Data_Preprocessing_and_Model_Training.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9nLz4tLqaeLa",
        "outputId": "0b3c9779-9624-4d5a-f4f1-8089daa19d6f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.9.2\n",
            "Done\n"
          ]
        }
      ],
      "source": [
        "# Imports\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras \n",
        "from keras.models import Model\n",
        "from keras.layers import Conv2D, Dense, Dropout, Input, Flatten, Conv1D, MaxPool1D\n",
        "\n",
        "from sklearn.model_selection import train_test_split \n",
        "from sklearn.metrics import f1_score\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sea\n",
        "import os\n",
        "\n",
        "# For audio analysis\n",
        "import librosa \n",
        "import librosa.display\n",
        "\n",
        "# Progress bar\n",
        "from tqdm import tqdm \n",
        "\n",
        "# For Audio and Spectrogram Augmentation\n",
        "\n",
        "# !pip install audiomentations \n",
        "from audiomentations import Compose, AddGaussianNoise, TimeStretch, \\\n",
        "                            PitchShift, Shift, AddShortNoises, TimeMask, AddGaussianSNR, TanhDistortion\n",
        "import soundfile as sf\n",
        "\n",
        "import random\n",
        "\n",
        "import json\n",
        "\n",
        "plt.style.use('seaborn')\n",
        "print(tf.__version__)\n",
        "print(\"Done\")"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load the folder with the audio\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "PATH = '/content/drive/Shareddrives/Megathon/WAV/new_converted/'\n",
        "labels = os.listdir(PATH)\n",
        "\n",
        "print(\"Done\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_bS5eQDFaia7",
        "outputId": "b6f2088b-1a85-4f47-fcb2-4d2f7e06b657"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n",
            "Done\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# We first converted all mp3 audio files to .wav format (faster processing and efficient use of libraries)\n",
        "\n",
        "# Script made to convert mp3 -> wav\n",
        "\n",
        "'''\n",
        "filenames = []\n",
        "outputnames = []\n",
        "\n",
        "for label in tqdm(labels[1:]):\n",
        "    i = 0\n",
        "    for audio_file in os.listdir(PATH + label):\n",
        "\n",
        "        filename = PATH + label + \"/\" + audio_file\n",
        "        name = audio_file.split(\".\")\n",
        "        outputname = \"/content/drive/Shareddrives/Megathon/WAV/new_converted/\" + label + \"/\" + name[0] + \".wav\"\n",
        "        sound = AudioSegment.from_mp3(filename)\n",
        "        sound.export(outputname, format=\"wav\")\n",
        "        print(outputname)\n",
        "        i = i + 1\n",
        "'''"
      ],
      "metadata": {
        "id": "_JOvyFPHaide"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Extract top k frequent audio classes (because rest have number of datapoints < 10)\n",
        "# Note that we can extend our approach to other labels as well, given more time and resources\n",
        "\n",
        "filenames = []\n",
        "outcome = []\n",
        "\n",
        "for label in labels:\n",
        "    for audio_file in os.listdir(PATH + label):\n",
        "        filenames.append(PATH + label + \"/\" + audio_file)\n",
        "        outcome.append(label)\n",
        "\n",
        "df = pd.DataFrame({\n",
        "    'filename' : filenames,\n",
        "    'outcome' : outcome\n",
        "})\n",
        "\n",
        "df_vc = df['outcome'].value_counts().reset_index()[:30]\n",
        "df_vc.columns = ['name', 'count']\n",
        "df_vc[:5]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "XkIUbtJ-aigZ",
        "outputId": "c2238e63-5556-459f-9f36-e34933226031"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "       name  count\n",
              "0   english    492\n",
              "1   spanish    129\n",
              "2    arabic     81\n",
              "3  mandarin     52\n",
              "4    french     50"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8e004aad-5b46-4d60-b05c-3805be4c46d6\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>name</th>\n",
              "      <th>count</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>english</td>\n",
              "      <td>492</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>spanish</td>\n",
              "      <td>129</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>arabic</td>\n",
              "      <td>81</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>mandarin</td>\n",
              "      <td>52</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>french</td>\n",
              "      <td>50</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8e004aad-5b46-4d60-b05c-3805be4c46d6')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-8e004aad-5b46-4d60-b05c-3805be4c46d6 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-8e004aad-5b46-4d60-b05c-3805be4c46d6');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Oversampling\n",
        "# Increasing the sample space by splitting the audio files into frames\n",
        "# Each Audio sample's time series data is stored as each element in frames\n",
        "\n",
        "frames = []\n",
        "new_labels = []\n",
        "\n",
        "for label in tqdm(df_vc['name']):\n",
        "    df_temp = df[df['outcome'] == label]\n",
        "    count = df_vc[df_vc['name'] == label]['count'].values[0]\n",
        "\n",
        "    if 50 >= count >= 10:\n",
        "        df1, df2 = train_test_split(df_temp, train_size = 10)\n",
        "        for name in df1['filename']:\n",
        "            sig, sr = librosa.load(name, res_type = 'kaiser_fast')\n",
        "            frame = librosa.util.frame(sig, frame_length = 10 * sr, hop_length = 5 * sr, axis = 0)\n",
        "            frames.extend(frame)\n",
        "            new_labels.extend([label] * len(frame))\n",
        "\n",
        "        for name in df2['filename']:\n",
        "            sig, sr = librosa.load(name, res_type = 'kaiser_fast', duration = 10)\n",
        "            frames.extend([sig])\n",
        "            new_labels.append(label)\n",
        "        \n",
        "        # print(len(frames), len(new_labels))\n",
        "\n",
        "    elif count < 10:\n",
        "        for name in df_temp['filename']:\n",
        "            sig, sr = librosa.load(name, res_type = 'kaiser_fast')\n",
        "            frame = librosa.util.frame(sig, frame_length = 10 * sr, hop_length = 5 * sr, axis = 0)\n",
        "            frames.extend(frame)\n",
        "            new_labels.extend([label] * len(frame))\n",
        "    else:\n",
        "        for name in df_temp.sample(n = min(count, 100))['filename']:\n",
        "            sig, sr = librosa.load(name, res_type = 'kaiser_fast', duration = 10)\n",
        "            frames.extend([sig])\n",
        "            new_labels.append(label)\n",
        "\n",
        "len(frames), len(new_labels)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O37ByOP_aijA",
        "outputId": "6801fbe3-0472-483b-96cf-054cfb4b86f1"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 30/30 [17:36<00:00, 35.22s/it]\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1932, 1932)"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Storing the audio files grouped by their label\n",
        "\n",
        "# key = label, value = frames_of_label\n",
        "frame_dict = {}\n",
        "\n",
        "for i in range(len(frames)):\n",
        "    if new_labels[i] not in frame_dict:\n",
        "        frame_dict[new_labels[i]] = [frames[i]]\n",
        "    else:\n",
        "        frame_dict[new_labels[i]].append(frames[i])\n",
        "\n",
        "print(\"Done\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NQnflHnXailt",
        "outputId": "b30220e0-2cc2-448d-e663-4f9cdc303ec7"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Done\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Audio File Augmentation\n",
        "# To make sure that all labels have 100 datapoints, we're synthesizing data\n",
        "# using Audio Augmentation (by adding noise, time shifting and more)\n",
        "\n",
        "def synthesize_data_audio(frames_of_label, label):\n",
        "    \"\"\"\n",
        "    Synthesizes audio files with respect to one label (one language)\n",
        "    \"\"\"\n",
        "    # df2 = df[df['outcome'] == label]\n",
        "\n",
        "    if len(frames_of_label) > 100:\n",
        "        synth_frames = frames_of_label[:100]\n",
        "        synth_labels = [label] * len(frames_of_label)\n",
        "        return synth_frames, synth_labels\n",
        "\n",
        "    augment = Compose([\n",
        "        AddGaussianNoise(min_amplitude=0.000001, max_amplitude=0.00005, p=0.5),\n",
        "        TimeStretch(min_rate=0.97, max_rate=1.05, p=0.5),\n",
        "        Shift(min_fraction=-0.1, max_fraction=0.1, p=0.3),\n",
        "        TimeMask(min_band_part=0.0005, max_band_part=0.001, fade=True, p=0.3),\n",
        "    ])\n",
        "\n",
        "    synth_frames = list(frames_of_label)\n",
        "    synth_labels = [label] * len(frames_of_label)\n",
        "\n",
        "    # number of frames to synthesize\n",
        "    num_synth = 100\n",
        "\n",
        "    # sampling rate (at what rate is it sampling)\n",
        "    sr = 22050\n",
        "\n",
        "    for i in range(num_synth - len(frames_of_label)):\n",
        "        # randomly choose an audio sample\n",
        "        frame = random.choice(frames_of_label)\n",
        "        augmented_signal = augment(frame, sr)\n",
        "        synth_frames.append(augmented_signal)\n",
        "        synth_labels.append(label)\n",
        "\n",
        "    return synth_frames, synth_labels\n",
        "\n",
        "# synthesize_data_audio(df2, 'hindi')\n",
        "\n",
        "augmented_frames = []\n",
        "augmented_labels = []\n",
        "\n",
        "top30 = df['outcome'].value_counts()[:30].index.tolist()\n",
        "\n",
        "for label in tqdm(top30):\n",
        "    \n",
        "    synth_frames, synth_labels = synthesize_data_audio(frame_dict[label], label)\n",
        "    augmented_frames.extend(synth_frames)\n",
        "    augmented_labels.extend(synth_labels)\n",
        "\n",
        "print(\"aug frames -\")\n",
        "print(augmented_frames[:5])\n",
        "print(\"aug labels - \")\n",
        "print(augmented_labels[:5])\n",
        "print(\"\\n--- Done ---\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SIrLMJH9aioM",
        "outputId": "b54d7283-8eb9-4671-c607-b08c109ad4fc"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 30/30 [01:11<00:00,  2.40s/it]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "aug frames -\n",
            "[array([-0.00050128,  0.00120264,  0.00398494, ...,  0.06498273,\n",
            "        0.07311676,  0.085912  ], dtype=float32), array([-2.5907140e-05, -2.5474981e-05,  1.1168807e-05, ...,\n",
            "        1.1118318e-02,  3.7805722e-03,  6.4079682e-03], dtype=float32), array([-0.00014149, -0.00020588, -0.00012798, ...,  0.0003859 ,\n",
            "        0.00038687,  0.00027097], dtype=float32), array([-0.00129746, -0.00168986, -0.00206407, ..., -0.01713795,\n",
            "       -0.01651185, -0.05745077], dtype=float32), array([-0.00032306, -0.00046929,  0.00068082, ..., -0.01146227,\n",
            "        0.01050493,  0.01390033], dtype=float32)]\n",
            "aug labels - \n",
            "['english', 'english', 'english', 'english', 'english']\n",
            "\n",
            "--- Done ---\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Feature Extraction and Model Compilation\n",
        "# Feature Extraction - Each audio file's time series data is converted to \n",
        "# Model Compilation  - CNN's layers are defined\n",
        "\n",
        "def extract_feature(data):\n",
        "    mfccs_features = librosa.feature.mfcc(y=data, sr=sr, n_mfcc=50)\n",
        "    mfccs_scaled_features = np.mean(mfccs_features.T,axis=0)\n",
        "\n",
        "    return mfccs_scaled_features\n",
        "\n",
        "def cnn_model():\n",
        "    inp = keras.layers.Input(shape = (50, 1))\n",
        "    x = Conv1D(128, 2, activation = 'relu', kernel_regularizer = keras.regularizers.L2(1e-4))(inp)\n",
        "    x = MaxPool1D(2)(x)\n",
        "    x = Dropout(0.2)(x)\n",
        "    \n",
        "    x = Conv1D(256, 2, activation = 'relu', kernel_regularizer = keras.regularizers.L2(1e-4))(x)\n",
        "    x = MaxPool1D(2)(x)\n",
        "    x = Dropout(0.2)(x)\n",
        "    \n",
        "    x = Conv1D(128, 2, activation = 'relu', kernel_regularizer = keras.regularizers.L2(1e-4))(x)\n",
        "    x = MaxPool1D(2)(x)\n",
        "    x = Dropout(0.2)(x)\n",
        "    \n",
        "    x = Flatten()(x)\n",
        "\n",
        "    x = Dense(256, activation = 'relu', kernel_regularizer = keras.regularizers.L2(1e-4))(x)\n",
        "    x = Dropout(0.2)(x)\n",
        "    out = Dense(30, activation = 'softmax')(x)\n",
        "\n",
        "    model = keras.models.Model(inputs = inp, outputs = out)\n",
        "    return model\n",
        "\n",
        "augmented_frames = np.array(augmented_frames)\n",
        "augmented_labels = np.array(augmented_labels)\n",
        "\n",
        "features = []\n",
        "for filename in tqdm(augmented_frames):\n",
        "    features.append(extract_feature(filename))\n",
        "print(\"Feature Extraction Complete\")\n",
        "\n",
        "features = np.array(features)\n",
        "features = np.reshape(features, (-1, 50, 1))\n",
        "\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "le = LabelEncoder()\n",
        "y = le.fit_transform(augmented_labels)\n",
        "y = keras.utils.to_categorical(y)\n",
        "print(\"Label Extraction Complete\")\n",
        "\n",
        "model = cnn_model()\n",
        "\n",
        "model.compile(\n",
        "    loss = keras.losses.CategoricalCrossentropy(),\n",
        "    optimizer = keras.optimizers.Adam(),\n",
        "    metrics = ['accuracy']\n",
        ")\n",
        "\n",
        "print(\"Model Compliled\")\n",
        "\n",
        "print(model.summary())\n",
        "\n",
        "train_x, test_x, train_y, test_y = train_test_split(features, y, train_size = 0.9)\n",
        "print(\"Shapes: \", train_x.shape, train_y.shape, test_x.shape, test_y.shape)\n",
        "\n",
        "print(\"Model Fit -\")\n",
        "model.fit(train_x, train_y, epochs = 100, batch_size = 128, validation_data = [test_x, test_y])\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9Gnl1c7Yaiqq",
        "outputId": "c3183a07-4f02-4b58-95ad-5f60635fcdfb"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 3000/3000 [01:46<00:00, 28.29it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Feature Extraction Complete\n",
            "Label Extraction Complete\n",
            "Model Compliled\n",
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 50, 1)]           0         \n",
            "                                                                 \n",
            " conv1d (Conv1D)             (None, 49, 128)           384       \n",
            "                                                                 \n",
            " max_pooling1d (MaxPooling1D  (None, 24, 128)          0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 24, 128)           0         \n",
            "                                                                 \n",
            " conv1d_1 (Conv1D)           (None, 23, 256)           65792     \n",
            "                                                                 \n",
            " max_pooling1d_1 (MaxPooling  (None, 11, 256)          0         \n",
            " 1D)                                                             \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 11, 256)           0         \n",
            "                                                                 \n",
            " conv1d_2 (Conv1D)           (None, 10, 128)           65664     \n",
            "                                                                 \n",
            " max_pooling1d_2 (MaxPooling  (None, 5, 128)           0         \n",
            " 1D)                                                             \n",
            "                                                                 \n",
            " dropout_2 (Dropout)         (None, 5, 128)            0         \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 640)               0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 256)               164096    \n",
            "                                                                 \n",
            " dropout_3 (Dropout)         (None, 256)               0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 30)                7710      \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 303,646\n",
            "Trainable params: 303,646\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "None\n",
            "Shapes:  (2700, 50, 1) (2700, 30) (300, 50, 1) (300, 30)\n",
            "Model Fit -\n",
            "Epoch 1/100\n",
            "22/22 [==============================] - 3s 91ms/step - loss: 4.4125 - accuracy: 0.0348 - val_loss: 3.4648 - val_accuracy: 0.0333\n",
            "Epoch 2/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 3.4731 - accuracy: 0.0496 - val_loss: 3.4512 - val_accuracy: 0.0467\n",
            "Epoch 3/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 3.4436 - accuracy: 0.0570 - val_loss: 3.4130 - val_accuracy: 0.0633\n",
            "Epoch 4/100\n",
            "22/22 [==============================] - 2s 77ms/step - loss: 3.3968 - accuracy: 0.0763 - val_loss: 3.3416 - val_accuracy: 0.1033\n",
            "Epoch 5/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 3.3409 - accuracy: 0.0922 - val_loss: 3.2782 - val_accuracy: 0.1300\n",
            "Epoch 6/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 3.2816 - accuracy: 0.1141 - val_loss: 3.1796 - val_accuracy: 0.1667\n",
            "Epoch 7/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 3.2170 - accuracy: 0.1244 - val_loss: 3.1007 - val_accuracy: 0.1867\n",
            "Epoch 8/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 3.1022 - accuracy: 0.1656 - val_loss: 2.9831 - val_accuracy: 0.1900\n",
            "Epoch 9/100\n",
            "22/22 [==============================] - 2s 77ms/step - loss: 3.0248 - accuracy: 0.1881 - val_loss: 2.8860 - val_accuracy: 0.2433\n",
            "Epoch 10/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 2.9252 - accuracy: 0.2170 - val_loss: 2.8128 - val_accuracy: 0.2600\n",
            "Epoch 11/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 2.8314 - accuracy: 0.2407 - val_loss: 2.7014 - val_accuracy: 0.2867\n",
            "Epoch 12/100\n",
            "22/22 [==============================] - 2s 111ms/step - loss: 2.7481 - accuracy: 0.2774 - val_loss: 2.6059 - val_accuracy: 0.3200\n",
            "Epoch 13/100\n",
            "22/22 [==============================] - 3s 154ms/step - loss: 2.6357 - accuracy: 0.2993 - val_loss: 2.5294 - val_accuracy: 0.3700\n",
            "Epoch 14/100\n",
            "22/22 [==============================] - 3s 142ms/step - loss: 2.5723 - accuracy: 0.3170 - val_loss: 2.4072 - val_accuracy: 0.4067\n",
            "Epoch 15/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 2.4624 - accuracy: 0.3530 - val_loss: 2.3050 - val_accuracy: 0.4433\n",
            "Epoch 16/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 2.3992 - accuracy: 0.3663 - val_loss: 2.2393 - val_accuracy: 0.4833\n",
            "Epoch 17/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 2.2351 - accuracy: 0.4119 - val_loss: 2.0965 - val_accuracy: 0.5267\n",
            "Epoch 18/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 2.1958 - accuracy: 0.4230 - val_loss: 2.0300 - val_accuracy: 0.5167\n",
            "Epoch 19/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 2.0634 - accuracy: 0.4544 - val_loss: 1.9299 - val_accuracy: 0.5633\n",
            "Epoch 20/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 1.9627 - accuracy: 0.4852 - val_loss: 1.8257 - val_accuracy: 0.5867\n",
            "Epoch 21/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 1.8816 - accuracy: 0.5089 - val_loss: 1.6975 - val_accuracy: 0.6133\n",
            "Epoch 22/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 1.7703 - accuracy: 0.5319 - val_loss: 1.6708 - val_accuracy: 0.6467\n",
            "Epoch 23/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 1.7190 - accuracy: 0.5430 - val_loss: 1.5494 - val_accuracy: 0.6567\n",
            "Epoch 24/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 1.6376 - accuracy: 0.5663 - val_loss: 1.5046 - val_accuracy: 0.6567\n",
            "Epoch 25/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 1.5550 - accuracy: 0.5878 - val_loss: 1.4500 - val_accuracy: 0.6667\n",
            "Epoch 26/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 1.4905 - accuracy: 0.6174 - val_loss: 1.4100 - val_accuracy: 0.6933\n",
            "Epoch 27/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 1.3960 - accuracy: 0.6393 - val_loss: 1.3372 - val_accuracy: 0.7100\n",
            "Epoch 28/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 1.3575 - accuracy: 0.6341 - val_loss: 1.2568 - val_accuracy: 0.7367\n",
            "Epoch 29/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 1.2732 - accuracy: 0.6600 - val_loss: 1.2388 - val_accuracy: 0.7300\n",
            "Epoch 30/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 1.1967 - accuracy: 0.6737 - val_loss: 1.2278 - val_accuracy: 0.7333\n",
            "Epoch 31/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 1.1283 - accuracy: 0.7037 - val_loss: 1.2165 - val_accuracy: 0.7333\n",
            "Epoch 32/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 1.1082 - accuracy: 0.6974 - val_loss: 1.1682 - val_accuracy: 0.7867\n",
            "Epoch 33/100\n",
            "22/22 [==============================] - 2s 81ms/step - loss: 1.0501 - accuracy: 0.7293 - val_loss: 1.1190 - val_accuracy: 0.7933\n",
            "Epoch 34/100\n",
            "22/22 [==============================] - 3s 119ms/step - loss: 1.0415 - accuracy: 0.7237 - val_loss: 1.0973 - val_accuracy: 0.7933\n",
            "Epoch 35/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.9796 - accuracy: 0.7393 - val_loss: 1.0880 - val_accuracy: 0.7967\n",
            "Epoch 36/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.9040 - accuracy: 0.7648 - val_loss: 1.0591 - val_accuracy: 0.7767\n",
            "Epoch 37/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 0.9272 - accuracy: 0.7481 - val_loss: 1.0556 - val_accuracy: 0.8067\n",
            "Epoch 38/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 0.8807 - accuracy: 0.7596 - val_loss: 1.0194 - val_accuracy: 0.8167\n",
            "Epoch 39/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.8105 - accuracy: 0.7907 - val_loss: 1.0260 - val_accuracy: 0.8167\n",
            "Epoch 40/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 0.7868 - accuracy: 0.7907 - val_loss: 1.0273 - val_accuracy: 0.8033\n",
            "Epoch 41/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 0.7796 - accuracy: 0.7837 - val_loss: 1.0475 - val_accuracy: 0.8133\n",
            "Epoch 42/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.7916 - accuracy: 0.7926 - val_loss: 1.0050 - val_accuracy: 0.8267\n",
            "Epoch 43/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.7373 - accuracy: 0.8019 - val_loss: 0.9956 - val_accuracy: 0.8333\n",
            "Epoch 44/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.7053 - accuracy: 0.8159 - val_loss: 0.9709 - val_accuracy: 0.8367\n",
            "Epoch 45/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.7193 - accuracy: 0.8167 - val_loss: 0.9920 - val_accuracy: 0.8200\n",
            "Epoch 46/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.6611 - accuracy: 0.8222 - val_loss: 1.0171 - val_accuracy: 0.8300\n",
            "Epoch 47/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.6380 - accuracy: 0.8400 - val_loss: 0.9700 - val_accuracy: 0.8400\n",
            "Epoch 48/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.6290 - accuracy: 0.8344 - val_loss: 0.9839 - val_accuracy: 0.8433\n",
            "Epoch 49/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.6015 - accuracy: 0.8430 - val_loss: 1.0234 - val_accuracy: 0.8367\n",
            "Epoch 50/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.5918 - accuracy: 0.8456 - val_loss: 1.0031 - val_accuracy: 0.8333\n",
            "Epoch 51/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.5620 - accuracy: 0.8600 - val_loss: 0.9851 - val_accuracy: 0.8300\n",
            "Epoch 52/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.5213 - accuracy: 0.8719 - val_loss: 0.9731 - val_accuracy: 0.8400\n",
            "Epoch 53/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 0.5299 - accuracy: 0.8644 - val_loss: 1.0244 - val_accuracy: 0.8333\n",
            "Epoch 54/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 0.5053 - accuracy: 0.8793 - val_loss: 1.0059 - val_accuracy: 0.8300\n",
            "Epoch 55/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.4668 - accuracy: 0.8904 - val_loss: 0.9735 - val_accuracy: 0.8400\n",
            "Epoch 56/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 0.5288 - accuracy: 0.8704 - val_loss: 0.9994 - val_accuracy: 0.8267\n",
            "Epoch 57/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.5099 - accuracy: 0.8652 - val_loss: 0.9972 - val_accuracy: 0.8500\n",
            "Epoch 58/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.5280 - accuracy: 0.8581 - val_loss: 1.0272 - val_accuracy: 0.8433\n",
            "Epoch 59/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 0.5080 - accuracy: 0.8756 - val_loss: 1.0004 - val_accuracy: 0.8467\n",
            "Epoch 60/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.4965 - accuracy: 0.8707 - val_loss: 1.0200 - val_accuracy: 0.8467\n",
            "Epoch 61/100\n",
            "22/22 [==============================] - 2s 78ms/step - loss: 0.5096 - accuracy: 0.8696 - val_loss: 1.0478 - val_accuracy: 0.8433\n",
            "Epoch 62/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.4731 - accuracy: 0.8815 - val_loss: 1.0185 - val_accuracy: 0.8533\n",
            "Epoch 63/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.4259 - accuracy: 0.8993 - val_loss: 1.0262 - val_accuracy: 0.8567\n",
            "Epoch 64/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.4126 - accuracy: 0.9048 - val_loss: 1.0512 - val_accuracy: 0.8433\n",
            "Epoch 65/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.4589 - accuracy: 0.8944 - val_loss: 1.0496 - val_accuracy: 0.8567\n",
            "Epoch 66/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.4227 - accuracy: 0.8952 - val_loss: 0.9679 - val_accuracy: 0.8567\n",
            "Epoch 67/100\n",
            "22/22 [==============================] - 3s 127ms/step - loss: 0.3957 - accuracy: 0.9033 - val_loss: 1.0078 - val_accuracy: 0.8467\n",
            "Epoch 68/100\n",
            "22/22 [==============================] - 3s 146ms/step - loss: 0.3763 - accuracy: 0.9119 - val_loss: 1.0393 - val_accuracy: 0.8467\n",
            "Epoch 69/100\n",
            "22/22 [==============================] - 3s 149ms/step - loss: 0.3631 - accuracy: 0.9189 - val_loss: 1.0783 - val_accuracy: 0.8533\n",
            "Epoch 70/100\n",
            "22/22 [==============================] - 2s 82ms/step - loss: 0.3944 - accuracy: 0.9089 - val_loss: 1.0405 - val_accuracy: 0.8467\n",
            "Epoch 71/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.3810 - accuracy: 0.9148 - val_loss: 1.1015 - val_accuracy: 0.8533\n",
            "Epoch 72/100\n",
            "22/22 [==============================] - 2s 97ms/step - loss: 0.3662 - accuracy: 0.9115 - val_loss: 1.1004 - val_accuracy: 0.8533\n",
            "Epoch 73/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.3582 - accuracy: 0.9259 - val_loss: 1.0198 - val_accuracy: 0.8433\n",
            "Epoch 74/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.4175 - accuracy: 0.9033 - val_loss: 1.0546 - val_accuracy: 0.8567\n",
            "Epoch 75/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.3927 - accuracy: 0.9085 - val_loss: 1.0569 - val_accuracy: 0.8533\n",
            "Epoch 76/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.3612 - accuracy: 0.9233 - val_loss: 1.0198 - val_accuracy: 0.8533\n",
            "Epoch 77/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.3490 - accuracy: 0.9230 - val_loss: 1.0472 - val_accuracy: 0.8567\n",
            "Epoch 78/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.3485 - accuracy: 0.9270 - val_loss: 1.0885 - val_accuracy: 0.8533\n",
            "Epoch 79/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.3374 - accuracy: 0.9248 - val_loss: 1.0869 - val_accuracy: 0.8533\n",
            "Epoch 80/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.3415 - accuracy: 0.9281 - val_loss: 1.1377 - val_accuracy: 0.8433\n",
            "Epoch 81/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.3317 - accuracy: 0.9267 - val_loss: 1.0693 - val_accuracy: 0.8533\n",
            "Epoch 82/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.3617 - accuracy: 0.9193 - val_loss: 1.0853 - val_accuracy: 0.8500\n",
            "Epoch 83/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.3238 - accuracy: 0.9311 - val_loss: 1.1296 - val_accuracy: 0.8533\n",
            "Epoch 84/100\n",
            "22/22 [==============================] - 2s 77ms/step - loss: 0.3244 - accuracy: 0.9330 - val_loss: 1.1308 - val_accuracy: 0.8533\n",
            "Epoch 85/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.3277 - accuracy: 0.9300 - val_loss: 1.0875 - val_accuracy: 0.8467\n",
            "Epoch 86/100\n",
            "22/22 [==============================] - 2s 94ms/step - loss: 0.3351 - accuracy: 0.9267 - val_loss: 1.1257 - val_accuracy: 0.8567\n",
            "Epoch 87/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.2904 - accuracy: 0.9396 - val_loss: 1.1101 - val_accuracy: 0.8467\n",
            "Epoch 88/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.3128 - accuracy: 0.9359 - val_loss: 1.1302 - val_accuracy: 0.8500\n",
            "Epoch 89/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.3399 - accuracy: 0.9293 - val_loss: 1.1054 - val_accuracy: 0.8467\n",
            "Epoch 90/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.3093 - accuracy: 0.9441 - val_loss: 1.1290 - val_accuracy: 0.8467\n",
            "Epoch 91/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.3382 - accuracy: 0.9300 - val_loss: 1.1697 - val_accuracy: 0.8433\n",
            "Epoch 92/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.3405 - accuracy: 0.9252 - val_loss: 1.1123 - val_accuracy: 0.8500\n",
            "Epoch 93/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.2674 - accuracy: 0.9489 - val_loss: 1.1446 - val_accuracy: 0.8467\n",
            "Epoch 94/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.2969 - accuracy: 0.9407 - val_loss: 1.1721 - val_accuracy: 0.8500\n",
            "Epoch 95/100\n",
            "22/22 [==============================] - 2s 80ms/step - loss: 0.2825 - accuracy: 0.9452 - val_loss: 1.1981 - val_accuracy: 0.8467\n",
            "Epoch 96/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.2843 - accuracy: 0.9478 - val_loss: 1.1876 - val_accuracy: 0.8467\n",
            "Epoch 97/100\n",
            "22/22 [==============================] - 2s 81ms/step - loss: 0.2916 - accuracy: 0.9478 - val_loss: 1.1841 - val_accuracy: 0.8367\n",
            "Epoch 98/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.2925 - accuracy: 0.9478 - val_loss: 1.1852 - val_accuracy: 0.8400\n",
            "Epoch 99/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.2553 - accuracy: 0.9519 - val_loss: 1.1464 - val_accuracy: 0.8467\n",
            "Epoch 100/100\n",
            "22/22 [==============================] - 2s 79ms/step - loss: 0.2625 - accuracy: 0.9500 - val_loss: 1.1357 - val_accuracy: 0.8467\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7fbcf8a94690>"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!mkdir -p saved_model\n",
        "model.save('/content/drive/Shareddrives/Megathon/saved_model/my_model.h5')\n",
        "\n",
        "print(\"saved!\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2ZQ2uVpKaitH",
        "outputId": "c1e6bd3d-f082-4dd8-e9a9-d62a2e6618cb"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "saved!\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.figure(figsize = (18, 7))\n",
        "\n",
        "plt.subplot(1,2,1)\n",
        "plt.title('Loss')\n",
        "plt.plot(model.history.history['loss'], label='training')\n",
        "plt.plot(model.history.history['val_loss'], label='validation')\n",
        "plt.xlabel('epoch')\n",
        "plt.ylabel('loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.subplot(1,2,2)\n",
        "plt.title('Accuracy')\n",
        "plt.plot(model.history.history['accuracy'], label='training')\n",
        "plt.plot(model.history.history['val_accuracy'], label='validation')\n",
        "plt.xlabel('epoch')\n",
        "plt.ylabel('accuracy')\n",
        "plt.legend()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 475
        },
        "id": "SL6aRylCaivq",
        "outputId": "0db1372c-2d61-4b52-b122-3a383ec5ade4"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fbcf8cf4610>"
            ]
          },
          "metadata": {},
          "execution_count": 12
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1296x504 with 2 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABBUAAAG4CAYAAAAaO9+QAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXhV1b3/8feZkpzMMyEJU8KQkIQZFFBAUBkVBLRoRapWq9W2t9P19vb29nft6L21g2NbS8WqdWZSFBUEZJKZkBFICBnJPM85w++PQCANMycETj6v5+nzNPvsvdfay51w9nd/13cZnE6nExERERERERGRS2Ts6Q6IiIiIiIiIyPVJQQURERERERERuSwKKoiIiIiIiIjIZVFQQUREREREREQui4IKIiIiIiIiInJZFFQQERERERERkcuioIJILzBs2DCKi4t7uhsiIiLihpYsWcKdd97Z090QkR6ioIKIiIiIiFyWI0eO4OfnR2RkJAcOHOjp7ohID1BQQaQXa2lp4b//+7+ZOXMms2fP5re//S12ux2AN954g9mzZzNr1iwWL17M0aNHz7tdREREep9Vq1Yxa9Ys5s2bx+rVqzu2r169mpkzZzJz5kx+/OMf09raes7tu3bt4rbbbus49syfn3/+ef7rv/6LxYsXs2LFChwOB//zP//DzJkzmT59Oj/+8Y9pa2sDoLKykscee4wZM2Zwxx13sG3bNjZv3sy8efM69XnhwoVs2LChu4dGpNcw93QHRKTnvPbaaxQXF7Nu3TpsNhv3338/H330ETNmzOBPf/oTmzZtwtfXl08++YTNmzfTt2/fs24fMmRIT1+KiIiIXGV2u53PP/+cJ554ApPJxLPPPktrayulpaU888wzrF69mvDwcL7zne/wj3/8g1mzZp11e1JS0nnb2bJlC2vWrCE4OJhPP/2UvXv38tFHH+FwOLjrrrv4+OOPmT9/Ps8++yyxsbH8+c9/Jj09nQcffJCtW7dSVlZGZmYmcXFxFBUVkZeXx5QpU67SKIm4PwUVRHqxzZs389BDD2E2mzGbzdxxxx1s376dOXPmYDAYeP/995k3bx6zZ88GoK2t7azbRUREpPfZtm0bSUlJ+Pr6AjBhwgQ2bdpEdXU1o0ePpk+fPgA8++yzmEwmPvjgg7Nu37dv33nbGTlyJMHBwQDMnDmTW265BYvFAkBSUhL5+flAe/DhlVdeAWD48OFs3LgRDw8PZs6cybp164iLi2PDhg3MmDEDDw8P1w+ISC+l6Q8ivVhlZSUBAQEdPwcEBFBRUYHFYmHFihXs37+fmTNnct9993H48OFzbhcREZHeZ+XKlWzevJlx48Yxbtw4PvvsM1atWkVVVRX+/v4d+3l6emI2m8+5/ULO/K5SWVnJU089xcyZM5k1axYbN27E6XQCUF1djZ+fX8e+p4Idc+fOZd26dQBs2LCBOXPmXNmFi0gnCiqI9GKhoaFUV1d3/FxdXU1oaCjQHuF/7rnn2LlzJzfddBM///nPz7tdREREeo+amhp2797Nrl272Lt3L3v37mXPnj2kpKRgNBqpqqrq2Le+vp7y8nKCgoLOut1kMnXUdAKora09Z7t/+MMfMJvNfPjhh6xfv56pU6d2fBYYGNjp/AUFBbS1tTF+/HhsNhubNm3i6NGjTJo0yVXDICIoqCDSq02bNo33338fu91OY2Mja9asYerUqRw+fJjvfve7tLa24uHhQWJiIgaD4ZzbRUREpHdZt24dN954Y6dpBGazmZtuuonW1lb2799PQUEBTqeTn//857z//vtMnTr1rNvDwsIoKyujoqICu93Ohx9+eM52KyoqGDp0KB4eHmRmZnLgwAEaGxsBmD59OqtWrQIgKyuLhQsXYrfbMRqNzJkzh1/84hdMnz69Y+qEiLiGaiqI9BJLly7FZDJ1/PzLX/6SpUuXkp+fz9y5czEYDMyaNaujTkJ0dDTz5s3DYrHg4+PDf//3fzN06NCzbhcREZHeZfXq1SxbtqzL9ttuu42XXnqJp59+mmXLlmEymUhKSuLBBx/E09PznNsXLVrEggULiIyMZP78+WRkZJy13YceeoinnnqKlStXMm7cOJ566il++tOfMmLECH784x/z1FNPMX36dHx8fPjd736Hl5cX0D4F4tVXX9XUB5FuYHCemoQkIiIiIiLihsrLy7nrrrvYvHlzp5csInLlNP1BRERERETc2nPPPce9996rgIJIN1BQQURERERE3FJ5eTkzZsygvLychx56qKe7I+KWNP1BRERERERERC6LMhVERERERERE5LJcM6s/lJXVufycQUHeVFU1uvy8vZHG0rU0nq6l8XQtjadrXel4hoX5ubA3ciGu/j6i3yfX0ni6lsbTtTSerqXxdK3u/D7i1pkKZrMKsbiKxtK1NJ6upfF0LY2na2k8ezf993ctjadraTxdS+PpWhpP1+rO8XTroIKIiIiIiIiIdB8FFURERERERETksiioICIiIiIiIiKXRUEFEREREREREbksCiqIiIiIiIiIyGVRUEFERERERERELouCCiIiIiIiIiJyWRRUEBEREREREZHLoqCCiIi4pc2bN17Ufn/607MUFRWe8/P/+I8fuKpLIpdN97OIiFyrFFQQERG3c+JEERs2fHpR+37vez8kMjLqnJ//9re/d1W3RC6L7mcREbmWmXu6AyIiIq72+98/Q0ZGGjffPJ7bb5/NiRNF/PGPL/Gb3zxNWVkpTU1NPPTQo0yefDNPPvkoP/jBv7Np00YaGurJy8ulsLCA7373h0ycOJm5c2ewbt1GnnzyUcaPv4H9+/dSXV3NM8/8gdDQUJ5++mcUF58gKWkEX3yxgVWrPu7pyxc3o/tZRESuZQoqiIhIt3r3iyz2ZJa69Jzj48K5Z/rgc35+771LWbnyXQYNiiUv7zgvvfQ3qqoqmTDhRmbPnkdhYQE/+9l/MHnyzZ2OKy0t4Xe/e46vvtrBmjUfMHHi5E6f+/j48Kc/vczLLz/Pl19+QWRkNK2tLfz1ryvYvn0r7777lkuvU649l3o/m0wG7HbneffR/SwiItczBRVERMStxccnAODn509GRhpr167EYDBSW1vTZd8RI0YBEB4eTn19fZfPR44c3fF5TU0Nubk5JCWNBGDixMmYTKbuugwRQPeziIhcexRUEBGRbnXP9MHnfQvb3SwWCwCff76e2tpaXnzxb9TW1vLNby7tsu+ZD1FOZ9e3y//6udPpxGhs32YwGDAYDK7uvlxjLvV+Dgvzo6yszmXt634WEZFrjVsWanQ6neQW12GzO3q6KyIi0gOMRiN2u73Tturqavr2jcRoNLJlyxe0tbVdcTtRUdEcPpwOwO7dX3VpU8QVdD+LiMi51DS0kpFbRWVt81kDyFeDW2YqZBfW8us39vHk3aMYExvc090REZGrbMCAQRw+nEnfvpEEBgYCMG3adP7jP35Aenoqc+feSXh4OK+++soVtTNp0s2sW7eWxx9/mNGjx+LvH+CK7ot0ovtZROT61GazY7M7sXq69rG7tc3OwaxydqQWk3qsEsfJYILV00RkiA8DIvxYOCUGby+LS9s9F4Ozp8IZ/8KVqYEZuVX831sHuPf2Ydw25tzLKsnFc3X6Zm+n8XQtjadraTwvXm1tDfv372XatBmUlZXyve89zj//+UGnfa50PMPC/K60m3IJXH3vX0+/TxdzP/e062k8rwcaT9fSeLqWO4zn7989yJG8au67bSg3j+h7zmlljc02jhXVcKSghpwTtbS0nSdTzAmF5fU0tbTvMzDCj/gBQZTVNFNU3kBJZSN2h5P/emAcMZH+HYd15/cRt8xUsHq2zwdsbLb1cE9ERMSdeXv78MUXG/jnP1/H6XTwne/8oKe7JHLZdD+LiJybw+HkSH41g6MDMJsuXEUgq7CG1GOVAKz4JJP045U8MDMOb6/2R/DaxlZ2pBTzVVox+aX1nPmm33iBmjZBfh5MHxPNxIQIIkN9On1msztoarHh5+1xaRd4BdwzqODRfllNLQoqiIhI9zGbzTz99G96uhsiLqH7WUTk3N7fnM363XmMjwvnsfkJFyxm+/HOXAAenhvPloNF7M4o5VhRLfNvGkTKsQr2HS7D7nBiNhkY0i+QIdEBDIkOZHCU/xVNWzCbjFc1oABuGlTwOjlnpbH5yosWiYiIiIiIiPtwOp18/FUuNQ2tjBsWzuDogPNmBxw8Ws763XkA7MksZVBff2bd0P+c+xeW1XMwq5zYSH8mJUZwY0If1mzLYd2OXJavywAgMtSHqSMjmZgYga/16tQ+6C7uGVTwODn9QZkKIiIiIiIicob1u/P4YMsxADbsLSDA14NxQ8OZmBjRqQ4BQEVNM8vXpWMxG/nOoiSWr8vgvc1ZDOjjS/zAsy8K8Mmu9gDEnIkDMBgMmAwGFk6JJX5AMAeOlDE+PpzBUQFus3SvWy4p6WE2YjQYaFJNBREREREREbfW1GKjpKqxYxWE89mbWcp7m7IJ8vPk2wsSuXlEX2w2Bxv3F/DLf+zlL2vTqKprAdrrE/x5TSoNzTa+fttQEgeF8MRdSRgNBl5ek0ZFTXOX85fXNLErvYTIUB9GDg7t9Fn8gCDuu20oQ6ID3SagAG6aqWAwGLB6mlRTQURERERExM0UljewNbmIovIGiioaqKxtDwL4Wi0MHxjE8IHBDB8QRGigtdNx2UU1vPJROp4eJr63eAT9+/gxLi6cpTOHkX68ijXbjrErvYSDWeXMnzyIyrpmsotquTGhDzeP6AvA4KgA7rt1CK9/doQXVqXwn/ePwWI2dbTx6e587A4ns2/of8GCi+7CLTMVoH0KhGoqiIjI+SxefAeNjY28/voKUlMPdfqssbGRxYvvOO/xmzdvBODjjz9ky5ZN3dZPkYuh+1lErke1Da3szSwl7XglpdVN2B2O8+6fX1rPb9/Yx2d78knNqcTphISBQUyID8diNrI7o5QVn2Ty73/eyU9f+Yp3N2VxJL+akqpGnn//EDa7g8fnJ9C/z+klEs0mIyNiQ/jpA+NYNmsYZqOBdzdlsWFvARHB3jwwc1inzIJpo6O4KakvucV1PL1iL18mF9HaZqe2sZWtyUWE+Htyw/A+3TZm1xq3zFSA9mKNNfWtPd0NERG5Dixd+o1LPubEiSI2bPiUadNmMGfO+R/WRK4m3c8icrlOVDSQkVvFtNFRV/yWvanFxr7DZexIPcGJykaiw3wZGOHHgD5+BPl7knG8iuSsco4V1XZZTjEkwJOpY/px6+hIPCynswBKKht59p2DNDTbWHr7UG4Y3qfTSglOp5PiykbSj1eRllNJ+vFK1u/KY/3JGgcAX79tKCNiO09LOLPtqaOiGDssnJVbssnIreLxBYl4eXR+bDYYDCydORSH08lXaSWs+CST9zdn0zfEm1abg5kT+l/UspPuwm2DClYPMydaGnE6nW41X0VERC7soYe+zq9//SwREREUF5/gJz/5IWFh4TQ1NdHc3Mz3v/9jhg9P7Nj/V7/6f0ybNoNRo0bz05/+O62trYwYMarj888++4T3338Hk8nIwIGxPPXUT/n9758hIyONV199BYfDQWBgIIsWfY2XXvoTKSnJ2Gx2Fi26h6VLl/Dkk48yfvwN7N+/l+rqap555g9ERET0xNDIdehaup9nzZqr+1nETdU3tfH7dw5SUduCn7cH4+PCL+q4nanF5JfWYzEbO/6XW1zH/iNltNrasw4CfT1Iy6kkLaey07FGQ/tyikkxwbTZHJRVN1NW08SJ8gbe/+Io25MLeWhOPLFRAVTWNvO7tw9Q29DK128byi1jorv0xWAw0DfEh74hPswYG01rm52M3CqSsytIz6lkfHw4M8Z2Pe5f+VotPDAr7rz7WMwmvjlvOAunxLDpQCFbDhZxtKAGX6uFm0dGXtTYuQu3DSp4eZpwOJy02RydolsiInJ1rcz6iAOlKS495+jwJBYOnnfOz6dMuYXt279k0aJ72Lp1C1Om3EJs7BCmTJnGvn17ePPN1/jVr/6vy3GffvoJMTGxfPe7P2Tjxs/YsOFTAJqamnj22efx8/PjiSceITs7i3vvXcrKle/y4IOPsHz5XwA4eHA/x45l8/LLf6epqYlly5Zw113t/fTx8eFPf3qZl19+ni+//IJ77rnPpWMiV8el3s8mowG74/yFw66n+3nKlGmA7mcRd+NwOvnrh2lUnKxNsHZ7DmOHhV0wW2HroSJe/TjzrJ+FB1mZlBjBxIQIwgKt1De1kVtSR25xHRU1zQyJDiAxJuSsyym2tNr5eE8+H249xq/f2Mdt4/pxKLuCitoWFk6JuajAAICHxcTIwaFdCia6UrC/F4umxnLHpIHsP1JGeJA3nr3s+dN9gwonU1SaWu0KKoiI9DJTptzCCy/8kUWL7mHbti08+eT3efvt13nrrddpa2vDy8vrrMcdP36MUaPGAjB69NiO7f7+/vzkJz8EIDc3h5qa6rMen5mZzqhRYwCwWq0MHBhDbm4uACNHjgYgPDycmpoa11yo9ArX0v2cn58P6H4WudY4nE62HChkS3IR8ycPYvTQsEs6/qPtx0k9VkliTDC+VgtfpZVw4EgZY4edO1uhoKyeNz87grenmccXJGI2GWi1OWizOQj09WRQX79OGeO+VgsJA4NJOMcyjGfy9DDx6IIkhvcL4NWPM/lsT/vfnlk39GfuxAGXdG1Xi4fFxI0JvTNry22DClaP9kBCc4uNAB+PHu6NiEjvtXDwvPO+he0OMTGxVFSUUVJSTF1dHVu3biY0NJyf/ewXZGam88ILfzzrcU4nGI3tX4AcJ98ut7W18fvf/y8rVvyTkJBQ/v3f/+2c7RoMBs5czcpma8NobJ9TaTKdDnA7L2LJK7k2Xer9HBbmR1lZ3RW1eW3dz+3n0/0scu0oLKvntfWHySpsD/C9sCqFpbcPY9roqC77Hi+uxemEARF+HVkIqccqWLMthxB/Tx69I4G6xlZ2pZWwdvtxxgwNO+tU8uZWGy+vTqXV5uDROxNIGHThQMHlGNY/iP95eALrdh7HYjIyb9JATW2/Brlt9QirZ3u8pLnV3sM9ERGRnjBx4k389a8vcfPNU6mpqSYqqj1VcsuWTdhsZ19yuH//AWRmZgCwf/9eABobGzCZTISEhFJSUkxmZgY2mw2j0Yjd3vnfmLi4BA4c2HfyuEYKCwsYMODafKMi15dr5X6Oju7fXZcoIpeozWZn5ZfH+H+v7iGrsIZxceF8b/EIfLws/OPTw6zeeqwj6He8uJZn3z7A0yv28ovX9vLDF7bz6scZ7Ewt5q8fpmMyGfj2XUn4Wi30DfFhwvA+5JfWczCrvEu7TqeT1z89womKRm4b148xl5gVcak8LSYWTonljsmDFFC4RrltpoLXyUyFppaz/0MrIiLuberUW3jssYdYseItmpub+OUvf86mTRtYtOgeNmz4jHXr1nY5Ztasufznf/6I733vcUaMGIXBYCAgIJDx42/gm998gMGDh3DffUt57rnf8/zzf+Hw4Uyee+5ZfHx8ARg5chTDhsXxxBOPYLPZeOyxJ/H29r7aly5u6Fq5n61Wa5d2ROTqK61q5KVVqeSV1hPs78n9tw9j1Mm6AT9dOpZn3znI2u3HqahtpqXNwd7MUqB96cUgPy8OZZez9dAJth46AcDS24cyqK9/x/nnTRrI7vT2bIVRg0M7PcxvO3SCnWnFDOrrz923xF7Fq5ZrlcF5jeSsXWlq4L9avyuPdzdl8Z1FSYwe0r3Rs97AFembcprG07U0nq6l8XStKx3PsDC/C+8kLuPqe1+/T66l8XQtjadrXa3x3H+kjOXrMmhqsTFlZF+WzBjSZcnDmvoW/vBeMnkl9QAM6uvP4qkxxJ+sZ+BwOsk5UUtyVgVWTxOzJvTvkgXw8upU9mSW8m93j2BEbCjV9S18vjefDXsLsJiM/L8HxxMa2H2BRt2frtWd30fcNlPB6nmqpoKmP4iIiIiIyLWjoLSe3ZmlTBnZl9CAi3swt9kdrNxyjPW78/AwG3l4bjyTk/qedd8AX0+eum8M63bmMqivX5faCEaDgdjIAGIjA87Z3h2TBrIns5RVX+Zw4Gg521NOYLM78fe28M15w7s1oCDXFzcOKpyqqaDpDyIiIiIi0vMKSutZuz2HvYfLANh2qIgfLhlNVKjPeY+rqGnmLx+mkVVQQ59gb55YkEh0uO95j7F6mlk87fKnJ0SH+zJ2aBj7jpSRW1JHeKCVWTf0Z3JSBBazVteT09w2qNBRU0GFGkVEREREpAdV1bXw1oYjHcGEQX39GBIdyGd78vntG/v4/j2jiIn0P+ux+w6X8urHmTS22JgQH86yWXEdL1C7293TB2OxGBk9JIyxQ8M6VoAROZMbBxXaL02FGkVEREREpKc4nU7+sjaNI/nVDOrrx/ybBpEUE4LBYCAq1IcV6zP5v7cP8N2FSR01DwBa2+y880UWmw4U4mE28o3Zcdw8ou9VXQEhPNDKo3ckXLX25PrktkGFjukPqqkgIiIiIiI9ZFvKCY7kVzNqcCjfWZTUKShw88hIvL3M/GVtGn94L5kh0YE0NttoaG6jrqmNllY7UWE+PDY/8YJTJER6ivsGFU5Of1BNBRERERGR3qu1zY7ZbMR4Fd/wn1Lb2Mq7X2ThaTFx/+1Dz5plMHZYON+728yfV6eSkVuFh8WIj5eF0AAvEgYGs3BKDB4W1TCQa5fbBhW8TmYqqKaCiIiIiEjvdDiviuc+OERSTAiPzU+86u2/+0UWDc02lswYQrC/1zn3SxgYzB++cxMAZpPxanVPxCXc9o7tKNSomgoiIiIiIr3O4bwq/vjeIZpa7OzOKCUzt+qqtp9xvJIdqcUMiPDj1rHRF9zfbDIqoCDXJbe9a80mIx5mo6Y/iIiIiIj0MqcCCja7gwU3DQLgnS+ycDidV6X9Npudf3x6GIMBls0aplUTxK25bVABwNvLQrOmP4iIiIiI9BqH86r4w3vJ2OwOvr0gkTtvGsSNw/uQW1LHrrSSq9KHVVtzKKlq4tax/RgYcfalIkXchVsHFayeZk1/EBERERHpJXKL6/jDe8nY7U6+vSCR0UPDAFg4NQazycgHX2bT2ta9Lx0/35PP+l15hAV6seDmQd3alsi1wL2DCl5mFWoUEREREekFmlpsvLw6ldY2B4/NT+gIKACEBli5bXw0lbUtfL43/5LO29JqZ93O42TmVuG8wPSJrclFvLXxKAG+HvxwyeiOZe5F3Jlb3+XeXmZaWu04nM4eWUJGRERERERcx+Fw0thiw9dq6bTd6XSy4pNMSqubmH1jf8YOC+9y7NwbB7I1+QTrduZy84hI/H08LqrN9zdns3F/AQDRYb7cOi6aG4f36bLM4/bkIlasz8TXauFHXxtFeKD1Mq9S5Pri1kGFU5HBlla7ooQiIiIiItexNpudP753iMN51cyZOIA7Jw/sWC1h04FC9mSWMjg6gIVTYs56vLeXmfk3DeLNz4/w5udHWDZrGN5elrPue0pWQQ1f7C8gItibfuG+7DtcxopPMnl/czZDogMI9PUkwNcDk9HAmm05eFpMfP+ekUSF+br8+kWuVW79pO3t2f5HoqnFpqCCiIiIiMh1ymZ38PLqNDJyqzCbDHy04zgHj5bx0Nx4DBh4e+NRfK0WHrszAZPx3DO8p46KZMvB9gDEoWMVTBsVye3j+xPk59ll3zabgxXrM3ECD86JY0h0IJW1zWw6UMiWg0UcOFreaX8Ps5HvLR7BoL4qzCi9S7c+aTc3NzNv3jy+/e1vs3Dhwu5s6qy8vdovT3UVRERERESuTw6nk7+vy+BgVjkJA4N49M4EVn15jM0Hi/jla/vwsZqx2Z08csdwgv29znsus8nIT+4fy5aDRXy6J49Pd+ezYW8Bk5P6snBqDP7ep6dEfPJVLkXlDdwyOooh0YEABPt7sWhqLAunxNDQbKO6roXqhhaq61oZkxCBt0lTrqX36dagwssvv0xAQEB3NnFep4IKza1aAUJERERE5HrjdDp587MjfJVeQmyUP08uHIGnh4kHZsUxNi6cFR9nUlHbzNyJA0iKCbmoc1o9zcy6oT8zxkazM62YT3bl8WVyEfuPlPG16YOZlBhBUUUjH+44TpCfJ4umxnY5h8FgwNdqwddqIZr2qQ5hYX6UldW59PpFrgfdFlTIzs4mKyuLadOmdVcTF3RqykNzizIVRERERESuN6u25rDpQCH9wn35/t0j8fQ4XRwxYWAwTz88gWNFtcQPCLrkc1vMRqaMjGRyUgQb9xWy6stjLF+XwY7UYppb7dgdTu6/bWjHi0oRObtu+w155pln+NnPfsbq1asvav+gIG/MZtOFd7wE1pN/ACxeFsLC/Fx67t5IY+haGk/X0ni6lsbTtTSeIiKXbt/hMj7acZzwICs/+NqosxZVtHqaSRgUfEXtmIxGbh/fjzFDQ3njsyMcyq4AYNywsE7LUl4uu8NOTm0eFU2VjOszCpOx6zOP0+kkuSyV8ubKTtuHBsbS3z/6ivtwuZptzewrTWaAXz+ifPtiuMCKek22Zg5XHsXmtBMXPARfi89V6un5OZ1OksvTiPLpS5j3xWW0yMXrlqDC6tWrGTVqFP369bvoY6qqGl3ej1OFGkvK6pSKdIWUzuVaGk/X0ni6lsbTta50PBWQEJHeqKSykb9/nI6HxciTC5MIuMjlH69EaICV7y0ewZ7MUpKzKrhn+uDLPlervZWU8nRSyjNJr8ikwdb+rJNfX8jiIXd22f/zvM2syf6ky3ajwcjS+HuYEDHmgm06nU4qm6sI9go658N/XWs9VS3VnbaFeAXjY/Husq/dYedvqW+QUXkEgCDPQBJD40kMicPf8/S/TU6nk2M1uaSWZ3C0+hh2Z3uWuAEDMQEDSQqNZ3DgoE7BFKvJet6H+5qWOnzazv2o2mxrprSpc6FML5Mn4d5dg0B2h523D69ix4ndWM1WvpW0jCFBZ18hRC5PtwQVNm/eTH5+Pps3b6a4uBgPDw8iIiKYNGlSdzR3TqdrKmj6g4iIiIjI9aClzc6Lq1JoarHzyLzhRF/F5RkNBgMT4vswIb7PZZ+jsrmKFw8up7ixFIBAzwBuCtA+WaEAACAASURBVL+BrOocNuVvo693HyZH3dCxf3JZGmuz1xPoGcDdQ+djMrSvXlHf1sgHR9fyWvrb1LbWcWv/qedss7ihhHeOrOFIVRZDAmO4Z+gCIn0jOj5vtbfyWe4mPs/djM3Z+dnIy+TJo0nLGBbcOYiyMusjMiqPMCQwhgBPf9IqDrO1cCdbC3eesx/9/KJIConHbDSTWpHBsZrjZNfknHXfGf2nsCB2DkZD59U69pce4rX0t3E6HQwOjCEpdDhJofGAgZTy9C7BizONDEtk0eA7CLEGdVz339PeJKU8g3DvUMqbKnkh+W88OPxeRoUnnfM65NJ0S1Dhj3/8Y8f/f/7554mKirrqAQU4s6aCCjWKiIiIiFzrnE4nb3x6mIKy9lUXJiZGXPigK7Qhbwt7ig8wZ9BtjAgdfsEU//Mpqi/mxeTlVLfUMDlyAjdHTSTaNxKDwUBZYwX/t+953j6yijDvUIYGxVJQV8SK9LewGM08NuIb9POL6nS+/n5RvHhwOauy1lHTUstdg+d2eghvtjXzyfGNfJG/FYfTQbg1lKPVx/jNnj8yLXoycwbdxpGqbN4/upbK5ioCPQMYHZbUcY1tDhs7i3bzYvJylg3/GmP7jAJga+FONhdsp69PH7414htYzV7YHXaO1RwnsyqLVntrp36Ge4eRFBpPoOfpIv0zB06nrrWetIpMCutPdNo/tSKDjXlfUttSz/3xizEb25/bNhds5/0ja/EwWYgM6MvhqiwOV2Xx/tG1nY7v5xtJTOBATIbT2Q/Ha/NILkslveIwMwdMZ2LkOP6W8gY5tbnEBQ3hkaSl5NTk8UrqP/hb6hvcM3QBN0fdSGH9CVLKM0ityKChrYH44KEkhg5naGAMFlPXKTfnklWdw5rsT2izt5IQGk9SaDz9/aIxGoyUNVaQUtEeEClrqmBIYAyJofHEBw/Faj7/iiXXA4PT6XR2ZwOnggoXWlKyO1JtKxvb+NFzW5l1Q3/uueXy05dE6dCupvF0LY2na2k8XUvTH64vrr739fvkWhpP17rWxtNmd7DpQCFvbTjKoL5+/MfXx2IxGy984BWoa63nZzt+Q5ujDYDhIcO4e8h8wr1DAahuqSGlPIOjVdnYHJ1fVPb1jej84Ogs5pmtL9Fka+auwXPPmllwtOoYzx98BS+TJ4+N/AZ/T/0nVS3VPJK49Jxvzs/MfBjkPwB/j9OZG8dr86lprSXEK4jFQ+4kKXQ4aRWZvHdkDeXNlXiYPGi1t2I0GJnRbwqzBs7Ay+zZ6fyHK7P4a8prNNtbWDzkTiJ9Ingh+W94m638eNx3CLVeWc2Ks6lvbeDPh14lpzaP+OChfDPxfj7L3cynuV/gZ/Hl26MeYmxMPEcLCkgrzyS1IhMnThJC4roEL05xOp3sLt7Pqux11LXWYzQYcTgdjO8zmvvj7+4IXOTVFvBS8t+pa6vH38OP2tb23wGjwYiH0YNmezMAHiYPYgMG4mE8HVgwGozEBAwgMTS+Y6pFTUsdq7PXsbt4PwBmg6kjG8TPwxdvs5WSxrKOc3ibrTTamgAwGUzEBAzA22zt1MbkqBuIDx56wXF0Op3sKTlAclkqZz7We5m9WDh4Hr4ep2tadOf3kW4PKlys7viD1uyAb//vF0wbFckDs+Jcfv7e5Fr7R+d6p/F0LY2na2k8XUtBheuLggrXNo2na13N8XQ6nRQ1FHOivpiE0DisZitOp5Mj+dWkH6/iaEE1x07U0trmwMfLzM8fHE9ogPXCJ75CH2avZ33uF8zoN4XC+hNkVh3FbDAxrs9oCuuLyK8vuuA5/D38GBoUS3JZKg6nk/vj7z5vDYQdRbt5M/P9jp/viJnFrIHTz9tGQ1sjr6T8g6PVxzpttxjN3Np/GrcPuAWPM96qt9nb+DxvM5/lbiY2YCB3D51PhE/4Oc+fX1fES8nLqW2tw2w043Q6+e7oRxkcOOhCl3/ZWu2tLE99k9SKDHws3jS0NRJmDeGJkd8kzDvksu/PJlsT63I+Z2vBTqb1u4n5sbO7TLEoa6zgpUPLqW9tYHjIMJJChzM8eCieJk+O1RwnpTyDlIp0ShvLz9EK9PEOIzZgEPtLD9Fsb6afbyT3DLuLSJ8IMquOklKeTlp5Js32FuKCh5AUGk9iSDx+Hr4U1BW119yoyCC/rrDLuY0GI/cNW8TEyPHnbL+groh3j6wmu+Z4l8/MBhM/HPcE/f1OF/lUUOEyGSxmHvzFZ9w4vA+P3png8vP3JvpH3LU0nq6l8XQtjadrKahwfVFQ4dqm8XSt7h7PNpuNg6WZHKs7Skp5RkeBQD8PX+4YOIf0g1Z2ppYAYACiwnwYHB3ItFGR9O9z9r99DqeDhrZG/Dwuvs5CY1sTXmbPLg+WTbYmfrbjN5gNZp6e9BMsRjMHylL44OiHVLfUYDKYOtLUE0KGdVrJwO50kF2dQ0pFBmnlmdS11eNl9uSbiUsv6g3zyqMfsTH/S8b3Gc2y4UsuasqF0+mk6eQb7lMsRst5U/TtDvtZV5s4m4qT9QZKG8u5P/4eJvYdd1HHXYn2Ioor2XFiDwP8+vH4yAc7/tte6f15oWt3OB0AXe6LMzXZmjplADTbW8isbL+fMyuP0Opow9ts5Y6YWdwUdUOXc11MG822Fhxn1IcorD/BKymv02Br5I6YWcwccEun+6OhrZGPcz5nS8EOnDgZGZbI/NjZ+J1xf5qNZjxMnYubduf3EbdedPVUocYm1VQQEREREblqGtua+J8tL1NvLAbAy+jFuD6jCPQMYHP+dv555F3stiCi+49j0YTRDI4OwOcsS0aeqb6tgT8nr+B4bR5Toycxd9DteFvOn82QXJbGq2n/ZIB/NE+OegSL8fTjz9aCr2iyNTM/ZnbHW/4x4SNICIkjv66QKN++553vPio8iVHhSTicDvLqCujfJxxj08XNj18weA5j+4zsqLdwMQwGA95nWaXhfC42oAAQYg3mqXHfpaypoktth+5iMpq4L24xkyInEOUb2SnbwhXnPp/zPeifYjV3vr+8Ld5MipzApMgJtNrbyKsrIMIn/JxLZ15MG/86HWVIUCw/GPs4LxxczofH1lPTUsu0fpNJK88gpSKTrOpjHfUz7h46n+Ehwy7YRndz66CCl8fJoIJWfxARERERuWSt9lYOV2WRUp7O0apj3Np/aqeVC86muqWG3+/5K/XGMhzV4bSdGEBTfRCNQ/vQv38QrWlGnBFpmIJLqfLfyLaGPGrK2yv8B3kFnvWcFU1VvJj8N0oay7CavdhcsJ19JcncNXguEyLGnPXBfFvhV7x9eBVOnGRV5/B25kruj78bg8FAq72VjflfYjV7cXP0jZ2O8zR5XFLav9FgZKB/f8J8/Shrurg3wUaDkQH+/S66javFy+x11QIKpxgMBgYFDLiqbbqCh8nSbdNDInz68KNxT/DiweV8WbiDLwt3dHw2wK8fY/uMZEr0pE5Bsp50bfSimxiNBrw8TDS3KlNBRERERORi5dTk8WnuRjIrj9J2RoHC946uYXDgIPqcY35+cUMJLxxcTlVrNbaS/jw17X7qG22s2ZbDvsNl7DtchqfFmweHL8W3TzVrsz8hveIw6RWHeefIKqJ9I0n8l8r5hfUnePHg36g5uazivEG3szF/K+uPb+QfGe+wtfArJkVOICEkjgBPP5xOJx8f38DHOZ/ja/Hhm4lLWZW1jq+K9xLhE85tA6ax48Qe6tsamDVgepe30SLXgkDPAL4/5nH+efgD7A47iaFxJIbEE+Dp39Nd68KtgwpAe1ChRZkKIiIiIiIXUtdaz9rsT9hxYg8AfX36kBQ6nMSQeKpbavh72pu8nvEePxj7eJfU7mM1ufw5+VUabI205Q9hWtRUYvq2Zx4kxYSQcqySA0fLuG1cPyJDfYA+JIQMo6KpkpSKDFJPrrJQUF/E+uMb8fPwZXjwMJLL0mi2N7No8Dym958CwKyB05kQMZoPjn7EwbIUcmpzARjg3w9/D19SyjMI8QrmyVEPE+4dxqMjHuD/9r7AmuxPCLWGsCF3CxajhWn9brp6gytyibwtVr6ZeH9Pd+OC3D6oYPU0U9/U1tPdEBERERG5ZjmcDrYV7uLDY+tptDUR6RPBPUMXMCQoptN+yWWp7CtN5ov8rZ2WTEwpT2d56pvYHDZsOUn4t8SwcMrpYw0GAyNiQxgRG9Kl7RBrMNOiJzMtejLNtmYyT063SCvPZFfxPkwGEw8m3Me4PqM6HRfsFcQjSUspbSwjpbw9KJFVk4PD6aCfbySPj3yYAM/24nKBngF8a8Qyfr/vZZanvoETJ7dE33RJBR9F5OzcPqjg5WGmrLq5p7shIiIiInJNcjqdvJ7xLruL9+Nl8mLxkDuZEjXxrIXu7hm6gCNV2Xx47FMSQ+KJ8Alne9Eu3spcicVoJqTiJvLLfPj6oqEd9c0uhZfZi1FhiYwKS+wogOhh9CDSN+Kcx4R7hzGjfxgz+k+hsa2JnNo8YgMGdimA198vmmXDl/C31NcxGUzMOJn1ICJXxu2DClZPEza7A5vdgdl04eqbIiIiIiK9yWe5m9hdvJ8B/v34VtI3Ot7un42vhw9Lht3FK6mv83rGuySEDGNdzuf4WLwZY5nLZ9l1jBkaxughYVfcr1MFEC+Ft8VKwnmq4Y8OT+LRpGUA5ywKKSKXxu2DCqcipM2tdnytCiqIiIiIiJxysDSFtcfWE+QZeMGAwimjwpMYGz6SfaXJHK/Nw88UgEfuRD4rqMPqaeK+W4dchZ5fvpFhCT3dBRG34vZBBatHe9pWU4sNX6vr1j0VEREREbkW2Bw2DpWn02w7PeXXYDCSFBqPr8XnnMflVOXzWvrbeJg8eGzEhQMKTS02KutaaGxuY7hlCpnGHFqaTJSmj4Q2GBcXzoKbBhHs7+WyaxORa5/bBxW8PNsvsalFy0qKiIiIiHtptjXzSsrrZFYd7fJZhE8ffjT2Cazmrg/5NS21PLvzZdocNh5JWkq0X+R528kqqOHZdw/S0nrGqmqGG8FpYFxcH+6cNJDocBU9FOmN3D+ocDJToblVy0qKiIiIiPuoba3jpeS/k19XSGJIPGPCR3R8dqQqm6+K9/L3tDd5fMSDnZZ/rG6p4aXkv1PRVMX8mNmMDEs8bzvV9S28uCqF1jY7N4/oi5+3Bz5eZry9zAyODiQq9NzZECLi/tw+qGD1PFVTQZkKIiIiIuIeShvLePHgcsqbK5kcOYGvDb2r02oN4yNGU9tWR3rFYVZlrWPRkDsAKGko5YXk5VQ2VzFr8DRu6zftvO3Y7A5eWp1KTUMr99wymFk3XFrhRBFxf25fufB0TQVlKoiIiIjI9a+ovphn971EeXMlcwbeyr3DFnVZ/tFoMPJQwn1E+PThi/ytbC/cRU5NLs/uf4nK5iruiJnJg2PuwWAwnLettzceJaughgnx4cyc0K87L0tErlNun6nQUVNBmQoiIiJu59e//jXJyckYDAb+8z//kxEjTqd/v/nmm6xduxaj0UhiYiI//elPe7CnIq6zLudz6tsa+NrQBUyJnnTO/axmK4+P+Ab/u/d53j6yCpPBhN1p5+txdzMpcvwFAwrbU07wxf5CosJ8eHB2/AX3F5Heye0zFTpqKihTQURExK3s3r2b3Nxc3nnnHX71q1/xq1/9quOz+vp6li9fzptvvslbb71FdnY2Bw8e7MHeirhGQ1sjqeXp9PXpw81REy+4f6g1hEcSH8BAe0DgW0nLmBQ5/oLHZRyv5LX1h/H2NPPkwiQ8PUwXPEZEeie3z1SweqimgoiIiDvauXMnt956KwCxsbHU1NRQX1+Pr68vFosFi8VCY2Mj3t7eNDU1ERAQ0MM9Frly+0oOYnPauSFi7EVnDgwJiuHH476Dp8lCuHfYBffffKCQNz8/AsCjdybQJ8j7ivosIu7N/YMKHUtKKlNBRETEnZSXl5OQkNDxc3BwMGVlZfj6+uLp6ckTTzzBrbfeiqenJ3PnzmXQoEEXPGdQkDdms2vfyIaF+bn0fL2du46n0+mkqK6EfUUpHDyRxqCgfiwdtajLfvsOHsRgMDA7YQpB1osfi7CwYefYfvocdruDv61N5aNtOfj7ePCTZeNJjA299Ivpxdz1/uwpGk/X6q7xdPuggpfnqSUllakgIiLizpxOZ8f/r6+v5y9/+Qvr16/H19eXZcuWkZmZSVxc3HnPUVXV6NI+hYX5UVZW59Jz9mbuMJ7lTRVsKdhBi721Y5vdaSe7OoeypoqObamlhxnsM4TBgaeDYcUNpWRVHmd48DBs9UbK6q9sLM4cz8bmNl5ek0ZaTiVRoT58d/EIwvw9r/vxvprc4f68lmg8XetKx/N8AQn3Dyp4nCrUqEwFERERdxIeHk55eXnHz6WlpYSFtad2Z2dn069fP4KDgwEYN24cqampFwwqiHSXVnsbn+du4rO8zdgcXV92eZo8GBWWRGJoPH4WH14+9Corj37Ej8Y9gdHQXgZtV/E+AG7oO/a8bTkcTtZsy6GyrrnT9tioAKaMiMRo7DxtorC8gRc+OERJVRMjYkP41p0JHdm+IiIX4vZ/LawdhRqVqSAiIuJOJk+ezPPPP8+SJUtIS0sjPDwcX19fAKKiosjOzqa5uRkvLy9SU1OZOnVqD/dYequU8nTeO7KWiuZKAjz8WDB4LgP8ok/vYDAQ7BWExXj6q/nY8JHsK01mX0ky4yNG43A62F28H6vZixGhCWdp5bT045V8uON4l+3bU4rZmnyCb8yOo194++/KvsOl/G1dBi2tdmbf2J9FU2K7BB1ERM7H7YMKFrMRk9GgJSVFRETczJgxY0hISGDJkiUYDAZ+/vOfs3LlSvz8/Ljtttt4+OGHeeCBBzCZTIwePZpx48b1dJelF9pbfIBX09/CaDAyo/8U5gy8FS+z1wWPuzN2NsllqazJ/oRRYYlk1eRQ3VLD5MgJeJgs5z12Z1oxAN9ZlER0WHvwoNXm4KMdx9mVXsLTK/Ywc0J/fHw8eG/jUTwtJh5fkMj4uPArv2AR6XXcPqhgMBjw8jDRrOkPIiIibudHP/pRp5/PnN6wZMkSlixZcrW7JNJJSkUGAN8f8zgxAQMu+rhQazBT+01mY96XbCrYRlF9e6DghojzB8daWu3sP1JOWKAXowaHdloh4lt3JjApMYLXPz3Mx1/lAhAeaOXJM4IPIiKXyu2DCtBeV0HTH0RERETkajtem4+P2ZtB/v0v+dhZA2bwVdFePj2+CbvTTpg15IKBiQNHy2hps3Pj8H5nXXIyKSaEXzx8Ax/tPE6zzcGCyQPx8Tp/5oOIyPkYe7oDV4PV06QlJUVERETkqmpoa6S8qYL+/tFnfcC/EG+LldmDbqXZ3kybo40bIsZe8Dw700oAuDGhzzn38fQwsWhqLP+2ZIwCCiJyxXpFUMHL00xTq63TUlMiIiIiIt0ptzYfgAH+/S77HDdH3Ui4NRQDBiZEjDnvvrUNraTlVDKorx99Q3wuu00RkUvRS6Y/mHA62wvUeFpMPd0dEREREekFcmsLADqv9HCJzEYzT4x6mMrmakKswefdd3dGCQ6nkxuHR1x2eyIil6pXBBWsHu2X2dxiU1BBRERERK6K3Lo84MoyFQBCrSGEWkMuuN/OtBKMBgMThp976oOIiKv1iukPVs/2QEKTVoAQERERkavA6XSSW1tAoGcAAZ7+3d5eSWUjOSdqGT4wiAAfj25vT0TklF4RVPA6manQpBUgREREROQqqG6poba1joFXmKUA8PnefJ7/4BBttnO/INuZ1r7k5MQETX0QkaurlwQV2jMVmpWpICIiIiJXQUeRRr8rCyo0NttY+eUxDhwtZ9XWnLPu43Q6+Sq9BA+LkdFDQ6+oPRGRS9UrggpWz9M1FUREREREultu3ckijVeYqbDtUBEtrXYMBvh0dx7ZhTVd9jmUXUFpVRNjhoR1ZOiKiFwtvSqo0NSqoIKIiIiIdL/jJzMV+vtHXfY5HA4nG/YV4GE28sRdSTidsHxdBq1tp7Nvj+RX8/KaVMwmAzPGXf4qEyIil6tXBBU0/UFERERErhaH00FebQF9vMOwmq2XfZ6DWeWU1zQzMTGCMUPDuHVsNMWVjaw+OQ0iu6iGP76XjN3u5PEFicRGBrjqEkRELlqvyI9SoUYRERERuVpKG8tptjczwn/4FZ1nw972bIdbx7ZnICyaGsuh7Ao+3Z1HWKAX7285RkubncfnJzJ6SNgV91tE5HL0ikyFU0tKKlNBRERERLqbK4o05pXUkZlXTcLAIKLCfAHw9DDx0Nx4AF7/7AjNLTYemTeccXHhV95pEZHL1DuCCspUEBEREZGrxBVFGjfsbT/HreM6n2Nov0Bm3tAfgwG+MSeOG7WEpIj0MLec/uBwOthXkkysIYoARwjeXu2XeSi7gvTjlQwfGNzDPRQRERERd5Vbm4/RYCTat+9lHV/b0MpX6cX0CbKSFBvS5fN7bhnMvIkD8PayXGlXRUSumFsGFfLrClmR/hakg5fJiyFBMSRNCCR1fyu/e/sgY4eF8bVbBhMaePmFc0RERERE/pXNYaOgvoho375YTJf30L/5YCE2u5Nbx/XDaDCcdR8FFETkWuGWQYX+ftE8PuJBshqyOFCUTkp5OgAhE3yxnhjPvsNlHMquYMHNg5g1oT+Gc/yxFhEREZHerbSxjCCvICzGi/vaXFRfjM1ho/9lTn2oqW9hw94CrJ4mJiVqaoOIXPvcMqhgMBhIDI3nlvgJlJXVUd5Uye7ifXxyfCNN4VuYNGgSqbuDeG9TNlW1LSy5dcg5o8AiIiIi0jsdrsziuYN/JdwayuKh80kIGXbBY3LrLr9Io8PpZPm6DOqb2rh3xhCsnm75VV1E3Eyv+EsVag1mzqDbGBo0mL+nvsmB+m0MmziMkuShbNhXQENzGw/Oicds6hV1K0VERETkImwq2ApAWVMFLyUvZ2RoAouG3IGfhx9HqrJIqcggtTyD6paaLscOvIxMhQ17C0jNqSQxJphbx0Vfcf9FRK6GXhFUOGVw4CB+MuHfWJH2FplVh4kYXk2M+SZ2ppXQ1GLnsfkJeFhMPd1NEREREelhFU2VpJZnMsC/H1+PW8w7h1eRXJ5GeuVhwECbow0Ab7OVwYGDMHA66zXcO4wIn7Mv8+hwONmVXkJsdADhZ9T3yiup4/3NWfh5W3h47nBNzxWR60avCioA+Hn48sSoh/ln5gfsPLGHmyecwHpwAAezyvnje8n8290jFVgQERER6eW2Fn6FEydToiYS5duX7495nD0lB1h37DNMRjNJofEkhQ5nkH9/TMaL/+64M62Y5esyMBoMTEqMYN6kAQT4evKXtWnY7E4enhtPgI9HN16ZiIhr9bqgAoDRYOSeofM5VpPL1qIdPDY9Dq9tZvYdKePPa9J4YmEiJqOmQoiIiIj0Rm32Nnac2I2PxZux4SOB9ppdEyLGMCFizBWd+6v0EgDCgqxsSznBjtRiIkO9OVHRyIyx0YyIDb3i/ouIXE299snZw+TBNxKWYDKY+OeR97hv9kDiBwRxMKuc1z89gtPp7OkuioiIiEgP2F96iIa2Rib1nXDZy0KeTU1DK+nHK4mJ9OdX37yBx+YnEBHiTUFZA1FhPtxzS6zL2hIRuVp6ZabCKf39opkXcztrsj/hvaxVPHHXvfzvWwf4MrmIQF8PFtwc09NdFBEREZGrbEvhDgwYuCnqRpeed29mKU4n3BDfB6PRwIT4PoyLCycjt4p+Yb5YzJqCKyLXn16bqXDKrf2nMjhwEMllqRysPMD37xlFWKAXa7cfZ9OBwp7unoiIiIhcRbm1+eTW5pMQEkeoNdil5/4qvRiDAcbHny7iaDQYSBgYjL/qKIjIdarXBxWMBiPLhi/BavbivSNraHBW8oOvjcLP28Ibnx7m4NHynu6iiIiIiFwlXxbuBGBK9CSXnresuonswlri+gcR6Ovp0nOLiPSkXh9UAAj2CuLrcXfT6mjjrymv4edr4N/uHonFbOQva9PIK6nr6S6KiIiISDerb2tgX8lBwqwhxAcPcem5d2e0F2i8cXgfl55XRKSnKahw0ujwJG4fcAtlTRWsSH+LARG+PHLHcFra7Pzp/UNU1bX0dBdFREREpBttzPuSNoeNm6MmYjS49mvyrvQSzCYDY4eFufS8IiI9TUGFM9wRM5P44KGkVWTycc4Gxg4L5+5psVTVtfDcB4doabX3dBdFREREpBvk1uazIW8LwV5BTI6c4NJzF5TVU1DWQFJMCN5erltNQkTkWqCgwhmMBiPfSLiXEK8gPjm+gUNlacy6oT83jehLbnEdf/0wDYeWmhQRERFxK20OG69nvIvD6eDrcYvxMntd9rm2p5zgD+8mczivqmPbrvSTUx8SIq64ryIi1xoFFf6Fr8WHR5KWYTFaeC39bapaqnlg5jDi+gdy4Gg5q7fm9HQXRURERMSFPsnZwImGEm6KupG4K6il4HA4+WBLNinHKnjmnwd4cWUKJVWN7EovwdPDxMjYEBf2WkTk2qCgwln084vk7iF30mxvYf3xjZhNRr59VxKhAV58tOM4B7O0IoSIiIiIO8itzefzvM0EewVxV+ycKzpX2vFKqutbGREbwuCoAPYdKeOnf91FeU0zY4aE4WExuajXIiLXDgUVzuHGvuPo4x3GzhN7KW+qwNdq4Ym7krCYjbzyYTqlVY093UURERERuQJnTnu4P+7uK5r2AO1THwDumDSQn9w/hsfmJxDk17585OSk/8/efYdHdd55/3+fmVHvvTcEAiTRRO8Ygwu2Y2dtx7gkccjjJM9uduNks082eZ5N1vtz22ycsptk7STrbNax40pcARs72DTRhEACCSRAHfXey8z8/sBWrNBUZnQk8Xldl66LOfc593w4cInhq3N/by19EJGpSUWFy7BarGxK3oDD6WB7yQcAJEUH8IUbZ9LdO8Av/niC3n41bhQRERGZzzT+UQAAIABJREFUrHaW7aK6s5bVccuZGTp9THN19QyQW9xAVKgv02IDMQyDJbOjePwrS3n0fy0lPTnURalFRCYWFRWuICtqHjF+URysyaG2qx6AlXNiWDc/loq6Dv5nx2mcatwoIiIiMun02/v5qHI/fjZf7ki9eczzHT5VS/+Ag5WZ0RiGMXjcw2YlNtxvzPOLiExUKipcgcWwcEvKDThxsq1k5+DxezekkRITQPbJGvbmVZuYUERERERG42hdHh39nSyPXTzmZQ8A+07UYAArMrXMQUSuLSoqXMW8iAzi/WPJqT3O+Y4aADxsFv76jjl42iy8tb8Uh0NPK4iIiIhMJrursjEwWB23fMxz1TZ3caaylVlJIYQGjr1AISIymaiocBUWw8Kt0z5+WqH0/cHjYUHerMiMpqG1h9ziehMTioiIiMhIlLdVUtpWTkbYTMJ9xt7rYH/+hR88qRmjiFyLVFQYhsyw2SQFJJBbl0dF+/nB4xsWJQCw83CFWdFEREREZIQ+qtoPwJr4FWOey+F0sv9EDV6eVhamRY55PhGRyUZFhWEwDIPbUm8E4MXTW3E4HQDEhvuROS2UospWSmvazIwoIiIiIsPQ0d9JTu0xwn3CmB2aNub5ispbaGzrYdHMCLw8rS5IKCIyuaioMEyzQ9NYGDmP0rZydldmDx6/QU8riIiIiEwaB6qP0O8YYHXcMizG2D8K7ztxoWn3ysyYMc8lIjIZqagwAnelfQZfmw9vnttOU08zABkpocSE+XKosI7m9l6TE4qIiIjI5TicDvZUHcDDYmN5zOIxz3eipJH9J2qIDPYhLTHYBQlFRCYfFRVGINAzgL+acRu99j5ePP1HnE4nhmGwcXECdoeTXblVZkcUERERkcsobCqiobuRRVEL8PPwHdNctc1dPP36SawWg4c+k47FMFyUUkRkclFRYYSWRS9kZsh0TjaeIqfuOADLM6Lx87bxYW4Vff12kxOKiIiIyF/qs/exo/RPAKyJH9s2kt29A/z7q3l09Q7wxZtmkRob5IqIIiKTkooKI2QYBvfOvBMPi41Xit6go78TLw8r6xbE0dHdz4GCWrMjioiIiMindPR38u+5v+ZcaynzwjNIDIgf9VwOp5Nfv1VAdWMXGxbFs3KOeimIyLVNRYVRiPAN45aUG+jo72RbyU4A1mfFY7UYvJNdSq+eVhARERGZEBq7m/lxzn9S0lbGoqj5bMm8f0zzvbGnhGNnGpidFMI966e7KKWIyOSlosIorU9YTYhXMNnVR+ge6CYkwIuNixKob+nh9T3nzI4nIiIics3p6u+iobtp8OtMSwlP5fyC2q46rk9YwxfTN2Oz2EY9f01TF2/vLyU8yJv/fUcmVos+SouIjP676jXOarGyJn45b5zdTnb1EdYnrOb21SkcLa7nvcMVLJ4VxbTYQLNjioiIiFwT9p0/yIun/4jD6bho7LPTb2FD4toxv8cHOZU4gbvWpeLv4zHm+UREpgKVV8dgRewSPCw2PqrYh8PpwMvDypdunoXTCb/dVsiA/eJ/1ERERETEdZxOJ9tL3ueFU6/hY/NmWfSiP3/FLOJrcx90SUGhu3eAvfnVhAR4kZUW4YLkIiJTg55UGAN/Dz8WRy1gf/VhTjaeYk54OjMTQ1i3II4Pc6t4e38pd6yeZnZMERERkSnJ4XTwUtHr7K06QJh3CH8z78tE+UW65b325lfT22fnlmVJ2Kz6uZyIyCf0HXGM1savBODDin2Dx+5el0pIgBfvZJdRWddhVjQRERGRKavf3s9v8p9jb9UB4vxj+PuFf+O2goLD6eSDnEpsVgtr5se65T1ERCYrtxUVuru7+cY3vsEDDzzA3Xffza5du9z1VqaKD4hlRvA0TjUXU9N5YTtJHy8bX7xpJnaHk99uL8ThdJqcUkRERGRq2VOVzfGGk6SFTOebWV8jyMt9vaxOnGukrrmbZelRBPp6uu19REQmI7cVFXbt2kVmZia///3v+elPf8qTTz7prrcy3SdPK3xUuX/w2NzUcJbMjqSkup2jp+vNiiYiIiIyJR2pO47FsLAl4z58bD5ufa/3j1QCcP3CeLe+j4jIZOS2osKmTZt46KGHAKiuriYqKspdb2W6ueHphHgFc6Amh+6B7sHjd6yehmHAG3tL9LSCiIiIiIs0djdR1lZBWnAqAZ7+Lpu3tbOPN/eVUFn/5+Wr1Y2dnChpIi0+iKToAJe9l4jIVOH2Ro2bN2+mpqaGp59++ornhYT4YrNZXf7+ERHj883/5pnreCHvdfLb8rll5vWD733dwgT+dKSCoqp2Vi+IG5cs7jJe9/JaofvpWrqfrqX76Vq6nyKulVufD0BW5FyXzvuH94s4VFjH63tKmJcaxs3LkjhUeGF56/WLElz6XiIiU4XbiwovvvgihYWF/MM//ANvvvkmhmFc8rzm5i6Xv3dERAD19e0un/dS5gXN4xXL27xZ+D4LgrPwsFy4tRsXxvFhTiXPbS8gLTYAi+XSv/+Jbjzv5bVA99O1dD9dS/fTtcZ6P1WQELnY0bo8LIaFeRGZLpuzsr6Dw4V1xIT54ufjwfGzjRw/24gBhAR4sWBGuMveS0RkKnHb8ocTJ05QXV0NwOzZs7Hb7TQ1Nbnr7Uzn7+HH6rjlNPe2sP/8ocHjUSG+rMiMprqxa7DSLSIiIiKj8+mlD/6efi6b9429JTiBz103ne89sJB/vD+LealhOIGblyZqG0kRkctw23fHI0eO8OyzzwLQ0NBAV1cXISEh7nq7CeGGpOvwtHqyo/QD+ux9g8dvXZmM1WLwxr5S7A6HiQlFREREJjd3LH0or20n53Q9KTGBzE0NAyAtIZhv3D2PX3xzjRo0iohcgduKCps3b6apqYn77ruPr3zlK3z/+9/HYpnaFd4AT3/Wx6+ira99yE4QkcE+rJwTTW1TFwcL9LSCiIiIyGgdrXX90ofX95QA8NnVKRct1fXxsl12+a6IiLixp4K3tzdPPfWUu6afsK5PXMNHVfvZWfYhq+KW4WPzBuDW5cnsy6/hzX2lLE2PwjrFCywiIiIirlbX2UhZewWzQma4bOlDSXUbx840MD0+iIyUUJfMKSJyLdH/bF3M18OXDYlr6Rzo4k8VewaPhwf7sHpeLHXN3ezLrzExoYiIiMjkdKAiB4CsKNctfXhj78dPKay6+CkFERG5OhUV3GBd/Cr8Pfz4U/keOvo7B4/ftiIZT5uF1/eco7ffbmJCERERkcknu+LohaUP4a5Z+nCmqpW8s43MSgxmdrKeUhARGQ0VFdzA2+bFjUnX0WPv4f2yjwaPhwR4sWFRAi0dfXyQU2liQhEREZHJpaG7ibNNZcwMme6ypQ/bD5QBcPuqFJfMJyJyLVJRwU1Wxy0n2CuIDyv3UttZN3h807JE/LxtbMsuo6O738SEIiIiIpNHbl0eAAsi57hkvrauPvLONpIQ6c/MxKm9Q5mIiDupqOAmHlYP7pxxG/2OAX5X+BJ2x4XlDr7eHtyyPJmu3gG2fVwdFxEREZHLczgd7D9/CA+LzWW7Phw8WYvd4WTVnBiXzCcicq1SUcGNsiLnsjhqAWVtFbxX9uHg8esXxhES4MUHOZU0tfWYF1BERERkEjjddIa67gaWJy7E38M1Sx/25ldjtRgszYhyyXwiItcqFRXc7HNpdxDsFcS20p2Ut13oo+Bhs3LH6hT6BxyDHYdFRERk5B5//HHuueceNm/eTF5e3pCx6upq7r33Xu666y6+//3vm5RQXOGjqv0A3DR9nUvmK69tp6Kug7mpYQT6erpkThGRa5WKCm7m6+HDA7PvxuF08LuCF+mzX+ijsDIzhthwP/bmV1NV32FyShERkcnn0KFDlJWV8dJLL/HYY4/x2GOPDRl/8skn2bJlC6+++ipWq5Xz58+blFTGorG7mRMNhSQGxDM9LNklc+7NrwbQ0gcRERdQUWEczA5NY238Cmq66njr3A4ALBaDu9am4nTCC+8X43Q6TU4pIiIyuWRnZ7NhwwYAUlNTaW1tpaPjQqHe4XCQk5PD+vXrAfjBD35AbGysaVll9PaeP4ATJ2viV7hkvgG7gwMnawnw9WBOaphL5hQRuZbZzA5wrbgjdROFTUXsqtjL0uiFxAfEMm96GHNTw8g728ihwjqWpmtNn4iIyHA1NDSQkZEx+Do0NJT6+nr8/f1pamrCz8+PJ554gpMnT7Jo0SL+/u///qpzhoT4YrNZXZozIiLApfNdS/rt/RzYdxh/Tz9uSl8JjP1+ZudX09Hdz+1rUomJDnJFzElNfz9dS/fTtXQ/Xctd91NFhXHiafXkzum38Z95v+WDit18MX0zhmFw34YZFJQ28+KfipmbGoaPl/5IRERERuPTT/05nU5qa2v5whe+QFxcHF/5ylf48MMPWbdu3RXnaG7ucmmmiIgA6uvbXTrnteRQzVHaeju4PnENrc29RER4jvl+btt7DoCs6WHX/J+N/n66lu6na+l+utZY7+eVChJa/jCOMsJmEe0XxZHaYzT3tAAQGeLLLcuTaO3oU9NGERGREYiMjKShoWHwdV1dHREREQCEhIQQGxtLYmIiVquV5cuXU1xcbFZUGaXdlfsxMFgTt9wl87V29pF3tpHEKH8SIv1dMqeIyLVORYVxZBgG1yesweF08GHlvsHjm5YlEhnsw/tHKqmsU9NGERGR4Vi5ciXvvvsuACdPniQyMhJ//wv/UbTZbCQkJFBaWjo4npKSYlZUGYXy9kpK2spJD5tJuI9reh8cPFmDw+lkpRo0ioi4jIoK42xx9AICPP3ZW3WQ7oEe4MIWk/dtTMPhdPL7906raaOIiMgwZGVlkZGRwebNm3n00Uf5wQ9+wNatW9m5cycA3/ve9/jud7/L5s2bCQgIGGzaKJPD7spsAJc9peBwOPno+HmsFoNl6mMlIuIyWsA/zjwsNtbFr+KtczvIPn+I9YlrAJibGkZWWgRHi+o5cLKW5ZnRJicVERGZ+L797W8PeT1r1qzBXyclJfGHP/xhvCOJC7T0tnK45ijhPmGkh810yZwHC2qpbuxi1dwYAnw9XTKniIjoSQVTrI5bhqfFgz9V7MXusA8e33z9dKwWg7ezS3HoaQURERG5Rn1QvpsBp50bEtdhMcb+cXXA7uD1veewWgw+szJ57AFFRGSQigom8PPwZXnsYpp7W8ityxs8Hh7kw9L0KKobu8g/22hiQhERERFzdPR3srfqAMFeQSyJWeiSOffmV1Pf0sPa+bGEB/m4ZE4REblARQWTXBe/GgOD9yt2D+mhcMPiBADeO1xhVjQRERER03xYsY8+Rz/XJ67BwzL2lbr9A3be2leKh83CrSuSxx5QRESGUFHBJBG+YcyPyKSivYqi5rODxxOjAkhPDqGwrJnyWu3LKiIiIteOnoEePqzch5+HLytjl7pkzg9zz9Pc3sv1WfEE+3u5ZE4REfkzFRVMtDFpHQBvnXv3L55WSATg3UPlZsQSERERMcWeqgN0D3RzXfxqvKxjb6bY22fnnexSvD2t3LwscewBRUTkIioqmCgpMIF5EZmUtJVxorFw8PicaaHEhvtxqLCO5vZeExOKiIiIjI9+ez8fVOzG2+rF2njXbCP5fk4FbV393LA4QTs+iIi4iYoKJrtt2o0YGLx5dgcOpwMAwzC4YXECdoeT93PUW0FERESmvuzqI7T3dbAmfgW+Hr5jnq+7d4AdB8vx87YNPgUqIiKup6KCyWL8olgSncX5zhqO1B4bPL48I4pAXw8+yj1PT9+AiQlFRERE3MvpdPJ++Ud4WGxcl7DKJXN+mFtFZ88ANyxOwNd77A0fRUTk0lRUmABuSdmI1bDyzrn3GHBcKCB42Kysz4qnq3eAPXnVJicUERERcZ+6rnoae5qYG55BoGfAmOfr67fz7uEKvD2tXL8w3gUJRUTkclRUmADCfEJZFbeMhp4m9p8/PHh8XVYcnjYLb+4toa6l28SEIiIiIu5zprUEgOnB01wy3978ato6+1ifFY+vt4dL5hQRkUtTUWGCuCl5PZ5WT3aUvk+fvQ+AQF9P7t+YRmfPAD9/LZ/ePrvJKUVERERc72xLKQDTg1PGPNeA3cH2A+V42CxsXJww5vlEROTKVFSYIAI9A1gfv4rWvnY+KN89eHz1vFiuWxBHZX0H/73j1JCtJ0VERESmgjMtJfjafIj2ixzzXAcLamls62HN3FiC/LTjg4iIu6moMIFsSFpLoGcAO0o/oKazdvD4vRtmMD0uiIMFtbx3WLtBiIiIyNTR0ttKY08TqcHJWIyxfTR1OJ1sO1CG1WJw01Lt+CAiMh5UVJhAfGw+bJ75WQacdn5f+MrgFpM2q4W//mwmQf6evLLrLIWlTSYnFREREXGNMy0X+imkBo196UNuUT3VjV0sz4gmLMh7zPOJiMjVqagwwcyLyGRh5DxK2sr5U8WewePB/l78zR1zMAz41VsF9PWrv4KIiIhMfp8UFcbaT8HpdPJ2dhkGcPMyPaUgIjJeVFSYgD6Xdgf+Hn68fe5darvqB49Pjw/ipqWJtHb2sf9EjYkJRUREXEs9g65dZ1tK8LB4kBAQN+o5nE4nr+8poaymnYUzI4gJ83NhQhERuRIVFSYgf08/7pn5WfodAzz/qWUQABsWxmOzGuw4VI7DoQ9gIiIyNVx33XX85Cc/oaJCvYOuJZ39XZzvrCElKAmbxTaqOT4pKLy1v5TIYB/u3ZDm4pQiInIlKipMUFmRc5kfMYezraXsrswePB7k78WKzGjqmrvJLa6/wgwiIiKTxyuvvEJERATf+973+NKXvsRbb71FX1+f2bHEzc61lgIwPSh5VNc7nU7++KmCwv+5bwEhAV6uCygiIlelosIEds/MO/CxebOj9AMGHAODx29ckogBbD9YrsdFRURkSoiIiOCBBx7gueee45//+Z/5wx/+wOrVq/nJT35Cb2+v2fHETQabNI6in8InBYW3P1VQCA1Uc0YRkfGmosIEFugZwPKYxbT3d3C8/sTg8ZgwP+bPCOfc+TaKK1tNTCgiIuI6hw8f5rvf/S4PPfQQWVlZvPDCCwQGBvKNb3zD7GjiJmdbSrAYFlKCkkZ8bfbJGhUUREQmABUVJrhVccsA2F2VPeT4J3sv7zhYPu6ZREREXG3jxo384he/YPXq1bzzzjt8+9vfJjU1lS9/+cu0tqqAPhW093UMed1n76OsvZKEgDi8rJ4jmsvucPDmvlKsFoO/3zxfBQURERONriOOjJso3whmhczgVHMx5ztqiPWPBmBGfDCpcYEcO9NAVUMnceHqciwiIpPXb37zG5xOJ8nJyQAUFBSQnp4OwAsvvGBiMnGF/IYCns77b9bELefutNuxGBZK28pxOB1MDxr50ofDhXXUNXezdn4sEcE+bkgsIiLDpScVJoHV8csB2FN1YMjxm5deeFTw3UN6WkFERCa3rVu38swzzwy+/tWvfsWPfvQjAAzDMCuWuEh+QwFw4cnL/zrxPP32/lH3U3A4nbydXYbFMLh52ciXTYiIiGupqDAJzAmbTbBXEIdqcugZ+HOzqvkzwokK9SX7RA1VDZ0mJhQRERmbgwcP8sQTTwy+/ulPf0pOTo6JicSVTjefxdvqzYzgaRyrz+fnx39DQWMRAKnBySOaK7eonvMNnSzPiCJSTymIiJhORYVJwGqxsjJ2CT32Xg7X5g4etxgGd61Nxe5w8vPX8ujq6TcxpYiIyOj19/cP2UKys7OTgYGBK1whk0VTTzMN3Y3MCJnG38z7Mgsi5nCmpYSStjKi/aLw9xj+Ek6n08lb+0sxgE3L9ZSCiMhEoKLCJLEidgkWw8Kequwh20gunBnBzcsSqW3u5ldvFeDQFpMiIjIJbd68mU2bNvHwww/zd3/3d9x6661s3rzZ7FjiAkXNZwFIC0nFw+rBlsz7WRu/AoCZIakjmivnVB3ltR0snh1JTJj6SYmITARq1DhJBHsFMS88g9z6fEraypgWlDw4dueaVMprO8g728ibe0u4Y/U084KKiIiMwt13383KlSvJz8/HMAy++93v4u/vb3YscYFPigozQ6YDYDEs3D3jdrIi5xH3cQPq4XA6nby48zQAtyxPdnlOEREZnRE/qdDX10d1dbU7sshVrI670LBxd+XQho0Wi8FXP5NBeJA3b+4rJbeo3ox4IiIiY9LV1UVoaCghISGcO3eOz33uc2ZHkjFyOp2cbj6Dv4cfMX5Rg8cNw2B6cAo+tuH3RDhV1szpsmYWzAgnIVIFJxGRiWJYTyo888wz+Pr6ctddd3HnnXfi5+fHypUrefjhh92dTz4lLSSVKN9IjtYd5/bUmwjxDh4c8/fx4Ot/NYfHn8vh128X8MiWJdpiSUREJo1HH32Uffv20dDQQGJiIhUVFWzZssXsWDJG9d2NtPS2siByLhZjbKtu384uA+DWFckuSCYiIq4yrO/uu3bt4oEHHmDHjh1cd911vPLKKxw9etTd2eQvGIbBxqR12J12dpZ/eNF4YlQAm6+fQU+fnf0nasY/oIiIyCjl5+ezfft2Zs2axWuvvcazzz5Ld3e32bFkjIqazwAj753wl0qq2ygsa2Z+WgQpMYGuiCYiIi4yrKKCzWbDMAx2797Nhg0bAHA4HG4NJpe2JGoBYd6h7Dt/iJbe1ovGl6ZHYbUYHCtuMCGdiIjI6Hh6egIXdoFwOp1kZmbqBxhTwGCTxuCxFRW2ffyUwl3rZ4w5k4iIuNawigoBAQF85Stf4ezZsyxYsIBdu3ZhGIa7s8klWC1Wbky+jgHHAO+XfXTRuI+XjdlJIZTVttPU1mNCQhERkZFLSUnh+eefZ9GiRXzpS1/ikUceob293exYMgZOp5Oi5rMEeQYS6Rsx6nmqGzs5WlRPSkwAc6eHuzChiIi4wrB6Kjz11FPs37+frKwsALy8vPjXf/1XtwaTy1savZDtJR+w9/wBNiZdR5BXwJDx+TPCOVHSRG5xA9cvjDcppYiIyPA98sgjtLa2EhgYyDvvvENjYyNf/epXzY4lY1DdWUt7fweLo7LG9MOo7QfKcQKbliXph1oiIhPQsJ5UaGpqIiQkhNDQUF5++WXefvttrXM0kc1i48bk6+h3DPBB+cVPK8z/uIp/7IyWQIiIyOTw+OOPExwcjMVi4bbbbuPBBx8kOnr42w3KxPPnrSRHv/Shqa2H7JM1RIf6siBt9E87iIiI+wyrqPDd734XDw8PCgoKeOWVV7jxxht59NFH3Z1NrmBZzGKCvYLYU5VNe1/HkLHQQG+SogI4VdZMV8+ASQlFRESGz2q1kp2dTW9vLw6HY/BLJq9PmjSmjaGo8N7hCuwOJzcvS8SipxRERCakYRUVDMNg7ty57Ny5k/vvv5+1a9fidDrdnU2uwMNi44ak6+hz9PNB+e6LxhfMCMfucHKipNGEdCIiIiPzyiuvsGXLFubNm0d6ejrp6elkZGSYHUtGyeF0UNxyjjDvUMJ8Qkc1R0d3Px8dO09IgBfLM/TUiojIRDWsokJXVxd5eXm8++67rFmzhr6+Ptra2tydTa5iRcxigjwD+KhqP809LUPG5s/4eAmEdoEQEZFJICcnh8LCQk6dOjX4VVhYaHYsGaXKjvN0DXSPaenDBzmV9PbbuXFxAjbrsD6yioiICYbVqHHLli380z/9E/fccw+hoaE89dRT3Hrrre7OJlfhYfXg1mk38vypV/mvE7/n4ayvYbNc+CNNiPQnLNCLvLONDNgd+sdYREQmtJ/97GeXPP6Nb3xjnJOIKwxuJRkyfVTXOxxOPjxWha+XjTXzY10ZTUREXGxY/9PctGkTb7zxBrfffjutra1861vfYsuWLe7OJsOwPGYxi6LmU9JWzutntg0eNwyD+dMj6OodoKii5QoziIiImM9qtQ5+ORwODh48qC0lJ7HCxiJg9P0UCsqaaO3oY0l6FN6ew/oZmIiImGRY36VzcnL4zne+Q2dnJw6Hg5CQEP7t3/6NOXPmuDufXIVhGNw7804qO6rZVbmXlKAkFkbNA2B+WjgfHK3kWHED6cmjW88oIiIyHr7+9a8PeW232/nbv/1bk9LIWPTZ+zjTWkKcfwxBXoGjmiP7RA0AyzOiXBlNRETcYFhPKvz4xz/ml7/8JdnZ2Rw8eJAf//jHPPnkk+7OJsPkbfPiocwH8LR68vypV6jtrANgZkIwPl42cosb1FhTREQmlYGBAcrLy82OIaNQ3FLCgGOA9NCZo7q+p2+AnKJ6IoK9mR4X5OJ0IiLiasMqKlgsFtLS0gZfp6enY7Va3RZKRi7aL4r7Z91Fr72PX594jl57HzarhTnTQmls66GiruPqk4iIiJhk7dq1rFu3bvBr2bJlLFmyxOxYMgqFTacBmB2adpUzLy23qIG+fgfLM6IxtI2kiMiEN6zlDxaLhXfffZeVK1cCsHv3bhUVJqBFUfM521LK7qr9ZFcfZl38ShbMiOBQYR2vfniWr92ega+3h9kxRURELvLCCy8M/towDPz9/QkMHN2j82KugsYiPC0eTAtOHtX12Sc/WfqgbSRFRCaDYT2p8Mgjj/Dyyy+zfv16rr/+el5//XX+5V/+xd3ZZBRuSl6PgcGRmmMAZKVFMDsphBMlTfzL745QqScWRERkAuru7ubFF18kLi6O2NhYnnjiCYqLi82OJSPU1NNMbVcdaSHT8bCMvMFiS0cvJ0ubmBYbSFSorxsSioiIq12xqHDfffdx//3383//7/+lu7ub6dOnk5qaSkdHB//4j/84XhllBIK8ApkZMp2StjIauhvxsFn41j3z2LQsibrmbh79nyODPwEQERGZKB555BHWrl07+PrOO+/UDzAmoU92fZgdNrqlDwcLanE69ZSCiMhkcsUS8sMPPzxeOcSFFkXN51RzMUdqj3FT8vVYLRbuWpfKtNhA/uudAn79VgHN7b1sWpZkdlQRERHgwm4PixYtGny9aNEiNRmehApnL1dgAAAgAElEQVSaLhQV0kfZTyH7ZA1Wi8GS2ZGujCUiIm50xaKCGiRNTvMjM3mx6I8crsnlxqT1g02OstIiiA1fzJO/z2FbdhkbF8XjYVNvDBERMV9AQAAvvPACS5cuxeFwsGfPHvz8/MyOJSNgd9g53VxMmHcoET7hI76+qr6D8toO5k8PJ8DX0w0JRUTEHYbVU0EmFx+bD5lhs6npqqOqo3rIWHSoLyvnxtDVO8DRogaTEoqIiAz1xBNPcPLkSR5++GG+9a1vUVZWxhNPPGF2LBmB0rYKugd6SA+bOapdG/Z/vDxzWUaUq6OJiIgbjbyDjkwKi6Pmc6w+n8O1ucQHxA4ZWzUnhu0HytmbX83SdP3DLSIi5gsNDeWhhx4iOTkZgIKCAkJDQ80NJSMylq0kHU4nB07W4uNlZf70kT/lICIi5tGTClNURtgsfGzeHKk9hsPpGDIWE+bH9LggCkqaaGrrMSmhiIjIn/3kJz/hmWeeGXz9q1/9ih/96EcmJpKRKmgqwmJYSAtJHfG1xRUtNLf3snBmJJ4eWpopIjKZqKgwRXlYPZgfMYeW3lbOtpReNL5qbgxOYF9+9UVjIiIi4+3gwYNDljv89Kc/JScnx8REMhIdfZ2Ut1UyLSgJH5v3iK/PPlkLwHI9QSkiMumoqDCFLYqaD8CR2tyLxhbPisTTZmFffo26a4uIiOn6+/vp6+sbfN3Z2cnAwICJiWQkTjUX48TJ7NCZI762f8BBzuk6gvw9mZkY4oZ0IiLiTuqpMIWlhaQS6BlAbl0+d6fdjs3y5z9uHy8bi2ZFsv9EDUUVLfpHXERETLV582Y2bdpEZmYmDoeD/Px8vvjFL5odS4apsPHjrSTDRt5P4cS5Rjp7BrhhcQIWy8gbPIqIiLlUVJjCLIaFhVHz2FWxl4LG08yNyBgyvmpODPtP1LA3v1pFBRERMdXdd99NcnIyzc3NGIbB+vXreeaZZ3jwwQfNjiZX4XA6ONl4igAPf+L9Y69+wV84UHBh6YN2fRARmZxUVJjilkYvZFfFXraeeZsZIalD1jmmJQYTHuTN4VN13LchDR8v/XUQERFzPPbYY+zdu5eGhgYSExOpqKhgy5YtZseSYShrq6C9v4MVMYuxGCNbWdvdO8CxMw1Eh/qSFBXgpoQiIuJO6qkwxSUExLExcR313Y08f+rVIf0TLIbBqrkx9PU7OHKqzsSUIiJyrcvLy2P79u3MmjWL1157jWeffZbu7m6zY8kw5DUUADAnPH3E1x4tqqd/wMGy9CgMQ0sfREQmIxUVrgG3TbuR1KAUcuvy+Khq/5CxlZkxGMCePO0CISIi5vH09AQuNGx0Op1kZmZy9OhRk1PJcOQ3FOBhsTErdMaIr/1k6cNSLX0QEZm0VFS4BlgtVrZk3oe/hx9bi9+mrK1icCwsyJs5qWGcqWrldHmziSlFRORalpKSwvPPP8+iRYv40pe+xCOPPEJ7e7vZseQqGrobqe6sZWbIDDytniO6trWjl4LSJlJiAokK8XVTQhERcTcVFa4RwV5BPJhxLw6ng9+c+D1d/V2DY7etSAbgzX2l5oQTEZFr3iOPPMItt9zCt771Le68806SkpJ4+umnzY4lV5HfUAjA3FEsfTh0qg6nUw0aRUQmO7cWFX74wx9yzz33cOedd/Lee++5861kGGaHpnFz8vU09TTzUtHrg8dT44LInBZKYVmznlYQERFTGIZBcHAwFouF2267jQcffJDo6GizY8lVfNJPITN89oivPXCyFsOAJbMiXR1LRETGkduKCgcOHKC4uJiXXnqJ3/zmNzz++OPueisZgZtTNpAYEEdO7XHOd9QMHr99ZQoAb+wtMSuaiIiITCJd/d2caTlHUkACQV6BI7o2t7iekuo20pNDCfL3clNCEREZD24rKixevJif/exnAAQGBtLd3Y3dbnfX28kwWQwLt6TcgBMn20p2Dh7/5GmFU+UtelpBRERErqqg6TQOp2PEuz7kFtfzyz+ewNPDwu2rUtyUTkRExovNXRNbrVZ8fS803Xn11VdZs2YNVqv1sueHhPhis11+fLQiIrTn8V9aF76YnZW7yK3Pp8ujlaTgeAC+eGsG//Dve9h+qIJVCxMvuk730rV0P11L99O1dD9dS/dTpqL8j5c+zI0YflEht6ieX75+AqvV4Jt3z2N6XJC74omIyDhxW1HhE++//z6vvvoqzz777BXPa27uuuL4aEREBFBfr87Rl3JDwnrOND3L80ff4KE5XwAgzNeDzGmh5J1pYG9OOTMTQwbP1710Ld1P19L9dC3dT9ca6/1UQUImIrvDzsnGU4R4BRPrN7zeF39ZUPj05wwREZm83Nqocc+ePTz99NP8+te/JiBAH4omkvTQmSQHJnKs/gQV7ecHj6u3goiIiFzNmZYSugd6mBuRjmEYVz2/rKZdBQURkSnKbUWF9vZ2fvjDH/LMM88QHBzsrreRUTIMg00pGwHYfpneCrlF9WbFExERkQksv/HC0ofh9lM4fKoOu8PJlk2zVVAQEZli3FZU2LZtG83NzTz88MN8/vOf5/Of/zznz5+/+oUybtJD00gJTOR4w0kq2qsGj29ePwOrxeD594vo6RswMaGIiMiVPf7449xzzz1s3ryZvLy8S57z1FNP8fnPf36ck01dTqeT/PoCvK1ezAieNqxrTlc0YzEM5kwLc3M6EREZb24rKtxzzz3s3buX5557bvArNjbWXW8no2AYBrek3ADAtpL3B4/Hhvtx87JEmtp6eX2PlkGIiMjEdOjQIcrKynjppZd47LHHeOyxxy4658yZMxw+fNiEdFNXdWctDT1NzA5Nw2a5enuu3n47pdXtJEUH4OPl9nZeIiIyztzaU0EmvlmhM5gWlERew0kqP9Vb4dblyUQG+/D+kUrKa9WwTUREJp7s7Gw2bNgAQGpqKq2trXR0dAw558knn+Sb3/ymGfGmrBMNhQDMjcgY1vlnq1qxO5zMTNByWBGRqUjl4mucYRjclHw9vzz+LO+W/YkvZz4AgKeHlQduTOPHLx3ndztOMz89xuSkIiIiQzU0NJCR8ef/2IaGhlJfX4+/vz8AW7duZcmSJcTFxQ17TndscT3VdvAoPH4ai2Fhbdoi/L38rnp+Zc6FJZaL58S45F5MtftpNt1P19L9dC3dT9dy1/1UUUFID51JQkAcuXX51HTWEe0XCUBmShhLZkdyqLCOHdmlLEkLNzeoiIjIFTidzsFft7S0sHXrVn77299SW1s77DlcvcX1VNuita2vneLGElKDk+luc9DN1X9vuadqMYCoAM8x34updj/NpvvpWrqfrqX76Vru3OJayx/kwtMKSetx4uS9sl1Dxu69fgY+Xjb+Z1sBrR29JiUUERG5WGRkJA0NDYOv6+rqiIiIAODAgQM0NTVx//338/Wvf52TJ0/y+OOPmxV1yjjRcAonzmHv+tA/YOfs+TYSIv3x9fZwczoRETGDigoCXFgXGe0XxeHaXBq6mwaPB/l78VdrptHVM8B7RypMTCgiIjLUypUreffddwE4efIkkZGRg0sfbrrpJrZt28bLL7/Mz3/+czIyMvje975nZtwp4UTDha0k5w6zqFBS3c6A3UFaovopiIhMVSoqCAAWw8JNSetxOB3s/IunFdbMiyHY34sPc8/T3astJkVEZGLIysoiIyODzZs38+ijj/KDH/yArVu3snPnTrOjTUl99n4Km4qI8o0k0jdiWNecLm8GUJNGEZEpTD0VZFBW5FzeKXmPA9VHuDllA8FeQQB42KzcsiqF53ecYk9eNTcsTjA5qYiIyAXf/va3h7yeNWvWRefEx8fz3HPPjVekKauo+Qx9jn7mhM8e/jUVLQDMUFFBRGTK0pMKMshqsXJD0nUMOO28X/7RkLGblyfjabOw83AFdofDpIQiIiJilvyPlz4Mt5/CgN3Bmao2YsP9CPT1dGc0ERExkYoKMsSS6CxCvILZW3WQjv7OweNB/l6snBtDY1sPOafrTUwoIiIi483hdJDfUIifhy/TgpKGdU1ZbTu9/XYtfRARmeJUVJAhbBYb6xJW0u/o53BN7pCxGxYnYADbD5YP2bZLREREpraK9ipa+9rIDJuNxRjex8ei8gtLH9JUVBARmdJUVJCLLI1eiNWwsu/8wSHFg6gQX7LSIiiraR9cIykiIiJTX35DITD8pQ8ApytUVBARuRaoqCAXCfD0Z15EBtWdtZS0lQ8Zu3FpIgA7DpZf6lIRERGZgvIbCrAZVmaHzhjW+Q6Hk+LKFiJDfAgJ8HJzOhERMZOKCnJJK2KXALD//KEhx6fHBTE9LojjZxs539B5qUtFRERkCmnpbaWy4zwzQlLxtnkP65qKug66e9VPQUTkWqCiglzSzJDphHmHkFN7jO6BniFjN338tMJ/vnGCtq4+M+KJiIjIODnbUgpc+GwwXCdLmwAtfRARuRaoqCCXZDEsLI9ZQp+jnyO1x4aMLZgRzoZF8VTVd/KjPxyjo7vfpJQiIiLibiVtZQCkDHPXh9KaNt7cW4KXp5XMlFB3RhMRkQlARQW5rOWxizAwLloCYRgG914/g+uy4qis7+BHL+bS2aPCgoiIyFRU0lqOxbCQGBB/1XOb23v5j9fy6R9w8NXbMgjyVz8FEZGpTkUFuaxgryAyw2dR3l5JaXPFkDHDMLh/Yxpr5sVSXtvBUy8eo0uFBRERkSml395PRXsV8f6xeFo9rnhub7+d/3gtj+b2Xu66LpX5M8LHKaWIiJhJRQW5ohUxFxo2fnBu30VjFsPgCzfNZNWcGEpr2vnVWwVDtqAUERGRya2iowq7037VpQ9Op5PfbiuktKadlXOiuWlJ4jglFBERs6moIFeUETaLIM8A9pQdoucvGjbChcLCgzfPYnZSCHlnG8k5XW9CShEREXGHc60X+ilMC7xykWD7wXIOFdYxIz6IL9w4C8MwxiOeiIhMACoqyBVZLVaWxy6hq7+b/+/gU+yrOojdYR9yjsVi8IUbZ2KzWnjh/SK6ewdMSisiIiKuVNJaDly5SaPD6WTn4Qr8vG38zWfn4GHTx0sRkWuJvuvLVd2UtJ47Zt9IZ38XL5x+jUcPPkVO7fEhSx2iQn25ZXkSLR19/HHPORPTioiIiCs4nU5KWssI9Awg1DvksueVnG+jtbOPBTMiCPTzHMeEIiIyEaioIFflYfXgvrl38Mjy77A6bjkNPU08e/J5dlXuHXLepmVJRIX68kFOJWU17SalFREREVdo7m2hta+NlKCkKy5nOFp8YenjgjQ1ZhQRuRapqCDDFuQVyOaZn+Wflv49nlZP/lS+B4fTMTjuYbPw+RvScDrhdztO4XCoaaOIiMhkVfJxP4WUq/RTOFbcgKfNQnpy6HjEEhGRCUZFBRmxSN8IlkQtoLm3hZONp4aMpSeHsiwjitKadnblVpmUUERERMZqOP0Uqhs7qW7sIiMlFC8P63hFExGRCURFBRmVNfErAPiocv9FY/esn4Gvl42tu8/S0tE73tFERETEBc61lWExLCQGxF/2nGPFDQAsmBExXrFERGSCUVFBRiXOP4bUoGQKm4qo62oYMhbk58md61Lp7rXz8p/OmJRQRERERqvf3k9l+3kS/OPwtHpc9rzc4gYMA+ZNDxvHdCIiMpGoqCCjtiZuOQB7qw5cNLZ2XiwpMQEcKKiloLRpvKOJiIjIGJS3V2F32kkJunw/hdbOPs5WtTIjLogAX+36ICJyrVJRQUZtXuQcAjz8ya4+TJ+9f8iYxWLw+RtnYhjw+/eK6B9wXGYWERERmWhK2j5u0niFfgrHzzTgBBakaemDiMi1TEUFGTUPi40VsUvoGugmp+74RePJ0YGsXxBPTVMX7x4qNyGhiIiIjMafd364fFEht+jjrSRnaCtJEZFrmYoKMiar4pZiYLCnMvuS459dk0Kgnydv7S+lvqV7nNOJiIjISDmdTkpaywjyDCDUO/iS5/T0DXCytJm4cD8iQ3zHOaGIiEwkKirImIR6hzAnPJ2y9grK2iouGvf19uCe9dPpH3Dw/M4inE6nCSlFRERkuJp6WmjtayclKAnDMC55zsmSJgbsDhak6SkFEZFrnYoKMmafNGz845l36Bm4eAvJZelRzE4KIe9sI386WjXe8URERGQEilvOAlfup5CrrSRFRORjKirImM0Mnc6c8NkUt5zjp7lP09rbPmTcMAy+fMtsAnw9ePGDYs5UtpqUVERERK4mty4fgDlhsy853tbVR25xA8H+niRFB4xnNBERmYBUVJAxsxgWHsr8AitiFlPRXsVTOb+gtrNuyDmhgd587fZMHE4nv3w9n9bOPpPSioiIyOV09XdzqqmIOP8YovwiL3nOSx+cobt3gJuXJmG5zPIIERG5dqioIC5htVi5b9ZdbErZSGNPE08d/SXnPu4c/YnZSSHctTaVlo4+nn79BHaHtpkUERGZSPIbChhw2smKnHvJ8YLSJrJP1pAUHcD1C+PHOZ2IiExEKiqIyxiGwS0pG7l/1l10D/Tw82O/vqh5401LE8lKi+B0RQuvfXjOpKQiIiJyKUfr8gBYcImiQv+Anf959zSGAQ/eNAuLRU8piIiIigriBitil/CljPvos/fzy+PPUvOppRCf9FeICvVlx6FyCsuaTUwqIiIin+jq76bwk6UPvhc3YHx7fxl1zd1sXJSgXgoiIjJIRQVxi6zIudw786/o6O/k58d+Q3NPy+CYj5eNr9yWDsCrH57RNpMiIiITQH5DAfbLLH2oauhk24EyQgO9uGN1ignpRERkolJRQdxmZdxSbp92M829LfzHsd/Q0dc5OJYSE8jiWZGUVLeTc7rexJQiIiICcLTuOHDx0geH08lzO05hdzi5f2Ma3p42M+KJiMgEpaKCuNXGpHVcn7CG2q46fnH8N0O2m/zsmmlYDIOtu8+paaOIiIiJLix9KL7k0ofconqKKlvJSotgwYyLl0WIiMi1TUUFcSvDMPjs9FtYEbOY8vYq/vXwzwZ3hYgO9WXNvBhqmrrYl19jclIREZFrV17DyY+XPswbctzhdPL63hIMA+5al2pSOhERmchUVBC3MwyD+2bdxR2pm2jra+enR59mT9UBnE4nt61MwdNm4Y29JfT1282OKiIick3K/XjXh6zIOUOOHzlVR1V9JysyookO9TUjmoiITHAqKsi4MAyDjUnr+Pr8/4W3zYsXT2/lhVOvEuhnY8OiBJrbe/ngaKXZMUVERK45nyx9iPePJfJTSx8cDidv7C3BYhjctjLZvIAiIjKhqagg42pW6Ay+s+jvSPCPZX/1YbaXfsCmZYn4edvYll1GV0+/2RFFRESuKZ8sffjLBo0HC2upbuxi1dxoIkP0lIKIiFyaigoy7sJ8QvlG1lcJ8Qrm3bI/UdtbzaZlSXT2DPCjF4/R0NJtdkQREZFrxr7zhwBY+Kl+CnaHgzf3lmC1GNy6PNmkZCIiMhmoqCCm8LH58IX0z+FwOvifghdZkxXFysxoSmvaeeS/D3P8TIPZEUVERKa8yvbznGstZXZoGhG+YYPHs0/UUtvczep5sYQH+5iYUEREJjoVFcQ0aSHTWZ+wmrruBt4u2c6WW2bz4M2z6O138LNX83jto7PaalJERMSNdlftB2Bt/IrBYwN2B2/uK8FmNbh1eZJZ0UREZJJQUUFM9ZlpNxHjF8XuqmwKm4pYMy+W//eFhUQG+/BOdhnPvVtkdkQREZEpqau/m8M1uYR6h5ARNmvw+NGiehpae1g7L47QQG8TE4qIyGSgooKYysPqwRfTN2M1rPy+8GU6+7tIjArg+w8uJj7Cn93Hz3OmqtXsmCIiIlPOgZoj9Dn6WR23DIvx54+EucUXliCunhdjVjQREZlEVFQQ0yUExHFLykZa+9p5rfgtAHy9bXz+xjQAnn+vCIfDaWZEERGRKcXhdLCnMhubxcaKmCWDxwfsDvLPNhIW6EVCpL+JCUVEZLJQUUEmhA2Ja0kIiONgTQ5FzWcAmBEfzIrMaMpq2/noWJXJCUVERKaO081nqOtuYGHkPPw9/QaPF1e20tU7wPzpERiGYWJCERGZLFRUkAnBarFy38w7MTD4w6mt9Nv7Abj7uun4eFnZuvsc7V19JqcUERGZGnZXZgOwJn75kOPHPl76MH9G+LhnEhGRyUlFBZkwEgPjWRe/krruBt4t2wVAkJ8nd6yaRmfPAK99dNbkhCIiIpNfU08z+Q0FJAbEkxyYOHjc6XRy7Ew93p5WZiYGm5hQREQmExUVZEK5ddoNBHsF8V7ZLmo6awFYvzCO+Ag/9hyv5tz5NpMTioiITG67K7Nx4mTNp7aRBDjf2EV9Sw+Z08KwWfURUUREhkf/YsiE4m3z5nNpd2B32nnh1FYcTgdWi4X7N6bhBJ579zR2h8PsmCIiIpPS+Y4adlXsIdAzgIWR84aMHSuuB2DBdC19EBGR4VNRQSaceREZzAvP4GxrCa8UvUltZx0zE0NY+XHTxh0Hy82OKCIiMunYHXaeK3yZAaede2f+FZ5WjyHjx840YDEM5qSGmZRQREQmI5vZAUQu5e602znXVsbuqv3srtpPUkAC89LnElg5wBt7S1gwI4LYcL+rTyQiIiIAvF/+EeXtlSyOymJuRMaQsbbOPs5VtTEjIRh/H4/LzCAiInIxFRVkQgrxDuafl32HvIaTHK7N5VRTMWXtFZAGtj4vnjqYxw0Zc5kWlEhSYAKeVk+zI4uIiExY5ztq2Fayk0DPAO5O+8xF48fPNuAE5mvpg4iIjJCKCjJhedu8WBKdxZLoLNr62jlal0dx81lO1J6lx7OKN89VAWAxLMT7x5ASlMzMkFTmhmdob20REZGPfXrZw32z7sTPw/eicz7ZSnKBtpIUEZERUlFBJoVAzwDWxa9kXfxK2qb18v/+5yP6PBpZsdSLut7zVLRXUd5exUeV+1gVt4x70u7AYqhliIiIyCfLHpZEZzEnPP2i8b5+OydLm4gJ8yUq9OKCg4iIyJWoqCCTTqCfFw+sm8fTb5zk/PEg/s99n8XuHKC8vYqXiv7I3qoD9Nv7uX/WXVgtVrPjisgE43A6yK3LZ2/VAfw8/UgKiCcpMIHEgDi8bd5mxxNxKYfTwc7yj/D38OPuGRcvewAoLGumr9/BPC19EBGRUVBRQSalxbMiOXyqjpzT9byTXcptK1NIDU7mGwu+yi+O/RcHa3Lod/TzYPq9KiyICABOp5O8hgLeKXmPqo7qweO5dXkAGBj4e/rhZ/PF18MXX5sPHhYbA047docdu9NOoGcAd0zfRLBXkFm/DZERqe9qoHugm8VRWfheYtkDwL4TNYCWPoiIyOioqCCTkmEYfPGmWZRUt/H63hJmJoaQlhCMn4cvf7vgIf7z+G85WpdHv6OfL2c8gIdVnaxFppITDYWEeAcT5x8zrPNPNRXzxtntlLdXYmCwNHohNydvwGJYKGuvoKytgvK2Slr6Wuno76SuuwGH03HpuZqL2ZJxP2khqa78LYm4RWlbBQBJgfGXHK9p6iLnVB1JUQFMj1OxTERERk5FBZm0/H08+OpnMvjX53N55s2TPLJlCf4+HvjYvPmb+V/mV3m/I7+hkGfyf8dX5nzxov24RWTyGXAM8HLRG+w7fxBvqxffzPrfxAfEXvb8qo5qXj+zjYKm0wBkRc7llpSNRPtFDZ4T5hNCVuTcIdc5nU567D0MOOxYDSs2ixWrYWV3VTZbz7zNv+f+is9Mu4kNSWvVv0UmtLL2SgCSAxMuOb7jYBlOYNPyJDU5FhGRUdEnIZnUZsQHc8fqFJrbe3n2nUKcTicAXlZPvjb3QTLDZlHYVMR/5v2WXnufyWlFZCxae9v5We6v2Hf+IBE+YfTa+/jl8Wdp7mm56NyW3laeK3iZJw79lIKm08wMmc53Fv8dX858YEhB4XIMw8DH5kOApz++Hj54Wj2xWqxcl7CKb2Z9jUDPAN44t51n8v6bY3X5tPS2uuO3LDJmZW0VH++SdHHxrbm9l335NUSF+LAwLcKEdCIiMhXoSQWZ9DYtS6KwrJljZxp4/0glGxdf+GmMh9WDh+Z8gWdPPM/xhpP84th/8dfzvqRGbCIf6+jr5OWi15kbns6i6AXDvq65p4U+e99grwGrxUqsX/Sof8ppd9jpHujB39Pvsuf8/+3deWBU5b3/8fcsmex7Mtk3AtkXQEA22UTccBdFxaVWW6u3tfd3vV389Wr780K1q621tXVplaLigopWQVE2IewQSEJICEnIRvZ9T2Z+f0SjlLAYkgyEz+sfmHPOnPOdhxnmme95nu9T3FTC3w68QkNnIxdZ01mSuIjNZdtYdfgDns18kf8z8UHcnFyx2W1sKd/Ou4c/oqO3g1D3YG4YezWJfnFDdhd2jHc0P53yQ/6e/SpZtblk1eYC4OPsTbRXJDeMvYoAV/8huZbI2eix9VDaUk6YR8iA0wA/3nmUXpudK6dGYTRqlIKIiAyOkgpy3jMaDXznmiQef2kHb6w/THFlM3PGhxEb5oXZaObbKUv4e85r7K3az5++SCycrFiVyIWipbuVP+77G2UtFeyrzsLqFkjkSeZcf6m7t5vlB99gd1XmCfvGB6Zyd9KtWEyW47a3drfxXsFHdPR0kBKQSIp/Qv/nr6a9ji3l28mo2ElzVwupAUlcGX0pUV8bpl3ZWsW6oxvZfmwPNruN62Kv5LLIORgMBuZFXEJdRz0bSrfw/IFXuDnuWlYeeoeCxiJczS7cFn8j00OnDMv0BE+LB/8x/j7y649Q1HSU4qYSipqOsq/6ABOsqUoqyDmhvOUYPbYeojxP/Gy3tHezYV853h4WpiUHOyA6EREZLZRUkFHB28OZB29I5aUPD7I16xhbs44RFuDO7PGhzB4fxreSbsNkMLKrch//s/WXTAuZzOzwGQS6qeMvF57W7jae2fs8ZS0VJPnFk1N3iBezV/CTyQ/jepKRPK3dbfx1/51pEWwAACAASURBVD8oaCwiwjOMSM8wzEYzJoOJwsa+H9P1ext4IO0evCyeABQ1HeXFrBXUddQDsLsqE6PByDifMRgNRnLr8rFjx9XsSoRHKAdqcjhQk0OyfwJTQyaxuzKTzOos7NixugVw87jrSPaP74/JYDBw07hrqO9sJLM6i2U7fg/0JThuibsOb2evYW1Ho8FIvN9Y4v3GAn11GDp7OzUaSs4Zxc1fFmmMPGHf+j2ldHb1ct2MGJzMmg0rIiKDp6SCjBpxET4s+85UDhXXs2FfOXvyqnl1XT55pY08cF0ydyctJswjhA0ln7O+9HM2lG4h2T+BSUHjCfcMxeoaoOUnZVjY7XbWFH1GVu3Br4r+GU2EuYdwzZjLR/R915dQ+BulLeXMDL2YW+NvYHXBGj45uoHXD63inqTbTnhOTXsdf858kcq2aiZa07gr8dbjhlL32Hp4Nfdtth/bzW92/Ynvpd/LobrDrDr8ATa7jatiLmNCYCr7a3LYX53NofrDAMR6RzMj9GImWNNwMprJqy/go6J1ZNfmkv3FlIJIz3AWRM0lPTB5wBEHRoORe5Ju48+ZL1LdXsuiuOsYH5gyTK13agaDQQkFOacUN/UVafz3lR86u3v5ZFcpbs5mZo8/eaFTERGRM6GkgowqRoOBxGg/EqP9aGrt4s/vHGBXbhVve7uwaO5YFkTNZV7EJeyrOsD60i1k1R4kq/YgAE5GMyHuwcT5xnJZ1Bw8nE4+v1vkm1hf+jkfFK7FgAE79v7tObWHcHNyZUHU3GG5bl59ARtLtwBgNpqxGJ0oaiqhvPUYM0KncGv8DRgNRq4Zczn5DUfYVbmPBN9xXGudB/QlCw7W5bHi4Fs0d7cwP3I218VeecKPe7PRzJ2JtxDo6s8HhR+zbMfvsdlteDi5863k20nwGwdAqEcwV0TPo6GzkR5bzwlTBL68659fX8CBmoMk+ccT7zv2tLUQLCYnfjDhOxgwqHq9yNcUN5VgMVkI+bfipJszy2lp72bh9GhcndUVFBGRs6NvEhm1vNwt/MdNaSxbvpuPth8lwMeVuRP6hmxPCp7ApOAJHG0q5XBjIWXNFZS1lFPeUsHR5lK2lu/gqpjLmBU2TaMX5LQKG4t55/CHTLCmMid8xnE/bLNrD7Eq/wO8LJ78ePIP8LZ4YbPbaOlu45c7fs+HheuYaE0b0jn4drudjWVbeTv/fWx22wn7p4dMYXH8jf3JAZPRxL3Jt/PLnU/zRt67uHlY2FuSQ1btQdp7OjBg4Ja465kdPv2k1zQYDFwZM58AV3/+mfsmMV5R3JtyOz7OJ657P9C2rxvnG8s439hv9Jq1rKPI8Tp6OqlorWSMd/Rxnw+73c6ne8owm4zMn3TqOioiIiJnQkkFGdU8XJ344S3pLH1lF//8+BD+Xs6kxQb074/0Cj+uOF23rYfNpVv5sGgdb+Wv5vOybVw9ZgFBboG4mV1xc3LDYnTS3VABvvrxvir/A3rtvRQ0FnKo/jB3Jd6Cm5Mbx1qreClrBSajie+k3t3/Y9pkMOHt7MlN467hHzmv8fqhd3go/dunfF/12nrpsnXjYnI+5XE9th7eyHuXLeU78HTy4N6UOwh1D6bb1k23rQejwTBgAsPf1Y/bE27mxax/8rddKwDwdfZhavAkJgdPOK544qlMDp5AakASziaLPiciDlTSXIYd+wlTH/JLG6msa2NqUhBebpaTPFtEROTMKakgo57Vx5Uf3JTGr17by1/ezeaWubFEBHkSFuB+wrBPJ6OZeZGzmBw8kQ8KP2ZL2XZezPrncce4mJy5IvpSLo2cNai7o81dLbTUN+CBz1m9LulT11FPa30j7pz67veXatpr+bxsO529nf1TApxMTkwITCXI3XrG1+3o6eS1Q2+zq3IfHk7u3BJ3PZ+Xb+dATQ6/3PkHFsffyJt579LR28HdSYuJ8T6xUNqkoPFsq9jFwbo89lRlclHQ+P59+6uzWX1kDU2dzXTauuix9QAQ6x3DPcmL8XPxPeF8TV3NvHBgOQWNRYR7hPLdtLsHPO5kJlrTaIm7gW5zB+PcxhHhGTaoxICL2fkbP0dEhtaXRRqj/y0huHl/OQCXpIWMeEwiIjI6KakgF4TYMG/uX5jEX97NYvnHef3b/bycmTM+jIXTo4873tPiwW3xNzIrbBp7qw7Q2t1GW08bbd3tHG0u5d2CD8mty+eupFvPqMK83W4nv6GAz8u2k1mdRY+9lwVRc7lmzOUatj1Ird1trCn6lI2lW+m193KRNZ0bxl6Nr8vAyZrqtlrWFH/Kji+WJvx3G0q28NjUR85oudGS5nL+kfMax1orifGK4tspd+Dr4sMEayofFX3KR4Xr+HPmiwAsiJrLlOCJA57HYDCwOP5Glu74LW/mrybRLx6DwcDb+e+TUbETs8FEoFsAziZnnE0Wunq7KWgs5Jc7nmZJ4iLSvyhI2NXbxfqSz/m4eAMdvR1MsKZxZ+ItOJu++V3IWeHTCAz0pLq6+Rs/V0TOHcVNX6788FVSob2zh525VQR4uxAfdeYJRxERkVMZ1qRCXl4eDz74IPfccw9LliwZzkuJnNakBCtP3Hcxh8saKa9ppay6haJjzazadAQvdwuz0k+sgB3mEUKYx/F3c5q7WvjnwTfIqs1l2Y7fsyRxEWO8o6lpr6W2o57a9jraetrptfXSY++hx9ZDfsMRqtpqAAh2s2I32vm4eD1VbdXclbR4UD/+zlZ5yzHcndzxdvYc1PNtdhs5tYcI9ww97Rz5M1XZWkVGxS7SA1MGvLMPfVNUNpZuYW3RZ7T1tOPv4ou3qye7qzLZX5PDgqg5zI+cQ1dvF1Xt1VS2VpPXUMCuyn3Y7DaC3YO4MmoeoR4h/VMC9ldn82nJJlYfWcvi+BtOGl9nbxf/KvyY9SWfY7PbmB0+gxvHXo3Z2PdfqdFg5OqYyxjrHcPyg28wxjuKa8ZcfsrXHOjmzxXR83n/yBpeznmNitZKajvqCfcI5e6kxYR6fLV+vN1uZ2v5Dt7Mf4+/HXiF2eEzCPcI4V+Fn9DQ2Yi7kxuLYq9jdth0TT2QC8ayZcvIzMzEYDDw6KOPkpaW1r9v27Zt/O53v8NoNBITE8PSpUsxGi+MRG5xUynuTm74u/j1b9uZW0VXt42ZaSEY9X+EiIgMkWFLKrS1tfHEE08wbdq04bqEyDcWGuBOaMBXqzpUNbTzxD92snztIUL83RgXfvopCZ4WDx5I+xabyjJYdfgDntv/j9M+x2w0MzloIjPDLibWOxo3bxO/3PBn9lVnUbvnLzyQds+gf5i3dbexvnQL1W21LIq7FvfT3Gm32W18WLiOj4rWYTQYSQtIYmboVOL9xp7xqAmb3cYrOW+ws3IPBgwk+I1jSvBE0gNTBpUg6ejpZE3Rp3xWspleey+fHN1AemAK1465nOAvqpbXtNeSUb6TjIpdNHY14Wp25YaxVzM7fAbBVm8+OLCR9wo+5F+Fn7Cm6DN67b3HXSPYPYirouczwZp6wuuM9oogu+4Qn5dt4+LgiwZMaGTVHGRl3rvUddQT4OLH4vgbSfSPG/D1xPuN5YnpPwU4ox/38yNnsbNyL1m1uRgwcEXUPK6Mmd+frPiSwWBgRtjFxHhH8WL2iv6VHZyMTiyImsuCqDm4ml1Pez2R0WLHjh0UFxezcuVKCgoKePTRR1m5cmX//scee4xXXnmF4OBgfvCDH7B582Zmz57twIhHRktXK7UddSR9MfrpS5szyzEAM1M19UFERIbOsCUVLBYLzz//PM8///xwXULkrFl9XHnw+hR+uzKTZ1cd4LF7JuPndfp15g0GA7PDpzPWJ4b3j6wBwN/FjwBXf/xdfHF3csdsNGE2mvuL8n39x56Hszv/Mf4+Xj/0DhkVO3lq5x8ZH5hCuEco4Z6hhLgHYzE5nTKG1u421pdsZn3JFjp6OwAobSnjP8bfd9IERWt3G//Ifo2cukP4ufjianZhX3UW+6qzCHDxY1roFCYHjcff1W/A50NfQmHFwbfYWbmHCI9QzEYzB+vyOFiXh7PJwuzwGVwZPX/A+Bs7mylrKcdisuBssmAxWShpKuWdgg9p6GzEz8WXSyNnsevYXjKrs9hfnc3k4Ak0dDaRV38YABeTC5dGzOLy6Hn9CRSjwci0kEmMD0xhbdFn5NQdwtfZhyC3QILcAgl2DyLGO/KkSROz0cziuBt4eu9zvH5oFT+a9P3+VT+6ert5I+9dMip2YjQYWRA196Sv7+u+yUgBs9HMPUm3sbb4M+ZFXMIY76hTHh/qEcyPJ32f94+spcfWw+XR84ZstIjI+SQjI4P58+cDEBsbS2NjIy0tLXh4eACwatWq/r/7+flRX1/vsFhH0pf1FL4+9aGsppWC8iZSxvid0feciIjImRq2pILZbMZsVskGOfclRvux+NKxvLoun2fePsBPlkzE2enMlpEM8wjhgbRvDeq6ZqOZOxJuJtjdyvsFa9hUltG/z2gwkh6YwpXRl54w/aK2vZ7NZRlsLsugo7cTDyd3boi5mrqOBjaWbuH3u//C9yfcf0KF/5LmMp4/sLz/7tU9ybfhZnalqKmEz8u3sbsyk/ePrOH9I2uI9Y5mcvAEJljT8HD6amSHzW7jtdxVbDu2iyjPCL4/4T5cza5UtlWz49getlXs4uPi9eyrOsDtCTf1LwvY3NXCx8Xr2VyWQfcXBQf/vS2ujJ7Pgqg5WEwWZodNZ39NDquPrGHHsT0AjPWJYXrIFCZYU7GcZDSEq9mF68dexfVc9Y3/Pcb5jmFqyCS2VexiY9lW5kVcQm17PS9kvcLR5jIiPMO4M/GWE/49hkqEZyj3pZz5NDGLycJN464ZllhEzhc1NTUkJyf3P/bz86O6uro/kfDln1VVVWzZsoWHH374tOf09XXDbB7apYQDAwc3zWywaqqqAEiLiOu/9uqMYgCunjlmxOMZaud7/OcatefQUnsOLbXn0Bqu9jxnfvUPx5c46I04lEZzWy6+IpHqpk4+2XGU1z8r4L/umDjsc9K/bM/brAu5KX0BJU0VFDeUUlRfSk51Pnur9rO3aj+Tw9K5KelKmrtaWZu/kd0VB7Db7Xg7e7IoZSGXjb0EF7MzdrudwGwf3sr+F0/vfY7/O/v7AOw7lkPmsRxyqvPptfVyc/JV3Jx0df+8Yqs1mSljk2ntamN76V42F+8gpyqfgsYiVh56l0jvUBICx5IYOJasqjy2VuwgxjeC/5nzMB6WvoRDIJ6kRI3h9u6FvJ71Ph/lrefpvX9l/piZeDp78GH+ejp7Oglw82N29FTs2Ojo6aKjpxOL0Ymr4+cR5BF4XPvMt05lXsIUsqoOEeDuR6hn0Bm159m4z+tWsj48yL8KPybEz5+X971Fc2cLc2Kmcd9Ft512dMJoMpo/746g9hwZdrv9hG21tbU88MADPP744/j6nr44YX1925DG5IjCpzkVfSO7fOz+VFc309Nr49OdR/FwdWKM1eO8LsSqQrJDS+05tNSeQ0vtObTOtj1P1Zc5Z5IKQ/0lDnojDqULoS1vnjWGI6UNbNxbSkKENxcnnfpH7NkYqD298CPV049UzzQWRtjJrs1lTdGn7CzLZGdZZv9xUZ4RzAqfxkRrOhaTE831XTTTBcDcoNnQZeKt/NX899ql2Pmqgx3mEcK1Y64gJSCR2trWAeNK9UwjNSWN+o4GdlXuI7s2l6KmoxQ3lrH28Mb+83wv5du0N9po58T3xNXhV5DkmcSruW+x7sjnAHhbPLk+7kqmhU7ByTjAfzvtUN0+8PsrxBQOHVDdcfL331C+P68bcxUrct/kT9v/gclgYnH8DcwMnUpjXQfQMSTXONddCJ/3kTScX+IXOqvVSk1NTf/jqqoqAgO/SlC2tLRw//3388Mf/pCZM2c6IsQR197TQXFTKb7OPnhZ+t47mYdraW7r5rJJETiZL4xClSIiMnLOmaSCiKM5mY3cf00Sj724gxWf5JEQ5Yu3+8ivygB98/FTAhJJ9k/gUP1hNpRuwd3sxiXhU4n2GnhVhC/NjZiJm9mVNUWfEuUVQaJfHAl+cd9olQdfFx8ui5rDZVFz6LH1cLS5jIKGQuo7G7gq+rLTFoOM8Y7kx5N/wMbSrRgNRmaEXnze3OWfGnIRe6v2U9Fayb0pd5y2voGIOM6MGTN45plnWLx4MdnZ2Vit1v4pDwBPPvkkd999N7NmzXJglMOvtr2O/TU5HKjJIb/hCDa7jYus6f37N+8vB+CSNBVoFBGRoTdsSYWsrCyeeuopysrKMJvNrF27lmeeeQYfn9NX1xdxFKuvGzfNieW1dfksX3uIh25IcejSfAZD38oKCX7jvtHzLg65iItDLhqSGMxGM2O8o77xj2uz0cylkedfR95oMPK99G/1/11Ezl0TJ04kOTmZxYsXYzAYePzxx1m1ahWenp7MnDmTd999l+LiYt566y0AFi5cyK233urgqIdWbl0+z2a+iM1uAyDSM4yUgCRmhl4MQFl1CwcKahkT6kW41eNUpxIRERmUYUsqpKSksHz58uE6vciwufSicHYfqmZPXjU7DlYdNw2ip9dGSVULUcGeWuN7FFMyQeT88cgjjxz3OCEhof/vWVlZIx3OiLLZbbyd/z52u51b4q4nPTD5hJVgPsgoxg4snBbtkBhFRGT00/QHkX9jNBi496oEHnvpq2kQFrORjfvK+WRXCfXNnUxKsHL/wiTNTRUREYfJqNhJeesxpoVMZnb49BP2V9S2siOnkkirB+lj/Qc4g4iIyNlTUkFkAFZfN26eHcur6/L57et7qW3qoL2zF2cnE6EB7uzKraK9o5uHbkzFxaKPkYiIjKyOnk4+OPIxFqMTC8csGPCYD7b2jVK4ZkaMQ6fyiYjI6KbbrCInMe+icOIifCitbsXJbOLGWWP4zUPTeezuSYwfG0B2UT2/eX0fLe3djg5VREQuMOuObqSpq5n5kbNPmPIAUFnXxracY4QHujMhLsABEYqIyIVCt1hFTsJoMPD9m1LJO9pAyhg/nMym/n0P3ZjC3z/MZWvWMZ5csYf/unU8vp7ODoxWREQuFA2djaw7uhEviyeXRs4e8JgPMoqw2/tGKagGkIiIDCeNVBA5BXcXJybEBR6XUAAwGY3ce3Uil02KoLymlT+/c4Bem81BUYqIyIXk/SNr6bZ1c82Yy3Exn5jQrmpoJyOrktAAdy6KD3RAhCIiciFRUkFkkIwGA4svHcvU5CAKyptY/XmRo0MSEZFRrrS5nO0Vuwl1D2ZqyKQBj/kwowib3c7C6VEapSAiIsNOSQWRs2AwGFhyWTwB3i58kFHEoaP1jg5JRERGsc3l27Bj59rYKwZc/raprYstB44R7OfGlISgAc4gIiIytJRUEDlLbi5mvnNtMgYMPP9BDq0dKtwoIiJDz263k1VzEHezG0l+8QMek1NUR6/NzozUYIxGjVIQEZHhp6SCyBAYG+bNtTOjqWvq5OWPcrHb7Y4OSURERpmSljIaOhtJ8k/AZDQNeExOYd+IueQYv5EMTURELmBKKogMkYXTookL92bXoWo+3FZMT68KN4qIyNA5UJ0DQFpg0oD77XY72UV1eLg6ERnkOZKhiYjIBUxJBZEhYjQauP+aZNxdzLy98Qg/+WsGn+wsobOr19GhiYjIKHCgJgeTwUSiX9yA+ytq26hv7iQp2lcFGkVEZMSYHR2AyGji7+3CL+6dwtodJWzMLOO1T/N5f2sRU5OCCAlwx+rrSpCvK2ZnJ3YfqiKvpJG80gaq69v51lWJWvpLREQGVN/RQElLOYl+cbiaXQY8JruoDoCkaE19EBGRkaOkgsgQ8/Ny4bb541g4PYpPd5fy6e5S1u0uPenxJqMBg8HAX1dn8fCidJLVGRQRkX9zoOYgACkBiSc9JruwL6mg7xERERlJSiqIDBNPNwvXXzKGK6dGUVrVQmV9G1X17VTVt2PDQFiAG3Hh3sSEeFFQ1sjv38zkT28f4JHbxhMb6u3o8EVE5BxyoKavnkKq/8D1FHp6bRw62kCwnxv+3gOPZBARERkOSiqIDDNnJxOxYd7Ehn2VKAgM9KS6urn/cWK0Hw9cl8Kz7xzg6Tcy+ckdEwkL9HBEuCIico7p6Okgr/4wYR4h+Lv6DnhMQVkjnd29GqUgIiIjToUaRc4RE+MC+daVibR29PDblfuoqG11dEgiInIOyK3Lp8feS2rAwKMU4Gv1FGIGTjqIiIgMFyUVRM4hM9NCWDxvLA0tXTz+0k4+2FqkpSlFRC5w+7+Y+pB2qqRCYT0mo4GESCUVRERkZGn6g8g5ZsGUSPy8XFjxSR6rNh1hx8FK7r4ygVB/dw4W15NVWEd2YS0+Hs78cFE6rs76GIuIjFY2u42s2oN4WzyJ8Awb8JiW9m6KjjUxNsxb3wkiIjLi9M0jcg6alGAlMdqXN9cXsCmznGWv7MZoNNBrswNgNhmobujgufey+cHNqZiMGnQkIjIaHWksprW7jRmhUzAaBv6/Pre4Hrtdqz6IiIhjKKkgco5yd3HinisTmJYcxJsbCgBIifEjJcafqGAPnll1gANHalnxST53LojDYDA4OGIRERlq/as+nEE9heQYJRVERGTkKakgco6Lj/TlZ3dNOmH7965L4ckVe9iwtwyrjytXXBzpgOhERGS49Nh62HFsD65mF+J9xw14jN1uJ7uwDldnM9EhniMcoYiIiAo1ipy3XJ3NPHxzGr6ezryx/jC7cqscHZKIiAyhzOosmrqamRo8CYvJacBjympaqWnsICnKV1PhRETEIfTtI3Ie8/Ny4eGb03C2mHjuvWz+ujqb4mPNjg5LRESGwKayDAAuCZ824P6K2lZ+/0YmABclBI5YXCIiIl+n6Q8i57nIIE9+eHMaKz7JY3tOJdtzKkmO9mXBlEjGhXvjYtHHXETkfFPWUsHhhkISfMcR5HZiwqD4WDO/XbmPlvZubp4Ty9SkYAdEKSIioqSCyKgQH+nLL+6dQlZhHWu2HyW7qJ7sonoAAn1cCA/0IMLqwfSUYKy+bg6OVkRETufLUQqzwqefsO/Q0Xr++PZ+Ojp7ueuKeOaMH3ipSRERkZGgpILIKGEwGEgd40/qGH8KK5rYll1JaXULJVUt7M2vYW9+Df/KKGbuhDCumRGNp5vF0SGLiMgA2nva2XFsD77OPqQGJB63L6+kgd+9kYnNZue71yUzJTHIQVGKiIj0UVJBZBSKCfEiJsQL6KsM3tTaRU5xPe9tLmTd7lK2ZFVw5cVRzJkQhofrwMW/RETEMbZX7KGrt4vLo+ZhNHxV/sput/Pap/n09Nh4eFEaabEBDoxSRESkj5IKIqOcwWDA28OZacnBTE6wsmFvGau3FLFq0xFWbTqCv5cLUcGeRAZ5kDrGvz8ZISIiI89ut7OpLAOzwcSM0CnH7dt3uIbiY81MSbQqoSAiIucMJRVELiBmk5H5kyKYnhLCZ3tKyStpoLiymT151ezJq+bdzYVMTwlm0ZxYvD2cHR2uiMgF51D9YSrbqpgcNAFPi0f/dpvdzrubCzEY4NoZMQ6MUERE5HhKKohcgNxczCycHg303RVraOniSHkT728tZGvWMfbkVXPtjBjmTwrHbNLKsyIiI2XzSQo07jlUTUlVC9OSgwgNcHdEaCIiIgPSrwWRC5zBYMDX05mL4gN57O7J3HV5PCajgTfWH+YXf99JRW2ro0MUEbkgdPV2kVWbS4h7EDFekf3bbTY7731eiNFg0CgFERE55yipICL9jEYDcyaE8cvvTmP2+FDKalr531d2sb+gxtGhiYiMevkNhfTYekjxT8RgMPRv35lbRVlNK9NSggjy07LAIiJyblFSQURO4OHqxN1XJHD/wiR6eu384c39/CujCLvd3n9Me2cPpdUt9NpsZ3zeXpuN9s6eUx5j+9o1REQuJAdrDwGQ6BfXv+3LUQomo4FrNEpBRETOQaqpICInNS0lmJAAN555+wBvbzzCweJ6TEYj5TUt1DZ1AmD1ceXqaVFMSwk+Zf2Fnl4bv1u5j/zSRtJi/ZmRGkJarD9mk5G2jm52HaomI+sYeSUNjIvwYfb4UCbFB+JkNo3UyxURcaicujwsRifG+ET3b9uWc4xjdW3MSg/F6uPquOBEREROQkkFETml6GAvHrtnMs++c4CconoAfDwsJMf44e5iZk9eNX//KJfVWwq5amoUM9NCcTKfmFx4dV0+uUcb8HB1Ym9+DXvza/BwdSI62JPco/X09PaNUAjxdyOvpIG8kgZe/cTMtJRgJo4LJCbUC2enrxIMdrudyvp2DhbXY/V3JzHCG+PXhguLiJxP6jrqqWyrIsU/ASdjX/esqbWLNz47jNlkYOH0KAdHKCIiMjAlFUTktLzdLfz49gmU17Th6+mMh6tT/7765k4+2l7Mxn3lLP84j/V7y3nohpTj5v2u31vGhr1lhAd68H/vvIjK+ja2Zh0jI/sYWYV1hAa4My05iKlJwfh7u1BV38amzAo+P1DBul2lrNtVisloIMLqwdhwb9o7ejh4tJ66L0ZLACRF+3LfwiR8tBSmiJyHDtbmAZDoFw/0JU7/8VEuTW3d3DJ3LAHeGqUgIiLnJiUVROSMmIxGIqweJ2z39XTm9vlxXD0tmnc2FbAps4L/9/JO7r0qkYvirV+MOMjDw9WJH9yUirPFRGSQJ5FBntw8J5am1i58PZ2PK0pm9XXj5jmxXH9JDFlH6sg9Wk9BWSNFx5opOtYM9NV9mBQfSGKUL7mljezMqeSxF3fw7asTSR8bMGLtIiIyFHLq+pIKSf599RQ2769g3+EaEiJ9WDAlwpGhiYiInJKSCiIyJLzdLdxzZSLxEb68vDaXZ9/JYs6EMHYfqgLgoRtSCPi3+cBmkxE/L5eTntNsMjJ+XADjx/UlCbq6eymubMbZyUS41aN/usOiBQm8vuYgb6wv4A9v7Wf+pHBunTcWk1G1aEXk3Ndr6+VQSUobQgAAFupJREFUfT7+Ln4EugZQWd/Ga+vycXU2c9/CJE3tEhGRc5qSCiIypKalBBMZ5MGf381iw94yAJYsiCM+0vesz21xMjEu3OeE7QaDgfmTIoiL8OGvq7NZt6uU2sYOHrgu+YwKPZbVtOLmbMbXU1MnRGTkFTWV0N7TwUVB47HZ7bzwfg6d3b1855qkUyZeRUREzgW6jSciQy4s0IP/uXsSl14Uzg2XxDB3QtiIXDcyyJP/uXsSiVG+7M2v4ek39592CcstByp4/MUdPPr8NrZmVYxInCIiX3ewrm8pySS/OD7MKKagvImLk4KYmhzs4MhEREROT0kFERkWLhYzd1wWxzUzYo6rlzAS1/3hojQmjAvgYHE9v125j5b27gGP/XjHUV7810FcLCYMwAsfHORv72fT1nHqRISIyFDKqcvDaDDiZwjl/a1F+Ho6s2RBnKPDEhEROSNKKojIqONkNvHgDSlMTwnmSHkTT63Yw46DlTS3dQF9VdXf3ljA658dxsfDwk+XTOTn35rMmFAvtmVX8vO/7yC/tMHBr0JELgQt3a0cbSolxiuKdzeW0tNr59Z5Y3F3cTr9k0VERM4BqqkgIqOSyWjk3qsTcXU28+nuUp57LxuAyCAPvNwsZBXWYfV15ZFbx/cXkPzJHRNZvaWQf20t5pf/3MPkBCs3zhpz3PKYI6W8ppW/f3SQuAgfLp8SiZebZcRjEJHhl1uXjx07AcYINuTXEBfuzeQEq6PDEhEROWNKKojIqGU0GLjjsjimpwSTXVjHweJ68ksbONrbQmSQB/95y3i83b/6sW42GblxViwpMf6s/CyfnblV7Mmr5pL0UK6dEY2Px8gUcmxq6+LpNzOpaeygoKyJT3eXMndCGFdMicR7hGIQkZFxsLZvKcmDB8wYgNvmx43olDEREZGzpaSCiIx6MSFexIR4sXB6NJ3dvVTUthIW4H7SlSHiInz42V2T2H2omrc3HWHD3jI+31/BpPhALkkLIT7K95RLvGUV1vL2hiPYsZMU7UdStC/jwn1wdjr9ShTdPb386e0D1DR2cPW0KHw8nPlwWzFrd5Tw2Z4yFkyOYOH06DM6l4ic22x2Gwfr8nA2uFJZ7sSs9FCigj0dHZaIiMg3oqSCiFxQnJ1MRAd7nfY4g8HApAQr48cF8Pn+CtbuLGFbTiXbcioJ8HZhZmoIaWP9ibR6YjT2JRgaWzp57dN8dhyswmgwYDQaOFrZwprtRzGbDIwL9yE91p+0sQEEDzClwm6389KHuRwua+TipCBunDUGg8HArPRQthyo4IOMIv6VUcz2nEqWLIgnLdZ/qJtHREbQhtItNHY1QW0Ers5mbpw1xtEhiYiIfGNKKoiInILZZGTOhDBmjw/lcFkjmzMr2JFbybufF/Lu54W4u5iJj/QlxN+Nz/aU0d7ZQ0yIF3dfEU+QnxuHSxvJKaoju6hv+sXB4npe/+wwVl9XkmP8CA9wJ8TfnZAAd9bvKWV7TiWxYV7ce1VC/xBoJ3NfDNOSg1m9pZC1O0p4+s1MJidYuWXuWPy9h28d+/bOHto7e/DzGr5riFyIqtqqWV2wBrPdhebisdxySQxe7qqdIiIi5x8lFUREzoDB0DfSYFy4D7fNH0fm4Zr+JMGevGoAXJ1NLFkQx5zxYf2jF5Jj/EiO8WMR0NDSyf6CWvYX1JJdVMf6PWUnXCfA24Xv35g24NQMZ4uJRXPHMjU5mFfW5LIzt4qduVUE+boSF+FDXIQPkUGemIx9oyQMhr6RGd7ulm88R9tmt7NlfwVvbSygraOHW+aOZf6kcM31FhkCNruNl7PfoNvWTVdBEkGePsyfFO7osERERAZFSQURkW/I1dnM1ORgpiYHA1Dd0E7xsWbGhXufspCij4czs9JDmZUeSnePjdLqFipqW6mobaO8ppXW9m7uvCLhtHcrI6we/PTOi9iyv4LdedXklzaweX8Fm/dXDHi8t7uFMaFejA3zJirYk16bnea2Lprbumlp78bHw5moIE8irB44W0zkl9TzzMp9FFY04exkwtXZzGuf5pN7tJ57r078xkvd2e32b5SMaOvo5sCROtJi/XF1Hvhr6lhdG8dq2+jo7qGr20ZHVy+hAW6kxGhKiJz7Xtv3MUXNxfTWBRFiGsd3r03CbNIq3yIicn5SUkFE5CwF+rgS+MWylGfKyWzsLyA5GEaDgUvSQ7kkPRSbzU5JVQuHjtZT2dCO3WbHZu8bbdDW0UNhRRN782vYm19zynMagEBfV6ob2rHbYUpi3/QKo9HA31Znsze/hl/8fSf3LUwiyNe17xo2O712O+0dPbR1dNPa0UNrRzc1jR0cq2ujsq6d6oZ2xoZ58d3rUvBwPXlCoqu7l0/3lPJhRjGtHT14e1i4dd5YLk4M6k9KNLZ08vamI2zZX4F9gDb56Z0TiQ31HlSbigy37h4bL67byX7zBuh1Yp71Cm68IVEJBREROa8pqSAicp4zGg1EBXuesmp8XVMHR8qbOFrVjMVswtPNCU83C+4uZuqaOimubKb4WDNHq5qJCfHmptljSIzy7X/+I4snsHpLIe9vKeLJFXvOODaLkxEvNwvZRfUsW76b/3NLOgH/loDptdnYeuAY735eSH1zJ27OZmakBrPjYBV/W53D5swKbp03lqzCOj7YWkRHVy9hge5MTw7GxdmMs5OR9s5eXv0kjxfez+Hn907R6hhyTlr5WR77Oj/FZLFxbeQNXB6X7OiQREREzpqSCiIiFwA/Lxf8vFyYlGAdcP+0lL6pHHa7HavVi+rq5uP2G40Grr9kDHERPmzKLMdu79tmNPT96epsxs3ZjLuLE24uZvy9XAjyc8PHw4IdeGt9AWt2HGXp8t38cFE6UcGetHf2sDmznHW7S6lp7MDJbOTKiyO5aloU7i5OXDMjhlc/yWN/QS0///tOADxcnbjz8rHMSg/BZDz+7m51Qzsf7yzhzfWHWbIgfugbUeQs7MqtYlPVepxCG0gPSOXyuIsdHZKIiMiQUFJBRET6na72QVK0H0nRft/snMAt88bi6+XM6+vyefLVPVycGMSOg5V0dPViMRuZOyGMq6dFHbfKhNXHlYdvTmNffg2rtxYRH+HDtTOicTtJTYebZo8hq7COz/aUMX5cQH99BZvNzsZ9ZWw/WEVEoAepsX7ER/qedDRDZ3cvVfXtVNW30dzWTa/NTm+vjV67HR93Zy5OCuovxClyJqrq2/j7tk9wiizEz+LPksSbHB2SiIjIkFFSQURERsRlkyLw9XDmb+/nsCmzHG8PC1dNjWLOhLCT1lowGAxMiAtkQlzgac/vZDZx/8Ik/veVXfz9w1z+37enUFrVwqvr8impagEgr6SBT/eU4mQ2EhfujYuzme4eG13dvXT12Khr6qChpeuU1/n8QAX3X5OEzymKcop8qau7lz98tAF7+H4sBmf+Y+K9uDm5OTosERGRIaOkgoiIjJhJCVYCfVypamhnwriAIS9QFxXsybUzonlncyE/f2kntU0dAMxMDeH6S2Koqm/nwJFaDhypJbuo/rjnmowGfDycSYzyJcjPjSBfV7zcLZhNRowGAyajgU2Z5ew7XMPjL+3g/oVJpIzRahNyas+uzqDefwsGAzyQfhdBbqdPkImIiJxPlFQQEZERdbqikmfrqmlR7C+opaC8iZgQL26/bFz/ihB+Xi4kRPmyaO5YWtq7sdntWMxGLGbTGU1pSB/rz7pdpbyx/jC/eyOTuRPDcHEyUfnFdInG1i6+fXUSabFKNgjsLzzGlqb3MLp1c/PY60nwG+fokERERIackgoiIjKqmIxGfnhLOkUVzSRG+2I8SZ2IUy1veTIGg4HLJkcwLsKb597LZv2esv59zhYTIX5ueLtbBh27jC7l3YUY3VqYHDCFuZHTHR2OiIjIsFBSQURERh13FyeSY75ZQclvIjrYi8fvmUxOUR2ebpb+qRKnK3QpF5Z5sZNIiw0n2Bjm6FBERESGjZIKIiIig+DqbOai+IGX6BQBsJicSA9MOmGJVhERkdFkaCtkiYiIiIiIiMgFQ0kFERERERERERkUJRVEREREREREZFCUVBARERERERGRQVFSQUREREREREQGRUkFERERERERERkUJRVEREREREREZFCUVBAREZHz1rJly7j11ltZvHgx+/fvP27f1q1bufnmm7n11lt59tlnHRShiIjI6KakgoiIiJyXduzYQXFxMStXrmTp0qUsXbr0uP3/+7//yzPPPMNrr73Gli1bOHz4sIMiFRERGb2UVBAREZHzUkZGBvPnzwcgNjaWxsZGWlpaACgpKcHb25uQkBCMRiOzZ88mIyPDkeGKiIiMSmZHByAiIiIyGDU1NSQnJ/c/9vPzo7q6Gg8PD6qrq/Hz8ztuX0lJyWnP6evrhtlsGtI4AwM9h/R8Fzq159BSew4ttefQUnsOreFqTyUVREREZFSw2+1nfY76+rYhiOQrgYGeVFc3D+k5L2Rqz6Gl9hxaas+hpfYcWmfbnqdKSGj6g4iIiJyXrFYrNTU1/Y+rqqoIDAwccF9lZSVWq3XEYxQRERntlFQQERGR89KMGTNYu3YtANnZ2VitVjw8PAAIDw+npaWF0tJSenp6WL9+PTNmzHBkuCIiIqOSpj+IiIjIeWnixIkkJyezePFiDAYDjz/+OKtWrcLT05PLLruMn//85/zXf/0XAFdddRUxMTEOjlhERGT0MdiHYgKiiIiIiIiIiFxwNP1BRERERERERAZFSQURERERERERGRQlFURERERERERkUJRUEBEREREREZFBUVJBRERERERERAZFSQURERERERERGRQlFURERERERERkUMyODmA4LFu2jMzMTAwGA48++ihpaWmODum886tf/Yrdu3fT09PDd7/7XVJTU/nRj35Eb28vgYGB/PrXv8ZisTg6zPNKR0cHCxcu5MEHH2TatGlqz7OwevVqXnjhBcxmMz/4wQ+Ij49Xew5Sa2srP/7xj2lsbKS7u5uHHnqIwMBAfv7znwMQHx/PL37xC8cGeR7Iy8vjwQcf5J577mHJkiVUVFQM+J5cvXo1L7/8MkajkVtuuYVFixY5OnQZRuqPnD31R4ae+iNDR/2RoaP+yNBwVH9k1I1U2LFjB8XFxaxcuZKlS5eydOlSR4d03tm2bRv5+fmsXLmSF154gWXLlvHHP/6R22+/nVdffZWoqCjeeustR4d53vnLX/6Ct7c3gNrzLNTX1/Pss8/y6quv8txzz/Hpp5+qPc/CO++8Q0xMDMuXL+cPf/hD//+bjz76KK+//jotLS1s3LjR0WGe09ra2njiiSeYNm1a/7aB3pNtbW08++yz/OMf/2D58uW8/PLLNDQ0ODByGU7qj5w99UeGh/ojQ0P9kaGl/sjZc2R/ZNQlFTIyMpg/fz4AsbGxNDY20tLS4uCozi+TJ0/mD3/4AwBeXl60t7ezfft2Lr30UgDmzp1LRkaGI0M87xQUFHD48GHmzJkDoPY8CxkZGUybNg0PDw+sVitPPPGE2vMs+Pr69n+RNDU14ePjQ1lZWf8dVbXn6VksFp5//nmsVmv/toHek5mZmaSmpuLp6YmLiwsTJ05kz549jgpbhpn6I2dP/ZGhp/7I0FF/ZGipP3L2HNkfGXVJhZqaGnx9ffsf+/n5UV1d7cCIzj8mkwk3NzcA3nrrLWbNmkV7e3v/8C1/f3+16Tf01FNP8ZOf/KT/sdpz8EpLS+no6OCBBx7g9ttvJyMjQ+15Fq6++mrKy8u57LLLWLJkCT/60Y/w8vLq36/2PD2z2YyLi8tx2wZ6T9bU1ODn59d/jL6fRjf1R86e+iNDT/2RoaP+yNBSf+TsObI/MiprKnyd3W53dAjnrXXr1vHWW2/x0ksvsWDBgv7tatNv5t1332X8+PFEREQMuF/t+c01NDTwpz/9ifLycu66667j2lDt+c289957hIaG8uKLL5Kbm8tDDz2Ep6dn/36159k7WRuqbS8s+vcePPVHhob6I0NP/ZGho/7I8BvO/sioSypYrVZqamr6H1dVVREYGOjAiM5Pmzdv5rnnnuOFF17A09MTNzc3Ojo6cHFxobKy8rhhNXJqGzZsoKSkhA0bNnDs2DEsFova8yz4+/szYcIEzGYzkZGRuLu7YzKZ1J6DtGfPHmbOnAlAQkICnZ2d9PT09O9Xew7OQJ/xgb6fxo8f78AoZTipPzI01B8ZOuqPDC31R4aW+iPDY6T6I6Nu+sOMGTNYu3YtANnZ2VitVjw8PBwc1fmlubmZX/3qV/z1r3/Fx8cHgOnTp/e368cff8wll1ziyBDPK08//TRvv/02b7zxBosWLeLBBx9Ue56FmTNnsm3bNmw2G/X19bS1tak9z0JUVBSZmZkAlJWV4e7uTmxsLLt27QLUnoM10HsyPT2dAwcO0NTURGtrK3v27GHSpEkOjlSGi/ojZ0/9kaGl/sjQUn9kaKk/MjxGqj9isI/CsSS/+c1v2LVrFwaDgccff5yEhARHh3ReWblyJc888wwxMTH925588kl+9rOf0dnZSWhoKL/85S9xcnJyYJTnp2eeeYawsDBmzpzJj3/8Y7XnIL3++uv9FZW/973vkZqaqvYcpNbWVh599FFqa2vp6enh4YcfJjAwkMceewybzUZ6ejo//elPHR3mOS0rK4unnnqKsrIyzGYzQUFB/OY3v+EnP/nJCe/JNWvW8OKLL2IwGFiyZAnXXnuto8OXYaT+yNlRf2T4qD8yNNQfGTrqj5w9R/ZHRmVSQURERERERESG36ib/iAiIiIiIiIiI0NJBREREREREREZFCUVRERERERERGRQlFQQERERERERkUFRUkFEREREREREBkVJBREZlFWrVvHII484OgwRERG5QKkvInJuUFJBRERERERERAbF7OgARGR4LV++nI8++oje3l7GjBnDfffdx3e/+11mzZpFbm4uAL///e8JCgpiw4YNPPvss7i4uODq6soTTzxBUFAQmZmZLFu2DCcnJ7y9vXnqqacAaGlp4ZFHHqGgoIDQ0FD+9Kc/YTAYHPlyRURE5ByjvojI6KaRCiKj2P79+/nkk09YsWIFK1euxNPTk61bt1JSUsKNN97Iq6++ypQpU3jppZdob2/nZz/7Gc888wzLly9n1qxZPP300wD893//N0888QT//Oc/mTx5Mhs3bgTg8OHDPPHEE6xatYr8/Hyys7Md+XJFRETkHKO+iMjop5EKIqPY9u3bOXr0KHfddRcAbW1tVFZW4uPjQ0pKCgATJ07k5ZdfpqioCH9/f4KDgwGYMmUKr7/+OnV1dTQ1NREXFwfAPffcA/TNY0xNTcXV1RWAoKAgmpubR/gVioiIyLlMfRGR0U9JBZFRzGKxMG/ePB577LH+baWlpdx44439j+12OwaD4YShgl/fbrfbBzy/yWQ64TkiIiIiX1JfRGT00/QHkVFs4sSJbNq0idbWVgBWrFhBdXU1jY2N5OTkALBnzx7i4+OJjo6mtraW8vJyADIyMkhPT8fX1xcfHx/2798PwEsvvcSKFSsc84JERETkvKK+iMjop5EKIqNYamoqd9xxB3feeSfOzs5YrVYuvvhigoKCWLVqFU8++SR2u53f/e53uLi4sHTpUv7zP/8Ti8WCm5sbS5cuBeDXv/41y5Ytw2w24+npya9//Ws+/vhjB786EREROdepLyIy+hnsGiMkckEpLS3l9ttvZ9OmTY4ORURERC5A6ouIjC6a/iAiIiIiIiIig6KRCiIiIiIiIiIyKBqpICIiIiIiIiKDoqSCiIiIiIiIiAyKkgoiIiIiIiIiMihKKoiIiIiIiIjIoCipICIiIiIiIiKD8v8Bxoy6NFnhleMAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred = model.predict(test_x)\n",
        "y_pred = np.argmax(y_pred, axis = 1)\n",
        "test_y_mod = np.argmax(test_y, axis = 1)\n",
        "\n",
        "f1_score(test_y_mod, y_pred, average='weighted')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8pAEtwt8aiyQ",
        "outputId": "b8596073-ec0d-4f53-ae47-9d6b9e2149ae"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "10/10 [==============================] - 0s 9ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8331698454849302"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "keras.utils.plot_model(model, 'model.png', show_shapes = True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "xb63-tCwai01",
        "outputId": "3449eb03-51fb-4662-9451-e325f09eb7cb"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def predict(path):\n",
        "    sig, sr = librosa.load(path, res_type = 'kaiser_fast')\n",
        "    mfccs_features = librosa.feature.mfcc(y=sig, sr=sr, n_mfcc=50)\n",
        "    mfcc = np.array(np.mean(mfccs_features.T,axis=0))\n",
        "    mfcc = np.reshape(mfcc, (1, -1, 1))\n",
        "\n",
        "    plt.plot(sig)\n",
        "    # print(mfcc.shape)\n",
        "    pred = np.argmax(model.predict(mfcc, verbose = 0))\n",
        "    pred = le.inverse_transform([pred])\n",
        "    return pred[0]\n",
        "\n",
        "datapoint_path = df[df['outcome'] == 'hindi'].iloc[0, 0]\n",
        "predict(datapoint_path)\n",
        "\n",
        "# Our model predicted Hindi correctly!"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 366
        },
        "id": "CVtQs6yNai3U",
        "outputId": "721babc3-71ea-42e0-853f-2ed88a201f0d"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'hindi'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 15
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x396 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "OnO1uLFgai56"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}