cDCGAN_Mnist_PyTorch.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "cDCGAN_Mnist_PyTorch.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mKsRcN_HY1tB",
        "colab_type": "text"
      },
      "source": [
        "cDCGAN Mnist Simple"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tuEg6bqdRKTf",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "from matplotlib import pyplot as plt\n",
        "from google.colab import drive, files\n",
        "import itertools"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jRgZmtP5XNXJ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.nn.functional as F\n",
        "from torch.autograd import Variable\n",
        "from torch.utils.data import DataLoader\n",
        "from torchvision import datasets, transforms\n",
        "import torchvision"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2vsOBlH3RKgl",
        "colab_type": "text"
      },
      "source": [
        "Generator for 32x32 image"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XKsOjNydY42U",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class Generator(nn.Module):\n",
        "\n",
        "    def __init__(self):\n",
        "        super(Generator, self).__init__()\n",
        "\n",
        "        d = 128 # latent\n",
        "        self.deconv1s = nn.ConvTranspose2d(100, d*2, 4, 1, 0)\n",
        "        self.deconv1s_bn = nn.BatchNorm2d(d*2)\n",
        "        self.deconv1t = nn.ConvTranspose2d(10, d*2, 4, 1, 0)\n",
        "        self.deconv1t_bn = nn.BatchNorm2d(d*2)\n",
        "        self.deconv2 = nn.ConvTranspose2d(d*4, d*2, 4, 2, 1)\n",
        "        self.deconv2_bn = nn.BatchNorm2d(d*2)\n",
        "        self.deconv3 = nn.ConvTranspose2d(d*2, d, 4, 2, 1)\n",
        "        self.deconv3_bn = nn.BatchNorm2d(d)\n",
        "        self.deconv4 = nn.ConvTranspose2d(d, 1, 4, 2, 1)\n",
        "\n",
        "    # weight_init\n",
        "    def weight_init(self, mean, std):\n",
        "        for m in self._modules:\n",
        "            normal_init(self._modules[m], mean, std)\n",
        "\n",
        "    # forward method\n",
        "    def forward(self, input_, label):\n",
        "        x = F.relu(self.deconv1s_bn(self.deconv1s(input_)))\n",
        "        y = F.relu(self.deconv1t_bn(self.deconv1t(label)))\n",
        "        x = torch.cat([x, y], 1)\n",
        "        x = F.relu(self.deconv2_bn(self.deconv2(x)))\n",
        "        x = F.relu(self.deconv3_bn(self.deconv3(x)))\n",
        "        x = torch.tanh(self.deconv4(x))\n",
        "\n",
        "        return x"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Db15unTEUmdL",
        "colab_type": "text"
      },
      "source": [
        "Discriminator for 32x32 image"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0AJspoBoUo1Y",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class Discriminator(nn.Module):\n",
        "    def __init__(self, in_channels=1, out_channels=1):\n",
        "        super(Discriminator, self).__init__()\n",
        "        d = 64 # Num of dimension.\n",
        "\n",
        "        self.conv1_1 = nn.Conv2d(1, d, 4, 2, 1)\n",
        "        self.conv1_2 = nn.Conv2d(10, d, 4, 2, 1)\n",
        "        self.conv2 = nn.Conv2d(d*2, d*4, 4, 2, 1)\n",
        "        self.conv2_bn = nn.BatchNorm2d(d*4)\n",
        "        self.conv3 = nn.Conv2d(d*4, d*8, 4, 2, 1)\n",
        "        self.conv3_bn = nn.BatchNorm2d(d*8)\n",
        "        self.conv4 = nn.Conv2d(d*8, 1, 4, 1, 0)\n",
        "\n",
        "    # weight_init\n",
        "    def weight_init(self, mean, std):\n",
        "        for m in self._modules:\n",
        "            normal_init(self._modules[m], mean, std)\n",
        "\n",
        "    def forward(self, input, label):\n",
        "        x = F.leaky_relu(self.conv1_1(input), 0.2)\n",
        "        y = F.leaky_relu(self.conv1_2(label), 0.2)\n",
        "        x = torch.cat([x, y], 1)\n",
        "        x = F.leaky_relu(self.conv2_bn(self.conv2(x)), 0.2)\n",
        "        x = F.leaky_relu(self.conv3_bn(self.conv3(x)), 0.2)\n",
        "        x = torch.sigmoid(self.conv4(x))\n",
        "        return x"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aytORPMl_i3_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def normal_init(m, mean, std):\n",
        "    if isinstance(m, nn.ConvTranspose2d) or isinstance(m, nn.Conv2d):\n",
        "        m.weight.data.normal_(mean, std)\n",
        "        m.bias.data.zero_()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tNOuSjpVRgti",
        "colab_type": "text"
      },
      "source": [
        "# Train data sets "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_LzKbe0YyXdl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Hyper parameters\n",
        "img_size = 32\n",
        "batch_size = 128"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tkSZ1nZm_Ma0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "data_transforms = transforms.Compose(\n",
        "    [\n",
        "     transforms.Resize(img_size),     \n",
        "     transforms.ToTensor(),\n",
        "     transforms.Normalize(0.5, 0.5)]\n",
        ")\n",
        "\n",
        "train_datasets = datasets.MNIST(root=\"./datasets\",\n",
        "                                        train=True, download=True,\n",
        "                                        transform=data_transforms)\n",
        "\n",
        "loaders = torch.utils.data.DataLoader(train_datasets, \n",
        "                                      batch_size=batch_size,\n",
        "                                      shuffle=True)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9noof5wjeoK6",
        "colab_type": "code",
        "outputId": "445e7a10-30ef-4fec-c469-7bba3a4be928",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "# Exercise 1\n",
        "iter(loaders).next()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[tensor([[[[-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           ...,\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.]]],\n",
              " \n",
              " \n",
              "         [[[-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           ...,\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.]]],\n",
              " \n",
              " \n",
              "         [[[-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           ...,\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.]]],\n",
              " \n",
              " \n",
              "         ...,\n",
              " \n",
              " \n",
              "         [[[-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           ...,\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.]]],\n",
              " \n",
              " \n",
              "         [[[-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           ...,\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.]]],\n",
              " \n",
              " \n",
              "         [[[-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           ...,\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.],\n",
              "           [-1., -1., -1.,  ..., -1., -1., -1.]]]]),\n",
              " tensor([7, 0, 9, 9, 3, 8, 9, 8, 9, 0, 4, 5, 1, 0, 9, 2, 0, 9, 1, 5, 8, 1, 5, 1,\n",
              "         7, 1, 4, 0, 7, 5, 3, 4, 3, 3, 3, 6, 8, 7, 1, 6, 2, 3, 0, 2, 1, 2, 4, 8,\n",
              "         6, 1, 3, 1, 3, 5, 0, 1, 9, 0, 6, 7, 4, 5, 5, 4, 2, 4, 7, 2, 8, 7, 0, 8,\n",
              "         9, 1, 5, 6, 4, 3, 7, 4, 9, 0, 1, 7, 8, 9, 4, 2, 7, 8, 4, 0, 3, 7, 9, 7,\n",
              "         7, 1, 8, 7, 6, 2, 5, 0, 8, 6, 3, 4, 9, 7, 7, 8, 2, 8, 1, 1, 6, 0, 9, 1,\n",
              "         9, 6, 3, 7, 2, 7, 4, 9])]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 682
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GZOJ0a6di-bF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def im_show(image):\n",
        "    print(image.shape)\n",
        "    mean = np.array([0.5, 0.5, 0.5])\n",
        "    std = np.array([0.5, 0.5, 0.5])\n",
        "    im = image.detach().numpy()\n",
        "    im = im.transpose(1, 2, 0) # c h w -> h w c\n",
        "    im = std * im + mean\n",
        "    plt.imshow(im)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pKPV9DhdfMOj",
        "colab_type": "code",
        "outputId": "f4132679-28a8-4e19-a425-9c5a54ce6a86",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "# Exercise 2\n",
        "iter(loaders).next()[0].shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "torch.Size([128, 1, 32, 32])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 683
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nTvrVgHMlt_a",
        "colab_type": "code",
        "outputId": "a0efcc93-9b5c-4658-ee44-3c058abc7021",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        }
      },
      "source": [
        "# Exercise 3\n",
        "im_show(iter(loaders).next()[0][0])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "torch.Size([1, 32, 32])\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Pq0Pbr7wpb6K",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def create_noise(batch_size):\n",
        "    z = torch.randn((batch_size, 100, 1, 1)) # 正規化 -3.0 to -3.0\n",
        "    return z.cuda()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Sl1frv7EvmAK",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def create_label(batch_size):\n",
        "    label = torch.zeros(batch_size, 10)\n",
        "    label.scatter_(1, label.type(torch.LongTensor), 1)\n",
        "    label = label.view(-1, 10, 1, 1).cuda()\n",
        "    return label.cuda()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gNgvXlm_orpb",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def make_log(epoch, i, loss_info):\n",
        "    log  = \"epoch: \" + str(epoch)\n",
        "    log += \" iter: \" + str(i)\n",
        "    log += \" loss: \" + str(loss_info)\n",
        "    return log"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "usmGWdp_EntR",
        "colab_type": "code",
        "outputId": "44286a3d-792e-4bf8-ae51-f30ad7c6d03b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "# Exercise 4\n",
        "epoch = 2\n",
        "i = 15\n",
        "loss_info = \"0.0001\"\n",
        "make_log(epoch, i, loss_info)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'epoch: 2 iter: 15 loss: 0.0001'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 687
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A-AiP0oJCr-l",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def show_train_history(data):\n",
        "    x = range(len(data['loss_D']))\n",
        "\n",
        "    y1 = data['loss_D']\n",
        "    y2 = data['loss_G']\n",
        "\n",
        "    plt.plot(x, y1, label='loss_D')\n",
        "    plt.plot(x, y2, label='loss_G')\n",
        "\n",
        "    plt.xlabel('Iter')\n",
        "    plt.ylabel('Loss')\n",
        "\n",
        "    plt.ylim(0, 10)\n",
        "\n",
        "    plt.grid(True)\n",
        "    plt.legend(loc=4)\n",
        "    plt.tight_layout()\n",
        "    plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VscMU33Ix1yG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def show_result(num_epoch, show = False, save = False, path = 'result.png'):\n",
        "\n",
        "    G.eval()\n",
        "    test_images = G(fixed_z_, fixed_y_label_)\n",
        "    G.train()\n",
        "\n",
        "    size_figure_grid = 10\n",
        "    fig, ax = plt.subplots(size_figure_grid, size_figure_grid, figsize=(5, 5))\n",
        "    for i, j in itertools.product(range(size_figure_grid), range(size_figure_grid)):\n",
        "        ax[i, j].get_xaxis().set_visible(False)\n",
        "        ax[i, j].get_yaxis().set_visible(False)\n",
        "\n",
        "    for k in range(10*10):\n",
        "        i = k // 10\n",
        "        j = k % 10\n",
        "        ax[i, j].cla()\n",
        "        ax[i, j].imshow(test_images[k, 0].cpu().data.numpy(), cmap='gray')\n",
        "\n",
        "    label = 'Epoch {0}'.format(num_epoch)\n",
        "    fig.text(0.5, 0.04, label, ha='center')\n",
        "    plt.savefig(path)\n",
        "\n",
        "    if show:\n",
        "        plt.show()\n",
        "    else:\n",
        "        plt.close()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "22F1uocEpyHe",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# target_real = target_fake = pred = fake = None # For debug\n",
        "# def train_batch(batch, z):\n",
        "#     pass"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ueQ4kwzCWXrI",
        "colab_type": "text"
      },
      "source": [
        "Settings"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WizjjANQtq2l",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Hyper parameters\n",
        "learning_rate = 0.0002"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jmmGo98fCNfl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Train history\n",
        "train_hist = {}\n",
        "train_hist['D_losses'] = []\n",
        "train_hist['G_losses'] = []"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SUiklXHGiDCG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Create filled data for discriminator.\n",
        "label_fills = torch.zeros(10, 10, img_size, img_size)\n",
        "ones = torch.ones(img_size, img_size)\n",
        "for i in range(10):\n",
        "    label_fills[i][i] = ones\n",
        "label_fills = label_fills.cuda()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rvWZoqFVln5G",
        "colab_type": "code",
        "outputId": "826a3ba5-f95f-4e16-f3b8-915977320769",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 899
        }
      },
      "source": [
        "# Exercise 5\n",
        "label_fills[1]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([[[0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         ...,\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.]],\n",
              "\n",
              "        [[1., 1., 1.,  ..., 1., 1., 1.],\n",
              "         [1., 1., 1.,  ..., 1., 1., 1.],\n",
              "         [1., 1., 1.,  ..., 1., 1., 1.],\n",
              "         ...,\n",
              "         [1., 1., 1.,  ..., 1., 1., 1.],\n",
              "         [1., 1., 1.,  ..., 1., 1., 1.],\n",
              "         [1., 1., 1.,  ..., 1., 1., 1.]],\n",
              "\n",
              "        [[0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         ...,\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.]],\n",
              "\n",
              "        ...,\n",
              "\n",
              "        [[0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         ...,\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.]],\n",
              "\n",
              "        [[0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         ...,\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.]],\n",
              "\n",
              "        [[0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         ...,\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.],\n",
              "         [0., 0., 0.,  ..., 0., 0., 0.]]], device='cuda:0')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 691
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3XSf8-SNO3lO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Create one-hot vector\n",
        "label_1hots = torch.zeros(10,10)\n",
        "for i in range(10):\n",
        "    label_1hots[i,i] = 1\n",
        "label_1hots = label_1hots.view(10,10,1,1).cuda()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L4FnsREsl8sD",
        "colab_type": "code",
        "outputId": "39f78242-60a6-4336-fbdb-4166cd5c4f38",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197
        }
      },
      "source": [
        "# Exercise 6\n",
        "label_1hots.view(10,10)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([[1., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 1., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 1., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 1., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 1., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 1.]], device='cuda:0')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 692
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "saixeuVgMkBv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Settings: Auto grad parameters for targets. \n",
        "#  Discriminatorの教師ラベル生成と自動微分化\n",
        "t_real = Variable(torch.ones(batch_size)).cuda()\n",
        "t_fake = Variable(torch.zeros(batch_size)).cuda()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZJpXbh2kSHkm",
        "colab_type": "text"
      },
      "source": [
        "Instance"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rhnkPWJIet35",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Criteiron: Binary Cross Entropy loss\n",
        "BCE_loss = torch.nn.BCELoss()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HHj1J9V5WSLY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Instance\n",
        "G = Generator().cuda()\n",
        "D = Discriminator().cuda()\n",
        "G.weight_init(mean=0.0, std=0.02)\n",
        "D.weight_init(mean=0.0, std=0.02)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UeyLX89HPusi",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Optimizer\n",
        "optimizer_G = torch.optim.Adam(G.parameters(), lr=learning_rate, betas=(0.5, 0.999))\n",
        "optimizer_D = torch.optim.Adam(D.parameters(), lr=learning_rate, betas=(0.5, 0.999))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TAw3ihf5-nAY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Train log.\n",
        "loss_list={}\n",
        "loss_list['loss_G']=[]\n",
        "loss_list['loss_D']=[]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "S_pqZmwnNTIo",
        "colab_type": "text"
      },
      "source": [
        "Train"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ry18m6PbiIfs",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Hyper parameters\n",
        "epochs = 10\n",
        "epoch_from = 1 # 継続の場合は、途中の番号を設定"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rJDgKVjHdmee",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Train loop\n",
        "for epoch in range(epoch_from, epochs + 1):\n",
        "\n",
        "    print('epoch', epoch)\n",
        "    \n",
        "    if epoch == 5:\n",
        "        optimizer_D.param_groups[0]['lr'] = learning_rate/2\n",
        "    elif epoch==7:\n",
        "        optimizer_D.param_groups[0]['lr'] = learning_rate/4\n",
        "    elif epoch==9:\n",
        "        optimizer_D.param_groups[0]['lr'] = learning_rate/8\n",
        "\n",
        "    for i, (x_batch, t_batch) in enumerate(loaders):\n",
        "\n",
        "        # 取得したミニバッチのサイズ\n",
        "        N = len(x_batch)\n",
        "\n",
        "        label_1hot = label_1hots[t_batch]\n",
        "        label_fill = label_fills[t_batch]\n",
        "\n",
        "        # Genarator loss\n",
        "        z = create_noise(N)\n",
        "        fake = G(z, label_1hot)\n",
        "        pred = D(fake, label_fill).squeeze()\n",
        "\n",
        "        loss_G = BCE_loss(pred, t_real[0:N])\n",
        "        loss_list['loss_G'].append(loss_G)\n",
        "\n",
        "        # Lock discriminator (Not to need on PyTorch.)\n",
        "        \"\"\"\n",
        "        for param in D.parameters():\n",
        "            param.requires_grad = False\n",
        "        \"\"\"\n",
        "\n",
        "        # Update Generator\n",
        "        optimizer_G.zero_grad()\n",
        "        loss_G.backward()\n",
        "        optimizer_G.step()\n",
        "\n",
        "\n",
        "        # Unlock discriminator (Not to need on PyTorch.)\n",
        "        \"\"\"\n",
        "        for param in D.parameters():\n",
        "            param.requires_grad = True\n",
        "        \"\"\"\n",
        "\n",
        "        # Discriminator loss\n",
        "\n",
        "        # Real loss\n",
        "        real = Variable(x_batch).cuda()\n",
        "        pred = D(real, label_fill).squeeze()\n",
        "        loss_D_real = BCE_loss(pred, t_real[0:N])\n",
        "\n",
        "        # Fake loss\n",
        "        z = create_noise(N)\n",
        "        fake = G(z, label_1hot)\n",
        "        pred = D(fake.detach(), label_fill).squeeze()\n",
        "        loss_D_fake = BCE_loss(pred, t_fake[0:N])\n",
        "\n",
        "        # Total loss\n",
        "        loss_D = (loss_D_real + loss_D_fake) * 0.5\n",
        "        loss_list['loss_D'].append(loss_D)\n",
        "\n",
        "        # Update Discriminator\n",
        "        optimizer_D.zero_grad()\n",
        "        loss_D.backward()\n",
        "        optimizer_D.step()\n",
        "\n",
        "        if i%100 == 0:\n",
        "            print ('G:%.4f D: %.4f' %(loss_G,loss_D))\n",
        "\n",
        "    epoch_from += 1"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-m7xkjguSW-L",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Labels for test.\n",
        "t_test = torch.zeros(100, dtype=int)\n",
        "\n",
        "n = 0\n",
        "for i in range(100):\n",
        "    t_test[i] = n\n",
        "    if (i+1)%10 == 0:\n",
        "        n += 1"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LmFJKdk6TsuL",
        "colab_type": "code",
        "outputId": "0de5cead-1de4-4a4b-c8d2-fbbedb3f8741",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 107
        }
      },
      "source": [
        "# Exercise 7\n",
        "t_test"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "tensor([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2,\n",
              "        2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4,\n",
              "        4, 4, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 7, 7,\n",
              "        7, 7, 7, 7, 7, 7, 7, 7, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9,\n",
              "        9, 9, 9, 9])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 694
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vSV66Bp_UlRO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def test():\n",
        "    G.eval()\n",
        "    z = create_noise(100)\n",
        "    label_1hot = label_1hots[t_test]\n",
        "    fake = G(z, label_1hot).cpu()\n",
        "    return fake.cpu()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y6J3GT5n-rBL",
        "colab_type": "code",
        "outputId": "ed3e17d2-7389-4e5f-8d7b-4e60782d3610",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        }
      },
      "source": [
        "# Test\n",
        "fake = test() # n:0-99\n",
        "im_show(fake[60])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "torch.Size([1, 32, 32])\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FPGNX8SX6iB_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def im_layout(images):\n",
        "    print(images.shape)\n",
        "\n",
        "    mean = np.array([0.5, 0.5, 0.5])\n",
        "    std = np.array([0.5, 0.5, 0.5])\n",
        "\n",
        "    im = images.detach().numpy()\n",
        "    im = im.transpose(0, 2, 3, 1) # c h w -> h w c\n",
        "    im = std * im + mean\n",
        "\n",
        "    plt.figure(figsize=(10,10))\n",
        "\n",
        "    row = 10\n",
        "    col = 10\n",
        "    n = 0\n",
        "    while n < row * col:\n",
        "        n += 1\n",
        "        plt.subplot(row, col, n)\n",
        "        plt.imshow(im[n-1])\n",
        "        plt.axis('off')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mtp8nhm7LmPi",
        "colab_type": "code",
        "outputId": "6451dfe6-0fde-4e60-bdb6-8f7cb5463297",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 592
        }
      },
      "source": [
        "im_layout(test())"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "torch.Size([100, 1, 32, 32])\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAItCAYAAADfWbvdAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOx9Z3Ob6Xn1Qe+dAAgCJEgCBHuRRHXtrmLF9jprx5OsM3Zm0j5l4plMfkm+JD8gk8kHZzJxSfy67lrxrnbVGyn2ikKCANF7I8r7QbnufQiRFCWBAKXgzGi44lLEcz93u8q5zsWr1Wpoo4022mijjTbaeFfBb/UDtNFGG2200UYbbZwk2sZOG2200UYbbbTxTqNt7LTRRhtttNFGG+802sZOG2200UYbbbTxTqNt7LTRRhtttNFGG+80hEf9Tx6Pd2ipFo/HAwCc1mquWq3GO87PHTXGZoPeKQDw+XzUajVUq9VDf/5tHCMXx1lDxxkjj8ernfb1eBhedQ55PN6pGSOPx2v4Ov2/Mo9vI9p78TlO4148LhoxxtM+v4eN8ZUjOzweb9+f04hXfa7DxtLMMdLn0Gfy+fymfHYr57FRn8t9dwf9v9O4Vl/1mY76+WauF+5nAs+Ncj6/MQHi0zZHzUar92IjPvuo33Ea9yHhde6Mw77/tp+nR/2+03qeHgdHRnbqwb2Ia7Xaabbs3vjn6yf0pMdKv5++HuUpNwL1i7ZarTZ9Phs1xvp3R+AaAXw+H+Vy+cTf63HRyDUqFArZntzb22vKGOl59vb2Gv4769HMfdgq1BuslUqlqWNt1Gcdtk7pK4/Ha8lZ8zK86X6k+aP/14rxNfozDxqjQCCAQCAAj8fD3t4eKpVKQz/zJHGkscP12LjhrNNyYTQS9QdqrVZraOShlZv7sHGctgPndcE9TLkHDXd8p/GAfRVwDVPuOGi8lEp628dIX7lz2Mi9eBpw2Dwe9N9vGw6aJxrPuzKHh42jWq2+k2Osv/sFAsGpDnYchpdxdl7477dxkEehPgVCB+vrjJG8bKFQCIFAwDxuuqhqtRrK5TL29vZQLpebHmk4yKDj/v0w8Pl8CIVCiMViCIXCfdEg+rf5fB6VSqVlhvBhkbh3Za0eND7uGKvV6mu9ez6fD5FIxP7Qei2XyyiVSmy9NuM9HrQXGwXySGmMNE56b7lcriln21FjrDfwXhe0X7njBYBSqcS88ZPap/VG+buy/+rxLp81hIMMnrfZ+TjS2KFwFV3SwLs3sVxjhMZI3ycc1/iRy+Uwm82w2Wwwm80wGo3Q6XSoVqsQiUTg8XjY2dmBx+NBIBDA9vY2stksyuXyib7Po9IAL/tcHo8HtVoNm82G4eFhGI1GZsRlMhkIBAJUKhV89tlniEajSKfT+95jM3EQUZa7Od+VNVs/xtd933w+H1qtFr29vejr64PVaoVarYZYLMbu7i5WV1fh9/uxubmJUqnUNEP2sHl8XQiFQuj1enR0dMBut8NgMEClUkEkEiGXyyEWi+Hzzz9HJpNBqVQ60XXCPVfqx9iI98vj8aBQKGCz2WCxWGC326HRaAAAy8vLWF9fRzQaRSqVOpEUBN0ZPB4PpVKJPRN3Tt/Gi/IwcNfKq54x9Vy305oxqR/j25S64uJIY4fC4lwLnbyhdwncCAU3CsPNwR4FgUAAsVgMp9OJ4eFhTE1NYXx8HFqtFgqFYl8+N5vNYnV1FWtra/j5z3+Ora0tpFKpEzUQ6r3H4x42fD4fEokEY2NjcLlcmJqawsjICCQSCfh8PqrVKiqVCvL5PJRKJebm5rC5uYlAIMCiVs28IA/7/qseQhSVA77yVKVSKdsHlUqFecjNNOwa6SnT3E5MTODixYuYnp6GzWaDVCqFQCBAKpXCzMwMFhcXkcvlEA6Hkc/nT3w+Gx0N4PP5MJlMcLlcGBoawpUrV2A0GqFSqbC3t4dEIoHd3V2IRCL4/X6kUinkcjkEAoETG+9JRTxonV67dg1DQ0Po7e2Fy+WCWCxGuVzG4OAgbt++jbW1NSwsLLBobCPBNWq4Zyj9nfu1EeA6q8Rb4/P5EIvF+/boSfFZuOcp94486vMkEgk0Gg0sFgu0Wi3jE66trSGXy6FYLLJoaisdtEbOF/F9+Hw+iziazWYWcVUoFACec9UymQxyuRwKhQJSqVTDeIjHNnbogd811Btz9eAaP4f9jEgkgkqlgtPpxNjYGM6dO4eJiQkoFApmGNBnVSoVaLVaaDQaLC4uIpvNolQqIZPJnNgY6bO5/31YipILoVAIlUqF/v5+dlmMjIxAKBSiXC4jmUxCIpFAIBAgHo+z9ZLP55HP51EqlVAoFE58wx7n9/P5/GMd7EKhEFKpFHK5HOVymaUmOzo6WEqnVCohn8+jWCyiUCi84LEeFJloBBp5+IlEIiiVSgwNDWFychLnzp1jhw+Px0OxWGSXxsLCAgqFQlOjO40AHbAmkwn9/f2YmJjAuXPnYDAYoFAoUCgUkEgk0NnZiVwuB4/Hg2g0ing8jr29PcRiMWYQNOq9v+y8eROIRCIoFAoMDw9jYmICTqcT/f394PP5KJVK0Gg07JzhRusa+SwHccfo/GxUdJXSkAKBgJ2v9KdSqUAgEECr1aJaraJQKCASibA5bMS7P+rfH8Q7q///CoUCXV1dGB8fh9VqhUAgwN7eHgqFAsLhMBKJBAAwZ7EVXLyjnEfg1arOJBIJS5PTeSqTyTAxMQGxWAyRSASNRoNa7XmBRSQSQTQaRSKRAI/HQyaTYQbPmxjnLzV2uKhfKK+yeLmkpqOMp6N4JSeBo8ZIG1QsFh96afN4PKhUKgwMDOCjjz7C1NQURkdHIRKJDvxZHo/HQsu5XI59f2Nj48QukoOeu37RHlQloVKpMDQ0hDNnzmBkZARjY2OQyWSIRqPw+Xz46U9/iitXruD69eu4ceMGDAYDenp6IJVKsbu7i1gshu3t7RNP0x0F8vhEItFLjRA+nw+VSgWz2Qyr1YpwOAyBQAC5XI5Lly4hkUggkUggmUwiHo8jnU4jHA6z1IdEIgHwPK2Uy+Uaekg1cm3weDzodDo4HA584xvfwNjYGGw2275qIIFAgJGRERgMBiSTSWbY0UF8Emj0+hcIBJBKpRgeHsbFixdx9epV9Pf3Mx6LQqGAXq9HT08PHA4H/H4/QqEQ3G431Go1Njc3sbGxwVKzjXi+kzQWVSoVbDYbJicncfbsWTgcDojFYva5Op0OWq0WPT09ePr0KQqFQsP3Zv1l1OhoDp/Ph1wuZw5Jd3c3M9ABIJfLQSKRYHh4GJ2dnQiFQvjv//5vRhcgh6WRqB8bGdn175bOIrvdjitXruDjjz/G0NAQRCIRisUiarUa5ufnsb6+jmQyiXK5jEqlglKphFKpdKrSR2S0HHXG0Xvo7u6GRqOBSqVCKpWCSCSCXq/H3//930Ov10Mul0MsFjMnklLnPp8PDx8+xNbWFtLpNHK5HLLZ7Gu/h1cqPa/HcRdwPSmv3iqkRVBPHKQDpt4AaRbos7neez2EQiE0Gg16enrQ29sLg8HAUiCHQSgUQq1WY3p6GpFIBHK5HDs7OygWiy0pOT3M6xKLxTAajejp6YHRaIRAIMDs7CwePHiAzz77DCsrK4jFYshkMvjoo48wMDCAzs5O9Pf3w+12Y3NzE7/5zW8QCoWQy+VaslnJwHlZVILWpEwmg9PpxHvvvcc8D5rjSqXCDh9Kf3zyySeYm5tDOp2G0WjE0NAQtFotZDIZlpeXEQ6Hsbu7C7/f3zIuUz1kMhn0ej26u7sxMjICk8kEgUAAYH/kQSKRwGAw4NKlS/B4PKhWq3jw4EGLn/544PP5kMlkMJlM+PjjjzE4OAir1coI9ty1zufzoVQqYbfb0dnZCYfDgYmJCXg8Hnz22We4desWIpEIstlsUyKVrwoejwetVssck+npaZhMJvB4PBQKBVYQATzf0wqFAnK5fJ+RcJJ40/dFKVepVAqFQgG73Q6bzQar1Yrp6WloNBrI5XJIJBJmyMhkMpaq/N73vofFxUUsLy/j9u3bWFlZQbFYbMTQXgB3/xw0bj6fj76+PjgcDnR3d0OtVkMgEEAmk+H73/8+vv71ryMWi+HJkyfY2tpCMBjExsYGYrEYc6BaDTpTX1a4IBQKGY91ZGQEU1NTGBoagkQigUQiYY4HlxtcqVRgMplgt9uRSqVw5coVLCwswO1248mTJ1hfX2eG66vijYyd44CsO+5iNRqNAMBemEgkglAoZCkR4Ll+RyaTQTabRTweRyaTaUmE4GXRJbVaDZPJxAwCpVJ54AFChhNdqiKRCCaTCU6nE4lEAhaLhfEEWnGYHuSdSKVSmEwmGI1GqNVqAMDW1haWl5fx9OlTxGIxqFQqyGQyjI2NwWw2w2w2QyaTQaPRQKfTIRQKYWZmBoFA4MRTdYfhZXNI0Tu5XA6Xy4XR0VFMTk7i4sWLLPQKYJ8hnslkEIlEkM/nodVqkUql0NHRgeHhYej1eigUCqjVang8HgBANBo9EY7E60CtVsNsNjPjXC6XA8C+SkGKfkgkEphMJnR2dkKv17f4yY8PsVgMvV7PeCsWiwUKhWKf4jPxOegPpfY0Gg2MRiP0ej0ymQzW19ext7eHYrF46ojudJZ0dXVhcHAQU1NTMJvN4PF4iMfjcLvdyOfzAACj0bhPb6oZkfNX4QgeBKoC7evrg9FoREdHB3p7e9Hd3Y2uri5MTExAo9FAJpNBIpGwSrNyuQy5XA4ejweXywWNRgOxWIzNzU2WwjvJsR9m6IjFYnamqlQqiMVidkc6HA709PQgn89DLBajo6MDm5ubzNAuFounwtgBXn6mikQiaLVaWK1WjI+PY2xsDGfOnGEO5GFipLVaDTKZDHK5HKVSCVarFSqVCjqdDvF4HIFAgFWJvipO3NghDoTRaER/fz/Gx8fx8ccfo1arIZfLIR6PQ6FQQKlUwmAwQCAQoFqtIplMwuPxwO124+bNm1hcXEQikWAM/9MAHo+HgYEBnDlzBleuXIHNZmNEKy5o81UqFRbF4vF4kEgkGBwchEAgwNbWFn79618jEAicigVNocaxsTH09vZCoVAgn89jbW0NHo8HsVgMpVIJCwsL8Pl8kMvluHHjBs6cOYOOjg4YDAaMjIxgYmIC//zP/4w7d+5gc3PzVIytHgKBAHq9Hg6HA3/3d3+HkZEROJ1OSCSSQw9qqVSKjo4ODA4OsmhcfaVNX18fZmdnUSwWEQgEUKvVkM1mmzWsA0Ge5blz53D9+nUolUqWYs7lcohEIsjlcjCZTJBKpWydqtVqZvC+DdBqtRgbG8PXvvY1WCwWNk4ArKw+k8kwRyqdTsNisUCn00Gn00Emk6GrqwsffPAB5ubmsLe3h2g02uJRvQgy0D744AN8/etfx3vvvQeFQgGfz4fl5WX84z/+I7LZLBQKBb7zne+gr6+P8SGaxcF6XaOConMdHR34q7/6K5w5cwaDg4NQKpWMB1IfkaRqIdK8obPWarUiEokwEj6l8JoJsVgMg8GA7u5u5hRy+ZxknGm1Wly7dg09PT2w2+1YX19HKpVCPp9veArupKDRaDA6OopvfOMb+Pjjj1kg4GVZDzL8lEolgOfvxWQyobu7G6VSCevr64wT+qo4EWNHLBZDrVZjcnISfX19GB0dhcvlgl6vh06nYx4iGQGkCcENZSmVSnR2dmJychKXLl3Cj3/8Y8zOzuLhw4enwjsmHsjg4CBcLhe6urr2RaaArzYgpT2y2SwkEgkLuQqFQiiVSnR1deH8+fN48uQJkskk0ul0y71HsVjMvI5arYZYLIbV1VUsLi4iGAwyz5A0Sj755BOsra2hp6cHg4ODsNlsMBqNMJvNeP/999HR0YHbt29jaWnpVIyPQIdLd3c3pqam0NPTA71evy+i87J/Xy9TQF8tFgvK5TLy+TzS6TRWVlawurp6YmN5GeggIc4DHTwU4SCOVSKRgEKhYPuyXC4jGAxiZ2fn1EU2DgJdkmazGS6Xi1WYAWBp6WQyiYWFBdy7dw9utxuZTAY9PT1wOp347ne/C4PBwKKvH330EXp6eqBSqfDll18ik8mcGqOdiLoU4ZHJZAAAt9uNR48eMQNNJBIhGo3C4XBAKpWeerI5j8dDR0cHhoaGMD09jatXr8JqtUKv17NzCXjO0dnd3UU4HMb6+joAMGLyjRs30NnZyX6ezqpWKanz+XxIpVJIpVJGzK3Vnsu6FAoFZDIZyOVyyOVylq7TarUAgIcPH2JhYQG3bt1CoVA4FXfgQRCLxdBqtXj//fcxOjqKqakpFj3m3o3HBXGDjEYjPvjgA6yurkIul+Px48evHJ1ruLFDfADyHh0OB4aHh5lFTdY4t4KF8srpdBr5fJ5Z3B0dHVAqlcygiEajmJmZORVKscTjsFqtrISwPgeezWaRTqcRj8extbWFaDQKjUaDrq4uGAwGmEwmiMViqFQq9PT0QKfTQaFQoFgsnniY9WWgNJRarUalUkEikYDP50MgEEAqlWLzRwYrhRfD4TCSySQikQi6u7shFouh0WjQ19eHXC6HUCiEUqnECHmtBKVp6JIbGBhgG/M4/Z7ImOWuR+4a4O6F7u5uhEKhEx3Py0DGNVcDiiqv0uk03G43NjY2EI/HGQ9NpVIxaQUyGprNK3sVUMrcYrGgs7MTHR0d++aEqj22trbw+PFjzMzMwOPxsDLXfD4Pu92O0dFRaLVatj6IOLm4uMjSmK0GEbA1Gg3MZjPjfxSLRezu7sLj8SCTyUAqlbK0ATkxXIPhtIE7hw6HA2NjY7BYLNBoNMxAyGazyGazCAaD8Hq92NnZwcrKCoDnZ1KxWITT6YRYLIbZbGZnzklUoL3KuGQyGUu5keOQyWQQDoextbXF7gXad1QkUqlUIJPJ4Pf74ff7kU6nT1WWA3g+Po1Gg8HBQSZXYrPZIJPJjuSHkcHHlSzglqkTpYLWQzQaxebmJqLR6CtRWxpu7JA3dePGDVy8eBFGoxFarRalUolZ1SKRiC0+uhgjkQjcbjdCoRAKhQIjRpJ2jcvlQigUglQqPRU9OWQyGWw2G1wuF+M+cC3XarXKNuDKygoeP34Mj8cDu92OS5cuYXR0lHmPWq0WTqeT8XbK5TLi8XjLCK08Ho8ZAL29vSiVStjd3cXS0hJ8Ph9isRh7/7RQM5kMCoUCQqEQY9Pb7XYAz1MKNpsNDoeDaUlQOWgrIRaLodPpcOnSJYyPj+PcuXOwWCwsfXMU6nkftFG50T36/VS6v729feJjOgpyuRwWiwUTExMYGxuD0+mEQCBAOByG1+vF73//e6ytrSEej6NQKODatWtwOByQyWTo6emB3++HRCI59Z6lXq/HmTNnMDw8DJPJtI+UnM1msbi4iIcPH+LnP/85dnZ2kEqlWMQnHo8jFovhz/7szzA0NAS73Q6j0YhKpYJsNov5+XkAOBXGDq2vvr4+nD9/HjabDbVaDalUCl6vF8vLy0gmkxCLxZBIJIxLRvIXdNaeBN4kAkjE+LGxMZw9exbnzp2DyWRi0dZ0Og2v1wufz4dHjx5hdXUVgUAAoVCIGeLVahWDg4MsKpDJZNgZ1SpjnQpTtFotS1mVy2Xs7u7i2bNn+Pzzz9Hf34+RkREMDQ2x4IDNZoNer4fL5UKhUMBnn30Gt9uNWCx2qqJzUqkUPT09+PDDD3Hx4kVYrVYW5DjsPKWiDyovp3NFoVAwB4uiljqdDsPDw8hms/D5fMhms69U+NIwY4e8homJCYyMjKC/vx+pVAqhUAiJRALb29ss32gwGJDL5ZBKpRAOh1EoFJDL5bC5uYl0Oo1KpQK9Xo9oNIqpqSk4nU7YbDb09vZCq9WyiohWTnRHRweuX7+O3t5eZolz86+lUgmbm5u4desW7t+/j+3tbcTjcWxvbzMDYXJyEiqVClKpFGazGWNjY6ycuVWgUOvk5CTGxsagVCqxsLCAhYUFxps6qJKByObEh8jlctjZ2UE0GsXVq1cxODiI/v5+fPjhh+js7MTNmzeRTCb3VYo0EzweD2azGTdu3MAHH3wAh8MBp9PJvJCjQEKKmUwGXq8XABjxkKIllN6isC6RKFsFUtU9d+4cLl++jN7eXkilUlSrVWxsbOCLL77AgwcPWJknHS4SiQSTk5M4c+YMyuUy/uu//uvUlcESlEoluru7MTk5iffffx9DQ0Po7Oxk81kqlVgl4f3797G6usoOWB6Ph2QyiUKhgN3dXQgEApw9exbf+c530NXVBZPJhOHhYYyNjSGbzTLieasgEAjY85w/fx6Tk5NQKpUoFou4c+cO5ubm4PP5UCqVIJfLYTKZ4HA4YDAYUK1Wj+SjNQKva0wYDAZYrVaMjIzge9/7HnO4JBIJMpkMYrEYfv3rX2N3dxehUAgPHz5EOBxm94ZYLGaioFQUw+PxEAqFmiaMeRioVJ6icHw+n+k7ffHFF7h9+zZmZ2dx584dXL58GXq9Hnq9HpOTk5DL5bBarfiLv/gL2O12zM7O4pNPPmEGa6sjrXw+H3/wB3+As2fP4tq1a3C5XFAqlcyo44KoKolEApFIBLFYDF6vl0WrJBIJbDYbOjs7MTw8vI/QPD4+znTdYrEYgsHgseUwGmbsCIVCaLVauFwu2O128Pl8bG9vIxwOY3t7G16vl4m0qdVqFItF5HI5JBIJVgUSDodZeqNcLsPv98NkMiGVSkEqlUKlUr127q+RIN5DZ2cndDrdvmeiicxkMtjZ2cHq6iozdMgKjcViiMViSKfTTGOALktKbR0njXISIO+jp6cHnZ2d4PGet7jY2trCzs7Okbn++rROpVKBx+NhXpnD4UBfXx8KhQKWlpZYOqAVxo5EImFesdVqRUdHx0sNHRoTGemRSAQrKytIp9OQyWSYnp6GQqHYp7FEPBkKy7YKZrOZRes6OzuhUqnA4/GQzWYRCASYd0xRnVQqhVQqhWw2y8LTHR0dMBqN+7zjVh+yBD6fzyrMSD3YYDCwdM3e3h5yuRyWlpawvr4Ov9+/75IgB4XGtb29DYPBgGAwiM7OTkgkEmi1WqaK3kqQeGBPTw8GBgYwNDTECgjC4TAWFxcRCASQzWZZdYtGo4Fer4dMJgOPx2t5mvwg8Pl8dHV1ob+/H0NDQ+jv72dnRyqVgsfjwfr6Op4+fYpQKIRoNAq/38/afHDVeeVyOfR6PWuVkUqlkEwmWQ+0ZoMqPqlViVQqZc8ViUQQCAQYcZzEWtVqNXQ6HUQiEfr7+2EwGGCxWDAyMgIAWF9fZ7IojVIZfh2QY+R0OuFwONj5Um9Q0x4jsdmlpSV4vV4Eg0HmZO3t7UGtViOdTqNQKMBqtUKr1bKonlqtZnIoBoMB6XS6+caOWCxGV1cXzpw5A6PRiFAohPn5ebjdbnbhkwclFAr31erTiyAQoz4UCiEYDCIUCkGn00GpVEKhUDRNH+IwkLFD5ancw490WCKRCDY3NzE3N4fd3V1mJJAiZDweRzQaZZUfRGi1WCz7cpzN3phkdFFabW9vD16vl4k8HfeQpOozv9+Px48fI5vNYnh4mHlpy8vLbKOSuGIzoVKpYDQa0dvbi66uLuh0uiMrBcgALxaL8Pv9WFlZgdfrxcLCAjY3N6FUKqHT6WCxWFgpN3A6VMeprHV0dBTj4+OsEqRSqSAajcLtdmN+fp5dkABQKBSQz+dRKBSYkJtOp0Nvby+rxGulWCQXFEEjfseZM2fQ19fHoq21Wg3FYhHxeBz379/H8vIydnZ2Xnh2Ki0HgFAohO3tbXg8HoyMjEClUkGlUrU8QgeAVbcODg5idHQUo6OjAIBAIMBSdFtbW8jn80yKX6/Xw2AwMOHLTCbDzmOg9b0OaQ4HBgZYStlut7Ny8q2tLdy7dw+3b9/Gs2fPWMaA7g+uPIREImEXIjlsdN4mEomWRCW5pHmNRsOiqru7u+yOo7ZB1GOQqpIoACCVSmG1WjE8PAy5XA6Px8NETbPZbNMa9nLB4/Egl8sZN9fpdLJKzvrq1EqlgmQyyeRkPv30UywvL8Pn87GoG/DcMUskEsjlcnC5XJDL5cyBlEqlTFLCYrEgFovB7/cf61nf2NihcL3BYMD169dZBcrGxgZmZ2extbXFeCg0EfVNRetBF0sikUAgEMDS0hIuXrzIGlJ6vd6WapYQW57CdKRSSpO5s7ODX//613j8+DGi0eg+fYRqtcoiA7u7u+ju7mZpj6GhIVYJ02xjh5p76vV6XLhwAV1dXRAKhXC73XC73fs8xeOiVnsu/x0MBsHn83Hnzh0YjUZYrVb8+Z//OcRiMebn5xGPx5s+l11dXewwJBHAowh0e3t7SKVSiEajePDgAT799FMsLi6yi58EFYmzw1WlpshlK9YrlciTEXDu3DmIRCJkMhlEo1H86le/wp07d7CxsYFsNsua1lI6kowaqVQKnU4Hu92OxcXFpo/jKFBLk0uXLmFsbIwZJ3RAFgoFxGIxbG1tYXZ2FsFg8MASXnJGeDweM/RovdMl2mrjVSAQQK1Wo7e3F8PDw7BarRCLxVhdXcWdO3fw5Zdf4u7du2yvisViKJVKqFQqNhZyME4T34MbqRoeHsbZs2chFAoRCoXg8/nwn//5n5iZmWE8JHKmCDQ2hUKB3t5efPe730VPTw/EYjGy2SzT1yFeT7NB1btccnI2m8WjR4/w6NEjLC0t7dNYCwaDLCqcyWRYWwy1Wg2ZTAaHw4G/+Zu/wfb2NsuSNPO+IOONUnM3btxg1axKpXKfoZPL5Rht5d69e4jFYohEIvjZz36GSCTCFMpJOoBa0ySTSVayTuKuJADa3d0Nu92OYDB47GduSGSHDgK9Xo+9vT3E43F4PB7s7u4yXgYXx5kQYtRnMhkkEgm2cakktpUpAYlEAoVC8cKzkNx1JpPBo0eP4PV6XxCCIi+TROkoUkKTqFarmVXcrIOV21JBJpPBYDCwruabm5vw+/2Ix+OvtZEodJlMJpleBEn02+12RKNRqNVqJJPJEz186SCgyGJnZyeriqM8/0HPzi2vj0Qi8Pv9WFhYYOu7UChAq9Wyrtp0kNHcURkshWWbDalUis7OTnR2djIhM8qXb21tMUOH6zzQnFH1Ch1ElKpWKBRMq6QV3iQXtPbiCbEAACAASURBVG8cDgdcLherauSW1ZdKpX1l9cepBKT9cByyejMhl8uZ6GFvby90Oh1KpRK2trawvr6OlZUVlhqmfU3vgKs/o1arWWTkNKQj5XI5ent70d/fD6vVCoVCgWw2i62tLTx9+hRPnjyBz+dDMpk8MLpMYyXR2vHxcahUKkYADoVCrNdZq8jJlJoSi8Wo1Z73EIxGoyy9xj3/KCBAtIf19XXo9XpYrVZYrVamTDw5OYl8Po9YLMaI9s0Yn1AohF6vh1qtRnd3N86ePQuz2cwcdeCrIg5Sm9/e3sbt27dZ/7nd3V3WG5Krk5TP55FMJhEKhbC2tob33nsPlUplX080KtGnQMOxnvlNB03WJ5UYZzIZBINBrK+vIxQKvbaIGhkFuVwOmUwGtVqNCWi12tgho4RKPekwJKJ1Op3Gw4cPWZk1FzSuVCqFYDC4b/PJZDIm5U4VJM0CHe5yuRwdHR2o1WqIx+NYWVlhFVivC6rWWl1dRSQSQU9PDyOch0IhplR7UsYOGTj0VSaTwW63w263o6Oj48AyXNqoFJkhcvnq6iqePHmC7e1tpFIp8Pl8FjI3m80v8K2oTxYdaM0GVVLZbDbGS0qn04hEIlhfX8etW7cQi8X2OSRUmUSprL29PfD5fHbAEXeO3k0rL0oSvxweHsbQ0BCsVisjf1KlHJGOqc/Vy4TZqNRVLpczL5XLRWvleLVaLUwmE7q6uuBwOCCRSJgjsbq6ytSeaRx8Pp81mSRjRyAQwGg0MmO91W1MeDwelEolnE4na+shEokQCASwtraGu3fvYmZm5qXRKOKWWSwWjI2NQaFQIJ1Ow+fzIRgMIhaLtYzMS2erTqeDWCxmfEXqsXdY+wpytNbW1iAQCNDX1we1Ws3W5oULF1CtVvd1sT/p8VFwgxyowcFBXLhwgWkgESgiHgqFsLq6imfPnuHmzZuIxWLMyKmXASBHK5vNIhqNYnV1lUV+qCqL25HhoB6Uh+GNjB26PCi0Ojk5Ca/Xy4ycNw3dk4Q/GRWkmdAqjQiyKi9fvozz588zbR0ArOTT5/NhdnaWbax6VKtVJvRFSsT5fJ7pKlBOUiqVslL9kwa3pUdnZyc+/PBD1Go11hiRq330OqDoxs7ODubm5piuUH9/P2KxGGZnZ+H3+09MHVQikbDyXKfTiYmJCcbZ4faFItAlmcvlmGH6b//2b/jyyy+xsLDA0j1CoRBWqxVTU1Osy33976L0VytSBiQ5PzU1hYsXL7Ku5uSMzM3NIZVKHWiQFwoFRKNR7O7uIpVKQafTQaPRYGJiAr/97W+ZcdyqfUiRz2vXrmF6ehrf+ta3MDAwwIoFqB8aOV+zs7P47LPPXiqmR9WIExMTmJ6eRl9fHyQSCft9rZS9EAgEuHr1Ki5duoQbN25ArVYzSYif/exn2Nzc3KcsS0aEVqtlhRTUBoT7DloduaJigf7+fvT390On06FSqeDhw4e4f/8+ZmdnX1pFJRQKYTKZcPXqVVy5coVVnVGfqWAw+Mpp+Eaiq6sLNpuNRR0zmQy2t7exsrKC3d3dI/9tpVLB+vo6YrEYkxEAwDieTqcTQ0ND7Fw66QgyFQJ873vfw/vvv4/u7m6WDeBGtPP5POLxOL744gvcunULv//979lZ+DKngRxDt9vNtNvUajWEQiGEQiFUKhU6OzsZAf04eOPIDh0OFFaKRqMIhUKMAPcmICvP7/ezdFArhZR4vOeNInt7e9HT07NPfK5WqyGRSGBjYwP3798/ksi7t7fHQrQ7OztMzp4sZiIti0SipqQ+yGKmig2dTsdEEMnrf9PLmuZuZWUFBoMBFy5cgE6ng9FohEajObFIHdeQm5qawvj4OPr6+lgotD6CRuKW8Xgc4XAYS0tLWFhYwN27d+Hz+ZBOpwGAcbaob1S9eF39uMlwahb4fD7rRTM6OsoE2crlMjY3NzE/P4+ZmZkDDUwKsROpM5PJMF6T1WpFZ2cn63XW7MuDeAt6vR4DAwP42te+xiI6xGuo1WqIRqNMcO7WrVt49uwZNjY2jjyTuL2LSHqB1LSJYAm0xjigtAWJmJpMJtbCwuPxYGdn50Blcur7VavVWIsF8o650ctWgcfjYXBwEBMTExgeHmaiqrT3Njc3sbu7e+QzUsSfKvH6+/vB5/OZsOLi4iLC4XDLtJH4fD5sNhtLsZK+Tj6fZ0KWLwOdSxsbG1haWmLrQa/Xw2azYXh4mBm7J31nUAso2odcp58Qj8cRDAaxsrKCe/fuYW1tjckDHOfMIN5uNpvdF5nn/iF9s+Nyld44skMGgEKhgEQiQSwWQzgcZlbmm6JUKiEYDDLjgSayFWksOgy7urpgNptZPp+s1HQ6DY/Hg0ePHh0ZpSA+kt/vRyAQQCwWQ29vLzN2SDn6uCV1bwoyCIh7IpfLGRmXpAHe9FKjxbu2tgaDwYBUKsWMq3pCWyNB6RelUonh4WFMTk4yCfb6Vg/A88shmUxie3sb6+vr+M1vfoPf/e53yGQy7OepmzY1JbTZbDCZTC/wrGjM+XyeVRc0A2S8njt3DtPT0+wSAZ6nWjc2NrC4uIi5ubkDI3a1Wo3JQhCBsFqtQqFQwGw2w2KxwOPxNF1rhg490oz5+te/jvfff581KaWQdqVSQTgcxtraGp48eYJ///d/ZxHUo0CVS729vezy1Wq1EAqFLK3H1eVppqFHnBZaayqVCtFoFOFwGB6PhzWl5YKcRW5zV/pD6YZWqtFTdHBkZARnzpxhHbFjsRh8Ph9WV1fh9XoRi8UOfUZKOep0OgwMDMDlcqG7uxvVahV+v5+l96jnWyvA4/GYwJ5Wq2WpReJ3HscJqlQqKBQKrOeZRqPB+fPnoVar0dXVhaGhIdy5c+eN6AbHBWVclEolZDLZC7yZarXK0uT37t3DkydPsLu7+8qRey5FhniV3OrBV20I+kbGDh3mVqsVdrud5SIbkTeki1+j0WBgYABSqRTpdJpdOq3Q2imXy/uMAMqRUqk8lRy/rPcTt+yeK6pHYefBwUE8efIE6XS6Kc0HKfTZ3d3Noh7JZJKRhht1GJJ+yfb2NvL5PFOQBk7OW1ar1bDb7fjud78Lh8MBjUZz6Nrh6lVUKhX85Cc/wfLyMjKZDLsglEolrl27hsuXL2N0dBTnzp1jHnO94VQqlRCNRrG4uIi7d+82TUFZJpOhu7ubdYam9FU8Hmee4dbWFiM01oMOkp2dHcjlcoRCIfT+r6Antf4g9dpmXZQ83vOGpGazGX/5l3+Jc+fOYWxsjHWOpnVEhqXX68XKygpmZ2cRiUSOvFBIA8VkMmFwcBB/+7d/i8nJScZxonTIzMwMHj9+DK/X21QDgYQpHQ4HLl++jO7ubtRqNbjdbszNzeHRo0cHRpLJOSQDjTxhiUQCh8OBhYWFpo2hHjwej7VCeO+99zA5OQm9Xs/01e7fv4+1tTVEo9GXGjpOpxMjIyP4wQ9+gP7+fgiFQkQiEfziF7/A48ePsbOz01LV71qtxmRT6O8khfAqFAGqUJ6ZmYFQKMS3v/1tSCQSKJVKpkkUCoWwu7t7YuuTz+fD6XTiwoULuH79OnQ63b7/T4KrKysrePr0KRN9fJXGnRR1dzqd+Id/+AcMDg4yRWWK6KTTaRYVO+5YG5LGIlItXfrc0Onrgrw4mkiRSIRcLsdyr63q/kphX27OkRtWqy+JPAzkodanUqh9BDUKbRa4c1WtVhEOh5kyaSPD3GSoco0cipCdhLdMYn7VanUf16MeZOiEQiHs7OzA5/MxkTMiICsUCqjVakxMTGBoaAjd3d2sm/ZBBOdwOMw81ONEFhoBirQ6nc59Yom1Wo3xBNxu95GXSP04KIXMVYamUPZx1/ubQiaTQavVMu+9r6+PCa7RnFLqzefzYWZmBs+ePcPq6uqhlwlx5ORyObq6uuByuTA8PIyRkRFotVrmsZKWx9raGra2tpoWcSWQIGRvby8jgWazWczNzWFxcREej+fQOaiPNJK4IlWC0s8003ijNWoymXDp0iW4XC5mkIfDYZY+JyHWg0CVuaOjozh79ixGRkbQ09MDgUCASCSC2dlZLCwswOfztVSmBPjqvVcqlRccrVfZP7T3yFmmO5aiPlTNROT8RoMixrRWDhL4peeh8VIV58uCE9wojtFoZA3EJycnmYNKUh6FQgGRSASpVIppYzUljUURGG5jM5qMN/m9QqEQcrmcdaSmNBFZia0wdmhshxlzFKV52diJQ8LVECAQB+qgSMFJgTtn1KuELOdG6nGQB00kO/q9JzlWupipgu+wCBIZO4lEgpVGUjNUhUKBsbExaDQapnFC5Lh6zg+9y729PVZu6fV6mcprM0C8MovFwgwCqnAIhUIIBAJIJpNH/g5K2daX0gNgVT3NqhikKIDJZMLIyAjsdjsTRqQzh6pbqM/X4uIi1tbWsL29va/pMD0zN3Ks0+nYwTo4OIju7m5mgJOyu9/vZ8UXxNtqBng8Hjo7O1n1IClfJxIJrKysYGNj41BOC11OxNEBwIwdEqNrRQqL5AIsFgumpqbQ3d3NIgSkwZJIJA6tUiKRy46ODkxMTODcuXMYHByEwWBAMplEMBjEo0ePmK5Oq5sOc9OJdAbRGjxu5JybQlcoFMyBobOGJF7I4Dkp0DNXKpUDCxTombhq1qSRwzVKyLihd0FVsiqVCk6nE5OTkxgdHUVvb+8+bmylUmHGOmVXmmbsiESifZeJSqVi5eGv+9IFAgE6OjqYkmZfXx/m5uawsLDAqpxaaamHQiEW2qfFShZnuVx+qRVLlUAWi4Xl4AlEzgPQ1FQdeQf5fB75fJ4JkTXqQqPQpMViQVdXF9RqNfss7qJv9KFEonrvvfce9Hr9gUYV11MCnnd8J9lzakZH+h06nY55/QdF3oh8HgqF8OjRI8zPzzO9pWYcuBRpJaJmZ2cn+Hw+499EIpGXkjW50aGzZ89ieHgYKpUKAJgq9s7OTlPKznm85wqtw8PDmJ6exg9/+MN9hg4ARmScnZ3F2toaZmdn8eDBA0Sj0X36HGKxGH19fewApqidxWLBe++9xwikVFZP8/jkyRMsLi5ifn4e0Wi0KRE6GrtIJMLk5CQuXryIq1evQiaTIRQK4dmzZ6y58GFGtFAoRFdXF7q7u2Gz2aBQKFjj5fX19Za0TiBDhc72S5cuwWw2QygUIp/PY2lpCYuLi9jc3DzwnOfxeNBqtRgcHITL5cKlS5cwNTUFq9WKSqWCpaUlPHv2DJ9++il8Ph8rlGm1jlA8Hmf8R+Arh5cc25etKZJ2OXv2LKampjAyMsKq1kKhEO7cucOKBk7qbuRGbfb29g68G8hwsVqtSCQS8Pl8CAQCrEUElZDTz1CghHSjbDYbJicn0d/fz7irZCBWq1VmzK6trWF5eRmBQODYzvgbc3YoElCr1VhumYTHXocsRYf12NgYLl++jOHhYRSLRUZ0IvXMVlYQJJPJfSV0wFcXBFWlHcbbEQqFMJvN6O/vh8PhYEJvwFfEUJIBb2b1AOX3KQdqsVhY07pkMvlGBwZZ8UqlEi6XC06nEwqFggl9venvP+pzNRoNzGYzOjo6jtRkoMvQaDRCLpczRVKhUMjE5Uhfhqp+CLQPisUilpeXsbu7C7/fj9///vdwu93wer1NiUTSe+bKF1DVTTgcRiAQwM7OzgtCl1wIBALIZDIMDw+z7uikDE0GAB3czRAUlEgkGBsbw8WLF3HmzJkXdJHI0KHU487ODmKxGDQaDSQSCcrlMmvzoNVqYbfbmbFjNBqZs8H1IKPRKNbX15lx+MUXX7C2N9TipBkgwnRfXx+L6sTjcbjdbjx58gR+v//QKBMJ2PX398PpdMJut0MoFCKRSLDWBK1wGOkcIH4g9b4iAdOlpSWsrq7C4/G8kH4kfsrk5CQmJyeZ0rJWq0U+n8fq6ioePnyIubk5eL1e1ouw1YYOABYpp/UjFouh1+thNptZH7qDQNFwm80Gq9XKNG2cTieEQiECgQDW19fZuXMYD69R4L5PbqSKQJH0jo4O2O12dlaYTCbk83mUSiXodDrWt4zEFi0WC/r7+1kAgFpq8Pl8pkCfyWSwtraGzc1NFrV9lZTyG5NCKIy2t7fHNiextF8nNSEWi5n8OzUtpJDt2toastlsyzsu08KlS4Mmn1uOF4lE9vF6KKxH7Pne3l7Y7XbGzwG+EqALBAIsRNcs8Pl8VnFSq9Wg0+lgNpvR2dkJn8/3Rt3JyUMl/lVPTw/4fD4rbSZjp9Hg8/nQ6XQwmUxMo+Gw56PKELVaDYlEwtQ5qYmnSCRiRGRuxI3SX8ViEclkkh20gUAA8/Pz2N3dfW316VcFlw9FDUkpx5/NZpFMJhGNRg+MyHCVSXU6HfOcqceUUChkhgU1CG3GRSIUCtHb24vBwUE4nc59IW0ArNotHo+zy0QgEKCnp4e9k46ODvaHBOuEQiE0Gg2sVivrb0dRue3tbczMzGBnZweRSATPnj1DMBhEOBxu6tlD82E2m2E0GiGRSLCzs4NgMIjNzU0kEokDy4zJYTQYDOjr60N3dzeLHlOErxU8FlqfJGFgtVqhVCpRqVSQzWbh9Xrh8/ng9/v3ReWIy6FSqWAwGDA+Po6xsTG4XC5YrVamRLy8vMwavcbj8VPVDoMakdK+obNGq9Xu66XHBaUgNRoNenp60NfXh4GBAcbHK5fLjINHqemTLjvnZjKKxeILFASaY7VaDbPZDIfDwZSSyU4guQ7qASkWi5lzbTAYmBQJcYCpMCkSiWB5eRlut5tp1L1Khd0bR3ZqtRrr4g08T9FotVqmYvoqqQmBQACTyYT+/n5MTU3BZDKhUqng8ePHWFlZgd/vb3lPl2q1yrq0BgIBOBwO5mkaDAYMDw/j+vXrCAQCzLMAnnslKpUKFy9exNTUFIaHh3H16tV91UHUC8zj8SAajb62+vSrgipdUqkUa81hMBjQ29uLM2fOwOPxsJ5er3q50e9WqVSwWq14//33MTg4iEQigbm5OWYcnESUQCAQYHBwEFNTU+zCPgy0VtVq9QscFRpHPcjQDwQCCIfDWF9fx7/+679iY2MDxWJxX9j6pEFRKAqNcxvx1WrP1cepTUT9/qFIR0dHB4s4fuMb38Dg4CA6OzsZKTGfz2N7exuhUGgfR+AkIRaLYbfb4XK50N/f/4KhSX13SqUS669kt9uZYUQVdDKZjJGcKa3FDcNXKhXs7u7C6/Xik08+weeff47t7W2Ew+EjI2EnBXIQqEUAV9Wczoj61CiX12G32zE0NMTmkeQeNjY2sL6+/sqaT41KZZPRMjU1hampKQgEAqTTaezu7mJubg7Ly8vw+/0swkxNTEmh3Gaz4Qc/+AG6urqg1WpRrVaxvr6OxcVF/OpXv8Ls7CxCodCpMnQAwOv1Qq/XY21tDV1dXWxuuamaem5ZV1cXOjs7GXHe4XDg2rVrUCqVKJfL8Pl8uHnz5r6Kw5M2YImzSqKj9el8MnYUCgXrWN7Z2YlisYhyucyI5USTEAgETCSQ9iUAli7LZrN4+PAh4z7evXuXGfyvKtb6xpGdcrnMeAAUFaAXclweBm3srq4u1jlVr9czbsCPfvQjeDyelhs6wPPJ3traglKphNvtxtTUFLtcdDodXC4XyuUy7ty5g3g8jkKhgGKxCLvdjv7+fqb0SrL21AGeVDJXVlYQCARY5KhZYyqXy6zqZHNzExaLBT09PeDxeIjFYlhYWMCzZ8/YHBznkiMSKPVP+da3vgWXywWJRILFxUUsLCxgbW2N8T8aCfJuyfM/TpTxVSQNiOPkdruxuLiI5eVl/L//9//g8XhY64tWeM4Utai/oJVKJQwGA2w2G7q7u9naqtVqTO9oZGQEY2NjTLafoly0NgKBAB48eMCKBE7a0KGDUCKRvHCokndJl3tnZyfTOhIKhWxv0d+pKq8+lUl9y5aWljA/P4+VlRX87ne/QzgcZkZUKwwdrmdLDVkrlQoikQji8TiKxSJUKhXrnl0qlZjafE9PDy5cuMC4HUqlEoVCAfPz83j8+DGWlpZe+ZkaNdcikYjpIikUClZCz43IUMRAp9Ohp6cHLpcL4+PjsNvt++QxqCLt/v37bGzcTuinCSRUGo1GUSgUGN3BarVie3sbGo0GuVyOcQRNJhNGR0cxNDTEeE0kDBoMBhEIBHD79m3cu3cPHo+nqY4VNREOBAIwGAwvRKYo9UYVmwqFglUxc/cjVyiQW9FK/bQoZfvLX/4Sm5ubCAaDLODxOunkhkR2iGhKHZPJEzzIQz4IFHanhdzd3Y1KpQK3242VlRXW96PVhg6BupaTXgxdarRIK5UKLly4wFJRqVQKTqcTTqcT4+Pj6OrqYj1SgK9Udj0eDzY3NxGLxZpipXNBKQoKFXZ0dEClUqG3txeTk5OoVqsIhULw+/3s4D1K/4LbL81ms7EIi16vRzqdxubmJnw+H3Z3d0+kLxZd/OQh0PcaATJOuTo6T548wdOnT1u6Rmk/0nPRRU2Gn9lshsvlQjKZZCWblF83m80YGRnBwMAAuru70dHRwUiBRDTc3NzEzMzMC720Tgp06XM7HteDqjiI+Eocq8OI9dzKQ2o4GA6HMTMzgydPnmB9fR3r6+stT5UDX4mP5vN5FsUh3orBYGDfp7VoNBrR1dWFiYkJRiynn4vFYlhbW2Mpj1aA5lMqlbKUMACWGo3H44z4qtVq4XQ64XK5MDY2hrGxMdhsNsYfi8fjCIVCmJ+fx/z8POOsnJY7oh7cJtGkzSaRSJgOFrU9oDns7+/HyMgInE4nxsbGWPXVzs4O4zQ9fPgQbrcbkUikqT2/qG9VIBCA0+l8oQKMzn9KqUul0mP/bir0oezJ48ePMTs7i62tLcTj8TeisbxxZIfKUAGwjtmUcztO525i13d1deHq1atwuVxQq9VYW1vDj3/8YywvLzctnXNcFItFVm3zp3/6p8xLEYlEMBqNMBqN+Kd/+ieUy2WUSiUEAgGo1WoWTuce3BRVSafT+PLLL3Hv3r19itHNAKVjyuUy3G43fvSjHzGehMlkwkcffcREI3/xi1/sy5XWPyNXH0mtVmNgYADf/OY3MTU1hcuXL2Nvbw9utxv37t3D4uIiIpHIiV4spPtEl1wjDB5uC4h0Oo3f/va3ePDgQcvLWyuVClKpFOsuPD09zYi6dKAODAxgfHyccexMJhPjhNTzYYgUODs7iy+//BKzs7P47W9/27RO5zRf1NOJG3XjrrNX/Z3EsfJ4PPB6vdjc3MRPf/pTLC4uNo1f9bJnpDVGsvskEHnhwgUYjUao1Wp88cUXTItGJBJhamoK09PTuHHjBpRKJXM619fXsbS0hFu3bmFxcRHBYLAl46L7oN75UCqVkEqlKBaLkEqlrKv7D37wAwwMDKCvr48VGPD5fCZa9/jxY/zmN7/B0tLSqTZ0gK/WHRk75AxeuHCBkbOlUin+8A//EH/0R3+ECxcusEotHo+HVCoFv9+Pf/mXf8HNmzfh8Xhadi8Gg0HweDw8fvwYExMTMJlMDTtXi8Uidnd3cffuXTx48AD/8R//sY8O8iZoWNdz4LnFR/yVSCRy5KFIh5Xdbsd7772Hqakp2O12BINBzMzM4ObNm/B6vU0l6b4KaFx3795FOp3GuXPnDgydi0Qi9PT0sL/Xh+uovPXzzz/H7OwsfD5fS/QvyBhIp9NYWFjAj3/8Y5w9exbvv/8+9Ho9HA4H/viP/5j1rKnvM0Ry9HK5HP39/bh8+TJGRkYwMTGBjo4OVqF27949PHz4EF9++SWLPpwESHdle3sbm5ubcLvdsFqth5IBjwIdVMRNW1tbw8rKCjY3N3H37t2mK+oe9ZzU7mJ2dhbLy8sQiUSsEoek3S9dusQMCdI9om7tZDQlk0lsbGzA6/Xi4cOHuH37NjweT9MMHUK5XEYwGGQ6RcdtAkwGAx2gFCHZ2dnB4uIiazBIKfi5ubmWNoqsB9fgWVtbg06ng81mY+rYOp0O09PTLMJaq9VYJaxGo0GpVEIikcDdu3dZ6ur+/fsv1Vc66TEBz8/+bDaLTCbDJADICB8eHt7XSZt0rihKm8/nce/ePdy9exdPnz7F7OzsiUSGTwLUDNnj8YDHe95CoqurC9/85jcxPj6ObDYLq9XKxEorlQoymQw2NjawsLCA9fV1PHv2jFFGWgWSR7l58yYcDgfy+TzGx8ePFdyoR71hH41GMTc3h1/+8pf7aBONQEMkeumApFAad9AHHR4kEqbT6XDhwgWMj4+jt7eXESBXV1fhdrsbrt7bKHAvlYWFBVQqFUgkEoyOjjIvhVJ4VJLOTenRBGezWQQCAWxubuLBgwdMHbqVY6aWGIuLixAKhbBYLHC5XKz88eLFi9Dr9bBYLEw3gcquZTIZlEolRkZGcOHCBQwMDMDhcLDu08vLy3j8+DHm5uaY2NdJjZUMlGg0Cq/Xi6WlJZbu4AoMHiaKRakBuiT9fj8rRV5YWMDy8jITrzsNKQ8uSqUSQqEQ1tfXWcEA9XiiNE992BkAM3ZTqRTjIlEjRWoM22xjoFKpIBaLsRYt5DSQk0XzxY3Y5PN5ZLNZ9pWqDHO5HPx+P5aWlljPpXQ6zcZ8WgwdLs+xWq2yM4JK5VUqFfR6PdONqlcIL5fLiEaj8Pv9ePr0KZ49e4bNzU1Eo9GWF3dQujydTiOXy0Gn0zGBwTNnzrBCj76+PkaO5/F4SKfTTCOIdI/cbveBhPvTCkrPuN1uSKVSFk21Wq0wGo0oFouQy+WMaxiPx9l4FxYWWMPXRkU6Xhd014dCIXg8HpZ602q1BwoNHgRKt1NbIlLsj0QiTBA0GAw2dJwNKT0nwaBcLsdq7zs6Oli4q145USQSsUqfH/7wh+xAu3nzJubn5+Hz+U4lo56LTCbD2gCQ4NgPf/hD9Pb2wmw2swk/iGdAfAGPx4MHDx5gdnYWP//5z5FKpZpGMjsKlUqF9YQi5j0Jk/31Pr5m8AAAIABJREFUX/81M15CoRCy2Szz0pRKJTQaDZxOJ0vXUa+btbU1/OQnP8Hq6mrTKlwqlQqTi6eWJlziLZHk6v8NredwOIxgMAifz4ef/vSnrPLH7XYjmUyeKh4ZgcfjsaKBL774AtlsFiqVCsPDw/vK6Qm0L2kfE0/uf/7nf5g6L6mVNpv4SZyh7e1trK2tMYkG4nsQOZcu0XA4jFAoBK/Xi9XVVfj9fmxtbbFKELpoyfiJxWKMf3bawI3+Upn5/Pw8rl+/zgjkRqNxnzItEZljsRiePXuG+fl5/M///A/jO7R6rdL7p0pWu90Oi8UCqVQKh8OB3v8VaSWHBABrmEkVV/Pz87h16xZroHwa5+4gkCOxtLQEjUaDYrHIojgymYz9DDnS5FQtLCzgl7/8JcLhMFN3Pw0EbHrWzc1NVoQyMTEBtVr9QmPQg/4t3fmPHj3C3NwcEwFNp9PMuWl09Ip3lDfD4/GO5eqIxWJYrVZ8+9vfxocffgiBQIDV1VV88sknTHGUWjxUq1X09fXh/PnzmJ6eRnd3NxYWFrC4uIhPP/0UwWAQyWTyjdMbtVrtWPG0447xkH/LUjc6nQ43btzA4OAgqyhTqVRQKBQQCASMvxONRuF2u7G9vY2FhQUsLS3B7/djY2ODvZ/TMEYqH+zs7ER/fz+sViv6+/tx/vx5xssiTgBdOP/7TKzsOh6PM8Ln1tYWnj17xhbxcdV3jzPGo8ZH1TxGoxEulwsmkwkmk4n1HqO2JBSdSqfTTA+C1mwul8ODBw/YpZnNZhsmbHkSc0gOhtFohNlsxvDwMKampmCz2dDf34/u7m4IBAJWbh0IBFg5MwnzcQ06it69buTjdcdIl57BYIDFYoFWq2V7isrIyYgh3htxCKk8nkTN6Pvc3naNVIA+qb1IkWEStSRlWZPJxJS8BQIBM9ALhQJKpRK2t7exs7ODtbW1V9pvR+FN9yLwvBjFYDDgww8/xPnz5/Enf/InLP1PRhudKVtbW1heXsbMzAxrILy1tcVUoxut4H3SdwaR7V0uF1wuF6anp/HNb36TlZ77/X72x+v1wuv1Ymdnh0kFNKLnZKPGSARrm80Go9EIu92O6elpWCwWVgihUChYlqNYLCKXy+HZs2cIBAKsbc3Ozg4rfEkkEixg8ibjPGyMDUlj0QXg8XiQTCZhMpngdDrZYZPJZJBIJJgXNTQ0BJvNBgDMC1teXkYwGGSXzdsA8qaozHhmZgaJRAKxWAypVAoajYZ1uqWqIFJn3drawtLSEnZ2dpjX3GrPiwvKj4dCIRYWj0QijNSq1+tZnx76Q4s0m81id3cX4XAYq6ur8Hq9iEQizPBtZvSKQq7hcJhpU3DbPRCXhbgdJH7FPVjIC6ELsxltEt4EdGFQOSqld3Z2dhCNRhEOh5mxEwwG2UVClR2JRAKJRIIdsK0cB5GuSc+IIjqkn8OdE6owEwgEyGQyzMmi1PBpnrPDQNWuZIDXajUEg0HWRZuI2zTHdMYmEol9ve1Oy9gpqu3z+aBSqTAzM7PP2KGfoWrc5eVlzM7OIhgM7msOelrG8yqgM4b2H6XsNBoNxGIxfD4fI81vb28zwVWic5ymMdPdF4lEkMvlmOYNKSAXi0XW95GEV/P5PJ4+fcocKlLypmo8kno4qXE2JLJDHJzh4WF8//vfZ+WrPB4PuVwO8XicpS7K5TKGhoZYemBrawtzc3NYX19HMBhs2IXfjMhOPQQCATQaDTo6OjA5Ocl0MKhZGQDkcjkmzEaS6K875maNkfRJZDIZxsfHodVqWcqKW3VHze6A54z9SCQCv9/P9IZeZ5yN8CY5P8e+ElmcDB6uEdMID+q4aFYE0mg0wmAwMKEy4rskEgns7u4iGAwiGAyeSCuW1xljvQQ9N3LInUf6HvdrK9CK84bb8Z0bYT3B3kgN2Yt8Pp+VkY+MjLAqQJpb2osUcSS+3EnsSS7FopnnKantT05OMseRyq3J0GlkA+aTHiOPx2NteUjtmdJZlKUhAUi/388yOCdhxB02xoYYO8DzjafVavHBBx+whUzaAHt7e4jH42wTqtVq+P1+eDweLC8vs+hPIyM6rTh8gK9k96m6pV5UkS5Trv7L66KZY6ToDY3rIGkB7qVDPdPo8jzJFMjrju80XJrNmkMS1eMSfAGweaK5Ogm0ai82E60c41HFII1EI/cirUPuOfm/n8G+0jlJX08azT5PKa1Fe5PuhJPs/XjS1AdyIA/qQ0i2AM1ns8fYMGOH1HIdDgdrF0GEQgpd0uDEYjErMwsEAigUCg1PXbUP2K/wro/xXR8f0B7jaUd7jM/xro8PaI/xtOPEjZ26f/fC9yhMWR/lOCn8X57UerzrY3zXxwe0x3ja0R7jc7zr4wPaYzztOFGC8gEf9sL33pYSwTbaeBM0K6XQRhtttNHG8XEixk4bbfxfA+Xg66OXbbxdqBcrPE0Vkm0cH22n491AI+fx5a2g22ijjWOBq8zcxtsJamNDpNE23k4cppD+LuG4Y3ub30N9Eckb/a625dtGG2200UYbbbzLaLsubbTRRhtttNHGO422sdNGG2200UYbbbzTaBs7bbTRRhtttNHGO422sdNGG2200UYbbbzTaBs7bbTRRhtttNHGO422sdNGG2200UYbbbzTaBs7bbTRRhtttNHGO422sdNGG2200UYbbbzTOLJdxLvYDKwe7TGebrSbDz5He4ynG+0xPse7Pj6gPcbTjsPG2I7stNFGG2200UYbJ4LT0nql4Y1A2w3Y3g0cNo/cPivtJolttNFatM/bNk47Tsud0XBzq931+d3AYfPI5/MhkUigVqvf2uZybbTxtoO6swuFQojF4lY/ThttHIpqtQo+nw+ZTNbSBq0Nj+y08W6DawDx+XxmqbcN3DbaaB5ov/F4PHaBtPdgG6cR1WoV5XJ5nwPdiohk29hpAmQyGQQCAXg8HqrVKvb29lCpVFCpVFr9aK8M7oJVKBTMuyyXyyiVSiiXy6hUKqhWq+00VxstB3mUlUoF5XIZ5XK51Y/UUFCEh/YinTG0F9+lfcjn88Hn8yEUCtm4KpVK28h7C8Bdh9zIjlAohFAoZHu0UqmgWCyyO6SRaLqxQ55ItVr9P7FIxWIxLl68CL1eD5lMhng8Do/Hg1AohFgs9lYeRNVqFTweD9euXUNvby8sFgtisRjW19extbWFZDKJaDSKbDb7Vhp0rwpuSPZtW9NisZgZsI26OE5LlEEkEkEul+PKlSuIxWIIBALY2dlpyDi5Uc1mgy4LgUAAsVgMhUKB8+fPo6OjA2q1GslkEmtra/D5fEgmk8jn89jb22vJszYKPB4PSqUSWq0WNpsNyWQSyWQSoVCIGXVtvB2o1WqoVqtQKBQwGo3o7OzElStXkEgkEIlE8PDhQyQSCWb0NApNM3b4fD6kUinUajU++OAD7OzswOPxIJPJIJfLoVgsNutRXguvG3YTCoUYHR3F5OQkbDYbwuEw7ty5g/n5eTx9+rThE3rSqNVq4PF4EAqF+Na3voXBwUFYLBbk83l4vV74fD7cu3cPHo8Hu7u7CAaD7+RhJBQKIRKJoFQqYfv/7H3Zb1vpef7D7XDfd1KkqH2x5N2Wnd9klnQSZCYt2iBoWhRtgP4D/QcK9KJ3vep9r9oiKIIGyU3SJsFMmsbwNh55JEvWbq0kRVLc9538Xbjv5yONJEvikUg5fABj7JF0dD5+27s87/P29ECtVqNWq2FlZQWZTKbj51QqlUKhUGB8fBwAUKlUEIlEkMlkUCgUWjIGOsHQAQCFQgGLxYI7d+4gkUhge3sbAJBIJJDP51t6z3auZ3pvsVgMjuOg0+nw6aefwuPxwGAwIB6PY3FxEevr61hfX8fu7i6SySTS6TTbi50yRyeBSCSCWq2G0+lEb28vPvjgAwSDQWxvb+PJkyfI5/OoVCodN6ZWUjV0X9KdWavVUK1WkUwm35nzVCaTYXx8HLdu3cKHH36IcDgMv9+PbDaL+fl5lEolQX/fhRk7MpkMer0ebrcbH3zwAXZ2dmAymbCyssKs806/IE4LCrm63W6MjIxgaGgI8XgcsVgMyWQSy8vLLKV1mSASidhCHRkZgd1uR7VaZVZ6JpNh35dKpVAoFN6JzckHGTo+nw/Xrl2D2WxGpVJBsVhEIBBAOp3u6BC7VCqFSqXC9evXIZPJUKlUMD09jXq9jlKpdOY1SVGHTphvmUwGrVaL0dFR5HI5GAwGBINBFItF5PP5lp7d7uhVs9mEWCxm0avJyUl4vV7o9XokEgmoVCpYrVaYzWZsbGxgd3cX6+vrzDDoROPgMJBjpVar4XK5MDY2hlu3bsFkMkEul2NhYYGN512CWCyG0WiExWKBx+NBpVJBPp/H2toaUqnUOzFeiUQCm82GoaEhDA0NQSaToVqtwuFwYHV1VfD1eWHGjsFgwMjICO7evYsf/vCHkEqlSCQS+Md//EdMT0+jWCyiWCxe1OucGqf94GmTKpVK2Gw2GI1G6PV6qNVq9PX1YW9vDyaTCaVS6dKFmCUSCRQKBex2O7RaLeMKWK1WKJVKVKtV6HQ66HQ6RKPRd44rIRaLodPp0Nvbi7/6q7/CH/3RH8HtdqPRaMBut+PJkyf4/e9/j1wu15HjFolE4DgOJpMJf/u3f4uenh4AwD/8wz+gWq22FJmi1EqxWGz7ZUrGzv3796HRaJBKpfDixQvE4/GWjBVyYqrVatvGSFEDjuOg1+sxMDAAk8nEzhy73Y6pqSmUy2X4/X6sr6/jxz/+MTY3NxGLxRCNRpkxTp9Fu+frMJAx53Q68f/+3//Dp59+iitXrsDpdMJkMuHFixcol8stRyPPA628j0wmw+TkJKampvD++++jWCwiEongZz/7Gaanp7G3tyfgm7YPKpUKOp0OjUYDqVQKyWQSJpMJMpnscho7xNNRKBTQ6/XgOA5yuRxmsxmffvopms0mSqUSXr16dRGvc64gz1YkEkGn08HtdmN4eBgmk4kR7JRKJTQaDWQy2aUr39ZqtTCZTHC5XGwMwBvjTqFQwGw2Q6PRQKvVwuPxIJvNfu3yE4vFHXvAHgexWAy9Xo+JiQlcvXoVU1NTsNvtUKlUaDabmJiYQDqdxhdffNFy9OC8wE8pO51OGAwGlMtlaDQaqFQqyOXyM4eQiWDYbhCnRSaTQaFQQCqVQiQSMQOllX3XbDbbasRSRKe3txcDAwO4du0aFAoFczro61Q84HK5oFQq8Td/8zfMqUwkEvjd736Hra0tFu3qxAgzcclSqRTkcjk8Hg+TvqBzRip9t+psxGIxrFYrvF4vBgYG4PF4kEgkUKlU0NPTg4WFhXa/YsuQyWS4c+cORkdH4XA4kM1msbOzg8XFRTx79gzxeFzw33niVdLq5USbkM+ol8vlGBgYgN1uh0ajaXtoWGioVCqYzWZGTqYSUYVCAZVKBalUeuHKkq18xvTuBoMBTqcTcrkcEomEfY28eoPBALfbzSpCUqkUqtUq0uk0+16xWHwuhyu/UkNokMaQ1+vF8PAwRkdH0dPTA5VKBYlEgmazCYVCAblczi7T81jTrT6TUlhmsxlqtRocx6FSqezz9M8KIh+2ex/TGcNxHFujVDFYr9dbHmOrnxNw9nmkvabRaOByuTAyMrLPcSJni85ackrorKlWq4jH49je3kahUEA0GkWpVBJ8rQrp0IhEIsjlchZJ5jgOSqWSVbq+SxCJRFCpVNDr9cygS6VSbTNGz+MMk0gk8Pl8sNvtUKvVrKhlb28PwWBQcL4OcEJjh4hwfH7JaclXIpEItVoNpVKJHRSUs9PpdG0XxhKqSoyvI2AwGOD1eqHT6SCXy9nho9PpYDKZoFAoLtTYIY/vrPwoqojo6enB5OQkM9j4z5fL5XC73bDb7SiVSggGg5BKpXj06BG+/PJL9pzz0OihC6BYLJ5LdIFSP9/73vdw69YtDA0NwWq1ss+gXq8jHA4jHA6jXC6fy4VPEbRWSjMpBUmXJPD6AozFYoKk3oQYd6sXpUQigVKphF6vh1gsRrlcRiaTQT6fFyz91MozDs7jac5T2j8SiQROpxO3bt1i80hfJ0gkEqhUKvZZNJtNlMtlcBwHl8uFSCQCpVKJfD4vmJFK7yeXy5nDc9bnEGw2G8sK8KN27Y6Qn5czI5fLodfrYTQaIZPJkEql4Pf7sbW1hUKhIOjve9u70J0hlANJzxwZGYHD4YBcLsfOzg7C4TCi0SiSyeS5GHZvvWnpguaHf0+7sJrNJqrVKrLZLPb29pgHSV6H2WyGzWZr64IVupSUeBFqtZpFQOigIYJkPB6/MKIZfx7pcOX/OSmIq0LEVj74z6lUKsjlcgiHw4hEIkgmk/s8To7jBI1syWQyyOVycBzHImdCQiQSwWw2w+fzwev1Ms4AzWupVEIgEMDz58+xvLzM5lzINU0XXKsRGK1WC4PBAJ1Oxz5/ukAo+thutBoRoLQVPYdSrAqFYl+Uox04bB75kZiTPsNut8Nut8NgMLx1zujrlUqFlfeura2hWCzi5s2bjPArxNxLJBJIJBJUKpV9htxZnyWXy3H9+nW43W4WxSkUCkgmk4jH4yiVSm0jxJ+HQyORSDAwMAC32w2TyYRarYZCoYBUKoVQKHQuUY/DQOvxYICjVZCR2tvbC61Wi1qthr29Pfj9fgSDwXMr7DjRzuKHpvmX42leiELI5XJ5nwchlUo7olEYfyxCyFnzDTnaoHQpplIpJBKJCyWw0pj4c3jaMVIJqMFggMViOXTO6Jn1eh2VSgXJZJJ51HTxCKnpQr+TLhAy5oTeLMQx6+npgcPhgMFggEqlYjyQXC4Hv9+PQCDASNnn4fEJMYdkEKrV6n17uZPELoX47EjAkxwrSnG2myvGn7ezzmOz2YROp2NpyJN8f6PRQDabRSwWw87ODnNCyCgRCkI5j/xKrIGBAZjNZvY5FYtFpNNpFAqFc9lr7QRlBdRqNauUzOVySKfTjBJwke8iOFH4/xwPpVIJACiVSkgkEkgmk8hms+c2l29NY9Ev5nuSZ/kASqUSy5fzPRqyHMvlcts8yoNjbEUfAXhzodhsNgwMDOxTNk2n0/D7/djY2GC6F+0AGSqnNTqMRiPMZjMMBsOxBmqj0UC5XEYkEkEsFmPl6PQ7i8WioBuJIhKZTEbwy0wkEkGr1cLn82FiYgIDAwOwWCwselQoFLC3t4eZmRmmL5TP5wXXwzjM6aB1dZrxEmeH5pCMz1wu15ECdGdZJ6QiTEY2GTt0DrWzNP4wYwc4+V4kI44uRP55ddQZSsZsKBTCxsYGVldXsbOzg2QyCYlEgkwmI8i8H7wjWjlPxWIxVCoVLBYLbt68Cbfbzc6cdDqNUCjE0pLthpBnmUQigdFohEqlglgsRi6XQzQaRTgcvtBsQKv34FGgjIdYLEapVEKpVILf78fe3h7jdZ4HjjV2jtpEB2WfT3JwlMtl7O7uQiKRwO/3QywWs8OWvPJ24LDD4eBledqFTN+rVCphNBrZBm00GsjlcojH44jH4xfmkVDkjE8mBsAMrZPOI1WhUITuuMO10WigVCpha2vr2INUiEOC4zj2Get0OhQKBUEPBFLB/uY3v4mbN2/C5XKxg6hYLGJjYwNLS0v43//9XwSDQRaxE5rsedjfDxo/J9mLMpmM/QHeeOLtTu8chYNVfCcZI789RKPRYOkQ4s+1y7GivUicFuD1+E6zXqnXUC6XQyKRQDgcZpfHcT+TyWSwsrKC+fl5rK2tIZFIIJvNAoBgEVGFQsE+W+KyHaxeO+me5xs7PT090Ov1bI3H43H4/f6OkXcQ0mmjP/Q5UVQnk8lc6J1B8yiTyQSL+IpEIuj1erhcLmi1WpTLZUSjUcRiMRQKhXONKp/4ZONHP87iOZOnn06nMT8/j3A4zKpmiC/QjuokPqGY/i1UCJ0WCB0iFEbOZDLIZDIXVrVCY+K/x8Fw/knfI5/PI5FIIBKJHHvp1Go1lMtl5nkd/F4hu9/yP2vyjoX6XGUyGdRqNUZGRuD7v9YYVAFSq9UQiUSwtraGxcVFbGxsnNvhe9j+O+v6obkhoiPNg0wma8sePCnIKDvJmjn4OQFvSMvEF2sH+PuPv25Pu2YbjQZisRi2trawuLh4LOmaInfZbJZFlUOhEEsB1Wq1fbyhVnBwTK1URtI70byRcd5sNlkaK5vNdoSxcxCtnGuUvqO9WCgUUCwWL5SbxN8/QkfKSQuK4ziUy2WkUikmZHqe9+GxO/7gLz7s3yedVLLu8/k8Pv/8c+h0OgwMDLAyUMrjtaNx3dsu3bNEH6jyLJfL7dv4iUSCKSgfN04hQ4j8qqfDLOeT/o5ms8ny/aurq3jvvfeOXAOkJlwsFr/mjfAPVfr5VsZJBz1Fk4RcP6QbdP36dQwODrLqAVIafvXqFZ49e4bZ2VmsrKywz5fvGQkxh3TgHPVZneZ3lEolZLNZ1puNflYulzODp11aOW/7zCQSyamNPH70i8qV6aC9aNC7tHqwN5tN7OzsMGmHjz76CAqF4sjvrdVqSCaTWF1dxcLCAvx+PzNyiFDMr0Y967sRHeG4KONpo+T1en2fhECz+Zr7mMlkGEH5onGaSOpZnk0SFlKpFJlMBtlslkU+jvr8hCz15xexCBlNorue4zhwHIdCoYBYLNaSavtJ0bJ7c5rJJiLn8+fPcevWLTZA4hAAJy+9JAg5scd9/SzPzOVyiMVi+w44CtmlUqlDf44f3iaSrxDgpzrO+pk1m01kMhlsb29DrVajWq0ykhn/e/jRLP5YgDd8Jp1Ox9SjWz2s6NARmnwqkUjg9Xpx48YNDA0NweFwQKvVAnjdX2l3dxezs7N49OgRM3RIP8pgMCCbzbKctFA4bmwnGTddfNlsFpFIhDkblD8H8LUqu4vE28ZH++Ft65hSPeQ50sVhtVpZ/yhK4Vw0TuMkHveMnZ0dZDIZ5HK5Y6UOKPVcq9UQCASws7ODarUKsVjM2jBIJBJks1ns7u62tH+ENPD5hQf8SjXieBYKhbZVYr3tnDnr+MkYMBqNTDAxk8kglUohnU4f+lyKytI5KBSHSeiIDo1NqVSyc5SqzCjKeJ4QJJZ70gVO3jdtPlqozWaTse5Po8sg5ES87Vl83s1JIBKJmLYHhXYp2lMul49ckHyvT+hNLETajNKNVNp7GKjlQDQaZcJlBEpD8EP5rUZ2+D8r5CGr0+kwODjIOkorlUqIRCJUKhVWBrq+vo54PI5qtQqLxQKNRsNSXO2IUp4EtBb5a1AkEjGDtFPTWMDr+SUP/zhPkH/W0EFKgpcajebIKMhp0IonLVS0j3hIR+3HZvO13EU2m0UwGEQ6nUalUoFGo2FiriTqJkQaQeg0xGGpS5pXIS7H86g24j8bON1nQsYdGTq1Wo015z3O8RVas0xokKNNWmUkI0DUlnQ6fe5E8ws1duh7aBILhQJbbNQolCpC3vaMo97jvCabH0J92++g0HC1WmViXRSlIcG744yd4/LvZ/WchDIC5HI51Go1E2s7eMCS951IJBAMBrG4uLjPg6afqVQqh3J5zgoh553Iug6HA5OTk3jvvfdgNpuhUChYCD0Wi8Hv92NpaQnpdBoSiQQOhwM2mw21Wo2lCTqhjPsgDvKbaA756Z5OxkmMHeBN1RLxqEg1WqfTMcO1FQ9cCBHSViCTyWA0GtHb27uPpM8HrddkMomVlRWWPjcajRgbG4NarWbVPuclhHlW8CM7/DVJa1coYwc4HyPhLM8mdWi9Xg+pVIpKpcKikEelXd+WFj3PMZ4G1CrJ4XBgYGAAMpmMrc2LSEcKYuzQ4XPSi4smMZ1Ow+12Q61WM9Jnq8Qu4HwmlSo4qIfMcb9DJBLB6XRiYGAAo6OjTNU3kUhgb28P8Xh8Xyn2Qbzt2fz/XjR0Oh36+/tx9+7dQ9MdRB6MRCLY3t5mIlEEMgCo63K7N+Bh0Gg0cDqd+Pu//3tcvXoVAwMDLB1Xq9WQSCQQCASwtraG6elpVCoVSCQSRCIR1gMsm812bEd7vvoyP/2n1WqZFk2rOM+9SOTit5Fyydjx+/2wWq0wGo2YmJjA9vY2NjY2WnoHPoeqHZEwkei1yOX4+Di+9a1vsXY0B0GOViKRwH/8x38gEAiwyEg4HEaz2UQoFGKdtDtpP1J6Rq1W7yPON5tNpgjdqrF5ntyb02YDZDIZTCYTfD4fRkZG0Gw2sb29jbW1NQQCAcRisSN/9m13RrvnlRxIWrPf+MY3WM+zaDTaUvNhYH+LlKPQsrFzmo1OWhf04VMYlrR2hGpGJ7QhQO98kuohkjHX6XSwWCyw2+2QyWQsYkXM806sIHgbSPmYOtUe5UnyycmHfZ3+2+4NeBhIPLC/vx+jo6Ow2+3Ma87n80gmk3j48CEeP36M+fn5fZot2WyWhZrL5XJHGjrA63cjvRmaA5lMhqGhIcTjcaRSKaRSqY5MaRA34aRRZNK2KpVKEIvFMBqN0Ov1+9IErXDYhKooPC1oLFarFTab7UhDh8rNZ2dnkUwmmQGey+Wwu7sL4LVmzUGx1/PESQ1hfsUaH1QcIFTbhPOIZp7WwKCoucViQW9vL0wmE+N27u3ttVTlSe/Rzogt8XX0ej30ej2USiWWlpYQiUSQSqUEW3vHPaNlY4e/cN82wTRgvuFAERNK+bRiBJxXKJJvoJ3k+2UyGVQqFYxGI2w22760HRk7Z/Uk2rVwKTVHyrs6ne7QFBYZrUQePIpH0Gkgr4A0IEZGRuD1eqHX61kFEPEeHjx4gIcPH2JjY2NfhKRQKFxo35qzolqtMoOH9q1MJsPY2BiTbBeSRyUUaC+e9Pm0HmnP0fySscNv2XAWtMvYIafRYrHAarXCbDZ/zdghInosFsPLly/x1VdfMe5SvV7fxwG5aAXi0xSY1Ov1fRE8Ghe13Gn1Pc6Tr3Mah44KU6xWK3w+H3Q6HTMEYrGYIIGAdho7FNkxmUxs721JduqfAAAgAElEQVRvbyMcDh9JvD4t3vZ5C1KNdVKrjFj01WqVpQqazSYKhQIymQwrUz4rms3mufT24add3kaiIq8jFotBLBbDZrMBAPb29rC6usr0Lc4KobyQ4zb6YV9TKBQwGo0YGRnByMgI+vv7v3bAkrTAixcv8OrVK4RCobZyGg4DGTX8g0MkEjE+x5/92Z/h9u3buHv3LsubNxoNJJNJzM7O4vnz53j8+DFCoVDLhLp2HT5kvJVKJRSLRda/7f3330c4HEYwGMTS0pIgv0vIMdI7nzQyQETy6elpDA0N4e7du3A4HDCZTFCr1S2vzYtwPI7ai2azGVevXsXY2Bg8Hs/X9mK5XEYymcTnn3+OJ0+eYGlpaV+vKtoD7Yiw8qtCj7vASVuHojv0nlRJxnFcS3NIYz9NT7LTPv8k4FMT3G43bty4gUajgd3dXaysrGBvb++tPNa3vUe7ozoqlQo2mw03btyA1WpFLpfDq1evEAgEEI/HBYsiHzfOlo2d04bqaFGRpUd5dSKsnmXQ/Lyr0JuXNuRpIzFUVk3vQikQIWTZzzvcd/BrVJnU39+PwcFBuFyufTL1BJLoj8fjrE3EaT43OnTOM8VHqQ0+tFotHA4HRkZGcPXqVQwODsJmszHPv1Qq4fnz5/jiiy8wMzODSCRyJjLdQX5Au4xAPmGer4ZORg9FJoVItwo9xrfJRPBB0chCoYBqtcr+TWuALsuzjJEu6/M25g8+m897GB4ehtPp/JoiNOl5bW1t4cWLF1hfX0c4HGbnEX8PHOfwnBeX5TTPJFoA8ObSJmI4ySacNULDJ5gLnXI+zRhpXUqlUmg0GhiNRlQqFaYjRMrJp8VBo7adoJJzt9sNjuOQTCaxvr6OaDTakiF3EOca2TkN6MMnhj2V6FIZ7FmMioM4j0k97cIViUQsXEyHCwm5tZqqI1zk4pVKpdDpdPB6vejt7YXVaj1Ucr9araJYLDJ2fSKROJMxfN6EOv6zifvg8XgwPj7OjDmdTgfgTZO6mZkZzM7OYmlpieWYTwv+Qd1O0EVRrVZZibxIJGLN+VQqFfR6/ZkrXoSYv+OecZrUAJW30jj4kV+O486sY8VPU1zkXlQoFLBYLBgcHERfXx8sFsu+QoFGo4FisYhwOIz19XWsrKwgEAggkUjsu9Df9s5nXasnmfvTngkHWwnx+7mdNWJxmlTaeYPGSD2jdDodisUiMpkM0un0mTMeBysG2zVOSj+rVCrY7XZIJBLW7oMa0V4EjjV2yAtqNpst57fpeTShdFmmUilW8niWyA6/bPYs4IfEz6LOehCUT1cqlZDL5eyZ1BNLqIZ7FwmtVgun04nR0VFG2j14AFFVx9raGnZ3dxEMBhGJRE78O8iLvMi0F8nQ37hxA1evXsW9e/cwOjrKBK8qlQoWFhbw7Nkz/PznP8f29vapo1UEviBaqxCCT0MaUIlEgpViSyQSDAwM4N69eywdGQwGz/T8s4CfSpDJZC2X7lMEI5/Ps0grec8mkwkKheLMKeWznjetqtx6vV4WgRwYGIDRaGRfIy7L7OwsHjx4gC+//BJzc3OnbpZ5khTTUeBfrq2uU74hTp8Zn7dFXz/L72hlDvj3hBDOCz/qYbPZoFarEQwGsbe3x7qcn/ZdaS+dlY/FvxepN9ZZx6lQKGAwGOB0OjExMYGdnR3Mzs6y3mYXhWMTlTQ4obxRsmC1Wi0T9drZ2WmJpCQER4cf3m4V/HJJqjQj7kA0Gj33ZmdCQyQSsU7nJpMJWq2WlWHzUa/XEY/HsbW1haWlJSa0d1K0y0smzYeBgQH09fVBpVKhXq8jmUxieXkZT58+xW9/+1vs7Owgm822RCwXanytPoMvcplIJFi5vEgkgsPhwOjoKBwOB9Rq9YU26OV/PkJ0jKfLP5vNskaKFMHSaDRQqVQXrhZ91suZLnmv18vWK1WU0XOJQP9f//VfePDgAebn588USW5lrfLnrNV1Shc1GYjA6QtGzgN83pMQ78Bvwkv7jaLIfr//zM9t5bzhf96tBjl0Oh1sNhvjlgUCATx//vzC78JjjR1+pYEQHjdFPVQqFTiOQ7PZxO7uLuLx+IVaeAfRSt73ICjHTNwHsViMarWKQqHAtFfancY4KcgI1Ov1MBqNMBgM+xryESgql06nEQ6Hsbu7y8QUOxXk4btcLni9XrjdblgsFtbJPB6PY3l5GfPz83j58iUSiURLolftDCMfBEU8qtUqUqkUy5mLRCIYDAa43W5YrVZoNBpwHNeWdxSCZ8BvK0BRrGazCYVCAZ1OB5VK1baGoKcFEXLtdjtcLhccDgeTRKC5TCaT2N7exuzsLIuwnpUHeVYctc7P4pCSQX4wDcPvCt5uCJGuJY4crcdqtYr19XWmrXPWyJVQaIWIT+RkqkymlkkbGxv7OIMXgWONHZVKxUJ2QoB6IhkMBigUCtRqNbx8+RJ+vx+ZTOZUuXh+24FWU2sABKtM4DiORT8oWkT9kdrVXPGskEgkUKlU8Hg88Pl86O3thVqt/pqx02g0kEqlEAwGsb29jWQy2XFqrHxQtKqvrw/vvfce7t+/j9HRUajVaiSTSWxubmJ6eho/+clP8Pvf/x6bm5sXfmkc9s5CHe7N5mstpGw2i52dHSSTSTY2ImtPTEzA4/HsS5OcN0g4TqjPmQjz6XQa6+vrmJ2dRbVahdlsZnwXhULR0e0xCBzHwWazwefzoa+vDx6PBzKZjDkaVC34q1/9Cmtra4jH4205b44yQs6ydqm6MxaL7UvHUP+5g335Lgp8Nf1WwXEcLBYLhoaGMD4+DrlcjkAggMXFRcZpOQtavcsONjLmpxFPC61WC7vdDp/Ph1QqhUgkgnA43FnGjtCdSCnCQYdLuVzG5uYmdnd3kUgkTvycg00fW5lUoSXy6RIhVchqtcp6nRiNRhbt6XSIRCJotVqMjY1hamoKN2/eRH9//6HE5Hq9jq2tLRalu4gOtmeFWCyGQqHA+Pg4bt26hZs3b8LhcEChUCCbzeL58+d4+PAhfve732FmZoapzAqFs64zoSND1KctGo0ywTJKGchkMmg0Guj1+kP1lM4L/AayQoAiWLVaDTs7O5ienkY+n4darYbT6YTJZDp0PXcayDi/desWxsbG4HK59il6k5zH5uYmnjx5wrpItwNHzd1ZLzWKzh2M9Gm1Wtav7iJTrfROQq1RjuOgUCigUqkwMjKCWq2GFy9eYHd3F9lstm3aOkftw9OSy6VSKZxOJ/r6+tDf34+trS0mv3LR0bljb10hqqP4INIbeVTVahWJRIJp7LQDQhNi+T2wiLNCjTP5kuedDlK7HBwcRH9/P9xuN4xG49daelAInYydWCx24Rb7SUEXuVqtxtDQEEZGRuDxeKBQKJDP57G+vo4XL17gxYsXWFhYwN7e3qUQCTwLms0m4+wUCoV9oWoqJFAqlayHFP/PeeE8yOl05qRSKZbWoeaZOp0OHMd1vLFD1ZB9fX1wOp0wGo2QSqX7jLlMJsNatFxEB+mjILRRTlWD/LtIJBKx9dkO51FI/p1UKoVCoYBer0dPTw+azSaLtpLoZzsgFH+V1q5Go4FcLsfW1ha7Iy4axyasD7LgW/kA6ECRSqX44IMP4PV6USqVLqy9+3HvdTBE12qlC+WZaSErlUqWzuoEzYOTQKPRoK+vD9/+9rcxPDwMm8126MVAB+2DBw/wxRdfYG1trSN5SXziuMvlwscff4yxsTGYzWaUSiW8ePECP//5z/H48WNEo1FB2iUchrM+8zwImY1Gg6Ud6dn8rvZUhn7e73Fez6Znkbo1tUuQy+UwGAywWq3MmOtkkKr3+Pg4enp6YDAY9lUv1mo1LC4uMt2Sy3C+nBT8rvV0ppDDRTymdjuQZ9UkIqPNarVieHgYIyMj8Pv9WF9fFyw61+pd1gqo2lWpVLJebL/73e/YHXHRd+Gxxo6QzHpimhOTPpvNolQqsRB6uw4c/riEqpThC1XV63WIxWI4HA4MDw+ziEGrUufnDZfLBY/HA4fDAavVCq1W+zVtinq9ju3tbSwuLjISbycRcQl0MPb09GBkZAR//ud/jqtXr0Kr1aJareL3v/89nj9/jmfPniEUCrXVozoKQr8PX9TzYCpXLBZDo9EwMiwJDJ734XSexiXJ8VPDTFLF7sS5PojBwUGMjY2x9iVUyVosFhEIBLCysoJ//ud/xsbGRseP5bSgc4YfLSaHslqtguO4C09jHYbTfu4UQe3r68PExARu3ryJzc1NzM3NYW5urqOpACcFaUL19/fDZDKhVqshGo2y9NxFO8QXZhKTZ61UKhkpkHr0XJZox2kgk8lY2orUac1mMyP4dro3yVfZJU+f/7VarYZsNovt7W3Mz8+z/kOdYOzw5d/J6xKLxXA6nRgeHsaNGzdgNpshkUiQTCaxtLSEtbU1Jst+GZu0nhZ0iRSLRSa6x09lUbNXnU7HyPadvmaPg1wuZz15ALBmmBfdF+osIMOUn7ahaBVVtqysrCAej7f7Vc8FByPuBx3Uds/fWaUEZDIZ3G43HA4HNBoNU7tOJpPvxJ0ok8mg1WqZplW5XEY2m2VNZy8aF2bsiMVi1n/IYrGwQ5Qa0nVa2qNV8POUAJiQGR1YneCNHIfd3V1sbGxgc3NzX2UVeVWFQgFbW1v44osv8Jvf/IZdmO02FMhj4utWSKVSyGQyTExM4M6dOxgcHIRGo0GxWMTKygq++uorrKysMFn2y37InAQ0h+FwGMlkknHmaB9Sw0wq0ab1ehkNHpFIBJ1OB4fDAeAN6TWVSrWFO3Dw3d72me7t7bH+QZTqq9VqiMfj2N7exsuXLwVpFNmJ4GvQ8GVQSMakU8Z92jODigAmJyfhdDpRKBTw7//+7/jyyy9ZH8bLDJHodZNvk8kEk8kEqVSKTCaz76xp9fmnxYWJTDSbTZbDMxqNbOO2I3d3nuBrmORyOaRSKZZj5ziORXU63bgrFouIxWJYXl5GIBCASCSCyWRi4oG7u7v4n//5H3z11VcIh8OIRqOC9FISAvwyasrrUxO60dFR1g16dXUVjx49ws7ODlMq/UMBRe7q9TpCoRAWFhYwOTmJZrOJYrGIFy9e4MmTJ5idnUUikWibNyYUkskktra2WPo8Ho+zIoJ2nj0n+d2U+q9UKmx/5XI5zM3N4cmTJ3jw4EFLRls7BfqOQ7P5ukn03t4eNjc3meTF1tYW1tbWsLm5KVj7HSFw0s+RquvGxsYwPDwMvV6PQqHAouPvAvifA/E6Wy03P1gKf1pcqLEDvFFjphRWJ4QhhQQZO4VCgXWP7u3tRSQSYf1pLsN4Kczv9/vx8uVLpFIpWCwW1Ot1JBIJBIPBfRpJZOh0wtj4ZFvgTcg4l8shEomg0WggEAjg1atX2NraQjqdvhTcDaFB+3B3dxdzc3OQSqUs1PzixQusra0hGAxeekMHAJv77e1t5PN5BINB5HK5ths7J0G5XEY6nUYgEGA9y+LxOJaWllh1S6eP4SzgR5FfvHiBYrEIu92OjY0NrK+vIxKJXIr5OwhK77hcLpjNZkilUuTzeVQqlY6IUgkFmr9sNotarYZ0Ot1yxWUrP3uhxg79Ie+KXwXyLqHZbCKZTOLFixeQSCS4d+8eNjc3sbGx0TFh17eB5OdXVlYgk8lgsVhgNpsBvL44YrEYnj9/zqQDOmVMB3ulUekqVVxFIhHodDpsbm4iFAqxHHmnCz6eR6USec5LS0vI5/OIRCKsievGxgaCwSCSyWTHeM5nRbPZRCaTwfb2Np4+fYpsNsucj04WvySkUins7Ozg+fPnCIfDyOfz8Pv92NjYgN/vRz6fb2vVzXmiXq8jn8/jl7/8JXp7e2G32xEIBLC0tITt7e2OOXeAk3+OSqUSZrMZXq8XJpOJCc4K0SKlU0DZi3K5jFAohEajgXg83tL4Wl2nouMeIBKJBNsFYrEYFosFHo8Hf/zHfwyxWIx0Oo1//dd/RS6XEzx33mw2T5TUOzhGIcrPAbA2CyaTCQ6HA4VCgV0oxWJRkLzsWcd4UpBOAlU8EOmXyK2lUolV85wXTjLGw8ZHpGQiiVM1DnF4qJqjUCi01Ts87zl8G4jXJJPJwHHcvr5ZrTbiJLR7jMCbaqwbN24gk8kgk8kgGo0K5k2f5xhJCFOv17OmjJVKhfEdLyr9eta92CqI/0Fcx1qtxsYtsANw7utULBYzTZ3JyUncuXOHGa8//vGP92lenQcuci8qlUoYDAaMjY0BAPL5PL766qtzX69HjfFCIzuFQgGRSAQPHz6EWCxGuVxm1VjvGqjbMkmeU4SBqn062ZsiNJuvpegvI5eF0olkiFUqFZYP5+tivItr7zSgz4gMv3cV1Dpie3t7X/uWy7APG40GSxPzdcEuyznSKkij7F3gszSbr1u1JBIJrK2tsWrIRCJxKc/Z41CtVpHJZLC5ucn+3c7I1YVFdi4aneBNnje6Y3yNo8ZHlVlEhO9EdOfwDbpj7Gy0K7JzUTjLHJ42vcxXIudHys87Qk74Q16nHdXy92CX9S66aAWUbuuiiy46t+rqsuIs5c/8gpzTnE1C0Sv+kNFRxg5wduntLro4DN3DoYs/dBzU8unuCWFwkZ8j/17szt/Z0HFdKc+72WAXXQiJ7nrtotNBJPR295A6T7zre5Dm8F3GeZ+lx3J2uuiiiy666KKLLi473l1Tv4suuuiiiy666AJdY6eLLrrooosuunjH0TV2uuiiiy666KKLdxpdY6eLLrrooosuunin0TV2uuiiiy666KKLdxpdY6eLLrrooosuunin0TV2uuiiiy666KKLdxpdY6eLLrrooosuunincWy7iHexGdhBtDrGdvYsuegx/t/vbOVRp0a3+eBrdMfY2eieN6/RnUPhIRaL9/XUagUXNUZSe6b3Pu7dJRIJLBYLstksSqVSy62ijhpjR0V2LqPkt1CLsJPBH59YLL6U89RFF+8C/hDOmy724zKet9TFvdlsvrUNhEQigV6vh1wub1+7iK6V3tm4yDFKJBLI5XLIZDLUajWUy2XU6/VzP3i73uRrdMfY2eiO8TXe9fEB3TGeFvxo5MHIJP1bJBKx/9fqnXLUGDuu6znweuAU2pLL5ajVakin06hUKqhWqx3r2ZAFKxKJIJVK91m29Xr90nZyl0gkMJvN8Pl8GBgYgFQqRbPZRKFQQCKRQDwex/r6OjOAuni3IBKJoFAoIJVKIZFIIJFIUK/XUSwWUalUOnY/ngQikQhisRgKhQJGoxEKhQISiQTFYhGlUomdO5d5jF28G6C7RSaTAUBHr0t6T6VSCYfDAbVaDbVaDYvFgkqlgnK5jGaziVKphHK5jFKphHq9zs6VYrGIcrmMXC4n2BjbZuyIxWKIxWLIZDJIpVL2b/qaXC6Hz+eDRqNBuVyG3+9HOp1GOp2+kIjC20CHpEwmYx1p6e8ikQgcx7FQnkKhQLlcZn8qlUpHjOGk4DgOFosFY2NjuHv3LlQqFaRSKbLZLHZ2drCxsYFYLIZMJoNisXhpjbou9kMqlUIqlUIul8NgMEChUEAmk0EikaBcLiMSiaBer6NWq7X7VU8FOnNkMhlkMhk4joNer4fT6YRarYZEIkEmk0EymUS5XEatVusa8V2cK+jOoLXJj4LQ1+m/MpkMjUYDmUymIw0esVgMlUoFtVoNo9GI0dFRmEwmmM1meL1elMtlFItFNJtN5HI55PN5pNNpVKtVlMtlxGIxpFIppNNpFAoFwfZeW4wdkUgEjUYDg8GAwcFB2Gw2GI1GqNVqFtXR6XRwuVxQq9UAgGfPnmFxcREPHjwQ9AM46/vL5XLodDr09/fDYDDAaDTCYrGwxVqr1SAWi8FxHAYGBhCLxRCJRLCwsICVlRWEw2GUSqW2jeGkEIvFcDgcuHLlCr7zne/g/v370Gq1kMvlKJfLWF1dxcLCAgqFAjY2NhAKhZBMJtv92l20CKlUip6eHtjtdgwNDaG3txd6vZ4dxMFgEJ999hnW19cvleEulUrR29uLnp4e9Pb2wmq1wmQywel0QqVSQSwWo1wuI51OY2NjA7/4xS9QLpdZlLbdIGdKJBJdOiOzi8NB98SVK1fQ29sLt9uNarUKiUQCsViMer0OqfT1VZ3P51EoFJBKpbC8vIxwOIxisdjmEbyBVCqFUqnE1NQU3G43enp6MDExAbvdDrvdDoPBwMYllUpRr9dRqVQQjUaRzWaRTCYxMzODnZ0d+P1+JBIJwRzoCzN2xGIxtFotTCYTHA4HXC4XXC4XBgcH4Xa7YbFYoNVqWYSHPjRKB2m1WqjVajx+/PiiXvlroPSUx+OB1WqFz+fD6OgoXC4XvF4vtFot8xTpQBKLxVCr1SiVSshms1hcXMRvf/tbzM3NYWFhoWMO0cNAkauhoSGMj49jaGgIarWaRbCkUinMZjP6+/tx584diEQiVCoVpFKpjh3T20BRRZVKhZGRETSbTVSrVbbZGo0GYrEYS+HQZqWvC5V3bhcokuN2uzEyMoLBwUF89NFHsNvt+4yBubk5fP755+1+3RPDYDDAZDJheHgY/f398Hq9GBkZgdVqhU6nYxEdMiLy+Tw0Gg1+9atfdRRBVCKRQKvVwmw2s7NSLBZDqVQik8kglUohl8sBeL0Ga7UaizDT/7vsaGd1qJAQi8Ww2+2wWq1wuVyYmprC0NAQfD7fvmwH/0yheyQSieDBgwf47LPPEA6HOyLyKBaLodfrMTAwgPfffx99fX3o6+tjjoRSqYRMJmPzJxaL0Wg0oFKpoFAoUK1WUSqVYLFYsLa2hpcvX2JmZmbf+doKztXYocueyK12ux0+nw9XrlyBx+OB2+2Gx+OB3W6H0WiERqPZx9wm0lKz2US9XkcoFIJcLmfW7kWCwoc6nQ4jIyPo7e3F8PAwhoaG4HQ60dvbC7lczsLjVHpHoEtRrVYjFouhVCrB7/cjl8uhWq1e6FhOCuIy9PT0wO12w2QyodlsshwrACgUCjgcDgwMDGBnZwfb29v7yGaXBRRRVKlUMBgMsNlsmJqaAoB9PLFarYadnZ19abtisYharYZms4lKpYJardbR3LKjIBKJoFQqYTAYMDw8jImJCYyOjuLGjRvQ6XSsWqJYLGJvb+9SjI/2rcvlwsjICG7dugWv1wun04mBgQEYjUaoVKqvlcqWSiWEw+F9qelOGC85T263GxqNBnK5HHK5HBqNBnt7ewiHw9jd3WXjyOfzbE02Go195ysZQZ3scPFBd4lCoWBjqFarLM14cAydPCZy5j0eDwYGBjA2NoZbt27B5/Ohp6eH3XP8O7DRaKBarSKfzyMUCiGRSODp06fY29vrGGNHp9NhfHwcExMT8Pl88Hq90Gg0zJFoNBos9c0/IxUKBZRKJQtqiEQi5nBks1lB7shzMXZoQ3EcxwbgdDpx5coVTE1N4Xvf+x4MBgOUSuWRXtPBkjWj0Qir1QqLxYJUKnWhBgIdmEajEdeuXcNf//VfY2RkBF6vFwaDgYUYjwNt0qGhIdTrdbhcLkSjUSwsLCAej3fkxpTL5bDZbBgbG0NPTw8AwO/3o1AooFQqQavVwuFwwO12o1AoYH5+HhzHQSKRXKoS2YNpyeHhYVy/fh2ffPIJNBoNi9bV63WUy2Vsbm5ibm4OW1tb8Pv9jNdRqVTg9/uRTCaRTCYF80guCnK5HD09PRgcHMQPfvAD3LhxAx6PBzqd7muXpEQi6Ui+AB/8Qofvfve7+Mu//EsMDw8zp+SgQwK8uSCVSiXjHXAc1zFzKRaLYTKZcOXKFYyOjsLhcMDhcEAmkyEUCsHv9+Phw4fsctzZ2UEmk0GhUEClUoFcLmcXTz6fR7lcRj6fZ8Z6J4LWnlKphEajwfDwMCPpEj0gm81+bQxkxHXCvPEhEolgMBjg8/nw7W9/G3fv3sU3vvEN6PX6fdIeNBYaB0XwOI6DVCrF7du38bOf/YxVyLZ7/iQSCbxeL/7iL/4C169f3+cgAa/HUywWGfc2l8uhVqtBJBLB6XRCo9FAoVBApVLB5/Mhn8/D7XYjkUgwQnMrEMzYoSiORqNhl+C1a9fg8XjQ398Pl8sFk8kEi8UCs9m8L5xFOChAxLdsq9UqOI7D2NgYdnd3BRn8SUHh4EqlgkKhwC54nU536IF5FCgN1tPTA47j8MMf/hA//elPsbCwgGg0eo4jOBsajQZqtRoSiQQikQi0Wi3LE0ejUVy/fh0cx8Fms8HlcsHn88Hn82FxcZFZ8J0OsViM27dvo7+/H+Pj47h37x4sFgszrulioHAyhV09Hg8KhQIz/Ihcl81mEQgE8OjRI8zNzSEejyORSLT9IDoOcrkcWq0Wf/Inf4LJyUkMDQ3h6tWrzCGhwgH+4SuXy9Hb24tYLMa8tE6DTCaDVqvFD37wA7z33nvweDxQKpWMM8AH/1Kky9Vms+G73/0uqtUqdnZ2EIlE2jGMfeBXj925cwd2ux06nQ5isRg+nw83btzAe++9xypXC4UCi+6IRCLGdyReYSgUwoMHD7C6uopkMolMJtPuIe4DpdL7+vowOTmJyclJ9Pf3g+M4AEA0GsX8/DwCgQCSySSL9NRqNWg0GlQqFYRCIaRSKRbhavdetFqt6Ovrw61bt/Dpp5/C4/EwCgfdNYVCAel0GtlsFplMBkajETqdDlarlZ2rlHolSkG7x0Znfr1eZ1WbfAepXq+jVCphZWUF8/PzSCQSSCQSqNVq+PDDDzE0NASHwwGr1cr27vDwMNbX15HJZFqOrgpi7JAHJZfL4XA4YLfbGY/D5/NhaGiIlZEfNHLokGk0Gvs8ReK9EEKhEMLhcNvSA3Txl8tlVnkllUpPnc8ncnaj0UBfXx8sFgvUanVHGjuUsioWiygUCsjlcshms9jb28PW1hYsFgt8Ph8AMOa9xWKBSqW6FBUsUqmUeYpXrlzBrVu3cPv2bajV6mPnVi6XQ6/Xsw3MT12Vy2V4PB5UKo50sFkAACAASURBVBUkEglUq1UkEokLHtnJIRaLYbVa0d/fj6mpKYyNjaG3txcOh4MdVvz9RinlgymRTgRVWV2/fh0+nw86ne5r80pjo0MaAIsaSKVS6HQ6lsrqBDQaDZRKJSSTSXAcB7VaDb1ev29Mbrd7X+qqVCqhVqtBIpEwb5qI5js7O0gmk8x77jRjhy69iYkJ3LlzBzdv3oTL5WL3SCKRgE6nw+7u7j6DplqtQq1WI5fLYWlpCSsrK8hkMh1B6lapVDAajXA6nbDb7dBqtWg0GkilUshms4yoG4lEkEgkUCqVMDAwgJ6eHlgsFpbOovROp0SuKCiRy+VQLpdRrVYZjzOXyyGVSmF1dZXxVYlj1mg04Ha7odfrWRSZzzMTCi0bOxR9kcvlsFqtuHfvHoaGhnDlyhUMDw+zkjOKgPBDWmTolEollEolduFzHAen07kvz/fZZ5/h4cOHePjwITKZTNsMHnrPZDLJrPHTgowlCs2qVKpzeNvWwd9QdICYTCYEg0Gk02msrq5ieHiYhSJ1Oh1sNhtsNhsqlQoqlUq7h3AstFotxsbGMDU1hStXrmBychJarfZE0TrysMnzJDQaDUZ4pZTf1tbWOY6iNXAch/v37+P73/8+7ty5wzgsfGOG9iDwer/TuggEAm2vjDwK5FR4PB7cunULHo8HcrkcwH4SOR2qdDjz5zOfz2N1dZWlSToB5XIZgUAADx8+xMcff8wMOr4RJ5FIoFQq2Th1Oh37eZvNxv5OhR/lcpl5z+Fw+GIH9BZotVr09fXhRz/6EUZGRtDT08POTpFIBK/Xi+vXr++7T8gxpajOw4cP8W//9m/Y3NxEPp9v63goFUXFOhSx4DgOS0tLmJ+fx8LCAiKRCJP0cDqd+PjjjzE1NYWRkRGmOzc9Pc2M1E7Yg/V6HdlsFpubm+jr60OtVoNer0coFMLS0hKePXuGf/mXf0Eul0O9XmcBA5VKhenpadhsNlitVsYLjMViWFxcRD6fF+S+b9nYaTQa4DgOZrMZH3zwAb7//e/D6/XCZDLBYDCwHDGfYOX3+xEMBrG7u4tgMMjKzgwGA2Oju1wu9gEWi0Vks1lmMbYrVFcqlbC9vY10Oo1yuXxkGq7RaKBQKLCNJ5FImIYCGXByuRyDg4O4c+cOarUaFhcX2zKm41Cv11EoFPCLX/wCRqORGa1kkXs8Hnbhk3AipXI6wYM6KaRSKRQKBdRq9aHpDQAsUsVPz1Hki6oM6MIh0jbJKbSDUP82yOVyqNVq3Lt3Dx999BGmpqZY2g4AqzajcZPcgFwuRzAYhN/vZ55yp3iWfMhkMqhUKmZ40gVJDksul8Pa2hrC4TByuRykUinjwOh0OmYAkWfdKeuZxDz9fj/+6Z/+Ce+//z4+/fRT3L59GyqV6mtRKL5hV6/XmZFA6Q+1Wg2v1wuXy4VQKNQxRGzgtWFgNpsxOjoKn88Hk8n0tfHxCbwU9T+Ybv74448xPz+PRqOBRCLR1r3YaDTY3llfX8cvf/lLdo4kk0lWWk4XfrPZhEajgVqthkKhQKPRQLFYRCQSwZMnT1hkpBPQbDYRCATwk5/8BI8fP2bnYqlUQiQSYQU5tJcowshxHKxWKwwGAzQaDXsWAJZ2FmJdCsbZkUql+0JzNDkikQjVahWZTAbZbBapVAqLi4vM4AkGgyz8ODExwbwrfv4ynU6zEF87NT3IiEmn00zQ6eDXisUicrkcdnd32aZSKpWw2WwwmUws5Ez9QKjSieO4jqzgqdfriEQiSKfTiEajjKjbaDSQTCaZUccnaXdKyP8koAgG/Z0u90qlwiKIjUaDVc3x/9BadLlcMBgMsFgsLFfNrxjppFQPRRPtdjt6e3tx+/ZtDA4Owmw2g+M41Go1lEolhEIhZtQTSVsul7MDjYjZnXLQHgSlKK1W674KTqpkCYVCmJ2dRSgUQrFYhMPhgNfrBQBWfl4qldg+76RxkgP46tUrpkxbq9VgNpuh0+nY2iMHBHij5aJSqfaRlOn/A51XvSSXy2GxWNDf3w+9Xs/GRZSHSqWCUqnE1HZLpRI4jmPpPblczs5ei8UCvV7PBPnaOdZCocDkDcLhMKNHlEol9l50nlLhhEajgVKpBPDaEUmlUvD7/ft+phNQLBaxs7ODdDrNyucptZVKpfY5DWSYikQiqNVqtjb5hg2f99MqBDF2+OWLHMdBoVCwDUS180tLS1hfX8fy8jILv2WzWeTzeabBY7Va4XA4YLPZ2KIslUrMk4xGo209dOjiC4VC2N3dRW9vL/MCS6USAoEAAoEANjc38cUXXzAjwGKx4M6dO5icnGRleKSX4fP5EA6HoVarOzYiQqrP2WyWkeFkMhlevXqFvb09VCoVFrWSyWTQaDRIpVLtfu23giKNZNwUCgXIZDJks1nE43EsLi4ysmMgEEA+n0epVEKlUmHVVgaDAffv38fIyMi+agqpVMo+t04xdsjIttvtmJqawre//W1cvXoVTqcTCoUCwOvUTSKRwPT0NDY3N1EqlXDt2jVG2m40Gpibm8P8/HzHqmVTWsBms2FkZIRdkpVKBcFgEE+ePMHMzAxmZmaQSqUgk8nw4YcfsqgcAObURKPRjors8JHP5/H8+XOsrKzg8ePHGBwchM/nY0a3SCRCuVxmhGaPxwOPxwOz2cwuTjq7qDqmU0BpSJ/Ph9u3bzNDBXh9pxDxn87ieDyOTCbD7o+RkRF2gSoUCuh0Ouh0OiiVyrZXEdJnTuKVwNcNzUqlwjICNCaTyQSJRMJ0dsjY6SRUq1WmfAxgX8TtIPjFSBqNBhqNBiqVap+xQ45zx0R2iFz1xRdfYGBgAH19fTCZTKjX60in09jd3cXLly+xubmJV69eIRgM7iNW6fV6lgpzuVxwOp0QiURIJpPw+/34/PPPsbq6yi7Wdi/Ux48fI5PJoFqtQqPRoFAoIBqNYn19HeFwGIFAAGtrayxdNTExAa/Xi97eXsZUp3J2/pgpbdCJOJgTp/xsKpVCKpWCWq2GRqNhG7JTLvjjUCwW2fra2trC5uYmOI5DLBZDIBDA+vo6uwQKhQLjdZABD2CfKN3du3f3pUoOioK1G6QT9f777+P27du4evUqrFYrOI5jpNfp6WnMzc1hZmYGCoUCTqcT/f390Gg0aDabyGazePnyJZaWltqaUj4OpMdCa3R5eZmJlv33f/83ZmZmsL6+jmw2C51OB4vFgomJCTgcDjbOL7/8Ek+fPkU0Gu24yA4flUoF6XQas7OzWFlZgVqthsvlYtFXhULB+n599NFHcLvdzKCjc3tubk6w8l4hwJcMMJvN0Ov1zMmoVqt49eoV1tbWsLW1hfn5eezs7CCVSkGpVDKNl1u3bjESdqVSYRGfTsLbPmvis1CVq9PpRLP5Wr18c3MTqVSq49LjAFjkjXDcOFUqFcxmM1NXJv4q0SHi8bhgJGxBjB0S4QqFQpibm0MsFoPBYIBIJEIqlcLu7i5WVlYQiUQQiUSQz+fZy1O42WazsXCdQqFAs9lEPB7H1tYWvvrqK8Tj8Y7ZjOFwmHkLKpUKpVIJiUQCW1tbrDFmKpVixk4ymUQul2PNzmgMfC0ig8GAYDDY5pGdDHzDh4iAAJjyMMdxlyKVVa/XkcvlWBPTYrEIuVyOeDyOUCjEojkUdqYUF/0hBWkSxKIxVyoVJjpYKpU65qKktEWpVGJ7K5/PM493eXkZz549w9zcHAKBAHp7e5mBJJPJUK1WmRfd7pTycaC1mc/nsbe3h4WFBUgkEpRKJTx//hybm5tMFJG8ZmpWSGH3nZ0d7OzsIJ/Pd+SFQqBUP0Uds9ksM86azSZL4xDnijRayHCNRCJYXl5GNBrtqMgO8CaC8+rVK5Y1aDQa+Oqrr7C0tITNzU1sbW0hHo+jVCpBr9ejXC4DABOKrNVqSKVSSCaTrNiiE9fsYaA0rMlkYunJYrGIeDyOeDzesXw54GQpUX5RC8lcUPSOmoCSsSMEBDN2yuUydnd38fjxY2i1WnbpFQoF7O3twe/3Hxq5kEqlcDgcjDeg0WgYdyAQCODly5d48OAB0ul0x0Q9KLSdSCSgVCpZDp36XfHL7hqNBosOFItFphdERFAq2TebzScSJ+wE8HVn+I0gSa+FGoV2Omje1tfXEY1G4ff7oVAokM1mWTM6/kXH38DUVsJkMsHtdrPqQeB1CmRzc5NxnTrpcG00GlhfX0cul0MgEGDkVL1ej9/85jd49uwZ1tbWwHEcPB4PNBoNK9nO5/NYXl4WtF/NeYC8ebooSWcmm83i6dOnrIKM5q+3txcej4cpt+ZyOWxsbGBjY4NVjnQ6yACvVqusouogWd5utzP+FXHx1tfX8eWXXzKeRaeg2WwinU5jZWUFALC9vQ2DwQCZTIbPPvsMS0tL8Pv9rBKULkkipVOlbD6fRzAYxNbWFkKhUMdxXI4CSUL09fUxLSWZTIZIJILd3V3WhPcyQyKRsDFarVa2Non/SoGSjjJ2gDe5ubW1NcZbkEgk+0ifBxeZRCKBx+PB8PAwJicncffuXej1evac6elpzM7OIplMdtTBSoZLPB7fJzNPJEhahMTdcDqdsFgsTISQX35PhpFCobgU0RAC8V1o3Hwr/bIYO8Cb6haKdlD64zivSSQSwWKxoKenB9/5zndw8+ZN9PX1MY7E3t4enj17hu3tbSSTyY44XCnnXS6Xsb29Db/fD4lEAp/Px/gpVHkEAC6XC1euXMHVq1eh1+vRbDaRSqWwtLTEvq8TxnUYKNIcDAaxt7fHBNeICAq8Tun5fL590vYcxyGdTmNmZgazs7N49epVR4/zKDSbTeZE9ff34/79+7hz5w7u3r0LpVLJOkv/9Kc/ZWnLg+TRdoIic5lMBqurqwgEAnj8+DGLIhOFgO9kGY1GfOtb38I3v/lNjI+PQy6Xo1arYW9vDw8ePMDCwgKCwWDHjPE4iEQiqFQquN1ujI+P4/bt2zAajSgUCnj06BEePnyI+fn5dr/mmUG2gdvtZm1pxsfHYTQa2TlDzoaQkTjBbyS+FXZQ9vogZDIZJicnWWkhGQP5fB7r6+tYXFzE+vp6Rxk6wJvQMZVyEvhtEsgIkMvlrH/WQVEzeg7lJy+LpU7j44uw0QK+LHwdPmgs/Pk8as1SFV1fXx9rkErVh8DrqN/29jZLDbRb14MPvgAiQSQSoVQqsegHpa1u3LjBBAYlEgkSiQTC4TBWVlYuxVol3gCNlS5Q6t9jsVhw7949XL9+nbWQoLTX+vo6IpFI2/S8zgIq8iAirkajgcFgwJUrV3D37l2MjIxApVIxY+HVq1eYn5/H6upqR6YkyXmmeaR3JKOVQOO22+0YHBxkav0ikQjxeBx+v78j9+JRIAfZ6/VicHAQg4OD0Ov1AMCiseFwGMlkss1venbQvUhNeXt6emAwGFhGZ29vD3t7e4Irz5+r+33ci1KFwL179zAxMYG+vj4olUpWpra8vIz5+Xmsra2d5yu2BD6bHMDXjDKO46DT6XDt2jX09vayCeXL75OEdiwW63gRPgKfr0MibIdpe1xGvG3NkuAliRBeuXIFVqsVSqWS9SIicbC9vT0Ui8ULfPujwTfo+DjIE1MoFLBarXj//fdx/fp19Pb2otlsIhaLYXNzE/Pz88hmsx3ngBwGvrYM8KYHmsvlwtDQED7++GNcv34dbrcbYrEYiUQCfr8fS0tLHX85knNBY6NO9b29vXC73axn1s2bN9Hb28sKB3K5HONWzs/PC+49CwkqhjgqjcEv8qAu9v39/XA4HABer+3V1VUsLCwgFouxCqFOBdED5HI5xsfHMTk5ifHxcajVahZ53dnZQTQa7Til65OC5kyr1eLmzZsYGxuDz+eDRqOBWCxGuVxm0hDxeFzQ333huQbSf+jp6cHo6Cg++eQTJpktEokQCoWwtraG5eVlRjy7bKAx3r9/H5988gm+9a1vQavVslQVpRTo8ikUClheXu6Yi/Ft4PfnoZJzSpGQINa7BsqhO51OfPLJJ7h79y7zTIgImcvl8Mtf/hKPHj1iZeudeIkcBbFYjIGBAdy8eZPJ8nMch1AohJmZGTx79gyvXr1iqaBOBFXAmc1mFhWQSqVMAuBP//RPce3aNYyMjGBycpI18iV+z+zsLH796193bIsPEks0GAzo7e2FSqWCRqPBJ598gsHBQaYwTBVppK4MvC5VX1lZwcLCAn71q18hEol0rKHzNlBxR39/P77xjW/gxo0beP/996HX65lR95//+Z94+PAhZmdnO2ov0hql0nIy6Ii76fV68Xd/93fwer0wGo0AgIWFBXz55Zd4+PCh4EbARYDuRI7jcOPGDdy/fx8/+tGPmPAqUV7I+V9cXMTLly8FfYdzM3b4Fx6F5kjkqdlsor+/H5OTk0z8SSqVol6vIxwOY21tjRELLxs4jmOiXUNDQ7h79y5MJhOrguAL2BWLRUSjUUQikY6q2nkbiA9gsVig0WjYYUoMer5K5mUHGXY6nQ7j4+O4cuUKbty4gYGBAdhsNkgkEiZtPj09jZmZGWxubnZkY8y3gTR4qCGfVCpFMpnEo0eP8PTpUywuLnZkGTZFOahLudFoxMTExD59nWq1ColEgg8//BB9fX1MCLLZbCKZTGJmZgbT09NYWFhgEhftHhNFTelM0el0GBwcZEZ3T08PM3auXLnCyrT5kSwSK83n81hYWMCTJ0+wsLCA1dVVFAqFjjEATgutVguz2Yzbt29jamoKExMT0Ol0qNfriMfjeP78Oebm5uD3+9u+F8mwoXOTnHudTsfWIEXJiQtIjaalUikTEKSih052Ng6C1jC1RVKr1ZiYmMDU1BTMZjMrZALAGqBSxafQ1YEtGzv8cCqJyvH1RfhRAJrcarWK0dFRXL16FQqFgpF8qZfJxsYG1tfXL52xQ7oIpPY5ODiIyclJVpbMNwD5st+RSOTSGAeUCtDpdPB4PDCZTKyU9aCx00nS88fhuPekjepyuXDt2jXcu3cPk5OTrLKFLsvNzU38+te/xsuXLzuuv9BhjS/5oP2q1WrR09ODwcFBGI1GpNNp1hX7+fPnCAQCHcPVoXA4telQKBQwm82wWCzwer345je/CaPRCIVCwarqpFIpbt26BZPJxLzJRCKB3d1dPH36lAkptrMqic5RMnD0ej2riHO5XHjvvffQ09MDr9fLmggrlUoolUrmPfNBtIBIJIJnz57h0aNHWF1dxfb2dptG2BrIcaYqnrt37+L27dsYGBiATCZDLBbD1tYWfvOb32BlZaWtDZYlkv/P3pf+xnle15/Z94WzD8khh/u+iJIlWZYtO3EUp02TuGhiJ02QAkFRoH9IP7RAPxVBgQb9UANp+qHNntiyHcu1JUuiKIkS93U4Q86+7/v8Pqj3+iVt2ZJFmiP95gBEFFkS+czzvs9z7r3nniuBVCqFSqVivyOdTofh4WG2O7Db7dy1W61W0dbWBpvNhra2NigUCtTrdTYQpIaHZgs2Pg1SqZSfY4PBwA7zer0e09PTmJmZYcdkYbCcSqXg8/mQSCQOvdLxhcmOMA3ndrthNpvhcDhw7tw52O12mM1mZDIZ7gwgXYNcLkckEoHRaITZbIbZbIZYLEa1WuVZWX6//9hnmDwsiMxJpVIMDQ2hvb0dLpcLo6OjOHv2LJskCQ8iSlumUikkEommavn8PJDJ1cTEBP72b/+W6600EiQejzdtV5lwUKJw34TzdUgnQGMkpqencerUKTz//PPo6emBw+Fg/UOpVEIkEsG7776L2dlZ/PrXv0Y6nT72w0g4+4jGPFCKmCwR6KvRaECn08Fms+H73/8+zp49i8HBQUSjUVy6dAnXrl3Db3/7W/ZvaQaIxWKYTCY888wzmJqaQldXFwYHB9nYklLjZFdPVvzlcnlf67XH48E777yDmzdv4u2330YmkznWqJmGRPb09ODChQsYHh7GzMwMHA4Hu9JLpVKeJ3QwgCIItYD37t3D3NwcFhcXcenSJbYNeBIhk8mgVCoxNjbGgzGHh4eh1Wq5++7SpUuYm5vD73//+2NzpKf3r7u7G263G88///w+F3Kqcuh0un1Gh8KsqVKp5Nlsv/rVr/DBBx9gZWXl2M+WzwPpGoeGhrjT+sUXX2QCR9kdmnlFd0W9XsfOzg4WFxcxNzeHZDJ56IHyFyI71GZsNBoxPDyMiYkJOJ1OtLe3o6Ojg22fyQSQWukoJatWq6FSqdjEi/QOi4uL7LB83PNLPg8kJCOXz46ODkxOTsJut8Nms3GUKUwp068prezz+XD79u1Dr00eFSQSCZxOJ4aGhnDq1Cn2YalWq4jFYtjd3WU3U7pUm2EPpVIp9Ho9uru7YTabYbVaOStALx11KuVyORQKBfZEeuaZZ3DmzBl2BdfpdJBKpWyxf+/ePdy4cQPz8/PH5slCESTNz9Hr9ejr6+MMI82GqlarKBQK3FZP4l2bzYaOjg48++yzMJvNyGaz+Oijj3Dt2jUWJB+3toMiep1Oh66uLvT29uLrX/86urq6YDQaYTKZoFAo+IwRTqSnA5iyO9SRNDs7i1u3bmFxcRGpVOrQnFq/6Pqoq+iFF15gck2lKiJuwi/h3z3464PBFe21MPsubCxQKpVcRjjKUiUFGRRoCIfoUpRPPztl9ilj3t3djZ6eHvT29mJ0dBQ9PT1Qq9VIp9OIxWK4ceMGLl++jLW1tWMtpUulUphMJpw5cwZjY2M4e/Ysm1aq1ep9iQIirPR8C2dC0TBaGr9w3OW4zwN1Ofb19WFychIulwv9/f3o6+vb578mHJpMz2ipVEIsFuPRUDQE9TDxhciOWCzmIZaURuzo6IDdbgfw8QMt9JsRzlSiB5xalinVur6+jp2dHYRCoaZnsNSi29/fj+7uboyMjGBycpKnvTcaDXaRPgi6dAKBANbW1rC1tXUMK3g00HBLiqLHxsZgMpkAgI0jQ6EQGy7S3znYsXZcP7fD4eDuop6eHp7grdPpWGBMBCaTySCdTqNUKmF6ehonTpzgwbaUFSiVSkgmk1hfX8fy8jK2traOJfNB2QAaE6DX62G323H27FnYbDaoVComAZVKhYcmUguvSCSC0+lER0cHOjo6kMvlEAqFMDs7i4WFBXg8nqbQCNCF53Q6MTU1hfHxcZw/f573j84LMoqkM4cueGEGj5yGFxYWsLq6ip2dnWNfIwWQ7e3tPMrD4XDwDCvCg6wdDhIeoSCUnM0pi05BqJBc6PV6vnDItfawiTud+TQ/kcrhNL+KAgnat3Q6zSTAYDBgenoaU1NTcDgcrBck4ur1enH16lXcuXMHfr//WLOQNPdwbGwMp06dwtjY2L7ZgQczq0ICKISQuNIzTNrWZggiCbRHdrsdvb29OHv2LCYnJ7nKQdlUKlfRWg9WOoi0RqPRI3kfH5nsUG1/cHAQk5OT+NrXvoaenh5+UGkSNo1GoEOK7LvJ6ZIeanLK3N3dxb1797CysgKv13voCz1MUBr91KlT+P73v4/u7m7O6Ag1Sw9KM1cqFWQyGWxtbbGfRzNDLBZz98d3vvMdTE5OYmxsDBKJBH6/n+vJsVgMMpkMnZ2dPFKAnoXj6oYwGo3o6OjAN77xDZw/f54HIdLBIZwATbojmitEREI4RZpKBOT9sbOzg2g0eixW+/QuTkxMYHBwEC+//DLq9Tp0Oh3Gx8c5ghTWxcldWHjwyuVyyGQy1h+Fw2HcuXOHNQLHDere7Orqwvnz53HhwgW4XC5YrVYexZJMJrGzs4NqtQqJRIKTJ08y2aN/g/aPLsilpSX4/f5jf/+EJmujo6N45pln4HQ62dFZOF7m0zI3D/o3KcMwOjoKq9WKkZER+P1+FItF1ixRdsVoNPKZdOnSJQSDwUPtSJNKpVwGdrvdXAGQy+XsnqvT6fjdazQasFgsvF+NRgPt7e1ob2/nTCXpXOLxOLa3t3Hr1i1EIpFjzYDQflWrVdjtdnR2drJPDgBOAlAQSKUcWo/wOVWpVLBarfjKV77CBGpxcZFLkc0i81Aqlejs7MRf/dVfYWJiAidOnNhXegXwmc8q8LERJvGDo8Ajkx2ZTIaOjg709vZyVoPMAAuFAtbW1hAKhRAIBJDJZDgDMjMzw3M+5HI5v8DkOEt23vl8nje9mdirEBqNBi6XCy+88AJGRkZgtVo/0Vr+WZtLWS4aG9GsGhcClQ7OnDmDqakpuFwuniFF5o8kQqcugmw2i1wuh0QiwcSH5rmQdkRowHjYEIlEUKlUGBoawvj4OC5cuID+/n4YjUYoFIp93S7Cz59SyaTfETosU4YEAFQqFSwWC8bGxrC8vIx0Oo1IJMJ/56gzWiQU7+vrw9TUFLsdU/ZUrVbzPDk6FBUKBdfLSdRLn0Oj0eCJ59FoFAD42TzOQ1UsFkOr1aK7uxtjY2N44YUX0NvbC5VKhXA4jLfffhs+n49Hc9C+u1wu6HS6T5TDaRI9AO4GoWzdcYHE1pTxEHY40nN4MBNA5wuR7/n5eR5WWygUOPuh0WjQ1dUFh8PB0TONrKG9J7IfjUZRr9dht9uRz+cPzf1bJpPBbrdjeHgYg4ODOH36NBNRiUTCmX6FQrGP2FHnLmXt6M/RmhuNBqRSKSwWC090J8drmvYu/MyEeryjhLADiZpTaF3Cc0Hot0bDg4XnI90PQ0NDAICuri5YrVZsb28jEokgHA7v8yIS+p99GZURem5tNhtefPFFnDlzhn2OKLnxoHtQ+FzTPprNZrhcLgwNDWF1dfVQ52IBj0h2KAIxmUys0TGZTJwiT6VS2NzchMfjgcfjQSKRgEwmg9FohNVqZV0BXRjEgPP5PNdYKdKgGrtwyrbwgTlOGI1GdHZ2YmxsDE6nk4fsPQxpIQZLKnXh8LNmhcFgQEdHB0ZGRuByudDW1gaxWMykNhQKoaOjg4W7QlEoOe/GYjHs7OzwgVwsFjlVXq1W95GfwwB5jPT392N8fBzDw8Mwm81QKBQANYk9OwAAIABJREFUwAfgwRdSKJjL5XLIZDLIZDJIpVIwGAzQ6/UwGAy8f319fejt7WUBr/DAOspsFmVsHA4H+vr6MDAwgI6ODvbtIHJJAniRSMSDBbu6uj6hEaADk3QOJpOJh0Mep0eJsGTe19eHwcFBts7f3t7G7OwsNjc3EQ6HUa1WeV8O7gN5WpGrMo1xMRgMUCgUx+7nJezAEmYahcaI9HvCC45Kcjdu3ODBs9lsFv39/ejp6cHIyAh3iFIbOmUthaScSAJ1f1FA8Lj7Tuuy2Wzo7+/H2NgYpqenec4VERIiXcB+XQ+BPJNyuRwTAaoOGAwGOJ1O9Pf3I5fLQSwWI5vNcolOmAUqFotcdThKHMzsCz9rIuBCw8t6vc5VETojzWYzjEYjbDYbRCIRTCYTxGIx2trasLe3h/X1dVQqFX4OiCQVi0XO/BzVeyssL7a3t+PEiRMYGBjgtvoH3YXC8VGlUokDSMpCdnR0YGhoiEfSHOZswUcmOzKZjNs7KcIXiURIJBKYnZ3FpUuXsLW1Bb/fz8P2qE1ZKpXydHPaZKlUCqPRyK29Op0O8Xgccrkc2WwW+XyeD26qNR93+m5qagqnT5/GyMgIDAbDJ7IDBOEDLtQQKJVKWCwWnD59GtevX2/KgZEEkUiEoaEhnDhxAs888wzX/XO5HC5fvoxcLgeDwYBTp05xS69Wq91HUuPxOCKRCBYXF7G7u8tdaNFoFLlcjm3QiQA9LsRiMdRqNU6cOIEXX3wRJ0+ehMvl4j2iQ+BgBk4YGRWLRczPz+PevXtYXFxEJBLB8PAwhoeHcfHiRahUKhgMBszMzAAAJiYm8Kc//Qnb29tIJpOs+zkqokDvlcPhYDdn+tyz2Sz8fj+uXLmClZUVzM/P84UwODiI119/HWq1et8eEclxuVysOXvrrbcwPz+Pra0tjhYPruXg53fYIOO4mZkZnDp1Ck6nEwCwvb2Nn/3sZ5ifn0cymUS5XObGB5PJxNPoi8UiT71Op9PQarXQarUwmUy4ePEiQqEQZx6PUydIlxgROa1Wy3rGcrnMmVMKKoPBILs93759G9euXUO9XmfCYLVa4XK58MMf/hA9PT3cNEHZPCIPRNCJ2NOleViBB+3H2NgYdyW1t7fvexeBTzpC0+8B4MGQHo8H29vbSCQSPGbAZrPBZDJBpVLhb/7mb/Dhhx/C4/EgGo3yfaXRaFAoFJBOp+H3+9nA9agEzI1GA4VCAalUCul0GhaLhX3UiHiVSiUmZkRQFxcXsbOzg93dXeTzeZw/fx6nTp3C1NQUa+pGRkYQi8UQCARw9epVbtEW2n7EYjEsLi4il8sdmRZNIpGgra0NJ0+exMzMDF5++WV0dHTw8/VpILsV0uUkEgm43W42LyWnbyLZSqUSi4uLKJfLh0LcHonsUOqbWsiEZkDJZBILCwtIJpOsG6COLbvdjq6uLpjN5k9kMWjqOWVz4vE4q/Cz2Sybtc3Pzx+JaO6LwGQywWQysUfQQaJDGatMJoNsNsueCbR26uRyuVz42te+BoPBgF/+8peIx+NNZ9pGhyxpV2q1GjKZDM9nsVqtcLvd6Orq4ojw4BBQvV4Pp9MJ9/8NnczlcohGowiHw4hEIvB4PHj77bcRjUYPhRxoNBrWKLjdbo6MgE8f8UEC+VgshvX1dSYqH3zwAXw+H4LBICqVCvx+P3Z2djA0NMS1eLVazfOx7HY7tra24PF4cPPmTaysrBxZdEWXyMDAAF9i1WoVH3zwATY3N7G5uYnd3V0kEgmIxWJ0d3ejv78f09PTXHauVCpcZkyn03C5XFCr1bDZbDhz5gxUKhUGBwfxxz/+EX6/H5lMhsW/dDkJy3aH3bElzF7p9XoOrHK5HNLpNHK5HDo7O9Hd3Q2tVovx8XFYLBZYLBbI5XIEg0FkMhme6h2LxXDmzBmcPHkSPT09TP6i0SgikQgfqsIz5uDBfRTvJulS1tbWUK/XYTKZMDMzA4PBgHw+z/PISqUSjwvY29vj5zSZTPK5S8hmswiHw8jlcjxC4uTJk1waIU+aTCaDeDyOWCyGZDKJQCCA1dXVQ9PraDQa2Gw2nDhxgjvnRCIRm8YlEgkYjUYW2dN8pGKxiEQiwYaQm5ub2NjYwM7ODnQ6HdLpNKLRKE6cOMFdiP39/VxCL5VK+4TApK+7du0agsEgBzWHDTr70+k0bty4gUwmg97eXqytrTEhpxll1L1ZKBRYLxWJRBCPx1GtVrGzs4MPP/wQr7/+OoaGhrg8ZLVaufOpXC5zxxb58ZDerlKpHBnZodLkSy+9hFOnTsHhcHBp9EGfSbFYxNzcHObm5nDv3j2oVCq88MILOHnyJNxuN5esJycnAQADAwOo1Wo8uoWsBL7oGfNIZOdg66MwrUopNMrcAPejC0rDuf/Pi+dguUcikbBQTSKR7Gt9zOfzLDjb3Nw8kt77LwJywxQ6Ih9EtVpFIBCAz+dDV1cXbDYbjEYjR16UAhwZGUG1WsXi4iLu3r3bNF4mwMciSOpcEna01Ot12Gw2dHZ2wu1289wvYUuh8O9TNE1RTSKRQDweRzgchl6vx9LSEpOOx91jpVLJ6VXyXDkYMQpLBPF4nMeUzM/PI5VKIZVK4ebNmyyAFYlEHEEtLi7yO0DEl5zASRB7586dzxXlPQ7IwoGiWpFIhHw+j9u3b2N+fh4+nw9tbW3Q6XQwGo0YGRlBT08Puru7uexMY0p2d3d5dhCRQ6fTiVqtBo1Gg3A4DIVCgVAoxO+gsJwiPAMO0zKCCBV9thQs1Go1dptta2vjoZeTk5OsnUsmkwiHw9ylQ8RPJpPBarXCZrNBp9PB7XYjGo1idXWVo2QhIRYKgglHoceq1WoIh8MQi8WYm5vj8kwmk8G9e/eQTCY5uxGLxZic0Wd/MEiizEGxWEQgEIDD4UCpVOKOKK1Wi729PaRSKSSTScRiMeRyOaRSKe6oPMzMDhEd8nyiEmsymeQuRypnp9NpzlxRWWd9fR1erxehUAgWiwWJRAKRSAQGgwEOhwNGoxFarRbt7e0AwPcTaWEoe07B6VG+m+RxtLq6imw2i729PayurjLZ0Wg0PCaISEA+n2dyTs7WiUQCgUAAHR0dyOfzPBiU7CUMBgPrk5LJJLRaLQDA5/NxIH5U2ldq3nC73XC73VCr1Z9JdEgPeOPGDVy/fh0LCwtoa2vjfevs7ORGCdJCKpVKbG1tYXFxEYFAALlc7rHW80hkhwhNPB7nvn868DQaDSYnJ/HKK69Ap9NBp9PBarXuM8J6kHCXfDBocjSVGKiuNzMzg2vXrsHn8zUF2aGU+IM2lyK1K1eu4Ne//jVmZmZw8eJFnDx5cp/TqVKpxPj4OOx2O2QyGf7hH/6hKWfyUHmAupeohDIyMsJ7KyyJAB8TJaFg+6C/hsPhQE9PD/r7++H3+yGTyRAKhR6b8JFfidvtZhdd+pnof4UE/ebNm7h27Rr++Mc/Ym9vj2vmByP8UqmEdDqNn/70p/jWt76Fs2fP4ty5c7wet9uNZDKJlZUVrK+vo1AoHEkmgDKD5NtBP1sgEMDvfvc7XLt2DbVaDX//93+P06dP49y5c2yyJ5FIWDju9/vx05/+FKurqwgEApiamsL3vvc9fOUrX4Hb7eZSQV9fH/7whz9geXkZXq+Xy8vUqUUXLml+Djv7Shke0luRBcL3vvc9nm+lVCpZZxWJRPCHP/wBs7OzuH79OrvONhoN+P1+PkO+/vWv49lnn0VHRwfi8Tjm5uYQDAaRy+X2ETnSk5BHj/C/HQYoKxaJRJBIJLC5uYlLly5BLBYz2RaWER/2+9ZqNdZsbWxs4MaNG5ypbW9vx+7uLpM7Ik6Heb5SMOByueB2uz/RyBKJRKBWq3kPG40GdnZ2cO/ePbz77rtYWlpCJBJhIk7rp+fBaDQiEAhgYmIC3d3d6O7u5uBGp9MBAJeT7t69i/n5eczOziIej3Pm5yjuEyIg165d47PmQX/u035NID+2f/3Xf8XAwABGR0fxk5/8BIODgzCbzftsCSiYzGQy6Ozs5OCUbDUOE8LznCo4n6VXzeVy8Hq9WFxcxL/8y78gGo2iWCzyKIytrS0899xz/PNSE5TNZsPY2Bh+9rOf4cqVK9jZ2Xms/XpkspPNZnH37l3YbDZUq1X+YE0mE77yla/s+5CptZXwaV0qn9ZKKYy8E4kEfve73/FIhWYgO9VqlWukNK0V+FjUGg6H8U//9E+4c+cOlzWIwV+4cGFfhkQkEqFarfIl32xdaHTgkukaACY4BMqO+P1+hEIhrKysALiffXA4HKjX63xBmc1mzuKVy2Xs7e3h1q1bmJ2d5dbhxwVFhiSMJh8P4L5LaSwWw9bWFoLBILa2tjgT4vF4WNh38MKmPalUKtjZ2cFbb70Fj8eDWq2Gzs5O9gt588038eGHHyKVSh1ZyZVIp0wmQyKRQKlUgkQiQVdXF1577TU8//zzcDgcOH/+POx2O/R6PTKZDE9k/9///V/4/X4EAgF4PB4OXJaWlvCLX/wC8/Pz+O53v7vPsO8b3/gGXnjhBeTzeS57bG1tYXl5GXt7e9je3j70Theq8S8uLqKtrQ2VSgUdHR2QyWQwmUx49tlnOWtXqVQwPz+PlZUV3LhxAwsLCwgEAp/Yh3K5jNnZWRQKBfT398NisaC3txd/93d/B6/Xi1wux4LldDqNvb09zM/PIxwOIxgMHqqeRQjKMlJ5xe/387oetzlDmHkjfdb29vah63OEoGCuo6MDfX19+8TCarUa586d43llMpmMB7G++eabWFtbw/LyMhKJBDtfC/eQMlmJRAKXL1/G3NzcPr2WRqOBRqNhsuD3+xGNRtnHhfQzR3nOCgXIj/N9iIiSWWsoFMLY2Bh3J5LuKpPJYHl5GR6PB4uLi9jb2zsyITb9THt7e/y5HvSUo/JcMpnEpUuXcOvWLdy4cWOfNQCVqNbX1zE7O4vh4WHY7XaWx1DV57nnnkOlUsF77733WLKARyI7FIFEo1Gsra3xKPr29nbodDru0hEKy+gFKxQKrBJvNO4bWimVyn3uoFTioQ8zHo9jY2MDV69ebSoBbzqdRigUwubm5r4W3lKpBL/fj9XVVVy9epXTxKVSCaurq3A4HKyFEGa6FAoFl/+aoTNEiEajwWlz0hRRhobSk5lMZl9NfXV1FcB9stPe3s5CdereI0MxMlacn59nT5fH3WO6IHK5HHZ3d1l/UK/XIZPJUC6XmeTQ/3q9XiQSCY7oH/Qz0KVB9f9qtQq3241gMMhR661bt7C5uXlk4seDwQF1X1QqFbS1tWF6ehr9/f2c2ZLL5TxzbmNjA/Pz85zBiEajHNED4G7KfD7PehaHwwGz2cwlBtrzXC7H4lb6OuwsFp03gUCACTQJGkkfRutPJBJYXl7GwsIC5ufnWdNyUPBer9cRiUSwtraGW7duYWJiAu3t7ejt7eV1krA5lUphZ2cHmUwGtVoNgUBgX3nrKAiPkJjQ7x3Gvwtg37/5Zbgky+VymEwmHphLBIDErdQdt7m5CZ/Ph3v37mFhYQE+nw/hcHgfGTsIIobBYBCRSITPJCpj0rNBglihf9aXicPYPwqi6X0rFApMNIgMFgoFbG5usm7nqLWflPjwer3c1EDJjVqtxk0ogUAAd+7cwb1791i3JCTutD83b97kyQtCHkFZns7OTiawX3Rdj0x2arUaW637fD6IxWK89NJLPEmYLLCBjz0tqE2UrKAbjQZP7SVxIQlg6YOMx+O4du0arl+/jl/84hdNJdqldFq1WsXXvvY11uKEw2HcvHkTf/rTn7CxscEbWywWWefx7W9/GxaLheuppCH45je/id///vcsPmwG0Bo3Nzd5BIHb7eYONCI5q6ureO+997CwsACv14tiscitiXq9nt16aYSEXC6HXq/n0Qwk1Dusy6NWqyGXy2F2dhbLy8uo1+vwer1oa2tDo9FgwVupVOL0+MEI+kEgzRGZQWazWU6ZR6NR7O3tHfnsIWEQAQCZTAbJZBImkwmnT58G8DFRoCGQ7733HlZXV3H37t192RzheqvVKqLRKE9Zdrvd6OzsxMTEBO+T3+/H5uYm75lQRHrYoDXs7e0hmUxiaWkJe3t7+PM//3O43W7odDoeVbKysoIPPvgAOzs78Pl8+zKRB0Fk9d/+7d/w6quv8nDX9vZ2Hh1SrVZhMplgMBhY7yPs4KE/Rz/nUaz9sP+9L2t8AmmtVCoVG+tR23uhUOCSY6FQQDAYxH/+539icXGRW6mFprSf9jkIL8t8Ps+/LxaL9wXFzRIcHwaE9gnLy8tYXV3Fu+++y6Xyo8jOfRZo/z788EMUCgUuZwH3732aOL+xsYGbN28iHA7vIzqEcrmMaDSKn//859zJS/Y0RHjoGbLZbMjlcl+4wiP6rL8kEoke+B+F4+qHh4d5oimVN+ggoIuElP/pdBoSiYRV6aQwpy4vqnFms1nMzs5ie3sbGxsbj7y4RqPxUAq0z1rjg0CdAxaLhX0spFIpYrEYwuEw9vb2EAqF9r2s5FJ64cIFPP/88xgdHcXg4CCUSiVnef793/8dH3zwAX7zm9881IYe5RqFUCgUcDqdGBkZQVdXFxuxUUdEOBzmadFEHv7v++4zzhJGxWR7/nn6h4dZo3B9dNBSxo2+f6VSYTJNc6+E3/tRDwtaG/mi0MX8aS/0Z+GL7KFIJGLh8czMDM6cOYPBwUFMTExwxqFcLmNhYQG7u7tYX1/HnTt3EA6HOWL+rM+cPkM6eBwOB3diUVcQPdv0nH7Wmh/3OaWfhzy7JiYmuLNTKpUik8lge3ubBa2f5zFC/57BYMD4+Dh6enowODjIayEiQ1kW0gt6PJ59ejLhZ/hlvYvHiYd9F4UZ68HBQfT09GBsbAzt7e1QKpU85DmdTmN5eRnr6+vcsEAX+nF03T4Jeyhs/KDn81FwmGsUi8VsXkmzAwFw+SqRSOzLID/oThOLxdBoNDh79ixGRkYwPj7ODukulwuhUAgLCwv4j//4D8zOzn6upceD1viFp54LI1wATFiE/glkXEWRNmV6SKQqlUqxu7vLZloGg4E3sVgsYn19HbFYrOkYeqlU4tp6o9HgbBYJN7PZ7Kc6t8bjcdy6dQtqtZrTkdTCTySRSlzNNP+EWpQ3NjYQjUb3CV3JJyeRSOwzpSM86NA6So8LoYcIQSQS8f9/kGfMo34fKs8Kf+/LAL0f5Buyvr7ORo3kMFsoFHD79m3s7e3B4/HA5/Oxq/XD/PtUrqKsTaFQ4M/08yJnYZv/Ya2XIttYLIa1tTUeqEhDWaPRKFKp1MMGCXwmeTyefe7e5FUjJOwej+cTrd1Hsc6nBUJ9WzAY5FlsXq8XMpmMs26ZTAYejwepVArFYpHP1RYejC87g/NZqNfrSCaT3LhBGW46J0gc/nn7Slmira0tziCHw2E4HA7WKW1tbX1mqfxhyspfOLPzad/sQfi8H4LqrdSpBIA7sb7opfhlsfRHPfDEYjGGhobQ1dWFnp4ezuyQnsTr9eLNN99k/chn4TgiEZVKta+Mcpgv3qcJ1B81s/Ok4YvuIZUDnU4njEYjNBrNvjb7XC6HlZUVxONx7kb6MvBpkWczR8yUoaOsn/D36ecXfjUa+80oBb/ftGs8LDzqu0haTLLZEM6cI/J83ANYhXha9/CAW3bTrpF0XqQ3tFqtcDgc7AW2ubmJvb09Jk/CSgGt8bPexUMjO4cB4SHyuAy2mTeVvC5oEi4dCNVqlc2mHuZyOo41ftnRbIvs3MenrZHKMcJhgsL9oS6W446Wm/ldFHzvz/zvzVJSPk580XfxYDlb2I7dLFkKoLWHQhzXGoUla5IjCLsUH0ag/ESQncNEs2+q4PvzRfWobaBPyhofBy2ycx+tNTY3Wmu8j6d9fUBrjV8mhBYtwMOR40PX7LRwOCD9QDOMwWihhRZaaKGFZsFhdtZ9/pjuFlpo4YkCdbt9XmmmhRZaaOH/F7TITgstPGUQCPWO+0dpoYUWWmgKtMhOCy08hThuUXILLTwJaGU////BZwqUW2ihhRZaaKGFFp50tDI7LbTQQgsttNDCU40W2WmhhRZaaKGFFp5qtMhOCy200EILLbTwVKNFdlpooYUWWmihhacaLbLTQgsttNBCCy081WiRnRZaaKGFFlpo4alGi+y00EILLbTQQgtPNVpkp4UWWmihhRZaeKrxmYNAm2Xy6RfBkzbd9Yugtcb7OIr1iUQiqNVqHr1QKpV4/IJYLH6k6fSfhePeQ4VCAZFIBJFIhGKxSN8LYrEYtVrtqVijWCyGRCKBRCJBqVQSfr+nZh+/DLSmnt9Ha43NjQetsZXZaaGFA6DLXy6XQ6FQQC6X7/t9sfjpeG1EIhFkMhnkcjlkMhmvj9b4tFjpi8ViSKVSSKVSXtfTtI8EWlcLLbTwSXzmuIinkd0dRGuNzY1WNHkfrTU2Nx52jWKxuPGkjuhpvYv30Vpjc+NBa/zMMlYLDwexWMwRY71eb02cfgpAJY6nCVTOUSgUnNXI5XJcynna1tuMaH3GLbRwPGiRnceAWCzmModKpWKdQ6VSQaVSQblcRqVSaU2gfoJA5Q4qeQAflweIyJZKJdRqtSdqX6VSKeRyOZRKJYxGIyQSCUQiEZLJJKrVKq+F1litVvnZfVIvaNLpyGQyDkjoHX3S1khlN1oL7R+VrRqNBq+nUqkc80/bQgvNh2MlO8K6OQlBnxTI5XLodDoMDw/D5XKhr68PMpkMpVIJ+XweoVAIa2tr2NzcRDqdbvrDtIX7z6PFYoHNZoPT6YTZbIZMJoNMJoNWq0U6nUYikcDCwgIikQiy2Syq1epx/9ifC4lEAofDAbPZjPb2doyMjEAmkwEAKpUK8vk8SqUSFAoFstksMpkMAoEAdnd3kUgkUCgUjnkFjw65XA6j0QibzQaXywWj0Qi9Xg+VSoVUKoVkMgm/3w+v14t4PL5PuNxsoKBKr9eju7sbJpMJVquV/5tYLEa1WoXf78fu7i42NzcPTVzeQgtPC740siMWizkDotfrodVqoVarOTKp1+vI5XKIx+PIZrMoFApNnVqXSCRQq9V48cUXMTo6iqGhIUilUtRqNZTLZezu7mJ1dRVLS0u4evUqEokEcrncE3E5fh4oolQoFFAqlVCpVJBKpcjlcigUCiiVSk9EtPxpMBqN6O7uxvj4OE6fPg2lUgmpVAq1Wo1MJoN4PA673Y6dnR2EQiH4fD6k02mUy+WmXC9lAdxuN0ZGRnDq1CkMDw/znjUaDZRKJZRKJchkMhSLReRyOWxsbODGjRtYWVnB0tISqtVqU65PCLFYDJVKBaPRiK6uLgwODmJiYgJDQ0NQKpUsOC8UCshkMlhZWcH169exurqKtbU1VCqVplojlRw7OzthNpvR2dmJiYkJOJ1OtLe3cycdcL8c6fP5sL29jatXryIUCiEejyORSBzzKr4YSDxPAnqNRgOFQgGZTIZGo4F0Oo1MJsNl2BaaF0TWVSoVtFotZ84VCgXy+TxyuRyfoUdJ0o+U7FDUQalkpVIJjUaDwcFB2O12mM1miMViTilHo1Fsbm4iGAwyaWjWB5nITk9PDwYHBzE0NMQvYq1Wg9VqhdFoRFtbG3w+H6eZn1SyQ1k44Z4aDAaYzWa0tbVBqVQiHA4jFoshkUigVCo9kWuVSqVMeKanp6FUKvnSKRQKSKVSKBaLsNls2Nvbg0gkgs/nQyqVasrsgFgshlKphM1mQ3d3N4aHh9Hf3w+1Wg25XI56vY5isYhyuQylUsnkx2QyIZ/Po1KpwOv1IpvNNnV5RCwWQ6PRwGw2w+VyYWpqCmNjY5iamkJPTw9nkBUKBarVKorFIrRaLQcg4XAYyWQS5XL5mFdyHxQcmkwmjI6OorOzE263G+Pj47BarbDb7Ux2Go0G8vk8bDYbrFYrstksVlZWUKvVkEqlnpiMuUgk4tKjRCKBVquFRqOBTqeDzWaDVquFSqVCpVKBz+dDMBjk0l2tVjvuH/+RcbB7Tvhr4b3XzEH/54GIjlar5SCE9lWpVCISiSAajWJnZwepVIqTHEeBxyY7B1sdhe2rxMjpMqSFv/7665icnMTAwACUSiWn0peWlnD58mXcvn0b169fRzKZbMoLBLh/aDqdTsjlcjQaDVQqFRZ+yuVy9PT0oKurCzMzM/B4PJBIJKhUKigWi0/ciykkq2q1GjKZDAqFAj09PRgbG8PIyAi0Wi3u3buH5eVlLCwsIBQKIZPJPFEvqUgkQi6Xg1wuR1dXF2w2277oWavVwmKxwOVyIZlMIhAI4He/+x0uX76Mzc1N+P3+pisfyOVyWK1WOJ1OGI1GiMVi5PN5VKtViMVijo5LpRI6Ozuh1WphMBgwNjaGWq0Gg8GA3d1drK2tIZlMHvdyPgG6IOVyOUZHRzE4OIhTp07hq1/9KkwmE7RaLWcby+UykzqdTofx8XGIxWI4HA6Uy2XMzc0hHA4f95L4DLFYLDh9+jR+8pOfoL+/H1arlQlptVpFuVyGSCSCVCqFxWKBxWLB8PAwdDodNBoNqtUqQqEQZ1qbGRRAtbW1QaPRQKvVorOzE06nE93d3RgdHYXFYoFer0c0GsX169dx584dzM3NIRqNIpfLHfcSHhoUOFKGg84X4a9JV0aJAKGu7kkBZW/MZjOsVis6Ojrwne98B319fejo6ECj0cDq6ipWV1dx6dIlLC8vIxgMHlnA8Vhkh1402ji9Xg+LxQKj0Qi73c4RZW9vL5xOJ3Q6HXQ6HVwuF3Q6HdRq9b7Mz9jYGB9Cer0ely9fxu7u7mGt9VBRLpcRjUbh8/nQ1taG9vZ2yGQySKXSfdGJWq3GX/zFX0AqlaJer3NmoNkJD+2tWq3mcpVOp4PdbsfIyAhmZmYwPDwMo9EInU4HqVSKyclJBINB/M///A+uX7+Ozc3NJy6NnsvlEIvF4PP5kM1m2UxQ6NEiFouh1WrhcrkLjH2rAAAgAElEQVTwzW9+E3K5HDdv3sRvfvOb4/7xP4FarYZsNguPx8NZG7VajVKphFwuh3A4jGg0ikQiAa1Wi6mpKYyOjuLEiRNQqVSwWq2wWCzweDzHvZRPBWU/+vr68Oqrr6Kvrw89PT2Qy+UIBAKIRCJ4//33EQqFkEwmmdw4nU6cPn0a5XIZarUaer0eUmlz9GtQdlgul2N4eBhOpxNtbW2Qy+UoFAoIhUJYXV3FrVu3oFKp4HA48I1vfANarRZKpRITExOo1+swmUyIRqPw+/1NHXhQNrW3txeTk5Po7OxEb28vHA4HtFotSx6oGaSjowMulwvnzp3Df/3Xf+HKlSvY3Nxs2sCYQAJzuifo13q9HmazGU6nExqNBjKZDGtra5xtzOfzKBaLTVdm/SyIRCIYjUb09PTg/PnzGBoawsjICNxuN1QqFQeRarUaLpeLGwVqtRr29vaO5Gd6bLIjl8v5sHC73bDZbFxfpvRjT08PHA4HdDodtFrtJ7JBRAza2tqY0drtdjZza0ZUq1Vks1m+LNLpNCQSCV+Ier0eEokEUqkUfX19GB4eRiQSgcfjQTQaRT6fP+4lfCakUuk+/UNbWxssFgscDgeTHbfbvS8aMRqNMJlMWFxchMfjQTgcfqLITqPRQLFYRDgcxtraGlZWViASiVAul7kkJ5PJ0NHRwdnKrq4u9Pf3IxwOQy6XN11mp1aroVgsIhgMcsmKsjvpdBrRaBTxeBzJZJI1EQaDAdPT07z/ZrO5Kd9FsVgMnU4Hp9OJ6elpTE9Pw263Q6PRYGtrC7u7u9ja2sKHH36IcDiMVCoFkUgEl8uFnp4ezoKYTCbO4jUDqFmD1ieXyyESiVCpVBCPx+Hz+bCwsIBr165BoVDAbrfDZrNxxGw0GuF2u1Eul9HT04NsNotcLte0AZZWq4Xdbsfo6ChOnz6Nrq4uuN1uGAwGAPf1nPSMNhoNDA4OorOzEwaDAb29vVhcXOTmkGYFBY+kWSXxvMFg4LNVSHbMZjM2Nzfh8/ng9XpRq9X460mAVCqFwWCA2+3G1NQUhoeHuZKTyWQQjUZhtVphMBj4TDWZTFCr1Uf3Mz3OX6YIt7+/H/39/fj6178OlUrFtXOHwwGNRgO1Wv1QUZNIJOJ6ej6fb+q0XbVaRSqVwsbGBrRaLZxOJ/L5PMrlMorFIkZGRqDT6aBSqWCxWHDmzBm0tbUhEong1q1bR1qbPAxotVqYTCZ0dXXhhz/8IUZGRljYSrodAJxibTQanH4fHBzEysoKAoEAvF7vMa/k4dFoNJDJZHDv3j3s7u7C5/Nxpmdrawu1Wg06nQ4//vGP8eKLL2JwcBA6nQ6dnZ3o6emBwWBgy4FmQa1WQy6Xw9raGnZ2dlizUi6XWVdFAQYdRJVKhYWElUoFfX19uH79+nEvZR/oWXO5XDh9+jR+9KMfobe3F4VCAT6fD2+88QaWlpawsbGBdDrNgnmZTAaVSsXlEpfLBalUimg0ig8//LBp/JUoyq1Wq8jn80ilUqjValhcXMTi4iKuXLmC+fl5PievXr2KH/zgB3jttdfQ0dEBp9MJhUKBl156CYlEAtFotGkvyq6uLkxOTuLixYt4+eWXodfruaQcj8fh9/vxzjvvYHt7G+l0Gv/4j/8Ik8kEiUTC5LZZsnIPAmnnrFYrBgcHMT09jYsXL2JiYoKlAcJmnWw2ixs3buDmzZv47//+by5nNeseHgSdizMzM5iZmYHZbIZUKkU6ncbNmzexuLiIV199FW1tbZzhoYzPUeGxnhDaFIVCAZvNhqmpKQD3U8sGg4EPTIlEAuBjoRVdCLVaDVqtdt/lKTQ4a4ZD50GgzM7du3fh9/tx584dmEwm7mx5/fXXMTU1hcHBQUgkEqhUKqjV6ieiI0skEuG1117D5OQkRkZG0NfXx8StWCwilUohHA7jrbfewt7eHrLZLF577TX09vbCaDSyAO0oWfpRodFoIJfLoVwu4/333+fnMpPJoF6vo1Kp4IMPPoDT6YTJZOL08+DgIM6cOYMPP/wQsVisaYg6lUQoFU5lY4oS6/U6Z2gHBgb4cJJIJEgkEvB4PPjoo4+aKkMnEolgMBgwMTGBH/zgBxgdHUVvby/q9Tp8Ph8uX76MO3fuIBAIIJPJcLaNSpAOhwMDAwMwGo2Qy+WoVCoIBoMoFotNc+Y0Gg3EYjG8+eabTOx8Ph+2trYQiUTg9/tZQF6r1eDz+eDxeODz+eB0OjlbXqlUuHxCs8+aCWKxGK+//jqeeeYZDA0NQaPRoFgsIplM4vLly1hYWMCdO3ewtbWFcrkMnU6HXC4HtVqNarXKzQHNFGB8Gtrb29Hb24vnn38ezz//POuRSMpB7yF9KZVKTE5Owm63o9Fo4MaNG1hdXcXKyspxL+WhUCwWEQqFcPfuXWi1WhSLRS6pks3FX/7lX/KZGgwGEQ6HEY/Hj+xneiyyQ8SFtCkWi4X1DSqVir08arUaCoUCisUiCoUC0uk0ALDolbqYAKBUKiGbzSKVSjV19weJBBOJBBMAg8HAgslCobCP1AjnDdFXsxysByEWiznNPzAwAJ1Oh3q9jkQiAZ/Ph0AggJ2dHVy5cgWhUAjlchlf/epXWQRL9ehmj7YeBDIOjMViTNbpWaQuQRIMUousUqmEVqvdF6E1C4jwEAET7guRUoPBgJmZGfT398Nms6FSqSASicDr9WJ3d7epvHakUilMJhPOnTuHyclJdHR0QCqVwuv1wuPxYHt7G5FIBJlM5hMt8zKZDDqdjjM7IpEIpVKpKc+bYrGInZ0dzM/PQyKR8LuXy+W4O46eQQAc/ROI2FLLfSaTOa6lPBBkYSGVSlEul9nKIRgM4vbt21haWsLy8jLi8Thn5ejsrFQqfP42e8ZDKpWyXtXtdqOtrQ0ymQzxeJyF85VKBXq9HjqdjjWSYrEYo6OjiEQiSKVSWF9fb/q1AveTAel0Gl6vFwqFgvWB2WwWTqeTCXmlUkEmk4Hf70cikThSecdjk51arcZCVur2EDrO0uXg8/kQCoUQCATg8XjQ1tYGp9MJh8PBmZ16vc5mX1tbW02va6EsABG4ZDKJtrY2dHR0wGw278ts0IFLJJDW22yg2jLpAMgLgYjO+++/j7W1NaytrWFrawsAYDAYUCgUWGcgbFN/knHQKoBInMPhgMFggFKpBAAuCVUqFSa7zXgg0TNYq9U4W9DZ2Ymuri709vbi1VdfhcvlgslkQi6Xw/r6Ou7cuQOv19tUZEehUKC7uxs//vGPYbfbIRKJkM1mce3aNSwvL2N9fR3xePwTmRqRSAStVsuaCb1ez4dtNBptuuxAqVTC3t4e3n//fV5jsVjkLh16V2UyGUwmEwwGw74zh8oDpJWMRqPHuJpPh0gkwtraGgAgFAohGAwiGAzC4/FgaWkJsVgMyWQS9XqdjRWJ8JRKJYTD4abWIxGISFOJjsxnl5aWEIlEeJ2jo6MYHx+Hy+WCTCaDRqPB2NgYa86uX7/+RHTXkb6sUqlgdXWVz1KXy4Xe3l68/PLLkMlkTHTu3bsHv99/pF11jx16kwBSmMmgyDiRSCCRSCAQCODq1atIJpPI5XKo1+t47rnn0N/fD4VCAYlEgnq9jnQ6jXQ6jVQqhWAw2NSCMwKJCUnP0d/fj5MnT2JsbAxWq5WjELoEzWYztFot5HJ5U5azKGK6c+cO8vk8TCYTMpkMIpEItre38dFHHyEej7MJlNVqRU9PD0ZGRtDW1oZqtYqdnR3s7u4iEokc93IODRKJhLU5Fy5cwNDQECwWC8rlMiKRCBu65fP5pj14KQghf6T29nY899xz6OrqQkdHBwYGBiCVSpFKpXDp0iVcuXIFi4uLTaefs9vt6OjogE6nQ6VSQSwWw/z8PK5du4bV1VUsLCxwFx2BLo7nnnsOJ06cQG9vL/L5PDsOb25uNmX7cr1eZ5JCuiM6T6gTq6enB8899xzOnDmDvr4+NoukDlkidnTONlNGuVar4aOPPsLKygpUKhWy2SySySSXFYH7GSryGzpx4gQ/o8FgkMt4zfrOERKJBDY2NvDWW2+xFUmxWMT6+jpbI+zs7CCdTsNoNKKzs3OfeatarYZarW6qvXsY1Ot1/rlFIhG+973v4eTJkxgcHITX68XCwgKWl5exvb2NVCp1pHfiodQZcrkcotEoZ2NKpRIbsCWTSYRCIezu7nL9uK2tjb9oxku1WkUymeSvdDrdlGTgIIQq+56eHgwNDWFsbAwGg2Gf6ZdMJoNarYbVauXopFkzV41GAysrK0in09Dr9cjlckilUgiFQkgkEqhUKly26evrw9TUFKxWK7soBwIB7lB7GkBdMZ2dnRgcHMTg4CDMZjPrWvb29uD1ehEOhz9xyTYLSLehVCrR3d3NI06mpqbY4FOhUCAWi8Hv92N5eRkejweRSKSpLkixWIzu7m50d3dzm6rX68Xt27extraGvb09DqgIVEKw2WwYHBxEe3s71Go1QqEQ1tfXsbq6ir29vabUtAD3M4zC0ihlc5xOJwcaExMT7JMkNP6k1m2NRgO5XN50z2ej0UAwGEQymeRSVj6f55IbGdGazWZ2AZdKpUgmk9jb20Mmk2lq81kCZTpWV1dRLBYhkUhQLpcRi8XYeoXeT5VKxfciaVqo1NXs6ySQV5LZbEZ3dzebQ546dQrt7e0QiURYWVnB6uoqtra2EIvFjvzZPBSyIxQy+nw+ZDKZfaQnm80CAPvwDA0Nobu7m/UdpH8Jh8Pw+/0IBAJPjPMnWdSbTCacPn2aBZ7kPQOAo7C2tjZ0d3djbW2NiUMzCrEbjQZu3rzJCvlKpcJpyGq1ysZzOp0Ozz77LJ5//nnY7XZkMhkkEglsbm4iEAg0pQndo4Iycg6HA0NDQ5iZmeEOCrpsV1ZWsLy8jN3d3aYdqyCVSqHRaGCxWNgjaWxsDKOjo3wR5nI57Ozs4O7du5ifn2ebhGZZD+3F5OQkxsbG9pWu5ubmsLy8zOVU+vP0fjocDvT392N6ehptbW2QSqXY2NjAzZs3sbCwwN12zQgiOrR+0nOMjY1heHgYk5OTmJiYYMJKZEcqlXJQaTQaoVar+fIEPi5rHvf+CjPAwp+FTOmoK3RsbAwzMzOQSqUIhUJYWVlBOBxuWpIqBOk7FxYWEI1G+W5QqVQwm80wmUzsS0OebUR2stksj8dopsDjQRCLxbBYLLDb7eju7saFCxfQ1dUFl8sFl8uFQqGAcDiMq1evYmdnB8FgENFo9Mg1c4dCdiKRCIvFvF4v15XJONBgMGBqagrj4+M4efIk+vr6oFarWfNQrVa5ZfTmzZuYm5t7IogOcD89brVaMTk5iVdffRVdXV0wmUz7hKpUO7fb7XjppZegVqvR3d2NX//61yyiFE5MbwZ9BM3VoTVQ9NHf388ZrGeffRZdXV1wOp2o1WpYXV3F/Pw8VldXuZb+pEOtVqOtrQ0vvvgiXnjhBUxNTUGn0yGbzSISieDGjRuchm9WogN8LOodGxvDd77zHbjdbjZtk0gkaDQa7Aq9urqK3d1dZLPZpiMAZM6ZSCRw7do1vPHGG9ja2mLBLn3+VMKxWq2YmppiQkD6B6/Xi0uXLmFhYQF7e3tN548kBF36er0eAwMDsNvt6OzsxEsvvcRdPQaDgY0vgY9nS/X396NYLKKjowOTk5O4c+cOkskkLBYLAoEAYrEYAoHAsRrWPej7SqVSWK1WjI2N4bvf/S76+vpgtVqxsLDAbdnUldbsIJJCQl3yd5qZmcHg4CBGR0fR09PDpEcqlbJFRCwWQ7lcZj1kM4OCqtOnT2N4eBjT09OYmZlh6YZIJEIqlUIul4PNZkMgEPjSMlaHQnaKxSISiQTK5TISiQSXn2jysNPpZCt3l8sFg8HA7ejFYnGfJsTn8zWFZfvDgnx0+vv7WY9D2SqqUwo9aKxWK0ZHR6FSqZDL5bC6uopEIoF6vc6dTc2Ag2STHuKenh6Mj49jenoaw8PD0Ov1kMvlWFpawuLiIpaWllho9qQQ1geBBK1OpxNTU1Nwu90wmUyo1WoIBALY2trCvXv34PV6EYvFmvayBMBOrWTuKZPJUKvVeAAolT1UKhUHKel0uumE9I1GA5ubm0in08hms/B6vftK3mTkabPZ4HA40N3djYsXL8LtdqO9vR0SiQSZTAbhcBiRSIStIJq1O5JsAdra2tDf34/z58+jo6MD7e3tGBoaYqsHMvekphE6c3Q6HRv0dXZ2cgZWoVBgcXER29vbiEajTUfUyYGXsgNutxsKhYIDjOXlZS5ZPglyB+BjUqdUKtlEkTRznZ2d3BBC9wf5LFGXXb1e5/e22YIQAmUfhfOwyL9M2PVYrVYxPDzM/l61Wg3xeBy5XO7IMjyHQnbIWyaVSvHvUWRhNpsxNDSEkydPwuVywWazcSRZq9WQSCQQDAaxu7uLlZUVeL3epuwaeBBIh0PZKiJxB9ONwuGZw8PD7Bj59ttvY3t7m83rmvUhpsuSDLHOnj0Ls9nM0+rn5+dx+/ZtLC4uMtlppsPzi4DEvG63G6dOnWJRLI1fmJ+fx9zcHAsLmxkk0DWbzdyBVK1WodFooNFooFKpIJfL2f3carUik8l8wkLhuFGr1XD37l00Go1PjJKhM0er1WJgYAC9vb0YGhrCt7/9bWg0GjQaDfj9fvaJKhQKLAAVzh9qludWOJOto6MDJ0+exCuvvML7Q5Ey8LEJIblkUwlPJpPxZVoulzE5OcmT7Wle4dLSUtN1+FAppLOzE319fXA4HDwo+r333sPOzg6i0SgKhULTnpmfhkajAb1ez6Lyr371qzAajVzlAPCJmVik3yE5RLMPPiXSTV/Uck5jeMhw+MSJE9yVXavVsLa2xvqto8ChkJ1POxzEYjGmpqZw4sQJnD9/HiMjI/ucLuPxOHZ3d/HP//zPPMgtl8txluNJQSqVwu7uLhYWFjA+Po7Ozk5YLBaeSZTP53mkhkqlglQqhVKphEKhwNmzZ2GxWOD1ejE3N4eVlZWmJXrFYhGRSATJZJLNICUSCarVKkqlEjKZDJLJJFKpFORyObvvNsvF8aigS7Cnpwdnzpzhab0ikQjxeBx7e3vweDzweDxNWe45iHg8jlu3bmF9fR2/+MUv2Lb+z/7sz/Dss89iZmYGer0eZ8+exdDQEEwmE/74xz/izp072NnZaYp3krR9D5qdYzAYYLfb0d/fj29961sYHR3F6OgoDAYD/125XM7lkJdeegkej4c9o3Z2dhCLxZrGQJEywz/60Y9w6tQpTE9Po6OjgztYhdko+lx2dnYwNzfHmjqySqDyiFar5RLKhQsXYDQasbW1haWlJRb7HjfoYrfZbOjv78fU1BTK5TJ++9vf4o033uCyYzM8k18EQusR4VelUkEymUQmk4HD4eDRLVNTU9Dr9ejq6kIgEGDfoWbzhQI+7sS+e/cuNjc38eabbyKTySCTySCbzUKj0eDEiRM4ffo0xsbGYLfbcfLkSZw4cQKXL1/G3Nwc3nnnnSMh30fm+kZDB2mg3cGXs1gsIh6P4+7du9zGTE6vTxJKpRKCwSBmZ2dRqVR4rktXVxd2d3exu7sLvV6PWq0GkUiEjo4O9Pb28lwQu93OXQg6na5pa7JkMXDv3j02ojt16hQblpHrrsPhwPz8PLxeL8rl8hN7IFGWQKfTsdU5Pb90ONFn0uxEBwBH/LVaDZlMhjs/3nnnHQQCAaytreGVV17hronx8XF4vV5ks1m2gWiGvfw0QT+ViIeHh/lynJ6ehtPphFarBfCxGWStVoNSqWQNmkajQXt7O/R6Pa5cuYL19fV9rsvNAGHGSiaTcfaYNB35fJ6F1qurqwgEAjxRm7pfacbdyMgI+5tZrVa43W5MTEwgnU4jEAg0hYSAOlx1Oh0MBgP7eKXTaZZLNMvefBHEYjGsr69z4ETTBsh2pVAoYHx8HCMjI+jq6uIxE6VSCTMzM8jn85wxaTbU63WUy2UEAgEmdcViEaVSCeVymacOZDIZGI1GjIyMoL29nRtAyuUyP8PZbPZQ9/lIyQ7V1emQqdfrXMIql8vIZDLsTUKXyJP2EFcqFYTDYcRiMSwvL8PlcmF4eBjnzp3DysoKFhcXYbPZkEgkUCqVMD09jRdffJEnu1OLukQigU6nY0O6Zvsc6CFeWlrivaIDU6vV4uTJk9xmSKMJyuXyE9NVJwQRTuFkYiI4B/flScpeHaz102G7sbGBu3fvYmBgAN3d3TCZTBgYGMDo6ChSqRTm5+eZLDQjaO7QyMgIl1iHh4fZ+qFUKvG8PSIyBKvVCrvdDqvVimw2i2q1Cq/X21RGdZlMBrlcDmKxmDNUwi6daDSKN998E7Ozs/uGYhIpUCgU0Gg0GBoawiuvvILJyUmYzWbo9Xp0dHRgYmICHo8H+Xy+KS5QKsEpFAoolUrI5XKk02m28BAGzc3Yzfp5iMVibNExPz8PnU4HvV6PaDTKz+C5c+dQq9Wg0WjgdDqh0+nQ3t6OmZkZBAIB5PN5RCKRpls73e1UoaCOMgIRHq/XC7lcjnK5DLlcDofDgd7eXjQaDdy9exeFQoHNWg8LR0p2FhcX2VTI6XTCZrNBr9dzVEnjFejPC3/9JEGYiqRp3x999BEfODRTioStlUoF6XQaFouF9RJWqxVtbW3QarVcKmo20AG7vLyMUCiEbDaLgYEBDAwM4Ny5c+ju7ubM1UcffYTbt2/jl7/8JXfKPCmg569cLuOjjz6C3+9He3s72yXodDpu8X0SD1sCGUgGg0EesPjaa6/hpZdegsPhwOTkJBQKBZuCBgKBplwrCas7OzvR29uL/v5+nnmVz+dx9+5dHntB72cmk4Fer8fY2Bh6enowMDCA8+fPo6+vDxKJBFevXkU0Gj12ot5oNPDuu++iWCxifHycSVswGMS7777LbubkTUYdkLRPIpGITetu376NRCKBGzduYG9vDxcvXoTBYMAzzzyD2dlZ+P3+phBqE6FbW1uDzWaDy+XCxMQELl68iLa2NrzxxhvIZDIoFovsQUME8ElAtVpFLpdDPp9nfyGZTMbjPwDg/fff58rHq6++ynrB559/HtVqFQaDAWtra00bbD3I1oC0uoVCAX/6058Qj8exvLyM73//+zCbzTh9+jQsFgt+/vOfc7bysHCkw4tKpRIikQgWFxdx48YNDA4OctcSMdazZ8/C4/EgFovxYLtm6wx4WFC9kkpWwlqssPUwHo+zTolEzTQ8VafTNbWfApUaG40G7t27x6Iz4P70YovFgra2NvT29qJcLmNjYwMrKyuIx+NNJXR9GNB+BYNB3LhxA8D9DkOa0aPRaHiu25Ny0B6E8PDZ3NzE4uIi7HY7LBYLdDodnE4n+vr6MD8/3xQX4adB6NO1t7eHvb09iEQihMNheDwezM7OIhQKIR6PIx6PI5vNolQqQaVSoVQqIRqNQiwWY3h4GF1dXZiZmWFTzWYoxYbDYSwsLOBXv/oVDAYDv4PUCbi7u8umc8JzQ7hfVMYkT5v3338f09PT6Orq4u47oQnqcYJKxMFgEMvLyxwgNxoNjIyM4K//+q+RSCSQTCaxsbGBvb09dnU/7p/9YUFBEhE14XNGNhDr6+soFotwOBw4efIk3G43tFot3G43EokE3G43t95/2c+ocCTUF0G9Xme7GZFIBKvVirNnz3LAPDExgWKxCI/Hc2gi9CMlO+SKvLW1hTt37gAA1yep8+Ps2bPQaDTweDycLaASyJPy4BLo0D14qdP/J1EamS3SGoVtvyqVqqlnSlE2oFarwePxsFhXo9GgWq1CKpXCYrHA4XCgUqlgdHQUoVCIu3+eJNAFkUwmsbq6ylosnU4HuVy+T4v2JIMIeSgU+n/sfelvo9d1/sN933dSokiK2keakWYfezJeYsduDCRNiiANigYF2nwrivwF/VMK5EtQJCmKpln6sx073jL7ohntokQtFElx33fy98E5x5Ss2SWRM+YDCIntGeq9vPe999znPOc5WFtbQ39/P86dOwe5XA6j0QiXywWlUtm146QU687ODld61Go1rK2tYXZ2Fjdv3kQ2m2WjU0oTSyQSvmXbbDaMjo7CbDazpQLdtjuNXC6HYDCIDz/8kAsDAHDZ+MNSxfv3T2rOSAdHOp2Gy+XiVFE3NbGlSt1gMLinStBqteLb3/420uk0EokEVCoVSyMo9fginRt0Id6PUqnE6aq+vj4YjUYWltvtdng8Hvh8PqTTaXYePq5xU6DzvMx2vV5njz61Wg2r1QqDwQC73Y7h4WHk83l88cUX2N3dPZTL/5EGO81mE5lMhhmbcDiMVCqFd955h7UeP/vZz7C5uYnNzU188sknmJubw9bWFiKRyB6tz4sMStGJRCJYLBZYrVZYrVY2H6QF0977pptBbMDu7i67gcZiMcRiMaRSKbz66qv8clarVWxsbLDw7kVBq9VCqVRikSfpVUioXKlUUCgUuuLmfxhotVrI5/MIhUJYXV1FJBKB2WzmKhByBO/G9Gq9Xkcul8PNmzcRCoX4RhwOh7nC6qA5qtfr2NnZQT6fh06nw4ULF+ByuVhPR0FFp9FsNpFIJHDt2rU9Ivmn3SvoMkZNKGntUs+s9g7q3YBqtYpIJMJaovHxcUxNTeGdd96Bw+GAQCCAx+OB1WrF7du32WW4GwLU5wXtP/V6HZ999hlEIhFKpRKGhoZgsVjQarXw3nvvYWtra4/ly3HhMNYJBagk8ZBIJNjd3cVPf/pTfgcDgQA++OADRKPR5957jjTYoQiw1Wohk8kgEAhwZ+zp6WlMTExAJpNxPlIikWBgYADBYBBXr15FKBRCNpvtukOSxgU8uUCOHFBJZe/z+bgUn7xq6vX6CxXc0aZLzRhJvGw0Gvl2rNFouNP7iwixWAylUomRkRHY7XbuaZbJZJDJZF64m+TjkMvlEIlEsLGxwf1tqNqJeit1I5rNJtLpNGq1GhuXkSD5UTseXsUAACAASURBVO8TuZYTm0zVTcRedsvhTxeM9n9+WrS3mzCZTMxQFovFPXqfbgFV9NK8kiyC/IK8Xi9sNhtGRka49H5lZaUjh/9Rgaon2wM5kjyQhlChULAvz3HsRXT+kcD4YezUk4DOkGw2i42NDWg0Guzu7nLhwKVLl/DgwYND8fw60mAH+EpZT/limqz2niAymQwWiwUikQh6vR52u52rD3Z2drom2KHNgm73ADg42R/00LhJj6PRaGA2mzE8PAyfzwen08mmYESl04bbrXqdg0DjLhQK2N7ehlAoxIkTJ+B2u9ksSyKRsL/SiwShUAiVSgWLxQKfz8cl6BTo5HK5rp8nKuOlFOrjbr3kmUQNX2ljo3XfzaD3hwKcJ7040PiI+SiVSnvEot2C511rlConTR2xV8VikVN83bae6QJIgWg2m4VIJOI2Ll6vF319fSiVSrDZbNjc3Oz0Iz8WT1uIQ2OnTAdVH9LZeZypdDrXqPk1nV/PY8hJ0o50Oo1YLIZ4PM6s8vDwMLv0k1nms67RI9+96CYhEAiQTqcRj8fZxTQWi2FmZgZOp5Nz7f39/dxM849//COuXbuG3/zmNx3feNpt27VaLac2aGOkhdjuXEo+EUS12u12vP766/B6vTCbzfwZ+Xwe29vb2NnZQTKZ7MpUweNAgsJisYjV1VWMj4/DYrFAqVTuCQ5fBLQ3kJyYmMClS5fwrW99CyqVCrVaDXNzc13XJPMgkOOqxWKBTCZjz5yHvUuUg9+/aRF71c1jBb5iP57m/SGvHa/XC5VKhUajgUwmg0qlcqD+7kUFrWfyIfqXf/kXeDwelEolrK6uMmvQ6X32INC8ErsRjUbZD2p4eBh9fX0AAKPRCKlU2uGnfTxIuvAkTAyZm5IRLaUaiW0lRue4qrKozyO5sZdKJW75RELiZ1lDdD7UajWsr6/D7XZzEZNMJgOA52auju2qRpsQsRi3b99m51K/34/h4WH88Ic/hFarhUwmg8PhwKuvvgqdToe1tTUsLS1x9/ROgDqbUy6RzL1IUE2UOfBl6qM9wJmenuaUjsPhgFKp5NLYeDyOYDCI3/72twiFQl3DYj0L6JCkclfyNKGN6UUBGc7RfL/22mvQaDScbtzc3MTGxsZDnXy7BRKJBE6nE5cvX0az2UQ4HIZQKEQkEjnwEKeAXqlUwmw282ZKgXy3BzvPAqqs02q1XDhBHiFisbgr0nbErrVXvzzNXNC8Wq1W1r34fD4oFArEYjE8ePCA/V1elDmm1CM1Ti4UCshms12v16FAhYwiqWn2QaCsAOk7dTod5HI5V3EVCgVmJY+rim5/S5b28v+NjQ32hGoPvh71XLSuyc9MJpPBZrNBJBKx8TCty+cd45EGO0R3UUkj5dzq9Tri8TgSiQRisRiLmC9fvszaAJVKBbfbjVqthlOnTiEUCnUs2BEKhTAajRgYGMCZM2f2VONQsFMqlZDJZDht4Pf7YbFYYLFYcOLECa54oCiVykfD4TACgQDu3LmDdDr9Qt8kKW1H1WS0QOm7ehFANxej0Yjh4WGMjo5icHAQEokEiUQCkUgEOzs7iMfjXd8Pi4LrkydPolarQaFQIBwOIxaL7Qle2jcctVoNg8EAk8nEjCwJdV+Ug/BJIRKJoFKpoNfrYTabuUdWt1nxkw+LXC7nW3y5XH4iVoCYAbVaDbvdDq/XC6/XC61Wy+W/0Wi063qgHQShUAihUAi1Wg21Wg2FQoF6vc69zrpt3g6CQCDgs0AsFiORSHyNUaN3TqVSQafTwe/3w+PxwOFwcJEAFUgQi3nczLlEIoHVauVqOEo3RqNRAGALmXbblf37DX0OBTher5e7C4hEIlQqFfZoO4y950iDHTIOJDPBWCyGra0t/iKImlxbW0O5XMann36Ky5cvw+/3c05SIpHgpz/9KW7fvs0laMcJOqxnZmbwyiuv4J/+6Z9YWCwQCFjISDlLkUjEufH2G1k7KDIPBoO4du0abt26hS+++KJrHWqfNMcsl8vZcE+j0UCv16PRaHR9eXb7s0kkEng8Hly4cAE/+MEP4Pf74XK5UK/XsbCwgBs3buDevXsIh8Ndz8INDQ3h7NmzePPNN9FoNGAwGLC6uor19XXeQChAFYvFMJlM8Pl8OHHiBEZGRiCTyVAul2EwGDg3/7KA9Fj9/f0YGxvDzMwMpFIpkskkAoEAMplM17yParUaRqMRbreb3Wnb5/AgUGBAHbapB9HFixcxOjrKWh1iRAqFQlewWA8D7cMqlQpjY2OYmprC+Pg4qtUq+7iRkLVbQaxOf38/tFot5HI5t04ol8s8Z7Q2h4aGMDIygh/84AcYHx+H3W6HWCxGKpVCMplENBplMf1xXURI0N9qtWCz2fDuu+/C6XSiVqtBJpNhfn4ea2trEAgE3NyaUsJUnEQ/AoEARqMRNpsNf/M3f4Pz589jcHAQVquVey1Saf1hMMtHFuyQUHV8fJzZkHv37uHjjz/GwsICBy30v+2C33q9zl+GUqnkmwjZoB8nqEeLy+WC0+mEUqnknh8P8xqgyWwHaSGKxSK2t7cRCoVw9epV/PnPf8bq6mpXljC3Czcfp9kQCoXc3+T8+fNwuVyQSCSIxWLY2dnpChv6dpBfCekBSFt26dIlXLhwAVeuXOFO9q1WCysrK1haWsLCwkLXb6oEo9HIDE29XofRaITf78fCwgKAL3PgCoUCAwMD8Pl8mJ6exuTkJIaGhiCXy1EsFhGNRtnrottvzU8KCgLGx8fx9ttvY3p6GkNDQwgGg3jw4AH+9Kc/IR6Pd0VwJxAIcPbsWbjdbjidTvzud79jX5397yPtR0KhEFqtFgaDARMTEzh9+jSGh4dx4sQJWCwWKBQKZLNZfP7557h9+zbm5+exs7PTddVY7RCJRPD5fJiamsJPf/pTWK1WSCQSfPzxx/j8888xNzf3tVYg3Qa1Wg23241XXnkFNpsNer0e5XIZm5ubiEajkEqlcLlc8Hg8OH36NCYmJjA4OAiPxwOlUgmhUIhUKoX5+XnMzc3hl7/8JYLB4LEH5SQmXlpawrlz52CxWGAymfD9738fZ8+e5b1yd3cXOzs7iEajfL45HA4MDAzA7XbD7Xajr6+PtbrUG5L8sZaWlvDRRx8hEAiwke3z4EiCHaLE+/r64Pf7MTExwQ35KJ9H6Z9WqwWLxQKHw8EBDeXm6LBVKBRQq9VQKpXHajbYXmJHXcv3/7fHMRatVotLO3O5HMLhMLue3rt3D4FAANFo9Gs0ZqdTBkQXOxwOpFIpzosfFJARNUtUZH9/PxQKBarVKmKxGLvWdgOoksPn83EptVQqZVbqwoULOHnyJLxeLzQaDYrFIlKpFGZnZ7G4uIhgMMgl592O9gsEUcWTk5OIxWLIZrOoVCqQy+Xwer3w+/2YmprCwMAArFYrWq0Wkskktra2sLS0tMeM7kWGQCCAWq2GyWTC1NQUxsbG4PF4AADBYBALCwvY2dl5ojTRcTyrWCzG2NgYhoeHYbVasbW1BZ1Oh1AohHQ6DeDLC5lWq4VYLGYW1Wg0wmw24/Tp05icnMTAwAD6+/tRqVSQTCYRDAZx8+ZNLCwsIJFIcKFFJ0GFHXSRpL1fpVLBZrMxozM2NoZCoYBoNIrFxUWsr68jHA53RXD6KEilUlgsFgwNDcHlcsFoNGJrawt2ux2hUAhSqRQDAwMYHBzEmTNn4PV64XA4oFKpWCaxurqKu3fv4sGDB1heXkaxWDx2PR1VLIZCIQQCASiVSu6zptFooNFooFarEY/HEY1G9zQSpmDO81eDVqq6IhPJdDqN+/fv4969e+y+f1jNlo8k2BGLxVCpVPD7/fD7/aystlgsfJBQNFutVuF2u3mSDQYDq9WBr5T4JpMJJpOJX/DjQHtlFZX4lUqlPS/jw0BMCDUWfPDgARYXF3H16lWk02nk83ns7u5y5cf+39np8nPyPPre976H69evY3NzE+vr618LNul7MJvNGBkZwczMDPr6+th7Z2VlBdFotGv0LXK5HB6PB3/7t3/Lng79/f0wGAwwm80YHByESqXiFM7q6iru37+P3/zmN1haWkIoFOr6TZVQKBSQyWSQzWbh9XrZ/+jb3/428vk8dxXW6/UsgKT0bLFYxOLiIm7duoXr168jmUx2HfP4LBCLxXC73ZiYmMCPfvQjuFwuyGQybG1t4YMPPsD169e7xqeFKqiuXLmCmZkZaLVazMzMYHV1FVevXsW9e/f4UjIzM8OVj7VaDSaTCRaLBePj45BKpRAKhWg0GtjY2MDi4iJ+/etfY3V1Fel0+muC0k6NlZySqfhDq9Wiv78fIyMjeOutt+BwOGA0GgEAa2truHfvHq5fv47V1VVug9HNIIuVqakpuN1uGAwGDA0NIRaLIRqN7jGdpXOQ5o0Ci9///ve4efMmtra2kE6nOzJnlKFYWVnB73//e4RCIeh0OrhcLrhcLq6opkae6XQa1WoVtVoNWq12TzEAWV3cvn0bc3NzCAQCuHXrFnZ3d/cQIoeBIwl2Go0GyuUyCzo3NzcxNDQElUoFrVaL9957j9u+12o1KJVKqNVqpljbjfbi8TiuXr2KYDCIbDZ7rKwHlfVls1n86U9/wsLCAv785z9jfHwcVqsVLpcLVqsVMpmMO2NTGd4nn3yCVCqFXC6HUCiESCSCRCKxpyUG6X3ax7PfPKxToMU7NTWFyclJJJNJbGxsIJlMslV7pVLhEsR3330XPp8PdrsdcrkcGxsbWF5extzcHFKpVMcDBLrRT01N4d///d/hdrshk8m4bQD9CAQC5PN5hMNhfPTRRwgGg9jY2MCtW7eQy+W6RsfxJNja2oJcLofBYMCbb77Jm6jNZoPVav2aYy4F4NFoFLdv38b9+/extraG7e1tlMvlDo7k0aAUpEaj4QtIuy6Qyst1Oh1eeeUVjI+PMwNZrVYRjUbx/vvvY35+HpFIpMOj+TrW19e5jYVOp8PExAT6+/u5yk4oFMJsNkOhUHALE7FYzD/km3Tnzh3Mzs5yoBCPx9lbp5OBLPVGOnXqFF599VXYbDZOv8pkMmb2yZj2iy++wPr6Ora3t7G0tMSGrN0Ouji3s1aka/V4PHsKPCqVCnZ3d5FKpXDr1i0sLy9ja2uLU475fL5jc0Ytg9LpNObm5hCPxxGLxeD1emEymeBwOOD5ax8vagJOa4yqyMLhMG7fvo1QKLQn3UVapKOoAD2SYIcYjXg8vifqo2okh8Oxp0kmmc5R/x2yyqaUz9zcHJLJZEcEdO0Ny9LpNKLRKAqFAux2OyKRCB/uVEpOCvJbt25x+4REIoF8Pr/HvKvTlPHjQPNQq9UwMDAAu90Op9OJdDqNTCbDqS2FQgG9Xo/JyUk2EcxkMgiFQlhfX0cwGHxo+uu4x0P9dYaGhtioql2302g0sLW1hWg0iu3tbX4Zw+Ewu/J2OrXxNEin0wiFQpibm4PZbIbNZoPRaOQxA18yHdSIkJqeRiIR3L17l9tGPKo8thsgk8lgNpvR19fH1Sp0GLRaLVitVqjVau7wTfR5rVZDKBTCxsYGZmdnEYvFui6oazQanCoQCAQwm80cwJKejPyU6KCky2Y2m0U2m0Umk0EikcDs7CwePHiAUCjEqSu6WHVyXYtEIi5qoFY6BoOB38tcLsftaMLhMG7cuMGBwIvUc4/amgQCAWY81Go1m36SOWs+n+e5SyaTuHXrFtbW1vjdJCauk2gPeKjCend3F2azGS6XC9lslvtgUlUn8KVDO4mPb9y4gVAohHg8zgxjqVRCqVQ6kvV4ZMFOo9HA5uYml8mNjo5CLpdDpVJBpVLt0RO0ew9QAJRMJjn1Q7TWUX0JjwIFOzs7OxyZp1IpLs/V6/XcRI/ykvV6HXNzc9jd3WUG5EVDtVpFJpPBxsYG/H4/nE4nxsbG9rh5UjWSSCSCzWZjZisQCGBhYQFzc3Occ+0k6Cbldrvh9Xq/1t2ZctCZTAaffPIJ1tbWsL6+jrm5OQ5YX0SkUimmkrPZLPR6Pfsd0WWDtFWUF6e8+fb2NmKxGKe6uhVUueLxeLiiqlwuY3d3d08lJWnuTp06BalUilarhY2NDdy+fRsPHjzA1atXkclkuurgpJvw9evXEQ6Hsb6+ziaslHqlgxIAz3UqlWI2mSrLqEP42toakslkV4mRKW1FTsnEoNLFKplMcnBDh/9hl8ofR7UoXZo/+OAD1lRREQS17dja2sLm5iZXIZXLZTac7TZrkmazyenwaDSKzc1NDliXl5eZ2SEyQyQSIZVKoVwuo1AoYHZ2FolE4tgCVsFjDH+ea5ejLq0ulwvnz5+H1WrlnCwFCOVyGWazGXq9nl/uUqmEYDCIO3fucA7vaYVYrVbriVbv046RRIN0iyLavP3wpNLy9lYSR4GjGiPwZQBjMBgwPDyMM2fOwGKxwGg0smV3+9hpvNRkkJiBYDCISCTyXON/kjE+yfiommN4eBjvvfcea1OazSa3gNje3sbnn3/OKUcKAI6S1TjKOfzr34NYLGbdRnuVIAnsSV9GARBVRB4WjXzU61StVsPpdMLj8UAul0MsFiOfz3OqwO1287gNBgOAL9Ncm5ubWFhYwNbWFnZ2dp4rfXyUY5TL5ez9dOLECZhMJmayaB+Vy+W8d25vbyMQCGB7extbW1solUqoVCqsm3jWeT2sd3E/2tuyuFwu9tHZ3NxkhoNMD+lCdRR76lG/i+36T7lcDrVazeNVKBSIRqMc4GWzWX4f29sxPC+Ocoy0v1AalQiM9kIeOg/bzUoPe3992BiP1GenVCrtETZqNBr2qKEfytfqdDoIhUKUSiUUi0Xs7OwgGAxy2qjTgl0C0XedphGPGpQKCAaDaDabvAFRv6T2YAcAlxaSxml7e7tjArqD0Gw2kUqlEAgE8P777/PhR95ItFaDwSByudyhlDp2Ax61Xttvs0cx1uO4LVOgnUwmWX8lEolQrVb5cEkkEvxn5XI5ALB5YCQSQTqd7gqd3MNQLpfZoI3M5qg3FI1XIpEwM5JKpZgJIb1cNztgUwCTTCZRrVYhk8kgk8k41UY9z44Kx+UBRj2gKCglbyPSfFLTa2J0unW+Hob24KwbsxlHyuw88hf/9cDU6XTQ6/WsziZtSyaTYRHvs0R+Rx2ldxp/ZSW6Yow0l4TDFDwe5m3ySc0RjwvdNIdHAbrRNRqNIxvjo+b0oEOsPUA/qIHvs+Bln0fCUTE7ncZxrNNuwDd9nXYs2Gn7HXvornZK/ajTH3/9/V0xqc9yy37RxvgseFk3WEJvDr9Cb4zdjd67+CW6aYxPe4F7Ecf4tOhIGutJ0C2l1p1GNxgJ9tBDDz308OKgd2Y8OYSP/yM9HDXa00C9xfviopv7f/XQQw89fJPxyDRWDz300EMPPfTQw4uOHrPTQw899NBDDz281OgFOz300EMPPfTQw0uNXrDTQw899NBDDz281OgFOz300EMPPfTQw0uNXrDTQw899NBDDz281OgFOz300EMPPfTQw0uNXrDTQw899NBDDz281OgFOz300EMPPfTQw0uNR7aLeBn7Y+xHb4zdjV4/ni9xmGOkPnRCoZA7FR822tuf9NbpV3gRx/g0jTJfpPHtd67vRCNQegahUMid6Y/S6PdlXqeEh43x0JkdejF6eLlx1HPc6XX0Mq9h2mBFItGRbqydnsMenh80hy/jPAoEAohEIg42OvkMYnHH21S+9Dj0b7jXfuKbgaOe56P+fKFQ+Mhb1Mu8jonNqdfrR/Y7iNl5mb/HbwJe5jmkd6CTY2w2m6hWqx353d809MLJHr6ReFk38OOCUCiERCIBwCkANBoN/u8SiQSNRuNIAyrgKwauN59Hg/Z05MsIsVj8tbXbw8uJXrDTQw9PCUoD0UHQbDZf6gNhP4RCIWQyGfR6ParVKmq1GqrVKn8PAoEAYrH4G/WdvMx4mQOeFzVN156Co8vGUWjvXiZ0ZbDTLtqiRVir1V7aF+6bAIFAsIcJIPq4U3jW3y0QCCCXy6HX6yGTyVCpVJBOp1GpVL4xm41UKoXT6cSVK1ewuLiIZDKJVCqFVCqFZrMJkUh0bAfIs8zjfoH2y5yqeV687N8LpZBepHEKBALIZDLYbDaoVCrUajVEIhEUi8UeQ/UIdFWwQ5OoVCrR19cHl8sFsViMaDSKxcVF5HK5F25RSiQS+P1+WK1WqFQqBINBhEIhZDKZF2oszwqRSASFQgGdTge/3w8AyGazCAQCKBQKLxwrIhKJ4HA44Pf7YTab0Wg0cO/ePYTDYWQymU4/3rFAp9PB4XBgeHgYo6OjyOVy2N7exvXr15FMJpHL5VCpVI5l420Xlj5J0CIUCqFQKKBWq+H1elGtVpHP57G+vt7xALyHx+OwWaYXbb5FIhE0Gg0cDge+853vQCQSIRKJ4E9/+hNfuF60MQkEAigUCigUCsjlcmg0GiSTSRQKBRQKhUP7PV0X7KjValitVkxPT2NwcBASiQQrKysIhUIv3OEoEomgVCpx6tQpDAwMQK1WQyKRoFgsolgsfiOEaUKhEEqlEhaLBWfPnoVAIEA4HMbu7i4qlUrHvoN21uFp1pNQKITRaITP58PAwACEQiFSqRTK5fI3JthRq9WwWCzw+/0wGo3I5XIwGAzY2NhAtVpFNps9tsChnaWh0t1HMWwCgQBKpRJmsxnT09Oo1+tIp9P88014J19UtJeKP8050M4yvsgsHo3faDTC6/Xi0qVLaDabWFpawueff/7CpeIIAoEAKpUKVqsVJpMJTqcT6+vriEajKBaLhzZfXRXsAIDX68Xp06fxs5/9DDabDY1GAw8ePMDc3BxSqRRKpVKnH/GJoVQq4XK58G//9m/MAiiVSuRyOeRyOcTj8U4/4pGC8so6nQ4jIyP453/+Z0ilUty/fx+rq6vIZrMdS0+2H5JPczCLRCJYLBacOnUK09PT0Ol0qNVqEIvFCAaDR/vQRwyRSPREh4her8fg4CDeeustCIVC7O7uotVqQa/XIxwOo1QqHeucSiQSqNVqFItF1g89DM1mE3q9HkNDQ/jHf/xH6HQ6pFIpVCoVXLt2DdFo9Nieu1vwpPPeKdAhTiXaCoUCxWIRlUrlif6uWCxmm4V6vd7VY30UBAIBpFIppqam8Oqrr+K73/0uyuUylEolfvGLXwB48Zgq4Ks99dVXX8Xk5CTOnDmDmzdv4vbt2/jFL37Bl5jnxaEHO89THSEWi6FSqaDT6WA2m6FUKtFoNOB0OiEWi184TYRIJIJMJoPFYoFGo0E+n0c8Hkc+n39hbpDPQxuTvsXlcsHr9cLpdPKhqNPpIJVKUS6XOzavOp0OXq8XGxsbKBQKKJfLT5wGkUqlEIvFUCqVsFqtsFqtx/TUhwcK+NrM/x75Z8mb55133sGVK1cgl8uRz+eRzWaxs7ODra0tJBKJQ5nPJ113rVYLBoMBp0+f5t8fiURQq9Ue+ufp8yUSCcRiMcRiMWQyGUQi0XM/94sGhUIBm82GZDKJarXKQvNuAO0fSqUSWq0WZrMZer0esVgM29vbTxTsGAwGTE5OIpVK8VrNZDIvnKCX2I/z58/j29/+NmZmZiCVSjnbIZFIWCvXyYCH3iUSTVer1cc+j1gshs1mg8fjgd/vh9frRb1eh1AoxPXr17G2toZisfj8z/bcn7D/A/9qjvS0i4lYAFrctPk0m809nijdELnSbUGpVKJcLnPp4v5nk8lk0Gg0UCgUXJ1SKBSOTc/wPKBbhFarRaFQQK1We+gB8qjPEAqFMBgMsFgsUCqVqNfrUCqVMBqNPMdP+7mHBZVKheHhYYjFYsTjcUSjURQKhceuMZFIxD90aJL4+kWAUCjkS4XJZGKvj1gsxuuzHXToyOVy6HQ6jI+Pw+fzQSQSIZfL8eGTSCQOZVN6GrRaLSiVSvj9fshkMg7AMpnMQ/cf8hlqN3STSCQvbBrgWaFQKGA2mzE+Po7V1VWkUimk0+muCQK0Wi1sNhtcLhf6+vpgNpuh1WqxsrKCQqHwWGZcJpPBaDRibGyM10Q4HEYwGHziy81RQSQScbBN6y6fzz/0eSQSCVQqFfr7++Hz+dDf3w+hUIhKpYJKpdIVa1cmk8FgMMDv96NWqyGbzWJ3d5cLFw4CvYN6vR5msxlms5nT5E6nEwMDA9jZ2em+YEckEkGlUgEAyuUyKpXKUy0mOlz1ej1HqNVqFdFotGMH4kEQiUQwGo0YGRnB+vo6crkc8vn81zxFDAYDPB4PxGIx6vU6SqUSSqUSqtXqkfuPPC+IWpyZmeGKm2Qy+dQbYbPZhM1mg8Ph4IWtVqvh8/lw584d5PN5lMvlIxrFo2EymXD58mVMT09jdXUVf/nLX7C4uPjYtdZoNPiALJfLKJVKvOF0QzD+OMjlcoyMjODChQt49913UalUEIvF8Mtf/hJLS0sIh8N7/jyJsp1OJ6ampnDy5Em4XC60Wi1sbGxgdnYWn3/+OSKRyKGlmZ/me9RqtZiamsLw8DCWl5dRqVQwNzd3IHsqEAhQr9dZzEl+QXK5vCtuxg/D/hLpx2mTngQOhwNTU1P4+7//e/zhD3/AwsIC7t271xVCbaFQiImJCVy6dAlvvfUWzp07B5FIhGq1itu3byOfz2NlZeWRn2GxWDA0NIRvfetbsNvtyGQymJ+fx+9+9ztsbW1hZ2enI5dO0jGaTCbo9Xq+CNN3fxBUKhXMZjMcDgdcLhfMZjNarRZrzWq12qGle54VJpMJFy9exM9//nOk02nMzc3hww8/xKeffvrQfYHeP6vVCofDAZvNxuy5xWLB6dOn8eDBAySTyed+vkMLdugQu3TpEiQSCXZ2djA3N/fE+XuhUAiNRgODwcALoFwuIxqN4tNPP0U8Hu+agIdEYufOncO7776LcDiMO3fu4MaNG3sO7lQqhfX1dZTLZQgEAtRqNb5BdnozeRQEAgEmJiYwMjKCS5cuQafTYXl5GdlsYOF2SwAAIABJREFU9qnTbwKBANFoFDs7O2g2m2w0V6/XO+pv0f4sU1NTsFgsEAgE2NraQjabPfAgaT9w6O/TvD5tYN8JEAt15swZnD59GpcuXcLExARXCCaTya8FnkKhEHK5HFqtFh6PBz/4wQ9gs9kAfHkTvXXrFm7evInV1dUnoqwPG+10uc/nY5ZudXX1QD0YXaByuRyCwSAkEglfRDp9WBwEgUCA4eFh1jLu7u5ia2sLc3NziEQiTx2YENsqk8nwox/9COfOncPMzAxEIhGcTieKxSIHjJ0AMcoajQYjIyMYGRmBx+OBVCpFsVhEJBLBf/zHf2Bubu7Av0+Hp9FoxHvvvYeZmRmcPXsWWq0WpVIJNpsNc3NzyOVy2NnZ6che7Ha74ff7cfbsWQwODiKfzyMUCmFhYeGhwQ6x4UNDQ9BoNACAQqGA27dv4+7duyiVSh1l5EQiEd5++21cvHgRfX19AL68hGg0mkfu8SKRCFKpFCqVilmu9jmRyWSH1srjUIMdhUIBr9fLt6RgMIh6vf5EB6RQKIRWq4VOp4NWq4VIJEKhUEAqlcLy8jLnJrsFMpkMVqsVo6OjMBgMSKVSWFhY2GP/XavVUCwWkcvlIJVKUa/X2QiqW0GbzcDAAAYGBqDX6yGVSp/5mQUCAbMfwJesSKVS4QqmTjJcNFdUCeDxeNgo71G0KXlA0XdSLpc7xk49DegQ8fl8GBwchNfrhVqtRrlcRjgcPrAaiVJYZrMZLpcLPp8PCoUCtVqND95wOIxsNtux1GytVkMqlcLQ0BD0ej3sdjuv2YOeqVKpoFAocNWYUCh8rjV+VKBqTr/fj/HxcZw+fRrLy8uoVqsIBALPdFEgew+Hw4GxsTGMjIzAaDTCarVymqjdMPM4QSlvnU4Hi8WCkZER9Pf3Q6/XQygUIp/P8yU6kUg8dHz0nU1MTGB4eBgmk4nPJI1GA61WC4VCwTKJ4x6jzWaDz+fDxMQEvF4vIpEIMpnMQ+eTmCCDwQCn0wm5XI5ms4lisYjV1VWsra11NNghosPtdsPlckGtVkMsFvM6etw6bZcCELvabDZRq9VQKBQObV851GCHonEa7MLCwmOrI9r/vs1mg91uh8VigUQiQalUwu7uLu7cuYNcLndYj3ookEgkMBgM8Hq9UKlUiMfjuHXrFhqNBt8oSf2/u7sLrVbLVTvdHPCIxWJoNBqcOHECbrcbjUYDsVgM6XT6mRYdpQ2IyanX68hkMqwR6GSQUK/XOYjWaDQYGhqC2+1+aLBDmz+J8ORyOQQCAesBuh2U85+YmOCDoFqtYmtrCzdv3jxQ2EvVdENDQxgfH0dfXx8EAgFisRgWFhYQCASws7PzSL3BUaNYLCIYDOLkyZMsiFepVA/dKEulEtLpNAqFAh+OBoOh63RXMpkMTqcTFy5cwMTEBEZHRxGLxfigBp6+EIQY9FOnTmF8fBxut5vF9mKxmNd0JwIdiUSCsbEx9Pf3w+v14rXXXkNfXx/0ej0ajQai0Sjm5uawsrJy4L5BWkqj0Yg33ngDMzMzGBgYgEajYSY2n89DpVJBrVbzBfS4xkrP5/P5MD4+jsnJSa4GfBibDHz5DhoMBrhcLgwNDUGlUqFeryOVSnHVUicvG7ROnU4nLBYLZDLZHub8UaA/I5VK+YfOiUKh8MRC9CfBoQU7RI263W5YrVbIZDJ88cUXyGQySKfTj/37EokEbrcbDocDRqMRAoEAkUgEwWAQsVisa1JYAPbc6hUKBYxGI487l8uxYSDlWUulEnQ6HRQKxWMbUHYSYrEYJpMJFy5cwOTkJKRSKZaWlhCLxR75Mj4M7UJupVIJ4Es2pVKpcFqyk99DvV5HLpfj0mWFQoHTp0+jVqshHA5/7dlIc6TVaiGXy/m2GYvFHnrT7BaQsJHSVw6HA7VaDZ9++imuX7+O5eXlr238EokEWq0WExMTuHDhAk6dOsVMXSgUwvvvv4+5uTlEo9GOu2G3Wi2eE5PJBKPR+FA9GLHNVFwglUrh8XigUqm6JsWsUqlgs9lw5swZzMzMoK+vDyKRiJniZznc6LsZGRnB3/3d38HhcHCAR5qPdsHscUGhUMBkMmFoaAhvvvkmfD4f/H4/fD4fpFIpKpUKtre38cknn+C3v/3tgfsGMdJmsxmDg4N488034fV6odPpIBAIUK1WkUqlsLa2hlAohEQiceyu51qtFj6fD1NTUxgaGoLVakUsFsPa2hpu3759YGBAGY++vj54vV5otVo0m01kMhk8ePAAkUgEuVyuY4GOXC6HxWLB9PQ0RkZG4HQ62XsskUg89twQCoUcZLdXQ1KT1sNMjR8qs0MRs1qt5nSUQqF47AZCJa1KpRIqlQpKpRLNZhPZbJa9dbophUWHuEKhYPqNbkcUBBFdrNVqoVKpmHrs1j4sAoEAWq0WTqcT09PTsFgsfJCn0+lnNnciipOE68BXOhCiLDsFWnftAtVHrVn681KpFBKJhNMkxFx1I8jOYWBgAENDQxgbG4PVaoVIJEIikcDdu3cRDAa/5k5OBp8WiwVerxculwtGoxHNZhOhUAiBQACBQADpdLrjKTyyeCBWYv+7uB+tVgu1Wg2ZTIY1hSaTCRqNBnK5vONeXsROuFwujIyMwG63M1OVTCaf6X2kVIHBYIDdbkd/fz9fvgAwA0+Hy3EEfXR5MJvN6Ovrw/j4OLOHLpeLDVjT6TTu3buHxcVFbG9vf+0soLHRWh0aGoLD4YBKpeJgLp1OIxKJsFndcTMhYrEYWq2W08dWqxVisRi7u7sIh8OIRCIP1QmqVCoYjUYYjUb+ThKJBBYXF4/VwHM/hEIhpxz7+/thMBigUCjQbDY5IH/c99zeDmp/BXe7RvIw1uOhBTtisZgHq1AooFKpYDAYnui2RAcI2bhTiXImk2HDr24KdijHr9Pp+NCjPGP7OBUKBZddq9VqlEqlr1VVdAtEIhHsdjtGR0dx5coV6PV6ZDIZbG9vszfQ04KCQr1eD4PBwJuoTCaD2WyGTCZjYehxo511orJqWoftJmTt8ykWiznQobLR9uaX3QayYXe5XDh58iQmJiYwPT0No9GIZDKJtbU1fPLJJ1hfX/9a2k4oFMJsNsPr9eLEiROcTqhUKnjw4AFu3bqF5eXlA6sQjxt0kCgUCrRaLUgkkkeyp6TVIq0EVQuSZqWTwU47Qz42NoaZmRnY7XY0m01sb29ja2sLkUgE+Xz+qa095HI5BzrEvhPTXC6XUSwWjzUdSZcej8eD0dFRnDt3DidPnuRzJJ/PIxKJYHV1Ff/v//0/zM7OsoHl/s9RKBTw+/04c+YMTpw4AbPZzExfq9XCzs4OVldXcf/+fQSDwWeqLH0eyOVyWK1WnDp1CmNjYzCZTKjX61hbW8P6+jrC4fCBzyMUCqHX62GxWGA2myGRSJDNZrG1tYUbN26wZ9BxgwJVi8UCt9sNn88Hg8EAmUyGRqOBTCaDRCLxxN8zsbC0l7RfREn28bzjPNRgh4zi5HI5+6g8jhZtv0HShiqRSJDL5bhMuxto5XZQKiaTyXAZazqd3nNTpDLJXC7HzIZAIODbI/kRPSsO83CluXvrrbdw9uxZuN1urKysYHFxEbOzs8jn849daPsD2nbfJFoHVO0ilUpht9v5e+hEipLYOBKmki/E/Pw8otEoBzTkoSSTyaBQKPhQpaBHoVDw+LoJAoEAXq8XHo8Hp0+fxhtvvIH+/n709fWhWCxibm4Of/jDH/h2SLo6CgJtNhvGx8dx6tQpvP7669Dr9SgWi1hbW8P//M//4O7du4807zsu7KfBC4UCdnd32U35IDSbTZTLZdy9exfDw8NQq9Uwm81wOp2w2Wwdc1Em3YpWq8XFixdx/vx5nD17FjKZDNvb27hz5w4WFhaeumSaLhher5fTYjabDTKZDAC4ZQbpIo86zU4ssk6ng8fjwXe/+12MjIxgcnISNpsNAoEApVIJV69exd27d3Ht2jXWpeyv7hUIBLDb7XC5XJiensZ3vvMdjI6OQq1WQygUolqtIpPJ4ObNm7h58yauX7+OSCRybKautA9OTExwCtlisaBUKmFjYwO3bt1CIBB4aJApFotht9vh8XjgdrtRr9extbXFNgGd6hdJommn04mRkRG88sorMBgMaDabiMViCAaD2NzcRCQSeehabbeX2dnZQTwe59YzCoUCVqsVZ8+exc2bNxGPx5/b+frQdmiihomKarVakMlkbBz3qJeT0giUFiKzpEKhcOwmZU8CijilUimAr26KpVKJN1j6PsrlMhvQ0a2NDsrnwWEtcHoZlUolBgYG0NfXh0ajgbW1NaytrbEw+aAcOX0HFH23G3URS0LCOrvdvqdku1qtcvTeCZ0EiYxJf0NeF0qlEnq9HlardY/pWL1eh0qlgkajgUqlglQq5XlVq9VcYtnpwJz8f8hLZ3R0FGfPnoXX64Ver4dIJMLGxgZWV1exvr7OAbhOp0OhUIBcLodKpcLIyAhmZmYwOTnJm1gqlcLs7CwCgQAfGJ1mXCmFRRWc1WqV0zyPCsRarRanAxKJBGw2G4xGI0wm02P3q6MCjYXacfh8vj3pq42NjWdyXycG0+l0shZGJpNBIBCg0WigVCqhUCiww/BRzikFp3SYXbx4ESdPnuQ0CAB25L59+zZmZ2extLTElYL70xxyuRw2mw1er5c/h9Z5vV5HsVhEOBzG5ubmnorD41q3xDqNjY1hYmICfX19aLVaiMfjmJ+fx9LSEqLR6EPXG3m6GQwGaDQa1Go1dglPp9MdY1Xp3LBarXC5XFxUVCwWkclkmKx4VIqt3TOqUqmwDx2xsyqVigXPh1HQc2jBTqPRYKfder2OWq3GwlRSvT9sgdENmYIdgUCASqXCZn3dBmKvVCoV02s0WSSea1ej06SSnkAmkz1TsHMUh2l7Ss7tdsNutyOfz2Nubg5LS0sHit8oSFGpVOylIJfLEYlEkEwmUavVIJPJoFarYbfb4fV6uWkmfVf5fJ5fmE4ECeTtYDAYoFQqOXhzOByoVquQSqWIx+M81nQ6DY1GA4vFAr1ez67Y5P5pMBi6Itgh3ZvRaMSpU6dw8uRJXLhwAWazmd+rlZUVrK2tIRKJQKfTsU4uGo3CYDDwjerSpUsYGhqCQqFAIpFAOBzG1atXEQwGD60txPNCIpGwB4lEIuHD4FFBAb2PxGgkEgkIBAIYjUaYzWYIhcKO9E8i+w673Y6hoSF4PB5muaPRKILBIOsXKT3TvrcQI7M/ICDLgb6+PoyMjHCDZfoOqKiCRKVHFeiRxw/5NvX39+ONN97AyZMnodPpIBaLWSf44MEDfPHFF1hZWcHGxsbXnon2Do1Gw9qms2fPslYHALO1a2trCAaDXDV4nHMrkUig1+sxPT2NU6dOweFwIB6PY2dnB3fu3MH8/DzS6fSBz0OBocVigdFohFqtZqfzZ0llHhZIhkHeTG63m+0BqtUqEokEO7Hv13/tl3C0+5bRGQp8RYBotVq+WD4vDi3YqdVq3Ja92WxCp9PB4XAgEonAYrEgHA4z80Ogxa/VamG322G1WjnvTu0JusHNcz9IBGexWLi/UygUQjab3VMm5/V6cfnyZabqKpUKT6BOp0MsFnuqsR3F9yCTyWCz2fDee+9hcHAQzWYTf/zjH3Hjxg1sbm6ycR5trhSYajQavPHGGzCbzTAajThx4gSSySSi0SiuXbvG/WzIu4YC4HA4jGg0yptsp/yTKG3hcDigVCqhUCig0+nwr//6ryzUbG8dQUxAo9FAX18fDAYD+3To9XoYjcaOr1OhUIjLly9jbGwMk5OTmJmZgcVi4WelNehwOPDaa6/hzJkzXHItEAiwubkJi8UCi8XCHktyuRzFYhG3bt3CtWvX8H//93+Ix+Nd0+7E6XQyxS8UCjmN1V6xs1/k2C7adjgc0Ol0vHd1iq0ilsLpdOLHP/4xszoCgQAbGxscoJJglcT07QE3MTOhUIjfWUqD9PX1wePxsH6QAt90Oo3l5WV8+OGHfLk5qvHTRU+v1+P73/8+Lly4gAsXLnClZrVaxebmJmZnZ/HRRx/h3r17D7W8oH1oYGAA09PTuHDhAvx+P6eTyd17YWEBf/zjH3H16lXEYrFjPU8oGPjJT36Ct99+Gw6HA61WC6urq1hYWMDS0tIjNW8SiQQajQb9/f0818ScPKodynGAgh2v18tmnq1WC/l8HktLS2yzolAoWArQbDaZvWs2m0in01Cr1TCZTBzQabVaPi9jsRg++OCDQ2sufOjMDpUvKpVKnDhxgkWc9+7d42ZsdICSt87U1BROnTrF5Z9Ea8Xj8a4s6VUoFHxAUgqrWCzyBkMpK6VSCZ1OxzcJsVgMh8MBj8eDTCaDUCh0rH2y9jMPQqEQDocDIyMjOHfuHEwmE5LJJBYXFxGPx1EqlZiqVCqVzNRQS4/XXnsNOp2O/32pVEIul4Pb7WbNi1qthlarZW+d69evY2lpCcvLyxw8dALU7TyZTKJUKrHPiEql4nnVaDT8olLHegrkRSIR8vk8/vKXv+DmzZtYWlrqaLAjkUhgMplw5swZTE9Pw+/3w263sydLuyDb5/OhUqmgXq/vEXFS1Q/1zSIPoeXlZczOzmJhYaGj1PlBoHVJPbEo4KbKRxLIDw8PY2hoiLU9crmcfZXUajWCwSCWlpawvr5+5Df/9tss/bNcLuey5OnpaXaRBwC73Y6pqSnUajUOsKn5I7E5Op0OtVoNpVIJq6urnHqu1WpwuVxwOp2YmZmByWQCAORyORaZX7t2DSsrK0gmk4cSDLSnJwgSiQR+vx8ejwfnz5/HhQsXMDAwwJdb8lO7ceMGAoEA5ufnORCgz2uv2qW2Ca+++irOnDmDwcFBPnBJp/PRRx9hdnYWs7OzSKfTx+ZyThpIpVKJoaEhXLlyBSaTidNXd+/eZcuGh33fQqFwD8NHXkHEvj2JnctRQalUwmKx4PLlyxgeHmbWuNlsQi6Xw+12QyAQwO/3Y3p6mpnSZrPJbYMajQaSySRfikdHR+H3+9l2Jh6PIxAI4Isvvngo8/W0OLRgp9ls7mkiqNFoMDExwZqIZrOJSCSCWCzGfYQkEglGR0dx+vRpTE9P8+2sVCoxnfm4Zm+dAOkipFIpT6JQKIRarWb7erIs1+v1PFESiQQulwt+vx/FYhELCwtIJBLHUv2xv3yP6O2+vj6MjY3xBhuLxbC7u8tGTyqVCn6/H2azGTabjR2jDQYDzp07x6lHClBrtRpGRkb2lLQmk0mEQiF8/PHHeP/997G5uYl4PN5xb5ZCoYDNzU0MDg6iUqlwKW77Zk1sCAXoFMTRzePTTz/F1atXsb293bGxAF+trVOnTuHMmTOskWoH6QfcbjeAr/or0Ty0j528PGKxGO7cuYO7d+9ieXn5UG5YhwkqOScTUkp9kw5LIpHA6/Xi9ddfx5tvvgm9Xg/gq2ozoVCIdDqNTz75hIOd4yi5pgOAnkWtVrNOamxsbE8Q6nK5uMGiWq1mrRix39Vqlf99s9nE4uIiv4v5fB4ulwtWqxVOp5MlBYlEAn/+859x/fp1fPLJJ+w5dJhjbF9XSqUSY2NjOH/+PH784x/DZDKxgVw+n8fm5ibef/99fPzxx+zrRewAVUgSM2Q0GrlM/Z133mGdTrsGKRQKMTu0u7v7REUWhzVuKsCwWq04ceIEzp07x1VU4XAYDx48YFPVVqu1JyVJP1SlduLECTaurdfrfCZ20sRUpVLB6XTi7bffht/vZ7aGNI+Dg4Po7+9nNp+6oLdaLQ6Ems0mkskkp6uouInWMAU7t27dOjQ27tCCnUqlgp2dHVy9ehUikQhXrlzhPlejo6N45513uENre2NPKqujKqxSqcStF8gToduQTCZZOU7usv/wD/8An8/Hgmq3243h4WH4/f49/UG8Xi8UCgUGBwehUqnwv//7v1hbWzuWXjTth59YLIbb7cb58+dx/vx5mM1mSKVSuN1u/PznP8fy8jIEAgFsNhvMZjN7KJE4l2jkRqPB1Plnn32GQCCAK1eucJpPLBZzo8kbN24gGAweWqT+PAiHw8hkMlhaWsKvfvUr6HQ6tgmgaqtMJoNIJILd3V2IRCJMTU1hfHwcFy5cwNLSEubn5/Hpp58iFAqhUCh0bCxk+3D58mX09fVx0E2BOPkA0YbTrikj0TwVEgBfphOWl5fx/vvv4/bt27hz5w5rBDo9b/tRLBZRLBbRbDah1WoxPj6OH/7whwiHw1w6f+rUKdYmtVotJJNJ7oAtFAqRTCa5HPk4CiIoWCGGTC6X48yZM3j77bdx5swZDnSAr6pVyaC0PYhoD1TbD8qLFy/ynFMQT1YJdGD+93//N37zm98gEAgc6WVLIPjSofq1117D5cuXMTU1tSd12mg08Ktf/Qr37t3DnTt3UCwW2Q9Ip9Nxi5LJyUmYzWbo9XpmizUaDVf90iFZKBQ4lb6ysoJQKHQkATpdDPYHUFRi/vrrr+N73/ses4nxeBzBYBCffvopGo0GN4mORCLM+pOrs1KpxMWLFzE2Ngafz4eBgQG0Wl82/Xzw4AESicSxVJMdxNDRfqFWq2Gz2fa4edNcu93uPcEbkQHtBSmkkWtfv7Tm6/U6e/QcZrXnoVZj1et1XL9+nYW6J06cYPGqRqOBWq3m9EB7xRaVJ9PnkMjwcRUVnQKlrZLJJJxOJ9RqNTweD2QyGT+vWq1mzUN700tSmSsUClSrVU6RHAfac7xkG0+bBpmwKZVKDA8Pw2q1AsCexQyAxZ/ZbBblcpn1LYlEAtevX0cgEGDr/tHRUXzrW99iYfLGxgYfTJ0G3WSp6o98nrRaLQvIqaVAqVSCUqmE2WyG2WzmQIFuzp3WlREDFQ6HsbS0xDqofD7PjrvkG0Q3X0oHUKBL/hiFQgHpdBqLi4tYW1vD9vY2l3J3i06nHdlsFolEAru7u/D5fHA4HFAoFCgWi6yPk8lkXHl1584dbG9vI51Ow263w2w2o1qtIh6Po1wuH8vaPMgUj1JWtC+2B6XtwQ1pH6iikXSNAsFXbVnEYjG7BCuVSlitVhiNRjgcDjQaDaRSKXz22WfMsh822t8FYi60Wi1sNhusVitX3NJ/HxwcZDaDClJI2EspcxoD/TuyfGiv1KF+SqlUCl988QVisdiRWZc8rDzfZDLB4/FgcnISfX19MBqNvK+azWaMjIxAqVRykE6GnMS6yuVyKJVKeL1erhAk3zkKpGw2G2KxGAKBwKGPa/8Y20EWBk6nE4ODg3C73WzASmM4yMxz/3qgdUs9E2u1GrRaLZRKJcsDjsKV/lDNQVqtFhYXF1EqlbjKx2q1Qq/Xs6iOJp5eeBLYtX8G9TA5Tj3L06BSqSCbzSIajWJwcBAymQw6nY5TVpTSAcC29MSGkGaA0nW1Wu1YNtiDFi49E+W6m80mxGIxrFYrLBYLMwPkH1QoFLjx4/b2NpLJJLN1xWIR8/Pz2NzcxPz8PBKJBDKZDM6ePQvgy40oGo12jW9SvV7fU0VHNwvyzBGLxchms3s261wut0ebJRAIjm3+HgfyqlhcXOQ0QDKZRC6XQyKRgFwu5wtJLpfjvPv4+DindiqVCpLJJMLhMJaXl7G9vY1YLIZcLtfxgO5hyOfzSCQS2NragsPhgFqtZst6WtP5fB7pdBq7u7v4v//7P2xubiKfz+Ps2bPweDwQiUTI5XLH1r6E3ivgq3QP9QJq95Khw5v+DgXYlB4gH7JKpcL7SblchkQiwfz8PLa3t2G32+H3+1Gr1WC329FoNJDL5XD//v0jLTPfzwaoVCro9XrWIrUzVKTV1Gg03CyZCgZoryTjSK1WywwIfU77d1MqlZBIJDA7O8s+aEc9PoJAIOBGtB6PB0ajkVPj5D1HuhR6rnZ/LqrylcvlrEOjgI4CDWqFQp3PjxN0Wbdarejr62OvpvaKwP19H2lt0/yQeWWhUOAS9VqtBo/Hw98Bae9SqdShPv+hO6FFo1GOzFZXVzkKNxgMPJEk/Gw2m1zuTJbTwFcBUDeU8h6EnZ0d9j8QCoXsIyOXy/nGFQwGmfnIZDLcEdbtdrMWIpPJdOzwp5zp7u4uotEo4vH4nlsSBWnpdBr3799HOBzG1tYWlz1ub28jl8uhXC5zUEqBnkAgwNraGmQyGTY2Njjd1W1O2PS87ZTwQZQ+BUHUo4ZulyqV6tA8IJ4H1AF5bm4OhUIBYrGYHVnJa6Q9jSWXy7kyRKfTQS6Xo1arYXV1FcvLy1hZWcGf//xnrKysIBaLHeiz1C3IZrNYWlrCr3/9a2xtbXEFk06nY5Y4m81ifX0dgUAAN2/eBPBln6IrV65wCTbpR44DB910FxcXYTAYkEql9niVRCIRAOB5JIaNGDoyIhSLxZxebzQa2NnZQS6Xw9jYGB/6k5OTKJVKyGQyfPE4atABSIUM1NKDQEUqJpMJg4ODfPGgNDkdnvSzv3SZDCKLxSJSqRRWV1dx584drKysHIs0oB0U1JDvVnuzVhItq9VqDmDbUzrtP/R39mvo6DJK73UnIJFIoNPpYDQav/bOtK/r9gCUKnB3dnawsbGBWCyGWCyGaDQKqVTK1Xm0TlKpFHZ3d7ub2QG+HHC5XEY0GkWlUmFBJ0WpcrkcUqmUN9D+/n5MTk5idHQUDoeDJ3X/zaebQK0sAoEAu6+SGRk9L1UTUMBz8uRJnDhxAlarlW/d9B11AqSGX1lZgUAgQKFQYEqyVCrxxkobCKUKyCyKUoztJpL7P58CJmKzSDPQbfP5ONDhSc34SBBLuotOj6edDV1fX4dQKGTtSXvqoz1XPjIygosXL0IqlSKRSCCVSuHzzz9nke7DjNy6DdRDb2FhgYXJlComzQo1JCSnaK1Wy3oek8nU8d57jUYD6XQaS0tLLGKt1WooFovcBJlMS9tT3+3pcSolp8tTuVxm1oMqXuiCQwHRcaxbeoZUKoVYLMbmc+36sPY0CO2yfCwoAAAgAElEQVRBxK7Sgd8e6BDrVa/XEY1GmYHc3d3F0tISVlZWOhKgt1otbrFz584dNBoN6PV6TtOQrqhdv9J+YWy1WqxHslgsPF4KGLa3t3Hjxg3Mzc1hc3PzWMdGaE+hZjIZlp9QOo7mJZPJcAqdPI42NzexurrKliPlchkKhQIWiwWJRIJTsAA4Y3CYOBKPe8odFwqFPeXYEomEb8K0GGOxGKeB2l/ibg52Go0GisUiMzxErwLgWz71DaJNhxYwaUQo4Okks1MoFLC9vb2nDLDVanG6hm6B4XCYjcee5Hnb89ntNgNyubzrD8+DQD1g7HY7bDYbp/3ahb+dBlkgPOq2Tq7RbrcbQ0NDGB8fR6vVQiQSQSAQwL1797CyssI6nUcZgXYL2i9XhUKB9xiqOiqXy0ilUhzwUUrS4XDA6XTusY/o1FipMpDYGGJzyA2aKqtIZ0YHzn6GiECHqVKphEwmg0aj4RRRIpFg6/3jAnmqRKNR6PV6btPSarVYkEzPTP+/3a5jf8shYnMKhQICgQAWFhYQCoUQjUaxtraGcDh85PKHg8S7wJdp1d3dXczNzbHuqFKpcC8y2gvpXKNKsXw+D6FQCJvNBqfTCaPRyMFRPp9HOBxGIBDA/fv3sbGxwUaYx733kPlfoVDgbABdOCiwrdVqCIVCbNhJBqahUIgzHsRMUVsP0nJS+vkomisfeUMfunVSNLi/aiWfz6Ovrw8+n4+pWop2jyM6f9YFUygUsLa2ho2NDX5Rgb03aYJUKsXY2BhH8LlcDslkEpFIpGMdo4k+J8+bSqXC7S+oTQJ9L8/y/ZDDNJXoq1QquN1urK6uHror9lG+9BSkXbx4EaOjo7BarVxOv7u7i0Kh0FW+Mw8DBdxutxs/+clPMDw8jP7+fkSjUXz22Wdcep1IJDrab+dhws9HgYKVZDL5yD9HOrWJiQm88sorGBwcZOuHVCrVsWIISrdFo1FEo9Fn+g7aQR48KpWKfVrMZjOSySTm5+cxPz9/bPokGtvOzg4+//xzBINBTE5OQqFQcGUYVVORXpBSi6TxMZlMnJ6klO3m5ibW19fxX//1X7h16xY2NjZYmN+pywddFGu1GmKxGJaXl9kkz2q1MmtDF8lsNstygEKhAJPJhMnJSUxPT2NsbIy1n4uLi7h58ybu3buH3/3udx1rvts+lwsLCzAajZwCp5QU8OX6m52dZTZnZ2fna4bC7difqmwX6B8muqZ7IQ2M6Nvd3d09vYmO+vc+C+jF2p/rP+gzqedSe0uMTrMCjUZjD/1IAdph3IroZa5Wq5DL5TCbzRgaGkI0Gj30QPaovkMq+7XZbBgYGOCqJepNEw6Hu0Zw/TjY7Xa43W5MTU1xGa9QKMR//ud/4v79+wgEAkgkEsdmvHYQjpptEIlE3KHZ5/NBLpcjHA6zsP44u53vD9APe3On5so2m41TKMFgEHfu3MH9+/ePLc1DB/bCwgK2tragUCjw4Ycf8qFGnmvE/FPJuc1mY7NAYs3pgnb9+nXcunUL9+/fx2effcY6p+MuZjloDinVmM/n2ay01WrtKXxoNyqlM44YR5VKxR5ZpO386KOP8Pnnn3O2oFNFOzRfS0tLCIVCuH//Pgeg5XJ5T4UddROg7+Nh7zZlfSgYIlLkKILWjgc7lObaXz5YKpW6shJrPx63SdFioB+6wVAl1HGJIulZ9r+c7d/xYX3f7WMmKppKuzst5n0atKc8qNJHIBCwPoD6CHV7sEN+ST6fDydOnIDNZoNIJEIqlcKDBw+wvr7OjtndPpZnBR2olL6yWq2o1+uIx+Pc6uVF2G+eFEqlkgOHdlNXaoh5nPNMafJ0Os1BDrB376QUFmkgPR4PfD4f+vr69njalMtl3L59Gzdv3sTCwgK2t7c7yubsR7uVyJOw9tT8VaPRwG63w+FwsO0HGevNzc0hEAggFAp1nEUmfVkmk8HOzs5zszB0JlGqtj3dfBj9sNrR8WAH+KrPSfvBf1xBwFGlQOhzKf+YSCQQCoVQrVahUCi46WCxWDyWqggAzEIdR2qQRJGUmy+VSgiHw1wZdFSgPPdhjJGsz0+fPo3x8XEYDAZUKhXcu3cP8/PzCAaDXbPJPgpCoRATExO4dOkS3n77bRiNRqyvr+PmzZvY2tpCKpXqKKNDeFhlx2GAypmnpqYwOjoKp9OJzc1NrK2tcQ+44wTd5v8/e1/y3GZ2XX8wz/NAAAQIcBZFUSI1Sy23uq1O7PJcsZ2qOOVNklX8J2SVdSq7bLJIXJWl7Szc+bk8tN2D263WLHGewAEkQYIYiXkGfgv1vfpIS2pKIgbSOFUqdUsi+T289713373nntOobJbFYuFsJPn3LS4uYn19HeFwuCE/82UQ0hmeB6H4JXl9EWGZRFmpfPzTn/4UoVCI7YlaAdrbX7f0ShCLxTAYDPD5fPjRj36E06dPw+v1olqt4tGjR/j973+Phw8fIhQKNe2cOIhGZiEpuRGNRtm8l6QKtFrtkfwMQsuv2eRaTE61QjfxZgQ8jdrghd+XSlaFQgGpVApms5lT6WSC1yw0K3VN2TrqiMlkMkxOaySOItNC3SAOh4PtTEhagJTCyeG9nSESidjD5ubNmzh37hxMJhPW1tZw//59fPjhh4hEIm0jGNioWj0dmvRZ9PX1QS6XY25urqUq7Y0KdCQSCa/dkZER6HQ65PN5LC0ttX0Gq1KpMIeKuj+Ff1csFvnG3w7jeNOLFbXej4+PY3x8HC6XC1KplMm9JPXRzBLrQTTyzKCMK0nUKJVKWCwWWCwWGAyGI/1ZLQ92iPQkvFkKNRVOAqjbIh6PI5VKsXox3bpOIoSbViqVYs+idtigvgyUUtfpdLDb7ejp6YFareb27mQyyTX5dgZtHKdPn2aRs2q1io2NDayvryMQCOzrjDipIHVeg8GAnp4eaDQalEolhMNhfidPEkgXioT8iDgbjUbbIoP3ZSDDWq1Wu09Zn7pDn6fSe1xBPCW32w2TycTdgSTKSgrux2HffB0Q3YE4TcTdor87SrS8jEWGiltbW6wnQYqgJwW1Wo2l/EOhEGsNORwOfplPEkQiEeuFzM3Nsf9QqwnZhwV1YOn1evbmkUql7FRPAVwzfJTeBCTUdv36dfh8PshkMkQiETx69Aizs7PY3t4+NgTrN4VWq4XT6YTb7WbtnWg0img02lIH6UZASBSVy+WIRCLY2dlBOBxuma7XYSEWi9lPcWJiAm63m8so5GFGhpHHHSKRCDqdjvl0VPIvFApYXl7mUl27yz+8CYQOA1QSJEuTtldQfhNQa3oikcDCwsKRtyi3EqSfsbm5CZPJBIVCwSnldtQSelNUKhVkMhn4/X5ut2+1eNthQOqter0e3/72tzExMQGDwYBcLoepqSncvn0bjx49QiwWa+uxiMVinD17FteuXcPly5fZ0X5ubg7379+H3+9HKBRq6zEcBWg+b968ib//+7+HUqnE3NwcG5xGIpETldkSiURsGkqaUJFIBKurq1hbW2v7AN1kMsHpdMLn87GbebVaxdraGu7cuYM//elP2N7ePvZBgFQqZUmLt99+G9evX0e9XsfKygqWl5fxi1/8AjMzM9jY2DhxZ4MQ9fozgchCocDn5Pb29pFzy1oe7NAthHr14/E4CxK1owno64I0hDY2NqDX66FUKhGPx0/UGAkUvNGclstlZLPZpml7vCmoQ6Krq4tLHpubm1hZWYHf70ckEmmZPtJhQHXw3t5e+Hw+aLVazmasra1xdqpVhMdmgT4H4ut0d3cjk8kgGAzyzTmXyx2LNXlYkJggtTiHw2Fsb29jZ2fnWJRDiJhKBtKkDL2yssJWJsdF2+plIJdzj8cDh8MBjUaDvb09LC0t4d69e2zV0u7B6ZuCGnhonVLXYDwe/zNNvjdFy4Md0iHIZrOYmZlBOBxGIBBoqIlbK1Cr1ZDP5zE1NYV0Og2VSoVgMNj2vI83QalUQjAYxPb29j4p+3YHBTskfBaLxXD//n1MTk5icXER8Xi8rQ8NEs8bGRlBf38/6vWnSuXr6+uYnp7mVutGzoWwlNIqkIGkyWSCWq2GSCTC9vY2FhcXMTs7y7YMr4sXqei2CiTjQUJ2iUQC09PTTMRu98sGZaW0Wi1359JFibR15ufnkc/nj/3ZQJ5QXq+XLSXW1tZw+/Zt/PrXv0YwGDyy7FW7rdODIG84hULBCY+D5PSjQMuDnVKpBL/fj2Qyib29Pa4tt7KtsBEoFouoVCp49OgRlpaWmHGfyWTadhG+LkhkiuqxlKo8DuOkmnkkEsFvfvMbvnV99NFH8Pv9TZGiPyqsrq5CIpEgGAxiY2MDS0tLmJ2dRSaTaXhGsR1KDJQeT6fTWFhYwO9//3usrKzg3r17mJ+ff+PDv93WM/EdIpEIJicnEQ6HIZVKsbKy0nRtndfFzs4O5HI59Ho9DAYDxGIx0uk0fvvb3yIQCCCRSBybveRlIIPaBw8esJ/WnTt38OjRI6ytrR2p5lU7f1bksfXgwQOsra1BqVSydcqJy+zUajWk02neHKnT5bgcKIcFvaCZTIYJ2BQAnTQIgxua13Z+4Q6iXC4jk8lgcnKS3duXl5cRj8fbnuAJPOtwfPToETY3N6HVavm2FAqFTuSaex5o/eXzeSwuLqJUKmF3dxcbGxvIZrPHak0eFpQlp+ydSCRiC5B2R73+1Ctrd3cXMzMznPUuFArY2Nho+7b5VwFxGufm5hCNRrG0tMTCgYVCoS0uC82A0Di0WCxCKpWiVCo1RA5D9CXqv8d2N6jX64fqW++Msb1xmDE2anzNSP82cg6FLtHU6dDsTfQLNe2WrtNGihXS92/1GJuBZr6LVJL74uc2fN22cg6F72mjNW3afZ0eRbPOi9ZpyzM7HXTQrjjuN3/h5tlKpdlWo9HP0A5jPGk4aGXTjJ/XKjQ6yBH+nHZHQ3mEDfvOHXTQQQcddNBBB22A1wp2Toqy8cvwlzDGDjro4M1AJYgOjjf+EubwL2GML8NLOTsddNBBBx100EEHxx2dMlYHHXTQQQcddHCi0Ql2Ouiggw466KCDE41OsNNBBx100EEHHZxodIKdDjrooIMOOujgRKMT7HTQQQcddNBBBycanWCngw466KCDDjo40egEOx100EEHHXTQwYlGJ9jpoIMOOuiggw5ONDrBTgcddNBBBx10cKLxUiPQ13E+lUgkr+1SKxaL2e1WqVQin8+jUqm88vcBGu8m/cXPeNUv3QexWPxGbr7t7Hp+VE7TjXJaPm5zKBaL+UEP+8xHMUaRSASJRIJqtfra3+dVxthsRXfab+RyOUql0mubTx7FPB7VmmwUDjPG1xmfWCx+IzPMZu41X/y8pu43wvF94Vz+yt+D0Mj95qjwps7nLxrjSzM7r+qlQT4xYvHrJYyEX69QKF77+7zOz2z01zTy+xzm57zO17zJszXLyRdozfiO+vs04ucc1XNJJJKm+eo0812krxOJRJDL5U2dx+f9rEavpXbda97k64Dm7jWvg5fN+at+fTO92E6al9ZLvbFeNYJ904jsKHGUUTrdPL74923zcrVzZueo0IjMzkldp+00LiE66/QZTvoYT/r4gMO/iwTK/rbDu3mUWdbjtt+8tIz1Gj/kKL/dl+Lgh92oNPDBtGE7TnAHh0er569R67TV4xKi3UsyHXTQSAjX/VEEOi8q1RH1Q/j/MpkMxWKRL+USiQS1Wu2Vyl+Hed52ercPs98cabDzqhCJRJBKpZBKpcz1AZ4+8Iu4OmKxGBKJhCe5VCrxRNLkNnISXvS9idsg/G96lnK53FYL4yhBi4zmhLJg1Wr1jXgeJwkUlAtT0Cf5c6FxNnOMtJfQOhSW3oSbfK1We2FZQDg3hUKhs4Y7OBK8CReJzhKpVMq0DuH3I7oH/SIayN7eHiqVCmq1Gp9LJ3kdH2a/aVmwI5VKoVAo0N3dje7ubphMJhSLRdRqNRSLRQSDwX0HaKVSgVgshkajgdVqhVKphFwux/LyMpLJJFKpFGKx2BuRt14XYrEYUqkUNpsNMpmM/7tUKiGbzWJzcxPlchnVanXf8x23xSfkYwlfRIlEAqPRCJPJBI1Gg3K5jN3dXSSTSWSz2WM3zqOCcIMCwJtWLpc7MZ+J8FZJAYZIJGpqgE+3WZ/PB6PRCIPBALvdDqn06faWTqf5udLpNGQyGeRyOZRKJZRKJf87pVIJAMjlcnj8+DF2d3cRj8dRKBTapgzRwV8OlEolVCoVjEYjrFYrtFotpFIpB+BisRjd3d1Qq9UQiUQwmUzI5/PY3t7GkydPkEqlkM/nkc/nWz2UhuKwpO2mBjt0OLrdbhiNRpjNZoyOjsJms0Gv10OhUAAAqtUqMpkMUqkUKpUKlEolRCIRZDIZtFotLBYL5HI5xGIxHjx4gKmpKXz++ect24wUCgWMRiMGBwdhNpthsVgwNDSEQqGAZDKJ6elpJBIJZLNZVCoV7vpQKpUc4Mnlcl6YqVSqKVmqw0IsFkOtVsNoNKKrqwsymYwzV2q1GjqdDgMDAxCLxSiVSpicnEQikWj1YzccFORSEKNSqaDT6SCVSlGv16HX66HVamEymZDNZhGJRLC5udkWmw8d/nTQ02ZBa46yIxKJBDKZjL8GeLreKfuhVquh1WphNpuhVqsRDofh9/ubOg6lUgmr1Ypr167B7Xaju7sbBoMBCoUCcrl8X6YxkUjwuHU6HbRaLeRyOarVKmQyGWq1GtLpNLq6uuD3+/H48WNsbW1xpuc443kNH+2yx3TwDCKRCAqFAk6nE3a7HV1dXXA4HDCZTLDZbFz1kMvl8Hg80Gq1UKlUEIlESCQSWF5eRjqdRiAQwMbGxokN1IVc2sOgacEOdTxoNBoMDQ3B5XLB4XDgzJkz0Ov10Gg00Gq1nI6qVCqIx+Mol8t8+5LJZFCr1dDr9dwOm8vlEI/HoVQq+cbczImlQMBut2N4eBhOpxMulwtDQ0MoFotIJpNQKBSIx+NIp9MoFovI5/OoVqtQq9XIZrOoVquQy+WIx+NIJpOo1WqcCaJUJNC6TJBCoYDZbIbP58PIyAhUKhWkUilKpRI0Gg10Oh36+/uRyWQQjUYxOzt74pj8wP5sFgUDcrkcCoUCKpUKVqsVVqsVKpUKMpkMFosFBoMBVqsVsVgMfr8fyWQS6XS6pRsQZR/lcjnUajVKpRLK5fK+UjL9vVwu5/eS3k2NRgMAKJVKMBqNMBqNcDgcUKvVWFlZwd7eHmKxGCqVSsMDBJFIBJ1Oh8HBQZw7dw5erxcejwcqlQoKhQJKpZL3inK5zBcJkUgEvV4PtVoNuVyOSqXC3IZMJgMAMBgMyGQyiMViXC4/LoeGMJgVlv0JtH6FJTshHeC4oV3Jsq8DsVgMo9GI3t5e9Pf3w+l0wmq1wmQyweFwoFKpoF6v8x5De3A+n4dMJkM0GoVOp4NSqdzX1t/ssrJwzQnLxXTxeN7XAC8/5yhgp45t4OlFjfaal2V4mhbsSKVSOBwODA4O4sc//jEGBgbg8XhgNBo53f+i2uLBg7Ner6NUKiGTyaC/vx/xeBwbGxu4d+8estksSqVSs4YFpVIJl8uFq1ev4gc/+AE8Hg+n0Cm9dubMGd5IUqkUisUiqtUqJBIJstksisUiCoUCNjc3sbm5iYcPHyKVSiGbzSKZTHImiBZ5s+FwODA2Nob33nsPX//61zl7EY1GOaVaKBSwuLiITCaDZDLJGauTsgFRsC6TyaDT6fiQFIvFfOOamJhAV1cXLBYLvF4vlzVFIhECgQDu37+PbDaLeDz+yoTBo4JYLIbZbIbVauXbYiQSQTQahclk4oORLh86nQ4+n483mXK5DLVaDbVazePWaDTQaDQolUpYWFhAV1cXfvvb3yIajSKdTjd0PDKZDAMDA/jJT36CiYkJGI1GaLXaPzvcXwVmsxlOpxPj4+Pw+XxYW1tDLpdre/4OjZf2UplMhq6uLs6a0/PTWrZYLCiVSlhfX8fe3h5nllu1zxwWwgOcLh8ikYjH187P/mUgzaeJiQl873vfw40bN2CxWPjvZTIZyuUyKpUKCoUCwuEwcrkcyuUyotEotra2sLy8jGg0ilqtBpPJhFwu17TnpyCH9krKAhM3l0rJqVQKwP6znvZKylwJgzT6vnRRk8lkcLvdfEGJRCJIp9N8UXkemhLsiMVinDlzBlevXsWtW7dw+fJljjyFvIaDRMEXgdLSuVwOU1NTWF1dRbFYRLFYfG0RwtdFd3c33G43rFYrdDodyuUyNjY2EAgE+GZVKpVgsVhgMpng9Xr5a4UkZgDI5/PI5XKIRqOIRCIIhUJYXl7G0tISgsEg1tbWUCgUmv4yezweeDwemM1mSCQSbG5uYmFhAe+//z6y2SxnBeLxOPb29hCNRpHP51Eul5v6nI0A3UgUCgX6+vrg8XjgdrsxPDwMl8sFs9kMjUbDtXV6sdVqNcRiMWcTpqensbi4iHA4jEKh0PR1KhKJYDQa0d3djWvXrmFkZARjY2PQ6/UolUoolUp/RrCnMhBlcggUEBH3hXh12WwWtVoNpVIJv/nNb5oyxkqlgmw2i62tLZw7d+65WYxXBY1fo9HA6/XC6/UinU5jc3PzCJ/8aECBDRFR6VDR6/UwGo04c+YMvF4v7HY7nE4nZ7zUajVkMhny+TyCwSAWFxextraGyclJbG5uolAotMX7S6VUIbGcyo31ep1LkBToPI8T+bzLcjtCJBJhcHAQY2Nj+OEPf4hTp06hq6sLEokEhUIB2WwWa2trmJ2dRSgUQiwWQzAYZO5OPB5HIpHAzs4OwuEwZ2yF1YFGQqPRwGazYWRkBOPj4zw3IpEIarUaGo0GyWQSDx8+xJMnT5DP51EqlVCpVPispDZ9uuDXajVODGg0GvT09MDj8eDMmTNwOp0olUqIRqP45S9/iUKh8NLna3iwQxuH1WqF1+vFyMgIurq6OMo7LCgooDRzLBbD+vo6pqenEQgEkEgkmjapQgi7yUqlEtLpNHZ2djA9Pb3v77u6umC329HX1welUskHBb28Op0OJpMJ9XodXq8X8Xgc4XAYZrMZADiIagWEN8VsNov19XXcvXsXd+7cQSaTQaVS4XZHYbq/XTeVw0JIvO7q6sLQ0BC8Xi+6u7sxMDAAu90OnU7H665WqyEWi3Hwl06nmYP16NEjrK2t8d83+7ORyWQwGAzo6enB2bNncebMGYyNjUGtVvO/EaaWhZvPQZ0pAv0ZlYiKxSKy2SwymQxnLxuNer2OYrGIeDzOa+91Ax3hTbJcLvM8KZVKKBSKtiuVUECqUqlgNpuZI6ZUKmEwGGA0GnHq1Cl4PB7YbDZ0d3dzsKNSqQAAhUIBPT09sFgssNvtTC6PRqMt5d0RF0ulUkGj0cBoNHLwTc9N2eNkMolcLreP+P+8NSDs9m1HiMVieDweXL58GaOjo3x5ikajCIfDCIfDmJubw9TUFLa3t/mMEIvF2NzcRDqdRi6XQzqdZlI90JzxikQiuN1u9PX14cqVK7h48SKrkhPVQ6PRIB6P8wUqm83ysyoUCr6oEGdXqGgul8uh0+l4Dx4bG4NOp+OmpMNkyhse7EgkEiiVSmi1Wmi1Wmg0mkPfvGiSaCCVSgXFYhHz8/OYmZnBBx98gD/+8Y9cGmrFIo5GozAYDEgmk1hbW8P8/Dw++eQTPHr0iCNWInEajUa89dZbcDgc+9LtOp0OIyMjMJvNUKlUUCqV6OrqgslkgtPpRKFQYGJlK7C9vc1R9NLSEj7++GP87Gc/43IMcHSS7e0ASplqNBpcuXIFN2/exLvvvstZHHqJ8/k8tra2sLS0hN3dXcRiMe6CEIlEWF1d5cyX8IVsxeej1WrhdDoxNjaGixcvwuPxwGAw7KulvwzCy4YwKKASZjwex+TkJO7fv4/f/e53iEajTckM0M/f2dlBMpmEyWSCTqd7rXeFMgTFYhHpdBrRaBTLy8vI5XKo1WpvbA1ylBCJRBzkuFwufO9738Po6CgGBwdhMBg43U+8jYPzTKUBCow8Hg/Onz+PCxcuQKFQ4MmTJ9jb22vJWqULstPpRHd3N3p7e/H1r38dDocDBoMB6XQawWAQoVAIgUAAMzMzCAQCfLASwV44Vlq37Uwyl0gkOHv2LP7mb/4Ger2es/yfffYZ7ty5g+XlZWxtbWFvb495VjQ/4XC4Zfsu7Zd//dd/jUuXLuHatWswGo08F0IZiEqlgkuXLkEkEiGXy2F7ext7e3uwWq1cxioUCggGg0in0yiXyzAYDFw69/l8nJWMRqPY29vD1tYWQqHQl5bMGx7s0I3S4XDAZrNBp9O91AaCJqxSqWBvbw/JZBJbW1tIp9OIxWJYXFzE6uoqIpEIgsEgUqkUSqVSyyaaunAMBgMTpzOZzD6yFKXr0uk0fve730Eul3PrPXX0aDQauN1u3ryoBl0oFPDxxx9jY2MD+Xy+JeMsFAooFAoolUrQ6XRQKBRMDiPQ5nncAx0i7RqNRnz/+9/H2NgYBgcHUa1Wsbq6inQ6zbcsShkTR6lSqSAWi/F8J5NJzg60MtMlFovh8/kwOjqKCxcuoLe3FzqdDgB4XovFInc/EmGeOrVyuRw2Nzext7eHXC6HTCaDQqGAfD6PeDyOfD6PbDaL1dVVvoU2q0xXr9eRz+d5s6My78Fgh0p19HkIuR7A00BnZWUF29vbmJubQywWQyQSwfLyMhYXF7lxoNWQSqUwmUzweDz41re+hf7+fvT29sLpdHI5lcqRwNO9hw6dYrGISCSCZDKJeDy+T2doe3sbm5ubmJqawoMHD1oW6BDRv7e3l8fX19cHm83Ga7JcLqOvrw/5fJ5LOfF4HLFYDMCzLkm6lEQiEaRSKaRSKUQiEdy/fx+ZTKYtynRCCEuwmUwGjx8/xvT0NG7fvo3NzU3EYjEOAFopdCtsVqBSotVqxdDQEHp7e6HRaJDL5ZBMJrG7u7uPv0n8MazVYUUAACAASURBVCqlUrlVp9NBJpNxbOBwOLgERp8JBU65XA6pVApPnjzB9PQ0Jicn92WyXoSGBztKpRI2m40Jm6RlQaBsDaXAi8UiyuUyCoUCIpEI4vE4b7SxWAxLS0vY2dlBNpvdd2tuFej2RClkmUz2Z0Q5ujFWKhWEw2EA+8sEtHh2dnZgNBpht9sBPBNXpC6eVt1KqERRLBY5ELBYLNjd3T3WukEHIRKJYDAYYLPZ0NPTg4sXL8JkMvFNIxKJIJFIYHd3Fzs7O0gkEtypQ3NDXAf61eqbpEQigVqthtfrRW9vL7xeLzQaDWq1GhKJBAKBABP7SIisXq/zoVmr1bh0ube3h0wmw2UqyuhQ+ZI2tmZfPug9EepAHfz7RCLBZE6CsDRFWcvNzU3Mzc0x/2FrawuJRKLlwqASiYQDAY/Hg7GxMVy6dAlutxt2u533HZof4msIlXSpFB6PxzkrS3+3vb2N3d1d+P1+RCKRpjZ5AE8vxRqNBk6nE16vF2fPnsXExATcbjccDgfTHoR8MgrutFotj5vKrpTZKpVKiMViSCaT2NvbQzAYxPr6OjeLtMOeReuWOgeLxSJ2d3exuLiIubk5rK2tIR6PI5fLtWVXIJWpDAYDtFotJBIJB6ELCwv79hWr1cpdngaDgeeQEgXCIJUydMJzni5Z6XQac3NzWFlZQTgcPtQ+29Bgh0iRw8PDuHLlCvr7+3mDAcAPHwqF9r1oFIXTrZkyPBQtUut2O9y05HI59Ho9urq6uMxxcMOl56RSHOFgWjmZTHI7s3DyhBtWK0Dt8plMBl1dXejt7cXw8DAWFxebTrRtJKhmPjo6iosXL+Ly5ctYXl7G/fv3cefOHb5ZFYtFhMNhpFKptrsdHoRSqYTD4cD58+cxMTGBwcFBSCQSRKNRBAIB/PKXv2SyI5XfJBIJ9Ho9KpUKE4BjsRiy2SwTzyl7IhTca9UmTN0fxFc5mNWhrM2dO3f2aebo9XoubRBRNxaLIRQK8cWLSNetPFyoFO5yuXD58mWcO3cOb7/9Nrq7uwE8vYyQLhnprNB80nwRgZmkAYrF4r6giLh3L2oLbiRovfX39+Py5cvczNLd3c2ljYP7IX0uJINBfEe5XM78QsooAE+zk1TymJ+fZz2ldggaaM+3WCyQSqVIJpP45JNP8PDhQywsLCAYDDaNA3dYCLOn9P6p1WooFArUajVsb29jenoaH3/8MTY3Nzl4sVqtfMHXaDTQ6/XQ6/V8wZdKpTAajfsukOvr61xK3tvb4z/f29t7JTHhhgY7RqMRTqcTg4OD8Hq9MJlMHAiQuvDdu3fx+PFjLC8vc/2OXlwhW1uoA9EOQQ6B+Dh6vZ5FykiX5Mtw8EWjMR48QFv9QlJJJhQKQSKRoL+/H++99x5+85vf8MZz3KFUKmE2m/GVr3wFFy9exIULF6BUKrG4uIhf/vKXWFlZ2XfQN0ND5ihgNptx7do1XLp0Cf39/VCr1UgkElhYWMDnn3+Ou3fvIhQKsaYVHezC9l7qiHge8bwdiOg6nQ6jo6OcCj94iahWqyxNQZksKgnTfkIBT7lc3nd7boe9xuVyobe3F9euXcO7774Ll8sFu92OjY0NbGxsMDeMLobhcBiZTAbZbHafbQAAviUfnE9ay82eS5FIhNHRUQwMDDDfgy6O4XAY0WgUoVAIW1tb3MpMTR1isZjLsPT8FouFeUrRaBRSqRRut5tLrpQlKZVKHBC1EpR5NZlMeO+996DT6TA1NYU//vGPWFtb4+7NdliHhINE71qtxmd1vV6HQqFgukOxWMTe3h5nCuPx+L5SOTXqGAwGnkODwYBcLsfaV4lEgmkCQlkEWteHTXw0NNgxm83o6uqC0+lkpVLgWQfF3t4epqenMTU1Bb/fj0QigVwuh2KxyINt9a3qy0C3CMrG0Gb5umjHsVarVQ5OSdHaZrNxarkdn/lVoVQqYbFYOG1uMpkQDoexs7ODnZ0dzuK0+3oUggioDocDXV1dMBgMAMBKzsvLywgGg0gkEshkMk3zlztKUJs/iTkSz4ggDNaIs0Kco4MHfjvuN9TlMjQ0hHPnzsHn80GpVCKbzWJxcZGlKSjzTd1/FASQaOnzxtjqcVI5anh4GGfOnMG5c+fgdrshk8mQTqe5rEhKwBTskKI38bWEh53RaOTsXiqVgkQiYese4ttFIhHO2LUaJFnhcDhw+vRplMtlBAIB5sK1u7YTAL4oUNBJgp1UBUilUvzOUaBJTgJC2QoKdhKJBH8vsVjMn8Ob2i01LNgRiUTo7u6Gz+eDz+fbR5yr1+vIZrPY3d3F3bt3MT09zamudliArwISOJLL5SgUCiyqB7xYL6jdF+9BULaJyhgKhQJ6vZ5tI47DC/kyUErV5XLB5XLBYrFAoVDA7/djY2ODZQ3a4YB4FYjFYqhUKthsNthsNmi1WtRqNQQCASwvL2NhYQFbW1st56O8Cah8ZbfbuUvjIOr1OquvC0s7wiCgHcdPJcXh4WFcuHABV69ehdlsRjKZxObmJj777DMsLi5iZWUFiUSCxyCcz4NjO4xKbTNA2RmTyYRLly7h4sWLOHfuHMRiMWdg/vSnP2F1dRVra2v7MjsKhYJbzw9mZhQKBUuBkAwBtTPTL8p2tTozKxaLuXmnv78f4+PjePLkCfx+P1vKEO+o1fP1MlSrVebQUhbKarViYGAAuVwOYrEYu7u7CIfDyGazAMASBxR8058DT8tTjUBDgh1qN7927RomJibQ29u7L71MkWAymWSF4ONoL0Bt4wqFgrM5RCIkJUcSxaLaOJGv212lVIh6vY50Oo319XWEQiF0dXVBo9HA5/NBKpVyHVVYajxOkMlkTDK32+0cFCSTSdTrdWi12n26K8dl3qiObjKZIJVKUalUkMlkMDc3B7/fv4/Yd1wzdG63Gz6fDw6H47klLODpgXHp0iX27woEAswNpEwy3SzbZe0SydZgMODKlSu4cOECTCYT0uk0VldXcfv2bXz++ecIhUJIJpPMnaPDXSj8CDzdh0wmE5RKJcrlMra2tlpCdqXno7b5M2fO4MqVKxgaGoJCoWDRvI8//hiffPIJ8+TojKBMwMF9hho+SFjuZW3m7bD3Uhev2+3G+fPncf36ddYR0uv1GB8f55bqYDDYFs0OLwIFKx988AHS6TQMBgPTAi5fvoyenh6USiUUCgX8+te/Ri6XQ6FQwN27d7ma0ww0JNiRyWQsSU+6F0JtnYOLnW5Yu7u7Le+uelWQ8JVcLodKpYLX68WNGzdgNpv3qX1St0AikWBT0FZpA70OhERVo9EIm82Gmzdvwu/3Y3NzE5FIBPl8nm9b7fxyHgTdBOm/aa329PRgdHQU2WwWc3Nz+0oFx2GNUmcDcRjokhGPx1mygQ5GIoEet2CV5u6gFw/wLIshkUj4fbx16xbLBmxubnIX0sLCQtuMm7pSjEYjxsbG4PP5YLFYUKlUsLKygrm5OczMzCAajTJxk7gfZMwr7PChPaq7uxtSqZRv0fTONnvPpWd1Op149913WWQ2nU5jdnYWk5OTmJub4/FRaYTEHSUSCWdmisUiy3h0dXWx3UA6nWauiDCga4egngjZRDgneQuhByFxWkiFnXhmX6YS3ApQqXhpaYn3j76+PpZjocuWWCxm8cd0Oo1IJMLuB82oDjQk2FEoFLDb7TCZTNySLexQooyI2+3GzZs3uS2StCzaZdP5MlD5Q6PRMFF5ZGSExQCJpV4oFBCNRhGPxxEIBBAIBADgWAUElUoFuVwOkUiEXeu//e1vY25uDvPz86xNkkqlsLe3x8HccZhL4h4d7JQ7deoUxGIx9Ho9qtUqNjc3WYb9OGR4KNghPScKdiijUa1W9ymXCn2R2n1sBCEXBfjz0rGwu4wc6EulEnK5HGZnZzE1NYX5+Xmsr6/v4/C0CnRBomDg7bffRm9vL4vMTU1N4cmTJ5icnOSMKgUP5JDtdDr5+5GYqclkQl9fH+r1OmKxGAqFAkQiEb+zzZpzGp/RaER/fz++9a1vwWazoVwu76M1kM8eaavI5XKYTCb2UiTORyqV4tblsbExiMViFItFFvsU6iO1y7pWKBSw2WwYHR3FX/3VX8Hn88HpdCIej3ObPdEj6N0MhUJIJBJtGewATz9bEj2cnJzEjRs30Nvbi8HBQTb7JlkPunCtr6+zV2AzbJAaEuyQOSaZIpJPEIFaKRUKBW7cuMFBA/lBtUOa8ctAmg8ulws9PT1wu90wmUwwm83weDyYmJgA8My0jtJ4m5ubeP/993H37l3cu3ePiZLtDCJek9mn2+2GWq3GxMQEzp49ywKQ9+7dw6NHj/C73/0OoVAI9Xq9bV9OIcidfnNzE36/n1OwZrMZ58+fx/DwMPr7+/Hhhx/ynFFXSzuDNkwibEqlUuh0OrhcLhY8NBqNqFQqyOfzWFhY4ExBu79/BBIxFHIBngdh0FOr1WA0GmE2m+FwONDT04NoNIqpqamWaMwIoVQq4fV68c1vfhMTExO4efMmjEYjIpEI7ty5g08//RSrq6sIhULseu1wOPCDH/wAfX19cLlc+zI7RqORxfhIkK9QKODmzZv4xS9+gSdPnuDevXvsUt9oyOVyOJ1O/OM//iOuXLkCn8+HSqWCpaUl/Pd//zc+++wzLl2RrYVer8fExAS8Xi8cDgdyuRwUCgW0Wi1GR0dhMpn40knv5s7ODv71X/8VU1NTSKVSbXXpun79Os6ePYu33noLV69eZW0aqVQKq9XKbd3Dw8OIx+MIBoMcBP72t79t23eTODiZTAb/+7//y0H7W2+9hZs3b+JrX/sa3G439Ho9LBYL/umf/ok7tlZWVvZ11TUCDQl2SBGYyGLPA20+KpUKXV1d8Hq9cLlczNxu1wk9CDLt3Nvbg0ql4nIAHYblcpnTyVKpFBaLBSMjIyzCRp4m7QwiKOdyOayvr2N2dhZWqxVKpRJOp5OD2tHRUW67n5ycxOrqKvx+f6sf/0shLO/cvXuXA1CPxwOlUgm5XA6HwwGPx4NQKIS1tTXmYLUzNBoNG+7S4adWqzE+Ps7O7MlkksUR6SYpVEhtd+h0Ouh0OiakCg0jnwdhuYtsWWq1Gr761a9CIpFgaWkJS0tLTR7FU1AmuLu7Gz09PXC5XDAYDJBKpUilUnj48CGWl5e5HVmlUmFgYAA3b97EtWvXYLPZOLgRmigSIZdKWtShNDIywkq9QpXaRoOMSg0GA5shKxQKGAwG6HQ61Go1DmTcbjd6enrQ19cHs9kMg8HA5VepVAqbzcYdWpQFoTOnq6sLRqNxX6dhKyH0iXQ6nXC73axET+KHJBlQLpdZxsRut8NoNO7zsWtXUPZMqMw+PT2NcrmMWCzG0gkWiwV6vR4jIyPcDRqNRpHNZhs2T0ce7NBmIpVKOb1cLpf3SZMLNyKJRMKMdLvdziWedodQeXRjYwNbW1vM+yCJcupeEovFsNlsMJlMMBgM6O/vh0gkwtLSEt9i2hk0j7lcDhsbG1Cr1RCJnjqBnz17lg0z3W43k5eJGLm6utpWt6rnoVarsRTC/fv3OcVfqVT4ACHH8EgkgsePHyOdTrd9QCA0sCQOgFKpxLlz59DT04NwOIzZ2VkEg0EAgMlk2udDd1AJvB2h1+uhVqu5W5BIuQczyURsFe5BxOUhITTKdrUq2AGe2UEQ35HGkslk9qnHU3l1ZGQEX//61zE0NMQGwxS85/N5lk/IZrMwGo0sQWA0GjEwMIBoNMqBQrMy6tQdR4RyiUQCrVYLr9eLjY0NZDIZiEQivPfeexgZGcHo6Cj0ej3zs0i6hNrJyQCUgkXqcqI9l9Z3q0HZVbqEaLVaFAoFJuyurKxgd3cX0WgUhUIBY2NjzH0hT77jBHIvJ3unxcVFSCQSTExM4NSpU1Cr1RgcHES9Xoff72fF70ZdIhsS7CSTSTx48AB2ux1DQ0MYGRlBT08P1Go1lEoll7WEDHqyhchms8emM6tWq2Fqaopdzs1mM7+MNBZazF6vF4ODg/iHf/gHaLVaDAwM4MKFC3xTa+cDBXi2cHd3d5HL5bCysoJCoQCr1QqXy4W33noLV65cwalTp7iF1Gw24969e6zO2q6gYIeImhTUvf/++xgfH8fFixfx1ltvYWxsDEajEY8ePUI6nWbdoXbF1tYWlEolPvzwQxgMBrhcLj7w7HY7BgYGcObMGVbInpqawu3btzE7O4tIJIJwOMyk7HaESCTC7u4unjx5wgKlFosFVqsVarWaSwMajYb/nLon6VCkjKvX68XAwABCoVDLglhSiP3973+Pr371q8xNEYvFsNvt+N73vge1Wo1qtQqr1YobN27A6/Wiu7ubNWcymQx3NPn9fszMzCAUCiGXy8FsNuPWrVsYHx/HrVu34Ha7Wew1Go02pXxXLBYRCASQTCY5g0rK5T/60Y/wne98B2KxmLlk9Isyy5QtX1tbw8rKCj788ENEIhHU63WcOXOGxTNHR0fhdDpht9ubmrV6Gei8W1hY+DMbC7pE0V7pdDrx/e9/H1qtlgMerVbb0ud/XdCeurOzg//8z//EyMgIJiYm8M///M8YHh5Gb28v5HI5/u///g8PHz7E6upqQ56jIWWsUqmEeDzONfBAIMA6H0ajEVevXoXRaIRGowHwzLiNNqhWqHiKRKJXzkCQXhApSFLXS71eZxE6Un2kjoFkMslRvU6ne2GZr91Anw11kNFGRQ7RJBAmkUj4lmkymTAwMIDl5WXWHmoUXncOCdRRkMvlsLu7i0wmw7dPlUqFa9eucard4XAgEAg0jefwuiiVSohGo3jy5Ak8Hg/6+/vR39/PQTll5+RyOTQaDUZHR2EwGHD+/HnO+qyvr2Nubg6pVKqtugcpW0Pk/3K5jFAoxB1JGo2GyzZ0eFBJhMquXV1dnPFRKpWw2+1wOBwt1Y6iNt7V1VV2+aZS3fj4OIxGI+r1OtRqNcxmM8RiMYLBIAKBAGudBAIBbG5ucpaAFGgzmQxWV1dhs9kAPDP4babIIDWj3Lt3j8VKDQYDl6KE73E8Hmc1fdJhq1arUKlUWF5extraGubn51n7S0hortfrbDPULiDu5vb2NkqlErsFkJEu+cwBz4jcMpmMydjN5pLRu/GmexxVBuj3lZUVrop0d3dzF3NPTw+2t7fZu+yocWQnrVDfoV6vI5PJwO/3Y3d3F+vr69BoNDAYDLDb7RzJUQ2Svo4Oy2bjZTX+L4MwnSpMnR/cQKhLiUokwtvlcYFQWp/GQSqt9XodNpsNBoMBbrcbUqkUBoMBfX192N7ebrjpnrA88WU/R/jvhP+WNiPq8pBKpRygl8tl1sYQZgianfV4lRtqpVJBKpXC8vIyJicnUSwWefMkDyFSAJfJZGxDQPV1q9UKi8WCvb29phqbftkYqfRB5VRqZQ2FQrxBU9BDQoLd3d3wer0YGRnB8PAwALAfj0gk4m4fsrRpVdMABQN+vx9GoxEul4u5Yz6fDz09Pbx+U6kUwuEwgsEgPv30U84UkO0OkbYpA1ksFhEKhRCJRAA89YtqhfdXtVrF9PQ0CoUCFAoFB3WkA0SZmO3tbR7fzMwMB9smkwmrq6vY2tpCLBZj4jmAfQFCuwU7VB4mbgp1xRG/iiQtiMtKkiZU3mlVsPM80DtILeXEmaML8MH1RGcHdd3VajVsbGwwt9dutzP/k4j0R70mjyTYoVqpwWDgyaFonB5YJpNBr9fD6XTinXfegdFohNVqBQBWHqY0ZbM5Hof5eS9THiUBq5dtkDqdDl1dXZzNqVQqvNDb5bZ8WAiDBLqdra2tQaVSccnObDajp6cH58+fx/z8PMLhcEOzIIedQxK8pA3xZXypSqWCra0tAEA4HGZStsfjQVdXF+u1NBOvslYoU7W9vY1PP/2UNVrI6oP4ct3d3XC73Sy6p1AoeA5Pnz7NJT4yjGw0vizQoa6VK1eusBM9mQYLRRKFvwcCATx58gTr6+sAAJvNhlqttq+cTgdMO4BS+hsbG/jhD38Ij8fD/CTKytVqNUQiEUxNTeH999/n0tBB7g1194jFYqTTac7EkgFzK1SkV1ZWsLW1hQcPHsDr9aKrqwt9fX2Qy+U8Z7FYDNvb2wgEAtwUQJo7wnOFSjyDg4NwuVzQarXY29vD5uYmd4W2yx5LBqSkMwPsX+8UeA8ODsLj8cBsNrM2G4mcNgsv2q9JCoAai4hwnU6nsbm5ic8///ylreTFYhHxeBy/+tWvUK1WWdC1r68P8Xgcf/zjH1kH7CjxxsEO1VdNJhNGR0fZXZZuFUIdDNK3IOdo4Fm0Sy2D7VYWoCibSIwkZf0qi04ul8Pj8eD06dPc6ptMJtnN9biDgr1oNIq1tTWsrq5yKyiVE+Ryecvnlm6Ap06dgkqlQjKZxOLi4kt9cohPNjMzg9OnT8NqtcLhcMDhcGBnZwdra2tNHsWrgfhI29vbSCaT2Nragkaj4WCHOpK6u7sxNjaGnp4eOJ1OPjQsFgt6enr4ItNqUBeVy+XCN77xDS7hELGctJ2EdgkAOLMhtA04CGqmaAeOR7VaRSKRwN27d2G1Wlnagqxp5HI5VlZWsLi4iMnJSdZgedHliT43KqGLRCL4/X4sLS3t889qFihLSF6CwWAQKysrUKlUnFHY29vjwICyHhS4CYnNPp8PQ0NDOHv2LLq6ugAAjx494sxJK6RMKCPyskDreX9OSYGxsTHY7XbU63UsLS0hEAggFAo19JkPCyrtkwHvwMAATp8+zTpyWq0W09PTPG8UpAr1jsrlMpaXl9HX1wer1Yrx8XEOeIeGhrCwsIB4PH6kz/1GwQ7dFvR6PRuZxeNx7OzsMKNcOECha/nBmwcFRO3UuUOdHUajEXq9njdMyj59WaqbNiaz2Qyv18vlu0wmg93dXWxsbLAHynHBizJc9XqdjfZ2dnYwNDQEo9HIrc+t5ibRBqnVanH69GmoVCrs7u4iFAq91FWYshkLCwuw2+2sv2O1WmE0Gps8ilcDbbh0m8xkMohEIvvadKkjaWNjg1tFgaclHplMBo1GA6vV2jbdIDSHDocD4+PjzAXc2dlBKBRCtVrdd7gd/J3GcTDzQ/+GDtN2QD6fx9raGh4+fIitrS0YjUbek8haYX19HX6/n/WRXgRSradgBwB2dnYQDAZboggu7PorFAosnEet8RKJBOl0el9XIPAss6dQKFj6or+/H8PDw6z/tbe3x4rnza4U0D5D3E3g8Bdj0puz2+3o6+vjbq319XVsb28f+eH/uhCqrptMJrjdbpw+fRqJRIJL/iRAm0qlOONI5S26TMRiMfbMqtfrHEd4PJ6GdGW/8QkkEokwMTGBiYkJfOMb30AikcDS0hLK5TJ3SVSrVU4TU1eIXq/nhUHCUcQjaBeoVCqYTCacP38eLpcLxWIRDx484Lbyl+kBSSQSdHV1wePx4Ktf/SrGxsbg9XqRzWYxOTmJqakpfPbZZ00pCxwVhJICzwv2stksE9KJJzE0NASLxQKlUolUKtWiJwe/ZHa7Hd/97nchFovh9/tZmfRFtz/qcvl//+//sSs68T92dnZangF4EYgXRvwTOjSo/i8EGS/Ozc3h4sWLuHjxIlwuF6xWK6er20Xno16v4/Tp07h06RIGBgbg8XjgcrlQq9UQCoWYH0YXKvqdPIeuXr0Kr9cLtVrNytl00aJDtF14dNVqFalUCh988AGAZ5dL4NkBeljFealUCrvdju7ubtjtdlSrVWxsbGB9fZ11eFoFIQ3g4NoEsC+TI5PJ4HA4+D0cHx/nQMdgMPD589FHHyESiTRdC0voSya84B8GUqkUp06dwo0bN3D27Fnkcjmsrq7i3r17mJ+f55J6q0EyJFtbW9je3kZPTw/zbnQ6HcxmMywWCyKRCKLRKObm5rC9vY3t7W0A4GBdIpEgl8vx/qvT6eBwONDX14dHjx4d+d76xpkdpVKJy5cv4/r16xgcHEQul4NcLsfy8jL8fj+nr0hNkVLjRC6km6Xb7YbH42mrA8RoNKKnpwfXr1+HyWRCuVyGyWTCo0ePsLOzg52dHa4rEm+JVGsVCgUmJiZw+fJlXLlyhQW0ZmZm8OTJE0xPT7d9S7YQND6v14t8Ps8p5udl6KrVKqefLRYLxsbGkEql8OGHH7ZwBE8PBqlUCqfTyYaI3d3dmJub+7NU68Gvi0ajCAaD2NrawujoKFQqFfMn2o13RUrJ165dg1KpRDKZhN/vRyaT2VfqobHS75lMBisrKxCLxfjKV77C7vZms5nbeBvVFnoYUMn82rVreOutt6BSqbikNTExgeXlZbb0IAn6QqGAfD4Pk8kEn8+HCxcuwOv1wmaz7ctSlstlJly+ScNCI0BCic87OA+z7qhc6XQ6cerUKQwMDHAZ6cu4ho3Gwc+auqmonCMUCjQYDLBYLBgfH0d/fz+LL5IGD3UQTk1NYXZ2dp+TdrNAmTe9Xs+dgMRXfB4RnLquiApCyspyuRxTU1N48OABFhYWkEgk2kZpn8ZSrVbZmX5mZgZ9fX18wb9y5QprX/X29mJxcZGDHqVSCY1Gw+KWNpsNcrmcfdooS9l2wY5YLOaav8lkglqtRiKRgM1mg0ajYTEzSv93d3dzqyHw7BDVarVMAG0HkKWFxWJBX18fTCYTAMBisaBWq8FgMEAkEjGHRyaTcUqdUsXj4+OYmJiAz+dDoVBAIpGA3+/nToJ2V+AVgjJzPT093GZPNzFhwEa3ZDKts1gs6O7uRnd3dwuf/tmzUYBOZqY9PT1QqVSssvqiF6xQKHBHnfD21srOnReBFJ8vXLjAt12ZTIZwOMwebaQALcwQkNQ7GdWSxgsFPTqdrqXjolKM0+lkDyGRSAS9Xo+enh6Mj4/D4XCwhQSl0PP5PLq7uzEwMACfzwebzcaWCsB+h2wKKtopeAX+vBx3WBDhVaPRwGazcZaOuntedAg3E8JOShLCVKvVfJ6o1WoYDAaYzWbY7XZMgIBaDQAAIABJREFUTEzA4/GwsjA1xPj9fg50YrFYy95L6gbs7+/nrrDnPYvw3aJgfHh4mM8L8vcKh8NtRXcQdhpvbW0hkUgwfaWnpwfFYhHnz5/n4LRcLrNBbTAYZNNsABgYGIDT6YRMJkM2m2VbpbZrPRduEhTRUpBA/heU+n/nnXdgsVg44KFgB3j24TmdzrZIlQPPzEqpfurz+djQ9OLFi1hcXMQHH3zAbskajQa3bt1Cb28vvF4vT6BYLEapVMLGxgampqbwq1/9Cmtra21Tfz0MaBOy2Wy4ceMGd4F8+umnzM8iEHeHjEAVCgU8Hg+8Xm8LR/AUQtsLqg2/8847+PWvf41kMsmHO6Wfgf03z3w+z91bVBpqBzLrQej1ely7dg3f//734Xa7US6Xcfv2bSwsLODhw4d48uQJd3cIn53auY1GI7RaLXNbRCIRa720CsTVsNlsrP1BJW+NRgOv14u/+7u/470klUpxWUssFsPr9cLr9TLvRViqIskBKvU1arNtBUjd3OVycUlZKpVib2+PdV5aOVZae5S9GRoa4osxOYP39vZiYmICOp2O1yfNX7VaRSgUwvz8PH76059iZmYGkUikpdIBwNM1Setxfn7+zy6FVAHo7u7G4OAgxsbGcOPGDfT19cFms+Gzzz7D7Owsc4/abT3S2U9K3bFYDEtLSxxU/+QnP8HAwACP79y5c9zpSdpz9XqdPdsA8AUlnU6zW/1R4o05OyQwR61mtVqNBbAMBgNUKhVcLhd8Ph93NR2si8diMfj9fty7d4/VMFuNWq2GQCCAYrHIomNdXV34yle+ApPJBI/Hg7/9279FMpnkSL6rq4t5R5ubm1wTn5ub4zbIlZWVLyUTthuIFOr1enHx4kU+cLRaLT755BPE43HOigDgz4y80TKZDGfAWjkG4Ckp87/+67/w4x//mFsnv/nNb7KbdCwW25dxozSz1WrlTgHi8VD2ox3WqxClUglbW1tIJpOw2+3QaDSsbv3WW2/h448/Zqd6ALz5kOHi+Pg4CwxSpo48e1oF4WdM+w2p71IgJFSYJb5df38/v7/EExCWTWjD3t7exieffIL79+83dVyHwcEyz2G/RiaToaurC9evX8f4+Djee+89aLVapFIpfPLJJ1hfX0c6nW7pXiQSidDT0wOj0QiLxYJbt26xwJxWq90nEkmfQy6XQzwex+7uLm7fvs2u9Y8fP265QS9dgorFIiKRCF/+KSNMcgekKzc0NMQK7UNDQyiXy1hdXcX//M//YH5+njVp3gTNuIxRKTidTqNQKOA//uM/2NtrdHQUo6OjGB8fh8vlYoqD2WxmsUvaj1ZWVvDgwYOGJAPeuIxFAkDEAwCebjTd3d3QaDRQqVTsLySTyViEiFCr1ZDJZBCPx5FIJFrqOCxEvV5nNd1Hjx4BADvSejwetr4AwJ/B7u4uyuUy8vk8Bzrr6+tYXl5GNBpFMpls+U3qdUBkV4PBwMKQWq0W586dQzAYxMbGBlKpFKdkyeRNqVSiWCwiGo0iFou1fAyUdXr8+DHGx8eRz+dht9sxODgIqVQKpVLJFgmpVAqVSoU7WIaHhzE4OAiTyYR4PM7loFaXAJ4HsvVYX1+HVqtljRYqFZN1gM1m43JWtVqFwWDAmTNncOrUKRiNRkgkEpaE2NvbaynBXNjlkkql2DSRav8HszVUvhF2f9DvQpAcRjgcxtLSUks5SS8CtZyTFQ/Nl3Dd0djpkkGu7sPDw7h06RJGRkbQ1dXFDQQPHz5EIpFoWWBA2cL+/n6MjY3BbDbDaDTygeh0OrnUQZIfdEhSI8Tu7i7u3r27r9GgHUDWOpVKBUajkbNSNAaVSsXv4rlz59Df3w+TyYR8Po9AIICFhQXWJjuK8lWz9ifaC0klWSqVMhcnn8+jUqnAYrEgm80il8vBYrEgn88jlUphfn4ey8vLCAaDCIfD7VfGooxGsVjkiJpuWD6fD5lMhm9VwDMCq7C9k1ReKZJrp4ODROempqY4sHv48CGLr2k0GpjNZm7lpdoxHYbk/EqHhjAgPE4Q1pbpJVYoFHj77bfZxG5+fh4WiwVOpxNXr15FX18fdDoddnd3sbi42Dbu59lsFjMzM/j5z3+O0dFR3Lx5k710fvCDH2Bvbw/Ly8tMIHc6nfD5fLh58yb7uN25cwdTU1NYWlpqq/mkgC6fz2NlZQUffPAB4vE4bt26BZfLxfYJ7777Lm7cuMG3TxJqo7o6GbmWSiXOemxsbGBnZ6fl45NKpQgEAtBoNPB4POjt7WWvPSGEJqbUDXrw39TrdT44V1ZWMDMz03ZzCgBnzpxhi49Hjx6xjhmNkbI4VApyOp24dOkSJiYmcOvWLZjNZqhUKpRKJUxNTeHu3bv44IMPmi5SJ4RUKkVfXx/+5V/+BZcuXYJer98nfkhK/LlcDslkEvfu3cPS0hJWVlbY64s0sGKxWEuzjkJQdSOfz0OlUuHUqVM4f/48Pv/8c6yurmJubg4ejwcejwcjIyM4c+YMd909efIEH374IT755BNWLT/KclwzSPdCDiC1/S8uLiIUCrGnIDUmqVQqFItF/pVKpThj2wi8UbBTq9VQKBQwNTUFpVKJ0dFRGI1GjuiMRuO+dklgv9gS+Z8kEgnEYjH2d2kFnpfqO9gpRcKI2WyWa64kgkVpPBLJEoqbtSvp8bAgifPZ2VnMzs4yidVqtbLxXjQa5exIT08PNBoNyuUyZmZm2EeqlaADjDgZjx8/RiAQwMzMDHw+H9xuNxNY+/r64HK5IJfLWc+DXtqVlRX84Q9/wNzcXMsP/4Og9UWlmc8++wzr6+tYWlrCqVOn4HK54PV64XK52ItISBynzAfdpGm8H330ERYWFvZxs5oNer8ikQjm5+fZMPjq1assMy8sjz9PE4g2eyLWLy8vY2ZmBouLi/jDH/6A5eXllypqtwoajQYDAwM4f/48vvGNb7AwK0lg5HI5dtI2GAwYGBiAw+FgjmSlUsH29jb+9Kc/4eOPP8bi4iJz1FoBkUiE4eFhTExMYHh4mLNz5DyfSqUQiUQwOzuLnZ0dblsmgn0qleLOyYO6Su0A6kb97LPPUC6X8Z3vfAfvvPMO3nrrLRSLRe42UyqV8Pv9WFlZwcLCAhYWFrC5ucmee0cddDfzMxImNsj/KxqN/hkXUigP8aJu2KPCGxOUi8UidnZ2EAgEsL6+Dq/Xy0JyVB8/GFESz4cM7+jWmE6nW5ZWfZHGysH/pxeM0ubkEUV/Lpy4kwLKFoRCIczOzmJvbw/hcBjnzp2DTqfj4AAAq5rm83kkk0l+edvpEKnVaixHTlYl9Izk10OdP8KbJmWo/H4/Z+7aFdVqlXVGqDsuFAohGo0iFApx27YwKCBOAV1i5ufn4ff7MT8/z63crUS1WkU2m8XOzg4qlQq3sFLJ0Waz8ZgO7jnUbZbP57l7ZGpqCpOTk2wqSSJ27QbaS7RaLdxuN0QiEcrlMhKJBJcIyOpCo9HA6XQydzAejyMcDmN7exuPHz/mzEgrAwR6p0qlElZWVrhbsF6vszXC7u4uHj9+zJYWJIqZzWZfakXQDiDC+8bGBvR6PYaGhqDVajm7SO705XIZ09PT8Pv9WFxcxMbGBp+B7Ty+VwEFPJTpaSXeOLNDaW61Wo07d+4AAJxOJ2w2G6dWhWqSwNOXNx6PIxgM4vbt23jw4AHf1o5DkCAUwWr1BDYLtMl8/PHH3AKqVquZC0BOyhTI7uzsYGNjA8vLy+xb1E6gg5PchikDsrKyArVazZpJ9LJKpVKsr68jGAxic3Pzz7or2hHkN0e3YYPBAKPRCKPRyIRdhULBm5FCoeCAXSwWIxAIIBgMYnp6ui1KsKQETWNKJBLI5XLo7+9HKpXico/JZOJOLbo9lstlFAoFBINBzM3NsTnq4uIitre3W5q1ehlEIhEymQySySSSySR8Ph8rkwvLdCSBQPtnOp1mqYvp6WmsrKxgdnYWwWDwpWKozQI5sP/85z9nKgAA7sZJJBJYWFh4rp5Xu4PW29raGnK5HIrFInp6eljIMpvNcmcnZa9CoRDy+fyxGudxg+hlH65IJPrST57qxWq1Gk6nEzdu3EB3dzecTie3o6tUKsTjcdbUIQ2BQCCABw8eIBqN7vPLeu3BCEpR9Xr9UAXKw4yxnfC6Y3yZkemrgASwtFotrl27hv7+fn6RKZNH2b5gMIiPPvoIoVCIndEP+bz8rIcZ45vOId00ab0+j8hKt2lh2+Sb/Dzg8OP74mte+wcS10VI5KVfdNsUiUR8UFLplaxRnudifJifSd+rEWOkMalUKpjNZvT29mJ8fBx2ux12u533ErpoUekqGAxiYWEB6+vr2Nra2mf/8iZo5DySPs7w8DB6e3s5aKUgh/gPJKAYj8eRy+XYBHZ1dZWJ929SHjmqd5HODFKsVigUvP6E/mU0L80Kso9yDomXIpVKoVarYbVaIZVKuSGHGgOIuNsoYVKhNMYXl5pXHmO7SWt8GV40j28c7Hzx73hS+/v7YTQaWXSP+DvpdJo3V0rBxmIxbGxsoFAoHHnq7qQGO8Czg+RVF+5RLVqqOft8PlitVphMpn1tvUQ6J+l2crM/LJod7DQbwkDqdTafo3wO2gypFAs8C3behGdGY/ziEGvYGKkBwmw2w+PxQKfTQa/X83oTvCss8BYKhRCPx4+UoNvI/UbYvUNkY7VazXIP9D4Sp5BKIaVSaV/n2pt2uh5lsEOgYA0AUwFacbB+sd4bEpSLxWJoNBouEwsDukaPVciNAVq73zQLDQ122g2NWrjtgtcNdhoJYWuw8KBsxo35OM4h8JexTr8QADyxYySc5MsVcPi1ehzHR3tXI4PydsBJ328IL3oXW2tF3SC8Scr9OKAdn1NIROvgcGjHeTxKnDSi/l8yTvJapb3rpOMvYYwvw4kMdl4Vf+mLoIMOOuiggw5OMsRf/k866KCDDjrooIMOji9eytnpoIMOOuiggw46OO7oZHY66KCDDjrooIMTjU6w00EHHXTQQQcdnGh0gp0OOuiggw466OBEoxPsdNBBBx100EEHJxqdYKeDDjrooIMOOjjR6AQ7HXTQQQcddNDBiUYn2Omggw466KCDDk40OsFOBx100EEHHXRwovFSu4hXdXclt+TnCRU223/qpBvzAZ0xEk76+IDWjvFF7/Rh0Y5jPOr9qB3HeFgcdn6P8l1sRz/Cdp5DoVP8SXsXBT+T/7sRY/yyYOdQP/RlDykWiyGRSKBSqSCRSCAWi5HP51Eul1Eulw/18CcVIpEIEomEP79KpfLSgLEdIBKJIJPJIJfLIZPJADyd83Q6vc9Qr12f/3VBDt4SiQRyuRxi8dOkaLlcRqlUOnGGl1KpFHK5HFKpFFKpFKVSCeVyGcVisdWP9koQi8VQKpWQyWSQSCTIZDKoVqstM39sl3ebnL6Bp58RfSbtCplMxr+q1SrK5TLvl82ey2bMIb13tN/QzywWiygWiw3fbxo9Rjr7hL+E5yDtqUf5DC8Ndg77g16UyZHJZDAajeju7sZ3v/tduN1umEwm/OxnP8PU1BSWlpZO3CFxWEgkEuj1elitVqhUKohEIgQCAWSzWVQqlVY/3nMhEomg1+sxOjqKy5cv48KFCwCAVCqFf/u3f0M0GkUulztxzucikQharRZ2ux0OhwPXr1+H0WhEtVrF559/jidPniASibTtvL0KxGIx5HI5BgYGcPnyZQwMDKCrqwuTk5OYn5/HH/7wh2Mzv0qlEgaDAV/72tcwPj4Oj8eDf//3f8f6+jqi0WhL5quVAQUFOHSAUjArk8mwt7e3L4Bo9HMc9meIxWLIZDKMjo7i/PnzOH36NLa3tzE5OYnl5WVEo1EUi8WmXpwb+fnQ++fz+TAwMACfz4eJiQmIRCKkUil8+OGHePjwIXZ3dxu6fhs1RpFIBLlcDqVSCZfLBZ/PB5fLBYvFAolEgmKxiPn5eczPzyMajSKTyRzZz26Y67lIJILZbIbb7cbo6Cj+P3vf9dzmmV5/0HvvAEEA7L2JEiXLltZx3zq7s5Ny4ZnMZDL5G3Kbq1zmNrlIMmk72WSyu951fl7XlVaWrEKRYidFgiRAFKL3Xn4XyvMapGVVkgC5ODMaWyRFfO/3tqec5zz9/f2Qy+Wo1WoolUqoVCqn5tA8SnA4HKjVami1WgwPD6O3txdarRZ8Ph8fffQRdnZ2EAwGUSqVUKvVWsLb4nA4kEgkUKvVmJ6exvDwMCYmJtDf349kMgmPxwOpVAo+/9iWU9MgFAohk8lw4cIF2O12WK1WTExMoF6vI5VKwe12QygUtsQ8vSy4XC4MBgOsViu+//3vY2BgAEajETweD4FAAH6/v9mP+MzgcrlQqVRwOp148803YTabWZSHx+Odifl6VnC5XPD5fKhUKohEIkilUlitVrhcLigUCojFYly7dg0+nw+BQODYHdBnffd8Ph9SqRQajQbvvfceBgYGYLfboVarkUqlkE6nEYvFjvVZTxJCoRAOhwNdXV2YmZmB0+mE2WyGxWJBOp3G/v4+tFotBALBqVu/ZGSbTCY4nU64XC6Mj49Do9FAJpMBAEKhEDPi6M9R4lhuJwr56/V6dHZ2YmBgAFarFaVSCcFgEOl0GoVCoeVTNkcNsmoNBgMcDgcuXLiAkZER6PV6CIVC+Hw+1Ot15HI55mm1wrvh8XiQy+Xo6OjA9PQ0BgYG0N/fD5PJhFwuh1wuBy6XeybnUiwWQ6fTYWJiAg6HA2azGR0dHUgkEsjlciz8ehbGLRAIYDAY0Nvbi1dffRVmsxlisRjhcBhcLrdpqZ8XAUWVOzs7MTY2hlqthmAwCODspVifBD6fD5FIBIlEAqvVCpVKBY1Gg97eXoyOjkKr1UIsFmN/fx/VahWRSKQlnCwulwuJRAKtVgu73Y6ZmRnY7XYolUrE43EolUrIZDKWTj7t4HK5kMvlcLlcuHDhAl5//XWYzWao1WrweDz4/X7EYjHweLxmP+oLgc/nQywWo7OzE+Pj45iamsLU1BQ4HA4qlQri8Tii0SgKhQJyuRzK5fKRG93HYuzweDzIZDKMj49jYmICly5dgl6vx507d/Dhhx/iwYMHSKVSAFonh33cIEOHrPbx8XG888470Gg0EIlEqNfrePfdd2G32yGVSnHv3j0kk0mUSqWmh77VajV6e3tx+fJlXLlyBRaLBRqNBuFwGL/73e/wwQcfIBQKMQO2FfEi64zD4cBqtWJsbAxXr16FRqMBj8fD+vo61tbWsL29jdnZWXZBnGZwuVwYjUaMjo5iZmYGg4ODqFariMVi+PTTT3Hjxg243e5TsV/J0ert7cX58+eh1Wpx69YtfPrpp/D5fMhmsy89huMk2PJ4vJc2OIhGYDAYYLFYYDQaMTg4iPHxcXR0dKCjowNqtZrxz7xeL0QiEUKhEAKBQFP5lHw+HxKJBJ2dnRgdHcWlS5cwPT0NgUCATCaDpaUlPHz4EH6/v+XX4rOAx+NBJBJheHgYr732Gt544w0MDw9DIBCgVqthb28PKysrWF5exsLCAlKp1KkaN4/HY1met99+GzMzMxgbG0O5XIbH48He3h4WFxextbWFvb09uN1upFKp02PsiMViGI1GWCwWWCwW5PN5RCIRuN1uZDIZtplO+yXxLOByudDpdDAajZiZmcHly5cxNDQEo9HIyIHpdJqRsiqVygECYbPA4XDA5/PR09OD4eFhjI+Pw2KxQCwWI5PJ4IMPPsCNGzewtbWFdDp9ZgjnFJmUSqWw2+0YGhqC2WxGPp9HKBTCjRs3sLKyAp/Px4y80wxKF0xPT+PChQuYmJgAh8PB5uYm1tbWcOPGDcaPaPX9SpwHi8WC7u5uDA4OIhqNYnl5GTdv3sT+/j4KhcJLj+M4L5uXiaBxOByIRCIWwRkcHITD4YDdbseFCxdgNBohk8kgFovB5/PB4XBQrVaRyWSQyWSayrnjcrnQaDTQaDQwmUwYHx9nKXOpVIpwOIytrS188cUX8Hg8jCPYirzPZ3UKKKKj1+sxNTWFkZERdHd3QywWo1aroVAoYH19HcvLy1hZWWG8zlbfh8DXqSuz2QyXy8WMObPZjFKphE8++QRLS0vY3t7G/v4+otEo0uk0u0uOeozHYuzQgSOXy6FQKKBUKlkFVjabPZaBtDIEAgGMRiN6enowMTGBwcFBdHZ2QiAQIJlMIplMYnt7G8FgEMlk8kCEpJkWPEXoOjo64HQ64XQ6IZPJUCwWEQ6HMTs7i62tLcTj8Zav5ngekJFH3khPTw/EYjFCoRC8Xi9WVlawtbWFSCSCUqnU7Md9aQiFQiiVSgwODqKnpwdWqxXZbBbb29tYWlrC5uYmwuEwcrncN/5tq0V6yEu2WCywWq0wm80IBoPweDzweDyngkD/ou+TuDlarRY2mw1WqxXnzp1DZ2cnrFYrhoeHIRQKUalUEIlEmNFXr9fh9/sZ2bcZ74eiOTabDRaLhRFzu7u7YbfbUa1WEQwGsbq6CrfbjWg02rKGzvNAJBJBqVTCbDZjaGgITqcTWq0WXC4XuVwO2WwWHo8Hu7u78Hg8jN5wGkBOlN1uR19fH8bGxuB0OlEsFuHxeHDv3j0sLS1hd3cXuVyO2QjHtf6OzdgRiUQHSpSJ4GqxWLC1tXWmLscngYiS4+PjeO211/CjH/0IcrkcABAIBDA7O4sHDx7gZz/7Gfr6+sDhcFi0oNkbWSaTweVyYWRkBENDQ+ju7kapVILb7cadO3cwNzeHUCh0KubyeZ6PpBImJiZw+fJlXL58GdFoFA8ePMDNmzcxNzeHQqFwZqqv1Go1XC4Xrl69iq6uLshkMly7dg3Xrl3D7Owsdnd3n6id1UqaKSKRiHGs+vr6oFKp8Otf/xpbW1vsIm+F53wSXuT5KKKjVCrx2muvYWRkBP39/bh8+TIUCgVEIhHjR4RCIfzrv/4reydyuRw3b95kkZKTfj8cDgcKhQI9PT24cuUKq0QaHh6GRCJhabbPP/8cv/71rw9EwVsVzyrZYjQaWdTj+9//PrRaLePlJJNJ7O3tYX19nTlYp8W54nK5UCqVcDgceP3113H+/HlcvHgR9Xod165dw29+8xssLCwgHA4zSsvTcDjT8bzr9FiMnUqlgmw2i3w+j0qlAoFAALFYDJlMBqVSeWpJVs8LOoCuXLmCK1eu4Pz581AoFEilUvD7/fj5z3+Oubk5eL1eRKNRyGQyljdvBTIon89nh1BHRweEQiHzkldXVxGPx1EsFpv+nEcJmjONRoMf/ehHGB8fh0wmw927d7G9vQ2v13siOhcnAQ6HA5VKhZGREVy+fBk9PT0QiUTIZrMIBoPY399HLBb71vlttXmnS7Orqws2mw1SqRS5XA7RaBTZbPZMzNlhUBn52NgYJicnMTk5if7+fuj1emg0mgOXZz6fx+7uLjY2NrCxsYGFhQUkk0kAQCKRaJoBQeTpV155BZOTk+js7ITL5YJKpWKpnNnZWSwsLGBzcxO5XO7UOxoUPTaZTJiensZ7770HtVrNtMtqtRoymQzC4TACgQDi8Tiy2WyTn/rZYTAY0N3djfPnz+Odd95BR0cHJBIJgsEgAoEAdnZ2WEr5efAyZ86xVWPxeDwmjHSYMd9qh+Rxgc/nQyaTob+/H06nEyaTCVwuF/F4HLu7u5ifn2d6ApVKBYlEAiKRiIVnmxlupxCkwWCAyWSCUqlEvV5HIBDA7u4udnZ2mDF7lkDETqlUir6+PhgMBtRqNWxubjKj9CS0SE4CRJq32WwYGRmBWq1GLpdDMpmE1+tFLBY7VQcsGapqtZqt13A4jL29PSSTyVMR1Xle0D4dGhrC+fPnceHCBVgsFkgkEkY+rlarKBQK8Pl87MwhiYt0Ot20PUz3BKU5hoeH0dXVBZPJBK1Wi3q9jkwmg1AohLm5OWxvbyOVSrV8GvJZQOlWnU6Hjo4O9Pb2QiQSsbuyWCwikUhgf38fwWAQmUzmVJy1xHekEvPBwUE4nU4IBAJEo1HMzs5iY2MD4XAY+Xz+ROfyWIwdoVAIrVYLo9EIpVIJDoeDQqGATCaDZDJ5Zi6Lp0Eul7O8ucPhgFwuR7FYxObmJu7fv8/4LqRKe+3atSY/8deQSqUwm82YnJxkJZ/5fB5fffUVE9Jr5eqrFwV5XKRFIhKJEI1G8dvf/hYPHz5kUbezBKfTicuXL0MqlSIYDMLtduPGjRuMDHlaQGXWMpkMKpUKyWQSDx8+xLVr105dBcuzgHSD9Ho93nrrLQwPD6OzsxNyufxAapGI9R9//DH+8z//E6urq0in002PdJGkxY9//GNWimyxWFh0IxqNYnNzE3fv3sW//Mu/MJX2swAydFwuFzo7O5kjDDyK6kQiEWxtbWFxcRGLi4uP5cu1IqiwY2RkhKWuVCoVtre3cefOHfzN3/wNKzF/Xrzs3B+LsSMWi2Gz2dDd3Q2TyQTgkcpuoVBoeoXRYRwXwZLH46GnpwevvPIKE1RMJBL44osvcO3aNSwuLsLv97dsDtZms6G/vx/T09NQKpXIZrPY2dnBzs4OwuFwSxk6RzmHEokEGo0GZrMZPB4Pi4uLuHXrFjKZTFN5AkdRjtwIKiLo7u6GxWKBVCpFMpnE2toa7t69C4/Hc+KGzstyfzQaDex2O0ZGRqDVapHP51lBxFlbqxwOBxqNBk6nE6Ojozh//jyMRiOkUimARxdmtVpFKBRiUgm/+tWvWDVss9YxzTE5E2+++SbefPNNuFwuGI1GCAQClEolpNNpzM7OYm5uDjdv3myp8+YoQOnW8+fPw+VysVQjpa9++ctf4ubNm8zQabZh+izgcrkwm80sddXT0wOz2YxUKoX19XVcv36dpUufB0fVM+vYIjs6nQ5qtRpSqRT1eh2JRALxeBzJZPJMLdrHgaxbUo9Wq9UoFouIRCJMI2Jvb6+l+S56vR5WqxVWqxV8Ph+5XA4+nw8+nw+JROInMyYbAAAgAElEQVRUbL4XAQmZ2Ww2cLlchMNhbGxsMLmEZmseHdXnC4VCKBQKDA8Pw2w2g8PhYG9vD9vb29je3m7ptiXfBpVKBZ1OB61Wi2q1ilQqhWAweCoI9M8DoVAIqVSKrq4uDA4OYmpqirWdIfHHfD6PTCbDKupWVlbgdrubmrai3nJ0Nvb09GB6ehqdnZ3QarUQCoUsfePz+bCwsIDl5WW43e4zdd4QvcFsNsNut0Oj0bCoTqlUQiqVwurqKivJPg3rl8PhwGQyobu7G1NTU+jt7YXRaASHw8H29jY2Njbw8OFD1hngeUAGfD6fby1jh1oLWCwWqNVqiMViVKtV+Hw+bG9vs0qsVsFRLyIq/9Tr9RgcHMTMzAwUCgX29/exvr6Oe/fuYWtrC+FwuGUXMIfDgd1uh8vlgsViAfCIwPjw4UOsra0hEom01LMf1bOQgKLT6cT4+Di4XC4ikQg2NzdfOPR6VKAD4igMHjLGLRYLvve976G/vx/lchm3bt3CvXv3sLq6ysQsT7K0nD7vRT/TbDYzflk8HsfW1hbm5+dbymijMdL/Py+4XC4UCgWcTifeeOMNTE1NYXp6mqkJ1+t1poS8t7eHGzdu4M6dO1hfX4fX620a34W4HBKJBA6HA1euXMHIyAgTKRUKhajVaohGo3C73bh//z4++ugj7O7uwufztdSd8bIQiUTQarVwuVxwOp3Q6XTse5lMBoFAAMvLy/D5fEfaG+q4QDzHc+fOYXp6Gu+++y66u7tZkOPzzz/HzZs3sbS09Exl842RHOL/1Ot1eL3elyrcOXJjh7wOlUrFNmA2m4XX68X+/j4ymcwTH5Y2BfUbosPhRSzCkwYZehqNBq+99hrGx8dhtVqRyWTYwbu5ucnIkq0IIg12d3fD5XJBLBaz0se5uTkkEomW63x9FBcycXU6OjowMDCAsbExxGIxeDweuN3ulolmHYXhoVQq0dvbiwsXLmB8fBxSqRSJRALr6+uIRqMAgM7OTvZ55XIZhUIBpVIJ+Xz+WPfiy4jpabVaxhF0u93Y2trC7u5uS8xbI15mDh0OB1wuF8bGxvDTn/6UqZlzuVxUKhUUi0Xs7Oxgbm4OS0tLuH37Nmt82swzh4w0rVaL7u5uXL58GcPDw8zQKZVKjBu3uLiIe/fu4eHDh6cywvg0dHV1MZFWnU4HsVgM4JFDQ5HkQCDAxAMp6kPnXCsR7TkcDqRSKRwOB9544w1MTEygu7sbXC4X29vb+PLLL/Hxxx9jd3cX+Xz+sb+DggPUEFSn07EO96TTl8lksLi4iPv37yOVSr3QmjhyY0ehUECj0UCv17M2CLlcDul0+lvZ13TByuVyqFQqSCQSiMVilMtlFtZr7BXVKhN9GOS5aDQa1mVZIBBgd3eXHb6pVKrlRaF4PB5LQwJAMBiE1+tFIBBoevuKx+EonocMbIVCAbVaDY1GA6/Xi3A4zDgOzR73UY1TrVbDbrdjcHAQSqUSxWKRFQ5otVpIpVJ22ABgKZFcLodYLIa9vb2WEucjQ1UmkzF14Gw2i2w2e2a4HrQ+HQ4H+vr6MDIywnpd8fl8NoexWAxLS0tYWlrC6uoqPB4PkskkyuVyU/ubke6KyWRCf38/Ojs7YTQaIRKJUKvVkEqlsLGxgeXlZayvr2N3d/eFU26H1efJSGiFu4O0daxWK2w224EKrEqlwpqbymQy1Ot1KJVKiEQi9nOVSgWpVAq5XK4lUus8Hg8KhQL9/f1MkJT6PLrdbqyursLv9z+WvkKRdLVazZxrk8kEs9kMqVTKeroJBALEYjHUajVsb2+jWCy2hrGj0+lgs9ngcDggEolYK4RMJvNYjgodVGKxGA6HAz09PTCZTOzASiaT2NjYQKlUOrZL56hE0Ug12mw248qVK+jo6AAALC0tYX5+HouLi0+NbDUbZHiSTke9Xsf29jbcbjd8Pt+Z87IIpOCqVCqhVCqhUCiYh9UKFyZ9/stEsWivmc1m9PX1YWpqCiKRCPF4HOFwGCKRCP39/VCpVFAqlZDL5eDxeEgkEkzpe3d3l5G1W4VcT4aAUqmESqWCQqFAoVBgiqxnAQKBACqVCsPDw5icnMTFixeh0WjA5/NRq9WQSCSYU3Xt2jWsra1hd3cXgUCA7Wkul8uiAid5+VO0XqfTsd6ALpcLOp0OXC4XxWIRoVAIt27dwuzsLDweD4LB4Av1suNyuYwb1Ch5QlIezZb04HA4sNlscDqdcDgcEAgEbE8TMTuZTKKjowPVahU8Hg9GoxFqtRoikQj5fB6bm5vw+/3MIGzWGifpCr1ej4sXL7Lm0PV6HWtra3jw4AHm5+cfq6dDqS+SHZiZmWGyAxStFIlEAIByuYxQKASBQIB79+4hk8m8UNn6kRs7nZ2d6OrqgsViAZ/PRzqdht/vZ/LejQuYw+FALpczfshPf/pTdHV1QafTMVJsOp3G7u4u/umf/gkbGxuIRqNHSuw9Cm+HDhPiely+fBk2mw2lUgkejwf//u//jvX1dQQCgaZfmk8Dn8+HXC5nWhe1Wg1+vx9+vx/xePyJC6xRUbcxRy8UCgHgQDfbVokKEChkqtVqIZFIUCqVsLa2xkrNG73FZniIR7FOxWIxNBoNuru70dvbi56eHvZ7+Xw+Xn31VXR0dMBsNkOhULDLIpfLsW7EoVAIfr8fxWIRsVisJVJEVP3pdDqh1+uZvk4ymWyJ5zsKyOVyDA4OYnR0FIODg3C5XBAIBMhms4jFYvj444+xvLyMtbU1LCwsIJfLoVQqgcPhHOiFJZVKwePxWKVaoVBg6ZLj7vdFGkgjIyMHxGVjsRj8fj8jyEej0Wd+FjLgyRjU6/XQarVwOBwwGAxM8Z0uSGrN04yICPWMpCgItYWg1FQ+n4fRaGRaSVKpFDKZDAaDASqVCmKxGKVSCRsbG9je3saNGzfw1VdfsTYfJwmSPaA2SG+88QaMRiMAIBwO486dO7h79y7m5+e/8WxisRgKhQITExMslX7hwgWm9E3visPhoFarMXmF0dFRuFwupFIpxOPx557DIzd2lEolNBoNlEoluFwuyuUyMpkMotHoN6IaPB4PDocD586dw9jYGCYmJqDX6yGVSsHhcBhXQKlUYnp6Gnw+H7Ozs6hUKkcWYTiKi0sgEEAmk6G7uxtDQ0MYHR0Fn8/H2toavvrqK2xvbyMej7d0VIQuc6qkk0gk4PF47CJ8muo1tQghnRO5XA6pVMryr8ViEbu7u00Rk3oayMtQKBSw2+3QarXgcDiIRCLIZrPs+xKJBCKRCEKhENVqFeVyGYlE4kR0o152nQoEAqjVanR1dWFiYgIulwtSqZQ1IjSbzdBqtYz3Qpy5arXKUlq0Jmw2GwKBABKJREvwBxpTkDKZDHw+H8lk8omtDw4b5q2S5vg2NDZW1mq1EAgEbP15PB7cvn0bW1tb8Hq9TH5fIBBAqVTCYrGwRsQqlYr1x4rH4wgGg3jw4AFLrx+XcUjvmVqxNJ4n2WwWiUSCnQ1POydpP1I0T6PRQKVSwWazoaOjAwaDgSkwl8tl1rA3FoshGAwin88jHo8jnU4fy1if9twajQYKhYI1YqU1x+fzodFoIBQKWXGPRCKBXC6HRCIBn89n0R5Sxt7b22N8upNau41pY71ezwoDqCu91+uF1+tlbY8OBzjUajU6OzsZSb2rq4ut6VqthlgsxritXC6X8QqlUilznl9EwuZYODtqtRoqlYrlGDOZDPb39w/0wKCJHxwcxHe+8x1cunQJXV1dBw4e4NEhr9frceXKFQgEgiMnrR0FsZUEokZGRlin3kqlgvn5efzP//wPvF5v03OrTwOXy2VjMZlMTH21VquxRUY/87hx0MGqVqtZMz+9Xg+73Q4ul4tkMgk+n89SC62UXiBDjYwBKpmkXkFEPDcYDNBqtVCpVCyt2tjc9jjxsmtHLBbDZDKx1hCUZq7ValCr1SxlRYcIpanK5TLy+TyUSiVzQux2O7xeLzY3N49iaC+NxsOX+EbJZBLZbPax760x8khjJiJ2K+7Rxn1ptVqh0WjA4XCYnMXDhw9ZR/dMJsPKu+VyORwOB4aGhuBwODA4OMgilzweD+FwGCsrK4jH40xu4LiMV3rnZPA0XlZk7Ozv7z8T+Z3mWqlUoru7G06nE1arFX19fUyvp7OzEzweD9VqFdlsFm63mwlmhkIhbG9vnzilgCI7jcU7jSBnUafToV6vQyAQfCMdBzy6Y202G0wmE7788kuEQiHE4/ETGwtVHCsUChiNRuYo8fl8ZDIZbG5uwuPxMJ5NIzgcDvR6PQYGBvDWW2+ho6ODyQ5UKhUUCgVsb2/jgw8+wPb2NoRCITQaDbq6upiBy+e/mNly5MYORQEaD5BsNsuIngSycL/3ve9hamoKHR0dLBJE/4Y8SqlUigsXLqBWq+HDDz9EOBxumc7pSqUSLpcL58+fxx//8R8z1d3//u//xieffIIHDx6wi7DVOkQ3otG4pH40hUIBPB4PAoGAeSGPA3F8xsfHMT4+jvfffx9yuZxt3mq1ikQigeHhYfzt3/4t4vH4SQ7tqZDL5dDr9ejs7ITNZoNQKEQ4HMbu7i5SqRREIhGmpqaYF6LX6xGNRhEKhaDT6TA/P9/yysoDAwOYnp7GO++8g8HBQaZ/RZ4iHSDlchnpdBqfffYZ5ufnEYlE0NXVhbfeegvO/+t6T1Gt44zqkOdP6/FJEIlEMBgMzOsTCoVIpVLfWv1Ba5rI2BwOB7FYjBmvrZb6ovSTVquFTqeDXC5HrVZjHCpaf9VqFRqNBqOjoyzd9frrr0OtVrPIADkswCMey/DwMHp6evAP//APWFtbw/b29pGPnz5PJBKBz+fD7/dDKpWy9HatVmOVZE/jZPJ4PFitVoyOjmJsbAzvv/8+K2qhi5B4OwBY5LK/v59VsfF4PFy7dg37+/snqnWmUqkwNTV1IKoDfB21bXz2wyTrRpBzRmTgkwbtyVKpxHpgUhYgmUxicXERgUDgsQ0+G7lbpMMnEAhQr9cRiUSwurqKv/7rv2bcQKFQiD/7sz9DV1cXpFIpc7peBEdu7ITDYdY8kLgOjQYQgdQ/u7q6oFKpUK/Xkc1mWfNBn88HsVgMrVaLqakpyGQyJvYWDAbZ5mg2TCYTXC4XBgcHYTAYmMWuVCohk8mYSB15yqTC2wrh/0bQs1AJKPFrSEqADqfDYUmZTAa1Wo2rV69ienoaIyMjTPK9kRTJ5XIxNDSEixcvgs/nY3FxsSWIvwDY8xH7XyaTgcPhQKfTsbTc97//fdhsNlalRpwH4mbV63WEQqFmD+UboHF1dXUxno5YLGbGRKNEPYWgd3d38dlnn7HKB7lczoyicrmMZDL5zJ2KXxTP48hQ76dyuXyABxiNRhGJRFCtVllEQalUwul0orOzE0NDQ2y/JhIJzM3NYWdnB5ubmy3hSBF0Oh0sFgs6OztZ6rFSqcDtdmNlZQVLS0uQy+Xo7e3FyMgIaxthNpthsVggEom+cf4CX1fSdHR0wGKxYH9//1hK9ev1OiqVCvx+P+RyOVZWVmCxWKBQKAAA6XQaqVTqqakYSldaLBb09/djdHQUJpPpG4Yc3T2VSoUZ73SpEgfGbDbDarXC6/WeWPsi4kM2GpyEw7zAp6VpKLpCIpp7e3snloYlJ6lQKLAzk/g1jYbr4Wg3h/Oo8bBGo4FGo2HZgnK5jFwuB7fbjfX1dUZmptQ50QcajcCWSGNFo1EWvqILvbEaAPhaE6O7uxt2ux0ymQzlchmxWAwbGxusMy+FYcfHxxmnwmq1YmNjA/l8vunGDofDgdlshtPpxMDAAOM6VCoVtpmsViu0Wi3S6TQrhyWtEgrZtsKFTxuFjJ18Ps8uOEoPiESibwg+qdVquFwuJnDW3d3N3gFV69CidblcuHjxIiqVCra2tprafqERtD6Jj0MkOUobWCwWvPPOOwcIc9ROore3FysrKyzS02rg8/lQKpXo6upi+62xAoTH47HDKRgMYmFhAQ8ePMD169eRyWQgEokwODjI3lEmk0EsFkMymXyhA+dZ8Tx7glIVxWKREXKdTid8Ph+2trYYmVIqlcJms2FmZgaTk5O4fPkyZDIZG5dKpcJXX32FnZ2dlkppGY1G2O12OBwOdrGUSiUm0ur1emG1WvHaa6/h3XffRU9PD3NODqdAGkGOqEKhYCRYWg9HOXa6HP1+P+r1OlZXVzEzM8O+TwT4p30mpYE6OzvR19eH/v7+xxo6VMJNnM9YLAa73c4iS0qlEnq9HhaLBYFA4MRkJbhc7oG9d5grdjii0/i9x13yXC6XGQ58Pv/EOKEU2SkWixCJRCzS0mhsVSqVb5ztHA4HSqUSWq2WKX7X64/UvgOBAFZWVrCyssIKW4inRuX3FIWj9/G8OBbOjlQqRblcZvXxq6urSCQSzGKr1+uw2+24cOECNBoNU878+7//e1Z6mEql4HA4WHqIrEDS4yFvoFlo9JgHBgbQ398PDoeDTCbDPP73338fP/nJT8DhcFh1CAAm3T43N4dQKPREIuVJgy4GiszQ5SGTyQ4Qw0hM6tKlS/jBD36AH/7wh4zU6vf7sbKywqrPzp07B5vNBrVajeHhYSSTSXz22WdN7dHTCJvNxiIfxWIRPp8P4XAYHR0dOH/+PF555RXw+Xx8+OGH+PLLLxGNRnH58mUMDAzA4XDAbDajq6sLi4uLLTOPBJ1Oh4sXL2JqagpdXV1MwKzxYEokEvD7/fjnf/5nzM/Pw+12IxaLsQqXoaEhpsfjdrvh8XgQDodbhnBfKpUQi8WQTqdRKpUgFAoxODgIn8+HlZUVdHR04O2338abb77JjNTGVA7w6CI9d+4c+Hw+7t27h0gk0hKl9RwOB5OTk6yHkkAgYJd5sVjE4OAg+vv78eqrrzKeHEV/nuVCKBaL8Pv9qFQqLJ15HE5kvf5Ib41aQZBxw+FwmIL33t4ea6PzuPdAEbs//dM/xeDgIGvpQpcvXZxbW1v4u7/7O0bw1Wq1+JM/+ROYTCZm3InFYnA4HIjFYnZxHzeKxSLC4TASiQRLsZIhSPclGTSNVIJUKsXSN4cdzsdxek4C9NyHK2vpzBAIBI99Lh6PB5fLhUuXLkEikSAajcLj8eAf//EfMTs7C5/Px+QxBgYG8Bd/8Rfo6+sD8OicomDBi6Qfj9zYKRQKzMMia5UsampmCDwKXQYCASaERdEcKnGuVqssRNbIAarVagc4Bs0CVS3Z7XaYTCZG8KzVahAKhRCLxZDL5az8U6vVolAooFarwWKxYGRkBEajETdv3oTH40EikWjqeAjkGTX+PZfLHTgMKK+s0WjgcDgwMDAALpeLWCyGWCyGubk5zM3NIRAIMNIyEc1IC4W8zlbgMTVqj8hkMva14eFhGI1GlMtlzM/PY25uDsvLy8jn8zAYDKx6QiaTQaPRNHUMjwNFQ20224EqM9pHtVoN6XQaq6urePDgAe7du4e9vT0kk0kmYDcwMMDSyLFYDL/73e8QjUZPLPT/LCChtcZyc5fLhZGREYTDYQwPD2N0dBQ6nQ6FQoH16EsmkxCLxZDJZLBYLJDL5exdkeHUTFA0TafTQa/Xs5RbtVoFl8uF3W4H8Kh3kNPphFwuZ5c4cFCb6XEgDbSNjQ1mwB7nmInnEQwG2VnI4/GgUqkYr0gkEjGDrhE8Hg8WiwVDQ0OwWq1MGoFSVUSkX15eZiX4ZOhQ+p3S8nSvkKFzUilLMhAOp6saCfP0c/l8nvVQLJVKcDqdTIPmcKqL/u1Jn6XE0zls8NC6PVzBS89LPFwA8Hq9mJ+fx+rqKnP8iTpAZe0SiQSpVIo1z06lUi80ziO3GPL5PCsfJEY5bUI+n8/y6oVCgVm4oVAIOzs7rMkk5QLlcvkBPQZ6sY3hvmYduCKRCHq9HjabDQaD4Rs5xcY8JkVByOgzmUxwOp3g8XjY399HLpdrmQapZOw0GgAUkmzclAKB4ECzUBJ+2t7exq1bt/DgwQOEQiF2QFGzSdKOaCVjp9EzJB0S4JFmlFgsRjwex/3797G+vs5y44FAABqNBtlslokRtsJYGiESiaBQKGA2m2EwGKBUKgE8uuRKpRIKhQK8Xi9WV1dx7949bG5usiijXq9njSapz004HMbs7GzLtTuhNFYsFmMl1GazGb29vUgkEhgdHYXBYECtVkMgEGCidfv7+5BKpSwN0FhBSE0zm4lGKQiqoqI9yOfzYbPZWHUPlSI3Xpj0Ox6Her3OUjxbW1vY29s7diOWClbi8Tjj55CDQUKeUqkUmUzmG5cnpSCHh4cPtFgoFotIpVKs6eutW7ewurqK/f19Vr5NBiKdxY3v6SQ1v+izD0t5NBordGdkMhns7u5ibW2NcY3kcvk3fiedWycd3aGxNDpOjffD48ZI/4aqqur1Ovb397GxsQGv18sUs0lSwGAwwGg0gs/nI5VKYWlpCcFgEOl0ujWMnUgkAp/PB4/Hg56eHqhUKhbup9IyHo8HjUYDu93OQl6k+0CpLio3c7lcBxrclctlRn5q1sXC5XKhUqkwOjqKyclJ2O12Rt4k1ed4PI719XWsr6+zyolKpYLx8XH09vZCr9djZmYGqVQKcrkcbre7JS5KCuk2EsysVitMJhOkUimrIpDL5ZiamoLD4YBYLEYwGMTs7Cxu376NX/ziF8jlcgAeHSaxWAzZbBYAmNVO897M5poEqhrj8/nsD4WUyQj/6KOPWGM+KqEn4TZqkdJKBgDwiOvR1dWF0dFRaDSaA/ssGAxie3sbv/3tb1krkGKxyITZ3nvvPVgsFtYIlhRRV1ZWXrg3zXGBzoZwOMwKHPR6PXp7exkfcHd3F36/H4uLi/B4PAiFQshkMkxSgMfjQa1WQ6/X49VXX2WVaM3ck1RxMzExwSroGi8NtVp9oHrncWj8emPFZblcxvLyMu7fv49bt25hbW0NqVTqWNcwPSfppBEnUCgUwmAwYHp6Gvfv3weHw0EwGDxg2E1MTODtt9/Gu+++C5PJBA6Hwxys+fl5LCws4D/+4z+YMywWizE9PY2uri709fXBaDSyaAIVwGxtbaFQKJxYKl0qlcLlcqGjo4NFghtL8amYo1QqYX19HX6/H7lcDn19fTCbzdDpdAcis6QS3gzuauNzEymZMhoUhSTaQyNvlxqEU+CDAh3kpAgEAthsNkxPT2NycpJF5Xw+Hz755BO43W5GB3leHLmxQ9LzkUiEhVa7urowOTkJ4BEZrVKpQKlUsm6vJNUvEolYr5exsTFW6kvhysYQ9LeVlZ4ESIX29ddfh9FoBI/HQzKZhN/vx/7+PlMDffjwIba2ttgCrtfrWF5exsWLFzE0NITp6WmmydNYCtlMUCSHhKq4XC6L4HR2diKZTEIgEECn0+HChQuw2+2oVCq4ffs2bt26hfn5eabXIRQKWesBOmiKxSIKhUJLpUGIoEiiXWRch8Nh3L9/H3Nzc/B6vd/oB0UpWhJJbLXITkdHB7q6uuB0OplhmUgksLCwgNXVVSwuLmJxcRGVSgVcLhejo6Ow2WywWq3o7++HUqlkjWBv376NxcVFhEKhpqd3DoMO/0wmg3g8jlAoxMpazWYzgsEgdnd38eWXXzI5AdLqoqrD+/fvw2q1MhXYxvRCM4xYDocDhUKBsbExFslo9Jwb0x5P4+Y0Fh9Q5LxYLOLu3buYm5vD5ubmC/eheh7Qc1AlT6FQYE6FRCKB1WrF+Pg4uFwu0uk0i+5rtVq8/vrrGBoaYqJ75Piura3hzp07WFhYQCqVYhkEsViMgYEBDAwMwGq1Qi6XQyAQoFqtMgVeEsU8KRDBmgjVjWg0WCmS5XQ6YTQaYbPZDojt0thzuRyi0egLqQm/LOieyGQyrBUUn8+HUCiE0WiEyWRCKpVCLBYD8LV8Qk9PD9MRymazjONaq9XY/T85OYn+/n7YbDZwOBx4vV643W7E43Fks9kXPn+O3NghElowGESpVIJUKoXBYMDQ0BAKhQIymQySySRr+EVVMHK5HAaDAQqFAhKJBMPDw6xXBpfLZbLm6XSaedTNglarhd1ux9DQEOPlhEIhLC8vY29vD16vF2tra/B4PPD7/SzMx+Vysb+/z9SFL1y4AKvVinQ6DYvFgng83hLGTrVaZY0fa7UalEolzGYzuru7EY1GIRKJWJpAq9WiWq1ic3OT9c8CwEKRdrudaYPU63VGLH8WTY2TAhHtyOOlg4j6Da2srLDQKZW/isVixi84CVHBFwEp7ur1egCPxNt8Ph8WFhawuLiIhYUFBAIBVto8PDwMl8vFInlU9bO5uYmVlRVsbGx8I8XQKqC1FY1GEQwG0dnZCYlEApVKhXA4jGg0io2NDabRRfNdqVTYgRoIBJjmB9BcXSyqguzu7mYX9bOW3R5+ZnK2qC0D8ZGoWSi9k+MeKxmlRLBOp9OMI0cOVH9/PwqFAiKRCOPckBHU2dnJqufoTKX5jkajrGpUKpVCr9fD+X/9p7RaLTMWiduTy+VOtDCE5qyxSrnxe41rjiLnpCis1WqhUCiYpk6xWGTNeZPJJDNUT3qt0j1BwQeKjut0OnR0dDA7gCpY1Wo1a5MBgDX6LpVKLFVnMpkwNDTEOIbVahV7e3usaIkc5RfBkRs7xO6/fv06JiYmYLPZIJPJ8MMf/hDnz5/HW2+9hdnZWbz66qu4evUq47BUq1X88Ic/RD6fh0QiwVtvvQWTycREzKjJXSwWa7qxMzg4iLGxMTgcDvD5fOzu7uJ3v/sdPvjgAwSDQcRiMSQSiW9c5vV6HalUCvv7+wiFQuBwOEzmfHx8HB6Pp+k8AfIANzc3odPpoNPpoFQqMTY2BpPJBJvNxrywvr4+CIVCpNNp7O3tsTEbjUYYDAY4HA68//776O3thU6nQ6lUwsrKCrtki8ViS1ycxWKRlVTT+6fmislkkimtksdI/d+cTif4fD6y2ewLk+aOCxwOB06nEy6XCwqFAvmBln8AACAASURBVMViEVtbW/j444/xm9/8Bvv7+6x8vLe3F9/5zncwMzMDo9EIuVzOegjt7Ozg97//Pe7evcsOrlZEvV7H5uYmO/QdDgcsFgvj/JEX2rjmKDpCzlQikWDl142X00mDDGqtVovJyUkW7X4WHNZaIQ86HA7j888/Z5EtEmglZ+ykQKToxcVFSKVSlorj8XgwGAy4fPkyent7ceHCBVSrVRgMBqYDRG0TGqHX6+FyuVgUmXh3vb29GB0dhdlsZrpZuVwOsViMpehPusVCoVBgKVStVguDwfBYsrFAIEBHRwdL/9DzA4/W497eHrsHA4FAU1oRkVPs9Xqh1Wrh8XgAPIrgOJ1OvP3229Dr9SiVSkgmk9BqtXC5XPjud78Li8UCoVB4oGDJ6XSio6MDPT09eOedd2Cz2SCVSpFOp3Hnzh3cvn2bBVBedM6O3Nip1WpIpVJYXl7GvXv3WM6Ryga1Wi0GBgaYYBuHw2HVOkqlErVajZEFKezo8/lw/fp13L17lzUhbNahS+FluVzOLHKSZydVWT6fD5FIhHK5zFIEdFF2dXXh6tWruHDhAqRSKcu7tpLeTrVaxVdffYVsNgsej8eqcTo7O/G9732P9bchQ1QgEODtt9+Gy+VipGQil01OTkImkzFS7/z8PJaXl5lh1AqgJpfpdJo9E23ajo4OGI1GlEolaDQaGI1GXL16FQMDAzAajahWq9je3m6Z1gkEIrfSRb69vY2VlRUm5kjE1+9+97s4f/48JiYmGI8gnU7jq6++wsLCAra2trC4uIhYLNaS0atGVKtVFtrPZDIoFAqsMMDhcGB6ehqzs7OsZQkZFEajEVeuXMHMzAyUSiXW19dfSMfjqECXHInQUeriSfoiRNImoz2bzSKXyyGTySAYDCIQCODWrVvw+XwsbRCJRE70HKXIaSQSwaeffopYLAaz2Qyz2QyxWAyhUAiz2QyNRsMU9UnHhfR0aPz0bii1lU6nGelVKBSyzEGjcF0ikcD6+joWFxdZocFJjr1QKDCuGEUoKGJHIAOcNGgaI3nVahXJZBLRaBQPHz7ERx99hM3NTcaHPEkQt7NWq8Hj8eD//b//h6tXr8JqtUKv12NsbIyp0geDQdZGyGw2QyQSsSKY/v5+AI8qKu12O2voKxQKkc/nsbS0hHv37mFtbe2lo4/HUr9dKBQQDAaxsrLCdHGoF49MJmP6CDSJJPVNFS1k6VP/oe3tbSwuLmJ9fb0ltFmo3LFSqbDNaLFY0NvbC4VCwXgDqVQKuVyOEVh1Oh2mpqYwPj6Onp4e8Pl8lEoldii1itdcr9exs7PDpL0NBgNMJhMTECRvmA4RkUiE4eFhaDQaJJNJmM1mqNVqJt5FFRNbW1vY3NyE1+tlpaetABLVo7QdVZuR/k4oFAKXy2V9YM6fPw+z2QyBQAC32429vT34/f5mD+MbOKzYTamCvr4+dilcuXIF3d3dMBqNjFwfDodZCe/Ozg6CwWDLpByfBJpH6uZNjoZcLofL5WLl5slkEsVikZ1FHR0dmJiYgN1uZxdKswnYjSkf2ieHozaNP5vP5+H3++H1ehkfhTgTfr+fncfRaLSpyuXU3dvtdjNNo76+Puj1ehiNxm9wNxsJ2Y2g6Bep1lP6mZxP0tKhf1epVFi00u12IxKJnOi4GzkuRMUgY+cwHjdeMi4ikQiCwSB2dnZw7949xtdpBhrTo7Ozs9BoNCgWi6w/JjXnpRSjXq9nxjsRmh0OB+NzGo1G6HQ6FolOpVJYX1+Hx+M5EsP8WIydYrGIUCiEjz/+mCl8vvfee8xLPixdTj2wgIMktlgshu3tbXz44Yf4/PPPsbe31xLKpltbW7BYLEilUlCr1XA4HLDb7Th//jzcbjcWFxdx/fp1bGxsYG9vD+fOncPFixcxMTGBV155hZU3FwoFJJNJ1qCumaTrw/D5fIhEItjY2EA8HsfExAQuXryIrq6uA2X2pCdE1jzw9RxSxcD+/j62trbwq1/9Crdu3UIgEGgZQwcA6/FCzfREIhGUSiXGx8ehVqsxMzODvr4+aLVapnWSTCYRCASwtraGhYUFbGxsNHsYB1Cv1+Hz+bC/vw8Oh4PBwUH09fXhnXfeQblcZvMmEAgQDocRCARw584d1sSP1m6rSCI8CyhKms/nWZSOyppNJhMmJyfR1dXF5B4sFgsjsFILELfbjbt37zY1LUnnXy6Xw+7uLsbHx78htdEol0C8hv/93//Fp59+CrlczlLpsVgMkUjkW5uiNgP1+qPWKolEAhsbG7h06RLGxsbw1ltvwW63s8j541pcEMjZUigUUCgU3xr1aiSvezwe3LhxA7Ozs03pz1cul5HJZOD3+2GxWJDL5Q7oIj0JJBOwuLiI27dvY2lpCVtbW02f03q9jmg0ik8++QSbm5s4d+4c/vzP/xx9fX1MmZvuC+BrfpJAIGA97WiOGlvXRKNRbG5u4pNPPoHX6z2S6NWxKfPVajX4fD6kUins7e1hfn4ePT09GBkZwRtvvMEs78Pt2uni2dzcxBdffIHl5WXcuXOH9cNqhcmNxWIIBALwer2szI7P57PGgpSqIrVHamBGOhKkq3P//n18+eWXWF1dxdraGivXbhWUSiXE43F89NFHbJO98sor6OnpQW9vL8u3A18T72q1Gmv8GYlEcP/+fdy9e5f1PYnFYi1XzUNaT6urq/j8889x7tw5TE1NQSKRoLu7Gw6Hg6XiKCU0NzeH1dVVfPbZZ/B4PC03pnq9ziQQ0uk007qSSqUHNDEymQyuX7+Ojz76CMvLy6xbOCn0Nnu/PQ+KxSKi0Siq1Sr+7d/+jRnoExMTjOz69ttvs2gXESpLpRJu376Nr776Cmtra7h9+3bTtYSq1Sr29/fxy1/+EkKhED09Pejo6IDP50M8HkckEmEyAHQR+Hw+JkJHkWf6b6vNI0UFotEofv/732NpaQnXr1+H3W6H3W7H4OAgpqenoVarWY/Bw0YBGTg0tkYqQK1Ww87ODgKBADY3N1l1JZHumxG5o0KWO3fuoFKpQCqVMopAY4SHokDU7mJzc5ORdCmA0GxZhMOo1Wrw+/2sd97MzAx6enowPDzM2pwQT4fmkSJxNI+U5fB4PMyge/DgwZGl6Z5o7LxsNQKJC+bzeXbwplIpRiajRUyleDwej3En3G73ASG3VulyDoCVcLrdbtYckjrZyuVyKBQKmEymA5c/8DVZ0OPxwOPx4M6dO6z54IsKJR0niKzs9/sZCZfL5SISiSAWi0GhULCFS7oZNF5Sxn7w4AHm5+fh9/tZKLLVxkklyJS+IUVkk8nE0pTVahXRaBThcBgPHjzA7Ows1tfXsbOz01JecyMOVz9SNI6amFJqcXZ2FouLi6zxJ+mgnDZQqD+RSGBzc5ONt1AoQKvVQqVSMU0dHo/H+oGRWOK9e/ews7ODWCzW1FQ5nbv5fB7b29vM2bNarQeMnYcPH7JUuVAoZOnwfD7fkuvxMBorqlKpFKLRKEvF0WWu0+mgUqlYFJKI2tTOhsfjsWpK6jlI5OOVlRX4fD643W7weDzm1FAlaDPGS+1o1Go1LBYLa3BNqvI0h5FIhFWM7ezswOPxYG9vDysrKwiHwy2VBSDk83mEw2EmQBuJRJBOp5FIJKDVaqFWq1mFGVVy0dxVq1VWZRYIBBhtJRKJHFmajvOkTcHn8+tHdTmRTLdIJGIhdFJIJuluqVSKZDLJ8plbW1usA/fzol6vPxPDkMvlssE96zhtNhtMJhP6+vrwgx/8AM7/66JMG/JwXxCa3EKhgJWVFXz66adMCItC6i/yjp91jDwer/5t+f5nBeWRTSYTCzOTAUoHTaMaKR3YxWKRlbq+iLH6LGPkcrn1l12jpOzpcDgwODiIqakp/NEf/RE0Gg2kUimy2Szu3r2LhYUF3LlzB16vlxFBX+azn3edPs9nXbp0CRMTE4yETFUde3t7cLvdWFpawn/9138hEokwFdrjUJR9njEexVnD4XCYthORWw0GAywWC65evcrUd6n5IF0k4XCY6e88L551jBwO54kDbCwvJ8E9IitTM0S6HIiQW61WDwj1HZdTeBJ7kZR3hUIhhoeHoVarGQeEouPUqqaxQWQul8Pe3h7jSiaTSSwvL7N0JPFEniQieFLnqV6vh9lsRk9PD2vw2t3dDZ1OB7fbjbW1Ndy6dYtVXInFYmY4vCzX8TjPG0KjUr5Go8Ho6ChcLhe6u7shlUpZ9ev6+jqrrCYjFXjE4V1fX0cwGEQoFHruZ/i2MT7R2HnaxnweNApiURSH/tACJ1XFRku9kaD3PDiqw+dxILlrmUyG8fFxGAwGphFEImT5fP6APDY159vZ2cHy8jK8Xi+CweBLRayOc4zfhkaPqnHDN7aXOPSML0VufZYxHtX4SNzMYDAwcrJIJGL8Kp/Ph2AwCL/fz7zIl71YjnMOtVotbDYbJicnodVqGeGTpA8CgQCL5jRqDR01mrFOG88YalmjUCjQ0dHBIskUem8UZntRj/+ox/i4kmRyIGiO6HuNXzvOiM5J7UUaK1XSEdGYnOJqtcpUzmUyGSNyE8E+n88z/lZjFRtxQ15mfEcxRhLRJSOOild4PB7S6TSSySTcbjeLsgJfR85fdn5Paoy090QiEXQ6HSvuAMC6IJCILvU2I0Od7k/6+vOi6cbOY373gY1MRhBdmi/boO24J5W8LVLnJKHARmIuGTtUds7j8RibPpFIvLSmTjMukYbfefhZjvoj6PeemLEDPLokidRK/YZIjCyVSiGdTiOTyRwZD+I455DSxR0dHcxoAx6JJWYyGRbNOe7UYrPXKUUKqFcYpbdI+fwoBOaaOcaTwknvxUZnEfi6kIUMUmoLQcYeUSbokjyqiMBjnuulx0hrkjqzU0USGQKUSj3qvXmSY6Q7nQw7kpZpFJZsDG40Eu5fJnLWcsbOcaMZh8/jOE6NRsFRXyjtA/YRzvr4gJcb4+FS1pMuI2+v069x1sd41scHtMfY6vi2MR5bNdYfIh53gZwGomAbZxvk9bbRxllHY3VoG200om3sHBEa8+dttNGKaMYabaYScRt/WDhNa42irZSuaRtnxw/u03/keNFYfXCaFmsjaOFSjvJx3z/qsZ3Wd3VacRzv+yTnsBl7jLh4bbRxHHjcWj4t5yJxdg5X7h4nmvVuWmVOmh7ZOQuRkKdVHB1ThcuR/87jwlmIep32OWwU7zrJz2x2a5c2zi4eV/lJa/xlNeKOG1RdBZzcOdCs99Eq89B0Y4eqmng8HivDbpWX8zw4jc981KAqFzp0zuI7afVD9Gk4zc9+VKAI7Gno93VcOO3rmGQUDlfunKaz5zQ843HjcZIKx4UTjzEfTltxOI9a2lNpdquEvJ4Vpzn99jJ43Jipu/vheTwth8/T8Ic614TnGX+rvKfHPTPJQTQ2I/5Dwmkb8+EULDnIQqHwGymgs3LWEE7bXD0vGptKH/dYn1h63kYbbbTRRhtttHHa0WYPttFGG2200UYbZxptY6eNNtpoo4022jjTaBs7bbTRRhtttNHGmUbb2GmjjTbaaKONNs402sZOG2200UYbbbRxptE2dtpoo4022mijjTONtrHTRhtttNFGG22cabSNnTbaaKONNtpo40zjie0iOBzOsSgOUrfXJ3V6PazA+7yo1+vPJMf4uDGeFin1Fx3jSfaqetl3+SxjbOb4XhYvs05PCs3eiy/6uSeJ07AXXxYvshdPEs1cp6cFrT5GDocDkUjEeocd5Tw+d2TnKCTza7XaUw0d6iJ+0nLZzfi8ZkiCn9RnNnN8J/G5zegmfvizjxvNaqtw1qXyCX8o4zwJ/CF09j7LbWuofRTd/0eJ524EehQeyLf9DjpUuVwu5HI5qtUqKpUKCoXCE42jo8RJd4ZuBk7SizypeWtEM/rjNMMzP6nI3Mt6zC+K0xDtOArQmfOHMt7jQjPfYXtfHB0qlQqAR/bAUTYGb3rX80Y0XlKFQoF9XSwWn5jB09iF9STQjIV71o054OQOvrN+8FC6+ayPs5loGzpHg2Y4VoTTlI5sZdTrdZTLZfYeG/fGy+6TliIok7FTq9VQKpVQqVRQq9VOtBt6Y2fds4yzPL6zPLZmgIyd9kF+9DjLKYk/FLTn7+hQr9dRrVZRrVYfa7y+zLtuemSHy+V+4yBtHDAAlMvlE01jnXUclaXcqjiLYzoJtOp6aNXnOgrQuI46ZN/GyaJZqd7TjG/b1/S1x9kFL4OmGzvPYsScZHjyD2WhPm6cAoEAHA6HseDb3nzr4CQu/Fad61Z9rqPAtx3sbZwe0NydZaP8OHDS7+pppefH+kBUZkapKyImNZKEKczbbK+n0XLn8Xjsa41GAT33YS/ttGwAYsGTcUkEcfr7WT6QD6cTDs/tHwLa3mlz8CLv+nHh/LMwZ41FKt92frbqvmy152llcDgcaDQadsfkcrljf39PNHaO/cP5fHR3d6NYLCKfzyMSiTAjgcvlsj9CoRCFQoFdvM0gotEG5HK5kMlk4PP5bENWKhUUi0XUajUIBAIIhULkcjmWdyQjDmjtDSGRSCCVSiGVSsHlclEsFhGPx5nmQblcRqVSOTCGRqP0tBl3BC6XC4FAAJFIhFqtBi6XCx6Px8ZcLpdZShU4u5VXfP7Xx8G3zfNpm9uz4G2Tc9WY8mosy6XzklL/p3W8HA4HEokEYrEYMpkMpVLpwBlKDgndBYfX6NN+N/1p3MttNAdisRjvvfcekskkwuEwHjx4gFKp9MS7nYIIL4qmpLFkMhn0ej20Wi0kEgkzYCwWC6RSKeRyOex2O5xOJ3Q6HbRaLe7evYvV1VVcu3YN2Wz2xDc1j8eDWq3G4OAgJiYm2LMXCgVkMhnE43EEAoEDG4kuS6/XyzamSCRi1mwikUA+n3+uTXuc4HK5sNls+MlPfgKNRoNcLge3241MJoNkMolgMIjFxUWUy2Xw+XxMTk6iUqkgm80iEokwFr1Go8H+/j7S6XTTI3JPApfLhUgkwvnz59l6E4lEAB7N3fb2NtxuN3Z3d5HJZNg8icXiA4dnLpdDqVRCsVhs8oieH2ScC4VC8Hg8SKVSCAQCFAoFtjbJEKxWq4hEIi17WTTOSWP0lQyeWq3G/k5OFPDIWMjn899KimwmuFwuc6y4XC40Gg3kcjkUCgXi8ThyuRzy+TxzusghpJ/n8/kolUrfMNhbDSKRCBqNBn/5l38Jq9UKqVSKcrmMfD6PTCaD3d1dJBIJJBIJPHz4EMlkkjmU5IwdBs2zSCQ64Dyn0+mmvovHOYgAmAFLGQIejwc+n8/eRaVSQT6fb7k1+jxQKpV45ZVXcOnSJVgsFuzt7cHr9SKXy2Fvbw+JROIAp5TL5bIzic/nI5PJMAf0edEUY6dxsnO5HJtYuVyOzs5O2Gw2jIyMoKurCzqdDmq1mkV/7ty5ww6mk35msVgMh8OB6elpmM1mSKVSFAoF5HI5xONx+Hw+FAoFlMtlNsZqtQqbzcYMAZFIxAyk3d1dxONxlMtliEQi5PN5lMvlA5VoJwl6p2q1GkNDQwAAi8WCbDaLeDwOv9+PQqGAaDSKbDYLPp+Per3OBKDoghEIBJBIJMwgJY+zkXTeCpBKpbBarZiYmEBPTw96e3shlUpZ+aPBYIBWq4VOpztwQEokkgNrOJPJIJ/PI5lMIhQKIZ/Po1QqtayRdxhkAAgEAmg0Gmg0GgiFQjYGuVzOoqurq6sIhUInqn31NPD5fIjFYqjVanYo0hyWSiUIhUJwOBzw+Xzo9XoIBAJm5FWrVRSLRQQCAUQiEWSzWZRKpaaOh8fjQavVQiaTQaFQQKlUshSzXq9nXw+FQshms8jlcqjVamx/lUoltj5zuRyi0ShSqRQymUxTx/UkCAQCaLVaDAwMoLu7G1qtFtVqFfl8Hul0GiqVCrFYDNFolN0ZtVqNrcPDUUi6JFUqFVwuFzgcDkqlEtLpNNxuNwqFwrGfRXQuksFCa5PmqvHS5nK5UCgU4PP57C6kKJdcLkc2m0U6ncbu7i7S6TSKxeKpOV8a0ZghiUQiiMfjSKfTBwwcABAKheDz+RAKhejr64NMJkO1WsXa2hpSqdTpMXYymQy77DkcDoRCIaRSKbq7u9HT04OLFy9iZmYGYrGYpRPUajWUSiVkMhkSicSJP3OtVoNQKGTGjs1mg0gkOhDeJy8kmUyiVCpBLBZDLBazEDNtymAwCL/fjxs3bmB3dxeVSgVOpxMPHz5EOBxGOBxGLBZ7aljvqJFMJuHxeHDjxg0MDw/D6XRiZGSEXRyZTAYOhwM3b97E73//e3z55ZcHDgzazPl8Hmq1GlqtFlwul3mfqVSqKYbqt8FiseDHP/4x3n77bdjtdlitVnZRVioVdHd3s2cvlUrsIGrkS1CIPZPJYGtrCz/72c+wubmJSCRyKgweOnC5XC70ej2sVitGRkYwPT3N9pzJZAIABINB/OIXv8AvfvEL7O3tnfj6fBw4HA7kcjlcLhdeeeUVZiQkEgmEQiH4/X5otVooFApotVr86Ec/glqtZpdPPB7H/v4+vvjiC3z++efY2tpCMBhs2rxRdO3q1asYGhrC+Pg4+vv7oVAomJFGDgR5uRTZoHVJUchsNotbt27hzp072NzcZEZRK0IikaCjo4NFR81mMwQCAZMhqdfrzLEEwLie+/v7LPIFfK2+L5FI0NPTg+npafzVX/0VarUaO9t+/vOfIxwOH6vxx+FwIP3/7H3Zb5tXev7D5eO+76QoipQoUtQuS7Jsx4mXOJNJZiaYtihatBgUHaD3BXrXf6LoVdGrYqaYaaeYTpNfNiSTSex4YsurdslaKImSuO+buJO/i/ScUI6dSDIpUq4ewIiTiNR3vrO9y/M+r0gEoVAIhUKBvr4+qFQqyOVy5PN5pFIpBINBJBIJegcODw9DJpNBJpNheHgYBoMBSqUSXC4XgUAA29vb+K//+i88evQIwWCw5Ub5cZBMJvHRRx/ho48+eub/J8ahRqOh7+If/uEfYDQa4fF48E//9E/Y3NxELpc78u8+MWOHeP0AaDiZw+FArVbD5XKhr68PV69eRU9PD/R6PQ2rk5/r6elBJBKBRCJpuIz0YVC/iZLJJEQiERQKBb0A2Ww2BAIBeDweFAoFPXwIr6eeaC2Xy2E0GmmEgM/nQ6fTIZlMYnd3F7du3cLnn3+OcDiMdDp9YmMkh2g+n0cgEIBUKj3gbTAMg0uXLsFut+PNN99ENptFLpejf8RiMcRiMVQqFVQqFUQiEbhcLrLZLDKZDHw+H/71X/8VGxsbLd+oLBYL5XIZqVSKjo2ME/g6WqDX66m3XCqV6DyTtCt5X6TXm91uB5fLxcOHD/H+++8jHA63fJzPAvGsOBwOlEolzGYzurq6cP36ddhsNhiNRqhUKvpOSLqHy+Xitddew/T0NKLR6LG8q0ah3igYGBjAxYsXYbFYwOfzweFwkM1mkUqlkEwmIRaLIZFIIJFIYDAYqNdIOCIKhYJGf1QqFf7whz9QDt5JCZlyOBxYrVZ0dXXhypUrGB8fh9FohFKphEAgQLFYRCAQQCKRoAaORqOBQqGARqOh46nVashms9Rg12q16O7uxsbGBqanp7G0tIRoNHqAR9gOyGQyWF1dxfT0NACgs7MTcrmcnrmDg4M0Tdfd3Y1MJoNsNguv14tMJoN8Pk8NJQ6HA7FYjGw2S9dyKpVCtVpFNptFoVBo2tpVKpXQ6/V47bXXMD4+DoPBQJ10hmEoF5KkvUmqimEYereRu0QoFNK0OolSxeNxJBIJ5PN5hMPhpoyhESBpSaPRiGq1ing8jkgkciAixefzqUOlVqthsVhgNBrR29sLpVJJI5gmkwnBYBAPHz5EMpk89tw11dgh4WWpVEpDzBwOB4lEAlwuF0KhEDabDU6nE3a7HX19fVCr1QcMGuKtEOIsIVGetOdVq9VouHt1dRXxeBwqlYqG+ImhU+/9E6+/UqmAy+XSPyKRCBwOB93d3eDz+RAIBJDJZCgUClAoFEilUlhaWsL+/j4ymcyJjbVWq9FDdXV1lYZWVSoVHZfRaIRcLofZbEYul0OhUEChUEA2m6WejFwuh1wuh1AoBJfLpR6ZyWTCu+++S7lNrc6bk8uBXOr13A6S3iCRj/39ffD5fPD5fDqPpHKEfFYikWBgYACZTAa3bt1CLBZrKUGWw+FQQ67+UOXz+XTNWiwWdHV1wWq1Ynh4mO6/+vXKMAyAr/cg4dWRMbcCPB4PMpkMdrsdU1NT6O/vx9DQEPWCAaBYLFLeERkrmTtyoZDUNIfDgdlshtPpRD6fx+bmJuURNLvqh6wflUoFp9MJl8uF8fFxdHV1QSQSoVgsIpFIIBaLYWtrixYMVKtVGAwGaDQa6HQ6yGQyutYqlQo9j/R6PQqFAgQCAfb395FKpVCpVBCPx9uKT1cul2lk2WAwwOPxQK1W03mrN87kcjmUSiUAQK/X08griTSSNFAsFqPpoGAwSI3fZkUkuVwudDodenp6MDU1hdHRUWi1WgiFQur0kogc4VHV700SrSI8JGJwk3NHJpPBYDBAKpVSI6gdQaLEHR0dcDgcEAqFSKVSiEajCIVC9NyXSCTQaDRQKpXQaDSw2WwwmUyw2+2QSCTg8/mU0xsIBBAOh18o/dhUY0csFsNkMqGvrw+Dg4NQKBQQi8WYmZmhJOWpqSlIpVIaKSEXZH0pcDuojFYqFcRiMdy6dQs+n49yG0wmE2QyGRQKBfR6Pd2cpLKHRAVISot4K3w+Hzab7UBeVyKR0M/fvXsXyWQSwWDwxMZYq9WQTqcxNzcHAPD5fGAYBmNjY5DJZBAIBFCr1VAqlahWqwcIucRLISk7oVBIo3NCoRBSqRRisRgWi4Vyf/L5fMsOW/LO+Xw+RCIRJR2XSiXk83lks1lEo1Gk02mk02kEg0EasdJqtZBKpRAKhZBIJAdSW06nE6lUCmq1Gnt7ey0Z4yBcdgAAIABJREFUF/ANx4xELkh0Q6lU0vCwSqXCuXPnaIRAKpVS73d/f586GOTd8Hg8egHVG3snOYcsFgtSqRR2ux1/9Vd/hddeew16vR5SqRQ8Ho8+k1gspp+pL1Wu9+rJz/P5fGi1WoyMjEAmk6FYLOKDDz6A1+ule7hZIPujr68P169fx8jICHp6elCtVpFKpeB2u7G9vY3t7W08ePCAPgtxRBQKBdRqNaxWKx27TCZDR0cH9Ho9TCYTOjo66BrY398HwzBYWFhoq5RypVLB/v4+tre3qQOo1+uhUCigUChQLpcPGN8kWqLX679VGUveDTGQGIZBIBDAzs4ONWIbHdlhs9kQCoVwOByYnJzEq6++CqVS+S0DJpfLIRaLHeBakf2jVqupg5zL5ehlr9frIRaLac9IoVBIHZB2AzHe+/r6qOHe3d0NLpeLUqmEhw8f0mg3OUuVSiWEQuGB1NXThPJqtYpEIoFCodBexg6fz4dKpcLly5fR29uLgYEBDAwMUGvt+vXrdJJ5PB4ymQxSqRQCgQC0Wi01HMjBXalU4Pf7EQgETlRNuR61Wg37+/twu90Ih8MQiUSQyWTUeGGxWDCbzTS6IRaLUS6XabiSWLCExyIUCtHf308nur5yQK1WQ6fTQS6Xn/g4y+Uy0uk0FhcXkUgksL+/j3Q6ja6uLlgsFshkMhoxkEgk9HMikeiAQVrvPdcLbhmNRhiNRni93pZGPWQyGYxGI1wuFyXklstlrK+vY3t7G+vr6zRtkE6nUSqVIBQK6aYcGBhAd3c3xsfHD6Ro66t96qNHzUR9ipV4fhKJBGNjY1Cr1dTQIeFisg8JqTcSiSAUCmFjYwNbW1vw+XwoFApwOp3o6enB5cuXweVyaYqzVaKTXC4XarUab731FsbGxnDjxg2aDn7aQaq/YFKpFHZ3d7Gzs4NAIEAvztHRUVgsFpqy02g0qFarGB0dxc2bN+n3nAQIh5HNZmN7exuxWAyBQAAPHjygTkg9T4qcnRwOBzweD93d3QBA0+TkrBkcHITZbKZ0ATabDZvNhng8jp2dHcqBaTXIWopGo1hYWMDe3h50Oh2NKBMDmzhUpHiAzBmHw4FWqwXDMJDL5RgaGqL8pkKhALfbDbfbjUgk0pQUHrnkyT3h8Xiwvb1N000k7UYMHVLBS9YtiZqT6Fs+n4dEIoFKpcKFCxfgdDohl8sPaLm1I5RKJVwuF9555x04HA44HA5IJBJ6DlosFjoGPp9Po+pE/oNEucjPl8tl7O3tYXt7G+FwmFaIHgcNN3Y4HA6kUiksFgsGBwfR29sLl8uFjo4OujnFYjGSySQSiQR8Ph/8fj9isRjlPbBYLOh0OppS2N/fRygUQigUornXVqCe/U9STLFYDADoRiUpAhLZKZfLKJVKUCgU9BBiGAZisRiZTAZ2ux16vR4dHR0H0iL16bCTNgiIdx8MBrGysgKVSkWZ8x0dHTRETkLJ5KAiC7fe6wdA00DRaJTm11utB6LVamE0GmEwGCAQCAAA2WwWW1tbWFxcxNzcHK12IREoUsFjMBio8UBSlCRtub+/j2w2S43ykxgjuQz0ej36+/uh0+mg0WgwMjJCybjEUJNIJFQpm5Dq/X4/NjY24PV6sbm5iWAwSC8NrVZL92GhUKCVeCTEflIgDkJXVxf6+/vhcrloiqDesycGDrlQiI4HuexCoRDK5TJNbZHoFTmbSJrvpCLJhPwfi8WozAObzUY8Hkc0GoXb7QabzUaxWDwgf0Dmg1x8hDdWq9UgFouphAeRRWCz2TCbzTAajUgmk9BoNPD7/W2jQ0QMVVKRms1mkUwmD1TnEEOHxWJBLpdT4x34OnrT0dEBo9F4oLKnVCohHo/Two94PN6UaBa5mBOJBPx+P9WNIXdXOp1GNpulfBsSPSZnPJvNphFv8g5ImsfpdNIKX7JeWsmXex7IneB0OtHb2wuz2QypVErPRpJeJXciMbRJxKreSSSEdL/fD4/Hg729PcoTPO6503Bjp97znZycRE9PD6xWK92MJGcZDAbpxbK+vo5gMEirPgiTvv4gIJYyOaxaAXKYkMWaSqUO/H+v1wvgG/2E+jQc0TKRy+WQSCQQi8VYWlrC1atXMTg4CIPBcCBCUK+10ArjrlgsIhaLUfKx0WiEzWbD4OAgTUm5XC4AoPNB/nv9wgWAXC6HeDyOra0teDweBIPBA51tTxosFgtWqxW9vb0wmUy0TD6RSGB+fh737t3DvXv3kEgkqMFC5pJhGCSTSRgMBhgMBqpDQ8TKotEoLY89KYOOYRhIpVL09vbilVdeQXd3NywWCzo7OyEWi2lFIPDN5UpKetfX1zEzM4PZ2VlsbGwgEAggnU7DZDKhu7v7gIYLSTPEYjE6vpMCSV0MDQ1hdHQUTqcTMpnsgBdIyq6z2Sz29vaoR+j1erGxsQG32001ukiqmUQIZDLZgfdzUoYqMVq8Xi+KxSKEQiFyuRw1UkjVybOiaPXvvz5lSqKqDMNgb28PXC4XYrEYw8PDUKlU1MhfW1trC2OnPiJH1hsAhMPhZ467XoNGLpdTbmhXVxelRZCzNJfLYWdnB16vl/I+mmXs5HI5GhWtT9UTPaR8Po90On0oXTVCnifGUb0uFHGo2g0SiQQmkwnnzp2D3W6HXC4Hi8WiDiOhOpBsAeGNcblcWK1WSg9gGIZyuB49eoTt7W2sra3B7/e/kJHXEGOHbCy5XE5DblNTUxgeHoZcLqcMdDLAubk5zM7OYnZ2llrAfD4fFouFagtwOByq4Lu0tIQ7d+5gfn7+wGZoN3yXUUJSAJlMhnJ2YrEYenp60NHRccAwIgu7XtH2pEEO/XK5jOXlZaytrWF6ehpdXV3QarXQ6/VwOByUGEgqScxmMxwOB4Cv30cul0MgEMDa2hr+8z//EzMzMy01WEmqsL+/H8PDw+ju7gaLxUIkEsHCwgJmZ2exubmJRCLxLdVk8tmenh50dXXBYDDQKECxWEQ0GsXS0hLm5ubgdrubLjJYz6Mxm824cOECzp07h46ODkqeJ1EPcqmSEvlgMAifz4dPP/0Ubrcbu7u7NFrD4/EOXIo8Hg+xWAw7Ozv48ssvT7zKjMViQavVwuFw4M0334TD4YBWq6X7g0RQt7e34ff7sb29jZWVFaytrWFpaQnxeBz5fJ4+M4kSud1u+Hw+WCwWmEwmmvbw+XyURNpsR4Nc5qVSCcFg8FvVfsc568jnK5UKgsEgZmZmAABTU1OQyWQHCLPtUIpODPHDaosRw7ZarSKZTFLnsaOjA0NDQ+jv7webzaacp/feew937tyB3+9v2rolz+T1einPsn4eyFwe9n3XajVqpNpsNnqPptNpJBKJbznarQQ5F69cuYJz585hcnKSyjpEIhEsLi4iHo8jlUpRKY/9/X3s7OxQ7u5Pf/pTCIVCyGQyVKtVhMNhrK2t4aOPPsLa2hoCgcAL3/0vfJsSIqRGo8Hk5CQmJydhs9ngcDio51UsFuki2N3dxaeffoqtrS3s7u4iFovRvLNOp6OeFhFri0aj2NvboxGBdjV0vg/1i52kREgprF6vp+HMekXQ+ohCK5+bVA4UCgXs7u4iHo8jEAjQqiriHcrlchgMBhqKLBaL8Pl88Hq98Hg8lAfUyhAswzCU0GkymahYHrnMfT4fksnkMw8lqVQKrVaL3t5edHd309QjiZSkUinMzc1haWnpxET3iHFMCO5kDdUTa+srenw+Hx4+fEjnZXl5mfII6ltlkMiVVqul3lkoFMLy8jItbT4JEA/XbrdT8i4R2avP62ezWezu7mJ1dRUzMzPY2NigPD+iOkskL8gYCdnx6bGcZGSHgFyWjYwG1r+bRCKBeDxO05ik1L4dQMZ71P1CxBdJ9uDy5cuw2+2U6Lu6uoq5uTnMz88jEok0nZ9U7yA2gtNWz18h9wOJ+LULsRz4xukymUyUa+v1ehEOh7G5uYm5uTnEYjGkUqkDFa7pdBodHR1Qq9WQSqUHghyhUAgejwcrKyvw+XxIpVIvvC9e2NghzHebzYaf/OQnlDOg1+vBMAwNK29sbGBjYwPLy8v43e9+dyDMTzgFnZ2dMJvN0Ol0AL4WIPL7/VSyv511BY4CQibt7OyEw+GgbHXgG14QkXc/KZLr94EcxiRNAxwkxup0Osq3Il4jIQZ6PB6sr6/jyZMnLfckGYaBVqulejLEWwqFQpScm0wmn/m+lUolrP8rtOhyudDV1QU2m01TmpFIBPfu3cPs7OyJRa6IAU0u52QySfle5HIvlUo0erOysoIvvvgCe3t7CAQC3zo0SVSRpMHIXkyn0wgEAlhaWjrREDrJ5w8ODuL8+fOw2Wy0MgX4Jg2USCSwtraGx48f4+bNm9jb23tmqrTeqHiW8m4r0Yw9Ts6TdDqNcDhMDXyxWHwgbd5KHOdMINGErq4uXLp0CaOjo7h8+TLVBSsUCnjw4AHu3buHhYWFE3OyGnm+EaOcnCXE6WhVkc7zQDI7pPghlUrRYoeFhQX6/uuVkkmPSbPZTAuSSPqR6M2tr69jZWWlYVWDxzZ26tMt165dw4ULF3Dt2jWo1WpKJiPlYhsbG/jqq69w//593L1791ueodlsRn9/P/r7+2kbBiIYNTc3h1/96leIRCItv/AbAR6Ph6GhIQwNDeFnP/sZXC4X1Go19chjsRju3btHW0mcdMXLUUCei8/nY2RkhLb4IKrJkUgEs7OzeP/997G6utoWG5TkggmROp/PY3V1FQsLC1Rw7VnqnAzDwOFw4PLly3j11VdhMBggFoupLsuTJ0/w29/+FouLi0gmkycyFhJ1CwaDtF0J0bXq7OyEWq2mxtj6+jqV20+lUs9NGZBKyqmpKSrbn8/naWToeVGvZoHI/pO2HUSYrV4t2O/3Y3Z2Fr/85S+xtbWFaDR6qMORiA+SfjxsNhsKhQKXL1/Gxx9/jFAo1NYtFg4DcgYHAgGEQiEUi0XI5XJcuXIFd+7cob3O2mFvHgVqtRqdnZ3467/+a5w/fx5Wq5X2KoxEIlhdXcW7775LL9rTNj7gm8hOJpNBoVCg7U3a7T4gZ+qnn36K27dvU9kOUiRAIqtPOxYGgwETExO4evUqrFYrGIZBOp3GrVu38OGHH2J+fr6hUeRjGzuEFKbX6zE1NYXJyckDSqSkaSCx0B49eoTNzc1vtXJnGAY6nY6mBUgPDMLCJmmTdlP8PCpIlZVarUZ3dzdcLhesViu1ZsnFlUqlsLW1hUQiQVsUtKJP1mEhl8tpE02VSkV1JTKZDCKRCPx+P63CagdUKhUkk0k8ePAA0WgUEokEq6urWF9fh9frfWY0gMPhoLe3F319fZQvQlRtnzx5gunpaSwuLmJhYYFqQpwUSKqC5Lm5XC6i0ShisRiVAyiXy5RsTqKGzztAiMAliQ7VajVEIhHs7e3B7/e3VPaB5PzrCeG3b9/GkydPMDs7i52dHSqadxhUKpUD1Vf10g9dXV0IhUJwu91tu/cOi3K5TKM71WoVAoEAJpOJitORasN2u0SfBTabDb1ej8HBQdrWhPCtyHxtbm7iyy+/hNvtRiaTObXzR6K2RJqEEPDbbZ5ItNTj8dA0FOlM/6yUG0l79ff306otHo+HXC6HaDSK9fV1uN1ueL3eho712MZOva7H8PAwBgcHIRKJqKR3Pp+Hx+OB2+3G8vIyzdvVLzySjzcYDOju7obBYIBQKKTVCSR9dZw+GO0EMrkikYimrpxOJ22LAYBWu8TjcXg8HiSTScrCb5dU1tMgSpl2ux1dXV20MoIQB0OhEPx+P+3h0w4gOkJ37tyB2+2GTCZDOBzG7u7uM9M6wNcG+cDAAFwuF+x2OxQKBRV9W15ext27d7G0tITNzc0WjOibw6aekxCJROjf69Mz37eGiLYHITZXKhWaP2+FSCIZWzweRygUQiwWg0AgQLVaRSaTwRdffIFHjx5heXn5W6Ty56Get0M0sYBvhCaFQiFMJhMMBgO2trZOjSHwPBA1XsLL4vF40Gq1kMvl9MyuP5fbdaws1tfNaru7uzExMYHz589jYGCAclm2t7dx9+5dzM3N4YMPPmiJk3iUvXYYkHQP6RFJSPPtdh8QlWOC73o+IpUxNDRE20Ox2WxapU1S7vVnWCNwbGOHXM4k1EtKOpeXl7Gzs4Pl5WVqrCwtLT1TH4fL5WJ8fBwXLlzA5OQkxGIxSqUSEokE3G43fv/732NhYeGFB9lKEEVTwvO4fv06hoaGYDabwefzaSk+uWyJVet2u6mX2k6LGviGq6PT6XD16lW88847tGN4uVymFXSPHj3C48ePD9W4lfB8gOYetiSC9oc//IGGiQnJ+FmdhEll0o9//GOMj4/DZrMBAPx+P7a2tigvYHt7u2nP/KI47Pvk8Xjo6+vDa6+9hsnJSYhEImQyGSwuLuLTTz/F4uJik5/02yDG1u9//3ssLy/D5/PhJz/5CTgcDt577z18+OGH8Hq9NFR+GBB5jImJCbhcLpjNZgBflylHIhE8fPgQ2WyWGkTt1FbhOCDGGuFPEt5Hd3c3UqkU5dIdpiS6lRCLxTAajfj7v/97OBwOGAwGAIDH48HGxgZ+/etf4+HDh7REuRVjaeTvVKlU6OrqgtFoBIfDoXIR7TpP9c/0vOdjsVhQKBSwWCx45ZVX0NPTA6lUing8joWFBczNzeHRo0c0tdxIHNvYIczzeDyO//7v/8bdu3dRq9Xg9/uRSCQQDoeRSCSQTCapR3HgF/9v9cjAwADsdjsVgwoEAtja2sLc3BzC4XDTy3ebDYZh0NfXB6fTCafTieHhYeh0OvD5fJRKJSrm9uTJEySTSVp5RkhZ7bioiShib28vJViLRCIqRri8vIyFhQWsrKx8b8NI4mWTDscnFQEiFW/lcpl6IU+/azabjc7OThoy12g0YLFYiEajWFlZoamrw0YU2hlsNpuWm3d0dEAikdAqtc3NTZoGO2mQyA7hXdy6dQvxeBwcDgeLi4uIRqNHFjiUSqU0mqxSqah69u7uLtxuNy1TJn21WimTQNLYLxJdIm0HOBwOtre34fP5sLa2hr29PeRyuQNqvO143pCKX4fDgbGxMaomXCqV6D5cXV2l500r29A8C8SRO+ozkVJsHo+HQqFAqynz+XyTnrR5II6DXC6n6UebzUYdqvn5eczOzmJpaYk6L43GsY0d4gmkUil8/vnn4PF4VDGSpLJI1cPThwVZvERtsbOzE0qlEqVSiaq5rqysIJlMnupLhGEYiEQi9Pb2wm63w2azoauri/Y1yWQylHE+Pz9PLxfSFqOdNiwBCfMrFAo4nU7YbDaqxZJKpRCPx7G6uorV1VW43e4DDPxngUSJiJDUSXot5CJ93nMR7aexsTF0dXVBIBCgVCrRRqnz8/PY2NhoS4Gvo4LF+rrjt16vp/NJlGd3dnaoR9kKEL0moiLs9XrB4XCQSqWQTqePfEZIpVLo9Xp0dnZSfhJJnZNyZZL2aWXFElmDZJ0eZ1+Qaja5XA4ej0e7hN+/f59G5RtRJt0M1J8NKpUKDocDU1NTMJlMKBQKSCaTWFlZwczMDJ48eYLd3d1nOtatBHHmjnOPkabKfD6fKkoHg8FTZ+yQ9CNJDw8ODuLixYvQ6XS0V9ji4iKWlpZoIUUz1uILRXaIMfO8PkfPemAWiwWRSISenh4MDQ3hjTfegE6noz1hHj58iLt37+L+/funblLrwWazYTQaaW8hImBH+AbZbBYrKyv44x//iNnZWSwuLlLy6LPSKe0AFosFjUYDo9GI7u5u/Mmf/AmcTifEYjGq1SpVxCaVd9vb299r6BCLvxVNJZ8HYtA5nU5cuXIFP/rRj2iFYDgcxieffILbt2/TtEo7PPOLgsvlUuK80+kEi8XCzs4OFQA9qQqzZ6Fev4SIkxIcNcXEZrNhMBjQ19dHVV6r1Sp8Ph/m5uZw//59KsJHhEBbdXlyuVwqsnac9hykISppr6HT6eB2u7G1tYWZmRlks9kD3cLbDaTxsMFgwIULF/Dqq6/i4sWLYBgGW1tbWF1dxQcffIDZ2Vns7u625X1BzrijrlMOhwO9Xo+enh5IJBKqkxQMBk9dtoPP50On06Grqws3btzApUuXcO7cObDZbLjdbiwuLuLOnTuYmZnB3t5e087Thkj0HtYrIOkKh8OB0dFRjI+PQ6PRgMvlIpvNYn5+HsvLy9ja2jrx3juNAp/Ppykeq9WKnp4eDAwMwGQyQSKRoFwuIxgMIhAI4NGjR1hcXDzAz2lXfgAJqZMu9mNjY3A4HFAqlahUKvD5fHjy5Anm5uawtrZ2qHLPeo+1XcpfCceqo6MDb731FiYnJ9HR0UG1ap48eYKZmRnaZ6kd5+qoqNfW0Wq1EIvFyGaztFP0SSsmPw9EZPO464TMrcVigdPphEajoSmCra0trK2tYWtr60CLhlatSVKUUN/z6qggZF6HwwG73Q6pVAq/3w+3200rXIlGU7uBzWbTSs+BgQH88Ic/hMViAYfDoUrnS0tLePLkCaLRaFusz6dRr4p/VDAMA4VCQZXCc7kcEolEy0VZjwo2mw21Wg273Y5z587h2rVrMJlMqNVqWFlZwePHjzE7O4vV1dWm0wFOtB8B4enY7Xa4XC709/dDJBLRJpEbGxtUKfk0pq9I/x6dTofR0VHYbDbYbDZ0dnZCoVCAw+HQ7sWbm5uUzH0auElcLhcikYhGq/r7+2mX4WKxCI/Hg83NTWxubsLv939LYuB5qI8QAq0XdiOEZJvNhomJCUqgSyQS2NnZwcrKCjV0Xpb0FYlkGY1GWlEXi8Vo88RWprCeBbKujhoJZLFYUCqVMBqNMJvNNCKZyWRoL61gMNg2Y613Ao5q7JDUgclkgtlshsFgQLVapWnydoyCEBBZE4PBAJvNRu8KojlTzwkkrT3a0WAjOKqxStKX9X0GSauh46RtWwUyjyaTiTr9vb29VMh1bW0NKysrWF1dpXdGU5+nqd/+FFQqFex2O37wgx9QohkAbGxsYHFxEY8ePaK519MGDocDrVYLp9OJc+fO4Z133qEKtISMnM1m8fjxY0xPT2NpaQkPHjxAKBRq+9J6FosFqVSKrq4uTExMYHJyElNTUxAIBMjlcgiHw/joo49oVO5F9FhaGSnhcDiUo/Paa6/h0qVL4PP5KBQKtDno3NwclpeXW5reaCT4fD5kMhlMJhNGR0ehVquRy+WwsLBAe0e1kyd5mIqP54FhGIyOjsLlctEoQSAQwPb2Nu1nRvoatRpEgfy4aVLinBCHy2AwYHFxEX6/v62V6DkcDlUqv3HjBoaHhzE6OgqpVEplTP7f//t/WFtbo4ZOu6K+Cu4oIBVLpI8Z4SdFIpFTUwxBWnnYbDb86Ec/okK6fD4fHo8HT548wUcffYSlpSV4PJ4Dqelm4cSMHblcjq6uLoyNjeHixYvQ6/Xgcrnw+/1YWlrCvXv3MDMzQ1sRnCaIxWKoVCoMDQ1hcnISV65cgcvlgkgkooJlyWQSPp8Pjx49woMHD7C2tnZq8q8k2uFwODA5OYne3l6IxWKUy2VaOXfnzh14vV5Eo9FTaQSQ/mw2mw2jo6N45ZVXIBQKEQ6HsbW1hXfffRcrKyvY2dk5Uplzu4KkJeVyOUZHR3H+/Hm4XC7w+Xyk02maJtja2mr1ox7Ai/C6OBwO1c/RaDS0+pN4mO1YEPEixj9RhJZKpWCxWFhYWEA8Hm+7MRIQkrzVasXExATeeOMNmEwmKJVKzM3N4e7du5iensb8/DySyWTbO4nHBXEu5XI55HI5arXagS7h7Z46J4177XY7Lly4gNdffx0GgwFSqRQzMzO4f/8+7t+/T7X3Tmoem27skLyzQqGgZbwdHR0QCoWUB7G2tob19XUEg8G2zL1+F7hcLpXnHxkZwdDQEJxOJ5RKJc23E+2Zvb09rK+vY2dnhzYoPA0QCoVQKpUwm82wWCxQqVTgcDgIhULY3NzE/Pw8dnZ2EIvFmh6KbBZIjrw+9Uguw6WlJSwuLtLGte1+2BwGRNhLr9fD6XRS/lwikUAkEoHH44Hf70c8Hm/1o1K8SIqT8AWVSiVkMhkEAgEqlQpSqRRCodCprHL5LnA4HPD5fGg0GqrnFYlE2rr4gSjMWywWuFwu2getVCphfn6eNtkNh8O0d+DLCJLGEggE1FkulUptqZ78NMg+02q1tI+g1WqlCsmzs7N4/PgxFhYWaBf6k5rHphs79W3qL126hD/7sz+DRCJBsVhELBbDb37zGzx69Ahut/tUHjYSiQRWqxVXrlzBz372MxgMBsjlcnowk9YJbrebCtB5vd4TCds1CjqdDjab7QDRulQq4e7du/jss89w8+bNtuI61L/7w/68SqXC4OAgrl27hoGBAUgkEoTDYdy/fx+/+93vsLa21rYXxVFBpB9UKhVeeeUVXLt2Da+99hr4fD68Xi8eP36M1dVVBAKBtlmnpFrvuBE10gbDYDBAIpHQnmHJZBLRaJT2jTppHHWtHhYSiQQGgwFjY2NgGAbRaLSt1y7hOzqdTpw/fx5vvPEGlEolEokEtre38e///u/Y3d1FIpFom3MG+LZicqPmkwiekrYt5O/tDNIpQKFQoL+/H+fPn8e1a9cglUoRCASwsrKCX/ziF/B6vUgmkyc+j019eyRvZ7fb8bOf/QwjIyO0Iunhw4e4ffs2pqen4ff7Tx3ZkwjrvfXWWxgbG8Ply5dpxIqE2kmn4V/+8peUA+HxeE5F6gr4xvu/dOkSzp8/jwsXLtAS7EAggN/85jdYWlrC7u5u26l6Ph0F+K5nk0gkNMJBGrNmMhm8//77uHnzJpaWlk5FdeBhUzxisRjd3d0YHR3F3/zN36CzsxNCoRCJRAIrKyuYnp7GyspKW+3J40rkEy/Z6XRibGwM4+PjVHnX5/PB5/MhGAwilUq1hJvUjD3DYrHQ19eHS5cuobOzE4FAANFoFKurq23JcVEoFOjXDuuJAAAgAElEQVTo6MDExATeeecddHd3Q6/XI5FIYH5+Hl9++SXldbRbNOcwcitHAZvNhkAggNPphMlkglgsht/vRyAQaGuuFQAolUp0dnZicnIS77zzDqxWK2QyGUKhEObm5vDFF19gb2+vZfPYVGNHLpfDZDLB4XDQqACLxUIqlcL29jbm5+dpVUu7LeLvAtEKslgstA0E6bpL5OXz+Tzcbjfm5+cxPT2N7e1thEKhU5XmIUz63t5edHd3Q6fT0fnb3d3FxsYGgsEgcrlc2xk6h70YiXaQ2WxGT08PlEolarUagsEg5ubmsL29TTt9t9MYj4t6XgTRfpJIJKhWq9jd3cXOzg6NPLbbnjyOoUPK6s1mM0ZHR2E0GmlbE5/Ph93dXUrCfhnmF/j6wpTJZNBqtRCJRGCz2bRfYbsZ7Gw2G2azmaZSnU4nFAoFSqUSdnZ2qOgqaXHxsoPH40Emk6G7uxtqtZrKsmSz2bZThiYgKUibzYa+vj7K65TJZMjn89jc3KRUlWw22zLHuGnGDovFgsPhwNDQEC5fvgy73Q6RSIRCoUDZ2EtLS1Rx+TShvi/U5OQkbDYbNBoNNXTK5TIikQjeffdd/OIXv0A6nUYulzs1ER0ClUqFixcvYmxsjF6K+XweXq8XMzMzCIfDhy4xbyaeNmwOa+yQ8ty+vj5MTEzg8uXL1JNaWFigpOt2bdvxNA4zXi6XC4fDgQsXLuDNN9+EVCqlqdabN29ieXkZoVCo7S7F44AYO0KhEC6XC6+//jqMRiNtPzE/P48HDx5gbm7uVMzvYcEwDE17EA4Fn8+HRCIBl8t9JvepFeMn++/ixYs4f/48rl69CpVKhVwuh729Pdy8eZMS5U/bHXFcEB2oCxcuoKOjA2w2G9lsFoVC4UBrm3YCh8OBWCzGm2++iYmJCYyPj0MikVC189///ve0pUcrnYqmGDtE+XJgYABjY2M4f/48GIahzRM/+eQTPHjwgPafOW0Hq9lsRnd3NywWC7q6umjPJNJKwO1247PPPsP09DRSqRSNXLXbIv0uED0d0rtMo9GgWCxifn4ejx49wsOHDxGPx9uCUP70e61UKt9r8BAZfZPJhMnJSQwPD8NoNGJzcxMPHz7E559/3pby8y8CsVgMnU6H8fFxDA4Owmw2U+Xr5eVlfP7551hfX39pVKFJV3Mi8KnX68HhcBCPx+H1erG2toZwOHxqCgUOAzabDalUCqlUCpFIRDtlVyoVKBQKqFQq2iW8FV3BgW+MbpVKhXPnzuHy5cu0qCOVSmFzcxPT09P47LPPsLu7C7/f33ZRxmaAZAxUKhXllgFANBql80Xa6rS6vQcpPCICrNevX8eVK1dgsVggk8mQSqWwsbGBO3fu4KuvvqIaVq08Sxtu7JAX0NPTQ3snqdVq5PN5BINBrK2tYWlpCT6fD/v7+6cuPcBms6HVatHZ2QmbzUYbtRHv2O/3Y3V1lUpfE4v8NI2RxWLBYDDAarWit7cXcrmcCkFtbW1Ryfl2lZkHDpIFnwXSDHJwcJDmxzkcDtxuN+3rdRqjjs8DMe4sFgt6enqg1+tpPzOPx4OFhQVabXYaCwWeBT6fD6VSSccrFAoBANlslpa8lkqltl3Dx4VAIIBIJKLNeRmGgVwux8DAANRqNYLBIGZmZlpmQLBYLAiFQuh0OoyNjcFms1Gl4GAwiO3tbdqoNBaLnbqI+HFAopACgQByuZzeK9VqFXw+HwqFgvatI1mCVCrV0ucl6eG+vj5MTU3RHpcsFguhUAg7OzvY2NiAz+dDIpFouWPccGOHx+NBp9PhwoULuHDhAjo7OyGVSqnCLik7i8fjp666hViznZ2d6Ovrw/j4OORyOTgcDsrlMgKBADY2Nmh5XSqVOpVj5HK5sNvtGBkZwcTEBKRSKTKZDAKBABYXF7G4uIi1tbW2NwS+670bjUb09/fjBz/4ASYnJ2mFGUlr7OzsvFQ8DoZhoNFoMDAwgIGBgQMkXZKy8/l8yGazL8WYSWsIIoRpsVjAMAxqtRptqNiuDTBfBOTClMlkVKOFGLlvv/02CoUCNjY2sLq6eqJlv08/o1KphM1mw/Xr19Hb20ujUKurq1hYWMDS0hL8fv//CUMH+MbYkUgk0Gq1tAFoqVSCTqeDw+EAn88HAMRiMUSj0ZYZO+QeFIlENHNz5coVyGQyqva8traGxcVFrKysUFmHVu+zhhs7HR0dtPcVaWJWLBapmN7c3Bwt82z14I8KsiBdLhf6+vpoZVk6nUYsFsPjx49x9+5d3Lp1i/aeOW1jZLPZ4PP5eOWVVzA1NQWFQoFcLoednR3cvXsXDx48wObm5qkiWteDbFSXy4WJiQlMTExAq9Uil8shEAhgd3cXwWAQyWTy1M3dd0GlUkGr1UKn01GdJMIru3PnDmZnZ9uSlHxckGiG2WyG1WqFQqEAABQKBaoPtbKyglQqdaBZ42lHrVajnj9Je6jVaqhUKuh0OhQKBbDZbPp3wjE8CdRzqIiQZW9vLwQCATKZDLa3tzE7O0srV09DBWQjwGazaUNkrVaL7u5uMAxD35fRaDzQwDaRSGB9fR1ut7sldwzDMBCJRFQ4kOghVatVJBIJrK2tUU2kjY2NtnH4G2rssNls6PV6WCwWWCwWCIVCVKtVpNNpKtwVDAab7jE3q79SfWO+fD6PSCRCBQO9Xi/u37+PlZUVhEKhU2noAN+QzVQqFSWvBoNBeDweGpIk1UmnFUSOXaPRQKvVgsfj0YqHUqn00gmWkfJrDoeDQqGAWCyGdDqNdDqNxcVF7O3tIZPJnDpe2XeBw+GAzWYD+FpagKQEkskkAoEA9vb22iK03gxUKhUkk0mEw2GUy2UIBAL6LpLJJNLpNL2ATroXHeHSCQQCKuwYjUYRDoepcGckEqEUh/8rIHcLIZaXSiU6Z6RPVq1Wo6m+Vjpj9VpCRPsqHo8jnU4jGAxicXERHo+nJQUs37WeG2bsECu0o6MDNpsNVquVls1FIhHa6bvZlR5kMzXre9lsNpLJJPx+PxiGQSqVQjAYxPr6Or788ksEg8G2Up09CkgKSyaTQSQSgcvlolAowO12Y2VlBSsrKyeqidRMo1UqlUKpVFKl60qlcoCM3OyD9iQvGVL1Qg6ljY0NlMtlhMPhUx1pfR7IxVGpVJDL5SAQCMDlclEulxEKhWhKvb7B6csyduBrYycUCmFrawv5fB4Mw4DNZqNSqcDn88Hj8RwQ5zvpCh/Sab1YLCIejyMajcLj8WB6ehpbW1uIRCInaoQ26844KogjXSqVsL+/f8C5JpEfDoeDzc1NrK6utsw5IQ1q8/k8stks4vE4tra2sLu7i729PRqZC4VCJxrV+b55bJixU6vV6AImegjkRRCux+7ubtNzsM3KwddP8K9+9SvqOVYqFZTLZZTL5ZblwBuJUqmEaDSK9fV1qtfx8ccf0yqsk8y9NmseSZsSi8VC5y6fz9Oox0kYcyd9uRDhvO3tbXzwwQcoFAr0sDptRQLfh1qthv39fXi9XhQKBaysrIDH44HL5WJmZgYzMzNYWlqifaJeprGTsvqVlRXkcjkMDg7CZDJBJBIhEAjgiy++wOLiIvb392ma6CT3c7VaRalUwuPHj7Gzs4PHjx9TInIwGMT+/v6Jz0mreVvE+KtUKlhcXES5XMbQ0BBUKhVqtRo++eQTrK+vIxKJgMPhYH5+nnLOWgFSxReJRPA///M/+OSTTygXLpfLIZ1O07uwneaxoWmsWq2GtbU1lEolJBIJAMD+/j5V72y2ZDmbzW7qwiUGD5HQbwXBsVny8uQ7K5UK9vf38fnnn2N+fh4CgQDz8/MIBAInZugQb6ZZ0ZVarYb19XVUq1UqiJhOp+H3+7Gzs9P0Fgmt0MrI5/PUYyQHUTO5Gs1cp4cBSTXHYjG8//77ePToERQKBZaWlvDkyRPKT3qZDB2CcrmMRCIBj8eD9957D3K5HAzDIJ1OY2NjA4FAgFZSnvT4Sa/ASCRCneFEIoF8Pt8SQ6eddGuq1SpisRjW1tbw61//GlKpFACwsLCAcDiMbDYLNpuNcDjc0opJci4XCgUEAgFwuVwqcVAul1tS4XiYu5/1nf+TxTrWKuBwOLTygVygJxHxqL8ka7XaoeKSxx1jq3CSY2zVhUX4FZVKBdVq9XvHeJzxkRAxn8+na7RcLp/IOq3fmC/zOiWXaSvH+HRou5GHcP1F2W7zSFLupKcYgPpz41jv4TBjPG3rtB33Yr3yN5vN/pZC/XGUxNt1nTYKXC6XnuPPG+N3GjtnOMMZznCGM5zhDKcd7FY/wBnOcIYznOEMZzhDM3Fm7JzhDGc4wxnOcIaXGmfGzhnOcIYznOEMZ3ipcWbsnOEMZzjDGc5whpcaZ8bOGc5whjOc4QxneKlxZuyc4QxnOMMZznCGlxpnxs4ZznCGM5zhDGd4qXFm7JzhDGc4wxnOcIaXGt/ZLuKkVRQbqdh7HKXIp5uINVpwkc1mQyQSUaXeF+2M3o5qmKSjMRnjiyrWHlW1tZ3k3w+DdpzDRqMd92KjcTaPX6N+fKRrd93n23oeX3SdPqVU3OjHawheZIztOqan8bwxHiuy04wOsWw2G2KxGFwu94W//7ifr5/MRnXCJW0JJBIJjEYjbty4gc7OTjAM0/JOu43+/TweDwqFAlNTU9Dr9eBwOC/0fUd5PjJfT/9pZ7T78zUCRx3j8+buNMznUfH0RfmygLQ7qJ/L+n9vJ9S31TjOZ0lD6Kdbc7QTyBiPeh4/PV/tfK7Wz8Fzf+a4X97oAbNYLPB4vBO9IJ+FRlqv9RtdKBRCqVRiaGgIGo0GXG5De7Ae+nmefrZGzSOLxQLDMBCLxXA4HFAoFOBwOMf6/qM+G/m5Z32uHTcm0LjnatfDBzieofO8f7brGF8U9WNrpIPVKjzP6WhnQ4A833E/97Rh187jPMrPH+W/txqHefdNaQTaKrRzI1AOhwMOhwOBQIBCodCQppPtOEaGYSCRSFAqlVAqlVAoFI7cXbk+bPoyNh+sR7vNYTPwomM8DanJs3n8Gmw2m47v6Uj52Ry2Hi/rXjxMQ9eTDy98D57l1ZEW8oT/8TweyHEnotk5SRK1EggEEIvFKJfLKBaLSCQSDe3C/H3PADQ378rj8SASiaDX6yEUClGr1bC3t0d/d7VaRSaTQalU+l6+0lGNI+JdlcvlFx7HGVqDeu+s3hFox8P1DM8Gh8OhVIRcLkf/+8s0h/Xpkpf1vDlte/Ewz9UWxg6XywWXywWfz6c5UOCbAXC5XKTTaZRKpaYZB82yWFksFvh8PqRSKaRSKTQaDUqlEnK5HEqlEvL5/AsTlY/yLM36PSRqJZPJYDabodFoaJ6YWN35fB4+nw/ZbBaVSuW5z3Jcg/U4efcztBfaNQ3QjqhPpfB4vAOpIrKH6gsFXjSSfBgQQ+BF6QjtjHblHzUaL9sYW27sMAwDk8kEk8mE/v5+KJVKCAQC5PN57O/vI5/PI5vN4t69ewiHw8jn88/9ruNY2fVRh0aDRHQcDgeMRiP0ej1GRkaQz+eRTCYxMzODlZUVRCKR7xxXI9DMSggWiwWVSgWdTgeTyYTXX38dFosFEokEfr8flUoF2WwWGxsbuHPnDnw+HwqFQkONL1Ld1izUR49qtRr951FTdC/y+9vVq2oUarXaiVzIpx1kLTIMA6FQCIlEgr6+PshkMohEIpRKJdRqNZTLZfj9fkQiESQSCUSj0aav12q1SlPXLzOq1er/iTGeljPnhSM7HA6naQMWCASQSCSwWq0YHh6G0+lEX18fBAIBWCwWkskkjX6USiXE43EAwN7eXkOfQyQSoVgsolQqNfR7eTweZDIZVCoVurq6YDAYoNfr4XA4AAC5XA5yuZwe7o0e19PPUi6Xm7I5RSIRxGIxbDYb1Go1dDodFAoFJZsrlUrweDxUq1UolUpkMhnweDw6n40AGV+jLkqxWAypVAqJRAKxWAyhUEjDudVqFQzD0L/n83nE43Fks1mk0+mmXdbkd77I95MIgFarhUKhgEKhQCQSoZFVEjmtVCrY399HPB7H/v7+iR3qjThvBAIBpFIp5HI5RCIRRCIR8vk8eDweuFwuYrEYyuUyyuUyYrEYPWNOCoc1WolhTSIlfD4fcrkcMpkMcrkcarUahUIBcrkcRqMRo6OjUKlUkEqlqFarKJfLKBQK8Hg82Nragt/vx/r6Ora2tpBMJps2p0KhEIVCoalGa73jQdYuKYYg77dUKqFYLDZE/uJpsNns74xMH/W7hEIhRCIRGIaBSCSCUChEqVQ6ELUjTnmlUqHZALKGm5Ed4PP59HecBMg4eTweJBIJgK/PdbVaDT6fj2KxiHA4jFAohGKxSB3No+D7dHaa4lFyuVzIZDLodDqMjo5ibGwMTqcTHR0dKBaL2N/fRyqVglgsBsMwkMvlsFqtiMfjDTcKmlElwGazIZVKYTabYbFY0NnZCaFQCB6PB+DrA1kgEKC7uxvLy8sIh8NNNXaaFY7kcrkQiUTQaDRwuVyQy+WQSqUoFosIhUJgsVgoFAqQSCQQiUQwmUywWq1IJBINfZ5GfVd91ZxGo4HBYIBOp4NWq6V8oGq1So2rUqmEZDKJ7e1thEIhFAqFpnm1jRgjn8+HTCZDV1cXOjo60NHRgWg0Ch6PBz6fT58/n88jGAxie3sbtVoN2Wy2ASNoPshZYTab0dnZCaVSCblcjkKhAIFAAB6PB5/Ph1QqhWQyCRaLhVgshmw221bci/oyWsL1UyqVsNlsMJvNUKvVMBgMyGazUCgUMBgM6O/vh0KhgEQioVGdQqFAI8q7u7vgcrmIRqNIp9OnLirBZrOpYSMUCun7EQqFYBjmAP2hWq0il8shkUigUCigVCq1VcTw6cicVCqFQqGAWCyGXq+HVqtFuVymP8fn8+nYSqUS0uk0crkcUqkU1tfXqbPVyDGeRPqK8LuIsUqczI6ODqpJZzabIRKJkMlksL6+jlwuR3mfR81WfKex04wDgKQ8ent7MTAwgJ///OcwGAyQSCSIRCLY2NjA6uoqPvvsMwwMDKC/vx/j4+OYmJhAoVDA7OxsQw2wdDrdkO8hIAeUy+XCa6+9hgsXLiCdTmNxcRGrq6vw+Xzo7OyEVquFTqejF87CwkJDn6MehUKh4d/JYrEgFothNBrhcDjw85//HAKBAJlMBv/2b/+G9fV1BAIBxGIxGmb/x3/8R3R3dyOZTDaUX9OIFCCHwwGfz0dfXx/0ej2MRiNGRkYwNjaGwcFBGnEEvkm35PN5uN1u3Lx5E/Pz89jf30c4HG7K+37R7+RwOFCr1XC5XLhw4QL6+vrQ398PnU4HkUh0wHsKBoO4f/8+bt++jeXlZayurjZoFN+NFzmsWSwWtFot+vr68Morr+DVV1+FRqOBUqmEQqGgF0Y0GsXDhw8xPz+P2dlZPHnyBD6fr6GRxu/C951bbDYbAoGARhBVKhU6Ozvhcrnwk5/8BOfPn4dUKv0WR+d5MBgM6OvrQzQahVarxdbWFiKRSNOMu0afp8A3TohKpYJGo0FPTw+9JNVq9YF3UK1WqROysLCASCSCZDJ5rKrQ56FYLL7Q53k8HsRiMdRqNXp7e6mTaDabMTU1hfHxcQiFQvrz9c9cLBaRTCYRiUQwNzeH9957D+vr6/B4PA11tOrJ5c0Am82GQqGAWq2GTCYDh8OBVquFyWTC9evXKce1o6MDAoEA4XAYt27dwv7+PrxeL5LJJDKZTOOMnUaDEHV//OMfY2JiAqOjo7BYLCgWi9ja2sK//Mu/YGdnB+FwGD6fj+ab3377bWQyGWSz2bYnTCkUClgsFvzd3/0d+vr6YDAYcOvWLWxsbODevXtgGAZ9fX1wuVwYHx+HRqOhYbvTApKi+4u/+AuMjo7C5XLBbDZjY2MDS0tLuHfvHhKJBPb396lnFQwGsbe3R0vTSdqkHXLCbDabes43btyAw+FAb28vTCYT5HI5xGLxAUOnWq1SD9NsNuPq1aswmUwIh8N48OBBU4ydF4VCoYDRaITT6cRbb70Fo9EIpVJJjQDiZRkMBsjlcggEAgBfRyE3NjbayjN+Gqz/1XgaGBjA+fPncePGDTidTur1c7lcahio1WpMTU3B6XRieHgYd+/exfz8PD777LO2WI+1Wo1epnq9Hj/96U8xMTGBrq4u9Pb2QqFQ0PEcBsRQUKvVGBkZwQ9/+EMolUp8+OGHzRxGwyCXy6HVatHb2wuz2Yyuri5cuXKFzi2Px6MePknfVSoVlEolRCIRhMNhrK2t4Q9/+AO8Xi+CwWBL1jIpVJHJZJRbJZPJ8Oabb9KIHUm/EukOkiojkh4kQq9WqyGRSCCTyVAoFKDT6ZBKpRCNRmmKp10hkUggFAohFotx7tw59Pf3o7u7G1arlVIG1Go1jfYwDEP120iGYHZ2Fjdv3oTb7UahUDj0eE/U2GEYBjKZDGNjYxgZGUFvby+KxSL8fj9WV1fx4MEDhMNhZDIZFAoFsNlsJBIJVCoVxONxpFKptq64YbFYEAqF0Gg0GB4eRkdHB7hcLjweD/b29hAMBsHhcKDRaJDNZunilkqlrX70I4FhGEilUpqC7OnpQSaTQTQaxdbWFiWSk4OnXmOI5NkZhmk6KfuwkEgklE81MjICu90Om81G+Q/5fB7hcJgeQGw2GxKJBHw+HzweD0ajEZVKBYODg1hbW0M6nW6rtAiLxYJer4fVaoXD4YDVaqXeVDabRT6fR7FYpNIBZA1rNBooFIpWP/73gih32+12OncqleqZopb1nCUWiwW/349gMNhWTlStVgOPx4NKpcLExATGxsag0+mgVCqPVeVEKl2VSiVNa50GwjuHw4FcLofFYsHU1BTMZjPMZjONtJIS96eNHRLdyOVyiMfj0Ol0iEQi4PF42N/fbyq37mmwWCx65hOHw2g00st7eHgYFosFRqMRXC4XqVQKu7u7NC1eqVQo/4xIl5Bzx2AwYHBwkEp8LCwsIJFIIJfLteXckvuxs7MTVqsVU1NT6Ovrg81mg9VqBcMwB0r6i8Ui4vE4VCoVBAIBnft8Po8nT55gZ2fnSCnKEzN2SHhWpVLh6tWrMBqNYBgG8/Pz+OKLL3D//n0aiqtUKjQsrdfrUa1WKbmOy+W2rfXK5XIhFouhUqlorjESieA//uM/4PV6USqV6MFDOARk8Z+Gw4eAcAj6+/thMpnA4/EQCASwuLiIx48fU1Ig8Uo0Gg2sVivOnz+P2dlZlEoliMViSn5t9bhtNhtGRkbw+uuv4+rVq5DL5WAYBuVyGYlEAoFAAL/97W8Rj8eRz+ehUCjQ39+Prq4udHd30434l3/5l5idnUUymUQikWj5uIBvdKomJiZw8eJF3LhxAwqFApVKBalUCjMzM9ja2oLP54PRaMTQ0BC6urpQKBSQy+WQz+fbmt/BYrEgk8kwNDSES5cuYXBwEAaD4TsVyskFpFQqKS/wpOQfDgMulwuNRgO73Y433ngDMpmM8lKOCxIVIAT0dj9vyMVoNptx/vx5/O3f/i2t1CXP/bTuGplXwt8hZFe9Xg8ul4sHDx7QizKXy53IuuZwOJBKpXjjjTfgcDjgcrkwMTFBo6ckYkOkOR4+fIibN2/ij3/8IyXTDw0NQa/Xw2AwYHJyEgaDAUqlEgaDAVNTU3A4HOjr68M///M/Y2VlBdvb223lbBGwWCxIpVLcuHEDf/qnf0ordvl8PjVYSVQuHo/TdPr169dhMpnoXNpsNrhcLiwuLiKfz7ePsUMO256eHoyNjeHVV1+FQqGA3+/H7u4ubt26hbW1NVoVIhAIIBQK0dXVhbfffhs2mw3z8/Nwu920ZLldD1+lUgmLxYL+/n5wOBx4PB7Mz88jmUyiXC5DIBDAbrdjYmIC4+PjYLFYiMfjiEajTT14GnmwkYoBlUoFpVKJWq2GSCSCr776CsFgkIbYCdlsbGwMb775JlwuFwQCAT7++GPcv3+fvpNWHrgkpdbX14eRkRGcO3cOYrEYxWIRsVgMH374ITweD7xeL5aWlpDNZlEoFMAwDB4/fgyj0YiJiQmMjIxAq9VCKpXi/PnzYBgGc3NziMViLU+NCAQCaLVaGI1GaDQaiMViJJNJLC8v486dO7h9+zY1aLRaLbhcLqRSKcRiMYDmSDI0GlKpFC6XC3a7HXq9Hmw2G9VqFcVikWo7EcJnT08PvWA4HA5EIhEkEklTK08PC9JDb2BgAIODgxgeHoZQKHxmhKpOLZbyVFKpFDKZDFQqFYRC4QFiK5vNhlwuh81mg9/vP6A4206QSqVQq9Uwm83UeB0dHYVarUatVkMsFsPHH3+M3d1dRKNRargxDAO1Wk3focPhQH9/P2QyGfh8PhwOB8rlMkKhEKLRKMLhcFOJ9yzW1w2RjUYjrl+/jrfffhsWiwV6vR4ymQxcLpeS/0lBzu3bt3H79m1MT09TcjWJVslkMkilUqyurkIsFqOzsxN//ud/DqPRCKlUiv7+fly7dg0ymYwWGJxkleH3gRifdrsdvb296O7uBo/HQyaToZWCJM24urpKKyWr1Sr6+vqg1WqpQZhOp+n7OUqErunGDkl5WK1W9PT0wGq1IhaLYXNzE8vLy1hYWEAwGEQymQTDMDR8TtJA9RdHs0saXwTEE5FIJJDL5WCxWEilUvD5fPQyYRgGBoOBXjqxWAw7Ozvwer2tfvxDg8PhgMfj0eoykuJZXFykxDEOh0OrREZGRijp1+fzYW9vD6FQqOWGDvA1h0yn06G7uxsWiwVqtRqVSgXRaBQejwf37t2j6cdAIED5R2w2G/v7+0gkEjQPz2azYbfbMTIyAgAIh8M0ctVKL4twNkgkkc1mIxQKwe12Y25uDisrKzRtwufz6fhIRcv+/n7Lnv0wIFwIlUoFhUJBiZ3FYhGpVIoSOYvFIr0QyQVI1jJJD5Aqj1aBpHp1Oh1sNhssFgsN7T9Nji+Xy8jn808K3EgAACAASURBVLR6rlAoIBaLIZ1OU4ejnttDnE4iOfD/2buy57av83qw7/tKAMTCnQBJcNFqUZYcO3Ga2G6TzKQznclMp9OHZqYvfehDHvrUl/4JzVOTyTSTtEmbpEk6E1uWFVkWJZISd5Dgho1YiR3EvvRBudcgtVgSF4AuzozHWijyd3F/997vft/5ziHt0+0C0tlptVphsVgo/6q7uxtqtRrZbBaZTAbRaBQPHjyA1+vF/v4+1bxis9m0dCmRSJBMJiGRSNDd3Q2VSkUFXU0mE+RyOeWAngbImtNoNLDZbJSbqlQqAQCRSIQG4uVyGdlsFvF4HDMzM1hbW8Pe3h6VmWg0GojH48jlcrQBhM/nI51OY2JiAmKxmHbh2Ww2JBIJbG5uIpFInErb/euCxWJBIBCgp6cHOp0OfD4fBwcHCAaDCAaDWF9fx9bWFvb29rC1tYV8Pg8Oh4Pe3l4qY1Kv1xGNRhEOhymtoG0IyqRjx2g0wul0ore3F1KpFI8fP8bc3Bzm5+fhcrnorUur1WJgYAADAwO4du0aisUidnZ28D//8z+Ix+MA2leummy6AoGAbriZTAahUIim2oRCIeWClMtlrK2tYWZmBouLiy1++pcH6VoSCoVgsViIRqNYWVnBnTt3kEqlUC6XIRQKYbFYMDQ0hOnpacjlcqRSKdy7dw/xePzUNH9eFWKxmKaVBwYGIJFIkE6n4XK58Mknn+CTTz6hNfDmOjiDwaDCkOVymbbCXrx4EV//+tdhNpsRCoXogmxlsEMyGM0bhtvtxsrKCpaWlhCJRAA8CfzsdjtkMhlkMhklPMZisbYueZD3US6XQyKR0DLHwcEBAoEAXC4XfvWrX6FarUIikaC3txcWi4Xq7vB4PJoeJ5perQDZP0iXjtVqRXd39yEeAwCascrlcohEIlR7pFQqIZlMolAoQKlUUs0SDodz6GeQDhhiwdMuAY9IJILVasVbb70Fp9OJN954A0qlEpVKBblcDoFAALu7u3C73fjoo48Qj8eRz+cPqRmT0iWPx8PS0hLkcjkajQYUCsUhWQK9Xk8D4ZMGEZLVarVwOp0YHR3FtWvXIJPJUKlUsLe3h42NDUSjUUqWJs9CODcHBweH1huRSSDdhHw+H9lsFnfu3IFer6fVEJvNhkKhgGAwiO3tbZTL5bbYZ4EnGWaVSoWLFy/CZDKhVqshEAjQGIB8LpFIBMViEQKBAFqtFm+//Ta6u7shFApRKpWwvLyMx48fY3Nz85XpLKca7BBtlcnJSdy4cQMSiQTVahV3797F9vY2YrEYRCIRZDIZlEolRkdHIRKJwOVysba2BpfLBa/Xi+3t7bbTwmgGOVBIaUcqlYLBYCCbzdLSjkwmg16vh81mA5fLRSKRwMrKCra2thCLxU79+YDjB4pks9Tr9TCbzQAAj8eDBw8eIJ/P0xLItWvX6LySFtBAIIDZ2VkqatYOmR2xWIz+/n709fVRG4+1tTXMzs7i3r17NJtztLxBbtelUgnpdPpQeYvNZkOj0WB8fBybm5uoVqsn0sZ5nLIDUXttLnkQTk69XodYLEZXVxe+/e1vw263g8vlIhaLIZFIIJ1Ot3yeXgSy9ths9lN8FNI1yOPxoNFoYDKZaCcg8HkmhbTet7r5gdzkSYZYKBQe+nsyD0T2YGdnB7u7u9je3kYoFILdbsf4+DgMBgMddzNI4E0aB0hWpPl7twISiQR9fX14//33cfPmTWg0GgiFQioL4PP5EI1GqUBiIBB4yjqo2ToDeLJeDg4OnsreNIsPnnQQTwIdmUwGu92O73znO7Db7TCbzVheXobL5cLt27eRSCSQTCZphpusR3IpftYzkT2nUCigWq0iHA7j3r17GBoaQqPRQG9vL60qyGQySt4+zlhO6rMhl9/x8XE4nU56+f3ss8/w2WefYXV1FbFYDOVyGQwGg4pkjo2N4Tvf+Q5UKhVisRhmZmbw4MEDrK+vIxKJvHI8cKrBDhFoMxgMUKvVlN8RiUSQz+cph0WtVlM9CfIiBINBbG1tIRgMIp/Pt80N5Fl4lnw7AHpzkkqlMBgMsNls6O3tPUTAanXq/GVBxkj8r9RqNbWByGQy0Gq1tE10bGyMLkqPxwOXywWfzwePx0OD1lYfoESsS6lUUvXSarVK30+i1Pm8zYdkFKRSKTQaDdRq9TM7Q07qvX3dQIeoIRPrFcIdI2UtQhrs6uqCzWaDTCajqfN0Oo18Pt/yufoiEFPdTCZD21oJH0epVEKlUkGtVlNxz+YuHkKIbDVfpzkQJdYOiUQCRqPxKU0dUqKq1WrIZrMIh8MIBAIYHByERqOhnUrNwRsZ68HBAXK5XNvc+AGgt7cXIyMjsNvtVMAzGAxieXkZHo+HagOR/56VsTh6GSFVBaFQCCaTSbNhsVgMmUwGxWLxVOablGvMZjPtsuLxePD5fNjY2MDGxgYymQzdN0mA/rLvH9lXiN7M1tbWIc9FcrFpF18rIrBrMBgwODhIuyDJJXhnZwfhcBgHBwc0y9rT04Px8XGMjIxAqVQik8nA6/Xi8ePH8Hg8VMvsVefvVIMdQjYzGAyUGBmPx2mtVavV4hvf+Aa0Wi3EYjElVvl8PgQCAfh8PmQymbYPBkggQF50qVRK/47FYtEWwbGxMTidTvh8PqTTaUSj0TPpdjmJ709u0AKBAHK5HFqtlt5IarUa7HY7bSM0m83Y2dmB3+/H48ePsby8jFAodOpE7FcBudWTlD8AGuwQL6FnBTpkEyGfg8Vigd1uR19fH1VZLhQKNFg4KXGu1/3cyOFJpBuIxYBcLodMJqNig729veju7oZIJEKpVEIgEMD+/v6piMSdJOr1Og4ODuD1ehEKhWj5hrSjs1gsatfS09NDBfmI/Qfx4GuHxgfCk/L7/djY2ACHw6EyAc3cHaLxRJS8E4kE9vf36V5DSpbNhx1RVSbvNzlcm7M7rQCTycTVq1dx8eJFjI6OgsfjIRwOY3l5GR9//DF2dnbg9XqRTCZfaX6ISJ1Go6FZnmg0ivX1dYRCIWQymRMfC+k+FYlE6O3tRVdXFy2lra+vY3l5mZ5px7kEkQxPMBjEwsICWCwWuru7wWaznyq5H2csxwUJylUqFXp6eijHKJ1OIxKJ4N69e3QuCM9Tq9XiwoULuH79Onp6esBms+HxeLCwsIB79+5hc3MTmUzmtao8pxrskAzG9vY2BAIBarUa0uk0tFotdDod/QCIBPbc3BwWFhaws7ODUChEb9btDhJtezwe2slC9D6+9rWvQSKR4MKFCxgdHaVeXMT64rSVKk8ahA1PZPbNZjMdo8FggEKhoAHtzs4OHjx4QDN57RLoAE8WItn8k8kk7Rbo7+9HNBqF3++nxNZ6vU4XLiE1OxwO9PX14dq1a3A4HNBoNCiVStja2sLS0hIWFxexv7/f8vklmZ3Hjx8DAL0FWq1WvPnmmygWixgbG8PVq1chlUqRzWap7tXe3h5SqVRLn/+L0Gg0kEwmMTs7CxaLRVvQu7u7weVyoVar8cEHH1ASslgspuTcWCyGWCxGLRRaXSYnh9ju7i4ajQYt37z11lvo6uqiwSmRuBgdHaUBez6fx9WrV+m4n5XVKZVKWF1dxeLi4qF9tVXrUiAQwGg04p133sHo6Cjkcjn29vbg8/mwsrKC+/fvU7mHlwl0SBm9q6sLly9fhs1mg1QqRTqdxqNHjzA7O4vbt2/Trt7TQKFQoAf51atXodPpKDGeSD4cF+TCVavV8PjxY+Tzedr04vP5sLa21hYVA5JdNRqNGBwcxOTkJMRiMdbW1vDpp5/S84/NZkOn06G3t5eKnprNZrBYLLjdbvziF7/A48ePMT8/f6zkwKkGO/l8nh56hHinUqnw5ptvQq1WQ6PRIB6PIxQK0SjV5/MhmUy2xU3rVVCv15HP5xGJRLC1tYVqtQqxWAyj0YiBgQHq1VMqlahCdDKZbPkL+bIgt8BSqYRcLod0Ok11MMgiZrPZSKVSWFtbg9vthsfjQTweb0utFhJ4u1wujI6OgslkoqurC93d3ZiYmKBfF41GqZUAqYmPjY1hbGwMPT09GBgYoMJ7sVgMGxsblGtWKBRaHqyT8kg0GsXu7i6Wl5cxNTUFsVgMu92OUqkEu92OgYEBMBgMKgxJOjqOK41/2iDZkHA4TDvLRCIRtYkgZoJsNptmPIhRJCH4ksC9HeaKlCiCwSAKhQINti0WCwYGBmjLLrkJW61W8Hg8+jVHAx3gybtObAbcbjfcbneLRvg5GAwGhEIhent7oVAoKMm/mRhfLBaf4uY873uRrLNer6f6V6S1eX9/HzMzM7Tz9zTlIAiBvLlcRi5RsVgM8/PzVCWbNGq86rM0lzQLhQLt6q1UKgiFQgiHwy2XvAA+D8qI0jXJThJyPVmrJBM2NDSE/v5+SKVSuqbv3r2LpaUl7O7uHlsb6VSDHZI6BEDT5mazGdPT0xCJRGg0GvjZz36GpaUl6olFyJ6tnqhXATlQSCDjdrtRrVapd1S1WqWlPFJf39nZQTKZPDfjbCbIZbNZJJNJqv0AgAasgUAA9+/fh9vtbuvMAMnqzM/PY2xsDHw+H0ajkbaI9vb2olarweVyYXd3F5VKherVfPWrX4XdbofRaIRcLke9Xkcul8P29jYeP35MF+dp8QJeFfV6nZY5SMrbYDDAYDDQX3d1daHRaCAUCsHlcmFtbQ3JZLLl2Y4vAtHeIJ0ZxWIRTCYTExMTlLNBsjnkkCBdV+RgiEajbcFjIftIsVik7bV+vx97e3vo6+vDjRs3oNVqqWWERCKBQCCAwWB4ygyz+XuSNvX9/X2sra1hfX29RSN8AnIICoVCDAwMUDuWer2OdDqNZDJJLxhkzl60jki3IdERInYvtVqNlsRu376N3d1dJBKJU12ThHeVSCRoZ5VMJsPIyAjy+Tzu379P9wvCX3yVgIcEDyRobzQaKBQKCIVCtB0/HA631bol7yS5TBDuFJvNhkKhgFarxdjYGIaHh2lGJxwOY3t7G7/73e+wuLiIVCp17PV56jo7pCPr61//OpWqJ/bx+/v7+OlPf0o3G0LYOs8g2Syfz4eenh7qr0Rq63/84x9x7949LC0ttcVB+LJoJmATgjIp7QCA3+/Hhx9+iIcPH9I0ajt6RDUjn89jZ2cH//Vf/4X9/X2qiyGXy6FQKPCP//iPtKOjWq1S5WuJREK7d6rVKra2tvDo0SP8y7/8Cy1dtVvAnsvlwGaz4fP5cPv2bVy+fBk3b96ExWKhfjUk0/Xxxx8jHo+fm6wj8OSQSSQSVNfL4/FQXlVzezIAeviXy2Xs7e3B4/G0BZmTBCbNHUREg2VjYwNLS0swmUxUt4rYrhDO2bNAMpjhcBjz8/NIJBJnvsc2ByvNt30Wi4VyuYxyuUzfta6uLmQyGezt7cHtdtMOsucFBCwWi9IixsbGMD09DavVCpVKhV/84hdYWFjA7Ows0un0mWQ7SNCWzWaxurpKg22hUIiRkRF8//vfx29+8xvs7u7SzPfLNB+QiwqPx4PVaoVarYZKpUKpVIJKpcKVK1fwwx/+kPpjtQMIJ45cktPpNCQSCaLRKNxuNwwGA0ZHRzE2Noa/+Iu/AIvFQqFQwMzMDH7xi19gZWWFCoKexLydarCjUChgNBoxNDQEs9kMhUJBBdmAJwdoT08PgCcLPZ1On+bjnCpIiYd0YP37v/873nvvPbz99tuUlEW8Wkin0nlC8yZFWqwzmQzVFpLJZOByufTFfpn0c6tBbmF+vx/Ly8vo6uqiYmASiQRisRh8Ph8ymQz1ep1uOCTAK5fLVAV8dnYW4XCYlq7aKdABQG+AsViMBqNerxcajQbT09O4cuUK+Hw+1Go1jEYjVlZWWi6I+Kpo7pB7Uest4Wd1dXXBaDQiEAjA6/XSzEqrcbSziGQCgsEgZmdnUalUqN/Zi2wxmv/d1tYWZmdnW5JpPRrokKxELpeDy+XCzs4OuFwuzaoODAxQ9WGS5SFcHr/fj2KxSC9efX19cDgc6OnpgcPhgMlkotIJ6+vrcLvdSKVSZ94BymQy4XK5wGQyodFooNVqweVyYbVa8Wd/9mdIJBJUvDSXyyGfz6NSqSAQCGBvbw8HBwfUCFOpVMJisaCrqwsDAwNUNVkkEiGbzdJAonmM7bD/kHeX8MWy2SwqlQr6+/vx/vvvo1qtHtK9CoVC8Pl8uHXrFtxuN9UpOymcWrDDZDIP+VhotVrw+XxaYySmkOPj47SGGQqF2mKSXhckPV4oFPAf//EfMBqNePPNN2mph/wdIbyeJ5AuELFYfCgTQIzpiL4Qn89veRvvq4CUeNxuN4RCIZxOJ/h8PrUQ4PF4T+mVAJ9nB7a3t3H37l3Mz88jl8u19bhLpRLi8TgODg7g8Xhw//59aLVaSKVSXL58GTweDyaTCYODg5idnUUymWx72YdmkMBUq9VSOQHg8MZPOie5XC5MJhN6e3sRi8WwvLzcFqWs54E0NczNzYHFYkGn00GtVn9hsEO4Dzs7O1haWjqVLqQvAsnsNAc71WoV2WyW6qkR81k2mw29Xg+9Xo/JyUnK4VlYWMDMzAzVtiKyAlevXsUbb7yBoaEh9PX1USqBx+OhirytyFAyGAxsbGygUChAr9djbGwMUqkUIpEIFy9epLSA3d1dpFIpehl+9OgRlX4gIpMWiwUXL16Ew+HAjRs36MUSAG0A+t3vfnfoItb8ubcKzXIKRHG+UqlgaGgIWq0W6XSa+tPlcjm43W4sLy/jzp07lLN1kjiVYId4DhGWvcPhQLVaxerqKlwuF+bm5jA6OoqJiQm89dZbaDQayOfz2NjYeG09kXYA8dwBnpBZ8/k8fD4fBgYGqHW9VquFUqk8V07nRCvBZDLhe9/7HoaGhtDT00N9ZhqNBiQSCVVNnpubeyU32laDCPaR4IUo6pKD8WhgSngVhPOzs7ND5RTaGaRMQrJUHA6HGgsWi0Xw+XyMjY1Bo9FALBbjk08+gcvlgt/vb/WjvxBknoaHhzE2NoYbN25gbGwMEokETCaTvouNRoOWH4ktw/T0NOXVkQ6RdsxmkblbX1+HVCrF4OAg+vr6XmgQymQyIRaLaSZWrVZTxeyzAtH7aW51JyVe4mz9s5/9DH/4wx+g0+koJ0mj0eDmzZuHujwHBgbwzjvvUA6kXq+nXEgejwcGg0ED1x//+MdUfPCsQTI0LBYL29vbuH37Nt577z2IxWJqeyQQCKBUKnHlyhUoFAqo1Wrw+XxcuHCBfk7EKZ3L5VLtHD6fT/crImKoVCrhcDiQSCSoKj8BkSBoVday0WjA6/XSMQ4NDUEul0MqlcLn88Hr9cLv92N2dhZLS0sIBAInws95Fl462HlZpU2BQEANEolIVKVSwcLCAlwuFx4/fgyfzweFQoHh4WEYDAZYLBbEYjEsLCwglUq1jNh53Ei4OaDRaDTgcDhwuVzURZvFYsFkMsFgMNDOnbMe5+so8Wq1WnR3d8PhcMBut0OlUqFYLGJmZgYWiwWDg4OwWCyUyyOVSg/V4c8Sr6oIy2QyYTQa0dPTg5GREQwMDFBhM+JBRIS6SBmLwWCgUCggl8shkUggn8+3TZ38ZUDeU6PRiLfffhtGo5F2jgBPys8TExOo1WpQKpW4c+cObQFuRxBtk76+PtrRwefzadv9p59+inA4jGQySb3QTCYTbQe2WCyYnp5GPB5HvV6Hz+dr9ZCeC5IRIITsF73n5DDU6XSwWq0wm83weDyndpg873mbFbzJnzX/PdFzikajtDtHLpfj4OAAer2etpJLpVJYrVbU63Uqigk8yVgSmZJQKIS9vb1DNj1nCTInzcFFo9HA/Pw8OBwO7Uwl5dZEIkFJun/1V38FiURCM5JkfKFQCDs7O0gkEvQM0el06O/vp11oarUa+XyeZiePftatSCKQDCrxu2wWcC0Wi/D5fFheXobb7cb8/Dyi0Siy2eypvZuvFOx80YfFYDAglUphsVjwxhtvwGKxgM1mI5vNUu2RR48eoVAoUFKySCSCXq+npLJCodCyTfU4wQ5Rou3u7sbw8DCUSiUVk+rq6qKpdZKiDQQCLUkzvooBIClddXV1UZVTk8lESz+PHz9GuVyGUqlET08PxGIxFAoF5SidNV5VAp5sFFarFYODg7Db7dS7jHCPyEbaaDSo8jCfz6dGfkSQrh2zAc8C6YLR6/UYGhrC1NQUmEwmzUyRUmRPTw9VW97c3KSE3rM6JF9lHjkcDhQKBaxWK2w2G4xGI1gsForFIuLxOO7evQu3241QKASz2YzLly/TTBaXy6X2HkTd1u/3t2WWrrkU9DKHF1m/5CA0Go0QiUQ003JWz/xFWYVcLodcLgfg8wNSLBajXq9Tv6e33377UNdWJpOhXaHN2ddkMkk7oV7VO+k4OPq+Hp2fnZ2dQ19LzJSDwSAUCgXMZjO+//3vQ6VS0SAnk8kgmUxifX0djx49gt/vB4PBgMPhwMDAACwWC/VUk8lk1Li3+bMmlImzfp/JGInukcFgoFQWErB7PB5sbGxgbW0NW1tbp851fOlg52W0DrhcLi5cuICpqSlcunQJbDYbgUAAjx8/xm9+8xt6uwKeTP78/DwuX74Mo9GIRqNBCWnZbLYlJZDX3QAIcZeIs01PTyMQCOCzzz7D3Nwc3G43vvWtb9EA0Gq1toyf9CqHMumku379Oux2O8bGxsBkMrG9vY2HDx9ifn4e1WoVMpkMFy9ehFwuh9FohEqlOhWTvS/Cq9xeGAwG1Go1xsfH8ed//ucYHBzE0NAQpFIpSqUSIpEIZmdnsbe3Rzu1zGYzzQI1Gg2qZNuKzeR1wWQyYTKZMDExgZs3b0KlUmFlZQVzc3PY3NzE5cuXMTY2Br1ej76+PkgkEni9Xqp5Qg6l08arfJ4qlQrXrl3DhQsX0N/fD5VKhWq1ikAggEePHuGjjz5COBxGNpvF3t4edbT/zne+g+HhYSpo9s1vfhMqlQpzc3NnelC+CgiPQy6XP2US2hxYkACA6PHU63WMjo5iZmYGwWDwzLg7r7qnkucvl8uUtEsCIHJRSyQS+Pjjj/Ho0SOardPpdBgZGaGCi2fNNXuVd4WU8srlMkqlEng8Hmq1Gi3HkZL6H//4R9y9exd/+MMfkEqlqP/e4uIiHA4Hent7YTQaqfci0V86Gsy2onzF4/HQ3d2N8fFxXLt2DSMjIxgfHweXy0Umk0EsFsOdO3ewsrICv99/JpfFE+Ps8Hg86PV6TE1Nwel0QqfTYXV1FRsbG5ifn6ebDak9BoNBPH78GLFYDEKhEDKZDBaLBX6/H9Fo9NzwPQBQgq7ZbIbJZIJGo8HS0hI1sSsUCpiamsLw8DCVD29XIiQBkfkeHx/HxMQETCYThEIhfD4fVldXMT8/j1QqRX18arXaodb0ZrfldoRGo0Fvby8uXbqEiYkJ6PV6CAQCJBIJRCIReL1efPjhh4hEIkin09BoNLDb7SgWizCZTNRugtTQzwNIcKZUKmEwGGC1WlEul7G1tYVPP/2UCn1ls1m88cYbtGWWpJ9fRIZtFdhsNqRSKd34FQoFGAwGFZPb29tDOBym5EgAlByr1Wpp+7bRaIRarYbBYIDRaEQoFGq5+vVRkEwNj8eDRCJ5yhKCdM/lcjkIhULaBEL4WUQ+oh3XJuFdEQd7o9FItVfGxsZQq9Wwv79PeSm3bt3Cw4cPUavV0Nvbi3K5jIsXL4LJZEIgEBziqrQjCJlYLpfDZrNhYmICjUaDXvZJMPCHP/yBkqwbjQadv0QigVAoRLMlhCdLLl+tArn4ExFMp9MJp9MJo9FIxS8jkQhcLhd1rj+rs/5Edi+SSiOtcRaLBQKBgDrV7uzsHPK4IpPq9XqRSCRo+7JSqTxXhwcBCXa6urqobHcsFqNWBLVajUrSKxSKc9HSKxQKodFoMDQ0RP15SOpxa2sLOzs7KBQKyGQyVKjr6MbariBtuzabDQ6HA2azGUKhEJVKhXaurK2tYWFhAYlEgnpdkRb7arVKSZ/NaqbtDlK2k8lk1CCTeOxsbGwgm82CzWbTllDiMaVSqai5ZLuBSNITYqtAIAAA2v1x1PiRlOMKhQLW1tbQ398Pg8EAs9kMmUwGtVoNrVaL/f39thGFbAbReiKEXAJygSLBDhHaIzd9NptNy3btFuwQPiN51/R6PXp7e2G329Hf3w+LxYJyuUxVwB88eIClpSW43W5wOBxqtEyyIs0+Yu0GUgEhppc2m42W0IlUQCQSwcLCAhYXF7GxsUEvxkdFMZu1vJoDoVaOjQRdVqsVfX196Ovrg8lkgkwmoxmndDqNYDCIXC53pqXxE9m9WCwWFAoFrly5gtHRUWg0GmSzWery6vF4njrcyaTu7OzQzVcgELywu+A0cNzonyiZEgIv6eyYn5+Hz+ej9grEjNFsNtP0ZbuCyWRSL5Nr167BYDAgnU5jY2MDv/71r7G1tQW/3496vY5gMIjV1VW60ZIAqF0Ju+R2fPHiRVy7dg1f+cpXqEmt1+vFRx99hNXVVSwvL2Nzc5MG6KlUCnK5HCqViuqclMtlqh3Rbofis8DlciGVStHd3Q2NRgMul4vNzU0EAgHqPJxIJLC7u4vBwUGIxWJ0dXVRUvqLxOtaBZJNlMlklPxINlRC7GzWfCL7UK1Ww/b2Nra2ttDV1YXx8XEolUrajr67u9t2JqjE0kSlUj3zMCcibvv7++BwONTxmwSEpOVXKBS2aARPg+jQkA6s69evw2g0UpdsUtqJRqOYnZ3F3bt38ctf/pKaXZLDVaFQgM/nt7UQJtl7LBYLDAYD+vr68MEHH1DCMRGA9Hq9+Kd/+iek0+lDgQAJJohYZk9PDwQCASqVyiF7jVZdpHk8HlWff++999Df34/e3l56ASFluFwuh3g8TvXYzmrvPHawQyZQpVLhxo0b1P/J4/FQ+/bnCTqRVkqTyQSxWEzNy87y4DjuzxIIBBCL+m13zgAAIABJREFUxZBKpVAqlWAwGIhEIggEAshkMmAymZBKpVAoFFAoFJSc1a5lOpIGvnTpEi5cuACr1YpcLofNzU3cuXMHbreblq2AJ9ompPuKEAaTyWTbBjtEonxqagqjo6MQiURUd+azzz7D7du36eFPNg3S8tnsnE3Iya/rwHvWIMRWMhY2m416vQ6/3494PE5viaTTh7ikK5VKurEST7d2AmlfLpVKlKdCDkEA1D6CXKDI4cFms9Hd3Q2VSkWJryRDTbI86XS6rTrQ6vU6XV9Hu1ZIGaj5dl+r1WiAymKxIJVKMTIygmQySc1GWwlydpBMztTUFMbHx2lHEik3JpNJ/PjHP8bKygrcbjfNuLFYLBgMBqhUKojFYnC5XFQqFdqF1erxHYVAIIBWq8XVq1cxPDyMyclJ9PX1gc/ng8ViIZPJYGdnh5p7Np8RpGxuNBoxMTGB8fFxaLVa2m5/584drK6uIhgMtmTcTCYTCoUCvb29eOeddzA1NQWpVIpKpYLNzU3KpTKbzdTaI5/Pn6nY44kEO82+JDweD8lkEuFwGPF4/LlCXSSjEgqFKPmK3MDa7SV9EcitQqvVQiKR0FIIGRN5CWQyGUQiEQBQVcl2BOlsIdYepBtne3sbbreblnUIQVcqlUKn0x3amNrVrZ7cjBQKBb1NcTgcFAoFaljr9XppLZkcmmSTIe27PB6P8pWauSDtDHKwCAQCandRrVYRi8WQy+XofJFySD6fpy31YrEYKpUKSqWyxaN4GrVaDZVK5SlCMQnCSVmZkFur1SrtEjGbzdDpdJDL5bTLh3SUtmsplkg6POswJ0EdOfibMz9EW6i7uxtms7kt+CyEWDswMAC73Q6n0wmr1UrVlb1eLxVinZmZgdfrpXxOUjKXSqWQyWRUNZtcQtot2CF6OHa7HZOTk7R0JRKJqFDi7u4u1tfX4XK5Dl36mzvULBYLHA4HhoaGIBKJkEwmEQgEsLKyglAohEwmc+bjJqVEu90Oh8NBAzHSXu7z+SASiaDRaGinJJfLpbIeZ4VjBTtkAyWHvU6no3wct9tNtTmOfvjkRWWxWPD5fAiFQtDr9W1ZY/0iGI1GOBwOvPHGG5DL5djZ2aHGZeVyGVwuF0NDQzAajZDJZFQqnSzYVrQEPu9nks3D6XRiamoKBoMBiUQCv//97ynZnAQ6JAM0ODiI69evo1wuIxwOw+v10p/TbiBt1zabDTqdDlKpFACQTqcRiUTg8/mo+zXhp6hUKnR3d+O9997DtWvXYLfbweFwEAwG4XK5sLW1dcjLqF1BNiSz2Yy+vj6qY0LMWslBT4id2WwWmUyGap3IZDKIxeJWD+MpkGCHdO6QuVOpVJiYmACfz8fW1hai0Sj1ONNqtTCZTLh+/TrNXjKZTBSLRepSfVq8uuOueXJQCIXCp9YYIfiSluTmYI2s14GBAWoKe9raKy/SvCJSHWNjY/ibv/kbDA4OQi6Xo1qt0vLjz3/+c+zv7+Pg4ADLy8uHLhXk7JFKpdTAl8fjIR6PIxgMtl1pWa1W48KFC/j+979PTWqJA7jf78fa2hp++9vfYmVlBVtbW4eyyqSjrru7G1/96leppx2bzcajR48wMzODe/fuIRQKnTk9glBYLl26hH/4h3/AwMAAFAoFotEo1tbW8Ktf/QrZbBa9vb2YnJx8Krg5yzk6VrBDFhdRvmz+M4VCQW9SzSZ8ZHAcDgcymQx/+Zd/CafTCblcTmuO7ZgVeBYYDAZkMhktbwDA7u4ubt++TQMd4rFEOgSIsSQJGs4az/qZ5OZANsO//du/hdlsRrFYxObmJjweDyKRCA4ODg4dimw2G1qtloq4MRgMVKtVRKPRtkr/E5Agk1h2kD+Ty+Xo7e3F9PQ05bKQTJzJZILZbIbNZoNUKgWXy4XP58Pjx4/x4MEDeL3etuZfEZDylVqthkajoRyHvr4+6uFDsqrkYFSr1dDpdBAIBFT2vd1AusfW19exsLAAm81GM5Ld3d1QKpXQaDTY29tDJBJBsViEXC6HUqnE8PAw9Ho9JBIJ5Z8tLi7il7/8JT0wT+N5j4Nml/ZmNWzg81JW837b/HObia7Ex+4096DnfW9SbtTpdPj2t78Nm80GPp+PRCKBf/3Xf4XL5cL29jYikQg9D5qDFwaDQd/PsbExTE1Nob+/H8lkEgsLC1hYWGirQAcA5YIR/SqXy4VHjx7hf//3fxGNRpFKpRCNRg+Vr0jpamhoCJcuXcLIyAhu3rwJpVKJfD6P2dlZ/Nu//RtWVlYQDofPjDpA9n+ZTIahoSE4nU789V//Nbq7uwEAXq8XP/3pT7G8vIzFxUWoVCpYrVZoNBrw+Xzs7+9jdXX1zOUBjl3GIrLf+Xwe0WgUUqkU9XqdlndIOaBZSVMsFlPdkrGxMSiVSsouP2vDtuOA2CjIZDLIZDLUajVKviLlPaFQiK6uLojFYjAYDFpzb0W68XkgXAWVSgWTyYTh4WFUq1UEg0EsLCwgFApR12Dg8xbmkZER2Gw2KBQKKoBF/JTakcfSaDRQqVSQTCaRSqWQzWbB4XDA4/Gg0WgwPDwMmUxGOwNFIhF0Oh10Oh0kEgnlJa2trVHiPTG1PQ9oPgxJhkChUNAsAZPJhEgkglKpRFdXF/0MSOakXcdKbFlcLhcAQCaTQaPR0M4lohsUjUZpWU4ikVCyLoPBoMTQzc1N2mnYLuuzGYRETgjlpMuKBDfPy6iSd59kr1o5NrLfKBQK2O12CIVClMtlBINBPHr0COvr6wgGg4coDeT9ZLPZlEhuNptht9uh1+vB4/Hg8Xjg8/nO3BLjZWA2m2E2m2lgfXBwgHA4DLfbTbNX5NJE+C1KpRJqtRoTExM0oFOpVNQY9e7du1hdXYXf7z/ziwjR0ZmcnMTk5CQcDgdqtRpSqRT29/cRiUSo6axQKIRAIKCt56T0dtZm0ccKdur1OkqlErxeL7hcLh4+fIjBwUF6W3Y6nVAqlQgEAtRPqFwuY3R0lIoPjo2NIRaLIRQK4eDg4EQyO2dRQiFpVK1WC7VaDYFAQG9b5GYsFAqhUqlw9epVGI1G1Ot1bG1tYXt7G3t7e6f+jC8LcsgNDQ3RjrKHDx/i3r17+MlPfoJ4PE5fTJIh0Ol0+Pu//3sMDQ1BJpPh448/xvLyMtbX1w9lTo6D05jHQqGA7e1tuFwuiEQicLlcysnRaDRUL4iQBknrLvBEndXj8eC///u/sbCw0PaeUc0gl5JcLod0Og2lUgmpVEoVk8lB0t3djYGBAVy+fJm25IfDYcrBazc0Gg2k02ksLCwAADUZFolEtDuLy+XS1nRy0PJ4PLDZbDQaT9zgNzc3ce/ePczPzyMej7dtsF4oFLC3t4cHDx7A4XBQs9ovAjlgo9Eo7dxphdYXCVokEgn0ej1GRkaoCfTc3Bw8Hg8SicShryf/hsvlQiQS4Y033sBbb72FyclJuq8mk0nMzMxgc3Oz7YIdBoOBS5cuYXJyEjweD8ViEWw2myqZk7Mkk8nQS/Lg4CAuX76M0dFRjIyMUFmWfD6PP/7xj3j48CF++ctf0jPnrMcjkUhw/fp1fOtb38LAwADl6pIuSLVajXq9TvXziHDwwcEBgsEg3G73mSc2jhXskA00m81ie3sbP/rRjzA+Pk51K65cuYILFy6gVCqBy+WiUCjg4OAANpuNlnXu37+PhYUFrK6uYmZmhhpLHve5ThvkphSJRBAMBhEOh6FQKOBwOPDd734XS0tL0Gq1sNlsGBoaApPJRCwWw+LiIvb29tqqrZUEFTabDT09PajVarBarVRJeGZmhpaw1Go1TCYT+vr6YLFYcHBwgEAggN/+9rdwuVzw+XwntomexjySAP3WrVvUi2VsbAwikYgKcjXrWWSzWWSzWTx8+JCm10m266zUhE8ClUoF8Xgc6+vrmJubQz6fx+DgIPR6PS5fvkx1kqx/slwgpclMJkNVwNutE4ugXC4jkUhgdXUV2WwWqVQK1WoVVqsVJpOJlmhJyzU5QEn20u/34/e//z0ePHiA3d3dts4uE8mOra0tWn4k5dXnXQ6IJEQsFsPu7i68Xm/LxtdsQaPValEoFGiX0vj4OG7cuIFgMEjJ/yQLWa/XKYmZZOrkcjnt9tna2sLCwkLbltAJpaNSqYDD4cBkMuGtt96CUqlEJBKhc6rT6WAymejaJJeScrmM/f19PHr0CB9//DFcLtep+ki9COQyIZPJIJfLIRKJ0Gg06DwymUzo9fpDvCNyXv7617/GwsIC5c+dm8wOcDjg2d3dBYfDQTKZRKlUgkqlokTPZvXPcrlMeSuhUIiagRFH2POCRqNBCXEej4fqmDgcDtrBotfraZ0yEAjQhdxO6qyEj0GM5Ii+A2nRJmJsTCYTMpmMEtLT6TTC4TB2d3fpjardybqknOrxeMBisSCRSOhNhUjSk245UlrN5XKYnZ3F5uYmfD4fotHoiWWvzgoksxqPx2krKJvNhtlsRm9vL9Wn0ev10Gq1EIlEKBaLSCQS2Nvbo52V7QiyB6VSKaqITCQT8vk81QjicDiUt0W+3uv1wuv1wuVyIRgMIp1Ot22gA4AG4MFgEOvr6xAIBFAoFNBoNGCz2ZS/Q97zSqVCS8wbGxtYX19vabBDno1w58LhMG2h1mg0mJiYgNlsRi6XQzKZPOTt1N/fj8HBQeh0OjCZTKTTaSQSCbjdbro2mzsL2wnRaBShUAhdXV3U2oG4gOt0OiSTSWg0GqjVaroGORwOarUaIpEI4vE4otEoFhYWKJ+pFeMkgQvJFKZSKWQyGdrYQOgQRFuHxWIhlUohGAxid3cXy8vL8Pv99PJ8ljgRUUFyW45EIrSstb29DYvFQtPFUqmUflA7OzvI5XLUdM/v9yMWi52rQAd4smj39vZoWpzwABwOB01ZcjgcpFIp+Hw+rKyswOv1Yn9/v60ODtJqvLOzA4lEQjuO+Hw+rl+/jt7eXlq+IrfEg4MDrK6uYnFxEUtLS9je3qYdMe2Oer0Or9dLswAej4eWtIgvD+E1MBhP3M8fPXpEvd3O23tKUCqVkEgksLS0hEwmg3Q6DYPBQDU/yuUylZ4nshB7e3vw+Xy0Rb2dkc/nEYlEkM/nweVysb29DZvNhoGBAUilUohEIlSrVRSLRaod5ff7KTE5Fou1LS+JgJhg7uzs4LPPPkOlUgGfz4fT6YREIqFu76SbLJVKwe/3w+fz4aOPPsL9+/dbWn4lByXhnaysrGB4eBgKhQJyuRxvvfUWvfETDatarUad7QUCAYrFIqLRKPb39w8FOsSwth2DVZfLBS6XS5shyFhIJr1SqWBycpLqJOXzecTjcSry6fV6EQwGsby8jEAg0NK1WK1WcXBwAL/fj83NTQBAb28vhEIhHaNMJgMAKvDp8/nwySefYGtri3qznXkn8ot+IIPBeOWnIRyHZtuAo90BzWRlovh40lFqo9F4KcLH64yxGUTPQ6vV4saNG7Db7RgZGYHBYKAqu4QLsLq6iqWlpRMLCk5yjCTLodPpcPnyZSqSaLFYqP8Kl8ulwen29jZmZ2cRjUapts5pROovM8bXmcPmjjKS0SE3yKPeZcS47zTEIM/qPW36PhCJRJSM/nd/93e0Fb9araJarVJR0K2tLezu7uL+/fvweDzIZrOvtUGd5RgJv4PP59MmCWJSy+fzEYvFKLeAeC2RDNBxupPOeh6ZTCaUSiVsNhvsdjsmJiaouF4ul0OpVEKpVEIymaQB3fz8PA3WX2etntRaJHpXpB374sWL0Ol0NBNOyKwky0oC1HQ6jXg8jocPHyISiSCRSCAajSIajSKdTiOXyx3rAD3NOezq6oJer8fQ0BDGxsag0+loJyDJIDdn2FdXVxGNRpFIJOD1eqntyXHLPycxRlKlIR5mJpMJV69epeXUZh5ZtVrF2toa1tbWsLi4iP39/VPvuH7eGE/c7IYcCGQCgae1Jdox8n5dlMtlZDIZ1Ot1GsgcHBxgZGQEHA4H5XIZ9+/fh9vtpum7dkyzkrQk8WURCoWQSCTY2dmhGTomkwmv14tIJIJQKIRAIICDg4O2rJF/EUjpg2ykwPE1UM4DCMk1mUyi0Whgbm4OxWIRpVLpUBlzZWWFtmuHw+G2vTEfBSmRkAtFoVDA8vIytaLJZDLI5XIoFArI5/MoFApn6s9zUqjX68hms7SrJZFIQCwWg8/nUwkPcnCS4C6RSLTFWAl/g1jQVKtVKBQKKJVKOBwOqunEZDLpHEYiEUSjUUQiEayvryOdTtP3lQR37fx+EmpApVJBLpejnVZisRi1Wo12IpN1SC4XpPuOjLEdzg6yd6bTadRqNWoPRHi4hEwOPIkHyD5CdPdahRPP7LQLzvqmBQAikQharRZmsxlvvvkmxGIxqtUqfvjDHyKRSCCfz5/YRvMnxddTGyPJznG5XFgsFirfvre3R7vqzmLTPK3MTrugFe8pAZvNxuDgIC5evAin04lkMolYLIZIJIK5uTlqoHncg6SVYzwrtMMYSQb9tMQCT2stEokAiUSC6elpqNVqyOVyNBoNasnicrkQCoWoFk2rxge8/hyS+REKheDxeBCJRODz+TQIKhQKNCN3WsHbaY6xuXpDOh1JBees8KJzsRPsnOAYSQqdpPFIyzKxzDjpl/csxkjKPQRnLcPeCXae4LTGSJoGeDweuFwuDW5IMHsSc93qMZ4FOmN8guMGAs0l5T/9TJpJIFyk09p/zmoOSVBw1I6EjPWUhR7/376nnWDnlG9af3qWU/n+7TDG00Yn2HmCs8gIkOzdeecltQKdMT7Bl318QGeM7Y4z4+x08DnauYbcQQcE5DZ5XrvMOuiggw6+CC/M7HTQQQcddNBBBx2cdzC/+Es66KCDDjrooIMOzi86wU4HHXTQQQcddPClRifY6aCDDjrooIMOvtToBDsddNBBBx100MGXGp1gp4MOOuiggw46+FKjE+x00EEHHXTQQQdfanSCnQ466KCDDjro4EuNTrDTQQcddNBBBx18qfFCBWUmk0kVB5vFB4+6Q5+2LQLxmyIeKS+Dk5bFPu0xvg7Om/T367iKnzeJ+ld9T87bHAKvPo/HHeN5cKM/j/P4qjhPa/G09po/fe+2eE9f50x6lTE+6/uf57X4wmCn2cWUDJD46JwVyM8jwc5JG1E+a4xH/7z596c50a0OqFr9818XL7sAz+v4gPZ5dvIcp7HpvcxabDQa52LDfR5edr951tec1rO0+2fZbvP9KnN42vN3WuvhRWf8eXlvjuKlvLGODorJZFI/HfJBn6LZJWq12ombEz7r5xAwGAxqUQ88cYYmVvWnPdazRvNLfVov8VkGx80gATKAQy7Kp/kunbJj8Qv//nmb8Fk+w0l/fwaDcWjuyP+PbvDnbeN90X7DYrHoXnPSl7vnPcNp4+gl+VnnxvOe53We8yzGdnQOORzOIVd2Mr56vX4qP/+0vm8zjo6xeU89Onftvga/kLNzdABkATZP6pcNzWMkv2cymXTT/TKhVRm7swCZv+ZsxJdtjEfxZRsjOSyaN3Ymk3nosDzv+9Czghqy35z3sRE07zNMJhNsNhtsNvupvz/PqFarh87F874Wn3X2N4+PzGNzANTOeKnT+1mDPhq9nmc86/mbszjNL+95H2sr0KrP7Oj8fRne1RfhdcbWTpvx857/WRnVdnruk8DRjMCXEUfn77SrAmeJ5oD1yzKmZ+FokuM8JQFe6HreLmSz10GHMPg5vuxj/LKPD+iMsd3RGeMTMJnMxhcd9Ewm80xKMK+Kzhx+jpcZY7te/p83xpfi7JwnsFgssFisVj/GqeJVx9iuL+X/ZzRzUV7268/bHL5qGv88jvH/C152Hl9mDtsx0Pmy4zRKaudtrZ5asEM+3KPpyrNI1Z5FipuQtQh/gIC0ybNYLFSrVZTL5UP1+JMadzuk8Zs/A1J/byauNxOeSb23Uqm03SJp7vgDPk/NMhgMSo5/WcmDV/2ZrQKLxQKbzT40R9Vqlc7Nl/FAOvqZczgc+vvmsZ+XUsSz9legtYfQaf3s5rEe/f1JcEhbuRbJxed5PLQv05nxLLxoLzzJfehUgh0mkwkOhwMulws2mw0Oh4NqtYpKpYJKpXKos+mkic71eh3lcvnEvt/zwOFwIBKJIJVKwWazKZlQLBZDLpdDKpUiHA4jGAwikUjQF5iM/zhjJkHDy+K0NiAejwexWAyJRAKlUknHWC6XweFwwGKxwGAwUC6XUS6XcXBwgFgshlKp1DYHCemi4PP5EIlEYDKZ4PP5EAqF4PF4yGazSKfT2N/ff4ooexyQGv+rfP1Jgc1mQyqVQqlUgsfj0bWZTqfpGi2Xy8d+T1/1357mO0EuISTAYbPZ0Ol0tAsqmUyiUqmgVqs9tU+1Y+DHYDDo/spmsw+RY09ijzmKl/1epzWHbDYbfD4fPB4P9XodXC6X7jHpdBr5fB6lUum1v3+r9iMmkwkejweBQPDcc7KZC/S6eNX95rTRHOCRz+BZQV6pVDqx5z7xYIcsQrlcjt7eXiiVSqhUKrDZbCQSCYRCIUSjUSQSCaTT6WeSDo/74p1VBMtisWA2m9HX1wen04m+vj6IRCLweDx6UEajUfzkJz/B9vY2otEouFwucrkcisXimQRlpwESHFitVgwODmJwcBBOp5My80UiEQ10KpUKDfru37+Pzz77DPF4vK26+cRiMTQaDaxWK95++23o9XqoVCpwOBwsLS1hfn4et27dwsHBAV18J/HsZ33TIpcQi8UCh8OBCxcuYHJyEhwOB+VyGXfv3sXi4iK2t7eRyWSQzWZRKBTaapN8VTAYDPB4PKhUKqhUKrBYLOh0Ouh0OgwNDaFUKqFYLCKRSCAQCCCdTiOTySAcDtOxl0qltgp42Gw2uFwuVCoVlEol1Go1JBIJstkskskkdnd3USgUaMB6EmhlVoDJZEIgEGBoaAh2ux1DQ0NQqVQQiUTIZDLY2NjA5uYmPvzwQ5TL5bbZV54HkkVms9mQy+VQqVTQ6XS4fv06xGIxGAwGPv74Y3g8HkQiEZRKJVSrVfrfeQa5aHC5XAiFQpjNZhgMBvT19aFYLKJUKiGXy8HlctEkQTKZpPN6nPf5xIIdMgi5XA6lUgmr1YqRkZFDm8z+/j729vbg9XoRDAYRiURQLpfpQiqXy8jn8/SG9To4ixedy+VCKpWiu7sbDocDdrsdly5dQn9/P/h8Pn2Ri8UiUqkU3G43FAoF9vb2kMvlsLe3h1qt9trBTisXM5vNhkQigVqtxqVLlzA0NITBwUE4HA5a/hEKhYeCHb1eD41Gg3w+j7W1NWQymbbYlEj5raurC1arFQ6HA1NTU9BoNJDL5eDxeGAymajVaohEIohGo0gmk4jFYvQmfZ4gEAigVCoxNDSE8fFxXLlyBZOTk+ByubSMIxQKoVAo4PF4EAgEEI/HkcvlWv3orwWSzdFoNBgcHERfXx+db51Oh56eHhSLRRQKBWQyGQQCASQSCcTjcSwuLiIWiyGTybR6GBQMBgNisRhisRgqlQo2mw06nQ56vR4KhQKZTAb7+/tgs9mIRCJIpVLIZDItX2evCxaLRTOuw8PDcDqdGBsbg91uh1KphFAoRCKRAJfLRa1Wo5mRdg3OWSwWDVKlUikUCgW9WOn1ekxPT0MqlYLJZNJ9c29vD9FolM5lMpk8d/sOCe7IWlSpVFAoFJDL5bBareju7kZvby+KxSKKxSIymQxEIhFisRjq9ToWFhaQSCSOveeeWLDDZrMhk8kwPT2N0dFRfOUrX8HQ0BAEAgFYLBYKhQIKhQKy2Sy2t7exu7sLj8eDWCwGDoeDRqMBj8eDnZ0dxGIxFAqF136W017cCoUCAwMDePfdd/HOO+/AYDBAp9MdIufV63VwOBwolUp873vfQy6XQzwex29/+1t8+umnKBaLODg4eO1naMUGxmAwIJFIYLPZMD4+jh/84AdQKBQ0tVwqlVAqlRCLxWiKnclkQiwWo7e3FwaDAQ8ePEAymUQikTjz5z8KDocDuVyOGzduYGJiApcvX6ZjyWQyEIvF6O7uhkqlgt1ux/LyMlZXV/Hhhx8inU4fO21+lnNIMhqXLl3C+++/T4P0Zq7VW2+9hYsXL9L39M6dO1hdXT1WsNOqg5ZcvoRCIa5cuYL3338f7777LsRi8SGuXXOJoFKpIJ/PIxQK4Uc/+hEePXqEhYWFljz/UZADo6+vDwMDA5iamsL09DS0Wi0UCgXEYjFqtRoODg7w85//HA8ePMDKygqWl5dPJBN51vPIYDAgEolgMpkwNDSEH/zgB3QtNotMKhQKxONxRCIRCAQCypFsN7BYLAiFQmi1Wnzzm9/EyMgIpqamYLVaIRAIwOVyD43L4XAglUohHo/j3r17mJubw/r6Oubm5k40Y3eaIFwcUgmQyWT41re+hcuXL2NiYgJ6vZ6W78jZWa/XUSwW8bWvfQ2ZTAb5fB7//M//jKWlJWQymWO9y8cOdsimodFo8O677+KDDz6AzWaD0WgEj8dDpVJBMpnE/fv3sb29ja2tLQSDQQwODmJ0dBTf/va36aaTTqfx8OFDrK6uYnZ2FvF4HMVisa1uJgwGA/39/RgdHcXIyAjMZjNkMhmYTCay2Sz8fj/cbjdWVlbgcDjgdDqh0+kgFoshk8nw7rvvQiaTwWq1YmZmBrFYDPl8vi0XKAGTyYRUKoVOp8Obb76JkZEROJ1OiEQihEIhBINBuN1uBAIBRKNRxONx+u9IlkskEqGnpwfBYLAt5pTJZEKlUuHixYuYnp6G0WhEo9HArVu3sL6+jrW1NRwcHECj0aC7uxvf/e53MTAwAB6PB7fbjd3dXVpXPw/o6uqCyWSCwWDAyMgIDAbDoeC80WiAw+HQYODatWuoVCrgcrkIh8NtV8p5Hki2zuFw0D3m5s2bMBgMtLxar9dRrVZpKblUKiGZTKJWq6FYLNKscyaTaYsxk/11ZGQp0+slAAAgAElEQVQE77//Pnp6emCz2cDhcJBKpeD3+yGXyyGXyyEUCjExMYFMJkNLPCfJezgLiEQiKJVKOJ1OTE5O4tq1a+jt7YVAIDhEyCYHqVarRX9/P27evIl79+4hGAy2egiHwGQyMTk5if7+fkxNTWFqagp6vZ4e9kebXMi4yHy+/fbb6O7uhs/nw4ULFxAIBLC3t4e5uTlwuVwAoM0wrd5XCQfHYDBAr9dTqoNYLIZAIMDg4CDUajX4fD4EAsGhJolm/o5GowGfz0cgEKBZMBaLhXg8/kp81WYcK9hhsViUoNrT04MLFy5gYGAAGo0GAoEAmUwGqVQK4XAY6+vrcLvd2NjYQCKRgEgkgtFoRLVahUAgAJ/Ph1KpRD6fp7yW7e1tJBIJHBwcHOqWaBXI7cpgMMBiscBgMNDDoVwuw+/3Y21tDY8ePcLi4iLN3ExPT1MSGrmpMJlMxONxCAQCJBIJxGKxU5WGf10wGAzI5XLo9XoMDAxgbGwMRqMRXC4XLpcLfr8fPp8P6+vrCAaD2N/fp6n/5mBHLBYjl8tRImirQUi6VqsVBoMBQqEQ6XQaGxsbWFlZwcrKCjKZDHQ6HQ4ODlCpVCAUCqFWq6HX62mJ4zzwrkhWx2w2o6enB2q1GgKBAPV6Hfl8nh4cpGzH4/Ho1+/v70Oj0WB/fx+FQqHt3s9mkNuzSqWC0+nE6OgoxsfH0dPTAwaDgVQqhWg0ikqlgmq1inw+j3K5jEKhgGAwSH9N+DvZbLYtxsvn86FWqzEyMgK73Q6tVguhUIhgMEgvG3w+H/39/ejp6aGcQULiPQ4t4KwhlUqhUqlgNptx+fJljI6OYnBwEEKhEI1GA8ViERwO5xC5VSQS0ZIQCR7aIUgFPicg9/T0YHh4GA6HA1arFTKZjDZEEBwV6yMEbJ1Oh1qtBqlUCplMhu7ubuzt7QEAYrEYcrkcMplMy6UbhEIhhEIhlEolhoeHYTabaflYKBTSEh6Px6N0j6NcMHLG8vl81Ot1iEQiSCQSiMViFIvFQ8Huq+K1gx2yOZpMJgwPD2N0dBRf//rXoVarwWQyUSqV4PF44PV6sba2hsXFRfh8Pvh8PlQqFbhcLhQKBaRSKRgMBnR1dWF0dBQOhwMmkwlSqRSffvopLXnlcrmWk7MI+bqvrw/Dw8Ow2Wzg8/moVqvIZrO4d+8eZmdn8eDBAwSDQQQCAWxtbcFkMkGv10MqlUKr1dJsT71eh9/vh9/vx6effkqDunYBKQXYbDbY7XZcvXoVTqcTmUwGy8vLuH37NkKhEGKxGM1QkcOfkJVJOzeLxUI0GqXdWK0GCa4HBweh1WppcL24uIjd3V0kEgnUajXE43Gw2WzEYjFIJBJIpVL09PRgb28P+/v7bR/sMJlMcLlcWvp48803oVAowGQyUS6XEQwGUa/XwWQyodPpwOVywWKxoFAoYLFYUCgUMDIygoWFBZTL5bZ6P4+Cx+Ohq6sLTqcTH3zwAfr7+2GxWFCtVhEOh+H1evH73/+ekncLhQINenZ2dpBMJg9lHZu7RlsJuVwOm82G69evo6+vD41GA7FYDPfu3cPW1ha2traQy+XwjW98A0KhkPKQyEWyVCq1xQXji8BgMCh/w+l04rvf/S60Wi1EIhHq9TpyuRwODg4gkUjA5/Np0CMUCiGXy2lJnWTv2gEsFgsSiQSjo6NwOp2UZ8Xlcp86tMmlg4BwIEkjjF6vh8ViQaVSQTabxRtvvIH//M//xPr6OvL5fMtJ5F1dXTCbzZiYmMC1a9dgsVjQ09NziMP5rGc8Om5yzjIYDOh0Omg0GshkMiQSiWMFdK8d7JDIU6FQ4ObNm3A6nVCr1WCxWEilUtjY2MD9+/exu7uL9fV17OzsIJ/Po1gsolarwev1IhqNIhqNwuFwwOFwYGxsDCqVCmq1GiKRCHK5HB6PB5ubm7h16xai0WhLX2KSDbBYLDCZTJBIJPTGuL6+js8++wwrKyvweDwolUrY3d1FOp3G8vIyGAwG+Hw+5HI5dDodpFIpBAIBgsEg/H4/JBIJbt++TUlZrQaDwaC3yenpaVitVvT29sLj8WBlZQWzs7NYXl6mAU5z62tz/ZX8ular0axOqw8P4AlZV6VSoaenBwAQjUaxsLCAzc1NSoYDgHw+j0QigXA4DJlMBr1eD5vNhq2tLezv77c9eVcikaC/vx9OpxN2ux0WiwVMJhPhcBjb29u4desWpFIpDAYDvva1r1GulUgkgtVqBYvFwubmJsLhMPL5PFKpVKuH9BRIQHfp0iU4HA5cvXoVExMT4PP5yGQy+N3vfofd3V14vV7Mz8/Td5a8i0TKoVarUemBk2j3PSlYrVYMDAygp6cH9XodPp8PMzMz+PDDDymBtVarYWxsDKlUCkajEd3d3YjH41haWqJdhO0wlueBZK+uXLkCp9OJq1evQqvVgsViIZfL0XWZz/9fe1fyFNd5fU8DPc9NjwzNKMlCAsvGkjzk59iuSjkuJ+XKIskulXVW+R+yySLbbJJtUpWqVGXhneWKLVmRLISEQQK6aZpuep4nep7o30K5Vw+EJCQQ/VD6VKlKxgLe1++977vDuedU8O6770IqlTLXQ6FQcJVWr9dzNUsMUCqVmJyc5KrbyMjInkCnWq3uqbxRNY7WRn/kcjlzX9rtNk/iraysoFAoIBKJoFKpnPj6JBIJVCoVRkZG8LOf/YyHVSYnJ6FWq6FQKHj/p6RKqIsklUo5Md7/c6lSa7fbYbPZEIvFjhTQHamNRWXEwcFBWCwWfsjogCgUCjzK2Wg00N/fD51Oh3K5DIlEwpsMVRAogJJKpbBYLJiYmOCpJipPdnMDoradTqfj/j/pk0QiEQSDQaTTadRqNSbsFotFhMNhDA8Pw2q1wmQyccTucDi4jOl0OkWxRoJSqYTFYsGFCxfwxhtvQK/Xo16vY2NjAxsbG/D5fMjn80/lrVBPlkYpW60W8vm8aIStpFIplEoltFotGo0GisUi0uk0yuXynmoNie3V63XeWI1GI0wmE/R6fRdXcDhQ0DI0NLRHV6dUKiGRSMDn88HhcECv1+/RviAJAaPRCIfDAa1Wy/wAsUGhUMDhcOD8+fOYmZnhqatMJoNwOIzFxUWEQiHEYjGeAKWpHVrzwMAALBYLTCYTc8symQzi8XjXW+g04arRaFCpVJDJZBCJRLi6WCqVePKx0+lAJpNxW8disaBYLIqaF0hTZtTqmZ6ehsPhQLvdxs7ODjKZDB4+fIhqtcqHorBKQPdPKBIpBtB+MTw8DJPJBK1WC7lczhxV4rMSd4z0vVQq1R4uC/0sItbTO6pWq/nfdGvtNJg0OzvL757T6YROp4NEIuHpKuH7RsGOwWDYI+YqBAVING5P2l9HeQ+P1MaigIXK+1Qar1arKJVKezgoGo0GCoUCGo0GiUSCSVhGoxFmsxkWi4VLdvSyjo2NQS6XI5vN8gPQTXEkCtZ0Oh1UKhUkEgkqlQrS6TR8Ph/8fj+LzwGPD0qfz8fVoE6nw5Gs0WgE8KhcPjw8zGsE0PV2gdFoxPj4OK5evYqLFy+iWCzC7XZjYWEBXq8XwWDwqdfY19fH93VwcBAymQzVahWrq6tHmrI7TlCWpNFoUC6Xkc/nWc/hoOCNAiCZTAaTyQSr1QqLxdL1PvmzQNNzMzMzGB4eZpJfpVLh4C4ej3OFUsiFkEgkUCqVMBgMGB4ehk6ng1wu7/aSDgRttleuXOEKSDqdxvr6Om7fvo0bN24gnU6jWCw+8czSu6hUKrnC/NOf/hTffvstVldXkc/nu952Ja0ylUqFeDyOZDKJeDyORCKBcrnMY9dCF2qz2YyJiQlMTEwgk8mgUCiINtih/WJ+fh5vv/02RkdHodfrkUqleODj1q1bUCqVsNvtvEZh5YOSRFI8FwOo6jE5OQmTybQnOKFWVDQaRbFYRK1Wg9lshtFo5CDpoLYPrZWS5HK5jFqtxtWik16fQqGAzWbj7o7D4YDFYgEApqlsbm7ygAPdI4lEgtnZWQ5S94OkWYrFIvL5PPL5PIrF4pHu7ZGCHSrDdTodftio/fHBBx9gdHQUOzs7KBQKkEqlzKouFovcR9br9TxJYDabucJDM/k0uv7Pf/4T29vbXeNICIM7IlRT7zwcDsPv9+8hrFLvWC6XY2dnB7VaDa1Wa082IpPJOKN+6623YDQaWQ21m4doX18f3n//fVy+fBkXL15Eu91GNBrF8vIyVlZWkMvlnhrokI7Jm2++ibGxMX7wY7EY7t27J4oWHQDkcjkObki7hA7z/Z/77u4uc5IGBga4rVAoFHDjxg3RbK77IRTwslqt0Ov1zENRq9UYGxvDr3/9ay6z788oBwYGoFQqmcAtRndjYQXy8uXLMBqNqNfrWFhYwO3bt/Htt99yYL7/2RsYGMDo6CisVitGR0fxi1/8AjMzM7BYLAiHw0in06JouxKhutPp8L5CX6NWsVarRbvdRjqdxhtvvMFk3ZmZGUSjUeTzeaRSqa6u42kwm80Y/68uGz2HxIEkTpLX68Xbb7+NqakpqNXqPZUMEn6kc6XbiSKB9vBms7knQGm32ygWiwgGg7h16xZyuRwqlQpPEFLF42nBC1WFarUaq7sT8f4k19bf3w+r1Yrp6WnMz89jbGyMK/npdBqhUAhra2v4/vvvUSgUUK1WWYNvenoaer0eCoXiiX2F+Fk7OzuIxWJwuVzY3Nx85rlzGLx0sCNstVDrSiKRoFAosMbM0NAQrFYrWq0WZDIZtzYo0CECnUwm414lPSCdToeDBQqSdDod9567sQHtJy6SJkC1WuWpFpKkJwVelUqFM2fOcGYtHDOkw0ipVHJUr9FouqoVQUHY1NQUJicnodPpsLW1hWAwiM3NTZRKpae+VPS9VqsVY2NjzPkIhULcPhBLsNNoNFjEiiqTNpsNcrl8z5g1TZMR2ZoIh0SIVKvVqFQqotlgDwJdN20qNNZKB73ZbIbJZDpwOkJYThZjUKfRaPiw1Ol0AIBsNguPx4Pt7W1kMpknAh2qgGi1WgwNDWFychJXr17F2bNnYTKZUCqV4Pf7eZhCDMEOcYxkMhlPqND9ov0wn88jEomwtQLwqCpkMBhYpE2MGBoawvj4OJxOJ1QqFQdtPp8PgUAAiUSCKwjj4+MckAuDCTr4xbTH0LtDiS5VNkj2IJfLIZ1O83VTq5HaOvuJu/S1ZrOJarWKQqGAVCqFbDbbted0/6g4URaCwSA2NjawtLSE1dVVPjccDgeGhoZgMBj2VOgIJEFDAsQ3btyA3+9HPp8/8v7z0sGO0LODlDt3dnaQzWbR19cHk8kEk8m0x7+FoFKp+O80JkiBA32NPjgKBuilzWQyXdl0hYGOMNMinY56vc6HvVwu57aUXq/njdRmsz1xc+l7iAei0+mQy+VOfH3C61EqlUyoUyqVXEp2u93MRzoIJOI2OjqKyclJOJ1OJBIJBINBeDyerrcDhKANslQqweFwcMahVqtZtpymAqiPTqOgKpWKx0D1er1op5ToENxvyUJjn8QhOGjToe/f3d1lgqtYSJ9CaLVaWK1WflZLpRJisRjcbjdCoRBKpdKeQ4N4gXK5HFarlSsKn376KQwGA5rNJjY3N3H//n2sra2JItip1WqoVCqoVqsszmY2m/ckh81mE8lkEn6/n99hapUbjUZR88ucTiempqbgdDohk8l4IGB7exuxWAw7OzuYnJzE5OQkpqenn3hWyXdPbLpsFNgQD5D2CKpElUolJigThKbKT/uZ9L3JZJLlPk56/xGeh8TbJZpJs9nE1tYWHjx4gLt378LtdrNTAu01DoeDuzjCyk6r1eL25fr6Ov7xj38gFAodC/3hSMFOvV5HKBTCH//4Ry5HkW+J0WhEuVxmgpnNZoNCoYBMJkO9Xmd2eSgUgs/nQyKRgFQqxdmzZ3Hx4kX87ne/Y16MXC7n0l03CWiNRgOpVIqjcTL8NJvNGBkZYY0Lq9WKmZkZ1sWgrx9EIhMaFJZKpa5re9D0gNPphEajQTQaxVdffYXV1VVEo9GnvlRSqRQTExM4d+4cfvvb32JwcBC7u7vw+/2sryQmEEk3FArB6XTCaDTinXfewZkzZyCRSBCLxSCXy9n7bHx8HGazmQ1CaSMTozYSodPpIJlM4uuvv8b//d//wWQycfmZkoiDskdCu91GtVpFKBRCPp/nZERMoIBUp9Oh2WwiHA7j9u3b2NjY4EoGtX5ompISi5mZGXzyySe4cOECbDYbcrkcPB4P/vznP2NpaQnpdFoU1SyPxwODwYBUKsVk89nZWSa7djodlMtlBINByGQylMtlDmgtFguGh4cRi8WwsrLS7aUcCEr41Go1v5fpdBoGg4Erjz//+c95BFl4OLbbbVanj8ViokqoqBBw//59RCIRVrumVj/xlIRiekSROKiNRYFOp9NBqVTCwsLCkUT2jgqqwAlNkkkLqVAooFwuM0eVkqyf/OQneO+99zA9Pc26XrROuvfr6+v429/+huXlZQSDQXEYgdKD1mg0OEJrNBpQq9VQKpVsEUHaCNSuIkKaRqOBx+NhtVKSPB8YGOCMk7hANG8fDoe7ymfZr4tA9gntdhvj4+OQy+XQ6/UYHh5mDx5S/tx/zcJ2Vn9/P2fS3TpAaSpibm6Op688Hg/i8Th77AjXQdetUChYwO38+fOw2+3I5XKIRCL49ttvkUgkRHFoCNHpPDKYW1paYqE9u92OL774glsYFMBNT09zjxl4VBWq1+uoVCpMEBUrKFi5du0aIpEIqyeTnk6xWGSSI2XWdJhQqTybzaJUKonqICHsb9GpVCo4HA5MTU1Bp9OhVCpxC5nEMUmkzG63Y3p6GiaTCbVaDWtra1x2J2l6MYBsA7LZLMbGxqDVajH+X2Xavr4+RKNRPgDj8Tg8Hg/OnDkDjUaDbDaLbDYrKo8vISQSCZP9KRlUKpVwOBz44IMPuKpPHCRhAEBVx1AohI2NDbjdbtFJQZBwZ7Va5UCFgjuqhAOPNXX2VzqEoIlWl8sFl8uFr776is0yuwFhSy2fz3N1RyaT4cyZM5BKpTw8QOfihx9+iOnpaS6OCLm/1Lq6e/cutra2WGj3uHCkYGd3d5cDGuDRDSuVSnyz9muvUAWDNhqTycQlKupVajQaNmGkh5/GtK1Wa9ezaCIpU0BCAld2u52nImiEkMqtVAWjNdLkxP6ROwp2utVzpmDnwoUL0Gq1KJfL2N7e5ikWYdl8/3jyBx98gHfffRfnzp1Df38/vF4v1tbWcOPGDVFWBACgUCjghx9+wOXLl6FUKjExMYHPP/8c0WgUwWAQwCONExLyoooO8X0okBdjC4tQq9UQj8fx1VdfsUpwu93m5CMWi/HIPU1EklddsVhEJpNhM1AxBjvCvQYAjzC/+eabyGQyKJfLmJubY6sMp9PJgnQ0CUIE/JWVFdy9exd+v180gQ4Aro6TbQe1imdnZ1Gr1ZhITTyQQCDAba5sNot0Ot3V1vjzQNUboZfe6Ogozp49u2cUez+IyBqJROD1erG5udkVrZlngegZQo05mpx6kS4F0SfS6TTu3buHhYUFfPPNN6/wyg+HdruNSqWCSCQClUrFgx4k1joxMYF6vQ6dTgeLxcI6Q8KKDlUmw+Ew3G437t69i2g0euyB67EZgQKPb4hwEfRHKC5Xr9dRrVYRj8e5BEfTWlarFVarlSs89H06nQ4ajaarfkQUWT98+BBqtRpWq5V5DxqNhtdJGzDxeeLxOBYXF+H3+zExMcGTSsJKATlpd3sKSy6Xw2QyodPpsAT9hQsXYDKZkM/nuTTZ39+PqakpXLp0CZcuXeJqUH9/Px48eIC1tTU8fPjwmRyfbqNarSISieD69esolUqc+VssFszOznI1q9lsIhQKcRmavJW6rb9yGND1ezweVuoml2y9Xg+v18syDx9//DEH541GA36/HysrK1hYWOhqBvks5HI5bG9v4/vvv2fSuMViwczMDJN6qfUok8lYy4oqxo1GA7lcDt988w2uX7+O9fV1UQU6wKMDJRQK4V//+hcMBgMmJyfhcDjwm9/8hsXbHjx4AKlUCpvNhkuXLsFms3HbTiKRiFbpm9obCoWCK/okeSD8N/S+tVotpjdQsEN8MjGRkwl0Zvh8PlgsFta72q+j8zw0Gg0UCgW43W5cu3YNS0tLr/CqD49Go4GtrS386U9/wu9//3ucO3cOJpMJZrOZZSuazSYn+Qdp6rRaLSQSCQQCAWxubuLhw4coFArHfq3HGuwAT7o509+FQQ+AJ4hZUqmUJ7aol0ffRy7UxWKR2z3dmsZqtVpwuVyQy+UwGAw4f/48V3GIeS/09aISdCAQQKFQgFqt5koWAD5YaPywmyRQKicGAgFcuHABOp0OTqcTn332GZPpSAW53W7DZrOxGjRVutrtNttC5HI50W0+QpCol8vlYpKjw+GASqXiaRaq3kSjUczNzeHMmTM8GnoaQM+s0AeqXC5DqVRCqVQin8/DZDJheHiYK3ftdhuxWAw+n4+DJLHoI+1HuVxGLBbD8vIybDYbxsbGMDo6ytmjUDdI+HcAXNHx+/3MD8hms11e0ZOgdo3f78e9e/c44VAoFJiamsJnn32GCxcuQK1Ws2YZSQUMDQ3BZrPBZDLB7/eLLmDtdDqsIj86OgqLxYKBgYE948f5fB7b29toNpuQSqX49NNP2TupXq+jUCjwWLLY1kcFgKWlJZTLZWSzWVy6dIkV9DUaDZ911FqmCheBKh/pdBputxvJZFI0FaxOp4NKpcIuB/F4HG+99RYcDgcnF8RBom7G/qpOu91GPp9HIBDA1tbWK6MGvJJg5zD/v9lssv4AfRAU4ROhUDhWSSqgAwMDXeO00I3x+XwYGBjA4OAgiwySejSNA37//feIRqNIJpMolUqoVCro7+/HxYsXueROa2s2m8jn8xwAdqu6Qw+u3+9HJpNhe4v5+Xlu3ZHxXLlc5qCU+CsU9KXTaRYyEzPa7TYTO2kzIU6HVquFSqVCqVTie6rVanlMe79TsVghFFoTco1oU5VKpVCr1RgdHeVskzKtcDjMTvZirQzUajVkMhm4XC7Y7XZUKhWeDqF3i9rGQlFEasHHYjF4vV5sbGwgmUyyea/YUKvVkEgk8PDhQ7TbbahUKjidTgwODrJNBkkikA5NX18frFYrVytJql9siEQi8Pv9GB0d5Wkkatkkk0mEw2HW6NJoNPjwww+5rVWtVtnhXayJ1e7uLjY2NlAoFJDP51n0UavVwmKx8Fmn0WigVqvZNFMI2p/IgkhM72Oj0UAsFsOdO3dQLBah0WigUqm46yHkAe7fM6nTk8lk2CdSKP1xnDj2YOdFQaJYEslj0y9hCRN4rKbYbrehUCi6eqN3d3fh8/mQzWYRi8WwurrKWVS1WmXFR7JTKBaLrDJNDHzajGht1Hcn875uBTu7u7vI5XK4ceMGDAYDj5/TyDwFB9vb23C5XBz06fV6/PKXv2RO0sbGBrxeL0Kh0Imv4WWQTCaRTqextbXFG41CoYBWq0W9XmenZTo0JicnOVg4TRAGPjQGOjk5yerDlEhUKhVsbGwgEonwOK9YD5JWq8UE6kqlwtY1Y2Nj0Ol0TOadmpqCVqvl76nVanC5XFhZWcHKygqWl5dFR24Vgu7LvXv32J/uvffew7lz59higare5F9GlZ65uTlUq1XcvHlzT1VZDOh0OlhcXEQkEoHP58P09DT76JGUSTKZxNbWFvR6PRu7UsUyFotxVVxsVR1Cp9NBLBZjYUcaxiGLErJIGh0dxczMDM6fP8+abMCjfTmRSLC4otgC8larhUKhgMXFRYTDYSQSCUSjUYyMjGBkZIR5cgepPO/s7CAajeLWrVtwuVxsTPwqIIrdmsiuQnHB/SOxRHR+0V7nqwBVcLa2tpDP5/maSH+H+qukrUCkO4vFwrL7tDFRIEf+IcCT0e9Jry2TyeDOnTsIBoPw+Xw8LtnpdBCPx5nAS1YDQ0NDrFxLZEqxC+3tB5Wby+Uy252QQGa73YZSqWRvlk6nwyVaclgW60Z7EPr6+thol8xBDQYD+vr6WDMrnU4jkUggnU6Lfm30DoVCIaRSKQQCAYTDYVZGnpycZH0POjhisRju37+PxcVFuN1u0U/VAY/bGUR6rdfrCAQCCAQCeOedd7gNHYlEWLfG4XCw6OLU1BQrvYsJpGReLBbh9Xo5sKMJJnofKakUBgEk1yHmQBV4TFROpVJYXl7m55HI8jKZDDabDRKJhM2Jgcfj5vl8nnktYmlhCUHPXjwex71793gKcmJiAj/60Y84USRrGuDRZxKJRLC6ugq3241wOPxKifSiCXYAMPl4vwjafhG+bgc7ws2mUCjssZIgCA96CnYGBwd5CkZIkBRyYbp9sNBGQ5yjXC7HpqckA04THsVike+HTqdDPp9HMplkTRaxHx77QW2P/aD7K1yPcKqum6Tyl4FUKuVK4/nz59kOAgCKxSKSySRPYeXz+VOxNur7UxuEJsfIk4h4AzSd5vV6sb6+Do/Hw0rJpwHEl6OKI7VGqP1IkzG1Wg0SiYTbshT4kYyEmEBBTbFYRDwe531RaPxIlUiNRrNnyrVSqXBbXcygSlSr1UK1Wt3DI6NEP51O48qVK3vOQAp2qGOQyWRE1cISotlsYmdnB5VKBfV6HdFoFKlUCkajEe12m215hMEOteYikQhX6A4DIefnsBBFsEPiZR6PBxcvXsTExMQeQrPQ4E5MJDR6gJ8H8hCZnJyE3W7nF7bVaiGXyyEUCmF5eRmFQuFAD5+Txu7uLh92Xq/3qcEleQtduXIFg4ODCAQC8Hq9SCQSoiW0vgwOUiImHKQ8LGZIJBLo9XpMTU1hfn4ev/rVr1gLqt1uIxAIYGlpCR6PB4FAAMlkstuX/FKQyWQwm804d+4cZmdnodFosLu7y6aSS0tLuHPnDhKJhOirAvtBxFziOrhcLpk+9BgAAA4mSURBVHz33XfcoiTH92q1irm5OWg0GjgcDszMzDDBVSx7KLC3vfo0mYr+/n4MDQ3h0qVLXFGlYYh4PI50On1i13vU5Ea4lwidvWu1GiubC6tX5ONXKBSe21LuduJFCWMkEkE8Hsfm5ibS6TQ+/fRT2O12mM1m/reNRoNVvwOBAEql0nPPPmGx40XPSVEEO8BjciwdssI2Fr0EdKPF9KI+DxKJhL2IHA4HT/sAj8mVoVAIDx48YPXkbj+w+7H/WkiLxW63Y35+Hp988gkKhQJ8Ph/W1tbg8/le6gChErUYVYlJLXl3d5c3pG5bmLwMaELn7NmzuHTpEqxWKz+PNNlEdgmZTKbrgfeLQq1Ww2az4erVq5ifn8d7770HpVKJcrmMVCqFv//971hdXcX29jbr1pxmUKBAnBX6b/Lrq1Qq7KWlVqu5hX6aWsy0h1osFkxMTPAeQe2dYrH4UsnVy+6zx7030X46MjKCwcHBPYbErVaLXQa2t7ef+T4KJ5i7DZqiJhoH6c8JKzLUGUmn04emPQgDxRddp2hsjMkWgkqYwsiW+rmlUklUlZ3DQCKRQKVSwWAw8PQWMe1pTL1QKCCRSLDCpthBEzxDQ0M86huLxRCNRhGPx480LSDWKolQDp0UUJVKJbf4TgOoSjo4OIjh4WHWaKFR30QigUgkwsatp606R7YRw8PDePPNN5mPBDwioa+vr+P+/fvweDyIRCJsvPg6gCojdMAQz4WSDiLckzL9aQJZglCCQS07mix8Gd+2FxX1e5UgzirJeAjvT6vVQjweRyQSQSKR6OJVvhxIk432S2GwI+RlHVY/71lV9udBFE89jZ2fPXuWNSKEY7Cks0Ajwqcp2yT15/HxcUxMTPDhQhov5XIZlUpljzz9qwx4iCt0lN9BCqeXL1/G+H9diG/cuAGXy3WkKgdlpWIEZYE0PUiZisFgODWtHhq/HhkZwfnz5zE/P8/DALVaDf/+979x8+ZNrKysiJYX8CyQOOLbb7+Nzz//nKtW2WwW3333Hb788kusrKywC/XrhP3aZqTtFYlEuKrjdDphMpmgUqlEq2r+NGi1WjbeBR6PnJPJ5oueCWJLKvv7+3H27FlYrdY9sivNZpOtMPx+/zN/hhjPxb6+vj3GycIWFFVWaQ961RBFsAOAhQMVCgVH70RA8/l8LGkvxhv6NNChOD09jeHhYZhMJp7CqtfriMViWFxcxNLS0omt66i/hypVNpsNX3zxBWtjuN1u+Hw+RKPRl/7ZL1uePAnQqDmNZ5Paa7cELl8GarUaw8PDuHz5MqamprjCWK1WkUql8OWXX+4x0DwtoKz/o48+wuzsLN59913Y7XZIpVJUq1Vcv34dCwsL2NjYYNuW1x00GZnNZtnqhVSz1Wq1KMUTDwJVUcfGxngP7evrQ6VSYT2vl21FiuW9JYHac+fOwWKx8NepkqxWq0WrkfQs0LSZ3W5/wsC12Wzi1q1bPA15EuefKIIdITOd+ssU1ZZKJQQCAdF7EB0ECnaGhoYwODgIjUYDAKzWG4vFsLW1hWAwKJoX73mgQ56E6EiFl7Kso2aMYv4cqPJEJfO+vr5TFXyTFQiVy2nzKRQKCAaDCAaDPAZ8WjAwMMBaOrOzs7hw4QImJiYglUqZp7OyssLaWGLkg70K0Dg+CbQRb+Jpkv1iBk1iyWQyFjKtVCrI5XIolUqnZpLuaaDpSKvVCrVazV8XVrpPU1JFoOm5/SPnwGPiNU1vncj1nMhveQ7oRSTnaYvFgna7jWKxiFgshqWlJVErZD4N5GpLlR29Xs8VLBIkfPjwIba2trp9qYeGQqGAUqlk07dCoQC/38+cltN2jw4LOjyIC0GB92nhkFE52WazwWg07umfb29v49atW8y1Og3rIVB75pNPPsFHH32E4eFhWCwWVgJfWlrC119/jUgkwgKf/wsQmhULJ1tP070ldDodfjbJ0iSXy7Euy2kKzveDKuVmsxk2m40TYgCcWJHY7Gl7dlUqFex2OyYnJ1nBG3gcuJG+0H616FcFUQQ7KpUKVqsV8/PzbJDZbrdZD2N5eRnlcvlUvaxEBJ2ensaVK1fYe6jVaiEajcLtdsPlciGdTp+q/rnRaMTIyAjOnDkDAFhdXcW1a9dYV+e0vZCHRafTYcE9siAQOtuLHXq9Hna7HdPT0xgbG2ND052dHbhcLty8eXNPECdWUBuYJuLef/99zM7O4uOPP+bWXKlUwvXr17G6uooffvgBPp8P1Wr1tX02D0KlUkGhUGDyp0QigcViwdTUFOLxOAKBwKnYS4m8q9VqYTQaYTQaWcD1dbmnzWaTvetI3Rx4bKVQLBZPXfWKtNemp6cxODi4J7mi6epCoYCdnZ0Tk37o+jQWRbaDg4OsQUO8CJLXTqVSp+pmU0vOZDJhfHx8z3hvqVRi36FYLCY6+fZnoa+vDzqdDna7HePj42g2m4jFYtjY2DiVL+SLgNqqFJjKZDKoVCpYLJYTy0yOAp1OB4vFwu+YVCrF7u4u255EIpFDT0R0GxKJBEqlEqOjo7h06RLm5uYwNjYGlUrFvl5utxtutxtbW1ucGf8vQeiFRmunsXxS6j0NINE98kykd43W9jqAqnBkkkmo1+s8wHJaKsgE6mqYTCYW9iRQAEfCtCd1H7se7ACPsk6HwwGbzQaVSsVTWJlMBvF4HNlsVhTqwoeBUBVzaGgIc3NzMBgMkMlkbKTp9/vh8XgQDodFKf19EKhvTt5Qc3NzqFQqCIVCcLlcSCaTp25U+UXRbDZZFkGhUMBoNOL8+fPsuSRmWK1WDA8PY2xsDEqlEn19fWi1WggGgwiFQojFYi811XLSIP6JXq/HW2+9hR//+Me4evUqnE4npFIpdnZ2sLGxgR9++AEPHz6E1+v9nwt0gMcaLaVSiXk71PJzOp2npiJJBGXyVhIOeFQqlVOxhmeBWjq7u7tQKpXsmUjWINlsFsVi8dS1l2UyGXuzjYyMsN8ljZzTOUjeeyeBrrex6EZT+Y78iVKpFDweD7a2tjiqPU1tLBIStNls2N3dRbFYRKFQwDfffIObN29ifX0dyWTy1FR1ZDIZ9Ho9JiYmMD4+DofDgY2NDWSzWfYVOq5DRYz3mSYDhdwBsV3j00B8HbJNIOO+XC6H27dvw+PxsA+YmEFJhFwuZ1Kyw+GAwWBAu91GNBrFxsYG7t+/j5WVFaRSqf/JQAcAa+1EIhHEYjEMDg6yei1JEBABVszPsXB4ha63VCohnU4jFoshFAodySqi23uNMDkWaiXV63V4PB4sLCxgcXERqVRK9O+nEKSro1KpmGdF7Su/34/FxUVsbm6yTdJJoOvBjkQi4bJWLpeDQqFANpvF1tYWVz9OS1UHeFwBoamBRqOBaDSKbDaLaDSK5eVl+Hy+U7UR01SZSqWC0+mE2WxGf38/+0PJ5fJjJ3+KMWMj756trS2o1WqUSiWEQiFRV7TosCDz2nA4DJPJxG7hZB9wGoJuyoJpmoh8oejg9nq93L7K5/MnUh4X43MKPBZjDQaD8Hq9kEqlyOVyiEQiyGazPNZ8WvZVMnGtVqsIh8OIRqNIJBLPtU84DLotMEgJfzabhclkglwuh9/vx+rqKlZXV5nXeZR7dZLro99FxqepVAqdTocndwOBALa3t3mS7qSeQVEEO1TJCQQCnGFSdhYOh58or7+MCdhJgXy8yGson89jbW0Nm5ub8Hq9WFxcFKUZ37NAwY5Go8HU1BTMZjN2d3dZlVWn02FnZ+dY74cYD5FcLge/34+FhQVoNBqUy2Wsrq6iWCy+8M86yfX19fWhXC5zsE0yDoVCASsrK0gkEq/kXXoVayRuQ6vVwvb2Nnw+HwqFAprNJh48eID19XWsrKyciFQFrU+Mzyq1ejY2NqDX61EoFGA2m1nq4jS0LIHHAW5fXx8H7Ovr69jc3ORE+Kjr6La5NPEBo9Eo9Ho9+vv7cefOHdy5cwdLS0vI5XJHSkYomDupNZJGXj6f53NvZ2cHGo2Gh3M2NzdPXNyz68EORbT5fB5/+MMfmKDVaDTY4Xf/wyzGIIdAGVWlUmFDRcpAhGsS2xqE7t37P2/hZkCmdVSSJIG941yPGDPOTqfDz6nf7+fqwmE9XQ76eccF4Ua2/7OjjXRtbQ1ut5uFEQHsuYevAq8qgKpUKgiHwwiHw7h+/Tq3OFqtFvOqTqJSJWYRTGqFhMNhJJNJHvOl6axuVpXpfgHP/+xarRbK5TL8fj9cLhe0Wi3+8pe/YHNzE/F4/MiBTrdV2zudDkqlEqLRKO7du4dGo4FUKoW//vWvrHt11Hv1qvbTp7UAacqzWq0iHo8jFApBo9FAoVBwlyOXy504wVzyrA9BIpGI7y0+JDqdzqHC2FexRuqJS6VSDAwMsBXEcT9wx7lGYdR/kPGnQqGATqfDhx9+CKvVCoPBgFQqhbW1NayvryOXyx25zHrQ9x9mjd14TinAOIbN9sj38Fn37mn//iQP6ON+TinQPAl7lf2/+2m/r5v7zSF/L4tgvuzndVzv4otU5ul+m81mTE9PY2RkBP/5z3/YaudlcJS95r/ff6z3kOgAFy9ehMPhgE6nw7Vr17Czs3PsAcFJrpHunV6vh0wm47OwVqu90kDnaWvsBTu9NR4a1M6izeq4JgSedViLNdg5LvSe08cQ8xqFJNmDRPqOusZuE2UPg9fhXaQk5aBqx+vwnD4P/8trfGaw00MPPfTQQw899HDaIQqdnR566KGHHnrooYdXhV6w00MPPfTQQw89vNboBTs99NBDDz300MNrjV6w00MPPfTQQw89vNboBTs99NBDDz300MNrjV6w00MPPfTQQw89vNb4f9PXnXo77VV5AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 720x720 with 100 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S7aR6rs-i7JB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def load_data():\n",
        "    G.load_state_dict(torch.load('cDCGAN_G.pkl'))\n",
        "    D.load_state_dict(torch.load('cDCGAN_D.pkl'))\n",
        "    optimizer_G.load_state_dict(torch.load('cDCGAN_optimizer_G.pkl'))\n",
        "    optimizer_D.load_state_dict(torch.load('cDCGAN_optimizer_D.pkl'))\n",
        "    print(\"Loaded.\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UCMTCpRZjwdX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# load_data()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hDCZ283y59ln",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def save_model():\n",
        "    torch.save(G.state_dict(), 'cDCGAN_G.pkl')\n",
        "    torch.save(D.state_dict(), 'cDCGAN_D.pkl')\n",
        "    torch.save(optimizer_G.state_dict(), 'cDCGAN_optimizer_G.pkl')\n",
        "    torch.save(optimizer_D.state_dict(), 'cDCGAN_optimizer_D.pkl')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4614ltcYkHt7",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "save_model()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6yYA4YxO63H9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "files.download('cDCGAN_G.pkl')\n",
        "files.download('cDCGAN_D.pkl')\n",
        "files.download('cDCGAN_optimizer_G.pkl')\n",
        "files.download('cDCGAN_optimizer_D.pkl')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TW8SAfAv3U6O",
        "colab_type": "text"
      },
      "source": [
        ""
      ]
    }
  ]
}