syL2lVJZCWJy2uDwnNZpOr0t7aZnrh4c8d+kKO6XDxRFSpzAWtWBLXNVaijUAFKhdvkSQhPmEBm7FjTDHGwXk/LFKzWHOzJ9fKvQcAyKkwIoam6eUQutaVNd7Sx4c1WGfYu8VVo8ucuzkMVYe63Dp2k2u3/xTT8Pb68/q+0AI4Sf+hPs+B3zu237326uGcPqgkUHgPQRpccHWiK7IY4NERxFxSPnDp84yLLZ49LFHETLh2Wde48TJdfZO76MtSGXmSHJqEhWBvKyIIlMj0lyFs3WPQaFrObcgaou6WsmFGg42/zW/SZBIEbCA97oW1RSgqLDz/oUwNRko5BphNCCwztGvSi5u7nHHiXXuWm7hkRweTKhCQacJSQapMyQhJnI1aEcQMSs8N2+M8bEj7fZ59PFT3PvIvawdW+XoiWMsrqxysNfnledP869/5fPsXBvTaie4EOG8wwaPUgrrPCNr2XeWtoQSSyw9KtIkpUHGgYOdPufOXWD1E49w9Ohd6F6P+NiMxsEQVZRoPK4qEUrSaHVIGhlO1DoBSiu8syRpClGG6DQxSy3ihQ6ql1HY2rtzfDhCTQSPrN7H8WSZ80+/wGF/yMFwwN7ogP5oyHjqmPiKXTtlb9an12qw+fKbvPT1b5L2p7RQVLMcV9XE6UgZZkVed9lV7Q2hTEyhFBNfMhrkDDzsexg7OHRQoOqmqDa1D4eSKO2ZKM8w5MxCSeE8VbBoozGdJsura4gk45uXbvDm5g5TKQlJDMoQbI0REA6cezvqMMx/CkDggphDwt9K6G8pJtzimdm3P92DFgoV5qWGZy75Luc6pvUrBDRaS6TylCVcvrHH9mDI8RNHuO++ezh+Ysbv/eFL39bZ+F2BaAyiVgEO0t+uxfS86SLRVK6W75aJRKQOITzPv7RFf/ASH//o+zk8ucvO7gZ33n2Uy1evY1VUd6tvqYNQU3tLZ9GRBjTBOTwBISWl87eZf97XZYIMohYsnaP6fLhF39YEUSH0DBCYIPDBoL2uu8feEhwIEkpTy8cTDHYiKSYl9vIezSXHseQIstElL0cMxS5JUrHmEjpRh4YWjG2OQzOwFbaR8qHvf4S/8eM/wN339Si1YGu/z8znXB5cYHVlif/07/x17r3vDn72p3+BN9/Yp9dqIuacfC/AaYmzsGPrBmQca/AVsY7paMPhbEoFXLq2TXhtFfMDHUKriz7paHtLK5bYmwmHm1sUTpGtLtFaaDOpCiyewlVEOiFqtZhZgeqkNJeXiLttdCMisZb+petUmwM+ds/jtZDMSHD59dMcDqf0ZwUlgdGs5HBUUCaQYyn6MxY7K2ydvsqlM1c53mjRzhS2Kim9r5WhqkBVBWYBCu8obe0RWlaeGYqJVEzxTJ1n6i2uEsRKkkYSIx1GCnSkKDxMbUkwEOZy6XHcptVsE2cpu0XBpYsXuDYs8Sai12oxKMfktiQIO/c31cz5pPWatw/qHKBGHvp3NP3eiUTW7xgSyLmI663NFPVLKW41KmQ9zbABK1RdymhJXngGb1zh5u4Bp+5Z/7bPx++KoAB1uSVkmHPBmZdbtWClUgpkLdQqAsRaEzS88eYmcfQSJ44doT8Y0FxYYGdvh0HuCU7WO9PP0VyhhuYGW08otNY4/1YweGvVo5zabMbhVc0GDHOpq7d9YmAusFkzWLDW3oanauoD05aBSGhMonBB1A2ivYBr7tOxmlQrDC2KScE4MoTGlNLCwd6UXJV8z488wmf+7t9k5cQ6r71xgc/+5gtMyxk6EbRaKY0sYrd5yMHuhPc8/ij/48/8t/zcP/1FnnnyLG2tiVVSa3+SzFl5lsqBUXktm25ihNaEKpClKaNZyZlz53nsxg0eetedFEyJTiyiUkG1nuG3Y1wOrcVFvASbW0zSIDOa2MTExuAmJUvra6TdNrPRhO0z5zn/8muMtvfInCZLGhyWY/K85Nr1bcbWM8wrBvmUEvAmpbSeUTEgayYsLvTodjsYo9g92Ccnrr8nJLPSMskrxs5S4ClDqJtyHlxQWK0phWBsHWNX4SRkcUwWxcRRhFIAngqHwyKURAiFjDVZkpI1GkxnORsbNxkMBpgko7fYJWpmeK05vDHGOf+W2rj3hD9nN5V3HqLy3/X683JFUmsqCC/oD4e8+tofGwD+W9d3TVC4pe0lJPUPVWvTCahn4YLbjb8gBAJDnMDZ169jK8v999+DiCXrq4tUm0OKyhIICFXLxN+SxXZVhVcKoyVSKpyr2Xzqlvzb7c9T/wu+7q4DOO/qzOHWdqGez/tQm9bU6gEKqeZCmS7U3WapQYDSAuck24UgiDE+gmKmiGSPNFtjpCx5Z8TVG/u072zxX/2D/4wPf/r9PPni0/yzn/2/2Nkb0F04zt133UWzETGtclIVM8orzl+9xOHhiI986OP89z/zj/j5f/oLqVZavAAAIABJREFUfP6zTyFlSSNuQRBI7YgkiNLW0w1RQlVhtKGdNJmUDlEJJv0JX/38F7jj/Q/SXu4y8GNE0iC0S7JValSliiisRVeKJGsQRwkKRbvZYpGEvaubfO2J3+fy2TeY9ce0kiZH144SXMHW7ICd0YTJbEa/dExCoDIam2jQAicDeZkjDNxx8igL3TYheEyacLB3yM60IjKGorJMc0vpwBuwsnZbrFA1TFsqcuuYzHLy2hwEHSfEQhFpAcpTOFubDOHQSYQyMUKa2skpjtk6GHH95k2qqkJFcc2yVILKe8ajIaWrsQi3pNAC77js/3mst1+M/ri++7eu+cN+Tl9HKpQK2L9stnHAbQXggERIXzf24DbL0c/HhnJ+1Xe+rqGiILl2bQtrLfc/copOp0vWryirEaJ+IhDQc4co6yHg8EqgpKL2RXV1Haprooqc+1BoredX/7l8dpjTVG+1HN72ZQkl8baq8elK1VTXSN0uFl0AoQVCGkYzha+GZEktYtQ/nJKVDRprgv3pAR/4a4/xE3/nx9kaHPBf/3f/hK8/f4HmSsJd99yBajfZG40YjgJZOyFtCYyuvSY3+7s88dXf5/3vey//8H/4+yx3O3z2576AnU1oRGntrxFp8BYpag6IzScgNYmMGeQFCk1HSi488wqvfvFJPvRjP0i71YLFGBqeamAxeUALhVENTOmJ4wZp0iQ1DQb9Ief+4Ot84998hb2NPZbSiKVOj5ZqMDwYsjseUlrPbFZR+ECVRKA0wWiE0thgGRd9XAisLS9xdHmJSAgODvo0FxfY2t3n0DqMhGluAYFVgSA1BTVGR+smpYO8zJkWFUEpktggTYSKDJpQA9JshYgUIjUIHSMiQ9ARSdqh8nD1+ib7B0OSJCJut4l0jE4MQQt2Dw/ZHwyo5NwAI4S30vnvAFL8bZ0bb9dGCHWJ+2/dNjiw80zY1VmsNqa+6H6bAeu7JijU+zMwtw5+y3DWA8rjKo8Mvta2FaKGrHhqWmmIuLqxx/L6CgFHYmISnaO0qq84t4fE9U4NIdQpH3WfQEf6LfyCqNWP1XxUdPt54q3b9Q6vA4lQEiVl/ZrzP0UpSTAGOy3RUtRitEoRfE2+IpFMSsFmGdBRjE4iNva3iaPAT/2jH+cHf/jDfOGLT/HPf/5zbB9UpJ01oqhDfySY2i3KrE0WZYjYMKs81gcaSYTHsjvc58tPfolH3/tufuof/hf0euv8H//bv+IgH7FoNKPckilFmpiaYSckMii81mRJQjHNMVaSFoI//NXfwVOS3dXDtzxRDNKXxFWdEU2mQ6oyoJhxsPP/U/fusZbl2V3fZ/0ee+9zzr23blVXVXdPP+c9YxjwGLAxAcsJBGGCYyeOnSgQxVaUP0KixICEkVEiQgJCCFmyRGSEhJKQh4KDDHaARLYHPzA2NthYtpnxvGeqqrtet+7rvPbj9/ut/LF+51R1T4+7R8xY7S11V9U5957H3r+9fmt91/f7XZ/m9qdf4Zd/8WMsP/oK10Z4/uljrh8eMwyJuydnnDOx8gXNwsx1SBegCbjg8WIL2gNBlHkTuXlwSJMhpZHleoObz+FwwfL8nEYhEyhZSWrEpUnAS4PGwNj39P0EIsy6BdFFsmbKqAw+4TxICISuQ4InO2FCaboZD84vOT0/oxSYXTvCu4D6wOQ8m37gcnXOqu8Z0T3hbWexLqY0eZO1/rqb+g3cmZ88yhOtcVe5DV/8NrIglbN5MFACKRVCiPwWGwZTd9968+ruHlaDe5wA3lE7hLbjBYxcJB7JtqheuXWX7qAlupY2BIoq0VvqXkoykpIqONmXBUWsteMaYUoZLYbWOxEyahc4FzKPA0YueR/ENGe8c5XvEI0xmQs+2LyEzjWkakuvrpALNHEgucjDzUgXI1dacIeJP/ln/iQf+v0f4s/+xe/jIz/+qwwjhPYpXLjGZqXkvOXouDC6iXljYpgpme6jjTZheLY44OGj+/zDn/jHTFvPt//n38GV61f5/r/6N3n42XOuScsQbHlsp4koAYegmpjNO862Pcsh0RB45aN3+Qf/+49w/OHnuQwrxCUWAsEJ3ndME2iOjOvCJ//Vbe588pLWw8vzBYurMwbpuHXWk0thDIW1d5QYiYsG1UhC2aYJXKGoYThSErMMh03DzfaQRWnYrDdcrAfUeabDI7aPztEsONcypgHFaMFd19CnkcvLS3Iq+BhZLDoQYTP21sKLHnUODdaJom1JYKQ54NYr9xjGkbab0zUNKZvmIrYzVpdLHj46sc6YM9VifqJR9daX+5dWYjj5UkAKCzjO2YAevCXL+Ut4z7dFUNg3Vry3mYXVTCLXMefiPI5sqO3OqjYIkrzl3xREIhcXWxQjKjnxoJmAkNWcgERNIJXqhdzt9M4JznsiMI5ljxnknPeMyCCB4hxaA4T3Duc90zQiSV7zZXbgpg1JqbtJgSyKBiGWrQ2kiVe4s5pYjhf8he/5Lj78te/nT3/PX+Kn/vmn6GYBaR09K3Q9cFU6Fu0Bc71OLKBlgqJE3wCZnOxzjyUxZuF8yPzoT3+Ei9NH/Nvf9k0c3zzif/juv8Zwe6BPxSYt5UQi0XYBCcLRYsFGlJPLFf0wENqWW597xOXViH+m4+L0AfOsaONJKsSwoN8odz59yvJh4sB3HMyPWaHc6pU2J6KXmklhVmMD9GliFRWCUKKAFFwxerkOA3PnudEcckRLWo2cn1xwuemZmsjKBzZ4coYuBJIvJoiKgYvNiu1oQcIvWnxsSMEMTHMISOMpMeAlIngbUFyUjJBT5nK5ZEjK4ugI7x1TVtrOgNjzizPOTi9RB7pzFqeQVPHSPN6930Lp8PpMQt7kl97s+de/OsqeSu2coxRF8m+xoCDCjte5r8/22UN92Dnrg9svgBe1prCwd6DYbAbEeY6cI0ZbOJLVFHEVdBEnKIVcbUUphZyl9tojvhTSlAx4BIILwG7Ixs4M02ip3jnyE2VF1mJZAzZwNURHHi0g2S9ah8UBeYLQHHJxdo9v+7bfx+/5xt/Bf/c//iV+7uduMzs4rmh2D9qT08B20yM5EaYZJYNrHFpsGIv3jmEcIQjD1JNRxEceXS75yD/9Kc4uL/l3/8i389/+5T/Pf/8df57D4LhxcMTiUGGzgtjSzDobYZYGNnnCeQc5cznA5tVzrh1cJ+eG7bZnmAr4lv5s4OTulosTOAgLfHPASGAkcV4SzVRMK6GFBiUmaL1HozCGDEXMFCVXC6JpgnHiaDHnxuFV+tWWV+8/ZJkzq5TYjhPn2w1DARcdXdfhmsBmSizPz+mTOTXFbm7XOXiKg4LHOzEqdJloXMTXNTVNmeV6zTia8/GV42PGMuLFszhoGafEo5OHLNdr63c4X9eOlbHh9WMZePO48BWAIl93VO9HNbEV3plA6i0eb5OgIITGvPQUtRFhZBseorsbyhaOqtVNfqqkEJlQJ7XcaNj2GTdbcjDrEJQoQLAMYdIEziHBmVsNxvjLquShp2laxDuYMtG7vaYs51zfWwnBBFa5uDqYpn48imURKvbZi1GMs2QC3prLOZMr137mW/yw4n0vXuO7/ov/iP/5h3+YH/3ZWyy65ylui7KxhKm0SFH6EaQklmFJTJGDcsA4ZcZxxAUIsZA8bKYNo24JpXD09E3Wy5Gf/IVfoQ03+JZv+nbmL3r80QvcHUbc4YJnX3yJ5ekjooNx7InHDcfBc/f8nNR5tng6DZRNYByEZVFkaNgslctHmcszIcqcFBxr1mxlxKlDaBgobFMmCrR4Oh/psjldTf1gQRdv4VltHcxmC64cHLMaR24/esgrp+do2zGMhWlUpmmiw9HGyKCJs82STT/iorA4nhGDJ2FkKhsobDiFlgQpVW8JIRPoy8B6s2VMia7t6GYztmm0cuFgwWq55uHJCcPY45wHV2xGozOFqN9BYE8cunv8NzjyF3QQXv/vf42wsePmZLD504aP+S+zycpv2uGc6fC9E/bWMgBajA9Q+QPWgLBsYgesINC2kWlMbDc9rY9VBDUZW1GrxRVKSgURT+MDzhXGcbIp1liwaGIwEYmO5PR4G7CJVVBE99dRnwCBrK1pWUvKBSTbjAIVdDL01wePpkBKA/2253u/909x9+45f/cHP0I3a4AtOfeI0/05Ma/KATaFtu2BlhhbnAsMw4jLBeciIgHBU7Kn5IZ+qwzjwOzQ82sf/zne9Y4FX/0HXuDTnxIuhi3jsmd+9Qj/1DGhbbl2dMjzBwuuLyc+++or/NpnPsX64oJ+vSVcXpjOQyemSbj/cGDswXegTIxi/XRHRkqHo1LP1BRyPTCgbH1CJgz5B+ucOI8XoQkR33acKJzcfoV756dsilKGEUUqKQwOFx2IIs7zwnPv4PRyxaQT3gecE1KyrE335iKFnMzf2AXPlJSU1vR9j/jIwWKBDwZEig+ENvLo/JyThye1DY0J4tSRpbypidEXzm/6zTtUgWzZqhfLcpHyGrDyzY63TVDIxXr9u8NurmyixZqeey+VjrxDZB1Oik08FmFKySZSJ2XbDwQf68lxFCe46lScdCKPmRSx1mMMBA2PuxKVHWLvXfaCFV8BH83m+Vj0cVAQcVVoJXUxZsT5x92L6AhF0Kx4P+Pi/Jyv/bp38g3f+Lv5nj/7lzk/g4PDBvUb4zeoodi7uOirBDdPRtrp2hlN05FSRtNECI44BUr25MkzDlKzGqGJQmbNr37iZ3n6fUf8yp0TLtyWu6dr1gLP3Xgatyn4cSJcLLm+uMLTL72Au3rEwau3uXX/DuvlltDCMCiXl9CvKyfMY8CtesYkOGYVoM14NUdi74UgsFLwmnCitpsqNMWYeeZuPOFGZcoTZxcXrPup2gAkfAiUXGgrxyR2kfbwiI2aSU5OjmTNBuvwZJPOO281qLETO6acWK3XJIXZbEaMLbkU+mFAgqedz3l0seTy/BwQfJ2ROZVsJjLypSkOf7MPYSf3l1qSK84/tgt4K8fbIigoxjQE9kHgjTga1vYxHoPtGhYYwP4wAqSQJ896NeAk0DXROOHZm5NPML3jlJLZslG7D07YOeJqSiQRSi77qP/k7mAtyyf/XaeFi5jOoFgmEbwwTcnGpVcCk2ohSuDgwPOd3/mt/NhP/ii/9Msfp5s9hQ+eoZwgRKBiKLuMwZt7csqZnJSShc26x4dMbBzTlJkmJScrbUy/39I0HU5tnMvDy571ZuD6c09x795dti7zYNgynV+Yg7E6cp64mu/TRE+z6Hj66jW0jLzy4FU2y8SosHkkeJ0RQiGPIxqVYRgYMwQ6clvQphBKNK+LbMNyS5koMuCKEJMV4zE5SukJzkDmlDJ5ikzDhIpx/GNwdL4hYxOzkMKUEpuLMybvkRgZ+oSWTNN2IImp7KY27/wePGfnK3N2ns85ms3MyVnVbNvayHy+4Pa9B/TTtN9Z99RiF1Et5N9g9sJ+PX+ZeQpf6mFO25XrUyeyfynZy9siKADghTa05JwYx0TK2YZlqrXxgrmFoFqq8QbsuOQCpi+t18p7T06FsZ9oYzT+k/c4Z3MPg/MQBYcwTiaOygJNCAiWceSc2ZmrWLsxMI4D3pkvf85VhecDJaVq0+4ts0DNEs4F66gIex6ED45pdckf+3e+kfe87yX+p7/xvzFVgknSCe9NNKNqoBbYwF2rngKryxXLyw7USgefM943bNY989kB4yD024nMQNGWcci2oGlZbhs+8tP/jA+994McP3XA6mKgT5lVn/CIpdglE0u2NuxDM+pwUWg0sF4rV28c8tyNq8wOZsQmU6Rntb4AAu9//1fx6iuv8nO/9inun8OVGaTJQVF8G8gRSs0K+upq3oJluDmjxbo/MWdcCLQuMu8is6ahiYFUBjRn3MyTsOzv4PiYB2cXbIeepp0xJLO4z9j5LwXyNNAPEygcHFyhmzVklM2wxTUtbRsZk3K2PAfv8NowpVzXSrBrkmwD8pWP8BU99oxZfUPOw244zI4TsbOBR4xVSTYRnriCSKbfKjduHnB5vnpLb/+2CApSmy65ovvOyQ4mQMXhpbYpxVSJTnLtOACiqAoSrG7PpeDVEULLMA7IasO1a0c4r4xlIBUbP2cxxdNFSKlYq3JKiDiCD1+QIUw1/RLvCDhy0f2Qmt0xTZOVPQBeqghmhz9k8+nbZA4azx//E9/M3/v7P8m/+ugJ168fk/qRKU/mnFynLdkverTyIqRqQTbbgWnKhCaahfsw4JxnHAqUBoiobMlFSRlEA30RxlE4WRdOTu9z45mnOHmwZH0xUdKagEerMcl5NklP1gEdCu1Bx+G1qzz14gHzK1fI4YJcLrncrGi6wuzKxIsvX+U//hNfT4iFf/EvH/ADf/3/5dMffUCHx6uNRNeSmcoEQOPniHiGUghi37Hve7zzdF1g0bXMY6TxHvLEZtww5UxsPMV5tuPIlRvXaa8csbx3n4SNYHdakEbNhm6aLLOaCm034/DgoAbIian0uLZBxLEdR4ZhRLwnpUTf2+B2LwZWZlW05qVProkvdrzZruz/NSmPu4AAVrbuylrnjF7vHbRezFUsKwfzlheee57PffbX39Lrvy2CAgBiklfnLAU3e6qd+YRHc7HyQATxnuLKHow0LYTWGqogrqkpUyKNE33f0y3aetYev6UTMWN9ybhkLk2qpUqpK5bgbGFoMg9HSmEH5eYnpgOLuJpdFLyj9octhXWAjx5UWa0T/8G3fhgv8A9/5Kc4OJjRpy1KJoZIzgss2u0sKQQRj2omTeb/sNmMbDY9uYzksuLoypy27chZiKGjiSOSEmgglxHVEecC0gjv/Z3vRNdLZBY4unGNk5N7JvrKHpcT0VWmIImmgStXjwjzjsuSuViecfngPkO5JCtstyb5vv6054Hc5f7/+b/wjheu8DUf+AN867//tfzflx/h4ee2iAqSFfEJVxISG1IdcR+8ZxxtvmOMHbNFRzsPxNhQMlxs1ox9jzpoZx1+0dGngnZzwtEVzrc9q6FH4k4vYyrEfhgZpwnvHEdHV4hNBwjDmFFfu1DekSYbSR87M8nZ9hekBBJsvqNm3bsnVTHMm+sP3nStv+7fX2Lqsf91sbF9gnmNTrngamC0O1s5OGx4+aUX2W76t/z6b8Vk5QXM3v0Z7Jb6m6r6/SJyDfg7wMvA54DvUNWz6vD8/cAfBTbAd6rqL73Z+zgMMS7FVY5A2X9778R2vHoxFHmcYUFtF9bOhHeImptt07ZQEperLS44/MzRurC3VQMTOYk4YuNhFKaScFon8WlBVYyrUEuDlHM1y6g9kJoq7PCEHfDonK+224YniCrTZmTewR/5pt/Dj3/kI9y/d8ns8CqDnoNPiA9oH0BfCwrlXCrByoLRetNz+/YrXLnSMT9wNG0ghJacCqGJCJGSA6gnhFhncQ5ITMyuCrE5ZJyUGy8+wyuvnjMsM0EVxwQIZdbRth2FzMOx5/JiyaPtwGCsWeAA5+aowGqbOb3TM1tN3B0ukM894Md+9DM83x7zVb/tBW7pKa9+9hzUxEKlQJFCwURowzAQRehiy2I+Qz2syoaL5QWxOCRnQtcSu4jOOi7HCVXH1avHSNtx59YtJk3Muw5NmSmNTGodqcODhQVLdSYIKgI1wcwlk4bBqM4hUlRZb7fGeg0RkV2GsGt9v/VA8JUHIu37mQO5fULvrNujiF1rB1eOFjx98zrrzZZXX3nwll/9rWQKCfgzqvpLInII/KKI/BjwncBHVPWviMifA/4c8D3AN2E2bO8Fvg74gfrnFz0UZSLvP41STKiUzW9PxBmtOSshRkNSS8TcTJ7QJ0QrvjWPJBGESqcde1ariQPf0LWOwRkbsFRnnF272cWAL1KZlMaqtPNu1OhdT1opBJG9t6OgoInobDYzKsTcseaCIpBK5MDPGIaJ9774FPPjQ/7xP/0FxtLQ5TnzLpH0HC0JIVGYUDdULZdA8NWFx9R4rsCdWyeEdz1jaj4fGKeB5eoCDoRczCdy21eDWme6C1cCOgaGIEhas2gSzxx33DvtSVODn3tS1zNm2Paw7EdW04YJRX2w8xscogq6IZAJZFKaGLbKxXaBiwserOCjH3/EOw8Sz19ZcOMpxzRNpBA52WbWveI7S3Vj8MQYCCEwlIE8ZbKbzPbcC+1iASGYbbnz9KVnEQ6Yz6/w6N4jLh6c0s1aSFa+IeDbjiY2+BgYS2Yae5z3SHQmlx+tPxIXHT461v2WaZhMtCa+8udqSlljQd4BvpgC8Tc6vlj42AntXl9fvP7V/I79uX8l23xKUXBmqCIl4QKQE5RkdpkCIYActFy7esBsPuNi03Py8BHrL2f3QVXvYi7NqOpSRD4GPAd8C2bTBvC/Aj+JBYVvAf622hb6z0TkWESera/zxY8de3HfUXjMbtRd1iDw+lMo7Fov7M9hUePSuxowxAXGoWcYlCYGnHeoF6gCHNvRPTFG0ICTgZJsYRQgTwm0WD/d1RvUKdX6uYKCAZzgnSB4dMzVldjwAXWKR/mar/5tnF9suX37jPl8gUqhn8zboNnNy9T8GMLef2ep58aQ9GkaWK1WvPTy09YlKYkNG5zE6lKtpLFiMTQoBoTG0LLtR7wIm2HFzWev8ujeQ9IWXNMxlWzzBKZkZZP3xg51ZkKapwJio2g17ww/FB1hfb6B0RFZ4NqGz98/Z90sOYimyQiuQ4dEYoBiHaYu2pj6nIzM5MWxmC0QMzrABUcqJvrSPNLEyNHhEdMwcufObaaUmNGRJis/2jaiztqGwzCRczIlvvdmdJsSMUS62QwJsJ22JE10846xz2jZWPn45PrysndHenwdvrTDPZFpvJkM4bWv/gQnplJyvLPrmlLGuUwTPd0sMusaZosZcRHptxN3XjlhterJCk0T2H4lLN7rUJgPAz8PPL270VX1rojcrD/2HHD7iV+7Ux/7okHBVQDxcU1gakir4bWKOYw/sBMjWbxw2BDZJz4jGA212HMijhAdaYTtpieGrk75tV3BB4z0XEFO7zA6tFdsyHIxfz0Xcd6MSjVnxjTuP+5OQZlztlapmHOTDRP2SAikbP30D3zgZT7xidtstiNNPEK8Ba1caxqVHuqoO6fVYENL7T4Yd8M5oeBYXq6Zxl0vHvoyIrLl+MoRTWzRoIZrQO23O4oKUwpo8Uxp4vBwweHTHWefu2R53jNpZhAzCvGtMffMGp/q9uMYasoUqF0SadCc6U8zCxdpOk+KyriARymzUphpQvstWwUXGoY8EXxgSCOoErzQNS2xbXBV5j7lgo4TIQQWXcd2GFgsDhjTyKuvPtzPgRzGkbZp6Lq5rZ1UqmHtjm4o9OMEXpnNO2bzBWMaWG5WxjiNDTln+n5LcL4iOU8sK+PN7dvQ/nX0wJxfexu736DU0P16/sLjsU9H3dBkR9ADA9oNoyrJoLBu1jCfLVgczS1wl0SfM/duXbJZb61PjrFux+krwFMQkQPMf/G7VfVSvvgXf6MnvuAsPDn3wXfyWuxGQTHyjzkj1Z1S7KQaCg876vP+HeuVdDibq+cePxiawDAkNutEFzx4wXsDBNsG+jExVcXdk+JXL44dlzmnRMG4ECa42nVEpGIi1vI0uoOrw2U8koCSuXp0wLve+Rw/9P98vFY9GWR8nAVVvYOkzj6HmEK01EW0S2lLbcX224kHDx7x7Duuo5T68wOb2FPIxGiErGkaODo6xImniR1T3nJxsYZpRJsV1164wq17l5xejFaDz+xdxpwIu3NdwGntjba7y+TQ7LA47SAXytLjGocPjsFBDjCFls2QKMNUMwvzwyyi4L1lbyjZgaaRKe9asCZTt/3a4/GsL5ZcnG1Yrze0Xawry9ykcyoMfU8R453YNEWbp1hQDhcHtLOO7bBmPaztvRtTWm43E8OQaHzDzvHfSja1Ybo7i+UvusSfWNvujZ/XotWp6/ViiR1ZruJTCmUflgpejAxWsPt8MQ/MDg7ouhY8jHng7GLLZjtYSTw0ON/ivGV3k+4cZN+a0cpbCgoiErGA8H+o6g/Vh+/vygIReRbYIRl3gBee+PXngVdf/5pPzn1ojrxKtTzbIyil8tA8SBbU6T5A7ADg/Ul0BgamXN2WsDReqcNBannhnZDGzDhkuoPO+PBiHYMmmgx5p80B9sBeyWabDlaf+zokZpwmaxe62iN2HvA458g51ZvYLm4uiedeeIajw4aP//odG1ITd2hKfm1g84pUJ6cC++BWiqHMvgrBxilx//4ph0cHNI23KcN5S9d15DIZ70CVKWWEgI3qMLVo0ZZSMqeXG24+dcTz77nJyfmdOhw2MRYrl7IWUIcHK8mEPWC3+3Z2uwopJ8bVBAcNolbmZC0kieCUzEDjjeEoxaTtKg6CdVdUrC3t1NytYgUBHY7tek2/7Tm/WKJFaGKwDCNEAPp+Mst5ta7BOCTLQkKh6SKLwwOKh4vVOUmTzRFtYcg9Q5rIOZNSHWXx2mVYF+PjztVup99lBK/PHH6j8e9OBNwb33blCeDaObX2usOk5RligKPDGQdHC7wE1v2G1XLDmNM+kHoJxLbFyotkm5/bGcV+mYJC7Sb8LeBjqvp9Tzz1I8B/CvyV+ucPP/H4fyUi/xcGMF68GZ6gr1FF7tBeDNASMSqtWuslKZZTl1KBlbBXMuIqQGOELvatPQW00LUt0zSx2Uz4trWdhrG6KWVrI+7giVoSWIrqCGKOyJozaSqI97RNJE12MyQg+IAP9ZSKXeScEqFYlvGud17n9NE9Xn3lAbELSBhQ8bgClBZhNJRpByFUfHNnM4nsGdhV8xJYXmw5fbTk5s2rjONkuwfgfVM19ZYNDcPEMExsNwNpnJDSMfSJy/ML4CHPvPAix7dOeHR/a3hLUfClLjQefyZkb13uxeTjqIGuTgw8XF1sIUCr0cJFtjJQYkCkmFu3CKJCImNadpvH4WqZEsXIz1mVoe9ZLpcALLoZse3MOdsJw5Aonv2WAAAgAElEQVQY+5GUyr5tXVTp+56RzMFBR2xb1EPKo/EZWqvNVv2aYVKaKLTzBtKE9pVepVRGpZ3txxOc6vnfgYZf5mOXBWdVgg/gEikrbQtXr14hdoFV6hn7nqFPNjLQBxpfBylPgEHD4Gz3rI36t/wZ3kqm8G8A/wnwqyLyy/Wx78WCwQ+KyH8G3AK+vT73j7B25KewluR3vflb6F5fYPfv4zbQLjAINknXu/o4fh9A0q6Ft2thek8R2yWtQFSCd6SsCIGcE9vNQBNb8IWiCdTVdLWtWVuyHaE8ob+APTnJidXT4j1RHP04VkWecfSTJqRyLoJ4VOHFF5/h/v2HjNNIs+hIZUR8tHaqi3bjYOWE1lpS99/LeuWllhqlgPeBYZx49PCMxWJeTUjFTF58YBoTTRuJMbBcriilsF6t6dc956eFi7ORsYehP2M2W/D+Dz7LL17eYttnmhgtQ0NfA7yB4OuOGarLcDbTe3BGfuq3VutLcsy8pfjTNIIa01BdIEgHqgwpMaUJyPgQ8OJoiXgPQ5pIq5GcMrPZnKODA4J4Rs3MujnDMHJxcWK075LNEIXHEcwDU04wDDgGXHSELuJiYjNtkQbaKJRJrZU7c/RDZXeqATpFdmShx8c+IAh2Tb5E05QvduzeRXaNrrrBHxw0XHvqCCWx3K5sRqWYNFwT5OIsS81SM67qsOSse1fU8eaUq8fHW+k+/AxfvIj6g2/w8wr8l2/5EzxxuBoh9xNz0H1gKJoNlCs20UE1245Te4p7H0WqMxB14rRXXJEaP6rpqgrbTU/TKE2n+OhqMLKdySPmuZhrpK0mLXZTKiU/FpiI9zjviDEwTcn6/d5+dqrmmUEC2QWef+4dvHLrs4wJvA4Q7LWKtmYtV22APVAMU6yAo9bP7nDOwNNcMl7MCuzR6QWhDbz4wtOklEw56TLr9YrDwwMWB9YOVc1cnJ9x+mDJZtkwbiMiM8Z+4O7dh3zoq97He9/7FL/2Kw+IdREpVYD2xJryOJxaCZVLJmsiBpOK+wgUG4rTukgYHXkyUliWhIa6mqqaVEVwzjGNEyWNFOchK0M/4hxEF5jPFxzMFzQhst2uCY11kFbrFcvlCi8e8RZQjL4EoQ1kCfjW0ivjqXnL6nIPjdB1LVogeWXezhhzIvktexce7BpkV2oHI9SA4NCKU3y5AgJUnP2JYJPyBCSOr9/ERcfy8pJEwXex4jhq9oJJrCukOzxs3Mun8eDVTH7e6vG2YDQKgi9iqKo4U9DtAMV65KJ4L6jLj3s2+lif4CiV+af0zib0uKJIERDBaUWPtdBqRymZ/tFIvDbD+UD2A6XNTLJhSi3kDvwuXVdKsVZh8XbTiwS7KXJizIoPgsdGoYuDEANeAlIK2+mCp5454Nl3v5sf/Ps/ChFU5pBz9X4pEEa0GLtuIpGdWClRxC58Bio/ArKlw+pwIZKz5979R0w58dI7n6U4q0klCGerR7j2OofHxzw8OeViDXfPEpoyQQpBIp6rPHqw5ufOPsYHP/gy3/TNz/FL//zXOXnQk7UF6cyKzk2oKNmPtGEy5iiCLy2kwrw5YOhHS8Pdmm1R2m5Gdp4hOZwc0Ym1GG0zzwRzPyB0HauLLXXgMsEHmiYS20iWwrpsKL5jduOQy83I51+5Tb/d0B50LJqOPCU0KbOuo3SJEjK4gLqMeCEB635kSAUfMq2a2W1w0HSB1sOD7Ya0k8oL+BhsAlMlCCUtFrxLIbsKjGKzOr0WfKkdI/8bC6ZkV5sAvlj7XbSOj0uKyo5UnVhc6bgcLhm2EwQhuI6grU2F0hEnaptLzWwcgqQWAZrO4XC4YIFm+/Ct3Y9vj6AgBhbqVMjOqK+GGTymG0MF2TzgPUZxdI9Zjk/We0ZF2v9bRU2IU0FLxDT8fSqsNz2u6+p7ZTOCFdhTpn1Nk4qgai5BRYE6k7JpI9OUQQuxsbYQis2sDAHNQk7Q+I5PffwVHt4/p4mtpd5q06lyHu0eqUHNwuTjEtaJSUDLPovaiWHMdss7zzAKp6crFosLclIWBw1NE5nGzGa9QWfCZrXBu4a2dWZ8m/J+tKH3nm0/8uufvMWH3vdOvvZrP8jP/8LHuP9wCw2M0jCVANIgMaF+RNJIwBMVYoHx1CYyvPz+Q97x0lUePuj55Cfvk3ND8JGkmcJYt8RoqH6wFH/mZ3SLjjSkylewSc0+21i7+XxGDFYGPTpZMq57jhaHBHFIMbm8AKGJpCZXJy3AO4Y8Mg3G1uy6iM+FMCSCQuMyXhXvthwDaGKbH68jTyS4KjAqhjsJwbqDuZBF9wCzrTV4rJL4oiu+dtsc+F3Q36HbBpIPw0icC7PDhp4lbevAQ562Rl2n2ERxhTRlKMZe7doWz2Alc7DWdtNEYvTc++xvIUGU9brNMmqXknqkGqVaMHgM/AneC0WsFNiZqX7h8eSF8Y8DQ6Ba3ZqL8DAm2j4Rnc0cEJwpMKWOqK8W83ipXRAqcGnCreA9WpSUTbnWdVbLW7JjPgBRW9Lk+Uf/4Cd4eG+Jc9FqQTV3KU+wckiNqFoqVqqy0xTuuLmw674U9ZZliMWz2LRoTjw6WaLF4f2RBZladmiBi8u1qSYFshQKqQq7Cs4pse3oh4lf/9Rnec97Xub3/v6v4rOfu80nP/OA5bIn1A5Gm45xWTBvqjUl9ZSgvO9dV/jmP/YH+aZ/78McXev4zGeW/I0f+Hv8zM98lDHV1mbMhuTjq/rT7fkETTTFqfeKi7sbxUq3KSXGYeDk9MyG0XSLKkwzr0TfCk0I+EaYXEFioTiY0kg/jWSlgsAFnQACeI+XQCPQiGd2GLm6uMYqmcnvNEyM48QwpTraHoRM0YzzVtoEL5QaoHet6zcPCk+s0WKbUCXW03gHWYkx0M08BEN1i9/xTcC5TGg9TiKSzQ9AaNBs5rfjtGUYR0rpycn2UPFfXqDxN+UoWqrGofaFg68eAmAGlNTZeVJvykr9rD8hlZIsUhWWux91WNuwojYZgWgtMPGgI2zXE6FrrAVVMgSp2V023UENWni7wXyxm7VoYprscgbvKKUQvMN1oc4kyEQX8E1LSspnPv0q63WGxqNDbzdFVNq2qe27TE6J3dfO2dkXeSLmFa3ELOz77USZu07JZp0QViwOZrRNxzCM9O1ICB2ryw0nJxu0TKRkdatzO8l3JqmCi1wOykc//XlefO4q73rpGs9d9wyrgkuOe6+e8uhz52hSQpvpDjPvev+zfP03fIg/+i3fwI1nj/mFf/Gz/MTf+Xn6TYs4eOGlZ7l954R+mkys4w1n8BIQNbboMI4s5nNGseul++gbUVU2qzXDNIAqi9jhvGMcB/BKCIHYRWLbsB025DASXcNUBrZDra8V0GyBuG1RPMu+JzVC6zzbskGy42jeMJ+1xCYSY1MViOYXsVyu2Gx6Li57xikhJSEEFEW9Y9qRcd+gJWlTxuzvyVVDYiv47bOZ25sBnJoIDYRWmEpPWNjvx9hwPJ/hCTgXKBNs1kq/zkz9RN9P9NuRnKbdXWFro0AZ3jr28bYICjvHI4cBQor5D6RpIgRP20brudZevbjakRCpN/zjoAD7zll9cUglW7QMYb84jEJgzMZxSAz9SBcDeHtNlYJKBqeUIqZhxyZUi3P4IqQioGb4qbuMQmxH6gTWW0PbBcHFuT3ngrU27WNYjzzZsBrvHV4CRdMbsL2q8xRmmGLrzkqcfeqCQ1Xo+8z56Yqjo0NCaLm8XNI1C7p2zjRszGnKGzOuoAhpn4wA4Fs2w8BnPnePiwcnvOOplq//8Af4+q/5avr1ivuf+zjHx4dcuXGVwRVefM+7iUeH/MrHf5W//rd/jl/5l7cYtjANULTh+Pg5Xn7ns3ziE59DUNKQKa52lirMblRzpVu0bC/XSIKubREvTOPIerMCqVT0IVfg2Wzz6cB1sJ4ukQihaxnywLYfDGCswVPEWsu5CP2Y6LeZVDxTo5AgTSNn64eQ1DQZIdJ2DW2MtG3kqevXeGE+J2Xl/PyC89MLzpdLpjQxJEcMtY36BZmCQ3Oq5CNfA/lu3qRFEuepoHYxG7vowCnzo4YrT8+JPqITuOIZ1hu2255xUKY+s15O9OtcL2CA3JlLudTPUXamQG9NKfk2CQrWsoqhNRMTKZRUXZDFvAtgd1NggMoTA1hUd/+rh9SyVYx+UqqTk3dWgpRcLdy84JO3qdSbROjEgCWydRmkGP4QAqKOlAslm/HGjoJaP9ATZCq7yMEH2piZxkQpGW0WFHIFLv2+U5LzxDSa45D31QUqyj4DKKVmToU9V8OYiwYvqya7savE3IuJny4ut1xebjg+bjg7X3Lzxjt45pl3cPvWClFrATrJ5jC0Z+xUxD1NeB/JGR5djFxerjk9/yU+eevzPHPzKjevL7l6E9JR5JOfvs8/+ejH+MSnH/HJz5yz2kIaPYtFx5gmSiqkswc8feNZ3vmu5/jsp+8Yeek1wiD7ey6ZxbwjDRvQBBKYhq2l70VZzI2D4fzImJMZ8HYO3yo9W3JTaNuGsdhw2d2SSLnS1xWb9zGBTsVaj1lIuUE8ZE0188gMKbHpM37sd0oc2nhK2zRcOT7gYDHj3e98B2ks3L3/kHsPz6zrEwP5dTijZbl1afiCYCVDVXfU9WMEOM3ggq2jbtbwzHPX2eqK5apnezEwrifKlBFpSENmu0qMYyWQSSBn2zjsvXakpS9Nq/G2CAohGBEoT4Ux2YIPfidFNoT/MaCotYzI1cy16iaeoI6K253qUlFkQwtLRZFL1TM4F8iCCaamgX4LB21Avekh7MWAyjj0zmrYlM3kdUdz9s5V+mo2j32jMtF2EZGRPKiZsZqlM0F2fTm1NFAzmjIpKdkldBKksV3NO1cXmVp7rOQquqIKtlKlBJtRp9QT0G97Hj54wHz+DDlnttst8/lVSgLB4YJDgsfrYIN9JVeQDGZqPJBJZ6g7ZApwZ7vmzsdP6O6csJBAdKdA4f79VIVhHTFcp/MHDFFY9xNOthS3ZDtuODm7x7te/AAXp1vun5wSZzsjG48Ta++lksjqCd7s5fq+BwQvLW0XjFyWEkWUECB0HhpHkglpHIvFjNV2zXpttWMIwZSngukbpsI0KS4VuhLwsWPKStBoPBHJTGmNBEhqtm9Z2etONtPIuh85W64IDg5jw1NXj3n+6ZvceOopXn34iAcnj8C7LxgC673fl4UqZgGHGKSs2VnXQB6T84J3THnk7PKcOw/OCQVC8XSywDvotyPrVWbqMW2ODwaAklGZHrPd9gDYWw8Mb4ugALCYzxiniTQmUilMU2GaJmKMZnmmipaqh8h1sLfIniOuOwxBDbizw65ocKU+l+wHgrHwyBnRQuts5IBuErSOIJh9WL1Ihnf5PfgpYi0l6xiIcRj2uqwd5mGiom42YyIzjIlOAogzuwTRin/UHTPvqg8z/mDKSBB89IZue1tMaVJUEyKpXnNvr6Hu8Xg8VXISHp2tOTpe4p1nu11D8UxpSeujCbW84ghIUVR256yAjDb5SlokerIr5OCgDeSF51JMYh5zgK4npEAjCyiOVCpluEz4kBEaQmy4WPacXZ7wzHMzHp4peUzYjpbNnNUVlERK3qZGA0GEtmmMA5ITm1TT3zYRmkA48vS5h9YxO2qZGBldtY3trYsVc0fUgG7VOkZpgGxzK0egVxO3NUFtMnhKiDpMchZtN698kOw86nf0d+VsyqxOTjlbrXnm5nVeev4qVw7g1fvnrPuEBCHVsnaYlNjuuA0L60r5XMtlYy8GAuSE0IKDKa+5/+Ac7xsOZkeEHEirxOXJks12wHtPDMbB0ZzJXnBe96DlbmNz+4z2rR1vi6CQc2a5XNI2gfm8NYuszcQw9BYY6uCTUsBLNaIsFiFNLCJMRXHBom3WHaQge5Yke4w41xvaIaKEKrSMXtCkTJtCGz0SLYoXta4FugOB6pi4nb+CZCP4VNak1MCRM/Q52a7sHM4HvG8oxezGvXf7QO6QPR9exMqHXDJlyGyHTGyaOiDXmXRbDUfY2Xhb9rMDsozoFduWrBse3j/h+GrH2A84XePdgKklFZdkz9B8jE3bYqWWOV4Hy7pSz9Rnwhzm17Llw0NL2yl5nShlsKxOoPhCoKDZlmPK0DYdt27f5sWXb3J8LfDoge32O2NaxJGzyX58DPjgiTXI9cOWIWcWxx1t11g3gcQmrZAOrtxYEGfKVJTZ1YYyBpaPEmWp+NLgkmOz3tYgZSTgflyjRRkqGdYHo0OH6JDs8Jj4qxQbE6jOg4+ox9QqapkmCmd9z+XnP8/NqzOeeuoKv/3wPXzm9l1Ozi8I0aHe0bSQMY5LLtVVS9TUqN7k6GggRshTh6IUyfgAR4tDZOxYna3Ynm9IfTH6eJHdgLR6vcqXWCi88fG2CApgkW6zSfT9RNNE5vMrzBcLVstL1us1MQSatiUns+5yT/gp71uSSlVWVlmUPNYOPBkrtSSjr7qIj54yZTzm6dhvQDqIjWEJzlVZ9a5EkV3aXizdLkoiEVxtsRXr/bsQYEzGciTQNA2NqwNmn+BX5PJYBAPgyWaRXqxd57yZvuSshIq0C1r9LMVwBqslqlBH8S7Ux4TVpieVgRhbZjcPodLDVTO52ti9lsQLot5ET+pQbwIj7xpwPdM2MfWe6CPeNeRY6HNPKj2u1OXkIDvY0eg0pb0d2nq15fDggPPTJZDx3uZvihe7LrkQ2oioUaaHvid2nps3rnPl+jGXqxXuvGM7XnDj6Ss88/wh6+mMfloziQHWXXuDg6ues/GC1XJLHsA3mHCogI+ZWQeL2PHSs8/x8N5DlkNvY+swYFIFywjroCGVjKo9UUO/bRRknETUC3fPtzza9rx8fcZ73vd+4u3b3Hr1gXlDeE8InXlQ+Ee2o9cSI3ggJ4pmGh+BiZGBa4cds2sNOi64d/+U9dmaPFWMS7wxbNnJ5r58x1snRH8FDxGq27K1xqz9c0maRg6PrnB4cGB69+2mkndsDNjjTD2zm7dgCsrHUuyirzVXBXaXFMRYac4HxHVAR55apl5gcjjrzOMoOGe6C+8qkl3pzdabrqmJE3CQNNtwEmeMOTST00DWhA82At45+77e78qf2o+uGZA491pdvtrg2l0Xxgxfdj+zaz1l0ISIknTan4ftRjk7XbHdDruKyv4rSsnWpnv8n5JLIGuwLgdUxSeU5BkH6DcFLxFU8DFQXMUDMNAz17aynXZHCA3TOJGBy8sNs8UBMUZSLQeN1OdJmfpYAjcx5p52HnjpPc/yzEs32aQlZ5tTJl1z5Tjwdb/7vXzt17ybZ48jh97x8lPP8sEXP4isA2e3Lzi/v6TfDIxpoM8DLnqOrx/w1NMdx9cC3/B7P8S3/eFv4He88x3c7CJtyWgVkeEU9VWM54wOl8tESRmyM04AQsEzec/gPWMwfP/Tr9zm83c+w3MvPsPv+pqvAiYTxomVhz53SG7xpYMUKKmDcoDLx+h4bN5bmvhDf/jf4vv+6l+jbODs1RVN8gTduYGV6t/x5Q0IdsXeJsdOBCVi/ISUJs7Pz9luNhweHnD16jHeNwzjQEqjKcLQmopBDIFaeL/2dfcdgdceBSx9k0wOlQPpG9DItIUygVPBYa1C50C8WjaAKe6KZrQi+KlYpBfnaJrqdCTWpBCM/5A147tohCF9nDaH4C0DeGK6sK8iMMFBcUhxOLUSJ03FbrxiKsUQLEjuYkgqZtHlBKIPRO/p+4nTR+evTZne8CgUNcsvCwi5SqQ9KHgNRI3oJKRUcY9W0JApoRjpSKxlWNTsx8dk8ypEAlOyFLc77MiqFJv5bKfCe3OgVpvY1c0bXnrP89x87iaX21PunNxnciOL6wOLaxP9eJ9P/fqvcvLqGUfxGgfuWU5vKZ/+1Tuc3D0nDUYy6haep24c8O73vsAH3v9uZl3Li8/d4MVnrtGfn/AHftdX82/+vq/j6sFBDbIKYrV5lly7UOyzO9ldW28GOpMWKzFDQxJHCYm7ZydIHPgLf/G/4U//qT/OYlFI5ZJhOGUYRjQr46RMCYahUGEkinimYlno0dEx4ybz2Y/e5igu7NxXe0KAHa+N2m51dcW84f31JWQTb5PyQaq3vvWOnBgDMZfEcnlJKYXFwQHXr7esVis7qWWiT5XK6YOBgYgNXzWK5BcYt+x21N2tV8hQJ12Lp+a8UEaYthBbxTVCExxjmWzeRHVFogYbcQZKpB346Ssw6DwehWzkoCmbgKiddWwebdg7ZoCVExirsNTW4/5j5xrUxFcxF3vtQPBlX0Y5MV1/LgU02SItWEbhoUyZ00eXvPakvPFhJHFlZySu1TUKsYG7jTSUcedq5QmzQD8mcvVIBGdqTq2fTwStevYpmaz56Kjj5KQy+VztzxXzwEiqdLOW5194Bzeeuc7HPvUxzjYrpIWDKy2zoxEpibPVJdJnSt9w99HA3bsf5eJ8AMk4habzLK5Ert28Stt2UOD89IzN2ZoPfPh30vjA5z/1Gd798nt4eP8+/WZLg6cwWBZqOKgBoQqu6k/YBQ6eiLFS++AqjM6cke68eovzy8/wX3/3f8jpxef4uz/003zjH/pq3vfbX+DGs0+Rcbxy9x6f+NRn+divfYrN5Zo8eSQpqcDP/JNf4Mf/v59FtEUJoKbafHJU4VfieJsEhcc7Ra0KEHG0TQsoy9WKzWbLfD5jPp9zeHjI2C9ZrnqmMYHY4nMuGrusgnBa+9JPZNgV8feoOONDUJWUFKR6MOYC4yZV6qyVC1WbZISmClmOWQmSbSxasDfwYkQnZVfiJDR78pgZSqZpOkIYyVNB3Q7ke+1F3nFOvO5A0l23pD7uHLl2WlLKlbNQ16XDShEFVc+UMx5fB61kwxveJFtQjIKNQFPnV2gx0KvxAYoJsQqZ7dDjm0BobILzlEYc/397ZxMjx3Hd8d/r6p7P/Sa5EkXJsmwzpmVHkRlZomFDMBLAjnVRcoqBIPbBQC42YB9yUOCLrzHgHAIEARL4G7ENBLERXwI4CIwIOVjxRyRKIs3V0hI/RFJLcjm7M73TH1VdObya3VmaK5GWpZkF+g8spremsfvvrurXr169968YiAOvSDUxK6syGF4YDFIWDxwgNsKolE9XLCy2gsXZee5aPshgmHL5xdNsbA2IuzC72Ma0DKmxdJOEaz1Lse7YuFKysT4kLw2IwcSO5YPzPPbRD9JLb9AfbGALx/lzV9i8VnBkeY73HDnK6q9OcWr1LM+d+jUXLg2xBqSj2ZIuPIBVENce2dII8JHVwQCI15xGXFC+AjBdiFI6MwkrKyd5z9GD/MVffpze1iU+dOJhDj3Q5sr1K6yt95CZjPm7HYcGCf0bjiKH/vWEKIt59v9WqXIQ5rFeNF72O44f3ArTYRTCOvDoxenHNlmpqopmo4H3FZubA/I8o93pMD/bot3usLGZspVmVM4hJERRjKdgbHdRnUKENWFk298arQHquT7ECmKQUsBCMSygiUattTQCW2pwUIw+MCIRMtojAtFphANiLfhxpcfmBUXhEVeQNLrEcYwrC0b7OdwOKr+jSDTaCKSq1M11QWciCgKjzmk+REiqD/oTMZGJ8XZshO/1v/ChL1TgQ0yQRYtVLGawmdGa0SrSzGZ0Wy0azYSqcFSOsAlOhUOQSk1eHMVachJ5SudwOJozCTZHNTRLjwmiIiLCZj9l/cZVfORodoTGXJO40yJuGcqkRX/guXg+ZfCqLolmWxA3K468Y5GDC7PEjQiJK3r9Nfr9PgvdBVrG4JoxH37oMaI84aWV82Re6JcFi3d1WVw8xPlL18iCpoWEsYnRZZVo20McKdaC8cGN8A5jVXzHmQPYzNFpz3Hk8BHOrvyKEx85wbFj9/ONb34HmfFkzjEotIusNxhJcEUbXKzebuQp8xysYGiFikxdtXhjX+/NYTqMwng8LSQExXGsm6ugBkNd1VF+f0qW9pnpdpntztJqtelv9MmGJXFSEZsG1qrIicq8BfWlkQGokvBWjdRT8A68Ji0BECXkrsIUDsmhOzsDkWVYZEHJKKRbhwCD7m+5M12JDHipKDKwGZSZvrUra6najqhhqLJKU3Sr4A1sv739zg0JMYnRcRWc8JGGhB9tw4xHs/A1xiCig1X32NblxySKxuyk3w42EoKR27mZXv++F0cVqc9sGkLcjECMrmYMLVWWErUMphkT+6bK0Cdgyxxbys61eKdCNWKIRA2nJgR5Ot0GaZlt7yaN0b0qhpsVWb+H4GjNCLOzTZqdNiJNbA5XL/S5sZbCACSFFpY/PHY3jz3+AXLpc/HljDMr53hx5Rxx2zG/2ADfozmj8arOfIvVc2e5cPU1Gp2EhXvuxviEAwuHubTWw1fZ9g0Z+XLbQzS4nd5HIQStGw97r4vescRYl4IR3vneY7Tml0iHr9BPr/L+D7yP7/7r06zf0PqLqoIobmMqw8Z6ji0jfOmJy4h8mEGpQneQByIj2YBbhQLHTcW4N3FnOQowJUYh5CfqW63St0xsKkxsxgqhQkk1+qb01tLrpaRpwUy3w+LCPDf8Os7nWBvhKtVMrJzVOECkSUc7KxS6NBczisY7nUvHKsmWWaFZGSgiumaJVjeiJ+sMsj4IJKYRNPUIS5/BnRSVubCSk+Ux2bBCrCGJYpzNsWVBs5PQT3UJUSSI01ahhn6kRzbeyZXXQqLRvapCzv+oAZAq0hoPH/bJICKWaFu0pkKXvqxVDSuiIExjVL59u7jMVWAcTizNlqHZaiinESUb0RoK1dBT5hbahrQsSYyhGbVwYonCAzMi6EXjEhKmPKBKzY0kJlOBNCpvsFbTw23uMKZkcanJ0d+7hwd//308f2qFU2cuMtiyDK9pdalEcOy9y/z5J/6Y4w+9m8vr5/nJz57m5dVr9G6UYaNdj5eC9j1t7rp/kcFGyeX+JeZm5uguz9PrbyGFw25ZXr34PNZqlWpVeaQKY86HmVtE0FsSgsEAAAVGSURBVAk1Qd8iUjMdG0zcoMyHmGYM7jrt+Rne/8iHKJuAy9lIL3DvfYfpzM2ylg5wUax/I7VkaUaZWrCRjocsIrYSkqwELxqDUtNkxk3Urodoe4Ts+no7ieG2MRVGAa/FORKCgR6PtSVx0tw5YdfpKhEWVY6yKLhRFkRiWF5eZJin9Dc1t99W0XZKcBXKjlUgBbbrFQgxBh8iyqLxv2ZiwOVkWznra9dZPDTD4vwCjWZEP9V9DQnJkVHIWlNJMKfTlRCfoFLZ9wrwXrUS5zot4liFagOVO4QOjt+4jZ4g9jnac1NL71wVNCkkgrBsGJt4e8oBUEWagIWP8FGMRIY4EUxDVXtKXY7BJELcamAzS+EcVWEZ5gVJ3KBlEkZD6vUEibyvsEWhXoPV+2di3cexdJZW17J0aJ5DSwu028tcenXI6up1Ll/OkEhoNbvMH2py7/2zPPFHj/DoHxzj5DPP8d//80vW+jmZtWxlOY1ujDfCbKfLQw8+TJE7Lp49zfqVl1hamqM3SOltZvR6GcZFxF5I4ubexPfoC+/UI0saDajUM3z8Y4/xjvuWMZxjbi6mEVkqyRGbQ46mteeeLLXYoQOr6d1Yrbb1ux7uO50w7ITSf5sFRvGv13tvE0TkKpAC1ybN5U3gIPubP+z/a9jv/OGtvYb7vfeH3uikqTAKACLyc+/9I5Pm8dtiv/OH/X8N+50/TMc1TE3yUo0aNaYDtVGoUaPGLkyTUfinSRN4k9jv/GH/X8N+5w9TcA1TE1OoUaPGdGCaPIUaNWpMASZuFETkT0TkjIisishTk+ZzuxCRV0TkeRF5VkR+HtqWROQ/ReSl8Lk4aZ7jEJGvi8iaiLww1nZLzqL4+9AvJ0Xk+OSYb3O9Ff8vi8iroR+eFZEnxr77m8D/jIh8YjKsdyAi94nIT0TktIi8KCJfCO3T1Qc+ZLxN4gfNwDkLvAtoAM8BD06S0x1wfwU4eFPbV4CnwvFTwN9OmudN/B4HjgMvvBFndD/Q/0AzZ04Az0wp/y8Df32Lcx8M46kJPBDGmZkw/8PA8XA8C6wEnlPVB5P2FB4FVr33v/beF8D3gScnzOnN4EngW+H4W8CfTpDLb8B7/zSwflPzXpyfBL7tFT8FFkTk8NvD9NbYg/9eeBL4vvc+996/jG54/OhbRu424L2/7L3/ZTjuA6eBI0xZH0zaKBwBLoz9fjG07Qd44Mci8gsR+avQdpf3/jLoAACWJ8bu9rEX5/3UN58P7vXXx6ZsU81fRN4JfBB4hinrg0kbhVslde+X5ZCPeO+PA58EPicij0+a0O8Y+6Vv/hF4N/AwcBn4amifWv4iMgP8G/BF7/3m6516i7a3/BombRQuAveN/X4vcGlCXO4I3vtL4XMN+CHqmr42cu/C59rkGN429uK8L/rGe/+a9955lSP6Z3amCFPJX0QS1CD8i/f+B6F5qvpg0kbhZ8BREXlARBrAp4AfTZjTG0JEuiIyOzoGPg68gHL/TDjtM8C/T4bhHWEvzj8CPh0i4CeAjZGLO024aY79Z2g/gPL/lIg0ReQB4Cjwv283v3GIijx+DTjtvf+7sa+mqw8mGY0di7CuoNHhL02az21yfhca2X4OeHHEGzgA/BfwUvhcmjTXm3h/D3WxS/Qt9Nm9OKOu6z+EfnkeeGRK+X8n8DuJPkSHx87/UuB/BvjkFPD/KOr+nwSeDT9PTFsf1BmNNWrU2IVJTx9q1KgxZaiNQo0aNXahNgo1atTYhdoo1KhRYxdqo1CjRo1dqI1CjRo1dqE2CjVq1NiF2ijUqFFjF/4fk0guGeG6naAAAAAASUVORK5CYII=\n", | |
| "text/plain": [ | |
| "<matplotlib.figure.Figure at 0x7fcbc279b898>" | |
| ] | |
| }, | |
| "metadata": { | |
| "needs_background": "light" | |
| }, | |
| "output_type": "display_data" | |
| } | |
| ], | |
| "source": [ | |
| "image = (\"flowers/test/13/image_05775.jpg\")\n", | |
| "img = process_image2(image)\n", | |
| "imshow(img)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "markdown", | |
| "metadata": {}, | |
| "source": [ | |
| "## alternatively this should also work" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 26, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "name": "stdout", | |
| "output_type": "stream", | |
| "text": [ | |
| "torch.Size([3, 224, 224])\n" | |
| ] | |
| } | |
| ], | |
| "source": [ | |
| "def process_image(image):\n", | |
| " ''' Scales, crops, and normalizes a PIL image for a PyTorch model,\n", | |
| " returns an Numpy array\n", | |
| " '''\n", | |
| " img_pil = Image.open(image)\n", | |
| " \n", | |
| " adjustments = transforms.Compose([\n", | |
| " transforms.Resize(256),\n", | |
| " transforms.CenterCrop(224),\n", | |
| " transforms.ToTensor(),\n", | |
| " transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])\n", | |
| " ])\n", | |
| " \n", | |
| " img_tensor = adjustments(img_pil)\n", | |
| " \n", | |
| " return img_tensor\n", | |
| " \n", | |
| " # TODO: Process a PIL image for use in a PyTorch model\n", | |
| "#img = (data_dir + '/test' + '/13/' + 'image_05775.jpg')\n", | |
| "img = (\"flowers/test/13/image_05775.jpg\")\n", | |
| "img = process_image(img)\n", | |
| "print(img.shape)" | |
| ] | |
| }, | |
| { | |
| "cell_type": "code", | |
| "execution_count": 27, | |
| "metadata": {}, | |
| "outputs": [ | |
| { | |
| "data": { |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment