Random_Forest.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Random_Forest.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PDgH8E3Fs1eh",
        "colab_type": "text"
      },
      "source": [
        "**Random Forest for Clickbait Detection in ClickbaitTR**\n",
        "\n",
        "---"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "41GGLUbBATEy",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!git clone -v https://github.com/ahmetax/kalbur.git "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jiHe16z7AXp5",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import preprocessing as pr\n",
        "import sys\n",
        "\n",
        "pr.current_path = \"/content/\"\n",
        "\n",
        "with open(pr.current_path + 'kalbur/kelime_bol.py', 'r') as file :\n",
        "  filedata = file.read()\n",
        "\n",
        "filedata = filedata.replace('veri/', pr.current_path + \"kalbur/veri/\")\n",
        "\n",
        "with open(pr.current_path + 'kalbur/kelime_bol.py', 'w') as file:\n",
        "  file.write(filedata)\n",
        "\n",
        "sys.path.append(pr.current_path + \"kalbur/\")\n",
        "\n",
        "import kelime_bol as kb"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AkywcANAATrZ",
        "colab_type": "code",
        "outputId": "489e860a-8310-4652-c76c-1d16cbaf8076",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 82
        }
      },
      "source": [
        "no_of_samples = 1000 # toy example\n",
        "\n",
        "csv_files = {\"limon\":\"dataset/limon_clickbait.csv\",\n",
        "             \"evrensel\":\"dataset/evrensel_non-clickbait.csv\",\n",
        "             \"spoiler\":\"dataset/spoiler_clickbait.csv\",\n",
        "             \"diken\":\"dataset/diken_non-clickbait.csv\"}\n",
        "\n",
        "clickbait, non_clickbait = pr.return_data(csv_files)\n",
        "\n",
        "special_characters = [\"#\", \"?\", \"!\", \".\", \"@\"]\n",
        "\n",
        "words_will_be_removed = [\"işçi\", \"eylem\", \"meteoroloji\", \"katliam\", \n",
        "                          \"murat\", \"altı\", \"seçim\", \"diren\", \"dev\", \n",
        "                          \"gazze\", \"blog\", \"protesto\", \"beş\", \n",
        "                          \"yaşam\", \"manşet\", \"günaydın\", \"türkiye\", \n",
        "                          \"sınır\",\"chp\", \"grev\", \"yaralı\", \"ateşkes\", \"yazı\", \"maden\", \"bayi\"]\n",
        "\n",
        "X_train, Y_train, X_test, Y_test, unique_word_list = pr.generatesample(clickbait[:no_of_samples], non_clickbait[:no_of_samples], \n",
        "                                                                   special_characters, words_will_be_removed, \n",
        "                                                                   isseparate=False, scaling=True)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "# of tweets in limon: 21791\n",
            "# of tweets in evrensel: 13093\n",
            "# of tweets in spoiler: 1898\n",
            "# of tweets in diken: 10936\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g4mks0DprQGX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import classification_report, confusion_matrix, accuracy_score\n",
        "\n",
        "def metrics(rf, X_train, Y_train, X_test, Y_test):\n",
        "  cf = classification_report(Y_test, rf.predict(X_test))\n",
        "  test_accuracy = accuracy_score(Y_test, rf.predict(X_test))\n",
        "  training_accuracy = accuracy_score(Y_train, rf.predict(X_train))\n",
        "  return cf, test_accuracy, training_accuracy\n",
        "\n",
        "def rf_training(X_train, Y_train):\n",
        "  rf = RandomForestClassifier(n_estimators=100, max_depth=None, random_state=None, verbose=1)\n",
        "  rf.fit(X_train, Y_train)\n",
        "  return rf"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "brlVYw_SrlEY",
        "colab_type": "code",
        "outputId": "caea876d-29cd-4ba8-e2fb-eb605815126c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 293
        }
      },
      "source": [
        "rf = rf_training(X_train, Y_train)\n",
        "cf, test_accuracy, training_accuracy = metrics(rf, X_train, Y_train, X_test, Y_test)\n",
        "print(cf)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:    4.1s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:    0.0s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n",
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:    0.0s finished\n",
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       0.79      0.91      0.85       184\n",
            "         1.0       0.91      0.79      0.85       216\n",
            "\n",
            "    accuracy                           0.85       400\n",
            "   macro avg       0.85      0.85      0.85       400\n",
            "weighted avg       0.86      0.85      0.85       400\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:    0.2s finished\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LR7wpDp3sEZA",
        "colab_type": "code",
        "outputId": "d0da797d-d164-4236-b2a3-9433caaf1147",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "from sklearn.datasets import make_classification\n",
        "from sklearn.ensemble import ExtraTreesClassifier\n",
        "from sklearn.metrics import roc_auc_score\n",
        "from sklearn.metrics import roc_curve\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "WORD_SPACE_WITH_OTHER_FEATURES = np.array(unique_word_list + special_characters + [\"o. special\", \"uppercases\", \"wordlength\", \"tweetlength\"])\n",
        "n_of_features_to_consider = 20\n",
        "\n",
        "importances = rf.feature_importances_\n",
        "std = np.std([tree.feature_importances_ for tree in rf.estimators_], axis=0)\n",
        "indices = np.argsort(importances)[::-1]\n",
        "\n",
        "# Print the feature ranking\n",
        "print(\"Feature ranking:\")\n",
        "\n",
        "features_all = []\n",
        "features_values = []\n",
        "for f in range(n_of_features_to_consider):\n",
        "  features_all.append(WORD_SPACE_WITH_OTHER_FEATURES.tolist()[indices[f]])\n",
        "  features_values.append(importances[indices[f]])\n",
        "\n",
        "# Plot the feature importances of the forest\n",
        "plt.figure(figsize=(20, 15))\n",
        "csfont = {'fontname':'Arial'}\n",
        "plt.title(\"Feature Importances\",**csfont)\n",
        "plt.xlabel('Features', **csfont)\n",
        "plt.ylabel('Feature Importance', **csfont)\n",
        "plt.rcParams['font.size'] = 30\n",
        "plt.bar(features_all, features_values)\n",
        "plt.xticks(rotation=90)\n",
        "plt.savefig('random_forest_graph.png', bbox_inches = 'tight')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Feature ranking:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABNUAAAQ8CAYAAAC1nahYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdebxt93w//tebm0RkkMhkiORKRZuIOa0am6JC9WsolbSECKWq6LeKtoaitP221ar6VVEkiOnLt4ZGaBNjKZUElVBTZEJkEDJLIp/fH2sdd93tDHvtc86959w8n4/Heuy11/5Me+29r5yXz1qfaq0FAAAAAJjejbb2AAAAAABgvRGqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagCwRlXVi6uqzbi9cmuPn7WjqjZOfD+O3tpjAgBY74RqAMCaU1VnCYC4oZkIPg/b2uMBABa3YWsPAACY2odHlP3Kqo0CAAAQqgHAetFae/DWHgMAANBx+ScAAAAAjCRUAwAAAICRhGoAcANTVRuq6oiqenNV/U9Vfb+qflRV366qk6vqOVV18xHt7da3949V9amqOr+qrq6qq/o2P1JVf1pV+y7RzmFzN2lPsv/gpTcttMrpRP3JFS43Tjn+JRdFWKjtqtq9qn6vf49n9+97sXaqqh5aVa+tqtOr6qKquqY/Z5+a5jyttvnOR1XduKoeXVXvq6oz+/f5g6r6ZFX9dlXdeJ52dqmqZ1fVf1TVJf37/E5VvaeqHrCMsWxXVUdW1b9W1bf6sXyvH8vvV9VOM7znfarquf3neF7f5iVV9ZWqekNVPXREWz+12EBV3bSqnlhVH+zP35X96y+uqqPn+z4n+egC3/uzFuh3RX6HU7yXDVX1G1X1gcH5v6iqPlNVL6iqm03b/qCfHarqqKp6S3X/Jl1cVdf237EvVtVxVfW4qtp5yvZuXd2/Yyf1v8srq+rSqvp638cjq6pGjO+2/Wf1scF5nRvfV/pz8aKqOnTsewdgnWut2Ww2m81mW4NbkhcnaXPbCrV5eJKvDdtdYLskyTFTtPfSJD+aor2W5JokL0lSC7R12JTtzHtOkmyceH3jlOfkrEGdoxco81NtJ/mVJN9ZYGw/1U6Sn09yyhTv68okf7LC36XJ8c/7Puc7H0luleRjS4z5pCQ7Dtq43yLnZm57xdjPJsltkvznEu1+K8m9RpybP0xy2RSfy6eT3H6K9oZ1DktytyRfXaDNF/fva8z3/qzV/B0u8V7278/DYu1/L8mhI87/UUnOm3LslyXZeZG2NiR5ebrf0FJtnZLkwCnG98IR57Yl+dWV/O3abDabbW1vFioAgBuIqnpakn9IMpxVdHGSrye5Ksm+SQ7sj++W5A1VdcvW2ssXafbgJNsPnp+f5Jx0f/zeJMkBSW7Zv7ZdkhcluUWSp87T1vezaYXTX+rrJ8npSb69xNvb0n4xyXHZ9N6/kS4Y2DXJz04WrqqHJ3l7kh0Hhy9Lt0rr5Un2SXcuqy/z8qra2Fp7ymq9gSntnOTfktyhf/6tJGcnuWmSu2TT+39AkrckeXRV3Tfd53iTdCHDGUkuSLJ3kkMGbf9BVZ3ZWvv/RozlpCS375+fn+67u0M/vrkZahuTfLiqHtRa+8/FGqyq1yaZPMfnJflm394h2fQ9vGeST1XV4a2106Yc8wFJXpHu95R0v42z+jbn3se3s+l7f/ig7ufS/SYmfW+eYyv5O1zI3um+8/v1z+fey3ZJ7pRN53/vdOf/Dq218xdrsKr+LsnvTxy+Ksn/pAv2d05yuyRzM2d3zgILrfWz2N6T5EETL309XcC7XZKfG7R19ySfrqoHtNb+e4E2X5AusBw6N91v4Op+PPtn07lNXAkEcMOytVM9m81ms9ls829ZwZlqSR6S5PpBe59Mct9MzFZJF6p9YFDu+iQPWqTdd/blH5dkrwXKHJLuj93hbI6HLTHeswZlj57yPW6c6GPjlPWW7Gueti/tH/8lye0myu6U5BaD53dJFxTM1f1SkocmufFEvVsleeNEP09eoe/S5PgXPKcT5+Oi/vFTSe46UW6PJO+daPfwbJqh9oYkt5znuzCctfXDLD7zaDiWC/rHc/rzV4NyN03y3Gw+o+jsJLss0vbTJsZ+RpL7TZTZNd2srusG5c5cot35viefmOf8bZ9kv0XqHjbi812t3+Gw7IX948fmeS83SfKyifKvX6LtP5wof1aS30pyk3nKHtx/Dhck2W2B9t4+aOu6JH+V5FYTZW6U5BHpgsy5sl9NstM87e098X06IcnPLdD33kme0H/OD12J36zNZrPZ1se21Qdgs9lsNptt/i0rFKqlm01xwaCt4zIR6EyUryRvHZQ/fbG2R4zjnwZtfmqJsmcNyh49ZfsbJ/5I3zhlvSX7mqftluRNWeISuv6P+DMGdU6aLzSYqPPnE0HGjtO8j5HnZsFzOnE+5kKUHRYou326WV1zZedCiL9YpP2DsnlI9fgRY/neYp9rksdMlJ93HEl2TzdDcBio7b5Iu8dMtPvXi5Sd/J78W5Ltp/ycZg3VVut3OPle3p9kwyLl/3lQ9rKFvrvpZuldOyh7apI9pxj7jpnn364kRwzauiZLXIKZ7jLi8wd1njNPmccNXj9zxGe44L+tNpvNZtv2NtOTAWCdWOCm5fNtk5dTPTnJXv3+N5M8pbX244X6aa21JL+b5Af9oTtU1f0XKHv5iLfw3HSXTCXJvapqnxF115oLkzyjP1eLeVi6WTZJdznbka21qxcpn3T3cPp6v79nkt+ceZTL9+MkT2yt/Wi+F1tr16SbkTZn+3T37HvhQg221r6S5KODQ/cZMZ7ntdbOWqTtd6WbjTXnyVW1/TxFn5xNlysm3f0DL1mk3Temmwk2bHeaBRGuTfKk/jytmi30O7wsXRh73SJl/mqwv3OSuy5Q7nnZdBnnFUke3Vq7aKkBtNauWuDfrj8a7P9la+2DS7RzbpLnDA49Y55itxns/9e0n+Fi/7YCsO0RqgHAtu/owf6rFgpIhlprl6a7tHHOA5c7iL7NMwaHfmG5bW5Fb5syyDh6sH/slMHBj9PNFJyz7HO/DP/eWvvWEmU+M/H8jUsEL5N1Dl6w1Oa+n+T4Kcq9arC/Z7r780165GD/P1prn52i3b8Z7O+W5JenqPPBPsBZM5bxO3xHa22+e7wN2/5auhlgc+4wWaaqNqSbUTjn2Cm+Ywuqqruku8Q66ULMv5+y6ruyKVy8TVXdfuL1qwb7d6p5VrgFAAsVAMD68eGliyTpLplLklTVbknuOHjt30f098XB/qFLFa6qn00XYBySbmbcLvnp/9Y4YLC/74ixrDWfWKpAVVW6+9bNWbVzv4oWvdF/b/Jm9NPU+e5gf/cpx3Jya+3aKcp9Mt2sql365/fI4Nz3M9fuNig/nIG2VLuXZNN475nkX5eos+T3ZKWt4u/wU1OWOy/dIgjJpgUahu6ebhbbnHdN2e5ChqHp51trF09TqbX2o6r6n2wK5A5NN8tyzucG+wcleWtVPXethaQAbF1CNQBYJ1prD56h2h2z+cz0V1XVtJcn3Xqwv9dCharqPulm8dxj5Njm+4N7vThzijL7ZtNKg0ny/Kp61pTtD+steO63gEVXb+xducw6N51yLF+aplBrrVXV6elCr2TTKptzbpNuxdA5X8wU+na/lOR+/aHbTVFtmu/JitgCv8NpPteku5xzznyf7UETz0+Zst2F3Gmwv39VfWhE3f0H+5v9zlprn6qqTye5V3/oyCSPqarPJPlIkk8n+c/W2g8CwA2WUA0Atm17TDyf9VLCm813sKqeku7G5zVDmzssXWTNunSKMpPn/t4z9jXvud9CZrkX2Ng60353ppqBNE/ZyZlwk88vHNHusOw0M+ym+Z4s2xb6HS552fg85hvPMDC+orU2GcqONfyd7ZNuBdpZzPc7e3S6mYx375/fKF3INhe0XV9Vp6WbbfeGpS6PBWDb455qALBtm+Zm6tP4qf9mqKq7JnlNNv3h/P0kr0jyq0kOTLJruhXzam5L8vEVGs/Wdv0UZVbq3M8SlGyLxoR1wwBoMjSafD5ruzeZovw035NlWYe/w+F5W2rRjmms2r9xrbXvppv5d0y6+wBOLkxyo3SXjf5VkrOq6ndWaCwArBNmqgHAtm3y0qSbL7bK4UjPz6Y/RM9Kcu/W2neWqLPLEq9vDat1A/LJc3+31trnV6mvG4Ix351dB/s/nHht8nOZtd21ctnfevsdDv/9WYlZmMPP4f2ttYevQJs/0S8c8qYkb6qqPdLdJ/HeSQ5LN4NtLszcJclrqqpaa69ZyTEAsHaZqQYA27bJ+yDtsxKN9jfhH15m9dIp/pBPVndxgskZR9tNWW/aG+WPtSrn/gbstiPKDm/E/72J1y6YeP4zI9odlp1sZ4tbo7/DpQwXqdhQVQcsWHI6w9/Zqv7GWmsXt9be21p7Tmvt59Pdd/JPs/mMu7+oqmnvEwjAOidUA4Bt239n8xuH33OhgiPdPJuv4PdfS1Woqtsn2XvK9oeXzU17+ePk/atuPm+pzcd026zc5WOb6VchHK4muFLn/oZqqhvwV9We2TxUO3X4emvtwiTDFRzvlSn0s5R+dnBouTfYn8/w8sJpvver/TtcDZOrwx62zPY+Pdi/S1XtuMz2ptZa+25r7aVJnjE4fLOMXywCgHVKqAYA27DW2rVJThocevIKNT3tLLChJ40oOwwCp/ojubV2ebr7Sc2500JlB359xJhmceJg/wlVNct5o/MLVTXNrLLfyuaB1CfmKTO8p9ijq2qa+6M9Lpv/t/N87S7X2O/9av8OV1xr7XtJvjA4tNz7kJ2U5Np+f4d0n9OW9p6J57fYCmMAYCsQqgHAtu+vB/v3WqGbaV+UzS+3vN9ihavq4CTPHNH+8BKxA0fUG84eOmKJMe2Z5A9HtD2LV2bTH/z7J/mzVe5vW/eXi71YVTdL8seDQx9rrX1znqKvH+zvNVFnvnb3SHfvsjn/0Vr78hJjncXY7/1q/w5XyysH+z9fVc9YsOQS+pmHxw0Ovayq9pt5ZL3+0tpp7Tzx3CqgADcQQjUA2Ma11j6V5B2DQ6+uqudV1aILFlXVdlX1sKr6aFXtP9HmdUk+OTj0ov5SyvnauWOSD2e61RLnDC/ZO6Kqpr0H1LsH+w+oqqMWGNPeSf41qzyjpLV2VrqVGOc8r6r+dqlL1KrqRlX1y1X1vqpaNCi5gXl0Vf3ZfIFHVe2W5H3Z/DN92XyNtNY+keSjg0MvqKonzle2D1//NV34lnSXaL5khrFPY/i9P6aqFr3f3xb4Ha6Wt2Xz9/rKqvrfVbXg3yZVtXNVPbuq5rtc+yXpAsaku7T141X1i0sNoqr26f8tPH6el19ZVX+91D3f+n9H/8/g0NX56UtcAdhGWf0TAG4YnpTk9knulm61y79M8rtV9c5092G6MN0lc7v15Q5N8qBsWp1vvlkbr0jygH7/FklOq6p/SndZ3OVJbpXkV9NdjrchyefTzdr6hSnG+7Ykz+v7vWWSb1bVaf04f3K/tdbaI+ap94IkczNVjquqByX5l77u3Op9T+rf20lJfi6re+P256e7FPVX++f/O8lRVfX2dH98n5/kx+nO/QHpPqMHZdNN1/9uFce2nrwzyWPSfb4Pqqo3Jflqukv+fiHJU9N95+a8vrV28iLtHZ3uO3nzdP9H8xur6jeTvD3Jt5LcNN391p6STYFakvx9a+2krI63Jjmy379TkvP67/33s+l+axe01p4yqLOav8NV0Vq7tqoek+SzSfZMd/7/NslvV9U7+jFekm4G2IFJ7pPkoenuf/iGedo7r6oenS403CHJxiT/WVUfSfLBJF9JtwrsTuk+yzumW8HzXn3fH59sM92/D09I8odVdUpf5gvpFr64Mt3v9U5Jjkpy0KDeK1trk/d3BGAbJVQDgBuA1tqVVfVLSY5N8qj+8H5JnrOMNk+sqr9N8gf9od2S/FG/TToz3f3Ljp2y7S9V1Z8meWl/aPskS848aa1d0c9OOzFdKFLp7rE0332WTksX0nx+mjHNqrV2fVU9Il049vT+8J7pbm4+82VvN0AfSrfwxsvTBUKLhULvTfK7izXWWjun/018KN0qjknyK/22kFcnefa0Ax6rtXZCVb0hm+57dtN0gdLQ2RN1Vu13uJpaa2dW1T3TzQKcWwDioMw4C7C19vGqum+S/5dNIfn9+225Du23pRyf5EUr0B8A64TLPwHgBqK1dnlr7dFJHpLkY+lmRy3mrCSvSXKf/jLG+dp8drrw4oIF2rgiyeuS3HWhNhYZ75+lm1X2xiRnpFvd8/pFK+Unl/b9UpLPLVDksnSze+7VWrtkzJhm1Vq7trX2e+lWAP1ANr8P1nzOT/LmJA/O5pf33aC11v48ySOTfGOBIhcneVaSR/WXRi7V3unpZi29Ij+9euzQqUl+tbX2jNbakt/B5WitPTnJr6WbMfe1dN/XtkSdVfsdrqbW2jeS3CXdvQ3PW6L4V9PN+rx8kfY+ly6Ye16Sc5Zo77p0M0Wfl24W36TXprv33llLtJP0AX1r7XH94jAA3EBUa4v+bzQAsI3qb+p+73SzOvZI94f7D9P9Efnl1trZC9f+qbZ2SDej5g7pLtm6OMm56W4Uf+XKjnx6VXVIuhlue6ULJ85O8pHW2hWLVlz9cc1dWrgx3bm/UbrxnZPkK621r2+90a0dVXVWugUekuSJrbVjB6/dI8nB6S6TvSxd6PKxacK0BfraLt3v4cB0MwmvSnep36daa0sFNGvCWv0dTqv/vd413e91x2z6zX5+ls+gqg5MN8Nsz3SXc16V7px8LcmXWmuXTdnOLdOFrxuT7J7uap/L0/1eTxvzbyUA2xahGgAAa9JioRoAwNbm8k8AAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGs/rmN2HPPPdvGjRu39jAAAAAAthmnnnrqRa21veZ7bcOWHgyrY+PGjTnllFO29jAAAAAAthlVdfZCr7n8EwAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjbdjaA4BJG//ohK09hCWd9ZcP3dpDAAAAALYiM9UAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMNK6DNWqc0RV/WtVnVdVP6qq71bVyVX15KrasAp97lRVf1BVn6qqC6rq6qo6u6reWVWHL7Pt7avq9Kpqg+2wFRo6AAAAACtsxcOn1VZVuyd5d5L7T7x0i367f5KnVdUjW2vnrFCfd+37PGDipf367TFVdXySY1pr18zQxR8nucPyRgkAAADAlrKuQrWq2j7J+5Lctz90bpLXJflGkn2THJPkoCR3S3JiVd2ztXbpMvvcP8mJSfbpD/1XkrcmuSjJHZM8JckeSR6bpCU5amT7Byf5k/7pFUl2Ws54AQAAAFh96+3yz6dlU6B2WpI7t9Ze1lp7R2vtb9KFaR/uXz84yQtXoM9XZlOg9sYk92yt/UNr7e2ttT9JcvckczPiHldVD5224aq6UZI3JNk+yQeSnLIC4wUAAABgla2bUK2/T9rz+6ctyeNba5cMy7TWrk7y+HQzvpLkGVW1xzL6vHOSR/RPz0ny9Nba9RN9np0u7Jvz4hFd/F6SX0w33t+bdZwAAAAAbFnrJlRLd6+0vfr9k1trZ8xXqLV2QZJ39E93SPLwZfR5xGD/dX1oN58T012CmiSHVtXkvdd+Sn9Z6cv7py9cqfu/AQAAALD61lOo9qDB/oeWKDt8/cGr3WdrrWXTZafT9vnaJDunu4z1VTONDgAAAICtYj2FaocM9k9douzw3mSHLFhqEf39zg7un16X5Isr1WdVHZXk8CQ/TvKU1tqPZxkjAAAAAFvHegrVbj/YP2uJsuelC6yS5MCqqhn62zfJjv3+t1tr1y1R/uzB/u0XKlRVeyX5u/7pq1prSwWEAAAAAKwx6ylU222wf9FiBfsA7NL+6YYkO61mf72LF6g76VVJ9khybpIXzTAuAAAAALay9RSq7TzYX2jBgKGrBvu7rIX+qurXkhzZP316a+3yGcY1bO8pVXVKVZ1y4YUXLqcpAAAAAEZYT6HaulZVuyR5Tf/0Pa21Dyy3zdba61prh7bWDt1rr72WrgAAAADAilhPodpwVtdNpii/42D/sjXQ3/9Jd5+2S5M8c4bxAAAAALBGrKdQ7QeD/T0XK1hVG5Ls2j+9NskVq9lfb48F6qaq7pvkd/qnf9xa+84M4wEAAABgjdiwtQcwwteS3Lbf35jFVwDdN8mN+/1vtNbaDP2dl+4+aTsmuXVVbVhiBdD9J8Y6dEyS6tvbs6peMEUbR1XVffr9d7XWJtsEAAAAYCtZT6Ha6UkO7/fvnuRji5Q9dKLeaK2166vqy31fG5LcOcmpM/ZZ/eOOSV4y5RCOmWhPqAYAAACwRqynyz8/PNg/fMFSnQcP9j+02n1WVU28vpw+AQAAAFjj1lOo9tEkF/b7D6yqO8xXqKr2TnJk//TqJO9bRp/vGuw/taoWWrDgIUlu1++f0lo7c/hia+3o1lottSX5+KDaLw9ee+8y3gMAAAAAK2zdhGr9/cxe3j+tJG+uqt2HZfrQ67gkO/WHXt1au3i+9qrq2Kpq/fbiBfr8YpK5QGu/JK+uqs3OWVXtl+Q1g0PztgUAAADAtmM93VMt6cKrRyW5b5K7JfliVb02yTfSLU7wpCQH9WW/nORlK9Dn7ye5Z5J9+vYPqaq3JLk4yR2TPDWbVv48vrV2wgr0CQAAAMAatq5CtdbaNVX18CTvTnL/JLfJ/MHZaUke2Vr74Qr0eXZVPaTv84Ak9+i3SW/L5osLAAAAALCNWjeXf85prV2S5IHp7pt2QpLvJLkmyfeSfCTJU5Lco7V2zgr2+fkkd0ry7CT/meSiJD9Kcm6S/5vkIa21x7bWrlmpPgEAAABYu9bVTLU5rbWW5J39NmsbRyc5ekT5K5L8bb+titbaYavVNgAAAAArZ93NVAMAAACArU2oBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADDShq09ANiWbfyjE7b2EJZ01l8+dGsPAQAAANYdM9UAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqACO0MqsAACAASURBVAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIG5ZTuapunOQ3kjwoyUFJdk+yXWvtZybKHZJk1yQ/bK2dsZw+AQAAAGBrmzlUq6rDkhyXZN/h4SRtnuKPSPKSJJdV1S1ba1fN2i8AAAAAbG0zXf5ZVf8ryb+nC9QqyY+T/HCRKq9Ncn2SXZI8dJY+AQAAAGCtGB2qVdWeSd6a5MZJLk3ypCS7JXniQnVaaxcm+VT/9IHjhwkAAAAAa8csM9WekW7G2TVJfqW19qbW2pVT1PtMulltd52hTwAAAABYM2YJ1R6S7r5p72qtnTKi3tf6xwNm6BMAAAAA1oxZQrW5lT0/OrLe3D3Xdp2hTwAAAABYM2YJ1XbqHy8dWW/H/vHqGfoEAAAAgDVjllDt4v5xn5H1DuwfL5yhTwAAAABYM2YJ1c7oHw8bWe9h6e7F9rkZ+gQAAACANWOWUO2EdKt4PqyqDp6mQlUdleTO/dP3z9AnAAAAAKwZs4Rq/5zkgiTbJTmhqu64WOGqenKS16abpXZmknfO0CcAAAAArBkbxlZorV1RVU9MN+NsvySnVtXJSS6bK1NVf5pk3yQPSLJ/upltP0ry2Nba9csddFVVksckOSrJXZLsleT7Sb6c5O1Jjm2tXbfcfib63CnJU5M8Kt394XZN8r0kn0nyxtbah5eov12S+yb5+X47MMmeSfZI8uN095r7QpL3JXlHa+2qlRw/AAAAACtndKiWJK21E6vqyCRvSBcuPWjupf7xRYPileQHSY5srf3XrAP9SWNVuyd5d5L7T7x0i367f5KnVdUjW2vnLLe/vs+79n0eMPHSfv32mKo6PskxrbVrFmjmZ5KcvEg3+/fbw5O8qKoe21r79PJGDgAAAMBqmClUS5LW2nuq6rNJnpvkt5LcfJ5ilyY5Psmft9a+PWtfc6pq+3Qzue7bHzo3yeuSfCPdzLhjkhyU5G5JTqyqe7bWLl1mn/snOTGbVjv9ryRvTXJRkjsmeUq62WaPTRcqHrVEk2cm+Wy6BR/OSXJlkl36to5IcuskG5P8e1Xdo7V2+nLGDwAAAMDKmzlUS5LW2nlJnpnkmVV1h3Rh0M2SXJ7k20k+vxKXew48LZsCtdOSPLC1dsnci1X16iTvTXJ4koOTvDDJc5bZ5yuzKVB7Y5LfHrynt1fVa5N8It2MtcdV1TtaayfM0865SQ5orX1roY6q6gVJ3pFupdSbJvnrJA9Z5vgBAAAAWGGzLFQwr9baGa21E1prb2utvb+1dupKBmpVtSHJ8+e6S/L4YaDWj+HqJI9PckV/6BlVtccy+rxzkkf0T89J8vTJ99RaOztd2DfnxfO11Vq7YrFArS9zVZInp7vHWpI8sJ+dBwAAAMAasmKh2hZw/3QLEiTJya21M+Yr1Fq7IN1sryTZId09ymZ1xGD/dX1oN58T012CmiSHVtXkvdem1lq7MN2iBUk3k3DPWdsCAAAAYHXMFKpV1W2qar+q2nXK8rv25fedpb/egwb7H1qi7PD1B692n621lmS4+ufMfVbVzdLdoy1Jrk23qikAAAAAa8joUK2qfiHJ2eluuH+3KavdJclZSc6qqjuN7bN3yGD/1CXKnrJAvalV1Y3S3ZctSa5L8sUt0OeGJK9Osl1/6IOLzI4DAAAAYCuZZaGCx/SPX2+tfWyaCq21T1TVV5L8XJLfTPLfM/R7+8H+WUuUPS/dfclunOTAqqp+NtkY+ybZsd//dmvtuiXKnz3Yv/2CpfKTwO5hg0M7p1u19IgkP9MfOyvJs6YdLAAAAABbziyh2n3SLRTwwZH1PpguOLrvUgUXsNtg/6LFCrbWrquqS5Psnu497pRuRdJV6a938QJ157N9kn9Z4LXLk7wryfNaa9P0CwAAAMAWNss91Q7sH08fWW9uYYFFZ3EtYufB/jSXRF412N9lHfQ35wtJPpLkh0sVrKqnVNUpVXXKhRdeuFRxAAAAAFbILKHaXGB06ch6l/WPN5uhz21Ka+3q1lq11irdZ7BHkgckeXu6mYBvTXJSVe21SDNprb2utXZoa+3QvfZatCgAAAAAK2iWUG0uTNt9ZL2b949XLVpqYcPLN28yRfkdB/uXLVhqK/fXOt9vrX2ktfZbSX67f+l+ST5QVTVtWwAAAABsGbOEauf1j/ceWe9e/eN3ZugzSX4w2N9zsYL9Kpq79k+vTXLFavbX22OBuqO01v45ycn903skefCsbQEAAACwOmYJ1T6epJI8pqpuPU2FqrpNupUtW19/Fl8b7G9couy+6Vb+TJJvzLDyZ9KFh3Oz6m7dB3WL2X+w/7UFS03nQ4P9w5bZFgAAAAArbJZQ7c39402SvL+q9l6scFXtk+S92XQJ5XEz9JlsvjDC3Zcoe+gC9abWWrs+yZf7pxuS3Hm1+xwYXj661EqiAAAAAGxho0O11tqp6W6oX0nukuT0qnpBVd25qrZPkqravn/+wiRf6su1JO9urX1mxrF+eLB/+BJlh5dMfmjBUivUZ3/fs+Hry+kzSW432L9omW0BAAAAsMJmmamWdDfTPyVdsLZHkpckOS3JVVV1TbrLJk9L8uJ09yOrJJ9L8sRljPWjSS7s9x9YVXeYr1A/c+7I/unVSd63jD7fNdh/alUttGDBQ7IpCDultXbmrB1W1c5Jfmtw6NOztgUAAADA6pgpVGutXZnkvkn+KcmP04Vmc9uGiefXJvnHJL/U15tJa+26JC/vn1aSN1fVZiuQ9qHXcUl26g+9urV28XztVdWxVdX67cUL9PnFdJeuJsl+SV5dVZuds6raL8lrBofmbauqXrhQEDgoM3ep7K36Q19P8u+L1QEAAABgy1vq5vsLaq39KMnvVtVfpFuE4D7pFgjYNcml6W70/8kk72ytnbdgQ+O8Jsmj0gV6d0vyxap6bZJv9H0/KclBfdkvJ3nZCvT5+0numWSfvv1DquotSS5OcsckT82mlT+Pb62dsEA7j0ry0qr6YpJP9OP7frqAcJ90K30+PJsCwcuTPKG1ds0KvAcAAAAAVtDModqc1tq5Sf6m31ZVa+2aqnp4kncnuX+S22T+4Oy0JI9srf1wBfo8u6oe0vd5QLrw6x7zFH1bkmOmaPLOWXrRgy8keVJr7bQxYwUAAABgy1h2qLaltdYuqaoHJnlMkqOS3DXdfdsuSXJGknckeVN/uehK9fn5qrpTullpj05yYJJdklyQ5DNJ3thaW2pxggenW8zg3ulCtdumW9mzpZvZd3aSU5O8J8lJ/eqjAAAAAKxB6y5US5LWWkvyzn6btY2jkxw9ovwVSf6232bp7/x093s7bpb6AAAAAKwds67+CQAAAAA3WDPPVKuqm6e7h9jhSQ5OsnuSHaao2lpr63KGHAAAAAAkM4ZqVXV4krcmufncoRUbEQAAAACscaNDtar62STvTbJ9NoVp5yb5dpIfrdzQAAAAAGBtmmWm2h+lu8yzpQvXntNa++aKjgoAAAAA1rBZQrVfTheofbq19usrPB4AAAAAWPNmWf3zFv3j8Ss5EAAAAABYL2YJ1S7pHy9ayYEAAAAAwHoxS6j2pf5xv5UcCAAAAACsF7OEam9Mt+rnESs8FgAAAABYF0aHaq21dyR5X5Kfr6qXrPyQAAAAAGBtm2WmWpIcmeTtSV5QVf9WVb9WVXuu4LgAAAAAYM3aMLZCVf14+DTJA/otVTVNE621NrpfAAAAAFgrZgm3JpOzqZI0AAAAANhWzBKqfSJJW+mBAAAAAMB6MTpUa60dtgrjAAAAAIB1Y9aFCgAAAADgBkuoBgAAAAAjCdUAAAAAYCShGgAAAACMNMvqnz9RVTsleXiSeyTZN8muSW68RLXWWnvAcvoFAAAAgK1p5lCtqp6Z5KVJdhlTLUmbtU8AAAAAWAtmCtWq6mVJ/jhdSLaUuRBtmrIAAAAAsOaNvqdaVd0pyZ/0T/8nyS8n2bF/3pI8IsnOSe6Q5NlJzutfOy7Jjq21pS4PBQAAAIA1bZaZar/TP16b5PDW2rlJUrVpIlpr7cokX0nylap6fZJ3J3l8unuuPWo5AwYAAACArW2W1T/vl25G2rvmArXFtNYuTxeknZ/kEVUlVAMAAABgXZslVNu3fzxtgdd3mDzQWrsiyZvS3Vft8TP0CQAAAABrxiyh2k37x29PHL+yf7zZAvVO7x/vMkOfAAAAALBmzBKqXdo/bjdx/JL+8WcWqLdL/7j3DH0CAAAAwJoxS6j2jf7x1hPHv5zu8s4HLlDvPv3jlQu8DgAAAADrwiyh2inpwrO7Thz/t/7xblV1zPCFqnp4ksemW+Dg8zP0CQAAAABrxiyh2kn9469U1bD+m5P8sN9/fVV9tqreVlWfTfL/Bn29frahAgAAAMDaMEuo9qEkZye5LoNLPVtrFyb5nXSz0SrJoUmO6B+rL3Z8a+2dyxkwAAAAAGxtG8ZWaK1dneS2C7z2zqr6bpIXp7uH2lz7X0/yD621V884TgAAAABYM0aHaktprX0iyf2rarskeyS5srV26RLVAAAAAGDdWPFQbU5r7dok569W+wAAAACwtYwO1arq8f3uR1pr542od6v092Brrb15bL8AAAAAsFbMMlPt2HSLETwyydShWpI79nWvT7dSKAAAAACsS7Os/rlctXQRAAAAAFi7tmSoNtfX9VuwTwAAAABYcVsyVLtl/3jZFuwTAAAAAFbcFgnVqmrfJE/rn35zS/QJAAAAAKtl0YUKqupZSZ61wMuvq6pXLtF+JdkpyR7985bkxFEjBAAAAIA1ZqnVP3dLsjFdGDZcYKCS7D1Df19N8ooZ6gEAAADAmrFUqPaDJGdPHNs/Xch2UZIrl6h/fZLLk3wryclJ3thau2KGcQIAAADAmrFoqNZa+/skfz88VlVzq3f+dmvt/as1MAAAAABYq5aaqTafc9LNVFtqlhoAAAAAbJNmCdWekC5U++EKjwUAAAAA1oUbzVDno/32mys8FgAAAABYF2YJ1eYu+/ziSg4EAAAAANaLWUK17/aPN17JgQAAAADAejFLqPbx/vFuKzkQAAAAAFgvZgnVXpPk+iRHV9U+KzweAAAAAFjzRodqrbVTk7wwyW5JTq6qQ1Z8VAAAAACwhm0YW6GqHp/kO0k+lOTBSb5QVZ9M8skk305y1VJttNbePLZfAAAAAFgrRodqSY5N0vr9lm622/36bRotiVANAAAAgHVrllAtSWqJ5wAAAACwzZolVHviio8CAAAAANaR0aFaa+241RgIAAAAAKwXo1f/BAAAAIAbOqEaAAAAAIw060IFP6Wqdk5yiyS7JLksyfmttctXqn0AAAAAWCuWFapV1T5Jnp7k15P8XDZfBbRV1f8keXeS17TWvrecvgAAAABgrZj58s+q+s0kX03y/CQH9W3VYLtRf/yFSb5aVUcue7QAAAAAsAbMNFOtqo5K8qZsCtBakq8k+VqSy5PsnOTAdLPXbpRk1yTHV9WNW2vHr8C4AQAAAGCrGR2qVdUtkvxjurDs+n7/r1pr585Tdt8kz03yu335f6qqk1tr5y9r1AAAAACwFc1y+efTk+yUbnbaMa21Z8wXqCVJa+281tozkzyxP3TTdAEbAAAAAKxbs4RqD04XqJ3YWnvzNBVaa29JckK6S0UfMkOfAAAAALBmzBKq3bZ/fP/IenPlb7toKQAAAABY42YJ1XbuHy8ZWe8HE/UBAAAAYF2aJVS7uH+83ch6c+UvXrQUAAAAAKxxs4RqX0h3b7Sjq2q7aSr05Z6Q7l5sX5ihTwAAAABYM2YJ1f6lf7xdkrdU1Q6LFa6q7ZMcm+T2/aH3zNAnAAAAAKwZs4Rqb0rytX7/N5J8uaqeVVUHV9WGJKmqDVV1UFU9M8kZSY5MN0vtq0mOW4FxAwAAAMBWs2Fshdbaj6vqYUn+I/n/2bv3cF3ruk78709uEAUVRNQEYeepUIlRKbPRDmiiMoZkqY2FpIZjjnOVWlH9MnJ0hhnLyaSaqBA0E5UxLRm0hNLJy1HxVB5KGQUETYFEBOX8+f3x3Mv9sNrrcD9rrb2etXi9ruu57sPz/d7fz80+sPd7f+/7m3sk2ZnkVQvfV9XNu7luJbkiyY929y0zVwsAAAAAc2CWmWrp7s8keViSd2USmE1/9trNufOSPLy7L1qHmgEAAABgU42eqbaguy9P8sSqOiLJjyX53iTfnuQuSb6e5EtJPpjkrd39D+tQKwAAAADMhZlDtQVDYCY0AwAAAOB2Y6bHPwEAAADg9kyoBgAAAAAjrfnxz6p6YJLjk3xPkvsk2S/JtUm+mORDSf68uz+71nEAAAAAYF7MHKpV1SFJfi/Jv1um2Y8l+a9V9ZdJ/mN3XzbreAAAAAAwL2Z6/LOqHpHkY5kEarWKz5OTfLSqHrYONQMAAADAphodqlXVAUnekeTumQRmFyf5pSRHJdk/yV7D9qgkv5jk80O7A5OcW1X7r0fhAAAAALBZZpmp9uIk90rSSd6Q5PDu/q3u/kh3X9Pdtwzbj3T3byc5fGiXod+L16VyAAAAANgks4Rqxw3bTyU5sbtvWK5xd9+Y5GeSfDKTGWvHzzAmAAAAAMyNWUK178hkltpZ3X3Lajp0981JzhoOd84wJgAAAADMjVlCtYWZaZeM7LfQftmZbQAAAAAw72YJ1T43bO89st+9FvUHAAAAgC1pllDtLZm8G+0ZI/v9ZCaPjb55hjEBAAAAYG7MEqqdluQfk3xfVf2X1XSoqlckeVQmixucNsOYAAAAADA3Rodq3f2NJE9IcmGSX66q/1NVx1fV/tPtqupuw/n3Jjk5yQeTPKm7v7kehQMAAADAZtkxtkNVLbwTba9MHgP9/uGTqvqXJN9Icuckd1/U9eAk76mqpS7d3X3/sfUAAAAAwJ42OlRLsjOTd6NlaruQlB04fKYttLnPVLvd6WW+AwAAAIC5MUuodmkEYAAAAADcjo0O1bp75wbUAQAAAABbxiyrfwIAAADA7ZpQDQAAAABGEqoBAAAAwEhCNQAAAAAYaZbVP5MkVbVXkuOSHJPkwUkOSLLPKrp2d99/1nEBAAAAYLPNFKpV1fcmeUOS+y3+ahXde5Yxgc218+RzN7uEFV186rGbXQIAAAC3E6NDtaq6f5J3J9k3u0K0m5JcleSG9SsNAAAAAObTLDPVXppkv0xmnP11klOSfLC7b1nHugAAAABgbs0Sqj0uk0DtfUme0N0e5wQAAADgdmWW1T/vPmxfL1ADAAAA4PZollDtS8P26vUsBAAAAAC2illCtQ8N2wetZyEAAAAAsFXMEqq9JpNVP59VVXutcz0AAAAAMPdGh2rd/XdJXpHkAUneVFX7rntVAAAAADDHZln9M93961V1dZKXJ/lsVb0+yQeTXJXk1lX0f+8s4wIAAADAPJgpVBt8OMlFSR6S5CUj+vUaxwUAAACATTXLO9VSVb+R5PwkD84kJKuRHwAAAADYskbPGKuqJyX5jalTn0nyviRfTnLDOtUFAAAAAHNrlscwXzRsb0zyM939xnWsBwAAAADm3iyPfx6RySOffyJQAwAAAOD2aJZQ7c7D1gqeAAAAANwuzRKqXTps91rPQgAAAABgq5glVHtHJit4PnqdawEAAACALWGWUO3VSa5K8qyqeug61wMAAAAAc290qNbdX0xyfJJrk5xfVT+x7lUBAAAAwBzbMbZDVZ0x7P59kh9OcnZV/W6SCzOZwXbrCpfo7n7O2HEBAAAAYF6MDtWSnJikh/2F7T2TPGnENYRqAAAAAGxZs4RqyWShgln1yk0AAAAAYH7NEqp9x7pXAQAAAABbyOhQrbsv2YhCAAAAAGCrGL36JwAAAADc3gnVAAAAAGAkoRoAAAAAjLTsO9Wq6tCNGLS7L92I6wIAAADAnrDSQgUXJ+l1HrNXMS4AAAAAzK3Vhlu1oVUAAAAAwBayUqh2adZ/phoAAAAAbGnLhmrdvXMP1QEAAAAAW4bVPwEAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEbasZbOVXX3JCcleXySw5MckGRHd+9Y1O7oJPdOcmV3/9VaxgQAAACAzTZzqFZVJyQ5Lcm+C6eGbe+m+UOSvDrJN6vqPt39tVnHBQAAAIDNNtPjn1V1UpLXJtkvkzDtS0k+s0yXs5LcmGSfJD86y5gAAAAAMC9Gh2pVdViS380kTLs0yWO7+5Akv7xUn+6+JsnfDodHjy8TAAAAAObHLDPVXphk7yTXJTm6u/9mlf0+mEkQd+QMYwIAAADA3JglVPuRTN6b9vru/tyIfgttD5thTAAAAACYG7OEaocO2/eP7HfNsL3LDGMCAAAAwNyYJVTbZ9h+c2S//YbtdTOMeRs18fSqekdVXVZVN1TVl6rq/Kp6blXNvKrpMmPuW1Uvqqr3VdVXqur6qrqkqt5UVces8hr3rKoTquqMqvpoVV1dVTdV1VVV9aGq+u2qOny9awcAAABgfc0SPl2R5ODhM8aDh+0/zzDmt1TVAUnOyb9e8ODew+foJM+vquO7+9K1jDU15sOGMe+36KtDh8/TquoNSZ7d3TcucY3fTfJzSe6wm6/vPnyOSvLzVfU/kvxyd9+yHvUDAAAAsL5mCdU+muSQJI9P8urVdKiqSvLUTN7FNvax0enr7J3k7UkeM5z6QpLTk1w01PTsJIcneXiS86rqUcPKozMbVjs9L8m9hlMfTPKnSa5MckSSk5IcmOSZmdzfTy9xqQdnV6D2ySQXJPmHJFcnuWeSY5M8MZPZgy9OcrckP7uW2gEAAADYGLOEam9P8uQkx1TVo7v771bR50VJ7p9J6PTWGcZc8PzsCtQ+kuRx3f3VhS+r6rQkb0tyTCYh1q8n+cU1jJckv5NdgdoZSX62u28djt9YVX+Y5L2ZzFj7qao6u7vP3c11bknyhiT/o7s/vJvvf6+qfjzJGzP5cXluVb2xuy9YY/0AAAAArLNZ3qn2+iQXD33fXlVPWqphVe1XVS9P8t8yCdQ+2t3vmKXQ4T1pvzYcdpITpgO1JOnu65OckF3vbXthVR04y3jDmEcmecpweGmSF0wFagtjXpJJ2LfglCUu94zu/qklArWFa52T287+e9boogEAAADYcKNDte6+KcnTknwjyf5J/rKqPp3kFxbaVNVrq+qvk3wpya8M41yT5CfXUOvRSQ4a9s/v7k8uUd9Xkpw9HN4xyXFrGPPpU/unD6Hd7pyXySOoSXJUVS1+91oWB4DLeMvU/hGr7AMAAADAHjTLTLV094VJfiTJ5UkqyXdm8lhmD01OyCQE23f4/pIkP9Tdn11DrY+f2n/nCm2nv3/CRo/Z3Z3kXes05ten9u+0husAAAAAsEFmCtWSpLvfn8l7y16S5GOZBGq16POpJL+c5CHd/fE11vrQqf0lH6EcXLhEv1Wrqm/LrhVLb06yUv1rHnM3fS9Zw3UAAAAA2CCzLFTwLd19bZJXJXlVVd01yX0zWbXy2iSXd/dVay/xWx40tX/xCm0vy2RhgDskeWBV1TCbbIxDsmum2OXdffMK7acDsAct2WplJ03t727BAwAAAAA22ehQrarOGHY/1t2/u3C+u69Jstv3nK2T/af2r1yuYXffXFXXJDkgk3vcN5Ogb0PGG0wHiPsv2WoZVfWMJI8dDr+cyWqjAAAAAMyZWR7/PDGTVSlrfUtZ0X5T+0stGDDtm1P7d5n38arqwUlOnzr1wu6+bqn2Q5+TqurCqrrwiiuuGDskAAAAADOaJVRbmJF1+XoWcntWVfdO8pfZFcb9fne/ZZkuSZLuPr27j+ruow466KCVmgMAAACwTmYJ1T4/bO+xnoWswvTjm/usov30yplfX7LVJo9XVXdP8ldJ7jecOifJf1ptfwAAAAD2vFlCtT/P5NHPJ61zLSu5emp/2UCvqnYkuetweFOSZR+jXOt4gwOX6LukqrpbJoHaEcOpv0zy77v7llVVCAAAAMCmmGX1zz9I8vwkx1bVj3X3W9e5pqV8Jsl3DPs7s/wKoIdksvJnklw0w8qfyWQF0W9mMgPt4KrascIKoIctqnVZVXWXJO9K8ojh1LuS/ER33zRDrcBIO0+e/8V1Lz712M0uAQAAgCWMnqnW3VcnOS6Td6qdXVWvrKrDVui2Hj4xtf+IJVtNHLVEv1Xr7luTfGo43JHkyPUas6r2S3JekkcOpy5I8pTuvmGGUgEAAADYw0bPVKuqC4bdqzOZEfaiJC+qqsszCdq+uVTfQXf3Y8eOm8lMrhcP+8ck+e1l2j5hav+dM4w1PeZCgHdMkg/vrlFV1fD9imNW1Z2TvCPJvx1OvTfJk7t7NSuMAgAAADAHZnn884eSLDxOubCtJAcPn+XUVJ+x/ibJFUkOSvK4qnpId3/yXw1Qdc8kzxgOr0/y9hnHS5I3J/nVYf95VfWqJcKvJyZ5wLB/YXd/bncXq6p9hnp+cDj1viTHdvc31lAjAAAAAHvYLAsVJJNwbPqzu3O7+8xseJ/ZK6bGel1VHXCboiah1VlJ9h1OndbdV+32BqrOrKoePqcsMebHk7xtODw0yWlVdZv/ZlV1aCbvmVuw22tV1d5J/leSxw2nPpDkSd197e7aAwAAADC/Rs9U6+5Zg7j18AdJnprkMUkenuTjVfWHSS7K5FHU5yQ5fGj7d+2vrAAAIABJREFUqSQvX4cxfz7Jo5Lca7j+Q6vq9UmuymTVzudl18qfb+jupd5+fmZ2rZj69eFejp48Obq07n7bsg0AAAAA2ONmefxz03T3jVV1XJJzkhyd5L7ZfXD2kSTHd/fX1mHMS6rqicOY98tkcYFH7qbpnyV59jKX+v6p/btkErKtxppm+AEAAACw/jZz1tlMuvurmTxC+Ywk5yb5YpIbk3w5k1U0T0ryyO6+dB3H/GiS785koYT3J7kyyQ1JvpDkLUme2N3P7O4b12tMAAAAAObXlpqptqC7O8mbhs+s1zgxyYkj2l+X5FXDZ5bxds7SD2AlO09e6qnz+XHxqcdudgkAAADrasvNVAMAAACAzTZ6plpVvXStg3b3y9Z6DQAAAADYLLM8/nlKkl7juEI1AAAAALasWd+ptpYVKdcayAEAAADAppolVPvhVbT5tiT3SPLIJCckOTDJ2UlOn2E8AAAAAJgro0O17n7PiOZvqar/nEmg9owkn+7ul48dEwAAAADmyYav/tndX0vy1CRfTHJKVT1qo8cEAAAAgI204aFaknT3N5K8dhjvP+6JMQEAAABgo+yRUG3wyWH76D04JgAAAACsuz0Zqu09bO+5B8cEAAAAgHW3J0O1Y4bt1/bgmAAAAACw7vZIqFZVL0jyk0k6yQf2xJgAAAAAsFF2jO1QVS9dZdO9k9wnyQ8lOSxJZRKqvWbsmAAAAAAwT0aHaklOySQcG6OG7X/u7nfPMCYAAAAAzI1ZQrVkV0i2GjcmuSDJb3X3BTOOBwAAAABzY5ZQ7YdX2e6GJFcnuai7b55hHAAAAACYS6NDte5+z0YUAgAAAABbxR5Z/RMAAAAAtpO1rP55dnd/ZkS/+yd5ZpJ098vGjgsAAAAA82Itq39+LMmqQ7UkD5jqK1QDAAAAYMvy+CcAAAAAjLQnQ7U7DNtb9uCYAAAAALDu9mSodtiwvWYPjgkAAAAA626Wd6ot6NU0qqo7J3l4kl8Y+vzjGsYEAAAAgE23bKhWVb+R5KW7+yrJ26pqljHfPksnAAAAAJgXq5mptlRyNkui9ndJXj1DPwAAAACYGyuFahcnec+icz+YyWOcn0py5Qr9b01ybZLPJzk/ybndfev4MgEAAABgfiwbqnX3WUnOmj5XVQuh2K91919sVGEAAAAAMK9mWajgvZnMVFtplhoAAAAAbEujQ7Xu/qENqAMAAAAAtoxv2+wCAAAAAGCrmeXxTwDYEDtPPnezS1jRxaceu9klAAAAc2BNoVpV7ZvkuCSPTHJIkrsmucMK3bq7H7uWcQEAAABgM80cqlXVf0rysiR3GdMtk0UOAAAAAGDLmilUq6qXJ/mVTEKylSyEaKtpCwAAAABzb/RCBVX13Ul+dTj8xyQ/nOROw3EneUqS/ZI8JMmLk1w2fHdWkjt190qPhwIAAADAXJtlptp/GLY3JTmmu7+QJFW7JqJ19zeSfDrJp6vqj5Kck+SETN659tS1FAwAAAAAm230TLUkP5DJjLQ3LwRqy+nuazMJ0v45yVOqSqgGAAAAwJY2S6h2yLD9yBLf33Hxie6+LslrM3mv2gkzjAkAAAAAc2OWUO3Ow/byRee/MWzvtkS/TwzbfzPDmAAAAAAwN2YJ1a4ZtnstOv/VYXv/JfrdZdjec4YxAQAAAGBuzBKqXTRsD150/lOZPN75uCX6PXrYfmOJ7wEAAABgS5glVLswk/DsYYvO/9WwfXhVPXv6i6o6LskzM1ng4KMzjAkAAAAAc2OWUO3dw/ZHqmq6/+uSfG3Y/6Oq+kBV/VlVfSDJW6fG+qPZSgUAAACA+TBLqPbOJJckuTlTj3p29xVJ/kMms9EqyVFJnj5sa2j2hu5+01oKBgAAAIDNtmNsh+6+Psl3LPHdm6rqS0lOyeQdagvX/2yS13T3aTPWCQAAAABzY3SotpLufm+So6tqryQHJvlGd1+zQjcAAAAA2DLWPVRb0N03Jfnnjbo+AAAAAGyWWd6pBgAAAAC3a2ueqVZVeyf5viSHJzkgyd7d/bK1XhcAAAAA5tXMoVpV7ZPkpUmen+Sui75+2aK2/y3J8Um+0N2PnXVMAAAAAJgHMz3+WVX3SvKhJL+c5G5JauqzO29N8oAkP1RV3zvLmAAAAAAwL0aHalVVSd6W5CGZhGj/J8nzsmh22rTu/kCSzw+HTxpfJgAAAADMj1lmqv1kkkcm6SSv6O4f7O4/SvLRFfqdn0kI96gZxgQAAACAuTFLqPb0Yfvh7v71Ef3+Ydh+1wxjAgAAAMDcmCVUe0Qms9TOHtnvimF7jxnGBAAAAIC5MUuothCKXTKy3y1rGBMAAAAA5sYsAdd1w/bOI/t9+7D9lxnGBAAAAIC5MUuodumwPXJkvx8Ytv80w5gAAAAAMDdmCdUuyGQVz2dU1T6r6VBVD0jyo5m8i+38GcYEAAAAgLkxS6j2J0luzeRxzv+5UuOquleSc5LsSHJDkj+eYUwAAAAAmBujQ7Xu/lSS38tkttpPV9X7q+rpSe6z0KaqDq2qf1tVv5nkE0mOyGSW2su7+8vrUzoAAAAAbI4dM/Z7UZL7JnlKku9N8mfD+R62n59qW8P2rO7+LzOOBwAAAABzY5bHP9Pdt3T3jyX5hSRXZBKcLfW5MskLu/tn1qViAAAAANhks85US5J096ur6n8mOSbJY5LsTHK3JNcmuTzJe5L87+7+xhrrBAAAAIC5saZQLUm6+4YkfzF8AAAAAGDbWzZUq6pXDbt/0t2f3AP1AAAAAMDcW2mm2s9nsvjA3ybZbahWVWcMu7/b3R9bv9IAAAAAYD7NtFDBIicmeVaSQ9fhWgAAAAAw99YjVAMAAACA2xWhGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASKsN1Xqd2gAAAADAlrdjle3eVlXLfV+raLOgu3u14wIAAADA3BkTbi2VmPUq2gAAAADAtrGaUG2loEyQBgAAAMDtyrKhWndbyAAAAAAAFhGaAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYKQdm10AAGxHO08+d7NLWNHFpx672SUAAMCWZaYaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhbMlSriadX1Tuq6rKquqGqvlRV51fVc6tqxwaMuW9Vvaiq3ldVX6mq66vqkqp6U1Uds8pr7F1Vj6iqk6rq9Kr6cFXdWFU9fE5c77oBAAAAWH/rHj5ttKo6IMk5SY5e9NW9h8/RSZ5fVcd396XrNObDhjHvt+irQ4fP06rqDUme3d03LnOp9yd5+HrUBAAAAMDm2VKhWlXtneTtSR4znPpCktOTXJTkkCTPTnJ4JsHVeVX1qO6+Zo1jHpbkvCT3Gk59MMmfJrkyyRFJTkpyYJJnJukkP73M5e6w6Pifk9yQ5LC11AgAAADAnrXVHv98fnYFah9JcmR3v7y7z+7u38okTHvX8P2Dk/z6Ooz5O9kVqJ2R5FHd/ZrufmN3/2qSRyRZmBH3U1V17DLXuiDJy5Icl+Tg7v72JGeuQ40AAAAA7EFbJlQb3pP2a8NhJzmhu7863aa7r09yQpLrhlMvrKoD1zDmkUmeMhxemuQF3X3rojEvySTsW3DKUtfr7hd1929091909xdnrQsAAACAzbVlQrVM3pV20LB/fnd/cneNuvsrSc4eDu+YyaywWT19av/0IbTbnfMyeQQ1SY6qqsXvXgMAAABgG9lKodrjp/bfuULb6e+fsNFjdndn12Onax0TAAAAgDm3lUK1h07tf3iFthcu0W/VqurbMnkvW5LcnOTjGz0mAAAAAFvDVgrVHjS1f/EKbS9Lcsuw/8CqqhnGOyTJnYb9y7v75hXaXzK1/6AlWwEAAACw5W2lUG3/qf0rl2s4BGDXDIc7kuy7keMNrlqi74apqpOq6sKquvCKK67YE0MCAAAAkK0Vqu03tb/UggHTvjm1f5ctMN5o3X16dx/V3UcddNBBK3cAAAAAYF1spVANAAAAAObCVgrVrp3a32cV7e80tf/1LTAeAAAAAFvEVgrVrp7av8dyDatqR5K7Doc3JbluI8cbHLhEXwAAAAC2ma0Uqn1man/nCm0PSXKHYf+i7u4Zxrssu96TdvAQ1C3nsKn9zyzZCgAAAIAtbyuFap+Y2n/ECm2PWqLfqnX3rUk+NRzuSHLkRo8JAAAAwNawlUK1d03tH7NC2ydM7b9zo8esqlr0/VrGBAAAAGDObaVQ7W+SXDHsP66qHrK7RlV1zyTPGA6vT/L2NYz55qn951XVUgsWPDHJA4b9C7v7c2sYEwAAAIA5t2VCte6+OckrhsNK8rqqOmC6zRB6nZVk3+HUad191e6uV1VnVlUPn1OWGPPjSd42HB6a5LSqus1/s6o6NMkfTJ3a7bUAAAAA2D5Wevn+vPmDJE9N8pgkD0/y8ar6wyQXZbI4wXOSHD60/VSSl6/DmD+f5FFJ7jVc/6FV9fokVyU5Isnzsmvlzzd097lLXaiqHjbUP+0HpvZ/rKoesOj7P+nuz6+hfgAAAADW2ZYK1br7xqo6Lsk5SY5Oct/sPjj7SJLju/tr6zDmJVX1xGHM+yV55PBZ7M+SPHuFyx2Z5NeW+f7Jw2fau5MI1QAAAADmyJZ5/HNBd381yeMyeW/auUm+mOTGJF9OckGSk5I8srsvXccxP5rku5O8OMn7k1yZ5IYkX0jyliRP7O5ndveN6zUmAAAAAPNrS81UW9DdneRNw2fWa5yY5MQR7a9L8qrhM+uYZyY5c9b+ALBZdp685NsN5sbFpx672SUAAHA7suVmqgEAAADAZhOqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAw0o7NLgAAYE/aefK5m13Cii4+9djNLgEAgBWYqQYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYacdmFwAAwGx2nnzuZpewootPPXazSwAA2BBmqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEg7NrsAAADYefK5m13Cii4+9djNLgEAmCNmqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADDSjs0uAAAAtpudJ5+72SWs6OJTj93sEgBgSzNTDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARtqx2QUAAADza+fJ5252CSu6+NRjN7sEAG6HzFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAI+3Y7AIAAAD2hJ0nn7vZJazo4lOP3ewSAFglM9UAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjLRjswsAAABgnJ0nn7vZJazo4lOPXXXb7XY/wO2DmWoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACNZ/RMAAADWiZVM4fbDTDUAAAAAGMlMNQAAAOBfMesOlmemGgAAAACMZKYaAAAAsK2ZdcdGMFMNAAAAAEYSqgEAAADASB7/BAAAANgiPMo6P8xUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASEI1AAAAABhJqAYAAAAAIwnVAAAAAGAkoRoAAAAAjCRUAwAAAICRhGoAAAAAMJJQDQAAAABGEqoBAAAAwEhCNQAAAAAYSagGAAAAACMJ1QAAAABgJKEaAAAAAIwkVAMAAACAkYRqAAAAADCSUA0AAAAARhKqAQAAAMBIQjUAAAAAGEmoBgAAAAAjCdUAAAAAYCShGgAAAACMJFQDAAAAgJGEagAAAAAwklANAAAAAEYSqgEAAADASFsyVKuJp1fVO6rqsqq6oaq+VFXnV9Vzq2rHBoy5b1W9qKreV1Vfqarrq+qSqnpTVR0z8lp7VdXPDvV+aaj/sqr6y+G+ar3rBwAAAGD9rHv4tNGq6oAk5yQ5etFX9x4+Ryd5flUd392XrtOYDxvGvN+irw4dPk+rqjckeXZ337jCtXYmeWuShy366uDh8++SPLeqfqK7r1579QAAAACsty0VqlXV3knenuQxw6kvJDk9yUVJDkny7CSHJ3l4kvOq6lHdfc0axzwsyXlJ7jWc+mCSP01yZZIjkpyU5MAkz0zSSX56mWvtP1zru4ZTn05yRpLLkjxguNZ9kzwuyVur6vHdffNa6gcAAABg/W2pUC3J87MrUPtIksd191cXvqyq05K8LckxSR6c5NeT/OIax/yd7ArUzkjys91963D8xqr6wyTvzWTG2k9V1dndfe4S1/qN7ArU3pnk+O6+fqr+30/y7kxmsf1wkucl+b011g8AAADAOtsy71Qb3pP2a8NhJzlhOlBLkiGgOiHJdcOpF1bVgWsY88gkTxkOL03ygqlAbWHMSzIJ+xacssS17pnk54bD65I8azpQG671L0P9PZz6/6rqDrPWDwAAAMDG2DKhWibvSjto2D+/uz+5u0bd/ZUkZw+Hd0xy3BrGfPrU/umLQ7Ap52XyCGqSHFVVi9+9lkzCub2H/TcOdf4r3f2JJBcMh/dO8oPjSgYAAABgo22lUO3xU/vvXKHt9PdP2Ogxu7uTvGuFMTejfgAAAAA2wFYK1R46tf/hFdpeuES/Vauqb8vkvWxJcnOSj69xzD1aPwAAAAAbZyuFag+a2r94hbaXJbll2H9gVdUM4x2S5E7D/uWrWIXzkqn96VoXArr7D4e3DPXNdC0AAAAANt9WCtX2n9q/crmGQwB2zXC4I8m+Gzne4Kol+ibJftm10urVqwjolrsWAAAAAJusJq8Dm39VdWOSvYbDvVYKpqrq8iT3GQ7v091fGjne9yd533D4vu5+9ArtH5jkM8PhZ7r7O6e+u0+Sy4fDy7v7kBWutVeSG4fDG7v7jku0OynJScPhdyb5p+Wuezt2j6wuGN0qttP9uJf5tJ3uJdle9+Ne5tN2updke92Pe5lP2+leku11P+5lPrmX+bWd7mc73ct6O6y7D9rdFzt2d5KtobtPT3L6Ztcx76rqwu4+arPrWC/b6X7cy3zaTveSbK/7cS/zaTvdS7K97se9zKftdC/J9rof9zKf3Mv82k73s53uZU/aSo9/Xju1v88q2t9pav/rmzzenq4dAAAAgA20lUK1q6f277Fcw6rakeSuw+FNSa7byPEGBy7RN5mEaguPq+4/1DfrtQAAAADYZFspVPvM1P7OFdoekuQOw/5FPduL4y5L8s1h/+BVBGGHTe1P15ruvjXJ/xsO7zDUN9O1mMl2e0R2O92Pe5lP2+leku11P+5lPm2ne0m21/24l/m0ne4l2V73417mk3uZX9vpfrbTvewxW2mhgt9K8uLh8CXd/dvLtP3xJG8ZDt/S3U+bccwLkzxiODyquz+8TNvTkrxgOHxBd//+ou/PSfLU4fDHu/t/LXOtlyR55XD4yu7+pVnqBwAAAGBjbKWZau+a2j9mhbZPmNp/50aPWVW16PvdjbkZ9QMAAACwAbbSTLUdSb6Y5KAkneSI7v7kbtrdM8nnkuyb5Pokh3T3VTOOeWSSjw2Hlyb5zu6+fjftnpTk3OHwwu7+nt20OSiTR0r3zuQda/fv7q/spt1Dk/x9kkryz0P9t8xSPwAAAAAbY8vMVOvum5O8YjisJK+rqgOm21TVPknOyiRQS5LTlgrUqurMqurhc8oSY348yduGw0OTnFZVt/lvVlWHJvmDqVNLXeuKJAuPhO6X5Myh3ulrHZDkdcP9JcnLBWoAAAAA82fLzFRLkqraO8m7kzxmOPWFJH+Y5KJMXv7/nCSHD999Ksn3d/fXlrjWmUmeNRz+ZnefskS7w5J8IMm9hlMfSPL6JFclOSLJ87Jrtc43dPdPLVP//knen+S7hlOfTvLHSS5P8oDhWvcdvvvbJI/v7puWuh4AAAAAm2NLhWrJt2ZznZPk6GWafSTJ8d196TLXOTOrCNWGtg8bxrzfMmP+WZKf6e4bl2mTqtqZ5K1JHrZMs3cn+Ynuvnq5awEA3F5V1UuH3auT/N5qZ/dX1XcmeXqSdPfLNqg8ALaRqvq+JI/PZBLPAUn26u7HLmpzj0xe93R9d//Lnq+SzbDlQrXkWwsDPC3JT2cSTt0jyVeTfDLJ2UleOzwuutw1zswqQ7Wh/b6ZzCT78SQPTHKXJF9J8n+TnNHdq15QoKr2SnJikmckeUgmvyivTPLRTGbBvbm34g8M666qDk7y4Ex+juyzQvMkSXe/bkOLYkuoqh/YqGt393s36trA5quqO2Yy+//RSXYk+XiS04dXWcyNqro1k/fsJsnfZLK6+or/IFlVxyX58yTd3XfYwBJHqaq7LfWEBZurqg7v7k+vof8vdvcr17Mmtrfh7wCPzW0DnOdsblW3T1X1XUnOSPLI6dPZzf9DquqVSV6Uyd/tD14pk2B72JKhGmx3VXVikpdk1+PMq9XdvWP9K1ofQsI9Z9FfNtfTXP8cA5Y3LML02kx+f3hBd//fRd/fJcl7khy5qOuVSX6ku/9+jxS6ClO/z9Ww/WySJ3f3Z1foN6+h2nVJ3pLkj7v77za7ntWoqs8Nu93d99/N+Vnd5nqbraouTfKo7r58hr6/lOS/ztPPte1k6veB13X3z2x2PWs1LLr3O5lM5Fj4ObNUgPP7SZ6b5Avz9Otlsaq6e5KTcttZXjsW/3myqo5Ocu8kV3b3X+3xQndjmJ3210nunF3vPV+wux+Tw5J8PpOfk0/u7v+9Rwpdo2HSz/0y+bHZ2z+gjyNUgzlTVWdk1yzKxb95r2Su/oKwYLuGhLszrPT7i7ntDI9XreVfuGes49YNuvRc/hzb6qrqiCS/kOThSW7NJNR4ZXd/cVMLWydV9ZhMZpc/MpN3oN41Ky+WtOV+/W8FVfWKJL+S5MtJ7rN4ZnxVvSbJC5bo/vkkD+7uGza2ytWZ+sv055Is/IXyq5m8QuOCZfrNa6g2/Y8h/5Tkj5KcNc+PEE39v+Y2/y0XBZ6zmMcfm08necyYH4+qekmS/545u5/FhqeAvmf43CeTRdWuTfLFJB9K8qF5fYqmqm7I5M9bz+nuMze5nDWpqgdm8l7te2d1Ac6Dk3wik19rj+3uv90DZY5SVSckOS27FhJcuK/d3c8Lk7w6yTcz+f/Tps7cHZ5UuyiTd6vflOTUJH+ayT86vTlL/Lquqg8meUSS13T3z++5iserqsdnMrPuB5LccTj9r/78VVU/l+TfJLm8u39zz1Y5//xhFeZIVT0jk0eDF7wv/z975x0mS1G+7fsByemICCJIFBQDoBIkSY6SUXKS9AmK+DMjBoyYUFQkIwhIRiRLTqKAKKAEQXLOCBxA4vP98VafmTM7szOzO6Fntu7r2qvOdlfNeWdnurvqqTfAZUQxi1IsYtplnCJhqZC0MnA2MXnZ2PY1NeffQYSEv6vq8EeA7SRtZPvSnhkL+YHXhDTRm0K/vCIlbUmE/lc/k5cCdpD0iTJOkltF0qxEVetNikN9NKdlqsOnq3drOxFWXYLd348S97CL6ghqsxNeDwbuA/ZO7V7pZyFgR0LsKRYc2H6xJ5Y35kvAMsB+xC77BZL2sX3QV7saAAAgAElEQVRYf81qm1uAD6R/vwf4GfBDSWcS3msNhcI+8gD1vaIbHR9U3iAKjZ0naU3bLzUbkAS1H6dfS7lBImkaYB/g88SGRyMelHQQ8Cvb3dq0GyuPEbY3/UzKTPIUOheYNx06nhBwFiNEqRHYvk3SbcSm9XqEIFcaJO0BHErl2f8IIdYu3mDI74CfEtEsGxN/g36yFyGovQlsbvs8mCJmjsafiWfSMt01b+xIegtwBK2v0R4lvoeW9Dvb93XRvIEji2qZTLnYLbWvAlvb/mM/jRkvQygSbgZMAu6vFdQSBwIL1Dk+I3CipMVsP99NAwvyLlJLHEtl0WdC/OkpkhYmQvGmq3N6TmIBt3GPBdlOciqwLjFZexH4FxVR5zZiN3pB4O2pv4G/p7795ArCFjP1XKk4PlZqX68fzJfaG+uc24DYqTawi+0r0/HPSlqREHs3lnQz8H1S0ShJVwLb2n68q5Y3xra/Kamoqj4j8Ju08Pl8CUWAutheUtJyRJjUloS30AxEUYWtUkjlUUTu4Cf6Z2kF2wu1c3yA2ZNYgC4HnJE2yhrmSpL0RUJQEyEkjFZgrS+kkLxzqeSJGm1R/S5ijrOlpI/bfrbb9rXBTYSo9p5+GzJOdiUENAN72j4CQNLMTcZdQaRWWb5Jv56SwiB/RXyvHiAK+l1e5Sk8AtvPS7oCWJu4Zvotqm1EfB5nF4Jai/w7te/uvEkd41Aqa7TngPOJ583mDfqfTXiCTyI2Sn/ZZfsGimahF5lMprcsTdy8jxl0QS1RLRJubnsV29+2fYTt37Xy00/j67As8flcXHsihX1umc7/g/A2mBX4auoyF7BLb8zMtInonwfVPsBMxPfmamAtYEngAOD1dO6Pklbqk31jRtKGxM45RJ6oeW2vWNVlP9vL2Z6H8Og8i/gcZiLCeFbvqcEjafS90Dh/+s1cqX20zrnVUvtQlaBWcBph/wbELvxqRNXNadK/f0yfsX0isRB7nLD1M8D5yQNvILB9ve3dCG+V/0eE3hXfnUWAHxJeQ6dLWq/xK2U6ie2jgG8Qn8M6hEdNXZKg9hOqBLVmef56TfJQu4DY5BDhiXM+4fW5KSFqbJp+P4/w1BMh3JyfxpeF3xG2bZ+8bwaVQsy4vBDUWuTW1Dby/uoXexNVMF8kroHLWxx3PfF51ub17AfvTW3LBQkTRYj4HB20pWOkOeWuxNzzQmBh29sxiojpqK59MfHZdK0Y2qAyyDeezARnPGE4NeE836o5993x2DVOit2ofocHdYphEwnfkdqb65zbkEgoa2A327el4z9N4sIqxGL0oK5bmWmVq+h/eNLaqb2XyIdSeD7sJ+k6QsiYhVjEbGH7kn4YOUa2S+0zwM62X27U0faNwGaSvkeE8J0laVnb/+uBnfVo5Ok5DB6ghcD0Wp1zKxLXRD3PyAer/v0r4Ie2n5F0DBE+slZHrRwjtq+VtDyxq74kcY39VdKGtu/tr3Wtk0JqjwSOTDkX9yCuqUmEZ+tmxDXzIBXvtbaT6Gdax/YPJc1DiAVbS3rK9j7VfSR9gYqg9ighJtzZe2ubsgeVjcLbiOiIWxv0/Xny+jyZ2DBcDtgdOLwXhjbD9h8k/ZEQAY+XtMtoz5sS80Hi82h3vvx0at/aWXPGzdrE+znedjuFS4q+C3bepLYpRLGn2hxXRB+80UFbOknh9PAwsEUr4eyJGwkHgvd3xaoBJotqmUHmCsa2IK4Nv9m/5nX6Kao9AixMObwZOsGwiYSFh0e9sJtC5P2P7Ztqzp1NiGr5IVQibK/WbxuISaOJfElThRLZPlvSdsBJhNfjhZL+QghwrwMvFAs6SZdNPdRr9sT60VmeyoS63gJnxH0uhfB9nNih3oMQb3pOo/DpIQmrfpEQ1uauPijpbVTuUfXC26eE7Nv+UtXxCwhR7W2dNXPs2H4g7cSfSITvvBe4PgnTA/c8sv0vYG9JXwY+SXgYFM+cBQix99uSLiCEuPMGJeR10LC9T/JM35oIi37K9vdgiqD2U8ovqAFsn9qngNVtjyoapNxdaxJ5/+YiCs+UQlSTtADwNSJ0bUtgxZTP92pCNGgqsNl+oKtGtsacqa3nRTwaZV0zFOlQ/trmuCJNymwdtGWsPEM8K9t9vi2c2nbFuF6xCjE/O7YNQQ3ieoJK3r9Mokyuu5nMWOhU+E1ZwnIK74Cl+2pF5ygS85bhb9sJZk1tvcVK4eFRL5F08XeYs865zMSmqBpV14PG9unE4qcIvVmR8FbZCdimqutqwKqpXa0rlrZPIdrULiqLTYwZG4z7PfFeP9kNozLcndpab+9Nqdyr64lqxedZu0goFj6lyPFVkDy9NiWS/YtYFF0kade+GjYObP/P9vFpQ+C9wNHplIh7yccJL5f7JH0tFQrJdJ4dqYRB7S9pjySoFd+1xwjP4zv6aGMzlqCyodPSwt/2k4RXpIgcXmXhPqIyaxEO/S7gW8RndBvxfB3tpx0vqm5SVLpsV0wqCkw8PWqv3lM849v1GizuW/3OrQpQhG2vOGqvkaxPXF/1cpeWgUIUu6XNccVn2Wj+NmHJnmqZQeZTHXqdMnkeHERM1naR9GPbZXtAtsulhIvx0oTHwKDzAhF6887qg5Lmo5Jc9i91xhUiXKnExZREdjvCo2h+wntlRGnwGmx70W7bNoF4ivg+Ncy7YfsUSY8RVZdG83Ys1feLyhzjyZrjk4lJ89upTxFmWOYEv4PMFcCHgS2S59YZkuYHvp7O32f733XGLZna2gXo+4D7iYTnpSJVN/1KKmBwKJHf5wgiT9nAktJf7E7kYCpE6urNwfmBHwBfkLRHv9MvSFqGKFjyPiJErZUFWVk8bkdg+3VJmwGXEyGUh1D52z9GeKjVu4bKxAyp/Web44r+9Yrr9JPa51/ZnoetcC8h/i9LFFJqlSL0vl2BpNs8SRTGma9ZxxoKwfaxzpozJv4ErAx8QtI3Wgmxl7QusBJxbz6/y/aNlSIstV0HqyLE+LlRe01AsqiWGVg6lcS+TOE8tm+XtCexE3ihpE8MeMniYRMJ7yAEqLWBX1cdr/ao+XOdcUUutlK8/5TI9ydEXpjigVo7AXWT45nOcBMhqq1ELPbrkpLGf1DSvMT3aXqmzonV76T+9XiasHWWmuOPE6Lae0eMCIrrpWz5YYaFw4DPEoviUyU9R3hGTENc379pMG4tKoVYpmD7i8AXu2ZtB7B9jKS7gDOI0LVl+2xS26SQw52pVAiEyv35buL+cQmwBbHpOC/xXk+XtJrtes+mrpLC8o4jQo3aGkrJnzW2X0qh6n+mkiD+CcJDreyCGsBDxPeoWWXJWor+D3XWnHFRtqJWY+Vi4t60dRJwmlZYTYL1OlQSzpeJGwmBfx1arBQpScQ9zLQfNtoNDicKjs1K5Hrd0HZDsU/SWoS3PcRcp+dV5VvkMWBR2t+8XCa1ZQiXLhVZVMtk+oCkHUc5/SZwOpEX4t+SzgWuJTxamuZIsV2aG/gQioQXE5WyPp6qex1FeA7tS0wAbm3w/opw3rt6YWQLHEmIndU76+8g3sNT6ficVAQ3E3kUyppwdZC5kAjZ2ljSbLZfGK2z7Uepk2+lTqXGMvBv4ntV69l4MzGR21DS5+rkfioqoJVChB42bN8l6bOEuDYN4X1bcDl18tilxP8LU6lSO3DYvlrSR4FziNC3gSB5PexOFMOZjsp9+3UiX+dhNQVMbpT0HWBPoiLrDMA3CU+xniFpEnAlkVdpYLyGkhDYDrsTn8O0hOj5YqPXKEneroLzgc8TgscxbYxbl5J54NjuVORKvzmc2KCYBJwkabPRCi5IWoJYL4jwHPptT6xsnbOInJbrSlq5RWH/C8ScwcAfumlcK9h+WtI+xN/2Q8S67CSq5sOSdqIiHq5MpZru7rZfGfmqpeAaYh62OfD9VgZImo1Ym5q4t2eqUHjGZzKZXiLpTVrbhW13t9a2ey6WNxEJIapebgm8SoQIDaRImLyE7mCk503xOe1m+5iaMSIEqXmAX9r+Qi9sbYSkVYiHoYmH6k627636Tm6WEuTPSnjkfR34CJErbqsh8DYsFZLmInb8ZgC+YfuAPpvUMSR9n/j+XGh7/arj2xO7tyZCXP7P9vOSZiEmd/ukc6fZ3rrnhjOi8EMnKU1Im6QPAbsQE+uXiE2Do22PqAoq6WdUvAc+0ooHRS+QtGr65y2t3pvSwuB7pJDrMi7IUzjuLoTHWSHQFMLUA8TGyNGjeUyk1zmA8LJ40vY8XTK30f/9QyJ5vImcVz8kniMP2361l7a0Qxvzs3bpy/ysESn9w82EB86mtpuGbyfPvLMJAWfpkomEQ4GkfYnQbROpEA4l5gf7p2OrEznj1gG2IrzWTQg4pRLVJE1HzJkXAp4FdrB9vqRNgDOJa2La1HdW4n7xNeJed6PtZeq+cB9IwtpPCYek0e4PIqII9izb51FN2qy5gHgv+9g+OB0f8dmk4yLS+GyVxnzY9s09N7zEZFEtk+kDadLWDaa6CfaKYRMJRyNNKotqjNX83vYOdfqvBVxEvO/NbZ/VfSsbI+kYIsn9ZGDBYnFcK6pV9Z+G8MjbiQhzWT1Xlesskr4FbEZ8JuvbntxnkzqCpJWJyr//A95h+/l0fHoi90vhwfY64ZU2N5W8UG8AK9m+vtd2Jxu7sbAWfbpHZwYDSZsSeUjXJbwICyHtTcIz6HDgfLc4eZe0MVG4oOffO0m3UMm39xHbz/Ty/x8rwzY/Gw1JawOnEBuFPyU2/mpzYBahx58Dvkwkj9/S9qW1/TKdQdIviM0laC7gAHynTKlsqknhqVcAM6VDdxJhkR8j3ttxhJfXR4nQ4sLrbjnb/6l9vX6S3su3CEeBernITORg+4btshYomIKkSwmR9k3g5+lneapENUnTEp/Vt6mE8Z9ie9s+mFxqsqiWyfSB5CrcFTqVa64dJtIkFEDSO4nKi1M8PGz/qUHfbwJrEA/bjVJFur4h6Q7C7iNs71l1vK6ols69haigtSh1vPEymUYkEXdG4GDb11Qd/yCR/6lesYI3gL1tH9YbK0cy0e5pmXJQdR8uFsuPEBU+j7L9YMOBjV9vijdCH0S1ycRCel/bP+nl/z0e0j2rK5TJK1JS4UWzEFEx2sS991YiVcVLhMjxbiLNRfH9uYIQShth2wNbYbcsJI+h71ApEFOPW4nrq3RFYqqRtAIh3hZVSuuJD8U9737Cc7K0XlAptH0l4tqZg9gQfRi4qp4oXVZSpMS1wCLEZ2Lgv0T6FxNe0XNTKSwjolDJysOy+dtJsqiWyWTGzbCJhMOMpOeJXekp7t7peLGY28b2qXXG7UeETF1qe+1e2ZsZXiS9lSiWsSYRHv0SUZXxkDJPqDOZblF1H76Y8Eo72/aYc1mma2xJ6H3eRUnPEAvOrWyf3sv/O9OcBt64jaIJ2ooy6OfGgaQ5gT2I8MgliII3b6mNepC0BpHz8ynbF/Xc0BaRtCThIbQQUws4V9q+oY+mtUUK79wD2J64J9V6et1GFJz4je2XemzehCU9I34LbFJ1uPpar86HeQawS7PcvxOVLKplMpnMBELSK0ROiC1tn1F1/AViV/oz9TyEJG0JnAw8Ynv+2vOZTCbTb1KoypKER8TsVLxrGlKyvJ0HEF7E9/bblvEi6W/Ah4ncQg0rG2f6wzB646b8vgdTyXtbCAIjbJK0N1GR8mXgnbaf65mhExxJsxN54aaIhGXN11tVdOQJ2/9rY9wMxGZh2QqU1CXlWN2BBgIucFy/0nEMCqXKVZTJTHQmys0701f+C8xFxZ274CkiIfZiDca9LbVzdcmuTCbTY1JxiE2IPCqtClGlKbZQkJ6d3yaSKM/UpHs1RU6fUmB7337b0EFOIYrcrAtkUa18LNxvAzqJpD2IpP7VodOTgcUbDPkdkUduRmBj4Phu29gMSacR96MLbL/eb3u6Rcqvemu/7WiR+4icY5sTRTpaZTUi9P5NBkBvSTngSp8HrsyU/kPOZCYY9zFEN+9hFgkHOMTgTkIYW6jm+L+ABYH1iZLutayb2rybm8kMAZI+B3wXmK2dYXSnMuKYkbQScA6xs64m3QeK5NGxDiF6zkt8Vi8QgsH1RGXdsobiHEJUL91E0vq2L+i3QZkKtkfLizZQpEqmvyKu/weAT9m+vKqS4QhSxekriCrna1ACUY2osLw58IykU4ATbF/bZ5sy43uuDNUzKdOY0iy+M5nMFIbp5n0fQyQSFowWYlCn+/tJIQaSyhBicB2RYPUjNcfPBzYE3iPpO7a/XZxIpcQ3Jt7fdb0yNJPJdAdJ3wf2pbVnRnFfK9vzpRCd/gBMIp4XxwN/AQ4j7D4YuIPYMFiXCA018HugtNULU4Xc7wGfZmSl6WomSzoU+LbtV3piXIvYfknShsB5wB8kfY/Il9TvZ2Bm+NgbmJ6oTLqG7XtaHHc9IVov1S3DxoCIyIA9gT0l3Q2cQFSYv7uvljUgzYm7QpnC8zOZ0cg51TKZEjFaBcYm4/pW4Ws0hu39QNMQg3p5O2YHngCmA3a23dfdUEnrECW/XwDmLhZikmYjFp/zpK5PAPcSVYHeTsVDZf2SeNxlSoKkb3XrtW1/t1uvPRqSPtat17Z9VbdeuxVS4uubiOv5DmLxdi2RW8jAZkRl1gWB9YDPE6GhxwGfLpN4I+krwI8Iu7ezfXI6XvfZkzxXjibCXHcvYyGclDj6MkIAbFX0vBlY0/az3bStHSRdlv45B/AhKtUl7yTSDTTL6VWqMOOq9zNWSvV+hglJNwMfAA63vVfV8cJTre5cUtLORJL2Z22/rfZ8r0lVsXcAtqZ+pcxrifvwqSW71usVvegEro3+6DXjWMdsBZwEvGi7HW/wnpHWJwAv236tzvn3EJs7KxMODjcBB9i+vHdWDg6l8wDJZDJjYlJqc8WcLjIkIQaXEklHZwRWBC4HsP2CpO2Ac4mcRPMQpbSrF3UHZEEtU4f96V5IYF9ENeAKurRIoP9zr0+n9jVgXdsPAkiVSz1VX7sduF3SkcDpwI6EGLVFT60dnfVT+/dCUBsN22dJegz4M3CIpOts/7urFrbPGVQ8Z14iFmYXEWLUZMJzbTHCw2YbwmN6KeIzKpNosxqVa6ho30KkS2hG6cKMmfr9tEsZ388wUaQa+Wub455PbSlED9v/Ar4i6avE92174n5biB8fTT+/lHQBMZ88p54g0gdK58ncZzZP7YN9taIBktYjvIhNrAWurzm/GHE9zVEcIp4va0ja0faJPTR3IOj3xC6TyXSGUt+8x0BZRcKBDzGw/QaweoNzl0taCvg68fCch/gM/gb82va5PTM0M2h0Y0Ld70XosC4SPkb8bU8tBLXRsD1Z0haEqLOppC2qKwf3mfcR7+WPDc6P8E6xfZ2kUwlB6tOEJ14pkLQpFfHm78AnGuQVvRk4PYVUng4sC6wmaRPbZ/XK3haodw0N8nU1cLZPkNC8ovDSy22OK0KrX+ygLePGEUZ2OXC5pL2I9BvbE57D0xHz0I3Tz3PpfnaC7T/3yeRPNTm/CbBp+vftxObuXcTffRbg3cSccwkq9/N2UsZ0hLRBvkmD059L9+dRX4J4P0sDixLv5erOWdhRNiTsvaNBVc+DqKzF3gCeJfIxCzhM0mW2H+uJpQNCFtUymT4xwW7e7VJWkXBt4u98fBuCGkDRd8HOm9RZbN8F7NJvOzIDRV2RdsD5Tr8N6CJFWNE/GpyfofaA7RclHUMI7jsS3lRloJj01z4rXiPmuLNQn0sJUW2dLtk1VrZJ7ePAOrb/O1pn2w8mj4PbCM/i7YBSiGq2p+m3DZ1kgN/PsXTP67YsotqTwHzppx3el9rSigMp3P404LRUIGsrQmBbIXWZBOwO7C7pPtuL9sHGhmH0kr5MCGqPAbvZPn+UvusDR6X+19j+eadtbcLSwM6MvF5E+/McEZvSvX4PrfJh4n2OyC2aisytl85fAnwyRd18kvCcngXYDfi+pJmo/G3+Wqaw5F6TRbVMpn8M1c17goiEQxFikMl0EttX9tuGTmN7mEW1mVP7cM3xl4jQ7zmozy2pXbobRo2RV4m5bG116ReIisyNFtmFR0u7i/Buszzx7PttM0GtwPazko4C9kvjM5laBs7Drk1uJDYL1iEKQzVFEe++BXG9tTun6wu2nyFy+h4qaWFCXNuOyOkLI6u69xVJHwUOIDzSVrX9n9H6275A0mrEhs+PJf25gRdVtxmvh+1zRF7M79q+ozMmdZwif/Itdc59nErI+l62nwewfVrKQ7gesK6kl4FvUAlPflrSBrZv6KrlJSWLaplM/xmWm/dQiYQNGKoQg0wmMyF5nhCcpqs5/iwhqjXydCg2Bebukl1j4WEiv1htkvF7iArHyzYY957Ulm0eXPxt/9XmuGJh9PYO2pIZDoYiNK8JZwEbEQv9lVsMg/wClQ3cP3TTuG5g+95UPGM+Qkybvr8W1eWzwDTA0c0EtQLb/5H0WyLdyt5E4YZecRDh2Vkg4lli4P8BFzcZ/yYweUC8teZK7VN1zhWFmv5Zp+LsRUQu0xWBlQhB+z4inHQu4IeUzwO8J5RtMpHJTCSG9eY9LCJhPYY2xCCT6TSSZrP9wjjGrzqMXnAl4C5CbKq9j92Wjq0F7Ftn3MqpLVOuy38Solpt8vtrgWWADSQtaPv+4oSkSUQuNRMVjsvEq0T47Uxtjiv6v9pZczKDzhCF5o3G8YTHzELAWZJ2aPReJM0KfC39GLhxkPLFpoqM2wPbUjLPtDqsRPyN2/U2u44Q1FZu1rGT2H6OWINMIRXwEfBE9XNkCCicBOqFta9IfG5X1DlXvY7ZuLh2JO1HVAr9cAdtHCiyqJbJ9IkhvHkPq0hYzcCEGFQnJ65OJtyJpMUlSk6cKTdnSVrPdtsL/RQCcg45ZLob3AAsB3yo5vhFRN7ID0vaxfZvixMpvH870iK0V4a2wJXAJ4jk/tWcQHhJzABcJelHwN2EZ8qXCI+w0Qoc9IsHiU2Y1Zn6edqMNarGlxJJ81HxeHorMJ3tXftr1cRlgEPzRmD7NUlbEiLAJOAcSXcSuQkBSDkh5yeqZ85MzFGfo5LHsLRImpuwc3sqokX1ZvVdxD3vhB6b1ox3pLbdfIRF/3lG7dUbFk7tE321ovP8l/AsW6D6oKR3A+8ino9/qTOu+GxeqRGj70xtuxtCQ0MW1TKZcjGwN+8hFAnrMUghBsem/7M2mXBxfKyUKTlxptysBpwo6ZOpmllLSFoVOJeST85S0ug9CJG9EAreYvstNf3WIBYXT9m+qOeGjuQSYC9gbUnT2H4zHT+OSn6UIyX9PypC1DJUcqwc2XuTG3I28GvgA5Leb/tWANvXSzqBWITODxxcZ+yDwIE9s7Q1LgXeD2wj6Qjb1zQbIGlFYsFdN+l0v0mCwEGE+FlUYy2+S7vW9D2ESID9YD8SricbJkK1TBi80LxRsX2DpLWBU4hr/j1ErrHi2VN8roUYdT+waavvvdekBPCbE/ewNZn62gF4mnivJ9i+tvcWtsR/iQ2MlWhP8Cs81J4ftVcPGJK1Sz1uJ8I8N2Tq5+C2Vf+ul9e68HCvDRudN7X3dcK4QURtzHMzmUymLSQV1S6fsN1uHrLSIWk64A7C5f5ZYAfb5ycvjjOJSujTpr7VIQYiQgyW6aGtxUJ5ik01x8fKVK+XyTQifdcMHGX7/7U4phDUZgFetT1jkyF9IS28D6ZSXbJY6Iy4PiTtTXi2vgy8M21A9A1JMxIT6hmBnaqFPklbAb8nFtvVE8Ti/Z1gu2uiw1hI35kZgdtsP1h1fDrgEKKacW0Kgr8DW9fJF9NXJL2XyKc2DeFB9CWiaMHrdfpOS+QxPZAQQl8HlrT9754Z3ARJixHeQ+9g5GdQ71p5H5EfzsCatq/ogZlTUXXf6jSuFdz7iaR7CS+VHWyf2Ma4bQmB5H7bCzfr32vS3GsPQoxakpFeUrcBvwN+Y7tMoexFZMM6hO2bMPL58grhwX0CcH69+0KZkHQ2Idq8Aqxg+6YWxixFhO9PD5xne+PuWjl+JM1AbBCsTDgs3QwcYfvJvho2CpK+SniqmnhOHkVs6BxOeHJeZ3vFOuNOIirQXmZ7rarjnyXymF5j+6juv4PykUW1TCaTaQNJyxCLhMKLpggx+BgVL656IQbL9XJHVNJOxb+rc6pUHx8ro+VoyWQKUkXCXYjr4gDb32jSv1pQe40QPc7suqFtImkPogJbsdB5BJhM8oqoIxTMTngfTwfsbPv4HprbNpI+BuxPZYEA8B/g17breXyVmrS5swYRSvQS8Dfbpa32J+k7wDepCDtPE8+cO6kkj1+M8ASdi4rX1/ds799baxuTRM1bCFsh8l6dkH4/mAYbNJJuITw/f2r7az0yt/r/H+/GUyNKtSGVKvdNTwjrLXsRSdqemOf8z/bMzfr3k3TvfRdR0Xgy8LDtp/trVWMkPUqlWMmUjRrgz8T1c1q/N2XaQdKGhDexCa+1fYDf1/NcT4LitsQG1JxpzEaj5fnrBUnkOybZ85lar0BJsxFpCJaqGfoUsLbtf/bE0DaRNAfxTJmr9hTxXje3fVbNmLcQ651JwE9s18u9OmHJolomUyLGEXbwJvAC8AzwL9v/7ZxVmVokrUAlxADq72rXhhjc3AvbMpmyIGkawoNzI+Ia+bztXzfo+zHgPMovqC1IeKtODzwAfMr25fW8VWvG/YnIV3ac7WbV+EpBEkXeBrxku+9hOBMJSQcAX2HqhXXdrsTz/8e29+uFba0i6dOEB4SBPW0fkY43u1YOJsKTr7TdbuXwcdPCxtOYq2WWaUOqSsA5wvaebYw7jPAEe8L2O5r1z7ROjaD7b1KeNNsP9MmkcSPpWCL0triHPU6IUHcRmxwzE9fMqsTGR3HPO872zr20tR6SfkAU7nmc8DR3zflfA59pMPxe4H22X+mulWND0rLEvfidVYcN/Kje80TS5sDpqXLX1LEAACAASURBVM/6JUlnURpK44acyWSA8ee7AkDSzcBhwJHt5DLqNMMqEtr+awpTGcgQg0ymF9h+MyWPvpjwevqFpKdsn1Tdr46gtk0ZBbXE3oSg9iKwhu17Whx3PRHWU7ubXVpsv0bJKxanUJTD+xEm2E1s75tCp74MrE8UXKjlFeK6+ant63ppX4tsntrLC0GtRW5N7eIdtqclRhO+hqhaJsDfiNC8nSUd3kZo3k6MraJjpjlPACcDx9v+e7+N6RC7EM/LQridB9iyQd9CUDsE+FyX7WqVjxLf94vqCGqzEzkgTeQS2zu1e6WfhQhB8cjUfxYA2y/2xPIm2P5bCtHfgBA2XwIusX17gyHzEeuaRpVBJzTZUy2TKREdDDsoLuy/Ahv2S5TqYG6SUoiEjRi0EINMppekMIOrgQ8QotlGxQ5nEtTOBWZN57a1fUa/bG1G2rD4ACHk7FV1vJn3zc7Ab4Fnbb+tR+YOPVXPmHuIhcuxtgeu0M9oSJqeEGPnJarhvgA8CtzsMVTW7RVVnlD7VIcNt3CtbEkIC6UKL1RUy/wzkRvxw62kc0gL1n8QOf9WckmqZcJwhOaNRiqQsSzhhTMrMTd7hAj/LuU9QtK0tt/otx3dQNIqwBeB9YiNqVpeBc4HDnQLBVp6haR/EyHrX7R9UM25rYETiethDdtXVp37B3HfPh/4HvB9KlWaryTmOo+TGRqyp1omUy6KsKCPE5WyAP5J7AjcQyW8YBEin8qSxM38dOAiIlRnOWL3cXpghXRuSjLJPlCbnHgsLEXkMNpRUt9Ewkak0Khbm3bMZCYgtp+TtC5Rnn1B4AxJaxLeN9WC2nZlFtQSRfn5dnNyFeGTs3XQlkwg4pl4APC95OF1lO0L+2tWZ0jC2d/6bccYmDO1j7Y5rhNzhm4wbNUyz5V0HOFJM4nwQPmJpFZC844vq6AmaTOiwMdHR+nzV+Bntv/YM8NaYFgFNQDbVwNXV20S1IqdZd0kKHKO1buPrZbah6oFtcRpwNKEF9i66dh/ifviasCPiUIzmSEhi2qZTImw/TtJXyAEtfuJpNa1N+opSFqNCBn9BPBX2z9JxxcidoKXAlaX9HHb53XV+PoMo0g40KScUJelXz/XyvdC0gZEYuk3gFVslzokLFM+bD8qaR3C02MuYvd2eiqC2va2T++jia1SVCNtt5rxrKntWdhH8gLsCrav6tZrt8lHiDD8bYgKmNMRYYebS3oAOBo4xvbD/TNxwvIc8QxvV0gucpWWzdt7JcYW9ngdIait3HGLxs+gh+ZNIYk1JwBbFIdG6b4CsblzOlH9tIxizlAygJsEs6f2tTrnViTuCZfWOfdg1b9/BfzQ9jOSjiFCqEu7jkkFCWYDXnDJK8yWiSyqZTIlQtJyxO7F88CqzZKT2r4iCWs3AT+VdI3tv9m+L7n230FUqdyayL3SU4ZQJGzKAJTW3hZYGHgS+FOLYy4kvkdzEznkftYd0zLDTPLa2AC4nIoXyxuEoHZa/yxriyeJvCLztTnufantpSB9BZ0Jv6/FlGT+aPtGYM/0nNmauPeumE4vCHwH+JakC4jw0PNtd6u6Y2Zq7iVEtWWJ53qrFIvNWzpt0DgpkvLX5k9tRtF/ng7a0hHStfAZSSczgKF5NZxBeAUVYtptxAZibRGJ1YH3pz6fIOY2G/fU0swg8SIhrM1dfVDS26h8j+pdE1OKE9j+UtXxCwhRrTRpIJIgvT2xIbU8lfkZkp4mNhJOB07MAnRj2n0wZDKZ7rI3MC0RXtBStR/b9xHlnqclwhOK4w8DpxITjOU7bmkL1BEJGwpqECIh4bH2AiESLpuO30d4qxUJ/7fujsWjI2kpSf+Q9PeUX6X2/GxEWNivga2IHdPvArdIWrK31jZkDWJRfG6roQap3znEd6m0u2uZ8pOSL29GLNJeJ/KKDIqgBnAjcR2s0+qAlI9oC+K6azdsdLyoSz+lwvbLto+xvTKx0Pkl4ekkQgDcEDgLeEDSd5PHbqa7XEz8/beW9NZWBkhahri2TGzmlIki7cRKbY4rPNRKW0HX9tW2NyXEg+WJe/QOqV0emN325mUV1FJuq4+nXx8hKhN+wPbnbP/K9tGp/ZztDxLi4cPE9/Pjkrbqk+mZ8nN3ams9vzel8iysd10UItxTNccLz91S5PVLHu13EptO6xNiX/Wzfq50/GjgjpQbL1OHUuw0ZjKZKaxMTCZvaHNc4Upde9O/lojZ79cO6ZhEwuQevQ8hEu6Ujj8s6VTi/fRFJCTCIpYmSmvXq7b2w3S+lrcDZ0oqQ2ntYmet3e/YP2rGZzIASGq1AmY1b6afH0v6cYM+tr3o2C3rCmcBGwHrSlrZ9p9bGPMFYFHi3v6HbhpXw3d6+H+VhlS57P8kfZXYed+N8E4RkcdnP+Drki4BjgDOziEuXeFwwvtpEnCSpM1sNwyblrQE4Q0hInT0tz2xsnWGvlrmAIbmFeya2heJDdy7R+ts+6IUGXEj4cG2G3BKVy3MDCpXAB8GtpC0he0zJM0PfD2dv8/2v+uMKzbSa+dH7yMid87thrHtIGltYsN8OioC4WRCSJxMpK1YlEr6igWBi1Nu60t6bG7pyaJaJlMuivCCERWxmtAovKDISVLPnb8XDJtION7S2jsAR/XM2voULuft7pIVu21zjdorMxFZiPjet+PBVFw/C44yrnSVfoHjgW8Q7/ksSTs0StgtaVbga+nHwI22ezmRPjO199p+oYf/bylIAsHJwMmSFibuzzsTVTQFrJ1+nkzJ5A+1/WCDl8u0ie0HJX0X+AHxd75d0qFEgRJgipfEuwjvtK2IuYqJSntl8+w6gkp+10sltVotcwbiPR3WQ1snGksRf+OjmwlqBbbvlnQ0sYFbbzM0k4G4bj9LCE+nSnqO8DabhvjO/abBuLXS+X9UH7T9RWKzoSdUbXpOtUmZ1iwnUlkfngf8CPhLnfXNSsBXqdz/TpK0aAnv0X0li2qZTLl4jvBqWoFIuNoqRQ6Z2hvcTKl9Zpx2jZVhEwmLPEo31jm3AZXJ8y5Voa6flbQiMenbhP6Lav8jdp1maXPczKkd2upUmTHzAOUUwDqO7dckbUnsXk8CzpF0J+G9CkDytJ2fEOFnpuJ5s02Pzb2J8AbcHDi7yr5vpX+ebPvOHtvUF2zfK+ly4L1ESFu1CDw3sWD4Yvrsvla2CtODiu0DJM1NCBcLEN7cULlfXF7Vvfg8vmu7bF5qQ1stE0DSdERRqPcBb6VSkGVUbH+3m3a1QeFJ066XXdF/5lF7ZSYstu+S9FlCXJuGuPYLLieKEEyFpOWJ3MUGru6FnaOwUGpr52h7EZvsBr5l+weNXiCFfW8s6evA94mca3sS6X0yiSyqZTLl4gZCnNlZ0sEpjGVUJL2f2H038Pea08WuRG1Mf68YNpFwvKW1y5BX7XFiArpUm+OK/mUpuJApCbYX6rcNvcT2DSls4hRCPHsPsDiVSeuOqS0W1PcDm9r+T08NndqGavYnbL2JyKUytEial6hCvQuxyJlyCrgduITIjfMuwhNhd2AtSSuUqLjMQGP7/yRdQYQjj/YMvBXYt8fenO0yNNUyYYqYth/hidNS3rsayiKqPUJc3+1u4Bb9H+msOZlhwvZRkv5OXP/vJgT0iwnPyHqpAz5JPPfLmBuyoMhBeM1oglo1tn8oaT0iCmkjsqg2FblQQSZTLo5O7YzAFZI+MVpnSZ8kqhsVu4pH1nRZhbipNxXnusQNxMRy55QvpSklFwnHW1r77d0wqk2uJT6TrSS1tBstqaggOxB5YTKZbmP7r4RXx5cIcarwfKr+uY3wgHq/7Zv7YGZR5XK6PvzffUPBhpL+SCxsvgcsQnwmrxIhL6vafr/tfYid/E2oPK8WBr5V77UzY8P2WbaXJjaX9gYOJLy2DwK+DCxn+4MlF9Sw/abtzxCeaGcTc4F6hTxeA/4IrGL7s2WsOCtpWiKf0jcJQW2Qi5Vcltp2k6gXc+TLmnXMTGxs32h7b9vr297C9mG2660FsP0l2wvbXsT2s722tYaF088iNceLjcBT23y9ov/i47Rr6MieaplMibB9pqTTiTLfcwGnSHoEuIpIdlmEFyxC5Bt7J5XJzem2/1i8lqT5qOQku7g372AERxOed4VI+BnbpzfqnETCg1N/Uz6RcLyltcuwkXEGUTp7XiKR9E4tjDk89TeRSDqTmfDYngz8HPh5yk/yLmAOIsHvw7afHm18D/gvsVieEJUuU0XPXQnPtHcWh1N7F5ET65jazyXljzlH0nnABUT+r/V7YvQEw/Y/gX/2247xYvtq4GpJ0xNe3O8kPMAnE15PN6ecfmVmDypVjF8nPG8vI6pi9rugUrv8ishZu6OkI203DQNNlWZ3It7riBC+TGYYsH1/g1NzpPaxNl+ySHUx+6i9JiBZVMtkyse2hHhTiB3vJLyE6lEsGI4lJkjVzEAsLqAqn04vGUKR8G7gQ8mO6qTDRWltM3pp7ee6al0L2D5L0o2Ex8D2aSH6jXpVDFPp7O9TKTjxL9un9dTgTGYASAl7b+23HTXcSmxEfFbS1cBNNTvrA58HT9JbiPvv7sCaTO1F8zpRrfUw2/U8iKfC9puSjiVEtQW6YvAEQtJJwOG2r+i3Ld1igKtlQmyuQQiBa7YiRJUV27dI2p3YyL1Y0peAY+uF5iUPvZ2BnxH3wN1sl+3encl0BElTnmW2H6g69SyxNmn3WTd/anPe0RpUp2hNJpMpAZLWBP6PmODXC995DbgI+IXt0rqup0XPkVREwtFuOlOJhNUTIkmLELnZAM7uRzU7ST8DvkD87betKq19JeFefW91dZ2qcYcTi77rbK9Qe77XSFqUCOOsTrj6X+AOKmW031N1XkSRiI+2Wlkrk5kopETsyzLSW+VvttutsttJu/YkKpPV3nOL++xYJ4C2XYpNWUmPU8l1WZ3D7kgi383jdQc2fr11CW812243P1OmCklvEt+xe4jP49h+Xg+ZqZH0LOFtcqDtr/TbnlaoKrLSiOWI6AgTosHVjCwisTKRaB3gfJIoWqKiC5lMx6i6D0/13JZ0GZEL+h+2l2nxtURcLx8CrrS9RuctHlyyqJbJlBxJM1AJL5iF8GIrwgsGxkV/GERCSe8GbqFif3VpbYAv2/55nXF3E3l7Dkv5WPqOpMWJUM4PVB2ufiBU50z5J/DJPiVazwwoLSyAmlLmhY6kzYicah8dpdtfgZ9Ve932CknTEF7KG3T4pUsjOKUFA0RV4vMJD+I/eYyTW0kfAL4IYPtTTbpnRqHqsyk+i9eJ7+NRtsuavHvCIGkyUfxpG9vt5lXqC1UCQUvdG/Ste7ws97RMf5BUVBy27V3rHB8rU71er6m+D1d/xyX9H5Hb0sBBtr/Ywmv9hJjzGPiC7V92weSBJYtqmUympwy6SChpNyqltau5HFi3Ntwgldb+K/EQ2s72yT0xtAXSontrIuR4ZabOkfAc8GeiautpZUy0nCk3bS6A6lLGhU7Ko3QCsEVxaJTuxfs/Hdih13mW0s7ypoSw9i4iLcCqya7bGGPRF9urd8rG8SDpQSLk60jbD/fbnkwFSR8i0lJsQ+XZUlwPDxCf2zGD+LmlqpnLEcVK3kqlWNSolGmTQNLtRLLxnW0f3297WqFKIOg0pdkoyPSH6vlKjfg00PMYSfdReV8LVx2fifDifEc69Ffgp8Altl+s6jczsBYhpq1EzHceARaz/XIP3sLAkEW1TCaTaZO0WKhXWntEJaAUMroF8VD7SAkqATVE0qzE4uf5lIQ9kxkzHVgAlXKhI+kcQqQqxLTbiATfdxGbBLMQ94bVqRQwMXCe7Y17a+1IqhYJm9nuS77NTiFpmiz4l5uq6tG7ElWyC0x4GF5AhIeeX/bPMolp+wGfJcS0tijT/UzSgUT0wKFl8aBvhqRVu/Xatq/s1mtnys8oHl1DOY8BkPRR4FKm3hQwUbygmMu8g8pcR8DLwBq2r+uhqQNBFtUymUwmk8l0nE4sgMq20JG0NXAiMfF8FNh1tFA2SesQHjnzpTHb2j6lF7aOYtPQiGqZwULSEoT32vbA29LhYiHyKPBbYoOqUcW6vpES3J9HpLCA0T1U61GqxbWkdxGpHaYDlrR9T59NymT6xiieauOunl3G+1lBchI4nvC6rcaMvMfdCmxv++Ze2DZoZFEtkykpwxBekMlkMsOEpIuJKpOTgQ+1UrgjFQa5kdj1vcz22k2GdJUqsfMW20/305bMxCSFUG8O7EZ4dFYXzzBwCXAEUZRoRAXHflBV+AMiP9wphIfqw0BLqStKuEmwLnAG8ASxWP5Ln03KZPpCI0+1iUBKE7EhcU9eDpiXyBf9ArHZcT1xnzivyFcqadYc0TI1WVTLZErGMIUXVJNFwnIiaT5CJFiC+Fym62dS1UymzEh6gvCw+ZXt/2tj3C+AfYCnbb+9W/ZlMoOGpIUJcW1nYjEHFe+1JwnvtUNtP9h76ypIuoaoQD4ZWNP23/ppT6u0UDDmQ8AmxN/8RuBaItdi07C3PD/LDAuSdir+bft3/bSlF0jaZ6yFBiTNBlxoe8WmnScQWVTLZErEsIUXwPCKhIOOpLmBg4BPAMXfWNT5Dkk6hFj0PGh70Z4amsmUCEkvEcn+d7B9YhvjtiWKG7xse5Zu2ZfJDCKS1gL2BDZjZNiRCc+wY4Cv2f5v7y0ESc8SOUcPtP2VftgwFtpItN6oWmZD8vwskxlMJL1OzGNOanPcrEQe6eXy9T81b+m3AZlMZir2ANZJ/x5TeEGZSCLhOYxDJOysRZ1D0izE7u7ywPzEZLvZA8a21+y2bc2QtBhwBVMnIB2Ng4FPAwtJWs32Fd2zLpMpNY8AC9P8Wq+l6P9IZ83JZAYTSfMCnyKK/ixcfQq4nQgB3ZSoWjsdsDuwlqQVbD/ZY3NJNgDc0If/e7y0OvdqZ45W2vkZgKQ5iVyWrczNsH1V143KZMrDNMCxkp62fVErA5KH2p+IdU+pr/9+kEW1TKZcbJ/agQovGIWhEgkLJH0O+C6Rc6DlYZTgIZQ8B8+lEmZzPOFBsxghno3A9m2SbiNCRNcjBLlMZlRSUvJbiO/95q0kxZe0EfBHIvRoMdv3ddXI9rmMqGK4CnHttMoqxN/hsm4YlckMAil3z8cJz+cNCLGjEHJeIfL2HG776tT/80Sun28CyxDi27eAvXtrOQAPAosTnqqDxOr9NqBXSJqdCLPfAWjHq97kNXFmYnEvcT89Q9Katq8frXOVoLZCOtTXgktlJN9AMply8T7i4X7YEAhqMHwiIZK+D+xLazu6hYjWrodeN9mVENAM7Gn7CABJMzcZdwXx/Vy+q9Zlholtie/+Q61WmbR9jqQHgAWA7YAfdNG+sfArYsG2o6QjW7mnSVoG2IkQDX7VZfsymdKRKujtSnimvbM4nNq7iKIEx9QWzkhJsc+RdB5wAeH1vn5PjB7J+cB7gI/SnqDeV8pWHKFbSHov8RktSLnmXJlMGVkH+DMwD3CepJVt31GvYxLULmBqQW37en0nMtP024BMJjMVgxxeUI+hEgklLQl8Pf36b2IHeKb0u4lQlVmB9wNfBB5K534HzFSS/AObp/byQlBrkVtTu3iH7ckML6sS18W5bY47h1gUralgW0mnpp8dOm5lG9i+hQhDE3CxpN0k1d2glDStpF2J/CMGdrN9a72+mcywIektkj4h6ULgbiK36nzEtfMG4ZW2tu3Fbf9stEq0tt8Ejk2/LtBdyxtyEPAcsJOkRfpkQ6YOkmYgnhsLEd+vq4AfpdMmRICfAacRRRiK46cD3yEiDzKZCUOqXL4B8DxRfOmiVLhsKlIOtQuAoijBqUSl4KaFTCYa2VMtkykXgxpe0IhhEwk/ndrXgHWLamQR0RLYfonIB3O7pCOJSduORF6PLXpqbX0+SEwm/9jmuGLB03axicyE5b2pvanNcf9M7arAP4Clqs5tIWmS7V+P17jRaKFi3sXEhPRw4EeSriY8bl4CZgbeDawMzJn6nw+8W9K3csW8zAThYWCu9O/iIXk/cCRwtO3H23y9Z1Lbl80p2w9K2poQAy+RtL3tv/TDlswIdiHCPQ18xfaBAJK+ls6fVHhLpxQYewEHEOksfmv7T703OZPpL7ZvlLQZMT+ZnxDWVrH9DEwR1P5ERVA7DdjO9ht9MbjkZFEtkykXAxleMArDJhJ+jJi0nVoIaqNhe7KkLYA7gU0lbWH7jG4b2YRikf9om+NyOEWmXSaltqEHSgOerfr3nMDngfuIRdAShLjdVVEN2J/mORCL83MCG9c5X51HcYP0A9krIjMxeHtq3yDmNocBf0ohnWPhYcLruy9UCe0XE0WKrpZ0I3At4f3U1HMjC+pdo7j/3lkIao2w/RrwS0n3AGcBJ0ha2vZDo43LZIYR25dL2p7w5nwvEQq6JhHNWCuobZsFtcZkUS2TKRcHETtuO0k60PY9/TZonAybSDh/av/R4PwI8dD2i5KOIcJGdyR2ufvJc4SrdztFFqDy3tsVSDITlxcJD83Z2xxX3f+9tl8GkPQ24GgixKcXtCMkN+pbe7zvxUoymR7xMHG9Hmn74fG+WAq9/tS4rRo7+1O5fk1c2x9KP62SRbXusBSVMM96jEh3lPJ3nksUztiTCE/OZCYcts+Q9BngUGA54ExgFiqC2ulkQa0pWVTLZErEEIYXDJtIWCTzr10gvETkVpujwbhbUrt0N4xqk3sJUW1ZKjlqWmGt1N4yaq9MpsKjhEDW7ndt2dTeUwhqiaJi8P/Gb1pTJkzFvEymSyw4hHl3akXydoT3LKh3j8ID//6a428QglqjQkznEdVlNySLapkJjO3DJc1DbB6sReXe9gdgmyyoNSeLaplMiRi28IIhFAmfJ3KKTVdz/FlCVGtUwr3wCpu7S3a1w8WEaLG1pG/YfrbZgFS9cB1iUXBhl+3LDA9XE+EE20jaz/Z/mw2Q9FZga+K7dk3N6fentuvC7kSpmJfJdIshFNSy0F5e3iDmZc/XHH+B2Oyct8G44pn0ri7ZlckMDLa/K+ntwGeIOdiZwFZZUGuNLKplMuVif4YovGDYREIiEfmyRAWzam5Lx9YC9q0zbuXUvtQ901rmcKIy6STgJEmb1XgDTYWkJQjXbxGho7/tiZWZYeBkolLmHMApkjZt8l2bKY2ZRNz/TqrpciZR1e2B7pibyWQy9clCe6l5jEgLUFtI6QGiOFOjKIGiiutMDc5nMgNNC0WXanmamOtPSxRd26+6GFs1JVuf9R2NPV9oJpPpNJLGu7Nr232pjFWP9H6qbzKizRCIkr2fg4mqUSfb3rbq+BeBnxLvbXfbv606twnhqSfgcttr0Wck7Qv8gLD3QSKPwgxURN3ViZ3bdYCtgOmp894ymWZIuhhYk/j+3AV8Dzi32mtN0iRgIyL8ZrF0+Erba/TY3Ewmk8kMGJLOIkI4f277y1XHjwR2JTzYFimqGqZz0xMboosAd9levLdWZzLdp846rGOUaX1WBrKolsmUCEmrjvc1yrSbOoQi4aZEfoGngXmK8JbkLn0nlQTrNwB3E+Ggy1ARE7e13SiRbk+R9Atgn/TraA+CYovqO7a/012rMsOGpLmA6wkvgmov3CeAycCsRFh08T0TcA+wgu0ne2psJpPJZAYOSV8CfgJcb/ujVcfXJCIlDNxMFIy6h5ibfQtYPp072PY+ta+byQw6HViHNaJU67MykEW1TCbTNYZQJJyRcIeeEdjJ9kVV57YCfk8kxa31zgM4wfaOvbK1FZIX3XeAJUfpdiuwr+1ze2NVZthIwtpxwHo1p4oQ92rOJ66tXGU2k8lkMk2R9B5ibvYmsKjt+6vOXQisTf3NQwHPAEt1okJtJlM2OrEOa0SZ1mdlIItqmUwm0yEkfYwIoVyZSs7K/wC/tn1wv+xqhqQlgVUIb6I5CA+ih4kQvBv6aFpmiJC0ArAdcX3MT3h2Pg88RBQ1OMH2df2zMJPJZFpH0pxEPtXZiRxEo2L7qq4bNUGRtCORG+0q27dXHZ+dSMGxZp1hDwOfyM+dTCYzXrKolslkMh1G0nTA24CXbNdWo8pkMplMJjOAJJFmH2AHGlf8rodt5wJxfULSKkQxqXmIolF/A860/b++GpbJZIaCLKplMiVH0gLAEkRVo+ltH9dnkzKZTCaTyWQmFJLeS4SoL8jI0PVm5BxEmUwmM6TkHZNMpqRI2gP4ApVqeAXH1fTbD1gVeMj2Lj0yb8xkkbC8JA+7WYHJtl/rtz2ZTCaTyZQBSTMA5xBpEgCuAv4CfI3I13UqUU17QaKC9lzp+BlEbtJMJpPJDCnZUy2TKRmSZgXOBNYoDlWdHrHTWVPdaEnbpZy8NRIJ67yfUouEKYRAwL22H2yh/wLEJPxN23/usnltIWlmIsfVZsCywJxVp58hwiPOAE6y/VLvLcwME5JWJKqtFfnUmnlt2PauXTcsk8lkmiBpT+A3xFzrK7YPTMffTMc2s312OjYdsBdwAPA6sKXtP/XF8Ewmk8l0nSyqZTIlQ9I5wMfTr/cApxBeXZ+mvqgmItnqPMDXbf+4h+Y2ZZhEQklrAJcQti1j+8YWxiwF3JjGfMz2Nd21sjUkbQgcDryjOFSnW/GAeAzYw/Z5vbAtM1ykKrM/AxZpd2wOl8pkMmVA0gXAusAdtpeoOj5CVKs6txFwFrFJtbTth3po8oRG0vTAJKJae1NsP9BdizKZzDAzTb8NyGQyFSRtQAhqBn4HvNf2fsCFjcY4lPGLCVFk5V7Y2SYnEVWXBNxL7NweNkr/ywgRB2DD7prWNpun9qZWBDUA2zcDf0+/frIrVrWJpF2APxKCmtLP88BNwDWpfb7q3LzAWZI+1ReDMwOLpL2APxCCmpr8UOf3TCaTKQNLEXOzUxqcH7Gmsn0OcC6xMbpn90zLAEhaXNLBkv5DFCN4lJh3Nvu5pz8WZzKZYSGLaplMudgxFiMn9QAAIABJREFUtXcCu9l+vcVxN6d2iVF79ZghFAlXIN5LQ/sbcCHxflbsuEVtImkJ4BDi/i9igbCc7Um2P2x7ldROIkJCT05DpwEOSYmaM5mmSFoEOCj9+jiwC5V7lIE9gA8Q94hfAZOp3CsWZQyebZlMJtMlivQI99ccfyO1MzcYdx7xrC3bJuFQkTb9bibEy0WozHFa/clkMpkxkwsVZDLlohBtjrP9RrPOVRSeXfN03qRxUSsStvqebibK1ZdKJCQW+gC3tznujprx/eQLwPTE9+xztn/TqKPtvwPbSrqKEOKmT+P36IWhmYFnL2Ke8Sawju1bACJiHYAnbN8G3AZcIOlAwoNyR+Al25/pvcmZTCZTlzeA6Qgv7mpeAOYgPLrr8d/UvqtLdk14JC0HHElFIHsZuIFIjfJKH03LZDIThCyqZTLlYu7U3tXmuFdTO30HbekEwyYSFjvR7Sbtfzm1s3XQlrGyFvGZXDSaoFaN7cNSXqx1gbW7aVxmqFid+K6dXQhqo2H7weTdejvwaUl/sH1pt43MZDKZFniMKDr01prjDwAfBJZuMK7wuJ2pO2ZlgC8RnmkmvJ6/YXtyf03KZDITiRz+mcmUi/+ldro2x709tc920JZOMGwiYbHjPPeovUZS9H+hg7aMlaIwwR/bHHdmzfhMphkLpfYvDc6PuL5tPw4cQ3gb7NYdszKZTKZtio2B99Qc/xtxv9pIUnUF7SJZflHBuGm18MyYWYkQ1C6w/fksqGUymV6TRbVMplw8ktp2wx5XSG3Zkq0Om0h4X2pXb3Pcaqktw6S6+Js+0+a4QlAs22eSKS+FZ2bt9/5/Nedr+Udql++4RZlMJjM2ribEs1Vqjhd5R2cDLpG0XkqYvz5wJeGpZuCCnlk68Xhbav/QVysymcyEJYtqmUy5uJKYtG0lqaXrU9I8wBbEpO3yLto2FoZNJLyc+Hw2lfS+VgZI+gCwGfH5XNZF21rl1tTW7rY3Y7HUNg3jy2QSL6a29l5WCLQLNRhXiPDZKzKTyZSFc1K7jKQFi4MpRL0orrQUUZjgdqLq53Kp27PAT3pn6oTjydSWIRogk8lMQLKolsmUi+NSuyjwg2adJc0EnEjk6ngDOLp7po2JYRMJjyL+ztMC5zUT1iS9Hzg79X8zje83xxKfyackNapWNhWp367EZ3JM90zLDBn3prY2N+K/ie/gqg3GLZvaVxucz2QymZ5i+w5gZ+AzjKz0+UngUupXlXwY+Ljth3tm7MTj76l9d1+tyGQyE5YsqmUyJcL2tcCpxETsK5JOlbQ8NeGTkuZL5cNvJEILDRxqu2yeXUMlEtr+D3Aw8fksAPxd0lGSNknhHu9M7SaSjiaqTy1IfD6HpUqHfcX274GzgIWBsyTNNVp/SW8j8q8tBJxp+6SuG5kZFv5BXCtL1hy/IrWrSFqn+oSkZYFdiGsme0VmMpnSYPs424fbvr3m+PO21yY2Cr4HHAEcBGwHLGb7ut5bO6E4lHjW7NDqBm4mk8l0Etnutw2ZTKaK5BV0KZFPqN4FamLyMGUIcAmwge3Xu29he0g6GdiSsPsM4ECitPypgG1PK2k+YB3gq1TCDA+2vU8fTB4VSdMSeTs2SodGu4kWn9PZwBZtVkDtCpIWAGYAfgxsSoTinUB85+4iKpvOTOz4rgFsD0wihLWvMor3kO0Huml7ZrCQtC3x3XrI9gJVxxckvNWmB14nvlt3E+L7Jum4gb1sH95ruzOZTCYzWEg6BPg04Y2/Rxnnw5lMZnjJolomU0JSxagfA3sydYW8WkHtVeDXwL5lnUAMm0hYIOkrwFeAOUfp9gzwI9s/641VzZH0JlN/DqK5MNjKg8K23zIe2zLDhaTZiFw30wPr2b6o6tzngZ+nX2u/jxD5B9ex/WYvbM1kMplOIEnE5uBbgLttv9Jnk4aGtCnYiGmIiIitiU2bQ4FrgaeI9BujkjcFM5nMeMiiWiZTYiS9g/DyWoUIv5sDmEzk6LgSONn2Q30zsEWGSSSsRtIswPrAysD8wOzA88BDRKWwC2y/1D8LR5JEtW5g29N26bUzA4qk6YjFzmu1Apmk7YDvEqHIBS8ChwP75cVoJpMpCyk9xdrp1xttj6jmLWkbYrNg7nRoMvAL2/v3xMghp86mYKfIm4KZTGZcZFEtk8n0jEEXCat2SV+x/XhfjRkjkrpWaMD2p7r12pnhRdLCRDGDl4Dbbb/WZ5MymUxmKiTtSIQWvgEsUiuqSVoXOL/4teqUgYNsf7EXdg4zeVMwk8mUlSyqZTKZTItU7ZIeYnvvftuTyWQymUym+0g6nig88BfbK9c5/0/gA+nXG4D7CM+2OYjww4/Yvrk31g4neVMwk8mUlezqmslkMq3zChG+en2/DclkMplMJtMz3kdsql1Ve0LShwlBzUS455fS8XcTVdpnJqoal6740iCRha9MJlNWctnhTCaTaZ1HU1v6fG+ZTCaTyWQ6xttTe0edc+um9jUiWT4Atu8iKp2LyL2ayWQymSEke6plMn1A0m+79NK2vWuXXjsD1wELUgnxGFpyBbNMJpPJZKYwV2qfr3OuEMz+YvvZmnN/Az7F1AVZMplMJjNEZFEtk+kPO9OdCkYAPRfVJpBIeBSwFbCzpANsT+63Qe0y1gpmknIFs0wmk8lMVIriAzNNdTA2oFagQWgo8FRqZ+2eaZlMJpPpJ1lUy2T6h5p3aZt+VR7ZuYv/d2lENduXSjoE2As4V9JWA1gF9JNUVTCrPZkqmJ1Q/Jra2YBvSpotVzDLZDKZzATkSWA+YPGa48sBk4g50F/qjJs5tf/rnmmZWiRNT3wuM7bS3/YD3bUok8kMM1lUy2T6wzCGAQyTSFgXSR8DTiM+v/WB/0g6A7gaeBh4udlr2K63k91LCi+16+p5qQE/pfJZ1lYw20fScbmCWSaTyWQmGDcB8wPbSPqx7eJ5v3tqXwOuqTOu2Lx6tM65TAeRtDjwOSLH3cK0Pi81eU2cyWTGQb6BZDJ9wPb9/bahwwyjSFiPK6gIfSbCOXZMP61QholbrmCWyWQymUx7nAZsCLwbuELSicD7iWeigfNsv1hn3PKpvb0nVk5QJH0KOISo0A7d2ejNZDKZuvR7cZfJZIaAIRQJR6N2ojZoE7cxVTCTdCqRbDlXMMtkMpnMROP3wN7AMlU/Ba8C+9cOkDQ7sDohul3bfRMnJpKWA44k5mMiogZuICIIcpGlTCbTdbKolsmUiBReCHCL7WfaGDcJWBJKEV74/9u78zDJyvLu49+bUZBdFFxYBFkiq+wisjiDIokLAomAILJFReKWV5O4AIoxMVHRRKMQAwqCCy6IJIBggBlBMaCRQUCEYWfYUZABAjJzv388p6zqprq7qruW093fz3XVdWp5nqq7uq+ZOfM7zzKTHT/sAnrAHcwkSepCZi6LiD+jbFi0N80LancB78jMX7Xpdhhl5FQC/z2IOmepDwDLUX7OnweOmY4bSUmavgzVpHqZTzkp2Bc4p4t+OwHnA8vwz3XfZOZMCNXcwUySpC5l5oPAvhGxFmWttMeAazNz2RhdrqdcjMrM/PmAypyNdqGcu5yfme8bdjGSZh//8y3NLNNtKqIGzx3MJEmapMy8n/Jv6UTtLhxAOYLnVsezhlqFpFnLUE2aGRph2lB2y2yZttpzTmftOXcwkyRJM8X9wNrAI8MuRNLsZKgmzQyNdbLa7Tw1CPPpT6BXh90yJxQRKwBrAMtn5u3DrmcC7mAmSZJmil9QQrWNh12IpNlpuWEXIGlqqkDn0OrhrcMspU+3WoqIzSLipIi4mbKuymLg5jbtDoiID0fEEQMvsr2vU3bFCsruZZ8FjqxecwczSZI0nZxIOac5JCL8v62kgav9CBBppoqIQ2mGYaN9IiImWmw1gJWBTatjAhf1rsKuTLSA/w7A66r7DwGXAYsoI+tWplxd3JXmml7nUq481lJEHAccS7kwMVHw9yzgE8BTEXFuZt7b7/rG4w5mkiRppsjMCyLiJOAo4OSIeHtmPjXsuiTNHpE5lCWYpFkvIj4KfJSR0yYnszZao899wHaZeVcPyuuZiDgYOIWyVtffASdn5pNt2i1PGTH1T5QA54jM/OYga+1ERBwPHEP5uS8FrqiOu1J2+Jozqv2zKOt9rAS8MzO/PNiKx9bpDmYR8RrghZTv97UBlihJkjSmiHgR5SLnPwAHUnZdPZEysv4BYKzdWf9oGizdIanGDNWkIWkJ1SYrgSXALZQRaifUMFDbnDLVcA4wNzMv76DPzsACylTEHTLz+v5W2bmI2JKy0H8AC4EDMvPGiHgj8H3ahGpVv7OBNwBnZuZBg6xZkiRppoqIZUxtXd/MTGdvSZo0551LQ5KZx2fmcq23xkvAPqNfa3Obk5mrZ+Y2mfn+ugVqlfdQpj+e0UmgBlC1O50ysuu9faxtMo6m/L35O2CvzLyxw36NNcy26ldhkiRJs9SMXL9X0vRgKi/Vz0z6x/3VlJBwQZf9FgCHV/3rpLFY/6mZeX8X/RrTCtbtfUmSJEmz1mnDLkDS7GaoJtVIy2i1mWLt6viHLvs12q89bqvBW6c6druJwpLquHIPa5EkSZrVMvPwYdcgaXabaf+Bl1Qvj1bHHbrs12j/WA9r6YXGemlLu+y3enV8pIe1SJIkSZKGyFBNqrGIWCci3hoRn4yIkyLilGHX1KWFlOmsR1a7M02oanckZZrlwj7WNhn3Vsf1u+y3dXVc3MNaJEmSJElDZKgm1VBEPC8ivkHZ2fOrwN8CbwcOa9P2SxHxZETcNNgqO3JqdVwVWBARu47XOCJ2AS4BVque+mr/SpuUKygh4es77RARzwTeRAkJL+tTXZIkSZKkAYvMqexALKnXImITYD7wAp6+aUFm5pxR7TcHrqGENq/KzPkDKLNjEXEBsCfN7c5/AVwELKJM71wJ2BjYg+a0zwAuzMw/HWy144uINwFnUr7LmzLzrOr5NwLfp/3v57PA+6o+czPz0sFWLUmSNDtExB7AfsBOwAspF3YfAe6iXBz9XmZeMrwKJc00hmpSjVSjmq4BNqmeOh04o3r8b7QJbap+1wCbAZ/OzA8OqNyORMSKwHeA11ZPjfeXTiNEPA/YPzNrtaZaRATwv5TpnP8HfAA4GfgzRoVqEbEh8HHgzVX3izLzNQMvWpIkaYaLiD+hnDe3ruPbenG69fzzCuCtmXnjIGqTNLM5/VOqlyMpAVoCR2XmoZn5IyZei2s+5cRhp/6W173MfDwzXw+8FbiKUudYt6uAt2Tm6+sWqEFJzIC/AB4EngV8gbLO2icbbSLikoi4EbiREqgF5fd3yMALliRJmuEiYhtKULYDI88rH6acgz086vmdgCsjYuu2byhJXXCkmlQjEXEh8Grg4sx8dcvzY04vrF5/J/BF4O7MXGdQ9U5GRGwA7AisDawCLKEakp+Ztw2vss5FxEbAt4Ftq6fa/UXauDp6JfDnmXnnIGqTJEmaLSJiBeA3QGNDrKuBT1OWEbm/pd2awF7A+4FtqqdvBTbLzCcGVrCkGecZwy5A0ghbUQKas7vs92B1XKO35UxNROxe3X04MxcCZOatlJOYaSszb4qIHYB9KCPQdgOe29JkCfBj4DTgu+nVC0mSpH54GyVQS+A/gKMzc9noRpn5APD1iPgm5UL0Oyi7uf9l9ViSJsVQTaqX51THu7vsN3pDg7qYTznJ+TSwcLil9FYVlH2/uhERKwOrA0sy8/fDrE2SJGmWeGN1/BVjBGqtMnNZRPwVsDPlYvY+GKpJmgJDNaleHqaMeFq1y37rVscHx201eI8BKzLDArV2MvNR4NFh19EPLSMOAcjMHw+rFkmSpBZbUC7gnj5RoNZQBWunUy76btnP4iTNfIZqUr3cQgnVdgRO7aJfY/21a3pd0BTdDWwIPG0dOE0r82muG5f4b4ckSaqHxvIb3a7Le3t1rNXSKZKmH/9jJNXLjyiB2oERcUxm/m6iDtXaXq+hhB0X9Lm+bi2ghGrbAWcMuZaei4ig/L7abbxwJXDlDFtPra7TjCVJ0uzUmOXx/C77rVUdXbJD0pS4+6dUIxGxHnADsDwlYNs3Mx8fa/fPiNgMOJ+yQOvDwPp1Ws8rIrYHfgY8Qtld6d4hl9QTEbEc8F7gfTSn3rZzB/AvwOc7nZJQRxFxKy07nGbmi4dXjSRJUhERPwVeDizIzHld9LsYmAv8LDNf0afyJM0Cyw27AElNmXkH8HHKiKA9gV9HxN8BWzfaRMTuEXFwRJwG/JLmjkfvr1OgBpCZvwCOBZ4NXBQR037dioh4DnAZ8BlKoBbj3NYDTgAui4hpO70gMzfIzBc3bsOuR5IkqXJ+ddw9It7bSYeIeBclUEvg3D7VJWmWcKSaVEMR8TnKSChoGSHUrml1PD4zj+9vVd2LiLdWdw8E/hRYBlxa3RYDj0/0Hpn5tb4V2KVqhNrllOmeAEspU24vBhZRNipYGdgYmAfsRZlmn8AVwC7TecSaJElSnVQXLW+i7MAOcCbwqcy8qk3blwJ/AxxEOYd+CNiok+VWJGkshmpSTVVTPo8HXjpOs2uBD2Xmfw2mqu5ExDJGhoLB+CHhaJmZtVn7MSKOAr5E+Q7XAQdm5rXjtN8c+BZlZ6mkbPX+74OoVZIkaTaozpm/y8hZWA8ANzLygmdjHbWgXBjdLzP/c4ClSpqBDNWkmquuqu0GbEC5CreEMsprQWb+fIilTagK1aZixBpywxYRlwGvAO4HtsjMBzrosxZlV9Y1gcszc9f+VilJkjS7RMRrga8Az2t5evSF3Yb7gMMz83wkaYpqMwJEUnuZeTVw9bDrmKTDh11Aj21GOUE7uZNADSAz74+Ik4EPAZv3s7jJiIjVKLvH7gS8EFiVsrHEXZQpqxdk5iPDq1CSJGl8mXleRGxCOffcl7JUx0otTR6j7Mz+PeDUzFwy+ColzUSOVJOkDkXEEmBF4KDMPLOLfgcA3wQezcxV+1VfNyJieeDvgaOAVcZpugQ4EfhoZj4xiNokSZKmqrpwuCrwSN0285I0c7j7pzQNRMQKEfGCiHjRsGuZ5e6sjiuN2+rpGu3vHLfVgFSL+v4P8AHKyeZ4O5iuSlnU9/LpvIOpJEmaXTLz95m52EBNUj8Zqkk1FRGbRcRJEXEzZcj6YuDmNu0OiIgPR8QRAy9y9jmPEjS9pst+e1GmjZ7X84om53vA1pTv8hhwCnAAsC2wSXXcHziZssBvVO2/O4xiJUmSJKmOnP4p1VBEHAccSwm+WxdWfdrC/RFxKPBV4Clgvcy8d2CFTlJErAK8gOb6XfdMh7UtImJ9YCFluuQ+ney6GhGvA84BHga2yczb+1vlhPXsA5xFCfl+AfzFeDVFxHqUMG3Hqs9+mfmDQdQqSZIkSXVmqCbVTEQcDxxDc7vvK6rjrrQP1Z5F2Y1yJeCdmfnlwVbcmYh4PvBXwH7ApowKC4HrKeHNiXUOBiNiT+BMyvbsnwb+NTPvb9NuLeA9lKmTjwL7Z+ZFg6y1nYg4E3gTcA+weWY+1EGfNYDrKDtqfS8z9+9vlZIkSd2JiB0oswM2B9YAntVBt8zMV/W1MEkzmqGaVCMRsSVwFSVwWggckJk3RsQbge/TJlSr+p0NvAE4MzMPGmTNnYiIN1MWu28s0h9tmjX+MnoEOCozvzWI2roREV+p7m4AzKXUvBS4FlhEmUq5ErAxsAXQ+F3NB24b560zM4/secFtRMStwHrAJzPzmC76/T3wEeCOzFy/T+VJkiR1JSI2BE4Fdum2K2OcW0tSp54x7AIkjXA0Zcrnb4G92o2AGsPPgb2BrfpV2GRFxCGU6amNhe8T+DVwA2VnyVUo63htSvnuqwFfj4g5mfn1oRQ9tsNohn+N4zOAl1a3VtHSZm4H7z2QUI0y2gzgV132u6Y6rtXDWiRJkiatmglxGfB82l+0laS+MlST6mUeJYg5tYtADaCxJta6vS9p8iLiBcCXKGHZsur+pzLzjjZt1wX+lmaweFJEXJSZ9wyw5E60O2Eb6ySu05O7QQ4ZfhJYAVixy36N9k/2thxJkqRJO46yTm9SLhj+EyVkuzczPWeR1HeGalK9rFMdf9Flv8Yi/yv3sJZe+CtKTQkckZlfG6thZt4JvCcirgROo0yjPJpyslQXLx52AT1wB2WtkXmUqRKd2qOlvyRJUh28jnKeeQ3w8sx8fMj1SJplDNWkemms6bC0y36rV8dHelhLL/wp5UTn/PECtVaZeXpE7E85SfozahSqZeZ466JNFxdR1nt7c0R8OTN/MlGHiHgF8GbK73Lomy1IkiRVnl8dv2ygJmkYlht2AZJGaOx62e1C8FtXx8U9rKUXGiO7zumyX6P9TBgZVjcnUULbOcD5EfH2iGh7gSUi5kTEkcB5VfulVX9JkqQ6aCyXUtud4yXNbI5Uk+rlCsrOkq8HPt1Jh4h4JvAmyiiiy/pW2eSsUh1/12W/h0b1V49k5vUR8Y/AsZSpuScCn4iI+ZTNIx6tnt+EssHCmjQ3XfjHzLx+CGVLkiS1czVl+RR3Jpc0FIZqUr18D9gf2DUi9svMszro88/ACymhxzf7WdwkPEhZPHbjLvs12j/Y23J6LyKeB+wIrE0JAZcAdwFXZuZ9w6xtLJn50YhYnrIxRFCCsz8fo3lQNpn458z82GAqlCRJ6siJwGuBg4EThlyLpFnI6Z9SvXwXWEgJMs6IiKOr8ONpImLDiDgDeC/VWleZeengSu3IVZTvclg1om5CVbtDKd/pqj7WNiURsW9E/AS4mzJd9STgM9XxHODuiLgsIvYZYpljyswPAbsCZwNPUH5Po29PAmcBu2TmR4ZUqiRJUluZeS5l46VtIuILEdHpzuuS1BORmcOuQVKLiNgI+BnwXEqw9HtKcLNp9fjHwLrAho0uwJ3AjplZq/UkIuIvgS9T6v4OcGhmPjFO++WBr9JcFP9tmfmVQdTaqarGM2iO7Brv5K3xF+x3gUPqurV79Z22pox4XJWy4cXdwMK61ixJkgRQrQ37Ocqu8VdTzj2voMx4WDZR/8y8va8FSprRDNWkGqqCtW8D21ZPtfuD2ghzrgT+PDPvHERt3YiIOcC1lPW5AG4FPg/8CLghM5+qToQ2AfYE3k0zLPwNsFVmdrsTal9FxH9Sphk0fv7XARcDi2iuR7YxMI+yyyaU39+5mbn3YKuVJEma+SJiY+BbwHa0P28eS2amSyJJmjRDNammquHr+wCHALtRRq41LKGMWDsN+G7W+A9yRPwJZQOFNXn6Sc5TPH1tx6Ds5LRLZi7qf4Wdi4gDgW9QvsfdwJGZecE47V8DnEJZQDeBgzLzzEHUKkmSNBtExGGU0WlzGk910T0zc87EzSSpPUM1aZqIiJWB1YElmfn7YdfTjYhYBzgZ2KuD5ucDb8/Mxf2tqnsR8SPgVZRQc9vMvKmDPhsBv6SMYLs4M/fsb5WSJEmzQ0S8AriUZpD2CPBz4F7KmrETyszD+1OdpNnAUE3SwETEVsB+wMt4+vpdVwBnZeavhlfh+CLiPsqIwc9n5l930e9zlA0lHszMtfpVnyRJ0mwSEecAr6esnXYscILrwUoaJOePSxqYKjCrbWjWgVWq45Vd9mu0X6mHtUiSJM1221OW2PhmZn5y2MVImn2WG3YBkpoiYnFEnB4RR1bTBqe1iNhk4lbTyl3Vsdu1Nxrt7xq3lSRJkrrx7Or4w6FWIWnWMlST6uWFwEGUxVZviIjbI+K0iDg8IjYYamWT85uIuLMlKNxw4i61dnF13K3LfrtRrqJePFFDSZIkdayxBu+yoVYhadZyTTWpRiLi58A2jAy8W/+Q3g5cAswHLsnMOwZXXfciYhlP3/HzTqr6Kd/htkHXNVkRsSVlKmcAu2XmhNNAI2IH4CfAUmDHzLy2v1VKkiTNDhHxReAo4HOZ+YFh1yNp9jFUk2omIlYHdgfmAvOAlzJ2yHYLIwOqWk0vjIjPUL7HeEHhbYz8DncOqr7JiIi3AKcAjwMfAE7NzKfatJsDHAZ8BlgROCIzvzHAUiVJkma0aqmRq4A/AFvUcfd4STOboZpUcxHxbOCVNEO2rWhuGw4jA6pFmfmSwVXXmZagcB7le2zN2N/hFqqADZhfp6AwIo6r7r4MeC2l7t9RtnJfBDxG2YxgY2BX4DlV+/OYYHODzPx4H0qWJEma0SJiP+DrlBkdb+lkJoEk9YqhmjTNRMRzGBmybUEzoMrM7HYR/YGLiDUYGbKNFRRmZtZml+IxprNGm+fGe76t6fB7kyRJqpOWC547AK+nnHv9Avgf4EE6WGvNC5uSpsJQTZpmImJFyiioecAewHaU3SWDaRKqjdYSFO4NHExNv08VqvVDrb6nJEnSdNDmgmdXFzXBC5uSpqY2I0AktRcRywOvoARo84AdgWc2Xq6OCVxNmTI5bVQB4S6U7zUP2J4SqNXVvGEXIEmSpBFigsfjcYSJpCkxVJNqJiKeAbycZtD0cmCFxsstTa+jufbYgsx8cJB1TkYVEO5Mc5Tdy2gfEF5D87vVRmYuGHYNkiRJ+iMveEoaKqd/SjUSERdSRqWt2Hiq5eXfUC3eT1nA/77BVte9KiDciWZAuDPtA8JfM3JzgtoHhJIkSZKk2c1QTaqRUWt2LaIEaJcAl2TmPUMpagoiYgntA8IbGBmi1T4glCRJkiSpldM/pfppJN2rjrpNu1ANWInm9/kp8CVKQHj38EqavIjYfarvkZk/7kUtkiRJkqThcqSaVCMRcTawG7BG9VTrH9C7aY7uuiQzbxlweV1rsyPTvYwcobZoKIVNUpvv063MTC9mSJIkSdIMYKgm1UxEBLANzXXIdgNWq15u/QN7OyNDtjsHWWcnIuJ9wFxgd+DZ1dOt3+Eumt/h4sy8baAFdmnU9NzJSLdtlyRJ6k5EvKhxPzNvb/f8ZLW+nyR1y1BNqrmIWA7YjmbItiuwSkuTxh/imyn4h8qmAAAMG0lEQVTB1DsGW+HEqqBwW0Z+h3ZB4W2MDAoXD7LOiUTERztothywJmWDhu0o3+8c4CqAzDy+bwVKkiTNQBGxtLo7YtS/swgkDZuhmjTNRMQcYEdKODW3OjZOBqbFSKgqKNyekSHbytXLrX8pLcrMlwy4vJ6JiF2AM4C1gP0z87whlyRJkjTttMwWGHGu6ywCScNmqCZNMxGxLs0wah7QGPYeTNMTg4h4BiUoPAI4lGkWEo4nIjYCFgJPAtvWfYqrJElS3UTEVxv3M/Pwds9PVuv7SVK3DNWkmouI51PCsz2q44ajm7TcvyYzXzqo2qYqItZhZEC4fuvLzIBQDSAiTgTeAfxLZv6/YdcjSZIkSZo6549LNRMRa9Kc1jkPaJ3+GKOa/xqYT3M3zQcGUOKkRcTzaH6vPYCNWl8e1fw6yveaCX5KCdVeBxiqSZIkSdIMYKgm1UhEXA1s0frUqCa/YWSIdt+ASpuUiHguIwPCTVtfHtX8eqrvRflu9w+gxEF5ojquO9QqJEmSJEk9Y6gm1cuWox7fyMgQ7Z6BVzQ199IMz0aHaDcwMkS7d4B1DdoO1fHJoVYhSZIkSeoZQzWpXm6mBE2NEO2uIdczVcu13F9EM0S7ZBoGhJMSEdsCR1F2Nb1myOVIkiRNOxHxlT69dWbmkX16b0mzgBsVSOqbiPgPmiHadA8IiYjdO2y6PLA2Zd24A6vHCbwzM7/cp/IkSZJmpIhYRjmX6rmZsCmWpOExVJNqJCKWUk4Y9svMc7rotxdwHuVqmyNQ+2SSJ3SNaa8XAq/NzGW9rUqSJGlmq87B+mFG7DQvaXj8z7dUL6PXHRtUX3Wu25/zQ8AXgH8wUJMkSZqUFw+7AElqx1BNkjp3fIftnqCEadcBP8tMNyiQJEmapMy8bdg1SFI7hmrSzLBqdXx8qFXMcJnZaagmSZIkSZrhlpu4iaRp4FXVcVbsqClJkiRJ0rA5Uk0akoh4JfDKMV4+MCK2megtgJWB7YB5lAX0L+9dhZIkSZIkaSzu/ikNSUR8FDhu9NPVcTI7TD4F7JKZV061NkmSJEmSND6nf0rDFaNuYz0/0e2XwN4GapIkSZIkDYYj1aQhiYj1gQ1anwIupoxSOxb4yQRvsQxYAtySmQ/1o0ZJkiRJktSeoZpUIxGxjBKq7ZuZ5wy7HkmSJEmS1J4bFUj1Mq86XjPUKiRJkiRJ0rgcqSZJkiRJkiR1yY0KJEmSJEmSpC4ZqkmSJEmSJEldMlSTJEmSJEmSuuRGBZIkSUMUEd0ucPuvmfm+vhTTAxFxGLABQGZ+bJi1SJIk9ZOhmiRJknrpMOCV1f2PDa8MSZKk/jJUkyRJqo99O2hzU9+rkCRJ0oQM1SRJkmoiM88edg2SJEnqjBsVSJIkSZIkSV0yVJMkSZoBImKLiPhsRFwVEb+NiCciYnFEnBMRB0fEuOd9EbFiROwbEV+MiP+JiAcj4g8R8XBEXBsRJ0bE1uP0n19tuvDKlueyze1jLa/Pbff8GO8/YduW1+dXj9eIiA9FxJUR8UD12qlt+i0fEUdWP6s7IuL/IuKhiLg6Ik6IiA3Gq616j7Uj4uMRcXn18/9DRPwuIm6MiEur381uE72PJEmaPpz+KUmSNI1FxDOAE4B38fQLpmtXtzcA746IfTLznjHe6jqqXTtHWQ3YvLodFRGfzMwP96L2foqI7YCzgfUmaLcD8G3gxaNeWgHYqrq9KyLek5n/PsZ7vA74FrDKqJeeXd02BnYFjqgeS5KkGcBQTZIkaZqKiKAEQo0NDu6mhDsLgceA9YEDge2BnYCLImLHzHyszdutCPwW+BHwS2Ax8AdgHWA7YH/gmcCHIuK+zPyXUf2PAdYEPgFsUT3XbuOF67v/pl17LvADYF3gPOBc4AHKd8lGo4jYGfhvYKXq+QuACynffUVgZ+CQ6vWTIuKJzDy19YMiYh1GBmrnUn6Gd1FCzucBWwN7Aqv3/JtKkqShMVSTJEmavt5DM7g6A3jH6MAsIk6gBF0fpow2Ow74YJv3Ogz478x8qt0HRcRHgB8CmwIfj4hTMvORxuuZeVnV7n0tzw1r44UtgaXA/pn5nXYNImJV4ExKYPYQsE9mLhjV7LSI+AxwEfAi4N8i4r8y84GWNm+mGaj9XWZ+aozPC8poNUmSNEO4ppokSVJNjLEGWevt1Ja2z6IEZQBXAoe2G4GWxUeAS6un3ln1Hd3uh2MFatXrtwFHVw9XBd44qS85OJ8fK1CrvI3m1NC3tgnUAMjMRcDh1cOVgbeParJxy/3/GOvDqt/DpWO9LkmSph9DNUmSpOlpL8rUQoDPZeayCdqfUR1XA14+yc/8acv9nSb5HoPyhQleP6Q63pCZ/zlew8y8mDKdE+A1o15uDTK3QJIkzRpO/5QkSaqPdmuQtbq95X7rTpJrRMQ+E/Rdp+X+ZsD80Q0i4nnAWynB0ebAGpTpke2sO8HnDdPizLxlrBcjYnXgpdXDezv42QEsqY6bjXr+R8BfV/fPiohPAt/JzDu7KViSJE0/hmqSJEk10eUaZBu03P9ilx+1xugnIuIA4N/pfDH91br8zEFaPMHr69GcsbEbIwPKiYz42WXm+RHxDeAgYC3gs8BnI+JGysi+HwP/lZn3dfEZkiRpGjBUkyRJmp6mspPk8q0PImJ34Bs0g6b/peyKeRPwMPBES/PvV8c5U/j8fnt8gten8rN7Zpvn3gJcTBmx1pgCukl1OxRYGhHfBt6fmXdP4bMlSVKNGKpJkiRNT0ta7m843nTHDnyMZqD29sxsu+B+RKw8hc+Yql6uBdz6s/taZh46lTfLzAROAU6JiA0pu3y+AtiDEqzNoewSumtE7JiZ907l8yRJUj24UYEkSdL01DrFcdLrm0XE8jSnP/58rECtsv5kP2cMrSPglh+zVbFmDz+3Jz+7djLz5sz8WmYelZl/AmwP/LJ6eT3gb3r5eZIkaXgM1SRJkqanBS33R+9I2Y3n0py9cNMEbffq4P3+uAtpRMQEbR9qub/2BG17tttoZj4AXFc9fHlE9G19uMz8X5o7jUIZxSZJkmYAQzVJkqTp6Tzgger+0RHxwkm+z2Mt9zcaq1FErEpzl8vxtE6tnGi66CLgyer+3LFCuIh4NmVX0l46rTquBHywx+892q0t911+RZKkGcJQTZIkaRrKzEeB46uHzwF+GBGbjNcnInaKiE+Nep+HgRurhztExL5t+q0CfIcyfXEirWu7bTdew8z8A2WBfyhTS9/V5rNXBr5Jb6d/Qtkx9bbq/gcj4m8iYsxz44hYPSLeExGvHvX8cRGx53h9gaNb7i+cfMmSJKlOoqyrKkmSpGGIiD+ejGXmRNMl2/U/jeYorqeAc4AfA3dTFshfC9gKeBXwYuCmzNx41Hu8G/h89XAZ8HXgMuARYEvgMMr0zK+1fNaCzJzbpp69gR9UDxcBn6OEV0sbz2Xmopb2r6LsNAqQlADtAsoIti2qz14X+BZwYNXu+Mz8WJvPbvws29bWpv02lGm0jemfNwHfo0wNXVI9vyHwMmAuZd23QzLzjJb3mA+8Erinqvuq6v5ylJ/Z3jTXrHsC2D4zr52oNkmSVH+GapIkSUPUg1AtgI8AxwArdNDlaYFT9R6nAweP0+8HlB0sG9NFxwrV5gDzGXvtsKcFYhHxceDYMdonZUTeAuCSsd6jep+uQrWqz0soQd62HTR/AtgnM3/Y0v8SSuA2kQeAgzPzwk7qkiRJ9eeaDpIkSdNYliukn4iIU4C/pIxIewllSugySphzPXA5cF5m/myM93hLRJwLvI0SMK0E3EcZeXV6Zn4bYKK9BzJzaUTsCbwXeCOwKWXE15xx+hwXEZcB76ZsSLB69dk/Af4tMy+LiLmd/Dy6lZm/iYjtgTcA+wE7Ay+grAf3CGWU3ULKNNVzMvN3o95ib+DVlNFq2wMbUzZ/SOC3wLXA+cApmfkQkiRpxnCkmiRJkiRJktQlNyqQJEmSJEmSumSoJkmSJEmSJHXJUE2SJEmSJEnqkqGaJEmSJEmS1CVDNUmSJEmSJKlLhmqSJEmSJElSlwzVJEmSJEmSpC4ZqkmSJEmSJEldMlSTJEmSJEmSumSoJkmSJEmSJHXJUE2SJEmSJEnqkqGaJEmSJEmS1KX/D8E070dRec7NAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1440x1080 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}