project03_CNN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "project03_CNN.ipynb",
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gjhdBsihMKJs"
      },
      "source": [
        "# Imports"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jc1DAIAG985D"
      },
      "source": [
        "test_path = \"/content/drive/MyDrive/Colab Notebooks/nlp-project-3-dataset/test.csv\"\n",
        "train_path = \"/content/drive/MyDrive/Colab Notebooks/nlp-project-3-dataset/train.csv\""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_Hw1PmQoDmYJ",
        "outputId": "713a1a56-cdd7-4dea-e4ca-5e74a4218cf8"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mKHnJDU1-LGB"
      },
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "from tensorflow.python.keras.models import Sequential\n",
        "from tensorflow.python.keras.models import Model\n",
        "from tensorflow.python.keras.layers import Input, Dense, Embedding, Conv2D, MaxPool2D, Conv1D, MaxPooling1D, GlobalMaxPooling1D,GlobalMaxPool1D\n",
        "from tensorflow.python.keras.layers import Reshape, Flatten, Dropout, Concatenate\n",
        "from keras.optimizers import RMSprop\n",
        "from tensorflow.python.keras.callbacks import ModelCheckpoint\n",
        "from keras.optimizers import Adam\n",
        "from tensorflow.python.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.python.keras.preprocessing.sequence import pad_sequences\n",
        "from tensorflow.python.keras import regularizers\n",
        "from tensorflow.keras.callbacks import EarlyStopping\n",
        "\n",
        "\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.svm import LinearSVC\n",
        "from sklearn.metrics import classification_report, f1_score, confusion_matrix\n",
        "from sklearn.model_selection import cross_val_score, train_test_split\n",
        "from sklearn.naive_bayes import MultinomialNB\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "from sklearn.metrics import classification_report, f1_score, accuracy_score, confusion_matrix\n",
        "\n",
        "\n",
        "import gensim, logging\n",
        "\n",
        "logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)\n",
        "import time\n",
        "\n",
        "from gensim.models.word2vec import Word2Vec\n",
        "from glob import glob"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZPFIuBGpL0Tk"
      },
      "source": [
        "# Reading data and Preprocessing"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_XgeJ_Wt-LWs",
        "outputId": "84356cd9-d950-40fa-f9ea-58560ff2831c"
      },
      "source": [
        "import nltk\n",
        "nltk.download('stopwords')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8W3PfMnw-NVy"
      },
      "source": [
        "test_df = pd.read_csv(test_path)\n",
        "train_df = pd.read_csv(train_path)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DaZevksL-O_T"
      },
      "source": [
        "trainx = train_df[\"text\"]\n",
        "testx = test_df[\"text\"]\n",
        "\n",
        "trainy_raw = train_df[\"label\"]\n",
        "trainy_integer = []\n",
        "for i in range(len(trainy_raw)):\n",
        "  if trainy_raw[i] == 'spor':\n",
        "    trainy_integer.append(0)\n",
        "  elif trainy_raw[i] == 'yazarlar':\n",
        "    trainy_integer.append(1)\n",
        "  elif trainy_raw[i] == 'video':\n",
        "    trainy_integer.append(2)\n",
        "  elif trainy_raw[i] == 'dunya':\n",
        "    trainy_integer.append(3)\n",
        "  elif trainy_raw[i] == 'turkiye':\n",
        "    trainy_integer.append(4)\n",
        "  \n",
        "testy_raw = test_df[\"label\"]\n",
        "testy_integer = []\n",
        "for i in range(len(testy_raw)):\n",
        "  if testy_raw[i] == 'spor':\n",
        "    testy_integer.append(0)\n",
        "  elif testy_raw[i] == 'yazarlar':\n",
        "    testy_integer.append(1)\n",
        "  elif testy_raw[i] == 'video':\n",
        "    testy_integer.append(2)\n",
        "  elif testy_raw[i] == 'dunya':\n",
        "    testy_integer.append(3)\n",
        "  elif testy_raw[i] == 'turkiye':\n",
        "    testy_integer.append(4)\n",
        "\n",
        "\n",
        "testy_integer = np.array(testy_integer)\n",
        "trainy_integer = np.array(trainy_integer)\n",
        "\n",
        "trainy= np.zeros((trainy_integer.size, trainy_integer.max()+1))\n",
        "trainy[np.arange(trainy_integer.size),trainy_integer] = 1\n",
        "\n",
        "testy= np.zeros((testy_integer.size, testy_integer.max()+1))\n",
        "testy[np.arange(testy_integer.size),testy_integer] = 1\n",
        "\n",
        "trainx = np.array(trainx)\n",
        "testx = np.array(testx)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZpE6vZ1t-kKi"
      },
      "source": [
        "#Preprocessing\n",
        "import re\n",
        "\n",
        "def splitIntoStem(message):\n",
        "  return [removeNumeric(removeBackSlash(stemWord(word))) for word in message.split()]\n",
        "\n",
        "def stemWord(value):\n",
        "    return value.lower()\n",
        "\n",
        "def removeBackSlash(value):\n",
        "  return re.sub(r'[\\\\]','',value)\n",
        "\n",
        "#Remove numeric character\n",
        "def removeNumeric(value):\n",
        "    blist2 = [item for item in value if not item.isdigit()]\n",
        "    blist3 = \"\".join(blist2)\n",
        "    return blist3\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8BLz2CLN-7rA",
        "outputId": "d6924880-3757-4726-fb03-69f3c34bf808"
      },
      "source": [
        "print(\"BEFORE: Key:\",testx[0],\"- Value:\",testy_raw[0])\n",
        "\n",
        "for i in range(len(testx)):\n",
        "  testx[i] = \" \".join(splitIntoStem(testx[i])).split()\n",
        "\n",
        "print(\"AFTER: Key:\",testx[0],\"- Value:\",testy[0])\n",
        "\n",
        "for i in range(len(trainx)):\n",
        "  trainx[i] = \" \".join(splitIntoStem(trainx[i])).split()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "BEFORE: Key: \"Filmimde oynadı ama 32 yıldır görmüyordum abimi\" diyen ve heyecanlanarak gözyaşlarına boğulan Cem Yılmaz'ı izleyenler kahkahalara boğuldu. - Value: video\n",
            "AFTER: Key: ['\"filmimde', 'oynadı', 'ama', 'yıldır', 'görmüyordum', 'abimi\"', 'diyen', 've', 'heyecanlanarak', 'gözyaşlarına', 'boğulan', 'cem', \"yılmaz'ı\", 'izleyenler', 'kahkahalara', 'boğuldu.'] - Value: [0. 0. 1. 0. 0.]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aCJ7C_i7-81w"
      },
      "source": [
        "#Remove stopwords\n",
        "from nltk.corpus import stopwords\n",
        "\n",
        "def removeStopWords(news_list):\n",
        "    \n",
        "    filtered_stopwords = []\n",
        "    filtered_stopwords_list = []\n",
        "    \n",
        "    stop_words = stopwords.words('turkish')\n",
        "\n",
        "    stop_words.append(\"bir\")\n",
        "    stop_words.append(\"iki\")\n",
        "    stop_words.append(\"üç\")\n",
        "    stop_words.append(\"dört\")\n",
        "    stop_words.append(\"beş\")\n",
        "    stop_words.append(\"altı\")\n",
        "    stop_words.append(\"yedi\")\n",
        "    stop_words.append(\"sekiz\")\n",
        "    stop_words.append(\"dokuz\")\n",
        "    stop_words.append(\"on\")\n",
        "    \n",
        "    \n",
        "    for i in news_list:\n",
        "        filtered_sentence = [w for w in i if not w in stop_words]\n",
        "        filtered_stopwords_list.append(filtered_sentence)                         #return list value\n",
        "        filtered_stopwords.append(\" \".join(filtered_sentence))                    #return string value\n",
        "    \n",
        "    return filtered_stopwords,filtered_stopwords_list                                      \n",
        "\n",
        "\n",
        "testx,testx_filtered_stopwords_list = removeStopWords(testx)\n",
        "trainx,trainx_filtered_stopwords_list = removeStopWords(trainx)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "H5qbqGOWL_Vj"
      },
      "source": [
        "# Tokenizing and Padding Words"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "auX2JNjx-_2S",
        "outputId": "07c3bb98-7a17-42e5-af72-0d11e0c9f922"
      },
      "source": [
        "import pandas as pd \n",
        "\n",
        "tokenizer = Tokenizer(oov_token='OOV', filters='\"#$%&()*+,-/:;<=>@[\\]^_`{|}~ ')             #Tokenizer(num_words=5000) => 5000 words of the highest frequency\n",
        "tokenizer.fit_on_texts(trainx)\n",
        "tokenizer "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras_preprocessing.text.Tokenizer at 0x7f9d4031ae10>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q_n6Wj9X_SSO",
        "outputId": "699eec2e-0ec8-4404-a0e5-bb223b11ce17"
      },
      "source": [
        "num_words = len(list(tokenizer.word_index))\n",
        "num_words"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "226460"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sFJhAj9z_U_a",
        "outputId": "5143367c-349c-4ef9-a823-e8d8586b8165"
      },
      "source": [
        "list(tokenizer.word_index)[0:5]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['OOV', 'olarak', '.', 'olan', 'sonra']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9cOjWaan_W4_"
      },
      "source": [
        "trainx_tokens = tokenizer.texts_to_sequences(trainx)\n",
        "testx_tokens = tokenizer.texts_to_sequences(testx)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "i6VMLYnP_YJK",
        "outputId": "450532ac-72f3-473d-ba70-ecaea56e8ada"
      },
      "source": [
        "len(trainx_tokens[0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "86"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gX5YDgQz_eyl",
        "outputId": "26e893b3-78bf-4149-83c1-233e8b103d50"
      },
      "source": [
        "len(trainx[0].split())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "86"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_xS5Oqem_0M8"
      },
      "source": [
        "num_tokens = [len(tokens) for tokens in trainx_tokens]\n",
        "num_tokens = np.array(num_tokens)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pArS9UU2DcMj",
        "outputId": "e3b6c433-c865-4ea9-c2de-d3b3e672f4c7"
      },
      "source": [
        "num_tokens[0]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "86"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eQi-n08yDcdu",
        "outputId": "48e7ac8b-8134-4049-a9e1-09dd6ae81d65"
      },
      "source": [
        "len(trainx[np.argmax(num_tokens)].split())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "6448"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7LfwE3ZIDepC",
        "outputId": "31d84cd7-e10b-4ce2-aade-f500338b2885"
      },
      "source": [
        "max_tokens = np.mean(num_tokens) + 2 * np.std(num_tokens)\n",
        "max_tokens = int(max_tokens)\n",
        "max_tokens"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "634"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "it2QA-IpDgeQ",
        "outputId": "53218c28-1142-4183-8256-62eb7c57c6f3"
      },
      "source": [
        "np.sum(num_tokens < max_tokens) / len(num_tokens)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.97475"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RGJQEOUHDhf-"
      },
      "source": [
        "trainx_pad = pad_sequences(trainx_tokens, maxlen=max_tokens,padding='post')\n",
        "testx_pad = pad_sequences(testx_tokens, maxlen=max_tokens,padding='post')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qCXzyEgKDjHa",
        "outputId": "b37cc164-9d12-4b6e-bc2e-54395a679d7b"
      },
      "source": [
        "print(\"trainx_tokens :\",trainx_tokens[0])\n",
        "print(\"trainx_pad :\",trainx_pad[0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "trainx_tokens : [1295, 81, 105, 606, 4964, 3498, 11, 5752, 356, 40178, 458, 91, 415, 1270, 9790, 485, 13886, 8519, 18798, 4, 415, 1270, 376, 1914, 5387, 5752, 3743, 13, 199, 4964, 3498, 18799, 893, 1295, 81, 105, 606, 4964, 3498, 716, 658, 100, 5621, 1720, 2889, 1602, 9106, 352, 415, 1270, 4964, 20264, 495, 446, 583, 96, 15457, 3803, 63799, 16442, 95160, 1721, 26667, 811, 301, 485, 802, 34262, 105, 606, 4, 4964, 3498, 1915, 4654, 862, 1416, 1055, 17511, 4850, 390, 583, 96, 17512, 49159, 63800]\n",
            "trainx_pad : [ 1295    81   105   606  4964  3498    11  5752   356 40178   458    91\n",
            "   415  1270  9790   485 13886  8519 18798     4   415  1270   376  1914\n",
            "  5387  5752  3743    13   199  4964  3498 18799   893  1295    81   105\n",
            "   606  4964  3498   716   658   100  5621  1720  2889  1602  9106   352\n",
            "   415  1270  4964 20264   495   446   583    96 15457  3803 63799 16442\n",
            " 95160  1721 26667   811   301   485   802 34262   105   606     4  4964\n",
            "  3498  1915  4654   862  1416  1055 17511  4850   390   583    96 17512\n",
            " 49159 63800     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0     0     0\n",
            "     0     0     0     0     0     0     0     0     0     0]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "G4YfUwpjMGtw"
      },
      "source": [
        "# CNN with Random Embeddings"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WDiQ0n6pD8gB",
        "outputId": "b862c2fb-228b-4c2c-dbcb-62de3d545195"
      },
      "source": [
        "weight_decay = 1e-4\n",
        "# Building the CNN Model\n",
        "model = Sequential()      # initilaizing the Sequential nature for CNN model\n",
        "# Adding the embedding layer which will take in maximum of 450 words as input and provide a 32 dimensional output of those words which belong in the top_words dictionary\n",
        "model.add(Embedding(num_words+1, 100, input_length=max_tokens))\n",
        "model.add(Conv1D(16, 4, padding='same', activation='relu'))\n",
        "model.add(MaxPooling1D())\n",
        "model.add(MaxPooling1D())\n",
        "model.add(Flatten())\n",
        "model.add(Dense(200, activation='relu',kernel_regularizer=regularizers.l2(weight_decay)))\n",
        "model.add(Dropout(0.75))\n",
        "model.add(Dense(5, activation='softmax'))\n",
        "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "model.summary()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_6\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_6 (Embedding)      (None, 634, 100)          22646100  \n",
            "_________________________________________________________________\n",
            "conv1d_6 (Conv1D)            (None, 634, 16)           6416      \n",
            "_________________________________________________________________\n",
            "max_pooling1d_12 (MaxPooling (None, 317, 16)           0         \n",
            "_________________________________________________________________\n",
            "max_pooling1d_13 (MaxPooling (None, 158, 16)           0         \n",
            "_________________________________________________________________\n",
            "flatten_6 (Flatten)          (None, 2528)              0         \n",
            "_________________________________________________________________\n",
            "dense_12 (Dense)             (None, 200)               505800    \n",
            "_________________________________________________________________\n",
            "dropout_6 (Dropout)          (None, 200)               0         \n",
            "_________________________________________________________________\n",
            "dense_13 (Dense)             (None, 5)                 1005      \n",
            "=================================================================\n",
            "Total params: 23,159,321\n",
            "Trainable params: 23,159,321\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 976
        },
        "id": "e2CBg_1Y8d33",
        "outputId": "e1bf120d-0e3f-4e03-d03b-405af9b2e942"
      },
      "source": [
        "early_stopping = EarlyStopping(monitor='val_loss', min_delta=0.01, patience=2, verbose=1)\n",
        "callbacks_list = [early_stopping]\n",
        "from tensorflow import keras\n",
        "keras.utils.plot_model(model, \"random_embeddings.png\", show_shapes=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Uqz5zwPPEEvQ",
        "outputId": "2b89098b-eefa-4613-92ee-5736c87c12a4"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "trainx_pad, valx_pad, trainy, valy = train_test_split(trainx_pad, trainy, test_size=0.1, random_state=42)\n",
        "\n",
        "# Fitting the data onto model\n",
        "model.fit(trainx_pad, trainy, validation_data=(valx_pad, valy), epochs=6, batch_size=16, verbose=1,callbacks=callbacks_list)\n",
        "# Getting score metrics from our model\n",
        "scores = model.evaluate(testx_pad, testy, verbose=0)\n",
        "# Displays the accuracy of correct sentiment prediction over test data\n",
        "print(\"Accuracy: %.2f%%\" % (scores[1]*100))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/6\n",
            "240/240 [==============================] - 52s 214ms/step - loss: 1.4576 - accuracy: 0.3695 - val_loss: 1.1028 - val_accuracy: 0.5200\n",
            "Epoch 2/6\n",
            "240/240 [==============================] - 52s 215ms/step - loss: 0.9433 - accuracy: 0.6111 - val_loss: 0.6995 - val_accuracy: 0.7129\n",
            "Epoch 3/6\n",
            "240/240 [==============================] - 51s 214ms/step - loss: 0.3915 - accuracy: 0.8717 - val_loss: 0.6382 - val_accuracy: 0.8376\n",
            "Epoch 4/6\n",
            "240/240 [==============================] - 51s 214ms/step - loss: 0.1820 - accuracy: 0.9517 - val_loss: 0.7709 - val_accuracy: 0.7953\n",
            "Epoch 5/6\n",
            "240/240 [==============================] - 51s 214ms/step - loss: 0.1067 - accuracy: 0.9768 - val_loss: 0.8171 - val_accuracy: 0.7976\n",
            "Epoch 00005: early stopping\n",
            "Accuracy: 82.15%\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 445
        },
        "id": "a3drvw9eBSfv",
        "outputId": "19563a83-ab64-4cb8-f218-7ae0a1406d48"
      },
      "source": [
        "import sklearn.metrics as metrics,classification_report\n",
        "\n",
        "# rounded_predictions = model.predict_classes(test_imag)\n",
        "y_pred_ohe = model.predict(testx_pad)  # shape=(n_samples, 12)\n",
        "y_pred_labels = np.argmax(y_pred_ohe, axis=1)  # only necessary if output has one-hot-encoding, shape=(n_samples)\n",
        "y_pred_labels[0:10]\n",
        "\n",
        "# y_pred= np.zeros((y_pred_labels.size, y_pred_labels.max()+1))\n",
        "# y_pred[np.arange(y_pred_labels.size),y_pred_labels] = 1\n",
        "\n",
        "confusion_matrix = metrics.confusion_matrix(y_true=testy_integer, y_pred=y_pred_labels)  # shape=(12, 12)\n",
        "\n",
        "import seaborn as sn\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "df_cm = pd.DataFrame(confusion_matrix, index = [i for i in [\"spor\",\"yazarlar\",\"video\",\"dunya\",\"turkiye\"]],\n",
        "                  columns = [i for i in [\"spor\",\"yazarlar\",\"video\",\"dunya\",\"turkiye\"]])\n",
        "plt.figure(figsize = (10,10))\n",
        "sn.heatmap(df_cm, annot=True)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f9d426485f8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 51
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x504 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qK3_2FISEaub",
        "outputId": "3c4a99b7-5cf9-4da5-b0c4-20fc305bcefe"
      },
      "source": [
        "target_names = [\"spor\",\"yazarlar\",\"video\",\"dunya\",\"turkiye\"]\n",
        "print(classification_report(y_pred_labels,testy_integer, target_names=target_names))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "        spor       0.93      0.94      0.94       383\n",
            "    yazarlar       0.95      0.91      0.93       411\n",
            "       video       0.66      0.88      0.75       306\n",
            "       dunya       0.77      0.81      0.79       376\n",
            "     turkiye       0.80      0.64      0.71       524\n",
            "\n",
            "    accuracy                           0.82      2000\n",
            "   macro avg       0.82      0.84      0.82      2000\n",
            "weighted avg       0.83      0.82      0.82      2000\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aCkTaySHMOmN"
      },
      "source": [
        "# CNN with akoksal Embeddings"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0xKzvSsfEbTP",
        "outputId": "e251ea2e-059b-4dec-fdcc-b5793315883a"
      },
      "source": [
        "from gensim.models import KeyedVectors\n",
        "word2vec = KeyedVectors.load_word2vec_format(\"/content/drive/MyDrive/Colab Notebooks/nlp-project-3-dataset/trmodel\", binary=True)\n",
        "\n",
        "embedding_matrix = np.zeros((num_words+1, 400))\n",
        "for word, i in tokenizer.word_index.items():\n",
        "    if i>=num_words:\n",
        "        continue\n",
        "    try:\n",
        "        embedding_vector = word2vec[word]\n",
        "        embedding_matrix[i] = embedding_vector\n",
        "    except KeyError:\n",
        "        embedding_matrix[i]=np.random.normal(0,np.sqrt(0.25),400)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "2021-01-05 01:12:52,202 : INFO : loading projection weights from /content/drive/MyDrive/Colab Notebooks/nlp-project-3-dataset/trmodel\n",
            "2021-01-05 01:13:01,287 : INFO : loaded (412457, 400) matrix from /content/drive/MyDrive/Colab Notebooks/nlp-project-3-dataset/trmodel\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "chexaNahEp1B",
        "outputId": "48630a31-bea5-4ea9-b45e-afaab0e07fd8"
      },
      "source": [
        "embedding_matrix.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(226461, 400)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LGBIO86pEqcc",
        "outputId": "d6f4094d-4471-4892-b7ae-756cdbb1eb0c"
      },
      "source": [
        "from keras.initializers import Constant\n",
        "weight_decay = 1e-4\n",
        "# # Building the CNN Model\n",
        "model = Sequential()      # initilaizing the Sequential nature for CNN model\n",
        "# Adding the embedding layer which will take in maximum of 450 words as input and provide a 32 dimensional output of those words which belong in the top_words dictionary\n",
        "model.add(Embedding(num_words+1,\n",
        "                            400,\n",
        "                            weights=[embedding_matrix],\n",
        "                            # embeddings_initializer = Constant(embedding_matrix),\n",
        "                            input_length=max_tokens,\n",
        "                            trainable=False))\n",
        "model.add(Conv1D(16, 4, padding='same', activation='relu'))\n",
        "model.add(MaxPooling1D())\n",
        "model.add(MaxPooling1D())\n",
        "model.add(Flatten())\n",
        "model.add(Dense(200, activation='relu',kernel_regularizer=regularizers.l2(weight_decay)))\n",
        "model.add(Dropout(0.75))\n",
        "model.add(Dense(5, activation='softmax'))\n",
        "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "model.summary()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_9\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_9 (Embedding)      (None, 634, 400)          90584400  \n",
            "_________________________________________________________________\n",
            "conv1d_9 (Conv1D)            (None, 634, 16)           25616     \n",
            "_________________________________________________________________\n",
            "max_pooling1d_18 (MaxPooling (None, 317, 16)           0         \n",
            "_________________________________________________________________\n",
            "max_pooling1d_19 (MaxPooling (None, 158, 16)           0         \n",
            "_________________________________________________________________\n",
            "flatten_9 (Flatten)          (None, 2528)              0         \n",
            "_________________________________________________________________\n",
            "dense_18 (Dense)             (None, 200)               505800    \n",
            "_________________________________________________________________\n",
            "dropout_9 (Dropout)          (None, 200)               0         \n",
            "_________________________________________________________________\n",
            "dense_19 (Dense)             (None, 5)                 1005      \n",
            "=================================================================\n",
            "Total params: 91,116,821\n",
            "Trainable params: 532,421\n",
            "Non-trainable params: 90,584,400\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 976
        },
        "id": "Jx9zQvP8G5Qo",
        "outputId": "4fecb563-c8a3-4b3d-ce79-afd19f1952a1"
      },
      "source": [
        "keras.utils.plot_model(model, \"github_embeddings.png\", show_shapes=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 66
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ivufdtUKEyO_",
        "outputId": "63f6547c-5cab-428d-bdf7-e8e80eb5383b"
      },
      "source": [
        "# Fitting the data onto model\n",
        "model.fit(trainx_pad, trainy, validation_data=(valx_pad, valy), epochs=30, batch_size=32, verbose=1,callbacks=callbacks_list)\n",
        "# Getting score metrics from our model\n",
        "scores = model.evaluate(testx_pad, testy, verbose=0)\n",
        "# Displays the accuracy of correct sentiment prediction over test data\n",
        "print(\"Accuracy: %.2f%%\" % (scores[1]*100))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/30\n",
            "120/120 [==============================] - 2s 16ms/step - loss: 1.6780 - accuracy: 0.4246 - val_loss: 0.7937 - val_accuracy: 0.7294\n",
            "Epoch 2/30\n",
            "120/120 [==============================] - 2s 13ms/step - loss: 0.8123 - accuracy: 0.7174 - val_loss: 0.6487 - val_accuracy: 0.7929\n",
            "Epoch 3/30\n",
            "120/120 [==============================] - 2s 13ms/step - loss: 0.6125 - accuracy: 0.7982 - val_loss: 0.5391 - val_accuracy: 0.8212\n",
            "Epoch 4/30\n",
            "120/120 [==============================] - 2s 14ms/step - loss: 0.4580 - accuracy: 0.8454 - val_loss: 0.5631 - val_accuracy: 0.8353\n",
            "Epoch 5/30\n",
            "120/120 [==============================] - 2s 13ms/step - loss: 0.3876 - accuracy: 0.8750 - val_loss: 0.5404 - val_accuracy: 0.8282\n",
            "Epoch 00005: early stopping\n",
            "Accuracy: 80.50%\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 609
        },
        "id": "ar3b1VX5F3nr",
        "outputId": "cf4723d4-7bfa-46ff-ab01-8f1cac4b2df4"
      },
      "source": [
        "\n",
        "# rounded_predictions = model.predict_classes(test_imag)\n",
        "y_pred_ohe = model.predict(testx_pad)  # shape=(n_samples, 12)\n",
        "y_pred_labels = np.argmax(y_pred_ohe, axis=1)  # only necessary if output has one-hot-encoding, shape=(n_samples)\n",
        "y_pred_labels[0:10]\n",
        "\n",
        "# y_pred= np.zeros((y_pred_labels.size, y_pred_labels.max()+1))\n",
        "# y_pred[np.arange(y_pred_labels.size),y_pred_labels] = 1\n",
        "\n",
        "confusion_matrix = metrics.confusion_matrix(y_true=testy_integer, y_pred=y_pred_labels)  # shape=(12, 12)\n",
        "\n",
        "import seaborn as sn\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "df_cm = pd.DataFrame(confusion_matrix, index = [i for i in [\"spor\",\"yazarlar\",\"video\",\"dunya\",\"turkiye\"]],\n",
        "                  columns = [i for i in [\"spor\",\"yazarlar\",\"video\",\"dunya\",\"turkiye\"]])\n",
        "plt.figure(figsize = (10,10))\n",
        "sn.heatmap(df_cm, annot=True)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f9d42444eb8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 70
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uwWo4DhmF6Vo",
        "outputId": "e09527b9-7e6d-4295-ea65-0529c56ad913"
      },
      "source": [
        "target_names = [\"spor\",\"yazarlar\",\"video\",\"dunya\",\"turkiye\"]\n",
        "print(classification_report(y_pred_labels,testy_integer, target_names=target_names))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "        spor       0.88      0.94      0.91       359\n",
            "    yazarlar       0.96      0.76      0.85       500\n",
            "       video       0.86      0.74      0.79       473\n",
            "       dunya       0.76      0.84      0.80       356\n",
            "     turkiye       0.58      0.79      0.67       312\n",
            "\n",
            "    accuracy                           0.81      2000\n",
            "   macro avg       0.81      0.81      0.80      2000\n",
            "weighted avg       0.83      0.81      0.81      2000\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fgYkUwsmMUip"
      },
      "source": [
        "# CNN with homework2 Embeddings"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JfVA7J3WE1EO"
      },
      "source": [
        "word2vec_hw2 = {}\n",
        "with open(\"/content/drive/MyDrive/Colab Notebooks/10kwe_32\", encoding='UTF-8') as f:\n",
        "    for line in f:\n",
        "        values = line.split()\n",
        "        word = values[0]\n",
        "        vec = np.asarray(values[1:], dtype='float32')\n",
        "        word2vec_hw2[word] = vec\n",
        "\n",
        "embedding_matrix_hw2 = np.zeros((num_words+1, 32))\n",
        "for word, i in tokenizer.word_index.items():\n",
        "    if i>=num_words:\n",
        "        continue\n",
        "    try:\n",
        "        embedding_vector = word2vec_hw2[word]\n",
        "        embedding_matrix_hw2[i] = embedding_vector\n",
        "    except KeyError:\n",
        "        embedding_matrix_hw2[i]=np.random.normal(0,np.sqrt(0.25),32)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "X4OVC5qMFSbX",
        "outputId": "1cc5bcd9-d91c-43b1-a5bd-1e01311a5d59"
      },
      "source": [
        "from keras.initializers import Constant\n",
        "weight_decay = 1e-4\n",
        "# # Building the CNN Model\n",
        "model = Sequential()      # initilaizing the Sequential nature for CNN model\n",
        "# Adding the embedding layer which will take in maximum of 450 words as input and provide a 32 dimensional output of those words which belong in the top_words dictionary\n",
        "model.add(Embedding(num_words + 1,\n",
        "                            32,\n",
        "                            weights=[embedding_matrix_hw2],\n",
        "                            # embeddings_initializer = Constant(embedding_matrix),\n",
        "                            input_length=max_tokens,\n",
        "                            trainable=True))\n",
        "model.add(Conv1D(32, 8, padding='same', activation='relu'))\n",
        "model.add(MaxPooling1D())\n",
        "model.add(MaxPooling1D())\n",
        "model.add(Flatten())\n",
        "model.add(Dense(200, activation='relu',kernel_regularizer=regularizers.l2(weight_decay)))\n",
        "model.add(Dropout(0.75))\n",
        "model.add(Dense(5, activation='softmax'))\n",
        "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
        "model.summary()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_10\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "embedding_10 (Embedding)     (None, 634, 32)           7246752   \n",
            "_________________________________________________________________\n",
            "conv1d_10 (Conv1D)           (None, 634, 32)           8224      \n",
            "_________________________________________________________________\n",
            "max_pooling1d_20 (MaxPooling (None, 317, 32)           0         \n",
            "_________________________________________________________________\n",
            "max_pooling1d_21 (MaxPooling (None, 158, 32)           0         \n",
            "_________________________________________________________________\n",
            "flatten_10 (Flatten)         (None, 5056)              0         \n",
            "_________________________________________________________________\n",
            "dense_20 (Dense)             (None, 200)               1011400   \n",
            "_________________________________________________________________\n",
            "dropout_10 (Dropout)         (None, 200)               0         \n",
            "_________________________________________________________________\n",
            "dense_21 (Dense)             (None, 5)                 1005      \n",
            "=================================================================\n",
            "Total params: 8,267,381\n",
            "Trainable params: 8,267,381\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 976
        },
        "id": "grcF8SDRIOPv",
        "outputId": "c6bc1e7e-40e1-4b73-f80d-9393a16e3aaa"
      },
      "source": [
        "keras.utils.plot_model(model, \"hw2_embeddings.png\", show_shapes=True)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhwAAAO/CAYAAACEAR+IAAAABmJLR0QA/wD/AP+gvaeTAAAgAElEQVR4nOzdeVyU9fo//tfADMwCDKCIHDYFFNyXtITimMcTH5OjYmSQ1Qn7amgaImouuOACuXSUhwr1KI06ueB6oJNSHi01Sz3mSlPHEEVUDFARkBlkgOv3h7+5cxyEGWCYQa/n48Ef3Pf7vt/Xfd+zXHPf70VERATGGGOMMTOysXQAjDHGGHv8ccLBGGOMMbPjhIMxxhhjZscJB2OMMcbMTtwWlaxZswbHjh1ri6oYY4wxZoLg4GAkJCSYvZ42ucNx7NgxHD9+vC2qYow9wXbt2oVr165ZOox25fjx4/z5/AQ7fvx4m90QaJM7HAAwZMgQ7Ny5s62qY4w9gUQiEWbMmIFXXnnF0qG0G+PGjQMA/nx+Qumuf1vgNhyMMcYYMztOOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzHGHrJv3z4olUr8+9//tnQoVmny5MkQiUTC3+uvv25Q5sCBA5g3bx52794NPz8/oewbb7xhUDYsLAyOjo6wtbVFr169cPr06bY4jBbRarVISUlBQEAA7Ozs4OzsjN69e6OgoEAos3LlSgQFBUEmk0GhUCAoKAgLFy5ERUVFo/uurq5GUFAQFixYICz78ssvsXLlStTV1emVzcrK0rsWHTt2bNXjbE2ccDDG2EN4Eu2mubq6IicnBxcuXMCmTZv01i1evBjr1q3D/PnzERkZiUuXLsHf3x8dOnTA5s2bsXfvXr3y+/fvx86dOzFq1CioVCoMHDiwLQ+lWaKiovDPf/4TW7ZsgVqtxq+//gp/f3/cvXtXKPP9999j0qRJKCwsRHFxMZYtW4aVK1fi5ZdfbnTfiYmJuHDhgt6y0aNHQyqVYvjw4bhz546wfMyYMbh27RqOHDmCkSNHtu5BtjJOOBhj7CHh4eEoLy/HqFGjLB0KNBoNQkJCLB2GAZlMhhEjRqB79+6wt7cXlq9YsQKZmZnYsWMHHB0d9bZZt24dbGxsEBsbi/Ly8rYOudVkZmYiKysLO3fuxDPPPAOxWAwPDw9kZ2ejd+/eQjk7OztMnToVbm5ucHBwwLhx4xAREYH//Oc/uHHjRoP7/vHHH/Hzzz83uG769Ono168fRo4cidraWgD3x57x9PREaGgounXr1voH24o44WCMMSu2adMmlJSUWDoMo1y8eBELFy7EkiVLIJVKDdaHhIQgPj4e169fx6xZsywQYev48MMPMXDgQPTp06fRcnv27DE4D56engCgdydER6PRYPbs2UhNTX3kPpOSknD27NlGy1grTjgYY+wBR48ehY+PD0QiETZs2AAASE9Ph0KhgFwuR3Z2Nl588UU4OTnBy8sL27ZtE7Zdt24dpFIpOnXqhMmTJ8PDwwNSqRQhISE4ceKEUC4uLg52dnbo3LmzsGzq1KlQKBQQiUS4efMmACA+Ph4zZ85Efn4+RCIRAgICAABff/01nJyckJyc3BanxGjr1q0DEWH06NGPLLN8+XJ0794dGzduxIEDBxrdHxFhzZo16NGjB+zt7eHi4oKIiAj873//E8oYe20AoK6uDosWLYKPjw9kMhn69u2L7du3m3SMNTU1OH78OPr372/Sdjp5eXlwdnaGr6+vwbrExEThjsijuLi4YOjQoUhNTW13j/444WCMsQc899xz+PHHH/WWvfPOO5gxYwY0Gg0cHR2xfft25Ofnw8/PD5MmTYJWqwVwP5GIiYmBWq3G9OnTUVBQgNOnT6O2thYvvPACrl69CuD+F/PDw6+npaVhyZIlestSU1MxatQo+Pv7g4hw8eJFABAaDtbX15vlHDTX3r17ERgYCLlc/sgyMpkMn332GWxsbDBp0iRUVVU9smxSUhLmzZuHxMRElJSU4MiRI7h69SpCQ0NRXFwMwPhrAwBz587FqlWrsHbtWty4cQOjRo3C+PHj8dNPPxl9jEVFRaipqcGpU6cwbNgwIans0aMH0tLSGkwCtFotrl+/jg0bNuDAgQNYv3497Ozs9Mr88MMPyM/Px/jx45uMYcCAAbh+/TrOnTtndNzWgBMOxhgzQUhICJycnODm5obo6GhUVVWhsLBQr4xYLBZ+lffs2RPp6emorKxERkZGq8QQHh6OiooKLFy4sFX21xqqqqpw+fJl+Pv7N1k2ODgYM2bMQEFBAebOndtgGY1GgzVr1uCll17C66+/DqVSiT59+uCjjz7CzZs38fHHHxts09i1qa6uRnp6OsaOHYvIyEg4OztjwYIFkEgkJl0X3aMQNzc3JCcnQ6VSobi4GBEREZg2bRq2bt1qsI23tze8vLyQlJSEVatWISoqyuBY4+PjkZ6eblQMurYaubm5RsdtDTjhYIyxZtL9Sn3wV3RDBg0aBLlcrvco4HFTUlICImr07saDli9fjsDAQKSlpeHo0aMG61UqFe7evYtBgwbpLR88eDDs7Oz0HlE15OFrc+HCBajVar1GnTKZDJ07dzbpuugayPbq1QshISFwdXWFUqnEkiVLoFQqG0yErl69ipKSEmzduhWff/45BgwYoNcuZ/78+Xj77beF9h1N0Z1j3V2e9oITDsYYawP29vYoLS21dBhmU11dDQB6PVYaI5VKkZGRAZFIhLfeegsajUZvva7rp4ODg8G2zs7OqKysNCk+3aObBQsW6I1bceXKFajVaqP34+HhAQBCOxsdOzs7+Pr6Ij8/32AbiUQCNzc3hIWFITMzEyqVCikpKQDutxnKzc3FxIkTjY5BJpMB+OOctxeccDDGmJlptVrcuXMHXl5elg7FbHRfgg8PTNWY4OBgJCQkIC8vD8uWLdNb5+zsDAANJhbNOZe6hphr164FEen9HTt2zOj9ODg4oFu3bvjll18M1tXW1kKpVDa6fUBAAGxtbaFSqQDc74V08OBB2NjYCEmQLtbk5GSIRCKDNiY1NTUA/jjn7QUnHIwxZmaHDh0CEWHIkCHCMrFY3OSjmPakU6dOEIlEJo+vsWzZMgQFBeHMmTN6y3v37g0HBweDL9sTJ06gpqYGTz31lEn1eHt7QyqV4uzZsyZt15CoqCicOXMGly5dEpap1WpcuXJF6Cp769atBhuA5uXloa6uDt7e3gCAjIwMgwRIdycsMTERRGTwWEl3jt3d3Vt8LG2JEw7GGGtl9fX1KCsrQ21tLc6fP4/4+Hj4+PggJiZGKBMQEIDbt28jKysLWq0WpaWluHLlisG+XF1dUVRUhIKCAlRWVkKr1SInJ8fqusXK5XL4+fnh2rVrJm2ne7Ria2trsHzmzJnYs2cPNm/ejIqKCuTm5mLKlCnw8PBAbGysyfVMmDAB27ZtQ3p6OioqKlBXV4dr164Jg3BFR0fD3d29yaHVExIS4Ovri5iYGBQWFuLWrVuYM2cONBqN0AhWoVBg//79+Pbbb1FRUQGtVoszZ87gzTffhEKhQEJCgknxP0h3jpsaB8TacMLBGGMP2LBhAwYPHgwAmDNnDsaMGYP09HSsXbsWANC3b19cunQJn3zyCWbOnAkAGDFiBPLy8oR9VFdXo0+fPpDJZAgNDUX37t3x3Xff6bVveOeddzBs2DC8+uqrCAwMxLJly4Rb5MHBwUIX2ilTpqBTp07o2bMnRo4cidu3b7fJeWiO8PBwqFQqvfYY//rXvxAQEID8/HwMHjwY7777rsF2Q4YMafALePHixUhJScHSpUvRsWNHDB06FF26dMGhQ4egUCgAwKRrk5qaihkzZmDlypXo0KEDPDw8EB8fj7KyMgD3H1WUlJQgOzu70eN0cXHB999/Dy8vL/Tv3x+enp7473//i7179wrjc0ilUjz77LOYOHEiPD094ejoiHHjxqFLly44fvy4XuNVU508eRKenp7o27dvs/dhEdQGXn75ZXr55ZfboirG2BMMAG3fvt2iMcTGxpKrq6tFYzBFcz6fY2NjydPT02B5Xl4eicVi+uKLL1orvDZVV1dHoaGhtGnTJkuH8kg3b94kqVRKH3zwgcG66dOnU4cOHUzaX1t+P/MdDsYYa2WmNJxsrzQaDb755hvk5eUJjRgDAgKwdOlSLF26tMGhu61ZXV0dsrKyUFlZiejoaEuH80hJSUno378/4uLiANwfjbWoqAhHjx4VBoazVpxwMMYYM9nt27eFydveeustYfm8efMwbtw4REdHt6sJ2g4dOoTdu3cjJyfH6LFE2tqaNWtw9uxZ7Nu3DxKJBACQnZ0tTN728Cy81uaJSzgGDx4MW1vbZo+D35iJEyfC0dERIpGo0ZbQjyq3b98+KJVK/Pvf/2712Jqrvr4ea9eubXS2yqNHj+LZZ5+FXC6Hh4cH5syZg3v37jWrPms8B6Y6fvw4evToIXRzc3d3x/Llyy0dlp7du3fDz89P6IbXuXNnvP7665YOq92bP38+MjIyUF5ejq5du2LXrl2WDsksPvroI71eFZs3b9Zbn5ycjLi4OLz//vsWitB0w4cPx5YtW/Tmt7Em2dnZuHfvHg4dOgQXFxdheUREhN61eHh8EGvyxCUcJ0+exLBhw8yy740bN+KTTz5pdjmysol48vLy8Oc//xkJCQmPHBhHpVIhLCwMw4cPR2lpKfbs2YNPP/0UU6ZMaVad1nYOmmPIkCH49ddfERYWBuD+CIcLFiywcFT6IiMjcenSJfj7+0OpVOL33383+NJgpktJScG9e/dARLh8+TJefvllS4dkMWFhYVixYoWlw3hsjBkzBvPmzTPozdOePHEJh45IJLJ0CAbCw8NRXl6OUaNGWToUnDt3DnPnzsWUKVMavRu0bNkydO7cGUuWLIFCoUBwcDDmzJmDzz77rFnDOFvTOdBoNI3e2WlPHqdjYYy1T09swqF7/tXajE1k2iLhISLs3LmzwbH9m9KvXz/s3r0br7322iOHKq6trcXevXsxdOhQveN58cUXQURNdi2zdps2bdKb76A9e5yOhTHWPlltwlFXV4dFixbBx8cHMpkMffv2xfbt2wHc70utUChgY2ODp556Cu7u7pBIJFAoFBg4cCBCQ0OFUeWcnZ3x3nvvGez/4sWLCAoKgkKhEPrKPzyBUGMxAPe/0FevXo3AwEDY29tDqVRi9uzZBnUZU+7o0aPw8fGBSCTChg0bANzvX65QKCCXy5GdnY0XX3wRTk5O8PLywrZt2wxiTUlJQWBgIGQyGTp27IiuXbsiJSXFYBrs1nLp0iXcvXsXPj4+est1s0WeP3/epP215BysW7cOUqkUnTp1wuTJk4Upo0NCQvQmeYqLi4OdnZ3ec9qpU6dCoVBAJBIJzz/j4+Mxc+ZM5OfnQyQSISAgAADw9ddfN3vAJWs7FlN9//336NmzJ5RKJaRSKfr06YNvvvkGwP12Sbr2IP7+/sKokRMmTIBcLodSqcSXX34JoPH31apVqyCXy+Ho6IiSkhLMnDkTnp6euHDhQrNiZoxZkbboe9ucfr6zZs0ie3t72rVrF5WVldH8+fPJxsaGTp48SUREixcvJgB04sQJqqqqops3b9KIESMIAO3du5dKS0upqqqK4uLiCACdPXtW2Pfw4cPJz8+PLl++TFqtln7++Wd65plnSCqV0m+//WZ0DImJiSQSiegf//gHlZWVkVqtprS0NAJAZ86cEfZjbLmrV68SAFq/fr3etgDo4MGDVF5eTiUlJRQaGkoKhYJqamqEcsnJyWRra0vZ2dmkVqvp1KlT5O7uTs8//7xJ570hzzzzDPXr189g+eHDhwkArV692mCdTCaj4cOHm1xXS85BbGwsKRQK+uWXX6i6uppUKhUNHjyYHB0dqbCwUCj32muvkbu7u169q1evJgBUWloqLIuMjCR/f3+9cl999RU5OjrS0qVLmzyW//u//yMAVFZWZpXHQkTk7+9PSqWyyWMhItq5cyclJSXR7du36datWzRkyBC9Pv+RkZFka2tL169f19tu/Pjx9OWXXwr/G/O+AkDTp0+n9evX00svvUS//vqrUTHCCsbhaG94nKQn2xM/Dkd1dTXS09MxduxYREZGwtnZGQsWLIBEIkFGRoZe2Z49e0Iul6NDhw549dVXAQA+Pj7o2LEj5HK50PL+4fYEjo6O6NKlC8RiMXr16oVPPvkE1dXVwuOHpmLQaDRYu3Yt/vrXvyIhIQHOzs6QyWRwdXXVq8fYck0JCQmBk5MT3NzcEB0djaqqKhQWFgrrs7Ky8NRTT2H06NGQyWQYOHAgxowZgyNHjgh95FubridKQ42YJBKJweyPLdXUOQDuz0/Ro0cP2Nvbo2fPnkhPT0dlZaXB66a5wsPDUVFRgYULF7ZoP9ZwLKZ6+eWXsXjxYri4uMDV1RWjR4/GrVu3hHkfpkyZgrq6Or34KioqcPLkSYwcORKAae/tFStWYNq0adi9ezeCgoLa7kAZY2YhtnQADblw4QLUarXe0K8ymQydO3dutCGinZ0dgPttC3R0bTWamiSpT58+UCqVwmOApmK4ePEi1Go1hg8f3uh+jS1nCt1xPnhM1dXVkEqleuXq6uogkUjM1qpZV9+D51unpqbGrDMZNnQOGjJo0CDI5fJmNWBtK+31WHTvLd0gV3/5y1/QvXt3fPrpp5g/fz5EIhEyMzMRHR0tvAab+942RVRUFKKiolplX08Sa2xIz9pGW/WmssqEo6qqCgCwYMECg+6EHh4eZqtXIpEIH/pNxaCbPEc3jfCjGFuupUaOHInVq1cjOzsbYWFhUKlUyMrKwt/+9jezJRy6tgMVFRV6y9VqNaqrq816rUxhb28v/Apv7yx5LHv37sXq1auhUqmEyageJBKJMHnyZCQkJODgwYP461//in/+85/YsmWLUKYt3tvx8fEIDg5ulX09CXTzkMyYMcPCkTBL0F3/tmCVCYfuy3nt2rWIj49vkzpra2tx+/ZtoQFkUzF89913ANDkAFe6uwDNHQjLWElJSTh16hRiYmJw9+5deHh44JVXXjHrbJJdu3aFo6OjwQyXuuF1rWFiIa1Wizt37sDLy8vSobRYWx/LkSNHcOrUKcyYMQOFhYUYO3YsXnrpJXz66af405/+hPXr1xs0yI6JicH8+fOxceNGeHt7w8nJCb6+vsL6tnhvBwcHm62h9ONo586dAMDn7Amlu/5twSoTDl0Pk8ZG62xt3333Herr6zFw4ECjYujduzdsbGxw+PDhRge5MrZcS6lUKuTn56O0tBRicdtcVrFYjJEjR+LIkSOor6+Hjc39JkE5OTkQiUQYPXp0m8TRmEOHDoGIMGTIEGGZWCxu8vGFNWrrYzl16pQwI2dubi60Wi3eeecd+Pn5AWj4FryLiwuioqKQmZkJR0dHTJo0SW+9Jd7bjDHrYJWNRqVSKSZMmIBt27YhPT0dFRUVqKurw7Vr13Djxo1WqaOmpgbl5eWora3F6dOnERcXB19fX8TExBgVg5ubGyIjI7Fr1y5s2rQJFRUVOH/+vMGYF8aWa6lp06bBx8enzSdMWrhwIYqLi7F48WJUVVXh2LFjWL16NWJiYhAYGNimsQD3h2IvKytDbW0tzp8/j/j4ePj4+AjXFbg/wdTt27eRlZUFrVaL0tJSg7s0AODq6oqioiIUFBSgsrISWq0WOTk5ze4Wa23H8iharRbFxcV6U4Dr7vwdOHAA1dXVyMvL0+ui+6ApU6bg3r17+OqrrwwGcGuL9zZjzEq1RVeY5nS7uXfvHs2ZM4d8fHxILBaTm5sbRUZGkkqlotTUVJLL5QSAunTpQt9//z2tWLGClEolASB3d3fasmULZWZmkru7OwEgFxcX2rZtGxERZWRk0LBhw6hTp04kFoupQ4cO9Oqrr9KVK1eMjoGIqLKykiZOnEgdOnQgBwcHeu6552jRokUEgLy8vOjcuXNGl1u/fj117tyZAJBcLqfRo0dTWlqacJzdunWj/Px8+vjjj8nJyYkAkK+vr9CN99tvv6UOHToQAOFPIpFQjx49aPfu3SZfs2PHjtGzzz5LHh4ewv46d+5MISEhdPjwYb2yhw8fpqeffprs7e3Jw8ODZs+eTdXV1SbX2dJzEBsbSxKJhDw9PUksFpOTkxNFRERQfn6+Xj23bt2iYcOGkVQqpa5du9K7775Ls2fPJgAUEBAgdDs9ffo0+fr6kkwmo+eee45+//132rdvHzk6OtLy5csfeRzHjx+nXr16kY2NjXDekpOTrepYPvzwQ/L399d7vTT0t2fPHqGuOXPmkKurKzk7O9O4ceNow4YNBID8/f31uuoSEQ0YMIDmzZvX4Plp7H21cuVKkslkBIC8vb1NnuYc3C3WZNwt9snWltdfRGT+ySvGjRsHoG2fFT1p0tPTkZeXp9cAqKamBnPnzkV6ejrKysrM2mvEGkyePBk7d+7ErVu3LB1Ki7X3YwkPD8eGDRvQtWvXNq1XJBJh+/bt3B7BBPz5/GRry+tvlW04mGl+//13xMXFGTwXt7Ozg4+PD7RaLbRa7WOfcAB/dNF8HLSnY9FqtUI32fPnz0MqlbZ5ssEYs25W2YaDmUYmk0EikWDTpk0oLi6GVqtFUVERNm7ciEWLFiE6OhpFRUXC0NON/UVHR7dqbP/73/8sUi9rW3PmzEFeXh5+++03TJgwAcuWLbN0SMyMJk+erPf+1Q2w+KADBw5g3rx52L17N/z8/ISyb7zxhkHZsLAwODo6wtbWFr169cLp06fb4jBaRKvVIiUlBQEBAbCzs4OzszN69+6NgoICoczKlSsRFBQEmUwGhUKBoKAgLFy40GAogYdVV1cjKChIr+v4l19+iZUrVxr8EMnKytK7Fh07dmzV42xVbfHchp8Rmt+RI0for3/9Kzk5OZGtrS0plUoKCQmhtLQ00mq1lg7P7ObNm0d2dnZCu56dO3daOqRma4/HkpiYSDY2NuTt7a03jHlbA7fhMFlzPp9jY2PJ1dWVcnJy6MKFCwZtthYtWkSjRo2iiooKYZm/v7/Qzuyrr74y2GdOTg6NGTOmeQdhAWPHjqXAwEA6fvw4abVaKioqotGjR1Nubq5QJjw8nD744AMqKSmhyspK2rFjB0kkEnrhhRca3XdCQgIBoMTERL3lqampNHToUL3pEurr6+natWt05MgRGjlypN50A8Zoy+9nTjgYY48NSyccarWagoOD21UdzU04PD09G1z3/vvvU/fu3Umj0egt9/f3py1btpCNjQ15enrSnTt39Na3p4Rj27ZtJBKJ6Pz5842WGzt2rMF5GDduHAGgoqKiBrf54YcfKCwsrMGEg4goLi6OgoODG/whOX36dKtOOPiRCmOMtZJNmzahpKSk3dfRXBcvXsTChQuxZMkSg6kWgPtzCMXHx+P69euYNWuWBSJsHR9++CEGDhyIPn36NFpuz549BufB09MTABocwkCj0WD27NlITU195D6TkpJw9uzZRstYK044GGNPLCLCmjVrhEnyXFxcEBERoTevS1xcHOzs7ISh/AFg6tSpUCgUEIlEuHnzJoD7Q6rPnDkT+fn5EIlECAgIwLp16yCVStGpUydMnjwZHh4ekEqlCAkJ0RvHpCV1AMDXX3/dZuPDNGbdunUgokYH/Vu+fDm6d++OjRs34sCBA43uz5jrk56eDoVCAblcjuzsbLz44otwcnKCl5cXtm3bpre/uro6LFq0CD4+PpDJZOjbty+2b99u0jHW1NTg+PHj6N+/v0nb6eTl5cHZ2VlvBF6dxMRETJ06tdGpMFxcXDB06FCkpqaCzN/JtFVxwsEYe2IlJSVh3rx5SExMRElJCY4cOYKrV68iNDQUxcXFAO5/iT7czTYtLQ1LlizRW5aamopRo0bB398fRISLFy8iLi4OMTExUKvVmD59OgoKCnD69GnU1tbihRdewNWrV1tcB/BHj6b6+vrWOznNsHfvXgQGBkIulz+yjEwmw2effQYbGxtMmjRJmF+nIcZcn3feeQczZsyARqOBo6Mjtm/fjvz8fPj5+WHSpEl6g9zNnTsXq1atwtq1a3Hjxg2MGjUK48ePx08//WT0MRYVFaGmpganTp3CsGHDhCSyR48eSEtLazAJ0Gq1uH79OjZs2IADBw5g/fr1wqSNOj/88APy8/Mxfvz4JmMYMGAArl+/jnPnzhkdtzXghIMx9kTSaDRYs2YNXnrpJbz++utQKpXo06cPPvroI9y8ebNVRwMWi8XCr/SePXsiPT0dlZWVyMjIaJX9h4eHo6KiAgsXLmyV/TVHVVUVLl++DH9//ybLBgcHY8aMGSgoKMDcuXMbLNOc6xMSEgInJye4ubkhOjoaVVVVKCwsBHC/50d6ejrGjh2LyMhIODs7Y8GCBZBIJCZdB92jEDc3NyQnJ0OlUqG4uBgRERGYNm0atm7darCNt7c3vLy8kJSUhFWrVhnMZqzRaBAfH4/09HSjYujWrRuA+1MOtCeccDDGnkgqlQp3797FoEGD9JYPHjwYdnZ2jxy6vTUMGjQIcrlc79FAe1dSUgIiavTuxoOWL1+OwMBApKWl4ejRowbrW3p9dHcQdHc4Lly4ALVajd69ewtlZDIZOnfubNJ1sLe3BwD06tULISEhcHV1hVKpxJIlS6BUKhtMhK5evYqSkhJs3boVn3/+OQYMGKDXDmf+/Pl4++23hfYdTdGdY91dnvaCEw7G2BPpzp07AAAHBweDdc7OzqisrDRr/fb29igtLTVrHW2puroawB9fyE2RSqXIyMiASCTCW2+9BY1Go7e+ta+P7tHNggUL9MatuHLlCtRqtdH78fDwAAChXY2OnZ0dfH19kZ+fb7CNRCKBm5sbwsLCkJmZCZVKhZSUFADA0aNHkZubi4kTJxodg24QR905by844WCMPZGcnZ0BoMEvrjt37sDLy8tsdWu1WrPX0dZ0X4KmjJAbHByMhIQE5OXlGQwW19rXR9cQc+3ataD7Q0IIf8eOHTN6Pw4ODujWrRt++eUXg3W1tbVQKpWNbh8QEABbW1uoVCoA93sdHTx4EDY2NkISpIs1OTkZIpHIoI1JTU0NALS70fCg8LIAACAASURBVKM54WCMPZF69+4NBwcHgw/zEydOoKamBk899ZSwTCwWNzrDrqkOHToEIsKQIUPMVkdb69SpE0QiEcrLy03abtmyZQgKCsKZM2f0lptyfYzh7e0NqVRqMAVEc0RFReHMmTO4dOmSsEytVuPKlStCV9lbt2412AA0Ly8PdXV18Pb2BgBkZGQYJEC6O1+JiYkgIoPHSrpz7O7u3uJjaUuccDDGnkhSqRQzZ87Enj17sHnzZlRUVCA3NxdTpkyBh4cHYmNjhbIBAQG4ffs2srKyoNVqUVpaiitXrhjs09XVFUVFRSgoKEBlZaWQQNTX16OsrAy1tbU4f/484uPj4ePjg5iYmFapIycnx+LdYuVyOfz8/HDt2jWTttM9WrG1tTVYbuz1MbaeCRMmYNu2bUhPT0dFRQXq6upw7do13LhxAwAQHR0Nd3f3JodWT0hIgK+vL2JiYlBYWIhbt25hzpw50Gg0QiNYhUKB/fv349tvv0VFRQW0Wi3OnDmDN998EwqFAgkJCSbF/yDdOW5qHBBrwwkHY+yJtXjxYqSkpGDp0qXo2LEjhg4dii5duuDQoUNQKBRCuXfeeQfDhg3Dq6++isDAQCxbtky4nR0cHCx0b50yZQo6deqEnj17YuTIkbh9+zaA+8/a+/TpA5lMhtDQUHTv3h3fffedXnuHltZhDcLDw6FSqfTaY/zrX/9CQEAA8vPzMXjwYLz77rsG2w0ZMqTBL2Bjrk96erowS3bfvn1x6dIlfPLJJ5g5cyYAYMSIEcjLywNwv1vxjBkzsHLlSnTo0AEeHh6Ij49HWVkZgPuPKkpKSpCdnd3ocbq4uOD777+Hl5cX+vfvD09PT/z3v//F3r17hfE5pFIpnn32WUycOBGenp5wdHTEuHHj0KVLFxw/flyv8aqpTp48CU9PT/Tt27fZ+7CIthjOlIc2Z4y1BVjhXCq6eUesVWsObZ6Xl0disZi++OKL1gqvTdXV1VFoaCht2rTJ0qE80s2bN0kqldIHH3xgsI6HNmeMsSecKQ0p2wuNRoNvvvkGeXl5QiPGgIAALF26FEuXLm1w6G5rVldXh6ysLFRWVlr17NVJSUno378/4uLiANwfjbWoqAhHjx4VBoKzVpxwMMYYM9nt27cxYsQIdO/eHW+99ZawfN68eRg3bhyio6NNbkBqSYcOHcLu3buRk5Nj9FgibW3NmjU4e/Ys9u3bB4lEAgDIzs6Gp6cnQkNDsXfvXgtH2DhOOBhjzEzmz5+PjIwMlJeXo2vXrti1a5elQ2oVH330kV6vis2bN+utT05ORlxcHN5//30LRWi64cOHY8uWLXrz2ViT7Oxs3Lt3D4cOHYKLi4uwPCIiQu9aPDw+iDURWzoAxhh7XKWkpAgDPD1pwsLCEBYWZukwHhtjxozBmDFjLB1Gi/AdDsYYY4yZHSccjDHGGDM7TjgYY4wxZnaccDDGGGPM7Nqs0ei1a9ewY8eOtqqOMfaEMmUiLvbHMNn8+fxkunbtWptNIigiIjJ3JePGjXtsuoMxxhhjj5OXX34ZO3fuNHs9bZJwMMYeH6+88goA/kXMGDMNt+FgjDHGmNlxwsEYY4wxs+OEgzHGGGNmxwkHY4wxxsyOEw7GGGOMmR0nHIwxxhgzO044GGOMMWZ2nHAwxhhjzOw44WCMMcaY2XHCwRhjjDGz44SDMcYYY2bHCQdjjDHGzI4TDsYYY4yZHSccjDHGGDM7TjgYY4wxZnaccDDGGGPM7DjhYIwxxpjZccLBGGOMMbPjhIMxxhhjZscJB2OMMcbMjhMOxhhjjJkdJxyMMcYYMztOOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzHGGGNmxwkHY4wxxsyOEw7GGGOMmR0nHIwxxhgzO044GGOMMWZ2nHAwxhhjzOw44WCMMcaY2YmIiCwdBGPMOm3ZsgWbNm1CfX29sOzy5csAgK5duwrLbGxs8P/+3//Da6+91uYxMsbaB044GGOPdP78efTr18+osufOnUPfvn3NHBFjrL3ihIMx1qigoCBcuHCh0TIBAQHIy8tro4gYY+0Rt+FgjDXqjTfegEQieeR6iUSCCRMmtGFEjLH2iO9wMMYadenSJQQEBKCxj4q8vDwEBAS0YVSMsfaG73Awxhrl5+eHgQMHQiQSGawTiUQYNGgQJxuMsSZxwsEYa9Lf//532NraGiy3tbXF3//+dwtExBhrb/iRCmOsSSUlJfDw8NDrHgvc7w5bVFQEd3d3C0XGGGsv+A4HY6xJnTp1wtChQ/Xuctja2uL555/nZIMxZhROOBhjRnnjjTcMGo6+8cYbFoqGMdbe8CMVxphRKioq4ObmhpqaGgD3u8OWlJTA2dnZwpExxtoDvsPBGDOKk5MTRowYAbFYDLFYjJEjR3KywRgzGiccjDGjvf7666irq0NdXR3Pm8IYMwk/UmGMGa26uhodO3YEEeHmzZuQyWSWDokx1k5YbcLR0CBDjDHGGGuclX6tQ2zpABoTHx+P4OBgS4fBGHvA2bNnIRKJGpxF9tixY0hNTcX27dstEFn7FRUVxZ93rMV07z9rZdV3OLZv345XXnnF0qEwxh5QW1sLABCLDX+v7NixA1FRUVb7C8ta8ecdaw3W/v6z6jscjDHr01CiwRhjTeFeKowxxhgzO044GGOMMWZ2nHAwxhhjzOw44WCMMcaY2XHCwRizOvv27YNSqcS///1vS4fy2Dtw4ADmzZuH3bt3w8/PDyKRCCKRqMGJ+cLCwuDo6AhbW1v06tULp0+ftkDEptFqtUhJSUFAQADs7Ozg7OyM3r17o6CgQCizcuVKBAUFQSaTQaFQICgoCAsXLkRFRUWj+66urkZQUBAWLFggLPvyyy+xcuVK1NXVmeuQ2i1OOBhjVsdau/U9bhYvXox169Zh/vz5iIyMxKVLl+Dv748OHTpg8+bN2Lt3r175/fv3Y+fOnRg1ahRUKhUGDhxoociNFxUVhX/+85/YsmUL1Go1fv31V/j7++Pu3btCme+//x6TJk1CYWEhiouLsWzZMqxcuRIvv/xyo/tOTEzEhQsX9JaNHj0aUqkUw4cPx507d8xyTO0VJxyMMasTHh6O8vJyjBo1ytKhQKPRICQkxNJhtLoVK1YgMzMTO3bsgKOjo966devWwcbGBrGxsSgvL7dQhC2XmZmJrKws7Ny5E8888wzEYjE8PDyQnZ2N3r17C+Xs7OwwdepUuLm5wcHBAePGjUNERAT+85//4MaNGw3u+8cff8TPP//c4Lrp06ejX79+GDlypDBuDeOEgzHGGrVp0yaUlJRYOoxWdfHiRSxcuBBLliyBVCo1WB8SEoL4+Hhcv34ds2bNskCErePDDz/EwIED0adPn0bL7dmzx+A8eHp6AoDenRAdjUaD2bNnNzqqZ1JSEs6ePWvVI3+2NU44GGNW5ejRo/Dx8YFIJMKGDRsAAOnp6VAoFJDL5cjOzsaLL74IJycneHl5Ydu2bcK269atg1QqRadOnTB58mR4eHhAKpUiJCQEJ06cEMrFxcXBzs4OnTt3FpZNnToVCoUCIpEIN2/eBHB/eoWZM2ciPz8fIpEIAQEBAICvv/4aTk5OSE5ObotT0urWrVsHIsLo0aMfWWb58uXo3r07Nm7ciAMHDjS6PyLCmjVr0KNHD9jb28PFxQURERH43//+J5Qx9hoCQF1dHRYtWgQfHx/IZDL07dvX5OHya2pqcPz4cfTv39+k7XTy8vLg7OwMX19fg3WJiYnCHZFHcXFxwdChQ5GamsqPCP9/nHAwxqzKc889hx9//FFv2TvvvIMZM2ZAo9HA0dER27dvR35+Pvz8/DBp0iRotVoA9xOJmJgYqNVqTJ8+HQUFBTh9+jRqa2vxwgsv4OrVqwDuf+E+PIx4WloalixZorcsNTUVo0aNgr+/P4gIFy9eBAChQWB9fb1ZzoG57d27F4GBgZDL5Y8sI5PJ8Nlnn8HGxgaTJk1CVVXVI8smJSVh3rx5SExMRElJCY4cOYKrV68iNDQUxcXFAIy/hgAwd+5crFq1CmvXrsWNGzcwatQojB8/Hj/99JPRx1hUVISamhqcOnUKw4YNE5LPHj16IC0trcEkQKvV4vr169iwYQMOHDiA9evXw87OTq/MDz/8gPz8fIwfP77JGAYMGIDr16/j3LlzRsf9OOOEgzHWroSEhMDJyQlubm6Ijo5GVVUVCgsL9cqIxWLh13bPnj2Rnp6OyspKZGRktEoM4eHhqKiowMKFC1tlf22pqqoKly9fhr+/f5Nlg4ODMWPGDBQUFGDu3LkNltFoNFizZg1eeuklvP7661AqlejTpw8++ugj3Lx5Ex9//LHBNo1dw+rqaqSnp2Ps2LGIjIyEs7MzFixYAIlEYtL10z0KcXNzQ3JyMlQqFYqLixEREYFp06Zh69atBtt4e3vDy8sLSUlJWLVqFaKiogyONT4+Hunp6UbF0K1bNwBAbm6u0XE/zjjhYIy1W7pfnw/+Om7IoEGDIJfL9W7xP6lKSkpARI3e3XjQ8uXLERgYiLS0NBw9etRgvUqlwt27dzFo0CC95YMHD4adnZ3eo6yGPHwNL1y4ALVardeoUyaToXPnziZdP3t7ewBAr169EBISAldXVyiVSixZsgRKpbLBROjq1asoKSnB1q1b8fnnn2PAgAF67Xfmz5+Pt99+W2jf0RTdOdbd5XnSccLBGHsi2Nvbo7S01NJhWFx1dTWAP76QmyKVSpGRkQGRSIS33noLGo1Gb72u66eDg4PBts7OzqisrDQpPt2jmwULFghjgohEIly5cgVqtdro/Xh4eACA0B5Hx87ODr6+vsjPzzfYRiKRwM3NDWFhYcjMzIRKpUJKSgqA+22LcnNzMXHiRKNjkMlkAP445086TjgYY489rVaLO3fuwMvLy9KhWJzuS9CUgamCg4ORkJCAvLw8LFu2TG+ds7MzADSYWDTnnOsaYq5duxZEpPd37Ngxo/fj4OCAbt264ZdffjFYV1tbC6VS2ej2AQEBsLW1hUqlAnC/t9LBgwdhY2MjJEG6WJOTkyESiQzamNTU1AD445w/6TjhYIw99g4dOgQiwpAhQ4RlYrG4yUcxj6NOnTpBJBKZPL7GsmXLEBQUhDNnzugt7927NxwcHAy+bE+cOIGamho89dRTJtXj7e0NqVSKs2fPmrRdQ6KionDmzBlcunRJWKZWq3HlyhWhq+ytW7cabACal5eHuro6eHt7AwAyMjIMEiDdHbPExEQQkcFjJd05dnd3b/GxPA444WCMPXbq6+tRVlaG2tpanD9/HvHx8fDx8UFMTIxQJiAgALdv30ZWVha0Wi1KS0tx5coVg325urqiqKgIBQUFqKyshFarRU5OTrvtFiuXy+Hn54dr166ZtJ3u0Yqtra3B8pkzZ2LPnj3YvHkzKioqkJubiylTpsDDwwOxsbEm1zNhwgRs27YN6enpqKioQF1dHa5duyYMwhUdHQ13d/cmh1ZPSEiAr68vYmJiUFhYiFu3bmHOnDnQaDRCI1iFQoH9+/fj22+/RUVFBbRaLc6cOYM333wTCoUCCQkJJsX/IN05bmockCcGWSkAtH37dkuHwRgzwfbt26mlHyvr16+nzp07EwCSy+U0evRoSktLI7lcTgCoW7dulJ+fTx9//DE5OTkRAPL19aXffvuNiIhiY2NJIpGQp6cnicVicnJyooiICMrPz9er59atWzRs2DCSSqXUtWtXevfdd2n27NkEgAICAqiwsJCIiE6fPk2+vr4kk8noueeeo99//5327dtHjo6OtHz58hYdq05bf97FxcWRRCIhtVotLNuzZw/5+/sTAOrYsSNNmzatwW1nz55NY8aM0VtWX19Pq1evpm7dupFEIiEXFxcaO3YsXbhwQShjyjW8d+8ezZkzh3x8fEgsFpObmxtFRkaSSqUiIqKxY8cSAFq0aFGTx3r16lV69dVXycXFhezt7enpp5+mnJwcvTKjR4+mrl27koODA9nb25O/vz9FR0dTbm5uo/suLS0lAJSYmNjg+vDwcPL09KT6+vom42wNrfH+MyerjYwTDsbaH2v4wIuNjSVXV1eLxmCqtv68y8vLI7FYTF988UWb1dma6urqKDQ0lDZt2mTpUB7p5s2bJJVK6YMPPmizOq3h/dcYfqTCGHvs8EydjQsICMDSpUuxdOnSBofutmZ1dXXIyspCZWUloqOjLR3OIyUlJaF///6Ii4uzdChWgxOOJgwePBi2trbNHh63MRMnToSjoyNEIlGjDaQeVc4ap/Cur6/H2rVrG53s6ujRo3j22Wchl8vh4eGBOXPm4N69eybX9fB02g39denSpQVH8wd+HbDHzbx58zBu3DhER0e3qwnaDh06hN27dyMnJ8fosUTa2po1a3D27Fns27cPEonE0uFYDU44mnDy5EkMGzbMLPveuHEjPvnkk2aXIysbnz8vLw9//vOfkZCQ8Mj+8iqVCmFhYRg+fDhKS0uxZ88efPrpp5gyZYrJ9T04nbZSqRRajtfW1kKtVqO4uLjVPpD4ddA+zJ8/HxkZGSgvL0fXrl2xa9cuS4dk1ZKTkxEXF4f333/f0qEYbfjw4diyZYvePDjWJDs7G/fu3cOhQ4fg4uJi6XCsitjSAbQXIpHI0iEY0E3hbQ3OnTuHpUuXYsqUKaiqqnrkl+CyZcvQuXNnLFmyBCKRCMHBwZgzZw7ee+89vPfeewgKCmpxLLa2tpDJZJDJZOjevXuL9/cgfh1Yt5SUFGGgJmacsLAwhIWFWTqMx8aYMWMwZswYS4dhlfgOh5HMdVvM2C+wtviiIyLs3LmzwSF/m9KvXz/s3r0br7322iNHMKytrcXevXsxdOhQveN58cUXQUTIzs5uduyPkpWV1ar749cBY4w1z2OTcDQ2nXFqaioUCgVsbGzw1FNPwd3dHRKJBAqFAgMHDkRoaKgw2IyzszPee+89g/1fvHgRQUFBUCgUkMlkCA0NNZhXoKkplYkIq1evRmBgIOzt7aFUKjF79myDuowp15IpvHWxpqSkIDAwEDKZDB07dkTXrl2RkpJiMItma7l06RLu3r0LHx8fveW6SaTOnz8vLDPH9N/8OrCO1wFj7Allmc4xTYOJ3cRmzZpF9vb2tGvXLiorK6P58+eTjY0NnTx5koiIFi9eTADoxIkTVFVVRTdv3qQRI0YQANq7dy+VlpZSVVUVxcXFEQA6e/assO/hw4eTn58fXb58mbRaLf3888/0zDPPkFQqFfqNGxNDYmIiiUQi+sc//kFlZWWkVqspLS2NANCZM2eE/Rhb7urVqwSA1q9fr7ctADp48CCVl5dTSUkJhYaGkkKhoJqaGqFccnIy2draUnZ2NqnVajp16hS5u7vT888/b8JVatgzzzxD/fr1M1h++PBhAkCrV682WCeTyWj48OHC/1999RU5OjrS0qVLm6zP39+flEql3rLp06c32IeeXwfmfR1Ye7c8a2Xq5x1jDbH295/VRmbKG1Cj0ZBcLqfo6GhhmVqtJnt7e3rnnXeI6I8vmsrKSqHM559/TgD0vpj++9//EgDKzMwUlg0fPtzgC/T8+fMEgGbNmmVUDGq1muRyOb3wwgt6+9m2bZveF4ix5Yga/6LRaDTCMt2X1MWLF4VlgwcPpqefflqvjrfffptsbGzo3r171BKPSjj2799PAGjNmjUG65ycnCgkJKRZ9ekGK3r4r7GEg18H97X268DaP/CsFSccrDVY+/vvsWg02tzpjHXTItfW1grLdM/om5pjoU+fPlAqlcJjgKZiuHjxItRqNYYPH97ofo0tZ4qGpvCurq6GVCrVK1dXVweJRGIwdHFr0dX34PnWqampadEER0qlUpi1EgDi4+ON3pZfB63/OtixY0ezt31SmTIxGWMNsfbX0GORcDw4nfGCBQv01ummKDYHiUQifHg3FYNuTH3d7IKPYmy5lho5ciRWr16N7OxshIWFQaVSISsrC3/729/MlnDourFVVFToLVer1aiurm7Va5Wamtpq+2oKvw4MRUVFtWKUT4bU1NQ2fd0y1tYei0ajrTWdsSlqa2tx+/ZtoQFkUzHofkU2NcCVseVaKikpCX/5y18QExMDJycnvPTSS3jllVeMGg+iubp27QpHR0eDCbIuXrwIAOjbt6/Z6jYXfh007OFj57/G/wBg+/btFo+D/9r334ON063RY5FwtOZ0xsb67rvvUF9fj4EDBxoVQ+/evWFjY4PDhw83ul9jy7WUSqVCfn4+SktLodVqUVhYiPT0dLMOVCMWizFy5EgcOXIE9fX1wvKcnByIRCKMHj261eu8ceMGJkyY0Or71eHXAWOMGeexSDiMmc64pWpqalBeXo7a2lqcPn0acXFxwrTHxsTg5uaGyMhI7Nq1C5s2bUJFRQXOnz9vMNaBseVaatq0afDx8WnzeRQWLlyI4uJiLF68GFVVVTh27BhWr16NmJgYBAYGCuVaOv03EUGj0WD37t1wcnJqrfD5dcAYY81FVgomttpubDrj1NRUYVrkLl260Pfff08rVqwgpVJJAMjd3Z22bNlCmZmZ5O7uTgDIxcWFtm3bRkREGRkZNGzYMOrUqROJxWLq0KEDvfrqq3TlyhWjYyAiqqyspIkTJ1KHDh3IwcGBnnvuOVq0aBEBIC8vLzp37pzR5Vo6hfe3335LHTp00OvVIZFIqEePHrR7926Tr9exY8fo2WefJQ8PD2F/nTt3ppCQEDp8+LBe2cOHD9PTTz9N9vb25OHhQbNnz6bq6mq9MsZM//3gdNqN/S1YsICIiF8HbfA6sPZW8tbK1M87xhpi7e8/ERFZ5UQMIpEI27dv58GHzCQ9PR15eXlYu3atsKympgZz585Feno6ysrKWtRrhLUPrf062LFjB6KiomClHytWiz/vWGuw9vffY9FLhZnm999/R1xcnEE7Azs7O/j4+ECr1UKr1XLC8Zjj1wFjrC09Fm04mGlkMhkkEgk2bdqE4uJiaLVaFBUVYePGjVi0aBGio6NRVFTU6LTvur/o6GhLHw5rJmNeB63Z/oUx9mTjhOMJpFQqsX//fvz888/o3r07ZDIZevbsiYyMDKxYsQKff/45goKCjOqGlZmZaenDYc1kzOuAWb8DBw5g3rx52L17N/z8/IQfA2+88YZB2bCwMDg6OsLW1ha9evXC6dOnLRCxabRaLVJSUhAQEAA7Ozs4Ozujd+/eKCgoEMqsXLkSQUFBkMlkUCgUCAoKwsKFCw3G/HlYdXU1goKCDMbMMZax9S5duhQ9e/aEk5MT7O3tERAQgPfee0+vsfaXX36JlStXoq6urlmxtAsWaTliBHAjKsbaHWtvtGatmvt5t2jRIho1ahRVVFQIy/z9/YWGwF999ZXBNjk5OTRmzJgWxduWxo4dS4GBgXT8+HHSarVUVFREo0eP1puKIDw8nD744AMqKSmhyspK2rFjB0kkEoOpAR6WkJBAACgxMbFZsRlb79ChQyktLY1u3bpFFRUVtH37dpJIJDRixAi9cqmpqTR06FAqKytrVjzW/v7jOxyMsceKRqNBSEhIu6+jKStWrEBmZiZ27NgBR0dHvXXr1q2DjY0NYmNjUV5ebqEIWy4zMxNZWVnYuXMnnnnmGYjFYnh4eCA7O1tv+gA7OztMnToVbm5ucHBwwLhx4xAREYH//Oc/jxwa4ccff8TPP//coviMrdfBwQGxsbFwdXWFo6MjXnnlFYwdOxZff/01rl69KpSbPn06+vXrh5EjRzY4BUR7xwkHY+yxsmnTJpSUlLT7Ohpz8eJFLFy4EEuWLDGYCwcAQkJCEB8fj+vXr2PWrFkWiLB1fPjhhxg4cCD69OnTaLk9e/YYnAdPT08AaHCMGY1Gg9mzZ7d4KHlj6/3qq68Mpgro2LEjgPtTOzwoKSkJZ8+efSyHueeEgzFmUUSENWvWoEePHrC3t4eLiwsiIiL0Jl6Mi4uDnZ2dMB8PAEydOhUKhQIikQg3b94EcH/SvpkzZyI/Px8ikQgBAQFYt24dpFIpOnXqhMmTJ8PDwwNSqRQhISE4ceJEq9QBAF9//XWLBqszxbp160BEjY7Ou3z5cnTv3h0bN27EgQMHGt2fMdcgPT0dCoUCcrkc2dnZePHFF+Hk5AQvLy9s27ZNb391dXVYtGgRfHx8IJPJ0LdvX5OH3a6pqcHx48fRv39/k7bTycvLg7OzM3x9fQ3WJSYmCncmWltj9T7o+vXrkMlk6Nq1q95yFxcXDB06FKmpqVbbvbXZLPtE59HAbTgYa3ea8wx50aJFZGdnR1988QXduXOHzp8/TwMHDqSOHTvS77//LpR77bXXyN3dXW/b1atXEwAqLS0VlkVGRpK/v79eudjYWFIoFPTLL79QdXU1qVQqGjx4MDk6OlJhYWGr1PHVV1+Ro6MjLV261KTjJzL9887Pz4969uzZ4Dp/f3+6fPkyERH9+OOPZGNjQ126dKG7d+8SUcNtOIy9BomJiQSADh48SOXl5VRSUkKhoaGkUCiopqZGKDdr1iyyt7enXbt2UVlZGc2fP59sbGzo5MmTRh/j5cuXCQD179+fnn/+eercuTPZ29tTUFAQbdiwgerr6w22qampoWvXrtH69evJ3t6evvjiC4MyR48epdGjRxMRUWlpaYvacJhS74OqqqrI0dGR4uLiGlw/b948AkBnzpwxKQ5uw8EYY4+g0WiwZs0avPTSS3j99dehVCrRp08ffPTRR7h582arDuUuFouFX/A9e/ZEeno6KisrkZGR0Sr7Dw8PR0VFBRYuXNgq+3uUqqoqXL58Gf7+/k2WDQ4OxowZM1BQUIC5c+c2WKY51yAkJAROTk5wc3NDdHQ0qqqqUFhYCOB+z4/09HSMHTsWkZGRcHZ2xoIFCyCRSEw617pHEm5ubkhOToZKpUJxcTEiIiIwbdo0bN261WAbb29veHl5ISkpCatWv7t+jgAAIABJREFUrTKYtVij0SA+Ph7p6elGx2GMpup9WEpKCjw8PLB8+fIG13fr1g0AkJub26pxWhonHIwxi1GpVLh79y4GDRqkt3zw4MGws7PTe+TR2gYNGgS5XK732KA9KCkpARFBLpcbVX758uUIDAxEWloajh49arC+pdfAzs4OwP3uqwBw4cIFqNVqvUadMpkMnTt3Nulc29vbAwB69eqFkJAQuLq6QqlUYsmSJVAqlQ0mQlevXkVJSQm2bt2Kzz//HAMGDNBrazN//ny8/fbbQjuL1tJUvQ/as2cPduzYgW+++cagsa+O7toWFxe3apyWxgkHY8xi7ty5A+B+K/6HOTs7o7Ky0qz129vbo7S01Kx1tLbq6moAf3whN0UqlSIjIwMikQhvvfUWNBqN3vrWvgZVVVUAgAULFugNEnjlyhWDBpKN8fDwAACh7YyOnZ0dfH19kZ+fb7CNRCKBm5sbwsLCkJmZCZVKhZSUFADA0aNHkZubi4kTJ5p0PMZorN4HZWZmYsWKFTh06BC6dOnyyP3pRvfVXevHBSccjDGLcXZ2BoAGv9Tu3LkDLy8vs9Wt1WrNXoc56L6MTBkgKjg4GAkJCcjLy8OyZcv01rX2NdA1xFy7dq3BQIHHjh0zej8ODg7o1q0bfvnlF4N1tbW1UCqVjW4fEBAAW1tbqFQqAPd7Fh08eBA2NjZCEqSLNTk5GSKRCD/99JPR8Rlbr8769euxefNmfPvtt/jTn/7U6D5qamoA4LGbVoATDsaYxfTu3RsODg4GH/QnTpxATU0NnnrqKWGZWCwWbtu3hkOHDoGIMGTIELPVYQ6dOnWCSCQyeXyNZcuWISgoCGfOnNFbbso1MIa3tzekUqnBHD3NERUVhTNnzuDSpUvCMrVajStXrghdZW/duoXx48cbbJuXl4e6ujp4e3sDADIyMgwSIN3drcTERBCRwWOlxhhbLxFhzpw5yM3NRVZWVoN3kh6mu7bu7u5Gx9MecMLBGLMYqVSKmTNnYs+ePdi8eTMqKiqQm5uLKVOmwMPDA7GxsULZgIAA3L59G1lZWdBqtSgtLcWVK1cM9unq6oqioiIUFBSgsrJSSCDq6+tRVlaG2tpanD9/HvHx8fDx8UFMTEyr1JGTk9Mm3WLlcjn8/Pxw7do1k7bTPVp5eDwIU66BsfVMmDAB27ZtQ3p6OioqKlBXV4dr164Jg2FFR0fD3d29yaHVExIS4Ovri5iYGBQWFuLWrVuYM2cONBqN0AhWoVBg//79+Pbbb1FRUQGtVoszZ87gzTffhEKhQEJCgknxGxufsfX+8ssvWLVqFT755BNIJBKD+ag++OADg33rrm1T44+0O5bpHNM0cLdYxtqd5nTLq6+vp9WrV1O3bt1IIpGQi4sLjR07li5cuKBX7tatWzRs2DCSSqXUtWtXevfdd2n27NkEgAICAoTuradPnyZfX1+SyWT03HPP0e+//06xsbEkkUjI09OTxGIxOTk5UUREBOXn57daHfv27SNHR0davny5yefN1M+7uLg4kkgkpFarhWV79uwhf39/AkAdO3akadOmNbjt7NmzDbrFGnMN0tLSSC6XEwDq1q0b5efn08cff0xOTk4EgHx9fem3334jIqJ79+7RnDlzyMfHh8RiMbm5uVFkZCSpVCoiuj9cOQBatGhRk8d69epVevXVV8nFxYXs7e3p6aefppycHL0yo0ePpq5du5KDgwPZ29uTv78/RUdH6w1/3pBHdYs1Nj5j6s3NzSUAj/xbvXq1wX7Dw8PJ09Ozwa6/jbH2brFWGxknHIy1P9b6gff/sXfncVHV+//AXweYBYZhE5QJBBFUwo1MTTC/WnZp8bqgoqR1s7LINMKVEFFEpBQvcl2om/r1dtVk/4KZmFeNylyqmyZiKeKCS8oSyC7b+/eHv5kcZ8AZYBjQ9/Px8A/P+Zzzec85s7w5ny0oKIjs7OyMHUaz9P2+y8vLIzMzswfO99BZNTY20qhRo2jbtm3GDkUrY8ZXXFxMUqmU1q1bp/exnfXzp8RNKoyxR8LDtAqnh4cHoqKiEBUVpXXq7s6ssbERGRkZqKioQGBgoLHD0WDs+CIjI+Ht7Y3g4OAOr9vQOOFgjLEuKCwsDAEBAQgMDOxSC7RlZ2cjLS0NWVlZOs8l0pGMGV9cXBxOnTqFffv2QSQSdWjdHYETDsbYQ23p0qXYvn07bt++DTc3N6Smpho7pHazevVqBAcH48MPPzR2KDobO3Ysdu3apbZmTWdirPgyMzNx584dZGdnw9bWtkPr7ihmxg6AMcYMKSYmRuskTA8LPz8/+Pn5GTsM1kYTJ07ExIkTjR2GQfETDsYYY4wZHCccjDHGGDM4TjgYY4wxZnCccDDGGGPM4AQiImMHoY0gCBgxYkSXW1iJsUfZtWvXcPz4cUydOtXYoXQpqamp/H3H2kz5+eukP+udN+EICAgwdgiMMS2Ui3898cQTRo6EMaZNSkqKsUPQqtMmHIyxzmnatGkAgOTkZCNHwhjrSrgPB2OMMcYMjhMOxhhjjBkcJxyMMcYYMzhOOBhjjDFmcJxwMMYYY8zgOOFgjDHGmMFxwsEYY4wxg+OEgzHGGGMGxwkHY4wxxgyOEw7GGGOMGRwnHIwxxhgzOE44GGOMMWZwnHAwxhhjzOA44WCMMcaYwXHCwRhjjDGD44SDMcYYYwbHCQdjjDHGDI4TDsYYY4wZHCccjDHGGDM4TjgYY4wxZnCccDDGGGPM4DjhYIwxxpjBccLBGGOMMYPjhIMxxhhjBscJB2OMMcYMjhMOxhhjjBkcJxyMMcYYMzhOOBhjjDFmcJxwMMYYY8zgOOFgjDHGmMFxwsEYY4wxg+OEgzHGGGMGZ2bsABhjnVd1dTXu3Lmjtq2urg4AUFpaqrZdIpHAwsKiw2JjjHUtAhGRsYNgjHVOCQkJmDt3rk5lN2/ejHfffdfAETHGuipOOBhjzSoqKoJCoUBjY2OL5UxNTfH777/DwcGhgyJjjHU13IeDMdYsBwcHjB07Fqamps2WMTU1xXPPPcfJBmOsRZxwMMZa9Morr6ClB6FEhFdeeaUDI2KMdUXcpMIYa1FFRQUcHBw0Oo8qicViFBUVwcrKqoMjY4x1JfyEgzHWIrlcjvHjx0MkEmnsMzMzw8SJEznZYIw9ECccjLEHmjlzJhoaGjS2NzY2YubMmUaIiDHW1XCTCmPsgerq6mBvb4+Kigq17ZaWliguLoZEIjFSZIyxroKfcDDGHkgsFiMgIABisVi1TSQSYfr06ZxsMMZ0wgkHY0wnM2bMUM0yCgD19fWYMWOGESNijHUl3KTCGNNJU1MTHB0dUVRUBACwt7fHzZs3W5yjgzHGlPgJB2NMJyYmJpgxYwbEYjFEIhFmzpzJyQZjTGeccDDGdPbyyy+jrq6Om1MYY3p75FeLTU5ONnYIjHUZRIRu3boBAC5duoTLly8bNyDGupBp06YZOwSjeuT7cAiCYOwQGGOMPQIe8Z9bfsIBAElJSY985skebsnJyZg+fXq7fOGdPXsWAODl5dXmc3V2giDw9wNrM+Xn71HHCQdjTC+PQqLBGGt/3GmUMcYYYwbHCQdjjDHGDI4TDsYYY4wZHCccjDHGGDM4TjgYY4wxZnCccDDGdLZv3z5YW1vjiy++MHYond7BgwcRFhaGtLQ09O7dG4IgQBAEvPrqqxpl/fz8IJfLYWpqiv79++Pnn382QsT6qa+vR0xMDDw8PCAWi2FjY4MBAwaoTQa3Zs0aeHp6wtzcHDKZDJ6enoiIiEB5eXmL566trYWnpyeWLVvWqth0rTcqKgpeXl6wsrKCRCKBh4cHlixZgsrKSlWZPXv2YM2aNWhsbGxVLOxPnHAwxnT2qE9cpKsVK1Zgw4YNWLp0KaZMmYKLFy/C3d0d3bp1w86dO/Hll1+qlT9w4ABSUlIwfvx45ObmYsiQIUaKXHfTp0/Hv//9b+zatQvV1dX49ddf4e7urvZj/d133+Gtt95CQUEBbt26hVWrVmHNmjWYOnVqi+cODw/HuXPnWh2brvUePnwY8+bNw+XLl1FcXIyYmBjEx8cjICBAVWbChAmQSqUYO3YsysrKWh0TA0CPOACUlJRk7DAYM6ikpCR62D7u1dXV5OPjY9A6WvP98OGHH1Lfvn2ppqZGbbu7uzvt2rWLTExMyMnJicrKytT2Z2Vl0cSJE9scc0fYvXs3CYJAp0+fbrGcv7+/xnUICAggAHTjxg2tx3z//ffk5+dHACg8PLxV8ela77hx46ihoUGt3LRp0wgAFRQUqG0PDg4mHx8fqq+v1zueh/Hz1xr8hIMx1iVt27YNhYWFxg5DzYULFxAREYGVK1dCKpVq7Pf19UVISAiuX7+ORYsWGSHC9vHxxx9jyJAhGDhwYIvl0tPTNa6Dk5MTAKg9CVGqqanB4sWLER8f36b4dK137969Gise29vbAwCqq6vVtkdGRuLUqVNtju1RxgkHY0wnR44cgYuLCwRBwKZNmwAACQkJkMlksLCwQGZmJl588UVYWVnB2dkZu3fvVh27YcMGSKVSdO/eHe+88w4UCgWkUil8fX1x4sQJVbng4GCIxWI4Ojqqts2dOxcymQyCIKC4uBgAEBISgoULFyI/Px+CIMDDwwMAsH//flhZWWH16tUdcUk0bNiwAUSECRMmNFsmOjoaffv2xdatW3Hw4MEWz0dEiIuLw+OPPw6JRAJbW1tMmjQJv/32m6qMrvcAABobG7F8+XK4uLjA3NwcgwYNQlJSkl6vsa6uDsePH4e3t7dexynl5eXBxsYGrq6uGvvCw8Mxd+5cODg4tOrcra33XtevX4e5uTnc3NzUttva2mL06NGIj4/npsXWMvITFqMDN6mwR0B7PdK9evUqAaCNGzeqtoWHhxMAOnToEN2+fZsKCwtp1KhRJJPJqK6uTlUuKCiIZDIZnT17lmprayk3N5eGDRtGcrlc7fH1zJkzqUePHmr1xsbGEgAqKipSbZsyZQq5u7urldu7dy/J5XKKiopq82sl0v/7oXfv3uTl5aV1n7u7O126dImIiI4ePUomJibUq1cvqqysJCLtTSrLly8nsVhMO3bsoLKyMjp9+jQNGTKE7O3t6ebNm6pyut6DRYsWkUQiodTUVCotLaWlS5eSiYkJ/fjjjzq/xkuXLhEA8vb2pjFjxpCjoyNJJBLy9PSkTZs2UVNTk8YxdXV1dO3aNdq4cSNJJBLasWOHRpkjR47QhAkTiIioqKioTU0q+tR7r6qqKpLL5RQcHKx1f1hYGAGgkydP6hUHN6ncxU84GGPtwtfXF1ZWVnBwcEBgYCCqqqpQUFCgVsbMzEz117qXlxcSEhJQUVGB7du3t0sM48aNQ3l5OSIiItrlfPqoqqrCpUuX4O7u/sCyPj4+mD9/Pi5fvowPPvhAa5mamhrExcVh8uTJeOWVV2BtbY2BAwfik08+QXFxMT799FONY1q6B7W1tUhISIC/vz+mTJkCGxsbLFu2DCKRSK/rr2yScHBwwOrVq5Gbm4tbt25h0qRJmDdvHj7//HONY3r27AlnZ2dERkZi7dq1GguZ1dTUICQkBAkJCTrHoYsH1Xu/mJgYKBQKREdHa93fp08fAEBOTk67xvmo4ISDMdbuxGIxgLtDJ1sydOhQWFhYqDURdFWFhYUgIlhYWOhUPjo6Gv369cPmzZtx5MgRjf25ubmorKzE0KFD1bYPGzYMYrFYrSlKm/vvwblz51BdXY0BAwaoypibm8PR0VGv6y+RSAAA/fv3h6+vL+zs7GBtbY2VK1fC2tpaayJ09epVFBYW4vPPP8dnn32GJ554Qq3/zdKlS/H222+r+lm0lwfVe6/09HQkJyfjq6++glwu11pGeW9v3brVrnE+KjjhYIwZlUQiQVFRkbHDaLPa2loAf/4gP4hUKsX27dshCALeeOMN1NTUqO1XDsG0tLTUONbGxgYVFRV6xVdVVQUAWLZsmWpOEEEQcOXKFY0Oki1RKBQAoOpPoyQWi+Hq6or8/HyNY0QiERwcHODn54fExETk5uYiJiYGwN2+QTk5OZg9e7Zer0cXLdV7r8TERHz00UfIzs5Gr169mj2fubk5gD/vNdMPJxyMMaOpr69HWVkZnJ2djR1Kmyl/jPSZIMrHxwcLFixAXl4eVq1apbbPxsYGALQmFq25ZsqOmOvXrwcRqf07duyYzuextLREnz59cPbsWY19DQ0NsLa2bvF4Dw8PmJqaIjc3F8Dd0UaHDh2CiYmJKglSxrp69WoIgoCffvpJ5/h0rVdp48aN2LlzJw4fPozHHnusxXPU1dUB+PNeM/1wwsEYM5rs7GwQEUaMGKHaZmZm9sCmmM6oe/fuEAQBt2/f1uu4VatWwdPTEydPnlTbPmDAAFhaWmr82J44cQJ1dXV48skn9aqnZ8+ekEqlOHXqlF7HaTN9+nScPHkSFy9eVG2rrq7GlStXVENlS0pKMGPGDI1j8/Ly0NjYiJ49ewIAtm/frpEAKZ94hYeHg4g0mpVaomu9RITQ0FDk5OQgIyND65Ok+ynvbY8ePXSOh/2JEw7GWIdpampCaWkpGhoacPr0aYSEhMDFxQWzZs1SlfHw8MAff/yBjIwM1NfXo6ioCFeuXNE4l52dHW7cuIHLly+joqIC9fX1yMrKMtqwWAsLC/Tu3RvXrl3T6zhl08r980FIpVIsXLgQ6enp2LlzJ8rLy5GTk4M5c+ZAoVAgKChI73pef/117N69GwkJCSgvL0djYyOuXbuG33//HQAQGBiIHj16PHBq9QULFsDV1RWzZs1CQUEBSkpKEBoaipqaGlUnWJlMhgMHDuDw4cMoLy9HfX09Tp48iddeew0ymQwLFizQK35d49O13rNnz2Lt2rXYsmULRCKRWjOTIAhYt26dxrmV9/ZB84+wZhhncEznAR4Wyx4B7TEsb+PGjeTo6EgAyMLCgiZMmECbN28mCwsLAkB9+vSh/Px8+vTTT8nKyooAkKurK50/f56I7g6LFYlE5OTkRGZmZmRlZUWTJk2i/Px8tXpKSkromWeeIalUSm5ubvTee+/R4sWLCQB5eHiohtD+/PPP5OrqSubm5vT000/TzZs3ad++fSSXyyk6OrpNr1VJ3++H4OBgEolEVF1drdqWnp5O7u7uBIDs7e1p3rx5Wo9dvHixxrDYpqYmio2NpT59+pBIJCJbW1vy9/enc+fOqcrocw/u3LlDoaGh5OLiQmZmZuTg4EBTpkyh3NxcIro7QycAWr58+QNf69WrV+nll18mW1tbkkgkNHz4cMrKylIrM2HCBHJzcyNLS0uSSCTk7u5OgYGBlJOT0+K5mxsWq2t8utSbk5NDAJr9Fxsbq3HecePGkZOTk9ahvy3hYbF3PfJXgBMO9ijoDF94QUFBZGdnZ9QY9KXv90NeXh6ZmZk9cL6HzqqxsZFGjRpF27ZtM3YoWhkzvuLiYpJKpbRu3Tq9j+0Mn7/OgJtUGGMd5mFfcdPDwwNRUVGIiorSOnV3Z9bY2IiMjAxUVFQgMDDQ2OFoMHZ8kZGR8Pb2RnBwcIfX/bDghOMh09TUhPXr18PX11en8rNnz4ZcLocgCG3qTKZLvUeOHMHIkSNhYWEBhUKB0NBQ3Llzp9V1AnfnFnjvvffQv39/yOVymJmZwdraGn379sW4ceP06n1vaC1do/uXMFf+E4vF6N69O8aMGYPY2FiUlpYaIXKmj7CwMAQEBCAwMFDvDqTGlJ2djbS0NGRlZek8l0hHMmZ8cXFxOHXqFPbt2weRSNShdT9UjP2IxdjwEDWpnD9/nkaOHEkAaPDgwToft3v37lZN16tPvWfOnCFzc3OKiIigyspKOnr0KNnb29Prr7/eqjqJiLZu3UoikYj+53/+h/bv30+lpaVUW1tL+fn5lJiYSL6+vvTPf/6z1edvT7reG3d3d7K2tiaiu+33paWl9PXXX9OsWbNIEARSKBR6TUOtZOxHumFhYSQWiwkA9erVi1JSUowWiz7a8v3w1VdfUWhoaDtHxDpaRkYGxcTEaKwqqw9jf/46i0f+CjwsCcepU6do8uTJtHPnTvL29u6whEPXeqdPn05ubm5qna1iY2NJEAT69ddf9a732LFjZGpqSs8++2yzy0Xv379fbc0PY9Hn3tybcNwvJSWFTExMqHv37hpLmz8If+G1zsPy/cCMiz9/d3GTykNi8ODBSEtLw8yZM3We6VBJEASD1tvQ0IAvv/wSo0ePVqvrxRdfBBEhMzNT73qjo6PR2NiIDz/8EGZmZlrLPP/885g3b57e525vbbk395o6dSpmzZqFwsJCfPLJJ+0YIWOMGR4nHK2wY8cODB06FFKpFDKZDL169VLNEkjtuJz0448/DkEQYGJigieffFI1/fCSJUtgbW0NqVSKf/3rX3rFTkSIjY1Fv379IJFIYG1tjcWLF7f9orTg4sWLqKyshIuLi9p25SJXp0+fVm3TZXnxuro6HDp0CN26dcPw4cN1jqOz3xtdKOeryMrKavdzM8aYIXHCoaf4+Hj87W9/w9SpU3Hjxg1cu3YNS5cuxblz5wDc7ckcFhaG8PBwFBYW4ttvv8XVq1cxatQo1YI/7777LubPn4+amhrI5XIkJSUhPz8fvXv3xltvvaWaZfHMmTPo1asXevbsiR9++EHVUWrt2rV488038dFHH6lNmKSLiIgIhIaGIigoCLdu3cLNmzebXa2yvdy8eRMANBZEkkqlMDc3V1sISTmKoampqdnzXblyBbW1taqVG3XV2e+NLry9vQFAbYZHxhjrCjjh0EN9fT1WrlyJZ555Bh988AHs7Oxga2uLN998E8OGDWv35aRNTU3x/vvvo6CgAOnp6apjqqurkZaWhjfeeEOv+GtqarB+/Xo899xzWLBgAWxsbGBubg47O7u2XZgHUI5EuX8mReDu4kr3Llqly/Li5eXlALQvatWczn5vdKUcUaTvwl2MMWZs2hu/mVanT59GWVkZnn/+ebXtyh+fn376qV2XkwbuDluNjIxEfHw8AgICAAA7d+7EpEmTYGVlpVf8Fy5cQHV1NcaOHavXcW0llUoB3O3Lcb+6ujq9F0JSJhr6rHDZ3kt9A+17b3RVVVUFImr1+ZVxMt2tX78eKSkpxg6DdWH6Tnf/sOInHHpQ/mWtXMXxfu29nLTyXG+//TaOHj2KH374AQDw8ccft2ryGeWbXrkSY0dxdHQE8Of1U6qurkZtba1quWtd9erVC1KpFOfPn9f5mM5+b3SlfM2enp4Gq4MxxgyBn3DoQbl0cXFxsdb97b2ctFJwcDDi4+Oxfv16zJkzBz179lR1uNSH8klDWyfb0pebmxvkcrnGAlwXLlwAAAwaNEiv80kkEjz//PPIzMzE999/j5EjR2ot98cff2DJkiXYunVrp783utq/fz+AuyN8WoP/UtePIAiYP38+pk2bZuxQWBeWnJyM6dOnGzsMo+MnHHro1asX7OzscODAAa3723s5aSVnZ2dMmzYNqampiIiIQEhISKvOM2DAAJiYmOCbb75p1fGtZWZmhpdeegnffvutWmfQrKwsCIKACRMm6H3OyMhISCQSLFiwQK0PyL3OnDmjGjLb2e+NLm7evIn169fD2dnZYH1EGGPMUDjh0INEIsHSpUvx7bffIjg4GNevX0dTUxMqKipw9uzZdl9O+l4LFy5EQ0MDSktL8eyzz7bqHA4ODpgyZQpSU1Oxbds2lJeX4/Tp01o7TLa3iIgI3Lp1CytWrEBVVRWOHTuG2NhYzJo1C/369VOV03V5cW9vb+zatQtnzpzBqFGjsG/fPty+fRv19fW4dOkStmzZgjfffFM1DXFnvzf3IiJUVlaiqakJRISioiIkJSVh5MiRMDU1RUZGhsH6iDDGmMEYcdKxTgGtmElw06ZNNHDgQJJKpSSVSumJJ56gzZs3E1H7Lyd9r2eeeYa2bt2qNaZjx47RyJEjSaFQqJZXdnR0JF9fX/rmm29U5SoqKmj27NnUrVs3srS0pKeffpqWL19OAMjZ2Zl++eUXva6FrvUSEX3zzTc0fPhwkkgkpFAoaPHixVRbW6tWRt/lxQsKCmjRokU0cOBAsrS0JFNTU7KxsaEnnniC3nzzTfr+++9VZTvzvdmzZw8NGjSILCwsSCwWk4mJCQEgQRDIxsaGhg8fTlFRUVRSUqLTdbkfz3TYOq35fmDsfvz5u0sgIur4NKfzEAQBSUlJ3EbLHmrKNuRH/OOuN/5+YO2BP393cZMKY4wxxgyOEw6m5rffftNYIl3bv8DAQGOHylindvDgQYSFhSEtLQ29e/dWfXZeffVVjbJ+fn6Qy+UwNTVF//798fPPPxshYv3U19cjJiYGHh4eEIvFsLGxwYABA3D58mVVmTVr1sDT0xPm5uaQyWTw9PRERESExhD5+9XW1sLT0xPLli1rVWy61hsVFQUvLy9YWVlBIpHAw8MDS5YsQWVlparMnj17sGbNGtUsyKwNjNuiY3zgNlr2COA25NZp7ffD8uXLafz48VReXq7a5u7uTt26dSMAtHfvXo1jsrKyaOLEiW2KtyP5+/tTv3796Pjx41RfX083btygCRMmUE5OjqrMuHHjaN26dVRYWEgVFRWUnJxMIpGI/vKXv7R47gULFhAACg8Pb1VsutY7evRo2rx5M5WUlFB5eTklJSWRSCSiF154Qa1cfHw8jR49mkpLS1sVD3/+7uInHIyxDlFTUwNfX98uX8eDfPTRR0hMTERycrLG+kEbNmyAiYkJgoKCcPv2bSNF2HaJiYnIyMhASkoKnnrqKZiZmUGhUCAzMxMDBgxQlROLxZg7dy4cHBxgaWmJgIAATJo0Cf/5z3/w+++/az330aNHcebMmTbFp2u9lpaWCAoKgp2dHeRyOaZNmwZ/f3/s378fV69eVZV7//33MXjwYLz00ktaZ0xmuuGEgzHWIbZt24bCwsIuX0dLLly4gIhKXoQtAAAgAElEQVSICKxcuVI10d69fH19ERISguvXr2PRokVGiLB9fPzxxxgyZAgGDhzYYrn09HSN6+Dk5AQAas0WSjU1NVi8eDHi4+PbFJ+u9e7du1djjSd7e3sAmksnREZG4tSpU22O7VHGCQdjTCsiQlxcHB5//HFIJBLY2tpi0qRJ+O2331RlgoODIRaLVdPXA8DcuXMhk8kgCIJqVt6QkBAsXLgQ+fn5EAQBHh4e2LBhA6RSKbp374533nkHCoUCUqkUvr6+amvbtKUO4O7srLrM7dIeNmzYACJqcTK76Oho9O3bF1u3bsXBgwdbPJ8u9yAhIQEymQwWFhbIzMzEiy++CCsrKzg7O2P37t1q52tsbMTy5cvh4uICc3NzDBo0CElJSXq9xrq6Ohw/fly1crG+8vLyYGNjA1dXV4194eHhqicT7a2leu91/fp1mJubw83NTW27ra0tRo8ejfj4+Ed+tEmrGbdFx/jAfTjYI6A1bcjLly8nsVhMO3bsoLKyMjp9+jQNGTKE7O3t6ebNm6pyM2fOpB49eqgdGxsbSwCoqKhItW3KlCnk7u6uVi4oKIhkMhmdPXuWamtrKTc3l4YNG0ZyuZwKCgrapY69e/eSXC6nqKgovV4/kf7fD7179yYvLy+t+9zd3enSpUtERHT06FEyMTGhXr16UWVlJRFp78Oh6z0IDw8nAHTo0CG6ffs2FRYW0qhRo0gmk1FdXZ2q3KJFi0gikVBqaiqVlpbS0qVLycTEhH788UedX+OlS5cIAHl7e9OYMWPI0dGRJBIJeXp60qZNm6ipqUnjmLq6Orp27Rpt3LiRJBIJ7dixQ6PMkSNHaMKECUREVFRU1KY+HPrUe6+qqiqSy+UUHBysdX9YWBgBoJMnT+oVB/fhuIufcDDGNNTU1CAuLg6TJ0/GK6+8AmtrawwcOBCffPIJiouL23V2WjMzM9Vf8F5eXkhISEBFRQW2b9/eLucfN24cysvLERER0S7na05VVRUuXbqk01o6Pj4+mD9/Pi5fvowPPvhAa5nW3ANfX19YWVnBwcEBgYGBqKqqQkFBAYC7Iz8SEhLg7++PKVOmwMbGBsuWLYNIJNLrWiubJBwcHLB69Wrk5ubi1q1bmDRpEubNm4fPP/9c45iePXvC2dkZkZGRWLt2rca6IjU1NQgJCUFCQoLOcejiQfXeLyYmBgqFAtHR0Vr39+nTBwCQk5PTrnE+KjjhYIxpyM3NRWVlJYYOHaq2fdiwYRCLxWpNHu1t6NChsLCwUGs26AoKCwtBRLCwsNCpfHR0NPr164fNmzfjyJEjGvvbeg/EYjGAu8NXAeDcuXOorq5W69Rpbm4OR0dHva61RCIBAPTv3x++vr6ws7ODtbU1Vq5cCWtra62J0NWrV1FYWIjPP/8cn332GZ544gm1vjZLly7F22+/repn0V4eVO+90tPTkZycjK+++kqjs6+S8t7eunWrXeN8VHDCwRjTUFZWBuBuL/772djYaF11tz1JJBIUFRUZtI72VltbC+DPH+QHkUql2L59OwRBwBtvvKGxCGF734OqqioAwLJly9Tm1Lly5YpGB8mWKBQKAJqrZovFYri6uiI/P1/jGJFIBAcHB/j5+SExMRG5ubmIiYkBABw5cgQ5OTmYPXu2Xq9HFy3Ve6/ExER89NFHyM7ORq9evZo9n7m5OYA/7zXTDyccjDENNjY2AKD1R62srAzOzs4Gq7u+vt7gdRiC8sdInwmifHx8sGDBAuTl5WHVqlVq+9r7Hig7Yq5fvx5EpPbv2LFjOp/H0tISffr0wdmzZzX2NTQ0wNrausXjPTw8YGpqitzcXAB3RxYdOnQIJiYmqiRIGevq1ashCILGKs+tcX+9Shs3bsTOnTtx+PBhPPbYYy2eo66uDsCf95rphxMOxpiGAQMGwNLSUuOL/sSJE6irq8OTTz6p2mZmZqZ6bN8esrOzQUQYMWKEweowhO7du0MQBL3n11i1ahU8PT1x8uRJte363ANd9OzZE1KpFKdOndLrOG2mT5+OkydP4uLFi6pt1dXVuHLlimqobElJCWbMmKFxbF5eHhobG9GzZ08AwPbt2zUSIOXTrfDwcBCRRrNSS3Stl4gQGhqKnJwcZGRkaH2SdD/lve3Ro4fO8bA/ccLBGNMglUqxcOFCpKenY+fOnSgvL0dOTg7mzJkDhUKBoKAgVVkPDw/88ccfyMjIQH19PYqKinDlyhWNc9rZ2eHGjRu4fPkyKioqVAlEU1MTSktL0dDQgNOnTyMkJAQuLi6YNWtWu9SRlZXVIcNiLSws0Lt3b1y7dk2v45RNK/fPB6HPPdC1ntdffx27d+9GQkICysvL0djYiGvXrqkmwwoMDESPHj0eOLX6ggUL4OrqilmzZqGgoAAlJSUIDQ1FTU2NqhOsTCbDgQMHcPjwYZSXl6O+vh4nT57Ea6+9BplMhgULFugVv67x6Vrv2bNnsXbtWmzZsgUikUhj+YZ169ZpnFt5bx80/whrhnEGx3Qe4GGx7BHQmmF5TU1NFBsbS3369CGRSES2trbk7+9P586dUytXUlJCzzzzDEmlUnJzc6P33nuPFi9eTADIw8NDNbz1559/JldXVzI3N6enn36abt68SUFBQSQSicjJyYnMzMzIysqKJk2aRPn5+e1Wx759+0gul1N0dLTe103f74fg4GASiURUXV2t2paenk7u7u4EgOzt7WnevHlaj128eLHGsFhd7sHmzZvJwsKCAFCfPn0oPz+fPv30U7KysiIA5OrqSufPnyciojt37lBoaCi5uLiQmZkZOTg40JQpUyg3N5eI7k5XDoCWL1/+wNd69epVevnll8nW1pYkEgkNHz6csrKy1MpMmDCB3NzcyNLSkiQSCbm7u1NgYKDa9OfaNDcsVtf4dKk3JyeHADT7LzY2VuO848aNIycnJ61Df1vCw2LveuSvACcc7FHQWb/wgoKCyM7OzthhNEvf74e8vDwyMzN74HwPnVVjYyONGjWKtm3bZuxQtDJmfMXFxSSVSmndunV6H9tZP38djZtUGGNG9TCtwunh4YGoqChERUVpnbq7M2tsbERGRgYqKio65WrQxo4vMjIS3t7eCA4O7vC6HxaccDDGWDsKCwtDQEAAAgMDu9QCbdnZ2UhLS0NWVpbOc4l0JGPGFxcXh1OnTmHfvn0QiUQdWvfDhBMOxphRLF26FNu3b8ft27fh5uaG1NRUY4fUblavXo3g4GB8+OGHxg5FZ2PHjsWuXbvU1qzpTIwVX2ZmJu7cuYPs7GzY2tp2aN0PGzNjB8AYezTFxMRonYTpYeHn5wc/Pz9jh8HaaOLEiZg4caKxw3go8BMOxhhjjBkcJxyMMcYYMzhOOBhjjDFmcJxwMMYYY8zgOOFgjDHGmMEJRETGDsKYBEEwdgiMMcYeAY/4zy0Pi01KSjJ2CIx1KevXrwcAzJ8/38iRMMa6kkf+CQdjTD/Tpk0DACQnJxs5EsZYV8J9OBhjjDFmcJxwMMYYY8zgOOFgjDHGmMFxwsEYY4wxg+OEgzHGGGMGxwkHY4wxxgyOEw7GGGOMGRwnHIwxxhgzOE44GGOMMWZwnHAwxhhjzOA44WCMMcaYwXHCwRhjjDGD44SDMcYYYwbHCQdjjDHGDI4TDsYYY4wZHCccjDHGGDM4TjgYY4wxZnCccDDGGGPM4DjhYIwxxpjBccLBGGOMMYPjhIMxxhhjBscJB2OMMcYMjhMOxhhjjBkcJxyMMcYYMzhOOBhjjDFmcJxwMMYYY8zgOOFgjDHGmMFxwsEYY4wxg+OEgzHGGGMGxwkHY4wxxgyOEw7GGGOMGRwnHIwxxhgzODNjB8AY67xOnDiBX375RW3bxYsXAQCffvqp2vbBgwfjqaee6rDYGGNdi0BEZOwgGGOd0969ezF+/HiYmprCxOTuA1HlV4YgCACApqYmNDY24osvvsBf//pXo8XKGOvcOOFgjDWrvr4e9vb2KC8vb7GclZUVioqKIBaLOygyxlhXw304GGPNEolEePnll1tMJHQpwxhjnHAwxlr08ssvo66urtn99fX1mDFjRgdGxBjrirhJhTHWoqamJjz22GO4deuW1v0ODg64efOmqo8HY4xpw98QjLEWmZiY4NVXX9XaZCIWizFr1ixONhhjD8TfEoyxB2quWaWurg4vv/yyESJijHU13KTCGNNJnz59cOHCBbVtvXv3Rn5+vpEiYox1JfyEgzGmk1deeQUikUj1f7FYjNdee82IETHGuhJ+wsEY08mFCxfQp08ftW3nzp1D3759jRQRY6wr4SccjDGdeHh4YPDgwRAEAYIgYPDgwZxsMMZ0xgkHY0xnf/vb32BqagpTU1P87W9/M3Y4jLEuhJtUGGM6u3HjBnr27AkiwtWrV+Hk5GTskBhjXYRGwnHs2DHExcUZKx7GWCeXnZ0NABgzZoxR42CMdV4LFiyAj4+P2jaNJpWrV68iNTW1w4JijHUtLi4ucHV1NXYYLUpNTcW1a9eMHUaXcvz4cRw/ftzYYbCHQGpqKq5evaqx3ay5A1JSUgwaEGOsa/rjjz8AAHZ2dkaOpHmCIGD+/PmYNm2asUPpMgICAgDwdz9rO0EQtG5vNuFgjDFtOnOiwRjrvHiUCmOMMcYMjhMOxhhjjBkcJxyMMcYYMzhOOBhjjDFmcJxwMMZYM/bt2wdra2t88cUXxg6l0zt48CDCwsKQlpaG3r17q6bAf/XVVzXK+vn5QS6Xw9TUFP3798fPP/9shIj1U19fj5iYGHh4eEAsFsPGxgYDBgzA5cuXVWXWrFkDT09PmJubQyaTwdPTExERESgvL2/x3LW1tfD09MSyZctaFZuu9UZFRcHLywtWVlaQSCTw8PDAkiVLUFlZqSqzZ88erFmzBo2Nja2KpSWccDDGWDN4ImbdrFixAhs2bMDSpUsxZcoUXLx4Ee7u7ujWrRt27tyJL7/8Uq38gQMHkJKSgvHjxyM3NxdDhgwxUuS6mz59Ov79739j165dqK6uxq+//gp3d3e1H+vvvvsOb731FgoKCnDr1i2sWrUKa9aswdSpU1s8d3h4OM6dO9fq2HSt9/Dhw5g3bx4uX76M4uJixMTEID4+XjUkGgAmTJgAqVSKsWPHoqysrNUxacMJB2OMNWPcuHG4ffs2xo8fb+xQUFNTA19fX2OHoeGjjz5CYmIikpOTIZfL1fZt2LABJiYmCAoKwu3bt40UYdslJiYiIyMDKSkpeOqpp2BmZgaFQoHMzEwMGDBAVU4sFmPu3LlwcHCApaUlAgICMGnSJPznP//B77//rvXcR48exZkzZ9oUn671WlpaIigoCHZ2dpDL5Zg2bRr8/f2xf/9+tYm63n//fQwePBgvvfQSGhoa2hTbvTjhYIyxLmDbtm0oLCw0dhhqLly4gIiICKxcuRJSqVRjv6+vL0JCQnD9+nUsWrTICBG2j48//hhDhgzBwIEDWyyXnp6ucR2U6w3d+yREqaamBosXL0Z8fHyb4tO13r1798LU1FStnL29PQCgurpabXtkZCROnTrV5tjuxQkHY4xpceTIEbi4uEAQBGzatAkAkJCQAJlMBgsLC2RmZuLFF1+ElZUVnJ2dsXv3btWxGzZsgFQqRffu3fHOO+9AoVBAKpXC19cXJ06cUJULDg6GWCyGo6OjatvcuXMhk8kgCAKKi4sBACEhIVi4cCHy8/MhCAI8PDwAAPv374eVlRVWr17dEZdEw4YNG0BEmDBhQrNloqOj0bdvX2zduhUHDx5s8XxEhLi4ODz++OOQSCSwtbXFpEmT8Ntvv6nK6HoPAKCxsRHLly+Hi4sLzM3NMWjQICQlJen1Guvq6nD8+HF4e3vrdZxSXl4ebGxstC4HEB4ernoy0d5aqvde169fh7m5Odzc3NS229raYvTo0YiPj2+/pkW6T1JSEmnZzBhjXQYASkpKavN5rl69SgBo48aNqm3h4eEEgA4dOkS3b9+mwsJCGjVqFMlkMqqrq1OVCwoKIplMRmfPnqXa2lrKzc2lYcOGkVwup4KCAlW5mTNnUo8ePdTqjY2NJQBUVFSk2jZlyhRyd3dXK7d3716Sy+UUFRXV5tc6depUmjp1ql7H9O7dm7y8vLTuc3d3p0uXLhER0dGjR8nExIR69epFlZWVRESUlZVFEydOVDtm+fLlJBaLaceOHVRWVkanT5+mIUOGkL29Pd28eVNVTtd7sGjRIpJIJJSamkqlpaW0dOlSMjExoR9//FHn13jp0iUCQN7e3jRmzBhydHQkiURCnp6etGnTJmpqatI4pq6ujq5du0YbN24kiURCO3bs0Chz5MgRmjBhAhERFRUVEQAKDw/XOS5tdKn3XlVVVSSXyyk4OFjr/rCwMAJAJ0+e1CuO5j5//ISDMcZawdfXF1ZWVnBwcEBgYCCqqqpQUFCgVsbMzEz117qXlxcSEhJQUVGB7du3t0sM48aNQ3l5OSIiItrlfPqoqqrCpUuX4O7u/sCyPj4+mD9/Pi5fvowPPvhAa5mamhrExcVh8uTJeOWVV2BtbY2BAwfik08+QXFxMT799FONY1q6B7W1tUhISIC/vz+mTJkCGxsbLFu2DCKRSK/rr2yScHBwwOrVq5Gbm4tbt25h0qRJmDdvHj7//HONY3r27AlnZ2dERkZi7dq1mD59usZrDQkJQUJCgs5x6OJB9d4vJiYGCoUC0dHRWvf36dMHAJCTk9Mu8XHCwRhjbSQWiwHcHTrZkqFDh8LCwkKtiaCrKiwsBBHBwsJCp/LR0dHo168fNm/ejCNHjmjsz83NRWVlJYYOHaq2fdiwYRCLxWpNUdrcfw/OnTuH6upqtU6d5ubmcHR01Ov6SyQSAED//v3h6+sLOzs7WFtbY+XKlbC2ttaaCF29ehWFhYX4/PPP8dlnn+GJJ55Q63+zdOlSvP3226p+Fu3lQfXeKz09HcnJyfjqq680OvsqKe/trVu32iU+TjgYY6wDSSQSFBUVGTuMNqutrQXw5w/yg0ilUmzfvh2CIOCNN95ATU2N2n7lEExLS0uNY21sbFBRUaFXfFVVVQCAZcuWqeYEEQQBV65c0egg2RKFQgEAqv40SmKxGK6ursjPz9c4RiQSwcHBAX5+fkhMTERubi5iYmIA3O0blJOTg9mzZ+v1enTRUr33SkxMxEcffYTs7Gz06tWr2fOZm5sD+PNetxUnHIwx1kHq6+tRVlYGZ2dnY4fSZsofI30miPLx8cGCBQuQl5eHVatWqe2zsbEBAK2JRWuumbIj5vr160FEav+OHTum83ksLS3Rp08fnD17VmNfQ0MDrK2tWzzew8MDpqamyM3NBXB3tNGhQ4dgYmKiSoKUsa5evRqCIOCnn37SOT5d61XauHEjdu7cicOHD+Oxxx5r8Rx1dXUA/rzXbcUJB2OMdZDs7GwQEUaMGKHaZmZm9sCmmM6oe/fuEARB7/k1Vq1aBU9PT5w8eVJt+4ABA2BpaanxY3vixAnU1dXhySef1Kuenj17QiqV4tSpU3odp8306dNx8uRJXLx4UbWturoaV65cUQ2VLSkpwYwZMzSOzcvLQ2NjI3r27AkA2L59u0YCpHziFR4eDiLSaFZqia71EhFCQ0ORk5ODjIwMrU+S7qe8tz169NA5npZwwsEYYwbS1NSE0tJSNDQ04PTp0wgJCYGLiwtmzZqlKuPh4YE//vgDGRkZqK+vR1FREa5cuaJxLjs7O9y4cQOXL19GRUUF6uvrkZWVZbRhsRYWFujduzeuXbum13HKppX754OQSqVYuHAh0tPTsXPnTpSXlyMnJwdz5syBQqFAUFCQ3vW8/vrr2L17NxISElBeXo7GxkZcu3ZNNRlWYGAgevTo8cCp1RcsWABXV1fMmjULBQUFKCkpQWhoKGpqalSdYGUyGQ4cOIDDhw+jvLwc9fX1OHnyJF577TXIZDIsWLBAr/h1jU/Xes+ePYu1a9diy5YtEIlEas1MgiBg3bp1GudW3tsHzT+iK044GGNMi02bNmHYsGEAgNDQUEycOBEJCQlYv349AGDQoEG4ePEitmzZgoULFwIAXnjhBeTl5anOUVtbi4EDB8Lc3ByjRo1C37598fXXX6v1e3j33XfxzDPP4OWXX0a/fv2watUq1SNsHx8f1QyQc+bMQffu3eHl5YWXXnoJf/zxR4dch5aMGzcOubm5av0x/u///g8eHh7Iz8/HsGHD8N5772kcN2LECK0/wCtWrEBMTAyioqJgb2+P0aNHo1evXsjOzoZMJgMAve5BfHw85s+fjzVr1qBbt25QKBQICQlBaWkpgLtNBoWFhcjMzGzxddra2uK7776Ds7MzvL294eTkhB9++AFffvmlan4OqVSKkSNHYvbs2XBycoJcLkdAQAB69eqF48ePq3Ve1ZUu8elaL7ViLo0ff/wRTk5OGDRokN7HanX/OFmeh4Mx1tWhnebhaIugoCCys7Mzagz6aM08HHl5eWRmZvbA+R46q8bGRho1ahRt27bN2KFoZcz4iouLSSqV0rp16/Q+trnPHz/hYIwxAzHEipudiYeHB6KiohAVFaV16u7OrLGxERkZGaioqEBgYKCxw9Fg7PgiIyPh7e2N4ODgdjsnJxyMMcZaLSwsDAEBAQgMDOxSC7RlZ2cjLS0NWVlZOs8l0pGMGV9cXBxOnTqFffv2QSQStdt5OeHoAOvWrVP16P7kk09U2/ft2wdra2t88cUXBo+hqakJ69ev13m1ydmzZ0Mul0MQhFb18o6KioKXlxesrKwgkUjg4eGBJUuWaP0r6MiRIxg5ciQsLCygUCgQGhqKO3fu6F1nWloaevfureoE9aDZF+Pi4iAIAkxMTODp6Ylvv/1W7zp1jUUQBIhEIjg5OWHmzJn49ddf262u+3X295u2ayMIAsRiMbp3744xY8YgNjZW1c7eFS1duhTbt2/H7du34ebmhtTUVGOHZFCrV69GcHAwPvzwQ2OHorOxY8di165dauvYdCbGii8zMxN37txBdnY2bG1t2/fk97excB8Ow8jLyyMA9PHHH6u27d27l6ysrGjPnj0Grfv8+fM0cuRIAkCDBw/W+bjdu3e3ah59IqLRo0fT5s2bqaSkhMrLyykpKYlEIhG98MILauXOnDlD5ubmFBERQZWVlXT06FGyt7en119/Xe86ldzd3QkAOTo6qq2rcK+GhgZydXUlADR27NhW16VLLNbW1kREVFlZSXv27CEXFxeytLSk3377zWD1doX3273XpqmpiUpLS+nrr7+mWbNmkSAIpFAo9Frz4l7oBH04uprW9OFgTJvmPn/8hMOIxo0bh9u3b2P8+PEGq+OXX37BBx98gDlz5rR6tcPWsLS0RFBQEOzs7CCXyzFt2jT4+/tj//79ql73wN0x+Y6Ojli5ciVkMhl8fHwQGhqKf/3rX22a/vnJJ5/EzZs3kZGRoXV/Wlpau08r/CAymQzjx4/HP/7xD1RWVmLjxo0dWn9nfr8JggAbGxuMGTMG27dvR3JyMm7duqWKmTHW9XHC8RAhIqSkpKjN7T948GCkpaVh5syZOk9BrCQIQqtj2bt3r8Y4e3t7ewBQTSvc0NCAL7/8EqNHj1ar68UXXwQRPXCoWkveffddAMDHH3+sdX9cXJxqGF1HGz58OADgzJkzRqm/vbT3++1eU6dOxaxZs1BYWKjWLMQY67ranHDEx8dDJpPBxMQETz75JHr06AGRSASZTIYhQ4Zg1KhRqhnfbGxssGTJErXjv/vuO3h5ecHa2hpSqRQDBw7EV199BQD417/+BUtLSwiCAFtbW2RkZOCnn36Cq6srTE1Ntc6u1pINGzZAKpWie/fueOedd6BQKCCVSuHr66uxMBARIS4uTrXSo62tLSZNmqTxV7eu5e535MgRuLi4QBAEbNq0CcDd8eUymQwWFhbIzMzEiy++CCsrKzg7O2P37t1qxzc2NiImJgb9+vWDubk57O3t4ebmhpiYGEybNk2v66J8HbGxsejXrx8kEgmsra2xePFivc/TkuvXr8Pc3Bxubm4AgIsXL6KyshIuLi5q5ZSrT54+fVq1bf/+/XpNcPTss8/i8ccfx9dff41z586p7fv+++9RXV0NPz8/rcca+j3Z0NAAQH0Nikft/aYL5eRYWVlZBjk/Y6yD3d/G0po+HCtWrCAAdOLECaqqqqLi4mJ64YUXCAB9+eWXVFRURFVVVRQcHEwA6NSpU6pjU1JSKDIykv744w8qKSmhESNGULdu3VT7z549SxYWFvTaa6+ptoWFhdHWrVv1ilEpKCiIZDIZnT17lmprayk3N5eGDRtGcrmcCgoKVOWWL19OYrGYduzYQWVlZXT69GkaMmQI2dvb082bN/Uup61N/erVqwSANm7cqNoWHh5OAOjQoUN0+/ZtKiwspFGjRpFMJlPrj7B69WoyNTWlzMxMqq6upv/+97/Uo0cPGjNmTLOv/amnnmq2TT08PJwEQaC///3vVFpaStXV1bR58+ZW9+G4X1VVFcnlcgoODlZt++abbwgAxcbGapQ3NzdX61uxd+9eksvlFBUV9cC63N3d6dKlS/SPf/yDAFBISIjafn9/f9q+fTtVVFRo7cPRnu/Je/spKO3YsYMA0OLFi1XbHrX3W3PX5l7l5eUEgHr27NlsmeaA+3DojftwsPbS3OevXROOiooK1bbPPvuMAFBOTo5q2w8//EAAKDExsdlzxcTEEAAqLCxUbfvnP/9JAGjnzp30+eef04IFC/SK715BQUEaX3I//vgjAaCVK1cSEVF1dTVZWlpSYGCgWjll/MofPV3LEen/A1BTU6Papvzhv3DhgmrbsGHDaPjw4Wr1vv3222RiYkJ37tzR+tqb+wGorq4mCwsL+stf/qK2vS2dRu8XHh5Offv2pfLyctW2AwcOEACKi4vTKG9lZUW+vr6tqkuZcJSVlZFMJiNbW1uqrq4mIqL8/HxydnamO3fuNJtw3EVqlrYAACAASURBVK8t78n7O42mpqZSjx49qHv37nTt2jUievTeb9quTXMEQSAbG5sWy2jDCYf+OOFg7aW5z5+ZYZ6b3F26F/jz8TEA1XjelhYqUpa5d8Kct99+G//5z3/wzjvv4Lnnnmv3IWZDhw6FhYWF6rF0bm4uKisrNRbQGTZsGMRisar5RddybaW8lvdet9raWkilUrVyjY2NEIlEGn0nHuTChQuorq7G2LFj2x6sFunp6UhOTsaBAwcgl8tV25Xx3/seUaqrq2vzCoXW1taYMWMGtmzZgsTERLz++utYv3493n33XYjFYtVKiA/S1vfk7du3IQgCTE1N4ejoiJdeegkrVqxQdVp91N5vuqqqqgIRwcrKqlXHT58+HdOnT2/nqB5+bem7xVhLDJZw6OrLL79EbGwscnNzVQvPaLN69WqkpqaisLDQIHFIJBLVin1lZWUAoHU1PRsbG9XyybqWM4SXXnoJsbGxyMzMhJ+fH3Jzc5GRkYG//vWvev8AKBfoUS6R3J4SExMRFxeH7OxsjaWQlePLy8vL1bZXV1ejtrYWCoWizfW/++672LJlCz755BP4+/sjJSXlgXNgtPd70traWvVe0eZRe7/p6vz58wAAT0/PVh0fEhICHx+f9gzpoaZcn2T+/PlGjoR1dc0l+kZNOAoKCuDv74/Jkyfjf//3f/HYY49h48aNGh1L6+vr8f7776tGFkRHR2PFihXtFkd9fT3Kysrg7OwM4O6XNwCtX+CtKWcIkZGR+O9//4tZs2ahsrISCoUC06ZNa9Wqkcq/XFsz2VZLNm7ciK+++gqHDx/W+iPp5uYGuVyusTLmhQsXAKBdFgzy9vbGiBEjcPz4cQQFBSEgIKDFyWyM8Z581N5vutq/fz+Au6OWWsPHx8dgHVofRikpKQDA14y1WadMOHJyclBfX493330XvXv3BqD9cd57772Ht956C5MnT8b169exatUq+Pn5tdtfL9nZ2SAijBgxAgAwYMAAWFpa4qefflIrd+LECdTV1eHJJ5/Uq5wh5ObmIj8/H0VFRTAza9ttHDBgAExMTPDNN99gzpw5bY6NiPDBBx+gtLQUGRkZzcZnZmaGl156Cd9++y2amppgYnJ30FRWVhYEQcCECRPaHAtw9ynH8ePHkZqaqraSpzbGeE8+au83Xdy8eRPr16+Hs7Mz3njjDYPXxxgzPKPOw6EcDnnw4EHU1tYiLy9Pox168+bNcHJywuTJkwEAMTEx8PLywsyZMzUexeuqqakJpaWlaGhowOnTpxESEgIXFxfVMDypVIqFCxciPT0dO3fuRHl5OXJycjBnzhwoFAoEBQXpVc4Q5s2bBxcXl3ZZMMnBwQFTpkxBamoqtm3bhvLycpw+fVptfgV9nD17FmvXrsWWLVsgEok0prBet26dqmxERARu3bqFFStWoKqqCseOHUNsbCxmzZqFfv36qcplZWXpNSz2XtOmTYO9vT38/f1VSURzjPGefNTeb/ciIlRWVqKpqQlEhKKiIiQlJWHkyJEwNTVFRkZGq/twMMY6mft7keo7SiU+Pp4sLCwIAPXq1Yu+++47+uijj8ja2poAUI8ePWjXrl2UmJhIPXr0IABka2tLu3fvJiKi0NBQsrOzIxsbGwoICKBNmzYRAHJ3dydvb28SBIHs7Ozo6NGjREQ0f/58MjExIQBkbW1NP/30k169Z4OCgkgkEpGTkxOZmZmRlZUVTZo0ifLz89XKNTU1UWxsLPXp04dEIhHZ2tqSv78/nTt3Tu9yf//731WvXSaT0eTJk2njxo3k6OhIAMjCwoImTJhAmzdvVl3LPn36UH5+Pn366adkZWVFAMjV1ZXOnz9PRESHDx+mbt26EQDVP5FIRI8//jilpaWp6j527BiNHDmSFAqFqpyjoyP5+vrSN998oypXUVFBs2fPpm7dupGlpSU9/fTTtHz5cgJAzs7O9Msvv+h8jXNyctTiuv/f/cNgv/nmGxo+fDhJJBJSKBS0ePFiqq2tVSuzb98+ksvlFB0d3Wy96enpqmnN7e3tad68eap9S5YsUb2HiIiWLVumuv4mJibk5eVF3333HRG1z3vy+++/p759+6pes0KhoICAgGZjf5Teb3v27KFBgwaRhYUFicVi1bVTjkgZPnw4RUVFUUlJSbPX60HAo1T0xqNUWHtp7vMn/P+dKsnJyZg+fTru2/zQeOedd5CSkoKSkhJjh9ImCQkJyMvLU3X0Au6O7Pjggw+QkJCA0tLSNo/yYEypq73fBEFAUlIS90fQQ0BAAIA/+3Iw1lrNff6MPkrFGO4d3tgV3bx5E8HBwRqruIrFYri4uKC+vh719fWd6geAdV38fmOMtYcuv5bKb7/9ptFHQNu/wMBAY4fabszNzSESibBt2zbcunUL9fX1uHHjBrZu3Yrly5cjMDCw3du9H8XrzO4yxvuNdT0HDx5EWFgY0tLS0Lt3b9V3wquvvqpR1s/PD3K5HKampujfvz9+/vlnI0SsuzVr1sDT0xPm5uaQyWTw9PREREREs322mpqasH79evj6+mrdP2bMmGa/Q7WN6Guv+KKiouDl5QUrKytIJBJ4eHhgyZIlan2z9uzZgzVr1hjmD/P721ge5uXpw8LCSCwWq/qbpKSkGDukVvv222/pueeeIysrKzI1NSVra2vy9fWlzZs3U319vbHDYw+ZrvZ+A/fh0Ftb+nAsX76cxo8frzaTsLu7u6rfz969ezWOycrKookTJ7Y63o40btw4WrduHRUWFlJFRQUlJyeTSCTSmJ2ZiOj8+fM0cuRIAtDsTLujR49uto/b888/b7D4Ro8eTZs3b6aSkhIqLy+npKQkEolE9MILL6iVi4+Pp9GjR1NpaanesRAZeGpzxhjrTIydcFRXV5OPj0+XqqO1CceHH35Iffv2VZsen+huwrFr1y4yMTEhJycnKisrU9vflRIOf39/jdcXEBBAAOjGjRuqbadOnaLJkyfTzp07ydvbu9mE4/nnn1dLzpSCgoLo0KFDBotv3Lhx1NDQoFZu2rRpBEBtLTEiouDgYPLx8WnVHxTNff66fJMKY4x1Ntu2bTPYrMgdWceDXLhwAREREVi5cqXG1PcA4Ovri5CQEFy/fh2LFi0yQoTtIz09XeP1KZcmuLc5YvDgwUhLS8PMmTPVVoO+3/79+9WWeQCAq1ev4syZM3j22WcNFt/evXs1Zga2t7cHcHeG53tFRkbi1KlTiI+P1zue5nDCwRh75BER4uLi8Pjjj0MikcDW1haTJk1Sra8EAMHBwRCLxaop+QFg7ty5kMlkEAQBxcXFAO5Oqb5w4ULk5+dDEAR4eHhgw4YNkEql6N69O9555x0oFApIpVL4+vqqzfPSljqAuz9krZ2vpjU2bNgAImpxkr7o6Gj07dsXW7duxcGDB1s8ny73ISEhATKZDBYWFsjMzMSLL774/9i797io6vx/4K+BucEwwyUQSC5y0Yg0L2krmKvlLpV+vSCi/MTWbDPSChE1xJQMiAUx5IvBtpqxfTUFVBZd0zJz6bK5Vqsm4qZkXkBFwLhfBIf37w+/M1+HGXBmmGEg38/HY/4553M+5z3nzDBvzvmczxsKhQIeHh7YtWuXRn9KpRIJCQnw8vKCjY0NHn30UeTn5/fuTf+vsrIyODg4wNvb2yT9paamYtmyZSbpC9A/vqtXr8LGxgY+Pj4ayx0dHTFp0iRkZmaa7qnVrpc8+JYKY2ygg4G3VBISEkgsFtP27duprq6OTp8+TWPGjCFnZ2eqrKxUt4uMjCRXV1eNbdPT0wkAVVdXq5eFhYWRn5+fRruoqCiSyWR09uxZamtro9LSUho3bhzJ5XKNy9m92ceBAwdILpdrVA7WlzG3VHx9fSkwMFDnOlXVZiKib775hqysrGjIkCHU1NRERLpvqeh7HlRVjj///HOqr6+nqqoqmjhxIslkMmpvb1e3W7lyJUkkEtqzZw/V1tbSmjVryMrKir777juD3qdKe3s7VVRU0ObNm0kikdD27du7bXuvasl3q6iooMDAQFIqlUbFZUx8RETNzc0kl8spOjpa5/r4+HijKoZ39/3jKxyMsftaa2srMjIyMHv2bCxYsAD29vYYMWIE3nvvPdTU1Bg9464uQqFQ/d97YGAgcnJy0NjYiNzcXJP0P23aNDQ0NGDdunUm6a8nzc3NuHjxIvz8/O7ZNigoCMuXL8elS5ewevVqnW2MOQ/BwcFQKBRwcXFBREQEmpubceXKFQB3Khzn5OQgNDQUYWFhcHBwwNq1ayESiYw+3p6envDw8MD69euxYcMGk1UjTk1NxWuvvaYu72AsQ+NLSUmBu7s7kpOTda4fOnQogDslH0yBEw7G2H2ttLQUTU1NGDt2rMbycePGQSwWa01tb0pjx46Fra2txi2DgaKqqgpEBFtbW73aJycn46GHHkJ2dja+/vprrfW9PQ9isRgA1NWdz507h5aWFgwfPlzdxsbGBm5ubkYf7/LyclRVVWHnzp348MMPMXr06F6Po7l27Rr279+vLq3RG4bEV1hYiIKCAnz66ada40lUVOf2xo0bvY4N4ISDMXafq6urAwCd8x84ODjorM5rShKJBNXV1Wbdhzm0tbUBQI+DI+8mlUqRm5sLgUCAF154Aa2trRrrTX0empubAQBr167VmOfi8uXLWgMk9SUSieDi4oKQkBDk5eWhtLQUKSkpRvWlkpaWhsWLF+scdGuu+PLy8pCamori4mIMGTKk2/5Uk/mpznVvccLBGLuvOTg4AIDOH7S6ujp4eHiYbd8dHR1m34e5qH6MDJkgKigoCLGxsSgrK0NSUpLGOlOfBxcXFwDApk2bQHemgFC/jh07ZlBfuvj7+8Pa2hqlpaVG91FZWYmdO3di6dKlvY6nq+7i27x5M3bs2IGjR4/iwQcf7LGP9vZ2ADDZLMKccDDG7mvDhw+HnZ0dvv/+e43lx48fR3t7Ox577DH1MqFQqL5kbwrFxcUgIowfP95s+zCXQYMGQSAQoL6+3qDtkpKSEBAQgJMnT2osN+Q86MPT0xNSqVRrSn5D3bx5E/Pnz9daXlZWBqVSCU9PT6P7TktLw4IFC+Dk5GT2+IgIcXFxKCkpQVFRkV4zmqrOraurq9Hx3Y0TDsbYfU0qlWLFihUoLCzEjh070NDQgJKSEixZsgTu7u6IiopSt/X398cvv/yCoqIidHR0oLq6GpcvX9bq08nJCdeuXcOlS5fQ2NioTiA6OztRW1uL27dv4/Tp04iJiYGXl5fG/fve7OPQoUN99lisra0tfH19UVFRYdB2qlsrXeeDMOQ86LufRYsWYdeuXcjJyUFDQwOUSiUqKipw/fp1AEBERARcXV17nFpdJpPh8OHDOHr0KBoaGtDR0YGTJ09i4cKFkMlkiI2NNSgulRs3buCDDz7A8uXLu21jyvjOnj2LDRs2YOvWrRCJRFpTqm/cuFGrb9W5HTFihFHvUUvXx1b4sVjG2EAHAx+L7ezspPT0dBo6dCiJRCJydHSk0NBQOnfunEa7mzdv0pNPPklSqZR8fHzotddeo1WrVhEA8vf3Vz/eeuLECfL29iYbGxt64oknqLKykqKiokgkEtHgwYNJKBSSQqGgWbNm0YULF0y2j4MHD5JcLqfk5GSDj5kxj8VGR0eTSCSilpYW9bLCwkLy8/MjAOTs7Eyvvvqqzm1XrVql9VisPuchOzubbG1tCQANHTqULly4QFu2bCGFQkEAyNvbm86fP09ERLdu3aK4uDjy8vIioVBILi4uFBYWRqWlpUR0Z4ZOAJSQkNDj+5wxYwb5+PiQnZ0dSSQS8vPzo4iICCopKdFod+zYMZowYQK5u7urpyp3c3Oj4OBg+uKLLzTaxsbG0oIFC3rcrynjKykp6XY6dQCUnp6u1e+0adNo8ODB1NnZ2eP+u+ru+8cJB2PsV8fQhKMvREVFkZOTk6XD6JYxCUdZWRkJhcJ7zvfQXymVSpo4cSJt27bN0qHoZMn4ampqSCqV0saNGw3etrvvH99SYYyxPmKWCpwW5O/vj8TERCQmJmpMoT0QKJVKFBUVobGxsV9WubZ0fOvXr8eoUaMQHR1tsj454WCMMWa0+Ph4hIeHIyIiwuABpJZUXFyMvXv34tChQ3rPJdKXLBlfRkYGTp06hYMHD0IkEpmsX044GGPMzNasWYPc3FzU19fDx8cHe/bssXRIJvX2228jOjoaf/rTnywdit6mTJmCjz76SKNuTX9iqfj27duHW7duobi4GI6OjibtW2jS3hhjjGlJSUnp9QRR/V1ISAhCQkIsHQbrpZkzZ2LmzJlm6ZuvcDDGGGPM7DjhYIwxxpjZccLBGGOMMbPjhIMxxhhjZtftoNGCgoK+jIMxxkzKFAW67ieqaaz5bz8zF8H/zgqmVlBQgHnz5lkqHsYYY4wNcPn5+Zg7d67GMq2EgzHGeqL6I8L/CTPGDMFjOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzHGGGNmxwkHY4wxxsyOEw7GGGOMmR0nHIwxxhgzO044GGOMMWZ2nHAwxhhjzOw44WCMMcaY2XHCwRhjjDGz44SDMcYYY2bHCQdjjDHGzI4TDsYYY4yZHSccjDHGGDM7TjgYY4wxZnaccDDGGGPM7DjhYIwxxpjZccLBGGOMMbPjhIMxxhhjZscJB2OMMcbMjhMOxhhjjJkdJxyMMcYYMztOOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzHGGGNmxwkHY4wxxsyOEw7GGGOMmR0nHIwxxhgzO044GGOMMWZ2AiIiSwfBGOufPvroI2zbtg2dnZ3qZRcvXgQA+Pj4qJdZWVnhj3/8IyIjI/s8RsbYwMAJB2OsW6dPn8bIkSP1avvDDz/g0UcfNXNEjLGBihMOxliPAgICcO7cuR7b+Pv7o6ysrI8iYowNRDyGgzHWo+eeew4ikajb9SKRCIsWLerDiBhjAxFf4WCM9ejnn3+Gv78/evpTUVZWBn9//z6MijE20PAVDsZYj3x9fTFmzBgIBAKtdQKBAGPHjuVkgzF2T5xwMMbu6Q9/+AOsra21lltbW+MPf/iDBSJijA00fEuFMXZPVVVVcHd313g8FrjzOOy1a9fg6upqocgYYwMFX+FgjN3ToEGDMGnSJI2rHNbW1pg8eTInG4wxvXDCwRjTy3PPPac1cPS5556zUDSMsYGGb6kwxvTS0NAAFxcXtLe3A7jzOGxVVRUcHBwsHBljbCDgKxyMMb0oFAo888wzEAqFEAqFmDp1KicbjDG9ccLBGNPbggULoFQqoVQquW4KY8wgfEuFMaa3trY2ODs7g4hQU1MDGxsbS4fEGBsgtBKOgoICzJs3z1LxMMYYY2yAy8/Px9y5czWWCXtqzBhjXZ06dQoCgUDvKrKWMG/ePMTExCAoKMjSoQwYmzZtAgAsX77cwpGwga67ixbdJhxdMxPGGAOA2bNnAwCEwm7/fFjcvHnzEBQUxH/HDLB7924A/Lef9Z7BCQdjjOnSnxMNxlj/xU+pMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzHGGGNmxwkHY4x14+DBg7C3t8ff//53S4fS7x05cgTx8fHYu3cvfH19IRAIIBAIdBb4CwkJgVwuh7W1NR555BGcOHHCAhHrLy0tDQEBAbCxsYFMJkNAQADWrVuHhoYGne07OzuxadMmBAcH61w/efJk9fHp+rKzszNbfImJiQgMDIRCoYBEIoG/vz9ef/11NDU1qdvs378faWlpUCqVBsdxL5xwMMZYN3giZv28+eabyMrKwpo1axAWFoaff/4Zfn5+eOCBB7Bjxw58/PHHGu0PHz6M3bt3Y/r06SgtLcWYMWMsFLl+vvrqKyxevBhXrlzBjRs3kJSUhLS0NMyZM0erbVlZGX77298iNjYWLS0tBu/riSeeMFt8R48exauvvopLly6hpqYGKSkpyMzMRHh4uLrNjBkzIJVKMWXKFNTV1RkcS0844WCMsW5MmzYN9fX1mD59uqVDQWtra7f/MVtSamoq8vLyUFBQALlcrrEuKysLVlZWiIqKQn19vYUi7D2xWIxXXnkFLi4usLOzQ3h4OGbNmoXPPvsM169fV7f74YcfsHr1aixZsgSjRo3qtj+pVIqGhgYQkcYrKioKr7/+utnis7OzQ1RUFJycnCCXyzF37lyEhobik08+QXl5ubrdsmXLMHLkSEydOhW3b982OJ7ucMLBGGMDwLZt21BVVWXpMDT89NNPWLduHd566y1IpVKt9cHBwYiJicHVq1excuVKC0RoGoWFhVrvb/DgwQCgcTti5MiR2Lt3LyIjIyGRSLrt75NPPtFKzsrLy3HmzBk89dRTZovvwIEDsLa21mjn7OwMAFpXY9avX49Tp04hMzPT4Hi6wwkHY4zp8PXXX8PLywsCgQDvvvsuACAnJwcymQy2trbYt28fnn32WSgUCnh4eGDXrl3qbbOysiCVSjFo0CC8/PLLcHd3h1QqRXBwMI4fP65uFx0dDbFYDDc3N/WyV155BTKZDAKBADU1NQCAmJgYrFixAhcuXIBAIIC/vz+AOz9cCoUCb7/9dl8cEi1ZWVkgIsyYMaPbNsnJyRg2bBjef/99HDlypMf+iAgZGRl4+OGHIZFI4OjoiFmzZuHHH39Ut9H3HACAUqlEQkICvLy8YGNjg0cffdRkZTvKysrg4OAAb29vk/SXmpqKZcuWmaQvQP/4rl69ChsbG/j4+Ggsd3R0xKRJk5CZmWm6W4vURX5+PulYzBhjAwYAys/P73U/5eXlBIA2b96sXvbGG28QAPr888+pvr6eqqqqaOLEiSSTyai9vV3dLioqimQyGZ09e5ba2tqotLSUxo0bR3K5nK5cuaJuFxkZSa6urhr7TU9PJwBUXV2tXhYWFkZ+fn4a7Q4cOEByuZwSExN7/V7nzJlDc+bMMWgbX19fCgwM1LnOz8+PLl68SERE33zzDVlZWdGQIUOoqamJiIgOHTpEM2fO1NgmISGBxGIxbd++nerq6uj06dM0ZswYcnZ2psrKSnU7fc/BypUrSSKR0J49e6i2tpbWrFlDVlZW9N133xn0PlXa29upoqKCNm/eTBKJhLZv395t29/85jc0cuRIvfqtqKigwMBAUiqVRsVlTHxERM3NzSSXyyk6Olrn+vj4eAJAJ0+eNCiO7r5/fIWDMcaMEBwcDIVCARcXF0RERKC5uRlXrlzRaCMUCtX/rQcGBiInJweNjY3Izc01SQzTpk1DQ0MD1q1bZ5L+DNHc3IyLFy/Cz8/vnm2DgoKwfPlyXLp0CatXr9bZprW1FRkZGZg9ezYWLFgAe3t7jBgxAu+99x5qamqwZcsWrW16OgdtbW3IyclBaGgowsLC4ODggLVr10IkEhl9/D09PeHh4YH169djw4YNJqusnpqaitdeew1WVr37STY0vpSUFLi7uyM5OVnn+qFDhwIASkpKehWXCiccjDHWS2KxGADQ0dHRY7uxY8fC1tZW4xbBQFVVVQUigq2trV7tk5OT8dBDDyE7Oxtff/211vrS0lI0NTVh7NixGsvHjRsHsViscStKl67n4Ny5c2hpacHw4cPVbWxsbODm5mb08S8vL0dVVRV27tyJDz/8EKNHj+71uJpr165h//79eP7553vVj6HxFRYWoqCgAJ9++qnWeBIV1bm9ceNGr2MDOOFgjLE+JZFIUF1dbekweq2trQ0AehwceTepVIrc3FwIBAK88MILaG1t1VivegRT1zwUDg4OaGxsNCi+5uZmAMDatWs15rm4fPmyUY+rAoBIJIKLiwtCQkKQl5eH0tJSpKSkGNWXSlpaGhYvXqxz0K254svLy0NqaiqKi4sxZMiQbvuzsbEB8H/nurc44WCMsT7S0dGBuro6eHh4WDqUXlP9GBkyQVRQUBBiY2NRVlaGpKQkjXUODg4AoDOxMOaYubi4AAA2bdqk9fjpsWPHDOpLF39/f1hbW6O0tNToPiorK7Fz504sXbq01/F01V18mzdvxo4dO3D06FE8+OCDPfbR3t4O4P/OdW9xwsEYY32kuLgYRITx48erlwmFwnveiumPBg0aBIFAYPD8GklJSQgICMDJkyc1lg8fPhx2dnb4/vvvNZYfP34c7e3teOyxxwzaj6enJ6RSKU6dOmXQdl3dvHkT8+fP11peVlYGpVIJT09Po/tOS0vDggUL4OTkZPb4iAhxcXEoKSlBUVGRXjOaqs6tq6ur0fHdjRMOxhgzk87OTtTW1uL27ds4ffo0YmJi4OXlpXG/3t/fH7/88guKiorQ0dGB6upqXL58WasvJycnXLt2DZcuXUJjYyM6Ojpw6NAhiz0Wa2trC19fX1RUVBi0nerWStf5IKRSKVasWIHCwkLs2LEDDQ0NKCkpwZIlS+Du7o6oqCiD97No0SLs2rULOTk5aGhogFKpREVFhXoyrIiICLi6uvY4tbpMJsPhw4dx9OhRNDQ0oKOjAydPnsTChQshk8kQGxtrUFwqN27cwAcffIDly5d328aU8Z09exYbNmzA1q1bIRKJtKZU37hxo1bfqnM7YsQIo96jlq6PrfBjsYyxgQ4meCx28+bN5ObmRgDI1taWZsyYQdnZ2WRra0sAaOjQoXThwgXasmULKRQKAkDe3t50/vx5IrrzWKxIJKLBgweTUCgkhUJBs2bNogsXLmjs5+bNm/Tkk0+SVColHx8feu2112jVqlUEgPz9/dWP0J44cYK8vb3JxsaGnnjiCaqsrKSDBw+SXC6n5OTkXr1XIuMei42OjiaRSEQtLS3qZYWFheTn50cAyNnZmV599VWd265atUrrsdjOzk5KT0+noUOHkkgkIkdHRwoNDaVz586p2xhyDm7dukVxcXHk5eVFQqGQXFxcKCwsjEpLS4mIKDQ0lABQQkJCj+9zxowZ5OPjQ3Z2diSRf5jHLAAAIABJREFUSMjPz48iIiKopKREo92xY8dowoQJ5O7uTgAIALm5uVFwcDB98cUXGm1jY2NpwYIFPe7XlPGVlJSoY9L1Sk9P1+p32rRpNHjwYOrs7Oxx/1119/3jhIMx9qtjioSjt6KiosjJycmiMRjCmISjrKyMhELhPed76K+USiVNnDiRtm3bZulQdLJkfDU1NSSVSmnjxo0Gb9vd949vqTDGmJmYo+Jmf+Lv74/ExEQkJiZqTKE9ECiVShQVFaGxsRERERGWDkeLpeNbv349Ro0ahejoaJP1yQlHH9i4caN6gNV7772nXt6Xpa/vVS65qxdffBFyuRwCgcCoQVf6lEHuTXy6dC2Lfa/JkDIyMiAQCGBlZYWAgAB8+eWXRu/7XrEIBAKIRCIMHjwYkZGR+M9//mOyfXXV3z9vuo6NQCCAWCzGoEGDMHnyZKSnp6O2ttbscbLei4+PR3h4OCIiIgZUgbbi4mLs3bsXhw4d0nsukb5kyfgyMjJw6tQpHDx4ECKRyHQdd73kwbdUzKOsrIwA0J///Gf1sgMHDpBCoaD9+/ebdd/nz5+nCRMmEAC9p9olItq1a5dR09oSEU2aNImys7Pp5s2b1NDQQPn5+SQSieiZZ54xWXzdUd0/dnNz05jm+G63b98mb29vAkBTpkzp9T57isXe3p6IiJqammj//v3k5eVFdnZ29OOPP5ptvwPh83b3sens7KTa2lr6xz/+Qc8//zwJBAJyd3c3egpqWPiWSnx8PInFYgJAQ4YMod27d1ssFn0Zc0vlbp9++inFxcWZMCJmCUVFRZSSkkK3b982uo/uvn9C06UuzFCq0tfm9MMPPyAxMRFLlixBc3Oz6Yrw3IOqDLJqJPrcuXOxd+9eFBQUoLy8XP2olrnie+yxx/Dvf/8bRUVFCA8P11q/d+9eDB48WOfTAOYik8kwffp0KJVKhIaGYvPmzeqiYH2hP3/eBAIBHBwcMHnyZEyePBnTpk3DvHnzMG3aNJw/fx729vZmjdvUUlJSej0h1EATEhKCkJAQS4fBemnmzJmYOXOmWfrmWyq/IkSE3bt3a9Qc0Ldcsi4CgcDoWPQtg9yb+Hqimkjnz3/+s871GRkZWLFihcn2Z4jHH38cAHDmzBmL7N9UTP15u9ucOXPw/PPPo6qqSuO2EGNs4Op1wpGZmQmZTAYrKys89thjcHV1hUgkgkwmw5gxYzBx4kT1BCwODg54/fXXNbb/6quvEBgYCHt7e0ilUowYMQKffvopAOCvf/0r7OzsIBAI4OjoiKKiInz//ffw9vaGtbW1zslOeqJvyWhAvzLJhrTrqjelr4E7A4pSUlLw0EMPwcbGBs7OzvDx8UFKSgrmzp1r0HFRvY/09HQ89NBDkEgksLe3x6pVqwzupyfdlUHWh6FluJ966ik8/PDD+Mc//oFz585prPvnP/+JlpaWbv8bM/dn8vbt2wA0p4S+3z5v+lDNVXHo0CGz9M8Y62Nd77EYM4bjzTffJAB0/Phxam5uppqaGnrmmWcIAH388cdUXV1Nzc3NFB0dTQDo1KlT6m13795N69evp19++YVu3rxJ48ePpwceeEC9/uzZs2Rra0sLFy5UL4uPj6f333/foBhV9C0ZrW+ZZH3b6bqn3pvS12+//TZZW1vTvn37qKWlhf7973+Tq6srTZ48udv33lO55DfeeIMEAgG98847VFtbSy0tLZSdnW30GI6u7lUG+V7xGVKGW1UW+7//+78JAMXExGisDw0NpdzcXGpsbNQ5hsOUn8m7xymobN++nQDQqlWr1Mvut89bd8fmbg0NDQSAPD09u23THfSDx2IHmt6O4WBMpbvvn0kTjsbGRvWyDz/8kABoTDzy7bffEgDKy8vrtq+UlBQCQFVVVeplf/nLXwgA7dixg3bu3EmxsbEGxXe3qKgorT9y3333HQGgt956i4iIWlpayM7OjiIiIjTaqeJX/ejp247I8B+A1tZW9TLVD/9PP/2kXjZu3Dh6/PHHNfb70ksvkZWVFd26dUvne+/uB6ClpYVsbW3p97//vcby3gwa7eqNN96gYcOGUUNDQ7dt7vUDpS9VwlFXV0cymYwcHR3VExNduHCBPDw86NatW90mHF315jPZddDonj17yNXVlQYNGkQVFRVEdP993nQdm+4IBAJycHDosY0unHAYjhMOZirdff/MNmhUVSpYdfkYgPrxmp7qBqja3P38+ksvvYTPPvsML7/8Mn73u99hz549Jo21a8lofcsk97acsr50lb5ua2vTqi6oVCohEom0xk7cy08//YSWlhZMmTKl98HqoCqDfPjw4W7LIJuDvb095s+fj61btyIvLw+LFi3Cpk2bsHTpUojFYnVhonvp7Weyvr4eAoEA1tbWcHNzw9SpU/Hmm29i8ODBAO6/z5u+VINOFQqFUdubokDX/UQ1jXVBQYGFI2G/VhZ/SuXjjz9Geno6SktL1fPA6/L2229jz549qKqqMkscd5eM1rdMsqnLKRti6tSpSE9Px759+xASEoLS0lIUFRXhv/7rvwz+AVD9oVFVVzSlvLw8ZGRkoLi4+J6VCc1h6dKl2Lp1K9577z2EhoZi9+7d95wDw9SfSXt7e/VnRZf77fOmr/PnzwMAAgICjNo+MzMTmZmZpgzpvjBv3jxLh8B+pSz6lMqVK1cQGhoKNzc3HD9+HPX19UhLS9Nq19HRgWXLliEjIwPHjh1DcnKySePoWjJa3zLJpi6nbIj169fjqaeewvPPPw+FQoHZs2dj7ty52Lp1q8F9qf5zvXXrlkljNKQMsrmMGjUK48ePx7fffouoqCiEh4fD0dGx2/aW+Ezeb583fX3yyScAgGeffdao7fPz87XKkvOr+9ecOXMwZ84ci8fBr4H/6o5Fr3CUlJSgo6MDS5cuha+vLwDdj2K+9tprWLx4MWbPno2rV68iKSkJISEhCAoKMkkcXUtG61sm2dTllA1RWlqKCxcuoLq6GkJh707j8OHDYWVlhS+++AJLlizpdWxEhNWrV6O2thZFRUW9jq+3li5din/961/Ys2cPysrKemxric/k/fZ500dlZSU2bdoEDw8PvPDCC2bfH2PM/Cx6hcPLywsAcOTIEbS1taGsrEzrPnR2djYGDx6M2bNnA7gzoU5gYCAiIyPR0NBg1H7vVTJa3zLJpi6nbIhXX30VXl5eJqlf4OLigrCwMOzZswfbtm1DQ0MDTp8+rTG/giGMKYN8L70pwz137lw4OzsjNDRUnUR0xxKfyfvt83Y3IkJTUxM6OztBRKiurkZ+fj4mTJgAa2trFBUVGT2GgzHWz1AXhj6lkpmZqS4VPGTIEPrqq68oNTWV7O3tCQC5urrSRx99RHl5eeTq6koAyNHRkXbt2kVERHFxceTk5EQODg4UHh5O7777LgEgPz8/GjVqFAkEAnJycqJvvvmGiIiWL19OVlZWBIDs7e3p+++/1ztWIv1LRutTJlnfdu+88476vctkMpo9e3avS18fPXqUHnjgAY3ywiKRiB5++GHau3evet/6lktubGykF198kR544AGys7OjJ554ghISEggAeXh40A8//KD3MTakDLK+8elThrunstivv/66+jNERLR27Vr18beysqLAwED66quviMg0n8l//vOfNGzYMPV7cnd3p/Dw8G5jv58+b/v376dHH32UbG1tSSwWq4+d6omUxx9/nBITE+nmzZvdHq97AT+lYjB+SoWZSnffP8H/rlQrKCjAvHnzerwPM5C9/PLL2L17N27evGnpUHolJycHZWVl2LRpk3pZe3s7Vq9ejZycHNTW1sLGxsaCEbJfk4H2eRMIBMjPzzfbpGS/RqoSALt377ZwJGyg6+77Z/GnVCxhoJeMrqysRHR0tFYVV7FYDC8vL3R0dKCjo6Nf/QCwgYs/b4wxUxjwtVR+/PFHrTECul4RERGWDtVkbGxsIBKJsG3bNty4cQMdHR24du0a3n//fSQkJCAiIsLk973vx+PM7rDE540x9usz4BOOgIAAvR7TycvLw5o1a5Cbm4v6+nr4+PiYfAKxvmJvb4/Dhw/jzJkzGDZsGGxsbBAYGIjc3Fykpqbiww8/NPk+DTnO7NfFEp83NvAcOXIE8fHx2Lt3L3x9fdX/hDz33HNabUNCQiCXy2FtbY1HHnkEJ06csEDEhuvs7MSmTZsQHBysc31ycrLOf8SGDx+u1Xbnzp0YN24c5HI5vL29sWjRIlRWVhoVV1paGgICAmBjYwOZTIaAgACsW7dOaxB7YmIiAgMDoVAoIJFI4O/vj9dff11jMPj+/fuRlpZmnjsBXQd1GDO1OWOM9SfgQaMG682g0YSEBJo+fbpG6QI/Pz/1QOMDBw5obXPo0CGaOXOm0fH2tfPnz9OECRMIQLdT9iclJekcKP/II49otMvLyyMAlJaWRnV1dXTy5Eny9fWlUaNGUUdHh8GxTZs2jTZu3EhVVVXU2NhIBQUFJBKJtMpVTJo0ibKzs+nmzZvU0NBA+fn5JBKJ6JlnntFol5mZSZMmTaLa2lqDYyHq/vs34K9wMMZYf9Pa2trtf8EDaR/6SE1NRV5eHgoKCrRKF2RlZcHKygpRUVGor6+3UIS998MPP2D16tVYsmQJRo0a1WPb7du3a135PXPmjEabv/zlL3jwwQexatUq2NvbY9SoUYiNjcWpU6eMKlEgFovxyiuvwMXFBXZ2dggPD8esWbPw2Wef4fr16+p2dnZ2iIqKgpOTE+RyOebOnYvQ0FB88sknKC8vV7dbtmwZRo4cialTp2qUJ+ktTjgYY8zEtm3bZrYyDH25j3v56aefsG7dOrz11ltatXYAIDg4GDExMbh69SpWrlxpgQhNY+TIkdi7dy8iIyMhkUh63V95eTnc3d01JhX09PQEAFy+fNng/goLC7WOv6pW0923Sw4cOKBVisDZ2RkA0NLSorF8/fr1OHXqlEnLA3DCwRi77xERMjIy8PDDD0MikcDR0RGzZs1SF3QEgOjoaIjFYri5uamXvfLKK5DJZBAIBKipqQEAxMTEYMWKFbhw4QIEAgH8/f2RlZUFqVSKQYMG4eWXX4a7uzukUimCg4M1/qPtzT6AO9PBGztBnjGysrJARJgxY0a3bZKTkzFs2DC8//77OHLkSI/96XMecnJyIJPJYGtri3379uHZZ5+FQqGAh4cHdu3apdGfUqlEQkICvLy8YGNjg0cffRT5+fm9e9Mm4Ovrq5UsqsZv3GtyQn2VlZXBwcEB3t7ePba7evUqbGxs4OPjo7Hc0dERkyZNQmZmpummyeh6j4XHcDDGBjoYOIYjISGBxGIxbd++nerq6uj06dM0ZswYcnZ2psrKSnW7yMhIcnV11dg2PT2dAFB1dbV6WVhYGPn5+Wm0i4qKIplMRmfPnqW2tjYqLS2lcePGkVwupytXrphkHwcOHCC5XE6JiYl6v3cVY8Zw+Pr6UmBgoM51fn5+dPHiRSIi+uabb8jKyoqGDBlCTU1NRKR7DIe+5+GNN94gAPT5559TfX09VVVV0cSJE0kmk1F7e7u63cqVK0kikdCePXuotraW1qxZQ1ZWVvTdd98Z9D7v9pvf/KbHMRweHh7k4OBAIpGIhgwZQjNnzqRvv/1Wo11xcTGJRCLKysqihoYGOnPmDD388MP09NNPGx0XEVF7eztVVFTQ5s2bSSKR0Pbt23ts39zcTHK5nKKjo3Wuj4+PJwB08uRJg+Lo7vvHVzgYY/e11tZWZGRkYPbs2ViwYAHs7e0xYsQIvPfee6ipqTF6in9dhEKh+r/3wMBA5OTkoLGxEbm5uSbpf9q0aWhoaMC6detM0l9PmpubcfHiRfj5+d2zbVBQEJYvX45Lly5h9erVOtsYcx6Cg4OhUCjg4uKCiIgINDc348qVKwCAtrY25OTkIDQ0FGFhYXBwcMDatWshEolMdry7WrhwIfbv34/y8nI0NTVh165duHLlCiZNmoTS0lJ1u0mTJiEuLg7R0dFQKBQYPnw4Ghsb8f777/dq/56envDw8MD69euxYcOGe1b+TUlJgbu7e7fFJ4cOHQrgTo0pU+CEgzF2XystLUVTUxPGjh2rsXzcuHEQi8VGDeLT19ixY2Fra6txy2CgqKqqAhHB1tZWr/bJycl46KGHkJ2dja+//lprfW/Pg1gsBnCnkjMAnDt3Di0tLRqPpNrY2MDNzc1sx9vT0xOjR4+GnZ0dxGIxxo8fj9zcXLS2tiI7O1vd7o033sCWLVvw+eefo6mpCT///DOCg4MRFBSkMXjTUOXl5aiqqsLOnTvx4YcfYvTo0d2O8yksLERBQQE+/fRTrcG+Kqpze+PGDaNjuhsnHIyx+1pdXR2AOyP4u3JwcEBjY6NZ9y+RSFBdXW3WfZhDW1sbAOg9iFIqlSI3NxcCgQAvvPACWltbNdab+jw0NzcDANauXasxJ8bly5e1Bkia04gRI2BtbY3z588DAK5fv460tDS89NJLeOqppyCTyeDj44OtW7fi2rVrSE9PN3pfIpEILi4uCAkJQV5eHkpLS5GSkqLVLi8vD6mpqSguLsaQIUO67U81e7DqXPcWJxyMsfuag4MDAOj8Qaurq4OHh4fZ9t3R0WH2fZiL6sfIkAmigoKCEBsbi7KyMiQlJWmsM/V5cHFxAQBs2rRJ6zHVY8eOGdRXb3R2dqKzs1OdmJWVlUGpVOLBBx/UaKdQKODk5KRx66U3/P39YW1trdXf5s2bsWPHDhw9elQrhq7a29sBwGRlCzjhYIzd14YPHw47Ozt8//33GsuPHz+O9vZ2PPbYY+plQqFQfcneFIqLi0FEGD9+vNn2YS6DBg2CQCAweH6NpKQkBAQE4OTJkxrLDTkP+vD09IRUKtWqAWROTz/9tNay7777DkSEoKAgAFAnTnfPjwHcSbR++eUX9eOx+rp58ybmz5+vtVyV2Kj6IyLExcWhpKQERUVFOq8kdaU6t66urgbF1B1OOBhj9zWpVIoVK1agsLAQO3bsQENDA0pKSrBkyRK4u7sjKipK3dbf3x+//PILioqK0NHRgerqap3zJjg5OeHatWu4dOkSGhsb1QlEZ2cnamtrcfv2bZw+fRoxMTHw8vLC888/b5J9HDp0qM8ei7W1tYWvry8qKioM2k51a6XrfBCGnAd997No0SLs2rULOTk5aGhogFKpREVFhfrHPiIiAq6uriabWv3q1avIy8tDXV0dOjo6cOzYMbz44ovw8vLCkiVLAAA+Pj548sknsXXrVnz55ZdobW1FeXm5+v398Y9/VPenT3wymQyHDx/G0aNH0dDQgI6ODpw8eRILFy6ETCZDbGwsAODs2bPYsGEDtm7dCpFIpDX9+saNG7X6Vp3bESNGmOT48GOxjLFfHRj4WGxnZyelp6fT0KFDSSQSkaOjI4WGhtK5c+c02t28eZOefPJJkkql5OPjQ6+99hqtWrWKAJC/v7/68dYTJ06Qt7c32djY0BNPPEGVlZUUFRVFIpGIBg8eTEKhkBQKBc2aNYsuXLhgsn0cPHiQ5HI5JScnG3zMjHksNjo6mkQiEbW0tKiXFRYWkp+fHwEgZ2dnevXVV3Vuu2rVKq3HYvU5D9nZ2WRra0sAaOjQoXThwgXasmULKRQKAkDe3t50/vx5IiK6desWxcXFkZeXFwmFQnJxcaGwsDAqLS0lIqLQ0FACQAkJCT2+z2PHjtGECRPI3d1dPV25m5sbBQcH0xdffKFut2LFCvLz8yOZTEZCoZA8PDxo8eLFdO3aNY3+ampqKCYmhvz9/UkikZCdnR1NmDCB/va3v2m00ze+GTNmkI+PD9nZ2ZFEIiE/Pz+KiIigkpISdZuSkhKd066rXunp6Vr9Tps2jQYPHkydnZ097r+r7r5/nHAwxn51DE04+kJUVBQ5OTlZOoxuGZNwlJWVkVAovOd8D/2VUqmkiRMn0rZt2ywdik6WjK+mpoakUilt3LjR4G27+/7xLRXGGOsjZqnAaUH+/v5ITExEYmKixhTaA4FSqURRUREaGxsRERFh6XC0WDq+9evXY9SoUYiOjjZZn5xwMMYYM1p8fDzCw8MRERExoAq0FRcXY+/evTh06JDec4n0JUvGl5GRgVOnTuHgwYMQiUQm65cTDsYYM7M1a9YgNzcX9fX18PHxwZ49eywdkkm9/fbbiI6Oxp/+9CdLh6K3KVOm4KOPPtKoW9OfWCq+ffv24datWyguLoajo6NJ+xaatDfGGGNaUlJSdE7A9GsSEhKCkJAQS4fBemnmzJmYOXOmWfrmKxyMMcYYMztOOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNl1O2g0PDy8L+NgjDGT2rRpE3bv3m3pMAaMf/3rXwD4bz8zH8H/zgqmduzYMWRkZFgqHsZYP6cqujV69GgLR8IY669iY2PVBetUtBIOxhjrydy5cwEABQUFFo6EMTaQ8BgOxhhjjJkdJxyMMcYYMztOOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzHGGGNmxwkHY4wxxsyOEw7GGGOMmR0nHIwxxhgzO044GGOMMWZ2nHAwxhhjzOw44WCMMcaY2XHCwRhjjDGz44SDMcYYY2bHCQdjjDHGzI4TDsYYY4yZHSccjDHGGDM7TjgYY4wxZnaccDDGGGPM7DjhYIwxxpjZccLBGGOMMbPjhIMxxhhjZscJB2OMMcbMjhMOxhhjjJkdJxyMMcYYMztOOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzHGGGNmxwkHY4wxxsxOaOkAGGP9V0tLC27duqWxrL29HQBQW1ursVwikcDW1rbPYmOMDSwCIiJLB8EY659ycnLwyiuv6NU2OzsbS5cuNXNEjLGBihMOxli3qqur4e7uDqVS2WM7a2trXL9+HS4uLn0UGWNsoOExHIyxbrm4uGDKlCmwtrbuto21tTV+97vfcbLBGOsRJxyMsR4tWLAAPV0IJSIsWLCgDyNijA1EfEuFMdajxsZGuLi4aA0eVRGLxaiuroZCoejjyBhjAwlf4WCM9Ugul2P69OkQiURa64RCIWbOnMnJBmPsnjjhYIzdU2RkJG7fvq21XKlUIjIy0gIRMcYGGr6lwhi7p/b2djg7O6OxsVFjuZ2dHWpqaiCRSCwUGWNsoOArHIyxexKLxQgPD4dYLFYvE4lEmDdvHicbjDG9cMLBGNPL/Pnz1bOMAkBHRwfmz59vwYgYYwMJ31JhjOmls7MTbm5uqK6uBgA4OzujsrKyxzk6GGNMha9wMMb0YmVlhfnz50MsFkMkEiEyMpKTDcaY3jjhYIzp7f/9v/+H9vZ2vp3CGDMYV4s1UEVFBb755htLh8GYRRARHnjgAQDAxYsXcenSJcsGxJiFBAcHw8PDw9JhDCg8hsNABQUFmDdvnqXDYIwxZkH5+fmYO3eupcMYUPgKh5E4T2OWJBAILPYH7+zZswCAwMDAPt93b4SHhwMAdu/ebeFI2EAnEAgsHcKAxAkHY8wgAy3RYIz1DzxolDHGGGNmxwkHY4wxxsyOEw7GGGOMmR0nHIwxxhgzO044GGOMMWZ2nHAwdh87ePAg7O3t8fe//93SofR7R44cQXx8PPbu3QtfX18IBAIIBAI899xzWm1DQkIgl8thbW2NRx55BCdOnLBAxIbr7OzEpk2bEBwcrHN9cnKy+n3f/Ro+fLhW2507d2LcuHGQy+Xw9vbGokWLUFlZaVRcaWlpCAgIgI2NDWQyGQICArBu3To0NDRotEtMTERgYCAUCgUkEgn8/f3x+uuvo6mpSd1m//79SEtLg1KpNCoWZjxOOBi7j/F8Mvp58803kZWVhTVr1iAsLAw///wz/Pz88MADD2DHjh34+OOPNdofPnwYu3fvxvTp01FaWooxY8ZYKHL9lZWV4be//S1iY2PR0tLSq77y8/MRGRmJ8PBwVFRUYN++ffjyyy/x7LPP4vbt2wb399VXX2Hx4sW4cuUKbty4gaSkJKSlpWHOnDka7Y4ePYpXX30Vly5dQk1NDVJSUpCZmamegwUAZsyYAalUiilTpqCurq5X75MZhhMOxu5j06ZNQ319PaZPn27pUNDa2trtf9aWlJqairy8PBQUFEAul2usy8rKgpWVFaKiolBfX2+hCHvvhx9+wOrVq7FkyRKMGjWqx7bbt28HEWm8zpw5o9HmL3/5Cx588EGsWrUK9vb2GDVqFGJjY3Hq1CkcP37c4PjEYjFeeeUVuLi4wM7ODuHh4Zg1axY+++wzXL9+Xd3Ozs4OUVFRcHJyglwux9y5cxEaGopPPvkE5eXl6nbLli3DyJEjMXXqVKMSIGYcTjgYY/3Ctm3bUFVVZekwNPz0009Yt24d3nrrLUilUq31wcHBiImJwdWrV7Fy5UoLRGgaI0eOxN69exEZGQmJRNLr/srLy+Hu7q4xI6enpycA4PLlywb3V1hYqHX8Bw8eDAAat0sOHDigVcHY2dkZALSu2qxfvx6nTp1CZmamwfEw43DCwdh96uuvv4aXlxcEAgHeffddAEBOTg5kMhlsbW2xb98+PPvss1AoFPDw8MCuXbvU22ZlZUEqlWLQoEF4+eWX4e7uDqlUiuDgYI3/YKOjoyEWi+Hm5qZe9sorr0Amk0EgEKCmpgYAEBMTgxUrVuDChQsQCATw9/cHAHzyySdQKBR4++23++KQaMnKygIRYcaMGd22SU5OxrBhw/D+++/jyJEjPfZHRMjIyMDDDz8MiUQCR0dHzJo1Cz/++KO6jb7nAACUSiUSEhLg5eUFGxsbPProo8jPz+/dmzYBX19freRRNX7D19fXJPsoKyuDg4MDvL29e2x39epV2NjYwMfHR2O5o6MjJk2ahMzMTL612FeIGSQ/P5/4sDFLA0D5+fm97qe8vJwA0ObNm9XL3njjDQJAn3/+OdXX11NVVRVNnDiRZDIZtbe3q9tFRUWRTCao6an4AAAgAElEQVSjs2fPUltbG5WWltK4ceNILpfTlStX1O0iIyPJ1dVVY7/p6ekEgKqrq9XLwsLCyM/PT6PdgQMHSC6XU2JiYq/f65w5c2jOnDkGbePr60uBgYE61/n5+dHFixeJiOibb74hKysrGjJkCDU1NRER0aFDh2jmzJka2yQkJJBYLKbt27dTXV0dnT59msaMGUPOzs5UWVmpbqfvOVi5ciVJJBLas2cP1dbW0po1a8jKyoq+++47g97n3X7zm9/QyJEjda5LSkoiDw8PcnBwIJFIREOGDKGZM2fSt99+q9GuuLiYRCIRZWVlUUNDA505c4Yefvhhevrpp42Oi4iovb2dKioqaPPmzSSRSGj79u09tm9ubia5XE7R0dE618fHxxMAOnnypEFxmOr7d7/hKxyMMZ2Cg4OhUCjg4uKCiIgINDc348qVKxpthEKh+r/1wMBA5OTkoLGxEbm5uSaJYdq0aWhoaMC6detM0p8hmpubcfHiRfj5+d2zbVBQEJYvX45Lly5h9erVOtu0trYiIyMDs2fPxoIFC2Bvb48RI0bgvffeQ01NDbZs2aK1TU/noK2tDTk5OQgNDUVYWBgcHBywdu1aiEQikx3/rhYuXIj9+/ejvLwcTU1N2LVrF65cuYJJkyahtLRU3W7SpEmIi4tDdHQ0FAoFhg8fjsbGRrz//vu92r+npyc8PDywfv16bNiw4Z6Vu1NSUuDu7o7k5GSd64cOHQoAKCkp6VVcTD+ccDDG7kksFgMAOjo6emw3duxY2NraatwiGKiqqqpARLC1tdWrfXJyMh566CFkZ2fj66+/1lpfWlqKpqYmjB07VmP5uHHjIBaL7zmYsus5OHfuHFpaWjQeSbWxsYGbm5vZjr+npydGjx4NOzs7iMVijB8/Hrm5uWhtbUV2dra63RtvvIEtW7bg888/R1NTE37++WcEBwcjKChIY/CmocrLy1FVVYWdO3fiww8/xOjRo7sd91NYWIiCggJ8+umnWoN9VVTn9saNG0bHxPTHCQdjzKQkEgmqq6stHUavtbW1AYDegyilUilyc3MhEAjwwgsvoLW1VWO96hFMOzs7rW0dHBzQ2NhoUHzNzc0AgLVr12rMiXH58uVeP9ZqiBEjRsDa2hrnz58HAFy/fh1paWl46aWX8NRTT0Emk8HHxwdbt27FtWvXkJ6ebvS+RCIRXFxcEBISgry8PJSWliIlJUWrXV5eHlJTU1FcXIwhQ4Z025+NjQ2A/zvXzLw44WCMmUxHRwfq6urg4eFh6VB6TfVjZMgEUUFBQYiNjUVZWRmSkpI01jk4OACAzsTCmGPm4uICANi0aZPWY6rHjh0zqK/e6OzsRGdnpzoxKysrg1KpxIMPPqjRTqFQwMnJSePWS2/4+/vD2tpaq7/Nmzdjx44dOHr0qFYMXbW3twP4v3PNzIsTDsaYyRQXF4OIMH78ePUyoVB4z1sx/dGgQYMgEAgMnl8jKSkJAQEBOHnypMby4cOHw87ODt9//73G8uPHj6O9vR2PPfaYQfvx9PSEVCrFqVOnDNquN55++mmtZd999x2ICEFBQQCgTpzunh8DuJNo/fLLL+rHY/V18+ZNzJ8/X2u5KrFR9UdEiIuLQ0lJCYqKinReSepKdW5dXV0NiokZhxMOxpjROjs7UVtbi9u3b+P06dOIiYmBl5cXnn/+eXUbf39//PLLLygqKkJHRweqq6t1zsXg5OSEa9eu4dKlS2hsbERHRwcOHTpkscdibW1t4evri4qKCoO2U91a6TofhFQqxYoVK1BYWIgdO3agoaEBJSUlWLJkCdzd3REVFWXwfhYtWoRdu3YhJycHDQ0NUCqVqKioUP/YR0REwNXV1WRTq1+9ehV5eXmoq6tDR0cHjh07hhdffBFeXl5YsmQJAMDHxwdPPvkktm7dii+//BKtra0oLy9Xv78//vGP6v70iU8mk+Hw4cM4evQoGhoa0NHRgZMnT2LhwoWQyWSIjY0FAJw9exYbNmzA1q1bIRKJtKZf37hxo1bfqnM7YsQIkxwfdg+We0BmYOLHYll/ABM8lrd582Zyc3MjAGRra0szZsyg7OxssrW1JQA0dOhQunDhAm3ZsoUUCgUBIG9vbzp//jwR3XksViQS0eDBg0koFJJCoaBZs2bRhQsXNPZz8+ZNevLJJ0kqlZKPjw+99tprtGrVKgJA/v7+6kdoT5w4Qd7e3mRjY0NPPPEEVVZW0sGDB0kul1NycnKv3iuRcY/FRkdHk0gkopaWFvWywsJC8vPzIwDk7OxMr776qs5tV61apfVYbGdnJ6Wnp9PQoUNJJBKRo6MjhYaG0rlz59RtDDkHt27dori4OPLy8iKhUEguLi4UFhZGpaWlREQUGhpKACghIaHH93ns2DGaMGECubu7EwACQG5ubhQcHExffPGFut2KFSvIz8+PZDIZCYVC8vDwoMWLF9O1a9c0+qupqaGYmBjy9/cniURCdnZ2NGHCBPrb3/6m0U7f+GbMmEE+Pj5kZ2dHEomE/Pz8KCIigkpKStRtSkpK1LHreqWnp2v1O23aNBo8eDB1dnb2uP+uTPH9ux/xL6eBOOFg/UF/+IMXFRVFTk5OFo3BEMYkHGVlZSQUCu8530N/pVQqaeLEibRt2zZLh6KTJeOrqakhqVRKGzduNHjb/vD9G4j4lgpjzGi/9oqb/v7+SExMRGJiosYU2gOBUqlEUVERGhsbERERYelwtFg6vvXr12PUqFGIjo7u833frzjh6CO3bt3CsmXL4ObmBltbW/zud79TD0p77733LB2eydyrvDVwZ0rtCRMmwNbWFu7u7oiLi8OtW7cM3lfXMuG6XqpH4jZu3PirPN7M/OLj4xEeHo6IiIgBVaCtuLgYe/fuxaFDh/SeS6QvWTK+jIwMnDp1CgcPHoRIJOrTfd/POOHoI++88w4++eQT/Pjjj8jMzMTLL7+Mb775xtJhmZQ+5a1LS0sREhKCKVOmoLq6GoWFhfjggw/UA84McXeZcHt7e/Ujgbdv30ZLSwtu3Lih/kO2cuXKX93xtqQ1a9YgNzcX9fX18PHxwZ49eywdklm9/fbbiI6Oxp/+9CdLh6K3KVOm4KOPPtKoY9OfWCq+ffv24datWyguLoajo2Of7vt+xwlHHykqKsLYsWPh4OCAl156CXPmzDGqH10lvPtDWW99y1snJSXBzc0Nb731FmQyGYKCghAXF4e//vWvJpsd0draGjY2Nhg0aBCGDRvWq7766/G2tJSUFNy6dQtEhIsXLxr9eR5IQkJCkJqaaukwWC/NnDkT8fHxWk8RMfPjhKOPVFRUmOTSna4S3v2hrLc+5a1v376Njz/+GJMmTdIoW/3ss8+CiLBv3z6Tx1VUVNSr7fvr8WaMsYGGEw4z++yzz+Dv74/r16/jww8/hEAg6HFCmq+++gqBgYGwt7eHVCrFiBEj8OmnnwLQXcK7u7LePZWtNqT8tSn9/PPPaGpqgpeXl8ZyVXGs06dPq5f1VVnyX/PxZoyx/oQTDjP7/e9/j59++gmurq5YuHAhiKjH0e43btzAvHnzcOnSJVy7dg12dnaIjIwEAGRmZmL69Onw8/MDEeGnn37SuQwAVq9ejQ0bNmDTpk24fv06pk+fjvnz5+P777/H0qVLsXz5crS2tkIulyM/Px8XLlyAr68vFi9ebLZZISsrKwFAq5CSVCqFjY2NRgEl1dMPnZ2dRu3r6NGjOif66erXfLwZY6w/4YSjn5kzZw7efPNNODo6wsnJCTNmzMDNmzcNKoZlSNlqfUqQm4rqSRRd905FIpFGsStDy5LX19drPJ0yZcoUvbb7NR9vxhjrT4SWDoD1TDXuw5D5DowtW61vCXJjSaVSAHfGcnTV3t7eqwJK9vb26mqcwJ1H7rrWrNDHQDremzZtwu7du43a9n70r3/9CwAQHh5u4UgYuz/xFY5+5uOPP8bkyZPh4uICiUSC119/3eA++kvZ6q5Uj781NDRoLG9paUFbWxvc3d1Ntq/Jkydj5cqV92z3az7ejDHWn/AVjn7kypUrCA0NxezZs/HBBx/gwQcfxObNmw3+Eby7bHVMTIw5QjWKj48P5HK5VuEu1TiIRx99tE/jGejHe/ny5Zg7d26f7W+gU13Z4KtCrLfufsqO6Y8Tjn6kpKQEHR0dWLp0KXx9fQEY98G2RNlqfQiFQkydOhVffvklOjs7YWV15wLboUOHIBAIMGPGjD6N59d+vBljrD/hWyr9iOpx0SNHjqCtrQ1lZWU4fvy4RhtdJby7LrO2tr5n2WpLWbduHW7cuIE333wTzc3NOHbsGNLT0/H888/joYceUrfri7Lk98PxZoyxfsNiZeMGKEOrxV66dIlGjx5NAEgoFNKYMWNoz5499M4775CrqysBIJlMRrNnzyYiori4OHJyciIHBwcKDw+nd999lwCQn58fXblyRWcJb13LeipbbUj5a33pW96aiOiLL76gxx9/nCQSCbm7u9OqVauora1No40+Zcn/+c9/0rBhwzT2N2XKFJ1tf23HG1yt0mDGVItlTBf+/hlHQETU92nOwFVQUIB58+aBDxuzJIFAgPz8fB7DYQAew8FMhb9/xuFbKowxxhgzO044mE4//vhjj2XfVa+IiAhLh8pYv3LkyBHEx8dj79698PX1VX9XnnvuOa22ISEhkMvlsLa2xiOPPIITJ05YIGL9JScn6/w7cPccNCpff/01JkyYAFtbW7i7uyMuLk49+Z8x/XV0dCAlJQX+/v4Qi8VwcHDA8OHDcenSJQDA/v37kZaWZtAcOqxvccLBdAoICFCXe+/plZeXZ+lQGes33nzzTWRlZWHNmjUICwvDzz//DD8/PzzwwAPYsWMHPv74Y432hw8fxu7duzF9+nSUlpZizJgxForctEpLSxESEoIpU6aguroahYWF+OCDD7BkyRKj+5w3bx7+53/+Bx999BFaWlrwn//8B35+fupSETNmzIBUKsWUKVM0JgFk/QcnHIwxg7W2tiI4OHjA78OUUlNTkZeXh4KCAq16QVlZWbCyskJUVBTq6+stFKFpbN++XesfjzNnzmi0SUpKgpubG9566y3IZDIEBQUhLi4Of/3rX7Vm39Wnv7y8PBQVFWH37t34zW9+A6FQCHd3d+zbt0/jasiyZcswcuRITJ06VeeMxsyyOOFgjBls27ZtqKqq+v/s3XtQU+e+P/53IAlJIAGsKKkI5SZe0Vp1C+oXO5y6t3q8ULzQVrvRUwdtLaKUIiooF62VHmDoht2xtTindiuoDFgV2/E46DilnvYowsZdFRTvcrEil4Bc8vn94S85xiAmkBCQz2uGP3zyrPV81nqQfNZaz3qeft+GqZSXlyM2Nhbx8fHaKfyf5u/vj4iICNy5c8egGXD7s/b2dhw7dgwBAQE689rMnj0bRIT8/Hyj9/n3v/8dEydOxLhx415Yd9u2bSguLkZaWprR7TDz4oSDsQGAiJCSkoJRo0bBxsYGjo6OWLhwoc7VZnh4OMRisXYKegD46KOPYGtrC4FAgNraWgBAREQEIiMjUVFRAYFAAC8vL6Snp0MikWDIkCFYvXo1lEolJBIJ/P39deY26UkbAHDixAmzz8/SHenp6SCiLievS0pKwogRI/DNN9/g5MmTXe7PkP7KzMyEra0tZDIZ8vPzMXv2bCgUCri4uGD//v06++vo6EBcXBxcXV0hlUrh6+uL7Ozsnh30c1y7dg2NjY3aeW40PD09AQAlJSVG7a+1tRW//PILJkyYYFB9R0dHBAQEIC0tjd8m7Gt67QXcl4Sx83AwZg4wch6AuLg4EovF9N1331FdXR2VlJTQxIkTafDgwXT//n1tvffee4+GDh2qs21ycjIBoJqaGm1ZcHAweXp66tQLCwsjW1tbunTpErW0tFBZWRlNnjyZ5HI53bx50yRtHD16lORyOSUkJBh87BrmnIfDw8ODRo8e3elnnp6edP36dSIi+vnnn8nKyopee+01amxsJCKigoICWrBggc42hvbX5s2bCQD993//Nz169Iiqq6tpxowZZGtrS62trdp6n3zyCdnY2NChQ4fo4cOHtGnTJrKysqJff/3VqONMTEwkFxcXcnBwIJFIRK+99hotWLCA/ud//kdb5/Tp0wSAkpOT9baXSqU6c+UYsr/r168TAJowYQLNnDmTnJ2dycbGhkaOHEl/+9vfSK1W67UTExNDAOjChQtGHZ+hjP3/x57gOxyMveSam5uRkpKCt99+G8uWLYO9vT3GjRuHr776CrW1tdi9e7fJ2hIKhdqr8tGjRyMzMxMNDQ3Iysoyyf7nzp2L+vp6xMbGmmR/ptDU1ITr169rr+C74ufnh/Xr16OyshIbN27stE53+svf3x8KhQJOTk4ICQlBU1MTbt68CQBoaWlBZmYmgoKCEBwcDAcHB2zZsgUikcjofvnrX/+KI0eO4NatW2hsbMT+/ftx8+ZNBAQEoKysDAC0b6JYW1vrbS8SidDc3GzU/jSDQp2cnLB9+3aUlZWhqqoKCxcuxNq1a/GPf/xDrx1vb28AT5YvYH0HJxyMveTKysrQ2NiISZMm6ZRPnjwZYrFYbzp3U5o0aRJkMpneQMGXSXV1NYgIMpnMoPpJSUnw8fFBRkYGzp49q/d5T/tLLBYDePIaKQBcvnwZKpVKZ3ClVCqFs7Oz0f0yfPhwvP7667Czs4NYLMbUqVORlZWF5uZmZGRkAIB2DEtngzZbW1shlUqN2p+NjQ0AYMyYMfD398egQYNgb2+P+Ph42Nvbd5qAafqiqqrKqONj5sUJB2MvOc0rgnZ2dnqfOTg4oKGhwazt29jYoKamxqxtWFJLSwuA//tifBGJRIKsrCwIBAKsXLlS54ofMH1/NTU1AQC2bNmiM9fFjRs3oFKpjNpXZ8aNGwdra2tcuXIFALTjc+rr63XqqVQqtLS0QKlUGrU/TX3N+B4NsVgMNzc3VFRU6O1Dk9Ro+ob1DZxwMPaSc3BwAIBOv6jq6urg4uJitrbb2trM3oalab7cjJlwys/PDxs2bMDVq1eRmJio85mp+8vJyQkAkJqaqvf6aVFRkVH76oxarYZardYmXO7u7pDL5bhx44ZOvfLycgCAr6+vUfuzs7ODt7c3Ll26pFe3vb0d9vb2euWtra0AoHM3hVkeJxyMveTGjh0LOzs7/Pbbbzrl586dQ2trK9544w1tmVAo1N6KN4XCwkIQEaZOnWq2NixtyJAhEAgERs+vkZiYiJEjR+LChQs65cb0lyGGDx8OiUSC4uJio7brzJ///Ge9sl9//RVEBD8/PwBP+nfOnDk4c+YM1Gq1tl5BQQEEAoHOmzyG7A94MunXhQsXcO3aNW2ZSqXCjRs3On1VVtMXQ4cO7cZRMnPhhIOxl5xEIkFkZCRyc3Oxb98+1NfXo7S0FGvWrIFSqURYWJi2rpeXF/744w/k5eWhra0NNTU1eleqADBo0CDcvXsXlZWVaGho0CYQarUaDx8+RHt7O0pKShAREQFXV1eEhoaapI2CgoI+91qsTCaDh4cHbt++bdR2mkcrzw6uNKa/DG1nxYoV2L9/PzIzM1FfX4+Ojg7cvn0b9+7dAwCEhIRg6NChL5xa/c6dOzhw4ADq6urQ1taGoqIifPDBB3B1ddWZRTQ2NhZVVVXYunUrmpqaUFRUhOTkZISGhsLHx8fo/W3YsAFubm4IDQ3FzZs38eDBA0RHR6O5ubnTwbeavjBk3g7Wiyz3gkz/xK/Fsr4ARr6Wp1arKTk5mby9vUkkEpGjoyMFBQXR5cuXdeo9ePCA3nzzTZJIJOTu7k4ff/wxRUVFEQDy8vLSvt56/vx5cnNzI6lUStOnT6f79+9TWFgYiUQiGjZsGAmFQlIoFLRw4UKqqKgwWRvHjx8nuVxOSUlJRp8zc74WGx4eTiKRiFQqlbYsNzeXPD09CQANHjyY1q5d2+m2UVFReq/FGtJfGRkZJJPJCAB5e3tTRUUF7d69mxQKBQEgNzc3unLlChERPX78mKKjo8nV1ZWEQiE5OTlRcHAwlZWVERFRUFAQAaC4uLgujzMyMpI8PT3J1taWhEIhubi40KpVq+ju3bt6dU+fPk1TpkwhGxsbUiqVFBUVRS0tLd3e361bt+idd94hR0dHsrGxoSlTplBBQUGncc6dO5eGDRvW6SuzpmDs/z/2BH9zGokTDtYX9MU/eGFhYTRo0CBLh/Fc5kw4rl69SkKhkL777juz7N/cOjo6aMaMGbRnzx5Lh9JjtbW1JJFI6IsvvjBbG33x/19/wI9UGGMmM1BX6vTy8kJCQgISEhK080b0Fx0dHcjLy0NDQ8NLsfrztm3bMGHCBISHh1s6FPYMTjgYY8wEYmJisHjxYoSEhPSrBdoKCwtx+PBhFBQUGDyXSF+VkpKC4uJiHD9+HCKRyNLhsGdwwsEY67FNmzYhKysLjx49gru7Ow4dOmTpkCxi+/btCA8Px2effWbpUAwWGBiI77//Xmd9m/4oPz8fjx8/RmFhIRwdHS0dDuuE0NIBMMb6vx07dmDHjh2WDqNPmDVrFmbNmmXpMAacBQsWYMGCBZYOg3WB73AwxhhjzOw44WCMMcaY2XHCwRhjjDGz44SDMcYYY2bHCQdjjDHGzI7fUukmgUBg6RDYALd06VIsXbrU0mH0O/x/lzHLEBARWTqI/uT27dv4+eefLR0GYxaTmpoKAFi/fr2FI2HMcvz9/eHi4mLpMPoVTjgYY0ZZsmQJACAnJ8fCkTDG+hMew8EYY4wxs+OEgzHGGGNmxwkHY4wxxsyOEw7GGGOMmR0nHIwxxhgzO044GGOMMWZ2nHAwxhhjzOw44WCMMcaY2XHCwRhjjDGz44SDMcYYY2bHCQdjjDHGzI4TDsYYY4yZHSccjDHGGDM7TjgYY4wxZnaccDDGGGPM7DjhYIwxxpjZccLBGGOMMbPjhIMxxhhjZscJB2OMMcbMjhMOxhhjjJkdJxyMMcYYMztOOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzHGGGNmxwkHY4wxxsyOEw7GGGOMmR0nHIwxxhgzO044GGOMMWZ2nHAwxhhjzOw44WCMMcaY2QktHQBjrO86d+4cLl68qFN27do1AMDu3bt1ysePH48//elPvRYbY6x/ERARWToIxljfdPToUcybNw/W1tawsnpyQ1TzJ0MgEAAA1Go1Ojo68MMPP+Df//3fLRYrY6xv44SDMfZcbW1tGDx4MOrr67usp1AoUFNTA7FY3EuRMcb6Gx7DwRh7LpFIhHfeeafLRMKQOowxxgkHY6xL77zzDlpbW5/7eVtbG959991ejIgx1h/xIxXGWJfUajVeffVVVFVVdfq5k5MT7t+/rx3jwRhjneG/EIyxLllZWWH58uWdPjIRi8UIDQ3lZIMx9kL8V4Ix9kLPe6zS2tqKd955xwIRMcb6G36kwhgziLe3N8rLy3XKPDw8UFFRYaGIGGP9Cd/hYIwZZNmyZRCJRNp/i8Vi/PWvf7VgRIyx/oTvcDDGDFJeXg5vb2+dssuXL2PEiBEWiogx1p/wHQ7GmEG8vLwwfvx4CAQCCAQCjB8/npMNxpjBOOFgjBns/fffh7W1NaytrfH+++9bOhzGWD/Cj1QYYwa7e/cuhg8fDiLCrVu3MGzYMEuHxBjrJzjh6EUpKSkoKiqydBiM9UhhYSEAYObMmRaNg7Ge8vPzw4YNGywdxoDBj1R6UVFREX755RdLh8FeUr/88kuv/H65urrCzc3N7O30htu3b+PQoUOWDoNZwC+//MIXgL1MaOkABpqpU6fi4MGDlg6DvYQWL14MAGb//frjjz8AAIMGDTJrO70hJycHS5cu5f+TA5Dm/wvrPZxwMMaM8jIkGoyx3sePVBhjjDFmdpxwMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzGm4/jx47C3t8cPP/xg6VD6vJMnTyImJgaHDx+Gh4eHdtr35cuX69WdNWsW5HI5rK2tMWbMGJw/f94CERsuKSlJezxP/4wdO1av7tmzZzFt2jTIZDIolUpER0fj8ePH3d5fW1sbduzYAS8vL4jFYjg4OGDs2LGorKwEABw5cgSff/45Ojo6zHLszDw44WCM6eC5AA2zdetWpKenY9OmTQgODsa1a9fg6emJV155Bfv27cOxY8d06v/00084ePAg5s2bh7KyMkycONFCkZtWWVkZZs2ahcDAQNTU1CA3Nxfffvst1qxZ0+19Ll26FP/1X/+F77//HiqVCv/617/g6emJxsZGAMD8+fMhkUgQGBiIuro6Ux0KMzNOOBhjOubOnYtHjx5h3rx5lg4Fzc3N8Pf3t3QYenbu3IkDBw4gJycHcrlc57P09HRYWVkhLCwMjx49slCEpvHdd9+BiHR+/vnPf+rUSUxMhLOzM+Lj42Fraws/Pz9ER0dj7969+P33343e34EDB5CXl4eDBw/iT3/6E4RCIZRKJfLz83Xuhqxbtw7jx4/HnDlz0N7ebr6TwEyGEw7GWJ+1Z88eVFdXWzoMHeXl5YiNjUV8fDwkEone5/7+/oiIiMCdO3fwySefWCDC3tPe3o5jx44hICAAAoFAWz579mwQEfLz843e59///ndMnDgR48aNe2Hdbdu2obi4GGlpaUa3w3ofJxyMMa2zZ8/C1dUVAoEAf/vb3wAAmZmZsLW1hUwmQ35+PmbPng2FQgEXFxfs379fu216ejokEgmGDBmC1atXQ6lUQiKRwN/fH+fOndPWCw8Ph1gshrOzs7bso48+gq2tLQQCAWprawEAERERiIyMREVFBQQCAby8vAAAJ06cgEKhwPbt23vjlOhJT08HEWH+/PnPrZOUlIQRI0bgm2++wcmTJ7vcHxEhJSUFo0aNgo2NDRwdHbFw4UKduwOG9gEAdHR0IC4uDq6urpBKpfD19UV2dnbPDvo5rl27hsbGRri6upRckgUAACAASURBVOqUe3p6AgBKSkqM2l9rayt++eUXTJgwwaD6jo6OCAgIQFpaGj8K7Ac44WCMaU2fPh0///yzTtmHH36I9evXo7m5GXK5HNnZ2aioqICHhwdWrVqFtrY2AE8SidDQUKhUKqxbtw6VlZU4f/482tvb8dZbb+HWrVsAnnxhL1myRKeNjIwMxMfH65SlpaVh3rx58PT0BBGhvLwcALQDBdVqtVnOwYscO3YMPj4+kMlkz60jlUqxd+9eWFlZYdWqVWhqanpu3W3btiEmJgabN29GdXU1zpw5g1u3bmHGjBmoqqoCYHgfAMDGjRuxa9cupKam4t69e5g3bx7effdd/Pbbb0Yfa0xMDBwdHSEWi+Hu7o6FCxfi119/1X5+//59ANB7rCSRSCCVSrXxG7q/u3fvorW1Ff/7v/+LN998U5u0jho1ChkZGZ0mFa+//jru3LmDixcvGn18rHdxwsEYM5i/vz8UCgWcnJwQEhKCpqYm3Lx5U6eOUCjUXq2PHj0amZmZaGhoQFZWlklimDt3Lurr6xEbG2uS/RmjqakJ169f117Bd8XPzw/r169HZWUlNm7c2Gmd5uZmpKSk4O2338ayZctgb2+PcePG4auvvkJtbS12796tt01XfdDS0oLMzEwEBQUhODgYDg4O2LJlC0QikdHn/69//SuOHDmCW7duobGxEfv378fNmzcREBCAsrIyANC+iWJtba23vUgkQnNzs1H70wwKdXJywvbt21FWVoaqqiosXLgQa9euxT/+8Q+9dry9vQEApaWlRh0f632ccDDGukUsFgOAztV1ZyZNmgSZTKY3gLA/qq6uBhF1eXfjaUlJSfDx8UFGRgbOnj2r93lZWRkaGxsxadIknfLJkydDLBbrPIrqzLN9cPnyZahUKp3BlVKpFM7Ozkaf/+HDh+P111+HnZ0dxGIxpk6diqysLDQ3NyMjIwMAtGNYOhu02draCqlUatT+bGxsAABjxoyBv78/Bg0aBHt7e8THx8Pe3r7TBEzTF8/eTWF9DyccjDGzs7GxQU1NjaXD6LGWlhYA//fF+CISiQRZWVkQCARYuXKlzhU/AO0rnXZ2dnrbOjg4oKGhwaj4NI9utmzZojPXxY0bN6BSqYzaV2fGjRsHa2trXLlyBQC043Dq6+t16qlUKrS0tECpVBq1P019zTgeDbFYDDc3N1RUVOjtQ5PUaPqG9V2ccDDGzKqtrQ11dXVwcXGxdCg9pvlyM2bCKT8/P2zYsAFXr15FYmKizmcODg4A0Gli0Z1z5uTkBABITU3Ve/20qKjIqH11Rq1WQ61WaxMud3d3yOVy3LhxQ6eeZryNr6+vUfuzs7ODt7c3Ll26pFe3vb0d9vb2euWtra0AoHM3hfVNnHAwxsyqsLAQRISpU6dqy4RC4QsfxfRFQ4YMgUAgMHp+jcTERIwcORIXLlzQKR87dizs7Oz0BnSeO3cOra2teOONN4xqZ/jw4ZBIJCguLjZqu878+c9/1iv79ddfQUTw8/MD8KQf58yZgzNnzugM4i0oKIBAINB5k8eQ/QFPJv26cOECrl27pi1TqVS4ceNGp6/Kavpi6NCh3ThK1ps44WCMmZRarcbDhw/R3t6OkpISREREwNXVFaGhodo6Xl5e+OOPP5CXl4e2tjbU1NToXSUDwKBBg3D37l1UVlaioaEBbW1tKCgosNhrsTKZDB4eHrh9+7ZR22kerTw7uFIikSAyMhK5ubnYt28f6uvrUVpaijVr1kCpVCIsLMzodlasWIH9+/cjMzMT9fX16OjowO3bt3Hv3j0AQEhICIYOHfrCqdXv3LmDAwcOoK6uDm1tbSgqKsIHH3wAV1dXnVlEY2NjUVVVha1bt6KpqQlFRUVITk5GaGgofHx8jN7fhg0b4ObmhtDQUNy8eRMPHjxAdHQ0mpubOx18q+kLQ+btYBZGrNcsWrSIFi1aZOkw2EvKFL9fX375JTk7OxMAkslkNH/+fMrIyCCZTEYAyNvbmyoqKmj37t2kUCgIALm5udGVK1eIiCgsLIxEIhENGzaMhEIhKRQKWrhwIVVUVOi08+DBA3rzzTdJIpGQu7s7ffzxxxQVFUUAyMvLi27evElEROfPnyc3NzeSSqU0ffp0un//Ph0/fpzkcjklJSX16FiJiLKzs8nYP4Ph4eEkEolIpVJpy3Jzc8nT05MA0ODBg2nt2rWdbhsVFUULFizQKVOr1ZScnEze3t4kEonI0dGRgoKC6PLly9o6xvTB48ePKTo6mlxdXUkoFJKTkxMFBwdTWVkZEREFBQURAIqLi+vyOCMjI8nT05NsbW1JKBSSi4sLrVq1iu7evatX9/Tp0zRlyhSysbEhpVJJUVFR1NLS0u393bp1i9555x1ydHQkGxsbmjJlChUUFHQa59y5c2nYsGGkVqu7PJ5n8d/j3icg4tlSesvixYsBAAcPHrRwJOxl1Bd+v1avXo2DBw/iwYMHFovBGDk5OVi6dKlRk0aVl5dj1KhRyMrKwrJly8wYnXmo1WrMnDkToaGhWLlypaXD6ZEHDx7AxcUFSUlJiIyMNGrbvvD/ZaDhRyqMMZN62Vfw9PLyQkJCAhISErTzRvQXHR0dyMvLQ0NDA0JCQiwdTo9t27YNEyZMQHh4uKVDYQbghIMxxowUExODxYsXIyQkpF8t0FZYWIjDhw+joKDA4LlE+qqUlBQUFxfj+PHjEIlElg6HGYATjn7mgw8+gFwuh0AgMMlIdEtISEjA6NGjoVAoYGNjAy8vL3z66aedXi2ePXsW06ZNg0wmg1KpRHR0tHZ2Q2McPnwYHh4eOnMTCAQCiMViDBkyBDNnzkRycjIePnxoikMckDZt2oSsrCw8evQI7u7uOHTokKVDMqvt27cjPDwcn332maVDMVhgYCC+//57nXVs+qP8/Hw8fvwYhYWFcHR0tHQ4zECccPQz33zzDb7++mtLh9Ejp06dwtq1a1FZWYna2lrs2LEDaWlp2meqGmVlZZg1axYCAwNRU1OD3NxcfPvttzoj2g0VHByMa9euwdPTE/b29iAiqNVqVFdXIycnB+7u7oiOjsaYMWO6teYEA3bs2IHHjx+DiHD9+nUsWrTI0iGZ3axZs7Bz505LhzHgLFiwADExMZ1Oqc76Lk44WK+zs7NDWFgYBg0aBLlcjiVLliAoKAgnTpzQLvAFPJm7wNnZGfHx8bC1tYWfnx+io6Oxd+9ek0yTLRAI4ODggJkzZyIrKws5OTmoqqrC3Llz+9VtcsYY6w844eiHBAKBpUPokaNHj+pdmQwePBgAtNMvt7e349ixYwgICNA53tmzZ4OIkJ+fb/K4Fi1ahNDQUFRXV+Orr74y+f4ZY2wg44SjjyMiJCcnw8fHBzY2NrC3t0dUVJRevY6ODsTFxcHV1RVSqRS+vr7Izs4GAGRmZsLW1hYymQz5+fmYPXs2FAoFXFxcsH//fp39nD59GlOmTIFMJoNCocC4ceO06yR01UZP3blzB1KpFO7u7gCAa9euobGxEa6urjr1NKt0lpSUaMtOnDhhsomgNJNTFRQUaMv6+7lljLG+gBOOPi42NhbR0dEICwtDVVUV7t+/3+lsexs3bsSuXbuQmpqKe/fuYd68eXj33Xfx22+/4cMPP8T69evR3NwMuVyO7OxsVFRUwMPDA6tWrdJOMd3U1IT58+dj0aJF+OOPP3D16lWMGDFCu1ZBV230hEqlwqlTp7Bq1Srt6pf3798HAMjlcp26EokEUqlUZ2VIzWuYT0+t3F0TJkwAAJ1plfvzuWWMsT7DgpOODTjGzmynUqlIJpPRW2+9pVO+f/9+AkAXLlwgIqLm5maSyWQUEhKis62NjQ19+OGHRES0efNmAkDNzc3aOhkZGQSAysvLiYjon//8JwGgo0eP6sViSBvdtXnzZhoxYgTV19dry3766ScCQCkpKXr1FQoF+fv7d6stT09Psre377KOQCAgBwcHIupf55ZnTjRed2YaZS8H/v/S+4QWy3TYC5WXl0OlUiEwMLDLepcvX4ZKpcLYsWO1ZVKpFM7Ozl0OrtTcTdBchXt4eGDIkCFYtmwZ1q1bh9DQULz22ms9auNFcnNzkZOTg59++knnboZEIgHwZCzHs1pbW822MmRTUxOICAqFAkD/O7eHDh3q92N8LIHP2cA0EN6k6ks44ejDNIsSaZacfp6mpiYAwJYtW7Blyxadz5RKpcHtSaVSnDp1Chs3bsT27duRkJCAJUuWICsry2RtPO3AgQNISUlBYWEhXn31VZ3PNPMEaMY4aKhUKrS0tHS7zRe5cuUKAGDkyJEA+t+5nTp1KtavX2/UNgNZUVER0tLSeLzMAJSammrpEAYcTjj6MM1V/osmutIkJKmpqYiIiOhRm2PGjMEPP/yAmpoapKSkYOfOnRgzZox2GmRTtAEAX375JX788UecOnUKdnZ2ep+7u7tDLpfrrSBaXl4OAPD19e1xDJ05ceIEgCdvwwD979y6uLhgyZIlPdrHQJOWlsbnbADiNVR6Hw8a7cPGjh0LKysrnD59ust6w4cPh0Qi6fHMo3fv3sWlS5cAPPmi/eyzzzBx4kRcunTJZG0QEaKjo1FaWoq8vLxOkw0AEAqFmDNnDs6cOaMzGLSgoAACgQDz58/vURyduX//PlJTU+Hi4qJd1Ko/nVvGGOvLOOHow5ycnBAcHIxDhw5hz549qK+vR0lJCXbv3q1TTyKRYMWKFdi/fz8yMzNRX1+Pjo4O3L59G/fu3TO4vbt372L16tX4/fff0draigsXLuDGjRuYOnWqydq4dOkSdu3aha+//hoikUhvqvEvvvhCWzc2NhZVVVXYunUrmpqaUFRUhOTkZISGhsLHx0dbr6CgwKjXYokIjY2NUKvVICLU1NQgOzsb06ZNg7W1NfLy8rRjOPrTuWWMsT7NsmNWB5bujIpuaGigDz74gF555RWys7Oj6dOnU1xcHAEgFxcXunjxIhERPX78mKKjo8nV1ZWEQiE5OTlRcHAwlZWVUUZGBslkMgJA3t7eVFFRQbt37yaFQkEAyM3Nja5cuUKVlZXk7+9Pjo6OZG1tTa+++ipt3ryZ2tvbX9iGoUpLSwnAc3+Sk5N16p8+fZqmTJlCNjY2pFQqKSoqilpaWnTqHD9+nORyOSUlJT233SNHjpCvry/JZDISi8VkZWVFALRvpEyZMoUSEhLowYMHetv2l3PLo+6Nx2+pDFz8/6X3CYiILJHoDESatUL42SEzB/79Ml5OTg6WLl0K/jM48PD/l97Hj1QYY4wxZnaccLAe+/333/XGYnT2o3kbg7GX2cmTJxETE4PDhw/Dw8ND+/u/fPlyvbqzZs2CXC6HtbU1xowZg/Pnz1sgYsMlJCRg9OjRUCgUsLGxgZeXFz799FM0Njbq1T179iymTZsGmUwGpVKJ6OjoTt+4e1G9I0eO4PPPP9fOKMz6L044WI+NHDkSRPTCnwMHDlg6VMbMauvWrUhPT8emTZsQHByMa9euwdPTE6+88gr27duHY8eO6dT/6aefcPDgQcybNw9lZWWYOHGihSI3zKlTp7B27VpUVlaitrYWO3bsQFpamvbxhEZZWRlmzZqFwMBA1NTUIDc3F99++y3WrFljdL358+dDIpEgMDAQdXV1vXKczDw44WCMmUxzczP8/f37fRvdsXPnThw4cAA5OTl6awClp6fDysoKYWFhePTokYUi7Dk7OzuEhYVh0KBBkMvlWLJkCYKCgnDixAncunVLWy8xMRHOzs6Ij4+Hra0t/Pz8EB0djb179+rMnmtovXXr1mH8+PGYM2dOp7MPs/6BEw7GmMns2bMH1dXV/b4NY5WXlyM2Nhbx8fHaCfue5u/vj4iICNy5cweffPKJBSI0jaNHj8La2lqnbPDgwQCezAIMPFmO4NixYwgICNCZMn727NkgIuTn5xtVT2Pbtm0oLi5GWlqaWY6NmR8nHIwNYESElJQUjBo1CjY2NnB0dMTChQt1ri7Dw8MhFou1080DwEcffQRbW1sIBALU1tYCACIiIhAZGYmKigoIBAJ4eXkhPT0dEokEQ4YMwerVq6FUKiGRSODv749z586ZpA3gyQyxxszFYmrp6ekgoi4npEtKSsKIESPwzTff4OTJk13uz5B+yczMhK2tLWQyGfLz8zF79mwoFAq4uLhg//79Ovvr6OhAXFwcXF1dIZVK4evra7Lp3O/cuQOpVAp3d3cAT1ZabmxshKurq049T09PAEBJSYlR9TQcHR0REBCAtLQ0fquon+KEg7EBbNu2bYiJicHmzZtRXV2NM2fO4NatW5gxYwaqqqoAPPkyfXbq74yMDMTHx+uUpaWlYd68efD09AQRoby8HOHh4QgNDYVKpcK6detQWVmJ8+fPo729HW+99Zb2NnxP2gCgHVD49Ky0venYsWPw8fGBTCZ7bh2pVIq9e/fCysoKq1at0q6h0xlD+uXDDz/E+vXr0dzcDLlcjuzsbFRUVMDDwwOrVq3SLhwIABs3bsSuXbuQmpqKe/fuYd68eXj33Xfx22+/9ei4VSoVTp06hVWrVmkXLLx//z4A6D1WkkgkkEql2vgNrfe0119/HXfu3MHFixd7FDezDE44GBugmpubkZKSgrfffhvLli2Dvb09xo0bh6+++gq1tbV6M9r2hFAo1F6tjx49GpmZmWhoaEBWVpZJ9j937lzU19cjNjbWJPszRlNTE65fv669Mu+Kn58f1q9fj8rKSmzcuLHTOt3pF39/fygUCjg5OSEkJARNTU24efMmAKClpQWZmZkICgpCcHAwHBwcsGXLFohEoh6f/x07dkCpVCIpKUlbpnnD5NlHLwAgEonQ3NxsVL2neXt7AwBKS0t7FDezDE44GBugysrK0NjYiEmTJumUT548GWKxWOeRh6lNmjQJMplM5xFBf1VdXQ0i6vLuxtOSkpLg4+ODjIwMnD17Vu/znvaL5k6D5g7H5cuXoVKpMHbsWG0dqVQKZ2fnHp3/3Nxc5OTk4Mcff9S5S6EZw9LZ4M7W1lZIpVKj6j1Nc447u/vB+j5OOBgboDSvGHa2gJ6DgwMaGhrM2r6NjQ1qamrM2kZvaGlpAfDkeAwhkUiQlZUFgUCAlStX6l3Jm7pfNI9utmzZojMvzo0bN7QDPY114MAB7Ny5E4WFhXjttdd0PtOMw6mvr9cpV6lUaGlpgVKpNKre0zRJiOacs/6FEw7GBigHBwcA6PQLrK6uDi4uLmZru62tzext9BbNl6AxE1P5+flhw4YNuHr1KhITE3U+M3W/ODk5AQBSU1P15sYpKioyal8A8OWXX2Lfvn04deoUXn31Vb3P3d3dIZfLcePGDZ1yzXgbX19fo+o9rbW1FQA6vfvB+j5OOBgboMaOHQs7Ozu9gYPnzp1Da2sr3njjDW2ZUCjUGYTYU4WFhSAiTJ061Wxt9JYhQ4ZAIBAYPb9GYmIiRo4ciQsXLuiUG9Mvhhg+fDgkEgmKi4uN2u5ZRITo6GiUlpYiLy+v0zswwJN+nDNnDs6cOaMziLegoAACgUD7Jo+h9Z6mOcdDhw7t0bEwy+CEg7EBSiKRIDIyErm5udi3bx/q6+tRWlqKNWvWQKlUIiwsTFvXy8sLf/zxB/Ly8tDW1oaamhq9K1MAGDRoEO7evYvKyko0NDRoEwi1Wo2HDx+ivb0dJSUliIiIgKurK0JDQ03SRkFBgcVei5XJZPDw8MDt27eN2k7zaOXZQZPG9Iuh7axYsQL79+9HZmYm6uvr0dHRgdu3b+PevXsAgJCQEAwdOrTLqdUvXbqEXbt24euvv4ZIJNJbuuCLL77Q1o2NjUVVVRW2bt2KpqYmFBUVITk5GaGhofDx8TG6nobmHI8bN86oc8D6iF5alZYRL4fMzKs7v19qtZqSk5PJ29ubRCIROTo6UlBQEF2+fFmn3oMHD+jNN98kiURC7u7u9PHHH1NUVBQBIC8vL7p58yYREZ0/f57c3NxIKpXS9OnT6f79+xQWFkYikYiGDRtGQqGQFAoFLVy4kCoqKkzWxvHjx0kul1NSUpJRx2+q5enDw8NJJBKRSqXSluXm5pKnpycBoMGDB9PatWs73TYqKooWLFigU2ZIv2RkZJBMJiMA5O3tTRUVFbR7925SKBQEgNzc3OjKlStERPT48WOKjo4mV1dXEgqF5OTkRMHBwVRWVkZEREFBQQSA4uLinnuMpaWlBOC5P8nJyTr1T58+TVOmTCEbGxtSKpUUFRVFLS0tevs1tB4R0dy5c2nYsGGkVqufG6eh+O9x7+Pl6XsRL4fMzKmv/n6tXr0aBw8exIMHDywdih5TLU9fXl6OUaNGISsrC8uWLTNRdL1HrVZj5syZCA0NxcqVKy0dTqcePHgAFxcXJCUlITIyssf766v/X15m/EiFMWZ2L/tKn15eXkhISEBCQkKnK6f2ZR0dHcjLy0NDQ0OfXtF527ZtmDBhAsLDwy0dCusmTjgYY8wEYmJisHjxYoSEhPSrBdoKCwtx+PBhFBQUGDyXSG9LSUlBcXExjh8/DpFIZOlwWDdxwsEYM5tNmzYhKysLjx49gru7Ow4dOmTpkMxq+/btCA8Px2effWbpUAwWGBiI77//Xmcdm74kPz8fjx8/RmFhIRwdHS0dDusBoaUDYIy9vHbs2IEdO3ZYOoxeNWvWLMyaNcvSYbw0FixYgAULFlg6DGYCfIeDMcYYY2bHCQdjjDHGzI4TDsYYY4yZHSccjDHGGDM7HjTay27fvo2cnBxLh8FeQpppn/n3y3Caxcv4nA08t2/ffikWD+xPeKbRXrR48eKX/rVAxhjrLxYtWsQzjfYiTjgYY0ZZsmQJAL4rwBgzDo/hYIwxxpjZccLBGGOMMbPjhIMxxhhjZscJB2OMMcbMjhMOxhhjjJkdJxyMMcYYMztOOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNlxwsEYY4wxs+OEgzHGGGNmxwkHY4wxxsyOEw7GGGOMmR0nHIwxxhgzO044GGOMMWZ2nHAwxhhjzOw44WCMMcaY2XHCwRhjjDGz44SDMcYYY2bHCQdjjDHGzI4TDsYYY4yZHSccjDHGGDM7TjgYY4wxZnaccDDGGGPM7DjhYIwxxpjZccLBGGOMMbPjhIMxxhhjZscJB2OMMcbMjhMOxhhjjJkdJxyMMcYYMztOOBhjjDFmdpxwMMYYY8zsOOFgjDHGmNkJiIgsHQRjrG/6/vvvsWfPHqjVam3Z9evXAQDu7u7aMisrK/zHf/wH3nvvvV6PkTHWP3DCwRh7rpKSEowfP96guhcvXoSvr6+ZI2KM9VeccDDGujRy5Ehcvny5yzpeXl64evVqL0XEGOuPeAwHY6xLy5cvh0gkeu7nIpEIK1as6MWIGGP9Ed/hYIx16dq1a/Dy8kJXfyquXr0KLy+vXoyKMdbf8B0OxliXPDw8MHHiRAgEAr3PBAIBJk2axMkGY+yFOOFgjL3Q+++/D2tra71ya2trvP/++xaIiDHW3/AjFcbYC1VXV0OpVOq8Hgs8eR327t27GDp0qIUiY4z1F3yHgzH2QkOGDEFAQIDOXQ5ra2vMnDmTkw3GmEE44WCMGWT58uV6A0eXL19uoWgYY/0NP1JhjBmkvr4eTk5OaG1tBfDkddjq6mo4ODhYODLGWH/AdzgYYwZRKBT4y1/+AqFQCKFQiDlz5nCywRgzGCccjDGDLVu2DB0dHejo6OB1UxhjRuFHKowxg7W0tGDw4MEgItTW1kIqlVo6JMZYP8EJRzfl5ORg6dKllg6DMcZYL8rOzsaSJUssHUa/JLR0AP1ddna2pUNgzOyKioqQlpaG7OxsFBcXQyAQGLyK7EC2dOlSREREwM/Pz9KhMBPgi8ye4YSjhzjTZQNFWloalixZgrfffhsAIBTyn48XWbp0Kfz8/PjvxEuCE46e4b8YjDGjcKLBGOsOfkuFMcYYY2bHCQdjjDHGzI4TDsYYY4yZHSccjDHGGDM7TjgYY73m+PHjsLe3xw8//GDpUPq8kydPIiYmBocPH4aHhwcEAgEEAkGnC+bNmjULcrkc1tbWGDNmDM6fP2+BiA2XkJCA0aNHQ6FQwMbGBl5eXvj000/R2NioV/fs2bOYNm0aZDIZlEoloqOj8fjxY6PrHTlyBJ9//jk6OjrMemzs+TjhYIz1Gp5n0DBbt25Feno6Nm3ahODgYFy7dg2enp545ZVXsG/fPhw7dkyn/k8//YSDBw9i3rx5KCsrw8SJEy0UuWFOnTqFtWvXorKyErW1tdixYwfS0tKwePFinXplZWWYNWsWAgMDUVNTg9zcXHz77bdYs2aN0fXmz58PiUSCwMBA1NXV9cpxsmcQ65bs7Gzi08cGipfx912lUpGfn59Z2wBA2dnZRm3z2Wef0YgRI6i5uVmn3NPTk77//nuysrKiYcOGUV1dnc7nBQUFtGDBgh7H3Bvmzp1L7e3tOmVLliwhAHTz5k1t2dKlS8nd3Z3UarW2LDk5mQQCAf3rX/8yuh4RUXh4OPn5+VFbW5vRcXenP9n/4TscjLEBac+ePaiurrZ0GDrKy8sRGxuL+Ph4SCQSvc/9/f0RERGBO3fu4JNPPrFAhKZx9OhRWFtb65QNHjwYAKBSqQAA7e3tOHbsGAICAiAQCLT1Zs+eDSJCfn6+UfU0tm3bhuLiYqSlpZnl2NjzccLBGOsVZ8+ehaurKwQCAf72t78BADIzM2FrawuZTIb8/HzMnj0bCoUCLi4u2L9/v3bb9PR0SCQSDBkyBKtXr4ZSqYREIoG/vz/OnTunrRceHg6xWAxnZ2dt2UcffQRbW1sIBALU1tYCACIiIhAZGYmKigoIBAJ4eXkBAE6cOAGFQoHt27f3xinRk56eDiLC/Pnzn1snKSkJI0aMwDfffIOTJ092uT8iQkpKCkaNGgUbGxs4Ojpi4cKF+P3337V1DO0DAOjo6EBcXBxcXV0hlUrh6+trsuUde1Y+FwAAIABJREFU7ty5A6lUCnd3dwDAtWvX0NjYCFdXV516np6eAICSkhKj6mk4OjoiICAAaWlp/Iivl3HCwRjrFdOnT8fPP/+sU/bhhx9i/fr1aG5uhlwuR3Z2NioqKuDh4YFVq1ahra0NwJNEIjQ0FCqVCuvWrUNlZSXOnz+P9vZ2vPXWW7h16xaAJ1/Yz04jnpGRgfj4eJ2ytLQ0zJs3D56eniAilJeXA4B2QKFarTbLOXiRY8eOwcfHBzKZ7Ll1pFIp9u7dCysrK6xatQpNTU3Prbtt2zbExMRg8+bNqK6uxpkzZ3Dr1i3MmDEDVVVVAAzvAwDYuHEjdu3ahdTUVNy7dw/z5s3Du+++i99++61Hx61SqXDq1CmsWrUKYrEYAHD//n0AgFwu16krkUgglUq18Rta72mvv/467ty5g4sXL/YobmYcTjgYY32Cv78/FAoFnJycEBISgqamJty8eVOnjlAo1F6tjx49GpmZmWhoaEBWVpZJYpg7dy7q6+sRGxtrkv0Zo6mpCdevX9demXfFz88P69evR2VlJTZu3NhpnebmZqSkpODtt9/GsmXLYG9vj3HjxuGrr75CbW0tdu/erbdNV33Q0tKCzMxMBAUFITg4GA4ODtiyZQtEIlGPz/+OHTugVCqRlJSkLdO8YfLsoxcAEIlEaG5uNqre07y9vQEApaWlPYqbGYcTDsZYn6O5yn366rozkyZNgkwm03lE0F9VV1eDiLq8u/G0pKQk+Pj4ICMjA2fPntX7vKysDI2NjZg0aZJO+eTJkyEWi3UeRXXm2T64fPkyVCoVxo4dq60jlUrh7Ozco/Ofm5uLnJwc/Pjjjzp3KTRjWNrb2/W2aW1thVQqNare0zTnuLO7H8x8OOFgjPVrNjY2qKmpsXQYPdbS0gLgyfEYQiKRICsrCwKBACtXrtS7kte8+mlnZ6e3rYODAxoaGoyKT/PoZsuWLdo5QQQCAW7cuKEd6GmsAwcOYOfOnSgsLMRrr72m85lmHE59fb1OuUqlQktLC5RKpVH1nqZJQjTnnPUOTjgYY/1WW1sb6urq4OLiYulQekzzJWjMxFR+fn7YsGEDrl69isTERJ3PHBwcAKDTxKI758zJyQkAkJqaCiLS+SkqKjJqXwDw5ZdfYt++fTh16hReffVVvc/d3d0hl8tx48YNnXLNeBtfX1+j6j2ttbUVADq9+8HMhxMOxli/VVhYCCLC1KlTtWVCofCFj2L6oiFDhkAgEODRo0dGbZeYmIiRI0fiwoULOuVjx46FnZ2d3oDOc+fOobW1FW+88YZR7QwfPhwSiQTFxcVGbfcsIkJ0dDRKS0uRl5fX6R0Y4Ek/zpkzB2fOnNEZxFtQUACBQKB9k8fQek/TnOOhQ4f26FiYcTjhYIz1G2q1Gg8fPkR7eztKSkoQEREBV1dXhIaGaut4eXnhjz/+QF5eHtra2lBTU6N39QsAgwYNwt27d1FZWYmGhga0tbWhoKDAYq/FymQyeHh44Pbt20Ztp3m08uygSYlEgsjISOTm5mLfvn2or69HaWkp1qxZA6VSibCwMKPbWbFiBfbv34/MzEzU19ejo6MDt2/fxr179wAAISEhGDp0aJdTq1+6dAm7du3C119/DZFIpPN4RiAQ4IsvvtDWjY2NRVVVFbZu3YqmpiYUFRUhOTkZoaGh8PHxMbqehuYcjxs3zqhzwHrIItONvQRexpkXGXseU/y+f/nll+Ts7EwASCaT0fz58ykjI4NkMhkBIG9vb6qoqKDdu3eTQqEgAOTm5kZXrlwhIqKwsDASiUQ0bNgwEgqFpFAoaOHChVRRUaHTzoMHD+jNN98kiURC7u7u9PHHH1NUVBQBIC8vL+1MlufPnyc3NzeSSqU0ffp0un//Ph0/fpzkcjklJSX16Fg1YOTMlOHh4SQSiUilUmnLcnNzydPTkwDQ4MGDae3atZ1uGxUVpTfTqFqtpuTkZPL29iaRSESOjo4UFBREly9f1tYxpg8eP35M0dHR5OrqSkKhkJycnCg4OJjKysqIiCgoKIgAUFxc3HOPsbS0lAA89yc5OVmn/unTp2nKlClkY2NDSqWSoqKiqKWlRW+/htYjejLT6bBhw3RmJjWEsf3JdPE3ZjdxwsEGkr7w+x4WFkaDBg2yaAzGMvYL6urVqyQUCum7774zY1Tm09HRQTNmzKA9e/ZYOpTnqq2tJYlEQl988YXR23LC0TP8SIUx1m+87Ct9enl5ISEhAQkJCZ2unNqXdXR0IC8vDw0NDQgJCbF0OM+1bds2TJgwAeHh4ZYOZcDhhMOCPvjgA8jlcggEgh4PxHpZqNVqpKamwt/f/7l1DF2u+kWeXfZb8yMWizFkyBDMnDkTycnJePjwYU8OiTGjxMTEYPHixQgJCTF6AKklFRYW4vDhwygoKDB4LpHelpKSguLiYhw/fhwikcjS4Qw4nHBY0DfffIOvv/7a0mH0GVevXsX/+3//Dxs2bHjue/2GLldtiKeX/ba3twcRQa1Wo7q6Gjk5OXB3d0d0dDTGjBnT46mbWc9s2rQJWVlZePToEdzd3XHo0CFLh2RW27dvR3h4OD777DNLh2KwwMBAfP/99zrr2PQl+fn5ePz4MQoLC+Ho6GjpcAYkTjiYyTQ3N3d5Z6IrFy9exMaNG7FmzRpMmDDhufUSExPh7OyM+Ph42Nraws/PD9HR0di7d69JZpsUCARwcHDAzJkzkZWVhZycHFRVVWHu3Ln96mrzeXrSR5a0Y8cOPH78GESE69evY9GiRZYOyexmzZqFnTt3WjqMl8aCBQsQExPT6RTorHdwwmFhTy+n3N/1ZLnv8ePH4/Dhw3jvvfeeO9OisctQm8KiRYsQGhqK6upqfPXVVybff2/ri0uyM8YGBk44ehERITk5GT4+PrCxsYG9vT2ioqJ06uzatQsymQxyuRzV1dWIjIzEsGHDcPnyZYOWmjZ0GW9NPC/aX0+X+zYlY5ahNuUy45o5HgoKCgBwHzHGWLdY7gWZ/q07rwlu3ryZBAIB/ed//ic9fPiQVCoVZWRkEAC6cOGCTj0AtG7dOvryyy/p7bffpn/9618UFxdHYrGYvvvuO6qrq6OSkhKaOHEiDR48mO7fv6/dPiwsjGxtbenSpUvU0tJCZWVlNHnyZJLL5do5CIjI4P299957NHToUJ1jSU5OJgBUU1OjLQsODiZPT0+jzkln/vSnP9H48eP1yk+fPt3pe/pERFKplAIDA7X/Pnr0KMnlckpISHhhe56enmRvb//cz+vr6wkADR8+XFs20PqoL7wW2x+BX6N8qXB/9gzf4eglzc3NSE1Nxb/9279hw4YNcHBwgFQqxaBBg567zc6dO7F27VocPnwYbm5uRi01/aJlvLuzdLWlGbMMtSmXGde8SdTZmhTcR4wxZhihpQMYKMrLy6FSqRAYGNit7Xu61PSzy3j3dH+W0J1lqE2hqakJRASFQtFlvYHQRzk5Ob3eZn/XnYXNGHsZccLRSzRz92tWXDSWKZaafnoZb1MvXd0burMMtSlcuXIFADBy5Mgu6w2EPlq6dGmvt9nfpaWlIS0tzdJhMGZxnHD0Es3VeXcmqAJ6vtT0s8t4m3rp6t7QnWWoTeHEiRMAnrwN05WB0EdE1Ott9mcCgQDZ2dlYsmSJpUNhJvAyvVVoCTyGo5eMHTsWVlZWOH36dLe378lS088u423M/vrKct/dWYa6p+7fv4/U1FS4uLhg5cqVXdblPmKMsefjhKOXODk5ITg4GIcOHcKePXtQX1+PkpISgwf+GbvU9IuW8TZmfz1Z7tvUDF2G2thlxokIjY2NUKvVICLU1NQgOzsb06ZNg7W1NfLy8l44hoP7iDHGumDJV2T6s+68JtjQ0EAffPABvfLKK2RnZ0fTp0+nuLg4AkAuLi508eJF+vzzz0kqlWpfw3x61UhDlpomMnwZb0P315Plvg1VVFRE06ZNI6VSqV2m2tnZmfz9/en06dM6dQ1ZhtqQZcaPHDlCvr6+JJPJSCwWk5WVFQEggUBADg4ONGXKFEpISKAHDx7obDcQ+4hfi+0e8GuULxXuz54REPFD2e7IycnB0qVL++Qz7dWrV+PgwYN48OCBpUNhz9Hf+qgv/773ZTyG4+XC/dkz/EjlJfWyL+P9MuA+YowNJJxwMLP4/fff9ZZ97+wnJCTE0qEyxhjrBZxwvGT6yjLeI0eOBBG98OfAgQMWic+S+kofsb7t5MmTiImJweHDh+Hh4aFN0pcvX65Xd9asWZDL5bC2tsaYMWNw/vx5C0RsuISEBIwePRoKhQI2Njbw8vLCp59+isbGRr26Z8+exbRp0yCTyaBUKhEdHd3p9AIvqnfkyBF8/vnnfGfRkiwycuQlwIPo2EDCv+/dg24OMoyLi6N58+ZRfX29tszT05NeeeUVAkBHjx7V26agoIAWLFjQo3h7S0BAAGVkZNCDBw+ovr6esrOzSSQS0V/+8hedev/85z9JKpVSbGwsNTY20s8//0yDBw+mFStWdKteWloaBQQE0MOHD7sVd3f7kz3BdzgYY/1Cc3Mz/P39+30bL7Jz504cOHAAOTk5kMvlOp+lp6fDysoKYWFhePTokYUi7Dk7OzuEhYVh0KBBkMvlWLJkCYKCgnDixAncunVLWy8xMRHOzs6Ij4+Hra0t/Pz8EB0djb179+qsmGxovXXr1mH8+PGYM2dOp0skMPPihIMx1i/s2bMH1dXV/b6NrpSXlyM2Nhbx8fHa2Ymf5u/vj4iICNy5cweffPKJBSI0jaNHj+otwjh48GAAT5YqAJ6smXTs2DEEBATozPA5e/ZsEBHy8/ONqqexbds2FBcX83TzFsAJB2PMLIgIKSkp2hVxHR0dsXDhQp0rzvDwcIjFYu06OQDw0UcfwdbWFgKBALW1tQCAiIgIREZGoqKiAgKBAF5eXkhPT4dEIsGQIUOwevVqKJVKSCQS+Pv76yxs15M2gCdT2xsziVxPpKeng4i6nDU3KSkJI0aMwDfffIOTJ092uT9D+iAzMxO2traQyWTIz8/H7NmzoVAo4OLigv379+vsr6OjA3FxcXB1dYVUKoWvry+ys7N7dtD/vzt37kAqlcLd3R0AcO3aNTQ2NsLV1VWnnqenJwCgpKTEqHoajo6OCAgIQFpaGr/m3dss+TynP+Nn2mwg6c7ve1xcHInFYvruu++orq6OSkpKaOLEiTR48GCdCcfee+89Gjp0qM62ycnJBIBqamq0ZcHBweTp6alTLywsjGxtbenSpUvU0tJCZWVlNHnyZJLL5drJznraxtGjR0kul1NCQoJRx09k/DN/Dw8PGj16dKefeXp60vXr14mI6OeffyYrKyt67bXXqLGxkYg6H8NhaB9s3ryZANB///d/06NHj6i6uppmzJhBtra21Nraqq33ySefkI2NDR06dIgePnxImzZtIisrK/r1118NPsbONDU1kVwup/DwcG3Z6dOnCQAlJyfr1ZdKpRQYGGhUvafFxMQQALpw4YJRcRrbn0wX3+FgjJlcc3MzUlJS8Pbbb2PZsmWwt7fHuHHj8NVXX6G2ttbgKf0NIRQKtVfwo0ePRmZmJhoaGpCVlWWS/c+dOxf19fWIjY01yf6ep6mpCdevX9demXfFz88P69evR2VlJTZu3Nhpne70gb+/PxQKBZycnBASEoKmpibcvHkTANDS0oLMzEwEBQUhODgYDg4O2LJlC0QiUY/P9Y4dO6BUKpGUlKQt07xh8uyjFwAQiURobm42qt7TvL29AQClpaU9ipsZhxMOxpjJlZWVobGxEZMmTdIpnzx5MsRisc4jD1ObNGkSZDKZzmOD/qC6uhpEBJlMZlD9pKQk+Pj4ICMjA2fPntX7vKd9IBaLAUC73s7ly5ehUqkwduxYbR2pVApnZ+cenevc3Fzk5OTgxx9/1BkkqxnD0tngztbWVkilUqPqPU1zjquqqrodNzMeJxyMMZOrq6sD8ORthGc5ODigoaHBrO3b2NigpqbGrG2YWktLC4AnsRtCIpEgKysLAoEAK1eu1LuSN3UfNDU1AQC2bNmiM3nfjRs3tAM9jXXgwAHs3LkThYWFeO2113Q+04y5qa+v1ylXqVRoaWmBUqk0qt7TNEmI5pyz3sEJB2PM5BwcHACg0y+1uro6uLi4mK3ttrY2s7dhDpovQWMmpvLz88OGDRtw9epVJCYm6nxm6j5wcnICAKSmpupN4FdUVGTUvgDgyy+/xL59+3Dq1Cm8+uqrep+7u7tDLpfrrXpcXl4OAPD19TWq3tNaW1sBoNO7H8x8OOFgjJnc2LFjYWdnh99++02n/Ny5c2htbcUbb7yhLRMKhdrb9qZQWFgIIsLUqVPN1oY5DBkyBAKBwOj5NRITEzFy5EhcuHBBp9yYPjDE8OHDIZFIUFxcbNR2zyIiREdHo7S0FHl5eZ3egQGe9NmcOXNw5swZqNVqbXlBQQEEAoH2TR5D6z1Nc46HDh3ao2NhxuGEgzFmchKJBJGRkcjNzcW+/6+9e4+J6sz/B/4emIFhlAGsqFTEclPrrba1rlBd27BlY4kipSqrNkubGmprKWpZ690iYlsMGrsQY+uyiRoLFYO2lbZxDTamxm2jVopbL1RUVARU5H6dz+8Pf8zXKRdnZA6Hwfcr4Z9nPuc8nzlnyHzmnPM8z+7dqKqqQkFBARYtWgQfHx/ExcWZY4OCgnD79m3k5uaiubkZ5eXl7X6tAsCAAQNw/fp1FBcXo7q62lxAmEwm3LlzBy0tLThz5gwSEhLg5+eH2NhYu/SRl5fXI8NiDQYDAgICUFJSYtN2bbdW/vjQpC3nwNp+Xn/9dezduxcZGRmoqqpCa2srSkpKcOPGDQBATEwMBg8e3OXU6mfPnsUnn3yCzz77DDqdrt36Sps3bzbHrlmzBjdv3sS6detQW1uL48ePIzU1FbGxsRg5cqTNcW3ajvG4ceNsOgbUTSqOkHFoHBZLj5KH+bybTCZJTU2V4OBg0el04uXlJVFRUXLu3DmLuFu3bsmLL74oer1e/P395d1335XExEQBIEFBQebhrSdPnpThw4eLm5ubTJkyRUpLSyUuLk50Op0MHTpUtFqtGI1GmTVrlhQVFdmtj0OHDom7u7skJyfbfNxg4zDK+Ph40el0UldXZ27bv3+/BAYGCgAZOHCgLF68uMNtExMT2w2LteYcpKeni8FgEAASHBwsRUVFsmPHDjEajQJAhg8fLufPnxcRkcbGRlm+fLn4+fmJVqsVb29viY6OlsLCQhERiYqKEgCydu3aTt9jQUGBAOj074/DW48ePSqTJk0SV1dX8fHxkcTERGloaGi3X2vjREQiIiJk6NChYjKZOs2zI7aeT7LEb8yHxIKDHiW99fMeFxcnAwYMUDuNTtn6BXXhwgXRarWya9cuBbNSTmtrq0ydOlV27typdiqdqqioEL1eL5s3b7Z5WxYc3cNbKkTk0PrS6p9BQUFISkpCUlJShyun9matra3Izc1FdXU1YmJi1E6nU+vXr8eECRMQHx+vdiqPHBYcRES9yIoVKzB79mzExMQ41AJt+fn5yMnJQV5entVzifS0tLQ0nD59GocOHYJOp1M7nUcOCw4ickgrV65EZmYm7t69C39/f+zbt0/tlOxm48aNiI+Px6ZNm9ROxWphYWHYs2ePxZo1vcmBAwfQ2NiI/Px8eHl5qZ3OI0mrdgJERA8jJSUFKSkpaqehmPDwcISHh6udRp8RGRmJyMhItdN4pPEKBxERESmOBQcREREpjgUHERERKY4FBxERESmOD4120+zZs9VOgUhxbVNB8/Nuuy1btuDLL79UOw0i1WlERNROwhEdP34caWlpaqdB1OPaFgl7+umnVc6EqOctXboUISEhaqfhkFhwEJFN5syZAwDIzs5WORMiciR8hoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFMeCg4iIiBTHgoOIiIgUx4KDiIiIFKdVOwEi6r3q6urQ2Nho0dbU1AQAuHPnjkW7q6srDAZDj+VGRI5FIyKidhJE1DtlZGTgnXfesSo2PT0db7/9tsIZEZGjYsFBRJ0qLy+Hj48PWltbu4xzdnbGjRs34O3t3UOZEZGj4TMcRNQpb29vhIWFwdnZudMYZ2dn/OUvf2GxQURdYsFBRF1asGABuroQKiJYsGBBD2ZERI6It1SIqEvV1dXw9vZu9/BoGxcXF5SXl8NoNPZwZkTkSHiFg4i65O7ujhkzZkCn07V7TavVIjIyksUGET0QCw4ieqD58+ejpaWlXXtrayvmz5+vQkZE5Gh4S4WIHqipqQkDBw5EdXW1RXv//v1RUVEBV1dXlTIjIkfBKxxE9EAuLi6YPXs2XFxczG06nQ5z585lsUFEVmHBQURWmTdvnnmWUQBobm7GvHnzVMyIiBwJb6kQkVVMJhOGDBmC8vJyAMDAgQNRWlra5RwdRERteIWDiKzi5OSEefPmwcXFBTqdDvPnz2exQURWY8FBRFb729/+hqamJt5OISKbcbVYlZWUlODHH39UOw0iq4gIHnvsMQDApUuXUFxcrG5CRFYKDQ2Fr6+v2mk80vgMh8qys7Mxd+5ctdMgIurTsrKyMGfOHLXTeKTxCkcvwbqPlNBW0Nrz83X27FkAwOjRo+22z95Go9HwC6oP0Wg0aqdAYMFBRDbqy4UGESmHD40SERGR4lhwEBERkeJYcBAREZHiWHAQERGR4lhwEBERkeJYcBDRAx06dAgeHh746quv1E6l1zt8+DBWrFiBnJwcBAQEQKPRQKPR4LXXXmsXGx4eDnd3dzg7O2PMmDE4efKkChlbLykpCaNHj4bRaISrqyuCgoLwj3/8AzU1Ne1ijx07hueffx4GgwE+Pj5Yvnw5GhsbbY47ePAgPv74Y7S2tir63kh5LDiI6IE4T4x11q1bh23btmHlypWIjo7G77//jsDAQDz22GPYvXs3vvnmG4v477//Hl9++SVmzJiBwsJCPPPMMyplbp0jR45g8eLFKC4uRkVFBVJSUrB161bMnj3bIq6wsBDh4eEICwtDeXk59u/fj3/9619YtGiRzXEzZ86EXq9HWFgYKisre+R9kkKEVJWVlSU8DaSUvvj5qqurk5CQEEX7ACBZWVk2bbNp0yYZMWKE1NfXW7QHBgbKnj17xMnJSYYOHSqVlZUWr+fl5UlkZGS3c+4JERER0tLSYtE2Z84cASBXrlwxt82dO1f8/f3FZDKZ21JTU0Wj0cj//vc/m+NEROLj4yUkJESam5ttzvthzifZH69wEJFD2blzJ8rKytROw8LFixexZs0afPjhh9Dr9e1eDw0NRUJCAq5du4b3339fhQzt4+uvv263QvDAgQMBAHV1dQCAlpYWfPPNN5g2bZrFDJ/Tp0+HiODAgQM2xbVZv349Tp8+ja1btyry3kh5LDiIqEvHjh2Dn58fNBoN/vnPfwIAMjIy0K9fPxgMBhw4cADTp0+H0WiEr68v9u7da95227Zt0Ov1GDRoEN566y34+PhAr9cjNDQUJ06cMMfFx8fDxcUFQ4YMMbe988476NevHzQaDSoqKgAACQkJWLZsGYqKiqDRaBAUFAQA+Pbbb2E0GrFx48aeOCTtbNu2DSKCmTNndhqTnJyMESNG4PPPP8fhw4e73J+IIC0tDU8++SRcXV3h5eWFWbNm4bfffjPHWHsOAKC1tRVr166Fn58f3NzcMH78eGRlZXXvTf9/165dg5ubG/z9/QEAv//+O2pqauDn52cRFxgYCAA4c+aMTXFtvLy8MG3aNGzdupW3+BwUCw4i6tKUKVParWj89ttvY8mSJaivr4e7uzuysrJQVFSEgIAALFy4EM3NzQDuFRKxsbGoq6vDe++9h+LiYpw8eRItLS146aWXcPXqVQD3vrD/uG5Jeno6PvzwQ4u2rVu3YsaMGQgMDISI4OLFiwBgfqDQZDIpcgwe5JtvvsHIkSNhMBg6jXFzc8O///1vODk5YeHChaitre00dv369VixYgVWrVqFsrIy/PDDD7h69SqmTp2KmzdvArD+HADABx98gE8++QRbtmzBjRs3MGPGDMybNw8///xzt953XV0djhw5goULF8LFxQUAUFpaCgBwd3e3iNXr9XBzczPnb23c/Z5++mlcu3YNv/zyS7fyJnWw4CCibgkNDYXRaIS3tzdiYmJQW1uLK1euWMRotVrzr/XRo0cjIyMD1dXVyMzMtEsOERERqKqqwpo1a+yyP1vU1tbi0qVL5l/mXQkJCcGSJUtQXFyMDz74oMOY+vp6pKWl4ZVXXsGCBQvg4eGBcePGYfv27aioqMCOHTvabdPVOWhoaEBGRgaioqIQHR0NT09PrF69GjqdrtvHPyUlBT4+PkhOTja3tY0w+eOtFwDQ6XSor6+3Ke5+wcHBAICCgoJu5U3qYMFBRHbT9iv3/l/XHZk4cSIMBoPFLQJHVVZWBhHp8urG/ZKTkzFy5Eikp6fj2LFj7V4vLCxETU0NJk6caNH+3HPPwcXFxeJWVEf+eA7OnTuHuro6jB071hzj5uaGIUOGdOv479+/H9nZ2fjuu+8srlK0PcPS0tLSbpumpia4ubnZFHe/tmPc0dUP6v1YcBCRKlxdXVFeXq52Gt3W0NAA4N77sYZer0dmZiY0Gg3eeOONdr/k24Z+9u/fv922np6eqK6utim/tls3q1evNs8JotFocPnyZfODnrb64osv8NFHHyE/Px9PPPGExWsCf3XhAAASeElEQVRtz+FUVVVZtNfV1aGhoQE+Pj42xd2vrQhpO+bkWFhwEFGPa25uRmVlJXx9fdVOpdvavgRtmZgqJCQES5cuxYULF7BhwwaL1zw9PQGgw8LiYY6Zt7c3AGDLli0QEYu/48eP27QvAPj000+xe/duHDlyBI8//ni71/39/eHu7o7Lly9btLc9bzN+/Hib4u7X1NQEAB1e/aDejwUHEfW4/Px8iAgmT55sbtNqtQ+8FdMbDRo0CBqNBnfv3rVpuw0bNmDUqFE4deqURfvYsWPRv3//dg90njhxAk1NTXj22Wdt6mfYsGHQ6/U4ffq0Tdv9kYhg+fLlKCgoQG5ubodXYIB75/Hll1/GDz/8YPEQb15eHjQajXkkj7Vx92s7xoMHD+7WeyF1sOAgIsWZTCbcuXMHLS0tOHPmDBISEuDn54fY2FhzTFBQEG7fvo3c3Fw0NzejvLy83a9fABgwYACuX7+O4uJiVFdXo7m5GXl5eaoNizUYDAgICEBJSYlN27XdWvnjQ5N6vR7Lli3D/v37sXv3blRVVaGgoACLFi2Cj48P4uLibO7n9ddfx969e5GRkYGqqiq0traipKQEN27cAADExMRg8ODBXU6tfvbsWXzyySf47LPPoNPpLG7PaDQabN682Ry7Zs0a3Lx5E+vWrUNtbS2OHz+O1NRUxMbGYuTIkTbHtWk7xuPGjbPpGFAvocp0Y2TWF2eCpN7DHp+vTz/9VIYMGSIAxGAwyMyZMyU9PV0MBoMAkODgYCkqKpIdO3aI0WgUADJ8+HA5f/68iIjExcWJTqeToUOHilarFaPRKLNmzZKioiKLfm7duiUvvvii6PV68ff3l3fffVcSExMFgAQFBZlnsjx58qQMHz5c3NzcZMqUKVJaWiqHDh0Sd3d3SU5O7tZ7bQMbZ6aMj48XnU4ndXV15rb9+/dLYGCgAJCBAwfK4sWLO9w2MTGx3UyjJpNJUlNTJTg4WHQ6nXh5eUlUVJScO3fOHGPLOWhsbJTly5eLn5+faLVa8fb2lujoaCksLBQRkaioKAEga9eu7fQ9FhQUCIBO/1JTUy3ijx49KpMmTRJXV1fx8fGRxMREaWhoaLdfa+NE7s10OnToUIuZSa1h6/kkZfCbTmUsOEhJveHzFRcXJwMGDFA1B1vZ+gV14cIF0Wq1smvXLgWzUk5ra6tMnTpVdu7cqXYqnaqoqBC9Xi+bN2+2eVsWHL0Db6kQkeL6+kqfQUFBSEpKQlJSUocrp/Zmra2tyM3NRXV1NWJiYtROp1Pr16/HhAkTEB8fr3Yq9JBYcPQBb775Jtzd3aHRaLr9YJhabFn2Grj3TMCWLVsQGhr60H3+cfnwtj8XFxcMGjQIL7zwAlJTU3Hnzp2H7oMeHStWrMDs2bMRExNj8wOkasrPz0dOTg7y8vKsnkukp6WlpeH06dM4dOgQdDqd2unQQ2LB0Qd8/vnn+Oyzz9ROo1usXfYaAC5cuIA///nPWLp06UPPIwDAYvlwDw8PiAhMJhPKysqQnZ0Nf39/LF++HGPGjOn2FNCPqpUrVyIzMxN3796Fv78/9u3bp3ZKitq4cSPi4+OxadMmtVOxWlhYGPbs2WOxjk1vcuDAATQ2NiI/Px9eXl5qp0PdoFU7ASLg3iRHcXFx5if258yZg5ycHGRnZ+Pq1asYNmwYAOCXX35BUlISFi1ahNraWrsv4qTRaODp6YkXXngBL7zwAiIiIjB37lxERETg/Pnz8PDwsGt/fV1KSgpSUlLUTqNHhYeHIzw8XO00+ozIyEhERkaqnQbZAa9w9BH3L+/siKxZ9hoAnnrqKeTk5GD+/PlWz+zYHa+++ipiY2NRVlaG7du3K94fEVFfxYLDAYkIUlNTMXLkSLi6usLDwwOJiYnt4rpaktqWpa2PHj2KSZMmwWAwwGg0Yty4cebpiHty2Wtb2HO58ra5IvLy8sxtjn5siYh6GgsOB7RmzRosX74ccXFxuHnzJkpLSztcebKrJamtXdq6trYWM2fOxKuvvorbt2/jwoULGDFihHmK4Z5c9toW9lyufMKECQCA33//3dzmyMeWiEgVKg/LfeTZOk9CXV2dGAwGeemllyza9+7dKwDk1KlTIiJSX18vBoNBYmJiLLZ1dXWVt99+W0REVq1aJQCkvr7eHJOeni4A5OLFiyIi8uuvvwoA+frrr9vlYk0fD2vVqlUyYsQIqaqq6jTmT3/6kzz11FPd6kdEJDAwUDw8PLqM0Wg04unpKSKOdWx7wzwcjgict6FP4fnsHfjQqIO5ePEi6urqEBYW1mXcwy5J/celrQMCAjBo0CAsWLAA7733HmJjY82rQyq97PX3339vsey1WtoeTjUajQAc89h2NNqHurZlyxZ8+eWXaqdB1GfwloqDaVtLoG0FyM7Ya0lqNzc3HDlyBFOmTMHGjRsREBCAmJgY1NfX9/iy12o5f/48AGDUqFEAHPfYEhGpiVc4HIxerwcANDY2dhl3/5LUCQkJ3epzzJgx+Oqrr1BeXo60tDR89NFHGDNmjHlWQnv0Adxb9vq7777DkSNHOl2JUg3ffvstAGD69OkAHPPY8pe6bTQaDZYsWYI5c+aonQrZgaOP4usreIXDwYwdOxZOTk44evRol3H2WpL6+vXrOHv2LIB7X7SbNm3CM888g7Nnz/b4stdqKC0txZYtW+Dr64s33ngDgGMdWyKi3oIFh4Px9vZGdHQ09u3bh507d6KqqgpnzpzBjh07LOKsWZLaGtevX8dbb72F3377DU1NTTh16hQuX76MyZMn260PW5a9tpaty5WLCGpqamAymSAiKC8vR1ZWFp5//nk4OzsjNzfX/AyHIx1bIqJeQ91nVulhRhFUV1fLm2++KY899pj0799fpkyZImvXrhUA4uvrK7/88ouIdL0ktbVLWxcXF0toaKh4eXmJs7OzPP7447Jq1SppaWl5YB/WsmXZ6+PHj8vzzz8vPj4+5teHDBkioaGhcvToUXOcNcuVHzx4UMaPHy8Gg0FcXFzEyclJAJhHpEyaNEmSkpLk1q1b7bZ1lGPLUSoPBxzV0KfwfPYOGhE7zw1NNsnOzsbcuXPtPkU3EcDP18PSaDTIysriMxx9BM9n78BbKkRERKQ4FhykiN9++63dsxgd/bWNxiB61Bw+fBgrVqxATk4OAgICzP8Tr732WrvY8PBwuLu7w9nZGWPGjMHJkydVyNh6ycnJHf6/3z+vzMGDB/Hxxx+bZwWmvo/DYkkRo0aN4mV8ok6sW7cOp06dwp49e+Du7o7o6GgEBQWhsrISu3fvRkxMDCIiIszx33//Pb799lts374dubm5KmZuPzNnzsSlS5cQFhaG3NxceHp6qp0SKYxXOIhIUfX19QgNDXX4Puzlo48+whdffIHs7Ox2M+lu27YNTk5OiIuLw927d1XK0D527doFEbH4+/XXXy1i3nvvPTz11FN4+eWX0dLSolKm1FNYcBCRonbu3ImysjKH78MeLl68iDVr1uDDDz80T+J3v9DQUCQkJODatWt4//33Vciw561fvx6nT5/G1q1b1U6FFMaCg4gsiAjS0tLw5JNPwtXVFV5eXpg1a5bFGi7x8fFwcXHBkCFDzG3vvPMO+vXrB41Gg4qKCgBAQkICli1bhqKiImg0GgQFBWHbtm3Q6/UYNGgQ3nrrLfj4+ECv1yM0NBQnTpywSx/AvRlibZmLpSds27YNIoKZM2d2GpOcnIwRI0bg888/x+HDh7vcnzXnKiMjA/369YPBYMCBAwcwffp0GI1G+Pr6Yu/evRb7a21txdq1a+Hn5wc3NzeMHz8eWVlZ3XvTD+Dl5YVp06Zh69atvA3b16kyGJfMOE8CKelhPl9r164VFxcX2bVrl1RWVsqZM2fkmWeekYEDB0ppaak5bv78+TJ48GCLbVNTUwWAlJeXm9uio6MlMDDQIi4uLk769esnZ8+elYaGBiksLJTnnntO3N3d5cqVK3bp4+uvvxZ3d3dJSkqy6f2LKDdvQ0BAgIwePbrD1wIDA+XSpUsiIvLjjz+Kk5OTPPHEE1JTUyMiInl5eRIZGWmxjbXnqm314v/85z9y9+5dKSsrk6lTp0q/fv2kqanJHPf++++Lq6ur7Nu3T+7cuSMrV64UJycn+emnn2x6nxs2bBBfX1/x9PQUnU4nTzzxhERGRsp///vfDuNXrFhhsdq1vSl1Psk2vMJBRGb19fVIS0vDK6+8ggULFsDDwwPjxo3D9u3bUVFR0W5G2+7QarXmX+ajR49GRkYGqqurkZmZaZf9R0REoKqqCmvWrLHL/rqrtrYWly5dQmBg4ANjQ0JCsGTJEhQXF+ODDz7oMOZhzlVoaCiMRiO8vb0RExOD2tpaXLlyBQDQ0NCAjIwMREVFITo6Gp6enli9ejV0Op3N5+Tvf/87Dh48iKtXr6KmpgZ79+7FlStXMG3aNBQWFraLDw4OBgAUFBTY1A85FhYcRGRWWFiImpoaTJw40aL9ueeeg4uLi8UtD3ubOHEiDAaDxe2AvqSsrAwiAoPBYFV8cnIyRo4cifT0dBw7dqzd6909Vy4uLgCA5uZmAMC5c+dQV1dnMXTVzc0NQ4YMsfmcDBs2DE8//TT69+8PFxcXTJ48GZmZmaivr0d6enq7+LZjcvPmTZv6IcfCgoOIzCorKwGgwwX0PD09UV1drWj/rq6uKC8vV7QPtTQ0NAC49x6todfrkZmZCY1GgzfeeAP19fUWr9v7XNXW1gIAVq9ebTF3xuXLl1FXV2fTvjoybtw4ODs74/z58+1ec3NzA/B/x4j6JhYcRGTWNhdCR19WlZWV8PX1Vazv5uZmxftQU9uXqi0TXYWEhGDp0qW4cOECNmzYYPGavc+Vt7c3AGDLli3thrMeP37cpn11xGQywWQydVhwNTU1Afi/Y0R9EwsOIjIbO3Ys+vfvj59//tmi/cSJE2hqasKzzz5rbtNqtebL8faQn58PEcHkyZMV60NNgwYNgkajsXl+jQ0bNmDUqFE4deqURbst58oaw4YNg16vx+nTp23ariN//etf27X99NNPEBGEhIS0e63tmAwePLjbfVPvxYKDiMz0ej2WLVuG/fv3Y/fu3aiqqkJBQQEWLVoEHx8fxMXFmWODgoJw+/Zt5Obmorm5GeXl5bh8+XK7fQ4YMADXr19HcXExqqurzQWEyWTCnTt30NLSgjNnziAhIQF+fn6IjY21Sx95eXm9aliswWBAQEAASkpKbNqu7daKs7Nzu3Zrz5W1/bz++uvYu3cvMjIyUFVVhdbWVpSUlODGjRsAgJiYGAwePPiBU6tfu3YNX3zxBSorK9Hc3Izjx4/jzTffhJ+fHxYtWtQuvu2YjBs3zqacycGoOEKGhMNiSVkP8/kymUySmpoqwcHBotPpxMvLS6KiouTcuXMWcbdu3ZIXX3xR9Hq9+Pv7y7vvviuJiYkCQIKCgszDW0+ePCnDhw8XNzc3mTJlipSWlkpcXJzodDoZOnSoaLVaMRqNMmvWLCkqKrJbH4cOHRJ3d3dJTk62+bhBoWGU8fHxotPppK6uzty2f/9+CQwMFAAycOBAWbx4cYfbJiYmthsWa825Sk9PF4PBIAAkODhYioqKZMeOHWI0GgWADB8+XM6fPy8iIo2NjbJ8+XLx8/MTrVYr3t7eEh0dLYWFhSIiEhUVJQBk7dq1Xb7PZcuWSWBgoPTr10+0Wq34+vrKwoUL5fr16x3GR0REyNChQ8VkMj34ID4Epc4n2YbfdCpjwUFK6q2fr7i4OBkwYIDaaXRKqS+oCxcuiFarlV27dtl93z2htbVVpk6dKjt37rTbPisqKkSv18vmzZvtts8/YsHRO/CWChGp4lFcJTQoKAhJSUlISkpCTU2N2unYpLW1Fbm5uaiurrbrKs/r16/HhAkTEB8fb7d9Uu/EgoOIqAetWLECs2fPRkxMjEMt0Jafn4+cnBzk5eVZPZfIg6SlpeH06dM4dOgQdDqdXfZJvRcLDiLqUStXrkRmZibu3r0Lf39/7Nu3T+2UetzGjRsRHx+PTZs2qZ2K1cLCwrBnzx6LtW2648CBA2hsbER+fj68vLzssk/q3bRqJ0BEj5aUlBSkpKSonYbqwsPDER4ernYaqomMjERkZKTaaVAP4hUOIiIiUhwLDiIiIlIcCw4iIiJSHAsOIiIiUhwLDiIiIlIcR6n0EhqNRu0UqA/j58t2c+fOxdy5c9VOg6jP0IiIqJ3Eo6ykpAQ//vij2mkQEfVpoaGh8PX1VTuNRxoLDiIiIlIcn+EgIiIixbHgICIiIsWx4CAiIiLFaQF8qXYSRERE1Lf9P1a0Jqn0pKHkAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 74
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eEWredPCFUKY",
        "outputId": "69aa727f-4126-4c60-bb63-c8ca80f6ea6d"
      },
      "source": [
        "# Fitting the data onto model\n",
        "model.fit(trainx_pad, trainy, validation_data=(valx_pad, valy), epochs=30, batch_size=32, verbose=1,callbacks=callbacks_list)\n",
        "# Getting score metrics from our model\n",
        "scores = model.evaluate(testx_pad, testy, verbose=0)\n",
        "# Displays the accuracy of correct sentiment prediction over test data\n",
        "print(\"Accuracy: %.2f%%\" % (scores[1]*100))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/30\n",
            "120/120 [==============================] - 10s 75ms/step - loss: 1.3548 - accuracy: 0.5165 - val_loss: 0.6046 - val_accuracy: 0.7953\n",
            "Epoch 2/30\n",
            "120/120 [==============================] - 9s 74ms/step - loss: 0.5935 - accuracy: 0.8059 - val_loss: 0.5472 - val_accuracy: 0.8212\n",
            "Epoch 3/30\n",
            "120/120 [==============================] - 9s 74ms/step - loss: 0.4021 - accuracy: 0.8781 - val_loss: 0.4951 - val_accuracy: 0.8306\n",
            "Epoch 4/30\n",
            "120/120 [==============================] - 9s 74ms/step - loss: 0.3025 - accuracy: 0.9132 - val_loss: 0.4998 - val_accuracy: 0.8329\n",
            "Epoch 5/30\n",
            "120/120 [==============================] - 9s 74ms/step - loss: 0.2250 - accuracy: 0.9316 - val_loss: 0.5989 - val_accuracy: 0.8376\n",
            "Epoch 00005: early stopping\n",
            "Accuracy: 83.45%\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 609
        },
        "id": "XoMIQjwnGm5E",
        "outputId": "ea4ace2e-4ac5-4de7-e68d-e01d46bd9ca9"
      },
      "source": [
        "\n",
        "# rounded_predictions = model.predict_classes(test_imag)\n",
        "y_pred_ohe = model.predict(testx_pad)  # shape=(n_samples, 12)\n",
        "y_pred_labels = np.argmax(y_pred_ohe, axis=1)  # only necessary if output has one-hot-encoding, shape=(n_samples)\n",
        "y_pred_labels[0:10]\n",
        "\n",
        "# y_pred= np.zeros((y_pred_labels.size, y_pred_labels.max()+1))\n",
        "# y_pred[np.arange(y_pred_labels.size),y_pred_labels] = 1\n",
        "\n",
        "confusion_matrix = metrics.confusion_matrix(y_true=testy_integer, y_pred=y_pred_labels)  # shape=(12, 12)\n",
        "\n",
        "import seaborn as sn\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "df_cm = pd.DataFrame(confusion_matrix, index = [i for i in [\"spor\",\"yazarlar\",\"video\",\"dunya\",\"turkiye\"]],\n",
        "                  columns = [i for i in [\"spor\",\"yazarlar\",\"video\",\"dunya\",\"turkiye\"]])\n",
        "plt.figure(figsize = (10,10))\n",
        "sn.heatmap(df_cm, annot=True)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f9d42de7eb8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 76
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0ldeG39KGn7F",
        "outputId": "008cc0f2-dc92-4c74-fe5a-344efe6128f2"
      },
      "source": [
        "target_names = [\"spor\",\"yazarlar\",\"video\",\"dunya\",\"turkiye\"]\n",
        "print(classification_report(y_pred_labels,testy_integer, target_names=target_names))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "        spor       0.94      0.90      0.92       401\n",
            "    yazarlar       0.97      0.82      0.89       461\n",
            "       video       0.77      0.83      0.80       381\n",
            "       dunya       0.88      0.81      0.84       425\n",
            "     turkiye       0.63      0.80      0.71       332\n",
            "\n",
            "    accuracy                           0.83      2000\n",
            "   macro avg       0.84      0.83      0.83      2000\n",
            "weighted avg       0.85      0.83      0.84      2000\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}