project3.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "machine_shape": "hm"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "pip install jamo"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MhZtzeOLQmcW",
        "outputId": "ec3bfdf2-d263-489a-e733-9a0a6dd2d977"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting jamo\n",
            "  Downloading jamo-0.4.1-py3-none-any.whl (9.5 kB)\n",
            "Installing collected packages: jamo\n",
            "Successfully installed jamo-0.4.1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.layers import Input, Conv1D, LSTM, Dense, Embedding, RepeatVector, Concatenate, Reshape, TimeDistributed\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras.optimizers import Adam\n",
        "from tensorflow.keras.losses import MeanSquaredError\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint\n",
        "from scipy.io.wavfile import read, write\n",
        "import librosa\n",
        "from tqdm import tqdm"
      ],
      "metadata": {
        "id": "qRYpRa2abVFj"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iVpcbJD4fUpD",
        "outputId": "ba69b461-ee48-4dfc-bf91-95395c39840d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data_path = '/content/drive/MyDrive/dataset/voice/ljs_audio_text_train_filelist.txt'\n",
        "val_data_path =  '/content/drive/MyDrive/dataset/voice/ljs_audio_text_val_filelist.txt'\n",
        "wav_path = '/content/drive/MyDrive/dataset/voice/'"
      ],
      "metadata": {
        "id": "_v_AcHeQ1nES"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import random\n",
        "\n",
        "def load_data(file_path, wav_path, subset_ratio=None):\n",
        "    data = []\n",
        "    with open(file_path, 'r') as f:\n",
        "        lines = f.readlines()\n",
        "        if subset_ratio:\n",
        "            random.shuffle(lines)\n",
        "            num_samples = int(len(lines) * subset_ratio)\n",
        "            lines = lines[:num_samples]\n",
        "        for line in lines:\n",
        "            wav_file, text = line.strip().split(\"|\")\n",
        "            wav_file_path = os.path.join(wav_path, wav_file[10:])\n",
        "            audio, _ = librosa.load(wav_file_path, sr=None)\n",
        "            data.append((audio, text))\n",
        "    return data\n",
        "\n",
        "# def load_wav(wav_path):\n",
        "#     sr, wav = read(wav_path)\n",
        "#     # Convert stereo audio to mono if necessary\n",
        "#     if wav.ndim > 1:\n",
        "#         wav = wav[:, 0]\n",
        "#     return sr, wav\n",
        "\n",
        "# def preprocess_data(data):\n",
        "#     preprocessed_data = []\n",
        "#     for wav_path, text in data:\n",
        "#         sr, wav = load_wav(wav_path)\n",
        "#         # Resample audio to 22,050Hz if necessary\n",
        "#         wav = librosa.resample(wav, sr, 22050)\n",
        "#         preprocessed_data.append((wav, text))\n",
        "#     return preprocessed_data"
      ],
      "metadata": {
        "id": "Gt-Tua-8pGJb"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# 데이터셋이 많아 학습 및 로딩시간이 매우 기므로 20%만 랜덤 추출 (시험용)\n",
        "\n",
        "train_data = load_data(train_data_path, wav_path, subset_ratio=0.2)\n",
        "val_data = load_data(val_data_path, wav_path, subset_ratio=0.2)"
      ],
      "metadata": {
        "id": "hqNv-kCQ3Hit"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(train_data[0][0].shape)\n",
        "print(train_data[0][1])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1oXK_-rmJnA6",
        "outputId": "10846df4-4593-4aa7-8afc-0a8af98771a7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(50936,)\n",
            "모르는 사람이랑도 페이스북 친구 하시나요?\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(val_data[0][0].shape)\n",
        "print(val_data[0][1])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ch86a6_2lS5F",
        "outputId": "06a4d24e-e1bf-4ab6-df28-b83826d1e12e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(44762,)\n",
            "오늘은 우리나라의 큰 명절입니다.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import librosa.display\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# 오디오 데이터 로드\n",
        "audio_data = train_data[0][0]\n",
        "\n",
        "# Mel Spectrogram 계산\n",
        "S = librosa.feature.melspectrogram(y=audio_data, sr=22050)\n",
        "\n",
        "# 로그 Mel Spectrogram 계산\n",
        "log_S = librosa.power_to_db(S, ref=np.max)\n",
        "\n",
        "# 그림 그리기\n",
        "plt.figure(figsize=(10, 4))\n",
        "librosa.display.specshow(log_S, sr=22050, x_axis='time', y_axis='mel')\n",
        "plt.colorbar(format='%+2.0f dB')\n",
        "plt.title('Mel-Spectrogram')\n",
        "plt.tight_layout()\n",
        "plt.savefig('Mel-Spectrogram example.png')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 407
        },
        "id": "9yt2yP3dJ2Z2",
        "outputId": "2390de12-7eb0-41bc-ffde-d21b5f8831ba"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x400 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def calculate_average_and_max_length(matrix_list):\n",
        "    total_length = 0\n",
        "    max_length = 0\n",
        "    num_matrices = len(matrix_list)\n",
        "\n",
        "    for matrix in matrix_list:\n",
        "        matrix_length = len(matrix)\n",
        "        total_length += matrix_length\n",
        "\n",
        "        if matrix_length > max_length:\n",
        "            max_length = matrix_length\n",
        "\n",
        "    average_length = total_length / len(matrix_list)\n",
        "\n",
        "    return average_length, max_length\n",
        "\n",
        "# 평균과 최대 길이 계산\n",
        "average_length, max_length = calculate_average_and_max_length(train_data[:][0])\n",
        "print(\"평균 길이:\", average_length)\n",
        "print(\"최대 길이:\", max_length)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OO6SMwGevvtM",
        "outputId": "a64c8570-0610-4806-c212-ec1c547e137e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "평균 길이: 25479.5\n",
            "최대 길이: 50936\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "max_text_length = 140 # 최대 문장 길이\n",
        "vocab_size = 10000 # 단어 사전 크기\n",
        "embedding_dim = 256 # 임베딩 차원\n",
        "latent_dim = 64 # 잠재 공간 차원\n",
        "output_length = 100 # 출력 길이\n",
        "output_dim = 400  # 출력 차원\n",
        "max_audio_length = 200   # 음성 길이"
      ],
      "metadata": {
        "id": "T2qN7KtIsGt-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.models import Model\n",
        "from keras import backend as K\n",
        "from keras.layers.core import Lambda\n",
        "\n",
        "'''\n",
        "latent_dim = 64\n",
        "\n",
        "# Encoder\n",
        "text_input = Input(shape=(max_text_length,), dtype='float32')  # Update the input type to float32\n",
        "text_encoded = Reshape((max_text_length, 1))(text_input)\n",
        "text_encoded = LSTM(latent_dim)(text_encoded)\n",
        "\n",
        "audio_input = Input(shape=(audio_length, output_dim))\n",
        "audio_encoded = LSTM(latent_dim)(audio_input)\n",
        "\n",
        "concatenated = Concatenate()([text_encoded, audio_encoded])\n",
        "\n",
        "# VAE part\n",
        "z_mean = Dense(latent_dim)(concatenated)\n",
        "z_log_var = Dense(latent_dim)(concatenated)\n",
        "\n",
        "# Sampling\n",
        "def sampling(args):\n",
        "    z_mean, z_log_var = args\n",
        "    epsilon = K.random_normal(shape=(K.shape(z_mean)[0], latent_dim), mean=0., stddev=0.1)\n",
        "    return z_mean + K.exp(z_log_var / 2) * epsilon\n",
        "\n",
        "z = Lambda(sampling, output_shape=(latent_dim,))([z_mean, z_log_var])\n",
        "\n",
        "# Decoder\n",
        "decoder_input = RepeatVector(max_text_length)(z)\n",
        "decoder_text_input = Input(shape=(max_text_length,), dtype='int32')\n",
        "decoder_embedded_text = Embedding(input_dim=vocab_size, output_dim=embedding_dim)(decoder_text_input)\n",
        "concatenated_decoder_input = Concatenate(axis=2)([decoder_input, decoder_embedded_text])\n",
        "decoded = LSTM(latent_dim, return_sequences=True)(concatenated_decoder_input)\n",
        "decoded = TimeDistributed(Dense(output_dim))(decoded)\n",
        "\n",
        "# Slice the output tensor to match desired sequence length (100, 256)\n",
        "decoded = Lambda(lambda x: x[:, :100, :])(decoded)\n",
        "\n",
        "# Create models\n",
        "encoder = Model([text_input, audio_input], z_mean)\n",
        "decoder = Model([decoder_text_input, decoder_input], decoded)\n",
        "\n",
        "output = decoder([decoder_text_input, decoder_input])\n",
        "vae = Model([text_input, decoder_text_input, audio_input], output)\n",
        "'''\n",
        "\n",
        "# # Encoder\n",
        "# text_input = Input(shape=(max_text_length,), dtype='float32')\n",
        "# text_encoded = Reshape((max_text_length, 1))(text_input)\n",
        "# text_encoded = LSTM(latent_dim)(text_encoded)\n",
        "\n",
        "# audio_input = Input(shape=(max_audio_length, output_dim))  # Update the audio input shape\n",
        "# audio_encoded = LSTM(latent_dim)(audio_input)\n",
        "\n",
        "# concatenated = Concatenate()([text_encoded, audio_encoded])\n",
        "\n",
        "# # VAE part\n",
        "# z_mean = Dense(latent_dim)(concatenated)\n",
        "# z_log_var = Dense(latent_dim)(concatenated)\n",
        "\n",
        "# # Sampling\n",
        "# def sampling(args):\n",
        "#     z_mean, z_log_var = args\n",
        "#     epsilon = K.random_normal(shape=(K.shape(z_mean)[0], latent_dim), mean=0., stddev=0.1)\n",
        "#     return z_mean + K.exp(z_log_var / 2) * epsilon\n",
        "\n",
        "# z = Lambda(sampling, output_shape=(latent_dim,))([z_mean, z_log_var])\n",
        "\n",
        "# # Decoder\n",
        "# decoder_input = RepeatVector(max_audio_length)(z)\n",
        "\n",
        "# decoder_text_input = Input(shape=(max_text_length,), dtype='int32')\n",
        "# # Pass the text input through an Embedding layer before the LSTM layer\n",
        "# # Here, replace 'embedding_dim' with the desired size of your embeddings\n",
        "# decoder_text_embedded = Embedding(input_dim=vocab_size, output_dim=embedding_dim)(decoder_text_input)\n",
        "# # Pass the embedded text input through an LSTM layer\n",
        "# # Set return_sequences=False so the output is 2D\n",
        "# decoder_text_encoded = LSTM(latent_dim, return_sequences=False)(decoder_text_embedded)\n",
        "\n",
        "# # Now that the output of the LSTM is 2D, we can use RepeatVector\n",
        "# decoder_text_encoded = RepeatVector(max_audio_length)(decoder_text_encoded)\n",
        "\n",
        "# # Concatenate the decoder input and the encoded text\n",
        "# concatenated_decoder_input = Concatenate(axis=2)([decoder_input, decoder_text_encoded])\n",
        "\n",
        "# # Pass the concatenated input through another LSTM layer\n",
        "# decoded = LSTM(latent_dim, return_sequences=True)(concatenated_decoder_input)\n",
        "\n",
        "# # Use a Dense layer to get the final output shape\n",
        "# decoded = TimeDistributed(Dense(output_dim))(decoded)  # Update the Dense layer's output dimension\n",
        "\n",
        "# # Adjust the output to match the desired shape\n",
        "# decoded = Reshape((max_audio_length, output_dim))(decoded)  # Update the reshape layer\n",
        "\n",
        "# # Create models\n",
        "# encoder = Model([text_input, audio_input], z_mean)\n",
        "# decoder = Model([decoder_text_input, decoder_input], decoded)\n",
        "\n",
        "# output = decoder([decoder_text_input, decoder_input])\n",
        "# vae = Model([text_input, decoder_text_input, audio_input], output)\n",
        "\n",
        "latent_dim = 64\n",
        "\n",
        "# Encoder\n",
        "text_input = Input(shape=(140,), dtype='float32')  # Updated text input shape\n",
        "text_encoded = Reshape((140, 1))(text_input)  # Updated reshape\n",
        "text_encoded = LSTM(latent_dim)(text_encoded)\n",
        "\n",
        "audio_input = Input(shape=(200, 400))  # Updated audio input shape\n",
        "audio_encoded = LSTM(latent_dim)(audio_input)\n",
        "\n",
        "# Concatenate the encoded text and audio\n",
        "encoded = Concatenate()([text_encoded, audio_encoded])\n",
        "\n",
        "# VAE part\n",
        "z_mean = Dense(latent_dim)(encoded)\n",
        "z_log_var = Dense(latent_dim)(encoded)\n",
        "\n",
        "# Sampling\n",
        "def sampling(args):\n",
        "    z_mean, z_log_var = args\n",
        "    epsilon = K.random_normal(shape=(K.shape(z_mean)[0], latent_dim), mean=0., stddev=0.1)\n",
        "    return z_mean + K.exp(z_log_var / 2) * epsilon\n",
        "\n",
        "z = Lambda(sampling, output_shape=(latent_dim,))([z_mean, z_log_var])\n",
        "\n",
        "# Decoder\n",
        "decoder_input = RepeatVector(200)(z)  # Updated to match the desired sequence length\n",
        "decoded = LSTM(latent_dim, return_sequences=True)(decoder_input)\n",
        "decoded = Dense(400)(decoded)  # The output dimension should match the audio data\n",
        "\n",
        "# Create the models\n",
        "encoder = Model([text_input, audio_input], z_mean)\n",
        "decoder = Model(z, decoded)\n",
        "vae = Model([text_input, audio_input], decoder(z))"
      ],
      "metadata": {
        "id": "8JsJN82vpgTF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# 모델 생성\n",
        "model = vae\n",
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0hdg9xcVvPBC",
        "outputId": "3b1c9c78-2501-4077-85f0-b232de809825"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_12\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " input_15 (InputLayer)          [(None, 140)]        0           []                               \n",
            "                                                                                                  \n",
            " reshape_7 (Reshape)            (None, 140, 1)       0           ['input_15[0][0]']               \n",
            "                                                                                                  \n",
            " input_16 (InputLayer)          [(None, 200, 400)]   0           []                               \n",
            "                                                                                                  \n",
            " lstm_13 (LSTM)                 (None, 64)           16896       ['reshape_7[0][0]']              \n",
            "                                                                                                  \n",
            " lstm_14 (LSTM)                 (None, 64)           119040      ['input_16[0][0]']               \n",
            "                                                                                                  \n",
            " concatenate_7 (Concatenate)    (None, 128)          0           ['lstm_13[0][0]',                \n",
            "                                                                  'lstm_14[0][0]']                \n",
            "                                                                                                  \n",
            " dense_11 (Dense)               (None, 64)           8256        ['concatenate_7[0][0]']          \n",
            "                                                                                                  \n",
            " dense_12 (Dense)               (None, 64)           8256        ['concatenate_7[0][0]']          \n",
            "                                                                                                  \n",
            " lambda_4 (Lambda)              (None, 64)           0           ['dense_11[0][0]',               \n",
            "                                                                  'dense_12[0][0]']               \n",
            "                                                                                                  \n",
            " model_11 (Functional)          (None, 200, 400)     59024       ['lambda_4[0][0]']               \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 211,472\n",
            "Trainable params: 211,472\n",
            "Non-trainable params: 0\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow import keras\n",
        "keras.utils.plot_model(model, show_shapes=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 754
        },
        "id": "00npcQitxn1Q",
        "outputId": "cf394138-8007-40ee-f3a7-37709105e3a7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAApsAAALhCAIAAACfSfiUAAAABmJLR0QA/wD/AP+gvaeTAAAgAElEQVR4nOzda1gT174w8DWBkAskEOQWQZSLSFHUemklSmnrqVZ8pCJYqdK9xa1FqyJeKCKCqKAibuBgpR53KT2t3QoWDliRuh/1QUtFj91KQayIKCIgdyFcEghk3g/z7jnZAULInfH/+8SsGWatWbPW/DMza2YwHMcRAAAAAMY5mr4LAAAAAAANgIgOAAAAUAFEdAAAAIAKIKIDAAAAVGAsO1FSUpKcnKyvogAAvLy8du3ape9SAADGpX87R3/x4sWPP/6or6KMU7dv3759+7a+S6EVdXV10B506fbt2yUlJfouBQBgvDIemnThwgXdl2P8Wr16NaJopWVnZ69Zs4aSm2aYiLYEAACqgfvoAAAAABVARAcAAACoACI6AAAAQAUQ0QEAAAAqgIgOAAAAUIEmI/rly5fNzc1/+uknDa5THVKpNCUlRSAQyCbGx8dj/27GjBk6LpihVZSaNm/eTFZmcHCw7KyrV69GRUXl5OQ4OzsTC3z66aeyCyxZsoTD4RgZGU2fPv3evXu6LThCIzQSWWKx2N3dff/+/WRKcXHxwoUL2Ww2n8+PjIzs6+tDCF28eDExMXFwcJBcLC8vj6wWKysrrW4FAAAgzUZ0g/qMW1VV1TvvvLNr167e3l59l0WeQVWURlhaWhYWFlZWVmZkZJCJBw4cSEtL27dvX0BAwNOnT11cXCZMmHD27NmCggJymX/84x8XLlxYsWJFRUXFnDlzdFxsZRpJdHR0ZWUlOVlRUbFkyZLFixe3tLTk5uZ+8803W7ZsQQj5+fkxmczFixd3dHQQS3700Ud1dXU3b9709fXV9oYAAADSbERfvnx5Z2fnihUrNLhOWSKRSMG5lKzff/997969W7ZsmT179tC533//PS7jwYMHmi7pKAynojSFxWJ9+OGHbm5uDAaDSDl27Nj58+ezs7M5HA65WFpaGo1GCw0N7ezs1GXxhqW4kRBu3bol1zwOHz5sZ2d38OBBU1NTLy+vyMjIb7/99tGjRwihHTt2zJo1y9fXd2BgACGEYZi9vb23t/fUqVO1vS0AAIDG1330jIyM5uZmZZacNWtWTk7OunXryADzWlG+orTkyZMnMTExBw8eZDKZsukCgSA8PLy+vn7Pnj36Khtp1EYiEokiIiJSU1PJlIGBgYKCAh8fHwzDiJRly5bhOJ6fn09MxsXFlZaWyv4LAADojMYienFxsaOjI4ZhX375JUIoPT3d1NSUzWbn5+cvW7aMy+U6ODicO3cOIZSWlsZkMm1sbDZv3szn85lMpkAguHPnDkIoLCzMxMTEzs6OWOfWrVtNTU0xDGttbQ0PD9+9e3d1dTWGYa6urpoqtu7pvqJ+/vlnLpebkJCgs21MS0vDcdzPz2/orPj4eDc3t6+//vrq1atD5+I4npyc/MYbbzAYDB6Pt3LlSuL0V0EtIYQGBwdjY2MdHR1ZLNbMmTOzsrI0shXR0dFbt261trYmU54+fdrd3e3o6EimuLi4IITKysqISR6P5+Pjk5qaSr0bKwAAw6exiL5o0aJbt26Rk59//vnOnTtFIhGHw8nKyqqurnZ2dt60aZNEIgkLC1u/fn1vb++OHTtqamru3bs3MDDwwQcfvHjxIi0t7eOPPyZXcurUqYMHDxJ/p6amrlixwsXFBcfxJ0+eqFPUqKgoHo9nYmLi5OS0cuXKu3fvqrO2sdJ9RRHDtaRSqc62saCgYNq0aWw2e+gsFov17bff0mi0TZs29fT0yM2Ni4uLioqKjo5ubm6+efPmixcvvL29m5qaFNQSQmjv3r3Hjx9PSUl5+fLlihUr1q5d+9tvv6m5Cb/++mt1dfXatWtlExsbGxFCsvcRmEwmi8VqamoiU9588836+vrff/9dzQIAAMBYaf2qu0Ag4HK51tbWQUFBPT09tbW1RLqxsTFxKubh4ZGent7V1ZWZmantwiCE/vznP1+8ePHFixfd3d3nzp2rra318fGpqKjQQdaKaa+ili9fLhQKY2JitFDqYfT09Dx79ow4eR2Wl5fXzp07a2pq9u7dK5suEomSk5NXrVoVHBxsbm7u6el5+vTp1tbWM2fOkMsMrSWxWJyenu7v7x8QEGBhYbF//346na5mWxKJROHh4enp6XLpxLB2IyMj2UQ6nS4SichJ4q55eXm5OgUAAAAV6O4+uomJCUKIOKmSM2/ePDabTVxf1bZJkya9+eabZmZmJiYmCxYsyMzMFIlEp06d0kHWSjKQilJZc3MzjuPDnqCT4uPjp02bdurUqeLiYjKxoqKiu7t73rx5ZMr8+fNNTEyIGw1yyFqqrKzs7e0lH0FksVh2dnZqVtG+ffs+++wze3t7uXRiWAAx8I3U39/PYrHISWLDZc/aAQBANwxlZByDwWhpadF9vp6enkZGRo8fP9Z91qrRV0UpTywWI4QUj0lkMpmZmZkYhm3YsIE8wSWe+zIzM5Nd0sLCoqurS8GqiEv3+/fvJx/+fv78uTqPLBYXF5eXl2/cuHHoLGLgglAoJFN6e3vFYjGfzydTiOhOVAIAAOiSQUR0iUTS0dHh4OCg+6ylUqlUKh0vQ+L1WFHKI0Ka7LtWhuXl5bVr166qqqrDhw8TKRYWFgghufg96vYSI9dSUlJkn0hU5yvjGRkZ165do9FoxO8DYv0JCQkYhrW1tXE4nOfPn5MLEyMVZs6cSab09/ejf1UCAADokkFE9KKiIhzHFyxYgBAyNjYe9oKzpixdulR28u7duziOe3l5aS9HDdJlRanMxsYGwzBlnjg/fPiwu7v7/fv3ickZM2aYmZnJDmq7c+dOf3//3LlzFaxk0qRJTCaztLRUzWKTMjMzZX8cEFdEoqOjiZr39fW9efMmOcywsLAQwzDZUf3Ehtva2mqqPAAAoCS9RXSpVPrq1auBgYGysrLw8HBHR8f169cjhFxdXdvb2/Py8iQSSUtLi+z5kKWlZUNDQ01NTVdXl8rBrL6+/vz58x0dHRKJpKSkZOPGjY6OjsRrvwyT+hVVWFioy6fX2Gy2s7NzXV3dqEsS197JgWZMJnP37t25ublnz54VCoXl5eVbtmzh8/mhoaGKVxISEnLu3Ln09HShUDg4OFhXV/fy5UuEUFBQkK2trWZfLhsTE9PU1HTgwIGenp6SkpKkpKT169dPmzaNXIDYcE9PTw1mCgAASpE9HSEe5MVVcvLkSeIuI5vN9vPzO3XqFDFEaOrUqdXV1WfOnOFyuQihyZMnP378ODQ0lE6n29vbGxsbc7nclStXVldXE+tpa2t77733mEymk5PT9u3bIyIiEEKurq61tbX37t2bPHkyi8VatGhRY2OjgsKUlJQsXLiQvLtpZ2cnEAhu3LiB4/ju3btdXFxMTU2NjY0dHBw2bdrU0NCg2iYTAgMDAwMDDbmiLl++zOFw4uPjx7ppSraH0NBQe3t72ZSwsDA6nd7b20tM5ubmEkPfraystm3bJvfvERERH330EfG3VCpNSkqaOnUqnU7n8Xj+/v6VlZU4jiuupb6+vsjISEdHR2NjY2tr64CAgIqKChzH/f39EUKxsbHDFltBI5Ele45OuHHjxltvvcVgMPh8fkREhFgsll1++fLl9vb2UqmUTNmxY8eECRNGrUZ87G0JAABkaSyij0loaKilpaUOMtIBrR6F9VtRKkf0qqoqY2Njubft6t7g4KC3t3dGRobOcmxtbWUymSdOnJBNhIgOANANvV11H3XkFCCMi4oSiURXrlypqqoixoW5uroeOnTo0KFD3d3d+irS4OBgXl5eV1dXUFCQzjKNi4ubPXt2WFgYQgjH8YaGhuLiYjVfiAQAAEoyiJFxY/Xo0SNsZLo8ggNCe3s78aWWDRs2EClRUVGrV68OCgrS10dZioqKcnJyCgsLFT8Zr0HJycmlpaWXL1+m0+kIofz8fOJLLbLfmgMAAO3RQ0Tft29fZmZmZ2enk5PTjz/+qMIa3N3dFVx2OH/+vMbLrBfqV5RunD59mqz8s2fPkukJCQlhYWFHjx7VS6kWL178ww8/kK++17b8/Py+vr6ioiIej0ekrFy5kqyW1tZW3RQDAPA6w3CZT0pkZ2evWbMGh49MjMXq1asRQhcuXNB3QTQP2oOOUbgtAQB0YFxedQcAAACAHIjoAAAAABVARAcAAACoACI6AAAAQAUQ0QEAAAAqMB6ahGGY7ssx3lG40ii8aQYoMDBQ30UAAIxXw0R04t2fQEkpKSkIoZ07d+q7IJpXUlKSmpoK7UFniLYEAACqGSaif/zxx7ovx/hFPD1M1UpLTU2l6qYZIHgSHQCgDriPDgAAAFABRHQAAACACiCiAwAAAFQAER0AAACgAojoAAAAABWMOaLfvn37jTfeoNFoGIbZ2trGx8dro1iycnJynJ2diW+f29nZBQcHaztHMCabN28mP04vt3euXr0aFRUluwc//fRT2QWWLFnC4XCMjIymT59+79493RYcIYSkUmlKSopAIBhpAbFY7O7uvn//fjKluLh44cKFbDabz+dHRkb29fUhhC5evJiYmDg4OEgulpeXR1aLlZWVVrcCAAAQQkj2y+LEk8cKPj1OWrp0KULo1atXyiysES4uLubm5jrLTnmBgYGBgYH6LoVWKNkeQkNDLS0tCwsLKysrxWIxmR4bG7tixQqhUEhMuri4TJgwASF06dIl2X8vLCz86KOPNFtyJT1+/HjhwoUIoVmzZo20zK5duxBC0dHRxOSDBw9YLFZMTEx3d/etW7esrKxCQkKIWampqT4+PmSnkEqldXV1N2/e9PX1nTBhgjLloXBbAgDogIFedReJRApOm14fGqkHHVQmi8X68MMP3dzcGAwGkXLs2LHz589nZ2dzOBxysbS0NBqNFhoa2tnZqdXyKOP333/fu3fvli1bZs+ePdIyt27devDggWzK4cOH7ezsDh48aGpq6uXlFRkZ+e233z569AghtGPHjlmzZvn6+g4MDCCEMAyzt7f39vaeOnWqtrcFAACQwd5Hz8jIaG5u1ncp9E8j9aD7ynzy5ElMTMzBgweZTKZsukAgCA8Pr6+v37Nnjy7LM6xZs2bl5OSsW7eO/BUiRyQSRUREpKamkikDAwMFBQU+Pj7km3GXLVuG43h+fj4xGRcXV1paKvsvAACgMxqI6Onp6aampmw2Oz8/f9myZVwu18HB4dy5cwihtLQ0JpNpY2OzefNmPp/PZDIFAsGdO3cQQmFhYSYmJnZ2dsRKtm7dampqimFYa2treHj47t27q6urMQxzdXVVpgy//PKLh4eHubk5k8n09PS8cuUKQmjjxo3EXUwXF5f79+8jhEJCQthstrm5+cWLFwcHB2NjYx0dHVks1syZM4krzMePH2ez2RwOp7m5effu3fb29pWVlepXEUIIx/Hk5OQ33niDwWDweLyVK1cSJ3bK14OmKvPnn3/mcrkJCQka2a5hpaWl4Tju5+c3dFZ8fLybm9vXX3999erVoXNHqiUFbQwhNOyuVF90dPTWrVutra3JlKdPn3Z3dzs6OpIpLi4uCKGysjJiksfj+fj4pKam4jiukTIAAMAYyF6CV/k+enR0NELo2rVrnZ2dzc3N3t7epqam/f39OI6Hhoaampo+fPhQLBZXVFTMnz+fw+HU1tbiOL5u3TpbW1tynUlJSQihlpYWHMcDAgJcXFxkc1R8H/3ChQtxcXHt7e1tbW0LFiwgb1sGBAQYGRnV19eTS65du/bixYs4ju/Zs4fBYPz444+vXr3at28fjUa7e/cuuS07duw4efLkqlWr/vjjD8VVoeS9z9jYWBMTk++//76jo6OsrGzOnDlWVlaNjY1jqgeNVOalS5c4HM6hQ4dGLbPy99Ht7e1lU5ydnT08POQWc3FxefbsGY7jt27dotFoU6ZM6e7uxv/9PrqCWlLQxkbalUp6++23h95HLy4u9vPzw3G8paUF/es++o0bNxBCSUlJskuyWKzFixeTk1FRUQih+/fvkyk7duyA++gAAB3Q5FV3gUDA5XKtra2DgoJ6enpqa2uJdGNjY+Ksy8PDIz09vaurKzMzU4P5IoQCAwMPHDjA4/EsLS39/Pza2tqIA/GWLVsGBwfJ7IRC4d27d319fcVicXp6ur+/f0BAgIWFxf79++l0umypjh07tm3btpycHHd3d/WLJxKJkpOTV61aFRwcbG5u7unpefr06dbW1jNnzox1VepX5vLly4VCYUxMzFizVlJPT8+zZ8+Ik9dheXl57dy5s6amZu/evbLpytTS0DY26q5UgUgkCg8PT09Pl0snhrUbGRnJJtLpdJFIRE4Sd83Ly8vVKQAAAKhAK/fRTUxMEEISiWTorHnz5rHZbOJSqpbQ6XSEEPEc0fvvv+/m5vbNN9/gOI4QOn/+fFBQkJGRUWVlZW9v74wZM4h/YbFYdnZ22itVRUVFd3f3vHnzyJT58+ebmJgQ18xVpoPKVEFzczOO42w2W8Ey8fHx06ZNO3XqVHFxMZk4ploi25g2duW+ffs+++wze3t7uXRiWAAx8I3U39/PYrHISWLDm5qa1CkAAACoQA8j4xgMBnECrUEFBQXvvvuutbU1g8H44osvyHQMwzZv3vz06dNr164hhL777ru//OUvCKGenh6E0P79+8knhp8/f97b26vZUpE6OjoQQmZmZrKJFhYWXV1daq5ZG5WpJrFYjBAaabgZgclkZmZmYhi2YcMG8gRXtVrS+K4sLi4uLy/fuHHj0FnESAWhUEim9Pb2isViPp9PphDRnagEAADQJV1HdIlE0tHR4eDgoJG13bx5MyUlpba21t/f387O7s6dO52dnYmJibLLrF+/nslkfv3115WVlVwud/LkyQghYrhTSkqK7B2IkpISjZRqKAsLC4SQXGRSvx40W5maQoQ02XetDMvLy2vXrl1VVVWHDx8mUlSrJY3vyoyMjGvXrhHvUMIwjFh/QkIChmFtbW0cDuf58+fkwk+ePEEIzZw5k0zp7+9H/6oEAADQJV1H9KKiIhzHFyxYgBAyNjYe9sq88v75z3+ampqWl5dLJJLPP//c2dmZyWSSTxYReDzemjVr8vLyTpw4sWnTJiJx0qRJTCaztLRUndyVN2PGDDMzs99++41MuXPnTn9//9y5c5Ea9aDZytQUGxsbDMOUeeL88OHD7u7uxGMIaLRaGonGd2VmZqbsjwPZkXELFizw9fW9efOmVColFi4sLMQwTHZUP7Hhtra2mioPAAAoSRcRXSqVvnr1amBgoKysLDw83NHRcf369QghV1fX9vb2vLw8iUTS0tIie+pjaWnZ0NBQU1PT1dU1bKCSSCRNTU1FRUWmpqbE00RXr14Vi8VVVVVDb7tu2bKlr6/v0qVLK1asIFKYTGZISMi5c+fS09OFQuHg4GBdXd3Lly+1VANMJnP37t25ublnz54VCoXl5eVbtmzh8/mhoaFjrQf1K7OwsFCrT6+x2WxnZ+e6urpRlySuvZMDzRTXkoKVjLQrg4KCbG1tNfty2ZiYmKampgMHDvT09JSUlCQlJa1fv37atGnkAsSGe3p6ajBTAABQiuzpiDJPK92+fXv69Ok0Gg0hZGdnl5CQcOrUKWI00NSpU6urq8+cOcPlchFCkydPfvz4cWhoKJ1Ot7e3NzY25nK5K1eurK6uJlbV1tb23nvvMZlMJyen7du3R0REIIRcXV1ra2vv3bs3efJkFou1aNGir776SsHA6dzcXBzHIyMjLS0tLSwsVq9e/eWXXyKEXFxciMe6CG+++WZUVJTshvT19UVGRjo6OhobG1tbWwcEBFRUVCQmJhLXSydNmvT999+P/qyA0k8cSaXSpKSkqVOn0ul0Ho/n7+9fWVk5pnpobGxUvzIbGxsvX77M4XDi4+NHLbPKT6+FhYXR6fTe3l5iMjc3l9iDVlZW27Ztk/v3iIgI8um1kWpJcRsbdlfiOO7v748Qio2NHbbYJSUlCxcuJG+B29nZCQSCGzduyC0me45OuHHjxltvvcVgMPh8fkREhOyLb3EcX758ub29vVQqJVPg6TUAgG6o+Dy68oiXfmt2nSrw9fV9+vSpNtasy6OwjitT5YheVVVlbGys5E8i7RkcHPT29s7IyNBZjq2trUwm88SJE7KJENEBALqhi6vuow6S0hLycn1ZWRlx8qqXYmiWvipTMZFIdOXKlaqqKmJcmKur66FDhw4dOtTd3a2vIg0ODubl5XV1dQUFBeks07i4uNmzZ4eFhSGEcBxvaGgoLi4mRs8BAIC2Geh73TUiMjKyqqrq8ePHISEh5IBqoA3t7e3El1o2bNhApERFRa1evTooKEhfH2UpKirKyckpLCxU/GS8BiUnJ5eWll6+fJl4I0J+fj7xpZaCggLdFAAA8JrTbkTft29fZmZmZ2enk5PTjz/+qNW8hmKz2e7u7v/xH/8RFxfn4eGh49w1Tr+VqcDp06fJaz5nz54l0xMSEsLCwo4ePaqXUi1evPiHH34g33Wvbfn5+X19fUVFRTwej0hZuXIlWS2tra26KQYA4HWG4TKflMjOzl6zZg0OH5kYi9WrVyOELly4oO+CaB60Bx2jcFsCAOgAla+6AwAAAK8PiOgAAAAAFUBEBwAAAKgAIjoAAABABcZDk7Kzs3VfjvGLeOsnJSuN+N4JJTfNMNXV1Rnad3cAAOPIMGPd9VgaAF5zgYGBMNYdAKAaDJ5NohIMw7Kysj7++GN9FwQAAICuwX10AAAAgAogogMAAABUABEdAAAAoAKI6AAAAAAVQEQHAAAAqAAiOgAAAEAFENEBAAAAKoCIDgAAAFABRHQAAACACiCiAwAAAFQAER0AAACgAojoAAAAABVARAcAAACoACI6AAAAQAUQ0QEAAAAqgIgOAAAAUAFEdAAAAIAKIKIDAAAAVAARHQAAAKACiOgAAAAAFUBEBwAAAKgAIjoAAABABRDRAQAAACqAiA4AAABQAUR0AAAAgAogogMAAABUABEdAAAAoAKI6AAAAAAVQEQHAAAAqAAiOgAAAEAFENEBAAAAKoCIDgAAAFABRHQAAACACoz1XQCglr/97W/t7e2yKfn5+c+ePSMnQ0JCbGxsdF4uAAAAuobhOK7vMgDVbd68+b/+678YDMbQWRKJhMfjNTY2GhvD7zYAAKA+uOo+vn3yyScIob7hGBkZrV27FsI5AAC8JuAcfXzDcdze3v7ly5fDzr1165aXl5eOiwQAAEAv4Bx9fMMwbN26dSYmJkNnTZw4ccGCBbovEgAAAL2AiD7uffLJJ/39/XKJJiYmf/7znzEM00uRAAAA6B5cdaeCqVOnPnnyRC6xrKzM09NTL+UBAACge3COTgXBwcF0Ol02xdXVFcI5AAC8ViCiU0FwcPDAwAA5SafTQ0JC9FgeAAAAugdX3Sli9uzZZWVlxN7EMKy6utrJyUnfhQIAAKA7cI5OEX/605+MjIwQQhiGzZ07F8I5AAC8biCiU8Qnn3wilUoRQkZGRn/605/0XRwAAAC6BhGdIvh8/sKFCzEMk0qlq1ev1ndxAAAA6BpEdOr49NNPcRx/99137ezs9F0WAAAAOof/u8DAQH2XCIDXF662rKwsfW8EAEBHAgMDZbv/MJ/xWLBgwc6dO3VfMsO3Zs2a8PBwQ35TekpKymeffWZqaqrCPyKEYL/rUUlJSWpqqqbWBnF9TCjc/ol2Be2Bkoh2K2uYiO7g4PDxxx/rpDzjzJo1a7y8vAy5chYtWjRx4kQV/vHChQsIIUPetNeBBiM67MoxoXb7T01NpeqmveaIdisL7qNTimrhHAAAAAVARAcAAACoACI6AAAAQAUQ0QEAAAAqgIgOAAAAUIEeInpISAiTycQwTCwW6z53wrvvvosNYWZmpvGMLl++bG5u/tNPP2l8zQbr6tWrUVFROTk5zs7ORMV++umnsgssWbKEw+EYGRlNnz793r17ui+hVCpNSUkRCAQjLSAWi93d3ffv30+mFBcXL1y4kM1m8/n8yMjIvr4+hNDFixcTExMHBwd1UWhdMbQWO9LOkkgkR44ccXV1NTExsbCwmDFjRk1NjS4LZmgVpabNmzeTR8Lg4GDZWYbcow8dOuTh4cHlchkMhqur6xdffNHd3U3OHbbbjjpLSUoeJRTMGnoAycvLI/eClZXVWIuE9BLRMzMz9+zZo/t8R7Vo0SKNr/N1+7TdgQMH0tLS9u3bFxAQ8PTpUxcXlwkTJpw9e7agoIBc5h//+MeFCxdWrFhRUVExZ84cHZewqqrqnXfe2bVrV29v70jLREdHV1ZWkpMVFRVLlixZvHhxS0tLbm7uN998s2XLFoSQn58fk8lcvHhxR0eHLoquEwbVYhXsrDVr1nz33Xc//PBDb2/vH3/84eLiInso1wGDqiiNsLS0LCwsrKyszMjIIBMNvEdfv35927ZtNTU1ra2tR44cSU1NJd+BPVK3VTxLeUoeJRTMGnoA+eijj+rq6m7evOnr66tijQx9Z5zcO2i0ITo6GiEkEom0ndFIli5dKhQKZVNCQ0OvXbum+L8QQllZWdos15j19vZ6eXmpvx6N7PejR4+6ubnJ7lYXF5cffviBRqPZ29t3dHSQ6YWFhR999JGa2amgtLR01apVZ8+enT179qxZs4Zd5tdff12yZAnRY4mUNWvWODk5SaVSYjIpKQnDsD/++IOYDAsL8/LykkgkapaNeAeImivR4Hq0RPkWq2BnnTt3DsMw4vPBGqGb496YaKprK9keQkND7e3t5RINv0cvX758YGCAnCQeu6+trcUVdlvFPVoZYzpKqHAA2bFjx4QJE0YtxtB2q8/76BiG6Svrn3/+mcPhkJMvXrx48ODB+++/r6/yqCwjI6O5uVnfpUAIoSdPnsTExBw8eJDJZMqmCwSC8PDw+vp6Q7gwM2vWrJycnHXr1jEYjGEXEIlEERERsm96GRgYKCgo8PHxIZvrsmXLcBzPz88nJuPi4kpLSzX4cv6Hbs4AACAASURBVBhqU77FKthZX3311Zw5czw9PbVQQEOh9649Lnr0pUuXiK9IE4gr1b29vQq67ag9elRjOkro+ACiSkQ/fvw4m83mcDjNzc27d++2t7evrKwcHByMjY11dHRksVgzZ84kXzp448aNt956i81mc7lcT09PoVD4/zOm0QoKCpYtW2Zubs7n87/55hsi/ZdffvHw8DA3N2cymZ6enleuXEEIpaWlMZlMGxubzZs38/l8JpMpEAju3LlD/MtIWSvv2LFjO3bsUKEqFCsuLnZ0dMQw7Msvv0QIpaenm5qastns/Pz8ZcuWcblcBweHc+fOIYUbGBYWZmJiQn58ZevWraamphiGtba2hoeH7969u7q6GsMwV1dXhNDPP//M5XITEhI0vi2jSktLw3Hcz89v6Kz4+Hg3N7evv/766tWrQ+fiOJ6cnPzGG28wGAwej7dy5cpHjx4hhdWFNLHThxUdHb1161Zra2sy5enTp93d3Y6OjmSKi4sLQqisrIyY5PF4Pj4+qamp+Pi/DKv7FquC/v7+27dvz549W0MbrYrXoWuPxx5dX1/PYrGcnJwUdNtRe/SoxnSU0PUBZNSz+GERl8137Nhx8uTJVatW/fHHH3v27GEwGD/++OOrV6/27dtHo9Hu3r3b3d3N5XITExNFIlFjY+OqVataWlrIf7927VpHR0d7e7uvry+Dwejp6cFx/MKFC3Fxce3t7W1tbQsWLCCvPISGhpqamj58+FAsFldUVMyfP5/D4RBXV4bNetRNINXV1Xl4eAwODo66JBr7VfcXL14ghE6ePClbb9euXevs7Gxubvb29jY1Ne3v71e8gevWrbO1tSXXmZSUhBAiajIgIMDFxYWcdenSJQ6Hc+jQoTEVEtfEVUdnZ2cPDw+5RBcXl2fPnuE4fuvWLRqNNmXKlO7ubvzfr9HFxsaamJh8//33HR0dZWVlc+bMsbKyamxsxBVWl5o7/e233x561b24uNjPzw/H8ZaWFvSv62k3btxACCUlJckuyWKxFi9eTE5GRUUhhO7fv698AYYykKvuOm6xypDbWc+ePUMIzZ49m/jMIIPBcHd3//LLL8mrmipQof2Pl66t8lX38dWjcRzv6enhcDhhYWG4wm6rTI9WYKxHCdUOIPq56n7s2LFt27bl5ORMmTIlPT3d398/ICDAwsJi//79dDo9MzOzpqZGKBROnz6dyWTa2trm5OTIjt8TCATm5uY8Hi8oKKivr4/oqIGBgQcOHODxeJaWln5+fm1tbUTFIYSMjY2J330eHh7p6eldXV2ZmZlisXjYrMe0Fdu3b6fRdHcDQiAQcLlca2vroKCgnp6e2tpaIn3YDRzTmpcvXy4UCmNiYrRQakV6enqePXtG/PYclpeX186dO2tqavbu3SubLhKJkpOTV61aFRwcbG5u7unpefr06dbW1jNnzpDLDK0u9Xf6UCKRKDw8PD09XS6dGJUqe2UPIUSn00UiETk5depUhFB5ebk6BTBk2muxKiBGwFlbWyckJFRUVDQ1Na1cuXLbtm1///vftZ31qCjTtcdjjz5y5Aifz4+Pj0cKu60yPXokKhwldHwA0UwYq6ys7O3tnTFjBjHJYrHs7OwePXrk7OxsY2MTHBwcFxen4NkSOp2OEJJIJMOmD/t00Lx589hs9qNHj0bKWsmSNzQ0XLx4cf369Uour1kmJiZouA1HMhuo80Kporm5GcdxNputYJn4+Php06adOnWquLiYTKyoqOju7p43bx6ZMn/+fBMTE/KWiiyyutTc6cPat2/fZ599Zm9vL5dO3EQcGBiQTezv72exWOQkseFNTU3qFGBcMIQWS9xWnz59ukAgsLS0NDc3P3jwoLm5uWzM0DtDqCh1jLsenZubm52dfeXKFWKAlIJuq0yPHokKRwkdH0A0E9F7enoQQvv37yefpXv+/Hlvby+Lxbp+/fqiRYsSEhKcnZ2DgoJG/R1UUFDw7rvvWltbMxiML774QsGSDAajpaVlpKyVLHliYuKmTZvkhn4YCGID9V0KpRCvFhhpuBmByWRmZmZiGLZhwwayGRCPbci9CcDCwqKrq0vBqtTc6UMVFxeXl5dv3Lhx6CziNic5/gMh1NvbKxaL+Xw+mUJ0Tj2+X8FA6KbFEjXf2tpKppiYmEyePLm6ulrbWWuK4Xft8dWjz58/f+zYsaKioilTphApCrqtMj16WKodJXR8ANFMRCfGCKSkpMhe0C8pKUEITZ8+/aeffmpoaIiMjMzKyjpx4oSC9dTW1vr7+9vZ2d25c6ezszMxMXGkJSUSSUdHh4ODg4KsR9XY2Pj3v//9888/H9vW6gS5gfouiFKIFjnqu1a8vLx27dpVVVV1+PBhIsXCwgIhJNfbR91wdXb6sDIyMq5du0aj0YijCbH+hIQEDMPa2to4HM7z58/JhZ88eYIQmjlzJpnS39+P/lUJry2dtVgzM7OpU6c+fPhQNnFgYMDc3FzbWWvEuOja46hHnzx58uzZs9evX5f98qSTk9NI3VbBLMUZqXaUUCY7DR5ANBPRJ02axGQyS0tL5dIbGhqIjmdtbX306NE5c+bI9UM55eXlEonk888/d3Z2Jt4rN9KSRUVFOI4vWLBgpKyVkZiYGBwcbGlpqcL/ahu5gQghY2PjYS/fGQ4bGxsMwzo7O0dd8vDhw+7u7vfv3ycmZ8yYYWZm9ttvv5EL3Llzp7+/f+7cuQpWos5OH1ZmZqbsoUR2zMuCBQt8fX1v3rwplUqJhQsLCzEMkx0DTGy4ra2tpsozHumyxa5Zs+b+/ftPnz4lJnt7e58/fz5eHmYbF117XPRoHMcjIyPLy8vz8vLkrgoYGxuP1G0VzFKcnWpHCWWy0+ABRDMRnclkhoSEnDt3Lj09XSgUDg4O1tXVvXz5sqGhYfPmzY8ePerv779///7z58+JdjwSYoj/1atXxWJxVVWV3K0XqVT66tWrgYGBsrKy8PBwR0fH9evXj5T1qGVuamr65ptvdu7cqea2a9CwG4gQcnV1bW9vz8vLk0gkLS0tsj/3LC0tGxoaampqurq6JBJJYWGhXp5eY7PZzs7OdXV1oy5JXKkjx4kwmczdu3fn5uaePXtWKBSWl5dv2bKFz+eHhoYqXslIOz0oKMjW1lazr6KMiYlpamo6cOBAT09PSUlJUlLS+vXrp02bRi5AbPh4iSgapH6LVS3fXbt2TZ48ef369bW1tW1tbZGRkSKRSG6IlkEZd117XPTohw8fHj9+/G9/+xudTpd9nzdxJVhBt1UwS+UDiGrZETR5ABl1NPxQiYmJxPWBSZMmff/990RiX19fZGSko6OjsbGxtbV1QEBARUVFTU2NQCDg8XhGRkYTJ06Mjo4eGBgg/33q1KnV1dVnz57l8XgIIQcHhwcPHkRGRlpaWlpYWKxevZp41tPFxaW2tjY0NJROp9vb2xsbG3O53JUrV1ZXVyvIWvEm4Di+a9eu4ODgUReThcb49NrJkyeJmyhsNtvPz+/UqVPECAhiw8+cOcPlchFCkydPfvz4sYINbGtre++995hMppOT0/bt2yMiIhBCrq6utbW19+7dmzx5MovFWrRoUWNj4+XLlzkcTnx8/Ji2C9fE02thYWF0Or23t5eYzM3NJQbKWllZbdu2TW7hiIgI8lkXqVSalJQ0depUOp3O4/H8/f0rKytxHFdcXSPtdH9/f4RQbGzssIUsKSlZuHAheQfLzs5OIBDcuHFDbjHZX98E4rUKDAaDz+dHRESIxWLZ5ZcvX25vb6/O01O4YTy9pvsWq6AwinfWixcvPvnkEx6Px2Aw3nrrrcLCQtU2mTDW9j+OurbKT68Zfo8eaXA4+aiYgm470izFBxBZYzpKjPUAovLTa/p5C6wKQkNDLS0t9VuGsUb0MdHvBqq/36uqqoyNjclfePoyODjo7e2dkZGhsxxbW1uZTOaJEyfUXI8hRPQxMYQuqSlaPe7pt6JUjuivZ482kAPIuHwL7FhR7CNXQ43rDXR1dT106NChQ4d0/MEMWYODg3l5eV1dXUFBQTrLNC4ubvbs2WFhYTrL0XCM6xarS+OiokQi0ZUrV6qqqoiBWq9hj9b7AQTH8YaGhuLiYmL0nArGU0RX3qNHj4Z+LJWky731WomKilq9enVQUJAyA2q0oaioKCcnp7CwUPFztBqUnJxcWlp6+fJl4t0JQHnQSQ1Ne3v7hx9+6ObmtmHDBiLldevRej+A5Ofn29vbe3t7y37abmxGPYs3BFFRUcS7CKZMmXLhwgV9FQNp7aq73jdQg/v9ypUrkZGRGlmVgcvLyzty5Ijsp5/UMb6uuuu9xWqW9o57eq8o9dvD69OjdUz9A8jQdovh//52eOLjshcuXFD5RweFYRiWlZVFfLCPYmC/6112dvaaNWtwtb/WoKn1vFYo3P6hPVDY0HZLzavuAAAAwOsGIjoAAABABRDRAQAAACqAiA4AAABQgfHQpLq6uuzsbN0XZVxQ53Mghox4DSHsdz3SbNOCXTkmFG7/RLui5KaBuro6+U/gDB0Nr6eyAQA09vQaAOB1IPf02jDn6IGBgZR8ikN98PQa0B7iKSNNrQ2eVhoTCrd/eHqNwoh2KwvuowMAAABUABEdAAAAoAKI6AAAAAAVQEQHAAAAqAAiOgAAAEAFENEBAAAAKtBMRM/JyXF2dpb9vDGTyXRyctqwYcOzZ89UWGFISAiTycQwTCwWa6SEQLOuXr0aFRUlu98//fRT2QWWLFnC4XCMjIymT59+79493ZdQKpWmpKQIBIKRFhCLxe7u7vv37ydTiouLFy5cyGaz+Xx+ZGRkX1+fOtldvHgxMTFxcHBQtfJr1e3bt9944w0ajYZhmK2tbXx8vLZzlG0qdnZ2wcHB2s4RjMnmzZvJo7fc3jHkzn7o0CEPDw8ul8tgMFxdXb/44ovu7m5yroIerU5nJ4zpADLsrKGHiLy8PHIvWFlZjbVICGn0++guLi7m5uY4jg8ODjY1NX333XdsNtvGxqa1tVWFtUVHRyOERCKRaoXRBqS176Pr3Zj2e2xs7IoVK4RCITHp4uIyYcIEhNClS5dkFyssLPzoo480XFDlPH78eOHChQihWbNmjbTMrl27EELR0dHE5IMHD1gsVkxMTHd3961bt6ysrEJCQtTMLjU11cfH59WrV8qsRPffR1+6dClCSMniaQR5iDA02vs+ut4p2R5CQ0MtLS0LCwsrKyvFYjGZbuCd3cfH59SpU21tbUKhMCsri06nf/jhh8QsBT1anc5OUv4AomCW3CFCKpXW1dXdvHnT19d3woQJo5ZhaLvVylV3Go1mY2Pz6aefbtu2rbm5+erVq9rIhUpEIpGCs0ldrmRUx44dO3/+fHZ2NofDIRPT0tJoNFpoaGhnZ6e2CzCq33//fe/evVu2bJk9e/ZIy9y6devBgweyKYcPH7azszt48KCpqamXl1dkZOS333776NEjdbLbsWPHrFmzfH19BwYGVNuWcU03DdLwjZfezWKxPvzwQzc3NwaDQaQYfmc3MzMjfotwOJyPP/7Y39//559/fvHiBVLYo1Xu7KQxHUAUzJI7RGAYZm9v7+3tPXXqVNUqRLv30V1dXRFCjY2NKq8BwzDNFcdwZWRkNDc3G8JKFHvy5ElMTMzBgweZTKZsukAgCA8Pr6+v37Nnj1YLoIxZs2bl5OSsW7eOPDDJEYlEERERqampZMrAwEBBQYGPjw/Z3pYtW4bjeH5+vprZxcXFlZaWyub1+tBBgxwXxkvvljMuOvulS5eMjIzISeJKdW9vr4IerU5nJ4zpADJqdpo9RGg3oldVVSGEZs2aRUwODg7GxsY6OjqyWKyZM2eSL6C+cePGW2+9xWazuVyup6enUCj8/4Wj0QoKCpYtW2Zubs7n87/55hsi/ZdffvHw8DA3N2cymZ6enleuXEEIpaWlMZlMGxubzZs38/l8JpMpEAju3LmjOGuNw3E8OTn5jTfeYDAYPB5v5cqVxG+xsLAwExMTOzs7YrGtW7eamppiGNba2hoeHr579+7q6moMw1xdXRVsiPIrQQj9/PPPXC43ISFBg1uXlpaG47ifn9/QWfHx8W5ubl9//fWwl2RGqpb09HRTU1M2m52fn79s2TIul+vg4HDu3Dniv7S016Kjo7du3WptbU2mPH36tLu729HRkUxxcXFBCJWVlamZF4/H8/HxSU1NxQ37NZwKdoSmGuSohu3XGzduJG4ruri43L9/HyEUEhLCZrPNzc0vXrw4bAs5fvw4m83mcDjNzc27d++2t7evrKzUSC1Ru3fLGY+dvb6+nsViOTk5KejR6nf2MR1ARs1Ow4eIUa/LK0/2JtmrV6++/fZbNpu9fPlycoE9e/YwGIwff/zx1atX+/bto9Fod+/e7e7u5nK5iYmJIpGosbFx1apVLS0t+L/uo1+7dq2jo6O9vd3X15fBYPT09OA4fuHChbi4uPb29ra2tgULFpD3G0JDQ01NTR8+fCgWiysqKubPn8/hcGpra0fKeqwbiJS4jx4bG2tiYvL99993dHSUlZXNmTPHysqqsbERx/F169bZ2tqSSyYlJSGEiI0NCAhwcXEhZynYEOVXcunSJQ6Hc+jQIWU2Tcn97uzs7OHhIZfo4uLy7NkzHMdv3bpFo9GmTJnS3d2N//utNQXVQu7ozs7O5uZmb29vU1PT/v5+XO299vbbbw+9j15cXOzn54fjeEtLC/rXbbAbN24ghJKSkmSXZLFYixcvVjM7HMejoqIQQvfv31f873q/j65gR2ikQeKj3UcfqV8HBAQYGRnV19eTS65du/bixYv4yC2E2JYdO3acPHly1apVf/zxh+KqULL9j8ferfx9dHt7e9mU8dXZcRzv6enhcDhhYWG4wh6tZmcf6wFEmeyGHiJ27NhhEPfROzs7iR/UPB4vJCRk3759//M//0PMEovF6enp/v7+AQEBFhYW+/fvp9PpmZmZNTU1QqFw+vTpTCbT1tY2JydHdoyfQCAwNzfn8XhBQUF9fX3EyPnAwMADBw7weDxLS0s/P7+2tjaichFCxsbGxG9DDw+P9PT0rq6uzMzMkbLW7LYjhEQiUXJy8qpVq4KDg83NzT09PU+fPt3a2nrmzJmxrmrYDRnTGpYvXy4UCmNiYsaa9Uh6enqePXtG/MAclpeX186dO2tqavbu3Subrky1CAQCLpdrbW0dFBTU09NTW1urjb0mEonCw8PT09Pl0omhp7KX7xBCdDpdJBKpkx2BuCVWXl6u/qp0YOiOINLVb5CjGqlfb9myZXBwkMxOKBTevXvX19d31BZy7Nixbdu25eTkuLu7q188avduOeOxsx85coTP5xMPbijo0ep0dhUOIMpkp8FDhIYjOvkDPCIiAsdxc3NzOp1OzKqsrOzt7Z0xYwYxyWKx7OzsHj165OzsbGNjExwcHBcXV1NTM9KaifVIJJJh04d9RmjevHlsNvvRo0cjZa3u1g5RUVHR3d09b948MmX+/PkmJibkxX/VkBuidgHV0tzcjOM4m81WsEx8fPy0adNOnTpVXFxMJo6pWkxMTBBCEolEG3tt3759n332mb29vVw6cadQbvxaf38/i8VSJzsCUWNNTU3qr0qXyB0xdJYOGqRsv37//ffd3Ny++eYbHMcRQufPnw8KCjIyMtJZvyZQu3fLGXedPTc3Nzs7+8qVK8QgPgU9Wp3OrsIBRJnsNHiI0NZ99JiYGDs7u3379hHDDhFCPT09CKH9+/eTz9s9f/68t7eXxWJdv3590aJFCQkJzs7OQUFBo/5WKigoePfdd62trRkMxhdffKFgSQaD0dLSMlLWGtrW/9PR0YEQMjMzk020sLDo6upSc83Ehqi5EjUR7wYYabgZgclkZmZmYhi2YcMGcj+qVi0a32vFxcXl5eUbN24cOou4eUkO4EAI9fb2isViPp+vcnYkoutS7M0K2miQI/VrDMM2b9789OnTa9euIYS+++67v/zlL0gLLUQxavduOeOrs58/f/7YsWNFRUVTpkwhUhT0aJU7u2oHEGWy0+AhQlsRncPhHDt2rKur6/PPPydSiHEEKSkpshf9S0pKEELTp0//6aefGhoaIiMjs7KyTpw4oWDNtbW1/v7+dnZ2d+7c6ezsTExMHGlJiUTS0dHh4OCgIGvNsrCwQAjJtV2iDOqsltwQtQqnNqLZjfrKFC8vr127dlVVVR0+fJhIUa1aNL7XMjIyrl27RrxWBcMwYv0JCQkYhrW1tXE4nOfPn5MLP3nyBCE0c+ZMlbMj9ff3o3/VHjVotkHevHkzJSVFcb9ev349k8n8+uuvKysruVzu5MmTkRZaiGLU7t1yxlFnP3ny5NmzZ69fvz5x4kQy0cnJaaQerWCW4oxUO4Aok50GDxFaHOv+pz/96e2337506VJ2djZCaNKkSUwms7S0VG6xhoaGhw8fIoSsra2PHj06Z84cYnIk5eXlEonk888/d3Z2Jt4rN9KSRUVFOI4vWLBgpKw1bsaMGWZmZr/99huZcufOnf7+/rlz5yKEjI2Nh72GOSpyQ9RZifpsbGwwDFPmIdTDhw+7u7sTI5PRaNUyEo3vtczMTNnjhezAlgULFvj6+t68eVMqlRILFxYWYhg27EDfsSJqzNbWVv1VGQjNNsh//vOfpqamivs1j8dbs2ZNXl7eiRMnNm3aRCTqrF8TqN275YyLzo7jeGRkZHl5eV5entxVAWNj45F6tIJZirNT7QCiTHYaPERoMaJjGJaWloZhWFhY2KtXr5hMZkhIyLlz59LT04VC4eDgYF1d3cuXLxsaGjZv3vzo0aP+/v779+8/f/6caNwjIR4DuHr1qlgsrqqqkrs9I5VKX716NTAwUFZWFh4e7ujoSPy6HzZrjW8yk8ncvXt3bm7u2bNnhUJheXn5li1b+Hx+aGgoQsjV1bW9vT0vL08ikbS0tMj+arO0tGxoaKipqenq6iK69LAbMqaVFBYWavb5Fjab7ezsXFdXp0w9ZGZmkoNBFFeLgpWMtNeCgoJsbW01+77JmJiYpqamAwcO9PT0lJSUJCUlrV+/ftq0aepnR9SYp6enBkure+o3yKHrlEgkTU1NRUVFpqamivs1QmjLli19fX2XLl1asWIFkaKzfk1mR+HeLWdcdPaHDx8eP378b3/7G51Ol30HOXGVV0GP1kZnVy07giYPEaOOhlfGr7/+6ubmRqxw4sSJmzdvJmcRLdXCwuLo0aN9fX2RkZGOjo7GxsbW1tYBAQEVFRU1NTUCgYDH4xkZGU2cODE6OnpgYCAxMZG4BDF16tTq6uqzZ8/yeDyEkIODw4MHDyIjIy0tLS0sLFavXv3ll18ihFxcXGpra0NDQ+l0ur29vbGxMZfLXblyZXV1NVGMYbMe62YiJZ5ek0qlSUlJU6dOpdPpPB7P39+/srKSmNXW1vbee+8Rb7zfvn17REQEQsjV1bW2tvbevXuTJ09msViLFi1qbGxUsCHKr+Ty5cscDic+Pl6ZTVNyv4eFhdHp9N7eXmIyNzeXGA1rZWW1bds2uYUjIiLIB1pGqpZTp04Ro0KIHX3mzBkul4sQmjx58uPHj0faa/7+/gih2NjYYQtZUlKycOFC8jaVnZ2dQCC4ceOG3GKyP7EJxHsRGAwGn8+PiIgg34WpZnbLly+3t7eXSqWK61aXT6/dvn17+vTpNBqNKHBCQoLiHaF+g/zqq68UDJzOzc3FcXykfk0W+80334yKipLdkGFbCHn0mDRp0vfff69MpSnZ/sdj71b56TXD7+wjDQ4nHxUbqUcrmKW4s8tS/gCieBY+3CFC5afXNPk8ut4RrwPU3vqViegaoe0NGUrJ/V5VVWVsbKzkUVJ7BgcHvb29MzIyDD+71tZWJpN54sSJUZfU/fPoytN9gxyWr6/v06dPtbFmXR73dFyZKkd06Oy6MewhwlCeR9c7w/zUlQoMc0NcXV0PHTp06NAh2Q8c6djg4GBeXl5XV1dQUJDhZxcXFzd79uywsDCNF0zH9NUgycv1ZWVlxMmrXoqhWYbZu0Ui0ZUrV6qqqoiBWtDZdUP2EIHjeENDQ3FxMTF6TgVUi+hA26KiolavXh0UFKSv7zQUFRXl5OQUFhYqfljWELJLTk4uLS29fPky+VYGMFaRkZFVVVWPHz8OCQkhB1QDbWhvbye+1LJhwwYiBTq7tskdIvLz84kvtRQUFKi4xlHP4seLqKgo4n0FU6ZMuXDhgjayQDq56q6DDRlqrPv9ypUrkZGR2isPBeTl5R05cmRgYEDJ5Q32qrteGiQpOjqaRqNNmjSJeO2rlujsuKf7ylS/PUBn15KxHiKGGtpuMfzf3w6/evVqhNCFCxdU/tFBYRiGZWVlffzxx/ouiObBfte77OzsNWvW4Gp/rUFT63mtULj9Q3ugsKHtFq66AwAAAFQAER0AAACgAojoAAAAABVARAcAAACowHho0u3bt4n77WColJQUSg6fuX37NvrXOAugF8q8cVN5sCvHhMLtn2hXlNw0cPv2bbmXpsuPdU9OTtbex4uAbgwODl6/fn327NnE14HAOKL+78WSkpLk5GSNFAaMC8XFxfb29tR49w4YK+Lzd+SkfEQH1GBmZpaWlka+KQIAQFV2dnbR0dHbt2/Xd0GA/sF9dGri8/na+wgVAMBwdHZ2El8lBwAiOjVBRAfgddDX1ycWiyGiAwJEdGqCiA7A6+DVq1cIIYjogAARnZr4fH5DQ4O+SwEA0K6Ojg6EkLm5ub4LAgwCRHRqgnN0AF4HxFfR4BwdECCiU9PEiRNfvnwJDzIAQG3EOTpEdECAiE5NfD6/v7+/vb1d3wUBAGhRR0eHkZERh8PRd0GAQYCITk18Ph8hBBfeAaC2jo4OLpeLYZi+CwIMAkR0apo4cSKCiA4A1XV0dMAld0CCiE5NPB6PxWLBcHcAqA1eLwNkQUSnLFtbWzhHB4DaIKIDWRDRn092+wAAIABJREFUKYsY7q7vUgAAtAiuugNZENEpCx5JB4DyOjo64PUygAQRnbIgogNAeXCODmRBRKcseBEsAJQH5+hAFkR0yoL76ABQHoyMA7IgolMWn88XiUTEa58BAJQEV92BLIjolEW8Ng4uvANAVRKJpKenByI6IEFEpyx4ESwA1AafUgVyIKJTlpWVlYmJCUR0AKgKPqUK5EBEpywMw2xtbeGqOwBUBZ9SBXIgolMZDHcHgMIgogM5ENGpDF4yAwCFdXR0YBjG5XL1XRBgKCCiUxlEdAAorLOzk8PhGBkZ6bsgwFBARKcyiOgAUBg8jA7kQESnMngRLAAUBi+MA3IgolPZxIkTu7u7u7u79V0QAIDmwTk6kAMRncrgtXEAUBh8pgXIgYhOZfDaOAAoDM7RgRyI6FRmY2NjbGwMER0ASoL76EAORHQqo9FoNjY2ENEBoCQ4RwdyIKJTHDzABgBVwX10IMdY3wUA2jVx4kRyZFxTU1NzczOdTnd3d9dvqQAAKmhqajIzMzM1NSUmIaIDORiO4/ouA9Ck+vr60tLSxsbG+vr6pqamoqKiV69eDQ4OtrW1DQ4OIoT+8z//MywsTN/FBACM2QcffHD16lUjIyMOh8Plcjs6OpycnNzc3CwsLCwsLMzNzTdu3Ghra6vvYgK9gYhONf/7v//79ttvGxkZGRsbS6VSiUQit8D9+/dnz56tl7IBANTx17/+NSIiYuhBm+jsEydOrKmpgZfCvs7gPjrVvPXWW97e3hiG9fX1DQ3nZmZmM2fO1EvBAABqWrp06bDnYAMDAzQabfv27RDOX3MQ0SkoOjp6YGBgaDqNRvP29qbRYKcDMC7NmDHDzs5u2Fk0Gu0vf/mLjssDDA0c3Clo6dKlnp6eQyO3kZHRu+++q48SAQA0Y8WKFSYmJnKJdDp97dq1EyZM0EuRgOGAiE5Ne/fuHXp1TiKR+Pj46KU8AACNWLp06dC7aRKJZOvWrXopDzAoMDKOmgYHB52dnV+8eCG7fxkMhlAoHPoDHwAwXgiFQktLS+K5FQKNRpszZ87du3f1WCpgIOAcnZqMjIwiIiJkL7xjGCYQCCCcAzCucbnc+fPnYxgmmxgeHq6v8gCDAhGdsjZs2MDlcslJOp3+3nvv6bE8AACNWLFihbHx/70czNzcPDAwUI/lAYYDIjplsdns8PBwsuf39/e/8847+i0SAEB9H374IXkrnU6nb9u2jcFg6LdIwEDAfXQqa29vt7e3F4vFCCE6nd7Z2clisfRdKACAWnAct7a2bmtrQwgZGRnV1NQ4ODjou1DAIMA5OpVZWlqGhobS6XSE0Ny5cyGcA0ABGIb5+vrS6XQ6ne7n5wfhHJAgolPcrl27pFIphmHvv/++vssCANCMZcuWDQwMSCSS7du367sswIBo4NtrdXV1t27dUn89QEsEAsEvv/yCEMrOztZ3WcAYfPzxxxpfJ/RWaujt7UUI8fn85uZm6NfjlEAg0Pz1FVxtWVlZGi4TAEATfRN6KwAGKysrS+MdXGPfR8dfsxF2GIZlZWVp4yxKGw4dOhQbG6vkwqtXr0YIXbhwQZslAopkZ2evWbNGe+uH3koBf/3rXzdt2kS8yx1667gj90YBTdFYRAeGbN++ffouAgBAk3bs2CH7VDoACEbGvSag5wNAMdCpwVAQ0QEAAAAqgIgOAAAAUAFEdAAAAIAKIKIDAAAAVKCjiH7ixAkbGxsMw06fPq2bHIeSSqUpKSkCgUA2MTEx0d3dncVimZqauru7x8TECIVCLRXg8uXL5ubmP/30k5bWry9Xr16NiorKyclxdnbGMAzDsE8//VR2gSVLlnA4HCMjo+nTp9+7d0/3JRx218sSi8Xu7u779+8nU4qLixcuXMhms/l8fmRkZF9fnzrZXbx4MTExUfab1obMYHurrKG7TLOgt+q+tx46dMjDw4PL5TIYDFdX1y+++KK7u5ucq6BLqtNbCWM6Agw7y1D6uPqPtBPvrBh1saqqKoTQV199pX6OKnj8+PHChQsRQrNmzZJNX758+YkTJ5qbm7u6urKzs+l0+gcffKDMCtHY3w9w6dIlLpd78eLFMf2X7gUGBgYGBiq5cGxs7IoVK4RCITHp4uIyYcIEhNClS5dkFyssLPzoo480XFDljLTrZe3atQshFB0dTUw+ePCAxWLFxMR0d3ffunXLysoqJCREzexSU1N9fHxevXqlzEqU7FMqGNe9VZbcLlMMeivBwHurj4/PqVOn2trahEJhVlYWnU7/8MMPiVkKuqQ6vZWk/BFAwawx9XEV2qRSq1V/FRo8RvT29np5ealfJDmlpaWrVq06e/bs7Nmz5Y4R/v7+IpGInCRertLQ0DDqOrW0P9ShqdpT/hhx9OhRNzc32Qp0cXH54YcfaDSavb19R0cHma6vY4SCXU/69ddflyxZItuf16xZ4+TkJJVKicmkpCQMw/744w81swsLC/Py8pJIJKOuZ1xEdN33VtLQXaYY9FZ8PPTW5cuXDwwMkJPEG4Fqa2txhV1S5d5KGtMRQHF2yvdxLbVJw7qPnpGR0dzcrPHVzpo1KycnZ926dUO/Ipybm8tkMslJe3t7hJDspZ5xREu1N5InT57ExMQcPHhQtgIRQgKBIDw8vL6+fs+ePTorzEgU7HqCSCSKiIhITU0lUwYGBgoKCnx8fMiXOi1btgzH8fz8fDWzi4uLKy0tlc1rXNN9byUM3WXjEfTWoS5dumRkZEROWllZIYR6e3sVdEl1eithTEeAUbPTex/XW0S/cePGW2+9xWazuVyup6enUCgMDw/fvXt3dXU1hmGurq6pqammpqY0Gm3u3Lm2trZ0Ot3U1HTOnDne3t6TJk1iMpkWFhZffPGFZktVVVVlYWExefJkza4WIVRcXOzo6Ihh2JdffokQSk9PNzU1ZbPZ+fn5y5Yt43K5Dg4O586dQwilpaUxmUwbG5vNmzfz+XwmkykQCO7cuYMQCgsLMzExsbOzI9a5detWU1NTDMNaW1vlag8h9PPPP3O53ISEBI1vCyEtLQ3HcT8/v6Gz4uPj3dzcvv7666tXrw6di+N4cnLyG2+8wWAweDzeypUrHz16pLhOEEKDg4OxsbGOjo4sFmvmzJmaej95dHT01q1bra2tyZSnT592d3c7OjqSKS4uLgihsrIyNfPi8Xg+Pj6pqan4eHsJq0H11qG7TOOgt5L02Fvr6+tZLJaTk5OCLql+bx3TEWDU7PTex/UT0Xt6evz8/AIDA9vb26uqqtzc3Pr7+1NTU1esWOHi4oLj+JMnT8LDwyMiInAc/+qrr549e9bY2PjOO+/cv38/Kirq/v377e3tf/7zn5OSkn7//Xf1yyORSOrr67/88surV6+ePHnSxMRE/XXKWbRokew3rz7//POdO3eKRCIOh5OVlVVdXe3s7Lxp0yaJRBIWFrZ+/fre3t4dO3bU1NTcu3dvYGDggw8+ePHiRVpamuy7qU+dOnXw4EHib7naQwgRYzSkUqnGt4VQUFAwbdo0Nps9dBaLxfr2229pNNqmTZt6enrk5sbFxUVFRUVHRzc3N9+8efPFixfe3t5NTU0K6gQhtHfv3uPHj6ekpLx8+XLFihVr16797bff1NyEX3/9tbq6eu3atbKJjY2NCCEOh0OmMJlMFovV1NSkZnYIoTfffLO+vl4jjVZnDKq3DrvLNA56K0lfvbW3t/f69eubNm0yMTFR0CXV7K1jPQIok51++7h+InpNTY1QKJw+fTqTybS1tc3JySEusAzLw8ODzWZPmDDhk08+QQg5OjpaWVmx2ezg4GCEEPGDUU2TJk1ycHCIi4s7fvy4Vr+QMZRAIOByudbW1kFBQT09PbW1tUS6sbEx8bvYw8MjPT29q6srMzNzTGtevny5UCiMiYnRQqlRT0/Ps2fPiN+nw/Ly8tq5c2dNTc3evXtl00UiUXJy8qpVq4KDg83NzT09PU+fPt3a2nrmzBlymaF1IhaL09PT/f39AwICLCws9u/fT6fTx1ohckQiUXh4eHp6ulw6MXJV9uofQohOp4tEInWyI0ydOhUhVF5erv6qdMZweutIu0xnoLfqrLceOXKEz+fHx8cjhV1Snd6qwhFAmez028f1E9GdnZ1tbGyCg4Pj4uJqamqU/C/i1HlgYICYpNPpCCHiJ6GaXrx40dzc/Pe///2///u/33zzTV3e3yIRWzfs5sybN4/NZmvkt4umNDc34zg+7E9+Unx8/LRp006dOlVcXEwmVlRUdHd3z5s3j0yZP3++iYkJcaFSDlknlZWVvb29M2bMINJZLJadnZ2aFbJv377PPvuMGDkhi7jRSDYzQn9/P4vFUic7AlFjGjnd1xnD6a0j7TLdg946dA0a7K25ubnZ2dlXrlwhzoYVdEl1eqsKRwBlstNvH9dPRGexWNevX1+0aFFCQoKzs3NQUJBGToBURqfTra2tlyxZcv78+YqKiiNHjuixMMNiMBgtLS36LsX/EYvFCKGRxi4RmExmZmYmhmEbNmwg929HRwdCyMzMTHZJCwuLrq4uBasiLgbu378f+5fnz5/39vaqXP7i4uLy8vKNGzcOnUXc+JR9LUFvb69YLObz+SpnRyJ6PlF744WB9FYFu8zQQG9FavTW8+fPHzt2rKioaMqUKUSKgi6pcm9V7QigTHb67eN6Gxk3ffr0n376qaGhITIyMisr68SJE/oqiSxXV1cjI6OKigp9F+TfSCSSjo4OBwcHfRfk/xCtdtTXKXh5ee3atauqqurw4cNEioWFBUJI7ogw6tYR41ZSUlJkn9MoKSlRufwZGRnXrl2j0WjEEYdYf0JCAoZhbW1tHA7n+fPn5MLEnc6ZM2eqnB2pv78f/av2xhFD6K0Kdpn6Iyo0CHqrOr315MmTZ8+evX79+sSJE8lEJyenkbqkglmKM1LtCKBMdvrt4/qJ6A0NDQ8fPkQIWVtbHz16dM6cOcSkjrW1tcmNiaiqqhocHJw0aZLuC6NAUVERjuMLFixACBkbG2vkRoOaiHeKdXZ2jrrk4cOH3d3d79+/T0zOmDHDzMxM9hB8586d/v7+uXPnKlgJMV66tLRUzWKTMjMzZQ83xBkV8TTqggULfH19b968SQ5TKiwsxDBs2HHCY0XUmK2trfqr0hkD6a0KdpnsZWG9g96qWm/F/x97dx7QxJU/APwFEhLCjSAgyhk8QVFRAaFeVWtZQUQEb227xquAIKJiKSIqLK4iClasddejKCjiUVCKivdFJXJViigCInKf4Qhhfn/Mbn4sRwwhySTh+/mLvHl585038+bLTObAsICAgOzs7KSkpG5nBchkcl9Dks8k/rMTbg8gyOyIHeOEZfQNGza8fv26vb09MzPz/fv3+ADQ1tYuKysrKipqbGyUwEhQUVFJTU29c+dOQ0MDh8PJzMxcs2aNiooK/vwgYnV2dtbW1nZ0dGRlZfn4+BgZGa1duxYhxGAwampqkpKSOBxOZWVl1/8Wu/VeSkqK+O6HodPpZmZmpaWln62Jn83jXUtCo9H8/PwSExPPnTvX0NCQnZ29ceNGAwMDJpPJv5F169bFxcXFxMQ0NDRwudzS0tKPHz8ihDw9PfX09ET7uMoffvjh06dPP/74Y3Nz85MnTyIiItauXTtq1KiBzw7vMSsrKxFGK25SMlqlGYzWbo0IMVrz8vL+8Y9/nDx5kkKhkLrATwjxGZLiGK3CzQ5H8BgX5rE0/0uQp1D985//xP9nUVFRWbx4cVFRkb29vZaWlqKi4rBhwwIDA/FHBb18+dLY2FhZWdnBwWHnzp34JQYmJiYPHjwICwvT0NBACOnp6Z0/f/7ChQt4g1paWnFxcfzn/uTJk+nTp/N+6tDX17e3t7937x6GYc7OzqampqqqqlQq1dzc3NPTMzs7W5ClRv184s/Ro0fx32DodLqzs3N0dDS+dBYWFoWFhbGxserq6gghY2Pjv/76i8lkUigUQ0NDMpmsrq6+aNGiwsJCvJ3q6upZs2bRaDRTU9Pvv//e398fIcRgMIqLi7v2Xnl5eXJyspqaWmhoqOBB4gR8CpWXlxeFQmGz2fjHxMRE/GJaHR2dLVu2dKvs7+/PewpVZ2dnRESEhYUFhULR0tJydXXNz8/HMIx/n7S1tQUEBBgZGZHJZF1dXTc3t9zcXAzDXF1dEUJBQUG9Bsln1XfV9T90HH4HNpVKNTAw8Pf3b21txcsHODsnJydDQ0PeA6f6Quwz46R2tHbVc5XxAaNV+kdrXxeHR0RE4BX6GpJ8JvEfrV0JvgfgPwkTeIz3d5sUkOSeAitnxLQ+cEwmU1tbW0yNf5aA+4iCggIymXz27FkJhMQHl8t1dHQ8deqU9M+uqqqKRqMdPHjwszUJfwqsnIHRCqNVMgQf42LaJqXrKbCAh/h3+HwOg8EICQkJCQkh8KG5XC43KSmpsbHR09NT+mcXHBxsbW3t5eUl8sAAsWC0CkK2RqtwCB/j8pDRX79+TeqbJFfnYLNz5053d3dPT09BLroRh/T09MuXL6ekpPC/11YaZnfo0CEWi5WcnIzfmT1owWglCoxWcZOGMU4masYiNHr0aEzWHpTNx65du06fPt3e3m5qahoREbFkyRKiI+Jn3759qampBw4cCAsLk/zc58yZM2fOHOmf3dWrV9va2tLT07s9bWoQgtFKIBit4iMlY1weMrqc2b9/vxQ+4oaPefPm4e8iBH1xcXFxcXEhOgogejBaAU5Kxrg8nHUHAAAAAGR0AAAAQB5ARgcAAADkAWR0AAAAQB6I7Mo4d3d3UTUlKw4fPpyQkEB0FKL39OlTNChXqPQQ5IGdAzEIVy6MVjAYwDE6AAAAIA9Edowul///8kEikbZu3bp06VKiAxE9/P/9wbZCpUp8fLyHh4f42h9sKxdGK5A2JBJJHM3CMToAAAAgDyCjAwAAAPIAMjoAAAAgDyCjAwAAAPIAMjoAAAAgD4jJ6JcvXzYzM8Pfn6ivr79y5cpeq7169crT09PU1JRKpero6EyYMCE0NBQh5OnpyeeFjCQSad26dbz2f/jhh14bP3ToEIlEUlBQGD169P3798W4tOC/0tLSdu7c2XXtr1q1qmuFefPmqampKSoqjhs37uXLlxIOj8Ph7N+/n8FgKCkpaWpqWlpaFhUVdavT2to6evTo3bt3I4SuXbsWHh4u/e/GHiAYraArGMVSDRuwixcvCteOubm5hoZGX1OzsrLodLq3t/e7d+9aWlry8/O3b98+Z84cDMM8PDxSU1Pr6uo4HM7Hjx8RQs7Ozu3t7c3NzRUVFX//+9+vX7+Ot48Q0tfXb29v79Z4R0eHsbExQghvUAgIoYsXLwr3XSm3ZMmSJUuWiLzZoKCghQsXNjQ04B/Nzc2HDBmCELpx40bXaikpKS4uLiKfuyBcXV1HjRr19OlTDodTVlbm7OycnZ3drY6vry9CKDAwEP8YGRk5Y8aM2tpa0UYi9JgSX8swWqWTmEZrX2AUi4qYtknpPet+8OBBTU3NyMhIExMTGo02cuTIvXv3KisrI4RIJNL06dM1NDTI5P/cT08ikSgUCp1O19XVnTx5Mq+RyZMnl5eXJyUldWv88uXLhoaGEluWfmlpabG3t5eGRkQoLCzswoUL8fHxampqvMKoqCgFBQUmk1lfX09gbLgLFy4kJSUlJCRMmzaNTCYbGBhcvXrV0tKya53Hjx/n5OR0LfH29p4wYcLXX3/d0dEh2XilC4xWwhuRABjF0k96M3p1dXV9fX1NTQ2vRElJ6fr16wihuLg4Op3e1xeZTObf/vY3/O9NmzYhhI4fP96tzqFDh/z8/EQftCicOnWqoqJCGhoRlTdv3vzwww979uyh0Whdy+3t7X18fD58+LBt2zaiYuM5fvz4pEmTrKys+qrQ0tLi7+8fGRnZrTw4OJjFYvUsH1RgtBLeiLjBKJYJ0pvRp0yZ0tzcPHv27EePHgndyOzZs8eMGXP37t38/Hxe4aNHj9hs9rx580QRJj8Yhh06dGjMmDFUKlVLS2vRokWvX79GCHl5eSkpKenr6+PVNm/erKKiQiKRqqqqfHx8/Pz8CgsLSSQSg8GIioqi0WhDhw7dsGGDgYEBjUazt7d/9uxZvxpBCN28eVNdXX3fvn3iXuReRUVFYRjm7Ozcc1JoaOjIkSN//vnntLS0nlP76sCYmBgVFRU6nX716tUFCxaoq6sPHz48Li4O/xaXyw0KCjIyMlJWVh4/fjx+npm/9vb2p0+fWltb86kTGBi4efNmXV3dbuVaWlozZsyIjIzEz6QNTjBa5Wa09gVGsWwY+Il7Mf0yx2azbWxs8CDHjh0bHh5eXV3dsxr+y1yvv9mYm5u/e/fuyJEjCCEfHx9euaur6+nTpxsbG5GYf5kLCgpSUlI6e/ZsXV1dVlbWpEmTdHR0ysvLMQxbsWKFnp4er2ZERARCqLKyEsMwNzc3c3Nz3iQmk6miopKXl9fa2pqbmztlyhQ1NbXi4uJ+NXLjxg01NbWQkBBBFk3kv8yZmZmNHTu2WyG+djAMe/z4sYKCgomJSVNTE/a/v8Dx6cDAwECE0O3bt+vr6ysqKhwdHVVUVPCfYLdt20alUi9dulRbW7tr1y4FBYUXL17wj/Ddu3cIIWtr65kzZ+rr61Op1NGjRx87dqyzsxOv8PDhQ2dnZwzDKisrUZdf4HA7d+5ECGVmZg64q/5D5n5Hh9GKk4PR2hcYxaIlyDYpBOk9RldWVn78+PGRI0dGjx6dl5cXEBAwZsyYe/fu9bedNWvWqKio/Pvf/25paUEIvX379sWLF8uXLxdDyP+jpaXl0KFDixcvXrlypYaGhpWV1U8//VRVVRUbG9vfpshkMv4f7tixY2NiYhobG0+fPt2vFpycnBoaGvq6kFismpub3717h1/31Cs7O7utW7cWFRXt2LGja7kgHWhvb6+urq6rq+vp6dnc3FxcXNza2hoTE+Pq6urm5qapqbl7924KhfLZ7mpqakII6erq7tu3Lzc399OnT4sWLdqyZcuvv/6KR+Lj4xMTE9PX1y0sLBBC2dnZgnWJHILRyiPTo7UvMIplhfRmdIQQhULx8vL6888/nz59umjRooqKCnd399ra2n41oqGhsXz58tra2gsXLiCEDh8+vGnTJiUlJfGE/P9yc3Obmpp4By4IoSlTpigpKeFn4YRmY2NDp9Px01YyoaKiAsMwPr+kIoRCQ0NHjRoVHR398OFDXmG/OhBfoRwOJz8/n81m866FUVZW1tfX/2x3UalUhNC4cePs7e21tbU1NDT27NmjoaGB73d27dq1fv16Phdn4Uv36dMn/nORbzBae5K50doXGMWyQqozOs+0adOuXLmycePGysrKu3fv9vfr+BU3P/30U11dXUJCwoYNG8QQY3d1dXUIIVVV1a6Fmpqa+PnDgaBSqfhZI5nQ2tqK/jvY+kKj0U6fPk0ikb755hv84AwJ24HNzc0Iod27d/Nud37//j2bzeYfpIGBAUKoqqqKV6KkpGRsbFxYWPjw4cPs7OzvvvuOz9fxi7rxJQUwWruSrdHaFxjFskLqMvr9+/cPHz6MEHJzc+t2LwH+HIPPrteerK2tbW1tnz9/zmQy3d3dtbS0RBUtH5qamgihbhtuXV3d8OHDB9Ish8MZeCOShI+Tzz7Awc7OztfXt6CgYO/evXiJcB2IX/Ny+PDhrr8tPXnyhP/cVVVVLSws8vLyuhZ2dHRoaGicOnXq9u3bCgoK+J4Fb3/fvn0kEikjIwOv2d7ezlvSQQVGK38yN1r7AqNYVkhdRv/jjz9UVFQQQm1tbd3WDX4F7Pjx44VoFv/H/9KlS1u3bhVFmJ9naWmpqqrK21wQQs+ePWtvb8dvwCWTyRwOR4hm09PTMQyztbUdSCOSNHToUBKJJMi9qnv37h09enRmZib+kX8H9mXEiBE0Go3FYvU3Tg8Pj8zMzLdv3+If2Wz2+/fvraysTp8+3XW30vWaGt65RHzp9PT0+jtTWQejlT+ZG619gVEsK6Qoo3M4nE+fPqWnp+P7CISQq6trfHx8XV1dfX391atXd+zY4eLiItw+YunSpTo6Oq6urmZmZiKNuk80Gs3Pzy8xMfHcuXMNDQ3Z2dkbN240MDBgMpkIIQaDUVNTk5SUxOFwKisr379/z/uitrZ2WVlZUVFRY2Mjvgvo7Oysra3t6OjIysry8fExMjJau3ZtvxpJSUkh6n4YOp1uZmZWWlr62Zr4WTtFRUXeRz4dyKeRdevWxcXFxcTENDQ0cLnc0tJS/BJrT09PPT29vh5L6evra2xsvHbt2uLi4urq6oCAgJaWlm6X+fQFXzo+d8HKHxitODkbrX2BUSwzBn65vBD3wyQmJvK5bDIxMRHDsNTUVA8PD3NzcyqVqqSkNGrUqODg4NbWVl4jDQ0NX3zxhba2NkJIQUGBwWDs27evW/s6OjpbtmzBC7dv3/748WP87927d+P3hiooKIwdO/bBgwf9XWokwL0HnZ2dERERFhYWFApFS0vL1dU1Pz8fn1RdXT1r1iwajWZqavr999/7+/sjhBgMRnFx8cuXL42NjZWVlR0cHMrLy5lMJoVCMTQ0JJPJ6urqixYtKiws7G8jycnJampqoaGhgiyayO+H8fLyolAobDYb/9jr2uHx9/fn3ffSVwdGR0fjl7FYWFgUFhbGxsaqq6sjhIyNjf/666+2traAgAAjIyMymayrq+vm5pabm4thmKurK0IoKCiorzhLSkqWLVumpaVFpVKnTp2akpLSs06v9704OTkZGhrybpIZOKm6ew1G66AarX2BUSxagmyTwjQ78CbEt/eRZmJaHz0xmUxtbW0JzIhH5PuIgoICMpl89uxZEbYpBC6X6+joeOrUKdE2W1VVRaPRDh48KMI2pSqjywEYrQMHo1i0xLRNStFZd9AXWX8vEIPBCAkJCQkJwW8YJQSXy01KSmpsbPT09BTFJfBfAAAgAElEQVRty8HBwdbW1l5eXqJtFsgoWR+tfYFRLBMgowNJ2Llzp7u7u6enJ1Gvc0hPT798+XJKSgr/e2r769ChQywWKzk5mUKhiLBZAKQQjGLpBxldqu3atev06dP19fWmpqaXLl0iOpwB2bdvn5eX14EDBwiZ+5w5c86fP897sLZIXL16ta2tLT09XTJ3WAEpJ0+jtS8wiqUcmegAAD/79+/fv38/0VGIzLx58yTwyg2JcXFxcXFxIToKIC3kbLT2BUaxNINjdAAAAEAeQEYHAAAA5AFkdAAAAEAeQEYHAAAA5AFkdAAAAEAuDPwhNfhTqAAAojXwsQmjFQCpJY5nxpEwDBtgWKWlpY8fPxbJEgK5l56eHhsbO27cuA0bNgwZMoTocKTa0qVLRd4mjFbhvH379vjx458+fdq2bZtwr58BoBt7e3uRv2lXBBkdgH55/vz5mjVrysrKIiIi1q9fT3Q4APDT2toaFha2f//+qVOn/vLLLyNHjiQ6IgD6BL+jA0mbOnVqZmbmxo0bN27c+PXXX5eVlREdEQC9e/r06aRJkyIiIvbu3Xv//n1I50DKQUYHBKDRaGFhYQ8ePCgoKLC0tDx37hzREQHwP/D3ajs4OOjq6r569SogIEBBAfaWQNrBNgoIY29v/+rVq1WrVq1evXrp0qXV1dVERwQAQgg9evRo4sSJP/30U0xMTHp6OoPBIDoiAAQCGR0QiU6nHzly5ObNm0+ePBk3bty1a9eIjggMamw2e8eOHV988YWZmVl2dvb69etJJBLRQQEgKMjogHjz5s3LycnBX5mwevXqxsZGoiMCg9GDBw+sra1PnDhx/Pjx5OTkESNGEB0RAP0DGR1IBQ0NjRMnTty4cSMtLc3KyurOnTtERwQGkYaGBm9v75kzZ44cOTInJwduwQAyCjI6kCJOTk4sFmvSpElffvklk8lsbm4mOiIg/27dumVlZXX27Nnjx4/fuHHD0NCQ6IgAEBJkdCBdhg4dmpiYePHixUuXLo0fP/7BgwdERwTkVn19PZPJXLBgwbRp0/Lz8+HQHMg6yOhAGrm7u+fm5o4dO3bWrFk7duxoa2sjOiIgb5KTky0tLa9evZqQkBAfH6+rq0t0RAAMFGR0IKX09fWvXbsWExMTHR1tY2Pz8uVLoiMCcqK2tpbJZDo5OdnZ2eXm5rq5uREdEQCiARkdSC8SibR+/frs7GwdHR1bW9sdO3ZwOByigwKy7fr165aWltevX79y5Up8fDy8XADIE8joQNqZmJjcuXPn2LFjR48edXBweP36NdERAZlUUVHh7u7u7Ow8ffr0nJycRYsWER0RACIGGR3IAPxgPSMjA8OwiRMnhoeHd3Z2Eh0UkCUJCQnjxo178eJFampqfHy8trY20REBIHqQ0YHMGDNmzOPHj4ODg4OCghwdHd+8eUN0REAGlJeXL1682MPDY/HixdnZ2XPnziU6IgDEBTI6kCVkMjkgICAjI4PNZk+YMOHIkSPwOmDAB35ozmKx0tLSTpw4oaamRnREAIgRZHQge6ysrJ4/f+7v7+/n57dgwYLS0lKiIwJSp6yszMXFxcPDY8mSJVlZWbNnzyY6IgDEDjI6kEkUCiU4OPjhw4dFRUWWlpaxsbFERwSkBYZhsbGxY8aMyc3NvXv37okTJ1RVVYkOCgBJgIwOZJitrW1mZuaGDRs2bty4ZMmSyspKoiMCBCsqKpo7d+7mzZvXrl376tWrGTNmEB0RAJIDGR3INmVl5bCwsHv37rFYLEtLyytXrhAdESAGfmhuZWVVXl7+6NGjI0eOqKioEB0UABIFGR3IAwcHh5cvXy5atGjx4sVLly6tqakhOiIgUW/fvp0zZ87mzZs3b978xx9/TJ06leiIACAAZHQgJ9TV1U+cOJGSkvL48eNx48bduHGD6IiAJHR2dsbGxo4fP76qqurp06dhYWFUKpXooAAgBmR0IFe++uqrnJycuXPnOjs7M5nMpqYmoiMCYpSXl2dvb79ly5YtW7ZkZGRMnjyZ6IgAIBJkdCBvNDU1z5w5c/HixcTERCsrq/T0dKIjAqLX0dERHh4+adKktra2Z8+ehYWFKSkpER0UAASDjA7kk7u7e05OzoQJE2bPns1kMtlsNtERAZHJzs62s7MLDg7es2dPRkbGxIkTiY4IAKkAGR3ILT09vaSkpIsXLyYkJEyZMiUjI4PoiMBAcTic8PBwGxsbCoWSmZkZEBCgqKhIdFAASAvI6EDOubu7s1gsAwMDOzu7HTt2tLe3Ex0RENKrV69sbW337NkTEhLy4MGD0aNHEx0RANIFMjqQf0ZGRr///nt0dHR0dLSNjQ2LxSI6ItA/+KH5lClTlJWVWSwWHJoD0CvI6GBQwN/H+urVKy0tralTpwYHB3O5XKKDAgJ5+vTphAkTQkJC9u7de//+/ZEjRxIdEQBSCjI6GETMzMzu3r0bERERFhbm4OCQn59PdESAn5aWlh07djg4OOjq6r569SogIEBBAXZZAPQJhgcYXBQUFLy9vf/44w8OhzNx4sTw8PDOzk6igwK9ePz48cSJE3/66aeYmJj09HQGg0F0RABIO8joYDAaN27c06dPf/zxx6CgoPnz5xcXFxMdEfh/bDZ7x44djo6OZmZm2dnZ69evJ5FIRAcFgAyAjA4GKTKZHBAQ8OLFi8rKSisrK3gfq5R48OCBtbX1iRMnjh8/npycPGLECKIjAkBmQEYHg9r48eOfPn26cePGTZs2LViw4MOHD0RHNHg1NDR4e3vPnDlz5MiROTk569evJzoiAGQMZHQw2NFotLCwsAcPHhQWFlpaWp49e5boiAajW7duWVlZnT179vjx4zdu3DA0NCQ6IgBkD2R0ABBCyM7OjsVirV69es2aNUuXLq2qqupZh8Ph3Lx5U/KxyYc//vij1/L6+nomk7lgwYJp06bl5+fDoTkAQoOMDsB/0On0I0eO3Lp16+nTp5aWllevXu1WITQ01NXVNTc3l5DwZFp0dPSsWbN6/qiRnJyMd3VCQkJ8fLyuri4h4QEgHyCjA/A/5s6dm52d7eLismjRoqVLl9bW1uLlGRkZoaGh7e3tixcvhve+9MuLFy+2bt3a1NT07bff8gpra2uZTKaTk5OdnV1OTo6bmxuBEQIgHyCjA9CdhobGiRMnfvvtt4cPH06cOPH27dttbW0rV65UUFDo7Ox8+/atl5cX0THKjNra2sWLF2MYhmFYamoqfpnC9evXLS0tr1+/fuXKlfj4eB0dHaLDBEAekDAMIzoGAKRURUXFxo0br1y5Mnv27Hv37nV0dPAmnTt3bsWKFQTGJhMwDHN2dr516xaHw0EIkUgkOp0+Y8aMlJSUb7755p///KeGhgbRMQIgPyCjA/AZQUFBoaGhXUcKiUSi0WgsFgueMc7fgQMHdu/e3fWpfBQKRVlZ+dKlS3PnziUwMADkEpx1B4Cf5ubmM2fOdHvTF4ZhHA7H3d0d3s3Kx71797qlc4QQh8NpaGioqakhKioA5BhkdAD48fX1LSsr63q+HdfR0ZGXlxcQEEBIVNLv06dP7u7uvT69lUQiMZnMiooKyUcFgHyDjA5An1JTU0+ePIn/BtxTR0fHkSNHrl+/LuGopB+Xy126dGldXV2vr6zFMKy5uXnLli2SDwwA+QYZHYA+/fvf/8bPtyspKfVagUQirVq1qqSkRLJxSbugoKBHjx719Z8QhULhcrkJCQl37tyRcGAAyDe4Mg4Afths9uPHj9PS0n777Tf82TKKiopdT8JTKJTJkyc/ePCATCYTF6YUSU5O/tvf/tZtx6KkpNTR0dHZ2WloaDh79mwHB4f58+cbGxsTFSQAcgkyOgCCKi8vT0tL+/3332/evFlRUUEmkzEMw08s7969e+/evUQHSLz3799PmDChoaEBwzAKhdLR0YFhmJGR0bx582bOnDlr1qxhw4YRHSMAcgsyukQ9efLk0KFDREcBRKChoaGioqK8vLyyshJP6l988cXQoUOJjotInZ2dd+/exZ+yR6fT9fT0dHV1dXV1lZWViQ4NDIivr6+dnR3RUYDPg/OEElVSUnLp0qUlS5YQHcigc+nSJVtb2+HDh4uqQXV1dXV1dQaDgWFYTU3Np0+f3r59q6mp2dcv7uLz9OlThJCtra2E59tTYWGhpqYmg8EYOnQoZHG5cenSJXd3d8joMgEyOgESEhKIDmHQIZFIW7duXbp0KdGBiJ67uzuCjQqITa+3IALpBNe6AwAAAPIAMjoAAAAgDyCjAwAAAPIAMjoAAAAgDyCjAwAAAPIAMjqQJTNnziT1oKqqKo55JScna2hoyM1j2zds2MDrsZUrV3adlJaWtnPnzsuXL5uZmeEVVq1a1bXCvHnz1NTUFBUVx40b9/LlS8kGjjgczv79+xkMhpKSkqampqWlZVFRUbc6ra2to0eP3r17N0Lo2rVr4eHhvT5S/rPkviuSkpJ4m4GOjo4kgwcSABkdyDwHBwdxNCt/D1/S1tZOSUnJz88/deoUr/DHH3+MioratWuXm5vb27dvzc3NhwwZcu7cud9++41XJzU1NSEhYeHChbm5uZMmTZJw2B4eHmfOnDl//jybzf7zzz/Nzc2bmpq61QkMDMzPz8f/dnZ2ptFoc+bMqaur69eMBkNXuLi4lJaW3r9//+uvv5Zo6EAiIKOD7lpaWuzt7aWzZRqNhj9hlIfJZG7fvl1UEXbl5ORUX1+/cOFCcTSOxNnPfVFWVv7qq69GjhxJpVLxkrCwsAsXLsTHx6upqfGqRUVFKSgoMJnM+vp6SYbXqwsXLiQlJSUkJEybNo1MJhsYGFy9etXS0rJrncePH+fk5HQt8fb2njBhwtdff93zNbh9GSRdQSKRDA0NHR0dLSwsJBo9kAjI6KC7U6dOiend1QNv+ebNm113uCUlJTk5ObNnzx5waAQQXz8L6M2bNz/88MOePXtoNFrXcnt7ex8fnw8fPmzbto2o2HiOHz8+adIkKyurviq0tLT4+/tHRkZ2Kw8ODmaxWD3LewVdAeQDZHQpdfbsWRsbGxqNpqKiYmJigr8FBMOwQ4cOjRkzhkqlamlpLVq06PXr1wihmJgYFRUVOp1+9erVBQsWqKurDx8+PC4ujn9rDx48GDt2rIaGBo1Gs7KyunXrFkLIx8fHz8+vsLCQRCIxGAyEEJfLDQoKMjIyUlZWHj9+/MWLFz87x4G03C9hYWHe3t4D7uxePHz40MjIiEQiHTt2DPFd3qioKBqNNnTo0A0bNhgYGNBoNHt7+2fPniGEvLy8lJSU9PX18TY3b96soqJCIpGqqqp69sbNmzfV1dX37dsnjsXpVVRUFIZhzs7OPSeFhoaOHDny559/TktL6zlVuO1QiNXd3t7+9OlTa2trPnUCAwM3b96sq6vbrVxLS2vGjBmRkZGC/HoCXQHkBAYkCB+6n612+PBhhNCBAweqq6trampOnDixYsUKDMOCgoKUlJTOnj1bV1eXlZU1adIkHR2d8vJyDMMCAwMRQrdv366vr6+oqHB0dFRRUWlvb+fTWkJCQnBwcE1NTXV1ta2t7ZAhQ/C5u7m5mZub84LZtm0blUq9dOlSbW3trl27FBQUXrx4wX+OA2xZQKWlpWPHjuVyuYJURghdvHhR8MYxDMPfen706FH8I5/lZTKZKioqeXl5ra2tubm5U6ZMUVNTKy4uxjBsxYoVenp6vDYjIiIQQpWVlViP3rhx44aamlpISEi/gsQwbMmSJUuWLPlsNSaTaWho2LXEzMxs7Nix3aqZm5u/e/cOw7DHjx8rKCiYmJg0NTVhGJaSkuLi4oLXEW47FGJ1v3v3DiFkbW09c+ZMfX19KpU6evToY8eOdXZ24hUePnzo7OyMYVhlZSVCKDAwsOvXd+7ciRDKzMz8bOcMwq7w9vbmDUz+hBg7gCiQ0SVKkIze3t6uqak5a9YsXklHR0dkZCSbzVZVVfX09OSVP3/+HCGE5wB899HS0oJPio6ORgi9efOmr9a6zXT//v0IoYqKCux/M01LSwudTufNlM1mU6nUTZs28ZnjwFsW0JYtW44fPy5gZVFl9F6Xl8lkamho8L744sULhNCePXuw/mR0oQmX0Zuamkgk0sKFC7tV46UxDMP8/PwQQlu2bMG6pDHhtkPhVnd2djZCaO7cuY8ePaqurq6rq9uxYwdC6Ny5c3gjNjY2paWlWB9p7JdffkEInTlzhv9cBmdXQEaXS3DWXepkZWXV1dXNnz+fV6KoqOjt7Z2bm9vU1GRjY8MrnzJlipKSEn6Ctxv8DWAcDqev1rrVp1AoCKGeN/zk5+ez2Wze1TfKysr6+vr4ecW+5ijylntVVlZ27dq1tWvXClhf5PpaXoSQjY0NnU4XfFkIgf+PRafT+dQJDQ0dNWpUdHT0w4cPeYXCbYfCrW78Cr5x48bZ29tra2traGjs2bNHQ0MjNjYWIbRr167169cbGhr29XV86T59+sR/LtAVQG5ARpc6DQ0NCCFNTc1u5fj9J93uvdbU1GxsbBSiNYTQb7/9NnPmTF1dXSqV2tfl4s3NzQih3bt38+5hff/+PZvN5r8I4muZJzw8/O9//3u365ikB5VKxQ+VpFZrayv6b57oC41GO336NIlE+uabb1paWvBC4bZD4Va3gYEBQqiqqopXoqSkZGxsXFhY+PDhw+zs7O+++47P1/E3uuJLygd0BZAbkNGlzrBhw9D/Dl0cnpW77Szq6ur4v/O7r9aKi4tdXV319fWfPXtWX18fHh7e69fxq2wOHz7c9cTOkydP+MxRfC3zlJeX//rrr5s2bRKksuRxOJzPrhfC4bv4zz6Gxc7OztfXt6CgAL+aEgm7HQq3ulVVVS0sLPLy8roWdnR0aGhonDp16vbt2woKCnhSxNvft28fiUTKyMjAa7a3t/OWlA/oCiA3IKNLHRMTE21t7dTU1G7llpaWqqqqvCGKEHr27Fl7e/vkyZOFaC07O5vD4WzatMnMzIxGo/X1CuQRI0bQaDQWiyV4/OJrmSc8PHzlypXa2tpCfFcC0tPTMQyztbVFCJHJ5F7PzBNu6NChJBJJkNus9+7dO3r06MzMTPyjcNuh0Kvbw8MjMzPz7du3+Ec2m/3+/XsrK6vTp093zYhdfzzmnQbHl05PT4//LKArgNyAjC51qFTqrl277t+/7+Xl9eHDh87OzsbGxry8PBqN5ufnl5iYeO7cuYaGhuzs7I0bNxoYGDCZTCFaMzIyQgilpaW1trYWFBR0/eVPW1u7rKysqKiosbFRUVFx3bp1cXFxMTExDQ0NXC63tLT048ePfOYovpZxnz59+uWXX7Zu3frZmpLU2dlZW1vb0dGRlZXl4+NjZGSE/8bPYDBqamqSkpI4HE5lZeX79+95X+naGxwOJyUlRZJ3r9HpdDMzs9LS0s/WxE84Kyoq8j4KsR3SaLS+Vrenp6eenl5fT1T19fU1NjZeu3ZtcXFxdXV1QEBAS0sLflHYZ+FLh9/AzWcug60rgDwb4JV1oF8EvHsNw7Bjx45ZWVnRaDQajTZx4sTo6GgMwzo7OyMiIiwsLCgUipaWlqura35+PoZh0dHR+JUvFhYWhYWFsbGx6urqCCFjY+O//vqrr9YCAgK0tbU1NTXd3d3xu67Nzc2Li4tfvnxpbGysrKzs4OBQXl7e1tYWEBBgZGREJpN1dXXd3Nxyc3P5z3EgLX+2Z3x9fVeuXNnfnkf9vF736NGj+H3kdDrd2dmZ//IymUwKhWJoaEgmk9XV1RctWlRYWIi3U11dPWvWLBqNZmpq+v333/v7+yOEGAxGz95ITk5WU1MLDQ3t76IJffeal5cXhUJhs9n4x8TERHNzc4SQjo4OflF3V/7+/rxbtoTbDvta3a6urgihoKCgviIvKSlZtmyZlpYWlUqdOnVqSkpKzzq9XuDt5ORkaGiI39/Ffy6DqitwcK27XIKMLlGCZ3QgWmLdKzGZTG1tbTE1/llCZ/SCggIymXz27FmxhSYQLpfr6Oh46tQp0TZbVVVFo9EOHjwoyFwGVVfgIKPLJTjrDoAICPemLwlraWm5detWQUEBfp0Ug8EICQkJCQnp+bYPieFyuUlJSY2NjZ6enqJtOTg42Nra2svLS5C5DJ6uwDCsrKzs4cOHb968Ee1cgDSAjA6kyOvXr3u+LJVH5Hu6waampgZ/U8s333yDl+zcudPd3d3T05OoN5Gkp6dfvnw5JSWF/+3g/XXo0CEWi5WcnIw/DkGQuQySrrh69Sr+ppauL5QD8oPokwSDC5x1JwoS25nDnTt34g8PMTExSUhIEMcs+BPwrDsft27dCggIEFU8hEtKStq/f39HR4cQ34Wu6El8YweIHAmDZ/dLUHx8vIeHB/S55JFIpIsXLy5dupToQETP3d0dIZSQkEB0IEA+yfHYkT9w1h0AAACQB5DRAQAAAHkAGR0AAACQB5DRAQAAAHkAGR0AAACQB2SiAxiM+np5CRArDw8PDw8PoqMQF9ioAACQ0QmA35UOJMnDw8PHx8fOzo7oQETv8OHDCCFpe3UNkBty/H+w/IGMTgC4s1PyPDw87Ozs5LLn8TvR5XLRgDSAjC5D4Hd0AAAAQB5ARgcAAADkAWR0AAAAQB5ARgcAAADkAWR0AAAAQB5ARpcx+fn533///bhx49TU1MhksoaGxsiRI52cnJ48eUJ0aAD0T1pa2s6dOy9fvmxmZkYikUgk0qpVq7pWmDdvnpqamqKi4rhx416+fCn5CDs7Ow8fPmxvb9+1MCQkZOzYserq6lQqlcFgbN++vampiTf1119/nTJlipqamrGx8bp168rLyxFC165dCw8P53K5kl4AMMhARpclp06dsrKyysrKOnToUElJSXNzc2Zm5t69e+vq6rKzs4mODoB++PHHH6Oionbt2uXm5vb27Vtzc/MhQ4acO3fut99+49VJTU1NSEhYuHBhbm7upEmTJBxhQUHBF1984evry2azu5bfuXNny5YtRUVFVVVV+/fvj4yMxF9oixC6ePHiihUr3N3dS0tLr169ev/+/QULFnR0dDg7O9NotDlz5tTV1Ul4KcCgAhldZjx9+pTJZDo6Ot6+fXv+/PmamppUKtXMzMzDwyMoKKi9vV1ikbS0tHQ7apH+lsVHJDHL4oIPRFhY2IULF+Lj49XU1HiFUVFRCgoKTCazvr6ewNhwr1692rFjx8aNG62trbtNUlVVZTKZ2traampqS5cudXV1vXnzZklJCULoxIkTw4YN8/f319DQsLa29vX1ZbFYz549Qwh5e3tPmDDh66+/7ujoIGB5wOAAGV1mhIaGcrncAwcOkMndnws0f/78LVu2SCySU6dOVVRUyFbL4iOSmGVxwYX25s2bH374Yc+ePTQarWu5vb29j4/Phw8ftm3bRlRsPBMmTLh8+fKKFSuoVGq3STdu3FBUVOR91NHRQQjhx/ElJSUGBga8J/KOGDECIfT+/Xv8Y3BwMIvFioyMlED8YHCCjC4b2tvbb9++PWTIkKlTp/KphmHYoUOHxowZQ6VStbS0Fi1a9Pr1a4RQTEyMiooKnU6/evXqggUL1NXVhw8fHhcXx/vi2bNnbWxsaDSaioqKiYnJ3r17EUIPHjwYO3ashoYGjUazsrK6desWQsjHx8fPz6+wsJBEIjEYDIQQl8sNCgoyMjJSVlYeP348/oxb/nMcSMvi01fveXl5KSkp6evr49U2b96soqJCIpGqqqq6xRwVFUWj0YYOHbphwwYDAwMajWZvb48fogneCELo5s2b6urq+/btE+vyEiUqKgrDMGdn556TQkNDR44c+fPPP6elpfWcKtzmLe6t6MOHD8rKyqampgghMzOzrv+Z4T+im5mZ4R+1tLRmzJgRGRmJYZhoYwDgPzAgQfjeRIgv/vXXXwghW1tb/tWCgoKUlJTOnj1bV1eXlZU1adIkHR2d8vJyDMMCAwMRQrdv366vr6+oqHB0dFRRUWlvb8cwDH8w+IEDB6qrq2tqak6cOLFixQoMwxISEoKDg2tqaqqrq21tbYcMGYLPxc3NzdzcnDfTbdu2UanUS5cu1dbW7tq1S0FB4cWLF/znOMCWhYAQunjxotC9t2LFCj09PV7NiIgIhFBlZWXPmJlMpoqKSl5eXmtra25uLn6RVHFxcb8auXHjhpqaWkhIiCCLtmTJkiVLlghSU0qYmZmNHTu2W6G5ufm7d+8wDHv8+LGCgoKJiUlTUxOGYSkpKS4uLngd4TbvAW5F06ZNmzBhQl9Tm5ub1dTUvLy88I/p6ekUCiUqKqqhoSEnJ2fMmDHz58/vWn/nzp0IoczMTMEDIJwgYwdICcjoEiV0Rs/IyEAIffnll3zqsNlsVVVVT09PXsnz588RQnhiwHd5LS0t+KTo6GiE0Js3b9rb2zU1NWfNmsX7VkdHB34Y0dX+/fsRQhUVFdj/pp+WlhY6nc6bKZvNplKpmzZt4jPHgbcshM/ulfj3Xr8yuoaGBu/jixcvEEJ79uzpVyP9IlsZvampiUQiLVy4sFs5L6NjGObn54cQ2rJlC9Ylowu3eQ98K+Kf0QMDA0eOHNnQ0MAr2b17N+94afjw4SUlJV3r//LLLwihM2fOCB4A4SCjyxA46y4bVFVV0X9/q+tLbm5uU1OTjY0Nr2TKlClKSkr4Wd9ulJSUEEIcDicrK6uurm7+/Pm8SYqKit7e3t3qUygUhFDP22/y8/PZbLalpSX+UVlZWV9fHz8X2tccRd6ySPSr9wRnY2NDp9PFF7bMwf91o9PpfOqEhoaOGjUqOjr64cOHvELhNm+xbkWJiYnx8fG3bt3iXd8XGBgYGxt7+/btpqamt2/f2tvb29nZ4RfN4fAF//Tpk0gCAKAbyOiywcTEhEaj4efe+4LfGIPnfh5NTc3GxkY+32poaMCr9Zz022+/zZw5U1dXl0qlbt++vdevNzc3I4R2795N+q/374RMcLYAACAASURBVN/z/89DrC0LTbjeEwSVSq2srBxgI3KjtbUVIdTzcrOuaDTa6dOnSSTSN99809LSghcKt4LEtxVduHAhLCwsPT3dxMQEL/n48WN4ePj69etnz56toqJiamp68uTJsrIy/GQMTllZGf23EwAQOcjosoFKpc6fP7+qqurRo0c9p9bU1Hz33Xd4Vu62g6urqxs+fDiflocNG4YQqqqq6lZeXFzs6uqqr6//7Nmz+vr68PDwXr+uq6uLEDp8+HDXMz/8H3cjvpYHQrje+ywOhzPwRuQJntI++6wVOzs7X1/fgoIC/CJNJOwKEtNWdPTo0XPnzt25cwcfPriCggIul9u1RF1dXVtbOzc3l1eC32WKdwIAIgcZXWYEBwdTqVRfX1/eUQtPTk4OmUy2tLRUVVXFf3HHPXv2rL29ffLkyXyaNTEx0dbWTk1N7VaenZ3N4XA2bdpkZmZGo9F4N+R0M2LECBqNxmKxBF8Q8bU8EPx7j0wm9/y9QBDp6ekYhtna2g6kEXkydOhQEokkyB3ne/fuHT16dGZmJv5RuM1b5FsRhmEBAQHZ2dlJSUndThjg/1t8/PiRV9LY2FhTU4Pfw4bDF1xPT09U8QDQFWR0mWFtbX3+/PmcnBxHR8fk5OT6+noOh/Pu3buTJ09+++23FAqFRqP5+fklJiaeO3euoaEhOzt748aNBgYGTCaTT7NUKnXXrl3379/38vL68OFDZ2dnY2NjXl6ekZERQigtLa21tbWgoKDrr5Xa2tplZWVFRUWNjY2Kiorr1q2Li4uLiYlpaGjgcrmlpaVdd2o9ia/lgeDfewwGo6amJikpicPhVFZW8u4w7hYznrA7Oztra2s7OjqysrJ8fHyMjIzWrl3br0ZSUlLk9e41Op1uZmZWWlr62Zr4uXfend/Cbd40Gq2vrcjT01NPT6+/D5fNy8v7xz/+cfLkSQqFQuri4MGDpqams2bNOnny5P3791taWkpKSvDYvv32W97X8QW3srLq10wBEJTYr70DXQh9rTtPcXHxtm3brKysVFVVFRUVNTU1J06c+O233z569AjDsM7OzoiICAsLCwqFoqWl5erqmp+fj2FYdHQ0fkmOhYVFYWFhbGysuro6QsjY2Pivv/7CMOzYsWNWVlY0Go1Go02cODE6OhrDsICAAG1tbU1NTXd392PHjiGEzM3Ni4uLX758aWxsrKys7ODgUF5e3tbWFhAQYGRkRCaTdXV13dzccnNz+c9xIC0L129IgOt1++o9DMOqq6tnzZpFo9FMTU2///57f39/hBCDwegZM5PJpFAohoaGZDJZXV190aJFhYWF/W0kOTlZTU0tNDRUkEWTrWvdMQzz8vKiUChsNhv/mJiYaG5ujhDS0dHBr2/vyt/fn3f3mnCbd19bkaurK0IoKCio1yCfPHkyffp0AwMDfD+pr69vb29/7969vh63HBERgWEY/nQBBoNBpVJVVVWnT59+5cqVrs06OTkZGhp2dnaKtEfFS5CxA6QEZHSJGnhGB8KR2F4Jfz6oBGbEI3MZvaCggEwmnz17ltgwuFyuo6PjqVOnJDbHqqoqGo128OBBic1RJCCjyxA46w6AiMErtvhjMBghISEhISFdX1kmYVwuNykpqbGx0dPTU2IzDQ4Otra29vLyktgcwWADGR0AIGk7d+50d3f39PQk6qUs6enply9fTklJ4X9nvAgdOnSIxWIlJyfjD2AAQBwgowMgMrt27Tp9+nR9fb2pqemlS5eIDkeq7du3z8vL68CBA4TMfc6cOefPn+c9Zl/crl692tbWlp6erqWlJZk5gsGp+1u8AABC279/P/5QWyCIefPmzZs3j+goJMHFxcXFxYXoKID8g2N0AAAAQB5ARgcAAADkAWR0AAAAQB5ARgcAAADkAVwZR4D4+HiiQxiMxPeWF2LhDxaFjQoAQMIwjOgYBpH4+HgPDw+iowAAgH64ePHi0qVLiY4CfB5kdABE4Ny5c5s3bzY3N4+Lixs1ahTR4UhUfn7+smXLCgsLo6OjV65cSXQ4AAxe8Ds6AAPS2Ni4evXq1atXr1279smTJ4MtnSOERo0a9ezZs61bt65Zs2bp0qV1dXVERwTAIAXH6AAILyMjY9myZfX19adPn3ZyciI6HIKlpaWtXr1aSUnp/Pnz06dPJzocAAYdOEYHQBgYhh05cmT69OnGxsYsFgvSOULoyy+/ZLFY48aNmzlzZnBwMLyxBgAJg2N0APrt06dPa9euTUtLCwwMDAoKUlCA/4z/H4ZhUVFRAQEB9vb2Z8+eNTQ0JDoiAAYLyOgA9E9qauqaNWuUlZV//fVXW1tbosORUi9fvly2bFlNTc0vv/yycOFCosMBYFCAYwsABNXa2urt7f3VV1/NnTs3KysL0jkfkyZNYrFYy5cvd3Z2Xr16NZvNJjoiAOQfHKMDIJDXr18vW7bs7du3MTExK1asIDocmXH58uW///3vBgYGcXFx48ePJzocAOQZHKMD8HlnzpyxsbFRUlJ6+fIlpPN+cXNzY7FY2tra06ZNO3LkCNHhACDPIKMDwE99ff2yZcvWrl377bffPnz40NzcnOiIZI+RkdHdu3cDAgL8/PxcXV2rq6uJjggA+QRn3QHo09OnT5cvX87hcM6dOzdjxgyiw5F50J8AiBUcowPQi46OjuDgYAcHhwkTJrBYLEg/ImFra5uZmeng4DBr1ixvb28Oh0N0RADIFThGB6C74uLiFStWZGRkhIWFeXt7Ex2OHDpz5symTZvGjRv366+/wg8ZAIgKHKMD8D8uX75sbW1dU1Pz7NkzSOdisnr16oyMjPb29kmTJp0/f57ocACQE5DRAfiPlpYWb2/vJUuW/O1vf3vx4gXcaiVWo0ePfvLkydq1a1etWrV69eqmpiaiIwJA5sFZdwAQgmecEQeewQeAqMAxOhjs8Heu2NvbGxoaslgsSOcSNm/ePBaLNWrUKEdHx+Dg4M7OTqIjAkBWwTE6GNQqKirWrVuXmpoK71whFv5+l+3btzs6Op45c2bYsGFERwSA7IGMDgYv/H3eVCr1/Pnz9vb2RIcDUEZGxvLly+vq6uB98wAIAY5IwGDE4XCCg4Pnz5/v4OCQmZkJ6VxK2NjY/PHHH66urgsXLvT29m5rayM6IgBkCRyjg0EnPz9/2bJlb968iY6OXrVqFdHhgF4kJCSsX7/e0NDwwoULlpaWRIcDgGyAY3QwuODvXFFUVPzjjz8gnUstd3f3zMxMDQ2NqVOnwvtdABAQZHQwWDQ0NKxYsWLt2rXffPPNo0ePLCwsiI4I8GNiYnLv3r3t27f7+vq6ubnV1NQQHREA0g7OuoNB4dmzZ8uXL29ubv7Xv/711VdfER0O6Ie7d++uWrVKUVHx3Llzjo6ORIcDgPSCY3Qg57hcbnh4uKOjI4PBYLFYkM5lzqxZs1gs1oQJE2bNmrVjxw54vwsAfYFjdCDPSkpKVq5c+fz587CwMC8vLxKJRHREQHhnzpzZuHHj+PHjz58/b2ZmRnQ4AEgdOEYHcuvKlSvW1tYVFRVPnjzx9vaGdC7rVq9e/eLFCzabPWnSpLi4OKLDAUDqQEYHcgh/58rixYudnJwyMjKsra2JjgiIxtixY589e7ZmzZrly5evXr26ubmZ6IgAkCJw1h3Im9zc3GXLlhUXF//000+enp5EhwPEIikp6bvvvtPV1f31118nTpxIdDgASAU4RgfyA8Ow2NjYqVOnqqioZGZmQjqXY4sWLWKxWHp6enZ2duHh4fB+FwAQZHQgNyorK52dnTdv3uzv7//w4UNTU1OiIwLiNXz48Lt374aHh//www9fffXVx48fiY4IAILBWXcgD+7cubNq1SoymXz+/HkHBweiwwES9fz58+XLlzc2Nv7rX/9asGAB0eEAQBg4RgeyraOjIzg4eO7cuXZ2diwWC9L5IDR16tSXL1/OnTvXycnJ29u7vb2d6IgAIAYcowMZVlRUtHz5chaLdeDAAW9vb6LDAQQ7c+bM5s2bR40aFRcXB0/5BYMQHKMDWZWQkDBx4sS2tjYWiwXpHCCEVq9enZWVRaFQJk+eHBsbS3Q4AEgaZHQgvVpaWnotb2hoWLVqlYeHx+rVqx8/fjxy5EgJBwaklqmp6YMHD3x9fTdu3Oju7l5bW9trtb42LQBkGmR0IKU+fvxoY2Pz6dOnbuUvXryYPHlyamrqjRs3jhw5QqVSCQkPSC0ymRwcHJyamvr48eOJEyc+evSoW4VPnz7Z2NjAtfFA/kBGB9IIw7C1a9fm5eWtWLGCd6kHhmFHjhxxcHAwMTFhsVhff/01sUECaTZnzhwWi2VpaTlz5szg4GAul4uXYxi2cuXKvLy8NWvWwFVEQN5gAEif6Oho/DHsCgoKBw8exDCsvLx8/vz5FArlxx9/5HK5RAcIZENnZ2dkZCSVSp05c2ZJSQmGYQcPHlRQUEAIkUik6OhoogMEQJTgWncgdQoLC62srHi/dCoqKh48eDA0NFRHRycuLg4e+Qn6KzMzc9myZVVVVT/++KOvr29HRwdeTqVSX716NWrUKGLDA0BUIKMD6dLR0WFra5uVlcV7DbaiouKQIUO++uqrmJgYFRUVYsMDMqq5uXnz5s2pqamVlZW8jE4mk8eOHZuRkUGhUIgNDwCRgN/RgXTZu3cvi8XipXOEEJfLrampUVBQgHQOhIZvPF3TOUKoo6MjNzd33759xMUFgCjBMTqQIn/88ce0adN4FzF1c/78+eXLl0s4JCAfLl265O7u3uskBQWFR48e2draSjgkAEQOMjqQFs3NzVZWViUlJV2PonhIJBKdTs/JyTExMZF4aEC2FRUVWVpastnsXnd3ioqKI0aMyMnJgZNAQNbBWXcgLfz8/PpK5wghRUXF5ubmdevWSTgqIAfWrVvX3NysqKjY61Qul1taWurn5yfhqAAQOThGB1IhNTX1q6++6rk1UigUDoejoaExd+7chQsXuri4aGhoEBIhkF1sNvv27dvXr19PSkqqrKxUUlLq+TYXEol0/fp1JycnQiIEQCQgowPiVVVVjRkzpqamprOzEyGkqKhIIpE6OjosLCwWL1785Zdfzpw5k0wmEx0mkHmdnZ2ZmZlpaWmJiYkvXrzAb0zHr9tQUFDQ0ND4888/9fT0iA4TACFBRgfEc3V1TUpKUlRU5HK5SkpKM2fOdHV1dXJyGjFiBNGhAblVUlLy22+/JSUlpaent7W1kcnkjo4OFxeXpKQkokMDQFiSfJzNxYsXiV5cAGTSkiVLBj4AlyxZQvRyACCTLl68OPABKAEEnMmEvM5z+PBhhNDWrVuJDkT0njx5EhkZ+dl1zeFwfv/99zFjxpiYmOCPfQU94duJSNja2srl9iYcDw8PHx8fOzs7/COGYUVFRX/++efcuXNl/ZkzcrxvkTwPDw+iQxAUARl96dKlkp+pdEpISEDy2yGRkZGCLNqKFSskEIxMw7cTkRg+fLi8bm9C8PDwsLOzk8sOke99i4TJUEaHu9cAAAAAeQAZHQAAAJAHkNEBAAAAeQAZHQAAAJAHkNEBAAAAeSDtGf27775TU1MjkUgsFovoWBBCqLOz8/Dhw/b29v2aJELJyckaGhrXr18X61wkLy0tbefOnZcvXzYzMyORSCQSadWqVV0rzJs3T01NTVFRcdy4cS9fvpR8hL2u35CQkLFjx6qrq1OpVAaDsX379qamJt7UX3/9dcqUKWpqasbGxuvWrSsvL0cIXbt2LTw8vK/3y0kbmRiA/NeCaMEAhAEo1SR58zt+d3J/vxUXF4cQyszMFEdI/fLXX39Nnz4dITRhwgTBJ/GxZMmS/j455MaNG+rq6teuXevXtySvX+s6KCho4cKFDQ0N+Edzc/MhQ4YghG7cuNG1WkpKiouLi4gDFUxf63fGjBnR0dHV1dUNDQ0XL16kUCj40+kxDLtw4QJCKDw8vK6uLjMz08zMzNramsPhYBgWGRk5Y8aM2tpaAecuxHYiwnakfwDyWQv8of4/OURWBmC/1jUMQP6E2E6IIu3H6NLj1atXO3bs2Lhxo7W1teCTRM7Jyam+vn7hwoViar+lpUXcpxm6CQsLu3DhQnx8vJqaGq8wKipKQUGByWTW19dLMphe8Vm/qqqqTCZTW1tbTU1t6dKlrq6uN2/eLCkpQQidOHFi2LBh/v7+Ghoa1tbWvr6+LBbr2bNnCCFvb+8JEyZ8/fXXfb1oDvQk3FoQORiAkgcDUHAykNGl5FFiEyZMuHz58ooVK6hUquCTZM6pU6cqKiokNrs3b9788MMPe/bsodFoXcvt7e19fHw+fPiwbds2iQXTFz7r98aNG13f0amjo4MQYrPZCKGSkhIDAwPe1os/o/79+/f4x+DgYBaLFRkZKYH4B0j6ByCftSBzYAD2NMgHYL9IY0bHMCwiImLUqFFUKlVDQ8Pf3583icvlBgUFGRkZKSsrjx8/Hj+1GxMTo6KiQqfTr169umDBAnV19eHDh+OnChFC9+7dmzp1Kp1OV1dXt7Kyamho6Ksd6ffw4UMjIyMSiXTs2DHEd8GjoqJoNNrQoUM3bNhgYGBAo9Hs7e3x/0+9vLyUlJT09fXxNjdv3qyiokIikaqqqnx8fPz8/AoLC0kkEoPBQAjdvHlTXV193759YlqiqKgoDMOcnZ17TgoNDR05cuTPP/+clpbWcyqGYYcOHRozZgyVStXS0lq0aNHr16/59wkS/3r/8OGDsrKyqakpQsjMzKzrrhn/Dc/MzAz/qKWlNWPGjMjISEz6XpUk6wOw61oQLRiAPDAApZQkT/EL+NtqYGAgiUT65z//WVtby2azo6Oj0X9/xtu2bRuVSr106VJtbe2uXbsUFBRevHiBfwUhdPv27fr6+oqKCkdHRxUVlfb29qamJnV19fDw8JaWlvLy8sWLF1dWVvJpRxDTpk3r68dyPpN6JcTvmvgJpaNHj+If+1pwDMOYTKaKikpeXl5ra2tubi5+hUhxcTGGYStWrNDT0+O1GRERgRDCe8bNzc3c3Jw36caNG2pqaiEhIf0KEhN4XZuZmY0dO7Zbobm5+bt37zAMe/z4sYKCgomJSVNTE/a/P+MFBQUpKSmdPXu2rq4uKytr0qRJOjo65eXl/PtkIOsd+9z6bW5uVlNT8/Lywj+mp6dTKJSoqKiGhoacnJwxY8bMnz+/a/2dO3ciwX6flvDv6LI7ALEea4E/1P/fR2VlAAq4rmEACjIAhdhOiCJ1GZ3NZtPp9Llz5/JKeBfmtLS00Ol0T09PXk0qlbpp0ybsv9tQS0sLPgnfB7158yYnJwf1uL6DTzuCkMKM3nPBMQxjMpkaGhq8L7548QIhtGfPHqw/OxShCbKum5qaSCTSwoULu5XzdigYhvn5+SGEtmzZgnXZobDZbFVVVd4axDDs+fPnCCF8x9dXnwxwvWOfW7+BgYEjR47kXV6EYdju3bt5/zoPHz68pKSka/1ffvkFIXTmzJnPzleSGV2mByDW21rgQ1QZXQoHoCDrGgaggANQhjK61J11f/PmDZvNnjNnTs9J+fn5bDbb0tIS/6isrKyvr4+f6ulGSUkJIcThcMzMzIYOHbpy5crg4OCioqL+tiNzeAvec5KNjQ2dTpeqxayoqMAwjE6n86kTGho6atSo6Ojohw8f8gpzc3ObmppsbGx4JVOmTFFSUsLPanbD6xOxrvfExMT4+Phbt27xLi8KDAyMjY29fft2U1PT27dv7e3t7ezsul6uhS/4p0+fRBKAqMj0AOy5FiQMBmDPFmAASpLUZfTS0lKEkK6ubs9Jzc3NCKHdu3eT/uv9+/f8r39RVla+c+eOg4PDvn37zMzMPD09W1pahGhHPlCp1MrKSqKj+H+tra0IIf6XE9JotNOnT5NIpG+++aalpQUvrKurQwipqqp2rampqdnY2MinKfGt9wsXLoSFhaWnp5uYmOAlHz9+DA8PX79+/ezZs1VUVExNTU+ePFlWVoYfjeGUlZXRfztBesjuAOy5FqQNDEAEA1DMpC6j45dctrW19ZyE72UOHz7c9STDkydP+Dc4bty469evl5WVBQQEXLx48eDBg8K1I+s4HE5dXd3w4cOJDuT/4SPqs496sLOz8/X1LSgo2Lt3L16iqamJEOq2+/js0olpvR89evTcuXN37twZNmwYr7CgoIDL5XYtUVdX19bWzs3N5ZW0t7ej/3aC9JDRAdjrWpAqMABhAEqA1GV0S0tLBQWFe/fu9Zw0YsQIGo3Wr2dXlZWV5eXlIYR0dXUPHDgwadKkvLw8IdqRA+np6RiG2draIoTIZHKvJwYlbOjQoSQSSZAbXvfu3Tt69OjMzEz8o6WlpaqqakZGBq/Cs2fP2tvbJ0+ezKcRka93DMMCAgKys7OTkpK6Ha/gu7aPHz/yShobG2tqavBbaHD4guvp6YkqHpGQuQHIZy1IFRiAMAAlQOoyuq6urpub26VLl06dOtXQ0JCVlRUbG4tPotFo69ati4uLi4mJaWho4HK5paWlXddZT2VlZRs2bHj9+nV7e3tmZub79+9tbW2FaEdGdXZ21tbWdnR0ZGVl+fj4GBkZrV27FiHEYDBqamqSkpI4HE5lZSXvHk2EkLa2dllZWVFRUWNjI4fDSUlJEd/NM3Q63czMDD/Nyx9+6o934ymNRvPz80tMTDx37lxDQ0N2dvbGjRsNDAyYTCb/Rvpa756ennp6ev19tmVeXt4//vGPkydPUigUUhcHDx40NTWdNWvWyZMn79+/39LSUlJSgsf27bff8r6OL7iVlVW/ZipuMjcA+awF4RoUIRiA3RqBASh2A7yyrl8EvKOpsbHxu+++GzJkiKqqqoODQ1BQEEJo+PDhr169amtrCwgIMDIyIpPJ+K4nNzc3Ojoav8bBwsKisLAwNjZWXV0dIWRsbPz777/b29traWkpKioOGzYsMDCwo6MDw7Be2+Ef1ZMnT6ZPn25gYID3m76+vr29/b179/hP4q+/1zAfPXoUv42VTqc7OzvzWfC//vqLyWRSKBRDQ0Mymayurr5o0aLCwkK8nerq6lmzZtFoNFNT0++//x6/4ZjBYBQXF798+dLY2FhZWdnBwaG8vDw5OVlNTS00NFTwIHECrmsvLy8KhcJms/GPiYmJ5ubmCCEdHR388tqu/P39eTfPdHZ2RkREWFhYUCgULS0tV1fX/Px8DMP490lf693V1RUhFBQU1GuQfa3f7OzsXsdUREQEhmH47cUMBoNKpaqqqk6fPv3KlStdm3VycjI0NOzs7PxsL0n47jXZGoD81wJ/qJ/XMMvQABRwXcMAFGQA9nc7IZA0ZvTBQ1R76l7hD0cUU+OfJeC6LigoIJPJZ8+elUBIfHC5XEdHx1OnTklsjlVVVTQa7eDBg4JUJva57nJMrHtqYgeggOsaBqAglWUoo0vdWXcgQtL/fiEGgxESEhISEiK+l2V9FpfLTUpKamxs9PT0lNhMg4ODra2tvby8JDZHIHkwAAUBA1CEIKP/x+vXr0l9k+SmNtjs3LnT3d3d09OTqHdCpKenX758OSUlhf+NuSJ06NAhFouVnJxMoVAkM0fpBwOQKDAA5Qlk9P8YPXo0n1MZ+Iv5ZMiuXbtOnz5dX19vamp66dKl/2PvzgOautLG8Z9AlpuEbOwUZEcpi1pcoTLqMHVUpiyiAi6t+vqWWi1SFZFFRASXF6sMFepYGd5RFBV0oFWxrba0ZbSMDkQQFRFZRKsssieRBO7vj/trXr4gIUDIDeH5/EXOPTnnuVse7nLuJTucISQmJoaGhh44cICU3r28vM6cOSN/zvZYy8vLe/36dUFBgUAgUE+P4wLsgCSCHVBrUMkOAIyJ/fv379+/n+wohmHRokWLFi0iOwp18PX19fX1JTsKMLZgB9RY2r0DwjE6AAAAoA0gowMAAADaADI6AAAAoA0gowMAAADagIQ74y5cuKD+TjUT8RhCrVwgxAsYtHLW1K++vl5Vb/ior6+HldKXtr6iSYt/W4Aiqn5kjSLEc8QAAMOlqmfGkT0fAIxL4+WZcSQco+M4rv5ONdOKFSsQQtnZ2WQHonoXLlwIDAyEda0SxHaiEsuXL9fK7W1kKBTK+fPnV65cSXYgqqfFvy3qR6FQyA5BWXAdHQAAANAGkNEBAAAAbQAZHQAAANAGkNEBAAAAbQAZHQAAANAGkNEBAAAAbTAuM/rFixdtbW37vj6ZTqcbGxsvWLAgKSmppaWF7ADBaF2/fj0yMrLvil67dm3fCosWLeJwOLq6us7OzsXFxWoOTyqV7t+/397enk6n8/l8FxeXmpqafnUkEomjo2NMTAxC6Ouvvz506FBPT4+a4xwjsANqN9j7xq9xmdEDAgKePHliZ2fH4/FwHO/t7W1oaLhw4YKNjU1ERISzs/OdO3fIjhGM3J49e1JSUqKiouQr2sDAIDMz88qVK/I63333XXZ29vvvv19eXu7m5qbmCAMDA0+dOnXmzBmRSPTgwQM7O7vOzs5+daKjoysqKoi/fXx8MAzz8vJqbW1Vc6hjAXZALQZ737g2LjN6PxQKhc/nL1iwICMj48KFCy9fvvT29m5rayM7LpKJxWIPDw9NaGRYDh48eO7cuQsXLnA4HHlhSkqKjo5OSEiIJqzWc+fO5ebmZmdnz5kzh0qlmpmZ5eXlubi49K1z8+bNe/fu9S3ZunXrtGnTli5dKpPJ1BvvmIMdcCDY+8YI7H2KaUNG72v58uXr1q1raGg4fvw42bGQLD09vaGhQRMaUd7jx4937969d+9eDMP6lnt4eISFhT179mzHjh1qC2YwX375pZubm6ur62AVxGJxeHh4cnJyv/K4uDihUDiwXJvADkiAvW+MwN6nmLZldITQunXrEEL5+fkIoZ6entjYWEtLSyaTOXXqVOLB8mlpaWw2m8Vi5eXlLVmyhMvlWlhYZGVlEV//6aefZs+ezWKxuFyuq6trTiFkYAAAIABJREFUe3v7YO2oDY7jR44cefvttxkMhkAg8PPze/jwIUIoNDSUTqebmpoS1TZv3sxmsykUSlNTU1hY2Pbt26uqqigUir29fUpKCoZhxsbGH3/8sZmZGYZhHh4eRUVFw2oEIXTt2jUul5uYmDhGc5qSkoLjuI+Pz8BJCQkJkydPPnny5PXr15VfRIrX9QhWa3d396+//jp9+nQFdaKjozdv3mxkZNSvXCAQzJ8/Pzk5WbsfjqtlOyDsfQj2vnFEnQ+RJ9aZqlqTX8brh/gJmDRpEo7jO3bsYDAYOTk5LS0tUVFROjo6t2/fxnE8OjoaIXTjxo22traGhgZPT082m93d3d3Z2cnlcg8dOiQWi1+8eLFs2bLGxkYF7YzS8uXLlXkDR2xsLJ1OP336dGtra2lpqZubm6Gh4YsXL3AcX716tYmJibxmUlISQoiIOSAgwM7OTj4pJCSEzWbfv39fIpGUl5fPmjWLw+HU1dUNq5HLly9zOJz4+PghYx7Zura1tXVycupXaGdnV11djeP4zZs3dXR0rK2tOzs7cRzPz8/39fUl6ihYRIOta3xEq7W6uhohNH369AULFpiamjIYDEdHx2PHjvX29hIVCgsLfXx8cBxvbGxECEVHR/f9emRkJEKopKRkWItFye1Ebe0QtGAHREq8gWM87n34iNY17H2DUWY70RBaeIzO4XAoFEpHR4dEIklLS/P39w8ICODz+TExMTQaLSMjQ17Tw8ODy+UaGRkFBQV1dXXV1dXV1NS0t7c7OztjGGZiYnLx4kVDQ8Mh2xlTYrH4yJEjy5YtW7NmDY/Hc3V1PX78eFNT04kTJ4bbFJVKJf6PdnJySktL6+joGO5ceHt7t7e37969e7hdK6Orq6u6utrOzm6wCu7u7p999llNTc2uXbv6liuziAau65GtVuIeHCMjo8TExPLy8pcvX/r5+W3ZsuXs2bNEJGFhYWlpaYN93cHBASFUVlam3CIZl7RpB4S9Tw72vnFBCzN6V1cXjuNcLreiokIkEslvmmAymaampsTpoH7odDpCSCqV2traGhsbr1mzJi4uTj4iQvl2xkJ5eXlnZ+fMmTPlJbNmzaLT6cRZuxGbOXMmi8VS21woo6GhAcdxFouloE5CQsKUKVNSU1MLCwvlhcNaRPJ1PbLVymAwEELOzs4eHh76+vo8Hm/v3r08Ho/4/YqKivroo4/Mzc0H+zoxdy9fvlTcy7imTTsg7H19wd6n+bQwoz969Agh5Ojo2NXVhRCKiYmRj5qtra0ViUQKvstkMn/44Yd58+YlJiba2toGBQWJxeIRtKNCxIgLPT29voV8Pr+jo2OULTMYDOLclIaQSCTo9512MBiGZWRkUCiUDRs2iMVionBki2hkq9XMzAwh1NTUJC+h0+lWVlZVVVWFhYVlZWUbN25U8HUmk4l+n1NtpU07IOx9fcHep/m0MKNfu3YNIbRkyRLi5oijR4/2vcxw69YtxV93dnb+5ptvnj9/HhERcf78+cOHD4+sHVXh8/kIoX67R2trq4WFxWialUqlo29EtYj9bcgHQbi7u2/btq2ysnLfvn1EycgW0chWq56enoODw/379/sWymQyHo+Xnp5+48YNHR0d4heKaD8xMZFCocjHZ3d3d8vnVFtp0w4Ie18/sPdpOG3L6C9evDh69KiFhcWGDRsmTZqEYZhQKFT+68+fPyc2FyMjowMHDri5ud2/f38E7aiQi4uLnp5e30d2FBUVdXd3z5gxAyFEpVKlUukImi0oKMBxfO7cuaNpRLWMjY0pFIoyY1737dvn6OhYUlJCfFS8iAYz4tUaGBhYUlLy5MkT4qNIJKqtrXV1dc3IyOj789T33hz5OUli7kxMTIbb6XihZTsg7H0Dwd6nycZ3RsdxvLOzk7jRsbGx8fz58++++66urm5ubi6Xy8UwbP369VlZWWlpae3t7T09PfX19b/99puCBp8/f/7xxx8/fPiwu7u7pKSktrZ27ty5I2hHhTAM2759+6VLlzIzM9vb28vKyjZt2mRmZhYSEoIQsre3f/XqVW5urlQqbWxsrK2tlX9RX1//+fPnNTU1HR0dxE9Gb29vS0uLTCYrLS0NCwuztLQkBhop30h+fv7YjZ9hsVi2trb19fVD1iTO/unq6so/KlhEChoZbLUGBQWZmJgM9njLbdu2WVlZrVu3rq6urrm5OSIiQiwW97tdaDDE3CkYTTu+aP0OCHvfQLD3aTRlbohXFVWNXvv666+nTp3KYrHodLqOjg76/alVs2fPjo+Pb25ultd8/fp1RESEpaUllUo1MjIKCAgoLy9PTU0lbpFwcHCoqqo6ceIEl8tFCFlZWX3//fceHh4CgUBXV/ett96Kjo6WyWSDtTP6GVFyhElvb29SUpKDgwONRhMIBP7+/hUVFcSk5ubmhQsXYhhmY2Pz6aefhoeHI4Ts7e3r6uqKi4utrKyYTOa8efNevHgREhJCo9HMzc2pVCqXy/Xz86uqqhpuI1evXuVwOAkJCUPGPLJ1HRoaSqPRRCIR8fHSpUvEzbeGhoZbtmzpVzk8PFw+fmawRaRgXT969Giw1erv748Qio2NHSzOp0+fBgcHCwQCBoMxe/bs/Pz8gXXeOH7G29vb3NxcPthGSZo2ek1rdkCkxKik8bj34SNa17D3DUaZ7URDjMuMrjVUOz5YsZCQEH19ffX0hY90XVdWVlKp1NOnT49FSMrr6enx9PRMT09XbbNNTU0Yhh0+fHi4X9S0jK411PZLrea9Dx/Ruoa9bzDjKKOP77PuYFg0/+1D9vb28fHx8fHxA1+9oDY9PT25ubkdHR1BQUGqbTkuLm769OmhoaGqbRaMC7D3KQP2vlGCjA40S2Rk5IoVK4KCgsh6LURBQcHFixfz8/MVj80driNHjgiFwqtXr9JoNBU2C4AKwd433kFGnxCioqIyMjLa2tpsbGxycnLIDmcIiYmJoaGhBw4cIKV3Ly+vM2fOyB+1rRJ5eXmvX78uKCgQCAQqbBaMC7D3KQ/2vlGikh0AUIf9+/fv37+f7CiGYdGiRYsWLSI7CpXx9fX19fUlOwpADtj7yDWh9j44RgcAAAC0AWR0AAAAQBtARgcAAAC0AWR0AAAAQBuQcGfcihUr1N+pZvr111+Rli4Q4oGLSs6aRCKhUqlUKtyn+Wa//vor8QxwlTSlldvbiB09ejQ7O5v4WyaTyWQyDMPIDUkltPi3BSigGxcXp7bO2tvbyRrmqJksLCw06v1LKsTlcp2cnJSsfPv27YcPHxoYGGj3a5FGzMLCwt3d3d3dfZTtKPPU7gnFycmJeC4pQqilpaWwsLClpWXSpEnkRqUSWvzbon5OTk6LFy8eFxsGBcdxsmMAE11DQ8O6deu+//776Ojo2NhY4lHhAKgHjuMpKSk7d+6cN2/eqVOnzM3NyY4IgBGCjA40gvxX1dPT89SpU2+99RbZEYEJoaGhYf369d999x38Nwm0AGR0oEHu3LmzatWq1tbWv//973/5y1/IDgdouevXr3/wwQcMBuPMmTMeHh5khwPAaME/pECDzJw58z//+Y+/v7+Pj8/WrVtfv35NdkRAO0ml0ri4uD//+c/z5s0rKSmBdA60AxyjA02UnZ390UcfmZubZ2Vlubq6kh0O0CoVFRXBwcGPHz9OTU1du3Yt2eEAoDJwjA400YoVK4RCIZ/PnzNnzl//+leywwHa49SpUzNnztTV1S0uLoZ0DrQMZHSgoaysrAoKCnbu3Ll9+/aAgIBXr16RHREY39rb21etWrVu3boNGzb861//sre3JzsiAFQMzroDTffjjz+uXbtWR0cnMzPzD3/4A9nhgHGpqKho1apVXV1d//u//7t48WKywwFgTMAxOtB0CxcuFAqF77zzzh//+Mddu3ZJpVKyIwLjSU9Pz6FDhzw9Pe3t7YVCIaRzoMXgGB2MG6dOndq0aZOrq+vZs2dtbW3JDgeMA0+fPl2zZs2///3vgwcPhoaGUigUsiMCYAzBMToYNz744IPbt2+LxWI3N7esrCyywwGa7p///Of06dMbGxtv3bq1detWSOdA60FGB+OJk5NTUVHRhx9+uHr16g8++KCrq4vsiIAmEovFW7duXbZsmbe39+3bt6dPn052RACoA5x1B+NSbm7uxo0bjYyMzp49+84775AdDtAg5eXlwcHBdXV1f/vb3wIDA8kOBwD1gWN0MC75+fndu3fP0tLS3d390KFDvb29ZEcEyIfj+IkTJ2bNmmVgYHDv3j1I52CigWN0MI4R73cJDw9fsGDBP/7xDzMzM7IjAqRpbGzcsGHDtWvXoqOjd+/eraurS3ZEAKgbZHQw7v373/9etWpVR0dHRkbG0qVLyQ4HkODGjRsffPABlUo9c+bMvHnzyA4HAHLAWXcw7s2ePbu4uPi99977y1/+snXr1u7ubrIjAuojk8ni4uIWLVrk7u4uFAohnYOJDI7RgfY4derU5s2bp0yZcvbs2cmTJ5MdDhhzNTU1q1atEgqFBw4c2Lp1K9nhAEAyOEYH2uODDz4oLS2l0+nTp0+H97tovVOnTrm6unZ3dwuFQkjnACDI6EDL2NjY/Pzzzzt37ty2bduKFStaWlrIjgioXnt7+9q1a4l3rty8eRPOxwBAgLPuQDsRt0rRaLTMzEy4tqpNbt++vWrVqvb2drgREoB+4BgdaCcvLy+hUOjq6rpw4cK4uLienh6yIwKjheP4X//613fffdfa2looFEI6B6AfOEYH2owYsB4REeHu7n769GkLCwuyIwIj9PLlyw8//PCHH36IioqKjY3V0YGjEQD6g70CaDMKhbJ169b//Oc/zc3NLi4u58+fJzsiMBJ5eXnOzs41NTVFRUVxcXGQzgF4I9gxgPZzdnYm3u8SFBQE73cZXyQSydatW/39/ZcuXXrnzh14hj8ACsBZdzCBXLp06b//+79NTEyysrKmTZtGdjhgCPfv3w8ODq6pqTl+/HhwcDDZ4QCg6eAYHUwgy5YtEwqFRkZGc+fO/etf/wr/zmqyU6dOzZo1i8lklpSUQDoHQBmQ0cHEMmnSpB9++CEuLi48PNzf37+5uZnsiEB/ra2tgYGBGzZs+PTTT3/55RdbW1uyIwJgfICz7mCCKioqWrVqVVdX1z/+8Y8///nPZIcD/n8//vjj2rVrdXR0MjMz//CHP5AdDgDjCRyjgwlqzpw5xcXFf/zjH5csWQLvd9EExDtX3nvvvdmzZwuFQkjnAAwXHKODie7UqVOffPKJk5PT2bNn7e3tyQ5ngqqtrV29enVxcTG8cwWAEYNjdDDRffDBB3fu3JHJZG5ubpmZmWSHMxHl5OS88847ra2tRUVFkM4BGDHI6AAgR0fHW7duffLJJx9++OHKlStbW1vJjmii6OjoCAkJWbly5YoVK27fvu3q6kp2RACMY3DWHYD/8/3333/44YcMBuPMmTMeHh5kh6Pl7ty5s2rVqtbW1r///e9/+ctfyA4HgHEPjtEB+D/vvfeeUCh0cnKaP39+XFxcb28v2RFpJ/k7VywtLYVCIaRzAFQCMjoA/w9jY+PLly8fPnz4wIED77333vPnzwfWkUgkd+/eVX9s48vdu3clEsnA8oaGBm9v7x07dkRGRn733XdvvfWW+mMDQCtBRgegP+L9Ljdv3qyvr582bdo333zTr0J4eLiPj09HRwcp4Y0LHR0dPj4+4eHh/cq///77adOmPXjw4Oeff4Z3rgCgWrA7AfBmM2bMEAqFq1at8vX1DQkJEYlERPnly5dTU1Pr6+tDQ0PJjVCThYaG1tfXp6amXr58mSh5/fr1rl27Fi9e/N5775WVlbm7u5MbIQDaB+6MA2AIOTk5H3300VtvvZWVlWVsbOzk5NTa2kpcYs/Ozl6+fDnZAWqcvLw8Pz8/hJCOjg6Xyy0vL29vbw8ODn7y5ElqauqaNWvIDhAA7QQZHYChVVdXr1q1qrS01NnZWSgUSqVS9Hu6evDggampKdkBapDGxkZHR0f5Pz00Gs3FxaWiomLq1Klnz561sbEhO0AAtBZkdACUIpPJgoODL1261PcGeBqNtmDBgm+//ZZCoZAYm0bx9vb+/vvviX96CDo6OsuWLcvKyqJSqSQGBoDWg+voAChFKBTm5ub2G88mlUqvX7/+1VdfkRWVpvnb3/6Wn5/fN50jhHp7e//5z3/euXOHrKgAmCDgGB2AoXV2dk6dOvXp06cymWzgVAzD7t69O3nyZPUHplGePHni4uIiFosHTtLV1X3rrbfu3bvH5XLVHxgAEwQcowMwtJCQkNra2jemc4SQTCZbvXp1T0+PmqPSKD09PUFBQYMtop6enmfPnm3atEnNUQEwoUBGB2AINTU1Dx48wHFcV1dXV1d3YAWZTFZSUnLw4EH1x6Y5Dh48WFxc3O98O4FYbjiOP3jwoKamRu2hATBRwFl3AJTS2NiYn5+fl5d37do1kUhEo9H6ZS9dXd2bN2/Onj2brAhJVFxcPGfOnH4H6MQiotPp7777rq+v7/Lly83NzcmKEICJADI6AMMjkUgKCgq++eabvLy8Z8+e0el0mUzW29urq6trY2NTWlrKZDLJjlGtxGLx1KlTa2pqZDIZhULR1dWVyWRmZmb+/v7vv//+ggULMAwjO0YAJgTI6BrtwoULZIcAFKmrqysuLv73v//95MkTYldavHjx+vXryY5LrTIyMq5du4YQolAoNjY2c+bMcXNzs7S0JDsuoMjKlSvJDgGoHmR0jQajnAEAYwF++bUS3Bmn6c6fP4+DAc6fP48QIjuKN+ju7i4vLx9lI+NovZeXl3d3d5MdBRgGYt8BWgke4QSAKtFoNCcnJ7KjUJ8JNbMAaDg4RgcAAAC0AWR0AAAAQBtARgcAAAC0AWR0AAAAQBtARgcAAAC0AWT08e3w4cPGxsYUCuX48eOqanPWrFm6urrTp08fsubGjRs5HA6FQhEKhaPvVyKRODo6xsTEjL6pwVy9epXH433zzTdj14XafPzxx5TfrVmzpu+k69evR0ZGXrx40dbWlqiwdu3avhUWLVrE4XB0dXWdnZ2Li4vVGziSSqX79++3t7en0+l8Pt/FxWXgw977bgxff/31oUOHRvYiHK1fFLm5ufLNwNDQUJ3BAw0EGX1827Fjx82bN1Xb5u3btxcuXKhMzZMnT6rw1eDR0dEVFRWqau2NcO16qoa+vn5+fn5FRUV6erq8cM+ePSkpKVFRUQEBAU+ePLGzszMwMMjMzLxy5Yq8znfffZednf3++++Xl5e7ubmpOezAwMBTp06dOXNGJBI9ePDAzs6us7OzX52+G4OPjw+GYV5eXq2trcPqaCIsCl9f3/r6+p9//nnp0qVqDR1oJMjo4M3U/Li6mzdv3rt3b6x78fb2bmtre//998eofbFY7OHhMUaND8RkMhcvXjx58mQGg0GUHDx48Ny5cxcuXOBwOPJqKSkpOjo6ISEhbW1taottMOfOncvNzc3Ozp4zZw6VSjUzM8vLy3NxcelbZ+DGsHXr1mnTpi1dunSwt7UONEEWBYVCMTc39/T0dHBwUGv0QCNBRgdvRqPRlKmmksQvFovDw8OTk5NH3xS50tPTGxoayOr98ePHu3fv3rt3b783o3h4eISFhT179mzHjh1kxSb35Zdfurm5ubq6DlZhsI0hLi5OKBQquZHAogATE2R0bfPLL784OTnxeDwMw1xdXb/99luEUHJyMpvN1tHRmTFjhomJCY1GY7PZbm5unp6ekyZNwjCMz+fv3LmzbzuPHz92dHRks9lMJtPT07OwsJAox3E8KSlpypQpDAaDx+OFh4cr7loZ0dHRmzdvNjIyUtEyeLPCwkJLS0sKhXLs2DGEUFpaGpvNZrFYeXl5S5Ys4XK5FhYWWVlZCKGUlBQMw4yNjT/++GMzMzMMwzw8PIqKihBCoaGhdDrd1NSUaHPz5s1sNptCoTQ1NYWFhW3fvr2qqopCodjb2yOErl27xuVyExMTx3S+5FJSUnAc9/HxGTgpISFh8uTJJ0+evH79+sCpOI4fOXLk7bffZjAYAoHAz8/v4cOHSOEiQgj19PTExsZaWloymcypU6cq82zR7u7uX3/9VfEtGoNtDAKBYP78+cnJycpcOoFFASYo8p4uDIaGlHi+d2VlJULoyy+/JD5mZ2fHxcW9evWqubl57ty5BgYGRPmePXsQQkVFRV1dXU1NTYsXL0YIXblypbGxsaurKzQ0FCEkFAqJyl5eXra2ttXV1VKp9N69e3PmzMEw7NGjRziOR0dHUyiUzz//vKWlRSQSpaamIoRKSkoUdK1YYWGhj48PjuONjY0IoejoaGW+NbLnuj99+hQh9MUXXxAfo6OjEUI3btxoa2traGjw9PRks9nEU8pDQkLYbPb9+/clEkl5efmsWbM4HE5dXR2O46tXrzYxMZG3mZSUhBBqbGzEcTwgIMDOzk4+6fLlyxwOJz4+frhxKrPeQ0JCzM3N+5bY2to6OTn1q2ZnZ1ddXY3j+M2bN3V0dKytrTs7O3Ecz8/P9/X1JerExsbS6fTTp0+3traWlpa6ubkZGhq+ePFC8SLasWMHg8HIyclpaWmJiorS0dG5ffu24pirq6sRQtOnT1+wYIGpqSmDwXB0dDx27Fhvby9RQfHGEBkZKd/YFJuAi2Lr1q1K7nEa+04EMHpwjK5tli9fvmfPHoFAoK+v7+Pj09zcTPwiEJycnFgsloGBQXBwMELI0tLS0NCQxWIRN0sThyMEDodjbW1NpVKdnZ2/+uoriURy4sQJsVh89OjRP/3pT9u2bePz+UwmU19fX8mu30gsFoeFhaWlpal4KQyHh4cHl8s1MjIKCgrq6uqqq6sjyqlUKnGs5uTklJaW1tHRkZGRMayWvb2929vbd+/ePQZR99fV1VVdXW1nZzdYBXd3988++6ympmbXrl19y8Vi8ZEjR5YtW7ZmzRoej+fq6nr8+PGmpqYTJ07I6wxcRBKJJC0tzd/fPyAggM/nx8TE0Gi0IZcPcduXkZFRYmJieXn5y5cv/fz8tmzZcvbsWaTExkBcKi4rK4NFoeSiABMNZHRtRlwLf+OwHzqdjhCS32dE1JRKpW9sx9XVlcfjlZaWPn78WCQSeXl5jabrvqKioj766CNzc/MhG1QDYpm8cSHMnDmTxWL1/Y9H0zQ0NOA4zmKxFNRJSEiYMmVKamqq/BoKQqi8vLyzs3PmzJnyklmzZtHpdOIqQz/yRVRRUSESieS3cTGZTFNT0yGXD3EHn7Ozs4eHh76+Po/H27t3L4/HI1LmkBsDMXcvX75U3AssCjBhQUbXNleuXFmwYIGRkRGDweh3aXw0aDSaVCqtr69HCA12wXu4XRcWFpaVlW3cuFFVQY4pBoMx5CkHEkkkEvR7nhgMhmEZGRkUCmXDhg1isZgoJMZB6enp9a3J5/M7OjoUNNXV1YUQiomJkQ+Grq2tFYlEioM0MzNDCDU1NclL6HS6lZVVVVWVMhsDk8lEv8+pArAowIQFGV2r1NXV+fv7m5qaFhUVtbW1HTp0SCXNymSyV69eWVpaEncOv379WiVdp6en37hxQ0dHh/gdJP5RSExMpFAod+7cUUnkqiKVSltbWy0sLMgOZFDET/yQJ0Xc3d23bdtWWVm5b98+ooTP5yOE+iWtIWeWWFlHjx7tew3v1q1binvX09NzcHC4f/9+30KZTMbj8ZTZGLq7u+VzqgAsCjBhQUbXKmVlZVKp9JNPPrG1tcUwTFVjyn/88cfe3l43NzcXFxcdHZ2ffvpJJV1nZGT0/RHsewdQ3zOfmqCgoADH8blz5yKEqFTqYJcnSEQ8OlCZYdb79u1zdHQsKSkhPrq4uOjp6fX9F6qoqKi7u3vGjBkKGiGGSIzgWYGBgYElJSVPnjwhPopEotraWldXV2U2BmLuTExMFHcBiwJMWJDRtYqlpSVC6Pr16xKJpLKy8o3X/5TU3d3d1tYmk8mKi4tDQ0OtrKzWrVtnZGQUEBCQk5OTnp7e3t5eWloqv2lIhV1riN7e3paWFplMVlpaGhYWZmlpuW7dOoSQvb39q1evcnNzpVJpY2NjbW2t/Cv6+vrPnz+vqanp6OiQSqX5+flqG73GYrFsbW2JyyKKESecdXV15R+3b99+6dKlzMzM9vb2srKyTZs2mZmZhYSEKG5k/fr1WVlZaWlp7e3tPT099fX1v/32G0IoKCjIxMRksCeqbtu2jdiW6urqmpubIyIixGJxvzvUBkPMHTGAW0EvE21RAPB/VHTPPBgTaKhRTJ9//jnxfzqbzV62bBmO4xEREfr6+nw+f8WKFcTAazs7u+3btxO30lhbW//yyy8HDx7k8XgIIRMTkzNnzpw7d45oRCAQZGVl4TiekZGxcOFCY2NjKpVK3BhfW1tL9NjR0bFx40YDAwM9Pb158+bFxsYihCwsLO7evfvGrokRX8oY69FrX3zxBTGOnMVi+fj4pKamEsvEwcGhqqrqxIkTXC4XIWRlZfXo0aOQkBAajWZubk6lUrlcrp+fX1VVFdFOc3PzwoULMQyzsbH59NNPiRH59vb2dXV1xcXFVlZWTCZz3rx5L168uHr1KofDSUhIGFac+EhHr4WGhtJoNJFIRHy8dOkScb+3oaHhli1b+n09PDxcPmSrt7c3KSnJwcGBRqMJBAJ/f/+KigocxxUvotevX0dERFhaWlKpVOJfvfLychzH/f39EUKxsbGDRf706dPg4GCBQMBgMGbPnp2fnz+wzhs3Bm9vb3Nzc2J8l+JeJtSiIMDoNYDjOKxXjabML/vENNa/SiEhIfr6+mPXvmIjy+iVlZVUKvX06dNjGdrQenp6PD0909PTVdtsU1MThmGHDx9WppcJtSgIkNEBDuPRARjMyF72pU5isfjbb7+trKwk7pOyt7ePj4+Pj48f+LYPtenp6cnNze3o6AgKClJty3FxcdOnTyeiOlA0AAAgAElEQVQehTRkLxNnUeA4/vz588LCwsePH6u2FzAeQUYHY+jhw4eUwan8l26iefXqFfGmlg0bNhAlkZGRK1asCAoKIutNJAUFBRcvXszPz1c8HHy4jhw5IhQKr169SjznQJleJsiiyMvLI97U0veFcmDiIvskAVAEwVn3QYzpmcPIyEji+SHW1tbZ2dlj1IsCo1zv3377bUREhArjIVdubu7+/ftlMtkIvguLYiA4667FKDg861+DUSiU8+fPr1y5kuxANM6FCxcCAwO1deuF9Q7GjnbvOxMcnHUHAAAAtAFkdAAAAEAbQEYHAAAAtAFkdAAAAEAbwJ1xGo1CocydO1eTXxBClvr6+l9//XX58uVkBzImcnJyYL2DMULsO/DLr5XgGB0AAADQBnCMrtFgFNNgtHsEDqx3MHa0e9+Z4OAYHQAAANAGkNEBAAAAbQAZHQAAANAGkNEBAAAAbQAZHQAAANAGkNG1zcWLF21tbYnXlZqamq5Zs2Ysepk1a5auru706dOHrLlx40YOh0OhUIRC4VhEAjTW9evXIyMj+26Qa9eu7Vth0aJFHA5HV1fX2dm5uLhYzeFJpdL9+/fb29vT6XQ+n+/i4lJTU9OvjkQicXR0jImJQQh9/fXXhw4d6unpUXOcACgPMrq2CQgIePLkiZ2dHY/He/HiRWZm5lj0cvv27YULFypT8+TJk1999dVYxAA02Z49e1JSUqKiouQbpIGBQWZmZt/XeH/33XfZ2dnvv/9+eXm5m5ubmiMMDAw8derUmTNnRCLRgwcP7OzsOjs7+9WJjo6uqKgg/vbx8cEwzMvLq7W1Vc2hAqAkyOhg5CgUCtkhjAmxWOzh4aEJjYxTBw8ePHfu3IULFzgcjrwwJSVFR0cnJCSkra2NxNgI586dy83Nzc7OnjNnDpVKNTMzy8vLc3Fx6Vvn5s2b9+7d61uydevWadOmLV26VCaTqTdeAJQCGR2MHI1GU6bauEv86enpDQ0NmtDIePT48ePdu3fv3bsXw7C+5R4eHmFhYc+ePduxYwdZscl9+eWXbm5urq6ug1UQi8Xh4eHJycn9yuPi4oRC4cByADQBZPSJ6JdffnFycuLxeBiGubq6fvvttwih5ORkNputo6MzY8YMExMTGo3GZrPd3Nw8PT0nTZqEYRifz9+5c2ffdh4/fuzo6Mhms5lMpqenZ2FhIVGO43hSUtKUKVMYDAaPxwsPD1fc9djBcfzIkSNvv/02g8EQCAR+fn4PHz5ECIWGhtLpdFNTU6La5s2b2Ww2hUJpamoKCwvbvn17VVUVhUKxt7dPSUnBMMzY2Pjjjz82MzPDMMzDw6OoqGhYjSCErl27xuVyExMTx3R+NUFKSgqO4z4+PgMnJSQkTJ48+eTJk9evXx84dbCVlZaWxmazWSxWXl7ekiVLuFyuhYVFVlYW8a2enp7Y2FhLS0smkzl16tTz588PGWF3d/evv/6q+C6Q6OjozZs3GxkZ9SsXCATz589PTk6GZ64BTYQDDYYQOn/+/Ai+SFxHH2xqdnZ2XFzcq1evmpub586da2BgQJTv2bMHIVRUVNTV1dXU1LR48WKE0JUrVxobG7u6ukJDQxFCQqGQqOzl5WVra1tdXS2VSu/duzdnzhwMwx49eoTjeHR0NIVC+fzzz1taWkQiUWpqKkKopKREQdfDRfxwD1ktNjaWTqefPn26tbW1tLTUzc3N0NDwxYsXOI6vXr3axMREXjMpKQkh1NjYiON4QECAnZ2dfFJISAibzb5//75EIikvL581axaHw6mrqxtWI5cvX+ZwOPHx8crM3YjXuyawtbV1cnLqV2hnZ1ddXY3j+M2bN3V0dKytrTs7O3Ecz8/P9/X1JeooWFnR0dEIoRs3brS1tTU0NHh6erLZ7O7ubhzHd+zYwWAwcnJyWlpaoqKidHR0bt++rTjC6upqhND06dMXLFhgamrKYDAcHR2PHTvW29tLVCgsLPTx8cFxvLGxESEUHR3d9+uRkZHy7Xk8UnLfAeMRHKNPRMuXL9+zZ49AINDX1/fx8WlubiZ+uQhOTk4sFsvAwCA4OBghZGlpaWhoyGKxiNvmicMmAofDsba2plKpzs7OX331lUQiOXHihFgsPnr06J/+9Kdt27bx+Xwmk6mvr69k16olFouPHDmybNmyNWvW8Hg8V1fX48ePNzU1nThxYrhNUalU4tjRyckpLS2to6MjIyNjWC14e3u3t7fv3r17uF2PL11dXdXV1XZ2doNVcHd3/+yzz2pqanbt2tW3XJmV5eHhweVyjYyMgoKCurq66urqJBJJWlqav79/QEAAn8+PiYmh0WhDrhriDjgjI6PExMTy8vKXL1/6+flt2bLl7NmzRCRhYWFpaWmDfd3BwQEhVFZWptwiAUB9IKNPdMS18DeOyaHT6Qgh+U1ARE2pVPrGdlxdXXk8Xmlp6ePHj0UikZeX12i6Vony8vLOzs6ZM2fKS2bNmkWn04lz5iM2c+ZMFovV9z8bINfQ0IDjOIvFUlAnISFhypQpqamp8ss0aJgri9gypVJpRUWFSCSS39HGZDJNTU2HXDUMBgMh5Ozs7OHhoa+vz+Px9u7dy+PxiP8eoqKiPvroI3Nz88G+Tszdy5cvFfcCgPpBRp+Irly5smDBAiMjIwaD0e/S+GjQaDSpVFpfX48QGngBcky7fiNilJGenl7fQj6f39HRMcqWGQzG2J1aGNckEgn6PWUOBsOwjIwMCoWyYcMGsVhMFI5sZXV1dSGEYmJiKL+rra0ViUSKgzQzM0MINTU1yUvodLqVlVVVVVVhYWFZWdnGjRsVfJ3JZKLf5xQAjQIZfQL5+eefjx49WldX5+/vb2pqWlRU1NbWdujQIZU0LpPJXr16ZWlpSdzh/Pr164F1xqjrwfD5fIRQv5TQ2tpqYWExmmalUunoG9FWRLYb8ryLu7v7tm3bKisr9+3bR5SMbGUR/zgePXq076XEW7duKe5dT0/PwcHh/v37fQtlMhmPx0tPT79x44aOjg7x/wHRfmJiIoVCuXPnDlGzu7tbPqcAaBTI6BPIf/7zHzabXVZWJpVKP/nkE1tbWwzDVDW07Mcff+zt7XVzc3NxcdHR0fnpp58G1hmjrgfj4uKip6cn/yFGCBUVFXV3d8+YMQMhRKVSB7uCoFhBQQGO43Pnzh1NI9rK2NiYQqEoM+J83759jo6OJSUlxEfFK2swxCiMETyOMDAwsKSk5MmTJ8RHkUhUW1vr6uqakZHR95+DvnfGya8IEHNnYmIy3E4BGGuQ0ScEqVT68uXLgoICNpttaWmJELp+/bpEIqmsrBzNReXu7u62tjaZTFZcXBwaGmplZbVu3TojI6OAgICcnJz09PT29vbS0lL5zU0q7FoZGIZt37790qVLmZmZ7e3tZWVlmzZtMjMzCwkJQQjZ29u/evUqNzdXKpU2NjbW1tbKv6ivr//8+fOampqOjg4iYff29ra0tMhkstLS0rCwMEtLy3Xr1g2rkfz8/Ikweo3FYtna2hJXXhQjzr3r6urKPypYWQoaWb9+fVZWVlpaWnt7e09PT319/W+//YYQCgoKMjExGezhstu2bSM217q6uubm5oiICLFY3O9mvcEQc6dgLDsApFHfbfVg+NDwRzFdunRJwZ3Gly5dwnE8IiJCX1+fz+evWLHi2LFjCCE7O7vt27cTt/xYW1v/8ssvBw8e5PF4CCETE5MzZ86cO3eOOCgRCARZWVk4jmdkZCxcuNDY2JhKpRI3xtfW1hIxdHR0bNy40cDAQE9Pb968ebGxsQghCwuLu3fvvrFrYiTYsCg5Aqe3tzcpKcnBwYFGowkEAn9//4qKCmJSc3PzwoULMQyzsbH59NNPiUHz9vb2dXV1xcXFVlZWTCZz3rx5L168CAkJodFo5ubmVCqVy+X6+flVVVUNt5GrV69yOJyEhARl5m4E611zhIaG0mg0kUhEfJRvkIaGhlu2bOlXOTw8XD56bbCVlZqaSmyZDg4OVVVVJ06c4HK5CCErK6tHjx69fv06IiLC0tKSSqUS/02Wl5fjOO7v748Qio2NHSzOp0+fBgcHCwQCBoMxe/bs/Pz8gXXeOHrN29vb3NxcPtRt3IHRa1oM1qtGG9e/7GNKnb9KISEh+vr66umLMK7Xe2VlJZVKPX36NLlh9PT0eHp6pqenq7bZpqYmDMMOHz6s2mbVCTK6FoOz7gAMDd64pTx7e/v4+Pj4+PiBLz5Rm56entzc3I6OjqCgINW2HBcXN336dOJpSwBoGsjoAAAVi4yMXLFiRVBQEFkvZSkoKLh48WJ+fr7ikfHDdeTIEaFQePXqVSXfaACAmkFGB0CRqKiojIyMtrY2GxubnJwcssMZNxITE0NDQw8cOEBK715eXmfOnJE/cl8l8vLyXr9+XVBQIBAIVNgsACpEJTsAADTa/v379+/fT3YU49KiRYsWLVpEdhQq4+vr6+vrS3YUACgCx+gAAACANoCMDgAAAGgDyOgAAACANoCMDgAAAGgDyOgAAACANqDgOE52DGBQY/0uEwDAxAS//FoJRq9pNOJ5jQAM6ejRowihzz77jOxAAACkgWN0ALTBypUrEUIXLlwgOxAAAGngOjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANqASnYAAICRaGpqam9vl3/s6upCCD158kRewuVyDQ0NSYgMAEASCo7jZMcAABi2jIyMDRs2KKjw97//ff369WqLBwBAOsjoAIxLbW1tRkZGUqn0jVNpNFpjYyOPx1NzVAAAEsF1dADGJR6Pt3TpUir1DRfOqFSqt7c3pHMAJhrI6ACMV2vWrOnp6RlY3tvbu2bNGvXHAwAgF5x1B2C8kkgkhoaGxD1xfbFYrKamJiaTSUpUAACywDE6AOMVhmHLli2j0Wh9C2k02vLlyyGdAzABQUYHYBxbtWpVv5vjpFLpqlWryIoHAEAiOOsOwDgmk8lMTExevXolL+Hz+Y2NjW+8Yw4AoN3gGB2AcYxKpQYHB8tPvNNotDVr1kA6B2BigowOwPgWHBwsP/EulUqDg4PJjQcAQBY46w7A+Ibj+KRJk549e4YQMjMze/bsGYVCITsoAAAJ4BgdgPGNQqGsXbuWTqfT6fQPP/wQ0jkAExYcowMw7pWWlk6bNo34w9XVlexwAADkgDtoJpAVK1aQHQIYK3p6egih+Ph4sgMBYyU7O5vsEICmg7PuE0hOTk59fT3ZUWi6+vr6nJwcsqMYNisrK2tr6yGrwTYwHo3TbRKoH5x1n0AoFMr58+dXrlxJdiAa7cKFC4GBgeNuvyDejG5ra6u4GmwD49E43SaB+sFZdwC0wZC5HACg9eCsOwAAAKANIKMDAAAA2gAyOgAAAKANIKMDAAAA2gAyOlCTjRs3cjgcCoUiFAoHq3P48GFjY2MKhXL8+HElm+3t7T169KiHh8ewJqnc1atXeTzeN998o4a+xpfr169HRkZevHjR1taWQqEQT7jrW2HRokUcDkdXV9fZ2bm4uFjN4Uml0v3799vb29PpdD6f7+LiUlNT06+ORCJxdHSMiYlBCH399deHDh3q6elRc5wAKAMyOlCTkydPfvXVV4rr7Nix4+bNm8q3WVlZ+Yc//GHbtm0ikUj5SWMBRha90Z49e1JSUqKiogICAp48eWJnZ2dgYJCZmXnlyhV5ne+++y47O/v9998vLy93c3NTc4SBgYGnTp06c+aMSCR68OCBnZ1dZ2dnvzrR0dEVFRXE3z4+PhiGeXl5tba2qjlUAIYEo9fAeHX37t34+PhNmzZ1dXX1S6gKJo0Rb2/vtra2sWtfLBZ7eXkN698d0h08ePDcuXN3797FMExemJKSsnbt2pCQkPLych6PR2J4CKFz587l5ubevXuXeHSumZlZXl5evzo3b968d+9e35KtW7c+efJk6dKlP//8M7y4FmgUOEYH6qPal4hMmzbt4sWLq1evZjAYyk8ap9LT0xsaGsiOYhgeP368e/fuvXv39k3nCCEPD4+wsLBnz57t2LGDrNjkvvzySzc3NwVPwheLxeHh4cnJyf3K4+LihELhwHIAyAUZHfSXnJzMZrN1dHRmzJhhYmJCo9HYbLabm5unp+ekSZMwDOPz+Tt37pTXx3H8yJEjb7/9NoPBEAgEfn5+Dx8+lE9KSkqaMmUKg8Hg8Xjh4eHyb/X09MTGxlpaWjKZzKlTp54/f17d86k6hYWFlpaWFArl2LFjCKG0tDQ2m81isfLy8pYsWcLlci0sLLKyshBCKSkpGIYZGxt//PHHZmZmGIZ5eHgUFRUhhEJDQ+l0uqmpKdHm5s2b2Ww2hUJpamoKCwvbvn17VVUVhUKxt7dHCF27do3L5SYmJpI300NISUnBcdzHx2fgpISEhMmTJ588efL69esDpw62OSlYqmhEm1N3d/evv/46ffp0BXWio6M3b95sZGTUr1wgEMyfPz85ORmutgDNgoMJAyF0/vx5ZWru2bMHIVRUVNTV1dXU1LR48WKE0JUrVxobG7u6ukJDQxFCQqGQqBwbG0un00+fPt3a2lpaWurm5mZoaPjixQscx6OjoykUyueff97S0iISiVJTUxFCJSUlOI7v2LGDwWDk5OS0tLRERUXp6Ojcvn0bx/HKykqE0Jdffqn8fM2ZM2fatGnDnTQYIhkM6ys4jj99+hQh9MUXXxAfo6OjEUI3btxoa2traGjw9PRks9nd3d04joeEhLDZ7Pv370skkvLy8lmzZnE4nLq6OhzHV69ebWJiIm8zKSkJIdTY2IjjeEBAgJ2dnXzS5cuXORxOfHz8cONUfhsYJVtbWycnp36FdnZ21dXVOI7fvHlTR0fH2tq6s7MTx/H8/HxfX1+ijuLNabClOtjmpEB1dTVCaPr06QsWLDA1NWUwGI6OjseOHevt7SUqFBYW+vj44Dje2NiIEIqOju779cjISPnGPNZGtk2CCQiO0cGgnJycWCyWgYFBcHAwQsjS0tLQ0JDFYq1ZswYhRBw5icXiI0eOLFu2bM2aNTwez9XV9fjx401NTSdOnBCLxUePHv3Tn/60bds2Pp/PZDL19fWJliUSSVpamr+/f0BAAJ/Pj4mJodFoGRkZJM7sWPDw8OByuUZGRkFBQV1dXXV1dUQ5lUoljkGdnJzS0tI6OjqGO+/e3t7t7e27d+8eg6hVoKurq7q62s7ObrAK7u7un332WU1Nza5du/qWK9ic5HUGLtWRbU7EHXBGRkaJiYnl5eUvX7708/PbsmXL2bNniUjCwsLS0tIG+7qDgwNCqKysTLlFAoA6QEYHQ6PT6QghmUxGfKTRaAghqVSKECovL+/s7Jw5c6a88qxZs+h0elFR0ePHj0UikZeX18AGKyoqRCKRi4sL8ZHJZJqamsrP1WsfYgESS6yfmTNnslgsLZv3hoYGHMdZLJaCOgkJCVOmTElNTS0sLJQXKticBrYgX6oj25yIeyycnZ09PDz09fV5PN7evXt5PB7x30NUVNRHH31kbm4+2NeJuXv58qXiXgBQJ8joYFSIMTzEy7nl+Hx+R0cH8dbOgdcgEUJdXV0IoZiYGMrvamtr1TPMTAMxGAzivK7WkEgk6PeUORgMwzIyMigUyoYNG8RiMVGoYHNS0NTINiczMzOEUFNTk7yETqdbWVlVVVUVFhaWlZVt3LhRwdeZTCb6fU4B0BCQ0cGo8Pl8hFC/H9zW1lYLCwviJufXr18P/BaR5o8ePdr3CtCtW7fUErJmkUqlxOIiOxBVIrLdkI9hcXd337ZtW2Vl5b59+4gSBZuTgnZGtjnp6ek5ODjcv3+/b6FMJuPxeOnp6Tdu3NDR0SH+PyDaT0xMpFAod+7cIWp2d3fL5xQADQEZHYyKi4uLnp6e/GcOIVRUVNTd3T1jxgwXFxcdHZ2ffvpp4LeIe+YVPDxu4igoKMBxfO7cuQghKpX6xjPz4w7x4D9lBujv27fP0dGxpKSE+Khgc1LQyIg3p8DAwJKSEuLV8gghkUhUW1vr6uqakZHR95+DvnfGya8IEHNnYmIy3E4BGDuQ0cGoYBi2ffv2S5cuZWZmtre3l5WVbdq0yczMLCQkxMjIKCAgICcnJz09vb29vbS0VH5/E4Zh69evz8rKSktLa29v7+npqa+v/+2338idF7Xp7e1taWmRyWSlpaVhYWGWlpbr1q1DCNnb27969So3N1cqlTY2NtbW1sq/oq+v//z585qamo6ODqlUmp+fr8mj11gslq2tLXHZRTHi3Luurq7842Cbk+JGBtucgoKCTExMBnu47LZt26ysrNatW1dXV9fc3BwRESEWi/vdrDcYYu4UjGUHgARjfTM90BxIuZFLycnJxF0/1tbWv/zyy8GDB4lne5mYmJw5c+bcuXPEcYlAIMjKysJxvLe3NykpycHBgUajCQQCf3//iooKoqmOjo6NGzcaGBjo6enNmzcvNjYWIWRhYXH37t3Xr19HRERYWlpSqVQi95eXl3/++edE42w2e9myZYrjvHXr1rvvvktcDUUImZqaenh4/PTTT4onDWkEI4W++OILYhw5i8Xy8fFJTU0lFqCDg0NVVdWJEye4XC5CyMrK6tGjRyEhITQazdzcnEqlcrlcPz+/qqoqop3m5uaFCxdiGGZjY/Ppp58Sw/ft7e3r6uqKi4utrKyYTOa8efNevHhx9epVDoeTkJAwrDhxNY5eCw0NpdFoIpGI+Hjp0iXi1ndDQ8MtW7b0qxweHi4fvTbY5qR4qb5xc8Jx3N/fHyEUGxs7WJxPnz4NDg4WCAQMBmP27Nn5+fkD67xx9Jq3t7e5ubl8qNuYgtFrQEmwlUwgavs1H9fG+tczJCREX19/7NpXTG3bQGVlJZVKPX36tBr6UqCnp8fT0zM9PV21zTY1NWEYdvjwYdU2OxjI6EBJcNYdAHWbCG/usre3j4+Pj4+PH/jiE7Xp6enJzc3t6OgICgpSbctxcXHTp08nHrUEgOaAjA401MOHDymDU/lvNFC5yMjIFStWBAUFjek7bBQoKCi4ePFifn6+4pHxw3XkyBGhUHj16lXiwQwAaA7I6EBDOTo6Kji5dO7cObIDHImoqKiMjIy2tjYbG5ucnByywxlziYmJoaGhBw4cIKV3Ly+vM2fOyB+VrxJ5eXmvX78uKCgQCAQqbBYAlYBXAQKgPvv379+/fz/ZUajVokWLFi1aRHYUKuPr6+vr60t2FAC8GRyjAwAAANoAMjoAAACgDSCjAwAAANoAMjoAAACgDSCjAwAAANqAguM42TEANaFQKGSHAAAYIfitBkOC0WsTS1hYmLu7O9lRaLRbt24lJycTz93UPoGBgbANjDvENkl2FGAcgIw+sbi7u69cuZLsKDRdcnKyti6lwMBA2AbGI8joQBlwHR0AAADQBpDRAQAAAG0AGR0AAADQBpDRAQAAAG0AGR0AAADQBpDRwf+5ePGira3tG99Hbm1tPda9X716lcfjffPNNypv+fDhw8bGxhQK5fjx4ypvXItdv349MjKy71axdu3avhUWLVrE4XB0dXWdnZ2Li4vVGVt8fLyTkxOXy2UwGPb29jt37uzs7JRPLSwsfPfdd1kslpmZWURExOvXr5WZpCSJROLo6BgTEzPi7r7++utDhw719PSMcOYBGIyCV1ADLYMQOn/+/JDV7OzseDwe8bdMJhOJRC9fvnz77bfHODr88uXLXC7366+/HovGKysrEUJffvnlkDWJkehjEYMmUHIbwHE8Njb2/fffb29vJz7a2dkZGBgghC5fvty3Wn5+vq+vr+oDHcr8+fNTU1Obm5vb29vPnz9Po9EWL15MTLp37x6Tydy9e3dnZ+fNmzcNDQ3Xr18/5CTlbdu2DSEUHR09mu6Sk5Pnz5/f0tKiTI/avU0CFYKtZAIZQUaXG4tfbZFI5O7urvJm30hDMrpKZnk0jSi5DRw4cGDy5MlisVheYmdnd+bMGR0dHXNz89bWVnk5WRnd29tbJpPJPxIj7Ovq6nAcDwwMtLGx6e3tJSYlJSVRKJQHDx4onqSkf/3rX8Tr3uUZfcTdhYaGuru7S6XSITuFjA6UBGfdgVJyc3NV3mZ6enpDQ4PKm9VkKpnlsV5ujx8/3r179969ezEM61vu4eERFhb27NmzHTt2jF3vSrp8+bKurq78o6GhIUJIJBLJZLIrV67Mnz9f/szjJUuW4Diel5enYJKSnYrF4vDw8L4PexlNd3FxcUKhEB4dA1QIMjpQVmhoKJ1ONzU1JT5u3ryZzWZTKJSmpqa0tDQ2m81isfLy8pYsWcLlci0sLLKysuTfPX369MyZMzEMY7PZ1tbW+/btCwsL2759e1VVFYVCsbe3LywstLS0pFAox44dI76C4/iRI0fefvttBoMhEAj8/PwePnyIEFLc1y+//OLk5MTj8TAMc3V1/fbbb8doaQwWnoKl1G+WU1JSMAwzNjb++OOPzczMMAzz8PAoKioaViMIoWvXrnG53MTERFXNWkpKCo7jPj4+AyclJCRMnjz55MmT169fV36ZKF5lPT09sbGxlpaWTCZz6tSpI3v+7rNnz5hMpo2NzZMnTzo7Oy0tLeWT7OzsEEKlpaUKJinZS3R09ObNm42MjOQlo+lOIBDMnz8/OTkZhwe2A1Uh7ewAUDs0orPuN27cSEpKIv5evXq1iYmJfFJSUhJCqKAWmd4AAAkDSURBVLGxEcfx6OhohNCNGzfa2toaGho8PT3ZbHZ3dzeO40ePHkUIHThwoLm5+dWrV3/7299Wr16N43hAQICdnZ28tadPnyKEvvjiC+JjbGwsnU4/ffp0a2traWmpm5uboaHhixcvFPeVnZ0dFxf36tWr5ubmuXPnGhgYEK2p/Ky7gvAULKV+sxwSEsJms+/fvy+RSMrLy2fNmsXhcIhTx8o3cvnyZQ6HEx8fP2TMuHLbgK2trZOTU79COzu76upqHMdv3rypo6NjbW3d2dmJ/79n3Ue2ynbs2MFgMHJyclpaWqKionR0dG7fvq3MvMh1dXVxOJzQ0FAcx3/66SeEkHyLJTCZTC8vLwWTlOmlsLDQx8cHx/HGxkb0+1n3UXYXGRmJECopKVHcNZx1B0qCY3TwBm1tbfK73L28vJT/ooeHB5fLNTIyCgoK6urqqqurk0qle/fuXbhw4a5du/T19QUCwX/913/NmjVLcTtisfjIkSPLli1bs2YNj8dzdXU9fvx4U1PTiRMnFPSFEFq+fPmePXsEAoG+vr6Pj09zczPx+6tayoSnJCqVShzUOjk5paWldXR0ZGRkDKsFb2/v9vb23bt3D7frN+rq6qquriaOJt/I3d39s88+q6mp2bVrV9/yka0yiUSSlpbm7+8fEBDA5/NjYmJoNNpwl8D+/fvNzMwSEhIQQsTN5H1PyCOEaDSaWCxWMGnILsRicVhYWFpaWr/yUXbn4OCAECorKxsyAACUARkdvEHfY/Qff/xxBC3Q6XSEkFQqLS0tbW1t/fOf/yyfpKuru3XrVsVfLy8v7+zsnDlzprxk1qxZdDqdOCk9WF/9ymk0GkJoLMYIDSs85c2cOZPFYhFnqsnS0NCA4ziLxVJQJyEhYcqUKampqYWFhfLCka2yiooKkUjk4uJClDOZTFNT02EtgUuXLl24cOHbb7/lcDgIIeLav0wm61unu7ubyWQqmDRkL1FRUR999JG5uXm/8lF2Ryznly9fDhkAAMqAjA6GsGDBgtHcCdXe3o4Q4vP5w/pWa2srQkhPT69vIZ/P7+joUPzFK1euLFiwwMjIiMFg7Ny5c5jBjnl4Q2IwGGNxUkF5EomECENBHQzDMjIyKBTKhg0b5EecI1smXV1dCKGYmBj5OaHa2lqRSKRktOfOnTt48GBBQYH8eQnEzQfEVkcQiUQSicTMzEzBJMW9FBYWlpWVbdy4ceCkUXZHZHdimQMwepDRwdh66623EEJNTU3D+hbxH0C/ZNDa2mphYaHgW3V1df7+/qampkVFRW1tbYcOHRp+vGMY3pCkUunoGxklIscMeWLD3d1927ZtlZWV+/btI0pGtkyIu8yOHj3a91rgrVu3lAn1iy++yMzM/OGHH4htjGBjY8PhcGpra+Uljx8/RghNnTpVwSTFHaWnp9+4cUNHR4f4n4OIOTExkUKhNDc3j6a77u5u9PsyB2D0IKODYaBSqQNPbitmbW2tr6//3XffDetbLi4uenp6d+7ckZcUFRV1d3fPmDFDwbfKysqkUuknn3xia2uLYZh81JDKKQ5vBEuJUFBQgOP43LlzR9PIKBEP12traxuy5r59+xwdHUtKSoiPI1tlkyZNwjBMKBQOK0gcxyMiIsrKynJzc/udFaBSqUuXLv355597e3uJkvz8fAqF4uPjo2CS4u4yMjL6/sPR9864uXPnjqY7YjmbmJgMa/YBGAxkdDAM9vb2r169ys3NlUqljY2NfY8/BsNgMKKion7++efQ0NBnz5719vZ2dHTcv38fIaSvr//8+fOampqOjo5+2QvDsO3bt1+6dCkzM7O9vb2srGzTpk1mZmYhISEK+iJGCl2/fl0ikVRWVo7yqrYCisNTsJQGznJvb29LS4tMJistLQ0LC7O0tFy3bt2wGsnPz1fh6DUWi2Vra1tfX6/MQsjIyJDf+TWyVYZh2Pr167OystLS0trb23t6eurr63/77TeEUFBQkImJyRsfLnv//v3/+Z//+eqrr2g0Wt9nFR8+fBghtHv37pcvX+7Zs6erq+vWrVtJSUnr1q2bMmWK4kkKulNsZN0RiOXs6uo63E4BeLOxv50eaAo01Milf/3rX5MnTyY2DFNT04GjepqbmxcuXIhhmI2NzaeffhoeHo4Qsre337VrF3GPj4ODQ1VV1YkTJ7hcLkLIysrq0aNHOI4fO3bM1dUVwzAMw955553U1FQcx4uLi62srJhM5rx582JiYojrjiwWixgj1Nvbm5SU5ODgQKPRBAKBv79/RUUFjuOpqakK+oqIiNDX1+fz+StWrCCGttvZ2YWFhRGHQWw2e9myZYqXkpIjhQYLT8FSqqur6zvLL168CAkJodFo5ubmVCqVy+X6+flVVVUNt5GrV69yOJyEhIQhY8aVG70WGhpKo9FEIhHx8dKlS8St74aGhlu2bOlXOTw8XD56bWSr7PXr1xEREZaWllQq1cjIKCAgoLy8HMdxf39/hFBsbOzACAe7OVw+VOynn36aPXs2g8EwMzMLDw+XSCTy7w42SUF3/fQ9Rh9xdwRvb29zc3P5Q+UGA6PXgJJgK5lAlPk1B+r89QwJCdHX11dPXwRltoHKykoqlXr69Gn1hDSYnp4eT0/P9PR0rewOx/GmpiYMww4fPjxkTcjoQElw1h0AMmngC7js7e3j4+Pj4+P7vs1Mzf6/9u4YVyIoCuP4vORtwBLswB4kWpXEKvRUGldnCyqNkNDQWoKtiEohppC8Yop5iuFmjv9vA+cE8eXGdc+6rk3TzPPs+768crs4ji3LCoLgsooQj0QH8CoMQ8/zfN8/skXuDMMw1HXd9/37P+O/tNzj8ciybBzHruv2UxOAjyDRAT2iKMrzfJom0zSrqtLdzqskSYIgSNNUS3Xbtoui+DvZXli5tm2XZRmGwTCMayriJn51NwDclFJKKaW7i3ccx9knh+KzXNd1XVd3FxCINToAABKQ6AAASECiAwAgAYkOAIAE7Iy7l4MzMO5sv0RlWepu5Cw8A1+HW4aDfrZt090DLnLe5BIAZ+NdjX+R6AAASMB3dAAAJCDRAQCQgEQHAEACEh0AAAme9yszDzUtuRUAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 102
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# 한글을 초성 중성 종성으로 분리한 후 각각 숫자로 임베딩\n",
        "\n",
        "import re\n",
        "import unicodedata\n",
        "\n",
        "CHOSUNGS = \"\".join([chr(_) for _ in range(0x1100, 0x1113)])\n",
        "JOONGSUNGS = \"\".join([chr(_) for _ in range(0x1161, 0x1176)])\n",
        "JONGSUNGS = \"\".join([chr(_) for _ in range(0x11A8, 0x11C3)])\n",
        "ALPHABETS = \"ABCDEFGHIJKLMNOPQRSTUVWXYZ\"\n",
        "NUMBERS = \"0123456789\"\n",
        "SPECIALS = \" ?!\"\n",
        "\n",
        "ALL_VOCABS = \"\".join([\n",
        "    CHOSUNGS,\n",
        "    JOONGSUNGS,\n",
        "    JONGSUNGS,\n",
        "    ALPHABETS,\n",
        "    NUMBERS,\n",
        "    SPECIALS\n",
        "])\n",
        "VOCAB_DICT = {\n",
        "    \"_\": 0,\n",
        "    \"~\": 1,\n",
        "}\n",
        "\n",
        "for idx, v in enumerate(ALL_VOCABS):\n",
        "    VOCAB_DICT[v] = idx + 2\n",
        "\n",
        "\n",
        "def normalize(text):\n",
        "    text = unicodedata.normalize('NFKD', text)\n",
        "    text = text.upper()\n",
        "    text = text.replace('%', unicodedata.normalize('NFKD', '퍼센트'))\n",
        "    regex = unicodedata.normalize('NFKD', r\"[^ \\u11A8-\\u11FF\\u1100-\\u115E\\u1161-\\u11A70-9A-Z?!]\")\n",
        "    text = re.sub(regex, '', text)\n",
        "    text = re.sub(' +', ' ', text)\n",
        "    text = text.strip()\n",
        "    return text\n",
        "\n",
        "\n",
        "def tokenize(text, encoding: bool = True):\n",
        "    tokens = list()\n",
        "\n",
        "    for t in text:\n",
        "        if encoding:\n",
        "            tokens.append(VOCAB_DICT[t])\n",
        "        else:\n",
        "            tokens.append(t)\n",
        "\n",
        "    if encoding:\n",
        "        tokens.append(VOCAB_DICT['~'])\n",
        "    else:\n",
        "        tokens.append('~')\n",
        "\n",
        "    return tokens\n",
        "\n",
        "\n",
        "def text_to_sequence(text):\n",
        "    text = normalize(text)\n",
        "    tokens = tokenize(text, encoding=True)\n",
        "    return tokens\n",
        "\n",
        "# Example usage\n",
        "text = \"안녕하세요\"\n",
        "embedding = text_to_sequence(text)\n",
        "print(embedding)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "p39iI8QBZGqZ",
        "outputId": "bd7f3b24-0101-4f99-f66c-f5e5530d16ea"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[13, 21, 45, 4, 27, 62, 20, 21, 11, 26, 13, 33, 1]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import librosa\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "def pad_2d(data, size):\n",
        "    if data.shape[1] < size:\n",
        "        padding = np.zeros((data.shape[0], size - data.shape[1]))\n",
        "        data = np.hstack((data, padding))\n",
        "    return data\n",
        "\n",
        "def audio_to_mel_spectrogram(audio, sr=16000, n_fft=400, hop_length=160, n_mels=128, length=256):\n",
        "    mel_spectrogram = librosa.feature.melspectrogram(y=audio, sr=sr, n_fft=n_fft, hop_length=hop_length, n_mels=n_mels)\n",
        "    mel_spectrogram = pad_2d(mel_spectrogram, length)  # Ensure the output has a fixed size\n",
        "    return librosa.power_to_db(mel_spectrogram, ref=np.max)"
      ],
      "metadata": {
        "id": "GYq1pqUS2gaQ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Preprocess data(text)\n",
        "train_wavs, train_texts = zip(*train_data)\n",
        "val_wavs, val_texts = zip(*val_data)\n",
        "\n",
        "max_audio_length = 200\n",
        "output_dim = 400\n",
        "\n",
        "# Convert text to sequence\n",
        "train_texts = [text_to_sequence(text) for text in train_texts]\n",
        "val_texts = [text_to_sequence(text) for text in val_texts]\n",
        "\n",
        "# Padding sequences for train and val texts\n",
        "train_texts = pad_sequences(train_texts, dtype='int32', padding='post', maxlen=140)\n",
        "val_texts = pad_sequences(val_texts, dtype='int32', padding='post', maxlen=140)"
      ],
      "metadata": {
        "id": "ylZKTqVJDt1V"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Padding and reshaping train and val wavs\n",
        "train_wavs_mel = []\n",
        "for audio in train_wavs:\n",
        "    if len(audio.shape) == 1:\n",
        "        audio = audio[:80000]  # Cut if size exceeds 80000\n",
        "        audio = np.pad(audio, (0, max(0, 80000 - len(audio))))  # Pad if size is less than 80000\n",
        "        audio = audio.reshape(-1, 400)[:200, :]  # Reshape to 2D and cut to fit into (200, 400)\n",
        "    train_wavs_mel.append(audio)\n",
        "\n",
        "val_wavs_mel = []\n",
        "for audio in val_wavs:\n",
        "    if len(audio.shape) == 1:\n",
        "        audio = audio[:80000]  # Cut if size exceeds 80000\n",
        "        audio = np.pad(audio, (0, max(0, 80000 - len(audio))))  # Pad if size is less than 80000\n",
        "        audio = audio.reshape(-1, 400)[:200, :]  # Reshape to 2D and cut to fit into (200, 400)\n",
        "    val_wavs_mel.append(audio)\n",
        "\n",
        "train_wavs_mel = np.array(train_wavs_mel)\n",
        "val_wavs_mel = np.array(val_wavs_mel)"
      ],
      "metadata": {
        "id": "YR-BQR1-D7Bn"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_wavs_mel[0].shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7UM8Ey4pE6bA",
        "outputId": "6033e93b-c882-4b1b-eca7-c0332ad678b2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(200, 400)"
            ]
          },
          "metadata": {},
          "execution_count": 107
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# train_wavs_1d = train_wavs_mel[0].flatten() # 원상복귀"
      ],
      "metadata": {
        "id": "t7ST1gHQFX_G"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# save_path = 'vae_model.h5'\n",
        "\n",
        "# def train_tts_model(train_data, val_data, model, num_epochs, batch_size, learning_rate, save_path):\n",
        "#     optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)\n",
        "#     criterion = tf.keras.losses.MeanSquaredError()\n",
        "\n",
        "#     # Preprocess data\n",
        "#     train_wavs, train_texts = zip(*train_data)\n",
        "#     train_texts = [text_to_sequence(text) for text in train_texts]\n",
        "\n",
        "#     val_wavs, val_texts = zip(*val_data)\n",
        "#     val_texts = [text_to_sequence(text) for text in val_texts]\n",
        "\n",
        "#     max_audio_length = 100  # Depending on your data\n",
        "#     output_dim = 256  # Depending on your data\n",
        "\n",
        "#     train_wavs_mel = [audio_to_mel_spectrogram(audio)[:max_audio_length, :output_dim] for audio in train_wavs]\n",
        "#     val_wavs_mel = [audio_to_mel_spectrogram(audio)[:max_audio_length, :output_dim] for audio in val_wavs]\n",
        "\n",
        "#     # In case the Mel Spectrogram is shorter than max_audio_length or output_dim, pad the sequences\n",
        "#     train_wavs_mel = pad_sequences(train_wavs_mel, dtype='float32', padding='post', maxlen=max_audio_length)\n",
        "#     val_wavs_mel = pad_sequences(val_wavs_mel, dtype='float32', padding='post', maxlen=max_audio_length)\n",
        "\n",
        "#     train_wavs_mel = np.reshape(train_wavs_mel, (-1, max_audio_length, output_dim))\n",
        "#     val_wavs_mel = np.reshape(val_wavs_mel, (-1, max_audio_length, output_dim))\n",
        "\n",
        "#     # Padding sequences\n",
        "#     train_texts = pad_sequences(train_texts, dtype='int32', padding='post', maxlen=140)\n",
        "#     val_texts = pad_sequences(val_texts, dtype='int32', padding='post', maxlen=140)\n",
        "\n",
        "#     model.compile(optimizer=optimizer, loss=criterion)\n",
        "\n",
        "#     checkpoint_callback = tf.keras.callbacks.ModelCheckpoint(save_path, save_best_only=True, save_weights_only=True)\n",
        "\n",
        "#     # Train the model\n",
        "#     history = model.fit([train_texts, train_texts, train_wavs_mel], train_wavs_mel,\n",
        "#                         validation_data=([val_texts, val_texts, val_wavs_mel], val_wavs_mel),\n",
        "#                         epochs=num_epochs, batch_size=batch_size,\n",
        "#                         callbacks=[checkpoint_callback], verbose=0)\n",
        "#     return history\n",
        "\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint, LambdaCallback\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "import os\n",
        "\n",
        "save_dir = 'vae_model'\n",
        "save_path = os.path.join(save_dir, 'vae_model.h5')\n",
        "\n",
        "def train_tts_model(train_wavs_mel, train_texts, val_wavs_mel, val_texts, model, encoder, num_epochs, batch_size, learning_rate, save_path):\n",
        "    optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate)\n",
        "    criterion = tf.keras.losses.MeanSquaredError()\n",
        "\n",
        "    model.compile(optimizer=optimizer, loss=criterion)\n",
        "\n",
        "    # ModelCheckpoint for saving the best model\n",
        "    checkpoint_callback = ModelCheckpoint(save_path, save_best_only=True, save_weights_only=True)\n",
        "\n",
        "    # LambdaCallback for saving the z_mean values\n",
        "    z_means = []\n",
        "    save_z_mean = LambdaCallback(on_epoch_end=lambda epoch, logs: z_means.append(encoder.predict([train_texts, train_wavs_mel])))\n",
        "\n",
        "    # Train the model\n",
        "    history = model.fit([train_texts, train_wavs_mel], train_wavs_mel,\n",
        "                        validation_data=([val_texts, val_wavs_mel], val_wavs_mel),\n",
        "                        epochs=num_epochs, batch_size=batch_size,\n",
        "                        callbacks=[checkpoint_callback, save_z_mean], verbose=0)\n",
        "\n",
        "    # Save the z_means\n",
        "    os.makedirs(save_dir, exist_ok=True)\n",
        "    np.save(os.path.join(save_dir, 'z_means.npy'), np.array(z_means))\n",
        "\n",
        "    return history\n"
      ],
      "metadata": {
        "id": "boXqFIR3sVk8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# def data_generator(data, batch_size):\n",
        "#     while True:\n",
        "#         np.random.shuffle(data)\n",
        "#         for i in range(0, len(data), batch_size):\n",
        "#             batch_data = data[i:i+batch_size]\n",
        "#             batch_wavs, batch_texts = [], []\n",
        "#             for wav, text in batch_data:\n",
        "#                 batch_wavs.append(wav)\n",
        "#                 batch_texts.append(text)\n",
        "#             yield np.array(batch_wavs), np.array"
      ],
      "metadata": {
        "id": "uEmhZI9iscGs"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "num_epochs = 100\n",
        "batch_size = 32\n",
        "learning_rate = 0.001\n",
        "\n",
        "# Train the model\n",
        "history = train_tts_model(train_wavs_mel, train_texts, val_wavs_mel, val_texts, vae, encoder, num_epochs, batch_size, learning_rate, save_path)\n"
      ],
      "metadata": {
        "id": "qRKdAYMe2WFJ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e8511609-b43b-4f94-f360-524f75102895"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "19/19 [==============================] - 1s 9ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 9ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 9ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 12ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 11ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 9ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 11ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 11ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 8ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n",
            "19/19 [==============================] - 0s 7ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Save loss and validation loss data\n",
        "np.savez(\"loss_history.npz\", loss=history.history['loss'], val_loss=history.history['val_loss'])\n",
        "\n",
        "# Plot the loss and validation loss curves\n",
        "epochs = range(1, len(history.history['loss']) + 1)\n",
        "\n",
        "plt.plot(epochs, history.history['loss'], 'b', label='Training Loss')\n",
        "plt.plot(epochs, history.history['val_loss'], 'r', label='Validation Loss')\n",
        "plt.title('Training and Validation Loss')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "plt.savefig(\"loss_curve.png\")\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "QxoMzVdnJw6m",
        "outputId": "687ff3fd-eab4-4abf-c1ac-9abfee56edad"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.save(save_path)"
      ],
      "metadata": {
        "id": "f3QtI1ueSlMq"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def load_z_means(file_path):\n",
        "    return np.load(file_path)\n",
        "\n",
        "# Usage\n",
        "z_means_file = os.path.join(save_dir, 'z_means.npy')\n",
        "z_means = load_z_means(z_means_file)"
      ],
      "metadata": {
        "id": "v-y2M5IlMuJ5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def generate_speech(text, model):\n",
        "    # Preprocess the input text\n",
        "    input_text = pad_sequences([text_to_sequence(text)], dtype='int32', padding='post', maxlen=140)\n",
        "\n",
        "    # Create fake audio data with Gaussian noise\n",
        "    fake_audio_data = np.random.normal(0, 1, (1, 200, 400))\n",
        "\n",
        "    # Generate the speech from the model\n",
        "    generated_speech = model.predict([input_text, fake_audio_data])\n",
        "\n",
        "    return generated_speech\n",
        "\n",
        "# Use the function to generate speech\n",
        "text = \"안녕하세요 반갑습니다. 저는 발표자입니다.\"\n",
        "generated_speech = generate_speech(text, model)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iWp0G1M_bhw7",
        "outputId": "c3011c1a-4c8d-4646-e61d-f9ae9635e349"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1/1 [==============================] - 0s 43ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "generated_speech.flatten().shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RqMXYDK-fBW-",
        "outputId": "07913853-1e5e-4142-dbb2-a9cbe2876d4a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(80000,)"
            ]
          },
          "metadata": {},
          "execution_count": 133
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import soundfile as sf\n",
        "reshaped_array = generated_speech.flatten()\n",
        "mp3_file_path = \"/content/drive/MyDrive/generated_voice.mp3\"  # Replace with the desired output file path\n",
        "sf.write(mp3_file_path, reshaped_array, 22050, format='mp3')"
      ],
      "metadata": {
        "id": "CLtMMl_XdwNo"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}