├── .gitattributes
├── LICENSE
├── README.md
├── images
    ├── perturbations.png
    └── reconstructions.png
├── net.py
└── reconstruction_visualization.ipynb


/.gitattributes:
--------------------------------------------------------------------------------
1 | *.ipynb linguist-documentation
2 | 


--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
 1 | BSD 3-Clause License
 2 | 
 3 | Copyright (c) 2019, Adam Bielski
 4 | All rights reserved.
 5 | 
 6 | Redistribution and use in source and binary forms, with or without
 7 | modification, are permitted provided that the following conditions are met:
 8 | 
 9 | 1. Redistributions of source code must retain the above copyright notice, this
10 |    list of conditions and the following disclaimer.
11 | 
12 | 2. Redistributions in binary form must reproduce the above copyright notice,
13 |    this list of conditions and the following disclaimer in the documentation
14 |    and/or other materials provided with the distribution.
15 | 
16 | 3. Neither the name of the copyright holder nor the names of its
17 |    contributors may be used to endorse or promote products derived from
18 |    this software without specific prior written permission.
19 | 
20 | THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
21 | AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 | IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
23 | DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
24 | FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 | DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
26 | SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
27 | CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
28 | OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
29 | OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
30 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | # Dynamic Routing Between Capsules - PyTorch implementation
 2 | 
 3 | PyTorch implementation of NIPS 2017 paper [Dynamic Routing Between Capsules](https://arxiv.org/abs/1710.09829) from Sara Sabour, Nicholas Frosst and Geoffrey E. Hinton.
 4 | 
 5 | The hyperparameters and data augmentation strategy strictly follow the paper.
 6 | 
 7 | ## Requirements
 8 | 
 9 | Only [PyTorch](http://pytorch.org/) with torchvision is required (tested on pytorch 0.2.0 and 0.3.0). Jupyter and matplotlib is required to run the notebook with visualizations.
10 | 
11 | ## Usage
12 | 
13 | Train the model by running
14 | 
15 |     python net.py
16 | Optional arguments and default values:
17 | 
18 | ```
19 |   --batch-size N          input batch size for training (default: 128)
20 |   --test-batch-size N     input batch size for testing (default: 1000)
21 |   --epochs N              number of epochs to train (default: 250)
22 |   --lr LR                 learning rate (default: 0.001)
23 |   --no-cuda               disables CUDA training
24 |   --seed S                random seed (default: 1)
25 |   --log-interval N        how many batches to wait before logging training
26 |                           status (default: 10)
27 |   --routing_iterations    number of iterations for routing algorithm (default: 3)
28 |   --with_reconstruction   should reconstruction layers be used
29 | ```
30 | 
31 | MNIST dataset will be downloaded automatically.
32 | 
33 | ## Results
34 | 
35 | The network trained with reconstruction and 3 routing iterations on MNIST dataset achieves **99.65%** accuracy on test set. The test loss is still slightly decreasing, so the accuracy could probably be improved with more training and more careful learning rate schedule.
36 | 
37 | ## Visualizations
38 | 
39 | We can create visualizations of digit reconstructions from DigitCaps (e.g. Figure 3 in the paper)
40 | 
41 | ![Reconstructions](images/reconstructions.png)
42 | 
43 | 
44 | 
45 | We can also visualize what each dimension of digit capsule represents (Section 5.1, Figure 4 in the paper). 
46 | 
47 | Below, each row shows the reconstruction when one of the 16 dimensions in the DigitCaps representation is tweaked by intervals of 0.05 in the range [−0.25, 0.25].
48 | 
49 | ![Perturbations](images/perturbations.png)
50 | 
51 | We can see what individual dimensions represent for digit 7,  e.g. dim6 - stroke thickness, dim11 - digit width, dim 15 - vertical shift.
52 | 
53 | Visualization examples are provided in a [jupyter notebook](reconstruction_visualization.ipynb)


--------------------------------------------------------------------------------
/images/perturbations.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/adambielski/CapsNet-pytorch/bee0cbee5bb8b30399e75c751d3085c2f261dea3/images/perturbations.png


--------------------------------------------------------------------------------
/images/reconstructions.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/adambielski/CapsNet-pytorch/bee0cbee5bb8b30399e75c751d3085c2f261dea3/images/reconstructions.png


--------------------------------------------------------------------------------
/net.py:
--------------------------------------------------------------------------------
  1 | from __future__ import print_function
  2 | 
  3 | import torch
  4 | import torch.nn as nn
  5 | import torch.nn.functional as F
  6 | import math
  7 | 
  8 | from torch.optim import lr_scheduler
  9 | from torch.autograd import Variable
 10 | 
 11 | 
 12 | def squash(x):
 13 |     lengths2 = x.pow(2).sum(dim=2)
 14 |     lengths = lengths2.sqrt()
 15 |     x = x * (lengths2 / (1 + lengths2) / lengths).view(x.size(0), x.size(1), 1)
 16 |     return x
 17 | 
 18 | 
 19 | class AgreementRouting(nn.Module):
 20 |     def __init__(self, input_caps, output_caps, n_iterations):
 21 |         super(AgreementRouting, self).__init__()
 22 |         self.n_iterations = n_iterations
 23 |         self.b = nn.Parameter(torch.zeros((input_caps, output_caps)))
 24 | 
 25 |     def forward(self, u_predict):
 26 |         batch_size, input_caps, output_caps, output_dim = u_predict.size()
 27 | 
 28 |         c = F.softmax(self.b)
 29 |         s = (c.unsqueeze(2) * u_predict).sum(dim=1)
 30 |         v = squash(s)
 31 | 
 32 |         if self.n_iterations > 0:
 33 |             b_batch = self.b.expand((batch_size, input_caps, output_caps))
 34 |             for r in range(self.n_iterations):
 35 |                 v = v.unsqueeze(1)
 36 |                 b_batch = b_batch + (u_predict * v).sum(-1)
 37 | 
 38 |                 c = F.softmax(b_batch.view(-1, output_caps)).view(-1, input_caps, output_caps, 1)
 39 |                 s = (c * u_predict).sum(dim=1)
 40 |                 v = squash(s)
 41 | 
 42 |         return v
 43 | 
 44 | 
 45 | class CapsLayer(nn.Module):
 46 |     def __init__(self, input_caps, input_dim, output_caps, output_dim, routing_module):
 47 |         super(CapsLayer, self).__init__()
 48 |         self.input_dim = input_dim
 49 |         self.input_caps = input_caps
 50 |         self.output_dim = output_dim
 51 |         self.output_caps = output_caps
 52 |         self.weights = nn.Parameter(torch.Tensor(input_caps, input_dim, output_caps * output_dim))
 53 |         self.routing_module = routing_module
 54 |         self.reset_parameters()
 55 | 
 56 |     def reset_parameters(self):
 57 |         stdv = 1. / math.sqrt(self.input_caps)
 58 |         self.weights.data.uniform_(-stdv, stdv)
 59 | 
 60 |     def forward(self, caps_output):
 61 |         caps_output = caps_output.unsqueeze(2)
 62 |         u_predict = caps_output.matmul(self.weights)
 63 |         u_predict = u_predict.view(u_predict.size(0), self.input_caps, self.output_caps, self.output_dim)
 64 |         v = self.routing_module(u_predict)
 65 |         return v
 66 | 
 67 | 
 68 | class PrimaryCapsLayer(nn.Module):
 69 |     def __init__(self, input_channels, output_caps, output_dim, kernel_size, stride):
 70 |         super(PrimaryCapsLayer, self).__init__()
 71 |         self.conv = nn.Conv2d(input_channels, output_caps * output_dim, kernel_size=kernel_size, stride=stride)
 72 |         self.input_channels = input_channels
 73 |         self.output_caps = output_caps
 74 |         self.output_dim = output_dim
 75 | 
 76 |     def forward(self, input):
 77 |         out = self.conv(input)
 78 |         N, C, H, W = out.size()
 79 |         out = out.view(N, self.output_caps, self.output_dim, H, W)
 80 | 
 81 |         # will output N x OUT_CAPS x OUT_DIM
 82 |         out = out.permute(0, 1, 3, 4, 2).contiguous()
 83 |         out = out.view(out.size(0), -1, out.size(4))
 84 |         out = squash(out)
 85 |         return out
 86 | 
 87 | 
 88 | class CapsNet(nn.Module):
 89 |     def __init__(self, routing_iterations, n_classes=10):
 90 |         super(CapsNet, self).__init__()
 91 |         self.conv1 = nn.Conv2d(1, 256, kernel_size=9, stride=1)
 92 |         self.primaryCaps = PrimaryCapsLayer(256, 32, 8, kernel_size=9, stride=2)  # outputs 6*6
 93 |         self.num_primaryCaps = 32 * 6 * 6
 94 |         routing_module = AgreementRouting(self.num_primaryCaps, n_classes, routing_iterations)
 95 |         self.digitCaps = CapsLayer(self.num_primaryCaps, 8, n_classes, 16, routing_module)
 96 | 
 97 |     def forward(self, input):
 98 |         x = self.conv1(input)
 99 |         x = F.relu(x)
100 |         x = self.primaryCaps(x)
101 |         x = self.digitCaps(x)
102 |         probs = x.pow(2).sum(dim=2).sqrt()
103 |         return x, probs
104 | 
105 | 
106 | class ReconstructionNet(nn.Module):
107 |     def __init__(self, n_dim=16, n_classes=10):
108 |         super(ReconstructionNet, self).__init__()
109 |         self.fc1 = nn.Linear(n_dim * n_classes, 512)
110 |         self.fc2 = nn.Linear(512, 1024)
111 |         self.fc3 = nn.Linear(1024, 784)
112 |         self.n_dim = n_dim
113 |         self.n_classes = n_classes
114 | 
115 |     def forward(self, x, target):
116 |         mask = Variable(torch.zeros((x.size()[0], self.n_classes)), requires_grad=False)
117 |         if next(self.parameters()).is_cuda:
118 |             mask = mask.cuda()
119 |         mask.scatter_(1, target.view(-1, 1), 1.)
120 |         mask = mask.unsqueeze(2)
121 |         x = x * mask
122 |         x = x.view(-1, self.n_dim * self.n_classes)
123 |         x = F.relu(self.fc1(x))
124 |         x = F.relu(self.fc2(x))
125 |         x = F.sigmoid(self.fc3(x))
126 |         return x
127 | 
128 | 
129 | class CapsNetWithReconstruction(nn.Module):
130 |     def __init__(self, capsnet, reconstruction_net):
131 |         super(CapsNetWithReconstruction, self).__init__()
132 |         self.capsnet = capsnet
133 |         self.reconstruction_net = reconstruction_net
134 | 
135 |     def forward(self, x, target):
136 |         x, probs = self.capsnet(x)
137 |         reconstruction = self.reconstruction_net(x, target)
138 |         return reconstruction, probs
139 | 
140 | 
141 | class MarginLoss(nn.Module):
142 |     def __init__(self, m_pos, m_neg, lambda_):
143 |         super(MarginLoss, self).__init__()
144 |         self.m_pos = m_pos
145 |         self.m_neg = m_neg
146 |         self.lambda_ = lambda_
147 | 
148 |     def forward(self, lengths, targets, size_average=True):
149 |         t = torch.zeros(lengths.size()).long()
150 |         if targets.is_cuda:
151 |             t = t.cuda()
152 |         t = t.scatter_(1, targets.data.view(-1, 1), 1)
153 |         targets = Variable(t)
154 |         losses = targets.float() * F.relu(self.m_pos - lengths).pow(2) + \
155 |                  self.lambda_ * (1. - targets.float()) * F.relu(lengths - self.m_neg).pow(2)
156 |         return losses.mean() if size_average else losses.sum()
157 | 
158 | 
159 | if __name__ == '__main__':
160 | 
161 |     import argparse
162 |     import torch.optim as optim
163 |     from torchvision import datasets, transforms
164 |     from torch.autograd import Variable
165 | 
166 |     # Training settings
167 |     parser = argparse.ArgumentParser(description='CapsNet with MNIST')
168 |     parser.add_argument('--batch-size', type=int, default=128, metavar='N',
169 |                         help='input batch size for training (default: 64)')
170 |     parser.add_argument('--test-batch-size', type=int, default=1000, metavar='N',
171 |                         help='input batch size for testing (default: 1000)')
172 |     parser.add_argument('--epochs', type=int, default=250, metavar='N',
173 |                         help='number of epochs to train (default: 10)')
174 |     parser.add_argument('--lr', type=float, default=0.001, metavar='LR',
175 |                         help='learning rate (default: 0.01)')
176 |     parser.add_argument('--no-cuda', action='store_true', default=False,
177 |                         help='disables CUDA training')
178 |     parser.add_argument('--seed', type=int, default=1, metavar='S',
179 |                         help='random seed (default: 1)')
180 |     parser.add_argument('--log-interval', type=int, default=10, metavar='N',
181 |                         help='how many batches to wait before logging training status')
182 |     parser.add_argument('--routing_iterations', type=int, default=3)
183 |     parser.add_argument('--with_reconstruction', action='store_true', default=False)
184 |     args = parser.parse_args()
185 |     args.cuda = not args.no_cuda and torch.cuda.is_available()
186 | 
187 |     torch.manual_seed(args.seed)
188 |     if args.cuda:
189 |         torch.cuda.manual_seed(args.seed)
190 | 
191 |     kwargs = {'num_workers': 1, 'pin_memory': True} if args.cuda else {}
192 | 
193 |     train_loader = torch.utils.data.DataLoader(
194 |         datasets.MNIST('../data', train=True, download=True,
195 |                        transform=transforms.Compose([
196 |                            transforms.Pad(2), transforms.RandomCrop(28),
197 |                            transforms.ToTensor()
198 |                        ])),
199 |         batch_size=args.batch_size, shuffle=True, **kwargs)
200 | 
201 |     test_loader = torch.utils.data.DataLoader(
202 |         datasets.MNIST('../data', train=False, transform=transforms.Compose([
203 |             transforms.ToTensor()
204 |         ])),
205 |         batch_size=args.test_batch_size, shuffle=False, **kwargs)
206 | 
207 |     model = CapsNet(args.routing_iterations)
208 | 
209 |     if args.with_reconstruction:
210 |         reconstruction_model = ReconstructionNet(16, 10)
211 |         reconstruction_alpha = 0.0005
212 |         model = CapsNetWithReconstruction(model, reconstruction_model)
213 | 
214 |     if args.cuda:
215 |         model.cuda()
216 | 
217 |     optimizer = optim.Adam(model.parameters(), lr=args.lr)
218 | 
219 |     scheduler = lr_scheduler.ReduceLROnPlateau(optimizer, verbose=True, patience=15, min_lr=1e-6)
220 | 
221 |     loss_fn = MarginLoss(0.9, 0.1, 0.5)
222 | 
223 | 
224 |     def train(epoch):
225 |         model.train()
226 |         for batch_idx, (data, target) in enumerate(train_loader):
227 |             if args.cuda:
228 |                 data, target = data.cuda(), target.cuda()
229 |             data, target = Variable(data), Variable(target, requires_grad=False)
230 |             optimizer.zero_grad()
231 |             if args.with_reconstruction:
232 |                 output, probs = model(data, target)
233 |                 reconstruction_loss = F.mse_loss(output, data.view(-1, 784))
234 |                 margin_loss = loss_fn(probs, target)
235 |                 loss = reconstruction_alpha * reconstruction_loss + margin_loss
236 |             else:
237 |                 output, probs = model(data)
238 |                 loss = loss_fn(probs, target)
239 |             loss.backward()
240 |             optimizer.step()
241 |             if batch_idx % args.log_interval == 0:
242 |                 print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
243 |                     epoch, batch_idx * len(data), len(train_loader.dataset),
244 |                            100. * batch_idx / len(train_loader), loss.data[0]))
245 | 
246 |     def test():
247 |         model.eval()
248 |         test_loss = 0
249 |         correct = 0
250 |         for data, target in test_loader:
251 |             if args.cuda:
252 |                 data, target = data.cuda(), target.cuda()
253 |             data, target = Variable(data, volatile=True), Variable(target)
254 | 
255 |             if args.with_reconstruction:
256 |                 output, probs = model(data, target)
257 |                 reconstruction_loss = F.mse_loss(output, data.view(-1, 784), size_average=False).data[0]
258 |                 test_loss += loss_fn(probs, target, size_average=False).data[0]
259 |                 test_loss += reconstruction_alpha * reconstruction_loss
260 |             else:
261 |                 output, probs = model(data)
262 |                 test_loss += loss_fn(probs, target, size_average=False).data[0]
263 | 
264 |             pred = probs.data.max(1, keepdim=True)[1]  # get the index of the max probability
265 |             correct += pred.eq(target.data.view_as(pred)).cpu().sum()
266 | 
267 |         test_loss /= len(test_loader.dataset)
268 |         print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
269 |             test_loss, correct, len(test_loader.dataset),
270 |             100. * correct / len(test_loader.dataset)))
271 |         return test_loss
272 | 
273 | 
274 |     for epoch in range(1, args.epochs + 1):
275 |         train(epoch)
276 |         test_loss = test()
277 |         scheduler.step(test_loss)
278 |         torch.save(model.state_dict(),
279 |                    '{:03d}_model_dict_{}routing_reconstruction{}.pth'.format(epoch, args.routing_iterations,
280 |                                                                              args.with_reconstruction))
281 | 


--------------------------------------------------------------------------------
/reconstruction_visualization.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |  "cells": [
  3 |   {
  4 |    "cell_type": "code",
  5 |    "execution_count": 1,
  6 |    "metadata": {},
  7 |    "outputs": [],
  8 |    "source": [
  9 |     "import torch\n",
 10 |     "from torch.autograd import Variable\n",
 11 |     "from torchvision.datasets import MNIST\n",
 12 |     "from torchvision.transforms import ToTensor"
 13 |    ]
 14 |   },
 15 |   {
 16 |    "cell_type": "code",
 17 |    "execution_count": 2,
 18 |    "metadata": {},
 19 |    "outputs": [],
 20 |    "source": [
 21 |     "from net import CapsNetWithReconstruction, CapsNet, ReconstructionNet"
 22 |    ]
 23 |   },
 24 |   {
 25 |    "cell_type": "markdown",
 26 |    "metadata": {},
 27 |    "source": [
 28 |     "# Load model"
 29 |    ]
 30 |   },
 31 |   {
 32 |    "cell_type": "code",
 33 |    "execution_count": 3,
 34 |    "metadata": {},
 35 |    "outputs": [],
 36 |    "source": [
 37 |     "# initialize network classes\n",
 38 |     "capsnet = CapsNet(3, 10)\n",
 39 |     "reconstructionnet = ReconstructionNet(16, 10)\n",
 40 |     "model = CapsNetWithReconstruction(capsnet, reconstructionnet)"
 41 |    ]
 42 |   },
 43 |   {
 44 |    "cell_type": "code",
 45 |    "execution_count": 4,
 46 |    "metadata": {},
 47 |    "outputs": [],
 48 |    "source": [
 49 |     "# Load trained model\n",
 50 |     "MODEL_PATH = '229_model_dict_3routing_reconstructionTrue.pth'\n",
 51 |     "model.load_state_dict(torch.load(MODEL_PATH))"
 52 |    ]
 53 |   },
 54 |   {
 55 |    "cell_type": "markdown",
 56 |    "metadata": {},
 57 |    "source": [
 58 |     "# Load MNIST test set"
 59 |    ]
 60 |   },
 61 |   {
 62 |    "cell_type": "code",
 63 |    "execution_count": 5,
 64 |    "metadata": {},
 65 |    "outputs": [],
 66 |    "source": [
 67 |     "dataset = MNIST('../data', train=False, transform=ToTensor())"
 68 |    ]
 69 |   },
 70 |   {
 71 |    "cell_type": "markdown",
 72 |    "metadata": {},
 73 |    "source": [
 74 |     "# Helper functions"
 75 |    ]
 76 |   },
 77 |   {
 78 |    "cell_type": "code",
 79 |    "execution_count": 6,
 80 |    "metadata": {},
 81 |    "outputs": [],
 82 |    "source": [
 83 |     "# (1x28x28 tensor input)\n",
 84 |     "def get_digit_caps(model, image):\n",
 85 |     "    input_ = Variable(image.unsqueeze(0), volatile=True)\n",
 86 |     "    digit_caps, probs = model.capsnet(input_)\n",
 87 |     "    return digit_caps\n",
 88 |     "\n",
 89 |     "# takes digit_caps output and target label\n",
 90 |     "def get_reconstruction(model, digit_caps, label):\n",
 91 |     "    target = Variable(torch.LongTensor([label]), volatile=True)\n",
 92 |     "    reconstruction = model.reconstruction_net(digit_caps, target)\n",
 93 |     "    return reconstruction.data.cpu().numpy()[0].reshape(28, 28)\n",
 94 |     "\n",
 95 |     "# create reconstructions with perturbed digit capsule\n",
 96 |     "def dimension_perturbation_reconstructions(model, digit_caps, label, dimension, dim_values):\n",
 97 |     "    reconstructions = []\n",
 98 |     "    for dim_value in dim_values:\n",
 99 |     "        digit_caps_perturbed = digit_caps.clone()\n",
100 |     "        digit_caps_perturbed[0, label, dimension] = dim_value\n",
101 |     "        reconstruction = get_reconstruction(model, digit_caps_perturbed, label)\n",
102 |     "        reconstructions.append(reconstruction)\n",
103 |     "    return reconstructions"
104 |    ]
105 |   },
106 |   {
107 |    "cell_type": "markdown",
108 |    "metadata": {},
109 |    "source": [
110 |     "# Visualizations"
111 |    ]
112 |   },
113 |   {
114 |    "cell_type": "code",
115 |    "execution_count": 7,
116 |    "metadata": {},
117 |    "outputs": [],
118 |    "source": [
119 |     "%matplotlib inline\n",
120 |     "import matplotlib.pyplot as plt"
121 |    ]
122 |   },
123 |   {
124 |    "cell_type": "markdown",
125 |    "metadata": {},
126 |    "source": [
127 |     "## Sample reconstructions"
128 |    ]
129 |   },
130 |   {
131 |    "cell_type": "code",
132 |    "execution_count": 8,
133 |    "metadata": {},
134 |    "outputs": [],
135 |    "source": [
136 |     "# Get reconstructions\n",
137 |     "images = []\n",
138 |     "reconstructions = []\n",
139 |     "for i in range(8):\n",
140 |     "    image_tensor, label = dataset[i]\n",
141 |     "    digit_caps = get_digit_caps(model, image_tensor)\n",
142 |     "    reconstruction = get_reconstruction(model, digit_caps, label)\n",
143 |     "    images.append(image_tensor.numpy()[0])\n",
144 |     "    reconstructions.append(reconstruction)"
145 |    ]
146 |   },
147 |   {
148 |    "cell_type": "code",
149 |    "execution_count": 9,
150 |    "metadata": {},
151 |    "outputs": [
152 |     {
153 |      "data": {
154 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAA58AAADqCAYAAAAlFvGTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/X9jR6j7CIUOdQAAAABJRU5ErkJggg==\n",
155 |       "text/plain": [
156 |        "<matplotlib.figure.Figure at 0x7f40574cba20>"
157 |       ]
158 |      },
159 |      "metadata": {},
160 |      "output_type": "display_data"
161 |     }
162 |    ],
163 |    "source": [
164 |     "# Plot reconstructions\n",
165 |     "fig, axs = plt.subplots(2, 8, figsize=(16, 4))\n",
166 |     "axs[0, 0].set_ylabel('Org image', size='large')\n",
167 |     "axs[1, 0].set_ylabel('Reconstruction', size='large')\n",
168 |     "for i in range(8):\n",
169 |     "    axs[0, i].imshow(images[i], cmap='gray')\n",
170 |     "    axs[1, i].imshow(reconstructions[i], cmap='gray')\n",
171 |     "    axs[0, i].set_yticks([])\n",
172 |     "    axs[0, i].set_xticks([])\n",
173 |     "    axs[1, i].set_yticks([])\n",
174 |     "    axs[1, i].set_xticks([])\n"
175 |    ]
176 |   },
177 |   {
178 |    "cell_type": "markdown",
179 |    "metadata": {},
180 |    "source": [
181 |     "## What the individual dimensions of a capsule represent\n",
182 |     "We can visualize what an individual dimension of a capsule represents by perturbing values of each dimension (sec. 5.1. of the paper, figure 4).  \n",
183 |     "Each row shows the reconstruction when one of the 16 dimensions in the DigitCaps representation is tweaked by intervals of 0.05 in the range [−0.25, 0.25]."
184 |    ]
185 |   },
186 |   {
187 |    "cell_type": "code",
188 |    "execution_count": 10,
189 |    "metadata": {},
190 |    "outputs": [],
191 |    "source": [
192 |     "digit, label = dataset[0]\n",
193 |     "perturbed_reconstructions = []\n",
194 |     "perturbation_values = [0.05*i for i in range(-5, 6)]\n",
195 |     "digit_caps = get_digit_caps(model, digit)\n",
196 |     "for dimension in range(16):\n",
197 |     "    perturbed_reconstructions.append(\n",
198 |     "        dimension_perturbation_reconstructions(model, digit_caps, label,\n",
199 |     "                                               dimension, perturbation_values)\n",
200 |     "    )"
201 |    ]
202 |   },
203 |   {
204 |    "cell_type": "code",
205 |    "execution_count": 11,
206 |    "metadata": {},
207 |    "outputs": [
208 |     {
209 |      "data": {
210 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7kAAAUuCAYAAAC/FPEBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzsnXegXFXV9n9Dk95DC6EFQkLvvQsI\nKFWsKPL6iiIKHyKCKCKgNBWlvWBDKQoIIr0ZmgRIaKFDgBASWiih96Lz/YHPPuvcu+/cSTJ3Wp7f\nP5msmTtz5pl99jlnr+esValWqxhjjDHGGGOMMd3ATK3eAGOMMcYYY4wxplH4ItcYY4wxxhhjTNfg\ni1xjjDHGGGOMMV2DL3KNMcYYY4wxxnQNvsg1xhhjjDHGGNM1+CLXGGOMMcYYY0zX4ItcY4wxxhhj\njDFdgy9yjTHGGGOMMcZ0Db7INcYYY4wxxhjTNfgi1xhjjDHGGGNM1zDLQH9ApVKpDvRndCBTqtXq\noGn9Y2uaxZo2HmvaeKxp47GmjceaNh5r2nisaeOxpo3HmjaeujR1Jrc1TGr1BnQh1rTxWNPGY00b\njzVtPNa08VjTxmNNG481bTzWtPHUpakvco0xxhhjjDHGdA2+yDXGGGOMMcYY0zX4ItcYY4wxxhhj\nTNfgi1xjjDHGGGOMMV2DL3KNMcYYY4wxxnQNvsg1xhhjjDHGGNM1+CLXGGOMMcYYY0zX4ItcY4wx\nxhhjjDFdgy9yjTHGGGOMMcZ0DbO0egNqUalUAKhWqy3eku7BmjYeaRqxvtNHTlNhbeunlo4Ra1o/\n1rTxWNPGU6+mEetbm6jp1J5LWds806NpDuvceE1zdIrOzuQaY4wxxhhjjOkafJFrjDHGGGOMMaZr\naKldOabUZ5ppptK/kE+H62/i63L85z//Kf0/vpce93zN1NCuqfr+NBVx+/V8zuIQqaWpnsvpXC+d\npOnMM8+cYvWO01q25p7/QqFp1L2WRvH9c+/XTvSnqYjbX0vT3HcXUT89/ve//13XZ+We7wRNpWW9\n+369mioWNZWWUdOcVt2i6bTs+7k5tufro6YfffRR6d++PqsW03N8G0j601T0t+/XGruKRQ0++OAD\nIK/ptFhO22nMTo+m8XU5TXvOz3E/l6YffvhhiuXOBWrRCZpKg1lnnTXFas1xUdPc79EzFjV9//33\nS//29Vk5cq/rJE1F3G9z5wezzDJL6d9cLGr6zjvvAPDee++lWL3XAd2iaW6cRv30N7PNNluK6Xk9\nF+dOaap/odA8p09Ov2Zo6kyuMcYYY4wxxpiuoSWZ3J6rAwBzzDEHAPPMM0+K6fmFF144xRZbbLFe\nMa3wzDfffCm20EILld5PK44AL7zwAgDPP/98iunxm2++mWJvvfUWAO+++26KvfbaawC88cYbKaYV\nt1aunOc0nXPOOYGypp/4xCcAGDRoUIrp8SKLLNLr/RZYYIEU0/PzzjsvUF4Vmzx5MgDPPvtsr5h0\nBHj99dd7/W1OUz3fytUzjau4siVN41jT83FMSlONV8iP50UXXbT0fnGsPfPMMwA89dRTKZbTVLrF\nFTXprH8hP047XdPZZ58dKI/dxRdfHCjG7ttvv52ek5ZPPvlkij333HNAed+XprlxGjXV8+2gadz3\n55prLgDmn3/+FNPzmhuh0FSaQfF7RJ2XXHJJABZccEGgrKm0HD9+fIppHoiaSreYoVBM2kKxD3SS\nptIFCk2XWGKJFJt77rkBGDx4cIottdRSQPF7xH16woQJADz22GMppvkgaird4jitpWku294scppq\nrOU0jccezQMah1Ach4YMGZJiyyyzTOn1UdMnnngCgEcffTTFpGk89rz66qtAeZxKSz0HxXzbDprG\njExunOr5GNO4W3rppVNMz0vH+FiaxvGncfrII4+kmDSN86R0y82nUVPNK+2mae4YldNUGkX9pPOy\nyy6bYnqs56KmGqdR09wx6pVXXgHKx31p+fLLL6eYNJ0eV8j0ktNU5/z9jVNpGvXTedNyyy2XYtJc\n80bc93VsGjduXIpJ07jvS7f4txqnU6ZMSbHcOG32+b+ytv2NU82nMaZj1PLLL59i0nTo0KEpJk11\nfJuWcSrdoqaKxXGq5xutqTO5xhhjjDHGGGO6hroyuZVKZRiwMjAP8CbwULVafaz2XxljjDHGGGOM\nMc2l5kVupVJZCvgbsDrwBPA6MC8wtFKp3Ad8sVqtPlXjLbIofR5ttLIcjRgxIsVWW201AIYNG5Zi\nK6ywQuk9oLA2KlUf31v23JgCl40m2mRk/ZCVGQobSLTd3HvvvQD885//TDFZ86IlutnWhVqarrzy\nyim26qqrAmVNZVnIaSrrExT2MD0Xi01IS+kYY1FT2ZWineHBBx8E4Jprrkmxp59+GihrmisWNJBI\nD1kNodB0pZVWSjFpqrEZH0dbrsZifD/9Xnpd/L6ydESbjDR98cUXU0y/Q7SQykJy9dVXp9ikSZN6\nfUarNI3jShbEuO+vssoqQFlTjVnpCIVdOY57jVN9VrQdaizGMan9+6WXXkoxaRrtdbLnXHXVVSkm\nq258XbM1lV0paipbbNRUYzZalFZccUWgsI7Fx9IxPpam8fvKohRv/5AFLI5daRrHn8ZkHKeyluU0\nbZbNTprGY4os3cOHD08x6Rs11fPxb/Xb1NI03qqgca9jIBR2rmjxkqZxLtbxKGoqi278jGZrKsui\n9lkobjOImupxtM1p7Mb9XJpGG15PTaOVU/NMnFP0fDxuydYZNdV8ETXVHBs/Q3/bSk1l78zNnXGc\n6rgVx6T0jXZRPa99ItoO9VnRhq/9O55fScs4N+oYdu2116aYzgXisaxVmsZjt+yacUxK02iZXWON\nNYCyfhqftTSN50Ma1/GWMukWz0Vzxek0N8Rzqfvuuw8oa6rfo1ma5goiSY9oQ85putZaawHlW0Kk\nabyloZamisW5Qt892pVraRrHqTSN+0KzNc3d/pG7hUPzQNR0nXXWAcq3g0nLWppGrTQm436y5ppr\nAmXt9bo4n2q+jddR99xzD9B4TfuzK/8ZGAUsXK1WV61Wq5tUq9XVgEX+Gz9zmj/ZGGOMMcYYY4xp\nMP3ZldcHtq9Wqx/EYLVafbtSqRwOvJL/s6knV1JaGZi4AjBx4kSgvCqlVYEY04qk3iNmbpRxiCsg\nWqmIKxvrrbceUKx6QLEaEousKFsRV3+ancnN6ZdbQdFKflzR13eJ+mn1KmZTpKlep2wrFNmcmGXT\n46jphhtuCMDaa6+dYlpdVqYMiqxa3PZW0Z+m0iPqp8xJrihETlOtkEUNVMAjrtLnCi1ttNFGAGy6\n6aYppgyd9hfIj9NmZx1z41TbEzN8WgmMK4LKnOQKwcW/laaaF2KhGY3ZqKkybjlN9S8U2aQ47nPj\ntNmaily7lP4KvD300EO9Xqfn4ziR+0BjSH8HRTGvqKl+g5iN0L6//vrrp5hcJpo/oMjwRE3jtjSD\nXHsJaRrHn/bpWOTp/vvvL70einEc308a6Xikv4NC05hh16p6HKfSUtpCoWnMrOv418pxmtv3tT1x\nntR+GzM3ypzEcRBX/IXGqcbT2LFj03PSNGbYldmJmuq4v/HGG6eYtIoOGunbruM0p2nMWMuVFv9W\n4zS2EFJmUa6Lu+++Oz2nY1Qcp8oExX1f43SzzTbr9R3iNrWDprkWSNI0HuM1F8YMrcZb/Fv9DlFT\n6SUn0J133pmeU3HJOE6laSwGuO666wJlTZXdiw4aaarfCpp/XpXb93Oaav+K7gJpE1vlSNPYVkjn\nmzqHuuuuu9Jzmv/iONX8ooJLUGi65ZZbppiuDeI41fvFc4FWaRr3X13vxLlRmkYXzO233w6Ux6SO\nazlNdY0Qx6nOffrTVNdPW221VYrJrZPTNBZabYSm/WVynwY+08dzOwBTbVU2xhhjjDHGGGMGiv4y\nud8FLqpUKgcC91Hck7sGHxei+uzAbp4xxhhjjDHGGFM/NS9yq9Xq9ZVKZSiwGx9f1C4CvAWcBVxc\nrVan1Pr7vpD9JKbUlfaPxV/GjBnT629zVhxZ7XJWrJz1RLaHmJbXzdPqXQhFoYoNNtggxfQ+0UoV\nU/6tQt83aip7SrSujB49utffSrdow5OmOWtrLU2jFrKDx36RsttEG6h+h2hvir9Nq9B3j3YaaZob\npzkbXrSB6nG0CdbSVOQ0jXYQ2cGjXVnW0WhvagdN9X2jprIWxnF6xx13AP1rqvlgejTN3aogu1S0\ngsm2FzWNPeqE9oVmFaDQOI1zoixqsfiLrEY5q3gsziWdc/t+znYmoqYaa7GIhcbk5ptvnmKyi8Y+\nvZqLoz2t2Zrq+8Y5UXaqqGnOnpiz39czTnNEDXLFU0Tc96Vl1FS2s/h+zUbfM2qqeTQW1JE1ub/b\nRBqhqcZptEfqfaNdWXbmXI/udtA0zom61SjelqCCTrm/jeM0VyBqWjWNBRa1ffFcSucFuR7drTyn\n0veMc6L2+Xj7jHqujhw5MsU0ZuM4baSm0cKsfSbe+qW+pvGcS8eteKxq9nya01TbH88FZIu96aab\nUkzbGI9Hejw9msp+H+22+p1VQAmKwk1xnMr62w6axrGmfT6eC+iWtVtvvbXXezRC04iKHj7++OMp\npvlIhe6guKUmatqzaCA0RtN+WwhVq9WXgT9M8ycYY4wxxhhjjDFNoq4+uY1GqwNxxSBXKjqXNaiV\nSaiXWitD8aZnPY6rzMo6qTgAFKuUzS42FalXU5Hb1kZomlsFikVltFIX22BoNT++rp00jdtQ7zjt\n+dz0kCssEIvKPPzww0C+/ZVW1qDIoLSbprkWEbX2/UaQK9QSs/PKfMRiHSLGpHN8v2at5Irciq40\njZmHZmqqx7lCNyq2AkWWIY5TuVHiXNIqTWOWq1Yrk4HSNOdkiJlkafrYY0Xberk84nyqbEmriqJB\noWkcJzkHkmiGpvpNY+GwBx54AChnP1VwMs6x0rTZhZEiOU01J/Wn30Brmsskq01IfF08l9I4aKWm\nufPDnCuoFZrG+VzHfbmeIvH8QJq2sohn7vwwOjp6vq6/2PRuR9yWXMHQmPWUqyAWnWuHcZobk7mM\n+UBrGpGmMZOsrO6//vWvFFO2NjefNnqctt5na4wxxhhjjDHGNAhf5BpjjDHGGGOM6RpaYleu1yrb\nbItaz8+U3SZawa6++mqgsIpAYWlrxfaKWpq2crugvG2yKUf9VAwr9omUXbmV2z5QdvlGEDVVwSYV\nbIGij2nsadgOduVO0VTWpKifLLXRHiaLTbtp2uwiGH2RG6fq0QeFfrfddluK5ezKzabWLR7tpKns\n4LGgoIqh3HLLLSkmO24rLeDtrGn8fGkaLYsqMnTzzTenmGzj7TZOW7k9kThONf7imNTtYFHTXC/3\nZtPOmsZtk3U5air9YtHWXI/sdtj3W73Pi7gdOvbEfV86R6u97ODxdpZWfZ+Buq1jesj1iY7HKJ3f\nyx4Oxe01jbYr13WRW6lU5gP2B9YE5o7PVavVbRu6RcYYY4wxxhhjzDRSbyb3QmBm4GKg993iDaDV\nKxDKfMTWLCpvHQtQ6ObpWFShXVaje9Iu2zP//POnx4MHDwbK2XGt5MaCKta0NirYA0W7gFjQRyuR\nscBXu6xG96RdNFWbFShahcTiPVqJjAWqahUjaiXtsj2xHYDasMQxqSI/sVBKrYJ5raRdtie2AlMr\nu1i4SfNAHLvWtDaxZY1aWcQsjc4B1OoQiuIurSw6185ETXW8ilrpHEAtTiBfcNKaFsQWOGphFXV+\n4okngPK5QM7BZU0LoqY6B4jHLY3PqGmukJ81LYiaqiVgPL9Ssbmoaa6YVyMccvVe5G4ALFytVj/o\n95XGGGOMMcYYY0yLqPci9xZgOHB/fy/sVLR6u/fee6eYGhZfccUVKTZhwgSgvNrgFZw8Wr3dY489\nUmzDDTcEyveRqK1Q9OK38h7HdkartzvvvHOKbb311kDRTgSs6dSglcZPfvKTKbbDDjsAxcp4fBzL\n81vTPLovdP3110+xnXbaCSi3YqilqefVMsrgrrLKKim2yy67AEUGDOCGG24Ayi06lHGwpmWUBRs6\ndGiK7bjjjkAxL0Bxb35sjdGuLo5WI02XXHLJFNtuu+2AwnkAcNVVVwHlFi7S1PNqmZzTcMsttwRg\nyJAhKXbdddcBZU3b1cXRaqRpHJObbropUJ4PRo0aBZTdMtLU47SMNNX1FMC6664LwIgRI1Lszjvv\nBMqayjnTaE3rvcjdC7iqUqncDrwQn6hWq0c1dIuMMcYYY4wxxphppN6L3KOBIcBEYN4Q99KQMcYY\nY4wxxpi2od6L3C8Cw6rV6uSB3JhmE2+E/vrXvw7AV7/61RSTRenGG29MMZUTt00hT7xh/3Of+xwA\n3/3ud1NMrRhUwAvg9ddfB6xpX8j6CbD99tsDcOihh6bYpEmTgHLZ+3Zob9HOxOI9sigddVRhSlHb\nhfPPPz/F1FrE4zRPLICyzjrrAPCLX/wixbSfn3DCCSmmIl4u4JEnaiq7129+85sUk23u//7v/1JM\nRbx8S02eWBRlueWWA8pjUsewM888M8Vy7S2saUHUVPbZX/7ylyk299wfN+WI86mKePmWmjxRUxVB\nPfbYY1NMNttLL700xdSWKd7+0S4t+tqBqOmgQYMAOPLII1NMhVHVKhSK8yvfUpMnaqox+cMf/jDF\nll12WaB8HaVxGm+zGShNZ+r/JQBMABrbvMgYY4wxxhhjjGkw9WZyzwEuq1Qqp9D7ntwbGr5VA4xW\nx1UECeDggw8Gylmz3//+90C5+IwzY3m0mrPaaqulmFbI5plnnhQ7++yzgXJbplY2fm9npOnyyy+f\nYscccwxQrOxCkS2Lhaca3VC7W5CmsSiK9FtmmWVS7IwzzgCKAglQFEbw6m0ZaaoWQVBkcVZcccUU\n09iNTeG1Om5N8yywwALpsTRdffXVU+zEE08Eys4YFZyypnliURSNyXgu8Lvf/Q6AkSNHppgKpDjT\nmEeZWoCf/vSnAGyxxRYpdtZZZwFw7bXXppjaMPqcKs+cc86ZHsu5pQJeUGTFY9bxtddeA6xpX6hw\nJ8D3v/99AHbbbbcU+8c//gEURdGgcHDZbZQnairX5pe//OUUu/zyywG45pprUmzKlClAc9xG9V7k\nfue//x7TI14Flmvc5hhjjDHGGGOMMdNOXRe51Wp12YHeEGOMMcYYY4wxZnqpN5PbVSy22GIAnHTS\nSSkmq93xxx+fYtdffz1g62c9LLTQQkBhn4PCEvrb3/42xVQkQdZP0zey1cUCHrIun3vuuSn2t7/9\nDSjfxG/yyAIWi0ypH7ZsNVDY62KvUVuU8siudMghh6TY2muvDRRzKMAf/vAHoNwbz5rmUVHEfffd\nN8U22mgjAMaMGZNiKjilol5gTftCfW+/9rWvpZh6Y99zzz0pdvLJJwNF8T6wTbkvVKRr9913T7FP\nf/rTADz00EMppmJpKjQH1rQvdMvcZz7zmRT77Gc/C8Bjjz2WYr/61a+AotAc2KbcF9J02223TTFZ\naidMmJBiOtd69tlnU8w9xvNI08033zzF9tprL6Ao1gXFOFWxKWhuj/E+L3Irlcoj1Wp1xH8fP00f\n7YKq1epSA7RtxhhjjDHGGGPMVFErk7t3ePyVgd6QgSa2tlFBpFgk6bbbbgPKWTNnG2sTi3T94Ac/\nAIpsA8ADDzwAlEu0O9tYm9ja5tvf/jZQXn3USu6Pf/zjFHvnnXeatHWdSWzDssceewDlYhNaYYxl\n75Vt9OptnqjpTjvtBJQzZMouRE2VbbSmeWIrhq222gooZ3JVAOVHP/pRr5g1zRM11bFJBWegKH50\n+OGHp5iyjc405omayrERx6SOR7mWbNY0T9RUheWOOOKIFJObUIXSoGhr5extnqjpSiutBMDPf/7z\nXq+Lbe50LuBiqHmipsOGDQPg6KOPTjGdv0aXrLK6rXLE9nmRW61WbwmP/9XX64wxxhhjjDHGmHah\nll35qL6ei1Sr1cP7f5UxxhhjjDHGGDPw1LIrDwmPZwc+C9wJTAKWAtYDLhq4TWsMSq/Hm/h1w3ks\nLKE+ZOozZvpGmm655ZYpts8++wDw9ttvp9hPfvIToFxswuSRphtssEGKHXjggUDZNn/ssccChVXJ\n9E2ud7PGZESFZiZOnJhitn/mkabDhw9PMVnA4i0h6jMcC6VY0zzSdOjQoSkmC91cc82VYmeeeSZQ\n3AYC1rQvpOlSSxUlQ3Qr0nzzzZdi0nTs2LEpZkttHmk6ePDgFDvhhBMAGDRoUIqpEOIdd9yRYrbU\n5pGmKoYKhaZLLLFEil122WUA3HrrrSlmTfPU6tse54PrrrsOKPcYt005jzRVgVkozkWXW67oInvz\nzTcDcOONN6ZYqwv31rIr/48eVyqV84EvVavVi0JsN+BzA7t5xhhjjDHGGGNM/dTbQmh7YI8escuA\nPzd2cxpHz9VxrTpA0Z7hvPPOS7HRo0cDXhmvRc/V8XjDvjIOahEEMHLkSMCa1kKaatU2ajr//PMD\ncNNNN6XYJZdcAjjbUIueK7nHHXdcek4ZhzvvvDPF/vrXvwJeGa9Fz5XcWFRGY3fcuHEppkxuq1dx\n2xlpusACCwBw2GGHpeeWXnppoOwuUCs2F0TsG2mqbO1BBx2UnlP7teiCOf300wEXRKyFNJ1nnnkA\n2G+//dJzK664IgAvvvhiiknT6OoyZaTp3HPPDcA3v/nN9JyKJKmoHBT7/ptvvtmsTew4pKnOReXY\nhMLNFVut/f73vwfKjk5TRprOMcccQNHKCmCNNdYAymNSzpgpU6Y0aQv7Z6b+XwLAeOA7PWLfBp5o\n7OYYY4wxxhhjjDHTTr2Z3G8AF1cqlYOBZ4HBwEfAbjX/yhhjjDHGGGOMaSJ1XeRWq9V7KpXKCsAG\nwBLAZGB0tVptWy/anHPOCRS9RuNN/OPHjwfKPbNsq+sfWRb22msvoHwTv3phxcI+toD1z+yzzw7A\n5z//eaB8E79sdbEHofq3mr7R7Qg77rgjUC6SpH6NcZy62Fz/zDbbbABsvfXWAKy88srpOVmTYl9H\n6Wz6RoW61L911VVXTc/Jqhj7Yj799NNN3LrORJquueaaQLnonDRVERoozgVM38wyy8enibLRrrLK\nKuk5WT1/85vfpNhDDz3UxK3rTKSpbqeTtlBo+oc//CHF7rrrLsC3ftVCPVqXXHJJIK9pvEVx1KhR\ngDWthTRVYbQVVlghPSdNVRQNimJe7XQ7Xb2ZXP57QTtqALfFGGOMMcYYY4yZLuq+yO0EZpqpuMVY\n2ZtNN90UKLexOfroowGvjNdD1FRZxs022wwo31yusvexZYjJEzUdMuTjTl05Tf/4xz8CcP/99zdx\n6zqTqKlWHTfeeGOgXMBD7S3GjBmTYl7JzRM1XXjhhYGixVUs1qHCXTfccEOKWdM8UVMVnFprrbWA\nsqYqMHfllVemWDutjrcTUVMVR5LTIO77yjJcfPHFKeZic3lUcAaKQj4q3PXyyy+n58466ywALrzw\nwhSzKy5P1FQOLrnh4jj9y1/+UvoX4P3332/GJnYcUVM5uBZffHGgPJ+ee+65QDFeAd59991mbGLH\nETWVM0bH/1i4S+dS+hfa02lYb+EpY4wxxhhjjDGm7fFFrjHGGGOMMcaYrqGr7MoqNgWFBUw3Tt9x\nxx3pOfUdtVWpf2SrgaIwiiwMd999d3pOVjBr2j+y1UDRZ1CFfe6999703LXXXgvY/lUP0g9g2WWX\nBYpCadHuLfun7V/9o/0cClu9NH3ggQfSc1dddRVg+1c9qOAMFLY62UFjwZ6rr74agHfeeaeJW9eZ\n6BgPha1Omj7yyCPpOfVtb0dLXbsRx6n6taun6+OPP56e0y0K0cZo8sRxqj7OGqcTJkxIz91+++2A\n+7fWQ9RU41O3LDzzzDPpOR2v4u1gvqUmT7z9Q+NTmsaCkk888XEH2dgjux01dSbXGGOMMcYYY0zX\n0BWZXK08qHQ4FG0ZtLrz7LPPpufeeOONJm5dZyJNlW0A2HDDDYGiYIrK2kPRhqUdV3LaBWm6yCKL\npJgK+QwaNAgoF+6KxShMHmm60EILpdh6660HFGN37Nix6Tlr2j/SVPs5wNprrw0UGd2HH344Paei\nft73+0bFPJTBgaK9zTLLLAOUsznKOFjTvpGmyjJA0TZErS4mT56cnrOm/SNNlcGBoojniBEjgPIc\nqsfWtG9ymmp8ag54++2303PS1IXm+kaaylkEhYNr9dVXB+Cjjz5Kzykrbk37RppG96auqdZYYw2g\n7PCQe6Pd3ZvO5BpjjDHGGGOM6Rp8kWuMMcYYY4wxpmvoWLty7OU077zzArDFFlukmGwgst69+eab\nzdu4DiVqKpu3LMpQ2ECkaSzgob+N72ELU77foIqiQTFOVUDBRVH6J2oqu9Iqq6ySYhqnsta4eE//\n5Ho4Dhs2LMVUdE4FvnJFprzvl8lpKksdFL1cVYguVwwtvofJ98WUhR4KTVWEMloWc+9h8poOHjw4\nxWRTli089snNvYcp66E5c9FFF00xFZxUUa84n+r8KhYAMvn+rfE2JVnAdZtNtNXr/CpqqvebkY9V\nOU0XXHDBFBs6dChQjNN4LtUpmnovMsYYY4wxxhjTNXRsJjeuHmjVNq6GPfnkkwB88MEHANx4443p\nuffee68Zm9hxRE2VeYgFEVQYRf9ef/316Tm9rp1WcNqB3IpuXA3TOFVhtKip3AfWtExu9TG376sg\nklqHQFEgzZSJmuYy4NJU+qllGLj4TF9ETbXqHTWdOHEiULg34jFKRZJcKKVMTtN4PH/qqaeAIis+\nevTo9JzaX1jTvtE5QNRUrVikW2wdqMJe7V58ppVozOY0FQ8++GB6rDGccyGYj5GmOr8HeO6554Bi\nXhg/fnx6ruf1APh41RfRUaT9W+dZGptQ6Btf346aOpNrjDHGGGOMMaZr8EWuMcYYY4wxxpiuoTLQ\n6eVKpTIgHxCttUqlxz5kKqAgi0i03X744YdAS1Prd1er1XWm9Y+boaksi7KCQ+8CKdF6Z037fN/0\nWJrG3m7SVPpZ07reNz2WprG3W09No01MdiVr2ut902NpKh2hsNrLQhctSta0z/dNj2Whk47xsWyg\nUVON3RZaaztGUx3/obem0Z4m9Eg9AAAgAElEQVRoTft83/Q4p6kea/+WjlDoa017vW96rPOqWppG\na7I17fN902NpGvu2Rn2hrKnGbAtt9R2jqeYA6K1pHJO5cdrkc4C6NHUm1xhjjDHGGGNM19CMTO5L\nwKQB/ZDOY+lqtTpoWv/Ymmaxpo3HmjYea9p4rGnjsaaNx5o2HmvaeKxp47GmjacuTQf8ItcYY4wx\nxhhjjGkWtisbY4wxxhhjjOkafJFrjDHGGGOMMaZr8EWuMcYYY4wxxpiuwRe5xhhjjDHGGGO6Bl/k\nGmOMMcYYY4zpGnyRa4wxxhhjjDGma/BFrjHGGGOMMcaYrsEXucYYY4wxxhhjugZf5BpjjDHGGGOM\n6Rp8kWuMMcYYY4wxpmvwRa4xxhhjjDHGmK7BF7nGGGOMMcYYY7qGWQb6AyqVSnWgP6MDmVKtVgdN\n6x9b0yzWtPFY08ZjTRuPNW081rTxWNPGY00bjzVtPNa08dSlqTO5rWFSqzegC7GmjceaNh5r2nis\naeOxpo3HmjYea9p4rGnjsaaNpy5NfZFrjDHGGGOMMaZr8EWuMcYYY4wxxpiuwRe5xhhjjDHGGGO6\nBl/kGmOMMcYYY4zpGnyRa4wxxhhjjDGma/BFrjHGGGOMMcaYrsEXucYYY4wxxhhjugZf5BpjjDHG\nGGOM6Rp8kWuMMcYYY4wxpmvwRa4xxhhjjDHGmK7BF7nGGGOMMcYYY7oGX+QaY4wxxhhjjOkafJFr\njDHGGGOMMaZrmKXVGyAqlUpdz+nxTDMV1+d6HF+n2Mwzz9zr9bPMMkuv1//73/8G4L333kux999/\nH4D//Oc/U/NVWkotHfv7m6iRYtVqNcXi457/r/dz9Tcziqa5v43fvaem00Pusxr5/gNFbqz1fK4/\nOuF7tjOdOnaMMcYYY3I4k2uMMcYYY4wxpmtoSSZX2dVchlZZ1vh40KBBKbbEEksAsPTSS6fYPPPM\nA8DgwYNTbPjw4QDMO++8ACy00ELpuZixFMrajho1KsVOPfVUAJ5++ukUa9fshrTKZbhzms4///wp\nJt2WXHLJFJt77rkBWGCBBXq9TprON9986bl3330XgJdeeinFXnnlFQAmTZqUYnfeeScAEydOTLGY\nPW8nplbTqMfCCy8MlHWWW0BaxZiYbbbZ0mNlfOPrpVUch3qdxjDAa6+9BsA777zT52e1Au37Ee37\n8TlpOsccc6SYxmTUXnpEjT744AOg+L7xe9fKFud+0/h7aPvieJW+H330UYo126VQrwtG2z/rrLOm\n2Oyzzw6Ux7j0i+NJ3y/33WppmnPXxM//xCc+Udo2KH6vqKk0j7FWkdM7N3ajph9++CFQHouNGCc5\nF44+X9pCoXnczjfeeAMo/87tdHzLjZ3IQLuC+js/0dwQ54i3334bKPahuJ3tRqvdG7l5X+M0apqb\nY9tV03Yhp2mcd3W87O/YaApyc2ycD3LOT2tam6ipxmwcu5pHp2eOdybXGGOMMcYYY0zXMF0XuZVK\nZalGbYgxxhhjjDHGGDO9TLNduVKpfAJ4EujtP+z/b4FyqlpWClkSAZZddlkAttxyyxTbbLPNSu8B\n8MILLwCF/QrgueeeA+CBBx4AyrZNWXDXWWedFFtxxRUB2HDDDVPsoosuAsp25XYlZ6+U1TNatUeM\nGAHAVlttlWL6ztFum7NnRfsblC1Zssfq30i0ksryfMYZZ6RYu9qVNSajZVZWbdmRodB0gw026BWL\n4zmH3lvvO9dcc6XnZM96/vnnU+ypp54C4K233koxWWIee+yxFDv//PMBePLJJ1NMlo9WWmhkRYnj\nVN853pagfX+VVVZJseWXXx4oLLaQLxjX01IeNdXjOPfIYvj666+nmPSNY/7FF18ECss9wNixYwF4\n+eWXU6zZBev0XXJ2qjj+NA8MHTo0xYYNGwYUt3xAsf36vlDs17oFIY4//R4rrLBCii211Mfrn7Lp\nArz66qtAed8ZMmQIULbS6fYGaQswevTo0ufDwOqbs3Lm5kTtv3HsapzGeUNj7Nlnn00xfReNuzif\n6rPi7SKrr746UNZPv8uCCy6YYuuvvz5QzEFQjGMdFwFOOeUUAO66664Ui9blZpKzrUX9FllkEaA8\nb2gMxmOO5gGNu9xcF38/3f6kcQjF7xHnjfXWWw+ArbfeOsX0N3Ff+OUvfwnADTfckGI692jlvJuz\nBuduVZBucb/VHFtr++P+orlkscUWSzH9LvH30zj9whe+kGKaj+J8fvTRRwNw7bXXptibb77Z7zY1\ni5ytPcYacdzV/hvnA50fRE3XXXddAPbZZ58UW3nllYHy/HLEEUcAcPnll6eYzp/bQdNmoHlA515Q\njPU4R6y55poAHHTQQSm22mqrAeX95Ec/+hEAV1xxRYrNaJpqLolzp7571HSNNdYA4Cc/+UmKSdM4\nTr/3ve8BcM0116TY1O77NS9yK5XKZjWe/kSN54wxxhhjjDHGmKbTXyb3JmAy0NAl89wKfK1iI3FV\n78EHHwTgoYceSjEVi5o8eXKK9cym5NrdxOJVJ510ElDO3GhFtxNWYWoV4Yirilq9jatXyrTGzI0y\n4HGVXxntKVOmAOUVdP1+sXjVF7/4RaCciV911VWBfAGidiPXfkrjI2a9F110UaBc+CyuZAmtQEXH\nwZxzzgkUv0H8rZR1idlYrRxqdRZgpZVWAorsJ8DIkSMBmDBhQoq1wziWpnFVT9nGxRdfPMW0oq/v\nBrDMMssA5UyuvlP8jbTard9IGsfXxRXYnllKKDKLMcOp3y0W61D2PP5ts3XOOWOUBdPYhGLMbLTR\nRimmFdVYIK1n+7X4fsoaxCybfo/4m0qjqIvm5/hZyi7HeV9uhbiie++99/b63s0g7o+57Lgy1tts\ns02KbbfddkA566LvEseGvrPGVRzXysLGrK1+l+hK0v4d/1bzUIzpuBB/j0ceeQSA+++/P8WancnN\nZce130ZnzJe+9CWgrKmKHGpeheL7Pf7440B5X/3GN74BwMYbb5xiuWJy2qfjmJTTKzevx7nkwAMP\nBMrjNf5ezSSOXe23MWO98847A+X9Ucf46EyRU077ZTzH+O53vwsUGRco5s64/z788MNAOestZ0Kc\nY7XN8TOOOeYYoOxUimO2VWhb4/FF4yTue9IvjgNpkxsbu+22GwB/+tOfUiynqcZY3Kc1tnOaxrnn\n17/+NVB2ltx0003Z79kK4rFH56pxPNcqnKVYfL1cAzfeeGOKaZ+I+6/Od6PjRe6NeM6X47TTTgPK\nY/yyyy6r+TfNJHcsi9RybCgW30OOsHgtJk3j3KlxGs9FdYyM1yE55PjU3A1wwQUX1PybnvR3kTsJ\n2KNard7W84lKpTI78PZUfZoxxhhjjDHGGDOA9Fd46i5gnT6e+w/wVGM3xxhjjDHGGGOMmXb6y+R+\nua8nqtXqB8CyfT1fi562AijsKTE2fvx4IF9sRJZZmPrCRfr8aG+WDSTaQeLz7U6uj6VisfCTrF3S\nFgoLwjPPPJNisi5HK0c9NsxYvEc2BtmioLDdREtHu5LTTxa5aDOStStaqGS3UbEdKMZT/O76DI37\nnN7RWiJ75FFHHZVisnyquA0Uv0M79MaN5PZ9bXcs8CYdYkx2oWhlkr0yFt9QwZOefZ0j0Y6nsR4L\npWyxxRalz4yfEWM9f79WkCt2IuJ40jwZx67m0bhvy4ocbXgag7LhxUJRuVsPtM/E309za7xNRNbf\nOG+piFO0ROf6Qw8ktfr8xlta9F022WSTXrFoGcz1VexZIC3Xp7nWtsXH0car3y3a16VbtFpL81b2\ndc5pKlt9tMAqFsdE7v00FhXrb5yK+PnSKFpOcwXrRPwMvS5+VqtuX4hjSLeynHjiiSm29tprA+Xv\nWQt99/j6nB4izsk6RsUCadI5N2/F95WmcX9q1a03cVv1XY499tgU+/SnPw2Ube3av3L7o36j3C01\nOeI41fEt7vu1NI0x7UfxFqt2uJ1JOsR9f8899wTKx10d1+K+J9303eJvUKuXfHwP3RIV/zaOu1ro\ntxk+fHiKXXrppXX97UCi77f77run2AEHHACUx47OC+L+rX1Yx+R6NY1zjzSN43pqNdVcBfC3v/2t\nrr9N21LryWq1+mGt540xxhhjjDHGmHZimlsITQ+1bmzuLxPZyDYocQVHqy9xlSAWCGp3cqvy0i9m\np7VaE7PUub+dVn3je6gIVcwe3HHHHaXtaGdy7RSkZczGKrurQi6QdyZI06nVNv5+ykCqpQYUq20T\nJ05MMRWUaHaWpj9yjgNpFDPRKvwQtcqN01rzQa0MZ0SrhSpOAUVxpvi32k4VYIEiUx8LLTR7RTyX\nHdeYjcXh5KyIxUa0Ot5f+yEVQFJ7HK3OQlG4JhbruPLKK4FyJlfFJuJKdy5roex8LDTTDi1DemZe\nofjdo6ZaHY9ZR61cxxVsPa/3i2Mt9z011mL7NbkQ9t133xSLc0NP4lyiubhZLdxy+6EynjGjoLZh\nsS2T9tH4Hjp+x6ypfqN6933pfPfdd6fYX//6VwC+8pWvpFgsqFgLHQNa6QKTBrFombI4sW1ibt/v\nLwM4NajgEsBxxx0HwE477ZRi0dFRCx3zWlV8DgoNYjbxm9/8JlAURYMi0zWtmvVHPO9Q+5VYSE0F\nG/tD+/xVV13VwK2bNmJ2cP/99wfgsMMOS7Hcvt9I4pz4/e9/HyjrqNY2/aFj7tlnn93ArZs2YsZf\nxeFUwA3KLouBIJ6L6NgUXRy//e1v63ofzc+nnnrqNG9Lf/fkGmOMMcYYY4wxHYMvco0xxhhjjDHG\ndA0tsSvXS86G10iibU7WntjXqh1uxK+XWtsan5tWy2y9RCvfDjvs0Ouz/v73vwPtZ6PNkeu3Vstq\n31+sESy33HJA0d8Uiu285JJLUqxdC3tpW3OW46hzzobcSJ1znxWLcKhnb7RCPv/88wDcdlvRUU02\n2laOZ2kabYeyYMW5U9sarbW5/ne5vrs9C/rk9olo+5KNNxb9Ut/sXPGK+LfqhT5mzJhe36dZ5MaV\ntIz71rhx44DilgUoLI3RgquCcdEGp0JnsoXHuVOfL9s3wFlnnQWUbf2aB2Khlhwan0888USK3XPP\nPaXvNdDUGmvR7i8L9i233JJi6hscrdiyv8Xx1NPCnLM4xuKIGpPxs2Qp33XXXev6XnFsyqrY7PEa\n0X4bb8eSVvEWLFkac4WkptYaGveXk08+GYAf/vCHKab3Ux9ZKG5f6O/9zj33XKC1tzhJl3i7Qezj\nLHJjvBGMHTsWKPc41z4TbzeI/URrMWrUKKC1t+RJ0zhPap+Lc+FA2ZRlg19iiSVSTJbjtdZaK8WO\nOOKIut5PvbljUdxmI62iNfhb3/oWULuYYaPQrUZxP5Gm9d7yEdFtHzr3mhbq+taVSmU+YH9gTWDu\n+Fy1Wt12mj/dGGOMMcYYY4xpIPVe2l8IzAxcDLzbz2s7htiGRavjsXhQNzLQ2emYYVSRi6uvvjrF\nYpuiTqGZmdocsUjAoYceCpQLYNx3330AXH755SkWMyPtSK6gVH9Z24FC7S1U9AKKYkoxa3bFFVcA\n5YJIyoK1g+sjV8wrVxBrWra11t/UKii0+uqrp9guu+wClDNH2uZYqEfFUJTRi69rNvF75wqk6XFu\nXouZcO3DsSCIsn2xsKKQprHonTSI7Uakaa7YVNx2ZZ+jUym25msG+k5xuzQ+Y4G0G264ASiyV1Bk\nu+P3VFucWMBIxc9UpChmVPX40UcfTTFlweIYVrYxFmnKoe8Rf/u77rqr13dsNhonsX2dMszRcaB9\nc+jQoSkmfeMYUwZ1woQJpf9DsY+qCBIUTpGogebTPfbYo67vEPeJCy+8sPS9WoG+S8x8XnDBBUB5\nTMppFR0syqjn3EP6ntFNodjee++dYpoTo6bKzP30pz+t6zvEz1DxulaeJ+i7RGfMddddB+QLG9ab\n3c2dTyh2yimnpNhBBx3U63V63+OPP36qvgMUjoN2OPeK2f0HHngAKNxEUByPpidLru9+6623pthm\nm21Wei5S7ziNaNzH4+DUUu9F7gbAwv/tjWuMMcYYY4wxxrQl9V7k3gIMB+4fwG1pGvKLb7/99il2\n+umnA9O3YjAjo6zFSSedlGJqkXHkkUemWDusck0PrVihX3fdddPjbbbZBiivpmvVMWa+OomBvk88\nR8yo/exnPwNg5ZVX7vU63bcIxT2RcTW/WfczTi3NzI7rfWPmUvUO/vCHP6SY7nOOq8caxzfddFOK\nyfnRyvvKa92Tm2u11p+2ys7UWjmv1QYLijH71a9+NcXUxiTeOy7iXKsWOfH3aPZcnPt+0lL3csXH\nMRMpoh65+0d7jvv+2uLpb2O7ImW54r2+OZQtOfzww1OsHdoO6jvHbI7uHVc2FgpHRWx1FR8LZdk1\nhuNvpfOlXJY13hOsOTbeK5hDv1F0HMTWeK1C3y8ed5Vliq2NVlhhBaC4hxyKeS/eC37zzTcDRTu6\nONb1uuhCyN3ru8UWW5Q+sz+efPLJ9DjOt61C3ylqqtYy0S31qU99Cihai0HhZIuaqn6BMoux1Z/G\nfTym5OYD/VbKSPaH6lxA4ZZopYtDnx3dRkcffTRQ3laNnTjvKVMe92+5aUaOHAkU9x0D/Otf/wLK\n55257y5XSDxu1SKO+1//+td9vm+91HuRuxdwVaVSuR14IT5RrVaPyv6FMcYYY4wxxhjTZOq9yD0a\nGAJMBOYN8dbfiGaMMcYYY4wxxvyXei9yvwgMq1ark/t9ZQegglMxBR4ttWbqWXXVVQFYb731Ukx2\nhm4v5jVQyDp23HHHpZgKBqg4CxTWrk5oyzRQLRbqRRbHz3zmMyn2+c9/Hijb61RMRjY7gEmTJgHt\nZ1HOadpMC3jO8nneeecB5aJAel20Q6lNkNqOQNEuoF3H8/RYwKf29dFWr8JoschPzlIrG7LGKxSW\n2lh4qJ3I/dbTMnZ7ziu594j7+WqrrQbAn/70pxTTbQu5OSpavK+//nqgXPCvHQrQ5Vov6nG0Ak7t\n7QDSI2cBj3Z5WUl/9KMfpZiKKMVbGnKoGNqPf/zjFGunW5xyLdmitXb8+PFAudimqGWdzx1T4vjT\nmI3F0FT4KnerQkSW8tjSqZXtmHoSf1/N/So2BsX+FQtP9SzcBb3bdsVbD3PHyFwLI51XxTkih94v\n3v7R7EJ+tYia6laFQw45JMV6tq+D4vaGWEhPdvBcm8BaLeFisdR//OMfQPlYVgtZ+aFsOZ9Was84\nBRMA36xqjDHGGGOMMaatqTeTew5wWaVSOYXe9+TekP+T9iKuAu25554AXHPNNSnWqUV7WklcwTnh\nhBOA8grhYYcdBriY17TyhS98ASg3JlcRkFjsJBYZ6BRalfEYMmQIAL/61a9STIXo4qrwz3/+c6Bc\nHr+dMgo5WqWpCkv8+c9/TrGVVloJKGdutAo8evToFPv+978PwMMPP5xi7apzM/VVJuHb3/52iqkF\nQ644UNRMBWYOPvjgFLvzzjuB9s2ORwZKZ50DqJANFJmY2AZGYzZuh45hDz74YIr94Ac/ANp3/u1P\nx3rbguUyuEJZxFj8SIW7oqsrnn/1/Pw472pefvrpp2tue6vItaWJ5OauaS02t/DCC6fYgQceCMB3\nvvOdFKtVGC1um/Z9OQ/6+tx2QPrlWirV6/7KFV3U30bNNt98cwD++Mc/ptiiiy5a12dMmTIFgN//\n/vcp1m4OLyHnVDwPf/nll4Hy8bmWUyPXKlGvj1lvtdCSywDyBT1zKJOsYlPQmHOBei9ytWcd0yNe\nBZab7q0wxhhjjDHGGGMaQF0XudVqddmB3hBjjDHGGGOMMWZ6qTeT2/GsscYa6bH6w8lOC+1r32hn\nBg8enB6rr9ioUaNS7I477mj6NnU6sRCAih5pvEJRIEV9L03/xCIIsicutdRSKSZ7zt///vcUU8+7\nnsUszMdETXWrwtZbb51iskFF65P6Se67774ppqItnWCjHWiiptIoFkSULTwiG1nsfyqLc7Tat+s4\nnp7jbj32xajp5z73OQBOPPHEFFtooYX6fK9cn+FoH1eRoU49d8h9Z8VyRexEtCduvPHGQNHfFGD5\n5ZcH8kWm4nvJ5h2Lfml+btdbFiK1Cu9EZOmOz/W0tsbiUbJ3RtvmpptuCuRt3xG9b+yJq9sWYp/U\nTiI3JrVfx5jmuFwxNBVAjOf8X/ziF4Hy+VWOXO/Zk046CSgXaeqkeSCnqY4vcb9VcTodn+NzstN/\n/etfTzEVN5t33tiEp2/ifnDjjTcCcNddd6VYI84L+txjKpXKI9VqdcR/Hz9NH+2CqtXqUrm4McYY\nY4wxxhjTbGotC+0dHn9loDdkoDn66KPTY7VWeOCBB1q1OV3BMccUt2hrRShmHto1e9DO7LPPPumx\niiDkWoF0wkp3q9GY3GOPPVJsk0026fU6ZRgPOOCAFIvFUEyBVnK/9a1vpdhXv/pVoJzh0ar2xIkT\nU+x///d/AXj00Ud7vW5GRONT2Zmvfe1r6Tm1uVNRtEjUbPLkj7v6xTF+zz33AN2VHc9lHmqhTM8u\nu+ySYqeeeioA888/f82/lW7SEYo2Y88991yv17UruSIxuRYquVj8W2VbVPDsk5/8ZHpOhXdi4a5a\nv1FsH6asmgpVQZEta9d5IW5XLkOrcRczXmqdEs+HVGRHz8XjkjRdcsklU6ze4lUqMvU///M/KSbH\nQbsWRorbr2NILsMYXRnah2N2WsWUlJmN7ZakqQojQf/trMSrr74KFEUSoWhx1K7nCVFT6RYz22qb\nFAtxqShnLMKrsaP3UKtQgNNPPx0oF5bqr52V0DxwzjnnpJjmAxX1ahR9XuRWq9VbwuN/NfRTjTHG\nGGOMMcaYAaCWXfmoet6gWq0e3v+rjDHGGGOMMcaYgaeWXXlIeDw78FngTmASsBSwHnDRwG1aY5Ad\nJKbZdUO/+7dOG7KP7Ljjjimm/oFjxoxpyTZ1OvPNNx8A3/3ud1NMlo4f//jHKSbrjOkfWZOOO+64\nFJOVSdYmKGyir7zyShO3rrOQDUn7fLz9Q/NBtEjJcvSlL30pxWKP0W6mlrUwWuQ0FtX7MhZFydmU\nxeuvv54e77zzzgCMHTt22ja2jcnZaKN+snnnNN1774/vtoq9gmvZlOPYVdGeLbfcMsVi//dOJmet\njYV3VDAm2oql8ze+8Q2gXDgu9nKthayysUjN3/72N6D9bd99oTETb9PQfhv7rS677MfNSWLvX9my\nZSvWrRxQv6bivvvuS4+32Wab0vvH7ewENE6ijVZarrLKKim25pprAoU9G4r99itf+fjuynj7x6BB\ng6ZqO+L+rn6648aN67WdnaCttjUWg1Jfa303KPpaqwAUFMd2Fen68pe/nJ6bWk2jXV7zcyz2qTmn\n0ZrWsisnU3+lUjkf+FK1Wr0oxHYDPtfQrTHGGGOMMcYYY6aDelsIbQ/s0SN2GfDnxm5O41EGN65+\nn3/++a3anK5g1113BcotLU4++WTAxaamFq2sK+Ow2GKLpedU+OSSSy5p/oZ1MFpNP/fcc4HyiqPc\nG8cff3yKPfTQQ03cus4htqtYf/31gaKAR2x1JWIRDq2mz8htxGLWTNmeuH8re/O9730PyGsaUZEV\nZTGgXNirm1G2NmZ4pKWyElAUiNptt92AfNulHLEYms4Zuqm4X67djbSMGTIV64lzptoErbXWWkD/\nLVdEzNB++tOfBuDaa6+d6m1vV3KtahZffHGgGIcAa6+9dq+/VUsbZXljUaV6UWZx3XXXTbFOHLMx\nc6fHKnIGxfiMxQ7l0oqF5ZSp1LzQX7ulHDo/GDZsWIqp2FwnZG1zaLsXXHDBFNM+reM0FO3UVl99\n9RT7wQ9+ABSFqqZFU32+5mSAK664AmiOi6O+8mIwHvhOj9i3gScauznGGGOMMcYYY8y0U+9l+TeA\niyuVysHAs8Bg4CNgt5p/ZYwxxhhjjDHGNJG6LnKr1eo9lUplBWADYAlgMjC6Wq22beUm2XK22mor\nAB577LH0XOwDZepHlhrZ7GIRpGuuuaYl29TpqBee+lxG+8aJJ54ItG8vtnYiWsZUGGW11Vbr9TpZ\nwNUzEzrXhjRQaO4cPHhwikkvWZqi7VG3KBxxxBEpNnLkyIHezLYnjknp9s1vfjPFNE6jBbcnsQ+k\nCiHNKBbliGxysR+r+rXutddeKSZrYy1LbdzfZVOOFr1OtHzWS7Qbylq7xRZbpNi2224LlO2aKopY\nbw9MFZiJRee6xaacKyoXi8RtsMEGAGy66aYptuKKK/Z6nQqi1tv3WcTzWB3fOn285grMqWcrwGc+\n8xkAhg8fnmKaT2PRr1zf4nqIBWh1q8Kzzz47Ve/RbkQNpEu0zev2gXhbgm7tiAWq6u0l3JM4x+6/\n//4AXH755dnnB5q6Ddb/vaAdNYDbYowxxhhjjDHGTBdTfxdxh6Ab11WkI2YW3Dpo2tAquVYQb7rp\npvTcSy+91IpN6kjiivi3v/1tAJZYYgkAHn/88fTcVVdd1dwN62CGDh2aHktTZXNiNkwtmt57770m\nbl1noRXdn/70pyk2YsQIoFjZje0ArrzySqBwHsCMnR3XKnosJKVCNGqdAMWKeS7z8MYbbwDllg13\n33134ze2jYm6SCtlb6EoRBOzjrUK+MglEwvNbbjhhkB3FkyM+ulxzMwos7PddtulmPbz6C6oJ5sT\nWw6p3Ug3FUzM7aPKisfsuFqyxGJoyoTHrGM92cbo6rruuuuAIqsJnX8eKw1yBfriOZLON2MLMO3n\ncWxObQZXbQTXWGONFHvmmWem6j3ajZwG0iqOF7m05CiAfJu2qUXzaGwvqmuvVp0TTPu3McYYY4wx\nxhhj2gxf5BpjjDHGGGOM6Rq61q6s/mOyLd97773puRnZSje1RNuI+uNKP/W6gs4vftBMVPADYIcd\ndgAK/S644IL03Ntvv93cDetAZMXZb7/9UkzFaWSpvf7669Nz9913XxO3rnOIljsVn4mWI9nIZKF7\n+umn03P/7//9P6Dz7avn7gEAACAASURBVHPTQ7SJyQKmIjQA++yzD1D0G4TetrC33norPT7ooIMA\nuPrqqxu/sW2OtIxjcplllgFgp512SrFcj9Gedr04JjUPyE4LM84cq7EWC3Kpn2js3azna1k/460K\nKoIW+8KquF83oXOenD02niNpPMVYPfbPaE2eMmUKAIccckiKnXPOOUBZ+04n17tZOsRbjJ588kmg\nmAOgvl7NUSvd/vG73/0uxY466iiguwp75q5tNAc+8UTR8VXnQeqXC8XtNbnbHETUVGP94osvTrED\nDjgAgNdee23avsAA4EyuMcYYY4wxxpiuoasyuXH1TMUUVNAnrpKb+ll44YXT46233hooVovGjRvX\nkm3qVDQ+1dYKiqzj66+/DpRbLcTVXVMQVxeXX355oNyyQbzyyisAnHzyySlmx0EZjUnNk1C0ZFEB\nKijGolZvY7ugTm+3MD3ksjnap7fffvtesZjVkabvvPMOAD/72c/Sc2eddVbpNd1OruVFzNCq1YWK\n+MTXxeyFHkvTWAztuOOOA2ac7G0kl5mdPHkyUMyTUOgcx7M0VXbt2GOPTc+ddtpppee6nTjWtG/q\n2A1FK8XlllsuxdZaay0g3yrs+eefB+DQQw9NMRXsiuO0m92HcY7T8fmpp55KMbWvi+fwOodSK6HI\nAw88ABQOIygyl7HgZDdrGr+bikE9/PDDKXbYYYcBhTsTYOeddwbKGXNlbuXaVPYbiiJd0S3Tjpo6\nk2uMMcYYY4wxpmvwRa4xxhhjjDHGmK6hq+zKsaiH+unJIhJvVJd1px1T6+2CbHWxL5v65MqeYE37\nJ9rEVOBjzz33TLF55pkHgBdeeAGwrb4W0lI9M6Gw20Tbksan7PSPPvpoem5GG585m2K0zKrYRCxA\nIbtSLDIhO+I///lPAC677LL03IxiqRU5a60KHAIsuuiiACy55JIpJi1jP1EVPDnzzDOBwpYH3dm3\ndWqJfUVl3YzWUBWTicehV199FSh6PJ933nnpuRmtMFqc6zT+ogaydY4dOzbFZBGP41m92w888EAA\nHnnkkfTcjLbv5zSVNR7g1ltvBcrW5PHjxwPl49bo0aMB+NOf/gTMOHbv/shpevvttwPFORLAhRde\nCJR1fvDBB4HiuN9NRbqmB+2j0aqtXuHRFq6xGAv+qQiaxmcnnj85k2uMMcYYY4wxpmvoqkxuXP05\n5ZRTgCLDE4skdeJqRLPRas5zzz2XYjfffDNQFEvQCq/pm7gqNnjwYKDIQEBRtEcruy+++GITt679\niVlHZRliCya1EYkr4Spfr2IJ7VTOvpUoA5kr6BOzjsowxoJSykZoXo1jeEYjZnJzrUXkznj55ZdT\nTK1WYob28ssvB+CMM84AuquVxfSgzEPUSsd2ZXWgyDLoeARw/vnnA3D//fcDM172NpIbpzGbI/1i\nC5BRo0YBMGnSpBRTplf7vM+fPkY6xGKG0ui6665LsX/9619A3oXgbGOe6BCQ+yW2wJkwYQJQ3r+t\nZW1yLoQ4JvW429wZzuQaY4wxxhhjjOkafJFrjDHGGGOMMaZr6Fq7cuw3Ct2Xgh9oZAO5/vrrU0y2\nG1kdop3MFqY80cqkQh/7779/isnmqF6FcQybvMUmFqBQsZ6LLrooxWSxl71pRraB5vbLaOtSQR8V\n7YBizOYKqkjTGXk+jd9dhadiTLYv9RGEQudYtEfHKNnpZ+Q5NH53zYnRAp57nSzgY8aMSTHZGG1d\nzGsai3mp6FyuQNpLL72UYtHibApkB4+3JM0///wAjBgxIsU0n8Z93+SRprniiCroB8U4Va9nKI7z\nM/I8WovcrV/SFopjWLwVKZ6/dirO5BpjjDHGGGOM6RoqA73qUalUXgIm9fvCGYulq9XqoGn9Y2ua\nxZo2HmvaeKxp47GmjceaNh5r2nisaeOxpo3HmjaeujQd8ItcY4wxxhhjjDGmWdiubIwxxhhjjDGm\na/BFrjHGGGOMMcaYrsEXucYYY4wxxhhjugZf5BpjjDHGGGOM6Rp8kWuMMcYYY4wxpmvwRa4xxhhj\njDHGmK7BF7nGGGOMMcYYY7oGX+QaY4wxxhhjjOkafJFrjDHGGGOMMaZr8EWuMcYYY4wxxpiuwRe5\nxhhjjDHGGGO6hlkG+gMqlUp1oD+jA5lSrVYHTesfW9Ms1rTxWNPGY00bjzVtPNa08VjTxmNNG481\nbTzWtPHUpakzua1hUqs3oAuxpo3HmjYea9p4rGnjsaaNx5o2HmvaeKxp47GmjacuTX2Ra4wxxhhj\njDGma/BFrjHGGGOMMcaYrsEXucYYY4wxxhhjugZf5BpjjDHGGGOM6Rp8kWuMMcYYY4wxpmvwRa4x\nxhhjjDHGmK7BF7nGGGOMMcYYY7oGX+QaY4wxxhhjjOkaZmn1BtSiUqmU/o3MNFNxfV6tVkv/Rv7z\nn/8M0NZ1Njn9ctR6zpTJjVPrZ4wxxhhjTHNxJtcYY4wxxhhjTNfQNpncueaaC4D55psvxWabbTYA\nll9++RTbfvvtAVhttdVSbNlllwVggQUWSLE555wTgJlnnrnXZ/373/8G4JVXXkmxQw89FIC//OUv\nKdbpWeDZZ58dgHnmmSfFZpnl4598oYUWSrGVVloJgKWXXjrFlltuOQCWWmqpFBs8eDAAc889N1DW\n9oMPPgDgscceS7Hjjz8egNGjR6eYtO9UZp11VqAYr1BkcOeYY44U01jUbwDwiU98Aij/Hnqdxr3e\nH+DDDz8EYOLEiSl23333AfDiiy+mWKePU7kKpE/uOSjGbsyO5757Pdnzjz76qNfj+F7dkoGXZv0R\nv2/OGdMtejSCnGMjhzUzxhhjWoczucYYY4wxxhhjugZf5BpjjDHGGGOM6Rpaaldecskl0+NPfepT\nAIwYMSLFdthhB6CwzkJh54w2xqlF77HEEkuk2JlnngkU1meAo446Cugs21nUdNNNNwVgmWWWSbHt\nttsOKNu9Ze2O1sZaRb9qMWzYsPRYv98f//jHFNt3332BzrIty6YNsNZaawHlsbP55psDsMEGG6SY\n7MfRrqxxF23eUzuO33//fQDOPvvsFNt///0BeO+996bqvVrJIosskh7LEh9vVdA42myzzVJMmsfb\nEvQ38847b4pJc2kbx7D25WhXnjx5MgCnn356iv3ud78D4K233prKb9Y6og1+/vnnB8pjTRrF8Txo\n0CCgbL/XfBA1XXDBBUvvG8e19JWOUNjqx4wZk2ITJkwA4N13302xdp9b4+0DOeu3vnt8nW6zidrr\ne+q2DijmwJzlXu8b9dHr49yZe49213Rqjynxb3J/O1Df3YUEjTGms3Em1xhjjDHGGGNM11AZ6JXJ\nSqXS6wOUBdh1111TbJdddgFgvfXWS7EhQ4YA+eJRA8UzzzyTHqvglbJnDeTuarW6zrT+cU5TZWI+\n+9nPppiytmuvvXaKKVOdy9oOFK+++mp6rEz9Cy+80OiPabimyhjuvvvuKbbFFlsA5Uy4xkmucFKP\nz5jWzevFm2++mR5vu+22QDlr1iAarqkKnmmbodjnV1lllRRbc801gXJ2MpeZzWV4plZnzYGvv/56\niu2zzz4AXHjhhSnWoAJfAzZO5TIAWGGFFQBYccUVU2yTTTYBinkVijEbNdN8G+eIWg4a6Rf1UcYy\nFk075phjAPjHP/6RYjGrOx00XFNlsxdeeOEUU2Z70UUXTbGVV14ZgDXWWCPF5KZRRhcKjaLOej/9\nq8+EIkMb9/Mnn3wSgDvuuCPFrrvuOgAefvjhFHvjjTeA6c5wNlxTjaeckyWXCZdrAIrfIToO5Fx5\n++23U0zF+vR9c63y9BooZ9Z7EvWToyM6O3p+Vh00XNNcxr/nc33F9DvkNIrfXY9zn1Hrc+P76veN\nxRn1fJwD9JtOxVzbcE07jbg/5Yoz5gor9sMMr2kkd45Rq21pH1jTxlOXps7kGmOMMcYYY4zpGnyR\na4wxxhhjjDGma6hZeKpSqSwG/BnYCHgIOKhard4Wnn+jWq3O29ff94XsSLFPrSzMsahMvdY42V2i\n7UUWDtmbovWulv052mm0TQNgV244+p5RA1k9VVwGiu+es17EgiayDcUCPT37mUYdc+8rohVN2zIA\nduWGo7EYx46sirF4T7Ql9iRqKntbjPW0hubsZDlNozV6nXU+dmzcfvvtKdauBVL0+y+22GIpJrt3\ntCuroFRuDoj7fhyfQn+T+9taBWyiXVQFr6K1tl37ES+++OJAYVGGogja+uuvn2Ky0caiUSJnba01\n/+aImmqfiIXwdtxxRwBGjhyZYg2yKzcc7fux6KGsyVHTddddFyiP51yBqpxdWa/Tv1Fv/R4q+AXF\nvhMLCQ4dOhSAM844I8V020ItK24r0JiI+5ksyXGcaB6QtlDcZhPnPWkUx66O37KAq6c7FL9BvH3m\nueee6xXTPK3nAK644goAxo4dm2K5QmDNJmcN1uN4fJb2UQ/NG9GSrzETx45003lQHN/6LeNvqs+N\nt5roFolY2FPnAJdeemmKjRs3DigXUeykIpW53yO37/e0mcd5Vc/F30/nUHHu1q0/8TYVHUvjLQ13\n3XUXUL71oVbRuxmFWrb6uE+stNJKQPl6RUUUc7dKzMia5uh53QBFEdG4T7z00ktAed/X+d30zAH9\nZXJPAp4DtgAuAC6vVCpfDs8P7M2cxhhjjDHGGGPMVNBfC6EtgKWr1ep7wD2VSuUG4KpKpTJXtVr9\nAzBN6SJdqT/77LMppse5rFhc5dLfxtW/Qw89FIDnn38+xbRCoJXL1VdfPT2nVe9YgEWvj4UtlE2K\nBWnalXfeeQeAp59+OsVefPFFoJxJzRWqkKY33nhjih155JEAPP74470+SxopiwBw0kknAbDqqqum\nWG4FR5o/+OCDdX6z1iFNn3rqqRR7+eWXgf411ar33XffnWKnnXYaAPfcc0+KadVPK+ExI3TwwQcD\nRcEgyBdviZq3O1qZi1kSZfPivp/LtErTOMZHjRoFlHXWeNaqd/yt1KostiaS9nFFd/jw4b1i7Yq2\nMWajtPIZV/5zGUb9Hq+99lqKaR6NRfiU2dHvov0ACv1ULAwKp0P8TBWdi24dzVHthsZMdArkCiJp\nLoz7o/ZpzR9QZFHisUS/W84xpN8jfr6OR1E/FRV85JFHUkyZm3ZzIEmjqJX0U1YKiqxuzO4qKxjn\nBWVR4sq/fiNpG4tMqWhUnHukc8zcLL300qX3B3jiiSeA8tzdDm6ZXDZK+1ycT5Xhjs4AHYtjITUR\n50y1eFPmNx739btER0bUVygbpveAYh6P2fHHHnus19+2A/UWM8y5iDTe43mQXBmaS+JxX8UCo86a\nS+KY1HE/nh/od7vgggtSTFnHOB91UnZ8apHe8dincbfRRhulmPYPZRCh2Pe//OUilyfnTBynxx13\nHFCeD2a07LjGczzua1zHdoKaI/bcc88UU3vTeC6nawid00Fx3Jye8drfRe4s8TXVavX+SqWyBTCy\nUqnM0+dfGWOMMcYYY4wxLaC/NMXdwKdioFqtjufjDO++wFyZvzHGGGOMMcYYY1pCf5ncnwAL9AxW\nq9VJlUplc+B/p+VDZSGK9olFFlnk4w3K9G+N6X/Zi5TahiLlnUtpy0Zzww03pNiJJ54IwAknnNDr\n9dHio+IVnYAsW7G4gLY/Wo9EtFrJevirX/0qxWTDiHa5noU7ot38gAMOAMpFZXIFMKItrd3RWItW\nzlr9QuM4ldXzZz/7WYrdfPPNQL6oR644hSyI9913X4rlCjLJvtgO9rn+kDUo2l2jpbAn8Tn1XP3F\nL36RYioIo96gUNs2dM455wBw1VVXpZgK3ERNZV/sBE1VwCXuW7Icx/lA3ynaWMePHw+U7W2XXXYZ\nULYcaR7VfBB10T6x/fbbp5jm52jD0xwfi/u1KxpP0comK2W0d8puGK1x0vS221KNxlQMKmoq+6x+\njzjXas6M9lJZ6Pbdd98U028eC/q0K/p+0V4su3ockyrwEq2AmhejJV9axn1faA6In6X5IJ5jyJ64\n3377pZh6oMfjpuah3NzdDuSK8cXzIekb5wMdv3PHsvjdNca1/8YiU9Iy2miFit9BcbtYLHylc624\nTe1avKfe3zp37Mmdl2oc6Xgei3Tpb6OVUz2y4xwva2gsLKrfPp7H9py7p+b7dBIaxxqnKsgJ8MlP\nfhIoz5OjR48GyudcGrOx77lsuZor4mfF37YbNRU6HkUL/Q9/+EOgfOuXznHjsW/YsGEA7LLLLimm\nOSXe+iNLeaN1rHmRW61Wb6/x3LPAUQ3dGmOMMcYYY4wxZjroL5M7oORWuHKtbeKVvVZyY5Gpeogr\na8oi5YoJxFVNrfp0QpEkfb+4Iq6Vu9z3jNprRVcFCuL71VpVic9NmTIFyK8Kx5hKh3cC0i8Wi4mr\nzkJaxayB2vncf//9KVbPKnXUVBmNXMur+JvmWsK0K1pVjgWHNP7iSrPGcSzbf9FFFwFw9dVXp5iy\n7PWuqOr3i79jbsxq9bbdMgo59F3iyr/GnbJSUIyjuJ+feeaZQNnpkiv2UEtTZSVUnAfyrdukadxP\n2hVlpuL31v4bt//RRx8FyscjOTDiGNd4zrVqyqF9QfNqjMVsjjI2sfBPLWdEK9F25fbzOMfKBZPL\nhsW/ze2b9WQB4typ98u124na6xyg3TK5OQ3qnbNy2dcccoporPeH9vmYdf/85z8PlLPuKhYY31f6\ntoO200Pcfj2O84bm6lyRrtx31zEqOjvU0kxFvaA4HsqNA8X5brdnHTXutd/GQqrKMEYNtO9HZ4Kc\nBptvvnmK6fzq+uuvTzEdX3NOkW5E4yU68E499VSg7KzLXZep2Oduu+2WYpojHnjggRTTeV08FjRC\n0/YvHWqMMcYYY4wxxtSJL3KNMcYYY4wxxnQNLbErK/WtwhtQ7kVaCxVPiUUNZKeJ1oGedoxoUYo3\nO9ci11eyXdH3jTZM2b5y1pkYU+Ei3SAOhdWulqaRXK/RHJ3Un01WiVh4SuM02uY0TqJVXLFYaEPW\n23otixqzsb9ejpyFul3R7x+tgCqqEecDjaPYq1UFfaIdMzeea6HXxX6NOTu/xn0nWJCkR7R2T5o0\nCShbk2VPvOmmm1LszjvvLD0HxW80tZa2WGRK4z7aQPV+cTvbFWmas3TGOU7jM45nWTIbUegljk0V\nT4lzirZTv2PPz20nNK7id6plOc4dtxpB7tiX6585bty4FFPRsXabD5phO53az5BGcY7VmNW8BEUR\nwGjrbzd9BwppWu++qn0ndzuajp8AV155JVCe4/U33WhRzqE5sd7bNuKYk91W5xpQnPP9+te/TjGd\nE84o41XfM9qV4+OexOO+bmN6+OGHU+zxxx8H4Cc/+UmK1SogPD3UdRVXqVTmA/YH1gTmjs9Vq9Vt\nG7pFxhhjjDHGGGPMNFJvqvJCYGbgYuDdfl7bL1rJjSstKj4QY1oNiCu/KrW+5557ptgdd9wBlLPB\ntbIRcdW2FrFFTruT01QrKHH1L6epCpnstNNOKaZCG3GVtdYKy6qrrlrXdtZbvKIdkEZRP61ExwIa\nuey12oyoOAQUjoN6MzxqAdWfo+Cuu+6q+Xw7EjVVNixmbdVaQZpBUVwntkeQ9vVmxzX+Bw8eXHP7\nxo4d2+97tRuxKI4yi/fee2+v18ViD41YkdZ+svPOO6dY/I1Ebt9pV/S7R100ZuMKtjSP43laM+E5\nYnZ8ww03BMqr5No/brnllhRr1+xCTo96Y40kzqd77703UJ4P5Cg5++yzU0xjtpPmg1ah84mDDz44\nxZTJvfTSS1NMbUbatVBaO6F9PhZEUsGpePw/99xzgfIc6zGbR8et2IJpvfXWA8qZXM0D8fykXefY\ndiG6D9XKKZ7LqYVjPD8ZKJdnvRe5GwALV6vVD/p9pTHGGGOMMcYY0yLqvci9BRgO3N/fC+tBKyix\n7YFarsSS6mq+HLOOytzExszrr78+kM/6iNhmZZtttulz23KthjoB6RLvIVOGO2ZPR4wYUXo9FCt9\nSy21VIpJ36hBz7ZCMaPwne98p89ti5pOnDixnq/TFuQ01arpmDFjUkyN7uOKtLIBuucLitXsd955\np8/PjGP9S1/6Uq+YiJrmsnXtijSN93rrfrdLLrkkxZRZie4MOSvid89pUwuNcTlCInElceTIkUBn\nrIJLg7j9at8Rm7Ir2xjvMc/dezy131lza2z2rt85bpPuF4sZ53ZFmkYt5MCITgw9bvTKvvT7+te/\nnmKqJRE/X20Xcq1I2pWoaU7ngWbo0KHp8V577dXr+VGjRgHFHADte59zuzDHHHOkx7/5zW+AIisG\nRcuV0047LcU0D3XCHNsK4jnaJptsAsD++++fYi+99BJQ1lQZyE6qfdIqdG62++67p5iuG84777wU\nk6vLc0D/yMEVWxfq2HjGGWekmM6jm6FpvRe5ewFXVSqV24EX4hPVavWoRm+UMcYYY4wxxhgzLdR7\nkXs0MASYCMwb4l6CM8YYY4wxxhjTNtR7kftFYFi1Wp3ciA+VPSVaKmS3uvHGG1Nst912A8pW41xr\nh1lnnRUotwZSilxFJr71rW+l56KFtK9tg7IFp93J2VNk77z44otTTDeExzYUstbGQlvSWdrGz5DF\nbO21107P1So8FbetkywfuZYXkyd/vAucf/75KRbbIggV/ok2Qn33aEPqaXNUsSmAQw45pM9tizrK\nttQJSNNo7db2X3vttSmm8RkLaOhxtDrXYxONhWZOOeWUXjERbbSxOFO7k9v3NT/mWk7EsdPzFoR6\nifvEdtttB5SLJPXcDoCLLrpomj6rFeS2MTd3DVQBEhVDicctzRvxFpITTzwR6NziPc0cCzr2HXVU\nYT7TuUAssHj44YcDndWarVVoTEbb96677gqU52m1Com3K7l4Tx7NrSosBXDMMcf0et2xxx4LFEVC\nwZr2Rzz333fffYHydcOpp54KlG+Tsqa1iUWm1No1ngvo+iO22WvmvF+7qWnBBKAzj6LGGGOMMcYY\nY2YY6s3kngNcVqlUTqH3Pbk3TO2H5q7iVYTg+uuvTzGtBqidCBRZRxWHAHjooYeAcoZsiSWWAGD4\n8OEA7Lfffum5WsVq4nPrrrtu6f372vZ2IJcdV/GZm2++OcX0fGzxIe1jAaNYLl0oo66VryOPPDI9\n11+bGzFkyBAAxo0bV9frW4lW8P4/e/cZL1dZtX/8N9ICJNQUIBAgEEIIBEINHUJv0gUEQf9SVZAH\nREDBR0AQCyAoVXhoUqT33kMnhF6ktxBKKKEIAjr/F/G699rn7JwzJ5me6/sm81lzTs6eNffeM3vf\na687zuAop48++miKaTH2+HNqLhVnIjV2ixonKX+6Cg5dVxzE/yNWOjS7opwqR3E2SnmJP6exG8d4\nVzOR+j8WXHDBFFtrrbWmuG2xWZ3eq1ZQtO8rLzGn+rmYq6k9nsXjh2YU4jFA/2+sDlHFQ7MeQ6Oi\nXNVa/Pw66qijgPzyFtqW2ExMjWZaIaeNpoaTm2yySYppPznnnHNSTJ/3zmn3llhiCQAOO+ywFNM4\n1hiGrELPs2Ldm2eeeQD4/e9/n2L6LrDnnnummJp5eZx2T9+RfvjDH6aYxq6qDKC4Ks+K6fNeSwRB\nltNLLrkkxSZNmlTfDeug0pNctc7tWDNRBgZXb3PMzMzMzMzMpl5FJ7nlcnnRWm+ImZmZmZmZ2bSq\ndCa3JmKZhcrqYvMZNSmINzZr6juuiVvUIGL22WcHsnUxVQLSnViuvOqqqwJw7rnnFm5zMyrK6WOP\nPZZir7zySqffUcOd2ChCsfj/aV3cAQMGADBy5MiKtinmVKWjMdbsOY0lVsppLMN85513gOIy0Pi7\nXZXWaozvscceKRbLFzuKZaitWAJWtC5rLBGudP3Mrp5X/uIazl2VdscmYbHMvFVUowy5UrqVA2Dw\n4M7FPHp/1WwK3MinO7rFBmDHHXcE8muRK38qD4d8YzHrTN8DAE488UQgv0a2mvYcd9xxKdaqTbzq\nJTajPOOMM4B8Wf3f/vY3IGviA163tTtxP9f+Hddz/t///V8A7rvvvhRr9u9NjRa/Y2677bZAdvsi\nwLHHHgu4RHlqDR06FMg3mbrsssuAxpcoR1M8yS2VSs+Vy+Vh/338JlNYLqhcLg+q0baZmZmZmZmZ\n9UhXM7l7hMe71HpDdFUqXgEYO3YsUNzUKF4ZLFrqRQ1r1FSp0sZIXW1bq9F2x4Y6elw0S9jdTJBi\najoz22yzVbQdcdZODa1aPafxar8eFzU06+516nc046CrY92Jy4i0+pXIrsbatNDs+Pbbb59iReNe\nx4/rr78+xZrpSmQz0YxDnE0sOrZqWagzzzwzxVpp+bB60jEgNprR8SAeO6+++moAxo0bl2Ktehyt\nl9hoRk0P1RQQYJ999gHgww8/rO+GtbDNN988PdYygs8991yK/c///A/gGfGeiJUx6623HpDNigFc\neumlgPf3nogzjKNHjwayKgPIj1mrTKyMUZOp2NhWjVmbyRTP/Mrl8r3h8d312RwzMzMzMzOzqddV\nufKRU3ouKpfLv6re5piZmZmZmZlNva5qeBcKj3sB2wKPAK8Dg4CVgcsLfm+aFDXqiaXJPW1Io9La\nrpr4FP0eZGuhtnqJSNH2V5rTWIKrx2qIENfK7EosD4tNsNrN1IwT5XTppZcGoHfv3hX9jdtuuy3F\nmrFEpBmonCY2RSmiWxpOPvnkFHOjlGIq+RwxYkSKFa1pfPrppwPw5ptv1nHrWtP8888PwIYbbtjp\nuYkTJ6bHBx98MOAy0Eqo0eQBBxzQ6bkrrrgiPX7ggQfqtk2tTjk96aSTUkzHyQMPPDDFfKtH5eaY\nYw4ATjvttBRT6fwf/vCHFPPnUeVmnXVWIL92s5rY3ntvKlBt+e/19aRmc9/+9rdT7OWXXwbyn1HN\nqKty5R/ocalUuhjYqVwuXx5i2wDbF/2umZmZmZmZWSNU2o1pE2DnDrFrgLOruznFKl0aI87WzjXX\nXADsssvknlmxuvm0yAAAIABJREFU7X1X4pIMF154YY+2s1V1ldM4k6uc7rXXXkDlOb3xxhvT42a/\n6lNvat6lWZruZse1tJOWFIDWXEKoltRw6ogjjgCyK7tRHPMan2qKZp2pudSRR06+i6Wo4iA2Q9Ns\nj8fmlKmJ19FHHw3AnHPOmZ7TrPif/vSnFJswYUIdt641KadHHXUUkF+WScsOHnrooSnmGbLuKaen\nnnoqkC0DCHDDDTcA+eYz1j0dT3WcHDhwYHpO4zMeT617yumvfjX5DsohQ4ak57R8mKtgekY5/clP\nfgLA3HPPnZ677rrrgOafEa+shhdeAn7cIbYP8HJ1N8fMzMzMzMxs6lU6k7s7cGWpVPo5MB4YCHwD\nbFOrDTMzMzMzMzPrqYpOcsvl8mOlUmkIMApYAJgAPFAul5tq7l9lNZCVKY8aNQooXsO0yCOPPJIe\nf/7551XcutYUc7rbbrsBsOKKKwLd51QldypxtMliWf1WW20FwOqrrw50n9MXX3wRgHfeeadGW9ea\nYt60Jt7666/f6TmJtyVozddmL7tppOWWWw6ALbfcEsiPYZV8XnLJJSnmZmjdW2aZZQDYYostgHxO\nVTofG9J4fHZv2LBhAGyzzeTr7zFn55xzDgDjx4+v+3a1Mn3eb7DBBkC+sZRum9FtNFaZlVdeGYC1\n114byK/Zevnlk1vfeH/vmTXWWAOAddddF8jfJvfqq682ZJtaUdF3Ke37umUBWuf8qNKZXP57Qjum\nhttiZmZmZmZmNk0qPsltBVreAmCfffYBYPbZZ6/odzUbEdu2+0oaLLzwwunxvvvuC0CvXr0q+t03\n3ngD8FW0jvr3758eH3744UDXOY1Ls+jn3dAnLzZE+P3vfw903XAqLmX1wgsv1HjrWlM8dqpxR58+\nfTr93CeffALA8ccfn2I+dhaLY1LNkdTQL+7np5xyCpAtb2VTpkZzkDXt0bJhb7/9dnpOS4R5bHYv\nfh7tt99+QLbczdVXX52ee/zxx+u7YS0s5nTnnSf3cVVjnzPPPDM9532+cmrcCbDpppsC8OWXXwLZ\njDi44VRPxJxqdvy9994D4KGHHkrPtcp30EobT5mZmZmZmZk1PZ/kmpmZmZmZWdtoq3LlWF6ncqWu\nGvnEsqUnnngCgKeeeirF9LvTc3lTLAGvJKcqFYGsKYXXIsxTcxTI1hwsyqnG3UsvvZRid9xxR423\nrjWttNJK6fHgwYOB4pxqfKrZFOTLRC2jhjOQNUlSTmOp0pgxk1s1uKFP95Zddtn0eJVVVgGynMZ1\nMS+++GJg+v7sqdTw4cPTYzXw0z597bXXpufef//9+m5YC1tqqaXS4xEjRgBZ/uItXT52dk/795JL\nLpligwYNAuDRRx8FsjVHwft8JZTTxRdfPMX0/f/6668HfBtSTymniyyySIp98cUXAPz9738HYOLE\niXXfrmnlmVwzMzMzMzNrG201k6tlFwDuvfdeANZaay0AZppppvScZnM0ewtZw4q4NEur3FhdSx9+\n+GF6rJvPBw4cCORnyj799FMALrvsshS77bbbAOdRlC81moFsKZvYPEU0s3Psscd2+nmbTMuujBw5\nstNzGnexkkBXzh9++OE6bF1rUjMUNfKIMeVU+zvAeeedB7hioyszzzwzANttt12KqQmVmqKMHTs2\nPRdnda2Yjplbb711iqkxmpa5ueKKK9Jz/hzqnsbk5ptvnmLK6TPPPAPkl7ux7qmRz0YbbZRiauJ1\n1VVXAVnzPquMZm3XW2+9FFOeH3jgAcDNpnpK+dPyVpH2/VY8hnom18zMzMzMzNqGT3LNzMzMzMys\nbbRVuXIsrVW5zQwzzNDp5zTl3opT7/UWS7rVjEJrkn711VfpOZXbxBIRN1DIUz7UGAGyktBRo0YB\n8Morr6Tn7r//fiA/rj1m85TTSy+9NMXUiEZNKe655570nNZ7Vem9daZjZrx1Q01nVHZ75ZVXpudu\nv/12wGOzo3g7xzzzzAPA0KFDU0zl3f/85z8BuPHGG9Nz8dhqGd2eADBkyBAgXwaq8an9O45hfx4V\ni9+R1Axtxx13TLF5550XgHHjxgG+LaESur0DYPTo0UA+pyq31a1zHpvdi7ccbrzxxkD+9o/PP/8c\nyI6dzmn34jjVrZ1bbrllir322mtAa38eeSbXzMzMzMzM2kZbzeRGuorjFvfVo3bi+temTrwq9uCD\nD+b+tZ7Rfv7yyy+n2K677lr4M1YZNTc75ZRTUuyiiy4CskoNNfYBH2Mrodna0047LcW0XN2TTz4J\n5Cs8nNM8zYrHBn1a6uLjjz9OsTfffBOAO++8E8jP5Fq+ukCz4vPNN1+K6djZv3//FNNYdPVLsZhT\nzYzFZa1+/vOfA/nlbvQdSs1+YoWCZ8rzOVV1hqoMAA477DAgv9yNxqcae8b/w98B8vnQrPiqq66a\nYscccwyQPx6oarN379712MSa8EyumZmZmZmZtQ2f5JqZmZmZmVnbaNtyZTObPrk0qTrimswu++y5\nOA61rvAtt9ySYrfeeiuQlSe6THHKlMt4q8fdd98NwPPPP59iKq2dOHEikJWJ22RFx8a45rVuUYjN\n+t59910AHnnkEcA57SiWgaqJ10cffZRiF154IQBjxoxJsWeffRaAm266CfDtCR3F8m2V1sYGnLq1\nQ+s6Azz00EOAb/2akphT3fYRb/W47bbbgPwxVseB8ePH12MTa8IzuWZmZmZmZtY2SrWe9SiVSu8D\nr9f0j7Sehcvlcr+p/WXntJBzWn3OafU5p9XnnFafc1p9zmn1OafV55xWn3NafRXltOYnuWZmZmZm\nZmb14nJlMzMzMzMzaxs+yTUzMzMzM7O24ZNcMzMzMzMzaxs+yTUzMzMzM7O24ZNcMzMzMzMzaxs+\nyTUzMzMzM7O24ZNcMzMzMzMzaxs+yTUzMzMzM7O24ZNcMzMzMzMzaxs+yTUzMzMzM7O24ZNcMzMz\nMzMzaxs+yTUzMzMzM7O2MWOt/0CpVCrX+m+0oInlcrnf1P6yc1rIOa0+57T6nNPqc06rzzmtPue0\n+pzT6nNOq885rb6KcuqZ3MZ4vdEb0Iac0+pzTqvPOa0+57T6nNPqc06rzzmtPue0+pzT6qsopzWf\nyTUzMzMzM6uHUqlU+FjK5c6To0Uxy8Q8futbnedI//Of/3SKNTqnnsk1MzMzMzOztuGTXDMzMzMz\nM2sbLlc2MzMzM7O2ozLbWGKr0tpYTtvo0tpmF8uVZ5xx8unjLLPMkmJffvklkC9b/ve//w00Lree\nyTUzMzMzM7O24ZlcMzMzM7MGKGqMVPS8Zx0rF3M6wwwzADDTTDOlmGYi//Wvf6XY119/DeRnIp3n\nYsppnz59UmzBBRcE4P3330+xSZMmAVluob459UyumZmZmZmZtQ2f5JqZmZmZmVnbcLmymZmZmXWi\nss+O/0LxuphFXPKZpxyqfDY271Guispt1cQHsjLbb775JsUqfT/amXLVu3fvFJt99tkB6NevX4r1\n7dsXgIkTJ6bYW2+9BcDHH3+cYjHn06uO4xVgjjnmAGDIkCEptuSSSwJZHgEef/xxACZMmJBi9cyp\nZ3LNzMzMzMysbXgm18zMzFpK0cxiR27UM3XijM28884LwCKLLALkG81odua9995LMc0sxnx/9dVX\nQL7JT6OXFqm3OE7nnHNOAJZbbjkAhg4dmp774osvgPzMl343zoC99tprQH7WLOZ3eqMZ3HnmmQeA\nRRddND232GKLAfkZc43rf/7znyl2//33A/DZZ5+lmGdys6WXYv6UZ+UdsvE311xzpVivXr2A7pur\n1Ypncs3MzMzMzKxt+CTXzMzMzMzM2kbTlSvHKe3ppYylViot47LKqWwjUi6d06nTVU67i1km7u9F\n6ypWynnuuaLcx3GtWNH6i9Ny/JjejvGxNG7WWWcFsgYokJXS6udiKaIeq3QWsvej6H2JDX2K1s9s\nRxqzsdzwO9/5DgDf//73gXwps5rKPPnkkyn2xhtvAPDRRx+lmEpqx48fn2IqCZ1eykHj8WDhhRcG\n4Hvf+x4AAwcOTM+9/fbbALz77rspprEb92nt+3FNUo3tdtz3u6P86ngw22yzpee0Jq4aUAHMPPPM\nnX5Ox5Si7ySW31e1/8b9XDmNZc06djZqTPqdNDMzMzMzs7bR0JncAQMGpMdrr702kF3ZAlhggQUA\nePnll1PsT3/6EwCPPPJIiulKwbTQVTFd8YHsiqWu7kB2dffTTz9NsWa4uqvtn2+++VJs5ZVXBmCb\nbbZJsQUXXBDIrrYCnHvuuQA89NBDKfbll18CPb/6UjSjEXOqx/FKj/LXbDf762pe//79U2zYsGEA\njB49OsWU03hF9ZZbbgHy41Svr6fjJeZU21Q0TnUVDbL8xZzGmYlG6dgcArLjwFJLLZViynkcBzoO\nPPPMMymmq4hFx4Cuxm68UqvHcYZC+Y15Vv5iTqtx7JlW2kY1eIBs/4pNYnQVO8b0fkyaNCnFtHyC\nGqBA9j5o7Mb8FS150XHb4uP4uzrOfP755ynWDLMRccawo6L9MeZeYzvOzmjfjLOIGk/KbcyLYsoP\n5HPUUdy3NT7jLGajr6bDtFUXFC1hocYxK664YoqtuuqquefiuNaxIjZJUizmVp/tauwD2axabOzT\nDJ/71ab3Jo5TvWblKo4rHSvifj548OBOsUGDBgFw2223pZj+n2b4rK+HOF40nq677jogf/zQexCP\n0/psXGihhVJM+/RTTz2VYtr3p8eZXB0Dldu4T6uJl8YhwCqrrAJk4xWyZlVPP/10bTe2xWjsxuPf\nBx98AOQ/D7VE09xzz51iGsct03iqVCr1LZVKy5ZKpZm7/2kzMzMzMzOz+unyJLdUKg0rlUpPlEql\nz0ql0qGlUmlT4HngAeCVUqm0dF220szMzMzMzKwC3ZUrnwicCfznv49/AvQDZgCOB34LbNHTP6o1\nq2J5rEq8iqa0l19++fR4u+22A/JriKkkNDZLUNmBbkKP5WcqHYuxomYTKqOJZTcqgTjooINS7Lzz\nzsv9fCMop7fffnuKqVyuq9I7gF133RWAN998M8XGjRsH5G/UV06V51g6pvLIojLGWOKg/MafU9nD\nAQcckGI33nhj7ucbQWUsF1xwQYots8wyQL68qKgMb//99wfg9ddfTzGVwMSyYpUfqbwjNkGIPyca\nY7FkrKi8U40+fvnLX6aY1oBrZE41Jg855JAU22ijjQCYf/75UyyOLdHrjM1LYtm9dCyZiTnV/xv3\nVZXlfvLJJymmMtFYsvvYY48BcNJJJ6WYSqgbmVOVA8Vy71GjRgGw2mqrpdgSSywB5EvFVYakssP4\nOOZIf0P7fu/evdNz2pfjmNQ+HctFVV4XG1WofPHaa69NsYkTJwKNzWnRGqw6jsbPA+Ulrsm45ppr\nAjB8+PAU0+dQ3Kd1G4xyGRso6X155513UuzVV18F8u+V9vn4c7feeisAL774YooVrV1ab9Pyt/W7\nseRT4zPmTbc+KBaP09r343uq3Mf3VPt+vBXhww8/BNp/HVLlOd6OdfXVVwPwj3/8A8iX0er9iHnW\nd5HFF188xTTW4/eJ6U0c/yqZ176qdXMB+vbtC8CSSy6ZYvqeEL8Lv/LKK53+3+mxTFn02vUdvejz\nIx4/tDbx0ktnc3X33HNPp5+zLLfxO4FyFI+nus0snos1+rOnu5PcFYCNmHxSewJwbnnyln5TKpUO\nA17s6pfNzMzMzMzM6qm7k9xSOKn9vFwufxGe+wyYdQq/16Xjjz8eyJpD9ISuGuhqF8C6664L5BtE\n6cpsT292jld0i35XM5Z77bVXil122WVA/upnvR122GFA1gQJKm+DXtRgSbMR8cqrXnvRLEcRXbmJ\ns3JFv6ur7t/97ndT7M477wTyTX7qTTPcyy67bIrFhlkdxdekcRSbq+lqd5wF0yxjpeO16Mq53r/4\nfmv/WGeddVJs7NixQGNnyDTDqNlbyJpBxH2viHKjmVrI3o84Q6b8FlUXSLxSW9RgTrH4u/q7V111\nVYqpOU0jc6pZgCFDhqSYGvnFsasZ3JgrzY7H16kZy3jVVn9D/8b/o2MDJShucqWZ9fhzeqwZjagZ\nlpMrmiWJsaJGXMpHrAIomolULpXv+Jxma2PTRcVi8xnNQsTPHs3gxpncdhFnV1Ut8MADD6SYxrM+\nU2LVh2bPYlMl/dzIkSNTTDOQsUGVlsiZXmbK4ljULLaOdbEyRj9XVC2jyhHIjsXx/ZteGk4VUd40\nFosa0hV9HsZGdKrsiA2WppfxWYmYC42/WC2mMauqN4DnnnsOaP+KjakVP5P1HSCeiylvsTmoqowa\ntb93dxb0cqlUGgRQLpfn7vDcCGB8518xMzMzMzMza4zuTnJ3BT6ewnN9gcOruzlmZmZmZmZmU6/L\n+sByufxcF8/dNqXnuhMbn/SUSrZ22mmnFFPDk+WWWy7FttxySyCbKr/55pvTc5pmV+kkZKU18SZ0\n3TxdtPbr1JRa15LKrqZmLSrlaPfdd08x5VlNJAA22GCD3HOxcZhK7kaMGJFi+t0VVlghxVQuUrRO\naSyXboaym6KS1UqpSUdsUKZ1dOP4VyMJlSE9//zz6Tm9l0XruK233noppgYKRaUksaFFM+RU5Wqx\nGVpXYzaWt2m8nXzyySnWVeMplcfEskOVicUSG+VUTe0gW2M6lozFEvGO/18jKUexZFXNm+JrV0mc\nyg8B7rvvPgDGjBmTYmqWFhtJdSydi88pB7GUXzndY489Umz99dfv9HPax2K5XlFZcL11VZocS9OV\ne5WzQlbWGceLfi7+bsdxX7QGa/wZrRu/9957p5gaucVmXkWNk5ph36+G+Dr0OuN6tmp0pmN2bHym\n3Mdjj27v2WyzzVJMzQUff/zxTr/bLnmcFvE4o+NAHOv6PIq3T6i0Nv6uc5l9Tsdyb31uxbGrW2ni\n2NVnn0tr83TMjLfa6VaZePuYbvvQdzXIcjo9l9J3Jd56qO9Z8fupvm/G22y6Wt+9Hnr+7d3MzMzM\nzMysSXXXeKom1GRkjTXWqOjn1YQIYPTo0VP8ObX/hvwyH1Ny5plnpse6+hObH5199tlA/uqFrrDf\nddddKdbI5kiimepvf/vbKdbV0kEXXXRRerzzzjsDxVdWNdMD2VJJldLV9JhT5TzOjmrGQcuJQL5p\nS6PccccdAPz4xz9OsaKlbTQmjjzyyBQ76qijcs9NiZZKqpTe0zjr+Ne//jX3HGRXz+I+0QxXfJ94\n4gkga/AA2cx2nOFTtcCee+6ZYjfccAOQn6GqBr2ncZZNM+xxnOoqb9HSLI2kWTzlFrJxF5vyaFZX\nDcggqy6Ir6Mas9NajkwN7CCbyY3NU7T8W5y1aIbZcW1Dd7NNRbO7RZ8HXf0/XT1XVPESl2fQbEWc\nzdSsWdymdpw102uKVQA6bhQ9J/EYrpmH1VdfPcU0wxOXFGuG/bxRNMuoSoI4S6N9uWhZvGHDhqWY\nmyTl6bNa1YfxM02z3XF2XPt8zJ+OsZ51zFN1RlxST2NypZVWSjHNRMbvYHGft4zGYmxkqeaWsapV\nn0OxsqjR49MzuWZmZmZmZtY2fJJrZmZmZmZmbaOicuVSqTQnsB8wEugdnyuXyxv29I8ee+yxAPzy\nl79MsVjuIpryjk12akWlM3GdWYnT7RMmTADg8MOzxtLNUHZzzjnnAHDMMcekmG62j81LXnjhBSAr\nUYbabb9K+WLzKonlX7rx/4wzzuj0u4109913AzBu3LgUU7OyWEZ46aWXAlmJMtRu+/X/xgZpKn2K\n5cgqp4+l/s2QU5X8xlsFVAoT16k94ogjgHzDuFrnNJYsqtFSLD09//zzAXj77bdTrBn2fZUMxu1S\nLJawqxlVvBWgVjnV39Ua5pCVicbS2uuvv77TNjVDTivdhlo3yYr/rxrzxdt89LkZj1FaG7YZ9vd6\niDlS2ac+s+Nz+hyM3zU0PnV7AmTfO2ITwEaX3DWSjss6vsTmUfocLGqmGNczf/bZZ4HmuA2pGeg2\nA+VNzf5iLH4XLVp/VN9Fm+F42Qz0mb3FFlsAsPDCC6fn1EwuNjd9+umnAbj33ntTLN5KY9kxU8fJ\n2PBXjeXibYa6DS02t2z051Cl9+ReCswAXAn4KGVmZmZmZmZNqdKT3FFA33K53LmLw1TQVdF4c7Ku\nsMSrUrr6Uo8rVVqC5+CDD04xXcVQMwuAX//610B+aY5moJnRl156KcXU1CDOmq611lpAfXKqG9L3\n3XffFNOVXzW8gWxGP46HZqBxqgZUAIMGDQLyDSB+9KMfAfW5YqUlmOLSLMppfO+PPvpooDmaokXK\nka6iAlx99dVAfj9TzuuRUx1n4jIi8vDDD6fHV111FVD9xlfTqqjJjvalmD9dpa7Vvh8rRjbffHMg\nyy1k+9MVV1yRYq+88kqn7bRMbJK06667AvkZHh1HL7zwwhSb3mbL4njuqkFUUeMuVW/EpneqglGz\nJJj+xmfcl9UUTjmIuejde3JhX5y17dOnD5DNNALccsstwPTdwCvSeNPyjXEMa5+P+7EqslSJB26S\n1NF8880HZJUusdpOM4vxe4dmcGMTzOltP++OvsMfdthhQL4Z2v333w/A7bffnmLKb1xisNEqvSf3\nXmDJWm6ImZmZmZmZ2bSqdCb3+8ANpVLpIeDd+ES5XD6y8De6oHti4pUoXTGIV1p0b1GtxHtzdE+E\nlhSAbOZB92YCXHDBBUDzXfHR1T9dGYRs5kYzZQDvvvsutRSv9OieJt0bDNmV3Ouuuy7FdJW32XKq\nK4HxirTy++c//znF4vIntRAXin/ooYeAfNt2vc9xiacnn3wSaL6caoYgXqXWvvfggw+mWK2XO4qz\nOddeey2QVXNANhMalyLTrFmz3gNVNKMVZ51rvd0DBw5Mj48//nig+Hjwf//3fynme6C6pqUvAHbY\nYQcgPxt2+eWXA9lSTDB93z/aFY1/3Q8J2WeTjpcAv/vd74Dmq4Kpp3is0D4aZ3dF1SOxikTHbvUw\ngKzKqFmPnfWm47JmGONxWrks6lMT78n1rHjeIossAmT795gxY9Jzup9cfSkAPvjgAyB/j7nHZ94m\nm2wCwGqrrQZkvX8gq3gpWlKsmb53VnqSezSwEPAaMEeIe0SYmZmZmZlZ06j0JHdHYIlyuTyh2580\nMzMzMzMza5BKT3JfAarWbUXlGLGsSqXLsUV1ragMNTaPimXKotKG2OSnWcvrVO4Sy4ZURnDIIYfU\n/O9ryZDYCj+W1Mpbb73VaZtqXZo6tTQ+4/ZpSZtYbl0ryukTTzyRYgsssACQL6t5/PHHATj11FNT\nrNmaI4nKWOK+r1LBWEpUK1pmIDYTU8OKuE1aFuqee+5JsWYtAy0qsVIpWz3KhlRW97e//S3F1Igm\nNqDQslDNtgRTM9LtJ3vuuWeKKadq1gVw3HHHAc37udRMdDwdPnx4iqm08ZprrkkxLWnXTCV3jdRx\nH41ly3o8YMCAFNMSOHFZK4/PfN50fNb3tVh6rCWbYk61DON9992XYj525ptL6bNGtzxquULIbj2I\nt87pNjB9JwU384J8TrXfquGZSrwhG7MxpxrjcQkhNWlt1Hit9CT3fOCaUqn0Zzrfk3tH8a+YmZmZ\nmZmZ1VelJ7k//u+/x3SIl4HBPf2jukJ66623ppiupugqai2pAUrR7G28orbKKqsA+asSzUo5veyy\ny1JMM2OxcVKt6ApjXGxb4kyoFpVutuWCiiincbFwjdl6NCW56KKLgOwqbhSbJWy33XZ126Zppat5\n8Yqg8lyP2edjjpl8CItL2+jqY1wy5Be/+AXQvFUGReLMUz2vmmq2Uc0pohtvvDE9VoOvZp0RbyYa\nn7vsskuKKW+x6Z0qZzyr072+ffsCsNtuu6WYlnK58sorU6yV9vlGiDM9auC3zTbbpNiQIUOA/Hc5\nz4rnKR+ayY0N+jROt9566xRbccUVgfxSYZafHddM7muvvQbkG3ZqP992221TTFVx++yzT4rpO/P0\n3NSraCZX3y0HD85O9/R9KX6XUs5vuOGGFBs7dmzu/6q3ik5yy+XyorXeEDMzMzMzM7NpVek6uWZm\nZmZmZmZNb4ozuaVS6blyuTzsv4/fZArLBZXL5UE9/aMqrVLpIGTrZtaq7OrYY49Nj1VOE6l8RGUh\nkL9xvdkpb9dff32KqfyzVjnddddd02OVdkcq+YjPtVJOJW6zclmrnK688srpscqQY0mOSj5GjRqV\nYrVe+7gWYklgrUst41qj+++/P5A1oYHsdoT1118/xWI5eKuoZ8lqXGf46KOPBrJmSQAvvvgikC8F\ncxlo11RSB9k6wyqpg2yNbK3VDtN3WV0l4pj83ve+B8Daa6+dYmok6NLaKdNxRZ9DcZzqFoWdd945\nxVSeGBv6uJw+T9/NdAtCPJ6OHDkSgK222irFlMtWuCWpnuK46rju7eKLL56e062Jm2++eYqpLDc2\nSC1aC3p6E49/anKo3Pbu3Ts9p8+mQYOyU0DlXM2mAJ577jmgvt/5oq7KlfcIj3eZ4k+ZmZmZmZmZ\nNYkpnuSWy+V7w+O7a/HHY/OhWl3l11Wagw8+uNNz8WrCUkstBdSn8VUtaUYcatfgRS3uzznnnE7P\nxZyutNJKQH4JnFYU81irK1Ba2iY2udJVxfj311xzTSBrntaq6nElT1dqH3744RRTnuNSW5rBHT9+\nfM23qdVpTJ5++ukp1qdPHyA/y6AqhFZo2tcs1llnnfRYx864BNMBBxwAwKRJk+q6Xa0szoSriVf8\nrqEZ85hnK6Zjdpz5UiOf+Bl1wgknAPDxxx93+t3pWcyBKjA07uLsuJZgeuyxx1JMjdHef//9wv9v\nehVnHfVZo7xpaTvIPvfPPffcFFOeY049k5vPqSqytE/Hqjh9r4p5VtVmXJ5VuY8NrerZfLKrcuUj\nK/kPyuXyr6q3OWZmZmZmZmZTr6ty5YXC417AtsAjwOvAIGBl4PLabZqZmZmZmZlZz3RVrvwDPS6V\nShcDO5X8AdXeAAAgAElEQVTL5ctDbBtg+2n547FksNo0hd5VU54f/CC9xJYvU5ZalgGoWY9KEYpK\nOw488MD0+PHHH6/ZttRTLcuClMOnnnoKyDdK0d894ogjUkxNPax7v/3tbwEYMGBAiqkU5w9/+EOK\ntcs4rQetibfFFlukmI45Z5xxRoppPFv3dPvHUUcdlWL6/IrrYqo8zGWK3dNnlcrmITsOXH311Smm\nfd857Z5yqsZIAMOGDQPyDS9vv/12wE3RuqLxps8jHQMgu23muuuuSzGtNx5vR/OYzdPnkL6fPvPM\nM+m5Tz75BMiX0Pfv3z/3HDinHalBmj57nnzyyfScyrzjd1blMq5RrONAo3Jb6RJCmwBXdYhdA2xa\n3c0xMzMzMzMzm3pdlStHLwE/Bk4KsX2Al6flj1f7zF43OANMmDAByC8VImqLHW9Ct2LxZvEHH3wQ\nyLcRl9deew3IblC3KYsz4L/+9a8BGDp0aKefe/XVV4FsiRarzLLLLgvAT3/6007PaZweeWRFLQfs\nvzTTcNFFFwHQq1ev9JxyGisOfEW8ezoO7LbbbkA2Sw7wzjvvAPCrX2UtLzwzVjnN0my00UYpNnHi\nRADOPPPMFNOSbNY9zc5o9hay2Zw4Ox4bilrXNAsWx6FmcmNVnmYnfQzonipE49KPHRt9QZZzzVaC\nlw/rSJ/jWhIoznorp/H7rGIxj7VeyrQ7lZ7k7g5cWSqVfg6MBwYC3wDb1GrDzMzMzMzMzHqqopPc\ncrn8WKlUGgKMAhYAJgAPlMvlr7v+TTMzMzMzM7P6qXQml/+e0I6p4bZMNU2X33TTTSmmNbAkNrla\neuml67NhbSCuL7z88svnnovrDa644op126ZWp3JagIMOOij3XCxb2nDDDQGX0FRi9tlnT4///ve/\nA9ntCzGnKg2NJUpWLJYh7bXXXgAMHjwYyB9PVU4fS5mse3379gVg3333BfK321x88cUAvPnmm/Xf\nsBam22vWXnttABZbbLH0nJqnPPvss/XfsDag25QWXHDBFNOxNTaaq+camK1O5Z2xDHnIkCFA/hjb\n6OY9rURjMt5qN9988wEwYsSITj8fv185v8X0XT9+J9CtoHPNNVeK6VaGuJZ7o9cerrTxlJmZmZmZ\nmVnTq3gmt5kttdRSQHb1NtJVxfhcbMNuxTQTfuihh6aYrsgop3vvvXd67oMPPqjj1rUmNZk4//zz\nU0xLhSinJ52U9XZ7+eVp6us2XdCYPOCAA1JsoYUmL/Gtq7KxwuOBBx6o49a1toEDB6bH++yzD5CN\n1zhzc8EFF9R3w1pYnF3YZZddAFhiiSWA/DH0tNNOA1zF0VOabdxpp52A/FIWt912G5A1UbHK6Bir\nygPNNEK2JIuaeoFnw3pCn/t9+vRJMVUavvTSSymm9yDOijnPxYoqCXSOMGrUqBR74YUXgPwxQsdn\nH3fzNNbi+FPelltuuRRbffXVgfySrHEpp0bwTK6ZmZmZmZm1DZ/kmpmZmZmZWdto2XLlWPb1P//z\nP0C+xEA3n5988slAtsarTVksRdhhhx06xXTzucq+zjvvvDpuXesbOXIkAAMGDEgxNZd48cUXgWzd\nXKvMnHPOCcBmm22WYjoOaK3R/fbbr9NzNmU6tm6//fYp1q9fPyAr9TzqqKPSc15rtHLKI3Q+xt57\n773pOTecqlz8jFJZ4vDhw4H8mq0PPfQQ4GNAT+k2G5V6Lr744um5u+++G/AxYGoVNe/RrWLxtjod\nkxvdxKeVxP1ceYtl9Z999hmQz7NLwLsW86Oc9urVK8WU81deeSXFtDZxo3LrmVwzMzMzMzNrGy03\nk6urB4MGDUqxRRZZBMiaIEB2hfEXv/hF/Tauxc0999zp8WqrrQZkV7sga4Tw/e9/H/AV8UrMOuus\n6fGmm24K5Jde+vDDD4FsiRY3Reuern5DNrugGV3IjgNnnXUWAG+//XYdt671abZx9OjRKaZmHjoG\n3HHHHfXfsBbWcWkbyD63tM+rQgbyS4pY1+Ixduuttway2TBVyAB8+umngJv39NS8884LZGM3NqTT\ncSEek5Vf57Z7ylVshvbcc88B+cZT8XuYdU3jLlYXjB07FoDx48en2LvvvgvA+++/n2L+Ttu1uE9r\nTOpcC7JqJM3eQuOrPDyTa2ZmZmZmZm3DJ7lmZmZmZmbWNpq6XFmlHGp8AFmZ8uGHH55iWmcwlnxc\ne+21QNbYx/JieZHKlPfYY48UW2mllQD4+uuvU0zNu2JZuGViMzTdjL/hhhummJojzThjttu99dZb\nADz//PP12MSWprypzBNg9913B/KNO3QcUBmNS5CmTMfYopLPESNGpJhyr/XvGl2C1GpU8qlmU5Ct\njanxqvI5q4zGrkqTARZYYAEg+9yaMGFCes4l4FNHt9fo8z9+d9D3rE8++STFXKZcOd2q8PDDD6fY\nE088AeRLlHW8LVoD1orF/V3fs2K5ssapx+vU0Vhs9vMBz+SamZmZmZlZ22iJmdxZZpklxZZeemkg\nP3Pz3nvvAdkSAQA33HBDPTax5SinM888c4oNHjwYyDfz0tWZp59+OsXOPfdcwDNjHSmncYa2f//+\nAAwbNizFVFUQG0qcf/75QP5GfcvEWYPevXsD+ZzONttsQHYMAHjssccAeOGFF+qxiS0nNt7RmO3b\nt2+KqTImziRoKRY1nPKsWPfi8UDLhsWcapmgZ599FoCnnnoqPefZhcrFY+fVV18NwDPPPAPAmDFj\n0nMffPAB4Nz2lL4LnHPOOQBceOGF6TnlPjZTdH4rp+9SkyZN6vLnnNNp41nb6Zdncs3MzMzMzKxt\n+CTXzMzMzMzM2kap1tP3pVJpmv9ALK9T+WIsBdPzsclUk9+g/2i5XF5xan+5VjmNDb70ODaYafIm\nXk2Z01gWrtLamEeVLzZpCXhT5VSNvWJOddtCbPqltYc1dpusRKnhOe3w/wHFOY3rZqucbuLEiUC+\nIV0TaKqcSlEjuvnmmy/FdIxVTmMDjyb4/GrKnHb4G0D+lgbd2qTcx88vldg38HjQ9DntShzP0gRl\noC2d0yblnFafc1p9FeXUM7lmZmZmZmbWNuoxk/s+8HpN/0jrWbhcLveb2l92Tgs5p9XnnFafc1p9\nzmn1OafV55xWn3Nafc5p9Tmn1VdRTmt+kmtmZmZmZmZWLy5XNjMzMzMzs7bhk1wzMzMzMzNrGz7J\nNTMzMzMzs7bhk1wzMzMzMzNrGz7JNTMzMzMzs7bhk1wzMzMzMzNrGz7JNTMzMzMzs7bhk1wzMzMz\nMzNrGz7JNTMzMzMzs7bhk1wzMzMzMzNrGz7JNTMzMzMzs7YxY63/QKlUKtf6b7SgieVyud/U/rJz\nWsg5rT7ntPqc0+pzTqvPOa0+57T6nNPqc06rzzmtvopy6pncxni90RvQhpzT6nNOq885rT7ntPqc\n0+pzTqvPOa0+57T6nNPqqyinPsk1MzMzMzOztuGTXDMzMzMzM2sbPsk1MzMzMzOztuGTXDMzMzMz\nM2sbPsk1MzMzMzOztuGTXDMzMzMzM2sbPsk1MzMzMzOztuGTXDMzMzMzM2sbMzZ6A3pqpplmAmD2\n2WdPMT2eccbs5Xz11VeF/wL861//AuDf//53ipXL5dy/UVGsnXzrW5OvdcwwwwwpViqVgPxr/89/\n/pOLxefaPUc9pfwVca7MzMzMzGrHM7lmZmZmZmbWNppuJlezigADBw4EYP3110+x733vewAstdRS\nKTbHHHN0+n80W/b1118D8M9//jM999577wHw5ptvppgeP/vssyn26quvAjB+/PhOP/fJJ5+k2Dff\nfJP7m80m5nSuueYCYMEFF0yxtddeG4CVV145xZZYYolOv6vX9/nnnwPwwQcfpOdeeuklAJ577rkU\ne/vtt4F8nj/99FMAvvzyyxTTe6P3CrJZ9lbI6cwzz5z7F2DuuecGsnwD9O7dG8jGC2SvU7GYg0mT\nJgHw2WefpVjRzxXlSLPu+jf+XLPmNM5+63F3Man0dfb0tRf9rWbNn5mZmZlN5plcMzMzMzMzaxs+\nyTUzMzMzM7O20TTlymooteqqq6bYwQcfDMBqq62WYipNjuWiUtQkSc2oYlMlxWIpqcp3Bw8enGIq\nrX3ttddS7Pbbbwfg4YcfTjGVlcZGVs1glllmAWDppZdOse233x6AzTbbLMX0mmedddZO/0csA+1Y\nAhtLZtdZZx0gK2UGeOONNwB4/vnnU+ytt94CslJwyErEX3755RRTiW4s7W2GMtFevXoB0L9//xQb\nOnQoAGuttVaKjR49GsiPJ43Z+Jo6juP4nBqkffzxxyn2xBNPAPDggw+mmMrCNQ4BJk6cCMA777yT\nYioLj38jvr/1FMuAte/rX8jGbtxHBw0aBMCiiy6aYtqXY0m3XpP+hvIIWQO6mKv333+/U0z7ctzO\nolsfFGuGJmxFpd1F4pgrajonReNkWkree1pubmZmZja1PJNrZmZmZmZmbaOhM7nxiv5ss80GwGKL\nLZZiCy20EJDN6kRx1vTDDz8E4Oabb06xa665BsiaRsUmSZp9ibNECy+8MJBvaKUZujhrt8giiwDw\n6KOPdvPqGiMuo6TXN2LEiBTTTLlyC1l+42yKZsZef/31FPvHP/4BZPmOM4xffPEFkH9P9R7FhkzK\nZWwWpt/RjGT8+80gjj81lFpuueVSTLPisUGaXmd8P5Sj2HRLj4tmuTrOSMa/v8wyy6SYxqn+f8hm\nxe+5554U04xmnKGrN70WzYhDtgRYbIamxmcbbrhhiuk1KweQ7cuxqkAzwnrf4nugMfnRRx+lmHIV\nm85pVjdWJqiBmmbTITuuNLLiQLOw8XXqcXd5XnzxxQEYMGBAium1qJIlxiT+v3q9cZ9VJUHMn8Z6\nfK/0PsTjs8Zxo6oMoHhmW+L+WDQTrsdFx8KuZsejopj+v6KqpPi3iprTNTKXou2u1rbUaj9rpeXf\nNP6a4f1tF0XLJ9q0cU6tUTyTa2ZmZmZmZm1jqk5yS6XSQaVSabZqb4yZmZmZmZnZtOiyXLlUKo2e\nwlMHAy+XSqWPy+XyHdXYEJWo3XrrrZ2ei6WZKqG75JJLUkyNjaal8dPjjz8O5MuQf/KTnwAwZMiQ\nFFMJWLM1RJK4LSoVfOSRR1Ksb9++QL40WCVPKvEGuOGGG4D8esBdldcVlduqbDSWRu+6664ALLnk\nkin2yiuvAPn3rxnWdC0qW9P7r7WWIcul1gqOzz/zzDMpduWVV3aKaTwrt7EUsai5msr6Y/MllUsP\nHz48xfTex5w2w9rDek2xyZRKX+ecc84UU7mymk1B9tpjmfyECRM6xbRvFpXLq/HUu+++m2JqJKV1\njAFWX331Tr971113Afn3TxqZU5WsqhwZstsBYv5GjhwJwBprrJFiKleOJfnKWxw7yoPeg/i3NHZV\nogxZgznd2gDZ/hTfq5tuugmAMWPGpJjK6hvZzEuvN45TjY94m8v8888PZHmE7NaXOJ407mJZqca9\nyu/nmWee9JxyFcvqY95E/18cz8ppXLO8GUrANcbicVXNDmPTQ/1cLImPj0X7rXIL2b5f9Fml9zL+\nX9pPBg4cmGJ6f+N4HjduHJBv5BfLwRul6Fauots0pKhcPo6Jjvkr+rmYU72X8bNK72W8zUv5jU37\ndJtI/I7RDM074zG/Y6zoFoQ4Doo+Y6vxXUbfC+Jxt1+/fkB+/KuJYmy22AzfT7Xvxfx11eyw6Pt1\nrV5H3A7lN47DZjh2Fikak3otMVcan0Xfr+tB7338m/W8Za67e3JvA94GOh7N5wL+BHwDDO74S2Zm\nZmZmZmaN0N1J7q+B7YCfl8vlmxQslUoTgBXL5fJ7U/rFSsQze12NUqMogHPPPbfT79Tqaor+33jl\nXg194tUzXU2PV8+a4UqZxKs1ms2LS/i8+OKLQHGjkmm5ulJ0tU1XE2OulFPNgMS/G6+INkNOtQ1x\nu/T+P/nkkymmGdyzzz47xXT1L16lVh6m5bXpinm8Sq9ZXTVFg2z2Pja5aqYrkXG7NGbV2Ayyxk8v\nvPBCiukqZVx+SstUdVVxUHRVPc48aP+OjcO0BFRscqVqj2Zr6FPUfEavs2hJqph7XfmPxw3NVsVq\nhY6zM3EGST8X3xf9rdjIb9SoUUD+yrmWZ7v33ns7bXsj6fVp5hqyipQ4a7vssssC+WXa1MSraPYx\n7reaRdS/cQZJ70ds3KUxXtSwMc7y6nd0rG8Wffr0AfL71LBhw4D8ONGya/PNN1+KaXY1fhYrX/p/\n4+OuZjPjWNe+EPcd5TfO5J588skAnHnmmSmmfauR41WVWbFaauWVVwayz1qAeeedF8jnQ/thzJ/y\nrFnC+Ly+GxUtQVY0c1QkjtPTTz8dgBNOOCHFNNPbyJzqu4nGJmTVL0XL18XPHr2+WJmwwAILAPnj\nRsf3Ix4TNa5jTGM25l7jNDbt++Mf/wjAaaedlmJxVrdRVFG00korpdjaa68N5KsotK2x4WlcwlP0\nO8svv3yK6X0rms3UfhKPyUWN/HQcUJNJgEMOOQSA66+/PsWaoeJAx8mNN944xZRTvV7IqgVjZc9T\nTz0F5MeGcqrPacjeN+Ulfh4p37ECT8fdos+yp59+OsX23XdfIN/Es1b7fJf35JbL5SOBrYCflkql\nK0ul0kJd/byZmZmZmZlZI3W7hFC5XH4F2KRUKm0P3F4qlc4Dpry+gpmZmZmZmVmDVLxObrlcvrRU\nKt3I5BLm8XS+T7cqGtVsRKU4v/jFL1JM0/ax3Fcllc12Y3+RopLFet7wrTKQnXfeOcVUTqEyScga\nT8WmFM1QBipFTTiKGhPEsiGp1diIZT9a+ziWhasZVlFOGzlei7ZBYzLuU2q4E0uY9buVrjXa022K\njYJiWZqoFC2uR9wMOS0qm1Qsrl2rMfHAAw+kmEq2Ysmdcl90jCtqViOx7EtNlHbfffcU23LLLae4\n7UV/qxnGadzPlcvY5EklYDGmktq4P6rMK74mlXmpRDTuq3o/4vun9Y033XTTFFPZY3w/ihriNMNn\nlErZYuOuxRZbDMhK7yB7nUXrW8dxov+vqEGfxlV8D5SPovWflUfISsBjIzD9rUauMV5EJYPx82Dr\nrbcG8mWgGpNF2x9vPdAxMJYbdmwQVPQdLf6/Rc0F9X/E90pl641qiDMlanS6/fbbp9i6664LZLcW\nQPZa4msvanao0uWi0vmuxFxo/y4a67G0XE3vmiGPkW7/+cEPfpBiusUj5kqfR0XfWeN40u8U5aNS\nylG8/UOx+PmvJqmxXLkZfPe73wWyZq7QuWQbstcXb8fSOO1uzfVqWmGFFdLjESNGAPly5Vrp0V5X\nLpc/A35Wo20xMzMzMzMzmyY9u7TUZuJVoN/85jcA7LTTTimmqzpjx45NMc06NtsV3Wa0wQYbAHDQ\nQQelmK6Sq4kPZE2cWmF2XBq1fWrG8utf/zrFNAsRZz21FFbRrGMjdTVLVzQbVfS71aYx+Z3vfCfF\nNMugJYogG6fNltOiJX80axWbTGl2UMv7QOcZ2hjrqThzqBlIzcpBNgsSG1qpGUXc95shp9qeeJwv\nWn5KYyK+dj0ualBWlOeu8h0/o9T0KzYZ0gxoXKpJn1fNdjzVTHWs4lH+4pjU7EJcrkdNoOK+VzRb\nq/Guf4s+p+PMkWaT4vF0lVVWAfI5ve+++zr9rWbIqbYnLjWlY1bcPo2juMydlvCJ40SzjjFHasLz\n7LPPAsXLgsVZNs167r333immBjax8ZSWKYw5bQZFDeOKxpoULZ0WZ8OKmqZpH1AlSPyc0fsWj91r\nrbUWkFVtQTbrHisTrr76aqA5lreK9L0lNvIrouNjHJMaM0VL/cRxqt9RbuM+IfH4oSqS2OBOs5mx\ngubmm28GmqPZVKRZ+9jkTPt50Wxs/OxRruLnS1HDM/2Ojt0xLxLzou+icZu0LbFp1T333APU5xg6\n9XP9ZmZmZmZmZk3GJ7lmZmZmZmbWNqbLcmVNn6tBA2TrNsXyB63BedJJJ6VYLHewzmLpx6mnngrk\nmzWotEbrDkLWsKkZyr+aUSwFO/zww4F8YwSVi1x77bUpprL6ZiuxqVStx0Is01lnnXWAfCmYSna0\nnhwUl6Y2A5UUFTXQKFKP/UyltZtsskmK6dga129V44lmu/1D73F3TfuKypCrWW5dVIKuMjvI3vO4\nBqFKIJuh7DvS2texDFMlbPEYp8+I2IirqLna1I7j+PeVqxhTKd/DDz+cYjoONFtOdZyPZYTKc1z/\nVt9lYhNN3b5Q1MhvWl6nypt1XIXsO8Btt92WYg899NA0/61aePDBB4H8+FJ+45qgep1XXXVViqkc\nPI5d7bdxX67kNcfj+QUXXJD7F7J1fGNDJI3ZZvsupW2MTedU+h1LmPVd8JJLLkmxO++8E8jf5qD8\nxrLmStatjp/7ulXh/PPPTzGtaRz/vsr0m422Uc2mAIYPHw7kj6c6xt50000ppjEb17ZXWXgsf9dx\nsaucxoZqG220EZD/fq995pxzzkmxuA5xrVV0klsqleYE9gNGAr3jc+VyecMabJeZmZmZmZlZj1U6\nk3spk9fGvRJoyanMeFVMjSX+8pe/pJiu8MQrogceeCBQ3KjF8nQ17k9/+lOK9e3bF8g3a/j73/8O\nZFfnoHVnG2tNY1bjFbLlQ+I41AzjX//61xSr5Arc9EyNpQAOOOAAIH9FWc2R4jFCVzqbNaeN3q5Y\nBbPffvsB+cZTukJ89tlnp5iu3DfbbE5RU6jumqXVQvzcWnPNNYFsBgKyY+vll1+eYspzo8dDR0VN\nZXTsj3nUzFc9xoSat2hJi7h9l112WYoVNVxpBjrOx+ZH48aNyz0H2axtPRpnabZuiSWWSDHNvMV9\nPzaiaSZqWKTPVchmETVLDvD66693ilVzzMb3R8fWWCmn/VwVc5B/z5uJcnXrrbemmCp64thVw0xV\nHkB1q3zi+6MxGZcO1H4Sc9pslVui5q3XXXddiqniRMsFAtx1112556C6zd7i+6Nzpdi8St+bTj/9\n9BSr53f+Sk9yRwF9y+Vyc7XBMzMzMzMzMwsqPcm9F1gSeLK7H2w2uhI+ZMiQFDvrrLOAfEv38ePH\nA/mlBHSPo2cai8Va/L322guA1VdfPcV09VtLBQAccsghQP4qbrPNODQLXbXVfbiQzTbGq3LKfbzP\nodlmxpqFxmy8H18zDvGen+OOOw6AMWPGpFizXtFtNB1j4wyjFqqPV3l1HIj3jrdiTuu5b8WKg2OO\nOQbIf2499thjQH6GpNnub5ai5ar02Vp0X2itzDLLLOnxiSeeCMCAAQNSTPdVapmLuJ3NRtsVZ2Y0\nQxU/Y/V8rT5r43cB3Y+nWXLIZu00U1fLbZlWmsmPS9Co0i8uf6YZ3FqN13j/6CmnnAJkSzFBNk7j\n/fjNmlPNOusecsi2Nd4nru/htTqGxZxqtnbgwIEppntU41JbzUpVUHGJU+U59hPQa6nVUl0xp2ec\ncQaQP56+9tprQH52vp4qPcn9PnBDqVR6CHg3PlEul4+s9kaZmZmZmZmZTY1KT3KPBhYCXgPmCPHm\nvGxkZmZmZmZm06VKT3J3BJYol8sTarkxtaByrz//+c8pppbbuhkeYI899gDgkUceSbFmLVFqNJUn\nrrHGGim29957d/o5NUJS6SfkmzRYZ7H5kcoTY1MUlfao7Buy0mWP12Kxec/iiy8OwA9+8IMUU9lP\nbN6jspu4FIQV05iNt3ooFkvRDjvsMCBr7gHNW16n7ap0WaZqU+OOQw89NMV0y00sO1OjPzX3aGYq\n64xNSXTMqmdud9xxx/Q4Lhsml156KZAvV21Wyl889qvcNpZ81jq/6623Xno8evRoIF/GeOONNwLN\n22wq0i0UsYmTykBjrNY5XX755dPjopzecccdQGssa1lUAq6GT3HJmlp/h4m3La611lpAvtRetyjE\n5njNSmMxNsbVUl0xz7W+JWjw4MHp8ciRI4F8Tu+//36gduXS3flW9z8CwCtA6908ZWZmZmZmZtOV\nSmdyzweuKZVKf6bzPbl3VH2rqqBXr14A/PznPwdgmWWWSc+peUCczVFDhGadWWgmWjD7yCOz27G1\n4PNFF12UYn/84x+B/MyNFdPsxq677ppiG2+8MZBvB6+cP/TQQynmGdyu9enTJz0+6qijgPyyAWrB\nH5e/8pjtWpxR0JiNlR3K3xFHHJFiaurRSsfYRm3roosuCsDuu++eYjpGaOkIgGuuuQZorUZzjTpe\nzTrrrAD85je/SbGZZpoJgHfeeSfF1JCmlXIaZ0k0c1OPsavjwPHHH59iymlcdkmVdK2U0zjrrFmz\nOHZrlV9Vj8TqQ+U0zsqdcMIJNd2OWlDVFGTjox7LWsnvfve79FjLMsX3tBVzqmZdkM3g1rPiIFYb\naZzG/TweGxqh0pPcH//332M6xMvAYMzMzMzMzMyaQEUnueVyedFab4iZmZmZmZnZtKp0JrclxPXv\ndtllFyBbrzGWJJx55plAtsYgtFZ5QqOo7POkk04CYOjQoek5rX/2hz/8IcViQwErptIkNUA56KCD\n0nMq+VC+Ae6++26gedfCbCYqnYk5VbOJcePGpdjpp58OwIcffljHrWttcW1BldDHEvC//e1vQH5N\nXJfVdy2WgP/2t78FYK655koxNZdSk0RojUY+zULNe9R4ErLjqJqiAUycOLG+G1YF8ftLPb/L9O/f\nH8g39JG478dbblpFLLmsZ5m1vscut9xynZ7TmqPQuHVHp0X8DKhnU0cdW2ODNIm3Jr3wwgt126Zq\niTmt53dufXfdcsstOz0XG3c9++yzddumIlM8yS2VSs+Vy+Vh/338JlNYLqhcLg8qipuZmZmZmZnV\nW1czuXuEx7vUekOmha7S6EotwC9/+Usga6l98803p+c0M+aZhe6pgRdkjaSU57gckJp4teIV20Za\nYoklADjrrLMAmGeeedJzJ554IgAXX3xxijWqDXsr0RXGzTbbDIB99tknPacrnXF2XFUIrubonpof\naRwytfoAACAASURBVIklgHnnnRfIGktBtsRVKyzF0CzU0A+yY2z8jNLxt9FXxltJnB0//PDDO8We\neeYZIKs8AB8HekLH1rhkiJa0+dnPfpZizmnlNIOrxkiQzSQffPDBnWLWPS0lquZzkZYMA+e0J1Qp\np3OsSMtaQuOrDqd4klsul+8Nj++uz+aYmZmZmZmZTb2uypWPnNJzUblc/lX1NsfMzMzMzMxs6nVV\nrrxQeNwL2BZ4BHgdGASsDFxeu03rmkoSAYYNGwYUr9uqNQXjmrgu+eyeyo922223FNtqq62ArARx\nr732Ss89//zzddy61qbSGYDzzjsPyEoVx44dm55TEy+XfPbMYostBmTrs8USpb/+9a8A3HrrrSnm\nUrrKrbvuurl/IStP1JrkkK1Fbt1T+ewxx2Qr9KkE7PXXX08xrZvp8Vq5ZZZZptPjWD6ndZx9jK1c\nLKNVE7T4fey5554D8msPW+V0HI051TE2fm5Z5TbddFMgf6uCSpNPPvnkhmxTqxs0aHI7Jt3CBNln\n0znnnNOITSrUVblyOisslUoXAzuVy+XLQ2wbYPvabp6ZmZmZmZlZ5SpdQmgTYOcOsWuAs6u7Od3T\n1S3N3gL85S9/AWDhhRdOMTWU2G677YD6titvVbF5xCabbALAgQcemGK6Aq4GKLfccksdt671aUZR\n+QNYcsklgWx5kFhx4OVBKhebH2gGfL755gPyjXqOPfZYoPHNEFqNKmNOOOEEID+bc/311wP5JUOs\nciuuuCIAW2yxRYqp4ZTGK+SXurCu6bNMSzFB1kTx/fffT7Ebb7yxvhvWBkaNGpUe9+vXD8g37FG1\njJv4VC4eT1dfffVOz7/55puAvxNMLS0pGmfHv/zyS8BViFNrnXXWAfI51T7fTN8FvtX9jwDwEvDj\nDrF9gJeruzlmZmZmZmZmU6/SmdzdgStLpdLPgfHAQOAbYJtabZiZmZmZmZlZT1V0klsulx8rlUpD\ngFHAAsAE4IFyufx1LTdO4s3iq622GgDnnntuig0cOBDI1sAEOO644wCYOHFiPTaxpam0a/vts1us\nf//73wNZmSLAY489BsCpp54KuAFKJbSWGMD+++8PwDbbZNeGNLZPOeUUAF577bX6bVwbUNODHXbY\nIcXWWmstAD777DMA9ttvv/RcXNvZuhaPu2oyt/jiiwPw0UcfpecOPfRQwA39ekoliocddhgAvXv3\nTs+p4VRcI9sqt+CCCwKw6qqrdnouNu/xbUyVU1liXKtV3x3UGAngqquuqu+GtQEdVyFrTBm/X115\n5ZWdYta1ePvdCiusAOTz9/bbbwPw9dd1OY1pOyoBj9TAL94S0miVzuTy3xPaMTXcFjMzMzMzM7Np\nUvFJbiNolmazzTZLsbPPntzrKi7DIrqRHOD222+v8da1tniVa8cddwSyJSogm8GNszOXXz65uXbM\nsxXTLNjmm2+eYloaoE+fPik2YcIEAE4//fQ6bl37GDJkCJCfXVCDmdtuuw2ARx55pP4b1gYGDx6c\nHu+5555ANvt4ySWXpOe0ZIj1zFJLLQVk1UmxGdpZZ50FZNUIVhnNNuozbbbZZkvPaZZBjSrBM2M9\noVxqVgyy/MWlrpppFqdVqEEqZN8d4nevZlqSpVXEc4RZZpkFyBr6QfYZ5mNA5WJ114ABA4B8g7kX\nX3wRaK7Z8UobT5mZmZmZmZk1PZ/kmpmZmZmZWdtounLluOZS//79gawJEsBcc83V6edUbhDXu3KZ\nVzGVG8S17o4++mgg32RK+Y1lB1dffXU9NrFlxVIOreMc12lUfmN5zJgxk29z1zq51r3YoEfHhoUW\nWijF1ExGtzY0U+lMK1BJ8o9+9KMUm3/++YFsnJ544onpuVgCZl3TLTiQ3YajUrp33303PXfRRRcB\nLqXrKTX6Gz16NJDP36uvvgrA008/Xf8NawNadzze6qQS8AsvvDDFfDyonL5nLbrooimmz6vx48en\nWCwHt8rE77Mq/Y5jM45Zq0z8jqvvWToGQHN+bnkm18zMzMzMzNpG083kFvnggw/SY13xilcTNWt7\n+OGHp1gzXUloJrrSPXTo0BSLzTlEN5PHJW3Uct3ydDU2zjDutttuAAwaNKjTz8UGM9deey2Qv3nf\nimmGceedd06xddZZJ/ccwDvvvAPAuHHj6rdxLS5Wxqgh0iabbJJiuoL7wgsvAPDSSy/VcevaR5xd\nWHHFFYFsluHhhx9Oz7333nv13bA2MccccwAw++yzA/mKrptvvhlw48SppUY+calGPdbnmPWMjqtx\nKStVy9x5550p5mqknovfCXQOEfMcZ8qtMnEmV+cD8847b4qp4W8znX95JtfMzMzMzMzahk9yzczM\nzMzMrG00XblynOZW2eFGG22UYksuuSQAiy22WIrdddddQL5xhxVTadyll16aYrpxfI899kgxlcvF\nmEtmuhbz89hjjwH5NQPnmWceAO65554Uu/LKK+u0da0plscofxtvvHGKqSQpril49913Az4eVEJl\nymroB/DDH/4QgAUXXDDFdFzWmrgx39Y9NZxaY401Ukxl4bpV4c0330zPuXlP5WKp/TLLLANA3759\nAfj888/Tc8888wzQXKV0rUBjd6WVVgLyt9tozPpYO3V0i1M8/uo7g9Z5B4/ZntDxYOGFF06xSZMm\nAfDggw+mmD/Dek63gQB88cUXADz66KMppuNAUWPgRvFMrpmZmZmZmbWNppvJjXQFIDaPGDt2bO5f\n6xnl9JNPPkmxCy64AMi3VHcjpMopp7GhyRVXXAFks4qQXQGfOHFiijnPldOYjcvXvPjiiwA8+eST\nKXbNNdcA+db2lolXWTVLE5vP6XirWVvIGnecd955gKs6KhHzrOYcq6++eor16tULyJr3xCZ/8Xet\na2qIBLDrrrsCWQOqoqUE4zJOPv52T80+d9hhByDfYFFjWFU2kM3mOLdTpsapO+20E5Bf0lHNQYcM\nGZJiOj57aczuaTnBn/3sZymmqqSYU/2cvkOAK2imRGNyr732SrFVVlkFgI8++ijFVl55ZSD/HVcz\nvo3imVwzMzMzMzNrGz7JNTMzMzMzs7bR1OXKVh8qt230DeKtLuZPpbJeW3jaxJI3lb2MGTMmxe69\n914gX2bkcdy1mB/lLa7LeuyxxwL5snA18NG/LkXsXiw51vqMseHf448/DmQNFp966qn0nJuidK+o\n1P6JJ54AsoZ1EyZMSM/FBinWtVlmmSU9Hjp0KJAdf2MpotYaVTkjZO+LjxF58XgwaNAgADbYYAMA\nZp111vScjsnzzz9/iqkcPN4SFRuAWdYU6ac//SkAw4cP7/ScSpQhK7eNn31ao9hjdzIdR7faaisA\n9t577/Rcv379gPzxd9tttwXg9ddfTzHdStaozzTP5JqZmZmZmVnbKNV61qNUKr0PvN7tD05fFi6X\ny/2m9ped00LOafU5p9XnnFafc1p9zmn1OafV55xWn3Nafc5p9VWU05qf5JqZmZmZmZnVi8uVzczM\nzMzMrG34JNfMzMzMzMzahk9yzczMzMzMrG34JNfMzMzMzMzahk9yzczMzMzMrG34JNfMzMzMzMza\nhk9yzczMzMzMrG34JNfMzMzMzMzahk9yzczMzMzMrG34JNfMzMzMzMzahk9yzczMzMzMrG34JNfM\nzMzMzMzaxoy1/gOlUqlc67/RgiaWy+V+U/vLzmkh57T6nNPqc06rzzmtPue0+pzT6nNOq885rT7n\ntPoqyqlnchvj9UZvQBtyTqvPOa0+57T6nNPqc06rzzmtPue0+pzT6nNOq6+inPok18zMzMzMzNqG\nT3LNzMzMzMysbfgk18zMzMzMzNqGT3LNzMzMzMysbfgk18zMzMzMzNqGT3LNzMzMzMysbfgk18zM\nzMzMzNrGjI3eADMzMzMzM6u+UqmUHpfL5QZuSX15JtfMzMzMzMzahmdyO4hXO2accXJ65phjjhQb\nMmQIAAsuuGCKvffeewA8/vjjKfbpp58C09cVkymJOf3WtyZfV5l99tlTbKGFFgKgf//+Kfbhhx8C\n8MILL6TYF198UdPtbFXKaa9evVKsX79+AMw111wpNmnSJADefvvtFPvqq6/qsYktR2NWxwCAWWed\nFYBZZpklxf71r38B8Nlnn6XYf/7zn3psYsuKx4MZZpgByMYwZPn797//nWI+jppZo8Vjl/jYNG3i\nsb8ol85vz8WcSsyjxvH0kFvP5JqZmZmZmVnb8EmumZmZmZmZtY22LVeeaaaZAJh//vkBWGuttdJz\nW2+9NQAjR47s9HvffPNNejzzzDMDMM8886TYbLPNBuTLVlRep3JQgNVXXx3Il9u2YmlAfJ0qh11g\ngQUAWGaZZdJza6yxBgArr7xyp9+N5Zt9+vQB8uXeKqmNf0ulihMmTEixNddcE4Dx48enWCvmVCWa\nkI2neeedF4CBAwem5xZbbDEgP041JmMZ7YABAwBYeumlU0zjPv4tlSY//fTTKbbFFlsAWXk4tFZO\n9fqKyoo1XpUzyPId86wxGUu7F110USAb1wBLLbVUp/9v4sSJAFxzzTUp9qtf/QqAL7/8MsWaPaex\nvEnHzhjT465eh34PstzPOeecKbbwwgsD2bERYJVVVgHy43TcuHEAXHnllSk2duxYIF/C3Kx0HIv5\n0+uLx0LlsuO/Rf9XfBzHusazjgGQHUtiCb1uUYifUc1wq0JXZXNFr71I/N1q7GdF758UvX/NoGhb\nI21rV3mMP9dMr61R4n7WVZly0fir9phsF/EzoqsxW3SrSrPue40Wc9rVsSvmVI+nhzx6JtfMzMzM\nzMzaRlvM5OpKhmZoAY477jgA5ptvPiA/U1DpVWFdOYpXRYp+V8/PPffcKXbkkUcC8N3vfjfFmn0W\nIr42zW796Ec/SrH99tsPgL59+wLFOS3KT7wCV5TToqtOuooaZ3wPPvhgAPbff/8Ua/acxqvBiyyy\nCAAHHHBAimkmVbMv8eeVl6LGDDGnykH83fjeiGYgl1tuuRT7f//v/wHZ/hL/RrPR69P4A1h//fUB\n2GGHHVJs+PDhQDaGY17UNCpe/dTrjVUcisWfi/+PaKZyxx13TLHbb78dgFtuuaXT/9dIRTOBaqoX\nqzI00zpixIgUU2WAZg7j+OrduzeQn7UtmqHT78QGaRqT8edWWGEFIGtIB3DIIYcA8NZbb6VYPXJa\ndDwrapyl6gHIcqWZf8g+h4quumtfjhUCOh6oagayvMX8Kef6ecia+r3zzjspdsUVVwD52fE33ngD\nyI/7WtJxrLvZaT3WuIJsn4/VFsrb559/nmJff/01kL2meOzU/xvHrsbQ/2fvvAPmqKr3/1krHUIn\n1CAJVXoH6aAg7UcvUhRRUUQRvjTpUhQURami9CoiGqRKJ4DUUKQEEkgEpAQIHRV1fn/Ac+fM+052\n9012951dns8/mZzZfffOmTt3Zu557jnR9/JvbNNbb70FFPvfpEmTCr8V/16n0HgW2692l41dun4h\nP87YfvlSCfUg93NZlFKfk9/j5yPdFI3TmBif5/qqgyA/ltifpZSIfVKKnnqJf6KqQuejlxIYapzS\nOAj5OBX7pHwV+5PGsbLrvCypqfwWfdpLvhR6vokqHo2Zcex87733Cv9C7tNo03Nk2fhc9ozULTiS\na4wxxhhjjDGmZxjwS26tVpu+VqvN2I7GGGOMMcYYY4wxU0NduXKtVvthlmXHfrQ9G3ARsBGQ1Wq1\nW4Cdsix7pf3N7E+U4kgqeNJJJyWb5FuNEi0IheNjshjVZY0SqXp/r0w6UVVJThmxHrDkgd/61rf6\n7W+U5EJIIhJ9KnlTlDbWo0wa2A0+lY9GjBiRbD/+8Y8BWHfddZOtLJFZPcp8KpmJ5CuTQ7/RrHS/\nCsS+JvnY7rvvnmySW8dEUpLulUnoyxKwyH9l0vdGvlL7ouSqSjXoyq6feJ0rIZ/8CLDMMssAxXEv\njrfQeAmHfBnlTWX1nMvGSUkDo7St09d+Wd/RdvSFfBml3VtttRVQXBag5IWxxrL6qY4t7iuTxsuX\nUUpaJkOVT6OEevHFFwfgtttuS7Yove0EOs7oP7UxSkOVcG+11VZLNvky9smyviMfyh+xr+t349j5\nxhtvAEXZnv5uHA/uu+8+AC6++OJkk3yybOlIp9DxxRrzCyywAJD7EWDRRRcF8iUzkN+D47G//fbb\nQHHcU9+VLDJKo5XQ7N577002JTaMcvlXXvnwMXHChAnJJrl31ZYcLbzwwkA+DkKe/DHezyWhj+d/\n4sSJADz77LPJpgRw0fdKbKh++vTTT6d9l19+OZAn24PcV/FcSY7bDVJcJcVcffXV+9nUNyHvd7Hv\nPPHEEwA8/vjjySZZeEy2qTFCPvrTn/6U9mkJUUy81w1+q4eWZcXnSfXPuFRG/WTMmDHJds899wDw\n6KOPJpvGwng+Vl11VSDvfzHB5tixY4HqS5gbva0cGLZPBN4G5gGGAq8CJ7SpXcYYY4wxxhhjzIBp\nlHgqTtFvAKyQZdlEgFqt9h3gkXY1rBFxlvLQQw8F6kcH42yvZg5jiYVrrrkGgKuuuirZlIxFCY8a\nEWckTz75ZKA7Zos0c73iiism24477gjkyQGgf4Qq+lSzRXGm7C9/+QsAt99+e7LpN/bYY4+m2hZn\nic4880ygO3yqiML666+fbCqvFGfC+0amYh9SpECz4JBHYjS7CbD22msDsOmmmzbVtjgb/Lvf/Q7o\nDp9KnbHSSisl2xxzzAEUIw99E6XEJBaKVKj0D8DDDz8M5DOZkJerigmZytBvvfzyy8mmqEYVIrmN\nylsosZH8CMUIoJAPdbxl/TRe+4rwxEioZt1jJLcMKWjiLHOc2e8kZcliyohRWCkq+ka/oTzBjKKx\n8be0TwmPII8Sxb6umfuyhFbxHGnWXX8DitdFJyhLPKVjj/cZRQ6VwAvye3vsm4pox8i2fqMs8Z76\n6WuvvZZsuubjs4OibHGclq8uu+yyZOs7zgwGigTGSPgiiywCFKM5ivrERI5CYyLk5yEekyK4GiPi\nPvXPWGZRvxvHaZ2P3//+98mm54KqRXKlCpLPII9AKsoL+THH/qzrKyaM0/02JghSFFi+jEmVhg8f\nDhTHCv2NeJ955plngOIYUVX0vB7vp1JqxASDGkd1D4C89FxUa+mYF1tssWTTNSAf6Xvx92P5Sd2v\n4m9V4Z7dLPJlVBGV+VRjYvycxoi77ror2VRGMpYBVT+VMiGWs9Q9Svd6yH1ZJT82esnNah9ewZ/g\nwxfe18K+14GZSr9ljDHGGGOMMcYMAo3kyjMA/wE+4EOJ8rJh33BgYtmXjDHGGGOMMcaYwaBRJHdY\nn/+/GrZnAQ5pbXMao0QLv/zlL5NNsoSyeqKSGT322GNp3wEHHADkMkXIZUtR9iUpWLPJeSQJA/jr\nX//a1HeqgCQJhxySn04lsojH3jfxyS233JL2HXXUUQCMHz8+2SSDilI+ScEa+VTn78EHH0y2eA6r\nSDwmSWu22WabZJMULMqzlARFEroLLrgg7TvnnHOAXEYSPx/lfZJQNfKpzt+1116bbJ1OPjNQouxQ\nCVWiFEeSwSg1lmRM1+Mll1yS9t1xxx1AcamC/BIlu0oSUiZXjlIcycjOOuusZIttqRJlEiLJ5KLk\nTdLgeBxPPfUUAE8++SRQTDQjP0fZl4iSyVNOOQUoyvbK6u+NGzcOgPPPPz/ZoiS1E5QlKFNfi23V\n+VebAa6//nog9xnk1+2LL76YbOqn6otR7q2xMx635LPrrLNOsh155JFAUS6qdkZpnhKNxKUPnZaU\nlflP/S9K3uSraJMUMfpDx1KWTEZjbFn91nhO1Rf33HPPZFPilShXlnQ0Sp11HIMpzZOvogxd98zY\n10aPHt3vu/pO/JyWcUS/aQyW3+LYKT/H5yZJSGPSSiUFitfJqFGjGh3eoKBzHJdLyM96foHcL/G5\nRM+U8TlI/orPp/Kvxsw4dupzMZHk8ssvD8AXv/jFZJP/4rNm1aTfQn0tJtPSMUefatzTfRrg7rvv\nBorjmcaB6NO+yxfiEiwtc9CzLuTLn1QvHLpD+i10rWpsh9ynkrxDfj+/9dZbk03L3qL8XX3y3HPP\nTTb5UuNffB7T0odVVlkl2ZR0Lo5Hg03dl9wsyybU2XcvcO/k9htjjDHGGGOMMZ2mUSS3EsQZ1dNP\nPx0ozmaXlVvQLI5mhg4//PC0T7MYcUZZs5RxRvy73/0u0LhkjmaSt95662SrelrtmMBDpW1iyQZF\nX+MstWZ6NPN12GGHpX2azYwzifKpZswAvvrVrwLlPo2/pdmnr3zlK8lWdZ/OMMMMaft73/seUFzs\nL5/GaIRmjW+88UYATjvttLRPs2FlyW9iyZAtttiisC8SZzOVgn+//fZLtqr6VP0j9p2NN94YKJbL\nUH+Ls7FKpqCIWkx8FpNu9f2tmBBJkYey8kPx/KlMwdlnn92vTVWgrIRQVAGoT8aombZjuRlFgrQv\nqgvKjldjcuxf8bwJ9c/4+yec8GHS/hghGaxoWdm1V5YgKiYyU2QlRljU72KkQLPj8l+8VssiyYqW\nxchlWaIgRY7++Mc/JpvOXyyf02nKjlPbMcmO2h+vaR1f7GvaLkuuVo/oUz0fRJ+q78Y26VzGSHgV\nIrlqd0xcpHtnVKqpjfGYptR/ZcT7ufppLPeka78sAWjVUKmV2P9uvvlmoBhx1bUU7wetTEYWr1X1\ntfiMJv+pvFW0VQ1FGx95JM9Vq1JJ8Z4s/0YlQSt8qXMU3yVWWGEFoNh3y55jq4ruz/G55aKLLgKK\n/VTbrT4mJT6LSYClOND1Ap1PcNiX5gqeGmOMMcYYY4wxXYBfco0xxhhjjDHG9AyVlitL9iLZMMDm\nm29e2Dc5JEFQqF4LoiEP28dF1KobpdquUFyk3peYfGG33XYDirVLq4rkGrvvvnuybbbZZkBR9lAm\nzZO/lKgiynnk0yj9UEKB6667Ltli0pm+xKQeO+ywA5DX56oaUfImyWesU7vhhhsCRVmxvhPl9erH\nSnIRk/2USRZVfy/Wc1YCgLIkOTH5jWofD1bN0UbEviO/RXmWEppEyZEkXWWy66effhoor4MXf0t9\nMiY6ijU6RVnirv333x+obsKKeJzyW1ySIZnjc889l2ySDsdrb9KkSYV/y/wd+59k5ieeeGKy9a0N\nCbncUksmIK9Z3ulkU1NCmYQ9JvMQkolGv+m79WpUR19pOcS+++6bbKrvGv/unXfeCeTSNcj752BK\na8sSP5WNcWV9q6zG85QS/4bk3ptsskmy6T4Yx05Jv6O0sgq1INWvyiSBjWpkt5L4d5UcdJlllkk2\nnd/4jFRVSaiulVg/uOxabff5j7+l6zzW7lXiqW6oca9xvixJXCevo7IEX0poCXlixar2zUhZgstO\n9gWN03rOgvweX1ZjfLBo6iW3VqvNDOwDLMeHZYUSWZZt1IZ2GWOMMcYYY4wxA6bZSO7lwCeBK4H+\ndSJaSIyuKlnBWmut1W9/WUKQOIuhGaODDz4YKEYJyyJvZ555JpCXKpgcmjlVchTIE9xUYWa3jBg5\n1ExLTIgk/8UZdM1yxUQVijZ+5zvfAYrJLhSRXH311ZNNkQRFHyeHouIxOZgSFXSDTzXLGqPUZYlm\n5NOYZEclg1QuKM566fMxHfxll10GFEvblEXdlQjn//7v/5JNpV6q5tOyCLcSo8X+p6QQ0aeabY/l\nKFTaStHJOC7Ip7FPqp+uuOKK/T4XZ3SlBlHZFsjHlar6NI6nssUkSSoFVFbaQVFbyK/1ejPcug4g\nLxcUywuoLTEapv58xRVXJNtgJkcaKBozo1/KomZTGkmLM+K6lqNP1U8nTMgLISh6HhUbVeqfZW1p\nRfKjZonXxK677grkKhHIx5zYJxWtqFrUrF6/6uQ5j4qR7bffHijeD5UEK5YyqpovRVm7BuP6if10\npZVW6md74IEHgOr6MTIYUdt67YiUKUu6gaqc95iMT/2zkdK2kzT7krsqMHuWZf9u+EljjDHGGGOM\nMWaQaPYldxSwGPBIow9OLTGao7UxSukOsMQSSwB5NBby2fRYsuGkk04C8uhPnK1RyZW4Bq/ezENc\nm7HddtsBecmX+PtVJfpUMy0xcqM1nRHp/W+44YZk09paRXiiTxUZjmtFY0mdvsQSJPJpLFlSVZ+W\nRcg0oxb7n9boxb6jqF/szypwruhB9Okcc8wB5EoBgPnnn7/f5zT7+PzzzyfbNttsA8CDDz6YbFX3\naVlh9zjzr2hKXNejz8W1NvJl2ays1uCpfAHAyiuvXGgH5Oc0+m+nnXYCYPz48clW1bU7ZdF9+SUW\natd2s5HIMjR2RnWLxth4TvX70fdSb0SlTVVn08t82q61jvotlc0C2HvvvYHivUrj6CGHHJJsGocG\ney3UQOjkOY8lL1ROLd4jFXU844wzkk3jS1X75mC3K+YwUL6PeL+RYiNGfQa7zY0Y7PbF8nnK8RHX\n+2vNeFUievUYbF+K+Ey6+OKLA7maCarTzm4ijp0aB+Jzr57XBsu3zb7k7g5cU6vV7gEKWTWyLDu6\n1Y0yxhhjjDHGGGOmhGZfco8F5gfGAzMFu6c9jDHGGGOMMcZUhmZfcncARmRZ9mI7GwNFyaCkiFEa\nLGlBlNwp6UaUvClRimSlKj0EcOGFFxb2TQ4lCFp22WWTLf5uNyLZlcpMANx9991AUQaq5D4xCYzk\nRzpHMSHSrbfeCtSXKEPuUyVSiLZuRT6KqdRVKiGWFpH/oryor9Qo+u+mm24CiinuRZR+SD675ppr\nJls3+bSsPIP8Fv0nqVuZRLgZOS3AySefDBST95QlrlOSqZicrpuu/XolL1qR5Kds+UcsS6axNZ4r\njRGxBE4cc6pOJ5P7zD333ECeEBHyZGzxmjj11FOB4jKRmKyt6nRSwqblTz/72c+STeW0YjmOgw46\nCOgOae1gt0vX+Te/+c1k0/KnuCxHz1xVXTITGWyfamxde+21k23BBRcE4Lzzzku2mEDQ1Ec+K24x\nOgAAIABJREFUnXfeeZNNcuWbb7452bpB+l01dF+CvGxYXNI12NfTJxp/BIBngO5Z4GOMMcYYY4wx\n5mNJs5HcC4CRtVrtV/Rfk3tz+VemjPjWr2jsQw89lGyaCYwRRs1cl5UK2WCDDQC49NJL0756EdwY\nAVPh7bKiy91E9IsiuTFJkXwaZ1nrpXxXlOGuu+5KtnoR3BgBW2GFFYBieYtupCxRT+yT8l+zxeOV\nSO13v/tdsinJWhnRpyqx1U3R2zIaRWgHmuRHY8DOO++cbLvtthtQnrgr+m+TTTYBuit6GymL5Pbd\nNzXEZHUqfxUj5vrdqG7YZZddgO6K3kbaPSMd/afyNUo+B7lPb7/99mRTuaBuit52knid77DDDgCs\nt956yaZzGpNQqhyZozqNUSLEqOLQOH7aaaclm8bWwY7qdANSHCh5JOTPbZdcckmy+ZpvHj0LxNKZ\nKs82bty4ZPM13zwaW2PSOakOY5nHwabZl9zvfPTvcX3sGbBw65pjjDHGGGOMMcZMOU295GZZNqzd\nDTHGGGOMMcYYY6aWZiO5HSNKMFTnL9ZarJc8JX5u1VVXBeBPf/oTUKzlVIaSTMRkSu+9997AD6CC\nxHqJ2o4yLlHm0/i5mWeeGYB77rkHyJN2TA7VypLsG7pX/inkl0YJNJqVZUk6/6Mf/QiAL33pS3U/\n/8477wB5bVcoSs+7Efmq1TVnR4wYAcBZZ52VbHGMEFoWse666yZbt0u/RavlgZJ4SdIJ5UsVdO3H\nxF1OlFKfPfbYI20rMVrsr6qR+bWvfS3ZeuUe1S5mmWWWtC1p93TTTZdsjz/+OJCPv9Bd9YUHg9gn\nVWc4Ll/Q2KBlDFDdeuJVZOGFPxRHxnrOWsYU67Zb+t08koBHn15zzTVA79zrO42eXeM7k94NqrQk\nabJvfrVa7Yksyxb/aPs5JlMuKMuy/qlfjTHGGGOMMcaYQaBeeHPPsP2VdjdElEUTm10MPtNMeQlf\nzdLEZB59iVHjYcM+VGT34sz41JS+iEm6NJsYZ237EmdslaK926O3ZbRqFnWNNdYAYP/995/sZ2L/\nX2eddYC8bJbpz/TTTw/ks4plKo7o06233hqAsWPHdqB13YkUHb/85S+B8rJWUd2w4447AvDMM890\noHXdzeyzzw4US9soWhbLBSmB2osvtr2SX9ej/vrTn/402aQ8ivf4r371qwBMmjSpg63rbmKiGZUP\ni5Gb7373u0Cu5jCNic9ZG264IVBUvpx00kmAk01NKXpmjSqNP/7xj4BVBlOK3rek7gK49tprgWr5\ndLIvuVmWjQrbt3WmOcYYY4wxxhhjzJRTT658dDN/IMuyw1vXHGOMMcYYY4wxZsqpJ1eeP2xPA2wN\n3AdMABYAVgauaF/TmkdyRCWZAphxxhkbfm/RRRdN270oU54aJPeS7BDymsP12H777dP2Cy+80PqG\n9QCx9uX1118PlCcCE8cdl1fueuCBB9rXsC4mSpJvvvnD0t1x+UJfLrjggrR9ww03tK9hPYJk8nvu\nuWe/fZLuq7YrwHXXXdeRdnUz6rOSeMWESJLTX3bZZcl26623dq5xXY7u7dtuu22yyacXX3xxst1/\n//2dbVgXIwn9AQcckGyS2p977rnJ9uijj3a0Xb2A6g0DfPGLXwTyJXfgZ6kpIcpoV199dQAee+yx\nZHvttdc63qZuJz6nLrTQQkAxcdcbb7zR6SY1pJ5c+avartVqlwI7Zll2RbBtBWxb9l1jjDHGGGOM\nMWYwaLaE0MbAzn1sI4FzSj7bcZS8R/82YuTIkQCMHz++XU3qeoYMGQLAaaedlmz1oo1jxowBitEc\nU0SRG/U/yFPbl/HSSy8BcMQRR7S3YV1MmeJgxRVXnOznNXv79a9/vb0N6wFi+ZVLLrkEKC/B9Prr\nrwOw++67J5vLW5QTx1AlPFt22WX7fU7lgvbee+9ks08b89nPfhaAs88+GyiWt1KptUMOOSTZmk1q\naWD55ZcH8gRokCeV/PGPf5xsVUo6U3UUbdx111372a666qpka1Sy0PRn7rnnTtuKlMfnU/t04MRE\nvkpAFxMhVrH8Wv8nlnLGAt/pY9sLGNfa5hhjjDHGGGOMMVNOs5HcrwNX1mq1A4AXgHmB/wBbtath\nxhhjjDHGGGPMQGnqJTfLstG1Wm04sCowFHgRuDvLskGLTcckHeeffz5QLqUTMbHUVlv53byMWKvt\nhBNOAIpyr77Emm2SMpnJs/LKKwP15bRRQrP00ksDltTVQ/3zJz/5SbL1HQeiT9dee+1+NlNEktrv\nfe97yRaTpUHRf5Iv/vOf/+xA67qbmBBREk+Nu1Hqtd9++wGuNdoMUQK+5ZZbArDMMssAxX76q1/9\nCujNuu3tREtqDjroIKCY0O+ccz5csSYpuBkYSt6z1lprJdvDDz8MwHPPPTcYTep6NJ5+/vOfTzY9\nQ8UkSWbgzDzzzGlbcvCYzKuKS2qajeTy0QvtHW1sizHGGGOMMcYYM1U0/ZJbFZRY4sQTT0y2mH69\nL5pZ2HTTTZPNiRGKKPK12mqrJdsuu+wy2c/Lp/vvv3+yuQRTOXHWW2VrYrkbIZ+ecsopyeaIQzkx\nUvt///d/AMw111z9PiefxhJBjz/+eJtb1/0oocT3v//9ZJPP5dOHHnoo7fvLX/7SwdZ1J4o2qr8C\nDB06tPAZJe8D+P3vf9+ZhvUAMUHawQcfDORKr2effTbt+/Wvf93ZhvUIiopLBRPvS7/5zW8Aq40G\niqKNq6yySuH/kJcOiko50zx6R4iJp55++mkA3nnnnUFpU7ej+1eM5EplVMWyQZFmE08ZY4wxxhhj\njDGVxy+5xhhjjDHGGGN6hkrLlSXhUM1WgOOPPx4o1mSsV79VNZxuu+22NrSw+5CvYr2rddZZB4CL\nL7442VSrrYw333wTgDPPPLMNLexeYj9Un1W9RsiTTJQhGc1hhx3Wnsb1ABoP1F8B9t13X6A86dz7\n778PwJ577plsVUyMUAWmnXbatH3eeecBRam9+Ne//gUU6wxbqtiY4cOHA8W+qGUL8unhhx+e9lWx\n3mDV0HgbE0kOGzYMyKWeSowE8NZbb3Wwdd1NvP9vsskmQJ6A6qabbkr7xo8f39F29QqS1C6yyCIA\nPPnkk2nfo48+CvheNaXoXhavd8mVvVRxytBYG+/1quWu56yq4kiuMcYYY4wxxpieoXKR3BgNU/II\nlagA2GGHHYDy5D0RzYJtvPHGwMc72hB9qhnaFVZYIdnOOOMMAGaddda6f0c+3WeffQAnRhCKMEb/\nSXGgWXAojzbKpyeffDLgxAhCfTZe5yrBdO655ybb9NNP3++78um1114L5GoO0x9FZ2KSqTXWWAMo\n9lf59JFHHgGKZQNMObH82o9+9COgqEqST1XW4pZbbulg67qf2WabDcifCSAfL15//XUArrzyyrTP\nkbHmkW8BlltuOSC/N1133XVpnxUHU4ZUMoo6alwFKw6mFD0z6JkgJkN96aWXAI8BU4qeBeIzwYQJ\nE4Dql2N0JNcYY4wxxhhjTM/gl1xjjDHGGGOMMT1D5eTKES1ovueee5JNNdqiTFEyhbiofLfddgOK\nMhCT+0hyLoBXX30VgAUXXDDZyhaaX3TRRUBe7/XjTFmysyitVWKJMqJkRvVGjzzyyNY1rkuJPtW2\n5LQAyy67LFCs1SaiTzVGKMmPJUofUiYBX3jhhYGi5FNLGqLf3n33XQCOOOIIoPoSpU5TtiRES2UA\n1lprLaAo95IPJVOWj83kiePqZpttBsCSSy6ZbPKvlijE+5xpjBJSrrnmmsmmpGn//Oc/AXjttdc6\n37AeINbCnXfeeYFctlz1WqPdgMZgjb9KkAqW1U8t8m1cTtctSbwcyTXGGGOMMcYY0zNULpIbowea\n6b733nuTbdVVVwWKs7eK3Dz++OPJ9nFONNWXMp8qpTrAFltsARRnb8XVV1+dtuNC/o870aea0Xrl\nlVeS7cADDwSKpasWWGABoJg46ZlnnmlnM7uK6FNtxz6nJDKxvMUXvvAFAK655ppkk9JApVnMh8in\ncWyUiiP2ye233x7Ix1WAo446CoAHH3yw3c3sSmIkV5Hyt99+O9nGjBkDwKRJk5JNCiVFxx1tmDyK\n0MZkXkOHDgXyPgx59OaPf/wj4AhZM8S+O8ssswDFcndKMKnkPfE+V698oykSVUnzzTcfkEfHY4RM\n0XTtA6uRmkGRco0VZT6NSppuiURWiXiP0n0u+lTPFlXqr47kGmOMMcYYY4zpGfySa4wxxhhjjDGm\nZ6i1O6xcq9WqE7euDg9kWbbilH7ZPi2lUj4tk3FVScLRJJX0aVm94S6SHlXKp/JlTEYl2Vesg11x\n/1bSp5LIQZ4oMUrFJaeraBKvSvm0LGmakvbMPvvsySafSsIc+3AFxt9K+TT83bStxF6SgkNe/11J\nvGLdcUlqB9G3lfRpn98AiknT1Genm246oLiMQduDOC5U3qdlaLyVT+P4q6R+cQlTh5c0dqVPhXwZ\nJfe65qOsXs8JHRoPmvKpI7nGGGOMMcYYY3qGTkRyJwIT2voj3ceCWZbNMaVftk9LsU9bj33aeuzT\n1mOfth77tPXYp63HPm099mnrsU9bT1M+bftLrjHGGGOMMcYY0yksVzbGGGOMMcYY0zP4JdcYY4wx\nxhhjTM/gl1xjjDHGGGOMMT2DX3KNMcYYY4wxxvQMfsk1xhhjjDHGGNMz+CXXGGOMMcYYY0zP4Jdc\nY4wxxhhjjDE9g19yjTHGGGOMMcb0DH7JNcYYY4wxxhjTM/gl1xhjjDHGGGNMz+CXXGOMMcYYY4wx\nPYNfco0xxhhjjDHG9AyfavcP1Gq1rN2/0YW8mmXZHFP6Zfu0FPu09dinrcc+bT32aeuxT1uPfdp6\n7NPWY5+2Hvu09TTlU0dyB4cJg92AHsQ+bT32aeuxT1uPfdp67NPWY5+2Hvu09dinrcc+bT1N+dQv\nucYYY4wxxhhjega/5BpjjDHGGGOM6Rn8kmuMMcYYY4wxpmfwS64xxhhjjDHGmJ7BL7nGGGOMMcYY\nY3qGtpcQ6iS1Wi1tf+ITnyj8+7///a/f57Ms67cdbaZI9C/YV8YYY4wxxpjq4UiuMcYYY4wxxpie\nwS+5xhhjjDHGGGN6hq6VK0uGHLdnmmmmZJt++ukB+OxnPwvAv//977Tv7bffBuCDDz7o93f/+9//\npm3tj1LnMtlzrxB9Kmmy/AfwqU99qvDvf/7zn7TvX//6F1D0X5kE/OMmC+8r8Z6cTZT5yhhjjDHG\nGNM8juQaY4wxxhhjjOkZuiKSW5ZQ6pOf/GSyzTLLLACsvfbayTb77LMDeTTs9ddfT/vGjRsHwD//\n+c9kUwRSUUqAN998s9939Z0Y0e2miJt8WebTeOyzzjorAGuttVayLbTQQoV977//ftp3//33A/DC\nCy8km/wWP/fee+8BeeQX8ohwt0bJy3yq7dhPp5tuOgDmnHPOZJtmmmkA+MxnPtPv706cOBGAN954\nI9nKIubyWy9FzOtFwKPiQP6Nfu577LFfyW/dev22mmZVBcYYY4wx3YQjucYYY4wxxhhjega/5Bpj\njDHGGGOM6Rm6Qq4cKZNhSr4YZcVCks/nn38+2SQNnXHGGZNt7rnnBoryPf29mGCpLBlV1SmT0Zbt\nj5LZhRdeGIB111032VZeeWUA3nnnHQDuvffetG+jjTYCiufltddeA+Dhhx9OtvHjxwNFWbNk4dHP\nVfVvszJaSb9jH1t66aWBoqz+c5/7HJBLmeXbyNixY9P2gw8+CMBTTz2VbJMmTQJyKTjksubo06rK\nT+vJvaNP1T+1PAFg/vnnB4oScEmXJU2O0vh3330XgFdeeSXZyvyn70aflcnqq+TTRte5KJN7xwRz\n8nPZuFdP7t2sX2Lb6iWnqwL1/NjoO2XfbVffKfutKvnRGGOM6TSO5BpjjDHGGGOM6Rm6IpJbNssf\nZ8RfffVVAO6+++5+31UEoizyqzJDkJcVmmuuuZJNSYFiaSIlASorP1RVGkVJZIsJop588kkATj/9\n9GQ7++yzgTxCGxN3yVeKiAMMHToUKEaJFlxwQaC8/JCibJBHJqoWjVB7yqJRjRIc6fiee+65fjZF\ncGOUbYYZZgCKSZXmmWceoOg/9d2XXnop2RSpbBSd7DT1IuFlydA+/elPJ5ui4iNGjEi2VVddFYBF\nFlkk2eQ3+V59E3IfRGWHlAbPPvtsssmnivJCruyI/VTnYTB9Kl+VlQCLfUfqAvkHYN555wVgiSWW\nSDZFxePf6xvFjj7VPqlmACZMmAAU/afrPI4zb731FlDsp2VjdqeR38pK1cUEfdqO9wgl5osqDvko\nHqfK2smnsf+rn8brXNtl10kcZ/QbURWi3xpMn8pXZf00+lTX/LTTTptsZeqCsiSGAx3jykrl6bdi\nm3Svq9q1rzaWjaux/erPUa2ldkf/qZ/ExIZ9P9+IsnOq343jufwXx4Oy3+00anc83maTHqr9sWRl\nWWLIKSX+vnwZz6l+Pz6bVUEVJx+V+aBsPIg2HVNZecpWEK8dnfsyn5ad08FEPirzRSNVVyv7ZCPK\nfKprP75HtastjuQaY4wxxhhjjOkZBhTJrdVqiwIbffTf67Mse6re540xxhhjjDHGmE5S9yW3Vqvd\nCuybZdnoWq22FXAecAuQAcfUarVdsiwb2f5m1keh7yhlGsj3IJc+RcndsGHDgHJ5Wi8hmUCUDkiS\nHGu09v18mbzg73//e9qWpHHIkCHJJlmk6hhDLq2NUrCqyZT7Uk/2DbkcREm1AB555BEAnn766X7f\nLUtqVJYQTBKlKLVXkjAlYYLcl1EeWVWf1utPZbJ2SVwhT2QWberHkheV/Y0o0ZM0OUrpllpqKaAo\n+Rw9ejRQ9GkVKJN7Sx4Wxyv1mfnmmy/Z1lhjDQBWWWWVZJtjjjmAoj/6ysdj/5Mvowx/zJgxQFGu\nLEmolpcA3HnnnUAxuVoVZKA6zihjlcw7jl3Dhw8HYMUVV0w2jXFRriziMWl81Ofi5/U5jcOQ+zeO\nyRovYtLF66+/Hshrl0M1kqapz8w888zJpr6o6w1g2WWXBYpLEPTdeD3qmKM0s+85iudK5zT+DY0b\nZcuZ4th95ZVXAnDttdf2+/2BPne0Et1bZ5tttmRbcsklgfzaBlh88cWBoj8kw4zLNHQdxv6ka0Df\njfcZ2eLYoz4W5eZKrBjH3auvvhqAc889N9lefPFFYHB9qvEvJjPU9b3++usnm5JGxuNUn3nggQeS\nTddhfLZUH1N/1TIuyM9V/H19TkshIL+O4jV92223AXD88ccnm87vYC6x0zJA+RbypUYbb7xxsskP\nUQL+8ssvA3DNNdckm5Ymxn6qPqjrPP6WxhT9C/nYE/tzfFYV9913HwAHHHBAsmk532D2U/WPmIhT\nPt1iiy2SbYEFFgCKY5yu80svvTTZlNQ0jo99l+fpOgZYdNFFgeI1sfzyyxf2QXFsEloittdeeyXb\nE088AbR+yUKjN7ZlgIc+2j4C+HKWZbcD1Gq1NYAzgUF/yTXGGGOMMcYYY6DxS+5/gCHA68BQ4K6w\n76/AgmVf6hbiDJhmGOMsuWaJ4mx+FRIjdALN4Ax0piouztd2nL1VopuyGeWqRhqnhug/HWdMtCEG\nmtQjzh5r9nOhhRZKNpUYqppPm21PWeIuzTAqSggwbtw4oNzP9RJWlCUUijO/monUrDrks7dVpSxK\nF/2tfhfHuDL/6XqNZZZ0LSsaEFUX+rvxOtdMfExotd122wHF2WBF4tWOqlCWzEsRgpi4S7PUSggH\neUQ2KjDkr7JrX38v+k8RxqiMUVQ3JkdcbLHFgOK4+49//AMoRpPEYI4HisbG5IQrrbQSAGuuuWay\n6dqL9wiNezFqK3/FKKKiW0oEFtUIZYmWtB3Ps+73ZWqPG2+8sfGBdhDdT1dbbbVk23zzzYFidFx+\nKUv8pIgk5OXtogJE9xr9G30lYr8qe0bSd+I+KRkUJYc8eWJZYsdOscwyywCw2WabJdsXv/hFoHid\n6/oui2LrbwDstttuQP1EXI1KldXzQdz35S9/GYALL7ww2TTGDqZP1a90D4D8mo+RyBjBFbqHRGWC\n+lFZwrqyvzE1qO3xepIabzAjuYqA77zzzsmmMp9RaVXWt5ZbbjmgGPFV343+k0/LrvmpQe1cYYUV\nkk3PrK1+x2rU8suBU2q12gzABcDBtQ/5BHAQ8GhLW2OMMcYYY4wxxkwFjV5y9wP+BzwPrA8cBbwP\nvAfsBuzR1tYZY4wxxhhjjDEDoK5cOcuy94GvfJRVeWVgPj58yX0EuD3LssGL1XcASZ6iRKoKSVG6\njegrLZaPEgrJwnrdp604vjIZqpLfxIQuVaiL2Qpi+yWVLZMIDfQ4y+rcxYRSkkxGaah+t131+qaU\nejX7YlvV/pggSkk9ohxJ12P0c9+6eo1q80nmFJMpSRYZpVR92za5v91p5LcoWZXMO9qUWOuxxx5L\nNskwY38qq1ut4yyrw6h9UTqmsXPPPfdMNslUy+T6VavrWNYu+S/KsnXMzzzzTLLJf/FzSmAWJYuS\n7yrZStyn5Q7x97XEQwlToH+dbSiv8VwFn+r4ouRTzy1lSTLjs4z6c0yIpH4U7yVaxlFWO1o+iP1P\n40CspV0md1R/iEkDO1m/c3JIJq/7KuTLEuIShLI6pWVLZMrqBterbywajbH1vhPPaRWWg6m2/ec/\n//lkUx+LY1zZ8dWrpd3ou61Ey0BgcGXKQskiozRe99ZG8uKy+uB993WCOJ63ayloU6mCsywbA4xp\n+EFjjDHGGGOMMWYQ6b16OFNJTIpxyCGHAHD44Ycn28cl8VQriQlEjj32WAAOPPDAZOuVqGMniWnZ\n9913XwB+/etfJ5siRlWINrSaVvcTzVxGnyrpyB133JFsijpVzadTk8wrRrUG+vfqfU+/Fa99RXJj\nuSDN5FYtkltWlk5KghiNVXLCeExqf/TzlEao4qx6WQkQJQNS6RXIS5XFiHMVfKpIaoyI6JhiUjf1\nHakMII/2xWOST5u9HstKsikacvrppyeboqIxaqvSLLG8RhXGAUW4VX4D8ui0kmVB3mdvueWWZLvr\nrg/ziEY/i3hsiqBJlRGvCUXUYlkylQXZcMMNk02RpdiHNbbGkmJVeL7SmBSvKfkvRhPlo5g0T2WR\nHn/88WST32KySCmElNiwrHRgHDuV9ChG2MsiyX/7298AmDBhQrJVwae65svGpNh+bccxVomz9C/k\nqpaoFNJ29JGQT2PkV1HMRonU5Ev5FqrhUz2PxHtEvShs7GPybxzPlBAyjo/yV71kXs3+fkQJ5h56\n6KFka5dPW5syyxhjjDHGGGOMGUT8kmuMMcYYY4wxpmewXPkjFI6PST0kwYnyqirIFLoFyUB22WWX\nZFPNs0cfzatPlcnwTDny6VZbbZVsSuogSR0UE4yY+pT5VLWHjzrqqGRTopYqSD+nhk60XzLGWMNP\nCXFGjRqVbJIGVkH6GdE436i2ZLO2KSX+LSUViTJQScui3FdJsKp2r5L8t6wmc0RjV73EZ1ND7Gu6\n70dpqIiSz9tvvx2o3r1KUuMHH3ww2cpq/0rOLJkg5H5u5NNmfB6ThGmpR5Q46m9EabJquVbtXiVZ\n6p///Odk07UXE/XcdNNNAFx11VXJ9sILLwCNkxOWLWkQ8tvCCy+cbCNHjgSKUlz9jSg5Pfnkk4Gi\nVL0KyFcrrrhismlZQJTHPvvsswCcddZZyXbdddcBxeOUj+L4LJ+rP0XfSsqsmsUARx99NFCUkYs4\nLp100kkAvPnmm3WPsdP88Y9/BGC99dZLNvXTeO1p3L377ruT7YwzzgCKSSh1f4514LWcSZ+Ly5uU\nROywww5LNiXBKpOAx7HzhBNOADrj06Zecmu12szAPsBywAxxX5ZlG7WhXcYYY4wxxhhjzIBpNpJ7\nOfBJ4Eo+LCHUc2g2KabiP+aYY4DqzYp1CyohsOmmmybb8ccfDxRnlLs9MtZJNFO36667Jttll10G\nwJgxeQL0qkVxqsyQIUMA+OY3v5lsiozceOONyVa1KE7ViLPqiy22GACbb755skkRc8EFFyTbu+++\nC1R3DGh3pLYRcUZ+u+22A3LfQu6/6FMlJKmaT9WeGKFVlCFGXdpd8iRG4773ve8BeRQD8ijO5Zdf\nnmy6X1XNp/KlEqBBnngsJu9R4q523Rei/3S/j0ma1M4YHZX6oGo+VcTwgQceSDa1P6r6pEaLCcpa\ncSw6R7ovAcw///xAcYzVNSOVAeQqmar5VEm8FH2EXO0Ty8ice+65QDHC2Io+q3t39EuMIAv59P77\n7082ReqrpjaSOiMqCZQ8Ko4Hv/jFLwC49tprk61MQVNGvX6kc6pEc1A/iVf0qcbWTvTTZl9yVwVm\nz7KsfypOY4wxxhhjjDGmIjT7kjsKWAx4pI1t6Thx1kEFz6NuXSnuHRVrnujTYcOGAfnMMuRlbspK\nl5hyok+XXHJJoLhe7Mc//jGQR3VMY2KEbIMNNgCK60O0FjeWkaja7HjVkHID8rJWcew85ZRTgOJ6\n/KrNjtdjMM7/PPPMk7Z/8IMfAMVojtbhX3/99ckWI6VVpKzUVOwH7fKz/PalL30p2XTtx/Hg+eef\nB4rR8arer+S32D6NYzHC2K5nGPnt4IMPTjblMon9VG3SmlHII81VG1d1/cT1ww8//DBQVKDFSHkr\nUQT8xBNPTDZF6CI6v1IcQnXzRqh/xjI86jtxPbnut61uv+5N3//+9/v9flk7Dz300GRTP6iaT/W8\nF9UROiatY4Y8gtrqMUDqjdVWW63u5+TTffbZJ9k66dNmX3J3B66p1Wr3AIWialmWHd3qRhljjDHG\nGGOMMVNCsy+5xwLzA+OBmYK9WlMbxhhjjDHGGGM+1jT7krsDMCLLshcbfrILkPxz6NChyabU1/fc\nc0+yxZTlpj6SJs0111zJtvLKKwNw0UUXJZtS7FdN+lFlYnkLpeA/77zzkk0JCLpJ+jmUL7F8AAAg\nAElEQVRYqJ9KUgew1lprAXDxxRcn28033wxUX/pZBeTT1VdfPdlWWWUVIC8dAfDb3/4WaD7pxccZ\nJWU55JBDkk3S5SiZPOKII4DuSI5YL4lXJ+4HSuQTJbNKQhX75LHHHgvAK6+80q+dVUPtilJEyVg7\nscxq8cUXB4pJ+/R8FX///PPPB2Ds2LHJVtX7ldoVSxsp4VQnxq4tt9wSgDXWWKPfvtgPlRRx9OjR\nyVZVn6ovvPbaa8l25513AnlSNGjtdRbl8kceeSRQvO+Xce+99wLFZYtVXa6odsXyXUrcFZdZtbL9\ncemcnpdiyaEy5MuHHnoo2TrZT/unwirnGcCpRY0xxhhjjDHGVJpmI7kXACNrtdqv6L8m9+aWt6oN\nxEXmw4cPB4ozupdccgnQvhmQXiTOlCmCe8ABByTbuHHjgGKiGZdhaR4t7N9+++2TTUkpYtkAR8aa\nR7OOO+20U799MZLrJF7NM+eccwKw3377JZuSTSgqBjBx4sTONqyLUTKPeO1r9vu0005Ltsceewyo\nbqSxjE62Nd6jlKBn3nnn7fe5qOC64oorgO66/3eyrbE0kHwVyzKJGAnXONBNypiYzEv9qF19Nybt\nO/vss4HycizxvrT33nv3a2fVicnQtN0un84888xpW4kQy4jPpF/72teA7vJpVPFIfdquSOkCCyyQ\nttdee+3Jfi6OR7vvvjsweNd+sy+53/no3+P62DNg4dY1xxhjjDHGGGOMmXKaesnNsmxYuxtijDHG\nGGOMMcZMLc1GcitNPSmJZMpzzz13so0aNQqA2WabLdkks7X0szHy94wzzphsv/jFLwD4/Oc/n2yS\n2ln62Tyf+cxn0vYee+wB5LUcAX74wx8CxWQNpj5xqYJkyptvvnmyqSZuTOjTTfLPwSBKFlVTMCZK\n0Xig+pJgnzZi2mmnTdunnnoqALPOOmuyKSmK9kF3yT8HA0npAXbZZRegKANVIhxJ6qAoqTT9UfJD\ngEUWWaTffsk/o08nTZrU9na1k3aPXVFOW5bIR78fa+IqGVY30cl7wGWXXZa2y2riiksvvTRtxyRO\n3UL0abv9q7rsUFwK0pcbbrghbf/9739va5saMdmX3Fqt9kSWZYt/tP0ckykXlGXZAmV2Y4wxxhhj\njDGm09SL5O4Ztr/S7oZMDWWzF5plUGRM5Wwgj4J9+9vfTrYnn3wSqG4K9iqhWbEYuVlvvfWA4ozk\nU089BdinzaD+uuyyyyabyjIoKgbw+OOPA/bpQFCiOciTdahEEMCtt94K2KcDYfnll0/bO+ywA5An\nmoM8qZ8TzTWP/AgwYsQIoJhUROOBVRyN0Xj685//PNmmm246oBj9VommCRMmdLB13Yl8qiSdUJ4c\n6frrrwfyEjdm8uhZ6vDDD6/7OSVEjf3Zyphy9MwfFXBlqETUXnvtlWz2aTlSF8TEU2Uo4dSOO+6Y\nbIPt08m+5GZZNips3za5zxljjDHGGGOMMVWhnlz56Gb+QJZl9aegjDHGGGOMMcaYDlFPrjx/2J4G\n2Bq4D5gALACsDFzRvqYNnLgQWjIa1cqKdch23XVXAO67775kc8Kpxsi/Q4YMAWDddddN+5R85sor\nr0y2bqo1NtioJu63vvWtZNPi/Vi/1T5tHskTlawL8kQzJ5xwQrJJtmQaM/300wNw4okn9tsXa2R3\ne6KZTqLxVGNo5Le//W3afuSRRzrWpm5HsrrNNtus376nn346bZ9//vnA4EvquoF11lkHgPnnn7/f\nvpisS7J6L/9ojK75+HwqYp/Uc4Hv/4256qqrgHIpfeQ3v/kN4MSozRATSNbjzjvvBODNN99sZ3MG\nRD258le1XavVLgV2zLLsimDbCti2vc0zxhhjjDHGGGOap9kSQhsDO/exjQTOaW1zpo44c6PI2Aor\nrADAmDFj0r5HH30UcFKUZojRcUVxtthiC6CYGlzRxvfee6+Dret+lCRBEYfYh4877jjAiWYGisrb\nbLfddgAsvfTSad8+++wDdGf5hcFECVKUuGullVZK+84558PbQCwv4MhYY+RTlbCKEbLx48cDcOyx\nxyabI2ON0bWvPik1B+RRsP322y/ZrOJozGc/+1kALrzwQqBYjkXXeSwZoiRJZvLMMsssABx00EGT\n/cwbb7yRtq+99tq2t6nbWXDBBQFYf/31J/sZJUYCOOyww9repm5HiVCHDRs22c/Ee73KXlaJ+vH8\nnLHAd/rY9gLGlXzWGGOMMcYYY4wZFJqN5H4duLJWqx0AvADMC/wH2KpdDTPGGGOMMcYYYwZKUy+5\nWZaNrtVqw4FVgaHAi8DdWZZVQu8r+YySTEFeF3f22WcH4Pbbb0/7LFNujGTKkioBrLLKKkDu2yil\n8+L95olyr8UWWwyAjTfeGIAzzzwz7XvllVc627AuJsrqF1lkESCXzkSp1z333ANYTjtQVGv4G9/4\nBgAvv/xy2nfSSScBHlcHyuKLLw7ANttsAxTlyL/85S+BPFGaaY4111wTyOX0cVxQwqmbbrqp8w3r\nYr773e8CMNdcc/XbVyYB99haTuyLSnoUn6/6cuqpp6btWNvZ5ESfKuFUfL7qi5aBQFEObnKiT/Xs\nFG19iUs+xo2rnri32UguH73Q3tHGthhjjDHGGGOMMVNF0y+5VUbJJoYOHZpsn/70pwG4++67AUca\nB4oSIKkUA+RJvG655RagmLzHs7eN6VuCCWDDDTcE4LnnngNg9OjRaZ8TzTTPDDPMkLZ32mmngu3c\nc89N+5xopnmmnXbatP2d73yYkkHjwTHHHJP2xdlxU58YuVHJpTnnnBMolrZRkh+Pq42JPj3iiCOA\nPOFULLmifS7D0hglmYT82leELPZJJfF89tlnO9i67kTJUAE22GADoDxCpnKWZWXaTJHZZpstbY8Y\nMWKyn1Of3XPPPdvepm4nKjbmmGOOhp+PZe6qeL9qNvGUMcYYY4wxxhhTefySa4wxxhhjjDGmZ+ha\nuXKUeUiarJqjAE899RSQy0At/RwY00wzDVCUgGihviTgToYwMCSrV7IpyOvlXX/99YDrDA8UyeqX\nWGKJZFNCHyWiiPWcTWM0ti655JLJtsYaawB5YomLLroo7Yu1B019ynwq+ez555+f9rk2dvMstdRS\naVv+lWwuSuljLVdTn+WWWy5tS04vYoI5Se79fNWYFVdcMW3H+s1QlHnqWcBjQGOUDBXy94AyXn/9\ndaBYy92Us9pqq6VtPV+VoXFg//33b3ubpgZHco0xxhhjjDHG9AxdG8mNaBZRUVvIE0050UzzxFkb\nRXLffvvtZHvggQeAvHxIFReZV42oOFDiifnmmy/ZVCZISWc8Iz4w5NM4oytFx7333gvkiTxMcyjp\njMpaQe5nqTheeumlzjesi1FypC233DLZ5NNXX30VKJa68jjQGCljVIIJ8vuWouN/+MMf0r7333+/\ng63rTpRcascdd0w2+Vl9Mipj7rvvvg62rjvRc1X0ad8IWXxOPfzwwzvTsC5Gz1U77LBDP5uIY6jK\nXXpcnTzy384779zU56+++mqg+s9XjuQaY4wxxhhjjOkZ/JJrjDHGGGOMMaZn6Fq5cpTKKllPlCNZ\nSjtwos/efPNNAP76178mmxLMOOFU80QJjWQdktFCLlOaOHEi4H7bDJLUQS75jDKk22+/HYC//e1v\n/faZciRJBFhkkUUAWHXVVZNNPlRdTI8BjYmSRCVDixJwSWsl/1RiPzN54nj6uc99DoBNN9002bRU\nQc8CTzzxRAdb1/1oKc1GG22UbOrHuv9Hn7rmcGNUa3T99ddPNvVj3e+1bAmcKLEZhgwZAsA666wz\n2c9oySLAyJEj292krkfPUmuuuWay9ZWAx+v9Zz/7WWcaNpU4kmuMMcYYY4wxpmfo2khuGY6CTR3R\nf4rUOGIzdcQoohQHsayFfO6+2zzRVyoNcPnllyebZhuVNM2+nTyaqY3RcfnrL3/5S7LdfPPNAPzp\nT38CHMGph3yqSC3AsGHDgGK0VuPAjTfeCMCkSZM61MLuQz6ddtppk00Rh1g6UBHcf/zjH0Ce0A88\nDvRFPlVSNID11lsPKPZdPQNIdTRq1Ki0zyqZcmI5G/k09l1FxeW/qJhzgrRy4j1q7bXXBor9VL7U\ndf7YY4+lfUqWaopEtZF8GsfTvtd3VBlE/1YZR3KNMcYYY4wxxvQMfsk1xhhjjDHGGNMz9JRc2Zgq\nY2lXa4h+VHIJScHBEvCBIB998MEHyabEMmPHjk02SRb1OfflxsT6gZJ7K3EX5D5V0hnLFBsT++l1\n110HFKWIkonKz1Fe5/GgiPwh6SzkSft+8pOfJNtcc80F5AkT77jjjrQvfteUM3r0aADOOuusZFNy\nv4ceegiASy65JO2ret3RwSImQdIyhEsvvTTZllxyycK+U089Ne2LzwcmJ/p0woQJQDFJ19JLLw3A\n888/D8AJJ5yQ9r311ludaOJU40iuMcYYY4wxxpieodbu2c1arTYRmNDWH+k+FsyybI4p/bJ9Wop9\n2nrs09Zjn7Ye+7T12Ketxz5tPfZp67FPW4992nqa8mnbX3KNMcYYY4wxxphOYbmyMcYYY4wxxpie\nwS+5xhhjjDHGGGN6Br/kGmOMMcYYY4zpGfySa4wxxhhjjDGmZ/BLrjHGGGOMMcaYnsEvucYYY4wx\nxhhjega/5BpjjDHGGGOM6Rn8kmuMMcYYY4wxpmfwS64xxhhjjDHGmJ7BL7nGGGOMMcYYY3oGv+Qa\nY4wxxhhjjOkZPtXuH6jValm7f6MLeTXLsjmm9Mv2aSn2aeuxT1uPfdp67NPWY5+2Hvu09dinrcc+\nbT32aetpyqeO5A4OEwa7AT2Ifdp67NPWY5+2Hvu09dinrcc+bT32aeuxT1uPfdp6mvKpX3KNMcYY\nY4wxxvQMfsk1xhhjjDHGGNMz+CXXGGOMMcYYY0zP4JdcY4wxxhhjjDE9g19yjTHGGGOMMcb0DH7J\nNcYYY4wxxhjTM/gl1xhjjDHGGGNMz+CXXGOMMcYYY4wxPYNfco0xxhhjjDHG9Ax+yTXGGGOMMcYY\n0zN8arAbYIwxxhhjeptPfKJ/XCXLstJt059ardZv+1Of+lQ/2wcffJBs//vf/zrUuu5CfTH69JOf\n/CQA0047bb/Pv/vuu2n7P//5T5tbVw2ib5pB/tO/AJ/5zGcAmGWWWZJN1/nLL7+cbLHPthJHco0x\nxhhjjDHG9Ax+yTXGGGOMMcYY0zNYrmyM6TgDlcFMDsvbcqbEp2XfkU/t2+YpkxFGaaa2o3RQ2/X8\nXPZ3o01SxU9/+tP9vvvPf/4zbUte103nNB6n/BePUzK4+Lm+fTeeA/nqs5/9bLJNN910hX0As88+\nOwBzzz13skl+d9999yXb3//+d6B7pYs6pmmmmSbZZp55ZgBmnHHGZJPP5bdZZ5017RsyZAgAQ4cO\nTbb5558fKPa1OeaYA4B555032SQJvfzyy5PtN7/5DQDvvPPOlB3UIKP+Nv300yebjvlzn/tcss00\n00yFf+ebb75+n19wwQWTbc455wSK44d+Q+cM8nN1//33J9vOO+8MFKWh3YSu7+jTxRZbDIAvfOEL\nySYfzTDDDAAssMACaZ/6Z7ym9bno07IxQtfJiy++mGxrrbUWAOPHj5+iYxps5NN47S+zzDIAbLfd\ndskmv+lz88wzT9o311xzAcXxQGNyRNdElDDr999+++1kW2211QB4/PHHB3w89XAk1xhjjDHGGGNM\nz+BIbo8xNRGybprl7wRlEZm++/puC80OxllC+zefKY2RExFn+soSGPz3v/8FigkKtF0WIfu4IF+W\nzZRGP2uWX9EryGez43ffe+89AN5///1kU7RK5yD2Zdn0b6QsmhnR+YvntOzvdJqyqKm2o6/k05hU\nY9iwYQAsscQSyaYoRIxE9u3jcUZc0Zl4rv71r38BRf+URYOfffZZAK699tpke/LJJwt/o+93Bosy\nP8tH0R/LLrssAJtttlmyKdKlKCHkflaUMEYq5Od4TZSN0+qn8XP//ve/AbjxxhuT7Yc//CGQR3T7\n/p3BpiyhTuyna665JgDf+MY3kk3Rxtlmmy3ZFJ0pG5PVh6Kt777YlrLvLrLIIsmmyNjIkSOTrUo+\njaifaAwFWGONNQA47LDDkk3HF6Pj8kM9JU3ZeBmp93wSI5w6v8cee2yyVdWnOqYYSdW1/8tf/jLZ\nFMmN42O9Z18db7M+LftbUYWgaz9eO1UYT+sRx7OFFloIgPPPPz/Z5Ofo+75+KEsS1+hZuN55kZIB\n4OijjwZg2223Lf29KcWRXGOMMcYYY4wxPUPdl9xarbZcpxpijDHGGGOMMcZMLY3kyg/UarVxwPnA\n+VmWTWjljzeqmVaPqksDOo0kAVEOVyYnqJdUpkxa+3H2s+QdURIjmVHZAvuYgETyyygP1P6Ps58l\nI4xyOEnolMQk2qK8TYl0XnrppWR74403Cvsg97MkhmXnJZ6DMimd9utvQDVktGVIXhRlc5K7xkQR\niy66KABLLrlksimxRBwjJFOOx65rQOcoJjuRT//xj38km7bLzkuUd956660APP/888mma2EwJXW6\nN8V7lK75KLGSNHmjjTZKNm3HJCcal+O4of6mfWXJo+I5UJ3GWK9Rctz4d+XLp59+OtnGjh3b7+8N\n1phTJm+L7VcCmS9/+cvJJllgTCaj75TVCS0bV3XsUYYft4UST8VxX30xJg+Ksr4qEv0sGfd6662X\nbEcddRRQ9Kn6YOz38qHGv9iHJk2aBMDrr7+ebLrm43gkv5X18bL7a6sSE7YTHctyy+WxoJ///OdA\nPi5AfkxlPpUvdR8DePTRR4H8moXcp7H/bbDBBkDRf31/M9INPpWPYiKu008/HSjet3TNlz3bylfj\nxo1L+/7whz8A8MgjjySbns1WWGGFZNtvv/2AorRcxN/Svazs96uG2hjvW2eccQYAK664YrLV86nG\nyTvuuCPtO/nkkwF45plnkk1+kfQZ4Oyzz+73+2WUjcWtoJFc+X3gGGBdYGytVru5VqvtWqvVpm/w\nPWOMMcYYY4wxpuM0iuT+N8uy84DzarXagsCuwKHAqbVa7Qrg3CzLbh3oj2q2Js6E1lsQHvdppiBG\nZ+pFJau6wL7VaFYxRsPk3zh7WhaNUqKZGHUpSzCjiE03lqOYEjSbF2e6FRmLJRM08xWjtopWxaij\n/Bz7rmZyFZ2J+8qSIKg/x/PYTedBUZrhw4cn24gRI/rZ9LkYpVG0IEYC33zzTaDYx/ueoxh1VBmA\np556Ktl0DuL50yzwPffck2xvvfUWUD1/K6IVx1NFo2JEQZFcJe2AvI/H/qTyHTEZlJL7yLexnIP6\ndTwHmrWNM+I6pzFqq3P5wgsvND7QQSCe67JrT7YYidT+GHHV2BpLJvRV2qh/Qe6XGB3XeYkJrTbd\ndFOg2MfLEoFVacwua0Mc9+S31157Ldl0nceIl/pntKlMyoMPPggUk28p4hD7tZ4tdG0AnHnmmUAx\ncqTPxd965ZVXJns8VSC2S/0vXmc6lpjgS31Gxwa5L3/9618DMHr06LQvRnX7EiO5F110EZD3V8j7\nvcYPgAceeADojuc29aM4nukeFVUcOs5XX3012f70pz8BeTKo2NfrEZ+FDz30UACOPPLIfr8Vr6er\nr74aqK4SKaLzHn2le3ZMUKZjidejrtvjjjsOqN83I3/+85/Ttvq2zk9Z2yDvz93QTzUOxLJcGgtX\nWWWVfp+P963jjz8egJ/+9KdA88cbo+jajuNG37ZBXj6s1eNp04mnsiybkGXZj7IsGwF8Cfg38IeW\ntsYYY4wxxhhjjJkKGr3klor4syy7M8uybwDzlO03xhhjjDHGGGMGg0Zy5W/W25ll2b/q7Z/sj34k\nQYxJUbQd67hpOy5Y1uLkKMeYf/75AVh44YWTTdKva665BoB777037YuSsV5ByQeWWmqpZJPMNkpr\no8xQyFdx4bckTFF2I0mNFp/HxCaS5UbpUTdIOeohH6y88srJJqmg6oxBnkxJSTgglxtG2Y2kRDHp\nkvqxalvGfi15Z5Qo3X///UB5f+4Gf8tX8dqXtDZe55J/xmtVfUzSxYj+BuT1M/VvlHtLkhzHCtW/\ni7K9xx57DCjKmtWWqskTJd2Kyw0kN4wSNV2/kh/G/TFxjL4bZZ1KcKR/yxKqRRZffHEAvva1ryWb\nJM9lideqVk+67FpSu6JE7vHHHwdg4sSJyaYan/FvSEYbvysf9q1BPDl031xppZWSbeONNwbKE7RF\nyZjOZRV8G1F7Yh+Sj66//vpkGzNmDFCUwGo8kG8hv4cNVJoZrwlJ+aIsXH8vyhirOh6IsqRb8Tj3\n3ntvIH9+grxObRz35NOBHmccu2+55RagmExMf++GG25INsmpq+rTiK7vCRPyfKzf+ta3gGIyIy05\nGDVqVLJFSeiU/CbAzTffDMARRxyRbLrX6XoB+Nvf/jZFvzUY6LzHe/xee+0FwCabbJJses684oor\nki3KcaeUhx9+uNAOyH0q2TQUlzF1C1G+feCBBwLFRFxaWnTqqacmW1xCM6U899xzQLlP4zPznXfe\nOdW/VUbdSG6WZRe35VeNMcYYY4wxxpg20CiS254f/WhGOiYlUcppRa8gT0gTIzyKLMa06YrEKEoE\nefRrn332AYozQ7vtthuQR3mhO2YO66EIS4yEy3/RpzGKKOTLmDhGi/xjhEzRgu9973tAMZqjWcVj\njjkm2RTlaDYBQNVQIp8Y/S6LhGtWNiaZ0gxVjIYpLb76a0QJgmKEUZHIyG233QYUZ8k1Gxej6FVF\nUYGY2ETXXkyKohlERbghj4rH4ywr6aRxQONGjIRr7IlJLHQ+ylQLZeVJqoYid7GvyRYjjJrdj1Gz\nskR+ihbUGxPLZmVjNFHRuJ122inZNL7EaKbOeaNEgp2mrA2K5sVoitodIzOKhrU6Ol1Wgkx9PJ77\n22+/HSiWIOmGpDNCbY1RhBhxEK3sJ/Halgoh+lnR3Ysvzuf9yxQMVaVvKRDIo7oxutsun2611Vb9\n/r7Gge9///vJFvtxtxD7gfqpygBB+8azH/zgB/1sGvel8IDufP6KPlOkXGVv+u5vJTEqLnR+V199\n9WTrxn4a0fPpKaec0vbfUvKqsgSqyyyzTLK1azxtOvGUMcYYY4wxxhhTdfySa4wxxhhjjDGmZxgU\nubJkrlHeqSQTMXGSpFhxQblsUXarMHiUJWq7rCZvWeKabkfSnyeffDLZVL+tTAIb5RaSM8dkQJJ/\nRp8qcYzkzVEiKmLym7Lz0g3JkYQSRkQ5lyTYMUlRWQ1CSe2ihFNScvkR8v6s2qWf+9zn0j7J9qLE\nTDLaKPXXeYi+r6qfyxIiSRIc2yw5dpSBqs+WHVuZlLhMRqt6orGvayyJSW00RnSDLEltLKuLGimr\nHd4K2VdZ8ij5dOmll042nYcnnngi2ZTMo6r9tYzos05Kq9Wf99xzz2TT8ol4L1VNwzhudCOd9O3a\na6+dtpWIKS6LUPKbXnp2aLd/Y3355ZdfHiiOp1/96leB5mvEdhPt8m1cLvWlL30JKI6dWkoW62v3\nCu3yaVymp+U18bcOP/xwIE+gZBoTn7l23313oOhT1YeONabbRVMvubVabWZgH2A5YIa4L8uyjdrQ\nLmOMMcYYY4wxZsA0G8m9HPgkcCUw1dPDmnmKCUi0XVaGRQl7IC+dEpPEKF17LK0wZMgQIF90f9ll\nl6V9f/3rX4FqJDhpFZohjbNN8rOSZkA+Exgj22URV0UYYyr89ddfH8iT9sQZ2JNOOgnIk67ENkX0\nW93ge0URY+RJ0bIYOVQfi8dbluhF34m+l5/ly1jOQX8jnlMlnoq2sjZV1c+KNsfSSiL6TD5tRdQx\n/g0lpVJfhrw/x0RqSmcf+3jVfNmXMl91Muqo5HcA3/72t4FikjX58g9/+EOySaVTdd9WAUXGlMQH\n8msmJkRS+Rf7tDFSwZx33nnJJmWJkilC+8pb9CK69+heBfk9LyZkiok/TXPo2RVyn8ao7W9/+9uO\nt6nbieWANB5ExcYJJ5zQ8TZ1O7GfKlIeFblHHnlkx9rS7EvuqsDsWZZ1X5o2Y4wxxhhjjDEfG5p9\nyR0FLAb0z+M/BWiGOUa+ZItlA1TyIhbb1lrEWNpDBdrjupqNNvpQRa11kj//+c/Tvm5K/d8smtGP\n64jk0xih0gxV2frlskhQLEOhiPBaa60FFNeAPvbYY/1+q9sjCeonsVxPs2sdRVmpldjv9R2dtxjl\nVVT8uuuuS7Y77rgDgNdffz3Z5PNu8LeON5Y1ULtjJLeVazTjWLHOOusA+fgAeQTyqquuSjaNKfF6\n6sZ1o53oExo/1lhjjWRTtDH67MYbbwTyEjfQm2NxK4lrm84991yguHZcY3K8v3XDOvKqoAhuLLMn\nBc93v/vdZHM/bR6VDImqJPXJ7bffPtm6qazVYKN1jUsssUSyaWz/f//v/yWbfdo8W265JZCXL418\n+ctfTtv2afNonXhUgKqfbrjhhsnWyWepZl9ydweuqdVq9wAvxx1Zlh3d6kYZY4wxxhhjjDFTQrMv\nuccC8wPjgZmCvfqhI2OMMcYYY4wxHxuafcndARiRZdmLrfzxGLKWlDNKAySHjfIrSTOjBFF/R+VJ\nAP72t78BufxTSax6nTJpRSPJYln5lTL5+NNPPw3k5W5ikilJvLpBMtssOpYyae2UJPQp+5x8rzJP\nKnEDMGrUKABuueWWZFPynrI2dQP1JN2tPg7JaGPiuj322AOA2WefPdkktT/ttNOSTWNJN0iU5bd4\nHXeyT6g01tFH56IeLSt59tlnk+1HP/oRUExA0U19dzBYd9110/Yqq6wCFPukygWVJXIz5UQZrWT1\n8drR8pC4TMrUJ0ro9913X6C4JOqRRz5c6aakaKYxsbTNmWeeCRR9qmfa+++/v7MN62Lie8Pll18O\nFK99PV/dddddnW1YFxOX1Pz5z38Gij7VO0RMRtVJPtH4IwA8A3ihjzHGGGOMMTw2ZxQAACAASURB\nVMaYStNsJPcCYGStVvsV/dfk3lz+lamnLCGNIrnRplmDmBxCyWQUDY6Rr16mLNHR1ERL4syhoo1K\nfhAjjL24OF9+a3U0L/pUJVa23Xbbfp9Tgp4YpZGqoVsjYPV82qpjUr+fbrrpANhtt93SPiWZmDhx\nYrIpgquILnRnf+5kn4izt/vssw9QTOChhF0//vGPk01KkG6Ijg82KmXxq1/9KtmkSorRsAsvvBDo\n3vFgMLj00kvTtvwckyh+4xvfAOzTgXDWWWelbfk0Xuebbrppx9vU7UilAblPY59UAlDTPD/5yU/S\ndozqitVXX72TzekJYtLD+FwgBtunzb7kfuejf4/rY8+AhVvXHGOMMcYYY4wxZspp6iU3y7Jh7W6I\nMcYYY4wxxhgztTQbye0YccFyWUIkhcMl34A86cHSSy+dbBtssAGQJ++xRK5ImZ+jTUkPhg8fnmzf\n+ta3AJhnnnkAePnlXLney9KuqTm2Mj/PNFOeoPzrX/86AKuuuioAV199ddonWWKsPdwr/bid/UVy\ncNXNlvwQ8vHjd7/7XbKNHDkSKPrZ1Ccm81I90Tgmy6cXX3xxsnWjBHyw+NrXvgbAiBEjkk3LcTQO\nA7z//vudbVgXM3ToUCBP4BWJcts33nijY23qdnTNb7PNNv32xYRIMSmoqY/uX3vttVe/fXGZjROj\nNY+evWLta6GlNeDEaFNCWT+NyYLjMrDBYLIvubVa7Yksyxb/aPs5JlMuKMuyBdrUNmOMMcYYY4wx\nZkDUi+TuGba/0u6GlKFoT0zUowhjjIatvPLKAGy99dbJNv300wNw/vnnF/7Wx4my0iJlaH9MWb/g\nggsCsN9++yXbOuusA8Bf/vIXAF5//fWWtbXXKPO5Zr1jwghFbPT5a665Ju3TrK0jYANjyJAhABx4\n4IEAzDnnnGnfHXfcAcDpp5+ebOrHH8cxYqBoLD7uuDw9g0oIPffcc8n2gx/8AHCkcSAoURrAEUcc\nARTvfSrPoIR0ZmBcccUVQDE5yttvvw3AAQccMCht6nZ+9rOfAcUkPrpfbbnlloPSpm5Hz7HxeUzs\nsMMOnW5OT7DCCisA5T794Q9/2Onm9AQLLbQQUJ7A64ILLuhwaybPZF9ysywbFbZv60xzjDHGGGOM\nMcaYKaeeXPnoZv5AlmWHt645xhhjjDHGGGPMlFNPrjx/2J4G2Bq4D5gALACsDFzRvqaVo+QbqhkI\nuRRh8cUXT7YxY8YAMG7cuA62rruJUo7VVlsNgE022aTf/nPOOQdwop5miLLlmWeeGYC999472SSt\nve+++wC49tpr075Y99nUJ/bdHXfcEciXMUiSCHD88ccDxaQdlik3z1JLLQUU614qccdRRx2VbP/4\nxz8627AeIMrm5phjDgDefPPNZFMCtV5JPtcJFllkkbS9/PLLA8Xr/Re/+AVQrJNr6hOfvfbYY49+\n+59//nkAXnzxxY61qduJzwmxNrb497//DXipwpRSJp/VOHDGGWd0ujk9gd4Dyth///072JL61JMr\nf1XbtVrtUmDHLMuuCLatgG3b2zxjjDHGGGOMMaZ5mi0htDGwcx/bSGDyr/IDRLMqcZa1LHmPbEos\nFb/z8MMPJ5tmw955551WNbFradanKsUEsMwyywDFGW4lRbrlllva0s5eJCY5USRhscUWSzaVq/jN\nb34DwKRJkzrYuu5HfTeWWtlzzw9z5inicNlll6V9t956K+BkXgNl2mmnBeDUU08FYMYZZ0z77rzz\nTqDoZ0fHm0elbdRvIfdfjEDEkm2mPhoXTjzxxGRTgpR333032eJ+0xwbb7xx2tYYG693JU0zzSOV\nF8Bss83Wb/+TTz4J+L41EOKzlxKpRjQOWMUxZSy33HL9bOqfVXqO/UTjjwAwFvhOH9tegLXAxhhj\njDHGGGMqQ7OR3K8DV9ZqtQOAF4B5gf8AW7WrYcYYY4wxxhhjzEBp6iU3y7LRtVptOLAqMBR4Ebg7\ny7IPWt2gMmltrBWoWoKzzz57sr366qsAjB49OtkeeeSRVjetZ5GMa9iwYcmm7Vj78uc//znghFMD\nIcqQttlmm362sWPHAnDVVVcBlnkOFMlod99992QbPnw4kCc/ijVdLU1qnri0YautPpzPlOQ+1r89\n5phjgDwBlWkO3ddUU1j1hiFPOHXsscd2vmE9gHz5hS98od8+LVmAYlI6Ux+NB4cddliyqQ/HZ4Lf\n//73nW1YD7DrrrumbT2PxWeBgw46qONt6nbi8+w000zTb/+VV17Zyeb0BHreAphpppn67R8/fnwH\nW9MczUZy+eiF9o42tsUYY4wxxhhjjJkqmn7J7RRx9krbmtmCPOFULKOg8iuPPfZYsn3wQcuDzD1F\njI4ricySSy6ZbFqUP2rUqGRT8gPTGPXZz3/+88mm7RhNHDlyJAATJ07sYOu6m9h3lcRr8803TzaN\nGzfddBMAzzzzTAdb1zuojA3AfvvtB+RKmrvuuivtu+222zrbsB5hzjnnBGD77bcHive0K674sJCB\nk01NGTvv/GGezJigUkqDGIk0zaPITUzyp7E2PhvExF6mPoqO77PPPv32xej4zTff3LE29QqHHnpo\n2paf4xgbS7aZ5oglRcuS2B588MGdbE5TNJt4yhhjjDHGGGOMqTx+yTXGGGOMMcYY0zNUTq5cRqx3\nJWltlCM///zz/5+99w60oyrX/z8jTSBUCQRSgNAhAgHpKKEqcJWiXBFQFAX12hBv1Z+CBb2iCApe\nEesVJChfyqVK7z1UA6GEkBAgtBCqdPbvD3jWvHPOYp99TvbZLc/nn0ze2WfP2u+sWTOz3me9L+Ak\nEoNhkUUWSdtjxowBYOzYscmmpD2xJq4TTtUnymiV+GTSpEnJpoRTMZmX5Mqvv/56C1rYG0QJ4oEH\nHgiUtUahrD08efJkwL4dLOrHe+yxR7IpiYckn8cff3za52RejRMlXnvuuSdQjgvqtwC/+tWvACei\nGyxKMHPQQQcBVX8/9NBDANx7772tb1gPcMghhwDV5DOqi3nqqae2pU3djpYs6BkMyms+LlV49dVX\nW9uwLkbvCx/+8IeTTT6NfvQSscGTS4QYJeA33nhjK5vTEI7kGmOMMcYYY4zpGTo6kquIgpKdQJkM\nZcSIEcmmGRnPeg+MZrliuYpNN90UgNVWWy3Z5s6dC8Bzzz3XusZ1KYoWxDT1G220EQCbbbZZsil6\nPmvWrGRzYpnGkf823HDDZNt6662BanK6Bx54ACgT0XlcGBzLL788AB/60IeSTX1cJQKcbGpoyLcA\n++23H1D23cceeyzti2oP0zi77747AGuuuSZQjeRqXHBSysEh9ZxKXUXFkiJjSv5pGkP9UoqYqKwT\n6q9mcPRVyEDp71jmzuNA44wePRqANdZYo9++GMl9+umnW9amRnEk1xhjjDHGGGNMz+CXXGOMMcYY\nY4wxPUNHy5UlMYhSjsUWWwyA8ePHJ5uktZZ3VMnVsZIvlfAAYNy4cQCsuuqqDX3HgkyUagn5dNSo\nUcm2xRZbAGXCnvi5J554ItmctCffx2SLSedWWmklAHbaaadkk89jQonp06cDltoPhkUXXTRtK1na\nxIkTk039XlL7l156qXWN6wHUj/faa69kU41nyenj/cv+bZyYiO5f/uVfgFIC/sILL6R9f/rTn4Ay\nWZJpjB133BEoE07Fsfbaa68F4Jprrml9w7oYLcHTM1eU0Srp57777ptsXnIzMHpmWH/99YFqIj+N\npzvvvHOyeRxoHC0TjUvtNO7GfhrH207BkVxjjDHGGGOMMT1Dx0VyYwIZzczEWW1tT5s2LdlUGsBU\n0exfTIik7VgGQEm8Ykr1+++/H6iWDdL5WJBnFbXIPiZDW3rppYHqovwJEyYApfIAyr4b/exIedmf\nYtRW40BMkKbI4iabbJJs6scvvvhissm/9m2e6Bf5PCZEUmKvqDJQgrQpU6YATtoxWNSfVe4O4Lrr\nrgNKP//gBz9I+6zwaJzoq2984xtAmUzx6quvTvuk8FiQ719D4dxzzwXKe1r0txLNxOQzZmB0v9pq\nq62AqkLMY+vQ0HX9/e9/H6iOp77m54/bb78dKBP6dROO5BpjjDHGGGOM6Rn8kmuMMcYYY4wxpmfo\nOLny66+/nrYl9YxyZdVpjPKDefPm9bOZkigvksw2Jj+66aabALj33nuTberUqUBVBmr5Z0lMFKF6\nbM8//3yyzZgxA6jKwiX5vOOOO5JNyQ8sBa8mgpC8M9qUYO6qq65KNtUWjfWGzz//fKDs97HfLsj+\nFdEHkhnGJF2///3vAbjsssuSTfuVHCn2fzMwWvZxySWXJNvll18OlH089nX308aJfpOsTv+a+UfP\nZFFqb5pD7to3zcFjqAFHco0xxhhjjDHG9BDFcM92FEXxJDBrwA8uWKxaq9VGDvWP7dMs9mnzsU+b\nj33afOzT5mOfNh/7tPnYp83HPm0+9mnzacinw/6Sa4wxxhhjjDHGtArLlY0xxhhjjDHG9Ax+yTXG\nGGOMMcYY0zP4JdcYY4wxxhhjTM/gl1xjjDHGGGOMMT2DX3KNMcYYY4wxxvQMfsk1xhhjjDHGGNMz\n+CXXGGOMMcYYY0zP4JdcY4wxxhhjjDE9g19yjTHGGGOMMcb0DH7JNcYYY4wxxhjTM/gl1xhjjDHG\nGGNMz+CXXGOMMcYYY4wxPcPCw32Aoihqw32MLuSpWq02cqh/bJ9msU+bj33afOzT5mOfNh/7tPnY\np83HPm0+9mnzsU+bT0M+dSS3PcxqdwN6EPu0+dinzcc+bT72afOxT5uPfdp87NPmY582H/u0+TTk\nU7/kGmOMMcYYY4zpGfySa4wxxhhjjDGmZ/BLrjHGGGOMMcaYnsEvucYYY4wxxhhjega/5BpjjDHG\nGGOM6Rn8kmuMMcYYY4wxpmfwS64xxhhjjDHGmJ7BL7nGGGOMMcYYY3oGv+QaY4wxxhhjjOkZ/JJr\njDHGGGOMMaZnWLjdDajHQgstBMDCC5fNXHzxxQFYZJFFku2ll14C4OWXX062N998E4BarQbAu95V\nvs+/+93vBmDkyJHJ9o9//AOAJ598Mtn0t72E/CDfQunf6KPXXnsNgDfeeCPZ5FNRFEW/71hsscWS\n7fXXXweq56WXfRr9EX0p+vbJvtt90ffF79Ln43npFaL/4nZf20D+a8SnOXqxb9b7ve9EL/rBGGOM\nMQsWjuQaY4wxxhhjjOkZ/JJrjDHGGGOMMaZn6Bi5smR1kiMDrLjiigBsscUWybbddtsBsNJKKyXb\nGWecAcDUqVOTbe7cuUApy11rrbXSvp/85CcATJgwIdkkJT3xxBOT7atf/SrQvfK9nE+XW245AFZf\nffVkmzhxIgBjx45Ntuuvvx6ARx55JNlefPFFoJSKr7baamnfoYceCsCmm26abJIrH3XUUcn24x//\nGOgvfe4WJB2OPl1iiSUAeM973pNs8s24ceOS7cEHHwRKP0J5jiShV58H2HvvvYGqT1944QUAfvrT\nnybbKaecApT+7jZ0jcoHUPax6OdlllkGqF77zz//PFCVdC+55JJAeV6WWmqptO+9730vAGuuuWay\nPfPMMwCcddZZyXbFFVcA8MorrwzpN7WDKE2WT+OyDu2Pyz+0vEA+g3K8i0saFl100cq/0d8aU6Kf\ndX3PnDkz2WbMmAGUS0O6gdwShHpSeij9Fv0n4rhXbwysd6z4d7rm4/KFTr9fDUVCr7/J/W1u+cL8\n+GBBXtLQi7/PGLPg4kiuMcYYY4wxxpieoWMiuYoujBo1Ktn22msvAA488MBkU+Th7rvvTrY77rgD\ngAceeCDZXn31VaCcmVQEDOCpp54C8smXttlmm2TTbHo3JfmJs7Ly1ZgxY5Jthx12AGC//fZLNkVg\nHn744WS7/PLLAXj00UeTTT5U9CD6T1G4GBFSWz7xiU8k27HHHguUycI6mVziJ0UHV1111WTbbLPN\ngLK/QpnU7Nlnn0228847D4Dp06cnmyKR8u0mm2yS9klpEI+ltnzzm99MtosuugiAxx57bDA/ry2o\n/TGauPTSSwPVqPfaa68NwE477ZRs6scx2nDXXXcB1f6k6K8iXuutt17aJ5/Ga0Jt2W233ZLt4x//\nOAA33nhjsnVqlCOX9E0+jVFvja2bb755sknhoig5lH1xxIgRybbCCisA5TUfVQu6JuK1rwiy1AsA\n3/rWtwA499xzk61T1Qdqf/SpxrjoF/3mNdZYI9nk06jKUL/XeYn75ft4DuQXqQzi9n333Zdsl112\nGQC33HJLsmlM6bT+qn4a7xvyS1QcSC2w7LLLJltfX0GZHDEqA/req+P3ipgIUeNGLhIeVRzz5s0D\nqiqcTui7uQSHuX31/JxLOKnnJyh902jixNx9M6cs0Xb0o3weba1WfeUSHNb7XPydOcWB2h9/RzOv\nzYFUJNqfU4A0mgSznQyk3mgVuf4c29b3Ool0qm87iVxS3Ny1k+vPdb93oA8URbFCURRbFEWx+Nv/\n37AoisOKothpoL81xhhjjDHGGGNaSd2X3KIo9gBmAecA9xZFsSvwN+D9wOSiKA4d/iYaY4wxxhhj\njDGNMZBc+Uhgn1qtdn5RFHsCpwGTarXalKIoJr79/2Ob0RCF+qO8U4lK/va3vyXbbbfdBsCZZ56Z\nbFF+9E4oERXABz/4QQAuuOCCZNtyyy0BOO2005KtW5MjCfk0SjklZZPEG0o/n3TSSckmSXc9okTu\n/e9/P1BN3iPb7bffnmydIPEaLFFqkusTki9GaaF8+sc//jHZJK2NPugrY4nyWCVBmzx5crLJp/E6\n6aZ+mpN4ScYYpYjrrLMOUJV8qh+fffbZyXbhhRcCVX9Icie/RPnL6NGjATjmmGOSTeOB5HtQJlPq\nBvT7orRWvpTsG8qlGFESr8/FpR4aY+fMmZNskmn2TegHpVT3Yx/7WLJpPF1++eWTTW2pJ7HsFNQX\ncvLi6FMlRdTvhbysXn03/nYdQ+N0/LzGaUnBoeyTUWovGXlMEKi/7TT0e6OsXf1vlVVWSbYNNtgA\nqCaclAQ8JqLLSQC1X0tw4rE0Hjz55JPJpiU6ufFUYzjAqaeeCsA999yTbLnz1mp0Heb6VUwEp+Uz\nMeGktmMfk0w5+kMSYv3e3LKImDRQ97fcvV79NR5r2rRpyXbrrbcC1aU3kqO3S7Yct+O4J7l19J8k\n9tEfej6V5B3K39R3WV0kSrvVj+M51f64xE/XThwjdP7i89rNN9/cr03tWpaX8/NA0uBmJNzLSfh1\nrHhOx48fD8CkSZOSTQks43I+PafF5SQ6950mVx5uCXhOwh+XqGk82nnnnZNtjz32AKpLJX77298C\nZX8FeO655wbXlgH2j63Vaue/vX02sFitVpsCUKvVbgNGDupoxhhjjDHGGGPMMDJQJPeRoih2qdVq\nFwG7Ay8XRTGxVqvdVhTFRsATzWqIZulixFVRwRgdbMZsg461zz77JNvXv/51AI4//vimHqvVxDYr\nohVn+TUjrX+hOTN4OlZMMqUyN7/4xS/6fa4byJWjUETm3nvvTTbN+P/ud79LNs3gzc/ss2bTo0//\n93//FyhL3ED1mul05I/YDxQB1yw+lBGTOHOuWeeYaKaRazRGFHSuPvWpTyXbOeecU2kblEqHbhgD\ncooN9YnoU806X3zxxcmmiEmM5GqmtNFEKUpSd/rppyfbL3/5S6Bapk3H6gY1Ry75kBJyxSiTksjF\nmf/7778fqEYCpWaJNvlZUfKoSNLxY+Rrzz33BKr3LSkeYuSoU8mVRcolLFT0P6oANA7EcUMR6xi5\nVqRLidFipOqJJ956XIllrXTNxBKDSlQXk4ldeeWVQDXq2AnkEvlJWZSLjr/vfe9LNpVRi303l1xK\n0Umdj6i40XHjNa3xPEbMFV2OChlFbOJ49NBDD/X726GUnJofcsmj9DujrxSNUlQPSv/GxHwaQ6LS\nS9e87mWxr6tk4Prrr59s6v8xcqj+HEs/ajveN+XLeAypdGLEvtWR3FzUVv0kV44x9vGnn34aqF7f\nGgd0Tcfzp2vhyCOPTDYpmuRHgDvvvBOoJmzceuut+7VJxHMqpc/hhx+ebDGZa7vIRaw13sbrXPef\neC3XS6al87H//vsnm8quxn511VVXAdVxQ+UxpayL7Yxj/MYbbwzA1772tWS75JJLKm0biIFeco8A\nzimKYh4wDfgacHFRFFfy1rrcIxo6ijHGGGOMMcYY0wLqvuTWarW/FkVxDTAGmFKr1d4siuIhYCPg\n57Va7apWNNIYY4wxxhhjjGmEAevk1mq1R4FHw/8vAS4ZzkaFYw3r90d5n0LqnZq0Y35oZcKG6FMl\nsInS3m4nV7druCXYsTaj6o7GxGvdVMdZ13Rss7ajXDPKXYaDWDdb0if1V+iOmsNC/os+lUROvw3K\nvpOrjdiMsTaeM8nhopRuypQpQHckSpOUMkq3JC2MfUOy9r/85S/JpvEg9udcXcpGiHI3Sek+97nP\nJZvG23ieO1ViLx/E5QbyZUwmouU111xzTbJJehjlgUqOGL9PfUvXQm5sjrJHJehRDWco5Ynx3Ou5\nYH4S3QwH9eqyxv4n/86ePTvZNEZEuaj6U7xGJdNceeWVgarkWLLYKBudNWsWUE2SpCRiUTIr/0ZJ\nfi5pVav9LH9Ev0hWHyWrm222GVCtry6Je04CHn3aN1lglHIqmVeUl6qPx+dTSXtjMjH9bS5ZZvR9\nTpY+nNSTgMfkfpIQH3TQQcm27rrrAlV/qO/GZyP1Xf2mVVddNe1T8qj4Hfqc+jWUPornWbL02B/0\nt7GPS64cj9Hqe53aGJNojhs3DoAvfOELyaZEiXEs1DNRvJa1rX0TJ05M+7baaisgX4s8jrs6D3FJ\njZKl5ZJQxu/T+YjPEYN93u78NJfGGGOMMcYYY0yDDBjJ7WU0QwPVxedm6MRZXs0ed1OyqU4kzgpr\n5q0TEhp0M3GmU7O2MRrXDdFGkZuNb0eUKc7SK3lPLEMUk1t1Oookxd9UbxwbLn/HfpiLpE2dOhWA\nxx9/fFiO30zkvxilU0QmRnIV3c1FowYq5zZYckl5FF2LEXslDOs01UyujJEiWjF5k2wxcZb+JkbC\ndT5iX9fnchFOEc+F7lFKoATV5Ex9/yaOEXoOi2VE2hUNi79TpaliQiIl7oqRSJ2PWIJRvyWOJbqn\n694Tz1+ujJOUhrGUykYbbQRUI7lqc05ppgg7lONFq/pzrtSVxjEpJwC+8pWvAGXCISh9nxtj4/f1\nPW+5MjYR/XYlCgT4+9//DsAOO+xQ929li/5TQrtWjcW5RH5KOrf99tsn2xFHHAGU92SoRnD7kivp\nNFiiUk59Nya9i9HzemjcuvTSS4fUDnAk1xhjjDHGGGNMD+GXXGOMMcYYY4wxPUNDcuWiKJYBvgpM\nBEbEfbVabZdhaFdL2HHHHdP2aaed1saW9A4777xz2r7hhhva2JLeYdKkSWlbSbw6TTbXbUT5nCR6\nqnlqhkZMwqFEHz/60Y+SLdac7RbanVwoSu4+8pGPAKWcFuA3v/kN0B1LQjRm5eRw7ZLcSza3+eab\n99t34YUXpm3JT9vdH/qSq2OpPhPbKvlg/FxO6lwvIVEuSV0uoVCu1qnqlMb6rUqWlqv53s5a2rl+\nmkuYKPlvlNrrOozXo+THcXzUUjnJ5OPYqFq48fnpvvvuq3weytrHuWRK8fiS0UY/SwrdrqReUNZf\njrXU1Wdi+3UeYt/J2frKd3P9Wv0L4Bvf+AZQ9bMk4O9///v7HSui74sJ/84991ygmnx1OFG7con0\nPvrRjyab5Ow5/+W+b7BEP8sHhx56aL/v/cUvfpFs8Zz3JV5j1113HVBd0jBYGl2TexqwEHAm0Joz\naIwxxhhjjDHGDJJGX3K3BFao1WqvDvjJLkAzGrvuumuynXTSSe1qTk+g2ZpPf/rTyXbqqae2qTW9\nxeGHH56299lnnza2pHc47rjj0rZmC9sZPegFlDQEypIY559/frua0xNoZh5KlUxMOteNaplcCatW\nEiNf//Zv/wZUy9io1NYpp5ySbJ0aKc/5L1eGJxehbUb5MP1tjAIpodCXvvSlZMtFLC+77DIAbrrp\npmTrJD/HiJLKMam8FZRJsnKfixFz+SYqMPQMqpJAMdGWvi+eP0UpVbYFyuRISswUjxsjX1deeSVQ\nJk/re7xWEn0ldYGS50HZT2JiIv1NTBapPhY/pxI1ilLHRFt/+9vfAPh//+//JZsi8DGRn1Rzo0eP\n7tf2eE6VrO32229PNpXIa5XKLpf8SlFkRe+hmoBOqD/F6K62cxHfnJJBfWz//fdPthtvvLHy/VAm\nnNp2223r/h4dIyat0ntZLIc2WBpdk3sNsO6Qj2KMMcYYY4wxxrSARiO5nwbOL4riRqCSH7tWq32v\n2Y0abn7wgx8A1XTwZv5QBFdp9cHrRucXzXytvfbayTZ79ux2NacnUGQsrsE77LDD2tWcnkAz7N/8\n5jeTTWvIYikN0ziaQf/lL3+ZbFqvpugjtG79Vy8R1+PvvffeQDWydeKJJwLVaEinrcXty0DrapsR\nta1HjFL++7//O1AtA6P+fNdddyXbr3/9a6A6RiiS205/546dK+uTK4mV831u3bK2652rGFFbd923\nYkwxx4HWOcfvVWRYETUoo5exJFaryzLpeLGtiqQqzwiUkfIYuZPvo4+kxtC63mjTOtmHHnoo7ZNf\n4u9W5Deq4/bbb7/Kd8XjRpWB1B6/+tWvki2u920FubXjasOtt96abEsuuSRQXc+tvhWvWykC9Hko\n7y86LzFPwcUXXwzAiy++2K9t8d3qpz/9ab/vFbnSWSo5BOU68vnpr42+5B4JjAVmAksHe2eP/MYY\nY4wxxhhjFigafcndF1i7VqsNPcWVMcYYY4wxxhgzzDT6kjsD6JyMAEMgSj+UEGGPPfZoV3N6giiT\nOProo4GypIWZfyZPngyU0k/orMQc3cgZZ5wBVJMbRGmXGTyf//zngVL+NG9cQwAAIABJREFUBWV5\nhk6XeXYqKvuw0047JZsSx5x++ultaVO3o/75k5/8JNmWWGIJAO64445kk1y5GxLR5RI/5aTJw3Ud\n6rkq9tNPfepTQFUKqSQ1P/7xj5Pt7rvvBqpS8U4YLySNjMlz1BfiEqxc+aZ67W+0RIs+FyWfJ5xw\nAgBrrbVWsql9SoIEZSKkP/7xj8kmOXC7kk1BfZ/GxEjajnJlPfPkJKs5CXjfv4PyvOh6B/jkJz8J\nwPe+V664VMLE3Pc8/ni5UvPkk08G4Prrr+/3G1tFrq+prZJTR1uUFUt2H21KChd9r3Ok6zzK9bUv\nXufrr78+AMcff3yyadlCPPdqexxj1eY4PjdjOU6jL7knAWcXRXEc/dfkXjbfrTDGGGOMMcYYY5pA\noy+5ygX/wz72GjC+ec0ZPiZOnJi2FcVRanUzNBRtgHJW84gjjmhTa3qDmEhBiQA+/vGPt6s5PUEs\nPTB+/FvD1Xe+851k64aITacRZ2W/8IUvAHDzzTcnW0weYRojRiKUHDFGI77//e8DZRIVMzik3Npy\nyy2TTZGMWKZt3rx5rW1YE2hXWaaxY8cCcOyxxyabEtzEKMzPfvYzoJq4Zn7KgrSCgSKHg/Vzo1He\npZd+K+1NjIYpUWJUJCqCG5VISvKjsnjxc50UJYfyvpuLMMdxr1H1mu5Jud85YsQIAA4++OBk0zUf\nEyLpb2OpKyXs+stf/pJsf/3rX4GqIqxd5JLOxWtLid3iuKaIeXz20TN8VCssvPDCFVs81lJLLQXA\nnnvumWy5pL59k6zF48aSXL///e+BUuHR92+GSkMvubVabfWBP2WMMcYYY4wxxrSXRuvkGmOMMcYY\nY4wxHc87RnKLophWq9XWe3t7Nu9QLqhWq40bprY1FSWbAjjzzDOB1i8U7zW+/OUvp21JFV23cf44\n4IAD0rbqvD3wwAPtak5PMGnSpLQteaKSy5ihsdpqq6XtUaNGAWXCGWhvkpNuZYUVVkjbu+22G1Ct\nIal6l50gO+wWohTxqKOOAqrJZ5TEK8po7d/6xEQzSogUxwM9V51zzjnJpoSU3fp8kKvzKprRX2LS\nPi1L2HXXXZNNMuUoL1Ut1LhETM9hnZbMK4d+S5TWxvq0QjLk+LyeOw99n+cl+4byWfWwww5LNkmY\nI5JGz549O9nUd0899dRke+KJJ4CqtLcTUPtjgij1regzLeGKPpM0OVe7VqhWO8DnPvc5oPpuNXLk\nyH7HytUZln9jolq9l8W2N6Pv1pMrHxy2D3jHTxljjDHGGGOMMR3CO77k1mq1a8J212Zo0ozCRhtt\nlGy77LJLu5rTE8in22+/fbLFCKQZOp/5zGfStsotdOpMbLfw3e9+N22fd955QDV1vhk8v/zlL9O2\nZrUvu8yJ9ucHJeeBcvY9JvTphCQn3YaSogGMHj0aKPsrwFe/+lXAyoNG0H1/n332SbYPfOADlX0A\nf//73wH42te+lmzPPfdcK5o47DT7XqwIbXx+OvDAA4FqwkRF3KZPn55s//Vf/wXADTfckGzdlERR\nvhzo2lMkNxcdzKkxlbBz3333TTZd57FEkL4vRmOVkEmqDygTTkUVQqdFcIXapWRTUP7mGElVlDsq\nXfS3sSSV+pM+/9GPfjTtUwRX0VuoH2GPJZjUd6+44opkU2KsZvu2nlz5e++0L1Kr1b4z8KeMMcYY\nY4wxxpjhp55ceWzYfjfwUeBmYBYwDtgccFV6Y4wxxhhjjDEdQz25ctJNFkVxKvCJWq12erDtDeyT\n+9tOQgulY03cGMo3g0c+nTp1arJNmzatXc3pCVRbcO7cucl29tlnt6s5PcFyyy0HVBPNHH300e1q\nTk+gOs5x+Yeky92aVKbdqJ9+5CMfSTYlnTvrrLPa0qZuRz499NBDk03Sux/96EfJNmfOnNY2rIsZ\nN+6tHKNRyilZfbxvHXTQQUBZX9T0RxLcHXbYAYD//u//TvtyCZEeffRRoEz2A2Ut3G5fzpRLdBTr\nsOcksNof6wYrIdr+++8PVJcp6fkqfpdktE8//XSyfeMb3wDKJH/xc91ElK3PnDkTqD4HyVeSdkN5\nb4++V1/8+Mc/DsBXvvKVtE+JEgdKAqb3rSh1vu2224DWyL4bLSG0K9D3bns2sFtzm2OMMcYYY4wx\nxgydenLlyHTgS8Avgu2LQMfXNllxxRUB+PWvf93mlvQOmtGN6b+7cbark5gwYQJQJjkAR8bml/e/\n//1AOeMN5aymGRp77bUXUI3c/O53v2tXc3oCJeiJiWaOP/54wAnSBouiEN/5zlupQlTeCuDuu+8G\n4Le//W2ydXsUrBUoAvSnP/0JgJVWWintUzKbqJBRaRtTJUYd11lnHQB+//vfA6XyAMrI2PPPP59s\nSqB27bXXDns720kuoZT8ESOM2l5qqaWSbaeddgLge997K51QLMmm74jXu5KhqVwbwJQpU5rwK9pP\n/J2vvPIKUPWpIr2xZJP65/jx45Ntq622AsoIbiwhFM9H3+M+/PDDyTZx4kSgGjFvJY2+5H4OOLMo\nin8HHgFGA68Dew9Xw4wxxhhjjDHGmMHS0EturVa7rSiKtYAtgVWAOcD1tVrttfp/aYwxxhhjjDHG\ntI5GI7m8/UJ79TC2pan0rYE1a9asdjanJ5BPJau7884729mcnkCSD8lAYt0wMzQWXvitYW299dYD\n4IILLkj7OrW+XaejRBUf/OAHAbjlllvSPifyGxpK9CEJeKzf+te//rUtbep2JLX753/+Z6B6vR9z\nzDGA6w03QpQiqm7rZpttBlQTzSj55C9+Ua5kswQ8T6zRquUIq6yyClD1qSTgJ5xwQrLFe9iCQK4P\nRZuktVtssUWy/fCHPwTKZ6lcQqSXX345bf/rv/4r0DsS5XdCfot1cnV9x/FR9/iYAFHjqHxaT6IM\npSR5u+2262drF40mnjLGGGOMMcYYYzqehiO53YZmejRrq9TkZuhopkezYZ4Rn3+Uol0zZC5pMf+M\nHDkSgKWXXhqAm2++uZ3N6QnWXXddAEaPHg3AkUcemfbFcgWmPjG6sPvuuwOlTydPnpz2xaiuqU9M\nnqISIEqOdN9996V9Z5xxRmsb1sWoTwJ8/etfB8pyIzEh4re+9S0A/vGPf7Swdd2FlEUHHHBAsikC\nqft+jIbde++9QLWs0IKc2FO+iWPn2LFjAfjmN7+ZbGPGjAHy0Ubdoy655JJkO+mkk5rf2A4m9jFt\nqwQYlIm7pCyCMnlXzqdCia2gTFDVScpZR3KNMcYYY4wxxvQMfsk1xhhjjDHGGNMz9KxcWdJayZad\nDGH+kfxTEocFWULTLCSxkRQsJgcwjROlTBtuuGHF9uyzz7alTd2OZHYAO+ywA1DWbb3jjjva0qZu\nR8sTAPbcc0+gXPbx5z//Oe3z2No4sa6j+qnG0ShJjHVHTR5d80o2BbDyyisD5TPUXXfdlfZdc801\nLWxd9xDvR/Lf5z//+WSLMlGoSsBVc3jevHnD2cSuQb5ccsklk00Jkd773vcmW6xDDNVnftVtPeyw\nw5JtQVvCGPuk5MerrbZasn3oQx8CYMUVV0y2vj6NyH9nnnlmsmm7k963HMk1xhhjjDHGGNMz9FQk\nN85UKOW1yt100sxCNxFncpR8RlFHMzTiLO7OO+8MuJ/OL8stt1za1iyvZisdFRsacZZ37733Bsr+\nGSMPZmB0b9p2222TbZNNNgHKRH6zZ89ufcO6GCWc2mOPPZJt+eWXB8qyFRdeeGHa57F1YJS0TxFx\nKKO7UhycfPLJaZ8TTuWJydDkyxgh61vWJZZkO/fccyufWdDRM6gi4gCTJk0CyuemiO73MRKuRImd\nlBCpnchv48aNSzYlm4vP/H37YEwydd111wHV5F9xf6fgSK4xxhhjjDHGmJ7BL7nGGGOMMcYYY3qG\nnpIrRyRX6sTwebfyyCOPAKUMxDLQoaGkaABz584FYNq0aQC88cYbbWlTtyIZ6CqrrJJskuLcdNNN\ngJN5DRbJE7fbbrtkk4xRCafs08Gx1FJLAdVamZIv3n///YDrDTdCrNe4xhprAPCxj30s2ZSc5skn\nnwTgueeea2HrupN4P9pmm22Asg4plP3yscceA+DGG29M+yyprSKpp5bLAWy++eZAdTmdkvbo/v+H\nP/wh7XOixKqvJP0eNWpUP1tMHqW+qARz0adKiLQgj7HRpyJev1qOEN+Z+i5VkJQe4KijjgKqy2w6\ncTxwJNcYY4wxxhhjTM/QU5Hc3KyEmT9iZPGBBx5oY0t6hzhTG5N4mMGTK2tx0EEHAWXfteJgcGi2\ne/Lkycl2+eWXA6VCRqWETGPIp7GkzYwZMwC46qqrgNK3pj+KQsSEPkqU8vjjjyebkiJefPHFADz1\n1FOtamJXkIuQrbTSSsm21VZbAdUx85lnngHK5EgLcoK0XDQsqgukJJDKAMoIZFQV6Pn06quvBsox\nABY8NddAPl1iiSUAWGaZZZJNasI5c+Ykm8bYSy+9FIATTzwx7VMf7sRI43CQ82lE/pXiBeDWW28F\nqkkltX3++ecD1TJ3UnZ2en91JNcYY4wxxhhjTM/gl1xjjDHGGGOMMT1DT8mVjTELJlGG5KRIzSHW\nwJS01gwN+VIy2rgtaeiCIqUbCn3rigJcf/31ANx7773J1rdGpmoQm/7Ip3Hpwemnnw7A3XffnWyS\net58880APPHEE/2+Y0Eh/l5JPqO0VkkPJZuHcimN/AjleHrJJZcAVcn9guxTJTqK9W+XXXZZAMaP\nH59s6rNK2AkwdepUoPTpgrxUIefTmGBOSQ+VFA1gxIgRAMycOTPZtERBsnol9YKy38elDZ3Ydx3J\nNcYYY4wxxhjTMxTD/eZdFMWTwKxhPUj3sWqtVhs51D+2T7PYp83HPm0+9mnzsU+bj33afOzT5mOf\nNh/7tPnYp82nIZ8O+0uuMcYYY4wxxhjTKixXNsYYY4wxxhjTM/gl1xhjjDHGGGNMz+CXXGOMMcYY\nY4wxPYNfco0xxhhjjDHG9Ax+yTXGGGOMMcYY0zP4JdcYY4wxxhhjTM/gl1xjjDHGGGOMMT2DX3KN\nMcYYY4wxxvQMfsk1xhhjjDHGGNMz+CXXGGOMMcYYY0zP4JdcY4wxxhhjjDE9w8LDfYCiKGrDfYwu\n5KlarTZyqH9sn2axT5uPfdp87NPmY582H/u0+dinzcc+bT72afOxT5tPQz51JLc9zGp3A3oQ+7T5\n2KfNxz5tPvZp87FPm4992nzs0+ZjnzYf+7T5NORTv+QaY4wxxhhjjOkZ/JJrjDHGGGOMMaZn8Euu\nMcYYY4wxxpiewS+5xhhjjDHGGGN6Br/kGmOMMcYYY4zpGfySa4wxxhhjjDGmZ/BLrjHGGGOMMcaY\nnsEvucYYY4wxxhhjega/5BpjjDHGGGOM6Rn8kmuMMcYYY4wxpmfwS64xxhhjjDHGmJ7BL7nGGGOM\nMcYYY3oGv+QaY4wxxhhjjOkZFm53A5pBURT9bLVarQ0t6R3kU/vRGGNMs4j36/m5z/jeVPKudw09\nXpHzo307tH6aexYVb775ZnMatoAhnw50PrQ9P31X35u7nuL56/brI/c7F1poIaD62/SbG/29uf6v\n71100UWTTdsvv/xysr3yyisAvPHGGw0dq1EcyTXGGGOMMcYY0zN0RSQ3zg5oVmDhhcumL7LIIpV9\nUM48aHZA/8Z9CzJxBkfb0afyZW5m5rXXXgPg1VdfTbYF2ae5mUb5L/bJ3OfkN81eybdxX6+T62O5\nmUZtNxq1kE/jzGAvz6bnZrpz+3M+jeRmxPvamj2D3mnUix7kbPE6F7kZ8UajAb3oU91foq/U/3K2\neD+S3+K1/Prrr/ez9Y085PrpQHSTzxURWWyxxZJNfTJGTuRfPStB+TvlRyifk+J9SP7NRXXqjae5\n+1w3RMNyz5Pqk9F/2l5yySWTTdtxXFW06h//+Eeyyec5n9a7b+XG7uhTPZPF89cJ97zcfSY3duqa\nHzFiRLKNGjUKgFVWWaXf38bfKZ/KFscPXR/xOon9vh7PPvssALNmzUq2J598st/xO6E/13sWjT5d\nffXVAdh8882TbfTo0QAss8wyybb44osD5fmLY8q73/3ufp9XX3/hhReS7aWXXgKq14nOzU033ZRs\nZ5xxBlD1c/TvUHEk1xhjjDHGGGNMz9BQJLcoitWBCcASwMPA1Fqt9uxwNswYY4wxxhhjjBksdV9y\ni6JYGZgMfOBt05vA88DCRVEcC3ynNowx+lyIfOTIkQAstdRSySbZSJSSiGeeeQaARx99NNmibGRB\nIEoX5FPJEABWWGEFoCon0N9EyYf+9rnnngPgscceS/uiPKGXyclB5KPov/e85z1AKemAvNxW2y++\n+CIATzzxRNqnvtsJMphmM5DkU9d8lMLo2s9J7iI6H5LOPPXUU2mf/NsMGUynUE+GHH2la37ZZZdN\nNvXT+Lc5aajOjWzqr1D6VP01fke3kpPMajuOnZKARZ9qO8rh1Bdjv+sr3433Od2jZsyYkWySzXXr\neCCpYLx3L7300gCsuOKKySZZovomlOPo888/n2zajn1NPpRvo6/k0zgePPTQQwDMmTMn2XSuusHP\n8qUknVD6csyYMcm22mqrAeW9HkofqV9B1b9C50j3tzh267kqPl9JnhjHiOnTpwNVKWJMOtNJqK/p\ndwMst9xyQCnpBFhvvfUq/0J5HqKPcte+xo3ll18eqN7n9LnoK21Hn+lz9913X7JdfvnlQPU5ohOW\nPuSWJWg8iGOnZLTbbbddsm277bZAdYzIScrVnzUGxHFa++I9TX093rfiEjwh/5500knJJpltvF6a\nnTBpIHJLkuTnKMuW37bffvtk++QnPwnAmmuumWy6vnPLHBpNyCX/xTFF98F4Pen74vX04IMPAqUU\nvO/3DJWB5Mq/A+4GRgNj3/7/fwObANsC35/vFhhjjDHGGGOMMU1iILnytsBHarXa6wBFURwKzKjV\naj8uiuJA4Abg/2tmg3JRxziDs+666wLVWRO97ceZc83GbbPNNkB19vb8888HumOmthlEn2qWZty4\nccm21lprAdXkXJrdirM08umGG24IlDO2AKeffjrQ+GL+biUXddQM2Prrr59sa6yxBlCdza7XT8eO\nHVv5foA///nP/b6jV8glRsjNPm622WbJJh/F2T310zhLLp+uvPLKQDkLD/DHP/4RqEbIup3cLGsu\nKYT65NZbb51sijw8/fTTyTZv3jygGjXQuLHSSitVvgvgwgsvBOCUU05Jtl6J5MZZbfkyRsje+973\nArDJJpsk2xJLLAFU7znqp3HMVDRC0TV9F5RRx2OOOSbZrr32WqB7x1j5Jd7PN9hgA6CaAGXjjTcG\nyqgOlPemnHoojiWKbCqKEyOX8neMFCjZya9+9atk69QIYw5FwWJERP1IvoXyHi81DJR9MV77+u2x\n3+ua1zga732KgMfv0HUS+/9pp50GwCOPPDKYn9cWdB+KkUD5YPz48cm29tprA7DOOuskm/p4vEYV\n7YtRQp0HfW88lpRyUfGg57V47Wjsvvvuu5Pt9ttvB+Dxxx8f+Ie2gfjMnVMLKsId+6mer2K/ky/j\nM6v6rM5BLkldjNpqHIjHV2QzRpcVgTzvvPOSbbCldYYDHXugBG/ajqpCbce/1TNUfJbSfn1H7h0h\n+lTKmPiMtsUWWwBlX4dyLI7nWcdtdQmhx4E1w//X4C25MrVa7SFgqdwfGWOMMcYYY4wx7WCgSO6P\ngSuLovgrUAD/DBwOUBTFBkDvhEOMMcYYY4wxxnQ9dV9ya7Xab4uimA7s/rbpE7Va7dK3t6dTJqQa\nFnKSMUk5JKmDqvxISFbymc98BqjK6yTd6SZZ0lCoVyszSjQkLcgl6InIp/vuuy9QlUNdfPHFQFW2\ntKCQS6qg/hllQ1FqJ+TT3Xd/6xLbdNNN074rr7wSgHvvvbfJLW4/OdlSJFcDUNd5lLw9/PDD/b5P\nsmb5dMcdd0z7brvtNqC35Mq5xCI5n0qGFz83d+5coOoPJYCIsiUlAZI0b5dddun3HSeffPJ8/IrO\nQv6LvtL1nbPFZIYaT6NPlTgmyr10PuTTD3ygvJ1K8hnPY7dLwPtK3+J2TFyoe0j0qa7ze+65J9k0\nHuQSpOn+FiXgGgei5FMS3FYnjWkWand8ltG9OyZC1JKX6GdJC2fOnJlsktbGhHWSNvZN6AelNDTe\nt/bcc0+gKjeX5LMbpPZqY+x/uufE61d+vvHGG5NNvonPT+qnUa4sX8rPcUmSron4jDZx4kQAvvzl\nLyeb5J/xuS2XNK0TluXl5L2Sy0dfqf3xWUnLCyP6m/ge0FfCnEsiFZf0qO/G54PvfOc7QFXaq/eP\n+Gyr7+4E3+baEH+7fHXBBRckm/qzEtJB2Qdnz56dbEoop/ORS7SVSyIal0l997vfBWDChAn9PheX\n72iMj9dYMxiwhFCtVrsCuCJjfwVobmuMMcYYY4wxxpj5oKE6ua0kzkpoRi1GbjQrFmfZNEsUZ2mU\ncEKJa2JSm26YTWwGuQiPZkkeeOCBZNOM7kA+1Sy5fBpna3o9Ki5yM5JSF9xxxx3JJl/FGVrNfMWI\nrxIcbLTRRkCZLAmqC/p7jei/XMIBXfPRB5r9i9EIXcuxn6ofa6Y7+rQbEp8MFvky+k8zpDEae8st\ntwDVkhO5hBy58kqa9VZytZiYQ5Ggbo2G5dBvibPKipjoeocy8U5UG8n38XOaWY9RTPVZlVdRqYz4\nfXFWvdsjuRoLo7pF1+pdd92VbLlyQYqi5MbT6BddCxpjr7rqqn7fG1UId955Z7/v7YToTKNINSD1\nBZRRl5tvvjnZdJ3HyFdujGgkoU6M3ChhUhx/d9ttN6B67UybNq3fsToV+SVGw3TPiaWmpk6dClR/\nk/wcnzFzz2GN9LH4nKCIb4wwKtIbI/FS1XTaWFHPB/G+oefInDozpwCZn2tV43O8H2rcjW2SCkd9\nGDorkhup98yvSClUS36JeuPpYImJp9R347ihY91www3JJsVis9/PBko8ZYwxxhhjjDHGdA1+yTXG\nGGOMMcYY0zN0nFw5orB5lMIq9J4Lo8dkCd/61reA+jWzFkTkhyhNziVZEVGGtP/++wOw/PLLA1U/\nxu9bEMjJ6qO8Lve5HErepWRJUSqSS/7Vi9STLeWSEOR8mqu7q9ql8fNRUt5r5CTg8RqVxCpKiXJ1\n9XJoHFA9zjguKOlcL5GTbsmXMVlG7prPSenq9Vmdjyh5ljS6U+tdDgVd01EGqt+eS14S+67OR6Py\nOX0+nh/VGo3SZEn4u0FGm0My2tgndW1GX8kf0afNkLTqO2JSGSWpU81WKGu5dpqMNkdu7JQtd003\nQ96ZI36XkqXqXyivo0suuSTZ1Lc7VUY72H2N7J9fYuIpJaKLzx2TJ08GqmNxN/RjIf+18r1H710A\nkyZNAqrPDPJlrAOvsazZ57uhl9yiKJYBvgpMBEbEfbVabZfsHxljjDHGGGOMMS2m0UjuacBCwJnA\nSwN8tuk0OiM+atSoZIsp7aG6uNw07tNYhuCf/umfKp+JyasWZHKL/esRE0p8+MMfrthiApFOm41t\nJYP1aYwEKWKj5EgxiUWz09N3OrlozlBQ/9xiiy2Aqr97eWzN9b9oa8aMvkrabLnllsn2s5/9DOgt\nhYx8NZDPmjnuqTwTwHbbbQfA//3f/yWbxttuisxEcooD0ex+mkMlSPbbb79kU3Kk8847L9k0BnfD\nPa1ekqRWluZR2SWAww47rJ9typQpQFVJ002KhHb3hZEjRwLw+c9/PtmkpokqhLPPPhvoLt+2Cz0X\nfPvb3062nErxL3/5C1AmXYTh6w+NvuRuCaxQq9X6F54yxhhjjDHGGGM6hEZfcq8B1gXuHMa2DBlp\nvVUwG8oZGc0OaH2eGRxaXwPlegXNHseSF6Zx4tpxrWXSeol99923LW3qdmJkMY4DAMcff3yrm9Nz\nKIKgEkJxBnZBKR/WbNRnP/OZzwBlySsoowfdGmHM0crIjcbYI444ItlU7ua3v/1tssVyZN1IPcXL\nsEVGFi4fGw866CCgulb06quvBso+DNV12N3IYJVF84NUMx//+MeTTQqauMb85z//OVAtB9Pu6Gin\nE1V0P/nJT4BqiUGNBz/84Q+TrZtUCO1GOWYOOeSQZNNYHJWfRx11FNCacaHRl9xPA+cXRXEjUMmE\nUavVvtfsRhljjDHGGGOMMUOh0ZfcI4GxwExg6WD31IYxxhhjjDHGmI6h0ZfcfYG1a7XanOFszGCI\n6aiVHGnvvfdONkkLtIDckrqBiZJPyb2V/htKn0+bNg0oy1yYxpD/xowZk2wrrLACADNnzgTg1ltv\nbXm7uhn1WSU7Adh1110BePrppwE4+uijW9+wHiBKuz74wQ8CsMwyywClTNEMHV37StoTk87NmDGj\nLW3qdjQeaHlSLA9yzTXXAHDTTTclW7dLENX+eO8e7t8U71/qu1F2eOKJJwIwd+7clrWpmbS7raus\nsgoA//Zv/5ZsenaI5YIkB1+QS2IOlpjcb8899wSq7xJXXXUVAFdccUWy9dKSkeEgLr8766yzABgx\noizCo7HhX//1X5Otle8O7xr4IwDMAJxazBhjjDHGGGNMR9NoJPck4OyiKI6j/5rcy5reqjpoxnKx\nxRZLtg033BCArbbaKtmU7ltRHTMwMaGEkqB87GMfSzbNGH7iE59obcO6mJziICaXks+Vcr3ds8jd\nQIxayH+xZNh6660HwDnnnAP0VhmWViD/xqRzX/jCF4AyCccFF1zQ+ob1AHGM3WuvvYAyoV8sZdHt\niXrahZJLffGLXwRKNQeUSah6UdXVivuG1F3x/q+kPeeee26yXXTRRcD8lSxb0Ijjwve+91aaG5Ve\nAXj44YcBOPzww5Ot25OmtRKNCyeccEKyKdo4b968ZFOpppdeanml1K5FiRMBVl999X77L7/8cgDO\nP//8ZGvlc26jL7lfevvfH/ax14DxzWuOMcYYY4wxxhgzdBp6ya25LZ00AAAgAElEQVTVav1fz40x\nxhhjjDHGmA6j0UhuxxHlyqrXqqQoAM888wwATzzxRGsb1sVEyYzqXUX5geQxSjxlBibKlUePHg1U\n6wtLVn/eeee1tmFdTJQrS3KkJBJQ1nT9zW9+09qG9QhKJKHajADve9/7ADjzzDOB3pR8Difqs/Ee\ndcABBwDlvUpJO8zgiOOBkiJtvvnmAJx++ulp37XXXgt4SchQ0RKmT3/608mm56sjjzwy2aL80zTG\nhAkT0raWiMWER0qeGJ+93I8b59BDDwXKOu9Q+vfHP/5xsjnhX+PoXhYTe+p5N9Zz/uQnPwm0L0Ha\nO77kFkUxrVarrff29mzeoVxQrVYbN0xtM8YYY4wxxhhjBkW9SO7BYfuA4W5Io2j2KrcwPEYXtIDc\ns12No6gilCVZYiRSi/adUKJx4mysfKpEM1CmqncSicaJ17QSSsTEU48++ihQlg8zg0Olg3bZZZdk\nU7Ts2GOPbUubuh35LyaTWW211QC48MILAY8BQyXeozbbbDOgfD743e9+l/ZZfTA05F+VERs1alTa\np2eC2267LdlccqVxpJ6L0UTd02IZm9///veAn70Gg5QHAN/4xjeA6lihkmLHHHNMsvl9oT5RNXPy\nyScD1XJB6p9f+cpXkk3JKtvFO77k1mq1a8L2la1pjjHGGGOMMcYYM3TqyZW/18gX1Gq17zSvOcYY\nY4wxxhhjzNCpJ1ceG7bfDXwUuBmYBYwDNgdOz/xdS4iLmCVLjIk7Yk0m0xhRZvTII48AcMMNNyTb\nn/70p5a3qduJ8hfJEWfNmpVskydP7vc5U5+cr2LflQzJ8sShIUnSWmutlWxz5swB4NZbb21Lm3oF\n1XSHUjr35z//GfAYMFRUfxxgjz32AMr+6iSJ84/qZR988Fsr2GJiKSX3e+WVV1rfsB5ACf1iMkr5\n9z//8z+TzbXeG0fLbf74xz8mm5aIxb6rxH+uSd44H/rQh9K2li/E+5aeDyRl7gTqyZVThd+iKE4F\nPlGr1U4Ptr2BfYa3ecYYY4wxxhhjTOM0WkJoV2D/PrazgT80tzmNE2cP7r//fgB+/vOfJ9uLL77Y\n8jZ1O9Gnc+fOBaqJO2bPnt3yNvUS8u8FF1yQbJdeemm7mtO1xOQHStIxderUZIvjgBk8Sy21FFCN\nzhx33HFAPuGfGZhFF10UgIkTJybb9OnTAZgyZUpb2tTtaByI5Vc22mgjAE455RTAybyGSiwnuP/+\nbz36qfxKjNLMnDmzpe3qFZZYYgmgTOSn+xjAGWecAThx4lDZa6+9AJg0aVKySel1/PHHJ1tU1Jn6\nqFzQr371q2TTGBGfCfbZ5624ZyclSHvXwB8BYDrwpT62LwIPNLc5xhhjjDHGGGPM0Gk0kvs54Myi\nKP4deAQYDbwO7D1cDTPGGGOMMcYYYwZLQy+5tVrttqIo1gK2BFYB5gDX12q11+r/5fAR611JVvfs\ns8+2qzk9QfSppGAPPFAG613/bvBEaa3q5N59993J5uRIgyf6VAkl7rnnnmSzrH7wRJ+OHj0aqCY7\nOeeccwAnRxoqK664IlBKPqGUI1pSOzQWWWQRAHbeeed+tssvvxzwPWuojBw5Mm0rQY/Gg5jQJyYA\nNfWJY6wSpCkR3fPPP5/2HX300YB9O1gkqT388MOBcokIwEMPPQTAT37yk9Y3rItRnz3kkEMAWHnl\nldM+ja1KnAidKQFvNJLL2y+0Vw9jW4wxxhhjjDHGmPmi4ZfcTkEzC4sttli/fU6zPn8o0gjlLNij\njz7arub0BLGfqtTFgw8+mGyOjA0eRWugjOTecccdyeaSAIMnJppZddVVgTIxEjg6PhSiMmbdddcF\nqtGFa6+9FuisJB3dxLLLLgvAlltumWwPP/wwUFUgmcZRn91xxx2TbezYt6pJzpgxA3BZpqGiexXA\noYceCpQJp6688sq076677mptw7qYGB2X4kCl7157rRSafve73wWqEXMzMCuttBJQ9td4T1OZ0f/4\nj/9ofcMGQaOJp4wxxhhjjDHGmI7HL7nGGGOMMcYYY3qGrpMrS1YnqRKUyXss+xoa8mlMNiHJZ6yV\naRpnoYUWAmCFFVZINvXPZ555pi1t6nYkTXrPe96TbJLPxIQHloAPnqWXXjptL7/88oATpM0vcfnH\nOuusA8DTTz+dbPfdd1/L29TtRLncmmuuCVTvW7fccgtgWeJQUY3s3XbbLdk0nl599VspWbwsbHDo\nvrX99tsn2+qrrw6USef+53/+J+2LMltTnygB/9SnPgWUY8T999+f9p111lmtbVgXo2dXKGXK8nNc\nCvanP/0J6PznWUdyjTHGGGOMMcb0DF0RyY2Ly5dYYgmgXBANZXrw+DlHc+oTfTVixAgAVltttWR7\n7LHHWt2krifXTxVtgHKGNn7ONI4iY9GnUnTkyl95DBgYJfFSZAHKMgExeY8+56ReA6O+GCOMSuYV\n0Rjh/to4MZHfeuutB1STeSnKGMcDU5/oK5UPW2WVVZJt3rx5QBklt2JucCjhZEzmJZ8rud9NN93U\n+oZ1MfLfP/3TPyWb7mEqvRSTeblMW+NMnDgxbe+///5AqfbUWABw7rnnAp1/3/KdwBhjjDHGGGNM\nz+CXXGOMMcYYY4wxPUPXyZUlV5KkDsqFzzGpR6eH0NtN9KnkNFGuLMnHzJkzk80ypfpEnyqRj2q2\nATz33HNANeGM5HXur3milE7JDzbYYINkk9Q+SmvnzJkDWFr7TkSfLrPMMgBsuummyTZ+/Higer3r\ncy+++GIrmth1xGtfsnrJaaGUgaouJpRyZp2PN998c9jb2a1ILheXKW288cZAef+C0vcxeYrJoz4b\nE6RtsskmQD4J5bPPPlv5O/B9652I/W/ChAkAbLXVVv32q66z71UDE/udajcffPDByaakaUqSePvt\nt6d9HlsHRs9X3/72t5NNST51natWNnRP4kRHco0xxhhjjDHG9AxdEcmNszAqDaB09uCU60Mh59OL\nL7442TSz6Jnaxom+Uumla665pt/nou+ddKZx5Ku77ror2ZR4SsqD+DmTJxfJlcoAYO7cuUB1XFUk\nzf01T4zcaPY7lrrS/liKSefB/TWP+hzklTFK5hWvffVZ+dtRxyrRH0omt+KKKybb+uuvD1Sv/Sef\nfBKAl156qd93mCq6pqO6QIoO+Q/KCO6UKVMAP8M2QhwPFMmN961HH30UgKlTpwJw4YUXpn2+9vPk\nFEixdKAUSPLtf/7nf6Z9nV46SDiSa4wxxhhjjDGmZ/BLrjHGGGOMMcaYnqEY7jB+URTWCfTnllqt\n9r6h/vFw+XQgGVKHSz46yqfy5UD1GiVd7lDfdpRPJUGUzO7tY/T7nKT2HZooraN8qhqjkoNCmRwp\n9t2nnnoK6NhEaW33aa5GdkyOOGrUKKDad5W4QzXJO6y/tt2nsf/Jp5LPQSmt1ZIFgDvvvBMoJXdR\nHt4BfbbtPo39VPJPJeyBMumc5KBQShWnTZsGVGuOdkBCn7b7tM/3AdXrXAl9xowZk2zql7Nnzwbs\n0wa/N20rAe3yyy/f73Oq5eprf3BovI11x+XzDn2masinjuQaY4wxxhhjjOkZWhHJfRKYNawH6T5W\nrdVqIwf+WB77NIt92nzs0+ZjnzYf+7T52KfNxz5tPvZp87FPm4992nwa8umwv+QaY4wxxhhjjDGt\nwnJlY4wxxhhjjDE9g19yjTHGGGOMMcb0DH7JNcYYY4wxxhjTM/gl1xhjjDHGGGNMz+CXXGOMMcYY\nY4wxPYNfco0xxhhjjDHG9Ax+yTXGGGOMMcYY0zP4JdcYY4wxxhhjTM/gl1xjjDHGGGOMMT2DX3KN\nMcYYY4wxxvQMfsk1xhhjjDHGGNMz+CXXGGOMMcYYY0zPsPBwH6AoitpwH6MLeapWq40c6h/bp1ns\n0+ZjnzYf+7T52KfNxz5tPvZp87FPm4992nzs0+bTkE8dyW0Ps9rdgB7EPm0+9mnzsU+bj33afOzT\n5mOfNh/7tPnYp83HPm0+DfnUL7nGGGOMMcYYY3oGv+QaY4wxxhhjjOkZ/JJrjDHGGGOMMaZn8Euu\nMcYYY4wxxpiewS+5xhhjjDHGGGN6Br/kGmOMMcYYY4zpGfySa4wxxhhjjDGmZ1i43Q2ox7ve9a7K\nvwBvvvlm5V8zOIqiAGChhRZKNvt0/pBPc/20VnMN7/lBvgX70hhjjDHGNIYjucYYY4wxxhhjeoaO\nieQqCjZq1KhkGzlyJACvvPJKsj355JMAPPvss8n2+uuvt6KJXYd8+p73vCfZlltuOQBeffXVZHvm\nmWcAeP7555PtjTfeaEUTuw75dKmllkq2JZdcEqj2wxdffBGAl19+Odns0zyK1i622GLJtvDCbw1N\nUV3w2muvAVU/O7pbn6jYiFFxkVMc2Kf1yfkxYv8ZY4wx7ceRXGOMMcYYY4wxPYNfco0xxhhjjDHG\n9AxtlStHyeekSZMA+OxnP5tsq6yyCgCPPfZYsp188skAXHTRRckmmW29xEk5iVkvSvSWXXbZtL3N\nNtsA8NGPfjTZxowZA8Cjjz6abKeffjoA11xzTbLJp/UktrmkQL3ix8jyyy+ftidOnAjAtttum2xj\nx44Fqj699NJLAbjzzjuTLddPG/FXL/o09lP1yfHjxyebJPaS0gPcd999ADz88MPJ9o9//AOo9tN6\n/upFX4olllgibS+99NL9bO9+97uB6lIF9ckXXngh2bS/nk972Y8RyeahKv0W9aTL0UfyZe7aX1B8\nmWMg6Xcjnxuu+/hAzwy9iH5zs3+nvndBeQ4bLqL/tHQqjkvyXxy7nVC0PtGnGu8XXXTRZJP/tFwK\nSv+6v+aJSVi1DE3L+qD05UsvvdTP1myfDhjJLYrikKIoriuK4tmiKN54+9/riqI4uKktMcYYY4wx\nxhhj5pO6kdyiKP4b+DBwNHAH8CywNLAxcFhRFONrtdp/DfagiyyyCAAHHXRQsh1yyCEArL766v0+\nF9/2n376aQBuvPHGZFNynzgDsPjiiwNlxGjEiBFpn74vRomU3CrOiikKEhNfKeFVpyURUpTm85//\nfLLtt99+QDVClvOpfvOMGTOSbfbs2UDVp5qJkU/jbNdzzz1X+RdKv8VZHf1NnBVTkqZO8+kyyywD\nwBe/+MVk22OPPYCqTzVTFZNMrbXWWgAcd9xxyTZ9+nSg6lOpGWJkU6h/xn6qY8TZR52/6D/5vtOS\nsq2wwgoA7LPPPsm28847A7DOOuskmyKRMep43XXXAfCXv/wl2RTdjehv5dvc+DFv3rxkUzQ4N4OY\nS3w12Ej8cKOo9/ve975k22STTQBYf/31k22llVYCqmPcPffcA8DVV1+dbOqnEY2FGmfkMyj7Z/Sp\nIsM5X8U+qe1GI/GtQmOdxgAor9EVV1wx2dSflSQx8tRTT6XtOXPmANWIucglWdP1G5MB5vppzqe6\nZnKRh3aifhfvBxrHYsRc21HpJd/HaIB+Z+yLssmXcZyULfpC2/GayEXsdYyc4qET+muOXNQ0/jY9\nC8RnI93Lcj7KRQRzJQn1Hbnzp2NCmURUzxpQ+rcT+uv8EPu4fnN8Xurbx3IR2qjC0Ri/xhprJNva\na68NVBM23nXXXQBce+21yfbEE08AnTfGNoNcf5Y/cvuUeBVg1VVXBar3zY033hgo73MAt99+OwCn\nnnpqst1///1AdYztFZ9G5EONu7FPatyIyYL13Lvddtsl23vf+16g6tO7774bgJ///OfJdvPNNwPV\nZ75m+HQgufJngQ1rtdqcPvZbi6L4G3AnMOiXXGOMMcYYY4wxZjgYSK480IKZxhbUGGOMMcYYY4wx\nLWCgSO7vgMuKougrV94IOAz4zVAOqkQ+SjYFMG7cOKCUGUei7ECS2igbkgwjSp4kWVxzzTUrx4RS\nJhNlHgqRr7vuusm2+eab9zvWKaecAlSTDHXCwn4l74k+laQ2SrxyiSUkjYuybO2PPpIsQbLSKF1Q\ncjDJ8qD0aZSgb7DBBkBVhnf22Wf3+9tO8Ol6660HVH2q3x6lWPJplGKpv0Xfq2/Hv5WcVDKPyCOP\nPALAgw8+mGySy0t6CmXiqygVl/xUUiXoDAmYpCs77rhjsm211VZAtZ6zfBmlK/rbKFHW55SkDmDC\nhAlAKSuVHB5g1qxZQCmXgdLPUU6mcxR9+ve//x0oJc/QGXJw9cnYT3fddVegeu1pTIzXlvpd9N/M\nmTOB6vnQ96hfx7FC1+3UqVOTTf6Nn9N1ovEX8j6NErB2kZMHbrnllkCZfA5KyWC8v+g6i+OZpMtR\nliips67leO+TxF79FUqfRl/p+oj3IyUQjH+bq4fcanStRtmarrMoAZdPo4xQtjie6jfF/iwfqq9H\nKa6I/tM5ivcjfV+U319wwQUA3HDDDcmma7+dyxdyCZ1yEnD5fPTo0cmmcVf/Qnke4r1dzwD6V76N\nx4jjtKTd0aaxNS4dOe+884DymSru77QlITnk5yhNlt9i3/3Yxz4GlFJYKMcL/W08V7pOclL7eCyd\n0zheXnHFFQBMmzYt2ebOnQt0xv1/foj3Z/ny0EMPTTaNzxojcku6ov/67oPS97H/aXy+5JJLkk3P\nZJ1wr2oW6rsHH1ymWzrssMOA8l4V/Ze7LmWLPlXfjp/XseIyn9xY0gzqvuTWarX/KIpiBvAZYANg\nBPACcBfwi1qt9uumtsYYY4wxxhhjjJkPBiwh9PaLbFNfZjUjk5sJyC2Ov+yyy5LthBNOAKqzrLm0\n6Y8//jhQJkXJzQzG2QbNTsZEI5qNi9+rxdGxrFEnRB01q5eLLOV+uxbTAxx77LEAPPTQQ8mm74nJ\nlLStEi65cxVnJOVLzQIBbL311v3aqQiFzlnfNrcLRQXibJ1msnLtUwQMyn6q/gKl/2Lf1SzrLbfc\nAlRnseSjnEJh5ZVXTrbtt9++8v1QnsuY/KYTZnI1gxd9quswjgf67TEadtZZZwFl5B9KX8aZ6zvu\nuAMoIw+KfkMZKcglqVtttdWSTdGNqOLQuYrnrxNQ+2OfzEUD5FP9DoCbbroJgAsvvDDZFBWM0RxF\nehUZi8nQ9H0xAZCuEylpoBxP498qUt5pPtU9KnePiEoM9bE4nuneoOQkUEZV4+ekXtJ1G69zXbex\nXJZ8Fa99KUGiWkH+jQl9OgH5MvpUfTdGwqVKitejIuvRR7qWo1pA/V7RnBj9kY/iNa1xKCoZdNw4\nFmsc0vUSaWekMRfJzUUH1Xc32mijZNt9992BUrEE5fmolzgpKpbkv6h40X0mnlNtR59KxRGviU64\n7zeKfBQVGB/4wAcA+Na3vpVs8m/8XL2SWPJH7Nfyc/yOXCRX5zze87q93I1+p5LTAnzzm98Eqs+W\nfaO0uRJuudJKsT/nknhq3IjjaacnnWsU3YOgVFbEZJW5yLfIJfITcYzPPTPrHhUVdervLS8hVI+i\nKMYN/CljjDHGGGOMMaY1DPkltyiKxYAHB/ygMcYYY4wxxhjTIgaqk/uBOrsXq7OvLpILRemPEqDE\nZCeScKreK1QTRPRloGRKfYmheMkPYpIfycji4mglIol1yDoByf3kMyjrgMXaYJJ1HnDAAcmmRDCN\n1gnN1XoU0aeSL0afSooTz7MSZOWSerQTSdSmTJmSbErMEdv/wAMPAPCFL3wh2fSbc1KOKCuOMq++\n5BJbKGlKbsF+TCoiyd1tt932jt/fDiT5iedaEtgow5RM84gjjkg29d2BEhPce++9QL4/5+o69q2p\nDaW8TtLJuB0lNp2AEmfFfioJlpKSQSkRivX+JO2Octd6kkH5NPq2njwyJhlSjekoQ9U5Vzs6Bd1n\nolz48ssvB6qJzyTpjjXGJVOOCY5yCXX6SsFy+6LcVmOOEtkA7LDDDkBV/ig5dadJ6TQWxnuyfBSv\nad3LYiIz9aN4/1dSvdh3+x4rfm+uJq6uj3/5l39Jtg033BCo3oN078slZ+wEYlvUj2L7dS+O/VRL\naXLJCeOSBv1m3bfiEhiNPfGZQEmBorxUyXtim9SWnNS5k3w7ELE/KTlnvG7Vd6KsWL5U/1d9WygT\nx8XxQ7XkNYZCfjmVlvZ12jKlwRLHRt0v9t9//2SL92qhfq9rPsqLlTQq+lRy/SjP1bmM1/kf/vAH\noOzr0J0+jeiZ65e//GWyKeluTqKs6zGOv1deeSVQ7Wvqp/FZTn8bfXrUUUcB1fFouJ75B1qTewUw\nB+iehRLGGGOMMcYYYxZYBnrJnQXsX6vVruu7oyiKdwP9p1AbQLMBmhmHcmYrzkoptXy96O38EGfO\nNfsTZxaUvCeW0cnNGncCisZedNFFyabZ2DiDcsYZZwDV2ZdmEn2q4yrSCWVSm+jTGNnsJHT+tSAf\nyghtTLKjfhxLozRjJjqXUE2+itFxneeYQGG4FvHPL2p3nLW75557gGo0QLY489pMn8bjK8oWE59p\nNjPO0ueimJ2AZpjjOKloYry21D9z5VLmB/XP+F06RkyAotn32J9zf9sJqN3xOu+bJA7K+0YsjdLM\naFQ8f7o3Rp/mSl3pPMc+3gl9Vu3PRUjjta/rMBedzCWOiTTyO2PUW+c33iMVhYvJ0KSIieejE/ps\nvTIe8VlK12NUocREiSKXzEvnKzf+aTtGf/Rs8dnPfjbZ5PM4xiq6Fq+dTuinjaLzH30lhVL0hxQu\nUVUl30tdExUHOZ/qufPDH/5wsulaiEkX/+///g+o9tNu8qnIJeaaPn162pZSKCaA1W//61//CuQT\nmeaSTMWyoerrKsUE5TNzPVVoNxB9KnVnVFrF61AoYeLXv/51oOoX9b/4LC9fSdEQP3f66acnm7Zb\nUYJpoDW5U4D3vcO+N4GH3mGfMcYYY4wxxhjTcgaK5O73TjtqtdqrwOrNbY4xxhhjjDHGGDN06r7k\n1mq1YYkl52QWkrhEmUWspzjcSM4QpWBXX301UJWsSW7baSjsH+VIamv0aUwoMdzIp1FKJ8lOrq5k\np8lqJE+RdBbKRDRRuiKftqL98mmU90kOnqvr2Gk+1bUkGQyU5z9KV/T7WunT6D+1sxv6qdoa/aex\nM8o7h1vCnpOXxjqGStIUZWS6njpB+hlRX4jnP+e/4ZZbx2NJbrbpppsmm+rBxpq8SrzWaclR1J4o\nm2skyVnf7fklfpfk3qpvCqWkUX4EuPPOO4HO82m95Hq5uuO5ZFQ5Pw/W3/HzEyZMAGDttddONvkt\nLqfq1H46WOK4q+U48TlRMtB4fxnsePHJT34SqCYR1fcqiQ/kpf7dSGy/ll+ceOKJyabfHp/NJMmv\n99ujBFwJ/KKEWcujvvKVryRbvUSr3Yr64jnnnJNsktpL9g3l83q9azTKlbfeemug6mddC1/+8pf7\nHb8VzFedXGOMMcYYY4wxppMYSK48rMSFzkpjn1uA30rirJwiZFrkDp0bIRMx6qxZ23aVPcglwFA0\nTCVaoHN9qvbEfqrf0u5+GmeCdc5jhEyzn50WIZOvov9yUYZ2+FSRRihnImNiC/m3U30a/aeZ1zgD\n20qfKqlHLBek48dEdEqa1ak+je3S9nBFGHPEqKei4ltttVWyaTxSOQcoZ847zadiIJ8Nt09jMrkD\nDzwQKMvYQZnk56STTko2KSM67R6Vo14yquFimWWWSduKLEplAKVyJ5Ys6dSEk42SSwypZ4X4HDZU\nVB4MYM899wSqfffGG28E4G9/+1uydeo1P1ji71ACOJVWivsH269jNFjlQON3HH300UCZdLSXyEXH\nTzjhhGTLPTPUQ/cmlSKD8l0pfsenP/1poKqSbSWO5BpjjDHGGGOM6RnaGsmNMwuKNg5XQeBGiW2S\n1jzOOHZ6sfLYrty6snYQoxGK4MZInqIR3eBTtbXd64i0lgzKaFksYdSps+S59WLtLiOjqO16662X\nbGPGjAFg6tSpydbKdcJDIRdhbJdPF198cQA23njjZFMpjeuuKyvSNbKOqp3k1sS2IyIO8KEPfQgo\n+yaUKpjJkycnW6de+/VopU9XX73Ml/mlL30JqN6j1D8vuOCCZGtFqYtuRGPncccdl2yKisd7vKK7\nsfRdp17z80MzfpNKrZ111lnJprJWsfzbwQcfDORLv/QSzbiH7bvvvgAccMAByaZrXnkhAI455pim\nHbOTyeUhGSwXX3wxAGuttVa/fTfddFPavuyyy4Z8jGbQ0EtuURTLAF8FJgIj4r5arbbLMLTLGGOM\nMcYYY4wZNI1Gck8DFgLOBHp72sgYY4wxxhhjTNfS6EvulsAKb9fGnW9yC/Y7RQYcU1+rTTGFeKdK\nFnPJZ9ohr8sRE/pIdhPLCmm70yQijZa3aCU6p1FaK3lTLB8lWVOn+jTKA4datqJZSFr7kY98JNkk\nrY1lznIlNzqBdiSayRHPqcqIbL755smmxF3XX399srWylMBgkP8aLXczXMQSTF/84heB6nh6/vnn\nAzB9+vRk67Rrvh6t7Ke69xx55JHJJv/GpR7f/va3geo9yuRRcqR99tkn2XROr7jiimT73//9X6D9\ny3w6lbgs4dprrwWq17n89vWvfz3ZYhk+05+Y9PDkk08GquO5npW32WabZPOyhPocccQRaXvHHXfs\nt1/PSJMmTUq2dj8vNZp46hpg3eFsiDHGGGOMMcYYM780Gsn9NHB+URQ3Ao/HHbVa7XtDPXi73/Bz\nxMLQIs7oKkFWJ7YdWlveYiA0axYLmCtSPnv27P+fvTsPtGs6/z/+PjUFMWdEEsQUiXmetWoIvr6m\noihKFTVVi9Kqtob2q1q0qmpWSrV+5nmsmOdQY0xJJARJzMTUnt8f8Vn72bk79557c+6Z7uf1j2Pt\nO+w8d+1prWc/K7WpcEK993dmGmlmRLOOcYZMs2FjxoxJbSo+02gxrfesrcQR3VVXXRWA4cOHp7Yn\nn3wSgKeeeiq11buA28w0SkzjkiFHHnkkkGUZAFxwwQUAvPTSS6mtkY6tIvWKqTIJDjvssNS28sor\nA/mlws477zyg9YvPVIMKd40cOTK1aYYsLi3y73//G6j/8RgaVTwAACAASURBVNSo4pKKmqFVf4Xs\n2r7vvvumtmYshlZLytIAWH756fNJsf9df/31AFx44YW13bEmpHv4mDGkmfJ4vfne974HtOZyQdWm\nYofHHntsm23xvkiz5400I17pQ+5JwCBgHDB/aPdVwMzMzMzMzBpGpQ+5uwLLlsvlSd25M2ZmZmZm\nZmazotKH3FeBxpl/7kax8JQKVRQVo3IqU8cUt969s1WnlO4di3kp3cEx7dgiiyyS+y9k6WFvvvlm\nanOBj/bp2IYsffHjjz9ObTfccAOQL0jT6Km19aLU73XWWSe1aX3cJ554IrUpXVnFKWzmll12WQC+\n//3vpzb1v1NOOSW1vfjii4DPnTMTU+hPP/10IL/GuPrnqaeemtoa9bWEetNxfuaZZ6a2AQMGAPlj\nWqm3WsPZZk7X8ZNPPjm16b5p3LhxqW333XcHfA2qxI9+9CMA+vfvn9p0frz88stT28UXX1zbHWtC\nSvO+7777cv8PWV9cd911U1u8X2oUlT7kXgJcVyqVzqDtO7n1XenXzMzMzMzM7CuVPuQe9NV/fz1D\nexlYqnq7U3+x8NTAgQMBeP/991ObZ3IrVzSTq/hqiRvwrGNHYiZBv379AOjVq1dq00zue++9l9o8\n4tu+OBO+5JJLAvllWB555BHABVMqoWJo22+/fWqbMmUKAL/97W9T28SJEwGfO2cmFu/RDG48d6r4\nTCyS5FnH9sWZ8EGDBgH5LCIVR/JyQR1bbrnlANhll13abLvmmmvS59tuu61m+9SsdE1XxlBcLkjZ\nbltvvXVqc/ZLx3Rv9Ktf/QrIF5dUlttee+2V2nwd6thxxx0HwODBg9ts+9Of/gTAY489VtN96qyK\nHnLL5fKS3b0jZmZmZmZmZrOq0nVyzczMzMzMzBreTGdyS6XS8+VyedhXnycwk+WCyuVy23nsJhZT\nQ5U2EtNAnR7WeTEtRGkjU6dOTW1OV25ffNlfabaxyNTzzz8P5FOanIpTTLEcMWJEm7Ybb7wxtSm+\nTvsuFs+TK6ywAgDDhg1LbbfffjsAo0aNSm0+ztun9TEBttpqKwDGjx+f2o4++mgg/6qHFevTpw+Q\nFaGB7JwYC85oTVybOb1idN111+X+H+Dtt98G4JBDDkltPmd2TK92xPXuRf3zueeeq+k+NaN4HVJx\npLnmmgvIX2/WXnttwPfvlVDaN8AxxxwDZKnf8Vnohz/8YW13rIvaS1feL3zeo7t3xMzMzMzMzGxW\nzfQht1wu3xc+j5rZ17UyFVSJy41o5Ci+1O5Zs2KKkWbEAT7//HMAvviiR6xIVRVxJlcFKt54443U\n9u677wKeKauERnlVRAWy0cmnn346tamfWjGdGyErkBJH1TXrE5dlsmI6puPI+IILLgjAueeem9pi\nYTQrpnOlCp5piRvICp+pMA141nFm4v2NincttdT0GqMxZr/85S+B/AyPFYvFDnVc65wZC58deuih\ntd2xJhYzNdQ/izI2XnvttdruWBPSufPee+9NbSqGqBnweN/ULM897aUrH1/JDyiXy8dVb3fMzMzM\nzMzMuq69dOVB4XMvYEfgUWA8MBhYC7iy+3bNzMzMzMzMrHPaS1f+rj6XSqXLgW+Xy+UrQ9sOwLe6\nd/dqR+k58803X2pbbLHFgKy4AmTpjjHd1mmixVSgIq4Bp7aY6hBTo6ytmBo6//zzA/n0JqWDN0v6\nSK3F/rXwwgsD2ZqZkKXaTZ48ObU5lsUUS60hDrDqqqsC2XrNAK+88grgOLZHsVxppZUA2HDDDdM2\nvYJw6aWXpjYXTenYKqusAsA222wD5K/Nf/7zn4H8cW7FVLgL4NhjjwWy/vrqq6+mbRdccEFtd6wJ\nKSX5rLPOSm0LLLAAkB3Tu+66a9rmtdk7tvjiiwPw05/+NLUpzpMmTQJgv/32a/uNlhPvjZQmv+SS\n2YqxejXh+OOnJ/bGZ6FmUekSQiOBa2Zouw7Yqrq7Y2ZmZmZmZtZ17aUrRy8DBwF/DG0HAq9UfY/q\nRC9dDx06NLXpRfZ33nkntWlJhzjaptEOz1rkR4Y066gZcchi1atXr9Sm4jQxfi4IUpxdoJi60Efl\nYkEkjQBrRhdg3LhxgIuhVUKxXGaZZVKb+mQs3OVlbjqmrKBddtkFyOIIcMcddwAwduzY2u9Yk4nX\nEi15oeNbxzbAeeedB/g63R4d3yeddFJqW2ihhYAsYygWRopFJa2YsgtGjhyZ2tQH77//fgBuueWW\n2u9Yk1ERJIBzzjkHyN8bTZs2DYCNNtoIcIZlJWJGmzI24v2SMrLi+aDZVPqQ+z3g6lKpdBTwOrAY\n8CWwQ3ftmJmZmZmZmVlnVfSQWy6XR5dKpWWAdYBFgUnAg+Vy2VMfZmZmZmZm1jAqncnlqwfaezv8\nwiallJwNNtggtQ0ZMgTIv2ytNKjYVlQ4qaemRMU1XZdddtncfwGmTJkCZOtAQpauHNNLFL+eGkfI\n0nNiSonSbd9///3Upv4XYy8xfj01lip2BjBixAggn66slBynN3VMqaFrr712m7b4Wodj2bF+/foB\nsO666wL5wlK33nor4HTQSigdFGCttdYCslcPLrvssrQtnjOtmF5D2HLLLVObrhujR48GslR6m7mY\nQn/yyScD+evQ1KlTAfjOd74D9NxrcyV0f7PFFluktnXWWQfInzNVBE3Xc5s5pXnHddjnnXdeIL+2\n/Te/+U2guV8frLTwlJmZmZmZmVnDq3gmtxXFma8VVlgByGZ6ICsM8vnnn6c2fS4a2fBoXFYaH2C1\n1VYD8iXJNfI255xzpjaN1Dl+eVp6SUuMQJZxEIvUaAROy45AcTG0nhrnWJxCx3nv3r1Tm2Z94ki7\nii94RnI69Z2+ffsCsPzyy6dtOo+q8Ef8esuL1xzNOvbv3x/IF+uKy7RYMc2W7bBDVhpE50JlFVx7\n7bVpWzPPRnQn3ecAHHjggUD++qJ+edpppwEu0NceHd+xyNSKK64I5PvfbbfdBsDrr79ew71rHvH6\nocKlcbkg9dmYnXHKKafUaO+aU1yK8uCDDwbyGVm6L3z44YdTWyv0T8/kmpmZmZmZWcvwQ66ZmZmZ\nmZm1jB6drhzTExdddFEgn/7wwgsvAPDcc8+lNqWExpQdp0Fl6SWLLLJIalNhlbim65gxYwD44IMP\nUpti6SJJ+TQdpX7HmCpNJ67THNPpZ9RT4whZyrHSQSFLt41xUdpjTKHX98ZjuyfHUkXQhg8fDuSL\nocV19UT9NKYw9+T4STyWt956ayA7zuM50deUYjHdW68Wbb755qlNx/CECROArMCPtaVjes0110xt\nKjgV743efPNNAJ544oka7l3ziH1ShSEPOOCA1KYU+nidvuaaa4CefZwXvdKitvg60U477QTA0KFD\n23z9Sy+9lD5PmjSp2rvYdBS/GFsdy7p2A2y77bZA/p5H9+Hnn39+amuF/umZXDMzMzMzM2sZPXom\nN5YfV3l8jQBDNgP51ltvpTYVtHBBmmJaIgjg5ptvBvIjmK+99hqQn93VbI9nevKUVXDXXXeltief\nfBKAF198MbUp5jG7wP0zG82M2Rm33347kJ+peOqpp4B8n1T83Cen06yP+tiDDz6YtmnJgdgWMw0s\ni19cukr9U+fEu+++O23zrESeYhWXZtFyN7Fg1xtvvAHAPffcA+SPacvPOmq2LBaR07Vas7eQXccn\nT55ci11saHGGrKhPatnJSNkEjz76aGobNWpUd+1i09C1NWYC6TypApuQ9dN4Hz5x4kQAjjjiiNTm\ngmjFMVVWVYypri+xTfeZV199dbfvZy15JtfMzMzMzMxahh9yzczMzMzMrGWUujsdr1QqNUW+n6b3\n4zS/YtMNxWceL5fLa3T1mxs5pjGdJ6ZGSdH6rY5pMfXFGEfFN6YjF8W0Spo6poqVUqAgS1OO8dPn\n2NaN58WmjKn6olKf4pp7SnGMRaa6sU8WafiYFqU2quCU+mQsPPXRRx8BdX3toKFiqvjFc6HWvx44\ncGBqmzHdVqn00BCvHtQ9pvH+Rv0upixqTdLPPvsstY0dOxaATz75BGiIOEZ1i6liGa8v6pOx2KG+\nbvz48alNx3eDxVLqFlMd3/H6olc8YjEqpc7HwnINXiSpbjHVcR7jp5jGFG+dM9srZNpgKoqpZ3LN\nzMzMzMysZdRiJncyML7DL+xZhpTL5b5d/WbHtJBjWn2OafU5ptXnmFafY1p9jmn1OabV55hWn2Na\nfRXFtNsfcs3MzMzMzMxqxenKZmZmZmZm1jL8kGtmZmZmZmYtww+5ZmZmZmZm1jL8kGtmZmZmZmYt\nww+5ZmZmZmZm1jL8kGtmZmZmZmYtww+5ZmZmZmZm1jL8kGtmZmZmZmYtww+5ZmZmZmZm1jL8kGtm\nZmZmZmYtww+5ZmZmZmZm1jL8kGtmZmZmZmYtY/bu/gWlUqnc3b+jCU0pl8t9u/rNjmkhx7T6HNPq\nc0yrzzGtPse0+hzT6nNMq88xrT7HtPoqiqlncutjfL13oAU5ptXnmFafY1p9jmn1OabV55hWn2Na\nfY5p9Tmm1VdRTP2Qa2ZmZmZmZi3DD7lmZmZmZmbWMvyQa2ZmZmZmZi3DD7lmZmZmZmbWMvyQa2Zm\nZmZmZi3DD7lmZmZmZmbWMvyQa2ZmZmZmZi3DD7lmZmZmZmbWMvyQa2ZmZmZmZi3DD7lmZmZmZmbW\nMmav9w6YmZmZmfVEpVKpTVu5XK7DnjSvr33ta4X/jf773/+mz//5z38Ax7k9s802GwBzzjknAL16\n9Wqz7dNPP01tn3zyCZCPc715JtfMzMzMzMxahh9yzczMzMzMrGV0OV25VCrNAdxaLpe/UY0dKUox\nUApHTCfQNHgjTYc3KqUTFMU0xs8xrVxRTKWon8Y2p8VkYnpWUSyLKH6OabEY06JzZ0ffI0Vx7i6V\n7mcz0b8p9mt9jufYGePclRj0tDTHGNO55poLgNlnz25jFN+iVMT24tKTr4exDykdceGFF05tvXv3\nBmCOOeZo871ffvllm7b33nsPgHfffTe1ff7550Br982ZUUyXXXbZ1Lbuuuvm2uaZZ560bdq0aQB8\n+OGHqe36668H4KmnnkptX3zxRTftcePTsb/JJpuktgMPPBCAESNGADD33HOnbTofqG8CHH/88QBc\nffXVbb6uJ9J5dNNNN01tipH6qeIexXTl4447DoAzzzwztdX7mJ+VmdyvARtXa0fMzMzMzMzMZlW7\nM7mlUunVdjbPcqpzHBkcMmQIAKuttlpqm2+++dp8j0ayXnjhhdT28ccfA9UZMSgaGY80khxHj4tG\n7jXKVqvRNu1DHGlZYoklAFhxxRVT27zzzpv7eoDnnnsOyMdUo4izMpqt31EU09im2dHYH/S3jDH9\n7LPPgNrFVL87jgguvvjiACy11FKpTSPd+ncAjBs3DoCXXnoptX3wwQdAfrSwkvgWzdDF3zXj/kIW\ny9hPNeoejxPFtGhEvjtov+PI9QILLABAnz59UptiGr9O8Zs0aVJqe+edd4BspgCy+LYX26Lshhir\noj6p4gvx76dCCzF++lyrmSDtd9H+x/OBPuscAFmcY5/46KOPgOzfBll828tQKDrOY/z0OfZd/Vz9\nTshGhotmPWulveyCoiyEWJCjb9++QH7mRn276FjWf9W/IDvHxZkH9fXYpmMitimWsZ/WezQdOj9r\nXzTDuNJKK6W2nXbaCchmbgAWXHBBIDsG43GpuOh+AeD9998H4PHHH09td955JwDjx49PbbU6P9aS\n4tu/f//UdtBBBwGw4447praBAwfmvi/GQrOO+i/A22+/DcAll1yS2i6//HIg66+tLp4P9t57bwCO\nPvro1KaYFmXW6diP17Rtt90WgEMOOSS1PfDAA22+t5XF68Zmm20GwFlnnZXaZoxppHNOv379UtsJ\nJ5wAwMMPP5zaJkyYUMU9bi5LL700kJ+FXXLJJYH2n4videvggw8G4KKLLkpt8dpeDx2lKy8MHAGM\nLdg2J3BD1ffIzMzMzMzMrIs6esh9AphWLpfvnHFDqVSaC2h/2tPMzMzMzMyshjp6yD0e+Hgm2z4H\nvt6lX/pVWt2aa66Z2k488UQAll9++dRWlK48ceJEAC644ILUdt111wHw1ltvpbYZ0zDjdLvST+Pv\nWm+99YB8ep3SQeJ0+/DhwwHYYIMNUttiiy0GwCuvvJLarrnmGgAeeuihNvvUHbTfG220UWo74ogj\nABg2bFhqm3/++dt87xtvvAHAP/7xj9R20003AVm8IUtJUjpNTB9ZaKGFAFhnnXVS26qrrgrk026e\neOKJ3M+CLKYxvU/pPi+++GJqu/nmmwF4+umnU1t3pi6rn3z961k332effYB82pz+7bGPqS/eeuut\nqe3uu+8GslRmyPqW+mlMc1L6iIpUAAwaNCj3fQBjxozJ/QzIUtVjaurkyZMBeOaZZ1Lbgw8+CGR9\nALo3/Un7s/rqq6e2kSNHAvm+o39nTIVRCv2zzz6b2pRqH/up+pv6p/4+AEOHDgXyKXj69yodFLJY\nxfOB+kNMY1Sq08svv9zme7vzeI+UhqxUTcjOSfGYWnnllQFYbrnlUptiE49RxTkeo/o76PwRY6rf\nH9NAFYPYT9U/Y6GKxx57DMhSRAFef/11IB+/eqUrF6UXxz6ha1RMmVVqYezjild89UExnfG/kJ3X\nYp/UMRrT9V977TUA7rrrrtR23333Afl082ZSFGelzSkdDrJrXez3+rsVpSurPxelca+wwgqpTf0u\nnhNbMV1Z92FF52KdfyHrlzpHxONSsdRrD5C9fqJ0csj6Z09JV46v2ej+RnGJFMv4uoFiGq/dul5t\nvfXWqe2RRx4B8ufunqLolSC16dwZr0dFr5Woz8ZXz3pyurJe3Zg6dWpqU79Tn4zHvq77Ra90xb7b\n0OnK5XL57na2lYFR1d4hMzMzMzMzs67q8hJCAKVSaXC5XH6ts9+n0RTNNEI2WxVHs4uKf2hEd489\n9khtmpmIo4SacdDoWRyt0ehEHH3UKK9GxyAb1YmzCCoiEGeh9e+JIyAqnx8LWnTnzI7+vfvvv39q\nW3/99XP7B8UxVcx32GGH1KaYxlEYzVpoeYHBgwenbYsssgiQL3SjmP773/9Obfobxa/73//9XyD7\n20I2CxdnxzVjGWfyupP+nXFEWrO6McugqAiUYr7FFlukNs2qxRkW9UEVRFDRmvg74s/X6GSc4Z5x\nfyGbRY8jyuqfsW306NFAxwXXqkWzLmuttVZq0+yBZp8hf2yK4hYLLClucdZRv0PFfopmHePMzJQp\nUwAYOzYrPaBzRJzR0Lnk1Vezenw6PuJMcq1nHXX8xllCzeTGGSrNKKjIH2SjsDF+RUWrFFPFO/59\nipZdUFziiK6KthX93EcffTS1xRm0eiv6WxYVxCpaxiYe52qL/W7GUe8YR2WCqG9CFrd4nGgGOR6/\nsZBKI6n0uNDXxXholiGe+3XNifcMuu7qGC0qUBmXHdl44+kLROh4gSyr68orr6xof5uV+mJclkbZ\nXLp+QHbNHjVqVJuv1znnmGOOSW0bbrghkD93FmWQtTL1V8gK+ShDBbLzxR133AHkrym69px99tmp\nTRlI8R6pEYrJ1VI8H/zrX/8CYJdddkltuh9VtmC8H9f9bCyIpPugomVxeqI333wTyDKRIDuG9SwU\nn2F07vzLX/6S2pRV0EhLMXW5QvJX7+QWFaQyMzMzMzMzq4uOlhDaqJ3NHv4wMzMzMzOzhtJRuvLd\nwCSgqpVoitKulCJXlE4b0zJUiCOmzOjnrbHGGqlt0UUXBbJp8/hzlaYT0+KuuOIKID/1rt+ltCjI\n1pGLKZ/a95iSo2n99tZbrCbFIBbUKSqeIjGdQIVMVIQIsrSDuIae0kH0c1WgBrJiMjFl/IYbpq8w\nFdO+lMYTi37ttttuQHGKakzB1c+uVZqO/q4xhbi9mMYCEEonVloNZP/2mMKp1FHFL6bM6vfH4jP3\n3HMPkKXkxH3ZbrvtUptSxorSxGLfVUplrWLa3vrHMX7ar/hvv/fee4F8OqbSOmMKqVJglTIef67S\nxGI6mYr3xBgoft/5zndSm9KeBwwYkNrUZ+PvqPW6hTqfxZRjpRfFVHetXxkLRSi+MZVO54P48/T3\n0n87Knai1O6tttoqtaloWyzAopT8onWO65mOV7QecNGalkqPja+5qLBc0RrPsRjKjGupF8U0niuU\n1l+03mZcY1rHU7OnMxalb59//vmp7eKLLwbyMdV5VDGNMVDfVbE6yApIxuJVOr5rtR57vSg2sZDZ\nn//859w2yPplUfEtnV9UQA6yV3Ti3y8WVuoJ4r/9pZdeAuD3v/99m+1F1wqdU+LP0Hkyng9asRha\npXSOi+dd3RcUnfeefPJJIH/t0/W8WQv0VZviFgv4xs8zUiHVGNOi18fqraOH3PHA7uVy+YEZN5RK\npV7MvPKymZmZmZmZWc119JD7GLAG0OYhl+mzu50uOgXZKEwsNqLlUuIMmUZl4+itlhqKBZE0Yh5n\n3DQKVjTLoRGLOIKj0eCiUSC9kA35Uc8Zf14cWdMIU61GibT/Wj4CsmJbsUiSZhTiqKKWY4oj4prx\nikWr9O8s+vcWFVbRCHDRbEgsIFJUYlw/L44AFy2V050UKxWHgOxF/FjMSLNgv/nNb1KbZnBjv9O/\nvb3Z/aIX9mP8imaYNJIWZ901axa/TsddLFql+NZq9lEzqJqVhax4ScyYUMEYLRsFWZzjDMuMfbJI\n0baimfg4Iqm/UZwd18+JS+Bon+LMcK2LLmh/4u9Vv9MsNWR/47j/RYUiivpCJbOCMabKLInF6RTT\n+PdTYTktOQTZ8V3PmciiftVeEar49y/KOCk6bisRC31o1j1e59Rn47JqrbikiPpELMQllcZUfTxm\nIBUVTNQx3YpxLBKP/XgPUAn1axW1g+x8Hq/rRfdNPYXiW+l1QTGNRQNVXC3GsdkzNaqhs/ctscCc\n7nF7ct+cFbr2xCyYoqWG6q2jh9zdZrahXC5/Diw5s+1mZmZmZmZmtdbROrmt/VKKmZmZmZmZtZRZ\nWie3q5QSc9ZZZ6U2pbbEwknHHXcckF8rtShFQ+l3emF/Zl/XVUUpETHlT2l6sZCVilvUKnVRKXJK\nPYYs7SqmK1944YVAfl+LYqV0g86mL3VEvyumLKp4RUztVkrZ/fffn9qU0ljr1NqrrroqtakvxtRM\nFYOKqdW1TCVSPIrW6437pBT6GNNap5WoP8X0ShVEiumBekUg7l81Y9rR+qdKCY2FPtQ/Y+GrotcS\nap1GplTOuP9FKf1FxU66O6YbbZQV6Fdak/7eALfddhuQT/etdeGuIrOypms1xfOMUudjMT4d3zGt\nv5HWKKy2avTXuEa3CnbFPqfzvdNBO6b+ueWWW6Y2XYfiKwjxfsnap2KRsUCf4nz11VfXZZ+andYT\nj68lFBVes8rtt99+QP6+s7uvh11Rm9K/ZmZmZmZmZjVQl5lcUSEhgFNPPRXIz4jEmdlKdNfIqwop\nQFZ4KM48aHbq+eefT23a91qPBmv5CoDLLrsMyBc/KiryVEsakVxuueVSm+IblzHRTHgsTlbr0eCi\nkuoqmBRjWu9Rai1jE5dr0Ux5XGro2muvBfKz+LUecSsq+jZx4kQgP7tfy+U71CdV/h5ghx12APJZ\nEFoKQscVZH2jEUYui2ZoOyqcVE1x1lHH90477ZTaVHjqoYceSm1atizO4nsGLROXq1KfjCPnjz/+\nOJBfesyKqf/FJZhU0CcufReXJbH2qX/G7AIZNWpU+uxjunKHHnookD+fKn7um11z7rnntmnTNbve\n92/N6phjjmnTplg20vHumVwzMzMzMzNrGRXN5JZKpQWAQ4FVgd5xW7lc3ryzv7RoKQktDdBIC7Br\nP9dff/3UVrS0jRaajsvy1Hr5AY36xRkljU430lIIegdqzz33TG3az/gOzzXXXAPALbfcktpq/a5e\nUUw1A1nvmGr2FmCXXXYB8u+ajR8/Hsi/f6vP8Z3IeokjfTrm67W4vd6BOvLII1ObYhmP80svvRTI\nsgwgy4xohPdIo0qWVqq2OOt98sknA/nlbtQnTz/99NQ2depUoPHiV2+69vz4xz9ObZoti1kQxx9/\nPNBYSzY0Kr0TPmzYsDbbYsZQXGLI2rfFFlsA+WXxdM4544wz6rJPzSjO2u64445ttuv47mx2Y0+m\nJYIAhg4d2ma7njl87alcfE9cS7dGehZqJJWmK18BzAZcDUzr4GvNzMzMzMzM6qLSh9x1gD5frY1r\nZmZmZmZm1pAqfci9D1ge+Hc1fqnSA2KKxozb6qVXr17p82qrrQbANtts0+brYjGiyy+/HIAXXnih\nm/du5oriprZ6vQSuAimrrrpqalOa8tJLL53alD4biypoWZ5Y4KvWivqpUmprGdP4+xdZZBEAfvWr\nX6W29dZbD8gXw3rggQcAuPPOO1Ob4lyvtGAoTqOtRz+NSwMdccQRAOy+++6pTSmhN954Y2q74YYb\ngHzKWCMUnFLcigqV1IKO83322Se16RWP+PrJmWeeCcAzzzyT2hohfo1Ix3nsk/qb3n333anNhWgq\nN3LkSCArNgVZ//zZz36W2hqpaEojiueZ3Xbbrc12pYG++uqrNdunZhdTa1XcNLr99tuB+t8fN5P4\nSldcOkiKCidZ+2Lf1Hkgni+/+93v1nyfOlLpQ+7ewE2lUulh4K24oVwuH1/tnTIzMzMzMzPrikof\nck8CBgHjgPlDe5eGPItGSusxoh+LJay44ooA/O53v0ttmu2JM18qknTHHXekNhWtqGfxD8W03iN9\nccTssMMOy/0XstlGLccC8NRTTwFw/fXXpzYVoarnv6feC1trpGzxxRdPbRdeeCGQLzKlkXPN3kIW\nyzia3ghF3erdTzWL88Mf/jC1qX/GGZ6bb74ZgLPOOiu1vfnmm0B9Z8LbU68ZqBEjRgBw1FFHpTYt\nCxbPk3/9618BF0mamThDdsghhwBZsSTIMghiMapG6iVQ2QAAIABJREFUOKYbWbzGH3DAAUD+OHn6\n6aeBbCkm61hcwmr48OFA/pyoApyNep5sRPHao8/x2D788MNrvk/NLs6OS7z2/OMf/6jl7rSEmC2o\n82hcgqkRszcqfcjdFVi2XC5P6s6dMTMzMzMzM5sVla6T+yrgIWMzMzMzMzNraJXO5F4CXFcqlc6g\n7Tu5d3X1l9crvU7FpeL6t3/84x+B/IvVH3/8MZCt7whZIR+t4wrZWo+NULCiXvuw0EILAfC9730v\ntSl9Ma6npYI+Y8aMSW2K5YsvvpjaGikNr15FppZbbjkgW3MUYO211wbyKdRam+ySSy5JbUrDi6kk\njdA/pZb7ElMWVfgsptArtfbBBx9MbVp/dMKECamt3q8CNJL4WsLZZ58NwMCBA1PbxIkTATj00ENT\nWz2LyDWDuM7w/vvvD+T7rtZpjudJa5+uSwArr7wykE+jPeigg9q0WfsGDx6cPqtAWrzO/OUvf6n5\nPjU79U3Izq3xfDlu3Lha71LTW3PNNdNnpS7HgrEx9dYqE5+ZdK/6xhtvpLZGvEeq9CH3oK/+++sZ\n2svAUtXbHTMzMzMzM7Ouq+ght1wuL9ndO1IpjR4ULZdRNDukIgkDBgxIbdtuuy0AO++8c2rr168f\nkB/dUaEZFaGBrHjK66+/ntp62jIYml1YaqlsfEMzD3G5JcU+zoZde+21AJx//vmpTaOUn3+eLcPc\nSLOOtaD+HJdb+vWvp48pqSgaZIXPtJwNwGmnnQbAyy+/nNo8M5HFdMMNN0xtWipEs7cAo0aNAmDf\nffdNba+99hrQ8/phpeK5U7MQ8dypwkmxwJy178gjj0yfNQOpAnwAv/jFL4DGHC1vVDELRhlcMYtI\nRSOtcroHgmyG7LHHHkttztjovIsvvjh91v3Vvffem9p62j1mNVxwwQXps+4FbrvttnrtTkv4wx/+\nkD6rn8aCsY2o0ndyzczMzMzMzBreTGdyS6XS8+VyedhXnycwk+WCyuXy4KJ2MzMzMzMzs1prL115\nv/B5j+7ekfbE1GSlwMY1sDRtHlM6tNaYUj332WeftK1///5AVgQJsjSGuM6TikzF9UdjSm2riPFt\nr00xXWONNYD8upiLLbYYkK3lCHD77bcD8M9//jO1jR49GsiKevV06s+rrLIKkF+XVYV8YtEJpYvc\ndNNNqe2jjz7q7t1sSkqnP++881Jbnz59ALjnnntSm9KU9XqCzZxe6zjhhBNSm/pwXHfwxhtvBJzu\nXYn555++9HxMl9e17PTTT09t8dxq7dPrCDvuuGNqU5r3iSeemNqcBlo5pXvHgj46vmNMrXLqp0OG\nDEltimk89q1yiqmuVdE555xT691pCTr2F1hggdSmfnrZZZfVZZ8qNdOH3HK5fF/4PKo2u2NmZmZm\nZmbWde2lKx9fyQ8ol8vHVW93OjbHHHMAMO+886a2hRdeGMhGGyAr3DFy5EggXwxBS4U888wzqU2z\nZVOmTEltPWWUV7O2cbmKopGbpZdeGsiKTMWlge6//34ArrzyytSmYhSetZ1OcY79dNlllwWyIlNx\n9PH5558H4Nhjj01tjz/+ONBz+malYuaBZmvPPfdcIMsygGz5FRVGAs/gVkJZHMokiEutqbDc0Ucf\nndoaaQmwRqU+e8QRRwDZciyQLcug5Zmsc3bbbTcgf4165513gHzRPqvcdtttB+Sz6JRFFDNjrHL/\n8z//A+TvvT777DMgu9Zb52yxxRZA/p5AhTjjPb9VbrPNNgPyMdU9aKMXl2wvXXlQ+NwL2BF4FBgP\nDAbWAq4s+D4zMzMzMzOzumgvXfm7+lwqlS4Hvl0ul68MbTsA3+re3TMzMzMzMzOrXEXr5AIjgd1n\naLsOuLC6u9MxFTlRSiJka4vG1KQFF1wQyNJplKIMWRro22+/ndp68rqiSkHo3bt3ahsxYgQAq622\nWmpTwQml1D7yyCNpm14+V2zBKbWQT0NSfNdaa63UduCBBwIwdOhQIFufFbI1HuMahF4js1jsu8cc\ncwyQFfPS2sIAJ510EgBjx46t4d41p5iatPHGGwOw+eabA/l0ZKUwK8XWKqPXbPbcc08gH28V8Yp9\n1zqmVFq9jhBjqpTaWHDSOqYYHnbYYW226fUPx7Rr9ttvvzZtemXOr3l1zV577dWmTf3TMe2anXba\nCcgXklQR3kYvfFrpOrkvAwfN0HYg8Ep1d8fMzMzMzMys6yqdyf0ecHWpVDoKeB1YDPgS2KG7diyK\no7Eq2qOZL8hmFuebb77U9umnnwLw9NNPA/DUU0+lbSpA4Vmx6TT6HWOqGZs466iZchXp+n//7/+l\nbS+88ALg2dsZqWAPwEorrQTAwQcfnNq0xJVGxWI59nvvvRdwP22P+q6KogDssMP009Jcc80FZEuB\nQVZ0xkvbdExL2wD85Cc/AbKCfjFj48ILpyf0OKYdi9cyLWs3YMAAIF8c8cwzzwQc087S8na6lqmI\nD8Cpp54KOKadpSUXV1hhBSB/PdKSd45p5+g+VvcEMabXXnttmzbrmLI8ldkZ++QTTzwB+P60s3S9\nUlZcjKmy4Ro9phU95JbL5dGlUmkZYB1gUWAS8GC5XHYJTTMzMzMzM2sYlc7k8tUD7b3duC9mZmZm\nZmZms6Tih9x60FS51saFLIUupoHqpfJY+OTVV18F4JVXpr82HAt4OA0kS+2AbC1crdkKMGTIECBL\nowUYM2YMANdccw0Azz77bNrWkwt3FVF849qX3/jGNwAYOHBganvvvfcAeOihh4D8OsMx9paJKZ+L\nL744ADvuuGNqm3vuuYFs/dvTTz89bdNrDDZziq+KTQEMGzYMgGnTpgFw3nnnpW0ujlQ5FUQE2HXX\nXYEs3o8++mjaNmnSpNruWBOL17LDDz8cyF5jmDhxYtrmNTK7RoV8dM8Vj/dbbrmlLvvU7FZffXUg\nu1bpvArw17/+tS771OyUVq9nhHj/dO6559Zln5qdXlHUK6HxPv/vf/97XfapsyotPGVmZmZmZmbW\n8BpuJjfO0mg0VjONkBXpiDO5KiQVR23Hjx8PwIcffgj07NnbGFMtaTPvvPOmtiWWWAKApZZaKrXp\nZfIJEyaktrvuugvIXuKPRT16MsU3LhekImgqLAXZ7HicTVTGwSWXXALA1KlTu3dnm5jiHPvuBhts\nAGQzupDNNKjgVCw65wIpHdNI+DbbbJPadL59/fXXAbjpppvStp58bq2U+u4mm2yS2gYNGgRkMw63\n3XZb2tboxTwayWKLLZY+r7322rltsUCar1eVU2EkyJYPUR9+66230jZncVSuqOicshTffffdtE3n\nWOucb33rW0DWd+NMrpa6ss5Zd911geyeIM7kxkzORuaZXDMzMzMzM2sZfsg1MzMzMzOzltFw6cqR\n0pX10jNkqUkxnUtFOmKxDqUpO+0rnyaj9JhFF100ta288soALLzwwqlNBZG0FhZkhTs+/vjj7tvZ\nJhFjqsIncZ1mFepZc801U5vS7uN6mKNHjwbg5ZdfBpz6KUUp4CqCoLUFIVvPOb7SoBQ6rTPsNMWO\naU1hgA033BDIUsHjdhX3i33YOta7d28A9thjj9SmojPqn05T7BydI1TQD7JzxBdfTF/d0GmKXaMi\nPpD1XaV/xtc/XHCycvEVO603rqKpjz32WNqme1frWLwPU0x1f/r000+nbXotzDpH51M9D8Tz6X33\n3VeXfeosz+SamZmZmZlZy2i4mdw4MqNRwrfffju1aVTxo48+Sm0qhBBHwLz8SjHNFMb4aVZGSwQB\nTJ48GciP3GimwbPj+X4qyjyAbFZXMwqQFUOLGQejRo0C8n+PnirGtGj5MI3UDh48OLWpyERcPuyl\nl14CssIInh3PK8pCiDPhw4cPB/LLWujYv//++wFnc1QixllLB8Vr2bhx4wB44YUXAHjkkUdqt3Mt\nQPGNSwPdfPPNQJbNccYZZ6Rtvm5VLhaX+tGPfgRks7s33nhj2ub7rMrFWH3/+98Hsow6LXUJPrd2\nRiwkedJJJwHwpz/9CchnG8X7MKvc9ddfD2RFEWMfbpb7Ks/kmpmZmZmZWcvwQ66ZmZmZmZm1jFJ3\nrxtZKpW6/AuU/hnXbFMBlJgKppf3Y4GZBk9NerxcLq/R1W+uRkxV9ASyNbBim9JnY8qH4tugaQp1\ni6lSPmNhCcVUKbZRjKnWx1MaSIPFtm4xLSo8pfOAUj8B+vbt2+brlA6uNYdjqlIDrJNbt5gWUdyK\n0sK1Jjlk5wPFVudccEwrofjG84HadD6IMW2A80DDx1Ri39X1Tdd/H/uzLp5boSH6ZtSUMW1wjmn1\nOabVV1FMPZNrZmZmZmZmLaMWM7mTgfHd+kuaz5Byudy3q9/smBZyTKvPMa0+x7T6HNPqc0yrzzGt\nPse0+hzT6nNMq6+imHb7Q66ZmZmZmZlZrThd2czMzMzMzFqGH3LNzMzMzMysZfgh18zMzMzMzFqG\nH3LNzMzMzMysZfgh18zMzMzMzFqGH3LNzMzMzMysZfgh18zMzMzMzFqGH3LNzMzMzMysZfgh18zM\nzMzMzFqGH3LNzMzMzMysZfgh18zMzMzMzFrG7N39C0qlUrm7f0cTmlIul/t29Zsd00KOafU5ptXn\nmFafY1p9jmn1OabV55hWn2NafY5p9VUUU8/k1sf4eu9AC3JMq88xrT7HtPoc0+pzTKvPMa0+x7T6\nHNPqc0yrr6KY+iHXzMzMzMzMWoYfcs3MzMzMzKxl+CHXzMzMzMzMWoYfcs3MzMzMzKxl+CHXzMzM\nzMzMWoYfcs3MzMzMzKxl+CHXzMzMzMzMWoYfcs3MzMzMzKxlzF7vHZjR176WPXfPOeecAMw111yp\nbdq0aQB8/vnntd2xFqH4zjbbbKntP//5DwD//e9/67JPraJUKqXP5XK5jntiZmZmZtZzeSbXzMzM\nzMzMWkZdZ3LjzNeiiy4KwO67757a+vfvD2QzjQD33nsvAKNGjUptH374IeDZsxnNPffcAKy44oqp\nbZ555gHg008/TW0TJkwA4O23305tX3zxRS12seloJnz++edv0/bll1+mts8++wzIZxy4f7YvZhfo\n3BBjps/OODAzMzOz9ngm18zMzMzMzFqGH3LNzMzMzMysZdQ1XfnCCy9MnzfbbDMgnwaqwlPR3nvv\nDcD111+f2q6++moAxo4dm9o+/vhjIEtl/uijj9I2pZC2Yvrofvvtlz5/61vfAmDxxRdPbfPNNx+Q\nT/l88cUXAbj00ktT26OPPgrA1KlTU5viVpSK28rFq1ZYYYX0ed111wWy9HqAhRZaCMinK48ZMwaA\nhx9+OLW9+eabQFY8LX6P4hbjp8+t2E/VDyGL3+yzZ6cjfY6vNKjf6ZiGLJYxvb4oltKKsZSY7q0U\n+qJ/b4xp/CxF8dPPaeX4VUtRTKW74tfe7+zO39uKFMui1yd6yvm52hS/OeaYI7Xpc7xu6jzeivcR\n1aaYxvtkXVfj9VD3wvG1P/fZYoppLHbbt29fIB+/d955B8juScAxnRnFtFevXqlt8ODBuW0AEydO\nBOCTTz5JbdU4D3Q4k1sqlXYulUp/KJVK3y+VSnPMsO3Ps7wHZmZmZmZmZlXS7kxuqVQ6AjgYuBY4\nADiwVCptVS6XJ331JXsAP+jsLz3wwAMB2GWXXVKbRk46GpFeZJFFANhmm21S2/Dhw4GsqBJAnz59\ncj93ypQpadtFF10EwCWXXJLaJk2a/k+KozXNZOuttwbgxBNPTG0LLrggkJ8hK5rhGThwIABLL710\nanvttdcAmHfeeVObYqoRGY1mAdx0000AXH755alNs5lxZKaZ4qt+ddppp6W25ZdfHshmHyEbSY19\nV6OnL7/8cmobN24ckB/RUuw1AhtnKe+77z4gn7Xw9NNPA/D++++ntmbKTNCo6MiRI1PbaqutBsCw\nYcNSm+LSu3fv1KbYxJgqeyOOXC+88MJAlhUS+5/ip0yF+DM++OCD1KYR2thfG3V2Qf0pzo7ruFUs\nIDt3DhkyJLXpe2J2wVtvvQXAe++9l9oUh6Lzh743xk9/q/h30efYpr4bZ3MaIc5xKbv2thXNpug6\nFM+dEkf+FVP9t+jaVzTrHs8f+l3xHK/YxywcHQONENuuKJpdbe9vpP4Z46evj38rxVLXNsiOj5it\no0KNo0ePTm06b8S/abPGtz2K4Yz/hSymcYZWfbJfv36pbbnllgNg1VVXTW3a/tBDD6W2m2++Gcj3\n3XhuaFVFMY3HtGKq6ydkGWabbrppaltllVUAeOmll1LbH//4RwCef/751NZM9wzVoPgWxTQe+7oH\n0f00wMYbbwzk++Qpp5wCwJ133pnadI7tKTGVeE5Wsdt4f6xjf4cddkhtW221FZC/N/v9738PwD//\n+c/UpgzcWYlpR+nKBwKbl8vlFwFKpdKvgPtKpdI3yuXyeKD9J1IzMzMzMzOzGuroIbcvkKZNyuXy\nL0ql0mTg3lKptBnQs4YszMzMzMzMrKF19JA7HlgJeFIN5XL5T6VS6RPgbmCumXxfu376058C+Ze7\nO0pTFqUDqVgSwL///W8gnwozaNAgIEsZi6ljhx12GABrrLFGarvsssuAfPqBUkKbIf3ghBNOAPLp\niUojqDQNLhaSUptSHCFLlVG6V0xJ2GuvvQBYe+21U9sjjzwCwFVXXZXaHn/8cSC/Tm+jOvzwwwFY\nffXVU5vSO2N6VhH1t5jypvRZpZHHNv28mO42dOhQANZff/3UpvjdddddqU1pzXGd40ZN8dpkk00A\n2GOPPVKbUoRi2lBMyRT9mxZbbLHUttZaa7X5esW+qHCdUmuVOg7w5JPTT2+PPfZYalOcX3jhhdT2\n7rvvAo2Xcq/jUGlBACuttBKQ77sjRowA8mmESt9Sej1kMYrnPcVU/TXGVunHsf8pviomAVka9HPP\nPZfatO65tsXfW8/zrlIGY0qsjlGlZEF2LMdU+zXXXBPIrkGQT5MT9Vml5Mdzra6N8e+idLmYHquv\ni+fuf/3rXwBcd911qU3bY6p4I13Xiq5RRWlw8djX53jeUAqiYqtzKMAyyywDwAILLJDaFJf4u/R3\niNe3119/HYA//zkrQ6I+W1T0rlkpDvF+SdewAQMGALDEEkukber38RUInRviOVl/KxWcieLvUupy\n0asSjdRfZ5XOJbqnisUt9ZrUkksumdrU/+M987LLLpv7esjOJfE+UIVHYwpzPF+0Gp1r9coTZPf6\nui5C/vVG0d8h3nPpPKDCoZD153gfqHNYK/XTGcWYrbjiigCss846qU39L16jdN+hFGXI+r3uqSC7\nL4nXymrEtKPCU38FvjljY7lcvgA4Ani9y7/ZzMzMzMzMrMrancktl8u/a2fbpcClM9venrhMUCXi\n6Oi5554LwM9//vPUppHUpZZaKrWpuJVeyo8/Qy87x1HFAw44AMiPKl555ZVAfjS9UcXRbGlvdjzG\nQ6Onxx57bGrTLLZmJQB23HFHICu+FIvVTJ48GcjPIK633npAflRR3/PUU0+ltkabGRONTlda7CT+\nOzTyH2exNfIfC3xphFGj5HEETLNgscCX+qyWh4JsVDEWUtOsT6ONKmr/4yi/RqmLZrtiP1Vho1h4\nSv0uzq7pd6jfxT6p2Ub9fSAbffz617+e2vR3iYXUNHsesxAaIb46Z8UYqD/pv5Cdd+NsgPpbLACh\n/hZH+zXyqr9b0d8lFvfTsaDZBsiW39IIMGTFe+IscCPQuTMe7+qfse9q1i8e0/ocZ8wVjxjTGYuD\nxSJrMy7XBlm/i+dTxTceO/p7xIJ1jdBPK1VUJEaxVOYGZDNY8dqnWVr9NxbqUb+PRfveeOMNIF+0\nTTMP8XvVD+LfT3+bZp+9jXFWf47nws033xzIZnTjPZLiEovOqcBRPE40ix6zmHTfFmdzdO6J5+xm\n6rvtiefn7bbbDsgymmJf0/2min8C3H///W1+nu5Piv4eui5Clo3U6svdaMZ1//33B+Ab3/hG2qa4\nxGVGb7jhBiDfd/UzijLJYkwfeOABIP9s0OzngSI6F2sG/Ic//GHapvPB+PHjU5syYuP1XLPd22+/\nfZufH2N66623AvniqzVZQqg9pVKpbe6JmZmZmZmZWZ10+SG3VCrNBYzt8AvNzMzMzMzMaqSjdXI3\namdzl4pOQTZFHdMsYkqoKKXimGOOSW2/+930DOqiaWytgQlw1FFHAVkaUkxp0hq7u+++e2pTMYWY\nNqIXzWMxqkZNrVXKQFGKVUxbVtzOOuus1HbkkUcC+XQWefbZZ9NnrV+l9IP4N1P89t1339T2zW9O\nf507FguRI444In2O6481kmeeeQbIUq0g+zfHwjvqE9dcc01qU/+LaRvqzzE9TGkxReuPqu/GlPE9\n99wTyBcUUspiLJJ022235fatUSjV+JVXXkltKmbQv3//Nl+v4mWQFX2JhYuUZhv7uPqnYhpT3/R3\ni0UptCZeXMdN54tYAEUpSo1WNE0pU1rrG7J1gGOqu1KXY4qVYhn/Hjo/x9TMGYvYFRWpi0UpVl55\nZQB+8INsGXWdB2IKsFJIGy19TvsTj5+i9FSlXGo9P8gKIUZKyYypcTP+jhhT/Y4YF/VJFfmDLPU7\nno+U7hXTvhq9eE/RfsXjVv1YBeEAXn31VSD/99BnvRYT/y46buNrNjoXx0IzOp/GtHDdC8S/rX5e\nK6Up6jiPKd1aG/iWW24B8vcEOlfEc4r+lip4FD/HNqWKx3UxG7W436zQ+S5e35R2r5RkrWULWR+P\n5wrFI55jlVoeiynpHBULpE2YMAFo3GKUsyLeS+m+XsfvhRdemLbpHj5eDxXT+DN076AU/dgW/0a6\n14pF51qRjteddtoJgAcffDBtU7p3UcHTeI3X+XHXXXdNbUoL1/McZPfb1e6nHVVXvhuYBLTOWdzM\nzMzMzMxaViVLCO1eLpcfmHFDqVTqBXSpItMdd9wB5GdSi0b0//73vwP5p/32Rk3j6J9GwTSaHYsb\naCQ3jqyp6EccKTvooIOA/NIicSSokWgWMRbTUpGHGJcLLrgAgKOPPjq1tTdyEkeq4ugu5Edr9Dku\nz6CZh1hmXQUt4pIbmiFrtBFxLcERZ/1U3CzGTAWf4shhe7N9MaZxVmFG6rtxuRsVYIkz9hoVi8s3\nqUhSo42Ia7me2Hc0mxiPfc0exJlczRbMSj/RrGOc4dHnuNyTzgfxeGqv6Fg96biMmRgqAhVnvtTv\n4gxB0exkV5fwiZkdOuZjMSUVXonHjvaz6PfXk46bmCGgtnj86jiP1xct1VE0wxj/bZ39d+rvFpdb\n0ixEzDjQEkKxjzfaubU9RbPo+rfHf5O+rr2+01GMdUyrWBJk5+S4ZIjO8bHISqvMjMU4a2b27rvv\nTm26lqjfV9qXYmaJPmv2HbJ7kaJzVCtRH4yF+U499VQgmwUryqIrEgsEqnhizGy66aabgGyWrTM/\nuxnF87OKc1199dVA/lzRnqLrUbzn1z2IiipBz4mprm+nnHIKkL93aE88R+jvcM8996S2MWPGAPC3\nv/0ttXXXslYd3bU9Bqwxk23/BV6byTYzMzMzMzOzmutoJne3mW0ol8ufA0vObLuZmZmZmZlZrXW0\nTm635I5oijqueae0tpjScfjhhwNdS7WaMU0pvlyu9fXiWlhKtYspiUqpjWmMSgtrtPSva6+9FsgX\nIVDhjLgW1YknnghUP9VKRcRUgAqyFPSYxqi1OmMxKsW80WKqtXz/8Y9/pDYV74npiTfeeCNQ/YJE\n6sNKR4asL8Y195Q6E+Pc3hrJ9aQUQBVGgix1JaarKBWr2mvSFv2MVVZZBcgXmtHXxTTQ7kqnmVWK\nUTymlXIZ/73aHtMTq5kaHH/WoEGDAFhuueVSm87Bcf1H9YdGO/aLUrZ1TMW2ov3urnRrpYBvtdVW\nqU3n2JiyqOIgzZ76GeOoPttdr19ssMEG6bPOA/EcdcUVVwCNV3SuGmKc1Weq0Xd0DgAYOnQokE9Z\nvPjii4HWjGmk+MZUY71609lzRVzPWQUpY0zPPvvsNr+rlcXzgV7r6uy1JBbtUwG6WGDp/PPPB3pO\nTCM9j83Kq0trrDE9GTieT6+66iqg/df1qqUxXzIzMzMzMzMz64KO0pW7hYoP6AVxyEYONXsGbQsd\ndUXRiLxGv4tmNOLXaeRcpdohm42II52NUChFy/DEmGo0JS4PUumL452lGKnYFRQvYaS4xREc7Wd3\nzTB1lUbuVAQJsr4TZ/ji8gnVpLhoJAyyWd0YU83kakYUsvjFr2uEmOpvHGfCVWArzpRqdL/a+6x4\nxEyCjTfeGMhne2ifYgEWxbkR4hhpf+JMrkazi4ofdZeYGXPAAQcA+UJ0Ok40igvZ+ajRYlqkHvsY\nj98tttgCgHXWWSe16Ti56KKLUltXR997Ii2REQsx6hx1zjnnpDYVCHJMO6asrpNPPrlNW+ynumfp\nKTGdlfOv7kV/+tOfpjbNlP/sZz9LbfG62hPMSkx1f7r33nuntlVXXRWAX//616mtGs8hzWRWjkdd\nr7bccsvUpozdP/3pT6mt0qJg1eCZXDMzMzMzM2sZdZnJ1QyZljqA7N2sWLa/mjMPmoGDbFSx6B3G\n+Ds1s6RlW6B4GYxGGInUDKkWqYZsNioue1TNmMb3l1dYYQUg/16jYhpnaDV6q32b8ec0Ev1d4+y3\n/u5xJrq7ZsgUy6J38GJM9f6qFtOO2xuhb0ZFs476XIvsCI2Ix+XLtERUjKmWFInnqGZaMqRoJre7\n6DiPNRY23XTTNr9fS4Vcf/31qa2ZYloPWoYNstmF+A7ZnXfeCcDNN9+c2hzT9sX4nXvuuUA2owtw\n6aWXAjBq1KjU1mhLsTWaeA1X3Y+4pJ3eFX3iiSdSW6O9h99oYhbHXnvtBcDWW2+d2jQzFu/5Gu16\n38g222wzAHbbLauve+aZZwL5pa4c08qpJlCM6VlnnQXklxSrpYoeckul0gLAocCqQO+4rVwub94N\n+2VmZmZmZmbWaZXO5F4BzAZcDXR/OSwzMzN5Qm7EAAAgAElEQVQzMzOzLqj0IXcdoM9Xa+POMr10\nHNMrlRpc7ZLSSqMZMmRIaltxxRWBfNqSxLQkFZuIL54rxabRUhgUN6WuQpb+qYI51RbL2e+8885A\ntkRQFH+/yoiPHz8+tSm9rtFiqvjFwlLdvZRFLH6k1KS4LJRSmOKL+5dffjkAY8eObbOfjUbHT1G6\nf3elr8W0r2WWWQbIF5vQawuxmNipp54KwBtvvJHaGjW9rt7Hjc4Dxx9/fGrTKyFxSbhf/OIXQL44\nSr33vVGpT/7kJz9JbbqGxZj+/Oc/B2pbyKPZ6VoFWdG5uMzeSSedBNRmeYtWEV9V0KsgMaZnnHEG\n0PzLWtVSLI540EEHATBhwoTUdskllwCNe61vRLEw6sEHHwzk0721FGejXusbUVzOcp999gHgscce\nS21ajqle1/pKX4a8D1i+O3fEzMzMzMzMbFZVOpO7N3BTqVR6GHgrbiiXy8cXfkc7NIsTiw8VLeFT\nDRpl2G+//VKbSq/HYgkauYkj4irq8eyzz6Y2je422gyERvPigtXdNeusWYZNNtkktW2wwQZAfiZS\n+xSLif3tb38D8jNkjVooRfErKojUXX//WOTs2GOPBWCeeeZps08xC0KFUoqOp0ZTNGvb3fsai879\n4Q9/AGDAgAGpTf00Fpm65ZZbgPyyRo2uln/zeO7cf//9gfxSV/r7XnHFFant4YcfBjzzUAllG333\nu99Nbfr7qjgKZOeBRj3eG4kK+akwEmT9WLO3AOPGjavpfjUznVtPP/301KYMubjcTZzVtfapTyrz\nBbLMmBhTZ29UTtlcKuAF2bNBPB90V9ZjK1ORScgKe15wwQWprd7X+0ofck8CBgHjgJiP6iurmZmZ\nmZmZNYxKH3J3BZYtl8v1qQFtZmZmZmZmVoFKH3JfBapeMaA7p7GVNrv55tNXONpjjz3abIuUkhpf\nmL7hhhuAfKpNo6bWSi1SA/r27QtkqZ+QpdPEtLlPP/0UyFKUIYtvTAtp9FS7WhQhUJpHTPvS2o2x\ncJKK9px88smpTcXGmqlYQi3+5orb9ttvn9o22mgjIH8OmDhxIgC/+tWvUltcG9na6t+/f/p81FFH\nAfkCFC+++CKQre0KTgXriM4BkK3fGtcdV9G+eN6tdypYo4vnzv/7v/8D8kUo9YrChRdemNqa6Txa\nbyo0s8oqq6S26667Dsjun6Dxr/GNREW8ttpqq9SmIlOjR4+uyz41OxXx+va3v53atHZzLOZllVOR\n2V122SW1nX/++UC+WG+9VfqQewlwXalUOoO27+TeVfW9MjMzMzMzM+uCSh9yD/rqv7+eob0MLFW9\n3Zk1cdS2T58+APzyl78E8rMM+ro4Cq7ZnFjUQwWnmqn4THfSTMMhhxwC5Iv3SIzpI488AuRfQtcI\nj0d2p1NfHD58OABbbrllm22x8NWVV14JwF13ZWNLXpahmLILYmEJ9eF4TP/+978H4Pnnn09t7p/F\nVBQlLhe00EILAfmid4cddhgAb72VGxO1dowcOTJ9VuGpWFxGs2axwJy1b/DgwemzZhziLMOee+4J\nOMugM3r37p0+qxBSXH7tgAMOAJxl0Bkxs0jL18V+qvOtr0uVi88Dxx13HJDvpzHD0DrvO9/5DpCP\n6b333luv3Zmpih5yy+Xykt29I2ZmZmZmZmazqtJ1cs3MzMzMzMwa3kxnckul0vPlcnnYV58nMJPl\ngsrl8uCi9lpSWkJcT3TnnXcGYOmll27z9Ur5iKlgKvhzxx13pDYVTurJ4nqYSy01PTNdaQpaLxey\nYh1TpkxJbT/72c+ArDBS/DqbrlevXgCccMIJQD6tXrGK6zbq62JqqOWpz+6+++5AVsALshS6WGBO\nRX6cXtcxnQO222671KZ+ev3116c2rTHu9LqOKa3+t7/9bWpTH77oootS23PPPVfT/WpmujbFV2V0\nfxBT7eO1ydqn+6zf/OY3qU2F0VR8DuCdd96p7Y61gFgQacklpydO6pUP8PW+K1ZbbbX0WcW8Dj74\n4NTm17w6b9FFF02f9WrdT37yk9TWiPdQ7aUr7xc+7zHTrzIzMzMzMzNrEDN9yC2Xy/eFz6Nqsztd\no1FbFesAOOig6bWy5pprrjZfrxEcLR8A2Yi5Z2/z5ptvvvT5mGOOAaBfv35tvk6FfM4555zUptky\nz97mxRlwjYZtsMEGQL5YwscffwzAz3/+89Q2aZKXqu6IRhs1whiXZtEsQxwlV5xt5nQe/eMf/wjA\nggsumLapT8bZHI+Sd0yztUcffTSQzZIDvPHGG0CWuQE+j3aGlg3TeRWyWdszzjgjtTnToHLLLbcc\nALvuumtqU8HOv/71r3XZp2an8+iPf/zj1KZ+GjNjrHJzzjknkL9vevvtt4GsGKp1ju5ZjzzyyNSm\n7AItF9io2ktXPn5m26JyuXxc9XbHzMzMzMzMrOvaS1ceFD73AnYEHgXGA4OBtYAru2/XzMzMzMzM\nzDqnvXTl7+pzqVS6HPh2uVy+MrTtAHyre3evMlpjLKbRKGVR6Z8x1Wv8+PEAHHHEEanNKYt5itsq\nq6yS2jbddFMgS12IL5k/+eSTAJx22mmprRFfQm8EMdVTabMqOPXll1+mbSqCdt1116U2p9cViynJ\nSqkZOHAgkE+dveKKKwAYPXp0DfeuOcXUeaXVr7feekD+2D7vvPMAmDBhQg33rvlpDde99toLyB/b\nZ599NgCTJ0+u/Y41MRXxUtHDuP6oCszFdR2tY7reK6Zxndw//OEPQL6Ip3VM59a9994byBf00T3U\ntGnTar5frWCzzTYDYOWVV05tZ555JpC9Vmedo1jGtdx1Po33rI2o0iWERgLXzNB2HbBVdXfHzMzM\nzMzMrOvaS1eOXgYOAv4Y2g4EXqn6HnXBgAEDgGy2AbKlWSSWYD/llFOA/NIslqdCM9///vdTm5YL\n0Cjk+++/n7ZplPeDDz6o1S42HRWa2XbbbVPb8OHDc9veeuuttE0x/eyzz2q1i01LcYRseRsVoFBx\nFMiWv3CWQccWWGCB9FlZL5rFee2119K2v/zlL4CzDCoRZxaVxaFZnDgTrgJ+jmnHYsbB1ltvDcCI\nESOArIAXwFlnnVXbHWsRQ4cOBbL7q5hd4Jh2jc6t++67L5C/b9IMmXWO7llVADHO2sal2Kxyul79\n4he/APLn2osvvrgu+9RZlT7kfg+4ulQqHQW8DiwGfAns0F07ZmZmZmZmZtZZFT3klsvl0aVSaRlg\nHWBRYBLwYLlc9joRZmZmZmZm1jAqncnlqwfae7txXzpF6Z0AK620EpBPrxO9FB3Xcrr55psBpyy2\nRyngK6ywQmpTzFXI5/HHH0/bnnjiCcDpde2ZZ555gPzL+0qrL1q7eezYsTXcu+akglM77rhjalNh\nL6Ur3XjjjWlbTF+0YkpJ2mSTTVLbsssuC2Tn02uuyUo0TJ06tXY71+SGDBmSPm+xxRZAduz//e9/\nT9vefffd2u5YE4uFkFTES/30kksuSdtccKpyMa1+9913B7LzQoyp+2nlYqrnxhtvDGQFJy+99NK0\nTWu5W+cMGzYMgPnnnx/I91Nfo7pm8cUXB6BPnz5APkV5ypQpddmnzqq08JSZmZmZmZlZw6t4JrfR\nxMJSmslVqXvIRsc//fRTAO6+++60zaM6xeLo7RprrAHAQgstlNo0Oq5CSHfeeWfaFgt7WSZmHGg2\nTP+FbGkrLWF16623pm1x6Rsr1r9/fwDWXXfdNts0cxNnHZ290THNjMUCaSripZmbq666Km1zTDum\nc+s222yT2hZZZBEgm7m59tpr07a45J0V08zY2muvndo0m6NZhthPHdPKxYyDzTffHMjum668Mq0k\n6Zh2Qsw03HnnnYGsn2ppO3BMO0Mz4ZDFVMU7Y2aMMwwrF5dj1PVKGXBxdrxZYuqZXDMzMzMzM2sZ\nfsg1MzMzMzOzltF06cpKUYppoC+//DIATz31VGrr168fAGPGjAHgvPPOS9vi+lmWxVTrjEGWsjBp\n0qTUpjQarS98yy23pG1OWSymNE/I1huM6UgfffQRkMU09mErFmO66qqrAtnxDlmat9bHjWu6WrH4\nqsLKK68MwOqrr57aZlzHOa6VacVioZmllloKgF122SW1qRCd+qdfo+kcFUM59NBDU5sKJo4fPx5o\nnuIojUJ98gc/+EFqW3rppYGsn7755pu137EmpnPr9ttvn9pWW201IIupCyJ2jq5Ha621Vmpbb731\nAHj11VcB99PO0vVK96kA66+/PgAvvfQSkF3/m4lncs3MzMzMzKxlNN1Mrl52VqEegKuvvhrIF+3R\nTKQKIsWvb5YXpmtF8Zg2bVpq0yzts88+m9pUkEYzZHH00cUS8jQqFme4teTS6aefntpUfOaRRx4B\nshEzcExnVJTFoUJIsbCcZnVvv/12IOuv4GN/RkVZHH379gXg9ddfT23qx3fccQcAb7/9dq12seko\nprE44oorrgjkzweaDX/ggQcAL8dSiZjFoUwDzehCNhv+4IMPAvD+++/XcO+aUyzYufzyywPZ7C1k\nMR01ahQAH3zwQQ33rjnFa9TAgQMBGDFiRGrTLKPus+L9qRWLmTFaJigucansDS0ZqAKpVhldr5ZY\nYonUpvvR2267DWjOYqieyTUzMzMzM7OW4YdcMzMzMzMzaxml7k7fK5VKNcsPjOkM0qDpiY+Xy+U1\nuvrNtYhpUWqoYjnjfxtEQ8Y09knFMhb5UQy1BnGDpSg3ZExjn9RrCXG9PMVQ6feKLTREn22omBYd\n5yo+o1R6yNJslboYX21wTNv8PCCfBjrffPMBsNhii6U29Uu99qEidNAQ54GGiqnEmGrd0cGDB6c2\nxU2pix9++GGbbXXUkDGNx776qVJs4/YJEyYA+X7qY3+mPzd91rUpptXrHkBpyz6fdo5er1HaMmQx\n1brjDZau3PAxVfx0/Y9tOo82WLpyRTH1TK6ZmZmZmZm1jFrM5E4GxnfrL2k+Q8rlct+ufrNjWsgx\nrT7HtPoc0+pzTKvPMa0+x7T6HNPqc0yrzzGtvopi2u0PuWZmZmZmZma14nRlMzMzMzMzaxl+yDUz\nMzMzM7OW4YdcMzMzMzMzaxl+yDUzMzMzM7OW4YdcMzMzMzMzaxl+yDUzMzMzM7OW4YdcMzMzMzMz\naxl+yDUzMzMzM7OW4YdcMzMzMzMzaxl+yDUzMzMzM7OW4YdcMzMzMzMzaxl+yDUzMzMzM7OWMXt3\n/4JSqVTu7t/RhKaUy+W+Xf1mx7SQY1p9jmn1OabV55hWn2NafY5p9Tmm1eeYVp9jWn0VxdQzufUx\nvt470IIc0+pzTKvPMa0+x7T6HNPqc0yrzzGtPse0+hzT6qsopn7INTMzMzMzs5bhh1wzMzMzMzNr\nGX7INTMzMzMzs5bhh1wzMzMzMzNrGX7INTMzMzMzs5bhh1wzMzMzMzNrGX7INTMzMzMzs5bhh1wz\nMzMzMzNrGbPXewcaRalUAuBrX/tam7boP//5DwDlcrk2O9ZiFFPHz8zMzMzMuoNncs3MzMzMzKxl\n9JiZ3Nlnz/6pc8wxR+6/APPPPz8ACy+8cJvvmTx5cmp79913AZg2bVpq++9//wv0nNlJzcbGmMYZ\ncNH2GOfZZpsNgM8//zy1ffbZZwB8+eWXqU0x7SmKMgnUn2K/ai/jIH5dUZ/sKf2zGoqyOBw/MzOz\n6vG1tvoc04xncs3MzMzMzKxl+CHXzMzMzMzMWka76cqlUulrwKHA0sDZwFvAWcBSwB3Az8rl8ucz\n/wn1p5TZPn36pLbll18egKFDh6a2AQMGAPk0WaUmjx8/PrWNGTMGgDfffDO1xdTlVqYUiLnnnhuA\nQYMGpW1LLLEEAPPNN19qm2eeeQD44osvUtunn34KwNSpU1Pb2LFjgXxauFKYWz3FQunbvXv3BmCR\nRRZJ2xS/aM455wTy6cr6rNhC1j/fe++91Ka/QyvGNMZjxjR5xTiKMdD3xq+LKfby8ccfA/k4t2Ih\nOsWh6BWEohT6GbdBdq6IaVP6ubFN8dN/oflfVSj6tyuWRa8USKV9qCgVraP0tEbqn0X7Oivf06j/\nzloquh5E7cWlvT7ZkxW94hYVnQuLzmc9tU8W0f0LwLzzzgvkj23FquieMb7OZplevXqlz3379gWK\n73k++uij9Pn9998H8nFuVR29k3sKsArwX6Y/1P4F+CcwB3A08J+v/mtmZmZmZmZWdx095O4CjABm\nAyYDF5fL5VcASqXSE8ANNPhD7lxzzQXAsGHDUttKK60E5GfNNMIURzsGDhwIZLOKAJMmTQLgrbfe\n6qY9blwaHdKs7be//e20TW1x1FKjRHE2USOir732WmrT6LFGlyArTNWKo6Bx5FKFzr75zW8CsOWW\nW6ZtmhWPRbr0+ZNPPklt+rtMmDAhtd1zzz0APPnkk6ntgw8+AForpoqlZsIhO9ZXWWUVAJZccsm0\nTaPzcQRTI8RxpFjng5ix8fDDDwNZNgdk54s4ct9MFL84U7HQQgsB+UwNZcLo62K840iyaPQ9zgzp\ndylDBuCVV14B8nFW367nrFLRzKvE47eoCJ8yMPr165faYrxE/U19J/4MHdOxX+nYj79f3xO/7p13\n3gHycdY1rJ4x1b7GWQbFOc7w6PP/Z+88AyWrqrT91IiSaXK0u0lNlBwFZEAER3KQAZHxAwdERBgV\nZRzHUWREURFRVBiiZMnQNrmVZmhoQpOj5NQ0WSQHx/P9aN591rl336q63XXrnrq8z5+uXqdu1a51\n9t7nnL3evdaoUaOSbf755+/3Pikr9C+U4zr3O2WLvtLrXJsiUmvlvms451Pd38SEmfLbRz/60WRT\nhEf/QvlbHn744WSTai3+zmZzm45Ff+t1K5/Kf/H+qg7ROkUYP/axjyXbKqusAsD666+fbLo/jL/z\n3nvvBWDChAnJputFVPw1U8Tk1EF6Hb8rp26Q7+P1rQ7ReY3fnXbaKdk++clPAuV1Gsq+G+95rrvu\nOgCOOeaYZHvwwQf7va8vrRQKOt6qbKjeV7fovMbyt79dPoLpPnLMmDHJpjkz+kr3h4ccckiySVGZ\nG++55KbN+nDuGhnR33Sjb7Z6yJ23KIpXABqNxmt6wAUoiuK+RqOxyMB/aowxxhhjjDHGdJdWiade\nbDQa873/+svxwPsPuG/0/xNjjDHGGGOMMWZ4aBXJPQZYCHi1KIqz+xzbHrhsSFrVARQil/RuqaWW\nSseef/55oJQfArz44otAVQ6y7LLLAlU5nmR4UWJTB+lCN5AfNt54YwBWW221dOyZZ54B4MYbb0w2\nSZ+iHEQSoOhT+S/KKeogsRkqoixxjTXWAGC//fYDqtLaxx9/HICbb7452R599NF+nycpVZSiqe5z\nlH+NxH4qXyqZHMBXv/pVADbaaCOg2tc0zmMyOcm8o0xHUr8o2dUWhShX7nWfamxqngTYdNNNAdh+\n++2Tbdy4cUAp5VP/gnKujdsSnn766X42SSslp4VS1qfzEj+vDuSkbNEmKVhMbLjeeusB8IlPfCLZ\ndP2R/6Dsl5I3x3Og61CUHKufRp/qfdqKAHDFFVcAcP311ydbTlrb7b6r8x+l2/KbtrtAuZ1orbXW\nSjZJQ6Pvc/XVlRRRssecTDz2P20/iv7TPBCTI15++eVAdS5+7bXXgOq1qts+lV+22mqrZNt8882B\nqgx0scUWA6rbEnS9nTZtWrLpdbwWSwot+X30qX577JNKICn/QCnVjd+lsX/rrbcmm7Z/DOf1X/c3\nUcqpex3JbqF6HRdbbLEFAFtvvXWyScIc+6muKzp/uSSTrXyqbR2aawH+8Ic/AFWf6n3Dea3aZ599\nAPjmN7+ZbOpX0Y+5uV++WnXVVZNNctv4t7q3HDt2LFC97mtMx7Gva06U5mtOif30wgsvBOC2225L\nNj0HDKdPf/rTnwKw++67J5vm2JwfYx9T/4zX8YsuugioXqN0L6Drvz4fymtKvHarn8bnI41lzbUA\n55xzDgC33HJLssW/6SRNH3KLoji6ybGTgJM63iJjjDHGGGOMMWYmaRXJbUqj0RhTFMWTrd/ZfbTi\nq1XHuAKrqGOuFEiM5GplIUbXtBLUq4lmZgWt8Gj17LjjjkvHtHIYV8S1cplb/Vl77bWTTStBH4R0\n5lDtY4q+Sl1w9tmlYGLixIlAtbSSfBRX4NSPY1IMrQLHFflejzrmkC9jFExj/4EHHgDg2muvTccU\n3YqJjuS/GLX9whe+AFTLjImR5NNc4inNmXE8yl/67THxnubTuPqtKNjyyy+fbDvssANQjdpNnToV\nqG8JoVxZpEjf6wyUEa9c1Ccmn+mbJCmuZCtiIzUHlD6P/XSdddYBqnOKEohMnjy53+8Yzv4qH8Tk\nUbkkZ1JQxcRJ8m+Mhul17Kf6bCXti9EcRbRyEYOYqEVtie+T79VfI8PpU/lIkRYoI16KEkIZ4Y59\nWP6LfUf+i+dI/VllFuMxfUacPxQxip+re4d4f6B7qTvvvDPZ6qDi2GSTTYDq3JVLfNYu6rtxjlAf\nl5IhRiQ1H8T5Q+cy9jX5PEZ3NRbuueeeZNOcM5z9dJtttgGqvylXlimH2h3HvvpknDfU7/UdUfWh\nz5CiIX5e9Iv6bIzuqp33339/stWhxKVUA/H+ut3xo3bHBKbyZbw31/VFfTdX7inO07mkafJpvPZp\nfr7vvvuSbaiSzbbakzsgjUZjduCxDrbFGGOMMcYYY4yZJZpGchuNxqZNDs/e5JgxxhhjjDHGGNN1\nWsmVJwHTgfrox9pEkhlJYmISiWZS4yiVyyX1GIm1RpsR5QmSGFx99dVAVUrXrL5dlJPJp1GCKynY\nB0UCHmU0kin/13/9F1DWDYXmPo3nRZ8Xk9QosdJIl4Crz8Raj8ceeywADz30EFBNjKDxHcev/Jer\nnxklNjo3I0murN8Z+84f//hHAG644YZkkzxL74t+yflUsr6YPGW33XYDSukklDLouibyi23J1QSV\nH2Ld7yuvvBKoJjaUJD7WYddvVn9q5QPNv7vuumuyKeFQnA90jcol/xhONJ9FiZyk7nFMKUFJ3Gag\n47nrRq5muMj5IEpDtRVp//33TzYlsMklQqzb2JdUVdcRKBOUxd8p2bak7AAXX3wxAFOmTEk2zZXx\nGiX550ILLQTkr0tRhiq5484775xskpfG86M21a3+qO4Zc/V7+/YvqPr+6KNnpLEZP358skn6GhP6\nrLDCCkCZMDF3PZc8HMo+qaRsUM6jsU2xNned0LyXS3yXk9jGefLQQw8FymRFUPaZmFB2yy23BEqJ\nbW7Oi/7RFoUoYZb0N/pUc0Qu0dhwMth7u/j+b3zjGwCcddZZyabfHOtDf+lLXwLKfhfHvs5BrGeu\nOSL2dX1unFNWXHFFoLqdJFeLtxO0OmtPAJ8viuKGvgcajcYcuISQMcYYY4wxxpga0eohdyqwLtDv\nIZcZ0d1aJZ3KrT4ONkoYN3FvuOGGQJkCHkZ+ZKwv0R9akVEZm2aRxkhMXf6pT30KqKa4j5v8PwjE\nhAuKQKoMQLv9NCaeUgKCmMBDK/J1WBkfSrRaG1fT+yaWa+UDHY/RCK1cxnJBUi6MJMWBfnuMzMp/\nEfm53YigIl6xnyqxRVSAKEFKu3NJt2j2O3PRiDiHSUURf2cu2j3Ysal+FyMKiu7GMiIqdVG36Hiu\n5I/GaIzQKsFLLslUPC85n7ZDjBzpvMUklPJvVDfccccdld/Qty3Dhc57jHprLoz3Q1Kh3HXXXcnW\niRJI8mWMfD344INAGVmLbYn+U2LA6Ps6+FQJ22LUVGXBItdccw0AP/zhD5NN191Wv0O/XSV/IvJp\nruTLkUcemWyKoMXv0txTt3562WUzqo3Gslax5KHQ9faf/umfkk1qn1zfjGVp4j0lVMe5+meM2iqa\nqfJG8W/id0kFWjcVh8p1xsRPuWRef/3rX4FqhDYmiezLpEmT0mvNK/JL9KmisPGc/uhHPwJggw02\nSLac+kHzbjdKXLZ6yN1joANFUbwLLDPQcWOMMcYYY4wxptu0qpP7wQpbGmOMMcYYY4zpaeq1k3oW\niTXMFKKPEoNmKAyvGmkAn/70p4Ey0RKMLKliO0RpnCQO7Uq2JVHaaaedkm2zzTYD4NRTT022ukkV\nhwr1sfh7Jbdpt1/pfOy9997Jtu666wLwm9/8Jtmi/HQkIylWTD4zWBmjJD7Rp6rleswxxySb5pSR\nSE4GOitom8Nee+2VbJKDx/lUc0odJHUzg/pYLmlZp5E8LCb0ke2WW25JNslF6+bTXOKunK86Ie1u\nRvws1X/caKONkk3zQUxmJ5lv3a5VmpPitgpdU+J1WsmUoq0TPs31f0noVQsWynuBKEuXvLRuW8BU\nu/Pkk09ONrU1yjyVoC/WqR0suXMgW5RxK8lUrh58TNKkZHe5WtDDyYUXXgiUCYeglIBr+xvAwQcf\nDFS3H7VLX1/G/2tOiQloNc7jljwRnxs0t8Z7jDrIlX/yk58AVbnwkksuCZTjHcptA80kypF2511J\njuPck9u+IOLceffddwPVsTNUPp3pOrnGGGOMMcYYY0zdGFGR3Lj6ElfB2kGrOXHDu5ICxBIkHzRy\nEbJ20Sr5jjvumGxa0YtJWeqwKtYN9DtnZUVQyRo++9nPJptKlijhDNQvijNU5CIJg0WJG3bYYYdk\nmzp1KgATJ05MtrpFcepGTEqx2mqrAWWiOSgjY6eddlqy1S3iUDeiT7UiL+UGlMmRTjzxxGTLleuo\nA7n25MbtULc7qpOk3pByA8r5+eyzz042RYDq5lPN81G5o8hojJAOdbQ0lgL52te+BlSjjpo7//Sn\nPyWbkmbVzae6j5QiAsrkRzESGCOoQ0FM2nfAAQcA1fJrOveKPEMZIavb9V+KA5WtArj99tuBaqm1\nGOkfCmLJId2XxvlAfTEmX7zuuuuA+qk4dQ+te2qAcePGAdVyS7FsWCfRtUklmwDWWGMNoJr0Tj6N\n5QSvuuoqoDv3VI7kGmOMMcYYY4wZMdhJzUwAACAASURBVLQVyW00GqOAg4C1gHnisaIothqCds0U\ns7J6pWiOUrsDXHHFFbP8ub3OYH97jDysssoqQLW0jaI4H+So2GBXruOqmPaMx37685//HPjg7MPt\nBHHPiCK4cRX5xz/+MTCy9+F2mpgTQZGHqFpQeYGnnnoq2eoWxakbKhMC8J3vfAeozrGKNsbyGXW/\nXsVznivZMdQst9xy6fWXv/zlSjugVMRccsklyVb361U854radqPN8lssF7TNNtsA1euW9gkff/zx\nyaa5oW5zQK6smnz66quv9ntfp1Fk8aCDDko2Rchy5a+iikOR5rrNAWpPLHWmfbdxT+5Q+VQR8KOO\nOirZllhiCaDqU0XxVfIIykho3Xyq8R3nfkXyY96LofLpQgstBFR9KvVm9Kki4GoblM8E3Zij2pUr\nnwd8CLgI8J20McYYY4wxxpha0u5D7obAwu/XxjXGGGOMMcYYY2pJuw+5k4GVgLuGsC1dJ27iV4IU\nJZyBoU8sMBKRXAFgiy22AGDSpEnJpkQppn1iAo9NN90UqErpovzTtIdS7QNsvvnmAFx66aXJppIh\ndZPS1ZmYEGnDDTcEyqQdAJdffjlQvwQedURSz3/5l39JtuWXXx6Ahx56KNmOPvpooHcTeHVzfCm5\npLYiQCm5iyU3vvvd7wLVRCl1nQckoYzywMGWUJsVlAgxShbnnntuoCr3/dnPfgZU+27d5J9C7YrJ\nupTIdKjmrnj+VBJGCbygLHcT5Z1KjKp5Fdovmdlt5Ms4pjRnDVVStCiX1zyqsqDxeOyH9957L1At\nHVjXRH7qkzFBmvwrKXuniVuSjjjiCABWWGGFZMttP9E2sMMPPzzZnnvuuX7vGyrafcjdC7is0Wjc\nBDwXDxRFcVinG2WMMcYYY4wxxswM7T7kHg6MBh4H5gv2ei1ttIlWGzbbbLNk02rEmWeemWx1W7mp\nM/JpLMOiyMNvf/vbZKvr6m0d0UrjF77whWQbM2YMANdee22yOTLWPloR/8Y3vpFsiy66KADnnXde\nstV1RbyOSL1x6KGHJptKivzqV79KtqFaXR6JjB49Gqj2U3HkkUem19OmTetam3qRGCHbbrvtgGqS\nJF2PTj/99GS75ZZbKsd6gW5eA2LSPvXFWIJJfps8eXKyyb9DXcqoE+i+L14DhvpecK655kqvzzjj\nDADmnXfeft8fy8F8+9vfBqrJsOpKrozMUCsOdK8E8NOf/hQor/+R2CYlonvyySeTra7PAfJf7BPq\ns51us+ZRqTOhjI7H+UDEEq7f+ta3ALj++uuTrZvzVbsPubsDKxRFMX0oG2OMMcYYY4wxxswK7dbJ\nfRSo/xKcMcYYY4wxxpgPNO1Gck8HxjcajWPovyf3Tx1v1RAz33wzFNdRBqp6g042NXMstthiAHzl\nK19JthNOOAFwrdGZZaWVVgLg3/7t35LtuOOOA6o1XU37SG6z1157Jdspp5wCwBNPPDEcTepJogxU\nNXHj9o/zzz8fKGuOQn1lX3Uhyr5++ctfAlUZqLYoyLfQW5La4UBJkKCsJx6TI95zzz1ANSlKL0hq\nh5OVV145vf7sZz8LVPuuaqF+6UtfSrZevK/qxtjSPKo5FMptXjFxkhJ2xntW1R7uJbpRF1V98bTT\nTks23fPH65YSox144IHJputVL82rURo8VNdYyelPPfXUZMtJvzV3aq4FOPfcc4HhqzXe7kOuRuCP\n+tgLYNnONccYY4wxxhhjjJl52nrILYpimaFuyFATV3A22mgjAF588cVk++Mf/wg42jAYok/33HNP\noLpie+GFFwK9tSo23MQV8R/84AdANbJw0kknAfbpYFASJChLWcSkIkqeMlwrjb3IwgsvnF4rOVIc\n+//xH/8BOIHXYFhrrbXSaykOXnvttWSTSsYJvNonRshUNiz203333Rdwabt20LXp2GOPTTbNrTGa\npH4ak/eYPFIVHHLIIcmmCG687n//+98HyqRo4HvVgZACTmXsoLxXjQmPTjzxRKCacLIX76u60Q/+\n9V//Fahe93Pff+WVVwLwwx/+MNmGWxnT7p5cY4wxxhhjjDGm9gwYyW00GvcXRbHy+6+fYoByQUVR\njMnZjTHGGGOMMcaYbtNMrrxveL3nUDdkqBk1alR6vdNOOwFw1llnJdubb77Z9Tb1OqrlCLD99tsD\n8Itf/CLZeqF+W91Yb7310uuNN94YgP/+7/9OtiixN+0Rk3Usu+yMFAKHHXZYsvViAo/hQlK6KEda\ncMEFgWo/tVSxfWaffXagWlM4Z3vggQe627AeZpFFFgHgq1/9arJJVhfliVH+aZqj69E666yTbJJ3\nTpw4Mdkuv/xywHLaZkg++73vfQ+A+eefPx2TT++4445kO/744yvHTH8+8pGPAGVN5tlmKx9v1Bfj\ndek73/kO4G1KzZA8WVvn4hZFEe9J9957bwDeeeedLrSuPQZ8yC2KYnJ4fW13mmOMMcYYY4wxxsw8\nzeTKhw10LFIUxfc615zOo2QJ2jgN5arZ3XffnWxedWwfrZhpJQzKxBMqcwH26WBQAoqf/OQnyfbM\nM88AZQp2sE8HwxJLLAFU+6nKBCnpBNing2HttdcG4HOf+1yyPfbYYwD8+te/TjZHHFqjVfEvfvGL\nAKy77rrpmMqwxPnAPm2NojdKMKfSdlAqNv7zP/8z2WIiGpNH1yaNb6kMoIzixDIsjoy1ZrnllgPK\nsR8jZEqMFksw1SkyViei33bZZRcAVllllX7vk//233//ZOvFslbdIJYGUikglWCKKKHUv//7vydb\nHZWGzeTKo8PrOYBdgFuAJ4AxwPrABUPXNGOMMcYYY4wxZnA0kyvvrdeNRuP3wOeKorgg2HYGdh3a\n5hljjDHGGGOMMe3TVp1c4DPA5/vYxgOndLY5nWe11VYDqhKlK664AnBipJllyy23BEqpDcDJJ58M\nwF//+tdhaVOvIrmN5PSbbrppOqYkXq7hODi0RUG1BceOHZuOKWHSyy+/3P2G9TCqhyn57LzzzpuO\n/fjHPwbcTweLkiOpzrCSegGccMIJQD3lX3VmzTXXBOAzn/kMUJV4y6fPPvts9xvWY0QZ6P/7f/8P\nKCW2UY58xhlnAOU2EDMw2uYFcMQRRwClFDzK5idMmADAXXfd1cXW9SYLLLBAeq3rvbYsxLE/ZcoU\noJogzeSJcu/tttsOKOeD6NP77rsPKOeAutJundyHgQP62PYHHulsc4wxxhhjjDHGmJmn3UjuPsBF\njUbjEGAasBTwN2DnoWrYrKLVHCVEiCs+Z555JlBunDbtMddccwHw3e9+F6huUFfadifyGBwqv6Jo\nzrvvvpuOKTruhDODY5lllgFg9913B+CVV15Jx0466STAyabaIUZzPvGJT1T+jdGwU06ZIeixT1sj\nlQGUiWUUIVOyKSgTo9mnrZHKAODb3/42AAsttBBQjTAed9xxgH3aDosuumh6/eUvfxkoE05NmzYt\nHZOywz5tzfrrr59eS7Gl+eCFF15Ix5TIx9f9gembtA9gqaWWqhyL133Ntb4/HRg9MykiDqXSQLzx\nxhvptRQedX+OaushtyiK2xuNxjhgQ2BJYDowpSiKev86Y4wxxhhjjDEfKNqN5PL+A+11Q9gWY4wx\nxhhjjDFmlmj7IbfXUF0nyZTHjx+fjt10003D0qZeZ8yYMUCZMOWqq65Kx+69995haVMvEmWg6623\nHlAm8rnyyivTMdUfNa2JMtCtt966Yrv44ovTMdUeNq2JMtA99tgDKCV0559/fjrm5EjtE2WgO+64\nI1Am8on1sJ0YrX2UbApKSagkdLGf2qetUfKznXcud6JJBqp+6rE/OCTz3nfffZNt7rnnBkqf+ho1\nOOaff34Adtttt2RTYi9Jki+77LJ0zPdSrdG2Gd2TQjkf6Lp/zTXXpGO9cs/fbuIpY4wxxhhjjDGm\n9ozYSK5WIBTBffjhh9Mxl7pon1jWQtHx666boVqP0fHXXnutuw3rYWLCrtGjRwPlqlhMx/722293\nt2E9jFbGAcaNGwfAo48+CsB5552XjsXyF6Y5iuAALL/88kAZZbjkkkvSMSfzaI3m0Y022ijZFNVV\nNCxGc+zT1khpsMsuuySb5gFd4y+66KJ0zIl8WiOVVvSprlfy6dlnn52O2ad5olpLSoM49nVcJReV\nvA/s04FQYiSAXXfdFSiTTEZef/11oEw0B55PByLeNx188MFAmbQPyoRyb775JgBHH310OtYr91KO\n5BpjjDHGGGOMGTH4IdcYY4wxxhhjzIhhxMqVJa1RkoR33nknHbN0oX1i/bsHHngAgCOOOAKo1nW0\nT9sn+vTWW28FyppucWO/fdoayb6U3APK2piXXnopUPoYLAVrB8nCJPuGsi9KVi8pOLhGZjsoAaKS\nokEpt33qqacAeP7557vfsB4jJphbY401APj0pz+dbEo+ozrOTjbVmjnnnDO9VoK51VdfPdkkV371\n1VcBb/dqhrYlxARz3/jGNwBYfPHFk039WNJaJ/AaGF2PYp/8yle+ApQJO6H0vXwa6zmbqoRe157t\nttsu2bbZZhugup1OfyOf9mICL0dyjTHGGGOMMcaMGEZsJFeRB61AmJkjRmnkSyXxcqRx5oh+U3T8\nz3/+MwBvvfXWsLSp14lJuiZMmFCxPffcc8PSpl5F0TAlRYHSp/fffz8A06dP737DeoyYKEUJfd54\n441kky//+Mc/Au6nzVBEIUYdV1xxRaDqU/nw+uuvB+Cll17qVhN7ghjNUTRx4YUXTrYVVlgBqF6H\ndL267bbbgFJ1ZGYQk3NKUaRyiwALLrggUL1Gyb9SGTk6XiX6dK655gJg7NixyaZ+HPupEiHdeOON\ngFUcfYk+nWeeeYCq4kBq19hPZbv22msBeOGFF4a8nZ3GkVxjjDHGGGOMMSMGP+QaY4wxxhhjjBkx\nNIY6aUij0XBWkv7cWhTFujP7x/Zplp7xaZSM6bXGYc2S+NTep/JfTEgjm5JM1UxWX3ufypdKTgGl\n1Enypffeey8dq0GfraVPczLGWINQSFYXpXf2ab/PA6rjXElnllxyyWRTv5ScPkqZa5B0rpY+1fYE\nKKXLqt8O5TYlJfSzTwdGYz4mQlxsscWAat1xbQV5/PHHgapPPfaryKdxq4J8GeW2mkflU8+nAyOf\nxjq5GvPzzz9/sikZopLMxgS+veJTR3KNMcYYY4wxxowYuhHJfQF4Yki/pPcYWxTFIjP7x/ZpFvu0\n89inncc+7Tz2aeexTzuPfdp57NPOY592Hvu087Tl0yF/yDXGGGOMMcYYY7qF5crGGGOMMcYYY0YM\nfsg1xhhjjDHGGDNi8EOuMcYYY4wxxpgRgx9yjTHGGGOMMcaMGPyQa4wxxhhjjDFmxOCHXGOMMcYY\nY4wxIwY/5BpjjDHGGGOMGTH4IdcYY4wxxhhjzIjBD7nGGGOMMcYYY0YMfsg1xhhjjDHGGDNi8EOu\nMcYYY4wxxpgRw2xD/QWNRqMY6u/oQV4simKRmf1j+zSLfdp57NPOY592Hvu089inncc+7Tz2aeex\nTzuPfdp52vKpI7nDwxPD3YARiH3aeezTzmOfdh77tPPYp53HPu089mnnsU87j33aedryqR9yjTHG\nGGOMMcaMGPyQa4wxxhhjjDFmxOCHXGOMMcYYY4wxIwY/5BpjjDHGGGOMGTH4IdcYY4wxxhhjzIhh\nyEsIdYJGo5Fe/8M/zHguL4oyo7ZeR5tpTvRpDvvSGGOMMcYY04s4kmuMMcYYY4wxZsTgh1xjjDHG\nGGOMMSOG2smVo4x29tlnB2DRRRdNtr///e9AVU77l7/8BYC//e1vyabXOdntB02KG30622wzTvkC\nCyyQbPLV//3f/yXbG2+8AZT+7vvalEhCP8cccySbfBV92qxPmjzybSS3VcEYY4wxxhjhSK4xxhhj\njDHGmBFDbSK5itgsueSSybbeeusBMHbs2GRT5OaRRx5JtocffhgoI7oA77zzDlBG1D70oQ+lY4qu\nKVoZ39cKRUVziZviZ9QhwiSfxkj4iiuuCMAiiyySbO+++y4Ajz76aLI999xzALz55pvJJr/lfJWL\nXLbrg5wvZatb9Fjtmm+++ZJt4YUXBmDOOedMNvn0pZdeSjb1t6g46Pv7PsjKg4985CPptRQHMZIr\nP8i3UPY3R3fzxLHVLNmc/WeMMcaYkYQjucYYY4wxxhhjRgwz9ZDbaDTGNFrVoDHGGGOMMcYYY7rM\nzMqVbwTWAF6YpS+frfx6SZI/85nPJNsWW2zR728ee+wxAG655ZZke/HFFwF46623+r3/wx/+cL/v\nki1KRfvKmyNR6qzPyb3vvffeS6+jbLeb6LcBLLfccgBssskmyfbxj38cqLZPMuVnnnkm2SRZjGsZ\nOV8KfZ78CM0TLbWqfZxjuKTLSoAGMHr0aABWWmmlZIuvxdNPPw3A3XffnWySgEe5rZA/csnTcvLm\n6KvcelMzn9ZBmhr76fzzzw/AvPPOm2ySg8exJ7n3K6+8kmyS08ex19dHueRprXww2DW8Okh8o6/k\n39w4a5XMS2O5mU8H+tt2yPm2Dv7LMTNrubktLbla7p38zb3k014hd/6MMcb0Dk0fchuNxpMDHFoU\nuL3RaPy9KIoxnW+WMcYYY4wxxhgzeFpFct8AngN+DLz9vq0BXAAcCLw8M1+qSIIiOFBGHVdfffWy\nce9HDF9+ufyaCy+8EIDbbrst2RQ9bBaRaZWAJRe51OuYEGeuueYC8hGmGDHqdiRXvoqJu9ZYYw0A\nPvnJTybbYostBsBf//rXZLvyyiuBajKv1157DWi+gt0qIpR7n3waI3nyb/S9zmmMenY7kqsIbkx8\ntv766wPw6U9/OtkWX3xxoNrWiRMnAtV+quPxd+g3y2/RV3od+1qzJF0xwp6Ltuv7Y2S42/1UybnU\nD6Ec++uuu26yffSjHwWqY08J5qZMmZJsipjH3xT91feY+lUrxYF82vezIF8WKtq63U9VuioXCVff\nhDJBWkw6pz72/PPPJ9v06dOBamK+6EOo+krHYv9vNidHFC1+++23+9mGM5FfLvGZXse5S69jIjpd\n13StgPL3RZ/qd6rvxLGt354bq7k2RaRoit+lczOcEclm6pLcdTfOYfJl9LN8n1Mc6N9W0XR9R0wa\nmFN6vfDCDNFaVJHE7+0VcsqOqFTS69y81/fv4vviPK3+r/kGSj/H5KBSkEWf9p1nepXoZ/kozhvy\nh/6Nx3QORo0alWzLL788AMsss0yyqX/++c9/TrapU6cCVT8Pl6qwG8S+KF+qL8b5V31y6aWXTjYp\nHMeNG5dsuu+96aabkm3ChAlAOQdA/RKiDhXyr3w6zzzzpGO6R9M9McDWW28NwJgxZexTzxrXXHNN\nsp144okAPPvss8k2VD5ttSd3deBq4ChgVFEU1xZFMQl4F7i+KIprh6RVxhhjjDHGGGPMTND0Ibco\niveKojgc2BbYt9FoXNpoNJYDvEHFGGOMMcYYY0ztaCvxVFEUjwHbNRqNnYDLgYVm5Utzsq+FFprx\nkTEcLolBTN5z1113AVW5YTMJVk6i1E7bIC91lgQiynMkRRtOCYN8GSWLkrgstdRSybbEEksA8Oqr\nrybbvffeC5RSDWhPNpSTweRkZzmJlCSWsc3Rf3WQLUnCFuXKG220EVBKwaGsQxzlLJIKRj+rz0a/\ntVO7NOe/XDK0eO7VP+M4kWxkOH2r8R0lQltuuSVQJkWDUlYU+4SSzUWJS06aqd+ekzPm5MqScuak\nT3E+0vvi9gmNmeEc+2pj7KdrrbUWABtssEGyqc9Gqbjmrih5e/zxx4GqT+eee26gnD9iX1N/jufl\noYceAqpSRPXn6PsHH3yw3/drzAxngjT1nSh5k3wwbglZZZVVAFhnnXWSTZLCeI1QH4xjT/OLPjf6\nVCipIpTS/LjVRH6J8sSrr74aqCZn1PcOpwRc4yvOZ30lhgALLLAAUE3ot9lmmwGw2mqrJduCCy4I\nlH0zvtbnRRmyvjdK4zV/5JJWKlEgwPHHHw+UW3ugHmO/2fUjJyuOvlpxxRUB+Od//udk0xwc+7jm\nl1ziydz356T2eh3niCOOOAIot6BBvu553Yl+Vr9be+21k+2AAw4AYNNNN0023e/mtkU0O6e5JIov\nvfRSsn3ve98D4Mwzz0y2XN/uJeSP2Hd33HFHAP7jP/4j2bTtKZd0cbA14nfbbbdk0xaJY489Ntni\nHNLLxH4niXH8nZtvvjmQ31I4WOL9nfruT3/602TLJWTtBIMqIVQUxUXAmsD6wIst3m6MMcYYY4wx\nxnSVQZcQKoriTeBOmFEvtyiKgTIwN/sM3v/7ZNOKeFypVYTshhtuSLZmZWk6Qfzc3IpkLhKZWyXv\nNjl/rLrqqgCssMIKyaaowRNPPJFsWunrRPtz7cglWoo2+TJGOPV6OH2q1e8YNV155ZWBatRMEZi4\nEqUoWFxFndl+0mpFMhcN0UpxXOVtVgamWyipRmyrfKrIApTJkVQiKP5NtClSmIsYaqzG8dtM2ZHz\nn5IrxM+JEc5c1LHbaPzEOUkqjjifao6N0UklmYoruvKlIodQRnPUh6IKRz546qmnkk2qhhidVCQ5\nJsPS6ryixzB49c1Q0DcxDJQ+iIm7NA/EhCaKMMb2y0fNriXxuxQljBFazSWKpkN5nuPnKvp76623\ntvyd3aRZEsg4x+p17DvyaUw8pfMRIzyyyafxc3VNib7SfBSTYGp8xGimxtEVV1zR+ofWhFyEKtrk\nK5XFg3K+i/6Qj3JJzkS8puWSUek8xzlK9yW9FLXNEduv3x6vZRtuuCFQVdBorLcbYdT74jnQcV2r\noOyzIzExkuYAgL333huoKsLiNWkw5OajeI3M3Z+OFOI151//9V+BaqLaOIZnlXh+dA/VDZ8OKpIb\naTQaswOPdbAtxhhjjDHGGGPMLNGqTu6mTQ7P3uSYMcYYY4wxxhjTdVrJlScB04Eh0T7EpBCSxMT6\nSpImdbM2Xasafrl6fbmapMNFlFlIChCT58jnUR4oSWunJS45iVROnqZEI/HcKkHFcEqZ5I94fl9/\n/fV+Nv2m+++/P9kmT55ceT/MvDSjlQRcEqYo21OysZjkp9nndYtcrV5JZmMyB72OSeeU/CXWyZUM\nNPq2Hblrq7qcGjNxPpLcJo6dvt85HOgcx772wAMPAOWWDyjbGBOk/eEPfwCqdQElU47no69Pc8nT\ncgmFooxXiSdiQqG6+VKof8btBvJb3Jaga9Ptt9+ebGp/TFCmPh6l9n3fH8eqxneUJ0qmvM8++ySb\nakvnEtLEdtZBAp67vsgW5375PEq1lRwxShYlI4w+6itTjvcO8keUQauuY0y+pHEe/ad+GsdEnSSh\nufOakyvHRIg33ngjUJ0PlFhRW0ignAPl00ceeSQd0zlS0h8ok9XEbQk6R9Fn+v52k4jWldhmjW/N\nq1D22T322CPZlJxOc2ZMyKXtBnFLlMZ+vG/S38Z+On78+H62Xkf+1RwKcPTRRwPVPqatG/JR7Fd6\nHe+Fm8mb4zhXEq9erIvdivibzjjjDAC23XbbZNM2jVyNc43lXJ/MEa99v/vd74DuyJVbPZ09AXy+\nKIob+h5oNBpzAG/0/xNjjDHGGGOMMWZ4aPWQOxVYF+j3kMuM6O6gk05BfuVaKwpxpVGvY4r54Sh/\nElcnFl54YaCMPkIZQYulHbqN/BJXrrVCG0shKDnMkUcemWy56MJQE6NmSnYR255LA9/tVV75JSbU\n0Sp2TEqiCNpBBx2UbIqGdWVj/furbDFCodX3uELcLHFIt1BCnRgNUKmTeH61Ev0///M/yaayNLkI\n1WDJRTniucoln5H/csk/hhP5dNq0acmm+eCxx8q0CRpLDz/8cLIpapCLpszKb9PqbixLpsQrUcGj\nBBS5CNlw+jZX8kf9Ls7zuXGu19E2s78pzn/yUTxXilzGaJzK7NUt6thuG+TzXPmzSLN+2szP8dqj\n87zrrrv2e18s3zRp0qTK++tCu/0p5ytdt6Ja5p577hnUZwgltAQ466yzgGqUTX/z5JPlLeN1110H\njKyEPvqdse8o6vjrX/862fqW+MslpFt//fWT7bLLLgOqCkJ9V1SRSPFQh+tSp4ljb8KECQBcddVV\nyabri/p1vPZontS5ANhrr72A/PX87LPPTrZ4/zfSiP1EZfxiv5NCQ9eSeJ2RuuBPf/pTssUkdn2/\nI5Z7iiqdoabVQ+4eAx0oiuJdYJnONscYY4wxxhhjjJl5mj7kFkVRr2VLY4wxxhhjjDGmCcOSMUnh\n6ygnkEwmJiKRRChKXLopw8jVdpNcOUqeJHdotul6qJHkJyY7UWIHJSGCUvYS/dxNKZtkjPIjlHUn\no1S9DolSJI+JCQ+uvPJKoNpW2R599NFk66YESzLlmCwkV7u3DrIwtSfKXlQHO9ZlffDBB4Hq2B9q\nqWBMRLHOOusA1SRJU6dOBarzVh2ktfJp7JOSv0XZnOT3cXvCUG//iDW6VSc3fr+kdtFWB6mdxkqc\n03PzZG6e6mT742dJ5r3FFlskmxI2Rvm/5MrDsbWnGYP1y1D1g3gelWgpJrTScc3rUEoW69A3O00n\nrv9xnCgZWkxIo774i1/8ItniPDqSyW1fyCWEFDofsR62rufRz7oefuMb3+hnG+nktjzGe4W+qP+p\nVjzkt28p6V2U1o7EMd+M2IeUwDKHniHi1s0c2oaoxKHdZvg36RljjDHGGGOMMR1iWCK5Wo2KKwZK\nkPK///u/yaaV1GarXt0gluXR6mSMOmmVaDgjuVqVislGlGDm8ssvT7Zrr70WGL4U8/JfTOijtsSV\nuJxPu72ipu+OZUQUMYkb51U6qJurqHEVUgk+4irlfffdB1STEdUh+YyIY1qJ0WLiMZW16kY0Suc5\nrkgq+YIiZQA333wzUD33dVrljZEC9dk4H2icdSOiL/WLkntA6d9YAko+7UQysU7SbjKjoW5rjIbt\nuOOOQFWxofMbE6Vo7NTBj3VEyVTZ/wAAIABJREFUCWoADjzwQKCq4pDK5Jhjjkm2D0qEbGY5/PDD\n02uptOK1WwkQzz333GRz/8wjleDJJ5+cbLmooxIF3Xnnnd1pWA+jUnobbLBB0/ep9FNURJo8StYX\nyzLlOPTQQ4Hhe+ZwJNcYY4wxxhhjzIihrUhuo9EYBRwErAVUHtuLothq0F86W/+v1Urf5MmTk01R\ns+Fe8Yur6do7FqNiwx1phrxPtSKtEi3RNlw+Ver8uM9ZbYpRJ62cD+e5zxULVymLGHVUGZRuEtum\nyE6M0KncToxC6/hw+jS3f13jJ+7L1KpfN9qqNsX9o9oPddNNNyWb9qfEqE4douM5BYnalStjM1TE\naIOKyG+yySbJpnEe9+Yo6liH/eI5Yv+Tn7s5fhSBAPjmN78JVK9HmtsvueSSZHPUMY/8dtJJJyWb\n8kHE/qcSOLHU1nDfg9QVlf/bd999k03jJM433/ve94Bq3gCT5+c//zlQLRckYj/cZ599AI/3dtC1\nu9m1Ekplh8d7nniNP+OMMwZ8X1TgxTKQw0G7cuXzgA8BFwFvtXivMcYYY4wxxhgzLLT7kLshsPD7\ntXGNMcYYY4wxxpha0u5D7mRgJeCuTnxpTpqmZC533HFHskkaOlzSAcmbYohe8s8ou5FtOCV3uZIX\nOWntcEiro/9GjRoFVDehq5RALCkgucNwykbUhtj+119/HahKhLp53nV+JRMDGDt2LFAtI6LEaPF8\n10GCI2lQTu7fTWltRH1yq63KnRcqG3X11Vcnm8ZTHSTKEbUn9tPc+Bnq8z/33HOn1/vvvz9QTeb1\ny1/+Eqhun6hbmZtmdHP8aPuJJMoAo0ePBkqJN8D3v/99oDrH12Gc15GNN94YqJZgErFMnJIo9VLf\n7CZRLq8yj7ntUjGRpCTgJk/clvDVr351wPfFki5xHjX90TUIYP755x/wfRMnTkyvY2lD0x8lioR8\nMjTxm9/8Jr0e7u2c7T7k7gVc1mg0bgKeiweKojis040yxhhjjDHGGGNmhnYfcg8HRgOPA3E3/Ewt\nGSvyEFeclXQoJsoZjohJjIRqdXLOOedMNrXzxRdfTLZuJskZiFxSoeFIPhORL2OER+WYYrKm559/\nvp+tDtEy+S1GnfWb4ip/N8+7isIrKgGl/+66qxRa5Erw1CHCo/OaS5bRzXMeSwNtv/32AIwZMybZ\nzjzzTKAs0wD1GOc51J7ov262UZGd3XffPdl23nlnAB588MFkO/HEE4GqYqNuvqwL6667LgBf+MIX\n+h077bTT0mtFc+owX9aROM5POOEEoEx+COWYPuSQQ5ItRspNfzbbbLP0eplllul3XHP7DjvskGzD\nVT6k7uh+4oorrki2XIRM1/HNN9882Tzm8+i+/eijj276PvXTnXbaacjb1OvovnOdddZp+j6N86hA\nGm7afcjdHVihKIrpLd9pjDHGGGOMMcYME+3WyX0UcJ5yY4wxxhhjjDG1pt1I7unA+EajcQz99+T+\nabBfKolalKrVRc4SkyrMM8+MksBR8qQEFUqUBfWo8ZiTLOZk4d1EtVyV2AfK+rhxg79qJNctSZLa\nkDu/3WxfrIm7/PLLA9VEM0888QQAjz32WLJJ+l0HP0ZyY7+bSAoWZTdrr702ABdffHGy3XjjjUBV\nQl83X/ZluNq36qqrAnDooYcmm7Z1HHzwwcmmZGh19+NwEa8zxx57LAALL7xwsmk7whFHHJFsdblu\n1hVtRQBYbrnlgGr/mzx5MgDnn39+srl/5tHcefLJJ/ezRZ9pq8fdd9/dxdb1JrrHXGONNZq+T2P+\nueeea/o+U271iNsScqgm7ptvvjnkbep1/vM//7Ot9+2yyy5AvZL2tfuQe8D7//6oj70Alu1cc4wx\nxhhjjDHGmJmnrYfcoij6ZxfoMMO9eqoVybj6o4RJMRGFkvzEFfThbntkuNsSkyYoahuRL2PiLkV9\nhrvtAzHcUceY/l5R5VtvvTXZXn75ZaCatK2uvhxullhiCQDWX3/9ZLv88ssBuO6665KtrpHwuqAE\ncgD/9V//BVQjkV//+teBMlIG9VC81Bkl6wJYeeWVgbJkGcB+++0HlOPdDIySzxx55JHJpvk0lv/7\nl3/5FyCfCM9UUWK+WL5OqLwawAEHzIiJeO5szXe/+10gn2wq+vSww1zEpF3OOeecAY/Fsa9EdGZg\nlBhN1/Mc0aeXXnrpkLdpsLS7J9cYY4wxxhhjjKk9A0ZyG43G/UVRrPz+66cYoFxQURRjcnZjjDHG\nGGOMMabbNJMr7xte7znUDRluYn1coeRSUQYqWZOlOFXkP8nEoJQnRrm3/Fenjel1IvZDyeVjneZH\nH30UqCZLsAy0OTHxmWo8Pvzww8k2adIkoDrOTR7J6vbcs7wkSPqtOrgA5557LuBx3g7qnz/4wQ+S\nTfPAWWedlWyqiWtaI+n3kksumWyaJ0855ZRkUyJJMzDqixdccAFQldYqueVvf/vbZHMin9YomeRB\nBx004HuOOeaY9Npy+tYssMACAIwePXrA9/zqV79Kr11nuDVKiNYsiddRRx2VXtfxuWjAh9yiKCaH\n19d2pznGGGOMMcYYY8zM00yu3NZO96Iovte55nSXGDXT67hipoRIXvFpjUovRZ+qJFCMNNZxpadO\nxOQ9Ki8Qy1UpEY392Bqtlq+22mrJpr4Yk0w5gts+K6ywAgD77LNPsv35z38Gqkl+Yjkwk0cRMZVn\nGDt2bDo2bdo0AL7//e8nmxUbrZH65eijjwaqyiLNo7EEk+fR1igJ2uqrr97vmBJw/vrXv+5qm3qd\nr371q0A+OafGuRMjDY5LLrkEyKsyNc5PPfXUrrapF4n+mzhxYsv3//73vx/K5swyzeTKMeY/B7AL\ncAvwBDAGWB+4YOiaZowxxhhjjDHGDI5mcuW99brRaPwe+FxRFBcE287ArkPbPGOMMcYYY4wxpn3a\nqpMLfAb4fB/beOCUzHtrj8LxMYGC5AyW1rZPlIJJZhvl3vKl/dga+XK++eZLNvnN9W8Hh8b3Yost\nBpS1cQFuv/12AF577bXuN6yHUV1c1RWNiSh+8YtfANUEc6Y148aNA2D33XcHqmP72GOPBeC5557r\nfsN6jCiv++Y3vwnAoosuClR9Kumdat2bgYn3Rueddx5Qvd4LJUJ89tlnu9OwHiZuRfrhD3844PtU\nd/Spp54a8jb1OgsttFB6vfHGGw/4Psnq1V/NwMTtXdG/fdH9fUziWUfarZP7MHBAH9v+wCOdbY4x\nxhhjjDHGGDPztBvJ3Qe4qNFoHAJMA5YC/gbsPFQN6zRxZVIrv3GVV68dKWuNVnRjaRuVCoklQ+zL\n1qhfKoKrZFNQro67DMvgUNRxzTXXBMpkXQBPPvkk4GRy7RDnzLXXXhuA7bffHigj4gCTJ89IxO/x\n3poYAT/44IOBUmnw2GOPpWMnn3wyYJ+2Q4w27LfffkCZCDGWsznkkEMA+7QdNN6hVByIOHfK3/Zp\na2K5oHjv1Bclm7NPWxNLrMXrVV/OPvtswNf9dpgwYUJb75syZQpQ/4SIbT3kFkVxe6PRGAdsCCwJ\nTAemFEXh4l3GGGOMMcYYY2pDu5Fc3n+gva7lG40xxhhjjDHGmGGi7YfcXkUShma1s/q+NnnkS0nu\nJAmDsi6m/dia2BflywUWWACoJu6Kr01zYlIUyT8XX3xxAKZOnZqOKQGFac2oUaPS6x122AEoZXZX\nXnllOhYloaY5qjMMsNlmmwHldoRzzjknHYu1sU0ezaOf/exnk03zqGSJktJDuVXBDIx8+q1vfSvZ\n4nUeqgmRbrzxxu40rIfRfVOsLd73fjRuqTn++OO707AeRn1ygw02GPA9cZtXlIqbPLqHUsLOHPH+\nPs67dabdxFPGGGOMMcYYY0ztGbGR3L4rZU4yNXNEP2r1TP8qegvlqpl92poYdVSiKdlefvnldKzu\nG/rrgPpnTNi1zDLLAKX/YhkWJ55ojcb3iiuumGxrrbUWUJZfueOOO9Ix+7Q1Umxst912yTb//PMD\n5Zi/9NJL0zH7tDUa8yrBBGXUTOoClWIC+7QdFllkEQA+8YlP9Duma/xvf/vbfjYzMGPGjAHgox/9\naL9jul8aP358ssX7KpNn5ZVXBmDuuece8D0xOWKMlJs8iop/+MMfHvA9Tz/9dHrdK6XYHMk1xhhj\njDHGGDNi8EOuMcYYY4wxxpgRw4iVK1uS3Bmi/yT/lBQsHrMUrDW55GeSJqkmbkziY5/myUnoVRsX\nyn55//33A/Daa691sXW9SfTpvPPOC8B6662XbPLvvffeC1Rl9SZPrNuoWqPbbrttss0xxxwAPPPM\nMwC8+OKLXWxdbxKTICkZ2hprrNHvuBLMaV41AxN9+oMf/AAoZctQzg26HsV6ziZP9OkJJ5wA5Gvj\n6loVt3+YPHE+Pf3004Hq1i8hn9500039bKZKzqe5+1QRk3j2ik8dyTXGGGOMMcYYM2IYsZFc03kc\nWewMMaFU3/Ir9vHAaIUxrjRqJfLtt99ONkVw5du33norHeuV1cduIV/GZBMqHfTqq68m28SJEwGY\nMmUKAK+88ko6Zp9WkU9nn332ZFtuueUAeOONN5JNpVgmTZoEODreDPk0JppRBDeWBdNYf+SRR4Ay\nSm4GJpYKk+Igl/RQY/6hhx7qTsN6mOhTlbKL1/a+CdJuvvnmLrauN5HCCGC++eYDqtcezRGaD2Ii\nP5MnKuCaJZxSgjlFe3sJR3KNMcYYY4wxxowY/JBrjDHGGGOMMWbE0BhqqVmj0bCWrT+3FkWx7sz+\nsX2apWd8mtvYX1PJZy19Gv2n1zGBgpA8rGYS8Fr5NCcBV03XKGWSDyUHjRLRGvTdWvo09knVdJV0\nEUoJmOo41yzpXK18KmKimUUXXRQoa2ZCmchPWxb+8pe/pGP2aZ4oU1x22WUB2GijjZJNY12y+unT\np6dj9mme6FP1zy233DLZNPYvvvhioNy6APbpQMSx/7GPfQyAbbbZJtnee+89AH7/+98D1Zqu9mme\nmCBt1VVXBWDHHXdMNm1bOPXUUwGYNm1aOtYr131Hco0xxhhjjDHGjBi6Ecl9AXhiSL+k9xhbFMUi\nrd+Wxz7NYp92Hvu089inncc+7Tz2aeexTzuPfdp57NPOY592nrZ8OuQPucYYY4wxxhhjTLewXNkY\nY4wxxhhjzIjBD7nGGGOMMcYYY0YMfsg1xhhjjDHGGDNi8EOuMcYYY4wxxpgRgx9yjTHGGGOMMcaM\nGPyQa4wxxhhjjDFmxOCHXGOMMcYYY4wxIwY/5BpjjDHGGGOMGTH4IdcYY4wxxhhjzIjBD7nGGGOM\nMcYYY0YMfsg1xhhjjDHGGDNi8EOuMcYYY4wxxpgRw2xD/QWNRqMY6u/oQV4simKRmf1j+zSLfdp5\n7NPOY592Hvu089inncc+7Tz2aeexTzuPfdp52vKpI7nDwxPD3YARiH3aeezTzmOfdh77tPPYp53H\nPu089mnnsU87j33aedryqR9yjTHGGGOMMcaMGPyQa4wxxhhjjDFmxOCHXGOMMcYYY4wxIwY/5Bpj\njDHGGGOMGTH4IdcYY4wxxhhjzIjBD7nGGGOMMcYYY0YMfsg1xhhjjDHGGDNi8EOuMcYYY4wxxpgR\ngx9yjTHGGGOMMcaMGPyQa4wxxhhjjDFmxOCHXGOMMcYYY4wxIwY/5BpjjDHGGGOMGTH4IdcYY4wx\nxhhjzIih6UNuo9FYus//d2s0Guc1Go3zG43GnkPZMGOMMcYYY4wxZrDM1uL4XcB8AI1G48vA94Bf\nAgVwRKPRGFUUxW8G+6Uf+tCHACiKoun7Go1G5f0As802Wz/b//3f/wHwt7/9rZ9N3xG/q9n36jsB\n/uEf/qGfTX/797//va3P6xZqa7tt0fvj62jT75Mf42fPyu+NvuxLHfwYyfk010b9pvjbcjb5tN2+\nOBKRP9r93bn+khuPHzQ/Rgbr02afMaufM1LohE+NMcYYM3y0esiNd5gHALsURTEFoNFoTAJ+Bwz6\nIdcYY4wxxhhjjBkKWu3JjcvYSwA3pgNFcTPw0aFolDHGGGOMMcYYMzO0iuTO0Wg0Tnv/9YeAxYBn\nARqNxvzAuzPzpZIaR4mcZMgf/vCHk22eeeYBYLHFFku2pZdeGoAFF1ww2SRTfuONN5Itymzjd8b3\n/fWvf+33GXPPPXeyqS2vvPJKsj366KMA/OUvf+n3t8MpbctJq3Ny74985CNA9XcuvPDCAMw333zJ\npt/02muvJds777xT+fe9997rdyzKuPW98bvmnHPOyvsBXnrpJQDeeuutZMtJe7tNTnIscnJv+RbK\n3zzHHHMkm3z65ptvJtu7784YQvJl7LfRl32/a/bZZ082vY5/K19GCX/u87pNM5/m3hf7rsaj5goo\nf7P8GG3t9iF9Vzyn+o74t7lxXic5ayuf5t7XbEtGbqvCrLQp176RKDcfji0ZlpsbY4wxVVo95B4e\nXh8NzM/7D7nApsBVQ9EoY4wxxhhjjDFmZmj6kFsUxQ+aHBsPjJ+pL80kj1KEL0YTl1pqKQDWWmut\nZFt99dUBWGSRRQb8XCijk4suuigA8847bzqmKOJzzz2XbIrMxgiZonGPPPJIsh111FEATJ06Ndn6\nJrkaDvTbYzRKkS/5FsoI+AorrJBsa6yxBgAf/WipPte5iedIfyvfzj///OmYIofTpk1LNkUso+/l\n07vuuivZjjvuOAAeeuihZKtD1DH2JyF/RMWBftOSSy6ZbCuttBJQ9Wkumc1cc81V+YzoU/lPkW4o\n+6f6NZRj5tZbb022s846C4Bnnnkm2eoQNYv9STRTHER/LLHEEgAsvvjiyaaxF1UAsunfeK76fieU\nc8no0aOTbaGFFgLgtttuS7aJEycCVWVHHRQHuUR+uYi53hcVB1LL6F8o/RbVFn0VB/G79LlxntEc\noTkcYIEFFgDgwQcfTLb77rsPqKob6jD2myXya5VgTv7IzR/xt/Xtp7nzF/uuxrn8COV5e/HFF5Nt\n+vTpQFXdUIfo7mCTI7arTMiRm+ty50rzaey7sr399tvJ9vrrrwNVZUwdfNpMHTFc5Pq/Xsf+r/5Z\ntySedUw6l1MbxdciN5fU6XfUkXbVRq1sJk8nlE2zolSapTq5jUZjzKz8vTHGGGOMMcYY00lm+iG3\n0WjMDjzWwbYYY4wxxhhjjDGzRFO5cqPR2LTJ4dmbHBs0kgFFiZBkgVHGqvdFeZFkLzHJjySNY8bM\nCDZHiZ4+9/nnn082yT833HDDfp8RP1fS27rJQXLJvJrVb41JoyQVjDJMSWGifFxIzhV/tyRHUXIq\neenyyy+fbPJflKGec845rX7esJCTG8qWk3xK4gqldDlKmCU9jHJD/Y36WpR2q6/H86J+L99C2T+X\nWWaZZJO0VtLFupCTDuck4Op3SjQH5baF+DtzidTkQ/k0+lvfFeXNkoxLYhvbcueddybbAw88AMCr\nr76abHWQ1uZ8qvEb+7B8FKXuq666KgDLLbdcsqk/xT6u10oCGPv1qFGj+n2/fBp9r7bErR4HHXQQ\nUJUwN5tPuzXXNpO452T1+r1Q9rvoU231iPOpzpH8F6Xdks5HGbn6dfS32jRlypRk+/a3vw1Ut9k0\n21JTB5/m5N7Rz/JvTDipbQZx+4KuTbltHerrse/q83Jy/WuvvTbZfvKTnwDw2GOP9XtfThLdrXlB\nPs21IZKz5bYvaI6I1xfNi3pfvEbJp6usskqy6W/jmNA92qRJk5Lt2GOPBeDJJ59Mtr5JA2Pbu3XP\nlUs62Iyc73NJDONcqHlU5y+ODW11iveiuubFe1ElUL3mmmuS7dxzzwXg6aefTrZcwsRcP62DT5tt\nqcrdx8Y+prGcS1Cpe9DNNtss2TQ/xy2Kute6/vrrk+2KK64Aqlu/dP8c/adz3omEje2Q2/rVbH5v\nNwFnbiuefBl9qmerzTffPNmWXXbZfu+TT+PWrxtvnFGoJ96f5pKvdsKnrRJPTQKmA8N/J2eMMcYY\nY4wxxrSg1UPuE8Dni6K4oe+BRqMxB/BG/z9pTd+EG1CuNsVkJ4qYPPHEE8mmFYCYAEKv+5YNer+d\nA7YjruAoShSjjlrxjUlRtPLQrdWadlF7WpUCkX9jMqN777233+dpVSVGfPuWr4krSVrRihE1rfAc\neOCByabVzJgURUm/6ubTXBv0O2P7VYoqKg5eeOEFoLqiq34aV/8UmdUqZC5KE1fa5dOtt9462bRq\nFs+9+mzdknrkkmXkytjIV7FUlyKpceU/FzGRD5X0K0Yloi+FktnlouO5CE8d/BjJRchyK6/67dEH\n8l8spyY1TTwf6p85H+j98VxIqREjl2pLjFjGyERfhtPP8dogchGCXIRR15Bx48Ylm64lsT/pbxSR\njMkU5ZfoU0XQ4mfoHCmiFj9X5e6gHn1Wc13sVzmfqn/G3zl27Fig+juVMFERBSh/u65Dsf/l+lru\n+0X02emnnw7ko44D/U03UP/L+TSXhDKOfUVoV1xxxWRTpGu99dZLNvXL3GfIp63KkqkfxyjwlVde\nCVSjObn7GNEt38qnOf9Fcr9Tv0/9FcprdbxmK5KbUy3Iv/H7c/7QNTJ+1z333AOU9x9QDxWHxmOu\nbGirduVUGdtuuy0AX/jCF5JNio6ciinn09x3yafrrLNOssmX8V6k7j6N1w29jvOV3hfvNz/1qU8B\n8LWvfS3Z1LdyyeQ0JnJ9M36/niGiCubHP/4xAFddVRboUcQ3d083Kz5ttSd3KrDuAMf+Djw5wDFj\njDHGGGOMMabrtIrk7jHQgaIo3gWWGei4McYYY4wxxhjTbVrVyX2v2fGZReHrdutTtZIEzGwoO0pO\nX3vtNaCafEZh8z//+c/Jpg39OanScJLzaa4uYUy4I9rdrN73fblzFmV+uaRfaqfqYwI8++yzQD2S\n+ERyPs2dd/k0ylli4geRq6na16e5OnhRtidfbrXVVskmuWisPawa0L3gU7UxSuNzPpX8MpfwLCfL\nzsmWJLuJ43zvvfcGqvJSnec77rgj2ZSorm4ScPktJ8WKPlU/idJkbQWJPmrm01ztbRHrDH/pS18C\nSllZ/A7JziHfT+vgU8nWol9y22y0/UPXD4CXX34ZqP5OyTrj5+l3SnYWpWM6V0rkAbDbbrsBVSmd\nPiNKwZqN/eH0rfwWrwdqT+yneh2TUOYkd+rHUcao98nf0ac6R1HevPHGG/ez6TtizXdJFlttqRku\n/0Y5be66r3uduM1LiYviPYHm26eeeirZ5Ad9Xjx/2hYTk3+pz8bkkvreKKPVlql47tu57xhq+m7L\nguaSzNi+vgn6oNyqEMej/KA+HuXK8lVMVCUZfpTcqy/GxJQ6f/HeNjcPdPu+IJewM5f8Knd/pTGs\nbUUAm2yyCVC9N1I/zvlUr2OiKl3Dcte+uEVR2ybjVsqcXLnbPs1tHxC5raDx/kD9ad11S6Hu7rvv\nDlS3ben3aa6I80wuYWhu65w+I/bJ3NyT82kntojNUp1cY4wxxhhjjDGmTrSSKw8JudW6bqeJh+oK\nhFbJlawGytWcCRMmJFssH1Incj5ttpLfCd/mPiOuin3mM58Bqivtr7/+OgCXXXZZP1vdaJbILNdP\nO+HTVqUTtPIWE3goQhH7aV19GiMJIvc7B7sq2q7vc2UzFMWJ84FWxy+55JJk09ivQ6QxkvOpaNen\nuZXXgT6n7/u1QhxXurViHlfT5dMLL7ww2RR5aHUt6Da5xHu5yEPOp/qduch6s7/NldGJ5bJ23nln\noOoX9cmLLroo2RQlqls/1Up+9Klo16dRIXPTTTf1e78+u9m1KUaEVlttNaA612s+jdcoJUcaznJB\nORS1ilEStTEXIYt9TH3nxRdfTLbbb7+939/qWpLzrebMf/zHf0w2JZWJ1yh9xtVXX51sOpd1iohD\nc5/GfqLXcZzrb6MKoW8kHEr1mvpa9LciyDvssEOyffOb3wSqEWLdn06ePDnZHn/88X6f1+1yQTkU\nsYvXCLUhXr9yPtXxu+++O9nOPPPMyr8ADz/8MFD25/gb1Rd1nw/wxS9+EahGzHXeYrkbfW4v+FTt\niT6VLaecjUl9L730UqA6RpVMVUqheF6khvvnf/7nZNt+++2BqoJG7YvqTX1uThnTaZ86kmuMMcYY\nY4wxZsQwLJHcHJ2MhrVCqxFrrrlmsmlfXozw3HzzzUA1zXWzleLhpFl7utFWRR422GCDZNtjjxl5\ny+LeFq2+x6LwWjHtVZ8OVbvl01gUXtGc2E+1Fzeu6PZSP+3mPiyN/dhP/+mf/gmoqhC0knvdddcl\nW137abP8AN3wqVaF4/6ejTbaqHIMyvl04sSJyZYrP1QHtBLeqpRJru82K4OS+5zc+9RPVdoO4GMf\n+xhQ9ZXmU63CQ7nfqW77nHP78UWrfiqfxujaYCP+et/KK6+cbFJxxDE0depUAC644IJkU9SzbmXu\n5I+Z8WkuF8Jgy/lpj2hUwGm/X/zcW2+9FYDf//73yaa9682UKMOBIk8xatXsGhXHmaKritRCmcsh\n+kO/OedbqWDiflPZYtRO1/3TTjut33dFn3bz3nogNM/PjE8VsYx75JV3IO6d7XstiWNCvoyfq/ur\n6FNFG0844YRkk+IgF9mvg0/bHfvR9/rNzzzzTLJJuRJzdkhpkLumKQK+5ZZb9vuuqGRQborjjz8+\n2RQdz0X2O+3Tth5yG43GKOAgYC1gnnisKIqtsn9kjDHGGGOMMcZ0mXYjuecBHwIuAvqn5zXGGGOM\nMcYYY2pAuw+5GwILv18bd5YZbpnPfPPNB8BPfvKTZFtooYWAasmS//7v/wbKVPcw/G0fiGbSt6Ei\nfpf896Mf/aifTbIkKJNSxGQXveTToZapxO9SkonDDz882bTZP/rvyCOP7Germ/xT5KREYqglylCm\nx8/5NErMfvGLXwC979NOo++I37XUUksBcNhhhyWbxn6UmP385z8HSkkd1K8Um2iWrKNdCXir/tz3\nfMV+Khntd7/73WRTSZa9+YX0AAAPzklEQVSYfOlnP/sZUJWdjUSfDtb3en/06dJLLw3AwQcfnGy6\nF4g+PeKII4BqUpa6SWrFYH2au5bl+kuzcxB9qsRoKhkGZUms6FNd9x955JFkyyUgqwO50kDNpPE5\nP7dKBNb3c6NPVYJJJV2gLHcTfap7rZjQpxd92gr9TSw3o8+Lfu77uTmfbrPNNsmmLV/xGqX7gph4\nqq5bv5qN/XaJcm9Jk6NN3yFf5ny62WabJZsS0T355JPJpn56ww03JFs3t361m3hqMrDSUDbEGGOM\nMcYYY4yZVdqN5O4FXNZoNG4CnosHiqI4LPsXNSOuQOy5555ANfmMVmrjhvP//d//Beq7Mp6jm6tN\nsTzIvvvuC1STeWkF7Ljjjku2a6+9FrBPByKWsfnKV74ClAlnoPTpsccem2zup3m0qhmTdH39618H\nqsln5NNf/epXyXb99dcD9ulAKLIAZWRsxRVXTDYllDjqqKOSbcqUKUB9o2I5Op24K7fSLpsiYFCW\nDFluueWSTck8FL2F3vfpYP+mVaSibxRHkVqAb33rWwCMHTs22RS1iKoujf2R5NNmyplWPu0bxYkl\nVzT2R48enWzy6U9/+tNkU1LEukYac8xKP20WCYfSl7qHkvIF4MADDwRKhQyUUbbo02uuuQbofZ+2\n6n+KJubKjMX7er3WPdQiiyySjun+dPHFF082JViK86mSzI4kn+b8mys1lCvfpHsnJY+V6g3g85//\nPFAtESolrNSFAJdffjkwfD5t9yH3cGA08DgwX7DXK35vjDHGGGOMMeYDTbsPubsDKxRFMX0oG2OM\nMcYYY4wxxswK7T7kPgr0Tvw+Q5TYHHLIIUC1fqsknz/84Q+TLbep3ZREmcKXv/xloCq3vfrqq4Ey\n6QTYp62I9QZVuznKwq+44gqgKq+zT5uz/PLLp9ef+9zngKokZ8KECQD88pe/TLZekit1E0mfotx7\n1113Bao+HT9+PFDdqmCf5qVlGt9rr712sm2//fZAVWp28cUXA3DSSSclW6/7tBMS8NxnqC9+/OMf\nT7attupf7fCiiy4C4He/+12yfVB8qr4VZaC5BEs6Lunipptumo4p6Uz8jPPOOw+A008/PdlGok+b\nyUBz78ttfdA96Cc/+cl0bJNNNun3WaovfOaZZybbSPRp7lhuy5DGd/SRbNr2scUWW6RjG264IVD1\n2VlnnVX5t+/xXqTdsa97xlzd4PgZ8qmSHkafrrPOOkDVZ+qf5557brINt0/bfcg9HRjfaDSOof+e\n3D91vFXGGGOMMcYYY8xM0O5D7gHv//ujPvYCWLZzzek8WkXbf//9k02lWWJ5kP322w8oN/ibgdHq\njhIkQLnJf/r0UtGuxEmvv/56F1vXm2gV7d/+7d+STeqDp556KtkOOGDGUHzjjTe62LreZLbZZkxv\nX/va15Jt1KhRADz++OPJpn4cU+ebPIrmxH6q5D6xPIgSfMWyDybPHHPMAZRjG2DeeecF4MEHH0w2\nKZA+yD5tN1KhaM4+++yTbOqn9957b7J95zvfAT6YPm1VDkfoHkr+22OPPdIx9dM77rgj2Q499FBg\n5Pu0WV+M0cdmiZUUIdtpp52STT6dOnVqsqms1Uj0aasx3W7Zxr5Rxy233DIdk09vvPHGZFOZwA+K\nT3O2nIojlyBN900bbbRROiafKlEfwG9/+1ugXj5t6yG3KIplhrohxhhjjDHGGGPMrNJunVxjjDHG\nGGOMMab2DBjJbTQa9xdFsfL7r59igHJBRVGMGaK2dQQl8olyZYXoTzzxxGR76KGHutuwHmbZZWco\n1Pfaa69ky9XEfeyxx7rarl5GiXx23333ZJNPf/Ob3yTbk08+2d2G9SCS2yiRj5L4QFm/NdbEffrp\np7vYut5EPlVt8ZjER9Kko48+OtmeeeaZLrauN5EUTIlmPvGJT6Rjks7HZGjPPvtsF1vXm8in//iP\n/wjA+uuvn469+uqrQHXs26et0VYaJZxac80107FXXnkFgGOOOSbZnn/++S62rv7kZKLaSqPEaDGR\n30svvQSU0k+wT6F18i/103XXXReAcePGpWMvvPACUL3nl+2DTC4ZWk6urP659NJLp2OaO2PSvrgF\ntC40kyvvG17vOdQNMcYYY4wxxhhjZpUBH3KLopgcXl/bneZ0DpWy0Uq4kk1BGWGMkYdcmnJTRenu\ntRK+8MILp2MPP/wwAP/zP/+TbLl0+qbKnHPOCZQ+XXDBBdOxBx54AIBTTz012ezT1ijpzM9+9jOg\nTEQBcN999wFw9tlnJ5t92holmTjssMOAMgkNwN133w3ABRdckGyeT1ujfqnkR/IxlIl8/vCHPySb\nfdoazZ9KNjfPPPOkY7feeisAV111VbLZp3liNGehhRYCYN99Z8Q9ok+nTJkClCUYwT4diOhT3Tsp\niZeuWQCTJ8+49Y5JkuzT1sinO+64I1Dtp9deO+MR5rbbbks2+zRPbux/+tOfBqr99Oabbwbgnnvu\nSbY6+rSZXPmwdj6gKIrvda45xhhjjDHGGGPMzNNMrjw6vJ4D2AW4BXgCGAOsD1yQ+TtjjDHGGGOM\nMWZYaCZX3luvG43G74HPFUVxQbDtDOw6tM2beTbbbDMAtthiC6C6wfrkk08GvJm/HaJ0QZIFJUqJ\nPpWk1pv5WxN9usMOOwCw3nrrAVXp7JlnngnYp+0QfbrzzjsDsNZaawFVn0qmXMcECXUj+nSXXXYB\nSp/GsX/eeecB9mk7KJEHwK67zrh8rr766kC1n1500UWAfdoOSjgD8NnPfhbI+1TSb/u0NbGfqoar\nfBrHvqTfSpZkBib202233RaoJvESkta+/PLL3WlYDxN9qmSIukZFJP3+y1/+0p2G9TDRp0rgpySe\nkTvvvBMok8/VlXZLCH0GuLiPbTywdWebY4wxxhhjjDHGzDzN5MqRh4EDgF8F2/7AIx1v0SwQN0X/\n+7//O1AmSIllQk466SQgn5LcVImb97/+9a8DZYKU6NNTTjkFsE/bISbtOeCAA4DSz9Gnp512GmCf\ntsOoUaPS6/322w/I+1TRcSebak1M2PXFL34RKH0aS1mdc845gH3aDjGx3J57zihaIJ8+8cQT6dj5\n558P1DORR92IPt1tt92AvE/Hjx8P2KftEJNKSsWh63706RVXXAHYp+0QfSoFl65b0aeTJk0C4G9/\n+1v3GtejKDESwNZbz4i5LbDAAkDVpzfccANgn7ZDvO5/6lOfAsq++/jjj6djU6dOBerv03YfcvcB\nLmo0GocA04ClgL8BOw9Vw4wxxhhjjDHGmMHS1kNuURS3NxqNccCGwJLAdGBKURTvDWXjjDHGGGOM\nMcaYwdBuJJf3H2ivG8K2zDLrrLNOer3KKqsA8N57M57DL7zwwnTMSRJao6Qz0acrrrgiUPr04ovL\nbdpO5tEa+VRJpgDGjRsHlD6NdTGfe+65LrauN5FPN9hgg2RbfvnlgdKnEyZMSMeeffbZLrauN5FP\nN9xww2RTP3333XcBuPzyy9Ox6dOnd7F1vUlu7C+33HIAvPPOOwBceeWV6di0adO62LreJHeNWnbZ\nZQF4++23Abj66qvTsaeeeqqLretNlHBKSaag9Kn66TXXXJOOxa0gJo98utJKKyVbX5/GOsNRZmvy\naOyPHTs22ZZZZhkA3nrrLaCsNwxVma3JI58utthiySafvvHGG0BZFxtKn9Z9O127iaeMMcYYY4wx\nxpja03Ykt8585CMfAeDjH/94smn1TFGG008/PR1zkoTWzDbbjK6x7rrr9jv2zDPPAPbpYJFP11hj\njWRTtFEr4irFBPZpO8inH/vYx5JN0UZFbuzTwSGfxshDX5+eccYZ6VjdE0/UgQ9/+MMArLDCCsmm\nKI58etZZZ6Vj9mlr1E9jNEc+1XyqpGhQzrVmYFQ+ZMkll0w2RcYUzbngglRJMs0LZmB0L7rIIosk\nm3wpxYGSokHZh83AyKcx6dzrr78OlAkQlRQN7NN2kE9jEs9XX321cuyPf/xjOqa+W3ccyTXGmP/f\n3p20xBGEYRx/5uSGIki8uMSrN/EuuKCgH9qjCIL7vn6AiAuCV3MIT9XbSRtH0oaZmv/vklCtBx96\nqrum334LAAAAxWCRCwAAAAAoRteWK/slaSnv4TQ/P5/GXEZzeXkpiYYz7YiZumTBTXwk6eXlRZJ0\ncXEhiaYTn+W9G+OL/W6Cdn5+LommE581MDAgSRocHExjbtpzdnYmSbq7u0vHOr1JQifo6+uTlEuU\nJOn29laSdHp6Kkm6urpKx8j0Yy5XjuWd/sw7U1+rJDJth8uVXfopSUdHR5LyNcpzgESm7fBn/vn5\nOY3t7u5KynPAyclJOkamH3OmDw8Pacx7jHq/8cPDw3SMTD/me9XHx8c0trOzIyk37Nzb20vH2MO9\nfU9PT+n/29vblTHPBVL3vPrFk1wAAAAAQDG67kmuvxXr7+9PY94+JL7Y//r6Kkm6ubmRVG064W+B\n+MbsF+cRM/VWF9PT02nML+/7yVj8JodMq5yHn4pJeVuG2CjFTT38LTmZvs95+KmYlLe2mZycTGNu\nQHF9fS2JTP/GefipmJSzHB8fT2OuOPCTsdgYiUyrnIeb+Ej52hSberi66Pj4WFL1KS+Z1ouZjo6O\nSsrVHFKuhDk4OJBUbThDpvVixYarjeJ84Huo/f19SfmahffFqjjfV8UxV3F4Po3VCKgX83Oz2ej3\nyi3fB8Tf5bNfFTP1Zz5m5EzdbNaNqOLvdnqmPMkFAAAAABSDRS4AAAAAoBhdUa4cH6nXlYK5RCnu\n3+by5K2tLUnVRgqoL1MYGhpKYy5HjA0R3Cxhc3NTUi4J73XOMmbqktqxsbE05oZTbuAl5eYIbpoQ\nSz+6pRzkK/hvj6V0LlGKjbvm5uYkVffLc+MJl9jEknHPG724D6mzjJk6m4mJiTS2uroqSZqdnU1j\nnk/dwGNkZCQdc1lYL+6Z6Szj9cjlifFVj42NDUnS4uJiGnOmPnfjee09CGO5ba/MA84yZuqS5Piq\nhzNdXl5OY3414f7+XpI0MzOTjvn1hVhu2ysNaXyNrztPY0br6+uSquepzzv/63lVyg2+YmlotzSk\n+VfONJZ2O9N4nq6trUmqZurrm0vu4z2BM40NgHrleuX7ppipr1FxPl1ZWZFUzdS/4+ZI8bPtTD0v\nSHn+LX1ercvU91JTU1NpbGFhQZK0tLT0x895HTA8PJyOOVPfb0n5utVJmfIkFwAAAABQjNZXr7hb\nrdYPSeyLUvX97e3t28c/Vo9Ma5Fp88i0eWTaPDJtHpk2j0ybR6bNI9PmkWnz2sr0yxe5AAAAAAD8\nL5QrAwAAAACKwSIXAAAAAFAMFrkAAAAAgGKwyAUAAAAAFINFLgAAAACgGCxyAQAAAADFYJELAAAA\nACgGi1wAAAAAQDFY5AIAAAAAivETJb2Jlz1Wc88AAAAASUVORK5CYII=\n",
211 |       "text/plain": [
212 |        "<matplotlib.figure.Figure at 0x7f405a247e80>"
213 |       ]
214 |      },
215 |      "metadata": {},
216 |      "output_type": "display_data"
217 |     }
218 |    ],
219 |    "source": [
220 |     "fig, axs = plt.subplots(16, 11, figsize=(11*1.5, 16*1.5))\n",
221 |     "for i in range(16):\n",
222 |     "    axs[i, 0].set_ylabel('dim {}'.format(i), size='large')\n",
223 |     "    for j in range(11):\n",
224 |     "        axs[i, j].imshow(perturbed_reconstructions[i][j], cmap='gray')\n",
225 |     "        axs[i, j].set_yticks([])\n",
226 |     "        axs[i, j].set_xticks([])\n",
227 |     "        "
228 |    ]
229 |   },
230 |   {
231 |    "cell_type": "markdown",
232 |    "metadata": {},
233 |    "source": [
234 |     "We can see what individual dimensions represent for digit 7,  e.g. dim6 - stroke thickness, dim11 - digit width, dim 15 - vertical shift"
235 |    ]
236 |   },
237 |   {
238 |    "cell_type": "code",
239 |    "execution_count": null,
240 |    "metadata": {},
241 |    "outputs": [],
242 |    "source": []
243 |   }
244 |  ],
245 |  "metadata": {
246 |   "kernelspec": {
247 |    "display_name": "Python 3",
248 |    "language": "python",
249 |    "name": "python3"
250 |   },
251 |   "language_info": {
252 |    "codemirror_mode": {
253 |     "name": "ipython",
254 |     "version": 3
255 |    },
256 |    "file_extension": ".py",
257 |    "mimetype": "text/x-python",
258 |    "name": "python",
259 |    "nbconvert_exporter": "python",
260 |    "pygments_lexer": "ipython3",
261 |    "version": "3.5.2"
262 |   }
263 |  },
264 |  "nbformat": 4,
265 |  "nbformat_minor": 2
266 | }
267 | 


--------------------------------------------------------------------------------