4_robustness/Chapter_4_Simple_Transparent_Adversarial_Examples.ipynb

{
  "cells": [
    {
      "attachments": {},
      "cell_type": "markdown",
      "id": "4bf358d5",
      "metadata": {
        "id": "4bf358d5"
      },
      "source": [
        "# Chapter 4: Simple Transparent Adversarial Examples\n",
        "\n",
        "\n",
        "\n",
        "| **Platform**           | **Link**                                                                                                                                                                                                                                                                                               | **Notes**                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    |\n",
        "|------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n",
        "| _Local Folder_         | [Deep Dive Example: Simple Transparent Adversarial Examples](4_robustness/Chapter_4_Simple_Transparent_Adversarial_Examples.ipynb)                                                                                                                                                                                                                                |   `4_robustness/`<br>`Chapter_4_Simple_Transparent_Adversarial_Examples.ipynb`                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           |\n",
        "| _Book Chapter_         | [Chapter 4: Robustness](https://learning.oreilly.com/library/view/practicing-trustworthy-machine/9781098120269/ch04.html)                                                                                                                                                                                 | Ebook available from [O'Reilly](https://www.oreilly.com/library/view/practicing-trustworthy-machine/9781098120269/).<br>Physical book available from [Amazon](https://www.amazon.com/Practicing-Trustworthy-Machine-Learning-Transparent/dp/1098120272)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      |\n",
        "| _GitHub_               | [![GitHub](https://img.shields.io/badge/-View%20on%20GitHub-181717?logo=github&logoColor=ffffff)](https://github.com/matthew-mcateer/practicing_trustworthy_machine_learning/blob/main/4_robustness/Chapter_4_Simple_Transparent_Adversarial_Examples.ipynb)                                                                          | Please star on GitHub so others can<br>find this repo and get help with implenting<br>their trustworthy AI pipelines<br> [![GitHub forks](https://img.shields.io/github/forks/matthew-mcateer/practicing_trustworthy_machine_learning.svg?style=social&label=Fork&maxAge=2592000)](https://github.com/matthew-mcateer/practicing_trustworthy_machine_learning/network/) <br> [![GitHub stars](https://img.shields.io/github/stars/matthew-mcateer/practicing_trustworthy_machine_learning.svg?style=social&label=Star&maxAge=2592000)](https://github.com/matthew-mcateer/practicing_trustworthy_machine_learning/stargazers/) <br> [![GitHub watchers](https://img.shields.io/github/watchers/matthew-mcateer/practicing_trustworthy_machine_learning.svg?style=social&label=Watch&maxAge=2592000)](https://github.com/matthew-mcateer/practicing_trustworthy_machine_learning/watchers/) <br>[`@matthew-mcateer`](https://github.com/matthew-mcateer) [![GitHub followers](https://img.shields.io/github/followers/matthew-mcateer?style=social&label=Follow&maxAge=2592000)](https://github.com/matthew-mcateer?tab=followers)<br>[`@pruksmhc`](https://github.com/pruksmhc) [![GitHub followers](https://img.shields.io/github/followers/pruksmhc?style=social&label=Follow&maxAge=2592000)](https://github.com/pruksmhc?tab=followers)<br>[`@shubhobm`](https://github.com/shubhobm) [![GitHub followers](https://img.shields.io/github/followers/shubhobm?style=social&label=Follow&maxAge=2592000)](https://github.com/shubhobm?tab=followers)  |\n",
        "| _Colab_                | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/matthew-mcateer/practicing_trustworthy_machine_learning/blob/main/4_robustness/Chapter_4_Simple_Transparent_Adversarial_Examples.ipynb)                                                                         | Enable GPU<br> Colab Pro recommended                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         |\n",
        "| _Kaggle_               | [ ![ Kaggle ]( https://kaggle.com/static/images/open-in-kaggle.svg ) ]( https://kaggle.com/kernels/welcome?src=https://github.com/matthew-mcateer/practicing_trustworthy_machine_learning/blob/main/4_robustness/Chapter_4_Simple_Transparent_Adversarial_Examples.ipynb )                                                            | Enable GPU<br> Enable Internet                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |\n",
        "| _Gradient_             | [ ![ Gradient ]( https://assets.paperspace.io/img/gradient-badge.svg ) ]( https://console.paperspace.com/github/matthew-mcateer/practicing_trustworthy_machine_learning/blob/main/4_robustness/Chapter_4_Simple_Transparent_Adversarial_Examples.ipynb )                                                                              | Enable GPU                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   |\n",
        "| _SageMaker Studio Lab_ | [![ Open In SageMaker Studio Lab ]( https://studiolab.sagemaker.aws/studiolab.svg ) ]( https://studiolab.sagemaker.aws/import/github/matthew-mcateer/practicing_trustworthy_machine_learning/blob/main/4_robustness/Chapter_4_Simple_Transparent_Adversarial_Examples.ipynb )                                                         | Recommend `ml.g4dn.xlarge` instance at minimum<br>(alternatively, `g4dn.2xlarge` if using EC2 instance)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |\n",
        "| _Binder_               | [ ![ Binder ]( https://mybinder.org/badge_logo.svg ) ]( https://mybinder.org/v2/gh/matthew-mcateer/practicing_trustworthy_machine_learning/HEAD?urlpath=https%3A%2F%2Fgithub.com%2Fmatthew-mcateer%2Fpracticing_trustworthy_machine_learning%2Fblob%2Fmain%2F4_robustness%2FChapter_4_Simple_Transparent_Adversarial_Examples.ipynb ) | May run slowly without GPU                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   |\n",
        "\n",
        "<!--\n",
        "Originally found on GitHub at https://github.com/matthew-mcateer/practicing_trustworthy_machine_learning/blob/main/4_robustness/Chapter_4_Simple_Transparent_Adversarial_Examples.ipynb\n",
        "-->\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "id": "adbc67b7",
      "metadata": {},
      "source": [
        "## ⚙️ Setup"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "id": "TWUW-gGYME8k",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TWUW-gGYME8k",
        "outputId": "483b20d6-aa41-401d-c99d-e6264b0e27c3"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\u001b[K     |████████████████████████████████| 1.4 MB 4.7 MB/s \n",
            "\u001b[K     |████████████████████████████████| 1.6 MB 27.6 MB/s \n",
            "\u001b[?25h  Building wheel for imagenet-stubs (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Author: Practicing Trustworthy machine Learning\n",
            "\n",
            "Last updated: 2022-11-27\n",
            "\n",
            "Python implementation: CPython\n",
            "Python version       : 3.7.15\n",
            "IPython version      : 7.9.0\n",
            "\n",
            "numpy         : 1.21.6\n",
            "tensorflow    : 2.9.2\n",
            "pandas        : 1.3.5\n",
            "imagenet_stubs: 0.0.7\n",
            "art           : 1.12.2\n",
            "textwrap      : unknown\n",
            "\n",
            "Compiler    : GCC 7.5.0\n",
            "OS          : Linux\n",
            "Release     : 5.10.133+\n",
            "Machine     : x86_64\n",
            "Processor   : x86_64\n",
            "CPU cores   : 2\n",
            "Architecture: 64bit\n",
            "\n",
            "Sun Nov 27 08:46:34 2022       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 460.32.03    Driver Version: 460.32.03    CUDA Version: 11.2     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|                               |                      |               MIG M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla T4            Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   52C    P8    10W /  70W |      0MiB / 15109MiB |      0%      Default |\n",
            "|                               |                      |                  N/A |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                                  |\n",
            "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
            "|        ID   ID                                                   Usage      |\n",
            "|=============================================================================|\n",
            "|  No running processes found                                                 |\n",
            "+-----------------------------------------------------------------------------+\n"
          ]
        }
      ],
      "source": [
        "#@title Installing dependencies { display-mode: \"form\" }\n",
        "!pip -qq install watermark adversarial-robustness-toolbox numpy tensorflow pandas git+https://github.com/nottombrown/imagenet_stubs\n",
        "%load_ext watermark\n",
        "%watermark -a \"Practicing Trustworthy machine Learning\" -u -d -v -m -p numpy,tensorflow,pandas,imagenet_stubs,art,textwrap\n",
        "!nvidia-smi"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "id": "QuZJA7j3ME6M",
      "metadata": {
        "id": "QuZJA7j3ME6M"
      },
      "outputs": [],
      "source": [
        "%matplotlib inline\n",
        "\n",
        "import os\n",
        "\n",
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "tf.compat.v1.disable_eager_execution()\n",
        "\n",
        "from tensorflow.keras.preprocessing import image\n",
        "from tensorflow.keras.applications.resnet50 import ResNet50\n",
        "from art.estimators.classification import KerasClassifier"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "hrjdpSmHMK0D",
      "metadata": {
        "id": "hrjdpSmHMK0D"
      },
      "source": [
        "We then load the classifier."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "STPaL2SxME31",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "STPaL2SxME31",
        "outputId": "4ef4a9f0-6771-4849-ae59-1f2c465b1b9a"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.7/dist-packages/keras/layers/normalization/batch_normalization.py:514: _colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Colocations handled automatically by placer.\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5\n",
            "102967424/102967424 [==============================] - 1s 0us/step\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "WARNING:art.estimators.classification.keras:Keras model has no loss set. Classifier tries to use `k.sparse_categorical_crossentropy`.\n"
          ]
        }
      ],
      "source": [
        "#@title model\n",
        "mean_imagenet = np.zeros([224, 224, 3])\n",
        "mean_imagenet[...,0].fill(103.939)\n",
        "mean_imagenet[...,1].fill(116.779)\n",
        "mean_imagenet[...,2].fill(123.68)\n",
        "model = ResNet50(weights='imagenet')\n",
        "classifier = KerasClassifier(\n",
        "    clip_values=(0, 255), \n",
        "    model=model, \n",
        "    preprocessing=(mean_imagenet, np.ones([224, 224, 3]))\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "MD0b_GT_MRvz",
      "metadata": {
        "id": "MD0b_GT_MRvz"
      },
      "source": [
        "We load the data now. Make sure to install [Imagenet Stubs](https://github.com/nottombrown/imagenet-stubs) before proceeding."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "id": "0MY0qkWFME1d",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0MY0qkWFME1d",
        "outputId": "f2feb705-0714-4103-a4e4-9f28efd669d5"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/engine/training_v1.py:2067: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n",
            "  updates=self.state_updates,\n"
          ]
        }
      ],
      "source": [
        "# @title data\n",
        "import pandas as pd\n",
        "from imagenet_stubs import get_image_paths\n",
        "from imagenet_stubs.imagenet_2012_labels import label_to_name, name_to_label\n",
        "\n",
        "all_paths = get_image_paths()\n",
        "all_imgs = []\n",
        "for path in all_paths:\n",
        "    img = image.load_img(path, target_size=(224, 224))\n",
        "    img = image.img_to_array(img)\n",
        "    all_imgs.append(img)\n",
        "all_names = [os.path.basename(path) for path in all_paths]\n",
        "all_probs = [np.max(classifier.predict(np.array([img]))) for img in all_imgs]\n",
        "all_labels = [\n",
        "    np.argmax(classifier.predict(np.array([img]))) for img in all_imgs\n",
        "]\n",
        "\n",
        "img_data = pd.DataFrame(\n",
        "    {\n",
        "        \"name\": [os.path.splitext(name)[0] for name in all_names],\n",
        "        \"label\": [label_to_name(label) for label in all_labels],\n",
        "        \"prob\": [round(p, 3) for p in all_probs],\n",
        "        \"img\": all_imgs,\n",
        "    }\n",
        ").set_index(\"name\")\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "b3PypJKPMaXR",
      "metadata": {
        "id": "b3PypJKPMaXR"
      },
      "source": [
        "Before proceeding, let us check the majority class probabilities for each image predicted by our classifier."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "id": "td5TNFaEMEzI",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 551
        },
        "id": "td5TNFaEMEzI",
        "outputId": "f01ed677-79e5-4bd0-b0a6-c8e6cb86818b"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-34a1657d-b25d-4c87-98d5-9dd132d43cf0\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>name</th>\n",
              "      <th>label</th>\n",
              "      <th>prob</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>beagle</td>\n",
              "      <td>beagle</td>\n",
              "      <td>0.476</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>malamute</td>\n",
              "      <td>Eskimo dog, husky</td>\n",
              "      <td>0.517</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>marmoset</td>\n",
              "      <td>titi, titi monkey</td>\n",
              "      <td>0.536</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>standard_poodle</td>\n",
              "      <td>standard poodle</td>\n",
              "      <td>0.537</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>tractor</td>\n",
              "      <td>tractor</td>\n",
              "      <td>0.756</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>notebook_computer</td>\n",
              "      <td>notebook, notebook computer</td>\n",
              "      <td>0.891</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>sleeping_bag</td>\n",
              "      <td>sleeping bag</td>\n",
              "      <td>0.944</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>unicycle</td>\n",
              "      <td>unicycle, monocycle</td>\n",
              "      <td>0.960</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>flagpole</td>\n",
              "      <td>flagpole, flagstaff</td>\n",
              "      <td>0.981</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>rock_crab</td>\n",
              "      <td>rock crab, Cancer irroratus</td>\n",
              "      <td>0.982</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>gazelle</td>\n",
              "      <td>gazelle</td>\n",
              "      <td>0.996</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>centipede</td>\n",
              "      <td>centipede</td>\n",
              "      <td>0.996</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>bagel</td>\n",
              "      <td>bagel, beigel</td>\n",
              "      <td>0.998</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>mitten</td>\n",
              "      <td>mitten</td>\n",
              "      <td>0.998</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>koala</td>\n",
              "      <td>koala, koala bear, kangaroo bear, native bear,...</td>\n",
              "      <td>0.999</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>manhole_cover</td>\n",
              "      <td>manhole cover</td>\n",
              "      <td>1.000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-34a1657d-b25d-4c87-98d5-9dd132d43cf0')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-34a1657d-b25d-4c87-98d5-9dd132d43cf0 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-34a1657d-b25d-4c87-98d5-9dd132d43cf0');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "                 name                                              label  \\\n",
              "0              beagle                                             beagle   \n",
              "1            malamute                                  Eskimo dog, husky   \n",
              "2            marmoset                                  titi, titi monkey   \n",
              "3     standard_poodle                                    standard poodle   \n",
              "4             tractor                                            tractor   \n",
              "5   notebook_computer                        notebook, notebook computer   \n",
              "6        sleeping_bag                                       sleeping bag   \n",
              "7            unicycle                                unicycle, monocycle   \n",
              "8            flagpole                                flagpole, flagstaff   \n",
              "9           rock_crab                        rock crab, Cancer irroratus   \n",
              "10            gazelle                                            gazelle   \n",
              "11          centipede                                          centipede   \n",
              "12              bagel                                      bagel, beigel   \n",
              "13             mitten                                             mitten   \n",
              "14              koala  koala, koala bear, kangaroo bear, native bear,...   \n",
              "15      manhole_cover                                      manhole cover   \n",
              "\n",
              "     prob  \n",
              "0   0.476  \n",
              "1   0.517  \n",
              "2   0.536  \n",
              "3   0.537  \n",
              "4   0.756  \n",
              "5   0.891  \n",
              "6   0.944  \n",
              "7   0.960  \n",
              "8   0.981  \n",
              "9   0.982  \n",
              "10  0.996  \n",
              "11  0.996  \n",
              "12  0.998  \n",
              "13  0.998  \n",
              "14  0.999  \n",
              "15  1.000  "
            ]
          },
          "execution_count": 5,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "img_data[['label','prob']].sort_values(['prob']).reset_index()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "p41-sYFmMeq6",
      "metadata": {
        "id": "p41-sYFmMeq6"
      },
      "source": [
        "We define `SimpleTransparentExamples` as our attack class, then apply the attack through setting 1 and 2."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "id": "EEKlzjRoMExH",
      "metadata": {
        "id": "EEKlzjRoMExH"
      },
      "outputs": [],
      "source": [
        "#@title simple transparent adversarial examples\n",
        "from PIL import Image, ImageDraw, ImageFont\n",
        "\n",
        "!wget -q -O sans_serif.ttf https://github.com/matthew-mcateer/practicing_trustworthy_machine_learning/raw/main/4_robustness/sans_serif.ttf\n",
        "\n",
        "class SimpleTransparentExamples:\n",
        "    def __init__(\n",
        "        self, image, text, font=ImageFont.truetype(\"sans_serif.ttf\", 16)\n",
        "    ):\n",
        "        self.image = image\n",
        "        self.text = text\n",
        "        self.font = font\n",
        "\n",
        "    def generate(\n",
        "        self, alpha=0.3, angle=0, x=0, y=0, image_init=None, resume=False\n",
        "    ):\n",
        "        # watermark\n",
        "        opacity = int(256 * alpha)\n",
        "        mark_width, mark_height = self.font.getsize(self.text)\n",
        "        patch = Image.new(\"RGBA\", (mark_width, mark_height), (0, 0, 0, 0))\n",
        "        ImageDraw.Draw(patch).text(\n",
        "            (0, 0), text=self.text, font=self.font, fill=(255, 0, 0, opacity)\n",
        "        )\n",
        "        patch = patch.rotate(angle, expand=1)\n",
        "\n",
        "        # merge\n",
        "        wx, wy = patch.size\n",
        "        if resume == True:\n",
        "            img = image.array_to_img(image_init)\n",
        "        else:\n",
        "            img = self.image.copy()\n",
        "        img.paste(patch, (x, y, x + wx, y + wy), patch)\n",
        "        img = image.img_to_array(img)\n",
        "        return img\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "id": "TS_F1uqbMEuu",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TS_F1uqbMEuu",
        "outputId": "996cda6c-32b4-46cd-a66a-e179a159907c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[0, 0, 'beagle']\n",
            "[0, 20, 'beagle']\n",
            "[0, 40, 'beagle']\n",
            "[0, 60, 'beagle']\n",
            "[0, 80, 'beagle']\n",
            "[0, 100, 'beagle']\n",
            "[0, 120, 'beagle']\n",
            "[0, 140, 'beagle']\n",
            "[0, 160, 'beagle']\n",
            "[0, 180, 'beagle']\n",
            "[0, 200, 'beagle']\n",
            "[0, 220, 'beagle']\n",
            "[20, 0, 'beagle']\n",
            "[20, 20, 'beagle']\n",
            "[20, 40, 'bluetick']\n",
            "[20, 60, 'beagle']\n",
            "[20, 80, 'beagle']\n",
            "[20, 100, 'beagle']\n",
            "[20, 120, 'beagle']\n",
            "[20, 140, 'beagle']\n",
            "[20, 160, 'bluetick']\n",
            "[20, 180, 'beagle']\n",
            "[20, 200, 'beagle']\n",
            "[20, 220, 'beagle']\n",
            "[40, 0, 'beagle']\n",
            "[40, 20, 'bluetick']\n",
            "[40, 40, 'beagle']\n",
            "[40, 60, 'beagle']\n",
            "[40, 80, 'beagle']\n",
            "[40, 100, 'beagle']\n",
            "[40, 120, 'beagle']\n",
            "[40, 140, 'beagle']\n",
            "[40, 160, 'beagle']\n",
            "[40, 180, 'beagle']\n",
            "[40, 200, 'beagle']\n",
            "[40, 220, 'beagle']\n",
            "[60, 0, 'beagle']\n",
            "[60, 20, 'beagle']\n",
            "[60, 40, 'beagle']\n",
            "[60, 60, 'beagle']\n",
            "[60, 80, 'beagle']\n",
            "[60, 100, 'beagle']\n",
            "[60, 120, 'beagle']\n",
            "[60, 140, 'beagle']\n",
            "[60, 160, 'beagle']\n",
            "[60, 180, 'beagle']\n",
            "[60, 200, 'beagle']\n",
            "[60, 220, 'beagle']\n",
            "[80, 0, 'beagle']\n",
            "[80, 20, 'beagle']\n",
            "[80, 40, 'beagle']\n",
            "[80, 60, 'beagle']\n",
            "[80, 80, 'beagle']\n",
            "[80, 100, 'beagle']\n",
            "[80, 120, 'beagle']\n",
            "[80, 140, 'beagle']\n",
            "[80, 160, 'beagle']\n",
            "[80, 180, 'beagle']\n",
            "[80, 200, 'beagle']\n",
            "[80, 220, 'beagle']\n",
            "[100, 0, 'beagle']\n",
            "[100, 20, 'beagle']\n",
            "[100, 40, 'beagle']\n",
            "[100, 60, 'beagle']\n",
            "[100, 80, 'beagle']\n",
            "[100, 100, 'beagle']\n",
            "[100, 120, 'beagle']\n",
            "[100, 140, 'beagle']\n",
            "[100, 160, 'beagle']\n",
            "[100, 180, 'beagle']\n",
            "[100, 200, 'beagle']\n",
            "[100, 220, 'beagle']\n",
            "[120, 0, 'beagle']\n",
            "[120, 20, 'beagle']\n",
            "[120, 40, 'beagle']\n",
            "[120, 60, 'beagle']\n",
            "[120, 80, 'beagle']\n",
            "[120, 100, 'beagle']\n",
            "[120, 120, 'beagle']\n",
            "[120, 140, 'beagle']\n",
            "[120, 160, 'beagle']\n",
            "[120, 180, 'beagle']\n",
            "[120, 200, 'beagle']\n",
            "[120, 220, 'beagle']\n",
            "[140, 0, 'beagle']\n",
            "[140, 20, 'beagle']\n",
            "[140, 40, 'beagle']\n",
            "[140, 60, 'beagle']\n",
            "[140, 80, 'beagle']\n",
            "[140, 100, 'beagle']\n",
            "[140, 120, 'beagle']\n",
            "[140, 140, 'beagle']\n",
            "[140, 160, 'beagle']\n",
            "[140, 180, 'beagle']\n",
            "[140, 200, 'beagle']\n",
            "[140, 220, 'beagle']\n",
            "[160, 0, 'beagle']\n",
            "[160, 20, 'beagle']\n",
            "[160, 40, 'beagle']\n",
            "[160, 60, 'beagle']\n",
            "[160, 80, 'beagle']\n",
            "[160, 100, 'beagle']\n",
            "[160, 120, 'beagle']\n",
            "[160, 140, 'beagle']\n",
            "[160, 160, 'beagle']\n",
            "[160, 180, 'beagle']\n",
            "[160, 200, 'beagle']\n",
            "[160, 220, 'beagle']\n",
            "[180, 0, 'beagle']\n",
            "[180, 20, 'beagle']\n",
            "[180, 40, 'beagle']\n",
            "[180, 60, 'beagle']\n",
            "[180, 80, 'beagle']\n",
            "[180, 100, 'beagle']\n",
            "[180, 120, 'beagle']\n",
            "[180, 140, 'beagle']\n",
            "[180, 160, 'beagle']\n",
            "[180, 180, 'beagle']\n",
            "[180, 200, 'beagle']\n",
            "[180, 220, 'beagle']\n",
            "[200, 0, 'beagle']\n",
            "[200, 20, 'beagle']\n",
            "[200, 40, 'beagle']\n",
            "[200, 60, 'beagle']\n",
            "[200, 80, 'beagle']\n",
            "[200, 100, 'beagle']\n",
            "[200, 120, 'beagle']\n",
            "[200, 140, 'beagle']\n",
            "[200, 160, 'beagle']\n",
            "[200, 180, 'beagle']\n",
            "[200, 200, 'beagle']\n",
            "[200, 220, 'beagle']\n",
            "[220, 0, 'beagle']\n",
            "[220, 20, 'beagle']\n",
            "[220, 40, 'beagle']\n",
            "[220, 60, 'beagle']\n",
            "[220, 80, 'beagle']\n",
            "[220, 100, 'beagle']\n",
            "[220, 120, 'beagle']\n",
            "[220, 140, 'beagle']\n",
            "[220, 160, 'beagle']\n",
            "[220, 180, 'beagle']\n",
            "[220, 200, 'beagle']\n",
            "[220, 220, 'beagle']\n"
          ]
        }
      ],
      "source": [
        "#@title setting 1: alpha=.1 (NO repeat text)\n",
        "nm = \"beagle\" #@param\n",
        "font_size_1 = 16 #@param\n",
        "font = ImageFont.truetype(\"sans_serif.ttf\", font_size_1)\n",
        "patch_text = \"Hello World\" #@param\n",
        "alpha = 0.1 #@param\n",
        "angle = 30 #@param\n",
        "\n",
        "# load images\n",
        "all_raw = [image.load_img(path, target_size=(224, 224)) for path in all_paths]\n",
        "raw_data = pd.DataFrame(\n",
        "    {\"name\": [os.path.splitext(name)[0] for name in all_names], \"img\": all_raw}\n",
        ").set_index(\"name\")\n",
        "\n",
        "im = raw_data.loc[nm, \"img\"].copy()\n",
        "\n",
        "si = SimpleTransparentExamples(image=im, text=patch_text, font=font)\n",
        "xlist = []\n",
        "ylist = []\n",
        "imlist = []\n",
        "for x in np.arange(0, 224, 20):\n",
        "    for y in np.arange(0, 224, 20):\n",
        "        img = si.generate(alpha=alpha, angle=angle, x=x, y=y, resume=False)\n",
        "        xlist.append(x)\n",
        "        ylist.append(y)\n",
        "        imlist.append(img)\n",
        "\n",
        "        print(\n",
        "            [\n",
        "                x,\n",
        "                y,\n",
        "                label_to_name(np.argmax(classifier.predict(np.array([img])))),\n",
        "            ]\n",
        "        )\n",
        "\n",
        "outputs_s1 = pd.DataFrame({\"x\": xlist, \"y\": ylist, \"img\": imlist})"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "id": "F08GVaKPMEsU",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "F08GVaKPMEsU",
        "outputId": "b24932f9-ef94-4111-9173-7caef8ce4580"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[0, 0, 'beagle']\n",
            "[0, 20, 'beagle']\n",
            "[0, 40, 'beagle']\n",
            "[0, 60, 'beagle']\n",
            "[0, 80, 'beagle']\n",
            "[0, 100, 'beagle']\n",
            "[0, 120, 'beagle']\n",
            "[0, 140, 'beagle']\n",
            "[0, 160, 'beagle']\n",
            "[0, 180, 'beagle']\n",
            "[0, 200, 'beagle']\n",
            "[0, 220, 'beagle']\n",
            "[20, 0, 'beagle']\n",
            "[20, 20, 'beagle']\n",
            "[20, 40, 'beagle']\n",
            "[20, 60, 'beagle']\n",
            "[20, 80, 'beagle']\n",
            "[20, 100, 'beagle']\n",
            "[20, 120, 'beagle']\n",
            "[20, 140, 'beagle']\n",
            "[20, 160, 'beagle']\n",
            "[20, 180, 'beagle']\n",
            "[20, 200, 'beagle']\n",
            "[20, 220, 'beagle']\n",
            "[40, 0, 'beagle']\n",
            "[40, 20, 'beagle']\n",
            "[40, 40, 'beagle']\n",
            "[40, 60, 'beagle']\n",
            "[40, 80, 'beagle']\n",
            "[40, 100, 'beagle']\n",
            "[40, 120, 'beagle']\n",
            "[40, 140, 'beagle']\n",
            "[40, 160, 'beagle']\n",
            "[40, 180, 'beagle']\n",
            "[40, 200, 'beagle']\n",
            "[40, 220, 'beagle']\n",
            "[60, 0, 'beagle']\n",
            "[60, 20, 'beagle']\n",
            "[60, 40, 'bluetick']\n",
            "[60, 60, 'bluetick']\n",
            "[60, 80, 'bluetick']\n",
            "[60, 100, 'bluetick']\n",
            "[60, 120, 'bluetick']\n",
            "[60, 140, 'bluetick']\n",
            "[60, 160, 'bluetick']\n",
            "[60, 180, 'bluetick']\n",
            "[60, 200, 'bluetick']\n",
            "[60, 220, 'bluetick']\n",
            "[80, 0, 'bluetick']\n",
            "[80, 20, 'bluetick']\n",
            "[80, 40, 'bluetick']\n",
            "[80, 60, 'bluetick']\n",
            "[80, 80, 'bluetick']\n",
            "[80, 100, 'bluetick']\n",
            "[80, 120, 'bluetick']\n",
            "[80, 140, 'bluetick']\n",
            "[80, 160, 'bluetick']\n",
            "[80, 180, 'bluetick']\n",
            "[80, 200, 'bluetick']\n",
            "[80, 220, 'bluetick']\n",
            "[100, 0, 'bluetick']\n",
            "[100, 20, 'bluetick']\n",
            "[100, 40, 'bluetick']\n",
            "[100, 60, 'bluetick']\n",
            "[100, 80, 'bluetick']\n",
            "[100, 100, 'bluetick']\n",
            "[100, 120, 'bluetick']\n",
            "[100, 140, 'bluetick']\n",
            "[100, 160, 'bluetick']\n",
            "[100, 180, 'bluetick']\n",
            "[100, 200, 'bluetick']\n",
            "[100, 220, 'bluetick']\n",
            "[120, 0, 'bluetick']\n",
            "[120, 20, 'bluetick']\n",
            "[120, 40, 'bluetick']\n",
            "[120, 60, 'bluetick']\n",
            "[120, 80, 'bluetick']\n",
            "[120, 100, 'bluetick']\n",
            "[120, 120, 'bluetick']\n",
            "[120, 140, 'bluetick']\n",
            "[120, 160, 'bluetick']\n",
            "[120, 180, 'bluetick']\n",
            "[120, 200, 'bluetick']\n",
            "[120, 220, 'bluetick']\n",
            "[140, 0, 'bluetick']\n",
            "[140, 20, 'bluetick']\n",
            "[140, 40, 'bluetick']\n",
            "[140, 60, 'bluetick']\n",
            "[140, 80, 'bluetick']\n",
            "[140, 100, 'bluetick']\n",
            "[140, 120, 'bluetick']\n",
            "[140, 140, 'bluetick']\n",
            "[140, 160, 'bluetick']\n",
            "[140, 180, 'bluetick']\n",
            "[140, 200, 'bluetick']\n",
            "[140, 220, 'bluetick']\n",
            "[160, 0, 'bluetick']\n",
            "[160, 20, 'bluetick']\n",
            "[160, 40, 'bluetick']\n",
            "[160, 60, 'bluetick']\n",
            "[160, 80, 'bluetick']\n",
            "[160, 100, 'bluetick']\n",
            "[160, 120, 'bluetick']\n",
            "[160, 140, 'bluetick']\n",
            "[160, 160, 'bluetick']\n",
            "[160, 180, 'bluetick']\n",
            "[160, 200, 'bluetick']\n",
            "[160, 220, 'bluetick']\n",
            "[180, 0, 'bluetick']\n",
            "[180, 20, 'bluetick']\n",
            "[180, 40, 'bluetick']\n",
            "[180, 60, 'bluetick']\n",
            "[180, 80, 'bluetick']\n",
            "[180, 100, 'bluetick']\n",
            "[180, 120, 'bluetick']\n",
            "[180, 140, 'bluetick']\n",
            "[180, 160, 'bluetick']\n",
            "[180, 180, 'bluetick']\n",
            "[180, 200, 'bluetick']\n",
            "[180, 220, 'bluetick']\n",
            "[200, 0, 'bluetick']\n",
            "[200, 20, 'bluetick']\n",
            "[200, 40, 'bluetick']\n",
            "[200, 60, 'bluetick']\n",
            "[200, 80, 'bluetick']\n",
            "[200, 100, 'bluetick']\n",
            "[200, 120, 'bluetick']\n",
            "[200, 140, 'bluetick']\n",
            "[200, 160, 'bluetick']\n",
            "[200, 180, 'bluetick']\n",
            "[200, 200, 'bluetick']\n",
            "[200, 220, 'bluetick']\n",
            "[220, 0, 'bluetick']\n",
            "[220, 20, 'bluetick']\n",
            "[220, 40, 'bluetick']\n",
            "[220, 60, 'bluetick']\n",
            "[220, 80, 'bluetick']\n",
            "[220, 100, 'bluetick']\n",
            "[220, 120, 'bluetick']\n",
            "[220, 140, 'bluetick']\n",
            "[220, 160, 'bluetick']\n",
            "[220, 180, 'bluetick']\n",
            "[220, 200, 'bluetick']\n",
            "[220, 220, 'bluetick']\n"
          ]
        }
      ],
      "source": [
        "#@title setting 2: alpha=.05 (repeat text)\n",
        "font_size_2 = 8 #@param\n",
        "font = ImageFont.truetype(\"sans_serif.ttf\", font_size_2)\n",
        "alpha = 0.05 #@param\n",
        "im = raw_data.loc[nm, \"img\"].copy()\n",
        "\n",
        "si = SimpleTransparentExamples(image=im, text=patch_text, font=font)\n",
        "xlist = []\n",
        "ylist = []\n",
        "imlist = []\n",
        "for x in np.arange(0, 224, 20):\n",
        "    for y in np.arange(0, 224, 20):\n",
        "        img = si.generate(\n",
        "            alpha=alpha, angle=angle, x=x, y=y, resume=True, image_init=img\n",
        "        )\n",
        "        xlist.append(x)\n",
        "        ylist.append(y)\n",
        "        imlist.append(img)\n",
        "\n",
        "        print(\n",
        "            [\n",
        "                x,\n",
        "                y,\n",
        "                label_to_name(np.argmax(classifier.predict(np.array([img])))),\n",
        "            ]\n",
        "        )\n",
        "\n",
        "outputs_s2 = pd.DataFrame({\"x\": xlist, \"y\": ylist, \"img\": imlist})"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "id": "Bs3e3fRQMERH",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 330
        },
        "id": "Bs3e3fRQMERH",
        "outputId": "c0e6d470-174e-4cc7-e6e3-32dec306f52c"
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAz8AAAE5CAYAAAC+k553AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9d7ymy1Hf+a1+nvc958zMvVdXV4GsSxDGZINZFkwQJof1LkGwxhgEGLAJRssSBRhhmwUMJsomWwIMSzLBpGWBtYhrQMIEIRYZgUSQBJKQru7MnTnnfZ6u/aOquvt5z5kc3ztd83nnnPOEDtVVv367qrpaVJVOnTp16tSpU6dOnTp1erRTut0N6NSpU6dOnTp16tSpU6dbQX3x06lTp06dOnXq1KlTp7uC+uKnU6dOnTp16tSpU6dOdwX1xU+nTp06derUqVOnTp3uCuqLn06dOnXq1KlTp06dOt0V1Bc/nTp16tSpU6dOnTp1uiuoL346derUqVOnTp06dep0V1Bf/HTq1KlTp06dOnXq1OmuoL746dSpU6dOnTp16tSp011BffHT6YaTiDxTRPQa332aiKiIPHhjW7Wo40Gv42k3q45One5kEpHniMhLbnc7biWJyHNF5Lk3uY6OLZ12ku5GTLgbSUSe4hj1lNvdlttJffHTaUEi8jYi8h9F5C9F5FBEXiYi3ycib3O729ap06OBROTtRORHROSlInLBde3nReSzrrG8jxWRp59w/Q3cEPGO19/qG0ci8j+IyL8XkeeLyOZaDSV3A4nIKR/Dp9yCuk6Uo043n+5mTBCR5EbP/ywify4i50TkBSLyJSKyf7vbd7tIRJ4hIv/Lbaz/3V1WHnOT67llGNdSX/x0KiQiHwH8NvC+wLOBTwe+C3gf4LdF5MOvsKh/DRxcYzO+19996TW+36nTHUsi8u7A84B3AL4D+EzgO4EMfPY1FvuxwElfWt8A+DLgpC86nwL8rWus73rpQ4B/AijwJ7ew3g/wzy7RKWwMn3IL6rqYHHW6idQxgVPY943HA9+Ktfs3gS8HflZE5Da06U6gZwC3bfEDvDsmKzd18cOtxbhC462srNOdSyLy5tjC40+A91LVVzb3vhH4FeB7ReTtVfXELywiclpVz6nqBEzX0g5VnYH5Wt7t1GkH6IuBh4B3UdXXtjdE5Am3qhGqurlVdZ1A3wJ8taqeF5FnAW95KypV1aNbUU+nTldJdzsmHAF/T1V/vbn2HR6C9+WYMfYXbkRF8R3lRpTVabepe346BX0etgL/1HbhA6CqrwI+DTgNfD7UfT0i8tYi8v0i8hrgV9t7bRkiciAi3yQirxKRh93F/YZexjOb547t+RGRl4jIT4nIe4jIb3pYwJ+IyMdv1fFYEflaEfl9ETkrIq8TkZ8VkXe4gXzq1Ol66M2BP9j+kgOgqn+9fU1EPs7Dw86LyN+IyA+IyBs3958LfCjwJNcbdX15CvBb/tizm3tP8/cW8f1S96p8roh8qoi82MNef0tE3uWEdj1VRF7ouvgCEfnw7TIvRqr6V6p6/nLPXQ2JyOuJyLNF5C+83S8XkZ/YwpHFnh+pse8fLSJf7O9eEJFfFJG3OKGOz3DcOe849J7bZV6ifW8lFtb0N17H80TkH1zmnQeBwOIva8bwmVdarog8QURe6e2U5vpbiIUX/WDwhhPk6HL96nRD6K7GBFU92lr4BP2Y//zbl3r/YuR1nxWRNxeRnxGRh4Hv83tJRJ4uIn/g7f0rEfk2Ebl/q4z47vEBIvI7/uwLxaJktut7jIh8g1jo3qGI/LGIfIGIpK3nPldEfl1EXu1j+HwR+aitZxT7vvUJzTg9p7n/hiLyH7zdh96PTzqhTW8kIj/uuv7XIvL1wN4V8O6ZwNf4n3/atOHB5pnLyeEn+juftFX2M/z6h1wJxt0s6p6fTkH/E/ASVf2Vk26q6i87iH3o1q0fBv475qK9lHv6OcBHY96l/wq8N/DTV9G+twB+BAvD+27gk4DniMjzVfUP/Jk3w9zEPwz8KfBEbNH2SyLy1qr6squor1Onm0EvBd5NRN5WVV9wqQdF5IuBfwX8EBYG83jgs4BfFpG/41+WvgK4D3gj4H/zV88Cfwj8C+BfAt+OeW4BTvqS0dLHAvcA34aFpX0+8KMi8mZhGRaRDwV+EPh94IuA+zG9/MsrYcBNov8EvA3wzcBLgCcA7w+8if99KfpCLMToazFefj72Jeld4wER+WfAszA+fj3wIPDjwGuAv7hU4WL7JX8N489XAecwLPxxEflIVf2xi7z6SuCfYZ6yHwN+1K//3pWWq6p/7W3/YUx2vsm/jD0HeBgLbYaLy1Gnm08dE06m1/Ofr7qOMkbg5zDD7OcCj/j1bwOehoXbfRPwpli44d8Rkb+35QV7Mta3b8W+e3wi8MMi8kGq+vNg+1aAXwLe0Mv+Myxs7CuB12cZgvjZwH/GMGYN/K9e3oepanwn+sfY+P4mNlYAL/a6noh9h1IMk14JfDDwXSJyr6p+gz93APwihoHfBLzMy/37V8C3H8U88v8Qk6EYg1d62ZeVQ1V9ti8Sv05Efl5V/1xE3g4LcfsuVf0ZETnNJTDuppKq9s9d/sGAUoEfv8xzP+HP3QM803///hOee6aJVvn7nfzZr9967tl+/ZnNtaf5tQebay/xa+/ZXHs8cAH42ubaHpC26njQn/vSrWsKPO12875/7q4P9oU8wkJ/HfhqbB/Kauu5J/kzz9i6/rbApr0O/BRmuNiu6+9eTM6xL78vaf4OnXgVcH9z/R/49Q9rrv0e8OfAmebae/tzx9pxGX48q8WKa+TpY7zuz73Mc88Fntv8/RR/74XAurn+z/362/rfa+fLbwJj89wn+HNtmcewBQvZ+T1gr7km2MLlRZdp8+O2MfJaygW+H1scPRn7EqjA/7z1zIly1D8399Mx4aJ8+XksHPAx1/j+c7z+r9y6/h5+/WO3rn/g9nXqd4+PaK7diy0kfru59iXYAvPJW2V+pY/ZGzfXDraeWWGLxl/cun4WeM4J/fpOr/+Brev/J/DaKB9bZCnw1OaZU5ixWoGnXIZ/gRMPXoccvh7wauD/xnD0t7HF/r3NMxfFuJv56WFvncAWM2CWwEtR3L+3ufatV1D+B/nPf791/Zuv4N2gF2rjlVILzfsjzNsT1w5VNQOIyCAiD2AA8kfYAqxTp9tKapbCd8Msf++AWVF/DvhLWYZBfQQWlvxDIvK4+ACvwCav97lJTfxBVX1N83fo3JuBZYsC3g74HlUtngFV/SVsAr8ddB7bN/CU7bCVK6Rn63I/0KLP2BfGB4DvUNvPGPR9mOfnoiQij8UsrT8E3NOM4wPYuD9ZRN7waht8DeV+JvZF8kcwi+33qupPXG29nW48dUw4TiLyDOD9gC/UE8IBr5K+Zevvp2K68PNbfHw+9n1hm48vo4bgoaqvA74H8xK9XlPmrwCv2SrzF4ABeK/m/RLy63h1n7972e8oIiLARwI/6X+2df2clxXlfAjwckzno+5HqJ6ka6UrlkNVfQXwGdgC/1ewRBuf5Dy8rdTD3jpBXdTcc8mnTl4k/ekVlP8kLKxk+9k/voJ3g/7shGuvwdzrgMXxYtaOT8fc2EPz7Kuvoq5OnW4aqepvAR8hImvsy86HY6EFPyIi76iqL8Qs9IJNJifRzdqcvNAzVX2NzbdFz57kP0/S3T/mNhgZVPVQRL4A+LfAX4nIf8Us39/jk+/laBtb4oveJfusqpNcfl/MW2Dj+K/8cxI9gasPD7qqclX1b0Tkn2Phb3+Febc63SHUMaGSiHwMljH2u1R1e+FytTRxPCz1ydgi4dh+KqftJBN/rO6iaOhF/vNB7Ev/k4G3p+5fuWiZIvJhmKfoHVnuv9mu4yR6PObp/lT/XKquJ12k7X90BfVciq5KDlX1B0Tk47AtE9+uqr94nfXfEOqLn06o6kMi8nJMeS9Fbw/8paq+Ture2Ru6cfkSdLEMcO0+o2dgXwT+A/ClwN9gi65voCf36HSHkXsbfgv4LRF5ERYG+lQsw1HCJsMP5mTZv1n7Ma5Ez+44UtVvEJGfxPb8fSCGA18kIn9fVf/bZV6/mX0O3PlazDJ7El2NEeh6yv1A/3k/th/kei3qnW4w3e2YICLvj3lVfhr4pzegyBIN0lDCFj7/6CLvXGwBcylKWJjev7nI/RcBiMh7Yh6+X8aMtC/HFgufiO2tupJ6AP4jtv/oJLrZ+2WuSg49Aufv+p9vLSLphDG55dQXP52Cfgr4FBF5D1X91e2brrQPYpv5rpZeiinMm7K0FhzLqHSd9FHAf1HVT24vih3SdT2bJjt1utn0PP/5+v7zxdiXiz9V1Red/Eqhi1kMr8SSeLUU52+dpLs3Wp+vilT1xZj359+KyJOB3wH+d+DjrrPots//JS6KyIhh4qW+bMSxABtVvZZ0vRcbw6sqV0Q+CDtb6d9gX/q+W0TedSuM72bIS6drp7sKE0TkXbHwsucBH70lmzeSXoyF1P2aXlnWybcQEdnyoER6/pc0ZZ65Al38SGwP8geq6mFcFJFPPOHZk8bqlVjkzXAFdb0UeNsT2n6lZzldTFauRg4B/h0WNfRF2B6opwNfdwX13FTq1vBOQV+DeXG+zVfqhTy+/FuxTClfc8K7l6OwTH761vVrOr36EjSzZY0SkadiGVg6dbrtJCLvI3LioX0f4j8jJOFHMXn+su3nxajV0XNYGMc2xXkWN+yQOrWMiS8APl5EzjRtem8s7v+Wk9gJ4dsnwb8Y+5Jw2bSuV0DPw8JmP8UXPEH/iCbs9iRSS1X8XODTROT1t++LyOMvU3dkp1qM4dWU68afyBz1DGwR9E7+e0sXk6NON5E6JoCI/G3M2/MSLJHCzYwo+SEsJP5LT2jH6PrS0htgYYjxzL3AxwO/04TV/hCWse8Dt96NFNiBGzP2ZX9o7j/IyYeZnuO43s9YZsuPFJG3PaGuFk9+xtv+Uc39U1w8XO6k+tluA1chh2IpvD8G27v1VcAPAP9aRNqz3U7EuJtN3fPTCQBV/e8i8gnYJt7fF5HvwvboPAh8MpaR4x+6dfVqy36+iPwn4OmuGJHqOhTgRq38fwr4FyLybCxrztthX1Bu5SnynTpdir4ZOCUiPwb8f1gGnHfHJoiXYGEuqOqLReRLMEvZgyLy49iX+TfFJuJvx0KewDbqfoyIfB0WMnNWVX8SWwC8FvinYmdcnAN+Q1WvZJ/epegZWObHX3Ndux/bUP8C4MylXgQQkSdhKVfBwyG8rwAvVdXvbZ59LvDeqnqpEJu3BH5RRH4Iy9w2YTx6IjbZXhep6pHYuRPfDPw/Xs+DWGbKF3N5/PoMLNXu74vId2B49ERsk/sbYXs8Llb3eRF5ITa+L8JCeV+glhL5Ssv9RiwRwvv5l6f/S0S+E/gSEfkJVf1df+5ictTp5tJdjQkicg9mIL0fM65+6NZ36her6v/bPP9cLo8JFyVV/SUR+TYsLPYdsUxkG2wvy1OxfcM/0rzyIiyN9Ltg++U+CdOz1lvzNVgWvJ8SO4/n+dg5PW+HLT4exKJPfhr4HEwHvx/bn/MZWIjq9raD5wPvJyKfgyVd+FNV/Q0sNf/7AL/hev9C4LGYQeP9/HeA78DG4HtE5J2xELt/TF1sXI6e7z+/QkR+wHn0k1cqh2IH9H4L5i1/lpf1md7253iUUb4Mxt08upmp5Ppn9z6Ysn4/pmxHmMJ8P572tXnumdik/7gTyngmW+lrsRSLz8IsqA9j7u239DK+oHnuaZyc6vqnTqjnuSzTzO5h4P8yTMF/FfgfT3juQXqq6/65DR8s8+F3YWduPAwcYqGg3wQ84YTnPwLLknOWelbHs4C3bJ45Tc08tkgti03If4BNXEXmuXha22PpojkhDSn2xewPsRCO38fOCfsR4A+vgAdP8TJP+jx369nnAS+/THkPOE/+0Hn0WszA8tSt57ZxINrxUVvPnYgPmKf6Jd7n38C+oD4P+NkrePfNsBj9l2O4+hdYxqaPvAJ+vZvXc7g9Fpcrl5qW+HO2yrzH+/I7eErlS8lR/9y8D3c5JjT1XOzznK3nL4sJTX/OXuL+p3hZjwCvw8JXvxp4/eaZl2BG1Q8Aftf79odsYYY/ewb4P3zsDrEQtV/DQm9XzXOfhC2ooqyncfJ3pr+FnR30yDYfsEXTs7BkFPE97ReAT9kq402wRek5b883UFN6P+UKePglGKaEx+rBK5VDzEP1OuBJW2UGJn1+c+2iGHezPuIVd+p0y8mtLv8N+DhV/b7b3Z5OnTpdO4nI7wCvVNX3v0Hl3YNZAZ+uqv/uRpR5I0ksu+QrgR9V1U+53e3p1OlOo13HBLFsji9Q1Q+72XV1urXU9/x0uiUkdtrwNj0dy8b2y7e4OZ06dbpGEpHV1t4XROQpWJjVc29gVe+FpWr+jhtY5jWRiOyfsC/j47EQk+fe+hZ16nTn0N2ICZ12m7rnp9MtIRH5MuCdsfjPCUuT+MFY3vdPu51t69Sp05WTb9D9BSzd6suAt8LS0j6Ehcc+6s7U8i9yX4+dk/NqLL7+k7FQj3fW5SGpnTrdVfRoxYTu+Xn0Uk940OlW0a9jp/x+KRYb+2dYnOtX3MY2derU6erpNdhm2H+CHbp3DtvI+4W7+iXnCuglwJ9jh4M+Fgu9+R6sz33h0+lup7sREzrtMN1Wz4+IfAbwecDrYZvJPktVf/O2NahTp047RR1DOnXqdL3UcaRTp7uLbtueHxH5GOygoy/HQgh+F/g5T4/XqVOnTpekjiGdOnW6Xuo40qnT3Ue3zfMjIr8B/Jaqfqb/nbCwgm9WOwypU6dOnS5KHUM6dep0vdRxpFOnu49uy54fEVljm9+/Mq6pahaRX8DyfW8/v8fxk7oj7rpTp063n+4BXqa3yJpytRji73Qc6dTpzqY7Gkc6hnTqdMfTFWHI7Up48DhgwE7LbemvsCwh2/RFwJfd7EZ16tTpuuiNsDSkt4KuFkOg40inTrtAdzKOdAzp1OnOp8tiyK5ke/tKLCY36B7gL77xq76O/b0DZhQQRAQEEGlPkzVSRRUUu55SQkRIqW57EiDPGVUlq9oFWJSTUmIYBlKK+vznCZT83e37UV5GGSSVusl2XQH1d1RqvW05qkqea5kiAgmyZlAFEYY0IM4LNFs5Xkb2v0vdWF+W/VWctYgIYsw1/jbtmPOMeD/HYbS2lnKjpNmvJBDK+My65G/Q7J2Lcqx7AmKHPkvhlPFvGIbCn2EYyt+qSs7ZPqq1t5JOrDeej/a1fEqSSElIkhFSOX56GFbGn6a+eZ597G04gueqCmJtG1MCvH2aydPsdTe8UyW5nKaUYGi26Un0RVDNZJ1Rb3uwJ6WBYUigmTxP5HkCMpqVPM/2yZl5mkq7rQ3Os3l2eQQhQRJEBoZxRNLIajUiInz2F3we2OnkdzJ1HOk40nGk48j1UMeQjiEdQx4FGHK7Fj+vAmbgiVvXnwi8YvthVT0EDuPvEKzV3j7rg317xm7YPW0VMwQ6+3PqymODLqmMG2Rlno3xi/K8zlD6Fqji92MABwwnAFE8Z+WbEqWUrKEFcJSMK4XUPkvTnlKORvuFlOzelK2vVVlrO+x3b6s3tzzjH9W86Its8cHabm0WlHmey70hDSQRJPpUehODZEK7DTjbvNOci6amJCSfUAyfjwNFGgZTyq1xqnzSBUiCNLLB4rns/Ntul4EYiFTggkQaBldEQTNkNUWO8ZwduKO+AI8y6TlzAqRyAxqqyugTh4gwDIMBdvQFyHlC/d0Yu2j2MAyM44DmienwkKPDmXmyNgkUOZ5FSMPAOAz2dwD1PDvaC3t7B4yrFeM4slrtwWCT79HmtmT6vSoMgY4jUU/HkY4jHUcK9e8i0DGkY8hdhyG3ZfGjqkci8nzgfYEfBxDbZPi+wLOuopwiPAigUhjdDphfQZIDDa2VJVMf0SqEjYK3ILOwzjTKf2L7qGK5LdjlgVQVOawQdYle+1esJugWgCwBBWBM6Vj/t58xy0m9VwGH0uqLWSMQKSDSCq207+VMgM2izwJJlxaurQqsVa01SBJCOzkcB3ITn0So7/Je07+mX2aF0YbXuVhmSr3S8jh4iv9u9zOKZNySBznk6Fgvt/gNi3YlV3JJpvQBRIODKSytbhmz4IgmEON3ijEQUBGGITEkIefEnGfOX7jAPE1hn3SZb8bZJxTBwWptlqQ0rNg7dZr1esU4rhjHNYiQc2Yz1bG6VXSjMMTL6jjScaQ803Gk4wj9u0jHkI4hj2oMuZ1hb18HfLeIPA/4TeDpwGng2VdaQPaPLWaljGvOiuqMKMWSYEJbhbFd0YooShU8EYHiUm5cwQBZq9AHMCVZlBs0X0Rh1ZVpcHdnztnqscYu62veW4KGLpRvqRh2XdyNvQAdQ9RFm4tlAoeIpg8FuNDSH81VmxK1zgqkdt+UwxUwLEmSF9aQClPQWnnijlkjwsrCUlFxUJdh2QcFKePp/aPKAmIWpXBbq87MjXu4unYFSVVF4n6t38Axz43P3zrpVdexqVY+rF5R0pCKhQwgq4AkJNBKDPSGYTQLVuGZt0EzolaOyohqJmG8zqoGhF43SUufJh+r1XpV3PIhMykNjKtVGSNVZUaZc2IYEuM4Mgwj6rI7Axvd6v+to+vGEOg40nGk40jHkf5dpGNIx5CguwFDbtviR1V/UEQeD/xL7GCx3wE+SFW3Nx5elDJCblWzyHusFylKF9cLE0O4UkICmZxCKNtVcShqxF1WwU9oap6jUdKLWSuoShxtyDlvlQHxZFyugCI0jwLH43BbiuvFUlRA89iDKGIrdc2omFVFMeOHAXnbp1b5jvdbFZLHo+YyOBxT2vJ+jiLnBX8r2Cz7E/dbUhRBUHVA0FwUPQCwctaASTUjOZPUWilpqJY1UUeM1uLVtkGRXNDBHk14G/zVBc9q22ss7pasUcd6QBhS2/daVi02NeOpkDJoNN0n20EY1yv29vYZx5GUBtbrtfMk0UYuSxPLqzkjGuEUMMWkTGNVyrfeYgs3BkOg40hLHUei+I4jHUf6d5FSHR1DOoY8ujDktiY8UNVncZUhKsv3G6Zr677z60IzkMaUdgNZAM8wSlnZI4KgCwEQEdRdkK3wiCTSoI1r8jjoXIzCMiD+rAFOXeEWBaUBirBBOOh4TceApgJAvda6yVVC2SqzNC9jURdM9Hq2SZWydbCtTx1U1qlaP6C6yltXb0qJFCAnaqjT8N5c1bXPS4VPtArcAt72mEdp0cbiUs7ZN+bVMQylFqkhCYqSEwxJICvDUOudSp11rFQdZDNApjjjJTjvdWhGEGJbpoq9klgC7NLq53wUNTBW0Cyk8gdIjglDfIOgMI4rDk6fImclSWIcV0XepmnyeOA6Cao4oCuknBxgTGdEmhCCi0x0t4KuF0OsjI4jHUc6jnQc6d9FOoZ0DLlbMGRXsr2dSDkrsdBrwl+BqnQzatk/wFxw+FBHDGXyv8sCXMsAmVXCFdE3vRWBd9dwq5JQhd7X90XAzCkX7QrlbTbB+VOxmI4MISJSsq3YBj1AZOHirYq0bE21Jkn5u/3dXql9CnCI/X0BFFZ+KHhBeNwkUN51yS8KolGPesYaqmu6WIDwTYFJGYYVk2f6CB4s26aMTSaVMMNo0wcD4Gj3kmwz5HGLURowd60Dr00/ipQNe1bW6GOexia2OCsiMzlHm7TIURIp2NK6lGuf/FlNqFuCBEui0tg7wDOniIjJHZTxFjX7EtIAOuDGJjQlB3AQWbESQWdzq0/TBINvdhwHa3q0rZEtMGAbMZe0hOBjE8q4ZfHaNeo40nGEpo5oV8eRjiNXSh1DOobQ1BHt6hhy52LITi9+gqq1olGmxao8xK8OelVWceAwpRVA1Fe9qiVF5PH3at0nWlYctOrzA0pe6kFu1DfKkHKLUPHW7VoLd8AQqcJH0ycCD2obLCbZ3YMhkf78sgtblheksdAEj638aAMn8EC3fkJtT2SlMSWqfawgk6BsnmvLTKUNJ1m3pF4oYBjWMvV/C5JGybZ+iplMXK1A0mhjopYWVABNypAt4tt4lhrZs/FqY1at/Mg4E2PuE1pMDjl7hp3g9Qnx0gvGxmyZS5/ELTsh12YpyYUPM2K8mH2sj+NzobQ9mzcsNDf+JV7eIeo4Ah1HOo50HLl26hgCHUM6huwChuz04qeV8ZISMAtZpzI4bRo/W4kvXciVQokr+xQhq1Z2isTiua56EeYCVO0mucEzMJrgF8Wk1dQq4AEwqCKSSLGalcSQxqKgpe8IhlheRu1FUWQloTKU+xdLm7gN1MUd36hncW9LtNF+X/JQoxcLEig5/aV5p7UWDcNolpYZ3Ny1SK1Z8o1oSVZJ2RrXKjNusdoCQS1WCSgmLYAmdjj6b10b6ubPaEO2PPNzEgaXhar8Bhq5abN4h7ctLcsJMVXwjvbqTGtvaTd9LixVIRNqbm0TKbMnik9EElzx58SHyeQjJqIq9e14Fmuf+rhqbX9KqWQHOgZIO0YdRzqOxP2OIx1HroU6hnQMifsdQ3YDQ3Z68RPKlXNmmiY0u+t24dY1Wrhc/W11pU7NVRATGtWqWKp1xW1SVNTKDoWy96XdgOiDUjcg2orYBhGv0ZS4tQBoGdRMStt9GH3w7X67t8uykzTAgJaSi7UlYoXREjMZSmFtjphTLet3FoKoBTDMjSpYRpA4iyChvu1TfIUf71bQXo6hqjLPmc1mg+/OQ1oXsyRIAcAtKIjHe5aR8P44wOvxetoJI+QGnSHOU/AxG8cKYHOTPaVYTRiYs7qcZGa1LCRhzZIkIEMzKXkLVUvWTJHBXeyhrM05Cj7GISjxXkoDSvL0k9LEB9t2W8qoeWyyS4E5zjOiBvw6JtCBFEwS60MBkxizBtiEOiYiwrBelcl8GHYaRjqOdBzpONJx5LqoYwjNex1DOobc+Riyu2hDFZz6aVasvnrdXtkbNlRwMcNHqJY/U8pxgGjAQqscNHVGmaEV9l/2FW7T2lKvrfiT1d0qZDyTZPHRRZ0BOBXgVGvMboEKEaR5Jmu2VXkoW20q2w7DANZglZb0J+Kg4dYGpDDEHK3ioaahBA5QbhtRqWMkDok1Q1PCrbUAACAASURBVIdbRJpxo/1Z+rekOlbHblXgsj+sPreKGeBkdJ4XwGJhCzYVtUoGDqJaQxtUPZVj9nEFaNNduoz4Deth3C8bPw10isUkGLX1e7UK1THPGdC5jE9MjIjaVOrx5Ojg17JdY/Q8/r4JMc9lQmoBJzaBIkPVJ08zKSIMw4p5Pj4mu0QdRzqOBE86jnQcuRbqGNIxJHjSMWQ3MGSnFz+zwjxVoAnnWUqJLEumkYYCNDWbhq+MGyG2FIqZcEdGir6hWX2q14sEEFg5CSnC4KU1myC3UgAqbCXHxBtFkoT6z6HZSJf1sMkQU5yDpa7lachSLCNtCkaDuFC8RlAk4pRbQdf6Kc9hADgkq8Pd3dVxGdakVMoqK3cH7HBdVp1yi1ex+jh/HNsk2kIFxjINNP01zqblxINjljhIz75ZNDfAXfgm5BmTnaSQ1IDT+ZhKkUMBpTnPTHkidqumITEMQnJQqcBmLw++6Q9tN4A6kDvYWoiAAgMCZD8DAEmWq0UtxjesPKo+BhJnFkBysFEf06xaaiFgKQ31nWTxw5oVzbO1JiXSYFadaGv8HIcVpGbS2WHqOAIdRzqOdBy5duoYAh1DOobsEobs9OInbybyaKv4USJLx1CyewDGFHF1kOpeTUMyQ4bHCob1ISVBNaE6A54tpAEbqAIsrhFF+LOg7nqsrjrLSpI9W0sBQADxdIWlXG+LLk9wjhzsedrUFS/JN6Wptz8yo7QHpkU5FCUsiqi6EJNtkWktGAXYwoIgYlYKcDc7mKrXPrTCWQu1rZaRvcPcwrV+Ebe3qAm9qqLDuIzj9AfFO7VtDYm2t5Yn0++wMlULmb1q8cg+qObGHUbE2z6rbwyVYksiK8xTWKvCxW+hAEMaSCVN6LKeysOYrBLTPBcLjllhsJOSHTgkmZ1KqOWFrM6zgWVYy4IPIokkQ0kPKZKaqN2YvDyeGgNBO1zMXew+qYW+qAgJdV2yOrLOMNtEOs0Tu0wdRzqOdBzpOHI91DGkY0jHkN3CkJ1e/KjGwUwwDIkUB0INcR9U2vSKvsqUOKzJrCgoDGlwC4QN7CIOFRaC3X5o7seKtpJs/azPxkp8plHS5rWLK1ADBlrXz/UZgaYVsdqP+tpnoz9t/9r3wDzdImbjcFYhqAHN8hUDi4Yv7e8mwDTtXvaxPOs8LP3JuUl1WGNdoz8FpHHM2GoT0FioDHAW/RUhpRUymNImSQzDCkkeBpA9RtWtR4pNSJHNRBgoMbOlXdamKWSoVmUWptaCluvp2wk7X8EYlZayRlpsNgyehVUF9TZqtNMjbJMU8BSR4k4mT5ZKM7GoK2cbJ+cSuLUpIQweN25tr3KyLfW7Rh1HOo50HOk4cj3UMaRjSMeQ3cKQnV78ILGaNleZjZWphLj0xYm0sUqOTV5JUs36kVrFMHezaBu7mstqHFwwG0tCzUiCCbC4NaRtqtSVqtWzBIX6jLjFpwEHqWWcBBSq9ZToZZ1b7Greb8s46RkRE9qwJpWeu0VEVV0qm3YA7aFcJwFbWdE7r8UqwNfzjmjuQvV/5Fx4s80zVd/cF2jYIGEoWAWbfIxHiG30G3x8i5UruYVOFYMCb2sziRTLRmw+dEDLki1zSzu2wUPVRltlcYAcMngbWhku1R2fNGJ8NB0f7PraYmwtK0v2JvhPnUGLTccyrohN0GHpGgdhtVr5pCBMbtWbphnNu2uxBTqO0HGk40jHkeuijiEdQzqG7BSG7PTiZ0jKWHhtwCDNShbsXriTk9JscsuE67Bd1eYMIplEcmFQV6LjloGFUinUOMelRSDeiRV6ERqhZjTRGndpFoEqcMMwlJjGrNoI97K8Zb+PK2cIXNuPuF8tF0JYRcLlmDElKRYQQEUK77bbkLaUYwF0ElaD5DyrH38AVSEzgpilYkZLNpBoZ+E94kBXLUFtWw2Ic8k/gsZz9pKooNnGYnB8ySowW73TbOWJQPZJzHJgQhoGUnLXsFLGpByituUibwEvLCitPJXD5LDwgjaDjDQTmLgsmFwLkf8/nt22SOk0kamH6qnOCDNDEoYkWKj5bBOACCIhiwMRupBEWa0im4ogR0fMszLpRJ52+EsLHUc6jnQc6ThyfdQxpGNIx5DdwpCdXvyEZaIFAPBVfCPApryx8rdPBksHKbb6FI+HbYU16liuWU3AIsF7xt3SqK2WRRwclqBk7dmyZMS7W4IImUglGRYDUSmb4UoZ/numujOX4OMK0vDrpJ+t9QOhKIz6RVV1JWisFSkt2BLXw92eVNkGNLMc2VkEkJEsFteqEDHIVkxFIFG3u3i1ubG8RHmBNqGsESYw5yZt4lY72ylEwE7cTiYD4oe2ac6LsRmGKl/WtnCHb4cBtBTXLeXj8r5Z8cISk1C3d9Tyan/V4nc91nUxpj4UwUObZK3u2DRYwM5l1TbYGqikZDaWrDDPNRZcUNRd0ptpLtYWVdhsNszzzNHhIRcunGeXqeNIx5GOIx1Hroc6hnQM6RiyWxiy84uf1qoQ1/w3wATAVq0h+K5Q8a5mMgmRxnLhAiphefCVdF2lO1g1Imur2GxCkTFX5TGlq+kWBQcQlpaBGOz295OsKZkKPuV+KDttnUu+tD9b0LIykvdDUUJAPTtIA4wqwmASXgBxyWtTGtu4tg2CbpkoZo9UxiHSFBqG+HXRYmdQbbOSGFnEqx6ThXKOQIx1IyOtxcavWJ1uuxFRT6m4bEtKyc4MGEIppVg8Wj4GBsasNCDMufLSYoKtV8MwUly/44AkA4lta5WI2MbY1FgTs5pFzG0p5bkEOYvLfL2PVp7UjD/ZN6iqD7OiArNmsgjTtGGaNujsG1yTyfrR0RGbzcxms+HC4SG7TB1HOo50HOk4cj3UMaRjSMeQ3cKQnV78WJq9oVn1i8fRGlMQyPNMpBYM92gIIbZIJ5GoLmpzyCVS2aDotpTFJq/jCuy1ZsiNANr9qkiL5s+ZBWDEdT3u3jZBmc3SAwugqsDhyhyuRpbKmecZok0mjQVgUAUZ7CAvqZYDs/R4Pc63lJJtyhSKValV4iQwDmNxnRqYmMt3ZPYxc2vUILahL9rg3KrpDPcJuBBR2+iGWUemebJxE2EYxjqB5KGJfXbrhKSFFaNaFDAgyBnIJDXFFhEyA3uy5/2yTazAIja2BQDVzNFRTQFq8mI/p2lj1qpsY5dsV2MBZbBJLP62dpn1ZRAhpYEhDd5/q2vKs/N1Ap2LTIeFUEQ8g49Qt0VS3OKaM3kG1Zk5G+i15zWICPMUh65tmM+etf5r5vDoiFCHeYfDVYCOIx1HOo50HLk+6hjSMaRjyE5hyG4vfoKcgYjFRopIyYlvKfTCtWbCrOAWBPs9S44tbmWgiXfUV9OcBAqy+Oml2fOhw1KVJYQg3g8li2vJHnRlrAAVShJA2QJUCIaIghqgjoO5D9PWBrh5novrkqirmmOsDQKWosaEOlGtPdG+lBLDOC7a1rZlTBYzulqtPIYzc3R0RM7CkDMyKMWCIWKu0mJ1cd56BhyGleuRWxukxhjroVsYSnrEGJfEMKTC/yjb1P844EiemecNOhs/Ip2oDLYBUQg+GoDOHJ8kAOJssrD0SFYHB7NKZFdwgJQ8602j4POYgNnAqIBsyGp23hSBtwlDs5dRATtj2X5SSjAYUIVsW6x5yJjPuF5+WFtMlKrFMYmgfp5A9A8ibaku44l3mTqOdBzpONJx5HqoY0jHkI4hO4EhO734aTuawTV8yYTW2oH6AIptZANzLXKC0qA+QJ4HHd+IFs/E6hq3rCQxK0OpyP9f1H9C+zXnQCaGVF2J1a6jRXGGYXDAVChAYIIyyIAkW5lbfndbeVvKTQPAlKScHhz9bF2kIL5xbvDNmCbM4kJerB7JFLHECAdYJWF05U+xgS0Zr8ckfmaAC7CE0jtPwdsc/fX2+j0Ru99mnzGBD9UMV2mxKTiGVmtTi6/tNVx2akxtNgtQzqADgmVOGQbj6ZTridYtzfPMNM3kyYA9JpMcuxi34p6PnT0QFrS2XVifsmeOaSefrBlEzd2tUuOjHThyzsxIyfVf+ywl5WS5hiDkOunVGyYPKkQ4RkrJksF4Oza7bLGl40jHkY4jHUeujzqGdAzpGLJbGLLTi5+9vTXrvRU555IeMKjEETowJI0NWvZM9kEJ1zKYK1VELO9KAI+lC2EcmhhU/9hmNq9PQsAb5U1STtENQQoqrXWNS0VRLQNJzbySET8kKk52VvENkYDlUvfDoSTci1Z2xuqPQ7jQocS4xoY1JdzYFltqFpoA1wRDKlwV74O9o7ifEnELF84Xs+oIaDYQKcqcmTD+qlp2Gya3fhAWFnXrGKSs9dRqhTklYvNczpl5M6OAifuRx4rOdiKxVr5Gvnqlnn+wbW3JeYacEVyBRezUYgXETh9OEq7miCs2vgS4aACRT0Q2edgJxLIeFu9UK1kFoQBqEaHNqBIvhrWs5EjFJzrfngi5ZOBJrg/HcvWniAmvE2fMeZplUWUCIgQjizIMo427qJ337H0ZhxW7TB1H6DhCx5GOI9dOHUPoGELHkF3CkJ1e/KxWA+M4kLOg7gYM12+s1svKPFeGQ1Uysz6ExcNPBg6rAhgyJcrqNPsqVxV3JTYWGqkr3Xh/4Vpu/7aLZcXdrryXK3B7V7xVJl41jhgx0Uzi98uqHFSEkjumsSaoKpq0mnTCYqGKMJvgpdoWofYlQDdrtqwkWjfygVkiov3z5MIfzwBZkm/gkxLvGSQ6F2uYiFvCUNvQqaGUlZ95tobP6jnts5KzAw6AtvGxPu5lYJu+5dmYoNk3aFr7ppyZcgbJiIyIb7yUFJYy43Oca5CizDQsAEUGGNNYeKRZC4i3XtpxjPjoOtoKxR2tHjYRk1qJEybAvnauypClTK18q3HjJqsxiW6PRYxD8Cr5M86DYai6JbsdrtJxhI4jHUfoOHLt1DGEjiEdQ9glDNnpxQ+Ee9LMC4K6CxOkuDQDhJRYeCp2PbtS4vpiSmcWBEXQnBY55OsQxkpzCxiavw0kGmHxwdHmfvwuqsSOrQBNVMvpwyZ0yoy5FgVBSypB67/WgguAqCiSZ8JFOEf8p0LSyNteLUaidqBXMENwgPbV+5CGAnIksfSQs1saSkaUKsDZs36EOAYokhXEbBaWJ95PMBYpShnKBpSNeQB51vJIhBcsFAehxATUwYiDjEnRV+87aFOl98GtJGG1sDhda5oqkAbnTfR1OcEsLCmCjVmRHyHLVMZI/EVBPFwhLGlmsSEAQgNowjTisbTi8cMalsM6MRgPZ58UZpdb1xFSmXRLuWFL0+QtUBtDjAc5z25hzKRcwem40333qOPIowBHjjZIzujBQceRjiO3nDqGdAzpGLI7GLLTi588zeQ5NlhRLAEKFicpMaS+KS2LD4YwT7nkYE8YICVPwReAM6sWwEleJhAVhW4XgAiJbJU/ACe7+zEktwBQAI+G8DtA+rOCWRNQEwDR2Q640irclHzrJmgBkAZUIUhNm1XRDLbxUb2/obXBXa/fXZ8pJSQNJWYzZVDJCwBHhAqjXgaUTCxIKuAji+vqbKn9sFIjBWeAeXZFFW9qrau0McanXI+ynTOpaVeZIHRxLRJaJknO2wZAzPTmY+5vDrVX9dkQy5gAKjARYyJhO/P2qYKPSYCB2OlzBpoha7mCy0wAZ+1JnT3xMQGNGOfGvFNlMNpeZUjxNJiRpnK2TCtZs1m5qOA253bMd486jjxKcGTeMGRFN4ewv1/GrONIx5GbTR1DOoaUmjqGlJ7cyRiy84sf9bz5UDfM2c3sp+C6IEVsol/Ls60c0RB8zDqBgwgerNoMQlCoVZzQXE7RjfsS62oKgNR3rQ1VgCnCZL/qogyw037tXashSisA2CjMYrWPkk1Ka9eR0s8KNo1AE0rSKJhqdLawJCwq4boPC8dS9L1ewvJQyw1LRnaEsPjepgz/L4Ap+lAsNuhC1ap1wxX0BIoFVh0LrRPU9jgVHrK4xoJLDmYL6fDJQmuqxgKN4aaNviULGoiMLjZBWSSytoe40Yyh5urer70nJght21Bek2PPxWS2lLewTAb7zX0vApPHVFt4RkakhhQsJ5ndo44jjxIcWa/g8Ihx2pDloONIx5FbRh1DOoZ0DIFdwpCdXvyYQqsPVqxg47ptvFqAkFTWpwIKVYliMIz3iUGSu65TGcwydI20SyO0Vop4HfFs1OBWkHguBNrb2OhC87s0dbbi36q0/SWyfE5QPwCt8sudsc1bBjriFh4RyColVlVQxBcmyRW/PaG3Al3A3rL0ll9SxsbGjQYAFl1zi4Jda7xCDrmum2F8IW+n7vSeFguHhlIYgLYgVTnZ/m5/x6FuAUpFGQN4vc+aG0CJ9tPKnSyulWz6GvxyYcq5eMlLtpdgGnWSEJYjb8+kMnEW8PARDDlbxnIXDNvSkQactIlHR93KY5/F6dYnZJvZJeo48mjBEWv7oIoeXkDXex1HOo7cEuoY0jGkY0gwdDcwZKcXPylV4CgD4WARrto828Ysd4xRh0Kbn/6OSMmugQjEABUFpoxSiHTEUC7AiHbgva1UjSuA5JlPKlhpGcTLUQi9CYaWa8aJbCDZbM6z69rE8+JtqYJWrCDxrMTmx+TKpajO7rZ3oZdauriSxphoYxVI3u8CPCGsEvxRB97GIqECkhdxulFX9KFxrpf+tM7p6HexCGguzxZFP87dws/2U1gdQFDGolpGWitIjGMcBldQMviUvP6t9+cycdaJw7LGUMptrUNZYuKIpysYVvmW6r4WQTIOHOou5MpTZLBmqTJNs7epptTUnIt7WVGOjjYn8HB3qOOI13aH48hweAEdBmS1WuAI08SYZ6aU0GEgzxm5cAEdRud1xxFrZceRm0UdQ7y2jiEdQ3YEQ3Z68VPAQ7eFvypSsUD4oVu2MFfE3auxok0iJWd8jQttwKYFAa3yqgg126PlQG/honUnx43WHdyKeywOaH5CCLwNbgiaFktHlKClGu8woGWjoj1RFTE+AOFVR9p+lsYSrAjLBVrBXZ13IfxDCmVKng7RXJSpaav1vRRd+iLi/ZprWIC04yrei0zDAxzocN5kktQY0BAPRcp7xPRQfm/KagA/DgirDMxtQ+rl1lrRjOjCmlMuV/5YzKvDYQAZFk+dhqFimuaQBMeUtn4Djeru9YmyaVuGClA5+ul1eWaauelbkgb8fIPn5Flk6icXADs6OmKXqePInY8jmmdWhxfYrPesvNVYcCQdHUJKpPUeDH6a+YULyIXzsN7rONJx5KZTx5COIcG3jiG7gSGPisWPNoMMYKtX05JWOevqnIXSpCQ17/62tcOBwuqSUk4UkyNmVUBVkO22UAVP/ffInx7KUIpzYBHXhaLkcY9WsKPE2sxakv1cWJPqQ/U9bYBHmme2+0/lX3ChJDFp+NUCaVQgoWy1EY3FZSsWVbfaW+7kqLWA71Lpl1aqY7qv8d+yX6qZ7PIjogUEIh/98vAwe18LWC2gZTnGZbIzt3YCs5Zkb3czkZXUL60Ui7fBATbPQrXIVGAE8b46WLuglsw8uiUvqtVK1kxeOddY8UZDUCAjRX9qRhZKnv+cM9O8u4cTQseRXcGRvN4jzbOp0comZJknVnnm/HpFErW9B0NCjyAdHlmr9/a8lI4jHUduDnUM6RhS6ukYwi5gyE4vfpDK/JaxRVC0iakkYiNtoCOGUsAyh3hKQWiUPGuREUvM4UhQAMwsOC0EtG3ZVsDkghY/a1mtvpuQLPukjRbJEhzK21WILH20kJiX7HIBVvWwskCbBmCsabWOkHFxfhhOahhh3HHaxMSqonkmLCWCQKpCG+UH/4NPC3RoEFC97+qVHYfz2qeWf5UtS6AWz46iWOaa6Et938oL5Vq2sfY5FxCzMoPvYc1A6jM5TlMWi1XWZCkhS2pUQKQBA4uhsMHIijKbqJQkPUIxkakikn1ydW5sxb0WKxlRR3U5x/04QMzSTmqMRJExA6YGcHJ7+N3Jo7Iz1HGkXrtDcQQEPThgfOghJsE2l4/jos8LHBlX6JSRoyN0b6/jSMeRm0sdQ+q1jiEL/lW2dAy5kzBkpxc/5RTbi1IV5NaAEJutIuVjBQVZMjB+l4jTtWuiLixpKFaamsqyqb0FG6+zXZXnJvXj0o1XDxyzHsz+WJwUXZoC2CnA2eTZ8rM74IhbBWrTA7i0aeo22OCnG1cBPsbVhplDUea4Zqig3ufYaVn7tbTOVOsGaLGSOcg4jwLYs0a7G4tOA8qqWrLmtICpaspvLmlx5VTqqPq16EHWuslvaxzVpy5tKhHzc1uYQpN20awRZtWJk7oR9SVhCzpRtr1rpzdrGS9jr/jGz1bOBE3iqH8CyFihWD7+4JGDRM42Pt63NAy13Udz7bszsQWVIqewALpdpY4jVtIdjSN5Js0z7O+zOjxkM/vZIOPIPI6sVdHYDK4ZXe+hR0cMm5n56Ii8qof2dRzpOHKjqWOIldQxxPjQMWRrrK3QOwpDdnrxkxVfGS5djSLucxA7qKmuNhsSs0dUwKk5x+uBXtVd6xdYcjcTq15xRVBdKmStzoGpAbMQ8aIyJwysaAXL8rSItcNjaZW8yJterB8nCYKD2MXaV7JwtLxsVubt80UpTqgoLBDRj7aO6kaNcgLsg6dLnkQbSi2CAX1UHUpVXkxNH11hVVDBzlOgykdYm9pJB4eTFkaP971SezJ2+0zhm3p7HcijBuNLtQhGD4VqEbMyWAx7sQxZRUW+YwJqZU21bgiNNKL1cDIpZYi3MXi+bUGJ/p00we+0xZaOI3cyjgzTjA4DTBMpZ6aDA5Iqg6pvmBY2e3ucOneO8/unPOY/oTr7Fx+FC+fR8VTHkY4jN406hnQM6RiyWxiy04sf43NolglLEmNiar58I9gm+sXJv8R5sS5w9qBg1ouFu9KpVWL3S5QHpPkU/svyy3ttd7Z3VcvbyglKHT1zb0jog4i5SKMiVWUcBltwBLieAAQtKBcGesfKvS2wadC0PL+tnMs62kWO93trMoj400xtv6ug8zY+WqqV5I7rBhghlE3ReT6xbVZCXVC1LuLSLnF5SYkIIE5pWCzaEDEXOsGS41AUY91a2ABkHByUvDXzhKrWQABfDAaYKLlmqSFbo9UOwJu1WpTKYW45DvuieNoq4Fj/i3WpnLzcAFe2ezUiu8qOlvGRRd/i93BL7zJ1HLlDcUSV1YXzzOMKTYnp1CkQIZ85w97rXsdmMyMHBzDBZlwhR4dMq1WdwIfELKMdEKgdRzqO3DzqGNIxpGPIbmHITi9+iiIQKRBtAFIjDDlWrqIMEu5AysAlsbW5201i8e2bz2hl08EJE3AfTM0z5slIeLp4h63jVF10s7e6ttLqqFeSeH409awtCVQMIrcXt0Oys3nFG1rTCDYsgpLSsFUWa1MASul1s4BJtk4PofN7kR3EVuzSYNeyfLM2KAxukXDwKEqjWlbwRcGTmBIoCMnLrplJglfxe6RHtPjSakGLRVXha8HNcN+m5nIy65vEOCwnCTD+Z6155Vsexs/jlhhhTEOjoJPJXp7Lqd4F0LxO0UwsEJG8OGRs9swmOaa3MhlYT8wbLWUyK1OCaJ2MCmU0J1TmYpESkZIxKFzSdWJPha8VVE+W9Z2ijiPAHYgjAOsV48Z1RaTgCHtrBpRZLW7/cH+f9cMPs9lbo2lgAPI4WirU1Soiaek40nHkplDHEKBjSMcQdgZDdnrxIydYKIxCoExhVdUPpJLlu8V0Mocq+Yq0gkMdoYiljRqgnOSEDaZoiiUzlrYZvy+m1JIcAO16FikHepmiW5ymCAw62cpYMcNMFqaVxYkmzYhmUlqZICX82RBSF5A4vAuDiiRCCjest1v9/7JoLjKpkGy/GwgSaQlDmV3YbSOjOBBq4WxKpsBxxRLeGP8WeUsKnwubg4WUxhfFqmDZvlO8TCKN8nsfisXIf1WML83QlvHwCaiW27YTFHOFa2NlK/e8vUOSes+Vvv6Zy2JvdstNzg7eGi5whZzJSYjU39ndxUpkQ8lueXKQdMBXlFRCh6XysJnaiDJjI2HwMTk4u1aYi9o3E6ptlBSfbI+5m09SwR2ijiO3EEfmDDkjw7jEEZ1J01S+vORhgGFgPn2G8fx5j5MHXNfn02cYH3otGwFWaxDhaL1mmDOTwpBn8qkzzYZbZ1bHkY4jN4E6hnQMgY4hu4Qhu734ab5Qt1SAwtNymKUk3Ja2+k+xqt2ySihmeanf0GVr8HDwAEmWhzxhh1+ZZSDcyJEOI3DDXc7zxKC2uXAjkHKGyT5ZRlQnkszo9DAjGVTQacWsA5u9kXF1gMiGpJnx4HHMK8hpYpwHmE1Y8gjISNaZZPlASmIBYfAdd9k+1jIDBO+feKKCNJj71fhk3IhtkebmVBKDK1Uq9+2nCWi4hMXfzKp22NsWYIBvupRQnsGtEJk42Gzbhdx6jDTAnND+uqiK94pFyMuIhVY2hLY2qpCS1uc0FBBm90YpE7g1TBRz83pfRUMO3D6hMItbJZp89BEWPDMvrDc1xEEKIE5Kscy0WU6SKEOy8csKmpUskFSrK97LEgayZjJuMZKw/MXAZjt/wvmcs7mRjzZeX6bwJeeMDGF5mVs420nqOHKLcCTPpGmDTjM5JVQGk76cGacZOdqQ1Xg6r9YkhbxasTl12iyy84ysVwSObJJYWfMF0sEBenDA7LP/URLnV8eRjiM3nzqGdAyJ8e4YshsYsuOLn2ELcFxhSsq+ujrPJDZZXYgcZEKYsg+2JJREVmHUDYK79MSEcSAj4jtV0oCk+OIvIDM5HdpqVxWyIJqQGSRn0vwIaT7iYJhZz2eZjy6g88R04RE2h+fZbDacn2ZkgM3ReYZxhmyZLo6OEpspM6aBvf0nwDCi6zXTveeQU6fRIbEZToNODENC80Aa9xh1tvb6brwpmUDNaqFkIn4CsiZUE5IM7AYZGNPIkCzrRtbMLBOTjIZVLwAAIABJREFUTmiybCEDM2Qpi57gvrhCpGFwbMgguKJk5lmbsRmwCaHmsddy8FUcvuWpKIvlweBGNSJDhWrScCBxOuZVon2u/i2EQmdUhDFXl2p4s7IqWe3A1kETgwxRCwFPJYuKo8k8z0wF9LSUJyIMw1Dy07euaqCmPnXZneepWj1yhCXYJJFzJtJPmnEpdCB4vHTFa8CyyoI/LTi3myUji5FmmKap6JQbiMxyNJfZeSep48gtwpFZmSdlIqP5kGHcY8iTHSY4K5tTp/Gd4zBNrB4+x9G998E4oENC8kSeNqhgOLJ/YF9oVntljOqE3nGk48ito44hHUM6huwWhuz04gcqqESsaGW4CYowNPLlsbXxnAtqdTq73SSJAUe2cK2IMWzrVHGlZTbXb8oMMnF6nVgxkzaHcHRoudTzzOboPOth4J79NbLJnJ+OWM+PcH56LcgM68zDHKGiDKeFxIajC+eZpoxqIs0TQmJ/WjMdKYfnlQtnH0LW+6z2T5FXZ1AGhjGxd+8DsD5DHk8h42BNdytJpsaWJpQB40dCmMX7Lh6Pm1w4xWJ9Y1OmuL/H1urNZk5wq5WDeoaS9jDnFguayWA5bjZKS0gAioKYFcXqyPjE4W2yl3zbqS43d1bgidKbGkQYJFmaSzKTZnfZik1Kom3TPTwg3LI+IYlE9cco5+mYVTAUe5qmBei0QBe/zzmbe7kAjlmfssdrt+8aJ2ziDOviPM9knesCE1BNx5i85Nkynnb7mUcbdRy5BTiyElRHhkeOmFkjDJaelgEV06e0Xpn8Ceg0INMGRcmnDlg99BAXcob1GiIl7TiyRpmk40jHkdtLHUM6hnQM2R0M2e3Fj4PNIjWkM6PIly/AB9SER1uBi1V7kAITqOXdl5Agz/GYGYpg21lOmUEn1jKxJnNmnHjMHuzphsRZ8uYhWyKMA7pas1qPrNbChXOQVkKaZ3Rl2TTW6z0O5sTRPDNNM+fPbSw5gCo6bViPidUZONAjVIUVM3r01/DIQLow8vChspnNxXjmsU9k78z9pDNvYC7eJGhKyPoUyIqUBlIy16EkW8oMKSFpMDd87Jb05Y26kooMRLyu2yjqgkikKqFbLgx9Gh1sjCHqFi8idjWUBrARGcpzMTb2q8ZfDvg+ltKMf6M4x0Cn6IqUX8MygVBcrFkj8WN0QB3itNm46u9rwJj/4qEEEVe7LZttv7ZdzAVk/AwCS4lZwUa1MrHtW0s5z97e5Pn951JGLDgTlNjqALjWuhRUNnJSQTDOJ2jv7zR1HLk5OOJhLqYZFt6AwpCFYTOT9wYslASm9ZphtTL5Qkl7a6YE+2fPcnjqNKoDOSUkZ+bDQ9g/AGA6fbpYZDuOdBy5bdQxpGNIx5BjanEnY8hOL36kAA6FkSIsU9551ogxT6jEzhMPoxWpJ+uSETKJGdHMnMwyMGgmoVis54BKIuXMqBvWTBwMM/eshdMr5f7Vw5xOM/PhOZTzzKsNF442yLDPfQ88lqOjide+6hUcvu5hhMy0VjaSGcfENCqkgXmamTNs5oTKmmFv4L7TK8ZxhPE1nMnKNM+M04zOGzazcva154E99lf7rHSEh17B5uFXsrf3Ms6dP8cjR0dMCKfvfwP2Tj+W8eAeVvfcB/sH5HENB2eQcY80DEgylcpq7kmyIllYqbASTAGSsklu5cijcTSsHlIFOM8GJAVKXI6L+7aNk20Ux85DCCWLxdTcjLctvFA1N3pOPp5VVXJuYlq1glTYZlyAomULcnsNs2oBHREQnW1KGgBmUM/m4yBjGUzsIxjvFIW07R2rYDiOY3Ett5v3cs7MsblQI0bYrGBRRqsDofgaFpZJrY2YpQo1C41l2EmgCfJkuuJ1m/Uqyl3q2TDWcxdaoMs5lzbtKnUcucE4kpLNwtMh8zSRV2vTlSww7KP7iaSZnGCWzFpnxo1wYQ/GIbzLCmJfgsbz59mMI/OpU4xnzzKvVjYkHUc6jtwh1DGkY0jHkN3CkJ1e/FQKRsXmKhNUAPyE4QAbkIUpxgwpWtyHg5iQaTJ3ctJsDujkLlLNDHrIfn6EPTningHuHdfsr5RhtJVxWo9MOnLu/CGzrlgNB0yzMM0ZspI3F9gcnuf8ShkHYb1akQbh8PARJFt87T2nTjGu1gyrxDwdoQmG4V5O6cDR4YZRBpADXrt5hIkNw/4DrE7fz7g6xWMfex85b9g//zcc6CMcyMwjh4dcePWfcu5Vf4GO++zf/wB79zyG1cE9nL7vCazOPJa0v4+s14TLPbtSGxgryX9DzfVsrLQkBYItjBIGFoLHxPriR0RYzAOt5URZAI7V6VaIPC1cpMNgClNOCs6gzMWqZq/rsU+8b3G5S+VPZnrytPq2ca4+H4BkZXjmReIgs5qa05d4msvvNXt+gMHS+hPpMoslZfJ6kzBrBl88Wh6IqvBtH2s55pVT9f1VxToVGxy9I24pkSwL8BCI3Dul/PhpoJIWgL+cIHb3S8uSOo7cCByR0Sy8acqkacO8yczrNYIiOpHUzpZgs0HXB1zYF/YuHJplWQVNIDIg4wgiXBhXdg/YnLkHbUes40jHkTuKOoZ0DKFjyA5gyE4vfiyW1lyiYXWBWFFqEQwF5sE3bqmaBUYtXjERqfYsrCshDAkGGciiDEXEEkgiMTPkc6yOXs2aQ/bTyJnxPlbjYNaSeeZgtcfR+QvkKbO/3mdcjWyOLnB04TyZDUf5POc358izcN/j7mccE3nKyCazIrG3t2bcWzMMA2kQzl/YMI6JeT5A9YjVODCkgbOMTGNCD/Z44pu9PQcPvDHrM/fzwL0HnH3NXyEvfyHz0euYZ4udXU8bjvIG5JB89hzz4SvYyMj0yvuY7308B495HOv7HkfaO43IiCLoMFiscIIsaoCShSHbBs+JQwgjg4QlwRXRc9QjeG7+JbC0Qr3teiXctu72rq7dMV4ujyFSNiequ0TnubG0SMBF/FLrMitaquAi1BSTqqVu+2m/D7RKbbHboqb4wzjaex7nraWo1pLhspYqDwYEHQeSJHLCNqZKYs6zY4QDhZexzLQykHzDrV1L3iYt71mntICLzbut9coxWwOgqvVnG+ja8TP+7vaXlo4j14cj+cIreGQLR1an7mPQ0TdyC/loxXTqgNVqhGFENrMdHJgTqiuOxon9s2e5cOYMOgyMQD60zE0MI8jQcYSOI3cqdQzpGNIxZLcwZMcXP0vXo/W9btLK1FXhRka3GChJ1bKloMxJ0DyQRciSyMniCGXODDn7UyBpIOUNaT5LeuQVcO7PODgz8JjTT+De+0d0msivfYij8+f464dfR542nDp1wLw5x7lHHuJo3jDPMxcunCMNE495YI+kK5SZQVas9takGfb29tjbO+AIYf/UKc4fnmejmTOn9hmn0/zJy/6Y/YP7SQePY++N34w3vu9+3iANvOlbvROznCKnER76E6ajl8J6wz1nRoZhYDMlHjl/ntU0mQClTFJlTIm0eSXpb17K0atXPKQjk+whe/ewuveJ7D/wJqzvuR/2z6CrfVQ9//x8hIgwRJSye3Y22ewTCmTxmGSRMKkgkhfK0oKC0FokANQimwMJMGuGZLXTs9UO6FLMLWxlLZWlyIhtLV0oWNQ/e3pHEbeW5NksHXmGnFGdmXPm6OiIrBPrw8yUZ6bIkDIkxtXIen+f9elTMBgP5gBNrZYVa9PQtM2sRgZ86qdHJytbMkNO3hcDx4h3Xca3SvkpkrBEe9pga/mr8FtnzxSkDmSjtWnwtJTRvnh+GIbFmEVKTMu+srtZmqDjyHXhSJ5YX5hJ6xE9+mvSq1/KxnFE55G0OkN6zBPZv+f1GPXx5HvuJ69PIUyszl9gOljZ95IZ8ytPh8xzImfYOzxiM4ywt+/fKzqOdBy5M6ljSMeQjiGwSxiy84ufGl9ozKixm8bLsjrMkLKJ3YC71QSLY2w22E06o3NmLQODJNu3Mk8MnGPIZ1nNDzNceBWSz5HSPeh6zeFqD93MnM4XODw6y95ayMPIw+ceZtbMlG0j4d5qxd7p00ieSMCEMrkzM6XE3t4eq9WKcUys9k4xS2LK8v+z9+ZAsiRpft/vc/c48qp69Y5+3T0Xdhe7xJpBoUEBCJHUKJMGo0QaNVKiCAmkRIUEBVKlRAESqFCg0UCjAhh3IazRbM12gZnp6enrdfc76sorIvz6KHhEZr7uGWKO7iFqOn0mu6oyIyI9Itx/8fw7/h91NWcYIkO/YzcEYmV5/M73uPyTf5eNwPLRFame4aPirNLYyLxRshswM6WyFX2fMLnBG4ura+rakPNAzmmc6IFWoAFCGOj3d/jhmvXmNe2j95g/+yHV5XtEcQRjECugmToXnXkwGJVS8lRkLERVZtsEnDzemEklBA7zsdzP8WcZ5GWb00S9U1gcJBM1F2XLw3bmLYtAkdA0HCooH44xeaiKRW4KVVAdZT3HVwqRlALeeza7LTkm2qFoy0fGOFoRctuUWGjk6M7WdLBsvD1wy3+mfhwsR1KMNFPl5Aksp2Nc0/F6HZpOVqaxGvNbRqURNnoCGy1X/bAdcuIufjthV+TogD6NAz61HD30dubIb86Rygpm6DAhEEgk62lt4Ujcd/j9HV2+Z9jf0G6fsnz8A8zzHxGzkMYaHpCpG0uoFxjvEcDGRHj0iJQnU8qZI2eO/Nvbzgw5M+TMEB4UQx744sci4sbVXxjf/erwLW8ZTRTn6eiSHm+MtQZNSiJiNGNz0ZeXVN4neTT0xP0bdHhDCmsa9lxezVhePaa9eIxpl0TvScmX1XJK9GlgiD0ygiYnqIylrht8H0fTwkBbL7CufG5dBvXsdx21QB5lHJ1r2K09wUesmVEvrpg9+T7V5WNaVZpmScyAEbJAEofYGbPFomj627Lad7bG+wgKzcwCFTEO+ODpe0+Kmba1zOeGVYpE3RPkFWG7I8Q1JnrM/BHqZuSqLcmXGsHYUbddStGxUbsfwJhSy6fch/QLB+10v0SOD42cS9LnVHBMTh8ejFYSjhMo5ymu9O2V/zRpxpuOObEmTLGyh9/Hl6ZICoEUPH7o8T7QdR3Xt7ekGGmSUM+aUnRNiixno0BV02bFikFOLH0FZOZgoTiO0zHGdYJAysXqp7wF2dNYXJFcCrch6Ghh1Am4lNMs1hw5QHTsAXpyvU6hdapQ9DUr2ASq/Pa1P3U/d13PQ25njvwWHJlbdFGR9h05eHwM5KC0rWW2skQfWcqOZJS03xD9GpGEaS9JtiaL4qoK0Vj+0VdXZITQNCVJ98yRM0ceQDsz5MyQM0MeFkO+8cWPiPzXwD/6yts/VtW/NX7eAv898A8oi/v/A/gvVPXlr/tdx9XyccE6XehDXCBaMKMlTKtUDHaorQ4X3aU9LuwwfgfdNSbskNzT9R2b7Z5+t0H3tzR5S+2gev4uq6u/wdU736eta7qXH7J985L1/WuMrUhRSTkS+wFBsTGwurgk5kwfAnZ5QbtYcuFWqJYKxdH3zGTLbv0S9VvWd3f0PuGqObVWuBhRq7Suol08ZvbOu4SmppVSc6eoXiYkCcausPNnzJ/9Ca69Ypl6ctizu3tDv9sRhg4nubi3TYuPwrYa6Po9YiJNO6OuZ/hhIPgOV0NMn3Lz4ScEWuziCctnf0DVLvHVAlPX5GxQ4xBjUJlilxkniaI5nlhapvkvp4+F6a6+9etb91fATIaJgxVAyWTSKFFpOFpUJkuFjBN2AtM05w/uWilx15qVMAxs72/Zbzfstltur9+w3/dsu47tbkdUJUlmPl9QuRpbVzR1w3K55Eoz9WJBVVXl2Gl0x5pYFGgObu8pUa+4sRljgs1YfCyOY3ty5R4kHg+wVtA8auZnso6Fz3KmSIcekwIPSZCcQIRjsuPxveM2KR3lPjUzCl9wCAWYthmGgb7v+Yu/+Itfdcr+Su13yRA4c+S35kjbYBYz+j5TrXv62GNspK4bFrMVPnpi6qgbCOkLbj74jE5a7PwJq/f+iHp+cebImSMPmiNnhpwZcmbIw2LIt+X5+SvgPzj5O578/j8A/yHwHwH3wP8E/K/A3/91v2Ragb69cvxKHGVmRE6G0UWqVjgk46viKKGRzoCViNGB2uxp7R6VLVk3DGlHCD3WNti6QY3jfr1le/eadP8Fab9GYiKGxOBTUbzIllnb4CrDMHhMXaNGaJdzFqsFvpuVYlaqxDTQZ8++29P3e0xVVtNoYHO/oalr1EXaalmS04whO4dVIUTAlfO0kjAacUaomhW525VaXkbpRBCNOAs5erxGxFVkGlw1o04dXb/l9n5gtXqEc3MihjTkEhvcGobYkeMNci+k4YJ+8X3qNCeKRWyDVu6gZFPiL8vK3hgp8asHq8loRaFYOowxxSmqb1th9ARMQlHzmO5xFsp3JEbrRpkcMloDJpe2YXJLA5QYXVOCUVERRI/WFx8Cm+2Wzd0d9zfXrDebMrF8ZBgGEpAbR58zIQXqJFht0JMYWNDSLz2xcDD2kdJnkfH6aFHdKSde+jcaOY5WjVw+m85bdbpOpcoSJ9B4Gx7H443vTI7nr0FmjAY4WLpGvh/uZTHcHPe5vrnm9uaW6+trPn/x4leftL96+50wBM4c+aY4kqioqhlKz95vuAsDF+2KyrR4EsOQMUa5nAs+DTBcY99Aunxy5siZIw+aI2eGnBlyZsjDYsi3tfiJqvrlV98UkUvgPwf+E1X9v8b3/jPgX4nI31XVP/91vmQctuOx7cn3jBcRKFWflCYaoqvIzuBEgQBqsYDTjFUQtYRhjek/Y1DBd/eY4ZqlyVAZujQnS8PFbMb3HzXMZobd/cD13Q2VKQmJu+0tIo6kDlvVZKskJyStcVxw9ehdVIVhr+Twkth3pNhTSSb0GyxK8pGuG+OFndDnnvV2A82cQM8jZ2mWl0TN2DzgxNHnBUkqMhEkkirBhAVVs8Sp0u93JI2Ic8zmLUoghJ6UB6x6JBfXbWVmhBTZ7zbU9QCq+BAx0TBvZ8wag2XH43nFEG/Y7b+g3z3CV+8QZu/QVQtyylQErCqJmoxgrZBFyRZSjBgUq4JRsEYP7k6ZqhtrCSFwRKZCXXBQRwRKAmK2RSVHUwQphbfelkIso8SInEQfKEWX0oz5gEXFBcmoUSJKMEKsa5jNEGup6ohdtIg11FVD5RzWGJyzPLq4ZDabsZw3NBYsmZShRHMLIm5MHpTxHOTQCwHEjnG0eQLw0dqRcyZNdQtGdzkiYAxGDaKKdZYQIzFOz3XDqXrKwaWuBa46xijn8bqeuvFzTsRYwgeOD+/yWRphtN1u+fDnP+PLF5+z3mwwevrviW+s/U4YAmeOfFMcqcRjaiUHpUoNXpT9bkttHNkKiaIaNG9nLJxgUseTq44hfnTmyJkjD5ojZ4acGXJmyMNiyLe1+PljEfkc6IE/A/6hqn4C/B2gAv7PaUNV/dci8gnw94BfCBwRaShu6amtvr7N4XikdFTXmGIUQy1UtqYUBwNDREiQTYnRHDbo9nPy+iUmRSBRWaG9fFwKfeU93u+ZNUptBrr7L9m8SYShx9bQ9Z4hWqIuCL64V2vjCQNUueLps0fM5g3W7th1AZdqBn9L6Hs0RzrfkaJnGAa6MLAfAiEk6rbBiCNmyOsBvXif+dX3yaZBMWRbE1PGoThVCB31/Q3V7g1+t4W0ZtvdEf2WuqppbDUmVCawjtjtiCmQNJEoyXOihhgjIUasczhbgcIweKr5HLFC13cYq7hhw0wyNbDXQLSW6FqycaixpdIyQjI1SRxOI5ItoqZUdB4tHmVdP1btFbCuWCxCzqPKib51v40xYIu7eYqdTWmEy0kMaHF1F5BN9RGm2NsSA3xitVOhrlreefacx1dPSGOF65xHJRVTwFlVYzKoGJCMM5apuNZkLbKVObiKkaJ7LwaMrbDWHvoWY+SAIj2tfDwex9px+9LnUmysWHIO2vyj9aiqKmKMbynYTMeBEoubcial4qZW8glUAEoFZhEpfUEwxpVz6DZ88dkXvLm55tX1G16/eY0P4Wv35Rts3yhD4MyR34ojcc22vyX63b+RI3kaV6oMOTKkiFNH1TRvccTUZ46cOfKwOHJmyJkhZ4b8fjDk21j8/EvgPwV+DLxHibn95yLyt4F3Aa+qd1/Z5+X42S9r/5Cvx+4Wl9x4g/Swlj263WR0USKCjqvHkqA1rnRVUU2EMKD7O1z3hoZQPpNEVRlihm4IJFUWi5Z3nj3iYtkQujV5LBSWclFKCVrAk5NSNxYfPLVUtM2Cpp6jCXq/xZmaWWOL1WXXA4oPHSEEsipJwMdEzGAzJIqKiCSQ5hJpLolqUWNJmgAp1X9zRoKH/ZpFHIh5S/RrnHjq2lLPHhNCwoeIWiGJYGPCh4APAWMctbUYaxn8QIgJMyasKToWZxvlsceCZy732Mrjmsxqbvg8remNkswSFQsKIlNIgGCSYLVYF8xoQSh365iMaFQLEGW0OL11z8eVf0rjQ8aM4+BtF/VX2yH21trRsnNMAiyu3aJY56qxgvUYJxxTJI1AyjmOEKionCtKLKojKMrL2tFKMRbzEpGx+sPbfTm4kUXGExSwFOlMDCkfLUyT6klRlZnUVwqoFWWsiPbWA/Y0fnb6PmCMyy2weXsuTS5pQ+Fy2afv9vRdx/7lCz779BO2+z37fYdDwVlijMSc+Ibbt8EQOHPkl3MkF+keI7+AI2lLuH+NDXuaHDHNnGgsMedfzJGmom4MMkToe0JK48P9zJEzRx48R84MOTNkvOdnhjxkhnzjix9V/d9P/vxLEfmXwMfAfwx0v+Fh/1vgH5/8vQI+G7+RIqt3BNBJb8aVtpAsQFlRi0KmFIQiKRp7Yr8h7dc4HTDllqMpM4RM7wPWCqv5nKePH7Gazwi7e4KPqDFsvSdlJUTwwZeJIIlmVtHWC5Ca6Cti8lRNTYyemAwx9IjJODEEa+m6niEkkkKIkFQZgmJMpmlasvdoc4lpl2RnSxVekQKCFJEMhkzs7sj+Hslr0nCPJo+pHLZuEOMQ24AzpXI0WvqcPIghJyWmhJgy8LyPGKMEH3DWkFJkMWtxzqA5FVe9LUmcqb8jDz1uCZiaaGcgJbaTHACBMclPjGLMdH8AneZ4uS8lxngEwtHbCSfWiKILz2HyTvf+1Npy+vekiDK5bFV14grGGJJOVg3QMeHPiDkooORxujhTjV+vnD7gJjdwcR3LAULToxA9WnbKqZwU75KyhB3/f4gLTkrR+IcjSDLjMcd424Kdg5WlHHdMuhwftKf7owWvmV8M6QlcMQa++OIFN69e429f4aqaHANmLHZWGUGyPSZrfkPtW2IInDnySzlibm+JdY1tHUblwBG/fYN0b0jaEcWTnbACap/pXQVN8ws5ogmyUUztqDol9oGkcubImSMPnSNnhpwZMg2wM0MeMEO+dalrVb0TkZ8AfxP4Z0AtIo++YnF5DnwtLvfkGAMwTH+/dZFklIwEitxeOlwwEcG4csErjWSNZQyrQ9WVyrNpT958Rtq8xObMPgzosKa2A1K17FNN1yeuVgv+5o/e5/HFHL/fkQYwybLzifsuMfhILYYnT+algKjAfH7BrH1E3VywWLxH5VqiDgz5NZ3fkPpI6+bc39+z2w1s+0TMSvCJbZdQhRnFKtR3HRcuUD15l9mz96AxVICII+aMNSW5T21F5YTUDdC/KhCIkewdfddRzy5x9RIfM1W9ZFXV+OgZfM92vQExzBdzfEjkGAg5oTEiYtj1niF4MIJrKiRHnLEIBiFiTE8T19y/vCbO3iNf/SmxnpXEQN1TCSRTF9BYMGaMlc2QYoZUAKBSLA8qU7IgTKWOi0uW4mae7rGxGOMKk05iVMtYKbr6h+S5EWoiRzgYsaATBMaJnKftjnGxRcK0jLfCllJsq9iLDCkHYioQPrqTIZNwzjG5nlNKh3F8+ooxHoGiMFUEh3gY80bsaGU5KsSU4uLFNV/O8VgBOufxGnC0TBpjEM04EcQ4VDj0K+fMZrPhi88+5eOPPuL6+jXGCPMK2nbOkCJ731PkSivaRYMf/C+d/99E+yYYMh7nzJG3OPLugSPGOFLvyblHZ4sDR8x+S9Q7UthASmTruI6euc6QuiFE/f/miEkEFUJIqD9zBM4cecgcOTPkzJAzQ34/GPKtL35EZAn8EfC/AH8BBODfB/7p+Pm/A/yQEo/7a7YTrfxxII3H/NrLqiXpdGESUCEIFYlhf03sNoiz5JiwBCCz6wa2QyQlqB4J1mS6fk8OEdTih4FuCMSkxKDMaoPvBpDMxeqKulpQ1w2qCR+2VJUriWQCZozXDCEAhqqdkfcDfggMsYSy5ZwgK+2shaRUFdTLFbapiDlhjSWkUrwqo0U60zkwwm63Rfd3pLCnaRpsZRFbYQSapsJKcRMOuz2iwqxuMBdCysXaYa3FWaXvI9aVOFE7TpquG6hqi7OCFcNutyV3nuXFBY0oc1GGsMUzEHNDMmZUMUlTVbdiJSGDjBaQo7lltAKNxcD0KIUIozNaOKiilPoKMs66wo184vo8urI5kbYcLS+mFNQqSpBaLHBpmqxvjeHxCo+jTvNB6QUgSZG9NOLAHN3Nk+v2aCoqvx+sQuKOsbKqpDQUYIyQkPH8KY5+Di7/0qnjLBitKKKlHxOMjuAt+07bnpwZR7d4eX+/3/Gzn/yYly+/5O7+dpw/EBX64Nl1e8QaKlcX6BhL/oYttl9t3y5D4LvKEbfd0OVAdfGIcHmJ3NyUUBJrQApH7sNA3t+DDLRt4YiVimRrltbgnCNaYdjHM0fOHPkOc+TMkDNDzgx5SAz5Nur8/HfA/0ZxL78P/DeUGf5PVPVeRP5n4B+LyA2wBv5H4M/0N1BpKgvpMZaWsaLsOECnVWVJQysa40bcONAtooIJAe3uCPdvMGkgZIP6PVXekbNjN8DglVnbcrGYYUjsd3t8F0g+4jAYZ2msYz5fcDVf0Lb16MIUjNSklAggaPY0AAAgAElEQVShY7fbsN68pqpqomwwJuFUqFzDYmUJd3cYU2JvrbWQI5UtA7KtK0IIiHHMlheErIgD0TSKJzKOqXHS5EwMAxoC6NFlqiYjNpPVMwwlMtkZi8UgWairmiEEuv2eECL7IRTgWEtd11R1TU6ZPvTIrmM1bzGVYV5V1IsVrqq53a7p1x3RDLRP/oCdALKEsbBWNgZrSyJj1lKXOE9a7gLTxEAMWcDwq8VwTrGzRWjn6PodbSMnbZRvzCVOt1hATmNu02gNkYOVokz4zBSbWp5vp/5vSpEzFCOWjEEYlWJMHjXz5cR6AmAOiYtpciVnIcWyhXMTUAVVMz6oRklNnZRYjjHmp3G1RszBVQ4wSW2mNNUVEFBIKaJaoB9z5u7ujo8//pgPPviAEAaMjEmV1qJG8SnhmpLr2zRtGZ/TA/8bbL9LhsB3mSPjwzePHHn8GHe9Jm63mFYgRcx+T0wB40444gDJpBwZggLVmSNnjnynOXJmyJkhx3F3ZshDYMi34fn5PvBPgCfAa+BfAH9XVV+Pn/9XlEXzP+WksNhv8kXGCNYUCb9SVVdLgtjY8niTUMhuBskDQsgJl25J15+w+fivmKUOVc8Q9sysp65r9rklZMVIZtVWGO1Z3+/L6r6qMdZhnaVxlst2TtXMMdQIDc44ILPZfIl2kaY2fP7ZpyCJ+XzOYrEgqqLZYaYFtZTK0DENxKw0s2LZmC9qco7M24bZ4m9wcfEOSI2qkAgIMqagRQxlAPS3d+jtHX3Yg8lkLFED7bJBTSZrxFqH94kYIjllYggMMdD5gWEIZD3GbXpfLEp5N6CaS4ytJJIOGJu5dIG8XZOqOY+WV5gs3N6/xrz6M9rmMal9B9P+kKqeEXAQy9RVhT6E0Ys8TmMpLuQ0ubAlv2U5KD8m128BRMlPKvddxI5GmnFqf2XS6WS9yAqaibFU4y77jAl4mkeXczqJ2c0HN7aSi/ubqbiZIGKxtiQqlq+VEWiQNJB0SjVUBDsqoxyvcc65QMsIlZmsNVNy5hEwOcmByUVRBsgF2uX8BcYkwcmSM7VyvAlOxcXvfc/r16/5y7/8S+7vbomxqKbEGLFGmNXzMo/UlH45oalr2rYux8yJ3n/jldl/ZwyB7zBHnryLub9myGvs6lF5eGjEUVFh6e/WhG6HbrbEuZD0yBFbA77kGYagJbT2zJEzR76jHDkz5MyQM0N4UAz5NgQP/sG/4fMe+C/H12/VjDGYcUVsrT0ZlMeVZpEXLGoiFaboiedM2L7h/vMPqIY1TQM+J5x4UhjYDZGOmv1+oK0sq+Wc+byhskVgoB8C/RBwVcVlvaRpGqqmwjUL2mqFEcNue03dQvAJVWG5vGDfrTGixUWrsBl6mqbhIOknGWulKKCoImJJGTLCar6C9hm2WYCtCCTAMs2FIuOYIUca4xDnMNUSHwaMacoEMAZnXbG29BHvE2EYyCmxmM1oaWBnSnG0fiCnjHOOwQd8iIQ4XkuKR1sS3G17Gtswb+sSXtcucM4y9GuG3St2t68I8iWy8syWV9jVM9rlCp+KMo2qIelYDdmMLuNx+a4c42Gn+3p04Y5bKJAzKnl0z06u4dPtj/Glp9KLBygd5/D4feNXnujTi5hDfOyxuvE0sccdjGFSO9HR5XsAXPkx7jROeuREMeXYx6kzpfLS9F3FajglB56ef8rpYLExxhTrGUeYTeCZwO29x4fIvttx8/o1X3zxBd1uW+KDnQOUYZTUDDmBKiaDkwoBUoh4SmVlROmHQwj8N9J+lwyB7zZHogqMio4oyOMr6rt7QojUFF7obM7eZDTbUhtQyjjvDQTNxHDmCHDmyHeYI2eGnBlyZsjDYsi3nvPzbbZpMEyr9KlNrsLjwDTkFEkpYYg0BG5efYq/e8nMRLyPBL9F8ISU6HxilwaquubxoyWzeU1dV1gRvI84W7NYtBhncM2culkwX16QRDDOcH93S7+/J8YOYwQxlufvfI/buwofurF/lsF3tLN6DBHN1HVNytCHrsxoU6FSM28XrC6ekC6+X2rxTBWhJaFANU0kyZSY3QWxqrE4nBSFj6TFAhBixNUVxgmtqcs5dXeIUWrXUA8JY7oyvFVxtcO6in034GP57piEfkgoBfj7IdC2YEbohxQZQqDvbklDJLBDdIXfXCPrO975/h8irgY1KA5jzTgX81gArFgXjCpgDvcY3lYnKZOUA5Cme12aInLc5jRWd9r3AI7RCjG1Scv+CJzxck/H0ENt4sODQWRMRjxJ6DtYQ0YAnEa4Hl3rXx/TE2RUT/aQCTdHm810nqfnMvXxdNec81tWF2MtPuz58Gcfsr67pdvtQMvDTrMS07i9lMRHYwytq6mqkjjpvce6mpSLjn+JFX+47bvMEX36Lvbmc9J2S7tckYEsJc5cpMLOGoZ6RrvbkxP4pLRZsNue2taobXBte+ZIeePMke8oR84MOTPkzJCHxZDfk8UPb62yJ9hMLjWANvfk7OnvXrP7/EPs7c953BSZ6H2/pmIAzfSpYpscQxBWjbJsHcu2xmDJSXBUtO2cer5Empa6XWFp6feJIa4R3WOMMHSJz19cM/iOprE8efKcpl6QouVu7wlhS1O5Ap+UwRo0C5oNxrR0g6GqZzxavM/y4gm2XWHe/yPMqkVNImTBisVmJcWMOkO2hlw32Kun0D2n7XrmKM6WysYxZzQbYoT5bAk4ejE0s5bt+h6/2zGEsmJvmwZXKVksWQwNhv2Q8CGXGE5xMGTmCp/v3nC32/I9ybz/3pwhCoPW/PSjO3L2ZLnh6fdmuOaCsP05n9z9jPnj77G4+h6s3iNjUclYq6gpVYkllcVaHBVzpvt9cL+qjjA3b42DmNOoSz9aWyheMeDgNi4wGKsw5zESd7J2jK7uKQ72INnI+HseXcWjVYbJGoMt7mdRDqWfDyaXMs5k7NPYOzS9XdgrjfHGVsBYc1BpOR3vqCBZyQeT0AS3dLDYnFqbciqJqlmKust6vebF51/wyYvPScGPCjfFYyjCKCEamLU1rm4R6zBGaG2mbVtEhGEYuL+5LeCxFsnKQ27fKY5ES37yfZLN2JEjTiyiMNxtMJcL9NkzePMGLq6Q/C6zbkDaFmsqshHYBdJyTm5nzJsLzhw5c+S7zpEzQ84MOTPkYTHkYS9+Rh/hYd2pUwwh42rzuJyVHNls1rz57GO4+Zx3q4yzhm2IBxcv2eK7hM/KEDyrpsRDphTZdR5RmFcz2mbO6vKK7CpiromDIwwBHz2usuSopGgIngPwck4MMdD7MVlNHJvNHZfLFUFDsa5gmM+X2FoYbgbqdsnjq3eYLy5YrS4ZHl0iTohERGqyZixQW0enpd/WOZJ1ZFcVTficEDK2qqitI2QlhoirAC2FqqqqYjafgXi6vmPoemJOiKswxuB9xIcIImNVX2UYBqImlskwu5yDZHzo2e7usG7FO+/9gL/+yc/JCrPGEuJAu8gsKsOGiMaObr+mXT0vLlo53sjiPc7kLGRR7Fcm0dFFfLy/eVSuyZpPxsFkk3h7QmRN5ZXz4aPCh9Eac+JKLrBR9KSgF7nEc4spZphJ5UWZ6jWAiB7cyGAwMiq9SLFoTKdgrXkbEDmC8pZ15LQvv6hNbuQprOLU+KQKeZSNvLm54YMPPuDVm2tUHMv5jMoIOXmMGIxA27Y457CVw1VNGS85kjQCmRgTw9AR4oAbx0dT/wKz0QNqvy8c8TmQUnlYfpUjT9ySpVZUT5/RWosMHeH1l6SLZ8jVI+rrW6SeE1XwWalchWTwMdOogjMY56iqCpZm5Eg6c+TMkTNHODPkzJAzQ+BhMeRBL36KALsdq/AGsgjeWFJIOM1YzWRb0Ss89Ur30f8Drz/lsnWYKhJzIPceAwwaMMYQOo/vAhtveHoFFxeGZdOSUyDojmSVQS9xQ0X0SpKAqSsyFRJa7u4+RTXQdwOZjv2+pwstndtQVR1ksHGgqZQ1Fb3PVNYQUawRrM2kEKjrAh/XPGY9e0L1/g/oF0uMLqiz0uYBbyq8MwyaqcQg2aJemA1bmnBHX80Qin69KNiYqCQS847h7g29D+RsEWlAV2jeIrKnqko1Z2MztsrUYmhtg82GTezpoxKyY8hwbT1XPpXPNpHnz1piGNhv1sVlKY5ZOwfN3K/X/Gh1xerqkpvNNWn/irh+gVQL7PyC+ft/wN7UeGkwsWcG9GJIWUFkVJ80RVkHMFIS/3IyiE2IlijUgoHjhM9aLCqaj9YWHWGjCllPKhzD+H4+WDGOBbs4gKIy7mDZmOCQc0ZTiUstjmiBLBQVvlP38QSTovwzhUmIlAecQ0/c3EIaO6oCipIoEpiGAkqbI7Uoah2iDpukuNlTwPs9Q7fmfrPm089f8OXrL8jiyBnutzuMEWIYSJKojMHHgDVzslowLZGeoOD3ihAJfsD3e5qmAuNIWfHxCOiH2R4+R3b7wNJHEsIGpTrhSONa9PJdbq7e4emPfkj/5Dk5LVjd37P0e7JbEW1NDj0MGTGXpMVT2tizCHfEuljZzhw5c+TMkV/Wzgw5M+TMkIfEkAe++AmoenzORWtdMrYsdQkKMWdm8Y79q5f865//NS33PL+aU5PIEUQNghJzJubEsBu4HYTX9x4fhKdXf8B77z4Hn9j3maZ5ynz5CGOXbMLAtt8Ts6GtIYTAqzdfkDRiTYXNjmyW2Nqy7XqGtWEYAslnGjoulxVVk7i/v8eZxHK5IqnS7QcE5cnVFcsn7+EePWX5vT9GLy7IZkajAcERpUJFcXkg54AxDXMdcLtrwt0LJGWsBeeKKzDFwH7o8L7Hx4Hb+3v2/YCIo23mLOaXrFYrFosVvR8IMeD9QDf0VLViKzBVTTNv2Xaem/UWZ4WYHTELGjLdvufFixdcXD7Ch0QXIstZjascmcR6fc0//+iGd34QsSbjSPSbF+yGjM/CxQ//lEff+0PmT96nrufscyjxrMYAtrhqRUawnMaSKhqVLJOU41G15NQCobnE8Z6CBEoc9DEGdzziyf7TZyKmqPpYe3ADn1p/ZHIvH/YfVU+QItByYpSYwDfFqIqU4xpjiqv5VB0lTdCElIoCCllJY/+yKHIotDa5mCND6PjixWd8+JOfsr5fQ4ZFPWeIA4PeY3DUrmZR1WyGcj5dDFSVR0Nms35NUwu1gYoaui2NFdqmqP64WkhJWdQPHCO/BxzZ7rbsNfLEtqQM2xOOzBbPWVRz0o/+BL24JJsZVQzlob3rSfMF5tECefMKzY4273H7G0K+Q2LG1oJrzhw5c+TMkV/ezgw5MwTODHk4DHnItAFN49qzpHEZLTKLUlXkkgLH7Scf8urnP8GlLaurhtaCxbDrMiF6YvKE5OmDL8mF3tIny9OrC2Z1Rbfb41RIEYYcwUbmSyFo4n634e62p9+/JGtmF3eEmJk3M5Z1xbCJ1LXFOEfvIzlX9MOAMUrXe4wmYu1QUfq+Z7sbsFVNZQ2zpmGxWjF//Izw6CnBgBqLjNV/o1gQjyVQCUDEdHek6w+R/p6YEkY9bd2QNWFsGaz90LHdbRn6nuImLSt4Y2E2m4HUmL4nxICrPK5uSZrphwFPByHiKsVZZd91qG2oqrZ4i60lhYhmpaprXN0QYijJnQasRHbMeLPteHIxZ9lWDPdvCN2OLJbN9edUsyX14hHGNjhrxwTMceJNE1omvXzlGGtwElYgx4RCYHQ5T/GzAuOIKUcolpfJ2vLVmN5TuVKREteLsaOaSkJODA1vFS6To7KLqmIni4xwcE0frTh62HZyWZ/25avNjOec5WiRETGIKqRIDp5ut+PN6zdcv3qDYnB1C2Lok6IiLCtL62rAYI0Sqgq1BskZHzo0RZCAy4rNgm3NAdzWVkgyuKomq6f6BW7xB9V+TzgiOXPvIul6j1/NaGvHrGl4XNfYd9/HPHpGMGBubmlzJj95n2BqVAacBMzTK+R2Td5fE3efYfp7YlNh6GmNOXPkzJEzR35ZOzPkzJAzQx4UQx704kcA1QTZoOONbaqKKDCrDGG/44uf/TWuW7N65LCacGLHolJFPWLwPTH1xJTxMbPdB3ywzBvDxaxiVs8IvUekrNCvr9/w+vUN6y5wt+3Y94LmGh88vWaMmbHZDNy7AYkD83miboTteo2hJWtGamXQyDLX+DiUQT0EirBHcc/O2xntcoFZXqDNAsUjYrG5OC2zMSAZN4W8+Y54/4Lh1U+Z9beIrdCY8MNAyhFjhJACISd6HxFb01YVIpbKNYgIIQasq2maOWIiiMNUNapFVa7zPSmXwqlVZTF9yY/0IbKYtzhXsd93dMNLfvrpS+43Wy4WNWqUWe3oB6CZs9kHVrPMIImrywtyjtzc77l/8wXV7BHzq+fML5+ON3i806ViWJFyZCqQlUcAlc9EivVMc0ZO/jd9VhzQowtYZASX/FLYTBaQyZ0s4z4qk2sZRMaiYAery6TGd5R0zLnIdIoI+UQJ6NRF/dX2Vjzxye9m7Lue7m8KOK0qTpT79R2ff/Ip682WNEqEmsqVegaSmM1aLsRQVzV91xNDZjWbYeqKIUXe3GyZ1ZbKtDSkkgxrLUmFEIvWv6kqggpSNVixX+v/Q2oPiSO7+zVVcGRrvs4R6+iGiLMQJGPEMG9n8OwJM1PRb/YYAvHpcxgfPH4YIG+o5o5KLBI6Yrihe/2zM0fOHDlz5FdsZ4acGXJmyMNiyINe/JQlZwVaLkIwJcpylva8+fADvvzZv+JCOi7fuUC1g6T0aaD3A10fGAaPSMZYSw6JnKCuDM9mF/zwvZaZC9ze7Li93TJksG1DjIn7zZZNnwiy5OWd5836ht2wZ9MlJLdUVUXbOJbzRFvv+OF7S+ZNQ9gVqJjWYKqWlBr2acfVco5zM8L2DiuBZd2Q7QJ38S5pdkkqnlVqlCw1WQUkkgQiljomNv/6/4abD9D1JySjZKmwTrjfbQgpgjHcrrfcb3d4H6hciw0F2CHcEtNrrLUslpdcXjyhbpfMFpdUtWPfbbFVxerRBc/6npv7Dby8RZNwu/Vc320YYmQ5a2id0tYtf+/f+/u8/Gf/gqp2hJBwM1jVguiA98JmY7h4PMPHwN12S+8HLi4ecf/6c9Q2XD59l6ppwDQYW5LZdHTzGoWc46GYFkCkXKMxGPcAidM2QcQY85areNp++jn9/lXryQSNqAV6qqPSSZ76oYxBvqAZES0WGTnG4xpKtejp+KfxvYefHL8/50yK6S23+lf7Kgh9t+PN61dcv/6S16++pO97mrYl+FIl24nhoq5YNJacAkIm9feo75m3LbPlgpube5If+NGzR8zaipQzfR/puw7fDbSzJbOmwlUNWSND30GKJP1m63P87tsD4ci7S1biCDEx04Rp7dc4UrsZ+3TDqu9xbUu2C+zFc9JsRUYJjy6pRcmmcMR1dwyNEsVRp8Tmy7+G25+dOXLmyJkjv1Y7M+TMkNLODHkYDHnQi59Ulo9FIs8KiCGjvPr0I17+9K+Id18iq5beW4SEisGYTOd7hqgkBStgxVFZS10nqqHj2bML3nl2QdU07HeZu33PZgDZldHUR8v13vPi+jUfv9ry5fU9WRTTzKldjXEB2wVWXriYCbOl8r1ZjZJQDfiYsQq7HLEkqq5D05ZH7YLaZKwosVpB85gsNYaEEQuaSaOb0RBJgGLww47h9ae0/Sty2GFmLSkMgKPrB/ZDT0iZPmb2nWe39zStYT6vMUZIlJW4oCSNiDMslgvadoF1Fa6uGYYd+/09bTPncmnodok0ykj6bsNuvwdV3KximzuaXUfdzkqV6JipbI1HmblMHMCZ4jaezee89/73+eizL9hsdiSnbO9uQSOqDjtaLKy1qJQK2nJw144xsCJwqBY8Jhl+xd0sYsZJKxhzGiebD3G58DZwirpOOlo6zBjLOlptYDQGHdzFR2hM8FGdXpPFpCQgnvbv68AZjzDG+qYUD/u/lQw5ttAP/PwnH/DhBz+m262x1mCsIY/qf/P5jNYJlUYqE8g5kTKIKItlzXw+AyNURqmamstZS0wRHyPd4OlDYtHMmM3m4CqapuXN7TXWOowzqH/IicoPiyN12yJ4NEd81K9xpNnccdHMqK2QJdPLnKvVu+wXj6lSxCTFOC0PaAGjCWyFIvihZ3jz2ZkjZ46cOfJrtjNDzgw5M+RhMeRBL35yVLQGxCC2QqTokn/047/Crl8zl4j3A1pVNKYMpBgSYg0hDcQQaCrFh8x+UHZDIPQ9pJ66fszi4pKYPUFvuNsF9sGQsmEfez65fsPHX6653YPYllk7w81WUF8SjRBFCd0WHwOzqmf5eEadHRCQWlGnePU8vljSzmxZuQalai1JM6t3/4g8e4bHFBxky2E1D8iosEEWJAZMfwe+FIhar9cYikt0s+9IYvAps+56dkNg7z3iWmQIhBgJIYwT1SPOEmJmu+9RaZm5GrGOlJR9N5BSSeh8tLokJdjrljgYEEs3DEjsqdoF9W5HStDte2YGYrD4vmcmNfchYOSCFBPJFXnLx0+e8NmLW8LQEe5u2e92XM6aw72eJisiXwOEjj5pzZQQ2sOEPBYZm2AwTfTpVSwq+a1JfLrd16ow5wzGonq0xOSTvljnjsc9AV45jiCSSRzjar9qAZreiyGgqchZTtWcVfWt6uExBO7v7/nko4/5+U9/zPb+jrZ2OGORLCUGOCuRRB+VqIEkAUmKy0WVhqx03R40lGrfxpF9YLPbEDQTolC5Gls7zKiqMuQAArZqsKJs991vP5n/f2wPiSOrxzNqWmzYMt9G4tzg88D7rmGRhc3K4tVha0PdeebP/xCdPSNWDXVaI0GRzRvULrEpkh9fUFcGUj5z5MyRM0d+w3ZmyJkhZ4Y8LIY86MUPksjaI2YcmH3Hzec/J919SkWHj5Hl7AmNOLabe7QOeL8lahxVVSJ+07PvE32y9AmeP3+XP/1bf8jV45oQIq6yPH32mE3suL8xbCJ8ft/x4TVc9zViDd97esXTyyWS4IthwGdImqjjgKstd7c9LyQiWbAm8zTtWbS2FP0i4vceTdAsGhIGmb/D8k/+Dn09xZqGcXXvMBpwImQUTYmq21GtXzKXwGa/xc0s68FjspCJ3O12dCETVeiisu0DQ1B8N+BCZvCRzW5PVdX84L3nZDvn4y/vWSyF58+XBJPpuz2bzZa7mzti8BgRwhCoqpq2rXnzxrMblOW8Zd7UDCHy8Uefst3uIAZisFiEylguGssXOrDfbVlWM7xG+piwKqzmDX2Cu37D/XrN6uoxtioxf0kVzQmjijNyAgsQY7AiIKWA1mmb3M3TxMYIGFPUTkQwYhHsiZVlBNvJ/qfAUSiuWikOYUuJe562LYXJBCzEGEeFF5jglzmqu5xadk4TDVNKxJRIMSFfsbAYgZQT2+2WP//zP+eTTz4hRY9Fadu25Im5mpAyPmcIGaMZlUxTO1btnNoJQiBGT105Bh/ZdoHF6pKYlbvtltliwUKEar2lcQ5bO3Z319R1Tds4YtyT/Jr90FO5hxurDzwojnwmEZOhzcqz2HHZDbimJT6GnUY0Cs2iJWao2kuePP9j7hdXmLoiLVqMQjBC1ZSK6mIUTfHMEc4cOXPkt2hnhpwZwpkhD4khD3rxI2NMo6KID+T9hpvPf47NAxjFtjNASH4A1RIeNvRgDSlHfPSkEPHZMCRhSGVCN1XxtlgxzOuGRVPxaKX0SWDrucxznqZ3qReJYb+hdYnHM1g4MEPmPkSyWp4+u+K9RyvWN2+43QXqek5thCFHGhUyNbaCylTkLAwxUS0ucMsfkGeroh2jDlUHEgCL0UBWJSKYMJDXb9i/egE54nPG+4xiSQa8D0QFNY7eJ3ZDpg+gYkkIMWZ8hKAOpeZ2n8giWNey6RW92WDv9/T7LdHvSCHSNDXOGtabNdn35GSxrqYfBrqQeWwrGrEMKXN7e8usAnOxIudMVdeYUJISO+/pB8fjxjD4BCHT1EL0UFfC0O1IMVKP7tqkCWstWRMpCXBicVFlKuB2qlTylua9Fm360xhbACOCQQ5WjBLbmkb38PE7Dk3HYmV6dK9+Ne51+t7iDocJYKWLcnIoffv3Ue//kBhppFiPdIzJFYgxc319zY9//GNevHjBMAxYyZi2pl3NqWZzhhDZe0/KicY5WldhNUEEm2usADaybObklBlywFlL8ANqLYvlnPl8Rux6LtuG5WLBEFOBlk1IivjNNSolPKFuVr/xHP63oT1EjnhnWGlgkYVd0/K0FpyxGD9yZH5F/+iPyM2cZBPWNCg1YgJmvsBoJFP+IXLmyJkjZ478du3MkDND4MyQh8SQB734UQRra2JOMOzYvPqM7asXmBRQ67B1UxIIY3Ev55AwIoixpFD04401RJ/xClGElBPRd2zuBqJpMBK4v7mmNjPef3bBfG6Y9S1SRza9sN1UxO415IFZpTyrDU2fUeDdqxYnCTGG9RCQ4Jk3jot2RpWh8rDrA9kYbDLsRXhy+QNW7/9tggASsFlAHVkC6BSlOcZ89h15fUPe3fD/svdmz5El2Znf7xx3vzcWIIFELrVX9VLd02RrRqTIkckkkx70oD9YD/MP6EFmMyOR1AxlZPewl2p2V9aWSACx3OvLOXrwC2T19JNUZEuQhZtVVVYiEIiIe/0H83O+833zdGTODTfDRQiaIETSGGhVoFWKzeSlfSkxUszYTZnmkVIDX7yeeHNrhJjQOLC+mUgh4JbZriNJRpAIQTCNTPNEbY0YB6Ipza07cATFmhFjYLsZ2WxXmDWm+UBtI7k2bg8z2zHg25FVGnCvHEthO0YOJXP91Su+//1PqbWimno1hXuH/f/sPnDH7G1b+D/Xoz4M+IWAxvBQQbl/XFx0t/fVm/YwvPiHPwuERSK83IN8CzDSnX6k2z0SnHv7yrfP5n/wjMoCTXNMKs1+X4Pr3je2iPDq1Rf83d/9Hb/+9a85Hg+I9Pct7m/h6bBSQAPiDX3mJWUAACAASURBVBaHF1T45rCnXu/ZrguXF5eIwZAGhpWynzNhcWLJhz0tZ67O18TgZJztWcKtMR12uGXisGa1Oef8+bvA//5/Y+f+f2s9Ro5sh8izuOIWOD8Y+8l+nyOXH7H54KccLp8w7l5TzxXiunOEt/fXiSPfug9OHDlx5P/hOjHkxBA4MeQxMeRRH36qJ0oRosDrz/4jv/w//zdsuiM6JImIC8MwYFpwlGm+QUWhBWozkH4Rh/VIzsZ0KHjJHK6/otKIYU1pb7g97ri4fMaLszOereDZ0RhMeHXrfOkr3vCSL+4KuzlzcSZc6IgQuX29Y6pwtJGL997l6y+/pE6Vryfh9WHHs5WDnvPizLkYKp988j+w/v5fkj75EwpGapVIocga80ikUjSiGiBn/OvPOc97jvXAwfqGbaVbQYpWHKi24jBV5jnQfMCYad447CpzqeSqHJdKj1IR3/fWbUoMw8CYIjXPKI2XT88YkzAkJ6lgHil5ppky517dWemRzZA4TpUhBtargaAC6gxjIubIPje8zmCZT59/yNlmDTJzu/+a2jJnY+Srz37O36XAj/7yf2R7PnR3kxCwZpQ8LZWNe63s2w36beeU+3ZuHw4UNASQt572Dw4lon/wPX3pYmd5X8W51/qCe3uAFq4PLWJ3EHNUnaCRoAFzW7TMby0rVZVaK1H0W3A0WmuU1kPJVBWNgTJl9rtbfv3rX/Ozn/09r19/w/F4pPUpW6Iqq2GkHAuUHeqV0QpiFZXGKg2cbc9pBt9MR2rL1LuZkt+wWa9ZrwL1sGOdEuuotKZ4FKokdvOOWgvH1pA6k8TZrldcPP0EGTts1lefAP/zH3Hn/9Oux8iRNlc+Xyeur294NjplvPwDjgwf/4SssHqyZqUzVcKJIyeOnDjyz7BODDkx5MSQx8WQR334cRSTiGL8+ud/Szm+wcqRYXWGelisBg1TmHPFHdSVYl23+TBwRgOFEGGICm0mF7BBaOpo6kGkwQur2Acb370YISmzJI62pVpgbntev/mS0RurQckhsKvO+bMXFDfG9UgCdHC8zRSfuNkrT0YYn0Sevf8p6Z1P2Q0bklZCA/EKoQAbhEwlEFxIolAL0jItT4So5JLZHfdstltojVyM3ITDDLe7zLEGqgvNhDkbh7lxnBu3B+cwg7eMtak7z6TA2faMly9fsF1vkVq528/UMdAa3Ex3lPnIKkSIa9ydYsY0zSQVSsnEqDjG7e0brlYrUooMIVKbMRdjEyP748wQR4ZhZLvZMN/e9KpHEX73y/+EXH7Mn//FXyLSqyWt9pazmz0MF76tYPx+deQBCNCh0vvHD9WUEMLi2CJLtab/0x//+8Fg/b/3ji724Lziy3MCNBzxt98bUuS+OvNtve490EIIXZcrb1+Dmy0OmW9B9/nnn/OrX/4Dr1694ubmmmmauB+6FJEl/GwZdhQ64N1obWK1ijy72JBS4HZ3INYZBUIY0Jho7hyPE+qZ7SqwDjDlSvWuXS6tUr3iubEeItvVwKBCGCJXL59x9uIdclr902/uP+J6rByR0ZnDSLEjxzfOPOrvc2TcMgyFUKU7MnHiyIkjJ478c6wTQ04MOTHkcTHkUR9+kjWupHD31S85vvkCdec4VVbjjKkgOpCLUApgBTRSlwuR3CjmoCCuaK1scDZj6PpBaQwDNEmUVqh1Ym57wrhhXA1cNKe2iXkzM+8nmmwobWDOT5ha4VCFot3zXe5uGEmMrmxWAc03pKD49gkXq8j58ASJLwkf/5Tp6ROqKskjk4LrQKyBjVXmaBiJ1GAwo23Oegs6nTMfZ+7miTe5cVRhTCO5hm4nOTXusnJTlWM1xGGejNt95W52bo7GYTYMZfYnTMWx1ji7nXl/fsV7z85592LLWoz57kgOwrEZxwyrVLncNEaDOQdeI0xqqAihGqEKadzS6swYA0lmrMFclUMWDjkz5j3jOLBdD0xlTZthyOBlx/XP/lfkJ58Sz19gFcz7IKGL9IjiauBdH2yqqPd2a0eR0azbgwuKetextmYPKdwuhmvrloutgXcD8b6fO5iEvvFrLagI3gxzW+DkgGAG6FIBAgiBFsICNV8qQ0ZAca8Pr0/DPTQBMTQKsTi1GaU07vY7/uo//Ee++vorSm1MxSneqzUxCkMIJBXWQ+RsDKQA3iqOEYeBlAbmAsfpyDzPtDp3y0oExSEFCMZ6tSLpQMmVUjKaEikOsDfynIlWSOmKwsBM5MWLTxlffISeXdDKH3ff/1OvR80RDdjmjIsxcD5c/SFHSEwqJ46cOHLiyD/jOjHkxJATQx4XQx714West/zu5/+B3/7ir6l5RzNFdUQ8IRbxqrToFHJ3q0CXG8UIqtTWZ7jMHGvGehxYDQMajKDOehOIwxqRxu4w8caO8GTk4vycD883vDBh+9UNwjfsc+b2sGN2o1btw30x4AnMM24ziUKsjefrwhjB6sj5+bus3/+Ij//iv8bPL4kO5y5dvxoEX27qmQYEtjhmEzkfOb5+wzBdcywHcslIVfa7ypvDnuQHClv2uXGozl12ro+FuWS8Fnb7AzeHzL6AhIEQAsUCRwuYB4JCtSOff3PLl9fX/HIV+fjpBVdnW0SNm3kChXOBValMZaY2OE4w1crZqIQA59sVmzVsRiFQSUPCvQ8aThMcp4lVElqrxGHg6ulz0mS83n+JeOB48wX//n/5N3z06U95+eH3sLClsLSHq2Gte+NXumtKFFuG+ZbWsd9fZKFafijOGIKV7v6ioQ8DundLSPd7O8f7QUIHDfSKBpgrjnf/ejdaM0KIS/BZpKc8O3Wx7lR9O1DYWiXE/gJ1+ZlmhgHH45HXr9/w+W8/5xe//BXTdGTOmVwKaRwwM0ou1NKtIMuciePAuN0Sg4BXgiRWmzVBVtRSUOFb1ZyGipEGZYywXidSSsSgKMaxNlrJHKcJEWWuve0tqgzbS7ZX73D+5CnrJ08ZnzwnDRsc7d3JR7weO0dc1mzOX7K9+pCP/7v//sSRE0dOHPkjrxNDTgw5MeRxMeRRH36uf/dz3vz275nuPgeM4uAmmAWCjIQw0CgUm3hoxblj9MEuc0MIWOuaxnFcEWIgBIFWEDFWY6BtVux2meOh8uRMWW9WbDYr1q5U70Fld9PMZmxcX8/g3bKwSaOYkF04zAWxjOYjl5eJIYLoGZ/8+M94+aN/wdn3P0XHAYojrfZYn+a9Zdz66XhNpLUj4hWphVZmdrs3tOOe3TxRTKhFmLyxL5WjC7ssTKbsqrCbGq1By8abQ2VfhKYD62FkMwRGIjILWRoBJ6kSdKDkia/eHJln56N3R87WkX02RCE5zKlSzWkiHI6ZmEBNWXkGCm4NvMMz1z4Eea9pzbmSS0VVmQ9Hxu2ABAHrSdnuzje/+zmQidpYv/NjTMfFfcQwA3PBURDFpCL6LdcTeGgF3wPl/q/MBa+O+n1S8du2tS8OLrroYM17FcXdKe7EEJdHKmiDJQBNQ+qaWzNynjur9G0SsrtDe/v6cinUZkzTxF/91V/zm9/8ht3uwN3dLbVWwIlpRI5HaqvUksEa4kBreK1MxwPUiLSEesOqQ6uoCOOYcFPcKoKRgqLSSAoxGClKlynMjVIKZXk95kZtxpPLp2w2G8btMz78+PsM6w0aV1RJ9DyGtsghHu96/Bw558Of/mve+dGPOfv+j08cOXHkxJE/8jox5MSQE0MeF0Me9eHnq9/8HDl8wcjETozWpF8kDQxpTUoDhzZRWiF4pNaCLFrJ7tXumFfMu8f6ty0GgwRSjAT1Hrgkwn7O7I97rm++Ybe7Zhg2bNKGD55fcLs/sEnKWRLc+qDZYSocqpElEZagKWmVSGA9jqw2z/jov/iv2Lz/MfMwQmvosjmrQa2ZYD0wSzVgh4oPFbFC3e2BSghGa4VDLjAMrMYV+11l8sghN24nONTAvjqHuWGt4q0ym2Jh6OFQUdmuB1bjyOZQuNkXRJx1iqzjOdM88s2bO77czfDmjhe2RYuxWQ0gwjEX5pwpLVAMaNBcyDnTbEZEyLkxauSQK3avWwVqbZTciBGm0mgyUVFiDMy5EKOyDsbx+re8+mzk46cfI+OINQc3HMMEQFEElg60CiAg94xxX/5u+cI9eKRbdYrTNzEgy/DpvQYW6ZJtpw9S5tI4Hg8cjkdqqWjoGtqUEqvV+kFne339BjPjbLtiu90SYySlRBNBQ09s/uyzz7i+fsP19TU/+/nP2B8OuAnurf9CFCGXGWuL974bob8khhgYoiLeP3Q3JU8TcytglWFICG3RTzdqmXuKeBCUxnw89OFHlNocWzTIIgEJgdVq4Onzd3n67Irx7CXbyysMWT7PBZ5mKI83mR3+f8KRf/UXJ46cOHLiyP9L68SQE0NODHlcDHnUh58wf81KMlWglUbOpQ8dBsUUHMFMwSOiAdGAlcWH3d6eqlWElAZSENwarfXTfwyRTYqkDezPBw555nZ/TWk7xqCcrc+4fHLFO2dPeOfJyO6wYndYk+eCVeduf+BuPtCkca0T013BxzVXH/4rXn7wEe/96M8Yvvcj5mFFLZlQ564NBewInnP3Qc8HzIzb7KxTotAo057D61eU21fcfvMV+yO4C+sx4AcDHzAqpVVydVpTQKgNpqlRPUEcQSNBDKszKcy8s3Iuhq4JjaoIAfFITCNK4liE613mIiWOR0e9QDOsCaUpc2lUA7zybIiUXLFVAAK5wcECzZdBPBzX2F9TbhzmRraJEBPrYWBMA4KQLGNT4cvP/hNPP/yXnL8zomFEguJhyQp70NYq5ovNpltPJsa70vZblpMmXWLQqyKZw+HAYX9gt7tjv9/RWndFERHmPDFNE9M8UUqlNafkgn07Wdl6y3i9XnerytbDv1ozUlLGNKAiDOOAhkBMCYAvvviiX6O5PAwPmnmHpXT7SzPrDjEIVg2vDVNnfb5hPUQE2KxGooLVQm0F8YY3wZuQxpFhtUJlBCuUkmkWmI4T1QrNnFoNo1uvxtWWi6vnvPvBR7z30fdJw4jGARkGvPbPllYRMzRAkMdbsYUTR04cOXHkxJHvtk4MOTHkxJDHxZBHffiJbUZwApGa90SNNHVMnNwKQQPiilfpQ4cSaBSaGREQVVQg0cOmxiEh988dAt4aVktPrE1KSsI+lz6MthpYpYJ4YaUNw3kyKpuwxjYr5lzRemCTElMp6Dryphiyec6f/Df/Ey8//gHpxRV5sybnhpj3QoBKd8g4NKQJ4hXLM81m3AI+3ycHC8fDHcebN9zuj+wnI1JZjYH1Cg43Rs4Tbo56YAgBN6E4NPNeYVJlPQ4kLbQ2k+eZYYTztEIkUA1ujzM3+wnXwHsv32FQhzKTW4UotOYc5toHADu+abUg7oTtCgi0CiYJIZA9UFvrGVreIZhrxTSQzaH54nQTQZWWjVaXtnQ7cPvNF2wunqEbhRB6VoKwVByWZnFd/tSVBb2K0jvTmDdKrewOPSn67m7HN9+85ubmDbvdjsPhsPj5943ezKilX/M5z8QY+3NKrwj14cP+eNy5eXONBmWe5iVVuRE1LPpdR4Ms1Qxdnsv759b68KMtg5Ddr1/A++OtGUYjBiUEIelbaKcYiAGiSod5UoIoKQZUnVIyWEEwvGXmXKmm5FxpS9ZCbUZtThxWvPjgPT745Ac8f+9DxvOrnhXgjWaAOKKC1NLfi1XCI67YwokjJ46cOHLiyHdbJ4acGHJiyONiyKM+/ISg0CK5VQIwRqGJdJ1gL7L0G7YYRNAYAL3vMvbhK5ZkXXE2m5EQMrih6kDDWkFFGYbAahzZT42pGGNUXJRqhcN0wzis2KzOmKcJjx0e52ul5IBNhaRrzq6ec/nxT3j/T/8l4/lTpkhvPXtFg6Mo3gpewVtGW6Mej5gUss1QlRoUkYiUig4DcbVhnGaub3YUqwxjYIhCbgesde/2QZ0gzrxoLderFYEAGkjaLQ1LbRwW28WV91P+MRe+uTtyqM5q+4Sr8zMGDKuR290NEgMxRuZpAl887c0xjFwb1UaMSC6Fw9ztGG+PuQPPuid+t7iBXBu5dE/6rQbGUXGNSHQOdzNzbhznzFe/+w3nz15yOa4wHYGl9ar33ePeRpYljTjI/WApvLl7w+evvuD1mze8ubnhzc0N0zRxPBzJOWPWHgZQY0qEEJimidba0oIVVMPyZ8g5s7xpmvfQOvc+SGi1EgMQItYaw5BoZizFHlotTLWSlqqL6pL0vNyb/c0IeA8666FnDlEJoXv+mxvmvcYk0K0pW0WBFAMphR4EdtwzuaE4QsO8V8ZK7bBxhOba28vbJ3zv0x/z/OWHxM0ZVRO9+NOorSH03r1q16qrSG/7P+J14siJIyeOnDjyXdaJISeGnBjyuBjyqA8/GhKtCrk11uuECRRT3AMuA225iKs0IilS2rwMowVE2qLBNIaYAKdME0ZDozMMXTvZDNKgbNeRp0/WNFOubyZ2h0ZMGfdKKYmry2ck6+1HURgSPD1PtKy4BdInf8LVj/6Msw9/SLi4Ym4GUybUCXEHDdTsMM+oV9q0I2fneDhCdDyC15kyzbTqBBHicEkeJ8IGxvWOfMhMx8JqDFxdpn4bTgZuHPOeaHA2DpBGXPuQnOA0ixQZMYscTZnqQIoRGYTzizM2COMQ2aaZs0EJEjhfbXDApsw0FbpHvdGs4WKUnJnLhpvdhG4UDY3DYeLVjT3oXrn3qkcozdhNFUfYjBUNA44QYsXTgJsSPPD61a/QKPxAnPWzD1GJfTMqEBU1w80Quo71iy+/4re/+5zX12+4vrnldrdjmmdyqb1iZH0j3jugxJhorZJzIYTWA82WX0zu3oPUWmMcR0IQaqkE9YcKTEqJVjKalM1mjYrSFuvKVt9qZ4OG7l5CH16sJVPKDEDSRFB9CCRrrWHLL4E5Z5rCaoiMKSFL+9wX6AUVxhiJKt1m0iohCCJxSabujjDVe9J4M6dYwIeBy6sXfPTJ97l8+T41KFZLr4qJgFWsFoSe3pxUlj8r92nUj3WdOHLiyIkjJ458l3ViyIkhJ4Y8LoY86sMPorgENESG1I9/UgRv4ASQgMZIEKUt7TJH+0lfe7VGJBJioFWn5okarYcu3Q+jEUAjSRtn60azkVLgzW7i7jBRi+NtJMUDtTkiiYgypgAtcGyNtN3w9E//nKc/+Qs4e4E3QcqRcDjQ8q4PF8qI+kCbJ/CC1j1TrsxlRurAahyYccgT5qDDCo1PiOsZamWzPcPnA8epEs+Fy61yOAZ2x+7SYXViFQfG7QAhISF1FxM35qZMMeBNMAKEiMRetdluIAgomZED50MiqjHgzLlyUxu5GiJGkP58hlFrH1ZrXnEJTAW+uZu4O0bc+wYW6PpU165ddaitUWthDAlEUDVMBWJinQbudm/48h9/wdnFBR9uny+hZvQXuZQydrsdb67f8ObNLb/+zW/47e9ecXt3x5wrxlJlC4EQIj1wrD1smp54HCilgBuSEraEiEGvoqWkQB8GBVssQOktYgy0DyjifYg1ilJbI4j0itoy7KjaswKa2QKH7utvCCn2qqDg3Z/f7xOk7x1qOqx7galXXlJQxhgZtFfrmhtuS5VEFZY2eS5dazs3oRIhrnj+wff48JPv886772Mh0dwIVsEqSG9hu7UuxZD+WYsKjjxULx/tOnHkxJETR04c+S7rxJATQ04MeVQMedSHn1YrMQZiE6J14DjK3Lw7pmhCUqTSB6NUlnYe9ze8EFVx70FTKYU+EDakpXoQMFFKM4ZB2a4G3CqHTWU3CaU2giqlKcep0BqkoXK2GRliIqsye0C2T7n6+Ef46oJGxPOekCeYD3jLkBtBIMaA4dSW8VawVnBtRIfUoMXQQ7UwrFbS5oxBK0ObaOs1edww10qtM0JjjGGpanTd8GYVOd+mrk9dqghm1v33s2EeadbTfUNwkhhRnRiVJM5ZDJwNTquVTOkVgsUuEnHMbUkF7pshxsB6HYkJqjlTcXLrw334MpznvfrSh0C162TdURWqg907f5gxjGtW48AxT3z+61/w3ic/RaTDUzShouR54u9+9nM+++w3TMeJKRemPPc2r95f7y7AlXtf+2Wo9P7zgK5zra32Te09miyG5f7Rrq3tacs8AKfWCtZTnwH2hx39V5+CQxoSIWgHjDsxKHMxas6I90TvZv3n5TI/gDHF2CskrfUKVAgE0Q7s1pitklaJYTV2v37rnvohBrK33kZv3S6ymVFrodXC3AJNAucXZ/zwJz/l6sU7DOMaa/2+dm9EEcwqze8ToPtHYCyuNHQ3nce8Thw5ceTEkRNHvss6MeTEkBNDHhdDHvXhJwRI0rAAqhuaNKQ4zZwxRsSFYXvBfHeH1Wm50RbULKfh5eBLGhNnmzVnW8ALZhWiQkxdT6qB1WCoTdyNM8Mg7CfnOFfu5ABmDCFwtk1shycIA1WfwpMzLt/7E8anHzLLCjsckP013jLipduyi6JiNI5INJrBPFfMDK2VWnYcCLAdkDAQW8OaU1ZK9TVpe0kdN4TVBimFkmfcnU0KjGPEVBiGxPNt4HLrjCtlvd0Q00htzrH0f0q1vkFLRsWRpQoQY1pO/EKtmWnKTFPjeCjkAiaOSuuwsK5phUAalJScEI3DwWmemGrXZQp9UzuGSA/30qSE2GGgdE1rPoKVPujnw4ogykpgvv6SL//xH3j6/g9JZ1f89te/5ZuvvmJ3uOWz33zG69fX9A5+oC2DfM0q4v3aC4B5z1igV1mGoYd33Q8HuhkhBtwVa4bQ06Y1xMX3vkOTBSDdi79XRlR6G7qvxjAOBJWuDXfD3AnAdr1Cl/ZvwxkQSquUWnFzAqHbRIowpIFAIgUlKSQxvFQyjQMNrDAEYR2FJB1c1nqFsZnRFmBJWBHihsv1E55cveDdj37ARx//gLTaIqLk+bjYsPY2vFnFXVERYicarbYH4CKPGiMnjpw4cuLIiSPfaZ0YcmLIiSGPiyGPlzaAihPESCLEtGa2mbnMqDTEGkqkohyrMyK4dM2gIyD3buzdtWIYEqshMQ4AuQNHDImKhAFcieKcDXBx1rienKkcyHlmtztAnTjfrIhPLhbYjVR5yurqEy6/9+e0sIaS8f0dMh+7vlKNgOAqmBpFM5iCCh4GBqnYNDHXws6ccdiwiuddQ0om14xJQIcNGlbENEIcurY3Q6SxGhKmSkqRZ+fC1Vo420Qun20ZV2c0E4opU4XD4cDtYce+QbWCZ0MmRTWSzfjmUCi162r3t41pcrIJzXslQ6TbUlo1YgqMq4TqREzCnDPVur1la0aK2qsXS59StENAQwDvLc0YI1MNSCu0WnuFownBQazym1/8PZMmfHXL3/zN3/LNl92qcS6FXAq2uNY4S0u7NlR7ToKKIhjNu9Noa/cbqLdx73++BsVrr/jUWpZW9Ftt6b2//L1tZVh0us29JywDurwfa30QEe8Vq6A9zXg1jrg7c86YGcPZht3+0O0mceY8k0IkpEAS7f744rSy2Egq5GKIFXSM9NqQoUHQEAgu1FIopQfApWHN5bOX/OBHf8rVi3e5fP4uq7NLUEVQMMeXgUsVB4v0TGkWm0untplqvX1O0D/qvv+nXieOnDhy4siJI99lnRhyYsiJIY+LIY/68GOuuAYk9BO6VcOb0zDmWiAG5ruM50ocB2qdUSpgiEaCNJRKlNZPvG64QQoB0da1j3mmqmPjBgsJx1kPIxfrSs6VO6u0bFSrpCRstpek9SU1XTClF1y8+zHx8ikH+sk3WG9rSwxghW5OIbTWW5ZzPhA0MW7WSK2UUgjujBixZVQKAE6jzhkkoprw9SUhF1bu3O2uqSWCVAJCwtgk4eXzSy5HYYiRs82aOAw0b6xcGbMRG8QwEGfjZjdTMEjdq34q89KydebqHHJjbj3JGo8EX7zfcZzKmBJKZRVgdKOYsG+JZl3rGkQYl8C25n1DW3NKgTZESqtIPZJpNDPUDa8TMUDOBgjHmy959Yu/5TYnvnh1w1QbDaNVo5kgKOINceuBXKpdb9qb9UvbtSGyDKjmPuTXvgWWWitWlyE/hKiRUuriptIrLL21fO+D36h12azaW7MCNDpbW+ktapOusa3zTIy9mhJjIpdCDGmBUZ+dLPNMlQ6M9TgQcLI3tM0Eelt+jAPjELvzS62YGkMccBGq93TvYspcDR0HLl58yEc/+DFxWBPTCqGnXiNtebUG3nABkZ48rYA+aKQdwRADecSDynDiyIkjJ46cOPLd1okhJ4acGPK4GPKoDz/NlDCOII08Ham5gCsSlGKVNs9YqYxLm44GKQpu3QkjuGF5j9VGYWZ/OLDSxLACJVBLo/oMCXy1oQah5EKrMyutXK4FNeUoyjgoY0okXZHrBrYvGd/9CeGDT5jGFW2ekNZIQ0JaIJpj3tgdd3RDdKfWmalMjOOK9eoJtTpoZDX0qsQ8T0y1EcKw+JsbRqVYoA0XxIvAJgy0333Ovky49MeODoMY59s1l082iPVKQG2Z5hVFCdZYjwIpMeUJrY18nJhbb4vOuVJL7TaPk3E0p7rR3ImSiIDiBIUwCBdPBlYJtmNksJlSlb0HajugqgwxMAZIqWuHj3MhSHdSyVWJpSKtcqxGs+4o0/JESG9v2ViPTF/9I7WtkBrIBao3ggRUAgrgRghCDAl3oVbvg4Z9BpVhpXgTWmuUknt7VbuF472tZG0N6FWF0rpOlUVjGhcwuRsxdrcXW1rPdanCOD3kS+jWlCGE7nqiitf2EGBmKOZQ5xk1ZxUj0KUThnVLSowo3Q41ASEqKejSAl4ea0ZpjeIZRzjmRqkgac2zq0tevPs+733vU8LqDJeAScAWeLpXWivAEsxmRn/lHbiy2FmqOMsfl+HIx7tOHDlx5MSRE0e+yzox5MSQE0MeF0Me9eGn30zdDaO0Sm0NjSNBA80b3hSrldV6zZACdQ7EMOBREHV600z7MFtpZDtyto5MTZBmX+/NEAAAIABJREFUII0UIIpi9CG65g3zwpCMjQhIYJ1GhpQY0gavkH1FWr/k4oNPaesn5CbEfIvY3F9bydi049h2vP7ic2qpjKuILMFQ6obURM2VOR85WxuCcUyJeTrSpgMaBtJwTnPtFoQV1uMFXh1JW5rsqG0JqBJ6m7H2ikMvODRarfTOolFbxRBKLeQ80cpMKzMl9/bkcW6UAsepcszddtNRglSG2KsmgpOqs9qMPH1yzvl6ZEwVz8pUe5hVqY11CKxSJIaAhEipjf1hYnO2JYpQWqb6CK2HoGnUXhGx3CsdukgFTFAzxiQkbWirqENMPWr5Po04xEBMgVqti2odggbGcei1odwWG8eui70fOLwfOpTFbefB494dkCW4S5nnmRDC0jX/w4E7s66j7UBob/Xe6e3XVqsVZr0dbTSSwjDE7vdPv8/c+3V0FWJKRAUNQoi9qmPu6H0Vp/RqICFSTWkuXD19xo9+8lOev/Mum4tnON0bH2FpsfsDeHqDvu8RgIdmsiz/WqQB5k7TxytXAU4cOXHkxJETR77bOjHkxJATQx4VQx714WcYEqUUpulIKZkQFgcSub/sXTcZYujug0vzTkQpNaMBNA6I9hvIgOrC3BTxSoqGObj0jSIiaEik1NOCUe+VGwIhJCRumXXDePaU+PI9fNjQGohXAgWf97RSuLn+HfPumrvbmdfffE2tlc12w9Nnl4h0+8vdYUetlcNxT6nG2fma1fY5yA27ekOeZobhYqkMLO85DsRhS1qdE8YdZZ4pzfHmeKsIQs4VsYZLRmNYwrNm9nd7pmYUEXKemaZjdx5ZhKrN4JgbuUpP10VAjCjeZzFjWEDpXGxHtuuEWCEINAnk1sFl3m09YxBi6M81l8ZxKpxfBDCj1EptA5oiMSheM+OYcA946wOA+JIMjUGbWYXEqAZNoLUFiI6o0Vy6DMEAUbwZEHGD4zTRqj0MDd7D5l532ysjibq0/e8dV0IIfVAVHr73Xnt7D6tcKyLdhz6G1OGyuLo0FeZ57qF2qv06yjJ4aZnVOII3ajGQ7rwiqoQhoiIMKZC0B4fd64O7ftew1gdVfan8GEpab3j3gw/55AefonGAEPGlYQz9/tGgQKDpEr6n33KVuf85LCgSQJeBSflDyD6mdeLIiSMnjpw48l3WiSEnhpwY8rgY8qgPPyEEWqto6CFMaRg55oZhlFoQcdYpLYNmRowDcym4F3wZzoJAbb3qICoc5kKtyirCelS6r0a/MadZWQ9p8d5XQpciMlVQGZB0jj3/IfH9P0Ev3yc3xdsBzTtef/05r7/6mv2ba7786rccpiO7g5PngkpgHGY21weiCuNq7K1SVaw1zla9ArS9UFI64+w8kOfcXTMwTCAMIwWhoKTNlnjY0sqBWjJBlBAGkEgzJx9n5jwTh0gcB25u7ri+vuHYHNKAWx++K/d62upMBQ5TIxfr1oIiDEFZxcBqCIgGvMKTs8jLpxu2KyWRSZKYDOamHKYZkD4A6sIQR+ZiHOdG897a9VrxZhxzJaAofdAvBCHogFuktUwtizsNELyySYk59UTtueSuwZaep1AXz/oUh6VV2lfXw/pDFUWXqkFrvQJiZoQYYfm606sh0Csw89xwur99LqXDCmABTrfJ7HaWZj10DfrwYojdxnNMidpqh0+IXd9r2t1YNJJCr6Z17/0O66ABla417tWzbndZzVgKUZgrhjKuzjl7+oznL9/lw+99Stqc91+i/nbg1pCenCz3r71/DqIs+QPdYr/Wyn2MgMkSaIZiUnnM68SRE0dOHDlx5LusE0NODDkx5HEx5FEffkqt6GLdl0sPorp3wsjtgEjkybNnvbUqTogDbgfcQLS3/QIBt4pqd2w5TjNHAnUUtuuBat4rBbXh5ihrVoOQkpMQahVqDcS0Jpy/R/rwvyS9/CHZB0IttMNrjq9f8Vf//v/gZ//4Od+8ueGwP4BH9ioogSGtSCFzPPwO8UoMwvl6YL0aWQ8jL6+es7lzLg+veP+dp6zHDcO44XjoQ3Ih9tTcWitVI+vLK8Z5j+/3zDUzqnQgoNQmHKbMNB2p+4aHwPXNjuubO47NcU2MaejWjmbc7Gf2c2WqkGu/MaNCirAeIudrZUwDpUGeGy+3I+9crlBx1uqkoByLcTQ4zjNoQrzbPMagHHJhqo6kgcPh0J08zLDcEAqboDy7uuCw24H39GIN+tCGFiCosaKxic6NGMa9T/9b734VaLVXaVQjOWfK0lrGeyVDVDHvvwQAUkrkXPqg51J1aK0Rw9KGXl6PqFJLxZYqjH+r6nLfvm2tYUv1TxbHG3WWtnOvjKDWdbuSMIcxJlKMHPa75fVX3BvqvfxVrDvBuDpKB2b//oBrYBhWvP/xD/j+j37CxdVzhu052Zbqivf7OUbAjNAMx5bP+B7KPV8CADWqdc01gGhY2s2Ck/8Iu/2fb504cuLIiSMnjnyXdWLIiSEnhjwuhjzqw09rrbc7YySlxJT7jWKtYhoQuQ8UWy4+gmokxgje24lDSqgHVB1VqK0P9hmQq3PMhbYMkIkKU517aNPgBFViEob1mrB6il5+xHpzRc5GGgop3zFdf82rf/ySf/s3v+SL3YEWIrQRNSfHiDcjt0yMgbmAW2/m7eaJoDNBlc9vjjy7uuKd6Y6UGi+fXbEeB9abkVAMDSP1cMSCITHwRN/lkAvj7Z7d7V0fBgSmXAliHHPBEKoLZW7cHQs3x0wlYDSmPBNCZM7GVISpCHN1UEVppChcng2sx8gmtu4iooJX52wdeLKOlJoZVNAYcI1MJff2sIEmfWjpluaYKBoCtda+8ZfWaQiRpMr52QZa4Xgs1FYolimldB2pCq4g0gjLpkMFd1ksK+NDG7e3Yru1o7nTvFc/xAXV3w8Vu18PbVx6KznGQAz9td87sPSN2YO4ulSgfkuf27WsGgNWa6/uWCPnxWnHfQljczQEonYLTrxfr9o6MGOMpBCopbAQiiiLbab3IVVbqkSqzrjacPnsBR9+/D1evvsBhEjzruPurfp+3VrrqdCFHljX33ND3UCEKvfuKV0mYAskg0LPKBDcH7FWnxNHThw5ceTEke+2Tgw5MeTEkMfFkEd9+HEzhnFYLvbyd/dDUr6EHnmfKnPrbhEhKKKRVkBFlw9PwXJ31vNIrY1ZAnezQSiMrbAaRlSUoxlsEqtxwFyQMLK9eIGcfUB++jGSEjE11I589au/5YtX3/C3v/iSz64PmCq4YLY4ZHii0T3gc62YaB8yVCF7H4hThNvrW745zrx+7by52fHROy9558UlLy7PiGGF4jzZrKjuHDLI8B6XtuZwmPnqiy+oNoMou+OMEJmt+7jnUply481+Zl/6/F0feqvUNlGaM1ehmCyaSwgCT7aR5xcjQ+jDmP31Z5IUrs4vWEVDrZFiZH/M7ObG8VhwB2tGHANDSiDatbfmjGNC2xJORq8axBBYxUirBQ3KfdLynCulNIr1ZOQQAgHvg3lJ0WUOUBb9quBdE3CvzzVoblTvD4wSaK3bNN6v+0rJvfsJ3OtQ+9feam79AUoaBpr1z4OlHd+/gZ6IvGzYUgpu3cUlLUOg2L2mN5Brpeaui9YASZ3oQhRdgsn6axWRnhaNLe3s/t7ONltevv8hH338PZ698z6axj5I612S0SW0S4p060OM/fPiAby2vI9aWpc80LpGeQERElBdHPcfcTI7nDhy4siJIyeOfLd1YsiJISeGPC6GPOrDzxATIUSmeX7QMoYQCB5odBeNYRgI0KfkMGxp5apFDMPc2IyJljN5nrEQqECzhJRA2xc2Yx8mDBL7JmmJ0roeVeIl28sfEq8+4e7yHUp0trGy/+IVf/Xv/pqfvfqGn325Y5aI52O/SGmNxDUahVYLrt3XfBzXkPpJ2urEaj3QmnE3Tdy9vuVGnF/86msuz37B+882/Ld//inPLy9IccPF0/dBIu6NqhcMT8559u6O4R/+nuPNkWkufHV9y91OcJupbSaXyt0+8+aQORSn+8IXSmlUM5orLgFEiBE2Y2A7Djx7sub5eUJxWonsp4JaZozGk23C8pFBoLXAvgjX+8ZcliE/b4zjhmEcqa7MtYEIQZY2rChz7pagYNSW2d31je8I1aCSOJbctc5qOD1IaztGnsTIsfbHuii1GaVkcEPpv6Tce8qy0S0TH9KBlz/fV0pUBWf5f+1Vky5l8IfBwr75wHEUJw0DNZcFMP23oCG0ag8e/LUuvvvW7TkBJPSwsmmayLWBQ0oRVJAoXU8ui7a8ZmIc6KSiA9T6MGJajXzyw3/Bn/3Fv+by6gpDyYulpsO3Bhr7Z11r7VkErSEO9N3Sr39ri/WpIqG33kXDw3N0oCrpfhDxka4TR04cOXHkxJHvsk4MOTHkxJDHxZBHffjRxaWi1rq0CXUZ6gKWVmAIioogVhFp4BVVpd1XZYA0DETpqb0VJaQBxicUjOAw4N29JKTuYOFOqQVrlRCMIv1nDApBDA533F1f8/NXd/zqqwNT7vdlw0lDYr19iumKpDvyIffXLDCmEasVt8zzFxeE4Hz99Wu0FEIzska8Kbf7wnT8knefbShzYTWsWK+fIMOKVmYYNlSDMK57krE5c20UA23WnVMU0F7tKLVvIjPHWqNZwwxEvINAnVVULjcD60EZQq9euLeuiXVHxPrXhkTzCm7MDaamHGp3CGm9CNFb/RqWITdYr1ZLBSMzpDWEbj3ptVA1kot260V3SjM0jqCZ/4u9dwu1LUvzvH7fuM0519q3c06ciDgRmVmZlZVN2YUgtNqUgg+K0C+CviqoL4I+dLcPKgiiiILggwgKQguKD0K/KO2LrdCiD4o0qCg2WlhamZVZFRkR55x9W5c557h9Poy51j6RXYUVFVlFHnMO2Gdz9rrOucb4zTW+y/8vVIxrNas1FiqZi9Cxd3CYYmvAPEUCpEV6mu59S41bY0BbRAs4z4dTpEUVrG11qefAici5IbHFLVq69vQi3vs270ozpsu5GYmV0+sKGNvKFmot4JpiysnITFUZOkeplc1mi3OONB1JNVFz4uZygwSP846cCikXqkIpiguej159wvd+7dd49uIDxDpKUZB2sRWap4Jdzm87aEALVU9p5tre4/JedGmYNGb53E4nczl21bq0Wb6/Y+XIypGVIytHvslYGbIyZGXI+8WQ93rzA+0kiAjTPC6gaWlGYwUjQo4RFwKUAjVjpGJFSLUpVJxOmO863JwZq0Gtw/fPyCXycDxSa8bJTG8z1lhiXhrmasX7RFJwxjW1j+Oe4+Nbvnzzmp/cj9wfFBELDly3YYwz1o+IM4hVgm0TtCo41zUX3DlxcXWBd8Ljbg82tbpK45gkMtXKYUr8Xz9+i2rgsj/w4cuP8VrpnSPOB9Jxak7NxlN0SdGmDLXJIwbfYhRWBGcs1ja4FNriMALOO7rg6Lxh23uuNx5vBaEwTxNaC6ZUYqJFO/p+UX1pe/aohl2E/XxqXLN40+pJYz7VhLa6Z9G2u09WQSyyRC/EeabUFHNyyaQCIXRgZ2qdABAFJ5BrZpDCpRPmMpFrBhuoS6mBCOd0qS4Nim2+PC2YkzSk6tMyeldnvzknN81+rbo8z6LFz1Mq2jlHrmWRlATvA7qknL1rUqZG2nOxRD6MaZ+Jd4ZYlFpyU8usDW2lNuOw4D3WeUqNGGOpVfFdz4cfv+LXf+Pv5NNvfxcb+tYUyNIcWAvGSNPjlxMsOL+uORGVU6TpdCz2/H/n/PI4pZTEyUCtUnnfx8qRlSMrR1aOfJOxMmRlyMqQ94ch7/3mp+97UkoE33GYRkptrsk1F8RVRBVnDWlKlDrhlkhBVdqHRatxdcFjbGAcM3ONXN9ccPXimsPbyvHwJeGYKabSdRbvIAj0XY+xG3BXuOE5xXnGNz/m85/+lP/nd3+PL2/vidVSpVLmPVXheJw4vP2c55dXlP4Fm8uhrRitGAtiB8Dyu7//lqvrS8xwjWwMkgt+mpmkMJZMEcvf+tEXPB5GPnmx5fvf+xUusS2qc5go+z2TDIjriFmhRKxmnCl4n9siqeBEGLoOE1rNZc6JECzeOTabjm3nCaYSrLINBmsMRS1jTC0lWpVYYBta3e08z9SSGbzlGJXXD4m3jxMVwTqPIZJyhtL0+Luuw1lDjXNLZVrHODf5T2eUagw5wf4YmeLEPhae+wtSNpia0ZIhtYtO8DBIwm4DRntup8rtNBNrg4cVg8Gg2lLGaFOXr8K5JvbUaAjtYzktqlbX2uIyp+hYUxd5itQ0Z+a01CqXRcN/SWUvUb+Wpm2NsCLSpEGXmteUEtZ4pBxbvW5KGOuW5yoYEfbHI8IiUWktRZvJ23e+/S3+3t/8+/ju93+A8T1Tbv4CuZaWhlbFmpPHRG0XhqXB0DqL901GsrwTIZIlCnQCqHPufJy6WCqLCC74P+VV//MfK0dWjqwcWTnyTcbKkJUhK0PeH4a815sfEXM+ccaANUJMGQiUVDCblr5LKS2TJ4O0lJ9W0Lb9ptaC4kCg0Iyu9scDV88+QPzAXFoEoAHGkXMlGeiCRY3H9jd0Fy8ovaFuekw/kHJbwEWEXCPj/ReknDjMmY3tGMqMPNsgnYCtCKZpoyNMMTGOM/vDiAiMc5t4zrRax6kkUimQM1/e3SLlyG63Y7BdS60r1Fw5zDvEWMQ5cpwR49vO2tglzVqxVjC5YrXV0gZrwQScswzB0VvFUnBo0/t3llqUeY7klCk54YzDGSV4g1tS2MZaxqjsx9RqbAUExVnHFCOuC2w3W2I5AjDNEecGECEET86F4A1pjhyOiTFGcq3kUtk97sglEaRJfJaS6XqPs0JwrWb12Dse52mJsrTFhLYE6ck5WJZIm76TWj1FVkotlFzOCw5aWrU5F7fa4zZOKVrFOUuMcVEwYdG9t7RTXc+PaSl9xYdWF92aEls6WLVFcrwPIHZJ4wuINFfo5QJaqlK1tPpYlI8+/IhXr75F32+YilIRxFqMKWdHZAGMtYBBS6urVgWjshz3SV1G3jlmPcPw3Xrk4P1iaCfvdbkKrBxZObJyZOXINxsrQ1aGrAx5vxjytTc/IvIPAP8i8OeAV8A/pqp/7Z3bBfjXgX8GuAH+B+CfU9Xffuc+z4F/D/hHljP9nwF/WVX3X+e9ZDVYdSgVoxErBecMLaNqMOJbujkntvaKqewQ0xQzrAktNWgMlSOxzEQqUy5Ua3Cm4ruecPExj3dvOM4PXHWCUqilECoUFcLWU15+xHGzZaj3BNkQzMBhPDK7Qi6CyVBny+H4wJgiEzNGOobNW9xUsS7gjKfEiTlnxunI3f1r5nlGKZSUcM4zhB5xHmcMYiBS2U+JTXDkOFHmO9Q8krnmMUWO+zfNdM0OZJ3OjWYW13bvgAsOn0eChU3nMNaDaS7CVhK9E6gFFzxjVYiFsiiIWOfBWC70wAcXA88ve3qZMM5xVOF29tzuRuboEEkYm/AyEGm54e3gkUNFtYALqEKOEyZ4MB0Fh0uFOUeOJdMMsJRpPqA0S6w+WHoHnVeMVwab6UxExRCzYVc60mzQXBFtqdeCIKKItrmg2HNjISypZ1WMLA15S03qySgsxSbNeFJXcb7VdlctaG0RqBZVaRKmYMip6c+fIhZNSaVgF5hoyQRnqGSy65jnGchNq78UWoIfrGnnPNeCtwZnDUOwvHj2jOA64lxJYpuajxSsMfTWoLWth2pDg1WdqDW3RkpVSmYpeVjqNTBnwFYtGFFizOfzEaxr0StjqPL1GpV/kRgCK0dWjqwcWTmyfhdZGbIy5JeJIX+czM8W+N+A/wj4z/+A2/8l4C8B/xTwQ+DfAP5rEfmzqjot9/lPabD6hwEP/MfAXwH+8a/zRsZxxNmmnZ6zLrtoOe98jTTPW9WK9ZVOO9RYasr0XUsUYh21GMgCVSgpo2ScWC6uP+DDb/0K1In5s/+dKSqdWLrOg3eMcsX24ld58dF3SGKIh+boezjs2T3eISUjte20vfc41+FRnASm4x4/b7FXBicGrZVpHJmmIw+Pt9zdveVw2DNOR5xzXF5uke0zfK/UnJpR1aIaUkohptYYWGtmyiP7/QMP9ztqbY1vSZXH3Z7eC9Ubrl9sGHrHfpy4vOgZup6rodWCPh4T0RhSzjwcR/oQqBkO04gxDkGYp6m5UVP47sc3fPryisteGI+RMcFjgi9uDzzsD0wJrDc4HKkknm2uuLrcgmmp11KboVYpzSxOtMLQU0oh00BTUnOQ9q45aVeUUhNu67m+HLjoLaUmBEMRy7XvKM5yl0cOaaZoM+FysKiqCKhrxabvBAtOqeZadUnn+iUFW3Cu1bSeam5P4JAlVd0iFbZpSC7DuRadOwFN5akpsdZWoZpokSi7uD6nGMk5n2tdvXXNI8JYRGkNf8Yji+fDR598ynB9w248InPEhg7xLf1eUCIGFcVpwccRaqXEVlZgTANNyfGpnnhJxZ9g2TwK6jkiVWtFczmnomP62uaEvzAMgZUjK0dWjqwcWb+LrAxZGfLLxJCvvflR1b8O/HV4qk08jSXS8s8D/6aq/hfL3/5J4AvgHwX+qoj8HcBfAP4eVf2flvv8ReC/FJF/QVU/+6O+l9OHnOOMalNYKbQT4p1grVBKavWNtkAyOOPJopQyUbXirWCNa812S9PfSFOiwDpsv2V78xHTF7/DYbxHpWKcI9Fhug+Q7cdgewyZRJO0HDYDN9dXkH8PYyyxFOZ0qu01PL9+xjxGamrGZyU13fngDLU246jjYcd43FNrQayhD6EpxSw1j1qbIzKlEHPFuPa6h/2eEDybzUBO10zzjGCw1qMlgrb2vaq1SUZuAtOUEM3UQqs7nicej5GqQh86umG7pCWbD4DWJp1oxeCC8OxqSzBKzTPGOOYCD0fl7hA55kJRxUl3rlENzjflkJSaxjtCLPN5OtpFscY5x5gyMUYMBmMNIq1xzyh4J/Sdox88g7d42zHnQkYQhaEqgxM6a4gKLIteBHLlXCN71plfIHKaW7VWamkLzRq7PPYJJqW0VK/D4P1TA57WegaXtXZpTrQ419yeM3mJ3DhqTq2GF5jnmbRAWBbfh1IKskAuOIddZC5PSigpJ+4eHtkdjlzFiEqiEyVIAGmyljG3i1OV1vzZUs6CPaeYn9bxuc54uVBWLRht5+fdZkxdIi2llGYa9zXGLxJDTu9h5cjKkZUjK0fW7yIrQ1aG/HIw5Odtqfw94GPgb5zfmOoD8DeB31z+9JvA/Qk2y/gbtJTzn/+DnlREOhG5Ov0AlwDbYcMQQjOpMq1BzVrfoi2mTVqtleAtYpYTqZamLFIxJKgZTilDY/HeUEsmx0SqBXWe7c2H2P6K/ZyJtbba2uE5Fy+/T/fsW2SFXBJVLS50XF1d8eknr6glEqexRUdqpVZBqzAeR+bxSIqZkhJxnqglojWDNlnCcTyQS0QMDENH33dnbXZRpdbWpNZ8noTd/sjb+0ced3tUFW8MzgaQxUxKKykXpjmTT3MqJ6RmrFG0ZEqesUZwvkPFUdShtgfTUXFY12o/S64YMXhvuRgCgzdQJubjnuNU2E3wxd3M41iaNLsz5wY/4wyddwTvzhcJFDoXYFm4m74njhMpJeZSSKm5H3vjmv57LUhJbIfA0HuGPtB7y9Y7LoaOyy6wsYbeKI6EISNGUQMqSiu4rUut6lPK+A9SWimLRr4xFq1CbW9liURwNgM7LciUEqWWd6Iviyb9omPfUsicb3PBE0LABX9uTsy5QTWnfFaGEdEGjVooJTd1H2vpnGUejzw83DGnCWSp6S3aamlLIddMLrnJfsriBr44kVtr2wVZzBk87b0vSix6Uqd5R3lm+XkyIPu5qjT9iTBkOa6VIytHVo6sHFm/i6wMWRnyS86Qn7fgwcfL7y9+5u9fvHPbx8CX796oqllEbt+5z8+Ofxn41372j7lG5mjPu/9cmyKH946LbYe3sB0CwTtSOlJKQotQS8T71qRYFeLU6i0vrwb8bqQcJvaHHTlPVKn01y9g+wE//fFvkfzI5sUruue/Rv/xn8Fef0DUgjWC6XrKeMD1PZ9+8gnPLi54OCT8ZsP1zXPuHt4wjUemhwc6H7jY9MRppJSM94E47ZnnmcfHO3KaoCrOeT7+8ONmZCUQYySlRE4REcVYR9dvUfFULBfXN9Suox+U/VhgSbWDgrFErUxJKRhKLaQccTbghw2qmTFWsloyjqlU0j5yd39ANKM1NSdgK5jaGhRvLjzbHiQp0Vi+3Ak/+jLxk/vKfl5qWpc0pUjAuow3BgtkUYw1WBWMKjOKVMvQ9czTkbvHOybj8cFjVDDOMB5HtGSuho7vvHzOzeWGi21HXxKSI73vwQqbUnC+cBFANJOzUFQwyjsRk9Zk2qIa+TQXgaemw1Nd7DzP53Rxra3O1nkDPCmrtJx1a7nTBTqyREaccygG0bLMXaXznlrzAi+l6wLee0rKGDGEIWBMa/BDK1ozxglyiggpBO9xRnn70x9zdXXBt7/3fbphi7UdVPB1pHfNvM05h/NdA8siffkUYWrSlue64uX+MaV2XniKyKgqSZ+A+3P+0vInxRBYObJyZOXIypH1u8jKkJUhv/QMeV/U3v4t4N955/+XwO/V2iIk4zwh1hBcIBfBScYg1JzowhZrDdNcKDW15i8UasE4g3JSnSg4bxfzpFbDaVCcGAgd3eUz/MUzqoNEQP0FYbhArBDThNUKqXDcH3Aom82Gy+2G47THh8CwveYwHZjjzPbqko3vCF0g5/aBWmtIKVJ1mfgiSzMZ3D/u8b7D+/Yh19KiMiiItcRU2F5eY0NgihNqUjNJy5FSM2IAGpAFQ6s4bcofIoaUKylHjBGmrNztE8cpU9VStVJiAk0tYkEzIJOa6ULHNjiMUcQbqgbuDpHXjzO7qT3WGGnn1yzmaV1AVCm5yRg6wIrDOY/miLXmHKlwzmHU0wUBqc0jIBdsVS76jqtQoIMiAAAgAElEQVQ+cNEFeucIVnEerPNY5+hrBVu5HCacGZuqCQ0QRgFVpApimvSKLJr7J0ic5BvfbT4spS3SljJ+KrU4Qar9epJlPKWf57mpk1QWZRdpr9lkHAud9zhnKWWmlNYY2CI3saX0l/8bgeAdblGr6UJrOLVamXb3vP78J1xf39B1VxgzLOfRNT8EY7DOgXEtkpMzMSXqEjWSFtJZVGEMYtyinlLOQPoqoDhHh9Iyh9+DsXJk5cjKkZUj32SsDFkZsjLk/wcM+Xlvfj5ffn8E/PSdv38E/K/v3OfDdx8kIg54/s7jvzJUdQbmd+4PLFKIixFSCAPHlIgpoirEODEMHSUlam1Sj7VkxHmEVj9JaWZNp42kasVZg1kiBFLBFoNYz/OPPmX/8APmx9fY4RkfvPoOlzc3mGBJCtSmTjHFhKkJFcOHH77ky7c7xuOI8Ru22wtAsTnSOU+kfdilFKbpiA+WVFpd6ma4WD7Ypi2vNWGNw9jFjVcV69rO2BrDGBOlBlQst3f3jNUSc1wiA4r3jlQsORemVJhTJgXIVZjjyWnYsU/KbkyMc6HWhK2KaHOkdmJJFaiZwQvb7cC280zTiAKvHyJvHuH2UNjP7XWlKhbXIGUKwXVAPadwxTTAeO/Jx5EuBErlDByfA4UJVJljAbHcXN1wc7FhsEJvLVs/kOcWjeqcBVHEChdD4KL3dM5iTCUXaR162uqbxQjOGbRKU5Sx9pxCPcHlNNdqrWfn45NEqTFPtabv1qGegGOtPevU55xR0SVt3BpcW5RCzxcHoVKNUPRJicX75oBcc8Zag/MeL4rU5mgtC3xMjRxvX/PT3/1tvN9w/UGH0o5HsBiaNGWpUAukouTS5hdAM1VujbrW28Vo7Cm1/G6a+XQ+TnM3p/z/TYY/+vgTYQisHFk5snJk5cj6XWRlyMqQlSE//83PD2nQ+IdYALPUxf554D9Y7vM/Ajci8udU9X9e/vYP0o78b369lytQE523zFMklkpMhaHfcjEEvAUtkVJoERFTQAvWKCm36INiUAPWtIYtv0hLzvGIJkOQC4pVXv3KD3j58lM+/90fcXXR8+zDVwQXQMF6ixbFBo978QFxHjHHkT/767/B7f2Bn/z0Dfv9Ix99/DEfvHjO69//CQbFaGEuU2tCqx3j1D64ruswptUMexe4uHxGCIGqkfu7N/ROuNxe0vcD43HCe8+cC0mFOEbu9zNjMUxzS6+nOZHn1KIszhOpPI6FzgWqWvZzZJwzY4qMWSlq0GyoKZJKxIhifVNCoSq9t3zwwTXf/vQlvd5xd4zso+Ht2PF/v94x1r4dV46IWJw0CdDtZmAwBmcs02HPnGeuXr7E20BMBW8DXRjaYlKDrYJUYc6ROY3MMfP88oYPn1/y/NJxsx3oncVppVqDilBJbPuBoopVw6++esYXu8wY73kUoapFtS4RNmkXniVFnHM5g+60oMwCxBOMck5naJwiDydInYaIwCItOc8zVZv+vJEW3YL2ko19zaHYGhDTFoExTbXF2qYAE+NMHzzOCKYWjFWsBWea74F3gc4JZt7x5nf+DyRX6g8Ev73GOIe1QM4YKShlUVCpyCILKdIUVWqtpFypx+lcO1zPlgBmiQgtFwp58iF4R1Dm5zH+lBkCK0dWjqwcWTmy/G39LrIyZGXILwFD/jg+PxfAr73zp++JyN8F3Krqj0Xk3wX+FRH5bZ7kJT8D/hqAqv6fIvJfAf+hiPyzNHnJfx/4q/o1VZq8dRiBXHJrvqqVqi1N2DlPsK2ZTFhUO9xS5yhL2kwruPah1rL81ISwuPVisaHH+IJIphuuuHnxKV4iKWUkJkRMU3KJM/PDI7evvySmGcRgljShM8KUJg77RzabHqW21GZ9irYYk1s0xbgmX8nMq1ef0oWewzG2yRtHRAuX24FXH79s+vgKznvG8ch+b/Cmpc+HYaBIobxukpu1ti7OrAXRQsyOca4UhTEpU6lEhayCnmeQUmpBDaCGUkA10znL8+tLNsHBrByOhV0x3O4zUQ1zzpSlgdBai6A4Zxh6h3eWlGdEhM57TvKSc06tPKC2hkhnTJNSpH2mMdWmyELBSAUtyPK75IhqQUWWKEkBAS/CzRD49geXvH4YmR8yEajSFE2aMV1Zzv9Tk90ppXqCzLvNh++sg69EV051uXaRO1XV5h4NLZr3TjOj8KRqYq3F0LwSrNRlcVd8CADkmLHL89eSES/NBVnASSuZACWXgqPJdO7vvuT2p7/Dy299j3D1HGP8+U0YWp2uYE8H0oCZcwNOSpQFsKdo0ika9m4UqpTydJ8TRP+I4xeJIbByZOXIypGVI+t3EVgZsjLkl4chf5zMz98N/Lfv/P9U//qfAP808G/T9Pf/Cs1Y7L8H/oI+6eoD/BM0yPw3PBmL/aWv+0asFYxpkzSXQnNXFqhtN9g5i5j2YXjb0so5ZRBB1CCiywmvTdc9NcnE4AQ1beubtOm1WwRqZbvd0rmBWgopzRhrMcWR54lxd0+cJ4pWrDOIQOcdzoLOiePhESOVWjNTrWitOGObtKJrk8oDKSW6rmMYWq3kNB44HA+oZi76wOV2yyZYxsOME2XoA9t+Q62VKSWmccTimaeZWgrWONR2xDKi2iAy58JchRQLh1iZipJy24kbFdCKUkGhFKhUsILUgusdGyek8YAtmaSe/Sy8fYikKsScWgTJWoxpaebL4OmdpfeBnEaGEDCiiMKUM4hFtZBzxrumuiIi9L1jrgbB4KxdVHESYCm6NAfmSDGANRTNzGlqzXLGMBjHh5eObz3fcD/umA6LFKJpEQeF88JpzYblZyInkHNTNTG2RU3adepvXzon1RXg/FxfdSmuC3BaFCfn3MoFjEVE0ZpRBapSUj5HM5xzzTG6G7BSsAjWCKIZqOTa0uxGFETJ447bz3+HvrdsNx0uXAOOJS+NopT6pJhSa6Us7/lUZ/sUZRKceXKWPo13z9FJLeZrjF8YhsDKkZUjK0dWjqzfRVaGrAz5ZWLIH8fn57/jXJn6B96uwL+6/Pxh97nlj2FG+LPDOUOwFi2ekDM9ljFHrBOMKMF74nxoO3ujxKpUAYrQ+55sCsVkcoSSYLefuPzgku9+50M+P1yRfaZsIkY6TCzUdFxqRT1pOoIVvN2S5onj3Rumu9fc391TxRKGYYGN4Cx0VonHHT+9fY2VNslDCBhj6LqBoR/Y7XcAiGkqJm+++JwpzhyOE947Nl3Ht1+95OXzC0gzOMPNq5c8f/aCbd/hJbPdbpHhmkNSxjTx4sWHjKZnNo6HQ2GaE2qEu8cD4+OIDR0TQqS290VFa6Gm3IovraHUVpdJUbbBc7XdcGGAaU+0mWJu+OmbW768jzweJoq2JkZnN82LQDI3m47BCbYW8IbQuaasYlrEYy6FYD3UioqhlMrN1TX7PHN8M2PEcbnt+PDKs/GVYeipxiBeqWkmRSFXR+0TRTNWlN4KWxv47nVP+NUPOCTl8OP7phxTAOtak2l9Ms06/TjXoiYxNWWVUwr6BIFT7WxrOmz3bU2J+SvRmZQSmJP5GAuIAQwXmw05zlTNeDGErue0rnMu5zmS48zxeKSGwOXgQDM1K8bo0jZZyGqoakgRej8RDp/z2W99STm+5Vd/4zepbkvCUaRdqLsFmApUmtRkl5vai+b0dC5yJkmF8tSAeQJMOV9Uvt6Xll8khsDKkZUjK0dWjnyzsTJkZcjKkPeLIe+L2tsfOnLOlFqIKRFTS7ttNhuGri32bAzOByTPWGfpVGh7bNBFfaTUgqpBEXLJGK8MoSN0liwRMlhVvG+puq73DF1PFogl0vmOLjii1CZlWWGaJlLKWITOebxJFE2IVnKpIBXoAOh9YNP37A47jJhzc1opra/SOYMxQugcNzc3aEnMxwPWCteXlzx79pw5zmy2PTlHVD21QkqFUtqxOuu4vLjC9I798RFZIg0pN/8AsbZNtFpAa6sJBfKSvkeW9xAc1hpqjtQ4Ukzm8XHk/nGi6FKLa+ySTjVUCsFafLA4AwaD7zqsMQQ1iFhUWwREpLkq4wSTEnm2uE7IpRJcoAue4FtTXS3tM3POgDhyaRGEduvSFEeDXWeV64uey81AF/aU1PTrSz15BDSN+aca0gUqFH52nKIOJ438U9QlL4vVOXN+DmMMtbbGPWMEI4J1HmObmVnOmZRSU0CxLfVbtSIYrq6uEGMoOZ8jLq2hsSK2RZ4slXaGm3GZEYeokvPErJEQPNPujofbL9k8/5RshAyIWLI+NQYaYxDrFqCyuGW3UUtZVFaeJDdPx63LbyN/6PeP92asHFk5snJk5cg3GStDVoasDHl/GPJ+b37EoqKkOqO06IATwVZp6impgvOot4jZ4spM0ojWCTS2hVVtM6uyoJ2S1XLln/Ph9TVba3F7EG1uwSIbqJE8JzR4euMhR1It2M01V592lIcHHu4fmXY7fOjohy191+Gniewcfugpi3KF+LYLH8eRdMhoy6+SSyHltsunKhcXG7z3fPriA4YA+92IuJ5PPvku282AM9A5MM4Twg3lYcfx8ZZ8OHB8fCTlyJRbpj+EC65tz2G341hTS3tqwaaKw2KwZDLVpsVluaXirRS8gBdL7wLj/IA1mR3P+fGbB+ZsydVSMDhpwAk64jRxc9nUUJyxlBTppcMZJVgoc6JTQzWGqRaMNaTDyOWFY5TING4QCn1XsFRq7XGdw2qisx1aDVU7UjmAZEzNaKxUKtEoedhgTGQDXJjYjLa04JohNSpPUZZTRKXVyxpySecoyrs1uO12e04ny2LAZYDFqaylpGkQdSLnWlShuSLn0qQjESgCsRRiKXjfGhxd0aYgJDD0A6ZmOid4jWiMZGtQ73DWY0XJKVKZsdZjXYcaTyqJw/2X3P/4b9Frwl++IoZLRmDKhb4kLELxAeM3VGl1zIKiYikIRcB7h+FUO9wiMV0/nKNBu93jn/rS/7mOlSMrR1aOrBz5JmNlyMqQlSHvFUPe682PtZah8xjnuD3scM5ifYeVincekUJnhaqFFCecQHAOqb41rIngvGOaM4Ky6Ts2Q0cw8Pbhjs04EXyHkWZeNs7xrCmv1iO+o2olT4mUlDTOPNzumWJimhJv3zxg1LIJl+xNIi8p5H7YMM8zx7gnxhZpcK65FVdJDMPQ6h9d29132wtevHjBRx9+Qsk7rm9e8PFHr/jw5SvG/Q5D4jsfPWcIHc739NstXe+xznJ3+yXH4wFBORz21Fro+57NZqBkT4oTWirWOEDIeULL3MyrSluUBUGKsrHCzWVguxUyHuyWL3fC7TFxLMr9ccL7DiOOwRkuvXK58VxtAxtvsQiz93TWUHPC+8CxZg4pNffqzpPGIyqF41SoKMfS1E+uL66wRIKzODHnZkTUkXNinueW/o0TQyeIXaQSo8H7ga4f2FxWhuHI/v6ACgRnz67KsEglLnKQKS8KKO80GZ5KA0SEaZrO6WOR1hD5bi3qqea2RVzqVxoZpzFi7UnXnzPA2lxQvLGM+z37UjAI28Fx2XsMle2mpwuGnBJVleBtq81t/yyRnkofAlY8VisPr3+Pt2++4OZbf4ZPfv030eqo1mNEKCWTYiSlW7quw1sDtkdL83DIOaJEhnB5PjZVJbjmBg2c9fnf17FyZOXIypGVI99krAxZGbIy5P1iyHu9+XG21T3WWjFim1SgVIKz9N7jBQpNdpJSKUahZqomxAiyKHpYa5ZUYKXkiUhBcwda24QEEEMRIXiH8Z5UFFdhHGe+/PI1J4n+h4cHfvKTn/D27VtiTqSUsCjbLiC05ys54Y20GspF090Yw3azoR8GpnnCdYEYIyLCsL0g9ANhs+Xx7pFaC8c58fLDD7mlootE5f3DHbvHI2OqzCXx7OaKi4st07QH1aYEUwrjeGSeDcE7um5Ac6LkTNGM0hRYyFCrUGjRH29g6AM3F4HgKmIDx+i5fYjMVTimuaVU1WBtaxLsgmXwwkXfYbWlK9U4vHVNgtNashaK0pRoUDAVvCEpaIFpmkkxkdJMCBYrQt8HumUhp5SIMVGKogKiikERowhKsAZvLck6MM0t2JhTpKRStTX9ncB/MgSr2tLApya8d9PHX1VXWaBUnyQmT+A516OWcq7RBfDeM2y6VqebmrlYA1JTNjnN6+Dd0jTbSgD6zYC1UJYaXKtKzWkh46nxbzGcQ1uzaK2IJEQm5scvqPs7Ns86ojW40BNTQucZoQHLGIdaT6mpPZfUr0D3dA5KKcQYse5rNyn/wo2VIytHVo6sHPkmY2XIypCVIe8XQ97rzQ9SqVWJMbKcW3ywhOAwUnBG0FqpacKYpW60tBPoXGj6685yHCeMcQx913atJWHVYbQ2qcIFCKH3bIOn6wNVFw16FeI8k0tL0VVNxDRyHHfkpUZSULYbj3MtrRxLZU4R6wNgmvRjaRKMaMEYKFnx3uNDT/A9WoXdYeLVp9+GnHh284zd4wHrOy6ut+i8Z3f/wP3jjv2cyLVSELwzDH2DV/BNjrOqEkKH1tqgRlt0lXa8kgpGDUJLe+qiq/7i+oKXNxdQR6ZsePswc7efOcREzLXpuJsmcRlE2HaW7cbSeYFkWlo3eNBC8B5rDEmVuRQ2gNGMNW1ip9Lcn0UKStOX3/ieUpsqS9d5gu8QUSQbCm3Ri1RiSSBNxjHU3I6zKMeYmyyogmiLwlVVrLFfAck8T5T6VGtq7Tsa9AtUGoCeIjXWPMlTvgunUyQFOM+HU22uUcCAVlmg1FLaQx/ahcoKXQiY3FoB52lkc9nTL/DKOZO1QddZh7cB1dRSw1agVKxUgqkEZyjTjtvf/yF9v8Ff9aesOBjBCVhhqS+2VH1XQ1+JaabvexwtKqiqKJWcn5RW3tuxcmTlyMqRlSPfZKwMWRmyMuS9Ysh7vflRTeRUsFIoqSJVCWK52W5xJEo8kOa5HaQxTUsfpWQl5plcFyMpayklMR4e2HQdfd8h4vB2KZuk0awzhr5z9J0nLylY4wP90LPfH5inxE8/+4I3r2+ZxkhetNVFmnFTihEjQjBC2Ay8nRPH45FtP2CswRppOurWU8sMGExV+tCDwv3jA9eXPd/6+GM+/eQVwTnu3n7JZ5+9oatH9o+PxBg5HiayNFnN29vXxHluacM0Lw6+lZISFxcX9N0FObX3hXjmQyVlSy6ZuRaKWLTCRe/5+IOBedxxHCN3o+XLh4ldCsxzwjuPIlgp9GJ4drnhcmvxTlvzZ4G+G+g6h0wzXXfBmCr7OFFKJhjYOEt1Pcd5prcbJBZMHXn18gMuOkvvhZuLDRfbLRdDj7dCjDOyRNpSybjO4oNHtTWfahXEekZ1PM6FWEtbjBj8AhAr9pw2LbWQc8VY3/7/TuPhKcpwTnPz1ca7r87Np2bF0xwAFinUxLh/uhjpMi+D67HGUPJMs9I2iIXOC14M3gIlosaS06LkoqV5TBhwRhDpUIWcCr2rOFOxVJw6Ypz4/If/C7uHz/jk1/9+Lj76Ls53eB9IcU/NkUOqxNlAyagmlIKxFmvsGZhdHxCR87l4tynxfRwrR1aOrBxZOfJNxsqQlSErQ94vhrzXm58SZ4I3bIZW/9j3Hu89Kc2cXF/P910mST1JTC7pTaXp8xvjccY2NRNRgjd4ZzFGUAFymxRGFapixOJCoNqT6ZIyT5m720fmKeN9wBjLNE1Mc6Tve7p+wziOHI4TVSsz7T1WafKFwbs2wWvFG0PfbTDeYQWmcWJzdc08z3RdR9cFxsORvh/onNDVgYf7ex52O45zJaqiNTXoqS7PqxgR/FKbOY4jXReopamDBKdYE5hFmbWQqlI0E6xh6C3OCiVbknZ8cf9IwhFTm5BKq8v0xjN0wuAFIzDPkSH0VBXEClIiznWUCvPiL7DtHJ6CqiDGoBisczBGUkpcfHCJ0wraFpl3TSc+zqlFr3I6T3oFcm7qMc56NheXiA+8fn3g9nFE1WCtNtWYpTygRVjmJaVaz5GWU5QEQLWSc1mkJtuycc61ZsNFevE0TnW1J/Ot0+KEBh7LU/1uFxxgcdIig0UVWzNDH5BasZrpQ09nITiDb52MBG/xzlFK5jBnYpyoiwu3NU2VJtZICIrxTX3FisHJzHy85f7N73P58ltnI77gffNWKIBo84ZQi9BhvUXEIedzoefjq7WSYvp5L+0/1bFyZOXIypGVI99krAxZGbIy5P1iyHu9+amaccZRES4uLjiOCe8MXQhomgCLcY6UIrlW5jhjLWhRrPOUWsk5Y2yT8VMtdL5f5PZabtrYxWRsaVB0QJ4n1HowlpwTvg9cPbtmvxtJJdNvBuZ5JpdMVlAjPBz3iAgpJaoRpjmRVHHBLunIk265IGLpXcfF9pJxjpiikDLUwu5xx+3dHddXF7x49hxvBYdyvPuSm+cvEGN5/OxL9rsjVTMxFgyKc0tqN0bSaXFqIaWId7al7KdI7weGy45p90CuBSuVTTBcDp4UC5XAYZqZEsyaGecCtTkxWzEMnSM4MJKw1eNtx9ANlFLoOocpM13XU8UzjiMvbq5xUhisY55HfLchGMs8RrwzOGvahQBlEzo0N93/YgTrDFkXQ7TafrKF40RrDA0GbKAYz+3ukdvHkSnFZu4mFtWTsVY5R1NAvjLHvuq23LT0RbRVJCyLLjh/Tv2eQKJa0Vb8uqirVBBBRKm1GXhxuuDU0q5nJdL5QNcF0Iw14J3Ham7OywreWkrMYA1D31GKA1Moy8XE+Y6Sa5OldAbrBeMsOVfyKU2eZuLuLVYKuRpyBGvAirRa8mCQkpsruOsQY8BaqM0r4OQpcIoove9j5cjKkZUjK0e+yVgZsjJkZcj7xZD3evMjVIwo3hm0JDofuOgGjocHtn0PYslRKGpwXkhZETEMQ4cxhphbLeNJJ90at0DGgjGYsCzSnAhG8MC831PTxDFGdnMi9AOu73l7f88Xbz/n/nDP8XhknmcOU4sEqDa1j1TaB5VKplQlVcOz62vG45HDNNI53yI9ISAKJUe2fYfV2gy5amIcZ370wx9y3D/ygx98nz501JzpyFzevCBsNoxmoN+N7B9eM08T4/FAzRGhYo2hdwHkabGVEmlrLXA4RvYaka5j2wfs/MjHL3r6zvCjz+65e4hM1TJiGFNirm3hWCtsQ8fzy8DQFTpX2QQl2MB0OHDzbAu6J1hH0kIscPfwyLc+eUEvBSOw8467w8jt444L385F11koic4LmxAYOiErxFSwtWnSh+BJ2ZFTxlnhOGacs3T9wBePkbdvH/jRm5FDUsR6jK2tkQ9PyeWcTq61oDzV0Z5Gu/1JMaVJSsp5wZ2UWYAnKUoB0VZLW0tGlgifEcG4FlkJIWBpOvvGKF23xXtL55SL/gLR3C4WFDrv2Q49WhM4i0EZD4dWB44sikJQ4oFST0Bskp/WW5yvaKqE2pNKJe0/4/Z3f4tnH3+fWT2TVsQ3i3ND61u00o5Fq5JqxNsWzYTmHXGC8SlN/76OlSMrR1aOrBz5JmNlyMqQlSHvF0Pe782PEUKwTPPcDJpcwFnDEDpKTUsq0FCrwTll0/XENDfVCSvNWIw2iYywKK20k5jLosBRCkYsJWcOKSPxyLS/Z7c/cEiF4fqGah2Pjzve3t9ze3/fVExiJOWMKlQB13WQF4WPXDDG4n3HRx99xGeffcbu4Z59SvQ+gBSMtEhQ31f6q0Ctmd1xx4vr5zx//gEff/QKay3jOFJzJpaJ3cMt++OB+31knCuHw0RaDK9AzylnapOzdNZSBWpt6hrWerCQpkyKiY0TbraBoXOkVHjYZaZqyeKJcaaW1shnBDrv2PQO7wzeVIJRtGScB4Nipb22CQEVbanhWrjoOpwU5nHE2id34lQrNU5QM2hLQxuzKIBUpRihppYaFWsJ3kE1qCYQYU6Z4zTjuy2jeo7xSEytYVFpkpBGBKyl1kqMM8qTe/IpjdrSzU8qI9bac5PgOdJQ9QyfkwOziDTPhnfuV0pBFgWVApSScc4jC+RyjNQsbC88Qwg4G/BOSMcd3hpyjMAp3d2g1nmHOIuKbU2WVKBSq1C1Na+WnLE2L6osFjWGMh/Z3X7O9fOPcd01uI6sGSNKLZm6QMta28oxRJpyzqKx33Xd36Y2876OlSMrR1aOrBz5JmNlyMqQlSHvF0Pe683PcRq5uejIWgnBsxm2WKRJDLpWI9l1XZs4VMQbUq6oFqxYOmdJtWBEzxKPp8awWqFqc+bVCiU3nXFXWw1nCBbxnv1hz++9uWOeC7tjRk2PtYqrhikVFLDW8fzFy3PjYW+ErusIzrDdbnn+/Dkiwu7hkVSVNM10zjcX4hjJtUUD9vs9v/rt73N1dYULzR/Ah9bwOM+J3eHAmzevOSZDKoYpzhRtLrwlKU4MIrU1oS1pUbs49GopzSegtsiHU6Uzlpc3N2w7eNgdyCZQrWmSjnNEMGBKg3kI9L5JHjoxBLtEwwxYK4tUY2WaRhh6VCvWKGGR+VThPMG9seQ0M6WIWxZ8yU2+URDgJBEpqKRF273irGBRNpuBnFsU5LhPvM6JMVeMCFohVdCacbYtyKoVH9w5fXqaC6eFrfrULAic08qnuWKQc03uKXJj7FMN79PjWJoK248TgzXQXO0qsjweddSameJEtQYni3yQaa+FCLUq277HWSGVRMlpgb/irIAxWOMWA7VILZGihmlWigpqCg9vP+fm+RcML3sIQ2uAdZYcZ+Yyk0pt79FydnXWU11uCF+pt32fx8qRlSMrR1aOfJOxMmRlyMqQ94sh7/XmZ0yF4zyTC3Rhw3bYoDESjOK8JSmkpFxuB6pOxDnSu/+XvbfbkSRJsvQ+EVU1M/f4yb+unin2TC+5Sy5I8B34DHxBArzhE/CGr7C8JUiAwDa43Oma6Z7uqsqMCHc3M1UV4YWYe2R1EyDIXJAMwBTIysxI/zM11eOlco6co7RmDEkYUmZtgtjmoCGvi8lFqUunXlYkZUaogdwAACAASURBVMyDIqYutPnM/PyFy1r5cm4sFX58OvHT589ceo0AqmnE1wvmztPzC3/88SdaazeP9XVtDBr64NY79/f3fPju1+Scw/LRnPl8YmmVP/30J8ZhZDre8fnpKapIQ+FXnx44Hu/5+fITIkpbVk6nZ/7405m5CkkXWl0oKozThFoHN2qLprCgVx0M3ODSz7wsM+ulMSV4fz9g5zM//PHE2YRnS8y10nrnMBayKiqZhztnLJ0hrQwMFBQ3J08Di63oNPC8OkOZGOj0dUbM+Pu/eQ91ZW2NL08nfDzgdYW6ojlhOXNIMCYheYSG9bEwDAVHqX1lrTVcRFTQJBzygB4GyjghJP7hcs/v/3DiH/74I8/nBaMgOtANqHM0kH71RQOv2tpaK7WG577IX4PK1z//2k5ym9zbv0OA3dUzP2/07DAMWK+0NRK0P314F9UeVmiNosphGFAJQN1ekWi0DBccQSk5o1rx1tBUtiZIJ6lzV5QpTeAj1SJ7oplRPbG+/Mi//93/xK8lk/5mIpUBw6AZvRFWntax7TpEUuiH/VZGAritp7c6dhzZcWTHkR1HvmXsGLJjyI4hbwtD3vThp3XnPM/gyt3dp9BAujNOI7WvlKHQ12iKcgvtZWIk5wi/muc5nFc2j/Dszt3xyDAMLDaCx4ldCD96SQnxTG2Vp+cvnM4rn8+VPz1XnpeVcztRfWG5XFiWhcvlElUGc0Th7u7ANT13KMq0WRiWUljXlZ++PMWCzJl/8Xe/wekkUS4vz7QelZsvz8/c398DYL1zuVwYxoGnzysvL8/M84xbULOtVVJOccLfNKR9c1NRUUQVRQJMa6VawxWUxHHIHEqizjMpjZSUoDXqstC78Xg8kMTJKIdRUGmUBFMKulk0tKW1NcpwAHGahdd/aFaFkhN9rXSHjrNcLozjyDQN/Hy5IKo83B+o64zRqGmk1coMDAXMbbs/xpQTZZyYUmd8fEBTZl07Ly8XvjydaK1f9wdXutTccfWgzDfK1N1vYW8pKb1fmw9juPsNWK7Vma+bDCE8581C060aTi7XETrrCBqzTWurAnVdyTnHF1INa0yVaGxEI97aHNqm17Xe8JxIY2HImVFHWm1Y39xbMIRY2yIZ8Ui1Vio5C26Z3p3L5cS8XDgQ+QUNJ2+Ba6oJVIC2XZvedMVrDSnHW5aqXMeOIzuO7Diy48i3jB1DdgzZMeRtYcibPvzYcsJLouaPHEVh+QlVY5ERSQmsgl2wtdLTzJQf6J6QLliu6NjJ1tGmdJmwNJCHkcMgfJYHvEykre1MBTSHPnPtys8vMFfl82Xm3M+s6wVZG7Ks1CW0vBkFLXRVUsrc3x159+4dh7Hwpz/9Ef/KseN8PuPtQu2Ol5Ef/vBH7u8fuHv3SEsD8+WCXSrlQ8a98fz8wpQSy7nhGP/rv/vfOV0aw7tfc9CfkcsFWxJ4QyJJK2wTBRKKmJBcyKIstrLUC7V3sIQUR8aE58T2BC5zZ5mNUQtl6Nxlw3un5M4hHbG2hViNNZw6WiOVY+h8fd5o0EK3LeW4R8jXmcxliRyBdvqJ+/KBwkqWgnVn0JFUwqlkygmVzjhMmC30VlE6GSNL5+E4cD8OHAel6cDncuQfLj/y+yfj0kcUI0mj024yAnEHSuihN21pGKNEwBsovUcT5XVcNbW3qkxfESfCuHilqq/+96M5lhINMIy7AsUr2aBdYo5VM+fTEx/fPyJMPD4oYivtUnm4HxiPYXX69HmlN2McGsdROG5VL9TJQwYUI/IMrPeoJgLDUKLJ0YTeDPHOcRi42Avr8z/xrv2XW5hcBe8RtjZklESvI8Z8A9iQDXTcMiLpL7flmxs7juw4suPIjiPfMnYM2TFkx5C3hSFv+vBzHr7jh6ef+fAYacTH+3s0C9Y7gwAWHuRNolKhnkg5sdLx1JCewTJSlIzSJZw31m74NDEcjpRhwNAIqfKw8TsvK3/++TPny8yXy4kvl1M4T5gzz3FjUs483r9j7Y1mThkmzIxeF/LdyKcPH3m8u+cPf/4D67ry3Xf3rLVTa+eyVMyML09PXOZL0NAf3pHLgJnx+fNnnp6e+CEp93d3iAh/+vNPJIHh8R5cSCRKGbicZ0SNIWd6vWpSHRCsN+bu0RDZhd6iVpDLQO3GZVl5Pw2sa8N7JeGMOfN4d2CQTlsX7qfClBXXQs4a6b09UpMhbAtLSYikoLtTNIK+PD3z9PQFT0Ebi8CnTx84n8+h3Wwd98Q6X/j4ODKkgeOQeHy4p9agNlUFM2FeK4cpgyREFMqElAd++rnz45eZyzpHWJhvQWBbcrOjAS7+apV4pZN9K5GEe4jfqlTxvq8uKwJ4B0nx8whue9XgAqH5FlAiWVpFcDHWWnk4hkWp9s4wHHk+XTgqnM5wKMrhkPnw8Y6xCMtywR8SSQaOh8xQFOtGKdH8Ny8Lra3h3rPOERg3DehYsK1p8pUSB+ghyVgvDDm+hEAZy4CkDCqsaw0wxrfk8Kt2OB5rZvT2drX6sOPIjiM7juw48m1jx5AdQ3YMeVsY8qYPP/ff/2vO//Q78ITQ0TLiKmTrDCnRrdMTqAveDGuO906ziuQeJ0kTUsloErIb3ZxOJg9H0pYUbL65VZjTu9O683w+89PPn6kep1pPmXEYGYb11qh2PIyMNrK0igO+LcKff/wpFmOHIoXFVuZNzzuOI2hmWerWcGY8ffmRaZoo091G1RZ6XflP/+W/RMz5+eefmaYJ641aK1kTqzsQTXJKLHLbmgzDS15BieuS+Gy6+acHDQu1O5e1sawVt85UMg93I/fjwHJ5ZpoGhhyuHiULOQUFK9vGFt+CzJLgDnljXXuvpCwUC6eRtS+klBjyAbm743T5KRoIYaPFD0xTYSyZVDKiwrKu9K2iICIcDkeGMSPqnNbG8zLz7//4ws9P5wgEcyM6GfUVXLQAjmwU89d+8QEWfnNMAW4U85VeDUoXlGheZNNpf63BVQQXieZXgajedJobY848PBxw20K6PDEvle4V98JQBsZhC7tTIaXMYYrKXymJlCLvwa1zmS84yjiOWwMnrBqpzfQOomCGtRZmOygaSwBrK7SFUg50v9qrSpDmFrQzwm1+rnT7dV6+bsB8i2PHkR1HdhzZceRbxo4hO4bsGPK2MORNH34+/fY/ZxRIn/94S6ptHSZVxpJYF+HSGrW3WLyt02rH1BALp5GUJrQMJO1h4SeK6T3l7pFUym2CgypVPBWm+wc+/frXSCl8fnpiefpCzplSBu7u7mhrRRSGnCjDxOl84nxZWOtKXSARfv5nPW9OFeGYcTyO5GEk1ca6fgmKmI5b43x6Jnfj4e6e87qgwLIszPPM6XRiKAncmU9n6BUxB6+oOPjWTKhhtdkJy8eoKXicmyVcOdydbkZH6C4YgiOUlDiOA+/uJtQ7VZzDWBilktTJOcH2WpL0ZkeYc2LIJTZUa8g0cDldUBEOx5HTZeEwDqzryrxcgtlOinujpMxYlFK2lGVgXdebBeV6vTdit02Qk/JS4c+Xyh8+z7xc6s0+0yQcQjBH1AiVsQK/dAkRCWR0F4z4eVhfhsY0qNZXS8nrxnO+Sk7ewCZ+/go+4TuTEA2qe13Om9NMopEppTBmIUlmnleSJFqbqEulzis5OdPhsAG5Iji1NaxVbKvmmEf/31DyprF2tESTYO9Bf+MdXENLXhvLfGLQuEa4funY63yIYtajSrNd85VqD9vJtzt2HNlxZMeRHUe+ZewYsmPIjiFvC0Pe9OEnP/4tj5+eaC8/MgxBt40lkZaZvnagMYwD3hPMznhQNMNlVZLHSVQyWKtIaoyHkZYf0ft/SXr3PSaRRhunakc0Fsqvv/8Nf/fb/5iUM7//hx/4H//Nv2FZFnJJHMaB5XJhXmbqfGEsme/ev6O9B5XwZf/hh3/g/eM9tTaGKZoZX85nvBtY5zBOyKPT6oL3xtxWJCmtLrz8vJBVGcbMP/zvv6OZMZ/PfHz/gaTCOl+QvtJrpXAha6KTaK40F85LZV2NaVCKtS07wBlybMYqjq2dtRunJehRMWdI8OGQGKRS28L7+7CILNJD5+kdSYpICp3pFkqVUkK883B3YF1XvpxOlJTwjfb81cd3N23qjz/9iWVpgPDu7g40cRijOnM5N3pJ4CWqBBLZCEmd4zSFF/12rZ974n/+x8/8Lz985sulRaMeoZF2QNwRiewEEITXioqZ3ShZANlABkIrfaWir0FiWFgsppQQEln8thGvgLTWlVQizE0k3lFTIm1VKG+d2SouHd/o3iEVsBVrTtaw/1zqyjgcoTeWuTI83FOGghaBA6CKarpVBL1fMxQEa43eG4pRkkZehBnejTafef7pH/k0DWiaNpI95jjnhCpbk6ZSSt4AJqpRDr9ownyLY8eRHUd2HNlx5FvGjiE7huwY8rYw5E0ffrBOnc8M6gzDhAyZLCASVKkURZrQTVFXujU8gSSh5CHasbrTBUpKSM704R45/ArTqyd/hIP1bnGEFQcVxuPAOE0c7+/5+PETzy/PgCFDuW1Asw7WMOscxhFQDscDj3f3iAj3hwN5HEg58+7dA607WgZqrdiycP/4wHI5U9RpbcU98fT0hJREGu/DEWNb7Ou6IBhYIxMfVVPGgHWxCF/rxmWe6X0llZEsoaUUF9JGpa/WMY/F1cyZa1hNDiVxHMOtheQkFcwixVi3PrNcyq1ylMcUtYybrtOx1nFXHh/fk5Pw8vTMVAprW8jKLYirtQYp0pI1GZoSKtHcNiQlaYrwq7YyTplpLJh1anV8KHw+G7//84k/v4Sf/JCjama8bhRBuCYlf00Ni8gNcErJ5JJuOQSttRuYqCre7bbhrq+hX2l1a63xOjnjKuHSg900t1H1GVBJJDcO04F1WcneESo5Ob1WLqdnPr674258hxCOPEKnDJmcCp6UaQp7SAfmZY01tEkVFEdSAFmPlAlq31xgSOQEp6efef/pb5A8sjHLUY/bKnVf0/CqaQNm3x71xseOIzuO7Diy48i3jB1DdgzZMeRNYcibPvzY/ISdfyJrnJZVlAGji0MKmrC1xuWycjCjJTARVEEsbBUFYcrCdJxguCMd3lPHO9AMCN071j00mqIgEYK1rCuIkEvh06fvEIRlOZGkoBIL6/z8hHfDemMo96RcaHVhHAqtN3IRpqGQStgKah7Di//pheHTB+4f7vnDD79H+sqYD2DCWiL1N/nm/IIzlAHzhvcO/dqAR1SLzFnqwstlxsxprZNToqQ4NUtSxIMCpXWk9tCGElada3Ny8K4oRttSqTUbKSlDyZv3enyu82UhbBWV+IfENB4oQ6atlSkNYE6ZBkrOtLqGdeVGY65zJaviAto7jYZYhiQcj0foNTSt1pkOhaEkhjFSp6Mokvjj5yd+fF5Z2ka7QiCwxwaJ/tPQj4qAymu14Grvea26fJ2ZdaWXrzrqbq+aW4Cs0JrdKjFXevnqztJw1J2iEoCRlXU1PAluxnFsZGkUhbF0DlMheaG3GSzx4f0jvcbnL0PmcJhozTifVloPa1B3YVlm1nUFd3JSNEXlJ+etwoPSVqP3ipni68r55TN1PSPDkb7BgpnRPdxm4NU5BoLyNzMg1tRbHjuO7Diy48iOI98ydgzZMWTHkLeFIW/78POn3/FenpgOUK1yNEdsxc3wDI0IgJrKxNBOXERp1ikFfDVKOXJ/f8d4UE5d+cNzhodfUQ6PSBm2ZqxOSoKmQk4jrkqPFc+6Lqgqf/fbv+fjxw98+fwjXz7/RCmROvv4+IiIMIwjOQ/x52HbaK3GJrFGXVbaKuSxklLh/cMRetDb7x7umUpoHMU6vU5xmlcwlGMuQS9ao1rFxOmtMl8uvKyw1ooSrjNZjeNd4TAO3B0nNF+TiRPdhfryQrcLbhaOGi4YxphC10pfOY6JlAaQ0NAWFdba6N1Y15j7Mg4ATNOBq5d8u1SadYYcStO2Vqx18qSIG7lkRJTD4YDhAVY0eltYVyGnhFjnOCbGUihlpOQjtS2cLjMCjMPI2Y3f/dMLf/i8cq4dzMJTfmuqDOAJHa0KIK9Nc9dx081u3v1XgBnHkZwjG6HXdtOamhkqjpvFFweCeViHGmBNQsudE0mcZJUEDJpp1bHaGbPz6XFk1ML55cL3v77n8eGO9dKop88kWfD+QmYkSUJoXM4rzoAmQbzTWwWCgre+pSyn0OOyhY1B4rKsZFWkDGgX5tbQemI5f+Hw+B1NFO9tA2g2atkZhljDvfdbBapvja1veew4suPIjiM7jnzL2DFkx5AdQ94Whrzpw4+c/sz7SfGueBJSss1NImjBS12YyggmqDtJlEYPulQFzQN5mFhx/unziZ/tPR+GB7QUIJrAIDq2UopGN5egZh3oPWwEx2EgPz4yDolxLNsNML779a95uL9nrY1/+2//Lcs8R7OdgZBwaVHxEah1pXfjcDiQp8KyLpg50zRStrukdl1U0H2jCnPERWGNuTfaMmNtoa4L85oxd/JGDU9D5jgNTGNhHAZSKRiJZkJdG7X7tsGCEnaULkJHEc0MgzKWcPuI5kQnEnwbjmMb7a1CNFqOB0TivV9ezkzTxDgWRMG7hcuKBk1Oiec5nZJHEg0lsVbFekNTCgcXixC0MiiIUg3O80oZJtYKP3x54Z+fLszV6BaaWnrHUKQkNLQIG60dG+prsvRKKUNsNNGgvs0sXGCGIewva3vV4orfqFm5amnt1V7SLYLdAsadsShTTmBOThK6aXUeDyMPk3LOA4eSGVPi/vHIhcqQK2PO2EpUyrLwfLrEl85xYhgKl3nlPM/bl1kQ67071qM641vVpOREWwE6JQ20DnM9s1yeEXFUAVHUBXrH1cJ5xyLcrFvfile+BbC97YyOHUd2HNlxZMeRbxk7huwYsmPI28KQN334GX1lUjDJ9Ay9L9S6oJrpvdFaRwdFfd20mpCIRXhdHC6ZH08XvrSB8Vd/Rznc427gfluIqtFsFoykIMRpPcVfcQRNyjiNfBq+226E8P33/xEfP/2K0+nMv//9P7KsYe3XPKjba0NZLMxKa5XeC6fTM+sSwJJLwtxIKePiHO+OQSt6WC8aGyXYO96iuoCHTnQsimhCw3uD493EcRxJqqQ8oGlAROndMF/ptjmEEBWRILKF1p3uTkqvFKRI0LTXOcopQq3k1oRXETcwQzLkHNaZ01BYzTFrm2YzNuY8z+SSoFauKcmCUcoUlZnewDOI0HEuy0rticvaOa2dQynU2fjdD5/505cIaDMgbzaKbgGILoKqo2yAIIJslpN/2RzoDtbDZcTdWderWwu3n4lIuKVsj8M7orEJb6nDfaPvbUtcbtBFKKo8HAbuxkKSireFw/iO7ApWWS8LJo3DWJjGQhahi9BqxSWsTh2nt8Y4DcghGgzdPO6xRCOtWcO7APbqJCORup1ToSTl3BfOpyfWdUbSMaxBkyIOLrEWrpWnr+0l4wtG/gPu6v/3x44jO47sOLLjyLeMHUN2DNkx5G1hyJs+/ByHoB8FoaSGSccHxZbQU47jhFlnTEaVgcFqeJKXA5oTKWUW77wM33P4F7+hPH4HKZM89IrdO2g0knXvSO+oCOKO9XCsMDfMNru9XFAv4envxk/PFw7vnPuP3/H9b/8T/B9/4PLywrpWcKPVQlEAI+fC5fLC5XwCItBJk9IpiCrLegmt65DRFjt9miZU89YMGRWlXjK9Km6dh9rQnJBcICU0D5BHOspimdQUEQfr0CKIqjukDUZMAEl0N5baWVvfGgyh5Fho7Ra2FXNQSg76tRTcGmMZgiaeDqhcU3uBbmQR8M5QlNoDwN69e0BVOHjFW6epID1eP2nUNMyM03mmAnM1VoMvz40fPz/zu3+eeVrCHlNEb7S5qEdGQnWQdqtyfG2neA0E+5p2FtFNXi03YJfNdjEuxEkIpAQeuQXXTV225kMXAevkrSaTckElqN9D6by/K9xPI++mxHx6JomhJtTLBVPh+HCA7nSDD+/egwrneeZOJrqHFee6Or05WCNt+u+SM+KNvhpDTiTNXC4nQMlZmaYBaxYABrw8/8SXn3/koMdowM1l+1YNwEwaXyjucgsZG8eR/IYrtrDjyI4jO47sOPJtY8eQHUN2DHlbGPKmDz9tXuCYo3JiFfOGlgFpkHE0K+oN6Z00HijNcBdIhbuHe9a68M8/f6F8/1+Q3v8dPoxRqbCgpSWBieMSPuPdDCdcK2KxGW4GkiI8ShPdo6vNu7I2Y2nGUTPf/+bvOV0utNoow4hZJ+lAazOtd/pmY8nW1HV1OamtUoYhqFw33IS+0YjSIbtSSuHueBcUby20BcwaY1lBoalgSWmaMU24Z5yEG6iEKwttiUrBVoG4aizNoXWjtXCZ8aSEHaTS6op5VFFaa+SUGHJGk0IOpxU2u8VxTPRuqF+Dx0DTNajLSEmYDhOOhkZULPzvJZE1Uq3XZaaUTPUA9Je1snTH04HTWvnnzye+XJylhyO+elSL3I3glhPmm4Sgd4SoEIHfAAW4/Q5smQG/HK+OLApEc2TM218/TmCTJ8CgSk4ZSXlrZoSSnZIqD8cjY0qsawNfGPIRHYbw3S+QRbBauVy+MIwjIhbWjwjratSlhaWkO+MwEHrbTlIlDwNWK5IT4zCwdkebbFU1oDdQoS4z63LhTnWrs9kmKOi4xetoyTfHlZtzzlfz9RbHjiM7juw4suPIt4wdQ3YM2THkbWHImz78VAdPkKxTV6fJSCojmiveK8la6CzzSCdBGhA3ulU+z86fz4kX/Vse7v8Wxodtj8Vid+2gSuThBsY4K83jxKxC0JWqm1NLWAaqOCagScKRZL2g4nz4+IGPHz8xn07MLy+0dcGkUlul9RY30TUoRIQ8DKFnNCF5ppuRU4qTsQvdobkhtjCYkzRTprtIlk4ZsY55o7YaTiCAewntJIZHKx/AtukF3UC0k659eEAN0HHF80AelEyjm1ObkXL48qvBkOBYwjGmuvBynqO5cYDjYWBdZlQhtUqY4MjN9WUcB44lAtcQNrBVnEQagC3Qix5zXrtxXhuLK5Izfzo5Py0T1WK+xcM+MoAg3zZ/3hpP8aig2XXjaro5oWyzcpubKyCpCjkrRRKX+RI0dFKsGyqBaTdtbTfkFrgVa8KyQk4UdXpfyeOIkDmdO4fBKXcJzwPeGrkMuLUtgyBThkIjMdfGZb0E9Z8zZRq4G0dO7cJlXegOkjNDyvQWkgNNQq2VulZSGWjWEKIy1HpDMxTPEUZ3eWZQwdIBzQW3GjaammjulI2iv9Hs8poi/VbHjiM7juw4suPIt4wdQ3YM2THkbWHImz78/Ol0YZDEIQW9J8NIypm5nlBvlHHi82nFpYBcMBHOc+O8dvTdyOHj3/Ph4dfo9B6X2OwY9O0EHSFWaftzaCSvk+sSmkyBm570dQtDUqX1xpfPP1OGkb/9/jf89re/Rd2Yn5+5AMtabyd0JzYGgGhUXITQMnZzQt+rqGZENo3tpou9rGHpmFLQ3VnCOrP3hqSG9i0sy6MiEG95rRgJtTYu80pt4ZTiPWw0Q2QZ1QKkkPKIaidpCi1nHri/P3IYBmQcNivFq+sIDGXY9LVDWFJK/HsptmlQlVpX1nXmGv5lLWwLp/s7yjDQmnM6P8d8qETAWjeqOUuHBaXOjc+nztxhGAZybdQWemNEN0/7aBu9QkrcUt8oZjB71Y5GFeH1Xsv2ucNRJOw5c0osmx77+qKxuRP9KknY3ic0vZByYhwHbDmRpFPEOUwjtJnz+cxdCRo/J2VZLmQVsjhtXbi0ShZhKAPu0KzjON4aojCWBHpg7UbrjUutJJHIJyByAl5lAUoCSgmq3wFdBVrl8vKFkgTLBS0DWRK9J9bVWNaVtumJdQObK8C+5bHjyI4jO47sOPItY8eQHUN2DHlbGPKmDz/LeMeP9YVxOUfVoU8Yjb6cKVlplvmynGkIUuHUlKdVkOk933/8Vxx+9RukHHFSTLw4pmzaRQmJ4W3ZxPirBFl5Dc56Xc4BD+4R5PXnP//Ir//me95/eMe6/A2//3f/G61Vagswcwe7nlglbQ18r5aHrW8JvqK4hCOLkmIBybZxeqNvet+0hW+llGnasH7Vf4afvOqWKO3RLGgeGy7ko19TptcFJayt8fRyggnuJ6WkxOFwZBwGRELzazUWr3WjrY3DdEBVozluXfHecM1B4wMlK8LA/TGqAaloVIjMaK0yjAO5KIfpwDyfN0tDo3tQyV0LaxVO88q6xvyVEmBQW7iJ+Aay+I09/6txBZKri8rrHvJb5Sfo5LCQbAiimaSOeVRvzMNJ5vomOafAazOSRoOheoBgStySqktyVJRMRb0xpkwWD3mEJpStmsUGYL2Rc2EcJ5rF/R9SaK0HDQeY87ywrg0V6K0hKpFk7c7a1iCRe1iOunuEv0lc77pcouqkggKqCdwRCdeWK8VcxvEX2uO3PHYc2XFkx5EdR75l7BiyY8iOIW8LQ9704efdb/4z8uWfOH3+Ab+c0fPM3IRixlBgtYXTurBYpy8HFr1j+tX3PP767zl893e0fMANskZoFu6I5jAB9KhEXMe1+exrbWHcrKveEoyv3DmIhdxqZ1kWrnrSYRg43t+Fo0jOXFNqVa/Upv0CuK5UJxsRjEvoNtWjLLMtdiAqLM0wj8+ay0DOA56uVYXXCk3YpYfVZkrha7/VFlBJm74ywE0Il5V5rkwKswjleEcphdZqXNvm3a4az00C41BuG631Te8sBVPAPa5fjePdgZfnZ/I4UIbCly9fSKos84VUBu7ujgjG5RyNloaw9kbzzKV2Xs6N7kpOipEiQKsKvcPmpcPVL+dWHfgKfK5Vl9v8+wY8W+Pkq41k0NyiyjAMaEq0HgnGV7vKr1/rOlQjXdqt0+vClEHd8b4iPTMWKJpQKkWVMWfcG0PW7VeAqJtR18qyNI7pSP7K5nPIGauVZZ0RnHHIiEOvctL7/QAAIABJREFUK63HXLs1lvmES8bIUYza7rlofKZwsumAYdZofm2ofQWVr6ssrb3tfA7YcWTHkR1Hdhz5trFjyI4hO4a8LQx504efw4d/xfDuew7v/zXry4+xgXqjzp9Dg5oL090dYxpA7tDDO8r9I+XuEc8DweBuFLEIIhmXgouGt7sGmMj11K0a4CJBLbs76NVrXkKnCBs9HIu2bZWDlBUz53h/x3d/+zexqDU2cW0N1UatS7i2EG4gbvG5to9H67GhISokSdN2+k3xvO0zdYvHdxOGPGyf2UlZUE8bPWqx6bbnywY6gm3aSSFALpZ0d4+AMItmwWVZcXNEVo7jyHk+oS6MYyQZD2PBtxwARClJ0ZQYUmI8TKy1UkphrSutrkzTiGGMJXEYC8OQeTnNmC3QE1MZYHQWjHapLB1OKzyfnedLR1IAjiOUFIDXPcD3Wj7ZCgq3sUEMQoBCVL5+WXnJWW8b7OtK29WNpeQh5r21W6Nl0ggbS6Lx1m5kTVHdMke9c3cYeLwfeX8/MeTOqM6nhxElnEse7t7hvaICWYVxKGRVlhSJ0zlFSJ5bCBxyypSe6BIe/t2uny+FVaQqZchMHo2czQqXebk9X8QRU8Qry/kFPSa6h6TArN2+BK+62nVdEZFfSB7e6thxZMeRHUd2HPmWsWPIjiE7hrwtDHnThx8td5T0ASZH7xdUO0rFlp/w1hEdWNM7cjmSSiENAybhG9G9g7eoJqTwmk9pQD2hHuFhCQPkVn2xbrhCbPivTu8bKFxP2teKiwLLunD+8oSokDWTU+L77/8WPNxZvryckWUFolkxMMCwGz+6XauGXtRFwIN2TBt9qCSMoI25Pt8c90bSvAU/yVYV2uwye4+gKTdMIOVMShnrNU7hW1jWVY3rHiFVknRrxjOWdSZpo+WMbzraUSZ661GV6jXmIRXKMEQFKCvHadqqLUrvQm2NlBLWwrKzlNjo93dH5jrj1kNfas6yrnSErpnnS+c0d1pPlCSIxCYfc2JOwlrDE/5aU7hijdz+E38QrlWtGK/30UgtNKrXitqVVm7requ+xRrQ14rcthGjIqGw6XGveQSHYeD+OPB4nBgSPB4nJjXe3Y9Yj2bAw1iQTQKRgmNmHAY0yntB8y4ztUYjYlRHYBwG+rIwz2eSKmMZKGXAcNSEUqA1Q+VqBeokh9UMoeN9Zb68MA4TxhgyCItrvgKvmWHb9d3W/xseO47sOLLjyI4j3zJ2DNkxZMeQt4Uhb/rw41bxUmju1DxSipKLMj0+0ms4i1xqwfOIp4SUDOaI90hfNsFFyXkkDyMpF7LESdmqYe0VQLr1G8V8o5y36kZoVIPSzDk2eO9BC768vCAaDh+qiqjw+O6RZZ6ZLzP//OcfAaEbpNvrGWrt68IAiJLEtwa2WPyxgTO99wj9MonFp5ulImFJqX4Fmw04rdOtY73dTtsppwiSEqLZ8hfvLbFvtooTGonDvs1F641mTskDKWdCruuUUiJ9GGccB9ydkpVujZyUlADv5BzNhr01Wq/h4Y4zlEL3NTwuAQTW3jENWvwyzyzVMBJmHe1Aio1Uckalbg2aftsUV8rZX1/yJguIS/2a5ueWsFxKNEVeN5i5BRpfqzVJcY/wrWvgFuZcdcrutlHXwvEwMeaYP3FnyBNjDnA5TgMXM3qvDDmTSyFJfI6+pUuzSRq82w0Aa11o3RnGcUt/HgNTk+LKZl8ajja1NVo3uvlWYQu3GxXAOst8YXwwUEclbeC86bs3QP06abrk8v9sA///ZOw4suPIjiM7jnzL2DFkx5AdQ94Whrzpw49pxnPGk2PVsZTwVCAV3Fd6rYg6Ij1OqR4UornjKGiKjaQpmu+sxcISp3VuVZagkvW2+MJxI9G9bqm7G8WcC+N2w2ttrK3zcH/Ph199B2yL3SHlwvuPH3AX/vCHf+bJn2lG+PITS9iIhq64UL/tDLfOdvwmmO5YgGaOiuMalSInKN/e2nbSv26kaIyzr8DNgVwGhnHEWqOtkSNgAi5K96C2RZVSBoYBsjipZMYxMc9zZAukzNobSQPcVB3UQXpQ+taBEhcikXJsftVwRsVAYKPlO0mdrIKmHJaSDqdqnFbnZYHzUllbbGLrjguoGiUpx2mkm/FyqfQruHwFJkE7+1cFrb+uGFxpZ4iNVsorMIVri2Nh4oIQeuneKtcazxXIxI08ZoYsjENUnFJSeq0kHTiMA2pzuPRUI6vivWEKtUL1zVq0d6ZhpHfbGi4dzGnW0JQZs8aHcSOnyJxw2FxXDFxoFvdSVZBNU2wW96I3cO2s8zkqX7qClK1y12+VpfiiK9s91jfdqAw7juw4suPIjiPfNnYM2TFkx5C3hSFv+vAjaULShJqRU0MlToJh6JdQ3RwsNFE0LCh7i2V9DXeSK3XcV2p3vOdNPxjOEQ5k8m2x3hbctmGvjYa/qE/YZmcowvvHd3x4955rc6KIkHLheH+HSuL9h4/MSzSC1bViuZCA5B2xrWLSw+7Se4vT8VZtaW3dHEGuLi+GO4iEllY1KEHv4eIR1Z4Ar1iMoQ82B0mZYRxpy0KrK8jXlYh4bMqZw3Tg7qAUbww5cbgbN7BLjOPE5XLhMI50a5QMSJzozcIdxcmkFNWV2mpUIWL3U0qiDDl+tmUYiAudoH0Xc+YufDmvPM3OWo1uglyrD6Ykd7LAWDJ1GJlXo7fO14By/VM0Vv5lU6BsmtstyM3lBsxXjWnv7fYa7k7erD3NNBryLO5DzmE1KQiaNqARIrhuc7lZ14WS4v6NRbBWySmHQ4tEZSzuvyElrCKdoLKvDajuHo2VqWA9mjBNobZG3fTXZmzPM65WpC5CNw+9rhvWHcxY5wttXelZQmrgjlu/NcUKYXN5pb1PLy//Ibbz/2djx5EdR3Yc2XHkW8aOITuG7BjytjDkbR9+cPCOeCeLkYmAKyQhaSSJkSUs+MQ6sjlKCASttuket+IA1py5VkLH6nRrbPf0dnIGNg0lN23pdeIdZ1nmzcP++howX+bb5rVtcfRmlFL4+PEjT09PrGtDVFFPuNqtihKb1HD+Wtvb7WqRIYgK9I3S3DZG2H44bK4tcfIPQAbZKOGwShRxcgmqWKTedL3X94pFF4tWNTEmZciZkgvHwx2qiZwz5/OJJEptC4cxXD4CfLfN46+BVmtdcaKK061TSmYYBtZludGZbnCuMyaZpcHchJdL5/nc4rOL070xbGagbn2zRhSyyjYX8BdfCfHXvy6w3FaWiKDi4K+A1FonJY2KmabbPOeUQidsxrqwVcleGxOTx1qNTQ34lootwnw+cT6f+PTuSNJI0EYS0zRQ2xrrdLsOzFiWmZwySYVpHBhKYqktgMEM1YjCs5SiYtaichb3PDSzKWlU3Nxxi2+WUgYGd1YRlnkhsh3k9qX6l/OjGk2juN88+9/q2HFkx5EdR3Yc+ZaxY8iOITuGvC0MedOHH7cG3hA6KW0LhKCgASQZSRe8d5qHNWASR22zV4ydgBB+6qab/3y30OLSt43dYJt8iAWV9LWhTK6vs9kIYpupoYB3Y5lnlmUll9DbuioqCU2dT9994un5mfNlIV8uuEOXjuhrmvN1Z6SSabXGh3dwCRcRESFJ0H+iKfzUJaov8bnjhbo7ZgGMsd80HGK2jZforKWguoSDzAbP8Vqd2sK+UkXCtQOBbgx5QFNGJU7hAEkF25oHW6sswFXru9aVZV3CmnCrBlw1qL13vHeqGeu60rpyXlaqdBYKP58qT5fOabGoEgXyB63ugpoTzYaZcSiUvMaGdL9hjPPaWHiton39++sGk9vfbaugpFSo60LJ163z6shybbpzdzp+c9lJvSMeulhSgCTueGuUIWw07W5AxqB/17pwf39ApdBbNH2WUrY1b6jC/fFA21xdulnQ6RL3VN2Dps+J1CInQiRh29y4X7MWwDw+c86ZUZS1EtdXMuVw4LTEtamEpvtKN0cjbFxTe8MWtbDjyI4jO47sOPJtY8eQHUN2DHlbGPKmDz+h/Ry2AKTtZgu4Bp0mCMniZLmqUERRi+rDtFUzOrd+Q5wI7jIBdSNJgBQim9Zwc08hHptyihRhicWrKVFS5qrPrK2BR3Xmcr4wTAMpJVQ8fs/Ku/fv+PDpI3/68Sfki/5isbtvzWwSqb6yhYmJRiOkC7dFHXRmIqXQUQa3GBRr4OF1kbCBrH5VbYhrSnlgGEZSutwo9GZs1aO00a5BPW9QhLWOkqnLQkqJwzhtGyrFok9Cb53LpfL48EjOmaWurOtKN2Ne59uC772zLDPeGuu6Ms8zzUbOtbLQOVnn88vMaTWqpbBQ1AjGimuGjKFJUYkNF82GawCObzD7VZXl/wx0rj+/Noa6X11rQn5Q1zUqdhpWo+vaN7o3/+I1au+kUiiSEAzFQyJthltHMD68f0dOyrouvDsc6LWyXM7U48Q4DKSSI3RNhTKMzOeZvtlYtlqpdY0GyxTNrW6CuZM2a1BV+arZka0qFI2fKWcUwUxpS4/7a0arLTIgjkeWfonXk9dQvev6Dmr6Lysxb2/sOLLjyI4jO458y9gxZMeQHUPeFoa86cMPoogWkpaobLgh0sF7LKorPWqKpvAsR0A0GghFwr++907f/i5uaChxEdGgfEUAI/m2mGR7702bGjaPFq4l1rBrM5cZQXuXbdE4Mo6b60dHJTMMIw9394xDCX/8DWBcBCOaBTXrVnnpW+XjdfMEAMjNNhMnHFYMXMKH/Xbhv9xpIK80YlSHwmazlIL3HgtWCHrYHSfobTNDskRVSkA09KIiEnaGKZPTQEnRx1mp4CspJ0y4ZRl0swD/acLdaesc9pWSmasxVzh152KZc3d+PF04zwu9R8qwSoCeyrV6YttmiysS75QUxSkzj/m4XitblYyvp+SVIo4N6jfqHqDWGhpb9/DRz4lNZXB7/tcAdv2iklxIdiFr3MssnaJQivDpwwPaL4xq9HpBRTiMA3jQyuMwxsYuQk6JtVaEkCzM60prFRMloajm0M6uM7XbZo4aeQlgUZFUxT1jJmSN6lxtztPLKdxrPNZha+GaoyJb0+imLQ+xQewt4t9zebsuTcCOIzuO7Diy48i3jR1DdgzZMeRNYcibPvzEqT70lr550wNID1rQLdJ8Ox1tsUnco7Ky2ub64bGoNBArqhRbc+G1CU8cVJRUNpoNgrYuA1oOSA56t7cZaZ0uCZOwazQUNLznp1LQkoml0HArqGrYDZaMe2iCHcElIWmjeS02P27RSGaN3i1oaFfMFZHEFTq6x69wannVSuptI8Xj3B3HbnaS5oKXkcPdAyrCvF5Ca0xohoNKTkxjIRPA7gBu5Ew087khnhk2u09VSJLIg2Cs9HqlS5VpHLG24mZYr1sIm9O88zw7tSU+95GneeG0NJ4vnaVu180GLFvOAIRmOgBWUA3AHzMch8TZjRrluevlx29fAwSvoHNbY3p9sNxAXjddrSh4twA0j9lmm+PeO0kzrRtVMw8JkgbQjtnJNnN/nDgUp6hzHKJ9cBoKh2nk7v6ey+WyNc4qvTm9EzpcQFIC1QCFpLRmrDUqI7XbJinYmghrhIxdG1JFNebQ4kvHum21oFj1bV04nc7osYWmV8CuFb6t3NZ6fIGb/GK63uTYcYQdR3Yc2XHkG8aOIewYsmPIm8KQN334gfAgd3dE44wdKtJOqzWqHRb+5eYOPW44ogh6o01TKqTNAcU2zSR4+M3r9b2up3AliWxgBMu60i6NulbEGochaFMTRbzjprRaUVGGcYKc6N5JRJVknVccYxgDfHxtuIUjiqS4k836tpNSVFJ66CSvi8oN8gZqTlgemrNRgnbTiH5tD/jKGet2eje6CJSBfK+kUvAnWNYnskMRZ1BnypC1Mw6C92jErLWRkjIOQ1Q/MnRbKJQA2ilRu7KuM2JG1kJ3GKaJ8/nCaa6s60JTZe1wWS+8LFFxOpvzx5+euKyN5kT1SbcmSbMNIPSGI7FnfKsyODllpgmaQ1varXHSxW+gA39NM8c8vVZkrvSqu0P2AEvvt+deadic001z61HyihdQ3bpZlWFUtIV3fmuVw1AYDoVxGPDemI4HpmliGAZajfe4XC7UdQ1971aNK6XE5xZlnhfq2mgeidyaCng4qqxrw3FyLlyTosMVB1ptLGsn54SLULtQa2VZFh7FSamgW3WsbDrx3hrLMnPNmNA3/D8tMXYc2XFkx5EdR75l7BiyY8iOIW8JQ/T/+iG/HCLyX4nIfy8i/ygiLiL/9V/8+3+7/fzrX//DXzzmo4j8dyLyJCKfReS/EZH7/7ufBYtKSSQEN3zTD0blxII2lmh4S5qQ64aTzYM8FLqopm0zaricbO4R0zRxPB45HA7kvE229VtTWa2V8+nE8/Mz83yJCohvNK8Zbo5Y1CrO55l1bZhdF7cQDi8BZMMw3A6tOSVUUkhlPfzzRZVQw+qtCuRoNBVuoBk2jo6mgkrervH1OHyjld3jOrcwMUcCpFBME6RMGkeGaaLkgiCh88yKqqE0xFewlUEKyWFUZSrKkIyiFWWFtuLW6MvK8jJDdTL/B3vv9mvblp13/VrrfYwx51y3vfe51KlT91S5fEnsxPEFGRQnQkJY/hMAOeINHuCJFyRQ3kBCKLzwBFIQQkAecHAwwjHgJKIiLCM7iWPHLmKX6+JTp86ps6/rMucco18aD62PsdY+VRauOlVQKzW6tPbaa17GHKNfvjH797X2tcLJtuf8dOcxq9UYi3F1zDzbJ64n4cjAdQ48vSk8vTpwMxbG7FaSc+Kj2ZwqaUjrR7f59GuUNj8E6GKg62IrWMbt8y/NSe5IzDSGav59O04iHjecc3YJWF4ONY0x0nWdx9xieJi2s3wiztRsNwOnZydAYy9E2Aw+1+bPq7USY2SzHRiGntk+dJ6f0j7LK2Yb/dCx2QxsG1CJSHPkcUccJxGdSSrFb8BznYnazuEuuN7th1wyJSdKztRSlqJi4G4tMXxrMPI9hSGw4siKI6w4suLI+l1kxZAVQ75/MOTbUX5OgN8G/gbwt/6E1/wd4N+88/f4vuf/W+DDwL+CV5r6r4D/AvjXvpUTMSu+9KR5h1ttC81lMO8z7zgNAdXIvADLnDTo/dbe74liMToAbbdb3/HmTM5eidiBQpr46NyO0IothYBZq4CLUUumWqDYxHvvvcfuZMujYYOVgjZ50ivgdpzsdsQQ6DQsVoGCF+uapeU5xha8gq9IaK4xIBrQAJY8WWwG1NmpxRkIv/YZWB2Xa+vL9iPq8qEacXvCSa2k45GeTOxdzi6lUKRCMbJUun6g7yOqkCyBeEwqGGUyxGCaEtvtltDBmI9UIlf7PaMp+6xc58hVNrIJ2ZTHlyOHQ2EqiWwRiQ6oVssiNXuxOB/fWyCRW9ARdzMJKgxdZIyFYq1a9e0/y/XPzQHCGlsyP39nESKM49huTOK1EEJz1+HWDSWnBPNxlnOC7dBz1gekJqZxJJ6cNXlc6fsOK5XSCoeFJmt3XUcphW4YiNETD2mS9pQSiKLBx9zESGP2+SpeywFxqXia3A0lBJ9nDjwz8yQIsoBmrZVSJ6iZjKClEO5IziJ+nPCtLFpv3zMYAiuOrDiy4siKI+t3kRVDVgz5fsKQb3nzY2a/AvzKPDB/QhvN7J1v9oSI/DDwc8BPmdlvtsf+HeB/EZF/z8ze/tOfi59DiIFaPLY0pdE7QvA411KYpcAQgjMLZpRUm0e+NJakYlaInQ9uCF5peZwmX2B1Pk7wxbxo0J6oiHlKoIQOEMiJPCVyGRkzHFLi5OyM3dkZu22k1uJxtCrEGLg4PyOqgLntX1R3iCnmDI6GiEYhTVNLGYPQdYTo1XNFfPJriBwPewRfGJ6IeeuyAj63Sp0XlIOutEQ1Bx6hCMTtjovdlpxG6v6KkT0v9iMRJeMVqItec7LbEsSrOI/J3BoyFbRmDuNISomLizPGPDFNgcv9JZnAmISDDTzPwmU23nm25/L6kmTCWAKzMNl1cWFDuuAysaFYaZWhZ/l4vrpqDphBPZkUTwgdcyWV0mT4WZV+mU2Z5XhPWCwLoL3EVMmc0FmRGqhURAKh88rbs5Rvjd2oJVEkE6Mn/D08P+NiyJQ8crx+Ts4DpfRcXj5n00efT9WYJpdzi8EwDA4Y48hSxXg530AuftMtVpiyy8tTLm1uz04xzrB0Xec3HMCqoFqpyci5kEtoVbv9xoIJeqefQvCb8TiOuG2lLnHaf/p1+72DIX4+K46sOLLiyIoj63eRFUNWDPl+wZDvVs7PXxGRrwPPgL8L/Adm9qQ99zPA8xlsWvvf8TX0LwD/4/sPJiIDMNx56AxuY0mNJpUplOS7cFOXk0tNTZZzh4jSJMRqZRkIGutQSgFxT/2UddmBzwvVWRhfCGZuHzhXvLVaKVIR3bTBLbPeiVAZc+I4JZd0q3vzu4tKbceOxObX7mBxO8lj2/nWUrhbXMwXWWMbVKGU5XpKS+RT0floLy0aaBK23FonitMtaAh4sbEM4jHBuqleeM2E61wp2ehSYnsC1zfXlDTQdz01GWUq5AzHcqBiFINMoIjx+EY45o4xG4nI48s9N8fMs8sbLg+ZQyogocm3lRhcAp3PWdztc3H7MDxGWEyZrSBfYlPs1hXELRLbDUKsMWZ8wzh73+rCRN29sc4sg5lhdwC8HYE5lMFuRxEV6KJXkzYqQ68MXaCKorsNuYUu9CGgCPtxbKEDnhg4jkePBQ+BQT2ZthQPdRiniWkqxM6dfsrUHkuZnKtbnIq7pHiMbaDr3OWn5EStXvBtSoVxKmTpSTm5I1BjKKPgIRMixK5nDlUAWeTq70L7jmIIrDiy4siKIyuOrN9FVgxZMWTFkO/O5ufv4BL0F4FPA/8R8Csi8jNmVoA3gK/ffYOZZRF52p77Zu3fB/7a+x80zAtT4QvF2kY7amiLRqkkn6BmqNTlnYKhAUKzH3T5NVBKYhxLYy/Ckkh1d9L5hKtuwVh8QaLqx28WhtDiIFUpdeRwGHlxfcPzyyt224fuBlNf3sGfnZ3y4sXlcm6R6Au2AVtt9pX+cdreZ4j4eTpg+iQtxXfaoZ0/4taZs13ibUzpnFCnUCuKIBLdPQRxFqtCtYDqlptaKWOmz4nTTQdVyNPIVIxNqUhLbDMil2OiIkypMmnGRLg8GBllTIXDuOedJ1dkM8YxuQOK+4Ki0mJQUY8sFmnFzPDkSxFyyRjaYk29zYXeRP3GooI75lglCPRBmayQhcVTpN0ZFlBbbhQv9ZM/5k/e+TzVZcHVavQxoO2VnsoqREDm0AGEk83A+aaQJ2PUiponag5xYLb8nKbJ5522ZMqaKeKONiEGzOyWdaHFdgePvbZlDKzJz87qqcwuMbfzLoSISG2hFIap33xTmhzwWghFtWavWQsYXsSuMVy1Vr7D7buBIbDiyIojK46sOLJ+F1kxZMWQ73sM+Y5vfszsb97583dE5J8AXwD+CvBr3+Zh/2Pgr9/5+wx4y+MRPekJQCU6cIQeVDFRusEncGjJXotAbCydL0JLlFJSwmVgUUKIbcBYXDNUfcdeKw4M6jtzlYpqaPG1BaPSxYjVQkmJr7/3BPodJydnvPrgjK4LpHR0v3VVuq7j0aOHPHn6jMvra8wUw4s3pTyRUyZlnwS+24WaZ9aEZWeec27Fwfz/KSVSnm6BE0DmJeZAO/M7ErxSsgaPR84pU8ZMKVBr4CpHbkqmjx0b7Xh8hLAvRITtIJyddPSdccyJXArPD8LNPvH8xQ3H4zXb7Y4sbnSTc8EQiuH2hqa3y9gMpS6JdVYa4yFGyT5eqKBBPWGuOdLMSX+lFOZCc1gzThTootCbS/OW2gKrjambgUVwpoy7LMs8T25Begb8UtydpJRMKYVNDFgralYSxDvVvq0ULCiPLk55/VTJ04HL68jN9SW51FbTYGLY7djvD5yfnzOO43IuIQSu9zeUfcWAlDNmUIqQSyKXiSkXUvK4bKS5sLRq1SV79fFpGskVak5UMmhj+oISw5YSBrrGqoQQlvmScmKcRjCWG1mtxjh9Zyuzf5cwBFYcWXFkxZEVR9bvIiuGrBjyfY8h33WrazP7IxF5DHwGB5x3gNfvvkZEIvCoPffNjjFyJ1HxJUkQ31EXaxJuk8dCc03RocesIOaWjNB2hzInf0HKtsThukzosacmgDqrc3dH6Tg1+50Ac7wqM48D4FKymRehGseJcSxcXh8YxwmzjiklqoC0KssPHj7k5OyU6/0NY0rkYg4YyS00mWVh8x1vqZlAJOfMsNGlT1SFUlhYiJw9brSU4rmH1Zw3aAtoYRkozZ3GJ3JtXvkqYBrRLkCoJMscjgeur6959vQ5UQqnm54HD07pO2F/PGAiPL8aEYnkpOSxMirEHmeOAGlgrqLUlJEm1XqNhILgIAMssrDIfJ2eJOkhrYaXGptff+tQcnfQgihdEIpWihTqfOzGdi1ytnDn8dtDzDedOdbUGa1yx+nEP8eqs2UiHhccVQFflB676uemNbLbnVDNb0rD9oQyCsfDka7ryC3R0G0m5/nn7EsqhZxL649ArpUxJUr1+VrafNXg7I8nwjYpPAiYUapRzV1uajVU/TjaBbqub/UDlFpctrZSF/tV6bo7oJu/2bL9jrXvBIa046w4suLIiiNtXq04sn4XWTFkxZDvVwz5rm9+ROSjwCvA19pDvw48EJGfMLPfao/9y/ja/Y1v5didKl10OY5UqcjCKqgqEgOxD5SiSC6YVGo1L2RVZ8/5u5OtDU2TKz1O1WW9XAuluq1eDLUNtLVJCDS5e97Ba2kTpHrCYNCeasrz55dcXd3w8OFDhthhNTcJuXL+4JzXX3+V55fPuTpckdLENOZF8nPv84RVjzAt7TPm61BxWbBQiRoodisjz5Vyrb6fSfDrBBDUJ2GtSz9oDIgJioNxtkLOR57fjLzz7hMur2+IovQx8fimEmLHOE1UhL4MgZJeAAAgAElEQVSLxC5CznQhEIP73edSkBAcms0XgzNXt3HGzgq1cVwAscnmzczd8FhmdYrEj/I+eVhVUDMqLlMHdcZMVNAqmOKLdAab5Z+X5vDy/JyoZ/Pnmce79lEJd2KaBQe3EG7jdasZXYhMaaKUwDBERjNOz84Yx9H7bLMjTD5W0zTRdR191xOjsyuqHgfdtTjXUioaAlHF52guCMpms6G086zNkcarhEuLcDByTnhBPjATdwYSryIuGsi5UOuBmhJT9SREP16hFsVCcBtNeV+HfYfbdxNDYMWRFUdWHFlxZP0usmLIiiHfTxjyLW9+xD3wP3PnoU+JyF8Anrafvwb8Is6cfBr4T4A/BH61Ddjvi3vt/5ci8m/h9pL/OfA37Vt0aRIMNffSt6jesVYaO+HwoarkUiiHI2Zl2SXXnJcEP9WOoNH97PFdJ8wFxryZ+U6z5Ey+O6FFUAmONwilCNKSEEs1pDkLxn7DcSx87e33+L+ON3z2s5/l4XnHw0cX7k4SAienOz704Td45+vv8vT5FV0XUQ1YG+iMx22WNtFjiFANlcB0TPR9T9/3jIeKRIVayHm6wyL4UhZ4aVHeXWyCUOdiWAJFnD9yD38lRGejYhjAIsmgmGK2wfIGlS1F4Hg8cKaBjsJgxjYWeo7kvIEKVjLShcaiCMocW1tBq99E1Nzv32zBADH/8WXdFrgZUtzhxpDGYDnYyAw0zQFGzP8fRDG1mcJh9opv2LXcdGae525/ieHVlENAQiDlkZwDaMCqourn28dAmOVeChUhpexMWSlUCTx85VWmlJimielwze5sx8lJ4fnz54vUayZMaQI8vjWV2hidiATDWuhDNJ+/Nd+yfinlltSqhCBY8WuotaAhkCteA0IVSrOrzKUVD5s83tcMiXFJcp1tMENQSk4cDodvZdl+T2EIrDiy4siKIyuOrN9FVgxZMeT7CUO+HeXnJ4G/d+fvOf71vwb+beDHgL8KPADeBv5X4D9scvHc/nUcZH4Nz8f6ReDf/VZPpFp1JxEaG2EA0iQ6UKuU2iosTyMhgFnwDsZ36e4UMUtzhWVjP685aU4k2i8Vc+fCYoK7aagGzNzIcPZRd0nSE/aMQAg9IfRUq+RSuLy64vz0IeNxZOh9GK6vr8kp8cqjR3zpK287cFahC74DnkaXKz0hUFw2niXuWtyPX4XbBLRbtmWW1a0xHHfbLNkvzIe63CuAhlnCVahGKpU8ZWo1Si6YCBp7l/Rjj/QDwYxA5erwgvNtYAgQo+IWnEqQQK4JqhMnIizxz67i+zUpBUxBtIGM3erAOKDMVpJiOKshYbm8BSDUCEiryuzsmEg7Hrfg6r3wsvz8/n6CFtJw5zX+o7fnbiAtSZHGTlgDsmzw+Nklr5w+RM0YCPTRk1j3hxvGKS+s036/B+DkRMk5tdhZXc5ncZkxY0qZaSqkVJiykas4s2heUKyUQm+RahUTCEGRvqeMySXxGCkt6XWaRqY0EYvHeYsUpLab6h120uN5C9M0/Ylr9E9o3zMYAiuOrDiy4siKI+t3kRVDVgz5fsKQb6fOz9+/HaVv2v7VP8UxnvJtFCP8huNopKDU5BaMGtog1MlncKlgSk0TpRwxm5MMhRg7tMVhppKpNWNWCE3erLg8KRp8cMwn+NzhOWcwAY1UCVRzNxKxaaEEAoUiwoFA7TZkDUi/JVum5Im9Rh72PakUujDQS8cWOB82xNBjnVsiHo8j2QIxRMZSCCEy9JHpcNOuplLrRNW2glUwjdg0OROBJ4WFocOIHj97Oxa+EqpR8N262ESUFoOZPJZTpDClTJoyaRwZx4zGjm3eELszNJ6QiUQ9A5voojGlkbDdeLyrBrJVEpN7sYs0KRw60RZz64+bgVgAiVDzwnx4rLGAVHcxKWkJEbAQFtSaJ6dVP36YZVmMUI2qLT5Xbq1JZxtKGsMxL6pZhp+Pa9UWhxMzQ8wI2rltqcCEMUjABHoRsAK9UMcAqhRVfv0Le7rT1/jEa8qFKRARNY59YAyRWBJBhKhCTRPT4YZcKmnKSOxJpTKNiRg7+hhJpRJFqGKkmlrQuYc9qCrD0CNi9EHpgt8cSzKOoafGiB0zHAVSQizSbfGCYmZQCwTDVCm1eihCynR9A+TG4H1L6/Z7CENgxZEVR1YcWXHkg7UVQ1YMWTHkfmHIdz3n57vdYoyEGImxX+z2nA3Jzqw0gAhdR5omQDzWEqDtVA1dJEcJ7p/vMrZXUZ53qyIQohJjxOMNPWaUUhvrIoQY6WIkqpGnAzVXppTphh2lxfEeD3u++vY7DGenPBwCVy8u2Z2ecNjvuT5cY2Ref+0BV5c3TOOeHCNCTymJs9MLRD35UII7e1h1iTbEjBAQCQQzpIuUeptImHNyoPTMvPY+o+ZCKdltEoMQQ0S1a3GpWww4HI48f37N9dUl+/01tWSmnIjdQLfZUaRn6LZU2uSviYcnA2fbgWilcRiBOU5VWlJnqe4CMzMWqq1/xQtyxXAbe6s0psO70W8M1WOnvdSyLOzSDBoqgCq5VKi1yfpGFyNVhJQLycrcJcyL526YwQw6ilD1NsbWT+zWtnEu1DaDY4jR3WOqQIzkWqDAu89e8Bu/8/vsfuoH+fTHL7yCMoFTraTsVqLDyRnZjMN+T61jSyqshNqkb8ydW2xm/to1dQYUIoFBXKY3c6nZKO7kopFsmf1hZJ8mZ6hqR9/3pAJIJQSlj7F5+hdSue1bESGltMSzn5zsvitr+//LtuLIiiMrjqw48kHaiiErhqwYcn8w5F5vfpZiXI0FuZs06LJgpRSf6qLaKsl6p1UTLDcw0llyZulQVxVbZd+mXZZasOzJiSml5qmuhNA5E4OBu+H7ImgxotkgxI7sIZZo7KgG11fXXJ9teH55hXY92kdOz04xCqe7LePhhjLiDhsWnQ0KgdoYiJwzRgWDoK1IQFsI2q612m1CZW0JhLW6a4pfK2RrixZPRJPWB6UYKXls5f4wcnl1w35/YJoSKhUJSgyd+8iHiPucALUgJXHSR/om7S6kTvUKxCa1xTQr1ipWA0tfz9J3KWVxL1mkY7nzwvc1Z1+sLfz5mNXZlPm5WVK22+RMk1sXnfn3N0uec9m9SfLtwpo43a7RfyP+fmdwPDFSTNDQUag8vZn4wluP+ekf/RHvi6K4QF+oIp4wGzoQITcmqJpL3F1QP6YVxuOR41SxZofad+6KMhXc4Qd321EBQYlt/huGKo2tgX4YCDUwJa8uXktuTFUAqW0cnLkBYZomn0f6cm2D+9hWHFlx5P1txZEVR76VtmLIiiHvbyuGfG9jyL3e/PR9T4wdxcoS+zrLgzPDMPukgxK0a+yK3FbMDUof+sUpopTSYm5bvOwyQX0ylVLdiaIV8VJtRbkEpBpWCskqxdpCE5cS6yytUinmQNR3PSVXHj95RjfsOD/rqRq4ePCAVw6JUhNaIe8zlJasJoKZvBRjKSrUJvmJiLtsNMCdWQegvX9DmiaP/TSfXCItIU8CFSGn4k4qqTBOmZQr1zcH9sdMKkIhYBLatQdSKWjwqswilZKPiGX66FKrS/e++Eqt7royA4vIAnC3ca4+RvMYOkC2kWh9DuIsBiyANo+RVbtb+wszB9egkIvhhcr8CVmYkxlAluFe2ITlPLk9n5eAXX1hi77s2kM7HOKAo3SEECk2cTPBl772nMurA9vO318JxK4nTePyPpFIDH5zTakssa16p29qKaBQKZ6cyQywQi6N3QpOJxW81kBQDx8Qgc3QMR4mJGxQMaQaaZpIeaRQwDJWhX4Y6PoegHEcl+v0pNz721YcWXHEj7HiyIoj315bMWTFED/GiiH3BUPu9eYHnzqUbIsPea2VzWYDcDvw0BLtIOfM7JHuzdouvP1VjZySzxGdFwCL3Z9RURVi1wEsPusCSBByqe7CUg0NHfvJ2I+VUn1iBgKhi2yGwNl2x6uvf4jnNxP//S/+MkriEx95jc988k26ruPVV1/lfHtC/dozxuzxp4fDTdv9JmLnkneM7q9PS3gUy9RSXYbG40ZjDIg4kxBjRAmYFXI2ajaqOGsQJBJ6n0AaMqkUpuPE9XHkMFU07IjdObWCUYA9ipHHa+CIpSO9Fi52gV7yvHa9D1vhL++3DqjUUpcEw1k6Bvffj123jI2zMvWlWOcQtF0Tc4Fqby+NrYNFbXI8czExc2eeoK0gmXlRtRZ5+xLYLYdVT2BU0Va9uQFe+6zbBM8MofP+bO49c7VkqZWCsmfLV59Xfvf//gI/9Kk32e22VItQC5vNhiDGUZWowtXzp5RpJMaOB+fnVDOOxyPH48im7+l6xaspeycUq6h6RWu/QRoxKkEENa/KLKHSF68BUCp0UZlyJqdErnC4viQfJ0oURI0udpyfn7scndICuiLC8Xj8U67X79W24siKIyuOrDjyQdqKISuGrBhynzDkXm9+lkSvO4P//ufgdtdcyq3cGENsUjGLdCwiUGcHCU9QnCVon0juWuJxuHd29jqfA0CT6qonL+6PmTFll4ENrDlvaBfp+0hOlc1myxe/9BZDr3QCDy8u+OTHP8p42LPdnfLKa5F33nuKFbzwVzX3zicRluJZLZa2+fRbrfS9+7ibFWquBO3oo19PtYJXJC7NvSO7Z3vXNTm7uu1hThzGI2NKxGFHCAMmimUQS1i9ISjUnJimG6JUdtsNmz5gFpgBW1RAPB4ZZhlXPA6UOxjRnqtWX/o7qDr/Ze4VL+KMydzvPkx3EwNbsuJy3DYfml1mwGseiCroLbCU2anmzgk1LsY/yzz51Mxvd+aHXq5H2ufU9pgGnyshuMd/FGW0Si3GJMrTF5dcXp8BzqhgLt9utzuCVaJUjteBWluiJQXV4Dai48g4TkAgdh3DpneZOU2knJxmCkrJYKUQYkfsAqKCFI85LrUQJDDWglU/zzolShoJAqHvMHEmLYTQ7C6Noe+Z3V6+DZem76m24siKIyuOrDjyQdqKISuGrBhyvzBE/99f8r3bZhCYGYfQCh3Nj7kM5s4dJSeMgjTRE6rLalLBihf4yhNQCC3nMJdbCXv+mf9ezqF6tVnmz2u6ZzVjzJn9YXImQG6l01IyQZWT0y2lJq5urtkfjqQMT55d8aUvv8325IJXX/sw/XbD+cUZn/70p9luvEK0WaEL7pU+f95ms1nOK7aMuTQdUarvskXbe9vCKsUn7HFs1ZObHBvUnSRrYZoSNzcj1/vJkxVDT5WAERHtQJzBCVKIJDpGzrfKxg1V0NhTRZEQvb8VhNswAJgXqi3j6Y8LwzC8xIKlUhY2Rpq0OY/JLZti7f/uDX+XVavFmRadJeUmNVtOCAVanQVBvDBzvb0ZzWcxx5XO1ZrvnN7t55gt8cxzDKq2eSECYgUNgCgmgXffe8bxeGQcR47HIyJK3/cEga7vlnoJu+3A2dkpQZScvfCYiBcQG4YBVQflLvoa2G4GtkPvMbmCsz00YDXIxYuUeZ/4xcQY6YLSd4Hx+kDfdXRdz2YYFgZznjv/PLUVR1YcWXFkxZEP0lYMWTFkxZD7hSH3WvlJOS1xkHM86d0fmknkEn+LLK8vDUxEhK7vnY3JiTSNPhERaAXDfMSUruuptSw77xgjKopI24WKEYIxDBskVS4PE4exINohAl0Uaoahi8QusNsN/NIv/U88eX7DX/7Zv8TnPvcPuHryhK9+5S0+8tGP8yM//AN87ON/hqeXV4Rh4JUHZ/zGr/+fqE0cj3uCVVLKTKWy2e7o+4Hj8cA0eRkDDcrV1SUAwzDQ9z01uSSfUvLr18CwPfE44xAoBscxcXM48vj5gaeXmcIO7SKhudgYhRDcZvFi+4h8vMa0oJuB7eCyagwKGqgGmeKOLwDiscouFTfguQM+IUQftfcBCYC0m4rHmVoDIJb31lrR6NWfZ/kXqwQViC2Br8XZBnGbyVwMqziLYhkEQhRn0RAaxbFI9hrcQ1/Vi6HVWula2MGMTGYG6ham4HGrXado12MlsRGlSoQKb733mC++8x4/dnZGTYlSPOnyZOjoh0g6TsTYMY1Hah2ZpkyuXrRuSl5DYrvdUq1wfX3pfRg7hmEDIkwvrrBciENPCM6KqRhD7KjViLuem+OB3aZnPyasZra6YVBD8oSqUKqzhy9evFjW3jAMpOlIrXavE5VhxZEVR1YcWXHkg7UVQ1YMWTHkfmHIvd78zG3evc8Ts9aKb5ntJQCapyHc7hhdxgweewjklCk5LzaTy/G4ndS12mIzeevaIkuspy8cLwbmLhctgrN6KlkMysOLc3YnG7761pfYnT6iTgewRK0jFiK/9/l/xiuvPEQodP2G4/FIyhOvvfqIr3/tSFFhTJPHfLaExhAiXTeQwtFjfbPL5zlnxnH04lJ9v5xf7Pp2zi6TmgilVg5T5eo6sT9kknlRNW0LHakEjN1W2Q0bIoKUESN5DKdv3bGa0QbEmGClUGshaHhJVr7bZmlYGoM09+UtS1UagybOkC10hwODV1GebzSG2TwvXOJXcb96BHIt2J05YOaJd9XME0TnGGu7FUf9da27GgjdfW6+gJlliX1PVKGMGQ1KF5VaBIpbmpoYyZSbfWKfMhe7LZthR1QvIDelIyaBftj6jWQcGZMXHouxY5oypZaWGOs3kjnu26ondIqIx+2qustKcNYnxoBMCWlsXK2ZKEIfFJsyaTowHvYM50axAkWWY89yc0rJx7j75wJGVhxZcWTFkRVHPlBbMWTFkBVD7geG3Gu0mVkO/3+LM20S4zzpZrcVd0HxnbPAIhfmkj0m0Z/yuEsJINqYlFspFGY59FZ2tur2hDUEjz+tAlJxm8bgv2WOwc3E4P78jx6eEaMyHfeYCe88ecGD0y1fv35G33d86Stf5uvvvc2HXn3Ev/gzP8PZxQUPz8/42Mc+wuH6BdN4QwzClKpLyw1Mh2FgPEbSNKGxI3TGYZpIpdADpcnSXTfgmjLtHJUpF548f87X33vCzX6i1Eg3nHjRtCYVm2U2nfDobGDoImWq6LBhqhnyiDu2KNSKSXZZ30siU3JFO3PAEFvk+XAH2BfWDABFpLYE0ZnZaF77otSagZmxuZWtb28uM5PTQAAHG6sQxB10VDz44OV51SxIX8pcvL3xaAzLXBBxV57ZoQdkiUftuo6h6xhlJMZA1wWQnjgWilVSGqnSc5ONKfmNJW52nGw7pv2V36C6AroH7TmOBxA/r2maEDVOtlv6vmeahFLmegUOBjlXUiqUWsAqXezphx7DSLl6cbpcvXhdzXRdh2kkTAVy4ub6kt0bfhfR0C99AyxJvQgv3cjvY1txZMWRFUdWHPkgbcWQFUNWDLlfGHKvNz++u+wWJ5RZnsxlwtouebPZeCdJ8E4qdXkv4s4jOVdymdxvPgSG4JMnRLcDNCDnQs65JfX54vMdrDJ71FdxGbJmT/bLpXCcElU6xAoBo9PKa48u+NBrj9h0Pee7nqdPv84PfeqT/NzP/zz/2V//T3ny7AVTUs4vXqGLkR/6wR/k4uGr7DbCi/c2nG4iX/7CH/J7v/d5+hiQ0COhWxIia0uUPI4T4zSRcqXWwmEcCUEZU0L0QJoqYypcXd1wfb1nqh3EE0LcIcMJlEo2Q4KhVtlEeHT+gPNtZCMJDUoazrHdlmk65Xi45nj5jE4MqQWbDo358v4X9RjoTuMtQIjLtnNsNOYuNKrBKxVXgxmIGgiI3GFlmB1XfE7kXAhxtqBkYWSiuqNMqnUptuVOjLbEHkMDm3YwVfEbyJ02hyosr73zN+IStZkxDAOvvPIKUZUkkT5EiJW+D8QElo+M+UguR0aDp1d7Xn9d2cRTkir0ELuB3ckZZl4jIldjGo9M40gp7pQjBoIscnfOhWkqpOwVkBGh63rme9I4jR6KIAr01DyhUjjZ9VQTypg4GZTJCo/feYtXPvUZT0psfbIwdTE2e9fI8TB+kGX8/3tbcWTFkRVHVhz5IG3FkBVDVgy5Xxhyrzc/cwymS4q17YbnRCrv7LlzxgKQl/fOk6XMBafMQBWx2yS4arX5sM+Mycya+IT26sMuNWPutzFPfDOjZAedOGzJuDtKCIGL81O22y2lTPz0T/4Ux2nkL/zET/ChN97gR37ks/z9/+NzaN2QUlqSyIa+X87/4YMH9D/wA/zxH3+VMRUqEek6EHfXqLUiMWA5cTweeXH5nJwzx+MBVWGz2XF6/gDRyH4/cnW1Z0oZui3QkYpRWyVkw+sFbIfIq6+cso1K0IyaWyVqUMZUEY303YYpBMwqQkHMwcIMiLcL06wl/xke0zx3/91duy0vbkBT6UJgShNd11Ft9sRvwKO+2KX9ruI3BNG7gOE3Jh8HRbJ44mNVsNvEuTlBcJap7zZVxQSQW6/9u3GmC8DFyG63g1rpup0zGTKx2XYMWSD3lLohU4j5kjFlLq+vkXhBzIHtEBlU6aJxenZBGg+c2RnPphERIUZ3r5kmv6mE6PLvNLo9pKrHBFfzhNeSPRl22HTs4pbjODlTqEIfeqYCtRghyJKUOqUMUiiLo40sfdA3j/2+7+9I/vezrTiy4sg8liuOrDjy7bQVQ1YMmcdyxZD7gSH3evNTSmZ2TGnu4vjcDpTS7PWskHPBElhOUHPbqSul4haLrRvEBFEQPJZWimJSfZ5Wc+vJZpBXC4zFk+26rkMUYuipFaweSdPEVAIxbhENbFGwxK4z3njjAtEEQfjZn/t5ZocYJfCX/8rP8/l/9hYvXlzy4Vdeo+s2PHtxzdnFI07Ccx4bBOl59OA1Pv6pz/L02VP2+ytGUYq5w8bZycDV4QjhhDE94+bmmpInxmTkEnl+k+nHkTgoGraw2xKmwiSBAmizNQyWiUxse+H8NHASpiWpMusJ2timYYCcR2qZQDxms2EwpYqDeslAoDcgGxqdmfK1q4jMVo3NMpSKluIxytaYDzNnDgQwX2Du7w9QqU3m9gJptiwoXyfzHHEWwmymbNxlR6V5rLT6Bz7G7vsvM8AEASrSFiDctTV1kKrF4523uzOG00fOLnUdQ98xO8TkzgibC2rOXOwC080LnubE+X4D3chm84hpd8HmwQXa9cQHGTl5i+7ZV+huDpATXewZMayP1DJiySArEYUyQpmIw5YpGvt0wBL0MhBCJHTCZtsxjhktxe0htSNPGSpspSI897mfA5ktkybOup4OozecVboTW36f24ojK46sOLLiyAdpK4asGLJiyP3CkHu9+akGY0qowIOL82ZJCE+ePKUpyuyPHndbU15iKxeLRRMgUiyRS0EFokSvQMttYht4NducfXJ7spUzPDk3tkYDhEont6yLNabA3yNEiQwDPHr0gGHYYHVC9Y4XPPBnPvMpfuEXfoG33voqn/vc5/jKV77Cxz/xCU5OTvjCzRVpzNzc3LAben7iL/0sZdqTxiO/9Vu/w+GQuLy55jBOmBROzwJvf/XgyWAIXXT5FqlYDeQcsEla7HFAKSgJpdKr8uD8hEcXrxPEUK10MRBiRINSCpgoasLV5Q3TeCCPR+c/RKkloygxBiKCRz4LuWYHBqtE1D30a7m1fSwu6W+HDSEEUsmouow6S8q13MZPC3N/z8mI3o+znSPM0ru/p4sdJfmNKqi0KssONn5fsQUYVB3sFsCZf7j1/p+TW82gNgvMruv48Ic/zOnpKd4dEcwTJIMZaCT0AwMwDIGTsweUwyWP93tyKAyh8kY8IdgJD0/fIPTn7F77cWLIvPpjf8zx8Vu8+8XPc3r4OunqXZ5fXjJNE2mcmGpGut6tPZMxHUey+OhOVpkO13TJE1ONwGazIQ4bHj99gUgkqBC0YCmhITMeLtmebrgxOKaRDIw50bGj1Mr1fk8+3t/6HLDiyIojK46sOPLB2oohK4asGHK/MOReb360SXwuiSqgLjk3IDGMIB0WjGDQCRju9DFL1IjSEwiiuCsL1OK2fRJ9xnpCYfbYTHWwUXWOp9bqVWwRrGSKQKnZk8hyxhCoFQkRs0wMHV0M1JKWBVFKabaJPslPTk74zGc+zW635bd+6x/y7jtv89Yfv8ofvvWc3SBkDBN48xMfZ7M55fzsAZ/62FNubkbefvyMd56+YJwKGgpTKqi4LeZ2t8NkJNVWj9rcVcUXjxFqZtMLZ6fn7DZbhl4JasTgiYCiBgKlVnI1Zj14u+kJUhktc7h07/YYYnMwEazJvQouATNL9CDJbmNxmatf25K4d3fxQ4ut1tvqy8gd+0lpDjjmxb8WxxWZ40Ob84lBqhmtDjop31IGcpc9EJyxu8OqGLYkn85JhvNnq3i88MnJCScnJ0u1a79BdQugSfS6BKiQzGXp7vQR5Qaej4ZewSM5I7DlJsPupEdjZKqF7uITPLj4MKevfZL83pcp1084ffYuV1//KvsX73F1uGSfJq/qrYrQNXZIHZql+DgWv9aUJqoEdic79odEnmpL9lSCGJYOBLJLytPkhdfMGELAVChTvh2Le9pWHFlxZMWRFUc+SFsxZMWQFUPuF4bc682PJ6l54bCcM6MItYKqg42ZoV2H4n7pWIWacFdBIUrwnW+IGILVQk5HcslUA7WAysue/T5pfcKVWpaFghkqxSvYmk/6XDIiHaGPiMDQDTx4cEaMoYGYJ8/VORYVd0gJUfjYxz7JG2+8wcnJCb/8y79MrZV3X1TOhsqTq2uOoeeH/9yf5fXzc2JQPvqRD5OzkTTy/B/+E7bdhmMpmOwgVErZgyjD0GNjwSyjZGoVrExIMHabyIPTLbvdBhWjj0IfA1BQFWq9ZRbmlvNEyROlJgxriW8C1ciWAUOqAxtAFyNLQTBoEnuzgWyPiQi5FkqqS1/PSYPOeN2tpH0LKIC/3gyvTtyKeYmftwObEkNA1Y8TgiLFAdcrIsvtxd0Bspfn3V2geVl27rqON998k9dffx0hELuAdr2HIzSPeguRiqCxI+fMocDpZkuOI7lmujpweax05xtq7Mg6oVRElawuiRMAACAASURBVFQVZEP/4GP0Z68RzLi4fsI7v/sPqHki1wkJgomSp8w4GmJeUTx2Qj8Er3sQAuOYMdwtpVRFNdJ1Ri6FISjFMtP1c3avfxS2g0d5V7+5DpuOXHzd5Xses7LiyIojK46sOPJB2oohK4asGHK/MOReb37qHGIpcBz37A/VBzn2iAo5VzoNEJRIJKUjU5qYptIcIgL9MFCr7+6LZSqVENwRI2gEdTvBro9YZZE4q9XFZtCb+8l76lsHKZFKJVmlpmvOd1u60PHKwwsHPvPj5pSZKz+L2JKcNqUjJ6dbfvwv/nkuHpzxS7/0S3zlD75EOV7xlWd7+lc+zj/+w+c8OBlhvOZf+rFP0J8NXByV8wdv8umPf4Tf/N3fh/gAC9BtNhyPV4RQOTsZEIEQM7vdlouLB2y3kT52BO0opVKyL0LEMFM03LJRZtCZMxuEjsv9JdN4oExjAw6wYrRsPJdpW/yyjkZQQSUiiMdK1+qFuDS4+wmtqBiQUiLG6DcPVYLq4mJy29RDqs2Lg83SsIOMOCA0m8p2Oi1Wt4Kpu68UozBbX8pyI5hjFvzmNnv38z7AgRmSfuSH/xx/8Sd/eqlyfTgckG7D6dkpIkJq88LEbUy132JWmVDi2YfoYgC75MXlU07Oz7h4eE4wQcyTI007EoUchKrnmER6OaV7+EecP/5DdnbJi6s91+OeTqBS2KeKBCUXo5sMi4B6n+Ypk804TkbOxnaz4+RkQGJEJ+Grf/S7bPqO3Sd/lKEfFil9u90wpUTJkVFfBuT71lYcWXHE24ojK458e23FkBVDvK0Ycl8w5F5vfswK0nbJpWRy9qJL0liXWgrCQGgzwvC4ynHKVIwQuyV29m7T0Dzxgy6e6WZGbXG6vrN2945aa/MZr0SBoBGrgln2nTXuwCFUPvTG63ziEx+ntuQ5q7e7dqB9TmW323B+7nHDAD/6oz/Kdrvlf/5b/wO/9r/9KjkFxstLPv8HX+Xi7IRnX/sSf/7Pfhqmyj/6/B8gXc84GofDSIxKiFCmxDBEohqYcXrSsdvtGIae3dbBV9XZpRg7ZGiOKGIYRs6JcTpSSl2At9ZCrRlRZ6Es+nK05ZoCIi7NxgAiASsFK61wl4DiINDHwJgmZ1bCrZQcgnvoi3hSIeZ9P8e3tpkANO97edkC0molN7ZIqy4SdxcCfQiMeXbqeVkuFRECQqHF95oxF5p7P+DNDJGq8uqrry7AohoQCY1V6uk699zPhyNhLsglzriICH0f25wNmAmSQZJhfUG183MJrWpAnYhmkAv5+jn5cImWI1ITXYz0Vhmnidoqi5sJxaBYpCRDNbWbo99sVSPbbQdU0njEslJqQHLk8vG79G98Bvo5rhnkYEzT0Rmm/l7DyIojK47MMwFYcWTFkW+9rRiyYkibCcCKIfcBQ+4v2gCboW8+6N517pVeqCVRGzCYJQSoVcmlgAZi33lXtzmzyJSqWIgU8yq10mI+Rbwy8RwzGUOEFmN7W2CsoC2OU1WQuRpvIxxOTra8/uprxKDY+2wu3y9lnp+fOyOgPtmjCj/02R/g0V/9N/i1v/d32WrkKh95+53HvPf4Gc/e/jKPL48cxpF/+gd/xPV7jzm8eIbakYcPNwR6aq6UZNAqE3cx0ne+GKxUqN3CKGDiiZbiC9YlZui7DfSAQcmZUgLJyrL4tTq45iktC1BadWuPna1E8UTFWWpWDUs/dq3vay7E3iv45pzd375J0fP53Y1xfakv37/zFxCCv9/8XGa7UP9/O6RBtYoodBKX2N8lmfHOZ82xvyklgkLse+Zqwx/5yEfauQQq0G8G0N6Lc2mk60D3RzQKQ9djoqTpiIkQwoCKkPMp11eZJ/0lpyenbE57jAQ0iVqcsYrZ6CSSSRyna26unpH21+xr5ViN0A9IBrOx3ScjqMdclzKS021V6VpaQTha3HWtDH6n4nD9nPHmhqE/oZiRxomUj8RghBjZ3OMvLbDiyIojK46sOPLB2oohK4bcbSuGfO9jyP1FG2C33RAUpmlif9wv1WxLdf/1OAQgUUqmyhaNPVIhCKhBrkY5HJbJOzuiVFGs+C7dzGMJS1uoMCfCuS3fNCWsQoxKFRCEYdiwKYrqJREwS0QqXYAYnS3wEN07GmVrIsJ2u10W8Vw/oNbKw9ff5G/8N/8dv/2PfpN//Nv/lM+/8zY1V87Z86t/+2/zoY99lL6M9PkJ0/VjzrtI0MybFw8BYzxO5IRLnEHRTomxI4QekQ6Zi6ThLiVWuZVwWwGu0JgMl/QFie6mkgWyFWwaCX1szjQtPtYKYhkJSlF8/99ClEOzN6mGLyIiRvEHLKMEd6+ZWa47HfYNSX7qB6vNuz+EgJl6bKj6NeUpIaqoCH1UUimIhDam4FWLX05qXD6jqecinlA49LcSuKpyfn7OK6+8wpgymFEMuhDZbLd0XUfJhcNhzzQdGXRDkErOE8ESKpE8HlqF74Fhe8KL45GvPnnC7tEJJ6FHDSKlYa6SVBiDsd10VA2UbkvdjlhO1ClzfXNkf8ikGkljofbCZgANQp4iXd+BQJkmtMUiixnnJztSzYzZGKRw8+wxl++9w4MHr1LM+c0hBsQyapX31V67d23FkRVHVhxZceSDtBVDVgxZMeR+Yci93vz4LtgdSi4vL+lDZHeyI8SeGmqT0NpOv0byvICrUPHVVNLktoDBkwnj0BO7DgmGlUqa3Jvffdt5SeLMOVNyk5TNkxRL2aM5cmy2libGZujY7jqGjVcTrlZ958s3JmctzMUdpmf+fTgmXnv1ET/54z/GR159xKPf+TJPnzyn7jss3VCvn/OwM+LWF2sfNgybDSEIVSANHktZTCiWMQwNSuwiIfaoqEugITrQ1EotmVILOSppmqhmgHohKxFSMo6HkelwhDzh0aotTrb1tcfV0titRAyRbNmrIkjnixgaMyII0Z1ABIKGl/p87pdvaC0+lpmZ4Q5Y3GFgZrAQDRStCC4zm5m/H14CnJdCAaCxdNaKmN0eM4ry5ptv0vc9qXj8bh8iXddxut0x9AMTU2PjGjCW5CydCCEIViuWC3EYGYYNKRWOxwM5C7WqX2Jw6bhSqRKckusHjvRcjkYdE1PaM03Za6VJdMCW5ONREqVmxlSJGgit4FsphRgiwzCw6RWtCdPKZjMw7jPH/SW1jJQcSAZn5xeepFoSU71lD+9jW3FkxZGlrTiy4si30VYMWTFkaSuG3AsMudebn/G492q15oltNQilGBqMWqpvjdsVjungE16bHFlxTsGMWrPLwxrpYmTY7AgiTMcj4zj6JEGWiZlS8tjZJm2m6cg4lSadJrSL7KdKwCjmBa0+9NprnOw2lJKIvbMAZt8Y4wsvT/JSCsfjEVWP+Y1BePPDH+bDr77K629+nHffeZe3vvhFHj+5osjEycMThnpCjBDlnIsHDwl9wMQYU2V/GDkcRw7jkZQmT/4LyjBE+q5HaTHG7fNzzqQ0Ug8Jw2sJiABdRETp4sCw2VLTkZRtkZBFPF7ZzGsWqAkmlV0wSvUCZKJCtkQQj92c6yQvkndb3HOf3P3/3Ee1PbY8hwOHV9Zu1YfFY4L1rmQsRojqCy7N7/UTnxkmVWXuiZc+s/jNSmfwqka/6fn0Jz/FZrNDQqQWKEsvNkcZcQeWGCPaip/NCZQGeJJpoIuG5T1alTIdKdmtSrWDKIZKRWrFjgfKOHG0DCFiOjCmSj6OlMndfVQjBE9oVQzMAVaQRXKPMcLkjKKHBJSGvYaoIdW4fvGEmhOl+Jy+O1en8f5+aYEVR1YcWXFkxZEP1lYMWTFkxZD7hSH3evPz4uo5D84viDHykY98jO12R5gHu2RySUx59F1mdhk5tCREQ8AUjfH/Ye9dYi1Ls/yu3/oee+9zzn1EREZkZlV2ZXd1dau7q7pHlloGIR7CNggZIZCQkCfYYgSMmIBaCIQQEmKAQTItj5BAjM2MARZCCAnaEwQSuG3L6nZ3vqoyMjPivs45e+/vsRis75x7MzKrnRGRlZW3tFfq5r33nH3Pfn3fb8e3Hv9FFzxD39OtBnzsUCy/E+cZVqsWslRymsklHfcfQsA7Ty2FlBTnhS52zDlzdXVBP5xxcnrG97//Dj/6zV9FawGZKRWc+Ftny51QnYigYoP6MMiHYTAvg5iCTNKA+p633oQnj8/4zR/9OjeX12yGFX038Pf+/t+llhnxkSdvvcOw3jCnQj+YdGUplT/6h3/MRx98xNXlFdvdNfM8MqWZVK0o0InlXtZiYd/grOAP51AC3kdEAhRlNZwieWbME+N4Y2o1pVoHZBSHELxwsjrhB+eRMGz45OKC692O67miEphyoVS59T85fzup3W1O82Hiioh1TuYAiwYk1Qb+BjGx/GlxgjRJHlUbD8F7hq7nZk6kkhu0qk3SdlsOjeGO90bNQ6VNY955z2q14nd/93f5jd/4LQC8i1TN9N1ADBGvhWna45wjDivWKN55TtZrxmlHygZpFUetypQcaEU0Me0Kn33yDAlvswodpAk37ZDtJfv3/5Dtx39KStfk7ZY+J1zYsJWZKW0pWpllYl8yFAd4tqkjhECVTKlKnSux85yfnDDOmZQSJWfwwpyUNJl3a7d7RghCFmU/FZ49e2YN51C245c/OO+LLRxZOLJwZOHI69jCkIUhC0PuF0Pu9eJHVZuCoSOEjhAiMfRUDylP5FopxQrJQnAtHKyoHHTcTU6wj5HNyYZ+vaYo7PYj42jqEV3X4UMrUquFlGdSSqhq8354NusNuXSoVJxE5pKpWnn86BFvvfMd3n33HRCbQ7UY+HIuBPlisPnu74cCt8O5Vi9IFeYsFPGsQ2fgcp7z0NF5h5PI4++8i+SJKpmzB+eEbiDmSuziMWX1V979Fd589BY3Vzc8/eTHXFx+xvX2mmcXF+3cTfKyaiaX0q6j0IWASgQJiI/4KgTWSN2jaUceQ/MMgVa7vtbN2LFarXhyvsZ1PcE5Hpye8icff8qYlS44xnRociZHLwRVyfmw+v+8J+pgdz0h7uClwbw3Rw1/71p+rV1l88ZYUWQIHl88tQHp9rrL59IL5HMhb9un956+7/nud7+LE9c8FuYtijGyXm8Imrm+2SLicd7jfSQEZ3KWYqot9XAezjNlRcl4rTgRrq93nJxl+m5N9IWQlTqO+O1nxPEn1P1zyi6jY0VrxfuO0FmqBDrhWs9ord4UVkrBK4QYQK1YslqrObrW96FqRatSmgqQqanY9YLCbjfiyNbHwA2vNY9/3rZwZOHI4bWFIwtHXsUWhiwMOby2MOR+MOReL36CWrFg1UKqBR0nYgBx1sU2RMGnAlIJXtCaTMcdBQ9FFUVwwxq6ARWP1kRJE1r2R+38Ug/FbRWpCrkg2srdxONDhwQFCq4ouzoSvPD9X3nE9955wqp3Lf8UxHVW9PYlsAEsGVU/n096kLAkWdqkSCGKa6HMQJkz3gWq8+SSOX2wYRo9rusIXU/nPZRkrgc8iHByvqbvI6tVJOsIXuk3Jzg/8OmnT5n2+wbqQK1qXhcCWgO4iJMexIPPEDzd6oQ074nbjppn1KoG7VqnkXWAt07OefJmoGPiQT+xm2E3duwn5XI/MReormfKCUfCSSYUJeBxYUDFMecZwfT3ozYPhUJVk8csXnGiNslqtnC+FmoVe58WRhYrpfQ6Eylsq1DxOKlQJrtPLlCdWlfiBp/gOE7CQ2Hjm0/eJviO692WrhvwoYNq3Z8Fpbpg90orseV6j2Niv58QEbL5igguoCURyEj1qATwjpv9Zzy7cKh/zJM3HpBDIFOZcqZMmXncU12mW5mMqJaCeNCcCQVWuiIBzhUiV2hRUvWULjAMPT57dvtrhr7HeWWaQaRHfCWQyZpZ18p8s8WfbRCf6TqFIoh6Mu7LRvK9sYUjC0cWjiwceR1bGLIwZGHI/WLIvV78dF1vE1eg6sxuPyKyJ7jQVtEBH2CeE/M8MXS9dZTF5BJRbbJ/phM+z3vSNDNPkxUWFqHk+djsyosNunjwvkillhl1Sug6gu/QObNeD/zwnR/yq7/6fboWMi2HkKhR6ug9+LPsbjHdsaFZC7VCZU75Nn+zdei17RzrzYbqvCnGaLEQvAhV1ZqiOUc39HTDQFwNPH7rbaZp5r333mNOE1vnSGlGtdL3nm7wIBGbyp5SbaIjnlIz4zRTMjjf4xBynplzxgt4D/0qsNn0rDtPTZl11xFd5nfefcjF9Y7LrfDZ6Hh+M7JTpRIQOqa+UnMhuopQ6EJBKGitxGjds821UhESvVQ2Q6DvIg5H0doeCpH9lNnPCRArDi0J7zwxOJyrxOZh6mOPqsc6N5hef/N5HWF1kJMMvuMv/At/ifOzBzjvmdrnhxDYbq/Zbq8RcXRdZNX1xBiZ9jtySvgu4kM8KgNJVaJ36JSJISI4C52nyrOnn3BzvQMV3jjf4M7exD35ZcuJveyYnn1Inm5AK1BNIpViz5gmhYpzZPGtSZkjp8K2FGotdLFrD7aC8z1ztjmiTa7Vu46ffPQ+310/pI/RvEnBE1zHlL/CYP4W28KRhSMLRxaOvI4tDFkYsjDkfjHkXi9+fPRWNCiHHMsKFPb7CVXo+57YR7uQKZHF4WPEtYZgFbHiQaCWTC2FaZzsJqrHWkt5rPmS3UgbnI6cZ1vJi4NsGZ/Fe5gL0zzx5GRtyibFmlrdmnLrZ/niKvVuaPnuawfgHArBDlr/YOdZSiGlRNd19snOHXsD1Oa9sWI3RZqePZUWVoRxnthut3Qx8vjxY4L3XF5csN/vydVUU5wDcRHnPD46qlrzspQyThybzQnl5hlzLjiE3nu8V0J1vHG+5uFJh5tgtXpE6TLsd7i8ZTMEfIyEfWHTKRfPb7ieEkhEvNBFWEVYd4EhRrQVhVaElLLBoFS8d5yvOs42AzGams12HJlzRcVz44XoBRUhZ0dOwlgUkUqtBXW+CbWYK0waaKzQUAALRQfnydUUSU5PT9lsThDnWJ9siHM53qO0S6ZaEx1DF+n7nuA9fdfRxcj5+TnqHNM0AjCPE/NUUMxD6JsMaU6JMjlS2vP84gbvA6fDiv7Jr1BdJHQ9mhO7Z4V5uibnRKqJrLV5o6r1fTiEtdX8dIcGbTYcPTnPlovuPdLyq2utOHGIFq6ePePxPOL7Dd5HHBaO/lyi+D20hSMLRxaOLBx5HVsYsjBkYcj9Ysi9XvzMc8K1Qj/xrbEYFa0ZE8GItwosOVOKUh2AkHNF24Qv3uQCUyo2gLUVEPpg8n1aIUMu1uRKS2KeZxRrrIQGCkpHjxflZLPh7OwMbXm8ZndB84+3L8snPcgl1pY/rArOWYjVOju3AeI8B5hpsW7B6lwrynPM8wxVkSqM+5H9ODJOM/v9nv1oxXCnZ6f0Q89uu2WaZ3KupFytuNBZgyswWcSL6YbdeANpZtVHOhlMljLtCRRWneOX3jjjycmat1cb5m5gO44t3L9HvElXrgfP+eaUN886nl/uSEWo3vPodMWDdc/ZZqB3ld1+b7CbrVDOeU+dG/ROekSUlFo+dCg4VeZaCaJ0QamqhM5To0NSIeytkRbS1HfaHLLu1zZp0YN6jKUnBOfp+4Hf+Z3fwYdoyiO54r3BPKWEViUOkRAtlxctTNPM0Ec2pyc8OH/AlDNXV6b5nzAlzNAkH8V5gnNISxEo4ri8uCbnwvzwhLcfPOF8fcrcdeTdFeP2CkkjinVTLpqpeLrQUXFWVFgFJ9bl+5A/HHyk7/rWMA/GOVOyUnIrpHSOkhI3Nxekac/p6QNwipYmvVrrF8bqfbKFIwtHFo4sHHkdWxiyMGRhyP1iyL1e/NQaCG6g6yNVTXav1EIYIngrPKxFcBJZ96ekPNoAqqZP3iKU7Lfzseis5oxIpdRKSTOlKrmUps7iqKVQsiJiBYbiTbtckBaWq8Q+cnKy/hK+HIbry5t5lIRp2pOSSQH2fd+8K+ZxCq1b8VFTvip5ntlPE955XAwUgaKVq4sr9ts9tShzStxst+z3W5wIp6enNkEwqKeU2j5tUvoQODk9ZbPZEIfI9fMLrq8uuXn+jJ/88T9gf3VBzYnTwbMZIg9Wge+de/rgCH5mzDNxEFYS2e2FVJQHJwPUSh+F0wfn6NsPCbGjaGEVgDxR52u8COsV+G5gV6w4ryjsbkbSnJjTlhAO3ZwzPggecFkJUYmi5GLeh9VqxZkEbvKWp88nipjySXDOcqvFcnO9OLTcSkCCFTP+4Ae/xm/96Le5vL5hnBL9zZbT01O0ZOZ5ZtxvqSUTUmCeJtqf4pxjzon9fk/se1KaySlxs71iHCdKzZQyEWOi6wa62LPd3vDs4jniPQ8ePmB3/RCp3+HRgzNWb/0Wm92O7eUzxt01Ps6so6dfQ06ZknIrZvSUbKo06hwlGyhKhe1+hGrSoebFSigQvLCfJlx1XF1+yvOnH3L26BG5eFNiqQL1/qo0wcKRhSMLRxaOvJ4tDFkYsjDkfjHkXi9+NienrNdniFP2+63lJGJSgsEHEEcqpgrho8UORUBF7X0VarWwLWrykOIqaLJwoThbRdfKql8zDCtKytaZF/DBpCiLAm2Al1LZRGtOZnVlh668wrEt7yuaQce+H7T978ouHjw7t4ogBs39zQ1djEQGspo2ep4T027PNE1UVRs8Cl0XWa9XdF1HjB0ijpQS0zQxzTMhBDabDY+fPGa9HvDekX/pHaZxZH99w/X33+Xy2SfUeWYzdGyGSJ2uke1TApn99Iz9uG2h/YkhCFRT+PDRcbruCQIpjQQUaiGNM0MndIMnxhXdsKLiYBy52Y2UnKhkQh8oKhYaLwXng+VD10zwgQ5IJVkjM61onvFdYOijgUTBKjzN4+JFwLXcaAVp80pVOTs74/HjxwzDAC6S8g0pZ+aUiM5yvFerlanwNI/f4T6N44h4R9d1bE5Pj7m7q9UKgP2+sk8zpUDKlb2MPH/+jKdPnzKseoKrRKdsH5yx2WwgRLQ7JQwnOB/xxeNcQL1nlpFpnqEWnFijs4owq5KpDUTWNfsAHHEBHxwOjx8LQxftmswz25sLaprBb8xB9dUdiN9aWziycGThyMKR17GFIQtDFobcL4bc68VPFyNdFwHlpum5O3eYjCYjeZjezitSrB+tcw7nPKpQc0EkIw5C9BA8OWVqrhQ1Kciu6+n7gb7rKa5p4lfFOU9VwXtb3ZeccM5xdn6OE2chSloYTg7/awWOHAoOv7qJ0HJ8c+sY/UXd90OhIQDVwu42aasVY0qbkDXjHXgnlDnjULponZ6naSKlxDAUhmHFMAycnp4yrHqDeQhsNmvT2leI0bHqV5ydnPH4wTl5/j61FGJ0aMmU+YZ08RGaRtLFU3x4Sp535HmP5kzKBVW7TmNOROd4eLoCKrt9YrXuWA0dwUfmuZCLMqcRTRM17yjzhOBbzqdd8uAjtVRqnpHWAdlhjcJ8806VlCl5T5py86zY6+Icmot50LA0BJPKdCiV4D0nJ6fNKxVYn5wj4o4N4Exes6PbrCzfOUbSOJElk8eJ7W5P1/WE0DFNM9Y/YcVqFQk+Igi10pRtEmNKpHmPI0GBeXfFvvPsdlvGccYNK1Kp5JZ/7IApZTRXpEIfg6Ui1GKpEiKkZPSsWtACwVl3bec92/3e5pIXYvSoVqQ9wKbtNfuba8LJYDnbQpOcvL+2cGThyMKRhSOvYwtDFoYsDLlfDLnXi5/9dIP4w0kLEjxewDlFSeQ0k7LiQqDohOKBiNaIpZ+aJyZ2Fev+q4gbqDVQSyI4wUdTwVCchQFzRqsVJNY8oeKIXU/sOrzr6H3kjYePEDF1D2funTth5xcLDr+6iViObK31eJMPCiuHbs8H5Y9SCq55HeQQfq6BUgulFqQUapqRkhEtaMnE1mNg6Ad8CPTDiq7riTHQDwOxCzb4RPAeTNWkBc+bmkvnz4j1hKqKBMv/9TzGrU/YXV/CXAn7GxxNuaNfUXcT+/0OSsTVyGa1puvWhDrTbQbWJwMpZ1KurNfnlFyQaprxWgKuFLRYIywnlVXf0XUdKSWkKKKFjDDlQi2KqqDFwCLOEVqOdhUHmPKIXW+h1tKKV+0BhRO6YcV3f+l7PHz4yLwUwKNHj47An/cjUE1q0gtzVa6vb45FoZvNiYGo5SqXUpn2I1or8zwzTVsGL0hw5JQJVQknK067xyhK7AJSMx989BFZIm+/cc747GN2l8/weWYIHu8Du2mipExwHomOuZjUpTgh1kJV0KLgIAQDrnfK2cmGuRR2u4ngHbNYoaXTxO76OdeXz3h48oQQIqqV7c3VK8zeb48tHFk4snBk4cjr2MKQhSELQ+4XQ+714mdOI34WRDwxRqJ3lidYrJNyyaYoQqmUkvC+A+nJxTPnjKB00bWiPEAdqgHRaMoYQQixg9YwKucEJRO8eVi0ZEqdmecCUvA4vDf1DIvBtaK1otxVUzEvy6vE+Std1x0VPG4/TyjFPDAhhCOAAIL3hGgqM0grVAQDkarlgXpHVUeaR6bR45u3Rms5eq8ODcIOoW5pBZ0HnXvbm1izAxFEoTqH+AjV4YcT+iJ0j9/mev8puY6QPVd5wpN5eHrG1U4Rt8LHM1TWxF44P1HGeUeabuj7wWBRHX23okwTfb8CPFUc4qHMW/rO41xFvdAPEQ2efaowV3CCVgFxqFiqgah1WC7YdSulIuJxwdRzBPNy1VZwt15veOONNxjWa0rJ7Lc31NXaws5NlQSsH4K42D6zMM8zToS+78mlkHMmdB6qMqUZUZiniVqTFbqKI4iCKBIC0Tu8dzjv0RC5nkY+u7rkdHCU/Q4tpoZC6/vgxYEP5glMmaJQxeNUCd5ZIa5Yc7PYBQszi3XxDlmZ50KqleAc3gleKjmN5DRZ4WFV85Ld85yVhSOHz1s4snBk4cir2MKQw+ctDFkYcj8Ycq8XP5ZXasV3ACqmhZ5zOk44JwKlIDhUHaVYQ7GUaRe4EqMVPElcDQAAIABJREFUEKKOWiyx0lbiNjDFeaTocR8iQvAORSlaKSUzjRWHo0OOE1vFwoVeBAs5v14Tt1orw9A1JZhknow7Yb5DmPPOC4Suoy/WRTqreWmceFT3OG9dfnMthOhbYWU+do3u+s7AItq8LObxOZgiIK1RG4q5I0yRXpwgatfIJAy99TxYnVH9ilQDOVeYRwYHMQj+4Snq10gY6B68bekC3Q7nYeUUUZB8SFWueKcEEVQiLhXKaN2sc8loKkzzjOsGfAyQZ9RVKko95ltDzpVail0XbSo0SvOEYOnRYgKTwQdiF4n9QGrKIrUq+92OcT+STzYEH5u3y5RgYs6kVoQnKDknnINcinUtzr0Vt5ZCdIEgQsGRy4x3wQpbvenhI47QmTzqrKYClOaZ/TgRxeFij2egzNZs7DA+nPdoKpTaYKrKOni6LpBzMQ9OtBzc0nKqiwrOO5zjmLrhnJDSzLjfMc8zuSo+0FIU7q8tHFk4snBk4cjr2MKQhSELQ+4XQ+4vbYDTkwes16eUQpvklZwnSvGWX4oQfA8YALIWKgUVwUVbWeIMRBaCLOQ8katJTKIBKogEqlSbRNavl9w8KDEOdGLNuyi55WpC07HgkFtrsGk5t4eXX8FExHTjsdDywfNhXpZgIcwWesdZB+WA4ETwLfxcVTl5IGg2WO7GPbkqIYzsbnakaabvLacWil0fdcf92ym0c6o2EeTu+dE8OtokG70gGhlOTtHVmvW7ifDgLS5//I/oZ8XnjOtWrM7fZjh9yOb8EWePv8OcCyudKLuPSdtPmK8vWKXMvNsTo0O2E5oqIQRubnZ4zSiC4ohdR6kwzolMboWXllJQkoJYaFybF8p7T5RArqbAA1bA6Z3DiyO4SH9ygjhB8Xzy6TMeP36TR48eU0s1b9w8E1ZWtDfPiSklnJvw3lRKN6vhGL4O7cHgW3haS6GIFYVWIuvVmiF6RGDeJ6q3nG+CyZ1GoAsrQtcxz5XN6RvU3RNufnJJnQ2sKtXOuQrO2o+Tc8ILpgykbXSqkqaZNE+gkKtS1JHmbJ5ExCRcmdFauLh4xhvzjIvRvC6v5Dn89tjCkYUjC0cWjryOLQxZGLIw5H4x5F4vfk4255ycnpGyqaBYp1xFXKIWW/3HviOESNXKbt5Ta7LQHRaiRgLgKbVQy0zRGZUMeCvyytUUQA7hO6ywrio4cYQQcd7yT0ve2UQUCzUCmLgh7TX3ahHmL5gSgum1e28AODQeg0MoWFBxFi7FmZ8neCs8bNt4NRnBAviqUGG82VnhpUL0Hh98A5q/DW+LNWNDFaedeSbEzhbK8TiqWA6uiEdDJUggh0x88x3C2TkaB87PH5JSpvoV8eFbrE/PWZ+eouuHaFZirnT7h/DZGtwKd3XBadzgUOYykbAu0X3IjC6Tc2WeCuDQpqDjnKeLjs5XOqeomLdKnIVXtTXecq3xlhOx1FoU7xzBBbquY71as5/35FK4vr7h8vKKnDJdNA9WKZXdboeImCxnsYJPh9J3fYvCCyW3a+fMy5fVdOqrWE6vCytW6xO64Mi1kKcb5gO8gShCdI7gHK4q85SJ5+eEB29y/ekHqI+EYE32pBbLCS/mVaJ68Ja/66qSiuX+4qFWO95SKiqxXRuhHPpQqOW0z9NoedzR0hrKPe7PAQtHFo4sHFk48nq2MGRhyMKQ+8WQe734KTjUeTSUNsdtIuZZqHhUrJGYGyJBoVAgmaygtkninG8XFpq8PiK3g6/mTKltxekD6ix38RDSrSilZKz5WEbSDUxX0D0C6RAHlQoaTKWCguKpBBz5pc/5MOlFhL7vPgeZu+/ZOTi8CNULuWTEBZzY4NBokoZJldD3SBuQJ+dnxBgZ1it8iPTdgPP+2HTqNr/3EHp/cbC55olpDFILRYvIEcEu9LAJnH1vjbz9LjkXcoGw2uBDpAaP+I7eAaHiwzt0rHHdGxT3J1AmyBNhviSWa8ZxRy4TiuBCtDBtTgTv6NVREKpKOzZrFqeuUFSoLd8WEqjiUUQzVW+9SM55+mFoHj1BNZNK5tmz51xeXrFery0HW03BxztPrVaYZymoYg+kYt2MSws9qwhzrqSi5Kqsu9hSAQIpJ0oxNZyaEwEbm7mAeG/3F3vIOCkktyZsvsPm4S9RZEKnZ5QGe3ERkWz3XmeUQFaPsyfP0WvnQ8S5wJQnvA8MQ0cuI6NaUaqjNi8kzDmhyQHaCnbvry0cWTiycGThyOvYwpCFIQtD7hdD7vXi5+bmGueFohUtGa0V0SYfGRylVuY0o6MVXNVqnaicNglBtRxcbeFh7z2UAFiRmU0yCwXWWk2eDwVn4eY2esi5AErIE6pbRE3pJBMQim2rDqHiULIcgrGvZge4hPD523e38NBA2rwiTmz/ziHVtgvRihVdCKzWa1SV1WqABw/oug7X8ned8xZmFPdCeLzlE3/hTITDhqKH9/QIxiCBSkUkInFFrae4UvCl4A8FkiIIFS+gXikMhLMnhPWKOQp1vCGkmXL1DNUObCqQxVlKQLCCQM3FeifUwpQzuVQrtENQxPJOq1LK4YGTcYBW8za54OlXazarFevNCWPKeG89E0rOPLt4znvvvc8PfvBr1qlaK0glY3Dt4oBzEYBpmnFNcWdO5r2LvW/jKeA8OBcQZ96j3W4074oIQxcsVF6VVBVEcD6wPj1hGEyGMyOMbo0MbyDhJ+h8ieKZs/WJyAVwHuerZQhgBagiYKPk9mEVvOX3FsXmExXn2qOkVoIPhK4D8a1T8+vlj/+8beHIwpGFIwtHXscWhiwMWRhyvxhyrxc/11cXbaiZUogVBirS1Da8WKfjtN8zVcvAVLULLm0FXNW2v1UROahGuKMnxmQdTW/etvOIM2UKcQ7vFQdIgZqVlAqhmjb5T4ss//R3vpodJvDBtIUuDz8ftqktDHhoRHZ4z3tPrfUILRFphZaHnF05qrK0Ufnax3gwU2rR49ehUdohXH73OHHOigqDw4UB/+gtmB4S54z7+ENkd0WIhUoi10TKCSdWNOl9R82ZPCWmVMi5otr2jYGtFqXWgghQ9dj3zXoIRDabEzarFbHrmAs4r5SaySWxH/f8+Mcf8YMf/Cp935HmyRR9BFNI8Q7nDWalWIi/qB2HhGCypdrO13lKzYhaT4iU8/G6xNihQK6ZLjig5U17QUSpFXb7PYMU+pNHpMsT9rN5cVI1sFaEgidRcVU51KIKwtB3x/tg4XWDbs7VHtJACBEfFJcdzgdiHEi1knLmZrt7+cHxLbKFI7e2cGThyMKRl7eFIbe2MGRhyH1gyL1e/Mz7LaN3RAdVldoW+t3Q26oxeDRbuHieE+LaRFRHdTYIwBHCoTkYTSvdwoy1aNvG1E1KKW0yVIKLxBgQ59HgLaRb9+RdZj+OrEtFXZuwd2brATOv4mv5aRP4xW3uel1e/Nu7Cize+895bCy/1AaeOMFJy939Cvv9KnYXhC/+fhc2d63KQUmnpRYMp/igFJ/Q/gztThCdUEzuMDfgdNHT9T3qPDJXECuuLCWTs1LVIT7gRRG5VYUBEOfx3jP0A30/mMJOSzWQXNo4MSWf/X7HOO45f3CGd5DSbai/62K73ybPSFVKnhHn6WJPCPE4vhDzDqJAuIUxQEq3SjoxBMAepPvdDfM0UnHM057ildXmFPoTdnNBs1Krb5Ka5mlKWXHOkh20goqal6t5xIIPqMI0F0pOKOYhiiESQ8UTmqpQZbef2O+2fPbJ069lfPy8bOHIl2+zcGThyMKRr2YLQ758m4UhC0O+rQy514ufLnicFvKUkeAQcWStpCTEGG1F3SQTqaUNJIcPgVIrVQ5hUVtVWlOugHNQqxyLrvwhJHeYDKqg2kJ1wn4/W7OuaUTFoZj+eaVNKgELLr+eh+Wn2V3AvAibFyfw3d8PHpgjpBTEa/OufHHyf912FzA/fV+HsHX7WXpK8NS+4B+9TazXbD/b2yR2ggsOHzwxNLnMUhFxRB+JAUKpppAzKym3LtzOivecE5z3dGGgHwb6YY24AOLRNulSqWi2Rm2lmITl+++/j3Pvsl6vjhCPTX3EDr4jBk/OiXkCdYew+q0qjTjBVPwVqKZ1jzt6pu72U3AOS6OYRruHLuCoTKnwbO/I2qPhBCrUkqyQtBRyqVQcoiaLWXLBBWuq1seAE0fVSnBCFuuiHEMgaWFK5knSqmxvdvzxn7zH1XbPbnvD5cWzn80A+YZs4cjhcBaOLBxZOPIqtjDkcDgLQxaG3A+GvNTiR0R+D/jXgN8E9sD/CfwHqvoP7mwzAP8l8G8APfA/A/+Oqn58Z5t3gb8J/HPADfDfA7+nqi9VdRejx4ky10LnAj4EyJmaswWgXUVLpdTavCkmk2dhVlPesDCg6aR77+n7DieCBQYdOOsibANIqKWQUyLNM6GFa2+ur3E1s2Lk9PSMfljbQAEbP3eA87NBzucn7Isel8OgBe6EFG/t+F5bcX9doHmx+dmXeVn+LBMRgyAWEhU16UgVj3SB+MYT5nyBSzeE4WNWYyb7qUlrQs2JWm6hYl9WXKdioVSvNuGcF6QYVM4fnNF1AyqerIeHkmswsVSEaZpQFWZmPvjgA0II/PIvv8vQR2KMdF0kZ2vMplgBX86J2n4+5DF/zovXTtY7IcYewR+9LqWU45cpx5jXx0kT1KwWWr7cJ6R2uHiGlEyZMuNoRbBFCyog3rW7rc3LpdYsT4RpKrgWbg7e4VyHMrGbEiaRCtubLX/yd/+QMVdqyZSSXmpcLBxZOPIytnBk4ciX3NOFIQtDvrItDFkY8qK9bIXhPwP8PvDngb8IROBvi8jmzjb/FfAvA/962/67wP94ZxB54H8COuCfBP5N4K8C/+lLHgvTOJFKQYLHx2BKIDGaBGRtK3QnSAyEYIVaORfSnMilAjYJSutwW0pqocjc8m+FWmFKiTllarUwpGug2e127HY7Suu823nP5uSUblhRq34uQitf+OlnhZ7P25eFnr/s90M+rn5jR/Zn261nqxyb9goOj4VJHRW6AVmfE8+fsHrwhG59iveBWjKUArXipeUPo60RWG0eMLEQbNufE/NsxK5jsz6xz9HDNVG0gcd7b0o7arCoqozjyNOnT9nvRwOVCvOcmaaZ3ELTMfoW6i7HnGLVSkoz8zyx3+/NA9jugz2wrMDV0iZM4vLo+eOQC5uYph3TfkdKM0kr+B78mkKw+1lNTcc7QbR1SQ7WPds5Zx2wczEVHrEcXueErusIwY41eEcXAyF4SilcXFyw2+2ZU6bUlx4xC0cWjnwjtnDkF5YjC0MWhnwjtjDkF5MhLxX5UdV/8YVB8VeBp8CfA/53ETkH/i3gr6jq/9q2+WvA3xORP6+qfwf4S8APgb/QPDD/j4j8R8B/ISL/iarOX/mAnEdiNK+Js3CgFQDSigghta5jzo7fpP1U8aEjxGDbMaNqALIBIqgoIr6F3w5NqQDUdPabp0bE0fWB3mUGVxnWJ/jYUdqNy7fulsNVaxfzhZe/2vU/XPfPvf5lXouf5sk4AOhLP+swkl/t8F7qGL6qBS0oQmmhfqcHkUohDqecPP5lxn6Nzg6JHyAfQ9o+A9QkM7VAaTml6kAEFwLkRK0JabnEsevwoWdzcgpNjafStPmFY/Gh94FaCzF21hCsKrlULq6u+OCDD3ny5E1qhVoyczp093akFA12IiZd2TwnB72XWpN5V4A65pbvfFs0egCl976NY21ewgIlESWYjGie6bs1J2+8Q+3AiYXGRSq5jviqROdxUqlaqWqN0qoR1LyPzfOI96TJVIXE3Xq/VDFvkbP83JxfTiZ14QgsHPnqtnBk4ciLtjAEFoZ8dVsYsjDkC/fzK2/55Xbevh8S7f4c5oH5Xw4bqOrfF5H3gH8C+Dvt+/97N/SMhaP/JvAj4P9+cSci0mNh64OdAnTrDd0wUFC7YAihraJrLaQ5M6YJRJDgUYRcKoJj1Q/EzrTpK8o8T20gGHhwajKDciiAMwWMnBO1qoWs22dKu5ABiP2A+oge5CyhbfN5AfKvJ5j7evZi6BfMQ/U1RZq/FvOqzE4pzfsQVXG1gDp83NhXf07fPUQ23yEGuPzxHzHvbkBnhLmFqm2i5lrJqBWTqk0k7wMn/SlhfUroeqZxb/dMTEa09YimYp2NvYvEWOz9WqwT85x4//0P+eEPf5uui9RkD7pSCvM8mbRlqSjClGbzcJRCDA68mHNIreg1zZmSldh1rIYBRJjnmXEcCaH1dZCDRCZQKl0fCd1AqokqHcPDd+hPOjwzOk/kMqNpwiOIFrRW1IGqQS4V8wqVatcodpE5lQbVaiO41hYOF2LXUyXa+6/vnls4snDkZ2oLR37hObIwZGHIz9QWhvxiMeSVhfXFloL/NfB/qOr/115+G5hV9eKFzT9u7x22+fhL3ufONi/a7wGXd74+ANMiR9xxcPhgco9FK3NOTNNISQkthaLWAEnF4X2k61b0vckG1qrMc2KcZ1LOzRshTPPMbrdjmqamr69Nck9ttSmO2hiVSyHNE6lWcIGqULUg7jZ0a5mNhwtYv3CSPy87el++hn/Ffv1m4eAqJv14kIW0d1rObVgjwznx/E1Wj34J+gdUP1CJqARC1xO7wZqOVdhNMynbw0AEQmuk1vUDqoeMaNuDhaRN0lHVFFAODyELw7bXsUl7dX1tvRbcYYKacg20PgXemo5JC1uHEI6h51q1pSi44/1QVWILM5uXpRr0YsT5iIpHcagP4AOqwtVu5CcXW65nj/r2QMYkU+2UrFt0DKHlAdP6DNgQEOdJtbYwcj6G1FXNN2SRZQdiD2Ln/Kvf3YUjC0e+EVs48ovKkYUhC0O+GVsY8ovEkNeJ/Pw+8NvAP/Uan/FV7T8H/vqd30+BD3yIeN+hJSNqTaRKLahWSi1UrXTBE7xHxeF8JPqOYbUmRGuMVIuF/5z3CCYvWdsYOxTGHXIfoYWyRRAX8aF1pkWQaQSp9MMK13ITpbabfGce62t4Mr6u4r+7djf39KC48m2ySkAA33TvRcUmokDVbA8b5yB0xM057skv03/2GXNSkCuK3pCnyRpyuQ7xUCfLaT3klOIsdxYRcrXkgKqmv98cG2jz0KgUnFjerrHIHXXwnXg+/OgjAM7PzywfW62ZXT/0II5cretyF1oDO22ejFjNn6PgcJSSKaUwjqPtW5UYA3gPzqMqpFTYbvfM+5GLXSapWu73zRZJI29vHG+GCRGsmZ3zOM2Iv5t73MLYtZrqi3N4PLm2AkRnY14F5qKkYuFpvHVgrgrkV/ahwMKRhSPfgC0c+YXmyMKQhSE/c1sY8ovFkFda/IjIfwP8ZeCfVtUP7rz1E6ATkQcveFzeau8dtvndFz7yrTvvfcFUdQKmO/sHLKzbxQ6HMO6bR4TC0Ec8kVXwRBGkQoqR1eqU2A10Xc9cCvM8U0rGhcB6c0Ithe3NlpxnIpHYRYJfUWuhlIrzdrlKNpD42BFCtBBmuoJUmHMm5Qp9BCqHJmV8yxwZB5C+mCP5rQOOBJxmXEslEG0TRSrqMoYCR+w8IZzih1+npsIwrLn55EPm8CmpXJD3W6YCRQKhW4HzgIVJ7cvkFkupR2+CCi3XtXkatCAVnIuoGqids9fnrCCO9957j5QSv/brP2CI3jwvXhj6FeI9FWGeJwu5akHz3AohA84VA05VlHJ8GMxpsqLAGADPnBLjOHJ5ecPFxSXXV1suxh27eaSOhTknchp53Ak/etPzndNMLjNSlFDBeQuDHwoYa62glT5Y3rkPAU0J8aYqk9t4TzVTqnlcfIxUAg5B5lf7R8vCkYUj35QtHPnF5MjCkIUh35QtDPnFYsjLSl0L8DeAfxX4Z1X1H72wyf8FJOCfB/5W+5vfAN4F/qBt8wfAfygib6rqoSPRXwSugD98mePJJVOctJVpZk4ztZqCBdVUM2gFg6itqI+KKnMil1YEVq2wKqdMzgXvPFTLeTzMP0VBLXSoWtDqqCWTtJJSYd5ukWnL+vqCszLjagRt+b/eQpW2uuYYMn0lrf0v+ZOf5sERlfYHylGmhENu7e1EsxW1NRP7/FZ3zv3u68oLW71wUPJFVZdXB5l9jrvzcz2EmsWOQeuhCFSpPnD+9jt4Xwl9z+XTgZWsGHhG2X7COO7JarrxXTvn5A+9uUHEOmbXYqo5Inax9JBR3a6Zdx7V2pqvcRxX4zTx9OlTHjw8560nb5BSofZQVIkuEH0kpUxOE7UktEzHEL+021SKhXfxHhesR4MLgVQKF5eXXN9sub654eL5FdfXN+x3E7s8Ual4DRSUosLVPvPZtvBgHahZ8MVyZF1WQpSWEmHn5ZxDEKKPjNPMOM3kClUcuRRyCTgFrwVHMulOJ1BvOy9/VVs4snBk4cjCkdfhyMKQhSELQxaGvA5DXjby8/vAXwH+FeBaRA55sZequlfVSxH5b4G/LiLPMIj8DeAPmroKwN/GwPI/iMi/j+XW/mfA7zevyle2m6tLSh+p1VaNPjjIQi1tgouwyxPeeYJ4xnFCXLLVZK3HIkIvVoAmePp+wIlF9CiZuRiEUEVCy5+talJ91VEr5HkipZlcwM3XdGUCVmRVPErVFwFzANDXZ4dg+GHda4NXDEZyyPBtg0scJWUrerOkT/MIiRXymYa9WIhVrbDMJlzLya0VpzZAq2DdiPX2+9d6alLa+bnbEwMEB2qviYA6KxisCvX0FOR7nDx4G330Ken9PyXKHxMutsj1FbVkohPW3lR5ihPz5tRKo0cbD3aeKkrRCgS82r0LsUcRtGS8hKaeYtdyt9/xwfsf8vjhG9Qi3Ewjstuzqh7nLCUC50CVeZrxTnE4ViGiCqMK6gKEgIrjZj+zv7xht9vz4Y9/wuXlFdM8M04jc0rUrDgEL45KJVclF9iK8Mnc86ics3ErnGxxbkJFSbUg3qNScd7hRalFqRTSOJtKjDpmhakIc/W4kgg10VPR+ZrseqgO//L52QtHFo4sHFk48jocWRiyMGRhyMKQV2bIyy5+/u32/X974fW/Bvx37ed/Dxv/f4s7jcUOG6pqEZG/jCmq/AGwxRqL/ccveSxsry7JQwTg4cOH1km5CqWYLrgIXFzsEFdZ+4jmQkWJak3CEJCmG+5DIFZHTkKak00eYJ4m5nk2ZQuVJglo+ZXeeyiW1zusVgzdGXAoEDv6OX6KvXyRoYId8/GX29/vgqa9THGlbSyAR7QeP2l7fcl+uwNgtV6zXq8hVvBy9MKYl6GaN4OKiDVhKynjxXJyi0CM0cCNP163n1fI2ok9GE5OTxEVHj94g8dvvMH69ISnzz6Fzz5j6CK9KwQpln+LnVdGqM5TioXftYW3y+H8VSkYlPouohqYSmrhWOvHEFJCRPjoxx/xvXe+y5tvvsmUElozKW1RrOjQIaAekb7lvFaqZhDFxQ5KJeXCdr/jj/7kfT777Bk32x3XN1tS2wdi+cAOh4sRXOskXRO1ZDJqHpe9EE7POO17Hvc7ppsL0jQhEtu9DRTXPIm1grPvVZVaAqUIWor9XrHwe82U6hCCFfu+nC0cWTiycGThyOtwZGHIwpCFIQtDXpkhL9vn5x87ilR1BP7d9vXTtvlT4F96mX1/mcUYiDEePhPvvamd5GwhNecoqtRU6LqZvu/swlUoxRqHWcOmFSFASplSzPFwG4oFcLa6b/tR1WMBmOUoqoWxNR6VOyyE69CXTAn6KmYh0c+/dhdCx7ekotZVAFSah8L0Q3bbG3784Yf0fc933v4O2vXWL0C0eVQaMVFqK9i0vGErPjt4s2otLd/0tYrevzaTRnoRj4gHL6zPH/Hke99n8+AJtf5DOl+JXnBUgvMEcRTnSdWR1VRMzJdk3idVPV7wA4hrrYjcdqEOMVp2Anrspvzjpx9zenZK7HssYH8oMIygUKrD4QgxMM8j+/0O562w8tPnFzy/uOSziwuefvqccRxbMzyzWqulCIgcw+FZldLC4gd/3n4ufHKdWA0dj9YDnc/4oWdkolY7f1HwEvFBrCgS8BJI8+G+WhM0XEuXqBaYN0+ieXdexhaOLBxZOLJw5HU4sjBkYcjCkIUhr8OQ1+3z83O1GCJd7EAgpYRWIYQO5yu1eUFiiORijZqCDxyk+3LOzTMglFIxnXSo1XwXVRUEYtfT9T3e+eNEtoZOtcGntgHuqM662FaLR5uihQstTHoM1t4JOb+63XG2fP6Fu++pGnRaWNi2tsLHYehx3lmIvElnUk0N5OiUOXiOMA/BYVw57wk+GnOqP8LmsO034Wn5sn1pg6l5uSzHuqhQfQfdirg+px/WhLyl93I8pyAO16+IYeBqP5KK9U+IznJ4wULbIuaJQm/BG7uOeZoJ0fTxDxNexHFxccFnz5/z9ttvkecJLRkNgegD3jlyKhQqNSX208x2StSaGeeZP37vAz799BnXu60VrQJaPn/O0v47jK6iSm3vew+CJ6vn2bawWhXe2vSwMqlNcZm5NTArpVizPHXMOZFKQaunFJjmmZILiN1n5wUtFdFizcmAl1z7fOts4cjCkYUjC0dexxaGLAxZGHK/GHKvFz/iAiH0iAjTNFKbXn0M/thxtouRvuvw3jMMJ+aRKZVpmpoHAUpucnvq0eLRAuKV0AdONhtOT07ou55SMtubG66vr0nzjIiQ0sw076liaZphPzKPI3E4wwVHzhnnHcgBNXL0YvDS8/IWGo1dX2rHDsDawsYvfooIDx49wofAhx9+yM1+x7DZEIo5FUw+8RYiB29K13WHDzjqqbvW5ffFosJv2g7HUFHmZGFaEQHnmFNhVIfv12xWG/pcWPeJMSfybgK/4ux0w8O33+XDp5/w4ceJmpvMY7WH0SFkL15uJ7x4uugoxfYtAuM0AsKUEtfX17z//vucn/Z4zgg+4kS4eP6ZFcXOiTB0bWInLreJ588u+fDjp/zk6ceW390ekO35h8PSHFS1FdGqAVCs07cHPBVriebvtDEUAAAgAElEQVTIEriahPx0y6N14AdvvQlzRBEk7KBkvAipFMbSpDWLkEtB2wNnGDrqbPKTURzMGZcnnIPgBiTea4wsHFk4crSFIwtHXsUWhnz5VgtDFoZ8Wxlyf2kDmBq5rTq7YUUMoeUs/v/tnXu0LFld3z+/vauqu8+5575m7ryAmeE5oCgoCoIoL13CikkwMb5igD8wIdEYNS4QNUuCK1HUaKIgSyOGmOVj+QorqIBiFgYFBwFREJCBmWEe9859nlef7nrsR/7Yu/v2PXPue5jTfe7vc1etc6urumrvqtrf6v377f37gc0xz0MIWGsoyx5FUeVETgHnPKHr0vbcyCIR79I2Iyl7bq8/YLC8TFkU1KMxEbDWEIuCSMRGQxlKrAgllqpMVgsjkc65acM8N6vy5P+XpzgSYVavIDJp5+dYHSaHDpKtLBMRivn/AmKoen1uvOlmxuMRm1ub9MMSVb+frVLJzT6xKuUVAmCzyIg5e/xZzmcJ2f7Z1bDTcSbi6HN0HCNC3TZsjcasb2xQj0cYIoZkjRMRjDWE4IjBs3/fMltdx0MnT4IJEIUohojPYw5iPm+ywgRiTjBWpvHXUTB5wqGIpDGv3ifxEktZlvR7SzTNKuPxCB8dvgl0DrZGDcdPrHP69Dpr65s4H8FYQowpzn06PbYwGFPC1P2fcwBgUo6BOHNt8p8I1J3nzGbDiCWWB5YYHM1WjQnpBR07TwhJaEpbUpCuZVEkMUqWqZjH3gYkOkwsidGDLLSMqI6ojjzsvKojqiOXg2qIasj286qGzLeGLK7aAC6Aj0IhJolKVSV3aUwPU2GF1nuij7TS0XYtresILkVN8d7Tde3UHOG9p2nGOOcoYknZK2nqlqHZQiQyGm6xtTUkeIcRQ4jh7KRDWxBcZH3tFNdvbdI/dAOmTDHYRZLoTO0R08Z/ZYIDSWxjjCmSymzjk4mzFaIUaZyvkKU5EQPJopRHk54+fZrRaMjBQ4fZf+Agvf4SZZksVJB698akyCyTWPy2tEyGn07OfzYB2+UJy4UE6lKPZUxyBXcxpgmhYhACIXbUBvCO0DaY0FFIxHUtTWyJUZAYsSL0+30OHjjI0tIyTd0QfCQ6n4QjeuxMtuMgYLLFLI15LZAYkaLC+o4QZDoGtyx7LC0tU5YF/UFJ01U4X+CjZdR0DLca1jfHPHD0OG0bCAgiFhc8k/HiyZ+b3ejTcdARsUloQgS6DoiIdzlaDGDCNHTkmVHH506NefLjjtALnnZ9HYmO4F2OlGMQSrwHYhoyYaxgY8QYmyY/BoAO8Y6iiGCg22VL29WiOqI6MkF1BNWRK0A1RDVkgmoIC6EhC935qXo9+v1BGv/oHF3XURihLC1FkSYgjrdqtoZDxt0wJYYCvO9SL9larIWm3WI09tmCECmrFKqyHY84sbU5HVdrptFDkjXHGIvkHnevP0C6ElrBN2NwDcGWU2vLWfKkvyvAIPgsICEGCB7nPf1eD9elHrEpLCKk+OpikvtdImJTJA6JEFykazwnHzrBXX//Kc6cPka/X+DrMYOqx8rKCpE0gawoSiTGlDyrdamxxUgIdRL43NOW3DBmhSK5X88VC5GzWaof6fG4IoLYMk2wiwHXdTSjmnY4otlcpYw1B5ZLpBHqNub8CIHgHcPhBqOtIUuDFW644QYefPAYnXP4EFMWYQyQ7qWPAZvdsOmFY0nWHUNVWjxgxEHwXH/99Tzh9qdy/ZGDDLdWGY2HlD1hIEuMa8f99xznoYdOcXptg86lyarpuheUtsJYm8aQO58mA5oc5SamiDdWhBgE16UxukjACBTGYE3KK1HiicDqVseHPnuSsHwDX/qYp+BW12nqmsKPWO5bxBb0ipJu3OBcg3MdoTA0XcuoFoZjQxcrEEF8g4k9vLf4q0kVPgeojqiObD+26ojqyOWgGqIasv3YqiHzrSEL3fkpy4KisIhkN5sEos+TB83E3WbTQxgdLvjkVvSBsjB4ST3WiEdMQJDUwCSNh43RYOIks6xLEwqzlSPkHnrqUee0U8ZSGKFrxrTNmGJpBef9NBvzWYQrER0xMvVSGyO0bRr36+ocLaMoKMoepkrhM33waRwsPoVIjJMI9YLrWtbOnOHkQ8dYXzvJvuU+/WqJZjwmOEfZ6xPF5HMZ8BEfOpx3EAVbFkTjiWV2pe8gLBNL1OQYE7HZSYjgXKvLlYpRSiqXJs9ZY+j3l9JEuvoAJ3olQxy2EIy34Fx24SZxqsdjlvr72LdvhX5vla4dJVe6cI5LfTa0Yyp3up8iUFhLYXoQC7xruO66w9x88y0Y69kaCZGA2ILONZw5vcHxh06xurrJeNwSJdC6muihrNKwCGA6iy8lfpu9RmctXCn586Q8Mb8wIjE4AgZblFhbMO4i9x49ze1HDnPkMY9nXWrazY7OjTDW4GOahFoUlkikdS0hBMqyj2kgBoOIIfpkcSl7QhEXWkZUR1RHHobqiOrI5aAaohqyHdWQ+daQxVUbILgW1zSIGHq9HDEketKzmiaXiS0YLO/DZLHpOvCho64bQkjWl7KcjCuN2TWcklaJWGxhMUUBRJwLhBDpug5IUV3KoqAsSuq6SWMT65oH7/s8rrePWw4eSUI4Q35cLtPJnL+bXehBYBKlJXSOU2fW2Fhbp9/vc+DQYfYdOoixBlOUuJAbjID3AR8jOMdwY5XVMydo6hGFMRACo81NhqvrHDx4CBGDKSuipImKhpSJWELA+yRu0QdiETHp4s2UM57jdo4x0rYtMJOTgMn1jg8Tmu0TFi8kPttd0jHEnEPA5xcDOBdxbUd0HcF31F2dhhj4HPsmBKJzBO/p9/qsrET27dvPaNQi4tI+UWbeE4Gcom1a/hBSZmprJbtsI8GnhGWj0RAfGkb1mBCErXHHsYdWeeD+o5w+s0HdOnzwiE0vvUgkBo/PYVKjT1YuHyPeuVznmDJhi6TJfkWyCPks6D6EFO4yv4xNThzXec+p02scPbXJzU84Qr36ANIsIfm4IYRzxhVXvSVsX2hbsG1E2nyxgyMGx5W8OOcN1RHVEdUR1ZGrQTVENUQ1ZLE0ZKE7P5WFwgR8iLiuwcQ0+Q1jCcbSdh1dm8aFSlmxb3k/3jlGW0PObJzAB0dRGKyV6cS6IltGgvd43+XF03YtzvscijFMLQe9Xo+qqmi7IdJ1XOe32HdwPyuDChMDIabe9dl+8pW7ml0IiDXTI/V6ParDBylCYP34KU6dWeOBz9/HuE2Jpw4cXuExj70NU1SEAE3bsDXc4OTxYzx4/z0cP/oABMfhgwdZ6S1TWkszHLJ26jTLnWOwskJv0E+Zin2AzqUMuiHix2OCLSBEBoPB1IIS4mRyWkHbtlNBsNZOBWUiKrOiNLHGADuKzqUQY8qELNGnSXhBGDeB9c2G1dUNNrc2aLpRikBiLDb6ZJWRNB57PBoTYmBpaZkDBw6wORzjQiAEl137yYox/RfJ7mbSy02EwWCQBMAKh/av4EPHPZ//FEXZY9x2jLbGfPquezl2/ARnVtcZjutUbxOxZAEJFu8c3vup+z7VLV2ndN1SpucYIjZ0KY+DkOPtF5j8DogUYEpcMBAc+Ib1uuW+E2t8w3PvYMlvsCYNW6s13jvEQq/qpxdQ07FWC3WTQrWGaGhdGnZBDIS2gaKhi9uHUywWqiOqI7OojqiOXC6qIaohs6iGzL+GLHTnR8zkwXU457BOKKuSLjikjXSdTxEyouCaDkJMVghjMbaYuomNsQyWlhj0+1hraduWrm3pcg9URHDBpx55DrU4SWhmjKHrOpyPWAJRYG19ncHJUxy47Y6UpTbC2aGIV27hOtsQw7RxFmXJ0vIyR45cz/79K9Tjms/e+3k2Njc4dfohHnjgGFXVx5iCzdEWW8MNgqsZba1RN1s459i3bxlMtt9IMil0bYdpGmxRJJd6iMQQ8Dmjr3MOigJjLf1+/6zlJAR8SBM4Y4xT68pkotvEzDTZvyiKh43PndT1iq5RtmT5EFKUEjFUVZ+irFKo0X4FPqZwjEHwdYPzybIRgqcsSsQUDAYDelVFXVQpWV1IFpHJWSCCpCgsIQS87xAxONdhbYp0UuzrURSWutmih2E8rjl5ao1jR0+wMRzhfaTf7xFCyvYt08sT8nUMhEhKWicpBOrE/ey9I/ouR5NJicQwOat1vpbGWCIWxGBtAaEFIsE7RnWDsRW3POZWBmGDY80q43qIcw1t12AQXOdpmkjrwbVj2raky9YWK4YuW+C4gpfDPKE6ojrysGukOqI6chmohqiGPOwaqYbMtYYsdOeni2V2LUagBSkwIlix+OAoJYIkN2HhfbKu9JeolpbZ19WMN4e0bU1A8IXFlQXeRUJMY3fLwhKiJ8Tk2nU+4KOhLJfo9fqE4InRE2NLaTymMIyKxxDqhlEtxOAprIHoiKYgxdVPcdBbY+hdQcJlyQESQYjREUJyCVfLvTxB0rLUt7jWMhw7zhx/iBhTll/fNaRMwBHamiKPKzaSLB1SGFrfpTCMhaGsDd4YglhiiOTEyrT1iOhdsqAEhyz3U2OLEYxBYqTLruUqWx+889jC5nInEZ+4nGetLtN6XqLgTPabCDEEmhDwId13EwXB0nlPgeOAhboNDDtPKwFfCBIKwCM4CkmWmF7VwxQFLieG86ElyhhCRKgQaYkxZeiWKBTW0PmWiCWGEpGCQVHQKwKhKRhJ4NipVe6+7z5WN9YJLoUxiQKhzHfWg43gJExfpMl6mF6KMQZcCMSYhNXaCmMhIhS2TO7/EGjbBh98Fj5PDC0+TMadC9EIW13NZ9ZKnva4L6EvhzncVbiN+9lYvZ8zdUP0BcG3IDVdYSlcRdc5hEhVVTjTo4uBMmyyssATlUF1RHVEdUR15OpQDVENUQ1ZLA1Z6M7P5toqvV6PwghL/TJFD/ERa4Wuaeg6T68/yOMoRxAsTeiwy/s4sDygJ4GmsRSVBdfQbDnwkdAlN5/3HjFgbOrl9vtp4l1hq2lcf2MM/bKfYo8Hx8ryPsygY2tUpwdDJhMWJ67Q1KM2F6nbjshkMpsQc2Zf5zuMFcqiwFtD27b0ej0OHDhAb8njvKOpxzRNjXMtVtJ43dFoTNO0VFWFMYaNjQ3OnFljed/+lGhs0Kfq9xkMllNOAlNSlhXOtfi2QwgYawnryX504ODh7DJNWX4lRGL0uKbFlAVl1c8uyjh10+9cxct7AZ5rpZlkPnbJPR4C4/GYrdEW/cES+w8d4czq/fhxQ9d2dNFT2ALvHbWrWV1dZXNjg+VDA/r9XhIdhoi1xGDpAogYJrmbJxmXrS2YBGHx3mOLZInrLQ0obMXSoUN88q67+dRdd7G2sUkIpCzGInTeEUPOjOxJ45rzU1IU1bRmziUrX/Ad3qfnrqoqyqLAGEsIyYqSxHwyrjq5iIkhhwhNli9blWxsrPKud72L5jlfxq2He3TB4rwh2ApjumSlKQtik561xhuqaomil66riQ4TPGUI3HzY8JHLumvzheqI6ojqiOrI1aAaohqiGrJYGrLQnZ/lpWXKwhC8p7QV1hqca4ku4pzD+0DbpnGyhU+uQucihRF6ZY9eUWKRHB0wuY+lgC7C2Dmc91S2oip7RDqMz5FU5GwEEGuFED2+8wTvacuWEqGqUu/UGpvDUgoiyRIRSV7Ky3U7T+e4EZPchUjwHmIASanLIpHWO1wM+OCoqhLvWpoGnPO0voPs5p24yU+cOJGExxZsDIecPnOK/mBAf2mZXm9Avz/g+uuu5/Dh65OrM0a867BFgYTA8aPHKKRgeWUFoqTJcTEyHG6ytBzY1z9I27YURTF11T+sbjNCc6HoK+fbPtknxuTXj1FwLtB5RyBSVH3K/j5MMcAUW9iiwI9bOhcRyQm8XJcabIxIFMqyxJgUdQQkN+zJxDqTopqEnDJOwOZQos61CPDZu+/hzOopbntSj6MnTjBqW0IUHEwteIWtMDa52kPXpkCQJg2jAKYRaSb1NpKsLiGEaZ4DkUDXubPRVrLg5FHLOTN0snJ573G1o+kaNssl1jeHFLdcR1w6AP4QVemp3UOM6i7fqwLjI010GFsQg8O7jsIKnfMYH7h+aXGHq4DqiOrIw/dRHVEduRxUQ1RDtu+jGjLfGrLQnZ+qMhSmoA0hR0lJ1gbvW3znEZOSixVFSYHBxkCFMOj3qMoewRYUztF6jxhDUVRYYxEMPkbEpky4tiwoYqRXpQl3ksMupjGigg8d0VhMJ4xGNSZG+r2K9fUNrr/pRkIKts50omEAshv8Qjxs7GmWnBTbQwgTl2Se/OhcS9s2lGWZH8iOpqlzRuExEAkhR+/w6f/ps0Bd1xhbUOXedNO2jMZjBv1lqqpHXTc0Tcfy8jJVVVLkyY5VUVAPa4bDIf1+n5CFLIZuKjyha6nbjl5vkLJXy85jai8kJBda374tBqGuW7aGQ0IU+ks96CAeuJ7NfYfY2FgDayl6FZ0LxC5bxWKKmjMcbtK2niJHMCEmAbA25VII4eyY5GlJYhL76QTKGPGd5/TqOt2993JqbY0oFiksoU2TFgWhV5VTIW6aDokGzNlX0cSqNr1G+cVlbaqrcy5vixiThGZi2Qsx2YUm1q1pNJjowcBwY5O1tQ1Mb8DguhsYtus0W2fw2YqESeOpm9YToqEwluA6QttS2InVxXKgt7PlbFFQHVEd2Wlf1RHVkUtFNUQ1ZKd9VUPmV0MWuvNjBPpVQa+wGJOqUlX9FP7RhrNWERFMUTEoKpquZbjV0O9belUfMRWxc2CFYNJkLtOzVBgK12FMihpibUFRQOc8diJE1ubxkCVlURCdY2triO86vLVsbm6xtG9M1e+lB0mSZBi5qNZMG9XUggCIOdv7NhF80xJ9RzsaUY+3cE3LeDSiNIZR2yVXr0vjYXu9Hm3bJm+1gaKyiE/5AnxwdF2LuI4ocaYBWNq6xtqS9bVVTh4/zr79Kxy54QiPu+02in6PUiw37V/BmILRaERhLF2XRG9lsISJkc2NDcrBEs61lLZKomTt1MowW+crnVwI2bXqI1tbLfXY0XYOBMQWFFWfauUG+oceQ7m+Sqi3CB6CzwLhHb7rCC5d06X+gAMH9tM7uYY4j0RDh0v3z4IPZnovwiQKS27cRGjbmv3799PUI+6+734ihqLXTy+1QX7ObBIx7z2u7fAxjaIOXhCTrT1VD9e1+RwRaw29ymJyvgefIwEZkemEzcm+MaaJhsaUeSKkz9GBIMaO1dUzfODOv+LwkUN840u/BnfkBk59bpmjD95NW49oXKDtPKUYip5la3MLG1quO1Cx1cH6Zk0IsNxfXIstqI6ojpyL6ojqyOWiGqIaMotqyPxryEJ3fmJM8cYl9zonYwmlsNAld1yIKQuxKfq5p5xCRxqxGFuCpOgrKe680CstRWEpypJRXaeY58bgvadrO5q2w1g7decZkyYkTm5wv99nHCNN09LUTcq+bNPkulzqS6rX9vWJVQWf6pXmCY6IPiXFck2bLS6O0Wice+JdsgQRUxSRnOjL2hIRKAqTrTKptx5CSBYSnx5KUwqElJQsOI/3gc51tF2HWMvNN9/MgX0rVNZkC4ChqccABOfoWpfiuhcF0o9nr0BM/99JbGbdque7JjttSw0v0DU+i2XL+voG/X5Br78MtsSWfVwsGLc+2ayswQIVBuMCYpIFLQLOdZTWUpSW1iXLXVkUtCFNDhQsRgxePCF008ZcVVW61SKsra3TNC3O9AFLCCmUZGHLaV3btqZpGtq6wWDBWowIVa+iLEtijDR1Cv04edFNXPWTq2DEEIKnaZppmM7JJM7tlqwQAohPLz0TGY7GfPIzn+WlL3sxrfTopKJXDdKLzLeAYAi44FkZ9AFhsFRy7/EtXPRgK6rFjVALqI6ojsw8B6ojqiNXgGqIasj0OVANWQgNWejOTwhx2uO0eMoyue2cczRNAxKpemmiVtc5JKYGZI1NkVVCICC0dUPrHZ01VAf3A+Bz423aduIgzpPkPETJWZZTtImiKJMYTF2NKcRiN66px2NsWaaxk3I22XMh5L77w9mp4YUQ8E2Dze50fIAQ6LqGZjSm69JDb60l+sB4a8Ta2hp1XdN1HU09xhhLUUzCaiaXpLUmP5yGrmtxbUcgRecojEUsTMb1hhAIbcfm+jr33utZW1vj1ltv5fbbnoC1EVwKk+jaDqlKEHBdx1JVMdxYZ/+BQ3j81CqQxq+ereNsnbeL0SSfwWTb7D5JkJMLtigMw1MbrK6ugkDVqxgMKgrTw/sVDt94M2dOHGJj9Sht0+KDoekgUqQkcWWZRgKYlCBs0B8wHDXpHmSrW4gdIRhSEjohSpg28IkYRIGuzpFeSNFSjAdTWLrQEJm4ilsmWZOLymaRsNP6TV4Sk7pOxtZOwneKpIRmMabnLoRwTjmS5SVMLUPp/huS793gvOPM2garGyOsEyiWMNZQlQV16yiLfH+KSGgjhbUsD3opJ0UhxOhZXlm6vIY7Z6iOqI5MtquOqI5cCaohqiGT7aohi6EhC935SWNqPW3TcHaCVepFT/7vJtElaBiOa0LXpgyzxuARfDQYW9Hv9SltycbqKdqmYVyPcC7f7BCn2ZNTZuaGut6CPOI1RIgiGIHQdQQJ2MLy4P33c+rkSW667XHceMvN07JVxqTkXJfoVp00OBsdo2ENMeKdZ7y5TtM0NO2Ytq0ZDkc0TcOZ06ucPHmKYb3FqBkTnQcMZVVNbT1JcMJMw7VU5QCLpWkauqajbVqMGVMUJVXVR6wHsdAKw9GIE8dPcM/9D3DvAw/y2Fsew8EDB1iqSoIPtF2DazvcJCQisIWhiSn51lQ47FnrANOynbs+uQaTbQ+/NjAajXCuYzzybA5H1HVLUUI9HNJsbYH08E3HRuOhGBClpG48IUbaDsadZ31zyOZwk8NH9lFWS/SWDIdHDRvDIeNxTcjWphg7fDRJNDhrDbO2xLkU4jIK1K2nbjtS4xYMOY+AsUSJ+BwOsxBDkSerBh8wadriVCggYIzQdQ3ONdN6TyKmhOCnQmTtJAN4N3UvMxFvkWSNMZLGk5uUofvEyYf42Mc+zpc//emsXPdYqsEKJnb0nRANFBZsEVnqD7AUjMZjHn/bTdiiohltUZlLe47nFdUR1RHVEdWRq0E1RDVENWSxNES2uzUXARF5DPDAbpdDUZRzeGyM8cHdLsSlIiK3A/fscjEURTmXhdER/S2iKHPJRTVkUTs/AjwF+DTwWGBzd0t0VayQxHPR6wFal3nl0ajLCnA0LpCgiMh+YB29x/OG1mU+UR3Zxh77LQJ753ndK/UArcuVnOOiGrKQw95ijFFEjuXVzRjjxq4W6CqYcZ0udD1A6zKvPEp1WeRrpPd4jtC6zCeqIw9nL/0Wgb3zvO6VeoDW5Qq4pONeUXJfRVEURVEURVGURUM7P4qiKIqiKIqiXBMscuenAf5j/rvI7JV6gNZlXtlLdXkk2UvXResyn2hd9j576brslbrslXqA1uULwkIGPFAURVEURVEURblcFtnzoyiKoiiKoiiKcslo50dRFEVRFEVRlGsC7fwoiqIoiqIoinJNoJ0fRVEURVEURVGuCRay8yMi3y0i94pILSJ3isizd7tMF0NE3iAicdvy6ZntfRF5i4icFpGhiPyeiNy4m2WeICJfKyLvFJGjudwv37ZdROSNInJMRMYi8l4RefK2fQ6LyK+LyIaIrInI20Rk36Nbk0uqy9t3uE/v3rbPrtdFRF4vIn8lIpsickJE3iEid2zb56LPlIjcKiJ/KCKjfJyfFpGFTH58uSyajqiG7H67y+XYExqSy6E6chUsmoaA6sgctb09oSOLqiEL1/kRkW8FfpYULu/Lgb8B3iMiN+xqwS6NvwNunlmeP7Pt54B/CPwz4AXALcDvP9oFPA/LpOv83efZ/lrge4HXAM8Btkj3pD+zz68DXwx8PfCNwNcCv/yFKvAFuFhdAN7Nuffp27dtn4e6vAB4C/BVuRwl8McisjyzzwWfKRGxwB8CFfA84JXAq4A3fuGLv7sssI6ohqiGPJKojlwhC6whoDoyD21vr+jIYmpIjHGhFuBO4M0z6wZ4EPih3S7bRcr9BuBj59l2AGiBb5757KlABL5qt8u+rawRePnMugDHgB/cVp8a+La8/rT8va+Y2eelQABumZe65M/eDrzjAt+Z17ocyeX62kt9poCXAR64cWaf1wDrQLXbz9oX+HotnI6ohsxlu9szGpLLoTpy6ddq4TQkl1N1ZM7a3l7SkUXRkIXy/IhIBTwLeO/ksxhjyOvP3a1yXQZPzi7Ou7Or8tb8+bNIveXZen0auI/5r9fjgZs4t+zrpBfDpOzPBdZijB+e+d57SY30OY9SOS+HF2a369+LyFtF5LqZbfNalwP575n891KeqecCH48xHp85znuA/SRr0p5kwXVENeQs89DuzsciagiojlwSC64hoDoyj21vJxZRRxZCQxaq8wNcD1jg+LbPj5Me+nnmTpIb76XAvyY11PeLyAqp7G2McW3bdxahXpPyXeie3AScmN0YY3SkxjFv9Xs38ArgJcDrSC7ad2W3LMxhXUTEAP8V+IsY4yfyx5fyTN3EzvcN5u++PJIsqo6ohsyw2+3uAiychoDqyGWyqBoCqiNz1/bOw8LpyCJpyJ6fkDgvxBjfNbP6tyJyJ/B54FuA8e6UStlOjPG3ZlY/LiJ/C3wOeCHwp7tSqIvzFuDpnDtuW9ljqIYsBguqIaA6ck2gOrIYLKiOLIyGLJrn5xR5XOC2z28EHnr0i3Pl5F7wZ4AnkcpeicjBbbstQr0m5bvQPXkIOGcSaI7icZg5r1+M8W7Sc/ek/NFc1UVE3kya6PiiGOMDM5su5Zl6iJ3vG8z5fblK9oSOqIaohjxSqI5cNntCQ0B1ZLfb3qUy7zqyaBqyUJ2fGGMLfITkBgSmbraXAB/crXJdCTkc4RNJE/Q+AnScW687gKuqT6gAAAnASURBVFuZ/3rdQ3o4Z8u+nzTmdFL2DwIHReRZM997Men5u/NRKucVISKPBa4j3SeYk7qkiJ7yZuCbgBfHGO/ZtsulPFMfBL5kW3Sirwc2gE9+ocq+2+wVHVENUQ15BMqmOnIF7BUNAdURVEeutlyLqSG7FRHiKiJJfCspescrSdEufglYZSZKxDwuwM+QxmzeTgrl9yfASeBI3v5Wkuv5RaQJYh8APrDb5c5l2wc8My8R+P78/1vz9tfle/CPgC8B3gHcDfRnjvEu4KPAs4GvJlmafmOe6pK3/TQpZOPtpMb6kVzW3jzVBfhFYC0/UzfNLIOZfS74TJHGrH+cNLHwGcA3kMYQ/+fdfuYeheu3cDqiGrL77e5idVkkDcnlUB258mu3cBqSy606Mh9tb0/oyKJqyK4/zFd4sb8nX8iG1MN9zm6X6RLK/FvA0VzmB/L6E2e290njJc+QYtP/PnDTbpc7l+2FuXFuX96etwspHvtD+WXwXuAp245xGPgNYJMUvvBXgX3zVBdgkBvfCVJoxntJMfNvnLe6nKcOEXjV5TxTwG3AHwEj0gvwZ4Bit5+5R+kaLpSOqIbsfru7WF0WSUNyOVRHru76LZSG5DKrjsxH29sTOrKoGiL5pIqiKIqiKIqiKHuahZrzoyiKoiiKoiiKcqVo50dRFEVRFEVRlGsC7fwoiqIoiqIoinJNoJ0fRVEURVEURVGuCbTzoyiKoiiKoijKNYF2fhRFURRFURRFuSbQzo+iKIqiKIqiKNcE2vlRFEVRFEVRFOWaQDs/CgAi8ioRWXsEjhNF5OWXsf/t+TvPvNpzXy0i8sJcloO7XRZFWQTmWTcezfYsIm8XkXdc4r5zo3mKstuohkzPpRryKKKdnz3C5TQcRVEUUN24XC7wo+PfAa/ahSIpyq6iGnJ5qIbMB8VuF0BRFEVRFpkY4/pul0FRlMVFNeTRRT0/1wgi8gMi8nER2RKR+0XkF0Vk3w77vVxE7hKRWkTeIyKP27b9H4vIR/P2u0Xkx0TkkehEP1VEPpCP+wkRecG28z5dRN4lIkMROS4i/0tErp/Z/lIR+XMRWROR0yLyByLyxG3HeJ6IfCyf48O5rhdzez9fRN4vIuN83X5eRJYfgfoqytyz6LqxrQxvEJGPbfvs+0Tk3m2fvVpEPpWP+WkR+Tczm+/Jf/86a8f78nfOsX6LiBGR14rIZ0WkEZH7RORHzlMuKyK/ms916+VVX1HmG9UQ1ZB5RDs/1w4B+F7gi4FXAi8GfmrbPkvAjwCvAL4aOAj81mSjiHwN8GvAfwO+CPhXJDftjg0yf+d9IvL2SyjfTwP/Bfgy4IPAO0XkunyMg8D/Bf4a+ArgpcCNwG/PfH8Z+Nm8/SW5vv9bREw+xn7gncDHgS8H/gPwpgsVKHee3g38HvClwLcCzwfefAn1UZS9wMLqxpUgIv8ceGMu29OAHwZ+XERemXd5dv77dcDNwD85z6F+Avgh4MdJdf4O4PgO5+sBvwM8E/iaGON9V1p2RZlTVENUQ+aPGKMue2AB3g684zL2/2bg1Mz6q4AIPGfms6fmz56d198LvH7bcb4TODqzHoGXz6z/GvATFyjH7fk7r5v5rADuB16b138UeM+27z02f+8p5znu9Xn70/P6a4BTQH9mn1fnfZ6Z11+Y1w/m9V8BfmnbcZ8P+Nnj6KLLoi57XDe2t+c3AB/bdpzvA+6dWf8s8O3b9vlR4APbzvvM811HYAWogVdfpOzPz9fm/cCB3X4WdNHlShbVENWQRVx0zs81goh8HfB6kqjsJzXyvogsxRhHeTcH/NXkOzHGT0uKwvI04EPAM4Cv3uZ6tTsch5ljvOISi/jBme84EflwPi/5vC8SkeEO33si8BkReTLJ2vIcUsdn4tW8FfgEcAfwtzHGeua7H7pImZ4BfGm25EyQfOzHA5+6lIopyqKy4LpxWeThrE8E3iYi/31mUwFcznj8pwE94E8vst9vAg8AL44xji+nrIqyKKiGAKohc4d2fq4BROR24A+At5JcsWdIFoO3ARXwMOE4D/uAHwN+f4dt9Q6fPVLsIw1Ze90O247lv+8EPg98F3CU1EH5BKl+V3PeXwJ+fodt6lpW9jR7QDe2E0jGi1nKmf9P5iF8F3Dntv38ZZznUn+E/BHJev1c0rBeRdlTqIacg2rIHKGdn2uDZ5E6A/8+xhgARORbdtivIM2Z+VDe5w7S2NuJh+OjwB0xxs9+Acr4VcD/y+ctcpknc2s+CvxTkmvZbf9iHp97B/BdMcb358+ev223vwe+U0R6McYmf/aVFynTR4Ev+gLVV1HmnUXXje2cBG4SEYl57AhpnDwAMcbjInIUeEKM8dfPc4w2/7UXKNNdpB8vLyENnT0fbyUZaP6PiPyDGOOfXWBfRVlEVEMejmrIHKCdn73FAXl45LLTpDGoJfBvReSdpAmFr9nh+x3wCyLyvSQ39JuBv4wxToaHvRH4AxG5D/hdkhXkGaR5NT+6U4FE5NeAB2OMr79I2b9bRO4iid33A4eAX83b3kKypPymiPwUyXr0JODbSPN2VnM9/6WIHCMNdfvJbcf/DeA/Ab8sIj+Z9/nBvC2yM28C/lJE3kwSoC3SxMOvjzF+z0XqoyiLwl7Vje28DzgCvFZEfpcUOOVlwMbMPj8G/LyIrJOCnfRIP8oOxRh/FjhB+lHyUhF5AKjjthC1McZaRN4E/JSItMBf5PN+cYzxbdv2/QURsfn6vCzG+OcXqa+izCOqIWdRDVkEdnvSkS6PzEKaLBd3WH4lb/9+0nCwEalB/gvOncj3KmCNFHnkcyRX8p8At247zzeQGuKINIb1TpLHZbJ9+6TD9wFvv0C5b8/f+fZ8rAb4O+BF2/Z7MsnlvZrP/Sng5wDJ278O+GQu998AL9ihLM/L2xrgw/mckWRRgm2TG/NnXwn8MbAJDPP3f3i377cuujwSy17WjfO059eQhqwOgf9JisR077ZjfwcpsmRDMrT8GfBNM9tfnY/hgffNXMd3zOxjSMN87iVZej9PnrDNDhOegR8g/YB63m4/E7rocjmLaohqyCIukx+OinLNkQMZ/A9SlBSdLKgoiqIoirLH0WFvyjWDiLwCuBt4kOQyfxPw29rxURRFURRFuTbQzo9yLXETaezwTaQocb/DBZKkKYqiKIqiKHsLHfamKIqiKIqiKMo1gbn4LoqiKIqiKIqiKIuPdn4URVEURVEURbkm0M6PoiiKoiiKoijXBNr5URRFURRFURTlmkA7P4qiKIqiKIqiXBNo50dRFEVRFEVRlGsC7fwoiqIoiqIoinJNoJ0fRVEURVEURVGuCf4/lu3D2Z5e360AAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 1000x1500 with 3 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        }
      ],
      "source": [
        "from textwrap import wrap\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.figure(figsize=(10, 15), dpi=100)\n",
        "\n",
        "plt.subplot(1, 3, 1)\n",
        "plt.imshow(img_data.loc[nm, \"img\"].astype(np.uint))\n",
        "plt.title(\"Original\")\n",
        "orig_label = img_data.loc[nm, \"label\"]\n",
        "plt.xlabel(\"Label: \" + \"\\n\".join(wrap(orig_label, 35)))  # wraps label\n",
        "\n",
        "plt.subplot(1, 3, 2)\n",
        "im1 = outputs_s1.loc[\n",
        "    (outputs_s1[\"x\"] == 40) & (outputs_s2[\"y\"] == 20), \"img\"\n",
        "].to_list()[0]\n",
        "lab1 = label_to_name(np.argmax(classifier.predict(np.array([im1]))))\n",
        "plt.imshow(im1.astype(np.uint))\n",
        "plt.title(\"Setting 1, single text\")\n",
        "plt.xlabel(\"Label: \" + \"\\n\".join(wrap(lab1, 35)))\n",
        "\n",
        "plt.subplot(1, 3, 3)\n",
        "im2 = outputs_s2.loc[\n",
        "    (outputs_s2[\"x\"] == 200) & (outputs_s2[\"y\"] == 20), \"img\"\n",
        "].to_list()[0]\n",
        "lab2 = label_to_name(np.argmax(classifier.predict(np.array([im2]))))\n",
        "plt.imshow(im1.astype(np.uint))\n",
        "plt.title(\"Setting 2, repeated text\")\n",
        "plt.xlabel(\"Label: \" + \"\\n\".join(wrap(lab2, 35)))\n",
        "\n",
        "plt;"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "id": "8e591aa6",
      "metadata": {},
      "source": [
        "## 📚 References, Resources, and Further Reading"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "id": "664064d7",
      "metadata": {},
      "source": [
        "- [ImageNet](https://www.image-net.org/)\n",
        "- [Trusted-AI/adversarial-robustness-toolbox - GitHub](https://github.com/Trusted-AI/adversarial-robustness-toolbox) - Adversarial Robustness Toolbox (ART) - Python Library for Machine Learning Security - Evasion, Poisoning, Extraction, Inference - Red and Blue Teams"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3.9.0 64-bit ('3.9.0')",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.9.0"
    },
    "vscode": {
      "interpreter": {
        "hash": "b6b7e97e50c754c7aee36d85160e6764033ec8a20165f676e018446c78d531c2"
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}